From 5f268563802a05d588b6dd2dd80640f771acf6a8 Mon Sep 17 00:00:00 2001 From: Max Ghenis Date: Thu, 29 Oct 2026 10:49:10 -0400 Subject: [PATCH 01/52] Add the Autumn Budget 2026 measure registry and stage HM Treasury's Tables 6.1, 6.2 and 6.4 Tranche 1 of #230, following the AB2025 port (#136). - data/uk/ab2026_measures.json: 104 measures. Every Table 6.1 line once, under a key written out per line in pipeline/ab2026_measure_keys.py, plus Table 6.2 line 4 (Sovereign AI) and five packages. The engine pin is policyengine-uk 2.151.0, which the dashboard's Budget-day receipts record. The seven dashboard measures carry pe_modifier with their receipt hashes. Savings rates and the BADR limit are parameter deltas; the bus fare cap is partial. 89 gaps each carry a computed name search at the pin. - pipeline/stage_ab2026_hmt_tables.py stages 762 rows from the two pinned workbooks. Table 6.4 recostings carry the earlier event's key. - The validator accepts executed_by_dashboard and package_of_registry_measures constructions and resolves replaced and read variables against the engine. - pipeline/harvest_counts.py writes and checks COUNTS.json. - The harvest README, the seed inventory and the tests for the registry and the harvest contract. Co-Authored-By: Claude Opus 5.5 --- data/uk/ab2026_measures.json | 2305 +++++++++++++++++ pipeline/ab2026_measure_keys.py | 203 ++ pipeline/build_ab2026_registry.py | 1170 +++++++++ pipeline/harvest_counts.py | 75 + pipeline/stage_ab2026_hmt_tables.py | 386 +++ pipeline/validate_budget_2026_registry.py | 83 +- .../harvest-uk-ab2026-2026-10-29/COUNTS.json | 16 + .../harvest-uk-ab2026-2026-10-29/README.md | 45 + .../REGISTRY_KEYS.md | 112 + .../SEED_INVENTORY/publications.md | 62 + .../uk_hmt_tables/NOTES.md | 47 + .../uk_hmt_tables/claims_staged.jsonl.gz | Bin 0 -> 38260 bytes .../uk_hmt_tables/manifest.jsonl | 2 + tests/test_ab2026_registry.py | 262 ++ tests/test_uk_ab2026_harvest.py | 321 +++ 15 files changed, 5083 insertions(+), 6 deletions(-) create mode 100644 data/uk/ab2026_measures.json create mode 100644 pipeline/ab2026_measure_keys.py create mode 100644 pipeline/build_ab2026_registry.py create mode 100644 pipeline/harvest_counts.py create mode 100644 pipeline/stage_ab2026_hmt_tables.py create mode 100644 sources/harvest-uk-ab2026-2026-10-29/COUNTS.json create mode 100644 sources/harvest-uk-ab2026-2026-10-29/README.md create mode 100644 sources/harvest-uk-ab2026-2026-10-29/REGISTRY_KEYS.md create mode 100644 sources/harvest-uk-ab2026-2026-10-29/SEED_INVENTORY/publications.md create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/NOTES.md create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/claims_staged.jsonl.gz create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/manifest.jsonl create mode 100644 tests/test_ab2026_registry.py create mode 100644 tests/test_uk_ab2026_harvest.py diff --git a/data/uk/ab2026_measures.json b/data/uk/ab2026_measures.json new file mode 100644 index 000000000..8efdc6ea0 --- /dev/null +++ b/data/uk/ab2026_measures.json @@ -0,0 +1,2305 @@ +{ + "schema_version": 1, + "fiscal_event": "autumn_budget_2026", + "event_date": "2026-10-28", + "chancellor": "John Healey", + "administration": "Burnham government (from 2026-07-20)", + "registry_rule": "One entry per measure ANNOUNCED at Autumn Budget 2026: every HM Treasury Table 6.1 line (including the lines a household model cannot touch), the one Table 6.2 financial transaction with no Table 6.1 line, and the packages producers score. It is the port's key space (#230): every staged AB2026 claim carries conditions.measure_key from this file, and every measure carries a computability verdict against the engine the Budget-day runs used. It stages NO value claims; values live in the harvest under their producers (sources/harvest-uk-ab2026-2026-10-29/).", + "measure_key_rule": "ab2026__, written out per Table 6.1 line in pipeline/ab2026_measure_keys.py so a reworded title cannot move a key; ab2026__package_ for packages; ab2026_option__ for options costed by others. The pre-event registry (data/uk/budget_2026_measures.json, #96) keeps its own keys; pre_event_outcomes records what became of each.", + "verification_note": "Every pe_reform_delta and parameter_legs path was RESOLVED against an installed policyengine-uk 2.151.0 on 2026-10-29, and engine_baseline_2026 records what that engine returns for 2026-06-01. Every not_expressible verdict carries a name_search COMPUTED over the pinned engine's parameter tree (4292 leaves) and variable list (1271), never typed. Mechanisms named in a verdict were read from the engine's source at the pin or from the dashboard's source at PolicyEngine/uk-autumn-budget-dashboard-2026@1eca3182.", + "engine_pin": { + "policyengine_uk": "2.151.0", + "resolved_on": "2026-10-29", + "pin_meaning": "The pin follows the Budget-day runs: every published receipt of the UK Autumn Budget dashboard records policyengine 6.11.1 with certified_for_model_version 2.151.0, and those runs are AB2026's PolicyEngine counterparts (tranche 3). It is NOT the certified populace pair data/uk/certified_pairs.json holds (policyengine 6.11.0, policyengine-uk 2.132.6); tranche 3 adds a 6.11.1 pair record before any counterpart is ingested.", + "budget_day_runs": { + "policyengine": "6.11.1", + "certified_for_model_version": "2.151.0", + "dataset_revision": "microcosm-uk-2024-25-national-20261025T091414Z-a6299504", + "dataset_sha256": "b96a8cd625b645a409c063b1de5158532708a2f0d3c3897a794a8fb0bbd5064e", + "basis": "static; OBR October 2026 EFO growth rates applied over policyengine-uk's March 2026 rates (receipt dashboard.basis)" + } + }, + "computability": { + "expressible": "every leg is a resolvable parameter at the pin, or the measure was executed by a receipt-pinned dashboard modifier (construction executed_by_dashboard)", + "partial": "some legs resolve; the missing ones are named in `missing`", + "not_expressible": "the engine has no representation; `why` names the gap, `name_search` records the computed search, `action_link` says where it goes" + }, + "construction_vocabulary": { + "executed_by_dashboard": "the measure ran on Budget day as a dashboard reform; pe_modifier names the module, function, arguments, the engine variables it replaces or reads, its parameter legs, and the published receipt (path, file sha256, reform sha256, run identity)", + "delta_on_current_law": "pe_reform_delta applied to current law at the pin", + "reversal_on_certified_world": "the measure is already in the engine's current law; a counterpart executes the pre-measure world as baseline (#56 convention)", + "package_of_registry_measures": "a composition of registry measures (`package_of`, a list of keys, as in AB2025); a PolicyEngine number covers the expressible legs, stated as constructed" + }, + "exogenous_levers_note": "The engine carries exogenous incidence levers that spread a £ amount supplied as an input: gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'); gov.contrib.policyengine.budget.corporate_incident_tax_change ('Tax increase incident on owners of capital'); gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'). Feeding a Table 6.1 figure through one restates HM Treasury's costing rather than estimating it, so no lever makes a line expressible here.", + "action_link_rule": "Every not_expressible measure links to #230, which lists the upstream development items; the bus fare cap's partial verdict links to #106, the bus gap record.", + "pre_event_outcomes": { + "ab2026__property_income_tax_rate_rise": "announced: Table 6.1 line 61 (ab2026__property_income_rates_plus_2pp)", + "ab2026__bank_profit_surcharge_increase": "announced: Table 6.1 line 55 (ab2026__bank_surcharge_6pc_temporary), temporary, 3% to 6% for 15 months", + "ab2026__business_rates_warehouse_up_pub_down": "partly: Table 6.1 line 12 raises pubs' relief (ab2026__business_rates_pubs_relief_35pc); no warehouse surcharge line in Table 6.1 or 6.4", + "ab2026__high_value_council_tax_surcharge": "an AB2025 measure; Table 6.1 line 93 changes its reliefs (ab2026__hvcts_deferrals_discounts_exemptions) and Table 6.4 line 11 recosts it", + "ab2026__cgt_align_with_income_tax_rates": "not in Table 6.1 or 6.4", + "ab2026__cgt_remove_uplift_on_death": "not in Table 6.1 or 6.4", + "ab2026__pension_tax_free_lump_sum_restriction": "not in Table 6.1 or 6.4", + "ab2026__ni_extended_to_investment_property_pension_income": "not in Table 6.1 or 6.4", + "ab2026__personal_allowance_freeze_extension": "not in Table 6.1 or 6.4", + "ab2026__wealth_tax_2pc_above_10m": "not in Table 6.1 or 6.4", + "ab2026__council_tax_and_sdlt_to_land_value_tax": "not in Table 6.1 or 6.4", + "ab2026__online_sales_levy": "not in Table 6.1 or 6.4 (line 46, online marketplace liability, is a VAT collection measure, not a levy)", + "ab2026__vape_duty_increase": "not in Table 6.1 or 6.4 (line 57 removes business rates reliefs for vape shops)", + "ab2026__ated_increase_and_sdlt_envelope_close": "not in Table 6.1 or 6.4" + }, + "triage": { + "expressible": 11, + "partial": 4, + "not_expressible": 89, + "executed_by_dashboard": 7, + "policyengine_uk_development_item": 13, + "out_of_model_scope": 76, + "announced": 104 + }, + "measures": [ + { + "measure_key": "ab2026__household_energy_payment_100_bands_a_to_d", + "title": "Household Energy Payment: £100 council tax rebate for bands A-D from January 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 1 (Spend, RDEL); exchequer impact GBP m by FY: {'2026-27': '-2625', '2027-28': '0', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 1, + "table_61_head": "Spend", + "table_61_heading": "Breathing space for families and businesses > Breathing space for families", + "program": "household_energy_payment", + "computability": "expressible", + "construction": "executed_by_dashboard", + "dashboard_measure_id": "household_energy_payment_2026", + "pe_reform_delta": null, + "pe_modifier": { + "module": "src/uk_budget_data/budget_measures/household_energy_payment_2026.py", + "function": "household_energy_payment_2026", + "arguments": { + "amount": 100, + "year": 2026, + "country": "ENGLAND", + "bands": [ + "A", + "B", + "C", + "D" + ] + }, + "replaced_variables": [ + "ebr_council_tax_rebate" + ], + "reads_variables": [ + "council_tax_band", + "country" + ], + "parameter_legs": [], + "receipt": { + "repo": "PolicyEngine/uk-autumn-budget-dashboard-2026", + "commit": "1eca3182d6983b6375f401a3565ac56c25fa401e", + "path": "public/data/artifacts/household_energy_payment_2026/101a07dc7238a10e/receipt.json", + "receipt_file_sha256": "0fb0414b45435799fb50afa9d79c2c7af8998de2ee406d8da6c1561bd46298d4", + "reform_sha256": "ceede8365a9462ac389dc5210882ea23850babfed6e5bff8cf264e774cd616ad", + "reform_binding": "code_revision", + "generated_at": "2026-10-28T17:07:18.666809+00:00", + "policyengine": "6.11.1", + "certified_for_model_version": "2.151.0", + "dataset_revision": "microcosm-uk-2024-25-national-20261025T091414Z-a6299504", + "dataset_sha256": "b96a8cd625b645a409c063b1de5158532708a2f0d3c3897a794a8fb0bbd5064e", + "basis": "static; OBR October 2026 EFO growth rates applied over policyengine-uk's March 2026 rates (receipt dashboard.basis)" + } + }, + "engine_baseline_2026": { + "gov.treasury.energy_bills_rebate.council_tax_rebate.amount": 0 + }, + "note": "The modifier adds £100 in 2026-27 to the native ebr_council_tax_rebate formula for bands A to D in England; the native rebate's own parameter is 0 at the pin, so the payment is the whole effect. Scotland, Wales and Northern Ireland get a Barnett share instead, not modelled." + }, + { + "measure_key": "ab2026__electricity_vat_zero_rate_to_july_2027", + "title": "Support for Energy Bills: Cut VAT on domestic electricity from 5% to 0% from 1 October 2026 to 1 July 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 2 (Tax, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '-755', '2027-28': '-305', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 2, + "table_61_head": "Tax", + "table_61_heading": "Breathing space for families and businesses > Breathing space for families", + "program": "vat", + "computability": "expressible", + "construction": "executed_by_dashboard", + "dashboard_measure_id": "domestic_electricity_zero_vat", + "pe_reform_delta": null, + "pe_modifier": { + "module": "src/uk_budget_data/budget_measures/zero_vat_on_domestic_electricity.py", + "function": "zero_vat_on_domestic_electricity", + "arguments": { + "rate_change_2026": -0.028342979605206865, + "rate_change_2027": -0.011149919361563782, + "geography": "GB", + "geography_2027": "UK" + }, + "replaced_variables": [], + "reads_variables": [ + "electricity_consumption" + ], + "parameter_legs": [], + "receipt": { + "repo": "PolicyEngine/uk-autumn-budget-dashboard-2026", + "commit": "1eca3182d6983b6375f401a3565ac56c25fa401e", + "path": "public/data/artifacts/domestic_electricity_zero_vat/2d9c65726915303e/receipt.json", + "receipt_file_sha256": "e464b000c2af360654462efd05180d0fed272728911dc7292d932b99117246ca", + "reform_sha256": "5ee4d8844432367d5557df38ee89e23b206e2c1236bce2451b0336ada02b58b6", + "reform_binding": "code_revision", + "generated_at": "2026-10-28T17:08:09.754595+00:00", + "policyengine": "6.11.1", + "certified_for_model_version": "2.151.0", + "dataset_revision": "microcosm-uk-2024-25-national-20261025T091414Z-a6299504", + "dataset_sha256": "b96a8cd625b645a409c063b1de5158532708a2f0d3c3897a794a8fb0bbd5064e", + "basis": "static; OBR October 2026 EFO growth rates applied over policyengine-uk's March 2026 rates (receipt dashboard.basis)" + } + }, + "engine_baseline_2026": { + "gov.hmrc.vat.reduced_rate": 0.05 + }, + "note": "The modifier lowers the VAT rate on electricity_consumption by 5 points times the share of each fiscal year's use that falls in the zero-rated period (about 57% of 2026-27 in Great Britain; about 22% of 2027-28 across the UK). Electricity only, not gas." + }, + { + "measure_key": "ab2026__fuel_duty_5p_cut_extended_and_rpi_rise_deferred", + "title": "Fuel Duty: Extend the temporary 5p-per-litre duty cut and defer the planned Retail Prices Index (RPI) increase to 1 July 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 3 (Tax, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '-575', '2027-28': '-780', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 3, + "table_61_head": "Tax", + "table_61_heading": "Breathing space for families and businesses > Breathing space for families", + "program": "fuel_duty", + "computability": "expressible", + "construction": "executed_by_dashboard", + "dashboard_measure_id": "fuel_duty_rise_deferral", + "pe_reform_delta": null, + "pe_modifier": { + "module": "src/uk_budget_data/reforms.py (_create_fuel_duty_rise_deferral) + src/uk_budget_data/budget_measures/pump_vat_on_fuel_duty_deferral.py", + "function": "fuel_duty_deferral_rates + pump_vat_on_fuel_duty_deferral", + "arguments": {}, + "replaced_variables": [], + "reads_variables": [ + "petrol_spending", + "diesel_spending" + ], + "parameter_legs": [ + "gov.hmrc.fuel_duty.petrol_and_diesel" + ], + "receipt": { + "repo": "PolicyEngine/uk-autumn-budget-dashboard-2026", + "commit": "1eca3182d6983b6375f401a3565ac56c25fa401e", + "path": "public/data/artifacts/fuel_duty_rise_deferral/e077f8d4b70440c8/receipt.json", + "receipt_file_sha256": "2f547d91b455b7e84cf8495cbfd0475ca626d7f1b4a95605de869cf61d52a30f", + "reform_sha256": "1dae41af83e4402f6b3af756d7dcdcc1dd525883dd32c3c8b9050456f52310f1", + "reform_binding": "code_revision", + "generated_at": "2026-10-28T17:06:46.194418+00:00", + "policyengine": "6.11.1", + "certified_for_model_version": "2.151.0", + "dataset_revision": "microcosm-uk-2024-25-national-20261025T091414Z-a6299504", + "dataset_sha256": "b96a8cd625b645a409c063b1de5158532708a2f0d3c3897a794a8fb0bbd5064e", + "basis": "static; OBR October 2026 EFO growth rates applied over policyengine-uk's March 2026 rates (receipt dashboard.basis)" + } + }, + "engine_baseline_2026": { + "gov.hmrc.fuel_duty.petrol_and_diesel": 0.539527397260274 + }, + "note": "Day-weighted annual fuel duty rates (fuel_duty_deferral_rates), plus the VAT on the lower pump price through a modifier on petrol and diesel spending." + }, + { + "measure_key": "ab2026__carers_allowance_earnings_taper", + "title": "Carer’s Allowance: Introduce a taper on earnings above the Weekly Earning Limit within 2030-31", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 7 (Spend, RAME); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '0', '2029-30': '0', '2030-31': '-40', '2031-32': '-80'}", + "table_61_line": 7, + "table_61_head": "Spend", + "table_61_heading": "Breathing space for families and businesses > Breathing space for families", + "program": "carers_allowance", + "computability": "expressible", + "construction": "executed_by_dashboard", + "dashboard_measure_id": "carers_allowance_earnings_taper", + "pe_reform_delta": null, + "pe_modifier": { + "module": "src/uk_budget_data/budget_measures/carers_allowance_earnings_taper.py", + "function": "carers_allowance_earnings_taper", + "arguments": { + "start_year": 2030, + "first_year_share": 0.5, + "taper_rate": 1.0 + }, + "replaced_variables": [ + "carers_allowance_pre_overlap" + ], + "reads_variables": [ + "carers_allowance_earnings", + "care_hours", + "would_claim_carers_allowance", + "carers_allowance_reported" + ], + "parameter_legs": [], + "receipt": { + "repo": "PolicyEngine/uk-autumn-budget-dashboard-2026", + "commit": "1eca3182d6983b6375f401a3565ac56c25fa401e", + "path": "public/data/artifacts/carers_allowance_earnings_taper/9458905cefca3e88/receipt.json", + "receipt_file_sha256": "0852c433f5f76cd2e7b27f1bcf6721610d8e36e80b5f416727901e8125d273f6", + "reform_sha256": "f45943c11aed787c570af3cc82d11af9cf181ce0877b5cb9350a9bc1e2f84dc8", + "reform_binding": "code_revision", + "generated_at": "2026-10-28T17:08:00.734195+00:00", + "policyengine": "6.11.1", + "certified_for_model_version": "2.151.0", + "dataset_revision": "microcosm-uk-2024-25-national-20261025T091414Z-a6299504", + "dataset_sha256": "b96a8cd625b645a409c063b1de5158532708a2f0d3c3897a794a8fb0bbd5064e", + "basis": "static; OBR October 2026 EFO growth rates applied over policyengine-uk's March 2026 rates (receipt dashboard.basis)" + } + }, + "engine_baseline_2026": { + "gov.dwp.carers_allowance.earnings_limit": 204, + "gov.dwp.carers_allowance.min_hours": 35, + "gov.dwp.carers_allowance.min_age": 16 + }, + "note": "A 100% taper above the Weekly Earnings Limit (16 hours at the National Living Wage) in England and Wales, half of 2030-31 counted because the start date within that year is not set." + }, + { + "measure_key": "ab2026__trading_allowance_3000", + "title": "Trading Allowance: Increase the Income Tax Trading Allowance to £3,000 from £1,000 from 6 April 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 15 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '-225', '2029-30': '-165', '2030-31': '-175', '2031-32': '-180'}", + "table_61_line": 15, + "table_61_head": "Tax", + "table_61_heading": "Breathing space for businesses", + "program": "income_tax", + "computability": "expressible", + "construction": "executed_by_dashboard", + "dashboard_measure_id": "trading_allowance_raise", + "pe_reform_delta": { + "gov.hmrc.income_tax.allowances.trading_allowance": { + "2027": 3000, + "2028": 3000, + "2029": 3000, + "2030": 3000, + "2031": 3000 + } + }, + "delta_kind": "absolute_value_by_year", + "pe_modifier": { + "module": "src/uk_budget_data/reforms.py", + "function": "_create_trading_allowance_raise", + "arguments": {}, + "replaced_variables": [], + "reads_variables": [], + "parameter_legs": [ + "gov.hmrc.income_tax.allowances.trading_allowance" + ], + "receipt": { + "repo": "PolicyEngine/uk-autumn-budget-dashboard-2026", + "commit": "1eca3182d6983b6375f401a3565ac56c25fa401e", + "path": "public/data/artifacts/trading_allowance_raise/eddde823a65b7be4/receipt.json", + "receipt_file_sha256": "59db02ec313c62f8c194580c806e782431f67a0f3b08df34adb91445a1eb9221", + "reform_sha256": "cdcc8b19d5234afe31ae857db16a8f2d6a18578a1d7da1c890b3c5f21832c268", + "reform_binding": "content", + "generated_at": "2026-10-28T17:07:22.701184+00:00", + "policyengine": "6.11.1", + "certified_for_model_version": "2.151.0", + "dataset_revision": "microcosm-uk-2024-25-national-20261025T091414Z-a6299504", + "dataset_sha256": "b96a8cd625b645a409c063b1de5158532708a2f0d3c3897a794a8fb0bbd5064e", + "basis": "static; OBR October 2026 EFO growth rates applied over policyengine-uk's March 2026 rates (receipt dashboard.basis)" + } + }, + "engine_baseline_2026": { + "gov.hmrc.income_tax.allowances.trading_allowance": 1000 + }, + "note": "The data hold profits, not gross trading receipts, so the engine relieves a profit up to £3,000 in full and gives no relief above it (the dashboard's own description)." + }, + { + "measure_key": "ab2026__pension_tax_allowance", + "title": "Pension Tax Allowance: Remove the income tax liability for pensioners with total income up to the value of the new State Pension from 6 April 2027 until 5 April 2030", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 52 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-45', '2028-29': '-90', '2029-30': '-150', '2030-31': '50', '2031-32': '-15'}", + "table_61_line": 52, + "table_61_head": "Tax", + "table_61_heading": "Funding the Budget fairly and responsibly > Modernising the tax and welfare system", + "program": "income_tax", + "computability": "expressible", + "construction": "executed_by_dashboard", + "dashboard_measure_id": "state_pension_income_tax_protection", + "pe_reform_delta": null, + "pe_modifier": { + "module": "src/uk_budget_data/budget_measures/pension_tax_allowance.py", + "function": "pension_tax_allowance", + "arguments": { + "first_year": 2027, + "last_year": 2029, + "basic_rate": 0.2, + "taper_divisor": 3, + "weekly_rounding": 0.05 + }, + "replaced_variables": [ + "income_tax", + "uc_income_tax_on_earnings", + "uc_income_tax_on_trade" + ], + "reads_variables": [ + "total_income", + "is_SP_age" + ], + "parameter_legs": [], + "receipt": { + "repo": "PolicyEngine/uk-autumn-budget-dashboard-2026", + "commit": "1eca3182d6983b6375f401a3565ac56c25fa401e", + "path": "public/data/artifacts/state_pension_income_tax_protection/1bfeed3c27a3c9b9/receipt.json", + "receipt_file_sha256": "4f20f3c69f9ddb3b7894d3281e9416774a1e12d747de1e573629570871b3e960", + "reform_sha256": "65a95c811e5d6c6dd0b54255dd20ab1439c542994f1918778b93769c14199524", + "reform_binding": "code_revision", + "generated_at": "2026-10-28T17:08:11.348926+00:00", + "policyengine": "6.11.1", + "certified_for_model_version": "2.151.0", + "dataset_revision": "microcosm-uk-2024-25-national-20261025T091414Z-a6299504", + "dataset_sha256": "b96a8cd625b645a409c063b1de5158532708a2f0d3c3897a794a8fb0bbd5064e", + "basis": "static; OBR October 2026 EFO growth rates applied over policyengine-uk's March 2026 rates (receipt dashboard.basis)" + } + }, + "engine_baseline_2026": { + "gov.dwp.state_pension.new_state_pension.amount": 241.3, + "gov.hmrc.income_tax.allowances.personal_allowance.amount": 12570 + }, + "note": "A non-refundable income tax reduction for people over State Pension age: 20% of the full new State Pension above the personal allowance, withdrawn £1 for every £3 of total income above the full new State Pension, 2027-28 to 2029-30." + }, + { + "measure_key": "ab2026__property_income_rates_plus_2pp", + "title": "Property Income: Increase tax rates on property income by 2ppts at the basic, higher and additional rate from 6 April 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 61 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '35', '2028-29': '575', '2029-30': '435', '2030-31': '445', '2031-32': '455'}", + "table_61_line": 61, + "table_61_head": "Tax", + "table_61_heading": "Making fair choices on revenue raising", + "program": "income_tax", + "computability": "expressible", + "construction": "executed_by_dashboard", + "dashboard_measure_id": "property_income_rates_up", + "pe_reform_delta": { + "gov.hmrc.income_tax.rates.property.basic": { + "2027": 0.24, + "2028": 0.24, + "2029": 0.24, + "2030": 0.24, + "2031": 0.24 + }, + "gov.hmrc.income_tax.rates.property.higher": { + "2027": 0.44, + "2028": 0.44, + "2029": 0.44, + "2030": 0.44, + "2031": 0.44 + }, + "gov.hmrc.income_tax.rates.property.additional": { + "2027": 0.49, + "2028": 0.49, + "2029": 0.49, + "2030": 0.49, + "2031": 0.49 + } + }, + "delta_kind": "absolute_value_by_year", + "pe_modifier": { + "module": "src/uk_budget_data/reforms.py", + "function": "property_income_rates_up", + "arguments": {}, + "replaced_variables": [], + "reads_variables": [], + "parameter_legs": [ + "gov.hmrc.income_tax.rates.property.basic", + "gov.hmrc.income_tax.rates.property.higher", + "gov.hmrc.income_tax.rates.property.additional" + ], + "receipt": { + "repo": "PolicyEngine/uk-autumn-budget-dashboard-2026", + "commit": "1eca3182d6983b6375f401a3565ac56c25fa401e", + "path": "public/data/artifacts/property_income_rates_up/7bf77b35e33233e0/receipt.json", + "receipt_file_sha256": "20df7073b4481a5f5b0073c6949e0d0c2894adac88d6e70778946456b2fe899d", + "reform_sha256": "d46bf10b10b298db4059d7b6a2dccac7997e09afec0a88542c2059ea70646aa0", + "reform_binding": "content", + "generated_at": "2026-10-28T17:10:45.376537+00:00", + "policyengine": "6.11.1", + "certified_for_model_version": "2.151.0", + "dataset_revision": "microcosm-uk-2024-25-national-20261025T091414Z-a6299504", + "dataset_sha256": "b96a8cd625b645a409c063b1de5158532708a2f0d3c3897a794a8fb0bbd5064e", + "basis": "static; OBR October 2026 EFO growth rates applied over policyengine-uk's March 2026 rates (receipt dashboard.basis)" + } + }, + "engine_baseline_2026": { + "gov.hmrc.income_tax.rates.property.basic": 0.2, + "gov.hmrc.income_tax.rates.property.higher": 0.4, + "gov.hmrc.income_tax.rates.property.additional": 0.45 + }, + "note": "Current law's property rates plus 2 points from 2027 (pe_reform_delta computed at the pin from each year's 6 April value). Scottish taxpayers' property income follows Scottish rates; finance cost relief follows the property basic rate UK-wide." + }, + { + "measure_key": "ab2026__savings_income_rates_plus_2pp", + "title": "Savings Income: Increase tax rates on savings income by 2ppts at the basic, higher and additional rate from 6 April 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 62 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '15', '2027-28': '125', '2028-29': '555', '2029-30': '520', '2030-31': '555', '2031-32': '590'}", + "table_61_line": 62, + "table_61_head": "Tax", + "table_61_heading": "Making fair choices on revenue raising", + "program": "income_tax", + "computability": "expressible", + "construction": "delta_on_current_law", + "dashboard_measure_id": "savings_income_rates_up", + "pe_reform_delta": { + "gov.hmrc.income_tax.rates.savings.basic": { + "2027": 0.24, + "2028": 0.24, + "2029": 0.24, + "2030": 0.24, + "2031": 0.24 + }, + "gov.hmrc.income_tax.rates.savings.higher": { + "2027": 0.44, + "2028": 0.44, + "2029": 0.44, + "2030": 0.44, + "2031": 0.44 + }, + "gov.hmrc.income_tax.rates.savings.additional": { + "2027": 0.49, + "2028": 0.49, + "2029": 0.49, + "2030": 0.49, + "2031": 0.49 + } + }, + "delta_kind": "absolute_value_by_year", + "engine_baseline_2026": { + "gov.hmrc.income_tax.rates.savings.basic": 0.2, + "gov.hmrc.income_tax.rates.savings.higher": 0.4, + "gov.hmrc.income_tax.rates.savings.additional": 0.45 + }, + "note": "Registered by the dashboard but not in its published release, so there is no published receipt to cite; tranche 3 computes it from this delta." + }, + { + "measure_key": "ab2026__badr_lifetime_limit_500000", + "title": "Business Asset Disposal Relief (BADR): Reduce the lifetime limit to £500,000 from 6 April 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 60 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '5', '2027-28': '60', '2028-29': '180', '2029-30': '255', '2030-31': '280', '2031-32': '305'}", + "table_61_line": 60, + "table_61_head": "Tax", + "table_61_heading": "Making fair choices on revenue raising", + "program": "capital_gains_tax", + "computability": "expressible", + "construction": "delta_on_current_law", + "pe_reform_delta": { + "gov.hmrc.cgt.badr.lifetime_limit": { + "2027": 500000, + "2028": 500000, + "2029": 500000, + "2030": 500000, + "2031": 500000 + } + }, + "delta_kind": "absolute_value_by_year", + "engine_baseline_2026": { + "gov.hmrc.cgt.badr.lifetime_limit": 1000000 + }, + "head_variables": [ + "capital_gains_tax", + "capital_gains_badr" + ], + "note": "capital_gains_tax reads gov.hmrc.cgt.badr.lifetime_limit at the pin (variables/gov/hmrc/capital_gains_tax/capital_gains_tax.py). Not run on Budget day." + }, + { + "measure_key": "ab2026__bus_fare_cap_2_england_outside_london_2027", + "title": "Bus Fares: Cap single bus fares at £2 in England outside London for one year from 1 January 2027, funded by reprioritising DESNZ budgets", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 4 (Spend, TDEL); exchequer impact GBP m by FY: {'2026-27': '-50', '2027-28': '-180', '2028-29': '65', '2029-30': '170', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 4, + "table_61_head": "Spend", + "table_61_heading": "Breathing space for families and businesses > Breathing space for families", + "program": "bus_fares", + "computability": "partial", + "intended_construction": "reversal_on_certified_world", + "pe_reform_delta": null, + "engine_baseline_2026": { + "gov.dft.bus.fares.cap": 3 + }, + "engine_parameter_history": { + "gov.dft.bus.fares.cap": { + "2023-01-01": 2, + "2025-01-01": 3, + "2027-01-01": 2, + "2028-01-01": "inf" + } + }, + "missing": "The pre-measure 2027 cap. The engine already carries the measure: gov.dft.bus.fares.cap is £2 from 1 January 2027 to 1 January 2028 (its metadata cites DfT's '£2 bus fares from January 2027'). A counterpart reverses it, which needs the cap 2027 would have had without the measure; the parameter has one upstream revision (policyengine-uk c8d1f89a9d, 5 October 2026) and records no pre-announcement path. Supersedes data/uk/bus_fares_gap.json's not_expressible verdict at 2.89.2, which predates the bus model.", + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/106" + }, + { + "measure_key": "ab2026__mileage_allowance_payments_55p", + "title": "Mileage Rates: Increase the tax-free and tax-deductible car and van rates to 55p for the first 10,000 business miles from 6 April 2026 and the higher mileage rate to 35p from 6 April 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 5 (Tax, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '-30', '2027-28': '-200', '2028-29': '-220', '2029-30': '-225', '2030-31': '-235', '2031-32': '-245'}", + "table_61_line": 5, + "table_61_head": "Tax", + "table_61_heading": "Breathing space for families and businesses > Breathing space for families", + "program": "income_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "No business mileage or mileage allowance payments: the approved rates are a tax-free amount per business mile, and the search finds no mileage in the engine.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /mileage|business_travel|approved_mileage/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "policyengine_uk_development_item": true, + "upstream_item": "policyengine-uk: business mileage and approved mileage allowance payments for employees' own cars and vans", + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__parental_allowance_fathers_partners_adopters", + "title": "Parental Pay: Establish a Parental Allowance for fathers, partners and adopters in Great Britain from April 2029", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 6 (Spend, RAME); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '0', '2029-30': '-70', '2030-31': '-70', '2031-32': '-70'}", + "table_61_line": 6, + "table_61_head": "Spend", + "table_61_heading": "Breathing space for families and businesses > Breathing space for families", + "program": "parental_pay", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "A new benefit with no parameter or variable at the pin.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /parental_allowance|paternity|parental_leave/. Parameters: 0 node(s) []. Variables: 2 ['statutory_paternity_pay', 'tax_free_childcare_on_shared_parental_leave']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "policyengine_uk_development_item": true, + "upstream_item": "policyengine-uk: a Parental Allowance for fathers, partners and adopters (statutory_paternity_pay is the nearest variable)", + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230", + "nearest_variable_and_why_it_does_not_work": { + "variable": "statutory_paternity_pay", + "why_not": "an existing statutory payment; the new Parental Allowance is a separate benefit with its own rules, none of which the engine carries" + } + }, + { + "measure_key": "ab2026__uc_surplus_earnings_threshold_2500_extended", + "title": "Universal Credit: Extend the £2,500 surplus earnings threshold for one year from 1 April 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 8 (Spend, RAME); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-140', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 8, + "table_61_head": "Spend", + "table_61_heading": "Breathing space for families and businesses > Breathing space for families", + "program": "universal_credit", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "No surplus earnings rule: the search finds no parameter or variable for it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /surplus/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "policyengine_uk_development_item": true, + "upstream_item": "policyengine-uk: Universal Credit surplus earnings carried between assessment periods", + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__cash_isa_limit_held_at_20000", + "title": "Cash ISAs: Maintain the cash ISA limit at £20,000 from 6 April 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 11 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-5', '2028-29': '-20', '2029-30': '-35', '2030-31': '-50', '2031-32': '-70'}", + "table_61_line": 11, + "table_61_head": "Tax", + "table_61_heading": "Breathing space for families and businesses > Breathing space for families", + "program": "savings", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "No ISA subscription or subscription limit: the engine holds ISA balances and interest, not subscriptions, so a limit has nothing to bind.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /(^|[._])isa([._]|$)|individual_savings|subscri/. Parameters: 2 node(s) ['gov.hmrc.isa.lifetime.charge_free_age', 'gov.hmrc.isa.lifetime.withdrawal_charge']. Variables: 7 ['lifetime_isa_balance', 'has_lifetime_isa', 'stocks_and_shares_isa', 'household_lifetime_isa_balance', 'cash_isa', 'lifetime_isa_countable_capital']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "policyengine_uk_development_item": true, + "upstream_item": "policyengine-uk: ISA subscriptions, a subscription limit, and Lifetime and first-time-buyer ISA withdrawals", + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230", + "nearest_variable_and_why_it_does_not_work": { + "variable": "cash_isa", + "why_not": "a holding, not an annual subscription" + } + }, + { + "measure_key": "ab2026__founders_relief_cgt", + "title": "Founders' Relief: Introduce a new CGT relief for qualifying founders of high-growth, innovative companies from 6 April 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 26 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-260', '2028-29': '-180', '2029-30': '-260', '2030-31': '-285', '2031-32': '-300'}", + "table_61_line": 26, + "table_61_head": "Tax", + "table_61_heading": "Backing British business", + "program": "capital_gains_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "No founders' gains: the engine's relief-specific gains variable is capital_gains_badr (Business Asset Disposal Relief), and nothing identifies qualifying founders of high-growth companies.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /founder|capital_gains_badr|investors_relief/. Parameters: 0 node(s) []. Variables: 3 ['capital_gains_badr_behavioural_response', 'capital_gains_badr', 'capital_gains_badr_elasticity']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "policyengine_uk_development_item": true, + "upstream_item": "policyengine-uk: capital gains by relief (founders' and other business reliefs beyond BADR)", + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__first_time_buyer_isa", + "title": "First Time Buyer ISA: Introduce a new ISA for first-time buyers, replacing the Lifetime ISA for new savers from 6 April 2029", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 38 (Spend, RAME); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '0', '2029-30': '110', '2030-31': '110', '2031-32': '235'}", + "table_61_line": 38, + "table_61_head": "Spend", + "table_61_heading": "Hope for the future > Housing", + "program": "savings", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "No first-time-buyer ISA, and no ISA subscriptions to move into one.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /first_time|(^|[._])isa([._]|$)/. Parameters: 40 node(s) ['gov.hmrc.isa.lifetime.charge_free_age', 'gov.hmrc.isa.lifetime.withdrawal_charge', 'gov.hmrc.stamp_duty.statistics.residential.household.first_time_buyers_relief[0].threshold', 'gov.hmrc.stamp_duty.statistics.residential.household.first_time_buyers_relief[0].amount', 'gov.hmrc.stamp_duty.statistics.residential.household.first_time_buyers_relief[1].threshold', 'gov.hmrc.stamp_duty.statistics.residential.household.first_time_buyers_relief[1].amount']. Variables: 6 ['lifetime_isa_balance', 'has_lifetime_isa', 'stocks_and_shares_isa', 'household_lifetime_isa_balance', 'cash_isa', 'lifetime_isa_countable_capital']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "policyengine_uk_development_item": true, + "upstream_item": "policyengine-uk: ISA subscriptions, a subscription limit, and Lifetime and first-time-buyer ISA withdrawals", + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__lifetime_isa_withdrawal_charge_20pc_above_cap", + "title": "Lifetime ISA: Reduce the withdrawal charge from 25% to 20% for first-time buyers purchasing properties above the £450,000 price cap from 29 October 2026", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 39 (Spend, RAME); exchequer impact GBP m by FY: {'2026-27': '-15', '2027-28': '-20', '2028-29': '-20', '2029-30': '-25', '2030-31': '-25', '2031-32': '-25'}", + "table_61_line": 39, + "table_61_head": "Spend", + "table_61_heading": "Hope for the future > Housing", + "program": "savings", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "The engine's withdrawal charge (gov.hmrc.isa.lifetime.withdrawal_charge, 0.25 at the pin) is read only by lifetime_isa_countable_capital, the means-test value of a Lifetime ISA. The engine has no withdrawal or home purchase, so it cannot apply a lower charge to first-time buyers above the £450,000 cap; lowering the parameter would instead revalue every holder's countable capital.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /withdrawal_charge|first_time|home_purchase/. Parameters: 39 node(s) ['gov.hmrc.isa.lifetime.withdrawal_charge', 'gov.hmrc.stamp_duty.statistics.residential.household.first_time_buyers_relief[0].threshold', 'gov.hmrc.stamp_duty.statistics.residential.household.first_time_buyers_relief[0].amount', 'gov.hmrc.stamp_duty.statistics.residential.household.first_time_buyers_relief[1].threshold', 'gov.hmrc.stamp_duty.statistics.residential.household.first_time_buyers_relief[1].amount', 'gov.hmrc.stamp_duty.statistics.residential.household.first_time_buyers_relief[2].threshold']. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "policyengine_uk_development_item": true, + "upstream_item": "policyengine-uk: ISA subscriptions, a subscription limit, and Lifetime and first-time-buyer ISA withdrawals", + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230", + "nearest_variable_and_why_it_does_not_work": { + "variable": "lifetime_isa_countable_capital", + "why_not": "applies the charge to a holder's balance for means tests; the measure changes the charge on one kind of withdrawal" + } + }, + { + "measure_key": "ab2026__pension_death_benefits_two_year_rule_removed", + "title": "Pensions Tax: Remove the two-year rule for taxation of pension and lump sum death benefits from 28 October 2026", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 53 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '-5', '2027-28': '-15', '2028-29': '-15', '2029-30': '-15', '2030-31': '-15', '2031-32': '-15'}", + "table_61_line": 53, + "table_61_head": "Tax", + "table_61_heading": "Funding the Budget fairly and responsibly > Modernising the tax and welfare system", + "program": "income_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "The measure changes when pension and lump sum death benefits are taxed after a member dies (the two-year rule). The engine has no death event; its one death-benefit variable, pension_annual_allowance_taxable_lump_sum_death_benefits, is an input with no formula and no rule about timing after a death.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /death|bereave|lump_sum/. Parameters: 0 node(s) []. Variables: 2 ['pension_annual_allowance_taxable_lump_sum_death_benefits', 'lump_sum_income']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "policyengine_uk_development_item": true, + "upstream_item": "policyengine-uk: estates, death benefits and their income tax treatment", + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__vat_5pc_family_days_out_summer_2026", + "title": "Great British Summer Savings Scheme: Reduce VAT to 5% on family days out from 25 June 2026 to 1 September 2026", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 66 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '-310', '2027-28': '0', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 66, + "table_61_head": "Tax", + "table_61_heading": "Other policy measures including pre-announced policies > Pre-announced breathing space measures and associated funding", + "program": "vat", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "The engine's consumption variables are by spending category (e.g. recreation_consumption); none isolates admissions to family days out, the base of this reduced rate.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /recreation|attraction|admission|day_out/. Parameters: 3 node(s) ['household.consumption.carbon.emissions.recreation', 'household.consumption.carbon.intensity.recreation_consumption', 'household.consumption.total_by_category.recreation_consumption']. Variables: 1 ['recreation_consumption']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "policyengine_uk_development_item": true, + "upstream_item": "policyengine-uk: consumption detailed enough to separate admissions to attractions and days out", + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230", + "nearest_variable_and_why_it_does_not_work": { + "variable": "recreation_consumption", + "why_not": "a whole spending category; the measure covers only part of it" + } + }, + { + "measure_key": "ab2026__free_bus_travel_children_august_2026", + "title": "Bus Travel: Fund free bus travel for children aged 5-15 in England from 1 August 2026 until 31 August 2026", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 67 (Spend, RDEL); exchequer impact GBP m by FY: {'2026-27': '-145', '2027-28': '0', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 67, + "table_61_head": "Spend", + "table_61_heading": "Other policy measures including pre-announced policies > Pre-announced breathing space measures and associated funding", + "program": "bus_fares", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Free travel enters the bus model only as bus_pass_eligible, a dataset input whose documentation says it 'is not a pass-eligibility calculator'; no rule gives free travel to children aged 5 to 15, and the measure runs for one month (August 2026).", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /bus_fare|fare_allocation|bus_pass|concession/. Parameters: 19 node(s) ['gov.dft.bus.fare_allocation_weight_by_age[0].threshold', 'gov.dft.bus.fare_allocation_weight_by_age[0].amount', 'gov.dft.bus.fare_allocation_weight_by_age[1].threshold', 'gov.dft.bus.fare_allocation_weight_by_age[1].amount', 'gov.dft.bus.fare_allocation_weight_by_age[2].threshold', 'gov.dft.bus.fare_allocation_weight_by_age[2].amount']. Variables: 5 ['bus_fare_spending', 'person_bus_fare_spending', 'household_bus_fare_age_weight', 'bus_pass_eligible', 'bus_fare_spending_reported']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "policyengine_uk_development_item": true, + "upstream_item": "policyengine-uk: a free-fare rule for children by age and month within the bus fare model", + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__student_loan_interest_cap_6pc_plan_2_and_3", + "title": "Student Loans: Cap interest rates for Plan 2 and Plan 3 student loans at 6% from 1 September 2026 to 31 August 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 83 (Spend, Tax/TME); exchequer impact GBP m by FY: {'2026-27': '-230', '2027-28': '-170', '2028-29': '-10', '2029-30': '-10', '2030-31': '-5', '2031-32': '-15'}", + "table_61_line": 83, + "table_61_head": "Spend", + "table_61_heading": "Priority Public Services Funding", + "program": "student_loans", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "No interest cap: plan_2_interest_rate is RPI plus up to gov.hmrc.student_loans.interest_rates.plan_2.additional_rate (3%), and postgraduate_interest_rate (Plan 3) is RPI plus postgraduate_additional_rate (3%); neither formula has a ceiling.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /student_loan.*interest|interest_rates\\.plan_2|postgraduate_additional_rate/. Parameters: 4 node(s) ['gov.hmrc.student_loans.interest_rates.plan_1.boe_margin', 'gov.hmrc.student_loans.interest_rates.plan_2.additional_rate', 'gov.hmrc.student_loans.interest_rates.plan_2.upper_threshold', 'gov.hmrc.student_loans.interest_rates.postgraduate_additional_rate']. Variables: 1 ['student_loan_interest_rate']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "policyengine_uk_development_item": true, + "upstream_item": "policyengine-uk: a cap on Plan 2 and postgraduate (Plan 3) student loan interest", + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__company_car_tax_rates_2030_31", + "title": "Company Car Tax: Set the Benefit-in-Kind rates (appropriate percentages) for electric vehicles, hybrids and internal combustion engine vehicles for 2030-31", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 86 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '0', '2029-30': '0', '2030-31': '310', '2031-32': '345'}", + "table_61_line": 86, + "table_61_head": "Tax", + "table_61_heading": "Other measures", + "program": "income_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "No company cars or benefit-in-kind appropriate percentages: the search's only hits are council tax discount percentages.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /company_car(?!ries)|benefit_in_kind|appropriate_percentage/. Parameters: 3 node(s) ['gov.local_authorities.council_tax.discounts.appropriate_percentage.england', 'gov.local_authorities.council_tax.discounts.appropriate_percentage.scotland', 'gov.local_authorities.council_tax.discounts.appropriate_percentage.wales']. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "policyengine_uk_development_item": true, + "upstream_item": "policyengine-uk: company cars and their benefit-in-kind appropriate percentages", + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__hvcts_deferrals_discounts_exemptions", + "title": "High Value Council Tax Surcharge: Confirm eligibility for deferrals, discounts and exemptions from 1 April 2028", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 93 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '-5', '2027-28': '-5', '2028-29': '-45', '2029-30': '-45', '2030-31': '-50', '2031-32': '-50'}", + "table_61_line": 93, + "table_61_head": "Tax", + "table_61_heading": "Other measures", + "program": "council_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "The engine carries the surcharge (gov.hmrc.council_tax.high_value_surcharge, variable high_value_council_tax_surcharge) but no deferral, discount or exemption within it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /high_value|surcharge|deferr/. Parameters: 11 node(s) ['gov.hmrc.council_tax.high_value_surcharge.amount[0].threshold', 'gov.hmrc.council_tax.high_value_surcharge.amount[0].amount', 'gov.hmrc.council_tax.high_value_surcharge.amount[1].threshold', 'gov.hmrc.council_tax.high_value_surcharge.amount[1].amount', 'gov.hmrc.council_tax.high_value_surcharge.amount[2].threshold', 'gov.hmrc.council_tax.high_value_surcharge.amount[2].amount']. Variables: 2 ['early_years_school_start_deferred', 'high_value_council_tax_surcharge']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "policyengine_uk_development_item": true, + "upstream_item": "policyengine-uk: deferrals, discounts and exemptions within the high value council tax surcharge", + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__early_years_books_access", + "title": "Early Years: Fund increased access to books for children aged 0 to 5", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 9 (Spend, RDEL); exchequer impact GBP m by FY: {'2026-27': '-5', '2027-28': '-5', '2028-29': '-5', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 9, + "table_61_head": "Spend", + "table_61_heading": "Breathing space for families and businesses > Breathing space for families", + "program": "public_spending", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Public spending (books for children aged 0 to 5), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /early_years|childcare|book/. Parameters: 73 node(s) ['gov.dfe.childcare_funding_rate[0].threshold', 'gov.dfe.childcare_funding_rate[0].amount', 'gov.dfe.childcare_funding_rate[1].threshold', 'gov.dfe.childcare_funding_rate[1].amount', 'gov.dfe.childcare_funding_rate[2].threshold', 'gov.dfe.childcare_funding_rate[2].amount']. Variables: 76 ['tax_free_childcare_uses_qualifying_provider', 'childcare_expenses', 'early_years_school_start_deferred', 'early_years_term_share_before_school', 'weekly_childcare_expenses', 'WTC_childcare_element']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__food_and_heating_oil_tariff_suspensions", + "title": "Tariff Suspension Packages: Suspend tariffs on 135 food products from 21 June 2026 until 31 December 2028; 101 food products and kerosene for heating oil from 26 November 2026 until 31 December 2028; and fertiliser from 26 November 2026 until 31 December 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 10 (Tax, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '-50', '2027-28': '-75', '2028-29': '-55', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 10, + "table_61_head": "Tax", + "table_61_heading": "Breathing space for families and businesses > Breathing space for families", + "program": "indirect_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "An indirect tax or duty collected from businesses (tariffs on food products, heating oil and fertiliser), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /tariff|customs|import/. Parameters: 26 node(s) ['gov.dwp.ESA.income.capital.tariff_income.amount', 'gov.dwp.ESA.income.capital.tariff_income.step', 'gov.dwp.ESA.income.capital.tariff_income.threshold', 'gov.dwp.JSA.income.capital.tariff_income.amount', 'gov.dwp.JSA.income.capital.tariff_income.step', 'gov.dwp.JSA.income.capital.tariff_income.threshold']. Variables: 5 ['jsa_income_tariff_income', 'esa_income_tariff_income', 'income_support_tariff_income', 'housing_benefit_tariff_income', 'uc_tariff_income']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__business_rates_pubs_relief_35pc", + "title": "Business Rates: Increase business rates relief for pubs, social clubs and live music venues by 20ppts to 35% from 1 April 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 12 (Tax, Tax/TME); exchequer impact GBP m by FY: {'2026-27': '5', '2027-28': '-160', '2028-29': '-115', '2029-30': '-5', '2030-31': '-5', '2031-32': '-5'}", + "table_61_line": 12, + "table_61_head": "Tax", + "table_61_heading": "Breathing space for businesses", + "program": "business_rates", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "A business rates measure (business rates relief for pubs and venues), out of the household model's scope. The engine carries business rates only as a household incidence, business_rates ('Total incidence from exposure to business rates via corporate shareholdings'), and its business rates parameters are revenue statistics (gov.hmrc.business_rates.statistics.revenue.*): it has no properties, reliefs or thresholds for the measure to change, so its £ change would have to be supplied as an input, restating the costing.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /business_rates|non_domestic|nndr/. Parameters: 4 node(s) ['gov.hmrc.business_rates.statistics.revenue.ENGLAND', 'gov.hmrc.business_rates.statistics.revenue.NORTHERN_IRELAND', 'gov.hmrc.business_rates.statistics.revenue.SCOTLAND', 'gov.hmrc.business_rates.statistics.revenue.WALES']. Variables: 2 ['business_rates', 'baseline_business_rates']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__business_rates_sbrr_rhl_thresholds_raised", + "title": "Business Rates: Increase valuation thresholds for Small Business Rate Relief for retail, hospitality and leisure (RHL) properties from £12,000 to £18,000 for 100% relief and from £15,000 to £21,000 for tapered relief from 1 April 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 13 (Tax, Tax/TME); exchequer impact GBP m by FY: {'2026-27': '5', '2027-28': '-315', '2028-29': '-320', '2029-30': '0', '2030-31': '-5', '2031-32': '-5'}", + "table_61_line": 13, + "table_61_head": "Tax", + "table_61_heading": "Breathing space for businesses", + "program": "business_rates", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "A business rates measure (small business rate relief thresholds), out of the household model's scope. The engine carries business rates only as a household incidence, business_rates ('Total incidence from exposure to business rates via corporate shareholdings'), and its business rates parameters are revenue statistics (gov.hmrc.business_rates.statistics.revenue.*): it has no properties, reliefs or thresholds for the measure to change, so its £ change would have to be supplied as an input, restating the costing.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /business_rates|small_business|non_domestic/. Parameters: 4 node(s) ['gov.hmrc.business_rates.statistics.revenue.ENGLAND', 'gov.hmrc.business_rates.statistics.revenue.NORTHERN_IRELAND', 'gov.hmrc.business_rates.statistics.revenue.SCOTLAND', 'gov.hmrc.business_rates.statistics.revenue.WALES']. Variables: 2 ['business_rates', 'baseline_business_rates']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__business_rates_rhl_transitional_support", + "title": "Business Rates: Introduce a transitional support package, including a two-year real terms bills freeze for RHL businesses and a one-year extension of the 2023 Supporting Small Business support, part funded by extending the Transitional Relief Supplement for two years from 1 April 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 14 (Tax, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-10', '2028-29': '-230', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 14, + "table_61_head": "Tax", + "table_61_heading": "Breathing space for businesses", + "program": "business_rates", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "A business rates measure (business rates transitional support), out of the household model's scope. The engine carries business rates only as a household incidence, business_rates ('Total incidence from exposure to business rates via corporate shareholdings'), and its business rates parameters are revenue statistics (gov.hmrc.business_rates.statistics.revenue.*): it has no properties, reliefs or thresholds for the measure to change, so its £ change would have to be supplied as an input, restating the costing.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /business_rates|transitional|non_domestic/. Parameters: 4 node(s) ['gov.hmrc.business_rates.statistics.revenue.ENGLAND', 'gov.hmrc.business_rates.statistics.revenue.NORTHERN_IRELAND', 'gov.hmrc.business_rates.statistics.revenue.SCOTLAND', 'gov.hmrc.business_rates.statistics.revenue.WALES']. Variables: 2 ['business_rates', 'baseline_business_rates']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__business_tariff_suspensions", + "title": "Business Tariff Suspensions: Suspend tariffs on 72 products from 5 August 2026 and extend suspensions for 361 products in 2027-28", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 16 (Tax, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '-65', '2027-28': '-215', '2028-29': '-155', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 16, + "table_61_head": "Tax", + "table_61_heading": "Breathing space for businesses", + "program": "indirect_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "An indirect tax or duty collected from businesses (tariff suspensions on business inputs), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /tariff|customs|import/. Parameters: 26 node(s) ['gov.dwp.ESA.income.capital.tariff_income.amount', 'gov.dwp.ESA.income.capital.tariff_income.step', 'gov.dwp.ESA.income.capital.tariff_income.threshold', 'gov.dwp.JSA.income.capital.tariff_income.amount', 'gov.dwp.JSA.income.capital.tariff_income.step', 'gov.dwp.JSA.income.capital.tariff_income.threshold']. Variables: 5 ['jsa_income_tariff_income', 'esa_income_tariff_income', 'income_support_tariff_income', 'housing_benefit_tariff_income', 'uc_tariff_income']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__climate_change_levy_rates_2028", + "title": "Climate Change Levy: Freeze electricity & liquefied petroleum gas main rates, and uprate gas and solid fuels main rates by RPI from 1 April 2028", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 17 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '-25', '2029-30': '-25', '2030-31': '-25', '2031-32': '-25'}", + "table_61_line": 17, + "table_61_head": "Tax", + "table_61_heading": "Breathing space for businesses", + "program": "indirect_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "An indirect tax or duty collected from businesses (Climate Change Levy rates), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /climate_change_levy|ccl/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__red_diesel_duty_cut", + "title": "Red Diesel: Reduce duty rates by 3.7p-per-litre from 15 June 2026 to 1 July 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 18 (Tax, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '-60', '2027-28': '-30', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 18, + "table_61_head": "Tax", + "table_61_heading": "Breathing space for businesses", + "program": "indirect_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "An indirect tax or duty collected from businesses (red diesel duty), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /red_diesel|rebated|fuel_duty/. Parameters: 4 node(s) ['gov.hmrc.fuel_duty.lpg', 'gov.hmrc.fuel_duty.natural_gas', 'gov.hmrc.fuel_duty.petrol_and_diesel', 'gov.hmrc.fuel_duty.rural_fuel_duty_relief']. Variables: 2 ['in_rural_fuel_duty_relief_area', 'fuel_duty']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__alcohol_duty_small_producer_relief", + "title": "Alcohol Duty: Increase the generosity of Small Producer Relief from 1 February 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 19 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-10', '2028-29': '-10', '2029-30': '-10', '2030-31': '-10', '2031-32': '-10'}", + "table_61_line": 19, + "table_61_head": "Tax", + "table_61_heading": "Breathing space for businesses", + "program": "indirect_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "An indirect tax or duty collected from businesses (alcohol duty small producer relief), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /alcohol_duty|small_producer/. Parameters: 18 node(s) ['gov.hmrc.alcohol_duty.high_abv', 'gov.hmrc.alcohol_duty.rates.beer', 'gov.hmrc.alcohol_duty.rates.cider', 'gov.hmrc.alcohol_duty.rates.draught_cider', 'gov.hmrc.alcohol_duty.rates.draught_low', 'gov.hmrc.alcohol_duty.rates.draught_other']. Variables: 1 ['alcohol_duty']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__good_growth_funds", + "title": "Good Growth Funds: Seed funding for Mayors and local leaders in England and £1.5bn finance for Mayoral Strategic Authorities with a mechanism for subsidised lending (resource impacts of subsidy)", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 20 (Spend, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-340', '2028-29': '-30', '2029-30': '-35', '2030-31': '-55', '2031-32': '-60'}", + "table_61_line": 20, + "table_61_head": "Spend", + "table_61_heading": "Good growth in every postcode > Delivering Fiscal Devolution", + "program": "financial_transactions", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Lending, guarantees, investment or a financial transaction (Good Growth Funds for mayoral authorities): no household tax or benefit rule changes.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /growth_fund|mayor|combined_authorit/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__business_rates_retention_mayoral_authorities", + "title": "Business Rates Retention: Devolve a share of business rates income and growth to Mayoral Strategic Authorities from April 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 21 (Spend, RAME); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-10', '2028-29': '-45', '2029-30': '-75', '2030-31': '-110', '2031-32': '-140'}", + "table_61_line": 21, + "table_61_head": "Spend", + "table_61_heading": "Good growth in every postcode > Delivering Fiscal Devolution", + "program": "public_spending", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Public spending (business rates retention by mayoral authorities), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /business_rates|retention|mayor/. Parameters: 4 node(s) ['gov.hmrc.business_rates.statistics.revenue.ENGLAND', 'gov.hmrc.business_rates.statistics.revenue.NORTHERN_IRELAND', 'gov.hmrc.business_rates.statistics.revenue.SCOTLAND', 'gov.hmrc.business_rates.statistics.revenue.WALES']. Variables: 3 ['business_rates', 'baseline_business_rates', 'relative_capital_gains_retention_rate_change']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__local_apprenticeship_services", + "title": "Youth Employment and Skills: Expand local apprenticeship services to all current mayors", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 22 (Spend, RDEL); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-50', '2028-29': '-70', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 22, + "table_61_head": "Spend", + "table_61_heading": "Good growth in every postcode > Delivering Fiscal Devolution", + "program": "public_spending", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Public spending (local apprenticeship services), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /apprentice/. Parameters: 11 node(s) ['gov.hmrc.minimum_wage.apprentice', 'gov.hmrc.minimum_wage.non_apprentice[0].threshold', 'gov.hmrc.minimum_wage.non_apprentice[0].amount', 'gov.hmrc.minimum_wage.non_apprentice[1].threshold', 'gov.hmrc.minimum_wage.non_apprentice[1].amount', 'gov.hmrc.minimum_wage.non_apprentice[2].threshold']. Variables: 1 ['is_apprentice']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__regional_growth_investments", + "title": "Regional Growth: Fund priority investments and local growth & regeneration projects across the UK", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 23 (Spend, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '-20', '2027-28': '-85', '2028-29': '-45', '2029-30': '-30', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 23, + "table_61_head": "Spend", + "table_61_heading": "Good growth in every postcode > Delivering Fiscal Devolution", + "program": "public_spending", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Public spending (regional growth projects), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /regional|regeneration/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__cambridge_development_corporation", + "title": "Cambridge Development Corporation: Pilot a new government financing model for land, housing and infrastructure development from April 2027 (capital grant element)", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 24 (Spend, CDEL); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '-50', '2029-30': '-60', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 24, + "table_61_head": "Spend", + "table_61_heading": "Good growth in every postcode > Delivering Fiscal Devolution", + "program": "financial_transactions", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Lending, guarantees, investment or a financial transaction (Cambridge Development Corporation): no household tax or benefit rule changes.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /development_corporation|cambridge/. Parameters: 3 node(s) ['gov.local_authorities.council_tax.band_d.amounts.CAMBRIDGE', 'gov.local_authorities.council_tax.band_d.amounts.EAST_CAMBRIDGESHIRE', 'gov.local_authorities.council_tax.band_d.amounts.SOUTH_CAMBRIDGESHIRE']. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__strategic_authorities_capacity_funding", + "title": "Strategic Authorities: Increase funding to build capacity to deliver new powers and investment responsibilities", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 25 (Spend, RDEL); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-180', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 25, + "table_61_head": "Spend", + "table_61_heading": "Good growth in every postcode > Delivering Fiscal Devolution", + "program": "public_spending", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Public spending (strategic authorities' capacity), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /strategic_authorit|mayor/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__rd_relief_claim_notification_form_removed", + "title": "R&D Tax Reliefs: Remove the Claim Notification Form from 1 April 2027 and simplify the Additional Information Form from Autumn 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 27 (Spend, RAME); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-45', '2028-29': '-60', '2029-30': '-65', '2030-31': '-65', '2031-32': '-70'}", + "table_61_line": 27, + "table_61_head": "Spend", + "table_61_heading": "Backing British business", + "program": "business_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "A tax on businesses (R&D relief claim forms), out of the household model's scope: it changes what companies owe. The engine's corporate channel, gov.contrib.policyengine.budget.corporate_incident_tax_change ('Tax increase incident on owners of capital'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /research_and_development|r_and_d|rd_relief/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__emi2_option_limit_120000", + "title": "Enterprise Management Incentives 2 (EMI2): Increase the individual option limit to £120,000, extending the exercise period to 15 years and remove interaction with EMI option limit from 6 April 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 28 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-25', '2028-29': '-35', '2029-30': '-45', '2030-31': '-145', '2031-32': '-140'}", + "table_61_line": 28, + "table_61_head": "Tax", + "table_61_heading": "Backing British business", + "program": "business_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "A tax on businesses (Enterprise Management Incentives share options), out of the household model's scope: it changes what companies owe. The engine's corporate channel, gov.contrib.policyengine.budget.corporate_incident_tax_change ('Tax increase incident on owners of capital'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /enterprise_management|share_option|emi/. Parameters: 27 node(s) ['gov.dft.vehicle_excise_duty.band_k_max_emissions', 'gov.dft.vehicle_excise_duty.zero_emission_exempt', 'gov.dft.vehicle_excise_duty.zero_emission_expensive_car_threshold', 'gov.dft.vehicle_excise_duty.zero_emission_supplement_registration_start', 'gov.dwp.carer_premium.single', 'gov.dwp.disability_premia.disability_couple']. Variables: 16 ['car_co2_emissions', 'severe_disability_premium_before_non_dependant_test', 'severe_disability_premium', 'enhanced_disability_premium', 'pension_age_severe_disability_premium', 'receives_severe_disability_premium_qualifying_benefit']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__british_business_bank_backing_innovation", + "title": "Backing Innovation: Bring forward £1bn investment and increase financial capacity by £2bn for the British Business Bank to support additional investment in the UK (resource impacts of increased lending)", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 29 (Spend, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '5', '2028-29': '0', '2029-30': '5', '2030-31': '10', '2031-32': '25'}", + "table_61_line": 29, + "table_61_head": "Spend", + "table_61_heading": "Backing British business", + "program": "financial_transactions", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Lending, guarantees, investment or a financial transaction (British Business Bank): no household tax or benefit rule changes.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /business_bank/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__national_wealth_fund_capital", + "title": "National Wealth Fund: Invest additional capital to support growth, including £2bn for the new Energy Wealth Fund (resource impacts of increased lending)", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 30 (Spend, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '35', '2029-30': '120', '2030-31': '250', '2031-32': '470'}", + "table_61_line": 30, + "table_61_head": "Spend", + "table_61_heading": "Backing British business", + "program": "financial_transactions", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Lending, guarantees, investment or a financial transaction (National Wealth Fund): no household tax or benefit rule changes.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /wealth_fund/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__artificial_intelligence_package", + "title": "Artificial Intelligence: Strengthen the UK’s AI security, accelerate adoption across the economy, and support strategically important UK AI capabilities", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 31 (Spend); exchequer impact GBP m by FY: {'2026-27': '-35', '2027-28': '-115', '2028-29': '-185', '2029-30': '-145', '2030-31': '-5', '2031-32': '-5'}", + "table_61_line": 31, + "table_61_head": "Spend", + "table_61_heading": "Backing British business", + "program": "public_spending", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Public spending (artificial intelligence), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /artificial_intelligence|\\bai\\b/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__ev_charging_public_chargers", + "title": "EV Charging: Reduce the cost of using local, lower-powered public chargers in Great Britain", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 32 (Spend, RDEL); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-50', '2028-29': '-50', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 32, + "table_61_head": "Spend", + "table_61_heading": "Backing British business", + "program": "public_spending", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Public spending (public EV charging costs), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /charg|electric_vehicle/. Parameters: 21 node(s) ['gov.dwp.housing_benefit.ineligible_charges.meals.at_least_three_a_day.aged_16_or_over', 'gov.dwp.housing_benefit.ineligible_charges.meals.at_least_three_a_day.under_16', 'gov.dwp.housing_benefit.ineligible_charges.meals.breakfast_only', 'gov.dwp.housing_benefit.ineligible_charges.meals.fewer_than_three_a_day.aged_16_or_over', 'gov.dwp.housing_benefit.ineligible_charges.meals.fewer_than_three_a_day.under_16', 'gov.hmrc.child_benefit.opt_out_charge_share']. Variables: 6 ['water_and_sewerage_charges', 'housing_service_charges', 'child_benefit_less_tax_charge', 'high_value_council_tax_surcharge', 'income_tax_pre_charges', 'pension_annual_allowance_chargeable_amount']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__rough_sleeping_investment", + "title": "Rough Sleeping: Invest in a route off the streets across the Parliament, alongside longer-term accommodation and intensive support", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 33 (Spend, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '-175', '2027-28': '-515', '2028-29': '-580', '2029-30': '-85', '2030-31': '-90', '2031-32': '-95'}", + "table_61_line": 33, + "table_61_head": "Spend", + "table_61_heading": "Hope for the future > Housing", + "program": "public_spending", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Public spending (rough sleeping), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /rough_sleep|homeless/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__rough_sleeping_reprioritised_mhclg_budgets", + "title": "Rough Sleeping: Reprioritise existing MHCLG budgets to tackle rough sleeping", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 34 (Spend, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '120', '2027-28': '405', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 34, + "table_61_head": "Spend", + "table_61_heading": "Hope for the future > Housing", + "program": "public_spending", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Public spending (reprioritised housing budgets), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /rough_sleep|homeless/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__social_and_affordable_homes_programme", + "title": "Social and Affordable Homes Programme: Fund delivery of around 10,000 social and affordable homes", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 35 (Spend, CDEL); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-250', '2028-29': '-500', '2029-30': '-500', '2030-31': '-500', '2031-32': '0'}", + "table_61_line": 35, + "table_61_head": "Spend", + "table_61_heading": "Hope for the future > Housing", + "program": "public_spending", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Public spending (social and affordable homes), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /social_housing|affordable/. Parameters: 0 node(s) []. Variables: 1 ['in_social_housing']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__vat_zero_rate_land_for_social_housing", + "title": "Social Housing: Introduce a zero rate of VAT for land sold to registered social housing providers (relevant housing associations) from 1 April 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 36 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-15', '2028-29': '-15', '2029-30': '-15', '2030-31': '-15', '2031-32': '-15'}", + "table_61_line": 36, + "table_61_head": "Tax", + "table_61_heading": "Hope for the future > Housing", + "program": "indirect_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "An indirect tax or duty collected from businesses (VAT on land sold to social housing providers), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /vat|social_housing/. Parameters: 64 node(s) ['gov.contrib.conservatives.cb_hitc_household', 'gov.contrib.conservatives.pensioner_personal_allowance', 'gov.contrib.labour.private_school_vat', 'gov.economic_assumptions.yoy_growth.obr.private_pension_index', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.UNITED_KINGDOM', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.ENGLAND']. Variables: 13 ['private_pension_wealth', 'private_pension_income', 'private_transfer_income', 'private_school_vat', 'attends_private_school_random_draw', 'attends_private_school']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__your_first_home_equity_loan", + "title": "Your First Home: Introduce a new equity loan scheme to support around 70,000 first-time buyers from April 2027 (net income from developer fees)", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 37 (Spend, RAME); exchequer impact GBP m by FY: {'2026-27': '-15', '2027-28': '50', '2028-29': '95', '2029-30': '95', '2030-31': '95', '2031-32': '95'}", + "table_61_line": 37, + "table_61_head": "Spend", + "table_61_heading": "Hope for the future > Housing", + "program": "financial_transactions", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Lending, guarantees, investment or a financial transaction (an equity loan scheme for first-time buyers): no household tax or benefit rule changes.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /equity_loan|help_to_buy|first_time/. Parameters: 38 node(s) ['gov.hmrc.stamp_duty.statistics.residential.household.first_time_buyers_relief[0].threshold', 'gov.hmrc.stamp_duty.statistics.residential.household.first_time_buyers_relief[0].amount', 'gov.hmrc.stamp_duty.statistics.residential.household.first_time_buyers_relief[1].threshold', 'gov.hmrc.stamp_duty.statistics.residential.household.first_time_buyers_relief[1].amount', 'gov.hmrc.stamp_duty.statistics.residential.household.first_time_buyers_relief[2].threshold', 'gov.hmrc.stamp_duty.statistics.residential.household.first_time_buyers_relief[2].amount']. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__defence_investment_plan_further_savings", + "title": "Defence Investment Plan (DIP): Commitment to further savings made at announcement on 30 June 2026", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 40 (Spend, TDEL); exchequer impact GBP m by FY: {'2026-27': '-1755', '2027-28': '-1090', '2028-29': '-1010', '2029-30': '-905', '2030-31': '-940', '2031-32': '-970'}", + "table_61_line": 40, + "table_61_head": "Spend", + "table_61_heading": "Investing in our national security", + "program": "public_spending", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Public spending (defence), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /defen[cs]e/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__defence_investment_plan_capital_reallocation", + "title": "Defence Investment Plan: Reallocate capital spending to fully fund the Defence Investment Plan", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 41 (Spend, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '145', '2027-28': '2135', '2028-29': '1420', '2029-30': '1100', '2030-31': '1135', '2031-32': '1180'}", + "table_61_line": 41, + "table_61_head": "Spend", + "table_61_heading": "Investing in our national security", + "program": "public_spending", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Public spending (defence), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /defen[cs]e/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__defence_admin_savings_reinvested", + "title": "Defence: Reinvest defence and security administration savings into frontline defence and national security", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 42 (Spend, RDEL); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-25', '2028-29': '-70', '2029-30': '-95', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 42, + "table_61_head": "Spend", + "table_61_heading": "Investing in our national security", + "program": "public_spending", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Public spending (defence administration), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /defen[cs]e/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__ukraine_winter_air_defence", + "title": "Ukraine: Emergency winter support for air defence", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 43 (Spend, TDEL); exchequer impact GBP m by FY: {'2026-27': '-500', '2027-28': '0', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 43, + "table_61_head": "Spend", + "table_61_heading": "Investing in our national security", + "program": "public_spending", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Public spending (air defence for Ukraine), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /ukraine|defen[cs]e/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__uk_export_finance_defence_capacity", + "title": "UK Export Finance: Increase capacity to support UK defence exports from June 2026 (increased income from additional activity)", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 44 (Tax, RAME); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '200', '2028-29': '200', '2029-30': '200', '2030-31': '200', '2031-32': '0'}", + "table_61_line": 44, + "table_61_head": "Tax", + "table_61_heading": "Investing in our national security", + "program": "financial_transactions", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Lending, guarantees, investment or a financial transaction (UK Export Finance): no household tax or benefit rule changes.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /export_finance/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__hmrc_tax_gap_package", + "title": "HMRC: Measures to close the tax gap4", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 45 (Tax); exchequer impact GBP m by FY: {'2026-27': '10', '2027-28': '100', '2028-29': '115', '2029-30': '1980', '2030-31': '2485', '2031-32': '3160'}", + "table_61_line": 45, + "table_61_head": "Tax", + "table_61_heading": "Funding the Budget fairly and responsibly > Modernising the tax and welfare system", + "program": "compliance_and_administration", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "An enforcement, compliance or administrative change (HMRC measures to close the tax gap): the rules households face do not change.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /tax_gap|compliance|evasion/. Parameters: 1 node(s) ['gov.dcms.bbc.tv_licence.evasion_rate']. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__online_marketplace_liability_uk_businesses", + "title": "Online Marketplace Liability: Extend to UK businesses with a minimum annual sales threshold per platform of £90,000 from 1 October 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 46 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '165', '2028-29': '320', '2029-30': '270', '2030-31': '250', '2031-32': '265'}", + "table_61_line": 46, + "table_61_head": "Tax", + "table_61_heading": "Funding the Budget fairly and responsibly > Modernising the tax and welfare system", + "program": "compliance_and_administration", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "An enforcement, compliance or administrative change (online marketplace VAT liability): the rules households face do not change.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /marketplace|online_sales/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__dwp_fraud_and_error_prevention", + "title": "DWP Fraud and Error: Expand use of data and targeted controls to prevent fraud from entering the benefits system4", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 47 (Spend); exchequer impact GBP m by FY: {'2026-27': '195', '2027-28': '-125', '2028-29': '60', '2029-30': '75', '2030-31': '155', '2031-32': '240'}", + "table_61_line": 47, + "table_61_head": "Spend", + "table_61_heading": "Funding the Budget fairly and responsibly > Modernising the tax and welfare system", + "program": "compliance_and_administration", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "An enforcement, compliance or administrative change (DWP fraud and error controls): the rules households face do not change.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /fraud|error/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__benefits_removed_hospitalised_serious_offenders", + "title": "Welfare Reform: Remove benefit entitlements from serious offenders hospitalised in Great Britain from 6 April 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 48 (Spend); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '5', '2028-29': '5', '2029-30': '5', '2030-31': '5', '2031-32': '5'}", + "table_61_line": 48, + "table_61_head": "Spend", + "table_61_heading": "Funding the Budget fairly and responsibly > Modernising the tax and welfare system", + "program": "benefits", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "A change to the benefit entitlements of serious offenders detained in hospital in Great Britain: the search finds no parameter or variable for hospital detention, prison or offending, so the engine cannot identify the people it affects.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /hospital|prison|detain|offender/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__distributions_framework_modernised", + "title": "Distributions: Modernise the distributions framework from 6 April 2028", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 49 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '55', '2029-30': '800', '2030-31': '775', '2031-32': '785'}", + "table_61_line": 49, + "table_61_head": "Tax", + "table_61_heading": "Funding the Budget fairly and responsibly > Modernising the tax and welfare system", + "program": "business_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "A tax on businesses (the distributions framework), out of the household model's scope: it changes what companies owe. The engine's corporate channel, gov.contrib.policyengine.budget.corporate_incident_tax_change ('Tax increase incident on owners of capital'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /distribution|dividend/. Parameters: 20 node(s) ['gov.hmrc.income_tax.allowances.dividend_allowance', 'gov.hmrc.income_tax.rates.dividends[0].threshold', 'gov.hmrc.income_tax.rates.dividends[0].rate', 'gov.hmrc.income_tax.rates.dividends[1].threshold', 'gov.hmrc.income_tax.rates.dividends[1].rate', 'gov.hmrc.income_tax.rates.dividends[2].threshold']. Variables: 6 ['dividend_income', 'received_allowances_dividend_income', 'dividend_income_tax', 'trade_loss_relief_against_savings_and_dividends', 'taxed_dividend_income', 'taxable_dividend_income']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__advance_corporation_tax_abolished", + "title": "Advance Corporation Tax (ACT): Abolish Advance Corporation Tax from 1 April 2031", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 50 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '0', '2029-30': '0', '2030-31': '5', '2031-32': '15'}", + "table_61_line": 50, + "table_61_head": "Tax", + "table_61_heading": "Funding the Budget fairly and responsibly > Modernising the tax and welfare system", + "program": "business_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "A tax on businesses (Advance Corporation Tax), out of the household model's scope: it changes what companies owe. The engine's corporate channel, gov.contrib.policyengine.budget.corporate_incident_tax_change ('Tax increase incident on owners of capital'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /corporation_tax|advance_corporation/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__vat_reverse_charge_overseas_services_restricted", + "title": "Cross-Border Services: Restrict the reverse charge on supplies of services from overseas to VAT-registered businesses from 1 April 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 51 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '70', '2028-29': '75', '2029-30': '75', '2030-31': '75', '2031-32': '75'}", + "table_61_line": 51, + "table_61_head": "Tax", + "table_61_heading": "Funding the Budget fairly and responsibly > Modernising the tax and welfare system", + "program": "indirect_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "An indirect tax or duty collected from businesses (the VAT reverse charge on imported services), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /reverse_charge|vat/. Parameters: 64 node(s) ['gov.contrib.conservatives.cb_hitc_household', 'gov.contrib.conservatives.pensioner_personal_allowance', 'gov.contrib.labour.private_school_vat', 'gov.economic_assumptions.yoy_growth.obr.private_pension_index', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.UNITED_KINGDOM', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.ENGLAND']. Variables: 12 ['private_pension_wealth', 'private_pension_income', 'private_transfer_income', 'private_school_vat', 'attends_private_school_random_draw', 'attends_private_school']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__foreign_sovereign_investments_taxation", + "title": "Modernising Taxation of Foreign Sovereign Investments: Introduce a new statutory regime for the direct taxation of foreign sovereigns from 1 January 2029", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 54 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '110', '2028-29': '495', '2029-30': '690', '2030-31': '675', '2031-32': '655'}", + "table_61_line": 54, + "table_61_head": "Tax", + "table_61_heading": "Funding the Budget fairly and responsibly > Modernising the tax and welfare system", + "program": "business_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "A tax on businesses (the taxation of foreign sovereign investors), out of the household model's scope: it changes what companies owe. The engine's corporate channel, gov.contrib.policyengine.budget.corporate_incident_tax_change ('Tax increase incident on owners of capital'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /sovereign|foreign/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__bank_surcharge_6pc_temporary", + "title": "Bank Surcharge: Temporarily increase the Bank Corporation Tax Surcharge rate from 3% to 6%, to fund breathing space measures for households, for 15 months from 1 January 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 55 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '305', '2027-28': '950', '2028-29': '220', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 55, + "table_61_head": "Tax", + "table_61_heading": "Making fair choices on revenue raising", + "program": "business_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "A tax on businesses (the Bank Corporation Tax Surcharge), out of the household model's scope: it changes what companies owe. The engine's corporate channel, gov.contrib.policyengine.budget.corporate_incident_tax_change ('Tax increase incident on owners of capital'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /bank_surcharge|corporation_tax|bank_levy/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__machine_games_duty_higher_stake_25pc", + "title": "Machine Games Duty: Increase the rate from 20% to 25% for higher stake machines from 1 April 2027 and move duty liability to the machine supplier from 1 April 2028", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 56 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '135', '2028-29': '150', '2029-30': '145', '2030-31': '145', '2031-32': '145'}", + "table_61_line": 56, + "table_61_head": "Tax", + "table_61_heading": "Making fair choices on revenue raising", + "program": "indirect_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "An indirect tax or duty collected from businesses (Machine Games Duty), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /gambling|machine_games|gaming/. Parameters: 23 node(s) ['gov.hmrc.gambling_duty.bingo.rate', 'gov.hmrc.gambling_duty.gaming.normal_six_month_bands[0].threshold', 'gov.hmrc.gambling_duty.gaming.normal_six_month_bands[0].rate', 'gov.hmrc.gambling_duty.gaming.normal_six_month_bands[1].threshold', 'gov.hmrc.gambling_duty.gaming.normal_six_month_bands[1].rate', 'gov.hmrc.gambling_duty.gaming.normal_six_month_bands[2].threshold']. Variables: 8 ['remote_gaming_net_losses', 'machine_games_higher_rate_net_takings', 'machine_games_lower_rate_net_takings', 'machine_games_standard_rate_net_takings', 'gaming_duty', 'remote_gaming_duty']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__business_rates_vape_shops_and_agcs_reliefs_removed", + "title": "Business Rates: Remove reliefs for vape shops and Adult Gaming Centres from 1 April 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 57 (Tax, Tax/TME); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '10', '2028-29': '10', '2029-30': '10', '2030-31': '10', '2031-32': '10'}", + "table_61_line": 57, + "table_61_head": "Tax", + "table_61_heading": "Making fair choices on revenue raising", + "program": "business_rates", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "A business rates measure (business rates reliefs for vape shops and adult gaming centres), out of the household model's scope. The engine carries business rates only as a household incidence, business_rates ('Total incidence from exposure to business rates via corporate shareholdings'), and its business rates parameters are revenue statistics (gov.hmrc.business_rates.statistics.revenue.*): it has no properties, reliefs or thresholds for the measure to change, so its £ change would have to be supplied as an input, restating the costing.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /business_rates|vape/. Parameters: 4 node(s) ['gov.hmrc.business_rates.statistics.revenue.ENGLAND', 'gov.hmrc.business_rates.statistics.revenue.NORTHERN_IRELAND', 'gov.hmrc.business_rates.statistics.revenue.SCOTLAND', 'gov.hmrc.business_rates.statistics.revenue.WALES']. Variables: 2 ['business_rates', 'baseline_business_rates']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__writing_down_allowance_12pc_and_fya_50pc", + "title": "Writing-down Allowances: Reduce the main rate writing-down allowance to 12% from April 2027 and increase the main rate first-year allowance to 50% from 1 January 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 58 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '125', '2027-28': '765', '2028-29': '1040', '2029-30': '1075', '2030-31': '1100', '2031-32': '1120'}", + "table_61_line": 58, + "table_61_head": "Tax", + "table_61_heading": "Making fair choices on revenue raising", + "program": "business_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "A tax on businesses (capital allowances), out of the household model's scope: it changes what companies owe. The engine's corporate channel, gov.contrib.policyengine.budget.corporate_incident_tax_change ('Tax increase incident on owners of capital'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /capital_allowance|writing_down|first_year_allowance/. Parameters: 0 node(s) []. Variables: 1 ['capital_allowances']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__private_hire_operators_vat_full_fare", + "title": "Private Hire Vehicle Operators: Deem larger operators with annual fare turnover over £50 million liable for VAT on the full value of their fares from 2 January 2028", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 59 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '95', '2028-29': '385', '2029-30': '395', '2030-31': '410', '2031-32': '425'}", + "table_61_line": 59, + "table_61_head": "Tax", + "table_61_heading": "Making fair choices on revenue raising", + "program": "indirect_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "An indirect tax or duty collected from businesses (VAT on private hire operators' fares), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /private_hire|taxi|vat/. Parameters: 64 node(s) ['gov.contrib.conservatives.cb_hitc_household', 'gov.contrib.conservatives.pensioner_personal_allowance', 'gov.contrib.labour.private_school_vat', 'gov.economic_assumptions.yoy_growth.obr.private_pension_index', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.UNITED_KINGDOM', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.ENGLAND']. Variables: 12 ['private_pension_wealth', 'private_pension_income', 'private_transfer_income', 'private_school_vat', 'attends_private_school_random_draw', 'attends_private_school']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__civil_service_admin_budgets_cut_20pc", + "title": "Future Civil Service: Reduce administration budgets by 20% by 2029-30 compared to 16% reduction in Spending Review 2025 plans", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 63 (Spend, RDEL); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '185', '2028-29': '405', '2029-30': '595', '2030-31': '615', '2031-32': '635'}", + "table_61_line": 63, + "table_61_head": "Spend", + "table_61_heading": "Transforming the Civil Service", + "program": "public_spending", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Public spending (civil service administration budgets), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /civil_service|administration/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__civil_service_ai_fund", + "title": "Future Civil Service: Reinvest part of the administration budget savings in a new AI Fund to reduce administration burdens and improve outcomes", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 64 (Spend, RDEL); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-25', '2028-29': '-25', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 64, + "table_61_head": "Spend", + "table_61_heading": "Transforming the Civil Service", + "program": "public_spending", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Public spending (a civil service AI fund), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /civil_service/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__hgv_ved_1_for_12_months", + "title": "Heavy Goods Vehicles: Reduce Vehicle Excise Duty to £1 for 12 months from 1 July 2026", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 65 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '-135', '2027-28': '-50', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 65, + "table_61_head": "Tax", + "table_61_heading": "Other policy measures including pre-announced policies > Pre-announced breathing space measures and associated funding", + "program": "indirect_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "An indirect tax or duty collected from businesses (Vehicle Excise Duty on heavy goods vehicles), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /heavy_goods|hgv|vehicle_excise/. Parameters: 93 node(s) ['gov.dft.vehicle_excise_duty.alternative_fuel_discount', 'gov.dft.vehicle_excise_duty.band_k_max_emissions', 'gov.dft.vehicle_excise_duty.band_k_registration_cutoff', 'gov.dft.vehicle_excise_duty.engine_size_threshold', 'gov.dft.vehicle_excise_duty.expensive_car_supplement', 'gov.dft.vehicle_excise_duty.expensive_car_threshold']. Variables: 2 ['car_vehicle_excise_duty', 'vehicle_excise_duty']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__foreign_permanent_establishment_exemption_mandatory", + "title": "Corporation Tax: Make the foreign permanent establishment exemption mandatory, preventing their losses from reducing UK corporation tax receipts for accounting periods beginning on or after 1 January 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 68 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '120', '2027-28': '480', '2028-29': '495', '2029-30': '450', '2030-31': '460', '2031-32': '470'}", + "table_61_line": 68, + "table_61_head": "Tax", + "table_61_heading": "Other policy measures including pre-announced policies > Pre-announced breathing space measures and associated funding", + "program": "business_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "A tax on businesses (the foreign permanent establishment exemption), out of the household model's scope: it changes what companies owe. The engine's corporate channel, gov.contrib.policyengine.budget.corporate_incident_tax_change ('Tax increase incident on owners of capital'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /permanent_establishment|corporation_tax/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__electricity_generator_levy_55pc", + "title": "Electricity Generator Levy: Increase the rate from 45% to 55% until 31 March 2028", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 69 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '60', '2027-28': '35', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 69, + "table_61_head": "Tax", + "table_61_heading": "Other policy measures including pre-announced policies > Pre-announced breathing space measures and associated funding", + "program": "business_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "A tax on businesses (the Electricity Generator Levy), out of the household model's scope: it changes what companies owe. The engine's corporate channel, gov.contrib.policyengine.budget.corporate_incident_tax_change ('Tax increase incident on owners of capital'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /generator_levy|electricity_generator/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__british_industrial_competitiveness_scheme", + "title": "Industrial Electricity: Fund the British Industrial Competitiveness Scheme starting from 1 April 2027, including a backdated payment for eligible businesses", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 70 (Spend, Tax/TME); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-1235', '2028-29': '-65', '2029-30': '-135', '2030-31': '-420', '2031-32': '-465'}", + "table_61_line": 70, + "table_61_head": "Spend", + "table_61_heading": "Pre-announced growth measures", + "program": "public_spending", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Public spending (the British Industrial Competitiveness Scheme), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /industrial|competitiveness/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__carbon_price_support_removed", + "title": "Carbon Price Support (CPS): Remove the CPS rates to reduce energy bills from 1 April 2028", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 71 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '-165', '2029-30': '-45', '2030-31': '-15', '2031-32': '-10'}", + "table_61_line": 71, + "table_61_head": "Tax", + "table_61_heading": "Pre-announced growth measures", + "program": "indirect_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "An indirect tax or duty collected from businesses (Carbon Price Support rates), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /carbon_price|carbon/. Parameters: 37 node(s) ['gov.contrib.ubi_center.carbon_tax.consumer_incidence', 'gov.contrib.ubi_center.carbon_tax.rate', 'gov.economic_assumptions.yoy_growth.carbon_intensity', 'gov.economic_assumptions.indices.carbon_intensity', 'household.consumption.carbon.consumption', 'household.consumption.carbon.emissions.food_and_non_alcoholic_beverages']. Variables: 2 ['carbon_tax', 'carbon_consumption']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__youth_guarantee_and_apprenticeships_spring_2026", + "title": "Spring 2026 Youth Employment and Skills Package: Expand Youth Guarantee and apprenticeship support", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 72 (Spend, TME); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-290', '2028-29': '-375', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 72, + "table_61_head": "Spend", + "table_61_heading": "Pre-announced growth measures", + "program": "public_spending", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Public spending (the Youth Guarantee and apprenticeships), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /youth_guarantee|apprentice/. Parameters: 11 node(s) ['gov.hmrc.minimum_wage.apprentice', 'gov.hmrc.minimum_wage.non_apprentice[0].threshold', 'gov.hmrc.minimum_wage.non_apprentice[0].amount', 'gov.hmrc.minimum_wage.non_apprentice[1].threshold', 'gov.hmrc.minimum_wage.non_apprentice[1].amount', 'gov.hmrc.minimum_wage.non_apprentice[2].threshold']. Variables: 1 ['is_apprentice']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__union_learning_fund", + "title": "Skills: Re-establish the Union Learning Fund in England", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 73 (Spend, RDEL); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-20', '2028-29': '-20', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 73, + "table_61_head": "Spend", + "table_61_heading": "Pre-announced growth measures", + "program": "public_spending", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Public spending (the Union Learning Fund), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /union_learning|skills/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__investment_for_growth", + "title": "Investment for Growth: Support strategic technologies, local growth and key industries", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 74 (Spend); exchequer impact GBP m by FY: {'2026-27': '-155', '2027-28': '-460', '2028-29': '-530', '2029-30': '-330', '2030-31': '-355', '2031-32': '-535'}", + "table_61_line": 74, + "table_61_head": "Spend", + "table_61_heading": "Pre-announced growth measures", + "program": "financial_transactions", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Lending, guarantees, investment or a financial transaction (investment for growth): no household tax or benefit rule changes.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /investment|growth/. Parameters: 45 node(s) ['gov.economic_assumptions.yoy_growth.obr.rpi', 'gov.economic_assumptions.yoy_growth.obr.average_earnings', 'gov.economic_assumptions.yoy_growth.obr.consumer_price_index', 'gov.economic_assumptions.yoy_growth.obr.road_fuel_volume', 'gov.economic_assumptions.yoy_growth.obr.petrol_spending_litre_proxy', 'gov.economic_assumptions.yoy_growth.obr.diesel_spending_litre_proxy']. Variables: 3 ['unit_and_investment_trusts', 'other_investment_income', 'charitable_investment_gifts']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__growth_guarantee_scheme_expansion", + "title": "British Business Bank: Expand the Growth Guarantee Scheme to support £6.5 billion of additional SME lending across this Parliament", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 75 (Spend, TME); exchequer impact GBP m by FY: {'2026-27': '-105', '2027-28': '-145', '2028-29': '-180', '2029-30': '-170', '2030-31': '-160', '2031-32': '-155'}", + "table_61_line": 75, + "table_61_head": "Spend", + "table_61_heading": "Pre-announced growth measures", + "program": "financial_transactions", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Lending, guarantees, investment or a financial transaction (the Growth Guarantee Scheme): no household tax or benefit rule changes.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /guarantee/. Parameters: 25 node(s) ['gov.dwp.pension_credit.guarantee_credit.additions', 'gov.dwp.pension_credit.guarantee_credit.carer.addition', 'gov.dwp.pension_credit.guarantee_credit.child.addition', 'gov.dwp.pension_credit.guarantee_credit.child.disability.addition', 'gov.dwp.pension_credit.guarantee_credit.child.disability.eligibility', 'gov.dwp.pension_credit.guarantee_credit.child.disability.severe.addition']. Variables: 9 ['is_guarantee_credit_eligible', 'guarantee_credit', 'in_receipt_of_guarantee_credit', 'standard_minimum_guarantee', 'minimum_guarantee', 'child_minimum_guarantee_addition']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__leeds_city_fund", + "title": "Leeds City Fund: Establish the Leeds City Fund with 100% retention of business rates growth", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 76 (Tax, Tax/TME); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '-25', '2029-30': '-25', '2030-31': '-25', '2031-32': '-30'}", + "table_61_line": 76, + "table_61_head": "Tax", + "table_61_heading": "Pre-announced growth measures", + "program": "business_rates", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "A business rates measure (business rates retention in Leeds), out of the household model's scope. The engine carries business rates only as a household incidence, business_rates ('Total incidence from exposure to business rates via corporate shareholdings'), and its business rates parameters are revenue statistics (gov.hmrc.business_rates.statistics.revenue.*): it has no properties, reliefs or thresholds for the measure to change, so its £ change would have to be supplied as an input, restating the costing.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /business_rates|retention/. Parameters: 4 node(s) ['gov.hmrc.business_rates.statistics.revenue.ENGLAND', 'gov.hmrc.business_rates.statistics.revenue.NORTHERN_IRELAND', 'gov.hmrc.business_rates.statistics.revenue.SCOTLAND', 'gov.hmrc.business_rates.statistics.revenue.WALES']. Variables: 3 ['business_rates', 'baseline_business_rates', 'relative_capital_gains_retention_rate_change']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__uk_india_ceta_and_double_contributions_convention", + "title": "UK-India: Bring into force the Comprehensive Economic and Trade Agreement and Double Contributions Convention from 15 July 2026", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 77 (Tax, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '-175', '2027-28': '-285', '2028-29': '-330', '2029-30': '-360', '2030-31': '-385', '2031-32': '-400'}", + "table_61_line": 77, + "table_61_head": "Tax", + "table_61_heading": "Pre-announced growth measures", + "program": "indirect_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "An indirect tax or duty collected from businesses (tariffs and contributions under the UK-India agreement), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /tariff|customs|double_contribution/. Parameters: 26 node(s) ['gov.dwp.ESA.income.capital.tariff_income.amount', 'gov.dwp.ESA.income.capital.tariff_income.step', 'gov.dwp.ESA.income.capital.tariff_income.threshold', 'gov.dwp.JSA.income.capital.tariff_income.amount', 'gov.dwp.JSA.income.capital.tariff_income.step', 'gov.dwp.JSA.income.capital.tariff_income.threshold']. Variables: 5 ['jsa_income_tariff_income', 'esa_income_tariff_income', 'income_support_tariff_income', 'housing_benefit_tariff_income', 'uc_tariff_income']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__steel_tariff_and_quotas", + "title": "Steel Tariff: Change tariff rate and quotas on imported steel from 1 July 2026", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 78 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '85', '2027-28': '110', '2028-29': '120', '2029-30': '120', '2030-31': '120', '2031-32': '115'}", + "table_61_line": 78, + "table_61_head": "Tax", + "table_61_heading": "Pre-announced growth measures", + "program": "indirect_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "An indirect tax or duty collected from businesses (steel tariffs and quotas), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /tariff|steel/. Parameters: 26 node(s) ['gov.dwp.ESA.income.capital.tariff_income.amount', 'gov.dwp.ESA.income.capital.tariff_income.step', 'gov.dwp.ESA.income.capital.tariff_income.threshold', 'gov.dwp.JSA.income.capital.tariff_income.amount', 'gov.dwp.JSA.income.capital.tariff_income.step', 'gov.dwp.JSA.income.capital.tariff_income.threshold']. Variables: 5 ['jsa_income_tariff_income', 'esa_income_tariff_income', 'income_support_tariff_income', 'housing_benefit_tariff_income', 'uc_tariff_income']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__securities_transfer_tax", + "title": "Stamp Taxes on Shares: Replace Stamp Duty and Stamp Duty Reserve Tax with the Securities Transfer Tax from late 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 79 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-15', '2028-29': '-30', '2029-30': '-30', '2030-31': '-30', '2031-32': '-30'}", + "table_61_line": 79, + "table_61_head": "Tax", + "table_61_heading": "Pre-announced growth measures", + "program": "business_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "A tax on businesses (stamp taxes on shares), out of the household model's scope: it changes what companies owe. The engine's corporate channel, gov.contrib.policyengine.budget.corporate_incident_tax_change ('Tax increase incident on owners of capital'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /stamp_duty|stamp_tax|securities/. Parameters: 115 node(s) ['gov.hmrc.stamp_duty.abolish', 'gov.hmrc.stamp_duty.non_residential.purchase[0].threshold', 'gov.hmrc.stamp_duty.non_residential.purchase[0].rate', 'gov.hmrc.stamp_duty.non_residential.purchase[1].threshold', 'gov.hmrc.stamp_duty.non_residential.purchase[1].rate', 'gov.hmrc.stamp_duty.non_residential.purchase[2].threshold']. Variables: 1 ['stamp_duty_land_tax']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__business_rates_ev_charging_relief", + "title": "Business Rates: Introduce 100% relief for eligible Electric Vehicle Charging Points (EVCPs) and electric vehicle-only forecourts for ten years from 1 April 2026", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 80 (Tax, Tax/TME); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-15', '2028-29': '-5', '2029-30': '-10', '2030-31': '-15', '2031-32': '-15'}", + "table_61_line": 80, + "table_61_head": "Tax", + "table_61_heading": "Pre-announced growth measures", + "program": "business_rates", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "A business rates measure (business rates relief for EV charge points), out of the household model's scope. The engine carries business rates only as a household incidence, business_rates ('Total incidence from exposure to business rates via corporate shareholdings'), and its business rates parameters are revenue statistics (gov.hmrc.business_rates.statistics.revenue.*): it has no properties, reliefs or thresholds for the measure to change, so its £ change would have to be supplied as an input, restating the costing.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /business_rates|charg/. Parameters: 25 node(s) ['gov.dwp.housing_benefit.ineligible_charges.meals.at_least_three_a_day.aged_16_or_over', 'gov.dwp.housing_benefit.ineligible_charges.meals.at_least_three_a_day.under_16', 'gov.dwp.housing_benefit.ineligible_charges.meals.breakfast_only', 'gov.dwp.housing_benefit.ineligible_charges.meals.fewer_than_three_a_day.aged_16_or_over', 'gov.dwp.housing_benefit.ineligible_charges.meals.fewer_than_three_a_day.under_16', 'gov.hmrc.business_rates.statistics.revenue.ENGLAND']. Variables: 8 ['water_and_sewerage_charges', 'housing_service_charges', 'child_benefit_less_tax_charge', 'business_rates', 'high_value_council_tax_surcharge', 'baseline_business_rates']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__misogyny_prevention_programmes", + "title": "Tackling Misogyny: Invest in grassroots misogyny prevention programmes", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 81 (Spend, RDEL); exchequer impact GBP m by FY: {'2026-27': '-5', '2027-28': '-5', '2028-29': '-5', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 81, + "table_61_head": "Spend", + "table_61_heading": "Priority Public Services Funding", + "program": "public_spending", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Public spending (misogyny prevention programmes), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /misogyny|violence/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__infected_blood_compensation_additional_report", + "title": "Infected Blood Compensation Scheme: Implement the Inquiry's Additional Report recommendations from 24 July 2026", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 82 (Spend, CAME); exchequer impact GBP m by FY: {'2026-27': '-360', '2027-28': '-355', '2028-29': '-265', '2029-30': '-90', '2030-31': '-75', '2031-32': '0'}", + "table_61_line": 82, + "table_61_head": "Spend", + "table_61_heading": "Priority Public Services Funding", + "program": "public_spending", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Public spending (infected blood compensation), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /infected_blood|compensation/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__waste_crime_landfill_tax_clearances", + "title": "Waste Crime: Fund Landfill Tax costs for six priority illegal waste site clearances until 31 March 2029 and a three-year pilot to fund Local Authorities' Landfill Tax liabilities arising from illegal waste site clearances from 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 84 (Spend, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-5', '2028-29': '-5', '2029-30': '-5', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 84, + "table_61_head": "Spend", + "table_61_heading": "Priority Public Services Funding", + "program": "public_spending", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Public spending (Landfill Tax costs of illegal waste clearances), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /landfill|waste/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__further_priorities_nhs_and_vawg", + "title": "Further Priorities: Fund key priorities, including pressures facing the NHS and delivery of the Violence against Women and Girls Strategy", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 85 (Spend); exchequer impact GBP m by FY: {'2026-27': '-355', '2027-28': '-175', '2028-29': '-140', '2029-30': '-5', '2030-31': '0', '2031-32': '-15'}", + "table_61_line": 85, + "table_61_head": "Spend", + "table_61_heading": "Priority Public Services Funding", + "program": "public_spending", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Public spending (the NHS and the Violence against Women and Girls Strategy), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /nhs|violence/. Parameters: 1 node(s) ['gov.contrib.policyengine.budget.nhs']. Variables: 7 ['nhs_budget_change', 'nhs_spending', 'nhs_outpatient_spending', 'nhs_a_and_e_spending', 'nhs_admitted_patient_spending', 'travel_grant_receives_means_tested_nhs_bursary']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__supported_housing_licensing", + "title": "Supported Housing: Introduce a licensing regime to enforce national standards for supported housing from 2027-28", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 87 (Spend, RAME); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-55', '2028-29': '-70', '2029-30': '-30', '2030-31': '-30', '2031-32': '-30'}", + "table_61_line": 87, + "table_61_head": "Spend", + "table_61_heading": "Other measures", + "program": "regulation", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "A regulatory change (licensing of supported housing) with no tax or benefit rule for households.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /supported_housing|exempt_accommodation/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__vat_zero_rate_donated_medicines", + "title": "Donated Medicines: Introduce a zero rate of VAT for the donation of medicines under qualifying circumstances from 23 June 2026", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 88 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '-15', '2027-28': '-20', '2028-29': '-20', '2029-30': '-25', '2030-31': '-25', '2031-32': '-25'}", + "table_61_line": 88, + "table_61_head": "Tax", + "table_61_heading": "Other measures", + "program": "indirect_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "An indirect tax or duty collected from businesses (VAT on donated medicines), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /vat|medicine/. Parameters: 64 node(s) ['gov.contrib.conservatives.cb_hitc_household', 'gov.contrib.conservatives.pensioner_personal_allowance', 'gov.contrib.labour.private_school_vat', 'gov.economic_assumptions.yoy_growth.obr.private_pension_index', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.UNITED_KINGDOM', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.ENGLAND']. Variables: 12 ['private_pension_wealth', 'private_pension_income', 'private_transfer_income', 'private_school_vat', 'attends_private_school_random_draw', 'attends_private_school']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__vat_medical_deputies_staffing", + "title": "Temporary Staffing: Amend Group 7 of Schedule 9 to the VAT Act 1994 to restore the tax treatment of medical deputies to its previously intended position", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 89 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '85', '2028-29': '95', '2029-30': '100', '2030-31': '100', '2031-32': '105'}", + "table_61_line": 89, + "table_61_head": "Tax", + "table_61_heading": "Other measures", + "program": "indirect_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "An indirect tax or duty collected from businesses (VAT on medical deputies' staffing supplies), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /vat|staffing/. Parameters: 64 node(s) ['gov.contrib.conservatives.cb_hitc_household', 'gov.contrib.conservatives.pensioner_personal_allowance', 'gov.contrib.labour.private_school_vat', 'gov.economic_assumptions.yoy_growth.obr.private_pension_index', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.UNITED_KINGDOM', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.ENGLAND']. Variables: 12 ['private_pension_wealth', 'private_pension_income', 'private_transfer_income', 'private_school_vat', 'attends_private_school_random_draw', 'attends_private_school']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__landfill_communities_fund_credit_cap", + "title": "Landfill Communities Fund: Increase the Landfill Tax credit cap from 1 April 2027", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 90 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-10', '2028-29': '-15', '2029-30': '-15', '2030-31': '-15', '2031-32': '-15'}", + "table_61_line": 90, + "table_61_head": "Tax", + "table_61_heading": "Other measures", + "program": "indirect_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "An indirect tax or duty collected from businesses (the Landfill Communities Fund credit cap), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /landfill/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__franchised_higher_education_registration", + "title": "Higher Education: Introduce registration requirements for franchise providers with 300 or more students from 1 September 2028, to strengthen quality and oversight of franchised provision", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 91 (Spend, Tax/TME); exchequer impact GBP m by FY: {'2026-27': '30', '2027-28': '70', '2028-29': '200', '2029-30': '220', '2030-31': '200', '2031-32': '225'}", + "table_61_line": 91, + "table_61_head": "Spend", + "table_61_heading": "Other measures", + "program": "regulation", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "A regulatory change (registration of franchised higher education providers) with no tax or benefit rule for households.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /higher_education|franchise|student/. Parameters: 13 node(s) ['gov.dfe.disabled_students_allowance.maximum', 'gov.dfe.maintenance_loans.independent_student_age', 'gov.hmrc.student_loans.interest_rates.plan_1.boe_margin', 'gov.hmrc.student_loans.interest_rates.plan_2.additional_rate', 'gov.hmrc.student_loans.interest_rates.plan_2.upper_threshold', 'gov.hmrc.student_loans.interest_rates.postgraduate_additional_rate']. Variables: 22 ['student_payments', 'student_loans', 'uc_is_student_with_no_work_related_requirements', 'is_dependent_child_for_student_support', 'maintenance_loan_in_higher_education', 'adult_dependants_grant_adult_receives_student_finance']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__high_needs_local_authority_spending", + "title": "Local Authority Spending: Impact of adjustments to high needs budget on local authority spending", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 92 (Spend, TME); exchequer impact GBP m by FY: {'2026-27': '105', '2027-28': '140', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "table_61_line": 92, + "table_61_head": "Spend", + "table_61_heading": "Other measures", + "program": "public_spending", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Public spending (local authority high needs budgets), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /high_needs|special_educational|send/. Parameters: 2 node(s) ['gov.dfe.childcare_grant.eligible_child.max_age.special_educational_needs', 'gov.local_authorities.council_tax.band_d.amounts.ROSSENDALE']. Variables: 1 ['childcare_grant_child_has_special_educational_needs']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__benefits_in_kind_payrolling_delay", + "title": "Benefits in Kind (BiKs): Delay mandatory payrolling of company cars, vans, fuel and medical benefits from 6 April 2026 to 6 April 2027, with remaining BiKs implemented from 6 April 2028", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 94 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '-30', '2027-28': '-5', '2028-29': '0', '2029-30': '0', '2030-31': '-5', '2031-32': '-5'}", + "table_61_line": 94, + "table_61_head": "Tax", + "table_61_heading": "Other measures", + "program": "compliance_and_administration", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "An enforcement, compliance or administrative change (the timing of mandatory payrolling of benefits in kind): the rules households face do not change.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /benefit_in_kind|payroll/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__low_value_imports_customs_relief_removed", + "title": "Low Value Imports: Remove the £135 customs duty relief and introduce new customs arrangements from October 2028", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 95 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '215', '2029-30': '0', '2030-31': '-10', '2031-32': '-15'}", + "table_61_line": 95, + "table_61_head": "Tax", + "table_61_heading": "Other measures", + "program": "indirect_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "An indirect tax or duty collected from businesses (customs duty relief on low value imports), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /customs|low_value|import/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__ets_energy_from_waste_delay", + "title": "Emissions Trading Scheme (ETS): Delay the inclusion of energy from waste in the ETS", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 96 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '-125', '2029-30': '-490', '2030-31': '-450', '2031-32': '-440'}", + "table_61_line": 96, + "table_61_head": "Tax", + "table_61_heading": "Other measures", + "program": "indirect_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "An indirect tax or duty collected from businesses (the Emissions Trading Scheme's coverage of energy from waste), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /emissions_trading|\\bets\\b|carbon/. Parameters: 37 node(s) ['gov.contrib.ubi_center.carbon_tax.consumer_incidence', 'gov.contrib.ubi_center.carbon_tax.rate', 'gov.economic_assumptions.yoy_growth.carbon_intensity', 'gov.economic_assumptions.indices.carbon_intensity', 'household.consumption.carbon.consumption', 'household.consumption.carbon.emissions.food_and_non_alcoholic_beverages']. Variables: 2 ['carbon_tax', 'carbon_consumption']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__gas_distribution_revenue_recognition_spread", + "title": "Corporation Tax: Spread the one-off impact of revenue recognition changes for qualifying gas distribution activities over five years for accounting periods ending on or after 31 December 2026", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 97 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '-160', '2027-28': '40', '2028-29': '40', '2029-30': '40', '2030-31': '40', '2031-32': '0'}", + "table_61_line": 97, + "table_61_head": "Tax", + "table_61_heading": "Other measures", + "program": "business_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "A tax on businesses (corporation tax on gas distribution revenue), out of the household model's scope: it changes what companies owe. The engine's corporate channel, gov.contrib.policyengine.budget.corporate_incident_tax_change ('Tax increase incident on owners of capital'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /corporation_tax|gas_distribution/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__us_llc_double_taxation_relief", + "title": "International Investors: Relieve double taxation for UK investors in US Limited Liability Companies and similar entities from early 2028", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 98 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '-5', '2029-30': '-10', '2030-31': '-5', '2031-32': '-5'}", + "table_61_line": 98, + "table_61_head": "Tax", + "table_61_heading": "Other measures", + "program": "business_tax", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "A tax on businesses (double taxation relief for investors in US LLCs), out of the household model's scope: it changes what companies owe. The engine's corporate channel, gov.contrib.policyengine.budget.corporate_incident_tax_change ('Tax increase incident on owners of capital'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /double_taxation|foreign_income|llc/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__sovereign_ai_financial_transactions", + "title": "SovAI: Additional funding for Sovereign AI to invest in and support the UK's promising AI companies", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.2 line 4 (financial transactions, no Table 6.1 line); cash impact GBP m by FY: {'2026-27': '-25', '2027-28': '-45', '2028-29': '-45', '2029-30': '-45', '2030-31': '0', '2031-32': '0'}", + "table_62_line": 4, + "program": "financial_transactions", + "computability": "not_expressible", + "pe_reform_delta": null, + "why": "Lending, guarantees, investment or a financial transaction (Sovereign AI): no household tax or benefit rule changes.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /sovereign|artificial_intelligence/. Parameters: 0 node(s) []. Variables: 0 []. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "out_of_model_scope": true, + "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" + }, + { + "measure_key": "ab2026__package_total_policy_decisions", + "title": "Table 6.1 total policy decisions", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 row 'Total policy decisions'; GBP m by FY: {'2026-27': '-7595', '2027-28': '-2475', '2028-29': '730', '2029-30': '5190', '2030-31': '6135', '2031-32': '7350'}", + "program": "package", + "construction": "package_of_registry_measures", + "package_of": [ + "ab2026__household_energy_payment_100_bands_a_to_d", + "ab2026__electricity_vat_zero_rate_to_july_2027", + "ab2026__fuel_duty_5p_cut_extended_and_rpi_rise_deferred", + "ab2026__bus_fare_cap_2_england_outside_london_2027", + "ab2026__mileage_allowance_payments_55p", + "ab2026__parental_allowance_fathers_partners_adopters", + "ab2026__carers_allowance_earnings_taper", + "ab2026__uc_surplus_earnings_threshold_2500_extended", + "ab2026__early_years_books_access", + "ab2026__food_and_heating_oil_tariff_suspensions", + "ab2026__cash_isa_limit_held_at_20000", + "ab2026__business_rates_pubs_relief_35pc", + "ab2026__business_rates_sbrr_rhl_thresholds_raised", + "ab2026__business_rates_rhl_transitional_support", + "ab2026__trading_allowance_3000", + "ab2026__business_tariff_suspensions", + "ab2026__climate_change_levy_rates_2028", + "ab2026__red_diesel_duty_cut", + "ab2026__alcohol_duty_small_producer_relief", + "ab2026__good_growth_funds", + "ab2026__business_rates_retention_mayoral_authorities", + "ab2026__local_apprenticeship_services", + "ab2026__regional_growth_investments", + "ab2026__cambridge_development_corporation", + "ab2026__strategic_authorities_capacity_funding", + "ab2026__founders_relief_cgt", + "ab2026__rd_relief_claim_notification_form_removed", + "ab2026__emi2_option_limit_120000", + "ab2026__british_business_bank_backing_innovation", + "ab2026__national_wealth_fund_capital", + "ab2026__artificial_intelligence_package", + "ab2026__ev_charging_public_chargers", + "ab2026__rough_sleeping_investment", + "ab2026__rough_sleeping_reprioritised_mhclg_budgets", + "ab2026__social_and_affordable_homes_programme", + "ab2026__vat_zero_rate_land_for_social_housing", + "ab2026__your_first_home_equity_loan", + "ab2026__first_time_buyer_isa", + "ab2026__lifetime_isa_withdrawal_charge_20pc_above_cap", + "ab2026__defence_investment_plan_further_savings", + "ab2026__defence_investment_plan_capital_reallocation", + "ab2026__defence_admin_savings_reinvested", + "ab2026__ukraine_winter_air_defence", + "ab2026__uk_export_finance_defence_capacity", + "ab2026__hmrc_tax_gap_package", + "ab2026__online_marketplace_liability_uk_businesses", + "ab2026__dwp_fraud_and_error_prevention", + "ab2026__benefits_removed_hospitalised_serious_offenders", + "ab2026__distributions_framework_modernised", + "ab2026__advance_corporation_tax_abolished", + "ab2026__vat_reverse_charge_overseas_services_restricted", + "ab2026__pension_tax_allowance", + "ab2026__pension_death_benefits_two_year_rule_removed", + "ab2026__foreign_sovereign_investments_taxation", + "ab2026__bank_surcharge_6pc_temporary", + "ab2026__machine_games_duty_higher_stake_25pc", + "ab2026__business_rates_vape_shops_and_agcs_reliefs_removed", + "ab2026__writing_down_allowance_12pc_and_fya_50pc", + "ab2026__private_hire_operators_vat_full_fare", + "ab2026__badr_lifetime_limit_500000", + "ab2026__property_income_rates_plus_2pp", + "ab2026__savings_income_rates_plus_2pp", + "ab2026__civil_service_admin_budgets_cut_20pc", + "ab2026__civil_service_ai_fund", + "ab2026__hgv_ved_1_for_12_months", + "ab2026__vat_5pc_family_days_out_summer_2026", + "ab2026__free_bus_travel_children_august_2026", + "ab2026__foreign_permanent_establishment_exemption_mandatory", + "ab2026__electricity_generator_levy_55pc", + "ab2026__british_industrial_competitiveness_scheme", + "ab2026__carbon_price_support_removed", + "ab2026__youth_guarantee_and_apprenticeships_spring_2026", + "ab2026__union_learning_fund", + "ab2026__investment_for_growth", + "ab2026__growth_guarantee_scheme_expansion", + "ab2026__leeds_city_fund", + "ab2026__uk_india_ceta_and_double_contributions_convention", + "ab2026__steel_tariff_and_quotas", + "ab2026__securities_transfer_tax", + "ab2026__business_rates_ev_charging_relief", + "ab2026__misogyny_prevention_programmes", + "ab2026__infected_blood_compensation_additional_report", + "ab2026__student_loan_interest_cap_6pc_plan_2_and_3", + "ab2026__waste_crime_landfill_tax_clearances", + "ab2026__further_priorities_nhs_and_vawg", + "ab2026__company_car_tax_rates_2030_31", + "ab2026__supported_housing_licensing", + "ab2026__vat_zero_rate_donated_medicines", + "ab2026__vat_medical_deputies_staffing", + "ab2026__landfill_communities_fund_credit_cap", + "ab2026__franchised_higher_education_registration", + "ab2026__high_needs_local_authority_spending", + "ab2026__hvcts_deferrals_discounts_exemptions", + "ab2026__benefits_in_kind_payrolling_delay", + "ab2026__low_value_imports_customs_relief_removed", + "ab2026__ets_energy_from_waste_delay", + "ab2026__gas_distribution_revenue_recognition_spread", + "ab2026__us_llc_double_taxation_relief" + ], + "pe_reform_delta": null, + "computability": "partial", + "missing": "89 of 98 legs are not expressible (see each leg's verdict); a PolicyEngine number covers only the 9 expressible legs: ['ab2026__household_energy_payment_100_bands_a_to_d', 'ab2026__electricity_vat_zero_rate_to_july_2027', 'ab2026__fuel_duty_5p_cut_extended_and_rpi_rise_deferred', 'ab2026__carers_allowance_earnings_taper', 'ab2026__trading_allowance_3000', 'ab2026__pension_tax_allowance', 'ab2026__badr_lifetime_limit_500000', 'ab2026__property_income_rates_plus_2pp', 'ab2026__savings_income_rates_plus_2pp']", + "expressible_legs": [ + "ab2026__household_energy_payment_100_bands_a_to_d", + "ab2026__electricity_vat_zero_rate_to_july_2027", + "ab2026__fuel_duty_5p_cut_extended_and_rpi_rise_deferred", + "ab2026__carers_allowance_earnings_taper", + "ab2026__trading_allowance_3000", + "ab2026__pension_tax_allowance", + "ab2026__badr_lifetime_limit_500000", + "ab2026__property_income_rates_plus_2pp", + "ab2026__savings_income_rates_plus_2pp" + ] + }, + { + "measure_key": "ab2026__package_total_tax_policy_decisions", + "title": "Table 6.1 total tax policy decisions", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 row 'Total tax policy decisions'; GBP m by FY: {'2026-27': '-1645', '2027-28': '695', '2028-29': '2865', '2029-30': '5550', '2030-31': '6495', '2031-32': '6970'}", + "program": "package", + "construction": "package_of_registry_measures", + "package_of": [ + "ab2026__electricity_vat_zero_rate_to_july_2027", + "ab2026__fuel_duty_5p_cut_extended_and_rpi_rise_deferred", + "ab2026__mileage_allowance_payments_55p", + "ab2026__food_and_heating_oil_tariff_suspensions", + "ab2026__cash_isa_limit_held_at_20000", + "ab2026__business_rates_pubs_relief_35pc", + "ab2026__business_rates_sbrr_rhl_thresholds_raised", + "ab2026__business_rates_rhl_transitional_support", + "ab2026__trading_allowance_3000", + "ab2026__business_tariff_suspensions", + "ab2026__climate_change_levy_rates_2028", + "ab2026__red_diesel_duty_cut", + "ab2026__alcohol_duty_small_producer_relief", + "ab2026__founders_relief_cgt", + "ab2026__emi2_option_limit_120000", + "ab2026__vat_zero_rate_land_for_social_housing", + "ab2026__uk_export_finance_defence_capacity", + "ab2026__hmrc_tax_gap_package", + "ab2026__online_marketplace_liability_uk_businesses", + "ab2026__distributions_framework_modernised", + "ab2026__advance_corporation_tax_abolished", + "ab2026__vat_reverse_charge_overseas_services_restricted", + "ab2026__pension_tax_allowance", + "ab2026__pension_death_benefits_two_year_rule_removed", + "ab2026__foreign_sovereign_investments_taxation", + "ab2026__bank_surcharge_6pc_temporary", + "ab2026__machine_games_duty_higher_stake_25pc", + "ab2026__business_rates_vape_shops_and_agcs_reliefs_removed", + "ab2026__writing_down_allowance_12pc_and_fya_50pc", + "ab2026__private_hire_operators_vat_full_fare", + "ab2026__badr_lifetime_limit_500000", + "ab2026__property_income_rates_plus_2pp", + "ab2026__savings_income_rates_plus_2pp", + "ab2026__hgv_ved_1_for_12_months", + "ab2026__vat_5pc_family_days_out_summer_2026", + "ab2026__foreign_permanent_establishment_exemption_mandatory", + "ab2026__electricity_generator_levy_55pc", + "ab2026__carbon_price_support_removed", + "ab2026__leeds_city_fund", + "ab2026__uk_india_ceta_and_double_contributions_convention", + "ab2026__steel_tariff_and_quotas", + "ab2026__securities_transfer_tax", + "ab2026__business_rates_ev_charging_relief", + "ab2026__company_car_tax_rates_2030_31", + "ab2026__vat_zero_rate_donated_medicines", + "ab2026__vat_medical_deputies_staffing", + "ab2026__landfill_communities_fund_credit_cap", + "ab2026__hvcts_deferrals_discounts_exemptions", + "ab2026__benefits_in_kind_payrolling_delay", + "ab2026__low_value_imports_customs_relief_removed", + "ab2026__ets_energy_from_waste_delay", + "ab2026__gas_distribution_revenue_recognition_spread", + "ab2026__us_llc_double_taxation_relief" + ], + "pe_reform_delta": null, + "computability": "partial", + "missing": "46 of 53 legs are not expressible (see each leg's verdict); a PolicyEngine number covers only the 7 expressible legs: ['ab2026__electricity_vat_zero_rate_to_july_2027', 'ab2026__fuel_duty_5p_cut_extended_and_rpi_rise_deferred', 'ab2026__trading_allowance_3000', 'ab2026__pension_tax_allowance', 'ab2026__badr_lifetime_limit_500000', 'ab2026__property_income_rates_plus_2pp', 'ab2026__savings_income_rates_plus_2pp']", + "expressible_legs": [ + "ab2026__electricity_vat_zero_rate_to_july_2027", + "ab2026__fuel_duty_5p_cut_extended_and_rpi_rise_deferred", + "ab2026__trading_allowance_3000", + "ab2026__pension_tax_allowance", + "ab2026__badr_lifetime_limit_500000", + "ab2026__property_income_rates_plus_2pp", + "ab2026__savings_income_rates_plus_2pp" + ] + }, + { + "measure_key": "ab2026__package_total_spending_policy_decisions", + "title": "Table 6.1 total spending policy decisions", + "reported_status": "announced", + "source_note": "HM Treasury, Budget 2026 Table 6.1 row 'Total spending policy decisions'; GBP m by FY: {'2026-27': '-5950', '2027-28': '-3170', '2028-29': '-2135', '2029-30': '-355', '2030-31': '-360', '2031-32': '380'}", + "program": "package", + "construction": "package_of_registry_measures", + "package_of": [ + "ab2026__household_energy_payment_100_bands_a_to_d", + "ab2026__bus_fare_cap_2_england_outside_london_2027", + "ab2026__parental_allowance_fathers_partners_adopters", + "ab2026__carers_allowance_earnings_taper", + "ab2026__uc_surplus_earnings_threshold_2500_extended", + "ab2026__early_years_books_access", + "ab2026__good_growth_funds", + "ab2026__business_rates_retention_mayoral_authorities", + "ab2026__local_apprenticeship_services", + "ab2026__regional_growth_investments", + "ab2026__cambridge_development_corporation", + "ab2026__strategic_authorities_capacity_funding", + "ab2026__rd_relief_claim_notification_form_removed", + "ab2026__british_business_bank_backing_innovation", + "ab2026__national_wealth_fund_capital", + "ab2026__artificial_intelligence_package", + "ab2026__ev_charging_public_chargers", + "ab2026__rough_sleeping_investment", + "ab2026__rough_sleeping_reprioritised_mhclg_budgets", + "ab2026__social_and_affordable_homes_programme", + "ab2026__your_first_home_equity_loan", + "ab2026__first_time_buyer_isa", + "ab2026__lifetime_isa_withdrawal_charge_20pc_above_cap", + "ab2026__defence_investment_plan_further_savings", + "ab2026__defence_investment_plan_capital_reallocation", + "ab2026__defence_admin_savings_reinvested", + "ab2026__ukraine_winter_air_defence", + "ab2026__dwp_fraud_and_error_prevention", + "ab2026__benefits_removed_hospitalised_serious_offenders", + "ab2026__civil_service_admin_budgets_cut_20pc", + "ab2026__civil_service_ai_fund", + "ab2026__free_bus_travel_children_august_2026", + "ab2026__british_industrial_competitiveness_scheme", + "ab2026__youth_guarantee_and_apprenticeships_spring_2026", + "ab2026__union_learning_fund", + "ab2026__investment_for_growth", + "ab2026__growth_guarantee_scheme_expansion", + "ab2026__misogyny_prevention_programmes", + "ab2026__infected_blood_compensation_additional_report", + "ab2026__student_loan_interest_cap_6pc_plan_2_and_3", + "ab2026__waste_crime_landfill_tax_clearances", + "ab2026__further_priorities_nhs_and_vawg", + "ab2026__supported_housing_licensing", + "ab2026__franchised_higher_education_registration", + "ab2026__high_needs_local_authority_spending" + ], + "pe_reform_delta": null, + "computability": "partial", + "missing": "43 of 45 legs are not expressible (see each leg's verdict); a PolicyEngine number covers only the 2 expressible legs: ['ab2026__household_energy_payment_100_bands_a_to_d', 'ab2026__carers_allowance_earnings_taper']", + "expressible_legs": [ + "ab2026__household_energy_payment_100_bands_a_to_d", + "ab2026__carers_allowance_earnings_taper" + ] + }, + { + "measure_key": "ab2026__package_household_energy_vat_and_fuel_duty", + "title": "The Household Energy Payment, zero VAT on domestic electricity and the fuel duty cut, together", + "reported_status": "announced", + "source_note": "A composition of Table 6.1 lines that producers score together; see `package_of`.", + "program": "package", + "construction": "package_of_registry_measures", + "package_of": [ + "ab2026__household_energy_payment_100_bands_a_to_d", + "ab2026__electricity_vat_zero_rate_to_july_2027", + "ab2026__fuel_duty_5p_cut_extended_and_rpi_rise_deferred" + ], + "pe_reform_delta": null, + "computability": "expressible", + "engine_baseline_2026": {}, + "note": "Every leg is expressible; a counterpart is the legs' runs combined, stated as constructed." + }, + { + "measure_key": "ab2026__package_policyengine_seven", + "title": "The seven household measures PolicyEngine modelled on Budget day, together", + "reported_status": "announced", + "source_note": "A composition of Table 6.1 lines that producers score together; see `package_of`.", + "program": "package", + "construction": "package_of_registry_measures", + "package_of": [ + "ab2026__household_energy_payment_100_bands_a_to_d", + "ab2026__electricity_vat_zero_rate_to_july_2027", + "ab2026__fuel_duty_5p_cut_extended_and_rpi_rise_deferred", + "ab2026__carers_allowance_earnings_taper", + "ab2026__trading_allowance_3000", + "ab2026__pension_tax_allowance", + "ab2026__property_income_rates_plus_2pp" + ], + "pe_reform_delta": null, + "computability": "expressible", + "engine_baseline_2026": {}, + "pe_modifier": { + "module": "src/uk_budget_data/reforms.py", + "function": "AUTUMN_BUDGET_2026_COMBINED_COMPONENTS", + "arguments": {}, + "replaced_variables": [], + "reads_variables": [], + "parameter_legs": [], + "receipt": { + "repo": "PolicyEngine/uk-autumn-budget-dashboard-2026", + "commit": "1eca3182d6983b6375f401a3565ac56c25fa401e", + "path": "public/data/artifacts/combined--f91fe847275730cb/4faed20984703c7d/receipt.json", + "receipt_file_sha256": "a00d0614eeed1c5cfaa63353d08a604a433bf881329bfd2d7b60c1681c243e67", + "reform_sha256": "09498c7b550716444725ef211725c894bc0efa13bd6c855efb8bdadc93b0e834", + "reform_binding": "code_revision", + "generated_at": "2026-10-28T17:54:34.931664+00:00", + "policyengine": "6.11.1", + "certified_for_model_version": "2.151.0", + "dataset_revision": "microcosm-uk-2024-25-national-20261025T091414Z-a6299504", + "dataset_sha256": "b96a8cd625b645a409c063b1de5158532708a2f0d3c3897a794a8fb0bbd5064e", + "basis": "static; OBR October 2026 EFO growth rates applied over policyengine-uk's March 2026 rates (receipt dashboard.basis)" + } + }, + "note": "Run as one simulation of all seven measures (the combined run), not as a sum of the seven runs." + } + ] +} diff --git a/pipeline/ab2026_measure_keys.py b/pipeline/ab2026_measure_keys.py new file mode 100644 index 000000000..ad842d218 --- /dev/null +++ b/pipeline/ab2026_measure_keys.py @@ -0,0 +1,203 @@ +"""Autumn Budget 2026 measure keys, one per HM Treasury Table 6.1 line. + +The key space every AB2026 claim carries in conditions.measure_key (#230). +Keys are written out here rather than derived from titles, so a reworded +title in a later edition of the table cannot move a key. Both the Table 6.1 +stager (pipeline/stage_ab2026_hmt_tables.py) and the registry builder +(pipeline/build_ab2026_registry.py) import this module, so a staged row and +its registry entry cannot disagree about a key. +""" + +PREFIX = "ab2026__" + +# Table 6.1 "Budget 2026 policy decisions (£ million)", lines 1-98. +TABLE_61 = { + 1: "household_energy_payment_100_bands_a_to_d", + 2: "electricity_vat_zero_rate_to_july_2027", + 3: "fuel_duty_5p_cut_extended_and_rpi_rise_deferred", + 4: "bus_fare_cap_2_england_outside_london_2027", + 5: "mileage_allowance_payments_55p", + 6: "parental_allowance_fathers_partners_adopters", + 7: "carers_allowance_earnings_taper", + 8: "uc_surplus_earnings_threshold_2500_extended", + 9: "early_years_books_access", + 10: "food_and_heating_oil_tariff_suspensions", + 11: "cash_isa_limit_held_at_20000", + 12: "business_rates_pubs_relief_35pc", + 13: "business_rates_sbrr_rhl_thresholds_raised", + 14: "business_rates_rhl_transitional_support", + 15: "trading_allowance_3000", + 16: "business_tariff_suspensions", + 17: "climate_change_levy_rates_2028", + 18: "red_diesel_duty_cut", + 19: "alcohol_duty_small_producer_relief", + 20: "good_growth_funds", + 21: "business_rates_retention_mayoral_authorities", + 22: "local_apprenticeship_services", + 23: "regional_growth_investments", + 24: "cambridge_development_corporation", + 25: "strategic_authorities_capacity_funding", + 26: "founders_relief_cgt", + 27: "rd_relief_claim_notification_form_removed", + 28: "emi2_option_limit_120000", + 29: "british_business_bank_backing_innovation", + 30: "national_wealth_fund_capital", + 31: "artificial_intelligence_package", + 32: "ev_charging_public_chargers", + 33: "rough_sleeping_investment", + 34: "rough_sleeping_reprioritised_mhclg_budgets", + 35: "social_and_affordable_homes_programme", + 36: "vat_zero_rate_land_for_social_housing", + 37: "your_first_home_equity_loan", + 38: "first_time_buyer_isa", + 39: "lifetime_isa_withdrawal_charge_20pc_above_cap", + 40: "defence_investment_plan_further_savings", + 41: "defence_investment_plan_capital_reallocation", + 42: "defence_admin_savings_reinvested", + 43: "ukraine_winter_air_defence", + 44: "uk_export_finance_defence_capacity", + 45: "hmrc_tax_gap_package", + 46: "online_marketplace_liability_uk_businesses", + 47: "dwp_fraud_and_error_prevention", + 48: "benefits_removed_hospitalised_serious_offenders", + 49: "distributions_framework_modernised", + 50: "advance_corporation_tax_abolished", + 51: "vat_reverse_charge_overseas_services_restricted", + 52: "pension_tax_allowance", + 53: "pension_death_benefits_two_year_rule_removed", + 54: "foreign_sovereign_investments_taxation", + 55: "bank_surcharge_6pc_temporary", + 56: "machine_games_duty_higher_stake_25pc", + 57: "business_rates_vape_shops_and_agcs_reliefs_removed", + 58: "writing_down_allowance_12pc_and_fya_50pc", + 59: "private_hire_operators_vat_full_fare", + 60: "badr_lifetime_limit_500000", + 61: "property_income_rates_plus_2pp", + 62: "savings_income_rates_plus_2pp", + 63: "civil_service_admin_budgets_cut_20pc", + 64: "civil_service_ai_fund", + 65: "hgv_ved_1_for_12_months", + 66: "vat_5pc_family_days_out_summer_2026", + 67: "free_bus_travel_children_august_2026", + 68: "foreign_permanent_establishment_exemption_mandatory", + 69: "electricity_generator_levy_55pc", + 70: "british_industrial_competitiveness_scheme", + 71: "carbon_price_support_removed", + 72: "youth_guarantee_and_apprenticeships_spring_2026", + 73: "union_learning_fund", + 74: "investment_for_growth", + 75: "growth_guarantee_scheme_expansion", + 76: "leeds_city_fund", + 77: "uk_india_ceta_and_double_contributions_convention", + 78: "steel_tariff_and_quotas", + 79: "securities_transfer_tax", + 80: "business_rates_ev_charging_relief", + 81: "misogyny_prevention_programmes", + 82: "infected_blood_compensation_additional_report", + 83: "student_loan_interest_cap_6pc_plan_2_and_3", + 84: "waste_crime_landfill_tax_clearances", + 85: "further_priorities_nhs_and_vawg", + 86: "company_car_tax_rates_2030_31", + 87: "supported_housing_licensing", + 88: "vat_zero_rate_donated_medicines", + 89: "vat_medical_deputies_staffing", + 90: "landfill_communities_fund_credit_cap", + 91: "franchised_higher_education_registration", + 92: "high_needs_local_authority_spending", + 93: "hvcts_deferrals_discounts_exemptions", + 94: "benefits_in_kind_payrolling_delay", + 95: "low_value_imports_customs_relief_removed", + 96: "ets_energy_from_waste_delay", + 97: "gas_distribution_revenue_recognition_spread", + 98: "us_llc_double_taxation_relief", +} + +# Table 6.2 "Budget 2026 Financial Transactions (cash impact, £ million)". +# Each line is the financial-transactions leg of a Table 6.1 measure, except +# line 4 (Sovereign AI), which has no Table 6.1 line and so its own key. +TABLE_62 = { + 1: TABLE_61[20], + 2: TABLE_61[30], + 3: TABLE_61[29], + 4: "sovereign_ai_financial_transactions", + 5: TABLE_61[37], + 6: TABLE_61[37], + 7: TABLE_61[74], + 8: TABLE_61[24], + 9: TABLE_61[47], +} + +# Table 6.4 "Measures announced at Spring Statement 2026 or earlier that will +# take effect from 29 October 2026 or later": recostings of earlier events' +# measures, keyed to those events' registries (full keys, not slugs). +TABLE_64 = { + 1: ("autumn_budget_2025", "ab2025__cbam_indirect_emissions_removed"), + 2: ("autumn_budget_2025", "ab2025__student_finance_lle_launch_and_loan_outlay"), + 3: ("autumn_budget_2025", "ab2025__ppf_and_fas_pre_1997_inflation_protection"), + 4: ( + "autumn_budget_2025", + "ab2025__fya_zero_emission_cars_and_charge_points_extended", + ), + 5: ("autumn_budget_2025", "ab2025__student_loans_plan2_threshold_freeze"), + 6: ("autumn_budget_2025", "ab2025__property_income_separate_rates"), + 7: ("autumn_budget_2025", "ab2025__apd_higher_rate_extended_to_private_jets"), + 8: ("autumn_budget_2025", "ab2025__db_pension_scheme_surplus_extraction"), + 9: ("autumn_budget_2025", "ab2025__ets_expansion_to_international_maritime"), + 10: ("autumn_budget_2025", "ab2025__sdil_threshold_cut_and_milk_based_drinks"), + 11: ("autumn_budget_2025", "ab2025__high_value_council_tax_surcharge"), + 12: ("autumn_budget_2025", "ab2025__eved_mileage_supplement_electric_and_phev"), + 13: ("autumn_budget_2025", "ab2025__low_value_imports_customs_reform"), + 14: ("autumn_budget_2025", "ab2025__salary_sacrifice_pension_nics_cap_2000"), + 15: ( + "autumn_budget_2024", + "autumn_budget_2024__company_car_tax_set_appropriate_percentages_for_electric_vehicles_hybrids_and_internal_combustion_engine_vehic_f0345c3d", + ), + 16: ( + "autumn_budget_2024", + "autumn_budget_2024__inheritance_tax_include_unused_pension_funds_and_death_benefits_payable_from_a_pension_in_the_value_of_estates_215d16bf", + ), + 17: ( + "spring_budget_2024", + "spring_budget_2024__carbon_border_adjustment_mechanism_cbam_introduce_from_1_january_2027_46125f17", + ), +} + +# Packages: compositions of Table 6.1 lines that HM Treasury totals or that +# outside producers score as one thing. A producer's package row joins here +# when its scope is exactly these lines; any other scope is a proposal. +PACKAGES = { + "package_total_policy_decisions": { + "title": "Table 6.1 total policy decisions", + "lines": sorted(TABLE_61), + "table_61_row": "Total policy decisions", + }, + "package_total_tax_policy_decisions": { + "title": "Table 6.1 total tax policy decisions", + "lines": None, # the table's own Tax/Spend head split + "table_61_row": "Total tax policy decisions", + }, + "package_total_spending_policy_decisions": { + "title": "Table 6.1 total spending policy decisions", + "lines": None, + "table_61_row": "Total spending policy decisions", + }, + "package_household_energy_vat_and_fuel_duty": { + "title": "The Household Energy Payment, zero VAT on domestic electricity and the fuel duty cut, together", + "lines": [1, 2, 3], + "table_61_row": None, + }, + "package_policyengine_seven": { + "title": "The seven household measures PolicyEngine modelled on Budget day, together", + "lines": [1, 2, 3, 7, 15, 52, 61], + "table_61_row": None, + }, +} + + +def key(line): + """The full measure key of a Table 6.1 line.""" + return PREFIX + TABLE_61[line] + + +def package_key(slug): + return PREFIX + slug diff --git a/pipeline/build_ab2026_registry.py b/pipeline/build_ab2026_registry.py new file mode 100644 index 000000000..74c816631 --- /dev/null +++ b/pipeline/build_ab2026_registry.py @@ -0,0 +1,1170 @@ +"""Build data/uk/ab2026_measures.json (#230) from the pinned engine. + +The Autumn Budget 2026 measure registry is generated, not typed, as the +AB2025 one is (pipeline/build_ab2025_registry.py): titles and exchequer +impacts come from the staged Table 6.1 rows, every `name_search` is computed +over the full parameter tree and variable list of the pinned engine, and every +`engine_baseline_2026` value is what that engine returns. + +The pin is policyengine-uk 2.151.0, the engine the UK Autumn Budget dashboard +ran on Budget day (policyengine.py 6.11.1, whose certificate records +certified_for_model_version 2.151.0 on every published receipt). It is not +the certified populace pair the AB2025 registry follows: AB2026's +PolicyEngine counterparts are the dashboard's own published runs, so a +computability verdict has to be true of the engine those runs used. + +Inputs: + - an importable policyengine-uk == 2.151.0 (refuses any other version) + - sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/claims_staged.jsonl.gz + - pipeline/ab2026_measure_keys.py + +Usage: + PYTHONPATH=. .venv-pe2151/bin/python pipeline/build_ab2026_registry.py + PYTHONPATH=. .venv-pe2151/bin/python pipeline/build_ab2026_registry.py --check + +Verdicts are authored below. Every mechanism a verdict names was read on +2026-10-29 from the engine at the pin or from the dashboard's source at +PolicyEngine/uk-autumn-budget-dashboard-2026@1eca3182. +""" + +import gzip +import json +import re +import sys +from pathlib import Path + +try: + from pipeline import ab2026_measure_keys as K + from pipeline.uk_engine_registry import engine_dumps +except ModuleNotFoundError: # direct script invocation without PYTHONPATH + import ab2026_measure_keys as K + from uk_engine_registry import engine_dumps + +REPO = Path(__file__).resolve().parent.parent +PIN = "2.151.0" +OUT = REPO / "data/uk/ab2026_measures.json" +KEYS_MD = REPO / "sources/harvest-uk-ab2026-2026-10-29/REGISTRY_KEYS.md" +STAGED = ( + REPO / "sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/claims_staged.jsonl.gz" +) + +ISSUE = "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" +BUS_GAP = "https://github.com/PolicyEngine/policyengine-scorecard/issues/106" +DASHBOARD_REPO = "PolicyEngine/uk-autumn-budget-dashboard-2026" +DASHBOARD_MAIN = "1eca3182d6983b6375f401a3565ac56c25fa401e" +POLICY_YEARS = range(2027, 2032) # the dashboard scores 2026-27 to 2031-32 + +RUN_IDENTITY = { + "policyengine": "6.11.1", + "certified_for_model_version": "2.151.0", + "dataset_revision": "microcosm-uk-2024-25-national-20261025T091414Z-a6299504", + "dataset_sha256": "b96a8cd625b645a409c063b1de5158532708a2f0d3c3897a794a8fb0bbd5064e", + "basis": "static; OBR October 2026 EFO growth rates applied over policyengine-uk's March 2026 rates (receipt dashboard.basis)", +} + + +def _t61(): + lines, totals = {}, {} + with gzip.open(STAGED, "rt") as fh: + for raw in fh: + r = json.loads(raw) + c = r["conditions"] + if c["table"] == "6.1" and c["row_type"] == "measure": + e = lines.setdefault( + int(c["table_line"]), + { + "title": r["reform_hint"], + "head": c["head"], + "fiscal_head": c.get("fiscal_head"), + "heading": c["heading"], + "values": {}, + }, + ) + e["values"][c["fy"]] = r["value_raw"] + elif c["table"] == "6.1" and c["row_type"] == "total": + totals.setdefault(r["reform_hint"], {})[c["fy"]] = r["value_raw"] + elif c["table"] == "6.2" and c["table_line"] == "4": + e = lines.setdefault("t62_4", {"title": r["reform_hint"], "values": {}}) + e["values"][c["fy"]] = r["value_raw"] + return lines, totals + + +P, V = engine_dumps(pin=PIN) +T61, TOTALS = _t61() + + +def ns(pattern): + rx = re.compile(pattern, re.I) + ph = [p for p in P if rx.search(p) and not p.startswith("baseline.")] + vh = [v for v in V if rx.search(v)] + return ( + f"Searched BOTH the parameter tree and the VARIABLE list for /{pattern}/. " + f"Parameters: {len(ph)} node(s) {ph[:6]}. Variables: {len(vh)} {vh[:6]}. " + "Both are searched because a parameter-only search misses a variable " + f"(#106). Re-run at policyengine-uk {PIN}." + ) + + +UPSTREAM = { + "mileage": "policyengine-uk: business mileage and approved mileage allowance payments for employees' own cars and vans", + "parental_allowance": "policyengine-uk: a Parental Allowance for fathers, partners and adopters (statutory_paternity_pay is the nearest variable)", + "uc_surplus": "policyengine-uk: Universal Credit surplus earnings carried between assessment periods", + "isa": "policyengine-uk: ISA subscriptions, a subscription limit, and Lifetime and first-time-buyer ISA withdrawals", + "cgt_reliefs": "policyengine-uk: capital gains by relief (founders' and other business reliefs beyond BADR)", + "estates": "policyengine-uk: estates, death benefits and their income tax treatment", + "vat_categories": "policyengine-uk: consumption detailed enough to separate admissions to attractions and days out", + "bus_age_fares": "policyengine-uk: a free-fare rule for children by age and month within the bus fare model", + "student_loan_interest_cap": "policyengine-uk: a cap on Plan 2 and postgraduate (Plan 3) student loan interest", + "bik": "policyengine-uk: company cars and their benefit-in-kind appropriate percentages", + "hvcts_reliefs": "policyengine-uk: deferrals, discounts and exemptions within the high value council tax surcharge", +} + +# The engine's exogenous incidence levers (descriptions read at the pin from +# parameters/gov/contrib/policyengine/budget/*.yaml). They spread a £ amount +# supplied as an input; none computes a measure's yield. +LEVERS = { + "consumer": "gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers')", + "corporate": "gov.contrib.policyengine.budget.corporate_incident_tax_change ('Tax increase incident on owners of capital')", + "spending": "gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset')", +} + +measures = [] + + +def entry(line, program, computability, **kw): + t = T61[line] + m = { + "measure_key": K.key(line), + "title": t["title"], + "reported_status": "announced", + "source_note": ( + f"HM Treasury, Budget 2026 Table 6.1 line {line} ({t['head']}" + + (f", {t['fiscal_head']}" if t.get("fiscal_head") else "") + + f"); exchequer impact GBP m by FY: {t['values']}" + ), + "table_61_line": line, + "table_61_head": t["head"], + "table_61_heading": t["heading"], + "program": program, + "computability": computability, + } + m.update(kw) + measures.append(m) + return m + + +def gap(why, pattern, *, upstream, nearest=None): + d = { + "pe_reform_delta": None, + "why": why, + "name_search": ns(pattern), + "policyengine_uk_development_item": True, + "upstream_item": UPSTREAM[upstream], + "action_link": ISSUE, + } + if nearest: + d["nearest_variable_and_why_it_does_not_work"] = nearest + return d + + +def scope(why, pattern, link=None): + return { + "pe_reform_delta": None, + "why": why, + "name_search": ns(pattern), + "out_of_model_scope": True, + "action_link": link or ISSUE, + } + + +def baseline(*paths): + return {p: P[p] for p in paths} + + +def receipt(dash_id, artifact, sha, reform_sha, binding, generated_at): + return { + "repo": DASHBOARD_REPO, + "commit": DASHBOARD_MAIN, + "path": f"public/data/artifacts/{dash_id}/{artifact}/receipt.json", + "receipt_file_sha256": sha, + "reform_sha256": reform_sha, + "reform_binding": binding, + "generated_at": generated_at, + **RUN_IDENTITY, + } + + +def rates_up(paths): + from policyengine_uk import CountryTaxBenefitSystem + + params = CountryTaxBenefitSystem().parameters + out = {} + for path in paths: + node = params + for part in path.split("."): + node = getattr(node, part) + out[path] = { + str(y): round(float(node(f"{y}-04-06")) + 0.02, 6) for y in POLICY_YEARS + } + return out + + +# --------------------------------------------------------------------------- +# Executed on Budget day by the dashboard (seven published runs, one more run +# outside the release) +# --------------------------------------------------------------------------- + +entry( + 1, + "household_energy_payment", + "expressible", + construction="executed_by_dashboard", + dashboard_measure_id="household_energy_payment_2026", + pe_reform_delta=None, + pe_modifier={ + "module": "src/uk_budget_data/budget_measures/household_energy_payment_2026.py", + "function": "household_energy_payment_2026", + "arguments": { + "amount": 100, + "year": 2026, + "country": "ENGLAND", + "bands": ["A", "B", "C", "D"], + }, + "replaced_variables": ["ebr_council_tax_rebate"], + "reads_variables": ["council_tax_band", "country"], + "parameter_legs": [], + "receipt": receipt( + "household_energy_payment_2026", + "101a07dc7238a10e", + "0fb0414b45435799fb50afa9d79c2c7af8998de2ee406d8da6c1561bd46298d4", + "ceede8365a9462ac389dc5210882ea23850babfed6e5bff8cf264e774cd616ad", + "code_revision", + "2026-10-28T17:07:18.666809+00:00", + ), + }, + engine_baseline_2026=baseline( + "gov.treasury.energy_bills_rebate.council_tax_rebate.amount" + ), + note="The modifier adds £100 in 2026-27 to the native ebr_council_tax_rebate formula for bands A to D in England; the native rebate's own parameter is 0 at the pin, so the payment is the whole effect. Scotland, Wales and Northern Ireland get a Barnett share instead, not modelled.", +) +entry( + 2, + "vat", + "expressible", + construction="executed_by_dashboard", + dashboard_measure_id="domestic_electricity_zero_vat", + pe_reform_delta=None, + pe_modifier={ + "module": "src/uk_budget_data/budget_measures/zero_vat_on_domestic_electricity.py", + "function": "zero_vat_on_domestic_electricity", + "arguments": { + "rate_change_2026": -0.028342979605206865, + "rate_change_2027": -0.011149919361563782, + "geography": "GB", + "geography_2027": "UK", + }, + "replaced_variables": [], + "reads_variables": ["electricity_consumption"], + "parameter_legs": [], + "receipt": receipt( + "domestic_electricity_zero_vat", + "2d9c65726915303e", + "e464b000c2af360654462efd05180d0fed272728911dc7292d932b99117246ca", + "5ee4d8844432367d5557df38ee89e23b206e2c1236bce2451b0336ada02b58b6", + "code_revision", + "2026-10-28T17:08:09.754595+00:00", + ), + }, + engine_baseline_2026=baseline("gov.hmrc.vat.reduced_rate"), + note="The modifier lowers the VAT rate on electricity_consumption by 5 points times the share of each fiscal year's use that falls in the zero-rated period (about 57% of 2026-27 in Great Britain; about 22% of 2027-28 across the UK). Electricity only, not gas.", +) +entry( + 3, + "fuel_duty", + "expressible", + construction="executed_by_dashboard", + dashboard_measure_id="fuel_duty_rise_deferral", + pe_reform_delta=None, + pe_modifier={ + "module": "src/uk_budget_data/reforms.py (_create_fuel_duty_rise_deferral) + src/uk_budget_data/budget_measures/pump_vat_on_fuel_duty_deferral.py", + "function": "fuel_duty_deferral_rates + pump_vat_on_fuel_duty_deferral", + "arguments": {}, + "replaced_variables": [], + "reads_variables": ["petrol_spending", "diesel_spending"], + "parameter_legs": ["gov.hmrc.fuel_duty.petrol_and_diesel"], + "receipt": receipt( + "fuel_duty_rise_deferral", + "e077f8d4b70440c8", + "2f547d91b455b7e84cf8495cbfd0475ca626d7f1b4a95605de869cf61d52a30f", + "1dae41af83e4402f6b3af756d7dcdcc1dd525883dd32c3c8b9050456f52310f1", + "code_revision", + "2026-10-28T17:06:46.194418+00:00", + ), + }, + engine_baseline_2026=baseline("gov.hmrc.fuel_duty.petrol_and_diesel"), + note="Day-weighted annual fuel duty rates (fuel_duty_deferral_rates), plus the VAT on the lower pump price through a modifier on petrol and diesel spending.", +) +entry( + 7, + "carers_allowance", + "expressible", + construction="executed_by_dashboard", + dashboard_measure_id="carers_allowance_earnings_taper", + pe_reform_delta=None, + pe_modifier={ + "module": "src/uk_budget_data/budget_measures/carers_allowance_earnings_taper.py", + "function": "carers_allowance_earnings_taper", + "arguments": {"start_year": 2030, "first_year_share": 0.5, "taper_rate": 1.0}, + "replaced_variables": ["carers_allowance_pre_overlap"], + "reads_variables": [ + "carers_allowance_earnings", + "care_hours", + "would_claim_carers_allowance", + "carers_allowance_reported", + ], + "parameter_legs": [], + "receipt": receipt( + "carers_allowance_earnings_taper", + "9458905cefca3e88", + "0852c433f5f76cd2e7b27f1bcf6721610d8e36e80b5f416727901e8125d273f6", + "f45943c11aed787c570af3cc82d11af9cf181ce0877b5cb9350a9bc1e2f84dc8", + "code_revision", + "2026-10-28T17:08:00.734195+00:00", + ), + }, + engine_baseline_2026=baseline( + "gov.dwp.carers_allowance.earnings_limit", + "gov.dwp.carers_allowance.min_hours", + "gov.dwp.carers_allowance.min_age", + ), + note="A 100% taper above the Weekly Earnings Limit (16 hours at the National Living Wage) in England and Wales, half of 2030-31 counted because the start date within that year is not set.", +) +entry( + 15, + "income_tax", + "expressible", + construction="executed_by_dashboard", + dashboard_measure_id="trading_allowance_raise", + pe_reform_delta={ + "gov.hmrc.income_tax.allowances.trading_allowance": { + str(y): 3000 for y in POLICY_YEARS + } + }, + delta_kind="absolute_value_by_year", + pe_modifier={ + "module": "src/uk_budget_data/reforms.py", + "function": "_create_trading_allowance_raise", + "arguments": {}, + "replaced_variables": [], + "reads_variables": [], + "parameter_legs": ["gov.hmrc.income_tax.allowances.trading_allowance"], + "receipt": receipt( + "trading_allowance_raise", + "eddde823a65b7be4", + "59db02ec313c62f8c194580c806e782431f67a0f3b08df34adb91445a1eb9221", + "cdcc8b19d5234afe31ae857db16a8f2d6a18578a1d7da1c890b3c5f21832c268", + "content", + "2026-10-28T17:07:22.701184+00:00", + ), + }, + engine_baseline_2026=baseline("gov.hmrc.income_tax.allowances.trading_allowance"), + note="The data hold profits, not gross trading receipts, so the engine relieves a profit up to £3,000 in full and gives no relief above it (the dashboard's own description).", +) +entry( + 52, + "income_tax", + "expressible", + construction="executed_by_dashboard", + dashboard_measure_id="state_pension_income_tax_protection", + pe_reform_delta=None, + pe_modifier={ + "module": "src/uk_budget_data/budget_measures/pension_tax_allowance.py", + "function": "pension_tax_allowance", + "arguments": { + "first_year": 2027, + "last_year": 2029, + "basic_rate": 0.2, + "taper_divisor": 3, + "weekly_rounding": 0.05, + }, + "replaced_variables": [ + "income_tax", + "uc_income_tax_on_earnings", + "uc_income_tax_on_trade", + ], + "reads_variables": ["total_income", "is_SP_age"], + "parameter_legs": [], + "receipt": receipt( + "state_pension_income_tax_protection", + "1bfeed3c27a3c9b9", + "4f20f3c69f9ddb3b7894d3281e9416774a1e12d747de1e573629570871b3e960", + "65a95c811e5d6c6dd0b54255dd20ab1439c542994f1918778b93769c14199524", + "code_revision", + "2026-10-28T17:08:11.348926+00:00", + ), + }, + engine_baseline_2026=baseline( + "gov.dwp.state_pension.new_state_pension.amount", + "gov.hmrc.income_tax.allowances.personal_allowance.amount", + ), + note="A non-refundable income tax reduction for people over State Pension age: 20% of the full new State Pension above the personal allowance, withdrawn £1 for every £3 of total income above the full new State Pension, 2027-28 to 2029-30.", +) +entry( + 61, + "income_tax", + "expressible", + construction="executed_by_dashboard", + dashboard_measure_id="property_income_rates_up", + pe_reform_delta=rates_up( + [ + f"gov.hmrc.income_tax.rates.property.{b}" + for b in ("basic", "higher", "additional") + ] + ), + delta_kind="absolute_value_by_year", + pe_modifier={ + "module": "src/uk_budget_data/reforms.py", + "function": "property_income_rates_up", + "arguments": {}, + "replaced_variables": [], + "reads_variables": [], + "parameter_legs": [ + f"gov.hmrc.income_tax.rates.property.{b}" + for b in ("basic", "higher", "additional") + ], + "receipt": receipt( + "property_income_rates_up", + "7bf77b35e33233e0", + "20df7073b4481a5f5b0073c6949e0d0c2894adac88d6e70778946456b2fe899d", + "d46bf10b10b298db4059d7b6a2dccac7997e09afec0a88542c2059ea70646aa0", + "content", + "2026-10-28T17:10:45.376537+00:00", + ), + }, + engine_baseline_2026=baseline( + *[ + f"gov.hmrc.income_tax.rates.property.{b}" + for b in ("basic", "higher", "additional") + ] + ), + note="Current law's property rates plus 2 points from 2027 (pe_reform_delta computed at the pin from each year's 6 April value). Scottish taxpayers' property income follows Scottish rates; finance cost relief follows the property basic rate UK-wide.", +) +entry( + 62, + "income_tax", + "expressible", + construction="delta_on_current_law", + dashboard_measure_id="savings_income_rates_up", + pe_reform_delta=rates_up( + [ + f"gov.hmrc.income_tax.rates.savings.{b}" + for b in ("basic", "higher", "additional") + ] + ), + delta_kind="absolute_value_by_year", + engine_baseline_2026=baseline( + *[ + f"gov.hmrc.income_tax.rates.savings.{b}" + for b in ("basic", "higher", "additional") + ] + ), + note="Registered by the dashboard but not in its published release, so there is no published receipt to cite; tranche 3 computes it from this delta.", +) + +# --------------------------------------------------------------------------- +# Expressible at the pin, not run on Budget day +# --------------------------------------------------------------------------- + +entry( + 60, + "capital_gains_tax", + "expressible", + construction="delta_on_current_law", + pe_reform_delta={ + "gov.hmrc.cgt.badr.lifetime_limit": {str(y): 500000 for y in POLICY_YEARS} + }, + delta_kind="absolute_value_by_year", + engine_baseline_2026=baseline("gov.hmrc.cgt.badr.lifetime_limit"), + head_variables=["capital_gains_tax", "capital_gains_badr"], + note="capital_gains_tax reads gov.hmrc.cgt.badr.lifetime_limit at the pin (variables/gov/hmrc/capital_gains_tax/capital_gains_tax.py). Not run on Budget day.", +) + + +def _cap_history(): + from policyengine_uk import CountryTaxBenefitSystem + + node = CountryTaxBenefitSystem().parameters.gov.dft.bus.fares.cap + return { + v.instant_str: ("inf" if v.value == float("inf") else v.value) + for v in reversed(node.values_list) + if v.instant_str >= "2023" + } + + +entry( + 4, + "bus_fares", + "partial", + # a reversal once the pre-measure world is known; until then no + # construction is claimed + intended_construction="reversal_on_certified_world", + pe_reform_delta=None, + engine_baseline_2026=baseline("gov.dft.bus.fares.cap"), + engine_parameter_history={"gov.dft.bus.fares.cap": _cap_history()}, + missing=( + "The pre-measure 2027 cap. The engine already carries the measure: " + "gov.dft.bus.fares.cap is £2 from 1 January 2027 to 1 January 2028 " + "(its metadata cites DfT's '£2 bus fares from January 2027'). A " + "counterpart reverses it, which needs the cap 2027 would have had " + "without the measure; the parameter has one upstream revision " + "(policyengine-uk c8d1f89a9d, 5 October 2026) and records no " + "pre-announcement path. Supersedes data/uk/bus_fares_gap.json's " + "not_expressible verdict at 2.89.2, which predates the bus model." + ), + action_link=BUS_GAP, +) + +# --------------------------------------------------------------------------- +# Household-facing, not expressible: policyengine-uk development items +# --------------------------------------------------------------------------- + +entry( + 5, + "income_tax", + "not_expressible", + **gap( + "No business mileage or mileage allowance payments: the approved rates are a tax-free amount per business mile, and the search finds no mileage in the engine.", + r"mileage|business_travel|approved_mileage", + upstream="mileage", + ), +) +entry( + 6, + "parental_pay", + "not_expressible", + **gap( + "A new benefit with no parameter or variable at the pin.", + r"parental_allowance|paternity|parental_leave", + upstream="parental_allowance", + nearest={ + "variable": "statutory_paternity_pay", + "why_not": "an existing statutory payment; the new Parental Allowance is a separate benefit with its own rules, none of which the engine carries", + }, + ), +) +entry( + 8, + "universal_credit", + "not_expressible", + **gap( + "No surplus earnings rule: the search finds no parameter or variable for it.", + r"surplus", + upstream="uc_surplus", + ), +) +entry( + 11, + "savings", + "not_expressible", + **gap( + "No ISA subscription or subscription limit: the engine holds ISA balances and interest, not subscriptions, so a limit has nothing to bind.", + r"(^|[._])isa([._]|$)|individual_savings|subscri", + upstream="isa", + nearest={ + "variable": "cash_isa", + "why_not": "a holding, not an annual subscription", + }, + ), +) +entry( + 26, + "capital_gains_tax", + "not_expressible", + **gap( + "No founders' gains: the engine's relief-specific gains variable is capital_gains_badr (Business Asset Disposal Relief), and nothing identifies qualifying founders of high-growth companies.", + r"founder|capital_gains_badr|investors_relief", + upstream="cgt_reliefs", + ), +) +entry( + 38, + "savings", + "not_expressible", + **gap( + "No first-time-buyer ISA, and no ISA subscriptions to move into one.", + r"first_time|(^|[._])isa([._]|$)", + upstream="isa", + ), +) +entry( + 39, + "savings", + "not_expressible", + **gap( + "The engine's withdrawal charge (gov.hmrc.isa.lifetime.withdrawal_charge, 0.25 at the pin) is read only by lifetime_isa_countable_capital, the means-test value of a Lifetime ISA. The engine has no withdrawal or home purchase, so it cannot apply a lower charge to first-time buyers above the £450,000 cap; lowering the parameter would instead revalue every holder's countable capital.", + r"withdrawal_charge|first_time|home_purchase", + upstream="isa", + nearest={ + "variable": "lifetime_isa_countable_capital", + "why_not": "applies the charge to a holder's balance for means tests; the measure changes the charge on one kind of withdrawal", + }, + ), +) +entry( + 53, + "income_tax", + "not_expressible", + **gap( + "The measure changes when pension and lump sum death benefits are taxed after a member dies (the two-year rule). The engine has no death event; its one death-benefit variable, pension_annual_allowance_taxable_lump_sum_death_benefits, is an input with no formula and no rule about timing after a death.", + r"death|bereave|lump_sum", + upstream="estates", + ), +) +entry( + 66, + "vat", + "not_expressible", + **gap( + "The engine's consumption variables are by spending category (e.g. recreation_consumption); none isolates admissions to family days out, the base of this reduced rate.", + r"recreation|attraction|admission|day_out", + upstream="vat_categories", + nearest={ + "variable": "recreation_consumption", + "why_not": "a whole spending category; the measure covers only part of it", + }, + ), +) +entry( + 67, + "bus_fares", + "not_expressible", + **gap( + "Free travel enters the bus model only as bus_pass_eligible, a dataset input whose documentation says it 'is not a pass-eligibility calculator'; no rule gives free travel to children aged 5 to 15, and the measure runs for one month (August 2026).", + r"bus_fare|fare_allocation|bus_pass|concession", + upstream="bus_age_fares", + ), +) +entry( + 83, + "student_loans", + "not_expressible", + **gap( + "No interest cap: plan_2_interest_rate is RPI plus up to gov.hmrc.student_loans.interest_rates.plan_2.additional_rate (3%), and postgraduate_interest_rate (Plan 3) is RPI plus postgraduate_additional_rate (3%); neither formula has a ceiling.", + r"student_loan.*interest|interest_rates\.plan_2|postgraduate_additional_rate", + upstream="student_loan_interest_cap", + ), +) +entry( + 86, + "income_tax", + "not_expressible", + **gap( + "No company cars or benefit-in-kind appropriate percentages: the search's only hits are council tax discount percentages.", + r"company_car(?!ries)|benefit_in_kind|appropriate_percentage", + upstream="bik", + ), +) +entry( + 93, + "council_tax", + "not_expressible", + **gap( + "The engine carries the surcharge (gov.hmrc.council_tax.high_value_surcharge, variable high_value_council_tax_surcharge) but no deferral, discount or exemption within it.", + r"high_value|surcharge|deferr", + upstream="hvcts_reliefs", + ), +) + +# --------------------------------------------------------------------------- +# Out of the household model's scope +# --------------------------------------------------------------------------- + +SPEND = ( + "Public spending ({what}), out of the household model's scope. The engine's spending channel, " + + LEVERS["spending"] + + ", distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it." +) +BUSINESS = ( + "A tax on businesses ({what}), out of the household model's scope: it changes what companies owe. The engine's corporate channel, " + + LEVERS["corporate"] + + ", spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds." +) +RATES = ( + "A business rates measure ({what}), out of the household model's scope. The engine carries business rates only as a household incidence, " + "business_rates ('Total incidence from exposure to business rates via corporate shareholdings'), and its business rates parameters are " + "revenue statistics (gov.hmrc.business_rates.statistics.revenue.*): it has no properties, reliefs or thresholds for the measure to change, " + "so its £ change would have to be supplied as an input, restating the costing." +) +INDIRECT = ( + "An indirect tax or duty collected from businesses ({what}), out of the household model's scope. The engine's consumer channel, " + + LEVERS["consumer"] + + ", spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds." +) +FINANCE = "Lending, guarantees, investment or a financial transaction ({what}): no household tax or benefit rule changes." +ENFORCE = "An enforcement, compliance or administrative change ({what}): the rules households face do not change." +REGULATE = "A regulatory change ({what}) with no tax or benefit rule for households." +DETAINED = "A change to the {what}: the search finds no parameter or variable for hospital detention, prison or offending, so the engine cannot identify the people it affects." + +SCOPE = { + 9: (SPEND, "books for children aged 0 to 5", r"early_years|childcare|book"), + 10: ( + INDIRECT, + "tariffs on food products, heating oil and fertiliser", + r"tariff|customs|import", + ), + 12: ( + RATES, + "business rates relief for pubs and venues", + r"business_rates|non_domestic|nndr", + ), + 13: ( + RATES, + "small business rate relief thresholds", + r"business_rates|small_business|non_domestic", + ), + 14: ( + RATES, + "business rates transitional support", + r"business_rates|transitional|non_domestic", + ), + 16: (INDIRECT, "tariff suspensions on business inputs", r"tariff|customs|import"), + 17: (INDIRECT, "Climate Change Levy rates", r"climate_change_levy|ccl"), + 18: (INDIRECT, "red diesel duty", r"red_diesel|rebated|fuel_duty"), + 19: ( + INDIRECT, + "alcohol duty small producer relief", + r"alcohol_duty|small_producer", + ), + 20: ( + FINANCE, + "Good Growth Funds for mayoral authorities", + r"growth_fund|mayor|combined_authorit", + ), + 21: ( + SPEND, + "business rates retention by mayoral authorities", + r"business_rates|retention|mayor", + ), + 22: (SPEND, "local apprenticeship services", r"apprentice"), + 23: (SPEND, "regional growth projects", r"regional|regeneration"), + 24: ( + FINANCE, + "Cambridge Development Corporation", + r"development_corporation|cambridge", + ), + 25: (SPEND, "strategic authorities' capacity", r"strategic_authorit|mayor"), + 27: ( + BUSINESS, + "R&D relief claim forms", + r"research_and_development|r_and_d|rd_relief", + ), + 28: ( + BUSINESS, + "Enterprise Management Incentives share options", + r"enterprise_management|share_option|emi", + ), + 29: (FINANCE, "British Business Bank", r"business_bank"), + 30: (FINANCE, "National Wealth Fund", r"wealth_fund"), + 31: (SPEND, "artificial intelligence", r"artificial_intelligence|\bai\b"), + 32: (SPEND, "public EV charging costs", r"charg|electric_vehicle"), + 33: (SPEND, "rough sleeping", r"rough_sleep|homeless"), + 34: (SPEND, "reprioritised housing budgets", r"rough_sleep|homeless"), + 35: (SPEND, "social and affordable homes", r"social_housing|affordable"), + 36: ( + INDIRECT, + "VAT on land sold to social housing providers", + r"vat|social_housing", + ), + 37: ( + FINANCE, + "an equity loan scheme for first-time buyers", + r"equity_loan|help_to_buy|first_time", + ), + 40: (SPEND, "defence", r"defen[cs]e"), + 41: (SPEND, "defence", r"defen[cs]e"), + 42: (SPEND, "defence administration", r"defen[cs]e"), + 43: (SPEND, "air defence for Ukraine", r"ukraine|defen[cs]e"), + 44: (FINANCE, "UK Export Finance", r"export_finance"), + 45: (ENFORCE, "HMRC measures to close the tax gap", r"tax_gap|compliance|evasion"), + 46: (ENFORCE, "online marketplace VAT liability", r"marketplace|online_sales"), + 47: (ENFORCE, "DWP fraud and error controls", r"fraud|error"), + 48: ( + DETAINED, + "benefit entitlements of serious offenders detained in hospital in Great Britain", + r"hospital|prison|detain|offender", + ), + 49: (BUSINESS, "the distributions framework", r"distribution|dividend"), + 50: (BUSINESS, "Advance Corporation Tax", r"corporation_tax|advance_corporation"), + 51: ( + INDIRECT, + "the VAT reverse charge on imported services", + r"reverse_charge|vat", + ), + 54: (BUSINESS, "the taxation of foreign sovereign investors", r"sovereign|foreign"), + 55: ( + BUSINESS, + "the Bank Corporation Tax Surcharge", + r"bank_surcharge|corporation_tax|bank_levy", + ), + 56: (INDIRECT, "Machine Games Duty", r"gambling|machine_games|gaming"), + 57: ( + RATES, + "business rates reliefs for vape shops and adult gaming centres", + r"business_rates|vape", + ), + 58: ( + BUSINESS, + "capital allowances", + r"capital_allowance|writing_down|first_year_allowance", + ), + 59: (INDIRECT, "VAT on private hire operators' fares", r"private_hire|taxi|vat"), + 63: ( + SPEND, + "civil service administration budgets", + r"civil_service|administration", + ), + 64: (SPEND, "a civil service AI fund", r"civil_service"), + 65: ( + INDIRECT, + "Vehicle Excise Duty on heavy goods vehicles", + r"heavy_goods|hgv|vehicle_excise", + ), + 68: ( + BUSINESS, + "the foreign permanent establishment exemption", + r"permanent_establishment|corporation_tax", + ), + 69: ( + BUSINESS, + "the Electricity Generator Levy", + r"generator_levy|electricity_generator", + ), + 70: ( + SPEND, + "the British Industrial Competitiveness Scheme", + r"industrial|competitiveness", + ), + 71: (INDIRECT, "Carbon Price Support rates", r"carbon_price|carbon"), + 72: ( + SPEND, + "the Youth Guarantee and apprenticeships", + r"youth_guarantee|apprentice", + ), + 73: (SPEND, "the Union Learning Fund", r"union_learning|skills"), + 74: (FINANCE, "investment for growth", r"investment|growth"), + 75: (FINANCE, "the Growth Guarantee Scheme", r"guarantee"), + 76: (RATES, "business rates retention in Leeds", r"business_rates|retention"), + 77: ( + INDIRECT, + "tariffs and contributions under the UK-India agreement", + r"tariff|customs|double_contribution", + ), + 78: (INDIRECT, "steel tariffs and quotas", r"tariff|steel"), + 79: (BUSINESS, "stamp taxes on shares", r"stamp_duty|stamp_tax|securities"), + 80: (RATES, "business rates relief for EV charge points", r"business_rates|charg"), + 81: (SPEND, "misogyny prevention programmes", r"misogyny|violence"), + 82: (SPEND, "infected blood compensation", r"infected_blood|compensation"), + 84: (SPEND, "Landfill Tax costs of illegal waste clearances", r"landfill|waste"), + 85: ( + SPEND, + "the NHS and the Violence against Women and Girls Strategy", + r"nhs|violence", + ), + 87: ( + REGULATE, + "licensing of supported housing", + r"supported_housing|exempt_accommodation", + ), + 88: (INDIRECT, "VAT on donated medicines", r"vat|medicine"), + 89: (INDIRECT, "VAT on medical deputies' staffing supplies", r"vat|staffing"), + 90: (INDIRECT, "the Landfill Communities Fund credit cap", r"landfill"), + 91: ( + REGULATE, + "registration of franchised higher education providers", + r"higher_education|franchise|student", + ), + 92: ( + SPEND, + "local authority high needs budgets", + r"high_needs|special_educational|send", + ), + 94: ( + ENFORCE, + "the timing of mandatory payrolling of benefits in kind", + r"benefit_in_kind|payroll", + ), + 95: ( + INDIRECT, + "customs duty relief on low value imports", + r"customs|low_value|import", + ), + 96: ( + INDIRECT, + "the Emissions Trading Scheme's coverage of energy from waste", + r"emissions_trading|\bets\b|carbon", + ), + 97: ( + BUSINESS, + "corporation tax on gas distribution revenue", + r"corporation_tax|gas_distribution", + ), + 98: ( + BUSINESS, + "double taxation relief for investors in US LLCs", + r"double_taxation|foreign_income|llc", + ), +} + +PROGRAM = { + SPEND: "public_spending", + BUSINESS: "business_tax", + INDIRECT: "indirect_tax", + FINANCE: "financial_transactions", + ENFORCE: "compliance_and_administration", + REGULATE: "regulation", + RATES: "business_rates", + DETAINED: "benefits", +} + +for line, (template, what, pattern) in SCOPE.items(): + entry( + line, + PROGRAM[template], + "not_expressible", + **scope(template.format(what=what), pattern), + ) + +# Table 6.2 line 4 has no Table 6.1 line of its own. +t = T61["t62_4"] +measures.append( + { + "measure_key": K.PREFIX + K.TABLE_62[4], + "title": t["title"], + "reported_status": "announced", + "source_note": f"HM Treasury, Budget 2026 Table 6.2 line 4 (financial transactions, no Table 6.1 line); cash impact GBP m by FY: {t['values']}", + "table_62_line": 4, + "program": "financial_transactions", + "computability": "not_expressible", + **scope( + FINANCE.format(what="Sovereign AI"), r"sovereign|artificial_intelligence" + ), + } +) + +# --------------------------------------------------------------------------- +# Packages +# --------------------------------------------------------------------------- + +BY_LINE = {m["table_61_line"]: m for m in measures if "table_61_line" in m} +for slug, spec in K.PACKAGES.items(): + lines = spec["lines"] + if lines is None: # Table 6.1's own Tax/Spend split + head = "Tax" if "tax" in slug else "Spend" + lines = sorted(l for l, m in BY_LINE.items() if m["table_61_head"] == head) + legs = [K.key(l) for l in lines] + comp = [BY_LINE[l]["computability"] for l in lines] + expressible = [K.key(l) for l, c in zip(lines, comp) if c == "expressible"] + m = { + "measure_key": K.package_key(slug), + "title": spec["title"], + "reported_status": "announced", + "source_note": ( + f"HM Treasury, Budget 2026 Table 6.1 row '{spec['table_61_row']}'; GBP m by FY: {TOTALS[spec['table_61_row']]}" + if spec["table_61_row"] + else "A composition of Table 6.1 lines that producers score together; see `package_of`." + ), + "program": "package", + "construction": "package_of_registry_measures", + "package_of": legs, + "pe_reform_delta": None, + } + if all(c == "expressible" for c in comp): + m["computability"] = "expressible" + m["engine_baseline_2026"] = {} + if slug == "package_policyengine_seven": + m["pe_modifier"] = { + "module": "src/uk_budget_data/reforms.py", + "function": "AUTUMN_BUDGET_2026_COMBINED_COMPONENTS", + "arguments": {}, + "replaced_variables": [], + "reads_variables": [], + "parameter_legs": [], + "receipt": receipt( + "combined--f91fe847275730cb", + "4faed20984703c7d", + "a00d0614eeed1c5cfaa63353d08a604a433bf881329bfd2d7b60c1681c243e67", + "09498c7b550716444725ef211725c894bc0efa13bd6c855efb8bdadc93b0e834", + "code_revision", + "2026-10-28T17:54:34.931664+00:00", + ), + } + m["note"] = ( + "Run as one simulation of all seven measures (the combined run), not as a sum of the seven runs." + ) + else: + m["note"] = ( + "Every leg is expressible; a counterpart is the legs' runs combined, stated as constructed." + ) + else: + m["computability"] = "partial" + m["missing"] = ( + f"{len(lines) - len(expressible)} of {len(lines)} legs are not expressible " + "(see each leg's verdict); a PolicyEngine number covers only " + f"the {len(expressible)} expressible legs: {expressible}" + ) + m["expressible_legs"] = expressible + measures.append(m) + + +def registry(): + lines = sorted(m["table_61_line"] for m in measures if "table_61_line" in m) + assert lines == sorted(K.TABLE_61), "every Table 6.1 line exactly once" + counts = {} + for m in measures: + counts[m["computability"]] = counts.get(m["computability"], 0) + 1 + return { + "schema_version": 1, + "fiscal_event": "autumn_budget_2026", + "event_date": "2026-10-28", + "chancellor": "John Healey", + "administration": "Burnham government (from 2026-07-20)", + "registry_rule": ( + "One entry per measure ANNOUNCED at Autumn Budget 2026: every HM Treasury Table 6.1 line " + "(including the lines a household model cannot touch), the one Table 6.2 financial " + "transaction with no Table 6.1 line, and the packages producers score. It is the port's " + "key space (#230): every staged AB2026 claim carries conditions.measure_key from this " + "file, and every measure carries a computability verdict against the engine the Budget-day " + "runs used. It stages NO value claims; values live in the harvest under their producers " + "(sources/harvest-uk-ab2026-2026-10-29/)." + ), + "measure_key_rule": ( + "ab2026__, written out per Table 6.1 line in pipeline/ab2026_measure_keys.py so a " + "reworded title cannot move a key; ab2026__package_ for packages; ab2026_option__ " + "for options costed by others. The pre-event registry (data/uk/budget_2026_measures.json, " + "#96) keeps its own keys; pre_event_outcomes records what became of each." + ), + "verification_note": ( + "Every pe_reform_delta and parameter_legs path was RESOLVED against an installed " + f"policyengine-uk {PIN} on 2026-10-29, and engine_baseline_2026 records what that engine " + "returns for 2026-06-01. Every not_expressible verdict carries a name_search COMPUTED over " + f"the pinned engine's parameter tree ({len(P)} leaves) and variable list ({len(V)}), never " + "typed. Mechanisms named in a verdict were read from the engine's source at the pin or from " + f"the dashboard's source at {DASHBOARD_REPO}@{DASHBOARD_MAIN[:8]}." + ), + "engine_pin": { + "policyengine_uk": PIN, + "resolved_on": "2026-10-29", + "pin_meaning": ( + "The pin follows the Budget-day runs: every published receipt of the UK Autumn Budget " + "dashboard records policyengine 6.11.1 with certified_for_model_version 2.151.0, and those " + "runs are AB2026's PolicyEngine counterparts (tranche 3). It is NOT the certified populace " + "pair data/uk/certified_pairs.json holds (policyengine 6.11.0, policyengine-uk 2.132.6); " + "tranche 3 adds a 6.11.1 pair record before any counterpart is ingested." + ), + "budget_day_runs": RUN_IDENTITY, + }, + "computability": { + "expressible": "every leg is a resolvable parameter at the pin, or the measure was executed by a receipt-pinned dashboard modifier (construction executed_by_dashboard)", + "partial": "some legs resolve; the missing ones are named in `missing`", + "not_expressible": "the engine has no representation; `why` names the gap, `name_search` records the computed search, `action_link` says where it goes", + }, + "construction_vocabulary": { + "executed_by_dashboard": "the measure ran on Budget day as a dashboard reform; pe_modifier names the module, function, arguments, the engine variables it replaces or reads, its parameter legs, and the published receipt (path, file sha256, reform sha256, run identity)", + "delta_on_current_law": "pe_reform_delta applied to current law at the pin", + "reversal_on_certified_world": "the measure is already in the engine's current law; a counterpart executes the pre-measure world as baseline (#56 convention)", + "package_of_registry_measures": "a composition of registry measures (`package_of`, a list of keys, as in AB2025); a PolicyEngine number covers the expressible legs, stated as constructed", + }, + "exogenous_levers_note": ( + "The engine carries exogenous incidence levers that spread a £ amount supplied as an input: " + + "; ".join(LEVERS.values()) + + ". Feeding a Table 6.1 figure through one restates HM Treasury's costing rather than " + "estimating it, so no lever makes a line expressible here." + ), + "action_link_rule": "Every not_expressible measure links to #230, which lists the upstream development items; the bus fare cap's partial verdict links to #106, the bus gap record.", + "pre_event_outcomes": PRE_EVENT, + "triage": { + **counts, + "executed_by_dashboard": sum( + 1 for m in measures if m.get("construction") == "executed_by_dashboard" + ), + "policyengine_uk_development_item": sum( + 1 for m in measures if m.get("policyengine_uk_development_item") + ), + "out_of_model_scope": sum( + 1 for m in measures if m.get("out_of_model_scope") + ), + "announced": sum( + 1 for m in measures if m["reported_status"] == "announced" + ), + }, + "measures": measures, + } + + +# What became of each measure the pre-event registry (#96) recorded as under +# consideration. "Not in Table 6.1 or 6.4" is a statement about those two +# tables, read in full, nothing more. +PRE_EVENT = { + "ab2026__property_income_tax_rate_rise": "announced: Table 6.1 line 61 (" + + K.key(61) + + ")", + "ab2026__bank_profit_surcharge_increase": "announced: Table 6.1 line 55 (" + + K.key(55) + + "), temporary, 3% to 6% for 15 months", + "ab2026__business_rates_warehouse_up_pub_down": "partly: Table 6.1 line 12 raises pubs' relief (" + + K.key(12) + + "); no warehouse surcharge line in Table 6.1 or 6.4", + "ab2026__high_value_council_tax_surcharge": "an AB2025 measure; Table 6.1 line 93 changes its reliefs (" + + K.key(93) + + ") and Table 6.4 line 11 recosts it", + "ab2026__cgt_align_with_income_tax_rates": "not in Table 6.1 or 6.4", + "ab2026__cgt_remove_uplift_on_death": "not in Table 6.1 or 6.4", + "ab2026__pension_tax_free_lump_sum_restriction": "not in Table 6.1 or 6.4", + "ab2026__ni_extended_to_investment_property_pension_income": "not in Table 6.1 or 6.4", + "ab2026__personal_allowance_freeze_extension": "not in Table 6.1 or 6.4", + "ab2026__wealth_tax_2pc_above_10m": "not in Table 6.1 or 6.4", + "ab2026__council_tax_and_sdlt_to_land_value_tax": "not in Table 6.1 or 6.4", + "ab2026__online_sales_levy": "not in Table 6.1 or 6.4 (line 46, online marketplace liability, is a VAT collection measure, not a levy)", + "ab2026__vape_duty_increase": "not in Table 6.1 or 6.4 (line 57 removes business rates reliefs for vape shops)", + "ab2026__ated_increase_and_sdlt_envelope_close": "not in Table 6.1 or 6.4", +} + + +def keys_md(reg): + out = [ + "# Autumn Budget 2026 registry keys", + "", + "Generated by `pipeline/build_ab2026_registry.py` from `data/uk/ab2026_measures.json`; do not edit.", + "Every staged row's `measure_key` is one of these, or (Table 6.4 recostings only) a key from the", + "earlier event's registry named in `conditions.recosting_of`.", + "", + "| key | Table 6.1 line | computability | title |", + "|---|--:|---|---|", + ] + for m in reg["measures"]: + line = m.get("table_61_line") or ( + f"6.2/{m['table_62_line']}" if m.get("table_62_line") else "" + ) + out.append( + f"| `{m['measure_key']}` | {line} | {m['computability']} | {m['title'].replace('|', '/')} |" + ) + return "\n".join(out) + "\n" + + +def main(argv=None): + reg = registry() + text = json.dumps(reg, indent=1, ensure_ascii=False) + "\n" + md = keys_md(reg) + if argv and "--check" in argv: + ok = OUT.read_text() == text and KEYS_MD.read_text() == md + print("byte-identical" if ok else "DIFFERS") + return 0 if ok else 1 + OUT.write_text(text) + KEYS_MD.parent.mkdir(parents=True, exist_ok=True) + KEYS_MD.write_text(md) + print(f"wrote {OUT.relative_to(REPO)}: {len(measures)} measures, {reg['triage']}") + return 0 + + +if __name__ == "__main__": + sys.exit(main(sys.argv[1:])) diff --git a/pipeline/harvest_counts.py b/pipeline/harvest_counts.py new file mode 100644 index 000000000..a65cae7bc --- /dev/null +++ b/pipeline/harvest_counts.py @@ -0,0 +1,75 @@ +"""Write or check a harvest's COUNTS.json from its family directories. + +COUNTS.json pins, per family, the row count the ingest's exact accounting +reads as `read`, the primary count, the family's producer slugs and the +sha256 of its claims_staged.jsonl.gz, plus `_totals`. The AB2025 file was +written by hand; this computes the same shape so a new family cannot be +added without its pin moving in the same commit. + + python pipeline/harvest_counts.py sources/harvest-uk-ab2026-2026-10-29 --pinned-on 2026-10-29 + python pipeline/harvest_counts.py sources/harvest-uk-ab2026-2026-10-29 --check +""" + +import argparse +import gzip +import hashlib +import json +import sys +from pathlib import Path + + +def counts(harvest, pinned_on): + out, rows_total, restated = {}, 0, 0 + families = sorted( + p for p in Path(harvest).iterdir() if (p / "claims_staged.jsonl.gz").exists() + ) + for fam in families: + gz = fam / "claims_staged.jsonl.gz" + with gzip.open(gz, "rt") as fh: + rows = [json.loads(line) for line in fh if line.strip()] + primaries = [ + line + for line in (fam / "manifest.jsonl").read_text().splitlines() + if line.strip() + ] + out[fam.name] = { + "rows": len(rows), + "primaries": len(primaries), + "claims_sha256": hashlib.sha256(gz.read_bytes()).hexdigest(), + "sources": sorted({r["source"] for r in rows}), + } + rows_total += len(rows) + restated += sum(1 for r in rows if r["attribution"] == "restated") + out["_totals"] = { + "rows": rows_total, + "restated": restated, + "families": len(families), + "pinned_on": pinned_on, + } + return out + + +def main(argv=None): + ap = argparse.ArgumentParser(description=__doc__.splitlines()[0]) + ap.add_argument("harvest") + ap.add_argument("--pinned-on") + ap.add_argument("--check", action="store_true") + args = ap.parse_args(argv) + path = Path(args.harvest) / "COUNTS.json" + if args.check: + pinned = json.loads(path.read_text()) + fresh = counts(args.harvest, pinned["_totals"]["pinned_on"]) + if fresh != pinned: + print("COUNTS.json is stale") + return 1 + print("COUNTS.json matches the families") + return 0 + if not args.pinned_on: + ap.error("--pinned-on is required when writing") + path.write_text(json.dumps(counts(args.harvest, args.pinned_on), indent=1) + "\n") + print(f"wrote {path}") + return 0 + + +if __name__ == "__main__": + sys.exit(main(sys.argv[1:])) diff --git a/pipeline/stage_ab2026_hmt_tables.py b/pipeline/stage_ab2026_hmt_tables.py new file mode 100644 index 000000000..94b871072 --- /dev/null +++ b/pipeline/stage_ab2026_hmt_tables.py @@ -0,0 +1,386 @@ +"""Stage HM Treasury's Budget 2026 Tables 6.1, 6.2 and 6.4, cell by cell (#230). + +Every value is read from the published XLSX workbooks as stored (openpyxl, +cached values) and staged verbatim: value_raw is the cell, value is the cell +times 1,000,000. Nothing is derived, re-signed or summed. Each workbook is +hashed and the hash must match the manifest pin before a row is written, so a +re-published file stops the stager instead of silently re-staging. + + uv run --with openpyxl python pipeline/stage_ab2026_hmt_tables.py \ + --t61 PATH/Table_6.1_and_6.2_Budget_2026_Policy_Decisions.xlsx \ + --t64 PATH/Table_6.4_Recostings_Table.xlsx + +The output gzip is written with mtime 0, so a re-run on the same workbooks +is byte-identical and COUNTS.json's claims_sha256 pin holds. +""" + +import argparse +import gzip +import hashlib +import io +import json +import re +import sys +from pathlib import Path + +try: + from pipeline import ab2026_measure_keys as K +except ModuleNotFoundError: # direct script invocation without PYTHONPATH + import ab2026_measure_keys as K + +ROOT = Path(__file__).resolve().parent.parent +FAMILY = ROOT / "sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables" +FISCAL_EVENT = "autumn_budget_2026" + +T61 = { + "url": "https://assets.publishing.service.gov.uk/media/6ae187a2779378370fd80a9a/Table_6.1_and_6.2_Budget_2026_Policy_Decisions.xlsx", + "title": "Table 6.1 and 6.2 Budget 2026 Policy Decisions", + "sha256": "359958df6b1ab9746073725f6f8017ebdbb55040e57cf28612891f1e294a7f15", + "bytes": 67126, +} +T64 = { + "url": "https://assets.publishing.service.gov.uk/media/6ae0f93fe4367e5c8b0467d6/Table_6.4_Recostings_Table.xlsx", + "title": "Table 6.4 Measures announced at Spring Statement 2026 or earlier that will take effect from 29 October 2026 or later", + "sha256": "b0449df1737f128fdc32ffeffbe7ed11876e7ab1c80d14dd9908e68a38386d3e", + "bytes": 76402, +} +PUBLISHED = "2026-10-28" +RETRIEVED = "2026-10-28" +REVERIFIED = "2026-10-29" +COLLECTION = ( + "https://www.gov.uk/government/publications/supporting-documents-for-budget-2026" +) + +SIGN = ( + "positive = yield to the Exchequer (reduces borrowing); negative = cost to " + "the Exchequer (the tables print no legend; convention observed from the " + "rows themselves, e.g. the bank surcharge rise is + and the Household " + "Energy Payment is -, as in the AB2025 Table 4.1 harvest)" +) +FT_SIGN = ( + "positive = cash inflow to the public sector; negative = cash outflow " + "(financial transactions: the table prints no legend; convention observed " + "from the rows, e.g. the National Wealth Fund's investment is -)" +) + + +def fy_end(fy): + m = re.fullmatch(r"(\d{4})-(\d{2})", fy) + assert m, fy + return int(m.group(1)) + 1 + + +def raw(v): + """The cell as stored, without Python's float decoration.""" + if isinstance(v, float) and v.is_integer(): + v = int(v) + return str(v) + + +def sha256(path): + return hashlib.sha256(Path(path).read_bytes()).hexdigest() + + +def publication(src, locator): + return { + "title": src["title"], + "url": src["url"], + "date": PUBLISHED, + "publisher": "HM Treasury", + "primary_sha256": src["sha256"], + "locator": locator, + "access": "direct", + } + + +def base_row(src, *, title, fy, value, locator, table, column, conditions, metric): + r = { + "source": "hm_treasury", + "source_model": "hmt_costing_obr_certified", + "value": int(value) * 1_000_000 if float(value).is_integer() else value * 1e6, + "value_raw": raw(value), + "normalization": "£ million as printed in the cell x 1,000,000; sign as printed", + "value_kind": "point", + "period": fy_end(fy), + "time_basis": "fiscal_year", + "conditions": { + "fiscal_event": FISCAL_EVENT, + "fy": fy, + "geography": "UK", + "basis": "forecast", + **conditions, + }, + "reform_hint": title, + "attribution": "own", + "benchmark_class": "different_model", + "baseline_policy": None, + "publication": publication(src, locator), + "source_table": table, + "source_column": column, + "quote": f"{title} | {column}: {raw(value)}", + "parse_confidence": "high", + "status": "ok", + "note": "", + } + r.update(metric) + return r + + +REVENUE = {"metric": "revenue_change", "unit_concept": "gbp"} +FT = {"proposed_metric": "financial_transaction_cash_impact", "unit_concept": "gbp"} + + +def header_years(ws, row): + cols = {} + for c in ws[row]: + if isinstance(c.value, str) and re.fullmatch(r"\d{4}-\d{2}", c.value.strip()): + cols[c.column] = c.value.strip() + assert len(cols) == 6, (ws.title, cols) + return cols + + +def stage_t61(wb, out): + ws = wb["Table 6.1"] + years = header_years(ws, 4) + # Headings sit in column B with nothing beside them. A run of headings + # with no measure line between them is recorded whole ("A > B"): the + # table marks no hierarchy beyond that adjacency. + run, heading = [], "" + seen = set() + for row in ws.iter_rows(min_row=5): + b, c = row[1].value, row[2].value + if ( + isinstance(b, str) + and c is None + and not any(isinstance(x.value, (int, float)) for x in row[3:]) + ): + text = " ".join(b.split()) + if text[0].isdigit(): + break # footnotes + run.append(text) + heading = " > ".join(run) + continue + run = [] + if isinstance(b, int) and isinstance(c, str): + line, title = b, " ".join(c.split()) + fiscal_head = (row[3].value or "").strip() or None + head = row[4].value.strip() + seen.add(line) + for col, fy in years.items(): + v = ws.cell(row=row[0].row, column=col).value + assert isinstance(v, (int, float)), (line, fy, v) + cond = { + "sign_convention": SIGN, + "scoring_method": "final", + "table": "6.1", + "table_line": str(line), + "head": head, + "heading": heading, + "row_type": "measure", + } + if fiscal_head: + cond["fiscal_head"] = fiscal_head + out.append( + { + **base_row( + T61, + title=title, + fy=fy, + value=v, + locator=f"Table 6.1 line {line}, {fy} (cell {ws.cell(row=row[0].row, column=col).coordinate})", + table="Table 6.1 Budget 2026 policy decisions (£ million)", + column=fy, + conditions=cond, + metric=REVENUE, + ), + "measure_key": K.key(line), + } + ) + continue + if not b and isinstance(c, str) and c.strip().startswith("Total"): + label = c.strip() + slug = { + "Total policy decisions": "package_total_policy_decisions", + "Total spending policy decisions": "package_total_spending_policy_decisions", + "Total tax policy decisions": "package_total_tax_policy_decisions", + }[label] + for col, fy in years.items(): + v = ws.cell(row=row[0].row, column=col).value + out.append( + { + **base_row( + T61, + title=label, + fy=fy, + value=v, + locator=f"Table 6.1 '{label}', {fy} (cell {ws.cell(row=row[0].row, column=col).coordinate})", + table="Table 6.1 Budget 2026 policy decisions (£ million)", + column=fy, + conditions={ + "sign_convention": SIGN, + "scoring_method": "final", + "table": "6.1", + "row_type": "total", + }, + metric=REVENUE, + ), + "measure_key": K.package_key(slug), + "note": "Table footnote 1: totals may not sum due to rounding.", + } + ) + assert seen == set(K.TABLE_61), sorted(set(K.TABLE_61) ^ seen) + + +def stage_t62(wb, out): + ws = wb["Table 6.2"] + years = header_years(ws, 4) + seen = set() + for row in ws.iter_rows(min_row=5): + b, c = row[1].value, row[2].value + if isinstance(b, int) and isinstance(c, str): + line, title = b, " ".join(c.split()) + seen.add(line) + fiscal_head = (row[3].value or "").strip() + for col, fy in years.items(): + v = ws.cell(row=row[0].row, column=col).value + assert isinstance(v, (int, float)), (line, fy, v) + cond = { + "sign_convention": FT_SIGN, + "table": "6.2", + "table_line": str(line), + "row_type": "measure", + } + if fiscal_head: + cond["fiscal_head"] = fiscal_head + out.append( + { + **base_row( + T61, + title=title, + fy=fy, + value=v, + locator=f"Table 6.2 line {line}, {fy} (cell {ws.cell(row=row[0].row, column=col).coordinate})", + table="Table 6.2 Budget 2026 Financial Transactions (cash impact, £ million)", + column=fy, + conditions=cond, + metric=FT, + ), + "measure_key": K.PREFIX + K.TABLE_62[line], + } + ) + assert seen == set(K.TABLE_62), seen + + +def stage_t64(wb, out): + ws = wb.active + years = header_years(ws, 3) + group = None + seen = set() + for row in ws.iter_rows(min_row=4): + f, g = row[5].value, row[6].value + if isinstance(f, str) and f.startswith("Measures announced at"): + group = f.strip() + continue + if isinstance(f, int) and isinstance(g, str): + line, title = f, " ".join(g.replace("\xa0", " ").split()) + seen.add(line) + event, mkey = K.TABLE_64[line] + head = row[7].value.strip() + for col, fy in years.items(): + v = ws.cell(row=row[0].row, column=col).value + assert isinstance(v, (int, float)), (line, fy, v) + out.append( + { + **base_row( + T64, + title=title, + fy=fy, + value=v, + locator=f"Table 6.4 line {line}, {fy} (cell {ws.cell(row=row[0].row, column=col).coordinate})", + table="Table 6.4 Measures announced at Spring Statement 2026 or earlier that will take effect from 29 October 2026 or later (£)", + column=fy, + conditions={ + "sign_convention": SIGN, + "scoring_method": "final", + "table": "6.4", + "table_line": str(line), + "head": head, + "row_type": "recosting", + "recosting_of": event, + "announced_at": group, + }, + metric=REVENUE, + ), + "measure_key": mkey, + "note": ( + "A recosting at Budget 2026 of a measure announced at an " + "earlier event; measure_key is that event's registry key. " + "The table's units header prints '(£)', but its " + "footnotes and magnitudes are those of Table 6.1 " + "(£ million), which this row assumes." + ), + "parse_confidence": "medium", + } + ) + assert seen == set(K.TABLE_64), seen + + +def write_gz(rows, path): + buf = io.BytesIO() + with gzip.GzipFile(fileobj=buf, mode="wb", mtime=0, filename="") as gz: + for r in rows: + gz.write( + (json.dumps(r, sort_keys=True, ensure_ascii=False) + "\n").encode() + ) + path.write_bytes(buf.getvalue()) + + +def main(argv=None): + import openpyxl + + ap = argparse.ArgumentParser(description=__doc__.splitlines()[0]) + ap.add_argument("--t61", required=True) + ap.add_argument("--t64", required=True) + args = ap.parse_args(argv) + for path, src in ((args.t61, T61), (args.t64, T64)): + got = sha256(path) + if got != src["sha256"]: + raise SystemExit(f"{path}: sha256 {got} != pinned {src['sha256']}") + rows = [] + wb = openpyxl.load_workbook(args.t61, data_only=True) + stage_t61(wb, rows) + stage_t62(wb, rows) + stage_t64(openpyxl.load_workbook(args.t64, data_only=True), rows) + FAMILY.mkdir(parents=True, exist_ok=True) + write_gz(rows, FAMILY / "claims_staged.jsonl.gz") + manifest = [ + { + "url": src["url"], + "title": src["title"], + "date": PUBLISHED, + "publisher": "HM Treasury", + "sha256": src["sha256"], + "bytes": src["bytes"], + "doc_type": "xlsx", + "access": "direct", + "retrieved": RETRIEVED, + "note": ( + f"Attachment of {COLLECTION}. Downloaded {RETRIEVED} and again " + f"{REVERIFIED} with a browser User-Agent; both downloads hash to " + "the pinned sha256." + ), + } + for src in (T61, T64) + ] + (FAMILY / "manifest.jsonl").write_text( + "".join(json.dumps(m, ensure_ascii=False) + "\n" for m in manifest) + ) + by = {} + for r in rows: + t = r["conditions"]["table"] + by[t] = by.get(t, 0) + 1 + print(f"staged {len(rows)} rows: {by}") + + +if __name__ == "__main__": + main(sys.argv[1:]) diff --git a/pipeline/validate_budget_2026_registry.py b/pipeline/validate_budget_2026_registry.py index 9786ea3fc..a3c561787 100644 --- a/pipeline/validate_budget_2026_registry.py +++ b/pipeline/validate_budget_2026_registry.py @@ -74,11 +74,44 @@ def key_prefix(reg): def _paths_of(m): """Every parameter path a measure asks the engine for.""" - return list((m.get("pe_reform_delta") or {}).keys()) + list( - (m.get("pe_baseline_modifier") or {}).keys() + return ( + list((m.get("pe_reform_delta") or {}).keys()) + + list((m.get("pe_baseline_modifier") or {}).keys()) + + list((m.get("pe_modifier") or {}).get("parameter_legs") or []) ) +def _variables_of(m): + """Every engine variable a dashboard modifier replaces or reads (AB2026).""" + mod = m.get("pe_modifier") or {} + return list(mod.get("replaced_variables") or []) + list( + mod.get("reads_variables") or [] + ) + + +SHA256 = re.compile(r"[0-9a-f]{64}") +RECEIPT_FIELDS = ("repo", "commit", "path", "receipt_file_sha256", "reform_sha256") + + +def _modifier_errors(k, mod): + """An executed_by_dashboard measure names the code and the published + receipt it ran under, so a counterpart can be traced to the exact run.""" + errors = [] + if not mod: + return [f"{k}: executed_by_dashboard without a pe_modifier"] + for field in ("module", "function"): + if not (mod.get(field) or "").strip(): + errors.append(f"{k}: pe_modifier missing {field}") + rec = mod.get("receipt") or {} + for field in RECEIPT_FIELDS: + if not rec.get(field): + errors.append(f"{k}: pe_modifier.receipt missing {field}") + for field in ("receipt_file_sha256", "reform_sha256"): + if rec.get(field) and not SHA256.fullmatch(rec[field]): + errors.append(f"{k}: pe_modifier.receipt.{field} is not a sha256") + return errors + + def load(path=REGISTRY): return json.loads(Path(path).read_text()) @@ -114,10 +147,23 @@ def validate(reg): errors.append(f"{k}: missing title") deltas = m.get("pe_reform_delta") + construction = m.get("construction") + if construction == "executed_by_dashboard": + errors.extend(_modifier_errors(k, m.get("pe_modifier"))) + if construction == "package_of_registry_measures" and not m.get("package_of"): + errors.append(f"{k}: a package names what it is a package of") if m["computability"] == "expressible": - if not deltas: + # A measure the dashboard executed as a modifier, or a package of + # expressible legs, has no single delta to carry; the receipt or + # the legs stand in for it. + if not deltas and construction not in ( + "executed_by_dashboard", + "package_of_registry_measures", + ): errors.append(f"{k}: expressible but carries no reform delta") - if not m.get("engine_baseline_2026"): + if construction != "package_of_registry_measures" and not m.get( + "engine_baseline_2026" + ): errors.append( f"{k}: expressible but records no engine baseline to " "check the registry against" @@ -182,6 +228,24 @@ def validate(reg): "the pre-measure world it executes as pe_baseline_modifier" ) + by_key = {m.get("measure_key"): m for m in reg["measures"]} + for m in reg["measures"]: + if m.get("construction") != "package_of_registry_measures": + continue + # AB2025's totals describe their scope in words; a list names keys. + legs = m.get("package_of") + for leg in legs if isinstance(legs, list) else []: + if leg not in by_key: + errors.append(f"{m['measure_key']}: leg {leg} is not in the registry") + elif ( + m["computability"] == "expressible" + and by_key[leg]["computability"] != "expressible" + ): + errors.append( + f"{m['measure_key']}: expressible package with a " + f"{by_key[leg]['computability']} leg {leg}" + ) + # The whole reported package must be represented, including the parts # this model cannot touch — a registry that silently listed only the # modellable half would overstate coverage. @@ -225,9 +289,14 @@ def resolve_against_engine(reg, year=2026): f"policyengine-uk {installed} installed, registry pinned to {pin}" ) - params = policyengine_uk.CountryTaxBenefitSystem().parameters + system = policyengine_uk.CountryTaxBenefitSystem() + params = system.parameters problems, checked = [], 0 for m in reg["measures"]: + for name in dict.fromkeys(_variables_of(m)): + if name not in system.variables: + problems.append(f"{m['measure_key']}: no engine variable {name}") + checked += 1 for path in dict.fromkeys(_paths_of(m)): try: node = _resolve(params, path) @@ -288,7 +357,9 @@ def main(argv=None): print(" computability:", by) if args.resolve: checked = resolve_against_engine(reg) - print(f" resolved {checked} parameter paths against the certified engine") + print( + f" resolved {checked} parameter paths and variables against the pinned engine" + ) return reg diff --git a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json new file mode 100644 index 000000000..e12e3861d --- /dev/null +++ b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json @@ -0,0 +1,16 @@ +{ + "uk_hmt_tables": { + "rows": 762, + "primaries": 2, + "claims_sha256": "c71472c6957ebc95406002f360da47bc344ba0da43d3702ad6f833d47898dd45", + "sources": [ + "hm_treasury" + ] + }, + "_totals": { + "rows": 762, + "restated": 0, + "families": 1, + "pinned_on": "2026-10-29" + } +} diff --git a/sources/harvest-uk-ab2026-2026-10-29/README.md b/sources/harvest-uk-ab2026-2026-10-29/README.md new file mode 100644 index 000000000..199fcdb4a --- /dev/null +++ b/sources/harvest-uk-ab2026-2026-10-29/README.md @@ -0,0 +1,45 @@ +# Harvest: Autumn Budget 2026 scores from every producer (#230) + +Staged from 2026-10-29, the day after the Budget, from the primary documents. The layout follows the AB2025 harvest (`sources/harvest-uk-ab2025-2026-09-24/`): + +- One family directory per producer group. +- Each family holds: + - `NOTES.md`: access recipe, coverage tally, and what was **not** staged and why; + - `manifest.jsonl`: one line per primary document, with its sha256; + - `claims_staged.jsonl.gz`: one line per claim. +- `COUNTS.json` pins each family's rows, primaries, producer slugs and the sha256 of its claims file. `pipeline/harvest_counts.py` writes and checks it. +- `REGISTRY_KEYS.md` lists every measure key. `pipeline/build_ab2026_registry.py` generates it. +- `SEED_INVENTORY/` lists the publications the harvest starts from, compiled while monitoring reactions on 28 and 29 October. + +Nothing reads this harvest yet. Tranche 2 of #230 adds the ingest. That ingest's exact accounting will read `read = ingested + dropped` against the rows pinned in `COUNTS.json`. `tests/test_uk_ab2026_harvest.py` checks every row against the contract below. + +## Rules + +The AB2025 harvest README's rules 1–10 and its row contract apply unchanged, except for the following: + +1. **`fiscal_event` is the slug `autumn_budget_2026`** on every row, whatever the publication calls it. +2. **Measure keys come from `data/uk/ab2026_measures.json`.** + - `ab2026__…` keys cover announced measures: every HM Treasury Table 6.1 line, Table 6.2 line 4, and the `ab2026__package_…` packages. + - `ab2026_option__…` keys cover options a producer costed. None is registered yet. A row on an unregistered option, or on a package whose scope is not one of the registered packages, has `measure_key: null`. Its proposed key goes in `conditions.proposed_measure_key`, and its NOTES says what the scope is. The registry adds it before ingest. + - Table 6.4 recostings are the one exception. They carry the earlier event's key, named in `conditions.recosting_of`. +3. **A measure's scope is the producer's, not ours.** A producer's figure for "the cost-of-living package" joins `ab2026__package_household_energy_vat_and_fuel_duty` only when its scope is exactly Table 6.1 lines 1, 2 and 3. Otherwise it is a proposal (rule 2). + - For example, RF's package also includes the electricity VAT cut that was announced before the Budget, and its own rows say so. +4. **HMT Impact on households, Figures 1.A and 1.B: no bar readings** (AB2025 rule 9, #61). Values HM Treasury prints as numbers, in text, tables or data labels, are staged as printed. +5. **PolicyEngine's own published figures are not harvested here.** They are tranche 3's counterparts, computed from the dashboard's receipts. A producer that restates a PolicyEngine figure is staged with `attribution: same_assumptions`, as the Fabian Society's row was for AB2025. +6. **Quotes from producers outside the Open Government Licence stay short.** `quote` is the sentence or cell carrying the number, cut to the words that carry it: under 15 words for think tanks, firms and the press. HM Treasury and OBR material is OGL, and its cells and sentences are quoted whole. +7. **A primary that changes is a second primary.** If a producer replaces a document, both versions are hashed in `manifest.jsonl` and rows cite the version they were read from. IFS replaced its cost-of-living deck on 29 October, and HM Treasury replaced the Budget 2026 document PDF after 28 October. NOTES says what changed. + +## Producer slugs and families + +| slug | producer | family dir | +|---|---|---| +| hm_treasury | HM Treasury: Tables 6.1, 6.2 and 6.4 | uk_hmt_tables | +| hm_treasury | HM Treasury: policy costings (static and post-behavioural legs, tax bases, behavioural parameters) | uk_hmt_costings | +| hm_treasury | HM Treasury: Budget 2026 document, Impact on households, Pension Tax Allowance explainer | uk_hmt_redbook | +| obr_efo | OBR: Economic and fiscal outlook, October 2026, and its detailed tables | uk_obr_tables | +| resolution_foundation | Resolution Foundation | uk_rf_ab2026 | +| ifs | Institute for Fiscal Studies | uk_ifs_ab2026 | +| tax_policy_associates, centax, cpag, ippr, niesr, fraser_of_allander, scottish_fiscal_commission, cps, institute_for_government, age_uk, … | other producers, each in the AB2025 family of the same name where one exists | uk_tpa, uk_centax, uk_cpag, uk_ippr, uk_niesr, uk_fai_sfc, uk_thinktank_misc, … | +| commercial firms and platforms | worked examples (mode 3, #63) | uk_cases_commercial | + +Macro calls by banks and City forecasters belong to the #55 lane and are not staged here (AB2025 rule 10). diff --git a/sources/harvest-uk-ab2026-2026-10-29/REGISTRY_KEYS.md b/sources/harvest-uk-ab2026-2026-10-29/REGISTRY_KEYS.md new file mode 100644 index 000000000..46f33ae28 --- /dev/null +++ b/sources/harvest-uk-ab2026-2026-10-29/REGISTRY_KEYS.md @@ -0,0 +1,112 @@ +# Autumn Budget 2026 registry keys + +Generated by `pipeline/build_ab2026_registry.py` from `data/uk/ab2026_measures.json`; do not edit. +Every staged row's `measure_key` is one of these, or (Table 6.4 recostings only) a key from the +earlier event's registry named in `conditions.recosting_of`. + +| key | Table 6.1 line | computability | title | +|---|--:|---|---| +| `ab2026__household_energy_payment_100_bands_a_to_d` | 1 | expressible | Household Energy Payment: £100 council tax rebate for bands A-D from January 2027 | +| `ab2026__electricity_vat_zero_rate_to_july_2027` | 2 | expressible | Support for Energy Bills: Cut VAT on domestic electricity from 5% to 0% from 1 October 2026 to 1 July 2027 | +| `ab2026__fuel_duty_5p_cut_extended_and_rpi_rise_deferred` | 3 | expressible | Fuel Duty: Extend the temporary 5p-per-litre duty cut and defer the planned Retail Prices Index (RPI) increase to 1 July 2027 | +| `ab2026__carers_allowance_earnings_taper` | 7 | expressible | Carer’s Allowance: Introduce a taper on earnings above the Weekly Earning Limit within 2030-31 | +| `ab2026__trading_allowance_3000` | 15 | expressible | Trading Allowance: Increase the Income Tax Trading Allowance to £3,000 from £1,000 from 6 April 2027 | +| `ab2026__pension_tax_allowance` | 52 | expressible | Pension Tax Allowance: Remove the income tax liability for pensioners with total income up to the value of the new State Pension from 6 April 2027 until 5 April 2030 | +| `ab2026__property_income_rates_plus_2pp` | 61 | expressible | Property Income: Increase tax rates on property income by 2ppts at the basic, higher and additional rate from 6 April 2027 | +| `ab2026__savings_income_rates_plus_2pp` | 62 | expressible | Savings Income: Increase tax rates on savings income by 2ppts at the basic, higher and additional rate from 6 April 2027 | +| `ab2026__badr_lifetime_limit_500000` | 60 | expressible | Business Asset Disposal Relief (BADR): Reduce the lifetime limit to £500,000 from 6 April 2027 | +| `ab2026__bus_fare_cap_2_england_outside_london_2027` | 4 | partial | Bus Fares: Cap single bus fares at £2 in England outside London for one year from 1 January 2027, funded by reprioritising DESNZ budgets | +| `ab2026__mileage_allowance_payments_55p` | 5 | not_expressible | Mileage Rates: Increase the tax-free and tax-deductible car and van rates to 55p for the first 10,000 business miles from 6 April 2026 and the higher mileage rate to 35p from 6 April 2027 | +| `ab2026__parental_allowance_fathers_partners_adopters` | 6 | not_expressible | Parental Pay: Establish a Parental Allowance for fathers, partners and adopters in Great Britain from April 2029 | +| `ab2026__uc_surplus_earnings_threshold_2500_extended` | 8 | not_expressible | Universal Credit: Extend the £2,500 surplus earnings threshold for one year from 1 April 2027 | +| `ab2026__cash_isa_limit_held_at_20000` | 11 | not_expressible | Cash ISAs: Maintain the cash ISA limit at £20,000 from 6 April 2027 | +| `ab2026__founders_relief_cgt` | 26 | not_expressible | Founders' Relief: Introduce a new CGT relief for qualifying founders of high-growth, innovative companies from 6 April 2027 | +| `ab2026__first_time_buyer_isa` | 38 | not_expressible | First Time Buyer ISA: Introduce a new ISA for first-time buyers, replacing the Lifetime ISA for new savers from 6 April 2029 | +| `ab2026__lifetime_isa_withdrawal_charge_20pc_above_cap` | 39 | not_expressible | Lifetime ISA: Reduce the withdrawal charge from 25% to 20% for first-time buyers purchasing properties above the £450,000 price cap from 29 October 2026 | +| `ab2026__pension_death_benefits_two_year_rule_removed` | 53 | not_expressible | Pensions Tax: Remove the two-year rule for taxation of pension and lump sum death benefits from 28 October 2026 | +| `ab2026__vat_5pc_family_days_out_summer_2026` | 66 | not_expressible | Great British Summer Savings Scheme: Reduce VAT to 5% on family days out from 25 June 2026 to 1 September 2026 | +| `ab2026__free_bus_travel_children_august_2026` | 67 | not_expressible | Bus Travel: Fund free bus travel for children aged 5-15 in England from 1 August 2026 until 31 August 2026 | +| `ab2026__student_loan_interest_cap_6pc_plan_2_and_3` | 83 | not_expressible | Student Loans: Cap interest rates for Plan 2 and Plan 3 student loans at 6% from 1 September 2026 to 31 August 2027 | +| `ab2026__company_car_tax_rates_2030_31` | 86 | not_expressible | Company Car Tax: Set the Benefit-in-Kind rates (appropriate percentages) for electric vehicles, hybrids and internal combustion engine vehicles for 2030-31 | +| `ab2026__hvcts_deferrals_discounts_exemptions` | 93 | not_expressible | High Value Council Tax Surcharge: Confirm eligibility for deferrals, discounts and exemptions from 1 April 2028 | +| `ab2026__early_years_books_access` | 9 | not_expressible | Early Years: Fund increased access to books for children aged 0 to 5 | +| `ab2026__food_and_heating_oil_tariff_suspensions` | 10 | not_expressible | Tariff Suspension Packages: Suspend tariffs on 135 food products from 21 June 2026 until 31 December 2028; 101 food products and kerosene for heating oil from 26 November 2026 until 31 December 2028; and fertiliser from 26 November 2026 until 31 December 2027 | +| `ab2026__business_rates_pubs_relief_35pc` | 12 | not_expressible | Business Rates: Increase business rates relief for pubs, social clubs and live music venues by 20ppts to 35% from 1 April 2027 | +| `ab2026__business_rates_sbrr_rhl_thresholds_raised` | 13 | not_expressible | Business Rates: Increase valuation thresholds for Small Business Rate Relief for retail, hospitality and leisure (RHL) properties from £12,000 to £18,000 for 100% relief and from £15,000 to £21,000 for tapered relief from 1 April 2027 | +| `ab2026__business_rates_rhl_transitional_support` | 14 | not_expressible | Business Rates: Introduce a transitional support package, including a two-year real terms bills freeze for RHL businesses and a one-year extension of the 2023 Supporting Small Business support, part funded by extending the Transitional Relief Supplement for two years from 1 April 2027 | +| `ab2026__business_tariff_suspensions` | 16 | not_expressible | Business Tariff Suspensions: Suspend tariffs on 72 products from 5 August 2026 and extend suspensions for 361 products in 2027-28 | +| `ab2026__climate_change_levy_rates_2028` | 17 | not_expressible | Climate Change Levy: Freeze electricity & liquefied petroleum gas main rates, and uprate gas and solid fuels main rates by RPI from 1 April 2028 | +| `ab2026__red_diesel_duty_cut` | 18 | not_expressible | Red Diesel: Reduce duty rates by 3.7p-per-litre from 15 June 2026 to 1 July 2027 | +| `ab2026__alcohol_duty_small_producer_relief` | 19 | not_expressible | Alcohol Duty: Increase the generosity of Small Producer Relief from 1 February 2027 | +| `ab2026__good_growth_funds` | 20 | not_expressible | Good Growth Funds: Seed funding for Mayors and local leaders in England and £1.5bn finance for Mayoral Strategic Authorities with a mechanism for subsidised lending (resource impacts of subsidy) | +| `ab2026__business_rates_retention_mayoral_authorities` | 21 | not_expressible | Business Rates Retention: Devolve a share of business rates income and growth to Mayoral Strategic Authorities from April 2027 | +| `ab2026__local_apprenticeship_services` | 22 | not_expressible | Youth Employment and Skills: Expand local apprenticeship services to all current mayors | +| `ab2026__regional_growth_investments` | 23 | not_expressible | Regional Growth: Fund priority investments and local growth & regeneration projects across the UK | +| `ab2026__cambridge_development_corporation` | 24 | not_expressible | Cambridge Development Corporation: Pilot a new government financing model for land, housing and infrastructure development from April 2027 (capital grant element) | +| `ab2026__strategic_authorities_capacity_funding` | 25 | not_expressible | Strategic Authorities: Increase funding to build capacity to deliver new powers and investment responsibilities | +| `ab2026__rd_relief_claim_notification_form_removed` | 27 | not_expressible | R&D Tax Reliefs: Remove the Claim Notification Form from 1 April 2027 and simplify the Additional Information Form from Autumn 2027 | +| `ab2026__emi2_option_limit_120000` | 28 | not_expressible | Enterprise Management Incentives 2 (EMI2): Increase the individual option limit to £120,000, extending the exercise period to 15 years and remove interaction with EMI option limit from 6 April 2027 | +| `ab2026__british_business_bank_backing_innovation` | 29 | not_expressible | Backing Innovation: Bring forward £1bn investment and increase financial capacity by £2bn for the British Business Bank to support additional investment in the UK (resource impacts of increased lending) | +| `ab2026__national_wealth_fund_capital` | 30 | not_expressible | National Wealth Fund: Invest additional capital to support growth, including £2bn for the new Energy Wealth Fund (resource impacts of increased lending) | +| `ab2026__artificial_intelligence_package` | 31 | not_expressible | Artificial Intelligence: Strengthen the UK’s AI security, accelerate adoption across the economy, and support strategically important UK AI capabilities | +| `ab2026__ev_charging_public_chargers` | 32 | not_expressible | EV Charging: Reduce the cost of using local, lower-powered public chargers in Great Britain | +| `ab2026__rough_sleeping_investment` | 33 | not_expressible | Rough Sleeping: Invest in a route off the streets across the Parliament, alongside longer-term accommodation and intensive support | +| `ab2026__rough_sleeping_reprioritised_mhclg_budgets` | 34 | not_expressible | Rough Sleeping: Reprioritise existing MHCLG budgets to tackle rough sleeping | +| `ab2026__social_and_affordable_homes_programme` | 35 | not_expressible | Social and Affordable Homes Programme: Fund delivery of around 10,000 social and affordable homes | +| `ab2026__vat_zero_rate_land_for_social_housing` | 36 | not_expressible | Social Housing: Introduce a zero rate of VAT for land sold to registered social housing providers (relevant housing associations) from 1 April 2027 | +| `ab2026__your_first_home_equity_loan` | 37 | not_expressible | Your First Home: Introduce a new equity loan scheme to support around 70,000 first-time buyers from April 2027 (net income from developer fees) | +| `ab2026__defence_investment_plan_further_savings` | 40 | not_expressible | Defence Investment Plan (DIP): Commitment to further savings made at announcement on 30 June 2026 | +| `ab2026__defence_investment_plan_capital_reallocation` | 41 | not_expressible | Defence Investment Plan: Reallocate capital spending to fully fund the Defence Investment Plan | +| `ab2026__defence_admin_savings_reinvested` | 42 | not_expressible | Defence: Reinvest defence and security administration savings into frontline defence and national security | +| `ab2026__ukraine_winter_air_defence` | 43 | not_expressible | Ukraine: Emergency winter support for air defence | +| `ab2026__uk_export_finance_defence_capacity` | 44 | not_expressible | UK Export Finance: Increase capacity to support UK defence exports from June 2026 (increased income from additional activity) | +| `ab2026__hmrc_tax_gap_package` | 45 | not_expressible | HMRC: Measures to close the tax gap4 | +| `ab2026__online_marketplace_liability_uk_businesses` | 46 | not_expressible | Online Marketplace Liability: Extend to UK businesses with a minimum annual sales threshold per platform of £90,000 from 1 October 2027 | +| `ab2026__dwp_fraud_and_error_prevention` | 47 | not_expressible | DWP Fraud and Error: Expand use of data and targeted controls to prevent fraud from entering the benefits system4 | +| `ab2026__benefits_removed_hospitalised_serious_offenders` | 48 | not_expressible | Welfare Reform: Remove benefit entitlements from serious offenders hospitalised in Great Britain from 6 April 2027 | +| `ab2026__distributions_framework_modernised` | 49 | not_expressible | Distributions: Modernise the distributions framework from 6 April 2028 | +| `ab2026__advance_corporation_tax_abolished` | 50 | not_expressible | Advance Corporation Tax (ACT): Abolish Advance Corporation Tax from 1 April 2031 | +| `ab2026__vat_reverse_charge_overseas_services_restricted` | 51 | not_expressible | Cross-Border Services: Restrict the reverse charge on supplies of services from overseas to VAT-registered businesses from 1 April 2027 | +| `ab2026__foreign_sovereign_investments_taxation` | 54 | not_expressible | Modernising Taxation of Foreign Sovereign Investments: Introduce a new statutory regime for the direct taxation of foreign sovereigns from 1 January 2029 | +| `ab2026__bank_surcharge_6pc_temporary` | 55 | not_expressible | Bank Surcharge: Temporarily increase the Bank Corporation Tax Surcharge rate from 3% to 6%, to fund breathing space measures for households, for 15 months from 1 January 2027 | +| `ab2026__machine_games_duty_higher_stake_25pc` | 56 | not_expressible | Machine Games Duty: Increase the rate from 20% to 25% for higher stake machines from 1 April 2027 and move duty liability to the machine supplier from 1 April 2028 | +| `ab2026__business_rates_vape_shops_and_agcs_reliefs_removed` | 57 | not_expressible | Business Rates: Remove reliefs for vape shops and Adult Gaming Centres from 1 April 2027 | +| `ab2026__writing_down_allowance_12pc_and_fya_50pc` | 58 | not_expressible | Writing-down Allowances: Reduce the main rate writing-down allowance to 12% from April 2027 and increase the main rate first-year allowance to 50% from 1 January 2027 | +| `ab2026__private_hire_operators_vat_full_fare` | 59 | not_expressible | Private Hire Vehicle Operators: Deem larger operators with annual fare turnover over £50 million liable for VAT on the full value of their fares from 2 January 2028 | +| `ab2026__civil_service_admin_budgets_cut_20pc` | 63 | not_expressible | Future Civil Service: Reduce administration budgets by 20% by 2029-30 compared to 16% reduction in Spending Review 2025 plans | +| `ab2026__civil_service_ai_fund` | 64 | not_expressible | Future Civil Service: Reinvest part of the administration budget savings in a new AI Fund to reduce administration burdens and improve outcomes | +| `ab2026__hgv_ved_1_for_12_months` | 65 | not_expressible | Heavy Goods Vehicles: Reduce Vehicle Excise Duty to £1 for 12 months from 1 July 2026 | +| `ab2026__foreign_permanent_establishment_exemption_mandatory` | 68 | not_expressible | Corporation Tax: Make the foreign permanent establishment exemption mandatory, preventing their losses from reducing UK corporation tax receipts for accounting periods beginning on or after 1 January 2027 | +| `ab2026__electricity_generator_levy_55pc` | 69 | not_expressible | Electricity Generator Levy: Increase the rate from 45% to 55% until 31 March 2028 | +| `ab2026__british_industrial_competitiveness_scheme` | 70 | not_expressible | Industrial Electricity: Fund the British Industrial Competitiveness Scheme starting from 1 April 2027, including a backdated payment for eligible businesses | +| `ab2026__carbon_price_support_removed` | 71 | not_expressible | Carbon Price Support (CPS): Remove the CPS rates to reduce energy bills from 1 April 2028 | +| `ab2026__youth_guarantee_and_apprenticeships_spring_2026` | 72 | not_expressible | Spring 2026 Youth Employment and Skills Package: Expand Youth Guarantee and apprenticeship support | +| `ab2026__union_learning_fund` | 73 | not_expressible | Skills: Re-establish the Union Learning Fund in England | +| `ab2026__investment_for_growth` | 74 | not_expressible | Investment for Growth: Support strategic technologies, local growth and key industries | +| `ab2026__growth_guarantee_scheme_expansion` | 75 | not_expressible | British Business Bank: Expand the Growth Guarantee Scheme to support £6.5 billion of additional SME lending across this Parliament | +| `ab2026__leeds_city_fund` | 76 | not_expressible | Leeds City Fund: Establish the Leeds City Fund with 100% retention of business rates growth | +| `ab2026__uk_india_ceta_and_double_contributions_convention` | 77 | not_expressible | UK-India: Bring into force the Comprehensive Economic and Trade Agreement and Double Contributions Convention from 15 July 2026 | +| `ab2026__steel_tariff_and_quotas` | 78 | not_expressible | Steel Tariff: Change tariff rate and quotas on imported steel from 1 July 2026 | +| `ab2026__securities_transfer_tax` | 79 | not_expressible | Stamp Taxes on Shares: Replace Stamp Duty and Stamp Duty Reserve Tax with the Securities Transfer Tax from late 2027 | +| `ab2026__business_rates_ev_charging_relief` | 80 | not_expressible | Business Rates: Introduce 100% relief for eligible Electric Vehicle Charging Points (EVCPs) and electric vehicle-only forecourts for ten years from 1 April 2026 | +| `ab2026__misogyny_prevention_programmes` | 81 | not_expressible | Tackling Misogyny: Invest in grassroots misogyny prevention programmes | +| `ab2026__infected_blood_compensation_additional_report` | 82 | not_expressible | Infected Blood Compensation Scheme: Implement the Inquiry's Additional Report recommendations from 24 July 2026 | +| `ab2026__waste_crime_landfill_tax_clearances` | 84 | not_expressible | Waste Crime: Fund Landfill Tax costs for six priority illegal waste site clearances until 31 March 2029 and a three-year pilot to fund Local Authorities' Landfill Tax liabilities arising from illegal waste site clearances from 2027 | +| `ab2026__further_priorities_nhs_and_vawg` | 85 | not_expressible | Further Priorities: Fund key priorities, including pressures facing the NHS and delivery of the Violence against Women and Girls Strategy | +| `ab2026__supported_housing_licensing` | 87 | not_expressible | Supported Housing: Introduce a licensing regime to enforce national standards for supported housing from 2027-28 | +| `ab2026__vat_zero_rate_donated_medicines` | 88 | not_expressible | Donated Medicines: Introduce a zero rate of VAT for the donation of medicines under qualifying circumstances from 23 June 2026 | +| `ab2026__vat_medical_deputies_staffing` | 89 | not_expressible | Temporary Staffing: Amend Group 7 of Schedule 9 to the VAT Act 1994 to restore the tax treatment of medical deputies to its previously intended position | +| `ab2026__landfill_communities_fund_credit_cap` | 90 | not_expressible | Landfill Communities Fund: Increase the Landfill Tax credit cap from 1 April 2027 | +| `ab2026__franchised_higher_education_registration` | 91 | not_expressible | Higher Education: Introduce registration requirements for franchise providers with 300 or more students from 1 September 2028, to strengthen quality and oversight of franchised provision | +| `ab2026__high_needs_local_authority_spending` | 92 | not_expressible | Local Authority Spending: Impact of adjustments to high needs budget on local authority spending | +| `ab2026__benefits_in_kind_payrolling_delay` | 94 | not_expressible | Benefits in Kind (BiKs): Delay mandatory payrolling of company cars, vans, fuel and medical benefits from 6 April 2026 to 6 April 2027, with remaining BiKs implemented from 6 April 2028 | +| `ab2026__low_value_imports_customs_relief_removed` | 95 | not_expressible | Low Value Imports: Remove the £135 customs duty relief and introduce new customs arrangements from October 2028 | +| `ab2026__ets_energy_from_waste_delay` | 96 | not_expressible | Emissions Trading Scheme (ETS): Delay the inclusion of energy from waste in the ETS | +| `ab2026__gas_distribution_revenue_recognition_spread` | 97 | not_expressible | Corporation Tax: Spread the one-off impact of revenue recognition changes for qualifying gas distribution activities over five years for accounting periods ending on or after 31 December 2026 | +| `ab2026__us_llc_double_taxation_relief` | 98 | not_expressible | International Investors: Relieve double taxation for UK investors in US Limited Liability Companies and similar entities from early 2028 | +| `ab2026__sovereign_ai_financial_transactions` | 6.2/4 | not_expressible | SovAI: Additional funding for Sovereign AI to invest in and support the UK's promising AI companies | +| `ab2026__package_total_policy_decisions` | | partial | Table 6.1 total policy decisions | +| `ab2026__package_total_tax_policy_decisions` | | partial | Table 6.1 total tax policy decisions | +| `ab2026__package_total_spending_policy_decisions` | | partial | Table 6.1 total spending policy decisions | +| `ab2026__package_household_energy_vat_and_fuel_duty` | | expressible | The Household Energy Payment, zero VAT on domestic electricity and the fuel duty cut, together | +| `ab2026__package_policyengine_seven` | | expressible | The seven household measures PolicyEngine modelled on Budget day, together | diff --git a/sources/harvest-uk-ab2026-2026-10-29/SEED_INVENTORY/publications.md b/sources/harvest-uk-ab2026-2026-10-29/SEED_INVENTORY/publications.md new file mode 100644 index 000000000..5469c1649 --- /dev/null +++ b/sources/harvest-uk-ab2026-2026-10-29/SEED_INVENTORY/publications.md @@ -0,0 +1,62 @@ +# Autumn Budget 2026: publications to harvest + +Compiled 29 October 2026 from the reaction monitor that ran from 28 October 19:05Z to 29 October 14:05Z. Each entry names a publication the monitor read and the kinds of figure it holds. Figures are paraphrased here; the harvest reads each primary itself and stages only what the primary prints. Times are UTC. + +## HM Treasury (28 October, 13:30Z) + +- Budget 2026 document, PDF and HTML. + - PDF: https://assets.publishing.service.gov.uk/media/6ae20b85460c1346f7be784e/E03668986_HMT_Budget_2026_ELAY__2_.pdf + - The PDF was replaced after 28 October: the 28 October download hashes to bed70546…, the 29 October download to 19b337ac…. + - Holds measure descriptions with numbers, for example the Household Energy Payment's "around 20 million households". +- Supporting documents (https://www.gov.uk/government/publications/supporting-documents-for-budget-2026): + - Table 6.1 and 6.2 (staged in `uk_hmt_tables`); + - Table 6.4 recostings (staged); + - Policy costings (AB26_Policy_Costing.pdf); + - Impact on households; + - Data Sources. +- Pension Tax Allowance explainer (https://www.gov.uk/government/publications/pension-tax-allowance-explainer/pension-tax-allowance-explainer). Holds counts of pensioners who gain and who pay no tax, plus worked cases. + +## OBR (28 October) + +- Economic and fiscal outlook, October 2026, with its charts-and-tables workbooks and the detailed forecast tables (the policy workbook). + - Chapter 3 costs each measure and states the behavioural adjustments, for example paragraph 3.17 on the savings and property income tax rises. + - The fuel duty path is in paragraph 3.19. + +## Resolution Foundation + +- Press release, 28 October: the Budget balancing act and a four-year living standards downturn. +- Press release, 29 October, 06:00Z. +- *Spread thin*, published 29 October at 08:20Z: https://www.resolutionfoundation.org/publications/spread-thin/ (PDF …/app/uploads/2026/10/Spread-thin-1.pdf). + - Holds: + - VAT cut per typical household; + - Energy Payment reach by half of the income distribution; + - fuel duty timing and cost; + - gas standing charge gainers and losers; + - the Carer's Allowance taper count; + - package support by half of the income distribution and by vigintile; + - poverty over the Parliament. +- A decile chart of Energy Payment eligibility, shown in the Guardian live blog, 28 October 15:30Z. The harvest should find RF's own copy. + +## Institute for Fiscal Studies + +- Initial response, 28 October: https://ifs.org.uk/articles/autumn-budget-2026-initial-response +- Post-Budget briefing, 29 October, 10:30Z to 12:00Z. Four decks: + - Tom Wernham, "Cost of living measures". The first version, TW_0, was replaced at 12:57Z by TW_1, with no figure changed. + - Isabel Stockton, fiscal. + - Bee Boileau, public services ("Spending on public services: the Achilles heel"). + - Stuart Adam, "Tax measures". +- Any data items or chart data IFS posted with the briefing. + +## Others + +- **Tax Policy Associates**, "Budget 2026 – what it means" (taxpolicy.org.uk, 28 October). Covers the Pension Tax Allowance design, its taper and marginal rates, and other measures. +- **CenTax**, initial reaction and LinkedIn post, 28 October. +- **CPAG**, Budget response, 28 October; child benefit item. +- **NIESR**, "Breathing space" blog, 28 October. +- **Fraser of Allander Institute**, UK Budget reaction, 28 October. +- **Centre for Policy Studies**, verdict, 28 October. +- **Institute for Government**, "underwhelming Budget", 28 October. +- **Age UK**, Budget response, 28 October (Pension Tax Allowance paragraph). +- **IPPR** (three pieces), **Citizens Advice**, **Trussell**, **Social Market Foundation**. +- **Welsh Government**, written statement on the Budget, 28 October: consequentials for Wales. +- **MoneySavingExpert** restates government figures; stage them as `restated`, if at all. diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/NOTES.md new file mode 100644 index 000000000..84abeae34 --- /dev/null +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/NOTES.md @@ -0,0 +1,47 @@ +# uk_hmt_tables: HM Treasury Budget 2026 Tables 6.1, 6.2 and 6.4 + +Staged 2026-10-29 by `pipeline/stage_ab2026_hmt_tables.py`, which is vendored, so the staging re-runs byte for byte. Source slug `hm_treasury`, source_model `hmt_costing_obr_certified`. + +- **Rows:** 762. + - Table 6.1: 98 measure lines and 3 total rows, each across 6 fiscal years (606 rows). + - Table 6.2: 9 lines across 6 fiscal years (54 rows). + - Table 6.4: 17 lines across 6 fiscal years (102 rows). +- **Primaries:** 2 workbooks. + +## Access recipe + +The two XLSX attachments of [Supporting documents for Budget 2026](https://www.gov.uk/government/publications/supporting-documents-for-budget-2026), listed through the gov.uk content API (`/api/content/government/publications/supporting-documents-for-budget-2026`). + +- **First download:** 2026-10-28, `curl -sSL` with a browser User-Agent. +- **Second download:** 2026-10-29. Both hash to the pinned sha256, which the stager checks before writing anything. +- **Reading:** openpyxl with cached values (`data_only=True`), so every value is the number HM Treasury stored, not a recomputed formula. + +## What is staged + +- **Table 6.1:** every cell of every measure line, 2026-27 to 2031-32. + - Fields: `metric: revenue_change`, `unit_concept: gbp`, value = cell × 1e6. + - Conditions carry the line (`table_line`), the Tax/Spend head (`head`), the fiscal head where printed (`fiscal_head`: RDEL, TDEL, RAME, SUBTOTAL, Tax, Tax/TME, …), and the run of headings above the line (`heading`). + - The table marks no hierarchy beyond adjacency. Two headings with no line between them are recorded as "A > B". + - The three total rows carry the package keys `ab2026__package_total_policy_decisions`, `…_total_tax_policy_decisions` and `…_total_spending_policy_decisions`. + - The lines don't sum exactly to the totals (by £5m to £10m a year). Footnote 1 says totals may not sum due to rounding, and the rows are staged as printed. +- **Table 6.2:** financial transactions, `proposed_metric: financial_transaction_cash_impact` (cash impact on the public sector, not borrowing). + - Each line carries the measure key of the Table 6.1 measure it belongs to. + - The exception is line 4 (Sovereign AI), which has no Table 6.1 line and so has its own key. +- **Table 6.4:** recostings at Budget 2026 of measures announced earlier. + - Each row carries **that event's** registry key, with `conditions.recosting_of`: + - lines 1–14: `autumn_budget_2025` (`data/uk/ab2025_measures.json`); + - lines 15–16: `autumn_budget_2024`; + - line 17: `spring_budget_2024` (`data/uk/events/*_measures.json`). + - `parse_confidence: medium`: the table's units header prints "(£)", but its magnitudes and footnotes match Table 6.1's £ million, which these rows assume. + +## Conventions (load-bearing) + +- **Sign.** The tables print no legend. `conditions.sign_convention` on every Table 6.1 and 6.4 row reads "positive = yield to the Exchequer (reduces borrowing); negative = cost to the Exchequer". The convention is observed from the rows themselves: the bank surcharge rise is positive, the Household Energy Payment negative. The AB2025 Table 4.1 harvest uses the same convention. Values are NOT re-signed. +- **scoring_method `final`.** Table 6.1 is the certified post-behavioural total. The static and behavioural legs are in the policy costings document (family `uk_hmt_costings`). +- **period** is the FY end year (2026-27 → 2027), with `conditions.fy` holding the label. +- **geography** is "UK" on every row: these are Exchequer impacts. A measure's own scope (for example England only for the Household Energy Payment) is in its title, verbatim in `reform_hint`, and in the registry. +- **quote** is the line's title and the cell, as printed (Open Government Licence). + +## Not staged + +- Table 6.1's footnote 2 names two measures with "no net increase to borrowing" that the table excludes: debt relief for councils in severe financial distress through the PWLB, and a landlord registration fee for the private rented sector. They have no figures, so there's nothing to stage. The registry doesn't list them. diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/claims_staged.jsonl.gz new file mode 100644 index 0000000000000000000000000000000000000000..d4a027ccda2c4d1cebf20a1dbfa3bae0c1da196d GIT binary patch literal 38260 zcmXt4B`~!@ofM`+a&pX+L#X8BD5p^sOK<6jyRpj!;gA_=2J|;Pm@Nd6&9%M8DWV ze1h<^NB~NdWB@LNajB)BuWU9F+A(a&T(*=9Ygv^&F%q#ns2St3Vce)x@e zn)hd0d3mFu-2#c}xpf)P`ZRKBC*OD9)`%;&*s?77G;p$?PGof)(5{Tw$k zk6Lw9>?6u<&8gZT(F3H_srk~1Z8LJ+o3qF8g(H1;3Ey}IZu@~N|5DEO;oRGLkC%Jh zUEw*;@r6qZ6~36ViXB7QT2AVUL*&@XJagKEu0CE3@^G?o(YmzI5r9l#*8a_YXSjD| z5@AL@Vb7Ke{)ntjzkJ7#!J{2NfW34fSZU9`miYR3Xyfo3S4LDL7gH*3UE z{ovbYjly7(Dm`Rf@IEpe?q0xY^)@xo(pe5X_y;txbiQK%sIZ2KCANP>mZ<;zsfmbaYPSV* zH6{AXR*L2Ew~e`l9PvtQlu|;TgjXff+pZcLA3lBL)e2*0l$I~q`oUv-DB!s5f4Cw$O`S@xmc!&{a0{`814tr;znmaCQ zJtSQEcCGwq_3tFm+xItfn%K{lt>QG)S^uy`3n7$=ZS%O&uR&`qLl`CCFau!L+v_-BcyJ~#bl{%czjlJpaYPJM*jgXn ze27Xy#=?shi>6{a7B1@|dOfeqqE`Fm3rI{72T5}w8&#}botu-8*G0l3d12=Ipl*HV z@;0B1*k-Qltj=45YYzgwlw613Upn7boU)@vsGtO(<+z~O4$z7m&=Lh+zY3hv$r*D1 z!vufoYmD_3^xnw*M&jIH;k2jdqPQ5s&x)71q!O18oyhW^Sld!`QMA2OAM5KX6nx0Y zaD&=FeK3lWwVkBm-=(C!e1v;Aq2o8^xW&ktlSc4L3zW6B+v574jZE@`QeIKtK0}wO z?qzXHapgl>dZ@o_;KgBv8~w#sFNcA`mH%_0xmeQl8n!v2<1bCcdN^U=pYZkdxLg6L zTN2)rJbm%2fhuZ6&b|x?5mJ&4Te%>o?s*TZO%Un~`N8%+U}BG}Qu`Z3U9lo(4>ftM z=87FFV1xg&DS1kjr!Dirun)rgy^TTD-dw+FDbZ(sd8vig^L;1qk~1=U$EV+Fc8r z4;H}tK3psqUHXDbt4EiML~YRL4%n1VKZXrUW-kh0kE`Yrv{z){c`vhF^bow{qTDh8 z5nNzHc6oQ*>F9SC+$csfOFU~rmtaq6O-_)>^xd~gtU73GKtcJ;4{OU`OlFoi*7rsI zulaOnMvCU1enAM!0S)9c^QQ?7D=^Ie#j?kL#A1ywCJAcBn;oC#oqC}%^ehFb9ar`q z7JS)`h6dX6D#n{N$OySlZqgQPj2Wzye(^+fcK>uj5Drd83)tfNvq2RphV4bo9_A+b zB>~>@<@W3C=@1Dq>sRcOfGs6pdDySpqgxV}YTkZ1yJpC+6fPJbn3%+Ij%^>*)D zs6VSp?7D3eCU*m^ze~jV_&uRbk0M;65>NQPjO($fS)-K(%BZ(na|h1hsc8>SZ0qUq zmM%R^&SKMHl0rj0O#WjqN-2Of9-0v_rCe1{`l`#ca>7|b{D_=T<0>ArI=y)?8C+eD z`t{4SoV^V}*8lZeEa(Wg!en4~xkrauqh%1A3Xj8NK$`N(4irvtJN3Fb*T=9}wNN?a>NaIV091>90<70! znt+g1UbIPBHV&6@J+c+@Ve@EEXuj1MO1}|)r5V6oE@V|8q+xP1i(;p4)rU#5f&R|& zIgD3g&|9CEUx}@Zf zqAjfSaN3O_Eg-LK?``3bbCm=@Es{g7tixIzWpK8H^_o>n{1jnT>#64t;JtaR-+gOZ zzZt+6cv6$-oFb-Jlehi$t^=mCSJ*z^9^wr*8-MC&!MyQD9RO--__C$w@E431H4Pdq{>7^saf)ll8fvuq|2m zcJJa3-iWVVkiXZkuvvrH9f(CEqN%JnCDllGJ0BzOOC>f=H97|g}P*=Il@fji{YC+kgA zT(6N2Vyok1xWU~&`qTM8LXGO=8K6(-?tkkpSp1OHl)38;neGf(L!-F|of*A+^&ivB z++|yVAfV&dCQ$i<(>}OiA6g58Xt2U66%zq43cm(yclWQk8-$w;e6IQQiNPCRblh8m zv5-HaR*PS`Z0BdLdJFvlVG@Hh78@pc)8O~0vF~e1kd&nFN^OJ2C)2SZ603ZUDt}D4 za*8gzNJ<`*3Uxq55fxcGtRnI93L-bjI0dc;=~k8Gi^i~C$J^B|kP!Rzeuyu*tb7-i zUwr~)E;ThjLA>CE>7JhAhJMUo?aB|L*6te%C@(VHAltDzSRX`C<@|PJ9|L9%nEEAG zg!jS{)`tgmSsajq|G4!}OvL_& z4{oRde@z3BQjE(rdl3b#6`0#pU=0Nd%58Eg9}{81{ki-Ltv9Lx6SVD3nK^wzV(tlR zfFnv{QJn!pcRA)j?6HcL%Qtzi_QW60bnpAN@IHPXOOA<&u1+0HM=bj%Gy7XffV@cZ zVWf`hO*E%ZXA?TZ4a=5ujWpaPaqP@E7!W@3Wm4Ps=Wo+56GYS@7R>U2GNBrd8G=f} zAx~8^t30_q`a!)2xdBBOW+U4L4G6h7LbJhayKCRkHJIy+zM+fMcZm>i0F1uPeap4D zUi^~la2a7uLVaTJEs;qcNFSp)MilTacjYmABY|+ccZT{e)!FQm zP|F5{oMum5!ZCMh({CNQOsTt5LcDqM>1c(lB*sm}SdYiD_g_V`DM@b#oL#?{6r?10 zh>yQ{h;T=Geaz0}REbH1R}YjK(zp%~R3RNKDc?XfHG|M8m%HCj6mBP{uNXR;QZ@3M^npsKF7O56<7rj71-2 zOwf-(H>#H=L`r*N9JKhPF6G|zZe&b~q$#R$&_h&;Z{*DL!xDfGqaI0AiV=)RM1yIb z!ls_p>4x~>n20QM7n>L~yrmME!rLZ2bMLOEFlfWzHhmM|t#_jq(Gk;4{-y}CQtZ~O zKJ<_>te7t`)lMFq4=2p#DqR^5iZGG>u02Hqb0z7!^2T@RI(cR36C-jS@);VSHUfUm0PBm zFNJd=WIg84DA*)Zr)Z&U{^$}X-%$wXl?b6jW8|Ns`>xlsbY5KS=QSkSL-;?1on#*% z+{xQ#7O=1l$yXkCr_-8co1j(bcMMUQ3@4rY7;ulsoR@KN^5F#+?r+*Jui{`DUgpl% zi(rAWYc#m0-WmDsZPSX%T%SHi`dtNzZ~dH3Z2GPcqS==CRBY<)h@^q=d@f1p#xbgG8#9t~R~ z!6V#NVr60_ZyPHF;CJqN11C*znTXUd-U~?`oH3m(p|IB0E%$ADo7_ z>$gijC?AuT0?cay?@`vzQT)5|D#zR11hJTrpylqTSB?UH;D#fbxn)CM*aYt^aD-{2{Wc@Ur-|H_lZoZYuYgHmnBjm zi~X0r<1utHOzZTtI(%Y2n1vX-MnQ>!-%i{=jIauL+Tb0I5PKXP8M5`N0^A}^j zf>^IZoOB+E2)Gd^!wD=vtylgmK@lV4CCGg#T8a@PtC#p$MUtpeK&RWf{r7?zgMZs6 zcTn;Ux^4|Wlsoi%U`nED#{1jNR%3YWg4(wp_G+vK30ftYSkY1~Xk3tFhkE5qS(P`} zU{y0-pP|&IVyjV_l=&rXD`p+DJG6)P%_A$x}eSQMf3HHs}d>IY6SbYvDGb{CgVPZA;6y^d(95L7QL#m79pW zk0);;{Yw_=3n>Czv^iqYZ5DJ(fK{e>%h$ZfVBX0+Cj~TQB8929$oz{ z&7fvXqf=I6!icsOa;VfoD17=H6B_L)GbXmSv;rn=wBV6}!O=8$D@EVY9zg-~FFm-B z-gJFwkEVVgo#=Js7*FEsxp!?0gn&n`de+@Llr5p3j?^3Hx1Zsf> zwLqurs>S7df)xQ;`OjLS3~z!J)6Nw=#>L;Rr3TiP+T{_bPo&x?*}2iVmkR9#b2CyJBcO_V-w2Je zl1FxN|6M8!I}{qlF3D2N+=sD0lpBBLGiGVH5w^&Z|9oe+N1-A9C-Hws$p54=m2XWQ zTtz#1Tdj1*tV1@;gy~fm{(#ba&rT{ZRJxeH;>t0__#ly$15>wgX``0sPXj-|m(8vM z@LS{hO{YH~zDoVnZ6ji=n>?!>rf5#nz;P5Z-7fyaLfNRZnB<>DE$zYi=e?KAp~3|I zdJicp7^s99Sf1yB@hQP$k1)!-D)?(a$9$5HO)DFZg+?s_#M9jTl~$1->i#{XtHzt4 zp)g-EeAwS=n#R~1NbQ6X4lnzL=H;DTsu7b(c%eR(gOZ^ET&{cDe9E%f3c8#-qdM2d zi~(*DOY4>RhDJAW{^6d$W<7dLDA!q5yKXJ|QM!@xRwU$unqqSq{OM_N2QMWySQme7 zdZF_os+&)(T2-1Y9Z?fdtyZjhM=?pRDc*$W&HmbY$>HC4E#SJl=rKF0{13B4-LY}cbO(F5d(u@GPaPcIS%0J7K!Cc_)SmTF%p_sojPJy!7zI*E>wxCR%p>c8F#J5%x=R>-U@2k+8w4Wg;A9&EDKgQE0k!kjk{KD-hXjY9RK2QCR8r4 z&@-2WxfWrKElTZ@`kW3ntrS;-W&rxFg7~b67KCK{Nqy&(WSOM`Jm}}M5a`SsXNTLb zoH|)@uvxSyLegJJbYE(6_~lI{h39jAVm4}Oo+TE(wCJ~OOn0wkQ8DE#k41p-M~=(K|l%kU9i$8I71He)$^B-HPn4vpq7@- z3N@$THuNj4i)&y(EBw=Kf)6q#hL6)zjtt|`=RlwAEd2tjNmgUQPM>L8+2-|u0>YB} zTn~l=*r(LJd>8-&`pq%bD*=7a^TCZUKznu;`hft|E8cq}9AhQ{NkEZ-BAZS~Tast| zOISEr=8DZ9nKYGyFL=C3fu`JNeU4$RikB8_iG|pu-*SjCO&t=}!<^{HjitjHjipg~ zjigmd(RKQhmx7+MiG$9B^kH>Fj~a?>;Bf+i?ox8%3$WpE?vjJ@@s4g1gRNfV3iAn% zviE2Qzc4d1i&6j-(Tq+Qut%n=+kcRzA62TveP5;=G{eGft(Bkvg#S+@srV;ym$Aov z|3x)uhKCK58B*hr7VKY(tEMhoR%N~_4zc)wos%9AQ4l+IniRqI{J+4Sf`5U_e^p{G zr7Yt!D?@R!Ag&5ofuS_HqV{yA!rb4&hnxn{%$0nKbW4mGGExm!M8wscDoHZ&UC&w&mOg&eFqbi_2@ zr5~`5IeA+z{e;Kyo}wL}#d51|(HNxId4Inxyw3A&MsY||OLTAXU})cMiz#KQh%4q? znY4rqI=w9mNj7|J3KbOA`0V&kUb1hLQ*YE2Du71gS4VjfWNGihVc~= z2t6v?Sy{4nqjz^A6#`;Y&On*Ii23>jWJ8802FT$|zwQ(Gy?h0+88s}ZRnVmc^JO}Qz_ zlY^|(yd`3;IQbeS^7yq&srN3z!^&Q+`(%KR+wB#h{S78+^9E|b%O1}ueA^mK$hqgg z?Ty}w2*n)ci5qFMcq_o7%PQL0csPpeeLXcQ=1SP`BI$YHD8QoRnw?6Q z)>^v(41)d-p{k0;&Q6W=4NQ;30qt%skA2C;6{S_}&KCs!il>x!#8g{(vi_cSR- z`cx%V34y9-ddy&>EA#4Enk;@pb=RBv$wIE9nJg64C)SXmkQBkjzOu(@B9qQg^{`2w ztkj`?yE{!CyC>R%uf5neMj{iEjW=afW7(#ihw3fE+x#=}k-vFuBqx~QJ%of{_#P)~ zd5#yYH3t3x#m-y&wejIUtY*L(P2EJ~6b9Ttlv;YIz?61!6Z6NrloYE~?b491x|n@N zBGQeT2J@e_ztyhy&F)!eg${uVccn*xJj6YD^`U8N{1zsc=sE><0nm9t5=x&ej zc}tBt`V@Csmi0tPRU3G_k~`=aCL&e$)v)r1m1UId45v3yBL{*0)FX2b#iySJ>J>lJ zOw23#vmag;th4U=z@Uk7U( zVx%W-U0FPu-BKd*8(qq9aB;j%sz$ACw~}H?OQnh3drDYMS3e873Vq7(pJFgqv(877 zpa{^nP(N8<*_UycH=~m*GcIGy=053x;BI+Lz_%^zV{28*O7T3QH z$g|=+av7OQ8-b5X>zN8JUwgAh^Ra#BdO~1xur@icBNQ&2x?+~1^ytC9-6zo=sVpR) zm@m$xwBqmz2J|jKNk;!TtffBLL6iRjbaNqeDcrKIk%0d`HTX(&P_$*0uS2>YG^aGq zWU+W% zhQM!GkR|uaT@IWeJua3cSa4072}a=Ve3e$|P*PP!_>vX+$R=~(?x-67VbU&J9yV=b z&~84;siu%4yG%X0=VxBwL_<3u$3kmuNid)kN2@XDI6sN6=J&Bx+nKynMKq@!z6?jx zc;vB6p;l!&c%y|*YdtUMI*)tR0aYET5|7rj-=R@6Y2n17Y85BxXIUYO5!Z_glq>r0 zbsN-?S8v+FrGNvML}_8Te9LBO@=!;T7w$B2koCu;taPqP-~0?1^8S)d2!C@;eb=|- zeF6+qEiL91v&i=hsOWLNC-eJ=gsvbdp1=x4W@xvy2UV7GobVX#=A-2?7h%Uh* zJL!i7P)1(!zQkXHRK8U_?u^SqpEo^318YKnPJFfnoZC>q2UTy9d#4}Vd|vb?X4Q9A zM}%eDwf%BO)ts~k0PAN#w2QZa1Q@qE8^<;*Ce`H?FCoD&;Kffd1bVnoe<7%N%-v=af*=*H z1vTLM&IX;5YGOmd4S&j3zr$7#Y@h^|Zcy|5aGO;M=vKaN)=&vvYiArFtEee{wi+S< z$~JrWaUxU%l%Bk-%g?++tB&nkr~{W?j+NP+lTX`p`1(*_RJ)a3YCd+r&jF?Vj`e81 z#n0Zg&BOgrN!n9!rL(Skj<*m$u(Y66mr~pF3YOq2(k7IxAE)axHy28+{^~BFpEglH zk9XCxo7(DlMTYeWSe{REuLaLrxxmhSyk-&(z(W)Vl**0qtG7wp-aOPTArNqkZ7}(H z(%P-RjyS>{MgN$*QhW-IU`HpNFcJuRUxbsmj0=!P<~79yUY%SUH78Zg^cokK4w(st zB*?!Uu1w5kxWy2!1`7YC8nt&mF6?{l>FFb)4;fJ)Cg4Cf_?C=pFO}$az@kr&o9DK8 zC3ZErFR;!)D}(S9&_OG@ZLi1*yfUMmXk&Z-io z1Nf&JrSt(nJ@(D6@VbIM0Za4s=ztSJW7{>aSdL+KBf|mWT@Hs1xlKw1x;f@hT#*0v zutX!+94Sr45n`G^Y8i8OVM!x0kYkFvB}}=hGZ#r`s5=#Y*ixs>efOD1^z0?8i{gwi{ER5#6FYQ)M=FfJDH!M@dM*z3ZjwQbCAhzzp4n02QkAh9%` zj}HH>h2~DW*(vb<7bJ>=#^YY%y>~iiU(r-v(S{8R-sRh|R5a24(yaV>l~i2JB0}Y4 zM=df5S9P^_p*BUTw-mc09yW<{O{Rgv#Xhkx8dL;Os||-5umM8iXUy;c@mnfdYtx=u z9?`2o!u=X-sC;y!TD*yQas9skN*xjWmp+lp^Y;m{PtTPO{8M%)LvbB{<#^ncf*JF; zgajXgC(NHS#?i0%h24Ekar~waB%ns6paN`W{L*HkSs}~?ND(JZI-^4V$q-pm z%znR4C16ISV93GbP;8roLlANPV$9j~zN@HuC2yLpvJgMEV_~q8VVoL~^rp37{r=CS z6w*%C?=TF4=lpAS0ACy=R6Jzij8S%n1jIiF*?514V!y!jPtwWooTsCL!6Rdkv$Nv- zdV2gDF6qd#D&L3*1^6ciVQ;uUVY#(}xh?slWt1)5GKy<)6gZa8c=lAbG%EI9;Z)h{ zUFnZsztwG0m0EqSUE{uT50k!Bt3I;Q%Vf*$C|_NirBSVZSNhc|$Ym_DXPh5S+#+Gf z89zsaI5Lq&vMna)79y;w&KQn+u^+;a0{$bxW>XvC8MytoPea zNBg^9JC{zvEW>n>UytTRGvvil?uY5bbIp_Xn%BC12mH_;UXWi=`Z5at_(T&H1=>=% ziXezEOL%=NeX(lJJY-g%EE5v;m1zT`FWV{J0{U&HgddoQ&c9 zF=c{=ie{@o(^FPH%gGeJ$a*vHErisWlyD0=&yrJ_4Y4UK4K_0?q)4EQxfVyMpy3%S z{oE1i8m&p5N>&|NwK`sq5F!$8uYRK@(gLl8T?$`tz1SGASBupm6R(27&g1?CWZ5)F zdwFu!>*$4Ak{6Gk4O#deRlDj^xoqt*M?dLBo2L6hoS6veO1)MbTe2-2HS9#c<7NND zsByBaN#bYU(vZd*gJ|5Du{@|SqAaT@7LIL6N^SJ zgIVL}6ZRi^7uD}xvlzDPhl8N*`om0WCoRS)2lNH&k|sTa^HjB#IzCepT+DfDa-m3Q z)YEwbZ~+E9`CV4%vgIicYO`B$MylJ}^Gx_U{dh-M{F7r#)cG&TiNlpzwU^ek_g(mp zWFa805jje|rMU&^S`@F+n2@X_qKWtOO1)BLasHFQcD0!oMK2&DdDBI;Av2dO*S?O* zAYfKYR9eO@#}KTecF8`BAxRb(AJk8xt-XteEA$q~sM5EjP%$cz%xWp-0qFnLulv2y zf$O3)5o^WR8+ch!PaN$Vb2opUPhhT&sej_Qd4}mxxk2AFBJO>gc0|)O64aVVe9X8> zPZ-}|U9PjP-wS5AG4Iy4OO94JW=qttOGb31xn{?@*q|pA<}fQgfU!Db@y#iOKMQE} zQqrz0J_=~HYSNtFMNH8;tD)*#X~6L5+BLYB64evHoZQ|1PM~{1uRPRE;TjBJGHV3S0n+85ira*!Tgj^Rt zwWv$)XO`TmU(#yn$YQD*e^%GEab3Jg6VJX_D7a`Xhd{d=5P=kKdwg6v&Jw{6u{aT% zxrgVv*=a330JA6H-^YY68g`6n%%!9?pko`W3b4Xp3yKx&T~?VB1V-?Y<++3eLv`93 z?T|nz&NPtZoxl}!5s?lK&LR@9gY{yngN98WL4*C`y~L|#ZedEaGj z&}O<#4?D#A#)}cKPkl4HnFansfcO=T@3H}|UGdUEm?4`h2FBKfMyf-+kIZ9GzMJ&w zS+*HjFXxln3J+Mce;m0>1AUHKONbwu9?)kzNk?zFMrCIXM?VtKP+<$hv@~1lykcX0 z`Y-JiM89E;&5XA>RCiPqlvO-XPt3sR{q*M0ql|td52203jQ`cFik5&gSJ)Q1==`SO8jj9=kM$`e*u05y>y((<02GAD*wn)9IREBh^;1s&_ zC5Zu*nn#FjEGw9F>&r-Sz&e)I-pJmxoXLm@`>)zaEB-!HnDlNFO{+cm8IzCorl$ES ztGOmMn$h}4I&xM1$6~aEq3SW~^wH|Dns;H-Wp`w%In}#vt>s{lyG&uo9C*I)lFFmc zpitK-x|Jek6H{0X3TM?=qBo9CSGfBu$B!}mE(uAYGwi1r9(3fV2_pO*B|uCnx5LVv zT#Z-uXwoiIo2wzlIzJMkw2?VTrK4qI^J8_kXm#uKp2i}>h%LWzdSZ++3X@gQ*nnU_ zq?L}?lOx8{rEC6ZF{uaZHUZMmQ4h}vemgA|5$pPQbwjnQiWWN%q$P5aG?yA#HnvK{ zWWf^Hw-@6!O&4nU#wIZu-?!YJY1c-*m#Zzitc@nCriGd}&LUxj5ZYQ>J^)z3A={HQ zu-2$0W5OI-7DLH1VU@7@77CP+)N(BqG)N+&Z?}jdGs#oZlr)&lf(acXu3q_YCjXyv zu8E`p{ipgw*TNjxS`z!J=suhtkC4G;V(F_v7h7&Kx0yGHFZ>wYX(pI)LbAqxLsj=p z0Tok$7D(od$DxB;w$IuzGxeA?~89X!!CXBX@(u&XVh6q7)5K0)WD#@|AdC5e}eEoLHa{bi3<#h z`%k!w#vgDiaj1?grqOtHnxTt;`{TX*VkU<&*Qy;OF=U)8&)*YyIPY#--ld?8cqMV7SZ%Tp`Zm<*OC zuFVXZGWHNA<<(fBN0d#Qu7bKCfM2u50EW3*T$DgGBa6#N5wY!Qz4LH+fxBH>)hXsJ z`jm0u%6=G`$(2&7WIZor3F}s-<@n^C=MK{UbSLpi$2j}}fGxS@*K}!a437FVO4)^A z1hr(j0^E_Dk%XyfCe?%=p*Ztnt7(21os~U}Iqk86sqwd}7{2~n%nP-&8e8sdb~d<| zE2SA`6*42x&^>rIUX;liX+f0EQ&Vr_y>->WG!6n4_*jHm&HDTo8h7`X${cEGG*%mD zbu@NN?7RVOX|mPdDn9nqCHy4q9?{5AydkNlnKV6|$p*HDA+yu%l`Y~@mFDYrY-?7E z5sL#FHbaHky-TO~A0WD>SG83Fq&s_0jENtgdPQ!tU(W2>LsT}s;-~IS^a;RnP}kEM-2Gr~*{9a0a1&6|?GQw@ipjhJDC7iGum z=Hou0-EzZm*b@0H$z`#D&Z|4a+ku3?L2w-PZc?F|ci*9+d~*5`qXP0W=+5Hu*urt~ z*#6|@N1h^(Ly>Y0_m?Kot;OZlL}S_Az=JL9(C~K4RFd(^i|00qsPO2}%}qB_y5-qx zYOO9C@SN2?pg*Snpc9@C=yCi%sC{>+D`KU^>a&r>g$r6DpAo-Er;;Fy2&DWtOij?o zIYSr`M1M156xkp#t?(=PfZ}dVA*aF3UpG?udID(Qc27Phe=Fze_A8?LMh9LST0{t` ziFNu9vLX@rzazQWO4?^c&K;ls&X@tw(CAnXFyR<$9_j9#fd!R3p8-6%#ah)0L75to zBzf{X4rlVMXc9$q7>zfbyK)|1LNseCSgh=2?`ga{>oe)uTw1}wLHw8@Ju}cEfMByp zIuXkndLoO@BK5_`+?kMQy9S5D^EQn`G27Lfl)Tz@vrbVmExTrbd1&#-jUskuU=Gj0 zPw{58mYRA^ov0Qoo1+gWam0Ny*P4o%1m|vfQqB^9cJqk)JJ+!2O#6yr2dLmmixW;6 z6OS!3Po@yB{JCYL=qnzbvG&04=N2nz?ZQ)U?cfP-?ZuD&OFZY+kHMDy8|=;*pW@*C zmJ(N8wZTOy3P#L^Jvgo91^bAgXVR=eT4nADgQU=;gdm%Jq}fJs{I+|nvJVhSg>RM{ zA?z$RmqDx~NA4b5EYq%ODij+xyxB;tL}=Yv&vb4T-r@=Igc#MQVSFzaR0BatrVTMG z;s*o%R7bw|DZ3xCj<)dl$%HropsdZqZY(US-IV|(O#$wyGui)C(BEBGR?NeG;*kI_ z|ETU@_0v{J=Nx}3fa$5@5nZsRZt*L&XDo+1#&IzoF6^>dd1}F{klsA`R3O$fdx%!x zEQ`s=hFFkaYdP;sh3mdhur5h_{3%=P)9fimHk%+O1RGMe#ylHBwpN+;I7GW&@d#n| z8o!^-nrSS;C~};xI@S+$kkSgicIDwgl;61F)MIN@qvEB3B&009Cymp=6z9Lw0N+@P zzrywM;8+(oP`r&_eJ%evqhWs6XRH2Qu$~jDLNFV8Vf3TKK(}CVMB8+yrZod|vMmr9 z-fuAnjkBZdH%c1=Xn|G-XhIJXz#A>|d*j8}%S)7h!tRsz8>NhKE{yO17^4w+(#Duy zU)u}xi>}vwHHu+}WbPH%bAqdBsrVYhhcbAlmJAF;xcVDP6v+EECKZ~wQ8Q+)PSWqB zTHvngb&+q7hzX|(ZsTMuya8of^yWJj3#?)9D;E5Av_oBR&D5@ipM`Q4tA_YY&Ayci z_Q$r)9w7rivR2ppglf2L*4~v`r~KpDQg*gNcXxq6Pivt8%we)8UN;@lB?xafBN0*^D?NocTdJAXokL*CJXVfOT4Q67bhT8l2;&g57bbR-_X>gUqCkf!|p+ zEJaFhppF@VY+m--_0ys!Ci5qQY>0{rt(7w5-FG4ncQrEh>rHVi#n^W=k4$s5{I~O6yDcYh zLS^>n?s51*p2u9U8&husJc)N2Qfqq`6p^a?oN<}w5!Iaggmf!apAJHlnoK0p`Dh$wZUXObaI zBB;TWIltE+7=##asTgz31#B;tIQv)&prB?#DyW8`#iBJbzTpD z;9Z4w+n`h-zTPl(8}|d6R`?IHpR)}q{SUg60!m(t(T_H{W!H=?Pf4@a2y#9%@1lO3 zJ3mPjdJRH@%kUP?P(IsZd{9|1@xqg`sC9|Uw=#;iyA$R9VF#13M5FbDxQW8~oWwb! z^LIa@V{Q%`H&DJoohGALjJ-;)i5vq@T%h9jsu|(VbRwJqcBOr_j)>x8{{gc)8K_*# z&#pNq>W58V^XcApkvTuNzl0%!wBH7iHSiw)Oxwy7XhUav_w5vYkZv6!ogajY=&}#sUBDEO10IojQrZTmnNs9ARSk%TUi+`+gu@ipla%$bS=?{E| zN$YIY=r1PywsaF7^rfXX^Z^z74Q=LU4oz=C4e&miedlLJYx0VF(4DE5t+Wz6CLmk( zdh-D~_}xhC3*1cR2W#|Bpv&NFto~vf%Nj_c082>phb-uNMqorVH&WYSHX+Q%Ln;J$1 zdWdqM=2=2?x=b@XKjFy!{<|BE&TxvN>4i{##&593o9MP+PnZYwbe>2Q9=0$Vv}h9j zY)vt2iVR~sOVC@kJ`#5rPry3XadQ4fh0c%4uqCH` zpiQVIOOK$IVrc2yIqCo#YO@xZ?Gj{pN~uSLI>6u~vt`3;iMPqooN0p8R?I2&J_SNc zJVk5K>$r?)kmo@*)(myg*;%sNwANM(0FH5**w7m_Dhblv|BQ&{5j3`T;6JYpb3-+D zm2WmVbAzn4<6vwMLDuHNJRPxZ#(8{-&yCOD7fe;%ar&U`5h*HG=uSVAnR?WKI( z&2|ut`E#$Tc7^gbA##~REYa)(z5B{d0vN-SsemeOz`f(YL=k&8;`G%GL%j(G&Z1x| zZ?=OYiTv(P4C)S< z7`}Y$7k#YzDrcyCj!Ql$Y9FbECFz$MdEiG_+D8Ym57_2X^E%UC*YL6wWGZ|)0`lAo zS$|E(CD0QxbcCLl$UM<@PuOvt0* zQZ@~R(Qk=H&!U*ioOF1!L4VHw>b4ygAQbpadFrbnJ)&dUOnNmb`??H=W4s;#ph z*nZP8uCccous*(z#?~?=Fjx;r)hoChb!s&>0W?Chc!H$MC!+!hnBv2+q#~E2QXzNL z7-!r}o&vo`BTP7^enmc50ZrgISj~QZka1mqJfbTRh~m7!NC2ORvChIhIWH#McV4joUEf_J(@11^`G z{@`w2tjEYfm7$w7hvEjxOsT6{_>cSKI?5KW42V&{EeTaY3qZ;}zkIwoX6wX&IHeM( zH$T3K%hpQJEW{VW|I(;e^oZbm$8KCQll+m~sFGr1(*@$CRWEuL-ACx-BXrQH#3z zIZiKA^FW;=8L|g$-<%T~ww3KsJhRAhs&Cmy5&U%tvxcDx(dxR_Ws}(|VvZtuMB;+If z3e#D>-MB*6X-UHQFc!XUoNprYVxoAs)bqSQIc~k**c!cl?3TDrx1d2gjqI=+cIwc) zk}xumeN&y_Qz{Ig`@>x-1ems!#8ZroD=Roum>>!#Ymvr~w5Kr>Lo-f5IqeJn+f*>6 zDH49Po9@h(Hg`hDMD?x)BJAVpCdHV;^ljPgRwvcTRav}sf?>z04cU?_TUbM#J7y&}JA#Zx zfSYT24-cp9_x)0jy?MDz&BU9)k-x}9BY_u0ZO*^iR+P2`-O|5Q{&_?lNL$}odbpvW zJ8sI#j^QDifVf&Cj>!Va+v6*o^k4rS?A;PB2~tk6KFDKc^H4Z9W@v0%YHP2>Ym{U$w4v#a>1=A&Y2jaOS(J=|_qn9&sA ztERo`zyXu1Cf4c59VXny)n5d?bgS ze#1gVf`vJ|i$w{7^~DNhz^7t-@Brh=$0%UDmR9g|`(9n( z+3I1%>3!@(`k5B5`%$5b(|!5v&br9C@#?SK@I&V`2^EEi#F>^s67&V4yS(o!>*7}K zA)8?0puAhwzb_U*lURZscHDEHUN=IXjhpHWA8J^TJGy=0RxU#0@2htswnU2ZWS?S8 z{*Mo&t>jM^?0tTw?L^^NeDhcz+T?vY3w@c$pl9zhCSULyGN;lN?;Bu#a1`x3wwRof zOZ4UGP%9|i^1y@~Six!IScJB%9Usu($>gp?pE5SpJR16UPi42kwpiyxEkKb4oEM{g zjO@|SaZqGd-7139mgaH)f5-^fR_*LRt{ z#yIhWRye!DyN`qX8jDEAlHJd$Hs$$q>j=DgNx%9#OEs3{#nFoDp0Lu*dyH-mfmvyX%eD^5$XD z+RSf}c1+*oeFMjW@4iKW?Y_k&nyD37FWTh!yBP|^xbQAR`U|>94w8BpqN8^P+j&?h zPo*JZ`AH;*?bg^Sl@{emmFsTAbs9pu67%c-z3u6bK%qcyagQ z?(SYFP~3yNySuvucXxLw);Ik<|Mh;EoHb|PbCax@Su^MEy~k&(C=1iO(bDU$JoNpl zCKe$A?*|*^oFH2yUG`~VbFUA&Y6g$q-#<+hlC&+_;-;P%#G9kFeXa*2FT~qg=+=!3 zTX~sR%sUP>?T3RR3Y3(T|l!D|?PGBj0$o_qJv0qmaHI^QzA}Idc}6(faPTyjG>l zn>N1e>=v?I1e{1z-O}$Ww8uNRL)O9cu}#kCZbp`H=mGPZnG=8r5g3(8;FLMon2+a) zV5F_=@@MLV6TEX!s(oXR4D&c@tBL8v76QB1C1M^j|2%5WNz+t|& z`$Di~cirI?Bh!&yUwmYXH66bJ2W+Oy)^f%szwPsczuDBbru0Zh2n|u}lAOB;D+2#= z_v|rFrpx-Lt1~Ws28rgWtG(TOEoAwpPJbGF_ty2^I@7-}nF66^C1($WnbsWId|OC8 zNdnwKK3lIbRF;yHv?~9Ti+&Lq(v_e8%-t0!;^+T$Tv)k`7$$jW=Z2K`SYXH~5K(LJu(0gjFw#xh?KQ1CXCYQ`CtW@YO|&0W>BD z$U3l(6?Uq#fH0;GIAIAmn!*nfr}#fF1vvqI@2bx$9>_z}&ad*Io$~@VGH{7GD$JqQ z>9f_N?w4{;oq&@M&HLxB%M*jJ#FGGjTW)ZMT!0=|1GICRzn;hDWOpWmGmyk%zn2)* zd3oXhmY9Uc-xdH~w++y%8-jK&^Ve%Vin782FE^jhSuq7%^x7eUUy$xbfZ4Mp8{j!1 zc5H#|{doVR$(v^RQ+)BE|P1wHVA2+`E|EXd1E2EHql_92h|5X)FmjSj@h0C1j z8Twum151PL7C;B29N^6ODr`w^Y#nYG51|YFA^aog_jZhNVJLqA63U`mbbWJhC8wbx z@cUZ+6VxHbn;Z>prxGmDS=+azV13egv!IZJ87TL_s|%SxcV#zl12eT*QS&X?V7E04 z1FtIrxbP!=W-cUdkSJY+K4Y5ERm+25_{eOr`y0A0;2#vyBRb@K7La#qqqi%AcDJ~{ z+4;=eE%Z2~aaT6DOy!n2Fwe9u-Z-wM327n@^*AK8KIfN|1?}QFwAasvMsW&;)j7(ylCsMv!Ilf%EihAeNroi~V&dB?k-Ae~>%J`O@j<`S7}q(eP7C z)gBo)-Em8Lw1jNN$bN3=i!#dod#8_d>mN@m)hpL}Q4go;kL7!L0mS569Lb*$+x_3H#K~5tvf&ehRXW z91&l0(bh9+h3?nI#O~kAJ5A~@KW-%}B

Zt>9waB8-px0=Z_}j+{V+GyYnZH{^Xu zZk7Do*^CxH&!Ic&&b-5~!vgR)s$6Wes+CiYCk_G)u{ch%(dd<=@Fo6bYWn;OSh|id zhpP0m5S&BDYzP0-I5wh~Rzf{fChG$MCWo(y5u{%w{}Jjw6aEEtW&Kp1^mJIkjKC$S zNBwG^xnaSgv|+34+IxPX;xG+OgIT0<@kSc2AqHB@KPffgl%qh$f59DW`D2lkNKT8H zp2bD9Eiu<KAu&fYqP5H{wR)b8sjSbs?~!9mlY!1T&~%u&$AD_mpT-3((-9&1_5jj2Ydt z;#<40KD5H$nu9R9O?)X*v3eH;oBg0Mmw6iVf;utUeKySN>gw=NgFdcxykvo>iOA{G zqrsY#>~#iMH)36|AB4z#9U?;*@voOGS|)z)iU=T>259OX)U0wu2Wa-OX?qTJPN$d{ z5jV9PoxM~?1!(FgB|D!13XJp@?8hzlm)sDkjglhlW!Ayybr(>*Y^*5bwKtLLg?6ap zpvdJJ0N}+kDbmV%f1XjtNFT4%0k)2F{!xD?P*C{0^mWxPjsM$5Iba!91@gsH8iqI@ z2rJoG$6)FT>_0yZ1|>zhe6FvLg3W?;!~jSA*IB$=vT`31Vn*4s{Nv8eO4_1>x@Ryp zaT5ZIH!L6|jPS29`PsN?n+&#vF-ALl82dJMY*0M2elDN-bBx!oDAYZemWQWdy*>#9 zM;@}KggA+)>tPck>S!&zNJ1XQ7%zX&e~{)8y*^n4M@=$ylu?Q)Z%9`(L4t|v4}k}e z(S)qYZkV`H$oAtTSF4%LLR_dU(=xPCo1t}WTr6YpB1-I}>hPK9$^y)1tZDyuvPr$L z#8IGU6bl$8n1>9 z@11AVish&^k%P;#E2K3DCDzQXzc z5)bJ-8xaRP0;|p}gpKS=RxKO+)+ZDCEZ-lJdfC4{)M8~<7_!uxRWI63c`V(f_vmGw z%^$E!JUFuD@YFrdR=lBuv(nPYU)6?lH%+n zT0PvwmUFF!&AHN0q30UOzdTX4wgd7Oyki5T*DhZRdR&_T#oj{bQao~*cH!DQ(tn?4 zv2Z}W4nNUl1+G3{kC{4@$@`2&0UcbKrB4nbD(=%w7@A(f7O%Oj3%vYqP&Wdbygi7V z$5uhoXd4fprUI;UjOPe#fMMOHEy|Z=x`Vz4s49YLcv=Pg6U z`#a35&W}$}1_b`_`pYnld{`@X;b181D(IK$sWt3wk?#Pzs%O}8V<M2%D`|G9RjY2GHT5H0HRf;fPq5XWq4`!2 zm(q$qyxnu8OK_0c^=Ue%g+jblZYLZG%-jOZ>#X@LM+brWUX0d$akruFplSWds(}_{ zE(bm4TAm`smiEo$Er|OJkN(>w+|Pt46>zP^a?`$(Y;CngQIqNwNj0m0Wg!KpX^>58OvpGY zq*bjOF_Blegvw2XbQt>cFtd-~3x}wAH-S9>cXC~7Ik<^4j>o0z9&A+w(q9Fz_4f)gS|IVws^my?Bv ziZ?z8hva4};PB$i!xYee7IPNE+RNhZM9sr5D}y+JsYWDCsuZ8FZ-QxcKUBTzR|MYK ztHU4Lxv}D@Cz3Q)&0JzcQwEK=BO}WKA8~YV*IYCC-$4@gcjc9PU&h{sny-WdNfud#DmbE=OjS{ zwT&M7!2Sh<6lhbc?lWiPgVt&^)-;3zfm&C;9tcNVl$BK95_TM_j#J;Y4cf{cW{ z?aOtI?m?9z!fg~biBds-P)apa0({-JA*e9^G+QKYS*Taq1soh};J=Q^%#^ZSUtzXJ zlr``|l_W(uHnV7yW1m;WIUtoENIUW=|B{cY=zyP_iXcSv`130r^@!>= z&=f=B^7vy4bYk8wD%9{;S6-$Gc`5AGKnM@xOWXgxdYt$ERQSF;Uy(1F0!ih4UxC+Y9?b@<84)#SgNWIJ!h zPQyhCY?Z*kY39yzbd>rWIMe{G9j3~3XZ{?oiZhMS{hUFMlsWeAle-v|H^tTSV+0q= z%5#64Ye`AnO*p2a&L;UsMsG$u>47%`JU!Dn&>t7cfQ}J5dpeFnh)MvzG*(?Sp0$Wt zYcPqSdsp}ybSA3MfKkcQ3G-P$Ue&08u5mX1{54&p-cvbA){f3VQ8s~~GJ{EJq;9bh zo{8j|*)tZ}= zG&Su4tmN{Igd*5d)PzkbX0XaocLXuE@}(|nrG$yP8mfFP2cL6)$1yIahi{=u6D7^sFlg1c+I3qqH1q-IUz>BC#6W z&qDu(j!)f~E)bLbQ;y6@$Id2LuoeY8)h zVWYb#7_=8GR%*MAHm@~GE9I*EocHodc$x51xW(UJe}RV5p^qIn)4}w7BlMo%2c|6Z}*RB^+1;$q9txeC?QpdJX@%Pcccp8)#I&?i8<)R%-TEhX2jJoF9fC+ z#?*0GPms|NO)pS8<^q_8EdV@_hbc`bTVXk);gWsrZ^I)4J^M0~Zc{(nLp;auGd6(E zN^3=B_QHAkR4@9Snwmd8Em)m2kX*TQkl=dl5yoJLuSyb!AjfQ)z=bcGz@3C2;qbAj zsiSyQW;vO5of(n$u0VZFsfJ1&Xe?W-*M+Vfg4Zml22zhcnU~m9sNg9CB_bDVT4#F7 z-r^{D=@cjO9_NWfMt-rRnbZFTV~MAotv3q;KSzRZ=`}ZL7tHYlItW1Rr4}-+OE~5% zknulL6mFcVe+7^Q0WH1^Z^^?LeO>zP14mFVmefypL7@iG z^HFrM=l?=-saj@cMo-(>!5rsJ(yY1=kyR)T%Tg)M*hp-gMA004rcVw3lBriUBj%Yn zJWcsxp(smpvql$4S8qz>3EKg=E8}-~)%~f_`p3{}LSlK;rYZF*5{ArK0t#%o?XZMH z0^O|#TYX1f^5+Fb2+&E{;5cUULBdoQS|kzx>)EOt#cej~GPgEg*Emeaf zAJnKV8Z*>z&pw1UIxbYvt7V_&L2rH677vIKP?M0)88rsu z14_g@knzNJg&r|OIS3pRX40U|({KGK`c_DD0vkK@<;Frhh`XzBcd{h$R)@bJ$5Gu# zQ#?FmQ7vcR2{*-YOAR;k@3Yfg9t-oB{`bQ#pkt&sbJzC+_{fdK^fXb9i5VUAY0)I7 zNj$0iK@C}Y;Y+SMeHi{Dwu zNBhE8mnI?G<^i`vg0i~!b-14$`Kzi}29z2@{rkx5&NmV4Q1(yWG_}peA?rFiYM2RS zt~Zu}%e6^vx}C01^9!ZJ=}ljk#rj}zjQ??@i-@8@t!(B?euS2h&6g0)G{UNQAB1lE zFL(Gi3iQtMFPEfOb^4edYVEAsv~Q!FJf=Gf8>}0HWxPjo7zJ`<9$xbMnlxi`IWjz0 z(2)2$bgBLyoqH;sm~%oirsmPO{P$94d~EcQB@+RLg08T5Ol;Nyucc)LnbF@7jgwNn zDL>skqnySK=YjrBg8y;gmaQaaVgh-?%sKF4LC>bg;++t-1M~0Ob6~ZO zGBFqDJlmEBBfsC@`@KZeKILnbNq8I*4j~KMv_X2A%z`l&ydOhlI_TT@b8p9AQ%mbQ zhI0ubqsthh=7(dW1=Pu9E?1BZ{-|??%!6IL|9g;HEmre)Yd>ANjg1EH?6mDRS_uyV z@zKBe!G}&9*n^outpXXj$AkJznDn@|x(xs>?j?L$8?@+4_Mdi`xUB)9kR|Kp!h^s* zer}zUwzj4xKxCiw=O2w;0i@jq=AB#B)1NQo;7GW{^zBm+L%(Y-5jkwJ66GyfX_JA8 z9JX2W4egig(fKkQn=6Tf`WU%&N=8URUuL{tO{&g)ETSiTENI8Ks<%Hw)}!oY9T7r3 z?3;0NLSYdW`@F4>x3l-}tIq@})ISO*9)xe{lLp>?fv}4YzJ>!h{YEv}Ms4tOg9DDm9J&Q;=wgVYKO(xv(U7(F>wXuT=E9_f%W|}wp|{F^QVh%yjCXr zNy)hZQ&9f)36SfA>|VCT7g+xexNN)#Vl%KHjN^20iGwz|f@{bh)L>Fn1T7nAL|K$i zsAp8L4Rfc>0>)(cV#$O}#IU;PsKIF{ko*>cEG+(id4<*Q2#(W66~<)k8mwU37f$n) zDt&!Y0*%pfA7e5i&Urb6UMYCk3V?i2p$9U;LtKCsrAzud0h@Kbg5_FL7vgh_NKZGA z&Av7{0K!033>xG?J2p>L9-e-BF>k}+wUExiX1#Jp5Fg`&zaG+HJ4_Gco|GSGbMN?7 zSOb<9+{9@*L2Y{X#xE15*(|l{8Y|}QLeXux6aHE089H<&$X6b=CxrMeX&>RQNEh9mT;MIhD1Rg zT8TxR52Xq6Qd&d+>7YyGOv2&W3-3j#SvPojv;zxO(*=#p8#*Q{a{2Aaz#`YSy7)pq zluXdrY@1_E;Z1^Km!%ANUhk(x2oPK7K$&ZPBx-_M##;dzad^^4FQ*9v=-<7ZncB8b2tgs+r>F?sr#0?AT7wzxG`s)q)c;xi zRb(H$$sH}l-;cw;ozlk^y%-AnNIKB|>wK4#tLcf;XRr>S*K7@tPHh|r76Zb{@Hda} zv+~OZDiq?W0z&i{1zUfqi`0@;E!3ndW-${c8oLEFUk+ioq!jV4331*TWUnx)Aga|eh@4*BCHmyS;IFhyMvi>CVnon=JZ(S zrL3o7>wfiG4MV7_FmR<);8R<}H>1#QTB|wU*X+wRp{mcvSckNg!c zn1s_S+(`gqZ)e}MKo8bl5U)AIXG@RT!_%d$6>Twm@ypM>`LwUkln=(-W0E=Gv0Fx2Bw1Da@Ru!|$DWpRx69Z2Kd0W24TOzoU07UK zZvGKyuh4$3OG}gJL_|-gxbMgOayOtI>5CChmUD`D`R-H{s?_IZKUzc=4~gCF$#j^RosYo;d5SAS zY|J{0IcX^S>#iJQ3)wACsa$L;x3I+7iJfqmHQVLtdRrWD)^^csnB%i{U1Kd<#VFl2F?I(axUOypZzWO;Vlk(n(2 zDxG^}qC0*0OpBt4o4}b~8*fXe?WQiQ$xf#&U3oDqI$C>e08ytasy*Ptj`oH@1{Yep zg_pN|i~X;T=0`Wf9W)~Vt}PAICB8iO8gqz0NqRb65RA&4TdLyNrD5OI_4;KpXxbN8-<(dE35{xD_ZKA<_*oP>1~#e2Ikc2vvYe$=%lFaIXIa6 zuX)a3nL`BIAfC5yjMI?SJSbFAQ(WO~v?()lqZq}sByEasw*5J%j@oJAJgLrQ`%^k+ za-9}0&_M>C3I9IDGkc$^BCJ^8@7wW!9llwW0ePM2xjm_spaRr1+2o#n&L2ZS5ti7u zYYH(+w8UQslD2KW{A)R+FA&CX%v)5|LNCv+Ht{B#nRX$8d4MMqvM~M zg|Y3J^tQ)z&si^$xI~0`C`dJ2SZKU1@fr>CqB+R}*J0`J1H78~mM|%0ECo?NFz?JS za#$pHUVoep)kx-!8RukTVMXe<2>25|zeJ0&?EVq;*l;OYi8HNUs5xj4w=jc&t|3GU zjztm$+rTJtmBtE@1(?SRCEzthqxJ&r)&kq!`Ph*pcZVr|k5f{wn|^fB{+tT;H4Km{ zYEeuPU#$?{#2N389WB$_FO$nh;M@EyM1nd%CRdhP5&3DDlF|8{X@WBQl_H*%J>H&o zf)eQP7Ar*-Zyqa^aEg*GvDB2`0}z0*B8U-6Nfl~dr^dV&LzN_p5vrFWBdJ`7f`d`r znUEr@Mh>z{+PWP3c#!rRD2OulCzLwC-X)_f^)L0ay~ztQs^1XuZ>+DuM(n}KyeIvu zZN&^oQqy!7->#A1C!K3zZxIEgJEWjx_QDEppa~TIUf;G@!h+B(tj(IfWRwm4m7Mwc z8y`zoY$QD7MNo-Q?W91EYWWNP;ZL1rNm$!9#`eR@XW<;KAxpa&_mdP?t~V2T5S(tg z4wQ3bkS&hP5p5^>DrV_y z1+fMLv`w4W^PGvkYfCP66f%xmDlBl1wyrBR+NuWCuF9@C}ixGJAxmsJCGAG+D z=U&;-tBjQZ5uHdQTk?FyGYZFC>RZu@6btd#N_=QUQVBwN*&COYNY1MhU@;#xcK$Q| z5=+&gbCO2#%=xls(H?#%iE}X?n;W?fNuM#^(3Y zsxOrMhelL zRy#1f@zgVZXrvQB!3#w$Bn|P^n?TJe_$SRL4TB}h_b-SQK&r?((uo2R;&QPkm@_=TB`d6}f&ySj>e<6z)Vw40z^uT%A^oX`k zxM}mh5v~NeaDLrthm~5y$#Q*1;^09uxA)jdM!PSz%w@uIdu)V`_VQi`Z*jfA^+z7a zF3X6r1v@(ntCKi0Me0JN4Z6!uZMI`p>iDXm|4frr0=UeP+p`%kT9bHL5L;_wy(;n< z=#hddT~|7UXD{G&J-`jlKDr2QA^FP3e@VX3GDbTuq z%zM;ka8-SU)%iNMvX9||D*2DJw0b`ujSdnt##Y*gL>8XzIXSD|Lb?29)Kd-~^pK=5 z1xnOx=-2jrf+!q~mOZgpAji!F&Ey!wP0kGGf&E%J$#%hzaobrA)~PXvJwwGv)$q(( z>x0=>g3%#!wH)7wpCrwsKaCY~#N`JX&B;qslI$QtRLg5wkvv<6GWt2}M^h1bI-I&J zx|7w#UGeK5C=VV#HBk{(k8tH(-3Rnpoe->cxvzM>r9^~(u5bH-?)_`npH1I7KZJid zha7fr+@KtOAIa*I6I(LPxjz5(HWTcie@*Pdjd5rL>L>QlnupX3=+l>39hpPRzWZ+u zjBU7u?Cz)}X-7Pe`jcC1b|lFN9Cjpi4teK$Y-{FMOXBRFwSB)4qEl}*_+}^lZ`UmW z08SU>?<9SN2+^d8L-9R85|$%{0}hSi#N_V;uW>`SXtMRnmf}2kxefbD_r}%VN?ijW zC(&yws=)Ob@UH(mvOo)~IsKF-_UVNS$Iu7j#K2r73?YehMH)txzkZV;?nBaq8k8Hej6k;}fDMMq8KHJS>TQ)m#eBdGHi_KH zcsb<`eAEV&(}8CP$q%6Xm*?5e42+TesZS*Q!48kCUL$Ls>{T=i{(~2C zFN8}fC_G%0Qk6&+T#byma$1o@R_;{>W$;5-1Mm2p?!)rjdILO!-wLtJtzh!(q_%_# zee2Ju!w9`)%=OQMP5b$g%>i$8P?;a8?umXRg0U3UAuU7EER_8OI&A36wHdC{ju@`f zD(7BPU-bLaVTh+ZdX!FLTo;NCa_U`3?w7Fv2*(z0L!a#EPk|0uEPh(5Z|K1i&%`;z zw7FA}L5%|i%++7UfIOO53H-5EIw)gz1be(S7Z?5_oZH3tH(#Zy>H8Gh<6QZ?eupbz z441yh6KsM8CQ0xi(;zQjzXs-pXRj5kn>EjNOG@XiYcQ!3`73lK%Lk0!TYzr{@gU0w zJAv&6tuu5+34Z+0WgAVy5n}s*S7;J;GK}Hd#?%aA06UqTz+MC3yfx`tW<=I`D--$U z*ML{@D8N4ipdo{}5I|~vw2O!l*#kZ2kr|H2ahefrPC9$Kg~|!S ztw!q^P($nvRJW-o$LuN4nHQ&3rfEQL(Xx8xH`Bio*O^tSkp+GtA+zB&?l4=6;Wn<$ zyj&2Yw*q0T)}6n*()*{P%Ylae!d7#z8x4G4_XSu>1S`jGa?x?YRY z)$Lcmejb4}qstHMrD5@rhOo#O`yTPMH*L32KNJ9Auei+!VUIX7b(w5hG3i#7VNmc! z-1O@A!8D85(7F^uh6ubnmhfPR_lNK&zu(EHxgd3$1@#W%=TZy~-^c>G9Sk1za5Ego zSF*oL0SBu6G=&ne>yK7NV3%)m`z5jzCXX31BpyO zst{Gve8a{X6$>j^up0uCBmN>G8-mI8r&Z@#Slu3k$u$*GDXTRVNr>RHq7ucZF^5(Z zN+=UOJ&%Dz6y#dG4K(lF{1Az#by-?V=;nk;HV$#_>VE}R7gh}*5v0Ia)4u|dkOJHL zglN;-z+X!F1Gm8`spTrPp{GZH?-o}Ip$aKtO^x4M`iPY-KEMeiWy_B>#|K<93}igL zTijT^_t6n^!e?mA42_bs;67f{CdbIk?(=G`htvIpc&K4l9Qhrlx#z41lho-5C=Jw; z_=0zD-!?_Fz_mXzxjaUQgk)`_}t z5$1(mjKsFL(G@?bG-aH4B^Y^Sqk9nCe2qcRPcvo4a7!5zYCV_qYeudX5S8#S_%uhx zBWW8B6j;lV587*tf87QRJj8q+h{92S~E?kn;m6poj@=7 zEZMX`OM1aN8sNMdtJiX@Mv;-WXFJPssIVq0-r9w6jBTeOpb%#BM8q4v3V(!wf@0|qLz?P>~z&7><;L4(<^ZXLr6d?cYx}?Ht@mdy+7GDYDe_S=PlY3Q)?P zv-exs+?Ize2+6%OIB$6*`&ogiP4EN&ff!|I*ABil>Hh9+!0lpp&8GEfqL%+eAs3WQ^CgmZ48H^IFaI9zPf^4~q9 zTtBoxA9*ke{9ugNk)Oj)1u^#XME004cKSIu_%<<%g~-%Ge5G3mot<)mRHAut^Huey z+kl5_khk-IEpz4ry|zfc#kwa`Te3amU!A^@b9W`C z=YjE7bC`}O;)py^IygFV$OV&c1Ja-Gwdo{hCZu~MkK9MY0>e`rmu7Eo12^_~YI$}| zS{M!M&vP=l^K?G|UQsLN^-SCvU(uZ5;PILfFL-Ecc8R4KA^+XxY16mv%;?kAn^E5n zhxpluM?Bo!yTs_MLKxu9mq922in?WSaGxc@5yff5U6h-~B$R-VpPdTq-S0IU6k!ZI zqHHe2JGr9lzx}%iN6^n+W}maojEh!3c%~df`561u>|UcQ6GO^gR=C?|M*G?+^Dxss zQr`+J4L!<|FyqbIC)w_Kg&!HIDCs4cHR;x3fZ{s%pr)g#o*7QgtP7s$wSK8H5DZ&D z-tm!LoX1w0r>HVre0BFyg(I=gU?TeF%dAH;COWuy`af|@W`XEvB4xUDT6*RW(bO0U zbpv@cP0$9!G??g2K&PRFWhhK(iuN9LFtCPk;vw6jF{ozk(~JF2b0dO?+8Kp$2$ygH zu7uxQdS9QBYcOfP*henLM7g`wNGF(1HNS03 zh>$J&3Q}wgm#s8g6;TL%y^UL*ez5G30}dkn)(wdG^iON7*VmUCq?e8SYEfLm7y%y9{P!+_~0<{RaH5a`JhTdIFW?B za8#i_KzX7<$#$~&pJdWQ&C&VSbMihQtZ#=#^h-s~Ap>g_Wkse)46u)&JxQPibR|Jv z)5=YNfK0jRJYwn1ROew5X6ZcZ0<|l}2}?|$hvSHBjHBC6cMhFQK0t?0!dcA6%7i~D zhE6XAiHQ2-8T~mE3ecqlCs8vHWTfai=(@bl(iwq3i zbiYLeZwO^o$!2+1pzQSez0iLu1>KTiO~OZaqDPZh--EDdWUCNI?zSYB$_Cd{9YQ0P zvHQs2IniTkr($1F^G|9&lik~pscJc8vcKcS80JllJXj*q&G&*2q3DE4r%2Y#$W-x9 zpi1muOKM5-+Ai@6s}Sh`i67S+yEu>WLd+2*p_P097eLsOR!>ApA0o*_7gT}2+N|NS zKEjrjNsEe|7z7x>Fgcz!`{y)sG@c|M?qxrys=nWfZ*HlPX2p5W=`bO25(+KB<_Hz* zrFmtMtLSdS=5(Uv)Q=s)Z^9_(qysH0)V$*=Q1UR`pGV#4bE9MabRwq~Q(@dZ@e3WF z3R$QJE#d&pIJ}<E12EWP69G0RT|5qy$Z`Nee)?L!5{ z$Kydb0xdXF$|MS=)z*T8(4RRBeTVc@3#di#br@Q=8i_@i&vmomSP(O z!U+lLQ1r4>YH^G32KkDy=8E$&+a%_X0@l$j@8&G)SFoo1c3zwh&cYANHi~a4&cYEn z0|{Lgempi4a#pw~c&EW}qn^+ygI9spU+-WS6 z*Vv?!aQ1Jizl$tdK1^&Mz7N5zuQ&Ek4t@%%God4S3q;xQ=H8x9rMdk=5bk6@xd5V6 z9~Y5|w-mWsoIl(5%O9z4H@P%_d#x-!V5ovW%t>iJCqLr5-=`iEzeekD@gRymCJw(C z2PBg;^5=~uM(srQSIB$o@LjWYabFvYduqO~)~NO+vQKA(GB=Guo!A%iz6!w)`(nb= zN~iN4O$nx72A%T6e5+xSQinv zlEG>aV2Z%x#1@@6&o7=Y4-^!hDukD%(}_ane>B!hCiu?zAa;a(d9i;r}k~n3(fJeSd_g zQG*#zQ}1h-NmVg==xhPFO7O3J3uY|nsyDqTOWh%1CBkpw^63l_ZZ?m~HI``xVKy&g z25^*LKmQgT1JErvcfdA&EI^Z!G|Go;_zIX#ASv0{f2j&bao%p22M!I|R{LdpRX zj&EmMG4)Dkyk<$x+NM&!GAa$#d8l<47nRRs9%OGKa;aT+z7mqV+RJItyUO7YR2t+4 zFImvLp7IQJD15JBUy&bS3Aems;t)IADy~;PTayddUDrXXnN>Zb4=uny_8^)y&!Xz0 z4L8KU|D8UGDpe*0YWhA?*;18?v*oD??u+gbzUe4vA&g|gGtT>~`lItjcW&S)#f$w}??wNV2(O0`D zc8S2chhO<^X7VzXTZ>T1ta?r?BcZkR?W0mr$6SSXN~=}Iw6Bb6jT475pF4AfdfVh5 zmX4&qA1#YIR#e4hii%oRQmu!bJITHEcg(H-&2s3hcut5QQ=D;O+2m*_?irCW9WpMd zc$%9T=F~*)g77_KRjQYvy#P z4UjlW0=BCDA!wLpiuZ{yJ;l?Ba6L_$60YdMBl-cG}(5-zThZhmJ}cL$i!M zzERO->R`u@;tEyY=1;pya4duil#B(H=x$^K#`uYtLg~KuZ*HX0K*on2OHBBRPvr78_K6cAuM>Xpq^PJJ1JM*GZ9$l4|kYz0TqP)=xSK~RQ{ofm*5b}JG z?u>Mz6JLp=OB$t$-1o#XfYF@tzRX78w0>fc#m#C1W3a`|k``X0OuQV1p{q073su2w zYS^34srLQCOXNsm?$-L)`JY^L(%4jx>E+Ala%_~u9zBV94--%%Lr~VVTuwMIor;SfMX)q>P8|9mcy^nG zJ_s!$X9q3`+&3AHNw4v-ORD%JN%TV^wFs5$uc*~U;Sy8;@eM=v0~c~q)PSQ)C$!%U9AR>mK5gi-ro=vH+r8Qi>2>Z2L%!7U5ceUHgPyNwH^@_ ze;%Hd-UQ=2$Bx~FwARXmHr-l-)@77VaD7t?Wfsbx&Q#V^V)Lj345~E82CUjPcAw&@Tpi~53vN8)^eO^?Oz!}-=KoJ-ajO$Mf%)_ARQvXNHWN@n(Tden~jH6fTR<&a;k{eB19uXW1}P;xzZV){&gg zQMxx(jde~r4L7y};Cf5~WeYVm&kv*}5)Yh8*lIBkxMUdRM4SqpajKSV8X2Fa{_y13 zPFZMlfTg6Yo1w|W2Bi%5$<{ovm5dZQmYs_| zH3wi;}h1-eD=t>=$}3xn%`9ukKt=p#ToAc?fzXwLKs8QT8-;D{fy< zlu!?dTYj9PY4YJ@C#jSw!=&OD0^hK%PX9B$qPsO;#oV6nFm~tXUdLUY6t2%cl4_5 z?i2D9@w$7n&oWiRCa*c3FE;nCB&wHWQK5??jD>Od^ms*eQsj6G-12728E1QuD&a1n zpa9_HE47cIMVJ2Rmu(llADretWi;*sKA~BvoP;zh zCLR0#N=b+%GA7?d=5`iW7*#DDu8v$Kk%$k8-hVsH)|i~xc^UE#89Q)noL}&g;kc#( zabujfTn0%D@3kJjOq-u!M*j}u9n8x)oaqIS{3v>J-isLywGeD^4GG#9zLvS?H+aSBwr({(Z{F)v!36pk*As#_Y$uPvrl|ZV7iaX_)E1HUlW9X_ zU6Qwx%A&y2pn2!A!71XVzg;&M%^^{0m;2Yw63990Z)d$tlv)Q)u!{nfL2agtbG7Xd zZV_FfJssS6@A6Y^&fT(+_HZR#$tGq^$J<4vB|~q1n&9FyGiqr4yd=29=$ zSA3TPQNq@U=(M1({&p&dPJD{~S92kDm3aB@Wa*XIQEd_S$jqZEAzwvhq8;)r@&yTmn&zlxb;8>hU*Kl zl6?Y$=5kl1nF67E{EG~PQgiz=y88$)W97WDy!C%$-yuTgAjmv+)=rfJlOA1C-ZhP; zv~G0;{lSP`G^OlUm1%GN&>jdU97rbYHB)8F@P_;m4k4QbI~g<6kUTZK|uYQ>7# zP}Hc{u~$)5s`hN{(yIRSdA`4Y&v~D7?tk~b?(5$3nfFkZ&k8YIi{|Y-7a#dns`Rl5 zGgFMAb~L@6yI19wu&G?mU*Qnh*0H%4e5?|*w=vQDq99e#=b04P{JHt=0Q=R<+UKnS z{$59^?k_A8RXlVt6(?w~D7dTm#!5EYu77%(@#3lNe<_xBghNm`j#IKseWFd?}k|Qag`qmF@C) zMt8{TR!Iwofe1x3=Ev`P%Q3|@=TXs7#tT*}qOtXMu@<42H;Wsqe0gn?at`ujj&m*; z>3hHuO5FsQ3wt<)c5JmeX1~y?*P1G|D*>a7_fg1_*>fvJJSzO*MeQpEXH+%2)nA^R z_z(qFh?-{H8xI+~YI)cc9s6QxbZ2^*o>aNDu=*X?uouGm2R>9Apm3tX&ngMZNTcOT z&=p%eylH^ZzD7%-w544f3P-C@kaCYG3CRUmILHUv4slDhp*e$ieBQ=>?$3- z#TRattkpcDyXukp{^VwBpSOxEov9H0q8|{eHSN^&Wc1gmwBX^_w(1hY4Y7s?ye|D1 zWXd5)P1|Je{BpHul!4a$JbBZXD4cm4yEt`G_&8HhorKOOW{UfuBn_JTA>BOpL!xxU z?;+%&FI~sX+SvK13jkwGNck*wKg2!qQw`mF@y&rsnH69S%G~e~$$^-kmBC^s48d5j z+&)BULbWmrMM&ZZo1VkF*nc8MNSFn=0qwj{IMrXzN5-HwF9kn+E_$*pF8g{xZkPLH z@F~9Ji(%kj8nd63h>aM5tWf1T8<)O~;@FmxaWGOz=A1x(>sh%v$*ox$kX^FTW{${8 z-x$hEpHHN4EqTTMC)|HsIY`ttHKf4U^&0#40$b^+TX2f@OG{nlEKAOOSKL|p-`DyX zk-WY@lzSp6q8$Z=Z?JH(E9)Y_Q0f$y&;Xfx(VPJBr-*nXNTAuMx~^7Y^{|+4+;d2@ zqj}c}LJZ0Vi9elb%}bx7%;(Y;lSQn>@CmVXecX&%gn#?E5d$KIoBmZi;~rY@RvK-Y zC{uLQ;P*v!HmY}*qY46;yBpT{!6K)BmDO59gjFbglnYJ1B}_td^YtZavtrt)aSI`~ zsI31+G13+1r8@!t9i_YcLsef4bRZ8Zs@4|^>3RZbm`X6|$f5K*NUW=Bc_u1wg^RO} zV&%g_nLoKew2Nz^sAqRTHvj1I>DaQxWHu_`NA@30O~n?|HtDf^qhVY+bDH{K)%UDs zYtuLp*SBEyw6Aci>bra&NrD2mLB2upA_|KMt=AgdNM`t&7a4|e9Myt?GARcEtC_r1 z;5%ASr{tteK?|Mm`OL^*Gd`aW*)n7f0G2{Ej4w?N=?u4^`uK(gi>45dm2WGLtTCtb zXG^*suCB{tGzZ+Lcv8V29$lHjn$oMgT*i9q^{RhBN~pV%zgqWUTFl|46k0OnvW)T( zTfMmnf}b9TVaM$s>GF4#y(BNe1)g1j8dEO=F1j!zn?+Z0vH#me_ICcfqbDr)$ca;< zJJ>2Vm?K3`YI)H8Prp6V*Lne8C4aLyd2iF;um>q+9wt2HYK=T;MT`371`LLt6R`8 z>M5X4eoJh^tcwSiz6dyeeDeYDuHic)XS`eYtot9QHd$cXYI z3tOg*rB>s$Lr3ADz2l>s#|GE0HnZLus6tP8PnEvR&?60~Ez!exOeUn#w)&n=-7)%O3PBQRzPjPekZK z1(z<}iNw0lqp4&+-FR0&&oh2G39|2=hugFl>}sLV3lk*as9CDDR2Qk0fV>Zgj<3Cb zB&C2oNVvrXaPRAOc5*ZznL>{P(?PRi#I*o=Tt}C1+B2uDomc_Td%u_?ctuN=#Bvcw zhsJu2GU5K>SH3n*mw93S^^;z8Z7ZNYx&MJp&r#!?e}R(iJzic|=~S3tw}CU0$i)cx z$ijP^yknP5xa#zUuB6)xo(koPdaF4bDX4z^O|3zct>kz+;m&`uetNp@eO){tlEc?9 z6?YQAD(7H#9(a^m^0l1Fp?l%av0UMoUF5fr*4&w1Lhg*z4AGAI%e$FX51;ne=OBn= zCzmu;KS^qV6WyOYAeCgqebVg|iZhV8bwa%%wM#GSxWKtf@8YO>TEy}T*v2&|B|E4v zO68O?N}wD>?7YGQRmio9cU&ar;_edD*%AX{Od{Iy4ARVU6}ovz#y?cGB4 zZ3o2#n>=HECww|Lb`s&rtyqUX+F>=#VLR8Q)Yg?s{d>~3@h-np=I3&29lP8`Ke-!f zI&HaSToG&h>XGv&R#d;tb6m>1j!Z1Nnq=rCjITa+ zEz&&$S5R36LZ|y|PbOTz(!LH%-f^OO!iat-JdOz_oG8y`i^ha_m2! zyPdRcB2g+8%u%{M@)IyI-%QpGpIL)Ch_7$N6Cz57 z*j8dGW%0&$?`V04*n*K7Z*zPf(pPr3t~mM0&n#5Z(DkHSZw*VFk8E4)I44$yxZkeq z{Pw%}p&6F``Sv2_+mubMsJaGvn91jMv%6wU03Fnah6zX;9w|we`)+5KTZwGWL}10- zdBk|XM@v6TK}mlvbkrs+ky{DGOT|PK@m#I%bX$PlRMr<6(G5gEqDZ3?>RMHDh#tzSYTU^mV5-N<&+RBYPb70KFX zFD3F*#DM7DFm2oXlm`0fVqX+G$;2%|QYs;!QL^9Ds>{2G5-`KfeeenQMgz2dO1Dm# z{*2>0WbE-Fn3lg&7$DU#qu=U|tQ8I@UO;rcJY&2aZ_vp=N;v zKSr*ai~aQl6lO*yK2LV!IU(KBH*n&WOoCwIZ3LT8RhSrpjjEDg=eh1EIM#)$ zz*4YDWl$^@sc=kh`S}{8vz{d3#9kA>VIWqW69p{i>XDHB%I2AqJ;`FN^|*`)9=^y` zL2zrUW>MQ^8Bvv@4xrOke^0n7Yv=uz>6#8XpFK z`ZeANOzj-wMGeTKsfvnkWED84v>Ri%eP-WiE@T1I=#wtNmWVjw4RyjCTspf)`I#}_jyp7sn zE<=_|n!gt+uI-j@HnA`xO=@}{Pycun;4v!Bb1=C-qG>l@e)v*(X;Z?)x{*#$Q5K20 z)0^V4Cj&88@%$OF7`rcb#iBIK5+MOi=gSvvwe;0{e+v?s_V zd^3kRP~(B-GknDTSr{D3p5o4F-Zwa9^P!feECt{JMd6hM<+Z^K`u{2%^}(!-^na?d z(DhiFy}n8k1lg%zh5sh(SpVt32KE0TDSB`$SqQf(Zg|p;*L)YhOKFLnG>vs#U;Cz zEr<1#?e)%y(*qH?ElN3GcF!Ttp7pAaFh89Kk=-y=()U)tbNnegr?{OY&!FR{kVvaT zA1qScbSube>(^@RIO>MlnBeSYVI78EmIL<8xLYH#_*;GJ0Jh=$kV@}PO(qwt0l{m(sjj&2A9 z(6_uD<~t55X5k!<_FLTi@j>!V3%{pxwO}!QyMvQd4QD~Z?8kC`d#a%D)+s=D=@qUX z2YuMoOhQ#PYtUX`O;l@gme$`S$pNJF(79Td#Ij9&Y}?C~<1iw9qHX;K;CKdvkG&`b z-PzrWbw>WZ(? ze)R7X0&F@6VHqNIrGEZp{-3u(OL5H18_32DoCDSE@NyC=5YO3(4P>!%h2QS$a^a$; zNn!Vn_ahc~5PT2frzuzyoFt_=xai}$3#F|D-++3E>2J`6V|%|u5F7<@N!6eb`ikf$%i)y-0WHdLuwwV7+g+(W#A(^AKSo`ms{WS1#yk zQKzd+XdY-JhORIg{227G2I9&+`Q(!{f_To$4ohl0YL-3>jA9#VoU;{ z^RH+PxVxM8P5DqNv!%`s1x_qO?IuG1{yR|Y|wwt zlXooaeq-b5p@oX%iVkZte_q*GO9d@mFHT)iVvnE?A`c9uOaYKTzD(O}@G+<09@5h7 h$@yHP1C1Ot@7&~(2n^9-lKsO>(--@Rv6+M7{{Rp$pb!86 literal 0 HcmV?d00001 diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/manifest.jsonl b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/manifest.jsonl new file mode 100644 index 000000000..f9e549154 --- /dev/null +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/manifest.jsonl @@ -0,0 +1,2 @@ +{"url": "https://assets.publishing.service.gov.uk/media/6ae187a2779378370fd80a9a/Table_6.1_and_6.2_Budget_2026_Policy_Decisions.xlsx", "title": "Table 6.1 and 6.2 Budget 2026 Policy Decisions", "date": "2026-10-28", "publisher": "HM Treasury", "sha256": "359958df6b1ab9746073725f6f8017ebdbb55040e57cf28612891f1e294a7f15", "bytes": 67126, "doc_type": "xlsx", "access": "direct", "retrieved": "2026-10-28", "note": "Attachment of https://www.gov.uk/government/publications/supporting-documents-for-budget-2026. Downloaded 2026-10-28 and again 2026-10-29 with a browser User-Agent; both downloads hash to the pinned sha256."} +{"url": "https://assets.publishing.service.gov.uk/media/6ae0f93fe4367e5c8b0467d6/Table_6.4_Recostings_Table.xlsx", "title": "Table 6.4 Measures announced at Spring Statement 2026 or earlier that will take effect from 29 October 2026 or later", "date": "2026-10-28", "publisher": "HM Treasury", "sha256": "b0449df1737f128fdc32ffeffbe7ed11876e7ab1c80d14dd9908e68a38386d3e", "bytes": 76402, "doc_type": "xlsx", "access": "direct", "retrieved": "2026-10-28", "note": "Attachment of https://www.gov.uk/government/publications/supporting-documents-for-budget-2026. Downloaded 2026-10-28 and again 2026-10-29 with a browser User-Agent; both downloads hash to the pinned sha256."} diff --git a/tests/test_ab2026_registry.py b/tests/test_ab2026_registry.py new file mode 100644 index 000000000..957d848f2 --- /dev/null +++ b/tests/test_ab2026_registry.py @@ -0,0 +1,262 @@ +"""The Autumn Budget 2026 measure registry (#230). + +data/uk/ab2026_measures.json is the key space every AB2026 claim carries in +measure_key, and the computability verdict that decides whether a producer's +row gets a PolicyEngine counterpart or a pe_gap finding. These tests pin the +rules the port depends on: every Table 6.1 line once under a key that cannot +move, every gap with a computed name search at the Budget-day engine, every +dashboard measure tied to a receipt by hash, packages that are exactly the +union of their legs, and an outcome for every pre-event key. +""" + +import hashlib +import json +import os +import re +import subprocess +from pathlib import Path + +import pytest + +from pipeline import ab2026_measure_keys as K +from pipeline.validate_budget_2026_registry import key_prefix, load, validate + +ROOT = Path(__file__).resolve().parent.parent +PATH = ROOT / "data" / "uk" / "ab2026_measures.json" +REG = load(PATH) +BY_KEY = {m["measure_key"]: m for m in REG["measures"]} +PIN = "2.151.0" + +# The seven measures the UK Autumn Budget dashboard ran on Budget day, by +# Table 6.1 line and dashboard id. +DASHBOARD = { + 1: "household_energy_payment_2026", + 2: "domestic_electricity_zero_vat", + 3: "fuel_duty_rise_deferral", + 7: "carers_allowance_earnings_taper", + 15: "trading_allowance_raise", + 52: "state_pension_income_tax_protection", + 61: "property_income_rates_up", +} + + +def by_line(line): + return BY_KEY[K.key(line)] + + +def test_registry_validates(): + assert validate(REG) == len(REG["measures"]) + assert REG["fiscal_event"] == "autumn_budget_2026" + assert key_prefix(REG) == "ab2026__" + + +def test_every_table_61_line_is_registered_once_under_its_written_out_key(): + lines = sorted(m["table_61_line"] for m in REG["measures"] if "table_61_line" in m) + assert lines == list(range(1, 99)) + for line, slug in K.TABLE_61.items(): + m = BY_KEY[K.PREFIX + slug] + assert m["table_61_line"] == line, slug + + +def test_table_62_maps_to_table_61_measures_except_sovereign_ai(): + """Table 6.2 lines are financial transactions of Table 6.1 measures; only + line 4 has no Table 6.1 line and so is a measure of its own.""" + own = [m for m in REG["measures"] if "table_62_line" in m] + assert [m["table_62_line"] for m in own] == [4] + for line, slug in K.TABLE_62.items(): + assert K.PREFIX + slug in BY_KEY, (line, slug) + + +def test_keys_are_unique_and_disjoint_from_the_pre_event_registry(): + keys = [m["measure_key"] for m in REG["measures"]] + assert len(keys) == len(set(keys)) + pre = load(ROOT / "data" / "uk" / "budget_2026_measures.json") + assert not set(keys) & {m["measure_key"] for m in pre["measures"]} + + +def test_every_pre_event_key_has_an_outcome_that_names_real_keys(): + pre = load(ROOT / "data" / "uk" / "budget_2026_measures.json") + outcomes = REG["pre_event_outcomes"] + assert set(outcomes) == {m["measure_key"] for m in pre["measures"]} + for k, text in outcomes.items(): + for named in re.findall(r"\(?(ab2026__[a-z0-9_]+)\)?", text): + assert named in BY_KEY, (k, named) + + +def test_every_gap_has_a_computed_name_search_at_the_pin_and_somewhere_to_go(): + for m in REG["measures"]: + if m["computability"] != "not_expressible": + continue + k = m["measure_key"] + assert re.match(r"https://github\.com/PolicyEngine/", m["action_link"]), k + assert m["pe_reform_delta"] is None, k + assert m["why"].strip(), k + ns = m["name_search"] + assert "Parameters:" in ns and "Variables:" in ns and "VARIABLE" in ns, k + assert f"policyengine-uk {PIN}" in ns, k + assert bool(m.get("out_of_model_scope")) != bool( + m.get("policyengine_uk_development_item") + ), k + + +def test_development_items_name_their_upstream_item(): + devs = [m for m in REG["measures"] if m.get("policyengine_uk_development_item")] + assert len(devs) == REG["triage"]["policyengine_uk_development_item"] + for m in devs: + assert m["upstream_item"].startswith("policyengine-uk:"), m["measure_key"] + + +def test_the_triage_counts_are_computed_from_the_measures(): + t = REG["triage"] + ms = REG["measures"] + for verdict in ("expressible", "partial", "not_expressible"): + assert t[verdict] == sum(m["computability"] == verdict for m in ms), verdict + assert t["executed_by_dashboard"] == len(DASHBOARD) + assert t["executed_by_dashboard"] == sum( + m.get("construction") == "executed_by_dashboard" for m in ms + ) + assert t["out_of_model_scope"] == sum(bool(m.get("out_of_model_scope")) for m in ms) + assert t["announced"] == sum(m["reported_status"] == "announced" for m in ms) + + +def test_the_seven_dashboard_measures_execute_by_receipt_at_the_budget_day_engine(): + run = REG["engine_pin"]["budget_day_runs"] + assert REG["engine_pin"]["policyengine_uk"] == PIN + assert run["certified_for_model_version"] == PIN + for line, dash_id in DASHBOARD.items(): + m = by_line(line) + assert m["computability"] == "expressible", line + assert m["construction"] == "executed_by_dashboard", line + assert m["dashboard_measure_id"] == dash_id, line + rc = m["pe_modifier"]["receipt"] + assert rc["path"].startswith(f"public/data/artifacts/{dash_id}/"), line + for field in ( + "policyengine", + "certified_for_model_version", + "dataset_revision", + "dataset_sha256", + ): + assert rc[field] == run[field], (line, field) + + +def test_the_remaining_expressible_lines_are_deltas_on_current_law(): + """Savings rates (line 62, off the dashboard release) and the BADR limit + (line 60) have no receipt; they are parameter deltas the engine resolves.""" + for line in (60, 62): + m = by_line(line) + assert m["computability"] == "expressible", line + assert m["construction"] == "delta_on_current_law", line + assert m["pe_reform_delta"], line + assert "pe_modifier" not in m, line + + +def test_the_bus_cap_is_partial_because_its_pre_measure_world_is_unknown(): + m = by_line(4) + assert m["computability"] == "partial" + assert m["intended_construction"] == "reversal_on_certified_world" + assert "construction" not in m + hist = m["engine_parameter_history"]["gov.dft.bus.fares.cap"] + assert hist["2027-01-01"] == 2 + assert "bus_fares_gap.json" in m["missing"] + + +def test_packages_are_exactly_their_legs(): + tables = { + "ab2026__package_total_policy_decisions": set(range(1, 99)), + "ab2026__package_household_energy_vat_and_fuel_duty": {1, 2, 3}, + "ab2026__package_policyengine_seven": set(DASHBOARD), + } + for k, lines in tables.items(): + assert set(BY_KEY[k]["package_of"]) == {K.key(n) for n in lines}, k + tax = set(BY_KEY["ab2026__package_total_tax_policy_decisions"]["package_of"]) + spend = set(BY_KEY["ab2026__package_total_spending_policy_decisions"]["package_of"]) + assert not tax & spend + assert tax | spend == set( + BY_KEY["ab2026__package_total_policy_decisions"]["package_of"] + ) + for leg in tax: + assert BY_KEY[leg]["table_61_head"] == "Tax", leg + for leg in spend: + assert BY_KEY[leg]["table_61_head"] == "Spend", leg + + +def test_a_package_is_expressible_only_when_every_leg_is(): + for m in REG["measures"]: + if m.get("construction") != "package_of_registry_measures": + continue + legs = [BY_KEY[leg] for leg in m["package_of"]] + expressible = [ + leg["measure_key"] for leg in legs if leg["computability"] == "expressible" + ] + if m["computability"] == "expressible": + assert len(expressible) == len(legs), m["measure_key"] + else: + assert m["computability"] == "partial", m["measure_key"] + assert sorted(m["expressible_legs"]) == sorted(expressible), m[ + "measure_key" + ] + + +def test_the_combined_receipt_covers_the_seven_package(): + m = BY_KEY["ab2026__package_policyengine_seven"] + assert m["construction"] == "package_of_registry_measures" + assert m["pe_modifier"]["receipt"]["path"].startswith( + "public/data/artifacts/combined--" + ) + + +def test_the_registry_is_regenerated_byte_for_byte(): + """The JSON and REGISTRY_KEYS.md are generated; --check needs the pinned + engine, so it runs only where its venv exists.""" + py = ROOT / ".venv-pe2151" / "bin" / "python" + if not py.exists(): + pytest.skip("the policyengine-uk 2.151.0 venv is not installed here") + out = subprocess.run( + [str(py), "pipeline/build_ab2026_registry.py", "--check"], + cwd=ROOT, + env={**os.environ, "PYTHONPATH": str(ROOT)}, + capture_output=True, + text=True, + ) + assert out.returncode == 0, out.stdout + out.stderr + + +def _dashboard_repo(): + path = Path( + os.environ.get( + "UK_BUDGET_DASHBOARD_REPO", + Path.home() / "PolicyEngine" / "uk-autumn-budget-dashboard-2026", + ) + ) + return path if (path / ".git").exists() else None + + +@pytest.mark.parametrize( + "key", + sorted(k for k, m in BY_KEY.items() if m.get("pe_modifier")), +) +def test_each_receipt_matches_the_dashboard_commit(key): + """Hash-check each receipt against the dashboard repository at the pinned + commit. CI has no checkout of that repository, so this runs where one + exists (UK_BUDGET_DASHBOARD_REPO, or the default clone path).""" + repo = _dashboard_repo() + if repo is None: + pytest.skip("no uk-autumn-budget-dashboard-2026 checkout") + rc = BY_KEY[key]["pe_modifier"]["receipt"] + blob = subprocess.run( + ["git", "-C", str(repo), "show", f"{rc['commit']}:{rc['path']}"], + capture_output=True, + check=True, + ).stdout + assert hashlib.sha256(blob).hexdigest() == rc["receipt_file_sha256"] + receipt = json.loads(blob) + ex = receipt["execution_receipt"] + assert ex["reform"]["reform_sha256"] == rc["reform_sha256"] + assert ex["reform"]["binding"] == rc["reform_binding"] + assert receipt["dashboard"]["generated_at"] == rc["generated_at"] + assert ( + ex["certificate"]["certified_for_model_version"] + == rc["certified_for_model_version"] + ) + assert ex["data"]["dataset_revision"] == rc["dataset_revision"] + assert ex["data"]["dataset_sha256"] == rc["dataset_sha256"] diff --git a/tests/test_uk_ab2026_harvest.py b/tests/test_uk_ab2026_harvest.py new file mode 100644 index 000000000..6cf15c782 --- /dev/null +++ b/tests/test_uk_ab2026_harvest.py @@ -0,0 +1,321 @@ +"""The Autumn Budget 2026 harvest (#230): vendored data, checked as data. + +sources/harvest-uk-ab2026-2026-10-29/ stages every producer's AB2026 claims +before the tranche-2 ingest exists. These tests hold each staged row to the +row contract in that directory's README.md (the AB2025 contract with the +AB2026 exceptions): field names, closed enums, the FY-end-year period rule, +the fiscal_event slug, measure keys that exist in the AB2026 registry or, for +a Table 6.4 recosting, in the earlier event's registry, sha256-shaped +manifests, and per-family pins that the ingest's exact accounting will read. +""" + +import gzip +import json +import re +from pathlib import Path + +import pytest + +from pipeline import ab2026_measure_keys as K +from pipeline.harvest_counts import counts + +ROOT = Path(__file__).resolve().parent.parent +HARVEST = ROOT / "sources" / "harvest-uk-ab2026-2026-10-29" +UK = ROOT / "data" / "uk" + + +def _keys(path): + return {m["measure_key"] for m in json.loads(path.read_text())["measures"]} + + +KEYS = _keys(UK / "ab2026_measures.json") +# A Table 6.4 recosting carries the earlier event's key (README rule 2). +# AB2025 has two key spaces: the #136 port registry and the events inventory. +EARLIER = { + p.name.removesuffix("_measures.json"): _keys(p) + for p in (UK / "events").glob("*_measures.json") +} +EARLIER["autumn_budget_2025"] |= _keys(UK / "ab2025_measures.json") + +COUNTS = json.loads((HARVEST / "COUNTS.json").read_text()) +FAMILIES = sorted(k for k in COUNTS if not k.startswith("_")) + +REQUIRED = { + "source", + "source_model", + "value", + "value_raw", + "normalization", + "value_kind", + "period", + "time_basis", + "conditions", + "reform_hint", + "measure_key", + "attribution", + "benchmark_class", + "baseline_policy", + "publication", + "source_table", + "source_column", + "quote", + "parse_confidence", + "status", +} +OPTIONAL = { + "metric", + "proposed_metric", + "unit_concept", + "proposed_unit", + "proposed_baseline", + "note", + "local_artifact", +} +ATTRIBUTION = {"own", "restated", "same_assumptions"} +BENCHMARK_CLASS = {"different_model", "administrative_fact", "same_assumptions"} +VALUE_KIND = { + "point", + "central", + "range_low", + "range_high", + "approx_chart_reading", + "cumulative", + "categorical", +} +TIME_BASIS = {"fiscal_year", "annual", "point_in_time"} +CONFIDENCE = {"high", "medium", "low"} +# README rule 6: quotes from producers outside the Open Government Licence +# stay under 15 words. +OGL = {"hm_treasury", "obr_efo", "hmrc", "dwp", "ons", "scottish_fiscal_commission"} + +from scorecard_db.models import Metric, UnitConcept # noqa: E402 + +METRICS = {m.value for m in Metric} +UNITS = {u.value for u in UnitConcept} + + +def families(): + return sorted( + p.name + for p in HARVEST.iterdir() + if p.is_dir() and (p / "claims_staged.jsonl.gz").exists() + ) + + +def rows(family): + with gzip.open(HARVEST / family / "claims_staged.jsonl.gz", "rt") as fh: + return [json.loads(line) for line in fh if line.strip()] + + +def manifest(family): + return [ + json.loads(line) + for line in (HARVEST / family / "manifest.jsonl").read_text().splitlines() + if line.strip() + ] + + +def test_every_pinned_family_is_present_and_complete(): + assert set(FAMILIES) == set(families()), set(FAMILIES) ^ set(families()) + for fam in families(): + for name in ("NOTES.md", "manifest.jsonl", "claims_staged.jsonl.gz"): + assert (HARVEST / fam / name).exists(), (fam, name) + + +def test_counts_json_is_what_harvest_counts_computes(): + """Rows, primaries, sources and the claims digest per family, plus the + totals: one function writes COUNTS.json and this re-derives it, so a + silent edit to a vendored row or a family added without its pin fails.""" + assert counts(HARVEST, COUNTS["_totals"]["pinned_on"]) == COUNTS + + +@pytest.mark.parametrize("family", FAMILIES) +def test_rows_follow_the_contract(family): + for i, r in enumerate(rows(family)): + where = (family, i) + keys = set(r) + assert REQUIRED <= keys, (where, REQUIRED - keys) + assert keys <= REQUIRED | OPTIONAL, (where, keys - REQUIRED - OPTIONAL) + assert r["status"] == "ok", where + assert r["attribution"] in ATTRIBUTION, where + assert r["benchmark_class"] in BENCHMARK_CLASS, where + assert r["value_kind"] in VALUE_KIND, where + assert r["time_basis"] in TIME_BASIS, where + assert r["parse_confidence"] in CONFIDENCE, where + assert isinstance(r["value"], (int, float)) and not isinstance( + r["value"], bool + ), where + assert isinstance(r["value_raw"], str) and r["value_raw"].strip(), where + assert isinstance(r["quote"], str) and r["quote"].strip(), where + assert bool(r.get("metric")) != bool(r.get("proposed_metric")), where + assert bool(r.get("unit_concept")) != bool(r.get("proposed_unit")), where + if r.get("metric"): + assert r["metric"] in METRICS, (where, r["metric"]) + if r.get("unit_concept"): + assert r["unit_concept"] in UNITS, (where, r["unit_concept"]) + c = r["conditions"] + assert isinstance(c, dict) and c, where + for k, v in c.items(): + assert isinstance(k, str) and isinstance(v, str), (where, k, v) + assert c["fiscal_event"] == "autumn_budget_2026", where + if "fy" in c: + m = re.fullmatch(r"(\d{4})-(\d{2})", c["fy"]) + assert m, (where, c["fy"]) + assert r["period"] == int(m.group(1)) + 1, (where, c["fy"], r["period"]) + assert r["time_basis"] == "fiscal_year", where + assert isinstance(r["period"], int) and 1945 <= r["period"] <= 2075, where + pub = r["publication"] + for k in ("title", "url", "date", "publisher", "primary_sha256"): + assert pub.get(k), (where, k) + assert re.fullmatch(r"[0-9a-f]{64}", pub["primary_sha256"]), where + assert re.fullmatch(r"\d{4}-\d{2}-\d{2}", pub["date"]), where + + +@pytest.mark.parametrize("family", FAMILIES) +def test_measure_keys_exist_where_the_row_says_they_live(family): + """README rule 2: an AB2026 key, an earlier event's key on a recosting, + or null with a proposed key and nothing else.""" + for i, r in enumerate(rows(family)): + where, c, k = (family, i), r["conditions"], r["measure_key"] + if "recosting_of" in c: + assert k in EARLIER[c["recosting_of"]], (where, c["recosting_of"], k) + elif k is not None: + assert k in KEYS, (where, k) + assert "proposed_measure_key" not in c, where + else: + pk = c.get("proposed_measure_key") + if pk: + assert re.fullmatch(r"ab2026(_option)?__[a-z0-9_]+", pk), (where, pk) + assert pk not in KEYS, (where, "use measure_key", pk) + + +@pytest.mark.parametrize("family", FAMILIES) +def test_every_row_cites_a_manifest_primary(family): + shas = {m["sha256"] for m in manifest(family)} + for m in manifest(family): + assert re.fullmatch(r"[0-9a-f]{64}", m["sha256"]), (family, m.get("url")) + assert m.get("url") and m.get("title") and m.get("retrieved"), (family, m) + assert len(shas) == len(manifest(family)), family + for i, r in enumerate(rows(family)): + assert r["publication"]["primary_sha256"] in shas, (family, i) + + +@pytest.mark.parametrize("family", FAMILIES) +def test_quotes_outside_the_ogl_stay_short(family): + for i, r in enumerate(rows(family)): + if r["source"] not in OGL: + assert len(r["quote"].split()) < 15, (family, i, r["quote"]) + + +def test_same_assumptions_rows_are_policyengine_computed(): + for fam in families(): + for r in rows(fam): + if r["attribution"] == "same_assumptions": + assert r["benchmark_class"] == "same_assumptions", fam + assert r["source_model"] == "policyengine_uk", fam + + +def test_no_hmt_decile_bar_readings_were_staged(): + """README rule 4 (AB2025 rule 9, #61).""" + for fam in families(): + for r in rows(fam): + if r["source"] == "hm_treasury": + assert "Impact on households" not in r["publication"]["title"] or ( + r["value_kind"] != "approx_chart_reading" + ), (fam, r["quote"][:80]) + + +def test_the_key_listing_is_derived_from_the_registry(): + text = (HARVEST / "REGISTRY_KEYS.md").read_text() + listed = set(re.findall(r"^\| `([a-z0-9_]+)` \|", text, flags=re.M)) + assert listed == KEYS, listed ^ KEYS + + +# --- uk_hmt_tables: the stager's output against the tables' own structure --- + +HMT = "uk_hmt_tables" + + +def _hmt(table): + return [r for r in rows(HMT) if r["conditions"]["table"] == table] + + +def test_table_61_stages_every_line_and_total_for_six_years(): + t61 = _hmt("6.1") + years = sorted({r["conditions"]["fy"] for r in t61}) + assert years == [f"{y}-{(y + 1) % 100:02d}" for y in range(2026, 2032)] + measures = [r for r in t61 if r["conditions"]["row_type"] == "measure"] + assert sorted({int(r["conditions"]["table_line"]) for r in measures}) == list( + range(1, 99) + ) + assert len(measures) == 98 * 6 + totals = [r for r in t61 if r["conditions"]["row_type"] == "total"] + assert {r["measure_key"] for r in totals} == { + K.package_key(s) + for s in ( + "package_total_policy_decisions", + "package_total_tax_policy_decisions", + "package_total_spending_policy_decisions", + ) + } + assert len(totals) == 3 * 6 + + +def test_table_61_rows_carry_the_written_out_key_for_their_line(): + for r in _hmt("6.1"): + c = r["conditions"] + if c["row_type"] == "measure": + assert r["measure_key"] == K.key(int(c["table_line"])), c["table_line"] + + +def test_every_hmt_cell_is_a_multiple_of_5m(): + for r in rows(HMT): + assert r["value"] % 5e6 == 0, (r["conditions"]["table"], r["value_raw"]) + + +def test_table_61_lines_sum_to_the_printed_totals_within_rounding(): + """Footnote 1: totals may not sum due to rounding. Every staged cell is a + multiple of £5m, so if HM Treasury rounds each line to the nearest £5m, 98 + lines drift at most 98 × £2.5m from the exact total; the observed gaps are + £5m to £10m. Tax plus spend equals the grand total within the same rule.""" + t61 = _hmt("6.1") + for fy in {r["conditions"]["fy"] for r in t61}: + year = [r for r in t61 if r["conditions"]["fy"] == fy] + total = { + r["measure_key"]: r["value"] + for r in year + if r["conditions"]["row_type"] == "total" + } + by_head = {"Tax": 0.0, "Spend": 0.0} + for r in year: + if r["conditions"]["row_type"] == "measure": + by_head[r["conditions"]["head"]] += r["value"] + tax = total[K.package_key("package_total_tax_policy_decisions")] + spend = total[K.package_key("package_total_spending_policy_decisions")] + grand = total[K.package_key("package_total_policy_decisions")] + assert abs(by_head["Tax"] - tax) <= 98 * 2.5e6, fy + assert abs(by_head["Spend"] - spend) <= 98 * 2.5e6, fy + assert abs(tax + spend - grand) <= 3 * 2.5e6, fy + + +def test_table_64_recostings_name_their_event(): + t64 = _hmt("6.4") + assert len(t64) == 17 * 6 + for r in t64: + c = r["conditions"] + event, key = K.TABLE_64[int(c["table_line"])] + assert c["recosting_of"] == event + assert r["measure_key"] == key + assert c["row_type"] == "recosting" + + +def test_table_62_rows_are_financial_transactions_not_revenue(): + t62 = _hmt("6.2") + assert len(t62) == 9 * 6 + for r in t62: + assert r["proposed_metric"] == "financial_transaction_cash_impact" + assert "metric" not in r + assert ( + r["measure_key"] + == K.PREFIX + K.TABLE_62[int(r["conditions"]["table_line"])] + ) From 9a0189cf397293ffd7078dec7db45c098e036a4a Mon Sep 17 00:00:00 2001 From: Max Ghenis Date: Thu, 29 Oct 2026 10:52:54 -0400 Subject: [PATCH 02/52] Check the AB2026 harvest contract with one function a lane can run on its own family pipeline/ab2026_harvest_contract.py holds the row contract the tests applied inline, so a family being staged can be checked before COUNTS.json pins it. Only HM Treasury and OBR quotes run whole (README rule 6). Co-Authored-By: Claude Opus 5.5 --- pipeline/ab2026_harvest_contract.py | 242 ++++++++++++++++++++++++++++ tests/test_ab2026_registry.py | 1 + tests/test_uk_ab2026_harvest.py | 185 ++------------------- 3 files changed, 259 insertions(+), 169 deletions(-) create mode 100644 pipeline/ab2026_harvest_contract.py diff --git a/pipeline/ab2026_harvest_contract.py b/pipeline/ab2026_harvest_contract.py new file mode 100644 index 000000000..e3e37ae0b --- /dev/null +++ b/pipeline/ab2026_harvest_contract.py @@ -0,0 +1,242 @@ +"""The AB2026 harvest row contract (#230), as one checkable function. + +sources/harvest-uk-ab2026-2026-10-29/README.md states the contract: the AB2025 +row contract with the AB2026 exceptions. tests/test_uk_ab2026_harvest.py +holds every pinned family to it, and a family being staged can be checked on +its own before COUNTS.json pins it: + + python pipeline/ab2026_harvest_contract.py sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026 +""" + +import gzip +import json +import re +import sys +from pathlib import Path + +ROOT = Path(__file__).resolve().parent.parent +UK = ROOT / "data" / "uk" +FISCAL_EVENT = "autumn_budget_2026" + +REQUIRED = { + "source", + "source_model", + "value", + "value_raw", + "normalization", + "value_kind", + "period", + "time_basis", + "conditions", + "reform_hint", + "measure_key", + "attribution", + "benchmark_class", + "baseline_policy", + "publication", + "source_table", + "source_column", + "quote", + "parse_confidence", + "status", +} +OPTIONAL = { + "metric", + "proposed_metric", + "unit_concept", + "proposed_unit", + "proposed_baseline", + "note", + "local_artifact", +} +ATTRIBUTION = {"own", "restated", "same_assumptions"} +BENCHMARK_CLASS = {"different_model", "administrative_fact", "same_assumptions"} +VALUE_KIND = { + "point", + "central", + "range_low", + "range_high", + "approx_chart_reading", + "cumulative", + "categorical", +} +TIME_BASIS = {"fiscal_year", "annual", "point_in_time"} +CONFIDENCE = {"high", "medium", "low"} +# README rule 6: HM Treasury and OBR material is under the Open Government +# Licence and quoted whole; every other producer's quote stays under 15 words. +OGL = {"hm_treasury", "obr_efo"} +PROPOSED_KEY = re.compile(r"ab2026(_option)?__[a-z0-9_]+") +SHA256 = re.compile(r"[0-9a-f]{64}") + + +def _keys(path): + return {m["measure_key"] for m in json.loads(path.read_text())["measures"]} + + +def registry_keys(): + """The AB2026 keys, and each earlier event's keys for Table 6.4 + recostings (README rule 2). AB2025 has two key spaces: the #136 port + registry and the events inventory.""" + earlier = { + p.name.removesuffix("_measures.json"): _keys(p) + for p in (UK / "events").glob("*_measures.json") + } + earlier["autumn_budget_2025"] |= _keys(UK / "ab2025_measures.json") + return _keys(UK / "ab2026_measures.json"), earlier + + +def enums(): + from scorecard_db.models import Metric, UnitConcept + + return {m.value for m in Metric}, {u.value for u in UnitConcept} + + +def read_rows(family_dir): + with gzip.open(Path(family_dir) / "claims_staged.jsonl.gz", "rt") as fh: + return [json.loads(line) for line in fh if line.strip()] + + +def read_manifest(family_dir): + text = (Path(family_dir) / "manifest.jsonl").read_text() + return [json.loads(line) for line in text.splitlines() if line.strip()] + + +def row_problems(r, keys, earlier, metrics, units): + """Every way one staged row breaks the contract, as strings.""" + out = [] + fields = set(r) + if not REQUIRED <= fields: + return [f"missing fields {sorted(REQUIRED - fields)}"] + if not fields <= REQUIRED | OPTIONAL: + out.append(f"unknown fields {sorted(fields - REQUIRED - OPTIONAL)}") + for field, allowed in ( + ("attribution", ATTRIBUTION), + ("benchmark_class", BENCHMARK_CLASS), + ("value_kind", VALUE_KIND), + ("time_basis", TIME_BASIS), + ("parse_confidence", CONFIDENCE), + ): + if r[field] not in allowed: + out.append(f"{field} {r[field]!r} not in {sorted(allowed)}") + if r["status"] != "ok": + out.append("status is not ok") + if not isinstance(r["value"], (int, float)) or isinstance(r["value"], bool): + out.append("value is not a number") + for field in ("value_raw", "quote"): + if not (isinstance(r[field], str) and r[field].strip()): + out.append(f"{field} is empty") + if bool(r.get("metric")) == bool(r.get("proposed_metric")): + out.append("exactly one of metric and proposed_metric") + if bool(r.get("unit_concept")) == bool(r.get("proposed_unit")): + out.append("exactly one of unit_concept and proposed_unit") + if r.get("metric") and r["metric"] not in metrics: + out.append(f"metric {r['metric']!r} is not a Metric; use proposed_metric") + if r.get("unit_concept") and r["unit_concept"] not in units: + out.append(f"unit_concept {r['unit_concept']!r} is not a UnitConcept") + + c = r["conditions"] + if not (isinstance(c, dict) and c): + return out + ["conditions is not a non-empty object"] + for k, v in c.items(): + if not (isinstance(k, str) and isinstance(v, str)): + out.append(f"condition {k!r} is not a string") + if c.get("fiscal_event") != FISCAL_EVENT: + out.append(f"conditions.fiscal_event is not {FISCAL_EVENT}") + if "fy" in c: + m = re.fullmatch(r"(\d{4})-(\d{2})", str(c["fy"])) + if not m: + out.append(f"conditions.fy {c['fy']!r} is not YYYY-YY") + else: + if r["period"] != int(m.group(1)) + 1: + out.append(f"period {r['period']} is not the end year of {c['fy']}") + if r["time_basis"] != "fiscal_year": + out.append("an fy row has time_basis fiscal_year") + if not (isinstance(r["period"], int) and 1945 <= r["period"] <= 2075): + out.append(f"period {r['period']!r} is not a year") + + k = r["measure_key"] + if "recosting_of" in c: + if k not in earlier.get(c["recosting_of"], ()): + out.append(f"{k} is not a {c['recosting_of']} key") + elif k is not None: + if k not in keys: + out.append(f"measure_key {k} is not in data/uk/ab2026_measures.json") + if "proposed_measure_key" in c: + out.append("a row with a measure_key carries no proposed_measure_key") + elif c.get("proposed_measure_key"): + pk = c["proposed_measure_key"] + if not PROPOSED_KEY.fullmatch(pk): + out.append(f"proposed_measure_key {pk!r} is not ab2026[_option]__slug") + if pk in keys: + out.append(f"{pk} is registered; put it in measure_key") + + pub = r["publication"] + for field in ("title", "url", "date", "publisher", "primary_sha256"): + if not pub.get(field): + out.append(f"publication.{field} is empty") + if pub.get("primary_sha256") and not SHA256.fullmatch(pub["primary_sha256"]): + out.append("publication.primary_sha256 is not a sha256") + if pub.get("date") and not re.fullmatch(r"\d{4}-\d{2}-\d{2}", pub["date"]): + out.append("publication.date is not YYYY-MM-DD") + + if r["source"] not in OGL and len(r["quote"].split()) >= 15: + out.append("quote has 15 or more words (README rule 6)") + if r["attribution"] == "same_assumptions" and ( + r["benchmark_class"] != "same_assumptions" + or r["source_model"] != "policyengine_uk" + ): + out.append("same_assumptions rows are PolicyEngine-computed") + if ( + r["source"] == "hm_treasury" + and "Impact on households" in pub.get("title", "") + and r["value_kind"] == "approx_chart_reading" + ): + out.append("no HMT Impact on households bar readings (README rule 4)") + return out + + +def family_problems(family_dir): + """Every contract breach in one family directory, as strings.""" + family_dir = Path(family_dir) + missing = [ + n + for n in ("NOTES.md", "manifest.jsonl", "claims_staged.jsonl.gz") + if not (family_dir / n).exists() + ] + if missing: + return [f"{family_dir.name}: missing {missing}"] + out = [] + manifest = read_manifest(family_dir) + shas = [m.get("sha256", "") for m in manifest] + for m in manifest: + if not SHA256.fullmatch(m.get("sha256", "")): + out.append(f"manifest {m.get('url')}: sha256 is not a sha256") + if not (m.get("url") and m.get("title") and m.get("retrieved")): + out.append(f"manifest {m.get('url')}: needs url, title and retrieved") + if len(set(shas)) != len(shas): + out.append("manifest lists a primary twice") + keys, earlier = registry_keys() + metrics, units = enums() + for i, r in enumerate(read_rows(family_dir)): + for p in row_problems(r, keys, earlier, metrics, units): + out.append(f"row {i}: {p}") + sha = (r.get("publication") or {}).get("primary_sha256") + if sha not in set(shas): + out.append(f"row {i}: primary_sha256 is not in manifest.jsonl") + return out + + +def main(argv): + bad = 0 + for d in argv: + problems = family_problems(d) + rows = len(read_rows(d)) if (Path(d) / "claims_staged.jsonl.gz").exists() else 0 + print(f"{d}: {rows} rows, {len(problems)} problems") + for p in problems[:200]: + print(" " + p) + bad += bool(problems) + return 1 if bad else 0 + + +if __name__ == "__main__": + sys.exit(main(sys.argv[1:])) diff --git a/tests/test_ab2026_registry.py b/tests/test_ab2026_registry.py index 957d848f2..57d3eb128 100644 --- a/tests/test_ab2026_registry.py +++ b/tests/test_ab2026_registry.py @@ -217,6 +217,7 @@ def test_the_registry_is_regenerated_byte_for_byte(): env={**os.environ, "PYTHONPATH": str(ROOT)}, capture_output=True, text=True, + check=False, ) assert out.returncode == 0, out.stdout + out.stderr diff --git a/tests/test_uk_ab2026_harvest.py b/tests/test_uk_ab2026_harvest.py index 6cf15c782..a3ff6cb17 100644 --- a/tests/test_uk_ab2026_harvest.py +++ b/tests/test_uk_ab2026_harvest.py @@ -9,7 +9,6 @@ manifests, and per-family pins that the ingest's exact accounting will read. """ -import gzip import json import re from pathlib import Path @@ -17,82 +16,21 @@ import pytest from pipeline import ab2026_measure_keys as K +from pipeline.ab2026_harvest_contract import ( + family_problems, + read_manifest, + read_rows, + registry_keys, +) from pipeline.harvest_counts import counts ROOT = Path(__file__).resolve().parent.parent HARVEST = ROOT / "sources" / "harvest-uk-ab2026-2026-10-29" -UK = ROOT / "data" / "uk" - - -def _keys(path): - return {m["measure_key"] for m in json.loads(path.read_text())["measures"]} - - -KEYS = _keys(UK / "ab2026_measures.json") -# A Table 6.4 recosting carries the earlier event's key (README rule 2). -# AB2025 has two key spaces: the #136 port registry and the events inventory. -EARLIER = { - p.name.removesuffix("_measures.json"): _keys(p) - for p in (UK / "events").glob("*_measures.json") -} -EARLIER["autumn_budget_2025"] |= _keys(UK / "ab2025_measures.json") +KEYS, _ = registry_keys() COUNTS = json.loads((HARVEST / "COUNTS.json").read_text()) FAMILIES = sorted(k for k in COUNTS if not k.startswith("_")) -REQUIRED = { - "source", - "source_model", - "value", - "value_raw", - "normalization", - "value_kind", - "period", - "time_basis", - "conditions", - "reform_hint", - "measure_key", - "attribution", - "benchmark_class", - "baseline_policy", - "publication", - "source_table", - "source_column", - "quote", - "parse_confidence", - "status", -} -OPTIONAL = { - "metric", - "proposed_metric", - "unit_concept", - "proposed_unit", - "proposed_baseline", - "note", - "local_artifact", -} -ATTRIBUTION = {"own", "restated", "same_assumptions"} -BENCHMARK_CLASS = {"different_model", "administrative_fact", "same_assumptions"} -VALUE_KIND = { - "point", - "central", - "range_low", - "range_high", - "approx_chart_reading", - "cumulative", - "categorical", -} -TIME_BASIS = {"fiscal_year", "annual", "point_in_time"} -CONFIDENCE = {"high", "medium", "low"} -# README rule 6: quotes from producers outside the Open Government Licence -# stay under 15 words. -OGL = {"hm_treasury", "obr_efo", "hmrc", "dwp", "ons", "scottish_fiscal_commission"} - -from scorecard_db.models import Metric, UnitConcept # noqa: E402 - -METRICS = {m.value for m in Metric} -UNITS = {u.value for u in UnitConcept} - def families(): return sorted( @@ -103,16 +41,11 @@ def families(): def rows(family): - with gzip.open(HARVEST / family / "claims_staged.jsonl.gz", "rt") as fh: - return [json.loads(line) for line in fh if line.strip()] + return read_rows(HARVEST / family) def manifest(family): - return [ - json.loads(line) - for line in (HARVEST / family / "manifest.jsonl").read_text().splitlines() - if line.strip() - ] + return read_manifest(HARVEST / family) def test_every_pinned_family_is_present_and_complete(): @@ -131,103 +64,17 @@ def test_counts_json_is_what_harvest_counts_computes(): @pytest.mark.parametrize("family", FAMILIES) def test_rows_follow_the_contract(family): - for i, r in enumerate(rows(family)): - where = (family, i) - keys = set(r) - assert REQUIRED <= keys, (where, REQUIRED - keys) - assert keys <= REQUIRED | OPTIONAL, (where, keys - REQUIRED - OPTIONAL) - assert r["status"] == "ok", where - assert r["attribution"] in ATTRIBUTION, where - assert r["benchmark_class"] in BENCHMARK_CLASS, where - assert r["value_kind"] in VALUE_KIND, where - assert r["time_basis"] in TIME_BASIS, where - assert r["parse_confidence"] in CONFIDENCE, where - assert isinstance(r["value"], (int, float)) and not isinstance( - r["value"], bool - ), where - assert isinstance(r["value_raw"], str) and r["value_raw"].strip(), where - assert isinstance(r["quote"], str) and r["quote"].strip(), where - assert bool(r.get("metric")) != bool(r.get("proposed_metric")), where - assert bool(r.get("unit_concept")) != bool(r.get("proposed_unit")), where - if r.get("metric"): - assert r["metric"] in METRICS, (where, r["metric"]) - if r.get("unit_concept"): - assert r["unit_concept"] in UNITS, (where, r["unit_concept"]) - c = r["conditions"] - assert isinstance(c, dict) and c, where - for k, v in c.items(): - assert isinstance(k, str) and isinstance(v, str), (where, k, v) - assert c["fiscal_event"] == "autumn_budget_2026", where - if "fy" in c: - m = re.fullmatch(r"(\d{4})-(\d{2})", c["fy"]) - assert m, (where, c["fy"]) - assert r["period"] == int(m.group(1)) + 1, (where, c["fy"], r["period"]) - assert r["time_basis"] == "fiscal_year", where - assert isinstance(r["period"], int) and 1945 <= r["period"] <= 2075, where - pub = r["publication"] - for k in ("title", "url", "date", "publisher", "primary_sha256"): - assert pub.get(k), (where, k) - assert re.fullmatch(r"[0-9a-f]{64}", pub["primary_sha256"]), where - assert re.fullmatch(r"\d{4}-\d{2}-\d{2}", pub["date"]), where - - -@pytest.mark.parametrize("family", FAMILIES) -def test_measure_keys_exist_where_the_row_says_they_live(family): - """README rule 2: an AB2026 key, an earlier event's key on a recosting, - or null with a proposed key and nothing else.""" - for i, r in enumerate(rows(family)): - where, c, k = (family, i), r["conditions"], r["measure_key"] - if "recosting_of" in c: - assert k in EARLIER[c["recosting_of"]], (where, c["recosting_of"], k) - elif k is not None: - assert k in KEYS, (where, k) - assert "proposed_measure_key" not in c, where - else: - pk = c.get("proposed_measure_key") - if pk: - assert re.fullmatch(r"ab2026(_option)?__[a-z0-9_]+", pk), (where, pk) - assert pk not in KEYS, (where, "use measure_key", pk) - - -@pytest.mark.parametrize("family", FAMILIES) -def test_every_row_cites_a_manifest_primary(family): - shas = {m["sha256"] for m in manifest(family)} - for m in manifest(family): - assert re.fullmatch(r"[0-9a-f]{64}", m["sha256"]), (family, m.get("url")) - assert m.get("url") and m.get("title") and m.get("retrieved"), (family, m) - assert len(shas) == len(manifest(family)), family - for i, r in enumerate(rows(family)): - assert r["publication"]["primary_sha256"] in shas, (family, i) - - -@pytest.mark.parametrize("family", FAMILIES) -def test_quotes_outside_the_ogl_stay_short(family): - for i, r in enumerate(rows(family)): - if r["source"] not in OGL: - assert len(r["quote"].split()) < 15, (family, i, r["quote"]) - - -def test_same_assumptions_rows_are_policyengine_computed(): - for fam in families(): - for r in rows(fam): - if r["attribution"] == "same_assumptions": - assert r["benchmark_class"] == "same_assumptions", fam - assert r["source_model"] == "policyengine_uk", fam - - -def test_no_hmt_decile_bar_readings_were_staged(): - """README rule 4 (AB2025 rule 9, #61).""" - for fam in families(): - for r in rows(fam): - if r["source"] == "hm_treasury": - assert "Impact on households" not in r["publication"]["title"] or ( - r["value_kind"] != "approx_chart_reading" - ), (fam, r["quote"][:80]) + """pipeline/ab2026_harvest_contract.py: field names, closed enums, the + FY-end-year period, the fiscal_event slug, keys where the row says they + live (README rule 2), a manifest primary for every row, short quotes + outside the OGL (rule 6), PolicyEngine-computed same_assumptions rows and + no Impact on households bar readings (rule 4).""" + assert family_problems(HARVEST / family) == [] def test_the_key_listing_is_derived_from_the_registry(): text = (HARVEST / "REGISTRY_KEYS.md").read_text() - listed = set(re.findall(r"^\| `([a-z0-9_]+)` \|", text, flags=re.M)) + listed = set(re.findall(r"^\| `([a-z0-9_]+)` \|", text, flags=re.MULTILINE)) assert listed == KEYS, listed ^ KEYS From d37c86f42d7d8aaeb0d1698fc22577f66695455f Mon Sep 17 00:00:00 2001 From: Max Ghenis Date: Thu, 29 Oct 2026 13:03:32 -0400 Subject: [PATCH 03/52] State which harvest numbers are estimates and how a measure's parts are keyed Rule 8 keeps a measure's design (rates, thresholds, payment amounts, dates, durations, allocations, financing remits, contingent liabilities) out of the rows. Rule 9 keys a printed part of one Table 6.1 measure to that measure, marked as a component that is never summed with the total. Co-Authored-By: Claude Opus 5.5 --- sources/harvest-uk-ab2026-2026-10-29/README.md | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) diff --git a/sources/harvest-uk-ab2026-2026-10-29/README.md b/sources/harvest-uk-ab2026-2026-10-29/README.md index 199fcdb4a..4207d213a 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/README.md +++ b/sources/harvest-uk-ab2026-2026-10-29/README.md @@ -15,7 +15,7 @@ Nothing reads this harvest yet. Tranche 2 of #230 adds the ingest. That ingest's ## Rules -The AB2025 harvest README's rules 1–10 and its row contract apply unchanged, except for the following: +The AB2025 harvest README's rules 1–10 and its row contract apply, with the AB2026 exceptions and additions below: 1. **`fiscal_event` is the slug `autumn_budget_2026`** on every row, whatever the publication calls it. 2. **Measure keys come from `data/uk/ab2026_measures.json`.** @@ -28,6 +28,17 @@ The AB2025 harvest README's rules 1–10 and its row contract apply unchanged, e 5. **PolicyEngine's own published figures are not harvested here.** They are tranche 3's counterparts, computed from the dashboard's receipts. A producer that restates a PolicyEngine figure is staged with `attribution: same_assumptions`, as the Fabian Society's row was for AB2025. 6. **Quotes from producers outside the Open Government Licence stay short.** `quote` is the sentence or cell carrying the number, cut to the words that carry it: under 15 words for think tanks, firms and the press. HM Treasury and OBR material is OGL, and its cells and sentences are quoted whole. 7. **A primary that changes is a second primary.** If a producer replaces a document, both versions are hashed in `manifest.jsonl` and rows cite the version they were read from. IFS replaced its cost-of-living deck on 29 October, and HM Treasury replaced the Budget 2026 document PDF after 28 October. NOTES says what changed. +8. **Stage estimates, not design.** A row is a number someone estimated, forecast or counted about the effect of a measure or of the Budget. That covers: + - a cost or yield; + - people or households affected; + - a gain, loss or bill change; + - a distributional or poverty effect; + - a behavioural adjustment or tax base; + - a fiscal aggregate; + - a consequential for a devolved government. + + A measure's own parameters are its design: rates, thresholds, payment amounts, dates and durations. So are spending allocations to programmes, financing remits and contingent liabilities. They go in `reform_hint` or `conditions`, never as rows (the AB2025 red book family kept "£9.90" in `reform_hint` the same way). A worked example's inputs follow this rule too, and its computed outcome is a row. +9. **A component of one measure keeps that measure's key.** A printed part of one Table 6.1 measure carries that measure's key, with `conditions.component` naming the part and `conditions.row_type: measure_component`. A part can be a fiscal head, a sub-measure within a line (the tax-gap package's items, for example) or a nation. Components are never summed with the measure's total. Proposed package keys are for scopes that cross Table 6.1 lines, or have none. ## Producer slugs and families From 7a76bb35fd0e4b735404836775ee3a0367e26491 Mon Sep 17 00:00:00 2001 From: Max Ghenis Date: Thu, 29 Oct 2026 13:06:26 -0400 Subject: [PATCH 04/52] Test the harvest contract on broken rows, and list every proposed key for tranche 2 The contract now also checks that a quote contains the value as printed, that a measure_component row names its part and its measure (rule 9), and that a family's NOTES.md gives the scope of every proposed key its rows name. tests/test_ab2026_harvest_contract.py breaks each rule on an otherwise valid row, and Hypothesis checks the FY-end-year and 15-word quote rules for every input. pipeline/ab2026_proposed_keys.py writes PROPOSED_KEYS.md, the one list tranche 2 registers or folds. Co-Authored-By: Claude Opus 5.5 --- pipeline/ab2026_harvest_contract.py | 23 ++- pipeline/ab2026_proposed_keys.py | 87 +++++++++ .../PROPOSED_KEYS.md | 7 + .../harvest-uk-ab2026-2026-10-29/README.md | 1 + tests/test_ab2026_harvest_contract.py | 180 ++++++++++++++++++ tests/test_uk_ab2026_harvest.py | 14 +- 6 files changed, 309 insertions(+), 3 deletions(-) create mode 100644 pipeline/ab2026_proposed_keys.py create mode 100644 sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md create mode 100644 tests/test_ab2026_harvest_contract.py diff --git a/pipeline/ab2026_harvest_contract.py b/pipeline/ab2026_harvest_contract.py index e3e37ae0b..bbcf7d573 100644 --- a/pipeline/ab2026_harvest_contract.py +++ b/pipeline/ab2026_harvest_contract.py @@ -101,6 +101,16 @@ def read_manifest(family_dir): return [json.loads(line) for line in text.splitlines() if line.strip()] +def proposed_keys(rows): + """The proposed measure keys a family's rows name (README rule 2).""" + return { + r["conditions"]["proposed_measure_key"] + for r in rows + if isinstance(r.get("conditions"), dict) + and r["conditions"].get("proposed_measure_key") + } + + def row_problems(r, keys, earlier, metrics, units): """Every way one staged row breaks the contract, as strings.""" out = [] @@ -169,6 +179,10 @@ def row_problems(r, keys, earlier, metrics, units): out.append(f"proposed_measure_key {pk!r} is not ab2026[_option]__slug") if pk in keys: out.append(f"{pk} is registered; put it in measure_key") + if c.get("row_type") == "measure_component" and not ( + c.get("component") and (k or c.get("proposed_measure_key")) + ): + out.append("a measure_component row names its component and measure (rule 9)") pub = r["publication"] for field in ("title", "url", "date", "publisher", "primary_sha256"): @@ -179,6 +193,8 @@ def row_problems(r, keys, earlier, metrics, units): if pub.get("date") and not re.fullmatch(r"\d{4}-\d{2}-\d{2}", pub["date"]): out.append("publication.date is not YYYY-MM-DD") + if r["value_raw"] not in r["quote"]: + out.append("quote does not contain value_raw") if r["source"] not in OGL and len(r["quote"].split()) >= 15: out.append("quote has 15 or more words (README rule 6)") if r["attribution"] == "same_assumptions" and ( @@ -217,12 +233,17 @@ def family_problems(family_dir): out.append("manifest lists a primary twice") keys, earlier = registry_keys() metrics, units = enums() - for i, r in enumerate(read_rows(family_dir)): + rows = read_rows(family_dir) + for i, r in enumerate(rows): for p in row_problems(r, keys, earlier, metrics, units): out.append(f"row {i}: {p}") sha = (r.get("publication") or {}).get("primary_sha256") if sha not in set(shas): out.append(f"row {i}: primary_sha256 is not in manifest.jsonl") + notes = (family_dir / "NOTES.md").read_text() + for pk in sorted(proposed_keys(rows)): + if pk not in notes: + out.append(f"NOTES.md does not give the scope of {pk}") return out diff --git a/pipeline/ab2026_proposed_keys.py b/pipeline/ab2026_proposed_keys.py new file mode 100644 index 000000000..fe599df3a --- /dev/null +++ b/pipeline/ab2026_proposed_keys.py @@ -0,0 +1,87 @@ +"""Write or check the AB2026 harvest's PROPOSED_KEYS.md. + +A row whose scope matches no registered measure has `measure_key: null` and +names a proposed key in `conditions.proposed_measure_key` (README rule 2). +Tranche 2 of #230 registers each proposed key, or folds it into a registered +one, before the ingest runs. This lists every proposed key once, with the +families and producers that use it and its row count, so that work starts +from a complete list. Each family's NOTES.md gives the scope of every key its +rows use; the contract checks that. + + python pipeline/ab2026_proposed_keys.py sources/harvest-uk-ab2026-2026-10-29 + python pipeline/ab2026_proposed_keys.py sources/harvest-uk-ab2026-2026-10-29 --check +""" + +import argparse +import sys +from collections import defaultdict +from pathlib import Path + +from pipeline.ab2026_harvest_contract import read_rows + +HEADER = """\ +# Proposed measure keys in the AB2026 harvest + +Generated by `pipeline/ab2026_proposed_keys.py` from the staged families; don't edit by hand. + +A row whose scope matches no registered measure has `measure_key: null` and names one of these keys in `conditions.proposed_measure_key` (README rule 2). Each family's NOTES.md gives the key's scope as Table 6.1 lines or the producer's own words. Tranche 2 registers each key in `data/uk/ab2026_measures.json`, or folds it into a registered key, before the ingest. +""" + + +def inventory(harvest): + """{proposed key: {"rows": n, "families": set, "sources": set}}.""" + out = defaultdict(lambda: {"rows": 0, "families": set(), "sources": set()}) + for fam in sorted(Path(harvest).iterdir()): + if not (fam / "claims_staged.jsonl.gz").exists(): + continue + for r in read_rows(fam): + pk = r["conditions"].get("proposed_measure_key") + if pk: + out[pk]["rows"] += 1 + out[pk]["families"].add(fam.name) + out[pk]["sources"].add(r["source"]) + return dict(out) + + +def render(harvest): + inv = inventory(harvest) + packages = sum(1 for k in inv if k.startswith("ab2026__")) + rows = sum(v["rows"] for v in inv.values()) + options = len(inv) - packages + tally = f"{len(inv)} proposed keys ({packages} packages, {options} options)" + lines = [HEADER, f"{tally} on {rows} rows.", ""] + if inv: + lines += [ + "| proposed key | rows | families | producers |", + "|---|--:|---|---|", + ] + for k in sorted(inv): + v = inv[k] + lines.append( + f"| `{k}` | {v['rows']} | {', '.join(sorted(v['families']))} " + f"| {', '.join(sorted(v['sources']))} |" + ) + lines.append("") + return "\n".join(lines) + + +def main(argv=None): + ap = argparse.ArgumentParser(description=__doc__.splitlines()[0]) + ap.add_argument("harvest") + ap.add_argument("--check", action="store_true") + args = ap.parse_args(argv) + path = Path(args.harvest) / "PROPOSED_KEYS.md" + text = render(args.harvest) + if args.check: + if not path.exists() or path.read_text() != text: + print("PROPOSED_KEYS.md is stale") + return 1 + print("PROPOSED_KEYS.md matches the families") + return 0 + path.write_text(text) + print(f"wrote {path}") + return 0 + + +if __name__ == "__main__": + sys.exit(main(sys.argv[1:])) diff --git a/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md b/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md new file mode 100644 index 000000000..d7d60610d --- /dev/null +++ b/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md @@ -0,0 +1,7 @@ +# Proposed measure keys in the AB2026 harvest + +Generated by `pipeline/ab2026_proposed_keys.py` from the staged families; don't edit by hand. + +A row whose scope matches no registered measure has `measure_key: null` and names one of these keys in `conditions.proposed_measure_key` (README rule 2). Each family's NOTES.md gives the key's scope as Table 6.1 lines or the producer's own words. Tranche 2 registers each key in `data/uk/ab2026_measures.json`, or folds it into a registered key, before the ingest. + +0 proposed keys (0 packages, 0 options) on 0 rows. diff --git a/sources/harvest-uk-ab2026-2026-10-29/README.md b/sources/harvest-uk-ab2026-2026-10-29/README.md index 4207d213a..063d94fad 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/README.md +++ b/sources/harvest-uk-ab2026-2026-10-29/README.md @@ -9,6 +9,7 @@ Staged from 2026-10-29, the day after the Budget, from the primary documents. Th - `claims_staged.jsonl.gz`: one line per claim. - `COUNTS.json` pins each family's rows, primaries, producer slugs and the sha256 of its claims file. `pipeline/harvest_counts.py` writes and checks it. - `REGISTRY_KEYS.md` lists every measure key. `pipeline/build_ab2026_registry.py` generates it. +- `PROPOSED_KEYS.md` lists every proposed key a row names (rule 2), for tranche 2 to register. `pipeline/ab2026_proposed_keys.py` generates it. - `SEED_INVENTORY/` lists the publications the harvest starts from, compiled while monitoring reactions on 28 and 29 October. Nothing reads this harvest yet. Tranche 2 of #230 adds the ingest. That ingest's exact accounting will read `read = ingested + dropped` against the rows pinned in `COUNTS.json`. `tests/test_uk_ab2026_harvest.py` checks every row against the contract below. diff --git a/tests/test_ab2026_harvest_contract.py b/tests/test_ab2026_harvest_contract.py new file mode 100644 index 000000000..304b1858d --- /dev/null +++ b/tests/test_ab2026_harvest_contract.py @@ -0,0 +1,180 @@ +"""The AB2026 harvest contract checker (#230), tested on rows built to break it. + +tests/test_uk_ab2026_harvest.py holds every staged family to +pipeline/ab2026_harvest_contract.py. These tests hold the checker itself to +the README: each rule, broken on an otherwise valid row, gives a problem, and +the period and quote-length rules hold for every input. +""" + +import copy +import gzip +import io +import json +import shutil +from pathlib import Path + +import pytest +from hypothesis import given, settings +from hypothesis import strategies as st + +from pipeline.ab2026_harvest_contract import ( + OGL, + enums, + family_problems, + read_rows, + registry_keys, + row_problems, +) + +ROOT = Path(__file__).resolve().parent.parent +FAMILY = ROOT / "sources" / "harvest-uk-ab2026-2026-10-29" / "uk_hmt_tables" +KEYS, EARLIER = registry_keys() +METRICS, UNITS = enums() +BASE = next( + r + for r in read_rows(FAMILY) + if r["measure_key"] and "recosting_of" not in r["conditions"] +) +REGISTERED = BASE["measure_key"] + + +def problems(row): + return row_problems(row, KEYS, EARLIER, METRICS, UNITS) + + +def mutated(**changes): + """BASE with dotted-path changes; a value of ... deletes the field.""" + row = copy.deepcopy(BASE) + for path, value in changes.items(): + *parents, leaf = path.split("__") + target = row + for p in parents: + target = target[p] + if value is ...: + del target[leaf] + else: + target[leaf] = value + return row + + +def test_a_staged_row_has_no_problems(): + assert problems(BASE) == [] + + +@pytest.mark.parametrize( + ("changes", "expected"), + [ + ({"quote": ...}, "missing fields"), + ({"invented": "x"}, "unknown fields"), + ({"attribution": "guess"}, "attribution"), + ({"status": "draft"}, "status is not ok"), + ({"value": "1"}, "value is not a number"), + ({"value": True}, "value is not a number"), + ({"proposed_metric": "x"}, "exactly one of metric"), + ({"metric": "not_a_metric"}, "is not a Metric"), + ({"conditions__fiscal_event": "autumn_budget_2025"}, "fiscal_event"), + ({"conditions__geography": 1}, "is not a string"), + ({"conditions__fy": "2026/27"}, "is not YYYY-YY"), + ({"period": 2026}, "is not the end year"), + ({"time_basis": "annual"}, "time_basis fiscal_year"), + ({"measure_key": "ab2026__not_a_measure"}, "is not in data/uk"), + ( + {"conditions__proposed_measure_key": "ab2026__package_x"}, + "carries no proposed_measure_key", + ), + ( + {"measure_key": None, "conditions__proposed_measure_key": "package_x"}, + "is not ab2026", + ), + ( + {"measure_key": None, "conditions__proposed_measure_key": REGISTERED}, + "is registered", + ), + ( + {"conditions__recosting_of": "autumn_budget_2025"}, + "is not a autumn_budget_2025 key", + ), + ({"conditions__row_type": "measure_component"}, "rule 9"), + ( + { + "measure_key": None, + "conditions__row_type": "measure_component", + "conditions__component": "Scotland", + }, + "rule 9", + ), + ({"value_raw": "-2,625"}, "quote does not contain value_raw"), + ({"publication__primary_sha256": "abc"}, "is not a sha256"), + ({"publication__date": "28 October 2026"}, "is not YYYY-MM-DD"), + ({"publication__url": ""}, "publication.url is empty"), + ({"attribution": "same_assumptions"}, "PolicyEngine-computed"), + ], +) +def test_each_rule_broken_on_a_valid_row_is_a_problem(changes, expected): + found = problems(mutated(**changes)) + assert any(expected in p for p in found), found + + +def test_a_component_with_its_measure_or_a_proposed_key_is_valid(): + """README rule 9: a part of one measure keeps that measure's key; a part + of a measure with no Table 6.1 line keeps that measure's proposed key.""" + keyed = mutated( + conditions__row_type="measure_component", conditions__component="Wales" + ) + proposed = mutated( + measure_key=None, + conditions__row_type="measure_component", + conditions__component="Capital AME", + conditions__proposed_measure_key="ab2026__package_fiscally_neutral_x", + ) + assert problems(keyed) == [] + assert problems(proposed) == [] + + +@settings(max_examples=200, deadline=None) +@given(start=st.integers(1945, 2073), offset=st.integers(-3, 3)) +def test_an_fy_row_is_valid_only_at_the_fy_end_year(start, offset): + fy = f"{start}-{(start + 1) % 100:02d}" + row = mutated(conditions__fy=fy, period=start + 1 + offset) + end_year_problem = any("is not the end year" in p for p in problems(row)) + assert end_year_problem == (offset != 0) + + +@settings(max_examples=200, deadline=None) +@given( + words=st.integers(1, 40), + source=st.sampled_from(["ifs", "resolution_foundation", *sorted(OGL)]), +) +def test_only_ogl_quotes_may_run_to_15_words(words, source): + """README rule 6: under 15 words outside the Open Government Licence.""" + row = mutated(source=source, quote=" ".join(["-2625"] * words)) + long_quote = any("15 or more words" in p for p in problems(row)) + assert long_quote == (source not in OGL and words >= 15) + + +def _family_with_a_proposed_key(tmp_path, notes_extra): + fam = tmp_path / "uk_test_family" + shutil.copytree(FAMILY, fam) + rows = read_rows(fam) + i = rows.index(BASE) + rows[i]["measure_key"] = None + rows[i]["conditions"]["proposed_measure_key"] = "ab2026__package_test_scope" + buf = io.BytesIO() + with gzip.GzipFile(fileobj=buf, mode="wb", mtime=0, filename="") as gz: + for r in rows: + gz.write((json.dumps(r, sort_keys=True) + "\n").encode()) + (fam / "claims_staged.jsonl.gz").write_bytes(buf.getvalue()) + with (fam / "NOTES.md").open("a") as fh: + fh.write(notes_extra) + return fam + + +def test_a_family_scopes_every_proposed_key_it_uses_in_its_notes(tmp_path): + unscoped = _family_with_a_proposed_key(tmp_path / "a", "") + scoped = _family_with_a_proposed_key( + tmp_path / "b", "\n- `ab2026__package_test_scope`: Table 6.1 line 1.\n" + ) + assert family_problems(unscoped) == [ + "NOTES.md does not give the scope of ab2026__package_test_scope" + ] + assert family_problems(scoped) == [] diff --git a/tests/test_uk_ab2026_harvest.py b/tests/test_uk_ab2026_harvest.py index a3ff6cb17..44d1cba32 100644 --- a/tests/test_uk_ab2026_harvest.py +++ b/tests/test_uk_ab2026_harvest.py @@ -22,6 +22,7 @@ read_rows, registry_keys, ) +from pipeline.ab2026_proposed_keys import render from pipeline.harvest_counts import counts ROOT = Path(__file__).resolve().parent.parent @@ -67,11 +68,20 @@ def test_rows_follow_the_contract(family): """pipeline/ab2026_harvest_contract.py: field names, closed enums, the FY-end-year period, the fiscal_event slug, keys where the row says they live (README rule 2), a manifest primary for every row, short quotes - outside the OGL (rule 6), PolicyEngine-computed same_assumptions rows and - no Impact on households bar readings (rule 4).""" + outside the OGL (rule 6) that contain the value as printed, components + that name their part and measure (rule 9), a NOTES.md scope for every + proposed key, PolicyEngine-computed same_assumptions rows and no Impact + on households bar readings (rule 4).""" assert family_problems(HARVEST / family) == [] +def test_the_proposed_key_listing_is_derived_from_the_families(): + """PROPOSED_KEYS.md lists every proposed_measure_key a staged row names, + once, with its rows, families and producers: tranche 2 registers or folds + each one, so a key added without the listing moving fails.""" + assert (HARVEST / "PROPOSED_KEYS.md").read_text() == render(HARVEST) + + def test_the_key_listing_is_derived_from_the_registry(): text = (HARVEST / "REGISTRY_KEYS.md").read_text() listed = set(re.findall(r"^\| `([a-z0-9_]+)` \|", text, flags=re.MULTILINE)) From 9dbb52d96f4f4d69203165d48f0aaf35dc36fdd6 Mon Sep 17 00:00:00 2001 From: Max Ghenis Date: Thu, 29 Oct 2026 13:29:42 -0400 Subject: [PATCH 05/52] Correct rule 3's line 2 example and name proposed keys by scope Table 6.1 line 2 costs the whole electricity VAT cut from 1 October 2026, so a package of the energy payment, the VAT cut and the fuel duty cut is exactly lines 1 to 3; the old example said otherwise. Rule 2 now names the three kinds of proposed key (a measure with no Table 6.1 line, a package named by its scope, a producer's option), and PROPOSED_KEYS.md shows each key's kind. Co-Authored-By: Claude Opus 5.5 --- pipeline/ab2026_proposed_keys.py | 27 +++++++++++++------ .../PROPOSED_KEYS.md | 4 +-- .../harvest-uk-ab2026-2026-10-29/README.md | 7 ++++- 3 files changed, 27 insertions(+), 11 deletions(-) diff --git a/pipeline/ab2026_proposed_keys.py b/pipeline/ab2026_proposed_keys.py index fe599df3a..f5deb0d07 100644 --- a/pipeline/ab2026_proposed_keys.py +++ b/pipeline/ab2026_proposed_keys.py @@ -24,10 +24,19 @@ Generated by `pipeline/ab2026_proposed_keys.py` from the staged families; don't edit by hand. -A row whose scope matches no registered measure has `measure_key: null` and names one of these keys in `conditions.proposed_measure_key` (README rule 2). Each family's NOTES.md gives the key's scope as Table 6.1 lines or the producer's own words. Tranche 2 registers each key in `data/uk/ab2026_measures.json`, or folds it into a registered key, before the ingest. +A row whose scope matches no registered measure has `measure_key: null` and names one of these keys in `conditions.proposed_measure_key` (README rule 2). A key is a measure with no Table 6.1 line (`ab2026__`), a package (`ab2026__package_`) or an option (`ab2026_option__`). Each family's NOTES.md gives the key's scope as Table 6.1 lines or the producer's own words. Tranche 2 registers each key in `data/uk/ab2026_measures.json`, or folds it into a registered key, before the ingest. """ +def kind(key): + """README rule 2's three kinds of proposed key.""" + if key.startswith("ab2026_option__"): + return "option" + if key.startswith("ab2026__package_"): + return "package" + return "measure" + + def inventory(harvest): """{proposed key: {"rows": n, "families": set, "sources": set}}.""" out = defaultdict(lambda: {"rows": 0, "families": set(), "sources": set()}) @@ -45,20 +54,22 @@ def inventory(harvest): def render(harvest): inv = inventory(harvest) - packages = sum(1 for k in inv if k.startswith("ab2026__")) + kinds = [kind(k) for k in inv] rows = sum(v["rows"] for v in inv.values()) - options = len(inv) - packages - tally = f"{len(inv)} proposed keys ({packages} packages, {options} options)" - lines = [HEADER, f"{tally} on {rows} rows.", ""] + counted = ", ".join( + f"{kinds.count(c)} {c}s" for c in ("measure", "package", "option") + ) + lines = [HEADER, f"{len(inv)} proposed keys ({counted}) on {rows} rows.", ""] if inv: lines += [ - "| proposed key | rows | families | producers |", - "|---|--:|---|---|", + "| proposed key | kind | rows | families | producers |", + "|---|---|--:|---|---|", ] for k in sorted(inv): v = inv[k] lines.append( - f"| `{k}` | {v['rows']} | {', '.join(sorted(v['families']))} " + f"| `{k}` | {kind(k)} | {v['rows']} " + f"| {', '.join(sorted(v['families']))} " f"| {', '.join(sorted(v['sources']))} |" ) lines.append("") diff --git a/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md b/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md index d7d60610d..2c519bfcb 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md +++ b/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md @@ -2,6 +2,6 @@ Generated by `pipeline/ab2026_proposed_keys.py` from the staged families; don't edit by hand. -A row whose scope matches no registered measure has `measure_key: null` and names one of these keys in `conditions.proposed_measure_key` (README rule 2). Each family's NOTES.md gives the key's scope as Table 6.1 lines or the producer's own words. Tranche 2 registers each key in `data/uk/ab2026_measures.json`, or folds it into a registered key, before the ingest. +A row whose scope matches no registered measure has `measure_key: null` and names one of these keys in `conditions.proposed_measure_key` (README rule 2). A key is a measure with no Table 6.1 line (`ab2026__`), a package (`ab2026__package_`) or an option (`ab2026_option__`). Each family's NOTES.md gives the key's scope as Table 6.1 lines or the producer's own words. Tranche 2 registers each key in `data/uk/ab2026_measures.json`, or folds it into a registered key, before the ingest. -0 proposed keys (0 packages, 0 options) on 0 rows. +0 proposed keys (0 measures, 0 packages, 0 options) on 0 rows. diff --git a/sources/harvest-uk-ab2026-2026-10-29/README.md b/sources/harvest-uk-ab2026-2026-10-29/README.md index 063d94fad..9c456d00b 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/README.md +++ b/sources/harvest-uk-ab2026-2026-10-29/README.md @@ -23,8 +23,13 @@ The AB2025 harvest README's rules 1–10 and its row contract apply, with the AB - `ab2026__…` keys cover announced measures: every HM Treasury Table 6.1 line, Table 6.2 line 4, and the `ab2026__package_…` packages. - `ab2026_option__…` keys cover options a producer costed. None is registered yet. A row on an unregistered option, or on a package whose scope is not one of the registered packages, has `measure_key: null`. Its proposed key goes in `conditions.proposed_measure_key`, and its NOTES says what the scope is. The registry adds it before ingest. - Table 6.4 recostings are the one exception. They carry the earlier event's key, named in `conditions.recosting_of`. + - **Naming a proposed key:** + - An announced measure with no Table 6.1 line is `ab2026__`, naming the measure. That covers a measure below Table 6.1's £10m threshold, a fiscally neutral one and one costed only elsewhere. + - A package is `ab2026__package_`, and the slug names the scope, not the producer, so that producers with the same scope share one key. Lines 61 and 62 together are `ab2026__package_property_and_savings_income_rates`. A producer's or the OBR's name goes in the slug only for a grouping that is its own, such as `ab2026__package_tpa_other_vat`. + - An option a producer costed is `ab2026_option__`. + - The coordinator gives each scope one key across families before merging. `PROPOSED_KEYS.md` lists the result. 3. **A measure's scope is the producer's, not ours.** A producer's figure for "the cost-of-living package" joins `ab2026__package_household_energy_vat_and_fuel_duty` only when its scope is exactly Table 6.1 lines 1, 2 and 3. Otherwise it is a proposal (rule 2). - - For example, RF's package also includes the electricity VAT cut that was announced before the Budget, and its own rows say so. + - Line 2 costs the whole electricity VAT cut, from 1 October 2026 to 1 July 2027 (£755m in 2026-27 and £305m in 2027-28), including the weeks before the Budget. A package of the energy payment, the VAT cut and the fuel duty cut is therefore exactly lines 1, 2 and 3. A package that also counts the gas standing-charge change, which has no Table 6.1 line, is a proposal. 4. **HMT Impact on households, Figures 1.A and 1.B: no bar readings** (AB2025 rule 9, #61). Values HM Treasury prints as numbers, in text, tables or data labels, are staged as printed. 5. **PolicyEngine's own published figures are not harvested here.** They are tranche 3's counterparts, computed from the dashboard's receipts. A producer that restates a PolicyEngine figure is staged with `attribution: same_assumptions`, as the Fabian Society's row was for AB2025. 6. **Quotes from producers outside the Open Government Licence stay short.** `quote` is the sentence or cell carrying the number, cut to the words that carry it: under 15 words for think tanks, firms and the press. HM Treasury and OBR material is OGL, and its cells and sentences are quoted whole. From 2ef78706c6eb848713cd98474deece8bbce64a20 Mon Sep 17 00:00:00 2001 From: Claude Date: Tue, 3 Nov 2026 13:07:38 +0000 Subject: [PATCH 06/52] Land the reworked HM Treasury red book family (432 rows) The 29 October rework cut the first staging's 1,699 rows to 432 under README rules 8 and 9 and stopped before its report. Checking each instruction of its brief against the rows found two loose ends, fixed here: Chart 1.6's six headroom-bridge labels used `component`, which rule 9 reserves for parts of a Table 6.1 measure, so they now name their bar in `chart_bar`; and two manifest notes named local machine paths, which a public repository should not carry. The family passes the contract with 0 problems and its stage.py (kept on uk/ab2026-harvest-staging) reproduces the gzip byte for byte. No primary was re-opened: this session's network policy refused gov.uk, so the 15-row spot check the brief asks for could not run, and NOTES.md says so. COUNTS.json and PROPOSED_KEYS.md are regenerated for the landed family. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_01LwQq6HuQUQ1obM9JvyhJbE --- .../harvest-uk-ab2026-2026-10-29/COUNTS.json | 14 +- .../PROPOSED_KEYS.md | 28 +- .../uk_hmt_redbook/NOTES.md | 307 ++++++++++++++++++ .../uk_hmt_redbook/claims_staged.jsonl.gz | Bin 0 -> 36363 bytes .../uk_hmt_redbook/manifest.jsonl | 5 + 5 files changed, 350 insertions(+), 4 deletions(-) create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/NOTES.md create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/claims_staged.jsonl.gz create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/manifest.jsonl diff --git a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json index e12e3861d..72bdcdffd 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json +++ b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json @@ -1,4 +1,12 @@ { + "uk_hmt_redbook": { + "rows": 432, + "primaries": 5, + "claims_sha256": "3dde1f25fa36df598b54340324e49787e5483bf12962371a25adc0319955a476", + "sources": [ + "hm_treasury" + ] + }, "uk_hmt_tables": { "rows": 762, "primaries": 2, @@ -8,9 +16,9 @@ ] }, "_totals": { - "rows": 762, - "restated": 0, - "families": 1, + "rows": 1194, + "restated": 169, + "families": 2, "pinned_on": "2026-10-29" } } diff --git a/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md b/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md index 2c519bfcb..349c3064e 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md +++ b/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md @@ -4,4 +4,30 @@ Generated by `pipeline/ab2026_proposed_keys.py` from the staged families; don't A row whose scope matches no registered measure has `measure_key: null` and names one of these keys in `conditions.proposed_measure_key` (README rule 2). A key is a measure with no Table 6.1 line (`ab2026__`), a package (`ab2026__package_`) or an option (`ab2026_option__`). Each family's NOTES.md gives the key's scope as Table 6.1 lines or the producer's own words. Tranche 2 registers each key in `data/uk/ab2026_measures.json`, or folds it into a registered key, before the ingest. -0 proposed keys (0 measures, 0 packages, 0 options) on 0 rows. +23 proposed keys (8 measures, 15 packages, 0 options) on 67 rows. + +| proposed key | kind | rows | families | producers | +|---|---|--:|---|---| +| `ab2026__defence_investment_plan` | measure | 1 | uk_hmt_redbook | hm_treasury | +| `ab2026__energy_debt_circuit_breaker` | measure | 3 | uk_hmt_redbook | hm_treasury | +| `ab2026__english_devolution_agreements` | measure | 1 | uk_hmt_redbook | hm_treasury | +| `ab2026__minimum_wage_uprating_april_2027` | measure | 2 | uk_hmt_redbook | hm_treasury | +| `ab2026__mod_counter_fraud_strategy` | measure | 1 | uk_hmt_redbook | hm_treasury | +| `ab2026__package_bbb_equity_community_lender_and_venture_debt` | package | 1 | uk_hmt_redbook | hm_treasury | +| `ab2026__package_breathing_space_for_households` | package | 1 | uk_hmt_redbook | hm_treasury | +| `ab2026__package_budget_barnett_consequentials` | package | 4 | uk_hmt_redbook | hm_treasury | +| `ab2026__package_business_rates_rhl_sbrr_and_transitional_support` | package | 2 | uk_hmt_redbook | hm_treasury | +| `ab2026__package_dwp_fraud_and_error_measures` | package | 1 | uk_hmt_redbook | hm_treasury | +| `ab2026__package_hmrc_tax_gap_and_online_marketplace` | package | 7 | uk_hmt_redbook | hm_treasury | +| `ab2026__package_hmt_distributional_budget_2026` | package | 1 | uk_hmt_redbook | hm_treasury | +| `ab2026__package_hmt_distributional_since_ab2024` | package | 1 | uk_hmt_redbook | hm_treasury | +| `ab2026__package_obr_forecast_policy_effects` | package | 24 | uk_hmt_redbook | hm_treasury | +| `ab2026__package_parliament_fuel_duty_decisions` | package | 3 | uk_hmt_redbook | hm_treasury | +| `ab2026__package_parliament_regulatory_burden_reduction` | package | 1 | uk_hmt_redbook | hm_treasury | +| `ab2026__package_parliament_tax_gap_measures` | package | 1 | uk_hmt_redbook | hm_treasury | +| `ab2026__package_parliament_welfare_fraud_and_error_savings` | package | 1 | uk_hmt_redbook | hm_treasury | +| `ab2026__package_property_and_savings_income_rates` | package | 2 | uk_hmt_redbook | hm_treasury | +| `ab2026__package_regulation_tariff_and_sandbox_business_savings` | package | 1 | uk_hmt_redbook | hm_treasury | +| `ab2026__regulatory_administrative_savings` | measure | 5 | uk_hmt_redbook | hm_treasury | +| `ab2026__rolls_royce_gas_turbine_programme` | measure | 1 | uk_hmt_redbook | hm_treasury | +| `ab2026__state_pension_triple_lock_adjustment` | measure | 2 | uk_hmt_redbook | hm_treasury | diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/NOTES.md new file mode 100644 index 000000000..7a8f15bb5 --- /dev/null +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/NOTES.md @@ -0,0 +1,307 @@ +# uk_hmt_redbook: Autumn Budget 2026 red book, Impact on households and Pension Tax Allowance explainer + +**Working files.** `stage.py`, which writes this family's two data files and is the file meant wherever these notes say `stage.py`, lives on branch `uk/ab2026-harvest-staging` at `harvest-staging/ab2026/lanes/hmt-redbook/uk_hmt_redbook/stage.py`; this branch carries only NOTES.md, `manifest.jsonl` and `claims_staged.jsonl.gz`. `.ab2026-primaries/uk_hmt_redbook/` was the lane's local primaries folder and was never committed: the repository carries no primary bytes, so fetch each primary from its `manifest.jsonl` URL and check its sha256. + +Source `hm_treasury`. **432 rows from 3 cited primaries (5 hashed versions), 23 proposed measure keys.** `stage.py` embeds every row and manifest line as data; `python3 stage.py` rewrites `claims_staged.jsonl.gz` and `manifest.jsonl` byte for byte with the standard library alone. No primary bytes are vendored. + +The first staging (29 October, morning) produced 1,699 rows and 250 proposed keys. This rework applies README rules 8 and 9: it keeps estimates of effects, drops design (parameters, allocations, financing, contingent liabilities), and gives each printed part of one Table 6.1 measure that measure's key. The section "Not staged (rules 8 and 9)" accounts for all 1,267 dropped rows. + +## Primaries + +| Primary version | SHA256 | Bytes | Access | Rows | +|---|---|---:|---|---:| +| Budget 2026 (HC 726), 28 October version | `bed70546c026487e3141799968f6df08210c182b0695fed0e8adf7efd6697e03` | 4,657,025 | local copy | 0 | +| Budget 2026 (HC 726), current version | `19b337aced8fe47de1d6ee462dabf75155110d347c38123b6380d90f09e97b93` | 4,679,545 | direct | 363 | +| Impact on households: distributional analysis to accompany Budget 2026 | `36cd5afb4fceca2cecba0d065974ffe8280888e15ed852c129339439e539469c` | 197,606 | direct | 57 | +| Pension Tax Allowance Explainer (content API, full body) | `cc3ec22eb55f4c827c301bc67b06aa7482e4970ef7e1cec644bc382dac0dbb25` | 24,883 | govuk_api | 12 | +| Pension Tax Allowance Explainer (content API, 29 October abbreviated body) | `791d2573bafc67fcfd89567dab55c0239dd26febc9feec44d10f2be4ae21a9e4` | 9,816 | local copy | 0 | + +## Access recipe + +Every fresh fetch used: + +```sh +curl -sSL -A "Mozilla/5.0 (Macintosh; Intel Mac OS X 14_0) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/17.0 Safari/605.1.15" URL -o PATH +shasum -a 256 PATH +``` + +All fresh fetches returned HTTP 200, and the files sit under `.ab2026-primaries/uk_hmt_redbook/`. No page refused access and none went unreached. The manifest labels the two historical copies `local copy`; they are kept for the changed-primary comparison (README rule 7), not as blocked-fetch fallbacks. + +- **Red book:** fresh PDF fetched about 2026-10-29T15:41Z. `19b337ac…` matches the supplied 29 October download. `pdftotext -layout` and `diff -u` against the supplied 28 October version `bed70546…` find one change: paragraph 6.207's heading lost the words "and potential for ULEV SalSac reforms", with resulting line wrapping. **No printed number changed.** Every red-book row cites the current version. Printed page = PDF page minus six. +- **HTML discovery:** the publication content API `https://www.gov.uk/api/content/government/publications/budget-2026-document` (hash `4568690c1ca115ff4a7aa5aa116bcdb39cebee39d8af4490b8dc3ad544b3ec6d`) names the HTML attachment, fetched as a cross-check (hash `33ee5a00b6383104f83790719c0a67e396cfb52e7ea93986b76ab15a19fdaaec`). Neither supplies rows; the PDF is the primary. +- **Impact on households:** fresh PDF fetched 2026-10-29T15:41:44Z, identical to the supplied 29 October copy, read with `pdftotext -layout`. Printed pages equal PDF pages. +- **Pension Tax Allowance explainer:** fresh content API JSON fetched 2026-10-29T15:41:44Z, matching the supplied 28 October response. Rows quote its decoded `details.body`. **The supplied 29 October response differs:** its body is a 353-character introduction without the counts, worked cases or FAQ of the 19,727-character full body. Both carry publication date 28 October. No row cites the abbreviated version. + +## Coverage + +Rows run in document order: red book (printed page, then position), Impact on households, Pension Tax Allowance explainer. + +- **Red book (363 rows):** + - 160 fiscal aggregates restated from the OBR, ONS and IMF: Chapter 1 text, Chart 1.6's six data labels, Tables 1.2 and 1.3; + - 84 cells of Table 6.3, which breaks Table 6.1 line 45 into eleven items, with its subtotal, the line 46 memo and the subtotal including memo; + - 30 costs, yields and savings: Table 6.1 measures where the red book prints the line's own Exchequer impact, printed parts of those lines, and saving estimates with their own scopes; + - 89 effects on people and businesses: counts affected; gains, losses and bill changes; GDP, earnings, price and jobs effects; Barnett consequentials. +- **Impact on households (57 rows):** the distributional and context statements of paragraphs 1.7, 1.9, 1.10, 2.4, 2.5 and 2.10, and the 50 cells of Table 2.C. +- **Pension Tax Allowance explainer (12 rows):** three pensioner counts and nine worked-case outcomes. + +Rows by measure key: + +| measure_key | rows | of which components (rule 9) | +|---|---:|---:| +| `null` with a proposed key | 67 | 18 | +| `null`, context (no proposed key) | 191 | 0 | +| `ab2026__hmrc_tax_gap_package` | 73 | 67 | +| `ab2026__household_energy_payment_100_bands_a_to_d` | 12 | 0 | +| `ab2026__pension_tax_allowance` | 12 | 0 | +| `ab2026__british_industrial_competitiveness_scheme` | 8 | 0 | +| `ab2026__business_rates_rhl_transitional_support` | 6 | 0 | +| `ab2026__business_rates_sbrr_rhl_thresholds_raised` | 6 | 0 | +| `ab2026__online_marketplace_liability_uk_businesses` | 6 | 0 | +| `ab2026__electricity_vat_zero_rate_to_july_2027` | 5 | 0 | +| `ab2026__alcohol_duty_small_producer_relief` | 4 | 0 | +| `ab2026__food_and_heating_oil_tariff_suspensions` | 4 | 4 | +| `ab2026__mileage_allowance_payments_55p` | 4 | 0 | +| `ab2026__uk_india_ceta_and_double_contributions_convention` | 4 | 3 | +| `ab2026__your_first_home_equity_loan` | 4 | 0 | +| `ab2026__defence_investment_plan_capital_reallocation` | 3 | 2 | +| `ab2026__ukraine_winter_air_defence` | 3 | 0 | +| `ab2026__civil_service_admin_budgets_cut_20pc` | 2 | 0 | +| `ab2026__civil_service_ai_fund` | 2 | 0 | +| `ab2026__early_years_books_access` | 2 | 0 | +| `ab2026__ev_charging_public_chargers` | 2 | 0 | +| `ab2026__fuel_duty_5p_cut_extended_and_rpi_rise_deferred` | 2 | 0 | +| `ab2026__misogyny_prevention_programmes` | 2 | 0 | +| `ab2026__badr_lifetime_limit_500000` | 1 | 0 | +| `ab2026__benefits_removed_hospitalised_serious_offenders` | 1 | 0 | +| `ab2026__business_rates_pubs_relief_35pc` | 1 | 0 | +| `ab2026__carbon_price_support_removed` | 1 | 0 | +| `ab2026__carers_allowance_earnings_taper` | 1 | 0 | +| `ab2026__hgv_ved_1_for_12_months` | 1 | 0 | +| `ab2026__national_wealth_fund_capital` | 1 | 0 | +| `ab2026__parental_allowance_fathers_partners_adopters` | 1 | 0 | + +The 191 context rows carry no proposed key (README rule 2): 136 restated fiscal aggregates about no measure, and 55 Impact on households rows (Table 2.C and the context statements of paragraphs 1.10, 2.4, 2.5 and 2.10). The other 24 Table 1.3 rows, the OBR's effect of policy decisions, carry `ab2026__package_obr_forecast_policy_effects`. + +Rows by metric: + +| metric | rows | +|---|---:| +| `revenue_change` | 114 | +| `income_statistic` | 52 | +| `affected_count` | 38 | +| proposed `government_borrowing` | 28 | +| proposed `government_debt` | 25 | +| proposed `fiscal_balance` | 21 | +| proposed `government_borrowing_change` | 20 | +| `decisions_effect_on_borrowing` | 18 | +| `fiscal_headroom` | 17 | +| proposed `fiscal_balance_change` | 13 | +| proposed `government_net_worth` | 9 | +| `energy_bill_change` | 7 | +| proposed `business_cost_saving` | 7 | +| proposed `debt_interest_change` | 7 | +| proposed `household_tax_change` | 7 | +| proposed `average_cumulative_gain` | 5 | +| `affected_share` | 4 | +| `gdp_level_effect` | 4 | +| `tax_liability` | 4 | +| proposed `aggregate_household_saving` | 4 | +| proposed `bill_change_percent` | 4 | +| proposed `block_grant_effect` | 4 | +| `average_annual_gain` | 3 | +| `gainer_count` | 2 | +| proposed `fiscal_consolidation_share` | 2 | +| proposed `share_in_non_gaining_deciles` | 2 | +| proposed `unaffected_share` | 2 | +| `cpi_inflation_effect` | 1 | +| `revenue_share` | 1 | +| proposed `gross_earnings_change` | 1 | +| proposed `housing_supply_change` | 1 | +| proposed `income_mobility_share` | 1 | +| proposed `share_in_net_recipient_deciles` | 1 | +| proposed `share_in_top_half_of_expenditure` | 1 | +| proposed `share_with_zero_liability` | 1 | +| proposed `spending_to_tax_ratio` | 1 | + +Attribution: 263 rows `own`, 169 `restated` (OBR, ONS, IMF, DWP, DBT, the British Business Bank). + +## Conventions + +- **Values as printed.** `value_raw` is the printed string, and `value` scales it to the unit (£ million × 1e6, £ billion × 1e9; percent stays percent; words such as "three-quarters" become fractions with unit `share`). Bounds keep the printed qualifier: "over" and "more than" are `range_low`, "up to" is `range_high`, "around" is `point`. +- **Sign.** Values are never re-signed. `conditions.sign_convention` states each row's direction in the phrasing `scorecard_db/signs.py` reads: + - Exchequer costs: "positive = cost to the Exchequer (as printed; not re-signed)"; + - savings: "positive = saving to the Exchequer"; + - yields, including Table 6.3: "positive = yield to the Exchequer (reduces borrowing)"; + - Table 1.3: "positive = increase in PSNB" (the opposite of Table 6.1's yield convention); + - Table 1.2: deficits and debt positive, surpluses negative, as printed. +- **Periods.** Fiscal-year figures carry `conditions.fy`, end-year `period` and `time_basis: fiscal_year`. A multi-year total keeps its whole horizon in `conditions.horizon`, sits at its final fiscal year and is never annualised. An undated figure ("by the late 2030s", "this winter") sits at the publication year 2026 with the printed horizon in conditions and, where the year matters, `parse_confidence: medium`. +- **Prices.** Every money figure is cash, as printed; Table 1.3's rows carry `price_basis: current prices`. No basis is converted. (The first staging tagged the £595 million administration saving `real terms`; paragraph 2.14's "real terms" belongs to the Spending Review's 16% cut, and the £595 million equals Table 6.1 line 63's cash +595.) +- **"Central".** HM Treasury prints single estimates; `point` means the printed figure. No row is an inferred scenario. The one OBR scenario row (borrowing £12 billion lower if energy prices follow a lower path) carries `conditions.scenario`. +- **Rule 9 components.** A printed part of one Table 6.1 measure carries the line's key, `conditions.component` naming the part and `conditions.row_type: measure_component`. Components are never summed with the line's total. This family has 94: the 66 Table 6.3 item cells and the 160,000 employees (line 45); the £3 billion and £1.7 billion parts of line 41's £4.7 billion; the two food-tariff consumer savings (line 10); the Scotland, Wales and Northern Ireland GDP effects of the India agreement (line 77); and, under proposed keys, the Tax and Spending parts of the OBR's policy effect, the three devolved Barnett consequentials and three parts of the regulatory administrative savings. +- **`component` means a part only.** Where the first staging used `component` for a Chapter 6 paragraph title on a whole-measure row, the condition is now `paragraph_heading`; in the pension worked cases it is `quantity`. Chart 1.6's six headroom-bridge data labels name their bar in `chart_bar` (changed on 3 November; the rework had left them under `component`). +- **Attribution.** OBR forecasts are `restated` with source model `obr_efo_forecast`; ONS, IMF, DWP, DBT and British Business Bank figures are `restated` and their notes name the original. HM Treasury's own costings, counts and worked cases are `own`. + +### Which spending figures stay (rule 8 exception) + +A spending figure is a measure's design unless the red book presents it as the measure's Exchequer impact. The test applied: the figure stays when it equals the printed Table 6.1 line (from `uk_hmt_tables`) for the years the red book states, within the red book's rounding. Tax costs and yields, savings estimates and Table 6.3 stay regardless. + +| Table 6.1 line | Red book | Table 6.1 prints (£ million) | Rows | +|---|---|---|---| +| 1 Household Energy Payment | £2.6 billion | -2,625 in 2026-27 | kept (4) | +| 9 Early years books | £5 million a year | -5 a year (2026-27 to 2028-29) | kept (2); years differ, see anomalies | +| 19 Small Producer Relief | £10 million per year | -10 a year from 2027-28 | kept (2) | +| 32 EV charging | £100 million over two years | -50, -50 | kept (1) | +| 41 DIP capital reallocation | £4.7 billion, 2027-28 to 2029-30 | +2,135, +1,420, +1,100 = 4,655 | kept (1), with its £3 billion and £1.7 billion parts as components (2) | +| 43 Ukraine air defence | £500 million this winter | -500 in 2026-27 | kept (3) | +| 48 Hospitalised offenders | around £5 million in 2031-32 | +5 | kept (1) | +| 63 Administration budgets | £595 million in 2029-30 | +595 | kept (2) | +| 64 Civil Service AI fund | £50 million across 2027-28 and 2028-29 | -25, -25 | kept (2) | +| 71 Carbon Price Support | £235 million over five years | -165, -45, -15, -10 | kept (1) | +| 81 Misogyny prevention | £15 million | -5 a year for three years | kept (2) | +| 20 Good Growth Funds | £285 million in 2027-28 | -340 in 2027-28 | dropped | +| 22 Local apprenticeship services | £100 million over two years | -50, -70 | dropped | +| 25 Strategic Authorities capacity | £150 million in 2027-28 | -180 | dropped | +| 33 Rough sleeping | £432 million across 2027-28 and 2028-29 | -515, -580 | dropped | +| 35 Social and Affordable Homes | £1.5 billion | -250, -500, -500, -500 = -1,750 | dropped | +| 73 Union Learning Fund | £15 million a year | -20 a year | dropped | +| 23, 31, 74, 85 and others | programme amounts inside a line (AI Security Institute, Cardiff Parkway, UKRI clusters, NHS winter funding…) | line totals only | dropped | +| 37 Your First Home (Table 6.2 line 5) | £4.7 billion | -4,700 of financial transactions | dropped: a financial-transactions allocation | +| 6.2 line 4 Sovereign AI | £60 million from 2027-28 | -45 a year | dropped | + +## Proposed measure keys (23) + +Each key below is the whole scope of its rows, stated as Table 6.1 lines or "no Table 6.1 line", with the red book's words. + +- `ab2026__package_obr_forecast_policy_effects` (24 rows): the OBR's "Total effect of policy decisions" on PSNB in Table 1.3, with its Tax and Spending parts (components) and the memo effect on the current budget deficit, 2026-27 to 2031-32. Crosses Table 6.1 lines 1–98 and adds the OBR's indirect effects; Table 1.3 footnote 1 says the OBR's EFO Table 3.1 explains the "Differences between the direct impacts of the policy measures as set out in Table 6.1 of the Budget, and how the OBR incorporate these into their forecast". Not `ab2026__package_total_policy_decisions`. +- `ab2026__package_hmrc_tax_gap_and_online_marketplace` (7 rows): exactly Table 6.1 lines 45 and 46. Table 6.3's "Subtotal including memo" (six years) and footnote 73's "£2.3 billion from this Budget": Table 6.3 prints +2,250m for lines 45 and 46 in 2029-30, against +1,980m for line 45 alone. +- `ab2026__package_parliament_tax_gap_measures` (1 row): lines 45 and 46 plus the tax-gap measures of Autumn Budget 2024, Spring Statement 2025 and Budget 2025 (footnote 73). Paragraph 2.46: "the total due to be raised from measures to close the tax gap announced during this Parliament to over £12 billion in 2029-30". +- `ab2026__package_property_and_savings_income_rates` (2 rows): exactly Table 6.1 lines 61 and 62. Paragraph 2.50: "84% of income tax payers will be unaffected by these changes"; "Around half of the revenue raised from this change is expected to come from the top 10% of households". +- `ab2026__package_breathing_space_for_households` (1 row): paragraph 3.2's "Government action at this Budget to provide breathing space for households", which the OBR expects to reduce CPI inflation by 0.1 percentage points in 2027. Crosses Table 6.1 lines; the red book lists none for this figure. Table 6.1 groups lines 1–11 under "Breathing space for families", and Box 3.A's "breathing space for households" names line 2 (the VAT cut) and line 1 (the Household Energy Payment). +- `ab2026__energy_debt_circuit_breaker` (3 rows): no Table 6.1 line. Paragraph 3.8: "a new debt strategy to address this and act as a circuit breaker on energy debt … via reforms to prevent debt building up and a one-off targeted debt relief scheme for eligible households". +- `ab2026__package_parliament_fuel_duty_decisions` (3 rows): Table 6.1 line 3 plus the fuel duty decisions of earlier events in this Parliament. Paragraph 3.13: "the government's decisions on fuel duty across this Parliament … compared with plans under the previous government"; footnote 98's baseline is the 5p cut reversed in April 2025 and annual RPI uprating. +- `ab2026__minimum_wage_uprating_april_2027` (2 rows): no Table 6.1 line (a regulation). Paragraph 3.18: "raising the National Living Wage and National Minimum Wages" from 1 April 2027. +- `ab2026__package_business_rates_rhl_sbrr_and_transitional_support` (2 rows): Table 6.1 lines 13 and 14, the three bullets of paragraph 3.22 (SBRR thresholds; the real-terms freeze; the SSB extension), to which paragraph 3.23's "As a result of this package" refers. The red book does not say whether paragraph 3.21's pubs relief (line 12) also counts. +- `ab2026__package_regulation_tariff_and_sandbox_business_savings` (1 row): the executive summary's "almost £800 million per year through further cuts to the administrative burdens of regulation, extending suspensions on tariffs, and introducing new cross-economy powers so firms can safely test new technologies faster". Crosses Table 6.1 line 16 (business tariff suspensions) with measures that have no Table 6.1 line. +- `ab2026__regulatory_administrative_savings` (5 rows): no Table 6.1 line. Chapter 6 decision "Regulatory administrative savings" (paragraph 6.36) and paragraph 4.52's "up to an estimated £640 million per annum of new gross administrative savings for business"; its £290 million, £116 million and £105 million parts are components. +- `ab2026__package_parliament_regulatory_burden_reduction` (1 row): no Table 6.1 line. Paragraph 4.51: the 25% administrative-burden target, "with over £2 billion of net progress secured" across this Parliament. +- `ab2026__package_bbb_equity_community_lender_and_venture_debt` (1 row): paragraph 4.46's "three measures": £1.4 billion of commercial equity commitments, £165 million for community lenders and £450 million for a venture debt product, which the British Business Bank's additional £5 billion of capacity enables. The red book does not map them to lines; the BBB's capacity sits in Table 6.1 line 29 (Table 6.2 line 3) and the Growth Guarantee Scheme in line 75. +- `ab2026__rolls_royce_gas_turbine_programme` (1 row): no Table 6.1 line named. Paragraph 4.80: "co-investment in a joint, 10-year, £2.5‑billion programme for next-generation gas turbine technologies for civil and defence aerospace"; paragraph 4.83 says it "has not been contracted". +- `ab2026__defence_investment_plan` (1 row): the Defence Investment Plan published in June 2026, which no single line covers; lines 40–42 carry its savings and funding. Paragraph 5.34: "The Defence Investment Plan will create approximately 60,000 new jobs by 2029‑30 compared with 2023‑24." +- `ab2026__english_devolution_agreements` (1 row): no Table 6.1 line. Paragraphs 4.67–4.68: "the final stages of agreeing 10 new devolution agreements", after which "over 90% of England's population will have a devolution agreement in place" (the 90% includes existing agreements). +- `ab2026__package_dwp_fraud_and_error_measures` (1 row): Table 6.1 line 47 plus the DWP fraud and error decisions of paragraphs 6.19–6.22, which have no lines of their own. Paragraph 5.17: "new measures expected to save an additional £1.4 billion over the forecast". Line 47 alone prints +600m over 2026-27 to 2031-32 (and Table 6.2 line 9 +340m). +- `ab2026__package_parliament_welfare_fraud_and_error_savings` (1 row): line 47's measures plus the fraud and error measures of earlier events in this Parliament. Paragraph 5.17: "bringing total savings from fraud and error to over £20 billion over the forecast". +- `ab2026__package_budget_barnett_consequentials` (4 rows): the Barnett consequentials of "decisions confirmed at Budget", 2026-27 to 2029-30, excluding the technical reduction from lower employer pension contributions (paragraph 4.150). Crosses Table 6.1's England spending lines; no line of its own. The Scottish, Welsh and Northern Ireland figures are components. +- `ab2026__mod_counter_fraud_strategy` (1 row): no Table 6.1 line. Paragraph 5.32: "a new Counter Fraud Strategy in November, with investment in fraud and error analytics and closer working with the Public Sector Fraud Authority set to deliver £250 million of benefits". +- `ab2026__state_pension_triple_lock_adjustment` (2 rows): no Table 6.1 line (its savings fall after the forecast). Paragraph 5.23: "adjusting the Triple Lock from April 2030 so that the State Pension rises each year by the highest of 2.5%, inflation, or the amount required to maintain its value relative to earnings as reached in April 2029"; DWP's estimate (footnote 61). +- `ab2026__package_hmt_distributional_budget_2026` (1 row): HM Treasury's own grouping, the Budget 2026 policies modelled in Impact on households (paragraph 1.4): Table 6.1 lines 2, 3, 4, 5, 8, 11, 15, 26, 39, 49, 52, 59, 60, 61, 62, 72 and 95, plus the private landlord registration fee, which has no line. "Decisions announced at Budget 2026 (including those pre-announced where funding is allocated at Budget 2026)". It excludes line 1, the Household Energy Payment. +- `ab2026__package_hmt_distributional_since_ab2024` (1 row): HM Treasury's own grouping, the modelled policies and public service settlements "announced from Autumn Budget 2024 onwards up to and including Budget 2026" (paragraph 1.8): the lines above plus earlier events' measures, which are not Table 6.1 lines. + +The coordinator's shared spellings: no kept row has the scope of `ab2026__gas_standing_charge_moved_to_unit_rates`, `ab2026__pwlb_debt_relief_councils_severe_distress`, `ab2026__private_landlord_register_fee_65`, `ab2026__british_business_bank_defence_commitments_accelerated` or `ab2026__package_household_energy_payment_and_electricity_vat`. The dropped rows on those subjects were design: the £108 current gas standing charge and the end-2027 delivery date (parameters), the £500 million and £200 million of PWLB debt relief (allocations to two councils, which Table 6.1's footnote 2 excludes as adding nothing to borrowing), the £65 fee (a parameter), and the "up to £5 billion" NWF and BBB defence commitment (an investment total). No red-book figure covers exactly lines 1 and 2, or exactly lines 1, 2 and 3 (`ab2026__package_household_energy_vat_and_fuel_duty`). + +## Quantity and baseline proposals + +- **Proposed metrics:** + - fiscal aggregates (restated): `government_borrowing`, `government_borrowing_change`, `government_debt`, `government_net_worth`, `fiscal_balance`, `fiscal_balance_change`, `debt_interest_change`, `fiscal_consolidation_share`; `block_grant_effect` (AB2025 list); + - household and business effects: `household_tax_change` (AB2025 list; worked cases), `average_cumulative_gain` (multi-year savings), `aggregate_household_saving` (AB2025 spelling; consumer savings), `bill_change_percent` (business electricity bills), `gross_earnings_change` (AB2025 spelling), `business_cost_saving` (administrative savings to business; replaces `operating_cost_change`, which the enum defines as an official operating-cost line), `housing_supply_change` (the 20,000 additional homes); + - shares with exact meanings: `unaffected_share` (the share printed as unaffected), `share_with_zero_liability` (half of RHL properties paying no rates), `share_in_non_gaining_deciles`, `share_in_net_recipient_deciles`, `spending_to_tax_ratio`, `income_mobility_share`, `share_in_top_half_of_expenditure` (Impact on households context). +- **Proposed units:** `businesses` (15 rows), `dwellings` (1), `ratio` (1). +- **Proposed baselines:** `hmt_no_budget_2026_modelled_policies` (paragraph 1.6, "a counterfactual in which the modelled policies were not introduced") and `hmt_no_policy_changes_from_ab2024_through_ab2026` (paragraph 1.8, "a counterfactual in which the modelled policies and spending settlements were not introduced"), one row each. Other rows have `baseline_policy: null`. + +## Not staged (rules 8 and 9) + +The rework dropped 1,267 of the first staging's 1,699 rows. 234 proposed keys named only dropped rows and are gone. + +| Category | Rows | Examples (locator) | +|---|---:|---| +| A. Measure parameters: rates, thresholds, payment amounts, dates, durations; worked-example inputs | 829 | 817 former `policy_parameter` rows, e.g. "£13.21" per hour (para 3.18), "£2,500" surplus earnings threshold (Impact on households para 1.4), "£13,319" withdrawal threshold (explainer FAQ). Also the £100 Household Energy Payment amount (4 rows, e.g. para 2.48); the £108 current gas standing charge (para 3.9); the £2,000 rise in the full new State Pension over the Parliament (para 5.23); six worked-case inputs, e.g. Tom's "£10,300" income. | +| B. Spending allocations: programme funding, capitalisations, investment totals, spending levels | 221 | 166 programme amounts that are not a Table 6.1 line's printed impact (table above), e.g. "£120 million" for the AI Security Institute (para 4.130), "£30 million" for Cardiff Parkway (para 4.157), "£1.5 billion" for the Social and Affordable Homes Programme (para 5.4); 46 investment totals, e.g. NWF "£15.4 billion" (para 4.44), Your First Home "£4.7 billion" (para 5.5); public sector net investment (Table 1.2, 6); departmental spending levels (2); £200 million of private AI model credits (para 4.132). | +| C. Financing: DMO remit, gilts, Treasury bills, NS&I, cash requirements, financial transactions | 177 | DMO net financing requirement "£258.6 billion" (Annex A.8); Table A.1 gilt sales by maturity; CGNCR "£144.1 billion" (para 1.47); Table 1.4 financial transactions by programme, e.g. Student Loans "15.9" (55 rows); Sovereign AI "£60 million" (para 4.135). | +| D. Contingent liabilities | 14 | Table 1.5, e.g. Department for Energy Security and Net Zero "1.2-1.7"; Total reasonable worst case "4-4.5". | +| E. Programme inputs, scope counts and targets | 24 | "670" HMRC compliance officers to recruit (paras 2.42, 6.151); tariffs on "over 100" food products (para 3.17); "around 10,000" homes the Social and Affordable Homes Programme funds (para 6.114); the 25% administrative-burden target worth "£5.6 billion" (para 4.51). | +| F. Effects of policies outside this Budget | 2 | "more than 100" businesses using the Global Talent funder pathway since Budget 2025 (para 4.112); "over £5 billion" of services exports from the Switzerland and South Korea deals (para 4.124). | + +Rule 9 moved 74 kept rows from proposed keys to a Table 6.1 line's key: the 66 Table 6.3 item cells and the 160,000 employees (proposed keys such as `ab2026__package_tax_gap_timely_payment`, now line 45 components); four food-tariff savings (now line 10 components); the £3 billion and £1.7 billion DIP parts (now line 41 components); and the Your First Home count (now line 37). One row moved the other way: the £1.4 billion of DWP fraud savings, whose scope is wider than line 47 (rule 3). + +### Other material not staged + +- **Tables 6.1, 6.2 and 6.4** belong to `uk_hmt_tables`. Table 6.3 is a separate printed breakdown and is staged here. +- **No Impact on households bar readings** (README rule 4). Figures 1.A–1.E print no data labels. Chart 1.6's six data labels are staged. +- **Words without numbers** ("hundreds of pounds a month", "thousands of jobs") are not turned into numbers. +- **Footnote 73's earlier-event yields** (£6.5 billion, over £1 billion, £2.4 billion) are other fiscal events' measures; only this Budget's £2.3 billion and the £12 billion total are staged. +- **Background macro forecasts** that score no measure (GDP growth, inflation, employment, Table 1.1), departmental efficiency histories, Annex B budgets and Annex C charts are not staged. +- **The abbreviated 29 October explainer** has no quantitative content. + +## Judgement calls for the coordinator + +- **`spending_to_tax_ratio` is kept, against the drop list.** Its only row is Impact on households paragraph 1.10's "households in the lowest income decile will receive over four times as much in public spending than they pay in tax": a distributional statement and one of the Impact on households printed values the brief keeps. The 60% row beside it stays too. Drop both if the drop list was meant literally. +- **Mislabelled shares beyond the 84%.** Besides the 84% (`unaffected_share`, never 16%), BADR's "three-quarters … will still receive relief on their entire gain" is the unaffected share too; "half of all RHL properties … will not pay business rates at all" includes properties already paying none (`share_with_zero_liability`); and Impact on households' 20%, 60%, 59% and approximately 20% are not affected shares (proposals above). The registered `share_no_change` is a household-income measure, so the 84% of income tax payers follows the brief's `unaffected_share`. +- **Registered metrics replace AB2025-era proposals.** The four "around £50 off the annual Ofgem price cap" rows and the three energy-debt "£30" rows now use the registered `energy_bill_change` (unit `gbp_per_household`), which the enum mints for HM Treasury bill figures; the India agreement's GDP effects use `gdp_level_effect` in GBP (the AB2025 IPPR precedent). +- **Financial transactions and public sector net investment are dropped as financing and spending levels.** Table 1.4 prints forecast levels of lending by programme, not Budget effects. Public sector net investment is an OBR aggregate, but the brief lists `public_investment_share` for removal. +- **Jobs and induced homes stay; delivery targets go.** Estimated jobs (NWF 100,000; BBB 40,000; Rolls-Royce 3,300; DIP 60,000) and Your First Home's "estimated 20,000 new homes" are estimated effects. The 10,000 homes the Social and Affordable Homes Programme funds (in line 35's title), robotic procedures rising to 500,000, 2,000 learners and the 650,000 workforce requirement of the infrastructure pipeline are delivery targets of allocations. +- **The BBB's 40,000 jobs is now `restated`.** Footnote 135 cites the British Business Bank's press release. + +## Printed problems and differences + +- **Spending figures that differ from Table 6.1:** + - Social and Affordable Homes Programme: £1.5 billion (paras 5.4, 6.114) against line 35's -1,750 over 2027-28 to 2030-31. + - Early years books: "£5 million a year for three years from 2027-28" (paras 3.20, 6.122) against line 9's -5 in 2026-27, 2027-28 and 2028-29. The amount agrees; the years do not. + - Local apprenticeship services £100 million (line 22: -120); Good Growth Funds £285 million in 2027-28 (line 20: -340); Strategic Authorities £150 million (line 25: -180); Union Learning Fund £15 million a year (line 73: -20 a year); Sovereign AI £60 million (Table 6.2 line 4: -45 a year). + - DWP fraud and error: "an additional £1.4 billion over the forecast" (para 5.17) against line 47's +600m in total. +- **Regulatory savings target:** paragraph 6.36 prints "£5.6 million"; paragraph 4.51 prints "£5.6 billion". The executive summary's "almost £800 million per year" and paragraph 4.52's "up to … £640 million per annum" have different printed scopes; both stay. +- **Repeated amounts that disagree (not staged after rule 8):** local AI partnerships £30 million (para 4.135) and £15 million (para 6.101); Leeds College of Building £28 million (para 4.71) and £15 million (para 6.53); business tariff suspensions on 269 goods (para 3.28), 350 goods (para 4.147) and 361 extensions plus 72 new suspensions (line 16's title). +- **Pubs relief:** the text prints an extra "20%"; Table 6.1 prints "20ppts". +- **Table 1.3** labels the new forecast "November 2026 forecast" in an October Budget. Chart 1.6 prints 14.5 where the text says £15 billion. "Debt is lower at the end of the forecast" fits public sector net financial liabilities (83.2% to 81.7%), while public sector net debt goes from 95.5% to 95.6%. +- **Paragraphs 6.225 and 6.226** repeat the Machine Games Duty measure. Annex A footnote 181 points to A.17 for green gilts, which A.19 covers. +- **Pension Tax Allowance explainer:** the FAQ prints "£13.319" for £13,319; the explainer prints both £13,036.40 and the rounded £13,037, and "up to £94" beside Steph's £93.28. The FAQ repeats a question and prints "part-time wok". + +## Verification + +- **Checker:** `PYTHONPATH=. uv run -q python pipeline/ab2026_harvest_contract.py sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook` reports 432 rows, 0 problems. +- **Determinism:** two runs of `stage.py` write the same `claims_staged.jsonl.gz` and an unchanged `manifest.jsonl`. COUNTS.json pins the gzip's sha256. +- **Spot check (the rework's, 29 October):** 30 rows drawn with `random.Random(20261029).sample(range(432), 30)` (0-based row numbers), each re-opened at its locator in the hashed primaries: red-book and Impact on households pages extracted with `pdftotext -layout -f N -l N`; table cells read against the printed row and column. All 30 match on value, sign, unit and period. + +| Row | Locator | Printed | Staged value and unit | Period | +|---:|---|---|---|---| +| 44 | Table 1.2, Current budget deficit, 2027-28 | 0.5 | 0.5 percent of GDP | 2028, fy 2027-28 | +| 53 | Table 1.2, Public sector net borrowing, 2030-31 | 1.8 | 1.8 percent of GDP | 2031, fy 2030-31 | +| 68 | Table 1.2, Cyclically adjusted primary deficit, 2027-28 | -0.5 | -0.5 percent of GDP | 2028, fy 2027-28 | +| 75 | Table 1.2, PSNFL, 2028-29 | 83.8 | 83.8 percent of GDP | 2029, fy 2028-29 | +| 92 | Table 1.2, Public sector net worth, 2027-28 | 63.4 | 63.4 percent of GDP (inverted, as printed) | 2028, fy 2027-28 | +| 115 | Table 1.3, November 2026 forecast, 2029-30 | 73.8 | £73.8 billion PSNB | 2030, fy 2029-30 | +| 118 | Table 1.3, Difference, 2026-27 | 17.7 | £17.7 billion (positive = higher PSNB) | 2027, fy 2026-27 | +| 125 | Table 1.3, Underlying forecast differences, 2028-29 | 11.4 | £11.4 billion | 2029, fy 2028-29 | +| 130 | Table 1.3, Debt interest, 2028-29 | 12.0 | £12.0 billion | 2029, fy 2028-29 | +| 133 | Table 1.3, Other forecast changes, 2026-27 | -4.0 | -£4.0 billion | 2027, fy 2026-27 | +| 138 | Table 1.3, Memo: current budget deficit (underlying), 2026-27 | 0.6 | £0.6 billion | 2027, fy 2026-27 | +| 145 | Table 1.3, Total effect of policy decisions, 2028-29 | 2.1 | £2.1 billion (positive = higher PSNB) | 2029, fy 2028-29 | +| 178 | para 2.46 | over £12 billion | £12 billion, range_low, yield | 2030, fy 2029-30 | +| 185 | para 3.2 | 0.1 percentage points | 0.1 pp reduction in CPI | 2027, annual | +| 216 | para 3.16 | around £150 per year | £150 a year per EV driver | 2027, annual | +| 217 | para 3.17 | £120-185 million per annum | £120 million, range_low, consumer saving | 2026, publication year | +| 226 | para 3.21 | over £3,000 per pub | £3,000, range_low | 2028, fy 2027-28 | +| 234 | para 3.22 | 120,000 shops | 120,000 businesses | 2029, two years to 2028-29 | +| 235 | para 3.22 | £3,250 over the two years | £3,250, cumulative | 2029, two years to 2028-29 | +| 256 | para 4.146 | Wales's economy by £80 million | £80 million GDP, Wales component of line 77 | 2026, publication year | +| 273 | para 6.36 | up to £640 million | £640 million, range_high, saving to business | 2026, publication year | +| 299 | Table 6.3 row 4, 2028-29 | +0m | £0, line 45 component | 2029, fy 2028-29 | +| 327 | Table 6.3 row 9, 2026-27 | +0m | £0, line 45 component | 2027, fy 2026-27 | +| 337 | Table 6.3 row 10, 2030-31 | +0m | £0, line 45 component | 2031, fy 2030-31 | +| 350 | Table 6.3 subtotal, 2031-32 | +3,160m | £3,160 million yield, line 45 | 2032, fy 2031-32 | +| 354 | Table 6.3 memo, 2029-30 | +270m | £270 million yield, line 46 | 2030, fy 2029-30 | +| 356 | Table 6.3 memo, 2031-32 | +265m | £265 million yield, line 46 | 2032, fy 2031-32 | +| 369 | Impact on households para 2.10 | £33,500 | £33,500 gross household income, decile 3 | 2029, fy 2028-29 | +| 395 | Table 2.C, fifth decile, 1 adult | 30,400 | £30,400 | 2029, fy 2028-29 | +| 400 | Table 2.C, fourth decile, 1 adult | 26,300 | £26,300 | 2029, fy 2028-29 | + +### Tranche 1 finish (3 November 2026) + +The rework stopped before its report, so on 3 November each instruction of `briefs/rework-hmt-redbook.md` (on the staging branch) was checked against the rows: + +- **Removed metrics.** No `policy_parameter` row remains, and of the nine allocation and financing metrics only `spending_to_tax_ratio` remains, on one row (paragraph 1.10's "over four times"). It stays: it is a distributional statement printed in Impact on households, which item 3 of the brief keeps, not the fiscal aggregate the drop list targets. See the judgement calls above. +- **Rule 9 and the "Also fix" section.** The 84% row is `unaffected_share` as printed. No row carries `ab2026__package_tax_gap_timely_payment`; line 45's items carry line 45's key as components. No measure without a Table 6.1 line is named as a package. No kept row has the scope of the coordinator's shared spellings (see "Proposed measure keys"). +- **`component`.** Six rows still used `component` for a chart bar; they now use `chart_bar`. +- **Stale files.** The family had no REPORT.md or spot-check files left. +- **Keys** were harmonised across families on 3 November; `stage.py`'s `KEY_MAP` lists each rename, and the proposed keys above give the final spellings. +- **Checker:** 0 problems. **Determinism:** a re-run of `stage.py` reproduces the gzip and manifest byte for byte. +- **Spot check: not repeated.** The brief asked for 15 random rows re-opened at their locators. The 3 November session's network policy refused gov.uk and assets.publishing.service.gov.uk (HTTP 403 at the proxy), so no primary could be fetched, and no row was re-opened. The 30-row check above is the rework's own; nobody independent has checked this rework yet. + diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/claims_staged.jsonl.gz new file mode 100644 index 0000000000000000000000000000000000000000..e33581a5eceb04d1c8868fa502069d938388f86e GIT binary patch literal 36363 zcmV*KKxMxliwFP!00002|LncnuHwj+HhRCGB9v09a;kv>bIw$^l*g*rmO&m_guK!$n}Z&XectiO2oG4EXp-g*_RXKDRB059Y3Ijh67Wv#@Raf;*$>%=GRC zBirB$FL zbhnA?@sa5V_~FFzN2WdC_i%H%a1ktK&S1D0-|}E!)Qt}Q>Xy5=o;jc5D4+cH7yM}I zde%S0!_@uoX_h;F@_*p#88`if#|L*j_GU8kDQ&&cd{rdQ}(xm#LvtnBy|va&B!-@6YC* z<-o+hkYi+)&*qCEOgufCepADz%uUbd1DM>2HHLMdC(iiTTFmHM@q*78P7MC-xpUX& z<3adlPsV&04DeaF)`QX1bZ$9a8a+)c*l5Jn;=-6|F~+8NMZ9~pQLh?3{M2^gGS|Z| z-hSaD5T<+1{KX7JWXYb%7i05U# zTODI;4kz7qqup*a>f>gsJ8JeCMswI{_Uhw)eNyl9es|b!hNJhVoJR5E&+M~D>$}A7 z0xPg-*sn-vXl~fEA78Vs(V-zNJR85B2Ep8a{{8nL(>(BN@|LhhHJ^L;)`-_`-TT_& z?t7eX^ZSm;jry?HZguLTMzhtKbcej#Yw_>j)|;J9uixtoe*F3K;8i54gSS7u{ExxF z7}VzD2}$AK7gA_GPhr79CLNyzQ_g<-3krZg=b%H*Et`SVFpvT<4*bLle1r}`}Vcb3dLP1 zDzhBMzXYDiZc(6}8R#XAgejR~BXCrSq^IGlevqt5Ps~L z4+3GLNq_Qx{-6Im9zyQ{zmugzg?_bJPs8KK5m!BF%FVIoy0a&H!tW-moLaZj6{Kv| zj{<*&hQBTF$K!YcO9uF~;66ggjKiQ2M}&khZ(+k11bZeME>NtWvp@g%ABHyp<&bEH zKVt-{0UIt6%-W&w0ES6nt*{|N5@(*}6C%3sOT$S1vRVK0kAF5BMIcy;lphoEjt?|@+Kvqi9Gz1nP?AA&VBf@uWP2$qUqhLIn^qWL=?1QWkPp}`+n&OMA0Wl?X? zG`#H%Zwj3vv08^GqqPc82s*HK^F=_%O*$sRwsoILmVz2z?a7+nAP$A9~5 zZo}#OIuUY*Q>t;IQ>wBf(i90yU_QwZb5Ian&vPF@el{8u0chNr;mwB~c8QH@ z)7TA0H(3@=z09ATdW8R!_Zagf%Ao39Z!r)O`?Uz%+zn0A`C3k0f2d zBW*9$9ZYA8pc@0+vHRdWnBF)BxzA&_n3%RrTh&Ib_Y}bu)B-^rMIAxhpj!$vpU-Uf z0iI}zJuQbX;DW^SpM}@mz!xU)0bS>~5BQbQfv=42O%YhTNvul-xS*gT9>Y;QSl4

BNI zE+9ER^vDR4z^BE0?s`F_Tz<5Z9l3zDfW1P7;OGB^t{Lew!XNzg9sN_n*F>TSqX(IC+0bL}7DcMTFT3mQ`JZ?z=TRjG)3flp= z$b`2?Egq0gq;k^YK0=Tk6buItf8fJ0Oe}UTUP^c2 z##1 zTmux6T#IwW=+5Q}#YMu7l#8UmMu&}NT{_Q7@K zE4a&sIrL%k50{P@IpciNINu%NP8zo7_%(hq#KXV{7fp@k(+t_*m3rz&cq;n{1LkP4 zk{Sg!!ayuMAt)v9iUY=Dy8i6rk3Ze4X#BdmV=x$uB8{cO$xYs=kD=zNxxfC>4Y~Qf|Y~CSI z3XuFDMMUAF3>JJTgo(WYVPgK$5BN-^+}|Q~V5$J1t}w{sz0e>Zib{00n<&dU3OU8( z8icSUolFWWQ&I|$Ir3azrTeELXX#BrY{iVP|C3 z+j0%H8fwpKr)q3=Oxo_l8sJ6~d}$ljH6;GUO`z>cWu}#SOl%=-6ptCd=jP}Rt!(g0 zWczhtp+FcWlL4Hx20r{}=zfXJBErTeEZie^&QpQud3daX&nb1s9M3Gr^5J|*zZ)(* zfK(rqC3% z7ns9|m?5Y`(qnMh`^U^jJ7>lK>Z zV^$2)p5bxy+t*$_gs?^0NG^k zP&V0*_(UO_Rg_I86$^hJ$$M4EyPzj7YGJs*URcBt-odn)(7M#1%8f5`@iCOe7}R$y zMUmwB0mfWy!)I8}VBtACq>i<#MI=2(+3yFgZu`q8_90Pr10@(%61Fs-U^)40$UV8g zp15v+*Rf`@96uK`s#HWTi9E z3L8NKr}Q&fU&v#scDSNW1qqXqA~rSeIaz?X0}}3Awk0aeqI(jlDDZv5f@ecZXj0>nxgVDuyi zB2_E0oDstPvyVT%d4ChTM)+mbNhDT_h+YLPQaBC=nTgm-4u|-G1rIKb#1}<0lEsUt zb_VG>lue?8 zktj7+Yr=vWw4kWXuzZ0U+|Sl{3{&+MM+A>G8A6yf0mY=;PgpHD`4APUFeV`jaC|vb zIgo89lO)CiJeXU#FO>FUR@vNam`h*C^Q#)%l|Y=vfFtg}#>9<9{!`WsmdmkWPGl{b zxk*wAGl3P8u+)kNTAV2MvXx53U5>_~xvM%Ez0-4ICqarmO^T0Ibz-1&Fg+L^L$No+ zEJ!z!mS86i1T*BUUTZU1A&7b+Acp}Ff*5d5!g5!TObF@WwOAG}lsF;A@(Kj<9<##y zxTM~gYnlTywU~+XJ046{mFNWUoU3Tl;jnp!`k{@puOAwgq1EL>*Pzg-C!nyep>P`% zdR*)TMWJKaSR&eXnizEMi&}CkwU12-*^ZSJjsJF~5@{I=cb;Uscu7UsbxC85Ng9*R z1C!3Ni10NvBfFw5WC{`+8WMA!Se0P-#k2yeq}UOEHoeib&{O3>x`O4v$?NF`26OAd z(cL657Yn(vS3AE(VJ{TuF*N@&+Z=McLYXL748f~TIqcrZcr#XA?2f+2dyVrBh?X$H z8TFeAix#1h@*W!SHUVBVZ?IC$TBiUQO6tQd;1e$MbA%s2F+|vv{ihOpr0o%pt$)(m zUF62DnYSz+O=!u`>NjW1J$o6pAql%xhWjx=9KHs#lYv?0HPTp@2kSCV>vL9Z7`1u< zSeQ%OE(a>B`F#aq#!Cj4zf#omB3eWZlBEI3N7qJ!V;*pPbUL52daYjoBxcg7%Ynjv zP8`HZE*T1b@OV`slg}=qDfW3IU+Q{jw9A8bsZ)p{4z+dx5VuibE(g{cMms04(cG2{ z+1^qFCK`6-12w3f3e;AYLgQH;JX@VTShYgVuW1T$K^DEdhtx72QAdHRGEFpmo$fRm zTLukXxelJOnM2oLAv>^Dt+=3#S1b3%t5u4P*VzKoNu$|}6{OKPvMqFDwD3Gi`4Yo~ z;m_@bzZTXegE3@@RO~Dh1$6{d%NtiItYgF7qVCcGi{_(UYxUUgQ|>TxG_^p=sV9Pk zZFXgb0_9m_E`uN4OTY%~L$FJ9-Ke$tY-=lQcw`7URQyBMK;@!1jA10R*qU{0KHbnU z!50`4EUy^cTa+!6)XsJSOM6wLpSAK+O#Ve5O4Od%jU!?^#TGF>5TV5uC9B3p{q8@Q zeC1rp)_slK8o4!cpAxxuhze|-ZL)8(UTrqcZ-d*=FsWfu!{kX5Om@h_H4c)8yL)LM zz!(>Ok$rnUx1JS{l4_fVXW1$b34uH68$PtfzqG z7T}IFD6*(pUqlt3TD!qsjz+FGmW6k5pznWvSCRFq;dMpnYE{)p)>eXP^Z~2JrYZqO zHI9sWRNaV1-nqd#x+?`#Ev%Rd5aBH3GpUvskJk(edqTqNaK~TB?l)#op&T+3VS%vH z6764tSpVj+D;yYOYNj0B)K%RgPgqpqbTxmZsKz-BJmlrG7$2zGn1@~o~9N}E!(I?bBg8^%_;kGO8Lkn;fJWTGExYNQJEMcK6gb>Mzc|E z8s|kB`CS@@Mt_a|8vQl;%lM&MFKb{W$U_PI&MQD4xF`}Ncd^#mXbVPRtAeP(V#lC8 z=OJQCEK#huSK?9Pn%$9JVdTU!#oGjt&Q<@70aALL8Q(YKJOI$bqDCI#{qZ=lT4f(~ zNpeQq6Vu8cJG9Oqa%L0u=>d&a^N8jV%_BQ_=Qrq5N!TPDPh^bhto&a_YKf%bgvX0y zc!)MlmB=(Dr*~-D_)P8ZI4!G0F7jYKMr+I`7KlK|D~&Z)k-M{6`>9}uwjveb`IZjz z4Y!q=xZ+7~vy^Bt%TLgdVk>gYQ-fk?eBn&;p>Xk^{`uPx$r;52UPHBP7CbUO=9 zYk<}OT{eI&w=o9hJacM(`>Eb5>%FqZV2!~VgNrqYw=)~W%ZkB|+RNp&RVf7tJ5#Y% z@^WJIp7!#NwwGTdXun#^hwF>ra*G`oZg=s9tQ~` zi?hfgKyS{9+1e*8k6Eh4O@ufG<`)iV-XS!X%NQbmScY%vKCszhG+liO!25WxKQq~b zxs(H;;#b1tEju++-{6Tm!Y0SJ%yrm9n7R?&5;?kJM8W8*v7x6BP(N#)8%W>@=|_oPjMViRK-qNfJUL`HB%v8?btvd2Xhad%4A320+Qa24vw0Y! zM)v0){{>WF0lg%$p+#$Zb+h%H=xW`dOR1;vGI9uu8CYpV1tJzD)2-aI*6xYJ%yNX{ zY}-p_j{mmsNf=iq^4aT`AKu)=hU|lu%zZp0sS@0jeJ4G{p;Or7n7}wznR&8b{fM5) z>2}yk(8dCNOE$7yk+VJIcr9QqVICh}k&7Ye3GQW1OX!uA#dLUSap9%fDy6ANQ_+P{ zQP~Z7-OPr(ZV4k>b8*b0@u^{O>f`l75|Ma^D;o0t=XdOP*8`n>s|5PjMPQ>`B+Cnj zcDGJF*Wsvl>J;iX30^WZFKy(dv_cq(ib;N07K6-iG9*+M_wa6!QXGJWDxk@1Lx~JP z18=?P2ZDdZYhmPgs4geSGqz@v!2>^L!`f6=bT6|j`cyl31zpGJ-;E4R6}bG=I-+mC zn4^FJUkmRYZHllKO4M09f)!DX%pIay2@wV1TR3Clc-wVutKNck^nc?c<&cTtSR8?$ zm_yR60ZsJ)HVi&mIdNNI^8o&%-rX#LuuhRq}iD`-hcL{5HqJE@)-m#2t+q`^AM>+UFzCG*uSLhZ~r%d6eb)l?4CxmlOiaI5kglk=(j! zukzaXVLnhoH6EzP;LdCcSh{dv#z)q~ia&Ti!XqpmSb&G{@rBjWJ|PFTGzV%9)EuZe zP;=nrabQG&**P#;>Jxf$UZ=)F9@L=&FlBUibLfPGBggqYhQGXE4pBOMEUn8W@aP!b zqk42cqp%~$Compe6X^?FVX<}4B#=e|=@ZBPUo~@RDsL8G=87%EE?$9f<)5=(?{QN2 zg9@+<>!%Bx4}D9|o}$t2<}luaW@hg}^U7eaE7kU*VXs$y*i-AEVJ|Q2;ml!_O1l`? zQwwqauvgEV{ZK!4o5hpE-tJirX9Row^24552Mv39VGm~xb=)Wh_S8aL0_+)?Vb8cS z*xS^a_L5*vbh0fs?5TCou$LG1aOR$~YQ5Gf2KUrToImd2oMZ$(n!c+-zKuN#FADh@ zrANMS9W?UgL_V51((+Cr$QQ1}MIc`%JMwj|4Ec7?R5>H$Gs=&AY8^E4w*ZwtP;WfL3ONxcsL%bIj3OBJ)YiKX=Di$^}V`1Y8vGA%L#pT38 z?K9pB3#H?@QR`{H@gDtBuMrEc*>zk}EYx1)y|7TYk_*Ga%kd?g+Ygu|hRKuVlU?-TUVngeNT+gZyAqMm1$I|Cl^$ zhxwzR|JW2Jz2@!9_L}D^4A{yXJ=8jB@Q?<54f^|me)s%9Uq=S)5Be%Rpnuun0l7fG znb{lGJgGNKgT4m+{XoBWexR>IAod4+5eLyMYNub;IEb8}-_8vBdZ({J|9nBee}15^ zV?FiEVeHT=w(q1~ko;mHa5XA;$&?4qhGKU_>j|CN>1VJnZv zDz0=CPSo(P>)H;h@rm`!@rm_oK{#E!|HKHVV>EIjoGRCE=!y;JS-1MA2&e1xpBUkE zKtgVW6Y&StdQpv-E~gL!&Pn!ih0SXLIbD_i#2}{~@VP-w*5NO#+uQ}!;m_V~uG+X7 zkkbYDPYiO};hY=fRLT8DQH7^3sN{Yld*SKEX$wyu6XbNQ{S$+ncCh9KIayu5QBdgQ zB~{mNWDcOD={sRxk>v->!o&2e|)vI|Ml^XBahEw+$UStebv`J z7R}5ai{>?>qxRF55FNGewh(j_{@kkWzkbHenNALn+Ra-+fYh$vLI5&!2luo)cnu(X z*#WY5%>b!g#3cks?KUn1AVXJjN4t{O43M4d2FT781Eh8@mk=Pe%efGMjNH(=^ScH} z?V3J^Ht-n%QoE}Q0Z8q_K7SYXdh3P>K%O(QV5V%#99lM9$^CP8W78VDqq~7Z7w(Kv zK7Ka6(ewq%0mC))K-67_#lJj{KJ)n6Cux(CDlOU#zC+8D)R~OC!h22KdwvyYaZWM< zEluB*1MT5G;^z&ty>kRw?NKibXoXiDkfNu(>T5vT%MP@?YX{oHd)ChzX#3{~wAwRY z7|;sud~rZ~Io^2L3FhWRNNxI;LO;7{!T;hYANn} z)DZ(`8!-@~rVbF;6*WaXK)+K|f6L2>2f#VWikh9P+iPm~|Jl0#L)6sX{#{X1IQRQq zU5R}SHM^Njp)`Hhg__!zf408-5H+>yepl2Kp8DeUnwR6L*LzLvn7>HJyh2Uwhu;-7 zrQ5yVuJyIsea)`f&u-W3Um0p@FZfu)V z*^^8AWlJtS3EXN&{#iTnH^Hs;?H35Q(%s)G%5il; z?*3MG$5pF@j;o`?t&RXV>j;2NaI1p?3WQr3C(zO@;8WnXnH_GMR}Z&3bl|*02R6X1 zjwUD&ZdHImtDxGemlR;o%3OPurtj+ER>vHib>Cda^i#IlMRABm}G<-hKL+(MT%H0?74H^^+*3@iia1Ry2P!hn^Q42c* zu8dLWG>USXT~3Sw&Pi6lZCstxY+FQv2HZmhF_Z;xb=<36H(Jlz9b! zQs%`Zy_vP=quwm2q-1FWG;)G{Gjlmf zF?Sm5HQ1je*!Ru{>~&Pgo?tIxLmKr~QMH1~8XS@n>|2?wIAZQJ*lVyqOR(>s57_I7 zls&;-#HAo$wJtj_B{$f&a)5nHgS`g(vjqF{luvpT>~%EEo?tIyVi2${JSZkN*w=G_ zeO-gS2K(~_`^Nczy^h4$6YNDi4%(eByrD^Mn|&vT&Hl*UOpad*~3LZL+Srd9xCZIYi|@arSU$Ix}}*z2h|2SR6! zLj($DFCK>zdm+fQG2r?v0+%&1X=J(rWNKvheZl0F0GTpkkoc9i5OWD599p&&EHkA% zKaNl8{&-mPTzB>)4XnO>W$;iN{$l_8e_4a~S@qx853fEHkyWhm=oEPL8Q^BL@K9)( z5IA87P&NvIKqy&=F0fED&I3d)X&ewT$WSWTgB47f3AaHlK*0L;wb51rQ(;&m60UOK z?jxKf?jwMG`)U}WPzfNbbvmvJk2BwE2By)rr&_@pc5-{ZnN2F8fSfBx}bM)#(|QUpOKc|8i)DYKSH?P|1Z?ee5{HJ>8_I=7rSC6qDdgF^AR z1U!ob|EBN_A_zuKtQ6CDks!IX+U-??WbceXQYT041d_6$Zl@@WEFvSS6d2j>tO28X@~%HqHkjb?(Ye2q`mG+Wj&m ztds&F+x@(INSegcA>=Qhm9ruDw7+J5kbJ>NC`FK?&(f!sjk!O!0?vLN1ulLY!@t_O zdwkMgSv!3mMcNIr68HNwn?*G#xE#MvvuWh@`_MFAD&Ag~>*u(5+dD73)&8Sh@m5yV zYZygYNH4~R1anf)?PoVGcz=6cj-=z_ZU4ORR(qp%#arQ!>i3Is*zAK20)UMSdJnS1NFVB0t= zu+=`YUBOm0JM0uxp8s;(WSvewZy-VEq;>ia3$|N(8)~rK3v7+^0$W{QZC9`rh1EK} zf*M~HQdg}UVB2fv1>4?9J7B$$KZm8#aR}wf8y$>Dl#hmzkr?Ozf=7`tnyf_7K)2S_ zF#|~-R=1nig@u!NNoc4OL<{iopBN3bdp{2vN|%0HXN9hnR?3|?f7-nH;?YnigBIX{ zKQS6==X)MBl%3?7+Tp$yPnsJIn-_+LI`Ojr7x{_NP>S4&tBp*MIO~~seSENxnjS{ zBeJ2LEsuicBQuyR*ICrQ@`9>$=H6E}PP(r=5}0bY%o)06vVf`f)$9$X!hzG%{+rdB zb$Jbzy_4E2kA$S!rE`WZoh(SI{XBaksdWAnwk^FJ=a1f&YLC#Rc!aWmsdf|X4W`m} z)GVq}$mRHsn)Upi^!mkvsdgxxqeCeJm})Q6-e4+SPlfGFFUR+Ec01EId>AlPBz&Is zHJ!-UG=^0es8u++6Fx2i+NJckA3i=`2?tVJJCN3b)?4|5)-MhZwYw;%yQpM%sJ%it z@lYo5baYwHRXohyxTta3#zhCgL+$j*>GUZX9%|oCPCS&J978v9TgAitRf>(v!$a-2 z$?3N#86Il)OinzM&KN^GV>BLW*USE{mvZ5u_ORr{L*-8~Y7OmAS;50bemrcP5D$M1 zKrkcEnNUTY#B)t(Nj3&Uyc&?*wAg_EnLeOxOz z-OB6YYMs=>RYsuJezBZ>u?qug?MEvTsD*E>r(J8SRYE(ttAy$uwf4vD<&V2Cpw|Ao zB7s`^^cn@FCR~tHuhGa~>ZzfzS~~;xaRy!%R%_>Bkyss-<*RFl;tE#R^JDcXLohzO zfoU@i*Wk)en0NxL_Q>+-iV}DUW9Sse%_ruN#CiZyduQJA0ZNM+qMWLMc(YT$_>*uB zelCq3XyUOqeQ~Jy8 z({4R?L)mLMJM7~S)$DMo?9eY2J7}lv{_G%JwW+m<_F#vL@!00BP;}Y?MaN=?eY~)m z9WIm|ga^7b?4Uix`?G`a4_A$}NS;0D;bPpxRU>yYlX23Jo+A;&UantF5SL02jWQ90 zc2Dn55Gu>HnHH13J3(AfvTHMU`KjhfgY%C?5RGcn&;+3gB1sTNsR%;Js3wTr2?FO3 zV6b1Tb2?m_^CyUYUhg!`NLKnmDdI zsLA}IsngbCe1DWp(vNt*_&XQq4Y%eqOT=&gI3ze+dGL zH2mv`qetOi2Nh+4e;GH_DmW$Wa^i+s&HR1bF4yW`hHxPb|2m%NG5FUZLz&=TMFq8V zRL~~)Z|8^q_T|EV34(z%{Of3;N8w)w1Z9GM8S~T9F+Ufw`~NiO@TGvA$2lPWYaRWg ziQs~wf0StJ5THldKuuLV4k(=w;&^eLVL-Y0;7aH~2Lx$4D04tif(~?4PHp+W6h zuShz$n8=`ZFDD(~L|y?M=oleQ2W5^CO3;B06v|EqB3h`GnkBF^9b8PTP^+H%bWpzr zI?ypgnhwewGnAkM9XOPo4rEuFTDM3#xSZBB-Q0Nt-7BC29Ydt)pwux$aXQd}MA_*; zHNGkIbZ|l4Z*n)kxr)<)jw#Y~Q0kbX1RdzWqU>}KMi=#qsa~z6L!?p|r`O8KI5?44&N$cC(oHi?JOt|U7)Qrg70EcN6RY(1w*Cmzdn zaipVz#fl^yRh-Hw)UtL#wW*9&e$u&i+u9Yjv(uy#j~Kf=($OJlMUsvTTI&{?({e#U zYu&s_Ei{!!C7pd}Lj1~Gh^e;#qy1negMc5kmSqU+l@FM5bL_e9Y~N7clcdvqFg+9G zu~m;jS`ENVgU}`rqRILN1~;)BAQ{WF0d$7Ien4#|tTbq`35>^J);x|aI)8ZcL69Cw z+n;~@vsouTMDT)9a?e4T#LH)#2Y5Mt6G%mDT z5Wx2gBV5UbOyHJ_@@ zn%HZ>ss5iv7Xhqe)QqRe9T;xl+WlH@8?G@SGX=w2LbS?@aX6t_!_sl?F9qYa249@f zW~{C;PGg+Lxa}ARbCwn3aC-HD^V|-c8sjv^T{p(zq-xnZl&mtIp!2pDr$E$cJKE6r8_Ry_9<`abEL1f<^%xx2%FXn9kGbv^o zN0RWicsN;$xKDQ{XA1yufEtdH$XX3pGN3GLynr?P_O(^(hIe=dG-A?hGo#kILDxMf zFY^~O06F$x0R-WU;C)V02{Gai48g+D;y}woBqiaGmku+>0y$`n8?iM;I2LF@3k0ib z+iVEXzW{_73rbmJ)BQjH`0q9CPoWHeZ7FSUgeI`}k*DVaXhL`o97AsR^vVF@15V?a z=?wAT$DdH3%Lw*n*2EG!tBr&c7quhRSR5??$0=|o#ibKl(TkyUc905;LF`+^{)WA*x{Nxr^H={2Hbf^gXbAT0Gq7tcc6;IVmyn^0QED zM;2;bAPXf9K>$aIkkxIVq|ZJ2+@pEr^t_^Y!W>O4SjI7$xh4w_7SI#Qj55A;cL5iG z;;XQZE*jH9O%paW_`@?TpU1Z_$7dND0=-}I4I<1cKlL>0yHQW`=)_YTjvp0;N7lV% z52Wr5%<;@pWA+EI)rNu3N4uc!ERMdN6!bkcP&PxLY_v2|ZbV9CVMH<^cK#~t;Yo{r z2j2sxt{9fu{dOn_2`2;YI!sTI=Nv@Jau*=KfO#9XDBbbeFu$o>3;2JLAd|tmr_}1MM=y-E%rJr{TJIKO91qm&u3i z`R-U0+P4cw`G!Z5OZ$ z76qZqM~gBFt6y17KqK6pnRgs^Aq?Kk0zOA)hY8pSkk#naE1;NC!a2w_us6-h7PrSK8)4cWpyaF6?wc$b z#Rb`VPi<}Jg?QZ3cG-1n10b8s{Rg!L#DZC1@dE#a=3r?oCG&=ob%Ez@?0KLGu@R0Jo`7!{ z<%dvZ@G6R;`S!J8#5~%{6I1IpQ}Fw)hHf)*4$0SPA1*sz#+_w3ERpGUB<4 zDT*Q>MZ^$D38tMKDf+js7?Bcp*5pclJRD#2G+(6fMY2xBtwkH@i3#g&;T{8CB$$%X z0`@epyLKNcQvUEm*nIOZfo$d`O6HKV{q^N%+OONNVjdRFSu)r?TQ>Xm1#HE|QX|)3 ze7!{%o1yP8AwOs)@fY*C>jj}H1)uYyQbZ7DR7AB|uLwi>2-;3T3nhFI+1P1pc?V%VT^HsFdEv+U4oBg4S!MO2!0wAeK0ybKhZpW-Q~ zx6g@SwTMP1e+`Y;yOo%Elmy`^Hk2(Eoe`7B`-1)U7v*{u8>R!0Oa>E6(uTdOnMK32 zhG`Ack5_XG4addVYxYCuXm%4Iz5RTk5FptK;rLeZ;~Q@hR&9{Of$Pxg;K+F}w?Wa) z)QI!Wbp!l2>ozg3%yq)$Q}Nvia{&%LIxoW&Fez0$V-uA{#|XUHcb^tRVG@YoXU1pl zcd=M_qg|h=V0T<6_6+CMXRS7&O$#FfMJyqFV3Q3(UfZbjd-YkZ1V*SyG=-Y4(t=*F z=$tv`i1Fk)?bO=IB!ZWuz$_Yms}h=flE3^kC%erj%w{LNhaa~q#6Yd&FQc2aW$uDe zYp&)lgtA>rHy>3K=A*6Jku!rz))S2)@}oP0nk6h>fq?M6_Vz0>T4;H^<_LOs+b`4Py`Egu)rAamh_hZcJDlLKIN zKRj8q>l&#yAhjt>`*N?3+jxxS1LQF0@V7adigN(U6@dAg^oOzF(SpjN2Y%>~NCVq< z@+=WNxK&vo07R=w2tv0Hi=5+bL50ewQKAmXvE;+y_3=mJP6u7s2nuH>fC=shL%exx z!!Bg5=*kJbR@Nq&t*2**wGs<&;ovqGA&+y`sFUa=k1$6--1WV8h*WD}tWam-nIe;k zf5DgTQitG5e!)bihZ7XoLB$8Fueeo6pkj8?0^iN;gd01)?A4N!yvlS5h{S^R~K`FSkztq=O6!sKj%{u0}x?U!-bCq zEFT}h9X7*|gIJaQun`MWku%Pg+#GsErLHt?U+6+kV?#Jek@-9pNg2&bf#&JQucyc;52Gah^}s>l2;^)2#)KQ(Pwa#xr=tL1Hn+gI zNC@HUdac2teJ&#f_hr0MHw`G|c)~$c#ULU-d}zFtk9+;$ee}qj$uIFs6Pkph+q=MG zTpITX!DI-HuHuX>;}A%KF^+`fBYBiG}We(uRaWaprRU*VI33L1pR{ou_p_TLD~2` zyjR9Ri!7lVX@LAzy>>j31m7K|A!pP?zwoFDsSgDQP4shlzhbX-wt$KvH9(OXwC7aO zLJ2!rRXx}rg8Xu-M&x;XtdUhCt47wskTra)97dLD7U7$O$E9{Y82%V%g&9C56^$DRetxrOD6!k65&=EV_${_N^KZmXjXmBzN0KzsgP7l#GOpOyD8diA;|(~Zd+vV{slG;i~qWM zTq^z(R{A7($pC{i;o2sdAh$2!FMFe1RctmHikzZ=u{*<}MWZ+uZ_T)2u#+zVu4CM>AJbw^lqy4kNK+FRO3jTfDUluQS`vwRPT>F6*GnI%p6-0*Jql ziU+S%Vgq=$O+e?V$fivu@ovITT>dGy=W(D8i}l0lAMZu_=uVhWp!2MLuqRBPXYm7l z-oCE$c1t7Y=It+rW&+y5Psz$E5}2a!O^_i+7+X!7f;>&zUfI2dmQ~tAJA7gVQ4&CC zXO+Hj1lEL+RWNXo=X_yEl~?yx6AJBOd#0H zj?mV2gc5kKHYZkTB-KcItw_3YQyzODDc8&Dn|#=*u8zF~4w}*;aJ=)?{rPFFAZdB9 zSm0xpXJ;`C^Ip*#T7^j#$yYHekOtut5Z2{@AE_ILj+_~0J8yNAhuw+*K6_0< zxlM8ygCJQSpqGM*_^_cH1cZPd=c;@Fl7XFG7hvPq#)vsmflWEee4w}SUFcqXVNU>=^;R?fg=`EOqCqkJyQxu6qn<{+-B54KNX`lPdRc+5 z_w=aUdQ8vq#Us!x?k=!K650@94dd+|c6Nou8mkSLRMaj@`ly7$F(1OjTHa^@6O1`? zcihX<9TUx*QIV_o{&dYWoF$QTFXQQVuf*$r$Ho{AN!6&#C1zcViWy-;Ej;;w1W}Va z=6WpZ@wCx(g;2DZ9ie<+s~w`Zu63J+-l6iW9N$H2OtvO=%7*amYqQRzf>bCPK{g~3 zsjH-S4Ap@R7nV&vSDDpsGS}f%vIdYd6`Mui?_|Qf4*)q6S+4NGt%dH@SQVFEFUQW4C*T5}dtgGsZxy$GBjRnZ(*kP- zdNzQ0S|I1b;E-Hk5%Ja-Q#}xZc*u?FdTo>#QV(LJd^k{S=xS8yb^26%h^iz%T@}~T@lg$8NdThOYiLCk5iIaa&sBN?duJ1d1pxjUE26qKd*_)m{@{9dq67gj z!d#{cV1r}+po#+!DPSc4*-#$G6*esNqi_U@PVnOu(FKBt&tEW6WIUt_GBAq^MXhIP zO%|qTszo~nNRlPR<8cpa=Z}5ySirqmtzSHtwGZVOv%SNalYOn@-4fY!B%Ln$Gm*HfXA9kI~*>A3K+&-X33wg_Z zUeIxFg;7kW!d|&wK!jgvg4vQ*EOR^j$4Yt)m(SUsfBX-7o5EE8f?c$Qs2mqS$ZQD!)9$D}QHg4q5Wl z;Q|op3KNF4?Ny`B-l+2qp@NQ+uJd4e=@Jq1+*gi6bKSdq%A~-P?HVK(OshUg53Srj%T=( z#=fHh*I?FKW=nV-kuBUVUbtHxH1 ztr}ZPfvt=Ayu8?&FHq`Cz_p}V4VZ}@`&OqkXh`4NA@g4O!2tjr{=l{+2T+}Y(VZQJ zC84NSln^APGxB*ynGdym$q0F!FU=CVzMI)yEzP>_z;b?`l5P|)s=EZNh4Vyx`2s$h zyB<0^R>OH@tVFDee^k1ETiO1!3W6gOmLG0{p_R7jP<((n9<%mb?CxAAUdn&n0YeW| zm94{AB8(%#D4x(`?OR`9kWMhA2Q;Ecq0P}fFoNj0qSO>-a)Ca4q&|M4Z+1S&J4C~j z#`SefF>h}@ej8#DiVke9Dds5&|a?CO3*==M&pPB~x!P!OsGAh=o7 z)&@bs-aYQvkABta8cg)9iGDcc=KWIs@Sabt5$yk;{rJOQZ`5TD2rlp+TwNiY7vRUo zvK0{VFdPNskx_wRYzk`qv^3>+;WWg$l&4Ne&bANh(-VXIq7s+0@F+3hz;XuoahCY4 zl+#2K;#Z@$j(DcfEP7%VT~}z$rkYJPn`$;a2Aif#z?rZrtWrK!g$p#RUJI*&YHBvs zY^vE*v*{7pH0{04j8$`n^UD<~nG0;Plql1c5R(>(6=MxPo7?V^^Q|FPGt1_PeWq=@ z59qK^8Abk}-JTcd8~fe$?s$m&vbbh|t?x*1d{S1NplTEM3@~=?*Vwy|iuzdaVuT55 zKE>D2EFkq*qbero4#5Nc|C1C-2Z@7YaTtb|d^8RCZF8U&1VN7i26Rbu~f}#1bp8ZK*Kd z9bqI@_;(l%g)ogbo&!)0=_?wGLQ3vsqTa>WD7~Cf=>K5&jmI6hkoGQrkWiH_?Z{lb0>OTF(-6oF<@!@dFajqD&VKBW z(QI_#=)Z0oiV7~^tVdzXnkY0;oE1?>l~wW4K_^tE0&F(>lf{a~reS1{RfgZ5o*)`+ z%@CR)G((i5n-S23Y5T4UWmd7%ark8B5Pq-n`;uTRd{IW5s_Qh{^9m!ipfGU|XV>AO z@JX3lxn2?i6=)PycSNCJOsuGH;R$-~&<&=cTgm~rC&&T%KMOop-XV$WFIebFZqys( zCidf*day%_jj#gR(Se^~vTMUoDGgEO#aaQG+b6(k-Sl`Bt6zrLXD@L#E;3eD^Jk)x zxF(20mBfL;(ph6u=E_9UHnOm5GI~It{q^me8xkREbp%=fyLE5jI5MBL5ig{RawkO* zMXyfRMU9g%+5$z?ol_miI7G)%d&pI51BAjzBZsugq48@V`DumJ;NEHDmoK9gmiPv?%`+Po4@7&z*@%Y#AyU4=nP8Zbhd=RSg7kW=S#RF`K9!%HyrjAY@hu?!+MS8A zMgXInf(*oV`O}=LB#9*nB+{?7ERDxfKBfg!)i_NPqEa6X*bJybFvo*uX-kRTqblE{ zu#I`Az>Rq$M>mW{`(*ihxQzHtGnFzx&8#T))pf_XQ@{#yKCBou#ln*q8Spyn7a;ms z*h=M@Y|vHTuTzoS*HiTT+=JbHIL_7my=T|K5+IsT9AySE!DL2LL#j-&UJdl=wSlKYIVyBZv5t#x~E_tjF zf27z4!0Q=qV-BY*z!UjjzZ zx)W3p;KoQw&Ydg;W1^pH$Yo_0_=a$2VaZZ4qp1tDfSv}tQ%cjOq-tPM7xHCR8@;Id zOo)7G3`XD=sVhA-RAuqXY(;Uh6B*Xul8H{rnWqL^C4)nx=8x4|&Bt%u1Drq}sp+3q zdv$FsH2G=r)8wbgPm`Y}zZm&FRuK2}0ZQ33 z{4hj(`JpTWOY(D*nL5cEDY+oQAy>ee=`XIyQk8$4iaf(nZtm{=$ZpVQ6KdStJR_t(_;hHA(AB4i)_ zwruw5{in}N0S0IfD2FwhW3@Ac0Np^EO$Vv!K78_}o!+t+@hE<&FsFzzbF zH~nJD?0}y{n!)cV+afbfm~7exd*J-ePu#M7bMl1u7)x$* zKHg5iGp#U@B~?6VpQFC`S!fM?g9(dMYcyR2cu;&kvBHoEqt*=f4RTazz@G2fQ6zLe zgjdrd#~5sio!BwEy?2mGCCyxwN}}m52(?!x1KX3GDRO6wIlF9RUf*%wB18;e)R@lT zA25GKqxcpS|J7lQMBl!Rh)hAt3IK8E+^qGViafwZvt~5tI1h&>ufq7>7Z$(+3MRIR zaiYs$2QLND5v)KwA-xhx7VuYbDOCdLY#iL_pgpEbKmD#=M7&k_I_o`X%7FPRTXtBK z%+Yi3+fxP*s5jcd-tz;tNTSx6G9WlnW6-v4ElkizmRgvYSi#fBsSh05-C)tzu{#ga zdj-Zy>*AZi2%|i7d9Q)G*C4!4-Yx$A4PpBSz}kL@wMJcI?S`}vcyfHlD!kThk-IHq zChGcFk$t`p!BYBrDM0}th3yxX8NfhdRc#ga0}t-PM&GtF3Ce&PxpPm$%@0y)v0*Ej zbTg|(Oq8<&=;{j>2$Dhs3+%W>x#W4w^m0{C1(qsLKZ-(-xe3BjkD2yie&4Z5eH5GD zhkIUPIDddE4|pm7>>$Z~Fk(-Mm~Srq_!P=GH8C-3iGKYk&rfXC;rt4l9GMJWN#y;> zgufcY=b1BcDo{OOO;}&KV4Q)0{vUx6}REDsdJ9E+kMc$q(%x~`qZE-xQ_$WAFLlLdu` z6`n$vg+lV}vv)F#m)#`ZtXG?jH9C0f+}ff+!Whn2_V(J{SiTJR!{Jsrv3sLQ?@RS~kSq{&gc)T22u~yvz zkSApa+v;Vv>54f%D<_>VKOc4%HLwWkgj_iwo*_5ye|BH8>EwDBnZ8f(d zU083As#{?iGyIb(xDA@^`KZV|X54rLSQJ2RKTQW4Fnd^v?kN+`ti<-J`1yPuga#{R zIc}1&sWedBjKZMFmK;9y)EiLD{ZyLM(8&5|GaB^4(fSLQ0G6)8Hb)Px7fex1)MBfs zdL;yE@u}1VWg|E0-Spnvuo&_YxXD{rq;;j)siTV;x*Xa50TDue3NG@NFSxIHw#b_u z_xz#f4ZNv6z(bCY0ay49*2y`6Ei~by!quc=qjJDPGK&VRMg&-f2CNJxW)NOWRJyS+ za_c$!^N;@l^17q5tJoHHg%Ne;!UhtTnaMf|(MmE5$OKReYh02r;z8S8+)h`0RY-rI zX2|xh)bBsi??uH`_UzNEj~{PXB*D@jCU&P({I4D|Eqg{sc+5t@uWx84@$IY82!Bjl zP>_z*-?zo@@h5-BT2hlGzVpW)f1naNTsLhi{1G*c;qc?5-};1{@EjEd z-M$hlH9q&>I2*a)Tj6v!N5n(7e%8C=l`XEuZH?PS;PyL2o^Wqip`By|1kFktw~!!j z75H}_{tXfom~Y zCH7}cqWY$i@jV)L9d?!^!k?p zdP2N3-f6tkcvmpq$u}t_)EP$h$bDRvOMtrVf}|l(L!gGhq78f72VP2OY;?1??8MbM zRqUorp`-puVSxpDrJZ9oF|2SKn#LMIqy!`ubDkz#g^PSCLxWgh;sT)c+t)6spfIgB zD@MCcJ=&zR(3iFHRf+rxyW{g9R;WR=YLJB;;?D0+pI)KH1YBfboAWtETSsot+>!Sh z)XS6vB?Y7+qw!m`=hp|E-HTH`2+LP;CtzJ^^Gq6Q2E@rLk80T9nO-9^ahhP$MK zJm5D>tz8YcH^b0S&i^5jn?4D3Lfl<(T*1Rcd4ha!AD;snxG0KVct+DcklxA}mESsL zP<}aqy_jgJ19C#XWZE<2f%TZL(PFl+g%+u}FHH-Y7D_=2D*7s#ix4H?8zV%rJwvgK zqC>H|X`Zi0(Nbcux<_?@o!cpZsV;>cT+d(l!RCZ1tQ%i&UJ##mLUq*~jXLG>93%njB(;4>ai#h^+j zuRNg!f#L{2-lUQz-yHSNjcB1w2ot)9%~np6f%v4La>m`S3EmHW|6 ze6A|Sfi&2F-f%66Wn$m5TDM7cNLhJwth~rR5v?oZJYd*rb8om0K}7+6z1Gf%_qfy= z?6ZRXDdPM!Dr!_LD=JbmE;TM1Wx0Sc`I@1FxRJ*}+&HO&_}kY;J<%?Dk2?29uf%NH zBX{cBgYhB&fWgDpK=dT!o-A&6kebYSAhvq|h&3XJ?Q4i#?+~S`94o5=77t$Q2SpG# zGw)Qbr4LeYRFx&!E99jRG4_U1jFx`DXQI;6EqCCAvGC)y@>7)wAzhhQK=8@*X2dNT z@(D_E*BV)g*733tN%RJ?q-bn-ZxL8T;bo80^VAw*@krOX^>J44E@CRv@cq3iM3p`( z;;nsoqRL${IcDoDwqkOBpQ8J#?cVG4JYA^bi+&*}ULfQiWgpu;zN=GXu*Tps!{80C zTI7y87Yv7W1M($O86~tIZ)x|m%`*r;0m&G^D=%QKEqD%g;^qNumdTptXva5*Eek#)PVP$Xeiui9u2kepUVnb_0Vj>3)^61W# zp<{uE>NH%;{JQyYF?|XOgGD9Id10}Ux9wB?ynyjhQ$f^v9J?2-3S42pdT8*BfE&jx zZ4U$Z`G^bj);Owh^fGWX`7ZdXCqvPcwj@g5!hL5uPd_ABjxuAbu)W5D<;aRZK_rzi zUP)LJa5cccy}8GOD7+v-qbePp#2Ybph{`Er!W7!b=P|qic?F^vG7v#{ODrtTopES6 zIihiRWO_r_8NihwRZ?XK9-o2Mjt^oMeg`ihrYvv+)2^-dOD|7o;I z%K_Bcv*;-|VdpiBX(TwPHyhQap&HTqiy3{|Z0Cxb=p2@B+L~`Rv|knBX5YTTf`yV4 zX4|ajYHIu$2`+n0Ld`y)P^-wZd?=k5LcxYuB>~$(n!o^%XMi7Pu*VwBc20wA{0+|u zv;C0^3TS#`CUh16aW=C8v^4oI*cje#3}a_#MncJf>6>__n8S^D4wUFFU~2Iu0-X{_ zK3mv$DCkiE5>^@~dt2(7G9LlsC1LhO;j1qW=c6ncBcJy5}!a#Mtj7V(cA;ImARP zI}+l<p@p8|?y=U@G6Yd|(*mqEbIvi8%guAxIq= z_VNUT4Tl5RyTU;T?G7_s&*i|zU~Ulm77lx$2GMzgH%MO?o@aJse zx&i){HGnbU;07nh@6xY%;5NJp55twvqayJ~tqgU{gm2lc=RQzu{Ww%`@;$(M#yasN zJS^W;Sn6{t?AKA?4$)l$Gkof;ZpVngzoWr_&FdF+`s4np-ckb6b`~I?UI2$M){5vL z+$CF)p4LEIx7an3k_+5ml1*YUhwF3hmW39xP>Y$+VkR}1p$0RB&qZ-Y=K!41%4tn* z6=_XgnFutGJs#75zFV~C%=U^#ixCQ}9HQ0-)3Yseri`zGK*k98v}hkz?H>u=pgf2b z=QAeK>kMeM2S!Tl;C=vVfC zH9`Rd&2>2T6b2PXZ1?#H1b7To&`mX$U~mJAl!qlThzVWFM+RxnFA(?JjRV+D>h(?0 zY};DFhSOde0zm(r>qzwf`N!8k{lI>N{U~rf<;_qcCKqwyA6kKp+okZXr>2BEMRwBL1EB{%@ z+7fa7M;BN0wt`+r+GbM*DQNUhE^eNFNQT=zBkjL=O#Rc$NWbt0b!gqp3^4`HP7N8b zs3?{&iVSsbbn?bJM~hreZg}_lB|~YlIh)x03;Eznw*b#7f$1p)GY z%mZNx#{&tTN6uyV5I0?>vrHq9<8sAgb5mBj=(M&Qo>GG>6EafH^Tmg(VxkWdnvOIb zX*#+rI$9rye1o#`5Ujy2JQQ2HGKY}ns8+?{r?#DJ)?bA05V@sj(CYO+tVQOhW4Cam>P!ls}&R z(XFd%+dlkvIl_K3{vcR*LmD-D*3H?BA2cHJLu{wpYXS!@eHqjvpRVeYVj*Fyu!?mX zK)BFWN3E!Q*ai>(1`nql#e?aME3jd>bpwl@EF4!h93`m4!icuG`MdQLi-6sFh=9Gi zM!=K^m1y}B8J^8bQ-vV{o_$2gt$sww6?Lcs42Sgs2Qxi~OL1BfL1tnd0bYQ0;bVvL zH!+Qq3m2$=N6DUemwO74mnLk2PAZ_yDFcV17pa7;%<5DHNy1Ghe)xua!(~PKRmTX1 z;?}P`KseV<%_oYz3uqyeS_sb_O&!;EZ!zexYQZ_qr~H#>hIRS}-O}Co^5dL7N@xz-B zl;nxQe|-F$eJA_

+TD7ySA=_C5O!R^Ja;tyYtH&dcw%=6^2kF#q2&2P1eF3cDY> z@FoPo;kx0sFVXvT=Qht0LhP1EF!no8%!z?T7@UcFkBf`_OoY>`FB41 z_l3Pg6luc@RNv!!uy>zds(D73^nJWTjUjx+r(g0{m@z9;$*7ROCCM&YQCV`IDstm! z_{j7h>E*B`2D4z=>W8$|(4`F4cLpNW$@;{ z?A5TM%1D&oz5;%Snj=|F0aO}Wg*qVKCS_-zg^TbBRd@moV_Q#Addfg3)gaQfEVR@x zTE$0kwtDrvH3?%Ad|a+d7|TD6+(6kUb|A6%nN*AiiJ8Zu{~~Ezj)?*)Yh(< zZ&&8)wIb%znP#acwiPCblN;7qq;q^^?MIIjDUpH{B1x%Df|45j^}3N2_Pc3wfBnb3 z@V6Bb-d|iQbLp`6nA4JW`*wImt zu_Fnl-A41F=0nYgnh!M}?#G8}q;jC=$BgZ~%-AkPfH4Kk#%mxcZI4x^k{tJf8XkWU zVv9cKiG|C*Qd)$68QBY8-kts!Q(HD!E5sM5YYGQ1ndwEjQ#KaOl5O_^D+tf#w!4IX z%w_xZ1V_l$bt3#wlqYS4zYH5+i@V!$tx~eC< zx^wnsjx@rd={YFPM5TRPYd}sMZW+Nu|@{UCjAB+`UjY zNzPsgvBeDx0Ex-sO7Z6(|BR^x{MAhDbS~#t-6^R6GTU-(@Of*{#$Ktwm=GzPMYc9? zG&j25MvK`()GtaiJ;q+}l4xely+^j>s6@ksQ_%|ND6XOU+;pX=Rvrs71ev@kj(;9K zYMv}x=ClOWi4l~J6v<3bAuY*+DV2PSf!^JIq0ywntXokFRpRWvnqc4Maa`&}$E2{Z4-(^8sy zP;*jAIH?=sq^{B^h!fCa4BAizH z3J7P_I0f4tnKpl+)bofnPP0%+SSaew)aYo_UFMU~jGQhhG_+aKw-b`lEu4_H45c7y ziVeC*YU&hjrh|d*LQV1aJCiK`6ZGxd5?io@r5;Qy&kqJu7nVT^u*^=*B*;5tFrh2cC9q;xNKBH z8Ua$p=O8Syq2*LmIc7FsGu48Ly5k6!5Yf_<#d8_7CphSbN^ z03M}0&%WELF5lhchvfCWx>R23BWo{Xonph^qBzpJKZw%b{u2%E8r;hW?$Oi3$Cd`v z6MaiE!nRR**p`KHF23dV%<|parL!E&siij67z^u$_=jn_E$i~gSNL@@!OqASO^koO z7vaEfWl$shXy#%y6`zOdi9dY+9dJl#Z!J%jWm#aKx4Fzs51WpDHvK!1Dv#HFHb-~V z=HX}g;wRtzH$sAF=$es_x1us_-^I!{_ZA7!x37(;9(`7cqMGs)ZXDG##WD$4ImeZi z5$011hy*P@bHGtQA&4?9*yx85ix61S7X1xh?G&PA&gS3u1HpR zF3K>{XR_GxzjW&L8I?+j?w9bfG#k~X0q=~dGf7idpZ!a-Rj10fp`DDy?PND6r!6C* zE$t~!^OV2Aj4hP%F`xr=P*j)>$IS`0)+POs%@c87ANJji_uTc(`?T=lR@7y2RZc5x zhz09}f}!MvHIUKANDn2PC#A(lsK$-;&G&qy!+YY5VmFUAl{qtAVg#%5y0#S0-;Z9p zMRr_iy()sE0%ln-WyPMi=@2d8^(Zm)^bz|6E9NK-7`?>;@ z_Yv%U0HF{LSJmNY4x;sie315%bs`ri6^kD(qx@vi_sEl8F8sWSSrM^ z_wXVBRDa>e)Tk(B3jlUzr0YpF5pG@ zAMT7zkI@XS8Vru;23?4;55y1$rsXNDJctYQ)Cd1wgo(N~QQGDaYtBSFGAnU(pJ8tg zY^oTBi?~e6J{1iD;bBsEd=?Y@)GF!XI{AzErQ4pE*E!eTo*%p4M(%(Gg7Q=dI{tnx z_Ih}?0c=-cYcvY0+$Z;YpD3jojtC$5y#@S6_0e$i-| zxr<=5DdKSp!el0@2yH^7DVDB6Z7e#M2?Qc9VfEXkBdpzx%K12F?Q}pA4D*kxPq{r< zj6@tYv-YJb^fa)-Q`P_nco1Srpn_6Cd^;k@)j%yU>w?>^mpk)>NePU9;Mi%{D zc;%nNTF=~jw3>z9A?yZ+NF_>95?}ZT)(cE-op6jerj!`ara3k&J^FD^IgbW3c0L}c zkyz-WkYgQ$V?FuF&o~6VHR_t*cHp-H<(^3K9saDtUczkK_`+R`_a6hQju%)X6;;Bb za2-td0_;F{F#bDnXEf+NOkIXyGGmp@6MqtMmE%%lM7Vy))mTOSaykCFxfElEOA{zi zW4~fIb;K>Q8(L{$L?ZbOU@^|K5@a^iKq?eL%6FV=6keMm-)U}rNsd}HDSXFU+SPdc z7OQ8t>l>F?5LU5Ay^mQQxpwN4=o9Bw+Aku-FdxJoq+ld=kg5v>W%Phn{C#1fHH0ug&A=0AM~ZVE1D^kF>#5k#|UIidsr9sJl6Z&Bht-RZ zWqvg3F_Ot*j*9T@Yp0bs#o_rZBO+C>G4W^`6-UF9*n{bf#ThLc2l>|m)4LV%Q=^5* z)n2(P*@9pM^0{;n1gVdyDxPiIPb;VkfZDZA)Cxpq*8%sb4815sHjP~~L0_&I29^WH ziwJFq)T#KhUZ=G^o>uL`6Lgv)PZw$;Rmq*PJQ{1JT1PK&cFH!AW-+3j9p&Uw!Brm zvV7KR-9TSZmDq%%S1|U_*d6^E9*hEOWX(+{aw!V4qAZQA+>2vQyGi&Pu8(D`U=dK* zVG2Js#yLy8Lok`XtRj8}B^Fmoz^gu<63S6H2<5_p>0OCL#k%am+h(P)SFmuUDf$bRcOl;J=Y zoyW|_fr6Sthz8wUHk5$-_$uo%eEuADZZvgWI~+=aJ6-SAbmUeH@gHQ^6t>DYu5B-g z8sVPrXXdTth%MSr*3hG15|C?r(a+u@JtcnwlH~0Y9D4=%UWYs{fykfM2UooFt!&1^ zw=?eD!Z4F%+cqVhY|g?WCHX#{n8K2bBlEMwfh2awkSDB5&&^WhDOJ@@P$Y!Dm8t-1 zp_Hes3lLYuRx;wmi2P!P_KR(1pM3hY;Y=14rd$h*tS7u<>q%rWeTtwTIaE)4GPF8Gd6kk@&D%Z&>EzK%Bu*xP)+Q1((lgh}L z_n=#JZZ?@Li-)8ki+nYrB41daWdVs))f{yl3aLQX6Lz!1=O|ZuWHgi| zX(-4;G`srU^=vq@n9d!uMG9*Lu4u-GppH~yHZozjXd|RC`~^D+C_WotPV8!<+|Z5@$%Q8XO25uVso!K=(aB(-Hdzh)bscRE)h;Q>M!7Nm{CBc zz=ao%&Z1G8_qI&#G|0?Ov+6F}wg-pe)|Td0j944;sd;a~l$rKg%gDD?%ei+gGP#HC zshZ4lA6VmN2P-4$l@%xGR1CYUlI`|S1=Xf-veA8B7torJe_$XkjpQFTXU}Je;&qQ8Lrh3)bGw;E8~pD zNZC2m3i!;2=~&oktqkYfhv~qtXME<0i$EjA#hqL>$8J&l9KKYNncE>VHybwRE7otT)E&}m{iIWaZPj+vD3Tplwu@|~e*?Jl38 zFfc$f)Ri$*|HKTXgibTm$r-BMJ~xI^!gF~H)y@~IL2GyU47H_5qh_cJWvJq+NAAc_ zO6W90o&F5fI=e%ZlAa4ZM78psqFUE*jB3`a&4wnaD z7f4c_yd>4R3X)2#pQuUdvPsG~F-a+*(5PN$MjB6H*kP zg_WVP+@?dt{F1a~=;*#w+A=h4oj$tH(9wO_iyv-L`seH;6{;pB`1aMPU@-D~2MfD{ zM15hOxIJMnv3!%C1Pe?z_NOlB-DIWAaBoA5-_ggRBmFKU((mxJbp^41nzl+1`)BCb zKTTUj)7IsN0czU1`hY-9TO|kxG;}~*t|7`$^H&L?3=JJ+sQIgK{<_?N!^_~WpIr}T2&Tbass`Xxsd7~~R+y}*0}oS~ ztf`so^nr&>9e9|z>*TunBW;LC^(|ZhQVc)GE>j>z@Grn(kIQb3mIs1jVwXs4zzC8i zifgNeC!6XEo?<9E{P05JIF3nUR}hG(X{-c+h)o@csA=plG^WE5HH}?CV4|k65(FkT zbzq{VvBS{VtDMD zZ{jD}bE-&+Qd4TOn`%+T4=`eY#IaL0*4+A7d|x@=BSGCgVx1rjAU@?%QPdCmXH$@t-1~lG z^^d++^I}&7`=T~)*n9z+b-=+0nzg{eXv4t>n!Ns1kGN8 zW-v5%m}VXDG=gR=@HE=+G+qVG%&W(lzQ#tg<~ya$&SI?#&PHIZ1_2nqyx% z)-*Oa*0t+9o%Mxu!Q}|7wZP?Q!{vAtShKGlYZ{v!>$)A=>D>1xiCB6oPuXJMQp-=H zPyXg!HT%R{eH0`{(UvW<2Z(cEeJn+s1H@h4`sl#=SQT*(G~oM{4D+8q{O7lv{;4F0 zFuJ4vm8vF1bw`@pzFS0M>JgphM44H(LokUT9eeFlqtA#%b1w+_ zqgf>4NFDYsTl)Kn_oQ9HmBUJ@%l;!TP09|~DnZH?*eV^^DnZI?BqeN`AZ1r9n;>Ni zESnB2n;_*ik`fkBkg_XwP>`|(c2Ea)P_0O5cLtUE+Dq-4_EH^@CAY z-I{B9F4pwAWpV_aT3~W?VRF1MolLvElP0Z=gmpJ}wlRI`mY)!SYJs27g`e=kfU;XW zf6{q7%eoUNyPY$rpC~;1hcNKyR^JM5rjFx<3QoJvL^PjM57|60ybs~oD+SaLfV#YA zZwSv`RX`m|!S|onETs=etP&mtVYrJn9|=D6U&>m(*#?PVZ}?729JjyIkI8*u%jVMQ z)demEem~8ge7+RfI^g#M*;?TD8^Z6`YO@1uPn6(a2vrYSqiri;C6Yd zO1E@KJVK?fy5Z zB%iHBf8>_Bj`)UJDU?}rWRH*5UN2min(PyDDL8h(Z3#HGz->8#+w%J1cvC(-)Q67x z8aHohP9kqGhwPYd6PRp)Z*v6SrUsZCXg-P5*8)mSR}Yf|%qF|#00kyn-~b)L0jj}L ztLBqQ?O*L(`T2GrvQs`%AhHEM(h+>5s7AINks3QvQH_KT6>w~;4|N0|D&W|5IO^+Y zy>2+ZUy>?|4JjF^U+CWkaVUjL(1E@CLk=2!cvee6X!JqC%bQk>K3uC+QCRHwZ^d`A zw{y}5%XKK}pkyGL@2LXs3LSczV1A!VDl5NwZi2oW6y;=cFDRu1>nQ)Qz>lH0DP1qU z>_x$Ux+@*lG}zc(Wm^X;%->;OyLo8mygw+IgX-uSpY7gB24{YlRnc4{9)D z0&3@l8f=vvhFS+Klz>_rJe2!TgNYJQJ1^8=hioU*eB&mwV3AbJnQws#;Js^7~`bK#(tEG67|PYds3VRx~9 zO@D}NF~AH~Nx-ZXR!IX^$=1MZbPX`md8^D(U6_pkGZ-TQv$hx`4HzR^12fCM0+?yM zp(QmyO_*6WUG@Df%<$&YEo#8*b?1Sb0&_Pku3jd~Le8Q5!-<`Z@~NHaWf4M;Pb zaaEoV$@ZvmQqT$G?DA8tbFl=-P zt+CI=q)^q9=rp?5DA6a(Cy;13Od61A_)MxvwADM})_<}@>+9aI>iX)l=)%8o35jMY zJ0;K2N=BZ8>?iAx6oG!~O@c5;?#J7Y6q%Sl=U3mt(b2qU^EJcGQTg@K6s@=7?=oY> zhh4^s|Hv}-Ep?5=U&-$EUw;+58vJf66HjhM^;nXOXF;6Di?09RdgcAaiBeqZ_#2;{a(sn8 zd+10KS27XQwWU|VCOa@q2f_o<_u@?R_GF$;lu03h&PyX;s=MduhyC_NIhRMbu`AwfjXqNbz;$7qs8cM6^v}>H_lm)0?{C zoJ>DI2DacfN|t3HD{JOwbBTK7)GzQx_RWe+q6WDl9-Sy?UzA65W@=^m#Q}hHK_HDD z0g}q&bT$@`S#Q#28z&38liP**zfN1=G-5azebDLobYk&%KAn;UPb;c0>f5dC0KD?> z5K-!2-$=PD#AqPWG(3zOaMoU&QL9t;i<652&B!w+{X?yp(H?;c$e|naRRXgnuY}S! z5wHD~ToRz~%LCjC3XTf7{2ooYpUs|c>4^U4M1&N~P;6Jk12NrhxJ}aj*GqZB%ZT#E z2tDeA@8n&jC^^Ec!hJrSF=c3=hNid$>wvT8oNzFxTg-z^Orr>e;3$m|(?oSW)By z=g=fdl1Qy6ErZ!ClzCPhyA^Is{Qk@T2uDGL*;43d{E39gb9&SR#-tspIvkEj0Htht zAvf8D+~g^|-)maS2&XC|obEEBh9-Z%y5(+0pD+3G(z|FgDz*kUpDA zP|EQEV$u}EJ)=ykbjqROO|T#PtKv!vCpXip+1XH zKOE46(VZVB`xESa0gmX$VoqWuSb1VWgOKHRx;U-LYwx}}F+My^dSx_A;8`;LdvY&p z7i%w=T}F0ukL&|!-S7l@Oi_LyuN=}ceckwcYBj{GBeLF%pOB}dlvAnk+*2L!n9$M` zC)sXR>MIn3O%PKejfq(A_);r49;vcgMdhRIuyPtpPiTA3Te0RU{mcf)%-~;uJB^Qlp@|)cG{whCr z%wlWwSr2%=+T&dhsMm0`-iQna`E6z+^i(h?1xW6adgJXuxv-}(UqU^kxgM9d%E8BSHCteULF3L+MUEvr@y-a2LohV4zjr>#wQf4eQ(s7`q z8I%rt9*&N(yUgHI2*sXqG`5D5U*<8c|8dz<%%R@Lgk1WQ}osRiV z$I*OTxXZI}!d9FB51zvVZ%xhlS&kCI`;ry;eYM^n-+LRsZie?HRs7H1avk~&inqG3 z@*cQd^QKP4E6Nn-1xEN%L1fQJ6{N}K_Xmxq(?W}k@Hn|&D@73AY#SzVg=LtTKc^W? zcKbwyIDSx}Ie`tUiqACt?G6X^H9GACJrd-#?-&7p^<4KK&pG#f-4XoF>2(zXMt%am%rcFAJl(nz6aw!sOaR!sOg!;X?`O#qR#elQj=rDaz2L7a-FT@d4udxwIz_A+vozU+g@o% z!F;JV`)y{JhGZpT@BW!>(HY$3lG}j%5JwW zW;bcI%peCYW)){9uCR(;aVeYRb5qXPE`4`OyOB;C%h77;WBx;L#2U$gt%+o*0^n=q zqnbM%PtT(qv?B30+g>!BP>#S}Cc$g}4jYF62DLQetDlfh4_c8zOX%)S7UXFSyJ)uJ zyotEed?X!3ieO?NwKrSOI+W6h;+K?47CQ;r{SvQ+4bdx(uh{z(p?j!sx7 znL$`YE*|_GGA<2UXIH4ogUbO20KXW1-7PYB5vZ#!jA)sd7#acB*ykT^}s`dno&0iwmOB1{G zzUgjOmhO)GdA&`sL0UA0R!W?`+Us+ka<%-KNqjZg*S%kRhMG&RYIaGDsvhpYAoGj^LH3wMQwn_E z_DPlg=LCQpz&r7eNWE^V;3zH2+Tm;0(yCDC5(*YxBf=$44K$yDqhvBA!2#yuC=$Wn zYMjvB=HYHpe++>l$^Xc%4!KATOnECNg#qleJ9!YQr~RNT{88V#j6t7cmpcT}B<0W4 z=^qdrigauTL*#;~22bZ{9|7dS^5%!S;K!r1JWP)Qs6jg3yBI#^POeM=$j%6SL=phI zf?#}-7o&Veyh#KaeO$8Elw8>*kdJu&{r6V_O}JA!!)cTT-Y_YEZ=>wm%bhbd$=u9S zVoY5QpK2KDZzMuFiNMdL8m{Kc=Ckr%fiZs&x6!5S&1*kKn*!r;4dn4U+J3wm@?R6! z74TU|)dbFnl`>L%F;0%lizG1qE!>^Yb)9 zRlf#j$>0%1sAAlT;oJJ;Wby@@r9jC%hT{L3+*+$8tbNr(%!Id>H;bEt|3iSadAksF z*?tMg_4!W}qW^WlaR6B~-is46MjQ4kg0pSXV(Cukq8cDKW}8Gnlk5mHzHeeL z{>zbk%t2pLc;5mZqZ!x7E&}9pu>afOQ{-1|<|?fdV>PGc&bH^Govn)QKTd%rxe`}f zq(_D$0oa{*Z_QLvQ4cFMFo{U(%lii){{nJ7tgA@q|GZpIEYQ3jQ`~F1DK^hom89cU z$*}Y9YA*Bo(7tvBl(VBi0i2C#<<$vi*{On;{kr7JYH;d zfMdH7x!y;hdJ8fR^kYnVrDXuKfV)&zaJb7aF1pf1TBX&*bBqwaR~SvZkZuA#T$jHL{J3@mSeR9B$_>NASxZWHR}$ zNgT)Tu^9KxCuNuXbuRg*yLFgMC#QrrSB0-aMf$g=*Y~a8nXm&zt3e%4s3$7?M0eqi z7R#(=G-h?W2ev1-NLvkSUGpZj7QEU!n5t)Y;mQ z>j`nUy0guZohk2A@m?G7Uxo^Yx9Qi|&Kdgc8avJ&|DJaS_Or|hyz||4J-{D0n_pJ% z4Fvn3cvuW%IOL9Ab#o}%9wpd2OeDz0cpSygZ_@9JGuxF8blLd|rUyzDO$6cN$2wjG zS8vZ<)}b){NtoD^YB$etvZZ|vUnr0KOM^?E&6rZTX8d-l7kA1Kd7-|i7z(DU!x#P8DgIqnL>dL)ac%?w%(9@JsfXchzOSzUp_`YH>Z`fia= zQ(mp4*u^q7>dADO5YSq-Uo)xaa;px!o*II;*TAcM%nlWD!0(nU~Y>dXDKazc~|jbz-*iu1DBO47Z>rLOtw5qc$WN+J<_K6fptJNZo%m5d?7qku+6P1AGZ z{mr5YkBmpR1=}W#XVB0ERWnb2V4Q&pY!xx8@RhDM&PUS~V5IHz(F1$AI7vk}) zE0Q@q@igb-Ss?k^N!>J6p_up;qwAd-IHAH@4Kp@I!|NX^bc47)aBkcW^|AC<*#7-E z;UzTZrefx-5RScZZBzBTbTCVSEigA5sC(2omU~57Or*wFPs0&(I5NXH6Uveyd(ED6 z+9t;v*BFL|=$Y@C)fc`w+bN^QyZ^XT_+nLLSj6wFKM;WD$syWWO84WLRga2Parp-! zktH)7XA1tLI3@hJiV@$$kmXy7xV^W$>yPXQO%m|#0s|fG#aH&XmU=>Tgo~!3iw4nR zSxTM*gI2v3?Z&=3;c25~dA_7(I`1~aVK8SFRTxUIrZ*AR(@p0l5=V24`*+qb? z(k;yB_2i;cGQUCGn{~6hlFzEzlhjV3fq(!W3y8<%CIJ;tT2!PQEC}=Ph(mY_e130X zF}cHKWDb&ke(&JTa|)dKN9EDkm1f-_!7ca00v7iR9cqOoVSl1B$lZToHO=Ea258vI z89_Kya*kb zxtE(7r!4ieT6l$Hz)Lx#ccwOb%rs1;HT`;>+70d~64GbqOmTOUUV;PADL&^b^Lf?>zWm(+v=!)@mjQ7dz6-$N-Uwh zhK@98+kaV58o%xO1%x;x!tY8ijAiL;f3mK7hnuSA^-$$U;zCAngKxIf7m?Bq9OKM5 zpPcV`VfL&Q;b1QV=IqwxnMS+WrW4L1;l!*H%ZGHVw(hmyuj<0LyeButfjgenQ1d8h zBKy}4Va`k@jk*lC+lzs~3H+;EsfS1EHnIIQg3{uxo$V5o8r+p?k2DOkyL^A2H&8rWqL`mPcAbc?n z;o%gFC0QEY<+1xVuE;aWT&MW~XXPah`p})J+ixZK2rK^*c63%$?zZLZ8vBX=0kLn0 AQUCw| literal 0 HcmV?d00001 diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/manifest.jsonl b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/manifest.jsonl new file mode 100644 index 000000000..540cbfd80 --- /dev/null +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/manifest.jsonl @@ -0,0 +1,5 @@ +{"access": "local copy", "bytes": 4657025, "date": "2026-10-28", "doc_type": "pdf", "note": "Historical 28 October download supplied by the coordinator as a local download; hashed from that copy. Retained under changed-primary rule 7, not a blocked-fetch fallback. pdftotext -layout diff finds only wording removed from paragraph 6.207 heading; no numerical changes. No rows cite this version.", "publisher": "HM Treasury", "retrieved": "2026-10-29", "sha256": "bed70546c026487e3141799968f6df08210c182b0695fed0e8adf7efd6697e03", "title": "Budget 2026 (HC 726), 28 October version", "url": "https://assets.publishing.service.gov.uk/media/6ae20b85460c1346f7be784e/E03668986_HMT_Budget_2026_ELAY__2_.pdf"} +{"access": "direct", "bytes": 4679545, "date": "2026-10-28", "doc_type": "pdf", "note": "Fresh HTTP 200 curl -sSL fetch with specified Safari User-Agent at approximately 2026-10-29T15:41Z. Hash matches supplied 29 October download. All red-book rows cite this version. Read with pdftotext -layout; compared against historical 28 October version.", "publisher": "HM Treasury", "retrieved": "2026-10-29", "sha256": "19b337aced8fe47de1d6ee462dabf75155110d347c38123b6380d90f09e97b93", "title": "Budget 2026 (HC 726), current version", "url": "https://assets.publishing.service.gov.uk/media/6ae20b85460c1346f7be784e/E03668986_HMT_Budget_2026_ELAY__2_.pdf"} +{"access": "direct", "bytes": 197606, "date": "2026-10-28", "doc_type": "pdf", "note": "Fresh HTTP 200 curl fetch with assignment browser User-Agent at 2026-10-29T15:41:44Z; identical to the supplied 29 October copy. Read with pdftotext -layout.", "publisher": "HM Treasury", "retrieved": "2026-10-29", "sha256": "36cd5afb4fceca2cecba0d065974ffe8280888e15ed852c129339439e539469c", "title": "Impact on households: distributional analysis to accompany Budget 2026", "url": "https://assets.publishing.service.gov.uk/media/6ae1850f16428df53760627d/Impact_on_households.pdf"} +{"access": "govuk_api", "bytes": 24883, "date": "2026-10-28", "doc_type": "govuk_api_json", "note": "Fresh HTTP 200 curl fetch with assignment browser User-Agent at 2026-10-29T15:41:44Z. Hash is API response bytes. Matches supplied 28 October API response, with full 19,727-character details.body; all explainer rows cite this freshly fetched version.", "publisher": "HM Treasury", "retrieved": "2026-10-29", "sha256": "cc3ec22eb55f4c827c301bc67b06aa7482e4970ef7e1cec644bc382dac0dbb25", "title": "Pension Tax Allowance Explainer", "url": "https://www.gov.uk/api/content/government/publications/pension-tax-allowance-explainer/pension-tax-allowance-explainer"} +{"access": "local copy", "bytes": 9816, "date": "2026-10-28", "doc_type": "govuk_api_json", "note": "Distinct supplied 29 October API response supplied by the coordinator as a local download. details.body is a 353-character introduction, omitting the full explanation, counts, examples and FAQ found in the fresh/28 October response. Not a blocked-fetch fallback; retained under changed-primary rule 7. No rows cite this abbreviated version.", "publisher": "HM Treasury", "retrieved": "2026-10-29", "sha256": "791d2573bafc67fcfd89567dab55c0239dd26febc9feec44d10f2be4ae21a9e4", "title": "Pension Tax Allowance Explainer", "url": "https://www.gov.uk/api/content/government/publications/pension-tax-allowance-explainer/pension-tax-allowance-explainer"} From 76931102d732eb3c8ec4419938277f544bb32add Mon Sep 17 00:00:00 2001 From: Claude Date: Tue, 3 Nov 2026 13:10:41 +0000 Subject: [PATCH 07/52] Land the IFS family with the addendum's rule 8 and 9 checks (106 to 87 rows) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The first IFS lane predates README rules 8 and 9. Applying the addendum's sections 3 and 4 drops 19 rows that are design, not estimates (rates, payment amounts, thresholds, government targets, scenario inputs, financing and the DIP funding allocation), and re-keys eight. IFS's three cost-of- living giveaways are exactly Table 6.1 lines 1 to 3 (5,040 against its £5bn), so they take the registered package; the full-period VAT row is line 2; the self-assessment timing and compliance-staff figures match Table 6.3 items, so they are line 45 components; and the two remaining proposals use the names rule 2 gives their scopes. Sections 1 and 2 could not run here. The browser-read pages are listed as pending for the coordinator, and the seven Flourish data items are listed as unreached: this session's network policy refused the Flourish hosts and ifs.org.uk, so no primary was re-opened and no spot check was possible. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_01LwQq6HuQUQ1obM9JvyhJbE --- .../harvest-uk-ab2026-2026-10-29/COUNTS.json | 14 +- .../PROPOSED_KEYS.md | 5 +- .../uk_ifs_ab2026/NOTES.md | 526 ++++++++++++++++++ .../uk_ifs_ab2026/claims_staged.jsonl.gz | Bin 0 -> 9694 bytes .../uk_ifs_ab2026/manifest.jsonl | 5 + 5 files changed, 545 insertions(+), 5 deletions(-) create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/NOTES.md create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/claims_staged.jsonl.gz create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/manifest.jsonl diff --git a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json index 72bdcdffd..ca4c26489 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json +++ b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json @@ -15,10 +15,18 @@ "hm_treasury" ] }, + "uk_ifs_ab2026": { + "rows": 87, + "primaries": 5, + "claims_sha256": "2e8b2dc5a85bb66fb11431f5c20565a40b2cf546e7edaabedd3a6c56ddb68d49", + "sources": [ + "ifs" + ] + }, "_totals": { - "rows": 1194, - "restated": 169, - "families": 2, + "rows": 1281, + "restated": 220, + "families": 3, "pinned_on": "2026-10-29" } } diff --git a/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md b/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md index 349c3064e..83eada253 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md +++ b/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md @@ -4,13 +4,14 @@ Generated by `pipeline/ab2026_proposed_keys.py` from the staged families; don't A row whose scope matches no registered measure has `measure_key: null` and names one of these keys in `conditions.proposed_measure_key` (README rule 2). A key is a measure with no Table 6.1 line (`ab2026__`), a package (`ab2026__package_`) or an option (`ab2026_option__`). Each family's NOTES.md gives the key's scope as Table 6.1 lines or the producer's own words. Tranche 2 registers each key in `data/uk/ab2026_measures.json`, or folds it into a registered key, before the ingest. -23 proposed keys (8 measures, 15 packages, 0 options) on 67 rows. +24 proposed keys (9 measures, 15 packages, 0 options) on 71 rows. | proposed key | kind | rows | families | producers | |---|---|--:|---|---| | `ab2026__defence_investment_plan` | measure | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__energy_debt_circuit_breaker` | measure | 3 | uk_hmt_redbook | hm_treasury | | `ab2026__english_devolution_agreements` | measure | 1 | uk_hmt_redbook | hm_treasury | +| `ab2026__gas_standing_charge_moved_to_unit_rates` | measure | 3 | uk_ifs_ab2026 | ifs | | `ab2026__minimum_wage_uprating_april_2027` | measure | 2 | uk_hmt_redbook | hm_treasury | | `ab2026__mod_counter_fraud_strategy` | measure | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_bbb_equity_community_lender_and_venture_debt` | package | 1 | uk_hmt_redbook | hm_treasury | @@ -26,7 +27,7 @@ A row whose scope matches no registered measure has `measure_key: null` and name | `ab2026__package_parliament_regulatory_burden_reduction` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_parliament_tax_gap_measures` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_parliament_welfare_fraud_and_error_savings` | package | 1 | uk_hmt_redbook | hm_treasury | -| `ab2026__package_property_and_savings_income_rates` | package | 2 | uk_hmt_redbook | hm_treasury | +| `ab2026__package_property_and_savings_income_rates` | package | 3 | uk_hmt_redbook, uk_ifs_ab2026 | hm_treasury, ifs | | `ab2026__package_regulation_tariff_and_sandbox_business_savings` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__regulatory_administrative_savings` | measure | 5 | uk_hmt_redbook | hm_treasury | | `ab2026__rolls_royce_gas_turbine_programme` | measure | 1 | uk_hmt_redbook | hm_treasury | diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/NOTES.md new file mode 100644 index 000000000..1f211e930 --- /dev/null +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/NOTES.md @@ -0,0 +1,526 @@ +# uk_ifs_ab2026 — Institute for Fiscal Studies, Autumn Budget 2026 + +**Working files.** `stage.py`, which writes `claims_staged.jsonl.gz` and is the file meant wherever these notes say `stage.py`, lives on branch `uk/ab2026-harvest-staging` at `harvest-staging/ab2026/lanes/ifs/uk_ifs_ab2026/stage.py`; this branch carries only NOTES.md, `manifest.jsonl` and `claims_staged.jsonl.gz`. `.ab2026-primaries/uk_ifs_ab2026/` was the lane's local primaries folder and was never committed: the repository carries no primary bytes, so fetch each primary from its `manifest.jsonl` URL and check its sha256. + +**87 rows from four current briefing decks** (five manifested PDF versions, including the superseded TW_0). `source: ifs`. The first lane staged 106 rows on 29 October. On 3 November the IFS addendum (`briefs/ifs-addendum.md` on the staging branch) was applied: sections 3 and 4 in full, section 2 as far as the network allowed (none), section 1 not at all (see "Pending" below). Its edits sit in an explicit block at the end of `stage.py`'s data. The sections below this summary keep the first lane's per-deck record; where they differ from the summary, the summary wins. + +## Pending: rows to be added from browser-read text (addendum section 1) + +These two IFS pages refuse non-browser clients, and their text is not on GitHub. The coordinator read them in a browser on 29 October and adds their rows locally after this family lands: + +- https://ifs.org.uk/articles/autumn-budget-2026-initial-response — pending: rows to be added from browser-read text. +- https://ifs.org.uk/collections/autumn-budget-2026 (Helen Miller's opening remarks) — pending: rows to be added from browser-read text. + +The addendum brief records that the coordinator's browser read found no estimates in the IFS explainer or the fiscal devolution comment, so those pages are no longer listed as unreached for want of access; they hold nothing to stage. + +## Unreached: the seven IFS data items (addendum section 2) + +The seven charts in the collection's "Key charts" section (listed in `harvest-staging/ab2026/ifs/data-items.json` on the staging branch) serve their data through Flourish. On 3 November the session's network policy refused public.flourish.studio and flo.uri.sh at the proxy (HTTP 403 to CONNECT), so no visualisation's data was read and none is staged. No value was read off an image. Each is unreached: + +| Data item | Flourish | +|---|---| +| https://ifs.org.uk/data-items/growth-real-household-disposable-income-over-decade-1950s-2020s | visualisation/30734907 | +| https://ifs.org.uk/data-items/current-budget-deficit-march-2026-and-october-2026-forecasts | visualisation/30726471 | +| https://ifs.org.uk/data-items/headroom-share-gdp | visualisation/30726788 | +| https://ifs.org.uk/data-items/real-gdp-growth-over-forecast | visualisation/30726917 | +| https://ifs.org.uk/data-items/average-annual-growth-real-household-disposable-income-parliament | visualisation/30726939 | +| https://ifs.org.uk/data-items/long-term-impact-announced-tax-changes-fiscal-event | visualisation/30726047 | +| https://ifs.org.uk/data-items/number-individuals-aged-16-paying-higher-or-additional-rate-income-tax-0 | visualisation/30727239 | + +## Coverage + +| Primary | Rows | +|---|--:| +| Weathering the fiscal storm? (Isabel Stockton) | 39 | +| Cost of living measures, TW_1 (Tom Wernham) | 15 | +| Cost of living measures, TW_0 (superseded) | 0 | +| Spending on public services: the Achilles' heel? (Bee Boileau) | 21 | +| Tax measures (Stuart Adam) | 12 | + +| measure_key | Rows | of which components (rule 9) | +|---|--:|--:| +| `null`, context (no proposed key) | 60 | 0 | +| `null` with a proposed key | 4 | 0 | +| `ab2026__founders_relief_cgt` | 4 | 0 | +| `ab2026__household_energy_payment_100_bands_a_to_d` | 3 | 0 | +| `ab2026__package_household_energy_vat_and_fuel_duty` | 3 | 0 | +| `ab2026__hmrc_tax_gap_package` | 2 | 2 | +| `ab2026__badr_lifetime_limit_500000` | 1 | 0 | +| `ab2026__business_rates_sbrr_rhl_thresholds_raised` | 1 | 0 | +| `ab2026__distributions_framework_modernised` | 1 | 0 | +| `ab2026__electricity_vat_zero_rate_to_july_2027` | 1 | 0 | +| `ab2026__fuel_duty_5p_cut_extended_and_rpi_rise_deferred` | 1 | 0 | +| `ab2026__package_total_tax_policy_decisions` | 1 | 0 | +| `ab2026__writing_down_allowance_12pc_and_fya_50pc` | 1 | 0 | + +Attribution: 36 `own`, 51 `restated`. + +## Proposed measure keys + +- `ab2026__package_property_and_savings_income_rates` (1 row): exactly Table 6.1 lines 61 and 62, Adam slide 2's "Increased income tax on savings & property income", £1.1bn in 2031–32 (lines 61 and 62 print 455 + 590 = 1,045). +- `ab2026__gas_standing_charge_moved_to_unit_rates` (3 rows): no Table 6.1 line. Wernham slide 8's "Shifting gas standing charge onto unit price": "This charge will be scrapped, but per-unit price will be increased to absorb this". The spelling is the RF family's. + +## Keys the addendum changed (rules 3 and 9) + +| Rows (first lane, 0-based) | Was | Now | Why | +|---|---|---|---| +| 45, 46, 52 (£5bn, £140, 35%) | proposed `ab2026__package_ifs_cost_of_living_support` | `ab2026__package_household_energy_vat_and_fuel_duty` | IFS's three measures are the energy payment, the whole electricity VAT cut and the fuel duty cut, exactly lines 1, 2 and 3 (README rule 3, corrected). Lines 1+2+3 print 5,040 over 2026-27 and 2027-28, IFS's £5bn. | +| 48 (£36) | proposed `ab2026__package_ifs_electricity_vat_oct_2026_june_2027` | `ab2026__electricity_vat_zero_rate_to_july_2027` | Line 2 costs the whole cut, 1 October 2026 to 1 July 2027, including the weeks before the Budget. | +| 89 (£1.8bn) | proposed `ab2026__package_ifs_self_assessment_payment_timing` | `ab2026__hmrc_tax_gap_package`, component | Table 6.3 row 3, "More timely payment" for Self Assessment, +1,815m in 2031–32: a part of line 45. | +| 92 (£1bn) | proposed `ab2026__package_ifs_hmrc_compliance_staff_and_debt_collection` | `ab2026__hmrc_tax_gap_package`, component | Table 6.3 rows 7 (debt management, +575m) and 8 (compliance officers, +415m), 990 in 2031–32: a part of line 45. | +| 91 (£1.1bn) | proposed `ab2026__package_ifs_savings_and_property_income_tax` | proposed `ab2026__package_property_and_savings_income_rates` | The name README rule 2 gives exactly lines 61 and 62. | +| 55–57 | proposed `ab2026__package_ifs_gas_standing_charge_reallocation` | proposed `ab2026__gas_standing_charge_moved_to_unit_rates` | A measure with no Table 6.1 line is `ab2026__`; the RF family's spelling. | +| 61 (£1.2bn per year) | proposed `ab2026__package_ifs_defence_investment_plan_funding` | dropped | See below. | + +**Defence Investment Plan funding.** The addendum asked whether IFS's "£1.2bn per year gap in Defence Investment Plan funding" is an estimate (keep, as context) or not. The first lane's own note calls it the government's DIP funding announcement, restated: the Budget "filled" the gap. That makes it a spending allocation, which rule 8 excludes, so it is dropped rather than kept as context. It is close to, but not exactly, Table 6.1 line 41 (capital reallocation to fully fund the DIP: +1,100, +1,135, +1,180 in 2029-30 to 2031-32), so it is not keyed there either. + +## Not staged (rules 8 and 9) + +The addendum dropped 19 of the first lane's 106 rows, all under rule 8: + +| Category | Rows | Examples (deck, slide) | +|---|--:|---| +| Measure parameters: rates, payment amounts, thresholds, a fare cap | 10 | £100 Household Energy Payment (Wernham 3); 5p fuel duty cut (Wernham 3); 0% VAT on electricity (Wernham 3, Adam 5); £2 bus fare cap (Wernham 15); Founders' Relief 10% rate, 24% comparator and £10m limit (Adam 3); RHL rateable values £12,000 and £21,000 (Adam 4) | +| Government targets | 3 | defence at 3% of GDP by 2030 (Boileau 4, repeated on 8); the DIP's earlier 2.7% in 2029–30 (Boileau 4) | +| Inputs of IFS's own spending scenarios | 2 | defence held at 2.7% of GDP, or at 3% by 2030–31 (Boileau 7 and 9); the scenarios' computed growth rates stay | +| Financing and allocations | 2 | ~£1bn a year of NWF and BBB financing for defence (Boileau 5); the £1.2bn a year of DIP funding (Boileau 4) | +| The base a measure moves | 2 | the £108 average gas standing charge and its 13% share of a typical gas bill (Wernham 8), as the red book dropped its £108 | + +Each dropped parameter that names a measure stays in the kept rows' `reform_hint`. The 25% capital and 40% day-to-day reserve shares (Boileau 11) stay: the addendum lists them as estimates. The bond spread and debt-ownership context rows (Stockton 7) are not design and stay as context. + +## Verification (3 November 2026) + +- **Checker:** `PYTHONPATH=. python pipeline/ab2026_harvest_contract.py sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026` reports 87 rows, 0 problems. +- **Determinism:** a re-run of `stage.py` reproduces `claims_staged.jsonl.gz` byte for byte. COUNTS.json pins its sha256. +- **Spot check: not done.** The addendum asked for 15 random rows re-opened at their locators, five from the new primaries. There are no new primaries (sections 1 and 2 above), and the session's network policy refused ifs.org.uk, so no deck could be re-fetched. Every kept row's printed value is unchanged from the first lane, whose spot checks are recorded in the deck sections below. +- **Stale files:** the family had no report or spot-check files. + +## Access recipe and primary inventory + +Fresh fetches began **2026-10-29 15:43:43 UTC**, using exactly: + +```sh +curl -sSL -A "Mozilla/5.0 (Macintosh; Intel Mac OS X 14_0) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/17.0 Safari/605.1.15" URL +``` + +The fetch also saved response headers and HTTP status and imposed a 50-second +timeout. Bytes are under `.ab2026-primaries/uk_ifs_ab2026/`, outside the vendored +family. SHA256s were checked with `shasum -a 256`. PDFs were read with +`pdftotext -layout` and rendered for visual checks of labels and embedded images. +None of the PDF annotations links to a chart-data download. + +| Primary | Access | Bytes | Rows | SHA256 | +|---|---|--:|--:|---| +| [Weathering the fiscal storm?](https://ifs.org.uk/system/files/2026-10/Weathering-the-fiscal-storm-IS.pdf) | direct | 318,348 | 39 | `fd09e707a4bc0a8e0a3de4a3fa902c8ccdc1e4c44ccdd0dee92866857aa4690b` | +| [Cost of living measures (TW_0, superseded)](https://ifs.org.uk/system/files/2026-10/Cost-of-living-measures-TW_0.pdf) | local copy | 231,180 | 0 | `abb63d9c4ecd21c3c9173077c5e92d8015c7cc30aebd12be159185d7d157f12a` | +| [Cost of living measures (TW_1, revised)](https://ifs.org.uk/system/files/2026-10/Cost-of-living-measures-TW_1.pdf) | direct | 222,191 | 21 | `6ed174b22b4d22a34c8bbd450194fb2b4181a9d566b137913bee52ca49a0299b` | +| [Spending on public services: the Achilles heel](https://ifs.org.uk/system/files/2026-10/Spending-on-public-services-the-Achilles-heel-BB.pdf) | direct | 157,838 | 28 | `7403ecc0641b23f5ffc0cb28f67435dbd1b8df6b124874ffae7ae7f75e357aa6` | +| [Tax measures](https://ifs.org.uk/system/files/2026-10/Tax-measures-SA.pdf) | direct | 133,005 | 18 | `6792dc448f40d34f174b6a3c839863b5c969a7334263f5d6b539be62a0428bb8` | + +Stockton's fresh file is byte-for-byte identical to assigned `stockton.pdf`, +confirming it is **Weathering-the-fiscal-storm-IS.pdf**. The other three current +decks also match their supplied local copies. TW_0 returned HTTP 403; the +supplied local PDF was copied to `TW_0-local.pdf` and hashed. All rows from the +cost-of-living presentation cite TW_1, not the superseded version. Full TW_0/TW_1 +changes are documented below; no numerical claim changed. + +HTML pages that returned a Cloudflare challenge (HTTP 403) to the first lane at 15:43:43 UTC; see "Pending" above for what has become of each: + +- https://ifs.org.uk/articles/autumn-budget-2026-initial-response +- https://ifs.org.uk/articles/autumn-budget-2026-explained +- https://ifs.org.uk/collections/autumn-budget-2026 +- https://ifs.org.uk/events/autumn-budget-2026-ifs-analysis + +No assigned local HTML copies exist. No attempt was made to bypass the blocks. +Challenge pages are retained only as fetch evidence in the ignored primaries +directory; they are not IFS publications and are not manifested. No rows are +staged from them. The explainer's existence/content cannot be confirmed from its +challenge response. The blocked collection and event pages prevent discovering +additional AB2026 items or linked xlsx/csv downloads. No data file was identified +inside the reachable PDFs; this is an access limit, not a finding that IFS posted +no data. No Wayback, alternative domain or other bypass was attempted. + +## Family conventions + +- Fiscal years use their end year and `conditions.fy`. Calendar years, + cross-fiscal-year nine-month support, and publication-date snapshots have + no `fy`; full horizons and missing dates are disclosed. Unspecified-year + Founders' Relief annual figures use annual 2026 publication markers and + medium confidence, not an inferred implementation year. +- Monetary magnitudes and signs are verbatim. £bn and £m are normalised to + GBP. Percentage forecasts and rates stay percent; recipient shares divide + by 100. Fiscal cells say percent of GDP or revenues as printed, rather + than inventing a real-GDP unit. 134 basis points is 1.34 percentage points. + The fuel-duty parameter keeps proposed unit `pence_per_litre`. +- Real/cash and price bases stay as published. Nine-month household support + is not annualised. A printed positive magnitude followed by “decrease” + remains positive, with the direction recorded in conditions. +- All figures are `point`: no producer scenario is substituted for a central + OBR estimate and no confidence band is invented. “Central” would mean a + producer-labelled central forecast; the staged figures contain no such label. +- `arithmetic` is used for IFS calculations where the deck does not name + TAXBEN. Official costings/forecasts are `restated`, with original source + named in notes where supplied or inferable; administrative policy parameters + are distinguished from IFS estimates. IFS does not cite PolicyEngine here. +- Current-law/default baseline is null. No new baseline world is inferred. + Every condition value is a string. Quotes are under 15 words, contain the + exact value_raw, and refer to printed text, a data label or a numeric cell. + +## Stockton fiscal deck + +39 rows: 38 printed numeric cells in slide 7's **Fiscal traffic lights** image +table, plus slide 8's “roughly 1 in 4 revisions” risk annotation. Text extraction +misses the entire image table; all cells were transcribed from the rendered +slide and independently audited. The table spans FY2026-27, 2027-28, 2029-30 and +2031-32; missing intermediate years were not invented. Net worth is explicitly +inverted and the signs were not reversed. The debt-interest denominator is +revenues, while debt ownership is a share of total debt. + +The table prints no source citation. OBR attribution for the 36 forecast cells +is inferred from the surrounding OBR forecast slides, not presented as an +explicit citation; these rows have medium confidence and disclose the inference. +Debt-ownership's administrative origin likewise has no named original provider +or observation date. The 134bp spread is an IFS market comparison with other +G7 economies whose source/weighting/date are unspecified. The latter two use a +2026 in-year snapshot. The 1-in-4 risk uses historical forecast revisions since +2010 excluding 2020-21; “Spring” is interpreted as Spring 2027 and disclosed. +All fiscal rows have null measure_key and no proposed measure. + +Not staged: slides 3–6 forecast/borrowing bars, diamonds and lines, and slide 8's +forecast/outturn lines, because they have no numeric data labels. Axis ticks, +chart origins and dates are not outcome rows. Slides 1–6 were visually checked +for hidden printed numbers. No headroom or tax/GDP number is printed there. + +## Proposed metrics and units + +The table lists every proposed metric used. Producer-specific meanings and +scopes are in the component sections and each row's statistic/conditions. +Existing AB2025 spellings are reused where they fit. `statutory_tax_rate` is a +single spelling across Adam and Wernham; `real_spending_growth_revision` +distinguishes a historical growth revision (percentage points) from growth. + +| Proposed metric | Rows | Unit(s) | Scope example | +|---|--:|---|---| +| `affected_share` | 2 | share | £100 Household Energy Payment (Jan 2027) | +| `average_energy_bill_change` | 1 | gbp | EDF forecast £470 increase in annual household gas and electricity prices in January | +| `average_household_support` | 4 | gbp | Cost of living support, Oct-June, England: energy payment, removal of VAT on electricity, fuel duty freeze and cut | +| `benefit_payment_amount` | 1 | gbp | £100 one-off household energy payment in England, January 2027 | +| `business_rates_bill_change` | 1 | gbp | maximum annual bill saving | +| `cpi_inflation_rate` | 2 | percent | The OBR forecasts 3.8% CPI inflation in Q1 2027 | +| `current_budget_deficit` | 4 | percent | Current budget deficit (% of GDP) | +| `current_budget_deficit_revision_risk` | 1 | share | Risks from the forecast | +| `debt_interest_share_of_revenue` | 4 | percent | Debt interest costs (% of revenues) | +| `debt_owned_by_non_residents_share` | 1 | percent | Debt owned by non-residents (% of total) | +| `energy_standing_charge` | 1 | gbp | Average standing charge for gas is £108 a year: | +| `energy_standing_charge_share` | 1 | share | 13% of typical gas bill | +| `fare_cap` | 1 | gbp | £2 cap on bus fares | +| `forecast_spending_change` | 1 | gbp | defence financing | +| `fuel_duty_rate_cut` | 1 | pence_per_litre | Temporary extension of 5p fuel duty cut and freeze to cover January to June 2027 inclusive | +| `gross_financing_requirement` | 4 | percent | Gross financing requirement (% of GDP) | +| `primary_deficit` | 4 | percent | Primary deficit (% of GDP) | +| `public_sector_net_borrowing` | 4 | percent | Net Borrowing (% of GDP) | +| `public_sector_net_debt_ex_boe` | 4 | percent | PSND ex BoE (% of GDP) | +| `public_sector_net_debt_ex_boe_change` | 4 | percent | PSND ex BoE (year-on-year change, % of GDP) | +| `public_sector_net_financial_liabilities_change` | 4 | percent | PSNFL (year-on-year change, % of GDP) | +| `public_sector_net_worth_inverted_change` | 4 | percent | PSNW (inverted, year-on-year change, % of GDP) | +| `real_spending_growth` | 11 | percent | average annual real-terms budget growth | +| `real_spending_growth_revision` | 1 | percentage_points | average revision to annual spending growth when budgets must be allocated | +| `reserve_allocated_share` | 2 | percent | share of reserve already allocated | +| `share_of_energy_price_increase_offset` | 1 | share | Three cost-of-living measures: household energy payment, electricity VAT relief and fuel duty freeze/cut | +| `spending_pressure` | 7 | gbp | total departmental spending | +| `statutory_tax_rate` | 4 | percent | Temporary 0% VAT on domestic electricity; previously announced for Oct to April, extended to end of June | +| `tax_threshold_level` | 3 | gbp | maximum qualifying gains | +| `taxpayer_gain` | 1 | gbp | maximum gain per beneficiary | +| `ten_year_bond_spread_other_g7` | 1 | percentage_points | 10-year bond spread (bps, vs. other G7 economies) | + +The only proposed unit is `pence_per_litre`: the printed 5p fuel duty cut's +per-litre amount, following the policy's named basis. No proposed baseline. + +## Wernham cost-of-living deck + +Tom Wernham component: 21 rows from the revised 17-slide Cost of living measures deck, cover dated 29 October 2026. Source slug `ifs`; own household-price calculations use `arithmetic` (the deck does not name TAXBEN), OBR forecasts use `obr_efo_forecast`, EDF's energy-price forecast uses `edf_energy_price_forecast`, and announced statutory parameters use `administrative_data` with restated attribution and administrative_fact benchmark. No chart bars were read for values. + +Access/version evidence: + +- TW_1 freshly fetched directly at 2026-10-29 15:43:43Z using the assigned Safari user agent. SHA256 `6ed174b22b4d22a34c8bbd450194fb2b4181a9d566b137913bee52ca49a0299b`, 222191 bytes. This hash is identical to the coordinator's local copy of the TW_1 PDF. +- TW_0 fresh URL returned HTTP403, so the assigned local file was copied to `.ab2026-primaries/uk_ifs_ab2026/TW_0-local.pdf`, hashed, and extracted using `pdftotext -layout`. SHA256 `abb63d9c4ecd21c3c9173077c5e92d8015c7cc30aebd12be159185d7d157f12a`, 231180 bytes; manifest access must be `local copy`. No bypass was attempted. +- Text diff: TW_0 has18 pages, TW_1 has17. TW_0 p4 already includes the80%/91% eligibility annotations; TW_1 p4 omits these. TW_0 p5 has those eligibility annotations overlapping the VAT/fuel commentary. TW_1 p5 relocates eligibility to the lower chart and VAT/fuel commentary to the upper chart. TW_0 p6 is another copy of the unannotated support chart and was removed. TW_0 p7–18 correspond to TW_1 p6–17. No printed numerical claim changes. All staged rows cite TW_1; TW_0 contributes zero rows but is manifested as the changed primary required by rule7. + +Coverage by measure_key: null14; `ab2026__household_energy_payment_100_bands_a_to_d`4; `ab2026__fuel_duty_5p_cut_extended_and_rpi_rise_deferred`2; `ab2026__bus_fare_cap_2_england_outside_london_2027`1. Within null: 6 unscored context rows; the remaining8 use proposed measure keys below. Rows per primary: TW_1=21, TW_0=0. + +Proposed measure scopes: the first lane's three Wernham proposals were re-keyed on 3 November (see "Keys the addendum changed" above). The cost-of-living package is exactly lines 1, 2 and 3, because line 2 costs the whole VAT cut including the weeks before the Budget; the full-period VAT row is line 2; the gas standing-charge rows use `ab2026__gas_standing_charge_moved_to_unit_rates`. + +Proposed metrics and units: + +- `cpi_inflation_rate`: printed OBR3.8% and1.9% CPI forecast levels for Q12027, not inflation effects from measures. +- `average_energy_bill_change`: existing AB2025 proposal; EDF£470 annual energy-price increase from January2027. +- `benefit_payment_amount`: £100 per eligible household one-off payment, rather than the average gain across all households. +- `fuel_duty_rate_cut`, proposed_unit `pence_per_litre`: printed5p extension; no negative sign is inserted. Per-litre basis follows the named policy, though the slide prints only5p. +- `statutory_tax_rate`:0% electricity VAT during October2026–June2027, not an effective tax rate. +- `average_household_support`: nine-month cash support, including indirect tax reductions; no conversion to annual gains or household income changes. Printed£140 headline and £24/£36/£80 all-household chart labels. +- `affected_share`: existing AB2025 proposal, payment receipt by80% of all England households and91% of the poorest30%; normalized to0.8 and0.91 shares. Household ranking is by spending, not income decile. +- `share_of_energy_price_increase_offset`: the printed "around35%" of forecast price increases covered by the three measures; normalized to0.35 share, no ratio derived from bars. +- `energy_standing_charge`: current annual gas standing charge£108; context. +- `energy_standing_charge_share`: current13% of a typical gas bill; normalized to0.13 share, context. +- `fare_cap`: £2 single bus fare cap, a restated policy parameter. Exact2027 timing and England-outside-London scope follow registered Table6.1 line4; the slide itself gives only "£2 cap on bus fares", and the row explicitly says this. + +Conventions: GBP amounts stay as printed in cash or2026-27 prices as specified; the £5bn aggregate is multiplied by1e9. Shares divide printed percentages by100, while CPI percentages and the printed1% income decrease remain percentages. Positive cost is cost to the Exchequer; positive household support is a gain. The printed1% decrease in real household disposable income is not re-signed; its direction is explicit in sign_convention. October2026–June2027 figures use annual time_basis, end-period marker2027 and horizon strings, with no fiscal-year condition because the nine months cross fiscal years. The1% RHDI decrease spansFY2024-25 toFY2026-27 and usesFY2026-27/end2027. No row is described as central: all figures are printed point estimates/parameters. For gas reform, year2026 is explicitly only a publication marker because no implementation year is printed. Baseline remains null; unsupported counterfactual definitions were not invented. + +Not staged: + +- Spending-decile bars in pp4–6, energy-price bars/percentiles pp7/16/17, carbon implicit-tax bars p9, and real-income decade/forecast lines pp11/14 have no data labels (apart from the separately staged£24/£36/£80 all-household labels and1% annotation). Visual inspection found no further printed values, embedded raster tables, or hidden numeric labels. +- Slide5 repeats p4's £140/£24/£36/£80 chart values; only its new80%/91% annotations are additional rows. Repeated£100 legend descriptions are not extra rows. +- Richest30% "lose a similar amount" is not an exact printed cash number; no negative£13 was inferred. +-30% subgroup labels,5–15 bus-travel age eligibility, dates, axis ticks, years and0% chart origins are conditions/context, not separate scored estimates. +- Other temporary announcements on p15 (food tariffs, family-days-out VAT, free bus travel) have no printed cost/effect values, apart from the£2 bus fare parameter staged. +- Gas-standing-charge reform's average "no give-away" and "does not change overall costs" are qualitative; no numerical zero was invented. Its future electricity counterpart has no printed effect. +- No data download was identified within the PDF; the collection/event fetches were blocked as documented above. + +Potential contradictions/limits: the package's modelled full electricity-VAT period includes already-announced relief, despite being discussed beside Budget measures. This is why the package and full-period electricity rows have proposed scopes. The printed£140 headline is rounded; do not derive a different total from a plotted stack. Gas reform has unspecified implementation year and geography. The positive1% magnitude is explicitly a decrease. No numerical difference was detected between TW_0 and TW_1. + +Component spot check: deterministic random sample seed230,10 of21 rows, reopened directly from TW_1 PDF text; labels and direction were also checked against rendered pp4,5,7,8,14. One-based component rows1,6,7,8,10,11,14,15,18,19: p2 3.8% CPI inQ12027; p3 0% VAT October2026–June2027; p3 £5bn positive Exchequer cost across nine months; p4 £140 nine-month household support; p4 £36 VAT label; p4 £80 payment label (all in2026-27 prices); p7 around35% offset→0.35; p8 current£108 annual gas charge; p8 40% worse off→0.4 losing share; p8 poorest30% gain£13 per year, date unspecified/publication marker2026. All values found verbatim. The21 component rows pass row_problems with0 problems using UV_CACHE_DIR=/tmp/ab2026-ifs-uv-cache (default uv cache is outside the sandbox). + + +## Bee Boileau public services deck + +Primary: *Spending on public services: the Achilles’ heel?*, 29 October 2026, +https://ifs.org.uk/system/files/2026-10/Spending-on-public-services-the-Achilles-heel-BB.pdf. +Fresh direct fetch 2026-10-29 15:43:43Z with the required Safari User-Agent; +157,838 bytes, sha256 `7403ecc0641b23f5ffc0cb28f67435dbd1b8df6b124874ffae7ae7f75e357aa6`. +The coordinator's assigned local copy and fresh copy are byte-identical. Read with +`pdftotext -layout`; visually inspected all 13 slides and the slide 7/9 data labels. +No raster-only numeric table was missed. + +Coverage: 28 rows, all `measure_key: null` (one has the proposed package below). +By slide: 3:1, 4:4, 5:1, 7:6, 8:1, 9:10, 10:3, 11:2. +The figures are 21 own arithmetic calculations/scenario assumptions and 7 restated +announced policy/target/reserve figures. No exact registered AB2026 measure is assigned. + +The first lane's one proposed key here, for the £1.2bn a year of DIP funding, went with that row, which the addendum dropped as an allocation (see "Keys the addendum changed" above). The rule 8 drops also removed slide 4's and 8's targets, slide 5's financing and the two scenario inputs on slides 7 and 9, so the deck now has 21 rows. + +Metric proposals reuse `spending_pressure`, `forecast_spending_change` and +`real_spending_growth` from AB2025. The historical one-percentage-point revision uses `real_spending_growth_revision`. New `reserve_allocated_share` means the +percentage of the specified day-to-day/capital reserve already allocated over +its printed multi-year window. No new unit or baseline proposal. + +`arithmetic` is the IFS's own scenario/growth calculation, or an announced target +restated without a named model; the £1.2bn funding amount is marked restated with +`hmt_costing_obr_certified`. Restated rows name their government/DIP origin in +`note`; two announced reserve allocation shares are `administrative_fact`. +Real growth remains percent, growth revisions are percentage points, cash +amounts are GBP, and signs are as printed. Positive pressure means additional +spending required; negative growth is a real cut. There are no central-value +rows: IFS's scenarios are point estimates, never substituted for OBR forecasts. + +The 2027–28 to 2030–31 chart rates are average annual real growth over the whole +interval: `growth_period` preserves that interval, and the row's fiscal period +is its terminal year, 2031. The reserve shares likewise retain their complete +allocation windows, ending 2028–29 and 2029–30 respectively. The published £1.2bn +and ~£1bn annual flows have no fiscal year: they carry a 2026 publication-date +snapshot, `time_basis: point_in_time`, `reference_date`, and medium confidence; +the rows do not claim these are 2026 annual outlays. The historical average +1-percentage-point growth revision since 2010 uses the same dated-snapshot +convention. Slide 4 states a calendar-2030 defence target, so that row has no fy; +slide 9 explicitly says 2030–31 and is recorded as fiscal. The difference in date +labels is preserved, not resolved by the harvest. + +Not staged: slide 3's unlabelled growth bars and axis ticks; all dates, document +address numbers and slide enumeration; qualitative/unquantified claims; the +second occurrence of the identical 3%-of-GDP scenario annotation within slide 9. +Slide 7 and slide 9 repeated printed labels are retained at their separate +locators, and slide 9's two health bars are separate scenario observations. +The £1bn financing flow is explicitly investment/financing, not military +purchases, matching the slide's caution. The deck identifies no downloadable +data file or distributional chart. No public-services primary was unreached. + +Validation: `row_problems` from the AB2026 checker, with registered keys and enums, +reported **28 rows, 0 row problems**. Full-family validation is recorded below. A full visual pass found no contradictory value. +The calendar/fiscal defence target wording and unspecified flow years above are +material limitations. + +Deterministic random spot-check (seed 230, 1-based indices within the Boileau component): + +| Row | Locator | Printed value | Normalized unit/value | Period check | +|---|---|---|---|---| +| 1 | slide 3 | £8 billion | GBP 8,000,000,000 | FY2030–31 | +| 6 | slide 5 | ~£1bn per year | GBP 1,000,000,000 | no year printed; dated snapshot, medium | +| 10 | slide 7 | 1.7% | percent 1.7 | average 2027–28 to2030–31 | +| 14 | slide 9 | 0.9% | percent 0.9 | average 2027–28 to2030–31 | +| 16 | slide 9 | £9 billion | GBP 9,000,000,000 | interval ends FY2030–31 | +| 18 | slide 9 | 2.7% | percent 2.7 | yellow health bar, average interval | +| 22 | slide 9 | -0.9% | percent -0.9 | green other-departments bar, average interval | +| 23 | slide 9 | -2.1% | percent -2.1 | yellow other-departments bar, average interval | +| 24 | slide 10 | £28 billion | GBP 28,000,000,000 | FY2030–31 | +| 25 | slide 10 | 1 percentage point | percentage_points 1 | historic since2010, dated snapshot | + +For every sampled row the primary PDF was reopened with `pdftotext -f/-l` and the +number/string, sign, unit and context checked; slide 7/9 labels were also viewed +at full-page resolution. All printed value strings matched. No bar reading or +arithmetic derived by the harvest was used. + + +## Stuart Adam: Tax measures + +18 rows from one six-slide PDF. Direct fetch with the required browser User-Agent, +2026-10-29 15:43:43Z; 133,005 bytes; +sha256 `6792dc448f40d34f174b6a3c839863b5c969a7334263f5d6b539be62a0428bb8`. +The fetched bytes match the assignment's local copy. Text read with +`pdftotext -layout`; all six slides rendered and visually read as a safeguard +against raster text. Slides 2–6 have no embedded raster images and all printed +numbers are in the text extraction. No distributional chart or data download +is linked inside this deck. + +Rows by registered key: `ab2026__package_total_tax_policy_decisions` 1; +`ab2026__writing_down_allowance_12pc_and_fya_50pc` 1; +`ab2026__distributions_framework_modernised` 1; +`ab2026__badr_lifetime_limit_500000` 1; +`ab2026__founders_relief_cgt` 6; +`ab2026__business_rates_sbrr_rhl_thresholds_raised` 3; +`ab2026__electricity_vat_zero_rate_to_july_2027` 1; null 4. +### Proposed measure scopes + +Superseded on 3 November (see "Keys the addendum changed" above): the self-assessment timing and the compliance-staff and debt-collection figures are line 45 components, matched to Table 6.3's items by value, and the savings and property income figure uses `ab2026__package_property_and_savings_income_rates`. Rule 8 dropped this deck's six parameter rows (the Founders' Relief rate, comparator and limit, the RHL rateable values and the 0% VAT rate), so it now has 12 rows. + +### Conventions and metric proposals + +All monetary figures preserve printed signs. Revenue-raising amounts are +positive yields; the Founders' Relief £0.3bn is a positive cost, explicitly +labelled as such. Amounts expressed in £bn/£m are multiplied by 1e9/1e6. +Tax rates remain percent. All quoted excerpts have fewer than 15 words and +contain their exact `value_raw`. + +The HMT-certified Budget costings are `restated` with +`source_model: hmt_costing_obr_certified`; rates and eligibility parameters +are `restated` announced-government design with `administrative_data`. +The deck does not cite a source for its beneficiary count or gain calculations; +these are presented as IFS `own`, `arithmetic`. Its maximum business-rate +saving is also IFS arithmetic. `point` is used for printed single estimates; +the £12,000–£21,000 benefiting-property range uses range endpoints. No +uncertainty range is inferred. There are no rows labelled `central` in this deck. + +- Reused `tax_threshold_level` (GBP): the £10m cap on qualifying Founders' + Relief gains and the lower/upper rateable-value bounds identifying RHL + properties benefiting from higher relief thresholds. +- Proposed `statutory_tax_rate` (percent): the statutory 10% Founders' Relief rate, + its current 24% standard-rate comparator, and the temporary 0% electricity + VAT rate. These are statutory parameters, not effective marginal rates. +- Proposed `taxpayer_gain` (GBP): the printed maximum £1.4m gain per + Founders' Relief beneficiary, with the maximum recorded in conditions. +- Proposed `business_rates_bill_change` (GBP): maximum annual bill saving + to an RHL property from the higher relief thresholds, £7,146 in 2027-28; + positive means a gain to the business. + +Fiscal-year impacts use their end year. The BADR £0.3bn bullet omits the +year; 2031–32 is inherited from the preceding slide's final-year comparison +and its adjacent Founders' Relief bullet. That row has medium parse confidence +and explains the inheritance. The annual count of 600 beneficiaries and +average £500,000 gain have no year printed; they use an annual 2026 +publication-year marker, with explicit `period_assignment` and `horizon` +conditions and medium confidence. This marker does not claim the relief operated in 2026. +Unspecified-year parameters and the maximum £1.4m gain use point-in-time +2026 as announcements, with no fiscal-year condition. No baseline proposal. + +### Omissions and source issues + +- No effective marginal rates, distributions or unlabelled charts occur in + this deck. No amounts are inferred from qualitative bullets. +- Dates and durations are descriptive scope rather than numerical impact + rows: the bank surcharge's 15-month period; two years of business-rate + discounts; the 2029 revaluation; April 2028 income-tax revenue devolution. + The devolution slide prints no quantified spending or revenue amount. +- The title slide prints “28 October 2026”, despite the assignment describing + the 29 October briefing and PDF metadata showing creation/modification on + 29 October. Publication date follows the printed cover (2026-10-28); the + contradictory date evidence is preserved here, not silently corrected. +- IFS's £1.1bn combined savings/property yield is retained verbatim. The + corresponding Table 6.1 line-61/62 final-year amounts are £455m and £590m, + totalling £1.045bn. Different rounding/aggregation could explain the + displayed £1.1bn; no correction or derivation is staged. +- The £1bn HMRC compliance/debt-collection figure is narrower than line 45's + full £3.160bn. It is not treated as a contradictory costing of the full package. + +### Verification + +The 18 individual rows pass `row_problems` from the harvest contract with +zero problems. All 18 short quotes were matched against whitespace-normalised +`pdftotext -layout` output and visually checked on the rendered slides. +Random spot-check seed 230 selected the following ten local rows. Each was +reopened at its slide and checked for exact value, sign, unit and period: + +| Local row | Slide | Printed value | Sign/basis | Period | +|--:|--:|---|---|---| +| 1 | 2 | £7bn | positive Exchequer yield, GBP | 2031-32 | +| 6 | 2 | £0.8bn | positive Exchequer yield, GBP | 2031-32 | +| 7 | 3 | £0.3bn | positive Exchequer yield, GBP | inherited 2031-32, disclosed | +| 8 | 3 | £0.3bn | positive Exchequer cost, GBP | explicit 2031-32 | +| 10 | 3 | 24% | standard statutory rate, percent | 2026 context marker | +| 11 | 3 | £10m | qualifying gains cap, GBP | 2026 announcement marker | +| 13 | 3 | £1.4m | maximum positive taxpayer gain, GBP | 2026 announcement marker | +| 14 | 3 | £500,000 | average positive beneficiary gain, GBP | annual, year unspecified; 2026 marker disclosed | +| 15 | 4 | £12,000 | lower rateable-value bound, GBP | 2027-28 | +| 16 | 4 | £21,000 | upper rateable-value bound, GBP | 2027-28 | + + +## Final family verification + +The required command, with its uv cache confined to the ignored family primaries +directory, completed successfully: + +```sh +UV_CACHE_DIR=.ab2026-primaries/uk_ifs_ab2026/uv-cache PYTHONPATH=. uv run -q python pipeline/ab2026_harvest_contract.py sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026 +``` + +```text +sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026: 106 rows, 0 problems +``` + +Re-running `stage.py` produced the same gzip SHA256 before and after: +`4f0c9cd197d970486608ea093bf3822ae1d92fa56ebc4a2ac1fae3086430b1cb`. +68 text-row value_raw strings were also matched at their exact PDF slide +locators. Sixty-five quotes match the whitespace-normalised extraction directly; +three Boileau chart annotations interleave with other columns in pdftotext and +were confirmed against rendered slides 7 and 9. All 38 embedded fiscal-table +cells were independently checked visually. All quotes have fewer than 15 words +and contain value_raw. + +The final family random sample uses seed 230, ten of 106 rows, one-based gzip +indices. Each primary was reopened at the locator and its number, sign, unit and +period checked. All ten passed; unspecified dates are disclosed markers, not +confirmed implementation dates. + +| Row | Primary / locator | Value and sign | Normalized unit | Period checked | +|--:|---|---|---|---| +| 3 | Stockton, slide 7, current budget deficit | -0.4, surplus direction | percent of GDP | FY2029-30 / 2030 | +| 21 | Stockton, slide 7, PSND ex BoE year-on-year change | 3.4, positive change | percent of GDP as printed | FY2026-27 / 2027 | +| 39 | Stockton, slide 8, forecast risk annotation | roughly 1 in 4 | share 0.25 | Spring 2027 inference disclosed | +| 56 | Wernham, slide 8, gas reform winners | 60%, better off | share 0.60 | implementation unspecified; 2026 publication marker | +| 69 | Boileau, slide 7, Health and Social Care | 2.7%, positive real growth | percent | annual average FY2027-28 to FY2030-31 / 2031 | +| 85 | Boileau, slide 10, historical top-ups | 1 percentage point increase | percentage_points 1 | historical since 2010; 2026 snapshot | +| 87 | Boileau, slide 11, day-to-day reserve | 40%, already allocated | percent | FY2027-28 to FY2028-29 / 2029 | +| 91 | Adam, slide 2, business investment allowances | £1.1bn, positive yield | GBP 1,100,000,000 | FY2031-32 / 2032 | +| 93 | Adam, slide 2, HMRC staffing/debt collection | £1bn, positive yield | GBP 1,000,000,000 | FY2031-32 / 2032 | +| 106 | Adam, slide 5, electricity VAT | 0%, statutory rate | percent 0 | announcement at publication in 2026; no fy | + +Registered-key scope review covered every non-null row: + +| Registered key suffix (all prefixed ab2026__) | Table 6.1 line | Printed scope checked | +|---|--:|---| +| household_energy_payment_100_bands_a_to_d | 1 | one-off payment and all-household mean/receipt estimates | +| fuel_duty_5p_cut_extended_and_rpi_rise_deferred | 3 | 5p cut/freeze extension and its Jan-Jun average support | +| bus_fare_cap_2_england_outside_london_2027 | 4 | £2 fare parameter; registry timing/geography explicitly sourced in note | +| founders_relief_cgt | 26 | relief costing, design, beneficiary count and gains | +| distributions_framework_modernised | 49 | reclassification of capital gains as dividends | +| badr_lifetime_limit_500000 | 60 | further cut to BADR | +| business_rates_sbrr_rhl_thresholds_raised | 13 | benefiting rateable-value range and maximum saving from threshold increase | +| writing_down_allowance_12pc_and_fya_50pc | 58 | business investment allowances | +| electricity_vat_zero_rate_to_july_2027 | 2 | Adam statutory zero-rate parameter, no full-period gain/cost attached | +| package_total_tax_policy_decisions | total | net Budget tax rise in final forecast year | + +Wernham's full-period VAT *impact* uses a proposal because it includes prior +relief; Adam's zero-rate *parameter* is directly stated for the measure and is +not a costing of the full period. All broader or unidentified packages use null +measure_key with proposals. Source issues are preserved: Adam's cover date, +savings/property rounding, Boileau's calendar/fiscal target wording and undated +flows, Wernham's undated gas reform, and Stockton's uncited attribution. None +was silently corrected. + diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/claims_staged.jsonl.gz new file mode 100644 index 0000000000000000000000000000000000000000..6958be62f23338852da20bd9ea93d117c451566a GIT binary patch literal 9694 zcmV<4B_Y}$iwFP!00002|LuM2lG{jf?tec;MMvmZifvK=$>v2J#~ac}ni+d%R+i*5 zaeg6+K#{}>1UR^Ki?iVfzr*=)!oT$<@|*ZcoS9X)kObKvcxkd(j&+Qh1d3Hzncr8H znU(eTD;~$8J4|9X@ULF6t58HSkA-t}&8~(#5}xY|YZiE}J%_LT#PjeA;oFlb55HKp z$D;_}bKKELgu;)lY2XMCU$+C_aj{)|;P2R)OOK9%P}n?*@fk<>!PND69EAAQk8eNV zrz1DAxo3$xXb(T;Nt{f5YnV7=5nC;z)upFYJJy}+$9ydC$-mlhFccxnp21#U;a`nb z4?h@-U>x$4x55g(; z+&_5QVIHwr2x5>E=La!K`?u_`Z$7eu9x*q9UjcOpp`}q6 zOj(SL!0>|vRKg$A3+eR9)}to-AfQ!{541Ye}Irozfm zRAS^@@Q@?~dWcqFiy1-YOd1n@b;rE~e+fpJ{$+h}eMbVFTX@I{`F$#nfB%QSUkW$$ z8m+$OhKuHgzUBs2umsHw88;Zso87n}mFY=y1B|jFH=sN&h#LltwxPM_8H^9C~P*jdn}Z!$s3WOVa}@Sc0a9B0cQ3cgTc# z>hy3^ogOY~_fR|PAT6FnVBDbMKA7DwHP00S_!Yj*jaTFT$%Yf;q*`)PEtphmVNy>L z{MRX$4keUhey7>GfKURAvLfQ6Jai~Q1O7$E2%a3^>-fM5;J0^m0i8vBpk3Jm)V?SO zQ2qFT2K*=J_B2eC@_B%MR^dfEOP| zSvfnjuYLGc0AHF=X>2$BZe39K!{WP|0#4YguaNT8!?^2JO8fHH0l#q(#{!JXz;9?@ zz6O5n-G5TRuYLL@IsdJT$oX$o&iQX?pS}iu?d5+`z^{G#s~MX+p9DUSg#PMBfg6`~ z7I+8uzQ+EGaQHt7?ALDoQtaz~!Qf4W|Nnqn)PHfw2$Pm%+wshalwm-9it*}VYoKa8>>8=yS2_g|BN_WVC}GSCr$ z6*3s?xHmYDh`^xo)?jd1JA>L00Zj%^J{s`E$v{U5R>+{YYk2r9f&{(F;o;szg@@~h z2s9Zy`S8HgCIcNdD035C!s&p1<Q%cetpZtSFqT38tc!g zoDa^aCB9n2Ks+|uzeV*zphw{jxd+$|LY@c_U$b)k(s6(_uAkPf6&}ta5P-w1>ilV4 zR*i?+p#Y8XPp^W{`D46x@2_Hfd&gD$nfUkHm6!1(5$)mE7=MXHd(I!@wbyR|*Ce4G z|4)-7bWotA{M#i|_BSh&gsCh0Ym(5Bg9|1J9aYd*0P8D&_Z3ypJfWixPm(8eWTKQO zT9}tf!6cWkfZ))oG)vqZ(V%h6~Sx1jiz7+y`Mid5|-H zW6#sKu9+r}GmmZ1oLlLHo(F4ep@BNOvCdO<3DFIsa&*Jc(GATVI>w>7Lq|C(aEEyj zQ4X^*cbGcLp}9lHIy86aXvf2DtHEVNI|kL=Rs$XF(Db2W9-2OM)Z^isqFa}7Q*^8P zP0=TKmLo*QNTlwZ+|>_=#V8(T2T>d`fNTy{%0q$HSM2!$g}g0!RhETIUwJ~ju|OI{ z!dM}TxB0{)VWhH4RQmiVCY6)mcJ}jDILF9 zcPr>c+ze6uVu({b`F&s(*zIx%%>t*(0)5Q_t1O@+5hX0py@dvPqV+w~Ee z1x}U)YJ?dsfCY3+p@ao4;Q+;;^7)CuDIT9VEDP*-!Gvak(`A8{W`PwJ(4m1c7Py4l zpa<2jk=t)ZD|f)A0s1h$dMB(Am*%jJ34$b!lh7|Af;fmlQmFIQsh|YaCqd}`A&2Cu zD4O!>-Lha9iSUl>n4!J0#E)jeCLAnB@!;MEY#Yu2qx=Sl2tg?`J?in{pq${kL8?S> zeqDyyKc|sG(pXS5fs-hiisruoQb6Te7(5UI3Pr_lktKX+FN5qR`!$1pu;2G4_C78U1UvtNx5+1Z4+<~ye;uLM^Eb7)H2%2(h zIJaQRt!&CAU7S8gR8ZLKOlh7g%XpzFWAQGE)=tStrQ%&?k0sm$)!=(15HfvO>sa`4 z)w?NLm2;CwKqVvs+IPTR1imQZA{@!5r+^7=q)7wKB z8;fQD(++z+oPkZ>!P0oO*SpXP2~47Y#9lJfFj(WCn3{3i0?4oGbI$uuESEaFcjHOP z2yQ_h0N~yntyoQ+)b49`FBG65gl&@J3c)22Y)o(pme`GcAUrrj#KNalnT7Lc$$7MB z?AEw<((JIc0&1Fguo7Uqb8waj!MH&EK>LnQ4gdvK4i$N2z;@;9DcYUmv&6*38FTHz zdKNdyZbXmni40(Yg&Ow>m%-yJQVM`zKrHFg3X<`}o9_VOCN9xC2;0)#WHehhMZ87` z&Ij5^$5(*mYbL(hUP7)I7!$yWiMs;1$&=7bL$INJx!|XNK?_0X;I2%di-rRN%R$PV zQNR{}4z@`(kDY|1CTcbvfxDB~N(E#g7#;_w zVLy%zwIO1$$6bfbJnjSBeuN%G`Q|6nQP_h8aki?_Fq^@wvL>c%V!@VcV(*$c*dBpj z!}17MR5DQ)JNSV#h<&1zWcDj8dfo(KC`}`BG4Z!$n8qGJI@^%u2h&=+J z9X$STKC)%6e+gy-E{Czy_kT4BQugkz!JfGO$dfDB0_P~K1+D{Z~)`= zr1J&UPk`v`7&qvgy`hma)xWiHIRUEQ16a;nE9*JvMtmd(#~K%PS2$*`Gi7qB%{v2bv<8D65blQ{BKdpn9qm@6 z^dt8CGa$roBJ?MGdQ1DB*QB87G!egXpL=uA?H4t$5P!H4j2iNX8|e=>KL2K!t4N4^ zQaqUvv39fnjQ#N8=Yj%)Wv7~68z2{++sfmezPSPIHpZZq?=azkrG~hznis%LwF-;c-Ee%T8Ew zj22GUQIAS$@T>=(sF#^4e*f0#8FfttA6K@yci0uqDAg?bu><1ocsMI!F9@U}_LA_9 z=A7Jz%hAW3qYqtTQ$UVM^lH(|76Int=~qyoCEK=Ai(#22`C`LF-aOff*W=Z|kb zEc-htvkX0t<8dg)h^sgXu$m@&4|Bu*&+E@j!Xg@lmpGv1IKCNrjG)^9$LM2AKzW>0 zc8%Adp&pBp*(?YFi=%+4k+R4RW&r9mx5`&Uwl+myrk4T$>PW=QMw@Xj@DGgJ%U>J} zpW7OrYvJ=>kyk+#ZPi^+KS1Vq5(-dO9!)5*lHc<=Snv05H$#6FdtsB0PeV^YBXQ-l z{9cak(gHzY=ctp{3{8-s2{Jp8@xKu?^@osQ);~lTkd0uNHJA!Y#wifP)eeZi(Mh^r zh`%w?|7%vqdX)4DWBogS?7j)b(x$x{50CYw@ z@;qhVc3}oo7&2W_KS&{C4oe8=R`%1;0r;a{=G;VZ2PgEF__9HKnPdDy&YSy(aA+t^ z_7<~oly+b~n?em1a{o(I7$z?LQz*PuWvz=4zv#O^^uFhj_XY*4oiNz|HE zE|^ume`}TGvR%DjgXb(Y0`?rxgf-B_Nnc)6M*3T}OXFw74y{gQa%gq*KIKv2zObC3 ztqJa$DXM72f%Uhv6!>oyg+IC;Qkf!1x=vz$&pynzWjcNWr%S zn&Y`+7YPAj3+V;6tcH=g05msfZYbl1zEQ#rQgeRFwDX@bxRAR}QKdyNSBc6V1&U5F26iI=11g2Uu^Xst5B@^hD2 zo_;ZwoyG?8Lm#j5!ZHOo2K*9I%Cyi|7%Y>H_*9+Vp^bAP${k@osUv;FFL6)#64wkb zQDNA~9TR|rVkUU3s^Poa^p&G0N`@pLLTXngn<}uUL_c|oJ4j-brN(WI+i3|b^O?#k*QJyU&%9m#U@9HSp?4-#S&{-!Ix|U^12cuC0Mu=%V%DHg7 z#HEJ`VL6Ej7W#9{EyC#tB96qLS9tc_4qYqPzX&1cc-jK~Fl6eUx5e{sKr zXBdDUkTRGZga8?{Kzia($LU3|4fzS~L9!#zQIq|fPp&5q4@vNlkyaPE>cGU5)pcA* zK~E>&l9PX=QZ1UVYM*nJNwZlG?_El(#Vn`X6ksT86KL&)U(6cXls{Xv8;mxnSQ{nM zF2$-}R!wTElQ5fA6;SXbd$1AHqCzZ0f@Xn*YdWS%*sEQ&zyM6nsj*igB@ZCE^>H>n zib3zBn%y#YI`iZy06K3?MSPY@e=Y7q3b2~4nvs0QvK-3-h|1r%T12)|UV^u+^leMN z4HCM87hrQjQ;AXQDo5iKVq%CxZN_SDV_jlSlbTAb5e-Vj~0 z`$>T7EZPxy0A(j-rd1O`_TwJ2sIz6M8kp`H8uQN#^F7!+ z9Sib5JPg$3a>mt93G`E_qA&*-hl+Dr(SUy8fR4H#9ikn8cG^D-1-lPI=>ce79FAAb zLhW2OjxU8P$_hF-n21K!xkc!rVrXKIG_aqnrWNW)b9AJDY2R>FJ#T55HPkf8BX}Rw zmLMyoA!kMw+g!6fz_0L#9-8VMSQaI-2FL}#GBin*HVIAkb1YqLFw99hHy&{qi`=Q} zAr(!71U&`!Ktz)?0>jG={fVy}_Tyw%nu#u)i9W22xezOn6@LF_woy%$T4x9Qp7?fS z`5axl`Z2oplPh0Y!NG%_M(YLvym)W%1Gie`)necu#ac0r8^MEwo|4f>9pV2Y z6?ZF^v)|_33Y3D)Svim1$CIn=>TfPYjo!Wg>xP0&X?!J<%9M{mD-!@X^O>xgPQXZ> zp;TPB*PuGaA<*nKmAnrUtXu}4kSTIDb^Qd)5yX#&!eyaPz~sNlrWVUQrHFx?20ZaZ zs)m?4eYR#zyl0uF?RH^d=fD5s|K*5)fAk)Ro>#fV>SJ|d#}ne7ya0$gVGghp#vyLJ zsv1Z?VjR7cBNp!`iGx`qnSqpkPh7kN38aJ%2tVRVozdn&o_Rr6Sv6V}TYsP8T1Fbx zC@8%T(P*^IHPiwC0Uu3^cX2_-V^!~{n5?y#W4I$_Qg|sF?|L#=0s9}{d?>FAbxR=2 z%E&wa4WX@G zrdvwLm7s#$vZ9VGun%w&_yUZ5v>bC(;k%r2a#Ms?yg44pqYydP*@qXls`$}1*Mz^7 z$StHR6!X=dOGB~eAKt!aGxWf<*W3uw>lkfe+eWuz4qNTf&1hsB_OR6-b$gxmjWcx2 zVc!`6+P6CWUS~Aoq6hyQ^=?EP%hh-51OBH};XuT4&;k6?Mms&E0#@2jJNf`$6#up|8IZ(z+&9&(b}ff>^(ae5nx#hU+p3{g1%xc^?c42hsJxf z*IX54p$k-Ibz&kf_@@Ns;`X?Np)e$kH=IONrhhthiT7TnAex7eJwiliuztbT$Aa!o z0lGUHx{n9l?_lu*2!tyR(7x~AOf%E;X%ObtvM)e?o$L3CLDq!w&MWq|tf@UBxoOsc z@?4gWi{<-+U+Dm`GQdEy?loF{xueS7TsqNA9+_B#(`d`JJnXAL0RplJXd|95*#SCI zfL9m6W+SuBkK)c1_qUl^eSxXOr;vmcZ>{)>(h_CL(~li(P)C^*j{3bVsDc;npWXLf zK;Ii| zScHxtjy|nJIDd@Fmn;}-E?LmbKu4sdTO(Ju=y05{asN=_7vpz<>lEq2PV(C=56CX{ zHzKLriga^JqNh8})x~+l2lBqeD8GU+qY(1(qt`T>R$308!v;gs#iRL#1_j?xx#i#& zrPf}PK0Zr}-Q7!c1O{2!X$a=ud&|!c#w}6tpM2#YjE-6)P^K z$xd?gS2=GmWev zfLuJhms4pcp1QhSy#y=srI+MYoCz)u3b;JbxU563t4KZ2q1TJ|so+EC5NF9q*DDNX{^KiTXk;%-m0A6T+lm*+!)ORIydGdNkiwx=-ikit@&@| zr8R1tRE={+1b|lvp*t#bMCrx72KD2oxr{?0{Igx^>tvS`M(VTm(`d}rn0?Baedhj| za{}$niB9_(KP7f25r=LB-lZ z>HUGm`XeN~pSKf7r@ia6_Y(qk4cQv9kDkMR-UzNU*-sh6>o{sOnrk$_2sCds&HbbK znLFrFC{^K{z=2 z`eS*=O{0L=rlBGGnEvrrUKM^%)q_7|mb_|XHKE%yUa_A5y6sRn)H2gvhF+RHD(% z8oP!y#!^Y(0(>@5IxO<}ESd!ILLqs&Ed=dyp=!W2fDHbsT5aH6ilmp(E;j{kUZPv+ zv}ovmD$xHrS7x)IN{mzq*IwC0Z{=pV3q#=g<_A!28`t(>lo^(^sz+go2d^4jv9?;5 zNv%lwQrG*FWl$|q!u*<=^(wgag8#Ucd~puxy59mzEACL0;+5y+cNibKTSCy7DG7yeHc{eKG*&!H=k=c&TM zD%L385XJU*LlpduONR{~4d1ui1sJt<0mi0XK;64#xifWrK#!1|$kvG4aS6Pu^)LLr z&V}CUD?7|&lHYEWza4E)3t7Vt&;p+~;s92=^J1nVU5WZIkkftig+>1I_CbCRYZ=ud zJc_sC%@4UR_?TV|JI2B>KHUP6|9mRYCi^AvVpm?{k$=qYL770Amta1h)!V|3!G(>E zl31Y#{PoiH;aS*V%j%G=!McqAjm64DXpxlnO#)_-1WZi=N8AErmyvbk)#3#cRg-_c zT%E4f$3F(UB-*U}NNqweU(p(>*vy{Y_noTf-zl$t@aUjFXta%kfPUKWX3)3!%#9as za@C0c=!oAoG~!=4;dw>0#+waONM;e@fibyM{y)y-b9ehc3X7;J0} z#f0CvASLeA@Zp$^c54l8ZwBPKD?F+fgC_9qt346FCtz8tAbZ7!_^XyR8-jb-7iIHXikAwzGCpUpL3c7mJR~cyf?;T7cCa3UTl;pi^fE6?WgiO&{e-X=1FnIufcFFee-7eV!v<>ja%`2rXqk6q4N$O z0PiQ$&?cB4^H~c1CyVgE2F$_Dgm1?Ml%^75h_m!YBwgw-r-<+ARo(38m`!2#IuvwB z6%3c&>cH{=6jHvTkdZlf#t;|n*kOV-yail*h5ej-E8id7%Ef(zvx2@-=;-| zLLfVd|6vN*L-VIaFJ^yhdLh~C#eD^^zlTj5POcN&`{XdV$9TSFwq~})ea9NjxpiaA z?2{#c$A!|rz`JZLd;x3yh-LDNWHC#;VjLfEeT?N!6<=Qxf^I7x37lp*aMls_!+2T0 z)f0nBvj$9>PZTB#UeS#{4SWvfYED=eL0=4qkfKt*S&in0^s4;y z%f}y>BZdevo+s?={7)gb4`1+HBwq;{xLH~BvP7&$FZHk4pKf|LQ~Kji9qn{EE?c}N z+!itVHcfVR}Xd z6C@Qb=qQMBGy+6Pq$hbrxnWIdE0Dec9TvFj8(%pLprg%hKZ?OpSx_MB8g zY;-%P0*b6r$fx2yfRW=cL*I?0sI<>y1x!B*QVw@Cmxs2~vu6<~qWD7%JD5x*0-+cBE$a!{@*_=j@`rRhG@4=}H~I%c(MTNxa-6x>As5SI`Uu+SK#rw1oTc)`#Nx|@%}ugk z&vA+~d`x{oM2Y2|;>~ic;uNmZ!jpx+gPcRJUzH>X-9OZE4RsnKpwj9SQAm4Ks)DZi zGzurJ*OIQnsruPs0S;QK;W$JnuOFBR8>Y>cOTF}jHd+SdcBw-bHvD3|6I$Y?M}U@4 zLz|DKEFUmeOG551^oH`!vXrwiq9R`!4GaPnl_)Vrz4PXjx=gyMhA3oXw=0E83UW$M zpK}IN?@r-SE^ApW%HNSj+UnLRO|su5>&j|d_=313ME4>5cy=lKCE;z?FktM#3{O)_ z1wo}e2$*g$19c9FQ~*+GDr-)IMCv#~cJpy%1cfKh3q6;>^5KOd>I5yIR4U-gVZ15- z>Q0lXd=O_cpWAs2h%8++PK%4iZpotYmtrS)q52X5m7i7`Y0~+u3?=EQB2217cqtZQ zYg?VBUQ<>nm7SFu*n#nF2g14s>lrP_dpg_Z z40NV?&A9EFF|++(mUtPuZDz(A@Vv{SRH39Ldz_)wHfvMvhU(-*THJ{)^IBPJln5>_ zny2!_p6c>5m5hdGcJ(|J#`kYG06_plFOGxF4^6_+*P%V~N;U~RXA9U&e`W2eR)IHq zEmc0?#_*MDPNo4{(J&0HaN=2c*WjBp%KreYR$X)U%h#MsUBLkqT(f&bxXv|OSba;_ zO(*=Kl9-@<62bRxy=$}6T@u9-rnyJ g*3U{EaO>>-1TcuG`q^I;D4^T_0H-{TgX01M03;gAeE Date: Tue, 3 Nov 2026 13:14:06 +0000 Subject: [PATCH 08/52] Land the devolved governments, misc and Trussell families with the thinktanks-b review applied The review approved all three with fixes, applied here as relabels; no printed value changes. - The Welsh Table 1 cells now say which are parts and which are totals, so an ingest cannot add Resource and Capital to their Total (finding 2). - The Welsh package key names its scope, the Budget's Barnett consequentials, so HM Treasury's figures for the same scope can share it (finding 3). - Real-terms settlement changes get their own metric, devolved_funding_real_terms_change, because AB2025's block_grant_effect means a policy's effect on the grant with the opposite sign (finding 4). - The Scottish Government's five rows join the Welsh ones in uk_devolved_govts (finding 7), leaving uk_fai_sfc empty. - CPS's PuFins figure names no Budget decision, so it is context with no proposed key (finding 1). - Trussell's row moves to a new uk_trussell_wpi family with AB2025's slug (finding 5). The README's slug table gains the three slugs. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_01LwQq6HuQUQ1obM9JvyhJbE --- .../harvest-uk-ab2026-2026-10-29/COUNTS.json | 34 +++- .../PROPOSED_KEYS.md | 3 +- .../harvest-uk-ab2026-2026-10-29/README.md | 2 + .../uk_devolved_govts/NOTES.md | 163 ++++++++++++++++++ .../uk_devolved_govts/claims_staged.jsonl.gz | Bin 0 -> 2566 bytes .../uk_devolved_govts/manifest.jsonl | 2 + .../uk_thinktank_misc/NOTES.md | 119 +++++++++++++ .../uk_thinktank_misc/claims_staged.jsonl.gz | Bin 0 -> 3963 bytes .../uk_thinktank_misc/manifest.jsonl | 5 + .../uk_trussell_wpi/NOTES.md | 29 ++++ .../uk_trussell_wpi/claims_staged.jsonl.gz | Bin 0 -> 927 bytes .../uk_trussell_wpi/manifest.jsonl | 1 + 12 files changed, 354 insertions(+), 4 deletions(-) create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_devolved_govts/NOTES.md create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_devolved_govts/claims_staged.jsonl.gz create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_devolved_govts/manifest.jsonl create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_thinktank_misc/NOTES.md create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_thinktank_misc/claims_staged.jsonl.gz create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_thinktank_misc/manifest.jsonl create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_trussell_wpi/NOTES.md create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_trussell_wpi/claims_staged.jsonl.gz create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_trussell_wpi/manifest.jsonl diff --git a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json index ca4c26489..038484aea 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json +++ b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json @@ -1,4 +1,13 @@ { + "uk_devolved_govts": { + "rows": 19, + "primaries": 2, + "claims_sha256": "745a5df6a7d0250658bafb2ad71ca49f90ec383a806633ef82adf381164d4440", + "sources": [ + "scottish_government", + "welsh_government" + ] + }, "uk_hmt_redbook": { "rows": 432, "primaries": 5, @@ -23,10 +32,29 @@ "ifs" ] }, + "uk_thinktank_misc": { + "rows": 12, + "primaries": 5, + "claims_sha256": "15a0efea0765f575d08a7184f1f7ca60edc20d53b3866edbe596638283ba3613", + "sources": [ + "citizens_advice", + "cps", + "institute_for_government", + "learning_and_work_institute" + ] + }, + "uk_trussell_wpi": { + "rows": 1, + "primaries": 1, + "claims_sha256": "8589f64e0c03ebaf6ab7f81e888577e89f723dbe26301537b57c5968a6679476", + "sources": [ + "trussell_wpi" + ] + }, "_totals": { - "rows": 1281, - "restated": 220, - "families": 3, + "rows": 1313, + "restated": 227, + "families": 6, "pinned_on": "2026-10-29" } } diff --git a/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md b/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md index 83eada253..5c13495ac 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md +++ b/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md @@ -4,7 +4,7 @@ Generated by `pipeline/ab2026_proposed_keys.py` from the staged families; don't A row whose scope matches no registered measure has `measure_key: null` and names one of these keys in `conditions.proposed_measure_key` (README rule 2). A key is a measure with no Table 6.1 line (`ab2026__`), a package (`ab2026__package_`) or an option (`ab2026_option__`). Each family's NOTES.md gives the key's scope as Table 6.1 lines or the producer's own words. Tranche 2 registers each key in `data/uk/ab2026_measures.json`, or folds it into a registered key, before the ingest. -24 proposed keys (9 measures, 15 packages, 0 options) on 71 rows. +25 proposed keys (9 measures, 16 packages, 0 options) on 83 rows. | proposed key | kind | rows | families | producers | |---|---|--:|---|---| @@ -14,6 +14,7 @@ A row whose scope matches no registered measure has `measure_key: null` and name | `ab2026__gas_standing_charge_moved_to_unit_rates` | measure | 3 | uk_ifs_ab2026 | ifs | | `ab2026__minimum_wage_uprating_april_2027` | measure | 2 | uk_hmt_redbook | hm_treasury | | `ab2026__mod_counter_fraud_strategy` | measure | 1 | uk_hmt_redbook | hm_treasury | +| `ab2026__package_barnett_consequentials` | package | 12 | uk_devolved_govts | welsh_government | | `ab2026__package_bbb_equity_community_lender_and_venture_debt` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_breathing_space_for_households` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_budget_barnett_consequentials` | package | 4 | uk_hmt_redbook | hm_treasury | diff --git a/sources/harvest-uk-ab2026-2026-10-29/README.md b/sources/harvest-uk-ab2026-2026-10-29/README.md index 9c456d00b..333a73d15 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/README.md +++ b/sources/harvest-uk-ab2026-2026-10-29/README.md @@ -57,6 +57,8 @@ The AB2025 harvest README's rules 1–10 and its row contract apply, with the AB | resolution_foundation | Resolution Foundation | uk_rf_ab2026 | | ifs | Institute for Fiscal Studies | uk_ifs_ab2026 | | tax_policy_associates, centax, cpag, ippr, niesr, fraser_of_allander, scottish_fiscal_commission, cps, institute_for_government, age_uk, … | other producers, each in the AB2025 family of the same name where one exists | uk_tpa, uk_centax, uk_cpag, uk_ippr, uk_niesr, uk_fai_sfc, uk_thinktank_misc, … | +| scottish_government, welsh_government | the devolved governments' responses | uk_devolved_govts | +| trussell_wpi | Trussell, as in AB2025 | uk_trussell_wpi | | commercial firms and platforms | worked examples (mode 3, #63) | uk_cases_commercial | Macro calls by banks and City forecasters belong to the #55 lane and are not staged here (AB2025 rule 10). diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_devolved_govts/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_devolved_govts/NOTES.md new file mode 100644 index 000000000..4e1b51211 --- /dev/null +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_devolved_govts/NOTES.md @@ -0,0 +1,163 @@ +# uk_devolved_govts — devolved governments, Autumn Budget 2026 + +**Working files.** `stage.py`, which writes `manifest.jsonl` and `claims_staged.jsonl.gz`, lives on branch `uk/ab2026-harvest-staging` at `harvest-staging/ab2026/lanes/thinktanks-b/uk_devolved_govts/stage.py`; this branch carries only NOTES.md, `manifest.jsonl` and `claims_staged.jsonl.gz`. The lane's local primaries folder (`.ab2026-primaries/`) was never committed: the repository carries no primary bytes, so fetch each primary from its `manifest.jsonl` URL and check its sha256. + +**19 rows from two primaries.** Rows 0–13 (0-based) are the Welsh Government's (source `welsh_government`); rows 14–18 are the Scottish Government's (source `scottish_government`). The Scottish rows were staged in `uk_fai_sfc` on 29 October. On 3 November the thinktanks-b review's finding 7 moved them here, so that both devolved governments sit in one family; `uk_fai_sfc` is left with no rows and is not carried (see the harvest README). Conventions follow the AB2025 `uk_fai_sfc` block-grant rows (`source_model: arithmetic`). + +## Changes on 3 November 2026 (thinktanks-b review) + +- **Finding 2:** the Welsh Table 1 Resource and Capital cells (rows 1–7) carry `conditions.row_type: "measure_component"` with their `component`; the four Total cells and the £448m headline (rows 0 and 8–11) carry `row_type: "total"` and no `component`. An ingest can no longer add parts to their total. +- **Finding 3:** the Welsh package key names its scope, not the nation: `ab2026__package_barnett_consequentials`, shared with HM Treasury's figures for the same scope in `uk_hmt_redbook`. Nation is `conditions.geography`. +- **Finding 4:** the seven real-terms settlement rows (Welsh rows 12–13, Scottish rows 14–18) use `proposed_metric: devolved_funding_real_terms_change`. AB2025's `block_grant_effect` means a policy's effect on the grant, positive for an increase; these rows are a year-on-year real-terms change in the whole settlement, positive for a reduction. +- **Finding 7:** the Scottish Government's rows moved here (above). Both slugs are now in the harvest README's slug table. +- **`component` on context rows:** the seven real-terms rows named their budget (Resource, Capital and so on) in `component`, which rule 9 reserves for parts of a measure. They now use `budget`, by the reasoning of the review's finding 6. +- **Attribution (coordinator check under finding 3):** the Welsh rows stay `own`. The page cites no source, and the coordinator's settled calls do not change this; HM Treasury's own consequentials are staged as HM Treasury's in `uk_hmt_redbook`. + +No primary was re-opened on 3 November: the session's network policy refused gov.wales and gov.scot. Every change relabels fields; no printed value changed. + +## Coverage + +| Primary | Position | Rows | Coverage | +|---|---|---:|---| +| Welsh written statement | Paragraph before Table 1 | 1 | £448m cumulative headline (`total`) | +| | Table 1, Resource row | 3 | 199 / 165 / 146 (£m); 2029-30 is N/A (components) | +| | Table 1, Capital row | 4 | -27 / -19 / 2 / -18 (£m) (components) | +| | Table 1, Total row | 4 | 172 / 146 / 148 / -18 (£m) (`total`) | +| | Paragraphs after Table 1 | 2 | 0.5% resource and almost 10% capital real-terms reductions (context) | +| Scottish Government news release | Opening paragraph and paragraph 2 | 5 | £650 million; 1.3% and £550 million (resource); 1.5% and £100 million (capital), all 2027-28 real-terms reductions (context) | + +By `measure_key`: **null 19**. Twelve rows carry the proposed key `ab2026__package_barnett_consequentials` (5 `total`, 7 `measure_component`); the seven real-terms rows are context with no proposed key. By metric: `block_grant_effect` 12, `devolved_funding_real_terms_change` 7. By unit: `gbp` 15, `percent` 4. Attribution: `own` 19; benchmark `different_model` 19. + +## Proposed measure key + +`ab2026__package_barnett_consequentials` (12 rows): the change in a devolved government's block grant from UK Government decisions at AB2026, by resource and capital, as the producer prints it; nation in `conditions.geography`. The Welsh rows exclude employer-pension-contribution savings ("Changes to the Welsh Government Settlement"; "an additional £448m of funding over the current spending review period", 2026-27 to 2029-30), as HM Treasury's do. No Table 6.1 line covers it: the consequentials follow from England's spending lines. + +## Welsh Government + +### Access + +The [written statement](https://www.gov.wales/written-statement-welsh-government-response-uk-autumn-budget-2026) +was first published and last updated on 28 October 2026. It was fetched fresh +with `curl -sSL` and the exact requested User-Agent: + +```text +Mozilla/5.0 (Macintosh; Intel Mac OS X 14_0) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/17.0 Safari/605.1.15 +``` + +HTTP 200; response Date 2026-10-29T15:22:18Z. The lane saved the unaltered +HTML and response headers in its local primaries folder. BeautifulSoup read the page's paragraphs and actual HTML +table cells, not chart geometry. `shasum -a 256` gives: + +```text +87490552c36d8ffd5787db18127704e467d93599e5708627adb6da7dcfe17c7c +``` + +The fetched primary is 26,908 bytes. No local copy was provided, no fallback +was used, and there is no earlier assigned version against which to establish +a change. No requested Welsh page was blocked or unreached. + +### Attribution + +These are the Welsh Government's own published settlement accounting and real-terms comparisons. The statement does not credit another producer for these calculations, so no HM Treasury attribution is inferred. They are prospective funding changes rather than administrative outturns or counts. The source does not name a formal model; `arithmetic` describes the settlement accounting. + +### Welsh numeric and period conventions + +- GBP values are the printed £m cells or headline multiplied by one million. + Signed table values are retained: positive means an increase in Welsh + funding. No totals or differences are derived. +- Table columns are fiscal years; `period` is the end year and `fy` retains + the printed year. The £448m headline is cumulative over the four displayed + years, `period: 2030`, with that horizon stated explicitly; it is not an + annual 2029-30 amount. +- Resource real-terms reduction “next financial year” is 2027-28 given the + publication date. Capital reduction is explicitly by 2029-30. Percentage + values retain the printed magnitude; conditions say positive means a + reduction. “Almost 10%” is preserved in `value_raw`, the upper endpoint + 10 is marked `range_high`, and the qualifier is retained. +- The monetary table has no real-terms basis stated; no deflation is + applied. The percentage context rows retain real terms with no invented + price year. All rows concern Wales. +- `value_kind: point` means a single published figure, not a separately + identified central-model scenario. No row uses `central`. +- Every quote is an exact page fragment or table cell, contains `value_raw` + and is fewer than 15 words. Tables are read by row and then column. + +### Welsh rows not staged + +- Resource 2029-30 is `N/A`, not a number and not zero. +- OBR GDP growth of 1.2% in 2026 and 1.0% in 2027 is restated macro context, + outside this assignment's devolved-consequentials scope; no own Welsh + estimate of a Budget measure is asserted. +- The £100 council-tax-band A-D payment is a Welsh implementation + announcement and policy parameter, with no implementation date, reach, + aggregate cost or estimated household gain printed. It is not attached + to the England-only Treasury line 1 key and is not fabricated into an + impact row. +- The £14 billion rail commitment and £302m already allocated over the + spending review are existing commitments, not newly estimated effects of + AB2026; the statement supplies no isolation of a Budget measure's effect. +- Dates, council tax band letters, and rounding caveats are not estimates. + There are no charts needing bar readings. No inaccessible Welsh page or + unavailable local copy is being used as evidence. + +## Scottish Government + +The [news release](https://www.gov.scot/news/uk-budget-delivers-real-cut-to-scotlands-funding/) ("UK Budget delivers real cut to Scotland's funding", 28 October 2026) was fetched fresh on 2026-10-29 with the same User-Agent; HTTP 200, 27,662 bytes, sha256 `f3d0183e470f711b6821e4956e108c5d17210a392e6a6e471eccf297c807a6ac`. + +- **Sign:** the primary prints unsigned magnitudes and describes reductions. Values retain those positive magnitudes, with `conditions.sign_convention: "positive = magnitude of reduction in Scottish Government funding"`. No minus sign is invented. The three million-pound amounts are multiplied by 1e6; the percentages remain percentages. +- **Period:** all rows use fiscal-year end year `2028`, `time_basis: fiscal_year`, and `conditions.fy: "2027-28"`. +- **Prices and comparison:** all rows say real terms, with an unspecified price base year. The primary does not explicitly name its comparison baseline. `conditions.price_basis` and `comparison_basis` preserve those limits. No cash-basis conversion is made. +- **Point/central:** the page gives one reported estimate per figure, not a named central model scenario, so `value_kind: point`. The money amounts retain the producer's “around” qualifier in `conditions.precision`. +- **Attribution:** all 5 are Scottish Government `own` estimates because the article reports its funding assessment without naming a separate original source. `arithmetic` follows the AB2025 family's funding convention and does not assert access to an unpublished model. None is an administrative outturn or count. +- **Baselines:** `baseline_policy: null`; no proposed baseline. These are context funding projections, not a comparison of a reform against a specified alternative policy. +- **Keys:** **`measure_key: null` — 5 rows**. No registered key is used, and **no proposed measure key** is introduced. The article does not isolate a set of Table 6.1 decisions or establish that its year-on-year real-terms funding assessment matches a registered package. `reform_hint: null` records that it is context. None of the 98 Table 6.1 lines is asserted to be the exact scope of these figures. + +### Scottish rows not staged + +- Every FAI figure: the primary refused the required fresh fetch, and no local fallback was supplied. The family therefore has no read primary evidence for FAI's quantitative or qualitative claims. +- SFC: no qualifying publication dated 28–29 October was found. Its earlier forecasts and funding commentary are not recast as post-Budget figures. +- The Scottish Government's £100 household energy-support commitment: it is a payment-policy parameter, not a separate quantified aggregate or model estimate. The article does not specify Scottish eligibility, rollout timing or a fully defined Scottish implementation; the registered Table 6.1 line 1 describes England's council-tax rebate. No England measure key is attached to Scotland's unspecified implementation. +- The £300 energy-bill reduction refers to an earlier political promise, not an AB2026 estimate or a newly costed option. +- The repeated £650 million in the minister's quotation and repeated £100 commitment are not staged twice. +- Dates, earlier political reference years and qualitative references to welfare policies do not form numerical claims. No chart-only numbers, macro forecasts or third-party restatements were used. + +### Scottish Government primary issues + +The response labels the figures as a consequence of the UK Budget but does not distinguish changes caused by newly announced decisions from the inherited funding path and inflation. It supplies no derivation, explicit comparison baseline or deflator year. Consequently the five context rows cannot be compared directly with a cash-terms Barnett-consequentials table. The two component cash magnitudes are consistent with the printed total; none was recomputed. + +The HTML's JSON-LD `dateModified` is `2026-10-29T08:22:59Z`, while the visible publication time and `dc.date.modified` show 28 October 2026 15:45. No earlier local copy was provided, and the metadata do not establish that the numerical content changed. The manifest pins the sole fetched version and records both timestamps; there is no invented second version. + +## Verification + +- **3 November:** the contract checker reports 19 rows, 0 problems; a re-run of `stage.py` reproduces `manifest.jsonl` and `claims_staged.jsonl.gz` byte for byte (COUNTS.json pins the gzip's sha256). The fixes relabel fields only, so no spot check was repeated; the lanes' checks below stand. +- **Welsh rows (29 October).** Ten rows selected with `random.Random(2302026).sample(range(14), 10)` +were checked against the saved, hash-verified HTML, reselecting the table +row/column or paragraph. Row numbers below are one-based. Every check +confirmed the number, sign/magnitude convention, unit and end-year period. + +| Row | Reopened locator | Printed value | Normalized value and period | +|---:|---|---|---| +| 4 | Table 1 Resource, 2028-29 | 146 | GBP 146,000,000; 2029 | +| 5 | Table 1 Capital, 2026-27 | -27 | GBP -27,000,000; 2027 | +| 6 | Table 1 Capital, 2027-28 | -19 | GBP -19,000,000; 2028 | +| 7 | Table 1 Capital, 2028-29 | 2 | GBP 2,000,000; 2029 | +| 8 | Table 1 Capital, 2029-30 | -18 | GBP -18,000,000; 2030 | +| 9 | Table 1 Total, 2026-27 | 172 | GBP 172,000,000; 2027 | +| 11 | Table 1 Total, 2028-29 | 148 | GBP 148,000,000; 2029 | +| 12 | Table 1 Total, 2029-30 | -18 | GBP -18,000,000; 2030 | +| 13 | First paragraph after table/rounding footnote, last sentence | 0.5% | 0.5 percent reduction; 2028 | +| 14 | Second paragraph after table, second sentence | almost 10% | Upper endpoint 10 percent reduction; 2030 | + +The headline and all remaining numeric cells were also read directly when +transcribing. Every row's quote contains its raw value. Scope review found +no assigned registry key: the proposed Welsh package mapping remains +explicitly unresolved, and the context rows carry no proposal. +- **Scottish rows (29 October, in `uk_fai_sfc`).** Because the family contains fewer than 10 rows, all 5 were checked. A reproducibly shuffled order (`random.Random(20261029)`) gave rows **3, 2, 1, 4, 5** (now rows 17, 16, 15, 18, 19 of this family, one-based). The saved primary was re-opened at every locator; its hash was also checked against each row's publication hash. + +| Row | Primary locator | Verbatim figure | Checked normalization | Sign and period | +| ---: | --- | --- | --- | --- | +| 3 | Paragraph 2, sentence 1; HTML 348 | £550 million | 550,000,000 `gbp` | Positive magnitude of reduction; FY2027-28, period 2028 | +| 2 | Paragraph 2, sentence 1; HTML 348 | 1.3% | 1.3 `percent` | Positive magnitude of reduction; FY2027-28, period 2028 | +| 1 | Opening paragraph; HTML 347 | £650 million | 650,000,000 `gbp` | Positive magnitude of reduction; FY2027-28, period 2028 | +| 4 | Paragraph 2, sentence 2; HTML 348 | 1.5% | 1.5 `percent` | Positive magnitude of reduction; FY2027-28, period 2028 | +| 5 | Paragraph 2, sentence 2; HTML 348 | £100 million | 100,000,000 `gbp` | Positive magnitude of reduction; FY2027-28, period 2028 | diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_devolved_govts/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_devolved_govts/claims_staged.jsonl.gz new file mode 100644 index 0000000000000000000000000000000000000000..0cf67caeb647877e068d2758e4ce1f33b613c3e9 GIT binary patch literal 2566 zcmV+h3iCCBO5l@(xiirDbIFsa2geO?yFJmgNJwKvK;UxhNLnc|0d9ess zgqQmw2{?9%4c>RyZ+?_3GXjg#NUWJ8DNFDefyNsVh@v+JaNSZED#?CJ^V*1*pko@;zJZ#t-x+D_>KF)gQ1M=dE&Lc*g zAF*l7{3T7eim~w}l?lr-lu5C7IgHF{q{tb;HVdY(mtR<jH+BR@0#gJ(K$}R z^HSWRylkB^yebnJ)wWQ4kziX?qcH&t1J=1=TKc&~zM-%*Uo$PoCD3#~?iY z!b|fxM007Z_Z{l{tc+o4+61&PiVhoR&>K$iD0D|NA@Rf;Dyfb*!sD0)9NI@~GH2k; ztWsder2W*fE@p$N;Pl9O}24i>X26JcX3|)6HVxx&0 z%+ANN85_HU>162AU_J?`8~7n}T;KIKQ&LMNcR3O^$T)*O-h7RjNIg}K)LY>|y~Spr zUO7-NzwQ-ztyko=o*n@M?psU`dDlD%bRy5=JffH@H|;%1S4FBrLv%j;@cvO4>WdgW#-Q*YCYNEm zhlFdmt`YQZ+MXyv%T6lQTb!OYPKWNJ&#{c2+MjN-iw%D<71tc5eq841F@;##c&O}q6wY`T3wS9ELtsEPZ_P#BSkC5 zr$uXPXq}7?6s=_*K0dTg#`~caOz+vVT~0jtS%b~(5`u!7usffeWZB*1D{ zgDqI?YVg5=)h-6N3RY)k7lT^{tFzxeE1KX*fYq)ATd>--;DZILT?%d!tV6pH+%8s! zdqH*hG?0qlrLjo0NVVUwIY^}Huew>JwpOc74-~3p8a~2Wb-HJ*I(;;-+Fexp#Z(K; z)89G05_aG8!5&C*hj!nz6lGf|!XCxD(mXsyqghhc+M;33);jAJ|mb1{wfi%hkdKXgVP^ zv=I1%<3IbEn1e3RJ^JKLWm?VOTb@T@JZ}r8sW`SZ8&nph$GH0a79+)h4 zl5M^&>+2H)nzLPwNpx_p43`GfDoq_)X;xo>BE@Od{X#p)i&yc!h zX8GU0{w0%*4O67;4tPtH^I8j_>`8Ddj6?@4MXDP}srI^-}K@*h!!p$ApFe96pw-)k zt4OaI>HFezSFPGVzwhHVs3^-&TbbI*)Kc_lDT)&&Fg*^r4ZU4Q#3p5zYP2lxQ_0~3Oce7Pwi_)(IDO6+Q zXZ3B;Mg+IS)6EN;j-@~y-@8CPxTk|ofu6QBwfwaFd>npm)u#R?S=-X|TS`-B@DQcx c4wN2tW$Fy}txW5R`|!j60jvU|w7H1@0B~9GMF0Q* literal 0 HcmV?d00001 diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_devolved_govts/manifest.jsonl b/sources/harvest-uk-ab2026-2026-10-29/uk_devolved_govts/manifest.jsonl new file mode 100644 index 000000000..99d59358a --- /dev/null +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_devolved_govts/manifest.jsonl @@ -0,0 +1,2 @@ +{"access": "direct", "bytes": 26908, "date": "2026-10-28", "doc_type": "html", "note": "Fresh curl -sSL fetch with the assigned Safari User-Agent; HTTP 200, completed 2026-10-29T15:22:18Z (HTTP Date header). SHA256 is of the unaltered HTML bytes saved as .ab2026-primaries/uk_devolved_govts/welsh-response.html. Publication and last-updated dates both print 28 October 2026. No assignment local copy was supplied.", "publisher": "Welsh Government", "retrieved": "2026-10-29", "sha256": "87490552c36d8ffd5787db18127704e467d93599e5708627adb6da7dcfe17c7c", "title": "Written Statement: Welsh Government response to the UK Autumn Budget 2026", "url": "https://www.gov.wales/written-statement-welsh-government-response-uk-autumn-budget-2026"} +{"access": "direct", "bytes": 27662, "date": "2026-10-28", "doc_type": "html", "note": "Fresh exact prescribed browser-UA curl fetched HTTP 200 at 2026-10-29T15:22:47Z. SHA256 covers the fetched HTML bytes. Visible publication timestamp is 28 October 2026 15:45. JSON-LD dateModified is 2026-10-29T08:22:59Z, while dc.date.modified repeats 2026-10-28 15:45; no earlier copy was supplied, so a numerical change cannot be established. Five rows, read from opening body paragraphs 1–2. Local file .ab2026-primaries/uk_fai_sfc/scottish_government_budget_response.html.", "publisher": "Scottish Government", "retrieved": "2026-10-29", "sha256": "f3d0183e470f711b6821e4956e108c5d17210a392e6a6e471eccf297c807a6ac", "title": "UK Budget delivers real cut to Scotland’s funding", "url": "https://www.gov.scot/news/uk-budget-delivers-real-cut-to-scotlands-funding/"} diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_thinktank_misc/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_thinktank_misc/NOTES.md new file mode 100644 index 000000000..a9edd248b --- /dev/null +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_thinktank_misc/NOTES.md @@ -0,0 +1,119 @@ +# Smaller producers: Autumn Budget 2026 + +**Working files.** `stage.py`, which writes `manifest.jsonl` and `claims_staged.jsonl.gz`, lives on branch `uk/ab2026-harvest-staging` at `harvest-staging/ab2026/lanes/thinktanks-b/uk_thinktank_misc/stage.py`; this branch carries only NOTES.md, `manifest.jsonl` and `claims_staged.jsonl.gz`. The lane's local primaries folder (`.ab2026-primaries/`) was never committed: the repository carries no primary bytes, so fetch each primary from its `manifest.jsonl` URL and check its sha256. + +Staged from five primary HTML responses published on 28 October 2026. Sources are `cps`, `institute_for_government`, `learning_and_work_institute` and `citizens_advice`; the manifested Age UK response contributes zero rows. There are **12 rows and 5 primaries**. `stage.py` embeds the readings and recreates the manifest and deterministic gzip using only Python's standard library. + +## Changes on 3 November 2026 (thinktanks-b review) + +- **Finding 1:** CPS's £3.1 billion PuFins row is context. It names no Budget decision, line or year and ties the figure to a CPS paper from before the Budget, so it carries no proposed key; `reform_hint` is null and `conditions.measure_type` names it. `ab2026__package_cps_public_financial_institutions` is withdrawn. +- **Finding 5:** Trussell's row and primary moved to the AB2026 family `uk_trussell_wpi`, with source `trussell_wpi`, as AB2025 staged Trussell's Budget responses. This family now has 12 rows and 5 primaries. +- **Finding 6:** IfG's count of one tax measure names the measure it counted in `conditions.subgroup`, not `component`, which rule 9 reserves for parts of a measure. + +No primary was re-opened on 3 November (the session's network policy refused every producer's site); every change relabels fields, and no printed value changed. The 29 October record below is updated for the new row set. + +## Access recipe and primary hashes + +Every primary was fetched afresh on 29 October 2026 using `curl -sSL -A "Mozilla/5.0 (Macintosh; Intel Mac OS X 14_0) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/17.0 Safari/605.1.15"`; every response was HTTP 200. The lane kept the HTML bytes, headers and text extracts in its local primaries folder. Hashes were checked using `shasum -a 256`. BeautifulSoup extracts were used to read the page's own body text; hashes refer to HTML, not extracts. There are no PDFs or spreadsheets in this family. + +No assigned local copies or earlier document versions were supplied. There was no blocked fetch, fallback or observed replacement. Fetch-completion timestamps are recorded in each manifest note because HTML can change. + +| Primary | Fetch completion (UTC) | Bytes | Rows | SHA256 | +|---|---|---:|---:|---| +| [CPS verdict](https://cps.org.uk/media/post/2026/a-dangerous-continuation-of-labours-tax-and-spend-fantasy-cps-verdict-on-the-budget/) | 2026-10-29 15:22:34 | 70,197 | 4 | `2c832e011866eb9f63f20597e7bf43bc01d44b6b1b5be43eee866ad6fc5ecd24` | +| [IfG commentary](https://www.instituteforgovernment.org.uk/comment/john-healeys-underwhelming-budget) | 2026-10-29 15:22:31 | 153,352 | 6 | `ecf19f14b4f808846fb8230b28a3244b0b65812484735842f9ab537a50195c69` | +| [Age UK response](https://www.ageuk.org.uk/latest-press/age-uk-response-to-the-budget/) | 2026-10-29 15:22:31 | 105,862 | 0 | `3035a8bbbb0f023dc24e805b430052163aa0dbfce3c975a6e0a0568983db6bf0` | +| [Learning and Work Institute response](https://learningandwork.org.uk/responding-to-autumn-budget-2026/) | 2026-10-29 15:22:31 | 122,809 | 1 | `c7fe077a0d9f4784aa4c91c906bcefa087cc9a4ece8b28026d24347c2a8b5dcf` | +| [Citizens Advice response](https://www.citizensadvice.org.uk/about-us/media-centre/press-releases/citizens-advice-responds-to-chancellors-budget/) | 2026-10-29 15:23:24 | 22,671 | 1 | `6a10336a67fc058512a4ccd9d0e80a27470fee8ba5775690fb1fcba39db67385` | + +Citizens Advice's press index linked its dated response. The other four URLs were assigned. CPS prints “Date Added: Wednesday 28th October 2026”; IfG prints “28 OCT 2026”; LWI prints “28 October 2026”; Age UK prints publication on 28 October 2026 at 04:14 PM and last update that date. + +## Coverage and conventions + +Reading order is CPS, IfG, Age UK, LWI, Citizens Advice, then top to bottom within each page. Only contiguous excerpts under 15 words are quoted. There are no chart readings or derived gaps, totals, ratios or annualisations. + +| Source | Rows | Own | Restated | +|---|---:|---:|---:| +| `cps` | 4 | 1 | 3 | +| `institute_for_government` | 6 | 3 | 3 | +| `learning_and_work_institute` | 1 | 0 | 1 | +| `citizens_advice` | 1 | 1 | 0 | +| `age_uk` (manifest only) | 0 | 0 | 0 | +| Total | 12 | 5 | 7 | + +| Measure key | Rows | Scope | +|---|---:|---| +| `ab2026__household_energy_payment_100_bands_a_to_d` | 1 | IfG share receiving the £100 council-tax rebate for bands A–D; Table 6.1 line 1 | +| `null` | 11 | context and document-tally observations; none carries a proposed key | + +The sole registered key is attached to the specific rebate's coverage, not to the entire cost-of-living package. IfG's exact identification of the £100 council-tax rebate for bands A–D identifies line 1; it gives no contrary scope. Its household denominator's geography is unspecified and is not inferred. Context forecasts, outturns and document counts, including CPS's PuFins assertion, have null measure keys and null reform hints. + +- GBP values preserve printed signs and cash magnitudes. “Tops £3 trillion” is the printed strict lower bound of debt stock. The six-month £61 billion debt-interest total is not multiplied to a year. Fiscal headroom is positive headroom; PuFins' £3.1 billion is a positive debt increase. No sign is reversed. +- Percentages remain percentages. “Almost 100” retains the printed upper threshold 100; no exact policy count is inferred. “Around 80%” retains the qualifier in conditions. +- `point`, rather than `central`, records each single published observation/forecast: none is labelled a central scenario. Thresholds use `range_low` or `range_high`. The 2% real-income growth is `cumulative` over the printed comparison window, not a 2% annual rate. +- Only IfG's fiscal-rule probabilities name a fiscal year: “2029/30” is normalised to `fy: 2029-30`, period 2030, `fiscal_year`. The income projection prints calendar 2030 and uses `annual` without `fy`. Other observations use the publication-year point-in-time anchor 2026 and preserve “last six months”, “this winter”, “each year”, “last 12 months”, or unspecified horizon explicitly. They do not claim a fiscal/calendar scoring year that the producer did not print. The PuFins annual-rate assertion has medium confidence because no scoring year or fiscal/calendar basis is supplied. +- Real income remains real; all other price bases are left unspecified where the page does not identify one. No baseline proposal is needed for these current observations and explicitly named context comparisons. +- CPS's own PuFins assertion uses `source_model: unstated` because it does not identify its calculation. Official debt/outturn context uses `administrative_data` and `administrative_fact`. CPS headroom and IfG fiscal probabilities/income forecasts are `restated`, `obr_efo_forecast`, `different_model`. The precise original release behind the unsourced CPS figures is not identified, so these originator assignments are explicit in rows. +- IfG's two counts are its own `arithmetic`, `different_model` observations. Its unsourced rebate share uses `source_model: unstated`, `own`, `different_model`. LWI's rounded NEET figure is treated as restated official labour-market context, `administrative_data`, `administrative_fact`; its response gives neither a quarter, age range nor a citation. Citizens Advice's unsourced bill forecast is treated as its own assertion with `source_model: unstated`, with missing method/geography acknowledged. These two method-unspecified estimates and CPS's PuFins assertion have medium confidence; the other nine rows have high confidence. + +## Proposed keys, metrics and units + +No measure key is proposed. CPS's PuFins row is context (see the 3 November changes above). + +Proposed metrics (counts and exact meaning): + +- `public_debt_level` (1; GBP): CPS's current national-debt stock lower bound; no series definition is supplied. +- `debt_interest_cost` (1; GBP): CPS's total debt interest over the last six months, not the change caused by a measure. +- `fiscal_headroom` (1; GBP; existing README proposal): CPS's current post-Budget buffer against the fiscal rule; target year unspecified. +- `public_debt_change` (1; GBP): CPS's asserted annual debt increase through PuFins, a context row; period limitations above. +- `measure_count` (2; proposed unit `measures`; AB2025 family precedent): IfG's approximate full scorecard count and exact count of tax measures costing/saving more than £1bn in the last forecast year. +- `fiscal_rule_probability` (2; percent): current and last-budget OBR probabilities of a current-budget surplus in 2029/30, as reported by IfG. +- `affected_share` (1; existing README proposal): IfG share of all households receiving the rebate (percent). +- `affected_count` (1; persons; existing README proposal): LWI's rounded current count of young people not in education, employment or training. +- `average_energy_bill_change` (1; GBP/month; AB2025 IFS/RF proposal spelling): Citizens Advice's forecast monthly household energy-bill increase, excluding any assertion that this is a Budget measure effect. + +Registered `real_income_growth` accounts for the remaining row (IfG's restated OBR 2% cumulative income growth). Proposed unit `measures` is used only for the two counts. Other units are registered. There are no proposed baselines. + +## Not staged and primary limitations + +- **Age UK: zero rows.** Its own response has no numeric Budget estimate. The exact “On Income Tax threshold.” subsection was read; it welcomes protecting low-income pensioners from cash tax demands but prints no number. Dates, contact numbers, navigation “Nearly 4,000 flu mortalities” and the unrelated £2 donation link are not Budget claims. +- IfG's £100 rebate is a bare policy parameter; its 80% coverage is staged separately. The £1bn cutoff defines its count of one tax measure and is preserved in conditions rather than misrepresented as that measure's costing. Other dates and the 1991 council-tax valuation year are context labels, not model outcomes. +- LWI mentions extra Mayoral apprenticeship funding and OBR employment effects but prints no amount or employment effect. Its “one million” NEET level is retained as context. Navigation cards and unrelated articles are not harvested. +- Citizens Advice gives no source, exact winter dates, geography or method for £60/month. Those are explicit limitations, not supplied from secondary material. +- CPS does not link its “paper released last week”. It is outside the assigned 28–29 October post-Budget window, and no unseen figures from it are staged. The comparison between six-month debt interest and an annual defence budget prints no second amount or exact ratio, so none is derived. +- No restated figure is silently removed: the seven official/OBR context rows are staged with `restated` and an originator note. No PolicyEngine restatement was identified. +- No pages were unreached. No charts contain unstaged numeric labels in these primaries. + +Potentially ambiguous source wording is retained: IfG describes the £100 rebate as reaching “around 80%” of households but does not identify the denominator's geography. No UK/England assumption or correction is made; the row identifies the precise rebate and flags the missing denominator scope. CPS's PuFins claim lacks components, annual dates and method, so it stays context; its £15 billion headroom lacks a target year. Age UK's Pension Tax Allowance paragraph contains no number despite the assignment's pointer. These gaps have not been repaired using other sources. + +## Verification + +- **3 November:** the contract checker reports 12 rows, 0 problems; a re-run of `stage.py` reproduces `manifest.jsonl` and `claims_staged.jsonl.gz` byte for byte (COUNTS.json pins the gzip's sha256). The fixes relabel fields only, so the lane's checks below stand for the rows that remain. +- **29 October (13 rows, including Trussell's, now in `uk_trussell_wpi`):** + + +Command: + +```text +PYTHONPATH=. uv run -q python pipeline/ab2026_harvest_contract.py sources/harvest-uk-ab2026-2026-10-29/uk_thinktank_misc +sources/harvest-uk-ab2026-2026-10-29/uk_thinktank_misc: 13 rows, 0 problems +``` + +Re-ran `python3 sources/harvest-uk-ab2026-2026-10-29/uk_thinktank_misc/stage.py`; claims SHA256 before and after was **`d677dde1f96e629097696f5d17b575c8f1d34baaca42b96297cd46be773669ea`**. The gzip header has mtime zero and no original filename. Manifest byte counts and all six primary hashes were checked against saved HTML. All 13 quotes were verified as contiguous text in their pinned primary and under 15 words; each contains its `value_raw`. + +Random sample was `random.Random(230).sample(range(13), 10)`, using zero-based indexes and reporting one-based row numbers below. For each, the saved primary was reopened, its SHA256 checked, and the paragraph at the locator read to verify the number, sign, unit and period. The three remaining rows (CPS debt interest, IfG count of one, Trussell share) were also checked for quote/value correspondence and basis. + +| Sample row | Primary and locator | Rechecked value, sign, unit and period | +|---:|---|---| +| 12 | Citizens Advice, second Moriarty paragraph | £60/month positive bill increase; winter forecast, no exact dates; publication-time 2026 anchor, no FY | +| 3 | CPS, final Halligan paragraph | £15 billion positive fiscal headroom → 15e9 GBP; target year absent; 2026 statement anchor | +| 5 | IfG, opening body paragraph | “almost 100” policies; upper threshold 100 measures, no exact count invented; 2026 scorecard | +| 9 | IfG, energy-support first paragraph | “around 80%” rebate coverage; percent remains 80; this winter, no FY; geography unspecified | +| 1 | CPS, national-debt paragraph | “tops £3 trillion”; positive stock strict lower bound → 3e12 GBP; current at publication in 2026 | +| 7 | IfG, fiscal-picture second paragraph | 56% current probability; FY 2029/30 → `fy: 2029-30`, period 2030 | +| 8 | IfG, same paragraph, last-budget clause | 59% previous-budget comparator for that fiscal rule; FY 2029/30 → period 2030; no change calculated | +| 10 | IfG, energy-support first paragraph, final sentence | 2% higher real incomes in calendar 2030 versus before pandemic; cumulative, positive; annual time basis, no FY | +| 4 | CPS, Herring's PuFins paragraph | £3.1 billion positive debt increase each year → 3.1e9 GBP; no start/end year; 2026 statement anchor with annual horizon retained | +| 11 | LWI, Evans's main quotation | “one million” young NEET people → 1e6 persons; current context at publication 2026; no quarter invented | + +Scope review found one non-null key: IfG's specifically identified bands A–D £100 rebate, Table 6.1 line 1. No broader support-package figure is attached to that key. (On 3 November the CPS row lost its proposed key; see above.) No git state-changing commands, commits or pushes were run. diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_thinktank_misc/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_thinktank_misc/claims_staged.jsonl.gz new file mode 100644 index 0000000000000000000000000000000000000000..0aa281cf2bd37508a44f6dc50d8c12fa18421480 GIT binary patch literal 3963 zcmV->4}|a^iwFP!00002|Lt7eZrsKdzVA~E7BJKROKJaqD)ho}(j;xt7)gp=5<_yh zI~AAQa7am;AV9AA0KI8@?Kg4XBu~=soSETrC9S=Z*LLjK2pow&!!u{j`M&cr;~zV+ zs!AQzl`itmA4I2ArjnIPI$hC;q*0m9Rk$uP9dB?uud|G9s61Y-WO)(BnKXvplgUcw z+Ek^i^rZ?@8CNt!T;z$SvFV{7X++JB7j;$DWloQz+Qc#o)uqZS9;vSCRUSrlvQSkx z=nW?H)It@DQm&UIMdxpSp-ao6)IV6M^Wv&2WF^*h1W6@N22bFDh0bMFl=dxAq^XXz z%-n+_#oatsg60h2b!^1%^P&>!dX3w<$~FRjtK=gZ&FDoQVnQUS)7H8$n&vD1h#`vj z0v`}Le;(gRR8)0-!iy^deAr{cS%5^K3oA}N&NCv&+{qQYgT%WSK|Qe~JJ-QlU?FQ}6j`y8)|RZ&8* zYZXJFSQu4m42frp2tQx)G(`eaqsizzRb{D?Q<8fvOQT59sZL;MJcp%TEcqUWDH2>E zyP)680&Ad>uu*5nE^SC6g&C|W>$IP%A>mf(ZF?d6lyR)AJtw+U)`Uq$xwF=8`T0K|m%5TV|K;cZnq6bT z6)UTTZ*4-el!NhvKON781J&#IXOoGF=ILaZ4tnGHR86CFG>qb2KN*dpNz{+Vks1w^ zQg~1%lQbTyI2nvOeyV1vSmY;=537@6Ii6Yz__nGOED$|e!R~p+Cy%i}`9hUNZ3Nb% z(s^yy4QBqOgvpgBMpW`Tka-edsbPVs%qwX&Cn69}o_;GXRhf{H;vF&~YjyOIy3BZ8 zt7>ijaCR22&1q3CPV0-al}fZc!)jG$r1CQvG(rSzA%Y?eG8V!FJ2C`Lh5&;GPJjT< zLx8}d${D-m_qB77y$b>}z7phd;`y^-^~YZiFLa(T8Ia;b08H~rSjsC4PAz@>^t&J54;m?8Eo`R{2Y9Kk0b+Tz zG;f+q*!HlHiqr@g(aEdvNYtj0d7B%))z<;fk^P%kqADtxIhM%rAG|FSas_&+2)*|) z$6pI!S3qo(Ea^#gjm0A+x(~oJ**Za8fF6JxzL%T}i|F`JjD;|){RTpBQ2H_0foZ|3 z7Q>k8U809B{}42$0l%q@`al|Qfi#{!oAevUq9It`o1VL&rG~|4MUFt}1@Q8|5Xz*# zpH6tJJE&x<$YWATq5(}M)bI520vdtpRV*zO9D_&f7(rQQh@E9gaLK4hpt_yRE~9+? z@i!hpy*Ll%kEN1HSrjWF7eEpWL`LTZC!t;!n>>a81&$xuI%wo^B;yhd3t8J-gE@g^ zbxLb7N>`Wy59ORLn0gJJFnf?m0S^N*PtPTIqJ`Z2`Z1`(S%#m?TjPzC)3814@fdI# zftJf6!Icjg#oHkVXSi4oZ!K5}5!Hl9u=gv-ULdFFj~n1SHU$(>8DhM!4Q+1%|NGndqUh`B@v zuAm&aw+7ssiUVUPvJg*g;_)X=JSY?S0$8sr)*OR@c+tx# zq>)1BKEZkQde(mCfJ|_Fa*Re&mwK>HY?eVFK|a5rKq@Ca;GL^w0fu?8Dx4C#qJrYd z^&~^dS_1X0nrYZ8r7i%bWDzZBITzYi!HA=L@{?c@4TkQD@5q_N=<`&3rK%<8LkZqlt-*LcW#Z_;F&`~UwIrgm(0&Tk zy@LtcS(VjiP*TawDw&pIH^b_(e8{TV)(9CF(3L|oh>HKglF5u?m^^O2kXs3N>(v9w zcC=m#^|}f66Pat3iO*HZQ#l|r+dSMIKQQszVPZQKTVrNKqm^S!0TMocHay)XaSe1I zos8>z8}+}Ds$T!z9NbdG8zbRk@GSum2)s;)<-8vxWRruL_&TnN2oUdwy2-bBy`tN& zc%Bb8DB8v>n$cf@_gtk5!y=OfStSCiEY_Ix*Mg z$j+;_>`4*V7S3$Phv$+2+u{rA@U&Y#wp3@w(KZ+J9T`iH06!}wVC;cX5G#!gP+s2wwwY2p^&%+3X{Mzf+*V1{(r~OehN@u;< zY&1!u*0CzR;Z%-${rNba%x?j^rxT0qpEW_y z4zB;XSmvNK@brc&h_D`2UM*F&;)%K1ZA~#nq!&#fV=RS1Gt)Z|{OaoJRQs>vRuwNvP>Rvm;Mmx_cOC%DjhK}Ago_;| z9a=pr5d*;vVTP{>uiYR`&a}7n;@xEjthb7Jr;wV9p0+t)!+4*1oo|a{4o>dEYJ>p| zXbP&^lbOH|jJ%|uRyJw}ZX%zIavLAJN^4V3^W<*oXn~kx%~KIJ+9Z>Ug$DfE_{`oJ__p@k2`4{PD9BpOUhP2bnQ(ebOCF~F_u(dO6z}5r2 zz3fzTDn4=eA8-@ANkAA*IXShK5xCsS^TLt4@oxdesS4(ij+80$%~qpu6sU?E2&%4g zyY`7san=YY@fdenQeAQh-?6|4kbfx1f1Q(6=Tz0bZX5smh~x*x+t9{4)9P?fao>P4 zpSq+kc`L(WuCtmfi|lHCHtg{`?kgd%xdN)ti;sD9ADuR%Z=5gB1N-V7ByvZi#El;S zrwRng(lkD7I}SjZNFtxn)#eo-T$(l`d9eOQZ5!0Um$p3Z?|N9~CCMP)wumJOR$!HN z2Hta`-3M%ky%%;UNCRP81FyM|awNbtii~!PxLx6F?e>y>z~TpDu@vLUBaV1t)fLun zkn*l~!X$R8Hxz6it54xf-$!^i3nXx!Shau{*T$0r!5SL-&4D$?&Gs$sA0*1en7nNu zAX@Dd$HPL9CCQEgxxDj%?*R+dRa2c zS?D|$4Nk(`cKX4GVSKBm!Y#ln-fwmTrVlHCZv_SL#Y$iLQp59SRClAKJ=Pf~Zl#I{ zOW1dv7uXbw3ek5Qk{#cRI;)}pex&(u5)it z6wbqu_(g8%x{-aS=Tv;wEV3(iHF^*&i}B@lH>aIlPE8qxYxH@&QVmP_V-;L5y0u=x zcD;D2tYq;x_@pa{nkt=cXq~sMYwlV1Z*Ah$jjoT=;+32L;Sy{34KUlMY;fE(FG>z@ zTdqlSgC|v*T@#C)U>WTeP2PQGw@=-w19hEdgRQ+lrL2z+y!Ozngh52idXL~tS>~`v zwJh6Na|`cob6rR7g)J)XwpJeTZ{ri4W7FenZ}#3njI%=HFb_TD063u(`|3rrJY0V zqzM2Kw-%eeaKL_Eq!GZSmy7fAe2Kpp+ULVXSzJ}i&fz@!q+Zodi}u~^o_0WaQpbvR zjck{fd^Mr7nf5pdZ|L$breg5uMJGz`+#>Axi~8CAXd&W}z^~%HH~a2M$2<|ei~MBw z2Y+CtS7oJw`WCQJ=Y3o>0giB!aSwpJJIR-I#slePaEo2y1_*9Z%;3@cb0E;y&+J|_ zL85U%>?J47o1=mDy}-VA;K{&NZ+e43W+B$9SZB)Z$`J}E>-K7@)jBISwhsYRD5a)N zDo1Kr8}3<)mE9BOde>drQr4p#8LBj-{B~P}VmQ3KJ7?c^<|g!w!{@EDgEpr`+J<!r%Yr2J}5@=2rJ9#T|b9VjJ<>0vhkrHLr9MW76ZgJ{`;62Ko%~xbxY^PYK`c z{Wik?tzG_Vx`mMN3B>+E0f-(amq>5iNscWl-BC(=N%hR{ZM8EFIBRN@Vrhh+zTHxH z5a8SC+X7s-VHdkR{I8A}x=MqxDcIu{GK)E=!Gbw~{IQExCn!H`i`4xJW44my3#7EZ zZ|n@jN{J^K&GRc5Hh6Z^S~u6#VFXO}Ziv-%Rn z5XX&sZk`BqAmtP8h-XLaU+1a`yhi%nDTafmv_lhX1F1P%-F{)I2>?-!3h(}MpM6{(+pG>3L5}w V4wv8KOgU|i{XY!YQhwe_008x*p~nCK literal 0 HcmV?d00001 diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_thinktank_misc/manifest.jsonl b/sources/harvest-uk-ab2026-2026-10-29/uk_thinktank_misc/manifest.jsonl new file mode 100644 index 000000000..e6cd9b72e --- /dev/null +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_thinktank_misc/manifest.jsonl @@ -0,0 +1,5 @@ +{"access": "direct", "bytes": 70197, "date": "2026-10-28", "doc_type": "html", "note": "Fetched HTML fresh with prescribed curl browser User-Agent; HTTP 200; fetch completed 2026-10-29T15:22:34Z. Hash is of fetched HTML bytes. No assignment local copy or earlier version was supplied.", "publisher": "Centre for Policy Studies", "retrieved": "2026-10-29", "sha256": "2c832e011866eb9f63f20597e7bf43bc01d44b6b1b5be43eee866ad6fc5ecd24", "title": "'A dangerous continuation of Labour's tax-and-spend fantasy' - CPS verdict on the Budget", "url": "https://cps.org.uk/media/post/2026/a-dangerous-continuation-of-labours-tax-and-spend-fantasy-cps-verdict-on-the-budget/"} +{"access": "direct", "bytes": 153352, "date": "2026-10-28", "doc_type": "html", "note": "Fetched HTML fresh with prescribed curl browser User-Agent; HTTP 200; fetch completed 2026-10-29T15:22:31Z. Hash is of fetched HTML bytes. No assignment local copy or earlier version was supplied.", "publisher": "Institute for Government", "retrieved": "2026-10-29", "sha256": "ecf19f14b4f808846fb8230b28a3244b0b65812484735842f9ab537a50195c69", "title": "John Healey’s underwhelming budget punts big decisions into 2027", "url": "https://www.instituteforgovernment.org.uk/comment/john-healeys-underwhelming-budget"} +{"access": "direct", "bytes": 105862, "date": "2026-10-28", "doc_type": "html", "note": "Fetched HTML fresh with prescribed curl browser User-Agent; HTTP 200; fetch completed 2026-10-29T15:22:31Z. Hash is of fetched HTML bytes. No assignment local copy or earlier version was supplied.", "publisher": "Age UK", "retrieved": "2026-10-29", "sha256": "3035a8bbbb0f023dc24e805b430052163aa0dbfce3c975a6e0a0568983db6bf0", "title": "Age UK's Response to the Chancellor's Autumn Budget", "url": "https://www.ageuk.org.uk/latest-press/age-uk-response-to-the-budget/"} +{"access": "direct", "bytes": 122809, "date": "2026-10-28", "doc_type": "html", "note": "Fetched HTML fresh with prescribed curl browser User-Agent; HTTP 200; fetch completed 2026-10-29T15:22:31Z. Hash is of fetched HTML bytes. No assignment local copy or earlier version was supplied.", "publisher": "Learning and Work Institute", "retrieved": "2026-10-29", "sha256": "c7fe077a0d9f4784aa4c91c906bcefa087cc9a4ece8b28026d24347c2a8b5dcf", "title": "Responding to Autumn Budget 2026", "url": "https://learningandwork.org.uk/responding-to-autumn-budget-2026/"} +{"access": "direct", "bytes": 22671, "date": "2026-10-28", "doc_type": "html", "note": "Fetched HTML fresh with prescribed curl browser User-Agent; HTTP 200; fetch completed 2026-10-29T15:23:24Z. Hash is of fetched HTML bytes. No assignment local copy or earlier version was supplied.", "publisher": "Citizens Advice", "retrieved": "2026-10-29", "sha256": "6a10336a67fc058512a4ccd9d0e80a27470fee8ba5775690fb1fcba39db67385", "title": "Citizens Advice responds to Chancellor's Budget", "url": "https://www.citizensadvice.org.uk/about-us/media-centre/press-releases/citizens-advice-responds-to-chancellors-budget/"} diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_trussell_wpi/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_trussell_wpi/NOTES.md new file mode 100644 index 000000000..7cbc84c91 --- /dev/null +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_trussell_wpi/NOTES.md @@ -0,0 +1,29 @@ +# uk_trussell_wpi — Trussell, Autumn Budget 2026 + +**Working files.** `stage.py`, which writes `manifest.jsonl` and `claims_staged.jsonl.gz`, lives on branch `uk/ab2026-harvest-staging` at `harvest-staging/ab2026/lanes/thinktanks-b/uk_trussell_wpi/stage.py`; this branch carries only NOTES.md, `manifest.jsonl` and `claims_staged.jsonl.gz`. The repository carries no primary bytes: fetch the primary from its `manifest.jsonl` URL and check its sha256. + +Source `trussell_wpi`, as in AB2025's `uk_trussell_wpi`. **1 row from 1 primary.** The thinktanks-b lane staged this row in `uk_thinktank_misc` on 29 October with source `trussell`. On 3 November its review's finding 5 set the slug to AB2025's `trussell_wpi`, and the coordinator moved the row and its manifest line here, as AB2025 keeps Trussell's Budget responses in their own family. The row is otherwise unchanged. `stage.py` is `uk_thinktank_misc`'s script, filtered to Trussell's row and primary. + +## Access + +The thinktanks-b lane found [Our response to the Autumn Budget 2026](https://www.trussell.org.uk/news-and-research/news/our-response-to-the-autumn-budget-2026) by searching Trussell's own news pages; the page prints “28th Oct 2026”. It was fetched fresh on 29 October 2026 with `curl -sSL` and the harvest's Safari User-Agent: HTTP 200, fetch completed 15:23:24Z, 171,343 bytes, sha256 `6cd2162d1382f03e937a382a39bd38c497a83d357bd969d2c85598740d17a95e`. The page's own body text was read; the hash is of the HTML. + +## Coverage and conventions + +| measure_key | Rows | What | +|---|---:|---| +| `null` | 1 | Helen Barnard's “over half of private renters on low incomes” in receipt of Local Housing Allowance ran out of food and could not afford more, in the last 12 months | + +- `proposed_metric: affected_share`, unit `share`: “over half” is the strict lower bound 0.5, `value_kind: range_low`. +- Trussell calls the finding its own research but gives no method, research dates or exact percentage: `source_model: unstated`, `own`, `administrative_fact` for a retrospective reported observation, `parse_confidence: medium`. +- Context, not the modelled effect of the LHA freeze or of any Budget measure: no measure key and no proposed key. The 12-month window is `conditions.horizon`; the period is the publication-year anchor 2026, `point_in_time`. + +## Not staged + +- Trussell prints no measured effect of the frozen LHA or of the Budget. +- Its statement that the government has already lifted the two-child limit is qualitative and has no row. + +## Verification + +- **3 November:** the contract checker reports 1 row, 0 problems; a re-run of `stage.py` reproduces `manifest.jsonl` and `claims_staged.jsonl.gz` byte for byte. No primary was re-opened (the session's network policy refused trussell.org.uk). +- **29 October (in `uk_thinktank_misc`):** the lane re-opened the saved page at the locator and checked the quote and value against it, and checked the quote is contiguous text under 15 words containing `value_raw`. diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_trussell_wpi/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_trussell_wpi/claims_staged.jsonl.gz new file mode 100644 index 0000000000000000000000000000000000000000..6b177800811cced4f4e4bfa4d65afe226e5d233f GIT binary patch literal 927 zcmV;Q17Q3giwFP!00002|9w=+j@w2Mz2_?myv#+^Xt73;eHxzzh=BlEkR#FTE{YA2 zU34!}Gy?-ZhytuJyfoRWF5#(c3yjORjhi1)W_ngQ8XeJL|9- zsI&UQ7MHqC4Cr**D}StXr$UerWqRvuh+ai|!m3d^N(xQRSV=2RFOtz_cS?>cr%M0d35^QwZ`&XX=OQZw*V_=2Y*j-?f63;5OHb7%p z0SqD%*vR8N%K!r^uiF{09~fd+p~;S@N>Y+%Zn+%G3kLL>Bh&%s z@CzmWV2TZPz*UnAkps*eerA3s45&qyc{)sZK%Kme7oa+zKV`0F$=UdA zoI|Ra*=d+Bh0+?wjWd=Nl9x=VkrydfE*Ir$C$Amv#@kLkVTTSrD(@8QkSQH_A#nyw z1LfJ02!uhYybMs|!8#{^U%&qSJHpZN1~#607P-HULYqNjYo0Z;eHv8T7-NG!lYQGuj5Zs|f zf2%VwL{?cw`_tG;h9c^cO?}Ku4j=Q9!?fhdcM{OXIE452_viC@9^W#~$sgwFc<=BW zO5WOX0+(Oz8E!e%C?l0zqZ|)OIUbTS*O2ARB<9n`3k9-kqMZM~X40HTJ`R1=T6Z(` zkmW4drbestV2i7TrPpy2-v?A#yAn3(|0X1=IukcVusw~p_&D;~V|BgH193l BzU%-1 literal 0 HcmV?d00001 diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_trussell_wpi/manifest.jsonl b/sources/harvest-uk-ab2026-2026-10-29/uk_trussell_wpi/manifest.jsonl new file mode 100644 index 000000000..d7e31c43b --- /dev/null +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_trussell_wpi/manifest.jsonl @@ -0,0 +1 @@ +{"access": "direct", "bytes": 171343, "date": "2026-10-28", "doc_type": "html", "note": "Fetched HTML fresh with prescribed curl browser User-Agent; HTTP 200; fetch completed 2026-10-29T15:23:24Z. Hash is of fetched HTML bytes. No assignment local copy or earlier version was supplied.", "publisher": "Trussell", "retrieved": "2026-10-29", "sha256": "6cd2162d1382f03e937a382a39bd38c497a83d357bd969d2c85598740d17a95e", "title": "Our response to the Autumn Budget 2026", "url": "https://www.trussell.org.uk/news-and-research/news/our-response-to-the-autumn-budget-2026"} From d5fccb466515b980fa25e868782ba95563b45949 Mon Sep 17 00:00:00 2001 From: Claude Date: Tue, 3 Nov 2026 13:15:59 +0000 Subject: [PATCH 09/52] Land the TPA, CenTax and IPPR families with the thinktanks-a review applied MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The review approved all three with fixes; each is applied in an explicit block in the family's stage.py (kept on uk/ab2026-harvest-staging). - TPA's "£93" is the most tax the Pension Tax Allowance removes, the same quantity as its "up to £94" maximum, so rule 8 makes it design: the row goes (306 to 305) and the figure moves into the 53% row's conditions. "About two-fifths" gets its own proposed metric, because the registered revenue_share is a group's share of one measure's yield. - CenTax's £3bn, £0.6bn and £0.4bn are parts of its £4bn and are now tagged as components, so an ingest cannot add them to it. Its timely- payment derivation matches the red book's Table 6.3 cells exactly. The £20 million yield is no longer re-signed, and a 25th-percentile EATR uses the registered average_tax_rate. - IPPR's 0.2pp CPI figure names no scope of its own, so under rule 3 it is context rather than a proposed package, and it is an upper bound. No printed value changes; the families' sites were unreachable from this session, so no primary was re-opened. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_01LwQq6HuQUQ1obM9JvyhJbE --- .../harvest-uk-ab2026-2026-10-29/COUNTS.json | 30 ++- .../PROPOSED_KEYS.md | 25 ++- .../uk_centax/NOTES.md | 134 ++++++++++++ .../uk_centax/claims_staged.jsonl.gz | Bin 0 -> 10529 bytes .../uk_centax/manifest.jsonl | 5 + .../uk_ippr/NOTES.md | 82 ++++++++ .../uk_ippr/claims_staged.jsonl.gz | Bin 0 -> 1588 bytes .../uk_ippr/manifest.jsonl | 3 + .../uk_tpa/NOTES.md | 195 ++++++++++++++++++ .../uk_tpa/claims_staged.jsonl.gz | Bin 0 -> 20674 bytes .../uk_tpa/manifest.jsonl | 10 + 11 files changed, 479 insertions(+), 5 deletions(-) create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_centax/NOTES.md create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_centax/claims_staged.jsonl.gz create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_centax/manifest.jsonl create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_ippr/NOTES.md create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_ippr/claims_staged.jsonl.gz create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_ippr/manifest.jsonl create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_tpa/NOTES.md create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_tpa/claims_staged.jsonl.gz create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_tpa/manifest.jsonl diff --git a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json index 038484aea..f2f5e2ab8 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json +++ b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json @@ -1,4 +1,12 @@ { + "uk_centax": { + "rows": 110, + "primaries": 5, + "claims_sha256": "ccac3eb6af8eff5eebba81f24ab7b8082a13526ac308943f72db18b9e445bdb6", + "sources": [ + "centax" + ] + }, "uk_devolved_govts": { "rows": 19, "primaries": 2, @@ -32,6 +40,14 @@ "ifs" ] }, + "uk_ippr": { + "rows": 3, + "primaries": 3, + "claims_sha256": "457ac8cabd815ca698b375f3271b3f6861f99bd958c6b7db779ede0a8e993c4b", + "sources": [ + "ippr" + ] + }, "uk_thinktank_misc": { "rows": 12, "primaries": 5, @@ -43,6 +59,14 @@ "learning_and_work_institute" ] }, + "uk_tpa": { + "rows": 305, + "primaries": 10, + "claims_sha256": "114f27d2e28955ced736752d4962525adc4d494160a0d2da689cd8fd18c58d4d", + "sources": [ + "tax_policy_associates" + ] + }, "uk_trussell_wpi": { "rows": 1, "primaries": 1, @@ -52,9 +76,9 @@ ] }, "_totals": { - "rows": 1313, - "restated": 227, - "families": 6, + "rows": 1731, + "restated": 292, + "families": 9, "pinned_on": "2026-10-29" } } diff --git a/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md b/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md index 5c13495ac..468a37d37 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md +++ b/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md @@ -4,7 +4,7 @@ Generated by `pipeline/ab2026_proposed_keys.py` from the staged families; don't A row whose scope matches no registered measure has `measure_key: null` and names one of these keys in `conditions.proposed_measure_key` (README rule 2). A key is a measure with no Table 6.1 line (`ab2026__`), a package (`ab2026__package_`) or an option (`ab2026_option__`). Each family's NOTES.md gives the key's scope as Table 6.1 lines or the producer's own words. Tranche 2 registers each key in `data/uk/ab2026_measures.json`, or folds it into a registered key, before the ingest. -25 proposed keys (9 measures, 16 packages, 0 options) on 83 rows. +46 proposed keys (9 measures, 29 packages, 8 options) on 201 rows. | proposed key | kind | rows | families | producers | |---|---|--:|---|---| @@ -28,8 +28,29 @@ A row whose scope matches no registered measure has `measure_key: null` and name | `ab2026__package_parliament_regulatory_burden_reduction` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_parliament_tax_gap_measures` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_parliament_welfare_fraud_and_error_savings` | package | 1 | uk_hmt_redbook | hm_treasury | -| `ab2026__package_property_and_savings_income_rates` | package | 3 | uk_hmt_redbook, uk_ifs_ab2026 | hm_treasury, ifs | +| `ab2026__package_property_and_savings_income_rates` | package | 8 | uk_hmt_redbook, uk_ifs_ab2026, uk_tpa | hm_treasury, ifs, tax_policy_associates | | `ab2026__package_regulation_tariff_and_sandbox_business_savings` | package | 1 | uk_hmt_redbook | hm_treasury | +| `ab2026__package_tpa_business_rates` | package | 8 | uk_tpa | tax_policy_associates | +| `ab2026__package_tpa_cgt_founders_relief_and_badr` | package | 1 | uk_tpa | tax_policy_associates | +| `ab2026__package_tpa_energy_and_environment` | package | 6 | uk_tpa | tax_policy_associates | +| `ab2026__package_tpa_fuel_duty_red_diesel_and_hgv_ved` | package | 1 | uk_tpa | tax_policy_associates | +| `ab2026__package_tpa_other_by_measure` | package | 1 | uk_tpa | tax_policy_associates | +| `ab2026__package_tpa_other_by_tax_2027_28` | package | 1 | uk_tpa | tax_policy_associates | +| `ab2026__package_tpa_other_by_year` | package | 6 | uk_tpa | tax_policy_associates | +| `ab2026__package_tpa_other_corporate_tax` | package | 7 | uk_tpa | tax_policy_associates | +| `ab2026__package_tpa_other_income_tax` | package | 7 | uk_tpa | tax_policy_associates | +| `ab2026__package_tpa_other_vat` | package | 7 | uk_tpa | tax_policy_associates | +| `ab2026__package_tpa_savings_property_income_and_cash_isa` | package | 7 | uk_tpa | tax_policy_associates | +| `ab2026__package_tpa_tariffs` | package | 2 | uk_tpa | tax_policy_associates | +| `ab2026__package_tpa_temporary_vat_and_fuel_cuts` | package | 6 | uk_tpa | tax_policy_associates | | `ab2026__regulatory_administrative_savings` | measure | 5 | uk_hmt_redbook | hm_treasury | | `ab2026__rolls_royce_gas_turbine_programme` | measure | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__state_pension_triple_lock_adjustment` | measure | 2 | uk_hmt_redbook | hm_treasury | +| `ab2026_option__cgt_reform_package` | option | 9 | uk_centax | centax | +| `ab2026_option__childcare_100k_entitlement_taper` | option | 3 | uk_centax | centax | +| `ab2026_option__childcare_100k_threshold_raised` | option | 2 | uk_centax | centax | +| `ab2026_option__childcare_100k_threshold_removed` | option | 3 | uk_centax | centax | +| `ab2026_option__childcare_100k_threshold_removed_first_15_hours_under_3s` | option | 1 | uk_centax | centax | +| `ab2026_option__income_tax_5pc_share_devolved_to_mayors` | option | 1 | uk_ippr | ippr | +| `ab2026_option__nics_on_investment_income` | option | 10 | uk_centax | centax | +| `ab2026_option__nics_on_partnership_income` | option | 24 | uk_centax | centax | diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_centax/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_centax/NOTES.md new file mode 100644 index 000000000..332ee6117 --- /dev/null +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_centax/NOTES.md @@ -0,0 +1,134 @@ +# uk_centax: CenTax, Autumn Budget 2026 + +This family stages three CenTax publications: + +- **"Autumn Budget 2026: Initial reaction"** (28 October 2026, centax.org.uk). +- **The two AB2026 briefings it links:** "Taxes at the top: Understanding what high earners pay and options for reform" (23 September 2026), which sets out "reforms available at the 2026 Budget", and "Capital Gains Tax on Entrepreneurs: Should the UK have a 'Unicorn Relief'?" (8 October 2026). + +The source slug is `centax`, as in the AB2025 `uk_centax` family, and `source_model` is `hmrc_administrative_microdata` for CenTax's estimates and descriptive statistics. + +**Working files.** `stage.py`, the file meant wherever these notes say `stage.py`, lives on branch `uk/ab2026-harvest-staging` at `harvest-staging/ab2026/lanes/thinktanks-a/uk_centax/stage.py`; this branch carries only NOTES.md, `manifest.jsonl` and `claims_staged.jsonl.gz`. The lane's local primaries folder (`.ab2026-primaries/uk_centax/`) was never committed: the repository carries no primary bytes, so fetch each primary from its `manifest.jsonl` URL and check its sha256. + +`stage.py` holds the rows as data and rewrites `manifest.jsonl` and `claims_staged.jsonl.gz` byte for byte. `stage.py --verify` checks each primary's sha256 and finds every quote in its primary's text (HTML text, or `pdftotext -layout`, whitespace-normalised). Result: 0 problems. + +- **Rows:** 110 (own 108, restated 2). +- **Primaries:** 5. + +## Access recipe + +Every primary was fetched directly on 2026-10-29 with `curl -sSL` and the browser User-Agent in the brief (HTTP 200). Nothing came from a local copy, and LinkedIn was not used. + +- **The reaction page** was fetched at 16:17Z. It says "This page is being updated live" and carries no published or modified time, so a later fetch may hash differently. +- **The two briefing landing pages** were fetched at 16:30Z. Each prints no figures (0 rows) and links its PDF; they are manifested as the route to the PDFs. +- **The two PDFs** were fetched at 16:31Z and read with `pdftotext -layout`. Locators are printed page numbers; in "Taxes at the top" the Key findings are p. 1. + +## Rows by primary + +| primary | sha256 | rows | +|---|---|--:| +| Autumn Budget 2026: Initial reaction (HTML, 28 Oct) | `92c6008835547d41…` | 3 | +| Taxes at the top, landing page (HTML) | `c954e08f2aa6de7a…` | 0 | +| Taxes at the top (PDF, 35 pp., 23 Sep) | `af4cab614220c286…` | 104 | +| Unicorn Relief, landing page (HTML) | `91a6c8dc749273b1…` | 0 | +| Unicorn Relief (PDF, 74 pp., 8 Oct) | `9fd8ce97f3343f7a…` | 3 | + +## Rows by measure key + +| measure_key | rows | +|---|--:| +| `null`: 52 with a proposed key, 56 context (descriptive HMRC statistics, current-policy childcare behaviour, OBR CGT forecasts) | 108 | +| `ab2026__pension_tax_allowance` (CenTax's 53% marginal rate) | 1 | +| `ab2026__hmrc_tax_gap_package` (the timely-payment component, `row_type: measure_component`, README rule 9) | 1 | + +## Proposed measure keys + +The Budget took none of the options below. Founders' Relief (Table 6.1 line 26) and the BADR cut (line 60) are a different CGT design from CenTax's package. Each option is CenTax's own design, in CenTax's words: + +| proposed key | scope | rows | +|---|---|--:| +| `ab2026_option__cgt_reform_package` | "Structural reform of CGT and equalising rates with Income Tax": equalise CGT rates with Income Tax rates; an investment allowance for the risk-free rate; remove uplift at death; rebasing on arrival with deemed disposal on departure. "We use administrative data from HMRC to cost the first four elements"; the fifth, more generous loss relief, isn't costed. No Table 6.1 line | 9 | +| `ab2026_option__nics_on_partnership_income` | "Applying an equivalent of Employer NICs to partnership profits": a top-up rate ("6.9% on partnership profits taxed at the additional rate, 7.6% … higher rate, and 9.6% … basic rate"), a £10,500 Partnership allowance and a £5,000 annual exempt amount per partner. No Table 6.1 line | 24 | +| `ab2026_option__nics_on_investment_income` | "Applying an equivalent of Class 1 NICs to investment income": rental income, savings income outside ISAs and other non-dividend investment income (`conditions.income_type`). No Table 6.1 line | 10 | +| `ab2026_option__childcare_100k_threshold_removed` | "Removing the £100,000 childcare cliff-edge": make the free childcare entitlement in England available to parents with incomes over £100,000. No Table 6.1 line | 3 | +| `ab2026_option__childcare_100k_threshold_removed_first_15_hours_under_3s` | "Remove the threshold for the first 15 hours of childcare for under-3s" | 1 | +| `ab2026_option__childcare_100k_threshold_raised` | "Increase the threshold at which entitlement to free childcare is withdrawn", to £125,000 or £150,000 (`conditions.threshold`) | 2 | +| `ab2026_option__childcare_100k_entitlement_taper` | "'Taper' the withdrawal of free childcare", at 8p, 18p or 28p per £1 over £100,000 (`conditions.taper_rate`) | 3 | + +Every childcare option is "costed on an actual ANI basis" (`conditions.costing_basis`). CenTax's actual-ANI threshold itself ("could be set higher (£111,000) … at the same overall fiscal cost") is that option's design and has no row (README rule 8). + +The £1.6bn timely-payment figure carries `ab2026__hmrc_tax_gap_package` under README rule 9, not a proposed key. More timely payment for Self Assessment is HM Treasury Table 6.3 line 3, a sub-measure of Table 6.1 line 45. The row has `conditions.row_type: measure_component`, with `conditions.component` naming the part. + +## Conventions + +- **Tax years:** CenTax gives tax years as the later year ("2022 is tax year 2021-22", Taxes at the top footnote 2). So "2030" is staged as `fy` 2029-30, "2028" as 2027-28, and the descriptive 2022 data as 2021-22. Every such row carries a `year_note`. This matches #207's reading of the same CenTax childcare figures. +- **Sign:** + - Revenue-raising options are `positive = yield to the Exchequer`. + - The childcare options are `positive = cost to the Exchequer`, except "raise £20 million" (28p taper), which is staged as +2e7 with `positive = yield to the Exchequer`: values are not re-signed (AB2025 README; changed on 3 November, review finding 4). +- **"Central":** CenTax's central scenario (`conditions.scenario: central`). It uses "elasticities drawn from the best available international evidence" for CGT, its "central revenue estimate, after accounting for behavioural responses" for partnership NICs, and its central bunching scenario for childcare. The investment-income NICs figures are static upper bounds ("naturally at the upper bound", footnote 13), so they are `range_high` with `basis: static`. +- **Units:** + - Shares are fractions (`share`). + - Effective and marginal rates are `percent`; rate gaps and top-up rates are `percentage_points`. + - The CGT package's "increase revenue by 62%" is `revenue_change` in `percent` of baseline CGT revenue. + - Income groups follow the AB2025 names (`top_10pct`, `top_1pct`, `top_0.1pct`, `top_0.01pct`, `decile_10`, `bottom_50pct`); percentile bands are `p50_p60` … `p99.5_p99.9`. +- **Attribution:** + - CenTax's model estimates are `own` / `different_model`. + - Descriptive HMRC statistics are `own` / `administrative_fact`, as in AB2025. + - The OBR's £32bn and £16bn CGT forecasts are `restated`. + - CenTax's 53% marginal rate and its £1.6bn three-year average of HM Treasury's timely-payment costing are `own` with `source_model: arithmetic`. The £1.6bn row carries `conditions.derived_from` naming Table 6.3 line 3, the same rule as uk_tpa. +- **Context rows:** the descriptive 2022 tax statistics and the current-policy childcare behaviour (bunching counts, response shares, partners leaving work) have no measure key and no proposal. The option keys carry only the options' costings and the distribution of their revenue or affected taxpayers. + +## Proposed metrics (registered Metrics used where they fit) + +Registered: `revenue_change`, `revenue_share` (also for "paid 60% of Income Tax and NICs"), `income_share`, `income_statistic`, `average_tax_rate`, `affected_count`, `affected_share`, `participation_rate`, `tax_collections`. + +Proposed, with CenTax's own words: + +- **`marginal_tax_rate` (percent):** "a new marginal rate of 53%". +- **`effective_tax_rate_gap` (percentage points):** "the gap between headline rates and the average EATRs". +- **`taxpayer_share` (share):** "one in ten pays close to the headline rate of 47%" and the other dispersion statements on p. 11. +- **`behavioural_response_share` (share):** "fewer than one in ten employees with eligible children responded to the cliff-edge". +- **`workforce_exit_share` and `female_share` (share):** "3% of lower paid partners … leaving the workforce, with 90% of those partners being women". +- **`partner_not_in_work_share` (share):** "the share of parents whose partner is not in work is about 6%". + +## Not staged, and why + +- **"Around four times as many parents would be bunching"** (p. 25), the halved-unresponsiveness sensitivity: a scenario ratio, not an estimate of an option's effect (added on 3 November, review finding 6). + +- **Duplicates within one primary,** staged once: + - in "Taxes at the top": 62%, 88%, £19.7bn, £2.1bn, 98%, 1,100, 90% and £640m (Key findings, repeated on pp. 17-27); 37%, 34% and 30% (p. 9); 17%→12%, 33%→36% and 37%→33% (p. 13); 40.5%, 60%, 43.9%, 61.5%, 47%, 20% and 23% (Conclusion); + - in the Unicorn Relief PDF: "almost £20 billion" restates the £19.7bn in the same footnote. +- **Design, not estimates** (README rule 8): + - the reaction page's "marginal tax rate of 33.3%" (the Pension Tax Allowance's £1-for-£3 withdrawal rate, restated as a percentage; kept in the 53% row's conditions) and "a 100% marginal tax rate" (the Carer's Allowance taper of £1 for every £1, restated); + - Taxes at the top's partnership top-up rates of 6.9%, 7.6% and 9.6% (the option's rates, quoted in its scope above) and its £111,000 actual-ANI threshold. +- **Policy parameters** (also design under rule 8): + - in the reaction page: the £12,570 Personal Allowance; the 24%, 44% and 49% rates; the £12,000 cash ISA cap from AB2025; the Direct Recovery of Debts thresholds (£1,000, £5,000, £10,000) and dates; + - in "Taxes at the top": Employer NICs at 15%, its "effective rate of Employer NICs is 13.04%", the £10,500 and £5,000 allowances, footnote 11's rates, and Appendix A's rates. +- **The reaction page's worked mechanics:** the payments-on-account example (a £10,000 liability, two £5,000 payments) explains the existing regime; it is not a figure about the Budget. +- **Charts without printed values:** Taxes at the top Figures 1-7, B1 and B2. +- **Unicorn Relief's illustrations and design parameters:** + - the risk-aversion example (£100,000 v £1 million at 10%); + - footnote 27's net-of-tax returns on a £100m gain (76% → 76.06% → 77.4%); + - the £20m break-even example at a 15% rate; + - the taper example (45% → 44% per £1m above £750m); + - the £750m / US$1bn threshold; + - EMI and SEIS limits. + These illustrate design choices rather than estimate outcomes, the same test AB2025 uk_centax applied ("thresholds/equivalences, not scoreable values"). The report says it does not cost Unicorn Relief ("It is difficult to model how much Unicorn Relief would cost"). +- **The September 2026 childcare report** ("Removing the £100,000 childcare cliff-edge: Impacts and cost of reform", linked from the reaction page) is not re-staged. Its Table 4.2 is already in `data/externals/centax-childcare-100k.json` (#207). That file differs from this family in three ways (thinktanks-a review, finding 3): + - **sign:** #207 stores the full removal's central net fiscal cost as −640m, negative for a cost; this family stores +640m with `positive = cost to the Exchequer`; + - **coverage:** #207 holds only Table 4.2's full removal; the cheaper options (rows 101–106) come only from "Taxes at the top"; + - **bunching counts:** row 27 has "11,980 by 2030", where #207's 2030 central "Number Bunching, Before Removal" is 16,280. This may be a different concept, or CenTax may differ between its two publications; row 27's note says so. The £520m and £640m staged here are CenTax repeating its own central net fiscal cost in AB2026 publications; the ingest should treat them as the same estimate as #207's. +- **LinkedIn:** CenTax's 28 October LinkedIn post (seed inventory) needs a login; not used. + +## Review fixes, 3 November 2026 + +The thinktanks-a review (`harvest-staging/ab2026/reviews/verify-thinktanks-a.md` on the staging branch) approved this family with fixes. Each sits in an explicit block at the end of `stage.py`'s rows. No primary was re-opened (the session's network policy refused centax.org.uk); no printed value changed. + +- **Finding 1:** rows 84–86 (£3bn, £0.6bn and £0.4bn, p. 22) are the parts of row 24's "up to £4 billion" (3 + 0.6 + 0.4 = 4.0). They carry `conditions.row_type: measure_component` and `conditions.component` naming the income type, so an ingest never adds them to the total. +- **Finding 2:** confirmed. Row 2's derivation cites Table 6.3 row 3 at 1,355, 1,605 and 1,815 (£m, 2029-30 to 2031-32); `uk_hmt_redbook` stages exactly those cells. +- **Finding 3:** the #207 file's three differences are recorded under "Not staged" and in row 27's note. +- **Finding 4 (optional, applied):** row 105 is +2e7 with `positive = yield to the Exchequer` and normalisation "£ million × 1e6". +- **Finding 5 (optional, applied):** row 18 (23%, a 25th-percentile EATR) uses the registered `average_tax_rate`, as rows 7–15 do. Rows 19–20 put `revenue_change` in `percent`; tranche 2 must not read them as £. +- **Finding 6:** p. 25's "around four times as many parents would be bunching" in the halved-unresponsiveness scenario is a scenario ratio and is listed under "Not staged". +- **Finding 7:** rows 20, 58, 108 and 109 are CenTax's 2024-era estimates at fy 2025-26 under the CGT option key; their `scenario` condition says so, for tranche 2. +- **Verification:** the contract checker reports 110 rows, 0 problems, and a re-run of `stage.py` reproduces both files byte for byte. COUNTS.json pins the gzip's sha256. + diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_centax/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_centax/claims_staged.jsonl.gz new file mode 100644 index 0000000000000000000000000000000000000000..d4f855416f313180cd19e16e908aca442adeb225 GIT binary patch literal 10529 zcmXZhRX`M6*EV2=p@tl~ySp2ty9K0`l#uRjhHj)AC8ay18w4IYhwkqF_5J>R?RB~j z_R(7RB9BIb`FCL7PmPY_Rz?%weydHV8*>neCwZlrQ@39@bqFGkjDT$So5d4ZBgvv> zDH-FxQ;4rqWo`Il@2gzV7|gA8Skc6m)qaH(GU=XNs&}~R==gonar2nlU|VHT3LarX z**e&wZI5s$ie_t1Z(+m;iHJ{gBJlk7K&u}{`PNPC(B+t%Rg2$d9Tor3<2s5>X0a`E zat$WG-x|ia`0m-a!B#AB)Vi4-5S(=Z^7vkfH|<$>Whp{ToVf{=(tH_*f8jkKlUN-n9sBDKyWF_-HKMnNf&El^ryhKLF{~#0@(~{EN zAGM&!)C2c~bKP_I@o&2nFY&Wiw~ebTae)eBo)_=~!ev=nvZ}B?bZr(!pTWs+l2zQt zZsKki@-7}QsN&fZyv$9=RrnUX=(+Ep#eZnYE7-7y?KHe#x%eEl*ISB^Ydgnf`15|a z?9Yd82i(@k@L<#i(@~*9u?3bz|Byfb8RKz+x!%0{(g@?^%FusZpPe3WyuPH;RwUGK zx{rQeRV%~3=z0cM_^&~=ey-1m4!~XT8AhTJ|n0$iuslF!68*%W+2fM?W1l9h3AyugVD_!5&2{= z3yzT=D@*(<1q+77QHQyh3t1`nRLKFrJOv zzc=CY+X+g+!tbgSI@2O5e}h`ag?r7H07+74k+J$%LI8SQqdJk!>SqECd+{c($Vx?l z8&}R;L@ZhNxJe~XwOioS**eeW#jZ?86xRcVt}A#VwA=Y07^33oKGxbFCk8Bz2>+6H z3K_K1^KGsHKbQ}K#nJmwgRC!^5*6783_k<2kR)bkpUTIuET`X4fYl-?t=OE4kK7lX z-=|(4WRH4-<1s<~yzmp>&ge{6CGe-PL~9HqYe9U-0Z%&+X5UJ?y?_{p;9-1^0f}9* znjV9WsF1t+_otO)4&`Zbm~dN>O_zI3m>I``PQ;l%rIt1ZH`jcbVNpU%;0BbgZCbu> zB6oWd2R>{sRo|8;1hVKwa)t`TR3BhnB|4bkH+u;NAE6Zp)Q~L(2&jYS5}G++8e^nl@ua*NfEK2PM#*y1^_aM(!|l9-SGwPhC9nu@36 z2TR@xpo+Pp{0h_a3n>G@z>TJB5mMUZpMs$iU#OD-7A@=#mI(yiQss`dIvuUyQfu$& zR6H+(QwR%w-R5+>k`1zeYVrqEZ$#6s&2aVnGSb-Pb#a&E_m}N2hpMJo!I$9b)7{1& zAhk^G7Hw<`9p9bKvSOE z>yIhuEsA6@Zxay`jI?0<^&6+EK+L%l#K@(--(Xyw%VfQ6BMqKF^;Yz#y(^lt5jMZ z`Mk7yWy1w%cg50B;L~ZI#XO_#+bK_T&yaV>pJyQaePi=HDg!U*Tngn`IR&G4?chV2DgpWvJ-3J8=eG4&=Ew{6q--q(=Z2NT zWZ1r*@|X`Vfm8qsDOIV1Bjc~*~HOXXxpt&{H z>HBDc9-BB0sUIW3nXba(muI#uz0Z*xp@lu67$9AYoWTDPan%xCF!i-X1uklj(=&~~ zgXGgo+&*XlTvZF-B^2Qe>jRcq`bIFM4x{#8PNT4LnecLfmR{mpV(HbSxyh0DlAkMK zb)D^ZX^e~IDWxSTCg(P-?^<}4Ts#zkN^hTA>ybEvvw-U>ADg`TI=$`l6dQ4BXqQy* zQshW{>!6LS*v%f~qFnH3>BTPM9pJ7Mpx0-=T+AVRTUAa!Q> z3{(xV`Jp}Zhti5fALW<5P6M2a^2huQC-$sa`l`-#qTxjrI=9Kb1Yl1%q=|A4YuuSO zqM9wJeSwCsH%y~ElHTIp0ywKN*A83tFxvhLIlD=FyJ5Rh8klcXvT7P~av%L+ADnQv zv`M$_06iReU=6e0^JGD-rNco1VZv>MK}m3~L7fRsW%Az{VKRI3{Jlf&upeTj&VB2C zlXsD-1T`07x?{VGp}6lW;ZJcjP(g%1Gp*k&s`RZN8DS?z)0hmRpsFe_2pTy6Z+hQa zSQ&ieZlNq)Hl6eya=04iiVEp(hg<|yz-vZxrk#Ch<98xu?IuE{DhCzlvCZWdE zrorTg-gszTlUI^m+>vTs7!6V}Ir@J-``f!D15xH$N(LqQ`=g43qEo3D1slLACjMvI>%{ktD~klt2PFwFt8o(2CfP@aGRg^EHJsH~jlYRZg;R6HI`B zR48phU@>FPe%vbu=OIa9R#0 zxRx3oMV^MhQWyl87MC$9pl4hOj6rJ=F91Lq9$KnCWA7_v@S>cdDf^|=fYDNGUNLLd zEPhw5!imx4uhoE7Kr6)LH|&jg&n}2G`pCPoYyBu;j=3zyZ$ULDH!%{)=yuvsJuRAA zObv9(nbdnoR`4LCXtj)S%)C0D?iuL_O@jjRzCqU=3q!;raf3ySVFpLT6q>6qwOq+v zNvb+Tu5tHFhA9^o+2-*iwTxFw0sYh1)*YMI&4F5VL=cLi_Rf<)(C2!=4vmyW!E89k z=5-sVGz5gu3(IsH#D2`d=3dlmA!bj;mY#-dj@Y~q_j)*6Ba&AxebQ3SBLth6Vw_jl z~@84mzz?nHfH@x!2-bQURn3`(izR zo0nK6G1OrS#5D}le}2GiK<{w9lZ6qOgMEPC+ z&BI?%Fp&{0jPSSl5%+Gk-Luq|jZW%3z-7F@<7@rauXGdm!`NT?kr=KS<**1;pkl(f zYz%jfq2q|4Db^I9BxqsC%a1=MqdVMU@&I-y)ytX7ugy^G{J2sLyzangA%{XLuxhIG zX)<9rX1va@pLM31KrKl^$MI7kuJ7KgXeyMTa*w;6o?V14Zx`lL?x5rx(h&s~FysGq zT-c(OrgCBA9Q9lK$1ql>_WGac2MsUmPkItp^WVYria!~z1S7dJHynS~!05tMvixK< z5@cgWGKFU&;0as^2{uRPJR(eW12}cRNA}}Wfulvq(b<)-I(NfzA8mqGsF>`?zoUoq zmf`Qa(#@{M_vKKEa%EHg5Oy(MDY0cSBH(}y1gk!LPWGRuMP z9RlsU>7^}$>zze6s@_U~F^h1_(H?jC2A&zD7e|h8%>Q$Bn}3?WNnrLVFIq(cAQQZc?GPLavhfc5w$`~JljfZI;KRtS6BhjG# z`BYV{<)X=rKF?-S%JNZIpqVIosivnEqiV8og0s}3gprMj<@L{~v{AHE2Ig8KEr0YL z;GiTYRt3ePb$yZgq8UQBRei!L4!55;Mhs>sJP;Bo& zWf4jJ6|@#SXVr`|SF{@CR+H7)s+qry`jHwlfLn|~!ea@83y*Hk*WUA15n&uZ&>VZx z@Dl^2Cg02eaMy?k=|Nm;Gt!3MwwGM)ffceauFbpl`;*%(U7Ml6*ZcPcaP z9V1z8G#9&H6~2ZF8}E{ae60kF_VOYRw&L80wTZ|wWv&Rc0@{JX&^(E>JO)a%9*^y# z=;&RKjB04X-iYKGUSg?at$y~XY(uCceSeSjONISdsV7-G>xwv%!x2cNt6F@VHpmn;O@TQPQHLI=$|_iFXA5H za@gEc_X#q^^OMB^0c^7ws=w~qv=z-;ztKotsz-}GV`>N(T_~HyLe}an$7)^2v+5Q! z{eR*i2UFKlU?Wfz__ZCgGtf1bBjm_RIUiS^mI2lVSrS|8A2Ew%6z9-Jf3xsLzH`(yZvxtHeK-ZofnB$1GXCryOt<~Q*du2 z+vGb)*)$l!*#oDKQ5@y9`qbBvFQChet)8}CyMZ_7Ik-E$gO}qjO5kqSOu^@6U^;}VP5uXcm`Sh zQ^2`ml0qCPS?ke;Oxi*TFUf0lns)vQs{R z9X^siyQwE>!hclstm885BOr&uFjeif=^dDQy{w*Fg0+3yEQ^x-M_*TXBM8ZQQHeBw z|I>Gd*8YhewnDk900OOIuR9?EpS0cV%5u@~N0o zw2P~|?ZDW_ID^ssBi;gmk*NaOe_c2OV<87w-)Gc_R`|hw4JFdH1RM56PZ4zh>|g(f zpLky7Ibn2zqQG*$*FV^Wd;H-x)1WetUZwQ|#BA!Z=|fe7K85br^OyN1tf=&0jY zavN`I=gUO3BJ^=h(Jnr#rnAFM5saa41MYWUPV5O|czF;nljBhy^ zvbI~Iv8nkxi1mp4^#TM6Q8xUJy7iHK-iSa1zwu+e)5;M`^*+e;cv@jlpi_!c`duf! zxs!!j51t*JtUY%0M%52pt8nt7%&DAW{Z|9ADxnW}in`kZr(>s!nX{`}xT7bv0+Ab5 zzA9$zU@)AuLH(h!DMG(Zo>wn%JVE_@0H^A{f|808p&VZzPT$C5E}BM#XZ6o0k0+IO?H=I`g3Tz0;HecznFHjnviV#%`V#``SHESgaZBB=qiy0Nn+3M$(s`J zCqPy89gn@rs1viFYxfZ>;LN7}s6()3a?)RJQL^AnB3SOv1GsYH%coGcOdqZ44dIb5 zMB!{Fih@t`W8%}b3~C}9;WL^Qr%smGHO0R$R+5L_)_I5+NtVE>w^JC@qcbeIm05%l z1_}K3=ZuvUzC;^+i;JDUq#^8M9@2*MeyNEnWoLNT5pMZ7x_u+Qd^UpOCf=qMf;(9g zl;_g;PR%poA@zCAXFh1bU<39jS#Em)6IPl)T$o?_#;Ds?0B+n8I8N@UvjW)TS$hzmW`^%ajNq`b2DI=US%njQ%53$MKs5?dnAf*l@{O zU8WYLMsb$4>|(o?e{9g^A+5@zC`z)JV~0&VO1#Vg$P4U-)k@HSy_s@MsX(U`4wMCsju z4D*?*MgbEUS7%ocJCQy=NyxzPG+4x#4UsRMb&B7ykE6?tx|)PteJ$&6V(gqO$aGJI zKoWR{_#dj=$i6+_&@py~EviK}4WPT>?+Eh zHTAW}uhQ=;E3?0&h50>MqGfl;xkPY5HT1_IXdF>~gFPw@J*IuH@M?gHi+rSbvJg)}JT{7LBE2!|M~ zuYG!Xv$Pexs+I5^GC?pFbI;YY?BmmyRX zI_PCxvwqabtD6o{RzO~~V_y@=zP>HevN=dI> zd~}luj9YSJ2w)htWLh=KMYy0*TV*g=$&IcXQJ$9AE#^gFe2tOGYit{s^Gz(w(Qmpb zkw`x`L$RtQ$xB{ff7D)v;RPkYEbEb?QjU5RCrrILCp>as-L9&MCJT&h4J^(NCq90- z8%aB9#JohmoNV^A++2s60-jqGxj*Su{HZl1WY;NZ7jBY%xsrb=oHJq~&(SH37T6N# zVH=l(&?8d0=DD;@*>0h1MSJ@rcj;1~TY9*H9;NK;G4m~B6fLgYjrpJ%H^Iv_^4**oD zAajYJ%ap?&nBpi?klmOc$+F7fFA#2A1Fca8TP4#|vgmV$njH6o|Jji{DFg+Ew<*|% zVP)g02#8~c@I7M|&BPP%CexAtWPqlTd$rP>Jjym$Frl*k($6mu8h&D?(s0vUVv@|S zNL2shOzqx|dD|JmZP1PoOu!z_dPgzP+R6lW0G~N0DBWC zU#Rj8tG8G?3=}IL{LgBn=q_7Bn{+`6v?IbaRC>C31)Q|ob3T42#zYWs7u9g zTBM2rFeQXtBZWJIgCe{9`{L3MkzK{H>)+nGL*{ z>GkhX8XI-ay##unG(l23v~pU7;#$~@MO-4vWhwUE<|!}HO_&JA(nh^tZ}HG+K76Ug zMpINNVoWpeUkQ0mVs-?L-v}64^OM%>oViaVHu$chmVql)bk@G=yjhX#nS=O`!yG^5 z;{@`Rx@0QkVS@_nPwfj3q_Iw&S+f<1FiOg-MIvY-Xohn<_q6Rx+?g=;6*RqY{CdfN zXOY5qV*(%hgD`;0X<%R1uOstHUDry8T(V%lE$gQ4$<%Wg7?uwSzl%@E(Oo&B9Zzb6 z6(=vIVQ%AIQ!9aFPcSmDiR=D%d&ZEDkNoMSjPRPBf}fODN3_2Du2BF(bid#H+l6gJ z1b5<_y(byTn5+R!_i84y?JwE3PU+n)5Vb!6?+IBJ+8;nV>)m(-Th>lZ1Fu&ne}~x5?6;= zYK|bd%~FK?dsWw^5f%IYSG}E_cFl2Cj;d<5SaB98N)$V12XjgI|1DiUj(aoY>8U{L zi!?QB19x25ksfsya+&m7T)Y?~;23_8*iAhTwqc=g?Oo4?1VIm(=W*ED_T>X|%@=KC zqf828W&cAcM`7E0z{2TX#h6xbwOFp zE}_UbMo>f~MKlS5?W|#w5j)dm;0rLu{f$gO-zGq1B;Xd`Ay9OE!w|tmMvI_5+e$S^ z@s|KCV$s=>_c1DIOo^d^nA9@CP6sE&`ac4+>jbmAu|aORxd={(tt8T<|NJ?3UOl{? z{hrzxFKTRCr;Q~V_yPRTbg=z5>M;oo3)D#t7HbiKMSWrb6AE#oT+saD$B-mXDWQBe zZBF2iYk}gCmBkop!}!-Ls-OEft1pKBkvNs8xByKvXf^66r$6lfLV22;)gkn~O_-=c zYSq3PFDrVWK_w=5#*%IBrIrKw)+0aQ0sQ-50!AGKkrCC~3adA`1nF=G_XpVA*h3X8 zNT3J<{GN{K1c|&Ndf06WCoKC>=lm z;Q1en2e08%i1i7jTx zcTm;S(*PF8$=(cxzNWrxwdc+W-UbeshHTe2D@3$D-@| z;S*q}&fpZG11(C!s?Y(zvIMIajkJou6Jp2M8yLXa0k5nqMe3<8`Y&v`5z|al)h15XkA0*`P z5m}tJ+wZj`kIba;O5VQDlPGXA*3CyGJnqLfuKw1d@<>k0LXKzC-|$C%!@~4o5$P(!=oPd=`F>lzhgR{NWHJ7IE4Q z3XqwS#=@Y8tg|`NXw82JCZS!cAwJ}3fY3rQrZfs8zs^XTz_8#|*fbTpXV_sQE?O2i zY<=xx1K1;mMBF?E)7%ZgDav9h+%XS&F54D-3$K%rY3s+Eus|`OZld_h3eF&9&~vLR z%&*451lxdQ91mAJc-pYsW^L#vif<+AWgm9m&r&q>OeR3L!`|LcyqO5pDXeIoBpLVV zFg-m;?yq~9pP^nL!OhWp3`Zm|w%WNfM7U>B+X9W+Y!oN$p%pKM@q_2>5>Et2ZX(Cs zg4L=yY6!zWX3<;%fdjo0<_Lt5IGg&~R-sE52p%*f7(7uth@Lwd_Ibw0({Uq=vhDJ_ z#;TNxf&g-9dRfZ9mP~Q-o|)BltmeXvDYks2NBKNubr%=d`<}NxVOU~x8l7;8q=-V$ z%;L9y@>GqX*4zGSus-zWuj+53ZpW`=0_2hZShVT!3#8WAyF}?a)>k98uMb9N&l+@s z_!$?i@o;25NyW#*OsR}zqZ>}T<-;2x#%%4)lIUVD5Tz#i%adKF7xw!D*q7nmoajJMQ}Iq^B=Ldb9N2r zfI|Oz(7@GUz}sc@JLAx9fRz7ASo51(tzz&xhH-Pg4e_sw=(K^2KcKc^0f1P`V=I$r z$Kz}GXOde!p?S^I^O3vrdx?uu!$wJw{HIbehFSObTqZ##9mbG9+!j~toNJuAFdc~1 zTfbE7eBKputGYSCI_(tIHfN?6fm;UnUd-g{H1ZDK!JZdEY??kT;^I9#QZAQ0-g};; zLEv-@G+G6YrMkaUso>e4`w*NOcrTV@z@7;H!D?;2xIH0ZD|Eh@43CfQutFl`K&e^Vt;NzFt-geNZq#wM>kqbvSpfjPE^K zS2C|DY6S<|-c$Nw=$O%Ms|tq#8Y8gtR=jaL0#|JhR)mOjYZ^-C+Zp~|np#)Jhiwq< zXQxV@^_2T!lnG4;JSC0efoUn*aa+ literal 0 HcmV?d00001 diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_centax/manifest.jsonl b/sources/harvest-uk-ab2026-2026-10-29/uk_centax/manifest.jsonl new file mode 100644 index 000000000..383a016c6 --- /dev/null +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_centax/manifest.jsonl @@ -0,0 +1,5 @@ +{"url": "https://centax.org.uk/autumn-budget-2026-reaction/", "title": "Autumn Budget 2026: Initial reaction", "date": "2026-10-28", "publisher": "CenTax (Centre for the Analysis of Taxation)", "sha256": "92c6008835547d411b67b9f23edecae77a45ba67985d5364d6bebde20f7be457", "bytes": 248542, "doc_type": "html", "access": "direct", "retrieved": "2026-10-29", "note": "Page dated 28 October 2026 and marked 'This page is being updated live'; the copy hashed is the fetch of 2026-10-29T16:17Z (curl, browser User-Agent, HTTP 200). No published or modified time in the page metadata."} +{"url": "https://centax.org.uk/taxes-at-the-top-understanding-what-high-earners-pay/", "title": "Taxes at the top: Understanding what high earners pay and options for reform (landing page)", "date": "2026-09-23", "publisher": "CenTax (Centre for the Analysis of Taxation)", "sha256": "c954e08f2aa6de7aaedb5e12f8df16ea93a44033e72c1bb08d25c5159e6aac33", "bytes": 267599, "doc_type": "html", "access": "direct", "retrieved": "2026-10-29", "note": "Linked from the reaction page's related-publications panel ('reforms available at the 2026 Budget'); 0 rows, the landing page prints no figures; links the PDF. Fetched 2026-10-29T16:30Z."} +{"url": "https://centax.org.uk/wp-content/uploads/2026/09/AdvaniHughsonSummers2026_Taxes-at-the-top.pdf", "title": "Taxes at the top: Understanding what high earners pay and options for reform (Advani, Hughson and Summers, September 2026)", "date": "2026-09-23", "publisher": "CenTax (Centre for the Analysis of Taxation)", "sha256": "af4cab614220c28640c0d875ab24aea263c54ec75957cfe263f6dbac171857f3", "bytes": 963857, "doc_type": "pdf", "access": "direct", "retrieved": "2026-10-29", "note": "35 pp.; PDF CreationDate 2026-09-23. Text read with pdftotext -layout; locators are printed page numbers (Key findings = p. 1). Fetched 2026-10-29T16:31Z."} +{"url": "https://centax.org.uk/capital-gains-tax-should-the-uk-have-a-unicorn-relief/", "title": "Capital Gains Tax on Entrepreneurs: Should the UK have a ‘Unicorn Relief’? (landing page)", "date": "2026-10-08", "publisher": "CenTax (Centre for the Analysis of Taxation)", "sha256": "91a6c8dc749273b1c6b6e330a2acaef4010a379d5e82e1bab22c4a820c5cb65a", "bytes": 267896, "doc_type": "html", "access": "direct", "retrieved": "2026-10-29", "note": "Linked from the reaction page's related-publications panel; 0 rows, the landing page prints no figures; links the PDF. Fetched 2026-10-29T16:30Z."} +{"url": "https://centax.org.uk/wp-content/uploads/2026/10/AdvaniFlewGazmuri-BarkerSummers2026_CGT-on-Entrepreneurs-Unicorn-Relief.pdf", "title": "Capital Gains Tax on Entrepreneurs: Should the UK have a ‘Unicorn Relief’? (Advani, Flew, Gazmuri-Barker and Summers, October 2026)", "date": "2026-10-08", "publisher": "CenTax (Centre for the Analysis of Taxation)", "sha256": "9fd8ce97f3343f7aee1cab11dede3f374ba473b8db4056268a0becc154e7a71b", "bytes": 568966, "doc_type": "pdf", "access": "direct", "retrieved": "2026-10-29", "note": "74 pp.; PDF CreationDate 2026-10-08. Text read with pdftotext -layout; locators are printed page numbers. Fetched 2026-10-29T16:31Z."} diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_ippr/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_ippr/NOTES.md new file mode 100644 index 000000000..dcbce2ee0 --- /dev/null +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_ippr/NOTES.md @@ -0,0 +1,82 @@ +# uk_ippr: IPPR, Autumn Budget 2026 + +This family stages IPPR's three responses to the Budget on 28 October 2026, found on the ippr.org media-office listing: + +- "‘Unfinished business’ on economic reform after sound Budget, says IPPR"; +- "IPPR North reacts to Andy Burnham’s first Budget"; +- "IPPR Scotland reacts to Andy Burnham's first Budget". + +The listing shows nothing on 29 October. The source slug is `ippr`, as in the AB2025 `uk_ippr` family. + +**Working files.** `stage.py`, the file meant wherever these notes say `stage.py`, lives on branch `uk/ab2026-harvest-staging` at `harvest-staging/ab2026/lanes/thinktanks-a/uk_ippr/stage.py`; this branch carries only NOTES.md, `manifest.jsonl` and `claims_staged.jsonl.gz`. The lane's local primaries folder (`.ab2026-primaries/uk_ippr/`) was never committed: the repository carries no primary bytes, so fetch each primary from its `manifest.jsonl` URL and check its sha256. + +`stage.py` holds the rows as data and rewrites `manifest.jsonl` and `claims_staged.jsonl.gz` byte for byte. `stage.py --verify` checks each primary's sha256 and finds every quote in its page's text. Result: 0 problems. + +- **Rows:** 3 (restated 2, own 1). +- **Primaries:** 3. + +## Access recipe + +- **The listing:** the media-office listing (https://www.ippr.org/media-office, pages 1 and 2) was fetched on 2026-10-29 to find the responses. `/news-and-media` returns 404, and `/articles` lists no Budget piece. +- **The releases:** all three were fetched directly (`curl -sSL`, browser User-Agent, HTTP 200), "Unfinished business" at 16:17Z and the North and Scotland releases at 16:35Z. +- **Edits after publication:** each page's JSON-LD shows a `dateModified` on 29 October (16:04Z, 16:21Z, 16:34Z), after its 28 October `datePublished`. The copies hashed are the fetches above. Nothing came from a local copy. + +## Rows by primary and measure key + +| primary | sha256 | rows | +|---|---|--:| +| ‘Unfinished business’ … says IPPR | `f834beaad7802472…` | 2 | +| IPPR North reacts … | `9b1e6ee72f3af251…` | 1 | +| IPPR Scotland reacts … | `9f328d542179e8f8…` | 0 | + +| measure_key | rows | +|---|--:| +| `null` (1 with a proposed key, 2 context) | 3 | + +## The three rows + +1. **"It will lower inflation by at least 0.2 percentage points"** (Carsten Jung, on "the cost of living support announced today"). + - Staged as `cpi_inflation_effect`, −0.2, `range_high`: "lower by at least 0.2" makes the signed −0.2 an upper bound. Context, with no proposed key: IPPR's words do not define the scope (changed on 3 November, thinktanks-a review finding 1). + - IPPR names no source and no period. The OBR's EFO paragraph 3.10 says the fuel duty changes and the temporary VAT cut on domestic electricity (Table 6.1 lines 2 and 3) "reduce CPI inflation by 0.22 percentage points in the second quarter of 2027". So the row is `restated` with `parse_confidence: medium`, at `period` 2027 (`point_in_time`). +2. **"the £18 billion annual taxpayer losses at the Bank of England"** (Carsten Jung). + - Staged as `apf_indemnity_losses`, the AB2025 uk_ippr proposal, which staged IPPR's "£22 billion" the same way. `restated`, source not stated. + - It has no year, so it sits at the scoring year 2026-27 with a `year_note`. This is context with no measure key. +3. **"Five per cent of income tax devolved to mayors would see an additional £4billion"** (IPPR North). + - IPPR North's own estimate of an option beyond the Budget's commitment on income tax devolution. Proposed key `ab2026_option__income_tax_5pc_share_devolved_to_mayors`, proposed metric `devolved_revenue` (gbp). + - It has no year, so it sits at the scoring year with a `year_note`. The release links the paper "Devolution that delivers: Fiscal power for regional growth". + +## Proposed keys and their scope + +- **`ab2026_option__income_tax_5pc_share_devolved_to_mayors`:** "Five per cent of income tax devolved to mayors". No Table 6.1 line. The Budget made commitments "on devolving business rates and a share of income tax to England's mayors" (line 21 covers the business rates share) without setting a 5% share. + +## Not staged, and why + +- **Design, not estimates** (README rule 8: spending allocations, financing remits and payment amounts are a measure's design): + - IPPR North's "£285million of good growth funds" is HM Treasury's allocation ("£285 million in 2027-28, comprising £165 million to support places with a mayor … to establish Good Growth Funds and £120 million of incubator capital funding for areas transitioning towards a devolution deal"), restated. It is a spending allocation, and part of Table 6.1 line 20, whose costing (−£340m in 2027-28) also includes the subsidy on the £1.5bn finance. The uk_hmt_redbook family stages HM Treasury's own £285m as a row keyed to `ab2026__good_growth_funds`, which rule 8 may also exclude. + - IPPR North's "£1.5billion borrowing facilities for building infrastructure" is the size of HM Treasury's lending facility for Mayoral Strategic Authorities, a financing remit (uk_hmt_redbook stages it as `policy_parameter`). + - IPPR Scotland's "Bill support of £100 for households in lower council tax bands" is the Household Energy Payment's amount. +- **IPPR Scotland** prints no other number, so it gives 0 rows; it is manifested for the tally. +- **Pre-Budget IPPR releases about this Budget,** outside the assigned "responses of 28–29 October". Each was fetched and hashed on 2026-10-29, and each is a candidate for a follow-up if the family is to include pre-Budget options, as AB2025's did: + - 27 October, "Chancellor’s £20bn to invest would cost less than £1 billion in headroom: IPPR analysis" (sha256 `e26f9b14989f14af…`): £20bn of NWF and BBB investment costing under £1bn in debt interest and running costs in 2029-30. + - 15 October, "Get started with tax reform to pay for priorities…" (sha256 `97750268ba93a3b6…`): + - a £16bn package: CGT reform about £10bn, deferring Bank of England losses £5bn, a bank windfall surcharge £1bn; + - headroom down £7-15bn, and a buffer above £20bn; + - an online sales tax raising £1.5bn; + - a £2,000 energy price ceiling; + - the £200 a month LHA shortfall. + The Budget's bank surcharge and the gambling and online-retail measures overlap these options. + - 8 October, "Tax online sales to fund a high street business rate cut" (sha256 `fa570119927dcc54…`): + - at least £1.5bn a year from a 2% online sales tax; + - average savings of £1,400 and £2,500; + - multipliers cut from 38.2p and 43p to 28.2p and 41p; + - property taxes at 10.5% of revenue against an OECD average of 5.1%; + - business rates of £27.6bn. + The papers behind them were not fetched: "Cost of living, economic growth and stability: How to make this autumn budget a success", "Bricks vs clicks", "Devolution that delivers" and "Beyond headroom". + +## Review fixes, 3 November 2026 + +The thinktanks-a review (`harvest-staging/ab2026/reviews/verify-thinktanks-a.md` on the staging branch) approved this family with one fix, applied in an explicit block at the end of `stage.py`'s rows: + +- **Finding 1:** row 0 (−0.2pp CPI) is context. Jung says only "The cost of living support announced today", and Quilter-Pinner "action to support families with the cost of living". The lines 2 and 3 scope came from matching the OBR's paragraph 3.10, which is our inference, and README rule 3 says the scope must be the producer's. So the proposed key `ab2026__package_cost_of_living_support` is withdrawn, the reform hint and note stay, and `value_kind` is `range_high`. +- **Verification:** no primary was re-opened (the session's network policy refused ippr.org); the contract checker reports 3 rows, 0 problems, and a re-run of `stage.py` reproduces both files byte for byte. + diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_ippr/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_ippr/claims_staged.jsonl.gz new file mode 100644 index 0000000000000000000000000000000000000000..afa3134524389a26273ce21c8928d96817bea8b5 GIT binary patch literal 1588 zcmV-42Fv*$iwFP!00002|Ls@XZrer>eeYKc^dcKDZHbaB$wmvrX#%tbnlwoIN-ViU zY9W%#>@KCSMu7Z9+h_lY`zQI5&MZYqjvVBI6itE353-g!JC}3j%<{S?oHHt}ozi9R z0(u6m6AoqX1bVTsn5z<_O6MwR=y6%+IX}QMSro!7qa+vBvYk|EiUvy;6=7}z) z;&XY!buU#m5qX3w+Qb&3c6CukaV;}+Q8)k9i_(YSI|%sEQ6aKYxmw~KljGJ=o9_=# zpFZ>WSfJFcuIaQ>c)rVvFz^6-Ai#>I#V@Tf2U)uYl7%QUG8d)vPaGDNHo`RU>x1Wz z)D9?WsmX&=31Ci{QV|qc9z7Rvj&OcD1g5Qp&>u2727|Ir@GexG`Fm(U*^y^uS`>&B z9EU|FG`y^Zap=)v{C0R+=z0g%$}NB>s_FvBC{>ag22Z(8l)uGjx5zbXclxKEDzw&8 zqgT~2bzTSFbx$M-cKv2H5KGZko+%y-2SGUF)m$fC$Gofw8%T{V;30W;Si+BWnVmqY z=+_PrAAnKR*G*&>BAlGFbvhf3V-!M8XM=D&4P`P(Lp)C>^XX`YqclE8oKHqljz^Pe zI!%*sG{(tjKAOiv6vJ_^OT$vLbPVA|r(Es>fBWW-7gUatcSwk9%h=hsZ~kPUx#+qe zpbe(PWe_PBpLj)3TzTm@X-U)_TbrB@T{u_Si?g%!dVQ*@%A9JGopGT>pwm<(cvf$H z2A$79mqF(+XdMRLVc;DGtyTr>FW^IcXZrH`#2f*i=kUoL#$mVKqJ?V3kqMV*G9a4e zYNg5yY+X?wIHGh(!6Z~Fr=@7TPWd)tJ0I0{g?N^e#_j}Tv)oEFZP&3OvCt9u#)RVA zGj-fcdzQrP+A~itTl5NyXh102sv^Zr0u=GGM}c&LS)O3!^V#Jb?Q&j;yk-|HF5S;) zsY>Yw5vxGN8oS*Z8L{R^_ZRW?H`ia=FiV?lV}{d@nr?0Jfrs2iOLVI~6=nM2k(6s| zc_8XmImN3=G-wWnQlB_fi@bly8trG#;c)Vjmr!&>J-?+( zFvR)(pm?ZDhp)IVR!G(q$mUR>LodDi2Up~C0`)RtT0<%dl{a08-0w6``nNYE;hmc< zel1 z|0VnjNw|=bG^4$d+};dlTVk~h^O5{I)bl&?`x!b-O%v^;>~j=)BUs$Y_vv#MSNn=Q zoDI6qfiQcyJ<#O+YTYNrdR=~>7Ikk9{CCiFoo8@I>{Dw$V~%^x7@@a70-zsW*HRGiMC)jR22(clBgnS9>Dns3TqZ zTRrJZLf2lx{Z%xn5^iwXDmkbX**YB+qS3~FcuW|sSdFm2T=Ml~XZP;A%fB=nc4a4t zC+YxS_UUyGxywWeZABo)MsMB{S|RE4D;yVzI~C*(FIp>9-qx$1d7ya>0^8W@}pFWvO5&!@WRR5v? literal 0 HcmV?d00001 diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_ippr/manifest.jsonl b/sources/harvest-uk-ab2026-2026-10-29/uk_ippr/manifest.jsonl new file mode 100644 index 000000000..e3622f43d --- /dev/null +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_ippr/manifest.jsonl @@ -0,0 +1,3 @@ +{"url": "https://www.ippr.org/media-office/unfinished-business-on-economic-reform-after-sound-budget-says-ippr", "title": "‘Unfinished business’ on economic reform after sound Budget, says IPPR", "date": "2026-10-28", "publisher": "IPPR", "sha256": "f834beaad7802472dc5f2e6f59738e3fb6ee953dd4357f7fc234e53939b1ea14", "bytes": 60580, "doc_type": "html", "access": "direct", "retrieved": "2026-10-29", "note": "Press release. datePublished 2026-10-28T13:54:00Z, dateModified 2026-10-29T16:04:19Z (JSON-LD); fetched 2026-10-29T16:17Z (curl, browser User-Agent, HTTP 200)."} +{"url": "https://www.ippr.org/media-office/ippr-north-reacts-to-andy-burnhams-first-budget", "title": "IPPR North reacts to Andy Burnham’s first Budget", "date": "2026-10-28", "publisher": "IPPR North", "sha256": "9b1e6ee72f3af251748526f7911a27e508d207d2f8909e9e9d0074a42f45cab9", "bytes": 58127, "doc_type": "html", "access": "direct", "retrieved": "2026-10-29", "note": "Press release, found on the ippr.org media-office listing; its £285m of Good Growth Funds and £1.5bn borrowing facilities are HM Treasury allocations and financing (design, README rule 8), not staged. datePublished 2026-10-28T14:12:00Z, dateModified 2026-10-29T16:21:46Z; fetched 2026-10-29T16:35Z."} +{"url": "https://www.ippr.org/media-office/ippr-scotland-reacts-to-andy-burnhams-first-budget", "title": "IPPR Scotland reacts to Andy Burnham's first Budget", "date": "2026-10-28", "publisher": "IPPR Scotland", "sha256": "9f328d542179e8f8f4a5a88d48a8f5a61680b3c2acb98bbc2eec95a25eb5b153", "bytes": 57561, "doc_type": "html", "access": "direct", "retrieved": "2026-10-29", "note": "Press release, found on the ippr.org media-office listing; 0 rows (its only figure is the £100 Household Energy Payment, a policy parameter). datePublished 2026-10-28T14:21:00Z, dateModified 2026-10-29T16:34:34Z; fetched 2026-10-29T16:35Z."} diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_tpa/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_tpa/NOTES.md new file mode 100644 index 000000000..3d73324e8 --- /dev/null +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_tpa/NOTES.md @@ -0,0 +1,195 @@ +# uk_tpa: Tax Policy Associates, Autumn Budget 2026 + +Dan Neidle's "Budget 2026 – what it means" (taxpolicy.org.uk, 28 October 2026), the seven chart data files the article embeds, and the calculator screenshot its footnote 1 links. Source slug `tax_policy_associates`, as in the AB2025 `uk_tpa` family. + +**Working files.** `stage.py`, the file meant wherever these notes say `stage.py`, lives on branch `uk/ab2026-harvest-staging` at `harvest-staging/ab2026/lanes/thinktanks-a/uk_tpa/stage.py`; this branch carries only NOTES.md, `manifest.jsonl` and `claims_staged.jsonl.gz`. The lane's local primaries folder (`.ab2026-primaries/uk_tpa/`) was never committed: the repository carries no primary bytes, so fetch each primary from its `manifest.jsonl` URL and check its sha256. `stage.py` reads `uk_hmt_tables` from its sibling directory, so run it from a checkout where the family sits beside that one. + +`stage.py` holds the rows as data and rewrites `manifest.jsonl` and `claims_staged.jsonl.gz` byte for byte. `stage.py --verify` checks the primaries' hashes, finds every quote in its primary, reads every chart value back from its JSON path, and re-derives each TPA group's Table 6.1 lines from `../uk_hmt_tables` (result: 0 problems). + +- **Rows:** 305 (own 244, restated 61). +- **Primaries:** 10 (the article, seven chart data files, one PNG, and the calculator page as a 0-row descriptor). + +## Access recipe + +Every primary was fetched directly on 2026-10-29 between 16:17Z and 16:19Z, with `curl -sSL` and the browser User-Agent in the brief (HTTP 200). The bytes are under `.ab2026-primaries/uk_tpa/`; nothing came from a local copy. + +- **The article** carries `article:published_time` 2026-10-28T15:51Z and `article:modified_time` 2026-10-29T09:59Z, so TPA edited it once after publication. The copy hashed here is the 16:17Z fetch. +- **The charts** are ECharts embeds. Each chart's `data-json` attribute names a JSON file under `/wp-content/assets/`, and those files are the primaries, URLs as named (three carry a `?v=20261028-n` cache-buster). As in AB2025, `doc_type` is `flourish_json`, the nearest enum for a chart-data file. +- **Values come from the data files, not from bar heights.** Where the chart prints a data label (`label.formatter`), the label is kept in `conditions.chart_label`. Each chart file also has an `aria.label.description` sentence, and the net totals are staged from it. +- **Quotes** from a JSON file are the element re-serialised with `json.dumps`, with `£` decoded to £ (the AB2025 `uk_tpa` precedent); the file itself may print the same element across several lines with `\u00a3` escapes. +- **The PNG** (`child_benefit_budget_2026.png`, linked from footnote 1) is a calculator screenshot. Its values are printed text in the tooltip panel, read off the image (`parse_confidence: medium`). +- **The "Download as PDF" button** builds a PDF in the browser (`tpa-pdf-exporter-loader.js`). It is not a separate document. + +## Rows by primary + +| primary | sha256 | rows | +|---|---|--:| +| Budget 2026 – what it means (HTML) | `9d8031655aa08411…` | 49 | +| budget_tax_rises_by_event.json | `62d64a909445ce61…` | 63 | +| tax_take_by_government.json | `b13087894131a927…` | 31 | +| budget_2026_tax_gdp_impact.json | `a96c8f8f29b1e0e7…` | 12 | +| budget_2026_chart_by_tax_2027_28.json | `dcf10b6e82b0848a…` | 14 | +| budget_2026_chart_by_year_v2.json | `db22d5e01bb59ba7…` | 72 | +| budget_2026_chart_by_measure_total_real.json | `7355f24b6b8a620c…` | 24 | +| bank-tax-compare-uk-budget-2026.json | `7a8454d612b6e46b…` | 34 | +| child_benefit_budget_2026.png | `e69c25ad27fbd19b…` | 6 | +| The UK tax calculator (HTML, descriptor) | `69a72c68640f3b2f…` | 0 | + +## Rows by measure key + +| measure_key | rows | +|---|--:| +| `null` (65 with a proposed key, 142 context) | 207 | +| `ab2026__package_total_tax_policy_decisions` | 37 | +| `ab2026__writing_down_allowance_12pc_and_fya_50pc` | 15 | +| `ab2026__bank_surcharge_6pc_temporary` | 11 | +| `ab2026__hmrc_tax_gap_package` | 8 | +| `ab2026__pension_tax_allowance` | 2 | +| `ab2026__foreign_permanent_establishment_exemption_mandatory` | 5 | +| `ab2026__foreign_sovereign_investments_taxation` | 4 | +| `ab2026__electricity_vat_zero_rate_to_july_2027` | 2 | +| 14 other single measures (Table 6.1 lines 3, 5, 15, 26, 44, 46, 49, 56, 59, 60, 61, 62, 86, 96), one row each | 14 | + +The 142 context rows have no measure key and no proposal: + +- the earlier fiscal events in the two history charts (`conditions.policy_event`), and the 2026 Budget series' 0.0 fills for 2024-25 and 2025-26, years before Table 6.1 starts; +- the other countries, the pre-Budget UK, and the after-Budget UK's bank levy and VAT components (which the surcharge leaves unchanged) in the bank chart; +- the fiscal-context levels in footnote 2; +- capital-allowance present values under pre-Budget regimes; +- the child benefit worked example. + +## Attribution: how TPA's numbers divide + +TPA builds no yield estimates of its own. Every £ yield in the article and its charts is HM Treasury's Table 6.1 costing (or the OBR's), and TPA uses it in one of two ways: + +- **Repeats it,** for example a chart group that is exactly one Table 6.1 line, or "about £0.7bn" for the £695m total. These rows are `restated`, with `source_model: hmt_costing_obr_certified` (or `obr_efo_forecast` for the OBR's paragraph 3.17 averages) and `conditions.originator`. +- **Re-expresses it.** TPA groups lines, sums them over years, deflates them to 2026-27 prices, or divides them by GDP. These rows are `own`, with `source_model: arithmetic` and a `conditions.derived_from` naming the HMT or OBR input, and every note says "arithmetic on HM Treasury's costings, not an independent estimate". This matches the IFS lane's treatment of IFS's own aggregate of HMT costings. The ingest can separate these rows on `derived_from` if it treats them as restatements. + +TPA's own calculations from statutory parameters are `own` with no `derived_from`: + +- capital-allowance present values; +- the Pension Tax Allowance's 53% marginal rate; +- the notional investment bank's tax rates; +- the calculator outputs. + +Under README rule 8, a measure's own parameters are its design, not rows. The allowance's "£282 withdrawal band", "up to £94 a year" maximum and the "£93" it removes therefore sit in the conditions of the 53% row. + +## Conventions + +- **Sign:** `positive = yield to the Exchequer` on every revenue row. The charts draw tax rises red and cuts green, which matches. +- **Units:** + - £ figures are in pounds (`£ billion × 1e9`). + - Charted % of GDP figures are percentages with `proposed_unit: percent_of_gdp`, the OBR lane's spelling. `conditions.denominator` says "nominal GDP". + - Present values are shares of cost. + - Marginal and effective rates are percent. +- **Period and price basis:** fiscal-year rows carry `fy`, with `period` the end year. Price bases are not converted: + - the history chart is in "2026 money"; + - the by-measure chart and footnote 7 are in "2026-27 prices"; + - every other £ figure is cash. +- **Cumulative rows:** the six-year totals sit at `fy` 2031-32 (`value_kind: cumulative`, `horizon` "2026-27 to 2031-32 (cumulative)"). +- **Present values and the bank chart:** the capital-allowance present values are properties of a regime, so they are `point_in_time` 2027, with the regime in `conditions.regime`. The bank chart is under "2027 rules" (`annual` 2027). +- **Rows with no stated year:** "about £1bn a year in the medium term" and similar figures sit at a stated year, with a `year_note` and `parse_confidence: medium`. Footnote 15's "£450m to £495m a year" (a range across 2028-29 to 2031-32) is staged at the years Table 6.1 line 68 shows those values: £450m in 2029-30 and £495m in 2028-29. +- **"Central":** TPA states no ranges for its own estimates. The one `central` row is "About two-fifths". + +## How each TPA group's Table 6.1 lines were found + +TPA's footnote 6 says "Our mapping of each row to a group is available on request", so the mapping isn't published. The lines below were inferred by matching TPA's printed values to Table 6.1 (`../uk_hmt_tables`). `stage.py --verify` repeats the match: + +- **By-year chart:** every group reproduces TPA's value in all six years exactly. Its 11 groups partition the 53 Table 6.1 tax-head lines, which sum to £700m in 2027-28, as footnote 5 says. +- **By-tax chart (2027-28 only):** each group matches its 2027-28 value exactly. Lines that are zero in 2027-28 are assigned by the by-year mapping and, for line 95, by the by-measure "Tariffs" bar. One line stays undetermined: line 66 (the summer 2026 cut to 5% VAT on family days out). It is zero in 2027-28, the by-year chart puts it in "Temporary VAT and fuel cuts", and the by-tax chart splits that group into electricity VAT and fuel, red diesel and HGV VED, neither of which it fits. It may sit in the by-tax "Other VAT" or "Other", and it is left out of both scopes below. The values are unaffected. +- **By-measure chart (real terms):** deflators fitted to its 19 single-measure bars (1.000, 0.9705, 0.9503, 0.9310, 0.9133, 0.8943 for 2026-27 to 2031-32) reproduce every bar to within £1m. That settles the multi-line bars: "Savings and property income tax rates +2ppts" is lines 61 and 62 (3.981; adding line 11 would give 3.816), "Tariffs" includes line 95 (−1.594; without it −1.776), and "Other" is the remainder. + +A chart group that is exactly one Table 6.1 line carries that line's registry key. Other groups carry a proposed key. + +## Proposed measure keys (all `measure_key: null`) + +| proposed key | Table 6.1 lines | TPA's words | rows | +|---|---|---|--:| +| `ab2026__package_property_and_savings_income_rates` | 61, 62 | "The savings and property rise"; "Savings and property income tax rates +2ppts" (by-measure chart). Same spelling and scope as the uk_hmt_redbook family | 5 | +| `ab2026__package_tpa_savings_property_income_and_cash_isa` | 11, 61, 62 | "Savings and property income" (by-year), "Savings and property income tax" (by-tax). The group includes line 11, the cash ISA limit held at £20,000, which TPA doesn't mention | 7 | +| `ab2026__package_tpa_temporary_vat_and_fuel_cuts` | 2, 3, 18, 65, 66 | "Temporary VAT and fuel cuts" (by-year) | 6 | +| `ab2026__package_tpa_fuel_duty_red_diesel_and_hgv_ved` | 3, 18, 65 | "Fuel duty cut, red diesel and HGV VED (temporary)" (by-tax) | 1 | +| `ab2026__package_tpa_business_rates` | 12, 13, 14, 57, 76, 80 | "Business rates" (all three charts) | 8 | +| `ab2026__package_tpa_energy_and_environment` | 17, 69, 71, 90, 96 | "Energy and environment" (by-year) | 6 | +| `ab2026__package_tpa_other_income_tax` | 5, 15, 28, 52, 53, 86, 94 | "Other income tax" (by-year); "Income tax: mileage rates, Pension Tax Allowance and other" (by-tax) | 7 | +| `ab2026__package_tpa_other_vat` | 36, 46, 51, 59, 88, 89 (the by-tax bar may also hold line 66, zero in 2027-28) | "Other VAT" (by-year, by-tax) | 7 Registry caveat (thinktanks-a review, finding 6): the by-tax lines that are zero in 2027-28 are placed by assumption and line 66 is undetermined; the values do not depend on it. | +| `ab2026__package_tpa_other_corporate_tax` | 49, 50, 54, 68, 97 | "Other corporate tax" (by-year, by-tax) | 7 | +| `ab2026__package_tpa_other_by_year` | 10, 16, 19, 26, 44, 56, 60, 77, 78, 79, 93, 95, 98 | "Other" (by-year) | 6 | +| `ab2026__package_tpa_tariffs` | 10, 16, 77, 78, 95 | "Tariffs (incl. UK-India)" (by-tax, by-measure) | 2 | +| `ab2026__package_tpa_cgt_founders_relief_and_badr` | 26, 60 | "CGT (Founders' Relief, BADR)" (by-tax) | 1 | +| `ab2026__package_tpa_other_by_tax_2027_28` | 17, 19, 44, 56, 69, 71, 79, 90, 93, 96, 98 (line 66 undetermined) | "Other" (by-tax). The non-zero 2027-28 lines are 19, 44, 56, 69, 79, 90 and 93; the zero lines come from the by-year mapping | 1 Registry caveat (thinktanks-a review, finding 6): the by-tax lines that are zero in 2027-28 are placed by assumption and line 66 is undetermined; the values do not depend on it. | +| `ab2026__package_tpa_other_by_measure` | 11, 17, 18, 19, 28, 36, 50, 51, 53, 65, 66, 69, 71, 79, 88, 89, 90, 93, 94, 97, 98 | "Other" (by-measure) | 1 | + +`ab2026__package_total_tax_policy_decisions` is registered. It carries TPA's rows for "today's Budget" and "the tax measures", the net totals, and the 2026 Budget series of the history charts from 2026-27, because their scope is all Table 6.1 tax lines. The totals come in two kinds: + +- **Sums of the 53 tax lines** (`own`, `arithmetic`): footnote 5's £700m, the by-tax chart's "Net £0.7bn", footnote 7's £19.1bn and £21.0bn, and the by-measure chart's net. These differ from HM Treasury's printed totals by rounding only (£700m against £695m in 2027-28). +- **HM Treasury's printed totals** (`restated`): the by-year chart's "Net total" (it prints £-1.65bn for 2026-27, HMT's −£1,645m; the 53 lines sum to −£1,635m), the GDP chart's £ figures, footnote 5's £695m and the article's "about £0.7bn". + +## Proposed metrics and units + +Reused from the AB2025 README list: `effective_tax_rate` (the notional bank's tax as a % of profit). + +New proposals, each with TPA's own words: + +- **`capital_allowance_present_value` (share):** "the present value of the allowances as a percentage of the cost, discounting at 5% a year" (footnote 12). +- **`marginal_tax_rate` (percent):** "a 53% marginal rate for pensioners on about £13,000", and the calculator's marginal rate. +- **`marginal_take_home_pay` (gbp):** "Of the next £1,000, you keep £417" (calculator screenshot; a level at the margin under current rules, not a policy change). +- **`marginal_tax_amount` (gbp):** "Income tax £400", "NI £20" and "Child benefit charge £163" on the next £1,000 of salary (same screenshot). +- **`unaffected_share` (share):** "84% of income tax payers will be unaffected". This is not `affected_share`: the value is the unaffected share. +- **`tax_to_gdp_ratio` (`percent_of_gdp`):** "tax burden … 33% of GDP", "Tax take". +- **Context levels from footnote 2:** `public_sector_net_debt_change` (percentage points), `programme_cost` (gbp) and `debt_interest_cost` (gbp, the uk_thinktank_misc spelling). + +- **`revenue_share_by_fiscal_event` (share):** "About two-fifths of that comes from Conservative Budgets", the share of the £123bn of extra tax in 2030-31 coming from a group of Budgets. The registered `revenue_share` is a group's share of one measure's yield, so it does not fit. + +Registered Metrics used: `revenue_change`, `caseload`. No baselines are proposed (`baseline_policy` is null throughout). + +## Not staged, and why + +- **Policy parameters** (README rule 8, "stage estimates, not design"): + - the Pension Tax Allowance's "up to £94 a year" maximum and "£282 withdrawal band" (in the conditions of the 53% row); + - "£93", the most tax the allowance removes (20% × (£13,036.40 − £12,570) = £93.28), the same quantity as the £94 maximum: staged on 29 October, dropped on 3 November and kept as `conditions.tax_removed` on the 53% row; + - savings and property rates of 24%, 44% and 49%; + - writing-down allowance rates of 18%, 14%, 12% and 6%, and first-year allowances of 40% and 50%; + - the £1m annual investment allowance; + - the full new State Pension of £13,036.40 (carried in a condition) and the £1-for-£3 taper; + - the 15% REIT rate in footnote 14; + - the £60-80k child benefit band in footnote 1; + - the 2.0% CPI assumption for 2032 (carried in `conditions.deflator`). +- **Duplicate phrasings** in one primary: "a new 53% band, for £94" repeats the paragraph's own 53% and £94. +- **Figures without a year or a measure:** + - "the UK's approximately £3,000bn GDP": an approximate OBR level with no year. + - "over twice its pre-pandemic share of GDP": a ratio with no printed base. +- **Not numeric:** the OBR's "High" uncertainty rating for the property costing (footnote 8) is categorical. It belongs to the OBR family. +- **Not TPA's:** + - image-6.png reproduces Red Book paragraph 2.54, HM Treasury's text, which holds only parameters. + - Reader comments under the article, for example "the cost is £1.2bn" for Founders' Relief, are not TPA's. +- **Further-reading thumbnails** (image-2, investment_bank_tax_as_of_profit-1, taxable_capital_gains_…) illustrate other articles. +- **The calculator page** (25 October, updated for the Budget) is a 0-row descriptor, as in AB2025: + - it calculates take-home pay, marginal and effective rates, and net-v-gross for employment and other income, current rules against Budget 2027/28; + - its worked examples (62% in the £100k taper, 67.3% and 73.3% with three children and student loans, 84.5% and 89.3% in Scotland, 2,000,000% at the childcare cliff) describe the tax system, not the Budget's measures; + - it restates CenTax's £640m childcare costing, which is staged in uk_centax. + - Its on-screen outputs are computed in the browser and are not printed text. +- **Pre-Budget TPA articles linked from this one, outside the assignment:** "Banks say they're overtaxed. The TUC says they're undertaxed" (22 October, linked from the bank-tax paragraph), "Capital gains tax reform was right. Now it's dead." (6 October), the Cebr VAT-free shopping article (28 September) and "37 ways Andy Burnham could raise £4.7bn" (3 July). They are candidates for a follow-up if the family is to include pre-Budget options, as AB2025's did. + +## Things in the primaries that look wrong or contradictory (reported, not fixed) + +- **Pension Tax Allowance maximum:** + - The article says "up to £94 a year" and, a paragraph later, "£93". + - OOTLAR paragraph 1.4, which TPA cites, gives the rule: 20% of the full new State Pension minus the Personal Allowance. On TPA's own £13,036.40, that is £93.28. + - The "£282 withdrawal band" is three times £94; three times £93.28 is £279.84. +- **Start year of the history total:** the £123bn sentence says tax changes "since 2021"; footnote 3 says measures "announced since March 2020". +- **OBR's property figure:** footnote 8 quotes the OBR (EFO paragraph 3.17) correctly, but the OBR's "£0.4 billion a year on average from 2028-29" for the property rise sits below Table 6.1 line 61's average over 2028-29 to 2031-32 (£478m). +- **Rounding:** the bank chart's stacked components sum within 0.1pp of the printed totals, for example the Netherlands at 43.7 against 43.8. + +## Review fixes, 3 November 2026 + +The thinktanks-a review (`harvest-staging/ab2026/reviews/verify-thinktanks-a.md` on the staging branch) approved this family with fixes. Each sits in an explicit block at the end of `stage.py`'s rows. No primary was re-opened (the session's network policy refused taxpolicy.org.uk); no printed value changed. + +- **Finding 1 (the coordinator's call):** row 16, "£93", is the allowance's design under rule 8 and is dropped; row 17 (53%) carries `conditions.tax_removed: "£93 (as printed)"`. 306 rows become 305. +- **Finding 2:** no change. The group, deflated and %-of-GDP rows carry `conditions.derived_from`; tranche 2 separates them from model estimates on it. +- **Finding 3:** no change. The 2027-28 total tax policy decisions appear in several rows (3, 26, 27, 147, 169 and 237, 0-based before the drop) and the six-year total in rows 4, 28 and 265; tranche 2 should deduplicate them. +- **Finding 4:** the quote convention above now says JSON elements are re-serialised. +- **Finding 5 (optional, applied):** row 1 ("About two-fifths") uses `proposed_metric: revenue_share_by_fiscal_event`, not the registered `revenue_share`. +- **Finding 6:** the two by-tax scopes carry the registry caveat in the proposed-key table. +- **Verification:** the contract checker reports 305 rows, 0 problems, and a re-run of `stage.py` reproduces both files byte for byte. COUNTS.json pins the gzip's sha256. + diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_tpa/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_tpa/claims_staged.jsonl.gz new file mode 100644 index 0000000000000000000000000000000000000000..c14eee2decd5498acb213d3e8ee1e853d0e65758 GIT binary patch literal 20674 zcmY(q19W9g(*`=hOl;e>lZhs_ZQC{`wohy)C$??d=ESyc-tWKbu60kZv)A6$&r`el zpmujvH9;igkN;kvUl&>HG5aHp#aG8vpg&QWQ3{lAy#EblN0;-pczLw}6;y1DcGw{$ z^O;W)#M26?X>2gBxz8h`K}@S&WjwbAIbfY#_GG6{3)AptD`v$oM99oCix%WQ%UWyWA-AqIq zy^js#$aapvAe-UKc3}DX;Qm4(KnN#z=Ek2qCddkj=WyKz=V!Sg{z~L@VG@6h8hs7n z``ED?KxH0p8Et#xZ+rRac+N)p9oX&{q|j(VVrHTvAR81WXOqywPooCSZe+dqIo032 zKhog5)MLF5es23Qt=12gAman7`w0Jz)Rg&!l3_69p66<>ZouKEQNj+vkc5E3GND#) z@5nHP{gsSJoi?`s~At?W~{@ctfo0M~fdQ-58sz zG~5N&BX*&SJ`bU7&5d*$!T4qV0k8T6Th)8`0BD^bh;*;*9w@$Vul7TB%Tc8>?~axA zzM^jX8(K$!uAQ29Rz;(cPAjJ^wOAC+i=`DpCc~qqCYF5QS#>V4ri=xwVx22dlBg>& zdf$mW&))Nj)mw*v+>pb?J$JzkH6pMjACe>UW!`Uo2pP_~Vg6d?|4QWDXotE(p`G&g zbHti-hfFj^^1C@jLk%XJWsJHE8IP^L&Zss%0T^~7n|O?q1P`}9Bx?C)s&$lqxumH| z{sB*3p)t=z!1wYW8yCd~zd*7gPp*JI{0$*^6VM&8A2OVweIPJ9o#n+( z28CqpAugxsjeXM z1J{ite~(1|#`H∋}D!`k$fksP_Z>H4fIRhxTypy`p#@3Wr3p>X2PA1@$}J0L36a zPuqcUiffdLOC!79eR~ego$;KFRb&LH-lV!O69#UA4}D_9;9jjx+*Slo$_{Yd_x=j7=`#+XT#BBtbz-#WSy-sPqC1%SF>b{7g68 zslN^#{0;S%HehTBN~I<3T@QuT3q+E)o*BoatnqX_F?sZ2m)0w7(vKOf%L#Q@N~HPLB3> zWZk{maxeB`9-9nHYMpz)s~v2@2D@PkIK;+XYPWPEuTDz~*q8($w>?7I7QDW8oPSRr z2E|riAVF>(eg{XWsZ7?TEoJ|@{^N+YEPbi7yoh5&&82Q6&3e(VC*7`gvLG^>-ZKW5 z<{ay!3$kOu*2phJ;+Efbj^s#`VitqaS_|-!q49pl5E|U$AOy& zcxX&dv)z%%fv2w56E9buC6yOdl}G&#e8>+DBzoOtG?cVN-%y>{l)J;I*b7@%=f!%v z>y-VC`*05=-bXvN=Kf3XOy~g`onJ`#+Zm8+2=DY8C$ z|Db+}bNU(?zK7g}&_43B9B6r?b{jZ5A~>h_#B=jySzldIRAbdyNv>Yvw{tb85#5P_ zhA7(bb90n=A_hEj&6t%E^bUk?%0XfbBiP^nnw_*9X96E8_kTLYtn0T!>o)B{#ItRz z6HC0~af#+}srDV9h*ZC-N+qb~o@A7N8fM|MfcHB(yfGc=dO4fR8f+i?*WGWzf8hgs z#nm0QaQheBTxOQF@rK02k;M@Bqv#=>qoJbjj-tyN^@hWL#7|>r1ttm3Vk)5G^gQpz z8LO9}6ed!HIA$N>3~XPcjwK?QG@H?`t~t@2U6TJa(Ec%Ye5um#DPtVssQRhKg-wPC z!5D2V*Im&+o>oLPW5km^XVAd44v9c@oU?RP@(y5*5wyUjczF8=6=jZ*H|FefxW8Y3 z&)?f?f8cReE9X);DI>Atq&zJ&FXxX?`62|B_7RFho40$5kJrA{rrbNt>1EuSKDTq+ z!S%D-y@{)XAwD+aOE z1EHGBJY;xIy5)@|iWr8$=7_%!MMhv2wk-XR+XzEb@@=MxKHQHo0zB)4IFZoH1>`iP zJ98Nb`;^VS%R%nVZq)Hx9`ccYWf3Qgsmpuu&X*&T#PLSs)-psB|1|e!$B7;Ls7`*I z$~XX9AEoDYT=hY%!T5n}iF%)mAlrtZxqmH}fCZ}#$@uBagWhpS#3sN z%~TaZ9=9j0+V-(L*u8^2m|4f-S65W!3GT=HGKeR6%4O6|xGiU|_rc$$i?gr&1PS5A3A>@>^1y%e@9Uz@Yqe!u~dafcAD zk)Qr?(A50jzojhIL`y7N!)Gm+xMzg%(-(w)k1(f}@nf}XX*a~KN^L(!LlVYr^n9UB zU~-Zl6#I=!*RSIm=s3PSvp&6`8d%$Z7bJd})N!bhs9C~Rdtg` ze%CBY$!BpHXK;V8(Tz7gB=hZ_Hp#K$2$+^vwrRKbJ@!+=P`f(9aKVN z)q-ZXH)gf52w^b=%+x~YrytcdaHC$*iDQ1Ad1BKq4ir;y~)EWvx>HwoIOPgM)2D& zJ#J~!Q!6!T)73Q?Jva1UR^VT9nPw5!Wtof;nJN8}Lk7W|(zvz({VPgHqr$nLch~ww z17>WtAqUaPeaiyHHiwQL6{v@`f;{v?`}4p14TtU?@vUgUqbN$T+D{frpH`1Upqn83 zv2a>h95j~Or!nqS%s9v>ZPKZVWSoxMQqf{$oFJRSQ;aQzUdWi-CbcOi+;LxEcqnj! zbG#-u?1W&UlQP`rvRwa`Aokzq5iGTk25H_qwu|Tx+RD@u+7i(RtqZpBI%WE ztxg;}0n!fOXskfVHNOQ;O%{|N`RdJ*SudhniWsgzSzi=9F@osL}HEJ(Nk(usq3 zqZrydUlVLz&$3tkJ!h#3Z;jOlTmMALTDbi)SJ>zxyK0lwPIa86RP-t_p>tt8xRX7) zDqVw`OZn~qhQ9NM;`ne((xqMY>dzJac??sx_DIfVUd5F+X#eyyP6~ZzM=3Z_3Fpb& zznn^xPSNp=kMrr24be{RFjHGAAUxZ>g+~PYRVvR)=()>^N2vM~2GLsdyfa?p!tWvU zdXoudz!z^nF5B5F7;0}E?*$=4#}6gz+;~XD=7@^lJA?76r)>B$wn-auxrEy1Ob{r$ zop0yDTRiUvGT_WME;OJm6_i~FW{HhxP}mq#BHdzJ^*IABuycvPA^F6)BEHmM2!vX` z`u^z&xgzzkqgHq_Y^38ID$%9-_4d+kzMV(b$2O8gQ>FcJLc?d2aq>xusObR0J;Nbk z?ZdEGfjKu16A2)q(+kg-X(|Q2Bwf4j)8tvT-*bqFlsx8Hh~M_16HO#|e|=SUA6aTR z`8Bn781Nce5T-XKNPMC%GzK6cuX*Hc15bB20fW#L0;ZI5^1Tv3`&+8tQhp>4+$7@G z$h|SuGXtE%oI4y^`^v>cR?PR4EK`AtFsyQp%kS^xf?iQ|vy0tNsol~2#(7h} zfTx81h05U6_&MAwc;KFP*UjhU#ODLD8LS^b+C4_v9q(Up4y(+O7QG}eA8GswSZ=0N-(o)m|%jMblhq! zR75@CbguI2GLwP@rDCGT@#Ar_UCqwVdwLE?c?ap}2Eo1FW%|pF=`F!TfOat&;Ob^n zn@yhQ#G#>2v4p^ZPG~?jTGoaldlLw7aaKc8^6aiTJl=Nd{`truHVz4$u+FSqORA!x z2G!9jI{HW>@_SgQZ&Eb$Wtk>lE(EL5d|UoPfBSld{6_|DuZ~XLUCmjiaTx;XIRr=V zXMdvgo7#;fOf8Kyclg_EyWw&<+rl{A)Jn4-^lXX4>+?<9sel~E9yjz2vgsLI`P2`b z^Mf9&Om8n>0;|24h(;jCcFfdxBH38(4>4GUyvuQsIim0RgmsA$!m1neqenUDe9Ad> z^MKP{bOKG&Cun|rcT1h(O(a#bv^7FFxFRklZ*S*skO3sQBpg3vr}B#%T7A;P$&VXL zvil%jD$fJCHLwJ>E}~l-$b10zzDwjZL~^y_#yasAZ4ns zRjA$cvW$z))83&Wk|<5~!1Poi)LZnk;=x{_f2_4D5&N zw7^ARcG$#*6SEYoLe&~TZd#>Y@*`YhsXFa!q0g1X=pRpF|GtWDUPyvK}unjRuXVH%mex z`>}t`dL|lUl3xh|b+Y1G}Dn6s`Gx}sgokieye<=DqRvW)nCsjS0%X~UbqEj zY#dZ@&9_XlPiq=p5f6FPHXi*1eg8<(YisT}Z;UWNk0@ecFV`~!$wJ#h$T4x(LOJYm zLnuQojgL_}J~D_i2Pjv`@0BC6`T3NH=9puggt&GOamRqs41Xq?PJP^F?kfhd%k?sC zN0LA!2_fW-ffi}1CA#qa>$I3$(lI-&w<6T-wum0@Ih1Y!92Ymk3= zKw>hxLHNBAseAhZgpH!#-C`DP{^m}4LLKFo=U>ePHX@IAC@lK&pg=f04!7H_l>s&< zWr2x|RTs<7ZCaD5joMLqrz>s#o@3uuW30Z=o{viTI>WOqeA^(KWPB|{;kOID&2EBw zk@J|&oJ??z0Ikr;Y{cM%ihT#G7Kre6@#+hu7rKJg2afG7nbvzx%5B(G<|JLi>_D<9 zVB{v#Ao}d^g>f`**MsO%NepB z)*UVM8;FyQ!Gwzg`jGpyL0dlp-#hhkR*G|E*Y;4I!#|qAfh%ZvR^~wnJd&C5Zdbmy zM1qD3+T-Yj7Wu!ldEpA}>pNkbwzoCg81+{+r>^7~9$yqf>paE+78g+ptPh`vjxN`} zMtU0%V(QeDBeiZyEl;<5O?Xx>O4I0a)x;Y-ubRi3PIkSX{}drNp6PwP^S7|hs-vCm zYIb?$udW0b$5~w7IEGND-F0f(tYXf%7_nYOh6>%yqhl0Cfl(@YetQ zNgDRPNBX6A-X);z3uDyGRdwT^pUw-v@|o0eVvpFYI#uT&BlP~`cd!RG%Hyk{Q}XdG z`xEvuzn~BLNmecPT`>8ZIG0aA!ol<=Rbl`$1A|^Gc zvs`oc0D8Qwe_50N{(#Betu6h0a_Si!(_){OA-sy9*RY$G)>8x8sRhdHG%FMFX_h|o zQh@!hIG^G{?dv7W!6(&%se7gYx`{PW4iyqil&EPqBZcD~o*BRix+80@-mO;9L40%sH>)PmBI&k|!|1NH}l2>r&Ht9^#rO5ZP(^`1f?-l7g z&d9(kx#-7{CqZ*~V5|lInpp!T*zbM$t&2Ou=idhCc6c_}Pm_TF%Is7QjZ%mfH`L2Y zq?WsTrg(y)&VSA)qL{lu`!4M3&%@Pv2usu9b7n#uCxc-uy20iH5!KHk_LBOiEjOJU zpOmf#VBr$g3s)5N0kIg=kJXq&kaI(&eVDu;iMp?sV#CM33Rj&9*poE(Hx zv8&g_M*3~>R;tl8{qd@tP388rGs6G3Fcp9PsiR)(Ugli<41XDsC)r#s&03Kd!aN&i zzT$$NUi37{_y9k5{pZsq-++u&IV?oDziM8dK}JaD>Ltd4_@QzU5$^9FCypL>gC38J z?EmH-cR#a7|29*&bH}DM=%rVnU?WARH*V$Mp$%bC%llWKDYngqnu4J$C%(zuy%a~` zh|sf&G?lYiU&`_{mH!6y<~mQq2<8nllsGroV6Y6Zr=fe*7hAJ+#Ur9k<&Np&MAw>z z#bgzq(QAUOS+2gDB3p|?$@S?_c6sn89-Q128&1`?n*%gaJ}9|Nw> zJNGEU^zir|$pL;y{?9)@eSlO#9zDJuNG(pEeGD4&{!jk6(6+#r@knZ!(l;$7Djel9 zM|TCLFd}Qm(X3ea!tg#hs*3GPwh825H^miBj$ji316rzldtOx4$-D*;16a-y#x5z; zEdhhYq%lQIBMDSrNrMg5Y)b95Wj|SHnt(oaMd96a$_m&-%xmHnkL2MK1`{6`|#t?Z!AFu zT2T!-WfYkRYE^XK5PC4v4puN+{Ld*rxvp**4vJ54`X$1tRd!}XZfJfW>cmfPEMkAq0a8cNf{Iyy}rWQck4@=VPb>&Kaei_ zHz@f(keSw|`qDjKGPHqpKbbolCpA69c)ISA+FuYd+8ljb658bx2bu_zy0*n=6P>j) z=eG+r@5Gu8jPW*Q^#8k|EyuU6;^uE%SKl{GcB2hPU*hPg)9#P>2h&%l?W)82E2+ji z;3K+5y9B0+GIgkL1Z)b#|z6isxO(ahJ8d2*b3EZ%3-JI^<{> z14c^|bSR{*DruxaM4Mw9*MAmghh{>^{z+oAQO2ptjLRQRi?ufuY@o3{%RL3_@NCM9 zxiy>67O1De5UB@Cy8EO|qJ4E^m5J4D>DdXfL-uXHu>!cr>k~qCmte1Z#_F9-Jd=@VkvvaESqJI$*+s%y zcMezkn7$_%T);jrK4C9i%k*8rzn>o7NJZw#v6nyKbpvn~viljZ=BNC(KZaPn6R+FB zx=lUa=L(t=;Viz{-rsDHO4s+bbuMPLU;hI{>$O`4NdIT|0^2Yc|97A~++e;>X4?H6P~F_u zK)^7Ai6*zl^32x$G7pRIsJkk+F8~k6xgyF;Qs>Rd8K+zFWE3KpoIrM)zTxS)aVkoW zIXera8{)rytzcK&Zb^^#gq!mNAHnZj&fE9Lu&=TC%oehwyN%T$cD)%;Z2&N42#TNa zIz7tu5wkliu0NiUDy}D=$?%wh>bz%3oxml0eyEbucP^uq5k<^tmMjOEGjp(P@Ap!7c6Ay=a>UJVgLZ80sB8f9Vj4@%1`%i=$TncNEXHVV0!{m;jSh56D%-b!M~Gb>2`aJtp`shY9rwz8LDILe6MLQVPr(*$}@_|jACjG_+I(4!i zIhW#Ry#AB*V5ztk4{qL*ogu7cAu^2gAU!4=y>aFUQPdOWoyXNKuAd25mA0R!+_yDm zx~$)*^OPRFPHTMEthpLB1L1ES7vDOf%p0fwlyVmns;=9YvqIFG!ROzT21(WPGG~ZN{S9+bmo!{#Y_qF#X)? zN-2+_G;hkXryk%rJ^Zm1j9lu zz!B|v0kR*!y%?dN+F|)%q`qpW0=#Q#ckR-ao28T8ppZCGx5TuyjR)Lmd-+`R*u!JycTZ4iO_aP%0qS&495AQLowzV=ouetHp zARu)|ig9|XF7$8I<_i4aOeLtfcbG|$){gjt-X!m?g)P}(K=$N2fwSa4NNlL^+%sew z#Aqgvx`_Xa2#4LGo5dmJU)Eyqgj?kl_!B72;;+y+scURhBhK6W61uNaxS}V=AjP40 zj#$j-(&Y<&t)4lX6oeUt$w15=KqBcm~ECun_g+2a* z?at+L@>3F2(86&2-zVvVd9CM)nHoY}!BRG(X{e{Mt8(zd{z3<{$^2NHcjkd6*^7C# zx3X=ak1pScsFvU9-A4w(*_$d$oHy!3d{xp^ylf=CtiyVOcY`|LLH#(ysvph@myyZW zX1mj`Q3#vbS;>^^Slv2m*)byz>KwWqC%$~+9NzVOST8a1rwRT@JJl}QeT(sv8AaIk zdva{{Lp%Oa%yqj7Y_mREE>Wwi_jHIlrM--xRTsS_+RZe!G7hf78*33kcl1wK5pjDLjgpZ*L7xIY&xLVtZwMTdguOp%VAZOrbt{Pgu zf5kC+1=QATr#618xjkFtHD-{=TlU^mvOU^ik43SK~kH2M~(@}XFcLa1_Eq8<99M0HD*PLc_!jq$E^O|W zfJuW=ch&J#<P>SzVqxkskfFvF*ZGID0q#@YVYaSWYB)sHIp8$V0s*-z z%a#FfoKRWqqoAksnNK30Mk|c&WKX$Ff84p-IC6c|I3v49&LAcoT>0_^0F0XXX=ldg^L<;W&rl%QTT zx*^KI%LZOl1cfV7C3nmu$fhcg$!N(wl%US~lm$JOssyP8sAq#D&B>GnNdTP9l1cUh z-{kXe@`FY*+&5Vmg*q=q9x6+DZ-ZK)Z(^PsfcPsLMiJ^lqX6;0ZW1G?f&ukDLI7vA zWYP`MgB;nxH~B*&7Uh4v)CS+=Th7Ln{d=^5sYt%o4Tto_vA-<@rA(>Dr3N{twxe8Y zi_Z8v<3HX(^T=}MxWpL{n2^$=G!_r@Y_9#*-b>t9OVMvcGL&&v=>=8nzZ^v<>YA5R z79i>}Dk-=^CiWmq+1*we6smZ3-AyY|?A(YTrr_jGMLO zTph0wSJ&r^nj_s-Es{G2W3uq$eC7-0CF>H;LPoaG>o9UmIR5LP&)|#GvE^MIr4ll> zgISH79!h1!nKv|>`2Wf<7zQ)Y6r^bK1-4Ret=0T5f&-*3Q=MjXu-B*!wH2;QOiK#Y zzB6}znv36YmU>H)ZIK%28$5nVZ@X&#Vghdt?jE<>5yb8FeLv-hHv%dKtwK9x6UzQ{-1|84u#QBK@S_MIix2*bEZ^-8X0%OZ~TjN zNcX+~p4FTe8tdie*G;QrCZzX1CV^KQ|`4WR}x-VrL{(^4_K93i?L6wQ_z9{`-9l`pO*nMK_eQ%iFJ&kLZ9S;!iZ5WaQ{{YZ|R*VD*w& zgSD}8E|nbo?#{0Obkz*{d{_)^piVq46*awmjG9q~3p!$8Fufef*oE=?`iZlOo8E-6 zyWQQ<6Wj5AC{iT5ToLEC%ZMuipE*3*{7}p;#R}aXab1w|YY?m|D>5*Dm5MR}P^?Os zE_BF6*(df2ErOqAC?6E7i;aqasXz}Op<0qFei@gZXU>lx(%voK5Ggd<#Y%~9ucnQQ zdLb(iPm9nfIEy_tEr7%~r$%}E4>ge3`+hFhy@%P}LVm+rbla11&gnfpuap--q_SIn zAwo!@oskmX6W-TBK8gFMh$p8G3+lshUg;A?gr@`Lsfyri){m)o&hY1Ck?))eodp4A zzexgZL9QkJ*kh$)MSWN%%m$Y*Kdk$`M^Qzz+xhy$V`0sTPt+;^CwN@HPx;|crOd&} ztSwBP|C3??KR{muk*BAl;i&#hyZjL$M*8v_I_iR>A3Txd;kT__R`p8J8zvW(aakERao3T{5;5sgEwXPeLb_$7+uTOSRBEdz&pyKBC|+K*kVZ2 zls6VZeb&ar?mUexS6o#X%PI$Y?iFGw)GfdkqyP5gM7c)Ff_GGvKK_Po#J-XZOzDu! z4S9WcN4x$=3%t7ccgsFakYmI^R=^YM`FiwfYU}r=f{$IWV~TO|^qVEa)i{}^Zr)Lm z(j-B#VOPUT2fY9{{Fm2r_qNH@xIQcd>}Acd$MhLFUgl{O`?K4!x zlUM;y*Yx-i2*EaDUigefW_2QCCfR4ea%7gz50NcKR5OSBZCvxZnDP!n__T@4L>-ZvBAC@f7vv>7kS>E0kL9ct$(NAy9 zA2w%GS}E&U{Q7dmm)?OEYlVRpd7;f@zU1^scKEAX{ZrMBDd7B=C^qV=yWHWW29oE{mexE&K*r(GRsq9L>Q(a$r z9B~q&s7ZyECh10dawE^@o$U!>BhZvs26|o zd|UZ;*QVRXKP{DO^YJpCmBZgane=)GSIm?QU)0OrpqNBjfIQ0BOd1HGV|+F#6(aEQ z^6}wOFYLEw zuf7pg1MnZlOdYLj^Cme_C3URKJW^N^$)>G!rvHlBu(6WM)s1Xev!cWj%S=bA%sk7> zWfp2#u_Wee^ULdO@DodD^$0j9!``ggav*3syFWY$u^0Zu*W>VCP^u1=*F2$qeV%Mz zgE2R2;3Go6werMJ{qps=iN%Ay*)3VNh(64S#iw00Dl+^cl1q-r7twAMU-9IM?lH71 z_#-X^Bm*_bk|V4T0=@ZBAlVTl@r=m%U&%>}JQS`5NRfSDILrT4h)B@FM+SGGM^Dn7!1hQA0_GTFbKZjP+%4yf)rbEK0O!zZf5KW6((#Cdd>32 zL=5{qpEsWIoSrXtwe5p0M*x-p0L`LhmUdamD8L#%7505e|4%;HG~9DDlr+a~Gu+?WJfLy^ zS<_-ekPUmX{>43hYG{FtHg70>dJ-!9y0nwKlcfL9@iBf4gJ|~A4uDiL}!S|f&~31^YUzVAe(Q=M%t-_{iOZb zGCaZc)9>l8j$ejq4b5ZBGULCcj?Y&-tnq0Y`{o}&Gt|qbjof4XUHl7t9{ZU~ST|ch zo!7O%&U%S01z&prQxohLk&`{YxDFm848fea41c+#*(QuT`bjc6G6N zm%fwVk-&qIIYuPJIR!%xS}&~1iHm9cQRhn>z6o1?ge#H;?I+_|P#wQ2Yb?J>3h5#} zLz(MaG2&g{+C6T*V-uC|;S6j(2iS7}^y?@IHc?i$hr&NdKfizPCR|u+n7rBYZH|ImGwt#|d&qTfH#|ggXtFeHpXiR;RnW zKk(Pmr5xQ|xfY@X9ZNBi`#VLvDjqMm^+Shb(-kyVS@%H!G9cH$Yl5{Aq)$Rn)xg+o z(Q(Xbnxnw1auM67&V-qVKmQzCe{U74NKZh-cVy{oom(zmNO<69^;zsQ>L}LT*88V# z@)6a`X%E0^L&Rai#Y_}S*&RtIj~0N1xsdx4KjUch*zm$OV+lP$}5g2+uwe#2Td*885nzNTVJV(8GMAySuX}=KjtNxP8DHD>k zz$yaFF{i7H-8zOLMfZie_v{^d)61ukxM)}Mx$;`PNSYeLV(6aQb*Cq=PG1b?Z6BBb z(<>><6JcAJ-9`jhGTm=DZ!z3EKmt8eYF306qIvuLHlM$cE}`RXV*++P>XW|QCzg<+ zix6nD+%rQPtYEjiErTquXU;n%7YQo~8m^t;Qp*MSnGJ#+XuN-@kX4L8*S(rak@fu{ zNe>ZYHH9O=j-Mk;G3Tx9eEUT~B9itmn>1Fmjn{FqjAxEQNo9^!1J5e?@>Homb)}eE zj%QVIqRiI3m@hqw%#HQo;N<2s=nv{5wdr!lfbfm}d<#{s^fT9SqGAg5A%dU%pMaFG zf@uD{D;PTTXJ>ruY}O?ET84K?y(HyNso8t}W?qiL-nz2hdfiu|__>CUyhTeD7f&)p z8bzSz(Kh>nB)VmAW6DCrLV1y8QF6|fA{;WX)9@!?fxMQuwp_L1ylqet$?zL85HV1m zWvKG|CBAUJ#j;y@wYgkaQAAOtOu0;PQ?UrTtZz-~`+8Baj2fuS3g)g4bU9H4NR)|} zm8(#<+7`_XOkJyc{dc|g8%*63R9;a6_-G0622vNQ6sjy*0u-B7PXvKAj|r;rs*Q_U zd4MNXa~(VlmBz6q;%0)%Ix?y4vO1D4533V0kGH96{#?deLM>I8ZdfZJH&VI}6=~nK z#^lF}0MeAnPWn+^3=+wgKZXm)!&|;g#Ao>73Qq&2krH^=%LNhdo`qt@V?epskt{LK z1p(fG5x^qJv*h&8GGaA+C}fSRQxwwMop?eQ`8ZEwms0VV^17{f-1zP)J}%+{yio;= z|M`84~*MqPQXHqepB$Cb!+!MMLteW548xHGVC$FoIU-J(_9`a(g z*Ry{)YD5EPup(>gC0o}|Y`)5y@~FaWAb&R7MD*xO+jDRyj$#S9!mw?zt^D*=$bRIocw1w~yIrF`SV%afe-jh0yf5 zao^bFz-=2l6MMU%ZAU1gZ%54WdK1)8Aa!YlUya!ayxqt&5%#&M-RKy?=^dB)j0_4b z*itke)+=Oeun61ayTX+D^qdR+Cj)%Tq{)RZvK8CGe8rr4V7+>z{|&C%H#f~{Z;>-x z=eLV>m2_Zu-xe^YdJ!+7dEz%;rM$;rB)3caqOD$tv%JUE83YC(6obvn@iD8&4gPEa% zG|S#EtiB55fmxQn<7y*b|lGb1J6qsg8?j<<&{$!phBUMMgNYKkFSmmIt2|X z7xRk$gBQQc@10x;#^78kguXznWi@q2;fnAW=)gG~aouWY_q2sW-r2@DpObunG9hdt z7!kKwokkZ(CR7?vjEr|`kuN0FL0MvVO7;jlg>fpe4si^>0O*t$qu^$Bo-2ts_brq= zcmnF=jGTzQE@=2Dj>YDnV)Lt*+>`ham*`!*3gh$7l}w!`>iXJ=ZRUe0tHARkQ`S8% z(#soPlKefr1kX|S{9$shn_YhU5YH1|TkYyWEJZM?(9j$$bAG z-G=@bRWuIt^&Tn5*^iQ*0YpjNZ=n6tqW zuY3*=LBqw+&;8>scX;h@rV81EJuAV&gS?`A_)($NnHn^o*GF zY_IvU^iL7>?Q{HhmDN=TwMYoRB9&s9Z>h41>h~G*Tb5eU$deeg2=yWgpnT-&`z)sW zt{gS(OT0+Ds4Z8qoJk-1y+p`QwHaap)LqMUe)R26Nx4cfZ?nX!vq2ZX(V;lRPTY7k z5l!2Kig@F7NzqNtb3ZE1h?XT3!f*6_iz6paK{Z&U!}(hm7ZqibIwcEPz68A6OBD+4 zg9Vb+6aVv4)m7eY72t8cl0`l*GM?kM3dQe(Mv~bR|9|S58nd+x*^#nEqj%!YBhqSF z|KeO?giB!ZNdV|szE-l0SU2KMBiL$LM{!6otR-OSV7pXhT-VXrK-nUTOL3QxQ`M}~ zxY}st5*QM^#(QzGi+_A4e+ct*r1ZAg zPNsr2R-|XLp7KQYyB z?+LCoVLT@O_n8;BL>2XITfBDs!^OFyh$&|Ra6BGZ=SgHuKuL4>rgSqe?vN4&O9ly;g)j&3i9>xs(2u8Bt?OSr1bRu|*# z)jJ`HKulpE*y*ryOq@IHHvHwEvIH-9vX-izER~MnwniZfRm zFviy60fXwKZvR_{TtlnX)BD^l7jZ{EMIdd>q|gE>qT5bse|)H2_Q`EvX$?j_{A+81hR8eS8m9guLP&`>so~Be2AnxH^VZ-RgDxZ|@Dr=jQ?jr1fATeo+I9 zzzN|x0_JYm#f4eBi)vHiUDS#%lr=~69F0UYeBFn?MbyfcZ_~P&33qEl*agO{T}ne8 zJe|ZD{x_>4?4r`l>u{;K-Nd8@H2`$-V+Ceovm zEXh#$4}T3o-3{{MLag0Uwc$H%Ipd3Et|nQ~Sul>s>eb`iB7DyxH-569j3klUFfY~4pskW=8uWt>EvfuRcFSpt zeqw=z#}Ts_Z$m7_N4;}%RZFkZCPaFIOg3IFmoic=+_s>h!z_7>%f5g=u2|$Ho^8O9%>2&|9RCHbx>bIIUt_Rb4%nYFgCODR*U% zoFf}^RLP-_Z;y5^>u)Pe!DC$YjD?V1v{jSn6Mm^UMIpaEif43DPfeQZE0jN?j@Ft;vA9+66@9d0bd_sp$#ef#y&cam+AD5x6BYX)%=oX$CkmM3EvY7^} z1Ti8rL_zG9GOFXL=c2Jh1w^fFXTd6jugT~T7`UXbZ>M&XpH2`(YHv>=O@4r?xDbOCqG1{<8lt4CYW1uB-@Uq^E{z%oc zh2!M+zW`(*GmMO?tVhHAM&DcS_yq9@+EUU7yUrby6IO4jWb}aakUn~Ta+!L^tas-c zmfIK;5?WLsWsVhN;X~!P-Q1nP^Qhk)h#ZdZ?%9W1S&UtQzX_=dJ8@z3LEC1i!|w5Y zL$eCz3%D+cjol@)Fqb@`c~%G2RR~E=>k3HJxsKpln76wjLhUeweYn@sPI;kW_qJ*g z!l`LAke+8pFW|<|+mF^AcvtsO-!!HG?5eU$Ufy?5_xfsSh!dnQK}6pF{cd?jaBGEi zr-abJZCfnl(Bi>fatFfiR{E0WZ<4G6p}_v}xF7eMwE})o!&Mybzvc=ItV+q&Kqi~D z8%8J-nm;8>{2lk*7BV%o7Yzl*mNsthG|qrOTaz7Kd@M4O@_uB z{se)bTzYh+zO?+L*f;=h@J-s#zF;~`FRzkAvoVgMU6wDrakr*SCzePzkRiVvSX5rD zqbMfwD5zyl@9Q=}lP%kC;4U6dN^BqG_%pU0GFczYWdO_RuVM;{&FYG0lNGD( z;RQd*X&59VaS|16%nei*%?gu9YIBqL^uE`h|LVmq3+)AiRmNc`7@C)$N*4HbSr0&1 zlB&YYW_gaExxS+ct#jWGE>whJgj|krHqtW9a1lh{wBwep&xOnFogip+#m~>qQh#so z#6@XxOOl+veYS)THzYh@-?`%eafDJ%*nPP-_lIYCkLosFQXax0w=)g%-x0LnN3 z1hpZ>%MaJQqtO$Cro1l%f)jssm3%bM(KHu^P8`>BZmgfQPV zi-uzB7$y6eUk|GpP6*Lpc5*ei1QHxLvobbX9Afb>7b%rvkRh8Wv6avfs0{)?RZKel z1Imlg?pvApRitvbd^uE901rmW--o3#OPo$QSAS{1B#b9Tz~2xD}_$-Sl$9~18i@k=gU4iTgPpD4<{Wli#k z+ZprS3Tp%r1%Ap1`ysg&y$9ah&E8>Lb>&*HYA&TxlBtF05yFiqT5KzgcZnqAe#QSy zaXDtg%w?swk+lwX1 zp*uGUzG($>Q=%k#ABH6Ju5^)H9Xe4e2>pKon-660)2GHEr3?dJF|ZDJ$ztZhRn2@C z^v+Kno#jV}M2rWDy_j^f5#6-by7`uu70Zlgb#l9Oj}^tOBXu*)kbXq61YRtylR&eD4<&mz)v6zyYMr~d#@}=9 z+_@J!lV$SgpbB`A1GW02=O{X57-~ucb18;?5XHoJF)=JAQo#mQ&@kl!Drgo~!5zhW ztPE}+s*^eh5k5u-;cKna!4{fRtsd&2bx;Q}Po*0?!oV-$No#r14F@w&J7OCW62S8d zU@isESE76vln+melH~^H8m2%cAp>C>Ms}^g1gSlElU&@fLWdKDUlB z!(P#q8#~MR=;a%`j@C&LyoE@YS3yxi?UTV4&X~$QT*e_ET*i?^?gaC)uT!$CWUy&W z2Xk3*K8S+JWzOb)2$h^QP0OKSSw-Zq_9lLH6srZ_^}j}k}tUfaY`9?I<# z#~uz54~`JM889ABipw-`S^wX>DC9G#01M|Y%h@Bwz$B?`Mk;f;Z_rzhhCztvB%I?t zj9ISmXx(Ewlv5r`y8OIyayAxJ&N3x`tdN#CM>lmly>oF!e_mX^uFT_vum zV-L&SQKUUBX-~_Xi02abyh-aKjai$Qa7B@;mGCEpdvv0Kjl*e0k;gNYL@d?)f_k%$Pj zhi1DQ(K{uVfaqH&F}{$Vn_9>k)e;Rz6oX zoIp*VPOE-QycIf{>7|+8oWAFhE`1OckX`%1cs67||Ni!SK!{5N-$s>PFXs0uo^fe! zF(18>cFzGjs^WkOOEq@#1mhFYtH+Hp9MP>>>R` zLK)JCb7{nnUXtDtz~9E@EyEq`$Et*FJPQN%uJAP;8xf~tp})F-Ztgr{%g1azlEwOL zoF9kFAc-E+fI}0QxxpkBun)qyA7o7w1{zbJ7bdbX;okscz=0sBJ`gibZ~-C1U5Se` zego-GE+AUYyo}Q(6MyX;T5NF=s0OeIQpMRE5%6GORdw0ETYyhNRrTT_DgL{Kh#~~aft7Z%?stiIEb8`%r34sUYHpUvVN*?>7Uew@}Gyi z+m6qxu5tykw~OvtnyyU@Ss59gW{h0j9Zhs~VvMwjCdryHndqh`nVPKW%G6Ly*;2=* zY-_SRQCvkj0>9sHmedykEwR51=)G+l&#?o$~^MaUYU==TuI$7cAsIyI5AL2}~rj4NIg z#LQZm2?Prm4>Z$odbmKXL5cPeL$oYsodC(&lA zIy851^~F{zVAnzTz|@yFEUVxC>~X;G5ek#pf4Bg$=#f`8oQ+gv_Ja#w#uuy;cCO4{ zYnMGgYa25;Q$>m-wg@ zo&Xi$YJiXL7p@;PL&w#riH!}i5%WS{@uDCy)=EYmW9TvtwQ9M@kJ}|GLIjFZkq4`h z0e<_(^dqc5-l0Lgx^qxZ9Ib_;h8X1&MC*Qv#Gvl0Qq{f7pMT#agi6IbmQq6<%b(vj zgwXfYrED05CP(3WLSg+Ve9siM>JH`4?>mG-y#0bG6p?DeWE6Sm-D=2Yv_3gOu|X(W z+-M{UgUDDb8Pg~;*%-a5XUp=?8c2B8(a`8*QAP7}+2rZ9W&7 z6$eqW-ucKZ&P>*ZcExj&*|EDsN{B*HQu2V&qMv;;l~_@*Qief#htc|^1`XMdU4m$CA^+}xW~=yVInY^2`U@7Fi;M67q)G@8O~OH` zs_>uEP#@q!4}MYg%R2Q&QvlwC-YBM5x#*4YQdZa{Jb+ z5pEmDEv;nhD{kL+8nhAQrRu4aFWpIa~B&l;EdJ>MX^0BpIir42tyOHT;X9l$g#Na-t~ zv(+dU0cRe?LTrXl(z28~m!)(8>$p|w`c5XH$B)3omg#WQK5liMo}_zejgHX)TFcUV zLD*V;auLd)6jlbMDP_21Cmwb3TL5?R=k=54gm9WDKQsgLeFWGEKY0y+O?jx`4=w#u5+BiPpB{g%8FpcEyfC8W zW)R(_gm{GIhOta3@_Gm4G&9~spt!KbuNjA5x1LfGj2nhAJ)!I$jk9Ny)H9|hlhr+$ zY(L6-@m#V5`{}7}7455f)%C2_; zg;k*Nu5gmuu0itV*8$4N*{2Hw_x-HLb+@y*b%MHuGwPjj9-5Z~R`kOkx1p>XbJu44BJ)`_+^Nk|d!h0~_#MW6%GD7dO85XA=g2@_3sJ zmh%zRB{(F11Fnx7+Twc%_07o$$3&1b@Af`i#@>Ah=Y0sHkKP1g<-(15>>A62sW*6m z?nLa}iMJMyBhL%qA`&Ef7|s%ZA9Zh6=q}rA{vPTok80)5OO;pcRRNJr>P1Q8THDt|_hF6{|-!E7e9LBSRy2-loJ^;No3v|H7`8LGTX2U*ZPg1Z9UMUu84SdZ> zzm{*y!$oj+q_7-3=B9QOXzEP2hU&{o^`R(Wp2$SF@B(M!#T{z#L%x`v*M$_SkDeP9 z6bB4PL3I$3YpFh%X)7(S4{05~&1x};L8ryQYFG?n4=XaMLGx>cRm!C41p z@bo^`j)Eh^VJBVMsswg%C9q#!#z-MUYx20@Vg3jKT0=1KK4V*uc~byD3zpLeqx+)5 zPdb9D`#6D!uy{Gg3xWXamWomH-0}UY!E#MVRHY*)e z9Xwx4MD%X7$O|ZhprC$=a9yN+V!H@~ixyAuC9eqeAM~eRcfKzq@B?r|m0#h&SmWzT!A0tt71^tt721Y=hZVO?kbm*7vgG aH0=~rxbwn_ic$K Date: Tue, 3 Nov 2026 13:17:07 +0000 Subject: [PATCH 10/52] Land the HM Treasury policy costings family with its review applied The review approved the 597 rows with fixes. p.50's UK tax gap is HMRC's whole-economy statistic, not a part of Table 6.1 line 45 nor even the costing's base, so its two rows become context with a null key, as the script's own context path intended. The p.41 British Business Bank costing keeps its proposed key, but its notes now warn that it very probably sits inside a Table 6.1 line and must never be added to one. The line 45 and 47 component notes now say that their notes do not sum to the line. No printed value changes. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_01LwQq6HuQUQ1obM9JvyhJbE --- .../harvest-uk-ab2026-2026-10-29/COUNTS.json | 14 +- .../PROPOSED_KEYS.md | 5 +- .../uk_hmt_costings/NOTES.md | 404 ++++++++++++++++++ .../uk_hmt_costings/claims_staged.jsonl.gz | Bin 0 -> 63405 bytes .../uk_hmt_costings/manifest.jsonl | 1 + 5 files changed, 420 insertions(+), 4 deletions(-) create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_hmt_costings/NOTES.md create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_hmt_costings/claims_staged.jsonl.gz create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_hmt_costings/manifest.jsonl diff --git a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json index f2f5e2ab8..0ddd4dc4a 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json +++ b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json @@ -16,6 +16,14 @@ "welsh_government" ] }, + "uk_hmt_costings": { + "rows": 597, + "primaries": 1, + "claims_sha256": "d84725f64c8ba4d14060c90b01df55f92f10490f0e042e5a2ad5fcbcc835bfb3", + "sources": [ + "hm_treasury" + ] + }, "uk_hmt_redbook": { "rows": 432, "primaries": 5, @@ -76,9 +84,9 @@ ] }, "_totals": { - "rows": 1731, - "restated": 292, - "families": 9, + "rows": 2328, + "restated": 294, + "families": 10, "pinned_on": "2026-10-29" } } diff --git a/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md b/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md index 468a37d37..3c276e183 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md +++ b/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md @@ -4,10 +4,11 @@ Generated by `pipeline/ab2026_proposed_keys.py` from the staged families; don't A row whose scope matches no registered measure has `measure_key: null` and names one of these keys in `conditions.proposed_measure_key` (README rule 2). A key is a measure with no Table 6.1 line (`ab2026__`), a package (`ab2026__package_`) or an option (`ab2026_option__`). Each family's NOTES.md gives the key's scope as Table 6.1 lines or the producer's own words. Tranche 2 registers each key in `data/uk/ab2026_measures.json`, or folds it into a registered key, before the ingest. -46 proposed keys (9 measures, 29 packages, 8 options) on 201 rows. +49 proposed keys (12 measures, 29 packages, 8 options) on 223 rows. | proposed key | kind | rows | families | producers | |---|---|--:|---|---| +| `ab2026__british_business_bank_defence_commitments_accelerated` | measure | 9 | uk_hmt_costings | hm_treasury | | `ab2026__defence_investment_plan` | measure | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__energy_debt_circuit_breaker` | measure | 3 | uk_hmt_redbook | hm_treasury | | `ab2026__english_devolution_agreements` | measure | 1 | uk_hmt_redbook | hm_treasury | @@ -43,6 +44,8 @@ A row whose scope matches no registered measure has `measure_key: null` and name | `ab2026__package_tpa_savings_property_income_and_cash_isa` | package | 7 | uk_tpa | tax_policy_associates | | `ab2026__package_tpa_tariffs` | package | 2 | uk_tpa | tax_policy_associates | | `ab2026__package_tpa_temporary_vat_and_fuel_cuts` | package | 6 | uk_tpa | tax_policy_associates | +| `ab2026__private_landlord_register_fee_65` | measure | 7 | uk_hmt_costings | hm_treasury | +| `ab2026__pwlb_debt_relief_councils_severe_distress` | measure | 6 | uk_hmt_costings | hm_treasury | | `ab2026__regulatory_administrative_savings` | measure | 5 | uk_hmt_redbook | hm_treasury | | `ab2026__rolls_royce_gas_turbine_programme` | measure | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__state_pension_triple_lock_adjustment` | measure | 2 | uk_hmt_redbook | hm_treasury | diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_costings/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_costings/NOTES.md new file mode 100644 index 000000000..77ce1fab2 --- /dev/null +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_costings/NOTES.md @@ -0,0 +1,404 @@ +# uk_hmt_costings: HM Treasury, Budget 2026: Policy Costings (PU 3699) + +**Working files.** `stage.py`, the file meant wherever these notes say `stage.py`, lives on branch `uk/ab2026-harvest-staging` at `harvest-staging/ab2026/lanes/hmt-costings/uk_hmt_costings/stage.py`; this branch carries only NOTES.md, `manifest.jsonl` and `claims_staged.jsonl.gz`. The repository carries no primary bytes: fetch the costings PDF from its `manifest.jsonl` URL and check its sha256. + +This family dates from 2026-10-29 and follows README rules 1 to 9, including rule 8 ("stage estimates, not design") and rule 9 ("a component of one measure keeps that measure's key"). `stage.py` in this directory holds the rows as data and rewrites `claims_staged.jsonl.gz` byte for byte. + +Source slug `hm_treasury` and source_model `hmt_costing_obr_certified`, as in the AB2025 family. The exception is the two tax gap rows, which use `administrative_data`. + +- **Rows:** 597. + - 576 Exchequer impact cells: 96 costing notes, each across six fiscal years from 2026-27 to 2031-32. + - 21 rows of estimates printed in the notes' text: + - behavioural parameters (8); + - PSND and PSNFL impacts (7); + - eligible counts (2); + - the tax gap behind one tax base (2); + - one tax base; + - one financial-transaction impact. +- **Primaries:** 1 PDF. +- **Measure keys:** 76 registry keys on 573 rows. + - 416 rows cost a whole Table 6.1 line: `conditions.row_type: measure`, 68 keys. + - 157 rows cost a printed part of one line: `row_type: measure_component`, 8 lines' keys. + - 22 rows carry `measure_key: null` and one of 3 proposed keys, for measures with no Table 6.1 line or none this family can identify. + +## Access recipe + +- **Fetch:** `curl -sSL` with a browser User-Agent to https://assets.publishing.service.gov.uk/media/6ae1883fe4367e5c8b0467df/AB26_Policy_Costing.pdf on 2026-10-29 at 16:14 UTC. + - HTTP 200, `application/pdf`, 745,488 bytes, sha256 `f716efada8a4d0726ce12b0f97d3f0dac671c31b2d005ecf300a48ff309638bf`. + - The site showed no bot mitigation, so the fetch needed no fallback. + - The bytes equal the harvest's local copy downloaded on 28 October. The primary has not changed since Budget day, so README rule 7 does not apply. +- **Reading:** `pdftotext -layout`. + - 122 pages: the introduction on p.10, costing notes on pp.12-109 and Annex A on pp.110-121. + - Two notes run onto a second page: EMI2 (pp.31-32, table on p.32) and Writing-down Allowances (pp.70-71, table on p.70). + - This file cites each note by its first page. A row's locator cites the page that prints the figure. +- **Where the data in `stage.py` comes from:** a scratch parser, not vendored, split the text into notes at each "Measure description" heading and read each note's sections. + - Every note prints exactly one "Exchequer impact (£m)" table: one row of six signed values for 2026-27 to 2031-32. + - I reviewed the parse against the page text. All 96 titles match the document's contents list. + - Prose joins a line-end hyphen without a space, except a suspended one: p.82's "18- to 24-year-old". +- **Re-check:** + - `python3 stage.py` rewrites the gzip. + - `python3 stage.py --verify ` writes nothing. It checks that the gzip on disk equals a fresh build and that the PDF has the pinned sha256. For every row it then checks the page its locator cites: + - **table rows:** the heading, the six years, the six values in order and the row's value in its year's column. The row label must equal the page's label exactly, and any footnote must appear. + - **text rows:** the quote, verbatim, with `value_raw` inside it. `value` must equal `value_raw`'s number at the scale and sign the `normalization` names. + - **every keyed row:** the measure's title, on the note's first page. + - All 597 rows pass. An in-memory mutation test confirms the check catches each of: a wrong table value; swapped columns; a wrong or shortened label (p.82 without "[AME savings only]"); a text figure off by sign, scale or value; and a wrong title. +- **Table quotes.** A table row's `quote` joins the table's cells in reading order: heading | years | label and six values | footnote. p.96 prints its label broken across two lines as "Exchequ / er impact"; the quote and `table_row` write it "Exchequer impact". + +## What the family stages + +### Exchequer impact tables (576 rows) + +- **One row per note and fiscal year:** + - `metric: revenue_change` and `unit_concept: gbp`. + - `value` is `value_raw` (£m, with the sign and trailing "m" as printed) × 1e6. +- **Every table is `scoring_method: final`.** + - No AB2026 note prints separate static and post-behavioural tables; in AB2025, 8 notes did. + - No note prints a behavioural adjustment in pounds; AB2025's thresholds note printed £180m. + - So the family has no static, post-behavioural or behavioural-effect legs to stage. Several notes mention a static costing, and p.108 a "pre-behavioural costing", but none prints a figure for it. +- **`conditions.behavioural_adjustment`** classifies the words of each note's costing section. The pages listed are each note's first page. + - **"included (see note)" on 56 notes:** pp.12, 13, 14, 16, 18, 19, 23, 24, 25, 26, 27, 29, 30, 31, 36, 38, 39, 40, 43, 44, 45, 46, 47, 48, 49, 51, 52, 53, 60, 62, 65, 66, 67, 68, 70, 72, 73, 74, 75, 77, 78, 79, 80, 81, 82, 85, 86, 87, 90, 95, 100, 102, 103, 104, 107 and 108. + - p.90 adds its sentence: the Infected Blood costing keeps the scheme's cost model, whose "behavioural assumptions" the changes "do not materially alter". + - **"none: ''" on 18 notes:** pp.17, 35, 37, 42, 50, 54, 55, 56, 58, 59, 61, 63, 64, 91, 94, 101, 106 and 109. + - **"not stated (the note's costing section does not use the word 'behaviour' or 'behavioural')" on 22 notes:** pp.15, 20, 21, 22, 28, 33, 34, 41, 57, 69, 76, 83, 84, 88, 89, 92, 93, 96, 97, 98, 99 and 105. + - This class is lexical and reads the costing section only; AB2025 used only the first two classes. + - Six of these notes name behaviour in their "Areas of uncertainty" instead: + - pp.20, 21, 22, 69 and 89: "the impact of behavioural responses"; + - p.93: "the behavioural response to the rebate scheme". + - Others describe a response in other words: + - p.15 assumes 89% take-up; + - p.28 puts 25% of income into reserves; + - p.76 "includes an assumed timing effect"; + - p.84 assumes the zone facilitates borrowing; + - p.88 says self-assessment "may affect how taxpayers interact with the tax system"; + - p.99 adjusts "for expected utilisation of the cap". +- **`conditions.static_impact`:** three notes say the measure has no static impact, so the whole costing is behavioural. + - p.19 (Cash ISAs) and p.43 (E-invoicing): "The costing has no static impact on taxpayer liabilities." + - p.44 (Employment expenses): "The measure has no static impact on taxpayer liabilities." +- **`conditions.table_row`:** "Exchequer impact" on 95 notes; "Exchequer impact [AME savings only]" on p.82. +- **`conditions.table_footnote`:** + - p.94: "*Note: This costing also includes delivery costs." + - p.108: "Note: this is offset by increased DEL spending so is fiscally neutral." +- **`conditions.fiscally_neutral`:** pp.108 and 109, each with the note's own sentence. +- **`note`:** the note's base section, then its costing section, then "Areas of uncertainty", verbatim and joined by " || ". Component and off-table notes have a mapping paragraph in front (see "Measure keys"). + +### Estimates from the notes' text (21 rows) + +| page | figure, verbatim | staged as | period | key | +|---|---|---|---|---| +| 15 | "an estimated 166,000 fathers and partners and around 750 self-employed adopters" | `eligible_count`, persons (2 rows) | 2026, point in time | `ab2026__parental_allowance_fathers_partners_adopters` | +| 15 | "an 89% take-up rate" | `behavioural_parameter`, percent | 2029-30 | same | +| 28 | "25% of the additional income above the baseline is placed into MSAs’ reserves" | `behavioural_parameter`, percent | 2027-28 | `ab2026__business_rates_retention_mayoral_authorities` | +| 33 | "public sector net debt (PSND) (approx. £386m) and public sector net financial liabilities (PSNFL) (approx. £375m)" | `psnd_impact` -386m, `psnfl_impact` -375m | to 2031-32, cumulative | line 29, component p.33 | +| 34 | "a cost of approximately £775m" (PSND); "add an additional cost of £68m" (PSNFL fair value) | `psnd_impact` -775m, `psnfl_fair_value_impact` -68m | to 2031-32, cumulative | line 29, component p.34 | +| 41 | "a cost of approximately £32m" (PSND); "the value of the portfolio increases by approximately £86m (total PSNFL benefit of £54m)" | `psnd_impact` -32m, `psnfl_fair_value_impact` +86m, `psnfl_impact` +54m | to 2031-32, cumulative | proposed, p.41 | +| 50 | "6.4% of total theoretical tax liabilities, or £59.2 billion in absolute terms" | `tax_gap`, percent and gbp | 2024-25, outturn | context (`measure_key: null`) | +| 58 | "an impact of +£130m estimated in 2029-30 in cash terms" | `financial_transaction_cash_impact` | 2029-30 | line 47, component p.58 | +| 67 | "An elasticity of -0.95 is used, with 25% of the reduction in stakes then assumed to substitute to Remote Gaming Duty" | `behavioural_parameter`, elasticity and percent | 2027-28 | line 56, component p.67 | +| 84 | "£20 million of additional capital borrowing annually from 2028-29 to 2031-32" | `behavioural_parameter`, gbp, one row per year | 2028-29 to 2031-32 | `ab2026__leeds_city_fund` | +| 108 | "approximately 4.9 million eligible properties are in scope" | `tax_base`, properties | 2026, point in time | proposed, p.108 | + +- **Labels.** Each text row names its quantity in `conditions.statistic`, or `conditions.parameter` for a behavioural parameter. Rule 9 reserves `conditions.component` for the part of the measure. +- **Per-year rows.** p.84's sentence gives one figure for each of four named years, so the family stages one row per year, each quoting the whole sentence. The family sums and splits nothing. +- **Year of a parameter.** The note gives no year for p.15's take-up, p.28's reserves share or p.67's two parameters. As in the AB2025 family, `fy` is the measure's first year, and the row's `note` says so. +- **Present-tense counts** (p.15 "There are an estimated", p.108 "The current estimate") print no year. They carry `period: 2026` and `time_basis: point_in_time`, with `basis: outturn`. +- **Metric choices that could go either way:** + - p.15's population is the people a new benefit would cover, so it takes the registered `eligible_count`. + - p.108's 4.9 million is the base the fee falls on, so it takes `tax_base`, as AB2025's 165,000 HVCTS properties did. + - p.84's £20 million is the council's estimated response to the measure ("Additional business rates income is assumed to facilitate borrowing", and "none is assumed without the zone"). So it is a `behavioural_parameter` in pounds, as uk_obr_tables stages costing assumptions, not a financing remit. +- **PSND and PSNFL figures.** + - They sum over "the period to 2031-32", so they carry `value_kind: cumulative`, `period: 2032`, `conditions.horizon` and no `fy`. + - The document prints them unsigned, with words. The family gives all seven one convention, matching the Exchequer rows: positive = lower PSND or PSNFL (an improvement); negative = higher (a cost). + - `normalization` and `sign_convention` quote the words each sign comes from: "net cost", "a cost of" and "add an additional cost of" are negative; "benefit" and "the value of the portfolio increases by" are positive. + - The £68m (p.34) and the £86m (p.41) are fair-value movements: increments to PSNFL beyond PSND, not PSNFL totals. They take their own proposal, `psnfl_fair_value_impact`. + - p.41's figures agree: -32 + 86 = +54. + - p.34 prints no PSNFL total, so the family stages none. +- **The tax gap.** + - p.50 prints HMRC's tax gap for 2024 to 2025 in its tax base section, as the base's starting point: "The tax base for this costing is the total tax gap less the additional tax revenues accounted for by previously announced Closing the Tax Gap (CTG) initiatives." + - Rule 8 counts a tax base as an estimate, but the total tax gap is a context figure, not a part of the measure, so the two rows carry `measure_key: null` (AB2025 rule 3). + - They are an official HMRC statistic repeated by HM Treasury, so they carry `attribution: restated`, `benchmark_class: administrative_fact`, `source_model: administrative_data` and `basis: outturn`, with the uk_rf_ab2025 proposal spelling `tax_gap`. + - The note does not cite the HMRC publication. + +## Conventions (load-bearing) + +- **Sign.** The document prints no legend. + - `conditions.sign_convention` on every Exchequer row reads "positive = yield to the Exchequer (reduces borrowing); negative = cost to the Exchequer". The full text also gives the evidence: + - p.38 says its figures "represent net income from developer fees" and prints +50m to +95m from 2027-28, after -15m in 2026-27. + - The Bank Surcharge rise prints +, and the VAT cut on domestic electricity prints -. + - Table 6.1 prints the same values with the same signs. + - Values are NOT re-signed. + - Text rows carry their own `sign_convention`: as printed for p.58's "+£130m", and for the PSND and PSNFL rows as described above. +- **period** is the FY end year (2026-27 → 2027), with `conditions.fy` holding the label. +- **geography** is "UK" on every row, because these are Exchequer impacts. `conditions.measure_scope` names the territory where the note or its title says the measure applies: + - England: pp.20, 21, 22, 36, 69, 96, 102 and 108. + - Great Britain: pp.15 and 59. + - England and Northern Ireland: p.99. + - pp.91 and 94 name England only for their cost bases ("English borrowers", "English undergraduate student loan outlay"), so they carry no `measure_scope`. +- **baseline_policy** is null throughout: current law at scoring, including the pre-announced indexation in Annex A. There is no `proposed_baseline`. +- **impact_channel** comes from the Table 6.1 line's Tax or Spend head in the registry: + - `tax_measure` or `spending_measure`; + - `not_on_table_6_1` for pp.108-109; + - `not_identified_on_table_6_1` for p.41. +- **national_accounts_basis**: "All costings are presented on a National Accounts basis (p.10)" on every Exchequer row. +- **parse_confidence** is "high" on every row. The document prints every value as text, and `--verify` finds each on its page. + +## Measure keys + +The document orders its costing notes by Table 6.1 line. Mapping each note to its line gives a non-decreasing sequence from line 2 to line 98, with the two Table 6.1 exclusions (pp.108-109) at the end. That ordering, together with titles and Table 6.1's footnotes, settles each note's line. + +- **Whole lines (68 keys, 416 rows, `row_type: measure`).** These cover every note that is the only costing for its Table 6.1 line and costs all of it. + - For 64 notes the title equals the line's title exactly. + - Four differ only by Table 6.1's parenthetical or the year: p.35 (line 30), p.38 (line 37), p.42 (line 44) and p.77 (line 66). +- **Components (8 lines, 25 notes, 157 rows, `row_type: measure_component`).** Under README rule 9, a note that costs a printed part of one Table 6.1 line carries that line's key. + - `conditions.component` names the part: the note's title, or for p.82 its fiscal head. + - `conditions.table_61_line` gives the line. + - The harvest never sums a component with the line's total. + +| line | key | component notes (first page) | +|---|---|---| +| 29 | `ab2026__british_business_bank_backing_innovation` | p.33 British Business Bank: bring forward activity; p.34 British Business Bank: increase total financial capacity | +| 45 | `ab2026__hmrc_tax_gap_package` | p.43 E-invoicing; p.44 Employment expenses; p.45 More Timely Payment; p.46 VAT and PAYE direct debit; p.47 Real Estate exchange of information; p.48 Debt Management, recovery from bank accounts; p.49 Debt Management, debt collection agencies; p.50 Tax Compliance, 670 officers; p.51 MTD penalty easement; p.52 Image Rights delay | +| 47 | `ab2026__dwp_fraud_and_error_prevention` | p.54 Abroad Fraud; p.55 Pension Credit capital redeclaration; p.56 UC specialist fraud and error checks; p.57 UC capital verification reversed; p.58 UC New Claim Advances | +| 56 | `ab2026__machine_games_duty_higher_stake_25pc` | p.67 rate rise for higher stake machines; p.68 liability moved to the supplier | +| 72 | `ab2026__youth_guarantee_and_apprenticeships_spring_2026` | p.82: AME savings only ("Exchequer impact [AME savings only]"), for the Youth Guarantee expansion to 22- to 24-year-olds | +| 77 | `ab2026__uk_india_ceta_and_double_contributions_convention` | p.85 CETA tariff changes; p.86 Double Contributions Convention | +| 84 | `ab2026__waste_crime_landfill_tax_clearances` | p.92 six priority site clearances; p.93 local authority pilot | +| 85 | `ab2026__further_priorities_nhs_and_vawg` | p.94 Erasmus+ (by position; see below) | + +- **Proposed keys (3 keys, 22 rows, `row_type: measure`).** Rule 9 keeps proposed keys "for scopes that cross Table 6.1 lines, or have none". The registry adds them before ingest; the table gives each key's scope as the producer's own words. + +| proposed key | rows | page | scope | +|---|--:|---|---| +| `ab2026__british_business_bank_defence_commitments_accelerated` | 9 | 41 | No identifiable Table 6.1 line (see below). "British Business Bank: Accelerate £450m of commitments to support defence-related and dual-use businesses" By Table 6.1's footnote 2 and its £10m bar, this costing is very probably inside a Table 6.1 line, most plausibly line 41 (lines 29 and 44 are fully accounted for by their own notes); tranche 2 must never add it to Table 6.1 lines or totals. | +| `ab2026__private_landlord_register_fee_65` | 7 | 108 | Not on Table 6.1 (footnote 2). "Renters' Rights: Roll out a £65 fee for the 'Register your rental property' service for private landlords and private properties in England from 15 December 2026" | +| `ab2026__pwlb_debt_relief_councils_severe_distress` | 6 | 109 | Not on Table 6.1 (footnote 2). "Public Works Loan Board (PWLB) Debt Relief for Councils in Severe Financial Distress" | + +How I settled each line: + +- **Lines 45 and 47.** Table 6.1's footnote 4 on both lines reads "Full costings for the measures in this package are set out in the Budget 2026 Policy Costings document." + - The ten notes on pp.43-52 sit between line 44's note (p.42) and line 46's (p.53). + - The five notes on pp.54-58 sit between line 46's and line 48's (p.59). +- **Lines 29, 56, 77 and 84.** The line's title names both notes' measures. +- **Line 72 (p.82).** The note's title equals the line's, but its figure is one fiscal head of the line: + - its table prints the AME savings only; + - its text costs the Youth Guarantee's expansion to 22- to 24-year-olds and never mentions the apprenticeship support the line also names. + + Table 6.1 prints -290 and -375 where the note prints +25 and +35 (2027-28 and 2028-29). +- **Line 85 (p.94, Erasmus+).** No Table 6.1 title names Erasmus+. + - The note sits where line 85 falls: "Further Priorities: Fund key priorities, including pressures facing the NHS and delivery of the Violence against Women and Girls Strategy". + - Line 85's 2029-30 to 2031-32 figures (-5, 0, -15) equal the note's. + - Table 6.1's footnote 2 leaves out only two measures, neither of them Erasmus+, and the note's figures exceed the footnote's £10m bar in four of the six years. + - `conditions.document_position` records the position. +- **p.41.** Position narrows the note to lines 40-43, the national-security block. + - Lines 40, 42 and 43 carry DEL heads (TDEL, RDEL, TDEL), outside the document's tax-and-AME scope. Line 41 alone has a SUBTOTAL head. + - But line 41's title, "Defence Investment Plan: Reallocate capital spending to fully fund the Defence Investment Plan", does not name the British Business Bank, and no figure ties the two. + - So the row carries a proposed key and `document_position`, not line 41's key. By Table 6.1's footnote 2 and its £10m bar, this costing is very probably inside a Table 6.1 line, most plausibly line 41 (lines 29 and 44 are fully accounted for by their own notes); tranche 2 must never add it to Table 6.1 lines or totals. +- **pp.108-109.** Table 6.1's footnote 2 reads "Measures with no net increase to borrowing are not included, these are debt relief for councils in severe financial distress through the Public Works Loan Board and a landlord registration fee for the Private Rented Sector." + +No `row_type: measure` row carries a registry key whose title covers more or less than the row's figure. + +## Cross-check against Table 6.1 (uk_hmt_tables) + +- **All 68 whole-line keys** equal Table 6.1 in all six years. +- **Component lines** (Table 6.1 against the components' sum, 2026-27 to 2031-32; computed for this check only, never staged; reported, not fixed): + - **Line 29 (pp.33+34):** equal. + - **Line 56 (pp.67+68):** the components sum to 140 in 2031-32 against Table 6.1's 145. The other years are equal. + - **Line 77 (pp.85+86):** the components sum to -170 in 2026-27 against Table 6.1's -175. The other years are equal. + - **Line 84 (pp.92+93):** the components sum to -4, -3, -3 in 2027-28 to 2029-30 against Table 6.1's -5, -5, -5. + - These three gaps are each £5m or less, and every Table 6.1 cell is a multiple of £5m. That fits Table 6.1 rounding each line from unrounded components, but neither document says so. + - **Line 72 (p.82):** the AME component prints +0, +25, +35, +0, +0, +0 against Table 6.1's 0, -290, -375, 0, 0, 0. + - **Line 45 (pp.43-52):** the ten components sum to 110, 290, 240, 2,140, 2,635, 3,360 against Table 6.1's 10, 100, 115, 1,980, 2,485, 3,160. Table 6.1 is lower by 100, 190, 125, 160, 150 and 200. + - **Line 47 (pp.54-58):** the five components sum to -10, 10, 125, 65, 145, 235 against Table 6.1's 195, -125, 60, 75, 155, 240. The differences are +205, -135, -65, +10, +10 and +5. + - **Line 85 (p.94):** equal from 2029-30 to 2031-32. Table 6.1 prints larger costs from 2026-27 to 2028-29: -355, -175, -140 against the note's -15, +70, +45. +- **Table 6.2:** p.58's "+£130m estimated in 2029-30 in cash terms" equals Table 6.2 line 9 (DWP Fraud and Error) for 2029-30. +- **Lines without a costing note:** 22 Table 6.1 lines have none: 1, 4, 9, 20, 22, 23, 24, 25, 31, 32, 34, 35, 40, 41, 42, 43, 63, 64, 67, 73, 74 and 81. + - The introduction limits the document to "tax and annually managed expenditure (AME) policy decisions". + - Table 6.1 gives these lines an RDEL, CDEL or TDEL head, a SUBTOTAL head, or none. + +## Things in the primary that look wrong or contradictory (reported, not fixed) + +- **The £10m threshold.** The introduction says "Measures that were certified by the OBR as below the negligible threshold of £10 million per annum in every year of the forecast period are not considered in this document." Five notes print costings below £10m in every year: + - p.59 (+0m, then +5m a year); + - p.92 (-2m, -2m, -3m); + - p.93 (-2m, -1m); + - p.33 and p.109 (+0m in all six years; p.33's PSND and PSNFL figures are not zero). + + Table 6.1's footnote 2 sets the same bar for its lines: "AME and tax costings with an impact of less than +-£10 million in every year are not included". Yet two such lines appear: + - line 48 (0, then +5 a year; Spend, no fiscal head); + - line 84 (0, -5, -5, -5, 0, 0; Spend SUBTOTAL). + + Lines 9 and 81 are also below £10m every year, but their RDEL heads fall outside the footnote's AME-and-tax bar. +- **The lines 45 and 47 differences.** Table 6.1's footnote 4 points to this document for the packages' full costings, but the components do not sum to the lines (above), and no note explains the gap. + - For line 45, the gap has the sign the HMRC spending the notes describe would give: £70 million on debt collection (p.49) and £485 million on compliance staff (p.50), "over the next five years". The document does not say that this spending is the difference. + - For line 47, the gap changes sign: +205 in 2026-27, -135 in 2027-28. +- **Notes with no Table 6.1 title.** BBB defence (p.41) and Erasmus+ (p.94) have costing notes, but no Table 6.1 title names them. +- **Minor title and wording differences:** + - p.77's title says "from 25 June to 1 September", while Table 6.1 says "from 25 June 2026 to 1 September 2026". + - p.16's title says "Weekly Earning Limit", while its text says "Weekly Earnings Limit" (Table 6.1 follows the title). + +## Not staged, and why + +The family stages every Exchequer table and every printed estimate in the notes' text: tax bases, behavioural parameters, eligible counts, PSND, PSNFL and financial-transaction impacts. The appendix lists everything else page by page. + +- **Design (README rule 8):** a measure's own rates, thresholds, payment amounts, dates and durations, and counts of products, sites or bodies. They stay in `reform_hint` and `note`. +- **Contingent liabilities (rule 8):** + - p.42's "£10 billion in new guarantees are issued each year between 2027-28 and 2030-31" (UK Export Finance); + - p.83's guaranteed lending "by £1.0 billion in 2026-27, £1.5 billion in 2027-28 and £2.0 billion in each of 2028-29 and 2029-30" and its £6.5 billion total (Growth Guarantee Scheme). + + Both are the volumes on which the notes cost fees and losses. Each note's `note` keeps the sentence. +- **Spending allocations to programmes and financing remits (rule 8):** + - p.34's split of the British Business Bank recapitalisation (£1,405 million for commercial equity, £165 million of CDFI lending and £450 million of venture debt, within c. £2 billion), kept in `note`; + - p.33's £1 billion reprofiled; p.35's £2bn Energy Wealth Fund; p.41's £450 million; p.42's £50 billion of capacity; + - p.109's £700m of PWLB debt repaid; + - HMRC's £70 million (p.49) and £485 million (p.50) and the staff numbers on pp.50 and 56. +- **Numbers printed only in titles:** AB2025 kept these in `reform_hint`, and Table 6.1 carries the same titles. +- **Baseline parameters:** p.91's RPI outturn and counterfactual maximum rate. +- **p.58's "-£35m in 2029-30 in expenditure terms":** the same figure as that note's Exchequer table cell for 2029-30, so staging it would count it twice. +- **p.94's "£0.6bn was allocated for the cost of Erasmus+ association for 2027-28":** an allocation made at Spring Forecast 2026, an earlier event, with a different scope from this note's student-loan and delivery costs. +- **Annex A (pp.110-121):** indexation assumptions in the forecast baseline, not measure costings. AB2025 did not stage them either. +- **The introduction (p.10):** the £10 million negligible threshold, and the list of five measures whose methodology "the government will publish further information on ... by the end of the year": + - More Timely Payment, Distributions, Foreign Sovereign Investments, Founders' Relief and BADR. + - Those publications will be new primaries when they appear. +- **References, not quantities:** scheme, form and statute numbers such as EMI2, Plan 2 and Plan 3, form P87, Class 1 and 1A, Group 7 of Schedule 9 to the VAT Act 1994, FRS 102 and the Five-Year Strategic Plan. +- **Uncertainty ratings:** the document prints none. Each row's `note` keeps the note's "Areas of uncertainty" text. p.101 says "there is no uncertainty over the figures involved". The OBR's ratings belong to uk_obr_tables. +- **Charts:** the document has none. +- **Unreached pages:** none. + +### Appendix: every unstaged number, by page + +D marks design, C a contingent liability, S a spending allocation, input or measure size, T a number printed only in a title, B a baseline parameter. + +- **p.10:** £10 million negligible threshold; BADR's £500,000 in the further-publications list (T). +- **p.12:** VAT 5% → 0% (D). +- **p.13:** 5p (5 pence) per litre cut extended by 6 months; 57.95 ppl main rate; 5ppl rise (D). +- **p.14:** 55p for the first 10,000 business miles; 35p above 10,000 miles (D). +- **p.15:** + - two weeks paid between £27 and the statutory rate of £194.32; + - 2 weeks at the lower of £194.32 or 90% of average weekly earnings; + - up to 39 weeks for adopters (D). +- **p.16:** Weekly Earnings Limit at 16 times the National Living Wage; taper set at 100% (D). +- **p.17:** £2,500 surplus earnings threshold for one year (D). +- **p.18:** two packages of suspensions (D); 135 and 101 food products (T). +- **p.19:** £20,000 and £12,000 cash ISA limits; individuals under 65 (D). +- **p.20:** relief up 20ppt from 15% to 35%; rateable value of £1 million (D). +- **p.21:** £12,000 → £18,000 for 100% relief; £15,000 → £21,000 for tapered relief (D). +- **p.22:** two-year real-terms bills freeze; one-year Supporting Small Business extension; 1p Transitional Relief Supplement for two years (D). +- **p.23:** £1,000 → £3,000 trading allowance (D). +- **p.24:** 72 products; 269 extended and 23 expiring suspensions; all 92; 361 products (D/T). +- **p.26:** 3.7ppl cut from 10.18 to 6.48 ppl, extended a further 6 months; the 5ppl cut (D). +- **p.27:** ABV bands 1.3%, 3.4%, 3.5% and 8.4%; RPI plus an additional 10% (D). +- **p.28:** 20% share of growth (D). +- **p.29:** 10% CGT rate; £10 million lifetime limit; at least 25% holding; two of three tests (D). +- **p.30:** no claim in the past 3 years (D). +- **p.31:** + - option limit £60,000 → £120,000; + - £250,000 EMI limit; + - exercise and exception periods 10 → 15 years; + - exercise window 90 → 180 days (D). +- **p.33:** £1 billion of commitments reprofiled into 2026-27 (S). +- **p.34:** c. £2 billion of financial capacity; £1,405 million, £165 million and £450 million allocations (S). +- **p.35:** £2bn for the Energy Wealth Fund (S); commitments over the next five years (D). +- **p.38:** around 70,000 first-time buyers (T); 20% equity loans; deposits as low as 2.5% (D). +- **p.39:** 25% bonus; £450,000 home value (D). +- **p.40:** withdrawal charge 25% → 20%; £450,000 price cap (D). +- **p.41:** £450 million of commitments accelerated (S). +- **p.42:** £50 billion of added capacity (S); £10 billion of new guarantees each year, 2027-28 to 2030-31 (C). +- **p.43:** two revenue channels (D). +- **p.44:** £2,500 Self Assessment expenses threshold (D). +- **p.45:** forecast liability above £1,000; less than 80% collected at source (D). +- **p.48:** debts up to £5,000 (individuals) and £10,000 (companies) (D). +- **p.49:** £70 million over the next five years (S); aged debt over 12 months old (D). +- **p.50:** £485 million over the next five years; 670 FTE (also in the title); 5,500 previously announced staff; 6,170 FTE total (S). +- **p.51:** 12 month penalty easement (D). +- **p.52:** delay by one year (D). +- **p.53:** £90,000 platform threshold (D). +- **p.56:** 2,000 specialist fraud staff (S). +- **p.58:** -£35m in 2029-30 in expenditure terms (the table cell). +- **p.64:** members under age 75; benefits paid more than two years after; 45% flat rate (D). +- **p.66:** surcharge 3% → 6%; combined rate 28% → 31%; 15 months (D). +- **p.67:** rate 20% → 25%; stake above £1; prize above £100 (D). +- **p.70:** writing-down allowance 14% → 12%; first-year allowance 40% → 50% (D). +- **p.72:** £50 million turnover threshold (D). +- **p.73:** £500,000 lifetime limit; 24% and 18% CGT rates (D). +- **p.74 and p.75:** 2 percentage points (D). +- **p.76:** VED of £1 for 12 months; vehicles over 3,500 kg (D). +- **p.77:** VAT 20% → 5% (D). +- **p.79:** levy 45% → 55% (D). +- **p.82:** + - 18- to 24-year-olds on Universal Credit for 18 months; + - a six-month job paid at 100% of costs for 25 hours a week; + - eligibility widened from 18- to 21-year-olds; cost base of 22- to 24-year-olds (D). +- **p.83:** + - £6.5 billion of lending; £1.0 billion, £1.5 billion and £2.0 billion a year (C); + - maximum loan term six → ten years; + - turnover £45 million → £54 million (D). +- **p.84:** 100% retention of business rates growth for 25 years (D). +- **p.86:** detached-worker period 52 weeks → 60 months; two countries; contributions in only one country at a time (D). +- **p.87:** 50% out-of-quota tariff (D). +- **p.89:** 100% relief for ten years (D). +- **p.91:** + - 6% interest cap; Plan 2 borrowers starting between academic years 2012/13 and 2022/23 (D); + - RPI outturn of 4.1% for the 12 months to March 2026, and a 7.1% maximum rate without the measure (B). +- **p.92:** six priority sites (T). +- **p.93:** three-year pilot (T). +- **p.94:** £0.6bn allocated at Spring Forecast 2026 (see above). +- **p.95:** rates 1 percentage point above 2029-30; maximum of 40% (D). +- **p.97:** VAT 20% → 0% (D). +- **p.99:** credit cap 5.3% → 8.3% (D). +- **p.100:** providers with 300 or more students (D). +- **p.102:** properties valued at £2 million and above (D). +- **p.104:** £135 relief; £39 private consignments (D). +- **p.106:** spread over five years (D). +- **p.108:** £65 annual fee (D). +- **p.109:** + - £700m of PWLB debt repaid, £500m for Woking and £200m for Thurrock (S); + - two local authorities; a transfer between two central government bodies (D). + +## Proposals used + +- `proposed_metric`: + - Reused from earlier families: `behavioural_parameter` (8 rows), `tax_gap` (2, the uk_rf_ab2025 spelling), `tax_base` (1) and `financial_transaction_cash_impact` (1, the uk_hmt_tables spelling). + - New: + - `psnd_impact` (3): a measure's cumulative impact on public sector net debt; + - `psnfl_impact` (2): its impact on net financial liabilities; + - `psnfl_fair_value_impact` (2): the fair-value part of the PSNFL impact beyond PSND. + + They parallel the existing `psnd_share_of_gdp` and `psnfl_share_of_gdp`. +- `proposed_unit`: `elasticity` (1). +- Registered values used beyond `revenue_change` and `gbp`: + - `eligible_count` with `persons`; + - `percent`; + - `properties`, a unit registered since the AB2025 harvest proposed it. +- `proposed_baseline`: none. + +## Attribution + +- **`own` and `different_model`:** 595 rows. They are HMT and HMRC costings that the OBR certified ("All of these costings were submitted to the independent Office for Budget Responsibility (OBR) for their certification", p.10). +- **`restated` and `administrative_fact`:** the two tax gap rows. + +## Distinct condition values + +- `row_type`: measure | measure_component +- `component` (25 values): each component note's title, and p.82's fiscal head +- `basis`: forecast | outturn (the tax gap and the two present-tense counts) +- `scoring_method`: final (every Exchequer and parameter row) +- `table_row`: Exchequer impact | Exchequer impact [AME savings only] +- `impact_channel`: tax_measure | spending_measure | not_on_table_6_1 | not_identified_on_table_6_1 +- `measure_scope`: England | Great Britain | England and Northern Ireland +- `behavioural_adjustment`: + - included (see note), plus p.90's variant; + - not stated (the note's costing section does not use the word 'behaviour' or 'behavioural'); + - 13 distinct "none: ''". +- `horizon`: the period to 2031-32 (cumulative) +- Also used: `table_61_line`, `proposed_measure_key`, `document_position`, `fiscally_neutral`, `static_impact`, `table_footnote`, `statistic` (12 labels) and `parameter` (5 labels). + +## Review fixes, 3 November 2026 + +The costings review (`harvest-staging/ab2026/reviews/verify-costings.md` on the staging branch) approved this family with fixes. They sit in an explicit block before `build()` in `stage.py`. No primary was re-opened (the session's network policy refused gov.uk); no printed value changed. + +- **Finding 1:** p.50's UK tax gap rows (6.4% and £59.2bn) are context: `measure_key` and `reform_hint` null, no `row_type`, `component` or `table_61_line`. They stay `restated`, `administrative_fact`, `administrative_data`, fy 2024-25. Registry keys now cover 573 rows, and components 157. +- **Finding 2:** p.41 keeps its proposed key; the double-counting caveat is in its `mapping_note`, the p.41 bullet and the proposed-key table. +- **Finding 3:** the line 45 and 47 component mapping notes now say their notes do not sum to the line, with Table 6.1's differences. +- **Finding 4:** no action. The coordinator found that the lane's copies of `pipeline/ab2026_harvest_contract.py` and the harvest README match #231's head byte for byte, so only this family directory moved. +- **Finding 5:** no change (p.84's £20m and p.58's +£130m are accepted as staged). +- **Verification:** the contract checker reports 597 rows, 0 problems, and a re-run of `stage.py` reproduces the gzip byte for byte. COUNTS.json pins its sha256. + diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_costings/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_costings/claims_staged.jsonl.gz new file mode 100644 index 0000000000000000000000000000000000000000..6126493f739f8894497a7bd1e75041d6d8fb018b GIT binary patch literal 63405 zcmZUbRY08EvaJ&!gy8P3!7aGEySqbhcXxMpcXxMdBoN#kfMF>Rr4Y zBK^9#+-j3Re%SZ|92{et&axd?8)Jhz%&fuj^ydR|&N2M-WcFf~2;}eiN!~rP9L?+E z;g2)wAh|^V+XcuZ&9+~@1LNq%kbUgYR)oKIZO^C;g&zQ%w%Zr9{uI54CMb!=aaw=1 zHP4E&hvOK}{T0x3d6P|1f5?%@0L;mV90}*casjhl#&WkIcuOhN^F4{~@-5cgzuzlg zlggs-ePvukf%I7)9PY;P)M1mE1s2W8JPFYsH_2*YcqIM0z}acaXgT<-IpalBI1=>m&>j{95LS?dyk^#LCte~GvJgkl-)Hw9oSbd6N|@om1Vv^ z{1xdDLl$@GVI}8;yS0?lqvG1wW0rftM|j7YB3*{3UGq4{o<7Q$dqkHB`iZlPx290< zVNl{@A|X#J3KyKwBp~|md24F&E}$ndNS5g;%kVx&SeNhIz0DVMC#CJz++|ico71`Z zY&e;jjNnJ+U75$DwAzg;3J+<%vtp*=Ee;W>tU4Ue88~=s;T?5^iBZVjCtokZ|1eybkMCJoVSsPfj%F zueTjMtRIaU=wr3iVz|WZ2JBIC3%@NwV6RqSTaU`s$t^00UnGP*GM;sXBW%GOvzpxfm<5`%Nd`cv2=k4KJpNo*~2 zYUUo$4lJOP#pZM7<&e1?;CS`L&9b0N|9Z!(4^`J@j1*=qU5Cn`xzh*amEi$W9nG7K za?6&^^^JZVaEJ58vUkJ9Q!!fe7VDr=8$4sd-TPW({U$k+-7o#h0X?^j6D-iX1K8hd zfXl*r@~+pXnaim%I8Hkq!VK!OA8oKYDA@QiLONXE!i6Vi;v}2Zj&EQf)i_!Mn}Eyf z*&GVjF{la8RX1%i%f$>yb3{&D<{fFYr1>g-2+=z}!@8xQD!}smbcX*hH3Dbd- z2JV*k)95aJ$`%d=72H|)HNjgxTz#Z-56(vVp5eOLig&$e>^HEXnbAeWW`uaQUyRo2 z7-O9acf0rnh&2Z;GmVMuo+flEnU|c!ZA8~hBBwluEKuHM>-~T?45eLsLOJ7 z>gfCrO73be>>zg9s4wW*U$mwY5B#;V@yX$^jTXbjY%0C;^UoXHci`50JP&g2>+WIiJb~{STg5ke+z;@5 z8hb!2H=ml%f0W3J_Q^lt5A0_?tgaw9G(j)O>3Uh<=xpm*6fKFT@>fWZ-ZS%kp`MRR z9iq4;XMYwZ#*4Z%#@x&*B6#Dxee8K!f?2ZE)yanIx^9e=s@>>3$;r{vzWwOy{R;NF zvR-Fo>&A4BFrWTL-Mnf4kjC*A|Dl$p<`ReK7;kB6b#u_`V$-XyL z97Deb6~vCo*UOXrlv{y_aB90QFs&@z9a;%TvZtgg*Y%a%QTJZJZ}|mTKp|l7Fsq5K z^j2?ZK<>@Z$C9)6S3+H6i#XqYX34cH1PKL}Ku*eRi3lNO@GEmxuB zjj}j9HSwU$e7wpLf>*5`r#kp456R2c+e2?QoGXugqOzL${fzp_GWC4%LL2jR^gMaJ z`Y`3MuJWnbI;!u9D3oN4ji+t+;wDd1B zGV7;2cN`@JZqg1vI?|4x9X~rEE+RbCN6+^k%}Rs72l@^A1<3cIJ_ZG{^!wA{54`^` za|;fi@f#567ohtCIQRS0VE=DG^WPJ|7a53gaS?H|c%b2h7n~j`zvuCs=RM$EDh0F$ zLh(~`&CFklt8sZFNO3aRG`b}DITC@tkTNxly%ZTy^9;W02>-lPzZM$M+Yh$^i=;TS zIKt%nG8%r(3hg^+oFB}Ftj^uYigH)TMrox~OIerbc(Vs}*vH-Z5pnpn`@T}GT>ZuJ=5)lTn{kMyWXfDF)*xhb z#I-^%5sKWj8h(pCh+kNC%l*)x9KwzRddU`%gqK5Zv)lVxdACM7LG4acXpJD+`Bj4v z)}Wb>8hs*v$n1eg((A9DWh9@eLH;=~^gA07*tNfDEUx9utH6tT9^EkgiI(sh+yO*7opK8>GcI5n3kB81y zQB7$LpOcQByAIxo$B9SEYCb6VN42_3t-o6kbeGh0kpHx)IFy~b3N5t~6rYazY_|*5 z?m-H;oeqRMm0U=hQeHHz&1sI`1HV9-rOk(+gklP5x6*6Mn=MBteDcA1m|Onpp>WAl zM*XqjY9t%eJ3X}}t0ex2*pX?HBKUURKW=O+7CSV&hK3}?Xu(gTT84&Yok+G5?8AzM z+w0tQt12WL;(M@0L!WWF>yGd1aAHy(Ar&@4S##IF zrPi)Etw9|DcI9Eev+&GQ^nB$T%7WZTaUquH)J{b8@Sd_3-E?Vq+`CIyT+Ne;Mg z_Sbd1!x*|WB`>xG` z@8DI8c5qIunXKQKNT}d$G>Uw~#q|hkG)D;P?fo|p3*+glyVS0AM8obzB^wrb3)iV$ ztQ6dsSaPUi=DS7Ql(ZW7jXm^j?u$J?kQYeVG8Y@;=s!BRh6@*yDPEWpk=M2D+4+b! z$yaonqnfocX}5k2QiqPnv$KUM#e#=3!5Z{UUdy_3eD!Ecp!se!8+w5%BWuGmO{1n2 z%*I!oi-2rJ22G~>L&-0MrJ+`3(epqTz=5q`49|P0de+v^FpsT%pGBk1h0Yte&1I8) zGM(~3X}8WfmwR|_BA!*9?Aq`%w?Egc$u#Q04bgou4Jf6T$%g`b(Z09aPL2;o1sOA3C6m_*NYX#o5bmjC8Ym?e)pW#C4g{G zxgRPRj%4YnV_n@;>5he_s=;J`=KOc88JP)af!Eep*OkL0={0hCZnfkOGg8MzXY&6K zq;>iYUXFF01f2WqpC{reBq?;9oibQkB#w;+g|J}0#T4V<0a>u^T5pn1JvyX?R7E8diC7*Ab=V%0 zQ7sFVGTe2K+8HlGjhsJ(%XuMxaG;Q~a{j2mMCH_x?tnB(%u-YHR?vBdTRNBgdURfp z!<2u{{UFb(UlI5;-|ukhoPmLH@{rmA#gYScT6(G7B8PEAtmPuP0bV7CS*K8)K`-7# z#4sBvmjV^o>Y~DBtFuB#ik|D2V??&Uh5D!omTawKly9y9Q|IvGxgcD?9il^*He+c_ zi&s)sUCR#nQmKI0kb2L@c3*9J2}ptW;HGgFN)SQTkh04seY;b?wR%S+_BKtkCK9>1 zv)y$8HMtD{sqG7U&9*dV=fwz}`Zzeant|q;cS%=IZ^Em?mqF=vqm|G(Zg^Ei#`X`H zVBK5`dUp)2LiSU#Bj84Bp)?_IFF5oF2CNE!N5Sq*PYynZu%;T^jn=#o*5so>JZyRi zKYdH9Sycy@5lk0eMHbedJ`QW!kS=TZ_A}fCGpKPdAY0Cgi5+zLxhu9gDje_h7;Xj5 zVknlYZv~Ve#5p;?^o2n%m!F=>Tn)*Oh-{_+ws#N5!(6>8x`*S z9a~#$Oa)1)!>DK`FB@Z|DG8Axf(qB+>_VHFe!W+`iE2$_L%Y@V^={rnb$Hu4Uh#*# zSWK?^RF^9Nl-OsyqMot|Ms3MkVoAzSi!B+ZN}42$iLU1zsElGcmHCWD>G@q49r~>v z%1<^tl-U`hSq+e+kV3vwSz6OiVeX9QIyW1!vz_=YTXGRgyAf)>S7z*@R0M;S(%^EL z5zcQw^T95!0Q)$<+Zp?vLA%@#X@89N?%nYCyD_#xO548sdUtPWwhj1gp&2}^^Q%tx zdqj&ev#m{RVH=ChLg2PPiQ~nzt*)fp_Sq9q@8EUR4-F5e>Tz?_jlhM}>N+~9qdWeJ zHo)|@cW>eDFSa)s7AGxFi>3f&$>PF)9-@R&Z2ee-aS9ci$`MhZWSvg&Z?_<*^9-+W-gy)s zG_6=swA{lpTgMD_^3>wMo9+`e%d&cCouE2V6{0E>bx`t4U^2NuaGir&o#I0bNf6q= zIQlL2-$S$kXakcs{bCUijZ0Xi-2DmM;NZLK7cZ}BIwlA#Lm^oEhtDBL{a_zZ90L4e z<2PCi+s4Z&Pc>ky`oP?7rdwTGOHadHp>-;oB3{5uuyrN3ekBd)O&Ijb3il1G8`2x> z`V<5!+wxRNEdfla163x+Wg2+KuN*2LuwQ0+({(TU|udTpZEUyEMO0hiJHkLG!Y+avKSOq5A6+{@hK%nGpM+NqPT(vVWd zsOPA(-GS+>uDAumyRA0Cy37rNc_AV(=! zj+I+7lfnUBbuAsFVb$r8h+nkLS_i+>lfn+{NYhp?M=L}Z9d-&xM#Ro?H6nn{(?Q%giL!~XJ-php zplqkLXKM)6Mh-66+kTxNAV6ib>O(}RmHfK%NkQf$+q!7*xibs<-AW-$A`$?~7=j55 z%Rh==l#nFuT(p@DR9I#6t2*1Wfvx0&pvt=c1DSLE2O{J82a=llKM)?@f1nj!uoeGz zD|-_s{|BlT{0&9(tGb&H=zudxZ(1zyOi7rY))0v^k)+2@ikp%$Ct!_58MojBV@a5j zHaaJ0EI4Dj#!S|zk~UUyVua`F*Twq4A96Q@cK$*{Z4Pl*x+Xu-rzr8Nhf6PLkR~(A z&)qUMr}lQ@$%i>LlZ#Owj?WUCqGrE5QOHWY<-B*pF0v2*DT+C-fccB7`u5m-^eIW7 zH5#Qxg2=7hBnJl;AnpBv=WXL}l9N`2!t8k4&2_tTy5(M zh!exuelUJCL1=3+5KWweisD)r^)sU8S9jY-1iF)12tqd@d#F*DjtDL@*_Wa6$>ti3 zY%S06h`m!47}bdow}hBZRwEi zZUnL$I?9gI&mFOuhcL5GYsCegGdMhh2Sc$HGj9)KmG75M)R(jX-`8)b+=L`Kh>47COpnlr2OZO`dh#c9!b%2bMkHf`eucj>A|s0<{8 zeZp%JPl>aS_IbyKX@Dd|2lI2d!=|+|tiJ22BaW&q&(YIiA@wK94!etPF{a^K)j&$# zY!}WE4N|e4C=(pdebI7Eaf}oRD^0PP%X(&5lC@!^ay`vw`uJyO6E&b`N^yIVM?p@ZU#>> zj+x;0TyjRS_FH2M3H~awi=pHD>XPd_RAfadcSDJR$h_#Qi|k>wLZ9Cvs5s4|Mj)xk(%k z1c~956`qVMWT7a2oY0Er6vxXghA$N5LZ^rzo;V07B#vZ+Q*B!{L6l_tE~n%gLI>Hi z7Ho@5W)|^z%Ow3rp*m!w$2`yoV+m)FCt6=0Pi+zNwU+K3IQ3A6r1I6cvw;MsB;$JA za_$#R2p>-W4BnovnVLnd1`)ABlhRJqqKJpi63w{BxxQbt;G7gAVl%)(G(B&~Mh4xl zU){UjF*q_9FdW$Ob$fnV;k%{ys%?yp^Q+ABZjy)`k{W6wXEYn>x$kmkAOCFTC%U^t zMX`D$-5rlJjH}dgjj@8GnJK)y?J-^P`zUHi0+Q%=2V3_q^KiSk>K-cqj^i)atiG=^ zoG(XZ=GB?nx6L$*Eat^2XvV=Rnpd&R#pnmA#cc_mab{C;hT}ES6)TSn6aDrmp!eYc zUs`myO%EF@V5F+6l~5;X0K)y!f7PWF@DLG@9^kGA;kZy7ck($cR_#Sg_rz2cxxl96 zuQnT@h3CJ`f4iyNQIgtN*R+T6Lua0%2@pDv{;MwAVaKwp@~l%epOTdE2dm+vYb?hp zntj245g+vbAq0!3E-mcVG|)-%9ZLRc)6c<1fkTPoGSZ=@16G_i`T8JY{Gp*|T!YGT zY*~CV$v&CgO}VWH6Mjgodv{nfn7TQ*&}Ojg68B(<$6>oORGpJ!?v5a6mVq76cBj$1UO6$VG58Cbh&y|1( zt*3Ukm+Itut+wLzY4)#5l?X(-R`LKM`-b*K0j8-^dP{DgNnep1{oq$@QOz$TCq>$M7|790H|!wh^6oOFmzA}Rp#Es{)hFGwuwEWyRGmSxknXMlu5{iD!oSKU#w{)a_{=_b_+0-kfz!OrL84Dip865&hJ(o>cZ0p z7Hjft24WLdD@DAlhI_}~eS?{~`)0ZZ?R(uzX80!3!oD@O>qUO588WA(-MqLnuIXEb zg2uMQnJtUrtY>!KoA!CF%JC)+8+H`?y-~B8W{-6uorT&^kdr_Mp-vo~I7U(A$g1cqRCKL7*V1iA=yl|wOPK*jDh zu;28Uh50byJ+9yuiN~?-6~N%c;=P`BBi}K%uRvdQ5`iSllLeXtt0`60`O%3_3MR7 z9{0Oeg4`7E#*o7fCqg50V8CriuVZgF*lRZq`_-YdR6y@$p)OvcDGkx61x zq{c9fP6ewdP_b$@xvFW;Y`7T6P}ywQZK{DZM1!tG@&_5BF+^vy{0Lq1n}3SZ(4zh+ zI_dGwl;>VXpaG3~EWR3`YE0dLwi030jI!_pR8;bqR-UW0l4#Yhf2cf0st;2arXfa0 ziVSluL%RbNRoneH(;u_d^zrbXLbYbz)Ra0EYr)i32ZRL$8>^$K!DkIpET8wN=v@E<0a0bt?njajZ z93aeb#mbE?2e&qo4|Q(z6^joF0+0LBbx(;Ot)03~c4+ttIGMg#IyAT;T$ZL1j-Ol3 zHL7z!2P1E^Hpd_My+VNV7^k;l>DVmh28KwGXUY02vrDJ?LtNu927v}~JV4x-lnEIt zJW5!U*yvj%^B+5Mp6sx1rE*tS-)<<9K)w?1OfVK>EY?7>5`QM{VmvJI$EnXfCS_Hh zEF(>rvA8AOhii`rus%CcSx;8Jk6#2a$FKRw<)u3kJV|rvbK^iyc>hp_6p4ab35Z zk}a~<@#`UhndTPu6g%QljY=I5Mi!gP&g7Nd@zBLD%gYZ!a>PzMZV<)RZ!Gy|U_7Pl zU>rCej3MA6OKBm+PO@g*u{4|xKa0iLj46XN$mP48CTVh?vmFEG5$mp{uPJ9~D^6$b zvppvrZHvD@M~lf>bwy=2%1(r&A*gDMoFoW|zf4cpPd9(b#k3)Kdy{(6C- z)<+!YYrK4^>hM+-?I|O!ZEQpkQJNMHb0!ro9#WOM?fiBX{FhwTLwrm|B1=n}LD8l}U4gwOyF1+R1{O$ zm3JVL_aD@O5H!O757`j^4+6bs6A&qV8gT%tWz97Cc!HQ<0AMi2WP-&CffWi_2|Nh~ z71AoOO3t)>cqW+AM-fl%I&lVz>X@;hbE3hxfAC5Cq&X5$L1maW9yxbTHK<)P&X8)H z#-eKRJO1y;#*a4}pdzgTv*g6WymOpEjoK!T5^Ls3X{v)UnjoupRBp^ z*b)aCR|t8$1XInB{1QR%7kvD-zW)1}9XrXLUM%E0pj}V5>*@0Cxt9HHJj|^e0dHUd zVX%&1;+&9g&l=6$hnJ!NMy~QJq;L|odTp|rNlxk{mc$Q~0Op2huqZSRGHR-1CegNx zvg_}stCicvShP_Ad83oZ5h#hTyK`D>CuV!a=N+HCe{C!Pk0ySFVw;(uMsjiLHaF*_ zGOq9{uW**i)@I7`sTyorR4~}K5@?4%1Zsp_Mv4F@UuRpTlq+NOT8z!57F;rwzH&{^ z5#a{3=wQ^8*KoCV8QcRKuk>HU7sW%*%Z~l7zn&v_1 zPyIX73(v7tm{tPGwqPCZhWmzlVoQGdGA^grD=Fvl0~KDi^r&G+dOsPdE)C3X1#dC; z4u=SSXaQQjoru3VaKO;-0H3etMWDz;Pm@8Mr@A`*h$?cf=CeckeJZ`L_V;r~?A0Jxk!ds0m+?ffXsXQNyIxj`o;wfo zB$gL5^`CEtB0C!;aE#R`O4^R_B%ATt60qbQcgC zuW|Vv%n25m6bJWA$^;;mRUN;G214-H#sp;5f&=r%(?Gx|to7L63N-h|8=$;%IJ4r=$L&tE{bu zww)E0sb2VULlEOeM{mBZ-&O@cJxp{^zwK#Nlk6qMU@#_jSuq;3LWeJjt%d0sP_JHf;4>1~ktq7O%?6Y%hoByq zt7?w8-Xs?FdTs|6rgYbu;=b8=irC4k)X6I+y3>#o?MuDzquJDx3YF?xiSWXdyv0$( zyYz6G-qf_wK&CTCmyPH01YW^DQ*y5tEl`11>c2L(we8~T zV3zN;a2M|Oi>P4q`b+$k$8JsR&`N|ubgyjuR=j=n4{xe!>VxNPF)nOv!b#-9TrQ(I zK#bNGYuVbBvG&3lXy+DnMupdNSRzTo=Q$)2>UsDd3l(LTS^YF9JT2eEh$l(D5^$6f zF^xm}ve(S1u3@zqfj;GWVzY8Y)tL0T**O_^H}q!OoAU;ap^htaA#boc%*@{EF_|c7 zrn?AKz%N`j9IL^9rk}f;Hv0U-izkQXlL-CUmHnpqLyPOmmHUsbD;q&V@-{A4gcZ%k z=FNtXw1Bi$K@+{7a3j@xp>Pm-NX=AEO0f|EE|S?pPf1_TD*N+z$)yLTe|0H7qr}en z=tl*OEwX4p-kk&5pj;>xvkB0Ykp@d2G79A1a@fn0sauKU|6G&lPye|lqn=7dt2Eh- z4sWF;-PHUhMz-0bIcr@X)`I2*GsP|EaeM&mzB7;5VJl~|#UwU0VjNWUsWJ9nvf$^t%i8wKKiyzKxNLWJ!2r%P^K9Y(u#E5OqgH zbw`mZqD4eYR{2G{-a6CVP5Vvj9*y!i;)f6~p&p{GI9mYQ$9}e@eVi+}2UGB2?7Jx2 zskKTVTvUix}e5t*S0X>Nk}0=5WZJ{Sj7h`(ja&zY=R1VXgTVx83kt9^?3~ z-x)18;0zUb%!KCClfc-1Bs?m;dN}_)ADrT>!i4v;&>x_uu14JF_kjN1#R}IAyjN}j z$LN@$qMQpT@&(&KQ(P?^T5lCFjeb^qji1{BOmr~xh5%LxMBv~EPFIdBL;^F{f*dO}o73to77 zSQSyRHRy5<@iSAXMNi6_U+d+$ZIFwsttS4+6C|y)WtOx}mc&xD@x{*hQ4T84oz7%u z0YGh0+AS>%A7knpDeJc()XPxExN2S}g;lCobuTh8`L?(IRPS!!Jce^kaLakUK2Z}k z|F|UDBot*3svtB`aGanx$5-)o;vz&vC`z0%s}`|1Vx`_yA+&n5tMMf@Qj*#@H30QK z+7-0>L7w|TMwg&2QLR;;;tDSi8diTewXX(19b3*I3QGTi<3!hzI@Gc8pf;Qz$raSG zzz0xoC3VQ}7YkEhJ#Y91t$&)^Tjdf4sjg3Y=zb-awDQlm_agQXTK+@} z_fj(yu`DhI%CB*)MI{k|+%j zy!ty>FHvDHvV|d<=Tgq&6Z?cVW`5;#2?}V#084VxD>UKC_zEQH`LWtp&~z#s@tyJtXegml6CZz04sqO zLajJj0J^hTq-DbgP*9&JO7XesOP zThMJ7j+myUyI>e+Wl|`rWQY% zoh}bRj(=mWjuT@N7C$lpfn4v^lkL{ep+lW1Yp z12pGD5oMQ{>+jB6uB~lQnpu|8HT;mLqvo=uJupw9YO%pSoYk3UH|1XnzydH}r5B^- zJMZ@YWE9kvTl;uwf!cEGzSdEzCc=CpxTyJ9ULfNC<%0GxuhBTOLM-zXr&h6J_E{vw zb#nfM<+HB#Rjr4H#vqc{$C?pMwaP%n$UNogt9DK5)@&lBxmWE7$y$?7(k%0D`?7Uu zCg(!qcfy_gDeO#R>eO<@aa9^c7IPxuXOnNI87S9p*9gK(A>$6^jy1<$c_?6DV(P6q zT2@#oX>=;cTh(-vM|!2~GCRliS~;pZ^cV5rs4Z>IHNVk1U~@!giOQ0c#?JxFX$XJ2 z^UtL!I>Av{EncMn6}E_dDbH9+V^MvQI;3SND+pFF9RFyQ6_9Je`_NJ$)rqv6Vw>Df zT$hcsjIV?j)Aw&uYU6hg7qpei(gaDJeP!Nx>iWOu^o>0FgC?%ofN33wKz*{3@~>{E zs%HteimfmfHzsL7S_!ifW+ldnk2b*yw-Ro_6KSciGk$)3ZL^)B^YZnD0lM_~ht$?K0#%c8+^ z(RUo8r>KKBBBU?_S~S_Mm%cs+-1ZbE=QD`m=bT%cH@1psYjx|~r7LZ=^d*cB0u^GkIl#9J{&8AygVjm-l&Jd`@nNEJeQ>^hh|d`QIK9!eR*Zh z;#X}H@qxu}4N3&}^tq5|G^MuxHm47s^ZSjd9m-0dTIE_#`beVBD46vP*u|u~p$atG zns>yvdnEg`y?XzBIVIS}QtAenSRS(-ow|7p^jnng#;87=0`h#F9e$?rDP4||NQ7Wp z=a2rtSBL0J->hDMiKka?C%Z<$vkOrQkM>;kL^-b{q9(WlGSru3e=f!m%a_X3%E)P` zE8l}tQg%;xbi<_y~c)z>OmqDap1cGgjboF*Y9EhZ^zh#;H zHP+q()jAf%I`v#Eyk+9CN0n%;WlVjna8)sHI%g&9<>8S*>nFiz2`d7%|2W5Zt>=E9 z|-#5AP>hrQom56Hdjtb*@yfJsNE@pP3jD3T+41Z^qOb*oBSo zbF=vtN`DQfjW4rTTOJ+-ABSF(Uyc^YJzd9|wszSTvNCuU=s6hkV31H=FK6#$p-+w7 zc59z-4)^UUduu(U`yTVUs60O+XnZBFT05Q%y&snP;BWs$E*85XzqHK_sd;J$?{rmK zGz7PJT5M&tkADN8g?{Bc7%p^ONASbnM>yVXwTS%#x&h*CjFx>)m*>m#f?RF(=V83s zT%W^zJuG_Gcp4T?&XjBs$He9(nSmj=AG$IcuLJVvY@SyQ0uI#l-AAKg3*JFHknmyG z%T_ti_x4`fA9Z~g1(Nw2wiw_RGAo?YHa;8#-*V>5{z*sfXasMUjtQyyIJN%lv&*0dO z6!Mhc0CLvd+h6F>XyYtv0((26KBmBXVKDnKoI68>q33$|Mmw)QmPl0E{r4$BJfivn=l z3;k!_lwDf#X$kOa7gdHv1UugAXL3iTb}9Nz#RhVEiIPdc5V^6&BXMce+$<$M&2$T~ z$nx-le)tM{TcqW7uymA$5`%2}XFBpm+eHrerP1U?qx6Z4pp-A|>bWa1eJ<|kgC1s+ zs=ndouf=5ALZZ^%Wb?NgR&B~ZFF|9irSt#3^xsKCA(tpKXtTKY z6`_hP*f(x*nQKCXF|kSOiSA!%Z?DsD4dagKZllk6B0WJl)8!7hU@|#7aTtzwKPp^o zRpBt*U!V7o>lQyV!F=GQkuP>|aG>X)C5j0>EvNmE6G~mm@-P-{wc1+UuYGvy zm8#dvv(@FSMKn^c40k@k*FmJR~`WO=hm~7i&8Oew&Yzl>@oa+FPf4ZU{G_ z#DMv2@VHd_vTS+NN4VWAi^NaQ*Rhe6^^=(;W|1?q+PyUpnB`cT+`j7A`60af%>+A` zI(AzJb6GqR89^?Tt4>*(6JzAJdMR|yQG2i-}`T8|tm0(s55c|bu*`ixP-1g15Z=sQQXC2vJu&*(Dc$$I2g4|_ zz^AIwXSa(Ixq6z={WVtFYwx+~JnCGEgDLxJ)gR;7%jgAQm8vUzPsep$!2XE`a8*9` zc^1Q4cNcbsl=(0jD8RY4RY~N|AdG2esH+R0bKB(oMz79<1Fjt7Ch6@tdP?rPY41=v zN0|8yYEu+n?}S=*^eRFLn1V)k+TI=SMMw{e6-hZ?c2jDQ0hm9)*UE+#)u1D5wBB^< zz>o6rzO{tQeWaiSJpfaBvwwig1(-RsGX$teBoPJZjAT9rSVvw zk((2@(xLgrd-mitb(e1Oz^mk`{oG3V_O*G`GpS}}6Mxz>PG?8Yk}-S0J2PF@fHU|Q zwL~fHe+);F(>U53rF1DncEe3|!5v}tRl92X{^Wmjfz7rewk zo!)X4c19NDDrtWPq5aF;GL_F7gNB2$mL7~-d{UK+(n=GR3@^2V?OCez!whq@5|uPn z1I^(7G0EQ2l}-O-$F<$ z0U0x0u^4%b;=|13VxMBYkO{+&yWv(?Y>1uc1h`uJ*O3smEisM?ukRomWdV|_RsAjX z9x_#=X`({!x1zhekOnGZ&%HR(eOd3*C?xdr+6;Z*3Qxr{6w-L#HDeF=CE=vPDb^&) zWubXpk7Mib2$?4d%%$yGp&Guq%5S1(54gg02B8%TT;%TK8&)}N&T@RS>L(uMwIZJ> z(X1`jOnu8PSGL8frsJEYmrIqnN0yx$A2L6uRM7Mo+3d0Nj}(hfoz1bi?Zvk z7>q6WKf3b>x?=%Vclpql6%*etx_h=@~3w# zq{t`Pzv{vNv{?Y^;Zy#k>nxF!ow}X2vOwCFAt*!EMms=Ozsmf{}!9_ z_qUR?nOBmwlD0Avw#;G^wlvuqP(J3cv9PE$g2^_|Y#u~p)1gmq36k{bxuM8Wec*k- zDXLjNSPD-njHq|g9dJ6x)YIc#SoPY?5gs>GJtyu~nskue(NeHL&qdEDqkpLVNN|cD;U*WvKP2dIq2o0@UgF@ z3=y|58lZPIBkmLVAR11K9P#~K_@-R`_eOS8YCO)8qeDy|%?gi5iI}q^$x?=^Hunk- z%$)T^XIuI>G@!QKNQ?HNw)Jp3Rr*(~jD(x7QHTY%g{>2YRqHa9=@lv-Gpo+t=2W9l z3t{$+!9Se3eWKvqOLz?t2YwV9v4;W&?2$|Nst_}lY>nr1uwM*^Ca&$cpsxjU(ikiy z($*&9^r_m4VDyyK)2KxPjT{pnT-0$jzOZJgwxAZ5mP1;kyVv^}nHy(>?M2N6 zo(h;ade@85U|TH1vN{?aj5H59ELM`IE2Bul^hHAJ1Y5v~)|GwFWk2TQz^Hl&-qXociyn(70 zykoc0+Ww7DHGjHs{!8h0aHPP-fG^m?1bp({N!S3wmg@eP-{ zTQ+;rKqp?MP3t79y0+^~$s{GFqj`G=YGr1^;sJpUB@V$a6`^%-pUTLTW%cKUEcq2qkwi0J0`RN~>q zz5Vl#OgD&|#+w(0+f)L@0ZgMN>)Qe@!`=iL3Znj;8-(bSO$Tq#U z_%#QCi4ozHP!q|ZM|othe<2oFB+^b{AOFB+>42t-8tDYph2xq);}{N9LUGQVa#WYf z!>iRXsOyNC!FJ#6H;Iu8N6M@q6OAGiMI;GB8iFz$%3#=yz!U%lR;5f!@AZd(70DeL zUNrUf6}tPMt{~h4xCL?u<*LFdYQ|7N!H|oEaXbY8Wv!}K1eC{t?C;dWt)S*Z1;~e= zUkfiR{BhQc9mM|uAlwaQxMxWJlxyXn|4Tq9XVrLt`gi*8DVDq8=zq#(Su~2PAz9#1 zN(+>Bx#C>IY}4<*`7L4Jxl+ht7;HF&ef})_?t>jCJ3{q2L_Sx!%zI7XAQS2_>Rj;H zqV0C!$8tLE7g=oFNJwV`{9JCN43Eh>b#N5t1p;DR{~39-oa__OgA7TGiplP7-*8LL ziO~6^W}MD12XFHZuWVY2g)7ksbP;)(wPJ~b;5gSpA|Wa(qPK645v(^ngQ&r0)-=T! zK{RfV_k15Yzq7D~z0?!5#iJAsGAtt3>zgErG}o+8RuAUAWsofx5b*@09V8O4S*2?h4jRqPYw49tvQ~5F zldWxpg2}3phCU{HiNMbBmnzu2jv_nVv#}$zNeaIg->YYQ7!koO+*18LK%l%h+)tkK~o28q4!rgk#>4h7MPt+H4x92R^PqF0^ z{|hYNthFpBKJh+zvAFU1e{6kaRGiDwb&#MTXn;U)7$CU2yAwRPySuvvcXxMpcbDMq z?(XspIrp6R%C~AQ)~>2uyL)~-!%Rt&IE@3wsM8D`hvT{JNO@GifVlU&gmbW-(3gZGYt;WGt5GZU$s6y14xd}xd7P~yu2s$$4l6DZ>;vbUB!%vtG8ovuE)r+Rtz`4~n~ zcC$+%XfWohO?>0mc=nl(@-2%I`b8Fp4z#gK>OrWbPu4t=b`LzW;5rmmb@wZkNWRM8 z(qNEU^!}6aLHFSI2_i68RtAsB-vF=~;Rr+#x@f6Jf{k#AF0J|8`J`LFZg~uYr#cn6 z;tY@?(eRcM;%r%3*bvWp{fA^7*kDqZsQu`-z$duF^z3_!+b_Dh& zM~THjpuR-;r-Xy62Okox*V3@-cuN0RFq>GcR&gs8%#E8`voVhfP6$KRfX`Q758ja| z9)1LTC`4Z>9Yq)+DTG6UEJuh3W`9Ac^a%nX_t^4oyvC zD<#N9F{>N1Q->`=N!7 zIs)&Z(u`1?P4dxxfCu(&KUjzqMIQ^eC71-gU+TFl* z9=kOrkxV@^TO&H#ttT$Kl5TX+8t|Cb@>^ALk)OO{Q}TT#nGm5h;-JCcqP?5#N6gvz z05-EdNDA?!U_gm-ALWqWn>Q7qN#XPx_a{Ox7QXsQ;5t_Y?^JGni@Z zC30t0oo1xSSlEx|-CH%P#vAr*5dpP?br~pHBiUnNFV+iPDF*D6^dpPXFPhrWZ1ejR zIEE#1LguEOB}>GH<`@MXb!oJBYht}BOuXfrh~`*g*?zPD95y!#TpBPs7VM{AGL@E4 zJq6R+$7=xych~jydWlI-3HXlQr93wOpRK>fu zWRV}~at-<0RB=;aXz7di22RPCK+B1snW!e+)kJ97ci{@xoF`q7JJi-PZ1pBQeYH|WZCOiP-Fd0P zvQBf&r`*cU@4=>78HQ>!`JCYTn2g%>CmS-X4 zF7<23*qLaob4ejWJ`AE8GeCEaFUJ)&PaijenL?X90C%Ov|w7=nWz ze9*_foSuAt2JWvq%Bc~}>TspxLu(`aMk~zu6F){B9p|{Ef>TNM<2a#<4pv#voQ{#n zAb}X&rtS9=o04u$M$3u0N+bYF9B6l<#ogvi%s8v-l8$_ItqSo4Cs9!>BfHWO3YS6Y zcwjod{@M?2G1F*oL->!s-OuoU4qo zkCr8=lCM8})a<&b4`A6K`-u+;-!|!MU3Dx`-HO(>(>VF@pnO2oPnaI9r15p(5PJnG zE`FOTi^V-17bizJDc*0U(5;6qh~^(uj@{@_EOY%I5fna7c{s~7I(`Gk`WhW}5sng7 z`z0#6@p+%aS|*KB!dWJf>wKjQ=RPUVF{|$m1@C8dO5-7hN`Jx`g|szXFzk<6UqpAo z;#lm{nCqn2On9YX2l_1XLgJ=p)L9tHY_a_VO48O^7R{kn)l)S!cJ-6ap@Q3j+q&%A zQ`Y)y7x@$xyel8E#tZd;cG8<$KX)@m0*D`cMDJD&Ai6Rn=KmN?mT>!U2qQ?5S9S&p zQ#15oP=!A?A-?q|9^>ZGuwn%y3*%C%gm-{VjK(kecLI>EwgGxqLVXLd5v`1Dh=gE-eu-7HPCiX+j3C zWDrez$+8xn(zM5#%7L9d8qp|+F-Ymg+Vc%tn9Vf^$TMZ2{V=Rv zk0d@*xgA99XrFZzuXFR&DkDmD zb?Z>4WYd2NlFnd}gp!LoKoy>zKA}~%nwcoA)e+l_DlueP_d!2L3*LkxDqw4}W28C2 z2;V2IAO6PG-b6J~M=fwc_Qge-7+&LhqQ(HG>DzD{grsy`(G;$);kN`gKw{z3F)@L_kN&hmof988+n=-C2~Su)p?Y&uO3dnWp!^vNp)Ap0o3F?h zgL!;?8F-e2%3p`am<}a-6O$AlV9CPowCE}UYeMBa`rT#=+DZ=SvpAPW3kQw?ND7$q zn5GVN((tRC_c1K_SVJtOH6UMYBu7|i_Igr~{|8KA*hnhACvBBj@{8(2mpAIu6pEzJ z&5{#=#q6aud_;6oBkVL+JtN5f1Ew(Oq~_j}wn~ZlMR#9TS4L!QYTdkhkrG@MQ@lSa;Y!RB4I#}7Pt@vd7o2oU>$tStCwEi(f z>!boh9aNgSl(HT#kjK&-mv?z&AGN=_q4}Qj_gJK0v^K7HV9_D?*8@8+9i3s(EmbZ8 zcTiHF5br)WxS9>vw_XQb$;xNGnd=Bl=o$6Z z@|tM)9gctO#ee7J3($L69eOHavMFe78Gm(l^5fFTnjszI`+nC6&RZt#L9$m<;izz$1xjIYLSC^XH@WI(-*%spT zu)D>UbkuR4qLUHaooQWnpHK1~s1PebFSg%AH|*uq62V$ZTvP1Q+X8pdFN6%%M9VBe zZ;0ZK0s78KV>S&b+={SAcO%_=`3b<45id<&N0RorN>^!I$Dg?(6zbwWEH!d;NV}|d z2K(BD+&Mg#&6bnyC@CR*KgtP;b@nCMnD4VqB6P%Gp`H z(ClPeacg1^f5SqyGIJSogAWBb9jx1sF3tyhbx)5|bWm8cNl~Me&9eVy`yht6A`OF= zT>B%6(IG~n&RjCAsF=63Ht*{DbgA{1Ls(hACb!!f)zfL)j~YgqsxCUDV!x)w!?!OI)FS6Ltlj{-O+J%t z=}R=~a>Ed+|8{jzD`hUCh;6(z9n}k|s6vz+T2B7uAxV(|l}!etZ!m$G%fVUObrN02Ak6VIX^8PBX!TqW*+5(+@!y;9%*TVn z@cRC%E}b1tlzN~#3_-+a={j+nq;Yk#AI4@~us3c>N*ZvF*Bz*i+JZzjEFd7Qj#QYG z{r!G|tT300lm8%lVzszdg;Q=k7>3X59fwnvioTa@-HTnGWp^>)kv%=7BoZA2<;RA| zB`Jm4gwKc0FrP$wPGnKX)6*q)c8;NNPnIVQt6r=TSU6gms9YVXT>OA|6!cc3+Q7ud zZSYMkm^aabM!^;arfTE5mk>Z}_AsZwByu+qzgiNh_&b)ia=|cRan$^miyltL@J@8d zhJSuwC_MV6X^_=^g^950cT*{6x~1VXHX6Uy+M1e9MIv)}BwSh=dO`i;Rp6Hu

dH^s|pzE(}w=GJOR~g`d8Ll{jX6hF$!yjVf`^UXu(a{ zj=?a)xz5M1M+(6v;TH=ldeYd9=(q;R^xt_@s_jP<)a zREwL8l26&DSRm8J7IK}<#P_IEyKO&DvHTYpW1VDCe2>|99;6f<>7H7$<~MGt+)L!y z8OO{n4St{qRs8f`J*=Q=~Ga99;74?q% zf6D#{mEe`i&WYZ=SIZhH=O1F+H$T?_x8#-|rChnCJ`&p77u|9qvxy&5S9bS&{x47+ z)*~4mgaxqwOjRlSkVf2`GT*4gP-)ZIBN>Xs)?70O^Irx3NW51D|EZ*|l)b@L1XdcA zC(Rq3Ich_0P^1)=#oQAoqjTDUtwD?(s-wDO^7{uE9#sEjIf z{M?epGA@d#vUWlVVvf2_d8?!3%bGG{S-hHRn;#fU7(1|69Kj$exzSM2RqGqH#tji} z0$h0Pp_fJfP`y(!9EeJ1M%$MkM&DhChMC8zGDrD=4P)jHw^)(*j8RdozTA1dDz-%Q z(`ExeMV2z=d@4o6blWPm39NLI$Rn+;^c;*#zH02$r7}SaP9gKRlS+7oJ!A!G&Z_O5 z%;cuJu(1ze0N@#Fj!NeXF(PN=3GNrK6H}LZ2&rwGZmelxHgUU0z(-e?qqHubngln* zgw&~Hb(0;%jL^DH`q^u9PU;i$ZMEwg)XW?5y7#-T=9$r--{5fdqdvIzo`kecZg%o# zh8HT9M2(lSpsx#4^p2AA2`ZlBznZM!JX(^4U4Mlz^tWuqbQC+~t4DqlUnEQu3l0<- zU_F7Pqd||3@Ow)Y(86uY+z{V*tXillR{cx^W;p4u^kyHf<(es5iH0ebN+&BbEDrtZ z;7Sw}CAMNOn)8{NxGOX#0a;7Zmj!W`|382(1xqB^dyFL` zMR=!~mWH@11p#HJU?_berk44(r8U9KS!s20VdD3#S2?bAo76l_{lj`c2q}*~VNz36=ZZ9Ryq_inp$RrQ;>ji5m9Epm!|C)IALnzg z!%15i{PtTfo5IAri%^KN(%q-~AC zp+>&&^d%X`vb6-r1tj}8HKfna^Nt4J+){Qvk>_qD@d<}8C$F^ZxKZ`3+gowe2m6nA zw;ao17zk6@#P5I}BT-I*{G{zT9%w;fIZ54^&7uYqmLU8z)e&kPDwFeNcA&?>J}k}T zXWVC6mookJlu%9PImKv|2|W%n2Ui^XuSUqWM=(wL8nbrR`GpaBxOT537 z%{Z!Rc{E2Q@Tr-I@IaBhf;#v$@u?9Lpv1vfLnVj;JNVZ}@hoT9BoKfp6+z8Y9v*PO zzw=b@IKcmAVZ#2+qUZi^mgogg_awEyDA&DzMa!0wNnDS zI5q%2I$*GSdt*@%k{Ae?`nqG|Kdr%H_wknxb!%g8Z#-U5ll}U`(fX%kt`(miSlz;c zUDhOit$t1oaFoYZNS@ciMtw*ZIMnV6A?9m1e1zKmL1;C=|1BtE$7Vgv&LZ-X{{Wof zCPF6Y%p??~fxL&8TIa%$mL@SlTpR#Ewnz$MFdvMp8Z&$5Eud>gpnzP}OXe+XM=fW$-~Jwi_`x{HpsX`LEUUy}vIMeGlMp^4(xZmo_q8A2bHJpX#Cp=la_D#Xe>}2)qj&cY3u*CRtlXOaVio%z zHO1|}>oS4N$)oM9qBZonmfzpap*Fiw$C{B4k}I3q21FXrgR|2kF!AsEA%8*;!Uf1RE?FKW zDKQ|5d=rj|E8e|XJ5l*UP9Ql1(2K4TS;#pqB3+t?q>MaKlZz?#fkAS#Rk<1R4mbplMzxoM5g$h5ju;yg z*Zy)wsI&>^X>=1$O7(Zp@%8yphq9mR=O2bO-BBI|?KY~Iq$j8Ypmbn&K=yLd1nR22 z33{n}uUt~)h1wg9bhlc{Z`!DWx@-G3;-0C6TVew0#ZhXD9$HT-*Kwqx1Aaav|KD@{ z{>(boi`u?Fz{#5%MZ}{3GyZX!Hf~1Pn1$Rk3 zgaLokn?shz!|+) zFhsu_=URjdgs3e=N=OWfrEEs#Qdzy)W02bExv}_u6U)ZCqd3|YcGViUNqFow{rR9A zEaV((KvKu#>&HGjD4RsfmT3) z2Ie^=M_lf?~eKusH52DwW2KIl;s|Y`0gcQsZX-RUv|_dDKUnt2%grigKuf zTL0_1C}$Ot7xC5Q*5Cs+Q0Ch2@vUS!eO$?a)9PnNtMmq`W#UuhyP&5Yx6N0|_17Q;@YKEjVae>9R=^a>LcV>brER1{MvEGjB zD_o-cIWp&Up%)!onAx}D_;DRN#)=wx;i}*`MrO<$3py$E3_8CEcxhmW?ww-JT})*4!dE$ zeJ2=}uT7=ilA?Dcoxo7%%ndP$Uhq90R|-V8b-3QWJ2Ef0PQGa-Ra1u`e;^!rP)ay{ zw9G1j71R}Os zJss&O&DNw37=%f@GW_=g!o}|PEdYR}37wR+iboWw+0Q;T3)I~wh{q&{3Dg9eFS$&W zh3ef!o)14`vb#Q*PlrW~9DES{9LA@Q-KKloZ)Mdm2)Q#U!{5O<{n)PGlwp4Wy5dp3 zFhmOxZ#IfktY+>18vLKRaEW7EL;+{>G`y8e1*5@*ZBg3fq*6Reysd(3jjHxO_PY`JlBUpDse zH_F{JI1c<0?M)(&+11SHi3STQe;J;cEA#Hf=}4+QzacDaafFnH8?u z)%6zADZ3UP`S|7+qLLxr=*jVeLDe)n5OeABt|J%RY-oNUMS(8^0c5w}PL9 zmFtxfM+wSY>Yy8B5@TOxW)=3O}!k^)oxv> zO99q!a={Ial$5s}p0ZrLqFdmb@f|5fuphCSDt)7P4QQ}4B%$_*^5VhWSv{vn4o?^h+k~5 z#*R3NvspcgQ)gcHk1dPNle0AVd=3mwaZ3F{>Ox{PCZ(c=11iNUQY!`(XiGLO!W<8J zW7({3ZEd^Pd9DD6QSrQ}q`|?w`skmJnK0`@n%?Thge5)p^OFsKi z)=D2P5Ij0Yqy8|va63FHU^{&(YiZ)Xn{1X?Htu5CMVKmUOM`JZY`$961~Oul^OW8l z?fBsH+*VFPLyHy<9@g>`F@Tw7u%%;Qj$>>G3l3`#dGF!!%37lMmfmkUMr=Q#AX;dB zwvfqM2+YGMP84)TJi#%{b$-+)}4N`tn!SEC=_ z7zMD&^5WPYD)koG>%*f_@`4|*mx-kbqdr=Knx?(i!pDCDv&m`=`oI{lj8d+ZVtLE} zje4$F53PqeL2cga7?deR=f5i6HMB|>{E|-f`j0`YVKIk*g$paE?ah{z#s@F3`e?u| zC>rvavIHhV?l3IhVy%up5>@rNmf6VPK1-~18H02RW z>SoJ<1;3(tyyb^uP<{K_V{hP2<7(K9D%=w?WJi62lzazTXs$+_-q_)u*2{pl{TN~{ zH^&<}pF2z+yUQIHDs<-Y%91VSB~98OUb|6Bek}ZZa73L$tullJU1IHO#FG66Q%vB zoL7@QA>FXd(?Q6b5GSi)&#Q(!HUl4$X8aj*c(etDqlD6gMK%06L+!49H`gEM|A=#@ zCO7h5U2b~oeF8mxN|VsnQ^9EhiHozE>pK)r`jBalA3EgEY)xFbRILm`S1fuWvNl97 z{^0ms=T(k(e7@<&GGYQJXr^J83E>2HHS(3TaUY;*=S+~ zl)d=*K~B@T3Btv>&nKBs^0!#&u*0w$>4$G|2yUeZ1q#h9(=Iu)33!f{`s*=HUTUMT zmXJ`p9~00t)jgd+{svyDqp+$pRKVX|uRv9nf;KcSLHo(tSS&}-AnloWIwZa&;A(n% zLW2UJ{sH!kJeH&2Ajsh22&QIqe%&GnDdh?c+$AB<-dGX{mN*dq#{H|vTTKFC+O8Yy zJy9UwopJbr@EO2Up}hmr##vFk0CJk$ar~hgY~(XF+drV9dBF?tSFED;<&%cB|y}K32e7@y4y<~P0sq@JR@BGT^qXIMt6CqXEI04iVn1kYa{(!>Yn)70`zqw zUwfs^j7E^hGK8br*zckX<+b`KF~fgW2b>+e$5Q`JS*Bh$Q81SfzzE!7yoNhUIf%jn z0G0Edl6SUF~(i zzP2r_oamuTbPs@=)NJ)QMZ11}!m2M<87GPi0<#{S#FojxQ7kbrZ_p&(WVIcbZL7^B zYhXp;f)1joW(j434s2ZQK-G0=`T6-g(!mrTJdAV9a@kai#g`<6?r11^1$e2HKL2i0 za9p~pT@^uGAyZ#gBHnU+%HY0nlvd+O=am20(%##xv$-plqTCO`X}vGA(Wy8AyYwr& z#adU`yRuV_U1=HZpj_2-I-7?)=0w+aq5Zvtj+kOl$49&16acWpxKGS+>>Ue6XtokA*ON->zDTIOb zW_Ap|7gbx^_rB{Y-2pw zq|((Xt&RhBh#eFVR(~Ms=m^xi%Z);4Kh$&D}S^6rKzcT_(ADS**#)-Y zbTkO&aJccfn~dGAPHe;#zNA@%*(tw)NY!yr3}B%VK5`Pba5gsXqy6Y&jo zfEU-b7Cy0%!yz@-nPBFULtFTA&D+QWL!BR*Ri5JoKa5f4)@p}TP6cX82Ka;{j85b{ zl5d3U1kNtL@;ZDS^=&jE@}P&5mhD2zT#yWrXE*(}}HEk0cD>?$SsIbomSV)nQd9WwkV_=eGo~&FhG!DE1KuGv&lN$i55_Pv zoU7Jsd(ly;HfIHTxXNFri6aCk;LNTOCy1}K9u)_DCT&5%9ae>X7XzS_9wHwK&>%- z_M;Fw+A3%(v)U7w4w|RUi^q7_f%{1Az(oBO>UWUctvuS+*J@1SHQl%|G*u-=>}Aae zQl&83lW6`v4ki;WgtoExLZ+CVJxCdw<#9hR%Q6k=F{Ixl&Y2yl^WetbSFZ?oM2}3O z{k0;wUu^1$K~;)NwM-VksT+4Yx1H>p@#6*)NS1$b*c#Wl-R`B82rkd2V8Wol;>P*; zH>4Lpt<;-OTib}MaW<^c*Qu<#*^3ug5Kfa#IfU()NVR>#Nuq-FbT9RsB+;q9xlI|< z1HTw$u0ZU8-UPSuX68x3l|MDB_Ia6t!jUHiFP%$qxAy&Y%j`yR&b>p-1|6!}2^ z?Z!6X$qmp4ByRWjtKQDgBj(IH7nE`PdlcyRxd2fXq*02OFho(vgF)N;*7z)uDZ-M3 z#8DnxQA{&WaX?%SH+aNYXGRfq_Z`V8oVzWRTYc0c{_lI^-nHL{Y6FjOfZ7JXTmijY z9L-XJ-R}P~ni2k>xUB-S9m`${{WtGl*UfE##5h=S1Yq@*}YRI$1`+7(&6IK-fiL929&YyYv}8Jrh3u1?e0zHw<#9*Z<4jETH6Hsy>0ELl>b@;J&Ph@d0zyTp)6bSZ!uGKl|G>= zhQAM(S@P?@{ziDuKZXV6U3}5+RDHfQ~!781-^sfN4{}45g z76YF8|4TBl*-Ul0^nR~T?eaq5g2t_3uzy?UlI9`7Rh$!m6CJLimm~oM^HltcxNjP_1iru)Q5&UWn3Y-J`0f;d?suV<6KcF@+N@fB(#m< z92g5s=`^6gYUhs!Qx&esk-QSisn%5M7xy6fkL%t%TvNc2UKN$Ib+o z_Av$CHy*yp8D<*z|-WjqvLO zZx&0`e}`mzyG?M_Dn29m@{tcUg99m|sPYFzLSHnYk4I(9ymhyVd+y|@3a90Gz!-Lq zzs#-pu#&y(L%2w_d&P47QmRB7!5Yt14>dhS&OlcYA=iVCmjp#wrw+;Col z5u-+sF|u2Ws@?pSbsm3ELFHy2Id=CmYQMgooZ?|Hm`tEa#_g~oZVs@5Pt$XB8_?4* zX|ZSSMS^R2@@a#kT+Gio5S*-`twFLaXO1{b=58{7HEbxgKUv9NU<^uLT_Q9`VhT#` zk=O)SMK_2nOAwyY0Rt{`fi5q7%cc0IzdXd;CidHN0X9>u$s-cG0GsH>!DU@@TBbyo z?h!XhVBieHrhIL&Sz<)(!e_@d+J($3d3O!m9F)l`xkF+dV2RuSy-Wsjnup=_j&4Bi z9#cD)z?V~tJccB;R|U88+?XSi2PC!u*2s-P%lzi}cfTT<`GD>jy0t}zN{Yvz#HRUI zt0}jXf2R3o)-DV^D@4FqJA^jOSxBq`HeYug>X}|3mo)@d`K{wpOlU5ar)e66FLy0g zORtng!>_Cz3HG>`nY_3(PWmg_!vf&HR*+ALu?Mzk3_lQph!&Kx0rn1Gc~H)RQTbAb zW~A?ou(6d3Y%V|b;%UfvBsAzRbwbdn7u83|q_X$X7xSQE!iQ@P>Fi(dz=j8g2ceNv zC44qWf2wk=Z5-&Se#NXb5tcv)or1G-I#2h&`9e}iY6-sV=$JvFOG4!ef=Ol6J#%7AaFb?buhD*tYz- z6aN@P?#=?g;i<_~l@i(MAcMCDhdrnp$Glmw-|Y2acUUuK0{#51lwWz}K70$#14 zXZ*#+t@Il*RT4OB@dKp?HG?&S;YrGk|w>hJr?b1ZVmk=pd*$-Z@@<; zwGyhvTG%4>@g382I(>t3II8j3qZ##!m%crbE`4A%CBf=#jIA`d?f=1FgBvc$c(pq4 z_*53nn$E`7M^5Hyf}8#wGs?L);0$c%%y%TOuceYOsbA`nnUNL)Pk%@{WK}0~*+~bJ zUd7UQ=&QqPa^2P2v8C!WTf~g4s}cI5#lBL^E`qB`-Koe4d!_HRq%m=A^rFZ~j=c<9 zVe0hiYP6#8NzN@%fg7?=aRe~Ie)T^jb;#QE_Zk(UlVs);KcL25C+WWg{?WU|C}=H& z?0AeII!||7 zPlA%cc)+An7VMwQzY$v%T+fhzRVo2z$;{6F_v!>~ZwEV%9~TghX(DUB8zfT{X~n0p zz)6Vt3CmTBhjvk6GbGY;!H@?cjSl)TLfZ`%1A2+7sL#1v3f>Z=S@~JA+r^fGuma821^j?p$T`pWo(i!8PJdnx_qUC1zXi6)OgBvR zk1z)Jclnr(LC{NU?rkL4QG1*Y>VTnD^AZp@n!rhv!%|m51NWRf7oUSTgOm8VMh+(w z*#L}42~dw~&)e-RYoScC5y4m9gJ94U`vJmYZG2t((GUECrk*(uHnro)RBq#`CnVWq zoIh;qD5!&7TX=`pIMb4Kz1qH?cagCW+Y(N3xiMcqI90BuT9{XpB#r)Rq>+6%cVAkk zaF?_2W|*{G|3k{g8mBRQS?HADF7j!>4Q{uM4j8bwYbZB=I=r>fsI7lVyHIOC$!yPU ztrYRd;hMxb2)`D3BzR#+cpU=<+E=pMo6I8K(d^{fB=+#XQ6gWH#5^SIhKXJxUk`KI zn@()sX?2rpm#D4(LBkODzybH*_|;uaY8H@IxOtFHpArG(SmniG&4{A{)& zDKzP3_Knpug!^i;0g1S*X7+3d(aOTqyvG}mfI8>45DHv}Rqo$V68vlSFkvzQV zXyE-N-VeQnHz`25H}EK0oIx^;?QBvYfVg2YTWoC`{cC9YnuqI2Lw6<4P({KG>2{Sx zC!ab!lr%$+p|eIm!TF90RwJw&E3PDHVm{1D3+-Tsp7L6V=$H9G**z+exJUZ^8>L5r ztO@E^(I;=uk;V_b#~YnwzQl_zIxauYR~as6+b5z+&a1{d>l}aKMzwr1;bnh#OI-Nk zcYQqrh6B?fCmhrY7X9h`*e9iUN*fi1LWBO>JR^4jw5591NuJ6BEZ87~7yk_R^)HTL zK6Mi$zuoHx%~mw#DoJeqq`Gt3S^5Owgb{V$Ul~+nSa}hyDxQgPXvEdvBJ^96uC-`* z@+kyhqL;0R6X;t=E=8*9Z`Xr}PI(^}#o=VUQ!?;tc6*)Y!S0C}$ZP@~JA1WN!MahC zx>c+_V^+WvKjT*l4MK^oWD?hXiy zq#-J~wLlUMv-bCLw@iNIxFaH@e9PmaPNs`Dc~cY=p7M)&O~?9MJtk(#6k??{d z_=WKa5u>0)dE51idrOrP#g)6e7GnI1n$J*v*rVf;t4(Cay}x7tUz*e?EjLDLB-z5 zRN}KIXN%Ga?8PwA2BBQF*{TZuh*3GL;qfiwXz}s2DbYO9ictob1Jw=p{;G;bh>r3LxSG<@*=GM6QtIj0gG2?frZ&P(xSsI6XWf zmthL8MF&LK;@tY4#Q3R?`w!xF@vRPfS?OiO-c8$DC|@fHPQ&NoC*Nx7{W|uar9vsc zieBc5(aU$>euce0Tm@eFh*U<4l2>KnyV;iip@hfH!Y5rX0Kd5szZqfae(onpxkl3T zGwvRHr?=Z0Cp&9aGO;^rD-oMPe@Ce6`(eD|7tPxvA7WQKc>k+0hTHBch<+F1Sx%;#*$A9fa~MkNs<6;}1=< zlwyk{DJ%VT0LctpD`G1u@nvq;XTn$O1rVdJ{)xr{D8ckR1%~+7wsyaM@f9YazECt7 zzI=-O0I+)$_W!Mf^y_LW86NY?(%PnplJ({Ee9B4}&aegI7i6_j+8g7*CS1bhVZvWj zI|mb!Mn84MWYgHF5c_!$R*e1np$bHD`Vwlyfz1ZD&IRM%n45k-=^J?m(UYm{lC>Li ze(x5KZM`s1o35>8#z>0UUCE77xnSR#{!S{{9Fz>7uQxu%j&g@n%Nw&0RXM5oRi#`` zjbv$kZIY;j9!Wv+M(pf0{$t0ba&SMHEIXxXr{jieP3$-ZoJmxLmqyrA1mF*z_<_aPm zr{xUDfXI-}QLdg8^}Rfq{7=nv)&D$doBLh`D=vrH)v1A~~IX>lnFJ97qk59W&CGJyKpS%Qr@zuk zkjN3)h-KG7LBEa3%RumN0>E{ep)MpTMa*{nXi6$G^B=(Z+`9B2vJ9cpL_jNz{B?gF zJ@#T3&8Or-I2LGU3i3k6-KdOny`l^;GK4%X}LH(nc6B_Av4MGidh`g%w z7+Wp{&0iKSbXSMneT}oV2GjdSLY7%m*quS^{FeC4h$;4;QpTVp!HKY>OG1nC)!5`~5uZOk zUt}hRPG)`dC;7;XG{lTW5b2s10&c_%C~0sK(CIo*x%dwPp7`7heYzI)Yrc=ScNvTP z(%r(v@<~fG81;*mO%1LKMzm)2X|*y_=VGMhSsJ@y2q;eL8V`XK7)QX1#spYUu}Z;% z{nL`w?;=@R5P=J+tAa^0k5+)WnVMx`vTG4Zfd;_gT#uKbeGQ%D>Q{r!$L^@F)eza7 zTbyP=^){%#&W?=WrEFyP7>B$!+_AeTv*fQYT6}KWbn}59qJ}Z|y-9c$h-i)C40)21 zdMPDHRkU?7;B_;wvK89R8_Mplmi&rdxHy=k5x_@+PNSe&J5#}H2obJ28&f%?raYw2 zd--8K*CBZoA&D01!I;i7IexDfk}u{U>=Xs+sI|CT0OX}$S3v{$OMw#CbI02T&ruxS=2X8^yW_aa2tu! zfoG;o3H=KL1d_u(FyicG z=azc#V9PSxqnJy|)%NBc0w1~DXfl}SYc>p(xFj$OHhMHGpeHswZ8@Rfn_6-o(y`!B|^jozuhmy zQ}r$z_$obk>{H~_Ayq#K3Ki7-BEh7mtep*rh6_k}Ks5NKHu*iKfCsABVO<<}FH}?y zN1D#fAf6n*>AY+@x^h4vKQn!>3wB&q_U36-@JqCvp`L+A@Sz-;$ZIYQOzq5_rOLX_ zdLVOz@ZO+KfYVVCl$iB)@B*cKG#$2*o!;Bmb_Um+t>?{v)diL0m!tLFuNbq+OygIk7~48F2yXf~*40$M=r#;W=3?@IX!hK8P%{ zl;YiCsKTIp|6304w_9hk*1D3NH63yWF7Z9uXJ1q7-pHvN~_p{z{TEeX_Cav3&n5A=!eBZ1FetYqG2I-4I# zZfIxyF`2R_zT<8^poCSgcmLdJzla+1)l%pj`4pGh@?ArMikboUUN1c?C}?wT|5=J_ zZ2uc;wLFy~&K?$R66Y2aDBn?y_4h;RIH}L0;b2~t7Z!m+vu~iKhVo-cU|kBIkm)r5 zw{-(Qk-X2+$D z$PMrN$m9YR%5vei73=ZRJT0UHHGL#vdsk;;Xad+zi8ckISB_01ZXJWZpD(IGNE3xF z4x8uCX}l3u#}9DRSjTkC8%>Db^q;H>z@eq4KzIfK0QeM$iFe&+NSd_R-~4p{FY6Ol z8p4i}Ngps$6Yo{6KM21@kvDqEp{xj?7IP~4VMFPng~qyignb1uNz^|lj&}!OWedF@ zaE>I4W3(#_P!sdP{~UP^0B{WaXn&u8&Tf%5i-eVp76#-Im}j>0>u+!`@(Z)MpEOSL;dcVwknEh`d-k!nk7^ppMEuChf;rv31vFD#vgb7Ve zaCc;^E0!{ThB~>!$E=Y6yPyK>r1AAj=nX*T~V7GS$j;~4JC%0Ax)JA(G|Q{ zMH{=#qOaI#f{&MyEuv`;n=-pnNw)t+hv0OrE9)^CxRZrR;?1x%_9I+>4J!JXEXe@N zbKWJ2Dvg3*t;7B5wH(E$Y3o5zJ^Yp^Hc3!!)z>ctYj2=7t|C>m7zfn~hn9O^G=ruT z$+vJ=jmzVkumD4Xt9IR3`_;g&g^0QiiT&rjH9Q`dJZHq*o60HQz(9mG?w4IVqbhcS z0RIjzlZ-&m{j&ZRhA7m+;ODqAhPH@)eZFWWr)!=Crwv3A#*9N?o{qAI*-Vkzs;=Gk z?fy>hP;%5IrrqltD59{W`8%U;`l1D-(RCI=MhzBr94xKOE>~w(bH}-3#xIGwJPlaP zwxBQaonl-!Dt$QJIdhdS(Q(z9Z1{Q{Sn#R3JI%F7{MI_0lvx>Jd6EL{t8k7xrxgXL z?P=+dC##g9iW1rlcOquUIG(O5)58xct6l`f_gURq^b(9tTg+#L@ID?QSJsYoQKTqr z!mNV0}RGc17SN@GhgS|S6193*0K?* zEOELUhvrhD6y*Ceah zBi^qW^m0iQBPhGjSAH(t++RHhK8z6pTr=;7mho0&q7ecWf0quXPMajF>1@p3$bV%# z;Bn0d018xs?;i0+ysNzfmj453^L`I`hd_7D_}Jdn9)sfl4@4F5Z>9Bc`Zducug$WPSX%e;6h$;Ag*vp^ko0Wd9A+a~ zpR#Seklh=;iyGVI0(P`#q*y>YhH?bs^1n5T0Y6UD0D&dRswF=R+uR=&niovT+90coz>%^8VB8@1z>(j@evqsqovVCsmT?#* z1E>VBHktb^R+w{6WrUIwr1LpAXZz0g0VwA`8LFV`4gjak0<10Exd0h3U@!bXK;zdR zNAF=xXy?@PNptjd<@!o+3^EuJ6L)#oExrZ-Pch+H;%T@%XJTQqPmur# z(?}#=WL*xZM?LlnyxOMdF16Yq>Gla#uQuMdlRp3Q291GYI62~W%CeVo_%Xk2{^?x) zj_lPMiOu2Cg;nJ;#W|q=Oh;7o(SGeh$5Rz^-b`vO89afnXz~C&7ZhO~QKs*HZGUaR z(g_8et-u7smoi?f`Rw@sg7NyYdyTmM03qEer>^HKO^9~$h5&!uz^_kxNHCDtGz$U# zL9l_p?vZOFDM(f;>}P&flTyrz4t`A({{6Z)L{=@P@$ct9p23jSP*$EGgB*tipyIC8hpwM}#* zLiOv2ZmX1jIbMF@!0mZ2V?@~Yf0~{$r%;T&w=^x^4Q?3)Ug}11&&(mxstzL}Z1&L- zs3260q8>yG-wm;|1#i|XZO*Z@;(S*Yy!P8L0PzpQoJ^f!>MshQ%Yvv68-z&!S#yS7 z?}7Jp04b8&oFfEq&6!kZPJ2b($x(;UzY}bSh}wxH0^9}DilQZ;5Uokv+_hb7cKK_? z>q@EXOd7IBD}7znNkr=QGYA2DYS#1@StrKNzxYyGURF0m;hnweKopf%GYi_sa4xt# zk{jnS5A^Qu_8e@J!5~h1<*F;%T7xML#F@3T*GR37Zf*oatw3Z96{5LG@tcBLmKgct zDnGqkpSqLqQplq^_cRZ%oFlf!T*jo~U1mf)V-gZ}+J{g@yOr9UW3*77JiML?6t~!z zY9u(v)FM>mB)p<|16#VM}II zpS4BE7;u(GfCLy^44I7E9X(}Iq;Xgp<8@RXJ$gk8mYAjF&`71{l1z4d_U3EZ4WWk9wV!0C~{##mB_p%S8kFa)$y)BYnJySow(ps_W(0hds%)M}q_&5HzI=YU6`^6prIADwV z{#qE%_YNTvVf0OIbetdc>g`l8uTJXV3wviAShKCw=~QZSI{eZ{n?}u>FR7N1KUz5h89?Z~6Msmplbs_w zL$`sgT7axF{SG~~|IlpQTCoJ7(<{S1m;8+EtZL{X_AjUNn`xE#;U@b#)QRDI)S$OK z>Ew*0g|@ai>r8r8F_I8HBDG6)72`70W}w4Jm+Tap)R--s0nk>|NsX4N6WZTar**#V z^6WgdQP;u!7nul1Tp`~*iF!|6Ju_Hcq%n7jbFteiff=E`5Zl#GPNzwJhyLcAu`*cw z6Y!+FnzN{ED;-%XA8XTQq{B-cYNIh&y~m-om<%dJ%3vhqqek95_6A-Gs=MZUB|6R)HB}$=X>e1yx zcR{7$Pol0C@EU;ql2}g?J{c#e63Ju~Br8U2$yt+7X4}uj>l7X6#jxX7r^k1u_4yRtMy&ZXR30nSC{8{$Zre5n< z@Ud^z{Q0x0pvm?!A=ZFY|yavA~w05{4T^)L8V_t4R+PX zdSK;l_PP^ftw3o9GwftJO{5Ql=mLJ`l+!tg?EJ?YZ%vjy#<*HX{KBxr+GN34k!~Jv zG2+p3;$@_t^M+kRi5^MVn!J@@6){4LN^+A4xinhl9DIVq^uEgQARA}FvmFiG+UGnd z$i^2(FQJc4va%yO4OxeC_lYX?)aEk1l&ldP39t@q0zDxv^DZo(*KP-R0WNMGYP#%J z+*Ze6IaELWduNgngAk1aGRPoa7*#ny)<&IJj6%nw7kQju&=PWoZbUEQfVazRDtb<^ z>e2L;*ru^1!c}@mSz-X!JW&Ob;uE;y^ynFJQ__0!UpuFy#p$ZgOh4^j;mO)Rn+I5D^4Okg*lEX(t4VT~z%}Si8 z0D4Xdu%)5Wo-MijaLHWEJu_S<|1U*Sd-9>yVemaoGxnL)DdtxM7h1X)>G`1f^&)9) z^snDTILQ`bN##<8B!EE5Hb!&m$Xxk7Z)yN;a=1>kK|=z=($I#&h~vQreJ!(XlNRV_ zhRcv4yl@hhJb6m#y&JjA%Q!di;6%)ma(%KrP;pbbm$F()nDMuRoi(fw!?CEOLXb3PZY0x;&1*`Ero4Se z=LvLWXN|LU=RxNa?9PZP?D9?zOb?4ELv$^VFOwUB&d%`!Nf!#7XKmG?QxC@I6DShl za(mJM=7+?w7(pw)XvtRf#?t zI2ZWSA-LO$jv@`paZ;D5k9I+tPdcqU>hZsJgmH`q;|Oz$hjd(z179ywrMmL<%r-C` zUEp5=^%w`~So@vPMmlzA9jTM9(x3a7(PKS`=lIz3xvKGVl~cm?b)39pp$C(zI8<@s zHHARJHDJT0q6k+^vey;@(;z=ny=7VeV&wcQX|k$=%bU5q`y}Huo}lR_Yc;fu!}zC7 zm?hA9nePCfiPHDO3d1=g^b1iDDqkrZ{5U>bG(OAWVB~f}+3GP_X0B&; z>k%awSou7kr2q3?R6`a&*#0cNICdVCoOF(W*)k@Kx^kj zRe?Ni)&0s07zgUlFKf}~qaJ>NqURDb!#|BXvPS{}_53&AQbqXg_750^hUEU(8t#yn z=x?ptN;7&Z{+e5|^s;UpYp)Y@wk47mh9|W49XYDk15cp~xj(j+ zHvJ8O>h;~s8i#-3UwCObY`rCMz1dG?-xvNdtEl_4&(Dego40JQ71?}iD!Jz?xG$te z=;QliUUt1+iJO@K>vY7)q(OdD=DChM$d(Hv2zg@-c{^3uaK4+6#zixOf=R zmBQ7P9X75)Xs7rkRA23q5LO01gs=`?JLf~#;t(x!XVlqRg+a$dUfC=TjBVQ-nKDG= zae^}1+tKQ*z6<vJB8NH;O;6?2V03Tw=U~6xaJKjhDq52e zwR9y$zULFYXhvSa!*B&D{90bDtqD*k=dFNZw{KrSv5?p#YP#x$A6hqwYmJ9+F2SOs zxGm;6n&BN#4w0C#A+Z%mGaXt&uuY*BkMPGj`1rK^uph7=YT7yyXBsDEjB`P&f7Uus zD~JY03CIx%LJ;{z2@Db%9FrTQAu&Q>`GqaMgf>7hwkT8ON{wWBAu;$S35*aLA~HFm zFpULc^}^|ip6Ey{`Czq|5F{U7kSN+q5(7qq-beZ^^+Xl5lL0!c-1gfd_Q*cIN^ck~ zf?h5K%75441h5bcxsGoX9jO8%Sx^5p2*=d&XRxriyesUgBhbBeco^QPKcvuj5 zapv@>HvJ-tVa6w28Czf-5+sLaMea3f1B;C&m!9eM7&wAnvm8&S|B-GsZdZMNG1zaE zug%O0(KzepO#4tT%&e3Z51l>d7@a!(?8Dq`j?P<*e+XzHLj9}9B|i*e{bOXKRn$>V zmr)ODW7A4KH=f`J*SWzPwzvD6ox5((imSwG_4G-ffWTEnQ4#{6?e5fKJ2a6Vu-x($ z*<%s7ENKoY3 z@b#Z7L6IA}pbq_TCzwC5?+5iAd!rT`UA>|2zhyd&6tF7|N46lN?{fEEdA+Vo z)-4`zlFz3StNru|uXQS^PNkkXWweX$ZIxx||S?_Q%y+q^~U4D!Wl zX~PAro%DgM8JVQhAG;Hkyt4!7yYl!`2Ti1Up%$+zqi;nE2MWS{vVUw z$6GYY+ZJt)qJ`oGlY61RK*xu^)oi^&y)~6SomPPlXk2Jq?RM|bKQ^De(5iH-bgWsr zaCDoJt+xhfaAo`~SiAyJ>`@ap(_u61X~no7xm-;k{Uj7-!}oKFH*ahqG*o$#xmJ>E z6Vg;LS)Bi{X-LS|j1QCet*DFG6_0dx>~)xa;B5SHd_bh*(a+H4-h(4_lCbL%g|2B3$zBPr`=&LE(2O5OwOCx z6>2@QuLW>UrNsba4*-d!2VTA*f!#w3C305kiHB`tJ1Ck;d>wQ4h7IM=r;U%sLfUO7 z$g^Jre1AMBfK_wTY;?vcb0dZBLZgGe0*WJ|aR8WUp_6+VQ(6?e}eYaSVcN z^@FS%CHiL!bhB%2Q$X>>{hd?Z5?I;g?e)d$26esuoeH`MeDNjpB1;HR>lh4w=9E;J zdx4BQiwN=J`=1FKFc4aL0c9-L1y&kFdO=e#kpOdzcKg=mpYtbV2}1^$9tYHzPTH$V zEvR|5T=kltpR8bNRZzBc>SiycCI z<~*T35uKAK{Vi(lLyg)LCNdjL>z0nd5S9Q9D8$JD@ePLnsiacxr@4I)Q)j%sO{3zj zSBY60ZzxV6PfCYb?uxD;%#5aunfr{*r%$txu^PpM-1jsg7d`%an*{}ZeMKygaLfL- zO@yGT79!#KB_ozZdpnMlCQznK6HvbX(?ut&-R{OWyi=K!1eMZ3#*aqa98L`IflsyS zP{d^*?x8z%2yjW8YeqH>;k5!K<*Y}&Q(2PWZ6m74p(A~x=Sh;Sl4A=T6nRXHqSKgL z>V?$OHtq$w`5=5;DrPfVC@-!pQLLwAlNe)e>PjwCvEQ`f=!`1dIYb_8LoVw9Ti$O1 zht$;E9B<5CXGN$Gvfbe%?^v+R*B@fJv64sf498l~NtLx8zEXBZlcHz}Oxgf$4y8YZ ze_O}iYp-+gZ54DCtW&9Ro;oKNo7@b>6^S4gi_RUC-!7Va5`Gsadqpkj#e?#ZcBq?m z*kTf2VsrQX|5B8;N=NYZ@A$2q(h9oHb`hQRj7uf47*XDkJTpAWKeZ@rQ8`L(#`L!V z_!CEoWhImC3di|L{#s;CD>g^S--$K9Oo{^5P@j=&%H~fZ=HZ%}Sq~ItLJ3Wxh~Y)Cd-pF$)r7w}x}dnXO} zj+_gJt)~cw>qZ;3Rp`EGWWi6sYx%>l@VDU054d%^?V*z878YqdIy^jDL|jT9umNX3 zCbA>8`E3a~qHx7OkOnriOhp5PRJ71afOR-UU~Y&Zh}bo976s0b_g{?W<|_$+PaqC# zp_#yz`@pRncdAbcfeae#-OH~>$WRu7nR3W33Q(U!;2cstqA~YtMF3tDrZ^k`>p4Xf z3Iu=)yn00|b0NL_VpM-qqc9xIklJl;^O3)hy%UJM* zXYwh($Z4NPbyc`y*&z9IW}fNj8Y|K=f(0kfE)0=-^GV`OZv;Y=&0FfhnMF$phpMxr z8YJ#d5Z>^uG}N9yb!j? zSuzH1&S;&J_47=0=fF=S|4a?~b$bd|62{}wE7#w*;NEM!D#=YYb{6$fb=SoJm-Z-%vSztTc^uqpelx>HNExyBfR_z*^fR{C~F6ZM( zwR0an?iL%Y{Yz_(Y#jJG?B2l0X<8d2cpnwd#f`gw4K1?;`$S811-uMFx_UP!+BVuo zp*pac{<%w$y3M9+XnLI9uW_$dez-?~YlMRpR4C_o^b zUTu+Vk$vBudXeSV0*RY@r#{Lr-(!JWLboWML45cxk#;qZfWRbFXhFJOWNgT`jo`6b z|84Msz%ijCB3BSzZ^Wp+m=I6$yW0IKw!$mddXpfw_!mOt#s9gYcUC)j)LGO{Foid4tzbr<5(Vd3@~dt-1qMs@TWBVsW`6 zIp7rY>r=lRxg}?lA=P;1+5jY%_-8+5XtZ zOxg_MqOIy(-sEL)6oS4Q2x%;4Ofvq>WGSzfD$;M-DavB5cNt&j+3=l&02`G)qY%5C zmTz;~FEX{2b6}t7k_D(6ty%@0EE~4KSu7ns&k_}|AWg%SQ03xL7ab(}>vX`P{D_?= zpv~y8<~0pxSz;idl|GHF|B_y30TPE-xD9NS#xhn_x|%U%QEFkSxlMG5=n}*&j8BXZ z9jMVSWCRFII9`2TNzmIS;`|;RI4JZNF=aqVj|BvlMui-f(&_^O4R$}KfS>2uNj1kI zg$IcWljI;x1{n)8n_^(I%ua^`lm;61tVQ1(J~L;qd_RhYOa`Vk7P9WqpuDKdkCh(ypwJBTaV{(t>9siq? zJK>fWB6v@pji|GX_4?^JxqneZZPDDz#}XQ~P~Duy;)7CajgA0TDErG7{C+umzb$AYvi!+-d0W|5w~1;uj77tA)l{$52nqoMT&Y>Qr`m*2 zgwyp}x1Dx=CAsbSJG=O2OHXdx;U>4I4>8UCMY7^9ww-)NKVPn6R<%opd$g<*7|RXX z*T(vil}KR6?fVU(cl~Oc#2%wQ5cjj#ezz5rMaaPj*_)`qs*Rc0_RXEncuD`5!+kS6 z3kRR<@M+zjbT`TTMBH%PjZ|*MDP~aIU6_rKC_?0RFH&z8sU2Cm0&wIqryOsG#{+~A zwvqAagW?c3;H|cw5SUq5MFbI&5%rjkD*W0W<>Ytmf+RBB@NCGNpG6#Ae&QiE#5yhF zsh%HeNx%h>NcUi)s4JfO3U^srFw46|oDj8#-4~)KhY~+p98XvX(v-* zK^)>fMpQ?+7a(9NUTV@|4gu{bTk48p#`Jey1h#Zr_7UkkjMm4P?LS{EMgw(o&Y(HC zoj`6ro;vq^bbA$9AHzEhFQ%_B#M>teeM8gUJ)#Y?z1gPI?)oqhPN@VxW!l!R3*dRBbsC)B*7hE{U|ai>xb1J&IO{uSn|3WF|*Eb`AN zMbI-~rXVcdDf}{7I><%i0QV3N{IKOkVF_|Z49)v7h<}9}Krc)FANoU4s%e1~AUywJ z#-a82GItg1;KtjYC07?P_x~*AI)-qC=zSq5d1g-E-d8veH5DJ0{b!y3VWoWR69Iq2 zs(e(%QeP@zJ8)D0y@P?c@OlALb_&IHFKz>{8h=CLo}j9+N_+TbcCD(G)3^ZR8ug0d z8C80dixM(Yli!c_{0PaGZnNF*^Tvmsz-l<;%UG^)2+}DSUx>G_gD_Jg!*de$p7YwN z$!y%7!(b$#^>I3F+hje9WHZ*og2ueKxHb$OW1kzEegW>>T|A2hc*Kb%RPEp9g9%&Q1pnih$|=I6a%1@I0p5TWGQ92T%P0TdE4x@_Ns1 zfLQEzF1O6wDBrH%Drs|+*2&GJcL{G3{gpq~L7Tw^*1r#>|g>puSwh;zN*V z)c_Rv4|(DsLRxM&;pJb9Z9u>}j!sj#K#2eUTC?iA>6>l$d)%tHZwrS8)=O zF)88C&FzO#tjThyMu)5MYIuI6F&BrO_vK;xT2MTdh3Y^KxT;YXhXlnurfKHus{i7S zP5!~GKmN+*5t5V-8#*L(^LoRZ9pxQ!c z&67_VazHW;KgBLkWxe+Ct_*{33S7-LPHg{L+UK0Qx`a!8=*#D8cY*Xb<(ThR8Vj#t zq`*qG!YUzg94^1w1L2wNKOlHTNhh9phMm%x`EJb~I!#ZVDG+xSQKX!h#&XUQO+~V|1$<7DU5gh50`RPXfoUVBC#_Jqscz@p!(d*j*{z2=ol5HOxV^o{az+g-34n4jBqGgREVph;Zc0zt&mO zYjv2c6?Y$ZPaS*jkXbDfEElYMHl+R@`PbsfJOcq4Uae%FCj*S~n~IzBZVSj{ynoH} zTku+bZ(;A6ZLsd01g}NRy%-vT5|8Pr(S^K;k9pCzH71ZX)2j=xhXwAv$WDT`9v-c- zNd}np`oO6V;1B3d5&cDJC4qWQ`y6zYdgIe5n;C7MPX{~G7-MG2_`&1`Y@$aypIZs( zXTjyd3SApAn`#BXzaVq*>?Oj1QzTy1ttZP1da=@M`={^tPv<)mJuTmTV<3?)v?P!@ zLdcb_@R#^lqp0v9V-%_HEL+~HQ9k=@n@F83imlzZHn!08h#f}@i*)aL=GU;@94rU1 ztkU_Xp+Q2`L4(UyHcnw@Ws{)oLhcfllApKh9&)2SL1btx-d_PH_2|ZY7lg;wT>Z$wF%e-G>Q7P&6u+E%Xh~3 zg{td-r?GmoUZ)7z8Vmaamuzmb3E;bc@x3MM@Yh2{!o)YmQA*j()btWf&g$zsRe~G# z#b+YBh>r_fgqNUuA438bb0`DiwD7W9N=b-tSDuQ$??E${r@{l_ zin4BHP$?+(GbwBIL9{M(iNrk$Eb>zXt4@m#p%vhUoAOu+7SG7c^hnV?S?m)pP|I_q zM(|+q5prHpI)7GEZq2%J5Mrj5Jpp=XaOj_AxEA%~42=fIke&N*r}v#63k zjJ{5B>OU%;O3&C|6hIWxg>UpZWGrbrHjE}>k2jc9k78%*IDWbh4K!`B;?+B3Ju@sWBuhHjWHDZmtO{> z%ZPms^s!*}SfJ3UyoA=GbaXDL>6pjf;r=c84#1!5DZYUIMz8c%Xni$%LkpGHbh&q) zLS=UPRIspO{-45*gYWZfVt=A;48j;Mx!lBZsjhBYAXkw7_w4}v zV$)@IJU659>8c2PsSOrMfEL0~hhry2ddiMrR5w;FAKEReVLw#zy%7(fFHh=b@8t!0 zg0jKT&{vH07UPim-LPC7tPLr;cC-|xU%G)ZM*O1P1X^1*E+#%WhvxZ{9k_{1{A|)9 zo!37R+7l%A?9;$mjA%cPei3K-20Yl)kZ+~a=}EhK_|sM7iDHrz_BuxGnFKWbHVM8B zNIb-NFj*;PV`8(=&G*M_f3pzB9gzU3nTB78Q+*|I>BMIj7#%l?X-8P3aHFEl@x5VL z@cpKy?JG!QTE|oIct&kIO80QTH#+7QGx4hL~?o#u? z$b3u$I%-!EPr!+Ux%1`RxP$%I?g`8gS8}ghw~t_x=;0iI9|c<;hk&X1U(i~wI7XxE3o zNG;QQhR;(p+#c*+w=~<)B&3LDol3MjT?W^j$A!QvWR=DjL%_#60%N|9_xhQaapbyc zKMi`%#o%YG?ZF7sC-?c~N;T*AhlhNfA$s=XE*v<4cNt}%d{|s^I-RZ$uO=O_Uv803 z@g9DEykhNRbC%#47i3Y9YFkZQ0FNwb7P{YXQzmR=$3)RcEvAek%V}1U{XoT0V1sKH zQ1e_i%@J3*0;$UdtG+9UF-S4`;lW$`Y6wKYT%p(eflw;N|C_z5FW#`({wfUBc3BR= z0~S(VpX2)9Lzjts>R<%M0LLsxm5W5RXve086DF0K!V$+xL79}3FcA463PSz!Q|=RG z2hav)M0u85a{)r)jUWB&p^0c%MacdG<;fsUDVa&UbKr;yjZUOQ2E#;o${k_X=>oq`)fuXY(H(m z*QlN7MklY(3qbH=x0}HXkGN1tJ%w#5U9gyOLhvT^`sbw?BzEI4c3~8AWndl(fKa_$ zPflm`kS=4{s6#;dlJa%F-M5_A%~H^u)9bWDSDTP7?wDLO^VhjZ3~4%n;`ctZmu)z1 z7K2vB3Qhz@g$r!ZrRlU5Z#5&{970(@(D!8Y) zC&lVQb@Z-~MTmcpwI{g>6ztr*Mp#mzZC^K0tNCf@Ws%7usg4AE?dlQKV)QaZWGPd( z@YWZ>QJ22dyZ!T5KnI3u!uWRn({wOez>dTSvN3|8u|DMXq@72(2I&KoI?p4?m<*N) zP%|OlDnZf*LsSmZpb6tpLP12LAVgs(Vz6in>lR{woC6}h;rd!nApC(-c^rO2lf-_7 zI66vPoRkz989JDB1!9RBa5aDzR_LUeLq{Bvujj*>bS8|R@Cg+bBqsU?jk$IE5)H^P zBaC_Esd>+Rw0ZY`*YFdbQV{wCD3a$%tnT|2J=jlJ(2f@11FpXdXdFg}07RL$_@UGI z`{c&eBE-Llh&i~lLd0oBNQIcM`*i5LHuudi$f5{S+^uGqx7L$E8fjGZdwZduV{+V{ zBIgNWJUp^#sr5Cu+{)X8jrhfujQNUOjHw=W)WS3_VucYDavxYNr=?9GWeXR~v9B|L2W1C&HhX^3gb zW)d+7Uw%P4N55Fsz$d8hk%sLe2RsALN~U35^@y*$h^*K$#>Kx)P)D7wGA&va5Qra5wLZbFoTv&7&<4Zs zf$j2N7qlj7QUXu3`|Z@YR92*eCWb5vnrD|{RQoRoniDl3YC7U-rW{U*05mhK<{|6# zV*8!u$Jeh?fdr_z4;X)nAZj_{X>tCp`t7Pi_c;rbk5V*}{_iA+S{errB7oVLVYLn$ zAW53)FtoBrea(xe`j{3julv%xcZ|kU)kXt=Fkxro6rF~C*{uC~x>`iwKEet~ZWhIx zHHe(b1aIJ8mQ;~5r^Lj{JXQ>WDDAF*}S-j9_@h9YOdHKbusE>orAL^XFQR9%LqjV?S{MdykW zs-uqxJ-AKX%OHxq>D;rNk5oviH^ZkQ5z_rt{ISc5k zl1$`fG|=lprVnzU?7Hf2m@i1)s;{1Fo7!AzqHrC602g<4n>@#BymjK>7mNV1^%Sbi zn(FN~FLsE?@PTs1Z7*TakR^rn6xPed#{Bz$Oh?*@3?iQ+A2;i)!On`z$OPr;FVV$n z{(1Y2Rf4c!u6PFoNehcvDsl#@e6u3wGyILwqT-n$+cROr6Icgrvs$fKY?7NMRtF-l zi+Nw`S?3lYdWY<$n9Os9n?(qI7ghN%*<~F!R(gne`MCXW_PzppQan|A1S4DdxVW0@ z$Bv%j5$4!p__*msiuQ8r)o#SP<2x9I_ou%EpY|*1*w&|<_1d>aL0*h@2NVa76Kl&J zGACI;-U;Q=>HX5n&UH!Iy~46PqGA zh`ttgb%@ak1UO4FQ>823H79va8W%F{iXDYwQ>DgU5`Q349zaf|?9@>VAcITN)j`v- zA?>;&{yc-_p{N#7zwNF#Evx?X0W>h|NpzptQjr~U7AE=SEjw&yu@7e&FID<})4r$<`P7CAaktA3K26FNwIN~X{Kb%He9@Jk-vlDU zkd3tveDJ;*)M-DwnmBI?ZBeX7_W=xYZ!6GAHZI>-*%+faZ0h0)(C!!sJe`)DO}a83 zbgfk_%C*^p0wXlV%8+(pT)rFrhG=?_dFgB8aF_Jj!uyAF&K)~r7I8#$IfT%vz@V~| z5tTnI%|CqQcbW(a6foZ4f8t^@*dY`@^{v#p-wt#O?~&%?t$B)Yj@2DfldYNg7|WL1 zDCy1z${B^dnv*a3Fd{;zgWMEg8Pb$=v2mwzB8kZ60k#2ijgWJG2;|F2stlp-;hQ6PU%9qf8q;n;|Cf9(5htaK+<0V4FvdqoeABrTZe&tIHzbK(5WYp{lLo*t`O8h{O z{HAzi!jyb~=)G%fODiF2g6?Q{CCH#nMMD@P4`}^2vwQ#^c@ywF=eAKkx()C=2jjEcitkl!VTzVv z4^_`*8XlLMnWXe0wd(UV&l!K;V=n@+HYQ0s)iu6~LjX%K&OLId-3gSts>NDgCGLLd zqmJ+8Fb&nFqg>k1tc_M)DoTsKcJ)C$Fc(%tyL(^OG#g0p<#|V}uxL1m@pn9%W6gzX zx`$8b_ky1gV#5FT5zk{Iqra3TED|y1yg6Js8?1rb3Ny&i*hK=cTywxD>gk|?0ZNG+ z@5+3cO7j4U+x}>S=w~%hEZ1i>YPYEMp zp6!8K6Tj$M?n~o8yv&OEGZL+BsyCIiOGoV6wJW7x0}0!l{h~tHg{W;v%E*()Y*BUj zjB0p$CnVYkaoO;xg~^pu>@C2Z4qR#n7YvE?_BdFcJP=+uHyYO%i1<@_F1^=R;4KML zHY}ZHhN-&Kwy@eC;64eiCH+^GYW-Mv3vz^&Ws1cDwumY^#)x)J#RmEjaQ6 zDVoQal7T1csGC>e)}vgE*WMk$y~j&>ieHXpPpIJbWe|-Wgh#u0Ki`5t}4sO zoIOp!4Ajf+&sN}?#?pQIE9Ea*;^cRf;tF)ia#K*y$x3#{r{Cp5?manK? z5J|Y)d$&=Fbwyl@KE`MD<1(bxk8w zNdvMPWcA=OVHINPXotP1o3uemgB2u$@}-|~0A)c`(=a9-4^V%xqY5H3MZ7P7l(~2k zvlYim3z4~8(?M5ZgB5ZEi_DEr(f@_dr3=k{00dfa1~ppAXy1b3%NEI1EYb<+{;qEb zu(ovlSz17#g``ZQQJ-GjfkNHEw>uoYn^XgtCOAP@Dkq&}k$+&mev69-ssq8m;RPoO zEm=Iuf_89hyy;Atdq~u0Fikraqtah&XJO!&pH#d^yoWIX!7>udN!s1?=o^*H&S1tA z>9LdrOigw;3LkAJ|FoZkSk8idcuH(G{wJ{M6pM2#UgBF;bT=;WOHZa-%Op7LQ_;4A zNg_d%?0Me_dscFQ2%Zo&CT2!TfoeyC@(}fV53!lXw245;=zff0hEJ53A~lI&mPjF$ zi6NIbgObRM%>#rE!d=;{P$Zm6qU{4o+kHu82#o(i@4yoOlnhHKxkPLpoAA_8W8CgQ zOYj7j_r!Jn|IUdAwer52;0c`iruPvi&}@@Lncu7UvoKd`#1^T4mUY-B<7bI37?_mQ zE|=qRX`71i!@A-z4z}kQd;AVR>vCy9reFQ?*#C-Ei44#=3Xjy7@H-y^1@VGlA|Dc5 zCVZibIuhE)*lz{p@AvvxQ)_p=RuLK=LN12KbnVx3SC>B2ANR9uWraBt>*`V4YEiN$ zuJB)4f@ipVX4-#Mipj_AGC%Tp_OKCu98M+y62o!`)Fc%6N)CA3G|rTl#`?cWdhV|JC)CL2P3JMd+9@9wQzT{U&i>C^qp{Fyx( zn(uP(fJvzKs-I@k6m5c%=OkgBz>$yBPgu^3_?T!-z8K+T?yGc{P*^v);reWoLXW`V zDcI%lOA9WFp|Cw>Ke*GZW%4D9_bRVOMC=V^m=T@E2oVu&p%t#IeNp~(HFsEXtoZ;NiDLX zUPy3Tet89TWh*o;_p|S#n>mTis>LjnL}c3&@D3pnP5a>lI;;kT`*S8+MvjTXO@2h9 z27$j+&jD2gn#yGR2-=N1GIi71?-LmWvjv(6xh8PqgXF4?FR5^x^zDTFEvj7ZXvE7i ze~H84pxB&SIjmN$7jF6o=Y&AUphK)edQBb&rk!EJRKs{KI58+?in+4phFe$;ObF}q z$6?6bST>&9RZr~6V^MB;97+!TI7*fc%j~xPvfT^4797R3TV@W`0_A&byQ{_0^mXBi z#5?U~{ySGE`@)B78v@F0@--wW0d$<-HwRiTZGEJjQdfF8_9!we-lj6Az7g(eUNV=g zL~RFv&ffYf!BC-YBH1UrN23Wlvy-Mn<|-QKL;}a!0b~5}(YV&!>2EK=uf}Q{4XQsS zxd+{$1sY28y8q-(x<{0IMr{((-U%=}tZAR@OvqLGO>b`p4$~e^pLjO$3&-oj5lD#mqJOvR|g%@_u1T zfGLy~&7Mv$-XrH)e{aZDx=jPnd$@P$9JA4{Hm|P}>>Sgf*65`aw$ZZIe=u~FVpAIw z_(BEB1=}LsBM)YW=8)u{3BlJWJ-zpj`0dX%nx#!k!WtMp@KS(4--P0l!>~aUI_NGz zJt8Zcw5PlDqbW^4=EOzloZz$w^4whE_uneIzs-kS#r)}<;Lk=B_2BCuRGg*oR7PTu zB|!$)3_gaS4J5hB*@gv&TLxdAJkr7=w&+kJQse%{ z9&;w~!W~|aKNK~G7FvJg`v`+GUpv+ZFD!&}57x>qhyc!w^7C$cUc{5zc zG2#{3ItX5%qGWg^uB>(!#`>lRN&1xC8vfb?tfX);oEvas=kNprQbrn}eLXJS4eu=V z)(FHuiHHw4L6u|)^Yr{qG9j4|Df0y0@M*b&4zLN;S7b8rv%D%xj?*$lS=^=U&kKiS z>PSSadxq3d+VHmjMoM)kPfXWE`V*wT%#>WApX3t?OE``=0@>go>P~dH2vD$@=Sc?w zt?EZ+)S!3e&4T|i+L;UrDAXx;s{?^aQ))4=>`i$u!ehy>o{$2TvqP~QyID~H`)ZuH zyb1p~piu0ca$28%7Yh&gTJCn51%Ye4^F%lYgCC683uPSiO=NkQ+r|a^`fdX!#{a|+B? zW`EL}tZ)`l#C-E?CFeGbd5WK=XaiqM{NFgit+jT_#5zhSky{5(?i39*tbgw`qQ*r6JJ z(Nsh+Em{YW&uy<#9{?UYyr#Bi#|6d=yb)RVIR`K3AYkQOCYcv~NpqBXzrW6zEE+5x zJ%I^v7%Yh5hg!UX4oOUFY~krK9?OB@BbuIuArF-)I{Hd?T=~x?)Y)=*&wCv9*VQ zm=LIkkEl2>2oO+U-~jI*CT91!aNUx zAkX1J#*Bhv4HepEo2i400K9^=72Y!qV*la2)0+zg=NKdu3E1{qRF6_SEczP)^`rXl zJmZ_1)T7j#fcBnurod}Ugy!rAw~z(&#AM#R?hPdHw%&z>cc{4zp;RMv7Dc4iNNSlR z??V6jRBFjW1O+nOt0HZ^Yvx9CSHgP2DzDAm7+N7V)d$Zp5Y;l7_e`AeZJjHK;rV(V z(=7n~^!|5Ch0lp41(^X{9HszqOlhfK#1Glh<1?KOgPJIAkw)NXOpW%p(U_mz9jZk! z=019mJthK zE)0K@Q+!MQZ*)3t`AEKF{cgpnRM-o<0m2JL_YVBc6|w&^Df~5UAck;-+z+nVQ+Lq( z4K9X`Olyocd#AycezX1fci4+aVveJ$LMlXug#62rzp>tOd9LK43XG+TZqYZpBNrB& z{olBVfxln`_T~z4e-Wg@SjSj*;mE&SSKspKYP?)L4KpJ2N|LiV{H@hE?T3HQqetED z+^39tn7iJs;rX@5+7(+WSUOAOoI+Cm$?UZoKoMSo78n}^Lbx01jT*`$8Qbban%6|9 zv;myAoS} zxUGV*_{`8JEA2uB)dKC$VrM;gQcD2-Q!m+l@)tpRDx0$U?-Phiu3VEF71-8BD`mPAy;HgO^*~V%969S^tE=G*I z-1z(g9P~ZCWr*iqPo7kj-{`E%{X8kdojtapxRFDiM05292HT<8*ViVGg!SlPeRQAn zAlj_!MUOre@3$JLD$~OLLU&DpP+imVhtt4V3R&Y8N|8jKhcn*PCEdgovQr4cWCg+X zM-YZ00iT}$u`mSzQlo}N#`azii-mRMcuR8l>FW@BVJs5(c*tR}+xi|JZGRwLuKG1z z^R35I0*m_t{B&sOza1P6!m9Fi!(gTX``?7upt#$6!Ls|TbCU54ia@<~;-RF#i$G^V ztnv)-4&O<@MBCLHU_T{A{G06=w9J#mBLPy*&zOyT?9-Jp;Lk-$*LuNDN5v_+x|jx# zDsX!n_R93X$z9f^i{Zlhp||_+09(%vG+*Mw2_HrLFLB~1Ghqdz>m8QED)42ToIk{{ zzeKB4CGb0LVRgp}@&sH|(3PYl80r)kbk|j2k%%%d=GSS)xm)(UyW z_XS*oipAD+h{d!y=X`m-Jf|u65dMuf&oO3M!JO_O1{%?3G*|D%)?;T#nOf+APe?KW zH!LpEF*~g=;Uq{Zrvo5a_vE6Eluup6075Nb|O`;shi}RX-HM zfelFhaRsb6=7|A`4JFe2`QN!$ErRf+{MV7o$hy8;Lji6*|bu1)s~oj1w$P;S3e!G z&o?unUp1yWrju^SKUS}T+8s|1o)qVaK>EVwm3h*;Tgi&a0+n)WmaQXo=~Ko-A>KiM zV|Sfu^s|s-pk)Vd25~oI+0Wo34AI)M6zhT#`RY>9%2QYhFh!N?$XtcUp9O9))j=&A1ukeA} z(ao-(4n?7$%b~cYI#lK+XuE|0H-l*gHm}JVhGoDf^RLo4m1|ATjGE%EYqdFu=fXNo>;4>7Egr-R__d{wi#?USv{O~NR_AtX=S`R=P0d(DJD=bgRD|%)obF_-h)bHBE`BALV(k(gv-Zx7`)j7||})y^TLQC$%!et^F_yXBE;YszX{G zufZuzhsfbe21F~8bgoZw2^w%ZM;gT`j6fqD?f-3*p(A{!_59WIc0m$4Ti$f|PNQ#? zZ;xREQ)Y#Yvy|_YE-$4`zfFp+;asi*uA zQ7txf7@I0B$8YN78)RE$-Sjeza^YQvb7Am=*0wiY;~CMxh2&e|xpid-2_;_-`PbvK z#61e&aGJ2CaAC+3s{0NmFY=xDpA)yzSnH{%Yl_6niffWS(^rL+!YBex)M@mrJ3FfZ zzaNBiYDqg(l@j z%=-uze;H4eCfc7I>oc9gufks8l9V{p5JN1~5fMKXs=EcTNLtDstI8Wjtv|&()Jq9` zX#D5XBxQ8DvvCmNRhyRlJ=$T=Lt03_Ss%@4s463VjMK4>Y?x-*9ZTlIe3zWif}nnL z*by;UdVnM;R)E^mO?AOa>EE!=UJ-i38?a4=&AMv8=KpY2 zezE@2CZN2d`Zt;tceob&BJgtY{F?B(?cyk{fd(&%{}{MF?zoo$$^q^oJ4@$s;xURb ziWa1AruTT>w|r*g5ajryVENqk@}&&mzBk__Bgk9st(kh>st}mPiS}gc5#q#LR8?AK zSXw2dqp-MvfRZhSJce6|wqXpK9)5dHQZpX7UR5*H#%7Qoj$ak5CXrEwC4Dg%Hp3{u z<{`$i`#ja%KDmqxhRFK4DAiyGRNILSS_$(BT#;sxq}C3%ss7kF6aUSQ9w+ha85vj+10OF zf1%g$c$*Y${ME?7Lh*IZ8N7e&b*lrziY&luNB)+D@Wu~A^VNqvP+{gH5$6LO06lfy z^1O<6@5Iof?r89$R&c+tR$2eb=%sC~f-fCnZZ%r zPq8DGY>>v`6wf^#(E?B`4e<_sHbnfXMIl70;CUdpkwTI6xQ-{0LZN^`gMbZ$nBffM zCKrh(aZ4aMz5^u_Ipq{lyLuu8Dt|2nwh)dAfg~cQWFqrt4D)V*+%@4SL#E+o&%whe zrZB7!xPb^Fp_aH}?2A?5Kvb4OdVI%7CU(j#rglw7CKgZn*H&PQFhdHYZt9v&!;zbp968{!zuq?PEe{%n9&Qbfx&dVp6`~uImN_a>FyHc>`%zAVl9LO|jwI7Gzj9-1hyg4li3GW9?Kt2f z5|xA+G3qk1Lto!lKof9`RK`F_O2~1FZf9!azKg-+;$Az-&LJ6K($j}G!2SlmXv_J) zY$Ro>eePy&mz?2VChvIsbd)M+)k4`hhSFvvq_L&{%#~u+mmikDHNqI=?CU%cDLQ>^ z4-bv{*Dx&*MCu^QhyzeOoEWQ-&335WlkvQ*6nAuYoJX@7T?BcNm7G>MZ6K4G2=z4G*9k-O-u7eV>;gV6=d@eiQIS_ zy%!LusRubzj1Ph14> z9tctHjQX)e_<8)sCAU^O*NSlc)Ysx+Fnw5Oz1%31&ur&s<)!}Jne2rWtIBNwr)@U8 zP}ktLT>cr;3yxo`xv+;eB`g1SeMYHe%##Y(dYZcCZ$44o&tLF{+frnQN-9BF}GJ1XvK+@P{xWe*`>X29`vrlLyn$m@S7b=^G^miHBe5%*aq8cSu z$ZQk7OX(Id%r5R&oy5NjO^Z{}jRP@+nwH^`<7DQj4dL4T^q;U;PG-I?=+tro(Pu*D zN~QO+eCjTd_ZoU>TcY#?ftN)x8S6WT@E-dK03cmrt z;G*hMG2PBqOFB{Jba8(Z(wF)_AhJy9rasgS%s5+q=u$(*bonQ%NNLZI$e&(W6%zBQ zbopaLa}j41FW`o*e$c02Kz4czg{=y9b7dJXj*}c-+>Gullc@c~>1_Xnu6Pa8dp0FM zmU4C_3o)EgZr@=QbIm;nMgP^?jT~7DLw2Mdde~luwTohq@+MOG+nnb~tUnvI2^c4G zUTUkV=TTo}Qmx^i;Yh;~;eP3k>^$p~q1F=B$CobPtXO{Mh&?8qly)_m=KBx+!O3V5 zG>;+iDSExc87sj-_HqJ}=v<@?v5hI_Jj}3|M*Z{UgtQ_JMXUYHU$r#4+2C6~6VkJJ zr<<1d60XI?yE?(&6M&z!$WV>u{y>T(tfktbBX)1yssOCQ@cj@FRsEGWQa5{@T|@a{ zd|f=s36hJ@BpevcfTv9W?|T;E+PRS6ZITp0uY<_ytD} z9k}>p7RG@Ryr*`3=6jrnC_iZqiemVO{v_9-B=Bumz^?SR^jw!`Kl<-g)HfX+s09OU zo)9?%Xh1;`O)-NrL)V@E2mR9T#I#UIkq`w(2+Tk@k&rx5CDJl_1T~${50XH{4o@qu z+XQgN;H~=KPA(C^153j{#OP*K^W_*hoxGSlGV+A1o9<#b!SRG-i7Jqm(kH6A4b$u< ze;A>W&kU*-DxGJ)wFWG8=4tZd;?MB1{w^@&m8hSHPZw1W8}4^HrDtn!NhuPmP0DR1 z|5TvbCKtAE)K9n_JCOa->Hs}ZN8B&M+jTysbyNGTA_yy!c?Ww3!|R1V!G>jUAW-_N z861L%C(4If4*HMX#|yXl6oR4x!eEMtiiwfdgDVw1Xr52XAOc(y*iLG)P`d+Ra3q{+m#~Ua9?( zKG<5HmdqSK!g*NxP2pUpnT(LRL=ef;T^p^RGj}j29FwQwI_Vkc9tkE^)U~i#m}k3^ zmCq#awF;0xRrniS3S9~qTNqpLs9W;CxRircgXaz-$g=*%GYi>sJP?0LlJJ1N4_R%B2`K!H zE_#9WVxm%_4I_c|mdA0OLx6$4z zd6v`(hlHgV&y@GkW8`Cv%7&P->E1^+85S(7Qap?(S~#BrghyU!CN#Pcq2IX|OFP7p zJ0OZ2S;2K#tMt%yc6#MvpR#YXWnm3B3WFNL^dPx-gYb^u^^`k!naMnDzZHhkVoNq|BQG3$crKqg%iV^Ab z9#>dVSK$93iSVlu21V*o`|zs_pQ{6#q&c~e$gKD1n23Cie@90Ig`xfv@yD{E*Z+fjl!ko&# z4>6YL5(72>wt$gXZ-j@#d@Qk@E4SkitDXp85qM6Wq`_FiSd!Eu9;d9a&=!h>dNF2A zLMMs#p{pxo#%WSk_@ynm4Kx3WTyQ(b*TE~!NJOTqyVRywy6(sgyGIn8pXYc~PK{Dp zvlq+yEO@1Q`y;76IX-KD_7|55*EC-o8_Pjd=}c3}Z@FNa#G(%a3XG(@x4lF0?MAOR zla?FLsI+Ro6)zMGyh}ycuYCOvM_4%|lkw4E{7*m1G#9+wr4N2RT*e@5)w0`lk43m- zf-o{}eb`zUSz_|TfQr;f-#rUwT&+OLs8T1Kj4ChJvxfA@5%04p?$~rX0MFXg7YTz} z%vQNhF~{^|MYNoPY+QdhfEUwmffr9W#PCF)4>Xvp0F|q8_w`8)MOyxjf&<=4zxV>t zrzP%>D3Ntee>pj8$|;yeZ%TYM*UrF)qZ&ak6d9}vb#+}_6eqW?8>`!aJf;BQyO}ac znv=f27SD(>q{_8Az`CSn^Eim~L>5DKAx7O$0}!R`wGH%~yx*-{^b(4^Hd%U^t`x0o z-X1wF`g)I5K;E}1s*ok+nv^MD&DLYJ(1YTp z1H$K0M0$$i2E!Lj*o$Obs=drZ@=nKCvB-eH@=k2|-}y;aTgO#eWMZ2bb zSRhe)n463#5i4?f*tpm}={0?Vr_M>41Q1pEduF%|tZ3~YEv}`;{=cmn!1jNVUdIE4u`4(7>#A7K#Gfv_vO9Rf z1s^4rUbw%CL|T&QMNywWNAKOuQk*e2EQmzb)7gn#98T1w>1(xj&?oe*9`Sqg?tIe& zk36)Khl9cU~wRZ9zd|WL8XE<#*0twT5|Bq|OHRMy;+grtw~n zeX5csa)*VP{FwD`C64My26WCq8tku?Q@`;RZ`Py4rb2A2&UT|D@&>CDu`w~P#Zxum z3h;H(qk48?nG)e@;f|8?JN6#4*aNp?f!-#9D-j#<`y?Q3j8r zwsz@sW;JQKcwavhF;e@D>qef| z`*A*n=VkAdha)BEzA3uW>OYI1oqBP8bqhEBMR}?ax`|zfEgDv4OTU?p;yNy}$iRB% zU1+D^2tQqAul6o6y#jB1eR;VGOM)yZrM+kB6wv?3X}djadk>YwV*AfES`{q|-J)p8>9vZ@y9*9YiNIYIm2FA4ch^ zMa8$|nkAYU!QN0YJ;IRndF&>)2_A{Dcg<#`sd>+-q)=~om@NL7V?iBh|NaXdg)QOm z$MdZ%ny{288hVD;dzI$3wMHK|ihSz$E8`kZcbL&BZ*OM1rt1cql3dW1utaDn@FQ`H z9F^Ce4iP~~YWL)*);hBf4Sm)^_0!88GUjC^T^caej9h77XKv5)C(d?vlx;wQu0NTT z*D@cQj}*8z4GX5+5Skg!Bsl2R^cgV3cx4478Tx~07;-L#|KoD)BB}5 zvWY?b+0^5MiTVSP=jE4vg`oJnVTL+tslR;dMg{-TLUz*Svw+=^g=K@~v5HK%r?_Oo zhbxK6!xWcT7Z+1OxNkI&b$X<7V>>3-MTlyu&&ALIGdc2mCaeP1{;NWlb2mCObm5Kx z0jKvQXxVeK>a~;T_Q8SLUC)wQ?8KLav5P+Qq{R087w31oQ(}*vAH}b(v-MxJ z4PKj@xe430I`b$WMC?U%OV>Xza=cniIRBZ;NO7^V{;AU7+&78H$91~;?SH`8ZKlOR zXm+Oj#VWh4-75qiS9@nhb*BCw^dIUscGj(@4eovGh~8YMYv29{oUI84b>iRIadx`T z-53xmesg!$y8p}-_Pjnp>&4aN25)>3`j%gA|6j1uj)?Ns(Z=hdVz*5(sB-+wEz2TQek#*J$@(KahPvfO83#k?TcB@J z)$Zs8{iFk4GMpKC`ZtnfRc>yUf$iSz@2;>-a8vD+w~XUXp834y6?miSB1(w?yy?r& zgJ+?+Tuk|V_yvrGSsA=^e4hBI5uZ&P;FU2-@*8f<=Z9FyG>V7`&Dn4)0#?~d*gDvq zpk^sBZ8u>ix%k3N_8m`eX#I^c4UE4!=z!N8(Q)tm^w2=ze8$-p#aN z<2IUw&w819yJ?_*GQaUmktQ^|4%3->-*?0DcQSPg&as0Y|3DL@L3=n+0TbeYeIDWu zAWjiisr=q$bxMjI?(c{D(nbD&`n&?}>wK{;cKL>3d?ADaiGSE<-`$Y_={|UOiTmUm zg7Ka3-&oc4L;05_Zu8@%ZfOiCu%i31r`GD153nw7a|z*LqSGD(h(7+SVYjOj=7+z{ z1lnQWPB0|j Date: Tue, 3 Nov 2026 13:18:24 +0000 Subject: [PATCH 11/52] Land the Resolution Foundation family with its review applied (767 to 766 rows) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The review approved RF's rows with thirteen mechanical fixes, all applied. The 25 rows for exactly Table 6.1 lines 61 and 62 take the name README rule 2 gives that scope. Restated rows name their sources; two RF statistics are pinned to the averages RF compares them with; six halves' averages carry Figure 8's cash basis; the £358 and 20 per cent rows are England, like the row beside them; and a workbook cell holding a floating- point zero (-7.46e-7 bn, with no bar drawn) is dropped. NOTES loses the passage contradicting README rule 3's corrected example, fixes a hash typo, and records Figure 8's lower rebate bars in v2 and the VAT-element wording as a discrepancy. No printed value changes. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_01LwQq6HuQUQ1obM9JvyhJbE --- .../harvest-uk-ab2026-2026-10-29/COUNTS.json | 14 +- .../PROPOSED_KEYS.md | 22 +- .../uk_rf_ab2026/NOTES.md | 372 ++++++++++++++++++ .../uk_rf_ab2026/claims_staged.jsonl.gz | Bin 0 -> 52874 bytes .../uk_rf_ab2026/manifest.jsonl | 7 + 5 files changed, 409 insertions(+), 6 deletions(-) create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/NOTES.md create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/claims_staged.jsonl.gz create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/manifest.jsonl diff --git a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json index 0ddd4dc4a..13fb60450 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json +++ b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json @@ -56,6 +56,14 @@ "ippr" ] }, + "uk_rf_ab2026": { + "rows": 766, + "primaries": 7, + "claims_sha256": "d4976001ec1985cbdd65ceddfc8768ca85bf5611ee480ac2b7e9fa138420f9c9", + "sources": [ + "resolution_foundation" + ] + }, "uk_thinktank_misc": { "rows": 12, "primaries": 5, @@ -84,9 +92,9 @@ ] }, "_totals": { - "rows": 2328, - "restated": 294, - "families": 10, + "rows": 3094, + "restated": 378, + "families": 11, "pinned_on": "2026-10-29" } } diff --git a/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md b/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md index 3c276e183..950c35200 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md +++ b/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md @@ -4,7 +4,7 @@ Generated by `pipeline/ab2026_proposed_keys.py` from the staged families; don't A row whose scope matches no registered measure has `measure_key: null` and names one of these keys in `conditions.proposed_measure_key` (README rule 2). A key is a measure with no Table 6.1 line (`ab2026__`), a package (`ab2026__package_`) or an option (`ab2026_option__`). Each family's NOTES.md gives the key's scope as Table 6.1 lines or the producer's own words. Tranche 2 registers each key in `data/uk/ab2026_measures.json`, or folds it into a registered key, before the ingest. -49 proposed keys (12 measures, 29 packages, 8 options) on 223 rows. +65 proposed keys (14 measures, 39 packages, 12 options) on 420 rows. | proposed key | kind | rows | families | producers | |---|---|--:|---|---| @@ -12,7 +12,9 @@ A row whose scope matches no registered measure has `measure_key: null` and name | `ab2026__defence_investment_plan` | measure | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__energy_debt_circuit_breaker` | measure | 3 | uk_hmt_redbook | hm_treasury | | `ab2026__english_devolution_agreements` | measure | 1 | uk_hmt_redbook | hm_treasury | -| `ab2026__gas_standing_charge_moved_to_unit_rates` | measure | 3 | uk_ifs_ab2026 | ifs | +| `ab2026__gas_standing_charge_moved_to_unit_rates` | measure | 34 | uk_ifs_ab2026, uk_rf_ab2026 | ifs, resolution_foundation | +| `ab2026__gas_supplier_levy_funded_from_general_taxation` | measure | 1 | uk_rf_ab2026 | resolution_foundation | +| `ab2026__minimum_wage_rates_april_2027` | measure | 2 | uk_rf_ab2026 | resolution_foundation | | `ab2026__minimum_wage_uprating_april_2027` | measure | 2 | uk_hmt_redbook | hm_treasury | | `ab2026__mod_counter_fraud_strategy` | measure | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_barnett_consequentials` | package | 12 | uk_devolved_govts | welsh_government | @@ -24,13 +26,23 @@ A row whose scope matches no registered measure has `measure_key: null` and name | `ab2026__package_hmrc_tax_gap_and_online_marketplace` | package | 7 | uk_hmt_redbook | hm_treasury | | `ab2026__package_hmt_distributional_budget_2026` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_hmt_distributional_since_ab2024` | package | 1 | uk_hmt_redbook | hm_treasury | +| `ab2026__package_household_energy_payment_and_electricity_vat` | package | 5 | uk_rf_ab2026 | resolution_foundation | +| `ab2026__package_obr_direct_effect_of_tax_measures` | package | 13 | uk_rf_ab2026 | resolution_foundation | | `ab2026__package_obr_forecast_policy_effects` | package | 24 | uk_hmt_redbook | hm_treasury | +| `ab2026__package_obr_tax_administration_compliance_and_debt_collection` | package | 7 | uk_rf_ab2026 | resolution_foundation | +| `ab2026__package_obr_total_effect_of_government_decisions` | package | 3 | uk_rf_ab2026 | resolution_foundation | | `ab2026__package_parliament_fuel_duty_decisions` | package | 3 | uk_hmt_redbook | hm_treasury | | `ab2026__package_parliament_regulatory_burden_reduction` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_parliament_tax_gap_measures` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_parliament_welfare_fraud_and_error_savings` | package | 1 | uk_hmt_redbook | hm_treasury | -| `ab2026__package_property_and_savings_income_rates` | package | 8 | uk_hmt_redbook, uk_ifs_ab2026, uk_tpa | hm_treasury, ifs, tax_policy_associates | +| `ab2026__package_property_and_savings_income_rates` | package | 33 | uk_hmt_redbook, uk_ifs_ab2026, uk_rf_ab2026, uk_tpa | hm_treasury, ifs, resolution_foundation, tax_policy_associates | +| `ab2026__package_rdel_decisions_excluding_household_energy_payment` | package | 3 | uk_rf_ab2026 | resolution_foundation | | `ab2026__package_regulation_tariff_and_sandbox_business_savings` | package | 1 | uk_hmt_redbook | hm_treasury | +| `ab2026__package_rf_ab26_permanent_personal_measures` | package | 19 | uk_rf_ab2026 | resolution_foundation | +| `ab2026__package_rf_absolute_policy_changes` | package | 2 | uk_rf_ab2026 | resolution_foundation | +| `ab2026__package_rf_energy_bill_support_announced_at_budget` | package | 2 | uk_rf_ab2026 | resolution_foundation | +| `ab2026__package_rf_other_tax_measures` | package | 6 | uk_rf_ab2026 | resolution_foundation | +| `ab2026__package_rf_parliament_permanent_tax_and_benefit_measures` | package | 42 | uk_rf_ab2026 | resolution_foundation | | `ab2026__package_tpa_business_rates` | package | 8 | uk_tpa | tax_policy_associates | | `ab2026__package_tpa_cgt_founders_relief_and_badr` | package | 1 | uk_tpa | tax_policy_associates | | `ab2026__package_tpa_energy_and_environment` | package | 6 | uk_tpa | tax_policy_associates | @@ -54,6 +66,10 @@ A row whose scope matches no registered measure has `measure_key: null` and name | `ab2026_option__childcare_100k_threshold_raised` | option | 2 | uk_centax | centax | | `ab2026_option__childcare_100k_threshold_removed` | option | 3 | uk_centax | centax | | `ab2026_option__childcare_100k_threshold_removed_first_15_hours_under_3s` | option | 1 | uk_centax | centax | +| `ab2026_option__fuel_duty_continued_freeze` | option | 6 | uk_rf_ab2026 | resolution_foundation | +| `ab2026_option__fuel_duty_rpi_uprating_since_2011` | option | 2 | uk_rf_ab2026 | resolution_foundation | | `ab2026_option__income_tax_5pc_share_devolved_to_mayors` | option | 1 | uk_ippr | ippr | +| `ab2026_option__lha_unfreeze_2027_28` | option | 17 | uk_rf_ab2026 | resolution_foundation | | `ab2026_option__nics_on_investment_income` | option | 10 | uk_centax | centax | | `ab2026_option__nics_on_partnership_income` | option | 24 | uk_centax | centax | +| `ab2026_option__obr_tax_measures_with_continued_fuel_duty_freeze` | option | 11 | uk_rf_ab2026 | resolution_foundation | diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/NOTES.md new file mode 100644 index 000000000..9ebf32db6 --- /dev/null +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/NOTES.md @@ -0,0 +1,372 @@ +# uk_rf_ab2026: Resolution Foundation, Autumn Budget 2026 + +**Working files.** `stage.py`, the file meant wherever these notes say `stage.py`, lives on branch `uk/ab2026-harvest-staging` at `harvest-staging/ab2026/lanes/rf/uk_rf_ab2026/stage.py`; this branch carries only NOTES.md, `manifest.jsonl` and `claims_staged.jsonl.gz`. The lane's local primaries folder (`.ab2026-primaries/uk_rf_ab2026/`, which the `--check-primaries` command below reads) was never committed: the repository carries no primary bytes, so fetch each primary from its `manifest.jsonl` URL and check its sha256. + +Staged 2026-10-29 by `stage.py`, which holds every row as data and rewrites `claims_staged.jsonl.gz` byte for byte. Source slug `resolution_foundation`. + +- **Rows:** 766 (767 before the review dropped row 397; see "Review fixes" at the end). + - Text: 212 rows from the two press releases, the Spread thin page, both versions of the Spread thin PDF and one comment. + - Workbook: 554 rows from the chart-data workbook RF links from the Spread thin page. +- **Primaries:** 7 (manifest.jsonl). +- **source_model:** + - `landman_ttm`: the IPPR tax-benefit model behind RF's distributional and living-standards figures (488 rows). + - `arithmetic`: RF's sums, averages and energy-bill construction on OBR, HMT and Ofgem inputs (148). + - `survey_microdata`: LCF, NEED and HBAI analysis, including HBAI outturn years (42). + - Restated rows carry the original engine: `obr_efo_forecast` (42), `hmt_costing_obr_certified` (38), `administrative_data` (5), `hm_government_publication` (3, Government figures that aren't Table 6.1 cells). + +`python3 stage.py --check-primaries ../../../.ab2026-primaries/uk_rf_ab2026` re-hashes every primary against the manifest. It also finds every text quote in its document (pdftotext `-layout`, `-raw` and default modes for the PDFs) and compares all 759 embedded workbook cells with the sheet XML. On 2026-10-29 it reported 0 problems. + +## Primaries and access recipe + +Every primary came from `curl -sSL` with the Safari User-Agent, HTTP 200, with no blocks or challenges. A second fetch at 17:02Z returned identical bytes for all seven. + +| primary | sha256 | bytes | fetched (UTC) | rows | +|---|---|--:|---|--:| +| Press release, 28 Oct: *Budget balancing act leaves families braced for tough winter…* | `ac5369e6…f98878` | 118,327 | 16:16:35 | 24 | +| Press release, 29 Oct: *Britain braced for painful second half of the Parliament* | `40b10268…c7c22d` | 116,817 | 16:16:36 | 19 | +| Spread thin publication page | `66a999dc…69d792` | 152,177 | 16:16:38 | 18 | +| Spread thin PDF, first version (`Spread-thin-1.pdf`) | `a669f39d…595603` | 1,410,169 | 16:16:41 | 3 | +| Spread thin PDF, replaced version (`Spread-thin.pdf`) | `9536b6cd…5d9ff` | 1,410,388 | 16:17:29 | 147 | +| Chart-data workbook (`Final-clean-charts-all-sections-for-comms.xlsx`) | `a4e1c072…6e0d1` | 14,637,565 | 16:17:32 | 555 | +| Comment, *It's good to be flexible* (Nye Cominetti, 29 Oct) | `3510a245…62fcfc` | 131,349 | 16:20:03 | 1 | + +- **HTML pages:** the stager quotes each page's own article text. Locators count body paragraphs after the dateline (press releases and comment) or name the bullet list ("measures bullet 2", "Key findings bullet 3"). +- **PDFs:** read with `pdftotext -layout`, and with `-raw` for the two-column boxes. Locators give the PDF page, which matches the printed folio, and the paragraph, bullet, box, footnote or figure label. +- **Workbook:** one sheet per figure (F1 to F23, plus "Extra spending stats"). + - `value_raw` is the cell's stored `` string from the sheet XML, at full precision, not the rounded display: F8's `_-"£"* #,##0_-` shows 145.33 as £145. + - The quote is "series label | category | stored value". + - Locators are `sheet F8, cell J8`. + - Each figure's plotted ranges come from the chart objects in the workbook. Figure 21 has no chart object; its column AD carries RF's own marker, "**USED IN CHART" (AD60). +- **Local copy:** the assignment's copy of `Spread-thin-1.pdf` (downloaded 08:46Z) hashes to the same `a669f39d…`, so v1 rows cite the fresh fetch. + +## The replaced PDF (README rule 7) + +RF replaced the briefing on the morning of 29 October. + +- **v1, `Spread-thin-1.pdf`:** PDF CreationDate 04:33 EDT; server Last-Modified 08:37:45 GMT. The URL still serves it, but the landing page no longer links it. +- **v2, `Spread-thin.pdf`:** CreationDate 08:13 EDT; Last-Modified 12:14:43 GMT. The landing page links this version. +- **Text changes, from a page-by-page word diff:** + - p.5: "the poorest half gain on average £158 … a paltry £14 more than the £144" became £145 / £15 / £130. + - p.19: "(equivalent to an average £158 for poorer households, compared to £144 for richer households)" became "(equivalent to under £145 on average for poorer households, compared to over £130 for richer households – a gap of less than £15)". + - Figure 8: the Council Tax rebate bars fall from about £100 to about £88 per household (vigintile 2); the other components are unchanged; the axis top moved accordingly, from +£180 / +3.6% to +£160 / +3.2%. + - Figure 4: the x-axis gains 2031. + - p.30: "restated the last Keir Starmer’s commitment" lost "the last". + - Elsewhere the lines only reflow. +- **Which version rows cite:** v2 is the source for every PDF row except the three p.5 v1 rows (£158, £14, £144), which carry `conditions.document_version: "v1 (superseded)"`. +- **Workbook:** uploaded at 12:16:43 GMT, it matches v2. Its Figure 8 net gains average about £145 over vigintiles 2–10. +- **Landing page:** its Key findings still printed v1's "gains £158 on average, against £144 for the richer half" at both fetches. Those two rows carry `document_version: "page text at fetch time; matches Spread thin v1, not v2"`. + +## Harvest rules 8 and 9 + +The coordinator added two rules to the AB2026 README during this family's staging, and `pipeline/ab2026_harvest_contract.py` now checks rule 9's rows. This family follows both: + +- **Rule 8: stage estimates, not design.** A row is a number someone estimated, forecast or counted about the effect of a measure or of the Budget. + - A measure's own parameters are design: rates, thresholds, payment amounts, dates and durations. So are spending allocations to programmes, financing remits and contingent liabilities. + - Design goes in `reform_hint` or `conditions`, never as rows. + - A worked example's inputs are design too; its computed outcome is a row. +- **Rule 9: a component of one measure keeps that measure's key.** The row carries `conditions.component` and `conditions.row_type: measure_component`. Proposed package keys are only for scopes that cross Table 6.1 lines, or have none. + +"Not staged" below lists the 36 design rows rule 8 removed and where their facts went. The electricity VAT cut's "original" six months and three-month "extension" are parts of Table 6.1 line 2, so their 39 rows carry line 2's key as measure components. + +## What the family stages + +- **Text:** every number RF prints that estimates, forecasts or counts the effect of the Budget's measures or of the Budget as a whole. That covers costs and yields, households affected, gains and losses, bill changes, distributional and poverty effects, and the public finances after the Budget. + - The text rows also include the living-standards and poverty projections and the energy-bill context the cost-of-living measures address, as `measure_key: null` context. + - Within one document, a number printed twice with the same meaning gets one row, at the occurrence that states its period or scope most completely. For example, p.16 states the £2,192 cap level as Q1 2027, which p.5 calls "next year", so the row cites p.16. + - Different documents' statements of the same figure are each staged (the AB2025 precedent; dedup belongs to the ingest). +- **Workbook:** the chart data behind figures of RF's own estimates. + - F7 (energy bill and support by year). + - F8 (temporary AB26 policies by vigintile). + - F9 (LHA unfreeze and Your First Home by decile). + - F10 (tax decisions and a continued Fuel Duty freeze by year). + - F20 (permanent tax and benefit changes this Parliament by vigintile, including the four AB26 component columns the chart doesn't plot). + - F21 (median non-pensioner income growth since 2023-24). + - F22 (non-pensioner income growth by vigintile and year). + - F23 (change in AHC poverty since 2023-24). +- **Structural zeros, not staged:** + - F7 support components in years when the support didn't exist (for example the Energy Price Guarantee in 2019-20). + - The 2023-24 zero base of F21 and F23. + - F22's vigintile 1, a zero placeholder for the excluded bottom 5 per cent. + - F8 and F20 leave vigintile 1 blank. + +## Coverage tally + +**Rows by workbook sheet:** F20 190, F8 114, F22 114, F10 54, F7 33, F9 26, F23 15, F21 8. + +**Rows by measure_key:** + +| measure_key | rows | +|---|--:| +| null (proposed key, or context) | 529 | +| `ab2026__package_household_energy_vat_and_fuel_duty` | 57 | +| `ab2026__electricity_vat_zero_rate_to_july_2027` | 42 (39 of them measure components) | +| `ab2026__household_energy_payment_100_bands_a_to_d` | 31 | +| `ab2026__fuel_duty_5p_cut_extended_and_rpi_rise_deferred` | 27 | +| `ab2026__pension_tax_allowance` | 21 | +| `ab2026__machine_games_duty_higher_stake_25pc` | 20 | +| `ab2026__your_first_home_equity_loan` | 16 | +| `ab2026__bank_surcharge_6pc_temporary` | 8 | +| `ab2026__writing_down_allowance_12pc_and_fya_50pc` | 6 | +| `ab2026__carers_allowance_earnings_taper` | 4 | +| `ab2026__business_rates_retention_mayoral_authorities` | 2 | +| `ab2026__package_total_policy_decisions` | 1 | +| `ab2026__founders_relief_cgt` | 1 | +| `ab2026__defence_investment_plan_further_savings` | 1 | + +The 529 null rows split into 197 with a `proposed_measure_key` (listed below) and 332 context rows. The context rows cover fiscal aggregates, energy bill levels, living standards and poverty. They also include the three pre-AB26 component columns of F20 (57 rows: "pre-AB26 Direct tax changes", "pre-AB26 Benefit changes" and "pre-AB26 Indirect tax changes"), which score earlier events' measures. + +## Conventions + +- **Sign.** Each row's `conditions.sign_convention` states it, and no value is re-signed from what the primary prints or stores. + - Household rows use "positive = gain to households". Text printed as a fall ("£2,000 lower", "fall by 3.0 per cent", "worse off by an average of £61") carries a negative value, as in AB2025. + - Exchequer rows say "positive = cost" or "positive = yield", matching how RF writes the figure. + - Figure 10 rows keep RF's plotted sign, positive = raises revenue. RF flipped the OBR's borrowing sign (workbook F10 I15, "Signs flipped from source"). + - F7 support components are negative, meaning a reduction in the typical bill, as stored. +- **Units.** £ values are in pounds (`gbp`), and `normalization` records the scaling, for example "£ billion × 1e9 → £". + - Per cent figures are `percent`; shares and "one-in-five"-style fractions are `share`. + - The stager multiplies workbook cells that store fractions (F8 K, F20 M, F21, F22) by 100 to give `percent`. F23 already stores percentage points. +- **Price basis:** + - F8 and the package's halves figures: cash terms. + - F7, F20 and the energy text: 2026-27 prices (Spread thin fn 13). + - £ living-standards figures: 2026-27 prices, as printed. + - `conditions.price_basis` records the basis on each row. +- **Period:** + - Fiscal-year figures carry `fy` and the end year. + - The temporary package (Figure 8, the 52/48 split, the halves' averages) covers 2026-27 and 2027-28 together. Those rows carry no `fy`, `period` 2028 and `conditions.horizon`. + - Averages over a span (for example "£11 billion a year on average over the next five years") take the span's end year with `statistic: annual average`. + - Calendar figures (the 2027 average bill, decade and Parliament RHDI growth) are `annual`; Q1-2027 cap levels and "currently" figures are `point_in_time`. +- **Value kinds.** RF prints point estimates throughout. + - "under", "up to" and "less than" give `range_high`; "over" and "just over" give `range_low`. + - Totals over several years and growth since a base year are `cumulative`. +- **Income groups.** `income_group` uses the AB2025 spellings `vigintile_N`, `decile_N`, `quintile_N`, `bottom_half` and `top_half`, the last two for RF's "poorer / richer half". The AB2025 NOTES flagged `bottom_half` / `top_half` for an ingest ruling. +- **Geography.** + - F8 rows are "UK", with `modelling_geography` quoting RF's note (Council Tax and VAT cuts modelled for GB, Fuel Duty for the UK). + - F9 and the Your First Home and LHA rows are England (Figure 9). + - Energy bills are GB. +- **Baselines.** + - F20's net and pre-AB26 columns, and the Parliament-wide text figures (99 rows), use `rf_start_of_parliament_policy` ("relative to the baseline inherited from Chancellor Hunt"). + - Every AB26 measure row is null: pre-Budget policy, current law at the scoring date. That includes F20's AB26 component columns, increments within RF's stack. +- **Attribution.** + - `restated` (84 rows): HM Treasury Table 6.1 costings, OBR forecasts and the Government's count of carers. `note` names the original on each. + - `own` (683): everything RF computes, including its sums and averages of official figures. + - `benchmark_class` is `administrative_fact` for outturns and official counts (53 rows), including the 2024-25 HBAI outturn year in F21–F23, and `different_model` otherwise. +- **Worked examples (mode 3):** the computed outcomes only (rule 8). + - The p.25 Pension Tax Allowance comparison: £34 Income Tax and £10 NICs for the worker, "nothing" for the pensioner. The £12,739 earnings input sits in `household_type`. + - The Box 4 Carer's Allowance cliff edge: net income of £280, then £216 a week. + - The NLW worker's £1,040 a year. + +## Measure keys: how RF's scopes map to the registry + +### Registered keys + +- **`ab2026__package_household_energy_vat_and_fuel_duty` (Table 6.1 lines 1, 2 and 3).** This key covers RF's "cost of living package": Figure 8's net change and per cent change, the 52 / 48 per cent split, the halves' averages and gap, the £4.4bn costing and the landing page's £158 / £144. + - RF's package is the Household Energy Payment, the electricity VAT cut ("original" six months plus the three-month "extension") and the Fuel Duty delay. + - Table 6.1 line 2's title covers the whole cut ("from 1 October 2026 to 1 July 2027"), and its costs (£755m in 2026-27, £305m in 2027-28) include the original six months. + - The registry's own modifier note gives the same reading: "about 57% of 2026-27 … about 22% of 2027-28". So RF's package equals lines 1, 2 and 3 exactly. + - The £4.4bn rows also carry `costing_basis`: RF counts Fuel Duty at the share accruing directly to households (fn 19). +- **`ab2026__electricity_vat_zero_rate_to_july_2027` (line 2):** + - **The whole cut:** the £40 "total nine-month benefit of the cut", and Figure 7's "Electricity VAT cut" series split by financial year (£26.77 in 2026-27, £12.80 in 2027-28). + - **Measure components (rule 9, `row_type: measure_component`):** Figure 8's "Energy VAT cut (original)" (the six months to the end of Q1 2027, 19 vigintile rows), "Energy VAT cut extension" (April to 1 July 2027, 19 rows), and p.17's extension "worth £12". `conditions.component` and `measure_part` name the part. Never sum them with a whole-cut row. +- **`ab2026__household_energy_payment_100_bands_a_to_d`:** Figure 8's "Council Tax rebate", the eligibility shares and counts, and the £2.6bn cost. The £100 payment itself is design (rule 8). +- **`ab2026__fuel_duty_5p_cut_extended_and_rpi_rise_deferred`:** Figure 8's "Fuel Duty freeze extension", Figure 10's "Fuel Duty" series (equal to line 3), the "just under £1.4 billion" cost and the "around half" household incidence. +- **`ab2026__pension_tax_allowance`:** F20's "Pensioner Personal Allowance" (the workbook's label; the PDF says "Pension Tax Allowance"), the £150m cost and the pensioner who "will pay nothing". +- **`ab2026__machine_games_duty_higher_stake_25pc`:** F20's "Machine Games Levy" and "around £145 million per year … through Gambling Duties … well-targeted at machine gaming" (Table 6.1 line 56 prints £145m a year from 2029-30). +- **`ab2026__your_first_home_equity_loan`:** Figure 9's "Could access Your First Home", the 76 per cent top-half share, "up to 70,000 families" and the £95m. +- **`ab2026__carers_allowance_earnings_taper`:** the £80m, the 48,000 carers and the Box 4 cliff edge the taper removes. +- **Restated lines.** These are each exactly one Table 6.1 line, restated by RF: + - Bank Surcharge (line 55): Figure 10 and "around £1.5 billion". + - Writing-down allowances (line 58): Figure 10's "Capital allowances". + - Founders' Relief (line 26): £300m. + - Business Rates Retention (line 21): £10m and £140m. + - Table 6.1 total policy decisions: £7.6bn in 2026-27. + - Defence Investment Plan line 40: RF's "addition to pay for the DIP … £6.7 billion between 2026-27 and 2031-32" equals line 40's sum (-£6,670m). The row stays at medium confidence because line 40's title reads "Commitment to further savings made at announcement on 30 June 2026". + +### Proposed keys + +Each row below has `measure_key: null` and this key in `conditions.proposed_measure_key`. Every one crosses Table 6.1 lines or has none (rule 9). + +| proposed key | rows | scope (Table 6.1 lines; RF's words) | +|---|--:|---| +| `ab2026__package_rf_parliament_permanent_tax_and_benefit_measures` | 42 | All permanent personal tax (direct and indirect) and benefit changes announced this Parliament, from the 2024 election to AB26 inclusive, against the policy inherited from Hunt. F20 "Net change in income" and "% change in income"; p.40 £320 / £2,000; p.41 0.9–2.2 per cent. Spans several events; AB26's part is lines 52, 56, 61, 62 and the gas standing charge. | +| `ab2026__gas_standing_charge_moved_to_unit_rates` | 31 | The announced abolition of the gas standing charge, "moving the £108 a year paid by the typical household onto unit rates by the end of 2027 or soon after". It costs the Exchequer nothing and has no Table 6.1 line. Covers p.18–19 (unit rates up around one-tenth; gainers and losers, by half and by children) and F20 "Remove standing charge on gas bills". The £108 sits in `reform_hint`. | +| `ab2026__package_property_and_savings_income_rates` | 25 | Table 6.1 lines 61 and 62 together: F10 "Savings & property" (their sum) and F20 "Property and savings tax rise". | +| `ab2026__package_rf_ab26_permanent_personal_measures` | 19 | F20 "AB26 Changes" = lines 52, 56, 61, 62 plus the gas standing charge (the workbook's column K equals columns F + G + I + J). | +| `ab2026_option__lha_unfreeze_2027_28` | 17 | Unfreezing Local Housing Allowance in 2027-28 instead of prolonging the freeze (no Table 6.1 line). Figure 9 "Would benefit if LHA were unfrozen"; "over a million families lose out, 91 per cent … in the bottom half". | +| `ab2026__package_obr_direct_effect_of_tax_measures` | 13 | The OBR's Table 3.1 "Direct effect of tax measures" since March, sign flipped. F10 "Net direct effects", the p.23 labels, and p.4 "about £5 billion annually from next year". Differs from Table 6.1's tax total (-£1,749m vs -£1,645m in 2026-27). | +| `ab2026_option__obr_tax_measures_with_continued_fuel_duty_freeze` | 11 | The OBR's direct effect of tax measures plus keeping Fuel Duty at current rates: F10 "Net effect if Fuel Duty frozen" and its p.23 labels. | +| `ab2026__package_obr_tax_administration_compliance_and_debt_collection` | 7 | The OBR's Table 3.1 group "Tax administration, compliance and debt collection measures" (it includes line 45 and timely payment for Self Assessment; RF doesn't list the lines). F10 row 52; p.4 "over £3 billion annually by 2030-31". | +| `ab2026_option__fuel_duty_continued_freeze` | 6 | Fuel Duty rates held at 52.95p a litre instead of rising to 60.59p in July 2027, relative to the announced policy: the OBR's constant-duty-rate alternative less its central forecast. F10 "… and if Fuel Duty frozen"; p.24 £4.3bn in 2029-30. | +| `ab2026__package_rf_other_tax_measures` | 6 | RF's residual in F10 "Other tax measures": OBR net direct effects less Fuel Duty, Bank Surcharge, capital allowances, savings and property, and tax administration. | +| `ab2026__package_household_energy_payment_and_electricity_vat` | 5 | Lines 1 and 2: "A £3.7 billion package of household support"; p.17 "£3.4 billion in 2026-27", "£0.3 billion in 2027-28", 45 per cent. The sums show line 2 is the whole cut; see the discrepancies below. | +| `ab2026__package_obr_total_effect_of_government_decisions` | 3 | The OBR's total effect of the Budget's decisions on borrowing, direct and indirect, including decisions not on Table 6.1: "an average of £2 billion a year" (two documents); fn 47 "no overall impact". | +| `ab2026__package_rdel_decisions_excluding_household_energy_payment` | 3 | All the Budget's RDEL decisions over the Spending Review period except the Household Energy Payment: "£3 billion a year", "£9 billion top up" (two documents). A fiscal aggregate, not a programme allocation. | +| `ab2026__package_rf_absolute_policy_changes` | 2 | Every policy decision counted at absolute value (RF's fiscal-event size measure): "£7 billion". RF lists no lines. | +| `ab2026__package_rf_energy_bill_support_announced_at_budget` | 2 | "Support of up to £116 announced today" and "less than a fifth" of the 2027 bill rise. RF doesn't itemise the £116; it includes the £100 Household Energy Payment. | +| `ab2026_option__fuel_duty_rpi_uprating_since_2011` | 2 | Fuel Duty uprated by RPI since January 2011: "around £20 billion" a year (two documents). | +| `ab2026__minimum_wage_rates_april_2027` | 2 | The April 2027 minimum wage rates the Chancellor confirmed (no Table 6.1 line): "worth £1,040 a year for someone working 40 hours a week" (PDF p.12 and the comment). The rates (3.9 per cent; £13.21; 50p) sit in `reform_hint`. | +| `ab2026__gas_supplier_levy_funded_from_general_taxation` | 1 | Funding the Gas Supplier Levy from general taxation: "save the typical household £7 in 2027-28" (fn 12). No Table 6.1 line found. | + +## Proposed metrics + +None of these is a registered Metric. Each row carries one as `proposed_metric`, reusing the AB2025 spelling where one exists: `forecast_borrowing_change`, `average_energy_bill_change`, `incidence_share`, `tax_to_gdp_ratio`, `real_spending_growth`, `spending_pressure` and `forecast_revenue_change`. The rest are new. Every unit is a registered UnitConcept. + +- **Energy:** + - `typical_energy_bill`: the typical household's annual bill level. + - `average_energy_bill_change`: a change in it from prices or support combined, not from one policy. The registered `energy_bill_change` carries single-policy effects. + - `unit_rate_change`. + - `bill_rise_offset_share`. +- **Public finances:** + - `debt_interest_spending`, `debt_interest_change`, `debt_interest_share_of_gdp`. + - `current_budget_balance` (positive = deficit). + - `primary_balance`, `primary_balance_change`. + - `tax_revenue_change`. + - `psnb_share_of_gdp`, `psnd_share_of_gdp`, `psni_share_of_gdp`. + - `headroom_share_of_previous`, `share_of_headroom_change`. + - `tax_to_gdp_ratio_change`. + - `absolute_policy_change`. + - `share_of_total_policy_cost`. +- **Spending:** + - `rdel_change`. + - `rdel_share_of_gdp`. + - `real_rdel_per_person_change`. + - `real_spending_change` (positive = size of a real cut). + - `defence_spending_share_of_gdp`. +- **Households:** + - `real_income_change` (£ change in a real income statistic). + - `income_change_gap`. + - `net_weekly_income`. + - `annual_earnings_change`. + +## Not staged, and why + +- **Design, not estimates (rule 8; 36 rows removed after the rule arrived):** + - **Fuel Duty rates:** PR28's "7.6p rise"; p.24's 52.95p to 60.59p a litre and "over 14 per cent". The rates sit in the continued-freeze option row's `conditions.counterfactual`. + - **Minimum wage rates and counterfactual upratings:** 3.9 per cent, £13.21 and 50p (p.12 and the comment), the "on-course" 5.7 per cent and the youth rate's 10 per cent (Box 1 and the comment). The rates sit in the £1,040 rows' `reform_hint`. + - **Financing remits:** NWF £15.4bn, BBB £5bn, the combined £20.4bn (PR28 and p.6), the 0.27 per cent of GDP of investment commitments, the £5bn defence ringfence and "around a quarter" (p.33). + - **Programme allocations:** + - The AI package's £700m, with £165m biosecurity, £120m AI Security Institute, £140m Sovereign AI Unit, £30m local partnerships and the £15m challenge prize. + - Good Growth Funds' £285m in 2027-28. + - The Social and Affordable Homes Programme's £1.5bn top-up and its 10,000 homes. + - **The £108 gas standing charge:** it sits in the gas rows' `reform_hint`. + - **A worked example's input:** the £12,739 part-time earnings, which sit in the worker rows' `household_type`. + - **Figure 7's flat payments:** the Council Tax rebates (£100 in 2026-27, the 2022 £150) and the 2022-23 £400 Energy Bills Support Scheme, as F7 cells E7, L7 and L11. + - **The 23 per cent rise in private rents since September 2023 (p.21):** an ONS statistic about no measure. +- **Charts without printed values.** The PDF's figures print no bar values except where staged: Figure 7's two 'After rebate' annotations (the two 'Net bill' labels are staged from p.17 bullet 2), Figure 10's eleven line labels and Figure 23's end labels. The exact values come from the workbook instead. Figure 5's decade labels for the 1960s–2010s are historical ONS outturn, not about the Budget; the text rows carry the 2020s values (1.4 per cent, £400). +- **Workbook sheets left out:** + - These sheets chart OBR, ONS, IMF or HMT aggregates and history, or RF's arithmetic on them. This follows AB2025, which staged these charts' headline text values instead. + - F1 (borrowing and debt, pre-measures). + - F2 (borrowing decomposition). + - F3 (gilt yields, debt interest). + - F4 (CPI, earnings). + - F5 (RHDI by decade). + - F6 (GDP decomposition). + - F11 (monthly real Fuel Duty rates, 2008–2032). + - F13 (real RDEL per person decomposition). + - F14 (spending growth by period). + - F15 (departmental RDEL indices). + - F16 (PSNB paths). + - F17 (current balance and PSNFL). + - F18 (current-balance forecast history). + - F19 (international borrowing). + - "Extra spending stats". + - F12 holds marginal tax rates, which are design. + - Within the staged sheets, the family leaves out: + - Working and input blocks: F9's raw winner tables, F10's HMT and OBR input rows 5–40, F21's percentile columns other than AD, and F23's level, raw and scenario blocks. + - F22's second chart (bottom 7 / middle 6 / top 7 vigintiles since 2019-20), which the briefing doesn't print. +- **Text numbers left out:** + - **Macro forecasts and statistics:** + - CPI paths (6.5 / 3.9 per cent, 43 / 22 per cent, 2.1 / 3.8 per cent, 16 / 13 per cent). + - Gilt yields (44 basis points). + - GDP growth (1.1 / 1.2 per cent, 1 per cent, 0.6 points, 0.5 per cent, 1.75 per cent). + - Unemployment and earnings (5 per cent; 68 / 26 per cent; 3.8 per cent real earnings; 0.1 per cent real wages). + - Migration levels (171,000; 331,000; 340,000 / 290,000 / 230,000; 70,000). + - Wholesale gas and pump prices (60 per cent, a quarter, £2 a litre, a fifth, 22 / 14 per cent). + - ONS business survey (15 per cent, 10 points). + - Rents forecast to rise 3.6 per cent in 2027-28. + - The OBR's 3.6 per cent Q2 2027 inflation. + - ASHE's 5.9 per cent and the LPC's 3.7 per cent April estimate (Box 1 and comment). + - The comment's 3.7 / 3.3 per cent earnings figures. + - **History and international comparisons:** + - Tax/GDP 37.2 per cent in 1948 and 4.6 points to 2019-20. + - Receipts margins (6.3 / 3.5 per cent, fn 24). + - Spending growth in earlier parliaments (1.8 / 1.4 / 3.6 per cent). + - Unprotected departments' 4.6 per cent low (2018-19) and 7 per cent (2009-10). + - Benefit and NI rates in past years (p.44). + - Fuel Duty's 40 per cent real fall since 2011. + - The EIB's 0.29 per cent of GDP. + - The G7, OECD and IMF borrowing comparison (3 and 0.3 points). + - The £72bn average of Reeves's two Budgets, which has no single period. + - The 19th of 34 fiscal events (a rank). + - **Other design:** £100, bands A-D, 5p, 3p / 2p, 2 points, 3 to 6 per cent, the 31 per cent headline bank rate, 20 / 25 per cent, the £1.5bn Good Growth lending facility, the 20 per cent equity loan and 2.5 per cent deposit, the £110,000 / £140,000 income caps, £86.45, £204, 16 hours, £1 for £1, the 55 per cent UC taper, Business Rates retention shares (20 / 50 / 70 / 100 per cent), and the 3 and 3.5 per cent defence targets. + - **Third-party figures:** + - Anthropic's $65bn (£48bn) round and the "1.5 per cent" comparison. + - The Health Foundation's £7.5bn. + - EDF's "top £2,000 in each quarter" (fn 9). + - Arin Dube's 70 per cent bite. + - **Ratios and qualitative multiples:** + - "twice the level of funding given to households". + - "twice the level of support" for a targeted scheme. + - "one-tenth of the growth rate achieved in the 2010s". + - "enough to cut employee National Insurance by 2p". + - "help half as many poorer households twice as much". + - "About half of that buffer was wiped out". + - The "0.3 percentage point gap" (3.0 minus 2.7). + - "35-year old property valuations". + - **Other context:** + - The 16 mayoral areas' £26bn (7 per cent of Income Tax). + - "at least 28 MSAs". + - 1.5 million homes. + - 33GW / 300GW and £750bn of infrastructure. + - "around half a million low-income families" (the AB2025 two-child limit repeal). + - The £24bn year-four headroom (a rounding of £23.6bn). + - p.15's "£7 billion in 2029-30" (a rounding of Box 2's £6.7bn). +- **Publications checked:** + - RF's publications, press-release, comment, Top of the Charts, events and Budgets listings were all fetched at 16:19–16:20Z. + - The only other RF Budget output from 28–29 October is the launch event page, *Pleasing people and satisfying markets*. It is an embedded YouTube video with no printed figures, so the family neither hashes nor stages it. + - The comment's three slide images (`Slide1-1-scaled.png` to `Slide3-1-scaled.png`) are charts, not staged; its numbers are minimum-wage design apart from the £1,040. + - No Budget response briefing, chart pack or Top of the Charts Budget special from 28–29 October appeared. + - The Spread thin page's "Further reading" links two pre-Budget briefings, outside this window and not staged: + - *Counting the cost* (8 Oct; page sha256 `415deda6…`). + - *Hitting the buffers* (13 Oct; `881bbd4d…`). + +## Discrepancies in the primaries (reported, not fixed) + +1. **v1 and v2 disagree on the income halves.** v1 prints £158 / £14 / £144; v2 prints £145 / £15 / £130, and v2 p.19 hedges them as "under £145 … over £130 … less than £15". The landing page still printed £158 / £144 when fetched. +2. **The £4.4bn package's years.** p.19 calls it "the Government’s £4.4 billion cost of living package for 2025-26 and 2026-27". The measures fall in 2026-27 and 2027-28 (Table 6.1 lines 1–3). The press release and p.47 print no years. The family stages the period as printed, at medium confidence. +3. **The £3.4bn's scope.** p.17 says "Together, these measures cost £3.4 billion in 2026-27" of the Household Energy Payment and "a three-month extension to the VAT cut". The extension falls in 2027-28, and £3.4bn equals lines 1 and 2 in 2026-27 (£2,625m + £755m), so the figure includes the original six months of the VAT cut. +4. **Your First Home's sign.** p.21 calls £95m in 2029-30 "a cost". Table 6.1 line 37 prints +95, a yield ("net income from developer fees"). +5. **SAHP top-up** (not staged, rule 8). p.21's "£1.5 billion top-up" differs from Table 6.1 line 35, which sums to £1,750m over 2027-28 to 2030-31. +6. **Good Growth Funds** (not staged, rule 8). p.33's "£285 million in 2027-28" differs from line 20, which prints -£340m that year (the line includes the lending subsidy). +7. **Tax gap measures.** p.4's "over £3 billion annually by 2030-31" doesn't match the workbook's OBR series (F10 row 52): 2.89 in 2030-31 and 3.64 in 2031-32. +8. **Figure 20's AB26 component.** The note says "‘AB26 Changes’ includes the Pension Tax Allowance, removal of standing charges on gas bills, and the Machine Games Levy". The workbook's AB26 Changes also includes "Property and savings tax rise" (column K = F + G + I + J). The workbook labels the Pension Tax Allowance "Pensioner Personal Allowance". +9. **Figure 10's Fuel Duty series.** Its note says the series shows "Fuel Duty changes announced in May and October", but the values (-0.575, -0.78) equal Table 6.1 line 3 alone. +10. **Primary balance years.** p.39 says the £107bn of higher tax revenues runs to the end of the Parliament (2028-29). Footnote 53 computes it "between 2023-24 and 2029-30, multiplied by forecasted GDP in 2029-30". +11. **Energy bill figures, release vs briefing:** + - The 28 October release's "£2,210 across 2027" (the 2027 average) differs from Spread thin's £2,192 (the Q1 2027 cap), which p.5 calls the bill "next year". + - The 29 October release's 2022-23 net bill of £1,685 differs from Figure 7's "After rebate: £1,686" (workbook 1,686.20). + - p.17's VAT extension "worth £12" differs from the workbook's 12.80 for the cut in 2027-28 (2026-27 prices). +12. **Footnote 50's figure reference.** It reads "From the forecasts shown in Figure 7", but the current-balance forecasts appear in Figure 18; Figure 7 is the energy bill chart. +13. **Capacity vs Table 6.2** (not staged, rule 8). The NWF's £15.4bn and BBB's £5bn "capacity" boosts don't match Table 6.2's financial transactions: NWF line 2 sums to £10.2bn over 2028-29 to 2031-32, BBB line 3 to £0.75bn. +14. *(Withdrawn on 3 November: the README's rule 3 example was corrected, so RF's package equals the registered one; see "Measure keys".)* +15. **The VAT element's description.** PR28 para 5 and p.17 bullet 2 describe the package's VAT element as the "three-month extension", and p.17 calls the six-month cut "support announced before this Budget". Yet Figure 8, the 52/48 split and the £4.4bn all include "Energy VAT cut (original)". + +## Review fixes, 3 November 2026 + +The RF review (`harvest-staging/ab2026/reviews/verify-rf.md` on the staging branch) approved this family with fixes; all thirteen are applied, finding 1 as the `P_SAV_PROP` constant and the rest in an explicit block before `write()` in `stage.py`. No primary was re-opened (the session's network policy refused resolutionfoundation.org); no printed value changed. + +1. The 25 rows for exactly lines 61 and 62 use `ab2026__package_property_and_savings_income_rates`, README rule 2's name (F10 row 51 equals lines 61 + 62 in every year). +2. The passage and discrepancy 14 that contradicted README rule 3's old example are gone; that example has been corrected. +3. Row 168's note names the Government's Defence Investment Plan projection. +4. Row 38's note names the ONS/OBR outturn. +5. The v1→v2 record notes Figure 8's lower Council Tax rebate bars. +6. "Figure 7's four annotations" now reads two, with the two "Net bill" labels staged from p.17. +7. The workbook hash typo is fixed. +8. Row 23 (−0.4 per cent) is a `point`, `statistic: RHDI per person, average annual growth`, `parse_confidence: medium`. +9. Row 167 (0.5 per cent) is `statistic: real RDEL per person, average annual change (Figure 14)`. +10. Rows 28, 29, 56, 57, 61 and 63 carry `price_basis: cash terms (Figure 8)`. +11. (Optional, applied.) Row 397, F10 C48's floating-point zero (−7.46e-7 bn), is dropped: F10 54 rows, `ab2026_option__fuel_duty_continued_freeze` 6, `arithmetic` 148, total 766. +12. (Optional, applied; England, the coordinator's call.) Rows 109 and 110 (£358 and 20 per cent) are `geography: England`, as row 108 is. +13. Discrepancy 15 records the VAT element's description. + +- **Verification:** the contract checker reports 766 rows, 0 problems, and a re-run of `stage.py` reproduces the gzip byte for byte. COUNTS.json pins its sha256. Proposed-key names may change again in the cross-family harmonisation recorded below. + diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/claims_staged.jsonl.gz new file mode 100644 index 0000000000000000000000000000000000000000..8c24bea78875c8685823658c99f709c10f866bf6 GIT binary patch literal 52874 zcmV)rK$*WEiwFP!00002|LncplHy2~E_#2S!Y$2?0X9JDUyx~Yk(E_hRoz|HIa%4= zHditMPM}I5(UOSBs5#rLuJ!}$eX*yn^G$NUNj=F~%iZ+@2tp771bEL>RRkd<*YdT# zwS4{l*9$jEB5$-zyfAq2ioA$uoVW>{yy%h_BR8hL7f@#r`rh~v{v0fQAOApu@oer! zpPaGp#xZ_x;!USCqCw)!!wL2A>*Fw(cz9pD;J@&09=miA(cYZ8@e-cJubVI2$cy;` z{yI*=5siqZXcoDT$xm)Hp7G!Q`ppM24I?^s;{?Ar_2RMXJMRz09g4p6|nN zUr(k}T9mK)B!F>!+l;Y20jGV#OkQ64G}aYTJah(+WDNlb!}h+TP2 z*2qfkfOauyZQGkxA72kxD5gmGsW$=kWUo5&?q=+7?8)#6?3{|fMT~ zeK!~b35YvR2>cDnKXvDx@4*kF$Q{G}0>MBQmUlA(#7bai81sb7sE3pIhwt>KpdANv4h>)#aE3f~V$kpoX~ zif|~dN!$_J-1M6u>`TV0;U~WLbLwPREFctLfNf0RB`8Bzl1)X%U9wZcGG?yIfBc zHA;3ZE;#_jgRHm}(C2jGE$8cyu9ZW2Ihl`u=Z3`dOTzPau!0%kOhJU~5jUdbA&in4 zu(zFYJs2RF$tyIy*0FRRdr)Co|x9VEB_L0Q(U9w|Bw+@y;9j=>TTj zZ_xp}P5N;Nwk@0r5Zb=`Jf6{iE*z?T)AATYqBeufFZi*BPl0PE<5UZ~e+zbNeoZi3!Qd0^9@1?mQr_mplT&lXAdsG2!rn9(qGs4DFTvpa}P8 zMovp2=OrPh8{vvbk*i2S6oCJQ*bvU1K0EA@c)3`B1}{NV7hNTa0(2h%({M`8;GC8c zZJR{ZY$mUw3Vh!^5~>1=N3@8mAdVH`w1oC2^OpcDgg~NHely5oTaFrW6edYH2V>2j zrl))_7>9F8CP>$GeU+W^m9d#W#c@A#@@Fe6dwiF~3&2a_CTxM|EN}z#hrEL0Sr~!f zz9fLI02%B>I)(e+0r3~i$-`dM!M`&WS&p(l`6vxQ4}?yTFyz&uaI`lW`8$Rb?kK#c z9rokg0}Kk62|5dLiCayJ`t^wmJb;dQCu7k~Z4w3)S&zaCCaE@2HB;&nsZXAzPl|j+ zo&zq8)Gi>T_w>=3d*diPt!e?;+)ud_H{#NiOL}!g@2*nSi1&g&`~~QT(WGF)ZvG1> zj2ZbZjq;Req@9|F{qz8*Gy)h7)EiU$R`awEXbv*QXvIr{GL71&aB(BFk1e&&C260( zanl*2k}I%uc79XXJDKINfRR2>DvUAD@IuUTjO%slAGv|Yk~{s+aMk*WVP zj6TIP06g{^dgp&w&POzo>O`s&sZJ`Z6P~9CubhObA)Pz`hdq+ES;aYAvEv zJ%N~~*CN?#efRCV43hvsD0%BHvdj~fOoQ2_D~XOjhb%=hB;Uau!-xm&4@fxu{u`b} zl)_x{xZp9W$AyP^mXoFH_ohqg=Qhb62|8}f1FSqsl)~C-j8(PBVwv!C&;L*vl#wyK zKd2!L>jHd`HcvS9Vmj%(78;YuSu1AOXsUv_Z}#E?-r`u<&-(!>>q-lcDKqntl-n zFioGKW+DIeL{Rx;b`E#evnX}^Y+$Vc?*xg8cz3+SL5zJ9Ub;L?WI z*9QgrQiP7y1@3w<5Z<~n1j4lv03`rQ0F(fD!2k&FHl_u9gU|G;X(I*FG_1U4FKo*_ zr`)B8ERDNdNG;lzXXbMXz$UWH===c->K2 zQ^@Np#l4F7`{5l(H;tL=)l29cPV9sytg3L#Z;Ky6Xtw!c$--a8B!*}CLMfzmL0G6D z2!s+Q0BYhscKYN!gLPibGmXMI-x;_a;(v>u!}!fWJ<|QEcaGs$m9T$Nu>Xq?GJyQ$ z0+0KgopXLnLg2%a{EzBiyhuYB-ep4xM2HwpYHzJ*f&vN76{zC?gjyScAO*@4;~7Hm zvJU%MCh`$9L3PWx<&vp4O=f9=oL0p^-T{4^CNf(r63_{!(-=sR`Q0T>-ZDy@KLdO#sV8>s3vc8QCI!~uWEb(M8gGR1PA>M&SZ&IC=1V7 zE-;5axN~ASs8M*Hsy=N*KYG-kR4I|%FqUhodEBEQ(I7Dv?Y2b1^)1N5nEHMg#f0OE|5$*J z6ks^4YzLFeEkxIp{VUN+sN7_aNu8~#_A#2V_%sO6A{UYusr zowC_gt`)K+WGf7umHQv8BvPNoD@NGtIiyvTRz{rsg|==4`Q!Khjm?7GAm1}|YHDjT z-XiUXndejPVPbU)i8v+=d%Duuzr9txI!0BclDVktiKkg^OWk{}R$4Jx74$366#Epm zpIkxY-4Vk$soEZK<=VAoi4AI$coA(8FRX{41m(*DWpQ={{st+G9(l_^=8wqwW5VB< z{C3NX22~#g7vaL$v!kGXP)9)|MiZM!r2qgtksw5^0d|*cpnK-2K~rxF3P(#WXyavs zixi?&$y65Wo>6xah2i|g8JfoKWxgsBDW#8u1+k63ISNS1xImU^@lbTVEmjTl5I|(q zYG`}aUCy;PO{e))`%+CV`1MK|-r~MIq3`N~Mn1wBr(fg|YVR0s!&=K+;@I4rr zBwkP;2b}&-gS+J&rMDn2uGC3adc?E6#NEqETOJ{{7QL(>CiHU>xof*+=4rn!3HdXA`pw-{8r!2svTH(6{rHIUv0#b@-Z-b4e(l04+3vq5m}2g zlAzj9P_5flAR$<c2KSxLprQ6F{n*sr(p0@;)ZXLnxAs+`I5uq5j=KdgrpuFEq~E zh|V)c%FFmB9^homMu~`CXzdR_eEVaEF&F%U?@=8ug$0VUq+AeO-zV*sA3o5iMP8B? ze!za100g!=ugTbrA`dPP=pAeMO@IaAv8|nN)H<}%+1dfVqJi7-pjXZ&xhUxUu(y<8 z=Oy9Q^>Ol`p(%v1{zbqrgHDhF2S2U=#65Vu%J>V{UX+E)DhJ-Xk?*;z7#A#A&ESbr z+OV^MsK^NgM(Ac4AtDh9jMoP&WT{d|kwcOA)^VCX43B_tvgi9nAsnkZ+sI)06;4n@ z>07yNxh)g*(mOZ6OI8H1A~|jd$4=UXW_V89Od{7sAy=Cnb#2?qPP&~m&v}L4udCe_ zyzd5HjhAya4DFtl&GWRa22Oi>3b82h(RsLvTIx`smT_^`S9yu4AyMq+y9CT=AP$p2 z2I~80jP;bhk=q(qXLI4}RW!O*Dbdi3$;!wQXB9G#hk^h2y3o&yPa>t{Z-357 zg!%Y1-Z?5hZNaiJ73G+B8fCv`c(jcV#;#XTwfL#KwF-A7^Kq`bJVMlN8&I`c!ioMSuRzWh0MaEx5X^(w? z`SU!lj!(|cWebF`?w+nV&2R1oPLj+|H#y6!I)JgTby8uRRAD6HLTU@CEpF4i4-=FP zBOfn7?n37A<9$3qHb_2(RBDPNY6_$3EC^#DHN`C-EjImDV>g}=Ui6S%-8nSfm$B+1 zT(HnE|4!k8BWS90R#5)Ejg*Zeq>M;qby>=a=XPTg6cZQ;*;s0;V|K6E zYhp#db5dX3+Q~*LE2*q*Pg$)Dk-z8Yrz`9&Qkj*?>SVnqw*@%%f?}pQ!tQbDLz=&r$O!S!-Uq&^)rhhGN*`-~Rl^@Bd=D z6^Ty@-AS)3-N!ztp`_i^hLR@$Ez$3h4R~H8am_GQ%zB&##ME&N;$n$S%e{n;El!{O z^80V!XZ?fIHxVW+cEcpd7g=vi0$2BI0Kvrl>}(FSlK^-@062%^mYCYG#F1;TS!}u^ zf05B6o4#!q+d!r>16g)2+xu&500}RBQ2(5qRu|>}O`9zLA~Gq+2YT<(2f{%hs}?AL zV`s0i#S=WZ=0nati(;G2qwYc!+Ej0~n|3zClg4Q~?e0^ZjRZJ0M9S0ikxtKFzYH@j z!7LFzmwbiwxU>~UQ>GG?@+4~`ok>*5(R-iAT=oF7;Vxv|S*SfcIk=)T1c54h)!avs zDoPQxVred@KcgjXHy32ks4H=t)jAt6>QoM-b+t&3i#^iQ157$7xV<_VSRJOLgIn+b zv!_|9#fVVfst~7Qo}vm6Sb$4csa*c`1-}{Duy!W0$k!0 zeoa2S<69mR-s+iQmYD6*H%0*|vIchbTqU>J75jav8Ou^%oGI1t0V{{xn9P^s8UB~^ zh|xDP4qbDxrj?sFnW}zj?PC}R*!6|YYp9olH4v^alSt3AUng{m&;v)g2)j(WxEnai+1^FzCtRcviVC zQT7-p`+(KHtZyo4Z`O;=DE9e*x=dr5-PrGpWgN}||J>lY3I3~EkgA)OOy^vY199|daaYz>*&N*L!oKHU8L%S(V9 zl8&=woWDz}d3g-F?FArj9t$X9!1_;sw)02w!!p>@N-`ucT_-TLwZ6FqM~rn&(wS^k zCgJ<;DIv+M3`u6=8$3QI$p*s#Y@8FDCW+6Qjh}2;w$a{JlnpWWmFQ@@F^pnMel|7- zW4%o8C}{^_&|7f#CyhE}XYTeTxA}e&@JtDK#{fKZAOQcPN0m(#fVKNcgqb#$q7(wi z075Y>e~BZ*;>lB`;l=k01j83D9thVGYMuv#&z*JEB=BtXXe^1#R~xq#{ZrFU8YeZ+@N^?(qW`wyVguY||-g~Y?^8s4ouEgET!`-qu5c)pA zt6zzmU?MHx2y5-J@#w65W$=~yW zL{*ZjPZgn7%CTrdM>v%?p-6Tos|2bGkK*N=RsZG~v?Y7mvfSZtI2r4vVOy?gj1@&6 zjfT^StI*MCq+7;dJTf&y(I@uE9*j-+8y!x~sc8)-_E0+jg9>@(ugIIg^&jJOkkK2) z89awc3>Rw`=mD>DM>~6iPkw{=Wy~Jl!-ub-$zlnwFk#Ec%sJd8lI{|p*ukq@uw`0a zX2_#JDc@d4PzMN-xTR@DmviyU^O zioix;UCox4M7q6^4x4CYOp{ks1Bx-ZxLyp?4}viLJP1}1hy(?ldssad7|{ks$y5PbytgY2xFOegcDfmne!1}XoiTT>XbSFp2wK~K{i@TV}X+h4zE zdfyraof@sK>Sj}m*AGs;X)=pj41Scpb!gev7Q-i{NS^Vxz_R*q8B6o$e4xY2p^N*E zz~Uw8v4GXhTtod;$s`M_BBE^4=<@r6ueB4faoncPD&g_6uV0kDF3_lIpO|g!U|4T} zLNzn?3#hL>K}i_&IPC^F`SO+nI*v@--g2OSmbTdv0QUyKH;nOf_myC0M8SYQiI9b7GJjy?F3UGw;Z%^w!uK>ndR2PXQw8nN!l#DN=C@=ofZ7W zt~{_;tjU3wwrwfzz44lCHpjuy5NkcjTXEXwCn1F?hI@unC03sgt8t(QKMOOpn}%bv zPULP7;Fu6P!Az5HmQgTs=S<>$psr6JWzX#6j}hHthQaDa%7nwc6Arhe6s}|u%GAN< z1u@;F2Oiron|ou zHjH@7zzy&e)h%NB*ZF#N>uNcqWf=h3E7Xl_f2NOQ>M=LRQ}YcZwx!rM1BM|Yr*OH* z`HR(1=RN$H(H~UZ|aLjpSO)xwItN18e}lv<-z1eyYQ`O=bk+4(J?AV#jro z(3z~l8ocbC2OH%uJyHs5^CG4lnqw`Y?5t3>GvV(pL>EOUYGXBpq&r@)7(iX8On6O= zQdgsHS+CUX=(1y``a1becM1*1OD5{~G>TIb3H4F}N&pSRDNVq7N5Z@&a~FPw{Vp0} z_i)HYi|t;k+n;|OiHC@oNPBE)?yS7{Tg=|_eDcx#%+@PZbWHg5qAADsK@OzME^u`xJ!PP0&CCd1$g~tS1gfrNe=RV`>-+!%=P=*T zjyCsdi`~>507v1?4e~If0T(zrIZsNxRFDTM$Ra<8o;fAp!bBjuU`=WrKJYQRs)kUCQNu`QZ(GPU&xGVevFuKJ^;@_(&eYCGZRb zPH)PteSig$WdOJOjRsGTZ!CB!o~Q*57mPm~f95)Pe_Rs2k?ru?mf$oX2w*MFIS()< zL#FWBK4?+W2IA5ws6PA>sE`U}4ET;T+TnX^xv4jSSD>->f2>pD);hHBQ%W!iVcMK+ zK@qow{u+|DAXPWrg3fD#4W36~_=&t5lAkm!9Yd~mO)%rTat-0teWySRBv-m9V(dQ(ZQGEo6PG;97}QXQnB4$iO7pHE*1A9 z5kf?{%{!iX{sib!1q9RO;q7*L7dGn8HfzkzURhfPilzj`-Jr;rCF3eD7BQ2xIxwwo z5s~nV?ouYmB;W(;enjV42DXinXUJDm(nxQ^0+G8`+KJ5|O*#T=TJq90%FjliUs;sx)& zeb*KA7F6z%!Up$Lw#EA@MR&1yxm@_6JBeT7b6={;%Z&e6=dC-#wJ*jr3+LdbehO@3jq)AtMx>l_^yI7pfD;q%1roCsxkXL{*V91u>L<*ZO zPLF8|kk&qf7NV)+ju=~{v=QjN+3OIe6^KWrUm*Iw<%0t+Mnkj!wcP;jPEO1QWxw@)-o18ez+QF1ybApqZtYIjLbCm>Ge` zGysx^IpH#Vi3R-dvkH8Mkhh@jC(cuT!lL)!eHYrB)vp8{Mbd zaZ3>18-!t)_P z4X{dD%xS;!wq26HE<1WLizR>ER)@)&)*lRc&sIiwcBS684DlIvAjBbQF=YPwB^z4a zBE0_&6FhO{p8fFvd4!E&aY5_5p93#u%;rM72D&a`)ksXu)A1}|QyDU450W%y z1wTlSKYssj#O6D9@yGA~y6|z1yU_WG@~i@l&y=`RZ1$ZU>&ZPfhYa(v`=_*gG+*DYL6~0#&_} z6UHFnQS7*B`vC{&j+u@z?An5dO7?Zdt+o~~am8g5w0`(0v+}#CTrnB@h#W(#zKZ={ zDX~er_w~y_nHQM3bFTQs3VO**n!3>IBPv%uWR&IOx-55^rN|8O=H-mPx!bH9(XjVD zYna15V`=xUKYsu3u@6#8D7^k3zyFJ6TZ6WC;Y7(r09gxDF1!Up&bRm>To~t_IqG$+ z&E|=dPJlt5`0Y&X1GMKgK`1y(ce3Fe+?td z$8P6y1K7|lWJ4gxTJ^<--=90k=3zh|dzkP?Vo>CBmWK*oMT7*bDwdou_9Smm@qhc` zR)#i|!_*ljH5{o*YW-|kobv;r~G?)e>DknjdqmHuy1Q|(jcoz7v(SObG5Qrb?H2s zo!IwC(HRmZ7#YB?v*Z*?e`8_PXbHYqfn>j(^VVUSd(>f?m)dVfHqzM_uMGy* z*U-8~V)y=uK!p-|bF+kAS7Zr&6I!M1(J4)yO=L?Nc`zwebcbp>sllWMlN#(<8ceSq z+=OeJFw_M>Klk&qKsfIOz^TPNs2mB%f5Io^9czg61K`vZ%yr4M+75VsgtvBB8z=`u zVGjN0(!;R{As>)X``DKMOqG2`B<@nEJZ$S{O(9C(wzP9;088ep976;AgAy78I`K>s z#`XxS|6ICJg6Sk}+sbmyI+;^po2Zv)f{!a%s4NxbkKg}$P3kd~WkQ)XC8ESvFe}Ek zD#CVw#HUT?%U%vRq*rt$?kESlkI7GNG@kL_u<|r55Ub0BwA#ji*Bv6)vC{4-THCk@ zF+(bL0OCyo()z>QW;%d%#F~hKVKWM&D15+PGKY3jIH-g-%NrCO5xV3%ST3m2F+*t4 zO($F?K45Q`6sf;gvY=Y-XiB%s$zH$y%Cmt?jXnlgN3Iw1wx7s1DV7$67lAa}*SvId z#4LO@FWu#*|6LQ}YtO01>_B*eQ-q5X7|-eCzp}oE zaewcrS({k4h=a+o$~z>69R9nO^ZOz&@Ovjb2Y@&VmnzSN1 zS&p%*6@F4SW$+o`OWLg;gB800=s<(XVmH=5_rnonrk;RgY##k;^0dloF;ny!( z-@JOj)XW3$vPhUyMX2_&R#)5kin6)5?zFf|^Ht*Pf(d2xiWz0IqzLU|49B57$sgw= zUkMS_ymj`7CJ&SbWHobXOXtALZ+@j6fT=lw-dzhjI~MGwkU~qrrx>S<|c4c33M0WtX|SMgGK__wQbl zX|S&erJM<+R7Dz4$AGEw!8SPPAP67w;4K^(;3i`}+tFTwwNx~&6V{A_rYsexbXr(D za@x$3c-W|kN~dzToo5!4#KfbSZh43KJX8o^$HeD)p-w*t>GT`1-)c`sIfCJJoHP^( zG)HXI=TrBd##{6u=AsyVbxShJbs1iz{Ycfy{Ycfz?nin?9C}SYyu(?8F;-<#ZlGb= zR62H@fK-hsQHQ40GwoKthMve0wl3M}PTfze@rN6+syP2KHR?H96?(m&v}5vFamuI; zRE_WMUS7_Y#uL+`-sEw?3pgiBKf%Tx-h4SH4;Xa8sx2CIiJ@R&P81WDJW%?n%J4LC zALoG$Mn^<*T##BH7iph#AjOs@89dgRH+Qc8Sf@$YI@(x@5o6ATIkLJ6@TVw4!20|+ zoYOGi+=~k@;oh~vvK6Lh%*sm|-yp}bDn>Dal!*vlfT2oLS&+8g9~O$*l91Ip_cgG0 z9zjXZ#NFWP;)O3Rd=ihb$FUN3=c0OH;6}rhN?bY$EU!?+VqX9i)|XwB|F%E;@a>PC zN+ZJ726*L#zaq=7A`t%d=U>nmOqNj~D%sP?xaEV_(u@o9fN()gu&7H|Jc>0Jg}?GK zE9m6M?D)|m8?Zvu{;DOHgb5xY($vcdaK?mQMA8J;*%ggBwPaAwcsXb7Q|{>oRH@CK z*0Tp*tu>%YvtxMq#8Cn52!nW(9TN{XTG2>&YBW5xu`LA%9AXf&k_<%=?ymxeb}k%- z)+{BXWM<(G2x3GzX~V}Xan%nWC@-O#dv{Sr(h~<0)9S8zs-R*3$z(GzR#vx|F(rlovPq>@|4hjkr#?IOEc<}iWPv=_2#?j#u1bHV8Zahv z`F>Jprfc$qsy;^kri z<{Dukhs6EFCn(~;M7VCcV^Z0+zNCz+-12^-DO8JYrI#He(07@6JCDHnnQXm0+4etF zCivObp=W};br+a;`yU3$%)0#{{QGapL$B~(56_j8rR(>w8*|QsmCw#dv-tVxCvYI< z!W^^leBZv~o*6QR_Xl~JcnpqW(w+ujg6+v`{tpBFC{erC+ap9Q}1U zUtpvm&#!ff==e$6nq7#EvWTjeWHLlJR$PMZ$?QhQ3mp8oV)#L;+JH&3Pj7x^Z?=wc zUBaE4*c`A+1QMPO?MVBt(>ppRAHRM3_g7@tH9?nt`p1mix!9vK`?X+!cAgD+i@r%F zG#&%dg&Qz-=h(Ac>#9SIxcoCjCDLMjLs12`Ub9cZ;!fkN5iAM|?6Jjp%)YIuO8RZ4 zPXx}@0hxk?MaYy`+57=rdG!_NEjWW0MC|fmaJ{()ykqsI9!QZhTYPBXsn>1bNqgsPlwF ztN1TTn`cjh(W{a_E&WF(y`8#-@@zx3{3yvs1qqNVj)Q_XiZ6oiq$L*`o4Q4|2%8U*?KA)JF0osWoRou#2pguL0-qG`$?cB|Z|$~6Ker~wr&C1E z;?wGC4Z^A-AngDaD~9lO>ca9XgX#=zyTt3$K3NT96Rj1Kc!5LZ^6_dmV5!x&?UmeG ze;IFpwFKvD^DgpgUOr6|D?CuZ1fBv0z+$s(AVgA;guP|pW`#RkHtZUEXJ$9G$QMO? zOR|tm_bI0P6cDkH`J)e%d6&eY7*xyiro3S6FDJBU^9l>3i_@OjSZJOLk{T-eM|F;Y zrZStdww^8_tZRnF@9^(9A2r3foAJq7#&#-)G9_qBN|13t=68Y2=Nd+wtHX2(wP4x_ z9eXiLhAALR`|B6GZ;5ovh9mje^3Jz;akIR0MsKm34wC(aM9iimW?|&}L)e>oQ=G2e zHq6bLzg(J_pkx@DZt@l8p&Mq^o-*b|^uF4KQ}tMe8?d&52N~|ZIEM9@HQ@T;C-O03 z%X;jlQpY4Pw=X#(^JNv(LAwXM-8F~iyr3C6(n7ql^Y#uTE4305Y!jH`#*uMuHcyz| z#33lyPisu0#G85?JdSVc+13HJUV(@g*icKJwKvb&NG>c!Qs$x(NmtByGyIMc2kR?t zo-QLiD)UOSwNr5_F;`Z{tRH}H1>#(kFY&1aqLs9voegMs)rih`G6pjjb}huK#5iv^ zcuY&KWCqe!3>(3l0T9T*NqtjmUB8#mXddVNnlaAI-E?5Ht&GnQowY!hS^7y$x|2hk zrB`>;S-OO@6-aZZSOJw5kh(ztTaIilppG#Pmk0@UPeGlu)#`#eSWs|>Cs8-aM}kg3 zOnEd+Q|&7RGEKFsSD^p}B#<2jWGpJ$1bjMjb({pIBeVy^PQe3XKzI=Z6)9irQ%mVw z<49L{+daA{!_`d%CatIGX8k`>MJPKs|eap3de7MLevp6oM=v7D5^K76OQ!&SQO!0fd&Qp|g*mi?v zpq(Mnt|qignM-*2i?L1JWCf0ebhhh`aGKSRr*Q2xy{3T-%+PcIUPmlGeHpfJm}KWqR&NK4#r5U}n_~vd{a8szMA~8e2U*B5gCskOddGJ5jXs)NV7h0GKJJNZKqOoUmK6 z<}bKU9f^N+^{8je#`)?M3o}>)t4^Ek#n15Gtg|iW>rHm<7X5d70ojX`M_!SSrI`JY zIuc!fj#=y=4)f)BR_;x7<27Gmtx*mj>mE@K+=Oe&5jieSZmWB~fkcrVK*=h?Bv2j= zC~aeK2B1vE?`nb4-YaLtK4H#E_D1P+foyEWv)ZOsIy@&XOB%{?+!dkL&_#>ZBw%F5XIL(tDEN55=&M{x2DWRIt#lq$IB;RB{4xOs%}7IXOV z!EYJsl$?O$G3*6mWq5Rw(21!_4N_#{4I`y8N zL4fY*qciu$Q3zjlPuZVUt?E|X*o`xFE`g2i!4&}?3!ld#8i9y>OK_wa=B0~nN=UUS z?+*ydEe=7FnZm8&Gvv)f?p!YrbJC99#bh7>v+00o+AE_lw@CzPt9oB2b6%a<$sL0u znxMT`KJ86H>hl_-m{mQK<{3sL-87aWiqZ(SZ<$)1Yfw5bHU_XHfJy+B0D5EqUB#Tt zDAo1-`VbWNNrGuPm~KhoS>@$=5jV}f;p6w18G8XCZwb0b>a)Lro?YH_z_3yfMCRf= zBZ0Nv&d}9kL53v}^mIpiKaAzcUSee0b_y8E&~xLoJ?syQ_9tv~hMR7NmB@o%zgVio z%=+7$R&5tk*fj~Cdl{e^S0BOt1PUTC$P zAZS7YddJ4!5X^>-NNXoGT2cW>1t1l`@%<4!ORZjKuAU3*5oA_7Q?T7;Sos{h7hvap z=A_GapZ|eOyf}N;E{PYu$3~Ol!{>D3;c&bU?_QBb7)HR&FA1;)*2T?SpUaWBe+KHK zFiApST=@G*^1rS!e?gW&)rG{!hoo&Pt3y=Eg-!5y%DNJQ>rGaunya9k-MeuuSe%3l z3B(eJ8v(?tE#C|)H;#pP6_{C~Z~e~YszJHJ&RoKG%b*d%wUuC8EW zG6RMz;oB~+=9G!CN5;siF4MxcT>(ah%R~aB1jdFQnhTz>60>ogcO@b=(S3RD8Pp|% zS7%}5{kGZ#p3NZ5dx6zFlp0&5IFn`a=$H#l}hxAU^6&JhrO>um0O zm(S(SApa;}j%|OC(_-R>;U_ZpV)3r8MC^<$I_8;Nyt~i3nj`IvN!&fX(<5>@> z5`o*&LACrQueSDh>ypAid-C@;ADPBzgRI|qKw}bnpTTkul3C0i&O9YO_5oowRm}3u z@J~5JF_m=L4OtE2fa&4J88k)B!;#LM%T>(X^-)ulk5B4-yGvDUsE7q;bw^bchuy(( z(R;FH_QR&!d9ZyT2h68(5)X;Q?Z)Hwz+Q73Q^L8HadrApiF}kG^v7M@Hm}5uF{s)y zW?(nIwSUv#vY2iqu~K4X6T@t_zJZs(wq!Sp-X>JqRqJtW{UTyuRlL2VEt=^L;OI=} zi!cJIBW%SUjb9TK&R3*WXqOvc+0*Bw)oFl%%82EoYi8xei%gB{73Zj!@asV)@f7zx zNKQh&fujtTYT>|)*M(CPUV+Q}+n;k15yZ*3a}38wyr4X&Sis~<4F^g9udS^OB{T+D zgB`%}pb0Ux4K#lMF5Bj4xGk>)nt^jSaFS$x!Z2agnWlxUlS<=kX1dfEQe#~A@M)Gn zM@h&cm_{M-_%38`5@u(BEtMBiL)2V}Q8oS8s6GAIoy@%e(Uf&5aHcMo&*OSw7)ci4 zV(Ay!D})OSFk{Bhz_7rjg1HH2BA*8V#pI?E0CiX96brimvzrB)?sB`-7^7R2{S$kg_Z$3Br2V;S zfH&)ZtH%|tKkUb6cK481i^74wvi$tH7smuMQo~^1A!g5o_N)=xc0-tr2m=11@D@o5 z?5bu;Xun?HcRo@k-FWgG5pN{MpU+ng!n6NupL8QGa2R9}6LGQ}E2)D`#>b;;j4=#|qv&DXfKlcP*0yx|w$KiwV zyHj@#R&BMClL{>>%e$q&g_9bb(=r#wUw(K)rf`J&4a>7;pk3If0tKrzCw$pwle8s+#EKhAqQZvwK z10ah;YQT>mvkLbxRgxBxsa5ox%?p04b&0Uu$LoT(SZ37CVh z6ZD@;JhB5{2CG&dh;-tOy@eYjd&Aj!rH9>$36o0_yN@wx36*7YzGkcGAzzTu@)6{% zwO=-{t{e-@WXVTeV20@pf{DGD(9a9rf;9;!zpg(J;{0Wgi=R5gK> z_~y&s+!Z zk4wTgvK@Zg(s0+7<4+ri3nG}X68h4#fc*zcw%l~L#rShntISGivkvBDgk0Y#V*B?g zB{=0pvcizG1ywq)v1RqA5!Tkd8d#3_jg4xyX|KbYH|LPNU&0Adn?zzWLY=y0dS-1w)ta~!wQefMBHxCGg zJq6Fr?q|MJJM80%!>addD<6e|8+}I&rxqU^T4T;B2Ti zy$()Sd%CtIaTm!3JI=x7XSOr0Pu`4${g$|&F=~%fZg{H-V3zN{ea|{lXvPkMwEnTW zy0(h0DI&rWG01;6Szw};7rv+wm^2v*ItDOBUMx~e^+CVT=zD~nYcg&ym+Gr6NEb_B4es{g8d;uKt`5_lR7JTYY+cI+OatwNChORTBlS-yhevhq4oy6HXn(g|wB_)8U@T1ul$<2)~eaL?=rC3oy;2D0~2^>2wkPKD;Bq z;lPf<>!vECjcobXZ$7jN7qd_Znfl&hfnBX~cN|>*qiT}YFYxy`y@2yF=c!gDx8gGz zalPhiD~IE(&7ydxF8K~97S6pf%N6B2i+%)b+|?z29Vg+4M#Tn0xI$(Nu z3lw0;QeEfr((bQuQf`$Rrv{Dl zZ(qMCeZA=#XVv?AI~qsdTi;~+*<^h)5J({Pjnp?)^v&?B`X=R8sc-7hH^47V*Ei{w z-<-Zt4^cRLLoY6cS}Gi=aJGbKl(Qf!+nUV@eDxC5p zw6iIklv}03sYl`1eY5EbCri-0F@`POt!jYjF3$-3E6L?!jFKcXx;2GPneH3BgGS z?rc@eCDT6 zP{Wpy-sN9X-gEY}4*;>Wmgt4WsIah0yRixoIhbw^WYhuD$s2~dac!LEkofQnn&gz%6WI5*0Hzf8nueQ zK}q%6e3EKh=sLd0^Ga}%x?|j7zO1+5?CEHDoT@v=tYhjLg?X`5soNBGKiCp>1slfl zhN45l#u_nT1FPkZz8v)0-O!_M^mkWogn9hFx z8IHT1EhlB&rMk)c6ZO3tI#cO+p{Jdy1RP$CZB$dFAQ@}eS)kBFaDuY|T6`hfIo4m^ zdzw8LGDJjGb`xJv0-D;I)btu@{}u-GbB)zuuj1z)uv7G|KW9fR+WwWlgi@@gNw8*7 z!}UJvOAZC)m;?m*{#>?2LngIl*$TfEHd!lUqw#kRrup{XJw)v!xciAo-^!~Z+3W?#5!85b? zVMG@GTRv5Y^)NT}g(}H%GXn3rV3s=^V-_5Y1)`cu%tnmAzOUaU)5h>ii`-0|{bWEV zKEFPF9)dW4vi&O-+cR$DI*LvzVP|s9XU!&z^?@&1B<(7?n~O=_fX{G#G5!)UHgI2< zQ0Y%O>m{B*t3AXU-Y}facM6U7$^Y5p>Y6;vHAPB~3rhOv5vD`4x|c#|Mx~hr`&MJ| zD#RjCRn$*W<-^WEiEiXG;?49uj|i9j7QvxAD+%S43UeMynRwea&&^H{iG)iXRq?E1 za|!elnTMYL7}pIB1@-COW2mt5!x5**;m`N6gU=J5F*t2^s6FZdrdNHBIT+u!Y2v&m zB8Ns9dONNf`3>)`2)qSf8kiU7pE#KuMG*0io}NuTaaj> zRwtGgsB$g2!kM5heNAx!ebL3{p=5MPq07k4((%u$517kup>KL@0S4qV;m(h*4RKKQdPbV5(3|S2VZ%0@ zFX}EdQ&BbIm8rcQlB+zD@QB9>e=#!H$6PRNF%Ulhz1Gj=$++ZcOiO&$rlZUi(7v7R zTrtLI+L2eV5qoDX7PK*^l@lW=i(%YYj8LY_JT};^vIJQUS?tVHxZ$$f?O$FNjHQDNKupF(w9a<$Q z+B!=@IezSV;{t&*W6fY3{1!lrHh1-PdxiBj{J;l?rMbswGL?SKE1r-psI^$l!Y7Rb zQDXmU+#nOQcLM7Te{OwB9iRJTs^7Vy z3?HKjA9Y6$Y1ED{{c_~{I>7b4Zs~s&4t{92#ku(4sjIM__~P=jOoirfP>t)G6PLL$ zPiA>&SxU7U5;Sr`Uz`zcGZBP`(=P^IxbkoF{c-wT!^^ecyA6F>0Hus=R?_QLj9Iur z*%5#cU&dX2KYOLn@?YyOVq?$0_@U3g1P@*+5oBR?3O-?&*=mKD-zEq$_db}5>}>Yt zemLxs`hc1Sp50Xa3~H8ft>fh3j2Wm~-++B%SGR-9GVzmX?=Ws-vOSPo*wvoL-8-7WDMhlg>b2JUA z8vN`;C2L|trolgymlWt>=zX4|lf zxPX<;%cy`Tf4!ZuQvF8IG7ZQ=}sO` zu#Nhs#@U3(G}8%KWq^1?t{ALk9e^8h{~PwvP2RZDW@Rj^4mVh#KQQ8Hhdhw{oQ*Lj zVl%{n$RLDmbzc1CX83la-xToj7Y)G{sk(K?=MNj-4mVQDw>(@Xdr>t0_6t#KcX9!7 z&5_*H!|wqp%{N!St#Q{$&nnYRe`D%8qaJ35Sbk?!Zhdg4MWMS85_^@n&{E zr%%(qFr{t$B{;%lg?Q;+GiKI$Iv}g0iR!Q7&aYQ|%TZ{Q=CIt2cUzwpB6z3NUN66D zY2vEP&N_gmrZw#W`+ziY!llw2nrOJ8Jj#NoD{ zOI8=TYH0A>$wTty+=IS9kW8Gh^dr#D{3~QN=}#DRyhgx0Pg&1`UbcL`+eLbxlS$4T zNzds{^uA!0@Y`42fU>nPDgb-M;j>DcPYTQUTW?aOAn!&!bVpOP@ZTySB)Xj*+os z9fiWPM%x69&V_r_zvqmOL^b|mupuj3iMF4nDKI*C?L1V?Y5GtFVtu#ox z9}5>iui;cckS*DtW1{W*eN3U{ugM0BBb`#3bELexXw-O(trO>BHppCD3iAYc@-5*n zK??$nhn=9CZ9G`0kY zBstDm7Z(d|?|G<~LPkn|H=`@L(?TOYrmbGS2-j1mLJ%qDc*sF0DCL9QN-@11-LtvIDsu!lbZEF{t zl~0A3$vj8)#dA*XhUVCm&hmSRlo-JwYHmy9QO+T-rB*S;NTP*l>R?569p7{CYhC%c z$j=do=BvL$uO9#AlaAQ_P7jdu;Hhvk?GV_-9<9cGsL7>%#eZ2hC(fu z_Vf+N$mk+s7AZm^O24yK8WKwCOZ6tuXZvz?kagzdn7~ardDnzrLS}wlT%i!)TSRTU zC{#NYB@u?Uwnr)XlfPGIr>UCM$%QoSlT{-aLy&L%v)^=ZhD!mNSDCB!_ruSd$B ze#U$TCYNUVovG`?)HrSNCV$+>t%fW06+GGa9V#CIiYwnH+Nhj7s`~}qh)=_==0At% z@uE|0&m?hcl<$z`B(9Gd?4X7_t>`?^2SkfmFCK)1^(CG9#oTkET(hfOM=@FX(^jb4 z$%ID6@dC-w9XJ)Wz>Ya>kn1U4PE|r+dZVAVNoZiB^Xptd#hX!Za8c^-d$MG5NvF0x zbhtdQe?~7n#trAs=OJ1-F3YskjMx@FTg>~4PJRTdXaY@m4%9A6vWRZQS9Wp~dpIMtc0VeZqrfHPi zEAZL1mAG%-nQkeWObZfj8O*IL=fx1YYGN)^+r+uAP?oBU=36Z;oND1sN(C!RpIUcX*3ZcsARJI!+3vV1&5q^@EsHIxCPS zlNa)-99KGKAr@n_7rC;}-(MS?7aZ)lG%>du1B@Tv_oalQp6c-GEtGftNE{c*cq8cS z%T?TjupD7N9l#6AA`$MnZ-=YUurj;Xn{#vHMY`ZYnN4x$GYgwjg6;jT`s-kI&PpaE z*7=t(9COrglQ<1%qQh$m1~{Jcj0GS5SP3e;X)IWorO7Pa9v5#kZJtejyY$q=ZxkAD zW&T+v>Ffbao4ORKwKBgVIJb>n#H+4~B26p}j*k?ru`7w-nYM4Olg=DCrRT{jm&N*h z%$_?YH?aYh9_%y?2aDTUaiz^tdvw8%Z+Y*q`%%0rJAY|a`6f}@_-%G`K9jaFy{US= zZ$ktRgPCD!Pl&bFABz_ZA^Y{rQe6QLcY-O1ppj&MUj5Z}8r6LzXdxm1OUd^~^aSep zQF1#-`)KxWtN0!S;+2OrR!4lV^Q~$m(rkRM$r*RDM9C| z@ZO^-<#>Emj^nF7QaEn#ay2Jz`R+Z!u5{`$|`i`6B;svGLR6W^5@aiWcmRnZb^IeBeUlb4T94O;;>@hBH< zW^ps(=y=~;v1IBnB%FpNVCG^<@0olyEL2_@sHVor2ALs$MwO@Be~_z25s0Si{0F>WD2c4*2{@MUq!UoK9F_>~mtopXqY z(DB(2FA{oRcBoyoIKIpin56j5v`@9v5L8Hele|@(C57%f<1EA1$pG#lkPw;T(BJSuI8Ni_`@Y$Qw?rvG#3uT8_F7zN^toBk!#=vw zdd)wH_p?f!p=W(A+o-CZ)&`zIL@gc5!&?^mXfi=5f9oHKOI~xcV~y-+PjzbWTOQ4`oNh5tn7N6Og_^OZ)!gmlp~JEX^B2lUSbsyWeGE|V$^%g zlb@UpzW+0|tmp~f%j~v3oizeqSC2~WwR{cek+jH<`}D(;BCrSk>Y<4dS76W=t;fM{ zpi0i+0g3K9AP&PjgAiC3w`ts>SW>x{k~tL8%%!W5u}B8_7L;bF8xb+^Wy zNEY8n>hjB~U#HY#dX{&>XDNHzOkn<2@rGGS*+%Mn#{oA*4V$&ih=_m`!-a`Cdlsn) zS5T?$A$WeTTSi|^^$NL)h1Ei*mK8mTI?IAxXSN*i@)8LDj=X-S&uSQ(;?vRgmZ*gI z=-cRrWW<()5E@p0I1Yz(LRwMi zhq@26Un(^YI04^PLg>QhB_;}Ai<&R@bX3Rns#AA(oqdvTVX2`+@}~WF;pItYgcsI| z?%yzpRAU5w+d+A>ufGZ4K5>C$=C(U{xXdcwM@58faoF_T4)MHG{=@ov#2`ZWZ%e)Y z6fJLTsZ`s$lo!_NpQhb|CgUWrH=$hR?Kmn0Pc`=Vf_W#%Gg9)MYJu7!k3YGjN7CXy zQy7YNhdhl3`BWF}0>XXrj`sED{dVaJ?d?iNj0H__)S3JJkmV8nJS2+>-KtQ>VfZD6 zm{OZE;HT!UxFe)J4SkoozeBmyGRb+T_qVlI3@xyyJ2Z)|4jbb_C)+LsT??6@u*YSr z5q|rFN4>WWrmp-c+({S?6V00&o`BFn>{n~I+6UqOL%hH2&!N{6F|>m-(PV zwkY{uR(!PK5HJ7cK;4=`FpY-ef4-QCSj{8#k>sM7O8>1MH=SnxLiDD&(J{mGok(su zE8EN6v;U=Hv*X_n^sp^ae@fH>$(4ySTlb5unZ=^iqr%t^q|NW0gsYajyhr>NoLIqv zPqc`81@H^00Sxf!2o-PU3lS=;)lonT)>ln^vycUo)sOsPPXkEu-n2n`Jb|U)JbF2< z4a7B}8rLoBA#E4w*_WNtenVJh^xe7BPoc!}TR?r=a@`>(mkYx37gd0 zh#$gpT7(hK*K{HSY=Dp%?iJ!05xL36gy5}kP{-Re zv_0b}@fx<231@zDAP|}-Uv{;p%?paxEo-!l#7gn^3>VnhE!O!c3-S!h_{TJVSX*yq z)c?x%ZW*hEMC#w1G@5^6t;b!hx6I?kJdU*3k><8P#0mIr+Rjlts zkVg|Y>SA%LZViC~;;DVEw}xp>ux!zf9+l+63inJolAgm@StOqi4U{ZaNk$)@U8&jJ+<3%2Ho7lT6( z_0mwL&i!(FsZw}2^?KHa<%}#9Dx69pHs!u34eSyLVRBP?p0s0K*5uK&-WgDD%{wvg z2fvRMTlC+Ad=2d0j2V249NWEAf1f)yQ(G1`+2~GZfJqu#ma5Zq3aWM|P-}+*IsWFi zAAHJ~`>BCQoPjn>nqMJ4wEA0;T^?fZ+kx1u!GV0qF3WBFS))H~$?~1`BOqjz6A%(E zn#xQeC%TwG!5IFjx%N-Mr$;U$^lx>BS?XPrJdCt|sN=A^@waRgKMt#AJq@-${w4A9 z`V8+Rj7#x0d0^dh{~j+-u9%&_UZ8WsV1u12t^!uZ$g8EQdI`xH1pR6tZ<1lADBwXB z>eOk_7}hOV?NI#XTeCBhE>kzHz##7WQ#E?C{h}YC9sqLOC+z_{7h_GL#O6~=8=t*& z#NBb&3fh-t)uKx8;&Bx>MnCJF6!y8?m&%IHU@8oVBO1T|PJp+LB4lP;i+~~aQ-AKf zW>S`l2##={CN2p zTc))CQ9{50ZWwt8&eAqdFmqbQ(XNp7Nqp0a&#K4Ep|V#RG58xNp4O$!#lGN*NqnzE8F1U<4%%F=s}wUY zS`(_v`q8dk*GU?x?SS{gcaLa!9?kjB|4N9c1T=92(Qa*Auq!=iT~8?n2C<(7Zqn}3 zMfsinD{9$&o^EH8vLl0s{xNm?eLDWl58gX;6|ofY3J z%M}<)7P7D<#}Mi&QIxmEUG{F{=TW%p#k>7WR&V;~sT}JlfX3yZr^dZ3h~&~y&QJqR z|6|75#N;h{6rhvzV`PT5!VcTcyej=KubxRXmIU{Fz95(D_H2H!!tmTVJnJo0SlOR` zWmgHrw0TB^8Z>*fKDK+JK1e&6HZ&PoZ9#?ps{j*Y`rDX(3(jeiu)BwMXjo z>W@a30uY1Sne3s?7C#*o2LiwcLAW=%VwD6n=Ic;h?Jkj2Q>2tm8j=I2PPm;35Mfa@ zQO9?8&m~6nGaKtPfiO9(4?U!M`kjA=LZ|J4H|fh^-{-bx8u?DGGP|03UQ5V+ZTEa! z6@mnu!YU)`Oau{a!1tJjk$c6NQ~#>|4i~leuHErUy6q_#o6Oj-w@2~pFw5uZ=Jyu7 zMMp_NywYp>yW;_xci5qRr-{Zl=h>O$I35anUXL%sF^a)uX0tCr_J5^+?jcy{36YBX z!J#6K8F2_=8^n7g;|Z`{-tD*H?f9G8YK@2@q~Q2E?Uu{)%CYCRz)m< zE~x^_Tt4pt^8lx1F#O?&FH#Y#>@_2L&Xk}O&2D9stpR|B{XbySp+ZNGs42Wo)jMZCBAf9+e#1UF5@ zyjUZJ8t$8@-UyaUPgO|!B}e|UjQnNsk)ttJu}rYRLNS!y?!jyU$d2^}NY_prqssw+J&3NMK-me;Fe0M@Bwux>l zhQIzLe9GZrUL;Ezk(^7USlXpFdzK^6FzFNu*hmSWb2}LcXUeTbtES-!2e}H&a`qzF zY%!}a8EiO)+TBc=RVUkNuBX%wvy)wCzq`pDO9=OPjHe#&{n~`hawoIYWW3d&E{8O` zis<5)(J)9>QVTM-ooOzB(yzp+Sz{U%N)V&E)dy{Decl*B+BSWrEi(D^J3D!j|H*?j zj#iE~r<`!vXi+0evDtK4Cx1^LKGMUj^YH;mn;*<~#Mx)Q-(#4z$QQ)&pmAJKcvRpl zpsga+m2Yy3H8-O^@W%QCQN3FIgPnH2LRpc*s>ZyGc_w8X8~aa69By?=7S;SL`k$1b zZ+xtML0sV(V3L8pnZ-DWg3oV0)9iFf_2H#`S{UMpEUH!@%r@!bAISl773x)i3!32r z`NpJ(E=D>xyofBCweDPH`s!!4_H)HCACuwJ7&Sjb?-5xL%MBFYI~rf|oYRdT+c5Md znH*P?FH}jlAnU1CwQHo*%Pdt;Epu`Ix}aiND%Um}+*`LkA$~_x9zBKY^zNao^w{0M zEV$X0v4U@&z{a!T+j#Ry5!R~4!oO8a|67I9|8EuZ$i8-Y+8Op0VlI~CY}?Mi8BBvy z)yWM>y+2-uMz>0*sIYBPs!OWqhds z{&+c&L97DvXf;u%(0wU&ou_Pp&bQ8E}{IGmX78Z41cc{Oh2QifzD*nCO8|JJqUVJfCr{8=ON*H$AxZtkXc zK#ji|b0Ji>NOsc^Db#bv*E-@k_M3(|zbMNztF~Bjef6;VaHi z(SK)yEQi%Bi`$wCX1`YOo!Ug&;b7`=<0}9bz1bNI`xGpjac4DZnzz=I8xNb$oRPk~ zyzEl}KI4@yC>AXkv{CuA?BXw;tbdRAcb#H8`ArC=tDpWuoV1hVHtB2t{S(d?sYn5t z)erC9e_PNzFHg24r(H#ZAb_Ag*&i>}0%na);U`r>Nb_I(;*Zb1b%D8dY}D_qG`3o> z+{&~N-STiy5Pj9|?GXEd`KFVZ4!K-+Xp1Y4`x6Bdg+w5o9|c2|Xg^C&Iv4y3ofqU` zMfTmOQdOl}G1i5XkaCTJJ4Gb!i27YZf`Onz04lv4Xf% zRF)Za$q@rhzj>T+D=>!LVp#@Vx1!QN5)dZwr z(d%JZ&vP}`Z)=yDrK86qDk}XuI*3%Rt(e>v-+`h&bbJz4g2SXPHum)_=hE!2l) zqH&F8|L6Z)@XjD#@jv1CCe`>DvA^TWSD)7Es))f#{%eXQ+>#o4N8Z}AjxIMcySx95 zq{i2H2XFoMMirK&m?U9H-c4EEzuD)3_`qxBrb>>2c~E!vMt3o~9;eO(Rrk*m2qEwWyddVL``8Q#g0s>nnx_4MRO_Q&kJ~2|$ z&*EybV4wD8oqmljh70z?VNHp5_yVW5J@jOQ`rnhNw>< znQd3}Jk5W!Iv?qdtOd{!@V{nzB9N&;|q=6zCXiDj5hn|kKb!NkQJbzQMP4m#mw00J< zEq8d|DQ`iPw`KE(i|ZYup#2`RT09x`b>u_b(Z~vjLxm@NTIYEI$$XW4XF=#BPS%b@?0c|nz%2Qxk zO9pK81A~#K^?1zzj>mU?k>MWiP}M3^dQE(lwt0niOK$BKsAbgMaI8)kTBUfP_E0Q5I$i_VHpQ^qH8-%~*@H2U$a+_x{`-%~tTFw4?Ei(|Drq;IMMD*X1`^L0^U>qNk1f-iZrrp8@)6e5J67+p!6@Y66+P9cpuPc~`^z!6=vJt*$M#UYpIP%; z<~n5VdNO&BEe$LVn*$QWz?@2(IxHQvH?hYyVFb7#P*}jiK)J`p5?jbrd+$>CpN=^- zxmr~hhxNLfDNjs`oG4?G)Ru;&cYk$Wvsq@AS%2`>{S=nJA_l8n+5G96tq(oy}IZ5GU9=!@A%B;N(g6VYAb*np1oxHVq0aa-rvvVKv=Fi&zZ z^Nr@vfHhTd?Z*$qR`nbOLSe<3Hv|LonBJD@h-wSf`PceNX|QKZ$CBrV>7V;Zlt2pp z%zHH2A6?)jqbXvDwLel#rdcDgS5W>=1FDY@cZy0jhm{y-RIr28)%k5gqn#=V*eku2 zyF|ghXJ(<%O?dwr!kT7LNgWJ}&k1av5z$=4z2)<)SNZmq&^j3!c4&3#NY!bC_P<9y zn}<#m1Ku@=O48=bMioa8u+N(CdPGFCYT>&D5afx?kywh&*(yd%OlC0J@T+bZXm81C zQ@Y&mM=~WOEJidbD(X{GRVM+LMu=_0qN@)(iXj;V1)qD^sLY~C$tfCJ$Wx?7svHWd z;uD;RwJE3+!YV7`6KDqja%PxZqR`{Y8|84dx@s-NjT!qEacJ9*ckCya%3dK8$Oj|1 z0IOiA&N69e=K>|9nY`S9y0JT@L3;mU zk^t7X8qX>_t5aH1hwgwmkxe^c!icrb%@kG2ajk_z&PFyR+N)5V-R4oHT^4Nm!5k8e zIW)o!u;S3ughX?YOli`ZHn0O4Ypsw4?8|UCT*@|W|Gy^;+oi$T4rVdY9O|&9Dxv)k zY&o?Y1(HxRPWBA&N{1jZL!{Ml1;xS0dNPfg0afD|9(a!g!fuer$xdW6q%|9v0E5A= zT#Kqu=5bG+lbvvJg!nn%`Z(2IyQ6v~%r3i86526I0%;a3)1zu^_h^umTt$O5IvAR| zR8-ZWE=wbM3wheU3$rL0)+ebY2!f>QR4$nsgZnw4N#2chd2K2eD~*tTX7?wT}*P)}UjZIh_^{OZ6N)#Q}IBUol5%WVd9Us5Gad49eIdh+S6??m&GuP zR?i{U_|H-`{#hz8ULzXj8gAr<&;eqrpDill0Y;(Go0|Bhc@RXzrfBR}UW?F)Z7xQ$ z#oIjV(24!cQFH(dH&_G1?VqEY(x(X42rjwhTzjvBIHYm9GsDqLoAQ5QsF-(8#2YgT zYHXY~Jj$4IHi`)o+Zre%Rez+f;n`34KiQbEZ0Y(nqqAAM0W*sw`_Y$5UrYtCM`sP-Hm?D zbW!sX&IQP|WXFXwdrA8!gZ$_HyHUrF5Y$L)sW&fG!K$jrfWfms2l&`)-Xyh6>*AQj zW{%M|jSm#SQL0w}``pJLw86S;mPsNDj(T?NWhALKNdN|9g>2JZZN4E59`?zWQ4XB~ zaiTBonewiCc?!r2=&NQkv@Yv*?DX0dmPt-xnb9g^Nt&Cygq!@XqfMZ(qVHz3Y2xl_ z@;_BYlP2KjS>JPFW|%2=nQ@I0gFa($j^1_4K;2c@WthJLOaCgetWfW+*(O53@oUt0 z7U8sjc{IyJk!Hsx1@2DeBvKD>B*)GwkZf zFG-kF`2}X>4Km0y2CjdP9O)ysr&%Is>zQ-Le3trE|xxklwLMjR3$$Pu)F^+=H<%!;sG>qL=- zI=!R`Wj4z>FO%|6wM9#@-$HUFKX7R};2JW`3K&sf;{F?&7P4!fC^D)lkre$Q!wf$> zze$xJH!)cRf9llzF7rWSF~V zjhTG!qH2#L*kcHBo#NlBmJ( zNnPTZy*ia?PWplrT%uB7ypM`b=sJ-nB;u>uX^MP3+CUBz^Aas^5mnamWt2%=*1Xh% zNJb^`#6RUgtP&3)eS!dv$vj7%h7tfi_BiKcv_$ZJY&DkY@9H@gJ(5ZH$tIm*am_F8 zYx1r~d3MN17!TkXZkJ!az%reR!%P4=@dnxluqktO!jxR!`xu}?3h-zOvJpku<3#Jm zmhZIx79Ai;5ex++vqEBkgftVXF+dKksrbS##1c@n(kz1}r!XD6tQ9PZ9w#u#dlCy} zhCC-^q=9p;sXQaCPN$O3WJQ0n)Fjcs7At%O6(km`Own8IV!eCA1{5M$nS6*oVxJYkLe2q`S7aBH#3HuS30Yl^JO=0sU&yzpVP%tb zk=RLq`yVu;PylB;y>kl9Q6*Y7FkE?58pM8>C4!Z-!n>^lqZFT|Eg*-taR_-vtivys5d5qyTp5U*>=TJzueSAsdY`h1PmX4T?`U1}!nEYLM zqBka|B3?ESMN>M@1sO$6r5Uus8(d+2GzH*wpF$m%6Uxdyi`lsz{oBdIRhDsq_}ha5*JCXHQq+jk@wlaBGTILu-$crH#i6>8+e$-docw7 zF}t}=jpVyPQ7MwypO1oBGlG8b&VhK7{$xS)@M*uh(7JKvyX?PZ0EEgSX9LTO+4(I5 ztkaA6EI@aTXj_+EU_>Upb;N4dPo9ed_NHGyCd6$xP}ulX+&^9S)K)ligvD zlUFahht&XwT=iM*LLIy7Z1KdcWBy58U!at7eVzj{5;}VFPhTLtGU1<071OQg#6_>2 z75-6GRZd@E#X>hk=HtCDaE|&1cQ8NuVdg)0V4Y#`WC9|Q70Bg^V{U?))TWx(xaTZW zaP{F=e1Wjy?JD~5*NBsi?w@$U9kh-ab>p=(??2mGi*G{I8y(Cg%wr8N@C0`lk3u2& z;F)!_>A_Vi<3}W??=ip_!cRX4;x~Bk^!Z@C?+b$6qMai1*u|_!l_jybn&A;MX6K4* zVzsT}M*pABfkH+l@*DXrR5kJVEU@zv*~-jdq$8nibP4;g)IL&B#3$>=aH_c?aLmOxB+mc>@sy>X3|dHRO5T0$Qui>>-$%>y{P z)T-1VzV?qgmYIF!|ESaQA9d`&`t@47F;vFO^|z;t)|plC6>0@jYRCFbQ=B*ijf+A6P~j zTV;dnHV~u_6r0Rz!@w}ITehU7Q6JF%f3eXm9!$fA4P2LJ1M4mY8@r-7wg$43BR>p9 zdC|I=7p58Hcb()=pQ8Q_{xv0$V8bY5D0c|q4dDw@8P2mnCXfeDM~fphf~GNV1%U;P zkvqx7uEIKF zM`Pi4lJdpE?>UYiT>r%?up`sISY-!`5UKW=+CgoVQdM?cK?#~S_rSd&q|3lG}y zg^P0z%LXZGE^wOa3I_3^(Y(Gr=1_o*u167A10a)=SWq=zf)Tm+J2ZLFptb_I$S!*laDw{76r#`b#jzhUoB+! z51ie9;5Z<2o6m6n1LxsCaJc`0;~G@0Y~WcM4YqG}D*3@c(}B>yZ&4tJQVpcnn&Asx zgitwTi!7*0olH1t(oWPv8?MP%-dcn}cfqR+7PHmB??e)Q2lB@zq6ZTc`0UuBn0yw> z{bue{mdUJgZc}Z4TdVkNvQW)bWqG=;F2Q<9{2!c9A?Bi#f=yF-8edK@8Nzn$vPFjK zJWu2ed8Ow<`3yW$HN2>*)zq6|{#a%(|Cc?_pjCm}{x&*V)6}*ZLq#fsu7z?l(*|E7 zi%v+3JasnVDZyi;jztEuwmW70J;6AKtm_mR%q2_#=IP2{yqU@q(r$Bs#$*G+2|kS|o1 zbw&E6@#(xcc(ws!&_}lml|LSU)&`Dx-UkRVOgN3P|Z`sE~#J|GAR5dJ2 zf&L3qk^hA$=lBlO-4tAdM5kXNqUo?QlOwo^C?s7eY26h-GF) zUuc3i0rRYoE?_hXIDZrHsej9Q!HTXEWPomi;Pfl83%I+!vc4#VFA7eHN5z2=lK@7{ zpdC*b4On3`0ODh-x`su>v`EU{D1xr%A(HIwvdsP$sH%ZH5LloZgMrG7(K;0^ zf#IRiI1|=+8`R}~-89vt(;!bdYJU?`D1I>U?my-F20avc9$+3Pne$tOU4)2RkbCbW zgGXmI=XmBU?Qr)Jwzz?#@9cnW6u~V35SG*;S8#=Kdke&E=u37zl>XCbl4d{m8rPg{ zOoKO)vU?1!J}=A<{PScmEB9K$gzZKpghm!y|C(`OA%isW-&8t)23bD0G#Q_L(oE z_17PaZ@niLJ3x<2aW{X?+c#A{w?;Db51jZt@OT!cKMiX(LzH&+AG$s3A0Ec(Ds%hw z8jgiIG+zbI9Nt8K_^up*M2%jaG=6vMpC2CdX_xs0{7Cp=s<Ny_4RWihr2RQuM#qNz7l&? zwBPK+wrN>&igEVyTJ#^RfySMWSLRL)#_OxaR*=7#5Y5l{;=C%R^zo7 zkpPi1-ba=Un~%E_+{?AiDjHsC>S#%aoo)0yuNf*1m^rq8x5eD?@qbtIXJB~x_;Qj5 zCLY6<8;5ZR&j-yH?tEv=*X3(QsFa5!KnV#7CD6@_SQ@FYFarWR~0NU#w)mPg}Tw%1^YP6e09%;-+td^m8 z`5lCR)H3MN{h-M~-~C7qA9O3KY<8&x=ZhnfJ6KUwt`0%+verr&{i{1vlvFx&qLm^8 z_X)m;t;v?5`oDG2R*ls$Fg7zFk`fuDT!T1@)YUP7j;up{LM687*pQ$=9HYGci>mTH z2x5@6R!(_$J$H0h>jKL(eqD0!EU^w(`XT^P0Y?Ha)j}&LMJFdP2IKSeSBAuHLd^_EZ!*H%TOA^SwdZiXke|n`Kjc4Gw zgTrPgw6tTMK_#v*4utNsE?O2>>gTcbCd19eYjdSTL*ProgU~QIm}VlGXTCr~i7W*K zde_H0{^SJ)f5OJ*HFt58Cm{2ILt*}J7;09~0Kz1;on^yMqzt6=M14!=IHc8d*4`^+ zR^z}hB%OLBl99H*2-Ow^P6m$Z7%*u*p&hx5wT57p2L$ny99JzLV`}Mnx-BBA z$TXBxECA@!^|G&d_iCrYU;~8Gqi9DXXykeeDYTRpwr!D z{f*f6D{E$*^Z;e`G9{Ly-ow$nAR&2s2cgwRW1Tbt=do7v>-^N8M(IK|2BA@I&%FM0 ztbPwW3_b_P<1ZTF7W^o?g@i3HZh*bN@26WUPyF~g z&|d3a%e+^E>Jx`3IH2P!*unsz+{%|rz>A`*L7n@UCu4}y)G_+m{)X51q?j|miXAF| zE?s6Ebj8_!x#DfzP&l|hD>qw+Qn=UAd&k_{w%qR+DPCShL!+&#rA$w$KP>x>xRWSY z#d?J-`Vva%tWC0M&;>=uWgPj1uG-(&d5V6aq*=>db(PINs3U#GnZPkCBv@Xj?{FC# zI7yt@eN`m%>oRY-q^WUMYYbK%8% zwT84SS>a$@zO_bRqp1A532nDhnfe>i^v>et@{C=BzO;5a=b1|0zf^NeKQ}gM`slt6 z(f#W+r5?Vpg8_H+_TvaFluOzr#U)|1a4`qpKI~Nu**V1-2 zd^#VNELt_ULJJCdbTs@)9^QkbDshHw%sCAh#l{C#rYk3#af7-uHw}E6;wGeUZNB(b z00Ocw;`F@CCc0o#zg~qm>iB%hVMy)~Hkxn|HloA0j*Gvpm{hG2-x|CX zQ7bcHjTJ8FzEaB`QiT|>#f70 z`oix~I;Fc|=te?1r9nUxX6P8YyOEGCDFJB_L}7@bQ;}{E7^IPsZt1!QKi^;6d;jn` z&*OQ|Is5GOuC?AhoW1wzUx{1F(y_6wFQ)I@k$jC7g0Dt)w10JV+@Bo(S-+tB;vQ-k zvQjy<(?tRuLFyve?QVH~=)8V6CUL%lx4IWvd4zC%({#NJWbFLdx!9<0Y0S}Dkmvc( zk@qj1JM&z2zRp$#<-O0Y5!gp2-`~|{f-^VkY5D8TH;wm#0)Ou1a*^f1=a%n!uz3Y8$g%oaAGZP4#&u=w6inHZ-I7Ad_fj;+*N9MTMhmFumq z9HBpa^SGTc-d_DBi4F)QT}Qo2$`A6)mq5Q#PjL;NsYAP3@=hgPm*hEF^tJhSx*$l1 zUKsrfC&ktCTOHEXl79ejTHs{SQ|Eu3&gc8@>M#GUwj^C2Vm7NU+JPO!j4x+l;2YivCQMC48^RB$NTs6DAX>*Ye)jucV zUaQwMq#rWKFRsnF(y-z?(rmSGRQS zW#;Hu@|ZDVP#1rhr#DIiWl8y4+^-+r?n`GaZn^i>nRGEZa(KfO)UDop;r5ie3G;w# zeJK}Hr=xwNsZl1?qeybQO&1oC+F9*Z?{;W{X^@l^81VS470>2OEV$Oe9S+IkRqxlA z68f||+IyNB#gmCYYc}*sY)8CifjF39UM%@1Y-BYzCKl1u2wW}1E5OnN2{H@90)LyT z9gO&0qAeID1@5kKSPXhO#YV7i7QfE^LVVD^OE~n@(-nA>P0Cn1ha$%qx#alO;_Qa3 zR@e6}+1llTf|Fg_(#FjyQ!|rjd;C&S2Csn0EW>n5`r+(&gMx^$k)3xrV$bfR)Z56`#n>82kQ`{V&MN;hvTjyZunUiHV0c z0`*FqIAeIvGle&tk2^E~cl^>l_KU_i{BD?0@7~2o`BL$<(sRZWlr3xjrFLLJt-+7$< zJ)5l!iP%1_d+=OAxwOu@bsc^41l%*S-DBPK_eQ?i#4JOn>REJ64%uX3?K;xe9MLM% zJlU#^ECys}&yV}>)}B;fo9XAoT5M)9c#}MQ2nk3P6zw+q^U+JAjfwV#W~)J~Xr5)I zhAMOLNKP9S#j|6fC@bgqh*1Dc>p9=`wmV*@;%YLy`zA0}K_4?fitsN=J9){kgi>tz zs@1=9`D`p#rtr&Ko~~I1U4q#VDaId{T@1&#b;!=OMb4vdW95w8>zosU^F&))!9%dg zY}wCSiT8-AJmC0FIoG$?`ALotQ=pgdZr7}LP{gZuciwJ+=cn7Vr(lXq3em=2 zT^r=R=ZM`*imtHB{GhS1@4J0B8$-v>QMTx){!T;hk9X`<|L!1qYLT7S5KS!so;QMR zIq$#^*L&wXt*5)KVy}^-8QvU~))s(UDRe@Ap5st|(+@$fSTdC__XdeF1icZ6O6;Jg^aL9H73*4b*lx_jJ27=xoHxJe(cbC-U)lx9y#C zd7@pX_=Dl{`tqb2y5cB2`%x>bZJfc@4E@x@Ke7WQ_kIE@JYGrm8-3o$=I3iO1WT zl9K8P9}35|{;Xc*)Mi>3qxdVot{Cllf63E! zWX(8p#A^1p*)+c$N4Dd|Z_}8QuH=ys^?3`+pZDGU4x`fCjGn+gtB$f%uuiM$igce* zkpi9XHB6SyFvk57)e4<8{u7-)tV^Tqo2g@!{4OC_47c&Neaf{YbNE4`NkM#Lx*zm# z>QuxMpA;h7ogz8+zU@LXbT6zRz=$4riL)Px=NQ!Glx`AXdnoZMa9C_y>UWi{ zE#Vq<|8p!A%O65VPUQI`{sClC+@rPowR?J@u|oO=8Nk`ci%K=#Q9fQYjapE_97kN9 zX7kRym{b^bhRvbWf?v15e{XbO`}R|J(t+_B zVt2)Iu7SddFJrQxb7P!fXJ4uR@6~pE^8EHG@GpM%^fV;@t*1(7b9sS&c2`a?NzqQo zp@Vqs7x}Vg)wBJp&k3`6sa+!HW2Zw~w?#hrdRf>dqDvC*JErfG=yj%Flt|v2w|$wa z;#)%mboLwc5gVy`tZZJj+D3NDytQPZ?ISR^@yQRo_@A(zw8VL-mZ!-;SIg5c4k&j# z5x5}}$T^+@nKy6@PqR~6Q^fhPro(&1%Ll;TyRaW=i5@vBq8O>p+DYNC@%&t*83OPx ziU-=|#Xm6<`v|S`==hJ$AwPU=Gatn89*Q1VLPFjQRu$we<$n(d_NHU|{%~|zC_#97 zoVg6U_mX}x!_QD9LsgmH`UZONC4_mC8!BVE*yew>bL-!D{pwP=aZ`2~BRg`o1k){~ zImqqVt@Gv2uR=LLuL|*}?(_o-UHGPgiDe+=vMw?ZWV?`-7U{s|CC{n*gDSKcAG151 z4$Ta}2x&jFR`hE4n0zY^$WMH3LOVsq- z*0(YOuOvDosP_73ayjJe=Z*N?oq=|4d86y0O(0eKw*YBlf;Gp3+pq-h#o2e8zk=76 z&h<)n8UDO^^Uai?OHAi(`HbS+D0dofyxZ%@L>8V`6@@Dll$nD#Yy>~-bjdE|n;1WM zoev4zuLK*LMK;#wn-*goiWp++^U>F{7+e_I0+ zHX4ZtWNYr2-oLHh$xLRS^C>Plv~~G;_-UKeVOhvoCZhGCePDSZIE|AudiVOOymn}E zeja?$0misIC$pt2Jw&nZ%)Q<@Gq}{Zpw0>!yubT?Co9?|xpng#)q2zn?b2mwguue` z>OhLe!OSe$UdH&^@mzX`>b?Dw=8M0&S9-XM+CD1e-RSP}3%cFGIIE-D3jlfdo9xJ?o@;+A+-(vpg6@|Wk6m-?-WFTmEH2#-tvDt?8+I5_8rp4I5otq+ zoS`?q56^Bn<{jG~++`^Lz_j~jSKdBX*x!`fcly%qijm;PP3#rCOnOb&xpeb+?vsIHV zb5d8w(3`#JAu+3#K=yN+9%17n9$5*kgEzE-Tm6jI zX}SB6_V+`~kiFw=_OsL7Z#CB{ertcugq)-pHXP2mkG!7c_j(#Wc{!QFZr}FG;>0_{ z38!d#;FxJ$v#kWfU7YUkUcCdTLdotb`V+L8|VfvMW5SWEbS?f#M^?P(*f5R1n1}X58Q9b9b%UMgMLJC4hmT zUHf4-r&i!BfdnzKbBcIqxdz=HmLGjzPPz#00$>0BGkWj-P}{bssKwBB?U#jMOIsWf z82MYJ#-n^q-*fanI^<5m`<-*(=2dcB=<-LbtO~3q$tyPMcM=mU{o=a`ES$!J++um( z8Bosi8h=_kz7!sxZC7$vtY3WCmu6YBJ3k$YK@F85`Rgk*yZT0KGU)2oE4U*3I$r0& z4eMG2OQIatL84WO1raB)A5lN;xX+X0Yfz|ry8e7o`)8{I0qnTR)=9OFWX#HVcdI3bbeMlFQTa(9Y(XXGA|A)a?DzekDv%ddZ`kHr)_D=nRZy=608M6hyrv*d!x@lB}N$&EHLHE^6O zk8M~FcLK7KC6pxo%b_xy(8JyHJvF{KH^RSf+U_HI8jDG0v$}{!hj<**O{ga==GArd}u#ZOzvz4d4c6MJRBCS;o!-g=&5ku5koWjx+wI<0Lj zWqlR z5i$`$0aD-il1(sTtE@iQKBjVx%mleqBersiqlMT|)mHpgS7Crv1YAZof;dRtG^%8K z`NQ2iMsmo_7`eil7A)r|l2jg6U*xHUYvGjDS1~mWuQKvPU?qPtt`Qmes<_f35ku4J zX7a-uk-Cf1Bj^7MAs0NWR{EZJPQ0W}FG-`dlKO|A)Oa(cwj$au1W_1$CAX^BoAzBC zG?P(pMA~bbS)#~?Ojx9@>1>G!*g4c&;$XI{T#WxqB@*OIrQB@bYjB|xSsAy8RJuH_ zs~V+agw6z&S;daJ$fuU*#!vT$Um72o(yh`9wSgI(s1C`njNV_1-En~v3q*xchRpli#Sk95Ypu$!%8FA_$PfJ7V z-TQLWArFj3%)j3z`1VwNMk{bTBJHy)M@dUM_%qQ4LO9ZCO^=iz0*u5Liu+qD5{z_E z(^;P1pdObbj9ZaTv_#cak5vV%8!dJJA@>Cu=9-?TT;wrMZ@JVOnR#p5aTzAx)o3mI zVZR@-A@MJpHV4!;cl_p?LHAGrQ-qQ!WtDlGgz*}gs!7=KV8UW&YNF?8+Q|ek6@9R+ z`gpKz;<;2T=zFBtAYmsr>?5+7lRA4OxX=~7vpnpZVi|Lvgy!5d_69lvz%AmNJ8ARf ziwLZ>N8AKVMV#7DbTS1^D}GfQ;xOf5bJEz3)by6y=$B8*kC{~}yZ`X>jtLxo>@#5h za7HJQUTz`ED$=wxUGWkF1J+$~(+|oK;540@Nbsb`Za?kqf>qD4XrKNv*w^V8h zRdWBtUczHd>;1}O(PBv0lOQxIwq{(htsdT~kyF_un?siIBJUaLy=A8u*T*J41X-iI zA$JnfMV&rKuD`Ow7@%Ctbz_$)> zfN!m%SS#r@dm{d&1Jj!=0N(%mf=genmshSn`Pky%>`Gl$9;FBVy{%iuO|J*m#}OQSMX9JXcamDh?V{ve zFS*6H>_t|BK8avD5a{wk?;sa6L9Rw*Rr(2-V6Q1YP|`kY@#X%~q6c1x*>Ol!Z`k2d zC#fud+ti5ri_94OqPbEBGu_wZtOK0)N=uA2;^Pt6=9Qrj0;6-yeKeTMjh4_0tPQxZKQ%nChvru>fl$dzPhGh0O~h6(WUz zNHXD>c;4;ezc@z~ask#V+hpERc~e@+XNJ1Lquldxrw*x#-%cH)|#BmRaEw0Rr*t)_tiSB)Cjw+-Xg+Xql2I>txZR3Lo zL5t?K{|uT*TBKz-;Y3O#*nyN7uk-+HTrZ!zlB3P{gt=&az+Mv$9e%u0(W;p7Mf5zQ&vsp&~t)~8If zHh#6Ohltd`^6|ai*t(JuO6jX~`}@HdfBB)kBSUJ*?Q=l-YV%8Ne;|;8+^AFaJFuDq zQMDGPj1ly`>Jw^iIdW?5aW-AhUN+w1zytH@TfW>GW+6 zPCvRw*uq!fXfkana?+5-1xo_v#njcpym_h1^;ISD7<1~}JqQg%qF=|r7;6Se43Ha# z;$Z>n*;>iEoq|G~YNo%~8USy5`Q*bw;@aEuGQK^9K=ggYibGZCHclP5Ya&Z>C2s;o z=lC(V@te%P9&`Mac4?{h^WN+JS&X zu5r}7v*=+`2tv$_3U`C?;;WFRO1i&1FNVF^!iZqTFbN)(M}OWDfEB1FU9D81Oa=( z*?yzeagg$zGg#x5{Mnc^G4ZEdd7K(A=sb;jwcGp~iWSfZ`$HkKusXf0s$Nh1y!u|T zNm8xv*$3Be>KoXnh;T)=FrhKAQ{#&3IQg{_1ehi2tZF=uiOD*| zFoTDkCW+k3Eb$bAI<%sOKiii_f~@cceJ8x7(8cOIig#wc8D2l4sFo)}Bo20#;plEA zqA<68w3$Dc{Tbs>FHku05@)qs>X=Nmwe8ZOu;R;z7Xr&hiCu7ji?OyWv83AIhmCjB zGw$j}lpI7;tLBs%G!NWuXtbZIkQnK$U$H09XRuBqQbfvB*EvXu)GOmTXNQggHm{?` zUSP<;OMB+enE+C2k@=*@jg(Z>zu}t86rn%jEZ${@sQ!r~XEJCfqWd};osj!04{nKp zVS@%&g(31Euc>|dcbwV1=p_jG1K}=v^j>lg!!S5PtfE)4F^|0=d$zi|H?I2ftC1sO zIB!vnuld5C#J|(Yn^H93V?+2{8D~9NT-b?2`57;Ck;=8}i@J|K%|={nMRqe2ZKBsX z-ijD=e?=1Zd9~4x=3^`$ZXGZ2Rzyqtubt6=uj$q`hHw*Lb63shhhFQoGy!Y8}zU$r^{z5m)A)Z&dAMO6Q3cdE0f6j?RyX`gYHo{(Qb)v{g+L{6#N=&RCvM;y*VkxUQQzj61 z1T)x2xw-oFgqFi@#Zd$GfF2Dj5w+jS86*yjm!MpVU_z6P*vgK197V9Sr-`it#I=NVkwL={A~A3~Cywc*ucoC|Ft;dDJ&xWY z{ka5ogY`x#6{gPk$VlPK_>t{83S6YhAMr}ioRr26PlDF z(vQ6kE@{*buqXt?e8S3CBUeKnkVm+>O|V=oF6 z_ba?(CSVF8?e%riN)=9~uI`-GR&~WKtNV2z7I2A6b$#{auo{UqvL^hn(udRtn1K)ys`1wl}KxI}Agdt9_{D*z8G~f5ao`U#|3o zQ*e*PqtEI8?%9bgm(X7#w6~@;u;uz|SuejN7NdF2MjSy|E_kVsfc|uPB@T|XOWh6W z$`Fp#I!_IeAB5O&Yerhalp9Lccxn!@`#3ujB1?pk5&+DoaM3D%B-FE2@5azMLSa*> zAGm~}BF}%@lV71F=|)hMH#<9I!Kt4?!@kCUfUn#2)=c?^2JBbZ^3(D<^rlv1!P!3L zyY|MdhEwe_isk_C1;zdFh}ba1!|$X}WYA#uiGd>-F1VRWcFLqx@7O6IH~Oia)Y-eF(|N16|;hlD!Q8N!aI z^fnVfrMbWsq#g~Hyctev@gj+=X7TTFNDX1EA7O47)V77?2uuGXKX={?>gf&^3F``T z_L~rn+h z;=AkC4l(z}Glf}bXIt6!C!24lPdIr+8a|@-iKE698z@QQBCUnD;w!f6NO3PKGHS+V!}m7qtP>~bmB_pQY;ro!Y|w{UnoS7lsD5kh{36SERny)i^DOJ%0{VNZ<@h| z(Dzym6z@GPMHGg&zt}dCv#y4d0MKQxbOX?dKSGxakWEP?s@`i!gI|?0O<3X=OOFHU zkUQT8h!#2y_q4^rRwNm&KI}6L@&$vu+OpyTCP_RO@Tm=I(^u0(UjBh@dt3GJxUkq> znGi}xvd7dLUcN@u4Jq6Tw}gd#*2*4DbJJ=`uG6IN;M(nRK&$-Ou~{Rc^~8`zhun~- z@Z6M_T*oMyT%)2MqN(LSrtf&N{-WKG2a!(hNIXPu2vaN^x8%jPPJA^khHg?$u7UFp zIr=(?mahUAn?=2F)QsNeo{2;|IGgyoBfz$gJ^$F|TLv66r!DdNMT%a5G-tF@J@RK; z9ois?9X_dY&m2hx9@+N#AKQlF(SJ_^Z2Raa`9eXXy!fZZ4x4q^sb&3-N4{PD-+cS= z$Tu#4Z^R$Iobu2aSSK7$V}U3TLgBJ`UeK@(y(YLaO6?A* zMny2h9sb&h%L`hs@S7vv38jxyAqlbYOk^FhKM3!b+E|C6s(P2Qh+AaU3r?t)P`rnV zgb&+IdB%Q-!S#T>U2uw`#ZBO4?n8lcr$X_@-{)0Z92~UY4*GvG-f)2_f1&M3)z}cz z{sJhZKK~R%2U=hJsE`rV8dV6$QZj!0vybhBqi*gZ87|6+B2?SbQ;^Ao!d?lsndgb< zW_?v)XjqX(8GKrptP=WyCE;|?Xq(&@P={aJ{hN~_gHL>TLdku2#?NIW$?c3{=^WdA zUu36OXV;Tjd}V1emCSJIOCsyBJQb~a71o(l&ytQeD!)eSRqw-t@{Xbr`XRr^)J?^y z&gwAIo5k7Mhi9@k)i3+`xqGEzXf>dVg?wLVX5n&+k^2x-j}H<9o!pLXf*CMj#0;D| z%sE$%&zWTQ$2$bQ{z_js1NYcFVCKW`z8!eELz5XmCw{+((#JS_~$H;c~H_1>v zWz&8*W1obvIJv6HTiT%nhon(B`Je-_%x&Ks&IaT&nMj{-e2|DrI%pjNQ9hTq5M69&!QH zis#7N9P<5euDAT_kYp$wu$NwSrF@defqLtxB zN|YvyPXwCkC-qeH!+lg1{|v#4nt*)iAF=5N;8O755on+npZ06aEUuS;`4)qAbJqre z=P%sk^`-_NiVl)^z`B|cxZ>pNkQW1R9b6v9K5G5BR0*}H1dv64Y;^)RfSQ%JG~M=| z%!(M(7XY<09jN^DF_r_5fS0_;wBHSw>nFC2eImu3*FFUpTJKRSN`1AL)9!Ag1Hh*9 z2wT}BY*UaneJGWRu7*DCJxKUhdGG4yDM%}PMtT7fE4R{XNIXbhTIi0`VknvcP|X)Q z`;&mz3Zg^Rlh4XI&>$$9;oL(Gz_v^dz$W@Xu+adp%_IV_4J9SIMHy)Dr+)%6G;!P~ zTdpBSkce&%vj39Ihi3+UvIk*wLHFU2%*}jeXlL8ztxUyQ3>!_drV(e7#sl3%c%?J$ zLEitMcnrq>m4puYW!_h_`*w7ca+k~>WKma7HV1bSQu)Yku;)%0kPX^{K-2-Y6>7~t zvTYBtmS@2&QPLTpQIo-vUW~I3bTM60P5(DZ+EOe@eb4!+YN&`)AId9=TIg&&oyLWw z!VCQxtCOL}yZXRdKk+@XZSs+A7uf*YECIHK{$rb`UZ1*#^Njj-3GpZxzyft}65=uc zhQz%y@v;10=?$kT=M*JZ4?|yorLo13B$vYl{4!5qN zE6Mto50A~X0&&`fE~iNq-Ow}ZzPUuDeTzPIG#p8wg!BA%H^S?kEG4_q7n84mD<$u#Xg7g4jOce)W#F zoM=Xir2kuUP7J;W?EDMPV^;cm@DMhlh-c8OL-tCJEe37K?hClnq0P$ckV(dx4am!I zKqWEE{+CLc11jkVVb(yXO=rjCT64dk{)IaPr#9g2Gq75wNO~1Pi>qOeL`gGcFHVkS z48*oQb#h|A?pqXhRP`gTQ2=B|H2M=zapM0UW5XK)jIC!KvbMjJq$)W*5%4MESe=in z&tZf75UBlKk9*yJBzY(2i~tS*kKBWpJ4yS%kXwpyJsaJ1Q*kvIEF#9=-@vI?JV05- zf~A6yEPzm6Dx{*T>AybCIQpvrZ4Q3KZX3W(4Zu$A5jz$DyXUDQ zd{9ZX?CVlwKPFaAr8D&!V~tdp8qZY{>~+4I6(E)Gy2izxm{c$iA@#*e*4+A$8eWek zn-ecdZnndQQ#OpmF#ys%Py+m5G3ej})KWQ>$K2OKZMwslS+Z-+(YZ};sisOL6@lf< zjyt@=Csy;jf6&tkxrt1yCSVJg1@JVg#NDdB)@nD+svdb+ELQXSGbnB<(p}`- zj%2P-t*zHOEy{o?CyUda?TI*o_#id-9<@|JSVJ-^ZbC(0s>srzwR58W#5p+zZ-VH1 zjT*z|o7Nm$+vVHD(+I|0frB%JWORA`SK&OQsJK`)y{*09NMMmV7RR5Z0>P%qh(YO< zZB^T4;w0v9hTSX(fh>6f-Mim?D2fz+p^2iQyst{@JroSIJ7+fT0FgKVJnqKFHJ9oZ zg2a0ZDco;=!!zE1FPmX@181=xn8>cynJBW6&2SxjS5GrU{|0clBPegb6{p)1>b}DA zGw4&dwmJrGV;SkV3b#BmbqVV{k?r|@+-rERKP+tXsiL@5m#7xu8y<6t8FU|JS@P;u z7=xo#<$2x_+wi=hgk!~Y{TAFX%$Zc~{O2}@_!b-(a{q;lLpQeJ9rGJ!5L8bkTeLJi zt+ZO*RK{01^yNgC%4+?4Bopot<50aQvn_ZbP@YJWyP1U1JJ#|$OKh}TmKQ%A4KypD zFVAb=ZZ){PP2A>C0LsZ!j9|EAo(~lHJ6RKa`ckCK(}E{aCn+ItjF~pW9V+s6`M-j) zGZC@CC$a#+YXx{$vH)RLebG0Kb(r43l4rS@2~?ipkE4Jp0}}TiiQdVQBW{nhTL5S$ z;sPUT36B0^Ep7m;clE>>B7tS`ODf!$U*KS?rZu zg09Ur_au#Zc=l{slCnggw!|n_z~h+vFNDxXDx4k|b8Ed5nRu;MC2>%Ien<-UoZzkK zbBL^*Oc^70X92yX!_AGP!>w)yd}Bu6qR+6(QgQ`GYAU%Bd6w^8?*oIPd(0(?#Ii0Y zy&pVvZvf_;d~a!LYie$n*)2Vuq_ABh6+k}9mqW0B;^W7j+KmrMI(&93?+x-EOOUMc zr*OX!>%5`;c=X@QILS*7e{cVjfJx!Zo-6@qn^Z9F?*f+))gtR#uy4%Cib2-RSk%~& z3TAYsb~M4>&Va47@pU|?t)1Y0E#!7AB@aAfrAmrNoY~|=#K~OKK&5G{T zLpZ;z{|RDeXJmgrXLl%=k!1w>JZ72@=kRB^g6GSv_?5W_zq65}0o?RiW88-Z!MvC1 zcKfuo?5Rs^?zQ(*(px$cmoGv*ee<&fnS;KCNIh))1wJ_4Er0Xn5Eb)oXlFP6p8SVj z>&o8s)yPrv%m;CO{!vqtwze%-`G{W&5sti?ulYMeoh5&E5hNFl&>m$axVPm8p8r7z zpeOCkt*A&8==#cDC`OA@J^YZ_nrXw>`a(gceSNa?Jwa&1wCy*Q|4K)EYTK4<>56ZQ zD<{_5!K8Wtt)H;A8BKfwV<;^>*XJ(o3kxNQ9Cd@AuA%!waU+f$am-ecc`0P{FphyZ z3(r@V|nV)}6eI|KPu{j<}^at1z9=__`%5^ztUcLGBb=fQV6164Ax0W_^hkD5a_W-R}8#nv6zgm)ICwTWN|X1 zlA7+{g2^z_C4c5RLd`itCR-CdC${J)ou^bwcRDG9pSAQk+>7kA^_|R6^=X;7?UeO7 zs*4!8qsF$I%8s0#M3*6qHfFM^4B&aRhKBl@QNCl|N}ybI3wd_+1Kju#N2GY>h{#+b za_VXh?D%_6iOxfw@U%n36&1-*Fl4 z=gqD!mO|nP-rTB(Zdd&y-Iy?Y~fdC+)2a&Gi6*FZdb!Q7h|y&LN$EgBILa(}ujAhS|S z|9Rzb^zfYSP22T0==MVi*_IHu$)B?TwC4qc!p9O9p1ZQ262!-nXLwJ>C8GGjaBhHC0q zkd09?HC9iAe%D~}scCBW;cWDC;g^#s&5h1z5`Pm3ZFXE+rn=gjgvWLx4R?01Adz$! z6K8274-0k_f4Q5m=TC$(N1VVm61rIkJsTf9e78>L`AJfcKSZWre(P4_3oal zj3ne`q5T}=d+HmVUz_qF?nysSf@ap6&|#&NE3#w*j1M+(nvTT?Ee>GCq0wuq6zthC zK0)8;C2M9OGOA<)$`93dNFU-sr;TzpR8}XXAEMqTPdrug$o`vv$A-27#IipxNF07` zC1+z{)p?0(#TPfcjQqhm+5QtrEl=B-X|*}I`l!Dp`DnpF?Cj#lG1;d%G5^}28uG6V zB9Cq03$TeL8BP1CAIKXyEd(N&D;RjL;S(+M@R2pPKXkaWKjH$j-E<$+ZOCuA#6o$iz#Ci8$GeeS=M2k7Wc>k6zG2%k!*;}2V<``BcjP=Kh+7a zL=uzg^Ej|ssW*ijfKC;?q^NE+?Y=SB_x<>D8c)GLTGbT@YzSSRVz1R3a%5_Se>xd6 zJmR>c#1`Yf-o1dCAVop>0INfxz;$!)F2c6j8N1HS(!%+a;GBF50Vt^Mj9H8?srLXREsK<3x<$3kkSSx$2koiPIyM2d!FiOhIUCdY%KD2-gS6XB+%QIY;iK|^v49CZKo@lM4mE=-AFO>WUF3Nr{I{c zZAjf7auG5}ay&@g-%(?fLi>pb)x<5(jYia?2Yi&P@KIxo7!Q(KMm8QV+305$W z2yl;%21t3egY|QQjS+B{t{{4v3to=v|Dez9W0Y zg(>9RLHR{4CQ+jz&}lM6ul&Pe(5*$c7BPu67>|V*;R%zNG>;$|zjTEVQKJ+EI0mV{M|S?TC9HDES zAFZC3UU!LY>*x;Gj2Obu}Y*t=Fydl60vN+;rIgQ#{xleBB*`0{}^#&DX zQl7=fN@M-fTxk1j%7zjvw@M!C!^d)OV2=kPXI#Zn#Mj(8!J#Gp@o~9_$v6B%{R!Xgb8Q}5;RgTT zZf~D2(VZFp3&w4+ACDrS&-lG!``dy_G%+imhiu}*yMSC5rXN)}Z3-Y&`vsBGj^>^u_ECi35J+i&5 zIxOXRI<-)FGCEc_J1nVqv~Gp(P9q|Glbk6bW9<~4$~>Jqo|y4AzP~U0blm>qsKwWR z9L0U)D0_Z3$DiKEIY}^a{nPkajJ{k#mUFal;`M#pmJn{u7X1?sW`N9Mc7+jnFeGi z6H0O~z%w2L~VsZLoo6;SfIFXA7tyhn`KvIm+r zm$fy~F*wL0067S?#OAoi*Wp1`qIv#Oe1C=z*S7Paoa!T1*oXOawGQx`D3;UhCD9w09s0M6C`IPLO4&>Z@7cN3Axdi0kjHDJd$fO!_$24;AC#1n> zN>z2QNAd4m;(Bs*B8LCqllu1MoqWu~d7{IEsf3+}R|r9gTSkdfD&$b|8djDpFBV=m zfq1KWoXWrKwDsJ~1r%uL*7Mod4z9=|=@dTPIBD!(I*+X52#+;i&4_j3V*^-+%0e!; zs;Wp+IOX~&|4Y(`h{-)~*NNA{Yp74BO>Vj_Txeo`e%U+`p!;WWcC(m|j=}%if{4Bh zo4}eqQE<+|Q4`|EF>!jV6Bg}A57^dnZq^{K91o_V@E&Rmdgvwi(O%8~dokdx6UZJ& zW|tRcAQTC0!myOuxx&bK{mg0T7Bb*lc12_$BDLoVsmz?L$H4qIq*b3tz+Z03 zEV=lR;IxE4%YbdC06#1ZwF< zd5Lvcr^k~p<+v%#6yEUhrqULoc{454W`xJ>oSB*HdqJjxH*iioeCTHs$Vm|!-WO|HHMZGN{usrZj<3TFR@Y;0)H zdO4j90@s8&h1OPHh5z_7dcHkceaKv`wvNt5ZUuVs%-j;2?Vendr>lF$V(mT1qPSL> z!`e@Ih*RFu)Aqv4cjjV7>cxM-QvU_x1Hp~kryAjyEsg~qaGU5aEuq5zlc4XjDZ+t-^gj1kwK?XdQ=%@dc2_;aG&n+$l5)tXp zsjzb$T`S37Y`)tAYa}AGYif^wq(-=IokY)>uTvbTN=KKv6JiOCGz*3 z3Faf5^hb`_2#sM2MuH;|KHxKwBZIUa zf+9shIUJP0Xa8b}W79?oye8M(xaq(KaI`sF@ju>)|6ja&N<>!+)^gySsm_Mu{d^na z8Y@P1CLNR2kE_VDRW9!+lMF$LEkKU74iW)b7P+O;J>_1@WvsCIi@=g|$L-m!-hBS_ zzakBd10@c<@RIZ~+2aa1B1JP`I#JTK2qB>8?BON--xbgQ_La>?Um2u~Tfzk`2mYwj z^U&f@F!IEF%p27Im0~Ubhi#huju1-bJmEo38zE*ArmAL74R_zE3t%hrK6DT{3ur;) zU)78a%3vx=iuqEk6P3WoJS!o+nMVAv%0IpvXsXcvG9e(%7bo4fISa#J1smnLr1J;F2u~!{!<3k|7t#JptE{rEI%bw zTV&UF8a9}Y`2Rlv$p<$0!Uh_fJL|;Buc8$huH6u$)&i$GDvCL0x28zenBLr$pv zp+j>)QNm<%2a1zn-vf$?zz!XX>O|^0P+Wa@yBqLFbz>5g1SRLE4sNdQQ=*0$$&b*Eo=U=$eiba9t9va0>vg1>pKu zz_kcmbv|5mClxzst#g{SHZv|=HQ4%ownlmEgVto2AnTH3g`farEkM@4fviQyYVyfS zofTm+7EX8)eOLU9cU2A0`hK*=f$YQ9bew?el5mCU0B|h;*S`aK-Us<{RDI^LRYsBT?J8C?l4*7oHt@dF@pnLbLct^WIwv5 z;sjlnq$@lQplb=behRu4p{viQtLi+34~q+2SQgV%JP z!0V!(Bq$-k*AjgF416u)NfP@kU$xX;DPc%EVYQvpKQBPoob~H&B5uFsYbs9gby2<& z_zJ!j;OkTHwG3ZrubZ!|3|r4^thau~ZrAX(fY+Rxue*WldG9q&6Lejau240Bt|jRD zIp|u3u6xLuZ0C((cG{^By>M>6-C=^ODQkP!P7>J%ugN%p*G2IPQxou70$t?f~+ZP^*T*tzvXKxPXKmNz{1u9z?J~)mjG-T!0LSftJn-D zVuW?1+^kP^gRvcsJ*> z#NlnqTft5e*>eS($P<8F6tECC0k9*0H!zBn6af?r2jeIm`il-<1MT}*v_6##bo%W40EjTh@UT&FBfU!6a{I2*6ji|;S{ zEc!RP9+befi~bL;{?qC@P;Y~U_19Zn_j6tvcKy5&-pWajgR|#vUfyN_Z}Ud`FMJw4 z@t5D9CyBSdO!_bW`f7m57JhZZZIfI0#Pu)x=lON`xP?n@N~le${U7}DlP6!T0`hL( zTwm^mF7w#vQX6#ro-N2V7rFj2esos-#rAL83*p~ZqSLHMN|Ih{p+uO@9u-V?zwdAE zPAgONgir8~=f9l3x_EYd)erT3{89h1`t*m3P$h5XjBr8Ugo`OmTra|itR_o?#{x+B z8BWKUVRu7xI)eMKBsKm3$=o1Wub2RmqXfzS4v@?TlHZRXod_h<8Ex4s-QU~sR;kB5 z26Ok@9w$fQ0o>DlHywUpWNwT+>o1<`^Gm?vc;N8|z+*mme7gR??2zL`bTX8^#E!Lb z##z?#80vrl$GOfvQB0Hh9bVOcD62goJjb_?pIwmY6MsBYYOo@T7lneGm zx_!B@c@1`%S|o&6m2ve6DdwaC(L97_gYe<#~--B zZWADp+HC?PO6@jn2qD7^rERK@#q5;*k}SEk{e9`M4^qQ4IO01BFs#J k-QWlczn^iV>8QvMkKCR+P0_&rAOF+;58#Vd_QZD-0Cl9^i~s-t literal 0 HcmV?d00001 diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/manifest.jsonl b/sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/manifest.jsonl new file mode 100644 index 000000000..9a5912ece --- /dev/null +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/manifest.jsonl @@ -0,0 +1,7 @@ +{"url": "https://www.resolutionfoundation.org/press-releases/budget-balancing-act-leaves-families-braced-for-tough-winter-and-four-year-living-standards-downturn/", "title": "Press release: Budget balancing act leaves families braced for tough winter and four-year living standards downturn", "date": "2026-10-28", "publisher": "Resolution Foundation", "sha256": "ac5369e6f73bbfddc53dfd9977a34a89e032be513e1a21f192a02d124ef98878", "bytes": 118327, "doc_type": "html", "access": "direct", "retrieved": "2026-10-29", "note": "Fetched with curl -sSL and a browser User-Agent (HTTP 200). HTML bytes fetched 2026-10-29T16:16:35Z; the page's HTML can change between fetches. Quotes are from the article text. Re-fetched 2026-10-29T17:02Z: identical bytes."} +{"url": "https://www.resolutionfoundation.org/press-releases/britain-braced-for-painful-second-half-of-the-parliament/", "title": "Press release: Britain braced for painful second half of the Parliament", "date": "2026-10-29", "publisher": "Resolution Foundation", "sha256": "40b102686bbee5dbaff30ea96bfa6bef3f73fe777142a826c811e7fa88c7c22d", "bytes": 116817, "doc_type": "html", "access": "direct", "retrieved": "2026-10-29", "note": "Fetched with curl -sSL and a browser User-Agent (HTTP 200). HTML bytes fetched 2026-10-29T16:16:36Z; the page's HTML can change between fetches. Re-fetched 2026-10-29T17:02Z: identical bytes."} +{"url": "https://www.resolutionfoundation.org/publications/spread-thin/", "title": "Spread thin: Analysing the Autumn 2026 Budget (publication page)", "date": "2026-10-29", "publisher": "Resolution Foundation", "sha256": "66a999dc35476a54c003bb9fda0ebbb3648cb52403d7b78c5b9fe9f5f569d792", "bytes": 152177, "doc_type": "html", "access": "direct", "retrieved": "2026-10-29", "note": "Fetched with curl -sSL and a browser User-Agent (HTTP 200). HTML bytes fetched 2026-10-29T16:16:38Z. At that time the page linked Spread-thin.pdf (v2) and the chart workbook, but its Key findings still printed v1's £158 / £144 for the income halves. Re-fetched 2026-10-29T17:02Z: identical bytes."} +{"url": "https://www.resolutionfoundation.org/app/uploads/2026/10/Spread-thin-1.pdf", "title": "Spread thin: Analysing the Autumn 2026 Budget (briefing PDF, first version, Spread-thin-1.pdf)", "date": "2026-10-29", "publisher": "Resolution Foundation", "sha256": "a669f39d30d5648697ce0234692bd9695fadafd85669ee809c02af0c90595603", "bytes": 1410169, "doc_type": "pdf", "access": "direct", "retrieved": "2026-10-29", "note": "Fetched with curl -sSL and a browser User-Agent (HTTP 200). Fetched 2026-10-29T16:16:41Z; byte-identical to the assignment's local copy downloaded 08:46Z. PDF CreationDate 04:33 EDT; server Last-Modified 08:37:45 GMT. Still served at this URL but no longer linked from the landing page. Replaced by Spread-thin.pdf (next line): p.5 and p.19 changed £158 / £14 / £144 to £145 / £15 / £130. Re-fetched 2026-10-29T17:02Z: identical bytes."} +{"url": "https://www.resolutionfoundation.org/app/uploads/2026/10/Spread-thin.pdf", "title": "Spread thin: Analysing the Autumn 2026 Budget (briefing PDF, replaced version, Spread-thin.pdf)", "date": "2026-10-29", "publisher": "Resolution Foundation", "sha256": "9536b6cd8478dea91f88a103c9004cc7b4cdde269a521f62048e3c523a15d9ff", "bytes": 1410388, "doc_type": "pdf", "access": "direct", "retrieved": "2026-10-29", "note": "Fetched with curl -sSL and a browser User-Agent (HTTP 200). Fetched 2026-10-29T16:17:29Z from the landing page's PDF link. PDF CreationDate 08:13 EDT; server Last-Modified 12:14:43 GMT. Differences from v1: p.5 and p.19 income-half figures, Figure 8's axis top (+£160 / +3.2% for +£180 / +3.6%), Figure 4's x-axis adds 2031, a p.30 typo fix and reflowed lines. Re-fetched 2026-10-29T17:02Z: identical bytes."} +{"url": "https://www.resolutionfoundation.org/app/uploads/2026/10/Final-clean-charts-all-sections-for-comms.xlsx", "title": "Spread thin: chart data workbook (Final-clean-charts-all-sections-for-comms.xlsx)", "date": "2026-10-29", "publisher": "Resolution Foundation", "sha256": "a4e1c072bf8f95f364a470be46ad1d1fad2c4f86fa83d371b45b577df1d6e0d1", "bytes": 14637565, "doc_type": "xlsx", "access": "direct", "retrieved": "2026-10-29", "note": "Fetched with curl -sSL and a browser User-Agent (HTTP 200). Fetched 2026-10-29T16:17:32Z from the landing page's data link. Server Last-Modified 12:16:43 GMT; core properties modified 11:51:06Z. One sheet per figure (F1-F23) plus 'Extra spending stats'. Read from the stored cell values in the sheet XML. Its Figure 8 values match v2. Re-fetched 2026-10-29T17:02Z: identical bytes."} +{"url": "https://www.resolutionfoundation.org/comment/its-good-to-be-flexible/", "title": "It's good to be flexible (Nye Cominetti, comment)", "date": "2026-10-29", "publisher": "Resolution Foundation", "sha256": "3510a245ad89667dee8e6dbb42d175970c05667a244c4034197c43000962fcfc", "bytes": 131349, "doc_type": "html", "access": "direct", "retrieved": "2026-10-29", "note": "Fetched with curl -sSL and a browser User-Agent (HTTP 200). HTML bytes fetched 2026-10-29T16:20:03Z; the page's HTML can change between fetches. A Budget-day comment on the April 2027 minimum wage rates, listed on RF's comment page; not linked from the press releases or the Spread thin page. Re-fetched 2026-10-29T17:02Z: identical bytes."} From 0d09200653f5ad1a6f2e57aa776e3804c87e91d8 Mon Sep 17 00:00:00 2001 From: Claude Date: Tue, 3 Nov 2026 13:23:04 +0000 Subject: [PATCH 12/52] Land the OBR family reworked under its review (2,891 to 2,797 rows) The review (verdict REWORK) found the OBR numbers sound but the keying, typing and design rows systematically off; every fix is applied, with the coordinator's calls on findings 5, 6, 9 and 10. - Rule 8: Table 3.2's 47 funding cells, six staffing inputs, two guarantee capacities, two loan totals and a financing requirement are design, not estimates, so they go; the staffing figures move into the yield rows' conditions. - Rule 9: 37 rows that cost a part of one Table 6.1 line (tax gap items, verification officers, the India FTA, the CSOP/EMI2 head, UKEF and the Growth Guarantee Scheme, WDA/FYA) now carry that line's key, and Table 3.3's tax-head rows and the behavioural-erosion rows are typed as components, so an ingest cannot add them to their totals. - 37 text and Table 3.1 rows that repeat a staged cell go, as the AB2025 OBR family kept only the table cells. - Proposed keys are renamed under rule 2: single measures drop `package_`, and scope slugs drop `obr`. Table 3.1's savings and property grouping turns out to be receipts-only, so it is not the lines 61-62 package. - Box 3.1's "a third of the yield" is staged from the review's quotation, the salary-sacrifice recosting carries its AB2025 key, REPORT.md goes, and no note names a local machine path. obr.uk was unreachable here, so no primary was re-opened. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_01LwQq6HuQUQ1obM9JvyhJbE --- .../harvest-uk-ab2026-2026-10-29/COUNTS.json | 14 +- .../PROPOSED_KEYS.md | 35 ++- .../uk_obr_tables/NOTES.md | 270 ++++++++++++++++++ .../uk_obr_tables/claims_staged.jsonl.gz | Bin 0 -> 111394 bytes .../uk_obr_tables/manifest.jsonl | 3 + 5 files changed, 314 insertions(+), 8 deletions(-) create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/NOTES.md create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/claims_staged.jsonl.gz create mode 100644 sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/manifest.jsonl diff --git a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json index 13fb60450..ce16d3c0e 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json +++ b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json @@ -56,6 +56,14 @@ "ippr" ] }, + "uk_obr_tables": { + "rows": 2797, + "primaries": 3, + "claims_sha256": "0fa9a6f30a898b4b0a8f89de52b1c878621144394d8f7d289778c36ba250e24c", + "sources": [ + "obr_efo" + ] + }, "uk_rf_ab2026": { "rows": 766, "primaries": 7, @@ -92,9 +100,9 @@ ] }, "_totals": { - "rows": 3094, - "restated": 378, - "families": 11, + "rows": 5891, + "restated": 401, + "families": 12, "pinned_on": "2026-10-29" } } diff --git a/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md b/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md index 950c35200..9d199dea5 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md +++ b/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md @@ -4,12 +4,13 @@ Generated by `pipeline/ab2026_proposed_keys.py` from the staged families; don't A row whose scope matches no registered measure has `measure_key: null` and names one of these keys in `conditions.proposed_measure_key` (README rule 2). A key is a measure with no Table 6.1 line (`ab2026__`), a package (`ab2026__package_`) or an option (`ab2026_option__`). Each family's NOTES.md gives the key's scope as Table 6.1 lines or the producer's own words. Tranche 2 registers each key in `data/uk/ab2026_measures.json`, or folds it into a registered key, before the ingest. -65 proposed keys (14 measures, 39 packages, 12 options) on 420 rows. +90 proposed keys (17 measures, 60 packages, 13 options) on 955 rows. | proposed key | kind | rows | families | producers | |---|---|--:|---|---| | `ab2026__british_business_bank_defence_commitments_accelerated` | measure | 9 | uk_hmt_costings | hm_treasury | | `ab2026__defence_investment_plan` | measure | 1 | uk_hmt_redbook | hm_treasury | +| `ab2026__departmental_spending_growth_2031_32` | measure | 13 | uk_obr_tables | obr_efo | | `ab2026__energy_debt_circuit_breaker` | measure | 3 | uk_hmt_redbook | hm_treasury | | `ab2026__english_devolution_agreements` | measure | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__gas_standing_charge_moved_to_unit_rates` | measure | 34 | uk_ifs_ab2026, uk_rf_ab2026 | ifs, resolution_foundation | @@ -22,20 +23,37 @@ A row whose scope matches no registered measure has `measure_key: null` and name | `ab2026__package_breathing_space_for_households` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_budget_barnett_consequentials` | package | 4 | uk_hmt_redbook | hm_treasury | | `ab2026__package_business_rates_rhl_sbrr_and_transitional_support` | package | 2 | uk_hmt_redbook | hm_treasury | +| `ab2026__package_carbon_price_support_and_cfd_levies` | package | 3 | uk_obr_tables | obr_efo | +| `ab2026__package_civil_service_transformation_programme` | package | 1 | uk_obr_tables | obr_efo | +| `ab2026__package_compliance_measures_announced` | package | 1 | uk_obr_tables | obr_efo | +| `ab2026__package_corporation_and_bank_tax_changes` | package | 7 | uk_obr_tables | obr_efo | +| `ab2026__package_defence_investment_plan` | package | 1 | uk_obr_tables | obr_efo | +| `ab2026__package_dividend_and_capital_gains_tax_changes` | package | 6 | uk_obr_tables | obr_efo | | `ab2026__package_dwp_fraud_and_error_measures` | package | 1 | uk_hmt_redbook | hm_treasury | +| `ab2026__package_fuel_duty_and_electricity_vat` | package | 11 | uk_obr_tables | obr_efo | +| `ab2026__package_fuel_duty_freezes_since_2010_11` | package | 1 | uk_obr_tables | obr_efo | +| `ab2026__package_government_decisions_since_march_2026` | package | 347 | uk_obr_tables | obr_efo | | `ab2026__package_hmrc_tax_gap_and_online_marketplace` | package | 7 | uk_hmt_redbook | hm_treasury | | `ab2026__package_hmt_distributional_budget_2026` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_hmt_distributional_since_ab2024` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_household_energy_payment_and_electricity_vat` | package | 5 | uk_rf_ab2026 | resolution_foundation | +| `ab2026__package_national_wealth_fund_and_british_business_bank` | package | 4 | uk_obr_tables | obr_efo | +| `ab2026__package_non_scorecard_decisions` | package | 12 | uk_obr_tables | obr_efo | | `ab2026__package_obr_direct_effect_of_tax_measures` | package | 13 | uk_rf_ab2026 | resolution_foundation | | `ab2026__package_obr_forecast_policy_effects` | package | 24 | uk_hmt_redbook | hm_treasury | | `ab2026__package_obr_tax_administration_compliance_and_debt_collection` | package | 7 | uk_rf_ab2026 | resolution_foundation | | `ab2026__package_obr_total_effect_of_government_decisions` | package | 3 | uk_rf_ab2026 | resolution_foundation | +| `ab2026__package_other_ame_spending_decisions` | package | 6 | uk_obr_tables | obr_efo | +| `ab2026__package_other_departmental_spending_changes` | package | 7 | uk_obr_tables | obr_efo | +| `ab2026__package_other_hmrc_compliance_and_debt` | package | 1 | uk_obr_tables | obr_efo | +| `ab2026__package_other_tax_measures` | package | 6 | uk_obr_tables | obr_efo | | `ab2026__package_parliament_fuel_duty_decisions` | package | 3 | uk_hmt_redbook | hm_treasury | | `ab2026__package_parliament_regulatory_burden_reduction` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_parliament_tax_gap_measures` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_parliament_welfare_fraud_and_error_savings` | package | 1 | uk_hmt_redbook | hm_treasury | +| `ab2026__package_planned_capital_spending_reductions` | package | 1 | uk_obr_tables | obr_efo | | `ab2026__package_property_and_savings_income_rates` | package | 33 | uk_hmt_redbook, uk_ifs_ab2026, uk_rf_ab2026, uk_tpa | hm_treasury, ifs, resolution_foundation, tax_policy_associates | +| `ab2026__package_property_and_savings_income_rates_receipts` | package | 7 | uk_obr_tables | obr_efo | | `ab2026__package_rdel_decisions_excluding_household_energy_payment` | package | 3 | uk_rf_ab2026 | resolution_foundation | | `ab2026__package_regulation_tariff_and_sandbox_business_savings` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_rf_ab26_permanent_personal_measures` | package | 19 | uk_rf_ab2026 | resolution_foundation | @@ -43,6 +61,10 @@ A row whose scope matches no registered measure has `measure_key: null` and name | `ab2026__package_rf_energy_bill_support_announced_at_budget` | package | 2 | uk_rf_ab2026 | resolution_foundation | | `ab2026__package_rf_other_tax_measures` | package | 6 | uk_rf_ab2026 | resolution_foundation | | `ab2026__package_rf_parliament_permanent_tax_and_benefit_measures` | package | 42 | uk_rf_ab2026 | resolution_foundation | +| `ab2026__package_spending_decisions_since_march_2026` | package | 7 | uk_obr_tables | obr_efo | +| `ab2026__package_tax_administration_compliance_and_debt_collection` | package | 6 | uk_obr_tables | obr_efo | +| `ab2026__package_tax_decisions_since_march_2026` | package | 6 | uk_obr_tables | obr_efo | +| `ab2026__package_temporary_business_rates_reliefs` | package | 1 | uk_obr_tables | obr_efo | | `ab2026__package_tpa_business_rates` | package | 8 | uk_tpa | tax_policy_associates | | `ab2026__package_tpa_cgt_founders_relief_and_badr` | package | 1 | uk_tpa | tax_policy_associates | | `ab2026__package_tpa_energy_and_environment` | package | 6 | uk_tpa | tax_policy_associates | @@ -56,18 +78,21 @@ A row whose scope matches no registered measure has `measure_key: null` and name | `ab2026__package_tpa_savings_property_income_and_cash_isa` | package | 7 | uk_tpa | tax_policy_associates | | `ab2026__package_tpa_tariffs` | package | 2 | uk_tpa | tax_policy_associates | | `ab2026__package_tpa_temporary_vat_and_fuel_cuts` | package | 6 | uk_tpa | tax_policy_associates | -| `ab2026__private_landlord_register_fee_65` | measure | 7 | uk_hmt_costings | hm_treasury | -| `ab2026__pwlb_debt_relief_councils_severe_distress` | measure | 6 | uk_hmt_costings | hm_treasury | +| `ab2026__private_landlord_register_fee_65` | measure | 25 | uk_hmt_costings, uk_obr_tables | hm_treasury, obr_efo | +| `ab2026__public_sector_pensions_2024_valuations` | measure | 27 | uk_obr_tables | obr_efo | +| `ab2026__pwlb_debt_relief_councils_severe_distress` | measure | 18 | uk_hmt_costings, uk_obr_tables | hm_treasury, obr_efo | | `ab2026__regulatory_administrative_savings` | measure | 5 | uk_hmt_redbook | hm_treasury | | `ab2026__rolls_royce_gas_turbine_programme` | measure | 1 | uk_hmt_redbook | hm_treasury | -| `ab2026__state_pension_triple_lock_adjustment` | measure | 2 | uk_hmt_redbook | hm_treasury | +| `ab2026__scottish_air_passenger_duty_devolution` | measure | 18 | uk_obr_tables | obr_efo | +| `ab2026__state_pension_triple_lock_adjustment` | measure | 4 | uk_hmt_redbook, uk_obr_tables | hm_treasury, obr_efo | | `ab2026_option__cgt_reform_package` | option | 9 | uk_centax | centax | | `ab2026_option__childcare_100k_entitlement_taper` | option | 3 | uk_centax | centax | | `ab2026_option__childcare_100k_threshold_raised` | option | 2 | uk_centax | centax | | `ab2026_option__childcare_100k_threshold_removed` | option | 3 | uk_centax | centax | | `ab2026_option__childcare_100k_threshold_removed_first_15_hours_under_3s` | option | 1 | uk_centax | centax | -| `ab2026_option__fuel_duty_continued_freeze` | option | 6 | uk_rf_ab2026 | resolution_foundation | +| `ab2026_option__fuel_duty_continued_freeze` | option | 7 | uk_obr_tables, uk_rf_ab2026 | obr_efo, resolution_foundation | | `ab2026_option__fuel_duty_rpi_uprating_since_2011` | option | 2 | uk_rf_ab2026 | resolution_foundation | +| `ab2026_option__health_foundation_social_care_reforms` | option | 2 | uk_obr_tables | obr_efo | | `ab2026_option__income_tax_5pc_share_devolved_to_mayors` | option | 1 | uk_ippr | ippr | | `ab2026_option__lha_unfreeze_2027_28` | option | 17 | uk_rf_ab2026 | resolution_foundation | | `ab2026_option__nics_on_investment_income` | option | 10 | uk_centax | centax | diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/NOTES.md new file mode 100644 index 000000000..1a74e2345 --- /dev/null +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/NOTES.md @@ -0,0 +1,270 @@ +# uk_obr_tables — October 2026 Economic and fiscal outlook + +**Working files.** `stage.py`, the file meant wherever these notes say `stage.py`, lives on branch `uk/ab2026-harvest-staging` at `harvest-staging/ab2026/lanes/obr-tables/uk_obr_tables/stage.py`; this branch carries only NOTES.md, `manifest.jsonl` and `claims_staged.jsonl.gz`. The repository carries no primary bytes: fetch each primary from its `manifest.jsonl` URL (or, while obr.uk refuses non-browser clients, the coordinator's local copies) and check its sha256. + +Staged 2026-10-29; reworked 3 November 2026 under the OBR review (see "Review fixes" at the end). Source `obr_efo`. **2,797 rows (2,891 before the review), 3 primaries, 98 individual registered measures plus the registered total-policy package.** All outputs are confined to this family. `stage.py` embeds all reviewed rows as metadata templates and cell data; rerunning needs only the Python standard library and reproduces the gzip without primary files or network access. + +## Access recipe and primaries + +Each resolved URL in the supplied OBR `PROVENANCE.md` was attempted once using `curl -sSL` and the exact Safari User-Agent requested in the assignment. The EFO landing page returned **403** at 2026-10-29T15:51:17Z; the chapter-2 workbook and detailed-table ZIP resolved URLs returned **403** at 15:51:18Z. No bypass was attempted. We used the supplied local policy and aggregate workbook members and PDF; every row and manifest entry records `access: "local copy"`. + +The coordinator supplied the local copies. The lane kept exact byte copies, failed-response headers and bodies and fetch timestamps in its local primaries folder, never committed. Workbooks were read with openpyxl `data_only=True`; the PDF with `pdftotext -layout`. `shasum -a 256` verified all three primaries. Both workbook member hashes match the browser-download provenance from 28 October, including the ZIP's recorded Last-Modified of Wed, 28 Oct 2026 14:05:59 GMT. + +The resolved ZIP URL is the publication URL for both members, with the member name in each row's note and the manifest. The ZIP container itself was not supplied locally and is not represented as if it had been read. The PDF attachment URL is absent from PROVENANCE; the blocked landing page prevented discovering it. Its manifest and rows therefore use the EFO landing URL and identify the supplied local PDF filename. The cover/imprint say October 2026; publication date 2026-10-28 follows the EFO-page date recorded in PROVENANCE. No replacement version could be compared after the refusals, and no changed-primary claim is made. + +| Primary file | Rows | Bytes | SHA-256 | +|---|---:|---:|---| +| efo-october-2026-detailed-forecast-tables-policy.xlsx | 2,520 | 106,602 | `cdcc2a06b100241bf3010bd159a4e5d43daef2348a3850797b1e88a308fd45fd` | +| economic-and-fiscal-outlook-october-2026-web-accessible.pdf | 264 | 1,754,344 | `9e0e9c340e35e603066f3124a614fcd68feadb3d5d10d1f76fb536e150e6225e` | +| efo-october-2026-detailed-forecast-tables-aggregates.xlsx | 13 | 218,000 | `3ffa3f1e63667b2ae9fb9626dc1da881673227d43aeeb52f8c3db168a5c75d6b` | + +## Coverage + +- Policy workbook Table 1.1: **2,520** cells, every numeric row series (420) across all six fiscal years, 2026-27 through 2031-32. This includes 2,112 numbered-measure component/total cells, 48 fiscally neutral cells, 30 non-scorecard cells and 330 summary cells. +- EFO chapter 3: **222** rows (316 before the review): Table 3.1 114 (its "Spending measures" and "Tax measures" rows repeat its "Direct effect" rows and were dropped); Table 3.3 35 costings plus 7 categorical uncertainty ratings; policy/behavioural prose and Box 3.1 66. Table 3.2's 47 funding cells are design (rule 8) and are not staged. +- Headline fiscal context: **55** rows: 13 workbook cells (PSNB cash six years; ex-BoE PSND/GDP seven years) and 42 PDF cells (Table 7.1 headroom 13; Annex A Table A.9 PSNB, PSNFL and ex-BoE PSND 29). The current-budget and PSNFL rules include current pre-measures, post-measures and the printed memo excluding fuel-duty rises; welfare-cap margin is included. Forecast duplicates between the PDF and aggregate workbook are represented once. + +## Conventions + +- **Values and precision:** policy workbook values use their displayed integer £million formats (`0`, `#,##0`; eight exact General-format zeros), multiplied by 1e6. Unprinted cached decimal precision is not a claim. Decimal rounding is half away from zero; a stored negative fraction that displays zero is `0`. A LibreOffice PDF rendering independently confirmed these display conventions. Aggregate workbook values use displayed one-decimal precision. PDF £billion values multiply by 1e9 and £million by 1e6; no ratios, sums or inferred annual paths are staged. +- **Signs:** policy workbook prints “This table uses the convention that a negative sign implies a loss to the Exchequer (and is therefore an increase in PSNB).” PDF Tables 3.1/3.3 print “A positive sign implies an increase in borrowing.” Both are retained. The indirect-effect workbook row's conflict with that workbook legend is explicitly noted in those rows. Prose uses positive magnitudes for costs, yields or behavioural erosion exactly as worded; per-row conditions state the meaning. Funding amounts are positive cost provision, except the printed negative change. Headroom is positive margin, negative breach. +- **Period:** FY end year, with the printed `fy`; quarterly CPI peaks use calendar-year `point_in_time` and quarter detail without `fy`. An annual average or recurring prose amount appears once at the terminal forecast FY, with its interval in `aggregation`, rather than manufacturing six yearly values. The 5,500 previously announced HMRC recruits have no separately printed delivery year and are an announcement input at point-in-time 2026; the separate 670 figure retains its printed 2029-30 horizon. Debt stocks preserve end-March timing and end-March-centred GDP denominators. +- **Central:** ordinary policy tables/prose are `point` estimates of the single final forecast; aggregate forecasts are `central`, meaning the OBR central forecast, and observed 2025-26 outturns are `point`. No statistical confidence interval is inferred. “High” ratings use categorical `value: 0` as the precedent's schema placeholder, not a numerical score. Health Foundation range endpoints span alternative reform options, not uncertainty for one policy. +- **Attribution:** 2,774 `own`, 23 `restated` (2,821 and 70 before Table 3.2's 47 rows went). OBR-certified/HMRC/DWP-scrutinised costings follow the AB2025 `own` convention; OBR fiscal judgments use `obr_efo_forecast`. Twelve workbook fuel cells and two PDF fiscal fuel costings are restated `hmt_costing_uncertified`: EFO p.59 footnote 1 says the final Government costing arrived after certification. Table 3.2's 47 funding cells name HM Treasury as original source; six latest ONS outturn aggregate cells name ONS. Two social-care options name Health Foundation; one FTA effect names the Government/DBT impact assessment. OBR's CPI judgments remain its own estimates. Funding/ONS facts use `administrative_fact`; modeled estimates use `different_model`. +- **Scope:** all 98 numbered workbook titles match registry titles after whitespace/footnote normalization. Fiscal-head components retain their full measure's key, with component conditions; totals and components must not be summed as independent estimates. Only workbook row552 and the PDF's explicit HMT-policy-table memo use the registered total-policy package. All other groupings use null plus a proposed key. Context levels use null without a proposed measure. The lone prior-policy salary-sacrifice recosting stays null as assigned; it is not recast as an AB2026 intervention. The AB2025 OBR family actually contains no headline fiscal aggregates; the live `Metric` enum supplies `fiscal_headroom`, and cross-family proposals supply the closest aggregate names. +- **Quotes/order:** full numeric/rating cells or full OGL prose sentences, preserving the printed number; PDF whitespace/line wraps normalized, footnote superscripts excluded from sentence text. Reading order is policy workbook row/column, PDF page/position, then aggregate workbook table/row/column. PDF printed page + 3 is the physical PDF page. + +## Coverage by measure_key, including null + +100 keys on 2,207 rows, 1,621 of them `row_type: measure_component`. Of the 590 null rows, 535 carry a proposed key and 55 are context (fiscal aggregates, headroom and ratings). Row 2828's recosting key is AB2025's (`conditions.recosting_of: autumn_budget_2025`). + +| measure_key | Rows | +|---|---:| +| `ab2025__salary_sacrifice_pension_nics_cap_2000` | 1 | +| `ab2026__advance_corporation_tax_abolished` | 12 | +| `ab2026__alcohol_duty_small_producer_relief` | 12 | +| `ab2026__artificial_intelligence_package` | 18 | +| `ab2026__badr_lifetime_limit_500000` | 43 | +| `ab2026__bank_surcharge_6pc_temporary` | 21 | +| `ab2026__benefits_in_kind_payrolling_delay` | 36 | +| `ab2026__benefits_removed_hospitalised_serious_offenders` | 18 | +| `ab2026__british_business_bank_backing_innovation` | 18 | +| `ab2026__british_industrial_competitiveness_scheme` | 24 | +| `ab2026__bus_fare_cap_2_england_outside_london_2027` | 18 | +| `ab2026__business_rates_ev_charging_relief` | 48 | +| `ab2026__business_rates_pubs_relief_35pc` | 48 | +| `ab2026__business_rates_retention_mayoral_authorities` | 31 | +| `ab2026__business_rates_rhl_transitional_support` | 48 | +| `ab2026__business_rates_sbrr_rhl_thresholds_raised` | 48 | +| `ab2026__business_rates_vape_shops_and_agcs_reliefs_removed` | 48 | +| `ab2026__business_tariff_suspensions` | 12 | +| `ab2026__cambridge_development_corporation` | 12 | +| `ab2026__carbon_price_support_removed` | 18 | +| `ab2026__carers_allowance_earnings_taper` | 18 | +| `ab2026__cash_isa_limit_held_at_20000` | 12 | +| `ab2026__civil_service_admin_budgets_cut_20pc` | 12 | +| `ab2026__civil_service_ai_fund` | 12 | +| `ab2026__climate_change_levy_rates_2028` | 12 | +| `ab2026__company_car_tax_rates_2030_31` | 36 | +| `ab2026__defence_admin_savings_reinvested` | 12 | +| `ab2026__defence_investment_plan_capital_reallocation` | 18 | +| `ab2026__defence_investment_plan_further_savings` | 18 | +| `ab2026__distributions_framework_modernised` | 19 | +| `ab2026__dwp_fraud_and_error_prevention` | 31 | +| `ab2026__early_years_books_access` | 12 | +| `ab2026__electricity_generator_levy_55pc` | 12 | +| `ab2026__electricity_vat_zero_rate_to_july_2027` | 20 | +| `ab2026__emi2_option_limit_120000` | 48 | +| `ab2026__ets_energy_from_waste_delay` | 12 | +| `ab2026__ev_charging_public_chargers` | 12 | +| `ab2026__first_time_buyer_isa` | 12 | +| `ab2026__food_and_heating_oil_tariff_suspensions` | 12 | +| `ab2026__foreign_permanent_establishment_exemption_mandatory` | 12 | +| `ab2026__foreign_sovereign_investments_taxation` | 61 | +| `ab2026__founders_relief_cgt` | 37 | +| `ab2026__franchised_higher_education_registration` | 18 | +| `ab2026__free_bus_travel_children_august_2026` | 12 | +| `ab2026__fuel_duty_5p_cut_extended_and_rpi_rise_deferred` | 15 | +| `ab2026__further_priorities_nhs_and_vawg` | 24 | +| `ab2026__gas_distribution_revenue_recognition_spread` | 12 | +| `ab2026__good_growth_funds` | 18 | +| `ab2026__growth_guarantee_scheme_expansion` | 21 | +| `ab2026__hgv_ved_1_for_12_months` | 12 | +| `ab2026__high_needs_local_authority_spending` | 18 | +| `ab2026__hmrc_tax_gap_package` | 105 | +| `ab2026__household_energy_payment_100_bands_a_to_d` | 18 | +| `ab2026__hvcts_deferrals_discounts_exemptions` | 42 | +| `ab2026__infected_blood_compensation_additional_report` | 12 | +| `ab2026__investment_for_growth` | 24 | +| `ab2026__landfill_communities_fund_credit_cap` | 24 | +| `ab2026__leeds_city_fund` | 24 | +| `ab2026__lifetime_isa_withdrawal_charge_20pc_above_cap` | 12 | +| `ab2026__local_apprenticeship_services` | 12 | +| `ab2026__low_value_imports_customs_relief_removed` | 60 | +| `ab2026__machine_games_duty_higher_stake_25pc` | 18 | +| `ab2026__mileage_allowance_payments_55p` | 36 | +| `ab2026__misogyny_prevention_programmes` | 12 | +| `ab2026__national_wealth_fund_capital` | 18 | +| `ab2026__online_marketplace_liability_uk_businesses` | 18 | +| `ab2026__package_total_policy_decisions` | 12 | +| `ab2026__parental_allowance_fathers_partners_adopters` | 12 | +| `ab2026__pension_death_benefits_two_year_rule_removed` | 30 | +| `ab2026__pension_tax_allowance` | 54 | +| `ab2026__private_hire_operators_vat_full_fare` | 12 | +| `ab2026__property_income_rates_plus_2pp` | 44 | +| `ab2026__rd_relief_claim_notification_form_removed` | 18 | +| `ab2026__red_diesel_duty_cut` | 12 | +| `ab2026__regional_growth_investments` | 18 | +| `ab2026__rough_sleeping_investment` | 24 | +| `ab2026__rough_sleeping_reprioritised_mhclg_budgets` | 18 | +| `ab2026__savings_income_rates_plus_2pp` | 20 | +| `ab2026__securities_transfer_tax` | 12 | +| `ab2026__social_and_affordable_homes_programme` | 12 | +| `ab2026__steel_tariff_and_quotas` | 18 | +| `ab2026__strategic_authorities_capacity_funding` | 12 | +| `ab2026__student_loan_interest_cap_6pc_plan_2_and_3` | 18 | +| `ab2026__supported_housing_licensing` | 12 | +| `ab2026__trading_allowance_3000` | 36 | +| `ab2026__uc_surplus_earnings_threshold_2500_extended` | 12 | +| `ab2026__uk_export_finance_defence_capacity` | 13 | +| `ab2026__uk_india_ceta_and_double_contributions_convention` | 19 | +| `ab2026__ukraine_winter_air_defence` | 18 | +| `ab2026__union_learning_fund` | 12 | +| `ab2026__us_llc_double_taxation_relief` | 12 | +| `ab2026__vat_5pc_family_days_out_summer_2026` | 12 | +| `ab2026__vat_medical_deputies_staffing` | 12 | +| `ab2026__vat_reverse_charge_overseas_services_restricted` | 12 | +| `ab2026__vat_zero_rate_donated_medicines` | 12 | +| `ab2026__vat_zero_rate_land_for_social_housing` | 12 | +| `ab2026__waste_crime_landfill_tax_clearances` | 18 | +| `ab2026__writing_down_allowance_12pc_and_fya_50pc` | 43 | +| `ab2026__your_first_home_equity_loan` | 19 | +| `ab2026__youth_guarantee_and_apprenticeships_spring_2026` | 18 | +| `null` | 590 | + +## Proposed measure keys and exact producer scopes + +All of these rows have `measure_key: null`. Names follow README rule 2 as harmonised on 3 November (review finding 9): a single measure is `ab2026__`, and a scope package's slug names the scope and carries no `obr`. Where the OBR publishes only a named residual group, its line decomposition stays unresolved; no guessed membership or registry key is attached. + +| Proposed key | Rows | Scope | +|---|---:|---| +| `ab2026__departmental_spending_growth_2031_32` | 13 | No Table 6.1 lines. “Departmental spending: Grow day-to-day spending by 0.7% in real terms in 2031-32, matching 0.7% average annual real terms growth from 2029-30 to 2030-31”. PDF describes the same baseline change as a £5.5bn reduction in the 2031-32 pre-measures RDEL assumption. (Was `ab2026__package_obr_departmental_spending_growth_2031_32`.) | +| `ab2026__package_carbon_price_support_and_cfd_levies` | 3 | Table 6.1 line 71 plus funding for Low Carbon Contracts Company to reduce CfD levies (no separately identifiable line in Table 6.1); related support for BICS/Supercharger includes line 70. OBR paragraph 3.19 combines these measures and dates CPS abolition April 2027, while HMT line 71 says April 2028. (Was `ab2026__package_obr_carbon_price_support_and_cfd_levies`.) | +| `ab2026__package_civil_service_transformation_programme` | 1 | Table 6.1 lines 63 and 64 net programme: “civil service transformation programme ... reduce resource spending”. (Was `ab2026__package_obr_civil_service_transformation_programme`.) | +| `ab2026__package_compliance_measures_announced` | 1 | No Table 6.1 line list. Box 3.1 (p.66): "Of the compliance measures announced in this Budget". The OBR does not list them; they are probably Table 6.1 line 45's items and line 46, but that is not stated, so the scope is the OBR's words. | +| `ab2026__package_corporation_and_bank_tax_changes` | 7 | OBR “Changes to corporation and bank taxation”; paragraph3.16 explicitly discusses line58 WDA/FYA, line55 bank surcharge, and line54 overseas sovereign immunity. Complete Table6.1 line decomposition and fiscal-head allocation are not published; unresolved, no inferred additional membership. (Was `ab2026__package_obr_corporation_and_bank_tax_changes`.) | +| `ab2026__package_defence_investment_plan` | 1 | OBR “funding for the defence investment plan” / “the defence investment plan, which increases departmental resource spending (RDEL)”; related to Table6.1 Defence Investment Plan entries, but complete line decomposition is not published and is unresolved. No inferred40–43 membership. (Was `ab2026__package_obr_defence_investment_plan`.) Shared with the red book's Defence Investment Plan row. | +| `ab2026__package_dividend_and_capital_gains_tax_changes` | 6 | Table 6.1 lines 26,49 and 60: OBR “Changes to dividend and capital gains taxation”. (Was `ab2026__package_obr_dividend_and_capital_gains_tax_changes`.) | +| `ab2026__package_fuel_duty_and_electricity_vat` | 11 | Table 6.1 lines 2 and 3 only; OBR “Fuel duty freeze and VAT off domestic energy bills”. Excludes household council-tax rebate line 1. (Was `ab2026__package_obr_fuel_duty_and_electricity_vat`.) | +| `ab2026__package_fuel_duty_freezes_since_2010_11` | 1 | OBR cumulative 2010-11 to 2027-28 fuel duty freezes and 5p cut relative to stated RPI uprating; includes Table6.1 line3 and earlier freezes since2010-11. Not the cost of the AB2026 measure alone. (Was `ab2026__package_obr_historical_fuel_duty_freezes_and_5p_cut`.) | +| `ab2026__package_government_decisions_since_march_2026` | 347 | Table 6.1 lines 1–98, plus the three fiscally neutral measures (PWLB debt relief, Scottish APD devolution, landlord registration fee) and the two non-scorecard policies listed above. OBR “Total effect of Government decisions”, with direct, indirect, receipt, expenditure and fiscal-head projections distinguished in conditions. Includes indirect effects where stated; not the HMT scorecard-only package. (Was `ab2026__package_obr_government_decisions_since_march_2026`.) Shared with the red book's Table 1.3 "Total effect of policy decisions" and RF's restatement. | +| `ab2026__package_national_wealth_fund_and_british_business_bank` | 4 | Table 6.1 lines 29 and 30 together, plus their associated financial transactions in Table 6.2; OBR “funding available for the National Wealth Fund and the British Business Bank ... to invest in financial assets”. (Was `ab2026__package_obr_national_wealth_fund_and_british_business_bank`.) | +| `ab2026__package_non_scorecard_decisions` | 12 | No Table 6.1 lines. “Direct effect of non-scorecard decisions”: exactly the departmental spending assumption and public-pension valuation changes above. (Was `ab2026__package_obr_non_scorecard_decisions`.) | +| `ab2026__package_other_ame_spending_decisions` | 6 | OBR Table 3.1 “Other AME spending decisions”; excludes net unfunded pensions valuation. Producer does not supply a full Table 6.1 line-number decomposition in PDF. (Was `ab2026__package_obr_other_ame_spending_decisions`.) | +| `ab2026__package_other_departmental_spending_changes` | 7 | OBR Table 3.1 “Other departmental spending changes”, excluding council tax rebate line 1; includes defence, capital spending and the £5.5bn 2031-32 pre-measures RDEL reduction, which has no Table 6.1 line. Producer does not supply a full line-number decomposition in PDF. (Was `ab2026__package_obr_other_departmental_spending_changes`.) | +| `ab2026__package_other_hmrc_compliance_and_debt` | 1 | OBR paragraph 3.13 residual “Other measures”: direct debt recovery, e-invoicing, debt collection agencies (components of Table 6.1 line 45), and online marketplace liability (line 46). (Was `ab2026__package_obr_other_hmrc_compliance_and_debt`.) | +| `ab2026__package_other_tax_measures` | 6 | OBR Table 3.1 “Other tax measures”; residual of Table 6.1 tax measures after tax administration/compliance/debt, corporation/bank, savings/property, dividend/CGT and fuel/electricity VAT groups. PDF does not enumerate a full line-number decomposition. (Was `ab2026__package_obr_other_tax_measures`.) | +| `ab2026__package_planned_capital_spending_reductions` | 1 | OBR “Reductions in planned capital spending” / “changes to departmental capital spending”; PDF does not enumerate the exact Table 6.1 lines. (Was `ab2026__package_obr_planned_capital_spending_reductions`.) | +| `ab2026__package_property_and_savings_income_rates_receipts` | 7 | The receipts of Table 6.1 lines 61 and 62: Table 3.1's “Changes to savings and property income tax rates”. (Was `ab2026__package_obr_savings_and_property_income_tax_rates`.) **Receipts only (3 November):** Table 3.1 prints 0.1 in 2027-28 and 0.9 in 2029-30, against lines 61 and 62's totals of 160 and 954 (£m), but in line with their receipts (Table 3.3 gives line 61's as 2, 543 and 400 from 2027-28). So it is not `ab2026__package_property_and_savings_income_rates`, which is exactly lines 61 and 62. | +| `ab2026__package_spending_decisions_since_march_2026` | 7 | OBR “Direct effect of spending measures” / “Spending measures”: expenditure-side effects of decisions since March, including unfunded public service pensions valuation and the £5.5bn 2031-32 RDEL assumption reduction. This is not the HMT total-spending package. Table 6.1 lines 1–98 can affect expenditure fiscal heads; a complete narrower line decomposition is not published and remains unresolved. (Was `ab2026__package_obr_spending_decisions_since_march`.) | +| `ab2026__package_tax_administration_compliance_and_debt_collection` | 6 | OBR “Tax administration, compliance and debt collection measures”, including Table6.1 line45 (HMRC tax-gap package) and line46 (online marketplace liability), with funding/receipt interactions. PDF group yield (£3.6bn2031-32) differs from standalone line45 (£3.160bn) and must not be attached to line45 alone. (Was `ab2026__package_obr_tax_administration_compliance_and_debt`.) Shared with RF's restatement. | +| `ab2026__package_tax_decisions_since_march_2026` | 6 | OBR Table 3.1 "Direct effect of tax measures": the direct effect of tax decisions since March 2026, as the OBR classifies them. Shared with RF's restatement of the same row. Before 3 November these rows sat under the total package with `component: tax measures (OBR classification)`. | +| `ab2026__package_temporary_business_rates_reliefs` | 1 | Table 6.1 lines 12,13,14: “package of temporary business rates measures” including real-terms freeze for certain sectors, pubs/live-music reliefs, small-business thresholds and extension of transitional support, partly offset by transitional relief supplement. (Was `ab2026__package_obr_temporary_business_rates_reliefs`.) | +| `ab2026__private_landlord_register_fee_65` | 18 | No Table 6.1 lines. “Landlord registration fee”; receipts, current DEL and total. (Was `ab2026__package_obr_landlord_registration_fee`.) The costings family's spelling for the same measure. | +| `ab2026__public_sector_pensions_2024_valuations` | 27 | No Table 6.1 lines. “Public Sector Pensions: Net impact of the 2024 valuations”; departmental reductions, net public service pension payments and their net total. PDF calls this “Net impact of unfunded pensions valuation”. (Was `ab2026__package_obr_public_sector_pensions_2024_valuations`.) | +| `ab2026__pwlb_debt_relief_councils_severe_distress` | 12 | No Table 6.1 lines. “Provide Public Works Loan Board (PWLB) debt relief for councils in severe financial distress”; capital AME and total. (Was `ab2026__package_obr_pwlb_distress_debt_relief`.) The costings family's spelling for the same measure. | +| `ab2026__scottish_air_passenger_duty_devolution` | 18 | No Table 6.1 lines. “Devolving Air Passenger Duty to the Scottish Government in April 2027”; Scottish BGA, self-financed expenditure and total. (Was `ab2026__package_obr_scottish_air_passenger_duty_devolution`.) | +| `ab2026__state_pension_triple_lock_adjustment` | 2 | No Table6.1 line: OBR paragraph3.49 adjusted state pension triple lock expected April2030, retaining CPI/2.5percent floor but linking catch-up to2029-30 earnings ratio. Figures3.51 are historical simulations1990–2025, not the medium-term policy costing. (Was `ab2026__package_obr_adjusted_state_pension_triple_lock`.) The red book's spelling for the same measure. | +| `ab2026_option__fuel_duty_continued_freeze` | 1 | No Table6.1 line as an announced measure: OBR paragraph3.43 counterfactual “if fuel duty were to remain frozen until 2029-30”, extending line3 beyond July2027. (Was `ab2026_option__obr_fuel_duty_frozen_until_2029_30`.) Shared with RF, whose series for the OBR's constant-duty-rate alternative prints the same £4.3bn in 2029-30. | +| `ab2026_option__health_foundation_social_care_reforms` | 2 | No Table6.1 line: Health Foundation illustrative future social care reforms cited by OBR paragraph3.41, from limited home-care reform to wider schemes including residential care; costs additional to current spending in2035-36 per footnote13. This is not a settled AB2026 policy. (Was `ab2026_option__obr_health_foundation_social_care_reforms`.) | + + +## Metric, unit and baseline proposals + +No unregistered enum literals are used. `fiscal_headroom` is registered in the live models even though the older README calls it a proposal. + +| proposed_metric | Rows | Scope | +|---|---:|---| +| `cash_receipts_change` | 1 | Self-assessment overlapping transition payments, £17bn in 2029-30; assumed extraordinary accruals adjustment excludes the temporary cash effect from PSNB. | +| `tax_gap_share_of_compliance_yield` | 1 | Box 3.1: the share of this Budget's compliance measures' yield the OBR expects to reduce the tax gap ("a third"). | +| `costing_uncertainty_rating` | 7 | Table 3.3 High final ratings; categorical placeholder zero. | +| `psnb_share_of_gdp` | 7 | PSNB as percent of nominal GDP. | +| `psnd_share_of_gdp` | 7 | Ex-Bank-of-England PSND as percent of nominal GDP; reuses AB2025 cross-family spelling. | +| `psnfl_share_of_gdp` | 7 | PSNFL as percent of nominal GDP; reuses AB2025 cross-family spelling. | +| `public_sector_net_borrowing` | 7 | Annual PSNB cash level in GBP. | +| `public_sector_net_debt` | 7 | End-March PSND stock explicitly excluding Bank of England, GBP. | +| `public_sector_net_debt_change` | 3 | Marginal stock debt effects of NWF/BBB investment and Your First Home lending, paragraphs 3.20/3.22. | +| `public_sector_net_financial_liabilities` | 7 | End-March PSNFL stock, GBP. | + +Proposed units: `percent_of_gdp` (28 rows, nominal GDP denominator, including one prose package effect and fiscal shares/headroom); `rating_category` (7 High uncertainty ratings). + +Every `baseline_policy` is null; effect estimates use the following explicit unregistered baseline proposals. Rating, funding-level and aggregate context rows have no proposed baseline. + +| proposed_baseline, verbatim | Rows | +|---|---:| +| OBR pre-measures Autumn Budget 2026 policy parameters | 2,729 | +| Fuel duty policy reflected in OBR October 2026 economy forecast, confirmed before the final Budget parameters | 1 | +| Current social care spending (Health Foundation2024 scenarios, additional spending in2035-36) | 2 | +| Fuel duty rates increase in line with RPI inflation: stated policy since 2010-11 | 1 | +| Current triple lock uprating applied to historical outturn1990–2025 | 1 | +| Earnings uprating applied to historical outturn1990–2025 | 1 | + +## Not staged and unreachable material + +- **Design (rule 8), dropped on 3 November (review finding 1), 58 rows:** + - Table 3.2's 47 funding cells: HMRC and DWP baseline, Budget 2025, Budget 2026 additional and total compliance, debt and fraud-and-error funding ("the funding provision", Source: HM Treasury). The table prints no yield, and its Budget 2026 lines are not the scored costs (its DWP line prints −16, 188, 178, 188, 210, 214; the workbook's line-47 departmental spending prints 218, −194, −184, −195, −217, −221). + - Six staffing inputs: the 2,000, 200 and 150 DWP staff (para 3.12); the 5,500 and 670 HMRC staff and the "23 per cent increase" (para 3.13). They now sit in `conditions.staffing` on the £0.5bn and £0.4bn yield rows. + - Guarantee capacity, a contingent liability: UK Export Finance's £40bn and the Growth Guarantee Scheme's £7bn (para 3.21). + - A lending allocation: Your First Home's £0.7bn and £1bn of loans (para 3.22). + - A financing allocation: the NWF and BBB's "£1.9 billion a year" financing requirement (para 3.20), which the review marked probable; applied. +- **Repeats, dropped on 3 November (review finding 5, the coordinator's call), 37 rows.** The table cells are the staged rows, as in the AB2025 OBR family: text that repeats a Table 3.1 cell (paras 3.2, 3.3, 3.4, 3.11, 3.13, 3.16, 3.17 and 3.19: 19 rows) or earlier text (paras 3.11, 3.12, 3.13 and 3.16: 6 rows), and Table 3.1's "Spending measures" and "Tax measures" rows, which repeat its "Direct effect of spending measures" and "Direct effect of tax measures" rows (12 rows). + +- AnnexA has no policy-costing narrative beyond introductionA.1; current receipts/spending/economy forecast TablesA.1–A.8/A.11–A.12 are outside the policy-family scope. Headline fiscal aggregates inA.9 are owned by aggregate lane. +- No printed numeric Pension Tax Allowance or Carer’s Allowance taper claim appears in chapter3 or AnnexA. Whole-PDF search finds no PTA phrase; Carer’s Allowance appears only in a generic welfare footnote, not taper costing. These measures are covered by workbook rows. +- Charts3.1–3.4 and Box3.1 ChartsA/B contain no printed data labels for the policy values; no bars or paths read. Tick labels and dates are not claims. Box3.1 tax-gap context restates HMRC and several-Budget historical compliance yields and is not staged; its one AB2026 estimate, "a third of the yield", is staged (review finding 10). +- Policy design parameters (tax rates/thresholds, dates, litres/pence) without costing or behavioural-estimate content were not separately staged. +- Older-policy employment/housing/childcare monitoring examples in3.29–3.36 do not cost an AB2026 Table6.1 measure; no new AB2026 measure key invented. Paragraph3.28’s salary-sacrifice recosting carries AB2025's key with `recosting_of` (review finding 6). +- Paragraph3.48 FRS long-run pension/debt scenarios are current-policy/earlier-publication context, outside requested headline medium-term aggregate scope. Paragraph3.52 adjusted-triple-lock debt effects are explicitly2075-76=>2076; checker caps periods at2075, so neither25pp nor10pp long-run debt effect was staged or misdated. +- No additional primary access attempts owned by text lane; root records direct/local recipe and unreachable EFO landing page in manifest/NOTES. + +- The policy workbook contains no static/post-behavioural split or rating matrix, and no recosting table. Fiscal-head rows are not labeled as static or behavioural. Contents/navigation cells are not additional claims. +- The supplied chapter-2 charts workbook and the other economy, receipts, expenditure and debt-interest workbooks are outside the policy/headline scope. The chapter-2 fresh URL was 403; it was not used. No economy-wide GDP/CPI paths were staged; only explicit policy CPI/GDP effects. +- The EFO landing page, resolved detailed-tables ZIP and chapter-2 workbook URLs were unreached (403). The PDF attachment URL could not be discovered. No missing online content, new attachments or separate latest ratings database was presumed available or silently substituted. The seven printed PDF ratings are included. + + +## Contradictions and limitations retained + +- Workbook row549 “Total effect of Government decisions” displays **-8,905 / -50 / 1,322 / 5,484 / 6,425 / 14,093 £m**, while PDF Table 3.1 prints **6.3 / 7.0 / 2.1 / -2.6 / -3.5 / -9.3 £bn**. This exceeds a sign flip or rounding difference. Workbook direct row551 follows its negative-loss legend; indirect row555 has the same signs as the PDF borrowing effects. The workbook appears to mix conventions in its total. Both sources remain unchanged; row notes call out the inconsistency. +- Workbook/HMT precision differs: Household Energy Payment is -2,627 £m here versus -2,625 £m in HMT Table 6.1. Rounded components need not sum to the separately rounded totals. +- Paragraph3.19 dates carbon-price-support abolition April2027; HMT line71 says April2028. Its combined CPS/CfD/BICS/Supercharger scope uses a proposal, not CPS key. +- Paragraph3.16 dates first-year-allowance increase April2027; HMT line58 dates1January2027. The review found the date difference a description difference, not a scope difference, so the WDA/FYA row takes line 58's key. +- Paragraph3.21 says£7bn Growth Guarantee expansion; HMT line75 says£6.5bn additional SME lending. The £7bn and £6.5bn are different quantities of the same measure (guarantee capacity against lending supported); the borrowing rows take line 75's key and the £7bn is not staged. +- Table3.3/paragraph3.24 call reform Company Share Option Plan (CSOP); HMT line28 says Enterprise Management Incentives2 (EMI2). The workbook's line 28 total (−26 / −33 / −47 / −144 / −142) matches Table 3.3, so the review re-keyed it to line 28 as its tax head. +- Paragraph3.11 says departmental spending+£5bn2026-27; Table3.1 “Other departmental spending changes”+£5.4bn and separate rebate+£2.6bn. Preserved without inventing reconciliation. +- OBR tax aggregates include fiscal-head/classification differences from HMT:2031-32 OBR -£7.5bn against HMT+£6.970bn. No total-tax registry key used. +- Paragraph3.2 prose rounds2031-32 borrowing reduction to£12bn; Table3.1 prints-11.7bn. Both preserved. +- Paragraph3.41 says social-care options; footnote13 calls cited amounts “extra costs of healthcare”. Reported as printed, no correction. +- Health Foundation range is across different reform options, not statistical uncertainty about one fixed policy. +- Whole-Scheme UKEF exports in paragraph3.21 are not proven identical to defence-export capacity in HMT line44. India-UK FTA GDP claim does not include Double Contributions Convention component in HMT line77. The UKEF borrowing row matches line 44 and takes its key; the India FTA GDP row is a component of line 77. + +## Validation + +- **3 November:** the contract checker reports 2,797 rows, 0 problems, and a re-run of `stage.py` reproduces the gzip byte for byte (COUNTS.json pins its sha256). No primary was re-opened: the session's network policy refused obr.uk, and this lane's primaries are the coordinator's local copies. The fixes relabel, re-key or drop rows; no printed value changed. +- **29 October:** the lane ran the full-family checker, a deterministic gzip digest and 40 seeded random row checks, plus ten aggregate checks. Independent audits verified all 2,520 workbook cells and all 316 chapter-3 rows. Its REPORT.md, the only report file any family carried, was deleted under review finding 7. + +## Review fixes, 3 November 2026 + +The OBR review (`harvest-staging/ab2026/reviews/verify-obr.md` on the staging branch, verdict REWORK) found the numbers sound and the keying, typing and design rows systematic. Every fix is applied in `review_fixes()` in `stage.py`, with the coordinator's calls on findings 5, 6, 9 and 10: + +1. **Design rows (rule 8):** 58 rows dropped (see "Not staged"); the staffing figures moved into the yield rows' conditions. +2. **Components (rule 9):** 37 rows re-keyed to their measure. Direct recovery of debt, e-invoicing mandation, timely payment for self assessment and HMRC's compliance workforce are line 45 components; verification assurance officers are line 47's; the India FTA GDP effect is line 77's; Table 3.3's CSOP reform is line 28's tax head; UKEF's borrowing effect is line 44 (moved to 2030-31, as line 44 is 0 in 2031-32); the Growth Guarantee Scheme's borrowing effects are line 75; the WDA and FYA yield is line 58. Rows 2693 and 2747 were dropped under finding 5, so their re-keys fell away. Nine proposed keys went with these re-keys, and the WDA/FYA one too. +3. **Row types:** Table 3.3's tax-head rows for lines 54, 61 and 46, the three behavioural-erosion rows and paragraph 3.12's welfare saving are `row_type: measure_component`; the bank surcharge's erosion row is `cumulative`. +4. **Sign conventions:** the workbook's indirect-effect rows say they follow PSNB; its total rows say they mix two conventions. +5. **Repeats:** 37 rows dropped (the coordinator's call; see "Not staged"). +6. **Field fixes:** row 2774 uses `decisions_effect_on_borrowing` with `component: gross additional investment income`; row 2828 carries `ab2025__salary_sacrifice_pension_nics_cap_2000` with `recosting_of: autumn_budget_2025`, which the contract accepts. +7. **Files and paths:** REPORT.md is deleted; the manifest notes and these notes name no local machine paths; the PDF's manifest note says its origin and hash are recorded only by this lane. +8. **NOTES.md:** tallies, keys and the not-staged list are regenerated from the rows. +9. **Key names:** renamed under rule 2, as the proposed-key table records. The Table 3.1 savings and property grouping is receipts-only, so it has its own key rather than the lines 61 and 62 package. +10. **Box 3.1:** "a third of the yield" is staged from the review's quotation, at medium confidence, under `ab2026__package_compliance_measures_announced`. + +Row indices above are the 29 October order. 2,797 rows: 2,891 − 58 − 37 + 1. diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/claims_staged.jsonl.gz new file mode 100644 index 0000000000000000000000000000000000000000..7ebbd4b44581634fb257d53bf980d86603b55231 GIT binary patch literal 111394 zcmY(q1ymhf&jtz<=iu({cF^MP4#nNAKqLYmE_>zsGCir{Ohgn5;^LtE77yv-M=eUQK^8(joi|Z7w~zkY>5A5=aVHLIcNcW z&g}BAMew0|p4heQ!#%Uk^XaK;GMY{9Iz&rAt!IlEwf6CZX%X(_gOPz~=kdL5^6!p$ zJ@>2&mS@hd<)<(65eYU&>H!2sKZJWm176*Qh{m-l$@8Y@Pu8d(R_&U+#!g?FmPuXb zYGY75jv61%>u$aL?`d=0k`?B3I~SDKy4Tu<&&pQ))G~zSzn+Nt%FCM2ZC)$?{`q=u zvJmh!*8g(yN5Y4D6Q}tLR}p!^Dj!Dt%-=F8FIS33yt$jP4F>HO=$)&@9NqE!`57p~huJopDrhn2zfHJjPnmX`vjwh@-A*S4ECWKZ!ZZFjyDK8^Jxl-&M z16AeuM$0r>vJ5BV?5IB2?Vl>Vzl-`l9ZXI1ly_!sx|CshEPRpN?DM=gIl$C-J*cH_ zw|HONr{yZ5L~rs#mb~NVr-O@JnSdwi8qNK!>4@$Zv6skC5Sp|e=D2;{;u#nEQpjma zE9BnNmH4?@Fvo-bCpACAl<4y%xpu$UuK4e~ond7ZT*HS^GGD zUGPu1TLWE&ofK5`diff)OU{%Q{rHpoR<>O*ZukZ^^PA4Y;#9-PQNWK^ijlv+zX(F8!A57tc|&)Ho@KX_elmbxt85@$%> zlYeP{sXF#}m{`OX&+||Z;9a~Z?`qp$sd>g-K4H&!&Ln;MDVy?BBtfR<`Ih_Z^!{*d z(cQYkTwKkZu?ZRcgW^7U=}GJU@cySHMYo<3oq#C+abr39ldwDc2blN^V+6>tKsr`JrQAfay^IuBndmPM6EHlqaYVtO=%EpJRpPk4B`E=?opSJz$vd|i4wCJ_s@AinyJ1}7WR!>ci^gXs z?kNADhb`eL;H6^y^xMN6FTt1zbHLBPogEum??ua3`(l{spZzncKUX$m$xU+`r>O;J(xD5^SOJo`rCb#{(WV(G5w-`m4V|T^OXDTmzTZf!h&kL1WqlzBG!Mjyp2b9Ww_e@gn8uL{4qlQ%UyKd)0O zlTROz_4@n7ej)qE*Z7)4lkl+snl$lFtb`vU=e3_NFYoUb_e@j*E)!}G5CW23-y0}4 z=p=M>r<5eI|6T3ySa1JzlpPtyqf~ua{6G_+Kk;)p$-wYg)@u!YzoMfamDNA}vhrl7 zJ$DSQjH`L7yz2N{ir#BSel1@NicLU6$JOTiCAT*<^TQ{dhQ_Mpqg`C*X3wAr7z#x{ zqUVj=JExoF(U`@eYtGI?JrCX)W3=L*H)Z{fYkH2q9+n=3&s7?z`Om5Z8Fl!l$;R39 zM@~co&e0A}oU9T&Y5(OP1e`xDkz4;e@LZT_oqG~h?dYkEhrGV@ykJ*+m;ctT>zjzI zAPGm9!sfgGzuqP5mM$3AAhzITX)o_g2eTCw6LVhQIsCbd-Ff_MSa7T^ z1r%%`)J_;3GmNb=5^MqE&g%^MLii`$h`OCdfFg99zw`RCE}0I{0|&bO!23zIm8(^N z1{A5aiGOH|*X?`&j)Si(imbCdB1b%2)0(w@i zy(Wrj<6ZE@yInC7>%ca7l4`afJ~&^TgL3C`mA{cE(5qlD>?#FuX=EECF>A3n?`U`z z{Hl-&Z!F0h&rjntE^`@{h}uv{Ey0_RiYloH@Mg(T*$t3>2Fnl!#8Si5I3d&)nlD+u z=UzPNR8xmk(ak4r^yLb{DAtr=ObaX%Lvp=T?1|?W74T{-`IsBumk{#}`RH}gLRa__uMZP5YUb1kc?=!pEZ2Iyb>A%?ED)90dTN#sC0VW&4?Zp`_vm z5yV1epx_EHh}xlWs+nLIpu#E<3uaNEaRMdS_o3A8_z(-AaCRvIC4RqCgF;?SKoNrw zkq5%4+4_!y;UG=Bj1M?cf1yRkKdf!U$t=9f77(Z-U0FDV@z3(tHrIuwgWFesA zHic#h3`@3TV`i~+L(#2^3$yFDyM7Mmbt%r@E2H7HKN z^O207xM#5nLlE?k_FsC)jlcaY!p|Vz>M`8x9wFY-ztEN4G%XXk72b;FU&rIvHbi85sc@psn@2?Zi>ez;eU={ zg)=TfhbIS|6KwBu1Ad&(EIgkY&*06RFnfU(POVQdjf7AyE*iZ-^e1ZQ6DgUR%~D z;!H5p4T;TH&;Wj1(BnwG)-R&}Z5nO= zvBTHLzaFl#@j=SShelK|j+fmFqhYBgsV_x!SKBeh;M3v!aaqlid>KT~<9ptN$^Nu| z`m1;K%SRfItAqfW6Z@G1`jb+thwSm+I7Xu>bUzQMy)>!s9ZaUPo((+sbgE?Y1dL|V zk{9R%*q!Lz;TIghi*>G{?56TYn6=V~+-rVdk6 zY9ZdyOWbN5=kn2x=*oM2Fq5|u=22DjK3`Ua9I39RYhQt+W@z+yL6Q3?lfLSj znmQ(V@`HTQ$i8D7Ap=a0Z9+LVnLQ%~K`~U!f53fgZY_A`Y=3Bs=3C)$ztK@}T(`i@c{XK0PWnz!@1lOJw zz0_b~O2kVeg2(V9LAs9ayN{-E7W&1wkvvTp$RBmd!E4csg}wy9KU9R+@gUR;;&3hO zjX+}Oz03_;gnm?3&R}9G87Y}9X&K%^vrH6d;4R^fM?{@T$;0Jt@LN|>p_zeis87^wKl4&Lh9m@Qs z^vfySQ+Q(c>^J&(6U9X@BwliefFv|y#3aNc=~|7^Ic=BZ>5HxwT!NZ*?ImTS$~Ghy)Xqob~CfL zO}G7f)+YOHCwoC-Z}vldhp6(EfK{jYcyW!W-DqNAHtB#%_co+;hdMt)=JD%O>T-#d z`*4*IW*zQ_#5&v-b8U@+mbnuO2di$hH7_D%J!f23RZGvqsi+SoIR*^1ndPvx+qhpc zVD{6rCHt8FUAuJ-b|JSFXvi*cFFOq|!*lmBdk$sXYr`C*Ed};5pLW@2JZx44HgdRG zqIxYJZZd}p0|>ZU-|3Ct*9tRR+iCaMwM?JAyl`EZ%i_Q9ROqA;7;{J=8w02`*m36S z#}J+0UW+UowJ<7hi%C2OTJxmF&o?g)9SGVJ!my5`2rz^T>CJl!?nagvMvp(tqRUq_TUC^EgO z#cZ?l$!lSA$&XwVm;~01R{mh)Azy%w9k~m1yegZCw^d&-V zUV=_jz!kDFbg-|Sq$n#OjxlwH9(+@asL(>yJBQ}8YC4mVGarP)nrjJHs# zmqqjQW&XTQr;O4rD=Q|?y_`y-~piDsw z>Udalu~gN62neg)r9OPr%)#IT6jt;xDeUe*18IT@xPYiKauZG&CXW?;0aLSbt z`bT7?n3x^LEg>@_bb(Ic!^C5C){hcYw6%1iNB2U0>;ruw-HK zhi~!K<(96;av6jrV_f{J{`2o3@k0kMI09rzvvPXEN+F14%b7z)q9c^W!$;P+*dUIXnM`+zSO}*8fyHxmCQV@p%58V%;FX}%;s;8@VS*(mF#K?Drj1SPk@Q+L zhrTtu#Ti$rj4d96g?(vkMc4$vJPv5Mmw@2*DK=`rGa}Wh!#y^NW;1|O4TYwICK{FZj!^JANsNio_@~vxp@@1uUCBcxfvw7 zL?87q$s79;W?o(&R)}tNKRX`$Q26heTOhxZywbLrc=LJx9t&r+=<}x5d!c9XbLGau zrw=U*uC+22&B_()0rY1w-veD@ph6>c*?UHb>iWxTOVWQNzeq_?GO<0a`CzVkUu zu%dp|_Y=FfL6qjpt8#?eZ^Y))3C}#gtYZ7|*<0OH%|Nuq^+2>)v!|~!vwD~10L!TQ z0PC9R4-d?{XRigyyjfhA6q$jE_xQXFH0fak^xGu73$2C>CHEc=x87(1Kv*5mZ52R5*=eK4!akR&_~`REqP-?qxL@ysM0Pr2mB(bvPu^#N04RTnx&qy zU!OuQAzm*j$3+3_`O-aqY#L@+USFgB4MIM&3LD)H-_QxH#s!1mS8(Dp4k_O`*tZ=% z%=Fa#NQQj5x-SKN?_*mD%i^w53S-4H?)#|axG0T2C(X=zTXE%2X@2|tzwR->v1j8`*1v!V zG^CkrLIIip;YJ_Fb*H40%y7IYJfx9XnjHcX$&~#BizYUpMGK+bA$X)A39&d%5ps!L zV%Q-dq9vjwk*o@}a0XQ;%f>^zfmW?lZK%a@x-%%g(BNi-!@D!o!Xf|GfZmAdBK2p- zDq)YCm}D=kEfgwrifjRdicZXenEP{BWCb5&F#H|CULXSaYpEXIo@WfYK=q6Wqn8fMCX;>wwfo(V#F(^a~>(mCSKJwfuMImJjEX2$Ctioi2Cc+}0!S&uG8qR+q3@QUVRZB7=cwEn zoK+MjfP>i-jX(z3849g82rL1imhzb%&ZWSP@1kVY+JV?YB|b8G36JT; zjp4me*7#RjM{Rr=GSYtnh$R^9GGG5PQ$Y58=8m(0D!nrlr zgO0s9JK}ban^|^3&Pr_vMK!7Cc=WGt>BchYO_fIeI{Q8OE zcJztxbKpON_;a@RNF|Ja);XK#yI2QGut6banO?g|W=+2E1F7raK`w%DMtTDH0r2G@ z`M~irdN^aVnS@gS7!;xm1_7buJ;NbIw4e`81F$44*R*sXZgKMC5C|QyLYqyKJTbD` zKQU1~z~z5aU0I z_n|8ggT_K*=zaP>5ml5AU7;AX0wP22+5d_3LwrgBR4@#~wgS$>&faRfehT}~Jd=%V z$?+ML;weQD#!y3JghmxKX(=BR%%jb*ZJ#^Q{R#o$IRCFzRmO&kTB)_SCuHTqAHR0z z`MiJbOCyPsJ6BYfey$@sZ&lW6pO^|DorsYtk2_x0Z_zV*8z&hcWyYJ^G0rqLqf=zuZ39v3`bW8)Lqt3wQDj5UmWY4iWZlL)dC^e@wGY z|2jR}o!J=Aow7>mNIP)M6TNOGefhuD1YtJ5pzVZ-F)21)UeoPJT6Oj72M;RMu;V>n z0hNf=t((I2f6B&1F2^^op0oi$9h2P2{U74~v4O*iv^C_FG#JS0Xke`PG+yS|h z2ergfi+tY9vghvggU$Z2y9!;`ohPafthuAc+><(5{97klBys!5H_T2_J+VKwj3%CS z@>m;A+}Nw!1XiCX0-kyV1AH#I8QmfTT@G@Gj7~?5PWRE}Y`p&zKoi2t;Gmc2=CMqq zae9BN?pm)GL*>#FAuU|mDx_O77)V;D96@VtS$)8{i46{j?FbCGLVw|CA4m+)H-ve0 zxUF3c4NyHY8U3u(Xg4hCCDSKr9p5YZ7vYtl(0o*sRVFxK$%5vu2kZ+R=?B7_TC`V! z7ns*=@+XV*IF8pQPt+G&=YEo^YRVVeOTM28m_mu@H{v36$b>1X?4*?B`(N^WL%${_ zh|B#aKNKJp6Jd=B^G%%byHUUAOpU%g4E%HzM`*ytahgYI;HD=1UJ;;)OmN`aX55FF zJP=E@5}a;o#T$W~-swW2m~up1{OX-sQw@~{wu^_PO~{~ z6*fUCG>Od0hofTIl33C6@W$?b{i`28KV#(8GWL4t@#XqwvG{8X`yUrNVyEN^ih$4W zio5U`XOa~N@EI_)M#z4VGPsCSQJwvQ)cf#`gHpA3Xw^x zE32`oBE2z%vFRB~r$DDbqlVDdVzXSJQLbE|Twpu(*~!@n==91%XIXXFM(4=VS7q4d z3hoU#PxXdO*;-xX|CO7v6{C2zRlcZVe4k|hrPPPB^BjdUs89@*#(=0zP%PsF%F4*n zUpc>-8ee(;SDLX6bK&>IdGexP<;yx%&tfmE2G{T>s%>dMPo!5{MB+pOGBO~#v`RNR znx_HN_#uCj;&r!zRNN5+PP7Fdx{oza_sCLBeI>7@O|Fn- z(YH^p5a<&JFc1^pxnxr8aJwvpAS7BmcW1R}{8{Ix9KmvNp4?him&LvX$HCm1lF8_K zTIAhR!QPtYagNYrMr09WA^_sr^-P4R#;plcGLcE9k2V7lda=7u{1g%VWIL3y6A@x) zql_QjVO=tzU|ht|_9S7XroENnT*OrLm|4)1Zcm-Pi+?XSs`>;Ln?xQWziqQ%n}2E6 z|JCACMX<6Grku7luD+rfr!rd6Pi)WZ`R-GB^~PbGwtgS+L{}8Ct9JY&of84HY&=bW zVYLJwVnoZh-HxphZht*%4wF_|RX>Lh3jLuli2_q-h%-Y434Xk`Iwi?;cVW@)2Js1_ z={tw=GNJG|V-;xPQK09LSm+ z{b?1hgRo2S!Z`q4KN3f_ezoiRswNE3vZxG1?ZW}a3D^&012QP3fpJzVGD{lg9` zo-m)vzip#-6{fpnzt?Al`?RerZ^7s($pmny6`E>Y5DiApff0=o4PO>%rZ?XesUvle z4{v8lzzOHTPy$3NOuQlF)u(*ZhNZICJsdo*{>aLjApam8LC>BzBHydY@3D+EYN7Hb4)?XZydL>F{eyb z7O-q9v<~LLQ37n$z<_PH(VK-k((o*E9|3`6?LN9ilLiv7C{v&{A-H`T^Vjg$fHB|U z^Pz)?`ZC$re*-@mO?LCg5)+vB+u{CWv>V@9Rd=)RaX$%KRRX)ulH;LYlyW4uL%%qq z1*NP{L-gr$ELQpHr!v>Z)PIfEa}Y%o#M2(O7liize1G%rQ=ec*&IsaVBW1n~|*_RwS#pE09TEL*=1fVdz1V^-5X-i;}x7N|yJTw#4_yGqt#Pb%4=?UG79EX!!p zL}lNhh4 zDRKRYH2ny%yOGxFTWw?_Gx8paR-fMj6!Ztjo}+#~}6LJ@J4;Wk}l1C=8+AtA7hkLT|u zYF2NMol%_nSaqx_Ya50v1Z(YK5IY0QlXdu!5gYZsF9s5tWyH*3l@X;L`-rBn}28N4`r6 z-O8mvT2C&=%xhzc`kF;!53Se_CKkt%gOMj;h@!!fw>^hX-WHB(7$*MFO<$kh@B@-Z zZ!q1a@s0o251}NkkP}?*1A^eWv?j*J(VT#b9rd+W<*qJf1hQ>1WE& zol{|^rQ!lQWXe|HV&W@%%*P)ZCC?6Ao}V9hBpK+?zLc-2wWMU)`b;%}tI^I04(Ph|{N0ImIXsm>*ozPV~;Kx&m=#&n@ z*#NlETv_XxL%${(QWfXsF8JAQk0pX=6}+sSOk~oNHHb9Lk&QSFWPEg{E9B5!R21X3 zbIx0*lAo&OrDWG|D(FL24ctyvLoPDgLoa}Qlf~M6NXL9XyQ|H!Bt!vs6Z0#pXgJR^ zMN7YI{ReCLLP6Jvi%j$Tx{D?d`sTIzcf}?PO+50UlN<{Myc{6oR8z9oMz))?CDG5% zcZ6}2`2css%Rl1E&WCWcZODHVEj-H4Di;Q1rf8=*x+ZC-p9NBn3WNftMa~Oz{3*S4 z2oUy6NxDxt2wU|nTJ@J@;QCi-4N?0c9oC^-|Ja&{5{q4d(FAQ&z2hQ$!}|bv-$9a1 zw2XWV(k3In9_ey3+2`U~t&28c#^14&n{Mz)zdpxpVzHh?`lC~ON7liKZhSN@pk$F9 z>R5L$xs0H+DNIHvSQ$9DMiV)ZH3WpJb6 zy9r3SIX|Gc)$}k>`KVb>LS8*regmoW2`yF+a|TWrQtpJQ@=24DH}-kv&~;eh@hiA%58{Fvy0!@C1x9usRuF8~EA0w>3pu=5g{paUa~YZfH6)e1U=v(Ks?6RMnz_4J>}vw=Op^ z$2~W=0gMbs-U$E6riCX`&RP*5;$+a6OA;lgRQR_U7gVeAgN{)QLabA&fdWgpxfl?j zVey;$$RNpjFC9LbIEzmRPb=wSY8eSiGW+$#RFcp>3k=w5_K$*T2sN;&B>_!E)Rbkp z&hVQbLh3g^v9bU26YBu@(U1ZB$Pf}__G&C}PO+)4A_k|*ByVT{XRZ;5XC&hF!etDG% zP6U8-LH7Mq;$xEr!Tt3r+c$9jx%ZK{u^)WcD?oW&l2q)JMIZm*ThBb9zRnvK?7rG z2}xVTYPrI+^5CiT>hsBF=(ID93beFQLoTMpxZnuyUQ^E4575}AFiKlN6}ZTmO%ga@ zJ@|Qf9+N9Dc9LyL9!SPvJWqIMd<=Xgr80`d??qqc&fqe~q_}Tsq=G~~iK)+akg5qf zNDY0W%r#?*0`lK3H_9qMj$)bl?YgO4c75r*rKfGY$(XXV!_|V7Xv}S^Snj)_#g?}Tn?pMkj6Xc>T%#@!v0ApJhWeg~x`>OVDVzaMjv`D%A zueN1o{~U!Z^!ouW=hTheQl)Ueeq$6%zzx%}OoP>Ua2LzW(cZN!XNqvupKOP8(eibY z$=Zc2mW(Gq*P^`po39x$_g;*Ld)$KE1c$+Snef1bP>I=|KI&p+nZ=~67Zq0JdA zCfiIbUKY!3hWtvgO|Nh`9~OxDhcHRtrziB+UQ}j+mfO3zp5aqOAS=dkzfCO5plirW zz_)a4O-oohAoH!3ieaFnfwa6TVHC8B2B?lX| zKPxL%lr9k>wJzsSB&Bd1lDfieVUA5dC%Pw<6N&vRKVe&{+HW{2Jy-W&Q)>JuBuAIW zN3eeUU(~`hA13{-l>LzfpE`5y=c4n=(Y~mX=?Fi`_$+KyXQ~P=mBA>!q=g4Wl?5)F z9NXtkoX=Fbw`M;}Z2t;J^Ce9a0@gWlN4~K1_HkS)C7(+H54}JLmxHpmV=umCpOcOIUvsDMk5GwIcm$Bz*LW`)7jf#wCDdeG-7C3!7*d4W zXbne`%nM}b5wYa669NbCn+R#vWXw@$G4&baBU zDrk-xfFQ#ENi#*1WmEx@cZf{ih+Y0j#YAH;4&Dhu*I9o2L7|HJ|C#rG#HjRJN9|G}Gf065Ry z8$7mn8mlU0nB+gWg_C-&w@*9(Kj;4szW6`*$s7Fa4PI9~U0nV_mnVBnQN;W9!tKty z{-tZsF3PWDO$>x8P~CRutV0Zyipa+<;PuF2-|2YX6k9u2a^|$WSdKtvMuCU z*>z7?85aFZyXJ+LOKqT)J-P-Z(4PFkjExrW2o1 zAk+Yrr;px1<>_6-%WVGpM`m@(-0IKkNUo&et5&+80l5Nnv&^oWg;yagi_g6TT$35V zdi^!b6?xD|c_jL7rjT2#SZSX9PV!_OOt@ttC^<~I!JUC-Uc*q% zI@?k{O_izQGdfq6kjJE2X`aJQJ*wzJ{nls#O- zUu`qO+{5WM5EJsJ;vef;M|sDWfT!Y0P*P9j>tETf$Bbd>XRe9bmU$y%M_E4ybdO9@ z1EJy*U?llQ@7>&z=$MhLtb70T-Me%sW*;M;WE=aiau?gTUudq#w`eHjFKj+|3_BlO zg`K~RZEYy9ja^*=6zQsBS8_j7D9p<1f3x$Pjw9h|>*teegV7-<)}_&D8U)qozOt2i zPodYq=6tLOG5y;tBSB2!=1pbRu5+W-9C2}!VLXKl z{Ped&x$GGINCZ=UukuK2hRDn!y3O2vlsfgEMIcXr$LBo*QgXML8Wc#Eg!Lm`6{SN6>Ipu86z|ZydVLGA)`%BIeGETv5MGUs~416 zNDRZ3Fa+!nWI}T_NyVF5cQe6oE%o$#lh)R}TW)Qd{L%d_%$}@}D4AL_#9l6*M19KJ zH`M-nRjIO{W|RGzb){YLSK0D#PNc;GqH!8o&BiT;LLyD<7f&k9FTRTyV4xw{yB}%a zlnNleUda*{5}{NUev#lQ51$^xT>Or&S`NP$DvHMlwK!UBoL^pT@*>w<9{<5j`erSl z#2F|pk;hpR7o<^t`FONt%!dlQYSkh9p4AE)BCDwv=NP66e2Z(c7o>&mU~8RaHm8sR!2{@a!i|13y2NXO8>-iDMTU#8J_ zT{M!e>ZeFu-B;FHs6?n_tiTc}T+J3n0b;iBCAs2W(NR8J#rHA-&)+XxKm~4A`ux?_veb;J{u1#AGIZr5zf%%5a9n zVmvk!PztUtkSY;R)EkS2(`SZ3acHD)N{qpohT%uFW>J%6)y2!CBdvqN>kX!8j1j~; zW`$#^n+~>4%PDfk<3|0-3P23;23de?oPYi2+rztDwOwSX6_=*|K-qUQ-&}*R;;Vg5fuz`+PnM1XgtBl zswRCIxob?4!9D}Sz@G(@$_#o2l2)TY3ofa?i>uoDNPlcg!z(PwU9(GS-Lp;FHQ+#i z-c@YH?m6;EukH5wa}%jOGHK7^ubFo49}j}ZV2>*_Ni*N8_@&$JT$mrP6YOWO8xMbP z7ERHhBd*z?5VpZDLq^-jR#?5zd8*uSUy3iznAGUr+vwPT# zqUb_1gMz$3Oc#VO^Pw0x$~ekoxXP2!LDfHekOktsnO zbJ1)iiIjMnWZIry+CUDGR1ia^#a=W+(*>Nm^AhZ)V${r6(^AyXK#a*Q4$4sDbTLFk zaXk%d3Ggetgc#!CUM7ab$M+Plp#_^)j4Gs~AGLn3`c@>s|61;jJGoHg;x7{m!0K)CCd|#D6R3Q5Oe9!~3=S-4n$OH#|%@@%P027}fS$Pb1dK z-Mx!&;Ll8Gv{iqWdp_07``REfJ>T~zT|D!%cmn{uXS~ckBv0 zq9mA$+Lbp%RDvJIKLbZ2T9|^uu}Mb=PL*+o zdkMI;2tszGfVl>^=2&-F7d;gWAA}W@JQIXf9CHP~Wty4Pxusj{cLm>!-kdPgjQ%D5 zO?s2mxHUUR*wyD%Si+TqKdwg874)7U#d3i7loWZZm)MmdsH8)53=+RJZSxPt=$unY zkx7ldnl;yOJm#r-!&rbXZSmTkP4yo$zeZJt(>?nX@Zo?weS`UU+(GTG_g(7)c*=yx z1>NFdx$n=}MweelqHx2Pv`uZD=V1%YCtQCDai*uS95h` zOkz}mK);`E2B4qFRw>KKAYLzI5HhGVyl$)!pnqE;8(b$L%6Th6t;d8jeZ$kk(@Lm4 z@7=4JO)biR!v+nWSddt7fVpdH<^IFp)fN7*{J7j8Y=<+4Go`32AB*MbbQ_B!2cW~> zUF%v7*^<-lKKpw)cFW5?>>B~YF}u~@dvjGz^^4oPr;E%j+hop6fKYjH6yM9wH+9n% z!!I*-A;zDI-M>9!d7_Y2^jkjt7rhAbJF)jY?IiC$xTWP@?=lS`pD^t}W0ObQ$Qpe7 zy|i3+-uUlSZ18tffFO;z#m?^2%TPr?xXnutmW*g1mRdhJl1ja35wT#9S7A3ggb9gX zE|S@rCCaNYv=5^VHUrF5q?Ta|4adg#O%bMs!5N8~#u7 zpU*3mn;x65@arOFca7(r5JLaIp6)g2`xC84_y#n5*{Tc8vNw(MwztU$9A@W4S+&%-*~68+H<6#xA`$KD_ZDDg^Rn?Q@>Qu(n$6@2!Nt<@lUl30 z$P=fUsM_nl5z*bA1<{Fkdh`S5`bg^@=AN6+rA5u`4cg;1U#QVvl%A7%{v23$-n21c z+?BPi)+!z4m-($~92kSXLsSthP?wwofq&(S^T~%HkIO~u6Nj?4j=dM9 zJZY5AzcPfB468nt$V(XlzTqG=HR;WOAkC#y-hrBJI=xJuWp(vNg&~Fx)2fi=>mWGo z8&4F&#ACa`BJV7+Pfj|C`9Nc9rQ@u05n_iJ{xLF*`bGj9t2(SvSF?1%sDS|-Lk!EC zX~UbTRTKk?6(()b_ahxVMk?Skl|-5`DXL8CkI;HQsUXcsCQo^Sz}iX|e!?{Cs5$oX z*YY^$X&QdI*&|7z&acuLy{1^Zt}v@QP`a1(#9)xF%*Tp4(3j283WfK}Q>9iQY(|z4 ztGY=p13IK3&eI{RbTe}zh(b;igQj)g^(Yjrqmx#LX~J z(dkAv;j`2}^{H^)$cViO_*~`MGMb!7s5420mha?Yz9JZ3sE9tGhiC-p9^41IFCZIE z_9`J0Zj%sBU;I$J@Z_%A`z@7(xcHCXLcNq&EViT=VkDv27P8!@CX7TYXr7Zi6o&1N zN;6M9)#CU-g8Q&Q42;`OA-qImcS?$Nn9AlSJhfisbP{4N=&?5kJHVtQa#jBavTZSf zq}UQ1jt^@=q9PkVZ#YaeCI)o_1_q4zMOqKc95W}ZrGc1Mk0=~U`)jg=1zka@~Wh(ZwMB?FD2GelVPAk(m1ongK z01gsyCd>yK2WV3T)%D)g``U1aR@?QQXYCR9iK~yK>#X2=DJAuRAV#n>Dn3-{odTND z>aZqPMe)_z44nE1c>%0@U^sw~Uiv9FOLfF!^9N|6@@^(uzB!OF+ z!dI~uzXSoYgR@%JP(h0ryDk5EalOiA>y@Q0Nbg6=XNLN7gTl_!gJQ|aK z2jT#bD?l6o8Ds;VOT3ZJ7c^#)$}Il~`~=fD)!4S&xUdzDe2T}rhLbCe$7@mOrcBzg z7XJONpvBgc+7HH4V*bTN)b)bUV!_8Df^Gw%kEcimsQvcC1ySuqCfr^g)HITPs16h2 zY??>{iY7D-W&^+JNS@MSN?9bhAi~)~f*?5ejoQr7L9Bf@uW2@(3&CTCmxj5|(EzSVZiR{Vt?yw%Mi^XRa2>*^QN-RkgW)_e`>ZC+19dl{ZJsxv z^^joBLHT;u(j1SyP7~gt^!h$Fl6l2VT4itgc{sPUtc`2IW8`l#?jetC79(?DmvK_V z`>Uj`5z1gcffiYlzd1jMR*ub>@yzwg?%udd=<-_4DOUN}8wE2}9ygz(71uaRD42 zzZqU{r2k8~dP&jONtkKcU4=4aGIHj!z9_@M*>IQ)KWV8iPC0fu|A>B(bRrRxrI zx(XW$s)U^Wa6(PwFFkh6sfUcohSNzt*moR%Meo3~VSn(lu_*BtsIb!#3?SL*lsxP^ zce}R?vSIgc8IiqI*p!{eARg`rt?q4-FYO=cXaRTq`*O+LZNY5XubA_pk<9s8f(O5D z3rLOk=Y$S&s7$0#g#UvGl7+Bcv0X7u7xm+hUu(~yeydXJgMikaVG8eH@8HO!`PG|q zj?k5Sj?fp^FS&XXCxOCh80_{%hs-;`5bPaL>R-Rv2$U#K>%W1|W#^A&qVk#ZTa#lL zAr^)2?IOm`v{zF)$0B!g)fFeUuj~?xRi1wsI}g+?6F+arSFfVIvR?bBs7AT48~KfE ziDoNH{jWBQWE;XYb;Yagi-+_@IkKVkST88zb@QV>sLm?iDu;372%|gH0F@tFuflin z8MYTT(1ssb#(|)TbVtyP_ZpC7jTfm=W)J*1dbl$|;USNdU7~7)y#hBPXzV>(ZE1wh zyM?m3cfW!gVU@Q;b?;v&8exMTA+_9&4H|PzW9fv5b!G$HN?3m%f90+nu@S$#u_LO0 za8nNO#Y9#|*%(cwvXx^SM~5vJjkf>XJNJ8+Ca~fj%rhB*YYYr;>KSvA6c8oqa%HBo zTqE-u1$tu>qQyjMz7Tv_3k%OHbKpEe5XB@vL`_k$&szu5JkC%rAFgmS?gKva%G|fyjAU)plpw%;C2TBP z$g$91Fjfpq@0Mv|+2^qz@kKti9#a>^0V;fv1%v}+Mj--JDtZspMEbeM2ke7IibMaQTCU*k9iMK$ zO~W_t{0;7+a^_9O#Pi_DWL=2n!Hak|KM-GC$0Ggw+OaQyIpCL1|&Y>>W#K z5LXbq%M8LR;4I)|^R+!0PCGYPn(j0Id2w1ILK=K#k~oTxFsb{ti;|#EI?1YI>i>gC z{{k%JMK~`{fi3n~u+%|C0?ePoZvQL%a2asd;rqhY=r7xmH~i31?8n5cY;dzT=jXI3 z=QE4UbS3XRv$LWWX^e-&_Ru#rV_N^VB-YOD&L7oJ-ZEvh`;#iuAq?eQoFc(Gd(k_W z1DiefH9bKuv;DAN0=u(>9qbkKB>Evb$=@n$30t*8baWmA4Fa)PCn|i;z8)oTxFNn43RYt!(QZP!@>D$lwm0B ztr_(N-27kNhcP}J#GBxIJnL|t;ePmQ*hoo~8*l*m44rgOovp%Bh(dQ%k3te?G%zfX z!3-xZxg1Ge3PO?#^%@j9G)J!n0afMe2(@h0aJ71Ta6;Z^4nLy^wdl&|%Fx)#?C5Z9 zH-fMXx#uyKg6kAc+)i%_oVat_U@ z)r!(2HNd;pW141tXVPI@z}4 zncbTc!n&Q49pPE-V=XQA~ucF)CmL$ty@6b@*2XUjj#<|`7$_>CuTx9|= ztv2v$EA&5-ii(ZY(uu~Kf;}41o!V`tC<20%36jVrLW+>mw~hFF9uFlLUsD2ARc&Ay z=9v8?+>wlUCVH7F!BU%@K~?3d-;rGm^*B_Uj>f1f*vx=Y0FIl(!Sff6TaHD2UvaD~ z=&ulKIlNk>degJIFPCKt&6M5ou$AeI5Krca2TZNwyK0N@Br!r0Li>EFsLwiZ00k$h zCjnt{GpGV=BZR_)O?my4GoTON(o;_lXf43rwAuTaloUc}5W?2G(0_`OSuR^*t<*hR z%H;mfV6*kD@w5R#VB`CGJj|Ctg5eS7*HbPLXWx-OjY7d5?VtD56kiRWaMpzen9MG; z>z`noD2~yL{mmYe3c7hxgZDY^i?Ik-v&!p$6v5XkzuA6TOJC*u)!lpWgJh01eImxG zS~qqvDskKR#mS??vk&8BcwvaxEbYx9$2K~(x_73n-%*?~po!D4(qX%VaPYDm{oVLT z5a3w)683Rjok+Jk;T*$N@mcOdcX%p>iN0}HmHB>kqcd}9KZ3cs&xdZOy7-N#*-((t zqi6o8*Jy2eFIy;h$%1kJpq}K^rk>;z(k=e^r>RAjYMZ}-ubswh_HlBr$~kR*eMn>i z6xlKJjNK~Iqz4oYX_@WNEA*Yr%d)rJ-y6;jij45`HjjEKZU)(hp$2949lfYOy>)$E zI{Aq}`kkT1AVyXPaDrIIB z;|Lo!sBy-<#Hr^3Og7iwU&@MRq`7{aomkDO05eSU!@-W4WlZ3NRUHt+QbC1@W@#>! zZnL#|j)3cQvL=#2M}=cM=joXg;f$GLRH-_{_E&z1(l0&_`y@&Xvhik;mD}uc67UFT z5IGGz2HBBY1TzSRRNgPT#VvXOJ6+Q2WL|<*A@6mS^Tx^d%YjU%k6(c~Q_AlHiN%gB zr6%)s^;BNZFc-}{!nuh>Kdx?bs|{~YedB1ApQ=B%RDN#ZKjSMv5LXwXJ>n((@Nl0Y zbm;~8%%1Z-+x_&JSkkZowl9%wGr_dnnji$bINsw(F?bPtQg^9|;P zrz|jPbhxT&;3ch*Gn*gDO}`TJ!ATm;PVXt+*(5KdmkZShx3@gfO!~M+E+=5~94{#^ zK_lJq2%qIak#A3Nf&ei~PqCQ#)1%!>AJxiREfqPma?_2Hw_40kR}X6X}|rdW^B+WhN=MSFseguP216VJj&)ke2bqjobyk%iQj-9$cxT%xn1M1Jex zw5VViJwsP=8JWv-X%E}+4{RFDqi&Z7I*uO5FD%yk6ht+@3TWX+Jy$RGn*~(iEKAo2|lT{>k^#L$q^EGe) znv^?6_wZN!gXC_=et@HKjOOx|N|n{QXuV z^!W>oAeyOW?97|%r`pQyLcvm3!d6`>t?>s9RmG93l&+l3X4B{I)b$$ugKp%n=RQ|m z9o@jb@SIy&p$tgh+IV@9giJao%SgxZkOcxEVe?jZ#uL4E$_X_!Y@T`n5+6 zdT=}-=a&}d3nQKJg(bc8&2GG{KlBL@O=6)yvW$Jz!iMR0OQtN0$mNdo9WepJIbF?W zhOg8fFHjxLg(LG+#`{!n>DgJEalU7^M7IGSFCMMt>zdgtK@r3w=r=@d=>WpZ~3sVby;Tu#3JTD4b$#SIE&DK=$c)-d86SU6LjlBe|`RzTkLqL0JKU3m=zlgxEoCos!?0Kh!8k#%I%RzDLNl-U zrtYSy&4OWQ!AgAh%Z0<^n z&$JTbP%W=L8>1AjIu+Yvwl2;YM?R$vE*dtGN`7-z{I=D#Iug`L>F}i>8kt>h)RADZ zxHU3o*j5~_RX=^21q-Q8>X-PCXe*Fz7XK*JjEuBQ<(>JHAbwFTvJ=p(nWYaS<&mNRui7xNtxW|xRO^8qgvm$`MCzPvRW0PqK5?ceqW zV&gU@+mhoZc=lCDN{&;`rB~Qqtw*>Ox8--;_J#Yi;=Hg&pBy-6*GzUnmW}q(mW6mN zbdtC&fQl1E=H)SAdG7E>pV|i69M-Qf$q-dEaIQ*1g#FC@ zhH+bo7xjd-j-*dXwLh^a+$-ND1_84xBhY!|i-Kuw6}a8?ionMJQUwVKx#LbtYx5Ku z4nf<9YV7+&KWyJIS64vPRrDLHXiYLlkZ^$-S1Z_}>WznnfLYk1lF*V+Zf#s5Zkh$Q zexNf13s7!hW`1(`Z&6*v1B@0uLt&9W9FDzd*x5hzT9^s8CcE`n_~}JASCdX?b(zw45fe%pvn1Ad_itc@c2< z%XD^`vu>sK^D&E?Xyu+HXjKcxzwHM?A?4+@%sh@>;r-=}gI{Y8`v+8ahn|lWo^C1L zdWqlr_|Hq+?f>N^KHKb_pc(`G5`pRpiu#++bq|_9*dFpx?(fElrHH+?wJYM>D0=X3 zbN_OcPl4@@yb;id(<67M=D2fTtv{lbEbGUc4e-YF^V>!=HMuGK28YRi0xW`nK&RQQ zfF&Tne#h@WfuEflLGI3~ae&1jMTnzTxnHdz8ZcaJoXDF8z#sdy+&1Q{7y7vY7l*B8 z`cB`UvVK!zwDDO#7x&L6cgm)2x=y(wM!aT3ieOZcy`cxRc7c=;-!0`{q{kD>MGo=N zA7=})6ub1>_&}2)4DTxJigrHlxA_!>18TrP#2A8NoS@?@AaM>&1l~B>oR54f3^w!p zwz10|^3oqxoPlHC4b6_2zOR5U-kAe5;!zaYG7$09`~L)BTH7kHrD(yBm(JO9A?CYa z_I-O;6}VBOJN|?s8zhdPx$k?)HhzXj2{br#5Iik@fkkH=!6g>WLwocRX#?9SnoHN$ zsyS%ldXN;yL{yLW59Rlq^Qh3fp}L_H4S28qWG5t3HX^0QX&i^!zM%YOV-|#7C5X=| zyUI@J8N{V9zTx#yaT!jSp3!mfJC9jTW}M>nI8}8iDHtYhV{d^!#N2qyG#YG0E|)8( zykhkzrdD$SI1!7nhZ#%icVm}@#%dy$u_x7^se4qh-?cUbkjadq6)%_1>Z#h;@4la= z7;#=f0)8ux??Apeoi9P8@*B6Q5du096^V#S3a^(7dySm$P#^|tZR;_y{!A%;n`~ue zdOomSY>|#2)#xrUNBY~FH&&PSSoX?!LN6u8KU`?%QLZ-()>DQ{*5iGm8p!eHtDXYO zo7udQoD--f2)tv{>@>*Ws~0}T&=*beshh#?R5K5Nszla2y+z@S zG0gNr|DmX!ZNSYMQ@nc^_8W7a<#ICpHh{|vN1o<91LgChED-FBwqdB)R9%W#)Y%!q z!hDu)C`viMk{7aIBo%Y-{L0^JYv!Afzm)A{orJ%|8IS={YQ+Pyy#d*)xWtICsJ$+CPrwY6?}4k|iB8YqS=a%~laC zvlnxRv&=928{9PVyT*QS6TUxE^Q{IE#6L9JNV zk~ZB>6ByBr#H~DNLA4!w?^PqpZhBDS<^KS*sD|(Z;)dA;mw;Jn@?gbnsn+q%^ zG}sL!6*GvQd9Q{wbkA)PzZJctwi?oiFJ1Z_B#i}GQ?3@#1ri11V12xrw_Obt)ObVx zsuMzc%9HzYJ-a<>9mW(%R4*3aT{<)Y&zPu!Q(&6tTz)&W8CJh>Yi*6U@$Rnv^>PqB&;;N2oOV*b@zY zXwDV~MEDty?nRh?Pzdq!!y`Jc)(;N#+s;NkmF6qNJ5N-VXD}~^oEKPTCmc|gmq8j0 zCdJMLH0!8(4s2pfm3)TCS;&QZ0jF>gO!S4@x-(TXFi4Zna>$WJXjsPq{@}et;%@^( zkIq(U*4gDemM91&H(>rVN6iO)2gfw49;p7#AMSfA(box|Btm2|yBJ8mrgKP23hHj6 z-`Zpcg)-NJy_b9S1gQ>oo4M!T#=n-G=T}3S-|f|tM;y=%CLthzZ&OSC3<>T%KUaM3 zUrq#mePTh);krNQXwW{8A8X8I9h?4dT22yKR@RwIjVM+!T<*V7JIEe%`yWwCNa*W? zKhOi%R0>v2JSDTU-w=$r?;arAy8W7q>c_4m%_{PWUB+ajjP(lElS3ahm^k`HbB^g; z`^9nW-=9tQvR70UC!$REU_lck8%|1QXYKoA1GH_i3lpS!gOKOvqhHKiBftHax`e$B zxcm^bJp`X}b1%Es=gSwI*`C+b}GH*V(viv zgFycA%a~YEpJ>7dqC72a<~YAI4wA6In+Qq!Ew=RDG|56 zpiN5mHp|)Hn1C*@$#hN#VH71Z|8l%k3pwH-#>vj*c~&_3!6EpCO2%D*4|rJXLc9VB z`jZ|FF(iGpB%WvzjJvI8ld@7&UPA~8MrjmJ_T3bcE(@rGWE0~Pz4>-A{&z4vxzeF` ze(dxw%tj%dA`~!?E9qK=fXPCWE6hI68PdPhVwuPF#0~f-^{Rp=O2!PG&7*;ZgT_kc z1apz#urX$kV!f3QuC^>ZId=PF)@dyjT+MQ{BT~9Tj0IxD+8V~vQU*omy~iRO!>W)J zEV=UsRPspz94$Nk(RT3PPRzi~?M{sMeYyV?NH7)wn6Jib%BHwii^ld3-puTeaTMT* zjyB%JEtcvPCyD!Fx`u0@jrWe~{+A;%a{`JSsCs8xz zsktTrzo_V#75_JWa>EAUHx|N=1#llzj<}0A8rI-tO4aw53)bUg(!JOg$vnd$6rw(H zejbA0U&9~ms(baERC~MvTN@E$Zv4kuOXOX5sW{CK$tK#ScqMaAt;H{%fw{)=o41FO*Ctv>mr^qq z%A7rNdniU2&9Jq(TniCe`m5ck=OU2u)DGXv?^wotb&&EJMy`J9Q*`-Wo~K2AQHV*8UQ4Dhauer4K#C*x` zi=zrbG{6AK$mtm9I(ev$*?VcOP{15I&!zc#d|W$hmYCd-)Qml{hmKH*ov%Ms*`kn7 zb-%?R`sN{Og9ntARW=Xge6pNNA))1n6r~{palrerRTm54s?Uni(SgxTdrc24l}qe? zMMld`uPgW0dth_GaiW=6IX+}*3Pw9lk(0e8O|i(>9oCMXh3b<&(2zBY&i<+bV6e!E z$YQ`D(UzxJUJlODYLfB~VQ^S?XgwsOUY{hKFK(=BZb?0my(`77E zg*$TpKJxUQ0XC5#n?@$%Rc#&EC+#v5sZ#-s;}>zPTYoQ~e5!bR)z!f8bxC{nVb`5| ziSNI^@DQ;-5da<#spYbyvu9;3tqXcV@~^kg*+_X5s6}$N3RAKf`Sq}$=LK_>P7K7Q zMize2-ypEb0iwu}gN90SQ;f_$*4Iz6f3tg2@v+~PI`2P7ppr2#h+x+;04#s&!vvMP z)TBoIlE%ulZUv)h6}vKI*S5pg$~8j1$p zmtzKOyt0IN1vzUs!Czngry%Mx-}2q~@W6bgkidL)j)Uvlvi|ya@E07ii@=Yk4qtdx zY(0i3`wdno`nDbaZ1Y@6YtFb|a0H9|)U#ajkY91+&^8ySOYSgx^^xbKTjz7suYg@W zyuby%$VS$pLxjX{v5Xpi!@q1qAqOliPK20UlRU?jfIpM4FYI=r#=$NM=MJnwJbSBN zrE&hBzd5`(^md0LaW~~-x9`unb!ag6ndR@U&iBwkXk_*xx3?}LzinOMogn$}Pm~mN zNRBlX{F*}oswZ7*c>av>ZoO9_NnWr1Bw)2X5f7|;0_apX(gMy*a4*fW^lQF%1t_~% zY+=!lY9#WVAZ-Z%72}e}8VY`eVF85FHbdeUWG8Ocq?`H$B(q5MSpC_B=v1Iu_#{*| z-Y`VgAj(%z>4C&6$<7n_@RjKm7xl@VvaV?d_QIjyH~PJ0HrGO` zdl3=!WEEN0G}B=sLbf-$g$yHEGp>Yl&@WV#pu%f*ReZc~G#PvLHQC~D4?5WkznP#n zH!^ws?w_VbxFc!!a6g)ijDzwytFoUEUN}mne1(nnIlZ!T-|3qrJ>7PnpO*4ZKYB}@ z+#0Uya@9ENtWIFRt7}utpuf?q)}m2IjTX5L>Osw`--r8i>W410l#HzP@+oY6l*zs# zduhM+$ax}JUTTQ+!jimBjRd%mEagMp@)?9D02Ania0JVUhJJT)lGkOwF$Jt_o6?kc zn^S}861Gls$;R3s61-Imu^Nv05o+I?Wo0AsxPSynI~yV!{^ZU>2hMMyK-Umh84P91 zmOKJ>`jcJSbeQEA|D%3<=Bcs@pRQ)7J(P%D8NrUq zr<|#tg3AQz>C7CBVa<^fXHi9}Q37+Xr7CI%urIPX?No>w6Eq2IoJk?%w zq&8un5xybdi{VbjzgU>8Eb@Xa!pkrmHYm2`QRX5$)L{7waH=SQq1@PT0bNnay+8Hr z)d=NdLa4^!)_AF=;>kjOy%L^1{}Nhk#YmIsqda+iLLVr{Q=B-ZC+q6ded6-L)?bSW zlS>KTAdo-orqtO&awK(8`hNU4m-f}eVdj1IWV55ixwgVnCU#2SJfW1uFp4%EJM+vT z`%hWVw7bq5P0i0vttRiRs$C*Se>?;hc)yKbQWd?Sg6fExx4*3V_+8ia$nIDuI}{rE zohm77(P?^3%=C${-BN{Caj4@#ERP<;;E<-({VgWz%iryAjckW7jU)VGC&H6xnvS&;Ltm~n7u+PKJ@V4=HFCLL zte2s3MRQ_4B}ObrD(JsY52uc=dq1b2bC|~H7WvoawG{eO3&1{sq>2+{nQ~&5GZaXI z8AV*WH%a;6Bw-fi@-;3JBbOYVt>p}hCZA#7aXuO@WGSrc;q(drP|B9H8%WIZ_8`UL zjUMt;h$BT~vHZqZ1Md}P5kZMkezqB|d-mzJb!4ekA+P-G_H1i(l~|@8G0NSxMi0m> z7y+Y+qch>jusT3+bUxDj{{r@Ar#VL_F5Ted*OGpOJ3&tCDiEed$;KMqs?&@=;$S4n#Kt$PY zos{1VDnQEb0%j59s)1s3MIQgycwkvyz8cj23gQmgsw#JiTiZsXy2E;7%Qr};@5J)E&By~Mt zOksvW`{;a z?Q12umHR@Th?Kg|1XM%sc*3BZ3bz>$w>039Q=FX8hZku+{7KnhWoxLsX5!e)zH zl0-aBKTSV6vQRt>76YFa{P~w|G6}e#ir}W1U>_SfV8*8tI{7V);iPiw9wAlbOMLyi z`{*B8{g%8>0wBuojfR2oGSOhSZmhHw>G@%vZjdzFD?V}(IuC4;Zjg$;_a^)Yc{fNN zTRv2Ue7mdjPZMtV|A`v<^-amZ%#w&wKksYAV4&r%eZHl3z2cvJ3#yq^zIiW-yUNlY zq-VIFI3@RG35S5au4ks!?yod!UAa2e+K^mBPWjJBU4qeR@}?U5P0=>{6PhzwoYpblAh7 zr+L2>wzJ%2cfo-$^bjO)w}<=+&WfY48|Ev2OIsg$r*O6i33RK4G?VW5!N~9f3rN-R z%|8vPJB0y@x(ebe#0D84GS1gfL)!WLHhbnr?rhA@f3c6%R=2T_sU!b8_E8gveH`8@ zE^B+!ww?KxT_kI_?2jsqgyaVuD=uI&AZ59-I@iNt%f8fu*X5yCe(CA4kHkx1NRH%b zi2dh1KfT#~jw`vpwKUNvkXh&NVpvFy;rGjt*0GkBXVyJS!iH zYa}O{Z=D=4aH)DTM>l-!6CHrI%4$od#g7%oHkKf`}ZE`>A1e{ajvd?$Zjq zQ&sC+xvF(v^V7~y#JuZV4U)OGBU-8`X#!sIx0gF537NN43#nY&F|BlGfcoIw&L%Hf_CYFr_0rf3aIhm2h~a~&lAz=6HYqS-g)72sC>cseNPs$glMWLN4jELJ z1p~Uap_uv5NaQK@|L9FMC@5^<1r#Lb`4AHG2;xFAKhc<-*lncw`F1@J`LTO5^m&OW z^7jzReZ>MA6wOm{P$b`6QF-k<)uq?SDiW-6bJ28t?y}-oZIXDQ@Ole2V&$@zDSPvj zHWQ{;;d9tT@+#jbOnFjKu>v!fKp2VCf@+7}bBjvR%`LKEhY`{3>u>pn}U#OxMu7WC+E;+XdR;(UfonC|9Tg zChDyU%d|uB>rjPv-yS~o)yXe#fEtf~u+wAK+U<*H2Oj5at}V{MvmBx;vr>J-nk$`r z^pdcQnEKfRjnubFwj0T~4e3tO2K#^3;aa?zi~;)}uY z1d38qLu3?^L1M6)9d@x@ZW3EGG+mmT8%*iT2DU{UepIo=Jfj(gPbg1pXifCyGXKJ2 z%`Sb&#wry{_6ts~OunH--h0M$OHFym#d=nWtzckmVC-XP45Q?gj%;e7YxrZKkYOzL zh_vBJtzvsy>Q2!jgUkCgkvLydt((a`1?~*RAC`heZ#(E*-!IfwO2N*Rje{qEOxC~s zfH3!(JpH9qZe^FM&&(pVZ8bGb*YRRz%TMjnKJF!YpWG9eEOxI;Td~%tW$u7!mHALL zlfEjM!O`*V@uax>0zI-~oHEWelg=c?saw$(UrN^&R*Ip@IZ-Hn$|_Y3(OvjtWJqB0 z5JW5OF+VPmaYX%7=S=CLBcU#?0Ue+T99~-eOn?Lc3~71HLnQ}VzKX-}bqo0=sNn%~ z3aD|ioQB#8r}`Yi*2UC5`*Qw|udMONbCe%C3ZQ9f0_V3lBl7kLqmnQ!ZTj6yl~$*-&iqn%7*<38fFgxt!1Oh8 z*kTv#534@sM(GR_r^sSOCO`P6hxr!rl~IO(Uf%2$G80=>tHQH;rpb(@GF+~LXM2ZE z2M2hFiX)c8<3GuOUOJ$48o$ZeD&NVR1;HYKI)&|Or)k|<aKiK#MWpLBQ{)3)51Jw2(>*>F14td5uzTV^Zk`TR!l-0UT6wiTM;gmC&iA?$28;bZ}pjTE7l(`-2V%H>T@pAu3l zzj3~bxZF#Q5X9*d!}|}@*+3D>QptvMtsG5hn4ug&##^eKtU^;u2JNE^N0X5$AGr_T z$M=JTlA(xt!XA=^vKZU=k@RX|c-gbn0Y_<;?y$AKxO*%p37VsBS<*^pjyMiq182sQ zf<=;SNMsL=92#=O_NyE@0PeIxLq7Rpbnxk7x0Gg%W^lc_b1Pt9#zuMLMuq(OAD?$s zs_hA+j+NdVquxivj17?L6-Nahv$;O5wE}UPlGtnfHuKIMRbF7Wt?FFfR?CrM9_uk$ zvVA8F#rpz8dcA(|^T&sXZ8A?Gms)Z2Rq*){2We?7)g6Y*o$dQnoqmX#r3d!221!rO z`;&Mlr)Z>e;-PYHQ@pLqZB&`n8{y35eV#N?6~(+f216HjU*FSvjowU+TzHYI!pyk` z>&4xJGnckX>o9j_gvh?@iW_s^!+AWC{d!|MQdWHyIZHGfBCF%?o-ONl2KEwq$m@PX zS&a_M??2$I4xg-#Z|XCmYYOwGgKpM)OEK^?**v=E)zVDq(UqP(Frmf#o$hHANX!xP z9i}pf{JyJ2v-q}#c}Q`HP-f`~HO&WJ&XclWX+WRXgY7e_r}9dKoi!o7Gl1 zNnTaXW6g_;X5>2fKkD_48QWX$iRzbrRb6wu0nP zQCOt%sQKWo7FGP)KPK@bXKa<_HeIwf(z1N_9>czdUfrf)-Q#tjFW%9E4NBQ;O|kn3 z9133N*q061MiC7@PL}KeQnr5D60nhy<)O2AC3>Qmt`i}u{%0s&bkD6Po5+uw4s5fV zk!u&v!uRjh8be(72SiIsImdw1bw}zp+tYnAdIq}ghNb*-_)BNE zF~5Y1m>h+i-zClc>vCNXN8RVzGD%|EssKajAAuCQA$GeeCJLB`WC4hqtuVqyS3zpc+b5O#?5?l zgi^0<3oX|#NmzxYexowdG>{4@Ua&pOm5ZpG3Y67?e7<+rXz^X+dj%c8c#GF{R>Ci; z`VODeXITx|B&V$CIJkdTj{ALQTf`aB6P;n*!vQyc6xV?Byt(7MrjP~SV6F89*;g4F z35z%#@gy~9D6*F@R0#$(b2ozl&o4rV=quOLUpCq3I5HU$b=TqY%}9H?rciqf^tnmy zT)bsX0sT_WQ@^uBo$i}at(k*OZm(qw^wb0c%|R~5Hs(w*GD_Gy694{QF^8b_UCJh+ z^rbiR1k4+EqJX}%R=VnC0tfztTdfYkDBH&j86%j^VA$U|$R8o30Qb6DW0W2g$ zZ~&&HZ;HVFM))l@?HgejP}M;k-WS!uyduknQz}z#*i!cXrq0Hqr0X)or~DIqap)?O z!1ZYMSJkYphNnh+Vnygi($)P;$xivDJa4nSm(t|Cp#|J1L5oZ(^zy~&!QBOYZxb5j z`dDm1B`rJqz~|yK!r^4xV!`0^Ie`HhSjZ`Pv~Z4_0(P-!0Z0OIGD^d+8w&!?SO8B} zr?gg%RbcU1`LECUX(<8c(lTiYKhl1;1?c|~^Izn8x%y>gMbCbqSNeU9kmbY@%1yDW zas4+-}8Qd}OT3q;o`r8>X3zlZr@N1JI)_DgNgea%Eii|9^n^Pq$j^(x=Wsp7<2-Q?@9R@@9&+dvQ+L z$9XSnZb?u~E^2PQOHf^n^&NUw%{>2h;Vol>%}<+!w_X9=iMj#Ze=A6y#=RTIpD$V{ zVr)<&wO%MXH@`tm<<3pgFi+Ds`vxgIl6d4P{%#jDS!m@lb>->%&QUMRA&$f`?w$Pc zYvCN>_w0{68AfQH$dUrBQ_Mn<%3_i5B%ec^eVofM&Fg^%wSCHtg(8`rSHg=(fs6J9 zv(&T}hmlFzQ+9?f6w@Eh7Fj*1fIdrc^ww&t$KRkwfewLHe_h*tY|LG~@WtNJzmWFr zY;BC2TG>_3lssjlse9UX(>G@Rj`!5HbYy7cd)5}~&%2vZSsyogvYGYs&W?ZM39d|Q z<}KEFbakN|`rQ72BUyJyQH7D~8^|6=DQI>w^`l2^viYHD$I@h%zw-iwxM!|3UKQS}hA=Y2Huy*??z9rF^FB}BmFq}f=7pa(^P4`R=LVXcM# zL`Vr=A-T|ua2rhjwY-}^x(7v*Bwbj=Dh=W-4_dc@_NdEdDn2Zp}7bJ z;#@T4c*omq)e42%G%;5=dN6H>UN^zlQ?uvFv`lWHGa~%b*lqJS-*yl0TMmwtP0b}N zJnpKY-@N%{9pH8M6$@`h%Ze6D+_i1*WDwTKXRH$|*Dv`jysos2QNLS>F55OYqDINn zY+ke^5af61+}Sd|aOP`a9IZUw*`*A-&hFR*NuOAy4OFIdm|4XNh$SnK$aJXWHT)ra zRuIwI+zn@HLetP0HMY}u9*59yV3hlf@hs*wh{ofUh#W)A@3cQtf-pPwyrr22JF;)= zax}|Uun$0~?Gden_T5mZ{K5+K{bDyXaRMfdrsudLh-YF?hlunNhehCsBk)SY4qPG) zb3*7Jm(d6Qbc}wVzq_0Ze523*Z`q3i#NkjoBe)yUhzQ(r#UC7%c@^Nv;H{l!nLj-4} zK<{A`4|Z{w71nS*zqKvDG|kwMBvzmbShi zy<`W>M}n__0#y<~0htV-;K^M;!Oop`OY&EdogXa$4TrzgPSgE2!-Mk!=CQ-0>I@~> zz_#!?H#7R-jmAuIxpC8Q91<-Ix5GRU&Xap2o{VuIY&;ki9zGokR&Q|!4f>hS+03ml*f!LZQ)+~!1LgOq}ja7jHJJV#BWeik510;Y!j}(EhJA_HF|2Gj;qc)4Dl$(X%AAR4oy~z>^)b6yp6%L< z3|=!~2nPu0pxAKc@Pa;^@U`E)!!pR;pwDU_g+=0W!*IlKld&j3naf?o3EIMQmQgqq`luXqB(ma)daxv%o`OK^^Hyh`^0kMn+J2F1LVh zV2;SnI{}Cn3t@ORXcP#F1~uRcLU$1$sPrH_$r8y5o!2!wkfZ}s5dy*nd}c^^3jw0r z>lKDmcV1ApeiOcEOG*ZGf*d>^u?>37z6SbCuN*~)tRpP6p=I~X*$2G`1D&-?QY~j7 zKyrlM7l(+2hIIlto(?`T4+DKkam&_(0>P%6*m7$D(G<=-LE`B`8hn?VA*3zA!&TQ( zZ5{c416|#eF74CtfT^Pa1Jfr+Rh5C7J7$m|aZ<2AdVaqNySi8mt_mp z*6K+q^Y+4MVS_A22W}Nm5^Rz6Y`^Bb!xh@;QS9JRkO?Ol4zaLimDzQ~mVwb>%czP#t_Zf! zOuM`nDJuNMaFX?43-ae`J!RBX)WaYd1EX0H2(t-W)Df@A(6TYky|-5@m)~BRPkrf< z^>0*FcDJ2&06jgr>nFeL|Dlnu`*wk!r|w@xNeU8_dTtwt)`I@dc@prZXY ziFS*ii6YFmSG14OiEtcEU(TO50DWK54XvF>X4%*v79FDcO5gNQWp4(TNe&6-$3>xu zOQ_l#lpLt)dt?|JNVo@rP{TrBP%wQNl#hxAh!Fv19%!sry8DkWYfI(dP&o6>R;7%X zQGzl0HfuXu(R>E+8#K0*w1dhkEr8$SVQ+k#-WPe7>h_H&1_p9OjGpo!#@xojFWd$5 zNyLGVq|W56!w!mp06vn`tghEwidN%Rxm_rn{*k4|P21C-2WkXuuG55kAZ|^;13q@5 zD1#zDx8hhpgr91e=3L+0*LfQ622 zv@uQy0wJZUw_V)#=kPM!ZoUEGTyH-QF}AJjFv{fmVDmcO-m(Sj^ZAtTzoIE?cU()G znEZ*?n(jR*`o4Kdaq>`+<)yr7kX9qtYQhBbNMSIC@Vg+zKnTzqXUxrFHk*XTQ+`mk zsC{($yiHfYtQboKlPy~^*dsZ_M}_2dI6RSVnEYIVWY}HWDkg9Z%}+}+*Xf?I+mxI=;lC&)su5Hvu5ZiJUc6xiN zs;jDX;(_rb%XQfco88BN#gxb;c0w@wl>$2rrEwy+p)i>T6;rYeghSF@x24*$aySQ|a7aYKIADU7@)`%7S` ze;x;Ux@1Ns0lwJ{-)N)V>?*3An`;!_>PV2?p>bYmIGnW-&YEEbu-YR5tn$V7R?^R| zxn^dYR&naM@7%=me>RDeTikDoAK`~&vhtE7jd0;-7R9>O*+fPxOnf2(5Wva^|k0oaB-lo#Ro!stF2fr?WI>Nb}UJ5 zEY!oTpp$3cS7XhS)m3^5MEjQeBifyW`DKgFeTA*3JhxX9RulHt6Xd0}1G~si^yXI+ z%w0#rUle=J->xO@lee!Y&FE482y*rh5`Ex%aF2L!rKSPT>S;OiFg=#Unky9~Swo5SXy{ zJ0apV3(3Ah0rI$RtN2mrx>kXq`%W<_ZBw=V<7iOPhyrSFxDVL@L5z=@ zBL0^9k3tEOp9SU9IIG(V`kt5zqvQ1fd~=T(o4ik1kqb$It!SK!&`)2SeJtbLU8HH=yMPl z$k4t60XGBF)O{_b*x+W_P`OYdi_li+#wT%U$!AXY4@R$CztVcVmFBxG8edLIzvS{6 zoS2m4=dzcsy8fDTc`W@itxsF&HhW~vM<{ngYx=F*t3RsLZGM&GLA>mr#3(g^#lS(9 zz9DB1e{+fM9wg*vel7XpY%eY^ldZ2#Ns`U}pp4PXoQXY26(;sM?~U)j^cmWH9d)@7 zWqoBFh4159WQj4zShD7@@E~YnlAK2B7+V)((d|3d;=`^@DSFqcbjI(KxPL^Kt8El_ zGc2fqF(}v>btGNXrSw6#bF{@NXhklvoL)Prq zcz4m)WGF{(eUxa!W)l!R@wJ>ibeApO890Rb=1zlhQ$v>Wd;7%v ztUbsbVv85e-Sc}(XLdYcL zAuKo36Sgu?Ff&b85hltToieF@a9DtK(D?~Tt4>{W{|m9(b8pwfobqTdA@%4+tRq^g zZg!r^@)P!oz2{;k@tu=2M>bi=pjMLkRZc2?j*4)v@%r8jEw=Y^VvUv04(FfWGZcpI z`->rOEvzzAy@La48hbO6UN@yASs%6cNX~P(Rxl_JsfisN`ms>?v90J6L;vXAm<#w?sVe9!O4A z#*xfjZoO=_AMjT3e}y6QYz9W~U+Jxa(I1uSSIcHM-73+C@Y^q|L#QmTuC>ZWqx&W7!^y+wB! zWD3?gp=$r}`6RQPl!uJ-ilHBY3_)I=sFaD!3k`+L>`RVDTLHm%Fm$%BVB=^z@CpRm$6QixU_%5&zy6I3A9@Ak#K2(AV*bpd)>Nz_~%E9gZ zwRtnKP&sjfg;Emd@PFyBzQr{8z2f-2*(? zOjR!Lz2EvU)%{-b`jq*X8Q+$fN9f+?QhP-BO*+VAy(#|c>w8PDdjfG+uF-RlK*aO32S$jcK;qwi zT|TUScbamXK?bcw!O4DpCGwC$B5yK9Wbb~w1&7D6q<8lHcqignTioA1!{MAuDVylLY1-G(tts)x_Op?KeVo+BPBud%lAH zkm3|+MMLFcNwfTG!r^^DYw~9E@R~ByZx@vvh zMPcnq)Dr_|Uh5K8(#GC$cZky4(UOn6F(de5D+BmRPqUTF7vuQe)h=EPzrt;u!fDQt zgH|^2OK*AAr4J8L0W_K*&zNnw06A@` zgT}2c!jpK;k$*}H)O^G^d5X?FzV_0GoWar|FWTCIOp!+)E=KN-UkgcUvAgzw#7S-^ zWD1|I`HUCF>gta6US>jCKV16ob)}MeUg+@{^|A8UtqM$&eE8LR<_e1JPT$!p1P6zW zBzL0_PLI^mu!I9He5cgcmqivUJ1Gia2b#><#pvncv9@4vE%j4x$y}tRPT&@kjSZ;o0<~h2pH(gN0PA z03m`lRuo0Z3maJ|FElYmkE~ht8cC=hZ7x1|Wwrx*#VsL`^GAEEImCZRV#4sd8(10D zpWuT9{;BMS2d#g`JCf6#6>5h*X+)cTyx6>0^=ZA7G!296XPCaAGH+To?bB07pM^au6mst*j(EN z9O%V%oWgGkt(t4ofe(6Lt?2==^D+DwU?G^&G-ttgK2G>+@u)1((_{Ok|LikxPzQK6 zZ*0D??ms*8_4)aMp9pzuy+>lq@S#VPdG7W3`#A(k3E&7gXZa=l`n(c!zq(E3dj9dn zXTQfJVPC(b1|eThjor;{t}oNC!r=%0pVRw&w5Bgvuflyk4%~WbWx>C=gzr8ngSvzJ za)X5Y=Z~Vfa*I9@>QNsf%WgeK9+=o)PQ7=M{#dxNsTGH6p0(pT@r9+D1|}?WzxTC1 ze4;9!nCImB!gf3Z(`78hklZpiW^j5bTP*d>N}GEiuDa72a+b z)(mbKc$OgwxOj7_IC$=oGW552^nBWB%}&)X!vjI057 z1NNBSZ-L`?5U2P@HOZADRp65=JfQUjIA|2DS}ZS7#3fZx#Uh;7C4Pe8gQ~TlpzUT&#ETu?FdHU@I4T zz#Y3Mo1=nkcU0Iy>|;iG_J?%!uAf&Iu)2D8k7gT(XQo&6cRO~TSzOGiGk%v21>+3G zTi<4ADtpbU@V^NJYkngUrXt2SOS+r0c|3Ate~9FK9h?)W`zv}xGn*$^G|80p;*9{M zltjeNBOMN@v~B0zjz{=CG8)PB`*TFjT1U+NOdVaY&G`O#vFcsx_8eWrTf!>jXVE*P zyyyqKG7PkUhrLTVFv!>P5QvHKh2?fwYlu}V(g;0wK zAK&2W3v;5MKY1JBR7yr=PgA8 ztUXm5cgeVgyCJw*OSB_9`nXXrxCT*D_oNqVEamGa}FP$CoC3=HlhV{WN24JdGJA z_GYFnU03$2?_v|#*2B-&4=>MI53Dgr`w(O-pX)rGPv}SWn4xhXge(1a|G1P;rvYpPAT@hlQYi<9kGUG#L zP7DfQQPNA58-=oAA*&~~ERqX^s`0T{7lsmNz(R%?(bxtUbB1{<-Eu2Dg>qnzl9j!L zT5+A;>GJmXNomR-*Z*16$s8_hDTKWul4f=HD1_S(3>RI@%@j73Hc5;&z1^^N+b!uh z73N6tBE6=fBAW@AC`rmphUAPfum#IZ{M%C+mzoJk$_$3&^c?`2WvPmsiTr{p`Q~cekY_ zc?CuR&>H}%@E^2M0f64x!=dG+3~K`aLDvKt%AuujXe5PLnc@GSh5%I92@b7StSVoa z+swRgyA$Z{&i!Ly?7Z*7w$e;wIH;D$vGUb$C@!S|KbIB+WIpgvU~Ux}N68&u3u8`i zTnBP0mc0(wXdG8KK?48{rn5Hl6O=Y#8jP+f6U4a~+W>$*1UpZoJ3L$?B^D2lUz8qX z0098{?y~8%;4xX|aRmr^u32iVJ|4d;GqkH@q9R@tXn^&Ao?+piWS!j$m#j6_t8YP` zKVt)ah_HS0VBu)kVCaCDg{d!k47fSi{vhIO;EpD4*g)V6WqK~!3`0=&@&Ig!a}%fu zOy(1zOk>7kY7?-*uW7vm@qZ-6ja$FdB0O|2ybuIM}+ zP={H+<9q|eQBHH%qh_Kq590s;(C1(qYgU@!3A&4cx**lhy84M~A;w7R6&{Gzs?qTF z2qzVznC6C@nR!0K_FY-vYuY#4*@@0#e*WGC&VPzNA~4-*N}!h@fG^D4Lee_a5ez=^ zHoo#IB*_Sa+IZ<4%Vf|ughO@W+PSgRYRB=^M5QlkHEl@mVzpg*K*ei0$Cy?0y+isr zap&AvdbKF{dZNMx$tXN-dN-#UOGEnFyu>EdRfGCAe!4kosBd#7152`X4X7a?i}7nJ-3` zV0(o&+AT${&sD0+&+e0%n~M!K&nIZvbhP|)BHi%@J(=X@rzxKhbygn2fEq$m~jL>t#(h_V} zGN)^1Q3r%+ zXZQ|Wsb>2C4d`x-^j8}|0(pap#X^cqGu8^4csVchC^-lRtY)Dc1$&(im{uF#yWsit(zkGS6JXNIGIA^M z$0#z2HOMhGD$f4|8KH8g1%1+qj`{k7N_W=J6kmZ4Pd9tt=JE`-l6gjreAL;gGe-juSJ* zm2`Q!WtL{=tl0rWVT!`&XS$r)3bFzTnQ{U#QC+%H@k+{aF{AhS;)VSxAkVXOZ_aO3 zM?R``Z%F^RJubL%)muX3yjUt>m7=tjAnWsgH9XBkt_kY zz)eq96O1mR5nZJLLOLW2*OA{+iZ+jero`8!D`z^Wxmj)}3glzR`!6NuP9YQD`@YAkI<| zuw{a=mbWROti|T#{;X!6aj-AY2gS2bQJ zOiX>#N-O3$XgYlICj2-9ZFbnHOqz?~|FH<^uprscJ@6it_7^7hu|Bj1nm1V9v{;l2-({_=(9m>U4q z^bBq%pY;L=hgppJ^LqyAo4r)Nu*Wnzzsx?&E(R0SB9+pHZ`v4W@fL)faZ)Jjomhqn zmVq$}U?uQDAeba6A!a0nke{9eTU$vCBCnKWs~~)km~gr!wp+;H$R7}qTr-ik&j9Ei z3;~N7%{VkvpOs;DcNjFO`H|ll3XA@cF9Gtj^0hrco-2?1UJlKVNW15}_fi z2QUiFBt^+yz5mE-9nAsl?+Wohv~}To^G{#)_?%AK#f!^ukE3=8)I70gCB>CEAy7hu ze@e!2yXM^%nO|+i8{=hd5_`bS2`vB233x>*;a7G0UMfY8U1nNimv2Q2V=gSnnL^P4 zpJVUAAzBg%K=hOdiDt&lL&6s)72ie#F3isf-Ws$GhilpPBm3IV=P;i)dwxcu1};`G z@f{d_S>G^j4~SBGs1XI9r3tEN_cS{NQoOB5X8*c5AtB`{dL0>dNi2 zEpg7@rsd1d>NcLsmHnJgpK>*ABGa;C{T_pvMmVhO(uODQ(t9ZmrXnQHr#Sr1>>Nk# z?%b9aM$Dx-2^=ny=Vm@~{vOGnTL(6A#2*^-Cwsv~v*C8phJeWsdK|6wec7hvEfbI7 z*TEYNM(xlqJKz7@^&rfrM~KnY7?SXvppy*HA0VPm(=YY+(;K&Q#M8gFwMeCkC%Zrf z28neDl3nUMe~VNn659q}ORzh7M#pnQnws1poYb64_*PHpAxJD~jybexPV{fk{n`}( z5fg~TP-OJXp(6?pq9szWBGXw>n6D&J@X)FVVlkv2+3=dMY-51-A*bftz-zb)kz+|a zo}*3kq<>Q$NXmBLp^2AImD)}HG2TmxGO$Y>NPWUnhpR9zTxJ#ypScCw%uE|2iT9tQ zrHUx#dvHDKwL94o?o)7vDP}k!wy9)0HILtP9vVh_a}CeUd($++dt0*t^rg%SyFjJe zDqN+Jkzvbg%LEME5xST3iHI23Lb%yr&5HbA*7$r*3E@`l9N?d z&IzUHEwMM0DXwheQnR>Q-p6Wz{JHCu`J=gzP896xM`nM9v zWQ*u8(|`(B1^7@0pmx3Q&9FIkJA|a@UogDDmVQx}sW|M^3~apB1Uvw$rUClp3yP!X zF-Lg@ooGqpUxAkky(u=8vY08d@^*|pE`+c?l)()pDUYN+)2$+05I8+t>L=`=5J6Jj zu5swd8`Z5fghP#SA*t{FgJ!)0pn7m9>r&ak9N@a<$lR2rU0#=egrA`AXZlXC1|_ou z`mC;_gQ7C)cA|)XCf?bSbTX8uond%o8TN~EY z#N39|jAQ$8IS9J}T0~|ZMYn;e>g{V|Opdu8OQvpNfYuDv>-nI@k$DNf82{ExL=FLJ zGOn#9H|xms!i@$R5HtgatHLswd3>%opAOX!ojz2A0&#DcEP`)5@}CoMV0A*6N*S?4 zG8CSo(sYQ_9AUH{-Y^~g!Ud>n!Wjy;|4|PV;M8L{^&LRH_#f3CPThi2-vZQ=|4|b@ z2r^}=31>v4v2N-RWm2lisj}AQ4sCtSte}Kb?KkLk=Ksms{{Lj{M3w(X)}{w!?L;_} zz-vo32U|oCYeQo!aiS?H&683 z)vjME;P()bcDN~L1g-UfHvrt2WHvwmeii_2uG;+{l?#aplhW>)Ul$Wku()c_r<^9l zKIiNC|KrcV#M0$TAp(miu-_icNawVmuRdOCDT_L>WW`a$jV^^y#L3~wTe4zx`QlWX zisJ0A=!iC$`9@7PeE$16brHm>pq0)8@+NSkM(O7q~X6=0~uPKMTQzCoYI(U3&c zL-Do%Hey*B31=%@3uTj9W9Zl`myC|0?;Fw~i+kjzmd})vz=t_Baa^ zC!^aO-AH0-J`b8O2Dyu=e-9GDewF&LQ!iGah!E4}MuKQj@Ohq(7AS8I5vjv_HHWqf zuu+m#w|TL^%L{2gzZ)>Q4hPug@JEm8q3Uod7v1M~LnZ=Ia4Om85$7f)oXeH$U7O+3 z^^UiY&!`s3^Pt-#!EI-EbPNn&`;}^p8rhK-Exr0SzQ}io(&OjbV-(Vy45GmKcJ#BL zYwMlp#)a+Z6GC?A_#8%ech+M52jXxHJj)TLk{fOtV|hX z30aX83v?M1Bj_t}FycC^RGO4Az_f~QCB@^$|J5{r_pc^$#a3h1GIes+-8_1~WJ6|| zI9L{V!vD|W1dp!$*Z6=}%-YH*@={Zz#`fvPY=+eVn8RVNMZjc(eDr z2ov(uQ*8p|nvIQpJU^Et!giyoKY*^5cIT$vwLXqksHnzg`#-Q;mGnqI%{(rT`TMm8 zkAIw&%FncaE1YJuvZrK6lK$nx<>L6K0MRGYi!TmE0bL(+0~?=9yj?fDik9f#*(+Xa z_V?i{kbLC-HiyH|d29sz14y{zX{42v?vj^BRpS#41#3UFA`6gA2_1s5v?f~GcQZtP zxy4;P2Q(Z$2M#T*T6;__(d4Y>eqTAYM+*2S2HKr+r{|RQapvXpW8!k3yT_?@7{yF> z)U;|1ZI#4{jHg*U?6Y~E?trgax4{KgUWNq)&?mA2kD1!?c9XogPB!t@p4rd3ag{l- zk(JHhy_Vhf47}}TeJWCwq--e8*rlUL7HNdl2604~j>f0BXdU*Aa2nfj$(|P zhnefP)sNcJ&O7O!2=}SI!n9^1JA@br#1139m(^VD`Z2t- zzE{#mm($&sWyMg;Bqc2MN<}XfL5-tL)+;G0)<%19k81Sk^!DzI63;IgK9>j5YjdX4`yy#?Ptf_MjSPYc@B2Z%atELv z_VX9h5_{`*Y22;@3Jnx=3+Ca|>vNI40b>)ct+pEJMz?K0$0kmeMlA1-zdRqAlAa3A z>+80CNE>4aeh8jC2?`QUZaM{-oHoTTziH_X>dJii6M81m^@(C8dw0b3*LmT@^`Fm= zyXhVNhpr1#Lq7P@xzW-B(v~8@W<$H~xNVPTNF%&jbwN~bZW!n9!;DiQ`#MMG`dF64 z75szB2~&g0fZ(i{BjoZc^{n7L)taWVPzyN%?7qQ>K6|VH@$Aok%@kz?ygY^#>Sr4& zhLYid{6WJ<3qV@}vbL%a>skhi=z_@1zUH?`8s$k$BfZ!MXY4JRLPo1w#j*$^3V zjtNB9&P+1DNZ{>z)hPD?S}Qc9_PJp(@tYc-vYiJ4SHq%9Z)T1syj&hAPtyWk+=g$_ zI26r#QC%v*kqSf_%BA~7!avIS zNK8*+(vs*-+fr}Kn|ud-l+^ryi)P+Kx6tUQpy`S$n2dh-QB-EEHGrh)EcQ__QCcg$ za_8hwi+cV0dXK}!v2<5&Ur{dc=xKHJWu3-5k~kNl`@Nl4c=6J4rCon5-T%sW{oOfS zhzzj(bG6?$Gt@QAgj4_So4BOpt{#^QvEnQOa+jb^ot(_)ug}>w3-$Z=^<$Z^#~ zTFR!@C>p8MjC4GQzFZ>m6wO$xWwPa_|48s$mg8FPjiX|+&|-s5Q>4Cqq%aImE+(8Mz2j}8Lpp#HKBbDHQ{z5p3sidR|!%H0(zE41N2NK z2WFy>6+FgVY3O#;{pe7$mC37O zz=##$!Qz{dgW6EGI5z5dm^pCj-&ittg zU~WZA()rq-`djq!?D2ZK{_f|?!jC4_6Guv(rR)_NdF6*2LHwD@XGA-VPDcq|)l^oHbW<&cJwNo8 zfW~Y2>Zt225xEbBKmVnN?T7Cu%LS99u2syI`ZtU^uVuOXYa_8RI!Z{`X_FVBSYbNq z^1vt}h=Ai;2mt)B9Dw(c{|`@zWys|^?dbOOH12$TIc3gFir=(<@u8=si^DI`b#WRl zLc?lro3lM&EV`7XkPLYj{T>e+E~0%2+ts?76tTl-tXM3}+tUN;cT5}M{xw-hZpn?g zZusd@Zun=bbCO=#o#?l-lBnC%oANutO@mJF%IkQv&oefBecSpa)zTKuiZa~;5ujAer5@6lX>+`p3Ep(K63s;oyX-I?qdI*zD4s!&Pdl`-#$tWd!~e3jZ+}pjnn?EY1(& zJlSX02KHz(vqq6stf%1H20r`apJIDFycY0DApAUK-wTMgr8JTGwkdXoNl)z^i+;@?2 znPv3o$4H1Cw6p{?D4Yz`vJ5J0pDhgAiL(Wb|zW zh`mg0D=b7ycUwkJl)ya>3Eg{I1t5nYgLdi#Y& zWKd!d2nA@feZjJ|(K0BS#In<}DD*3$ul`pcY8)pcj2hG5fdMTG8;Ymx0LdO}v)?}? zLcILuih_*W2LE#kVGic`tt38Y>HaD~VO5*pNl^Cjm51#Mape*99oX;Rzj&T?zVpp| zxi#g~|9H;B(6wF`^x$;!wS1&uCS0Sf_@L>0|D?wli<&0)Q+cux$RueRTj87L)^CRv zokwdgPo9%Ue96o6oTvjnT$-QGId(pzkt$L_G`Lq^H&L-H>(!pRkN)OylC;4_Cg+U_ zwXWxqtbf?}q52@15MuHj_)dC=OHS}bYfoE*0@ij>Zubk-@66{De8q1!gPz@i-9OQf z&BkdTPq!l+Fa>DKXq>>?TP)16~Q6fvo3_;l? zUULy81=x;3YBrByaos7C{#jvZFRKuMD~bMcZZG@ zXs+g5L1(J6rEHLrwq_`qr)(qtChBo>9H2-*QW1 zNGcpZYX*7K_ryZ{hmmz@>z*h)m zo$OU1yn#=%r!4=bS%$^`dzzil0jAkp^l}TZ?Tguj=O$kipmpjtih0?)tn(7{z2ox| z+9-+z(ZBANIwX^2=kbOn-8b?ktnpv*ZK;e+ z^h1E{iFFZ`>Nq!%%Q!O>@nc%?>NwS8Ga_4AQOgYsYsnP=tQ`RyyBGn-g5X$UIM#@) zs})uqB*70(eL2TD;bZNo-0C5rYX&U!9TNnW@?0yv>1=FD#D1k1x}szSjseHCki_he zCGZo5Tk=j0E0Jj!#gR?73?z+%<&{@bHy33H-CNY>qcDApDUhlB}!;d~w{ylA82bd;ejPzrc<1s6s*RzDrU>T1HVC2(v6 zKX*dI{PEBa=cLra{`SW+D$v;KE}jLEMwqHNZ>=o3Rm+-wdn8rD?j1Y6iIzFp?>E$~ z8)s_x9@Bfaaki)!dVrxu6s-5Vg;(*r5AO({-rG&?C-ZHMRBQVYu;)Jz4f_8;XaFeM zfp_Glhpj;?UV!Uv%Fg?__s{B_a_T01&OJUrFiM#Wtsr{4-H{R3_&z1hv-bL}m+63d z<#Jm_9OTBn(#q73ju?R9F=Yx6)n=qISL{A#zr-ODjO5OYzOpaoCiYz1%Z#)2jsOHG zsAz_Mt8qn{KTXB189Vui_8oMB^Ltfp`I%=?ligVVzP1kX zfq+TNN6zU?d(?|XD~bl~`G!NM2r-(VYo|{Wg=q>B>Gc4+8d+Hwttib~k}7$)RLE8n zD#eMMq{-XAVibkp3K9*q!x2e_D@yE4wqz46sbUSXpaig|4s-PEq{=~x@RG`O=b#E% z?1iBvrRtb7*+fTbt5U#M63q3hI)*NoHGjFTCn zs>xt_z6(=`)#{YN#3aQu(#Iry_HGta%u>Q2MQrW~ELK`vRvgz*Yz5?OLS)sb#3Eu9 zdx69{tUj#XGNP_gfdq)Tmilx^mHK#y;7Pj4(v?X%G+jh+$RN%;+R(B#st~#N^6x>W z@7vOZij&S7VU@)J6y}>b4G-&Cvt5_(r>ov23vYgYr28utW+RqqaGiOY`aw_B z?&p58<0Iew{r*JVZC{P9_p%Yq*uj^Vf5x)wa{WT=N?C;-?(07F@b|x_)c8YYt`!yU zPf2nsG#yWh2C>Z2|1o{^UcajMeU5>EEb_JI9rO4NmeLI9Sc5gWez2FdKFlEJr+&48 z7kh*O#FXia8iEz4iROuI%PoUR*}z#7D-3;cdS zCEP<1gPF(NH69SAr>nJQ%8v?vJ5HV@ykNReymxX|1g2W_SA zfY@mR!nA8bj;~@TgU@3!$T2XgESvhsU?i!1L zF74m#;vM(HXUIjN%Ednmz9_vsOPhMQ=dcdfzD{I?y9<*ozN`~s^%_)tbX+K7|v(87QaI&p8dIa2-^I) z4O*6(nDBADirt(ZD2Egi0F(f^UcF*%uikKNvxf@xUcG$nZf~ihUC=HSaPKhyR+VjQ zfCcwfb@~ENiVirvi|}QkEk5)`FYf%Mn(SLGRJ>yT2pcbtFYbj$xjgj$i*kwhf1+Hh zCMPA+E@+7>dW>gwz2ela|XkqhkiO(YNg*V6xhU$W!Z3z)P^ANz!cN}QyQ%PsQs}m zH`3ZFqGHMKd&Sc=9=PyCOA3>_^|bFApSPyppM29$hTv#ThDD1LZW&}x_747C6BfXm za2C#+U_In@7=FyJF?O8BLB)85GNFS-~s7D>ibf}KL7STzvDWWAm zre_SdsfiCrNjOVL(HLR^xxb3ADKh5=tEiw&=+NHN=0f8mZ1T8rb4(c-uO8D%p>q5x zQEC(D6NfJ_Cj`v+32)VQ)Ro{4Ic=L$P4-{s0UeFOZ$esE5O_=HVeZgE9IN^x+uRfD zSo6i~qg!KJF~93b>_7_SFTjbOUQZF8X2}K2!S7sQ?dE^I77-;PbVJb336vC3sHs#D zi>ks|cGa-XAw&|_dH{V^`yP#Y6Z#)L2cQ#3*cbxz6q9?jBB%$dF9lvvPCiG!r^>mG zH6nGteBwT(uU9UrZysJz%tHNaf7@g>u*Q8K)dy0``u5RUOwe=#5irBI=vB!2>9z3c z3#g0Yxt)Xv%|Oj@>`mE9sQ8dLd*}%ekYr>6etaQ96gxyjOd_0hA-rC!40$~jhq!4t zjMqHv;)a0Nk9fhvr;dL385Lc`TW0M9)V+v`=y%z`Xn|+g3vR~I1*`@kNEajY>R5+J zC|Dv@GFcx~Wsb{2B<>BtyyzUyu~wKVbO0+t=pBEYGFHYK2J3bn(nmyh1sCuk?kOuR zI>!qv<5BGYAV}d%s9u5ysAk1nD~{>7i$0%9_Q}H=-b+`)PAEe|A&+w(1R4LI1@LFU z0{A`W!yKJ?`Ox3|kLgtMFND6)BL9{4aTbMm#U3#6j^11a&B9^L+2NKTx=QW!7A zbCm7>7;^u~$_Z4ZuJlXcST#$XHtDju*o(TN5$2>GJ3sp!B$h!F&qcNk?K6P|^N>^l z`Vqg8uEaI(^iFmAV)y%0`3SMWO2b9K9ey!{qN?LAmcraEghFRwyc>COr4&dvRD!n1 zn>wPscnCpV6j9RLqK=r(i!mh@@zkjm(y?WYGP4gZ>`CH6$EVyC|GUw_rJ79OL zmFnLawsNec)#7X<7F>h^J=Rgb2W~_?lw#Say*%sk%F4$(yE|M+KyMapZC1nYzX-8q z8Wmj)zqg+fm!frNqn=}}$xu47B|DTmvDxpvRpV>O_yS5O7V9Tln^l#iw3lZA5=E&M zg4^9S@})8}vfEqo$2X#dz)+w{9TlKTa}`<2v36Bc6zGdJW*3tk=9@!}-K*8ZDRU1$ z>EKo8*~CwTe&B9d;X~$b3CFH!1DGzdmprv1HL(J<0f*ywxeRU%UGgPJP%Ihl4P*%| ziQ>fq9?D>ijBfc5s}A5}M6MJ{kp@3gDYKGE)G7lw9BhbC9b-!2yisuof|XqdGV3^m8d;*F z$H2`B3z4gMp@V^y`W97|&LBvZ?iOr?msJgJrHshDDV+teD$t@umJoy*6M&!p7%55d z$HK)JeXN-6Pm)8%XJdr>?_9wBcN#uY=8LB>?Zk4f{eD8RCBUSYtyl?fO^mltG|L)f z+U4JEZQgBl%TVxwFL2M9xT(z}QmA>uQ#Q=xW zd+QF9YpOm08~=MTn+d$PKoF z5Q$`vR8Em%IZ0FwWsp@mqZGCn=FUMxU7OdAo{S$=pMMV)JaDy5B>Oo=y2o)364km+ z^d=hH{Y)m>P9xEZ-77M@h*#%0>7f$W37<{Q6HcZa+J}H85urOUmW}!09U?QLo980I&F(n)FQR28 zA?$a|@9~G<<4m-i@$?9Yvc1_fyk>O^vrUBkw^y(3cg z$eQRl_o#3}>3AbL=+Iio<&9&E7QoFM=g~tX1sA)>OQu*ZuzSfg^jq$%XItXou(Z$PL8@e+_y~ zjA3yMZC4XRnmy7?psS~m+i_&EKFXb|Kfrz!#&NAHl!m_ zW_h7P{YR^6DBk~S%)=*Cp6sPwpX*v5Gl#nVrCh^XVq#UA>rO@OrG3F;^Y4c$ST_;| z;#~E5t$3c!@Ey;~690ak8^HeRPStyTRnh_Syqf>;P%sWr+qjD269!D>yywikw}&d| zw<6x!)x_`p>Ce1iv&LOx{}5~s z*$@ix+jf7PucgS!JcpG#4Bl(`%BMI8sP16*j4^lc&>I<)7>ns7qnOzx!k7}VGBK=5 zF{^nzK-8MvxWEDfYcXJ$1Wck7CXuiSsIy^+5-&pKeE37cw4$__Gqj?}B|?yNcoWB1 zbTW{@GO{6ZAw+5+gEG`Ih^%BYd&I0>tB9InKuV%&nW6@x+Ef>k|!+$F0M8h!#R3T@*nXG~r5Q zvwD3VV&2Ga=fS`47D^U77IiCmTccAY5io#U_p^GrNt=t*2rvp6WfN-hPwgOH2a!cTb4&7ULxHgv&A(F*{PgFb*wcC8|u?44~m1R<`!i2oM;} zpg}+hLD0EVH5hw6i;sXL<7I>p<1YQLnR8Hc(&jK6a5&s)BkdUsShFmzzu;}G@cw3* zNkMI;?U5tFL6##K5A&1pAD6q;7!T9)J<1)FjLgQAWkt5-WgHi_J=DRlEy0na%NSUp z%?Rbsfn1wS&Z6Q?ZtD?YL8O>0g9ut>ND>Ivj~!&Zsoco%W3tH*v#iA;IadoD0mhzPwb$qJdBCbv>)sBCBryA6H)iR7VqZi@UqK z2X}W59)bl3?hd&GcXxLuNCE^2?(R;|;1><<&Rg>R_1AlkuAA(gJ~OigTRYP|-KYQN zsMR)U(;**QCV7w3z4DOxJm)@vyu1VqNNNtYvupY(g|oEkK?Z%)|3s_$ zT&y@o`9619K_(VctvnP7+S4CN&3I;Fv!IZgC|{#jPKJz~b_*+1)%&I&5&Dgx$oOJ> z)46bk*l9*7JATSPNtxd`9hD!FUg?{Fu!`b68oxD-N*Pdp!Xlg93NG6pn{I%FjJ^c| zihjHgkrKW4vwUy?Y^9@J^H>QZtALQmN0g7z@E=j?bQkZI?g(qQ?6wyo5Q)Vi5F-?h zBM|E}7O$7C2|c&$78cs!T1DI89FMqu2Iu4gA{xqMGP{8Z^4x+4dG3Z)&(;Al7*AkF z1kpU^}@mzq}5o^)=bYjBaxn0n_UGV4JNl&}xLbmsT!#&B4q zXFj|c^N9@0z%tV--&z0_srnulCbxKq2k{#&=?%9C58$XMEL`IsdoT!Q_U}no$1jB*7y_Wepa!?0!kWqQ zA-UNfi?q1+pJ2Jq=ArK2o37~@p&c`|Q%uJxR}+>DP8 z&p+z;Yyu1mD=jt!BLj(Ycv?aeVcug^xXQnV&+GD8oYH`S>pr)ZbmZ3OuVOjEFE$Ze< zVM&+rVb_!>YK>Wuh5g~r@-w7#J&D6=yP&^{xA%zSeT6{ZkdP{^x#CqZ1p?XqnS&LW z-}cNwp|QpX_9VSf=o^sju+%ymkU($LoX(jQem__sdR-EZ&2?6^QbC2J%>Jt!1XC2NWD~k|*RKQ;TtaV<~cK=qSbh z`p9AO!l_0`ZK_7dBxQb{#`)=E$ta=bu8w(*Z|De-`rc4NgGR>vsJMA_mxOs4N=iRD zgZ;-0v{bbW+Z5E6vI zKGsaVG$f>s{5OsA(0N07Me<$&=N{8`Tdf6+ze_R8#{m!XXno1IId}>J|9s~(kSXL` zVETJ*sLM6-USI~q^w%6EX$uP#IA`K$AE^xGRsMOHPp&7YH)zoii!vxFwh#X-y-whe zIS}G&g=CFHyo|&7i2R&SeueEb+GxE<@KCT?v}$YBI%vgp734dmdNuy1MAv3NOg2`w zsvw3WX=MuiK|*Z0(K<`EzkpniTkltBgB41N)PDo}+h+!c`C4^Ybr3CkS0>+CE(9X! zChdGn_a<78rr%|=&D4+U^j%NfecfnHAJZ->g|p4o6VdBriPgH0MVoiN@O1vUi;0ut zRs(Xo$Q$b&+36&Mw-@cmeppfCGJL59f;0!Fp3G4$f#|j_cBG%a1gM%Mo#s40_WM2G z!ZTZ}VY74nP>YtBu4a@g(xTiH9&T8;4G24Icw4ZMu;2SL{E?RD?_`aiCmByWT8 zTiJI|0sX-D5F`9mn}2!_xwLTC^3u$2u(|{RaKOUOH7H3Dg;qiFetQVZc^gJIH-mW$ zK8QI!BtXAD@tdZML~2O_?JhfFtuUW1LGIgDVD`HWZInvJe?63OkaW|DVc8+zf--Z- zp|aq3u9x-@9T6M}Qu;zC0U=kNsI}(BBo?hlsz;g)SUg(-gxW@&H)h+om;QVs$9s39 zda@3XFme?;BH><9Rs3$z$v*rW4?ynEY=e}kwHEgx5Eu6{?0ZBf0U>{3e8>uosGf9% z6Mgi~ncf&*_%VDOLx@A|w^Q5od7N)txqc%wi>t9n^xc)7t=3l`R}Zn2PF4ZgcWgJ? z;M3a+(=klv&+_)7CoYmi-$3`=|ACTPN?jEDK~5tgr59!EIg!n)04DK6 zv-YV<41Hb){RLEOD*eez+`(^i+%lC@`rSPmddJD=a$cWFmG64Zj*+59uoVmsfx*8= z1hp2w+ay#BJ!rVe%*)N=zn|Y@DMfe!f`X~eqGCU^Tm*fnTCX_#M#|JFzFv`NYI$ps zBCxoBIM`BnE2X_3xyB}FaIlYVaBhJ+Ozx=Fyo)nT4*WuB{HC21-Ti>u#eD8&9L%P= z5BlH~31$;iNof(|hb;|%*+{lkNqFjJLV)2S!8#;^atAiAqe34OCqSS^hcPCjhz`aG4q(UX zfum583!xah$F0`L>``O$Pth}cWr8)+4D&~chpd&=4%4ETC8eTpMXim-;82S*2a$|U z5lmANNts$zOivf447d%159r%6mzttznl|SQxNyr}h7U{;5ZSt2S{2o3z(x#YqbHqA zNrVc%U%q%K6*YL_c8XV5hT;^LV?jkFoApe7aN#40cGmE`TqrR>!E2n8av{gwB8}+sJsP zzQSp&`=%sHv^_Pg{%Gyep0b(D@43Sw?RGor{P(JzXSQ9_wMvQZpLazgH$}=VvRSj! zU~~_LQ#uPCkC9T4PJmy0J{v6aTSE);0X(&LRuI>E9VJgccknq^ zAjH}bc+%Lte~6auqac)W&og4~3GthME@cNJ0S}vn#_v6x9GFd<+cZ~OulFhGUSu9? z_G|<5p)ACUJ=_e7P@f3aG|2(xbC^%rb>Dv9rn0EtidJ^eegcbkrz$qE^CC8M;gyj5 z9!%9xhxw#m_if?wBa8Z=czZ6X9+=%Y)uxVJ7opAw4-WFTB_1iUG2R}2%;il^YMD?$ z)ebR>LEM_9z@xUy&89HnD*j3{$sH{XZc4XXyJd0^sObhD@U8;$2c4S0Mi*9BbY3Fnq zOX>9f8r>E`%<0YPnfg{vKR#%-VCfCmHVmx$=rj!_7(FCQ6Wb8I8cDOJx240I z`UwbpKBJkT%6Wi@Rc)C!AN(x~QO*NCVgt(SoN~sqmVm^)Bh+^VZQ-xR0bLNE{J7Pp5(8f6gvLu@HRrb0|vK0-s>ZtDrGcBWcaHuUN}m*f81(D&c} z+tBm>KO6c0V_-v1SW8Y?%OjYbWq&pMg#Pk6LUkh`kOc4a_#EXiFdirPTKAQ9tqI8o^G=U(IUv0($s($he{vYfm7uJ~{!$FujBG~slkEXs~7FG63vKci|V=4Gem zg1}01b{+?<#jaXxScRafm0AoA(c59l41o+4cshuJ*I5Y1mu#QWZZY~ts^GPnjq;4+nE*xi;UU(%TsL)Mn2)Hix^ggk!WKMF3 z*{@xF`i~Il29UiFSP;-g1S(tPMs6iAMi5GU^&0;E_mg)I+wz7ZE)}JPN`3&FUhm=QEDLAMh%rPsqwoy z5B7KG%FGJ6QB7)F&er#%<|1F~Qus8L&*0Y7csxdlarwH57uu{F6|!u=$}0|ERa#LU z8!keJ5ZAhhspuU&iuCBe_F8KEQt_s{hW@Q4T0o>Jbq}tzaf1i}rhvrxexG17RG^0X zoigk~i|2c!Bu`l=(|6zB{3NSZpSXE^+^C5psFlCRk_`%`juhe$s^O{Q^)o|{G9eIB zK$}egWds?a4fZIY7Z!FWAD|zy1%HmS!FQc3KkbJ1Bg2k%z%f9;=1*nASs=o$HbcPf zNg;X}D5mL!0d-c)%nPgRIF*MR&uEINT5fA$RG} z*JO$E7IC=jJozt}^(f3yh-Sc=B;oo!^c#v>^V{;Y0|k9B0)3FaVd5Bh)4~|Ja9&gK zMQ*$NKC&XLU-fAr#v|Mj9CtTR0pr)%$q3r6%TisqKxmNIT=U*Ld)O2TBi|9~hA7fL zl#_jwO9CsJOA~v{QN)qr)40I%*Wl9=JJoCHC3x zFp9?zExrJrm+tr&OR#kLr9ZZ_OW;>%l?vBDNFHweRfweDK?`e-Tpv>RbPepE@toG+V54xvoN)LmU^ObfrbDKXjdw<@oAMZj;mY zwCj>HP@h4y6DeVi3{4lwr)x3nqj{3U(8d%fv~f~B%=pBVV{!ApB3nw~ZT0g?CI7gd z&1y1FYIJi8Wro5a%o-74<+mso)ln#IKUHRCGw*LW{?_?k0&fd4N4>|W2;c!reW{bB zm#H4OKT9H0p56>Ax?)z1+)xoHi(p!>uDD>TJl3Ul|v4zoi#qsc(ztBMus*Nmc)PB7d?SJ$I8#)e;g7Zcz?xJlAv&hl1q_*G& ze1h>64#&sTCD;{FKzvcG`&?EQaA8(D{@Qnnyv}7KZN;y22p@n}Is_D_9UdD*p#KLr z-o4%Mc?5a!Cdcv7Bdud&W5w4&;Pp4jx37&pqgwvttM zyyOYrZyQr5zf>`~a7fkP6lSJIyI2e8DFMqWWD_8xL?f0xhCB95Czq3A{4V&F6 zMIYRS%*5wBGqR9-&L8$u)=Di0cA`*CYiwY~MdzrT@3yL*HQ)q69M;-aC3A92dtA&} zQ4CBf6BT(iDh+FRDi@Z*3R36@stIwt%iTOHDuLX5E8<83PN+z6%x#50$poa&;t94m zUuFo@aGYso6)M$e2C;GMr7-K1#W5L7m=x6*G;}c-Tw)4G-uc0X6^3I|js6JU$yWP? zD-*2=DU%;=%q&qLFZUsTh$pgz!`mCr)#!Kq<6Xt*pyi5sA3MGpOe6s`Hcq6b23%xz zQ5imTB-O-Zi--`;wYSvtH{J6{L3OmYBgNh7$+65czy@QTV zV$Z_oymw|@GCaYvaoYWnE-by8RrIIDjOUqquAVb-R6i^eae<;kl3INZo1=>O zPD3Nj6)grt{FCOt>)R9y1xkhvJhXA?YTphOoU`-J*PF_WLftaOk_J?(_-pMWz+Mqe4-F9GE30FC(Wj64yb>WYhZfI3YS~xafzr2`)|E%-1;2p=;&1`&9x^MRrQL!v+f*hAUwJU-6XHbSm z!%-5#Jhji9U|`Bz{Tlv8K3nAdzqT ze1}^gWrzL#@p=3kGyI3kFLf&Ed7mC!E*4TM&}+qCla{_le-r`8ze!2iYS$iQ;#D zclskBj8f z4lZ{ETXVjRiB$rO34(vKW(Q5v`$)iUM@BQ}@O1yUctrnk@my1W{l7@X8YO_!Z5kj& za|#H^AXL_xCz>8Uy!bf1?N);*f+m7SQ!o@j6CX@fik7%f1E0v9PJ@Xr5O1cPIJG5? zB`+D}DxSDc@Lzi)|69A-=39HnTRX~s?b<-QT;kOFTe~07o(x4IKeJ(5-Y>F1w`g1L zVS#e@$sAlSZjSgp4n>YEGImr8m%2H1LM;Y!%u{V+baoj>^<+k8d-;C}d(*(!gGx}3 z(SHT-{t_dblI4?_3Z+9>7SCF>mXd<8&qqZ~fd4gZkNL~lSL8Xlr7Rcu@RKiZ%K7G8 z$5_GpK!6XMbr^aV2uj@Lw?uOdvN4H01_}`F9u1P{(WgVZTIC2Esgwr<3QT15;uL14 zn+k+7*!E_NWJvM8QF1RO!;ei1;Ep*@kXR7??jsp3kkwuN5TiAG0gw!=Z@HTc0gO0y zIRGOL1o;NFpf`7!9FtaHWMA22hpI`RN0a)=b?P~~mk1y}FaZ-TKNrSPz1lBd<5b-~ zOuqIvKKb2YIzFW8qGX8z;tO*y;xfCygOuf)H2FEaRTlC12{!Gq!<@dSV-gT@iP4%r zw_kF*JJ!MISKAX@cjA6~PJupofQSV)Z91>BT_YVmPTW4Two8gU`YyPvpUSqKeoBQo zdVCYi?BmWqx7UbI|1gmBXA56l|<-zORWqCL3Q7lY*w_y042QSgcDmq(K zE8pb9#ff$tBv9X6@z~|yIg>f{2Uw=&sB#ct6))7;HvKyn?xEpd1U#gj#hlS8hia#E z>p&+Av?TQ&j95i>3>6Ve(@?YQGNBm%pZs5(_SiRI_BKccii0;vKIUwY>`ffpK)I`o znJYG0?vV(jkAFeKR&hd`Fji3{m{3naop3(MexY#M>(k%50@z>=`0mC?WV1I&vU~HB z*|Up&P-Sob5wL!DhF(>u5(nO2R2U?g*wqN*pjU%GWM`&DIV_7uIwbKIT;THu44Am4;in!ohp!f+}QJRL0hj(TB+>MDBBTV6Hpg2Z2=$#tWK`1g*GrgKh z-wsEB%O|KEjPMF%C<%(%DsjZ(OI8GMws?DT$LMFMYrA#W!*f=Qeteifah%TqSY?{N z5)^wdhnC5oU$DyZbg|^o=wS+Bf`D$SEcoBveFvfk)`#b`)=nX4y7W zedPZiC^UNNQw}YsTLv^|HnpiQt=R~PSi-rJ;=pVIWf+&9*!XeB_(wfEMW^PBwlZ>p|;|>g}fSkW7C)Sou3O*Ee*@dRFQ(O9<^W+sG-Hv zZ=l5?5U4%T5HLvVATS0Y6!sq5#h@bil@s=sBPn(ljW7m|jWAGpqO+liq#IcY`syVF zSvpa$oSPFtM-z-ZgK$CYr*J_=7@>0{Qxt+B-4qk3Gv}iem|-jz`k#o3gCq#}gCy7y zKA50c#cMDINn{iE{hTy0IWYt`H*h25^|PPD0=)-*k<3vD7I#xrp(d9464Goj%jVh& z1%X9T*L*xKKv^PW9ZE1k0_^aZ(F?or_1~#{higLrTBY6XIKfsK_^lXOVYEV^=1UxH zVGs$rKoU(1_FPBkqw#_(9R;>4*Ntp3>WxqT0h+Jk#vtWOi9J> zTzTa9xh?IGPKr{#@-H4?w?0wtBK~affd3@?$$Pu;&X@U98yef1E&|(G_BqT+lgTy9Tv9F z@$pOKiAP=8Fk=C|LgN|Orn>SdgK?#|G_q{jogUfA^z=sD} zCGRqI*dQd_^q@=QD|`Jmt4Xmq@U2j4hzf|q19PnNNio*o{Lt`TP8Vqq+hgjYa`>}$ z-DH2SF!|>{s$g{alcQcUH?EYdI%7&0u2dxHsrXZBny`v#CaHq(~WPRFiHZvjf!o1j_v`s;wW{*cnG*%0xPn?ED zPIUIw9i|Zb5b04i(EEu~weyZ9kCnp+8W>=}>5e8t<6Dqf_Io;1HBR-oBdgNXM1=&D zZK1aThQd@6r)jYgomn%OrH7?OF#24ktc6S|5I%BpO3*1E2J^mjmP_$6Iwoq6^ z=EEd+ffF*5^B;+m#r>rGwmtlMy|McI3DQk8k_g=P0txo@_w|KO_0zuu5@0}MtJ`1& zohV>Hiy(?b?1><{e!ze>F=|m1F=6;Z93J2hzZ6eaxUBW1;eN`$E0P;w;TOVN>_u!B-ze-x|ml!FJbA^aY{iI!}T$`TCCxB5?ZTy zi4ji*F=BY{OP+oL$9-u6t*MrSb~*o59rJT$dN)X4XWYT%*E$Q{tMkk2$a>LWCf@Fk z>6~!%cJp$5_UTX@LH_7qcjC$0p)iJQ*^>V1M%CJVjZQ!Ge15hV)}NR5q)zCK6KBTn z_Xi=wgS~#VbY0_HnIZPVN}OS2DB8L&Of`l^WU!#b05Thg0vy{WEkNS2LHn_LgZ36z ziOrWoW_9d}IV*z?c9j9YN&hV1nm|q?bLChNn-F zfuz566y(`3ZAqv8KF3+8c;6MfDGX%sa?H{53LS`}2>o_lYH2#c`J#gNN9M%44tv`l zNEJ4TR)Ya)l#_s597`RZK#>9!?^{zrhYzNu|AHkNd>)LWvaVtf4%x9SdX&ad4^#gd z%P6I7(l#0@MjSxRC$5@DX%tHiF&dy~0A`qm3-dQ#OoJ|s+DGZCm}G>#cq|R7 z>?jFE!AvQ|wo+-XK`+H*7kNZR%5So1F0_mXifD`isT#7>_|?frtdO~sv@QI_!O8nu zRPh?+5mJis5mac5>fMyIKqHYscS>*< zC8;=Wzy5ZckMpvMD2YWrQC9K#m9mANy+A(SA&V0=z=uX{4Dg{vOP_`&0C|8mR;OB= zi6^VTWubeoe6j-pBDHaGXsQD(^});=R1tt~{||J#Bu3Nb&l}XaSANd^Kj^$Y0Bw1L z9yX*$UrjEQu9R%Bb|{U$B{QPtz7BVh8B>U-fNILrga zeW9+Dw3`jkMp+2=#U_qwCWMGeQX0o5Je&y?lafk$E}!XGUAjdM=oe0giaDJRv;~I~ z&E>O5Ykr@68$eV_8pra#0dOsW0V)ZB0e)FkXs!IfF)j^@s zZ_*(BI?Mi58;+Qwwa_0)z!C;zm=mCKituH~&rx8TQK?4-D79z>Da8)E&CzKiFDl6$ zzfrc$D8%xDJf+4$fuwPcPi~+Y17eVieQ9zvi#2w4kZ0!)=h(iHeXIf|7ez%$4&QX>+Qi zjY`+~%lOw^revYt$jsBs6Vu!?U?!>p-BW_Cwj=%Eo2%G9Ywg1$=HWwY|JmNvwMtYmHr>6C!Ijt%?C>ct5S+ zTn}xP!A#K{QU+}*Qv@MxY7Q;~ByD~lt)ir3K`O&^4K{=Giv?oyhegNKU(D;w5qH0^ z09v554J`ewIa+HPRz@DFroQI)%bHRWtw7LvBB+nGro#7srbz7H!pd3Oz}jOyz*>hn zPOSR&zU9Sj{GZ{@^!krPm#6MOBJiJ~9?s+_^UJ#22bor)K+H!QFWXI{_Uwx4x?@Kc z&-P!PR3U)OM6Yyq__(sIs#Dl|$xbn9kc~L)&(3L4?jysxm2eu3Q{ddqv z@#v6_Rt9h;*EH{%d;iMS#fRpTq%GwxHjF=^g+b7#r^RXr|>TV;A z^i;M<|HNjp>H9l7Rrez{xc7y2C-TXm#rik76WPNgOQML0SlGb~bP+TO=qE26v-aREZ4@og7Vxrow01w!_!)K=$U2(E%O>fHKFg8sSgu8z;l-EIyqa@P=8d!t`>n>6uA6`Kde#B?5CApj_Y#(f zWa$Ei&0io!MT`@f9~vM3o=!UG@)(Mybm#4sg9mM|HR0kgmy9xB8Opj1u}bNqAhhIH(XdNT1GU{eIWd9!r>}r z_C$B;`q{4NVu7LfqD`w}P-MalH@&Wy zCYR{Z!3m4pH2l;{JLs|(_Qi%`>{N`g!C<2%Lwf3q}eh+)3?tuF84CgY^pu?RVitY(Z- zQ~aZuKgvdlrF;Ry`OpRU1==}MY&X%67!ZN&7Kcy>t$|P&nrfPA^;P+cnQ(y3)$pqv z8F=mx#N!bX5Pl0{n!u9)58|RgDXV|=BO!!@n3ULub>Glq&=j<%QC4o|egQ^WKKvHe zR^r0|?a^@%*%Y)F)NdJdkVgO+0~-UIjh@WqA^BiffpOPAMahka@(!< zcF;uRseZ=1S)M1vx3xoI*YHNC`Rxjc4)Ngarr=L??j^=wZJBJBw;rp;Z)GmL++G$;=)#NQTm->CZnkie)+u7lGakIfMi)v?cvG3l zK}KYtYEP=5sLj=US9gy1YY$F&xC?GLMmpom{Mk?RKDd>;CHtk87M+RN#X(`-pT`nX zeF$75b>oW7JS5}}?;@fFf`^7=WRD1fqvJ6@!NJl}wpLec zB0D8)9uD^jjK}0ODay|`GlLf&8fuLegzS-bg~o9wF_;UGQDHW*rO<(iS|ny*z+2MA zX1R>?9Mj7r)@!E3G#M-4E-WHyn@&)3o1uQy6?$NRQfBmboiS7?o$PQ^F@ zJ8BUQ0k50t1`;V!1{4l(V5`sC{PNema(=#ZT*qKMg7bCu#Bwb2e%XN2gm|!eGK>4uWXrf@O01mKd!-y(& z$7&wg(zi1N2(Pts(zihgu5dsF4yn~Qci$K7@eI>jeRlX=?(*G00eH?ZJ>=ak5}V3Q zuF@&XZj#Bdh)-bMFD?@ocww!Qa%ki;jEZ3p$OjapP!*Z)2UV)#V{hA88LAHJXG69{ z@jV;Sz8sfic6?#blX&-dLQ3Sha#ihaRbXAxYFh|%7vu0_u<3OW6>#ja_yN<0C9;~B1ZUF# z6A@RX9b?l5VLaFg2s?oZ_Q!a{q`ErXNx%TDLs7P27s_mEuu1a!renT)VeYNh3c$dC zk~~rf*FmI2Iz|XSQ$}dGDNmyyaWu|jx(PCc4*XZhYZg5MPi#>J&iCvEPdbaH$k0*Z z7wLMS)dIbk58F>zj~s*1_fIRSIS^5wwU#>k?0mN;V(o%RUlXZ?BG-m0gggm>5EHV; zO@ALE->5OfhI9yvZAS~jRO2kaIPlLMh0k(+JL@9Eqp!_PsS8hPh6i=~hQ95#t|MaE zvd`XI1S3hgv9f!+N8u*}3$+i!KiyMar@S)~b`GQNAuk?Z^KLGi$j~o`1;{fF_&=vz z4qH5O4R?YC4#De!P3%(aO@=RO83G-sWaJ3j)w2@^$H}@`C{?NzSw4!LAtg^X-;c)A zOU6oIIGWf0ti3AK5DPsR@g0S(W;Xgvae?Z-QylF+v)cKb&0nE0f>x>-PI~!3Oo5JP zs2w~2i=5@1#g5L|xX^mbS>*w&&Yv&apYS_sps~G!9M38}RzyU4`QC*DIj-H$hdL_% zJSF+*z1_nHO8u|DGx3jO?>a8&yWK2DmaF-z3uok;;cRr}HtSRNn(BCw+=V+^(U$f}WADK>bz-K;+r1NIWmb)c z^fX={)ED!{Y_1@lJcq1j*1zf*-4g%E+a6MJCW-BJRvuBwbIvx%-Qs_3U508Na12N* zu6o>;Y=+vAS1TY%4suaeRLd*KHYku3I8EN>`p8SAMB)i%>;~~R;6xz2pYjX)MSv`6VFlHQ&a7jF6-gD zgImv>D--SI6Z#P1cow6C+u^aMvdY76{{T9qKnaGearfF+-|t;J4te3~mT3xq!p8Ri zC75`ryZrN)!Sf(bBO$RVyPn(8C;2F!B(H6d@4B9e5N=DOBQH z1IE&kfB28A#9g9M|0Rw>LNH!YoMFm?9^run=zv^Y5HDUITht#W9zp{_JTe8MTx|}L zhE=4nYiyZ7(+U7miA#iUKU z*$*Dlg9dzGh~KTe%X-FJZU4P?HotySx|>iahBq7i1l~!eVbC2Hxd^k|mD6AUoGBo# z)w*O--b!ScF2=5UcH8NvPvYv4F@oUzS!U;zWVFyhrD1b#^D7lwwBPDlG~IpLi(szd zWMQ0ZQnKGmhqAPG6urP#Z_%LPv?Xyev$HRE3zpn{4&kNRS}V)Lo{!QxScB&BM|W61 zCXeJgLQ{nGB1BMYu(V7DyVWYq;9YfwA1fz*8!5QX;0diol5h(YM65zfU|X3ucB)OA z$+_hp;4FlPMbL3JF6;M`5@jxyqH_n(fu@mhl|ef2Kq$rmvDHNs9jNb((DV6FQsx6? z!`mep8n=fm_Ty9l-Bcf|b3x%S>$y0#6;&!Sq4M=&ZW=Qh@P4mtBC4@^+-(j)tBMKo z7Anh0=n0`HXxL7r-R&wYxD9r+LO{jr2q~v*b(50n;7K+^to#H)l(-Epp>#Exp7M(1 zi*7@F7#d zV1^=%*|;w)qb@vB%{QV?-$NR|>Dj|bi-h`PU|j3)L)Bh#z<_ndb41T|o)PB`gxU#d z;Bk>~8Kni`Lh>;Vbog;18Kn8}&_a%!fV(fj zSk*zmaTyyp)QtHZBI#xdB(hJ(gh&-g^23FAe<)Mt!-EJ}cEGWbDnR}J`re}rE(e=J z2AewI2#WE9%-3HYfI7r`4XZt#PDG%q?++n2P_hHK@GOaNtamQu6=HwG5*!`LB;f8& z7>Z_W>gaZ(q(gF{S<0#+a~T72!qH{vVuMB23QH;}VPhyIa97z>L%=1<7J?PPIP z&)$0yVHE0aVa6e&jxaW4B4;ra@OuW zdubfaG((e4r?WOSEzTU-isCq>GV|^(tvFkVCMO4zF8$K*&CDj2GoG%5>hP2YJ@^C? zFKtb~+u{A{D%g>|gYB=YRGQ=+(0qUCn;O?`b)lR4 zr9s9qzu4INgzi`Gq!6eKC3e*Udp&q}Hs~)$B5j&P?Z3 z&D41#_4ms$IW+7;I2n3e%q%QQD(OBt8Oq2Zl`nv+W%-kIcN1k=M!EC1C@|J0J3t#WMnPGNJt7_EC-xCR3KI)b?Fcj zWXyZk2D~4UD%j;!O^LCe25I4mNNYSW1tsO|EO5TZ9&MqJ-p{MgYgbVBqh^xu^N|h* zRK>*l`rTSI^vnF{ED`$8kyIhXHM=hBoLaA^6Fxb?Fv0B&0=d1aWo(i^NtxrN`prLb zJ_Y&hkDj3`6TWnd-N`wBxY@|N&O~|LAh+Rsl9CAZJ41g_*N+#uvur2r6}t-qp}bB- z0o{ZAaBfZBz3O};csW1G^Oa=4qw;@nc=WkWf|Cm8MnQn2oT;v+_(c{1NqLG9`3X^7 zCmbsgdQY0tHQ&~>Kuom5!xk?dy|c}y)3}iD_QU0V0+5zic9-IHZgq^Q(2s#7Nrj6n zEpN+}Mq%y4$y`2QQr%Z-p=nA4 zEHX{25VSl*#0d(;M|c_=ZeUF*n+}9cBrSoY@t01Ai)Vty0kTjFx zf^m!sk~oRkzN~4sqsFp|rw2=b$FR_HmXcP;{>{}`tDKe`jCR?a?J+&s*%9m8q@x5c zCTd->(Al1Xzmc@=8lMZUv0k0R_ALB@tfb268j(9v+MjsH(yi`IBX?c0N!?6z8G(f+ z+D)&>59?i<#eWBrXSKDF^jUOzxEo+Wq%ZYX1#!HQK=qS&UyRZ-2?Bt!eVnu5lYFj&4wNXajnEy~%PTp0nRRGYowIRVos*%#^^Yk#kL z2kGyUe&iW-dp@~(+kJ-na_v1h;*0m0<#!HdWbzi#q{{WzkJeXbxE(#(;|^E3*;Ma6 z+i`5q6Uf_hzsEiGP~wr2_XX;>d5?sf^Z5V~ZPQm06|)!sXP}7$v!0lFUu~;cIBd=S z3&p&3xC}Z%4F#S^KBvi&Jq6I~R{+Len!ZB9U>%Go3pK(Tq-To0iw=0@Jb(5vWvhqS z)~TyaJZnlI2h{0@g7qfIVJU<1&&7CUu*hox5_cDNhim$y_$Ft9W_f^?pi{%o5`SN~+0UBPsl1 z=yNbs3wI1LV;p}Euo`Rdg$xB$DVsd~oWB0;WP>-Fp=3MVf9Bw&q5pDyl>0?*)Qyk% z^8DrG1OzBM*LRV3M)z0EL{skHtDtzGaSp0nT)PTBF#q(q^p?fGufeAy19IhDJybYF zqw0Kb&cZ+A#Ss`NJ34CN!#hBa^tz*^t_ystMe4&?= z%UcMyk!;sBK(+3z`Zxu*T#fnh*~%0iD8zn?s|r@YMrdQk zTxId;rVG;SgM6+n&Q@dUB#6V)PNph+p#n#1Fo2Op%n931B4iLK(2xYvaayRO!&Ji_r|syNi@~ zZ^q%WIs^Z1brFU*B!eB6o2I;Qy@DD(3kD-5W6dSHq&NaXL3!4_Ti)6jJmu8G)WclU zwQ(caFMqs6|2=8F>LkKn;+#^PAGXaxNkLOkdPcL^*w)>ws#8!qh&u`~qu8K+31q8) z7zdT25Xr6pyr$X&|Dp_S-dZnIJ#l0^`8s7}5PSgM6cnj6LGzy2XTN2LzOd>(%US{OyDO=a>vc*0BbSd%`@~SCAOt&fW~Kfj8*{ z{O;hM&?+h`{-o>KSsezP*GRS@YrFQejYFg_C%wEv_H-C~E+TfCfv3k-MvRUrWai?z zwgUr1UIqiZ9?lAY?UDEm71wsh}Y@Rjdq)+S&m@B;G5^ zehsZmNd(31U($?-nZHEHeRK+C1%WEmF^A3#T7a=?3!*~?rhQjR;|ilZ;K0T`?F3uJ z{q;3%E$OkdkA+*pMA`&L(WGyJ-3m&%T3gAU+xS|EXQ;TCew&5W7(=F z(sl~Q8P!>t-x*bC6kx#bHsq?G4*uAXkH`x1%?@!-QVo;~Y?1@ptquzjh2FhgR=_A% zbD?I6V;_Cou0Z^8AdbyV%1ye-O?sq{9nbebymrT(sGc#RN$iu2ZZ6C1rP(V zJ#8eb!|&XZUo|XvQg{yk-re|uo<%;Zn2$6i&pM`df^(CGE;!(S43dncs*;CQ7mXkI z%=-DtW${A_`WDPG8LQau32`Qqh`08X_P_8}!{2m~-L#jBcpb6A-F&rB{aN{2Br=H5 zL%Fr>lzK4z*a=vOyT$C1KDfUw(xShdALVuEy)vkHKRCP`G%kt2ziyB}^;QYi-^Q(y zZIZ9Q99Un+Z84OuyI!isjf4K{+yivRv0;4LUJ`SDr-ei@PMSi(Ll2ozGLD>xHB&9HR6W9i5$%F+h5t=P4swMNwTiWhm2`|<)I{+1N9QH}E+Trg z0q>3Rgz*$i_EZgzc?Us$q^#-gjblT6CSfmSUzFcqyg>C7AhtMOcJosG0sl$jQ1Y;l z;*0JNpY>5|YPJE}%f!XRXS~knoH!NBRA1X(X7sSV@#FCrkLJw0=f=n3Z%cP}_xZbs z!rrU&84%#foo`qMD5rQ{>HX8T8Gn9Wjy^XSZ+qTT?CMnbE^a9LE*hPOLwiV%bx+99 zH!5|SU$_2s)lrb2mE{{mxBi~j#9$=3>a`EXG3KLUfRb5M-iN`dc_d=k@>!sECw;F{}~%{5~`&MpgmfJes6^6b()<> z3rd21Pc($p&x6Wr{GLir@XJ##9K4XMsn4z2a7PN6_u!Y0T}0qOnXE?jCCRztY} z?fau}_|&hK3R*&`3{{S1O{BUSMGw1iagJ-52d}+sj_Z!q;PK}MXEHBA`n0ULkYbEP&x>RfPe@BqJlK(HJ}tlIs#GzX;Sq+ z;r+gU-Fw$rt~ET*-usM`bLQ;XvuB3St5Bb2Mj_YDG0SWPhbQ82$^Y=mYaFwT!{cx` z?|*m*fSYDvmvDFtfYV5^7_D4Av=u5`T{O7AQRC3158{8KummYITPY;uN2Y*H^5dze zL`~`yGNvA?KusYooS^z(8fTaN`7-8WXi&+rT}5v6NgQ5Q*l)cAt9U~Avq@M#d|?Pg|hT2Ze6O_!9*whj zwm-zH2g%8}K35+!Q4@S3qo`^VknLL5AJl*)2r(E<75`#5S@t zdIVhRM%JD=o~5zENNxaEN1e@Dz;cT#oGx;Y=i&{{oTIDL;A=W!f5m};BvIB+^^rK*MMAGiKkiyMEV1k_XGed zB>4Yh#rEum*yYLH%WE@YGr!pbm-m#!F6Ym+687Bkvl!iqJAR4xJC3FVF%Hh8^@SXM zC8Smp`Pbd%UVPE_k>|Jb6=+t(&Cjfdx#YC9R^p(_PkfJ_1Q7|Xlp-6r*kiorr`er{u4kohnDW4?McR*3Hyq2Frjpk7{6+!`M)$<&7aMZ< z)rJN$#pi}=rq*V%o=N0~XFGau4ew!%y3Ud}Z2HUz*9?otR4;3Xdnu$h4&~mfL|!c! z{v{#Tsd4nL|9h*H9?7TpF=?&iHQzfRxgT@PF=u3aaP(g-gDG9QH zG*`NC%*u$Auzl3s`Vr<~nik3W5sKm2-T&dl)&Nd0`adqm@R}PW{7-M>2f&^F4;KQ6 zkPOehAiCxvy*GxWl>1`)JF6?!oKg%Xa_M{_SV-hUY^JK4-ntEB86@4ex$4 zaIIu^8mPewrk&C@O8j6(zcmTBRI3zIex~J6JT+y!$xmBdv;DT&PF~%Cd8_QHmO~Xu zu$L$AGs_$Y-7>N9t)vUpb3OD<=IYj(*r5{M9qmwaO?>esf;O2inDMURU+hsi9-+Je z>G&3$ZENq8x`x}xQMRBvM`_n7=CmJIrT2ws(mIFb+w6%`pX-D+e^>L8c@18&j-}V; z#^tITYEGQ8wC%)EhMvXQPL{fSfxhm!Z|HTXWY+qk>0Zw4k?4yq>QJShf2T@YVkj4mx=V`jt_#MQ4M5%zrzC!eJ32!{(NrBp?;8|IWnG#}|cs_gi9=*8F{M?UPtC=zHe zeAH~$C-$wLRGV4*7I%`|v%&z8&(~2C{xze`<5_Nx9TUKh&7SFq4JsZpVk_U z1m#0MC0}Z>-TB+`r}4~%Dr3WST0U&@`S8sq1^74gx z=AvKiZLi+WMTGtEj4JsA*Ln$wM!;WK6j8d-8*vQCM^Cbmef|l5qU39WIYD5fpkU(rQC@`mEsjTMbt}%`7>$?v-#%|CrB&EHk zKm5A1>`X+Ft`r)s^`(nVs&dBYT*!V@Z^CqZ>>=;57R4dUqDp`9mYyI7#sj~>l@$09 zcQ~jaQ9D);AcdqNh#Nk@9yKdY2>f;@>BP|pBt1Avg?dwEmoSc_w={)?v@&#>MMYt4 zh!dF+rRksGa5s7h!%Eb}xyQ0a(8t-^jD5fQQEP0wsl{J_#l??N0TMrZj}J;t3lmgW zE7H8GS14D^%ZynbC{<&5unt!q>*Q4EA(v729~kRd!uXNjjdGAIf6ts1v)or&t209m z3Cj#^WoK`-u?b|Qqhl_aYc+L z#x0?BOMp>%71&?7LXwp}vvdNt)3S~s+HHr4WbibbRMIAbXLZ&kID<5dHd9?{%{ur= z3abUMx%213G}xe-4sM%g#uo@FG$Cw|BX6C@OfRmAc#wm!jE^j{;gA=&L%yXrS~ySs zQ|s^BCNblme)rUVU3`nGIJzg8=TBC*T+OJ;`?*i1g;7RS%>gCP7Di1iGZiP_!0hJ| zs<{C1TOtdylxS-Z;tvJ;Fb2m!4^U{j5Ti8nlg?2{v@1til!+_?smxY%6uWC2fU+Jq&}D371~nP^fNoKe7a)JtZrbTxsedw z_-qDEahJ5QSNT4v1FjTgp0DC+YaU8CF2kGDEgfqHl;9d(2PsAB%Zw_YpTJ<_W@xde7u~a=PI_7^1Q(Pu=2}!uK)s9$M%@zOSQfXef<CGb00F8G+@cUmG)G(Aa{VR z{7P8!Yae=f%u7D=-5?>@;eFb_x6oLg|C$Z&*;3Z}dAWMCrP|#mYetM_`y}W?a#c%* za{mRdAOF5ml0MixO6qFR(I8C#Ny@xrct{gd1mPrHblIGQOzDt1$V&#%CU-S#s{hA- z2m|;j4uBu{7AxD#sX2|5j$u*x=;=@DiU51!kr^l1vQ0S62$pmyGkrJ`JXVicWhTp8 z31PZ@L4^h45}H{|7$O-OE(+`tE19ZP`guv|MnYV%i;RLh6A%?a#qqtMN%yS3MU3Mk zyb}R_FCi{6LTDm{nzKjm&i6$*h2PFWl1z)E9^%{=$tRP&TV1oS)+OBBmuu+m)n0m_P9XqtKpNU_|j&)Q1Y}vEvior~g@|Z-G0)a$B zBbr}3dd_dK)>N%O4ek(bCd~(6X+LP#s`E?V3ckxPvpLjKH`%vh5SWtn;j0kZ#-|tj zs)V)))++kVK)6*l+Gb(#owvv<8C}D~)j1n%Qp96`jka;7x4Y$am;NReW`9=pbiIkh zepgdijxidL_?Q$lS&e6u@8#XOlm2MKa$ZxfZPXe6~$=z#>!)$U{7 z46AVvlnY#-p4`wgKg0=)R+3qu({Q|!8pjVxi>60_fr(#$0f${2-#o*kclG=C=&i#K z7r3c>=8}hjH)PY<1{Xi3Bssp%c{RCpN8mJZ3KBtSuU0bHA%79o#z^qph+l!HL`h0i z)G0i9Z!B?g$zwmsdBU0RSjHQaaGtRK)-VwN3ptAE2bUSx`$=+YO#ZwyzU+eGbC0!; zO-fX$CzNICGT-|9C`ZM(IGMDbC&YRN^DrIrXvKJb$x7wUpSRMGNzkz*p77R_IX6ro zoY)j+)@{;wN<6{AWX-2l!#yFOfYV#@VVT&eqXvw97~`Yd-Eef&y>VN_J&irY?Q7p3 zte=LY0AcT@e2e8szJh$gXe~jIJ!OECyC6kmWy_@F(B?(q8H@7RO0Q><6EY` z7NE$L}qAqi=S$fe}zWgB<8t{cI3f$pEZ-$0E2k~!;`Ylt08-g zVq>-xc)57gPD*eivLuqjyvlO6l#Jw!hyDS(@UxOP;SxVvk7Ev`ur&<*I~2XZ_J@J6 z-lHagKHojF6u(}#o)8#i)_X=iOCyHc4CICU&cbDQ}tM?a2a$V;^@h>s~H%d_jyKmR=+ zegXuAzTR25smjv&nzgU^gz(YA;$PaU&uu+rAyO)qfjw#|zP>-c=CWVh_{UngyXmqE zAG#@Kx+nBZaE(^!DWmCF_Tc90_}5G(l{ysbBaUON_df5~yt+MXeyiyiM8MCtOHH`Z?Xat&pPu$>0Va(=^!&QWxG-)t51I|k8O-OpK8dqTW1kgFpZ)u zcaUNJBR{%)NO3#8%qdMd$RbtI-r+SBtohzFx~2hh#9)#(@i5d1RdoZ3LPtG3LY zo#Y$P=W}o3J9wsfYv*WqLma|C^Nx9@Zn@?t1b#|p`B<@qehd9rLjSGI{%5aA5;i&H zV~}T(cX>|Q^E=)_M7bfKuq!luYk;BkMzo|v++S#)8Fv>0gJvoXsq2maao_1*RANMy_t9DA+*|z( z;$lv7oNuJ{@@;X0Aun(yu(X50i}TPv(Kl*)D~um_ao$q))>fWId#p#BBOXhZ_|zu; z((G^iC5B13v^tZX=7`m;$ysCS^4});PbyyOlYhl_!W}Yz6*u!iiw7>%K_C;0`#8I0Zfl z%|+bf9${&hD|oFS!v_ev;Z!lQxVp<;*S%U0F>)DrbT^z+Q}LXSTkheQjA^*+ed?wt zIjqjjIV-8Tr#vRAu`&L7g>dpN>TY!f2(ZG@HN+H3w2W2dWDyboxT+EvXn$HMoV=Se zPD7#G9N;iIUhp&rJx-QbO-{7`<_mx;&9UGp1hR7c;pAUwx;e`<6rk}51~L9S*|22} z7W?XR*v56B8NGZRXvVJpUxDUYT%cLVEyl;Q%A2d8IZxleQSIx~Bg+V@81KzWvr*&} z));srHRY{uP0fzdRIAI%SXQkwPGJpjs4xyK!l6MH*QhQIWyhiSacGlym3ynVXBqHI zl_pld@&-=u2GAq3Ygu)iED=tY?Aor`v307*A@G^h{@o(t0gGbOJBk|Rcsv4VGvF~fXrNxV^MUh;ho)OaCX4>bz zalG{%svfB?-La550v?p9=Z<|Fc>_%^fUjMm7qWc3Mq_93s24BMw5RIh*1tqM1+g-G zTsj;~M1uaSHiKBVHjVSU1PfC7Wv<)h?`h_KakVnxcyCgA^DF?1O_rH)n{#l>x%&=E zZ=Su(l_U*d{u%2lJ~hd(V@9!q9j=we^LGQg2lwjQ`Sdk)HxqbXsVluXlnIJWW3sG7 zzjz+J9sZ$6Jg7d$2^aW&-~$A{fq8tUU)h9H4rguVnLfNFQR(@5Z?}%kvrNSl2(A4= zVx0sbVnU1E7l_${4YCPOseME%1*bV(|7lf47UziTah-Aji28N;m7%9*I#q}Ci2 z!$QWdnESjkhdj-gJP1XO?aCuB4UtX{s&{4B?Bu8Ekr+LT?UuEUA_2_53*@xvMr&~6 z)3-g=$S_O%vdNv)^IJMeOOlmEnS2wEwIoKFl}a?I9t_)DDk29$X-pHPyL$1+2(Hg9 z|0q9R9Z51RY;3=2;}O%rrHyMc0BPgDrM?}oKKi-(=^R|2I=%q?>BF(MI~w{{fnCiw zKF6KlQsBl-Mzs;WE9qmrat&iQ_ zFxrzTkPrgB`@n>NR#_M%WE=}CSi*biQQEB|k07Y3e?0oK#3e)o(fes-#Uzi;((e3$q%}j_^XBE}%Ze`2)hK-5(ECWdzB+89AD+aN*+FE&QUfh!Tqan4H z<~wNb)288*BIlXVe5_!)TRG{z%>RDbz~;or`Gp5?*}|JtnrUIes-wWa*24P9M&P<9 zpi%s-E%s^gX3Xa5(FD<>C-wV{`;Es5{piIyKD*PGd{ulVj>h$KP~412LBi>+QHpZBwO32-mg3(<|QFfrvl&~4PP_&b^qEmZQZGamgC{%x|w zJa^%DJGC%CtYgXIyB|pGq!|z+@s|xZEV{oWwB`ep!w7QLeaZbBI4J-XRL$}u(0ziO zB1-%^>hH|&^cO4&gxyrt_lewXNwnzmO^iIm-1X0$+9bn?AG?Ns7IPHhuUmQH6#T`iUjnHS7jRb&@>o5j6qz0i6ij0`C-uXh@ z3G6c)CUSe0#zGV*>NTUt<_Z9fPh;MoPkkn3C^CH-{U^V8yhT+X%Tvy$tU=D+@~XCU zBsk2OY(86vd%Q4cp-@DcGqI-}=IcI~_54Krj_w0Y+3?(>!^EgjfM~cdfv#G-~?!{4ala%efGXkBimaAM$lGgvHrn(PT3wre(#}sW-+#3!YAu;YI89^IF`mP_!4U{bb{1 zN3>l(oTsyGH%L3(rBrTd7w<5HJH8`E^6qqlkVx(7)i1?7x*dVBmt)c`x0c*W;)}Ww z6r0XjuO+aqwDu+a|3tyAqFd(Q?FY>tv>iI8JesX1`Lo1!@(tFlitI!`YHQ`U?q(b1 z%|?yss|s#9e-^=lAI|OxksjFv&~JiN-R!$c<5_DFqzTT4FY}#N6GtnRB;PGrIKL4< z_PWgp7)1%r{`L~LoIRtg=%iWG&5dR)kRYgVYJ72J*recmAbV-;8uuB8O|!EbF+@6YKbvWIpq;6yQKRoAhm>(;oFf5BT18gW8vSWtDBCJeM*qpV6D!a%7IHKx3tTF zhZF~eX2)Rmk9#axEQNjR;$=UAPVajD6+6H1o2c|s`@Wd7{b)m*B~O3dIYU04^D&Pi z`Pmo0;+@Somy35bhhJmXe{{^Jj@Z_jXuYNUW@@WuT5mh?ZG`E{O3owai-Lnl$fdBy z$FW9h8o6OV=2~!cvQt5MgJ=LbZdYK^R02o3DRoz5e{5fVZA0ZZ9+~N&)cb?^z@ESW zEpp0NL(bGlA|W&Fma0k{T-s1^&&QA`uJ4-e()9_(<`#{VHt^lsc?l&?mA&mcSFI>> zO_Qjd)kO8dqBxoE)%O+q5Kh>AmVL`Bt%bX1X~Popppxf^nAC(_mh6d@TVmf6-}gTp zEFR5YfwW7j_=J+&f8*_{G#lDcOkjDr@+*8!(VlR!GGcR?pvR_3eZCiPvit01UL#0X zV6@MX57*XI{!{L`qeeb~?dQzz+X;xf7B`bQ@-R*A^BsYcr7{BIb>9Cg{^|IMC#;CB ztn{;nOzPJmvI!0*E$TAQ&5~l8)OjUNCVPIZz{!Kl)!tnoENbvSVNoDyf7xK=u-8#! z<*>5DqvP-8?}nP{3in|T1%|ZB7-hy+86vMgcm24!Y{mSnX@eb_^Q$e!J$D>IM9AJd z-~KIg?kbx!Aug3v^Xy>X;`G=iGI=mnG3_AX^<9gOvDXD!>Q?Kii^6r8(m$=6=18&vBsF5F1}p;uk-9Q5QJVzTZ|H<*b0ut)k@wC5@tFW8;!g*3fcVAkaZ?8at&i zu7s&7_xM9}o+Wg`8StN*0&>Wjgw8gS6uXInPL_MUhY#12x}Yg3lu?`#5S_lzgie?~ zH~yiRM81w6{&s65?CXZ5p0_R}4$yC2eF<^TNr#TKbn3BxBpzvwP`MFtxfzhP@%csK z&VEX2{RHTiqPzlo_-x~QEQGH!mJh=xVU$={JFCg12T5e%%3z}sEPtR^PDwQ}Gsu}i zoqZ30IWjbPCUv>cZlsAa8eHcOC3zlnN0TISQ1U!LX))ctscO(;Qi;*y<|%t!puxp2 zl81p(sdL5a@#t`&W1)#M>Rhl3PC%!O=vpU1i;BlW&+9*h=fwW>$$t(Ww6{Oi_fa}m zDIWC6fBnq%`KagFr%e}`_JZ*{I=6t+Lb0`Mt<^ByI~d0wIV`uupZ#?DZG@IR5Fsj( zi#V{EP>Dd9!$t!>w_H3t*a-#J4K{36-gj)y*mOhni-)dA{q}L_WBNYd>!>;?77v%^XWC4V%E}628ZXgzdAx_+G)#1BDnz*I z&|_6d@-+{55MWO{vN^sfS{XEw7<)=Mx?)VjUSe*?GwAs~N(U!Kkq;K)zZPqSPtnOB z@d-S`WB7I-f>Yb;^>v2HWTZYcgGSle7u#g-4aPezv_F0ztaH|yWByy1;#*L_F~geC z54pl)Srq<8CIu<(OkaLEe;i#Gcl->v978S+-u<1+BiVjjzunQ^hd%ztgNAM20IrT- zNoxUDEishS|D0wXM>IpX|LNOPOh3k5^QwKT=YEy+wg0;6{+)jynf{k$MP2(3cYPR+ zi$CswOVgu(8g=_Gso-$vlt_Jy#`JXR)hx}rc{&CKA(zWH184Y4*EWOc|K zW_706*qU+qR|7TvMq;dkbY$44J~YqTUhHqA^38%OTWPO^2d8)MpO#JSe@uyd+L0y} zzc{|pvT^W_?JwW?;iE4PC$1dVDSwo{=Dx||>%RV2bbPZ>Z+a?OeBJQyHS^G(Xw3_i zg29jBXW#yakedg|e*AdzSPXHRY_iq6CWUq&RIDw%?69>EZMHNR#|KlJ!dBPpS@8Puv+!30SH#ylYQu-Hs1^eY+D%-0gy6aJTyftWiDpZVFdn^>Rx?-1l@N z-*-93DBeGxDx@`Z7+^V^|B~Lw-+Kh47$(~ekW=`>%h!m^UxNfW2+FHw+<@bhcO`bS zweGWqYk$%C5?7(Ogr&*XuuL^w`Yv-fC%|*4v0`l1WZEyI&swLt7p3&ku~~%>z72QD zm{b=@&^91yDCW&}m{k|~*RzGRq-qsPyOmp^d4aNW~A$TZUR1HO1zr@%r zBu+j3%{f(rv2#bTCO%~cCj{d)=v{Eo^+elA_B>3pL$6S{oblUyOV|Hf&67=9CEPF~ ziL#4_Cu_l7)J3S&6OKh1b{@b1ne5uV@)&C>L&7gl?baEp=Iu^{oE<;4(8Z}?`|s~9 z#`T6!6dck;e-PW@sN$|7(PiLIo`(Zd4h#^lagtAIBk;IC@^bdOY^gWVd-n!HmD*o9 zr=jiy7S{glX2}Hynt7`Fk+**6W?2}e?ANlj&e!bwJi)RK-IVPgd>}9q1#C?=IC;2U zFU>QuyvUMlC{Tr|meS3Cm#>+IvJWh*qqw$$-B-DnrjZk)cv&BbBum}vgElnE;G4W1g4K~*fk!m|N%?l@b=6Cl+ zajwe!eg6`|vnZzTx!~h{%^69_)(*)-lrZ*7dTRzjbA-o zY`zUY%Bw6PIx6{)C^w#xS;t2}-@hnQ;Wp+%pNof#D$N8|VSq$Ecbtl`(GDL<(ECj# zuiYkI5}eaEq38sxG^V5HOa4P80JMw>KnHXHlr`=e6~8chNnk($pq`2v(;B7I2$C)@ zb?4+#OOiEuz%bk{ww6G-6hX4=rS8a4dYA>xPdN8lK3}n!HY!9>S8>{y*eUZHC?$|` zo-R0+r$D!){b|M6GpkQ|-)xcpzv&ihD8w5-lNULToPCk|4KG zbSn4MD-w|Qujr4C$s1BKs2S*<8yw*=a)8}99_Lf}DjxfTCtjE)14Df)p~TX~n;4Or1TZWxZbbC4jH~`I9w`bW!QoQH zM2P%|M;iXAB+Ci{j`$j#lFCe~MEn(B0#HZYDSf~&TZ%8SPZooG&BY+#L-Wm7SuMYz z+!AM;0WM%g8l#biN80wO#FC7ZMu%8RE=|HIkkAQdU0dT=qTUkX1ScguvW5;hBNELt z8FFbY_+~tX0)LXSFM*>WzoUX}o&*s)&jIz zT}F4MG2_MfN$^ZPPfV3apAQSUfoNSCKT`Vzx(+1i@dI zrBXlVL2rTx$DIV`$Q}j`c_YaKe!b;B3NxYu=9|FoB`FF?j5`YW(xDwerBl_n$zT@= zKaNSrZ6%uQk0=K~iX22#iO&bR>Uc0QA&OT2kTKmRTmj)7oQW@aoWMRPvREoz=a_ke zPQpEQi^3OIkv|@om-U!a^3Ip6uc?t z|HWh0Spun`h&(x}k5EObOQ1crP1|J<`i17s!)BLflOPI!MtnCbCE6(lQG||jWC^L1 z1Pl|+sIt!<{YSn6NLUm;+y1X8p#^p<0}oHy9tBT?Jp?ejk9VeyW~)yIt1Qidy=Z4B zJVi(?`nQQJR`6B)bBXHi7H4g=*lj%Do|xI&$791XK`G=+Q&W#hKEAV-u-~slPd<{& z8L=NFr$q8$~a~#PQ->k5U{c+_AIL zM<42h58Rda?yU&WQs^H1n1XAp;qi7?e+xsV;yk zPVb&h_UMJ8H-apPR)_4~hL*k=kOKH@}`|eQL0FNskt$>h< zMVEQdM zBo2p?a>?+|UV%;acVq9=81M^_vS-g3?CiweN!|zXv#i9n%p434Y6u;g907@5#-lt+ z&x>Qpg5ZX7K8o4~ZxSHdr`B5hV(L9_y=uRvYVjkv&?#aOkTL{KLiFtwtFroSVy)$y zG+zDB1&&_Oyqt??^omcxH|Kz9sapp~na^erlqqyg+gL|wIQo|cCJS?saDaF`#e&qk zO(alSofzzlYgH_S7PpXDi}3@8xu*mA07!DllTl7~aBm_Kkh>#`uYob)z@m4YIh z)cehE=oz)ARkoa=hW3O!iZ9`iC=RHC7MyoONB@0>Mm#mP)MvpP*--kK^a)cgNE+iA z>r2==+DW<)@^&-KIi`ltl{0nc6wa6*p`v#uIr}@Q?C=fE+*szRkWZp`+Z@R= zj#O@HIOpI5w6%Z~{A&S>JIfpuS_y1;`c~^(`jwrAYY8CWPAvTa!UX6$5i_N_-s&Ec zsgDgYJfbqqHnZ~{a8k6_(8;j?8?rbhcFYfdI2yhxBzubm-;di|UXB%bDfun5>ta-J zxNToSUskzG;?wFw)1g=vFs?-qsAX*^@<9|loTQY#|0%Kh@>_Z-6<-aBRsF#0J%LKz z>~lg|YYOGS#I6-@)M}FR$rx?=@qk@A}p0^fRQ)Ma-%drMF{a$?0rOP~8%MhhO9>$E{G(WOj% zK3}fHw`s^M#mKJ9mrMD=7e!1Ysow6a!k$mBWYY=bAvYAZtJK!tz=BEpJ9L8s-n!I~ zy!~)pT{FcmU}EGJ9u>YO{)G)(0R;j2iYk^%&Mp&HgoYE~PAYF|;R8M=9&s$WGvKlQ zhvq0(Y2l-Bs2hNmVwF&_^_UU~BsLLpj_Yj_Sh}^M9RRcwQz9OWJqrK_Ym#MfQ19uj z6CGaVtiE>t&=G#BEeUq#KuWmJBD0?~QA9wO&&bF0zX@9RO#iH1X1+0y`+k1vE2-37 z2;hF<#h9%$b=J2Mqn0C*6HFGxC1iY=Cj!1NeIu2!Cb$+DzzGzPpq3dKk~KHd>Cc%` z$~R(37@h1%0f)YXz^l*8%Q=S$+bu1+(zHE8cRoqRD8+v}my>S=LBEObfDJ0NqDS|D z9YPuZayCAQ=PQRNEUp?o?DQn3#?EcZT1cMlDC{ZhTL{Ohy2&Eec+xq5|vk zR)RxYxNaWp-!#51=@NnBAqdvDFW)$t8 z)2>*f^2aynB>t9|<=;&%xD@MW@??SM4U@^{wvbEzg=U}X$y=PKZvwY3jH5;dgZRT2 z1C>tx{EBUdatE|^N+l9YUf0@+k$+i>a{@)(u9$HY5uma9i#BntsB~m=@*eO-2Jt7z zn{*MkuHks|4~y=yh1i^6p0r!{`JC@J>rb2CFuDBui|p1NP9hk2MvNl8flDfcCv;3T zi;8=V+On~gl+mwVi zV)|7#7}k1fx_t4V6hj9Uqhd3I2z@Eu!(|@s zi0X7ggGddzYoAf`@@g{khPDs8BvRIfRWHxUzMFT-z~Jn}B8HaCp~w*wT!p06;RXC? ziqvTr|A+fSMp+GX7R9C8NKJ@6;dPH7 z8*_t`<{@D$90R90A3Iq%wJZY*wp^ErWphb43!shcLm_o^GgM_b2O>7As!FF;%hT9D`B zt0ZAu&;6u;P0F=m5n=UI-Rm#SEx3XxWA>;xML&*|xUG0U%)ua88e^L@pWwzWksKBZ z`RqHMpc|o!wGBU_D9*X!4)1<`cTtxr%0&pg*qvbWPZRr2y5vh;(rgwKL0Ec~aua_& zK~^U)Fi?F$q3asspwUUVC~vUeNf==rJLf=4{QwFis5pVcdKzZU26&w`Lw(G0O+4Mf zqYywvhQ_%O7?lDjyM@4xu*_pUbvyGLaw#64Hoax?qX_B|CY*M5MQ~n9Z3?baUX7o; z7z$~t#@0j!eQeoNpdZ1bGSds?jGEU*L_6t~yhda7@nKNlL%hk>%^BsO{2v+spj(6h z>dAydrT#-5&Ghky2yn8r-JI&B7AQz52H{*#k_E9M2JB+FVl{#F6evg+2H_Y|au^eX zO^!Jpv`&YeoXP$M+R8NHL?KvCGp-yh6TUclR-1Ed@UCHxbaCm5>zh<&T1)sC(KAx~ z$|agY4QWb@W$g$2%Fqylhtu0g(TD#Tf0H9?p(B(WB*%n5$I z`90CJlO`gc|&OAS%>#5aboKq`rq>(!}?Whwck6y0dO zH5?VFlX<21jYWv8)AG$W_Hu?V4cDxaPe&XOu%30)2`KAq9J;cHt-jDD$C zI0$|3-A!7}W(>~+_Bv7Lpep3s*#{ekPHRmnZCi!WR&ffEnIiSL!{U5mp*zu?iJyTN9K z)%&ybG(2oPY=99w7*o1Esyj^uulFTa`bTy3AEhM+p;vRagwmbz%>$eEA;laA3DYKt znSIZ}{3?Mib48Ll$p$fp0O3+Gq{ODNa`&%{5Y(bnQcg`V8(r za8uX6q`rn65dbmcnfFDPU~Pj37zW{rG2Ns!OlH_%JQzZKnvz8TW=&M0(5K5*RIX$u zP^&be%T_Zi;h=Vaah?LhKHwSys1+CuESpZ^MwKn53PaHC)tY^JaRZDr?sh+w@X)qq z!0@9meB7h#*5PPIc=BbobSOx?N+p)ww1|UuRF~f8v{&aRowMnf(ambl#khID~ z(K%0QIJ!+D>@nGZId$8zSUPkS#2%tgf>zb1_EOiz2mF5GL@N8)D`*om3vl@}_#!zp zBrJ68+vEfHkN*Tp!==*R@QG3^etS(KgCaZ|sPW7Re8I!_^AmZ;PQtBSs?Z&oMpV!ke(H9=5b zPnxwv;=f0UCuz=9sTUPTkLu*RF{od;?0UrHCTU))nzi~xes2MRYZCd!%f8| zn1rjEvbJk=t;b8n>AYOBR5fHGWVaI`Lk0;&*rdP>S=k^*JU)u;O>6xY3=>6H=Wvt^ zo4)W(JakB2`+{ZwE=Oe>6c~xI_9fZ4uVh_5{tH^@5BZJ6Lg%nZfSngEh!6k=Y8zu} zu);OiBc#}51r(6>i5c3XxtpHg88%|u75$GhB~QiFkc?~K^fja8eML}?O@h-=BWxN# zVb(@%5AspLC_*^<`-aZ!0>+pPNGATE7!=iaMRu~dOfxYL%KJgc`3xKo9^pM;@UKwjA1DE>$g1r*0Jz=aDZ7cvKB{0L0+)4Ofu!#I~ zQ3A_tD2RdY56K@B0+&KY%|u71`A~p{42DY}!z`4k4A4cHN_jTa9e7{096E3&_G?TT z%hw6*Kh#s&*TwLd>}24PlJq?e26!SXX_B)c061+jMd;uPV_=YEah{p#%i8w5zzW4n z2u?5w#+%7!BzKh#5{>KioS44D0$=6G_n zP~48@O_TYuIAMxW$jpRENUCHJ$zZvzZ7zOZRQd0Ar%Bvf9Q>r9$PuJiX#=E~cMyzn z%W<9?Q-RIy&%EOhoGl{PO>}ZRR_k5+r1ihDsq?zlJ)VBS| zv8vtIj{A*-Inq{8&k%$$^77&owwp{av(I)vo$ae|SdW85*!Ts7nPs*Ev{~QR3}FgY zPO}x|uannPF>c#PtoQ6pEZ^-WEgpaTYxD{LurazChXLFAzHU#RnE7QCIt+q_Y=iUjo=)9~Pj6D3R{KOp=M9}&anfK3&~Ft=V z{peZj;&2iJDN2P&1=w#Hxpk&)S<@o_%+txZpFsE~Om{{7E4O|xi{BqC}0MXDd7JUau74B9#t4FcBeyojqy==4d*Vu4i1Q1Vi~ zVuq5$5d%$BjV&-LC)Oe?t?NE!wqD5p`y-`%Eex@U|0Rn$8ep_fWJNmc9?#Dkv;Qq4 zpADiYk!S{u7_`{U#Iz+O3^}P1Nwe2Jmx*?Zrh(M{_Y`5O;A*3HfC#U^XobJcO4wm8 zn+)6MZm|EBmIq*_5dSR<%?%Sm*dDW7hM>PMC4R5fabEkDh&kBVpq1+H`I^U6^3bd) zsGR_=f*c}dT`ndKJ0FbdM?FWoPF_-9!PfTWwEY5cUKSlC2o+L92o^qJ2i7p15P5)= zIH{>*p-u_qr6^tudUOmWns`YRZXAgUjGRg!s5k`ZB?Pzq*u@h~TN=_cN*aex-GHsa zpHgrifT|QUK*}YvVhWRTtr*#?MofM$j_;V$qD!6Fa<3HB2q7Z`7ZtZdrA)$@phP2JxQ}u75`Cgy61HKx2b?XbT+$!;_iCgp7Ix65Ez%M~{}N{|WW+>6ak4@Y_8~*1tq0ZLiC==y zSgq{7MhH=M1Rx`e^q_tb zftN>sy#c+D*7!Q1BIs@m$C1V?*+{IJ2>&#MJdl40UlIO_cyp0Zy}wEPLIMEUKx7`2 zBMlby;W>zp?*%RQdJo&ebeRvKOU;^XTaAoKD5jXqPzeOt2&AAAP^f^bx)ef_Y_ls`u2caf^y0)MM*nt1 z>W$KmfEvT+ogf!dvWlT7`XH$#o84qmQt;r?u7Ocyt7$4O?bU3a6Q}^95h}JRbm5>J z2&pe{R&hH-h9h`F1Ds?$sUjk~zew>lq+GrW2POSOMgF1gAe5I(TMdM65wi%zC0JMn zL8AN{=gx-L4=QeP1%y1tb2PezCMdPNN*-kdACXor#eW*LzJ4zQqqcM1_0*U+@6Lxg*H%GQ@Em2fH> z36+s-qr*Wt)*w+3g_sPkkJt_P?|ULU^>B|EI1u}v0%?W+6u3}<6vU}IMA(Plzw`i# zm^4rSZlaR}%hiH|4lQh<;SWI(c*Fm`H@hPMF^+M1t?{3Ea#zp`0vR&f3mJ&a8wEz( zgM#|X5iy`Z;YKP2`p$&9;19Yp3+j6FvjY zWJ+ulf94+jG#g>*q$v?6g7{~UI0+7p*o_K8M8P@G>JU*GdxGtaM9l~w7bMLj-PS)j zP0>&cklXvDAo=G}8gh}6g6b@Zh`V1YKu&3h2L)O33fMP_8~db7X1SvJqQpdJcIosu zXplE5CnAfMAPvv|AG(DiSHxWT9~xi%KU5nnN&37CI@yC$`Vo8ir&Wy2{NXxTXLiXt zNpnA$p8PG+-M1r6aX~u8RYGD|$iga5jMN+)a)3OpBc5O}b@tB+vkEh0s3IYBF_yr{ z0XrjP@qnV)VM_a0@N<5a{?%3NGf>W z3uy>(Vf!G^?e8glT|~7=7D2+NOhqS|MFnq#A{Iy$AsU1gx*0?i^+mj}?$u(ioS>QD zFUIhpl-ZwlfF|!u5P&K8i%BxzpC_g!1>^m+bwsa_cQAses@r6Z!s->L(=+Cv8V4G+ z9HclZ4SpZFnX!X*B|zQno{p7Iwc$Y?{qyhR579!s8%IytUrH27gB3wIaSq%p zWoQ<3zQ?1q7hS%^Mm|V-j?yxJ4;rK$G@cQyNb#Hl<{$j3nSW)mQS~2wf%uLnaC4R% zf9KtlF;yG78p_A{;^FJ~~E@j-qO5 zN*Q6JuiuL{~*6N#bXh%ON&s%%M_TD8y;SYIjloMc-{Bl z`64E~-e8L^Bj?Ea!O`IZ7)y>=yWo1+1w;6J?T10)cw+A8?8&F^ZNYyh+Go=9yMRXK z@Ts40j&XpnTN(Q{_h)HA7f%?#?+Lc`jglpCypEq;(%eXIB@`w3hb^5klc0VN@)YL` z2|y>)>htN)ej-1AoJSz~qbqmZTEp+{d9dpBW`jE!f{VqUN>JEuXD$ATFDs_Z32$;@ zH0E8`Zz{Y=#xHR!`5|q2f`KxUFEaD_W$aB`J=O;`R~)oC4EL2Y;seh&J(yh3#%xkB zYBG{FH*@&S>c+b_!ml^>c5E{ybCT1{L*@z6pWC;277f6iW4PrE1%5 zyz#1w-&x}+b^!vH2e?ZB;w)H)YCvrrwu_qt^obAUskvnmUZ~3pc#~$>bdCB9j869% zxWbGQex;~u#U*4VLc_+omaE%r_!(bSt#8`j#^MSE{j=q?LT8mtqxzp!3x8^uMnTHrqDytYp+S+ zipsvoLe9$wz}s44&b+}c5v8?uC|jQ16Gb4^z)L-j_^XVH!-JD^zsU#cTQ!O>enoMA ztissy_oNfDf?|fqX|yVo(6~yGavWzHI2R_Eqdq(`PFI$2;D!O>_L0 z8HQOno?VAQcTW1hm2YE&tPL6cnGd9FgrwcKTx>fZ4q{^dxjNr z%p|{r|HAh(Ilyy&Vm7WJz|$GW>;nVwYGiWz49jT(1|Q3*%}kmp z#zs1R?PP2yOm~;lbRIR|nm{8d{?g-doVc{}JjEwPYph}6$McgbI* zw^k1hP6JBMTR}_Gte8Q^^`UTyoT!k-tu!Fu-QahQp3hz4PmsX}r)!4I(0gdbc#uY@ z^i#0@_w00J;GhiK852eM7Y#Xj(G^i`7*eF|I+I5MZX}RPgM7_E&rKlJ0UJ9>RPLfs z3MsN_YfvkE(a9gH5USeOdj^RvII$l_#Q@M-MFbC~;2WV5;XpKw5-}eaFNPjU9*10t z&x=-y&lyaGan1$-;)oHi{y`0uPLI8= zq;dyeg;n%RcdU*3>@EyHfzdi0!fiUa*#2n9K~rU=948%Rap&&In5lMU>PTC1a(do} z=5ga!%7Ab_H;Oy?duGWHR}hoe<#H8Nz2VdeQ3C9%^2-_Y=4G~$OWT%IJ8wV|I@`fgT-c9`IX(a`sr~eHHjD5!y^N;J9=uaZ#VEs(N7C#!woSwAu z|M}FdFV?#)@%(1@OU?|JmBs1n;a?(DR(%pi+{4|4I5bu1$3C4fbT=T6`b1g>xq|19XlUsP_P7!bB z4f%Q?6da&s&}{{Aoqfd3ceAAlg3so7)$K;BH5G5n?d7&p6MAOMEqd23=z_-LsviFz zx0bmDff?x1kPBom;=O8_NII8lSy-n3#o`gy!aBJ5WnB%0-iX&G{GuM{Y=c$R0wjJO za{I7N$3OMu4ed9l3qH2$tVX$7S5x->e_HxUOYY{l5${Zk>~!OkX3IkQ__+1V6v%Sz zO$}e#etVXvRa^(E<+nT6y?HuW>ysA&0-(Nx;8=>cXTJBe)S=FZq6xERunr(7 z9EJ*FzE1EH<-NT*^k%k#F!ysi@$Iz3|KqrK|6sQY^YJ1x`0^X{?Yy4fRrJa6`g6L^k92kejx0Z*rO7k@GyIFg`BWhX5ssdIzF;2j#7+DM z4no+~@zHPMVtf6t$0i^2yK=)t1HFPHUVAL-U$k1G#bdKyL*=}l^`}k1@f8o2xR;vR zGQ(4Y6T}2Jx-B04>GU&Riho!5Cz{nLuMxN&dmAFPb1{GKOM z8o6L3t?zX>g7vq%x>fOv}?_Dv%)Wr)t+3o(a zACXA+dK>TdoF1JMj8r!}qV+m62U_o7aqCc)rEkGbKEcM|p_x0uGV(FwpFP;@^lPoy|e4Ih+D?sC9S((lRQ1|H1pSG#;~6RUdY>eluc(IQZvR_%DeOiO2`v}Zq# zna{sQdfN&J{ezJPdV;RNFQ}LS30MKW^dG=XLEzoY)ZrfvFQ3h$yT<2bi)Ym9YZCg* z!+osoaIAGE7H759?_K+<6=HioGh+AiI!6g58}phR@yoV+wnhbuJ97~aOxx?*-b^`| z2Dx+WpXK5J8Gt7j`((Z@zNs!x!S*D8a@i_|qzbN zY~%A&KBA(;9B7;yGxZ32W(wi<*tB4(ZfCBrJXhCS>*fxqLp*OH-p^y^;AY!H0j=rp z@Atvp<&1*Nf~x!iLB5`E#EdOVo#+!9xBGZGGaWTp?|apqeiCW&t)2W`d66wwd^5G* ztS*jL)wFe!Js*tTTvJ^Id&UEuD|;6-uzK#=V6dXjt} zNx3V#j*k86H0}Z9>wl=NUZ__Zx9%pE9s% z^?hm+T&1wrA+4YLdwW!w<|B8DZxHvxDgx_dWBY`;!|&3tyWQ5G z6TtS01nIXQ(tNDR9_7PidK{6xWBN4FwY$49G~Imr-XCcWtq$b5&u?$LFVSzQ+5R^# ztr*)HmwbF7+@YBE6xmWTVds8X!Cq!P*X^8{viZ+r6pz(sW8M3HbHH*x)z0?Wc~7fP zvOzanv(T$&$<`yeBuZb?-wv6ullf}Ahjpz!Re*;ZIM3Vx!n9OI&IJTdWnWw$f~}h0 z8K1#pxnu9Mk3zNKs+r`+6988ZuPz6#Oqfw!mTq9T?vw_<_k*~jt&oh9R)^ihvd&m^Sxm)-ob0R zyMAoHi{TtW{5vBKK7v%>7k<1M3EVr^MmMnLbu~g>{%`!Ou1Q=oi`K?E$(b)JSl$P{ zb32Z;*k8?qmu&GSLvuc7{*3s0AI)6+e7pGR8DvH>RZjZ6^=7z$<#?>-2LaBid7;OK z>z6P7$&s>c|FY>bl6dg3?2^-qEHA>x1lZRXy`xad*#I8=8C!bXE!gbnye@wBk(;ci ze__3w$lzY;8R^?6>h@}z8d&?rrdO}dc9swM+uG;(Hue_$WA~7E{Y}ZGL3?AY;WO>? z)b%4w$+4y_H)Wds=r`{EALX8~Nl8vK#|*z#9`-xVXBJ{c0&eO9LVD|T>RP+NMjx!o zYl6k5t6*!m9U2ktqgkHhoITdvChDL|97v4I4bX<9YIztkzd-LTk;x%Zh_14*Qe?t`bW?TBwUc}DUnR`C_`TA3T z4rWEwV4ad)=JRkCgBo2tSSIWTdGaLE2d|Fr)}5dIM=InP@Y&+09!sW=DR>NY+C4M-$i68^Egf^G*idH-ZUDH4hxgbx+T~ zYo;|TcNf@qW0!MgZ}hFpTe@HUINk5wg$Z61GxzQMd;r~ky6CsO$l2cKNA7HSDcdZ6 zhJW!lxObR8qF^DjK4xdHPoJ-I9z81QZr(V$rtiG9#PyEnZ;7+Sxtkra*;7qUO2V2y zceUkYm{~sxre@kw{814w_sKfyU39;@I67qM*4|*~&DhHlT*R zBIJOfgQ{{4EG?)~UTJYzk4uaK+m>#MM304zc6W&c9{)HwzXsNwV zMM)DrX?W<&n_4s0pFcfl=x>Xm*`kOxy!tC`6M8**#fsubK&gQ%Lp?frTF~8r_%XJ1 zziM{(vN1iEqV8Rt&xS~c&NMTpfjRqZBhCUppmTT@6x-o6e zakMr#b5n&jO~qyiof$~hS`eG{@7-4YUGL=Po#B+%UH2(j&AxSpM34tegoEpYw|SgI zk`wvPJ@!H_zRnjey1t_Q!(IO2VgGQ8lQHsw3xBwZM{?2Wd<4jc1|f{}uMLn6?Mgec zU$7AP8z6@mJ3i6_y!5Hj-SN}E{1aCIiCc#Z=z+J1oSc#u?EAu1JRb^A>p%?!0%^4S zhlT0Dbg|K&Q2R`U6@i#3-BNFLVC=nARL+~BM1()XfaoS}&TzJ1Ozs*sZyWhnb_Nn8 z{!?S8KQkgZ-8nOkm%+2@e4d0E55pN}wugVAm^B4!xor8)eK=j@Ad|ebmm!_MhTU^um zP-PlE1Fg5y>gl-~aJ6SbVe^V>2{hcwpxhHH^*>eStI{s?yt?BLi;WAL^i~_*t3RIA zY=FnO_-+uiIJiG$w|KpSR{@;7HskcQZv3~pz!q2T2wB0??dRrs=h9Wr^S2-VlVzUO zUpJm zyyy_*yG7F$8vDe*D>5LgHl&4SZPU{_>tV9EEg5U=)N1WAAm|>ryqFeVy3fw{(6L#0 z^QVQDZAnj9YVDAeP^N_*cua_NGfK+nriUIZP7DlR#CBCnkZZ)+^6TjiHt%CJSCf~f z&ok7y=K2F>io(-DZQ~7Zr`Inppphi$i~%Qg4R|xJ?cB)!^QYgi56hh-&0L)GGOoIS zbm-4l9DuA~GFJVMbk%t3)f|A~Iy%A1Nc3zhPkM-u$1Q zjEQaJtr(n?@rRn`79_dX3<`0^2u`tqFJdE}7Qbn>G3Ml0vtrpx0o=Way$?}7`ZO;+ z)1po4tHTjyS#vf;)KjDN>cUNSr`r43Evri=SU8;*nie*xW3dKhSd%z8MKLyR;d)}m zIpJnel=OdaZ`0?n>Az@qo_j*FO$C^pz_GEwUP|?>@*~WCR}4?1=a4bkJ#PWCsA0iG zon?c_`awNAx~7?_WFC*OzGh$>5``^IxoEitSj86@j&3@DUi3VR+(;_)j&`b_ij7Gv zh)?t;e2P0{*toX)cCG7dG$4j6Vz4WgW;+}>92~LnXX^gNHXQQ##qD)*!pD~!UzQLl zpF72^1PN4Xmq?Nyo5UemHPNfuxV-w!xN3#F^sqgDb(WMKlah5C%`Wz1EW=fyRDC_9 zv`Xd5K{-)Hb-XT&ioHE<`p2RbZVa2+8SsHf#cF(D>1!zk?-X~cw&QV3txWgeqlk=r zWCKpo;sF|Wxk}Xq?40Z0;xH36FbhcwZU&cS>{65-e(et3EY$bt6yy*6`kZP_g|xKf z(!-LR@T_);2Md?R#Pj(V;Zu|Y-fkS#{2u2D|-pzS9T54E;XiYdr4UXS06jF$Bi) z$(9Fk0A0D)SHFn(of8Iq2AefyI*`LAQ7yt>`0n;DaG|mN=%A~GSN(mNH7sp z3|jwQ!iySri;jXi`n?1TbySasXK@MJATK+k|f z?%Z0IU(87HsL(8e|2dRqrdG`)k(h|0LcSR+f2KE0sGo;Sp~m_QYYOIzA+IWA;%u^iIf{s4TJ8jF_Z|cO zW@t|fo)Fpn!MJ7EY%1Y#!zI?7XO z2mgq=5O=U`5W3I-M+Lsir}smh7PWMrco8Y43$MS4i(wVX4)RNn%!ks+F1Y~o5*Vy> zvTDO`T>vdpi(>zY!n%S)Db;+ZYy5>fsQgy5R?`Kcm+iAtqpn3fEZg4Cq`DG#tzq;i z(g&k&-P`03Zdfd{1Ec=~Pyc9kB~VtwGj_|?#{TeQ*H!d#tad*gX0|CcNUT3(Q-P)u%p#u^sFj*$;Og`t4vuQbuE-=*-b?G>qnr#t#Ea2 z8nt`{OFQ*5nji4(06w{w$u3?%f^*msHVmJF861MUn+*~I(7^x){nJ(lGR*f42no$% zX$cE`YXn67YQL5tZT(%HMt*lC7W&wUObaw`Tv8Pnb zdfGL7h$(<9PP4gxYb}~lS2|RwGI8@-A|-Sfj5a0xk%5je_;&uiSi2+n!PfJZ3O_-| zAz_W zP99S56UJP=@Dws}yOSQLaA-K|i4Rd<8p+29L((oyQ>1=C<&lX%6&h9Z(~Q=Txj&vs zYZbOxOTQv$SToV1ut(saUHlMsD?2lGFfR|=YoQoS5JWd`p874eQg9Jn`>;YZsQTB$ z*vZa@%FZWiRFQ+i_KyFbSlHzvk>q0B*su0h))d8s#ngNom-V9<5*uU=xr+wjJnGQ; zLa#lF%?9Og_wk#aqB1T#TvhMW;p=zzmez>xElXvkNBub|SvM-l?JeJGfAFv6^QlUD zmOY#r9%NVr8b^H*)}!_X6IS>kBb0Fr6|%-bz1rlt8Ag6n(N88`K;)g$Q$Xar*q^r6 zAzj$PF5@n0`)!)a2+FhQ-V}SAc#ag;ayr^J>gjzpiX7yJD-c;%9U?MyWE$5zRWo!MQyFNJz=$SlXDmC9)*~8u2CWM@#am{fU**6GCK*)(g4u<8 zWcQA)(EKN0m|Wbk=Vjzlt6&+FUnNx!nQ|g4T;4a z4-$!lRcRfGgk5QJiHZx)dDogB(B-OM*3g_FY%1Yo0(P_Djx2Q zm-s=IC-+!Ev?Q&_MZU_6Ag46ai!qp6YGwa8Ml@N2$nnOagyk zUcW3Wp|`f(mq2wbb5tR~aXo7+qB`7`OKlQ3lz?Oe=ib8t@C8zA;qp4~c6{L92zoA{ zocR{HjzmoR{DZWV2QkmfA2Ypt32j{hngz+Tl6qx_CR@_%1E~?2rgA)bQ@cNT=U;!V zsVe}KO#)WMut~n2ng;T4hmFdq zg%%2*>cz2OtGtD^2A=mI*V)TgYYh?aCFm*CsyavT8K&alq!Glbhl*Bi6L6;VqoVuH^$GH$|5g4Oz0^$V zlA$%(P@+YpiDE(0T#m1B)x@X(wKj)r9@!Gd{SJPyPv@uH<5?J z9!2?cq6^B~eko&$2VW-~H_kw`-m>fS!eb?2#1_WpyQZ)zzjN+tdboVAtvW~0xBjx5 zqn|hiqR>Cjnq5gT42r$HIud=nI_`kDhTEH=mp}@G2U|!r1XNo_Oj-*eTO8yvI-XbUK=`CpYtKFv_x)W6K+4E%Ahuqax1 zYM7k}(=D4iE~z5#c#&~oDsf-S&#>5{eJaE{lph+*+gxYTHfl9A;QheyI-@~sI2M3E zF~ahDh{y0TYh7T!2{vSEM8h@N% zZ(+3ZeKEQ^+@lzrebzAn2X;?o%QQ~OGgGA*tA?=gmrUN! zpCRpxF6OXEz-cS8_k|Ou)2rnBbm`AyZ+CiseyFReNvaApcEi@_aJM~xTp_wjh4WEB90-4hR$0WFmaDxK2#bh(xvlhki*GoiW zL?lu%E7`*3@aoY&j+<6Kpc@?#dN1l{E;*60Qcwym23%Kyv9^Pi3Jnw-gX{h)Z zo>@;?I&G2TDjmNooy}iQc+32jM@IL56Bspmn}bKoN}ZJK-O$w($-9|ZeH+?6%5J+m zfon^_e179bXp*X0U*kEE(b0jy;%u>d(JG`ruHCOW>icl09ngZTWyJhyCPYaW0ocV1 z{TCa_@=o9OU&$sPh_-+t#>k@OXNW?- z>O@8;z=X_i3w;~HA;byJXCud5K9`F;1eImX5BEqT8%<4RP77*wkdTnyc2}W3hj?M0 zWvqGkcWem6jNOU$O_Oq&Nw*WR>1s}C zS~lz(qW8&qL{UKJ_+H2#AM1=Bomh2UV1V*1C!;2fYjm&Fe~d9Za5|!@%6V4pWY~@F ztJL0s;MX)8!2yi18a0U@p#_E*3|ssUV#10e;(AzEMa;6lj+Qth8ELGCXn*zchBz8^ z1oSByV%Ta>ZqH=OC2E%~DA7EmxbOE_Pb|&?&->S!|-%zB9Xs&agxHOnN44C8~Sm66jlwN5q z-l!Eel)VRFRnJ`j{iW&U3$Ob2J@HRF!;fuJ!EgTkzPhb?NlP%rKaU3ruEM8=S7Lqe z#EA;Co*6E!qID4}4)yAzd~}sGcaAvjbNVsZ=J;Mg z0+-|;nA6Mti~aHWyxco6Otji5tfhGTj>lERYUhTK(%thm{W^Wsizn{os#@azMf^ox z+oXyWEXBdn@nT{y2x$ucu|?f(HV+Q&oO-f7yV#t#m;`<9_xG}7TyDQi>VlABqIo zjWlFC@}8*ky7ui$S{@4xW{Xr0e(i_)bTP(NJinB>!d*k94eRy?Onr!W-16IY(Ry8I z!x)?}1;c7n=C%=*o2>}JY{@6EXamRG$MJNfgE~8OO2XEA+YO~w(Vn_NNlBC_+LmI7 z!7_sva)9Qr606NGR=h#li9Dg|A2+>$v#(?kk_3LVoj4>ow{P0im9NFFo_B~nqszXm**M+k_k6;6!kxSVwj)lcB z7DfGpvE=?crZYuC%{31E-_w;zj{xrws5B3_JxwlzAzSuE&26S6N%j+cbwXWiF4Z>k zNgRpc6z)EpQg$b0zBNo?GTSe{B)sH*pq>!Vt(ktV1P71xwyr#O$434UIrbLQYh4Kb zWY()7dDY`?1Agsp9T5mHno{;@LOUO2%gJv?Vxkm>oATkQ3$~Xovv0te4&WL*W_<1x z=t_~9NwB05<0bd`M!w>a~xN$4GnIJLBh{F%x!_ph<2KWZX6&>+sQ8 z;R6n8LTkkKgBl%$O$9;M{WTOjEXNIBTV5Zp=GvIJKZ8Cvo!e=WW%|(aWo)4sRvd{E zDmS~0lIeJ|y-Xp=#aMnx8KbFDcI^@EMq)^0G;>>CuvKZWgq_PN&b7adO~h(T#n#phT*jmw1YPNjHbMkO>Fk*XwsO zm+Adulu+-(D9<+=)m0h|r~c4u!eiQ3)24u#)gME8k>7|x`>o@DMVM<1MHeRYe@y6A zUr)USkaty@AJG2VzuUIX6W6%73p%4u)diQOguxO_QAw~6xSKYX(Jr#z^v2~R7gIzG z1`urI!enH@DJ}_j!GUe942^AtXw-%Xr+iIm>euz{_`Lr;_1jb_bO1k(4vJ>cOYa;syT>(P+pYE$F2ilzRdN0tpAxUu;%~q_fzkPakuIVaV3(1 zX_PGGDj9`a?=ZfRdN$uOxRQrDbWq_==K(@QvK?m9p)bsf7WyP`Iy2&GyMWM~p)Cw& zP^DkG1bXno-XA5uzbe!i-al$L^viH^p9i7`}%&QoXsWv>h&es}jeh$H>oTy5B3o)~{{!0Xjt2O{3yEwOrg8 z9bel84+KA7IPZ>}(c{P%%U247r(B+w+;z3esYag+|0N)X5D-W|;1p9}>qc_7-!k?u zc-YfY^6j{R{WJc@Ph(=DE=&N^r4G2f(*;xGUan(L35*9Zd^T+Az!S+Szqw?By&Bj2Qo=%X~=y~IW`|rdmbw@s#A#5g-utpyy`KU zgSTx%c?EnwsX8uL@j<(AGmk@sv&F%q*w2 zZKxV-?b(YU;AD+NSt3mfZ2y~=W@C4g%G?lI0M-_pTNl;LLYNl+5d#6$ zp&}lG!Q0_%*bRXZzeEFpt8{ffmr$#=qa1%+xVNUFCZC5V zk`HWVZ4Hz4koAnzX$ftA4Q9iht3pKE9TBSAt>a5rIh_s4-7Q(`{~ttc)tY8EZ2rmx30vdX?ubaajAHRNyk zo4w7q8;g2luPL_d8?IH$F*v^{0nZx~n7fL7tkjgL!{5 zgM8BiLB$eU`W%bT<4)`w_OH+_zeJ47$lK2P!tC-H%SG3ck#TqaR-whxXZ!80P zJPEM%Ra|;Cp9duy*nvDeM|$`E`>=4B{Pyj37O0cz$u=Xh{4S688~nL`x#DRhb3lqL z2%GNt=K}4?hQMoGDx(U=qMtt1(P4W?-123=hR(rOCT28I6#!-3dl>Dq@@O_Uz6j zVD#vH?J=W=-|9en1TU^m$KiXgsim27cM4;^)a4^b9eBfgIc8R~4e!{LpDe7sL;b$| zsRKtDxh%8FYehKd^eKTI~lE?@9s72cQ{B#(ZcvsiUdQIQ$&F1Yht*GC0&IK8*$) zke=^3>F$sNDLb*aiu%u#yz(?NH6?ZV?YP#}L4M9cIE{JM&_M*ZPa5oo%nSq>WqT;2 zwjFrB|(;6qiH zg-A}p2d}L^%OeLy2l1J^s93ZGO~G|Nw;yIa z*N~oq$nLe!jo~iqD{mEudi?ggOthy<{?uFG2gX<*=-5pa+2Us#An2H+HHf8oB7D2_ zK|9BUpz4?*$6|y?MD_Lq+NI$iVH&SYDP~Y=xQa_|sI5WPZ1oJQj<#a-55g2ep$^y! zGcx_K^X2R;xl%mwcG3{-Vir1Hb}9bp*l)j#xanWZhiyf98~d0pU)l6w`sh!&v6CCv zTchEzZkNkwV*j4tqWZAC#}eVrWiev!LEk~W=t6Q;cQc(=vdC4xSH6GetC%lsX&p3oj-&D31Xm^^31hyV$mJDY4se>vp86p>B3J7& zU;M;62kXf5j;%Xnhef)$LkPPBZEcEPaV_?0C?JQ%;oI1*7vByAGi#y_(_GV*ZnlJp z%SDm}<${rgBZ$f=rC1rL=lux+Cw|~YOEhzcS&L$Y2EPvvfAv*Tp!&qnCmbO9do+uy zQs>Gn8kt_AA^)2%I$Gd8nepYMeRM(4AW<^4ljyeDudM>saL=QPJNQGe?hm%yV~4pe z1KRmnG*mDe9|tKW-f*P*+tp$iLDIc>>90RwFZ-^+WY0eiU&w$vzXLWlv=c`B!W-#d zw%8x|p|$%QYzV#HxDJkv?3|h!g_yR%`;d>1sfE`H^XWaesC=$xqkH&>cl}=xyeGV2 zy(zt<3{4nH=Pf1c#7kq{^lHk+xik4}+`qc8CSq99)X>?{IiA6%E+4h}=ZhOF6MR1T zx^x<>KC&q9OLNRj=>|Y}+|;uMq+^+NoVT?+r(bR2+WuE^pSRAM0lG3)wD|;t2@u)r?HDRBbctku> zlSlRLanS17MW;DW?GR@~Sws6-GcSx{>mILoM*nuaw$L~6-6gh=Pcf%ZVGG$L-QWB8 zSiU3tD4zn?N{m&$Z*f)7Vg9RR^jExVnST@?lc0YIqnhtOCocg-+b|Jvw-8?PvgRJA zG8O(G=r8OYFT+o$*Kc`$VSmMUUV08-Qhu6p*XU>a8^aU3>{)tu*7GYdNQG+a)hp5I z$>7i8B5&Zb6|a}Pol*(T%ZBYW4MJ>X5T|>tf#2~&;T3BC0W-Kg`|5_bpiT4VFW9SG z_3`ECy6`=o$8@6~WTT#rXm*Sz8=7!!i35`O**~Eg@4QYgbg@Jc?icdE75so!ddu!X z`E}++f(-S|uam_br-2Dk_&&Lz8^@1WuMw1uWi_1LU3^tc4NClY#1E+JL5WuHS-0mG z><#IuQ?Rijq!gRioW?@EBYO@B^`(V+qN`A+QAa}kSg8A>$(e+@mOMG96KbdkIibdy z(^#l?To6q{ePN+i`OfO&ggT8n66(i7JsgkEB-FKB3w=7FhKi6AYOFbZg?g8no`m|+ zLTzX&)LGP#P;UwKXmlQ-uE$+}LZKdwB!wDkPGh0o;Z7l;zO+!Ax(am~btKeVLOnQ_ zP}gGQIH6DvWQBS#YJ!2IL}-O@S@yTIqEI#Km9Hve#uW3Mf^0*~ z_ltN#wK9oI-f0zk?-9zz1qu3~g+DIQwj}6NY1>w`LNDCXH8GYUlY~|SP?R%x+u9O- z&Lju>(qBcExm=PtLQ0%JCY70*V?ix=X6~%j$QH44$Z3{cOz~8m>STv7N{$Ep-pK`7aoYn_ft(~`ossQUIXG0!dMZIwLy+i89$3juK@uFjR`NdGhsS)vXIq z1(Kp{nk}&|NJte~7EY~WWL7xHtWcwtkxk)51Isra{32%X2bxGQdxuE!@hqz^9A}Eb z=W}u`>^IK5pNCVlmsxlAGQGrxyb4mPToIS#P}rrzWlOIW6{(ReVc*eXw?(qsN_T>q zvOCd5cFBn#V+^Z9cNKRZG}sPyYW;1c5kX1)ZI=CDm;RFJpwbC;s)4WKPJ@A@(O}S` zA#giw2Wm>Np@{^O#h`i!JQ$xwgH<;cz=KN;g9qalsckOtYVp{wTRpYyRVHH_Dd9bpNiG$&WK@O z^3DjTHcFI0GED}BI%S&N3YCil#`qjEt+;ssYC&?vz!+2N1d?g;G1My4WNN6~Q5$v7 zA=8Rm9L$lV#{sL2k|~f(lh>g}nI_}I4&%c)WLkX}gse;tx|CjlWSYznwaPR(BPwP3 zENj!M+ayfcwQ2LxYtv^lIvgloWPdn=vXK|!ATLC{>PE(h6AeE94&8f}^e+aVy9#Z1 zOcwAE45JMb*r%lnF?Paa4Ezf;8<#o{OKTR|E60LvWx4Qyp|#|LPBkuWZDkKBdG&Hl@U*OD~@q_REsDV1+T%`v6l>zr{g>tN#2fwydAX*C3!qfB$02gy~WqSn?mPf zb7*@zwU3cuqoJ48HOs6~8h>QX*nMuVEE7g$N7m+Y5Xt-(Mpr|bZKbh7O_`nOA+uzm zsFK;i@FX&;c=n2coZ}5^Nk5rwr2#_CtCzca$SheMDo3-u(Mfbx_1qJ^k>q%`H_}gO z+v!zMQ)vw?l$Klx)k-@UpF(L>&l~}dEvdAFv4%?9N>72BN^9z&wB#PBR$AjEN~?I@ z2Y75*r8V?X+8Xx$pOh=w_dkV}(Juc%yZlG$8SUwx%Ju}0=;H`sJ%^w0AcF$TFv$jV zqvHH%;vw0(hB(J%7YKX&qN`BSHJ08cqO$&{;^@xwkiy&YZG-W z0d#qDkcc`7c8OqrzIR~`1z{KZ_Hv1FmWV-g#o_=;4^PD&ycojp-o-}6Fojn~SkM{& zVcz%ymIMa;+>R5acO7Ul2R|8#!@hbpf78cT|MQkb>j+}~Dxb?e?!Re(v)eWM<9B0} z_P9<7W1iMBUddaJz)K=Zh!uAJmE$Zep@Tso>J!_}i%@mb*uYjTtSyyN4=T*}+|WuXP|dPIy-U+sxvgMU{e8Qs zmaT#OnXNt7ci0b53C?QDy^_V{k4n%DEDNHY3!gpZ$+V#D+yz#nweY4Ozd%3p&=V9- zV({0j@k%5^6zn=HSgIjc44weQBMz7cV({uUe>MC}$Wb0donjrcXTQFE_2&Is7Or5x zXMP_QBFFuP9R=RTcd)Ob^v{AI#upmB7wjkg z;2rL^K-1M$5zP+14ZCv$1IE!hF^XQ2BNR>GM-?uWxYf4T*bY z+cLG&|I>Z?ppT8v?7Rgu)iD|F`mS%IdtgWG2 z+`91tg6)srd!0eny|4m`GYjYhDBM@fantw053z)I?!pbR_pkYz9GCCx zn1!zUn=3`GY*BL8Hry7t9CykWcOy$|cUe%Pe0lR=j%jj?AnheRlKpDge-*Ll9{MM) znTc)xQOCGq-i%>nGmLEZ;EDym(4E700u}K$Q0*0jw4biCT`|#YEP^nECttv0=P~`j z8E}h>W&KVs3qXt~H!PFrciDG~?~F>nw3f}Mef$tvB-Osu){|aqD^%d6_ME1x`N|jg zG`rCA3L5@j2%bwIn7ebR!$5 z6Gay=zDDc~{rd4|_`%HQ1$@E8Ts75ZUSnx9lRRHyH>S|tFbZ*i%i|f9b@r%*(;;=* zFU9r;?@avfI5HkA;UnWKkk;f=q4V+)^G3t-F+KbkHxPY)pU@FoHK7;5_rpiRps+XD zWkI;fpDTZiq^Fh$3opMYdoAe~E-L&`HPs{Syur&x+Igj(cdhAZf85XTdi(lse_`)F zzhcjB-hO`doc;LbW8%d+&hY!B6>eW4hbZ=WsrK=^t-e?uK9n_eW1HUy=L8@oT$J;j zz~c9Az~IhVT0T6iZ{=06g@aAIrrcW6@CAG0qDLn)@w{GvVuK*_-tsxL9hdKIdUJqc zTyXbv_y^7HFVNQ(0nGfD5F%d8;9I=a{2B)1^MctrT{jsfHo$alA6^cchko8nvHB_c$i4Y|GZF2&D}XscV` z=l|zl|Bgx`wrnY9S262yOy6Y%vts-POBBu8wEb40WArpS#Ik)EHK5w2ZebdL4r-{* zJtqqrHLr+4?cR%G;r521rNCUUFeZ~Ogm&uhI%4{8z-)o@GV^yq{VaXl=b3)7a{ZGa z%-p_%MqD|_8b5%kb82nj-;-HMQs3x4a|&n%cjgIe$gYV+IZ0$@5`(7}`$hFItRA*n zibE#MUg9}J?ma6sFDOmC3Lm#^-BLafi~3`d^3v5E-s||1Is8qSl6iB3t}>fF+y}VY zUmlsss?9z0T;ab8a>td?>F1IzqTIEo_k5uqXAZbf<0M}=EeRd0kACqp#8Oc4t?)^> z;}vCLR;K?ym8l6%kVMyonB07}LK!6JAg^FY%ryo^&A@=R+BeB?p+yK6=f_3T33TAa z+sD0^AVH8INGJ=5{@#%2Ujh=xY$%9V#dZikEe>VH>rjtw`Kl%;>e63DVuqS+7o}fl zeTB-v(~j7^AN9z+wv%}^Sg8Sq*Kc7dq=FcQbt|Z{J3PIzh7RdX#R0QKEyS}iFJi!e zeE^Zv38p6rgJAvy96&qM9tfQV6=osk`_GHmxC!93-{Q2K_8o59!TdcfB%UyNiAznm zp!U^su(za#3Gf`N>vR~dJZvlwp#;p`dBDQp0saLqbTqidbi4WkfcfU=2 zfNE4+Aorr=6Vo%Mv#MQ2D8D#M>T*NNeHg6n?(sYxuk<7$I_zT@+}d2xG0RF?Tb?kd zL^h^l4<49smY@?W|0UO6h`>0tZQZ$CZ6jRFLoI^7;mxDNkD;b%0Fth51vBf;S=P(e zjxhwTeth!|Xn=A*>_5LM?ji5r(XKWiw;2LybziZMg(!sRd%@oGP$Q<7F#YNJWO!)9 zFgpV_7;Hl`k&OM=_w<*&+|_u8XDEO$dx4*fCAV>(GF6?;^~CR^O{I*E-OYKN)Sv{O z_^e?`N&|Q;QBz-P^1b~9`&O~5Pmojrl8q(cAHSztzXZP~!LPjCn#a8@7lT=SUZx6w zSnEWS)1JR{35oaCyRG@~qMXlO~vlv-60qM=sWj7*F8ICRIkCwsXdQUw+QmZ|Z3fb@@jgTEZvIFl!<# zxoXeq8XCfvcY5TirEN`uFHKd6wfam|$$6_6L$#g7c6;)(*rpy&`ES>s@k?v%ZRS}8X?TmXMpL-rw^~WcHP=OcC%|xK~TIj zE4`~SfaIgKa)p!tBe^On0`S`GjA>C+1}6~a(m3I@T;pALwlnt@UU~(%HM6JjF0LeZ z6!1Rl-o}>W!syP+)OWwJ-vKwE4z80+Dx4*Jr&!=iqyaD6;=h2nU7?F_AM-Zo_BS0JT5YIVq{E(iih#QexbUx&UwWOJ(d{yh@pGdDjVaS zA>&@h#6QN_9|jwnogV&nbR=yHjr5qMkEj5lWLM)9V&w&U_eeYHUb-Cbe{~bS^s+Oq zv!Zc$ZH@?yghoj;${=jiPK2dBix8$T!qP@Md^&g@enBnlJ;}otw}@crt@K1NBG`!t zHf|@u(w;>GQ z1`&)1b`pYZy8qilu(W3p!4xJ~+DPgX%#b0N(H?^Jy4O8}2u1`uA;FAR5-jUkL@;#; zmNk;<1T!UWcEA_6h+xKb&m@8o!A?jpvz-J>dlnH)VS=TNq&mS2i4p%%;*TWJqxxi+cdHKs>cM|a z7&DoGA57+K9DVMrQBL^&-bVPQl2vJ-Q(WllY|b3NHWeseY1&=8AxU&3MDZ^(6U zTNZIb1EFE3pIc8fkda$WG&ID!-7A?hRH32QuvtUqy(Tme8g?3`^+W?Xq}4=2Lrl>o zhTgup>^hT1pDLBgod4-Sv}D}W?41S&=7xY4QL>D>=WU31RS2Ym4Ixp zO4(p}BUqvlENBF4BV3d~N%3|bvbY|~52M%b^x#iCVqq9&r^_Z2h}jVyv=|NnCByO2 z_x7l5bA5a}0@O(LZ9CH7TLlN)J0eRI#aB|Hr`D@03ra%okh_JKy?13IRH}iOv)C3pcW1R5WKC8 zR|^n;2taZtUUUKwopNaofK)yCvIQV`{zCu+Z)@Yt00bZckleW%od85D;vAOeuw=^c##L??Ed10ZEj@jM1l51v^bK)kBjn3tabL;#XIBcv06 z=v+{90Ho?!p)CNx(;fmKtfc1nyzxcM{l;F~i!XT6?(O2vhY(ZK>CFpU3|o8xD75eW zbLBaO(1yBm<9(1+_4ouO%{|}P@Rw#=Mevcy`*6BdR4H(5H0Dej8_2PK7ydLSJ@~WV z#lH_s{BziA*EuE=Dn;Jr4R30|468n{ox$i$U~8S_yGZ~dk9U&*)EJjH)WtztOk<-} z{5gZqoA}c@$9I$XLl*BQ@uw*!?_R-TC$+%W@3x9RXRvt_e_ChwZW4dU;oT(uG{xiH zgXcelKk&9z@#hRKZ{knu{N7FC4;j3h#Gj^EynBV|j|_j{ZEfPuIZWQfpVryE8^j;- zcQ=VY&2e}aa?Hu_2j12y{+z+%P5fz{+q+5pA$xa|_|q7JH$3ej{J~0U(p957{A=~y z2=U=OcR+khE)R0NXzAN)wuD;oBGl4ni`5+a8b&VY>YuLu^rvp83knq4Ak-nKwY^~X zo&)L$on#6^;^*(~B;s_PJS%Wv>5AJLQ04iPzT`JF;6Lx&g$to^eHVOn!^Ir$sM*Yz z?=7;s9eN%1_8SD!6idB=2(TB~MG*K|0^P22&A%gJ|M%yQK-~ym4WIJy*I)kr_QR_W zH*Y(0=QKXF)7;wd>E4AJ`F_|Le1FA0Y~IO`gya>!B7@x+{c>sV&1QCQX1K%Na5%a( zZFe%eoea&PGc_E0JT``-o@tJpe$RH@+krV7PkYX-F&x{2>1f~#Z_g*~FS1t5(9e?Q z#Fx&>gsdC|PB4vrhyrJIw=n-{cKu1$2-e2rt1vN{c*(WFqo!N#xHc@^}Z^ zWxY~o*7(U}AT9B{lA5IOJG@S$@R}<8{+^|aC_IK_qVVrX;m@e>c){jn-AHFv_)%{x zt?;~(8l>>M+(e}Cx+=V(vcmI_Ocee{Qus3}Jl>vsS-;Pj6@ECJNGm+Aq$Vl+4!;g5 zysipwDy{Gsl8M6qL<)aSg&%h=?5H`n!jF6M3Xhf4>XeDj622AVf01_RMX&&Iw4-2Q z`~I5Q&OcUMq>3iN4TEPh-J!u-_RPMX{@Hnc1jFnI$RarP44#?S4-a%+p+45x^Mjx! z12h4kqRsPzOcU0KEtT^yMhgd#15F&vEC~DWimyXy;jeN-te@gbYUiZU`+lb*H4n@z z1!HPhGJ6EjvjKEGu3?IFY)(cD7Xi^kKS5l|xo|8_lLt#_ici01UPfL~txU^E@FaLD z<*{ul#wlenZYtvnrsTK8xN|bD z7B|Cjp5#k04jz4f#$DXdaH;f?#5iJ{QV#2;GOpmsen*TuC*%5^(WO1vmtq_|`u>c= zdTM0~OJW=`PAQLhQyEuqrT<8bJ1^sUmv*IJj&VI1#`W5nCX*OPj8n>G-%!RCefd8T z(Vw_Uv44TTg!pVf6iE-y;T=&u^6E4TNt_3{6!lyHc>%}w literal 0 HcmV?d00001 diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/manifest.jsonl b/sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/manifest.jsonl new file mode 100644 index 000000000..1316faf93 --- /dev/null +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/manifest.jsonl @@ -0,0 +1,3 @@ +{"access": "local copy", "bytes": 106602, "date": "2026-10-28", "doc_type": "xlsx", "note": "Policy workbook ZIP member. Fresh resolved ZIP GET at 2026-10-29T15:51:18Z returned HTTP 403; local member bytes match PROVENANCE.md. The local ZIP container is not supplied; only the extracted member was read and hashed. Local provenance: the coordinator's OBR PROVENANCE.md (browser download 2026-10-28; server Last-Modified Wed, 28 Oct 2026 14:05:59 GMT for the ZIP). The lane hashed the bytes from its local primaries folder.", "publisher": "Office for Budget Responsibility", "retrieved": "2026-10-29", "sha256": "cdcc2a06b100241bf3010bd159a4e5d43daef2348a3850797b1e88a308fd45fd", "title": "Economic and fiscal outlook – October 2026: detailed forecast tables – policy", "url": "https://obr.uk/docs/sodugn2q20l901oy19v1cz5e/efo-october-2026-detailed-forecast-tables-zipfile.zip"} +{"access": "local copy", "bytes": 1754344, "date": "2026-10-28", "doc_type": "pdf", "note": "Supplied local PDF; its resolved URL is absent from PROVENANCE.md. Fresh EFO landing-page GET at 2026-10-29T15:51:17Z returned HTTP 403, so the PDF link could not be discovered. URL is the publication landing page, not an invented attachment URL. Local filename: economic-and-fiscal-outlook-october-2026-web-accessible.pdf. PDF cover and imprint state October 2026; publication date is the EFO-page date the coordinator's PROVENANCE.md records for the publication. Local provenance: the coordinator's OBR PROVENANCE.md (browser download 2026-10-28; server Last-Modified Wed, 28 Oct 2026 14:05:59 GMT for the ZIP). The lane hashed the bytes from its local primaries folder. PROVENANCE.md does not list this PDF: its origin and hash are recorded only by this lane (review finding 7).", "publisher": "Office for Budget Responsibility", "retrieved": "2026-10-29", "sha256": "9e0e9c340e35e603066f3124a614fcd68feadb3d5d10d1f76fb536e150e6225e", "title": "Economic and fiscal outlook – October 2026", "url": "https://obr.uk/efo/economic-and-fiscal-outlook-october-2026/"} +{"access": "local copy", "bytes": 218000, "date": "2026-10-28", "doc_type": "xlsx", "note": "Fiscal aggregates workbook ZIP member. Fresh resolved ZIP GET at 2026-10-29T15:51:18Z returned HTTP 403; local member bytes match PROVENANCE.md. The local ZIP container is not supplied; only the extracted member was read and hashed. Local provenance: the coordinator's OBR PROVENANCE.md (browser download 2026-10-28; server Last-Modified Wed, 28 Oct 2026 14:05:59 GMT for the ZIP). The lane hashed the bytes from its local primaries folder.", "publisher": "Office for Budget Responsibility", "retrieved": "2026-10-29", "sha256": "3ffa3f1e63667b2ae9fb9626dc1da881673227d43aeeb52f8c3db168a5c75d6b", "title": "Economic and fiscal outlook – October 2026: detailed forecast tables – fiscal aggregates", "url": "https://obr.uk/docs/sodugn2q20l901oy19v1cz5e/efo-october-2026-detailed-forecast-tables-zipfile.zip"} From 5bd845baa6042a07c051900b72e2d7578e3694c0 Mon Sep 17 00:00:00 2001 From: Claude Date: Tue, 3 Nov 2026 13:26:10 +0000 Subject: [PATCH 13/52] Give each proposed scope one key across families, and record the families in the README MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit README rule 2 asks the coordinator to give each scope one key before merging. Three red book keys and five RF keys named scopes another family already names: - the red book's Table 1.3 total effect of policy decisions and RF's "total effect" restatement print the OBR's Table 3.1 total, so all three share ab2026__package_government_decisions_since_march_2026; - RF's restatements of the OBR's tax-measures and tax-administration groupings take the OBR family's scope names, which carry no `obr`; - the red book's Barnett consequentials share the Welsh rows' key (Wales is £450 million in one, £448m in the other); - the Defence Investment Plan crosses lines 40-42, so the red book's row joins the OBR's package key; - RF's minimum wage key takes the red book's spelling. PROPOSED_KEYS.md is regenerated (and checks), and COUNTS.json is pinned for the twelve landed families. The README gains a family table, the four families with no rows and why (cpag, smf, niesr, fai_sfc; AB2025 carries no empty family, so their files stay on the staging branch), and the IFS rows still to come. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_01LwQq6HuQUQ1obM9JvyhJbE --- .../harvest-uk-ab2026-2026-10-29/COUNTS.json | 4 +-- .../PROPOSED_KEYS.md | 23 +++++--------- .../harvest-uk-ab2026-2026-10-29/README.md | 30 ++++++++++++++++++ .../uk_hmt_redbook/NOTES.md | 17 +++++++--- .../uk_hmt_redbook/claims_staged.jsonl.gz | Bin 36363 -> 36365 bytes .../uk_rf_ab2026/NOTES.md | 21 +++++++++--- .../uk_rf_ab2026/claims_staged.jsonl.gz | Bin 52874 -> 52874 bytes 7 files changed, 69 insertions(+), 26 deletions(-) diff --git a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json index ce16d3c0e..177d7765c 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json +++ b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json @@ -27,7 +27,7 @@ "uk_hmt_redbook": { "rows": 432, "primaries": 5, - "claims_sha256": "3dde1f25fa36df598b54340324e49787e5483bf12962371a25adc0319955a476", + "claims_sha256": "1ea59fa28e9914b0c473a30fce88cc335d64917f77ea972dcdfc58d0723513da", "sources": [ "hm_treasury" ] @@ -67,7 +67,7 @@ "uk_rf_ab2026": { "rows": 766, "primaries": 7, - "claims_sha256": "d4976001ec1985cbdd65ceddfc8768ca85bf5611ee480ac2b7e9fa138420f9c9", + "claims_sha256": "a251c700f0f10d4ae47ca8c1d2c2a8ba9ba660f768d0169406a14dd28e8c3f85", "sources": [ "resolution_foundation" ] diff --git a/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md b/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md index 9d199dea5..c0427858d 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md +++ b/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md @@ -4,45 +4,38 @@ Generated by `pipeline/ab2026_proposed_keys.py` from the staged families; don't A row whose scope matches no registered measure has `measure_key: null` and names one of these keys in `conditions.proposed_measure_key` (README rule 2). A key is a measure with no Table 6.1 line (`ab2026__`), a package (`ab2026__package_`) or an option (`ab2026_option__`). Each family's NOTES.md gives the key's scope as Table 6.1 lines or the producer's own words. Tranche 2 registers each key in `data/uk/ab2026_measures.json`, or folds it into a registered key, before the ingest. -90 proposed keys (17 measures, 60 packages, 13 options) on 955 rows. +83 proposed keys (15 measures, 55 packages, 13 options) on 955 rows. | proposed key | kind | rows | families | producers | |---|---|--:|---|---| | `ab2026__british_business_bank_defence_commitments_accelerated` | measure | 9 | uk_hmt_costings | hm_treasury | -| `ab2026__defence_investment_plan` | measure | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__departmental_spending_growth_2031_32` | measure | 13 | uk_obr_tables | obr_efo | | `ab2026__energy_debt_circuit_breaker` | measure | 3 | uk_hmt_redbook | hm_treasury | | `ab2026__english_devolution_agreements` | measure | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__gas_standing_charge_moved_to_unit_rates` | measure | 34 | uk_ifs_ab2026, uk_rf_ab2026 | ifs, resolution_foundation | | `ab2026__gas_supplier_levy_funded_from_general_taxation` | measure | 1 | uk_rf_ab2026 | resolution_foundation | -| `ab2026__minimum_wage_rates_april_2027` | measure | 2 | uk_rf_ab2026 | resolution_foundation | -| `ab2026__minimum_wage_uprating_april_2027` | measure | 2 | uk_hmt_redbook | hm_treasury | +| `ab2026__minimum_wage_uprating_april_2027` | measure | 4 | uk_hmt_redbook, uk_rf_ab2026 | hm_treasury, resolution_foundation | | `ab2026__mod_counter_fraud_strategy` | measure | 1 | uk_hmt_redbook | hm_treasury | -| `ab2026__package_barnett_consequentials` | package | 12 | uk_devolved_govts | welsh_government | +| `ab2026__package_barnett_consequentials` | package | 16 | uk_devolved_govts, uk_hmt_redbook | hm_treasury, welsh_government | | `ab2026__package_bbb_equity_community_lender_and_venture_debt` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_breathing_space_for_households` | package | 1 | uk_hmt_redbook | hm_treasury | -| `ab2026__package_budget_barnett_consequentials` | package | 4 | uk_hmt_redbook | hm_treasury | | `ab2026__package_business_rates_rhl_sbrr_and_transitional_support` | package | 2 | uk_hmt_redbook | hm_treasury | | `ab2026__package_carbon_price_support_and_cfd_levies` | package | 3 | uk_obr_tables | obr_efo | | `ab2026__package_civil_service_transformation_programme` | package | 1 | uk_obr_tables | obr_efo | | `ab2026__package_compliance_measures_announced` | package | 1 | uk_obr_tables | obr_efo | | `ab2026__package_corporation_and_bank_tax_changes` | package | 7 | uk_obr_tables | obr_efo | -| `ab2026__package_defence_investment_plan` | package | 1 | uk_obr_tables | obr_efo | +| `ab2026__package_defence_investment_plan` | package | 2 | uk_hmt_redbook, uk_obr_tables | hm_treasury, obr_efo | | `ab2026__package_dividend_and_capital_gains_tax_changes` | package | 6 | uk_obr_tables | obr_efo | | `ab2026__package_dwp_fraud_and_error_measures` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_fuel_duty_and_electricity_vat` | package | 11 | uk_obr_tables | obr_efo | | `ab2026__package_fuel_duty_freezes_since_2010_11` | package | 1 | uk_obr_tables | obr_efo | -| `ab2026__package_government_decisions_since_march_2026` | package | 347 | uk_obr_tables | obr_efo | +| `ab2026__package_government_decisions_since_march_2026` | package | 374 | uk_hmt_redbook, uk_obr_tables, uk_rf_ab2026 | hm_treasury, obr_efo, resolution_foundation | | `ab2026__package_hmrc_tax_gap_and_online_marketplace` | package | 7 | uk_hmt_redbook | hm_treasury | | `ab2026__package_hmt_distributional_budget_2026` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_hmt_distributional_since_ab2024` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_household_energy_payment_and_electricity_vat` | package | 5 | uk_rf_ab2026 | resolution_foundation | | `ab2026__package_national_wealth_fund_and_british_business_bank` | package | 4 | uk_obr_tables | obr_efo | | `ab2026__package_non_scorecard_decisions` | package | 12 | uk_obr_tables | obr_efo | -| `ab2026__package_obr_direct_effect_of_tax_measures` | package | 13 | uk_rf_ab2026 | resolution_foundation | -| `ab2026__package_obr_forecast_policy_effects` | package | 24 | uk_hmt_redbook | hm_treasury | -| `ab2026__package_obr_tax_administration_compliance_and_debt_collection` | package | 7 | uk_rf_ab2026 | resolution_foundation | -| `ab2026__package_obr_total_effect_of_government_decisions` | package | 3 | uk_rf_ab2026 | resolution_foundation | | `ab2026__package_other_ame_spending_decisions` | package | 6 | uk_obr_tables | obr_efo | | `ab2026__package_other_departmental_spending_changes` | package | 7 | uk_obr_tables | obr_efo | | `ab2026__package_other_hmrc_compliance_and_debt` | package | 1 | uk_obr_tables | obr_efo | @@ -62,8 +55,8 @@ A row whose scope matches no registered measure has `measure_key: null` and name | `ab2026__package_rf_other_tax_measures` | package | 6 | uk_rf_ab2026 | resolution_foundation | | `ab2026__package_rf_parliament_permanent_tax_and_benefit_measures` | package | 42 | uk_rf_ab2026 | resolution_foundation | | `ab2026__package_spending_decisions_since_march_2026` | package | 7 | uk_obr_tables | obr_efo | -| `ab2026__package_tax_administration_compliance_and_debt_collection` | package | 6 | uk_obr_tables | obr_efo | -| `ab2026__package_tax_decisions_since_march_2026` | package | 6 | uk_obr_tables | obr_efo | +| `ab2026__package_tax_administration_compliance_and_debt_collection` | package | 13 | uk_obr_tables, uk_rf_ab2026 | obr_efo, resolution_foundation | +| `ab2026__package_tax_decisions_since_march_2026` | package | 19 | uk_obr_tables, uk_rf_ab2026 | obr_efo, resolution_foundation | | `ab2026__package_temporary_business_rates_reliefs` | package | 1 | uk_obr_tables | obr_efo | | `ab2026__package_tpa_business_rates` | package | 8 | uk_tpa | tax_policy_associates | | `ab2026__package_tpa_cgt_founders_relief_and_badr` | package | 1 | uk_tpa | tax_policy_associates | @@ -97,4 +90,4 @@ A row whose scope matches no registered measure has `measure_key: null` and name | `ab2026_option__lha_unfreeze_2027_28` | option | 17 | uk_rf_ab2026 | resolution_foundation | | `ab2026_option__nics_on_investment_income` | option | 10 | uk_centax | centax | | `ab2026_option__nics_on_partnership_income` | option | 24 | uk_centax | centax | -| `ab2026_option__obr_tax_measures_with_continued_fuel_duty_freeze` | option | 11 | uk_rf_ab2026 | resolution_foundation | +| `ab2026_option__tax_decisions_with_continued_fuel_duty_freeze` | option | 11 | uk_rf_ab2026 | resolution_foundation | diff --git a/sources/harvest-uk-ab2026-2026-10-29/README.md b/sources/harvest-uk-ab2026-2026-10-29/README.md index 333a73d15..e9fbddcd7 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/README.md +++ b/sources/harvest-uk-ab2026-2026-10-29/README.md @@ -46,6 +46,36 @@ The AB2025 harvest README's rules 1–10 and its row contract apply, with the AB A measure's own parameters are its design: rates, thresholds, payment amounts, dates and durations. So are spending allocations to programmes, financing remits and contingent liabilities. They go in `reform_hint` or `conditions`, never as rows (the AB2025 red book family kept "£9.90" in `reform_hint` the same way). A worked example's inputs follow this rule too, and its computed outcome is a row. 9. **A component of one measure keeps that measure's key.** A printed part of one Table 6.1 measure carries that measure's key, with `conditions.component` naming the part and `conditions.row_type: measure_component`. A part can be a fiscal head, a sub-measure within a line (the tax-gap package's items, for example) or a nation. Components are never summed with the measure's total. Proposed package keys are for scopes that cross Table 6.1 lines, or have none. +## Families staged + +12 families, 5,891 rows (401 `restated`), as `COUNTS.json` pins them. + +| family | rows | primaries | producer slugs | +|---|--:|--:|---| +| `uk_centax` | 110 | 5 | `centax` | +| `uk_devolved_govts` | 19 | 2 | `scottish_government`, `welsh_government` | +| `uk_hmt_costings` | 597 | 1 | `hm_treasury` | +| `uk_hmt_redbook` | 432 | 5 | `hm_treasury` | +| `uk_hmt_tables` | 762 | 2 | `hm_treasury` | +| `uk_ifs_ab2026` | 87 | 5 | `ifs` | +| `uk_ippr` | 3 | 3 | `ippr` | +| `uk_obr_tables` | 2,797 | 3 | `obr_efo` | +| `uk_rf_ab2026` | 766 | 7 | `resolution_foundation` | +| `uk_thinktank_misc` | 12 | 5 | `citizens_advice`, `cps`, `institute_for_government`, `learning_and_work_institute` | +| `uk_tpa` | 305 | 10 | `tax_policy_associates` | +| `uk_trussell_wpi` | 1 | 1 | `trussell_wpi` | + +Each family's `stage.py`, which writes its claims file, is kept on branch `uk/ab2026-harvest-staging` under `harvest-staging/ab2026/lanes///`, with the lane briefs and the five reviews; this branch carries only each family's three files, as AB2025 does. The review fixes, the IFS addendum's sections 3 and 4 and the cross-family key harmonisation were applied on 3 November 2026; each family's NOTES.md records them. + +**Families with no rows** are not carried here, as AB2025 has no empty family. Their working files stay on the staging branch: + +- `uk_cpag`: both CPAG pages (its Budget response and a pre-Budget call) returned Cloudflare's challenge (HTTP 403) to every fetch on 29 October. +- `uk_smf`: the Social Market Foundation page returned a challenge to every fetch on 29 October. +- `uk_niesr`: NIESR's Budget-day reaction ("Breathing Space, Not a Strategy") was fetched, but it holds macro and fiscal-stance commentary for the #55 lane and restated OBR figures, which the AB2025 NIESR family did not stage either. +- `uk_fai_sfc`: the Fraser of Allander Institute's reaction refused the fetch (HTTP 403), and the Scottish Fiscal Commission published nothing in scope on 28 or 29 October. The Scottish Government's five rows, first staged here, moved to `uk_devolved_govts` on 3 November. + +**Still to come from the coordinator:** rows from the two IFS pages read in a browser on 29 October (the initial response and the collection's opening remarks), which are not on GitHub. `uk_ifs_ab2026/NOTES.md` lists them as pending. + ## Producer slugs and families | slug | producer | family dir | diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/NOTES.md index 7a8f15bb5..ed1e29ad7 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/NOTES.md @@ -80,7 +80,7 @@ Rows by measure key: | `ab2026__national_wealth_fund_capital` | 1 | 0 | | `ab2026__parental_allowance_fathers_partners_adopters` | 1 | 0 | -The 191 context rows carry no proposed key (README rule 2): 136 restated fiscal aggregates about no measure, and 55 Impact on households rows (Table 2.C and the context statements of paragraphs 1.10, 2.4, 2.5 and 2.10). The other 24 Table 1.3 rows, the OBR's effect of policy decisions, carry `ab2026__package_obr_forecast_policy_effects`. +The 191 context rows carry no proposed key (README rule 2): 136 restated fiscal aggregates about no measure, and 55 Impact on households rows (Table 2.C and the context statements of paragraphs 1.10, 2.4, 2.5 and 2.10). The other 24 Table 1.3 rows, the OBR's effect of policy decisions, carry `ab2026__package_government_decisions_since_march_2026`. Rows by metric: @@ -172,7 +172,7 @@ A spending figure is a measure's design unless the red book presents it as the m Each key below is the whole scope of its rows, stated as Table 6.1 lines or "no Table 6.1 line", with the red book's words. -- `ab2026__package_obr_forecast_policy_effects` (24 rows): the OBR's "Total effect of policy decisions" on PSNB in Table 1.3, with its Tax and Spending parts (components) and the memo effect on the current budget deficit, 2026-27 to 2031-32. Crosses Table 6.1 lines 1–98 and adds the OBR's indirect effects; Table 1.3 footnote 1 says the OBR's EFO Table 3.1 explains the "Differences between the direct impacts of the policy measures as set out in Table 6.1 of the Budget, and how the OBR incorporate these into their forecast". Not `ab2026__package_total_policy_decisions`. +- `ab2026__package_government_decisions_since_march_2026` (24 rows): the OBR's "Total effect of policy decisions" on PSNB in Table 1.3, with its Tax and Spending parts (components) and the memo effect on the current budget deficit, 2026-27 to 2031-32. Crosses Table 6.1 lines 1–98 and adds the OBR's indirect effects; Table 1.3 footnote 1 says the OBR's EFO Table 3.1 explains the "Differences between the direct impacts of the policy measures as set out in Table 6.1 of the Budget, and how the OBR incorporate these into their forecast". Not `ab2026__package_total_policy_decisions`. - `ab2026__package_hmrc_tax_gap_and_online_marketplace` (7 rows): exactly Table 6.1 lines 45 and 46. Table 6.3's "Subtotal including memo" (six years) and footnote 73's "£2.3 billion from this Budget": Table 6.3 prints +2,250m for lines 45 and 46 in 2029-30, against +1,980m for line 45 alone. - `ab2026__package_parliament_tax_gap_measures` (1 row): lines 45 and 46 plus the tax-gap measures of Autumn Budget 2024, Spring Statement 2025 and Budget 2025 (footnote 73). Paragraph 2.46: "the total due to be raised from measures to close the tax gap announced during this Parliament to over £12 billion in 2029-30". - `ab2026__package_property_and_savings_income_rates` (2 rows): exactly Table 6.1 lines 61 and 62. Paragraph 2.50: "84% of income tax payers will be unaffected by these changes"; "Around half of the revenue raised from this change is expected to come from the top 10% of households". @@ -186,11 +186,11 @@ Each key below is the whole scope of its rows, stated as Table 6.1 lines or "no - `ab2026__package_parliament_regulatory_burden_reduction` (1 row): no Table 6.1 line. Paragraph 4.51: the 25% administrative-burden target, "with over £2 billion of net progress secured" across this Parliament. - `ab2026__package_bbb_equity_community_lender_and_venture_debt` (1 row): paragraph 4.46's "three measures": £1.4 billion of commercial equity commitments, £165 million for community lenders and £450 million for a venture debt product, which the British Business Bank's additional £5 billion of capacity enables. The red book does not map them to lines; the BBB's capacity sits in Table 6.1 line 29 (Table 6.2 line 3) and the Growth Guarantee Scheme in line 75. - `ab2026__rolls_royce_gas_turbine_programme` (1 row): no Table 6.1 line named. Paragraph 4.80: "co-investment in a joint, 10-year, £2.5‑billion programme for next-generation gas turbine technologies for civil and defence aerospace"; paragraph 4.83 says it "has not been contracted". -- `ab2026__defence_investment_plan` (1 row): the Defence Investment Plan published in June 2026, which no single line covers; lines 40–42 carry its savings and funding. Paragraph 5.34: "The Defence Investment Plan will create approximately 60,000 new jobs by 2029‑30 compared with 2023‑24." +- `ab2026__package_defence_investment_plan` (1 row): the Defence Investment Plan published in June 2026, which no single line covers; lines 40–42 carry its savings and funding. Paragraph 5.34: "The Defence Investment Plan will create approximately 60,000 new jobs by 2029‑30 compared with 2023‑24." - `ab2026__english_devolution_agreements` (1 row): no Table 6.1 line. Paragraphs 4.67–4.68: "the final stages of agreeing 10 new devolution agreements", after which "over 90% of England's population will have a devolution agreement in place" (the 90% includes existing agreements). - `ab2026__package_dwp_fraud_and_error_measures` (1 row): Table 6.1 line 47 plus the DWP fraud and error decisions of paragraphs 6.19–6.22, which have no lines of their own. Paragraph 5.17: "new measures expected to save an additional £1.4 billion over the forecast". Line 47 alone prints +600m over 2026-27 to 2031-32 (and Table 6.2 line 9 +340m). - `ab2026__package_parliament_welfare_fraud_and_error_savings` (1 row): line 47's measures plus the fraud and error measures of earlier events in this Parliament. Paragraph 5.17: "bringing total savings from fraud and error to over £20 billion over the forecast". -- `ab2026__package_budget_barnett_consequentials` (4 rows): the Barnett consequentials of "decisions confirmed at Budget", 2026-27 to 2029-30, excluding the technical reduction from lower employer pension contributions (paragraph 4.150). Crosses Table 6.1's England spending lines; no line of its own. The Scottish, Welsh and Northern Ireland figures are components. +- `ab2026__package_barnett_consequentials` (4 rows): the Barnett consequentials of "decisions confirmed at Budget", 2026-27 to 2029-30, excluding the technical reduction from lower employer pension contributions (paragraph 4.150). Crosses Table 6.1's England spending lines; no line of its own. The Scottish, Welsh and Northern Ireland figures are components. - `ab2026__mod_counter_fraud_strategy` (1 row): no Table 6.1 line. Paragraph 5.32: "a new Counter Fraud Strategy in November, with investment in fraud and error analytics and closer working with the Public Sector Fraud Authority set to deliver £250 million of benefits". - `ab2026__state_pension_triple_lock_adjustment` (2 rows): no Table 6.1 line (its savings fall after the forecast). Paragraph 5.23: "adjusting the Triple Lock from April 2030 so that the State Pension rises each year by the highest of 2.5%, inflation, or the amount required to maintain its value relative to earnings as reached in April 2029"; DWP's estimate (footnote 61). - `ab2026__package_hmt_distributional_budget_2026` (1 row): HM Treasury's own grouping, the Budget 2026 policies modelled in Impact on households (paragraph 1.4): Table 6.1 lines 2, 3, 4, 5, 8, 11, 15, 26, 39, 49, 52, 59, 60, 61, 62, 72 and 95, plus the private landlord registration fee, which has no line. "Decisions announced at Budget 2026 (including those pre-announced where funding is allocated at Budget 2026)". It excludes line 1, the Household Energy Payment. @@ -305,3 +305,12 @@ The rework stopped before its report, so on 3 November each instruction of `brie - **Checker:** 0 problems. **Determinism:** a re-run of `stage.py` reproduces the gzip and manifest byte for byte. - **Spot check: not repeated.** The brief asked for 15 random rows re-opened at their locators. The 3 November session's network policy refused gov.uk and assets.publishing.service.gov.uk (HTTP 403 at the proxy), so no primary could be fetched, and no row was re-opened. The 30-row check above is the rework's own; nobody independent has checked this rework yet. +## Key harmonisation, 3 November 2026 + +The coordinator gave each scope one key across families (tranche 1, task 3.4). This family's renames: + +- `ab2026__package_obr_forecast_policy_effects` → `ab2026__package_government_decisions_since_march_2026` +- `ab2026__package_budget_barnett_consequentials` → `ab2026__package_barnett_consequentials` +- `ab2026__defence_investment_plan` → `ab2026__package_defence_investment_plan` + +The OBR's Table 3.1 "Total effect of Government decisions" prints the same figures as Table 1.3's "Total effect of policy decisions" (2.1 in 2028-29 in both), so the red book's restatement shares the OBR family's key. The Barnett key names the scope the Welsh Government's figures share (£450 million here, £448m there). The Defence Investment Plan crosses Table 6.1 lines 40–42, so it is a package, shared with the OBR's DIP funding row. Table 1.3's Tax and Spending parts stay components of the total, as staged. diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/claims_staged.jsonl.gz index e33581a5eceb04d1c8868fa502069d938388f86e..2742cf735c797a3c25141397af4cd98131b17b3a 100644 GIT binary patch delta 35146 zcmYIvV{~Orv~6tLwrxA<*tYGYqZ1pQoTQVE?R0G0HaoWMynOfFG4B35XU$q;@7inB ztXeB36D%$h3_u(M1M=Sk_I2L59=|U0o6Dy=hs?N#hFVJX#IJ@G{8{_pgb$~w0^Uth zS(Ro`F6bKaodO=6M-6}UAUr?Ih6&=P3pE26BkPGws)n5u9!T!8eGeI}+%9V=xdmTD zVlKH{1clf>K!h{3KSvr61CY5(;IVs64{`bjpMOJ3)o-2eRy(%(p8dI_jlUgr0`w*n z|Fr^qhzrPXBIj7P?ooe1cV)ZG8+mWE?A3B-xTM9Y+&Ay}u+KMrq9VPmYjHK$;1Tl5KMMYR_`93!Dbwg{*j3pPP@S z4000Ut|zl&UszbKby}8&tZn(4GnwMDsj%!hc5PiuCK_yqpk#os z9i>Ujk{E1GsD8}~tN7;@;hl@EH0~jTsrAjn6#_+{EUA)t>xJPePp5@b?(`~?=Edp+ z=m%N<9_Y@UT+H~b99O)-1mXH=qN*#jZurWcVv}yNyu4PE=>We;pGhV}o1N9?2M(CS z&%3K`=s$-+M!@D!&hnRy_GYOL&ohm!Wh}3Mdwv=JycD)7k>9dBn}uu});49GMtncG zzCIl~7cU;6KmOkAP3OdEB^Kg20eoJF!)G~c?`}d1udm=~IdV#Bwn>2T)4{Y>-5-#U zJvnYGz;-Z7fXSB*Nu4Ih574Ba=yD`ml0v7T66uqZlrr`dRp^fwb_Imd?Q5kIt~g(4 z-c;?9L7tX1%r+GAqZs(D?7vjCZ+nWU{lrjPC{G@E1OhC33ZZ71?l!}yA%60)2e0%+ zF^jA?tlX}fYkevkBfn1(`alVcMuJDZD;Oh%vG|380+0-zkhUx(El-v~Lg14Eo)kxB zV)02uqOuh7_q0JThfs6S1vZh6o^LvBn6GK$JPkf_UKd*G2{!VS4wFrCovioek?p%yg-5fg}|^&bjpxrejQrn1jryEAHBZ8sIQe${CR!@YBI zzB>d71Biyvwb^<~AfHghED50Ec*rP1f5PK;k*u(Zmq@1nFKaPmMF6vtm z0VP*w5+j6)Ws~Sxswsqt#WKcVce3GWA;HH!*>51X^|QG!B70K|(xiRJPD1Colw9Ri zhmI_5qjlD=V-PLWnq2df0f{yEQM_|CKDJe;OKI2C4XSTyPoim{^C;1vX^c_=N$#Kt zg@)z)X;V0&+)yoMPrW}`I$aQ4bA}yR0jv%g>WflUUl_fZO_0HS(R{()Ku6jMKcV_f zXzzSlcU>f-9>}O8VfgCyAZGb(@U|%z;_C`a>@M3!$LO9l1ycjIo-jndT~>m@5Ra-H zRC^bNIT5Vi_knmcUCbiTThPuzbv|`Q)Xje3#*kt>Y)9uhHJN(em4Cp>Zp-Wv08YEP zH)sJidxvKi2U9mydeBxF%*j{CC|0v{9~CR;K??%P7~FA~#*e)ZLw@g>{@7_=*W%L# zfe?VVwXzT#rLKb7x0oG9KOt%u#Vs<(-JCS>wL_?86RTe2oomaDjG!Cj%hsAm(2+E| zcLx(@L{3ChCbtOR<{#ikJeD(tfCA{5JRJ#C=7pLsBxV~0hLt!XWOEcDslkdK z>n-vj8mWGq>w7BQs` z!k_)7YS8_DXB%rg0?*WjoWZQnhu9sceqVL3#bB;K)Z)pFCS2IIH&}t<3708-hDvqJ zOF(@FYSG~0^}v&@-B&h2fHz4`OyDIMZT!e+4FS0DmUB4^h-&Qh#h&}}s82&A>?W-& zbID$X9%fH=Mym`%VZr@H>TL3tKdAL5L;M5C5y+KB2QswdrkKdi1iqZ6O7-c0vVfKaH`F8Ud2g%v*F7!R2|+ zy|JTzqqtpq7l*&iYa7TbnEWC7ZuCI!8BAu%Jkrvn6I9IpZnAxmF(Z=zKT&gGU=>k_ zYS;yczIosJ9ut?fQlaZCgQA21kOqbTpkXgAsvNAKY1RHdoRd5s529ffdq~3%#arW| zyukzKg0mlM*zqEO;L3F~vYoK)&kI}H6OKui)J{mcc;iBAMRJYqhQT0EXRfz3@o(sd zI0_~Rr}o@UUqowIvjj2e2ohX4c%;*_GXi={R!r)g0Z*%r%!{M)jEt}DGZs>e?A!3o zas!8~14WrvWKC!#Q>z*sy<<9#TW8{w-Fq)RsoS|fjoZ10>fs7?QY+kGU;zR5DP|z@ zZ^<=ClNqXzT8)gX64*BUeoY_(tUA1Uo(NCtQT1FftRBz~mH=Q6RZG5K%9<#BpxEPy z$7l(^K7c)*I^|Y~LxcFaZkD^nA9R%TwP}t7P{w=5zmM(4az#{@ab=uBE#S|f$H0c= zS7k~9!zbj@^*HgaE)l&fz&;AqO1vdbw)1Bhw#nf%hf~#@^-%R5@+%A7G-$C_B zWX9d{J$-efX93XS{E2Va5EHRTl!{IV8!OlgCuVc^c3SQ^U>@Z}T6>(iLvXUi?_Jmv z$|~H(A3xf&TkKiK3*`>J_N7L6mk7*q^^FI^K_5k=g3c=+Mv(U=MuvO@mFO!ldc$GB zY+rQ#Gjot)U_;3OAD_S;ldpun91|C4ER-0}s#c`Pbp}u}Bc(63=qGs1eeW+Kmzv_q z@yxu6R|Y~o5OfbIg?r=R?Btit`lXZNcc?0Tz$jS|oAeMcYPw4I#<8HlE5XtJSP{r5vZ@4z(3o*=s#w|z zJ_^+#^GD#c#MlR;Ds(X%$SGWiz!p5>ZCRO0(*>UFo8-uV%WfTfNzw zC-D*pcl!1ZHSOhJ&)D2SP4p>$G*4M;!x~fsEZ?&;3iuo9dg&7{mczopOw!~Lcx-Ti zQ}A>E?jQFX`c028hTOXIUaI-RG7?3pZ(Er=0VZNp<_O?3!J})9ia({tT?AqQ_NM&YOwy3hxDvxhL73-h3QT{T?n^N%>$Takg>Y6QG~P*0KYR6PXlVI}c)##t_ThgQA7syA#s+8AK|R zL8X*}NmLuj#&JnEgx{X`GM=~rwRWLKnpqO2jgiM1Gh|_V&cFUdcJ4bz5^G;?&&l|y zNRaDvMA};QII%{$5g2$)@1+LfMR0QKM=YnGM;!mc(0Hoxqq_$EstgUu)hkI(J~MWR zca&SD;KPYhb!)_8um)mbc(F4WQd^KPcxqS5{e(FqzJrZp$V zJ}d&^rcRMMzZ@wTgQ$=I5bXPvhG#4~q>(;kBoUyf!8p!JeXdHsTlTwHmpzdt7EGz8uDn@vHYKP7kMt!okRm4e7myt?{Xb@E5UQAn+n~ZG zeZb(L@!*<8^M=EyXy6FPhtLw&Y)ofbnQcd_7KFHH&G|JoIX>4_QsfFP7&d;QdE6vL z*PSH|;Q=|QvY@H5%usjiY^6LU4Ms31xHu>d1#0PfuVZW=HDznbKka=AZ1PIRLfj$8 zgMiEPhzN&EAqDNSrc79_l0Z=uT!6Kcc?WI{Sm}HEy>`!Dw z?+g&;bZ+f3%_=<*1j?s=rfqblJGW3`5mt#6%n!9F!MVrC;U!_43ntZCy934;X>?t- zhSy14w`s$02bn&{lsqA-$o^$xPd1kE4k^5|TVsPm8~>^DILj(S5Uj&B2bNUul*RWh zFxEXv@NpyB)1Rn}`B0;Cf%^q^1--1uJVJa~(>k48G$3f~7iHr?eWE~ER5E}fC`z+S zzdlkWXL?p8IG_@0_`p%&xFmACGxl$4(NoLBY@VXnGAo(i&cDL~FHUL@%#`&C3?7rqjz5 zxR~o54_WX0OunTIhwaZwQ4SC9eUxxeWQCo-oPe$yNiT< zi8AUQIBNS!zWL%x>Aw$XV09+_s!wjB9(l9Wr5f22l1k7;y~3lzVnx-^3vCM`A`p*? zQ$Cxc*E~XC<52Y0*^3zw0%xzU5e^~1`)(%l?W;}b;Q$OZ{!ulyr>zz&O0_DfOy!rXSP?!E%IIs?Q$a1paK zm?32_2+;x@!|k&4mP4JEnyyCeY++TJ3Eo+15Ysvg(3gW|AOJtUo7;;|kvXx>x*z{H z)7gHa(fmR8stgYqgl9>u#}(dL8V^H9y^1+MTB&N@m|2>98GoF%ubpS+Nw~XW#Aork zE#JMd+?K+QzL(=QP{T|u1*8B^deTRIGg4}+6R3o)1bPL`LC1k9^EXK)8}z9GcB9h) zyk-4R~V&=nVsRrS&a%$GjojERF~3+hhlmJA9Bw`lNz zgmLbn4U)Uq2ixrMuwzY5p?rB?2kygGZ%Tch&G5{OMFJ4%$kII=d67zn%=^;ezJ}s& zpXXVjX-oXPD;G;>{2eopY7FsZM*I2zz3eM(LDXjy4kv$$%bTTue4fOaXJtPyMq^0# zEv%??umu$u({-CvVh%ATY17pn50Y;aW&4m-?5qk=FbddMrTi|Vb7M6N)@-w}E%v=^ zI5=PJ?w60NXN`uC7hiS2=|zd;w!I8NUwb4INm^qXm!$Mq;Pfyh5V827E_jgvsCuSR zrh}bm324jb@%M1#)Y-IKRFRMR5C3pTiiUOW-ZV3zL3FhjM#MXWiKsOz&i4$HGEtuf z8t#UXs@NklS3~{Ra%XLcW(D$pnnPc`hj*%@alsCw;}*2hf2AhslG*U=+>qfCzV3OzN?1TfEMR0_!i{>_e1I#WypaVlO%L%CtT}LdJ|AYGE9-h zfOPWGTiWH+2w0Y3aii|yq#5z#{!E?GXEKX?H^J>Hd+V``nRS1l1i9Ml{hTmIK2${t zjW&V>HlHuZ`-f0p!ekHt{;@5aAe)>@x&PQlm~QlcL7841n{OL#%L|sTs>BFkJYz5-Lyj+ME<(kO)J~RfcAeHOVMbyfM~$v` zVE)7Vx!y-@qW@>as@f>P+YKGR4!Pa@a*tFtWxM#C|AZ7NtB+_HVyJd(0{C$%eJ`)r z+^-qTg>|e~mv|8k$h;sOwgYv8y?cI+)<=+4au#;e73;2KDDM9e1JAsuNvj~49?cXR zJ?1!aX2G@pAcwP>9%Dunr-T@yMqG`Z6gj}JKl|%D1(R%%jE(LAh^aP>pGmR zz0Q{02;*N%6!QXr+eO8PMR6U;nlR)ltfR=|roV6%!$ZdayF^=zs?p#x%8(p?9B2|j zt+%VNKSZt*ycuH&kekP0Z)2-g2fVk3?X@#bqc5#IC3ECLv}QbFKQ~8u)Pu&As>oj- zV|d|kt{bnn5>-PRmGh_8P%9&C%`Z`!Q&3b;y29A0;+qYQnob#I3cm1(X(QwjhIe7! z2MFe-*sRL|h{`yA4X*jRCT`#$CsQbEBl1jFp8HReZ!1K8Zx$Q}0HdCBp&!b9)P7iR zB-xm@wA?y!FF#wLVxDn5OXLEa0={b4qW>LguRvB!WV95hn02?XRQQ>{=wr>l5PcF$ zIh7AB3AKksIiq+`G(dFlLGj|bQCXOz3LQ%><>FBQc#4X6E!azbZO7b75=bTJ1%E*3 zfi2_Kz#hueE!M%Qw&|0LQY(d2kDDpi7a@0f$>bIgFom;lvgIQXbqt3?f#$)>tR!z1nN^X zcQtAOemcCfa3?FrAwJ+NOcVze!9!d6nJZL2hdhxB^qKa934|svSN?v4<*ljWRd&lg z6qyW42WL(ICx@Gu3Yt~TnE~GMabn%(A`u3bSvqk?D9Z4-X8tpK9)~|fDe51dYtsiy zP^|Y-Pnfd8@Pv4Q7%H$M^ADnDXhKSdn3Ev^`0uWKRahb^98`f#=+}m=4AP_geRq32 z6I6r)D1&;Me7FbQ^F5Gsi^p(Djq2Y=zNaNOGTiU~Q(#~i8^x?2 z6(iE6=(*2dl<4&1#T0O{ESPNclp9CF6|()iafxZG6V8A5z2hRE0g$?Pd6{zDZuDmfccy)ohO{{U)hT*9=y7vpzbzfs(8bk1d? z^r7*ecS1nfa%9-@IUf@(mM!zGp*D^FXqd^)fp^k0Bs^dn>B(5pRPqcNIg<9dS(A0I zk)e~M+++r&23=mgHt$31ua!n5O32HtxlLRcbN#@NxN_!|}u|mk)0hFuQggtC_b#R%-W{wz;k?JK-TzU`Vk$%$`pa%f{@MiFDQS+*SDa6 zt9KD2X{dALS2_V>=hYq1R*~+MRs1EsGH&^hE~t+Wk#0ahZnZecBeXv%^emm+JhL5v zgyPJVIh23ZiQSB_(F}1W@$s2`|wR zo+^te#7xEcUt?pZH7Ubpm|A>r)V9yd9u5(U&jUV(&jYWGB$-TG0MpQUR=ieJ#gsFr z(`dvao#xk4n3A*m9Go3@5i!S5AS@v&XcZ8j&;;O>{=Zqi#30|lDs@fp<7RQ`b9tCQ zUO*_uCCA?2NyF$#yMfJ@L3cSeeoaYF%zdG2U2I0>*{!1O>lK&VijP;smfHyyRJA<- z9&}`eaAWy$n=|m-3VP&|6THw9H5G-ValHj45W!&7;6l!jUBt5k{M)<1%2^dY70A!2k>EddjBeJe&XiBn zn{f(u=kb>aVN4si1%B_}?&kzP89dzXIH8lJ@i_??*L#L-}>zcreUqX5N8IinvX=%$air$uqlVo`fVAh zA>yerhkFKT3e~#Ns*5ETbL(X{@X@v74$u0>VVlbErs#;CmQo)%7j7Dn~;q z0MQzMHK@tuK*Se9>FX!B!yZDArZgM|R0ErC?mq?^qm$r#NN2@5Fo(N_6kfm>NJB&d z4v$l=LpJ%$X=F%Wpr!=jy~jc?-;8Af-*GDAsmO&6p?Piyf@+xfFvdAZ0guZKO0Z~2 zso(54ENKt#Xkzaj z6uReJIJcCZbUaZ8-Y}o1oNVl1ba^;B%;VZMw|Q1LLCj(NoXvg&7}R?BGoepCSyokL zUC}$kyCq>dy1f1+a%>d-to*pi*w$=6h7Z*{X;fqk$ZVYo8M z#@I-xXVqVF^rh{lY(mGgqTqGtW%y$181b}z-+l$Lp4Xktk{sN)!ajyTedOi#F|M$W!QK|Y#!phrU=5{f)Kc7nojQ#2d3zM?(Ylq^81r|6`et!EQ0Di`!gxfk97 zRG+ZAhJ8Ew7(yhrDPB<#S3w=|8^kx`sarv2+sptiP-(SHQU+ait9V_v_utX953#<9RuxP@j za&r;t;iFZ*mmW+Dhkpt9260QjU|Nc`Rh{1S!$&`6j;y$j7?Rlp_#$F z7u$P$Te`;6F|c%Cjd|ZyGYo$kJfwj(m*p@-D1<~U=L9BvFbJEU78tsZS@YcrKJVn z?)8qCiO6S5xtW0qWH{zqFM*x$6D#QFl73PJTXdy`<~HCQuEGBFZoIOrUU=u6D>;-a zPt(xn9_l06#>}STcn_Z6)_4q$APN0^{ZAE-=_SXCU5)UM8sbdkm9xlX!Cu3GA`;_9 zW_}h*?v-Y@Kz}{pD47s|J7V&4WK%xg2!5Zz))v8F+Kt%Q4iid&SbiGfr#+%(uB`YJ z)kxl`PZ3tI;M?^{sus7s=~~mYbCESWeQi}u)X`KRT85HKj+tNi!HnJaRjY7@AXW>@ z%SXv>zb3elJ*X^Avi0TyTwtZbZ9RN|Kg0`8+a;l|Am@?Dait3&Ql$5MaL?J48R8EY zlzxuncAXSeoV-?`eB5~#b6N))iJBAwTK6rMDU(U4_;^#WMRD@c@Qv&yF|yr%H^PNM zi_O&tM7I4#=YTI}yKB&jPyDx?#R#&xXilOmummCSjaV!qI7W0ZY6Q?RDN+;|O88kw zEZd=Nrc0p%nxzZ!lI-Z!pIHy0|ic z3=}kgF(w+=x$m27*yOW?#V-qq_Geptj8lgRekSfyPL5NXh56D{_=oh#_fqW1d)O?} zvq2(H>m zjUWWA9*D|hh1a8tjpf%=u^ntBgORpZc`AkqxZH*GlkPu))o8{2io5$Jwby-XmY6g$MyGR7G{^zRe|+J(m^r*hOVnU zFNt#QwU!ifq!T?uDhED<09BeQ3UExrSM#=z3k_1c50t{on9B{5MuasFu{Ly!`$0F|h4EU?tGzg#!W zH!n$?NJO6sJzkOFoAi9~ZEH5+?K6%bo7%e8{ui5g3A(XR;L0dsBl1~fd^E? zaJp+dU06!#_=tKyBx*MJ>__eDaOeXM?I%Ux4{8cPvf)qBMLxH)lsMpNiZqzo#cy8m zW_c`@#jfEw@2Og&m)xDi4UDT zd*}CA!ThRCvC}h)aZBR$&$VTDGNXAM&0ggSXLo8peTar1CoaN%eqCrth4G#wRghK9 zchFTv1Vpx8DvDkzqu0v5VXnJ*^1COeb(t03Y3(j5dkt|hNWl?nC&2j=H8z2TuB=RN z8bMx#%@GGo(lsEW*nzL59uN;-BN_e6lTi?EoVS?&(2;He&6#zpxI?7MMVG(y z(Oua9N6{`qGm(ceQ{UlYPq+8P(-2$ht&r$18(8CG%)xz(0gkerux_FVqo1Asg|1!| z+S;T6rMlh18gX-~@MT}1?%P!~gzd?^9Y%7xv#4UH6gi-*Bz(yB0%2nq7vQa(njLuI z<5b6i6hi^DNk1eL|f}oo)4>jaKG>kL*S@M~|)(T~KU&cl$tW`)2OwIvSm@rX8>bEelObukO~;#`?0gH!kr>O2pjn`w*sAWl99G|9lasO-NO} z8#uU==vEiV^Dlq93%xFUoQp^Ey&f3kQTKh=P_V4t`DOvtzgd90f!uGJLEAUYdB0eb zX)T1BX(%V7Jz&;m9GftP`tfGgHoL_;8zuX6?mG05=4u+fFG1g-Sn~Z`6a#xt@h_Jf zwIHVw-m4o$`O3ItW!dg+S-uKY`3%y{G1T{Y~*f*e0~V|Bx$ zMR$Ywi0BFe%miYbx3js;>OMYQ>@Ppb*n7TNHjx$V-4B7vGM$?CSJS?9*X9&n2dzKg z^&(~Ye>mWeolB&3M5H+^OhH$@nwAzsf2_C1So{@V1$5|R(P&gAS%MNy+|$4xdl$(w zkeoGf+kz7UlWj3<5pMCqoSB$^!^mOA7DkCk(+kQ$@R87>qmHB%c=nDurlPwoI38SK zdbMr^xrX7{kj>P}B%lT%agRwr(B+p_kwd$zk|p{i9HtTvMP@G|u`9Sp z(P3g6Qy$=R0T1gjN#XfGuI@dCk7-M;kYMZjz5Za7=rH+zxOX80>v8Tkjv_^c7;PR7 z4=0bek1UUuDcd}`xTeY7)~QGP>S3t7Jy7kUA{L=UpyyiMpflp{p06RphMHodxuDCacc${o9I|b(rn_yl~)> z5I;?qXIE=WOuiEm=K`7?SLD5vkx^W^#;+$i z!))5a>Ma~JheGfw*})H%6(dp)geSKRe3O?7YNBrfoeZf=8;2YN%}WvD(qL-mpVR>j~O5E{d+=} za!Ixio1;92-)#HzsuyTP%|vU4Zl6BkTTU9oiVmYiE;=mgDy7s>xij)Ce$}svgPd{M zO_snP8$%51{*O!(`^W&MJ zv8d~AbtF@dCnJaPi~hFr*CjYbD9h+`uVu~^6= zhPsOn;jibb#?2ftT#$9uwXMjko-OI@Y~&jBKHqH|*Y*-EC=_C6ylHu>$S^m$-T6aA)Q$&k~}eky(m>nO*y zH@1*%7%|Fa59kh76<__4V3@yheRu{5(@}fB4oh5b=a0$`*m)Ur z37x{lOraWr@sj~S%zU`w?Dm$dzG@R&lWyqWN-+`9Q_(F1IY2aWW?}e7FY~-d$U5a+ z6bR>J7@bWhxz~+?X~A?Y#cBFz>WSHfU<-nRKX!H3#-M_=e^iE=ZLys}?fo#9X2s2P zTjoO%5kIfx$B$ev51$#t#dDv2bcu%Q9F$~h`g5P!5M}|qqx)|P@~=nQn(f6e1^Wr6 z88|KffoR}E2@ziech&Q}daBJlEr_hb1=YP4&U<2X#u}n&b?S;JhQ{&fK7KLz3l>M4Cq`3MPl7 zg-XN3v&92?U;qA58n}sNtKGoW$~KmSP9l;SM`*-5ayM7(_qYk*31wggBCp?V*3vXw zVZ=Z3p8&1dNjOcj-+I!yS3w0KyPGpOn*z_>595q!yK$Ni^$&QdS(gg!4TQqq7$NE{ zV>@{{%RR-_9U3`+o$UC41{5eKOVFHx`hEy|h!I=&Gt8AV)jr-4&KhBR9J?Wr((S)Z zgVbrs8BLHd%vP73V~%buGhSFM*-tsboe&L;t(^l2))1zJ?oa9zAGn@HN4dqHy@k>* z&-F+))*SR(a$Wvv@B$s0@&Oehm{8#ZG6C~Ah$^0Bm)hA-rR?OEIq!S3e3XW1 zWXadoPF#3kuc>MLuO2?gs`;g`y`Bg|8l==^*9dB)2#zU1G*)nG_dU1}`t@Nh4r{{% zbI^-PTdMm^14unU;WrZ;Q`iM&xi1Rra7u`~0D^CFu8PmN0te5<8^YGPlkobZPMelMfkkux2x& zp(WK)_M)qGvLj3vy_l&KQ*D_De`9Cs0u3a)Wz~A&tVRMbwZbh=?i_a``fh&}dPhbV z&VKmx*ME9=G3~G9`aD4NivGvJh-lx?-N@DLQfs40C1(>_4cbtJdPa(4JrF@L5p`Na zUIhb`mMmOEj>X2cVYb_UFfzHNo=WR>Z|6C``h_fvImv^}kx{jfnyRurj1Q_qhJ~^K z@|YvnpdA^2Neg%;rTO>%g9N`fDv7sY_4DR^_o|>JPxKW7GH(-4ATV0Xd;Z5wP7g~KQ(%H1QBG8 ze|I-bp-~Iub(aUXZ&@llXgQ{oc@#xWUB$c(yDNQtQ2JQ9Q?;s+e9Q6wBa} zf{NA_d+wpa5AA4D!M#7L*M*H$23o^$8Wlly8qe~&O`z_LdaPA>B*{WB+kn?wKVQgM zU&?;o&S#!;<*;OAT@wXtl%aq(0=r`!ml9jyCw1*Oy-1TUblYA^(^-feHVJvzOU(sm zX~={DZS3~-p$GIRC|#+J-DCw}zlZHEa#vG@emm>UD|sm@nzCy5aNX||5P|Z+Jh@^9 z4*lz`V~kQO{Q8ID6}k%7()|aPjZfk2qNL@ErYo?rvHSh?PuZPl*r0>aGyX%|`%S&n z(+Q$sf+uU3PfNzWG>LSxpS}e7WEeV;?(wV=22$W+$hG-wnO)&l;6gFW#Sa7GMIasz zC!nAySIlC8#szt_cE5k;;YNB-5!yf*BCH<02ZFN?#feC5C2RuIfH&{o)S7(#$2~%= zrs)nN!NwukzRMJz2U_Po>LE;I`(NJ(nMSA~I1j39TDCvW*bKTMmn?xBD;_TnO#X4P z(nt=dd9-Tt*`>;Yv>>p^)F4Q)80vqnWJzt}3*myb+Y%=^W0Zeq!_U$D}94v+hJypIU$>CWjI>*>As zQN|-=%I9|rdmXP&VJDCyh0Yv#=Fkl*H`>R)VMs*KVPRIQYdk#>+(TecyJg-d!<62D;bUqa5}*fHh(*wXc+ zyvs<_nFV?MXwOYqTdI=U^csu9w;wQY4@{B=x`}vb2!i}3|D`#Hyum7HyYj&SKrG$h zV`D;5y8bTXG0eH)Iyf5*;Xlj;s>V|jNN6CC-S$Te>Akcp6K_c;O#FFA>*lBhI=lN~ zYsdcBBQjZWy+mO3l0|SZ51m=CplQ|YkUWdyZH)6}XZh?EOhtOb;k#)L>*QDfGn9gn zY{xXs&}44L6l?RS&9C4w8|Q%lBpJsRkW{HbvW=W4ACfQ%kp2#QTQ?doke8OyG3M-6 z0L8*wiY{?Y5M?ds!(_+S#Nc4c`QBp#5{t`{<+@&3-h-lTWx_q=MPOf{Ax)LA2tw*^ zetK8O24U%lanFIgRQX(ZQ^0od%ZpEyMv3mIpt}lV5OA=L9DGD~F0FzA=%SN@$Sm9s z)n~8qOEU$@OW!{T9n!y| z%P}8%!2>zkB&DNV*Jt>0qV@f{f?6`x`U`aI9V}>LE z0tG;fiELOg0Cnw{5jdrmsE{GQyBV6(@~-AMy(dD5brvx%ycP3Uz;9v-o@MnAB53j| z5JP-?_vn$;ypG39&Rx-slMZVQElf#YA!8if$1NI8U}Ded0C&sb;4A5i$XHPb+!i1J zXrZI*(s3c;f{XQ|A@8(=jTx#-^c1_q$cJ~hlc2*q){(n65avN_RZK={;K_0(8PX>? zs*jl=7fwFZy;2c-mRFcD2ZaY-$kiD4qhkh}j*0JRpHd`;)Sx9y+ zwdQWw4t4)REFi_hnD5rZd-(4_e;x6mDRw_IusR3x`Bf=lt!nt5!AJmXa`zGdW=D~S zh`PkGb!lRyrCnLO<}53DQCFBCM?-K&DqndJhsDyH#_l!TDX_^pJAsSY8hut`SG>az zRp^VW)a^@Y_{GmPRHrmH>ygKxVticArizQMX|rjk$+_m@62fAAdU0DL_Q zrPHHoo#P&fdwT4~pkc_|&a?zz5A*GSPURhc!AX*jgiDZlq&yP#U-X2-k5lDTOq_E6 zIlHFQ0U!AjJxO-$#b&Yv9he!#p!K}HB=%%m8U3R>#}X71=M&25PWN2sQSQ9!8uBja z`PH{tSKg};@Q7^rWq$yt)B`E-)Ds%|kIwN!YGB>41ab-C@ek--gUA#hte3R4hitcs zwRjCB*jq-@qmxSxp)3xEmKX^GzYR~rCrb#EIG=75x%IlEzt4w#_E;LpAcg5Cr9^Zm zvSf6pVl>7ZMa{)(#Z$HLrt&78veD8$mWf7qfJgS+MVY#tIF}=NWyP*8pO9mSHdvmBn15XoZF%h=;kUQ)C4rJdnmf}@8s6c!&u>#tt=W*kXV+~6R?%-wC}on5D7SK zSP`z|VV5vO_>b>2n~(nPE$JRavU`UukAaI^7K|{lE2eU$qYDh9O}Cwm2Td}Uz46`% znR~BgCW3F_u0TAsfkdcd`Z4wBLSOM)zI$`sXT21=a+;gJ)pQxar!>ty!-d^M-0ENn z_T;h(^UzON*=rcm3H5dx^l@mxgL@cZaOC=!VMox6d^{_*<(h--o~dVoF<(`qb%~9# zlbUYaRERmgf%H|IWY~Y;_W0e;*S|%@=B05z7%pFn@@gWxO^x_?M&lfji3f0B)DNvww^fw=PW_lW+z*sa~n&V;22Z2Y8Dz++` zYV{e`6JZ?^vVrUs0|rg8kWnIB8HhMJYmomKyzpr^%&sqI_D==WwcUixfRVpWMKyvv znT|lTpipX5E#G`?+rq4n6wgAj!nm0C?vBTA0M5pKaizRg!_1Dh#)CB{6Kk-V_K1jJ5OIv+^4F$$SG+J<>yOhus&0XGa zW?a(hyCMy-WiR^niXx3=LhY$(wYivGNsrXPM(6BOW}_Y}w`FnZVOySr*px_(n*_Nh zC;IgqO;Km_zT18UmWDYl!f0fcVY8(_@E21ayG8GSWkQyF5sg*8Sjs*gve;c%!KJPT zNT?GCH6Ic3XpA+2O>pwaY%=I@wETbrY&9>)sQN) zB|;&TmhJGTujIUigwz?$L{$YNT}eJnu6WBdI0y0^1v!KChJo(EbDpO_Yjl3WgJ_F1 z{0v|cR5(%2KohKPLsOU;)=7o# zZ0on`s8B#$XdI$FYtlb1gMigEUD8NR?5(`ie zb@=tW_?;;;Ts!r40d{ot(Zt`!^p}zPt}7N1B@_q4a8E-0fKyQ>RT4@!UZ=+Zdzj@c zdj0Wh_>z73s(O8y<-A#q%@{-`Y(*d4ftf~&Nd0MJ4V}HhWO2FzXNE18AX!Fn=klLg z+bQRORxwJqAcfy(7CvaEb7ZxqmAwEfw{-*8F3tg4%T~@352f!Y1zuRi8i)v+Y=iTXJXs7F~Jktwr$(V#F!*cY))+R z=Ka>X>)tex@+$qO3hI{OkxPVNM`ih&}@H{YDE{X7$M51xx1icN5Z;}aE(JHD=RNOMa{gAuc>`C+E=~lzc z3SjH-!ky^LV{4~}2m{-6!^NAHDF+O}P8>Jc967IY{K^bZ!R}r#RoibLaf;c;8{Oy5 zO^=#4@gm1g<%%Wt6)86_=b%t6`PIdWT_;?>y~z&YEZXsjffn5_9oR(7XUdMR$NgHj z9$FX(&wZw}(ntj-hhZU(;XGOSe4LMOUkYZ+`Ihu2@=xj96j2pIpMZb}J1-!UlwH&^ zltz+^JRg4HQ4e#X+PG~0dVGrXUH|ZjazOLBFq5#dcHwl2KDre$BqE5r3kf+yVPa$A zArYlqxTw6791haNSlkz0;jc@WJqzNdv7a3aHzx;5YDyZu_QS>|>rAhpxVc6H-Nf5I zGN2~jk`%GNEv82NkNs7QcfWLVC@8BrduFH<^%&@@5In!e7HPy8v*xk1JrMtZ>Ty5g zcYeVi3C)1bxsW`gMw_AvWxT?Ps1`8RHU9lu7vJg0`5I-@vPC~2ST*$!eA9N!}2K4Fq4yy5GrMCPs5AUc! z<4W~`cZqeg*981C<>*S#qx83XmHa*z!c@MAC5q;H0ThJ zPPm|CP9{vSj4K$Bs#e7rs+Zdc6JhkyeuF8)>yP#ZiR?6^JP(@-p7#i1^x|ckn%`%a zPZjUy2LQ@Nj&KLio8?Q{WDYQ#Lkv?A;Gu#`Wgv= zV;C?I>V?Ud!yY+?5$`;167pOW<11WyVpXfvLIxQqJzTeNS|`O$0!m3+SPWW?@UU)> z9)f8oCd3a3NAG>`oow=}A4c0FkPd7EM|!*?sl-tB-UYFv`W$Q<<&uFxi_VSZrzedr&s6kd#LyBi3Ff!Z;NnLL)*7 zuX%bHY9t>gJOOes4(0CM%s(G@9RQQ&bz6YfDGbB@%YbR4=_Abx1k(N|GXi>aH9&dc zTazY**qg+C#n+?T-}0!xWw6m*GB8yM^)CuO-d=sr>vglM<-#}SMzkk&Bs~tRE zgWd?GZ+NB4Iw_22*g{cyqT7apdjKGwoCYWn zr(m9@(U@4%65In6PbItI?9(|78ZYr+l~sZX0_Pw3^YqOvA*qr?^;#LqIalo9HTtB0 zi}4wZ4z`hOJxAS=qdH5B?ZCZ0-s*(o^)}Q5T1zcZ66=|owVjGF zaNB0To!=1S(x}ZfdF)$fEh@@?;^r;0U^UE3hNwN@IZ79D7&+z zL5LXBijas1On%?&*M8CqsHoo9O|SSZdKPtU>}>=IoWp`iZ`@OT34mu_H{Itaq7gK7 z|C&g6%XH_+8ln~H^+EdxUH1}ULKmYc)#lk ze^6KbTw9H7AZxya^`%w&i--&(r!8fY{gE9i`r$c9=j$M3U@^)77@xfolNS+NkM5#M zK*_{HTp^07ePBJ%1W@XmzU@)1O#J8QsbJ<8!aw}>TmLe!ex$?=X!6J8@Ry_cIE8{} z6?F8b)MKvj+xOr~q<AJ2q0$R_vh=$%V2PLhKjQd@G37dY%SZfR^MRajZQx(Q zInu;H9GHBc1G-(XyL-JHi@M7wWiZh2S6^ccGGHVs%x;ad=({I*R_WVuEFJIExY$HN z!J2TJtghR508Cz3fnlO*Jk1CYN4lO2aaF`^cz9F_`NY3&Jcqm2v0&4FTE;6cn=&H$ z=aFNQz6P;#S}1(<{alp(o_7Bk`^r%WXNkoOmY zD1WdM+V-Ct(RP-}@;9L^sNLMX@WUXgPdINBPM6LLE8&KSBr)>%oK|2ePJO8Il@%mB z%$c(v2NgYFGl4Sbh1kUs=`yDmrkDRwS9*|v0q&_(z)ncm1HDgrI*fiCkSmVsY)hSG znU}>f5n3NRo4!Gd!F=Nx$8>skdY4wo9NahV!g4JjOVr-Ge zebYF7B0hO4+@te;JXsu#<6=lOJV8cZqs9EH3X`48okJyRQo26FpQrdf>);$DPJ)A* zZ+Lx+eHQpU09(^mMzP+`OXJg_Lvb30;B+Jc6;$*o3ojen{QS*#(kuhb!sSjQ>uv8%)N zgLam9HV_GD@tRX!nU0uDwBN)hTr{2O5AC z!aV4PpNNHOK@8_89)5t=dpHl-esQQ)o;;Oiw*#nYLa#cJ9$9HcWnns2ygs=oBB6OS zxU8pIj@Uw)waOBZ45zLkiD1EE*PFo$28S===acymMYLEa~!;h7a=j{ z1?q+z6b}sojcXRAToU4q(SQC>Op$;>?A*C#0nIKIqd!x=l!@Ft>`-;f8vwD*_A@+0zH@q_@eU;M5g z>P<;1%VjS@8Wxra7t7*gkY5HU%^~ys#&O{$d_Eh624HKNE>Sk#0DIuF{D$)YKUHkE zcn3D4neiTOYBK@TpBbDgx)fC#Z?7!WffL{b|s7ZcGUeFz1GuY0BTm5jw(jM%nO8 zk;FrH!bMTAt~6s6v;b%9>~tqb?cvGNK1=CAIPcXs5@Se?0SvcGZaA)slmK4|h!R{W zws`Y=AX}U!jG(-Ir*f`WW6~T?S%wA83PTo6ae$z>r?F6?>iw+#7!QDb->Wp<>eEbq z*U{vF0faYhv||U|Pxz0W2s@X>m4uNCz4XKpMLZ|yi#;Y5bC!kDF4Ei+rMtz#$eQlt zaq9{h_+yKB$V$T!(Ivl{WFnNFn3qLoLV%e0dysdnkSvr4NVN2Eu_@TA)P-(l+0Y0&b{B|Gq~$dIb<4cCH$-XLLB8tsLgLwd65=tD zX5!-3BPAo^MGV+1R0eFWC|tMP;rP^UDBqMw<Vct7C%LZ0Jb~%MeSD-NJ%xV z#Gaq}aHu8@G=D5lpna(51U8+a<#FJJ^y!7%*hq`%@8DnrGqy6FqC$NF6E|8U)C=9Z zZN<=f0vKrV@V5Xu5Lil{tNZu-EuotH^KSOO*_ho?=@aZefI2v(lG#X48-v_(XARld zRoFddR}Cf=hSZq^7(I6me$-Mea0z;c%XM@=pg z12VOy;KzN&i`-T(s9kCjT&V6_u%orWe}>fdMDdNXSdf5Stqvq9HO9NO7IPg@Yi(Hn z#Z<4lV33BV0+5DfX;9<5U>kXYXG*Y05>X=@m^X%LS&M)VChW>~-=ER+nit&h@*?o- zYy)6H{9rKBX9JG3U6Jg33gX&=)#r@*s9!b5{sN79ww5tcvIqSlM4K6`t_;_TIiXP* z23+5ODFuLtQGNL^4>JSjZL4o(9yW`hHC?!NoXJz6S^zzQCNg3!s^flp$;E(wUR=Ad zL(Dg}oEcm)=_;`9H!Fsm)vv-wopKNN+k4T#LmIYLxOE!_&DiQ^bTMVz{ghx1Pz|Wh zi)jHOv0X@1JhhW!immVt(T!|YD%pYxbZ_H~%>0cGP6S`zx{>`6hJQmqGpE-N+>?XgrHuFaPJ8<0 zcL^#l$Nhk50Pdc*4+p)WZ$G3f>CVGN)DEzMRsrXu07ID-d1_EF9OedDbHgWHYiV)M zSrVaCs!3M&YpEg9<^qqsEoHajK1;;<;EJCHLgmg0020VX{HRe)u2`l>sfP>88U zSz{5YKwaPHYA)PCJ;TBEJQU4hn9CiS;S~cbe70``jFO{dJ}fEsp0`6t+)P&{(E%u& z1bF#fgCZ~&NNl>ZB}OJJZ7^^{Rs)lxKAtOvEN=7gqVonBGlW5A&{J^FE}TxAAb?4{ z{&Yvjprn0|lVyHKeEpEHU72I0gy$mlP+s%Fw(}(yVwzSz0(-|>f&Skkyf+m54BKj& zr!Ms8%St;)b9&%xKGte&>(TvvRe%A&^cl9>>;C6yZwW;41%t3)4+xS`HQ}5{Ii=2= zU`;@~eS-}JC3gD&Mb@w}((*rG>K`WaT_$s5Y_^Ud+)KmK>kgWkVj%G;Wc{|QMEW4dH8g`zRoyT-sD4jXl6 z@6(^VnMl^#D01MvaZnVsXHxbH^o%!+pE2#d1f7VoVMe%UP7Y#gZVNGML#9~6>re<@ zONXrX>v%76OZ_jrv^9X!EPgQdb6ouJB+RsZlWpyTE6(nVoo$}^U4&4Zi%rxP+@dTL zK~oju7kdUpF!?-Fruk1agq4twa!9d1ta zF!4_&L@+AcL{fljerKkkuyWUo8P*u#*h$Q|X*sQv!u2hCPZ`8U*)b`eS#09Q$Ou7M zOWt|c;*;aXV@qF>Tf~V8(R!$^Yen0W34))QMcSBQdB(~Z1gSz}b=Q2r?e(3u)=eqfURoF9WEDv9hYyVofO= z`fB4%C=>uw^P&QUh3k#6qwHPcU`=y&|-dhRBF;gYX1h z3W+0}*dAt=XVT68B&77-)4QY<)6cAj<*o4NECV1AMCuo#wX#;JU{b)Q$iH7BK7~$3 z-dDwPC&tz+8vj8`b{0Z% z6**kBn^2?v=nhAN{tsI|PFi$Ihnsmv7`dKmVvix)a%`3DCpv1vo;_AJ+`N6~CV~Go zbS7Zrn=n%`ks9jj=EmhQ=lm&b{$t(X`s`}u>6}l~SIDqMTkeR5xom47@SSfH|}a+%h#{s2p3>9|8#@I5qgc8=QnO(G*T@~S@iQWm7>d1 z+-%&paxA83mwW+GoHmq-GZkoMgEFr!n#QWRU-J~pFv)~nO*S5&>P7(?jEm4hV3N)2 zh+y^SXX;D0U&rX^BjX^h9p?nbC5@-E)qC@9ucX~Y=8L;ouOFYWV2)MlZ)1RtQ6P7RQI0t8sN|C0J@*`ya{NyUT92gs;t|6h zd()5)`?U>&FDIdQyO7`~?6e78-gAMY=Hks>bbTco*G?X8B*tg5eXZ0d+#e0hA2~AD zdk~7Ly4<|kX@y$#@dT}lAjKk=gWc;f&yr*!Xa$70EZyo0d8I;ffXq+!j|*N&K%dJS zK}TI)+Uu*CVXs=lN&>`!ebI~JUBQv)bXX?x)4nK&GSA;TlM#7+^Ydh~{&X<&eU5mb zeNuwHJRqzO0pH+&`z*8u2YON6x9COVocA#tkL6$h{2%bTN5}3_Rl(Ek-&9MN?CdV6 zbavc>MF6T?*Kr{&&z^|(1ESJcKZXaCEY^8yczP67uDlpOAtBGL6yN9%VlBZ7bZs?< z&<0TvGO>)n&CA_88g-tf7*uaE$rPl3AnO{%RDddxz}BDaS<#k|H7VyunQ%sHCVvvD zsH@QL9A^t#p?-rJAl5%;zr~zHeBn~9T0}vIo7EM1OdJHVkfyQRqnIU$ymiYTI^l!j zR-B_N@cL+d)h`cY;RK83?T$#=r1;{js_9v^7-kbDb0H*2T#zWUMPZXkhojl48J$1O z=K&3*85y((CR|=aSDnSW-TZFVpLJ z_RftOczh*@teiA%&_25Q^51y^1=&ER%%sjPCFS=Irqsk>qo=(WnrYJV@nfJVFe*Vw zAxTMlH;q5IxihV;NvR9ngxvoNY|Pva1OWL%&bB;+Rj$sm)XKx%IMe+fT)O!p669;_ zRD7S}jMwmW>B&FfJ-VBrwQEZX)18*3L~IJi=W6#*8=F#-1~thhBx&!;gDT^l`gppH($5g~mz^xbEOK-O{3JDB)=KRO25SHD#OMAx*9?~T;uU`X|c)zmF#d-ZbhiK+?->b3%RnQ7y(H59%Crz&d5+FZaGr_ zQRg3Xsz#<5%-GnD*q-6kh;2FS865M8Q31ZNC%m`qoKD1X;gUOd>&{($Z59wLN9lM!;Fb2zkANkD%K{!*MYkJc6Nx%w6oHIYl$MaOE>W zusCN;cBIV66qIf0Z)PQEg58e-yfM{#Zl&l_TB4}^5yn*QSmy^!Qya96(5dYLJ^GkF z8N0YpcvTJ)W<8(5T8u>j->*58{o%ipyI)r+y}uq;(L^Nb;~t#JwaRixp8)&T@v&?C zv%Tw`dc9ra$m3_7og3Xi`X&95&VkTAai?N80Lo#{2Q}UXM|&FTMp0Q<93x^m_3tFp z;0yB;BOR>1j_1}jwn-0%w3+#Im37-na;zn&8{`E|ou#dWy6ZF3z|FWH#yDR-%s>6M zC;G*o=cfKd(h4HCQGecpJ_7tjvmTdjO;1uFADy!%-|y-hjx84Y&`+*CH!WOTC2I)$ zQkngnFZxi?#6*`ve7ZN2t=^c_>pI$`!oPRU#5n#n$zUwTl*R9r+ta50>_6qMt@dNd z(-x+%yvYEMeT5Z?&akFRV4=V9QDr~5nKw^Xlcl%(=Io#`H6xpIf&_31e!f3lCpdid zvWR;XuoTMgtt@2pZL(JybtTltATd32K<$xroh zGqjQJd1Ne$OCxMlTgG&Ahk>*eHoP=3W=38vIpigdFtr9;Tw!J*GfTFHDN^mCMnlVu zQjWUMU?_iQ-u&$=aR^W>e51^!uaRh34E7kgPcN>p>HjQQtW|kf`zt@P=iZo=w4lyb z$aq1f;F8ccP|c8h`y(IRO=O4b++V#9G{|4pZIYp=4tU^`mI@N&mAc3q$W>q z8!mn_3?(`ae@0N`l0YQGt&oL9T#(7O#I=2wz;qeG6}ic;OegMQ-ZTuOerRt;8{bFV zb02wLIX|PfM%|j4sjc35DAS>BlG)CX{bSA&VHw9y8zByl$6@$7SUnAQ;7|E1e0R}h zC(rs>5}qjuO%!ry#|>oQWzkU?aS2E4gm6@Q{WmH2_`$6vhL*PR%ZbYv^d?G|PLi=H z89#Uh^+K z!7C6;I_R9sCvaZkaLlMJn6h-&0ZyiS6gu4IR-`qX7_TamkRyIDpi9kCC4NI|e|qMo zD7<_*(&EJ^-K>hPG71Y7n1Q${dKME;3LsB(HL=f{5&8A-52=p;g@+NvcYIKTAiWmt81+Qm~}rQ^4Jk`rb>lm6uz9jl;ZFV4%CmS_z*lK354KK_Q|F8MXK^pimReRGZWnZqRuV;g#c>bCJ>TM(YaA0F_*Z_xI+Z zbLDJz6G2f&gaa=G^CIc@HnmJ?D4=Li($4|3Pl=ITwf8+xO(dnxWak9STRU-Fu#i9A z!fCxILrFB5aDem==K&HHp~4~k>xgB*6%TO^o&!8kWN50w9#P?f^c~E((SmrZy9nU! zFah3wz8Y7jRYx1D8k(G^+EDU{6;V6`G)Q>mJs`7C7*=3EQ$U>*I4=Zs1n|LgEL-b3 zV8S})&`pIUf2pC{Tl|f10f!7M**sd1Y#FtQ&~_q&8Ga~IFP<~}GZ+2pMe_!EiT5Zy z#p(T`&mEi~F$`})0~g}luVE4k8)mFs*gbwgXlQ6GzcT2=4cFSSrRLgxZJ0K8M9dc5 zwL7H;Q6!fTr-MSprX+5O4j_|jaK*eFDRjvcCb>mh%2(kXT-$8Vj&oU(ho%@{`D%Yn z-VZX98dhSnL9YC%08Y@Ah z87u6Azj}AkJj?KoQ*ZkFD>!Exr9$jMdJmp>;CFy@EqZLiXfhsnmsRYvbe!LYiabTo z$joRcUPMSO9FICV!P&Ti-?Zpooub+oJ7_<8#mfkc?cLeVbIPgK80L+k_OBoLr=9_3 zN>0BtFd0g>dR=Tp0J8Z7G3k|#NT@gWVOz^)`2j~$Q^i9EGj<3Ee-(+5J<3i2UJ*7hbb-f1zsbnL%@>`;XhC?N2VAU} z#Q8%GCkr|ByH`S0LvFA>naT!|wWXaUELA3HeVt*7qZtVOK0_j_t*>i4a z9vxIk4(;&lic_ykF2F@vkfDf12$6SwTL5<-6D5w|vBFx%d0hsW8K($eAq5 z{!OC~4LJ0_A^VA6c%k~PRkyxHq*2v;4Utniwh)pjr3Of}U@(xNOeISB!OVBn&7i8q z{`K%%`tvO0ec9IPbA1e;!>NaNHEQ&ulK<8*vgjET;ilDWf@x>IUy`e18a_0Qa1W*fn(Ie1Bgo5AQ*GWjI2 zx~xUtrxH-83C^5g)~tR8{Tf&qj25duup(;Wac|?v0#}{D!Id*sLNgWF$l2=59Lr;s zbj_tpJgGZioS$z5C&W1-1V%d8l08Y3cdAK)XH#&4-ezB)tUc3OTFO^^>557 z^;I_yj;7tyKT@~BP89@k!1T2-`WZJ_TekwV?l-_)hDLIiBpIH*^oaru^Q;lxAuDA> ze_-u}eGIkSnIv&00a=}e_A&8%y*D8603YoSvQzO-mfPgv*Q5FEc6MfU_%a05jAU!* z@${_v0i&x4Fj`EvoKu(OZ<5nO>UthgXT+@bNcGBI8VH75`e1sEkmOI1mAX~gMgwx= zzFdHLU(9)1Zx}1VSm9mz3%8=82h3b_)?_T2Be8**LFjDWelEV8-ZD~H;*R`s7YAWQ zBQ0Kf1|PotEk*=3tzJ8oy>Y-11H&4{_9(iFW}=g+Un+O}FcR)eFTJrqut$d(7BJN?SRwLGoTti?jdJ z<8kWc1C|C;hW~(j%{@g!qOvmxFKIDHGwrQI<$-5vUZ&-iW_LPJ(B4whfl0m;>SMj7 z!(vw2v&435x9EJS7bSaXlvRm!w*kG>zB(`w{^ktM>~IwyPS4@oph_aKB?o|g%~PD6 z9JO+_a`Mg|EE?3k3bORPc5Ht@yr#`C=4xV&2v4L>kGysOPYaVW#D`U|*n&B@jdHS? zQA;Qs!eJO{!5e(+x?7&$Df*$@iqrxocxS|m#i>m{#PejtdwqtR*Re;olCZaD&y;0S zN8^WNKsP>>6Q7|IlNE_F0|!9I88ZG^2Fv$=gJXB|?|d{*s8_v*Y~16|f5{XiX+nmc~bi|=OY3I4Ei0$3S$s)8Gz<|Wz98Ubu6<;`x2#U8jsnVP(E>C;B(N-yI?<)3BJr}q?#Jba1D z^DY4ti{rnPbCeT-l4w-LG=lK-q9z@3VUaJCR#g>6Z_3}&>g;~bSB`Kqt><|ckSh0; z?OXhFu^IMzrLWF{^U@O(vMEy7Td%N@Tc4)cJFXB2S&ILrg8jH@S~BGb5RKlR8!aIJ zV4|46x^8xZYyf1o%AmY01OGE3OKT<(!GX9mU>U`%{G|#)DO4^g)PWRG0*`ug zgw7EINTKC#C>cVZlrxEuq3?OV)KXl9OyjmatQn%$xUKZMz#GKks7oQS{eB5+a>i!! zb@g`5OhbHX+NL>idD=z;Wv1M#24Xjwk#I)SUH^=0CT*FiJ>Ix4A#2%^a&pwh0Y-Br z9hp(YQqvoq`etbI%d>03+>UTQ{ab$D<^({tvOS9*skX)1-bZza-d8mIO$0~Y7}^&L zPlSw>o3sP3<%tcLk4Xc8>dzgrGXj8Wl<>{Va}W@&d6LdF4<)T&LaG zf=w{LGJQsRre6j)$1B9)C`ciRY^W*1nN+Z}kh8!h7*9`{s!g7#C2>?z(SilTUg!Nb zR*fpoorAEM111EW7t9GPAPpx`Yx+=GUa&!{t8AC%0;0YE-v8S6+nq{oG-z>P_`$12 z$!HVj5!^vX-Q%U*(M&Evm;sP7h1wehV&cW6WiE#vD^cRLA5>dhH%aF_Y82uXv>dB) zE#-UAI1LMl&;NiGR~l;nOd68SIW672_6rgBjUV(v?)<*g-**!3<%3$#KzEe@?fYV+ zb}4p}1`qaJy>;1F3Hg^q4f63&D0$8mG^fxXO|(+KE`9qNUM07yyZ{aKfxb#vT(-R+ z)4V%|5YyT_hRwU*T0hYI`iFbEIbl6(pgChb>)L;sS%eJ-k59G<$rDec3Ck1D&J(XL z(EMm|U)I2+Nb4?V-rS;L3DSbRehJK?2+?9+&2YM*j=%;FpWtlmA!p|GXXlPH#FF-o zbMvlqH(d3qMjU`6W;)hwGxoyu@-0b>p(mxxw!xv|uxySR;8K_AyH_I!hgM`cnYs!- z%u_FzN;U(!$K>GaYUo=X*$$0&{UHq<=4Y3IAx&?(?bm-Uj{OM3l_>z{Er zfOTvC_-X%mB_;=B6Fr+;OihFOCN(bO9q&;lXaKc*aSE?y#lb#;2`rqKkS5wGxM%0} zSnA*cBtP?2g4_3oIOHrcB&{cN2}x4GHsD30y4a8!dxKuth(uuPIS`g(9eg3-AJ(ih zD~=L}H`$G{>#098rmJ-T^I{0|!=NiGKwfr)DVExIGf!gkBq`uTW%N8BL>2x#pXodw z8%j0m24eD1!(X{AJI!AiK0A#mI}JNDaea?Nz;`poewsr7+vfKsD^H+2BMVO_3y*%f zeCg(vbkdT>DAaAZQS2OL+YjV9R*@ zhPeA=!?8Nm>sw>FuuLx%qx@6?a7u4=o1}+nZ|SG!4%Vn&bZm;Q%4s_}V=tDdcp+#A z$k1uy5A)Z7=MNM0L|PYcajC_!$LqUTuQ-Eo0%^0?!UAo>(!>g@XK2N@zib(lv*;?) z!M11GSz_!W-&q3d+DCC8RUfi5)U}zGGpVBBkP&DfSGOhJ5PZMS$u~s-Txr}}7&t0_ z9EZ8wAU@P`r&={X(o!aRi!^AYu7S757^X0Utiw2Vb*ni;RGF%^`(A(Fdr$)5gS`O- z>$HP8K6EB|AE5Um8F_Et6nJN{7X;LN3;fGqME01 zRMMsnyY`pdBWwo-0pCt50D?egrLT=R8-P8CX0)MI)N)))P1TAR<^aj zjVI7S1zZ-1Jx}b^2DGHKDr_9oJZaO}A1sN{(LZ(vWYMrf?QXpIvF@?<-G>ilm6n(m zWfhG01V&Iq5m{iGV; z5rWhs$I>@TnUZ24z$LPc80pN~ft{=z=#~u7+!}WPHa8si6gc+~%p&%|^9hP-CD*j4EyX-Ctd1|Qd((Ef05^KdVgsP%2T!O1A z2CXu+G34xoN#iB5ouPN)gUFf#z1V&D3kh?h?_#emNeM2f~@ z6G1xfPuj=FH%rR`a`~yWE8eC} z@-aH#2wjSufC^EH-G^$CV(paB=}btg@WIVN)6Mrm%XN3VJy@(MIVbU-B_;W+$YYY5 zRHU%b1Cpt8;^{Jy&)!(KLAtQx71FF zy*ksU*Sfw)z1zqx5-pfGG-phPHinccMp^*iJfyk;UU^AetS50^&D_UrIm~j&5lv+{;8@X0>beg{!1`D?l$awn0egMr)Zi7{iu%iS} z#}mC)ReIAO!iKOBsB!*XM(wV*(a3R-Iupy&#ng5#c<#J|h2R?J1b@^~wU-+Ez!+l} zY`*!LdUD(b4g8Hh2m8MI41=QFb>AD7nOPQ$p>d2y3|(c_o>M{3L}{`X>`0h9O6T_% zgoSL$J@y}{s@L@xlM1v!7#DJud;kE_P=ubdmUjRQ=OcF*Y=MU!;l8Ce!;P4QmGHy8 z0}RPONQvGF`oUIk9BiyclgS$1u4!YuMZRkUO=(+} zWmjt2i2``ye1dJ#{TI&1ZwBd+CX$(b@nwKx&+X@)7%h|Ad(5mnPw&q>0auC$`V5%U zwfQQ;q@1*HEea`9NXV5u=m-D_>Z9NbK5w1pLkV_h;7mm#jke0{=g>gF71(Fw>qS~! z9Qba)vA0vHmp7{d^+a*J!pKhsZ}Dusu4tPnnVH*ZjTxlDlhD8I8=*vBEXW{~CfBs- z4OR7y<3L7Dm*H))AH^!=4hs$C^r#dT5AYLOZ06%x50`4Hw$v#lq`m-!BlmpSkr0Bh za9ZZ@;9S4H!d=n&J%(Rd=TW79MY)d0qp?Qt)2{E1q#)OvOWD@56rk=ZgdSvH^CFh> z9tgb;kjmc>r;PyI1h1iTjt$clMzIVw?DZ18a$n983lry_eZ{V6Cv7&BZN@?u74TlW z0uNS8f}M6`U2UD%kmmsYD9(|;m$Qf#hDhIdJm;KVlg_5P#gMsYyY8;~Fdwfxja%;! zCLsF5@A%thx{IV9c5_B=m7ir~)=k4}AB9B4&z-mMYgZOzYg9knK)pUria@#P^lrL7i{TUYzFh9sh`Yc$QvWydkJop>)LKmQwr1A44D}4D= zh6(oX(;?_3B%~xDMAt8_bYs?hw#`-S%Z2PGoH@ z+n~p7t#UGfk@~J+-b^|xyP~MdMxa~&s*5}~4v_I^fU0y3Onga!f?l?MkEfdEHo(Oa zjZ(T;Jvnv27Hfi^K6Vq1i^}io5D;8MGH6SC&-N)cc)$Y?I0ZGCM^Yn1lBDog=In{j zI{#^Zi?MaZQDYXYl7cGlfvY5%)Mv2i%vCg-_U?J2*PP*`XFOz3UN(OPjPsPrlcS_L zeD-fZbA%gV&!H<8Wia>)_HEAeAcOCG1G%;#~{YCE52auA- zXWqc^S`E!_aB))$4Yc9bb`XOKdgp9H{Y{6H3`XH0-E;Gpue-94&RpaQj*dIcHrDoH zPI3tNeJx)un7Xtz%>_sXw2|!3COGLE={7P3Eyj%?yreWf&sl8005M?ykwrX6yc5_7 zP7FLFf66cxcNZe~F=8ce-UN;~I>W82^h9b{)9B~DdvhE~y~!%96bYU8CZH7+@0ndtEt8jPn>!hBT(I^5ucv9i(lGnE=q4yrk_k^>xcN1HnznUN% z1}8G;+jIr)2EE(_Rr^H*-)B z73UFT!UT{dC0F&#uW5sg6?pJ;hQ^uia->R>H_)a#3`rir9WfBAIoA(K39==zOpXEa zqdsN(mV!(hhg#f5^z#1Dv%3LprAO}!lCZnJ$bBkE8ZC<9<_~V)f<_js3dAKmvp4Ee$Y2~=5tG^nS*ND* z77(bqd<%aP2$lcM?hjS_W0Cpk9u@?6#mjLIB!azNdqU&5*^F`T!&DJM*dEtjilJ@~wrd2;^wIKc=B`z6y)tCb~CLerJ zZOz|qw}XE#^lXlK(`N&&9_LY7OfcWw8;??A^tvurRbC=OB0F3v8pWe#fb`Z9JpH})M;SrR8+^BL9)kHAi zy)Xqvs{iq5Gn21x4*J_yf3*Y$k?zbMOL5aoqjm+?xD!UTo@8J%WoJm&pFl_})~-#9 zN6;2;vv}3rM#=?Gm%>?g6-O3?>|$$1oVAG;1ps9CcBUb&Din?sf=}Bd_eaHC9Ey z#2QTmjeIv1C(;rqQW9VC)#Y!DK0VErXv06Xt&}LE4;B{9N4^8a`%sPHgubOzm$SL| zyR)QP=&se(Ds2KO%74<;A`(eiFPtDxzRScgl3aXOn&zwRkVixhSwMvUGg+%G?7Lpo ztJ!EN6Cvp?TC{7AaVr6Ee+Y2I(@wr2`NJ#oBhoYG*kOTwgg4SYMto1c2GxWC0aW6n zcHlL~>Yf02WevQN9jegp`&AV6N_fy=a~plCr!rzv^sn%@#)6M+JIB;0U4QwOfJSl{ z%3*8xEy`jSmH5gfSF?G6!^6y_hC!p$P{b)A@ViY>kWnK(7)*dl4>x{H`-f9KJZ;sh z{{>O$A$MHI*N^TQ7g5Vs{t2Rwr|@0UuZL=g5Q^WV%#C;+J$7$sa1>Gj@ zmLWJMMldWg9@7ywf4N6NJXX#WycIBF}{r z)(gjwPTl3`>dp2aJmPxXfVI5<&4f^NH@!2PThTNQ?$zLHTGj>JNvF*4o}_V_poGsoVp{vP${-~0~w4KfBed~hX%b&1TkUR z(ERrPcVWGFVI3WsBbeJ3#Oa}&_5A@$9j|?YR=Aoocl#K<^$q*;HFrO8qW^URIPfA< z)o{Ve643H?UGvLcZFgANknGM zGn&pKyh6Uyzqc$SgD8gI4fT2M*mhBDtGXd^-LCi~X^vz(b^m)*YsGfz9Bfht0KE9W6Vv;7)%w}2sPjtDNNpRP zCg%Tz@P7ip2S4}?*(%I)JPxgO-Oe_vuUDkuMcwYXiflrW+oH%Of9iHkDzGs`-R=>^ zHHOvec4;_Kw}bk2Lw&JS(OBKi!u##&)a|V5@3%`si@F`8SVt*F-L4V+bsN%Ojq7%% zTSIi+Zip_%!VqB8?V2`GIM^q%mZ8F_@gTH=eKMiVoz!O3?HaOGpl(;&Sy7{IH$dI4 z8AUdr$Zb(%19iJbf9rN;&p_SobluK0s@LsQI8nEQ`gTKou~gBR5<95d)lp@q`Vw*) zTGZ_z#X3qc>UIq|*aoQEJ!WAW)UmJ)P`85scSwLS7;ezS=Av#_$HnH>ak06m+d-Q< zsLiO`HRf?MQMapUc{A%+-mdNG4>#?bB?=vHC%ICt{Z&lofAu%9xLZWxE(k-ZPiya9 z3RB{JCUu#56hPLhWZ{b+)KA1u^wBqA7~Od+EDt613va?#H6Ewupyc>G_VOR>sjWvL znUaiB>upv1D{AdebV2^-b?8y>H)0m~vAld%yCj(3%M12bxAZC926c)V{e?O(^*w*H zTGE$5yoYq4~RuhWdUp^Ui6GvX$1f5*o@-`BqPE0N}V?|6Qy%%%ig z(RV!JFLeM`xRHFfuqbwer*?rJ{WMu*N9KKS_{agBFiHp>qy#jYNfHM1d=o8c2*5|) z__M8S|2GfS`$+IvJ(ER5m*Y(+z~V>J$lW!0w{#^Ne;Iv}0|+I!e5Nq)CP5e^_hP$d zg7J;af46+I{@wmDD2^UwJB>Ef34)}Tg_wWH_wQuw0}kNdi|plfh<#J z{+3BP@?E>6;fxdxn+e=ZhPK z4@>H{+{*v=zryexar5t@&@N}7h2Uf5Dt#M#qGNqoe(PVRt4Aw&?qC`ua?bpYa()_Z zX57ta>MjZjdBdNJXu_)w@(*gsNl>29xo0A4vQ%=?;1eHx{skY9E@Tx^@24c-eyrM6 zf1dUz=XgzHyF4VF^J1hkLApms_s1+l%d)JOk;f2OUpS;%D!i%a9^+UyLBuPzDy{|1 zY8{FsLB;_|WuO+(E5OL;KY>Ph61^CY^6Vc++eFp~m`$7(Pr+>Kg@r1K%A`1tiMLwO7ldp&PTB`uN0Nae$sLbEMfW9>;vZX&P?y~AswL_8 zR@NfChE@joqrEn%IGg)d`bV)v@sQP%`ql25Ur`4(d$!wuE$F7jU9=5n zV&Tybe*1M~y&(`$?<@2q!V*m!e>G?5Ddi|(rnjL!N$Q{qvM)V9yPd3yIFyZ65r%0K zgMnROVEMu}F@}?ggOaxVI*O^us3-l~Zqb65c?yU6TJU_8yI0{xo`O`}xw(lbnlz5y z4IK)}LBW_Vyw$D$1rU3Ui2ZYPM`E(+ahl19{r=0payJ2S*Z&81-y*yYe|t24Py#dw zVxG`aKKen*?&C69ASlH&=XP1w?+&*smGJ=X7K8hn)yh9Rx6K}jRYTAwG9*N!5?N&` zc_Jp%g$$%m{8@TjOjm6ajC z3y5Rsq;}ZHjP|-_w8MU7e<}|9K%&k_#2glLSP=IG=d~O?gn9}0lqP}y@I*z6(KxBR{0YJ^?3ak8lxs8k1w$NYz!M* ze&!h5aC?i`3rQ#UXss~Cy^zu1Zzv7b5#Akiq?xbWIfpzy8d-q6f72k}P4CK8mH1uk zP+<5ehO>9-xKVh*GvUFYCmq9T;aGjZreOX=P*cTwvt)`oe``0Nmh)@Pp?h4a+Duz+ z*ow3r3_x4GX}5EnLWe|c=(88z4|sKFNNgk@q72Frk-TFNsSbUR&uUC2(Bt^b}X!{+O} zoLK29FZE`}DPX393$?As7`0Gkj?s4kvMUAI+6!i27#P+yhRrF-c)rD?$aqvhMjA&o zElZ6;)N@U1z<{9x@*ZiO!Il&W5DUb*G_gF{Z7C(6iL7VrBAW?nfm)ze+o`n^iq>Px WqUNa%F4>yIuKx#AS6__0@d*Gop42S> delta 35174 zcmV*0KzYB7oC1rS0)HQi2mk;800092?7iEr;>eaZdcU6{lv1j4s(}J?&Q!OQ$ExgA zhg4?PD)kjI5KO$Lv01j6Gg-ZVsa|2JFMjoH=hc3V>^G?=*<*}|@Bs|d1_K77rLL6| zupN#V%=GRCBirB$FLbhnA?@sa5V_~FFzN2WdC_i%H%a1ktK&S1D0-|}E!)Qt}Q>Xy5=o;jc5 zD4+cH7yM}IdVkhG#KYA6@M)Gie)50d>lruwg~tbXJoaWY4zn?e{&rk%X4=+2!eM{= z%7&J0!(Z%w|1WFsKCAv4`{C7xC-mch<8OX_^^v`O_v;I0I?lqhnR-y=STyX+=(@Yb)YBC_}E&^=v(oE z&lye({_VMQ*XQFw_-0SWd>9PyS-94N(bRNqIb9k(O)S`G#MR=$m}xP_rg%lXd$m!o z8aHH0tAKt2=7; z8b)*2Y4+;letlB!^L}^OZ-%4yr<_Ld(Psu-9&N>Z3-p z)tPjMyxVK>@88y&oldXc>kNMU`SajaB&mb9KfU~q!N3^Q=Hm%T;olchXg*J2!9gY+ zp9NFSe)|gwfIsJ;lez*We?k8&5LqjLr2xsi2_+EkA|LetBF*OJ5+pW~)(QLewb2U2 zT`4NF9LB!{p2==epq&}$C5?nBnPMY%22u)39=P;`PbdPnB)uMo$Ik{ZsX?S=zPs>* zAWvt5AhvMg3lMf2RDPT6${z~sf#119Zvamk#DPTqCNPIW7(T)Pe}}Gn2XC_?v-E3# z!39abdnaUQ#s?ul3C$+aajXEfam42YPq)K4etd7*3w%nWRagI~J{{ay4r=I=xr?(9 ze(aeK0%4*_fAW9+pZ`1_Lhk{;lchw3ezjRo!{f&hS3POU&9UdYvnPAP?NtWvp@g%ABHyp z<&bEHKVt-{0UIt6%-W&w0ES6nt*{|N5@(*}6C%3sOT$S1vRVK0kAF5BMIcy;lphoEjt?|@+Kvqi9Gz1nP?AA&VBf@uWP2$qUqhLIn^f1>$29|RM>LZQJQSuPtV;&Cpr9ij z!%;j~*LoSj7S|*VWMN4UxC6!a1?a&00M18NPxkBNmd_BZKnO?XoIMMca|@3(Qy^77 zb?xy@e;hPk!Y^%9zV2S`9N)Tia7OV$`ThWL7Om313Y`e#&I3hp(^vs!A6#$D$Isb! zfq4f%^xOqV@VBq+I*9S)I~u#c7~aC?bAsUHEm9!BN^4lfb9NeC~Qdq+EWqlO4H$wSc`shT!M_g{~RtGr}MI_2p*< zi-Eu%lTG~d-$5GXvct}MG(~vkU*_-|>9 z`*#yY=!v%@Z&+;VV3=>6TN{T5|HP|CFy_Ggf;WQiUsm5#KyGQuapa3RP6nJiW^(A! zy;9igN%qBhl8faeNYHc-V7j-szDy4ue+cj*pu0$lBXSy8&Hz8&aH<#pqzuBhkJDW{ z>(L@6bf+Y}oQ|(fRRDW#<|46>j?%YZMpORx1)%w}IMxH41H4!9J{-7p3ju{QKTr}Q zlx6@5G$08dn8-!rBnNr;maqa__$P~D=iNdekL8T)#h9am`rftg`M4T0i~{DvQsM9g zy{}N^P_qNmgSqvUhSoHFY5LOibpiARn9L`{6^Vn_5jZQS9juez3>L}7DcMTFT3mQ` zJZ?z=TRjG)3flp=$b`2?Egq0gq;kSJanr<@>y30?;}2d5nPcxmnPab$GRHoV%vC4` z9ObxCuK;|oEYdCrgOm)$RyEpB6LdEQRg(+}R14fffpBuqU!?^TRFi879DgW-&##1Tmux6T#IwW z=+5Q}#YMu7l#8UmMu&}NT{_kGO?FecBPfDEI2;SEQ&r&TrU&>!F(beW2Q=-G=J((V<9WJ%Z54h zVe=1{ju<)Pe9}1I9pO$Iw&(aYeloPL7g`v?Q(Xt9zS1vkP# zEIc77CGUy@#$&qv?BkC=-K=Q*y1HX97>pu~u~wgWvHSmSj{Sf4xcmQ=F7EQV6f{?C-?0B|-XTy5ko+J; zMB$?h7JMm$iM;_~V*b((_)MhS-y(HjssNy_Fv#P*&>$a*N_4iHD9bqtImP4}gs>!? zObRShQVNhc@?2k~`==pi>T~kJ=epP*4?eRe$>+{5)h6<-`%CyoBrY{iVP|C3+j0%H z8fwpKr)q3=Oxo_l8sJ6~d}$ljH6;GUO`z>cWu}#SOl%=-6ptCd=jP}Rt!(g0Wczht zp+FcWlL4Hx29wMSDu0zFdTuUTdjzi=t?4^$qCZGPrI-8OoT0ujArmr&0cgQwtyYQ0$N8L@3?&8656USZP290?M8I%-L)VdDj$h#~ZF4;4UWJh;cr$VIXw|x5 z%qH6C@dCu${{Y!!?@%_`kN89(n^lxeCKU^R9?5%E$bY+_CoXDXxWHam#1Y=Xw3*Pl z)S$|ZFLUuRl*Jg-cP&McG>lue?8ktj7+Yr=vWw4kWXuzZ0U+|Sl{3{&+MM+A>G8A6yf0mY=; zPk&e~IQbA2sW2uX3vhfnR5_4sCzB+`13Z{pxi6IVV^-PRY?w=5$n&cj+?7C_#(*R4 zz{bRlMgCLP4VKHXVNPT%nz>0*3NwKfld#l^2U?sc_Og{q#a)iZp}DI%8NJhUVkbe0 zJxz*_Rdr&ZbTB;_9z(G=#4JcRl9pg64u1qQsdLkf)0TF^2a8JT=SCC8y z>EX3l7B7@IA;$6w1o9rU!u+_T-k58e12eUliSs)iOjecX1n``zXw%`ad58L;jkB*G z8kV8e;y%jW7$|D+IE^4bnc5*)ptbL+v;-6Sy=3%RnBKM)yz?B2+DGge*fj=slxjq?tOmN3B? z^_vQd7NL^z9vbjA0bVq3uu{!hrvMmA>ccMJ6E5>}gdadLMA()6rxJUl?GcZyf704r z`!PWrz6P_Cfm!A?(pZ-V>oQO4b5?B_wR!B`F#aq#!Cj4zf#omB3eWZlBEI3N7qJ!V;*pPbUL52daYjoBxcg7%Ynjv zlLrwZe|7;7w^3m(2i6)!J14Nw+?EX4-ckf68g}IaHK?5m)K-^5<5?a&Tb(^vwL;FX zX$o>d7QMWO)G{7XM}ey{O*DL+?lc-(1`S=g4xX`@L)Ty-JFr!)xS)+!EBD5$Rf>(* z*#gr^quGoVq|rFCEp%hF@H|TS62pYy&+Uc3e-_pzgE3@@RO~Dh1$6{d%NtiItYgF7 zqVCcGi{_(UYxUUgQ|>TxG_^p=sV9PkZFXgb0_9m_E`uN4OTY%~L$FJ9-Ke$tY-=lQ zcw`7URQyBMK;@!1jA10R*qU{0KHbnU!50`4EUy^cTa+!6)XsJSOM6wLpSAK+O#Ve5 ze@fJz*o`A%JH-|;J`kbB7A333M*Z$Tn0)12$<}?1+#0zxa-R~pcZdpXoo%vjvtDgB z&ToU;&@icCQp4m)6HIo&ah@CmCqzFq(VATwUeo7%hIoA zylw)p0ajO_3bIrd2NcQ%DCXQzUUTo&MriYw3x|Aap!x_B0r#X*BN zNI03dW5^v{IpNrtYsu=h?o%}$0C23QfaMn8jx;E;s9Ik{6`xwW!CsC=t~Zv2D0gw7 z?|*(*k@c$Kbw%lFRnLbPP7o<2f9#NR05)5-R>&ckbuIy$Qx#L%%B{gEe$%M+p2|Z4o;{jI zt=(jwAKWS(fWn$_iQ%ToEeK|XNZ41ctG-m*KQ&IwP_q^WyHVw6%oxkw;YF4 zV=`e)g@0p5dHm^^gOpP~Jl5({ls>mj2UxEKq-0^?f8ZIYZs5r}Nn5(Bsp_rKq>kMO z=X&ZAMe;Hb^0G6Ntc#we7ELYNs6}&%<`m5-`*KS8$Ry#1sI@Xu2#Qge7$ZJ+MNmew zQEeLMMH%^B8iq!Hjs6<_HTui=p;|9%U?s>y3H;70Kp(g$5+rxA*4k(bMq#UhsKH{# zpgrdye_~54QLMOE;!)z7-H~2l@?bqie{0Mq7KlK|D~&Z)k-M{6`>9}uwjveb z`IZjz4Y!q=xZ+7~vy^Bt%TLgdVk>gYQ-fk?eBn&;p>Xk^{`uPx$r;52UPHBP7C zbUO=9Yk<}OT{eI&w=o9hJacM(`>Eb5>%FqZV2!~VgNrqYw=)~W%ZkB|+RNp&RVf7t ze>+pLR`POU^q%(ej<%OyBxt`{#*N8I+ixr-VeC4b_3KZS>H5L-?kM72;9Cmf%T~pR zJRS!LB8#)gB0z7>irLyHERR{L#Z81b2IdzIXxOQd9Vl-WS3Bdb! zus<`|gSnIgq2gD<DS|F-Z+7*{6p+3S}d-rU57?1PrfeLN(o65N!1Cq2ZWQ`qB}z&KW! zd9q*qh@Q#mcGybL#sYmyHnLrjf3rR0cr9QqVICh}k&7Ye3GQW1OX!uA#dLUSap9%f zDy6ANQ_+P{QP~Z7-OPr(ZV4k>b8*b0@u^{O>f`l75|Ma^D;o0t=XdOP*8`n>s|5Pj zMPQ>`B+CnjcDGJF*Wsvl>J;iX30^WZFKy(dv_cq(ib;N07K6-iG9*+MfA{ciky0Fh zhAN=RY(t3*Km%{R=m&y-#A{*Xc&IKX$TPNPlfeT&X2aT4S9CA4EBaJBcm-X@=--VD zOcl8N)H_%pCdAB3KZ{Ejr@a>n;l>dDJ zv->PT37_y;?1Ok#xJ$XoQp(T6HCV*Yusc=qd-MB;l~MdQ$HFdXW!}UcjT-yKg<0hz z%jdF~&iC;DFnUupw)WXAOoW4|BcUbW)iKD?m@_F0yYxXXvIbQ@f59nD%_*BXg#=%e zFXAwIR0;MIs06MvU=z%l+*HddNa=Eo43VfdXbNE&E<+I!xKL4tiEm%q_3giI)%{wh z0n6E})8QdJF~mnXTv#@yM<5jNd&^y5Y61u^!XS>?=OfTGRTj#J8Jxf$UZ=)F9@L=&FlBUibLfPGBggqYhQGXE z4pBOMEUn8W@aP!bqk42cqp%~$Compe6X^?FVX<}4B#=e|f9Vs){$Dk7Xew_OVCIS~ z#4cWeaOIz~U+-~J_=5_t3+tx~oDY3V&z_>u?&dJwgJx#$LG#LBuPfE|qG7LBe%Mp% zpkXgB?BUE|luEl8*i#E}{;*fioc&NgcALeM!`|*$4rc^={qn<}S_ch#d0`J{4t3lp z2KLlKTmtMFf0<#=xH8z=)SC8^U{7?iEjR3`bEgcF$BfBjhv6 zk9=w!H1g#|KAbty@?uPU;YwTt^0l)g-zklJTgn$}f4}ab{JIBi1hE+ex2&~x@$OaV zJLczY6=nIm96xU>v*oXKQp?}Tu<)9ly(Ps$?f2aa3x)HyS(G*Ka-6@-%*MdxNo|2A z!@_I!0GAXCwKI4xEEFDLqt@3R;nn=te&+nw{xxFZHM@mNiiO%kycZSzk}EYx1)y|7TYk_*Ga z%kd?g+Ygu|hRKuVlU?-TUVngeNT+gZyAqMm1$pszz9_6L0t2hl8Qr(f1Mh@7C`&J6l`r>{Z(d_li|exR>oJ@yBE74Xr$ z?0}D)eIL4+V?DYW^fl<8E9jRjwl53lf9s%@{Xt(wwipFvf0Q<|B`4^&GJ}3ggT4m+ zvjzRe`GLNU-q|1YMF>w(q1~ko;mHa5XA;$&?4qhGKU_>j|CN>1VJnZvDz0=CPSo(P z>)H;h@rm`!@rm_oK{#E!|HKHVV>EIjoGRCE=!y;JS-1MA2&e1xpBUkEKtgVWlL;LW z0nL*e9btbp&9;}!9RF?Mla^>->!o&2e|)vI|Ml^XBahEw+$UStebv`J7R}5ai{>?> zqxRF55FNGewh(j_{@kkWzkbHenNALn+Ra-+fYh$vLI5&!2luo)cnu(X*#WY5%>b!g z#3cks?KUn1AVXJjN4t{O43M4d2FT781Eh8@mk@s-wad8>fQ;PGy7RjRNbQ`&cVp8UyQ8~-LKp6gQ9gb)z0vdq z$^pYQ^gz^IhsD1AwyTW@--Fto&XmL(50xeD7l>_bJ zJ>q}o4Ya*;1X}G;FAQjfR~?X|r@iWHK-w7qKw+QWO+&l_m_=LodgGhZ0c3h#Vz zKzljfc@4BB^U$9)&}JynTSB1K9{$3BR(SaVDH__#zXr68>_FSNa-cner~kZxwsD3) zt0M&p16mm^&@&3kSGu5R0i2VpKx>?~GDm+IfcEea1LqC2#yJA5jyNa`XhqaPPe&cB z18w%GgI?L94wTO7AOjs_kSZKHreFsc#fgiD5cJzc<%?cU2m#JXM%1L~yDrq!5d&u% zF%Y7r4iMNCHAOr?zf)9y%gc!ez&Xi^nw_iLYijrZ*}DHj)YRVoT~Sjw_xoL4iG6yepl2Kp8DeUnwR6L*LzLvn7>HJyh2Uwhu;-7rQ5yV zuJyIsea)`f&u-W3Um0p@FZYC^@9N=J`{U2r zAHNB1wTr$$xRqXeoyETnx7m|R`(;ZmJqg@uNB&tm@;AY)_U#u4x6<9;D#~$nLGJ!m zcE?q#gpRAD!>x`0IO_<2O>nD&0t$p%87I)vE#On&wwWDnn^zCFI&|Q?LkE90z^#rZ zC=hN{fI+LE+N+loV9?53dzGf|>fu($9GrE`!6vxXVF(4nt&B`))#^GjVGVBU+2OWc zD!6@cyID|R8H{2uytRLQe`kB#4D$D}N^-%SbD5mi^Q|JP$?I2v+qk}R%Sw}1E z0KqhK@z{h;vnX@I<-{i7oMe9lT$;Ws1Kcf<2pVt?6}(Uuz}4{zI|HtWP5{Jc>F9(t zz-?s*+}5=L?v_{t4Y-F2UnmRU>WGD%0awH)bP6kheL3+7o$MvBJ7p_@eFVVW5{;k% z_fP>0WdU3rv#>MZiU@^{QB=Fa%ZX6HImrsRM#)@ehX>qk@dz4l4;6pHP!hn^Q42c* zu8dLWG>USXT~3Sw&Pi6lZCstxY+FQv2HZmhF_Z;xb=<}M^!A6h1WR0nbU}B0G`~+|40WV4BK?#q&yp(wb zfKukgB)wn>tz#a}w}ZDZ12*O67%^oN$|JZeW~-*-PFaH0w#u=Nj(8O4x6WTYMxs$S zE|X3o7-a7GfW3~)*c0qUd`6?*EU2VpX#+HJf_*b{IY}{h8tgULpC#D$&IjyuRLGuS zFJeO)^;S`}g31~kk`wG(nXNct?ljnIus=(%@1GCY>xh&+!Cu6rAYipFJ1`|T*tc?k zeM^(dB0hg7ht2-T-As;O3*;Nz=g^0Jry4|8}MNh zNJ61R^QKk+WNnh4UGVE8$j8ui@7U|9IR`>#jYEF~3T7`JhZB1t$h0xw`YZyMH8N>r zx&mZsWcPi+ZnN z2F8fSfBx}bM)#(|QUpOKc|Cs$*eSD?NbPF0Ywhx+b~T?P0y?*xI3<)Z=7U1C%rj4UD}suUR6@8pF@n#7Al z$gSmRuNooyXM~VCqi83D6nR9wPC>zjg`^Rc0wH^yya#TY#EV48t@VFouNfgT1ZI~9 zA$1zgP6#QB!Zyp4ds7O8>@-_B5%R>%4-SWr+Y7*6H9|Jd2O)Lt%1#I=GgjLDG9|2( z0wLS|yn9HR#M2?mM=eT&=J1@M|{-a&-R#w$( z7)4n~FUE%ib5hUkXE!c*e|ueyq~qdk|Ge;4d!u&6Tj7xE_lt7bUyMVl-_Px`@1K(^ zCK(EKn?)57y%^tC7n4pV9Ay+#p8s;(WSvewZy-VEq;>ia3$|N(8)~rK3v7+^0$W{Q zZC9`rh1EK}f*M~HQdg}UVB2fv1>4?9J7B$$KZm8#aR}wf8y$>Dl#hmzkr?Ozf=7`t znyf_7K)2S_F#|~-R=1mz&n6fFTa)M}9}1e<;l376ni~z9lMyE}3$FH9uZ9QZby#;# zlTIfjf7HJ6f~s}q-d8qGy01JEm}AlPBz&IsHJ!-UG=^0e zs8u++6Fx2i+NJckA3i=`2?tVJJCN3b)?4|5)-MhZwYw;%yQpM%sJ%it@lYo5baYwH zRXohyxTta3#zhCgL+$j*>GUZX9%|oCPCS&J978v9TgAitRf>(v!$a-2$?3N#86Il) ze@sq1l+GAKJ7Y8+YS+vDu9tG*q4u!k#6#szF=`F%Pg%jkMt(eOoDdIx4L~p>&zVp~ zoy2p^Nj&Kyp&c!c0%iKd#XT<7R#7|1i}AQrTX}boG>MmrnA-JnhM78B5L3Hic129# zlc{!fV$LdJcJpTD&?H_eVru8i8D`{ce?d&`tl1SYh1aH9uXVK7CW)Axynu;wc&UJ? zy*Fo=b+ZL9wKr#1z?5zs?bTT;&z9GSdATQY?c6!Xyqe8$shvH$!lm^36t=&-9Iwy0 z?Js}kGxs@rKbxDQfB`7LnllAn0tVTbPppv@uxE?y`BgXCRdga(k(}S5X`2pef7WpU zwn!1uuBAs|ImdLz+)ag90xrhelozNqPHP9%o(`%D!)fi%DiWuKldGkDTq`);%Io54 zoz%lsMxfSyv7COf3j=EHM=KJjg>S8=U2CgVLOZ#ugz6o&_Q&nzkGn9S*8aO9fm-_X z8U>{$T#!?*(a2xwsiCo2I|KJ|e+FI_R%_>Bkyss-<*RFl;tE#R^JDcXLohzOfoU@i z*Wk)en0NxL_Q>+-iV}DUW9Sse%_ruN#CiZyduQJA0ZNM+qMWLMc(YT$_>*uBUC-j=8xPt{~l+!mfJ&m3_f{G%(W-;QI2?Q3giCE zPOZ644ACMzxn4JBGbcUZOkO@a?Bko%>~N{<&?^-?X#eT{>>#|R-LzxF9_(;2&eLu^ zcSG50I6Lg)5Y_B(sqD}%e-%4ur|tgiAY8SnwTbp%hl}yp=B`k5+5$z#VuyXau$mn% zlpTZzx-{&dJ;nR8gYXYmjkHLfJ?PsngbCe1DWp(vNt*_&XQq4Y%eqOT=&gI3ze+dGL zH2mv`qetOi2Nh+4e;GH_DmW$Wa^i+s&HR1bF4yW`hHxPb|2m%NG5FUZLz&=TMFq8V zRL~~)Z|8^q_T|EV34(z%{Of3;N8w)w1Z9GM8S~T9F+UfwfBXM5=kTR~oyR#K{%alm zqlw^xqJNZV>kyzv*+5NIJPs(G5#o4ponb(^`QS?EKnDbAIw*5MP=XG0R8V$05TQZs zTCYetxR}VGb}uI#;6z>l9q1S#O$TL;5lYa34iw5x2O?Uim6|26GaXz^tWc|-`*cvh z20G9&Lz)iCe;hNEpaUH^l${P_SDIS4NIJNj)->JRc>~=mpaUI4r0Jm4F+_1X(1AqR z=|DBUDfD!3LEUe1H@~@x(}9jD(sWSjn4$z7=)j`vbPz@t^^4?#3kojk=MOIGpGoJ+ z)f!Vbx>ecK;@X$H;;#^Y-9i3Eoqh!@$t)b(Wf6_XvHF9@YYg{|yTwnV&%{cM+ zq|0L*9RO7%(GO*2kB1nTk_N5@zd$vCPNTw6zB zrIl)J=PuRSzIMjBzBX~1apF-{m&Z6djH^h-3ERunb);8nce#50?sD}jC!H&7J*P=0 z9?Nxcf25;>#fl^yRh-Hw)UtL#wW*9&e$u&i+u9Yjv(uy#j~Kf=($OJlMUsvTTI&{? z({e#UYu&s_Ei{!!C7pd}Lj1~Gh^e;#qy1negMc5kmSqU+l@FM5bL_e9Y~N7clcdvq zFg+9Gu~m;jS`ENVgU}`rqRILN1~;)BAQ{WFe*tuczQF6~gnZ(OyoCkP0eiKP6LjgvVh%O&4;r_5Rx9}-E;M-TD zrKGI3j&xFPvtwL}TIeu6u__HxFPB(J0(IkUL2`K2;?EGSG2%24WxX-*@+|XLQd!F>S+^F4Be-VCFC-DH->!9@B4#13 z!JZ2t3;(GUlRhsSf5ZLH2dFm1uE%5$bYt+~dcjnqRf*7wR)pZ{_Zw!-U~n8t)9}Z*}@L2D!wtBx}Nn9*Li{dN&W@C1DUV_pA5M+6LZ7V+$1WZ z{02WDC5DS&9B8G5F)vY;@L5WwzVq40oLe}Oj~q3Fhbdb4tl0LjxLu`usaUGf%A@wI zhvJ*}WRrI3e@)+h8BO`$7a(oVA{c(b0EC&xt9U0qzDdFM(5*h^6NKbJWZxdlZ4;g^ z=4}8oDP|f+lJK^8I9ZFhPj@G03jlC{8jh04S`An-pe$>=fHnK}wN>kecX$RgV$y6g zqt>}W*F7jN^A|G!Ird-y1mTR}eNIyeG2#yl!NSqve?ZGaBqiaGmku+>0y$`n8?iM; zI2LF@3k0ib+iVEXzW{_73rbmJ)BQjH`0q9CPoWHeZ7FSUgeI`}k*DVaXhL`o97AsR z^vVF@15V?a=?wAT$DdH3%Lw*n*2EG!tBr&c7quhRSR5??$0=|o#ibKl(TkyW74|QUO3M&CL-=Q-UZ)t0nT)}}U118RSJ=bU8cmTM@Y7bM z({9(;hd6^Ue{SP)<1{9j1|ILiN6qj9!W}p!JY}OrfcG17&^ekd5D}&0LKZ}DRT9?d zdVEeChcrKxlb=3MEuYC@(du|J6cRI_72L2ne_0`_VCuPx+bR4SrWW))uzXrP+~}-` z%keoWG4k@WP-{mPYF!`;B@RIVM~RTtZJ?yjJ^I|EdFAxHqIkj_O)Xf)F`Bt13lA31 z6UvM-zIAs27l7icu#PSo(?U%XHZ=IdGcBLTw=lDBbbeFu$o>3;2JLAd|tmr_}1MM=y8{Km{ zGN<9Xct0FMmY2zg?fLFl6xz27NBf=jlWQ{|HDsRqz>J&hfS!Wi+sztVKz2IwMzEA$ z2Dj#1Wj7}=)}(A*K8E`(1hZz0y*V|ic0yEbHZ_yLGaG+)Aq?Kk0zOA)hY8pSkk#na zE1;NC!a2w_ussiVPJV zmKL|iDH~zh9-!o_)b5)s8N~(JdQWX_=!JOP(stQ(YXcyg%>4(o1;m0`VC3$L9DxXg zA;zI-A$a*qKj1R~Zw`tPL(BXUm>;HlNe4{}Fx!8G|A^${Kw>Eq=h5*3|AppYX)Go4 zhLUxG=WgtIpb4=Nju)PQZy4o=P-XBcilX`UwPD0O+R77C>o!yH`?I`z`0U3Xlh@-I zjivBSZ>#vYmn{|NJQS3H?LMT%y%;66`{p+D7HUnHy0ER7y7qX$9q!#O0pN+72aS3i zmdJk@`vY?jxP$RQ{A}-p^CBvSGv@cMeb2|$*j|g&(|-Xm1#HE|QX|)3e7!{%o1yP8AwOs)@fY*C>jj}H1)uYyQbZ7DR7AB| zuLwi>2-;3T3nhFI+1P1pc?V%VT^HsFdEv+O(2 zYa_$J>qS(WcC^?u=)4RRo1fw-sJG9FV6})wCw~o%*t?aOd6Wdozfy-!?3N zc%xmPsbF_pC-w~I)n~0Xp-l@T14S$$d|;CeLSEab^n3MLtprA>Ni>C;uhN2Eu;`pQ z=7{m+I_=cj$s~f8q`)j1eyb9idy>EWG$*^wC(LFiyoVpRE5typ+L1GZO4bvNBJ!huJA;}fEMI|u@V)l-D>7PWdA#NhJNwed5^ihZ z>CVYXECvhMhCY1UN5CG0d%4O!K7vUqIH{ zUw-|})cK&o;}4di0YN17LMO zJXy5s8mTuRwJA*da<7lulN~rUf8%aJh03T=q7KQigTQitG5e!)bihZ7XoLB$8Fueeo6pkj8?0^ev(yu^1=y67 zdYLhkUF067md%x=nk?w_j(nK>$drLHt?U+6+k zV?#Jek@-9pNg2&bf#&JQucyc;52Gah^}s>l z2;^)2#)KQ(Pwa#xr=tLWU^chFxJU@$>w2xhqJ1tS1@~pVQ8x`J<#@tDRK*}7KYVDs zm5+P<;eGVToXIcoOB0%equaZ{Vq6;c2*Z|hHxP4|#Lz}Y2u)pvlJAPQ?0REP?ZLen z4F17Amk#?FI`OxKy&R|oyo?ZvRx`vly}jI42NP;nrf_9xMkU?gr z82z;HgjfRZZ+{VYF&!CYfz|vr1z1#AGoH#lhK(5&u9&&7qOjxg*`I&>7wt6FrjoBd z41b`a8j@ih6W9d(g95Q93ywkA_&mH<#z2cKp&My{{8qhoJdy<89i|~?)I`7Vs0pbL z1qV&^b9ujFuXVP6fQlkDK#>}>=Ty-`2|HO;J=hfx94%7 z4vY1}=^yV!`shxWP@wayey}G@pJ(v{ecryV^L9%mBj@JrFNS6U+QCoB$|@3=qVP?S zAx9WnO`C!|P1|1Cy@r-m+LPcs9Di055aL^G8Zriih9w@taf>Z`Q4s-~#w{`#^3s7! zAlS=}(AIW@5_qpRCst`B)ku1+NV;)T9(y1u*URdgeAub3j=cm9n$jY0yz|xl`Dv{n zX?d_&(G`~K9KaQ4XE6-(UeOy`g-I64S1~J)2H_MC*5!a7sT+oloEc_2Z+~@^huw+* zK6_0NU>=^;R?fg=`EOqCqkJyQxu6qn<{+-B54KNX`lP zdRc+5_w=aUdQ8vq#Us!x?td<@MiSZ(VGZN$9(Hzx#Tu&(msHd)OZupU!Z9Dh#9H2H z0TYZlb9da!(;XAdoKcah`2KXwG@K=obT8xScdx|jf5*lc4oTIh%q3=Bi;5XxLoGb{ zfdo;LJLY;U>hZMEb%jv0m>r>fV5=RXx2|=YhTfs_tQ_A(YfQE#c7Mu-@a=1}&ZL4= zC>lXFBoe8sq<0L}fejayO+Htd)o(J_;Z?E*kTVsVMd0sb!n_XvITKl~@WHKx?$uZo zmtHT;0xB^44c_Lp&_4D`XkYk(`9OUAy;m)!yV^f{2eW@RbA>110-k$dLc(tqw|*ny zZLiY;YX*8YfO%RV=YPWBkX&F9@zxkqJrII;$c^fHZIl>N4`QTzI8bcpYE&--?A31bYy5yA*DfT`TDfT)kQtXq!wr8*2zy5SXb*o_evL?3(t1|Sf z{giG`rCJ-ZaT{S>71z@7Q4L~A0HW4wXhjtfEbvRuReAz@XMYoh1pxjUE26qKd*_)m z{@{9dq67gj!d#{cV1r}+po#+!DPSc4*-#$G6*esNqi_U@PVnOu(FKBt&tEW6WIUt_ zGBAq^MXhIPO%|qTszo~nNRlPR<8cpa=Z}5ySirqmtzS4nZ8H+hX%h_+P zaoj$jM+ zp~3wS;Eo{uq3C{Xmzx1n3@Tq=(@Qc)d>{6=oNNNIm@D4bnaCQ&GNRaT*($$0#4CSi zZ4O!T)ZqdU=?W8uwe3}-&fci=4xxgMldkh%dgHb9S;QN04t^sV%;qz^Rw(ZQd2*92 zbp|*!F@HzFf((yn>Xr2vaHdQcTNO>R@e}MA)CC2!BKi~v5>BKoM21k44b9PAqV|H^ z{nw6XxRl1eqXOi1!({rl?8rt*9ga|{a0HE#8YNGOl4*mNjgf|ElsWANGtHTw$KfMZ zNMozUR*kJ1TT6khi}}2~*qSd;>P*13q*)D^iGLpZR;M&*NZ;Eb^IrME0RSETz_uj^ zP@RI&ogIcHp{Q4s5G18D@_9#@54C*B2zi|^%@Vr4o7r70&ARTua(-a)Ca4q&|M4 zZ+1S&J4C~j#`SefF>h}@ej8#DiVke9Dds5&|a?CO3*==M&pPB~x!P z!OsGAh=yJNwe2uYvFwJlzgNYr<#^gSGHgVkmf5-a0o63f@h0m$4aX{xmA!Qgk zf-nFojEalFRURhZ$B&Y2k*q1=0a2qDb5uN4uqgUnjs4=v-&wwB-HExIa52e{<$nl< z%oWGfNw=BE=^aE;Fi>5UH4C1$;#iEf{`}*={aE=i*NO2&x{Ok})Yfq!9a3TpkdH05{UG{m};r%p)Dwh!ym6NCJs5|^{^ zC^6u`at8QumiVod(?k;DSEIL%c&5-SdSVt`S7^_nXoFXQa)CN z3pA@<3#)=^YBtqus@YVt=@Hp9?Y+*7Rda^(%M~h_3v9BKDASb?lNN~;V}A`ko7?V^ z^Q|FPGt1_PeWq=@59qK^8Abk}-JTcd8~fe$?s$m&vbbh|t?x*1d{S1NplTEM3@~=? z*Vwy|iuzdaVuT55KE>D2EFkq*qbero^M1eXpU#Y^sOs$MijJ<6Cr_~#@#L1b)U&O?0q&c; ze4kZ%_&=mdu3Q>qZt4Mi1s^>Q4u7P+kku$}=FBl+1n50TKk?K&R8kEG=v*<)=m^}| zMUaP3XzT4On0E1v$<$vxu@J(T;y80bl@QKpabjjG9D|f0?^1MA>ie;_C>bs=u9MS1 z#7nYiW($!P*><;m&wmfl1UKoz+gxS!n=7@D-FK@^XGnnp>u@ zA=5H?m5o=9IL960jIzg%I`c7OmWk5T{3E&Njogz9#K3WX86>L#>CZwY|IDc88_MvDpA8L9}nvQXKYgT*=#?0p;sA( zuURrWb0)J&(17XybtM}~NB}&7S_pW#qxn2$egI%%@md;&y<0|LSOUGm06~!A{&dJv zaoJnq;Ec*KMS{Z4jLM^bcE*rtuMn-%th7)$&!biMlQ+niJ-T$TG)Bsw^s0RiD^Kz~ zC@L540Rc``E=_3Z^+{YR=+7afhSZCjYy&(!*PR#Mfs|lsCuc`H7Nx-+`*?O`d|7J7OjGP#z=R0l@Dax;VrChv`LMSUD%$aCOAi{G}um5iT2Py1= zIU_~nS46NpzLlSBFNp?&NTSZr$*ax=Sv_pFQ*$&+zD|x5~3a+-yC68LDLI}SGS1!Ud8e(ZJ8?{qJvs+Xj z=Uvtp=S?cf87uK&_U79j`>Md2$|5vfP4&F&KH9F@2jePGjm17Zkc@?W%X5WE^sc=o zbDUu}r@mXK`fCfYs`~i36|(KmHM9KHV^@K`dGbeT4ZBoq51vj^0=P>pmN?d7)a%8G zpYO#ONRq$97F~-_9s*`~EaxEz};1abz53WIkw z5}`aw*Kj9G<8c+{q6*mNz+8xKRhh-oVjCl6%zJYU+@A7x zxkx{2I0?WWW(6%jyPEmkgq^r^BP#aFZc>!OL`E5e9(jmp@kzq6?)0eRHfXjGLB)BV zhOfIlJ=>f0{o+yCxMFf1#UHFUehs9T$D>u-Q%5URoM%zbKoRieerzXzR55dd1F+0G z>nD?#-&+A4sO-IKat6;hE_DY{Ke1+C#yMl{xGEs|bXc1@qTn0qUr(q{wPF1TUW#=* z4^ad0c5p}7N~}iL#@`4jA2$}<;oaXUdObJm1ZPGvo?LS^bfKke^46IPY$et*T_7^F zU(E2OuD?w;D!6Le*?x~!+8e)L*fHbs8QBHH#1N)yQNiRbPA~YF)y2N7F1?1^jDC=q z5q<+&?Vc~b%4X$&A*>LoBG|J`TUbMZc7Hop%c*@8ME{*N`Gr)2J74y_xf)Q!Ad*MH z^p|Fs^UGx7Dlkthv;T%_lY~`1-EITIckgpXajVha^*;0kOYVhDuC9$p>H5exLMJC{ zMXLl?U`L^kwT9^zxjLU!UQzF>wiUd82y_fU)s{^k+)K$%pkW{px@ZON1>Qx?x-a); zU75^HAXIaaWE9ITI%kc!tH(1@hJ0EtRi^Nq0jfol0(3JuhuXz5mb=T&%wX4fmqZNC zXeoVWkB|ucc-5(dcmbgTL)FPg7UqRE2oj0-V>vTopY7ZU@} z2MR;c8s%I#X#~j#K95{2#7EqNB7pR7Z#5D4 zMyuOe0yCzYm?p03ruI9h>-x3O(B zzuLfyxZn#uB6E3=)elv~=jL4()VT^E>Q>yL|Ne$I{($V;2#9Qv&^NsfZ$|QR@f)S- zUa|kxMQ=uN@vimDzZa=r<=vNB7??Qh8Z!;MPsl{w9{m{<@$Z1dZYvA)L?BLwtK#4^ zqKZv}F>Qt$Yn8ZMN)zbW1uQ01k_Z^=Rl9h{k@wW2i@G9s8jYf0v`AtH^neN=A>-LN zV7EcWA1!M)!q>or3lK<)=tTo`MKtcuWTF_A?#S@u|bVGxTx9U|HwUQ8jeSB79_XKl_110$jT zm;yCTGI&@7dag)%n8^KglP62;?`^W3YR#-Tb{?-En7dGT0Bc6;`5CaWt1LuM-U1@YtHSE)L6l$v=uv$Kc0ps@U(=hH1(yHxi~lWZpDi8*gyXc`x_{&c7Q_? zu@+NKS#HUss@#Vh%;v6Q3Cx_7^J4pPXR99T9=Uy7SF_l?WF1gZ6ob==xBd-`ED`|2 zaG%E5UgzC9mWBWFS#S#9UBTZc8lM2H`3k=>&K*S0wqG=`Tx?S0(kH?a15qsjrWFoR z4M%B#WFV{7*}&v-$^G$=@DKRDv#69A`s;J3e&na|_|-**2gV*IHKlxVE5_BKUaRed*z z*c!S|>a{LfSJhITCV#e+FMO|XPH63BV(BJ_s}&PkwX+=I)@hOr>_{)SZeFx1o3Ax$ zj_J{p76#Pifj++f2KEu@X4|puUi8vr+8O7hCSSsqcpmfDOOp-kybdlVRuat*?<<7& z-Gnrdv9fb(YP%X*=2pKdl`AOvmNpIo_bJISE05siH-sagTe$^|1;e;K1&} zZ>52eW;wUT$IipMii7#~IOS^e;oCU}=agpt5hrB{K)Z`<={xK74~lsQ`#BNjf#S4N zMFR}HBo+k{cP6k?cX`Z}0=R>E(yb~$T||1#Hdv2>p3>{)O2Fe=&^CEWvaHmL)TG;jVreH(&)nS=CTd1`xp zOTj^t_`9F>13&yQnoYj1rymGHkFqQ{&vJs8tX@ts$ppvX0rp4?Zl8qrd(DAw7` z5`Mylnu1^9-(w|E9L|@atpBU_8?wvw;CFrgRwiU4SmYeT4Qm^EY?z1h_gNIV18_yCGebL70NiW{zP?<(sIY?kTe6Om$wEDV>Qcuc)mF16#4|Q>iUrZMP-c+IET0MSD_4bE3EzVmWs$1_ z>#S0}HixnlsDJeBPykuQSzb?~s}*46L5_Iwb258ie`HJ>$c-SuwV(1cFnTlC%3)?F zZs0^CDP>xk$R6Z=n}*JSW43Srzwk!(YNj-kah@?|T?K0|b}HTXD4kyn12z3Pj-p${_hcmP36fw$E_1%!?LdK_>_BJHJ6 zlRI^ssRkj}5M&0OE?pRv`Y|u>S4%Sz#I_c=<3t~2ZnE2`(e!AbEDjvdvoaI;(!PXq)fD_ieelUnW1~h#OQXWdz|MeU-Vwo5!b#qmMf1bf7f#*fCZU|tvLF#$25pqWY6xctwD)dNpluPH}nV31{P1NcO6v<8! zd2Ji1e}hy~iAP%nf`Mpo2`P@m#d@PwQ-Y}Cq?Ss)PJ2{>#QYqt5}Ov#XIZ$gNbk89 z+HiaLQ)!+GO{|}xDRevjXfZ^Oa}O=!vWlY}wQ|>pEd2=4SZrnI^^>>I65BUVSbSK* zF}agk?Raw5xMeV*S-kz~p%F_JpgpMZ_b!~!XXduJq-!;C0pt~aR9){}KJn?7AH)O@ zQyCB)5ttKh^7mG$n%0l(u@3IR*XbO^^I*b7xP#u4t6-Bi6FOSIG*7ca2?CE}_0cjK z)zW(*cpRJn+oz2e5&`&VSAi=k%?+w7n%Rm=Q-QSXGhSq%kbDVYkbHB{h@Pd}PlkdF zN%Ems(87c2XBVd2FxGBd3ZgeDhuVXe<3Z2Sn(O&d>6pey0~4DG;C?q&Ma~D%EZiHL zJKjaVI*LPH{kk?0wG}fVfke1Jn@K4!4T`v6)0_J2d~;&%y#o(KGJzvAT^7#2ux#M+AL+i-dh#;i`0gWN~79V|cBk6rE3LC6n15 z@VDeu6eQD%XECj+?->a`9*Y(fCQD^Dkk&$VV(8##qLWbQdcpDB*+U|iAOZS2!LcH zE~RqM@V_d`9afCCq zw2}}R4?55CDor1Z^Fl~DV=%J0qX5Z4krL2d3nCYM0V<=&evhsL?wWND19f0*Gp;4= z%0?t;w+28i0((e-5#!5gZ7u;;U&N z!d~EBgQOVI?6&zCVI3E$2~1E2SGV{vry-~$I%UgIG8$D zL~bPEWDxhTd#if=kpot5(sBsvL?E6#H5178hbbmkmuG~=pS@6Ih$}<3p zU|NYjBibU@{gt~Sq}R5_o|FW;Wt1du!W?nw`Zp~Mqe&?yW6p$vJ-q;eHEdSgS>hBZ zdw#-N-T~cO-XH^tErFk4F|b#+`E=?zS|SD+rW&ts1#+>8-nBHgE?8&;>a`hQMW}I2 z2)&cf7m0p2x^_%b(jU~tf(nKZCdMRNy&^*Zk`ZBw*Oy{+4uq*>(S8eb2T`EU=c-`3 zHxkGxSvlTTXEp{~L-2)GHOlDlI54pgg4;XyDHXedDZ*R={K6HDHM_ zYW8^G2<8(?PmFAAVi`%EpiE@N!You)1WZPhe9A8}FlnL41KGN>la~zpA+C z|Dqu|(|iW7ivWX7yK_CGL7+?8y}UtzdBLg8$+Q19asrJ>i6Cc=gw~&sgs`4vM0(kb z16@8K211w{oBzF`X6xo;@6AZ7724ChRK zRW&^jl3vesGXiqITVE9_&)=+c~=tKOw6SeP@4iK z%G*?_LKI@DMsMgw-6Ti^hWkYseLJ9T*YN40V5EMXqdCvodm=kmDMUogjAz5Q=INy4(IJfV3u(MZ zs_tjtL(}bKiAX#t*BIU}OGg6-M3`&I`&|Gr<*A0Ihd{?UMPDbqNX#(Cd41ft7t%jk{1btkrTN%b)FhXN)gQcny;w>*lr4ZdU6j0 zuznM1d5Pk%`>af_#}cxU+>6?kZQ6oBGWcUJk>=H2B%2T$SJy|o;|jIUX-L+T`~57i ze-Mae_`@ORakTZu?e3s4xpa7%9C+mALj#em?QdOAXfc7?K9PYvof3U1P^ngLsMxvF zGOA(3Ai4fx<23uT9TZ^;6MSHUK=*yqhiC4XRu!fx zqkCYjh?8}oX7T0mS3Mk=&l+s{xo65gldZx%pc)0m%zBWH^nZB)*!ddQr*;&8%?~~Y z&Y7(vyu(#I&=+P~o|KC9FtDRJNvrRfRX?+p<8BzKzQ6espM3j-vVjexKR? zlpkLOZvTFya^HEM`9Baeb`iB|G?Kk%CU06#01SVb{_V0rl6qmnbP;ylQh@dEcfQ^A zlVhED@DZ++r+_K|^d8JsfyNV5=t$aDaOG%_biyddne?#LpPMY>-#i6AUc!3B(5wgH z)OHVj>pk-fE4?Msf0Tbki2z}+1YKSxVuCMW={pPnPy;sZig%eUK zyS*R*5t*dg^1y01x?wy-I~3-Yd0{cP(CS_w@TR(o%03f9RLevQEp*LGOgh7o&aFOA z-R#h&sD65YS{bZK!MH7f2i-qjC%#K7wHc0@CzmZG zCyN;t6_l;$6p_ieE84}berDH_38>V+g3(B%E98TdaK4^8f1y8jS-S$hMtg4C&{1{_ z^=Mqr>tE6rlRQ2=yu@qf{~qmeHSfj5T59iD4uu*3-g#5^hN1K6I-K0N7{!~6h`H^m zfTg~+!o1sX?-Im<$wYWrjJ#?}1?2&m-)vvj@0$tV0jJO0E{2SZPfrtlepMz7q|l|? z5)TC@qSFzXi0p)aMSCnZc(BAt6VywS@Lqy(2mr_?F#@Mi zVUvH7^9^_bVSnIo4Uf1k!)vfUmM#2?-ZZW7ea�j`~4Qmu`Dh**hyKyE|Qk7FZ%^WF)E7C)Wf52_k8x^4JaRE^4<(<>e~?ICJTe#Y6q6H@>YNB~IxX1L%ff zcsQV_@dKH=A51B>0{nAX?%IvvRCAlU0*GbP#s1dYc5cL=qAwkZI$OMz!;y`%b4Ma( zU1W(-hLwNB;5Nz1V!B7N^{8BIFidqG=9H?o;C;)jq+EjlmIW-Mn>rpqz;BaFr+d;2 zwj2dOPKiewms2999}l4=5sOxVwF!3Lx2n}DDO2F*q4uZfj+%M89o=NiPrG%c`~7$Jgyx$gQcE z6Tj3jI@D#O)++BuRi~~ix~ngR z_zNS{;m|N2Rjq9rR%BXh%Gi)|;JFya8lG{lYr)yC=zMvgzB`*)3+AT<> zjr$`*|0^7J%X!40^VQ6Ny3%(9Tud|;Rg_cGPN?k>zj-zdiY3^;dKPz_$2uIaG`b8# zP{KFcyIIMwyL}XK^wiU-W^h4j%V?ToGPqmFwa_UTxv=|<3jY^pYZB`3{Nmtf8pIN+ zzbwl@tFq5pT9`9K@A320{q|-_6W&xs3CA)P>}B|CgmowV<^7oEt2;~o%ovt@ET9*2 zzt7QJiS*Oz#DpC^FXAFf&>skZ?<8gINUxY?u5Fap)-ffIm=zzu@x}X#CSlc|L(+5U zWZ&}bfO7YXQzY2xEb~`FTOo`6?(hW13kLLjTU}f^fxkvnVdg!W1Ku**FblL6_h@P= z@`);JBpAJbYoD?paU`?qG#QEod2T{91;4>YY-6-F`1B+ONLaNFS!V+v@ZfM3fJpfE zd;_bRLLhTkl3KA=sMJw3h_IZmS0UaXRnY-^m}lvGUr$;mli*dL;vf`8kv4t!hT%>3 zaTxJtbdV9lY+eQw){)yeG^?uSMTnwH%ltbopCIlI9huoyRnzoOUu24MC2_1uO|6uM z7sCY8f@R~{{U5iBC+t*!Ow`}sw+(DRyDH$D*a}u@Rn5FF*(UX}iPnOg*x_q4PRcTR z(n9)cYQ=|?KmD=vDfi6zpzca)=?J6K(5)-vU%#eNoec?}x8q*6o~A#VZx1w<%$?O0 zXze1wFGGV1j>6az=SQUw$+Ia-h9%2K;-02o{E7#C58_NX6F5@Exmop1z^d*$ThV5B z6AU`VJ@l-P89q>TW##EDN2mdWPDI+XlkCzdPxvi(>933Be&|=coS$I2tnEc2=>dHzs zFe3qkviNmGM^N;=u+&Qo)U!4UXYsL|Bhl|YQ87P0B7cCYjU2LOgePB45xVHGsJf7qF4&sbJ~}~C zUXZW+feMfiG?(|S5DK5d{1Ts&+(|I0&|?T0iayQeU&fLR7(TnBw5Mw9m0I7;+z+?x zUv5c&)ByXVOI1X>R<8f->X%WTbMVrassBtq&WOHPOgZvMM3hkpljy7v{pnWfV%P(S zzY_k9l~0(5tHptfKZDSZY(N}Bf-F6zuIBqe0+t~E8H7f4awu<{AluG}@v_i5yX6@> zBHC4Ag~znZO$eaFgX|ltc==(VelF3RAMm`zyq0X{rmu3Z?&K17?MnO&1P{JfgW~xP zV0iZNY6G^@G`3}fsoJ7^@y(+u!HEYz5x(_+e@Z;fJ3Sm@c7v;(^MLX#o8-x|EK8<9 z-qShwvOQQwHj;8>uYp$d7s6Khi36B)_Izg~%SJXqUh<6()XKuko{$27x!0Qz9e3sh zc{z`Pjh?SYSF3F!iBpM}Db?*Nj>SpJSf4bT{qbFXHBuf>#J;Nzxghe|0&M|i26AlK zdmYi=9It7nB2xaTFWPtQ#=C-o2bOG~EJ~IfwupUkA%E__EBh{3)gM?%@C&4DgS3SE zqclw$^kT*v6(ipbUEZN3iZ$G#APR>os$ zk)o>)6guI{y?|T^c`^aMd7UlU&iLi@!;M)?Q>Rb?D9)*eDyCxX0b|Qr;$JS@-yI!j zrV+BJR7$3QOgP<#=o@i$tc}G;-r$XxXPQvVIF^KH(1a&Evu)7~vPgEV9sH%^b=au= zMs+uqy*@FjEZZ+}e`_IWG)VR&XcKgJTqK{No@#6`k_GbOjroC*#hPrm3&DW!KleJ= zev`xx+?zMxFdCSNI)vR}d^42wr{Mg^5K2`XrwDL^HF0Pv#7{?Re`IQNvRoET$oO?6 zc|}Y|EXO362mFXnGbfFk#7F-yILlJGhWFp-NJ-S|neVJ{&?N_x*M7ED!`SpVkZ-FX zi(Tk%m4~_Z@CXU=;}!G`aShinQghfg3FfOE6*f?Z%20rXmB_Fks?km5CMkVw4 z&(&*+w>sSWy!oZiwqO5Rsq}Fk=6@+1Wj(Ntanz+yZ+{~DO)A7R(Q(q;ORZlUn9O+3 zJRB)=Ikl+Gl-2XiIG^1Vr=l>XX6`Yi?NN|6`rvF_bYx^Lq!P!a&GA4*kjuSAvgjqb zH?PDfsG|XF_sARlR&2I~43$#XLW?w+p6Ax^ziqGJ2~?0wBmSG-dj#1abi=6Utmr#{5@Or)&|uWa_3d|RLdMm z%$&mg!k*^FDbbw@>^+z2alL1#b~A~3gCn(tJ(Y|wKryE<`xENeC7wvntg&Pt>?F^Z zHqU^~v4BgSlzkzLoYAS#px{i=NQiAcj+DTUSDH%}zi&Ku2R3(oUObs#-v2VXM;dT< zgl21KtrMKJ_KZX{dcZUd zZ~X*?{6-%kYx2Zc`b)m9n*&D%T}(tVaf}~(A&H?@o`tZ#58y)<=VEdZaT(7YAErWR z71g6@@RUl^*NiQN+?r}kl&$N1`^k>$M^+HONgT!&yw++YB0pZul!L{%-HNoB`4gNM z`Y>`w8Cg{!B5$;Usd;0v`+N!22vCi{_IBL4Pk(p6nr|7;t7qa`$9>kyNVpyAd$aWL z3dVq}zcTbgN*ulaO1({rv34MPTRF8lp$exTZ$D1^9k}*Wrwe|u) zFk*yVyNtASGh;}57&{`h&bUF3y&BCE?A72W8TTCie!pi$72NBMJcney z#iXI8-uH6^d?K_#$4Hi83630+}y!TE54tpFTiZ?46ruoR5Q8ANxrow!BqP<1XmKO zz`8V!UAZm58Ll$m>J7I%lT7VE0O08oO-agyVOy$>}Cr?tn>>smUvszG#%5;9{1hE?6XU_erwJ!~pq9QMg+^6D%L_{_XxJ(h9sW#eHa|P-Roh@;v(FLeoSmEZUymVl04TqoUXh|# zQjotsFP9rJ-TOkqc7=%f8|St28#6ZeCFg;`%5grIF`R9il}tG~ipK8Fj~0-AF_10X z+%~&|HvlqQWsp8B&(;Lxb;TD~j>Ca%Hagz}h2(X~p`h5@gp9nct2S)&C>`}3oQUY8 z8GMQ9_t{a_(1&1cKxl=#h4Kpy!DapaUgVzlU%H4wc5#2{Z(s5S>WfkpoYbL&qTukP z)DL}{XP*LJ1sB1@Wv^?Gxaa<R8T(8UCLQ9_AcIXzeoNW)K_yW{+g62BW?jJMno#k4 z-ijP>#Y>xNB1M*4SAzvb%H#hyS4?cBgFhSqePN@*(gqYT3J&-GoWZ%ZV+l$doW=d9 z%n`r@TJGA?e$;VkgG5P&f!_;V`Bkti9>>GVek!f^a<>P(aZKi_3{vabl5vdrX^X;_ zo%eNP70Jm7yUzIKNeeBMf+Zbq-4o8~q%EeNRKx1nv~6?J@nH)m7@gfz zL>duWbx&0C>;5sITi?7X4Z&LO=W5Zk2#8>7b(1pH;E1{O8p1u~LQBt$7>KOiuR9DL z4GB9XdlkXh8P0P%AP)qtKXuZbW8ABBlia8XFA~LN9Ln(ULg@+)l#n+ehqAIe|Ihh> z>G9XRYREjm_hU1MWRPao7XEV(WKDRg?{uWEV1lvPqf&{|8%jd3ClyWN016+~)>pxz zIK1BcW;vIS3HqXFMHI~_V$7{>Xr^z_*VX4&!KB!(N$aU9>*!z%HZ-)tdnU4wKMpTl zs=mYR0lCt;q{Bw+VKtB)ci|H!2*p2=K=;(??|;uTj-$?pGKY04Oh{f38VTYmWF^>F zAo15k$ya0}8?c5jDPk>vcG0M4DS5@yiJN-N7`+eqZpQ=4b=OL%XcbkP#w;7QIjnA5NCn%92P^Bf3Ua*%P~Hkqw&7 z_}+_c2R}pZT~gs~ev=#`cp`yg73C$wcT3g8w5^W-)2D=9%?A^JCfG+IhvS+baHe}- zKU8K*U%%DhkL>}Q-ZQ;on&vSw2b5_Q`84__$VgWs ze>(AO&>{Vu_p`otEp%r@w!M#3xQ9H!&(pT-?+UU=rp}^&sg)Gs?h=1w`R}(xpId@G zzo*;4dUeH2E|7Ch#4@y7MVX?A@O`gUi)JImT^XU?07cql2U7UYjJoe1Q7^^%{i%a8 z6_ekjjPer*Yc#?W7;7~6|5;lya=T0-b4|J|B6C=}EQB9>%$|gTp_JH2>+y2|_fUfL zP^yu#_Jl}%-;lk|d{)67yNX;!8?y^qUJ6=PqczJsKEa32sA%w9uD-lyHhwsdPDWn> z*aamo2PLc1+UQtlT4dtrn$~nGvgn++PqBl7C}vC1y4S0Wk7AFYpxp=5lPy8sy6yj> z4=aQ9GT0%#93a(&%_W3UxiydxrU3o(@4ik;x?h2zU9)&V$fx@b9Mz^3vXpd|KKIy+ zsnF(G?4aaq3D}V4Yzf4$64vk{?A$CHkRGd#r}*9}7}vQ-3A$Jozb*`1jqT4V*?R$9*V%i) z9k^MwY0U8d(3vl-(@exE`73(}V6wkV(?oKz4bpUisxzwEx4_oqbzEGvmWoq(;5Gqe zZFBHO0B*zZLWsD}(`duAA@6y3R~rkg?yP72= ztQaur{m#5Xb)n|+=(C3gaJBZ?a|dtUzwdjg3h*yQ#J}TkI@&0ZI5I_Cr8}ag+XXqg zg6)<cp6x<7 zQ-#US1@(N0BH;LYX~+s&Jqiy@W@*F;=vApJYhArM4tqAXxh_CXx>KF!n-OK+RDj&T z;QM451?0)n(Ejmp2=2v2J2rD+q$&Zsr21Fimr&47tS0-L62?oeAt1Vc0Y8^ezNBcny1b zu9ET^U4boa>|Jd|;8f=XZg>}z)=8r(g`lG<*@j`dBHA#Sg4mpf;n=0(hwDJnr3C8W z@}+zzg+!q!@iHzMMc7d;9gQr7)ff@NTIL#%T-Q%l6C=C{%X9Ph%ZG&7F0$m$v z{%Mh}8PAe9%Mke>4!3#|;Z*KUD!T6Hr_qsbc`L%+m>PRF1jqv-?U?)EG^G2g(KVz& zHO5!^?A+j)7QL&cl=ai&fLO6efty;fNa3+abwODtdqL8zy9>zKh2+z`oIj1+=^6b{(CID1P}Y4=>p1D_nn0{^HL-8F-ZtGykW3-K zC``67zzc!Ejr2^cADZ=3c9x^Jd(tq^)ZQDoH318Fm6#a z#K#k?hlVsI&cP@?@SeN%a^ieY;j!0jQ%+DlxTOHBe9rJvtl7@$B^}!KE35VztK47e zzS)b@Z-4SWwD~5m8soS9hyZDpXu|BbvWG;4>`TAFH66w{T>#bAzgBm>n)wT24M=>s+Op zr}@TetTL#t7$v9VKE(!&>xBZH^%Zz?;bZ~ zIR5v^5P;rwgzhQah(thsM5E=PO{1x+^1TLt)mFtUu8xWHAukEZ7o~s$-pRjhGnD)y zkk0WDZ6f{_kS~PxKV9N2(RKgD)T(zg6P3f-9H?&J+K$%<^PFaDlDfZ!vmj$ym7v;| zezjKr9;nOok8gw?3s4)Sn^tj1Cu@j`)}lGREZl^Miq;OUyJWf*On#BsZr;h`B5Me6 z#cI)%KMaDk#A*kxTl;BKyy|Q>w;iL;!FTo0cUbpdJMN>$IwLs5m~zz|rpLe3#o2_K z?tG$OoOeVlf1)TveQml!CMxzg_eW%BSA?dnonse3&{%ZhRn^p!8EJ;z=iy6G|I7zA z605$(n*}Yu-}|LsiqMPTNWhT;0t@I1*09ra4+dv?=Z^v}b=AN+v-M}a6fiR5dB1i6 z#{+^a(l`M6?x5T^GFm?!`Vp^g6rTQ#QYmG7LIsMzlMo{7FzpTkSzp3$?=i=#V(kb6 zg`?YYvD)pcsrw|bzXF2V2UZj(7kGSGA?Gy9BR#hcajetAiQr2A#eo2d+6Mq@N|g)D zxsaBt_W^yVmYee`YYehp-L8;>_!!RU^OSd4vf}qfZb-h?Xf5CyHzS@EI!95KbL5=& zpl4jngCHv*U8gwz$Nc`Jl@x0!oIM&-`)|+Z_-i;yMepJjH!sJaOf&DG_By_%kyeVkR7jUT!QU*OyU&;M;q03(jOrukUTvXwIr?nD)Ek?|;qS|@nP z`#grP^n#l?`0d2?2%oiQlBGI`r@dlj6z7-qeH>ewx_%cRea^7xxT@wj9I&B^c|YKL zu~_ZzwkhV}=*|eU<_*BGk1g2F!=35Fn>lGc=J6kOvox%NExOzE@-#+#yBDZCdV$d| zr@VL~I<{8dp!Bt#(s?coRS;jYh-&!|=a;&3KgMj{ofT`AH@N`0e_zb%+|QO16ohEq zYyF6B3fYMi^X+XG00Y~QhkX@jsck-W2O8Ot;v`t2!a_azf}@7n52LY2{wX8&a=45tA6={$Cj$YXrFTj;W4J?tC`|D&W>vRgrTmne`Wqcm{` zY)Za|7HiF;E)}8Gn~(fXVJ()n=tn^Li%#m8GZFCHRW{A$UU8;iv0=6XxIiFANf2)? zbUufW@&kZ+%Sx0~R7Kjj6PuSO0LKE;z`##!HIZM}Wd2zpML z*gk3_di8D`>LZq~IyLV^!a=;*TaZ4gF;Rg_{Y~;YycX`LFJm^8IESEM$+S##e3%xR zcmnL@DEZUX^kg}HQ~h%g9?2s;>>g-0(}k!ajKjXCHNPvG?R}`AMwt;uxr}?|-tbEa zxg-2c)sgZR$Ntozqv|>YkJ~>&Yc2CZuUk^;7I=HHnfmpdky|ebSo0?(myjNvd_T(h zvwjmvts(P%nDujrYwb~~lbd%d-QuKD{$GF#9i{+iCG<{c zDyybLhtEma!-ZdYV2Iyq4(+Db38euN4a0;irT#vaY>%Azdk&SZtho@iO{?K5pj+-uBm`0)#Kgp2`G+c!lwykPdrE_Kvvq-$~^*N;Qcx6}qo~iONeboX_oQ4W)tx$wVn0WS|gz4;2CweDz0`u?(DL^$7pj4V=NqEJ}<0IlY|FtZW0*k~`(4 zGy`0u`~7+~TexQmG5q2_x$T#iLU z7f%j~_BI~P*XScr6kYw%?q!C(yv!Zvz`3-p7N?*I6%;N+cmm4yUV&qea3ouuxpMY( zGA~!^t>4ioX9+AJeACg$7fjP8o+?Ro%1%8HU-?#)>8a+TdEoBj<$^xGnAu4I--RI68o+n z^2ED%#$CRcg%J52z$A!GN0P+Bj*w(EJoZLm7W^l=E93+>teZ!t+{;X ze?h3-WX>6S22$P@PhQYv!OY;v>YVwdr>4pi>^iK7eYxHYmdWc7C6WAuRr3$BgS1wwEq7{j()0ebgIr2jvbCP4C|x@V>-88(bt1W z+Qfh1FI{J<8BLB{-Vpd+mh1aB07&2|FFcfX9m*NgXmDJW*2*`RomJPCNBga)fg_4zv#XXuBeZRjcjVgwAb%se4yIT&F&v`_F5M zSi+P=k6XTeU!!SK*sdMH&2tCrKTIxP6Aja+m@Hl6+{@=Xyp(GXRGOv5Sg$+pVnwgk&&v2OgrAimiw-aF7Tc+xQ{V zs%T7!N2uG?QRNy-$mR%jJ4ms%QjEG?BXv8gXQOWSNZpRz5>@MVDxRp@L4dm=z!(fS z_VBaS@XJPZ-+jWWz9&!`Th#5K&0W)G)a@FwRhZ{^99rwToo!ZMuSmm-y4`aX*@Pmu zMUhR^f9;x7U}K27-6M)?46E1e(r}_~2lefS`eLb~vAUgw_uJK}+ga7$ZNVhUmK85M7LgA;75HHEp7BuuoUNFRf9=ekfx6x4x}9lMuiL3`qHYKE?S}ed zsiH9@c2KvgqsmV8CFC@;sM|q`b(CV%?HY2h4N$jx%)&OPV__ShZU+JGkN{&a+@Oig zMcuBBi_NX$VslZqgEn_in^CuG%;RRFZdcRtX4bL1UE9+iZrV3X6gu8ca;04RtC-H~ ze{W)Ow}`}D5Qb8p*517oro{V9>N52xfUH%?!WTcNpNOC6qi@16y7O3A9!lyL-h{7e zJWkI+$?J&5j3w2=Xd;Vs%q%R4d)O|^^e^yH`hbQ%tj=udk_SQiq$&BgFIi6-uOekqc zO6;D9>C*TDgi$Vx)2Zc8P1B#)_IxlMn(ox<8 zuYK=VBF*>S@%&VoO$oZ9?|8&t>Hw^8Bl&J&QS1g!?E*deX|l+U%=_T*kpntmln^{f z31~EvBn;^JCR)-EfRDWKXIt6+Zyu`mk>Im>CX0wJ$D2@q#gC+syKC}p=}I>KGWsM3 z5K3_QOkvf9+Au@tVeVc}P0v#Yks@bdQkkk6DJ6WmzvHk0G+Ya7eXO zcvI0m#<6aKh*xY?Tnn1jIuuEQj02L&KrNzIfRWLE0*&${dNCg5**}i9iL4PYn>a1X zqo!t`@qR|7m!*Pr7i~UL>iD0%sfO#Jnt*=dwqe=;4XRSx>=Z@(H= zuWvbT2L19v3orhM{;sc5PniXug4xy!3sn-8NpT(%Z?&Q?2-$R;v=6$DBm+f~J069K z?n@@cKeisBF1g=TOVaVJtVMVYtqk%v2Huv*hi-}j9D>s(l9<~X*@bx?4EqMf> zR=VThc18ci*PS0Ne|16#moSM0tV)lrSV(4Bl0oY_`07sWe7V`%kx_mk-4+ zJQe(2#UK{1f_#4Sn)Ecf6JP$RR4hfb3h!yGkta1+dV;0j#0sGFk7A4BA*(0#tKBuf zq7H2KY`6bf&`pcGXdBMN!lNJj_Up)cLm;BwSLjKEC7L*Df6mZT%2C2hZ$o{O)Ik+w zUwVFaJ6RWTC>yOJ4AUkC1G~V$@`Y_;3?~x@C2je26jPH?Px`mrq6IJW6b|*Z;Q1N~J?Xs@l9d1`D;{n_)2KP6sm49|_ zn>`Y%hM-MkNQg!yvdUEQL`vbD#O44{azK&zA?%w`!5 z$~G{OC7fnc*H4vHE~b!TgD!ri%Aw$rN?| z)^0v6=hvD;_qbHGnYP@p6=^#dfVO(mw$PXMJ?MZfU<=p+wr!l@o_tyctbH15NdlX| z?BBdu;JpB2(|y*O0POzyvO+~9dCNT@E^ONJf7rS+YOq&OgDutx%fuiImH|JtlxH&N zcDew&kew`B|2su>dlT*z)S}hYFm#nYN5&;qwfM_R|>MV7tFvgFsy3~ zn^ThUe2Ynu@u+}|G>&RomKueq=bF}l0YeAmJ<>XZEh!Qp7Kn9eVtKOLQc6A(S?Vm9tr>e diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/NOTES.md index 9ebf32db6..1d7f3d467 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/NOTES.md @@ -192,18 +192,18 @@ Each row below has `measure_key: null` and this key in `conditions.proposed_meas | `ab2026__package_property_and_savings_income_rates` | 25 | Table 6.1 lines 61 and 62 together: F10 "Savings & property" (their sum) and F20 "Property and savings tax rise". | | `ab2026__package_rf_ab26_permanent_personal_measures` | 19 | F20 "AB26 Changes" = lines 52, 56, 61, 62 plus the gas standing charge (the workbook's column K equals columns F + G + I + J). | | `ab2026_option__lha_unfreeze_2027_28` | 17 | Unfreezing Local Housing Allowance in 2027-28 instead of prolonging the freeze (no Table 6.1 line). Figure 9 "Would benefit if LHA were unfrozen"; "over a million families lose out, 91 per cent … in the bottom half". | -| `ab2026__package_obr_direct_effect_of_tax_measures` | 13 | The OBR's Table 3.1 "Direct effect of tax measures" since March, sign flipped. F10 "Net direct effects", the p.23 labels, and p.4 "about £5 billion annually from next year". Differs from Table 6.1's tax total (-£1,749m vs -£1,645m in 2026-27). | -| `ab2026_option__obr_tax_measures_with_continued_fuel_duty_freeze` | 11 | The OBR's direct effect of tax measures plus keeping Fuel Duty at current rates: F10 "Net effect if Fuel Duty frozen" and its p.23 labels. | -| `ab2026__package_obr_tax_administration_compliance_and_debt_collection` | 7 | The OBR's Table 3.1 group "Tax administration, compliance and debt collection measures" (it includes line 45 and timely payment for Self Assessment; RF doesn't list the lines). F10 row 52; p.4 "over £3 billion annually by 2030-31". | +| `ab2026__package_tax_decisions_since_march_2026` | 13 | The OBR's Table 3.1 "Direct effect of tax measures" since March, sign flipped. F10 "Net direct effects", the p.23 labels, and p.4 "about £5 billion annually from next year". Differs from Table 6.1's tax total (-£1,749m vs -£1,645m in 2026-27). | +| `ab2026_option__tax_decisions_with_continued_fuel_duty_freeze` | 11 | The OBR's direct effect of tax measures plus keeping Fuel Duty at current rates: F10 "Net effect if Fuel Duty frozen" and its p.23 labels. | +| `ab2026__package_tax_administration_compliance_and_debt_collection` | 7 | The OBR's Table 3.1 group "Tax administration, compliance and debt collection measures" (it includes line 45 and timely payment for Self Assessment; RF doesn't list the lines). F10 row 52; p.4 "over £3 billion annually by 2030-31". | | `ab2026_option__fuel_duty_continued_freeze` | 6 | Fuel Duty rates held at 52.95p a litre instead of rising to 60.59p in July 2027, relative to the announced policy: the OBR's constant-duty-rate alternative less its central forecast. F10 "… and if Fuel Duty frozen"; p.24 £4.3bn in 2029-30. | | `ab2026__package_rf_other_tax_measures` | 6 | RF's residual in F10 "Other tax measures": OBR net direct effects less Fuel Duty, Bank Surcharge, capital allowances, savings and property, and tax administration. | | `ab2026__package_household_energy_payment_and_electricity_vat` | 5 | Lines 1 and 2: "A £3.7 billion package of household support"; p.17 "£3.4 billion in 2026-27", "£0.3 billion in 2027-28", 45 per cent. The sums show line 2 is the whole cut; see the discrepancies below. | -| `ab2026__package_obr_total_effect_of_government_decisions` | 3 | The OBR's total effect of the Budget's decisions on borrowing, direct and indirect, including decisions not on Table 6.1: "an average of £2 billion a year" (two documents); fn 47 "no overall impact". | +| `ab2026__package_government_decisions_since_march_2026` | 3 | The OBR's total effect of the Budget's decisions on borrowing, direct and indirect, including decisions not on Table 6.1: "an average of £2 billion a year" (two documents); fn 47 "no overall impact". | | `ab2026__package_rdel_decisions_excluding_household_energy_payment` | 3 | All the Budget's RDEL decisions over the Spending Review period except the Household Energy Payment: "£3 billion a year", "£9 billion top up" (two documents). A fiscal aggregate, not a programme allocation. | | `ab2026__package_rf_absolute_policy_changes` | 2 | Every policy decision counted at absolute value (RF's fiscal-event size measure): "£7 billion". RF lists no lines. | | `ab2026__package_rf_energy_bill_support_announced_at_budget` | 2 | "Support of up to £116 announced today" and "less than a fifth" of the 2027 bill rise. RF doesn't itemise the £116; it includes the £100 Household Energy Payment. | | `ab2026_option__fuel_duty_rpi_uprating_since_2011` | 2 | Fuel Duty uprated by RPI since January 2011: "around £20 billion" a year (two documents). | -| `ab2026__minimum_wage_rates_april_2027` | 2 | The April 2027 minimum wage rates the Chancellor confirmed (no Table 6.1 line): "worth £1,040 a year for someone working 40 hours a week" (PDF p.12 and the comment). The rates (3.9 per cent; £13.21; 50p) sit in `reform_hint`. | +| `ab2026__minimum_wage_uprating_april_2027` | 2 | The April 2027 minimum wage rates the Chancellor confirmed (no Table 6.1 line): "worth £1,040 a year for someone working 40 hours a week" (PDF p.12 and the comment). The rates (3.9 per cent; £13.21; 50p) sit in `reform_hint`. | | `ab2026__gas_supplier_levy_funded_from_general_taxation` | 1 | Funding the Gas Supplier Levy from general taxation: "save the typical household £7 in 2027-28" (fn 12). No Table 6.1 line found. | ## Proposed metrics @@ -370,3 +370,14 @@ The RF review (`harvest-staging/ab2026/reviews/verify-rf.md` on the staging bran - **Verification:** the contract checker reports 766 rows, 0 problems, and a re-run of `stage.py` reproduces the gzip byte for byte. COUNTS.json pins its sha256. Proposed-key names may change again in the cross-family harmonisation recorded below. +## Key harmonisation, 3 November 2026 + +The coordinator gave each scope one key across families (tranche 1, task 3.4). This family's renames: + +- `ab2026__minimum_wage_rates_april_2027` → `ab2026__minimum_wage_uprating_april_2027` +- `ab2026__package_obr_total_effect_of_government_decisions` → `ab2026__package_government_decisions_since_march_2026` +- `ab2026__package_obr_direct_effect_of_tax_measures` → `ab2026__package_tax_decisions_since_march_2026` +- `ab2026__package_obr_tax_administration_compliance_and_debt_collection` → `ab2026__package_tax_administration_compliance_and_debt_collection` +- `ab2026_option__obr_tax_measures_with_continued_fuel_duty_freeze` → `ab2026_option__tax_decisions_with_continued_fuel_duty_freeze` + +RF's three OBR groupings are restatements of the OBR's own Table 3.1 rows, so they take the OBR family's scope names, which carry no `obr` (OBR review, finding 9). The minimum wage key takes the red book's spelling for the same measure. `ab2026_option__fuel_duty_continued_freeze` is now shared with the OBR family, whose paragraph 3.43 prints the same alternative's £4.3bn in 2029-30. diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/claims_staged.jsonl.gz index 8c24bea78875c8685823658c99f709c10f866bf6..40eabde2ed247d4c9f8010fe0ea20db80fcb9577 100644 GIT binary patch delta 28439 zcmV)dK&QWoodb%U1F#GWfA8?d0ZYP~(6JZeI|#U+V=u03wWsWKm8*n&O_U{pIl!_j z_di%eqz;W&jH=mlNUJESNtobCCTj__Z5PNNzyELS6yyf^hM`AOTNCgWX+O+7pK|vS zD_Tgz8EM$lmCpXXt?E@Uswy?iMWs$W&2n3c-gBMOigWJ27B#U?e^vX*wL{(-F^rR{ zy%ATgUF($Cpe~6Q(G&5)8VE{IzAR7{2S?yU?7%u?2&Bn$qiDREh7s@F1sT!gfzp6^x_FC4xB-^RvGy7<;0&3&zhY~e zSvdX_SLouG!mG%8pi^(`C1*49Hs{{j~6UazB7|kR!mlNIx;-Y|Ic731qK@^FNjB2|Q!TP`0q)MJ&Ga{DcSX z;ZB_@;@Mu}?q%gHj}TjnURDqj`Zj9Ctb(&oi zzNWTk5I$r*LrS{eW6KfN^}fPb7}lU)d_SVD-($_jL7#85{8r>-7rN$cMCTbJ ziRf#(9jgZSpOnmm_a87vTvmw-maJy*L@909 z*+5j}gaRXUvy2dt2nELL16HwADWb@sNPO!!9Uq2AKsed({h|<#m7Hy4F#QTAD5CVO z+_v17iF)a68{j1?0$7n8H-uv+Z9y|Ur)?&Y>!OgW&5pXZZDl9jPMYVu!tdABYYX0Y ze*>?^%Q>5bc2CQOdD>P3N4-6TSd{qaJX}RBbtq8FxH#*pyuQ?sD0cHL0%kN2he;p< z^?fwPdP?8OZH=q5x$yNW8eOZDXlTY{Y0dtDy#W{4?mDMyTAsr`j!C;%{Ql0n0~jNA zXFVM#psd4==$_I=_VSmLv_aP%UH=d7e`1}k?G)KYgV&00(t;#x1^{W%1DJ<46M_?$8=dU28} z@jz9nJ#`0zu|3wbNevJSa*U4B0*Wr+_vz9nF-0CrG{((!=P} zC=5T5cQ*OS1{-i-_0{2UxK}Z{f2K=hy;5Y&&Qk(fnOyKn3kH@biXFmCQIrCXybmDl zg2&8j1X6gdJ%AL6a7j+T6aw+!gjN%r8A)5f%itAB=p0ZEq>Qi?!@1@)P97P8y_V}y z{4N%Zw>k}&oW+QRQ{A1F7cT-buJ`Y6KJxqse!Y;@?*U?yOo+q4e|%kNf9A!@{Sy+Q(0k$Rq`xt<2YI)>EEb#$2 zjIx|`{6u&_1Pg9y5K?~etF0%%E`iq=Bb$046+)7*ows-A1z>!|-e(Z2AeqG?<1eDL z$3DRPd7f9t$7biU1wvSNe?wOs@}_;-#MwTZtY|vm6cRhx2LRDCdl7&^wSmg7OBiiWp%RNlUtOPA3kj8s(iD~ zRPmSEH4N?qcc=CZGW^^NmTbOB=HM+Tfx5Z?*RT0bgww}@-p%Ot3hFy|?&5`n=!=4A zm$er}fXxO5VXB5if8*n#v3b}mC$=>WO-LFD8GaH|^G5xW z{^R$5G2M#9Cxz~$SC;N$AJkCNZfZlx6M&ZJ_sFI@FOs-sf0!y}Jx&8+>No>&vBak3 zUczSE zfaC465L)L z3#<;)xxp=XfZ5Zm)M7-aZ&ip>F;7v25y;PuMAbDinYE>|`YJ&NiHX|NNxYonD}4Q8 zb`?XJa{(@K3cn^F-tjGu32*hxFiXsK=^LYf6j=kidajb&?27$9)qrKGFV2)|_<)r| zZcOIOfAI|e%X!4;8ySbLxmeT6&6`YBKehHTj05cY!sa#9%fT85SC~np=h?3lIz{M# zqg;etCtwfQZu7R6@GoG5%zC*Kqk*;ZhF8VL!>UgH`2D{N?T7#P{a?in#Et0Gj`ldy z*iJZg?k0=lS>?J!*<+yW16KR8zNw(SSuZxDf7s_2R?S>ukE9?PSJ*|}sN6-}c%qB? z*+)&L?pj;)?pX)2=fcJ_CO(ojD-VV@K-pl!+djBp@A~8eg%9Ia3}26c2YCaZ{c6g$ zMBNv0;GGDf3<*)kfT+mD#*uu>aJ@i&Ry*#&aVloIkonX4!wuNtbIVdexQ?-qCEx|G ze_j#1ibQHMcx|=9&In%p8tf$rMwbpoXex)!1=(ZPtym>63hym+A0ou+V$q&n5PbX4ne+n3zt`$@nvCEy(c@X&z({Er@0 zHdO%D?k5pu+E|KG2p|Ip#kBk-jtq+@PnCuj-!BjhU$}T6TuZ2V9uPiv)>V_hf3wk} zu_P{EZQNG$Pfa_GulLbqoS4p05UG(}#Z*W;KQMN#5uJf6%{}p%5$2W<`jP>7@3{`m z2WW}A5_c~TcgyBL==%V#ekE>#om{&$8~xcBoOMv&0Nw^0uWv|%+)OcXhdFKMg^sAH zsH8HKBH5lfh^IGtKP;0OdAp2oe@HoDP|+p%1Rt4GbEHgn*yLaN$auy%O3pbOG7(>? zsjKX7>Xthr8k10iC4bKg5>-jAK2?NTDaWD-9pO~ogd*9UtP-d$Jc^fdR{fh}(3b3J z%W{Xq;bg3vhHbf~F;*0PG#XAPu0ltnk!~4-@yOH+MW5IsdoVWPZ*({{f2XE3oY+I{ z01PVRnZF`$0@r_x(?LdW7-#StCNW&BU7!cN&K>RS4L15qd6^-P0;PO=9YN`h#&U-h7#ceZ43CF7tF3!hlH=}U@{=2l zXZ$ye6cTV4|B_C`8vqLndCUQrDw#^mC9F-$)Q!u0bX zSVbTb6m;%k^;nFvhIVia3FD-fcCM<5UD1+ra9S3cb`h!IloRWo0rN#+IDc^wpe2#$ zEJ!3~#o{u)$4GL(oBf6SK#Vg=$Dr2L<5 zO<~Ml!Or>xJxzDOpTe|mfBmBAeQOkSYP7nln@ufVKRETK$t-R$_)+@Sp=DcJ44;%D zdB)!Y%j&~rEX|+ufetT+F77`9i$FpPibRCTX+qDj6ZecUJHhyYj$Zu_gyz+P0;<_r`0of7u)dM?SXeUv@3k3U9q zj~NE58z~bG_f9z6l2W*mMJQ7TpO4GB#O2+%{88jyt#IB-PGAr|5K%kB7zglK?<;xf zU|S?RU2!hpe|}bSkvg(krv7S~XJV(Udyyb2K~xs96u>ikFWBLg5Hn`qz+Q4lEzNWh zp3=DPxmXNxb)~1P6=t{u(#BR<&c(T0A#OFH?sYi zK9Z@&+#FBMH;~wtV%rQDhKQWPXNT;~D<2nw{J>`s#YizRp$VESEMMRVz0fRZoBX zyyO&`jHO-@RZ*C6QwV}wQW8xW2+7DqmB@t4iuqQPY1&pEY3kfiYK>MAXbR+!b3v-i z1qu0Fe~<-h7;x)DLC!RdqYNUYf9om#)>D%-(##t#W>I?-GdA3SPUKLH8yKo_lpd=H zd>bhHU2uo)%U!s(HJK5NJD_tgi5=HTLT9oHYw)sn9&D7u^hhbJ&5M|JXpXgnva>?j z&V;|a5M30ZsEySWlJ0oHVgPlWGT}8fN?nb*e`UQ=x1-CBndm3R6n#^7J8TPwqh~2{>8!fhbt!{t*btE1lVj}IarMa{6;%_l~ z%k#-c_cL3sP|-2r*Ndhc-vg&Ex#) ze}ntcl*E&Fw(xqk6)pVch3{=db1V8*(Kvhm?K{Hwg>gIV&Zq2X!$7dVm`uCC)t&T| zfu1!pBX}XxQq&Nrx|031z#y;l0}PzQd_Ozd+^a2iQ*!_ug*P|I!;A)8;OOK$DfLo8 z9;hIT{2+SflzfTG%y-B!`inpnXcZGSif@Uq( zol*y0PGv3AY>6;rO7>FfxFh!tF!tyeU^FesHge~ffQ1qOVDE=GP5}0m4dPiup&tf! zDVP7{hd1y!rN5zt#nWv0)NA0o>|08azF|vEZqAf1(yR zTrmD{{F&?E{c%b7Mz+ImTY}SoAb_QUV+Bi|FKSqTkFuiPbtA9glTiO1x4Hz`fEtqf>hmf3p%d}Hh3O|;V1HL zNPg0^bPT!LHEpe{np$%Z-H?d>e?*WMYn3tx0RwcGd^mFB+Y~AaJS-1wy?Ci_88SLg9n-54!c?yot?8SSI zHmCsSF6u@zQ4@7Iu>K;6v^1!+8DTmTcgKkMCf6lOK%HL4E z%{g_bV3J1%Uy5%suP1RV#mP&>c6%ivFV47B+>b;E5#=`Tc;@*Npi31HOqYkZ+vQ!@ zs6X4RF*|!@Z5b$<5)^lXB4d_}tGrmmOxEhaw7x||!Y{f@nS=u;jH57)c~IxdhS23N z>+qLdsTI>odVB3C6)UCQe_5U0_bO?}OIQ<_S9H>m&TJV-z7T(Q#iUwgUc)qEdHY_JSP`G3>!5EuP1*iDS&@M zghU~1hB%*V97bmfBQLNuN?r}Uihg{xHv*GliWj{9_FY%dTTr=63LD&0*%t4s6y3$* z<#OSN?j(MR&wZ&XFE{SGCDsfpVP}IoF3nk0A&)I*XY)1xda+VhdgE^{;=DQ^oE3%XYHtaInqS4=83&apI3Vd}SP6lO+< zp>2kNzJNP0@c0Bu8vGhQ@}6{fMG`J#`1cgy-#wzg@~kHlm*hcD9`vNzm1g-~1Hm_^|J)8EP5kp=9%77tTf)gohx;QJ_KSujNlP5YUx%i`swr-#ekiEqp7Q;Q*+y||3oS> zYywRqXy08W53UBK3JJbnzizx$*}1Enf;Aq#bU?|f9lnRAZjG$*WMJ5X33Z3+bTDvL zMF+T7jPckWe;MP+glg8%H8pi=X^JtR`qGagvAXDdM3gH;)pxo=X%F??y zIti16w;lr%ObpY?XAo>^ge~{E>^kp*W|m&&q=t21W&|G707xF@gv;l@(3#OPaQm$Vpf(Mb#cYVOsEQmYS%jqcOExFrJbjli(Z8Jt>1 zu*Y{Z#MZTTwtQ^S*Dv|lR?hYQBUu&D5+n4l+_iz+76YGLw()5m0LtsMYpSy#ZLnPe z()&Peu}@2g@H$`k9%d#phc2_!U{}%Wly3DQe<*w=n2o{*9J6lF;*hi$Du4ZwO)GB^-gt)zow#yOM}jKB zxZz+t!j7=Gp!MC)K^IeHbAeq0T$iviB&Oo&cowj644I+_NgA_CAEd`0zyCMl@}0Z* zf8+OmUHBZwUC8`IT~>j{XUf+Ew9*-w}`1ZTMuo?`EtcAg|8>a8q9dp-^m0`v0?Gdyo$Q#4UiN~kcu5t4< zj$}~wb{4=_np!DU+V_$pSiBr%85YX`BtBtPmRM;AFw%z8zQ1H!jec@(OuNKq1#@?Z z_(3+)_}6fRZqkUxoudKO*3OqDFog(pl%Wi30ew}1Q%&IXfOSlh2aq+Na>Ly0e;ezw z8AzUt1O}MVxAP*!E5M+V84YH?Y5=`RtqY8!7vnL=vu|g3`1AcR!V)(gwIjIsh&x#3 zmKkrCjbPg%(h|WgF9^MgNGW9H&WJ0;D0u10b7btD7}b=aRXc&I-jNAoknkvW+_dk2 z19Zm}#~5~P!NVl`y5d%QikGn1>ZB~wG*z}%t%Hf`|v~AZP zzyJ5x2Pq{KUjL8Z|Haa*L0h|UqU0igtcAH2-hv_LTl^3%jPsTp^*YvOfAhpiN59|< zt}%@gZ|aHckfR?k2m2@A4C?NFE@v#9W5z`C6B;{W!=txRhuhp97+ zXgE@N)cV=7IO1z138b?Nf7Y_f8^>HY<7MKoeE$iY^)QSZEj$O?1vwK6FBzq$$n{2* zSgF8!R4RS*S>4P^OHTRs^2TZs<{IrNn_WxTsMEYS<;BI1BsBMP@3UH=3kNjW+eems)>!g+f3IetZr)P`aq0Ey z;7Lq%jCf6FtH}kJQVktsZ9mm$2j>CFMmy&%z%=(Lz%(zl(T?n(voBs746d)CXN|<} z{S$!-we#j??YyqY+W98*NZX@Dnmn7xUNrJxQmW_<)pSyWNew17*t0a4UOl)8*EV6A z3xa;`=VyU%-UNVCe~WogITDcngipvj)(Pncz^N-3=#pu*9q|4LZ|$)DPY#B{9Qx0t zhw~CbJ{h6*c`g5$D*KE`+@(-?%GS@CLX^I3Y3I@amdsZw<6@Arv(lVWL6co9gmea%ZZN6f-k z^U_^@`rkDgzV@7I%npPnI7PTPf$^*!(x59RRYL%|rkP&Pt;l!=balxH`73L9829%c zm$ivyi#V7ZtGq*E*ikTy?+%G!BffDUK9f8oFa9d0T9$hi7CfU2YfWLUKV zWN0_hVhjU;PNFE9Iac8aA@b24!r#4SK2|YN9{a@f90MYMzdFrdKYQ5p?C7>M*i8zfg>T4 z>{(hDGNA)sDR$~FFzmHQIra;4C>H@X!=A1*8f-|LHN9GGhqYo*cA2YN4Va+&byi$Qmr-ij6 zf2YkniHD6ErF1HX+j)jCNlZMN>6Ukx&qIX(c1(P(7wYtbkWRl5`>pnLlp`2k$4Ntx zKy$=KeLi*XX}m=rVlIl&SGOdiT$kZh+K*JN+>cbf?0%$Y#G%*Z!#f-@7-Ll?gC+(PgR-7`b z16AX@yO)=rL?#%l;*MG9pByt^XGR27TXTp40T?JTFe^er1 ziGCc;X&7(@#)X$K@LFxz3g5G7(!8qi4RS0iWE2xfnUvrK7_Bsw1!?R3VWF-q30b*w zUjvip5%ly-+zqZSUijj|hw&JDoGfv7E~*^{ZZuq}#HG`~@*+j7_XSX4t=UBdaQnj# z-~QOCG$m|phF4zuD-!K00^whOfBpsS!DJZ)qL@7$j9Wf?Els*G4~Q7l1dF zQTQvLw1R$qOpqTvvKcEx?XMbhNtoa%BF(;>0B20-MdVFzon6tMQ%h#`jF)rPLFJxq zK$Y6ufjxWR)mjglG(m<}P#hJ|jxdWy*)jBRqZ^Hcr$)n58#_~gz##@Pe=EsQ6yX6Y zaA@bkVQ9@#GD^l4?tmahl#}*+%yL)#@PYCwy1923Wh6auJTWcss;3Yt29QjquJ7+C zg?dMTlYK9DxwQ)MRfBvJekMlWr zY;}BMOjJ%X5-;ccETVU~%&glV!oUBfJo*a%_3&IdS-O4?8#3oSSo!RX zG>e~~fdU6&F3d5T(D&^-?wKKDcz=+miO1k5ChcheCg@&K1U|E|BoZrHO$$|nFLE4v zQ~LGl$I)Mx^94p4@)TQ_h^C*Ut=WaxD2u3iNhU*tW5p%df1b>4biBa9k1K{Bw5mOr zH2d`CXZB|67}q7-xrzM&yF?)2>Cld}|2n;+bMo=qw|{>{hFueM>8F3p$eoLAIPOhkCJ>^kN~;jL@0Q(csXCB zJQ0Nt*|$>^%is1%B#d1)m8y*~{>hR%>%Aa5Gj?`O?^%naIKR;-bTDJwNkZpWpqZ1N z4RWL_N4Sbei$hfM4I5*?BPQ8C{b?ESqiUc0BOBRclOI%#%)~+V0Z2>P_PPmCwf-PY z;^NV#f3<9`vjt9U3Did^y+}(gG&XpPY!NmeB->~H<6UC69yuuo6%jT}K?Oc3-jmxO z7v9}%kCtvtj!&nEoW-Zr)f$9VML^mCELP0n>(qtiR|eG?+IET8r+u;-$R=7VCh-F2 z%H%goIR zcerfWHTKTTZfcz`iujgfA(`$|O!p}uVj=TKA1L!Ki9<1~mgh}*!Ps9;Xwl{s7DyKd zKC{WtJQpN2RQ8YR90N^dHf3#XT|!vb4C~+F-*G-|igP#PleLWPR1RfI(3X@SM}A zm`z8_!qoSNus8LlIAFbPn44pNxim3B$uKnCah$rU~L5t zGTeP}4C^uL!S%yWv#kScy#f(0u&0(hYj2*lkz81e zq|8MnlCGHZX80W?4%S!PJY7b3ROXdtYp3E=Vy>)?Sw8^Z3dFf6U*b~)q{ z@Tw7=@nj5UF6>%}SBZ(Yjo+ zX{*%*b+Dk|4o{+Pl1~MlfSB@Vf0(A)R|sU9YFDp90SZVUI}FHJRI~~BbmZzd2~0<5 z4~U(D2gZQ#A_yu{zSyUh(z(W|uJE>dbWw(@n+i-?Pt(r;OkxhASU`$U-5i)OokKMh z@)i-x^T|i|GqCyyCH#~irAv^KAa$`IW$n}I9j-o8wOk4r_i+jcUBXJOf4TXBZ6``# zpr#a!?}9tu4faH`>L)(DdO5m_!sTKmE1Kn<6yIv(VhhqYXv=?#e@oYOgGWX=1H0C7 z8$wz6k0eg;_l1idFiYcYUvTJ#@Rnr`v25I8!9ch&aI~sdZdls+aFJDJaa>H%tB$DW z*+4I*VwCZi;`fA|rzq#Jf9(d%Ks!UCT}^10GMDi37h{{c$qF0`>1@{>;WVosPvP2Y z#y1ID61F64H59hWTJKdovues^PI27Rxwbno?aqxIbe>^v4acYVkIoeK8Lr2Mjm{l{ z0jxK1Kf6b(vfao|dsgM_8S~4~)Rq3W`Z8?cFv-rJtlkb7i|fq|e>TSq+?}w-l6Aia zcKJmxI{6DQI(Z0}K1Q+#=Xfl_3YT{?BAA}6F#W|^a@C`s_Em)#wlubSdPLf0kRb~$ zaCf3;>8agjXaO)&Op&x%JUC&uWX)f2pE?r%>grL?n2qz*D;8$323DOm+l!y!y;)~l z&exml+%5X=_5!jOe<_c=A|FdJ`yq8Cy8axq*g+iT%kiw-o9M=CzQkIi96;7Rq8zvh z*OVi2T%6oi_k071B0GSRRfI{PJQ`5i#^4MODJy0Nv9^XYP%o5Wei5vOlX@)vdU(8)xcV0vp|f zD*`?iK95B-e*zKtmf%P;%u5&Dl#psu-X9Q@TO5KUGlg5lXULm}+__#L=A<3Hi^)I& zX43)Fv{y!9Zj%VoR`tG4=Da$ylRE}SG(mf>eA=6Y)aNxuF{^qe%`=Qhx@jy$6r~Yt z-!ip2*PwJ>Yz$yY0F?kL0rbcKx{5iOQL5|v^&u$kf0G2$axmSJ!n4ZD^&)PXd&9@? zF*Ei8Lf#T|kJM*>13kOE>40ISAc)Mxc}4HI0+XLh$Rp=0*F^zz8O|-91HO(FtbG8`kl*F ze}i&`owo7uvRG4 ztCnCDw&^OM5GUlK|CIF{l%Oa12{2CvS-$op-z3p|@an)d>$c_=luN^vI3 zcZ$Tg!YrTP8^Z)q$UorXc{ztN00JP#$mmqsS`P0&WWTN@v zn#NtL#*T5Y5emAK3eS3zlmhur*Fl*RDtp;{|5)-TCJyna4bvk8+Yc99+0e0CM{tGn56#vmR0<0=K1uYWYoGZSC>aeDK7-{RB(s=3oOw!m>;uAVs+i@Q;h%DdVk+sf8?qY40n@{cGiZvM zha;Uim#dh&>!YS9AD`6wc9*KyP!S8x>W-=?4!eWnqW5IY?1xRc^I-cx4wz5nBpwop z+l|NVfxYH7ri61XQb*W|JsQ6zD4efItI#et zz_O>$NvqQU1C=re=OC)ff=t0rzX4t zm-)9p=OiMClX2%5j*)mlc~G%{$(I@qlmK2^TN_Gf46p_}fa5_EVrm;`{s3II&CzgM zUI{b<=WgI6$^3+2!m2Y(3tK0Z#@WntsWGI+xbET8EP;-akVP+!R5E*D7)m@4XdL2nAw;=?4A{-6O45}eN|)9sFgmSXroaqmfiVS!5pa}&-)J`Vzl z$xS5y>aNTw7IpzaC-yq;H~67Q`*YO*e{a_RR*x%Of7p-D?Cv42 z7KHDYFOCUjq=v!1L(HBF?O7wX?S?QL5d{22;VqIB*j3Gx(0;wX?|h_8y7A;W zBHl=hKcBB0glGTRKIukW;4sJ{CgNl{R#FF>jE_gx9A_tC6X4H}@TW0F^khZ`vK~vn zhl3Ru!G8%cf3N-r@!=ca4<9Da-9IV=EB_|bHR+e{5 ze+wryIHzSUj=%iyhD_lI_ZyaH%|N@bPX!kEbjoV{83d4aD*@JgZ0(mNT_?b*5@3&& zK!^Er1W43X017KGDYAudpVc{R@@%L5O!$g@Ct04_f0m?XpwR|E7KzyS3X+1ATJRCa zU|-JBsg$_>YmS6)OB8SccwQpXaECoz*<}Mh!X`OWFNzZ|2Vp1ZKbLr92fhqetv(Rx z#2b4HH%Rt|v-L_3yA=~Amn3!{W6}~T%jSH|R?|biAfx3Y$Xjc_Y+zkE7MRJBkGjAN z(;Ea6e|s^ZpBKCZYZ7KEeYwEjQkE#leTnzqb2a=lpgHe@B`N17I4#scHf%@zGU4rj*95PM5?X z`S1?FwuP_Tf@2q_LbzaL0}%v*_Q~IoS+HXDj(6les0Kr#?ggCD6T$hs5iT%R%9aIV z;rtML)=Igdxp(11PRFJlv?Did^o{;WAN?ihGh}G7h(bRM?otNMkC?y=fTAMx;59#6 zfAe`ytjD%DW!FBy0?9IfTm8n%;WHFZi<-Bg_!B)J!Uf|U$Dg?l-XE8QZ)7|Cwx!{& zEytfW5En!+VI}mXX#x8WmTbA{Zj15fs8*Sk(q+#r^C}d}yn(u)_|VVIM#LIa!J5yShc~~P;Gi0oUZnCe{DXwT7m62NPXX#p_cGko33 zy=D8>B?HS!MSwc&X$oQbA^Xl@f5_8B#3{H=CGa#Fcw)*t?ASeyPuAjD?`EIXvZJJ7 z%N8b(S;~*9iPk~vAR~DBitq_*N%$KVkqSsswN9ywc#RIzl&0-))OxC)Yg&AR5HO)* z5BWHDKnQT>+-N+@p0+#I6S=^GF%hvB(vIk42}l8Uc@%{YATpgUV&8{%e+1YYcu~0A zRE4yWC;$4*hgRWY76&0y-&-uOrB&{WgX@1(P15=W{vM|na7N|~)vDxEd`1_p*L-c| zSe*4(6z|j}-vPzKxi@BsqI_G?iGW?Zy5z6pBplJG*nLs&tD6={3E6eYb|;0R3Q zU$un2Bd|k0%Pzn|e zcZILGaZ;jQUS!<$iu{x|$8&=KGgN`{482nDf`^Tk%zqZqtoOQ0e`c7%GcA;7MRe|Z zfI@CU#>FQ?H?|VDyM&Dc6Wc%4+AeD_&Hcu*>;aQbU%BoV?)T3;GUz)CL(k;I^ zWuqRVZTN;>T;{aYe>PIvl!axKvuc}^Ri(D6N86bFLDRKOy5+Z~ZOp+T+Jww7X&d#N+9qRFscq`fHdfzixVFi*{MNLMWgVhz_=YxF+w4e^lG;XUo062X zvuT@@Ri(D6N88wav+3F&!{oD{l_8A*Ru3Xl{Ub z)fExvCD$@f;23FXT3^|!aJKKob5zRqyY9>i*qPmY+39DRa!C3a-?@frmcx?)q-MD( z%`$9^W|2t&fAwk>Wk$8As0aDN)CT*)o)O}gishyhi`o>$B18SPE0$rS6wB7Y{uL{h z;bDqp*j&YOcr;%smYY&6T4NN8jOy2`ShR+@!nVfnf3H`uX!aqBMQdt_|L*=-d{R^d zkQv=kKsbP5F&np?b=`lJH8SB{7|}c0d0Kh!R{ZNj2X@UB7`eeGCtHnUEyhkZ()&eS zP+yw0^}|o(W5gEq*iG@Fkeii_9xzSP{`y5%%sKft)+JuyKWG~K@b5T>t!vok{Pn7G z`;##we`+88aMpl_oEmV1sR6+xHIl@j9_G79Ij;oB(U|xK_?-Nb@;Od<0Vx@=HmS&j zsqLfZ-3#IGg^qpKG*e)we`vg%vsU)^blWJuRdS!ou4jD&X@oPZmoB4XK~FJ2!8UMg zbOeJ9Hg@mB`yUBQuO)wSgC&R_Hhfd-8rizne=#9H2}BFeSuU{EAZ$NR{!Ld^6S~)F zrZ-_msv(i|D1BRTLgfMuPkfvP+JLEqYf8oqeDp8AWlU0lSp(P~r(}}L;S6X}E{iWN ztE1w&k*e&U2~88jiF#4#o5KZZv$+MTk!i`!0m%CNka3Qz1;0y0Z%gTrt`j0^?&o;If#Cyiaz_We zN96j&lcPxK=3Tt)HA5JE9mb$bdPZLx76vPwFJ5cN@c_p_^B;>R_f(kq37%WG%)NlG z!YV3~)Bky0)GI4I3H+r*~EFYbtRAoP!;8i8_D- z1pbKaz#l06^qL@aaSuo0erCMJ=U04$-zvZ{I^liKg{{FE$&*p0^aAGX>oyNBZ2B&;N_Iv&vHvBys?|rtUep9v| zKL+A5&na?#&i3e>!1?fBzvu?aK+eVNPjH=rA5-AA@R3sT^N(-iYiK^c-q*;e44%3n zF|#hr%-JfPzwl!KsOj=S+(Aa+fB1BeO4_AR2+NrRsk}H0IuaHoEZznzs?}{uRk;dS z%xx#;19F4ZZekOEH*l}2{g-!C;P+=q7;XwKFZRT-mayOW099odm>~^`E+OmMVoC)g zpLs>^)8!X!eFkg@V}f6{%%%?LL%IEb*+=85*p(9JLik}*@h?dQd^rUbAbf0mvbv9WVO$mqBgRdGo5a(+x)wWQ5>3|xJ;jM%inEK@sA zj{$y2xi6Kp4w<J&bOr*h*L(yGBR9cqcgdkKg}AJ}_p!D!4l& z1YIEErOs-t8SZjlBn@bE_vkF#$2q=##34Ai=oX3je-w4>Xv zirE@te>s8sO@W_de@yvdf1A+z(7y*^y2Dhg0N*b@+Z59k+%zO^))h_(!Op}C!oHR9 z9Y43H!c=l@{~_gm@*U%T@}6`5RRN`D9e|VtGBzy2vir!M7GV9dpm1;C&kyw8qYs2j zGih(B7pmHlI+|b-Wmd*YKO^4YzHYuyAH?-a0GW(IkAhe z&A1s>uL1iGr$H<=KncS9_INpGrm)quY-LVZiNPk}z(oG=j=+xim=(`NZ^E|=FTxmY zY9fF8WF*+z2m|+U= z1pmY+e;BE~A*|=SD11m}dl=8<8Ju<4vanJoUB4sNRU_^VWlZZBxQ83HY09b~dRbY# zaL|IinsQ^0b0bM;G{1_hO?aQQ=06KT_3erlYLQqM1yt<#@!pMm&&`e&MU)!53lC9g z3q-vwK-6u+kqjBt=K5p}iZ)p)1*aHWysY%-f4EBMF)Blku@ib4AM}1MQaocK-2Tz# zO_W*0frkPQs}HBn-DDvlrtuIniM%O0`D^_`?;kLFzW?@pAsgWY*ivCa+mR3kUcDm1 z6p3_cr!YKvt-Ice_6<`r%$K|4gvA|Xq@6a~Gdj-(t?X8!p=aq6<3oH*C%1zzZBZV} zf9eM*`gsAHKv`aGdPMtm87D;Pn>lC|+)<3zceOv*V&L?6Mb4LI`6stJV zicTY8PqD1NUF9ZQc(HC5xRa+#IYlNrf1$-?SGKqs>rSdNoFj&HgJuW|@}9E_D@p=L z+AzFpPXE|@e#9nLWf%nl5#zEXJ~bYne9D5j#o#d5)2!D($}$oC=MuJx0S!+%PI>w2 zIu9I@YoVT_Y2kbk1_E~qfw8q%*#3|C3Si1x6PMdiCE4^ArK|U?lsod9&e?pV-tQRpx&2nQ`gVR^&&(f+M3a{lzSd74Nup@0z zHsA$QpY0h4pRi`I?#Z73&jL+CLO**fjg9a%#2tx~5+|>_(jb|!NlYwWsb712dsBEN zLDz2VWG0%}$;7suiEZ1qcbv(@oM^|kZQFJxwlU$q=lj3soO5+9dfim7UaEIhch_56 zqKl~o{)2`i>4!HW(~T-8z;GRZZ+ zVsHW_Pfjp)n~C9eX<+FI702Dj_cyv9ZVNERS~tu!%oY}!USl|XBDHI|jLE5uait3A z?hrIgOn=2rf(G&i8MZVECi2j%$XCQV=hn^T}M__{X0_b-9vBb!QN>lFXAZ;g6NB7 zO$9XMR)^80&mH^RBJNFVDt)HRe%v-G&|*ZCDA%WB?+U>sXe1fYaoSBsE;Fv!eY;CE zlHKX=RuILsE6YDQu`=jZ0ZDq=%I(-Q;hr*=Y{k7GURoqFgQ>w-9<=!UF)j?#)%?{J z-+NwDMv)JX%%nayxq)OQoqP7SGA%EUc5gUT{ay>bUUp-y#jbF^qHUKlV}{fD;?g*7 zGVN`2z1seAnBY1OhXKOJ3L2oQXon7JcD^Y`-ivj=T!f&K<<#;wAcki>OjaM^6p8#) zSB_C~2eH%g{2XCaA9d?Exe~u~*DpNl+T7{)H2IT>oRN6iu?2GhY>x2r48?(-uftO5 zB7%qnHy7Sl2bAixq^MaxnDYd88DHuQ0`%r7tOe*eAk07wMaL3UH4-e}l3jMQW>&$C zZtXr!4dbdIOB8Dbfa%a}=en2GV|iqLL^g$O&f`mUfG3p3H{)uXP)D_+2MHMlSgV)2 zk%v{^IRrz>6n@CVs6NM!x27YA6753f1DDCg#e=uuy-W!W+>Kkn*z(D5?n`+FnG+Sx`$D5m5kWO}fGZGF0?;ow1FqHF4RUZP?< zPlcbF)lxE+0$KZCXWDJ`OU3H&Bv^b|h?67re4|CQzBDKBxhluwbPzK|q#m=jitL|i z4GCBJMk9t*75{j@l;2Mee%ro7xxXOybz2q5+zMppjoP>eK0>)B_bfd(fMsvZ7ZYA3qpVYs3ccJb@xG+A;DaYe9A>wct5`fb;5L1CoA)=sb%l->9hd4!q znHsvnej4-OYW&5+!j`fFY$t{1(n5Y846vdZ62o`?6lIWKKQo+U*%8;X4d(5)h6zvj zrz>Qev)gDXoAW&=(Yt3jlwXb(XJy!lV?lM#(x5SqM4N&e+Hv?VKZ?k=o>IsH*jm}l zWI`xjc6f8?yvtF2!6{99RE8A)A}$0h)t_s>Sl}pMQ^-mh^OZL5*1s%cH?r@;WdX?4 zJZo8SrjVKOy5{NcnNOX!3#zE8y&=`HQduiCutTNLrdV<4t>y#&{!I(`slRf))1Vg~ zm*l$dkV~3p`X=r-2977$myk)GmMy^xMnPK+)uLV&Lv548| zt$#s^1<{M0l2i@5HI+ce&ZjPBt*)BN3!5VI3F z5`f1*4O;>eqNZTQ9NR9sfIu%#g*~1ZVVD{dNvF@+9JbRS!)S#dX7cBygIWum9K@hC zEiSo^pE=X2sn~$Gk8ZCFZLa4p7`x8kyJLga$llDY*an~Ub*wgd@81Cj#}HR~`k%H9 zH<|aq`t-0~LjS}-=h*;j8KL9=8gMm;QzdY<#xlqdTWjLSefB8@hO#y=I@y#adqyb9PgaJ{1HRS2~>aJ%+w44h4wsb6x$pbf|m#K9U@W z7h{qg&;T1^rdv+e@gp97)qx7}*c}%6$IToAnE6eB-VnUfYw$*;`y1Oqy(jK20jHyNMDK$zg@;%{i%%NBD zy8(o#3V6E&(#6SZV2OV|@yoqeBPx|oAQesQw1$_jyl6Kd50_(l4d}^i%ax~XJzJ;J zxTYq1fIB_Y6MmRgY$+e9gKTK|9p53O3=Pw!8E=%ytnJ|hWkGuudt#hwGo(`!-atQJ zyyU{-tM#(!S3(4>rVu)Ktsh|(_(cm{ZIDIh>jBineg01B4;*_HsV#{hrlPWMHFqGs zjMFXp(PcltX8)iQYT#WmL`V_EWiRbWA%G!y?0n$%+TF)(e1)%!1Lb9ZwXR7YYeWg> z%!@g~DuMfE!wF;P#(NU?<*}hF8G>YKtk$;nIsN*pjN@H1%|e^jtCph~=v$m@b94i- zNA?%jF=%I1X1ZT0q}vZMownC7+VJtu3~DHK2P7xp^LjFa$ce?eE($_GpYG^NUDlE4 zIj*)7fetCcLvFa8gEo5RybHMQRdP8QgzECk7v1pD=jC{UKS1-j&Zq4rB;8J9;!!Vm zPF&mlcZ}xTO2Upyyt*pRWvaE>26OeH7>EO`O!5@iW@!4Ig;d(6`#a zV=YqL&15fTpWf5RC~E!q=63l_GkW6~_dGSgV|nSw2%Xa3mJ+%+*}U6XC9lLewB$1p zvyIO+B`=r$DwO1S`_~?4i6isgTYC)iPC2osWCKGL%8}OpoXvv!0Y0}iLB3Hv$z8FP zHd7mpgwA_^w_@9uMr5sg&W{`oCA6n!hrvm~$nb3QLe-j`| zl!&^TM-M;sdruj53S<$sc}LLu&~HSHa7>teXN{#NLlA}Gl01L6id%x$1tMH78^hwM zp%;bd$U2g)2n~KH>~NghONc%q3@5s6K#@b+mOr%Mj{wTnuoeT;3e@#UP|ItJ{A&Sw zZ?JRRo>z_6ReIaaC3V5PoSGARzbk+;IM-|yT^e>FJxU^sL&?s?lGbD}WO^6P#FnZp z7v-Wh7cw7(-i{|L=8I4p5wo-#R#+Z0LfXrHeYjWmMK%$&7hYe}Ecs0v7rA4WH|A3> z!%gRJ!;hzH?oRUaXiG-Fsd2R(2pAc2zIklV3A0?*6an;b7hD)NUfd+d9bAB6B@;uI ziSgFtDRLctdFj61ELX-J^iRv8I%2!dnVrUdJcCki1C{oDX5GG*=gpndKcS2azP=f? z!!sNkK$sudi$)6TpbhPJ&I2ykPw3glQ7F@HPQ5u}(hke7&jj@sBR&=6Bo3R<*z~a@ zq4_`h4u}?Z;B+8A+`NAr?e77a9X?UG=GiereP4#76p0uj92xFSek0Y#UTvI*wAe6B zuRjUev*C&HCpKZb2LC`~#Ns-8*)5cI7L-vl=_VSH;k)_j3P}Ok`!3a*^pmAQQ|M5s6EO)3s9FLU?u0AOT|O zJanhD8O-&ojVL8C6)a(2txX;B90W|>;-WLK!^(m5OF)}|g4W^4OX z6=$xQ0#gceJ4Kb8^41P0;Zp&6{#YE1C{J_2Qpsj3MI%5$B~yn|opv!(NBK%ObW2K# zvQwwIK>z;Er`fhtU;c7a#3(%5RSxfwUZ+u&xRRdZ3sq>);cQTovUyN zj6*vlP!Wsv`Tcz^S{F=U$w7oBW$saGb)+8vduLf^`X|jg5ZT;r@WhZetP-Sk3*cAV!>$iS1%Kr@_(>9&AJ? z$`6+!S(7geQePyRBs>7f`*w^75%aINGC zwQ~_=5d7l|Oc$5XovLh#L3N5BJElX$%8m6dvwsTb(Hw8eD-+eL4Wo=~(QTJs%dt}f zsscc9;Qx!m_5Y1S)NW=OCcW^ah=(nF2^ErIxr&9uQfu+1u@+X@V8k<06$f*VUfGO| zV#`?@s7${>nmmO`MUY>d#BNls4=5*8y^j(r#h#zcChsGF-%J@Itx$??zIY{ zr(V)^@bw2)CGXwg;Q>h1!tQOP#FYoDDKJMP-n5Hn3Cm;XOIbkEq>8?j-P*%Ws&W=% zc`0I1nteNEfmb8fV}kHg$h+BIs#Ejm$-ea00wbk|Lv91BTwb}~CjOe}JfUR~GuP}w zC5CO4hDF7EW8Kus%GXlX0v{Rc1e#0wQYra^ZJy{C%?Af5Ab`Xz?A}4DQK=muQx-kM z)goV^QO4a;@Q1tcPnzdEb+k}6oUx$J39+Tj(Ng8fw|4L;dbQJ(MBe7q21A(Znd7sy zuLI+3ZYGXpB7Z>!IafVEk=6Tga_p(BGGGlp&s#u^a>{b;#bUYjix44!%?FX<-Ek?%xp=8y7Psq|B|7lsVpw* z?>>h)U#Wx$yHRh6Pj?3G&arPcVRM!AN4kJL+zrrXOaubi|TT9x#qt z$ZTR6o<$Iv3VDcd24B1d_4Eq1^dn})#vHv%JvrO_!ERkK{1>|-?29AQ3}^B-$5+0R zJK517dr>TV<+d^gx>d3h^rTV5IhAMA+n@zHN@;z`&0t8AnPuel1zELb`bj=>r>VtP zC{Dmnt=8$!AqGi55_GJBv66%qVl@dBG%RJHakfOUs*G-nccBSUnd`ZX5|6tA-pD|x z&OCiE3^{|Pa&so16L=v3V9qt)VI6k~c}-M}b8?=Xp}^3kk>G>#Si7I_LeU_%9!g-YVSWI5H+sQH;B#EvUd*u8!_PjhAdy9 zQU-FbaGtNQ{!a~U@50ze8z%dGv!Dw*hFAAZW3lHIOkx!Oy0HapwWNpKV%II|9EN_UOi z9k=1*QA3fX&<+>?erF^7GsJhT)Uv0C!GPsCevXII&nuxt^t(Z@7$E%gzePIREF*OAf+YlJR`$V zU4tq4qCB2k$R|coE#5leuw(@1lR^>Qxo}*9MUQdfhPeTCz%r_u9z!%aDofb zGVJM-*}e6VhAcx`w|rpe^<`$A^Z(doAu;GK3G>>L+!Ttr6KHd)iCE8hWlbf;edDNk z6}3GvL2?7kFGUG!A?$s76C7Zn(wZIr^J{DZ>}$yA1~TZE##IfZkj@)+9R67wR1Z{B zmh5KhkxA;;dtzXpu-PP`P)4?ljJw(Q-JcIs>|GwQGu@8E*}5@;Zfme-K8kLc`lDOG zK|U&~qMviyRN!W^!J5|rZ0H4P*CA;EF|Za8GVT+AP)LUpCYbb|KurD3h;Lcw;157} z=qe0vU0lRwOFCc5=+Jx|#C0X#e#;p)MZUJGV9YNFk~qg^$*b*+eI!HuF__Y1#x4pp z7VY-7hhB9-EehX}4V#5uw@G+C3Q%tvP)Y7-c38>2)agkJqWNU|+%ZX-XI9btJ>!D^ zag9L^fU|b$z+IrfJqDCSu>oOpzAHV)g)+`PP2rp@6pfQusm>X<5=0&V8#H5n|7?^q zvBjm6*j5c|GwQMODmfF?p&uVvD-1vx8((V^u&S~>msGPcbD_*Itdxu$zKcpWh56Q5 zqwHMX^$rs-|HG~_K`@XZxhbMSuE0(?!|t{d&>$yomSOu;i1aEdu5q4QlzytXXxs0* z#^%*TX?Av6U{u#dq!t+_^LRrS27kEaSRnzmw?iL|h7(q5$Osp)HnAqO2TsPUl_{}? z)NIZACbfq2WAZ>Z{1Vv*oGZ85rpSt>EEhYBg{~f}seD{xKU{WyZo)n*agn~r+JT22 zFcU4i0#0u(sp)NeRwH;Y1B-nk(B&!d?F5FqAh2!}Ol{rJ&Us3UF6q<5MPt3kl$DsJ z#0TdxerSw2I(d)_)jgvB=a-r5)zQg)WKn2U-2v6wm5vEMlFz5luWCvoM4xwn+#8l} zgoIYLY!k+R>p^)mujKGDAVOE6J+oljxqj zO?}`m(I%{K^{C^MNQVarT*PH)Y-+% zh1Se(;y{PCPCx>ea*m7J%`b{geoBWZxGAFSv9>wc7(1PR0ZBzILz&1WBV}~RUv+YM zW#fky5!oISNM{X6Lk&PgW*2t1ILzeJyYTvNag&?3i+a=H)FE=ohZ%!`RBcq`;Qp~R zENLRXG(7T>T3!_!|4oMM)I_t*%79sRvL-x?n@MdGS9|-xRg1`lwMbUVFKUjc;zn%o zepy*R0h$HNZ~%MmXFN2^oNuW*aX!tQKetL~R4=Wms*NRX5ulVtR86h1UVQ|+S(~K3 z4R(>%bm>W6eMFL17$j_VwP{TY8XX_eaDso2SO$Ytt2wbqZ609aT4_jp*2ixRf(#S?dYh4rB;zja9 zWfBF;Z-7J92pWzRQ1Wblfr4d@hubAOs@HsEJpieP8Sv9NXFs-auNU0ItU%&M0d>X^h$YvjjP zt&CsY;MF1{yMO|T%Fyykkm0^o8`Bx-$o!ga-m@3>BE%YgI;s9&mCxiz)1|)>OCyph z;!7hVxM^oq@$rw+WY(s#>SXy=DV6KNxp}EYRk+SE6LMNC2kb26d`9Y3@zi@1T4XNZ z>QNMx*#ENe`xiS)dV|4~Dw&B!ByHsMMGbntoYE$HE1H>lC3co(D{^>n2sGAFHz^v~DAKE|B~Y7J;A~P#86Khf z*Mz=sn)OBQ1WUaiSTF1~aE7P3S6Be04kX->mc99tb zSxtg`nY#noMP?55;eafq@DyVzo;L;dg?jbjm1Zr1`X*Sl@WYgIU`0&4dk{!0^)0Fo zgK)b6(zqJ7r!51$rLIJeVRMvm4RtMsotNtsgKR2%CM4{0<5AY1d~BEl>s^E zd^m?~R2A?`Ma`}R32BiH#@of(#C}Qp-;_HXdr==dO&#adFm;P+kOxh++f!^SU}iJV zkolg8t@twi%9!to{Yc<$_kZkvY~R%N z1+0Ft`^@5KIsQ^^@aEa|qELQTA5fI>)!=#ANd1FJcE+JExmAFKNb^>yXI@SaH9Xb~ z4&av-uj#j90QLDYN+i&77q0QnsnhA{b@TG@_YUC2>)=cH@up#_PV~n7_L?zCgl+Kv zxDojvwlQn)&nPw`Yzpevke9=#IX+ zy;N{>(s>d&{j!JYQ?DKSceuQmWVT!bYdS0?&FyFx?oLNMj8Um5-e_zvaA+1a^cNrV$azU2C*%6iO&MbM2b+UX*- zaZiU|TzK7%ep=z4Rd~Bbdyz2B72Q1~+M4Rg*{!viMqbR<@pAdX*75t=tlwtJ-u^0( zGykeYw{@EiTn9)#!#;arXrG&=Z=7!HmQ7iv@Y z>l@Nnx-Y%{+PL=B)Rob9!Y@Uy%ies z95l*bJwi;^J0OB4^EDz<4Hsm%74=&TZ~wlem#Aqarae5A{n5zM*OB#Jz}3*~w3uP0 z_9RpRI)$*%*Xtd4sw2{!-^7Mq5Fl&MzzZZVIqg@bx~)1*Q(9MEsirWYc2H8VKQ`11 zNt+<~vAlcs*PLI(TO8%H!x4`bIt2V44{rA9X4;9RraBc5oByJj7ZIW=5@86`LNmy4(g~>btBp=RSfh zu>YDk+a}5{f`9=LIvm@F6PpDQTBz6};6#I9c&#@qmCfnKP~gOC6;9Oe6ziS~bn zW&T&42Bhu*#D@NRJEBpU!c)#}d?ys#tP6idZu;@P!`Qb2yp1#lG6=AS;s(d!>(9`S z2j`Wo`{aFH|GW8U`L~5=4x`UIz;QbNBT!@+T--r0F=Kv@WRK8oEZg!%iGEla|CkL6 z`0?KT-Yq&=$13bVH0^kl^XuHZvcP>;(6*t0bP*9St>Hs8f9VtaL}B8!4G z9dk(+xgjx&thG;QDP^ij{Yc301M)GTGhrp;1`c#s6aR53Tv=Y#BKd9zfGtJNPhFR6 zc%}D4D+b%g8c&?ll*Fa=H(agqCI3b^%w?btbSvi(IEJGG_hn^lLFTbMPc;T@ZUGaW zF8zkjvTzF7Vb<0^W6!#m6e1OgKHYtR|FkGtIrnbR9UYi2?NACrs1qw|<#|Xj;ZxR@ zdpkV{O}Nc3?edSq0CchfPkCQy54QRzEp`Z7G!lGR2?fBPQuel9xTI5A2y{qyN4cq! zzn6vhT=5JzPi(A9%tmYNyD%c4j@1vMxcYmz|Bao$W_+1Ge>y6Dds#iX_*=fLv22vy ze=-RR++r*GR+tJ~e`vf!x$9E@vUliGzlVtqL~E}qTe!7-0wD4w$tE^~_pUta=J~xF zmyYCMK_;b#&P5%OgBRv0t_d(KW#nO4>mMFOuVjd92^~P6H$(z)rl}!##cS(t%-oc} z4a(o$(osJR$0S`_g7Z!qTTGb*^~fuoZ;+Lz$8yi83+q77fU!kpjYC>W2OSDaM_yq&iZ)H6U@{&QEJ&7jafe12M7|B<#< zUlxW10cf@m%+BHmxH)+_x%lwwS{vyabozTS+A91)o|!Xl$UchGxoRt^?azFwy*Bu?e2;c!Bl+Sg$3}udNulhLw~ia&e^qj z(3m;Bz--W_biV8T(U_T^By^}d;@k7UJWGur1CT&W>4V>`kYRTFbhuSK6xbPVeM&N) z-o365alm8PuFqH}JO4vCf5vfhkZm3jBzT*jX2`7ROH7w_Q$wk#OAy%{F5Cz|PWHVJ z=sZR3lS=2mygFxMal_GRA2n^3&o%WvGMxr7D0CB;<2jUAF687!wIx8g#+5G>@(^z# z4ETkDHgLs*?Q>4NWxgD}Yv$sK`w~wS0Uav5+n;lt8Ez}QdrH+I5JJmKucaf0@lD)F ze(8{y`K^kzV>rgBRjwLuZ$6D!e&2r|)>GIx?pa8ZZ!g|3u7v~5AqHuD%BwPzZAyDI zC7M?Zt@{q?b~e#7L!WbKc$GZHS@0Hp6hO=@=}SzWbrZ!ZsY?*CY>}fjY$T8(EosM1 z7k4gfF9%Y3s7=(gn?Du}deP52k-WMo;J0gK(0JjO3>NfxLG>MEqi= zlls4sye<2)Xoij&>QG-~8VTh3tBDS*X$r5&#A^eG4EA#;WIBL8)5O_&1{)t=1)f1& zA=(_YoIcy*|JBVZWGTn6jCH)l7J{5P1Ye7S4FzxbBcwwFg_4+$v4)7Z^c=QI)ebkk zeWQ>WRL&sn<ZF8#y|FO#>zDL_8c{)H^^dVm5wjtZh^*AJ=0>{VyPSq-J+t74mA27 zrDQ_pe@d0G0y?}H6zY_dgnLxfmN4m)qV+fm{z2x2R*#PWucxrr#K&LB9DO!YGBf8u zO4is^^L3}Ip}^@dEey>&aPqFKF5;7D88ci>=9!#Dn>}ntXxPK&tdU?AShr77<0$qa- zfByFXeW0kyWo``x;+~4H5YmobW>j*eO3C;r-pZNUyuHpxq2$`PSeN+ z&*3=$rJ2q?7@rmWJCE#|d`g_qC7b00&Px$w1NZCz-QmsY6YED_lL?&pQuM zT=G09aNaMhLNG_pvP@d%Yh#YnO&OFT+>DUOK>|?uP>q@MmBA&^|Ehba9YmjHMz^}? zW1|XzSL>`qFuwp=@U6nkMMvy!vC5cxxx(TGvhJJ*a%J2VkUBO|O*IvN>pw;)jZ#wP z<(N~2u7^=zwZA)F3N+njiA>;Yn~xI0QG8TPn)H-KrP4t&bBLwm878rq+pSh_*`LSx zCJNAs{DIT#}o?v+y;31DiZuWdLDOB2LRd#Va|WYYGfm6jOk@Cdn|Ov-z75>RoicM zc)NXc=)6<==C9vnJ3vnMozFqGbn6Om;eBwt9WSI!9{Vc7p8A*bD{^9;3cwRpED~`N zRe&`J>TPCq>RxeJ1D-S>Y}>2-sYbqE?Q-tr1=+>843O>`<^>=6Mk;GV7 zd%nJ=I*}4c-8`t^=c^eS4*FRScOK3bP6=xQXCUau;1lqil#jEo^idw~k5U>Xyx5*& zoMtc%je27b4!18S#>e18aMCAM$MAxGOFWBV1T3#>=x`iYdU0UH0xP z4a}pAeAl)&ZyYikf)mVMH@7~%%YwL2H5`qO9NV_ delta 28439 zcmV)ZK&!usodb%U1F#GWf3f&92@_l`%d!F^)Vafl1uQXZLdRa5X49Rr*;TF;vL$3I z44jqwAFL!&pT;Xj*z7r^Rg_jnocx8hZUp(`_y3K}g4`hAGjwWdYck#8Pk7ue zPV1Ey)cu5K5zNm61_SK<1gHHGeGEn2JeY+ol+pXD{$X(@n$QvQ9`HRFnIv9NAP1cO zPlLPV9i_J*e=n}oNmqKrv%SRK%Su}wA+{F1tRN=za}v30yp8=m<7>p$1A=kuG`lE# zO>NI0e8}2{lr+G{t|P4ReTB&|tVh52enefr$NG(fKHup2vkU;55RLIE}s49vx( zP5XC!@~0)bOKF#UjKD6&e77$7M;Lu#>wHfGhRvN*e^!EQ>(7sG>7;OfL^PG{`7f)@ z!VFW~f;*PM1}uUi5R2aYR^QI59awl3r~;>7ZN!uEF)(`#@LEz20&iXsS&KB1pxRJS zt=m-~Ayz`HgxF`eT@OkVK&qUn{1`#Ol`p(%v1{zbqrgHDhF2S2U=#65Vu%J>V{e_oV@%PI%nyOHm?tQZ$8SyQrfVy zfvCs{1xDy*86hGO3XInWEM%!tN0CF3_||cnJ`9h5aI)w7MIjukI@`!#`V~%4MCn_( zZMiKI_0l^xz)MyHup&8b2**y^g=Tn8+e{+YMIl$49d&Kn%1*kSG|zd3-><9P7QF8U ze_oB3b2bd^o|et?w5hOC(2U8_vi$?w11_-Rbxzl`JcoT8lXkKG{hfCQFh=ao zdOA=*S(6>nK&6Z9z>;E7E#1MFHkRpI)JJx@#MQ787ytc1`O{ zJzPWAtY&Kj^Z?p!Nl`bHkqWmMSfdf8=49kfr@BJjp*5Pi)`(7Zz}_jfZCf=occ5A0 zfvQq_>JA2Dd#q`b8Xy+r7#*br6kWjY)1^;hiafqDf@9T0dJb&z#}<)MJhsOlf7`>y z_QGio9t(d;AG-#HABAM_E#!Sr$SNko5Ql;P_`1-~e~V8drQ~ma z&Pjy%_%z-*Dn4z&vN09w-o>tkle`(%>Nu<{!{>D4um;BhY)u08F#y}t^2m!=;sbCP zWjX2iiSU337TnSxr2OJnTTg&p0Q2xA~M#VsE#HvLv(H=Yq*^pIWMIW*mu zvFajRu+T97PT_(hXsUEpQ2xD*l#L^#j7Vj5S;~s%c4HG16DOCXe_?v|7F>Q;*;s0; zV|K6EYhp#db5dX3+Q~*LE2*q*Pg$)Dk-z8Yrz`9&Qkj*?>SVnqw*@%q)kqvlWByr6!e^tzSoCd_yaSY;OiA~GB zgpVywpZxOsZ{KJAgVQ$=CM|ZuB*+(8Z%hJL_iF&b#Qp4S4z!a1ctHR-hvSx*+OWit zYp_{tx+8y)(IcC_Z5P`>rZWRsb}!reYis}sFMUw|oSaq{<^N5aEdL@hDaZ$U@6iXs zK_IIZD1c*Uf3LB{6Fj))L(V>nVw=sQ?m`sWRByGLb~eM4#%Vk4?o*wO1UNQC%G2|a zPS0Pz3^OjlED=7Je1-M6v=v5ErV^F%Bx@s`NmR+vd!NT#_5ictE@a(Vs69M6xS}%z zfhv2|+((frN)feUX)dTgqa|)P7i7?=D{-9FIvX(Re^d^nb+t&3i#^iQ157$7xV<_V zSRJOLgIn+bv!_|9#fVVfst~7Qo}vm6Sb$4csa*c`1-}{ zDuy!W0$k!0eoa2S<69mR-s+iQmYD6*H%0*|vIchbTqU>J75jav8Ou^%oGI1t0V{{x zn9P^se;NLl^N7(mG7epHv8I)qH<_w_YVBhf2iWz6&1Qah18e0CuZoR_Rh|6t`+pbO5C8G|zlt4*8_}s9?Qy2D z*)ZtI;&@iME>ZRvDEolbzN~L5Xm8ew%_#Qye}z>um)IjIh{hFmQ8y}gQ8%9GqJH*K zlc~Gb7QK7cf$X`k@r;R&q|M5M;SEqW*zmRwF4((1`9R^rxD~_KBj7>az-Paj@-0#K zMI3l1f+#~m)G;6`aA<3)^NoM04e>^@X$p*s#Y@8FDCW+6Qjh}2;w$a{JlnpWW zmFQ@@F^pnMel|7-W4%o8C}{^_&|7f#CyhE}XYTeTxA}e&@JtDK#{fKZAOQcPN0m(# zfVKNcgqb#$q7(wi075Y>e~BZ*;>lB`;l=k01j83D9thVGYMuv#&z*JEB=BtXe`qX; z%U2t>75!7wPUGu+bQve6a}-2sWLGg2(#{WzoohsA;7W5(d}f5XC4|0Y0N#79L-PSz z;;zKq%fsEWIS~3jz^h-0n_wr`Zp}u2HU?)M)Hi^)!N%(w5+OHJOx$5k+j*fQYAPzJ z45di6XAa`&jouH-WJcaDBOFpre;8DBNj||x=F}W1(;YVXS3WYHagLI6&W23HS8D1i z`!x8_u4#-FMIViZ(}}Cl(P*Sw#$Y@$HAB%S_Q)QLP52ufPR*%le+?)0P&)vF z3VG(Q$eY0RALDe8(Hq7YJcmgP7i$;j0k3mMJ9~pqeuMaB%pTsuhp(Z@VhOJ>Vav$O zIou_Z?h>Ea!K+-bWm;Zl$fH0h-(E*hdZV%2VFiZ9&H}^ZVa{sno|WXd`7|{ks$y5PbytgY2xFOegcDe}PzmI0h;Ir(07P zvsbXQenC&u9q^|xt=nI}XnNlo1)UnLuIgq}i`NfMy=gLwTMT}bzIABX))vDjrAVIf zx4^Rca2ZSU=X{{U%b|<=kHF$3>9K&-&0ItMRmmg^t0JOo(&+O0gRiv{uyNd`&MM*Y zvaes1zAn(HYM+>Gf9+scZ-7EIGxiIpuRTFY81y*p1~~chmIFGDOx)gbpnsOO*%AQv z2EaFr@pJc;U}r?ZfIj4bo3zk>OM(BkLVi)6Ps{50FS{0Bw-41u@;F2Oiron|ouHjH@7zzy&e)h%NB*ZF#N>uNcqWf=h3E7Xl_f2NOQ z>M=LRQ}YcZwx!rM1BM|Yr*OH*`HR(1=RdE{J>Dsw?X zJ{M%ce;Nkd`cRNFP2(toNa^2t%D?s0B#kul#*10h9>t6eH=q+aRO1GQY8<7hvSX(DI{8j_3Ju3gChGSzic=E_ z^-=;#01d+_O~86b!n`JP7k-BQE*fI@aL7iB?OvRhlrR+du(a$ti1SJ%--^R z^3naw)+LlydfhVum1%-@P@X6mfiEuk;+$o7KkKGuyl1B45zxv>Q ze>5fW@OzME^u`xJ!PP0 z&CCd1$g~tS1gfrNe=RV`>-+!%=P=*TjyCsdi`~>507v1?4e~If0T(zrIZsNxRFDTM z$Ra<8o;fAp!bB6%XO#J zftOQR%QRae44IO>lsfLny#tIrItCa`OR|mJIVNDC1OV9kA&wJ(ePx4q7E$Pj!ClJb zfBE4Jd`{_aXkqa*TR!y~{`g29!X@ww15R(su6=+7l4St5`i%xpk8dn^DxRnXe-0On zKOBGNI(UCv626h`@Y|N)G$06IEzUU)FeXE$@Y+6TQPKwD(kZAu{1T{;3S|uVjx^fg zduzF=H-T56vG#wgQ{vV-wC__&FbQGWoNYl7w}t*1lC~gKH{F8HYl01)M`8Ghyc?3A zG%X!Nu69jZ>#C;K97H!HqCXMjf5lp*jD^Tn)=h=A1mNGVNp6TKOF!^jc=E;vF2v44 zWzu|UXq-}G1(Kg|1eR5atrGoQxhPxr`e7EXfipY%@@)pYImhM$(o&v+<1>5lo}&#a zz_|;1gzap1R1?>nbtCchtoS-y&CCPQnzOnGkEc9v?t99ibkm?)l8G~{f12_)RBv-m z9V(dQ(ZQGEo6PG;97}QXQnB4$iO7pHE*1A95kf?{%{!iX{sib!1q9RO;q7*L7dGn8 zHfzkzURhfPilzj`-Jr;rCF3eD7BQ2xIxwwo5s~nV?ouYUXLYr;54=H z*{ltqGicB$qg%{dp7P)_ivx@Q=nBut1rWnV%>nFspZtOnJ0UU8S)FkpvLvJ|?xRy1X>cfeyOiWO5*3cePux1Qvf2fTnLu+VG-HAJ$ z3{1F=(t$D@E1El1#zVy%nwFv;U+s;+q?qCb@4tQ574#NV?vla=_f)pU`zl3uv3R*$ z_@O(AU*dCLs>;iayKad!!%EoM;EqdkR#nJj%h}m{&A(o()Ro@&n~ONFjz&mXhO^Lg zBd=xd1xvK2+-ykGe_A}}qILO+)XoiSlFfpy6+P?Rx%Cy3N{w?YjZ>KVtr~@y5n^bY zVW2PI4h%d#fszKlhL5}_9bS=y3mN`BMfi7*=&wBM$;2gj(31x}sdlB>m1_5fv#vy> zNmqNiR;@a_Se(r(8$i#dy=TObSAa5L$d=$l3Y#uYk7)~#f7U*O7NV)+ju=~{v=QjN z+3OIe6^KWrUmczyuS+wDK7Qn;K!ueJ;Ds`=FVnmpQ3n9hez`$20(vhdJRge2E48@UsehhLE?Q z?kCPue!`;pakJ)_?gxYCm`1t<1WgHoHGv?UuX%Pif9tR40(mBB6QZwQwBkf&usI`l zjW2m7%=I-_JfOGYCebc!Fb#9>@P4moKanapBNeyc`s7YXA-PWP7T7T!T#ia z&c}|Ve>1?=_rQyZ(YMSlX)*SqlNSEf+^ZR-Rv#D}-KX1eOAy{0gkhaCIJk^pkMCv( zuWKD``Q)OnU-HSVoa_BZvMQq`cIaQZcLTXC20poL`_nuCm)B|6RDVI*W4i>W_d#6* z*)_P4X{dD%xS;!wq26HE<1WL zizR>ER)@)&)*lRc&sIiwcBS684DlIvAjBbQF=YPwB^z4aBE0_&6FhO{p8fFv zd4!E&aY5_5p93#u%;rM72D&a`)ksXu)A1}|QyDU450W%y1wTlSKYssj#O6D9@yGA~ zf4cB-j=RwLiSn!hjn9<0Q*8E~9qY+GHir=Xpg+)ik3Qf9CIJ==;yPP&u5{>Zvv|TX z!sTx~pF>G!!xF|>EIl4p{P68}e_^E<7+?#7XZKCtuSw>vSuqi8g##Bmd}a0f3ycb( zGl__MBVu;E#UeT3qhg^m7c7Rcx4^Lhf4tQazS75tG20_(Q;;`?mlKZ0Z4qpiY>9=4q&A11!I57wi?~#-k5fY&+6vx5b=X- zwDGUu2)(5djXOsJs;vz$OJE8S>L{}r*5dlA1gDz7=>f}`CJ!KMKIMkF**Dfle>9Lh zCkYHNqi^T6idTR^C8HY5e$_a7ky;lRPA`UKkZ0e{@bKsRVT6@#Ja|WN^AUHj+ATBQ zE*rtN1*Ro}U0y7D6_HZN%AFBcic#>=mB-20J29#$v#WLjRlSoF#vtKQ?6_(B0SD-g znT|2++Jc8l_I1UrwiYjO#bpz;e}4EWv+}#CTrnB@h#W(#zKZ={DX~er_w~y_nHQM3 zbFTQs3VO**n!3>IBPv%uWR&IOx-55^rN|8O=H-mPx!bH9(XjVDYna15V`=xUKYsu3 zu@6#8D7^k3zyFJ6TZ6WC;Y7(r09gxDF1!Up&bRm>To~t_IqG$+&E|=de@=iwpZl0b zi8u8`lE~2yn1lUua0YcZL6@Tz&M|7CT4uOJu!C{|@6q2ocYh5d%*Sr$a|77WEo4I= z$XfNqhToq%$mU@{AA6YaM`BRqbC!n+UqyrjtSXkAF!m&GQ1O5J;#P(>l*7~+CN&(X zN^1RVS)BB>k_gh-1#4LWe~x1=obfVoSla&t&UzTejTWARU4xtng_n%dQ{;NTN-S95 zttyqi`K+d9rIn}rdwG8~33H8hl+CbjYjM&bt4SB-FY9x)vRQTMJer-@_ejwh5+)cK zz^}996iR<%Vbp0}qw?b7Cli|cx%XMg(1inZQfg1Gd0b^au# zJ4U>;v(@A}OsR$rvbLYh!K4P08thpbOs^i?gln5H)CEC5 z_w%zrIPV3(sl_~~e;f(Of5Io^9czg61K`vZ%yr4M+75VsgtvBB8z=`uVGjN0(!;R{ zAs>)X``DKMOqG2`B<@nEJZ$S{O(9C(wzP9;088ep976;AgAy78I`K>s#`XxS|6ICJ zg6Sk}+sbmyI+;^po2Zv)f{!a%s4NxbkKg}$P3kd~WkQ)XeEawkpDQfyAdx z=gVFWIHXr}CGIE(yN}6FZZw|p->~vDEfA~AgS6VlfY%)&*Rj&>DO%gO2{A(|b^zi{ z0@C`!-DWy~b;O#8fnhTWqbPj9UNVPvQaGrDH_ICo9TB?ZJ6JBL(lJA5(M=~@CO%+q zmlUbLSF)g5f9_~Xx68?1zy8XzflQ4)23SX~7xT8C$TumL7KImqG~3s_baTWkd^IoK z<){B$6XI*nsmAO;c!E=eixU{n>LGo)a#A$}kZT(4_1ub#XFyk%jF7*wzK3yt@2Odv zShk3R$+5~iB!(RY!}#uy7&an|K&L0XD+TCq)*=opf6(EU;*X50uLGz`Ns<(XVmH=5_rnonrk;RgY##k;^0dloF;ny!(-@JOj z)XW3$vPhUyMX2_&R#)5kin6)5?zFf|^Ht*Pf(d2xiWz0IqzLU|49B57$sgw=UkMS_ zymj`7CJ&SbWHobXOXtALZ+@j6WtlWDN8 z38kC~rBp>4P{)9&^1(JZ=pYCm^588T8Q>;kKHJe=gSAvNt`pXbgQhGMsB~IbJ9660 zf0KCFsEJCaa=4vm7L&xpqnU1bhxt5I2w=y==X#+|KM3jc8?oPNPe(a|;dPue6bUp( zY}Dsd_nyXE^daV=7=3k1GRk!sUZwp=)yn-y)ywWjdPW?2O+LKCS%WcFWm0aSVcAqV zcAbD!jVV!wrqwg;R=N#2!dcB{tWAa&X%BT)h zjqmPWUe1=r6VsyJDal!*vlfT2oLS&+8g9~O$*l91Ip_cgG0 z9zjXZ#NFWP;)O3Rd=ihb$FUN3=c0OH;6}rhN?bY$EU!?+VqX9i)|XwB|F%E;@a>PC zN+ZJ726*L#zaq=7A`t%d=U>nme@vE9AS&6@$++c%*V2p&^MG(cO|Ym-SUid~7lps_ zF)Qfg$L#pgBO9#32?@QUPRIa*Vz?~IkjX^&v-d!?NjdQ22`od zoz}AlUad8tNwZ^k`NUBH?FfT-lpPZfH(Jq1cxp5}wXrP)2pnP%vyu!&e-ZAl0*7`k z9ER2`C8K0!;SLC5L^)~0$1HKx4<9Hmp__YmQAW}e2NToku6nAVVgSix>iYhUI;eL9 zINA4de_N{%Ts6oy;b&s>O-_^`N!@So9@+k|PG}}!O{rtzG45`C{ZbZOZ`Bbg$g*5F zM-SBnfmWBjxJ>)6uQyg!f47)1C58a9Nu^f*OvxChK0Ed-`+(SFfjrj;kJZtxN`Iyr zFeY>4&w!_+Fq}(}X+Fq=QHm#)vJzLU?$z&YrvV3rb$(MCKbqa`Au6v zvg}u&QS`ns;x3`DUzEN+3OZH1_FG$prlY^GU1*`TBDN5%Zv`F>Jprfc$qsy z;^kri<{Dukhs6EFCn(~;M7VCcV^Z0+zNCz+-12^-DO8JYrI#He(07@6JCDHnnQXm0 z+4etFCivObp=W};e{~m_c>5m)$;`U_A^iJq%0sX4Uk}fflcnqTup4vEgO$(DNVE9) z=_ha?=E5Aa@qFLDF>% zejNREIbUF;A)eGIwv2$ef#%UWY{%9mwx)kjNG}{qci)pV1ag? z4S9>cNhUNN1JQ*WFm~tIvs~+{LyfrnGejlQVtqqV1-4$ZPr>3&_@ZWD~6wlX!tcHYd#krgD$y&yCDu*&9XiG|vaX{vGfz0O` ze@2|E!*mL@VA=^CdofFfDIiPx>leFkiFC_`Bl+3#&bN7Sv%GUgZ?T&WlKq85%%&q| zVdVQm*qeG&oUYzB%*~mGUi3}zS@OT^;m`*u(pB+8ScI~ zhV_^=;QHYw@-bq|dhDiB$0RVfFF7Oge`OWaLAwXM-8F~iyr3C6(n7ql^Y#uTE4305 zY!jH`#*uMuHcyz|#33lyPisu0#G85?JdSVc+13HJUV(@g*icKJwKvb&NG>c!Qs$x( zNmtByGyIMc2kR?to-QLiD)UOSwNr5_F;`Z{tRH}H1>#(kFY&1aqLs9voegMsf7OW2 zcrpev7j`YgtHd~OH+W1-u4D$%Rty`#n*k8Wz)5{mYhAyW&}bg#{hBe(%-wWgv#pHJ z5S_I^ms$ErO}djqoTXQH(^0IMTS9se!x+ufdO$8>cr|D+^CNTt2EFeXwZVpVC&Y_wLd5ehU z`Q)Sf8CZRU5`IdM(j`bqkh)lqvi5294p*P4S}ui*`#1%JE@7qC+I9k;!H!SUZxX3EAI4-8>RY%nGY@ioY zF~@jJ@q5C~Qy{3T-%+PcIUPmlGeHpfJm}KWqR&NK4#r5U}n_~vH|=TN+zEJtA#0$dCmWxI0m_ z^we%Mv;deXrbyZ>9-OdSvgR+iPaTPWb@ixc%*OfZ6$>+11FKG(?ZwaV-mJ4N=j%;& z?iT%bdjZ*tlt*5Xe~+b@{g65mU4M>Q>>v*F<#<-^O?2ZmUt+CM4j}6uQ4ZXMYswKh zE>3Q%d%l50ksUzED#9dC9t|jMV{itbOvUeNfzsY9XT?5Y&Pw)1>2!f?Y{j$MrdB#U zCoW4G$CGMwhueM+Ur0iSQIidT-vjJgVL*EctaHZ4R;tR%f7tj#(9bGIzifF&aPkdg zkEf%QD!Az31EwRmd4%y6bNKPWZyD>9oPgso>;-#M{Cdi>g!62d0Mn+FK)3WDZMwQ} zy$L>{{M;P|NmyJ;dqes<^`4zUfbQv|Gxx?(2w!$j*`HOd>Q>y?jWcyFfsO9L6#*X$ zpT{B^frxxde{iH3=B0~nN=UUS?+*ydEe=7FnZm8&Gvv)f?p!YrbJC99#bh7>v+00o z+AE_lw@CzPt9oB2b6%a<$sL0unxMT`KJ86H>hl_-m{mQK<{3sL-87aWiqZ(SZ<$)1 zYfw5bHU_XHfJy+B0D5EqUB#TtDAo1-`VbWNNrGuPf0%Ab;aTP7dJ#9xz2W2cm>GKk zA#Vw~N9wb`fu3F7bilAu5JcwUJR^a%-p1&nJb%jz#^HQt=T_k%z+@1YOkLk6 z?Uo-t(5M9latl9TKTLp@tG%b*d%wUuC8EWG6RMz;oB~+=9G!CN5;siF4MxcT>(ah%R~aB z1jdFQnhTz>60>ogcO@b=(S3RD8Pp|%e^+N=Jd_$+r8tvi^5~cg zXS_@t`Z=MjV;+uhjTWAReHfhyg*T2ksm?m38uy6kY$~;Bn|u^2PN*Iz^Db+ap_}k& z62eyjnL7POfUxvj*49N&Y%CG;pb0y}^Ar!*{l1Y6CEY|_5LNe)HO8X^IX8*{e?OUd zONqH4!2BpV@342!w;&LVwcfz4`G5TWub6yz|07v009x^>OOQRE{AuYwGSPf-P2;Xr zW5+ny2nF3qg=f7!8dDmA!1fe=PYE6Nh-zhUpOk@*FL33%NMM2CgIm@doe1 z&Viw&G7vX=AW#Ct8KAfGvZu}we-M4^Z0>uP&*jb_|0rONZGVu{V&aG4Co=b9@vg5# z?2Iis=9yf)yU)6sBkhez+&#V1C+`_nlTX06B&Xq4h{rJ9a$ed!>lh#}r*zcUDwUG& z;p7$t*jtUE@7FIz4lZ0K0J(jS8Onm=Sr4faf!oqSwfrWpw)S}IlEOfHfAaS?ADPBz zgRI|qKw}bnpTTkul3C0i&O9YO_5oowRm}3u@J~5JF_m=L4OtE2fa&4J88k)B!;#LM z%T>(X^-)ulk5B4-yGvDUsE7q;bw^bchuy(((R;FH_QR&!d9ZyT2h68(5)X;Q?Z)Hw zz+Q73Q^L8HadrApiF}kGfAq&)-8QepjWMX&GG<^mzO{eT;If!*C9zUsWfQ|}w!VRv zz_w&Ji{2(w+EwduZT%u*U{$=mq%E514&dla=Zi1`sUvK~9*tiU6wX(qRcMzRVA<2> zq}6GFfy#*Gqibg6#fwah>lNpynDFaCCh-*aJxESMzJa3*mTKX^e~j0KQxjf+%lzA) za}p86$+&Y2$4I=OJg8W}7_4wI2 zmy4tYOcnLKpf?3-@nI55f6xF23C`z~>2^m#OEG+)xc8*Mu)w8)xd~??p9cZOb;;jlo#w)V@Ct`lHY39!dXpu_w* z0wn4x0EHEp6xl+!&*~gDdA8GjCVa)dlPphdOHwn?e`o_Bi$rXE1xdk5E%=CIurKH6 zR7za`HAh0YB?`CzJTDPxxWk^V?6LtLVUwJx7sUyfgRm3ypG!Ql178NKRv(CT;*Gt9 z8zg(f*?Ogi-HHj5OA@<}F=+{vWpln}tLY(MkkRrH77zFgpMDN7XOzQlX_$M66BpTNiD9dZt6`8oaL_kW4|*ImZp++|73 zPLat@2YB`lyelo@D3Y>?fS1{W46Jq_-~mMycrquj20vz>PkEfkWfym90mf=27OSnv z%StTPF_HsmO#W?X7O=dFb)z1IbAG#pBh7^Ye=v>UR5gK>_~@EMAyMa|n#{E40q;ezpwv;|cz9e#&C1f*1D>xovY?~e-(;l-auVY{JxU(VP$0y^J9O;b~$^e0$qaH z)$ae|3eNxP6Lrm7T>$R6$4_j9t%q5-2G*>5GT%~Q+k7_<2!=fc&&}>atwNChORTBlS-yhevhq4oy z6HXn(g|wB_)8U@T1ul$<2)~eaL?=rC3oy;2D0~2^>2wkPKD;Bq;lPfXI( z*Ka)-Uk)IK6=LGUusQCAZ=;8gaemYb%H2tj(f$ zr!M&pC>GAWG0PR@JBxkM12y|f9RfVLoySMzS`{Pp4zurB&$f<^JOXN-2z8oE+1chm$ZHc zkXs>0-0BwFPDxEtgC^-=XF+>Zc0}H!e^Fw0vg)DDa7&)KvgPZ#1>p-84R?jFxN%aV zVP0h1^@{wI_Q!LB0FzXK@(jIF@Pdbpmdu3~(X9QtOJsQWPLLbNFeo%)HhZ1&G4-HCgoPCZ|c!Ez%NbLH|dt&oW4;HQ8;`< zFD`{zDjcbBwuESuf3qr_lv}03sYl_M{Xx?ePP*l{rf|%`Aqt0YXoD4wuJm+ODjcbB zswy1yoC+u7R;h66Q8-rLYPiD5w*1xf!+nUV@eDxC5pw6iIklv}03 zsYl`1eY5EbCri-0F@nmGz&i1`{j#k-z*PTfMJ5!!7J0VR|4oOJkJJ(Ry}}obj#K_{}t<&;bFRE z*j(LmcyM3pmRr*;T4QvJ4D#2jTeODx#I}a?uUEHdfA%4|MQdsUfZYwY_$a9eATzq7 zfN%iAVt8&lYs>#CYjeWeG@_fd^R)8dz53Uu66_j0Fmi)WPPQ7yT8y1+=J$)bpuRMT z>xZAn$A~TJv76#UAyq4zO<8km~--Ptc|?Ff6z4e;ooskTi3A7`Ri5X_9tUV z)IR*-f2@%YIac5ZV+DdqY9xtCKFouWa$X6LqcQOf@HzP<<#QYZ15z?#eN>SNQ`<++ z+Z)2)3oZSwY3RUCFVT28XT9$4>9$#atK>eFY0nxB(g=rKFI`5(f}Uc4f_>uHYzYP% zY%br2_djCWs4*pfa)Tv^9(IIN>ze7h*GVBie+fhj&si?87a?pvTb+}xswQ-=(@bx| z5LH7W=}{WH;t0zH9G>{R4YUDM3D=a28~AKse9M@m0J8?LKTgRcm%|y*q+AwXTvp4) zbt6^TKP#w&pl3u(u|J3+VA7jfe`qei?)SZfjXRAKWM|P}9Ea^pmOe1(pO*e3lW5+F ze-rhh(l>_-(q@AVR3p=podb~d`LW|1Sqpxbir$veDPAW;)ZEYUgag9|;N*@Dc#p{S zizi2s;mzB8+iQj}`Z|n3m-LLjHY`kDI$zY*kn;nMf#yFJPwuHO^AkL`Zkc-lUxzV$ z&IvjEfBjJoJT2MGKT z`-MMH`sp=6=;9uZ#Qn^8jSsl^2)|W;WAwrMo(sF+C69Z|ip9w)z`FFaJ74(pe-(Q@ z{<1CB)ciFpa6tnU>HPg;cb*O4Sl~!mY)p-B4-8KE2<`X$J8U|9Hvju<3I3++Pks!< zWu8;y{G9F4Ie`P@zkbmTl!2U!*`MG#1wW?1ZQ&!ONSE-*tH5?w;pwZ)VQMn3b3 z-lxki+WHLG5XJ<*Y#Cl1(1&vS|FVz9Rk152&V}}^tjAxt4?tCOCC-bke?+U6z`91- zI7R^j>P_$qnlj3B3~UPnA2WH}@k|^xOeJHceA~~9Wlaf)@8R?-kn!?TkT76_v?*j9 z2~4nFA&VI198Q_+3Nc5GW5Y2rkmigtsdERKHs*009&&Vmf}g0PnTv5l0MjCERR@@Y zy@!7pX8x~?OlZY+`Yb&+e`0g#f{@X1E2`oY>*f5IxN1q8@ff)JZW*z0gjuF`o*o1I zkaAxtX&o|iV=|&N;K@t);&A=ljC}9Tfo||(-@>wJ$+7pv+7+bV?~@1W`@IBnnE+%k znk&SJfEWqW^m^dgF!zV&d=D;&CEz#KqJ#4N_Z%nbxPpRtv&I(E&Le(_Fl@*ltdi+o_rd{uCFNC>(> z!b=_IS~J|`zDOF-=wOYCF3yFXG$YGhWf-V+EzO*FSZGJLUlp@8 z#{O~w_nQJg$C&cNfBrV1_o06e!gPnJSOLCYe6|^ABuYB=kjrC-enz~(ecgtmPeZmy9G*~&iQ`vXf1ku+^=Sc_7bhHd-5^-P z-(8S1SkCtN^g(DW(3D#ZxPoq;G>P#?ww1uR90Uo+{+RsbYA6KuiZPLQ^X#}gB!*Sb{&uv;0W%na$*-_ zn{hL&UIX?WPJ>u#fD(lH?eTKXOkt~Q*~*--5`#^`frf049)FQDi3aHre-alaUeE;qHLN>w)u%*Jlwj&`7yn021 zDH7?@PGNZTT6et_?Hi_Mm@jw735z?(NIPw|XLOzoTG_2cL(kGD#)tTrPHqQb+M+y` zf7K6C^z#BXfwH{X^oaKBGERunH*?S`xT6@a?{r69g=qw@UlF*9gu2Melv+)r@3U-? zR=m~u2=MVMa2<+OyC6UW#r%LGkxej4BNZ7Q9m#ZBMI~uE)rCr!olnFj=8gb>C{}Tz z6`e-Ho?=;jyUIBhiN|cf))yVSPDS4QAW6bc)^Lx{t{3QZ#e}&*= z9fD0(!&~lE#hYL7uyzOBoi_3O$&G*;@OB-7nOpJGNn)}08%_JRqKl`;La}7LOn6-? z>qwVH_TbF7pf20BcsaU@!sP<*_s8%5y$k?U66R{a+L}Dl7mL6iRv+MMa}U`mqkG$6 zZ`G?+dj>5&R(iFLf_z%8Q@O7le}yLFSubLYn&rl>2B)vkpQTkj6kf}buo!{kU`N`b zY`_bqKHD=8K4HyZ-IG57o&}nOgnsr|8XMtjh&vJ|B~D&3fQ_wMz5y>?fp za?e9&cqNO}dt_fc$K-Bku1(n-pNB}PAsD2_wpc#-JOoQ>HA9Rze3*tda#YvJGdrIq z;Lg`Yb{^|{ff^bDaQvrJs{5CB#x}^&EW!eE>n12{SUmTYCk&cG_?}ue50U}6CFJoT z)r6*3c}U*ED|;~OrSv$D6ke5iCEJ5{xhJjIlXB$2hQB6;CR0;&H_2?CA;!(3247ih z$W-#E&!AKd4R0dmpd-Yh^g3%LGeQac02EMxzT3BRgUmChCs=NZ$-BmUVp0q1Kjez= z{E8`Umjr94qQt`B*Y?T9|MK-}?=kLc;D`?#f8U3w0WzGHX{pIfFn45|iGq_>SD-ay+=%!mtsy{UmbW7e zZ{oG-jZa~`{Cw#+_d~f+Blb6rGGi3-QET#eY^?e%S>G9VuFSGU&XHEg4DCWzZSsUU zp-(eaqxgQI@2;)bbMwx0OX*}-%hDdRi))O4Kh}xX0j@Uj$vH zyePJG9&DO};?YoJVF$AYvZ$#BT=N1ui#N~>ep8PQ0&hdz!7a+bS!^TfILH`+A>!`W z4`WhktW-4^ykO2`IMb1e0Lb*wUPOw%S9=>AH|(r=R56bm1N5ZCeJP=EXWBfvixpk7 ziQ^)fPy)_=oFz>VD-mYX0X!fnicrsUJ6NT(vGC6vfGVqOd;CYEN%LIt zm*ux6KEu#>D>Ejkq{|m`>eS^(%|8o&1m?F9ih0zOVI^@T&Eq4#*VvUta8KK})=6d! zoY8XUS4bnDpRnfrmYLWvmmKUg2{->?YsHy1N9oZ8KECC%!|G3lR(}1_s`OK=w(;lO z=0X;AV|o*Cy>CMV1AzGL>(ss=b1eXw$2@~@@0}4?2@ZCOBmkkFWPc6xww;D^{}Z$r z5rC}VHyb?xcYU1PuBvsyX$zgj*!R#-!>{XZ(VZHz;F!xBJAG8JF;v!^H@C1_+vt!3 zlk{!pgce4ntuyO6PORn7d_Mn9Hr-ipciLtmhKMkkfX~u17*M7M+h(lw`NEPXM+O}+ zED)<+aV|e}Zn|Ow%h4ju{x(-D@GG#we z85Mk16b){a!ti`G1~I%su7|f`953)#?xexX)eN`gm(M7x_LJ~NqtWjcC4~xK5^&i) zxYQzzfjCX|PJlY0D-$Z~LASG&!swoT;oGkgl zi)_u8cJEXtMNxy@pn}umz@A?i?FXcvkDaJ~eDEaemH_m!fJLj$7`KohJ9tG2@bb77 z;hR$&J_UJN=Uf~d1WXpln}ptv9ZFYCcHspq;}pM{_NkT{tV&5A{4XjrN52h>pYptJk_QxvGD~Epy#) z@ZI8Ems_8nGWVd0Xpb3QN4q>SuYLMxPc}?&uMeDAM9!&4&Ehp7fvROHtbiPk)BOHU zev4d;lP&a>8L{4Df#~#XkoaP1Mcxy<_nX`LbT(iVysjRO$ZO>uuSeXXAdYUgeIohCMpBo5cKtfJF2lR56DDK%mu4)pFY*uHBou8Vesvsjk=3wR z+l&H4_$9vH7@M(X6Bu&_mFXOrFYI?q>8U8+!BjIcTWHrZBP3B~Td-=+RY2a}vVo%! z)$jCKji8Xx9dB=a7yB{xCHf^9vLzveibsl7D8?O3T3sF95vU4%UkRa786>J2I0N}c z8MwhL`D;HMmEAfatvHmVj)YpcO8t-n@0SuFgeH7JY@+D1xcPQpTX|d;n7YH`?3?@u z(wpMQn)E+~S0tHY-&o7L|3t!3j^RJI8|K!!|H+U3#;Gbbzum#jX<9`b6%n?@ZqxTT z#EqtS!F)cdA0c$rQm;2f%@bQD(T0}t!94xfq$lG|3FR*wQFWfi(_&|5Tf0w4n-mY}iNWd6X?OVS; zj4Z^(OY(QYMn5lVx+#%B5hJHyrKf&H=8t0^|<_Q1m7m{sQ1wZAnC}Sft`k-GEhN1a0i4AqI_Dr)xHSzA7Nav zzMoCu2vXTm^3C$quJ8^E+QJqH|MArVgS@@U9d>I9K{6bP|NCJgVzmI>ho1*)B6$TI zH<@M?#(`>Xbjq8y1&mCB1o=+>oxdFK*SmU~7J*4F#Irp(s)^7k3T|Zqr{S{2Cc*Lu3Td6zZ z;B;Z1l+W{>V2*zV7_&%xiI^1v%xhvtIN#HV46xvRhv*cU|s75ccflX&w)h+gMK#VNeYPTL3PhDAfn`+O9Ns*6-rDc7U zCdS`X_Vz4PQnz4fp7_u|Ql(rT%F=#bNiS0f52swuCRxeMR-!$(z*~pWQC}+P70@`1%@V&g4z@3E9$-X2Hta~0jW8}+}u=3UOcW&r!uyV##f;uaC zwNwL_p`BHSz4c{{Gfn0BJqSacI`tdFx&?p^CBi?Oof&i(x~cgG(bwOA2+j6O{@A*B z87F;`9-z}R<|J|yUX`@*xm!o{9fz%;18HVWiu5jS*B?d*fXm)VA>Z2riR|c1hN6Ht zobl&hSQzWDf~K~$5Qw6G_2#c@CZ#FZe>ohw6Ic2)8V+WYZ$YKM=L+(=pps&DpXJ)~ z7nrYMHTr3kyC$S=lu)Fyw@^Om7GtS}@s(<0c}!ICC`HyB_LiDRBRr@DPbH%oL9ID0 zi$X5uFuYUN0PbDsv^&zcXFvYNmMa{Pit#&ujlc|nS=#0cWKBys+7&Us{n)hPwd(P5 zsOpu34F1W1p?Pa_b0D<6n>)u}X>A9S=2W8U| z6z~(qH?UKk#yj<=xQChkgnr!%xk;qP-g%9RhgMX25g^Ah<=y%#E31sEAmdj=%A|5% zSJlnprXTwqig=Gb?y8&XbycEf#cP6f*|Y6hb)5vES`HYqe)~8p3-Hbt|7~WX6wt)Q z2LEW|f>Px{?RrKwFo^Qb|B&{S{+-Y1-?0!{_jS6RP0EfG1_Ej7wubJ%GndIMh!HqN z0H84M+wP3S?<#p5*i!hquVMQEbYXb4hCSE&5JN_x;O*-v_W~=Ar)hj+mTfk%vs2l< zM)Nd7e+2C621eRf$)hTM1l1QjJIfO*$`l%j7csIV$6)IykX5wB-S%!{=99VW#=HHK z!Z(q6D#kkU!*e?5s&K6cK)ZBQe62B02V9skH!=8pKMv4NnjM{?uCzn3GpkO&;?XsZ zMi%2*C=lRu-JUBbksF!624{Yx2rIwnS9BEvy-xW*Fdw(R-%?OHh_0|@QtIQ-KuxKn z{T1w`>V8DG{UjDFOAF~jJ1?2`YLC?J)f7>Fxbm|1#nNY8AdbrNXB0}|$ z+xL3?7>SV&E;pU~2(rJD{JRffp({x60{~wBy%GA6z-2|(&~8h?4>=%QY0d<8^KAVC zbOr_dpbSB*{i-Q1O!K9ib1F5o41VpI;}^mirN2;jbUKvLHEj&*CDkj@H6Z1i3j6a% z=@L(34W}e*J~ZSh(MnykyhE$&xifNQES3&|9PE4n&*HZM4qLMy$m2QX7;Y!>P4z59 znkpXo3a&atcn70R=VPilnyMM))6E7v1sYs6?^BRfEnE*31a~XYPN*nl0@)>z&db5y zj>x=6-lzTF+OebDdL(Kp!vGEl3`{FIZ-5_gd~e10r5mufTP?lbjc%gL*7umn0v}X# zhLiemxg7|?u=T{Q5JAq{eNL`FSqJYAjk5ED?Yar4J#|=O#V4YAF^lQ(m#wrF<_#8r zc$`PtN(#xgKUMJ#Rq$4pFuv$r*|(MpY#NJtF-HnEJU3B52~wIK2yytCiP^j4a_++Z#7=4tkbg6I2+zM=XWDX19685V@;;)#B#9>$?j#kzE)?N%r; z<}VpTmZPH_r|M)cB>-^7KU|7P9*K03u|}$3DRz+g8nf!Bqin@K*y?5tR32-kTAWfV zCu%OUnrn8mSry~!u^CTX-QTPEhUtz?!7|Zp_4Nu+%BvV2=7qnk9?7vxfUH$!vu`;% z9447U2KuK2)oVK`F=z6vC99^9N(Y%rqzcwzP_&p;P!cu(_Mvt+<7VZ_cB=a+wWFM5 z*STMAGACj}Js#t!C;NMwpfFET%S}dG4QevbbE}Xpj+qUEgr&8rX0|iUg~POe;#91W z42#5Y;oRzjwzilyMxnP&-l>a?>CSVKC;8qynB%BrsBhG&}N5A@9}#bwC(p7WaI zq>HPKEFVyVK!;>hwo*lElPsA{4v?u-s}5Y$0O=14j0hlI47DG4AQ?4k-8swk)Glr9 z=SyNplfl#IHJFB=pHx;XR!g?P=qgvYtEbdUEmu;k zaB}V4P%ti6XqgV~uUntuqTy6TPoX=Zy_A=oxC6?An{DYUc^9y3JR5$FH=hjb z5{)PChYy#JwRr@4>d64gW|BRrs+4MP+R6#pqqT1M=k+HmiC;x4k&agrwTs-BQy1{% zoa}NmfvUH5Qe7@s-Xmw%sKp+qsS8nV{+FqhlJCyWkJ!?D^OPN{%aa8H?B{{z_`w)* z8`6CT-fqLZdi1B!F~|wLq8+Wp)D?d>6EpWzKY;V)acZBb3wQu!3(3z^Rf6W%*5Xym z=-|^q?~=@5nXkDjphI-7oH@9=2-1hf6qKMAR}jtY>j}2kl}qe-+Fhcg5Y53*W4ddQ zMLK0wxM0e@CM$pfCi(rVY|X<&RKA2sJ#ufWkqJs>UCDc9 z6KRKvq{D?NhqvUz`qiLM&axSOPQ9jiYdyK~sQJ^+9dM|O3PU71Ha{;t(sKNk=NPekR5;WqUiyGGz$(BUatMC~R zs>8lHq_-*obB1T&lS&~31t0$LCzn6F%sF>#)Sj)>w_1?h$~7U~@=;+S{ZyXqAp3%O zr;{0uI9+$BOR7%#69p0lMKU@`1ws|5nItEj3->}71h|=D{5I<3CRN?m<|a(A4|Tf~ z9D@PrDX(u1XF&@%tWF}8(O)AFmCe|xoX8gG9;{rYn+HRyG9w^>%+b(d z;1ri#jSWIqXvrtH#dpA}4V|2Zm7+4J)!XST#-+D8-V6328>`>L+h6>jiUI$(&y2*k4n|&UTFFaOm%0#e8hjmX-I!n>@sCCA0I!FEod_Sc_K3tB$y99)XA-1j% z$GSXVb^bm#$pF@^zEhCfPe77{@@;}{9R#*cbnh^Cnk3&l(a}@V&Y^2CqMY?+pLs`@ zfCc-bGN;5l2!rWvk27D97$kX==l1CJlabU42fl#hlg4NZ>Lt8xXTxmyrtER8p12}h zp8W%!;UM`|Hlfq-R+MA?9h;n1n_+ko3m11tF1S}V1z@cr5&{v4x_-B?$Eo4O!u8S$ zinrm)P2_Y#ek62-sypsU=j&S9w$Gc-Z^0i_y{i;k$bu!OQaFA5$t?ThFsV)$U`>Bn zFF59gnHau%3gm~KD$G(%9FJ|fU@~Id+(%ibAI2;o?{Qxi5x{1(A$JG`(-rnPK>iXM zOcfkfOz;n4`yP#wqXmJi@&rdc2M8KdF@Y)pf(FjU1<(ZD>sd z2N_4wzQ>>|4zvG+MwMg}$%I!(hd=Ukq^>g^(FJf{rJ|VR!xL)mEaq75@SrJfWhib- z7mSqDJ4Oxnd(3HaXV%vd4RJ-o$UzPjo$_j4=LaMQ>Z@mT>Ch7z^Pg!Ju#7CX0` z4x8uF(Eofjoda_xX!;~-aFM`AQy@V5UJD_}c0;;Waz|1B*v^F-Y&v{gnb;GX^t6D8I8Bc$#$ZdpK6LS)J3P0e&g+&Xa#yQ_9?M7LlH+ z^<>QfjN5Nv>FYDbk+M~mgGSxTQY#PJ=K;mcZ1(`LaMGq#MYG+C~A|O6J>Hq z7Cy=?HS+k>r>{!QmTjga)5htB7sw;lKbTQA=V%j*(h(kl;>#zik=cov(UXR(iYFbb zI0y1TUkkcZZbzGSH~|!s7hF&Z`VSfIQUE^J39&3 zG6M{$LNZbNme}JMirrS7cUbccasa#hKad-o3?@%{3cu7}_v!ww=hLb5j!k51nMC|v zlBJ^`yGci^Gjg>W7`?%@>QbeT!;?PUm)&RKO&vjc5fxKY8q++OnF%}L88^9GV~Eq* z3a;E|$?~BYH!02_1(|wB;64jz8F_d`Pc_P^ZJ5trvCl8L6)EkryZ{dXSvBWU_aXE5 zlgaxmY34sbK{A5AeoLiJ9g&RMpV()aFvNSnl3PSZgndRq7G2CzA$BSH&pdNzaJDKh zjp%kalb@OtJCVmEsVt93?(TKovsh-8Ta)9*J(F1UjhXaI zzGq=Yp#b?zI;!0RSmw;R4g8QAab&wdViTQZ_Ko4u9=FDO7B1_2TV`Q3M6)CpQ@?0- zb&#o!Yd?9xwW?ED0Y1} z_q#L=jc&sDXRvFUzl&=lS}-NBct%8X;`UZ7Fy9r}TMpMrQLzpKQ%9@MBDDUwFqwr; zl;EK?d>5xKkd7*ez+#;<=JAM#X4b@X3&6@3oyWHnowt>bn3&9@x8YOX(AV0M)*^R# z1{_2(BqS_FG|9{Bky8MZ*p^3eZNs90M;#>@nT3T+JuDQa(F8yrumS`^G{{xYshY)1eEY*%!oj)(g09Kpx0lq?sBQq3 zi8Q_V(x~KXiE&KyR}gZGqv@Iyklw?b;gI9Xz(64K`3{eb>BL%fg4SBF$Rd8mWLFXg z>zsT%$Y{5%;^SF;%l;(}1B8*gM1>7Ax5>25EG$?lXC*QPoAmdK)&Fk#4C=Vk&Y;keeqE@LB|675x_#%lAb)GlpqLSmK? zjWjgMif6^HtC11Sjt`j9pf+h>#cQm!g5kF>M`d>@-?aT-7lU?5^Bf1$m}qu2kf}~+ zzW^0f%U&oxY|6o!Y5u1}0QYO8)k!7U;ploYm5M$^<8KV}ef$ho{X`B{9K#{axyS?% z5dIaK6h%_6`?4IY*h{0h?*-N;srFhOz&~MjIe_BC%kql{5`vqOCt@z5N-^1d+Au`5W)l37U*#UCJ zR?asNcvhL&t->(hFw~9-R%aBEL9r8pHz=&;u8znbxST(8hoxGt*NilqVeSz1lUl2b?0tN4{>GwCd-w21yu_8W+@L?FsWF0yp6 z8%(JZnV7gC9pSlI(uEZce5#5H*tkXeWi7JU_$azr+K+9Ho#0y0jDpNJT{|}rZ>RK# zOm6+pS_-ux-2yhsqm7CpP0n=eM!P$x7GLXY?MfJ8+SxI_rvipY-&8g4^Xl5NTJ@ zQ3$9=<>=hF!F?4UH(A2VclP_oSI~3-K?byWZ9xCy>yJPXU&R6WNSNa-Sc)RUj@-DQ z8PsVolSVX+R?Edz{|{l+{}GlgUOgJLMLft1A>fIwGFeo{;~9oVZ)#wgqmUY;)3^EIGI6&4N~Zbc*+8~KEZZFShu>I=zxaMn{kIvZ2QE$x4V;7%AG8?!Mj zc5>c?!_NTEBP}`}NSe1Xgy2bPWVcE*>HO&>qkLArb2lj~v%Ip}jmZ%+G%8PepIQX-J^IZ``>!J(-5R zEoCrWFZ|M=@yq#F-_PBi+HRd%mj2gnyo%e;_bqzf$11?5&}Z%9=hNlq+7sjD zhbm61TW!e)vwjm-)qOgk(;FYX%iHbR_DCAHwsUbYYb|YO!?*Kg*`ifrE3~k%M_V0$ z|3a)PQH45UW5%IRFFHQ(XS!+<(2O3`owcd&+Y~n;fo>!0Ux^oxjTpZbR^HH?8b-p5i_>l09s&TVN^MnYC4YF_EM~`H*Db$+R;bww%4OzBwR<{@L(i z?w1jcI$`EXG2IXAoBqnFvNbsZ9^U#6%g|{keCXj8kq*cjWHO|hqd9H0t@LJ#HQc<2?4?`r-9I4};5)26k&q8ej~7oD>jAXChiz0Q_H5}f7Lz=LEj(L)d}lolMZZheq1qy#|-1sfL- zEIl#QH{I8FLwIIUsxQ2uwVQHn^mS2OhPoXNmh0BkhLC-2_HuY<`d)gZUM*ULN*)tD zxObbyF#e!_8>iRM2?+&+B9TYfSi()%s5bpNI_A235>P3=HFygUQ^mM^J>@E+F>&xQ?ucr z?wPO8>{<3#tepDQt@j~#OM8PiMfL9Yi^p8@_npOaz|7d9hwFWtue-mW&-Gv9Z1dO? zZmXxOs2vS=3^CS9NC9aCMTq1P)Zcj1c@W7Z!tH|Y_~r(u2=Ui?htg|gCQtE~3gV|I z_uO=faG1nG2p0Uo2$-qgcV3PvXao8@a_H?~f}ig&!o349_aL8RbA4TMd7+=B6CC_j znxLLH0d9$ydwdKx8y@EW8PD@&BW8zwMo4gQ>23mh-t_VY4Kv?txG4Vba4zG2-OK;E zyBTBdjZ+vj7oJpK{G6mf*>tArOp1Tbx{ACGgguJnMuGdV;#(jay?Sm(avbjTU(Y@SQ0(w7fu{ZGEYWNzW&!b)8tB(_LDl->$Y1AvA}@s+r~J7hR`c z##vk(K(Wjdrn_i=&`x9X%{*`X*%y)@=s6N<+{(+lIkhNjJ*?F?&^`1}H zPBtc<$c)m;3z?ogiOZUvjMb2Jl|JjJR32~KU;MPBPm*A zOS9!BK_y3iw|YI&bcj?g_sU4)GIl`lf=t1wn`d-4=P!dh!yK9#PFs}H!sMGAZ!BIX z;ZyD}f<9x9%HfHFhXA!Tafz$87CaEvfwsdEhMZyWW<<@WgU@_jS8NP~Nu|{+1j7$7 z|3eq-2qvdAvtDGJiOL?&N)!?FF{)7e?&`;)8#P_)-@%Ve(pZ%@^}VpHu{sO~0`TtU z=<{%U{O48!Ck=}@_!-ow{goplnhjM`?+fMop{eF(N5s3K-28<%)m z|K*HL6N&ou(&d=zV_94pXQ3O&u6!0tUtv(+w{&JpU4j^{k)LLVolD>GEr6!u!$jmU zTV9OMM^>G3wFJAnw)hh5B!Fi+vl(1=K~nr_&j&pAn%Lpf#@WhUW*Ql);ub<|=?Bu* ziY>2}mldn(mSZ;)-wbB?!Q6oBMTN8x$Mlz%p<^|YKVZ9)dvH_4@D@U!`O`-Sf1k4& z(~NW>yu?+3Qv)ea44qeIq$^ic26yfHxm^fCy>i?7KW_%&Cnu%Xi~z>7AA_UTQv&p) zE}%C5HbQvS{ib!mWcMkKJjI>{GGn`H((@8YpyDpmjp6~-=2X{V;&cs@Jphj6Ey`j< zyb)s+#g{AAmvLG-Lk*!xf;$E&AKda5%x*Y%08HB{ABYMQKBk0l9*S(^J7Alv7h-YA zyX-T;zZUDmh2kqX1t^VBlWvsYoEW*K*3Vn_f(e*M>W}6yO0?r$ud_ z>Cu?P;aIFWU2;P9NDLgYH%NOuz)2YQv0d3>=T3TgWAW2ZHJ&$~sRcx^scWTzM*lgL z$;9FQb6SGT15!kyS0$OpITd5FLrxwP{E03T5^0KkQAWhlx{avpUUnvch+@r`Yx7ge#yV? zImrCqnm!tXiw0vabaxj|cZ^qCaBLY<=kxmaXHZYh#YXR*kE7z(j&DJqsMFh9KrSHI zMWVmGDo;IgAj=P<@C0ya&C^&eQrRwffBsw={U;}Jfa78McKqnA&^=cz9lnHXlQ*Sz z>D}O}0M2K6&Gu1umpa})XO#F5zf6mGqg@;fS@;!8IO}?XJfMYkhVu}|j3%UEn4Ttu zg2_#d-{wlUX#HR|CUEaBgZ$0W6bB&4JpWH^{-0`GWs}{i?PjX&!a-d#1d?<54 zORL&-bysi<&6;orOTv1V#TY}}g(<~Pm_mVUOP5@2s99{P)dwgOTs^WB7HR7d@ytrk zrR}j{JK_!A__H+&r9=sj1~d@f{?L=={LOHg-=rE_X>fCPCxCEb$<;MTj@X#CwsLU$L08t{2D+TWBYiXmHJNB0Z^+>xX%&#x&ectNW5-oUGt9MASW>S5Nu8Uv96{09(}`t^!CaOeD2}cs0qt{7o*(0ThSyCW87KonH4RZ(gl0 zDo^6AheBI0nW2A5U>yPNzK)!4c2BbvY+19<`6vtT>OT4Qj0=9~f|^xA_JW$=mVk~9 z0iX6wmxYguTBsFwgLlPvxy=EG;m7#?CDhuNicji=^(|Da=f>Mp^0g=&oLygEQ-Wx5 zq#j;0@UuTz8V>ro_qQI-0H?%NL79dm%~(RBpyR4>j+K6zqrEX2qr?~6Goq6UQR)V@OHV*d! zF!5aCN;5y~OD;Gs^6wrG)85~3sJFNB47PMb>D}-5_Q$L8UJhx%%UpKvfKKMoM!p|g z8`lomO~J)xuN#{m<2yFdCM_o<5G_f;@dDr%NqzD7@bo%9UBvx6bgfBy@q*~dU61ku zsK(b_4o!W)eeOtsyGHf|RB((!tPixjFgkAC(E<1O7s&zQxP>J3L`!h`#-79hKDg~D L^@l+ZOtAj}q^Ndr From d5e323dcfc467a6186856db702eb9e95a5cf1456 Mon Sep 17 00:00:00 2001 From: Max Ghenis Date: Tue, 3 Nov 2026 16:38:17 -0500 Subject: [PATCH 14/52] Land the IFS addendum's sections 1 and 2 (87 to 144 rows) The IFS family gains 57 rows after its 87 deck rows, which don't change: - 11 from the two pages we read in a browser on 29 October, IFS's initial response and the collection's opening remarks; - 46 from seven Flourish charts, read from each chart's public embed. In 12 of the 57 rows, IFS gives its own calculations, and in the other 45 it restates OBR forecasts. We propose one new package key for them, `ab2026__package_ifs_announced_tax_changes`, and add two rows to `ab2026__package_government_decisions_since_march_2026`. I regenerated COUNTS.json and PROPOSED_KEYS.md and updated the README: 12 families, 5,948 rows, 446 restated. I put the family's stage.py on uk/ab2026-harvest-staging, with the other families' scripts. Co-Authored-By: Claude Opus 5.5 --- .../harvest-uk-ab2026-2026-10-29/COUNTS.json | 10 +- .../PROPOSED_KEYS.md | 5 +- .../harvest-uk-ab2026-2026-10-29/README.md | 8 +- .../uk_ifs_ab2026/NOTES.md | 320 ++++++++++++------ .../uk_ifs_ab2026/claims_staged.jsonl.gz | Bin 9694 -> 17171 bytes .../uk_ifs_ab2026/manifest.jsonl | 9 + 6 files changed, 231 insertions(+), 121 deletions(-) diff --git a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json index 177d7765c..87be47b88 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json +++ b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json @@ -41,9 +41,9 @@ ] }, "uk_ifs_ab2026": { - "rows": 87, - "primaries": 5, - "claims_sha256": "2e8b2dc5a85bb66fb11431f5c20565a40b2cf546e7edaabedd3a6c56ddb68d49", + "rows": 144, + "primaries": 14, + "claims_sha256": "40331bc6beb36fc78d2e10731730c4c8a15942d4419bd840d0cebc95988e6125", "sources": [ "ifs" ] @@ -100,8 +100,8 @@ ] }, "_totals": { - "rows": 5891, - "restated": 401, + "rows": 5948, + "restated": 446, "families": 12, "pinned_on": "2026-10-29" } diff --git a/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md b/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md index c0427858d..6621ddf2c 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md +++ b/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md @@ -4,7 +4,7 @@ Generated by `pipeline/ab2026_proposed_keys.py` from the staged families; don't A row whose scope matches no registered measure has `measure_key: null` and names one of these keys in `conditions.proposed_measure_key` (README rule 2). A key is a measure with no Table 6.1 line (`ab2026__`), a package (`ab2026__package_`) or an option (`ab2026_option__`). Each family's NOTES.md gives the key's scope as Table 6.1 lines or the producer's own words. Tranche 2 registers each key in `data/uk/ab2026_measures.json`, or folds it into a registered key, before the ingest. -83 proposed keys (15 measures, 55 packages, 13 options) on 955 rows. +84 proposed keys (15 measures, 56 packages, 13 options) on 958 rows. | proposed key | kind | rows | families | producers | |---|---|--:|---|---| @@ -29,11 +29,12 @@ A row whose scope matches no registered measure has `measure_key: null` and name | `ab2026__package_dwp_fraud_and_error_measures` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_fuel_duty_and_electricity_vat` | package | 11 | uk_obr_tables | obr_efo | | `ab2026__package_fuel_duty_freezes_since_2010_11` | package | 1 | uk_obr_tables | obr_efo | -| `ab2026__package_government_decisions_since_march_2026` | package | 374 | uk_hmt_redbook, uk_obr_tables, uk_rf_ab2026 | hm_treasury, obr_efo, resolution_foundation | +| `ab2026__package_government_decisions_since_march_2026` | package | 376 | uk_hmt_redbook, uk_ifs_ab2026, uk_obr_tables, uk_rf_ab2026 | hm_treasury, ifs, obr_efo, resolution_foundation | | `ab2026__package_hmrc_tax_gap_and_online_marketplace` | package | 7 | uk_hmt_redbook | hm_treasury | | `ab2026__package_hmt_distributional_budget_2026` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_hmt_distributional_since_ab2024` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_household_energy_payment_and_electricity_vat` | package | 5 | uk_rf_ab2026 | resolution_foundation | +| `ab2026__package_ifs_announced_tax_changes` | package | 1 | uk_ifs_ab2026 | ifs | | `ab2026__package_national_wealth_fund_and_british_business_bank` | package | 4 | uk_obr_tables | obr_efo | | `ab2026__package_non_scorecard_decisions` | package | 12 | uk_obr_tables | obr_efo | | `ab2026__package_other_ame_spending_decisions` | package | 6 | uk_obr_tables | obr_efo | diff --git a/sources/harvest-uk-ab2026-2026-10-29/README.md b/sources/harvest-uk-ab2026-2026-10-29/README.md index e9fbddcd7..d085d63a8 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/README.md +++ b/sources/harvest-uk-ab2026-2026-10-29/README.md @@ -48,7 +48,7 @@ The AB2025 harvest README's rules 1–10 and its row contract apply, with the AB ## Families staged -12 families, 5,891 rows (401 `restated`), as `COUNTS.json` pins them. +12 families, 5,948 rows (446 `restated`), as `COUNTS.json` pins them. | family | rows | primaries | producer slugs | |---|--:|--:|---| @@ -57,7 +57,7 @@ The AB2025 harvest README's rules 1–10 and its row contract apply, with the AB | `uk_hmt_costings` | 597 | 1 | `hm_treasury` | | `uk_hmt_redbook` | 432 | 5 | `hm_treasury` | | `uk_hmt_tables` | 762 | 2 | `hm_treasury` | -| `uk_ifs_ab2026` | 87 | 5 | `ifs` | +| `uk_ifs_ab2026` | 144 | 14 | `ifs` | | `uk_ippr` | 3 | 3 | `ippr` | | `uk_obr_tables` | 2,797 | 3 | `obr_efo` | | `uk_rf_ab2026` | 766 | 7 | `resolution_foundation` | @@ -65,7 +65,7 @@ The AB2025 harvest README's rules 1–10 and its row contract apply, with the AB | `uk_tpa` | 305 | 10 | `tax_policy_associates` | | `uk_trussell_wpi` | 1 | 1 | `trussell_wpi` | -Each family's `stage.py`, which writes its claims file, is kept on branch `uk/ab2026-harvest-staging` under `harvest-staging/ab2026/lanes///`, with the lane briefs and the five reviews; this branch carries only each family's three files, as AB2025 does. The review fixes, the IFS addendum's sections 3 and 4 and the cross-family key harmonisation were applied on 3 November 2026; each family's NOTES.md records them. +Each family's `stage.py`, which writes its claims file, is kept on branch `uk/ab2026-harvest-staging` under `harvest-staging/ab2026/lanes///`, with the lane briefs and the five reviews; this branch carries only each family's three files, as AB2025 does. On 3 November 2026 we applied the review fixes, all four sections of the IFS addendum and the cross-family key harmonisation, and each family's NOTES.md records them. **Families with no rows** are not carried here, as AB2025 has no empty family. Their working files stay on the staging branch: @@ -74,8 +74,6 @@ Each family's `stage.py`, which writes its claims file, is kept on branch `uk/ab - `uk_niesr`: NIESR's Budget-day reaction ("Breathing Space, Not a Strategy") was fetched, but it holds macro and fiscal-stance commentary for the #55 lane and restated OBR figures, which the AB2025 NIESR family did not stage either. - `uk_fai_sfc`: the Fraser of Allander Institute's reaction refused the fetch (HTTP 403), and the Scottish Fiscal Commission published nothing in scope on 28 or 29 October. The Scottish Government's five rows, first staged here, moved to `uk_devolved_govts` on 3 November. -**Still to come from the coordinator:** rows from the two IFS pages read in a browser on 29 October (the initial response and the collection's opening remarks), which are not on GitHub. `uk_ifs_ab2026/NOTES.md` lists them as pending. - ## Producer slugs and families | slug | producer | family dir | diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/NOTES.md index 1f211e930..a6987febc 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/NOTES.md @@ -1,64 +1,55 @@ # uk_ifs_ab2026 — Institute for Fiscal Studies, Autumn Budget 2026 -**Working files.** `stage.py`, which writes `claims_staged.jsonl.gz` and is the file meant wherever these notes say `stage.py`, lives on branch `uk/ab2026-harvest-staging` at `harvest-staging/ab2026/lanes/ifs/uk_ifs_ab2026/stage.py`; this branch carries only NOTES.md, `manifest.jsonl` and `claims_staged.jsonl.gz`. `.ab2026-primaries/uk_ifs_ab2026/` was the lane's local primaries folder and was never committed: the repository carries no primary bytes, so fetch each primary from its `manifest.jsonl` URL and check its sha256. +**Working files.** `stage.py`, which writes `claims_staged.jsonl.gz` and is the file meant wherever these notes say `stage.py`, lives on branch `uk/ab2026-harvest-staging` at `harvest-staging/ab2026/lanes/ifs/uk_ifs_ab2026/stage.py`; this branch carries only NOTES.md, `manifest.jsonl` and `claims_staged.jsonl.gz`. `stage.py` embeds every row, so it rebuilds the gzip with no network and no primary file. `.ab2026-primaries/uk_ifs_ab2026/` is the harvest's local primaries folder, which no commit carries: the repository holds no primary bytes. Fetch each deck from its `manifest.jsonl` URL, and each Flourish embed from the embed URL in its manifest note, and check its sha256. The two browser-text primaries are each page's `
` text as saved on 29 October; ifs.org.uk refuses curl and its HTML embeds per-request tokens, so only the saved texts match those two hashes. -**87 rows from four current briefing decks** (five manifested PDF versions, including the superseded TW_0). `source: ifs`. The first lane staged 106 rows on 29 October. On 3 November the IFS addendum (`briefs/ifs-addendum.md` on the staging branch) was applied: sections 3 and 4 in full, section 2 as far as the network allowed (none), section 1 not at all (see "Pending" below). Its edits sit in an explicit block at the end of `stage.py`'s data. The sections below this summary keep the first lane's per-deck record; where they differ from the summary, the summary wins. - -## Pending: rows to be added from browser-read text (addendum section 1) - -These two IFS pages refuse non-browser clients, and their text is not on GitHub. The coordinator read them in a browser on 29 October and adds their rows locally after this family lands: - -- https://ifs.org.uk/articles/autumn-budget-2026-initial-response — pending: rows to be added from browser-read text. -- https://ifs.org.uk/collections/autumn-budget-2026 (Helen Miller's opening remarks) — pending: rows to be added from browser-read text. - -The addendum brief records that the coordinator's browser read found no estimates in the IFS explainer or the fiscal devolution comment, so those pages are no longer listed as unreached for want of access; they hold nothing to stage. - -## Unreached: the seven IFS data items (addendum section 2) - -The seven charts in the collection's "Key charts" section (listed in `harvest-staging/ab2026/ifs/data-items.json` on the staging branch) serve their data through Flourish. On 3 November the session's network policy refused public.flourish.studio and flo.uri.sh at the proxy (HTTP 403 to CONNECT), so no visualisation's data was read and none is staged. No value was read off an image. Each is unreached: - -| Data item | Flourish | -|---|---| -| https://ifs.org.uk/data-items/growth-real-household-disposable-income-over-decade-1950s-2020s | visualisation/30734907 | -| https://ifs.org.uk/data-items/current-budget-deficit-march-2026-and-october-2026-forecasts | visualisation/30726471 | -| https://ifs.org.uk/data-items/headroom-share-gdp | visualisation/30726788 | -| https://ifs.org.uk/data-items/real-gdp-growth-over-forecast | visualisation/30726917 | -| https://ifs.org.uk/data-items/average-annual-growth-real-household-disposable-income-parliament | visualisation/30726939 | -| https://ifs.org.uk/data-items/long-term-impact-announced-tax-changes-fiscal-event | visualisation/30726047 | -| https://ifs.org.uk/data-items/number-individuals-aged-16-paying-higher-or-additional-rate-income-tax-0 | visualisation/30727239 | +**144 rows from 13 primaries.** `source: ifs`. The manifest lists 14 primaries: four current briefing decks, the superseded TW_0 deck (no rows), two IFS pages read as browser text, and the public data of seven Flourish visualisations. The first lane staged 106 rows from the decks on 29 October. On 3 November the harvest applied the IFS addendum (`briefs/ifs-addendum.md` on the staging branch). Its sections 3 and 4 cut the deck rows to 87, through an explicit block that follows the deck rows in `stage.py`. Its sections 1 and 2 then appended 57 rows in a second JSON block: 11 from the two pages and 46 from the seven visualisations. Attribution: 48 `own`, 96 `restated`. No primary mentions PolicyEngine. From "Family conventions" on, the sections keep the first lane's record of 29 October; where they differ from the sections above them, the sections above win. ## Coverage -| Primary | Rows | -|---|--:| -| Weathering the fiscal storm? (Isabel Stockton) | 39 | -| Cost of living measures, TW_1 (Tom Wernham) | 15 | -| Cost of living measures, TW_0 (superseded) | 0 | -| Spending on public services: the Achilles' heel? (Bee Boileau) | 21 | -| Tax measures (Stuart Adam) | 12 | - -| measure_key | Rows | of which components (rule 9) | -|---|--:|--:| -| `null`, context (no proposed key) | 60 | 0 | -| `null` with a proposed key | 4 | 0 | -| `ab2026__founders_relief_cgt` | 4 | 0 | -| `ab2026__household_energy_payment_100_bands_a_to_d` | 3 | 0 | -| `ab2026__package_household_energy_vat_and_fuel_duty` | 3 | 0 | -| `ab2026__hmrc_tax_gap_package` | 2 | 2 | -| `ab2026__badr_lifetime_limit_500000` | 1 | 0 | -| `ab2026__business_rates_sbrr_rhl_thresholds_raised` | 1 | 0 | -| `ab2026__distributions_framework_modernised` | 1 | 0 | -| `ab2026__electricity_vat_zero_rate_to_july_2027` | 1 | 0 | -| `ab2026__fuel_duty_5p_cut_extended_and_rpi_rise_deferred` | 1 | 0 | -| `ab2026__package_total_tax_policy_decisions` | 1 | 0 | -| `ab2026__writing_down_allowance_12pc_and_fya_50pc` | 1 | 0 | - -Attribution: 36 `own`, 51 `restated`. +| Primary | Access | Rows | own | restated | +|---|---|--:|--:|--:| +| Weathering the fiscal storm? (Isabel Stockton) | direct | 39 | 2 | 37 | +| Cost of living measures, TW_0 (superseded) | local copy | 0 | 0 | 0 | +| Cost of living measures, TW_1 (Tom Wernham) | direct | 15 | 11 | 4 | +| Spending on public services: the Achilles' heel? (Bee Boileau) | direct | 21 | 19 | 2 | +| Tax measures (Stuart Adam) | direct | 12 | 4 | 8 | +| Autumn Budget 2026: initial response | browser text | 5 | 1 | 4 | +| Collection page: opening remarks by Helen Miller | browser text | 6 | 4 | 2 | +| Flourish 30734907: real household disposable income by decade | Flourish public data | 5 | 5 | 0 | +| Flourish 30726471: current budget deficit, March and October 2026 forecasts | Flourish public data | 11 | 0 | 11 | +| Flourish 30726788: headroom as a share of GDP | Flourish public data | 2 | 0 | 2 | +| Flourish 30726917: real GDP over the forecast | Flourish public data | 11 | 0 | 11 | +| Flourish 30726939: real household disposable income growth by parliament | Flourish public data | 1 | 1 | 0 | +| Flourish 30726047: announced tax changes by fiscal event | Flourish public data | 1 | 1 | 0 | +| Flourish 30727239: higher and additional rate taxpayers | Flourish public data | 15 | 0 | 15 | +| **Total** | | **144** | **48** | **96** | + +Rows 0 to 86 come from the decks, by deck, slide and position. Rows 87 to 143 follow in read order: the initial response, the opening remarks, then the seven visualisations in the order above, each by data row and then series. + +| measure_key | Rows | of which components (rule 9) | of which from sections 1 and 2 | +|---|--:|--:|--:| +| `null`, context (no proposed key) | 116 | 0 | 52 | +| `null` with a proposed key | 7 | 0 | 3 | +| `ab2026__household_energy_payment_100_bands_a_to_d` | 5 | 0 | 2 | +| `ab2026__founders_relief_cgt` | 4 | 0 | 0 | +| `ab2026__package_household_energy_vat_and_fuel_duty` | 3 | 0 | 0 | +| `ab2026__hmrc_tax_gap_package` | 2 | 2 | 0 | +| `ab2026__badr_lifetime_limit_500000` | 1 | 0 | 0 | +| `ab2026__business_rates_sbrr_rhl_thresholds_raised` | 1 | 0 | 0 | +| `ab2026__distributions_framework_modernised` | 1 | 0 | 0 | +| `ab2026__electricity_vat_zero_rate_to_july_2027` | 1 | 0 | 0 | +| `ab2026__fuel_duty_5p_cut_extended_and_rpi_rise_deferred` | 1 | 0 | 0 | +| `ab2026__package_total_tax_policy_decisions` | 1 | 0 | 0 | +| `ab2026__writing_down_allowance_12pc_and_fya_50pc` | 1 | 0 | 0 | ## Proposed measure keys +Four proposed keys name seven rows. README rule 2 sets their spelling, and each takes the producer's printed scope (rule 3). + - `ab2026__package_property_and_savings_income_rates` (1 row): exactly Table 6.1 lines 61 and 62, Adam slide 2's "Increased income tax on savings & property income", £1.1bn in 2031–32 (lines 61 and 62 print 455 + 590 = 1,045). - `ab2026__gas_standing_charge_moved_to_unit_rates` (3 rows): no Table 6.1 line. Wernham slide 8's "Shifting gas standing charge onto unit price": "This charge will be scrapped, but per-unit price will be increased to absorb this". The spelling is the RF family's. +- `ab2026__package_government_decisions_since_march_2026` (2 rows, from the initial response): the key the OBR, HM Treasury red book and RF families share for all Government decisions since the March 2026 forecast, the OBR's "Total effect of Government decisions". The page's £8 billion this year and £4 billion next year measure how much the Budget's new giveaways add to borrowing. To the nearest billion they match the OBR's direct effects of Government decisions in EFO Table 3.1, 7.6 in 2026-27 and 3.5 in 2027-28 (uk_obr_tables), so both rows name that component in `conditions.component`, as the OBR family does. Neither the total effect including indirect effects (6.3 and 7.0) nor Table 6.1's total of policy decisions (7,595 and 2,475 £m, uk_hmt_tables) matches both years. +- `ab2026__package_ifs_announced_tax_changes` (1 row, from Flourish 30726047): the tax change that Budget 2026 announced, at the forecast horizon, as a share of GDP as forecast at the event. The chart's footer credits IFS calculations from the OBR's October 2026 EFO, policy measures database and historical official forecasts database. The chart lists no Table 6.1 line and names no OBR grouping, and one rounded figure cannot tie it to the scope of `ab2026__package_total_tax_policy_decisions` or of the OBR's tax decisions since March, so the row proposes a key in IFS's own terms. ## Keys the addendum changed (rules 3 and 9) @@ -74,8 +65,56 @@ Attribution: 36 `own`, 51 `restated`. **Defence Investment Plan funding.** The addendum asked whether IFS's "£1.2bn per year gap in Defence Investment Plan funding" is an estimate (keep, as context) or not. The first lane's own note calls it the government's DIP funding announcement, restated: the Budget "filled" the gap. That makes it a spending allocation, which rule 8 excludes, so it is dropped rather than kept as context. It is close to, but not exactly, Table 6.1 line 41 (capital reallocation to fully fund the DIP: +1,100, +1,135, +1,180 in 2029-30 to 2031-32), so it is not keyed there either. + +## Rows from addendum sections 1 and 2 + +### The two pages (11 rows) + +The rows cite each page's saved `
` text by line. + +The initial response (rows 87 to 91): + +- Line 16: £13 billion and £15 billion, `fiscal_headroom` in 2029–30 before and after the Budget's policy decisions (`policy_scenario` pre-measures and post-measures). They restate the OBR, which the sentence names. The OBR family stages £12.5 billion and £14.5 billion against the target "Current budget to be in surplus by 2029-30" (uk_obr_tables); the rows keep the page's rounded figures. +- Line 18: £8 billion and £4 billion, `decisions_effect_on_borrowing` in 2026-27 and 2027-28 (this year and next at the 28 October Budget), restated, keyed as above. +- Line 28: 80%, `affected_share` 0.8. IFS's estimate of the share of households in England that receive the £100 payment, keyed to `ab2026__household_energy_payment_100_bands_a_to_d`; `reform_hint` keeps the £100. + +The opening remarks (rows 92 to 97; the saved text holds the whole speech on line 6): + +- 98: proposed metric `measure_count` in proposed unit `measures`, `measure_key` null, no proposed key. A count of the Budget's measures, staged as the AB2025 and AB2026 uk_thinktank_misc families stage such counts (`own`, `different_model`). Table 6.1 numbers its lines 1 to 98 (`data/uk/ab2026_measures.json`). +- £17 billion: proposed metric `spending_pressure`, IFS's estimate of the top-up that avoiding real-terms cuts could need. The page prints its year as "2030-21", a typo, most likely for 2030–31. The row keeps "2030-21" in the quote and `conditions.horizon`, sets no `fy`, takes 2026 as a publication-year marker, and has medium confidence. +- £11 billion: `spending_pressure` for defence plans, on top of the £17 billion. The page prints no year; Boileau slide 4 prints £11 billion more defence spending in 2030–31 (row 55). Medium confidence and a 2026 marker. +- 25% and 40%: proposed metric `reserve_allocated_share`, the shares of the capital and day-to-day reserves already spent. Restated `administrative_fact`, like the Boileau slide 11 rows, which print the same two shares. +- 80%: `affected_share` 0.8 of households, for the £100 payment through council tax bands A-D, keyed to `ab2026__household_energy_payment_100_bands_a_to_d`. + +### The seven visualisations (46 rows) + +`.ab2026-primaries/uk_ifs_ab2026/browser/data-items.json` lists the seven data items in the collection's key charts, each with its Flourish id and lead sentence. The rows read each chart's data cells from its public embed (see "Access recipe and primary inventory"). Following the addendum, they stage the October 2026 forecast values, and the March 2026 values where a chart compares the two, as restated OBR figures unless IFS computed them; IFS's own calculations are `own`. Outturns stay unstaged unless a lead states one as its finding. + +| Flourish id | Rows | Metric and unit | What the rows hold | Attribution | +|---|--:|---|---|---| +| 30734907 | 5 | `real_household_disposable_income_index` (proposed), `index` (proposed) | the 2020s forecast path, years 6 to 10 of the decade, read as 2026 to 2030; the lead's finding is the year-10 value | `own`: IFS builds the decade index from OBR and ONS data, and the addendum classes this chart as own | +| 30726471 | 11 | `current_budget_deficit` (proposed), `gbp` | the October 2026 forecast for 2026-27 to 2031-32 and the March 2026 forecast for 2026-27 to 2030-31 | `restated` OBR | +| 30726788 | 2 | `fiscal_headroom`, `share` | headroom at the March 2026 and October 2026 forecasts, as fractions of GDP | `restated` OBR | +| 30726917 | 11 | `real_gdp_index` (proposed), `index` (proposed) | the March 2026 forecast for 2026-27 to 2030-31 and the October 2026 forecast for 2026-27 to 2031-32, with 2025-26 = 100 | `restated` OBR; the addendum classes this chart as restated | +| 30726939 | 1 | `real_income_growth`, `percent` | 0.4% a year over the 2024-2029 parliament | `own`: the footer credits IFS calculations | +| 30726047 | 1 | `revenue_change`, `percent` | Budget 2026's tax change at the forecast horizon, 0.2% of GDP, with the proposed key above | `own`: the footer credits IFS calculations | +| 30727239 | 15 | `taxpayer_count`, `persons` | the three forecast series (the freeze at the March and October 2026 forecasts, and CPI uprating at the October forecast), 2026-27 to 2030-31 | `restated` OBR: the footer cites the OBR EFO, October 2026 | + +Only the tax-change row carries a key. The other 45 hold forecast levels or IFS indicators, not the effect of one measure, so they have `measure_key` null and no proposed key. + +### Conventions for these rows + +- **Quotes and locators.** Each Flourish row quotes "row, series: value", and its locator names the visualisation, the data row and the series. `value_raw` is the cell as the embed serves it, at full precision, whatever the chart displays. +- **Units.** The deficit chart's y-axis title is "£ billion", so its cells scale by 1e9 to GBP. The taxpayer chart's cells are thousands: the lead's 10.8 million is the 2030-31 October freeze cell, 10800, and the chart's number format sets a constant of 1000. They scale by 1,000 to persons. The headroom cells are fractions of GDP, staged as served in unit `share`; the chart's number format sets a constant of 100, one decimal place and a % suffix. Index levels keep proposed unit `index`, not the registered `index_0_1`. The parliament and tax-change cells are percentages: the number format adds a % suffix, and the tax column's name says "% of GDP". +- **Periods.** Fiscal-year rows take the end year and `conditions.fy`. The decade chart labels its rows "Year of decade" 0 to 10; the rows read years 6 to 10 of the 2020s as calendar 2026 to 2030, which the chart does not print, so they have medium confidence. The headroom and tax-change cells are fiscal-event snapshots (`point_in_time`, period 2026). The parliament average takes the parliament's end year, 2029, and keeps 2024-2029 in `conditions.horizon`. Text rows with no printed year take 2026 as a disclosed publication-year marker and import no year from the decks. +- **Vintages.** `conditions.forecast_vintage` follows the series or row label, March 2026 or October 2026. The decade chart's footer names the OBR's October 2026 EFO, and the parliament chart's lead names OBR forecasts, so those rows read October 2026. +- **Signs.** As served. A negative deficit cell is a surplus; the chart shades the area below zero "Current budget surplus". +- **Sources.** Restated OBR rows carry `source_model: obr_efo_forecast` and `benchmark_class: different_model`; IFS's own calculations carry `arithmetic`. The two reserve shares are `administrative_fact`, as in the Boileau rows. + ## Not staged (rules 8 and 9) +### Deck rows the addendum dropped + The addendum dropped 19 of the first lane's 106 rows, all under rule 8: | Category | Rows | Examples (deck, slide) | @@ -88,15 +127,61 @@ The addendum dropped 19 of the first lane's 106 rows, all under rule 8: Each dropped parameter that names a measure stays in the kept rows' `reform_hint`. The 25% capital and 40% day-to-day reserve shares (Boileau 11) stay: the addendum lists them as estimates. The bond spread and debt-ownership context rows (Stockton 7) are not design and stay as context. +### Figures on the two pages (11) + +| Page | Figure | Times printed | Why the rows leave it out | +|---|---|--:|---| +| Initial response | 3% of GDP on defence by 2030 | 2 | a government target (rule 8) | +| Initial response | the £100 payment | 1 | a measure parameter (rule 8); row 91's `reform_hint` keeps it | +| Opening remarks | 3% of GDP on defence | 3 | a government target | +| Opening remarks | 3.5% by 2035 | 1 | a government target | +| Opening remarks | 2.7% of GDP from next year to 2029–30 | 1 | the government's defence spending plan, a target | +| Opening remarks | the £100 payment | 1 | a measure parameter; row 97's `reform_hint` keeps it | +| Opening remarks | property values 35 years out of date | 1 | it describes council tax valuations and estimates no Budget effect | +| Opening remarks | local areas to retain 100% of business rates | 1 | the design of a devolution plan (rule 8) | + +The pages' other numbers are not figures to stage: publication dates, titles and target dates (2030, 2035); horizons ("next five years", and "next three years", which row 95 keeps in `conditions.horizon`); the fiscal rule's year, 2029–30, which the headroom rows carry as `fy`; "2030-21", which row 93 keeps; the name of the "10 year plan"; the G7; the count of "two big problems"; and "one-off". The remarks' "almost halved" prints no number, and no row turns it into one. + +### Data cells in the seven visualisations (256 of 302) + +| Flourish id | Non-null cells | Staged | Excluded | The excluded cells | +|---|--:|--:|--:|---| +| 30734907 | 89 | 5 | 84 | the 1950s to 2010s paths (77, including seven base-100 cells); the 2020s series, years 0 to 5 (6, including its base-100 cell), which the chart does not mark as forecast; and the forecast series' year-5 cell (1), which repeats the 2020s series' year-5 value where the two lines meet | +| 30726471 | 13 | 11 | 2 | 2025-26 in both series, 47.4 each: a year that had ended before the October forecast, with one value for both series. The March series has no 2031-32 cell | +| 30726788 | 32 | 2 | 30 | headroom at the 30 earlier fiscal events, Jun 10 to Nov 25 | +| 30726917 | 13 | 11 | 2 | the 2025-26 base cells, 100 in both series. The March series has no 2031-32 cell | +| 30726939 | 12 | 1 | 11 | the 11 earlier parliaments, 1979-1983 to 2019-2024 | +| 30726047 | 85 | 1 | 84 | every earlier fiscal event, Budget 1971 to Spring Forecast 2026; one further data row is blank | +| 30727239 | 58 | 15 | 43 | the Current plans series, 1990-91 to 2025-26 (36); the two freeze series' 2025-26 cells (2), each 8510, the Current plans value for that year; and the CPI-uprating series for 2021-22 to 2025-26 (5), a counterfactual path for the years before the forecast years, whose 2021-22 cell equals Current plans' 4950 | +| **Total** | **302** | **46** | **256** | | + +Outside the data cells, the headroom chart draws a reference line labelled with the 2010s average headroom, 0.9% of GDP: a historical average, which no row stages. The lead sentences in `data-items.json` belong to no manifested primary. Each estimate a lead states repeats a staged cell (0.4% a year is row 127, around 0.2% of GDP is row 128, 10.8 million is row 142), and their other numbers are dates. Axis titles define index bases; tick labels, guide positions and chart settings hold no estimates. No row reads a value off an image, a bar, a line or a tick. + +## Unreached + +Nothing the addendum names remains unreached. + +- **The seven IFS data-item pages.** ifs.org.uk refuses curl, and the addendum rules out any other route, so each manifest line names its data item's URL and hashes the chart's public Flourish embed instead. The rows do not use the "Download data" link that each chart's footer template names. +- **The explainer and the fiscal devolution comment.** The browser read of 29 October found only an embedded video in the first and only design parameters in the second, so neither has a manifest line or rows. +- **The briefing event page** (https://ifs.org.uk/events/autumn-budget-2026-ifs-analysis) refused the first lane's curl with a Cloudflare challenge (see below); no primary holds it. + ## Verification (3 November 2026) -- **Checker:** `PYTHONPATH=. python pipeline/ab2026_harvest_contract.py sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026` reports 87 rows, 0 problems. -- **Determinism:** a re-run of `stage.py` reproduces `claims_staged.jsonl.gz` byte for byte. COUNTS.json pins its sha256. -- **Spot check: not done.** The addendum asked for 15 random rows re-opened at their locators, five from the new primaries. There are no new primaries (sections 1 and 2 above), and the session's network policy refused ifs.org.uk, so no deck could be re-fetched. Every kept row's printed value is unchanged from the first lane, whose spot checks are recorded in the deck sections below. -- **Stale files:** the family had no report or spot-check files. +- **Checker:** `PYTHONPATH=. python pipeline/ab2026_harvest_contract.py sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026` reports 144 rows, 0 problems. +- **Determinism:** two runs of `stage.py` write the same gzip, sha256 `40331bc6beb36fc78d2e10731730c4c8a15942d4419bd840d0cebc95988e6125`. The 87-row file before sections 1 and 2 had `2e8b2dc5a85bb66fb11431f5c20565a40b2cf546e7edaabedd3a6c56ddb68d49`. COUNTS.json pins the new hash. +- **The 87 deck rows:** the new gzip's first 87 serialised lines match the 87-row file byte for byte (3,247 leaf fields, 0 differences), and manifest lines 1 to 5 match theirs. +- **Every new row:** a script re-opened all 57 rows at their locators: the saved text's line, or the embed's data row and series, decoding the embed afresh with every number kept as served text. It checked the primary's sha256 against the row and the manifest, the quote against the line or cell, `value_raw` against the cell, and the value, sign, unit, period and forecast vintage. All 57 pass. Nine planted errors (value, sign, period, `fy`, series, unit, vintage, quote and hash) each made it fail. +- **Cell accounting:** the same decoder lists every non-null cell in the seven embeds: 302, of which the rows stage 46. +- **Flourish bytes:** fetches on 29 October and 3 November returned byte-identical embeds. +- **Tests:** `PYTHONPATH=. python -m pytest -n 4 --noconftest tests/test_ab2026_harvest_contract.py tests/test_uk_ab2026_harvest.py`: 52 passed. `--noconftest` skips the conftest's build of the derived database, which neither file uses. +- **Generated files:** `pipeline/ab2026_proposed_keys.py` and `pipeline/harvest_counts.py --pinned-on 2026-10-29` regenerate PROPOSED_KEYS.md and COUNTS.json, and both `--check` runs pass. + +The scripts and their outputs sit in the ignored `.ab2026-primaries/uk_ifs_ab2026/a3/`. ## Access recipe and primary inventory +### Decks (first lane, 29 October) + Fresh fetches began **2026-10-29 15:43:43 UTC**, using exactly: ```sh @@ -109,14 +194,6 @@ family. SHA256s were checked with `shasum -a 256`. PDFs were read with `pdftotext -layout` and rendered for visual checks of labels and embedded images. None of the PDF annotations links to a chart-data download. -| Primary | Access | Bytes | Rows | SHA256 | -|---|---|--:|--:|---| -| [Weathering the fiscal storm?](https://ifs.org.uk/system/files/2026-10/Weathering-the-fiscal-storm-IS.pdf) | direct | 318,348 | 39 | `fd09e707a4bc0a8e0a3de4a3fa902c8ccdc1e4c44ccdd0dee92866857aa4690b` | -| [Cost of living measures (TW_0, superseded)](https://ifs.org.uk/system/files/2026-10/Cost-of-living-measures-TW_0.pdf) | local copy | 231,180 | 0 | `abb63d9c4ecd21c3c9173077c5e92d8015c7cc30aebd12be159185d7d157f12a` | -| [Cost of living measures (TW_1, revised)](https://ifs.org.uk/system/files/2026-10/Cost-of-living-measures-TW_1.pdf) | direct | 222,191 | 21 | `6ed174b22b4d22a34c8bbd450194fb2b4181a9d566b137913bee52ca49a0299b` | -| [Spending on public services: the Achilles heel](https://ifs.org.uk/system/files/2026-10/Spending-on-public-services-the-Achilles-heel-BB.pdf) | direct | 157,838 | 28 | `7403ecc0641b23f5ffc0cb28f67435dbd1b8df6b124874ffae7ae7f75e357aa6` | -| [Tax measures](https://ifs.org.uk/system/files/2026-10/Tax-measures-SA.pdf) | direct | 133,005 | 18 | `6792dc448f40d34f174b6a3c839863b5c969a7334263f5d6b539be62a0428bb8` | - Stockton's fresh file is byte-for-byte identical to assigned `stockton.pdf`, confirming it is **Weathering-the-fiscal-storm-IS.pdf**. The other three current decks also match their supplied local copies. TW_0 returned HTTP 403; the @@ -124,7 +201,42 @@ supplied local PDF was copied to `TW_0-local.pdf` and hashed. All rows from the cost-of-living presentation cite TW_1, not the superseded version. Full TW_0/TW_1 changes are documented below; no numerical claim changed. -HTML pages that returned a Cloudflare challenge (HTTP 403) to the first lane at 15:43:43 UTC; see "Pending" above for what has become of each: +### Browser text (the initial response and the opening remarks) + +On 29 October, between 16:56Z and 16:58Z, the harvest read the two pages in the Claude desktop app's built-in browser, as an ordinary page visit with no challenge solved. It fetched each page's HTML in-page with `fetch(location.href)` and hashed it. Two fetches of the initial response gave the same length, 257,037 bytes, with different hashes: the HTML embeds per-request tokens. So the primary is the text of each page's `
` element, and each manifest note records the HTML's byte count and one fetch's hash. Both pages print publication on 28 October 2026; the remarks are Helen Miller's opening remarks at the 29 October briefing. + +### Flourish public data (the seven visualisations) + +```sh +UA="Mozilla/5.0 (Macintosh; Intel Mac OS X 14_0) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/17.0 Safari/605.1.15" +curl -sSL --max-time 50 -A "$UA" https://public.flourish.studio/visualisation/ID/ # names the embed URL +curl -sSL --max-time 50 -A "$UA" https://flo.uri.sh/visualisation/ID/embed # the hashed primary +``` + +Both answered HTTP 200 for all seven ids on 3 November. Each embed is an HTML page whose inline script assigns `_Flourish_data`, `_Flourish_data_column_names` and `_Flourish_settings` as JSON; a JSON decoder reads them from the assignment onward. `_Flourish_data_timestamps` holds a JavaScript `new Date(...)`, which needs its own parse. Each manifest note gives the data item's URL, the public page's bytes and hash, the embed URL, the fetch time, the Flourish version, the data's last update and the lead sentence from `data-items.json`. + +### Inventory + +| Primary | Access | Bytes | Rows | SHA256 | +|---|---|--:|--:|---| +| [Weathering the fiscal storm?](https://ifs.org.uk/system/files/2026-10/Weathering-the-fiscal-storm-IS.pdf) | direct | 318,348 | 39 | `fd09e707a4bc0a8e0a3de4a3fa902c8ccdc1e4c44ccdd0dee92866857aa4690b` | +| [Cost of living measures (TW_0, superseded)](https://ifs.org.uk/system/files/2026-10/Cost-of-living-measures-TW_0.pdf) | local copy | 231,180 | 0 | `abb63d9c4ecd21c3c9173077c5e92d8015c7cc30aebd12be159185d7d157f12a` | +| [Cost of living measures (TW_1, revised)](https://ifs.org.uk/system/files/2026-10/Cost-of-living-measures-TW_1.pdf) | direct | 222,191 | 15 | `6ed174b22b4d22a34c8bbd450194fb2b4181a9d566b137913bee52ca49a0299b` | +| [Spending on public services: the Achilles heel](https://ifs.org.uk/system/files/2026-10/Spending-on-public-services-the-Achilles-heel-BB.pdf) | direct | 157,838 | 21 | `7403ecc0641b23f5ffc0cb28f67435dbd1b8df6b124874ffae7ae7f75e357aa6` | +| [Tax measures](https://ifs.org.uk/system/files/2026-10/Tax-measures-SA.pdf) | direct | 133,005 | 12 | `6792dc448f40d34f174b6a3c839863b5c969a7334263f5d6b539be62a0428bb8` | +| [Autumn Budget 2026: initial response](https://ifs.org.uk/articles/autumn-budget-2026-initial-response) | browser text | 5,401 | 5 | `962f737e4dfb70416f76a7fc3104f9c4d6b04ef6566a58e0e55fd9db709136ff` | +| [Collection page: opening remarks by Helen Miller](https://ifs.org.uk/collections/autumn-budget-2026) | browser text | 12,403 | 6 | `5991ccec0e0ebb90952937f23400039a8386c047a4b44c9cd4391f535504723c` | +| [Flourish 30734907](https://flo.uri.sh/visualisation/30734907/embed) | Flourish public data | 912,453 | 5 | `99d8b19e36132340356adfcd9206cc26beb37484b3213cbfaa416dd39c9a8f9a` | +| [Flourish 30726471](https://flo.uri.sh/visualisation/30726471/embed) | Flourish public data | 909,632 | 11 | `db33c42dce0a78f12eef4c58497dbb9d2b3f0b3b304018fdf26f2052995f3618` | +| [Flourish 30726788](https://flo.uri.sh/visualisation/30726788/embed) | Flourish public data | 911,264 | 2 | `ce839de3a697429a34906b427d87f40ae9a9adccaf7b522a3b7d6d6bfb3bdf82` | +| [Flourish 30726917](https://flo.uri.sh/visualisation/30726917/embed) | Flourish public data | 909,037 | 11 | `a0a5eb60d2b40f76bbd78389bbb8ef3592ebee447e2746e6becc56c31fd60a5c` | +| [Flourish 30726939](https://flo.uri.sh/visualisation/30726939/embed) | Flourish public data | 909,837 | 1 | `96086a29b440505c6b4010e7844320b5d8179572c2f9d7585389b8a298a7d253` | +| [Flourish 30726047](https://flo.uri.sh/visualisation/30726047/embed) | Flourish public data | 920,130 | 1 | `3da8f416456fc61e99144bf23563bc553eefb8fa751251aa9e3cacc73af10c44` | +| [Flourish 30727239](https://flo.uri.sh/visualisation/30727239/embed) | Flourish public data | 912,450 | 15 | `b4303da3a3371585ab0c000a6f6419c07ae67391fd9c37efcdbc495a0f51ea6b` | + +### HTML pages that refused curl + +The first lane's record of the HTML pages that returned a Cloudflare challenge (HTTP 403) to its curl at 15:43:43 UTC; "Browser text" above and "Unreached" give what has become of each: - https://ifs.org.uk/articles/autumn-budget-2026-initial-response - https://ifs.org.uk/articles/autumn-budget-2026-explained @@ -140,6 +252,41 @@ additional AB2026 items or linked xlsx/csv downloads. No data file was identifie inside the reachable PDFs; this is an access limit, not a finding that IFS posted no data. No Wayback, alternative domain or other bypass was attempted. +## Proposed metrics and units + +The table lists every proposed metric in the 144 rows. + +| Proposed metric | Rows | Unit(s) | Where and what | +|---|--:|---|---| +| `affected_share` | 2 | share | Wernham slide 5: households receiving the £100 payment, 80% of all households in England and 91% of the poorest 30% | +| `average_energy_bill_change` | 1 | gbp | Wernham slide 2: EDF's forecast £470 rise in annual household energy prices | +| `average_household_support` | 4 | gbp | Wernham slide 4: nine-month cost-of-living support per household, £140 and the £24, £36 and £80 labels | +| `business_rates_bill_change` | 1 | gbp | Adam slide 4: the maximum annual bill saving, £7,146 in 2027-28 | +| `cpi_inflation_rate` | 2 | percent | Wernham slide 2: the OBR's CPI inflation forecasts for Q1 2027, 3.8% and March's 1.9% | +| `current_budget_deficit` | 15 | percent, gbp | Stockton slide 7, % of GDP (4); Flourish 30726471, £ (11) | +| `current_budget_deficit_revision_risk` | 1 | share | Stockton slide 8: roughly 1 in 4 revisions | +| `debt_interest_share_of_revenue` | 4 | percent | Stockton slide 7 | +| `debt_owned_by_non_residents_share` | 1 | percent | Stockton slide 7 | +| `gross_financing_requirement` | 4 | percent | Stockton slide 7 | +| `measure_count` | 1 | measures (proposed) | opening remarks: the Budget's 98 measures | +| `primary_deficit` | 4 | percent | Stockton slide 7 | +| `public_sector_net_borrowing` | 4 | percent | Stockton slide 7 | +| `public_sector_net_debt_ex_boe` | 4 | percent | Stockton slide 7 | +| `public_sector_net_debt_ex_boe_change` | 4 | percent | Stockton slide 7 | +| `public_sector_net_financial_liabilities_change` | 4 | percent | Stockton slide 7 | +| `public_sector_net_worth_inverted_change` | 4 | percent | Stockton slide 7 | +| `real_gdp_index` | 11 | index (proposed) | Flourish 30726917: real GDP, 2025-26 = 100 | +| `real_household_disposable_income_index` | 5 | index (proposed) | Flourish 30734907: real household disposable income, start of decade = 100 | +| `real_spending_growth` | 11 | percent | Boileau slides 7 and 9: average annual real-terms budget growth | +| `real_spending_growth_revision` | 1 | percentage_points | Boileau slide 10: the average revision to annual spending growth | +| `reserve_allocated_share` | 4 | percent | Boileau slide 11 (2) and the opening remarks (2): shares of the reserves already allocated | +| `share_of_energy_price_increase_offset` | 1 | share | Wernham slide 7: around 35% | +| `spending_pressure` | 9 | gbp | Boileau slides 3, 4, 7, 9 and 10 (7) and the opening remarks (2) | +| `taxpayer_gain` | 1 | gbp | Adam slide 3: the maximum Founders' Relief gain, £1.4m | +| `ten_year_bond_spread_other_g7` | 1 | percentage_points | Stockton slide 7: 134bp | + +The proposed units are `index` (16 rows, the two index charts) and `measures` (the measure count); no row proposes a baseline. The first lane's rows propose `affected_share` and `cpi_inflation_rate`, which the Metric registry now carries. They keep their proposals, because this extension leaves the 87 deck rows byte for byte as they were; the new rows use the registered `affected_share`. The new rows also use the registered metrics `taxpayer_count` (15), `fiscal_headroom` (4), `decisions_effect_on_borrowing` (2), `real_income_growth` (1) and `revenue_change` (1). + ## Family conventions - Fiscal years use their end year and `conditions.fy`. Calendar years, @@ -191,51 +338,6 @@ forecast/outturn lines, because they have no numeric data labels. Axis ticks, chart origins and dates are not outcome rows. Slides 1–6 were visually checked for hidden printed numbers. No headroom or tax/GDP number is printed there. -## Proposed metrics and units - -The table lists every proposed metric used. Producer-specific meanings and -scopes are in the component sections and each row's statistic/conditions. -Existing AB2025 spellings are reused where they fit. `statutory_tax_rate` is a -single spelling across Adam and Wernham; `real_spending_growth_revision` -distinguishes a historical growth revision (percentage points) from growth. - -| Proposed metric | Rows | Unit(s) | Scope example | -|---|--:|---|---| -| `affected_share` | 2 | share | £100 Household Energy Payment (Jan 2027) | -| `average_energy_bill_change` | 1 | gbp | EDF forecast £470 increase in annual household gas and electricity prices in January | -| `average_household_support` | 4 | gbp | Cost of living support, Oct-June, England: energy payment, removal of VAT on electricity, fuel duty freeze and cut | -| `benefit_payment_amount` | 1 | gbp | £100 one-off household energy payment in England, January 2027 | -| `business_rates_bill_change` | 1 | gbp | maximum annual bill saving | -| `cpi_inflation_rate` | 2 | percent | The OBR forecasts 3.8% CPI inflation in Q1 2027 | -| `current_budget_deficit` | 4 | percent | Current budget deficit (% of GDP) | -| `current_budget_deficit_revision_risk` | 1 | share | Risks from the forecast | -| `debt_interest_share_of_revenue` | 4 | percent | Debt interest costs (% of revenues) | -| `debt_owned_by_non_residents_share` | 1 | percent | Debt owned by non-residents (% of total) | -| `energy_standing_charge` | 1 | gbp | Average standing charge for gas is £108 a year: | -| `energy_standing_charge_share` | 1 | share | 13% of typical gas bill | -| `fare_cap` | 1 | gbp | £2 cap on bus fares | -| `forecast_spending_change` | 1 | gbp | defence financing | -| `fuel_duty_rate_cut` | 1 | pence_per_litre | Temporary extension of 5p fuel duty cut and freeze to cover January to June 2027 inclusive | -| `gross_financing_requirement` | 4 | percent | Gross financing requirement (% of GDP) | -| `primary_deficit` | 4 | percent | Primary deficit (% of GDP) | -| `public_sector_net_borrowing` | 4 | percent | Net Borrowing (% of GDP) | -| `public_sector_net_debt_ex_boe` | 4 | percent | PSND ex BoE (% of GDP) | -| `public_sector_net_debt_ex_boe_change` | 4 | percent | PSND ex BoE (year-on-year change, % of GDP) | -| `public_sector_net_financial_liabilities_change` | 4 | percent | PSNFL (year-on-year change, % of GDP) | -| `public_sector_net_worth_inverted_change` | 4 | percent | PSNW (inverted, year-on-year change, % of GDP) | -| `real_spending_growth` | 11 | percent | average annual real-terms budget growth | -| `real_spending_growth_revision` | 1 | percentage_points | average revision to annual spending growth when budgets must be allocated | -| `reserve_allocated_share` | 2 | percent | share of reserve already allocated | -| `share_of_energy_price_increase_offset` | 1 | share | Three cost-of-living measures: household energy payment, electricity VAT relief and fuel duty freeze/cut | -| `spending_pressure` | 7 | gbp | total departmental spending | -| `statutory_tax_rate` | 4 | percent | Temporary 0% VAT on domestic electricity; previously announced for Oct to April, extended to end of June | -| `tax_threshold_level` | 3 | gbp | maximum qualifying gains | -| `taxpayer_gain` | 1 | gbp | maximum gain per beneficiary | -| `ten_year_bond_spread_other_g7` | 1 | percentage_points | 10-year bond spread (bps, vs. other G7 economies) | - -The only proposed unit is `pence_per_litre`: the printed 5p fuel duty cut's -per-litre amount, following the policy's named basis. No proposed baseline. - ## Wernham cost-of-living deck Tom Wernham component: 21 rows from the revised 17-slide Cost of living measures deck, cover dated 29 October 2026. Source slug `ifs`; own household-price calculations use `arithmetic` (the deck does not name TAXBEN), OBR forecasts use `obr_efo_forecast`, EDF's energy-price forecast uses `edf_energy_price_forecast`, and announced statutory parameters use `administrative_data` with restated attribution and administrative_fact benchmark. No chart bars were read for values. @@ -461,7 +563,7 @@ reopened at its slide and checked for exact value, sign, unit and period: | 16 | 4 | £21,000 | upper rateable-value bound, GBP | 2027-28 | -## Final family verification +## First lane's family verification (29 October 2026, 106 rows) The required command, with its uv cache confined to the ignored family primaries directory, completed successfully: diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/claims_staged.jsonl.gz index 6958be62f23338852da20bd9ea93d117c451566a..61b4dade50590ede8e7bcc59ac66b7989dbd6635 100644 GIT binary patch literal 17171 zcmZU)Q;;q^w65E>ZOyiA+qP}n_Sg1o+qP}ncF(rmeg3s;?Q^qplS+-`smgniWW1F@ z2!I0mp9b=K?YEV*C7E>c=_{t0X~kkDr`&#D+PuS5p;}q_x0br5c%#)T_y9$=Pz;0? zNR_(EZu*q)#mFx2Qo;JywAXZpj|3IM={oJxd zfw>~mUe6hN<=hkf*&zT%*IiGHLSaT%lPLup z%yZ|fL{pSgML-T8de=GYE3j!XSTSEGBLV#irQ0F6j<2wFP`;B!Nx=`doKj`Gq#rvf z3|D5hX~9^-!y{7ii*H2y!3J_%U%BAE%tlJn-5K{4b3;hIfKnpvG@qYPgQhyi4a#7J zKJkt1EZH0GtH+EcTa>WuoqkIe3<_GAQ>%XLZeG6M%yP1`CWO!agIXfOw-zKxSb{W`_>wP((D8eFs#kKmx9FK8%G-TS z3;j%|>|W$;G6c(Q6$0@`DnD3IY2i0G)%^H?GU#bKM7@%&5?MeUjFG{3lv*@}#Z{p< zQ?JAcfeWE}bY^ORuBTA{%17V=OzP|Kae{KD_!iD#Su&|2Ku{8Q_l1uRG_7nGieQ?2 zVTN-~`^pX^xLb_Ln`s-D(fX=-xxJA>9EU>p9rPrmY!NSd_nNh@K{8N0-<*WY= zV*BnYfhU*mAv_!%E|>^T_695t^cMq%_)NrR{t~{^Z_dSAWT;OOWqdSS{6Ph-GjGl% zF1aTBkt#0&x$FekKwc6f;E14HGTKB~4f!(9j40tk;3(|TH8?%?(KoHf|Ct)cEQhq` zV!E$;@)S|9f(jc|XW=LS`(A}<6@3TxHM~kzt~OK zN%yaE!IdXE?{ncB5UV{%Msq0r0}pV}o!Vg;{q}N=aBg?FI-b4?&lC*rX)*5vCUFgw zhYT7zme*KM3MrU(WEIgqOYrnxP1&~<4F~2iiUGVD@XmL^!&D>W5~kA%^19Nsl*IC<7D`D-~;C$d#qMF5tkt@mk>Kk zd&AV!{EmYC@YV&Vmp_nX%SGgnc;+JRL{Zs@eCy_0t*m_6Y0Q1!o0T$mtx>q9brV#@D(Yis-#@oOA4nps5dxZz)1AGH_If0Gy zR{_o6e1hWKDKOpzI0x2+{LlnCxJ|P}z7J92g78m@NPMyF=b2IdE&~}v++Q||Kw2P; zL@2S(BH*86`J?t@%9Oi1X!y0+tGo^!IadjcVb+GU9;t z6_~u^yt%pl76%%PJ=r+_)HIF1x8a<9>2R0U1sO=ZE5N+D*#@wgxZEc9&;2Rrdso_j z@yY?;{nK67@JkEwLlN}NJ#7H^{&pE(jWA*#`{|8);^_a$Vx4Goeg-D}W{+UwPc1a) z{~%m7sB-%CEc#*{v1hWoSwtfwgJe}33lS>wfLI8?G817J^g2K9BL~HQ7jR>q z=K-NHW!zrk@Fd^?85*q}xa;N($?*VV#m@e+?~h6j+9lH@4*BWb8k#ds&foL&X%A4} z8tQUDwq)jfzYba=2ko&_#3bjASON|iEdz!4a25Q+4AqfWlm=m--WDS8{SbHWP;cC~ zrumya`M#gs`IU7^@l~<#gI|dW!UwbVN*;Jkbj)<-1z5RE--Xmf{b+cpxx1rf2XV}h z3lUiLTJc=>q7gLtx=_3G`oXFj`mDloxV0J)NPg?xS&yNJI)~~TeRHAReEee4m$=qo z+T{Xl`{S9b#CW*c{}*AYS$e!;(uX{Tw5ZJ)w)>)h9o6?8(vfu;Xo}hDGXBYhC$nAU!(R%>FPxbpDXs&s6LRO* zNeM&#C?u0Ul?uwex4D#GIk1yMt7yEekrsF$%w~{2)dt%GSL*-~T5AYBhVJ@tO6AV; zptkTD zT*O94T<|^WY9pQRvLLo86>uHdzS7F)u{`yP$-M^eYPm9;v)$;=FMVQUvOHmgpThOT zHt_SV2*lgP_&?sM8aS9(ZqD=IWU@cij(TsN=Zw{g*KyfB{WnT$Mjl|J8ep2+YLm*0 zzdecRYoK|q!dN=KpMGVxoYiA1m8T))O*N7R9TwvQ-9>^UpA5V`f<-lN*& zj+b9hf0SR4@J#z7Hktvu!_~Qp7&$^ofI+I2%M@sfs9urotn0VJVUVBT!6txicCf4< zAIX6sStN}`V|rhk?MVtp1}lUOWQN#VB6Rk?+}+L^Z_n&}0>A?&L(>k~7NpJ)-$*}U zuu+q~HK@LX@y5s@&g2N^98KOsoCf)1i&m^@unyY2luvE!0lYi_^%%W z`D}^*FWgL9Kg2$ST_3osQf?IF!YRmP>+5gq88b#fkOux@%UGbb@2S3 z2R>VX`&93R+c9L`>`h#470ADB@IAScNRc+gO)h!r#jkLLLo%L>M;gYxmkndB=$Bo* zaBD1?xGk2f)MJzOhON-I$Zg)Pzrj0e&P#+>EpdO$4q^Lv3siK-9tkB>q3QiJiU@fE zwq~bH7W}A=Z~j~C4kKkEw7@A}Xh@y;%Pt9=YGZUK2uF$55dfe?9{PqEzo|EXIge=R zA!#l5bV#70b_WyCR|Nysr|f0Lm0(eiY&Ac^US=c7upL+`#8r%f-WGU=8I-S5X*R1b zZ^IRuPwD!}fmOM|P@z9ReB&w%8f=9r=}4kEDk+T_(MMa=pO9=h)H9DK+Xh>?Ly#nD zei~SHx<4xIp{gG3kRr{;K>@z@5o7=f7=NV3i>fLMgjOQo1ER?Mp0~ie0eNbSHMvi) z1*yvMzaCoGWCzVa*hm!!Vf#jxJz8nojZl{~1o>upiT)ctu=yJ@fh~b4=-a|`*8Mvy zUD32fNw^lXk8=zR#NRQbMq4It=>gbrhPH$<$~tZrc6@}xL(L8*b1#A{Mw~_Bil3l% zaU#%Til>j=>A4_e;a|3?LhYV!?8Jo^xD_ zk^nG#lm0t|^(TTi&Z8vz`wb#Kw?!&9vN%p6`_%%s^xvt7iD2c$IP;Gk z+#KZu*%smo`99z?kxt^ExPCIJWKU~BVy<${egZ`D=jbNL>O<>sDzB{uu0J3c*WGnCY zG_y)F(pQ=9Zz59{i|0V^=befeu#p*!+#WU&X6+XBVU2$W@|I|OgVWTs&{5jo5X?D( zCI6R2!>{2;@?E0?I3#+a_jWbYDfE*#$CM%oj}cEr?Mao^CL>? z=);Zu6%dpIZic;vL(HjIQh`j_P_J(W0^fvPbl>GK%Chbi&c7DE*|Ng!Az9i^<8NT< z*3tGm_`#Uf`yrtFueZA_1dmUrwQ>rYHY;k$)Vs8>*{7-!DeYyMv1+ zQx&84!Mmr&T8`HMrf7_C+cI3hvoogZH2%!{F#H);g-R4FGhan8Jq^;=8R0PIdDmCp z?~pA2aQzvgg8=Oq4^elZ`}wfy57gzxL#*|t`tS}QI8kX!54X~;m^4UeF9>dpBaYC_y1}#=fbTx!?SCHhty%$RR>C%5ZCc4x zWv1CMOuAP=Z@3}F5n*q&1wTtxanF<&e?~uUkFua;7f|*cLOlXWGDOxs$vJvpO2*A~ zkN3iPf#hm+eL#V@i#Rf6k3SHoS%}(*Q>FYE!nQpPh5K#L<7Wv11>C2yaa|aOp?k=C zZBY%K#?^b_*og$lZdTC(v@K~?IRG>JpF0dkxoiVUGQ%$dFv!dAGnphW&jez!>kxjK_K^8C3qk$hOfNe(%WT-0`B~ydI z9+B>Wu-OjN030Ccty@7hycr}5!X(>MfT2VbS#cBkU>m|%00_9eeVB%-r1@WER;LwY zjR6^ZkJDi4uKHfi58`3%3);aHkhmy_loel~7Zfk73?%Bv{BS|*r3Zg0>Sho;$qm6P z3|26f&_h}vx2*)x8m*HUt&=%{>P+2Mz^lz&c7<4fEs&fhZaU}1)YPbCz|2d>dy_Hb zl$9((TG4k3Mc8LFmq;zN_e^uu`DxVW>-+Jn+8-Xq$u(i_krvhR=;O>_h`R;pz$!W; z6Q=5tK|RYZJ=i`8Y7Aa78S?XzKvRB|PLDbhWmbSzCJC$9vQns?it5SX5qW%|UXXA! z=1%ONm(I^766mhA4AQGy@+kBqB!bE+K|`;-cO@xPvhkALSynjD2+nC;#P?&QU67bQ zJ(z0bs`BEMU{UG@y7_D=S5#tRB3t5~z$t4iYOwqcAr}#R4!9T`d!)=qlY*Q*faM}& zf%IVaUK+A{`&rXrkkF&#;{H8|6bM><7$vYCf4E!uG1$0d#^^4QaPO8tN;AzcMbmq+ z=uBAP#8E-ZG!)I89){P)N{?*|ha$m;ryfRHML*L%CZ;mEh=FmvB=&{u7#JZ2f}4dL>s8pri-vS7)sjF+;)~2m@gJFh~ABihgf_$Ygy(67in> zy;Vm36d8nA3>eRk2TTFivDA^xuCD()1q0Gqc;%9JjOcTtFdggM3?kl}NDHXtS`XBn zZQF`~yb#Un%C8rif&x8#op=YZrUVt(U!?>?mlGw~QwII_6dZi--|^dS@*CIv(7kq~B14o{*9=KHW05K{Xlt-SH#-SEq=NmeMO(J)@{<)6}E2 zKhwB^jSLZxHdA)@DHZ@d#|XqnWL8fk1gwrG1F-iBPVVw|SZfo4c)gK8NP zMuEYxnIfR0$I-q;^twrz=x$Z_p|(_9s)u3@s02aD9mcgCZmw>`BoS;H9OQv0wO*gH z*ER$qKU-h7p(jNo>UefycSs4@g=XtAUvOndE(1F6W<$T%qW*3a(4{5Op<@`R(|fkj_`)H5LiY!ng_a|`(dOX7p+H7$<1A)rEy4wD(rIVT(B~gv z!P#XkI7f$oTR9K&_!HDX%dShI+0WvZUHgNT!+S53NvO0kh0Utx|4B(jq{j?eX7v!nNbOHicc2hjU(56FrHy&#$V#7!C2n0_nO8j*R36 zv^f!Q$JHgLj0}99jUUG`BkVSdJC@!`^ACnQG$T^H$^}pZ2-iV{EDjnnWvfw)m|Q>x zA@4gV^x9oywi&lVk+|cROwXq*AvPCB+tNKE*w^qBMNI#vp{bQJok@fVG&N7k=8VF9 zRi+!dX_#Tc#({dsm?pMl8LY}GcZx;T4o-qog^yKtGf((?8YCcpDb2F*mt`TM#fN@a zx*6v8Yp7XCu0$IX@>70$N0VQ|f3na@F>32P!?V)j<}&dwK-8{^vk{h2!`8vjz{IwfrMZ zoLb=ls?7x%$YKy#c>xxKhOK-kY!LDft4y$oj@m14y1^OSS6F7jWQj6&$QkVZsK8O_ z9F_6?_D~&QmWyUgR0gb&ExCfEQZ{#^iFzwmVD-sH>-uvk@aqu;^(OI~yYomcSIy6E zaOElNOM(NmAd?~D3w=b)tNH1i`;4;fR(g^uo#e~Z zIYQ#W5`cGkHH6q&`DT=5)7n=uu2)Qh_BawGBhpZXc)Qz#Exv!nhA;FHXY@$K=pA>V z^HR=(11`R=6QWdiyS_7zC>MYf2XswmG4CP%C;s)IugE`rOS87Ojb}eV83vLBBgA}s zBV+zClXrw;0LwKxIXH$Z?~A3&dy2~^3&_{!(We8*>0^pMG+)2^KY!ug*zVa_72Kdl z1J|JP`^D}W8(_>QiB`aF-wKRGgYi6@BZ;<*rWhJ$=nE$U|Ma_~K&TlvOR&MNUiN!95D7G; zJ-i6mhWbcqIesS7;bq-D@}^T$LhBSXOQRs43=U8>MD848gA3DHTt6T!V*XyT677DK z4zlbB<0Tdi#l6w5f3C>ist{}|D)_w5|MX$>2MU@$1iJowL|md}^~)?ivF1X{@1uJ5 zF%!ChI+?Y3sl8E$6m?$O2A`DC;-^a^MDP~_8m0pRO?UgxKMa*n`*OYzWf3!V&XsSX zv-7&E5(GxkMs|$%ogRjrPXw{m1Dh3ZU>A6(L+depA8o)MNze<3vMjt)58TI5HE!V3p7JX9y^`Z2XJ@^+#dyk)kKXclP zbRIW64~a`C>r=A7;oveIKcg8QM2u*njfnikgu|i5xXP+EKP@OxET&nc+IDroX;k#9 zh>H2xORl6@0x;BfZ}Sy&-`GJ`a;{pOVT;Q3&YT+c8ip)1Fjj9hq@@fs{=#CT`!9_g zINLY0?DTA0pRY&3tu)|z6T7zFOXTMVpmznNoFikIPAp;F99qJ)CcVSUC?$sYANxP#IF<7(HlnE;ZZqmN=G}HZS$Czl@befl~kAhZZ^^{Po zGZ;pH_Sd%ukt`zaP2Og$UjzQ^nj{DmcaA;AnIHt8nE6i76}!x%am|V?*jk?NLFnA$ zbql2!6}KZ9Y&wt2MY$8u?H51#yg2m9B3bgWeORrzwt#)TK1Kd+MYtd|Zd!vpNi48< zc5>$}|4RIqY@FE}YJsDb1==B7&rr53&+F#avk6ueS+aJl#*3;;pzi2Z?xf|=Q%mx! z=+KkvCkxI{Qj^2%u-sSGyU@hk_?vT7OL;OpwT$}-PQH9pA1%vH#SnXhGsbb?S!aSr z;_pnM>KxGZVKv{m<{6QET(Y1|G0E##C_7_B2}V_7KpN|o5kU8eVW@G^=pyvXbv$ZG zFyFx>A%(-uA$~T>n=}(Zm$^Fbq|7#pjSSR^H0Y>`9Jx5!1Mxhj;L;(*6`xmAQJ1V2 zhgFehmo0A6I(lU?dSyY7=Ov%tqg$I#`{WYO=iO{c$ghv(Z=x6xFr`vgsy*8<<`@mC zP*Nr%W*10WE}mIFfb_Fdq`au6kexsowh+PXS;S=`a%bM5F3d`esW*7Jwhg+hFrMci zDzTW#cbfm6@{v)^&B7jEYt6E#b7=nqg0PM0CH_(GxG`{YNpn37wSpRVldqh~m zIq6dPrz0=9L76&q{6VZj)I@c=8k~JpS!y4~w^fOo7SN-`F8>fF{IE<1wss=C5{MB9 z$FHW1yDpRX#Jp9evTu&MqqbC7@T;2tD_1dXE|!W_fwHMDR@C-k){ZEW@ArsL=@HoK zmx&H5Q(Z}66y>7>Ut8UplPOgr`JsFLea~FJN@=uFYTg55Z+9nPPiBxUV*RRPWHo`` zsN5aX{RMF}i#o4ANIv*X(+_M&#x%5j0CB&jy9`^R zgm{vqioA;JCaMeau=*X`yA(cO6|5Z+CO7jadRW_dX;$2RBhbFP+4KTgz0qS+Nz^Xf zMLb@P?=?PV=t4a7sW&^uKTy7k_IOS3ACOqVd4Yf(8(s!4_U%TojQHlwJ5IDbQu3d+ z7y2vsH{Gx|d*w2E3@NDS51Bd@$ey+}p}sAytrq?wPOS@=C9OTNaC)(?B?@B|si8HB zrzmE=rTA&&>R~DwHAP>GZq6*uKH-obik&(^BMWgx-9&ih?rHylPZ`w*&+`gmF5Xjw z-IgcPyavZE+(!@x;9*3*7Uw<>@7_-2&(GgKy(?tH8ka6~+a{p)4f7r~YOQafe@4Bl zww@@}Uh_5k#I7JPY;Qh%ul!;KrFTsuu;MmaYuMU*uoG@c_zS$5+QEAN*V)jjBJw$S ze^dQG&PGhS(zRH|yn`|hJuOf)dGe_00c0+mF^1DEphlRPjWK6p<@MiK-o{yq{(W2& zN51;jI&;?uNbV`gTo?&uGwv3!XQWycnP&YL?pq!3K4{5DMms6(uzk0#t1-%V=Xb;T zFg)D;2~mEGUn&1EIv0uBM*O)xnt4S$?}-X-?R_?DC^sPCNE zIy)mXFmXpZ-sNA1u6kTr;5v_}4vA|(IPO*2G$AjU*126XN(@A`aK~Bsz zSUNx{kve@3Ap1Y>S>d5h9lHO5{1%D58K7VkvLHixftOT-Vja?clb{m2WvC2$tW;)w zr!-M{*728#RMtCQXdq-D<|iCdpkQl7TEHz>j$x8(4$<%br@70aMoEApUCbw(1-MDc zU18f1*e-0H4UZngqQ9}Qpq#x=33gt(vj=e`letV4%4_wygXcDA#;QEnu_V;X zqT$R|hT=XymJ}8)1|({4!{YY}%xd)2Sh85$Y0?8fxGJ3_44fR4d)k7y4Zajk9Fj^K z->o<^gv^x>M{|UICd9AErsSvuXlDV`9e!^iQlZB*f}lC6Ba||uA5TR>fbWv6TltIl zK2GlQhxY$6UdOC6tUX|XAo;DnN@PkRc{c5G$&G+8?i$&hQ>k1+-gmc99_4m=q#UzT z%LqW)TG`^b)emNiE8^jx-4znOqJHdkesYhA(?gu(f%);*-Mt+35o+?Yv~2YVH2m8tBzLgsbiAFsq9?m%1p`h72M!L2^a(`luz5#`>p^u^n4QhDuk( zdy!o$RlY{Cu;ZTNhjm2n(8&QdQC^kD8msB4gpH9&O+Px zh5{QX{&Sk>(7;k3g!)|p*y|^OlAuKBGG;SY39s3rR0kyrM8Wp{wg6LJ!fAG2U)bdI7 z)WW2^pcIUVckQZ{mTe}fsWv*SgZLPG?Xr_tM7WX$e8Z_@x^`w$oEWeK5GHpx5D`P& zURPPPcW<3qXRsYhH_f+K&1qNtnX?pNT(_oUc={jt#VbUmG`gLF{<}R&^;0I3Q7hX~ z*&ekU*sQ@cFlLgyjLWS3sV6Kw%`r8ku>yV;z@b2}yx%$3efLOtJMlZgCU>qGwP0?? z4Li!Q7SbNl59@s_u;W$J=K%R|i( zBOT=dbeel0?kk~rYAC4Nd7^h`gx_fwPidkJp;k(XZqdgheNG#A9rNfo2=rkz)#1BB z{L_{uUuHwg#m`NRuE%2*8VS;;2)x&tURhSXc{Fh#Dda_F&NUF`x;l3y7wgL zYW(Bi*#p`cZ(9Z-yaP$`I_L+;B^VgiS7qILE8jDBku7kGU`VS)$8bglD>3;KLEu!d zZc_ldDmx4ZweF?mF5r>O5NZ4B<>ZRxlBpQAFipH#OmIDBB%oAPh_omc_jfIfle4{k z&Y30|BS2gsuC+`;YTdz+GsaD*WM%LfZMp|ieB^sux3Bz--+|__gm>|d@%}F58Jb5% zt0*Lp!(>w<0>Bg5wK)){Id>=~>O@@3zc6A;=%a zy7iSW*lS}X2vY_#-i*m6j0kpHH=F5eI&Vx_Lo3NAsEPHDo@O5|6i-A3p)HN82_zfr5kGFD@a!n!nBBMOVEDF_9zHQtqY zFV#wWH!c)-?|wr}K1Kd>m+>H5tl0~Uj)8da5%JzT`g4~E?E_sqN}O;9n*~;{;zc_Z z&Mlz}Hj0Rvk%gIaaN&a-D&>&mI)P7ANR4!=HF6-S9@dy1+W8A9MJ0rdG9}aLuLB9X z(qY0kN{#2hr*403o%9CuwpU09$xeR8DR|8VOr*-1)$br@URyl=J=h3{jx4zpnJJ>o zOaZ!MT$K=16A!(U?5GY&GKV7=?mCI8TduTn0?j^Lsh>KABZCVAynmtfp?E%NnXDxX zLUo&Kqx;hhHEt>DM;o9b?Vf6KmpF796%Y+4Njv;1GFs%~#eC^URb~40sIgQ@q3i)D&a{4z4TGPzDmclBm z#_P8f&Fbu)UwNYqFihxRVLO*I#=wohj}AxJAFW$Xq>KjGh!-9TghTB?NT}@VX-d+> zP4A6u+H?a0=uP}(umyH&Jks@|uQ8Q^Fm@#SmoZ*Y;qS?nubyHnF=$^ENa2WIKy=6# z-JTWsrgni${2@j2Adg@G5o%=GWg?#WomlD8u5&hAe~r8L`hL~OZ8z~z_qz8k8fe}> z7%sl;_9tk#BGae9c8vP&A|EpEi3>rBe1{r4h_vNlVfSI?;EsAe@sG&rEw#yG_01pX z-PrIxYi<;CX@4FOt4^M}mePggv0CLHm}Vrzck(UkV0^GWt#G7VBCW5(JuO=#`H>o* z$&=mRNu8Km;Sl!S0E-6Y)Z0|6-ZFp1(5*~xL?bFXM|DoPp5nI|v6`uE$_II2VM}x= zHt8v^gEx3_iu2c^o?RMOur{f)cD6D zn=|38wW^_ljW0)t#BN9_c_gxghBGeYbSjzl`#p(C&H{dd6jJbGPTCXOxiZC~V`yOU zA|vMl*C}5(JgHc?sSfBrq%U4g$y1hu>ol&@WPL!88~IB;+M?`7=zofXyM&Xrt4nH>e-bAG_;u*kquQ(_wm>$@_mQ`sYwAzHuJi znQ3>~x5zO_dvWW>NG9xL87M))q@l=MXy^jf-CnKtT!lDz{i zo7G@YB9zHq(R|uQB5UdIT@;&e?`+XTDxCCsePb`arDEDCZ;#(6d>5uOv7dVI$RCwbuVPJiugw;3d9^*+g3zCM-lLFMKC)C

+@SNmORL*l3>?AnTk%(J=^8SQ#)N{FH|BFjuA{MVctP8A`wC zI$%${^)ed0Pp=Rn$h*Z1fv7%y=CRys3U>Z`>*3tI%;Tv?r?hwG85D#|6S(*5K`tngZ}_|ntX z8Z%vYm2^MHTn~85IB(@W9V?G~va~aIr9k2 z(pe#ObcCK3p5&ZHilLVl9AbYyVHz2+bwu)Moq`^ziByN5QnPtKyw6ti+?s2KlNJ>6 zYA%MFP^i?=&2+jo(^Z&IG^Kfgy+&P0L~Fm1>fo8%Woo#UCeR2dgb+8pyf>dP}*MxJ11DcFI~4@_Ok{$gqjZTB~qLZWIS>7VkU{Kk8G zb2zKX-qbqmlo2X7FP4J?Uy;Pj;i){WJ-;Y@Dkcb|v0DIQ$1uK^P;b=i6bvg_zInt$ z^~6ZO%kd!F6i%WqhEJ1i_VHTqRzr)%yEPo^AF%$KRSHTeeKqI0dh)<|4S$F8cm1mT z_2v|hP>sR=Ot@!jpA9hhI4Zh%8!@~)aV)Vk2esqTeLA zZLxL$rz6tlruDzLoQ~|$+Kk7@4&K+}^1)<#k~^59?2PivaWUwuE4j~|iql2}H~Oks zGrvk*bqMAqUOG#;u zWx}|IZxEXEFwEKu2a-Qm>u5Qu_d@OBAl%8U)hvr?JD$p?It1>W1B5QgFP@fP$ zr4e+guw+ccVfqlKB4gkL$BC?b!Mak;fWA-5hAmcAt-hp1mr$cpQ$dWYp-zb%GGlzW zvmv(k8ZOpfCWSzxA}M&VePnoCt5dn@!+EA4--z>WhPchAG*&%j=i2Ll5-K8X0T!g{ z%w_L}2Xw5BZVBNl@r=uy0q)ueXmr)`{Mugeo2V3<=te(VY$_jQoz5JgOPo9{-n8Wz z1AVeAwq&DGL%3-(Wk{8`PLIEiE_ZG8tYxHW^QpYmb2j+7WVna+Y|WPQ>Iu?s%=zkv zU?8OPeQTD!Q^M_q(zq}CSQnw)s?|kIbW@O90<@wIp45)W3iM(EuK4FKj-;O*5npN0 zave*8?DP>%aDU8!7nWoyFb8WwCCMVhA{UM%=>JTY!j)7oCxV|{7R0dxytGPdLp3tK|?rl7YhsFkJz(O>*60{5;o zKY9pdru|U}WhxRoqMV{2Q=CoVMU14@gxpEt&EC4*C{4-I0=eQb)w&Q=CHa!U{?j{~2={4oC-x`kmijG`0} zBys|NMu7iWsy%hOj%z{9%Xg~Y7DJzo^=WRPR8)US?&V%Z>rHxL#($yZpfOtcl=94CigFbl_l2biW>pZ1`pM3>eKQ4HOcnp z+phZ?1_N>R0~ndw{{~s9T%NrZ$z$Do>O(S3IVllK>%FQ5+H?IjvjfQMprgIn=y`e@ijZ#_u1avraWOWoy)U+Q=+&IEEC zROEV)AlE3>R0gwk4?8;^a*NP*qLwlT^~r#(&DwrtyRt*=vkq~vR_(XN&g06(*>J11 zhiohTxqV4A`Lyo5s{Ny0_Y{*-nL}#$0=*86azu%e%C=xR!f0 zprt<9ayX`KUKSwkUv2i%cB+vwfAQXssmIo3?Y?qV*|R?MczI}E?PTB8F#599FdB6; zw)sDmD^u2KE3{qXjjI<;wDf-0Lj<`dVY*fP)eRaeQ5I)II72mlnn#qC@<1e{OO)zW z7ZDzW2p1cxinfO)UdO?E{-e&FJ)a$I`{32AaDr2bM8{k7Xo2($DXzq$gNyHAmuR38 zqADLw#9^NkS`#8WpT>EEbDB+_CuH{@shDE9$qZ9`36%(# zE1vvf^TMik^_os`U=U3RuaOm@Em1O*Q5D*>h7X}LV2j0jeIw77`)FF_e9%QmN~5?)&QsX-cW zEM7KWLJ{AMiptphnWA?|%~$=5irS584-c#G+v$Ct%A+@k5a^>A`-(^}@#t`O>WYS` zgP{EN5yWjhyY;u4B8XF`FTP{{B0_$~nRxl1!u7GhcFS))=%$*!{nufAexNM*@ZhU= zsJ_PFEmtCQR+if?)~>T0TgN-3nJ>j2a9np8z5QI;=iIN&6A3djjb3ic%nwyKqV>Mo zK& zwPB25!!d$+>JG;NGim#q_PBv9mW^PrwB^&FZl{XTjt0ME3rJBVz9fTyLcX!X?==r? z;bsi{RDv>>#c;=0HZg&QnmAkgVjOTiy~E!}2bO5RB z3|>cNNn-Qj@~`JO2!by)5P!$U}c{+Y&V?O+d!S)RG zGfL|rc@SNw<^{^m&VX8eHCo%@?B#w)@-jNy55>M*aGijHQYW8|LMHa}CJ4Q&1&`*e zcE+hsh`%;C*po^Q3*(9t7u%UnmW9ir6;ThNQO?+CtZLN2YGi*t_ioxyyNJ%tgi>^# zKlqv!hcj{5*m4Qk5acs)oH}d>o#~ih?rA>~)GaJE_X)TxPDC(@b_*A(l9oDJ#uxs> zDWjf1qn5l7uPL20B3ij+vpf z$gll#gB;zj_8m7~j)iVaS?YK$|9@B)uGRmSy${G>6SxY4{#mF@q#$ILR>*e$g|J|5$0DhnH& zZ}~lU2=zV>b$(N(f1_YplmBVf^#5z7{F(6nF-r~0Z`#Mg$wG)C-VH8>FkAbti(Eo3 z?o1+XRVw+&xx_;*?ozsyA>qhz4`XvK4lg3-)F)x+<7H#Q?klic(oD2RiZKsLYPyNC zm-w6xDc>+bcH0sL3&~9CYIiUY#ru8l6v&Q|jenNb!gsL3n6ehzxI_spgn(Wz%o^*_ z!OQ7AWGxq;!s@9jfm-*(a4(2B%Vp9d+yk_ zsnZ=`9t{mqcP%y~(tsm_2Q5o9)fT1h1LG4NMvTU=;MP4dMgvCLIfYZMIKj&Q%GrLv zVm;SsH;O7jd72{tI+?}|7}=j91z%ORJXhEuxfbw#V@#dm9Ou9nnGm;#cw?<_eYgnI zKlNK}AU3RUpF2XBe?5T&uspx#|7~_PU1#a7`NQf!(}5`cQBE^l!a!7AmQ~f`tFoI{$5p6f2mV#iG4Iktg9ij8nQFm1e!og&zG|@qZp?%Sb4HLt!j72@o|dGJ1PjJ_PctH1i<4%x8>Lybj#J7UYKeSE^1e zQX<>H;NFs_or6|6Pklc8T~~T?&#*ISF|j6DxWJHwYokZxtI|zoIMC0EU=*`)4UP7} z`4TYlF4oxDKsmDrg%vT;8)`bS5-H247z~!!AS2s7r4di97GZvAcwMN+?pjwGr30Y; zW@a>nG|A_g&^*~2r}IRW*vf1s)vZMkEnh9q9z1i&uV62tR$sWIjwk4r1s$7&;CuOfL{XXS>PnT^Tq*CDeC6wmaYdx8y6PJFkB(-_B5{IUzOwPeztd$Jd+dY}WUl&z`+^ zTSZK?&APcu-hL{~t(o)b`o$81Ya&nf*7L<}KA4d4=HXHE+0U5U9|wN@_HK#R!tJN- zYji*B4e=~m?)+%C$nNVV>oQ|5|M+>KBG;=$?UTFOr(cl_zofqLyFY!s{MXJ2mLAF6 zM(#H=jlxs@ncqqO`|hxRjvccs>))$A>m*~1e`y;0I(6yVmVY5z?p0{T8K-BwKm0b@ z_m{C^g@|WylRD$PniI9o!HjQLXQy31V`rh87XLzPf%7|m&zDJBvs&&<++K1bcEXMa zAC{^3O|oXzkG;HOS)uRaS;tCG`~_}^nj39bQf2idZW7!X*A>gpzL<5eB9&86rjR zXInj4x97{goxQ+=5-=^yJyC0LLGk;fGmcWx6D&{pP2S13Gu!QYPu-=Sb zx$`@A?$doDMeGxE1dH4u9z5*@@}PZn+Wmd4PYaEIJDvBEEvonO5)Zzp@+41c+0x{9 zu3O6m+=`{dH(4{g#~SZUD?0mR*2&T$JA;x3dk;NlPS&=QmzZ_F z^g-<8U20#2)k|e7?4{D~Ykj!){^fz+K@a9Cz3*jyuW;c1kLCaQSHEhjPu$AP000Qr BPayyR literal 9694 zcmV<4B_Y}$iwFP!00002|LuM2lG{jf?tec;MMvmZifvK=$>v2J#~ac}ni+d%R+i*5 zaeg6+K#{}>1UR^Ki?iVfzr*=)!oT$<@|*ZcoS9X)kObKvcxkd(j&+Qh1d3Hzncr8H znU(eTD;~$8J4|9X@ULF6t58HSkA-t}&8~(#5}xY|YZiE}J%_LT#PjeA;oFlb55HKp z$D;_}bKKELgu;)lY2XMCU$+C_aj{)|;P2R)OOK9%P}n?*@fk<>!PND69EAAQk8eNV zrz1DAxo3$xXb(T;Nt{f5YnV7=5nC;z)upFYJJy}+$9ydC$-mlhFccxnp21#U;a`nb z4?h@-U>x$4x55g(; z+&_5QVIHwr2x5>E=La!K`?u_`Z$7eu9x*q9UjcOpp`}q6 zOj(SL!0>|vRKg$A3+eR9)}to-AfQ!{541Ye}Irozfm zRAS^@@Q@?~dWcqFiy1-YOd1n@b;rE~e+fpJ{$+h}eMbVFTX@I{`F$#nfB%QSUkW$$ z8m+$OhKuHgzUBs2umsHw88;Zso87n}mFY=y1B|jFH=sN&h#LltwxPM_8H^9C~P*jdn}Z!$s3WOVa}@Sc0a9B0cQ3cgTc# z>hy3^ogOY~_fR|PAT6FnVBDbMKA7DwHP00S_!Yj*jaTFT$%Yf;q*`)PEtphmVNy>L z{MRX$4keUhey7>GfKURAvLfQ6Jai~Q1O7$E2%a3^>-fM5;J0^m0i8vBpk3Jm)V?SO zQ2qFT2K*=J_B2eC@_B%MR^dfEOP| zSvfnjuYLGc0AHF=X>2$BZe39K!{WP|0#4YguaNT8!?^2JO8fHH0l#q(#{!JXz;9?@ zz6O5n-G5TRuYLL@IsdJT$oX$o&iQX?pS}iu?d5+`z^{G#s~MX+p9DUSg#PMBfg6`~ z7I+8uzQ+EGaQHt7?ALDoQtaz~!Qf4W|Nnqn)PHfw2$Pm%+wshalwm-9it*}VYoKa8>>8=yS2_g|BN_WVC}GSCr$ z6*3s?xHmYDh`^xo)?jd1JA>L00Zj%^J{s`E$v{U5R>+{YYk2r9f&{(F;o;szg@@~h z2s9Zy`S8HgCIcNdD035C!s&p1<Q%cetpZtSFqT38tc!g zoDa^aCB9n2Ks+|uzeV*zphw{jxd+$|LY@c_U$b)k(s6(_uAkPf6&}ta5P-w1>ilV4 zR*i?+p#Y8XPp^W{`D46x@2_Hfd&gD$nfUkHm6!1(5$)mE7=MXHd(I!@wbyR|*Ce4G z|4)-7bWotA{M#i|_BSh&gsCh0Ym(5Bg9|1J9aYd*0P8D&_Z3ypJfWixPm(8eWTKQO zT9}tf!6cWkfZ))oG)vqZ(V%h6~Sx1jiz7+y`Mid5|-H zW6#sKu9+r}GmmZ1oLlLHo(F4ep@BNOvCdO<3DFIsa&*Jc(GATVI>w>7Lq|C(aEEyj zQ4X^*cbGcLp}9lHIy86aXvf2DtHEVNI|kL=Rs$XF(Db2W9-2OM)Z^isqFa}7Q*^8P zP0=TKmLo*QNTlwZ+|>_=#V8(T2T>d`fNTy{%0q$HSM2!$g}g0!RhETIUwJ~ju|OI{ z!dM}TxB0{)VWhH4RQmiVCY6)mcJ}jDILF9 zcPr>c+ze6uVu({b`F&s(*zIx%%>t*(0)5Q_t1O@+5hX0py@dvPqV+w~Ee z1x}U)YJ?dsfCY3+p@ao4;Q+;;^7)CuDIT9VEDP*-!Gvak(`A8{W`PwJ(4m1c7Py4l zpa<2jk=t)ZD|f)A0s1h$dMB(Am*%jJ34$b!lh7|Af;fmlQmFIQsh|YaCqd}`A&2Cu zD4O!>-Lha9iSUl>n4!J0#E)jeCLAnB@!;MEY#Yu2qx=Sl2tg?`J?in{pq${kL8?S> zeqDyyKc|sG(pXS5fs-hiisruoQb6Te7(5UI3Pr_lktKX+FN5qR`!$1pu;2G4_C78U1UvtNx5+1Z4+<~ye;uLM^Eb7)H2%2(h zIJaQRt!&CAU7S8gR8ZLKOlh7g%XpzFWAQGE)=tStrQ%&?k0sm$)!=(15HfvO>sa`4 z)w?NLm2;CwKqVvs+IPTR1imQZA{@!5r+^7=q)7wKB z8;fQD(++z+oPkZ>!P0oO*SpXP2~47Y#9lJfFj(WCn3{3i0?4oGbI$uuESEaFcjHOP z2yQ_h0N~yntyoQ+)b49`FBG65gl&@J3c)22Y)o(pme`GcAUrrj#KNalnT7Lc$$7MB z?AEw<((JIc0&1Fguo7Uqb8waj!MH&EK>LnQ4gdvK4i$N2z;@;9DcYUmv&6*38FTHz zdKNdyZbXmni40(Yg&Ow>m%-yJQVM`zKrHFg3X<`}o9_VOCN9xC2;0)#WHehhMZ87` z&Ij5^$5(*mYbL(hUP7)I7!$yWiMs;1$&=7bL$INJx!|XNK?_0X;I2%di-rRN%R$PV zQNR{}4z@`(kDY|1CTcbvfxDB~N(E#g7#;_w zVLy%zwIO1$$6bfbJnjSBeuN%G`Q|6nQP_h8aki?_Fq^@wvL>c%V!@VcV(*$c*dBpj z!}17MR5DQ)JNSV#h<&1zWcDj8dfo(KC`}`BG4Z!$n8qGJI@^%u2h&=+J z9X$STKC)%6e+gy-E{Czy_kT4BQugkz!JfGO$dfDB0_P~K1+D{Z~)`= zr1J&UPk`v`7&qvgy`hma)xWiHIRUEQ16a;nE9*JvMtmd(#~K%PS2$*`Gi7qB%{v2bv<8D65blQ{BKdpn9qm@6 z^dt8CGa$roBJ?MGdQ1DB*QB87G!egXpL=uA?H4t$5P!H4j2iNX8|e=>KL2K!t4N4^ zQaqUvv39fnjQ#N8=Yj%)Wv7~68z2{++sfmezPSPIHpZZq?=azkrG~hznis%LwF-;c-Ee%T8Ew zj22GUQIAS$@T>=(sF#^4e*f0#8FfttA6K@yci0uqDAg?bu><1ocsMI!F9@U}_LA_9 z=A7Jz%hAW3qYqtTQ$UVM^lH(|76Int=~qyoCEK=Ai(#22`C`LF-aOff*W=Z|kb zEc-htvkX0t<8dg)h^sgXu$m@&4|Bu*&+E@j!Xg@lmpGv1IKCNrjG)^9$LM2AKzW>0 zc8%Adp&pBp*(?YFi=%+4k+R4RW&r9mx5`&Uwl+myrk4T$>PW=QMw@Xj@DGgJ%U>J} zpW7OrYvJ=>kyk+#ZPi^+KS1Vq5(-dO9!)5*lHc<=Snv05H$#6FdtsB0PeV^YBXQ-l z{9cak(gHzY=ctp{3{8-s2{Jp8@xKu?^@osQ);~lTkd0uNHJA!Y#wifP)eeZi(Mh^r zh`%w?|7%vqdX)4DWBogS?7j)b(x$x{50CYw@ z@;qhVc3}oo7&2W_KS&{C4oe8=R`%1;0r;a{=G;VZ2PgEF__9HKnPdDy&YSy(aA+t^ z_7<~oly+b~n?em1a{o(I7$z?LQz*PuWvz=4zv#O^^uFhj_XY*4oiNz|HE zE|^ume`}TGvR%DjgXb(Y0`?rxgf-B_Nnc)6M*3T}OXFw74y{gQa%gq*KIKv2zObC3 ztqJa$DXM72f%Uhv6!>oyg+IC;Qkf!1x=vz$&pynzWjcNWr%S zn&Y`+7YPAj3+V;6tcH=g05msfZYbl1zEQ#rQgeRFwDX@bxRAR}QKdyNSBc6V1&U5F26iI=11g2Uu^Xst5B@^hD2 zo_;ZwoyG?8Lm#j5!ZHOo2K*9I%Cyi|7%Y>H_*9+Vp^bAP${k@osUv;FFL6)#64wkb zQDNA~9TR|rVkUU3s^Poa^p&G0N`@pLLTXngn<}uUL_c|oJ4j-brN(WI+i3|b^O?#k*QJyU&%9m#U@9HSp?4-#S&{-!Ix|U^12cuC0Mu=%V%DHg7 z#HEJ`VL6Ej7W#9{EyC#tB96qLS9tc_4qYqPzX&1cc-jK~Fl6eUx5e{sKr zXBdDUkTRGZga8?{Kzia($LU3|4fzS~L9!#zQIq|fPp&5q4@vNlkyaPE>cGU5)pcA* zK~E>&l9PX=QZ1UVYM*nJNwZlG?_El(#Vn`X6ksT86KL&)U(6cXls{Xv8;mxnSQ{nM zF2$-}R!wTElQ5fA6;SXbd$1AHqCzZ0f@Xn*YdWS%*sEQ&zyM6nsj*igB@ZCE^>H>n zib3zBn%y#YI`iZy06K3?MSPY@e=Y7q3b2~4nvs0QvK-3-h|1r%T12)|UV^u+^leMN z4HCM87hrQjQ;AXQDo5iKVq%CxZN_SDV_jlSlbTAb5e-Vj~0 z`$>T7EZPxy0A(j-rd1O`_TwJ2sIz6M8kp`H8uQN#^F7!+ z9Sib5JPg$3a>mt93G`E_qA&*-hl+Dr(SUy8fR4H#9ikn8cG^D-1-lPI=>ce79FAAb zLhW2OjxU8P$_hF-n21K!xkc!rVrXKIG_aqnrWNW)b9AJDY2R>FJ#T55HPkf8BX}Rw zmLMyoA!kMw+g!6fz_0L#9-8VMSQaI-2FL}#GBin*HVIAkb1YqLFw99hHy&{qi`=Q} zAr(!71U&`!Ktz)?0>jG={fVy}_Tyw%nu#u)i9W22xezOn6@LF_woy%$T4x9Qp7?fS z`5axl`Z2oplPh0Y!NG%_M(YLvym)W%1Gie`)necu#ac0r8^MEwo|4f>9pV2Y z6?ZF^v)|_33Y3D)Svim1$CIn=>TfPYjo!Wg>xP0&X?!J<%9M{mD-!@X^O>xgPQXZ> zp;TPB*PuGaA<*nKmAnrUtXu}4kSTIDb^Qd)5yX#&!eyaPz~sNlrWVUQrHFx?20ZaZ zs)m?4eYR#zyl0uF?RH^d=fD5s|K*5)fAk)Ro>#fV>SJ|d#}ne7ya0$gVGghp#vyLJ zsv1Z?VjR7cBNp!`iGx`qnSqpkPh7kN38aJ%2tVRVozdn&o_Rr6Sv6V}TYsP8T1Fbx zC@8%T(P*^IHPiwC0Uu3^cX2_-V^!~{n5?y#W4I$_Qg|sF?|L#=0s9}{d?>FAbxR=2 z%E&wa4WX@G zrdvwLm7s#$vZ9VGun%w&_yUZ5v>bC(;k%r2a#Ms?yg44pqYydP*@qXls`$}1*Mz^7 z$StHR6!X=dOGB~eAKt!aGxWf<*W3uw>lkfe+eWuz4qNTf&1hsB_OR6-b$gxmjWcx2 zVc!`6+P6CWUS~Aoq6hyQ^=?EP%hh-51OBH};XuT4&;k6?Mms&E0#@2jJNf`$6#up|8IZ(z+&9&(b}ff>^(ae5nx#hU+p3{g1%xc^?c42hsJxf z*IX54p$k-Ibz&kf_@@Ns;`X?Np)e$kH=IONrhhthiT7TnAex7eJwiliuztbT$Aa!o z0lGUHx{n9l?_lu*2!tyR(7x~AOf%E;X%ObtvM)e?o$L3CLDq!w&MWq|tf@UBxoOsc z@?4gWi{<-+U+Dm`GQdEy?loF{xueS7TsqNA9+_B#(`d`JJnXAL0RplJXd|95*#SCI zfL9m6W+SuBkK)c1_qUl^eSxXOr;vmcZ>{)>(h_CL(~li(P)C^*j{3bVsDc;npWXLf zK;Ii| zScHxtjy|nJIDd@Fmn;}-E?LmbKu4sdTO(Ju=y05{asN=_7vpz<>lEq2PV(C=56CX{ zHzKLriga^JqNh8})x~+l2lBqeD8GU+qY(1(qt`T>R$308!v;gs#iRL#1_j?xx#i#& zrPf}PK0Zr}-Q7!c1O{2!X$a=ud&|!c#w}6tpM2#YjE-6)P^K z$xd?gS2=GmWev zfLuJhms4pcp1QhSy#y=srI+MYoCz)u3b;JbxU563t4KZ2q1TJ|so+EC5NF9q*DDNX{^KiTXk;%-m0A6T+lm*+!)ORIydGdNkiwx=-ikit@&@| zr8R1tRE={+1b|lvp*t#bMCrx72KD2oxr{?0{Igx^>tvS`M(VTm(`d}rn0?Baedhj| za{}$niB9_(KP7f25r=LB-lZ z>HUGm`XeN~pSKf7r@ia6_Y(qk4cQv9kDkMR-UzNU*-sh6>o{sOnrk$_2sCds&HbbK znLFrFC{^K{z=2 z`eS*=O{0L=rlBGGnEvrrUKM^%)q_7|mb_|XHKE%yUa_A5y6sRn)H2gvhF+RHD(% z8oP!y#!^Y(0(>@5IxO<}ESd!ILLqs&Ed=dyp=!W2fDHbsT5aH6ilmp(E;j{kUZPv+ zv}ovmD$xHrS7x)IN{mzq*IwC0Z{=pV3q#=g<_A!28`t(>lo^(^sz+go2d^4jv9?;5 zNv%lwQrG*FWl$|q!u*<=^(wgag8#Ucd~puxy59mzEACL0;+5y+cNibKTSCy7DG7yeHc{eKG*&!H=k=c&TM zD%L385XJU*LlpduONR{~4d1ui1sJt<0mi0XK;64#xifWrK#!1|$kvG4aS6Pu^)LLr z&V}CUD?7|&lHYEWza4E)3t7Vt&;p+~;s92=^J1nVU5WZIkkftig+>1I_CbCRYZ=ud zJc_sC%@4UR_?TV|JI2B>KHUP6|9mRYCi^AvVpm?{k$=qYL770Amta1h)!V|3!G(>E zl31Y#{PoiH;aS*V%j%G=!McqAjm64DXpxlnO#)_-1WZi=N8AErmyvbk)#3#cRg-_c zT%E4f$3F(UB-*U}NNqweU(p(>*vy{Y_noTf-zl$t@aUjFXta%kfPUKWX3)3!%#9as za@C0c=!oAoG~!=4;dw>0#+waONM;e@fibyM{y)y-b9ehc3X7;J0} z#f0CvASLeA@Zp$^c54l8ZwBPKD?F+fgC_9qt346FCtz8tAbZ7!_^XyR8-jb-7iIHXikAwzGCpUpL3c7mJR~cyf?;T7cCa3UTl;pi^fE6?WgiO&{e-X=1FnIufcFFee-7eV!v<>ja%`2rXqk6q4N$O z0PiQ$&?cB4^H~c1CyVgE2F$_Dgm1?Ml%^75h_m!YBwgw-r-<+ARo(38m`!2#IuvwB z6%3c&>cH{=6jHvTkdZlf#t;|n*kOV-yail*h5ej-E8id7%Ef(zvx2@-=;-| zLLfVd|6vN*L-VIaFJ^yhdLh~C#eD^^zlTj5POcN&`{XdV$9TSFwq~})ea9NjxpiaA z?2{#c$A!|rz`JZLd;x3yh-LDNWHC#;VjLfEeT?N!6<=Qxf^I7x37lp*aMls_!+2T0 z)f0nBvj$9>PZTB#UeS#{4SWvfYED=eL0=4qkfKt*S&in0^s4;y z%f}y>BZdevo+s?={7)gb4`1+HBwq;{xLH~BvP7&$FZHk4pKf|LQ~Kji9qn{EE?c}N z+!itVHcfVR}Xd z6C@Qb=qQMBGy+6Pq$hbrxnWIdE0Dec9TvFj8(%pLprg%hKZ?OpSx_MB8g zY;-%P0*b6r$fx2yfRW=cL*I?0sI<>y1x!B*QVw@Cmxs2~vu6<~qWD7%JD5x*0-+cBE$a!{@*_=j@`rRhG@4=}H~I%c(MTNxa-6x>As5SI`Uu+SK#rw1oTc)`#Nx|@%}ugk z&vA+~d`x{oM2Y2|;>~ic;uNmZ!jpx+gPcRJUzH>X-9OZE4RsnKpwj9SQAm4Ks)DZi zGzurJ*OIQnsruPs0S;QK;W$JnuOFBR8>Y>cOTF}jHd+SdcBw-bHvD3|6I$Y?M}U@4 zLz|DKEFUmeOG551^oH`!vXrwiq9R`!4GaPnl_)Vrz4PXjx=gyMhA3oXw=0E83UW$M zpK}IN?@r-SE^ApW%HNSj+UnLRO|su5>&j|d_=313ME4>5cy=lKCE;z?FktM#3{O)_ z1wo}e2$*g$19c9FQ~*+GDr-)IMCv#~cJpy%1cfKh3q6;>^5KOd>I5yIR4U-gVZ15- z>Q0lXd=O_cpWAs2h%8++PK%4iZpotYmtrS)q52X5m7i7`Y0~+u3?=EQB2217cqtZQ zYg?VBUQ<>nm7SFu*n#nF2g14s>lrP_dpg_Z z40NV?&A9EFF|++(mUtPuZDz(A@Vv{SRH39Ldz_)wHfvMvhU(-*THJ{)^IBPJln5>_ zny2!_p6c>5m5hdGcJ(|J#`kYG06_plFOGxF4^6_+*P%V~N;U~RXA9U&e`W2eR)IHq zEmc0?#_*MDPNo4{(J&0HaN=2c*WjBp%KreYR$X)U%h#MsUBLkqT(f&bxXv|OSba;_ zO(*=Kl9-@<62bRxy=$}6T@u9-rnyJ g*3U{EaO>>-1TcuG`q^I;D4^T_0H-{TgX01M03;gAeE element as ifs-initial-response-main.txt. This line hashes that text, and rows cite it. An in-page fetch of the HTML gave 257,037 bytes, sha256 909aadc20622e80e3adf20cb2f8002efec0b7279ebff70b658d55e2ac6ac6455. ifs.org.uk returns Cloudflare 403 to curl, and the HTML embeds per-request tokens: two fetches of the same length gave different hashes. The page prints publication on 28 October 2026.", "publisher": "Institute for Fiscal Studies", "retrieved": "2026-10-29", "sha256": "962f737e4dfb70416f76a7fc3104f9c4d6b04ef6566a58e0e55fd9db709136ff", "title": "Autumn Budget 2026: initial response", "url": "https://ifs.org.uk/articles/autumn-budget-2026-initial-response"} +{"access": "browser text", "bytes": 12403, "date": "2026-10-28", "doc_type": "web page text", "note": "The harvest read https://ifs.org.uk/collections/autumn-budget-2026 in the Claude desktop app's built-in browser on 2026-10-29 between 16:56Z and 16:58Z, as an ordinary page visit with no challenge solved, and saved the text of the page's

element, which prints Helen Miller's opening remarks at the 29 October briefing, as ifs-collection-opening-remarks-main.txt. This line hashes that text, and rows cite it. An in-page fetch of the HTML gave 237,388 bytes, sha256 0a7a5c7c439d2e46703cea8fc5040e0b9f6a11451bf604439ee42198ce89a291. ifs.org.uk returns Cloudflare 403 to curl, and the HTML embeds per-request tokens, so rows cite the text, not the HTML. The page prints publication on 28 October 2026.", "publisher": "Institute for Fiscal Studies", "retrieved": "2026-10-29", "sha256": "5991ccec0e0ebb90952937f23400039a8386c047a4b44c9cd4391f535504723c", "title": "Autumn Budget 2026 (IFS collection page: opening remarks by Helen Miller)", "url": "https://ifs.org.uk/collections/autumn-budget-2026"} +{"access": "flourish public data", "bytes": 912453, "date": "2026-10-29", "doc_type": "flourish_json", "note": "IFS data item https://ifs.org.uk/data-items/growth-real-household-disposable-income-over-decade-1950s-2020s, which this line does not hash: ifs.org.uk refuses curl. Its chart is Flourish visualisation 30734907. The public page https://public.flourish.studio/visualisation/30734907/ (HTTP 200, 15,106 bytes, sha256 3bdcec5cc59a3104c668a9a92470cc8bc6fc9e30f4498337ccf730122c7c6809) names the embed https://flo.uri.sh/visualisation/30734907/embed. This line hashes the embed, an HTML page whose inline _Flourish_data, _Flourish_data_column_names and _Flourish_settings carry the chart's data and settings; the rows read them with a JSON decoder, never off an image. curl with a Safari User-Agent fetched it at 2026-11-03T21:00:23Z (HTTP 200); the bytes match the fetches of 2026-10-29T18:20Z and of 2026-11-03 between 18:05Z and 18:31Z. Flourish version 179; data last updated 2026-10-29T11:55Z. Lead sentence (data-items.json): The 2020s are set to be the worst decade for real income growth since 1950.", "publisher": "Institute for Fiscal Studies", "retrieved": "2026-11-03", "sha256": "99d8b19e36132340356adfcd9206cc26beb37484b3213cbfaa416dd39c9a8f9a", "title": "Growth in real household disposable income over the decade, 1950s to 2020s", "url": "https://ifs.org.uk/data-items/growth-real-household-disposable-income-over-decade-1950s-2020s"} +{"access": "flourish public data", "bytes": 909632, "date": "2026-10-28", "doc_type": "flourish_json", "note": "IFS data item https://ifs.org.uk/data-items/current-budget-deficit-march-2026-and-october-2026-forecasts, which this line does not hash: ifs.org.uk refuses curl. Its chart is Flourish visualisation 30726471. The public page https://public.flourish.studio/visualisation/30726471/ (HTTP 200, 14,818 bytes, sha256 affe39197809c2177e28148b5a04e8c2335f66990462dd4559aa892a54d84aff) names the embed https://flo.uri.sh/visualisation/30726471/embed. This line hashes the embed, an HTML page whose inline _Flourish_data, _Flourish_data_column_names and _Flourish_settings carry the chart's data and settings; the rows read them with a JSON decoder, never off an image. curl with a Safari User-Agent fetched it at 2026-11-03T21:00:23Z (HTTP 200); the bytes match the fetches of 2026-10-29T18:20Z and of 2026-11-03 between 18:05Z and 18:31Z. Flourish version 105; data last updated 2026-10-28T18:25Z. Lead sentence (data-items.json): The government is still on course to run a current budget surplus by 2029-30, but smaller, and with more of the fall in borrowing in later years.", "publisher": "Institute for Fiscal Studies", "retrieved": "2026-11-03", "sha256": "db33c42dce0a78f12eef4c58497dbb9d2b3f0b3b304018fdf26f2052995f3618", "title": "Current budget deficit, March 2026 and October 2026 forecasts", "url": "https://ifs.org.uk/data-items/current-budget-deficit-march-2026-and-october-2026-forecasts"} +{"access": "flourish public data", "bytes": 911264, "date": "2026-10-28", "doc_type": "flourish_json", "note": "IFS data item https://ifs.org.uk/data-items/headroom-share-gdp, which this line does not hash: ifs.org.uk refuses curl. Its chart is Flourish visualisation 30726788. The public page https://public.flourish.studio/visualisation/30726788/ (HTTP 200, 14,816 bytes, sha256 97b32b756569612128c96cd708cda8da9f6667ac8909229af53e49fec529715e) names the embed https://flo.uri.sh/visualisation/30726788/embed. This line hashes the embed, an HTML page whose inline _Flourish_data, _Flourish_data_column_names and _Flourish_settings carry the chart's data and settings; the rows read them with a JSON decoder, never off an image. curl with a Safari User-Agent fetched it at 2026-11-03T21:00:24Z (HTTP 200); the bytes match the fetches of 2026-10-29T18:20Z and of 2026-11-03 between 18:05Z and 18:31Z. Flourish version 78; data last updated 2026-10-28T18:40Z. Lead sentence (data-items.json): Faced with a worse forecast, the Chancellor has chosen to allow borrowing to increase, which leaves him less scope to absorb bad news in future.", "publisher": "Institute for Fiscal Studies", "retrieved": "2026-11-03", "sha256": "ce839de3a697429a34906b427d87f40ae9a9adccaf7b522a3b7d6d6bfb3bdf82", "title": "Headroom as share of GDP", "url": "https://ifs.org.uk/data-items/headroom-share-gdp"} +{"access": "flourish public data", "bytes": 909037, "date": "2026-10-28", "doc_type": "flourish_json", "note": "IFS data item https://ifs.org.uk/data-items/real-gdp-growth-over-forecast, which this line does not hash: ifs.org.uk refuses curl. Its chart is Flourish visualisation 30726917. The public page https://public.flourish.studio/visualisation/30726917/ (HTTP 200, 14,860 bytes, sha256 b76d9bd5810ca9c03e95b979681813deb73747c61e7dadae1b3108131fe7ae3b) names the embed https://flo.uri.sh/visualisation/30726917/embed. This line hashes the embed, an HTML page whose inline _Flourish_data, _Flourish_data_column_names and _Flourish_settings carry the chart's data and settings; the rows read them with a JSON decoder, never off an image. curl with a Safari User-Agent fetched it at 2026-11-03T21:00:24Z (HTTP 200); the bytes match the fetches of 2026-10-29T18:20Z and of 2026-11-03 between 18:05Z and 18:31Z. Flourish version 34; data last updated 2026-10-28T18:45Z. Lead sentence (data-items.json): OBR expects growth to slow significantly in 2027, but by 2030-31 they expect the level of GDP to fully catch up with their expectations from March.", "publisher": "Institute for Fiscal Studies", "retrieved": "2026-11-03", "sha256": "a0a5eb60d2b40f76bbd78389bbb8ef3592ebee447e2746e6becc56c31fd60a5c", "title": "Real GDP growth over the forecast", "url": "https://ifs.org.uk/data-items/real-gdp-growth-over-forecast"} +{"access": "flourish public data", "bytes": 909837, "date": "2026-10-28", "doc_type": "flourish_json", "note": "IFS data item https://ifs.org.uk/data-items/average-annual-growth-real-household-disposable-income-parliament, which this line does not hash: ifs.org.uk refuses curl. Its chart is Flourish visualisation 30726939. The public page https://public.flourish.studio/visualisation/30726939/ (HTTP 200, 15,104 bytes, sha256 33e963b9c5d403b849097d33833667c2681f8317967d4de6ad6e93f7e2708b39) names the embed https://flo.uri.sh/visualisation/30726939/embed. This line hashes the embed, an HTML page whose inline _Flourish_data, _Flourish_data_column_names and _Flourish_settings carry the chart's data and settings; the rows read them with a JSON decoder, never off an image. curl with a Safari User-Agent fetched it at 2026-11-03T21:00:25Z (HTTP 200); the bytes match the fetches of 2026-10-29T18:20Z and of 2026-11-03 between 18:05Z and 18:31Z. Flourish version 37; data last updated 2026-10-28T18:48Z. Lead sentence (data-items.json): OBR forecasts household disposable incomes to grow by 0.4% per year during this parliament, making it the second worst of any parliament since 1979.", "publisher": "Institute for Fiscal Studies", "retrieved": "2026-11-03", "sha256": "96086a29b440505c6b4010e7844320b5d8179572c2f9d7585389b8a298a7d253", "title": "Average annual growth in real household disposable income, by parliament", "url": "https://ifs.org.uk/data-items/average-annual-growth-real-household-disposable-income-parliament"} +{"access": "flourish public data", "bytes": 920130, "date": "2026-10-28", "doc_type": "flourish_json", "note": "IFS data item https://ifs.org.uk/data-items/long-term-impact-announced-tax-changes-fiscal-event, which this line does not hash: ifs.org.uk refuses curl. Its chart is Flourish visualisation 30726047. The public page https://public.flourish.studio/visualisation/30726047/ (HTTP 200, 15,014 bytes, sha256 7830f65c93c4ce3a1d709188047b119f345b0eb9bcfd0fdfab592df3041b00bb) names the embed https://flo.uri.sh/visualisation/30726047/embed. This line hashes the embed, an HTML page whose inline _Flourish_data, _Flourish_data_column_names and _Flourish_settings carry the chart's data and settings; the rows read them with a JSON decoder, never off an image. curl with a Safari User-Agent fetched it at 2026-11-03T21:00:25Z (HTTP 200); the bytes match the fetches of 2026-10-29T18:20Z and of 2026-11-03 between 18:05Z and 18:31Z. Flourish version 187; data last updated 2026-10-28T18:22Z. Lead sentence (data-items.json): Today's Budget will increase the overall tax take by around 0.2% of GDP: much smaller increases than the previous two Budgets of this parliament.", "publisher": "Institute for Fiscal Studies", "retrieved": "2026-11-03", "sha256": "3da8f416456fc61e99144bf23563bc553eefb8fa751251aa9e3cacc73af10c44", "title": "Long-term impact of announced tax changes by fiscal event", "url": "https://ifs.org.uk/data-items/long-term-impact-announced-tax-changes-fiscal-event"} +{"access": "flourish public data", "bytes": 912450, "date": "2026-10-28", "doc_type": "flourish_json", "note": "IFS data item https://ifs.org.uk/data-items/number-individuals-aged-16-paying-higher-or-additional-rate-income-tax-0, which this line does not hash: ifs.org.uk refuses curl. Its chart is Flourish visualisation 30727239. The public page https://public.flourish.studio/visualisation/30727239/ (HTTP 200, 15,106 bytes, sha256 4954c827579c86800908939fda087afb48d1fac758c282917259c938918e1f90) names the embed https://flo.uri.sh/visualisation/30727239/embed. This line hashes the embed, an HTML page whose inline _Flourish_data, _Flourish_data_column_names and _Flourish_settings carry the chart's data and settings; the rows read them with a JSON decoder, never off an image. curl with a Safari User-Agent fetched it at 2026-11-03T21:00:25Z (HTTP 200); the bytes match the fetches of 2026-10-29T18:20Z and of 2026-11-03 between 18:05Z and 18:31Z. Flourish version 69; data last updated 2026-10-28T19:22Z. Lead sentence (data-items.json): The OBR now forecasts 10.8 million people will pay the higher or additional rate of income tax by March 2031.", "publisher": "Institute for Fiscal Studies", "retrieved": "2026-11-03", "sha256": "b4303da3a3371585ab0c000a6f6419c07ae67391fd9c37efcdbc495a0f51ea6b", "title": "Number of individuals aged 16+ paying higher or additional rate income tax", "url": "https://ifs.org.uk/data-items/number-individuals-aged-16-paying-higher-or-additional-rate-income-tax-0"} From f4c7c94a03d4927ac39bd86d7f6146442d39b6a6 Mon Sep 17 00:00:00 2001 From: Max Ghenis Date: Tue, 3 Nov 2026 18:13:43 -0500 Subject: [PATCH 15/52] Answer the IFS review of d5e323dc: findings 1, 2 and 4 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit - I moved IFS's £8 billion and £4 billion of new giveaways (rows 90 and 91) from the OBR's key for all Government decisions since March, and its "direct effects" component, to a proposed key in IFS's own terms, ab2026__package_ifs_new_giveaways. The page names no OBR line and doesn't say the giveaways cover every decision. We infer the OBR attribution from the rounded match to EFO Table 3.1 (7.6 and 3.5), so both rows now have medium confidence. - Neither the opening remarks nor Boileau slide 11 names a source for the 25% and 40% reserve shares, so rows 74, 75, 96 and 97 now infer the restated attribution and have medium confidence. - In NOTES.md, I corrected two sentences. One said no guide position holds an estimate, but the headroom chart labels its reference line 0.9% of GDP. The other said we reached everything the addendum names, but we couldn't fetch the seven data-item pages or the briefing event page. I changed no value, sign or year. Co-Authored-By: Claude Opus 5.5 --- .../harvest-uk-ab2026-2026-10-29/COUNTS.json | 2 +- .../PROPOSED_KEYS.md | 5 +++-- .../uk_ifs_ab2026/NOTES.md | 13 +++++++------ .../uk_ifs_ab2026/claims_staged.jsonl.gz | Bin 17171 -> 17261 bytes 4 files changed, 11 insertions(+), 9 deletions(-) diff --git a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json index 87be47b88..eccbe6eae 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json +++ b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json @@ -43,7 +43,7 @@ "uk_ifs_ab2026": { "rows": 144, "primaries": 14, - "claims_sha256": "40331bc6beb36fc78d2e10731730c4c8a15942d4419bd840d0cebc95988e6125", + "claims_sha256": "b7b612261417ffaa580a43d938bba218d1b1894adb34d81b977e33930c1b2061", "sources": [ "ifs" ] diff --git a/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md b/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md index 6621ddf2c..8552c3b3e 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md +++ b/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md @@ -4,7 +4,7 @@ Generated by `pipeline/ab2026_proposed_keys.py` from the staged families; don't A row whose scope matches no registered measure has `measure_key: null` and names one of these keys in `conditions.proposed_measure_key` (README rule 2). A key is a measure with no Table 6.1 line (`ab2026__`), a package (`ab2026__package_`) or an option (`ab2026_option__`). Each family's NOTES.md gives the key's scope as Table 6.1 lines or the producer's own words. Tranche 2 registers each key in `data/uk/ab2026_measures.json`, or folds it into a registered key, before the ingest. -84 proposed keys (15 measures, 56 packages, 13 options) on 958 rows. +85 proposed keys (15 measures, 57 packages, 13 options) on 958 rows. | proposed key | kind | rows | families | producers | |---|---|--:|---|---| @@ -29,12 +29,13 @@ A row whose scope matches no registered measure has `measure_key: null` and name | `ab2026__package_dwp_fraud_and_error_measures` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_fuel_duty_and_electricity_vat` | package | 11 | uk_obr_tables | obr_efo | | `ab2026__package_fuel_duty_freezes_since_2010_11` | package | 1 | uk_obr_tables | obr_efo | -| `ab2026__package_government_decisions_since_march_2026` | package | 376 | uk_hmt_redbook, uk_ifs_ab2026, uk_obr_tables, uk_rf_ab2026 | hm_treasury, ifs, obr_efo, resolution_foundation | +| `ab2026__package_government_decisions_since_march_2026` | package | 374 | uk_hmt_redbook, uk_obr_tables, uk_rf_ab2026 | hm_treasury, obr_efo, resolution_foundation | | `ab2026__package_hmrc_tax_gap_and_online_marketplace` | package | 7 | uk_hmt_redbook | hm_treasury | | `ab2026__package_hmt_distributional_budget_2026` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_hmt_distributional_since_ab2024` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_household_energy_payment_and_electricity_vat` | package | 5 | uk_rf_ab2026 | resolution_foundation | | `ab2026__package_ifs_announced_tax_changes` | package | 1 | uk_ifs_ab2026 | ifs | +| `ab2026__package_ifs_new_giveaways` | package | 2 | uk_ifs_ab2026 | ifs | | `ab2026__package_national_wealth_fund_and_british_business_bank` | package | 4 | uk_obr_tables | obr_efo | | `ab2026__package_non_scorecard_decisions` | package | 12 | uk_obr_tables | obr_efo | | `ab2026__package_other_ame_spending_decisions` | package | 6 | uk_obr_tables | obr_efo | diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/NOTES.md index a6987febc..f745d9f14 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/NOTES.md @@ -48,7 +48,7 @@ Four proposed keys name seven rows. README rule 2 sets their spelling, and each - `ab2026__package_property_and_savings_income_rates` (1 row): exactly Table 6.1 lines 61 and 62, Adam slide 2's "Increased income tax on savings & property income", £1.1bn in 2031–32 (lines 61 and 62 print 455 + 590 = 1,045). - `ab2026__gas_standing_charge_moved_to_unit_rates` (3 rows): no Table 6.1 line. Wernham slide 8's "Shifting gas standing charge onto unit price": "This charge will be scrapped, but per-unit price will be increased to absorb this". The spelling is the RF family's. -- `ab2026__package_government_decisions_since_march_2026` (2 rows, from the initial response): the key the OBR, HM Treasury red book and RF families share for all Government decisions since the March 2026 forecast, the OBR's "Total effect of Government decisions". The page's £8 billion this year and £4 billion next year measure how much the Budget's new giveaways add to borrowing. To the nearest billion they match the OBR's direct effects of Government decisions in EFO Table 3.1, 7.6 in 2026-27 and 3.5 in 2027-28 (uk_obr_tables), so both rows name that component in `conditions.component`, as the OBR family does. Neither the total effect including indirect effects (6.3 and 7.0) nor Table 6.1's total of policy decisions (7,595 and 2,475 £m, uk_hmt_tables) matches both years. +- `ab2026__package_ifs_new_giveaways` (2 rows, from the initial response): IFS's new giveaways, which push up borrowing by £8 billion this year and £4 billion next. The page names no OBR line and doesn't say the giveaways cover every Government decision since March, so the rows don't take the key the OBR, HM Treasury red book and RF families share for those decisions. To the nearest billion they match the OBR's direct effects of Government decisions in EFO Table 3.1, 7.6 in 2026-27 and 3.5 in 2027-28 (uk_obr_tables). We infer from that match that IFS restates the OBR, so both rows have medium confidence. Neither the total effect including indirect effects (6.3 and 7.0) nor Table 6.1's total of policy decisions (7,595 and 2,475 £m, uk_hmt_tables) matches both years. - `ab2026__package_ifs_announced_tax_changes` (1 row, from Flourish 30726047): the tax change that Budget 2026 announced, at the forecast horizon, as a share of GDP as forecast at the event. The chart's footer credits IFS calculations from the OBR's October 2026 EFO, policy measures database and historical official forecasts database. The chart lists no Table 6.1 line and names no OBR grouping, and one rounded figure cannot tie it to the scope of `ab2026__package_total_tax_policy_decisions` or of the OBR's tax decisions since March, so the row proposes a key in IFS's own terms. ## Keys the addendum changed (rules 3 and 9) @@ -75,7 +75,7 @@ The rows cite each page's saved `
` text by line. The initial response (rows 87 to 91): - Line 16: £13 billion and £15 billion, `fiscal_headroom` in 2029–30 before and after the Budget's policy decisions (`policy_scenario` pre-measures and post-measures). They restate the OBR, which the sentence names. The OBR family stages £12.5 billion and £14.5 billion against the target "Current budget to be in surplus by 2029-30" (uk_obr_tables); the rows keep the page's rounded figures. -- Line 18: £8 billion and £4 billion, `decisions_effect_on_borrowing` in 2026-27 and 2027-28 (this year and next at the 28 October Budget), restated, keyed as above. +- Line 18: £8 billion and £4 billion, `decisions_effect_on_borrowing` in 2026-27 and 2027-28 (this year and next at the 28 October Budget), under the proposed key above. We infer the OBR attribution from the rounded match to Table 3.1, so both rows have medium confidence. - Line 28: 80%, `affected_share` 0.8. IFS's estimate of the share of households in England that receive the £100 payment, keyed to `ab2026__household_energy_payment_100_bands_a_to_d`; `reform_hint` keeps the £100. The opening remarks (rows 92 to 97; the saved text holds the whole speech on line 6): @@ -83,7 +83,7 @@ The opening remarks (rows 92 to 97; the saved text holds the whole speech on lin - 98: proposed metric `measure_count` in proposed unit `measures`, `measure_key` null, no proposed key. A count of the Budget's measures, staged as the AB2025 and AB2026 uk_thinktank_misc families stage such counts (`own`, `different_model`). Table 6.1 numbers its lines 1 to 98 (`data/uk/ab2026_measures.json`). - £17 billion: proposed metric `spending_pressure`, IFS's estimate of the top-up that avoiding real-terms cuts could need. The page prints its year as "2030-21", a typo, most likely for 2030–31. The row keeps "2030-21" in the quote and `conditions.horizon`, sets no `fy`, takes 2026 as a publication-year marker, and has medium confidence. - £11 billion: `spending_pressure` for defence plans, on top of the £17 billion. The page prints no year; Boileau slide 4 prints £11 billion more defence spending in 2030–31 (row 55). Medium confidence and a 2026 marker. -- 25% and 40%: proposed metric `reserve_allocated_share`, the shares of the capital and day-to-day reserves already spent. Restated `administrative_fact`, like the Boileau slide 11 rows, which print the same two shares. +- 25% and 40%: proposed metric `reserve_allocated_share`, the shares of the capital and day-to-day reserves already spent, restated `administrative_fact`. Neither the remarks nor Boileau slide 11, which prints the same two shares, names a source, so all four rows infer the restated attribution and have medium confidence. - 80%: `affected_share` 0.8 of households, for the £100 payment through council tax bands A-D, keyed to `ab2026__household_energy_payment_100_bands_a_to_d`. ### The seven visualisations (46 rows) @@ -155,11 +155,11 @@ The pages' other numbers are not figures to stage: publication dates, titles and | 30727239 | 58 | 15 | 43 | the Current plans series, 1990-91 to 2025-26 (36); the two freeze series' 2025-26 cells (2), each 8510, the Current plans value for that year; and the CPI-uprating series for 2021-22 to 2025-26 (5), a counterfactual path for the years before the forecast years, whose 2021-22 cell equals Current plans' 4950 | | **Total** | **302** | **46** | **256** | | -Outside the data cells, the headroom chart draws a reference line labelled with the 2010s average headroom, 0.9% of GDP: a historical average, which no row stages. The lead sentences in `data-items.json` belong to no manifested primary. Each estimate a lead states repeats a staged cell (0.4% a year is row 127, around 0.2% of GDP is row 128, 10.8 million is row 142), and their other numbers are dates. Axis titles define index bases; tick labels, guide positions and chart settings hold no estimates. No row reads a value off an image, a bar, a line or a tick. +Outside the data cells, the headroom chart draws a reference line labelled with the 2010s average headroom, 0.9% of GDP: a historical average, which no row stages. The lead sentences in `data-items.json` belong to no manifested primary. Each estimate a lead states repeats a staged cell (0.4% a year is row 127, around 0.2% of GDP is row 128, 10.8 million is row 142), and their other numbers are dates. Axis titles define index bases, and apart from the labelled 0.9% line, no tick or guide adds an estimate. No row reads a value off an image, a bar, a line or a tick. ## Unreached -Nothing the addendum names remains unreached. +We couldn't fetch two of the sources the addendum names, and two others held nothing to stage: - **The seven IFS data-item pages.** ifs.org.uk refuses curl, and the addendum rules out any other route, so each manifest line names its data item's URL and hashes the chart's public Flourish embed instead. The rows do not use the "Download data" link that each chart's footer template names. - **The explainer and the fiscal devolution comment.** The browser read of 29 October found only an embedded video in the first and only design parameters in the second, so neither has a manifest line or rows. @@ -408,7 +408,8 @@ its printed multi-year window. No new unit or baseline proposal. `arithmetic` is the IFS's own scenario/growth calculation, or an announced target restated without a named model; the £1.2bn funding amount is marked restated with `hmt_costing_obr_certified`. Restated rows name their government/DIP origin in -`note`; two announced reserve allocation shares are `administrative_fact`. +`note`; two announced reserve allocation shares are `administrative_fact`, with +medium confidence, since the slide names no source for them. Real growth remains percent, growth revisions are percentage points, cash amounts are GBP, and signs are as printed. Positive pressure means additional spending required; negative growth is a real cut. There are no central-value diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/claims_staged.jsonl.gz index 61b4dade50590ede8e7bcc59ac66b7989dbd6635..204b5272f4acdf056f87588e4af34d8dc678654b 100644 GIT binary patch delta 9347 zcmZYEQ*@qNvZYHjUY^XYa#{b3ey3*SoIPJ63WS z_+=P)eGws05#UugfHN^cGmwapa|7!VvYXc(qSxFm*@&pH;L}n6N~qfVkCqOEO&K*qx~j;e>S(y<^G&%7F&0hgqhwSz&_Ys*9M7zm*!3yMR-5 zt?KLf4x-*it~hv8RQo6kZjix1BXD~72|o2ol2AG|iZM*uZ>{1vt|PvBpyn&@;Oj%| z0SLh8dmB%6wFXJwjjwOj@7eBsgwci;-fzHe0#(!G z_0g~(cnukOWg~l!KDEDNzy>_IepMb7EXD$w4974+*>S&!^oWu|xj&u}_pH`C)mIYW zLSuRUaRnKtw9w0<*s9<;QV)QdWljGeNF_+()=6C&0@k5{M~ISfsGGCn7kvJ6O~HWz z1P}Io_=rAYbxvDxE0FRiKK^S&uiaK&VP;6aKs}!=5cZtpopBjZb93MN>h-pitpWV3 zjw4LGpA&i&y!?oh)d-9SsT@iScdygT&UkbZTVCgaEanN7P*H3!+avQkM|id42UZJ* zBtFP=m$@~eM1{yMMq-aB()!=n@{+||z~ra9`5J{)Y=mwnB6=oRwId%kQX!!6v0lG5 z>0K~c+8Mi-Goc`yIlFQ7Wy`QBlK|ge22lc@k8JS1PVafLL4Qg`-}F&1h=)@m-@(hO z(XtQgyh>2D!Wb)}?rBW9JgLo=?UTOcAa4y0mKyOG3kHZr^wH&p*_qs0ans}6hhp~-Qr{SQ8f^VyO?W~_hp=_sqUXR336}|AJ^>1@`AeaU z{&4v|(@;Ku6J&_<3ve~A>nK|zWaI`C_f?Z7^b7++p#=L+V-ifK#&+JM8+J#%;Ol!q zOJT|QRBbHKh>k2w0~|Fv``#`;`G*-w$Qs~n`}i~^Eq0l;L$cyPW!9sX%A zV-eG9phM(k6=E~ugfFs7q-y4>ZN>UCK_!(K`20hN&=F6f@#O!Ll$@+SrADdvxw#j+2SAjbE-NIJw)ARR$#~p(UT-DdMNB( z&u2Mzv!a8a0w}Z(14`&jP>M1yNRD%Rgd)Qsra>ixn1s+QagqTvZm2SbDvtHEkZ_A< z!{~40Q8dMQ%ul&NZyiU>O5Ch?8-nak=i@Tu!=v?0Xv(K?s!o$m=r}ycY_Y}i5wc_Q zn*9&t)F%;=9)~v_vWD%qMchNpB(8*>1cPt){6BN>103mz0VoACgL*G)lRcEwuD!?h zeV4cK#zOKbq}zq(!R1_Waz%9Z$TDVh1GFc|BU9+FMkWq{WMHH%3L%W@s?r!kukmT# zWIE}Y>!aATfhPl|uH0CYcfI8l@$6kt`q|>X!@e1|dBvCn&_}kjkuixFmDGAvZ-h7z zmsmbRN|n3PK)_b64DRpYnPP%A4U5eHgBnWU)I~4kA1S4jqCcNLnWl zSRjrS?hl%gyA&1(e8};Rh%h!vBY6fvrS)wJ1ou!iK!3kD3L2GiRVs>WszpY*eS%OC z>yzl1|Kn;!BZ>XhfCw&8iUN%eRp>OeS*nr_r%J2K+4QlKOr1M5z6J}8R1IltKy5+N zWE3)^RXD@@yrhGhC`AH2&k+PCu@@_e$ql5U zqBeHC3vQ-kO+c`DT&h=MEQ%bY9Ir_@+AUcyZib$bkCs|i-k4H{a# zrX%H45s3$fqEGM6LfQm|0vf2<_%Rui`6%#kxVa>i+wZ&z&(O4YooI>_6Tm6ez_pOy zlN~5X05S<-Oi&apA#c7$0fm7^sEik@k_qXq13sjSffo%WpVqf}*wMCFHCIa-Db|cR z`WI}&)qSa zRGME=T#WO!_*=`-s$Q<5n3Y9k=Jc(SvHO#U0Klqrp+2KTiHkIMaVQISG$+&vCITkL zL=2mKogE_8CYCPT(M3olcR+~CfUQCM>gva^|L}Y}GaNmnq5S^m$Y_eWdAdhTNJL&m z@dV9zPhzz_4MB)FI4L$+rXAY8s=T_zmeSK~S@jgsal1FE4c(^S@wB`$JC54%#r7r^ z*gWZj)y&?^3HHu*`=#LwmO6g{jyh}80R3C9zc%ZpN!B#zE&0pmd+{D??rsuI8uK^0 zM#aSLKU|F2{MenMLj$GGp&C1Uk`Ih15 z`G-~C8KI=%q83DRvEQQ%Qt1s#iY8$@fB-h_C3N7;E`eIR(-kBtRJ1%OS{IPg2I}m? zC2B|$$pRbnPRSv?0A-=F^VE}A6gqk}Sggz+)HKKEp=c%wrrm?~~?SI%!Uw-pumN&MMpASFEQeI^4fO(`Q>CxhQ7n)QC}}nte_1ghjj9u* zx9Z)P*W?)PV<&JZicdr=hhXl~CJNSvdmD1f;S4>hRi|jPz=~faM(Rvd=vJ@SppW#lx=OZQA48=RkuJ?ZTf^0&O75o=EU@+fBt z{mD#`cWkDbZU5K%{)4>yO$5|9PUqWLpEMa-ve9e)w!x2IynNae_2q+4xkAK3MiZ(H zqx#cAWcY_i*ZLs)x(QkQF6(+F?BRa%WzZ&IAmo5;-6gyQJvYIKlM`Yf+{L`fsO@NZ zfg$H4A}rPa7vb9N;r;qz@bBzWi!h9QFlB(3E7V=1>pb#C+6n_vB99JLk)w3hu%LGi zo}uchnBRHHy1RkG*E=e7PQ_DPrbRzac10q4hQgAoLsOQqY(6c`Yj=gAfWTpsQ*TEg zuU+m35C7nmZE_0cmau!Kab0~Ld=_-794I6{r-CYYlLExajseRAL;LefN^Y?V>6eJz z=F%)wphtD?>Xl730t?s*}egVn(C0-}MY{pEek*QCL{d3CO5ji!567fY#4j3C`6PAQr zs{`gsl$W7)Y*uMVgdDVyF1qz?>P^$IPTy#EZj7WWPes-H3yz~ZTbY}q2t}u>@!o@F zy!E)V&&;d)gpNmXB%hZ*tI>C{WMjs5*}SnNL)0tzQ4b4`(7TXc>vaa+l1RG7b;)of z*bDe)btBTeFYsD*8!tX-!@`C@{%~2TD z)tD5<`{UiBYKnO(FLrC=JV+n`hU!f?$TLMZy%~9J$BZezf0#_3XEyPf5oFz91%j9S zd=oQUZwYe;%r~stx5S)`QM>~W9t6fba*T08;sn^G47#>LQNHnUQ@AzKBhoYc8vSh= zVcYo81?|vY?x640#T#Zop{}fuu=?#YKRlJoip`giq1Ke;vg@NOd4K^MuvJ|cQRB$9 zl89Sy6x@O5FlCkAVr_H18<$ccGHB^4?bbmMfQbhW#0ng`vu31;;tf~6438ybsVwj> z^`M>cpe&z!%M$mjN12w1;xIKSj(JzmSyV1Nm$;SoOeqq?_Whzx0Gy!Dm&2Hx$|4=g zVz2C-s&rP4{OaM~p(yDFZWl1!|8}Qvq=3*YG5E@4jJ0$yj5*O~#G(~gOsvuT1|&7f zU}ahxv3CRI(O^FlZtoLuQ_3>S#}V<+Ztv9I=iDw?Z)<4Lmf z8)t2)#NqMZGWnhRi$AmFtVxr@@UsTheQ#YbbNQaT zF_yDJ8ceQ1;duyvv?9Z%;WAuaIL@C{txZpR*>psuP1s8}_empKVM=b{arn+m!yESM zPEqi#ys=>c)|0mfrP~Cxy_Dt~LLi!D6&P*DW)U<;sB+?d!})~j_5C^J4g8(H5oLKh z#R#qH$>gzYJy~#oK_Sch<>CPef0q#pyYNv+rPdXkmT51*vKMz>X<~+pr(6r^91v9* zE0D=3S_3(oT)bi~^Mlr<{6bY z0j)1w|1ggKz`C~hzEM{mu41E*-_DigX1dLJrcsh2e`J>dpZcX`-`##lqZbcy<_owp zs0B(kNlOD*qpyY|c}U&E_wHSzqv8B^LU@qMtny2-XVZ{V#P5%{O|h%VZ;UB^%(_PD zzG3Y1OXaUc-lmC!8nlr(CsQ&a`X`Wqyea4FwBs7}bTp15zh?O_=ze5Jv;~)5D%_OS z^Rg&kRl*uZI!eTi!<WQEE4vPABZ)<_q?e89bq&b7G(}Y0fnn0@O!kKif}{(jCw4x zs4BkCEelIjXNR~>%B<^%UMDv826(oiK?=@bK6!iB#Q3+7ovpbuisS@%kMv8cijYO1 z5nFai7fT?^i;P7_nGQ+Zb2M9O(X>U>N-KqW<*Ao6k_@fNe3!*gKo*U`hm!G(N1$Bl#Ue2_sTj6bn%@ogmMJXBHSK^EJf8>J0>*x?Ap%3Lg_Y#@$N2jF(7&@ z9U8|7F%!5gRb^+9V;0r+V#ZU7`cjM9QaY*dGFqKCXAEPFMc)My2*xTmhb@yBN(%S5Yk<)=fl0R; zPA!lGRdOD5y<66-u&Z*+rLDU9^@B_WFh>@9B{v9+$6&X!mhRo83(h*fBk=w=*|=k=E36QYrM^)b?cbp=1Xs=opRM zZ=T$59(phCk9sJI)rxXE#5<)9qZB~0r*?er4h3= z#0RHhEsqBJbvn~d-LG=}qZtaP8$?uN`M@%Emo6t4lt3ypkakR~nM26HPFVvr=p8A_ zk{e7o1tSpjeDiy=wkxv$j$$YDIx+~WbD>o{T34r64ZeWiGh3q1O~X&8`PydNtZj2e z`>CYw-LX=`4}yvkBHg6oT5Y|r%=l8Wn)*x^#@JqOYk#oLIF;E^okb0{PZ|@P zrcQ;lz2}t$RCj}XN953|&mbFDjHp*Gl6KeVQ)9*ZNX)wvxobo3Lpr-xY$`9Cj7nNT zD>QrG%gOs=Z9L~Q!YuGlg2HaG3W_u{FCnV*`-kYj!+RZI<1k^iZ(f5FZCnIhgRc5< zpTzcv!_u7M5%-=PhFGQtXb5zA5-q;h9MtncED~J+iv5px^R}(3_-?g10zcAZ1xp1+ z!XtVSZW7$dh{yzKh#Q65F(>E}4^`F`olxRxW@i{L{jL_Ghq(Uys)W#tRRRn4&0`x# z+pPGdu>>m9QmB)A==}0E%NRwp{)E2QvUO*J8r#&xKM#Cj^T}B*i4rbX?^$1=Da&#$ z%@Vd8P-^lk2?~5NMNw*$Z7jvMd*a)TH--4>l^_E61}2fiFP@Gs}!)%r z4a`!vpij`L=1fhN9!Hz)-$#-4mUE!5^)k*+5!rsLwe46tXeQ*K>rJO1M{}Svq@-(;(X!?A_!NLt=dxia$NAs7$bLMnxCOw z1M|KEmsx0x;zD{$Y^#H9*@`?U(>>8f&K;JRWG7{#<>J zLgN(>(rf}*YR%MU8MI*gc={V`$8gL#aILQU^R03(d}wdkZ>3r4%rs_MVC$^%`lXGO z*7>auxm73Ft%P$tRq@_l_-dQAx1iV&NUXpKhnVAy*m8W_Tn&*5Y~e*I_%g)n-)zlk zE4FzEJ!5qTq=tc8^Pq+JZULbX^CzK6jKkX6y!pntruz}rvO)DGdAWipJgUEIMFSHJr9ZNsYZMObNt&s@bAWa*Ud$h_4BOlUA=%*PBuXX$IK9 z2Z*)c0;x*u1T2|EEqWq!4^;_c&6vJWh<;3Rs`E~USj1|)x$-Z%5db_97X55uzwzn; z5ld=4rwbe(8;S2rS4qk)T(3JJc`O)cz zK8wYpC1cqmZ5TX@FIne={Jc@S!*D02#y0)=u`KU2Ene)?+E883#hAdKJtP$f92T?xVmXYe-QN_f%;<)9(KJswfw+T2#Dv^j&fD>s4>r zPwLi{#F<$2;sW5G<_3H^ND4%l7vterhLP%~D6vBj#NJQjmZW~3OtD_X61q@Kdi~N2 zybm)uO@oi-vdIxYs)Lg^!!C|7@K<+peN4_MT|G4K?Z$w-NB~J2bIZE7F3y=|K0jT=nDRq@w7= zbh3z~{euy4{68!=?#F*)&9l|4b^l^kQn7p~$9{^-5mN?l33ibx5mpP~7eD0Per8yu zP3is*M#b{~Fdz_Wb~-!+)k!STN)uD-zfwxxr&PLh$+lZb#c-V1D-`ej+M4Y@f1aHT zXOP^zn0SJACAiuWecMPps$cR&an;@f1o=8*i4(9ZV=B1dPSHsty|6~Opn9MoFc6a> z@fE-!ix?tz^l+3#nf`7O6d;w2j%A+2C~<2bj&O)ovqzcE50!$Hk<327F);!+n@vX) zvHsYqp&)__K4x>a>Q1;4@>6p+Ox?p78W>#Xjv`H()!-i@C(1`S81glIB-F3KC3mKk zZ;m{eS23bvj5n>*dI|D3laluBZ}~=Jlf#IEMXCgdYDnAA=jypPYiLOI-V6w;L}X5k zw9wAlS@JTpf}ek`%A-W3H(6j9BAt3LtS`>jj0E8-Fq#nU88~2;;fY(S>9p6~Rg4^q zeHjoAIT!FmJY4^IYk=|6H*47f;~Yq`f!5M(g=^y9gWIf}EQAoRYnzG3NhZ4ajL*)a z#u-5{LB~$-a)Jo4K~90%2&1B{%JWZMP^SoCCodEHW2n*^1yTd~1*j<5ls@V| zjKiOtUF6glR)gFCniL?_AzZ+0{=T~Knp<-1CyW4z!P5Rm;L&>8fyWRKxIygxrfD1- zX41#^lMEeY9~QEXSl7VEqt&nJ_okBxeW3)2t@tFG{pMj{R90kFba~5EdCSPsW-9f! z>liv71aPyfB}mTI%TKR{CrK+q8k~D}69UCZ*&?A{lh$2A+bwk^lJ%{IeFJ z^JwrprCUbPSM~jDgg8}S^0ZY;UYH*+26GkETJjcMeKoY|`WDvRviw0X&FuS@?1`nk@<8YfyaZBB_C9XX-@dS0(=$ z_#bAajOJGA7Ql{k)n|3H?{Ec0pZ{*V z-Z-Jg3&**qV8^vn-(wU>!#+D*gQfehZDe5*zb*Vs$$jpw`+KA0KV`95Gi1y%kmpd! zAOz{Ll|*)`<4A8&J^n(2?Ovodr>U(r&~J;xAZ#kLdw_w&Wre_-w6}s7bKCsjlVAer zR`)Sf8+=HT0pPaa$w>vH^AAwHYZ(n%S=aVF7UNn(<@5m(pZzyo?{FiKshA6u4!`>4 zr`+$od~`D!P+pWD_+@)}@Xwr!ZY-?Pgm}UB^nzoCnFPc6h5J*HaefL)a79CfASutc zrS$S`1Fdj#pz@*Y*%7F^8~(xy>$FI4XXZ`;d%>U`b-geXV<8RYBE*+-crBdHlz#K{ zvIX4w%8Q%I&WLe?($WfcxY)b}(=rz3m`lDY943(FxKw2zf=S+5sv{*G0a%7+*x>|P zWpCz9@`M1gfNPaXDqEPXo-Ro=6GC135Mrbez>G4S7H{I)_(OJ-@-IfNbwV^%-rPVY z#L769NR@M&*}>dV4XbN?%K4A=>^s*69ad_E=`k95FAX_uYG+tvIr#+$pZm1qzH;JS zU^Y#3PxomEd-7OQi=(T`zJ?~Vt^v$7;u0{=CBKXM%3Lh$U7~fM^ihNS!h!-}_y9&1T% zHV~|hjaU2R%|af{3hI8I8?fWVL5vmX{$si1yYI|jwY!_EPWhnuf^V_Vok%Bh^; z_a}13s-U=_FKdCL*iKRiTGB{oGixqjWe~ z(-rvrn7PQ>fGr=nKrn#*KK?;VYalO?oWOUNhTMUGhE1@WGA!e2wO?w7x5i!z*yt~{ zBXLU$!bv?p{mLhJ~ZZvE0x45lP_ z+*=5n{ItT+WM@g7ht1v6Ik%1^0k@zY#ni(#Bk|#_tOePoIlok2G3eMF!Q&{PTdGS*Gg(- zK89rsV0YugHrn93pTF<>@!afg(E&HRd>{R9WrBN~25Zx`nLGTnmPYmuHIUaz(!X5k zvY>an1PDEu2p%1s=%spmu)-HPlK*?(KUV0}8tg7?7;?APA75kpkuylK`wrx8JJq!k s|HG16$ZM&g3q9!Fw|@j9_-?i19+!PWs6WJiAO3d_y7O^ delta 9289 zcmZwLQ*R`w0*z9y{+uX5jCp&i1aniAE+qP}nR-b?7%*@TqMcq|BwchvU zt5sK_;EAE&wfTe~i?A==oWppF10}ynJE6aVe^U**GnXzRN6-Qaep0B@K1Ivi?Gg|8qT8wy z)w7YN)r^IfZy)#Wd6&?<^E@q~W)nPy*=~5C&1!INBfSTJIHT@|zt({7LKeNJwUW~h+I8x>7@>c)f7P81A|ve{k>p1E7W2P!#nK=svZ4ig zg$To4p3ew)RT=IWsBp15je5R1(aE;BiJb{CNDo|r$UTw)B1oV;xOwV>r_!wp)GB6& z%)U_;;SJosz9ToDZp^h|LD^UCL5B9HPqrI%UjAo>GdJrZIWP1x4wV5nsBRKhGq-xm zs11a*Sbp2;0`?%U2867ox}Q$pZxTBtd<#ltw6~5N?d{=d*!V+jZ}Kk#mtD?HNbLtS z`=nJM2+li|7ERa-mQ`+-v?6^`ExZv;!BdIguGYQuL39t$@PjEeihsmkcZp?E+AAu4 z+&WGz7`k}=4&_{f0pT7~x2Ufn8|2CB$0Nf%0=w&DceXZ2YP5ErJ=orlJ9cEaWBbm( z5dR9qUiHxd`D{O6J)w)rf^qigzsN91UDH%RgYHY^8DB{aG#)klB_idu4rdxDX{fnz z`&8(-TH)qMa~31ml$ryKyZ_bhysuFNa$t!5Kr)9mF1pQcIe^+mtg+?+!p!>_2@A^E zc^6Rz-7*!0HGXhkUXsO`_{WA3!Tbd$*~Kt(*ksZj4KSm1v=6cEEl3b#8uDOTZ)^qcB@$ zQ_HI7=i?jWc@R!65gGzvZXhHN5Qaf6ECd;ZXwF5evFKtwR4RbbD-}T=#M%z=B!<6> zxEFtu=C&JD*JaSB_hQM$vsucNS88?pPpvVHRC#b?$Y>WuLs=~i#J#^QsLY-9$<$s4 z#qShY)fg*rWO2Atr2D?{RocmzesfaqXba-k`A|7>N-C{=HRDYYvzFf<%n*Zmj7eX9 zm{4L6VV;0!+k9Vxq=FCW1R=AMhp1(S-yaKvqP~hYZsgDY^!(;Ny>GR-3w6ji!P(J8 zXg7F0bVFL45YN>}kd!1CMR6Os#O>xT2)wU>PkpuA59y_$>FKL_==d3)rYIYy+L~WL zGCT)|X39u5MmLW9%alauJ<|hZM6C70bP;2?Tg5@Zo!}T=Lj31O*s&ZTmu2uOj+;8f z_@>gi+?y@L&k@67$0T^Av$QdUG{3__g1J+9f_3#VL{2?S^R#-AYREq#9-wU4EN94B z+oPad%SyywH)NL|c}_1KMA*D`#9F^xOxj5K-}Tbk6D(lZ7QO-hIx03^IR;Fq zi3*^f0t0l{TsGMcCNfnU5#R7#`MG2p-9P^_0Uo`_N2iox-q zUB_Vsr2zm9Eu zmEz2$UKoRy^!%}>)FY@4+#ezSryZJSdQJEaZxrcN9_|0ZSpTb0_w@}Y$tF{_pTR(d z&$LR7t`EN;g_YRh;5oT7OA^wJO+aqTh1YRHBgh`K$oP-1c75@fU~wZ;rX~=wG}sI^ z+(AByL30PfZ7DcM4IN`MNA&iD_$%e?F-5d4*itFpHS%z%$8n9IZ59g;g)xM#GIU#r zf6~JE({y05@Ts95v6Tx}033$7Zb}CizcGgLNp1gv7v#iXkf{zFewpiZf{zEqm!N9 z-%|}ThEcKk_*OD;$u)ZiPJruR$@0Jx`c!6jv_Ii*EuB8{*S>q2hZ0_eTSmLv)F+tk zY0aXrVD{q;^(avuKYW_vIb$f@^T!1ecPX(rtp9Gx+z~_kRIFKD`b4}kLW413GUZJh zZvce-vT?PZykhXemNl@H9sh=DWP@coz>+9+uTEs3A_xDQs7zG?b8jsmI<%1ZkBXh- zgPz4oGA~%Su{yFh)j;Ehs&=>zF5LF9Dtm!br(OXyQHA;!h>xsX_qX9&Y(8J~KvwTs zUUQ16=R!!Co^Y@o z1R0H8EnQKHxXGQ7b&IY)h|$dd=uMX#(uD(Ij{!^QDJHn+9e_$x$QWq(k?UBTz`!^c6$D)QQE8% zV(fJ8oYm32y)&JC+3t?fafPQ&Ky4ZJ+=kz$-;w4+6!;9(w~=VeBO>l1%pe_feGnc{ z)LLj$#^{;dGrDpRd{kX4X48K^pjI3`b}Xa{%i}bIYq@2U8 zu0lO5nk4zr>YpeR-QLI@S(}kic3q?9^-F0tX_mcY{)%B)8smwEm9-D+9C1DV*<{9P zq_r;X=S4&;(xqBwq`nGV; zhLjz$s|i`a!0Ev!yAaXi^_q;C_N~m~Mu+JWbvVh5j1+j}r>JNW^Z5x?Y>uia3Bl#- z9a(2Q;)NUYBJK=~wJ2S&u@ObLH{0sTTCMRwFi7MmY%9(WgTV5qy&vD|_%)S^J(s$D zdWJColdth2HU5$C#&jertxC8+ql+ODv1@W_9*GR0!L&0uopP4lUJp{T6VP{nLK0#0 zQEPlFSGrhaG#w&Ac=(L}D)lp`2MrrH%^u^s^x2CEWzvFhjmA}qtT)7udj4k`6YCQY zFrHM_FX1?urvO>-aX5%+?b`BlL_%INn?KE%3)~DQg+4N6a*g&L>!Gt2k3$CjM=C;h zBW3S*S?>(G`4`@U8!O(}Q7-M*iE-=NRNC}3bo}XK0N-Q=Uq0!;701N}h0m3g(@VAh zwW7?AtT|lv!xkamnx+CiFR4rh*zFT@_9YeFN_}(q zHs&)oAsg7XOY1pY@krG5T79{ynEj4Dl_(a(q4XD^^H96z58UIwq8r2z+B*kv@PK9# za7Hf@S4)$LW;0qGlu|RP-&cqrM}DLprEW&`w5llL?~&%&ixZW3;j;j6zR*Kz>vaza zBpL}mT7v258%p(7Yz+B!JQ6NcViCzyraL(3ALn3eoPx0cL`m#S02p7T zfgJ3m$#9Vd>Q1KOPlh(cBQFpzt={YS5+#hXQ_KL2=KXsX$E_-F>%Zp--u3e={%Vk( z!EfK0J3=ZPm^@?DDW#Qbw>^;w6qK>dON5yn`m^B+qj0PGuMTM2))FkBX9ssEkX%(! zE^e9OJ11<3*PP9Vk-pl9<*K8o^C|kO&r8N>Bj@o@5$gas1AW`!_7&tu&{U_9g{B7i znqT)7%fE%dVH7B}!P??p0 z(x91gr%;&qNEh|cM4KGTELlK`qA)$QA7_nV{bGug=r6$J=-Cy?-G8Oa@#p6ly(wSh zsCJS2vL!0Y&`#f;SOCe94dW974PRzt{z|sdMv3PcrbD?Ui-DU-&ha*`VkX8>&SBfc zQrxGT=y>*-|CZ?2QSE{4SPDzc$MPyS~R1y;J~k^BNoumkc|VHck=}NP*u1( z@`Rf8+x}gqn#aaWBa*bBkY{5d+?YbSj&8c+jj67}n4$^YGvXD-QaonswNx9=%r;Bi zjWm%4sD!urdtpeOI4%*tp@tWl?HSHYEm=BrL9ek!*H8nZuc}l5{G?8K*Gn7BM7HaG zgykfUT7a`(a25nXs>;b6p%4#0I;UWE?uHnpUHSW5kXoTvB_hG7;8EE{x$aXl`5~aA zGx5)7nN7+gc((kF%9Vp3H}#i>b5#~LAztk0MtS-R$7MkckFZnPmVnfE9G zvqUnPE|IP;5(&aBWLU681~#^qNN{KtEC!=k&lIvT`VBPf!!un%8E0xC*oM*zPFwuz z*~AFm_HQ-C&b>xDvQg*QD`gg~a|7s<&#H zidsrf%?VUfOX)wW;b(vPrdN@>+L+`XtkKU~$n#|7y#|FGOT#d0EkZ0n@#P>svoTcLCJWRLI1kZ3&As^w7I!jIe+gmLN;W}^ z86JMbWrR#lZbp(1gG0Pwu9FzDX00h0w^eT;%?r3yNEu1cU4p~#E)G$-os9-~xnJ>G z4hXQ9St@%I6*dc*bJSc7TE{BgYKyf-{dPcGU$;7s&1%amu1#BQnDNMinl_te zux$RJ@vaO)dgFv(NN~R{YCA27VnU5gq5-W#*xfgu0#@<|QGOtoYG;;--|QP3VY)g; z)Kh02rPsPJF+R6*cC}OY2MI_xxF^QzM0T{~SW8fnCO0$UH!8;G9bpotFW1x0PGVR# z2P#{V^n-8lCIBk1xI6MIbDCd}r7+_ai;83K&Xp1j0(f%MZCV9Spf7LKNUx1i$ zR!4|;zC^v<7N{}VHSt)mG=50PA@W?*at1d1&_@hQBoOlqWufEJ?6zMEsZ*=BRv){h zAq{?eut=!TwdOvi9?~G(9E2?T!8LdYrvKfsX@!N_X#&IDQmho`Nhv?ATp9H=m?}Lt zNZ$u%c+_XXOGDNwuBuPl5)cjsMXr{nI~UOOgwJw<#>2s%X_h?c*;2YK-1I zJ~#`TaIyLue=3q*h=a^>x2kmG?EoX=r*I7(4f(T1p*z5-Ws zXJJDryj(PI7ygl$5ze%6KjR zR9>#(@7pQT7VqL1^`xyU&pm3mu#`DOh>8>EoogP*krsvpl+XARE;A;iD?_m1WsB1* zJH;=OVqB7Iy-cx*T!>W$GZavVIAzKolcpyC)_6%w(OSKRaKn1ifGTedkiUi@du8~f zX{cfSv9#G^I`FAzu#5g=#pd_rBdqU;)8#i&Ur_to#x!HQgzGc4QBUTfE=sFqv$L4! zx*)d%WLXU|xh;t$Zmfc|(Wg0SDlB|LN6TEfdc|byQ zT=ux|dUwrsgr;OMC{M0%ggsty+x1t%Qe7ypg7Y#bDn=ElwJt_A@^p>8g-`LKYsy^X z;v-Y9A#KOwDb+uQC1bLIPc~NIUn$d8qm3P$7wrTze8|ttk$S0lX_CrR>fmK?hOL~s zr9ZoTpY6>|wv-kSQto@+`MDu}RSa{3YJh?y2sBC}eY&`Gi#ZK{q-Ue)tgGH`>b zN5|?o+g~c8wyxpwE}1}VI=Kxi@nr-uQxOcjjbNTK(JxT#g?k}00qVy>^Pe+ zUKsGfqmY>Fo?PQ075L;M*!+G`m=2u(@G7d0KgAA?=0qQ9a>!OpC2p?U3R!e(!aK^i zc7bYTK}N*`E!5hu>&`N7L#Qs&HUI^a5j`|1c4u3zdh7cA@%2E=tPIl<^0S1$2mDSy z24?TC;`cVXZGVSy;e`CG;H;0X<{)sB5L^%1f^XG>)SgneGM-LuD4A9sD`_lyNIfM3 zL$##ZC*2RaCq)d_iLXSbAqu&AeBNC>-@Kf!JSOS_{r2fd&!!z&e1&2jJ>kNzDT_(h z3QYbDG|dDY>s_gJMbv3obO z^ZfQrzxs)V)QIivk)W2q(52b1bC&%F+xodf2c3TG1gc9@uI7KydY3_HHcL^i&yg4) z0UoO5HEzYv$%8>_;A((u`7A*Nm%8&MztrJOtTF5;q{vl2QMO^Si41o04sPZj*bQQv zv1;lpj7NQr7Aw1@&GI(2j~dkeYPEk&w(ghCP6ivzT@)h;!SeGVVzW!^6_)A?m)#{k z*@RQHSWAqTc~a_9%`S%UFw-LuA$xY94pcgFm%!LcGtH^mbVH5-D~V^J-)?>JO_vYr zks;vr9k_HRE|J}pwvDzbY*l93Q;kuw*4exZ^Ybg&mwj65<4yY`+GZsnalcB_=ayrQ zq}j8#x^y5%hn3sXWqH@?z{AD9S*4?0N8RxAM%{44^~k!}Ml<__Rmu{5$7nrh`K*DS z(f4YADBCzhw}QX2PGc#;{A2)cpvqVCfVx~BjEsDNTD{^d%$*qJY;9T5X5ZNJFmT6j z*r~niqpf8Zx{@79a3Y@MaDx%kpOGoag><-o{tfB^6H-D{<=v4q0Ol0fBD5rl$ zv%%wt;_f{eTP!<~X@W1V-0J`YKmebh!#E&WH_lR`Klw}tetX=!(zQ&{B?mE?3K?<1 z6=Js%LGej<`8P^Nc52_HwIX9H}|LFp&#*fLj-HD{9xP-QO=GZ>IKmC=Xww!e9*pb`{Z{|6n2Asw)~K_e1j6hLN^(ZP(stil9y$ zKl_ZFhXs9)vheahhU$TEptehI+!-bszWi1Zy}#is_y~|Iw`e{`kS&(Nvz8WH&Q`86 z9GXYlq**V-?(tlA;MaQUR1sewFv zes;64MiF^qfCZ!~Ty3^Svs2Shz-}(CS8`sm$+fWsY)WDot~Ln42r(28$WyaF8a0)& zyKaZ?-(*n_g-BmI3F&&Q80nz@k75ojqR5A=KPsPZWdCc$U0b*jK$uKa;yfSf@WLS` zP*)XeV^@d=4FvB0d$4Ca7b3@B>Sbm_n#=de|FLV4sc=nR^~j{3VrYLE{!?jMrUfCB z6N#-wRkKs-fE0#H3172%}i_`@BRinfs15KtVvd3Y9$ zpo*u65$MY)=W&odb~eK+&Sgke)%puKsF_&7LdAVb+Jc(*X7bkbX~Q?S^9kzBJ%HCQ z(X;o2#)~&^eyhZ*L3uYHko@%>B2gciJn}g@z9l2x{2v-5(fNt#hz?v9Z+7eqjy_A%dgW07~ticdt*i`^TY?tUu!(vaV7h? zQN^*d&Gbi${6*2Si2L9OCtOT+HCk{rik}}l*DV+wBqt|ANjgvOd=2x1>3AF**~Agzex|OX>JAzhtZxcMn4Ibe5v= zADB0hmcF|iaTid-6rPV|F_7}PC=xW*==r{v5-dtHH83t?$yI+XHC5~DZ;pC*!ap+B zLTiawhHFp!iM9+mtCiQKEK+T$h#)G`j@(6=kKf7XP4{8w;a>D52qSg81fAFr(g868>O%sXcPNzM*0~R_WIY${~3g z!=iN3jp_iIA0m056*fGVxIx+G$i5>i?c$uL;AiPDH>d<7%}71?D3jl{8!Zsl>_{J5 zLfC&jzyxqSzGlzYI~uMsfGd88I`9k-NE;Ws*AF!K;9}KKd6c!d%2p|($idZ zObZUjy5!pVS3_p8X+4bK=t!=&QZ0wv#=wWAHe6vfH+^Nizs76o`2cqe-2k_O{M^T~ z?apKrwB7g&+C=0`XMXx`zJqk|LwWeZvTqJFWIDn(gnm`7tNw>OIbHW-XnjF<;4C3_ zG2Pz*%#d7&9bJc{eGqlW< zj;X#hnzhzvK*ra?|9(T8hJyL)@?&txz-aM*V6~>?!u*()X6^qavt4LXem_sQ=ZBb` zRJ09mFc6HxWWy>`I$d$`BA>t@_L%2y`?PBg`QGW&szRR4V*6Akp`6vvSINqUc-&j7 zm*Ug8lJXSLUWBSVnHL@X6sLH5dKl*Duhzrt0{gW)FyZCbEMm{>G?~3{6%&{}7ACUb z_!)2N?%ho6(gFH)@)N{-{|;O@PtWE&bOVDp+t?muhZtGQ%K4W*$Z5N10|=19Q>60O z1`2Xl#}=}XDlB5G@55j1KijtY>TGudflN`Doy@oy=HL+($FqocVqc}Zl!z}Ul6K)M z_tc^=t*)ZM+%v}=)**!Tx_ z7d>w6HOqPE@eyvjFj9JioWP5THOM0Q2hCj>KK!^WUS~yu{3r`VHyzc`XziaZf}-qT zkD2zDGX;eyEQyI;(=vz^Nm)Eb1K45$3~hFlhCH;Ig!!eBb>SX5t6k`n_P}}@Suqt- zB%h{&b7Zd_Pvcc$$}?G1H|8Oe?f@PKO`}M_YDNqBKfbTq zZ|7#r_vL!xZbizL_VP6~Q){s?pBU&}o@Q`$|Eo2(9Vpn%hZwM=#fuoPST3aK?h$Wa zZ;RM|LxYz$YPhdnTsJ=BMUJ+W2lo~*KbkmG`%eoi^zf&(YTlZd8S9L$W)EE*-s-Hv zE6}~8txx9t%A5=HGUj0y2t}`<&%okxZ-_JF7&7cY>FN2pkT$Pne1|&fI7U^VYZc97qM8?KA2t zq5^(ZVQscD$^Vdc5XJ$N&|#bp#2*SdRKOfH|m1IYwjXf1Y>J4hdP z9sQCF_sP=XckxP86eb3J7#Li{eHe`riq?yb5<+|2lfEGHr8U-6{{^yr+%~59=D{bT zvMM8yv+|M32@N=%rPPTKXM3oo9%Rm967Rm2+v3pI)S96u~p{C z5AE=Zi;S`0(fRSw71(gXT2ZFyIEv6F_1STz5iZI{-IW!hk^{7Ai@4W%hwB( From 8b7b23db03e41c953b177a2856bc8928ec7d9c3d Mon Sep 17 00:00:00 2001 From: Max Ghenis Date: Tue, 3 Nov 2026 18:50:32 -0500 Subject: [PATCH 16/52] Answer finding 3 of the IFS review of d5e323dc: README rule 10 - README rule 10: totals of the Budget's decisions take decisions_effect_on_borrowing, whoever prints them. models.py defines the metric as PSNB, and ingest_obr_policy_effects.py maps March 2026 Table B.1's totals to it. Single measures, groups of measures and a total's lines by head keep revenue_change, unless their producer states the effect on borrowing, as the OBR does in paragraphs 3.20, 3.21 and 3.43 and IFS does for the Budget's new giveaways. - I moved 172 rows to decisions_effect_on_borrowing: OBR 87 (Table 3.1, the policy workbook's Table 1.1 and paragraphs 3.3 to 3.5), HMT tables 18 (Table 6.1's three totals), TPA 40, RF 25 and IFS 2. - The OBR's Table 3.1 memo on the current budget deficit (6 rows) now proposes fiscal_balance_change. Its sign convention, and those of the red book's two Table 1.3 memo lines (11 rows), now name that deficit, not PSNB. - scorecard_db/signs.py reads the wordings these rows print. The borrowing phrases gain the OBR workbook's legend, "positive = yield", "reduces borrowing", "positive = increase in PSNB", "positive = cost" and the OBR's and IFS's prose, so the red book's 18 borrowing rows read too. A total that adds rows under two conventions, or an average of absolute values, reads as unknown. The exchequer phrases gain the OBR workbook's legend and "reduces borrowing", so 2,576 OBR workbook and RF Figure 10 revenue rows stop reading as unrecognised. No exported row changes class. - Tests cover the borrowing conventions, and a Hypothesis property and a check over every published convention that revenue and borrowing read one convention opposite ways. Rule 10 tests on the harvest cover Table 6.1's totals, the total keys, the OBR's lines by head, the stated exceptions, the memos and every borrowing row's sign. I changed no value, and no sign a producer printed. Co-Authored-By: Claude Opus 5.5 --- pipeline/stage_ab2026_hmt_tables.py | 6 +- scorecard_db/signs.py | 24 ++- .../harvest-uk-ab2026-2026-10-29/COUNTS.json | 12 +- .../harvest-uk-ab2026-2026-10-29/README.md | 1 + .../uk_hmt_redbook/NOTES.md | 6 +- .../uk_hmt_redbook/claims_staged.jsonl.gz | Bin 36365 -> 36385 bytes .../uk_hmt_tables/NOTES.md | 2 +- .../uk_hmt_tables/claims_staged.jsonl.gz | Bin 38260 -> 38274 bytes .../uk_ifs_ab2026/NOTES.md | 4 +- .../uk_ifs_ab2026/claims_staged.jsonl.gz | Bin 17261 -> 17278 bytes .../uk_obr_tables/NOTES.md | 13 ++ .../uk_obr_tables/claims_staged.jsonl.gz | Bin 111394 -> 111510 bytes .../uk_rf_ab2026/NOTES.md | 4 + .../uk_rf_ab2026/claims_staged.jsonl.gz | Bin 52874 -> 52938 bytes .../uk_tpa/NOTES.md | 5 +- .../uk_tpa/claims_staged.jsonl.gz | Bin 20674 -> 20792 bytes tests/test_signs.py | 152 ++++++++++++++++++ tests/test_uk_ab2026_harvest.py | 117 ++++++++++++++ 18 files changed, 332 insertions(+), 14 deletions(-) diff --git a/pipeline/stage_ab2026_hmt_tables.py b/pipeline/stage_ab2026_hmt_tables.py index 94b871072..e7a360dd6 100644 --- a/pipeline/stage_ab2026_hmt_tables.py +++ b/pipeline/stage_ab2026_hmt_tables.py @@ -127,6 +127,10 @@ def base_row(src, *, title, fy, value, locator, table, column, conditions, metri REVENUE = {"metric": "revenue_change", "unit_concept": "gbp"} +# README rule 10: Table 6.1's three total rows give the Budget's effect on +# PSNB, which models.py keeps apart from revenue_change, a receipts line, so +# they take this metric. +BORROWING = {"metric": "decisions_effect_on_borrowing", "unit_concept": "gbp"} FT = {"proposed_metric": "financial_transaction_cash_impact", "unit_concept": "gbp"} @@ -222,7 +226,7 @@ def stage_t61(wb, out): "table": "6.1", "row_type": "total", }, - metric=REVENUE, + metric=BORROWING, ), "measure_key": K.package_key(slug), "note": "Table footnote 1: totals may not sum due to rounding.", diff --git a/scorecard_db/signs.py b/scorecard_db/signs.py index 142278fcc..daeff9aee 100644 --- a/scorecard_db/signs.py +++ b/scorecard_db/signs.py @@ -95,6 +95,7 @@ "(positive = yield", "(positive = increase in yield)", "reduction in borrowing", + "reduces borrowing", "higher government revenue", "here we present higher taxes as positive numbers", "reduction in public spending", @@ -102,6 +103,8 @@ "positive = saving to the exchequer", "positive = saving against the current budget rule", "positive = fiscal tightening", + # the OBR's Table 1.1 workbook legend + "a negative sign implies a loss to the exchequer", ), ), "spending": ( @@ -114,9 +117,24 @@ ), ), "borrowing": ( - (), - ("reduction in borrowing",), - ("increase in borrowing",), + # a printed total that adds rows under two conventions, or an + # average of absolute values, which has no direction + ("mixed:", "no directional sign"), + ( + "reduction in borrowing", + "reduces borrowing", + "borrowing reduction", + "positive = yield", + "a negative sign implies a loss to the exchequer", + "positive = increase in net tax", + "positive = additional investment income", + ), + ( + "increase in borrowing", + "borrowing increase", + "positive = increase in psnb", + "positive = cost", + ), ), "household_gain": ( ("reduction in benefit entitlement from the two-child limit",), diff --git a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json index eccbe6eae..260f396db 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json +++ b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json @@ -27,7 +27,7 @@ "uk_hmt_redbook": { "rows": 432, "primaries": 5, - "claims_sha256": "1ea59fa28e9914b0c473a30fce88cc335d64917f77ea972dcdfc58d0723513da", + "claims_sha256": "fbb4c6e185201f8904007545f17b750bbae35182833fe2b77887dbc385d23b69", "sources": [ "hm_treasury" ] @@ -35,7 +35,7 @@ "uk_hmt_tables": { "rows": 762, "primaries": 2, - "claims_sha256": "c71472c6957ebc95406002f360da47bc344ba0da43d3702ad6f833d47898dd45", + "claims_sha256": "b962aabb0064fe51860db293bee21aaff644894d472e409d05bd96dd8f2276e8", "sources": [ "hm_treasury" ] @@ -43,7 +43,7 @@ "uk_ifs_ab2026": { "rows": 144, "primaries": 14, - "claims_sha256": "b7b612261417ffaa580a43d938bba218d1b1894adb34d81b977e33930c1b2061", + "claims_sha256": "bb1ce115de1d141db8af91dd84ec8f94417c051f6994a85fe2d91220c19704b5", "sources": [ "ifs" ] @@ -59,7 +59,7 @@ "uk_obr_tables": { "rows": 2797, "primaries": 3, - "claims_sha256": "0fa9a6f30a898b4b0a8f89de52b1c878621144394d8f7d289778c36ba250e24c", + "claims_sha256": "abe826475c2e5da312fb55a69bbe08b5e5dc71ab12e3b655b8923ff782cdd23a", "sources": [ "obr_efo" ] @@ -67,7 +67,7 @@ "uk_rf_ab2026": { "rows": 766, "primaries": 7, - "claims_sha256": "a251c700f0f10d4ae47ca8c1d2c2a8ba9ba660f768d0169406a14dd28e8c3f85", + "claims_sha256": "48e9f2889097817a177bc98f9b0a903020222be6c034b2f74fb61c4bf35dee6b", "sources": [ "resolution_foundation" ] @@ -86,7 +86,7 @@ "uk_tpa": { "rows": 305, "primaries": 10, - "claims_sha256": "114f27d2e28955ced736752d4962525adc4d494160a0d2da689cd8fd18c58d4d", + "claims_sha256": "4225c7cddab4e4d85a94efd4e624319d4457ed040f7b2656aae1ee6a5b5826df", "sources": [ "tax_policy_associates" ] diff --git a/sources/harvest-uk-ab2026-2026-10-29/README.md b/sources/harvest-uk-ab2026-2026-10-29/README.md index d085d63a8..592b218ea 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/README.md +++ b/sources/harvest-uk-ab2026-2026-10-29/README.md @@ -45,6 +45,7 @@ The AB2025 harvest README's rules 1–10 and its row contract apply, with the AB A measure's own parameters are its design: rates, thresholds, payment amounts, dates and durations. So are spending allocations to programmes, financing remits and contingent liabilities. They go in `reform_hint` or `conditions`, never as rows (the AB2025 red book family kept "£9.90" in `reform_hint` the same way). A worked example's inputs follow this rule too, and its computed outcome is a row. 9. **A component of one measure keeps that measure's key.** A printed part of one Table 6.1 measure carries that measure's key, with `conditions.component` naming the part and `conditions.row_type: measure_component`. A part can be a fiscal head, a sub-measure within a line (the tax-gap package's items, for example) or a nation. Components are never summed with the measure's total. Proposed package keys are for scopes that cross Table 6.1 lines, or have none. +10. **Totals of the Budget's decisions take `decisions_effect_on_borrowing`,** whoever prints them. That covers the total effect of Government decisions, its direct and indirect effects and any part of the indirect effects, and its splits into tax and spending and into scorecard and non-scorecard decisions. `models.py` defines the metric as PSNB and keeps it apart from `revenue_change`, a receipts line, and `ingest_obr_policy_effects.py` maps March 2026 Table B.1's totals, its direct effects and their tax and spending split to it. The AB2025 harvest kept several producers' tax-package totals on `revenue_change`, the metric its README gives exchequer costings; here they move. Single measures, groups of measures and a total's lines by tax or spending head keep `revenue_change`, unless their producer states the effect on borrowing, as the OBR does for four policies in paragraphs 3.20, 3.21 and 3.43 and IFS does for the Budget's new giveaways. A row giving the effect on the current budget deficit proposes `fiscal_balance_change`, and its sign convention names that deficit. Each row keeps the sign its producer prints, and `conditions.sign_convention` says which way it runs. ## Families staged diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/NOTES.md index ed1e29ad7..f2ae6b9b6 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/NOTES.md @@ -132,7 +132,7 @@ Attribution: 263 rows `own`, 169 `restated` (OBR, ONS, IMF, DWP, DBT, the Britis - Exchequer costs: "positive = cost to the Exchequer (as printed; not re-signed)"; - savings: "positive = saving to the Exchequer"; - yields, including Table 6.3: "positive = yield to the Exchequer (reduces borrowing)"; - - Table 1.3: "positive = increase in PSNB" (the opposite of Table 6.1's yield convention); + - Table 1.3: "positive = increase in PSNB" (the opposite of Table 6.1's yield convention), except on its two memo lines, which give the change in the current budget deficit: "positive = increase in current budget deficit" (README rule 10, since 3 November); - Table 1.2: deficits and debt positive, surpluses negative, as printed. - **Periods.** Fiscal-year figures carry `conditions.fy`, end-year `period` and `time_basis: fiscal_year`. A multi-year total keeps its whole horizon in `conditions.horizon`, sits at its final fiscal year and is never annualised. An undated figure ("by the late 2030s", "this winter") sits at the publication year 2026 with the printed horizon in conditions and, where the year matters, `parse_confidence: medium`. - **Prices.** Every money figure is cash, as printed; Table 1.3's rows carry `price_basis: current prices`. No basis is converted. (The first staging tagged the £595 million administration saving `real terms`; paragraph 2.14's "real terms" belongs to the Spending Review's 16% cut, and the £595 million equals Table 6.1 line 63's cash +595.) @@ -314,3 +314,7 @@ The coordinator gave each scope one key across families (tranche 1, task 3.4). T - `ab2026__defence_investment_plan` → `ab2026__package_defence_investment_plan` The OBR's Table 3.1 "Total effect of Government decisions" prints the same figures as Table 1.3's "Total effect of policy decisions" (2.1 in 2028-29 in both), so the red book's restatement shares the OBR family's key. The Barnett key names the scope the Welsh Government's figures share (£450 million here, £448m there). The Defence Investment Plan crosses Table 6.1 lines 40–42, so it is a package, shared with the OBR's DIP funding row. Table 1.3's Tax and Spending parts stay components of the total, as staged. + +## README rule 10, 3 November 2026 + +Table 1.3's total effect of policy decisions and its tax and spending parts (18 rows) already took `decisions_effect_on_borrowing`, and its two memo lines on the current budget deficit (11 rows) already proposed `fiscal_balance_change`. The memo rows' sign convention named PSNB, so `stage.py` now states which way the current budget deficit moves on rows 138–142 (underlying forecast differences) and 161–166 (policy decisions). No value or sign changed. The contract checker reports 432 rows, 0 problems, and a re-run of `stage.py` reproduces the gzip byte for byte. diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/claims_staged.jsonl.gz index 2742cf735c797a3c25141397af4cd98131b17b3a..c9487174a0e7f6b6817eb11dfcbfce7fb65a9130 100644 GIT binary patch delta 32592 zcmV);K!(4KoC2Yo0+3z<>-!?HVORox&>`oR%|L1xNP!p!eqsea!G;HNQ?K zFh-OxCSUX;_vfI07Kp4Bz*2x@-h>i}cae{J0Fh>Ma|sd~N$Z4t``T!S;;s~xSq|f0 z0?%Z(DA3Lf^pZxxluWUaI|C_&B@bMB!Y34gTasRn!{cWLOllBmneQ$?NZG0%1^${E{x9AsTNoyVwZeu7Nt}6> zPl)KkFPqKeFI)9L|M+LCQ3QgeNcpkBOAbNpx*kKY&JGCHIa>s4)vK+>`5{!5J(S$4~0DsMT_y9Pbnq#3k z0I0E11$dDDV0yM?N+6bp(eRu+_xQc#F8ml=|M|y%`)qE*>H9hna)(oD^F*gqWk;kb z5}3ezk|E}xAiAFCK7jmeG$;blxHHR9i15Rk4?FA<8`V~GHyGVySu}lR0^AcTo9Sl1 zNZ5_Ue;jF8J@u{;)|l`ov{L(1_d%Ee(-4XQm@P&;l5_!&w7pb!Fr6`iZVYh8?t}AS zdgB=6K9AjEV%j!sRU5VbQv_E~3j}o(bp-Jy-BOtOd}g~3@I+JWX*qlW7bKqlEWGXp zzA%9g=sL%Jz^|HJ_^R2vDFSOZiFL^U7Zh~Fe`7d`2kTltBiQ1aq=76f=>d13_`U!g zcpt#|$m+>{o!s&nf)xni$egoh!E$clv1SUS%BQY9zKMgzOZcUY%GcY=o#R`#4$dfE zDBm9-&Z1TNSD_Pu+~eE->%lhn~9t3I6uAQwK4gd`DyV7sFfl ze|%05oV-N}#6K~Iq~->^4lfb9NeC~Qd zq+EWqlO4H$wSc`shT!M_g{~RtGr}MIfA!^O28)5f9+OS{^WQ-l<+8)hdo)FO=3nOU z8|k-6(T#121!@;5dWG1+HWz?DI202P2eodvdH+$p3wtX@{=r-#gsZU#f0XCmgUHgw zjd~sZQ^MCoq6ni0nR40CbnY0)DGW3K3iaF>`*#yY=!v%@Z&+;VV3=>6TN{T5fB(d* zMlj~U{DL=v?_XBmR6uTN%5mh2IZg(gJ7#j|(Y;dG>q+*-dXkIfBuLP78!+8lTwkUK z4+MA-&|M_O5jh!_W8lXdP89|~iV?njobEbVj}|eZJ0aOIU#|{FBA7 z^KK!K$8tvYV$9J&eec@$d|ZtgMgjBTaQK4WSEzER*@5Z7-1OFCYE_%OHdmgpk{*+Ic@0PzhUj$vZ4bMaET3nw=VDc=nW8ttAI zzJ3 zD94R@1>l2ak#<2Cq+~F*YP0h+L3d+NWt7^c}Yk0?_P;_|Bj7a?vshmH!+HZgE9z>kfX*# z9+f)*>NHx+7Pg6hFP1Nz=SVS78vM7%K)rVY&n5r6$sGQmcJOb|GyYo0H9#TBwKzwN z?rg45TqNvBxkw6Zbl7OsrQ^Mx2Dy)3oD*I}wvQo>HO8%JIp^e5XytY)w9XB5W9FEH zZ>=W_kI;*J$?)D&*PmN~Nx{f;q~#U@DvNfcxKluRblmWNLxv9T?K^P!OLC3FZhPo? zhBvj1VCr$7PEzoV<@52i?!em`gg1ildnXM1M1Y-gdyWDa6UhxG2eDO;8JO@s9t!LO zh=fPK_Ayv0cB|nsv6okNrIoNOI6lrSiat$TFBAd6d?FoVrb?Y`)}O{gR&bXMbLhk7 zA1)m+a>n_8q;bAG!ksj1&+%*gWQd1>5iXh<&8Hc%!7KIDkMLCX5eCfBVkI>SZiInY zctTK0-V+Cm$8`PK#~**XS<(1)b;n>Z7)2Umtv>N$_y4^d`~Ti?_x~%&_o+$u1sn~I z=;=1?kqd%tj28hQBx-;$7aKF+5)%GhT-@uPg%uZncllfjnk%+%*nc+f5GVymevl%f z@KFW}z7)d5-hePMf9VH&CQ|NikvcF{08m#Lsu=gu$HCi1QO_(vozHBVt@WYybp4Ye9- z&uXWC+T84zwB3g_z>OyO(l)GXNc@YNK--ndOe^)6*h1PU9y5N=&CwlN+2EDP_Upnz zfiO%a2As4EAO16RzeHvcVdE1P?vXp^slfC+JXXQyl)7V%XO?66aK5D94Hq6js*lPN zJvW!FJ%ZPb*7Th=(H|tD(#w5s&QM>NkO>)o!vM5kvUaW0WY2_C{M*-VJ5?5Y`(qnMh`^U^jJ7>lK>ZV^$2) zp5bxy+t+?Qg;r5aRSZ52b2>58~xVC$1!Ang{;8Sml+u)H2$VYgf zyf_EWtx%P+rX<8q8vXhXVL7QYb>^O15#hpYYr@sk9p1ZOd45pE$AKJfaMFg9xdbUO z=n7-%!jkj9FEFdlc1`D0$B%;(Ku;4}z}EA{kKCS%!@``qa~sR)0EQxt-f=!>CPPWV z^Mi5kD~I)ecwGsUTre zQpBd_JtqqgcR<2@%eF*?S#(cg_~YL|^) z+tUv{+Dp^^-tt9BPorzmGe2Sb$h51B{*oL8NL$mNP=QfA;amH}7v^*9gC? zI*G(;5z(ulMGD7%;UF^+d&%JtKd|7zrIGleh(@w_5!KFs-(nEwIJ)kh$1j0x9M;@$ z9bRj)AEHB2RNce!slW&Pvl0%1Zki0L2htSN5Mu?2)!dJhuKxYj=?wyH?(^cr>9UL#yAMG573c*oGwR zRvGTc1abI(8q7`xW|`MWV_hDs%RH^mS+&uu)eFGFT-tUyP+86ID-bhYGO+xWqMjGg zB5IH<4M;w^HX0oBfa9an`JB~jg90EilTKX@6b^FYAWm}0Q1FAts}h-fb`edn&l~wt z*F&RS9<)oHLJV=JbqavEjS6!)u-0gHasnI8ZOM>-?JY%MqG4A)P=ng3Ky7s?G@j+b zv(?#yRV(EDnx-HZWYNoeNG;);uiIdlycvIASyiVNC! zwR3O0+NIcdoh>k(G@7lsf;1XOwuNqt7M@2bUt*Xr{JFjG*TULlForCVik)SmppIZ_ zdE+X7g>`JWThv`TV9|VZYVAJzeaanXj;0n!IrT)au+6T_P@p_(%w_PSdkNTpeF%1m zt~YD#0o&RN8y*=#4i*29HBh-I4%2KVv)Ed7Y(Cx4F~JuY6D+SV?k&oeNor>&fu;Rw zbC9+2Q%wFvA4=4o*o`A%JH-|;J`kbB7A32H#zy__KbU;wT*=l0jocc!HFBR4xp#;P zY@cniZ>wHyHO_B?+t4tnVN%27NfS(V$irprRMK#W8ASapQqk2zIlO!nuM?U@XDv=g>T2hNQjS0)(`QOWb-4}K# zQmhAh0c7FxBU0TO(~Yb^)Kz)qegSZKeJ*q-hnpZfT?}tabX7te93E$eXQ)9pae_!G z`D2Hi1F+eywL=cUtaAz2oT`}8R&EVW@ta1i|5P3l@a)lQ);cZr`N6Hy0Vu2)mna^I zV-rjZ$CfmTNRYfurQa6c9h4TjyXs< z<-=pGK1Jzs+jM~S+CWMc77m_)>IR;yleDF~nyTI!P3qWvaIU8=Q6w*hke8jAWL@+$ zwP=?A?JVb1XC5jdIN<2zj zvpdo&jGTC;c$*;7x$3_$KuV9Z=J%~Q4*;~VsF8_gCR@p~glNl8qe+FGD37drD ziHuR*mH*2~Es->w@OY6757DNn5}Bst^bSoMpQ-&Fr)8DMMINljXpQ;A0ucy#rLo2; za(7niJQeKFQKTX~-_l{e;kHr}S3Kz*mJ%&y`3V|QY(yCeWxZF{7_2c^V{ows@lIxg zcv&&{QG2<(wko9{VP`7VN?uNk-qT**)%Nm>1npPLxG_0t`;Da}j9rJbLH(&RT|c%DV{rSg#0Toz4FNthu(b`_!Y&|EsT5r;&)YEtwIfTUw ztTdtm5sQ-PR_mHuY{4Uxj5$0_|z~s_3?TkiAcP|6%Bd+ z^E>vt>w(U`e^mng>msmGE|TSiL%UnIp6hVbKXnTAn*=X4H7{-CrL;mAiHb>nSQdlK za55xR7WeROky0FhhAN=RY(t3*Km%{R=m&y-#A{*Xc&IKX$TPNP6XSs&vtezjE4rWA z6@985yn?P{^zTLnrV3nsY8}zHU(8X!fUkx3jy6SDe+wn*tR2CMs7B@vQLTiCg77Vz zF>$=@I=5ACK|A`t@sV=K#BeN*z)#E}X_i4#ZNP@XM=K|8D{O86Al(^aH!_pUyB+y` z^FF47Z@-MD{O=2x-De3(_=L}5AH=i5UCK?CQhpw;!6JT!-Kmn_o8LdIjN-RB7Ir}^ z^Cs?Sf7IA7F3c()Sw5G=biRlGhtZp=v9-@`VImwv9SJP~uZ}^E#+*r6*rgA8ku|6W z2~KHgPT9;UB>19y5r@&EO0b_mC2*Yqn_$l5rdn1(N|$S7h(xtPQwYm&8H$L&g^D^% zeEZs|Z~t|>?$^2vSk6|R4iDjpAwJ6C!m=?ve*&R^-&^hiQxiaV5e9M8J|BUmsj^T$ z+`xp*qb%32B>1<#q!3ufsdbe5hP{6IVNb1thP}M7hcky!DxG3re@`vM`NLj4bM{01*liY14tu+2Ih+yf4ayID zY8^D}<%K<*In?oHF|el=;u2u5nHlz)R|b2VTGL(<}RmiuoXW>O5U!(NM7p{XwzMRNMGe=t9Ed=?(mADAxf9qyP zzV4MF-|m?zXM}vs@*|&G2aSArkq>8%w7eJ-U$_z%fqb3p$ahL3-mYOf>);x(@S5GiCB;JR zA>IoMg`2opYiKX=Di$^}V`1Y8vGA%L#pT38?K9pB3#H?@QR{2J@gDtBuMrEc*>zk} zEYx1)y|7TYk_*Ga%kd?g+Ygu|hRKuVlLtrPEsA5C= zg+);Nfqpx?<1`g*6YLH~R~e{g=FuVX#-2YnUr(Yx$`kDPrU zdYNNAdK&aK=$|X-mn^m~3+U^hmi<9rMz%By%Kj*AWJ^xaZ)XPmwg!C-`ezIJjq?M2 z9lf(Z=!+1ZqC&e%8^V(l^v@)!JK055SAMvdivBArtHV|ve~(pM=_s72;a}Ia9aiHL z>zU&d>(_#Cx_JMI5l+WwZ2l@uG4>Fgwp{Dxe-ppA5`l_HDbD) zLJT-3*~=BSt_9?DRsIu$oOZzH202-Wzp!p|7gUEod%L-6<7z-o7vMiJ$Z3ajZje(Y z_ZvkOp1z=xIQxz4g{K>*Ej)cpkkhsHPYiO}!I~T7WOe;UL7|hER9(N3Ie?O;?}U9t zmLt?O+g>tr{I`WqTC;~8l>tZTtPV2JK?bS9p<@bmfKi;dcnHCuQ&hg_<%AI6oMc2z zn!f8oO&u|Cwh;p%YU%)iT~SlS0}Q%F^|!p7cmSM}tf<+&y1kQQA2EM!uX#D1dcD`w zj`@pp%q!H?e)wHcQ@Y&;o!UUV-Pi1zgY0(A!Ihz=_OhR?mpw#H?Nr|tHH8ly)V`~I z=xeCi&5oMg(xE2fU)U?txx*=Z>)hI=GuCc(?N;9iq$%>EJ?*<<>-t7Z=Mr6++~?Z`iCNB$A&hEHsm(X!_bhyPd_snNrhPexgM&M&LHrA zNHf{3=lcHPiqE8>EFp(6=7DL^ioZ5(Ks3{<;+zY-W$+_dJ2ya7V(k}lIjcf~7)+KI zF#7qzFBFImgr+ap=+T#~F*OlPOc8^h;0-*7=LaNHprrOAnGT9cN+M8KIu$ zBO}&}xzk{;!Tv14zIQ%guOl<|1bY#m(WtiyDk)jo0F9hr-^yH0Qp}wOdkywy3HJT- z0ec-4vM1P!*pNoOT~w{0vId9b1p9VoD~^~u4fY!B&l2nh=L7aSB4tmo7jY?n2w3gQ z4ot}n_U#;C-_~HS!Tv14zC7iV9tC?H4YMcM%a|AhtP2l{$qn}P9AID9V6VacJi)$k zK47mSarOjz5s!m*=L>IWlG|qA&0(`YayOIX*8(|m$g=&~@+F>!~>hLT8Ob1PW#^9)}Zq zA;`2b;QA~Amo+kJWV!-mYGn6)!Q_)4XnO>W$;iN{$l_8e_4YMSoPo753fEHkyWhm=oEPL8Q^BL@K9)(5IA87P&NvI zKqy&=F0fED&I3d)X&ewT$WSWTgB47f3AaHlK*0L;wb4-mQ(;&m60UOK?jxKf?jwMG z``T=VLM4E#j*o)l|HIUO8cpG@w#|l|9vGM&j*!d|We`B5*L=@U9bD8E(cLVO1ILB>p# zb3u=yxmF@iANv~|8KLvyUqg{a3B`j-%!To<%;V$5D3J1t-|lgL2tEwLp2S!ep1xvW zjA;DlAOF?t-BehLAm}8oM*%x!))J{*jZUpop46_^b3{PrmJ_FhGRAyRC?1!9XOZCF z6uv7KM>TWJHw$BM05Q5J{7GkqEi9 zJndB@0Y>qrk;)WB6A)caKl{ zD{H6Eqe#0!R^oo2R;#Ec1()OZX|AlPBz&IsHJ!-UG=^0$)G8d^2_F{$?NWN&4+SqO>lcTI+Fg{> zT~snW)Lx;Scqo&2y1FdqDjw!;T+}#iqPE$8>TgAit zRf?OJhlkp4lhbcgGCb7of0>+kD4j7)?TpcQs9i7nyI#tLhuXuE6AzU?rCDoef659T zHuB?PLi|PPU1-y3GHZk6e!atF79!uwu{<9UW~`3+RnR!q)EI~ z#MG{rGtAW4f|%MBvnyf>pG>u<6LVG(vzIqBhbHk-5mP&7&M+fqe+y!2XU(pNDZDn- zdabLyHc7sl7S70;Y89Xs^y%dA7Vp%*#EIYv;~6=GAP5 zOYQ906)vUMr?CCy<#>J0ZGZVQpSjQ3``O$a1q?s|)|@Hu5-`Zdd}58PfIVAm&#$^U zFIQ3eTW$v4^xvP2e>+L8!O*l#2et3GAY7!FY4_8k(4AwtV=k$}tOXb2k;)6^8p}`P z#JW^0?_UL$YsXlTST3ApZS5;t!SZ%qS6Ta{p0YB6`EmSfmkQ>Cs{nKDvnvwJg&(i4 z-FT}NNxQi#lIorLDSUqy3g(;qgBJ$o+Mie?m`fjHqo7oYe+zOlHX8X$Pc<~2pTIeJ zsd(PF20Yge&m!?WDkWIgPS6!Rujj|}Q-+Ovb_3I99InBYpD^(RSnZMJ(-kG~Y{t+j zj+;-+A&IpCQhR6Kaw8UV5su%W8_zz9cu z_6tnHY{)$!f9nw-a(`|Goc%fqT>LhM;x+b8{;DchUy1`vguNJ(J_%@-U--;)#u8$E zHu2mU6H3^vHJCqg=lpw|;aYCr?J=0;F@e`U)<@amu_=suR=c&c_^yGRStgW0x zfirpeJh6}0RP)5;^F+T?JfZ!!`}2hG+V;|p5_|B(f5kX&d-dFnaj)S#v5!Mm^Tg%z z#Gq6>p22jM2C~&UY3jg}UMpoh!Z~dxoCC8!siK)Q3+QmCM_E8eG-YA|*($9~caCdZJ3pOiTRF{7Cy9}| zJbuzaSVi)a2-E7<+B!@tg`fKEocx3ndFA|crtR!BKb<60>+<+X$AT5fPpVa3M@NgL zm8k9HE>YXLc78h3HhP+$P7*PHc6t1yL(+=mr?5p|T}Q2@cId0;@6cDja*8_F_I;Y7 zP7=FzaTKM4UcHV`6=q!t$gR&Hc(U4Nuua3kD_##UXc_f1NnM| z=KfqzAYU(U@()erQ7LL4QW3xM7Gmlxz;QpA$-wAG?RFW0o#g|j+#Gv%-tl(a6`($d?iMKG3u?A7o{0n>sGGPTi8FFnV=7y=cNz{w^ z4SqmM3>U#T&`JwqUZO1Fvy@7G=d+PHw{RpMIcf$EQ_JvwS+VV5al1)xOC$^#&LqA6(i0i`tDDp@$fC_~Amu98W>QvajG%D?;*crk<8}&NHV9aC;2JQ+| zIK9FirjvXzEPsQ}(QJW;C>0Nr{n{pM}~xvQYa1StxM`0ys(pv2Ft; zeeTic9?dJK=M}{h=4ficGLF&AHCcGDfSyohl<}>*3xBu(6kmmP^w5|VYMQX2!5^M! z`8>XbIX=tK5a|7qZxCTt`KhN>-;H`&M<<@*aQvt!JhJXB+mO0vnB$qH#_Sug)rNu3 zN4uc!ERMe26!bkcP_{y#Y_v5}ZbV9CVMH<^cK#~t;Yo{r2j2sxt{9d&gH9+22`2;Y zI!sTI=YJeT%5oPVzkqogw&Q?UJPFGrtCk;lVqY(XHXlUBMk`;ia7Yx=RcK4YK__Q~ zZ)_c|)f}{?SXOM`EXSv|&nVsTDmGhz871UUuxuH)12{Df7uWMI9fvy_hgWd;2lxKv zdjYVf?YnGdI_53lESx|d*lUy#_DKwa!S4VH9e*EjQy?OWj7Wy))I&LlNqB_dDC77S ztnZrnL>$j36U&`(8A zj|TV&%g9GsXA}lLo7?V^jx_wsoI5`E*)wHwZZXe&V9lHCfS!Wi+sztVKz2IwMzEA$ zjDK5mu5zK17;93tE+51F7J^wb#@?J7Rl6aowptohSN)&13)lsVf>7q8MHz+FudF7Z z5$?{+I}W=L25)8opQE$G1Z)JzYIN%rP)sS|9ON3<9~8!#F%T-at8MWKCgrB70O94# zy;s?74j22K1&rtY=T9#IZEbEZ@go|67;Y)uJrzlR2CeiQ+^K&s=}|sD`Q?ZNmk>4@ zMOBU>KFlO$=R?|P0%&|2v_1`M<4?Wk21vM>1!aXyZ%0|vaTcb%A*RPi(2Dz6yLs9= ziFH3i%R<(A5HQWQWs*r&!K1vwe|#Ujw75M^*$C72laMqXf2b`W7R&-8cVFZPL?8?? z4n+&W%U}8dp9y$#P>dK_=9j?yFx^W!Xj*{TCj3Vv9|saknK+M*7x*tU2TNlqnKzWI z3p{sY&jU?}jc~m11bo9NKZGiSS5Xwrx37(6%%iP5F|}SR1;0PbyNAzy>@j&gj?q{O z-}JVNk9*ltf6dNAK{0IiAuaBuSyH=iZZmJ8)`Y1G+lr}cj}7i{?{*0QPuy%Y>UCHm zXY3oM5xBgm6r_m!n`?DIzWW;k7Qxru&iijbQ5==WeQuJ?Me=#B@?ySj`{CGIN=xe@6;frLQ zh+B&`(i0Qb-NHQvyht!5qXq0~V0Z04RHXdjhp_qPUjo_SPL#|cW&7*P&$M5+VZ}Tw znzLlEd$w%$?+e(9i={@c!T5TME;d8oVM2b;OyV!*bJq((Qwl!kN2Q1$%&3TJvtAK~ z@WovCBoRYvO{ln36h|rmNbiV?7dDly0QA90L}?7M*uR^w0rteOL6c85K|(f62OyaY zCYGcPdsj1yhG`Ac8m1qw<`x=`i?i44htARLCO~@o`9L8+vMIyyt>VWw-XyHrAcuzQ z(CgsHc`&y@lhQVMe@TH^H2iiYH1{Nb`Dspen@^a{PIwPL?o^0@+R0x=H|xmUg=Vd_ zn!6Cnb}ij}R85$VwrWSt3@TYqG>XWN?hI;{uzUpq!uQ(SugGYj-CkfU2ZE~2sL!U+#8^ynuC!0JJGvgp(`Qg1+NQ<(PUULUvd7|RF9Vb0-ib2Js_ z0F)~L^EK%YW5J^Zl|>Ky&>@iqw(sOwB6x7CvOoZcR+SKhZXXsx$K8Sol~JQa9g<_o zhr{dRkH(!2y08%x&Q1Um+!2O&^Vo)eUC3O~l@t2ytj#^!PtOo*B^KVo!EG)=9_Or4 zC(%nDVUB>f>wE7Isn)<)q0YoJMJ5yff<*h#%B7eI!8ilyMO zUh8DU$Ji>D-_M!cg1&nB=A#G`GT~^v&7M3o$DUzZ6Hc>gTQitG5e!)bihZ7XoLB$8 zFueeo6pkj8?0^_j(nK>$drLHt?U+6+kV?#Jek@-9pNg2&bf#&JXUg3l(INr+&^AZ?!=uQtK^eB1i<${K56cnF=O88O<#tywniQm!2HH6n3!~ z+*V25ty)t<@Op}j@-Rx`Uk@AXY8&@vVElu7E*{=T z{vz&TIx@-vtNCpTu&A(BJe7M48#5|gF>_%>VHn5dvp@g%FWPCUO(kD_82&&-H6+72 zCa{wVJ18Gg9|{hd807MP#a`=d0To5cK#?*!bE;^egq^Ia9_*7*J3BwKWj&8LUknk8 zUXT4cx#csoD2_N*-($~$<=n#ZMd3^CTJz!uG#{~7o)st~cy=QyIi)rXFf^+^XWvnl ztW=ZBJ2Zbbn+!!xQNY-pVbP*doC|VRhKmmdhq1M2rU~je1}ICETN#tEpQn zo+F2m*s+(@H}NgrT87t|?dLi=Z%dbT&}AJoh#!9e#NS87gV!pt0leELpz~B@(|R65D(#^iKCyx*2_UqyO5ZpFYr@DX7`VuDzA&W9 ztNVYe2?+5mHVqjALcNq zRwUiHDUUsnlUWTUq37G^`9QKTaW2kzIX(h#od1e z)<{AdBCKJ&-NVkVuvlZY;gX8lWl0~EP&np8m{`jjEntE%XYP)BdAehwnKLSK72lt( znTE3@lI~?Z{qB``{qNWq!y%~}mAS;MYf&*HY^a4NKae15a>rbcMLnK2x~>q47PBLi z4{Ucr^wza*)6hFqo|WUfXpPC%#7=+N5WanF)tOX~3PmHxhD0KDmGq9GIpYoo-F+K7?z;XtvWt5K!b=~M9`s*?PW3(q?&tTz*le&o5^AM zq}cmVq}c1ENU=`>+n&98|N7Gn)vbc<%bMIGtjf@@&QrQQm1=Fs#%+XkRa{HQM>U8g z0f<_!p%qm`u)r@pSLq4tolSoj76ABbtcdE`?44)M_=D@+i4p|B2y>Y(fDMlMgDMU{ zq=1zGWJ7rzSJ<%7kHQfsI>C=uL>CAmK7YYPk@1i!$iOTv6t$kEHCdRVsTS=RAW4=K zkH-)V*&SOwLv|tfI-$}$=N<`Wz~}Dw|iS}8})+s^UY4K*tLJo)AJ3q&4ukp zZkw6FeOxv|{ZNB7*lV!YU|&A4-@FVb!~C%8T+V)TjpOzKJzB_H?(>3?| znqHDY;`^|_YxAu@!TY-o<|615lP?!R_C!=*I#9TgzA8z$4YWk)tj z>TrZog(GN`)F^pMluR4EY>YHSqs(bHm}$=ZJPseRLK<5&wrXtE*jfr~UCig@#nya* zQfC6LCCzHUO!R-)w>qUkL;BthnfJ;M4gl!z2evIafa(;C?(8ru2}Ql4gdi!MkM)8Cm zYv1|;gLHx^J)jXq3T=+=fe}Q{6{V&ylMD3WBlYnUeY5jH-XR*MG_J2>ig|nM@!Jq< zqTNd*Zy*vMOwYDV;Zt^8kPtYDFu73_JD=DVDw%qV34Ru^Lp*GT#IvZ>+fCz)UL{Xc z*Liq{7_K3Q1(GsvWnm*om4=d=%fx;bW&;~Kj$rb}9v&Gt?`bpclaWF%e^6mmTnw)A zF!4Tql8oijK;;Di~(eG;P7gzqy@qx($bXQh0_r0Ql2^?Iom$0PfrZ;i%MM1!lT51hUFOeahCY4 zl+#2K;#Z@$j(DcfEP7%VT~}z$rkYJPn`$;a2Aif#z?rZrtWrK!e}xM)t6mGMf@*3u z)oiNSRI}+3*);9F&Wu%ahV#o6DwzvxvXm&(lMs^@i4|iFKAYR_lJl)0Rx``yh<&DQ zyASBFP#HzO(doZXcmxqtWgycfAsNQ5mbnWu7LU3bp$KA$71*{^ znD34-k}CW=42MFP#v9K8D2MbFjm78Gx1d=E@swMXu)!GIeL7VVRk0!5{qD$k-2 z(=^}Rd+C?MTUD_cIFriDQ2za#!CfIZ6zHzY&}ddd9Zpb2pItMT%dT0{M6A5vohGXq z$2E>?9M?D=$MM8mlpf0)xv;!({OAipHAAf-p(+S&zb&HBo4yI4h!%Dy!n7gKnry1=wu%CyNz}O{1ATRvCVK zdV*+lG(%{H&CrtP~blv%}2$KjKiL-@VQ?@NNQ@I^H{R9&alnO7L8lb1zx zf3QP~jj#gR(S@I3vTLKMQW~Pli?sqWw@-l8y6N#MR=*6f&tBqgTx6`Q=FdbWaZM11 zDv1MwrL)GS%$13xZDe8BWb}YO`|H~`HzY#T>Ik#|cI)24ab!MgBVI@sUA1_*_be?|^zl|$p#K=RWHslmO|#xGxRw~j}6)56@a zodRn;c5@VY?4AQy$3lu+Rb0T~Uji09xJ3Pa1W3l12c|_!@F9Yks>n|sqPm%?iT|uaI12&6qvA*e>hSV z_Jbo|Dbw3(hIM*7b=C6#PY9^Y(MSDibC4+En=koWL<$v53jK%_bUV`YZXURE#;6#Z ze_?{Yaf3WHW#dt*=Z_dfJvl&&VHrI%+@xe*(*7R=$z3=jf$)ubR3~$T*2YSzQyUCo z1YAuJOsXnWo#TITJxMs2QjIk8XU$J(7rgx|Q|hHk3+GKOkG;pH9z081O7tF8`5uLB%)13{%o{nnVKh1?%iqIg#CMvhlmTjHMX|50JI0*? zR+#f)#i%J3p2Wz2*JZx|(a*wGD$ittuKIqRisZhYqUYxx?C!&HuI}$;ORd;8AOD5I z&M+kcZ(unx3fQG$3}JLT#=D2>;kge69QEX`le0)9f8c%rH{ZDmMd5A?U>&wl&l2_i z)aPP9we{qIL!=TdQ*Mggi)Dizcpxo?SlKYIVyBZv5t#x~E_tjFf27z4!0Q=qV-BY*z!UjjzZx)W3p;KoQw&Ydg; zW1^pHf5>HJ7x;#7XJN@wF{7yqw1A!lyi-ckrle|MQ5W)MRvZ1O`b>y?X$(f-7pW^f zHB@Es%4|h(vJ)BB-;#+=%9*DITqT1;q~?#+Tg}IBy#t&;9;xY{R(o}AEj0OQ^3&v} z$xoA?CchZ@JysC+^yJs?1Am>vRB6YYifK$wc~vevJZb-HhcB{(`Tjt z1GEQ}!!48=p-&SSca`FselcZsz)vF0 z;CGa5k(nk;Hk~GW;QY={+_H0X@`U#oe@yo^Os}^fQ0?SlhgzMR)>x$*KHg5iGp#U@ zB~?6VpQFC`S!fM?g9(dMYcyR2cu;&kvBHpvW~~+O8|0|cfIZ){qe$p{2(P9^jxpF2 zJF#PSd;cJnN?N%pl|<8B5NfYX2DT?VQ{>JVb9ULtyuRbUMTjt9)R@lTA25GKf1~&o z6#vy>jYQwRjfhM^%L)K-=iI6do{BucMyu9r&~Y9PPhN%bzb`C+2NX&v(%Cq;(?NSomwx(Py@+_L@O9RE(3Ao5SGMf1D4C<@;J2p? zAW(0#gT3bmY>`B*Gi5+D zz7WAu`gI)YLl0pOv?6^g_KgP0>9r!1N=&MmaAN?x$AN| z5aV2iS8^<7y~?JU4Fh7VuBzk&vR^mmA1B%#)m%}GPhSsGfKO_ zh6WyuGoQ(#jyzbN?-2scQ6TY%jjfJF*~ zPs@vjU1)VA(u1*+lansU3(W3r3EfiH(&g+VFNi_B2_P5{lkFGYaR_^5ok@RZHp~b3 zpORX5-h&q`hrn%v%na0(->CdFI~*;^%gx=BEE|YzKOd7aDCRHQk)oKc6{bP z4`cyBJ%Fo|3;KFr7E|7o&Whk33#+C$v6f^Ia4L|MVWc#Q7_UWp`1)yCHO%StJlyJ6 z;@hYATW3kjSeCcP6z+`7yd(o3vmOdi`-boLy3dlx6O2gqm@|Z@MhK4g(O8HRk$wIcer6rsNeD6>6>? zkcyoCdm z1p0=6hM54j2?Tj5(X^aP^||Z*&ZhIx2ab%dkS7Azx*M9fNn`1R`c^HDMmRQ@B?CD! zE6!UK&&uWN_F7xLJ10yA|J78Fa+kD@m1DbJUAp?bqpmEZAa_ z!T-YnVF$yoU0LV9QTlD1pWJCmSv6H^#AynH|0oXwOE*J%iB}(L7`Q&ewgg2&PE?z6 z&$f$1qjMmQ&0-W73hTRMukTq9udP4{V5it20bS3Sre5Rz>wgs0v7DnA8#}u81bJM)k~o(TButKonvRlZ^&+>ou<%kYU#%O)~=O)4#B`=iMNYZ*5uRuAli9U?ZE0aJu0{MLVE9YN$-btW2+a>7fWrpJGr82}}GU`e=AT)`a zi1I+Xh)XGDxDyoJy_!M4>`2?U+`n zAwi>JOA+7T*ae2@b@3h0`MoR%NEKup(p|~~u!;{P=|ICD4qm?__JMc|SBDCs3Z1%q zaDCmcoHwUsW_Wt1eLFe_4bAufbjg%%zQKjGso?3JLb#gxhw0zQb0XQBTbiDa#k?;E z=rJdZ_eQi6JZF_CD6$MC$9I)1hBQ=53$)hLg&wh?Cg7L&c327I2Xmw-Ypzt1A=@4I zl527{(je==LS`5qW{hT85m_G!98*1n(hcVv%PFA(vLZMkCH7UYHlarVgWR0SC!yqa zz@-6y7+R=&;HLfYb!Bpdh2&&kh&JWOv3mS`6LG z5YPY2G4|~ajc5Wn5kxxzdMQ7A!`A&Q?54IN`;OG&5+)^yY(efeY3TGgrt|vn^RA^Y zrivmPTa#vOYegYW8bt$FYL^JCvdTy${TcE)ST{!mNm##n`Ic+w9W7SXYItufiwj%L z8oMblUVv(KR{o7x4lhYd`4$!Nx$BTYPnRynx9Y@?`BHPl&;hfeOS0Ae6=(~cRvqMu zfxw-?ts<|Lxrj$mt_P^5BF!bbgM($u*iW{I3iAs|UTKli&T{KbU(@U2a znY-Xcw|uRsYX-Q9el+KSK8?84=CYG8$AUt-JDHEo)baBSI(TPpr_`RM#5(s@s!Ck~ED_M*%E>>0rM!b3VHv1!9U=s1)YM zyOI9X_WUxHN%@WY${WNR)~>(-4Hwp-@(KpKy!4Pk@Ix5udC*{OUTL+=fi@0pmYeiBhd~;*N%%y`GdNeQb6Isgc>AiR3`F* zu^{;gF&>aT#miVcko(tr6#Es_(;~z%a^zH=)!V5uj zQkW*q)bdYstUnHGAb=r0UxNzJ=Y!Ugb!SF}93Reg8_mEwm2oYK8x8CXm#Dra+XMH9 z(Z)a0;8$`F`4wuXc;E%!o^}Ed4oH}g@?|af<_#l}C0YHKQBVuY)1uj~kc>u#>ptky zMH0szEey%_J_VsJ8X=D^jK*^7fo1CfW9AK2s51mkq#emkloO}TU)or9y~D(BIuqN_ zmaqJ9S`Td@Ml#!aYJW7gH%d%Sc>%Ki%dFU^ZQi0UOS1y~7k};sJF}n^ka&P5Ex<6| zT|8ClCtl7C#PNOtYIZLpwi=sB$fY(0jFq@5QTfY7Pz_(x4LgaE@DKJ1GkUf{-K^l# zg~CY0o}&Gkwsl3=>nuMaOr}T5R}({uY-rF?ygrx;KTU|Dm4m&;>16lFwm$P9np+&K zURH81Dw=lqBU%yyXUGO_6fdM)AOo8k}i-JVqjm{XqA+~FiF;5jE8a@mgJ}Rnknzgm(V8gfMo^x{2bZX#7dWA4OB-ZfPb)#x>GDRSsRICi^muIGi0U(Umq`jzs zGG!|VBxKXVP^%f=(Ea9UCysG>$B~!^Sr8@R=fqcZOU$VDe^Uk%rZzOOPi=R!GWLz~ z=rFR{BGnT`8T=jE2UpMv&zc4yZ{Iy3tC)#RfYk1B5H?r*N<>X(h)sw;lwDej1||F3 z7e?kc&cUCFzPJQ|tk2oXa)9+Z|A*_ZJ|EUDWV#OGNF2WNBby;a9OPHxruoZeU(n<( zxU*P!^+(Ct7$!B1!8Ujz9gFgOH5Gmv|6qWB@O;!>@52!K3e!g0KT&y%C|cYo6cr;w z5$%nSO^-0q9z8yh9v@v&1E~P1hL2FVX|=+_J|y62{=5Sv|VilVIBo3S1=8h7xHmA@p} zgn7VL2}$bopzf+9IsiBXt5&A=#l9?me0CpF*akWR??DhAfUNWArussIGWbZ@kddjc z%z8_SE{oPDfuR63^Dn!TkWzO8I|BSZQ(f_&UkYu2cUYfK{P!$x$ntB*3bpc4-s3}8 z?Rx=C*OYJj)FG&?z`mUhZZ{@F9+zLR{Uvr+{C$^0RX$a3w+E6)gZ}V>(1T+ zkX%)_IJo{qMNLhzBV3)1uXwJ^;*KyrJSX925E-Rq4U5QtsR zk{AK=svUQchn`6OTuQXLlam>1#SwhXg`xvte>JWLPg@ot@_I?T<6817ZV^qeuvWDMbkA1F&zDqA6t za)znt5v{No>vw5{#B$w?QvNCfe^~uy^>L*nR{GBVUw~?##+T&*#29}qXy_oLelCla zxTfk)`Qe|?%RlRHgRI;6O)isO?JWA;_{`Z;Y|?zSl}thz8#PlTnfs*H0v;I>p=bq2 zl2W!P2@6;i%aIi|+p_8J=K@#3-IXynUwcZa(!rg>8A%Q6u6?vtk#d}n!@1!4CcNW! z{G=2QbAatqFh)5b)6u~Ees}6LaM)2pZFoVP1BQwA%@6k#E~|7IKCv)B}W<}JE0Z-f>+6_g8;t@X(A|}fZ4@?>5v>X zNCdS3haRY;rDp|8;z0}M!p%!KlJ>(dc{-i1M8F)$$M}fe%iYz(aLQWUs1|XTpsq5P z+o%lVSY}kk7ON1E{SEYTf}GUivecMdL?TZ@ZizPI~FR%&GV)&s@_M*`*!%ePkEME83!3? z;?#QSX^^L7FOWtLIko&ZFw4YNE?jt9uTMqhdBVD*nMX@g1tm_jZ`O;~*_!dJB%CS+ zR?k$z5q|Qdghaf(Avr({U7B9J^$n(u9lDJ8-qm7RYQI`(J%1OZIjuICf!|O)>@2J6 z+alzT(aGV>_p{VC`MXUh5$VdXM{8_FyqScz4*Zgop6yOa zLm6vk%J{W493g;|p|!fFD6c?$O4JLV2;}~kV^8F}cQpuxzGyQBHJMxYlB}0`$Qj_^ zgm`r$62RRifRrpGs0#Mgt(|)QNw-4J(LNUG=e-_cHD#ViRfa21JgB&-L>K70_(w zC(}&SJ(o>SD>aNBCXol}LP2&UQ41}@E3N(I6X4!NgxG#nh8BWJQwgDBVYJGdr5rn4 zAmEdNi#n|nK@AE{HSxGW?vvHm`yR8Rh23fc_GrOX9r8~zy$$CbT&y5M3EnWl>RIG> zpQEBxEPp(B;&-8T^>hiIlv8(rW8r@h%gs~@K|Gr?{7QGjnsb>R?%JrvhMbGz-!&^S zKtrMKfhEHoXG&tA5Z&oz zK=@j!Oa-Qdj+!ZEZp_o%fct`m7pTRI+OF5EXs18&9f}u6vqdcC9MTa6s#i-%Ag}MU z($ynXG?yEN-bU7?rKW}2RSD(nMWN$;0BqFAA&1D_;=!Ae*n2TJ`HM3a(iW^o_kXSn zx}4^on{Tf`FUCZ0h-hDWSCR|lq$J(CSHMkoL8`=BH+^14hPS;;oLz!5-U}ZO`3*yg z*cQl`9EV3-sX7GVB+}A!0^SI(Ik-ly(&>?NHSfYNf@Uae-55$ETd`<9aYm7lfChaa zT65tmNxbMp$sk7c+ZnJIJ%}XI`}=M14z9 zC+03LU^KSGZ=Qhzesk%ma_0XES;4#$U6Wpl)X7!B@@a`siW7|{mG@}BxyPVjXl@o- zJBwe8bZlPf4+qlt3?x*p_wy>K{Jif`8((yFG*EB;sG8?N+{|G9KE=Y%2IOsoSjhYH z-v7Sr@+I5_w6JpUiLtg=adGAGI1vxoh5RHq98y>ChR{ANK=}Fugtj3lu2G^q`M~4F0HB@r*!a)642s_#1ehq&qT+d2bmRw%V!fxLVy7C?L zy;A|@{_0|Q@^Iq>ZK|wc2jB#vU&g<8$tTFbiw8!3Y=QXXYgx3s-ob9clzHKU$~Ujf z$H_b)AO7PFUniK1z{$ISLZEIUX4i1yO{_k_%?Ix~{*jLApm#vTF_Uw)cC$j<#f*vP zd%rot|E%{^CIB*<_VJqZkYz4lDflP|qi<}MAagGI3TvtsX88Lg0{9TA|3a(X=x4wI zYOZ%U){4B=76s8MhPfv`1S|%uV>|6_2>vs$Y<_)EJawa2q?H5v*W$Af#)ebDm8K|w zXxc4UX_!4=Q=fbtD2!z((!mszr$7bi#eS6HKvhIOmM*ge3_)=eK#v!kZ;*bjD;f-o z*dW63onp8FZ+=CBrD_}Iwc*E<8cOp!9mz7Ruv`H*e998!--xt^zuYO5wd?+0{wYl! z{1~wQkNq>{D8wXUWbw%gz7iNd_==kZ(;Xnh-!jlD3S*2krAj`3{vwcTbs*8ii<+AQ z*4T!R3d9Ajs43%wjE{|lBL{w~2jZ0@M?D@;b{`R+R4OiiG=X-bQ$34>UEi4NKB1Vc z4`E&zZhT`YICu2am6V`h1YR1fVH!ZT1>SCl^5hF$8=kSiC1aNtWWB_Pl;b6*5%oO^ zR}IlvC63=tPZnbF=%fk$!+A&NDJHc8#BTmn(NLmDIsk?L$x0-u&xKO@S-SRh@^5g^ zU*#OnrT;FJN$&D&%~cKUH*1LY>u$FU_|=B-oq12(jWjA}i)3>(&-oxz2fY$K5rLD| z`m5kmQq@B?>|Mj6lDn3sMFr)LEwu9ERo1ff@&qJ{g^FCHrhGqOw@|l~%mxD3zL-J4v;3+w0`|+A~D0a zFxwodh^C>vJ(L0=;J{I1(UX*Id2nvWFIYTptml(n~RrZ>fS0FK>ic z#ED_vV1aj_4=Z!6HlP@ZL1GT@JghQ-^+O&xo)bbU;4dn3Yr5pE%wYhsT5eNZM@%(^ z+$d&w%u`rpbD0G5ZZ*QvK)_-oX&_cA-kbK=e0plpYJ`=au?*wYEeTa2cOSYlRWUkM z#J;b)pBn$2D0;77oqW)(Hv*VUI_i%6Xcj%4ONowhBDv1U7Fu#Il(7UHNAGC6Xl0nT zD#nVB>0P5ZPCI5%M)&fEJSE6z|mlJl=?LwCfU=N}`Kc@dF7UP6I|NFIi#F@EIw6 zZ`9~8CYUH$>Jk&t3JrtMMhjb@vfx}ct=qkZ7n0R;1UJfVvqz&<`j&@O<5q=KrGu{Y z25H!B#yaW~8S?3o_Q_GV-G*Sw6(f4geE%mW=L#tgn==`oe22eltP3t%Ks*!oXVMNK zIBrL7xt$%q6TdnZDEuj&{edb4>;Omqp}tu{esbHF z%$)~hSWl>HnRr^7=&ty&7ya6cHof-dGQzW&X~N{SNj8)+y}m|5vMZXhsD^BR1 zHJZNKB+c?DjqM~2Pt6#G>{<^YYy8+~%5#pklO0ED#pb^P;&5;FJR^OLVlyFp55T)L z&e`}p;vxaS86T!pa2d6!zW z`qxPlBjIM4=k>z<3m5~k9OoJ9&Yu)T!J?i(%%~|qo?_Z_lgt&{$b?+|8N+%LWbocn z(~(+)2JT&{yM1p$>XX*=NVDpAt1CTAYp`vZ)`TOW#=aROE$P`Igx=N`F|xYVfl;Yo zL`OQxgFtRpa`gPk+};ycpm<>G7Ub;B&aT5b?v66cpq-gM3NpDm6Z-Bp1OsAh4+l!t zR0lHh$RXzbQbsw6XfU%zkTpBlV?3KJ5nx8*&mYd@$y#+Zfndf&lOws2iOL~aQ7H33 z=mxPbw4(2sch&Ar%%JgCHtib^wY?p4jp7wH~d?b0yElB+CA1#lZy1#V?#nmU-Hd zp-WAJ$Ek2KN;+5j6)2rkTE(^5#93vf6eP@8#&PpWI!{FvBs}O4Njl%fsT3!WJ;nw= zQ-@9Wjm>D-&F32VNUm9(ophtOJ(K!2M&l zc{En=$e^xD6jhcZLM{Ak_U$P`fTClxRejIAKd2msM+hJvHDxRy?+^oN_j41H zH6Rbcv^jI@MagdWF!uo5v~A4v>d67ruelKomXV}NUdzF917z`~G>pEivW|f-{PJPa zf7REgExUiU1yKo4>XV1C*qr^>3A|rEguv5-0DN9}FbOlkKPj;T6-y4P;0FycaB?fory$5Lu&EW@3?w1Q({VV(RSGBsL>d>x`BkHCjVy zFLbE+)c^*Q5VYxds-4fx$E}lf(V01`XQlKwt96CA@{bx(0$7`zxQtibE4R7y_1brS zhkc6K>E$QV5mt8+8yFh%H-P0DUdm<+AToaMU(%#EI!hS;tl85u`3q_*g>mZZGutndR(C5UHHdvU_mS+BD$(#&pOblwg2FIpBaI4zlMXY zXWVkYLuVRhJIM-WqkI1m>`VnHg4j$2ID#1NPkTXb0YPXSY`FwXOKPzMLYoi&(>|fF z`RCU5vz_}EBePlmE1nH1zNN7+)8m_ZVCP6<-=?;8!2nq6I&gqk(>rjiIdJHKFX6q9 z)fUzE)>i3mpH?syz%qy%RDv^z<1!4F_oR70YRs=(Xe?i@$Q^eHNDjhS#}!A@7KvvIt|=+Zoe zMhGWKkj2kx0IPG`@*R|yFjaVJQo2us?v_d}J!;TOO+9L-|FwFg)Q0n5)a^#AVAR?M zt4;<#ZFOuyjjvyVZmqvx0)34?ca6W5s1%@`wPayg&uY0Lc}>tvyrM|Z47Q>Op`u8O zc%F{)+G0c|va{SwH#r(QU0yw-BF79iqXHqLLQ8R;rJZGdVFkx}(K~LIrP{a(8_8`? zM61+4QHp4}x99$jKogT-RqN(MA;#}n35EsxC|Zp0q9jy|@uH;rqQodexj$fXvFA3o z{Asz6$&a1F?>vD3BiPu3aMp+5VJO&$C10cC15V{U?hZ}W*yDcM=k9Ju)k?27Kfk2x#Gu z&r{g#sSM;y0jBG_!;IRS%P8@N3aVeX;Z@=$_giV{ZjwU+%(LHBYp>-q(+41P1*^@G zELY9oKz;LW`E+n!II;xXO*pay(!Gu3)2$Xid9l76#^K%Rw?O5rj}vUE-Q3~ls}0-X zC+r5U4bM5=vgV%NB`0i^kT#8so6_n$?g-V=*yG3*a9p#=%6V+HTXehLmDv`B1nF>| zWQy!y{m=9d)VNXm?vWnth7+KPr_Fiy2*hr)jt9h!zJ>?E(A9>017fIgX@jjvN-2NA zKiy{VsrRy9ej8c};?_qL@B~h|812J_xvg*=W}GT!ovlRDiizfFTFU3U4y87#)Dj8Q zJg!YHUL&OzxDFZpqx3A*Q~ekKX>lG{ftx5w2>H<`279N5h6^jq`zJZ=660Kir!X()x)kc^x zhUmX$4mZ)~TZnp#@Y|WpAiAjsnNe**gN{L8SxMq@v5)GIK2p-7@^7_vz}B{QaCkIO%P`J*1oAz+NOZ0vpyTJlwK_8l*U0U%O0#% zoD|I`^d6agc=}RctGY`qh3ViG+S^kNdA zVk1jE6I~!FmNupn+at>RiM+npa527`+Hf%ta51+3QdhhEOvNJUBQ}5@4JfoL(3z~K z4BW_H03prsWKt(>K(XQNQ1wl|$u%@mJ~E^fT`W?ROsmH=$A@P&3+WakJ$m_U*)`)r zX}h7;f@!-E3(pl5K`N}|qJkwKGDS&zc4-u~2?P{(mC|xx zqL^Qi(h_XatJLmMYAknnCIXtl1yB#l_9y{De+9OD3PuJVes#ZLL`5BmrdMOzMU7P_ z1Bxs<(E-WWwEPz=Y9HcBwf3dfWM5J))!%RBE9BKJuyKJmRK)7Nf5(+t*W&Jm^+wX# zh)i`w7HFyQ!SGoOp>7}oyFaIV;FRBM`bP|M+lmtxd#w7oOh22F;80E>_tey_*OcD*l-vKw_W_~x2#X*ngGR#gaT z-YGdf^ecG;`xtbJ%LDZDi?P~oY5hMo97G+i{vLX_!5@>C(efa@wy@plMN0ko-|?${ z21vVJ!Dp;ygRFKMyHkzrxfL;srAqtMC?8M3N>1d=p6tkIAmT>pYtNEWcPW6xpTMfy zKQ(DV>#1#fN-RaPP|LgS6?>5IK%_>t$8H3IiICA3EY3I1Zud2(Xc|Bm4MzKAsC2%w z*OGKhne}~w`seAI&%mVfv-ZifJ3w4mjVO>e(|pj7pyT|M2oKH)@K#+33Zh!9;@A!| z?sz(Xt*cPEeCaPO>;1Q4r91)3f-@^+s@+GM?Q3Pv(9IYrKonTECLnbB@}_>COOk$5 z98}9?oqtlSPn-W>{TNd!O+YSG<&PT&T zGHsgc;14A!VBI04l0%F<+C!VH-BZ8Vh^X;LIV8WMXRC+fOn71UcmaA&pODM6*Nk!r zdrHIGFBb{>v=6Jo;&Cl;p*dCboiH4$?x887ej;iLzMt+fP$-7i@!h%iVe0xGhEyjR zG@*Iyj~4=XcrOe))g{|LuWwvP1R-6R=4s88_fB~-Fs~+b19j>HtBvC=Llp;f+g^tk zL>YEak>j=}w}|&Z3IG;}Y(MOF7}#W)O+R+=4|)_~?!Y$={O~pup}%U~IT`H6NgbNk zdZL3aN}4!x&n9+|VeQIvkieLG*ZO`C61^4ILuYh5$X8>@HM(V^stq;~87|q!# zC!(IErCd{e+KW7BpXRgLVlcOqv&}B^*D2~e%RLo)rs^$>;V!>693YMyIGX8~PkD-X z-nDeGBDaUyVLeBZpWVs3)lYu&DON*L%a_mbuSy#))p3$L(d2PRS8Ot#358YDcISNh zP~_R7C}YMfX9Knt;Ezq-UYTpan-%NmKs7a41=6J#NQ-^`an{+ZnV_(bpx>T|2)*a= z=CLZ%jx!dkBY$j)N#Iwd2sv%S7A(z~+0d5uOVZ;7U3 zFOze81bWbX%6*0eKdh?wDMcY-(E=A6$jq1i<^~Mhfk)18N|M%P9Nlw;nh4Q!|NCgq z(5_R72mD3IjPjO)hkDcuc-!@*h`9RTTFo}NhIJ4Tz;HyN?Hxmf#R;Hmo*%T7)$@Fw z487g8oTJUx1euOI-H!LvY}fGYy*Ka)GO)dV%hkmpipDNoYe$*8*|Ce7@mvKUnQK}X^Eo#plSO1Fnls)D=@I5W1q-1NE^e|3dr`NG3KMjs+yXVLhyn26D* z4kz?}%9NLlbzq3oupgM7J<7+6_}=-E937(_+gUaeTE%CyfTJv~;aei#9Ni>5&sC`D(`^t=&}zM-hdX#G3B$jX?Zek=@*chG;WvpU2oaO#ST105f?x2_@hQ+uvHVTw{G6&$aa_``juxSdB{hR73A-|SA8HU7Dtc*yOp&KAz!JCPp zWfCwv`l&_jcVo8B6iQ3iKk@Nv(x7ggl4ju6A7&ET=0~=x=+xeotE76?lJX>$sM?!R z5p{dh;#r&dvHvEs!gN!^Z`}>mZPzr5i(!nRx+>;sZD#ZMW33LY17 zu8jKdfkDbqKdOx{mxVjgs1+cZY5I8KqkpAK8BD)Q7I$Y!oxG?Ts2^D@iseN{t|2lj zvzQ$xs&GMM37@oVrB<)AjoE9~#P}^+!nwH~+0oqh`fSGtugIg;Yfovdx4P1Jp2e4# ztx;04t{HwxyBs97dSBjnXLSKTZu+g$851Tz^*J8+8k{(nd2v8K9|>rF8}ZhW?eb0o zD8+%X)g(RULPBOas4E-eekH;klRsw@iAW;NBoN`dFk{bAg27xOit#2Qi)~-PzkNbD zHO=5PnmBP{y%O;(c5#j8c5qJ8A?+>NcOSu>nPTKDA1CUD&F0iynE@~Ax*UdO5*LLF zg2SS0=tA>PfREh99|Bff(i8Xli)LlhY@37FihBn^ARaX2`4}jGL|#Toc(EQQ#6((F z2->Kx4NTP3x7hO5Xc@%@6%7B-Vdspwf%-- z1AAj4%$b#6v~VQSH(G2EyXK5GW0^c3OEl$%WrFdTPCP_bX$8=i@&5i*Es5-+GD*1D zv7dInz@*5vS)Z=y~JRh%n|ji^+SkCq>@9P-7{ zX;hJavZO(FwSdDs)Q4oSaE(*3ohmjh^jRhLImMQ7_SnwWR43#URVGtNAf3Tj195!N=?ZN*Nja_8^^aSGc3X@xxfrHXCRpb>LFI93Cxnne?I z6Ry(GcK^-d<~q45IYJJV)EyMxt;0VKT8SD{gc;h0%;`D{4?1XUD&+nSpTjTE z5U2h*D)M>2>Wx)xlusxQ?ii0*uv!Q+Mi5ZG({ctDVx;56XwkKWn}rHzUE%*h0{Gd9 z*!-ssLrc_^#+9`B%|jGNZ{aeld;|fll|Jku$8Na*M4Lko@Vufeet8y-Zpo$C?EmH> z>|Lueo8sRmwpkUG-dDwsASWE&_P;1LtB|(?7V8z7JqPFvH5VsQXtGMyg?!sdda&R) z&GN7~Oen7;Hm9)Fo-vAR0iv7#%gAOGv3B6%%@f+&@@0_P<0j4AO`hGej%ZWvh^?Xt zu5-KCxr_J0B|-W6qxHh&+s_fbc?!BT$}B4EmWS0+ozGm@w_YmVVYBtC>>#P_a?$eb zmSgNeoZH1GQPlA3!OmQeb{L?UAo?R0FW=m@s?;{ChAdW20sHPe7LJsS*v{=iO^KXS zhUGJDcE?LAjgyx{Rkp|K2f+W)z$>^}B?-=v8Kg%0P`REe_c$2#;}|~>{&S2niI~{M z&hq8NnLJnVuE^&9I)sv_u@vu*J&I;o<2Q%+53I1pM5BdEoYHkoz8~Vm&s;8&O0=4l zZ-4gq{~fb7IJ+`PpQG#@aD}S&$oFsV|A7!w*~dG)KclsLX;c4y-GSmi9&Gada|c}E zpX;EgW316>mmGxZ%u`~|xqvB+{zEKX$99$6=2Y32Pdi(~e?XiQ{sS_r-%R|u*kWhi z#xC-6bx(}`mk3MLK!1QYi#0Y)#t#=fE0m~f8kj}Q3U`nGKv zV>Aij!>$84^FTQSgJJpaLde=MoVhsb0Br2=G1Grsx1rU4Euu}=n1B=2xjik3vp(wZ zaxLr2VZKWWoKy(M1?uB77{wLE_DmL@wr zTn^VZBhW~i$+%A4vUwW)mK0VpdKQsM9>OGw_-mLUM-Q+{`xMJ zbLE}Jhm!>`{eA!1c!XDyHD^iz$W-S3x-a@%H8EOC46JQv9DLRskw&||x?-4b(oN}R z5TX6zo(#bq3rXiCwYziiy14hg^z25l?0SwmE*`UynnCv-CjC-^Gzi2Ub%hN#;&ijZ zJeie`7j-e&f((wvn}N;`*b$Nv%ugjV6rARqjSB+uI*1hbk>UW{D|AK!z_J%vHAS3$ zZ3TeP`H)3!=*GKBtJmEo&r<&KRqz`(hXMVc@SU;AngVNk#A}@MeEX8nfA2)GY2HFO zlSsltGvqaxSFfe^O@9GoTrYrH=3SkvdRc%Qd4$E8W6I+^rNJ0^a%f$(gCc0czmxD7 ziv}@Ae`3oVIVcaU-U>i?#iNc+_ikD8%Yeo;(7f+Jt%+U%Bmsy#E&o7>?8#XKG~;F+ z@_dg>HpRL}@)Tr_;-$DvE89-&33YCjumipjAGW9Uwx+;tyz&F^O$)b;%=yg#bNLrS z3yNQ@YOcALl6xunu=&fh&7&di7@bQ8;Y-ls0=n9aR6T|t@8bZ-=M)aMN zlmT`Ko#{zYMZ`3kd)41^5FGewL)B?`{ zC%^wmV$xW1w(ZVPAUK6?2p`sz`@X`JUEH#Zs;WiV0qf3WODMCrCuZ7pg@9 z(_XYiOZiL!f_I_e9!ng5W zHRhyT{z2R+SLNRseZc!-jfeNRJL^d@rk2~Sz=NxIqQ@BB3))1Jo<>PriA>mi%8yR8 zx6nw@eu4MTUmLNh;x=xnRGjo#sK06-p`1rQtF+>x>`ejmebyGHF#>DW@nQ9?+0R-s z9eX;hwP4RRvO9s*+tBu-q(F^x_>j%CZ;V}69~RWeq28=-(D^snUle1gWH|xEvqm>n zdnWaVhth@2ertd>_)ae>pWV`(XBvoFH}r%1^yPx$+02(a`9dS=JXJS&<4E4p-Z@md zPYN}lLI40w8x2vDuT=lC>J{DDpx(UwbJ@>MYeRJ1fdYfzbEjEfd`LXM_JPFB?i4)YOzB}HTbl8t0%hM0ky5UO&_RsIb=mZ z+?9+XouD{1V&%H~gVq%l$@LT9@J!!#UL$@FNiG;qbOcR*m`_-6!Osv6?x!TR5bN>{?*+q8zYCRasNE zggdy7>$EBBddPjOCW!)y3en4eTxP4^cP{{t@#&$O-b3@KUijwbkj@yVztKxD${VoQ z@Lxf53i!L5LgHz!W0{U4_xZ8lBRueTn-5hws-fG12;f24MqHl7QMiGWyr=|l9&^kU z7*UImHZVu>VCpc3I*Zh(Np;wHL|UBb4J zj0LI{tdK?>kKasEx=S7L80t#R!6*T@YRuH)$+h5XqO;06+li}`ym$jB$jf0#+(QHX Ee=o%e=l}o! delta 32497 zcmYIvV{j%+v~+CSwrxAv*tTsaPi)(^ZQHgswzF}vx%=MlR(-#_XXaE*%}k%^a|&}n zVsk(e^g$o|Xd8_*K>!<~bM)DPD@+L>%m`m2Fuq~5T1JTKOr0wv3^iIDZ0f2-OU>q zt;l32(QrFi6O=U{2TCb@MUXqlFP^#h{LKeSAQl;(c0(z_yu^7}gEodD8HG0NHXgT4 zHU26ZB7V)`dqeO|Mu0|s$r&PqGWmu8Bj~*%?3s(3U#tTK!zTH^%1;eS|#!)p{2Fk%-kwwh$AY!>l$U?kear=li zST`L5jlXlg-^YKLdZ29)BHiCuB?@pF4p0VbP+<^t3^UNyU~oyuoKCd_h>KwRZ8?(r zZC8K3j`C|IkfG8@PR1b^xxq@^_`?iy@sVm=+Mw2|G<0I0L>s~+m|_ia29=R(HfQ1u z7r$~5gCY$I7xv@n&IlwTn<~(g9l+QCJ|iN<&|309f3u$NfgSnS$BoZ`3dSVR_k}-# zFE-Pt12qXGhFu)HA*U&AX7EMbxg^2nGY-k!OOi_1%GtviqQ2b54Z! z)#jLe)-U!^16y*&dMpE>@H~eAX1vj0x7vJUTPL|CA&j&?0XS5Bj6#rmkd6XP-c1IS z?Y?1#;38ZsXV+Q{IXa#-nIM&ql{WDXhXoHP{#Hk)S2rhf54Adw7HEtqcZf(9i!|TW z8>oRRe2QqCvFL^`gU@5WUpao586Nkd^LhbbfX}VUU@V2c;>OSD1A1S8040?C9trqi zeun7QDMY)G#USFrz4=ad-~-}qXF~+=M25|moe?8EJ3KOjOSn(_FVHhC^9B84$b~{J zF=WP-h93k*D>=H2SOP>@q#!V#pUT0t% zV4qO9!TBFpJ=P_Q5WzRV)$}|Z252BZMfTfSe=ysW8RV=Fo4Dtx!RbBjMZ5GS?luR8 zZMS8yMg+aB3-DX8DD#7(;#JtxXZ)q!h7R*1;wcKTZmjrQG2bhJSQ>-<%GG&3N*@J^ zI2Qjz%5+$n%eaudQ30@G#o63Oe>RhdLJz*zu(1l{u9I3&K6U+V%bFygsZR(J|LKEX!vMwE^a8TKHz{TF-2MfO^(Li*-EeUnp z#AE{=sNkMs6%&wh%>B)g>-wa3O9b>TwKQY-QMC^GKwfsI6kSR2pPTf>lpjA}%fEDS z&p>BDcWOO|koLQhB0>i;WJ$kT5&AAEv z7;c`5JH5}m_-f{1DnkK01cQaos9L<+>w!Op_I|C&N06L=BL)sjg%amQI(acYvLY$N z=LgBB@Y8tMqj)!2s>MpmNPgl}TM`sRHCs_(#UMF#i|()S{FLQ5AT^ViQz~8v?gl5t z9WEFrtm7E{o_9VtXO4%7{rG)9Zs>}EFm#fnZhYd^2PbL^qI*;qG&;T}Q=Pq;-($Z- zkoiL*T#dw^T=mCgn z;#=TG3lzbPYT3Ev&|SF0>Jpi?1?zhIA2526pxp(4szUBA*hz>BQ5?>td;fBc9=!AP z1Y90hXFR--%RKUd@ki^B;D*0s7f$jXu@S65;P>%#4lMzhctbX(9`b%w50JeR z8L{_#E9%Y4ZC&?H*tf8Oi^rO+J1+lmp{FDAAq}#j~|&wKkFqwV98)o?=Cr?;~+!a zBXJ<8NZx@A*ipESkPrYS=4+RQ(dZ~ZPar+$H0lQCbN&<`q!efpQCSR4V9IbXT_j@z z7m4DO@jGB%Z0eg{8L|up=n^KFZx0s!v9LnSqC2uFr#L}H);<|2tt44G&`d@Km_;)0 zs@nC3?O4`NUf@j^^YduX;)U#e=e=}GtWPZ-Aa%hRGS&Mw^i~{*{+q%-hm5NSk=G9vTW_lpzbpWrZ#9 zWpVBwIYTluMoG6r>M^Ghc;V$)6EsilHjXz>H~x6wE8UOFz!#uO$ocEfj&9ovls>2S zvWIe!pp;l?`sZHGfxnRmg((~WbRlSRt5rUkA65Rr!JzxbF?g?M08f5|twVZ*el%qq zMgusW)|om^lu=KPQLCCZ;T?k1M}#({VJ~4)j3pR+YlXj7!_qlki&@quY<$td;PV+R z2)ay2ynVTqUIJMNyG@H6BjT8`AtpV3A#9Df@u8G?JZTS%rSK6_f=vx@LezH)XNk0Y z?aQ_c#d$a3x-Unuqu&H{WqW`QOelxH?(RVESs4lKduiu{%f|xz4u#0UI2Y(b+a{x> zf^tY1>+cudkh7A><1$|x;b_#xeqeoxaHrS6Cs=WKB!s}pWhFYL84iJVVm+8ot1 zc|10hFZrmC{>X418Lf@Q(13~I6ym{TN8zq#t#~q)-uZ?#Wef+n>`p6^dh7{g*4Vzs zUSrktoI|KI$+bq7o>zML46NS%`xxmbm9i@Y9hM4~dWuR%nGfGd`Fs_51>uQrb24>u zX|~Polj0V9;%u9NsugW+pOi8r zYdA_q*Lpa}kSvIZkKAx!=*`3zX6z85m0(XsA4rjGhsdCF+{e7`jxe)gOfh%o$)vR_ zMj3FXE1r!IKFhm;XqV&nz0d$x%Rtx#5jyJ)0t1Ny(=J>%9!fz4gFiinlDK7MywJ&L zJz2XVz(H-wtFF%WcRN+4#Efpo3&kz@1yK?puEE4 zAVz4F2S#dl8rDUqVb99V8IW+%VKD{x{6fyK+(<=SDR8Gz)X{^9weY&Pb$c;fh2icm zvrB+&IzQnrNqdOkfzeI1Dmu7wEnnPVu{@u0yN{X0>wmiLY1X))+qxt$HThC zcS}JrW+ltlZkvCX?--q0S_h`6D1gCkUr)h-Pc@I}LI`%H2f#0gHvDq!)v`g{SX1A) zpFnc1cnhxyoDz4+FlWsiU8^#Ax(JJd?iHP_Hakh+^Xk*)S?v}JvR7L zII@e{?IdPfu84dDirl?XWV#C2sQCR2nOvVkyXjq!s6*N=ajQb|il7vD(=7L5w_060 z_DrTG}{JohqrH+EzI&q0Zl8pK@whVh=(6>Cl4rywI);ggcTaDGC>}odShz6)%2$t zyFMGpHeeOKB8V<^G!Whl6wT$f;*m|2nv$kLmB6jq6Jr_ugQQxI-omd ztxUs;Ux$P75gEF295v(8KskqdN(iz(E0j{ zRq2>?C89@k(bVrS0x{Ijhu$7LqMDXRAJ|3K$wUMXAI;puc`V4~5Gl+2-5@>#vTd`u zNf2ea(gootrztC$svV^EU#@u0pArSh`>zY3S~b+86|bH4YE4mV#S!9 zL;CT&4>*Rd-<1ITy;|T}n2G?v(~xAj*>fY5kC_f-!TgNHJ-#h7LsFOfde*F#Q~B9v zBQ)saN=*(84YF=@1yWv-*m{2ik#c?XDYqvgj+WX_t9TDWjzp>c zkt>Y6N7H@_0KyV}!3^}=8P~*cjuL%om>t3fj$w^xZ_*ZtkrqPGnlXJ{ zy%de?T6R=>ayoL+XKT89OW!}S;M*FTlY-K&zcD4G@yd~-LQt*j3Ua->IKm^)fjWXw z2c7iM6OSx{ZjK}xsF$18+$O()&%6?Y6ZsG;!+@&*;FZ3h%U}}mWrW>V`8i-7z3BQ( z4s>@k_;@l+WhB+39fx&3-Y)-Ray7AhO>H_N&aqS2*<>#t+D`BuLdxG)J29U~JL zE%+UpF9l1p`i1jib!pB&oIgd6@F?KL^orOK zXh>HM0N{*z%5-pQDS~&i&&134h>Ls1Mlklp+t)BHn<_qTQ%1Iw?)EoGF0$fmWn-IHAZ;yzjQ}@fRc`t|&a)$`U z!NwY=W|F@zCI86Ew-2iaabTS5G$r0d<=haD+W@;jKfS$0>B31XI10LGi}csfl@0%j zhGkq;rGgw)}(aN~ziUn6)+b1ars@QPTzEzU*SMlPU07jsBk6S530Z`yyd(M~{SwKV4OcBK|?@;bpG_WVb@Un^oW}W`z zgZ==&)2>1b&<_?_U0C_1c=bYar zZ_}`+NTtK$OI^Ak@$$_cs_}CcXzpO|U}MELBz}RkbZrP(U`tX&euo=B)#eCtC$%6m zx{aY5qqInMUoJ0K{PHjWMPPTMZ`Z)5LWhe@6}GKiyfL4JscU{qJ?!h}E_T zF#jGM#LJ3^F32@I>#+XQl%a!)&DO6V{>90*o&QhsnX7u%?#9|tp>_J(5$=+N{AT%4 z2lLMHLf1oeY?DIpKIFDAk&gOqd1{&N0V~|+NX`KA4(QrUtoJdSX!j|b!X`Fo{+jr= zEUAJ1@o;9bo_h(2!>@Pan|1yagV%ysB97qwwAM9-yVncb|_@q2u4Ss3r*gxTp5iO{@)`O8R?KP0v|^ z{UqZ;fIGxpXy_2>P7QpEq$m52TLSlwn`7)h*#6)j{DSl!Dz{U^-jeQ#4vU-F*x(eZ z>A4zr===}F{s%5O?NmIs{`QRbuUwA^UBBjGpvJIfeAL&Z(wz)7Iymu6ng@pk>?6Dy z${UMcAtFXlzqP9~AGOl;5>**3Ac;1h%BnUNev8yus6`+JzdxE<#fCCAkNk?QV%#`i zOs`>PN0yaEF56e34d8bFg~h7kN*>YMgZ9y&(T3#BEX}BCyq$c7YX*>6X0*W*lV7+q zh48Druvri^SRl+HJipS6{}wg?_x;1byH9HIJ6!)%Gd>T(5B}o}=7T)?W*X#!r@{=d zke*0C8`LBygS(l8`e!>+xW87|Ao%ZQkpIz_kb&ggHR1NvmzU4sXP()w-P%`SZPZN&TZZRDesD2HJWU>vf{jMIrMpL*qR z83li))&5ZdRetp+AM3zXNW?w_^#O9#TPZe!p&SWxPWr_;wAKcgTQBuzgs2?SZE~f{iAmQ=~-^FQY;sNf-h%7e0_0TLH<9%CF2$7W)V36*jvIs~l*XTW*@h%WMe{Rfc$z7o$d#lM`68t9^_EtiFjg=JpSK)Gw1^ZX_?o*;} zJr_r_ipt(=iz_okB^BOP4{^Bz(Qv!ps{()6SFX(^XYDT(lOGtbQ!jQ7(E8l$?Uu1^ z+PmDV9l(~*ycY9b^cvm#7*OY4&1-9O?`R!iU6Ro4oj&Ra>{|u@0&2b=viH@y&tXG! zE?U>MzOEdz`8!<|+m4r#2Ws7q^IJ&)s5{ob{R16ocfW6n@H8-wZyhW=bX$L&ckuA+ zTtWpj#{q4CwewIF2z8`7f!OVW2!Pw`Y#y|^Jv&y5LeBBJ8iMXzS_Xj}sDop*z*hBo zO9MmP%(QA){TBkC+svTWIG$~g4t|&-(azLFhG|%fy-_BZCA&a1hsf z;8&8D;*bd=qnSQ%L5kP^o|$!K#Ix-|oebb?XP20aCg>AnU5sY6Wv6z!L2H1{iKe6^ zAIu+}Ge!c^#R^VFeyJhXoKP!^5bvz`7){1g`QToN&t|aCz4w6LxS0*qYjIzR;ys#* z5>qQsHs>HeS{H6$`0m+>vz zhE0RuuLi;##Eq+n6#hZ|kx~GWVJjmq6FKKbyGwwd&i}3Tn4;i~cJ0PC^KT*(W26QK zJ}L^%jrD&Fz5f_bV4q|EMKiMbFB(3Ctz5>QLTUSKKhE&kzZ1K%aR#u*bk^2zdh;%X zhBoLBa)h$;U|zQH>IKrGa}*PWlisBmLHwWh7wHQ=Zn}5cdU^YKS2h;m^`QflSsDLf7G*yb_?g~smjof21tiK=FJ66{%z7Ie9 znb3KSQ-sjq_2`kKF(X(e2ZV0E_TgK^I$ccq$!G>AQvQF0CXn? zg8+se6@(HFcuIm835pzc5gfyM%qw7mJQh?00$YaMry+=%^?;)T7OA1n&(1YvKox5( zBBUCx;}gmv@)OGNcOOS~p&~R<^-=Nu+Yb|qXeFW(%>G@# z5t1)f2!CWF1E1gM0M_E$Kb>2>*9$+E|~*RZ9x-@6DalTC^i@oTZxqlP55)(1!n(09{;9| z!~|U6=83RWZdfy_$W&27Ez8MX3J7s`t-E{(pVL$DF!AvVNP|XfLhRE&Qg_ooQcUVx zfh;ubzp3Vk>Jg-qJ$XXDz}-)5C-`sbHwnoMm{4SpkI$R`RGOLGs?OI2G#f|;M*kkW zuXn#C$p53UCZ8{v=pI}%@-0xSt|SM;Fn+ge&HnV^XNkjz?odDd?4sLqlaIM){j*r- zNn3r%o+Yj&#W)rY=At}*W8t9R;NJ{IVF?UK(f1Nw<#D-6jZMBxl>||_N#ho6m&IUO z?aS&NRONE~*a9?bH#5`^HUX4hgO!Ro&H+3s``X5 zjFP5+%eH1)Ixwa76{{EY;g0=j*a1!+O?L$E*iE}5a1+Q&?f~+;b zBe3>RlJ`)WyjS!Ibv`VVJ-oPV%Bl9u=yp;%YKe_T2nyf2NM62BVc}cp%gynm;^&sz zov=ekeA1w{*C!yb?w6DAm-{!m+XC=!xa$~OE33p&p?mN3oi{Zmfi>bUc|O&@Ddx3> z-_*388x!&2v8aa!^}tZ($K0qoN|X<}E+`6JQkIg_#X^|^PEPT^TJ=&OyqF2nt$FJq z{%vJtM5P|1MMh;aICCqlxyA^ZK&G_<bSvZM|YM(tSwQ=W<Fz;b!F-mj6P0IJpJb6`oA*gch0WAZSCVgAPRtN=FmK$u}n<_8YR>R^61;n zG+-1C{_B1mnj?8;yJwmlk+PoDnu8MkC z@;2e>iCpwKx^_25+JhJF`j{G@B?NUWAgwRaCx4>#uoP_swG)NteQo@1v~_Axwr2In zRc%(b2;0*IZ-)Z3KkuTztS^>r&{DD-h2?uCNCB1QVPn=ea69AJ08cH=+T3t)V>i{b z`a1k5T1~zA_|BT+fLF zoj);NzworfjN|*T5z%JLK3(}~jfo#k`SP%6o!8-+hm?1@bRY6UbvKVX6t8PnCjNCTjD~q6U&rA>$8$Ih;1 zS+yaS>`kl1m&$FYkqO!?R9TeC27B&UE~zCX$!2B@yy?-lwj%s%yF1#fPIR+J7lTTz zCdnKaf94Mr?73&DEFIBRTQVXY6EJ$R5ZyR5d-)n|5*9XLuW4?-BS9DyU)GgHJ_!efp!~M4NG|ef?=de*OeVm`ivr6g~ z`9SSkTPIwj)KQ^7H`&LLq&~-|t`k1@G|%&G<-{pIZk{H|NADtl3QbB>WDl6n$#?VFJJ|G0kvi|DmB8)9`J-SpFP@8{}awh2zF7}@}H{E zaXS6}}t!0OQQL#LVr|o3!x@tkdu-CLKsqS5#tF^0>jXg5gKeBm>{a# zD&TXF6&&K-gQ~@9kqc4kA^VrFxaPBl(8A*=5vz8q+6t*n6t47Kt3S<~ zqCi(1HnZi@@NnoIa|LU^!)#sTID`fI&!NMC7(6rhgqK{>Fz&e4*fA@ zUw5zxZ5GII{f0n8Ge>GYsD!^u41LZG37?=vnvgf!V>|7Jo<@7HnUammoXDUZqQjG0 zBR*X{oZEnaAUlSYP_(*%e?g>}#bP#?|L$wE>Agpo50H?}vRkWhb69Xg%uaug#tuNn z^JxGIa+J6iD6g!+dVqv5Nj_tu0o;5G{=D6_?&gbN18pmB?M37c>`CV3AvT~6`Rrpk zcb98GAQ3v^%*$Ft{0RIV5t_pev^STdE7Yo^$rn~B_}Y`0fzI6 zH$jk+3tzyS6c3TW98;jQb|GXwc8cfuvoz%AX`?7- z7FUAI@QO2Sn(hsO`5QBp#@ekh0gk}7zUXUeEHC5lKvnQyDe6$Y3sLTAqIwB(n~2wjr!rBWx+>(3Ejb zKTtnom-6asnbR^Ahv31a(Sx!Fs2n*u^*$^)brd^!f`bl2CChae%>(x=e+UhZvT|9c+y} zLvhF?0;y@Z)-zXA`C+#Qf36TZ#$?3pr`<-XmOHe#7v78h#VA7fd(^bV_?9^;IA4zT zU2=S@w{&g~-`i%3spB2vyw8Cw&P`x`@c#B}_O^g)*V9-->VB+tz^U#DHzo60iLIVM z*e5+$vw2J}H+z-4sH$BnTUFhMpdDrdECzVtjAS$mOtpcQoDk_wZl(ub3lfBbIdFb) z^Dvk#*o3v~75YYo@(}j~dxM}Ymd%(@#eQAeC}l=Ub{jYpqs49Sl)X#Gf(Hgu-fO;K zFIY=!XYWY7C75vu!1Yy?>>Jy?^sJ!l?_i1K``d9kDT=Phh9#OBchnnGaz-wqRWWqI zgWblnl_oCWlAlv(XqJwO*3_GKDl4v;=?0#eUm!?*XMSsT`L9&XC$=_5)0%gHRU3V> zY!eK@3Hsk7@B5|07cOyRif15lgHre3rjt|*6Kg)!Hll(ffTOmy={g-;piR?TL0cUm zx(slM+rA0p2q7$EyeN#I^!`7f0;u<=1z5~2GmL@nMqTNy3oSs+a&Pfc0tjSjEKUUY zjQ0kX2x()=EDls!n4Y&|d62xs977L7sQ2*2Ur6@YJyAuL{}MYk69T)F$jWDJ>bGyL zJ$~O8llaLifK$wva>vp(gTSrXFLWM6@X?0d@Rp8rbLpGD&e@(&ZPYS`3Urlq0^FU0 zy&DvuypB!Fm8%A^xeZQPQpeZ_;V;{}kS7wFFxJ!8I=}hx)r>mv?PY-Q9qq4^3E`o! zr-{4Awa!+f8um7n29&XC)$COJ=43eeMC5rjStT@J05wUNkPMTRbIW4C-)KZiM>B=S zML|3D1(_$LgtpW5Zv^r!wGnxCEmlsP`OqOppSjBB6qmPf5n6aKt9GB z4aue)!Xjv8pb4@NsK1uNdaElO=M9(bUQHgrLeG8C?HmMHY z)E|y;9Z=Nm5=4jH&fOf3`7*W}lFvB209K&QmTxyQmbcP>AD45kIr5pZbMFcKcPf!U zTa)`^?AH=IVP~~%*gc3-ZnV4JE3%jf?RN3GSu0HW7pX`DyI5_TLr!RskowZ?`$>vJ z)5dLX3O3UP(j0Y`6+GnSjhQtEIQ|?I69Dr-y*i@@jQt*LqK{N5`T33H5wZ!>0n?ue z&B7!1c~jnze8EWq=*T{nB$90R)fFS14Mio;K3`No2v`liH(ji>DcK8HDPy|%rAN5x z8FK=9%yf3&&YczI0yKpl@SgQ2ym5aqXsQU(|2#noblc;R4wC1A)p0Aan*@kGt^q+?H(f_aC@M zb>jniyq!~$W2ZSBHVTj|Ik>s@lEfSlzti zMwG95oj@!*08JQzt5$gl-T3DA;9b}z?;+%It3S&mt1+gXEu zA`K@<6tACFj(R>`LodK5N*vh=O(9!09yHI>pooM}p`kWwt+>%oyWz;~$zzFt$J2FeH*acvG50E;-^oVWM*u%;92FoKB1^( z`<{%n36$g~9jizlkbhz>ZO(U@C%S`6w`0&n_N;^DOV)Je#1f#2-h;1VS8_Mg$PPmEGV`U$wG@B)2je`h!be?rS?I`hCF1DU(P z#zco8_0_H8&@H)O+a+7^5Ca;M9bJ7eHDi7q5g2Va-@-9_NW_D1hT_v9Y6UJsyI@f)=BFe%7<-N>mq z%D50%S!(&scT~UB?UN6A*#IeWVI<{LhSzfqmWxz#;M}ysqmaSF8`^BkAvav$lWh`O zicP&1bvR>mVlo;{nu;%3j?Yj0%LeU3mD445^w+dy>-seH@l@5#;^f`!vP1dnEVGv` z=fCPhc(OXPx>^spov|YEeYE^gH6GSjdgf&Fug-hu_11*b-4=-a!2n1ALUcs^>Jf;0 z`|N-@l|;F0*~8tC{Ejab`}rdwe2lB`Wx>7ZmttNcW6)fS=U{%L558!^^QRZL-1co8 zZqDj{tSo3tNI`PiYAM60Ax_~iJR@6nJD5i{J0Ec$c>3xRQC?qTa3vHWON-P#C$P1U zL^USI#4}$bsZ@Ejk%JA_r~PkaI3* z$Kf?yn9&}sw`eufB*{CCIrtJ$4#-d!-aVmK z>#_a>_+nxlwB>#soag_(Ft_2K+hUG$0_yY8-#!%Lw`#}# z&>8T7%pTr?0Bpz#;gQ#v_HK;3{LWs0k6!aJz8SymJN635#eZ5j*dDhKIaa z7tbXT^imnT$iTH48> z)LGVxallEY(hr_H!AoC_j0CV9oYnA`R^V_=48P`HoM@}lWczowy|*hcYvwt5JB`K(s_?w(H*7o5_YBu}yFqitOBJHCt{>s@KMVW*P>zxbe-&N( z+oorMs3B78>o3T$$@&mKb>uk$HD}Q8u4GMPjj@%k+ig<~2o6*Gn>oViye=JqBV0qF zhqqp;lq?)pyuuZQ$EDZ*wZuGVf6yJqazXYZ!<#>0>(i#Z2Jm%GrtyI0WMvcs428oa z*`BAo;kZ;NVrru(Hs7G#;kLoSTSz|8pityX=*7Ymli_C<^@?7CR{q+Iv+By2{8j>X z?lxl4qvvf>QVu6gp}`X_E|Hj2DKg#KH!~?A#xawxHZ0?QdgAgOfw8h(-Kc6*GqIs= zaAVF-$JBWOm&t|hNMox8W?eJ^{JuS`WtAVYhO zLJ2Byl@Pe2dMZp~#3rtPDpeC%_n_^rE>&B{*PNSIS&BXscS{dwbMMo8+YDHD4OYui51K#$WD`*ne$J2DDX4!ltxk`i) zurbu(#N078Jo5Rb)I^%ZF3v}Edp?TCnMb3E_FXsT?{soUXV7vxvLn}#VLIUqCVCoW za#^-5d;A9{5e=)Vy2*TSSsgEWaV;P4)lNt&c*?_^CzY;rxr!ezm23RMTt*Zxu6nuE zHiSgGB6jq9i%4ptdmR6Bv3{$U0@cq3P(&!->D8Yi*mI_01NQOLObB?pitU;K9t8gz zTumGD1oHK!*#<&F5LO(($3nvtBPVQ^kXH7^kQ1hv{ojTeYd$Txq^+(qJ_OmIGZqL~~*SQ0owC zzHDuvv6Y)G%aUWyw&vg`$u8?%&up}wbM)^NA$JLsOG7d9MloI@X*8}Gv~bzhbM9jw zu{PhtYn*Sfdj-bFpedf<}&k3cNZ5v!5D1=vGEZHHeNm(g{o>-I0s~M<{#S& zTWJ{p&M1*laH`@9KMSH9&t7E=2ol(f?&&4VmZH*&+5uj&bUimI#BS( zaG2MxYPTS)OwnUdzaUJbbnRcCy}_&$47xnV@46NlEFZNG$sbP3RXs^(n%)7&j+WGm zRRn7sfvKSXCpk%DN??8vh6|C(9>)IZb7DVnlLzi+ddoFozS1ZZ4y<6n=ZN?&3BMJ! zqUu!TBOMI0+|zDvx<7Z18hUmhYo{0XRDV8e2Q64Q$gUeU&a6zuF!a=Uv(2gVCi}0f zh*Zp;WmC1Ijxnd$Bb@OguDp!sB@-`F%rwq8LO@EQ*pkC?Z5tM*;_zv`kb^r^J>xni_0a zyoK_=pKr2pQHZVGz`B82+#C+LSn^wu8B3y_DaDb+>*d$QW0RS-K*U#aRq!Xc#Kx6< zCe&;|Lx!YC!&{~bA8d%;twLA!_pFZ689Q{wNqek}>^I~CTM4?kqxjtC0l+mT#Nu@{ z4Ey(tz_0|mMgD>yWdCc~lfNMi%BmexB062d`k7z;1I61$OUmYUNFGl_ZW>^kH35?Oqjo{=I06!Zm5dDMw0 zRm1|<1j9VjDcvdfG5o2lr9{LdNwv&$R=k{+pFzO0vGBLXI;E-C@8fqULI$OuI@6cgkFhoSX3{^xA}8i3OF z2%-Vm!IGZrOfEH){o?`;=&ZR0`Hgf-N2f(&&gK?~v9KKAA!D_bvV)_!66%||Dj`@U zfJ$P36O8;OzMYzfV{i5_Wove1O36~wPmY?K!m0q@Ap;R^nJuX9z)slzjNl03RZ<4f zzFBQ;T{teER7lRJ5R~)J`Z7L#)fC=YPB!RI?k|+H+Cba|hCx>aKw1oZ6DN+1bb(Wu z&EF3rJ)DF$!O>05%b8!2Ig}g+fraY=Ta@*8P$zvbFH@cXxmbq^_Z~)( zmjj1s^ZGr&#|)a`z-{2H(d?<_4IJ^ns~H|Gsv4l8=(AaqOzczQPvy^x+d@V3LOFCy zx6FfFe+CQhK4qf$J?~W8ge)pW_P-`8NB=WQ`^`>nuOS}*T-hg1*}4u=h5KVcq54@F zyZS{~d^_sBEUm1mG&PUw&mj;` zp}`RjW_ed^%X?)c{6Uxz)Fs+NQ(ypkx-xmf+}L0)!j_58Mfva5=!kY01B+frIcz3? zxLLUPBpyFxKff7Jpm#C;HNT)?G?{fSHE0kjJA|G1)!GnPtZ8LvWF$JDU(Uw>@eM?D zU)-)&!X7P?x;Ex-I0>wi!f7AuOMMB?{2rRWf8mWFAqTdEBU)#>#j!Du+wmcM2%H#FdqZK-3^QFwX#7h1rpY`RvqucGV~m_uL^e7;;^*>F#byVHV z8!e2xyF)4N#ofKQySwWFiX7bC-QC^YT@UV1D6U0|T;AWe?z-PUGqW<8WY1(LD|zzl zO>F6?N2o-~!tc-5lb6w!`kmDU*cb|;z>v7WmcFrc+=or^R7U>>F)Y`fasvMotxuOT zQ|A%S9~j1z>*B2t@q^6=deW_de+}TLn!e#s8-dN~mT$SP+r(C}B^U=1>0 zB`VJCOt2ULl02&n>^W9W_G(;hqhR1oxlPwMZ9Pn1SV3W8>O9RzP{(?n3~^PYZ3F~V ziut5JZ#{>*H*w(8M6KdgR?QfZeGCrgDjr1DfPD1)C1ED1L$1^$%wrnW>TO`ZWfqJ! z-!C}ze~eik1t+N}8Bwv>ff-_}21A0Baz7fr|06fzq_i9SJkYm9td^ImJk}^iJik^ef8mEi)|hnJDx4`@ z1PUwThlnIG^7x!pU@J|(tMQc;Bs?rwu%84KJ>W8dGw6rd$CBtVrx&J|k7y`A$ijgR zR4d>oB^^NCr#&6UKYo8#n$X>qKF_i!i)FGoaWVS`D-QRiy&2c--RoUhCwByF-G}8` zda7>ri^ax)Yd#~dOz;FSvhNoSE*F~=0bTk4~uI1aqUZZC_;j-HHzc#bIQu|m-G!P zhpxbpjts(HY=tKmasIz&mNmq6HL|ek((!EwZ{A>^ zFz^V|(B-1wKA>v$Q#qlR-A_emoW8pL*rahqGGIkF3u`weQmwe~>Y3Rnk51E1|Bm!> z>*gh^vV~fs=BOS$84<^L$iFaY;J-w>*>|kFmpnCDb|$!JDM1`r>))T8wVA-rS`iaN zg^>dYAfh-g>`6Uo3mO)tb{a7?|EM)&v4+IHQinj}oTLC^*N5u|?X3uGp%O6SwWhr? zozSqE?n;pG8wQ7=h0ojS1U8a9l&FCoa{Ur8tvq?E&F;TpW{AD&M0(_8lvIT2SPA;% zqez4nF%YtzYB^#HX_A|juWZ*szS;B(qwQh|(t;hlly<@KY&)f?#lyaZR74mz z$1`UC6-4Y$RjOHc>_^HiAOe3>+Rdgw8w%%;zf{+0n1D&uwqlXoEYNt>uIcIl99tv+5 zILW&A!l_{BU5EnBT7Rd9kOWTkN)6^M(*JIq!9x^KAI^X;a$`$+K)3)KW@jjWnMCLc z(-~(YGDVUO-3u2*!Mo9nSI`1oaI@2$opgq$O8cy21`)j1!YuecGoE>i-0 zC80|2rP<;w@_ynQEUzVvUFl^ek166g!CxG(ahS6#mG@B=o+ts9 zOQRclQzxx!Qad?`r=-eoe2Tr7H>h`wZ^)~^Nlq|UVBRlKV4^R zfbtAC;3ggQkO@n%HEGFU^77ty?a~;zs#Vmk9aJgcs&-`B5Ej^pRwhS&ZkTyhkF1s;C1h-M}f*+2b{FHfMkZ`>P^9qhm4oGtzIzu z)MWTD-FFbj8-ag}s2xb+8|84I_q97vq}3VkH(D%o#cXuo{g+d{>Vm-3(RM|$ z^C?Q`2v%P(0u9hVYfk(H8ujh0Vx;AM_luBhXRx|4+$iOQMr9aseWkuDiWt|I5A!fH zaNf20R_5Wd2-?tv>%^Hp6{-i&BWa-`=b}3uwwGKE_~*s73p>Vq=?Ix2rIM}#>wd9f z$y@&{eAF%X0Nni*3p}D>Yem?!WzdSPe#R76!9Pq126BLF!hBrL2#|>HL!%R@pPo?c zgm;K-WwTPr6~y?E3=zg_l0|-lL=SEo9z58>4@*t@P7o$l%H>83SqoK>sJn9Wj0CuR zCbQW0xWO=)gy3Si+^`YqHTZ5fzIqaPp}zT?Cm9Cd@#po%5jdi{=nnl60|oo=pZ=~= z^6!TUP%^KTx8s1SGwdiU_BadqQd6RRy{)d^sDL75os-_1gX1F_Dy@gAz=Bk%T>ZDU zV6p;Ifst5sqW>kCyb;*A>E0OrMZ=5+b(sj~N9dp6^^froWUF7MRF4c`2X+?FR@Qfh z^*_3I@kVFoqk|J6m$+`_euNR;645N^_k#}Pfl$1Z@jhQUDnI?Mz!l`VAFvG}0eOcA zuv-QWLwZsG9xjr0YgkoCJ_-o5IgzIZMWbPEunl*@ZyPNwfSeT(O68hl4ZoHe5*;qc z*t=48Ywq(zoOiDH84yhFf&egqeAJH`-SnDe`WuzWdIB0L)fj6mG8MSnKf0Pr0Jvv( z(xwKrR({u@2;3zKo8DZBu_;R%9Kw+Gz|@$J=b90V`y!&)qG84?agaIeG{Unhr}H)# zaEhQm-N`8^>CoeJmEVcbAS7&G_Cz`1xkw|F*J7~kV#Sq|rqz$gAxTG*5NM;`wi(^u zS2X~fImdN>J^VQ9ErBY&WDpkY0YfpaCSDLJr_@~#tO;m$Xt1TA#BCp-$Qm|5S^Wb_ z9bvNAXR7bcCA@OJ1;HjCoB(IsE#|$`b7&9f9_lSDTUN=~<`XL&S zA_jUj`qgSan|!N|HjW3L2gJD?s(k-GK$r~MVDg67rB}?(?WAcD*k2G2XA!ZiGa*KH z?-UIq2cL@K1a2ORh>A7`cX^QEN&v|P)6JtI-|I^$W?=PzSF9!A=UjxVkxHWWEP9WM z&sr!`*Pn}@VY=~5f-8Yuc^E?NbL=E1$^0RiFSUk&(E#&6`v|~JlZzSC7MmVP3ThOU z@(A!Pt{|6&kyQ39r}o#9yu@xOtAu^33ePuO|7^iWAiK~DGgUNRI|#oNBn<$I3qbf) zag2fFV6<>-EQ4fX!kpBrMe`MPLeNw6VoMI6kjQ?P!X>eseJYF@^XCnnq`#&z3T?*) z8GYB9rF4q;Cw{H`%{%THhtfaP-(+5mw>6*e1DYbnY{hUIO>?k!gMmLBKI-1Vr$2W) zk-WE2AI~wzdGB%v%@q@7+6B0+KVP@^yY#Z-farU3=Y>UiqBE&jeY+`nhNtxR8K#_^= zBDs>?kHLA-7(s9UP5xyx4s)x?`Q!#6a}XbGvIwrVu&kK}P zap>lRQ!M8>dk>U@p7U8DCS5YgnrJctYwM1}orxUWC0b(5Y@8jAU$%b*S3wWL#Z^&; zbtQLl@_G9QkxJaVz2F19Bic$i!&GdHE)t}{~lX- z${i0dS3BID>EYs^OiAEWxk;pN_+6NW!phw;W?5r|W2dm=X5_U`3paNhJY|s^<;JCX z=5R?HBO?UmtaumQice1(Ppo{&?vN)Z#p+?YZj|g!Cy7Kei*&HV@=R2)fJCXr9JduP zMuKH9+GHCBjj4HAnVyam zw70*zw_mKxpT5%$fiP4Vo#N#;uxGK3?;gO=5 zXDb~eb|{?ZBRXZ|hnr~rk9Xa$jI`tR3kjI|6@Mua<*C-_Q89P#8XaF|_zNiY9{ zn9}<||B6=JAhRBxx5A&Z42dYxpctc-wMrG60zO6I?FRWNbSm<1RV+HvNR6SFzf)~a zrXrR%?plc4m8j>H)meM1ca%t+I^wLe;8C79TeDdF`#19Q5VGsY;i~?CVIkw15P%=qC@94k^c>B=IEEtoKvQ`qABrs2)`_1e<~pO&wXQHzfJF%NU!X69!{`_LP3ie2)C2@47bx!A>h#857V zWK;5uyB^r_^{Y6>H(q?YMd1j&!OruWFf<;mmZK~Z{Ya(gvXU^L@U0w=Dcb+O1T0P) zO2wNFw6;ZC)DTN!)jF(sie;E$!mTEs2vBpU01w7TY9TVs=5<1{9uZBQ7aVjhbG0i2 z%S2)jU>W!d@@uLA^ZWmTrPJtPa(?>^gd=GK7?C#uVVl`hW{z)&r2?n+W_;JHjLqr4 zEBpOJcqYCgny^90UtFWxf;_NSVgJXV7pz!5h-rnl`73f!6{(>j+N#dHphU7s-v`UG z%WzINiG_1u?5Hxh;caOgIoa0)79+~ASJb>Ywz2X{rFe(|+&pZ3h_dXJvcJrH0hsst z@tFwbSf~Cn(KQa_?l8`g;2o1%@%ziYK&28dN-MJ|F-K#b4Qt+A9>DkKlg{9u2-lmhm-{v+zB%S(HGH8<*2Z&*u! zT5>3QQMxZU7MlsnM149G<51z5zc(FKFtE5tCht#&usGz12RX#4mnITD1rm?|(M_TR z|3rjL^{MCt1M$}p4R>-r@Kkj!gsD=Ric~0aN({O;xl{^DK#)z1QmPt>z|KhaoLEc9hP2D0 zY&fG0lRp_%)OF}rj;W=cP`_af2^4W zDK#U3mzr ze4SOPHP8cKk_q^OkZ!Sz0{t309p9%k>ot5+din?S7t`Iy#;qlV>0aAPGByS4W4(K* zjZL{plN-|>Usfa=i_k408^pLp4hy**l5}wGL6z}FeKONV>1PE0(_W5Y4mG+0af+HR zYpwPa3$z4J+@9Vzk65!^C_q1d!$^S6ur=@%d_<~8Xgql?EmmEkk{gc7tqAp&UvR2( z{jQ=UP6QTyz!(a?H#*dbUyjm$+&N-F)yNcs9UI#b+cTURu`7=|i)S%8Ccqc=MDTAn zrxSTXxa1yS)46Y;!vcomBr~}L^hav)pcyK3s3*1k!+Z3Yok3E|cpfqOy_C9vXf}%Mo+?WzL%9yGX>+*nYW{a^EI=x$-Us4Ogw5xJcDSCU!qrNybZu`W(u$8+lj z+mwf6+U#Pw>ZV=gcbpZNTht{j-Ibk$x|?&e!0osnCU~Dd%%Xm~ll>CU3)3T!w1TK@ z)E|GrAN|F$k{nIQfi9PQ=osQ+t06w!+sW4dm^A7-+N8t3VlLtwf0|^mmSf7|56T^A zQ$_pFxNEEZSn_m)DXeZYAY)(QMWQoos1jJ{Z++C*Pj443lGWwtt-hqhG-l@H3(nA- zf*)^BHwliPy(|)51?&bo8Ig>HbRJ}D1)1?x_R3Co)>j8lz(4DO{1=^7Dr5%b%>-+r zm{Z5M*^&HI@3%u+>7K_Xa`-gD#zK+9{)Eac|N zHgH91UDOyDxlt-ne=``$-ys*!XYilWLx3dy-#4eir|Xe=2xL}wP+fK z(>QXlr;YC;?fDydQMov)zd_xanyI7Fc_iDRW188{0L=ceV1=}b=cj`dN5JDad=spZ zhCeV;K8M&{bk)hTdA>3S+fNE4I9H!c~p15Quw>i5P_%+HngV zcvW;`C^=IUPV_K1;jxIWgu^hoyWv~`=01#>X0=n^7G*j zN*@Cri#NK_8B`NHUl&B`6Z)GkSOHqB=JJdc3i)G*Pr_rO`*%$m>`+psqGYp$!||jY zrnl~ph^b5a_|{i5_rop5#~X4m4dDJ*tG0;X;>w4YQPt5|7f*eauHWiuPME`N`lnlT zEKpI|LA<$4B-g8;nR!p@ne-}JK5hZ2mH;L81Vul*fzTv8)Y=Py9`yi0pandi)VQ!# zEV)$N#5Y5ltqDWw8UOLNA12axEqto@LN$O4jf5&>%<{*8Vof+xZBk#jVdrIpSB_iG zWk$n4S~pk*nB+2o`P+xiwe$ULBqd!D4j@4Y_GQwqU256V(4s*pKS%IBWkz%w z$cmTW-XL5WElGE}i+}*fNyz?-^|(6iI@(aR(BwR|hLT5|h~in0VZtl#Z*oh;VW1-W zxgz?Mz(par6CWbSs*Ro_HoQ{~-E>&;r#hyC<$Q!IBy?cO_VJQb%b0D1jx#yj@I#SC z@q*FFLiDQ_%|GZXf=8KYPVW~3?%)K;VMJ3JgbTc~fMrq^6q-@b$`_uYRMe6biak^+!Y|0W=nB?*eZrE3&g|3;xWOo=V`Kr8w z8{6&Kajq*0uoMFdKT4*zYY}1 zYhCng0^wFsI|!=E)tSbGxYA56oU9?BZ7wt~vSZtU5&$uJ#%yVd0dR&B@;3qEv^lzU+A`-jQ`e}4D^gtaw@0!NB%A~q%z~+2i|ugwpx>8Hl^G#L-gH%NK8-#!;K=`$T$HfzQteHGnl}6A!!MbS^N_byJL`|laiA`zKEeG$g2L~dd$w`jx@215 zvU<5$=acJVY)s~q%srOWy;2La2Roap zKNb$MyGYi}!7C!%48~_t$)}0cWi19im4#Z6%n9br8t1UDftA4+u?7QcVx}H{Z9Q2K zsuMW4a>h$&rXw3UTV0rAd90Iexb#SYQ+fj?`T52OLY$*Q5Z?w{vZqM$&a`L>JiKWi6E8M<0|I{&VvL{yor^_T z?vjUJj~92_*_kyE%aGJEl5Joo(zEIZjISplXtCXM&Rkc&yxR(?>v_ankh9t&H7a{) zpcr!LgXuLxl0QV&>el5N4ZoZ898B0`v4IfGGXVP3 zS;RrMAPm<=6WyZcMjMR|1r2Lkjr9O^&dXxtw;ya3+0kC8-chqNR7ov-@$6;|b&_fq zi02@Na*;H=*T86I<%1;8QLqnRI{%u7g-8tsl&8Y z+PJ14yXF}UlwdP0Nri-KAIu)wXM}1@w7^GaF?o7(&y5|b;hn*i^9v$+If$xHkGY%D z;Dg{X;}9in?YAw8_hwq0!$^R-XJ1S%5s5b##@)lgTTz9 zOxr!p{%oM2y``oD`}<0$kIjxQi+O3!3frCivdfizl-!kZRwd5;7VJv<`oLuPKNm=5 z$LsiTdJdNcH8RN^dHB~nrMamwYd33W@9e>%LA~oBE6*FJ_6Ot}+6)t}CgzCnMEdl| z8%M~rFli$~ctuOh9k}0jQO>rr>IsEIcnm`=1cQ%Bvi4-awS{w9Rr=O(<0b{r8GdYZ*E$ zUU`)=>WZt51T`7~TxpfG_BqV#X#s%!hfo`kK1M?(@sFK=-q{}g75E46pPn4 z&F;_*z|2-zw12Ch{~D2{HB-otpl0_6@PPj|qpg9!1CB~+Z2DYtyWKkm0_rI`E<&1v zSEY?K(U{BOtD$ZK<%yZK4qP>_<$$OLX@tCMQ>WKw`hUPL2t?sIfGK3j^%Y|9Kc@|6 z?)m-ooz#rkECCw#ymN-DXy)ZFRZz;I^2Gl&oIX0Y{D-FY>hknaE#OKA3@Tyv)GA!m zxi!+OOCV8IGpb86F8&6DF%o;GcbxxzeU#cjOE}YQhbYGLi`6d`j@~l_`j3#LZ4)t~ zLX8YFC9L_QQ^;ejY9K3eF5~i#D7iiFhYULtC5_gSp(!tE!9$?y@_seejIHFtzS|IV z1@?=}n^MBb*`5B^2rFoqictWumkwca#H7&jHu_o{)~k7gvE)dCot zbIqo$GPTE>^d)4iT2W4o**d~$t)-(fidboRV^ZG^O?`THOQW-*@Dvw;B{Dt zd}#8ZA9C;asqwa#@K+(I1p|B^s0?gB6#r(QVlQR*;K0>emwla(e?`)u5D$Zv=Tbp) z2K(MbEB*7zx3A$UOtpFG%eD;&snsv1U9TdTYj%!fv_w> zwmeifnrWybvLzrSI^TH6nSB-Q+;f3i(b;op-goIns9x8MEJ?bi4YOB+k&2 zQfAlSSaDRgKn--Q%k({{5k$Z!vYJX=haBdq7fdCe1wUYN^mQ}xt&VJmCAfK)4Sks- z{kYdv_*BCX$J<``1j?!>74jDCmXLT|efy^~!1sBq3$e>c{^KEB!L@LjVe?zMczkTn zW5t5yaA5~-@Y@0dhI28>zcj}8_hIkF+Pz8R5|H_4LK>AhtX&G3Iig((%2$2Fk%+-5 zAKyry`c*)u%ENxucLXxj&#sl;ZTak-BTod&IS*{ry~q}RW=jZ7fewDwP&^-ok{b1*j1v&qNQG-zy7<1^m# z9%q6FP%9Lt@M=~3K14EwNAMET!Z?HU?7SII9bAIuXTDBw{|XsH&!IxodNP-gB?asP zUo@+M#YWV)Tl6Z%WCA$yTw__Y;I0SHQe{Hk!1S&AH@N}~97^EuzOSgAql2#-KJG1yU(*@n96ybzh`V~(G z6+MlG*BNt#PP=o=x8akwXYoObSR@}}c%E)mjv-vCW4EHvW!skInR69kb|iQ^VK)~k z5H@ha`a4Zr!IpJ&`6wlO|K@B{7sI}5z{dr$Y)NFrL6N>Pn=|g(1&P;l`IJ${Ztk7} z$A0(Lo4cO=J%dPt_)?T83K~Y*5#Rg^uw%UWhrIt}%dtM)>sw>Bv`Q}>qw-YZoZjj_ zMGx2B(ofGFtXaS8)D&Hn({_5!UMyMhLevnDq1(nE=C6y$A13CBvMJ!|T8raA(0936 zaSrDU*5;s%1Kx(Cg%eiK&`RiV)iNj#wCpO<#dToXTVd?_zPAF=b%+KS(-^Wh(zBhB zH?5-JkQHd3(6A%j5`6oclW&H0t@+o|&`IU}B+T^|`Jt9O)w=nSmNL;>q(L)v1F}8F zD1{+p6V9otTipe!%1pi8_vXupxdh4wcMAsIc@KMH=v?Y=fd2CY)d2|wzi~PRaJtIu zR*B=8=F9QtGfORTPNPuB<)^-sk5~=XWHnFYn3QcDZtYL`NB9mbBEG#=C?egxR-yrW zuXm?z+B}joF)81eEG`b}1g>;3z;qkDTx)+DPoSeJq#O!+p7@z9cu8qh*aWym(zc5~ zL=vNuf9xLEvQdTl{Y3F&-DB+=(1#CgotBgqZ5@L26i!e@?YnNQbPRDtx#VCN0(a|j z7;hjdwv!zn(Sv)l0To>yjR-b3#vSg zYT)H_KlfM1^0W=obR%_SiE71jG&~jN>Ne@_I!_>G0YKQ;I{~i5kXIChFT0hNR*T`w z2O{CiTLz;#U^&u%fPcM>h$6R6qKMT%;>ZfuAd^h`mJf)-m*?Qx)Qh!Z+PIjW4QkZK zg|I0yCji#WSJ1-V*!Hp+{^WswwDj{>_w#6{0ISh2>MawvDbaeuOg8z@ke9EI$~v5s zex_nvb(e)aHPm-$^_2>Vx8fPWR8bhMAXF8DSDD!wadyI_pLVMQkIR{+CH?WE6a2Mx zqZ44F6KVo7%niY(*>~m>3G%6?831G(Bw; z3^l!B63lKCY==C3LGPb2sxkT&!_Q$b0hvAk2A|$C0%qM0wvLmY@lLO00a_azM^-|$ zXdKZ2Z#qT_)eHp2dGg5=2G=5@b+kvZ*dltuR~7yqV>daVFLHxO@VWJU{2GI?w*76qr3delt|EOz$R8vefFKt86mns+@h@ zC_^eb?m{uQs#DHU#LLW|bz@NMWG|DiJ)2r^8FBNIAK@`zpu%6Dxz} zm(2a+^0!|=>&O+|TaS!}gCQo<5uh3Vztm32gF3UP*Sfw){kzC6GHtjx3>R#MHincc zMq1!SNOc9G%8G<|PvWA;{5Q-{Fh`~5&AKALxMGE#f^{@t9@Ro{uj^UPVoj$so^Gwj zpW#`2o&HKrT|E?SXzwYGhDj$oWb2ZK<*^D~Ie;Bd%eS8~9Y}JB6D_nl%s2&dPA#c} zkKCv_KFePZgGbm3WIX%fFo5AMzs0Ig+)+};6TMMYdfOkuhO`x^dGXan4bb0eK;^ zvkVxFrFlX?3R`8}o>M{3L}|Ja>_nV9M&~yV#zMXVi2Vbu=5;g9qzY>o#)X=t5CCl? zLeE*tI{<XKaA7?U*+Ja&0k}lkbGMq8Z2q4HZ+P^I*HvQ&saT&&f6b-Y9TElB)mi zL5-^2GKb>CF36>3gD|Wk)sHajzs_}8s(ytB3M#Fh051(Ht8Ufx{^s*5HL9TZ1G&8} zDju8W-0twT2>|`3Qm$EJyk)*y1WjpMmQ`13+NmOB;$nhb(%(7z?S{`f!8TzB(3n6cv z=R*l@Xy9x`B8`sf+{e&Bz%|52#ZaHM(*}l80KFr4}PxH<@gb9TC@GJhVneH;FhuwnFTUE4-+@@)G&b)+CW7S;NO)*K(+Y^_f?}lZJP6R))qXkd? z`tUY*^5-c`1eRf|<2_tz0Kg2qJi}?+>m7@?uua&}tO-a)k-2S|fwSey75_#iBmyjc zXC!?TtV~|T>E1UN_c9G8aUb-!t5r!RGS=7^%$rSTWmghY-3oN?Uw2jD z#se`P4^WjZKuD}80x9U_>JNCTY3>4CtuQEMiZzl`2kdYr`RQY~5xA)QK97LGMP!3^ zw13$?#0P)#2%LeNE~2OtqexNst8n(jXI+Ro++popbJUmztEQkUc;G9GCG{C@yKoiF zrM-Ec>NjUN>zfQ2mX|GFgW^1;^W-UMj-LG+Fq{xZ*>mWMfnp4X^KiaN?Br9{f^kXG zIK_8_<9$;dls`FMsQ=8LuKfY|E9FU)lcHD+p3KH2y~98Df=T@~{XyM;QO)0wFEZ}J zr=Pv@yTxNS8;>-Xs)sd`@`fSfL!cwAC8`B?n(2F2hty+0vx6suKq-w-ye5iuMz4d9 zU<<*)GP<_oC{S@+Rb{dQuIY1Roo8VHAk7B#V6?zXv-0A>2 zP>4VwJ-ZZYO~$ZMBe~gP6Ec7C%*Nz=SK-_lkL~Y!f;uNQx4n&9hZgWie&3!G`aA5s zEZ^H@=rv^|OLr6SYb-Hpl>J5i$_JQ|$7j*N@mdYb4>Y{It%U{Ia_cyX!vwu?HlhEb z!%K#s@Q~@beazQeTS{jxa)ZRgpJ5wsd$Axp`gNmFE||KqGs6W;2DXtM&Lub-80$4M z1}(>pBE6(EJ}+4AzJRdc|By#KNWKx-3r-F^qkhOT7IzmS`7z=oZ{G%vI=LWhs`f-` z+Rzw$7C@MZQ_r06lz3>>uqNjYnm50<*(sWVf?&Q}^JvFCM z^598DtV`b9p@cnnEI$y=*4a;Nf&FZPavYq@pl{O?xF7U#7gXyP5qz6XvFE?XWQN1W zJt|46b&4)AC}n+Dz$(6w^rM5MWlJSEs#E=2Ae4I5a(#lAyY7iG_B;k8PV0(@Yd4IGF23=3)!EJ$~Q6Tid1P+fw~}Vjk&9*qaAmbwM2Px+PqX3ZGH9UWlx>p z0V$P039K2lcUoahk ze`j{`cO8W-5oeTtN_qq(pf;*rMl}&ldM`~wd^31|w4Kd2umJz%Yp`B|hfH_wfTOf+ zu35W=Z_){;R!=stow7Hi=T9W04UDyK)8-MhBiJrp2iQux66i4)3CBHv4ox70gYy^$ zgOum&#Dk+Q3$pG|BrV*JO0v-aK2y~Kf$wqi( ze?)r5oH#DgkMc%3#7G<{)S#O(Ac0H1*ABeqSpOvgUE6@J<%TK@`hFHgy%PWKu)T{u z(^nleE&5aVOLNJ`uAO6g3`o~sz9XQS9ENt(8h(ei+(jj^cE#0fQQ-J6d!=dEC_NN$ zMhyCzR1|d7s1FuXrH4B|w!_1jK7o$f_5VOrddQ#D@%3YR#zoZfm4AR45Ga22^y{M= zA%zk)sc<7-$}A)tY>!v;QcDo%CST>x=r36$UlGu4E0-&G0;{w;*|G^ry4d?^w!d)b zhr8&CuS=ZP%Ak|r^`OgoP;+<~MbrIsAW--k+Pke}SgpZsQHb{#O0%ubrG^x?jEqg! zOtUQ#>7H!zb<_rhbmTgm^OLmlhJtQWfK>>dsWBXjtjA2m?L7BbuH9>FXqcq?cisPs z1o$5m7mar08g;ia7cQ5oI?dvJiS4V7$P3|w&BAe%Gk`o@z4_sTM_i9PsJ2%tAr#Y{ zuiTLfg!>aahKI%uba~cJdvSKS$KG7N`9Bn(OM+W@EdfOLE~#mZRP}9!f`f=oVyJd{ zi0A<)Q>sW~F~C7A=rY^b-O-ug>;Aj^-->*VYE;)kc2q0?ZaY}1@79nbGoJzG)XfMZ zw{Rp{mD_Ba9ag2buNpe_`y{t(<2t{tir86HN#y7mz}GHCyUaoJFlMf^@8;#`3=n^S zBm3hFfc!$G)aDIhFkb%gGv5Ig{3;R5lx0inYxJK>o8?QJ=+GRXU~XFwr-w?`*9ok2 zy$*?55o*o>4l(+hTMic+06z(m|1AR~=rU8yXvx~DMRd0_=TyzG`imI4-p=HoWxNyt4?f?p-ig>w)BL5($i7<7zfDkOgk5 zowsMJKZ0Ee6uwsOKSD65vzHEk|GEz4E-{p5uECwe#aga**%pJGw}0g?I-2>QD_;i6&8 zIiOTm%x&8toaz5th)hmpZr~bqQ)urMYBHPj<3wFQBBNc;c5bC@4KE_Bcw1CAW8I5%{BQ8$aPe5QPp}d>VYqnd&Vy;;gw3W z{U}n9;;`c11ukQ}Z7&7}RLTcL$3aaFlCnSfzn}RZ??^8yYNWnpBu1;8+pE9P+gvKT z0YyIkxSuHBaGBt|G<5rDbWx=5P9Z*Ow*95uhH;O?e-Ghhc@X$=QeI38Ry^ya(5@;; z8%p{mBem;CXg2Ehhe)xBnkt!{f)b!O>}Kr)K1hQ*dJ4KLD?L@T^A=1;k@v89 zh?xL6Q~QqHZw<`TBcu)YbLL3+WcWy}G($ zwciYRITvj{s9TC|FNkiRU^}3492*ndUG06MPdN6z_VlZqE``!x@ZdjSb_dgFPw;80 zrvU_&0Rl+g*|p82j(we;$H8-0WZ_7Pq1ZEJFkt@rU@}_6^-JwMH`UmEa-lB&14`d$ z3nWR1rya`E^dCqqj&E~#wYgCe0`YwlA>k<7eM(tw{`dKA(q$2N8#CSJ)AX_Z@q;Fo z#GEgN69gYvoEDM}96t;ptooE-tiO;*A8ZCn-2Hkl!VF+gwLGkDi=LFcbQg^FFYtw9x>TO^EvQtaWSh^Qmv}hS>T3 zx^SM!GCyR`^oY)v(5KMUSKII0u(V2w-s{SUfWl?=E!M@4z}{kO#_9~ngn-~Rlow?+ zScP(jYBM3Dlsn&HNp9Py=CZhlN$UFnO&ZT+hOK;EMzQBty`i4_Nb|>2-n5IKBIn+u zm`$WpO8RMxL2RnAIxS%gP1}&WSck5=;)LW-i%VpfzLl;Pdd7x0l0sw@T*f>fXMw(d z{6r`~(hH*pFYfMwRB96_z5-rf?bv=BX)@n_WhyF5V>l#Et@OI&#NE=_frhGs##Gs9jG{9cODW#7cp7P z9T(B2vw$0H$E&abt1p!pKt4enCo1tonRGy^8>f`EbzMn$`m6z5A@2^=fO#qE;jG}! z`sYBVmoJD31l<^VG4BUj#xH z4}%frnofaJ*PvI8;CuOd+2;II*a2=Xf?u5pT9l@UV4s)#FM{m|=1(=cRC1k=C-#-` zBmxG3fkj6XXr))6RD;BA{BqLk@>QmxGAyTe>n4!#q;y2T8QIcge*x9+pSxMlH#FsI z`o9!|81h6MmM64FG{&WE1%fhX`3Q~70zzFm)luBZF&gNh)|f#7`bI`CgPcD?lOlo= zTi_8<5c6gKAc@@D|k;$t=R&6qx z5%xa-?bU0PHu{uXuq&LPe_6|uLVRsABoXRj^Hi)9;zbzR=5$`MXly7~A;ijj>fVHvc$4-mp zAwP3Elz+m7v(m#vIIdb~Ptbo_hy;$V*Jo_01&Q}$l5}#h!UpX7IekJF4h@PVSoF=B zjIPd3w;sWJf3)=kc3fhA;}_F-!GwPBe8(jyqN0ccGS|P^Arb2N9ph~|c}J}w%1}R9 z)v%28#4qeQ(jYIR?-08cXVh)+gz*azX#00Bk?qP~PpYRp>t84tJ=2CZd7QN?dk(Vg z59Cu5583R0h0{unKE#z&N4%TNB6)$fTzRSYwlf9ZdX*b;x(;iR&BRE;x;@sqCqMs@ zxb@=z>!IzyH#ujW6a^U}R|I2PC<5e-9vycWZVdbzP6ch{!2m#C?E><|?jsU3Tuys0 zyKQ`qHTv=aobzCF6Ev79|2&;ZHheceeSiINh18nQT)Ls-^X_b|Q#$4ci!oZG(jdf^ znlZcml0x1s0pE05!S9#lh{{qN6 z66>6SR4jNj#Nkf%2L5$>TFShLYRii(c$)y7e2vYD7R<;M)2YJa3KXXG%;DNkST#+< K=S+~)yl2_ diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/NOTES.md index 84abeae34..8114761e6 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/NOTES.md @@ -19,7 +19,7 @@ The two XLSX attachments of [Supporting documents for Budget 2026](https://www.g ## What is staged - **Table 6.1:** every cell of every measure line, 2026-27 to 2031-32. - - Fields: `metric: revenue_change`, `unit_concept: gbp`, value = cell × 1e6. + - Fields: `metric: revenue_change`, `unit_concept: gbp`, value = cell × 1e6. Since 3 November, the three total rows take `decisions_effect_on_borrowing` (README rule 10), with no value or sign changed. - Conditions carry the line (`table_line`), the Tax/Spend head (`head`), the fiscal head where printed (`fiscal_head`: RDEL, TDEL, RAME, SUBTOTAL, Tax, Tax/TME, …), and the run of headings above the line (`heading`). - The table marks no hierarchy beyond adjacency. Two headings with no line between them are recorded as "A > B". - The three total rows carry the package keys `ab2026__package_total_policy_decisions`, `…_total_tax_policy_decisions` and `…_total_spending_policy_decisions`. diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/claims_staged.jsonl.gz index d4a027ccda2c4d1cebf20a1dbfa3bae0c1da196d..4c6d32a83246aafaecda27b8d8d4304823892407 100644 GIT binary patch delta 9463 zcmXw;byOAY_x9=T?v@TYNFBPnIUwC#lF~yQLO=wO?uJ7*(nyCiNP~1L?GZso2 zD&N3aPGm3E)rvYUF^POtna(Zlr%L2VJZBm0xLT}}oed%_o3%nv5)g{%`bK}ZzPy)@ zK}qO<>Ts|n($sZ!`ncN4tfgv*Rt z#V9qnmUetg6JE$!JT! z3GY$fXaqE@=NsIFN^$5Pb%@``PEMB{N%enDI8{WtLa)KfKr@$mkEyQxE z&C0;YZ1%brwQ*TrqU69iAa(!G%mV^9!UW5fhg7{J;S)l&2U0amyioWa?HB6r1>9!{ zvw9YRlOYV%GII|l6z#RIq9{S%y-?p=S^;@5wz+2mYGsS@chy#2Q<<3)u6bCjq%83n zlKw(HI%NbM#jP(zOI5#%SAhVPWH9=hZ=L72NOpQH>oFECKkyr8j?#ZWrif_t8%Ztfa-Q zQAY6y&(xh`^>ntnjuer}xp@PZ;MLUdrXM^W8>(Dw;aevBEfFU|+2n0eKygP}PL?Ja zOFXE8FT7Q}bJGWmqI(89dm5kb97b64V z;W64=VEMc4Y80o`yL~TAuj%i5FxI6Uy+f1J1hmMHkv3_6tjOYh^c8$}Pwxcc@mLWv zmg?n#InBE~22}942eCY8e*Qexs1q#bVYWGJ6Btt@ma%s1e!fkfGIE5ZDCu^;7QVu^A`kQQ73@Pt@A$eU^@%z@v`!7;n-^HE05i!Gh ze7^SS*o2jse3Jq90jZjm-SBS@K;m?$^h*h2Yw{H)x0e#D9Bz@CswNvx7*6w$RXSlA zKa16E=qttNrCD#a=wFKj^Iwb8p=AE_n8Snh4|8rFOh(sKQD6`cPlao!*4tjC%`OeP ze9Y;WYBT2>A~7^&yT#JW?rhD*TTEz;eppg!edwceM#ZWcn_C80x6xIsW4Dlq6+eD@ zD5_B0lGPi;OH>sbnmR98Ui_L$AZ9jH>-9P?Ij@cX{hiTOd=a~!&h+T#>7?eYl7LQ@ z^Q=EvDu}h|s1jZ5);%Gh$LWi=o0uuT#XD215j^&7t`1tdQbR`93a{>rmtN4eRjb`) zom}N~A7rO_Yru1Wz2ZQ=i)od)el5jup&Aw06dbnw5Hlxdu)YVYQ48?rGymes8hGKi z4?6mQo!L+Q5#=Cz&E)7+%4)wzO_~w=K@=R)10R{m*0Gaeydexc7r~F zSyj1X(dviPU(i|crv8i3Bwwtx>VvKx*ZOg>*_X5}Q@V{NkhJigWQs&n>pIjge9ryU z(eX&6Ti(4;6PyvarZ?GY!#T|sMmqnd^7^fcS@Om#WcQFgbI@cFX`qYVP_%dd3yt(X z9^`Y?8SrE0QM7Zp`d<7>cMPM`n3TKIf7-<04!KVJZ-C)P#?bIu!a;7Su;Zx$al`VE+A(&|y#n7h(|FMw^XlgG3L6YRMT1tOPF` zZVVQ04K%wd=L{2(`|9qTM9VAZq!G0NLArh;f_OHK{H)ZR4hiGFQ#};TXYtyiJP`LL zf#PURhatZNG8m2e2R~TOyh5r=25q4i7`E=Lq{suAItrnvF%VsTQKK)~mcnV+rXsuF ztpC|rKv4oRwZ}t&Yap7z+G{o;o5A{_#?ETu)$t93y%Kqog)TThb9t++2jT&AZ)KQr zU~P7<=kc9f5N1}cu;jj6R_aU)LU^O-J{ZRv_EG;5J354%QU8r>Xr%UHZV~~f22?eW zKKJqOD7ZnJ^2}F1*RX7*9R#O5CxKO@$U=Sz$X2`}-k)Zus1X`}A;MBCllsK!# zD2#Y3|81e3t+t^s&_<&)JnJ7@GBm{m+UMD7bzau-V~QnD5@D%P!D52RZY5(jN_EgXQ9a!JY8n&*qpi`!z43|mXxcj2fz%A$}tJ7?$} zKzYX9qVZ>&tp)O@NYmM3KppE`K>klj)B?u2z_O5Y_FX@>bLE}y6bDsHZqbaSTytI4 z$8*ZUP*$zkJCd00faoIznxdtXf_Vj6*HyW~Nc*F0XQVcp2s|o@LKA-RBayNr2G_#Y zlY%uBT33JB!pP_&OTmp!2Pj^a_+w`|P!ZXEINxt8Ii2RS-CQ3h_r(5fE20MI|M<}i zgRweyp#c1`Jcqi6XxSo#ud(l&-OJ7UFa2@>W>RvdGgiYc>2+gm)(vDXS2w4tK~gH^ z#i=Lep_H3tPo`y}LCNLCFKVXGz{Lmm>U@ulm0xa_O!8XNHdLKH*`_~@B-H*j1B?x~ zPi_B-6~_y8?T1WWVYUc6{QyQc%g`S(bVj!9Z#Z}lokG2*C13np#Hr4o%-j%WTfctadA-B>) zzK_2Cm>0(u`$_yN_%1G_Cl|-#*vdy;U&;qBZl9Z{2f%S`cja|3aqAud#TI$%K!`iq0s;HS>E}|HHpQP$f!Dje7<3QQ97WwaApJo~7#*#J86wi8N7we97 zsvIWx|PFo+m+8fD~>tGFZzBAgTR5{Nee<_AOyNz5Ca;y>AQ3bxk?x--nLc z?1s+i^)@3?tcT6J0zxh?_eK9G+;Exrf=u2vK0uoXdF4f~g!Z(OrTmt)#)4H0o9(tF zuG>65{i;UyG4B20Z9ru~lV^DqbF%%RkE(se3LR}7!5u-kJSMw81^A-z6?i1x9wuwA z^h5MRlfXSbs7Q2uJdWlq+Chs&4szc-s}GL!=o8+~p1;<(DGAoQUVg&qX^8SLSQBv> z6ISxwi1bJ}$fj)#y}*T|1$%q-B>#?Q+s!sk+(Cv9ll3HvVr%?9sQ8R^@qR40iK`Z| z2XQORLvvbqO7BWT0-(Pp47S1C(*I2<@%x}Xfb!z~Y;Y5EH1hvZKT@tVcz)d&UcBcC zX7<)`&@3tJgtv>bBf}MaCEJKmF|^#^2hK7l_+O5}-`hJe%D0Qu?bzi8Z&>~gLQhT& z_6*C7Kp&dyL^XaOiW#t0t{Z&hL^%_eIx`)UthpKnjzxn8T$u9tIfPt4rH#qsefxfX zYX#CQr_6=o-((&>xSCE4{U$z?6A~#N4h|FXBAQOhzxhNO9B`;rIs%zM**%r%SDg4% zm+!TJ@4MI~QoMS&A))4f9VchYLVc5vfSoH+dm@N+N&o91O4MP7*wNQaH`DMa1*dY} zK80}?0P%9m0EO{asr@)<`E+}eIBUB4kv^11_&4bH0^-gGtwr_)QZrjlM_P(oG}~L> z;V}%GtrDN^Hf1px#Vr|L5>Pm5FipF?xC#sJ>zlVMP{%+k=!@ha@Eii8!a~St7%DK( zP>>@KYgBkTY8IF%2ypzV&^~dfu49VlAQBz}gXI0_!y+=kb4dR8OwG2`W0c46znZc@ zRYE29U$KlQf&e!~S<|7xs`x%GWdarqqGqlS@?mIq1CblBE^Q$+)Q zAx2E&M|kKN6CLk%r^%(Pl>3;E@}K2joh8Oa)&!-w1sYtS`rcChbZ)K4YdYya5rlJ9 z(@i_nHIe5_b_a2A1_=Cu7cnSsp!S;1Mu@7d#J?ps5%4>!p_-if}>#N?Z6^pWBjcV7EaT`8t}r7^to0 zKp$zHU|=IiN^t#JVIjSNno*%iH*yf9F-Q>eJM4N*DX|U>V7n zPgckQiEeEAb}rC*5w@d?EHlQGlCdK}-okj-7r zoPwax$Y9Q9N$!JNmW|$>k1rrg|MyQI>{FuG+Aw2TumOq+{v^PBL-MokGEg>yxwdK*sN@kLDSGC}CTYpAD|t!Dkt7yP9=XVfjP}-KR)>L#;9Ofi z#YhJo8tr;y11)Q>iisS9Z3|pb!Y3PpE4IdCBBuKsVy-fApdIimV#JvdxVq_fQQqN#fn1l3jm(jRSYnP2Jqf$heV!tiN2SpNuj$ag2|@Vrcty7+uo)$K*A= zEqbxScA|fvd&vt!wb^J!A`s`-?4shQTe>VH0}9W?Oh!;>e&HtDh+}BO1$AXfJav+7 zP)ZPV5jYNF$RiE>36rcbN=8u%$x%0KNCx3239{+Rm^$zm3JLg7Bw575jQkC3;}r>l zI7TQFvQOx;jI2yTjV61~PGW9oECV$L>NC~A*iF_v=fXfDhoPW2`oOy>CzDhq!yX6x z(hd&#QE2&*{u#6O4drs&Zj_h#GnH8j_H&Mp194f!|Np-UHaT*u=Esdo)JBU#<+C-RFIdM@@CqHK@|&r|hwA zOQ8mj6d9)^lVsU)n%liSaG)N{`7ug>2gdNfL#0KD2ovq2u}mD3DX1q!K{{BD6$J^B z9K9DEC12!M`brt;N@$5*kG#w;sz=faQV64^wo*b%icr4p1Y>Xs0&kRN9f_B1n)?@c z81R!u#*=iHu{aO|T;=mPXZM^!W6+PwqrFG;)TIKHr4R$c#HHy{^t7zFZc+9c2m%ZU zIV?Q(WO88d#Cw>o;D|f=(sch^wkaUXaXSCs#hdixafs~72;c9!-lRj+r~_iajcndu z-aQqv;@7o+Fp}4j9^WpAnI>!fGXGIAUVM9Gyu_e4cfkumY8qdPJS8U$El|n1EvMTo zEn76HE8h{-ES72asp3NB0;$Gp;Eb1ap3!J=I^mZ9u@r94&w06n>t@wZF)YFwVj7}G zqfvXtY}7%-9fne0+d-$2<}eW<#>4~%(GCz^gpn^W?78scP((t$sSj*PTc1eESQ9E% zg1Xf3&es;KX0jy1iCrD3PWBqTI?`t;w0d6IA>7y1nwD~6F3p!<1nHC^0d)!Slp(E! z{PdNLktJ>X>+*4%<8kn)Rk&}`7o^iL-=vY%6kn&~ol z`+1iDO0$u8AM+XXcx*7SxOdMa`Mh0!C7V|CPIaWWYl1oM>MR#|&hS;}GKXZ8jTJAaHnB?SDdvVRM1+q~LTNDsTvO&<1 ztyNl)p;+US%V{@b%`66E9U4{Trl9O3_8SxtBpM@vOR8pO^DySVS=K-MlyA2%D(lf` zIPhxUoKc#Fn^J6Jys`CUb)A`lA|@gzW!GnTtmiThj#9nZM-BHAMmXb&=;n|dSn5)2D+`Fk z`SlalTta2y<*#%yhjO&80+>dKhxMC?H!ZfAx#pd9#A*)0Kjh_q;B`oe5O$O)XQ6!e zfsKE6Xur;oFMz7&1lWe){PmC$cpk3zHoyJJ45_})eEi5W0lCP{cIy!$KRf<_&<{#L zlq#FIb1pJb03Cdw93FYl7g$~oa zQL%dAc6}f}Hdv8g5f--%?}WZ6M_oDk1jFNkZDtQ_QNz`J(EZRya{M2zmf9SEHoOzO#epZGIR~=g(5#y-GWYqgNC+P2;CERk!{l z=D`W;A`VB^8;{S^cC}woP1y}{ ztGoU0@CfRjrQciV!hmOB_lo|4qw&}KNq^eEp{!K7V~Xj5>Qbdqr&)Ngb%dCKP}Kvq zj-#Oc?2fmCqjLRP3F0IKn6Xg#g;~N%!gokah{<$-ddkx+l+;wt;L{)s$)+Gi&f*=X3Ez{l^!Gb#bXwn0)m_ua9MzCiM+u!*Hsh zaEjNM)lh_$yy^w@r0~=}ReU0c=td$WXUi9Z^Z$e>3wQRVin6%RA%RJb*T@1!n27BF z8dv51R`B3fpK4aa$ENsHcQ)a+2JTD7I!;Z^v2_2C8A}#18xKvfJF>9ky!_bVfomvlpy8i(!4#1sK{W>>#nD zgXT#0Fed5b$+le4L?eBurAcHQ&5&bu1v+%CPhp$sP4Y^iP&utLGI@ABj62)vH5A8W z_m6-A=M`G82(d9FD!31S>XWD6Oy3uPl%vUb~u~kM~C}-QpQ6w3HSFnmrXU7O^??O z5-2kW3V|}oo!%MTAtZEPwNkP_X*)J##3jC;=kR4IOp5)%=RPkslboyasx4O@TSkd3 zgUv*cDx@Z~q# z%+WE~+_X4MA*9wF3waTG;GQcOG`A~vaQ5B@Y&rtSVVmI%fyn5`i_A5={P0Nb z{Q^$U*4w2ddm88`qsT~-pXj-$qZ#3meKOk&E??~3#az+w8b3t>wqGVaX^`D97!8*) znAN`NyM5&5Lt>c-la*#HNP0mi&_>ITjAaxj3o<-_v#t}%W>tsF!dMwcn9>QZY@Zo| z2s?MN*Jvb-#8I$UnF16?fFTBDBP-UdAViMQ97Q=9G3uX}+K#QenC3G^5@j1b>8l~l z$TD)mK3E4-5@Ee#lJ>)4FCoPW=+t&pHB?xF+qRBgHO9tF6iwPIa2tXNl5&&z>u4!Y zBg69Y!z8txMw9M?R`P?@IX%=@v>FX5zdh;(+j5FHJ5fqIK-B8`zT3!hCBuuV=up@0 z?7q!+aUq$be;@Dp6rp1+O~QahFm&@O{hOvYq7DuqN0K>r^0J5_2vuHY2P+my{`ns! zzq}JlO&EJLqpDz0)j&eTYkiwu>7jEZepl>jBm=~%j6CSostS93oK-awv1b3$m=SF_ zDN6cDJrR3iKh{e$>~> zxXRdoxGujUjCN~^+R+ki0zV2v0tDx6)rwdy#vQ3X(ywoZwxYNcK%_eI`O#5oqU=#JGKd(puRMSUh2O#7rx(zM2k>7&n>0w~MTeU7z zptkl38kPOhi&@X7)feY889g#%=|CADdm_8kHGdSAGHMBu?h2vt)mq^T2GS&Fa$QIY zOg1P9s}P2!%duGsZaU&6-GNARGNf)Wa`nv5hEV}9{Q?xlv>9)5 zVvm)_nzQsGxERGy4xui0utwZ1kEo%WO6fvr0T&QR}= zf*3p6-*=zPDggE5%PzFq3HRYvXHg{9x!p(ioDJl4=xWQS8IDfPwB^9B_1F0LN6H<7{s<#_@Kvrv|GAE*l?B7&nn7$4H&4)T>y zgE@Lx%ETRZMfN4ss}I2aqnLFgeSBhi@2$iZj}UtE=zBLkR?U&V z2Ce*3zk3WUN1Odsgr~681lekW{oF5CU;0o-<<*TYzxeCQ>a>RP-a-wO2fs)%hi20q zE4(2E+ObSal>+%)%gOs8tzN0>+|fFg2u6(Mui3ad%2b8x{Bao?FouzA!W)uai(}ac z62)6`n^3Bik=|?XNhY14#NkG*)@e8_;6j@( zZBS)Kp_lQzFe>{?PHDXVaWVcllg_{aLAX(x^%7^$@>`tt?5Hj);-;e_m=#$a(jIQO zQ6IM@*;TYsf3@N!L53ZdJ%n-~)V6Pfdk>4=(xAj&INngE?PV2sC3l>ZY$qekh_=7_ zuh68kMRA75;Tw(D^5yTLQksgrJn>$vs;~BE1UT$3FTk0XxYSn(`PP{Fnh$^p$imBP z&qenvCXqUcDI)JI0sC&vQwIOWCt`z2UbyL7QTEDENA@v^{>J40eSS0%ZhnxL8uN26 z{kvYX#!a*Dmb57rYiruFrr>UgI}{ughfjhWhd8uMHFerH5Kutzl|#|~ET^WYgxCtF z$m=~d2=;?%ByWv>iw;Od9!0Q40JXAi-=CwZ0kannnq1ypu0Fjg)xFyZ1q`Q5{CwF;?@WpdVtGp3BY<0N?1xfXJXmj0 zx|8*igHekPK23lh^vA6Q-wvbbUL1VYidM?=^eZd-?Mey-We0N1Ufw^ zpqXbuIj#MO%7AhpfI{Z?4rLBvih|HZu@|`Mql=fq08m08BDzv2rQssDKwmNl3LFQ; z=LL#Jhd|&_HW$h)XJCVfe(_J*teEYNqd$ZnJ1m zw&qj;o%(u)`w;zW^#aEt{WG0$7{LGLFeaDi;F9XfrGwXB)`dzw=k=j3O&Igw@vi|| z+Fq?DF!x`Ctw@xtub(Uu`^x*pU-syTnTBOL|L+{m+jz1O>t2`G z$TeV2wMLF#CH__q{GN-dEc23qcmnft2jORW$LmaDc8zN6;j0am#qxpG43~(QCDL97 zOcBC_y2-@*7llW4FBz_#7$C-O*+Neg3DyOZ!#6N&&bc1sJ_;wWY?6M}4$c6fJVGgT zMlY?SFKGQuSzL|HDVj>M+SDm>ND=-XTNM@TFZxRsy|l##Th*UdHPz#V%s&(UXbpdP zSXJ7BP6rV+Q7BrTkJIic`z*T;JDH!IpMAYF9X3!ba5*UJUj;4 z8-y&7$=V6~-4?G^eOht&JzNa?@h;NQbkP@uByd*O<+l~XT$T(@y&D7>&Qze2BZ+gvxtK1R+YJfpjATH z)hw)GnRpVub??4+61`u(hPY%tfk$JD1k-rTTm1P>BU0(1UBGU@d{~BG@HF$}1dT0- z&*AK^juT#5$&-kB9A6nFI$5n=-)bt?GF<0WaSv>6xQZ)T+D~m8x(+9AXn=a_hTewM z+i=o92IH^!iEqwjuwOD!N4tHTn9pP}7*kM7hACVv%$@AQ3x^xpY<|Cggg;av_PDDt z_w&-fpE4duUGK7uDxVXydwA1+KBSBOH75FnzDc@ZG&O!JuCG$d&s6GMpi}%Ds_bL* zv{DP{O%Jh@+v?T?Y>_(!FM zHjpzyo-Zvl`^PuxCvSyGP52_u?xGN%Mf_jL%u7 z$Hj;Clt*Ke5yVjc5r@waxvzrKiop;_r&fm7M_YhY&k?CcUT(C#Cf_ftCt%+=quC!1 zTMdcM0{7pz|B>$pY1bl>x%A!U9#P0=wEN2{FWa&`6+HFUEt*45;Ep0sp8sT6eTpNO zv)SY$Gg0%|DnDhBHfBovR*B3E&u?-oZ`X`j);SF^BNfqYzPK+#oF0pWA~G0Ynv_bF zzs`M$bY`?>^;<77vp?i}no?R1D)MTtD#)CHh#O^iV*nzgisPq+h`EYjIr&*(G7)>3 z9n!zTlc)U9kVxK6ILm5z!|>#Y|qe0wv0?-xm<_x0;@yYtryi>eG1$y`Gy<=guH zZYKJJfWMxI4yW>4w;OXZ>9@+Rsd*L#yTi0ihOa4%%mt%tsu6?{GeyX1xqlyj01`9L zdcKw97`UV?$H2#b-=1J&%oMPCCh|^W%@kx!iyjm<%sqZifQy^pc_U!Bn+d8@F!%`ZSejDL8CmRV3;1fx^G3 zvxNoS6VQ#uMNB@M=awK1qvuCWV=hlk%Kj>={Ln>vF3)3$!FKJpwL;5U!+g=OD_&ux zlMO_JF8t)DX0+L=DQ@iyOt%RA-QF z_!uF3W0>~tS!e#efbVDVXH3>hablh6%d$C}kD4q+w9#=nSKF9PyNtF#t4m)`E;)3( zl~@R!RFf_HgZA$m3*7^>&~YY{MR@olcx@rIxETKC+BOMt-<(5RkltRIoUZlhBfshV zwWh~Sz1`k{9U`k6CU#J`Ja?L=EpqeYVmsy~QDGDgXqjbraT=@_Joq&^mF)*Pix|1Xj|%3{eJ^l>AuEa2#^) zV1)ZwZoPN>FJ|9e=adEugvj~@Yj(9(_KC@w3LjI|InK|Ha%*Pvzb z{C0IDqO@72-%+ag!Vk?*fZ&TH{L3u)5}lbyKzx!~8-$mC?A|5y(&Y6Rs@^Z{)Yg$- zuh(+wW5&xSnfX#>Aa~S_Q@qLk@EF@; z3eyJx&z`rmm45mNtCs#-^?9%mXg`a z7lEO2V9pDTpiyaB-U!aOeQO=?I#KFs4gD9+Ge*m2S^L!M!Qc6~K1U_VZolPjHD4?7 zX1qTMu0rbR%`RZs$o-z`GK=I~LJF4)J0aCH%3ELL*&D-*ZNyLh2&af7k4N~=X-NN- z0gZrt}f4Y0Aad+5+gs_fs;A|2k(kX3BeGq z`i_I|^G3R_-j|F>nCblpUn1rzDHy5rgwBF@W7i_4V5uQy)9R^iQl+jQUs?1t-=>y} zzMIwE1+4{7S*C{AU#v9pgxX&$8dEfZk~N8}J>3QFS&J?+q8|N^O|IwfC1zx5Oyh?5HynS5W24yZC|4Yyaj21ZS}uMfR={N65yOG3;^O8_D(qF)KlkCO>kcoGY0i@Bt5D|GcK=Uq&V5 z(agBf`$zigw|;N1b=G=r{~*+_R|e#;xjYc+?dxPB>&Abub;9;f$}U1F-4jQz!dvP< z5zUv5nX4f15ji{{qZl+_aeJcoQ!lZARmQ5?XjJR6qvp=4O%P*PTb{-poIrymhXXFb zEM1;KuKZm9tB?)vwByV4%O&-t1{V-z&KhY%OO~n;@Vyc)oylyVNjUMit!62RVDc@S zVYONdVe>Fdmb659k`(>fN=k zd7B+Rd?J~Hw24%4_b6Uc;F#kdEq$}wL&&%G4lMHC_uN!jf_tB)KyaV+)4%N0DEE}9!&+eVt2CYMW-g7+sQkLybqK1Mk8N(ll}(HjhK5> zxO|EyiEW9(4F>d5`-v|bWRen8ws&;?auJZKgv6z% zt!jty&e*ECw{S#h#IUhU39D5&X4$Ic56fz=un<>i$FNSF7P5%G%ui6=<` zVK2*0@;W~!(VXpN69LnuBpfn0s;xn#Jq1t*KG*Eh;MGZ6MmCRbWz z%t_SdISs`{+qfP!E*2?=pgBLEqwSF0S@YB9L;sZ49ehs{p!+_Dahct$SowWU_XK#Q zfhYiM(#UsIcq1}ah?ITPpk=HIk3-6aI2AP6zZ%O=gIVl<%hy@q#5)yIyf5Qi$|gRt zyC8GYa#{Rp7fIN|2&*KNG@I=%mLZdTA5K`s_EC!MAa{zig}gT;{#`U-4+X58uqA7@ zJDmL)L=x)HwEHR_FSMt4-PFc7X>cgb7ZS%a#^y8**sAbdYtZM(0>@I0s`fK;g zrqka^o6LGjZ%O0nzhhNGu&T+cX}+&qAm{d7BZsBR6lFEfJNVea(Rz^wW{2ipr%=oL zUX!&!o*Zr8d*Jo;v2Pfs6E!XTlu|FreuATaO2BueBkT$PzE#_pioteL& z>LHtlJfC}~YfpN}Z-)4Yg<+oJBuXxdl6o~pig|G@yQQp?ojAQ{kBH15*@*`>e{1_r z%VVZZ4|5BV_hBHz>!8flgfVK7CwGMf-IP0c-SYk(dt+OXN|Xhh7Pxpw@Ta-My?<5RG*vtL!3=dfwI#(&+I3k$-nBax(42W;j z%K-T``O~!-<2<%9xu_~(533I^Tr*TX*h&6#PlRhx(=lbeTMnd$s+`vu}bS$n%V@AozUt`T=7IqFSq9vfhy|MN*; zmsgzA_N4#o0_7`Kty$AcKmg?IRg%oDzH&0}D?;;*k#=f{KncS=j!S9v5 zCw+h=*T=GUMd$sSxNL9);S)|p&{sN7%PxWqtdDL!->IcJf6hKE$f1644EhSAw6sIc zBz%ZJ8k*1f06$>j`Vc5yv%fb^I4X9F<%ixOA+xTiH6kYR`EhcC*5uZN?f9n3T2n2K@giSDnN5FH z-9vIk*}`E#J!W}#C_gb)!>c5s`X%TI4ut2*^%mHE;D%xSMSsi+e}Z0(qdw8a!*RBp zR^HhE=1l?gpktw-2L;WuT>R8$BBH{1Ax(fa5^9NpC!LN)Y*3+_yZoseH8|Q5$j?B9 z%$&_%QRfXt?G7xl0%qj*Wp(vZ0}?g;i2W>MNw2Vk58RJD@~O$3pz&STP)v7=*L$Kw+{@r zqD8EUvTF_!!WsVR`(ZD8pE%)@zYyY!Ep|)Ac9<60207Gdh?UY}g9lVD%;VkI4v?J90N|J_+U#S2M#hj{^h z+q$^V{@E?BCUpCW`_%x)_%P1Y^~ zfuHSvwslwI`G5e*11xv+mZhZLYhwx1Tw3Jm(1Cv3#+46H3~Kl1u}N=AY~KpWjwE*EoxY{iE zTOMeh@(Y$N-pKnqe&(yq&gv5NLyl&rm)0Y8N(on8(h6GVPc+r+p=$E>xx(6+kVJ4N zIVWaO5PX`jsDbOvG7*YBPMj>3y?;Ngy?>~3(0iaTL#EA?PA54XW6 zrgn7azCpbHeETOSC)TFj$v>3`nn+9~N|qU``km2gs11)|vuVU5Pu}GAl=_k9##1Ij zt(K=&yan;J3uuxbJB)v^nHv&lXMvPwz)43RL)}Y%iGgP#&tMrJm(HyFCTJ}^#7~FL zZT6eM5Cy6yic}z7vXGJ$w=TwFEwLvQg)>sPvpl2*jQ)u%PBchHnHIE)c7ftu@q#m~ zGJzyvP#L~cN}`dy_sJ_E04DxPy8kLU;)*l@dkB3;K;oLL39JRA$4|5)PrSHRrtd4w z9NOk)`uvJP6LSakdCiM>ov=zO^5-=>rGf@|YZi`-IoFnyvi*Mgk$w~jDttSKzZJ5f z0@1rZ%nWui*MhHwdh&+p{w+B-450^9fMC}va5Y}+v&)AwH-TY$``k~3o})`EHa{}Z zhwr2GkzpKyPoME#Velcj#KwMs1?Z9>k7HhPLgzCZ_p&ljS*`M&<^sT!WH%$@*Dyzl zIwRmq212m0it;yQb^OS8q+3H?X$9Qd4CxTNmnx?L@iT9P@hi~hQI8FYr;nQ77+cbP zg&V3A*EiVE71ni04nc!Fx+GQYP6s99+$jtP+2#{#d7O-}U3h8;dkih$UQ!rhb3XSCvAru~j7ky*(T*4UBn2ajc#jzZG3VtUM(F3MPpq@1 zJla(EG5u@4=L&TT?T`8im{EN5xqx!OCnUuO2M*?P>DpbpB_<~&oOO)}54YPj_pdBi zea~;M-`=Q+^0ty_j8;KsT;~y)X)0Z#(%blqv|2NK$;ZPf^BS6f2lp|GnG@M@o23yN zt2yXIaCKb0yJuF_p38?xsKm<6+TnCJQ>6zLb*a<9HioKGP%KcHfq9$MB*YVLjSs~T zTvq(iv_?Bk69>3}1C1Mg+%S&xxile=BssK8*?|XDxnZtfvVE6mBsu@Ua*V!xeg78bVL#+{)fN1 z8(Gr4Wc<`L=hD!*i!&?)|G(nM5viH=KSJF2{mYk(i1|R~el;sXw&x7tkb?J=FQ;g< zs2WsRUZon1-Ld4byGmPuM5Hz!F=gunD=8XhVDz^KcXtP_qMTkK6z=LtiML)PLguDV zKG#Vlo8^S&{Tu&ogbLQrlz#cQ=PV7Df3(HfJ8L`ja$q^zB7D5iw(*%d}|A-yCULyv`(rqKEW{&B!*=IvJ8C; zc>1*vb!*VH;9QKO-O6GHO0*(|XH^kz3 z_bDIF1!FJk65Oi;r6yna_|&|Z*3w8j9hmr*%tV!L^m-)Vd( z#2sa*(6gg%aa8B&IRA)t zYWSA!jpAu6w?w0Z^<%=?YEHdHAB=q&rY>iBY@7Cq zfA^0eMQUdKLjOj4w zDD_o|l_q=W8+wD97^teMNJp1QlgPn90yvZp9T7-LO!(20?g_2#QnzF+tiG`+QhvaL zS77M-02VF?Y*_nRk17hqt$S43@Slrl+(sm4Wrn4RXK36(sq`#8ARAVUp)U2v&q1kO z>Np9f!|LN0)P}efR}3z;9Ndh#So#Xm1kZyCKqqH?7|ZN)N)dLHs2qQCDj?9ICdv3| zT;+kFqL7+fqSf%w)76c3l;qU~yliVH#V=|Bn?Siv=w`3soii%c;T`p%%8}O1r?s4u zxDzcd1fL%T;$-qK4zjRm{w($A@|)J5LXtwEe*DsJEV^a8j@_wQn%VE~iAR4Oy!m$* zA+>GhBhi_00jpAdCqv$l=qVM`w5oD6mdCW`2_Tu8gIF_}6X3L=Mx!VN!kOwrK4?lE zU~!<44g0NRY^N(J;WUXLhIC?^qI`hg2~06wqfuti!Q7VfL_ZRLM~u`_hbs-$h;+#5 z7Ba2NM-ymcJdb8%o4YL^iU5go5qOo}K|0!@HOA=C(n|$rLJX6{b{ge_iNs$o?S*3Izi}W&b2cqcq^gGHuiE*U?UvhY7-|>=dLlw zHhjSZtb4%$#~?$)Ga?aiTOk(8O(^1EakGhc?vY_E9H!wYgps)NxZ*NuVLcm0!@Z^u z8z7^U4njRyA|&YPQ&<9{Lq5PvSA=X!DUgg>8;1jftd2! zk#M|7YpQ%b<`&5@jz}eu8+^T=uK~SHU0jyrHPMG^9MxH&XysG`T!Ncqj#=O4hz;dG zmJ)IYZBbRAysHW1?jqlaKDTGB&eBJ#N)O|P;Bn%##Bk$Kiny?9_obptF>T;|#QbjU zlRxCZ9v#RXz8MX|!?cQU!9>l)stS$%P6B6oLN}dyVenw~9n7N3C_Btw;#tCGQFs|F zYbTxa0@o}F)2!2+2x)DHGvpi*h!vQ&ACbmARdfQIE80Klg^F@2iK!{%k|bA-%~wiW zL-;Ot9V!hVr$nVn2p(_;OL2X_aI2QCPS>2A>&?O+XDMaYg=pXKiwjQfA0o8#4QYg+Wtbq4I z^?y_%=H-eJ|NpAW0=Uf$KY2dPqyUCzBr>*$1_>$yLOd1cA3UK1dYEO)@(>8xM4YJu z0xzm%FE)AzQoyCdJv95D%H^U>@(7SJ!Y)IDoaS>;&t59{@<4E&L=e#yKCERVjngh= z7gAJ@m+}zy!ZJ&4NMn@_h~1}2b?;9fSis9FwIol zv;0E6J^39`jNt?)Pv4H-go=cj%2VCKXmC!Oc-H#a6X5gbL77RS@HLzvTe~X$_q%wj z^a-bTiKlUnzBl235dAARPbC@cV_*H_`@P6%$a-fPR>}yMnTYI)uxxT%XF#x5MylCo zIlF6_OSl1BlE;B|BA_oeq2OYk*ez&-_l&{|Kd;xR4MId0+}vd%HCTQ)W)evh+?>rR zjI)O39Mqhk`g6eKT9L^MzpZhC?#8#6m~uMIW9#V8ERRVWy`xn%b1{Cmxg}o>Wq!=c zVmZAThHp^&JZeAU71e;bD&zc8OjR|B@3~J+D7-yBL;k;RmpP932dir3F5(l+8{Bq7ReG(GlihN{FP2*XN2zbUoVN0f|r$_5~=w1$*tn13)|vKTV@1m71~sP^-IAo zg-F?t2etMD{6k?QZldwJYrTE2D$>|*XFe|;>V)Q~nsnm`&yMfTj~|OqS&?mD{3oI7 z)t{dK;NUybzLGJymMQ;h9xwuZye(r~@CEAm^Sx?X%gp=N$g&+pM0BY1{V>rS!yu@< z3*u8q+3V>;Y-+=8Fx+QNg^86w?ivmEsLOXKbH9vUKp}~->3T1KPbQ^BlVPHO~B%9ycNHS~m z`h)y3D~f8g9l-xO*T})H3R}Ww$7x$x7vVn`*lILHD=HB4x{Sv59L@z?1Egb)q1iGz zdy$R7b_jxunymFx1x{43evT_touS$)MSrUIB6r8HFH2r6HML@l$4#Fj9y@aGt+)}1 zzVjqp2vu@lR=wSZUaiz|*yStg63Qf7QxH`rQ?J)BPTq2+0eAa4Sua@^lsee@0}j5U zaxxtZPkguRq@%n7!nN2JE9KpVe0f_k%qm(lH09^U|NrMlk2j-QSILh1`bU$iE#o~p zYgmWvA*utARHGTAq#%CxH8>;9r)cN#MFj7xMUfG3G znZ0g+*$tSB$AF%II<%OXeAMg)Mxh>;S02dvS@VV=1PSgvqUG}Oe?usU_m2o`IC7bw Il>+7e0Em(rwg3PC diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/NOTES.md index f745d9f14..a2af4f9a8 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/NOTES.md @@ -97,7 +97,7 @@ The opening remarks (rows 92 to 97; the saved text holds the whole speech on lin | 30726788 | 2 | `fiscal_headroom`, `share` | headroom at the March 2026 and October 2026 forecasts, as fractions of GDP | `restated` OBR | | 30726917 | 11 | `real_gdp_index` (proposed), `index` (proposed) | the March 2026 forecast for 2026-27 to 2030-31 and the October 2026 forecast for 2026-27 to 2031-32, with 2025-26 = 100 | `restated` OBR; the addendum classes this chart as restated | | 30726939 | 1 | `real_income_growth`, `percent` | 0.4% a year over the 2024-2029 parliament | `own`: the footer credits IFS calculations | -| 30726047 | 1 | `revenue_change`, `percent` | Budget 2026's tax change at the forecast horizon, 0.2% of GDP, with the proposed key above | `own`: the footer credits IFS calculations | +| 30726047 | 1 | `decisions_effect_on_borrowing` (README rule 10; `revenue_change` until 3 November), `percent` | Budget 2026's tax change at the forecast horizon, 0.2% of GDP, with the proposed key above | `own`: the footer credits IFS calculations | | 30727239 | 15 | `taxpayer_count`, `persons` | the three forecast series (the freeze at the March and October 2026 forecasts, and CPI uprating at the October forecast), 2026-27 to 2030-31 | `restated` OBR: the footer cites the OBR EFO, October 2026 | Only the tax-change row carries a key. The other 45 hold forecast levels or IFS indicators, not the effect of one measure, so they have `measure_key` null and no proposed key. @@ -175,6 +175,7 @@ We couldn't fetch two of the sources the addendum names, and two others held not - **Flourish bytes:** fetches on 29 October and 3 November returned byte-identical embeds. - **Tests:** `PYTHONPATH=. python -m pytest -n 4 --noconftest tests/test_ab2026_harvest_contract.py tests/test_uk_ab2026_harvest.py`: 52 passed. `--noconftest` skips the conftest's build of the derived database, which neither file uses. - **Generated files:** `pipeline/ab2026_proposed_keys.py` and `pipeline/harvest_counts.py --pinned-on 2026-10-29` regenerate PROPOSED_KEYS.md and COUNTS.json, and both `--check` runs pass. +- **After the review:** commit f4c7c94a's fixes and README rule 10's two moves changed no value, sign or year. COUNTS.json pins the gzip's current sha256, and `stage.py` still writes it byte for byte. The scripts and their outputs sit in the ignored `.ab2026-primaries/uk_ifs_ab2026/a3/`. @@ -480,6 +481,7 @@ Rows by registered key: `ab2026__package_total_tax_policy_decisions` 1; `ab2026__founders_relief_cgt` 6; `ab2026__business_rates_sbrr_rhl_thresholds_raised` 3; `ab2026__electricity_vat_zero_rate_to_july_2027` 1; null 4. +Since 3 November, the £7bn net tax rise, Table 6.1's total of tax policy decisions, takes `decisions_effect_on_borrowing` (README rule 10). ### Proposed measure scopes Superseded on 3 November (see "Keys the addendum changed" above): the self-assessment timing and the compliance-staff and debt-collection figures are line 45 components, matched to Table 6.3's items by value, and the savings and property income figure uses `ab2026__package_property_and_savings_income_rates`. Rule 8 dropped this deck's six parameter rows (the Founders' Relief rate, comparator and limit, the RHL rateable values and the 0% VAT rate), so it now has 12 rows. diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/claims_staged.jsonl.gz index 204b5272f4acdf056f87588e4af34d8dc678654b..7d900fbc785fcfc50748d8db0a1bf5fe53fb4275 100644 GIT binary patch delta 9226 zcmZwLRal(ck_KQTxFtA2gS%^R4^HDwkl^nAOb%A#`8DE-m3luA6)2@@kvr(^kRM=4uR zaZvPo<9|zERX86bhXzQg z#=MNnF*`q53h_ql!DM83dzvQ{Ee7l*fACHI)_{G|%!Gi-%Y3JA(Y|Ut{i7eMY)po~ zxqIjkz?ZC52m`$dBZv7jw!}@b{QcSjjD`;@I$|>52F8HU0n$&#*!V?%v5~wHN-H@( z?lHen;s_$Jmq%V!!3(@FSgJc^_$ymFn(QvtlQhtZeHu}r=4@i$HGPCN&iIOvoqg-L zBLOx}6&Y@Ig5k`;Lx&xT-ipKVV(Y0pcO&}p5`_ZG(|Mkng+4fxBbMPZ@0q98$M#Cx z=bM848wOL663ELV)`c?Dw;)W<_QxClct38m_dQ<%)9K|`5rJz1Pj3RWwRRf@>NpTF zG=Y%}$=E|`&nVl;`oZkf%kL)ZnFryq7+ApWUy?X6-ff45kCE8D#o*mcNyREdbK(f3 zI2=GKG{*Tj!46F1usIgOP8XEWmur{DZZaapEW!Q~z`KZD7bVF7-Fe5IUE4F%ou~(m zB9@78X(YkkLu@+LR8SuGEl*LGuuVn$lDWU^6R6Ex&9cT)ttenlZB0myKSB`PA23=X zZ#d+FQd|#;OuRJb23P#SC7LV=A}Zx9$jzDYkJ8x1As!K(K<{rbWjCSLWkh&!R6GIq z>}(VOezvgo!#BCZn-sSej}RMfV7y{7UKDtZ3)_%Q$W-)FC9$n4Vcx@{{g6}-gW;xy zU(8H&wiJ)Z@{sQEU9eR2&)#vt)UT!y;z9d7Ge1{`gagdQ5t869TK|4weNkrx-h*ql9jF#!^ z@5E#msbWEAe8U6I4KGlRN!t5POd-tOH*YexSq!oERIK2#M%LAy=!VOP$<0e2gR2g} z-+XO8@gV%^*w^G#x#&YVN|uy*MieW`OOSITTx5O2ve<{fgmqd80!9XEu*sXO0!vcO zl%n0QX!L=b91Z~zG)sv?9&)COci?i)xAWhSv566Mp`IV2%;4>GV`fkuKOvCpVRvJcj&={NQPOgc0G z%cUpa(UeK<;~T3=_^+MH<0|%F zu!{N=io)(!208cxK2&>OWnfgMA0-sreE6+UHmz>uG7emC7pd>|!rBX8n<)>Vh@b`i zj~M81G4d&d#qHS}pwpYZAi^$S?Oy&Z6`ZqN=o3G%EUVtJJatf*S@kN}_@#D>6;B(} zwy91BB6K)R!w`^ca^GwEMVCL+ToKn<3Eruf=5f(~^tDuGP^mo^2R?ro~ zSW#``!D`%8$^x(^Y$_({Pe#?Gx~jnvp^EZ`_m^$4K^=8^DM`82_^a^0AfpTdTq#sU z)0P#39guTHqQ~ItUj-Zfw^w31o{pJi2XADSkQ`XcD*I7;`%%+#y5{E+vG$qi4tSwU zvlRIhw-I6jtJYZiQ)!@5NTAS03l!gkr(Z1N^o>OCeR$)Fx z8GZRFO2;Log3(l1e!Td^=mhl!9Pf(y54~3G9u6CZTUb5^;|Tx~#r>7kv~x0>LC#MsAVO+qv` zx6$KurYMwPOoZiU-#=*1AZL;2Bs8hE_pCNWT3u_62rBgwtJVlq?2qn=Wy%|NP3vez zj_mq?wrSvDRZFeDX7Zb*=YOk*{euy+B-D-Dc1+(@=m4PUK7SNpwcJ&XEMfGTb@4{J zj?4h2rjGWv2;#y5OJvts5HDg~K>k+i*s&ND<{(g-bPsah`gpcXvfx4w42|ln-Lr>A zX@AS-RETyNXGVF&?2TKa8T6MQZljgeqAGl8yH`jVUZaGv)O0m>eB46ko*WSHW z)mheTN!d1*wwURDs!N?wOZof>rFPlB3G(GEl04WTlJ}hK(?p>;FqmynvSBP z1eIsEf~b1-`U30|(w9Q!%IbGCg8XmL(t!+{gGCRtyqeS{v{$&M(I|SNL?57OP>^O3 z=|OdFY751zBLJL=&2$(TDa^^iM06i#=ub4=XSx0TuV$I+`s(MV?UP0(hSGFyEn@yV z`N`7*MsKzsF9^DOhnijH1sYcEj9ttp4aQzl0!v;&6&LhIU5oJ_YMa7#4R*cF{n`V@ zyik&p(@-kYGkX*a{l^2U1yg|Xw7;)IHUe)Kcct%1!-tdo6m62vjRX3CeK%1nO?=%j z6?`qV&IE2qx}9jYrCG|lpVo&$011_+f_VbGlNP-4ojJ1O!ht|P15r?Me?cEnuvLmA zr_E5c`>gyN&Yr-kdq-dmAtbp0&5A$C3j8Gtu1~s{IGroge$Zpec91OaaP<_h>c5uGod=qdm2gP|L ze>OAH-RWQOLrjry?i!c^_Xb8MLJ@4f8z?HQam&(9tG|-n4FZ}|-aA+BDA>ZW^kE02 z@t^(7Xir}F!iuDsDoQ>y<1p$r%MvN2jcHfB1^s|Q%_?B4lGOcJBNn6wK8~3! z`1Bp)RKJn*oFj;t!i8$aYV&*SlXLzPUSG#!3ujC zL3`p>XOU?zcjBoITBn*fPw;@X{=AAMP2R`8&c34AG%P$hy7BT9u zdFp7bsy@wjM<(Xx3ZtKPCc@iZNoVSDXq4((y*$i!3o2A1gEq)m@>v9EA$=9QbI~^V zLxsAyTLeqDX&YfKy{wA5!mb_>k~mwG?18xcfE9Z8kNJyuW?~Fa-(w7@I2(IT-_cVX z4tmIZBM@;|Ve8p_Rrn1l!d%%S%p{Cv)5{PYq|^qq7hpWPB{Df`k2OSxqE?|Sc8&bK zZMZqK3!e)c(PS?3ZwK1*8ijQnpdrZ^^i80Bx<0X|CvwW(^?rO$4)@RD~ z3uP1$t@RJJl*yjyCyT3jyHogV3&(aiDCQ}h03EU!1ZmLG`>ONCMsZra<@y%R(&xH; z4R7OQWUPlE0l7vQX1kQc?+gkbCpCKM$x!0emXCt5g22D({osBtQ%oCFiW^v0{e}q1 z^SKN4rk7`jrtpb$4PNA@H-)=s4#t>{aic&Pqv574ioC5F8EDa!ir7}i<@YHjW*>s* z0H?Un_KE;3k7_J7QE5qCM)Zl~hVVa1$}}@f>qGS%&jP;&--tJ(6su@tU23s(%Xsmq z#vBFhdbx7xnjidrRMFanD160N8#M|4>IP~eB>b)GUE;*yN<5r*?vmxiRHQlNvZB}L zC1i=W^}xjvf~~4b#=;Un$R4h(C(4zHf_)qcaRjQZzY#Ob-8M^Xx0&3(h`wh1y0NFnp zyu80*RLUQN_?Y3^aZA5_QXN#pqoC2&cPlA-&xkFgC9G=`X4L!6NIdVXdLK8vgkQSK zAkzRv%lv%<6E|X);b)Ri%SWbgM5`S3`$^$l`DwmGh_eWf9kV3-q&p0zDo1N0-I8>L zd%w3!OBQn_E6qiUAy~oo3)e$$4A9#^!do~La{cgw%V(L{;bM62P}2O1_a`R`t-Z$t zg^ffyk{Ne5Bdur#SRTO?OC?ssyl4D6-5-(r&8ykpeo=qN--xsCnAA2aAo>L3(3&PD zcT;uh6UGA@+IQnCIrpfCFj7|9L)C*#%eYyo4Vj{?z%pjD*I>2FHO(yF(**EO(w`)H zNl_& z-}f^u)Q*2x6Eh@|L^Ojhvk^gP7fOCqAHr^zY}5XM;cipP?-Ai&amD1dBlppwNjnai zNPD@$!puXn5Z}Z4QH5ze&l5Ok*pw)oJbxTBGqx^kUF}M3snE@`zWk=w2v>W7_Dz9l z0^cTX^PI~-tBP}*(}s-FsCQ_X3~zTbG@0>4Z!=sGZBSGg{#JYNv| z_~4($Bu=-NOfAWIOWqN5bm+rfDdPYd%a4jT0k-i?aB*IRdU-M)t4p;!TV=N%O?j5p^J@x7&tBRKN}m>s|H)l9Rov%Bm>KL; zL)WJPrnNwk0g#!nWPZtg7~m;t)99>ZCF@&G{{@Z-1dmOHS~6L}Z~Rqt+rZUdvO&r) z+kI8=p`x&m8DD1HK1v#>e(x|bnI|8GAF%ojHvcP(%cw9B!uw*~F8Qsuv>R({t_$A8 z=Tr4K4B?Y`4+$J9rZ{QuMEQu@jJ?m{Ob$Da^Lr`KexSul=jKXT?BWz2@v5W8K%jLhaM-~P|2xdgrJq%PRGaGpkI(QYotHpwO(DB8P2M`jJo}d z^o8nU=WL@bTTejt*EtAh3Q@kgv-N9Et0!2$;|^BegoBaXZ`*SX-QrK7pdvtkrH906 z>GcX6*%IUyhp1*u{AFEi5pgk@=*t@yPr~nYF~CO=zS7u|A~WNe_gxUt&@FocBfNtZ zg{E`~Mu{tT0^+~*NxTVp(_$ZUDucM^5f{xqehG0$T)YWDf-CB;Tq6Ea1gidC}mADI?_f#8x z6>WXdXFSo*vDyI2Dq-`45M*fo={EkisrJ-ictLqnf0ED$xH%-OtRz)ThYdtu=GtJk zRv>#t7GMZ*=fmL+D(Suw2F*+Gt)~H2Af_o#dDs*r$L3?tx+=El2jN&>s8;vKsl2=N zZI;$5mHrZ`9ksdpOO2)9Q|lfgcsN1JT{lJ}^9Tf5u`DEzZvtHDnz!#@ghHL?lm7Xj zdJ$AfPW4Q!_mT^~_Y-UTeO8iXxcK5z+LU-q5E{dWJ=W@!r<&gWnWfeLA$l3`X#IGQ zec|@5g98KW<5GiNvB5n17M#X1W3!=HDw>xL&CY!Ld4Jt2|H2KNN8M zGBkg4p12F?v*!;NAPl3b=4*FMPN_?yCWHz=PR<7!S31{9d*_6GoSzUyg*E%Du2H@w%(>F8suA<_ z+W4&8Rh_s;84`O4u&T&!bLdYECny3q1&+@9RwU(f3i_o1r z>$Z3e_IV@duAG)!+nwo5K)J7oQO(+=gIoi3KF*!#e^gChpb<-=L6hyr8k6!XG)Kqs zOzfH~9;e>j1c|d!!nNj5XQuImI#E)#c(yi(n*BO4rrk1E^t3WU0&W6HIQKnfr(6xiM{RJIfd(QGPUL zPt38#UCW11-uYVojhC| zZ3PoFEV9F}$Ob1kcB{g4^{6g}5uzy^;nb+3lYCG*84JU`ptGdGa=(oBM-?@j*-K3y zKGQY=aJ`lLm>ls%Z4Ky%hIwF8gIQ@Bpsto%F zEo&ilnN4Hh1P*VuR}i~_&;wkD12?!YiqEhWkP3b$#*!VhrS89$G#wi^6nM77c#TUq z8VpT`@>(19^OxK=*a^9NC|z555IyAsTT^@7G=dG$bEaV!X8jqY=R#W zv!2kJn9}ESGXAqCN7OA!s6nV7C@H=ws>R-~FCL_ciuKe_7QXS|Gt_5>A6Z@FF_bTq&Ze^Z(crjsvK2sIkJ8?I7lwXna6=+gj%XGN)O(tBM0`s`BHUbE=>DKdP+0QKVTsZ*Rh zk|vy1Uy-ZD!fIwdt1z8NFEGU7STkG;Na~C)B-K$av#?@_WFgc;zUTo{vGM8$+jiT- z{3x2hW^O3p#9GdjWEk{+OP7+QPP>UhW&=NvL_ZpDTca%KxoP<1%*qiDZvMo58*cmD z)R}05`_C=+ckcU41yZ7&2K>-8yHxZ|luTf;vq)!h%i?!G-SFNp9-{Ovlen6x8h>mb~pe$B*6()VIFRIP7o;i3^1KqZBXZ?wyJ-mZJ+HkZVd zXAJ23(TfT26)U{&66 z_K#I8^S>7GA8e`dKiH2<&BK*{F=&NzL{iUO)JNw|TXGi`4%BPr-wC~O)|jNX(=NLI zt5N@-_CFYPLVxCS0{fml{mpdMo{i1&zag#iIC&hjQ68a@jb%Bt)^~#b5F(v7U`zKo zVWff>$DHAxwapg&XKnK5l80x4)yz)`Z?ikoeqRW$@o@$CyQ?m4!W6D4*i0uF)d%2J%1~GecZQ0~}&f71S-3TLs|!(JO-0ivK1)DY-vG zQwoErkj%g^GDg1rrU62h6C4esxvouzi#2@vJw*qbYMK@lTRIx3w_BX1iGT)SJ)N>o z`KNb-R*&jfIZjLmOUyFtK^~*COq}68#tH9TR;0t#1GE+zvAdc208d1n>4^k2F#NFg zZAC2MH5Qn#1PVc_D>B#N(W#CDI^+1L>jCNgegh3cqM1khaFwWc^iMe4dnV21@l}}6 zc`{Z9Yw(Dk(*;)8O?}gOgC)989?u==*zw2B0;Ce#k3FJJBoNM52Y&XKA%|_$ngw>K zCz7H09_;JaoqNe@+soNDM-Rgn7CV|2WU1IM|C~+wfRHMP3*>&?;_=+l=B3wtQ*w|C zqOtt+wI;V~Zf)8nD{bN<%qQ}Y3+aI$-1b3igDMt?K2^(AGDy&jqy2eW)k?pZn}$m z8f4U&1DNW#!Ki+&10b;D@Z!4SSix+;8PWYdxYF*iohJ-jjUKBVB|9IOhPV+L-pK78 zj)U-Uh^#Y?U^cFzrt0yv)1OE;!T!XKwWH#nZ_?J0P@l%4Y9cteQ?hG(udD*R56Wts zgLMj9CDe5{i%h$*XzMAd+bE7hp~&2pU)a34fi4_`Un^1sdxJ!D4%j8|?+K_Y_r(jT z?bWb#PkPW{aIwazSoNY1CSq!a=RzZUU@7pq`tR+L&8vxNSOBUDR@i4p&fb25fbhudYNF@$@5hc8KfiIa0m5{4^}GJr-kw4;r<9O|pY);LNWFcq zc;O}?(*6++bd}k9E0vW}@Sw!zEtr;Zu*aMe*ATK* z16t#o5NGPGn4C$KB~GM%y~^f85R8bPUC;Wqku6wz#G_31e${f`w5Oc9uV&x4Gj zGkI$&=+HSKnQWryeRbSQdVO@8dYM~O1CI5l9G)dsHcqQ7v{Q5OX>NJ&120z*bzs{p zA^te8&U1Wls%rIorviuU?7zt_@6QVbP$PmikBTYo;4N40V?aOmjpnvD;hsvx(t59p zE9EGDSa#p^@QT#iKdkRL+h2EhB(+VAwO%_<`pYg2=&0vE{t@t+K@2Huj8NaU1G2D4 za%EQ_-HlbDTZ}+4bJn9K%|tF{)TP>s(Nag5v(m34$zQER*jeM`nd4`cu*bHmq9m#H z`KLz`hZM0a<0Rk8Ckzswx1AvO-B_ZH&**L%h)q1jwlY6s{%jEWf3vpL0~qxjh4AV# zxmq4~A8KjeGT#ekxgQ57^qlY#o~F32{d2l5)89*P8t9u(mW$ZPPg)$& z;709yR3dfJHuBlt7*&w{etc-B z8SIdX7)dy22dOH&2sT!K$cA&QS%{(T7C}MS#$0?hRI#D_c(O!q+X&Q zEw&l2ww8h(5Bhx_tNVQ~@8B=9*8^-l4Y|54JZ?)e{2MRZF?&Qml}=S!;d@;^xb5^h zEqC|UlH6bDs*=e6RQ#WL|5*5}#^he)grBdBc5j0k@$#|IXO2E!)yWwq|Bg!T!t5c( lXm-T&x_oro@n^VFPJA2<@_fGoye|Dks`(a3O4AMx^It&o6~+Jn delta 9211 zcmZwKV_2pE6EE=G=FPUT*|u%lwb{C>&9?1&vYu?)*2dj7yA6BZ_d5D;&iA=~GynNC z^BDno83EZiOaj#0K&!y9#lC!uY8RIiFiaeu`#&DdU%1OR_>7Mbs!qE&e(ahB(!3RINeN`P3F2!jv z9>)sj#Q!4JD^3pM^>{|wyVmI3SWScvi{ta+1~N@+V+7>U?9~XIXa_;9^5!o{GD*_- z4KkO;fNgll5wdgw`qtdUg+S1J%a0Htq6f!*LS%o52IuXBRcJ*t|Dbh}zdbg-5f&%` zK%;;?5b>PimvtFjcXQwV*Y|BXSL;`80&(*FyvSeS%Z~(kt&l{J+M(1)&j#Jxtalfw z^>seT3RoakK}WO0?uahz8s*nb99%0NminO3U*Xk;5f`Dj7>z%o${2X#C`grXg;1R7 z5oi`wvlF?UjOv|a*NuMIOoxIc#QFQJ#o&U)+QHPTk_8Rv%*CCjKUa=JmCWOH2rc;e z$d2Ia%)So??59lZO+OWrWF$4}9iqGjJ?DttUqFhc9nMr4eP3(Z^+{u{;(+`u4|RKJ zsN96lR5(~Xrcl<>(rjqd3%ZYWkid)MY=6K8tDhl1!olp;hL;iVJ{-4~nD)ll$7uUE zd(smoCX~I~HX|2)SfpGq>q*hIU^$G*3s1l^9rXhwQI@2j5Kr@lo~kWUR(=R+e=T`Z zFE9d#gcBV+jZ3kdnmYKAZ#o?HL2m4aE=Qyi(tz1uk)7CB2D$3=4*XnyC0q%D9cDsH zse>!Sm->gCH&GJ#HvZx>9%n|E7 z7i=V6;y770r#oLZ^IlNCW!8B9w(RH6>)+gZBqJGVt3LOOKp&zJxvapW6DTW zfG_tAqk4#zI|fLh2h2uIhZdSN5uf(6mau&XJH=ktpthn;1fsjeYG$v%tG1s>fLb~! z@cBZF)R{=ek$qFdJFYpYHbSaML1h9-@T-cG{Vx^2F3|(mf>r;N{Tjvq03NJgsrZA`T`Gg$h z$XH_whU#g8y7QDXCLSLuM|`Pbl>E4&_P_%r?Malh_u);aym7~E3GZ+#nH#YW(a_ty z;IBNwU?)aW7$pm%MqgaBeYEuM{l|_2*SCr0V#;Zh+r{Uhm3&D`WlT;Wx`Gwc2;&L* z$Q<^oiP`rM3JCHxr7&hqby=+8zlj-s6ndH28)LZiAt!_8ZoD{CcYT#qiJaXqhPjdf zBLP|V1*O}mpJ|+D%E?k!Q1(A_WhJ0FLK1RpJhwi-bUlOONOXn%)@%5rSuf?@2d$m&OJ~2tCmm6~RGmqR1qyvav&j zszCa2mlNvw?XNpdXsakZ+I!ui*T1R)uap~7H57dcIDm8oXH ztI_FpF@Nl$(Bw@|ti!<|*FYH`)L4`@8-vbj7tQiJFYDwb35As&k&A;U$hr>C`1Rnj zfYG&hTFNpFe^M^@YZS>@;?+iKY7?cT1k6c*2ZNVR$0qH}(^k1{N9)ZldK7 z#Ae2VNxu^t@QB6VI0=@8Cc{>+nzc((B$1)x=jD-B?YQ$TKEu%6bEYd%PJ*P?fYd>K zSL$_M^bp3zJygta_Xk5ut`Wp9*DP*;*a(37zA&9?3MoJo8x%uN%wMQgL}jEEF6Yax zVn)9AoeAzz(QfXwIrosM4z*3 zg8gGKP@i+tB5xl0mip!My>y>Fe=mhDgY_FjvvP9J3lDRyAa0lV@L;)1xYn+~L?r(A zxPHg1oYo+1FG!xV1KMCQ$=UhO64652LhH!$!owO5W*Aw7m__k?-1iuxbVlQ{k}3Gk zU=Bcc854MONTSv4as!Erl&lPiHw34FL0$d$BuyD&IS@mBX?f%qpd55gzD6>uVkh4w ztJQ_Wx)|>qKb|68ch$-G)nNAYE|2V{ny<0Lk0nJ4a8N}{3j^$~W6UDcr%>;Nc$L0h z^p1;a9uk+>&39Y@hI2#tU*)uJ@=oqh4*_(xe;Jwclf5iNjQ?a4TMdwR2v16O7FwzK z@m1qXN>F2c&;#0%dE#ljOAH|8i}E(_F0152>^vF34T@H*Dr%2fv3WZRx1>;%stp^f z??{*CU*RlStOT>BW^f|89X%apF2DJ)Dw^6UP6?y5a=^13Vars3H(d;>w92%-t9dLz zt>AZ9an+Nfw%Rj`0$zF2iW>B1wY_G97cHZC-TLS?a|OHcW^qHt+)3zJ)q;USB7uVHT%GMzo9!lUs9#XSrZbX0z5>yvyCG) zug(-wiu+aULvG6w<}$M7W+IFkbW__DKcn?EZZvkD6IceZvBLwVDxsAE!hK=WWv@5T ztWb=o={B2x+pwUFX%c0&8{Amd_w7(5M2 z)X?SP-CB!Qa*-cHlvCMbNRpm1w@c}8aklB-kk1h3DUZ4z1;!Mz@>&&TduM=Sqw{l# zCxhK$!M67^Qtj#A-jyujzgVdXjxE%49sl{=e^6GwiGi9Y7y_CbQ>MbpHv25!HU$Yw zS58}EzI+I%R*6~3X~TfAYCo;SMm)W{H-g&B!?47iNY!VpbqwRgSl&VTR`2+ zSY;wf=F_7oagxm$5%$X?Fjila2s|&_@G?^RdPjrFt$d2lvJ}Y8sZ8R?R9tp-XwEjC zE1;u&?WHsv95P~d>gQC@q2S3UICN#7nufhC>XmKU&{%+&1Dmb@3QGj?s%U;}QGvKQ zvEW(Y=zo1l%P&Skgl6%+Lbcatu=QA|9%PzKH-$32rFh8 z#aR0S)-cYJMYaFLTz`t0kKL@`@cBjx!@^mz-D|HrFgUfa%=ct2Yc|}(+&|6nIqmI; zl9oz^^r9^ff&;n*M@j)`|Bn3{<7=!NpHm(dr2uQ9k7;|Ge$z6d*FV;iA1Cd`S5@== zis$6TQQ__+M%CqJy8mFEXglHJKl|5fQqMnSr`HA+aYbFy#sd!J^}Fi6|D|mrf+ZwA zCjG0neYz^G`3)!|-*A*x0ruZlXb`Hd>pXu@+WLZm&bJ0nMQ%snV03R^{(E}p2&DR? z(5SEjTM0<)!s0Z#l(Uk2!oP&O29XW-NZv8)s(qIzF<7X82w6d%-hO4 z{4T7|c7ut(ESjNtLpl-#{sQq?(}cX>3!+ZL=Bt0kh^TSlNJY&KT+lrZ^La|DJ*8I|TYuJ*#p-rc)4;Dxy0G+e;xpb7SPJ?u8ogHa$GWaS1G8+Y z$1ZE6Kd)n{hJ|j1U$gJ7>dcB0g9vWXwIzI1mwI&Z9;8qJW6c&k)Y+1ozN`Z9F>~4< zPqV4>>=pr{oEw}F$g*E=5*8b65nh1hrftWzgtIA{U&v43L1^4N&lE2#L5NevsCzpc z?VA8Em3uQIG9%OPu|MWf_RSyNu-`i>zZ?2?^G6s_X{stEt$q6}j7;aT;Rs-6syAo5 z?EdI}jJq84Q*hfZd^lFcK0?XG*mJpg2_mvrUmVro%D0h9SacHJMdUJPm)&LuyWLI5 zsF4`8b(i-5dPqWWiI5>UA;WjJ%ycpQk*b%G@x*M^MM34>^t0a7l~Zpyl0J=SGjcIp z=4PdF?@D@0sukx_x3WHICBnD?Uo?q;6U>E5II~lEltX#k)%{bouIka>y@I<`Wj(ix z*k1p7(zwz<7}i(<6>_FJdRV61n6nbGN^EAf7=eR;v^E92TzfNaK!iJ0kJO`oqd}n0 zD?P!Kx|wW27}tZd8-tzkB1usE}}O}d_mi0fAW6|X1ATS zXmgnn8bY-sv7>Upl*Xov4lK-1QHIB6Z<9SaW-f;rpw|&(fY$CtkN7}0?S*#uX~Cg) zwsPn|N}(pWgY#%_x}zYac!KEycZ-$*pWbCR)y=!0*TQ^1JrDG$%%}5F8CxezM$2Pu z;%jbofj(0clhXR8-I`-)rUNxtSET}Z)}(&mrw?JN*n2n5c2-P>%`+sr07YJvW!H2W zsVJHd#I6B0&|NVfRRfFq>gPXc#VSoJEItAwcNSWH@K<-r!grO;O^fh8{Jm&BX6PN| zbl;FdFsy4J=sUMcV7bColMkBCC)Ka-+^?+i_-D?6x0>D5oBj^!IEegqp8vn^aM z{Uj6YHeurwJqoMVxkAt}?_=9fxUV*|z$Z|xhjs~$sg4)QW)`o5o=YuVwUqOucdZ02 z)D86|a+8L0_fQIgS3F6TD#X%#8y7D8E#?sc7V9Q@yH&H;#6aN|=Bt1{xR3}p8H83@VomLJlbq-%k>(<70U-iyjSAg> z{^C;)W8@_cqLx#?2+5l~N8RXJU76JElA=G$atyYurwN^bzFfPi`-M~C&t~=X*9^7$ zueayKG*4AubZGK4<<`%+E%m>CPGUHwfW$4#IlwpCBsumH*Y)|dmkZs?j-XBLmP=#D zZZkH!q1|W~;!?7m7#Cc~b21>HYjuTKpqyVRU*`By{K6#v<4tq3|HiK{L&Lb)ey?&$emCzCGzk)8zv;yv5W!LZSAZk%)<{m> zF)+G|5$v7wQd-hylk#lfpN8!+g0yZQ%OjT&^} z;%-9b8XKkI4HZ&$bWcuv8{OTWKch-bLiEnOw5bYPYQ~jc*2fXb@ugtXQ)NJr^cl;Q zSu$@Ex6w(XU47~!j|M0(YApBIjD_Sem`o^lbOdF(RY%U>p8MN38u|mwNr8Z8Oy7)U zIxJ3I0ygY_=rT%30)HBLLR-*HkXX`sWPBmG8ToYkoHs3+wVEBYC&L-YVFIpfQ$QCU z3}MuxI>b98(B&L_G-NxERX*(I&JD{z{MLar+~dPb+v&GDC9WIJXKo}Mey zRIDa{-Z<7e8(`=P4gA2Vw1lsa8cvDyzH5RrG=s}@7)dXZ232$a?0&bdTjf;en$K8s z3mgEM3t^8g^+|6MnNnc>8RP!j&cXTjo0=|4xjZ7`5e_D$&2RbH^J%W}s)(NhJdrzB z+1&bSUf{FvQZgcaX_haw1@nR3*T_Ws?TP|7G6iQ@y#Q-#BqW|eX)(B4?D0pIb#RBj zYsTDOX7H~0X`_Rs`Kzq9fwNk1z?QBLvp+Q_NJ-CR^nUB)e(TWpdw{WlAV?|EKK6`B z*F=72+AU}{Vpthi@Mapnx<(hQ%I&|rDDl^{NyPY=w)=S!oO>FnBUy%0=FXY)cT02$ zE`>X(0GJUyFRTt(S%+J)Advmwhn^gn0h523!7GRi(ZqL=fUahs_)eBz8p65o-}=^& z2TQo614Wdw5bh+BWl@4!jwCqL_iwn@yKyV28~!b24mbCr2_Moni@YiH|_GqWBxtcXB(V-_=ZIm)&p$ zQQ?*ZTatc?m985c^xApEOq|qpFhhRP;%xaph^OI%LZ5H`Y}I#X7a>pq zyW!W-p*UTO?UJ$jdVLy*MS?!LQvL2)fqJdib~_f}tySHpvi^6cYOO#h8fvIazj~lD zXU$}-{k}TuOW9ibGeZP(N29&t&kg43?9SR8T8IPkxE~psG$=d!zB#o$pny>YjM_8k z=2a8g)r*w9b;k5~$$m1+o@Czo@cS^Jt7p}&`m)8OtR1vUx9?F&IS_B>vyc^GMR*b# zaf?$_qMdySRbx0X%m5kL_Z>a~8*b<3?~i0VS7Eo%s{#BciG9+D4Ch4T{ipB4Y%_y& zM0&l+RvvXf8wH@2NG_BI9tjrg+tmr(>+^&>GvtNKg+?Qz`jBptyeLR0gy~3uX3-Ap zNrvP@wGCxww1m33S?0^YtHs!1o|oU%P}=b-5I+J6I0iGes(x!NgDP~C8WbM7zI@Fw zMH6p4VeGSR-`%9eHFpgvfSlZVa#2X3M#wjKHdJcKv7XPchObm<2`mc@c``>+X_jv; z$8~rT*uB3spAGU>2&I0_E5Zc<|C`;44g4Bf=hrl6?0RT(Ho|WJA5-X?=TnCykR*ql zW5G3UpRh|C{l4jX0pV*}WwbHOQJdoUd+15v$NjD8+mCOX*yZk_pP*Ci*}5D9F0lQ- zN3o5z^ZxeBgg|9f$L)6Tv2MzOTvy|lt{iKwS^J#b=d}5jQazx&WGh^VbCO{)gPCb? z;FY(Db?5uDS$LT8Lh>py2w$6{)>`*+Lio!#GjDT7ps_&{>%n&(i|{z*#mu(&_V4x; ztBRm(uVlN~R2zN$zgww7ywmcX+V8s47Z39jiET5VYws}_{6ZqyEg);1+4>x#He7$7 zAfuf)uDS0#Ya2k2z#8x3hwirHc80ayY;%qkuHG7dV8&>9L*T}+du@`#Y9!ZF4gc*$ zfUZSH8=3==)GC5#m?goeJ=e$0)iAlxHbIP108^sj&Gx*mGT2+>8K);WJp$5}4V#L< za^4XqH}$nLCzAB9<%l`SGijM7y(R`u+I`F2ZpxWtnczbnpw@pB$yDPe;m9THFp^+; zt4o<`$MuIp4PaB!oOdzBBi9nlSANltLL|XqoJ$@sUHeJGmfpziia^b#6$`6a(6~8* z6!!D52^SdfELD3@>iJzmjX?!rzlr=%a6X-3>EuBKX9%woT{*ytW7%&Bmb-@v8)!;S)gY@XKs-r{G%m!)1Y#FJe4$ZvpAY z0O8|Dy_IkKKAOMTJSk*dWNAp7K={87$U1_imYl!bA63^3s6Ly5)JIFDKmk{mjiC>ToMXmj)M#0E@xUqzD=7LhTGqv8*!8}OhaJPeT(LZ zGu5ltE>>UR#LijgoTeyU6=R@?rSvK_^V$zgXc3NDmj0um=N+oz1IBvC) zAU$7iG5y1`!s*1K!#M^QxK5k=D5f2f>CY&$hhA|J4aWev|{YObZk&ggOb_%Qlg$(n!G@Y zbm<%!1!;%!tTtf6e8qs%Ps@!TjKlX|&%Q*^tYk&_n7_vGiM%es-CCF`&Ic*RW3E=P zu|irY{!Au*IJH9B@M&H=?U;@wd(7 zqsHX`G&kLSVS&zg(j?sKxGElmQ%v$`Uz|}Mm|j>YEaa4ELM2G(5`Zas*8opdoaNs( zQ4vbT*m(9yoC>cN@+g-?EoY4R!f-iA6~*H78w)dVv(<7$6(5AF9u6Y95MZ_BsO>^1 zqdc{A$JRfbrGvwF=`7KvTl?WDcA|QOhox92Kt}ubk~iBXAWsp(w-nha&X3-CqYU+% zSy{)Y_dnd^z1xl0$>Kgg&sSQDCASz(Y5X>Kplc zw+FWT*>W~x1d64|>?Nl#9JdZbVsfHuVk_IGE89kwx6*09UB@x-AwgPPt_03mXDqiW zu0ecx#q(f6Y_G8M_YqM-3R%&L4ClhQLHws4s9Wi3rF%9>@3-G|nd{p$p!AAwu6LWQ zqlsB*W5^M_Lm$Gq+7DM^R!(Ummc>wMG{@0?;s%1nMmv2dVCU&=FzAm}awsm1#XS?5>nf2Sd0!PN_KxMxPohbbTm8py zUC3c+CVo!cKseh%7lyTgw{rC}{rIs6(}n0CMa z`q9G-7*StT{tV3Z^%k5x7vEf5rwj9i@9l%cjxhUy5EvOmL&5ziEX5NG6NaL?08Z-@ z*l9BE%Ognpd!~#$5{3;IV)J?FpIc5G1pL^>im{1|o5Js>VH`Iwn?!x1@nX1P^%L- zVl^&ci=XpHb?k19Y3H6Bxp!_&dhE1HGvjoOzFG>pv@Y-)WB9VgKtcDyaM)kwHL zK2hV7KL>RzC$#5ze$as%4>?|_=fxUW4mfZTtl8Vk*Q9>XekHV8>`7*jyYi3?$TWY> z8RLp~AP^xXXi$n?ko=sGG!{V;H5vZ=Srzc1DT2!mZ|5j>N++On8Z+}R;;nW+@FjZI zrl_>2KWCAv)InMVR@y{wD`%}RW{pr~)ZrBQGO!~MBnQyIBvRNZ-%t&tqICk!whTpq zp0gJ@oA8yR7f4319upt*^hSzOsYwEN8K|8|7`Q}xX(MuOHV5Sn1nZpjc81FxsGXz; z$qnw)k;*EOI|~1I4*uhMd?Sv(avWX%sX(c7#&jb^b|iK)LGJt^+y14q6hcM#xUU#K z^=Xx>#lf1i0H+0D;WS$!f37U^V8f0VzY5fd+)2;e;{o3;sr`RCqI9Cc@gEzf@X1;r zvCC}Qv9FVR&;H6;RLNQNocOS{4a`mQ?+3N)WQt#rrF;{=qWf<|XIs7M|0cAVzt{#h z3OZSv{L-tIxKfe(ZO)JbgzYE|w|dOB4jq) z&nOLZdv5d&owavj7j9Rax)b}Xm(|OCjK~?m?+e1?jC#93SeSuT^A!yD{Xz?sW?h`>+r_Iyp1S^!nqdMlNxs{;z)jvBIa< z;rHMpP`4$JDHuVscW3}E-({zI4`_GldUx*ib2ypaBT P_y-9CV-)}ohKBeb7LmfQ diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/NOTES.md index 1a74e2345..39a58d3b9 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/NOTES.md @@ -186,6 +186,7 @@ No unregistered enum literals are used. `fiscal_headroom` is registered in the l | proposed_metric | Rows | Scope | |---|---:|---| | `cash_receipts_change` | 1 | Self-assessment overlapping transition payments, £17bn in 2029-30; assumed extraordinary accruals adjustment excludes the temporary cash effect from PSNB. | +| `fiscal_balance_change` | 6 | Table 3.1's memo line, the total effect of Government decisions on the current budget deficit (README rule 10, 3 November). | | `tax_gap_share_of_compliance_yield` | 1 | Box 3.1: the share of this Budget's compliance measures' yield the OBR expects to reduce the tax gap ("a third"). | | `costing_uncertainty_rating` | 7 | Table 3.3 High final ratings; categorical placeholder zero. | | `psnb_share_of_gdp` | 7 | PSNB as percent of nominal GDP. | @@ -268,3 +269,15 @@ The OBR review (`harvest-staging/ab2026/reviews/verify-obr.md` on the staging br 10. **Box 3.1:** "a third of the yield" is staged from the review's quotation, at medium confidence, under `ab2026__package_compliance_measures_announced`. Row indices above are the 29 October order. 2,797 rows: 2,891 − 58 − 37 + 1. + +## README rule 10, 3 November 2026 + +The IFS review of d5e323dc (finding 3) found this family's totals of the Budget's decisions on `revenue_change`. `rule10()` in `stage.py` applies README rule 10 after `review_fixes()`, and moves 87 rows to `decisions_effect_on_borrowing`: + +- Table 3.1 (42 rows): the total, direct and indirect effects of Government decisions, the direct effects of tax and of spending measures, and the two memo lines that split the direct effect between decisions on and off HM Treasury's policy decisions table. +- The policy workbook's Table 1.1 (42 rows): the total, direct and indirect effects of Government decisions, the direct effects of scorecard and of non-scorecard decisions, and the totals of receipts and of expenditure measures. +- Chapter 3's text (3 rows): paragraph 3.3's spending policies, paragraph 3.4's tax policy changes and paragraph 3.5's fall in departmental underspends, a part of the indirect effects. + +Table 3.1's memo on the current budget deficit (6 rows) now proposes `fiscal_balance_change`, and its sign convention adds that, on this line, a positive value increases that deficit. The workbook's 288 lines by tax and spending head, Table 3.1's groups of measures, Table 3.3's costings and the prose costings of measures keep `revenue_change`. Paragraphs 3.20, 3.21 and 3.43 give four policies' effects on borrowing, and their seven rows already took `decisions_effect_on_borrowing`. No value or printed sign changed. Counting from 0 in the 3 November order, the moved rows are 2190–2195, 2352–2357, 2490–2519, 2521, 2527, 2533–2563 and 2630–2641, and the memo rows 2642–2647. The contract checker reports 2,797 rows, 0 problems, and a re-run of `stage.py` reproduces the gzip byte for byte. + +`scorecard_db/signs.py` now reads the workbook's legend on both metrics: on `revenue_change` it runs the way PolicyEngine's sign does, and on `decisions_effect_on_borrowing` the other way. It reads the six workbook total rows whose sign convention starts "mixed:", and paragraph 3.6's two absolute averages, as having no single sign. diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/claims_staged.jsonl.gz index 7ebbd4b44581634fb257d53bf980d86603b55231..ef0a63666186bc78a126c1e79244140f89a135df 100644 GIT binary patch delta 26598 zcmZ6vWmH^E6D^Dk?(Q&1a3>_VyF-9LaCav-9NgU@SdajL;2PXr0t9z=cl{>MyY9Ev zy+3Eyu06f#bnmX}{9X9eJ$U~*=yE=wDTs+2V|3FAq46@YDSKEsIU$UQpR0#?zO3hB zdjH#`wvr6doeb^lfVTDh!_`cyI*9o5k?gNLdrr=%tHbxP(ImTDWoHkyUt-rxoSwuq zD|WtrtSq|8Z5(yEn#+Awz9JnHe|~9HwwcYq=y-WQ(jg6R`G6deT7075eYNQY5ZirV zjUqXx#?Lk8Ju){lU9ibDN(&h9KM-NZ}H3iY+mM1+Dsv5q`lnG zfFMmV>D1slZdTx@%=KiwRt!1#uTr?2tTqQ%PXAK~jiX&ZHBxY=1Xgm?Z~&MKdS zJ~<*QSp|REH0SBF_i&7KD0UrR%-$YUQcXW~HT@qkKFe7q>i-ZELwWiR{r)3PEM8Qv zDXXSGx&BYl-I^lH+5eM%Trp?ygn2=PwiVl%8IY5Y6G{#e__Ld#nz&!MzIneinLMg} z%!bqX$_e^t3C)rR7lSs2^QDT3s17p+|5}5cuU_OS5PuSeSAOtIRS@2P=ng#u4fuzu z%0kdutp89+O2&E-gny`&^xzlzUJ^@69u|59E?4USVpxbUOjJBjZ%Bu0Nh!dxk3!?> zhb@YuD$Js5usoa@f#V3n-8+nAvnL^okU$Aq(@9CjGV;y0e#EDVeTs`0^9RL4VTXeS zsJg}}L;@XXu764dwqI^6pY$*82_hpxf+8>{LDDqye_>_LV}g}ZWF`p&vfP*BCnxsVWqnt+2p8lT9U2Uq8hk4W}x-K7MOO$J)94C4M#D!U16U!_dtaE9qI zR*xVD!6(jo3)S+6Thqfu>srVIy`Z(Mw?RbltTer|^jy!6h`c6^OtFKnkmr7kGXQY6c0v(? z9S~t*-R7X+eGJ^<&G4K=jK<@ecjC;Mt%&@g*`eSiP&IlwD+v4p90*+#BMtM|bmFU> zf<);kb)euO>61+A=&=0iQ1Iu8{Q^XG-J))fb|iH z2Mbk=M<%j3jSzrU))@Q~lVh-1WlHs0 zG|+-3xxE80r2P?Ej1zlCXIhA3`0o=~M<8<;B48VKKJml1F}}}Zj|kUWqcH426!1ql zheDG`A+M@M5LjPT>yvf8^sT`1nXh-U<5!cB5;zv^s&jj-U})3gxv<#fxn>QM+t` zvb)065(^CVZ?o)U#lnZ)2tc!gkl=lP1uC(t&_i&MD$EarvrllHj?p>QM?e=Uv3IEc z9Jk}K+WU2cXMYMr(%9Xwn^kCG(k=|E!!8WQI|muE7cjLvbO6OUtRFa>Cy}CD2sh_u>d@6%%>g7vF+zF*bGp!KjzfCW{k$~(b0gcz%QKr(4(%Y^+sin;0fY9ZK zsq*hs)F4%=S9v3qtbapZQA$yR>_At3dzE_zLRpGsWbqU75IZ9Gd*6l6e~963N!agIDvbYA6eofk4x<_`pCyOuCKy03 z5Jd%?c;JS#CsTcWVy8ZdSN-n}FJV-oMdq&&!KIsD{K=@YkHYF|?<8poP2&$53t^R+ zM}m(>B*cF7LQRR!>xRM>38WV%ph1p-paj^+L1@|W;OT#;H3U5lhm`@LBKb;&jmLf1 z=Q%`(7@E7d{y?Gfptm!cL=gxEFhS5^i22*g^b8->_8npPlE-w$SU`@vNvsE8ZiFtcaEz_|84e`Hsn zC}HIj{gotFVIZamP(>GD<*OKoAih1h z+zDWcZD|s5s#QF>-j%R}LBal*+XT`y!7F+qh-bl;PaXtfTdc&KYO%qV(*`2pJ%|?+ zf0mRf3O0OSb%dWo>0u<}4a2F1mE#hTsbR6{O4xJK+YmVrIXYoIpfKtPZ+e5WP5i?| zF7nt0CL<8AQO}_O@)CB}^)7Ud&Y(LKhNM_JVftte(<;)+A_cx}LM<41L!`*)0-f$y z&Z^+M5z53II(hnI&VPh(8826&{|HmjCOY{l7KTV`sWMO16!Mm+?@g9pGLURw@Gue!IlmBXh$li+BrNoLE=sa#@8-z90PqH%UU|HV?@F)r=xJ2 zVOg2Af6iPoe7i9z#gg$zNKFLU6#0+Ko~&%!BsAs#LT~LA(KorOQcy)FUQ=KbCGO|> zG1v;k6};n*qEG%zj&K#=2!N-E=~xZsvquO2r@2OE+*64JH!_E*nRh=r2?=3MCbecd z3EjK`IG)#_ry4nMPgN3bbsa#m*#R8v@crOLo;D6-7KHxu4wDcIll|r9?yp*j%A?Qw#K!GRPuXD zz2)_%wg|Tfl3DZ8&@;<~2C}sf%TIy-)!b4@0I92j|6k2}t#rw-aP6%qx3c!PVaK(a4j(1e)G#N!IG zu=ON?Mj^l{q$HyiO|J*1;E%)}5ZH}^gZvAmaQ-6Itv%(S(GLpIlSsW@Vf~^TO>YJI zkAv8afpfeJP3vJH*6~e!0OV^WPP?@=2;$10L1W(l(++-&-V?tP8$i|y=ryu}JF25U zAocR!xW6l59oMw|+NDYG-+CFtyuknZKM^rQ8N)SXn18*e2uwi^ACi){93exi!V#;OV6S{qbW^Z-C`g8CE7us(}%@<3e3nBdcA3^&c z!3COK0M^85|M^19ASmHF%aIaRAqtuW((Z!Ccvx7D)nLDP{E*CYds!4S_)RIk8;rsQ z^h%KK!|3IqWX2{P=Er;Fp`1DB-uv!}8C^jL_p_35kFb!at0=0ovEIBL;_vhFD7Vzd zzai~f5}~8A1m@vGQj>kqzdhL2==RRIjTlnF*xFjgFIn>nVw?ER=dC2}5hUfFKNoJo zzG12T#(+2)CQM40G^Lgzg4j7rFyNOWHkszoR8JC20@);B5GPu$SXDl1@cPSi7-L9Y zWv69}@3;sM07(t1M7?W`IS5gP15T&Ud7)Q)H-2diD^Lh` zFlzjkUcQ<&VwOLo+lEXbbXrBQO6O<6<#KfUN5rQ7q}lO*xFOeAHsI*X&u#T*??OSd z^_f98Mk){bJti;nq>5KyGA;&3;JHt3@hpx%eQtN=if}JWBZ)vgJ-7;#+KIi5Nmm`b zB4vnlb{DPCW+Z07NLjGs(K48<3!?Pa4Z0E0~f ztN4%!pc6?fYTB_Is9;?WKJ!)@cwNn&Bru|hCD5YfpoIUo?m)&Znk;%6V#n@}bcyY3 zN|wuQR8B2i$wYcyXpjqf9dyypz+;p~`Y$ebw^SaA^H>wx>r^&IQ~vA25Z~cI#M)Y$ z(FoJFNEf9u`<|F=d4fM^qS->tIM~P7+AWQ8>j2jipminsukFK*hO~Wi_=cSy;TN6o zh$jg&2RoabMBo#vU@##4n~1n-Ho~zk495MfNQXfP!gUH3C%ZlQ{v87~qO3Vo_rLQNPhDLaA_6@w!gF!Px`&`1X)4X2K@Jxt=S(&z#2gyi<82UJCITXEOtRz$ zR%w4{C(!-ZA<|<(+D9_7b?7zn!QUDlA{z}@89D}r@QLkE7^Ld~fP%Tf=;w}yU}P2O zl&-S<3y342soAbSoJPsE^upWJs`oSo$}HM9>LhQ_@cqwo$p3i`7z|ny$K=B#x-pK%DHN-~PA!)Yg<}Y#09HTm z0r7wKM{0NQ_n1A^90Y~%a}M4&tz5HHsMf%x|07)l642se!>6xjBSDPKb+Tz%Z0jV&J!7XrQ2kx^KP zT7vjLJC;BvQsC(dMYc)8U1F^I1q11?&;7fhA;=2lM zAyFuDl@IqD0eT-gVb8ojxc|HgXs-|r9dJt8MOTG|923NVlCC7o;}U4fEYrAt!H_&h zYItCUpL>c63Vf-{Ybp9GCUxpH=nT|iom>oq^K>pSWLEcG60l=Bb`AwRda+hHZyZB zNN7w%)Y$37k&cU+2Zieo(;f0t#qbU}uk&stWlJB9YLb9)u&HT9lh=F;jvWfV6DmgR zD`C|%h<*J@Jg8Q);}bN(wFXi(YV;BU{DTTM9T-N!AJG!~Cw4$K6&U7?YTpbm91t-G z!5ryqiWCV=B!tF-P(*@M5)hFOpsCJT19=X}Sz$u{iZa7)(Zg^h_acNtK`5C)KP3!{ z;ln#y5EGT)&ag_NQTZy)u+qO5z83+7g&SU*`A{mGUMs@8NTh^g(qm|f%?ohL$py0- zpX*T-7Nrp-E9feoVNb=Hq{aho;n>GE3z`3YFuVcF{vHMzm zwPf$V1MUH~<{>6O9)uR~=odc>Kz|Gjn?XWnD1!ZOjJ@2jw5^gyoL_&ELL#5ftBiDP zDKi#BhbA$EwFSrV;FBOXgKQDw(2%E*#%%L4{!PB2RJI@H6#+O~#4~Z^5Z{np3H8K(#O;rn` z3??elbv-7ZtOztT5*T_#oQI~$2H`)H8y$kKKtWK&f2chCf9MSgqpV2tKNM&ahVjIAc%&<|@XRT!z^+wh^!T0hCkzw=XjWeSm`$#l_ zkP8Ff0QoyHK^$hC<8k-~(g2o|Q>0Zp)%2aL0nWqd}T653H6_Ls2X{Uu|%s4dzRCg<))W_tgIH;roP@Cb z=VGLg9hs~l#AfSm)Su*vlPP&f??u7bS+r?ds?b+3ONEV!YS8?cfSXg#mL)5>ts~RG z1?1C8g`!BkPsI9j^mkg9RFp{2Q*afHjc&dgyab~b9RTbBoJ z3MI}5Nm^w!U46C|AdBmg#(ejP?#f16;*$3(zMi6k!|@vU(dAkXJ@vUN{N9E@R@jyGdTWmH-hwX*RZuFZ&K zJ#CKB7N7;K)Vg&XqZ{+HnJX$uwm@n^)@9su1AHsu{Y7uy#hfl_bko$LU4D#J>+wUNU>;*A{~t}~`%0H_pO}B2H#2VTS1)PL zC`l7$+>!a_KC!5cifZjqIW_~nw(Gyhl+i08s&4Eb0;<2aH529TFj*SL{F}vQgmsiE z6YuG|z+WSP9IY%%K)qi0FQzFV{|GAzMg&prmU4(5ry>O2L(XOD*QEh)cy}^g!gm+y zLF2l=IHvML|IxmW{iD4{2=t(pac@k$R{{1>Tz_ZFg+gdAN z@Ye>Auip7ZK1p|lb^ppW-6n1H<23oTLa{_e@*E(1jbwzE1u{XbR9;>?PRG^jcf;PM;a?A-6#aNWN5D=`#lczirWtPU2=NC%U!u=pb zmFjlPoa(lWTCHBij}9S)hY*&u8&?N`dpCVaI^GKsYh6o!)8dt;&6*{=OBnj=Fy zaU}4-!i|V1CSnbOX#QMTaWhXKr`M%iJBgC1eTsp%b?msQ>q8BX@z&1BrO!jn`!#{6 zO9J2P0XD$fXJ}+jsI`BbLMdqg=l^m$vGK4!>ajtdL>bk&Bb5*9e*era&_JJ!KkM(V z?C9v6^6p$s{C3hI^Y6@6V0FGj^m=cftPljQZ2U9$3> z`Kak72Q~G8iarJm*VVhr8r2;26CULhTn)SOy_XdRq7>|7UU1$J31%OsSA~O82DC%q zEu&@mB>AGaw`FDZ;?couj$X-k!HGrri5h(`n&X3`>217Sx9U4%py``P8}jD+5`>#P z@);ZEo5$a!7RA0Q6I?f&wf0QB*f-Z9zSXEXI5+q`E1*LSl^apPe_Z`fr3A*f2V#OP z5y4)orD^XL5=x3I!=ug4NlPt4TcBsksKSE1R4@)j+xpemG!@p(M8Ke6#owRjA5mKtHKKGTmvv|1wa^}|lXjue=yGGhe-gdvW-}w3c)Ra` zGCge*tI#JLlN#sB8t2+hb9k9hj<-kW&CP+^;bShRofTBT{m60}$<>-Dj{owYr$3Vs z@b@q)y?nvc_9~T0PxYE{Ly?cb*=ejLmweIobCykNmM;*{tt(4@}n^u7? zwI9dD#mB;d#8#@%7&T>Mnh7@!c!qQhAfKW^lV5so-fBWrv=T$$OePHk5%|dZ_=#3>x zC%Sh%%S?N-f##!{9~UmvZQ&AA3gj0K?^qQt)3;mHG zRX54=*7*4=bM2Ak<5%0*9ao+V7FOG|Qd?*}KpA@4qe`H=B$cSap6j?JEN)=jl&VhO zGpa0p7JYq-E}(YdWo!jT|!u6N+5KX<8r_yyr7$_Avov0 z%r@sxicH_)#aC5vs z@N#43b78inm)sWxT#5IMe>%%aAKH+#lG}cazE*g0vWT0&<=1s{?|fQooNie<*gUc4 zZQJfteC5iGAiy~}9xSxTw8&grJGWumeLhGhsNJ|IIriG$@c6KpUYS&X^nHaDfLS=O z+RrFW>h=7gPqLIbIy2CIOZ^rDALr6%$U{S|1*|*8>kBRi8=c$|vuQCA-zdWzaku+C zNO@AconL0#`APlOVmaNO2S(nm`2g{fCe39fxjlRrlh}(JdIRy@0+L!Dg$^COZx7#m zGU5s1vT!fsQtQj`KYx_bZb#AZz&KnOF>QFxiG1CulI|he9H?Bnbr9$VTyx*9);GBM z_6M2Y&I5%WNIraBO^aE`SXlTiWj;WmK6Gk7I{Ek2`@8uZduC-4^`judo$b8_`jsu1 zt!#;eEA{G+ZIxHdr9HZh$ja4lp9W{#)zKx+_;ZNukMkWA0U)&OvtOM@z3_O^QPO3< z_hrQ#7x{G-`1T;e&bOTpXmI#ze#&#_p=i9YB9C_F55J~98g}ID=sn-O6g^|um^bLx#?1nU>M^0YVOvAY9TrC(#Cqr*xf2c( zJ8y#eSXO-B7l?-oMxOp0+81w;@IGB)L{N12d}Wy2^rJJ66c{<%2fT#8WqkW2#EkUk z8(R;*3sc2ytwZj}YJR{k0+PGw)dRn^VjbmCHADMo{#u8Z^fkcgYvbTw zvd}9Xu~$aGi(sf~708H~X+EmdceUyFx z|IUokePKcOom0-gaxffq26n4<+?O%eFMF^RRY$;s(`G~+Ott%^$2!fFD` z&P8_a$S7AJ+Wi-ma(-2ytct_^*+dwD7l4xB`U_Dv^tv|^vt$3!0PD~fO>OzBF4kx| z_c1b!Ca6M4t)9y9!#p`nC*29Zp(V&k;y&lX+%C z;_g_ccUh64q^J+ovp4-vh-4?o{xRH?p(-uVrvsjg{Xt%3yrHkRwb`IYpksg1XQv0l zi$U1OlaH;Tn|VJ7GoBp(++<8jZewC?Zt-~*TyP#(;j{#5VJCJ-oNrGQX^brAGb_p{d2rs zxtOlN?F?g`iA&QRqx#ji&%Ae`djW7^8oGS_jJ=Gmi5m2p^~g?Li%DgGibW&S?5$qN zp=g?x|5%G|%s(xfGL?KAyiKa+V$JWT_I5j~Wh!HpaYuY>Wz{5FS_>tDuxkq~ zXX8x84ZUKK|CaoKF(@CC8>kieF&R$u6wevclYv*^!8 zWUe*KKly$kuFnRSfdKb&XM=Z5&m#;$7d3J-Ph#vhk_rV4_@*iA75b-Tpfn4?W;xS= zFYm1V!YW5SQwHZv;xz*KeM`09HTfJ~mr1OgJr?<80$f)oYG7^X%up z*dXCwsP7=bmGH+l#x#Ud;oM`aO!)btH}-9`U|Y7m{aF0R#3yXV-8a?`u72HOUYu-^>v64?QL=0%r?xsDW914-!EjxDai( zT=j1OO@T`;4KLdAmz{m8rc$)V>U>9P<%QQqzk~t>Dd`2P^;t%0;x0wICPmy3673DS z^;s6(bf2XU)VhB)hV1z@-|3j;Qa3P+HNY2ZP{XdZ698wH#R1>8BW54J+>E%_;h+D~ z>lfbqw&*iGvrVFn(YTFqk`wGuQ|u`JnU;wL`=lYj;a3ea_S($uw+i~T!S~vkHx^+} zhl_@p!I6X-j1pR{!P^roj#QE& zkR_E#&03Co7PE*}d5R(Bi~UnVzFgawfP|dlL3<|feP+9Yw@6LizED-8v~k{0rj%hv zQ#?7-hpJfJx|xjfXNBKPJp2g42_ou9=qd5{_uY+QDBt8il>^^#yWfyThAwwZWqGyL z#>rYEw~4J1>&_X!q{+~j&wVHJ`|BEu`;ytk7QcC zIb{)Op!pYLratf5@6`s9*Kin9XDb1?!5cNXfG#9@@B^A(&@e7x-3h!fB+v3+-b`{f z_6ffELn#StV)1=x)<&`SDXMU>@?UXMu(#4QuZUkZ3n@=me*AH%N?w?rDmRrpC*%+`Pp*s2i<_mSNHwX7df)E_@6)ih#!kdz+9p* zNWnOxrJNGj&`}3$uTuWHtf-zPKEtF87!{AC3}BYP28%jVCmhP~^vX!cfawy9WdN{n z0!Su)NxhH{8~qX&3wuz#5FdN+i~30uTAw}d4*3UCg9&Q zOcZRBB(%5nMdVb~bXO`|y*bZfMS#ikAC7u`=cvh&%PPN_@oTZd>un_K(5WP}lI!Ts z?Q}V|#w!ao)&1(4FBziY!8ut%W0+o?cge^v=jmc>jky+D+WTQA3Rn_i%CFf{h=ST~ z9hqwjUZreWnzpM8RrAbDALO^A*Yh&<_WTXOA1BC53os6R zkTZPMl^OKH=EC;HT&w!`<7&}260VE+I#=m$>Y<5x4R<1(hFxryQ3kKBdB3c znKtu~`aNi8^_x_)c%;hi2Q83~@~3D+szm%iOsMqI#r_47d^s|E?RR71_|l+VipUn6 z7P?f4-iy*)+puotEYvCkge?{F>C@VAIb!J}vY~zSC5t4u3{Kf7av!hfSf~s=($FV8 z%Jz0b)fu`oZtw{SbFcnZaZWBwo}}VW6jbS-6CqqSX9YiHCs79H0HjC&wJ@@xDVxO2 zqkji!1=R9aCxRFT4zinN&L3foikjW3EeZm3jsCJlFMW^uqv(VjXcQ$I-6^8P2G=DL zTHz8XYQKaZU_<~A1OzC2K@JRP)!G9EEyps!^svq{!@xe2;eucnN`^mB)ftm$w3)^) zvlMB#$y@EY$&-;`152oyAS7qmQUoMUal3%t^Zoevm8(LuGI6mR#RIz!41pleY}{aw zcIGf8hma5rjXSH=)?qwhBg+OJ;z)NKS+!+Q8+V(7u55_PaKg&@m*fyhp^=odJ62}4 zpv$S30=eHwZg!rR^d|{ArhE29Dw&KsI1w}PKO)mb`e`)M0g>CnOYyI5pEyJ>x7EBV zACZ3RFV)Cn;SMGjBI6Ebs-JWp7Kn&mW~m$+=mlkp-ts~`eFWd zW-E}8V>ze6VnoC<*-;g>FS2&8Bu~LhY)MV-aIA-)##Rj~;ZV&XHnm_BhIe9>l6Qt8qF! zK|Pp$|Ful{!*4qmDXb5OB)>1`!4t_6^YMaCNr&?#J^W_87$pq&1N}3)JXk(&*#{oq zM;jIkX9Uu>)VTDmIp-R>xa-BAk}M9v6Em!W(WIYP92Ky@gG@!N;6VlrWW<*`|M(?Z zjT>A2M!-l`z?xkai++&2h?Rbjw7#*Z!Cr09K1OXtMJI?WQcE<4{;RfVY>BgGwDJh@ z(2L5}VL4&r7=~mym@c`d988h?zC3;fu1J8%nq3r&Y4Ej>m1*#q291fPQK)o0k}pfS zw&;X-meXx?k+p;fk)gNY!f@CrZbozZVAf!P z;LcPrwVkLjeyuJDKX_k%F-MSsv~5Q+BFWe`#}-^pr7&8&M+NyRV7YD>j$$kIv;o_MN#{Z$HIoP@E6 z;|Ob;kEG_mfrB4aOfDZK*j!EWv->&8D{4)9q*qcDa8m)hoaEzKi3gT)!bL~S-aEQuJ$)10!Xq#(*bft zOR(s;j?FK0Ki4p;`od}$x?H?ZbCm*wI~&`KQ-g>HY)eq@hr z)*aINp0cfSA`Lw}Vs0A^%Wt)z=@JTRi)k*dvl3-UxZAu|E|fR!VY%GFi5qX)j~JRGuS+3^Aju zAOCXRVGeesgQTrq@VoBBsUKA)CK1PGMOwu8IW?(p@-+q$AIp7uZmP8D8X7bHBFYE9 zEBa+GJ%F9DJ6!diu})Av=-tp+D~U$E%GIO2HX}VoA&*Ag*O$~B+s=SY;Gkh+nAW-U zc#5Chf3Wnk;Jb}K{PleHf*Lr4v+=KyC6Y7Nzv3#s2!n3Z5>{Z*2#MZoSoL@`H|o(3y&`6 zJu@VXW^ti8CqE?$V5L^)xtgDy?e(KJ76zOSe#ej{)hTHs@WaN~!+gkddM=;J)XvfM z13A9Jw$^fghh1?e`f=}J6E+jYEpfw{n-UBuw36JMF+Y^_Za;Qs48JcC6kTJ0({d#u zPx;?eB0?lb5D$ZI%n20f0>#~6ueVs|ke!pd(3U+KzVv^&dnS-v_@NpWKHrhjLmX_S zONoe5AjWjhj;BpKtmNoi6niV&No|o@r;Eo~)!rxzbtJjQLT~2jFVURCH|<11>FA^> z@dfn$gugWc^;$0YQ{j$n_lKB0C z!2>3p;)xNB*$%oeX$nvAX2|&H=(7kSjKXM~Urpb~8iY|9-g9k@e~I5r+H91$*iy>* zQ63_@_-wI@47VtnQH(MmvxZsb0tblt?LA?=W~1o`fWX+sE1i2L_dW+h!mFH= zD0-8Jh>vT`=XX>AXQe_cFn6uHZ1nq0z0^ z(yz_ece0OBg>Oe~ZP*D8l>-7Tdf#zlv`5X8^(w@Cb~sL3TIVB>@!FAnhY2tP2!SMK zcBAvn6+A!Hj%(^kBu$ABoA%!b^Or8!AQaq%s}KqiuG$06ZiBa%L#B!wt=DCA>XPFu zImgrB`Ye{8ov6RO&)-v)v53#iVjQ6oosGu_ez0G%qRt=GX5gfa#FSi=OsIhsa8@lm zR4DAiE!~h&_Qf4KGs2M|&fWY576UMO;U$LAhX&yiUQaO(KH^YGj;;@zYPp&6*>^v` z6$C-e?>45fWPO|bspehyoHU?tJ5fX=a??}ge^-uTE;S0x(1-5-gzRP+)gYE;(0qFI zE;6XzZ4HZYf`3py2CNg5jy}Zyk+X*nndrj62Pz zQL?(ADTMpUkya-^;l_OnL~eZ|LI+}$F+6gPLGrkFad(+zUoWRtm6zdTK0IO?pe3SG z1@?w#5xFt~2a05AlDqElF@Vm!U6^luM^IBaQyk$acJqEEc>7w+Jpo<5Z5m9DpG^ET56cC}Jc1 zbB%+cUZ$$opRS7c5yGam(^-l1*4e_7u4*U0NT8WxH0;l}|6-T~u0vB$qYWOk3GUHJ zuvkK~)x9-KP;4(U(nSyI0U`LZYw~?CKWO6a#;=89w&VHbrlGfg(kSPmehUl4{g}CPN5h$FNF3P9{js_L(v=> zx~yOj8v7N-f8133$p-@=QXn$`5C7n`((}UaNAm1czcjX){0NXdDXQH?aY}jLlZ5h` z;ErCCGGj~0RX@`zlggMz4jM^r{4=tLo;NneUvgg*!lh5RUQQul;tt-DOw8t_UP>Tj zjy~YzJU&Z^*`jF|$6_X>b`$|vNDf&rDL*Xig|iUw3~m+yW0i`5#~86ww||-B$0zh`^lKsMZgx5+?aEH}YA=2I`C91kr#y z)K;%4bU<$R@KQx=doG7Ont?=?^JeSa4?q8rr;_Mq&B&#fKL)??9rvN*2DC@(9|HJ5 zl1A4>19OJ^9jNmSvHGBRcO75PPZUA|;RNDI6#ZO@#Q>H^Dp|PhV74KbF)SE5mMf+( zw^DVx*^Qq~yUNIyt$(=SWPB9-QO(2YW@C2;PRGRAcNz(+$|wFbg0^F;SZdlh9mA*k zQ_1|&w~m?xPQ7Z>XKCs$EVc=_a(VB((-}WS#}tqMAf`$bQD5i3aAGw)?=(z@H^mJc z%DYX2{sF)TUXWQUV;V>ER30OY_+X}5Bci9Y=zUd46k$KWOR;!v-Eav)E}svnmACzJEP=M0*&?S zeBRwFkHnn6*37>pp4pBo-6cI0FOM8w;(b2NNhDClUbEMPVgVS0S1W%&SlX}j`B=C_ zblgO7+1yXX_+FS;J%?=Pr!H0aPD6{|!}-xo93@3-6FPxa{F_5loXwnk&7)zRvn}EE zOI;2@u({i(Hfef}ycLctnUSu}EQ;(n zQEfco%5csBOW57NdOs(u;N2|=<^{PC1!zXh6_{ao%id>>ZAdQ=6O+UDwWZ1rc%PHm zXASS7zZ!2KU|AFLg;b~Zt&@-RdQ-pJl5Zk-w7X~A49}@-Bk)IW#_2TC+LwHTUIXSf z3JNg$@Rt4r8YKz8;A50il@NW{l10$=M9mqp(T>c%NmTtd{1nAB^*)e@`7%MeIiz1K_4u7Er^IXf;p0w z&~Ka$dgIV95qPR#USmE>s)8CLcChCKCAHkx&;st5;&PXOx<4d|2)4enQ5xl78*w_GQIca^*GHjHOY~t?Sw$Rc9i4?_41NIq1#Aie_uzJP7&AclO6cQ$^zZ2Y`0Qr)shdT#*Pkga zCu0ioo?Cc0x}Ta@(+X$+2p6;<4?7$>UmX2r?0X{t6(t0hn}gRYosT5$Dm%!HKn-pQ zcu8ACdDbfE266};E-l|4=H9%s+kz`$ zKqW#vDMM`8g-PjyujF6Dwe2o-QA9Shi4a4x&Jx+cO24gS@b3V!eT(nHjl33O?Ld%U zR`WWsHL5&_E|nOyvExP32fh+Nmf@KplmG@-8D@)j{nxyAcm4H!rr2@jaacw9ka~00 zD+mXcna58)ynF}1i5lBq$8geFPVUM1`>xrpZdq)hX^U)bcUsFNuHj&_0m?q0b3%;7 zu6c9YaFcWUH6D0-K3W^im{8@}Y|Szbg6E#z{k-O6mYomY3R}sDC5R;0m3qd>p~upX z7 z!$yKT!3h%FU4pwyaCvZdf;&lo;1Vq8#@*fBT{nLC-v3-rSIt^;F>AWJs;9lzR=fmz zF;PM-Tk5{8OIXiySyiD|U_JM;#NiLw8Re{a2 zWbw!O>}4X#W-Y_XDn7uZUx>zQcK#)lj0B~3FFL}~ox9h}9zs=vK(faZdHF2ZQJR}9 z3y76=N8sES3xd4zR3juG$(W|Bmz;L2VM{?F-BCZXe>`hnhwVJ;rKoser7ahs`9|&! zsze@5hk4xoP^A3r(03T+Izp76!@4gVQTYkW{8t7p+gg5j?=m28T;z}`g5i!RP~5LL zwf+!h=T7M9OAj*JB;jteDg=kWNnS^{J-&qwoF#Ln6kQ{!E82%g!m- z_S{o29Y(4Xv`?Qkb=VG6eeDTE*k+C9iSfWVW{Uu@hxuuZp)@ByZ4SU9{_y8`hlFh+%THZ`uUoRVRr779Z zPAU7^=AxVnPCZ&uu#!&@3Mji))7Kxe+i}B(=Ufp? z3q*GdM*RZmF3k17L}xFA?udijOJA4VH;T9gGt-IUuB;QcqS~xKCt)!7gJ+mS(z^0m zq9+NfU;x#@Z#+OcEEP3P@DzF6jw?qeELsPnbQLxb>@vjk)v@Mh=K@ln5XRl~EF+%n z?abcY&zljfj!2f{8vK(qbVI_9?+y!-5PK+U=6foCSWwH|e&e%4Ge*dg1>5RP-RyXW zfuUjD@!~RQ**-g2FlywmEuG8|cy&4SIb+~k1IF?>_@kH$l@O)%0&(_|h%cUK<{T2} z7d%&QX(P+@tky$Z3W+(_Vs%zxc_~uN5_!Mt)a0mj+KrEvsU@;lBBDw4#=IhHfosc2M{1??qVkhpL3-Lg3BN%Lf}sOx74-w! zAtdsk78_cfj=Hzx_W0brO_PQU*;BRtMu-2O>g%nvTcLzzWLg>4jF*PJXMh# zmrw}d`f^}7=xnGDTbqbm<^t1D<})Ns*x#V*=0Gn(Iy#&9!o8o0Oa*Z;H%2)DF-$;7 zk##INg_b<~6vt1uMWlPF&4qCjblF0OEvJ#Q4CN&|yz^+DYwO_cIBsvuGH6Jc%9Qj3 zX0hFw&+B;pF68rsJ#?;QJH^<}{)$Hi+1I-lE_~#H)UqLIb=V6|Kv5xn8}ekM-sTm! zcmNYqU&R8QzN6B6w)>>$ycU474t3~CssnqTN>22)H%nzuedJo4E%y8{eS1!b&MS#l zUl9}js!0#$fZ!+PfRO?haMEnoPgcG>$LnfBzJHmuYb!XrB*`vGF0xhG?~=1YzP2hC zS{eC|>D|SVov}NDqP|Ibkis&B4R)cro@7ryv7;qVz|BGYc5P&R-W1q7$BD1SSs7%< zcP{5|0Y3>}q$w1=o}kXrt6ujMEr@;dOs$=o&D5O7ow`fg2y_l zb=54!$9U*ch~RQm`)CWjrEx!j(ER19TjA8YknPGz3Y9e~X8tX7LNxNCNT{{a<#@8Y zn2J2ok!x=cd253EDi5eDnpe)a8^LszV5|Zyg)=o4rE!^j7Qt-y)W;~I{5X{W_wcRg zC4I;d+W0a@&t^p<5h`u1@kLnf3EB`NTo{F#2~4V~F^AMMUzL+4KSbz3idQL?;2i2Z zT%q%~!W4UiQ__M^+$@84Vq=UK?7$w-^(60d$&nyyGF~;TPzQ9cn4m5%B3(mDD=z|C zWV7;Tvl=~-A9%z(Jw@0=#URWFe6+I_=iX^IFNA*a=Qa^@@CVPC4u5&PpWRP8U-zd1 z3*4F3J%Pnvg2#s5ZyqCPYlM(x@8(AP`3L>Gxp-$@^EYOeV9)Si_vbL)!Yna`{;OB` z^OH!qSXf4Xen7aza@UT(%EQ)>K0_VvQGYdgVxIOUqo{a)!9z!^=Hw6dRkND0wD z470@Ou^qxA7GJf0Kh7R05x`S;(+1O~@e{Wva=y$}UjR?5G()!UjUj#aRLdWadxMe) ztS+2@Q9M6{7k=)J&_JI9QdRy+!I(qch~rBTx=&CXSH084#9wZQJ}=L^NQiD z!~GjJynbGQVd#u%Tv65cT+xL?d{Zr@Wg?-z0xyQ{=2p3z%mQbwYqn;hUO~#d^ zbWbR~DG`EnbaFU4AY^qWY#gLVR}?l3ZEHA-P24O(0MA+%P~6J%!z}*jx3>Dg{C-F9 z0*-5fZk3j6!Nu7VL**xjiTAN`JkIb&*X4Y+jDnEaLyw|iR>-vcK{7X4Nvzazq(TON9ggUN4C8Edtj z-)-2Lp6cZBnbqU+f>dOKgFn6RKU>|N0Oix!`|c5Hh2{A#F&Wq)iO%s_f&2K!3X0Ky_$&^4YnsFVixnf09z5ElnT8dSlbTA={gV;N$Rm{fJ0Zyv#&Lh|1`F z-ga6*VMRtnlO7_gmyUD1S|air`VIbFT-ql7%rWtr;ez31&VDZJ!H>M61@6aL;Cbw# zhs7sxCFk}^3{|yY$@^CJy9)=TtZ9I_$?kt0+G2^Ix>{o&g%>D2!k)ui+YNL<70<{e zpa<1~mF#2#@2~ai6>iIrI=Nw7CCdD!m;C(W7yhHe2#bDCjshgDu+J2cc`qoXmvhHA zFvM~=TmF=9dSmETpvIH46hliOmu^6L^_ky##F^*-Bb0oL_@DE0#4VB7qPA}L;c74b z!%TipJl%7^KQbJv+c-BLZ?HjfF8>E&#=<0%>r6H6l18&}$^zj=WhHgoBhZm_NglEr zz{QuSJfB3j?Dn;U+%l;?UHY%?q=I2l_hl6{>^-jqf;?2b$t31D5DN?8${}nfs$VSn)U~NnB zzm1Pek=?M|Kr_T*LZ4HV=S_A#7b>HdBHtR#z`;rdP9Se`jGW6z7TC-&8k+aTeA?st zuoxE9vBy_w-%8yHa2MaTor{f}wpa-;sZ=(Re6WbVWi+105qg71C1&X_9NW68GMjAZ zF#k>G$2iewxTVe4&<|&*dQu|FO>qPhH=_!rHYH_zIIg196O;0DFIs?#0N8by*^_$#H@j z>%a7^W`2#qj|E!9R!3%`6nwNOJ;pBbd*K{>WH&xi_-2Fl=*k~vL4?OKcIcoXGmp>G zKL!t!kED;?<5`B1^hioA(0AOv;vry}*Yc214Vl@{!+r&W+o{4k#ah$yy*EG#|?!=b=(+n&P#|XG4>YIoggvyrcW2NmVFBEl4 z%2m<%N+i30mr^HWa#nGl3vM zqjad=A!mYGsGEj*y5I+MhLvd_DOKaa9Jq^(S#<(iSufm5aQH>}kHn{q&-_xk!EJ&F z2)3kr02)1IP>;CW`nG-8sHv2`BHBE=Bm#HpdVZ}K#E&4VY;Dk~x()-_!X0+O9iECZ zFPsx|A-C(`F?O+Mt7yGSAzYF*2mQ>a=^ah_Z=i|3;soo#Q50nq zTrJeU?7tN$GHA)gD>xL2!lp>;Rqk;(n45dQ0OT4spN;BOI6fpwi{%m`q4|hr?Pck}bM8fKro& zUufV)ZQ7Fvp9h4W2}M8Q2akV)rxm2V_Ckc;%tSgF?g}FiFyVlI{SXx^SJ5BJjxqxL zHWzK=u!D4VYF~UQCc$sy9iqFG|C=Of^2GMns%9gDLE{2GaIvfHq5_`A}P+R#omSApp7Lh<9$O zWVS_{PP%N6&ACJ~S<56^%fv~)B6S_0WiPQwCitsb=5e8vpqXsDWmG}iE#buwIha_l z`}hMZx%3kat^L#yksv}rBUlNY_VIDxoZR|;-y0{c?Q?bzHeNRAl*8AP(#PvJXrUqlRC}Px z)#}+{+h5iwfqeQiH3r-Y&>7Lp3g)ESmouBh3?6bz7}m|Y=o8Nk&Lysx8}WZ?G1r?{ zNnNcjh2TvMdDJlxYV2f{6VhC*lIKXb@A6pPNTuu$+r;j)R7A4uq&O<5JkU0N&G5B- z!A9$#;k6ePQpa*C=T*IE*y~IHw7_l}xY>dvzmY3*q`|ulF7M0gP8GAr2Elcvbc7(K zerv9zR#G1?0ULbTD^C&`Dn}O5bFfFz#KQu~(wh8MYxejMT-ZJA* zh{{r8N;AnXlUJ0Aviy;TDQ(K)5yhDVom%A|gC|CwI2kE|KE3&w+h~BTj8KOCZ)mZx zva}YP0VW+%S*g0VNtN%KcPrR!D!rF>j=yrJ8m7xAil=O3>6x3chEm+wgu|wWa`|A$ zS_-E=_8~LaGYWlhVW_AKn@aM@*SD_!Nnf$!MZQr}#!yidH#Jv%|5mO)+5X1IN<7|kJtb|9aOlhzA!K=aI(naK6ABIR)OTZg z+ZR*9udR=;K4&7GjMT(7piZsJeN6OmJFBx#e^oZd7x!7of%T0}-pbSYk)00IWMs(u za~8dOT!Jz{tK&4=i>6-mwZ=HPVF<0%7Tb*p`n#7OMl?T?d!P}})@c0kgE4;7o=KoH zUi24FFcrBhx>7G$4r15ETHn`Rwot8b?Fqs+<2eW{9s>OHN{q?xMrJDx6cNIxZZeYO zj>GHOx0Pxq09UNs(9E^D_=OkA$-Tt{c&?01BAfqYG;*!m$LO_<=_wmt!}gH6tMg^W`tN8vz_{J+LS?UikZg6T+F8m7EAS)8 z!H59kR39?(NvY_Ay`_(KY}mg17ym&2$96jU+}@2U>up@=F!}{^-D(MO76L)&2&o-T zgr1EfG=TeAen16JBFC4s4!>hfzwO1TJBS7qazoy&zq7FZUP{n2aL3U(Rs3Z-f;ms! zlS07)obmdes7Ep7c?rAvMchD%edsfq{AvdyEKvo%kV|35A%DJn;TD~4{gA!|V_t6h zem;hx{Xtcsh~Q6mA6mlO{cM+W6F49fyD-q<=aG`=TaftmmK^`3`fRag#xA4R$!ZDI zwYFo^h}wP{;g>FdqS5-ik5w^X{Z)Ktj7$!IWBc$x#>gsizSB@#o;u2A&r|t4czCTE z#~b%ks;?$=uS@rRp^myQd~Z{u0`I@@NC-#n7IA@yTRx&1_8K(|cX>}s5CUs}zW(NI z@Uhy)3g7dvR|yBY#7b9t)Rf5e3!-7ktwD_;B)QJxFgcj2i-aZU=rB51LO(7)x&{h} zhEsx7h#pN3lklAij2wca2^_Z|u=x0$9A|WepD@i}T~0?Q($bvKO+=p+)&OJD3YC}{ zr4gSvK4QGD>krzsPQTRdGTtxNcyL>LUo$W04A=QQC!U0J9|gZpRl04IqHUkiR+JJT z>{>xmM?1o>wDWz%^UwyF3WJJXv9y8AvBjx|Fb8K_sd&T&5&v^Hcf>zXYnho|hZo z8KBaodmiw`Yy%b=MjJ^fdGwYljhTPq&N2L*$690mU&W;bt9eRt(0;) zTFt7tXk_i~Xd7SA0DI7{v>WNWVe>ZzvrVBJC4Elp<=3Ba?%-^~?g0ye#gj=q6#qb> z@Jl#`Eyw&NIYet-9m^J@LSRuF(HK$b3~Xx2{xU%56di5;qR5IS8bc>QpLF1H=n;L6 ziqi2nb+q`~Gln&2wtZ%km%=QXS{sLfmVk89HoC-l#& z?>c4J!RTdSon`TF+~w^ChjSFUn_UraGLiR$M7uO#r@yUn6RhZ-OMLUF{MtB;RCy9AIuc!PF0(jMedDb z#@GXVVII!h|B?kmjMRJ?TO~pe|57%3tkFOo2<_d-{9l3B+w*D)lju~J2HZN9{+n;Q zY}8$jY!Z_(fZMVoA^+F$pW;l*_$sia_BQ)P&?)tc8y1T$wjY*-GC1iHv0 z54k)U-zooM1M|?+;X)QD*=c_|P>w7&w}i)UvJwvsKXi?Z3Rw004lQ@~OXsJX!+RC_ zlLESoJ>w`2u&B0-s}8fv(#A>tH6(9xv0~CuLKZgL4Wyzt4}@f(2wD!bSvK!|_#K%F zo^Ghn8$vKq0pwK;n*MeBkQ#D?cOa>8I66#dSk(6WkZV0fQ*#>BH;3mkK5O~!HBIrq z$WKZBSl+wIWu4g`fyTEfhoBUHCRQ#Xdql7m7Dl3arooq%lBizoS@A4^VtZm3<&k)*$Y7>s79gK;s@V=0?)_u0%5Hz zr(g@F^SeAv+TS?hV{OR|o`htFtk*UhNJrL_#kP===bL6iP;AiSLfdHg}O(>vFGc`#>z~OTLsLzbb=JiPtx># zC}Do~G`Ly1)}uMIqt_!ohH;|bi-cwVf{tkUw>cgoV5`EaC$NOk^!RWpkPkidQvG&C z)e+w6Bh{iEyyA@rn)L$U;`^;CJ4AYt>*F&#gN651-IB>LnE1{tVvV z3Ej1sY2{ITf&~%Fb&B{aHTWa>D+C!jw{XY%5VOFyJ-5meLe=+?5*YF)aEU+Y9OYKl(ut2+AitCU>Zx))MK z2vjI)E~@P0goGijc&c2SAg+e*77%EnQ;iiHr8rsRx=E56Wcj>5@|`yE)~j*;6l!@7 zM6-0UZOF-eWPvH2vTE{~CS>?P64#;~4tYwBpZo1mol^PvSFt3$yP9DKS`BbnoM3M4 zkQ+zfFcw*|WeRON7kDKv2jRPrOHFR?0H@i%Il{UxdHUrDiX!2-MS0>>j1RgV7oBW_fG+yuBjX zJF3&-HAYVgJyhiUF$kAFNUh3M`H06^-Nn=9^6-=um)} zS}CKY;s@CWO1*PQ~sjd3Sw+y<_sv6vOPrp z1YNXMT7uXq*Tz{SImz%jkgX~P+hCY`&COERwcj;~u`@Pk!@)%S z)kCqgy(RYLt_j-BiO^8hn0Z5>7sikY&2tg2iuvG7SiXQFpHjN$-rQ&t>jOEAxD&59 z{N7Rg+&La@sujukV4=Q5?T>1)Nuo4nKX?d+k)t8%Qxyk;7h5#gWG-y16OL%{dJ_zAgpm~G z!oQ~uAvig2JWP3$AevZ^2>34p-1L%owZVODG3Nh5Kqg%~oc0jKK1?LGI3W7ls64u==0 zYN)9uSlX%iTKvYeyDd6kK-PvSpF*F4Du1VPWZoZm4bFTo>JF zW5v2tpIt)2?h1^fVRaO^5^l8>xwv`&`*JNd;^X~0qVr=sS}qKt23?;N30N=hT?uhD zyH?%Mr0Zd{C1z?lW748ac*gE3-_ zM5zgEH>p0zUNv4P^tu`PTebNYBLG{NC<@a8-*_@aN8&E>dz}#J3FFY)fkr8ZtmElr z%>1uIj?15TTCZg7u1^u|1(BKw*~V-M3|$q&-6t0yNJLzr2{RO>B;D5)?_| zZ=WH)W?Mn-2hDoC@mD9IUT5IDHK7&5We+t$Bvpev1`eWTw%ChDpJ@uQ0uRijT@gXY z)`@7(qTjxk8CphZEcCS?0f?e-`8Jj+>M$5NBGw-R%!HC_p(#p2{h;6+ZMSD(;3!~} z>#(r^RZBj=#>ehQZ(7!*sF4)fV(ULO! z{x?bttjz=vYWf)o1Wsmk&hZa>BtuD%ozp^>u(LRpM7K4h_vw~FTw(HfFN@vc&S@{h znbs7xl+@(c+Om5qr(g^_mKk2NJNFcxt#dY8WS%L*J(3~QIPo6nxd8cV=dx~Z9(Ge} zq)5fJ5AVCC^uhrP$TYb;+bt5&fWSK1VgrDp@fGds-?6h*ir0jWcLwXYZh#-Q*O0Eb!`;aS&VI=&@ zW)qYW+;?57-zM|^!aSSH7p=GNm&GD(0^gMLJK+jGidHDZuL|Im(0xz(gczCC2DD&D zPn$oNcZP;CJJ^b*>?=KHu1^HKwbls{i)~+x0oAaMaUL_}mLndmHxA(j;j6ykqM0Ut z(lDTl(hT4N^JpgQ_0paN#O{D~0RAWpVgvax-r@iFG}>PlKhK-f>_$Sqi<4aW&YeOLi{SoFcDk{)ce=`y|>`3Eov8ovMIFV_S%136~ zt)et9rP*P*bayJ67h5cJtVtX<3jWSj9oAT!e5Ym2Sae|N?7AgqDJiBI12dc#CKB3S z-+Fxm9DVbcnte{;@tP>=K(mAEB|>GH-jHeMi7Oml&+bL_pA3AfgbiK?www&23D-Pa zPR9$j3bsR@ud0fNRL{Sm3ANP@MBCUT4LGt#CJvM_T$%onBopVrduRk*72|ld{@QO9 zu2hR|eK$T1{$HH4$$xR!|38ix>j67Ta1HEkP&E_D?ax7|p6Y2Jqo4|5HlS1Wl*IM- z7@yL?0p2+^As)Xj=+r%1jjmfAeql2{m+)4WU$iC>{km9Qk@$5{%Fs$%*L9r`gZpZ! z3c+R&T+8(}=opX0_5Yl$uKE8-{Yh0lHE<2mbT`>SrlD`4IIlmDMb5GpRa|XC$?r=U z1vOR6)M#~4Aeu1%>)}N1dEn>Fg7XZZvc;t6Hl7CmPpgd6=*FVFM+wkvke)UzqQ{DE zluihv16OR&$$9;_892r`)64JEj6YKfr2s|mc>S#X8_BjOb;OH6d|Yft6#I!2w(B+g z`B+&OQg#A$O!tLk4|g21Ocgn@&ZVnT**jDar-RDPg3U zu#iMeL%>*>Thr(1X^C3ujME%gf=IclUUpf?&Do6i%wNua_&j(Q0}g6u8HnNW?N9E) zb83ugBL6uybYilGnzBDVgb{ei_RYS32Q!lF%hCt_`EHJ{BPJS4G5F8AF=V`|-nT2b zcpl+gY^LAA$=^u#q#xrREVZ8FtgTcbZ!8l--6yll?#Qt1QhAdX>0E5Cf4)+AbIo3< z73$pPUNi{m!`%MmvdH_x*~{>iwWn#CLs6uW_02bCx{OWH(UL*YgoItO)D>RAT>s3h i%g1EJns8p|elX;nK`+!f9dz&p%L|oXc>k;q_5T1rD%x=X delta 26420 zcmYgXRa9L~v&7x~;O-8=-66P>;K7~Xx^Z_15F7%5;1=B7Ex5Zo1ebgA{V(^my1MqP zW~Qcl)*TJe!^v58b=($9a&?>0S*FTh8Cy##84T{Hzz18{%=vm)AxW z+otd6ov+-Zow9INk4X`!2SeEiC@v2J=tAY{C>j9l|{@KS>=SN%03N-j#C*_N-v3Z`Fj{~GMu%j;C`s6VDBH8@FIWY1 zFlxmwF)xunFNWN0Ep8H>uYtUa4;IGI4Jn5e%oju6vKBWY&%a4V$cT8d-U6PswdWPH z0zh{0@OBD%jpVEIzz>(F4` zenUTSxE~R@P!j_OA+W5-gebc!NyUO3sa;E@f;u+0+IGT1(6*rEAP~s|ce|;uP8T7m zVM1a2vV(1?rv(2g9$^54SV*zZatMfsNH=h6a^x=m$XyiN)Qi^8a+k3o?%(95bUtJ? zt|d(Ubbvq%C*L~Q+7+dGG%y2Yj1-__8$wzx z=o^5NGn`J-##uo255EX=3NS2)Sv+EZy_O>0gJv;kd+ ze2{brVh?T-6hU2x9UEFDvxrDc5q)PQu!IQU!a70WhFH=qlE5e7fu2)F1^A^k_HSR$?o-B-N8r*Y-+a}w9{2dO;>xGo4UNdkiERyg*}=h|TA zvIuA-m#{GwfcuO5CA32&3s#ISM<`^Q&_OU)fo|FUjmoqdf0cmKso#2z=lBe_FB;`(aj8?oTtnqwUJ@iN{< zlIQhMXnJzB5oK;%-#7_z04m2spTv+|F{%n;v42p99zw!R8q=?!db^C+xT86rH&ESr zDdGi!*r6@*V)&v!r+iRco}I4wbEW#INY4>_f05C2*y}&tVh%y_lfz?0{ut`6q)rI- z;MaV25AIQ>boT9}i2Oxu3~d*H8iFNj61=9+1$RNFo>lbfzxP!&9Z>A!RF2El+Lips z4+$=)FL@3h6dBShjPK=ScvH64ID9Nfnt5k;1fHvY7UvRDqzgVM>dR%f1LI~d=7Xv? z?@;eU9(zT^mjyuy*09tbC=ueG@LZ$!Faiu{wa^IhiO{6D0(2}Ps{GqmST2qaLfKBf zq7+ow5YXU1f$B$1AOcb_0Rnn`gX2&uA2L*4>h$Rif=Ff)QYZl4ioEnTAX|77@`ChG zXA3g4Nb=PCw`uMSBsw@OwyK&Bf0_e{?m0=G%qW!SJPN-cOtyIYzhDJ8h!hA#dU+nm zxK1`jKusNYfllrB2?i{C3GpZ%N?IKaEws4;4OrtYpj$Dcs1KgNt8YUIIi%|# zmfxd2Y$5o|X`uV7HUpB!hwvahB!rg$vP)?3zQh;fznK5RRy1?Z;(KstEQ0|vG0goi z{$1z~fdSjP{<;?om%)JOEf>14YP!xX$;JHYi9FijTovct_-T} zft0)|W-shTqb+m$s)=8z%|j)C#wk(B>u`#t4c@d6xb%!d>fOo=c|rH~xu3wBdMg#3 zQzGuO*zXo<-2XQHC!~1IYAU3X%~ylaYXkDTqM!lq9xp+JyWjvMFd;$7vnXioqZsUl z=AnCON>Gn-!HRQm5DNV_7-g_g1q$_KVj~%Vt!M<1&<9~pIS&Z?2cuAO(=EP2D_@?4 z7?vr}(W8*1t|*+Ih8RX)Ledd$hcqr+j}A*P-YDOCaf6h`sRe%IgrK@&eB`m?msY5a zfRTRClA>c3?`yCwUy7BYbEgcAVfBM45W|TG(%;rBs6IiI+S0?Z9{fRnTZHp#B^+cO z9s{LDjwB@xXgPDLEhfSi@O95?YXwReIpw=VGf^UwLsDC@x{CKGvj<3ZB}^@RDG7nW zg#%wRQlJ)cH7o7_sip)~0=vTynD75kO`;ZZ6Io(0N@KZ0CIMIfdhtOuQ8~qqxZ=jg zfRPJE%bXVsm_Tv_J}C^E0~z_|M2T?hseHS28^FCL{|`zEVLBvNf>nqETQgpH8HLJ^pJr=4i!V>kqmZWPLYn}uQE z>cyCs;Gz3DPVZM*BRS4_IA(F2qj-9LQGWmi*c35Lh+{EApP(cjHga1J0x<|1M-qU< zuWQAvvZ0U^Bl#-U{hS zn56w9NK4K982yisU_rcs6)FJz@u5L#IsR2w;!7TwV1dZh0z(MSH%26k{qnhIg1b7z zmaPWzX*CafZ>h zOFsDzg5B)BJKicIA;i%CY9bb}C5_0AD13ssag9Q7Iyi4G}s^-;R zQoyv9i6PKMaNh3(cEm_nhRF*+B9cSaNbylu-yq0gpdjAPVKB;zLHCHbi7HPI5t zaj@uP2m#lmfCNY^F{qhPCpqImqy%yTtV3IDSAQq5cvV5H7X#Ysn->}yl`a3T934DN z64ag9i zh|BQ$flN-Y@_h^jCTUnrJ+KPXN#4pR63P16>3J!2<~Ug8AQfQl@3XB0s$@|HHOoJ zcu=-Y0L-M^(6+yu%oM))&?O&cHwv8^9Qf`>2z=>@$2gsWqVU9%@SeZWeozbUD53mz z|4338{^`aC zXzbmB*iUPP0~31e`NDR?La1#Uzk z5JS*RzUHEpNLqNUUKL7D^Fg4~_mV$nS{r5@c6Ve1iXLVlS+t8f0O9M~cf z9ZGLrn_EKRG=~AVWDvos+x1=5t%w0ze$f%i8*J>{4n)(=tDDOxoLaE|5r+N|GLZwe zKm^f}2l4n%gfWE&A;=)(PblC_{SK9-(6xt^KseTb^CJn}RsmVj|2yvK2ucX8w3IgcNv;@SB7RaG@4?|)q!PnSv$jQnt`Dl@N%i8Tdl`R`? zDAc?_V8|peYzDiq&@%!vvAwS>xtCyb_p&yoYecMbn0 z0A-9%Sz*}jZjf}o8!iHO_vaxU2vp8(U7V2sBFBI0)&$XQNl{D(V0Z8*-ugL9?4099 zu4Lzr9nRQ-FK9@*2wvrG#Ma$^akZ8Wiw{{-X>QbUpY3T)?0?J8tBESVyq%+dHl%<@ zLQ4$h$5$|og`xJG5KhLT+QrtWYefO66c2Ibk3v9$+r9*2%?QZW1_R8D50L!zh5&r2F+}l;fBm_F&h;^!f0^z|88F^626tw$A z0DRj9QSI%z@&x(R0#ACPa~qSHZsXSi2k_|%NXS2@rsqaZ=z5lg(}Sg)YRZ;cx#F2V zixGkZe?SO4Nr1b@LkM)>;ZAJ6q6ZqOF?MM635qM>7S?kU0{!I+=d)?Y(>R@G08$Bx zp>({Zl)P<%DfFa=k>WB9k8WlWVF#|%jC<=^0SetTMDlO|60V%hfPp+(0z68jS{rdqo`~>zcm2?T7r}fo!~% zI5GidnUoAjS};TG&t^EH`w|593m1j0)*wu%AiZl{Did}sULv$2UkD=EL2qX3;kBU! z=|3@m_@5tBe;qRJLFCcY%`9Bh4z$MwS114;2hD9W(15QCdQ#}{t`vN;wdJq4YzLyy znnVBVq62EX(a}$9AxS+q>Y6pO(Qn;QIfddfxjiBX!N)`lK?*e|#8?&;VT;;_`@FJQtgWMv{oevt!gW60_W^bZg~9 z)0%N&!ato5v5`$R1oyE33?bxvgCdOEITVKr%6eoB~vki?B9 zH&X@gtp0xx9J@!wECstpPX39zB+Mtc|D54NuqXwNH(Urt0tU_0etcODG;lP&1z6VYd#T1$!S7AF8jruZ+Z=^3F6)?!P z`uQ{Lq8np$s0lVMy1URA0yYl<2ZL2J6slDR0=DC#lK7qxBd|kNe4ziVVT6N-EdUZc zSyn0@m1PMGxNXr>?nS3RXS|rX*_5+id>0B$5hNv_6wD8{9b_-cAC8>qRHafOVqVo)J^R1Z z)G1vtgDq6=!Q~nJ#{^aU>8DytckfNR&$!*p2z0 zrsv977D_nq0q=rV>z%$dmPOm_<|0)`cb&o)DpYooV<0hx29L2R=(1me4J5kYHh2{- zyWsq^LRJiLvA-fLLJ%24d-c^Ust(Z@OM2R=rYl=nK##+KQ@C+V>cAr_7$sAuKV{+u z9WMI2(#4iD0Jxq$MYKe`9&_$`fiF;g8isdy~41hw;HL^h2!U^LJNIV zn9@HL9+1+n1ALdC^_|Xq{`Dw7&E~j~3^kX-ejJDP@*PKg;_ zE1q`9W(p>QJPlD9+Z`^L*4LI&*p|jRG-E2QvF}$*f?6gm-an~6wtrH;|4C_!V|RQ= z%>MhN)xn$3CK1U6qTp3xhSQ3rmDBtuB@C7dErd~eSdbER&?!a?{Uw(*jL>Y;gHlSW zgpriyt_JxG^<2>Cq7EsD0XRWBEDobipUnNyCl&_@dqJ+F(s3q^R0)5OwKO$c48=_T zWzYPDUDT)pW1M%A67>1I&HKZi<7jT~h=54+TU+*st&ac8V{h5>)hd57G#@8GFDmZ8 zxg7r>lo9juEAiOqP|T~o|9E(vf`8(0@?Gk}C@Wp0P~`W=r{R}R+CZ!?R<^XxpB18K z-iSA1zqDW)Q9Fw<(WtRVuI%sqFE&>`oe}<>vDd?EG2h2{EzEWaMoz@J%{!zFc}D)E z8_2X3rXNA`PTD(MlyBIiE!sD_6>fcAxHUUnU0&^XDhoI0sz9xVcJD%4%!YCo?ys&x zrZqkwaxa3__*<&cv<1X}S#tG1u00~iL+0~@at%P6f^5+UsH`Gzb(euV^rbtpv`!)p zb$x=W(+!)b(4K_<+r_3tv)T?CA|d;!hyW99r7-Kt|GYnsxV@|Fp`RN}n?fly%DTsCY%-YvRYyQ(F? zMG?#cOACe`WA&HTQR!$NJYz4Omd~lg_XzmC+GY(eHv7S~5}ZGEou~O`{0v$%a{t{? zH{cMCk6Iw!AQ0^?pvMzD(O*$59yNIK)@pg?zZBcBc*sWK(gopM$jMJGTYtqVq(8;h015IIGKYN@RnO312*OE#Qs=&PZ>8!- zIv0xq3U5EjQbcvEPoQM~Q1z~Wdnn{@ClTqUdytakZ)jC1cSXs4xRZCEKkk!-jOuV@ zPR@6M8ZoV&LoO0OBajfMN063;{avSeih`8?fUnRAq>)O}BXN}|hE1r^TU7r=F5)M_ z`5$e%u2r?X5I)~i#e2I>`GVfDz)I2E6eQSKXvwkqK_*IXdGBL!T1OPAT#X>Z2-?6; zdLB<+-kmyMSk!V%agvfB-LcBUkD5t`)OkNxBPX!SFhk=?C5j1M>=0eq;rF|Usd?Qv zfpCLWW3-KZxrthbXM%lIf>mlxSBHJwm#%PoKhv%seq)2Ff zkBSR57iaT==M@8lpO-=9fgh^jJ}73Fn+&q+_CAX!IKR^`t2V_Z#boX9C|H#L4*v2P z37QWAev{nc7bYK)J@V#dN54vzg)9IOK0xAcwoB~s!11Mml=$-qUKb>oSfP;JWm3!- z(0gCZV0mO3O}4Q)gY6*tVyhh-oN6FHV*@A4xM%?v*M-R^ai~re2Q1bg6JL?E z4xLV)lZ;X$1*hqTO|p3E#duOhsCUzHVZef4sanzrG5a09kVgRpea_y@1@x;Xw5eMz zG*O@{8H}PJKMU|nRUrK?_Xo{+oCDh%_+phl#E!xNhj`pV~9R+%U$3YWn4(Q_PVK>{iJuUXKl->hs2|>j_&ABq}y*)qt5MJ-7kErhOY~!sa z3tNMWZiH%x(}TYweX`t*y(M+->(|E}I01KI)pTf=Q`5M^zsJ@9Yg}GTPMhmR81{F? zy@i~NA6y`V4(7N;&qo0vqYgB%=M}tYk-lDyZjc)7T>l2vmG|bDMBiSv+B}6^CUzm> z#>hrg63CkFsCHopl=k@2(-J|?%`X|~wP{9X*VIRQe`|ZK?Q;6Yd|%D{S(s!+q9+Q< zj?Q9ViKKwIj!~dHiHwKqJ%vDezwi6Rm<{mzq1{NTb3^9w#bH3vf{2UrE6Uzqax5+* zN;aastv_+t8uItEa31R$K$056barp)#_qVUDM)ysF?TG0H_!t&MfDPGy<5<8fq4C| z>&?XZ&}O#fdXc3bn{d+kt(Ms}l1rAF=HZgSW?ddnf`jM3_%nxp~Dt>kDx<}h%r$u?j$6})H|O*)lp__M$= z!08#+z}343v7&{^Vk&4CepYSZVuxSWU_kNM=l*7&MsnmQ7}{;g9DHEYT8??KqNV2Z z9}W1&OFL}HUK=qLoNV|wQTw3V@F#6##CCEV{OQW83Zc}U<_t-jxE3sH)El?$8Rn0+ zhfgA;21YWX!zn&qxjqwednRW~(o*p%W*oY~pFu(4@bu6#Rif{hug&$LS5qZq+3y>P zFGnrF+d=2f?p7Jz{aJ#%J{&{F`uYKIa$WX&juyUaY*lh!df^4`ddATIBKq)PXNK|&`_~+`HzK4GcPLL8^ZMV7yn2pkbryuhr zK%zy3+A^uzZf;%ngcc#)QZ=@v4t`qp`ZvCd$=g~#P$TP3nZv>Hr28_c|7^Y_nfh${ z3h!s}0`d1KfRKgkNT2k`SAKSnN{?wm<<7>G-v4ndrIrs~*7o-EYPp3cO*r$@O>0{T z7j)SiKZl-Qh6gS1r{P@)Xe9^U?aywkYLoh+4IH4=I1fn7l|Hu@Fam=n&9rA7%oXwL z)7r&zqCwi&_hw(&suuDUUoFr+&8H1q*g7JAtF#(?m$Lrr__H*%`0qVTmUa=zTXx4E z&cfaCy*{t;F22sS^ybb*Q#^g*z{9O}h@FT;=I85p_s29KIx85xYHCpL zd2;%tYTD}B>1T$K)d!kUZdOm->`~6aw@Dzgx7L>=XZU{qp0>4>$I(+?fQtTYy5I$*8MP(dmv*kEFX`p9C}QkWu*&>-e6VTr+&PJqCa(~JRmffl_)Vpm zEN$$`@!?`i=w)=t5Jy|g))6;a0`}denb_dFwF`Udqd(8&`(IPT4Q1r+-biCZQ8xn! zEWCgO{D4lDH-qn@1~-%AdnY{JzH9q8wT}x{X=}ES1RPS^ztdxPa$1MTWNUNkoJoGJ z`>qd({%Flc+cj^lZhA52VejS7a(q;b1Eqr=Tpg3sXzd^cD2FFRzv;JTi!1T9y)>?E zntJl)jK=$WPVQc9cq(Jp{hfbJ3$lac#s8$A>~&}Q;>KKeV)f)^7q{oi_noEEEN#jT z(f~UCbtaA79ug04G;SVkbB8XlZs~P=4NB}@!=-4T7dB1b9+Ch+<{GPQ9<_9uFABTa zYWzKF&|H!r|ME*-h%4E%xS!gPC$e+ch%vfyYb%DaU1-PWE!C;fi6;B;YowpZP9Vr%=`Ug?&C1+efjZl`;Py#L9xF^YxC5M zm(4qMue-fP=*1%-+qkcoMCWHd@AMsEELTf#ud30v40Lye=#||=mimp2_Ybm{njeub z>3YR{`g`zj_VDY}y;x7QIH#tcbk*xBd+y%i34B&&<9AAaAG6%< zmlEu~RJ`dX@IUL%5+%7Y>b zT)Jn5+(iCBuJk)k%7)|9n(tuZ?%RTERwss%gb+Ju$4K&qRyAwYVE1Hr?tUw8t)=zy z$D^;}ST)NN*Ue};|6Ipl*A98Rchh*!GAg%WwKn&0F3h~G@8fmsHPqYIp5V%hs%y=s z)!~}=)b}H|w=h-bil*$836}kNqMZ}9j<7LVUJGX+-M>-b!wv5vCj}cRKf^8=i)|W1 zr9)t?FMjDI>1^FXzuluS?(3*YyaHF>3wv}~kH|}kTch}%ZzwudFIHDJK-R|vy1m%@T>~~_p+kQ0(?8fN zPrhz;8!_Xn8s}G>eM5bp{=BsOYPs+N|9l|Pwli1pEMAAP+;HVX_B?c+U#7Jm(^>3P zm8I9CzaPNx-kT9sfqz7Go-4rn1N?IcPtH+CpjRMACaw44;AX|;5r|(9%D}Q}_<4DF z?O)##c=RBn-1wrgHOOXU@LbpQBl0`;gY%?+K)A-E#qu5#AA{>YGjnC)bcJ{SUR8hX%GoV#^Q9rKb0l|NnlsMB zVxQZQer!w@!Sb=KDJ$K=_FnYccYE3sbrDP7jQ!48kDIgoJtJrK+{wE1Kw1nN7sfMg>LM-_`A0!-`h8%nIms> z7A_lw_3xGp#taA-WIXW9uw^cRMS0aYu1587H8`5r@&S!Sc_&X*&|EeYEU;zh2)anR zNbW{RPmF0nbIo1qs=7!?{e8zi47#y^a9U8`-v$$lS&2_b+RsdNIMp~Miz;scK#{Q< zYc&o|YS7KD^Z|iww?=0B!WR}k75$qk-&Kj0&l4QH#+DzZY6<%Yb%MiP0_XCY2$@9_ z{HAKY+2@)3S+{5Ge{t7;@vwh!tHWWMyfXl?=`x>}9 z6gB(R`^1*xM@OGrn(mH%0T=ZOToK23+1^I?uIE{z^-!Z^j7^u3TH#s;d}Yc;3#Gk# z+A771H49qB>uH~|i)(y>*P5QWYEG+9iOrV~Uj%@vSo83YZ8@LZ-D6(w4Bc(~spt=; zE`+LQs-lqFblr8`fQ#nS8!39i6%xEOm@3uU+uXId{Ap7klA<`%lYqN#2!3 zP=DlZw)W2&i_NOf(yvD?JA(rv5_f1OLZaUv*9Coomq5INb|Wm6?!wpl1`Tff5g$d5 zfQ`rc8JD6ZuhW-bz}Qc(a+K9a#=~!cwq02c{@dmzC65Nx^&3z{YD9-M%-nw+bd^V; z@U8jx=nYdty^Si1IuxyEExehLmD(j!|I~I#YpSp!FV&=mW^Ax9xftScy3c*l+pN^v zVnx>9b$v3=KlhlL>0suz@exi9{kbkb3RvrHQWnyshVFWfO0=`dD(I($?#_<(^qi5>~;MSMZ6s66uR=H&Zpvi*pv^-F0&CAi{)6>gUO&VLkVO0(B8qRmBg?;RoKHARdPHFGBsJ(|HP5(# zJFUo-5&NMC#NUb5c^Bnt#0WffOi0#gFZD-QWK7!?FpLjXYm3)89DUl!Y*?Bz!zXM# z)3vhuHXLjG6Mu{_t02bCJ={?0OIEl=6dlV6(RJE10Sm;Z)>AJS?(qPNLqu*)gmbx$ zC1K<#O_lIe79It&tL(96~RW0L@SC)hai^Z+Y>FK4cGZd#tOM4yc#0|$G{;T+nzbQ{;nq4fxFc7$ z8R?lrDWVk|=l}M}`5>lJp}qHBLP06AhOl6E7n{FWz3dEO+U@U;Ff%PkD_JXkR@ViB zBFqlqPc8ZxSg+A37;i?^S>?LQsj10Ddxcry8O<_xR<5;)r!!B0_z~u=-&2xEM^fM^ zpKo-Nd=BaTmRWbXrC=-QZao!d8G*YzyMb_A2h-O1Z26jGZ zSe#A)M{=b=@s-L6Dx`$KpI12KvPyFd(+<|8xh))do>*i@#3E}S4=(uuPSS0NC6|l| z$m^5S%`OmLsOBW#_hrPT%Wh@Qvk#Kt9lkqlNTOml7K5^B9O-}Q7gj3 zhzTkmNdmb*tdJwv=#HfvM{oiQ!5{;YjQ#UG@lgMtuN8P)Q+f*~*5CEC^(g}#4D;JF zbBHCc&!=9Wzln5E7%2T1TnvqX*~OoUf!QT2U1j5$kb&w4+9ktoD{H0+FEA(q$0Rc- z0-2?d$i;;wNpv4%7Y@v(+JQG2YOQp>5X z8Wle0*Znw^4uQGGw+@5(3StO=`CTEioGV$$LWk*kxoXworShFn`;X4W#{N&f7D}}n zW~sdjC5wV=U|-#3kRmwh{6~D?;c|+Ip)rr*sqIH$DN~jGe2WO-$56(>haVa??b~?y ziMQ4MvlF)xg}1wiHeoZ#s3q4iUArj?RSh?mlnl4JdKF{jQW7f4xfTe*-+dTE{XG@( zbyg<1Qt+Nk1T|n1r;>qUObM0L>`{dFZB~-C(2fRTK#cl!cf>kuDjD;DQ=+QjUdAQA zs@EHPWrKO;yALv3l;pZczKB|^eTgQrDXQx`-pS&9Xb?mGTke^oUEQn>GaJu^ovvLf zerC~E8KiwMUcE?iHtXW?=Mx!J`x1wgkFz zB)0aCm9fA4Kj^D?JC}PL^>f%283T#cdt5?o@oTQJ_YYU;Ay8ukzOc87FM93DTnP?L zhHPGCg-@CIa)_lV3qx-CG7Vdan1B6AmHCw~PR*khE)bZ;e@6?0PIuRwAH0>jHHMD7#fN z7${~xN|z(pN9;BKCW$*Yrcj)va{iz&{bemY1BM<9({v zdmI~fvaORmulX#5#8=#9=323WAb#zSr!sgc6Ms`{zIw_bjegG*>0v)Y5t6J}Dh}m~ieTWB2qp95TtGbj6kF?e)Y(=*@NWTaXoev}cWjD76z_<&Crx9_2yW z^bUSA*|>lkU#DXfi5p@hM_zs?p`;1DgdS{9%*BoL4#Retp3L40yQ-5wbL_xLuaIs-U1m)eZ($(b9 zDGk-+*qSuVF*m&u_EagZ+vq?1jtUU)%z|d{@XUaaBms42Q{9rY(3!uw(x_&;<6d77pn%cS&5AhqzDNCNSnbOs>t0lIGl`AzroIx@4Ho|G~POY)y(61L=NWJ=Dw zX51=%!4+#jEMtF#wkt(c2-w1B##0ea#f=Uz;(DHOCrTEU_aAI7EVeQeo$-@xhI__j z;PV2aX4RdpQRG32@tRKk!s0DNa<-Q+i9yNJvMjK4R4RMhBZdC3Ix8hyiOP<=)|n>Z zFj@M1Hi&%9c8svTx}5=KX9XYy4mi84AwzgO)Y4Jc5b6?KT`G0)QEE z%ZZ*%EMX>Ds%#^y{#_!6B~W;#|Bn;_CCZU`pa6f^kfK&-zW9-094CSLOITyz zX$MA?qf)tEAH{Zpp>m~$ON5ZgH)6t6(pc?K$o){|v~>T#G=5jtnimvh}rU#~u->DUlGx*lkefaW7~`pWM3UTro3+~(I(FLTBT zrhZf)jA5Lq&K_~^e^IEDP26UvGX22UUj4uWnAGiwu8w-H>J1UgO#_H@y|J1?J$hYC zE6Vy}66K3JHZ5r5RZl~N7qYr-M}aTJWf-=(kr}}oLq}yLJkKUi@3d>Q7Q; z6Wi6uRB~@kN=%-C$}H*`3~NHigR($wv!ycf@uojIf!sDL*8_g?GsYDSW2v%*6f@lG z5jxP;Lv|iw0`pu+GWL?$ch+D>o5i$_;ErvaniC_e9oDknNf(S8&B6Q-MfZEO4c5nT zE`3~p4CL3#^=LWRKclIz^$XNDlcuZDeM4wQNyYQxX^2T_$e+r`p z)0;~3eTdUt(H|kx7rMeNt#Q)pValVu6n<8ve7C@S(T14J>N#<;vMTuVP%||ervG#N zps;4Wdid>8`TkPOLHdY|;dnb8IgXg18bP}CoUDWlpZr7io(%0GO z@^H@|P#>n869^HurI&*#;7EPx+ynp{di}C-Tj@(%iQm5I9BN9Ijq7=K9U6!SV1qW2&ONBL~Rg=V${?vfs=!m zkF_`IK$|%w-ecYz$8Dz`!I;+uZad(YeQ$CV_xWLg@oVzrcv1S22}1liyARxZNHd$O zB?7v^gbnrUpTn<57sL6T)2LXCxfQjA}i&^z_B264&eC%f#0q}aSlj~eD zh`8^7BWaGW8RSA&xEJCwlc&Z`ug%w+LAA5c01`5t`W>s>qbDuM{(mv{EDK-QR}4 z0;`L6FDZ}jgD{XzXKa(7`FQiSitgkd*^n*jfXY*hQnE1EHcJBzK?_nj26u>69VaTC zdfer+Qn!tARaasQDP0k|S6Y3@!A*HTYf~7l>%;-oYpTJKh&PK3ftvH1j6e~`t$1`* zf0y^8U~8L-mmOrLcbWt2dJL%^m7WboV2d}*2N}>yuD3@Q40u3>B$$%mAdJ~u`B5|) z`w+PzDxiDj;xPq^C-4nX-T4-HWl=q3xbqnBa;5v^$=3X-nCdA9n4Ez+#yt5v+#bEr6VAXtMW#=Hr-YRw{K%DuJ z>#(8K9wV;QSyT0R=5o|WVO|LX2Y3+~GJTmw!p`_MCfm8HuPc#rHM#WCx3!@p?j(4)|)ZAYIXPER19C)X%RZNznDStw|gACl>CHd zsg76$DyQI^Yeo>MzKZJpkml4iGlkSh3iomJ&XxY!M^dC0(EC{@ddUV-*yLo?!MP;& zFF;R>VUp6%mcgO+IwK#75pIhU&RO&RET4P*QoLAuS^0|7CzpfkRYwm%A$JIVT6|hS z$JIk0DM%&|kKx0mLbCL}l7h;K74HvAbg3<==LiUE*tO9j9%Q$36EWYlIfF;nm!xv| zStT(@3n^xKo$pr@4lPZ3Iu$wM`o2)$DXOMmxwlOfS!-^taFkcEiH?r(vb6;J-X{qE zs9@8P@LkU)r|9FiGFTX3G^KI+rWh@6{7kUq;af15A;G2{sniz_@*VS}Y!0f8L;~iD zBgTeR$C~IJ>JCXv@Oi!!14!_XS#T&cRz!N}Ub50FQu&6qi-2La%)p6=vND$`t;2qI zW)!*YT~U-&JJBB8;R-F8U!i#>xUB2KcT(aigVKih_yrsv2ljzE-bgk^+dig&PQego z)0TiP6%$;09lDLlbn`GaO63pE{)URd7-{!vAmoaa`e(`NYm&_=9U;;K_KiLGV+@?4 z=`<3f;QV(+vsM9bkl6a!?Vb|!!zH^OiV~1MCPyl}DeX>Ee2!Z%SrQh#q|=m0vC>h10xA`K2ESnV;^EhHIl-My$0CE~ z>G(=#?GoeX8kAso@L*(+m@S`20}y;?7?{^M&_KnMCuh9Oe&Ew#_cUf#gyux*HdB3O zvU*`qURCqyLC`pL1p!1m(0o37PLYuQfvhxP+6jIH0OR#R-D8c;8LeWH@j;q zngN*xp}k*NkgK9U@RJ?cgn%8(T3zkPS20#4_$+HjPgc#Eb@qU)@ zEhY_IpF^9FhiiF)`NPlnD2`3!rKl8YE1Tj>J>puO&Krveo_AfUE8a#tBwMn(bDk;4 zY?%vWfSVtIx2Gkldr0Et)oXwMJR3Q7#6N8I-#5txzX08S`i+K3bMQZYAN1s1gpcm!%!8CFO6>Z|H*?(^Jcb>nigR7IOA6*gggg4EwbM51_Trw$x) zN>N97!Y4cEMh*>!||S5_7? zimBlTbnj2&CmN7$zbcpA02gsD(`b%ZIq>VMGjQs8xQG9+7wT+`tsjTFjP8Zi1ZrPU zsl?ZTISQHvSG=d^8(Z~FV`ue>77SYqCB87ygAz}s53*xN#*RkcL)&Hkl0GsBM1!DCTPQsdLYE=86S7WpV2WU9M(RN3U%Q-{c?YZ zpg69cl;?2LK@&|9jIKq;HXG2wf2+zWTv!c&`HLEB*gQ-VBk$wrLXvyTLDo`;ExS&d zeqDp{S%}$&1hiWbS|PO)RO>8mE(*5(OV|$xJXba zmlKz_^3lPN-`TM0&v-rO^&0q@0q+IMLYD%*NL42%ut`bD8TE*PGgBt=WBQ9CD{Sfvw2;>EQ82Q`8fzL~bU&D10h# zT=OB~rIAuW3kH#{X2YjvAgauXw;@8VFZKh^L^W)OG9owE*w8^W*d8s-Zn3 zIn1z(Lw{3MaYd`M(S4M2#-fElso6at!*caOtYk`(^u7))Aj<{5*8`*XOp#^a?eDwc zKHF9q1Vt&jvU$`;x+Q8__s)J2Q|(Nl1t?WdZMdNP&DLGyh-3%6q&+|QCq0}ogNg4E z7aMtGmaI)-guUwB@<1V6@0lZjncH0!`UMGy+6sl}Pww(W^Q5~Yb!HLl_=(v{8=lYP z$Z2q{B-+~tfgHDiMPd86(Pzpf@IbYS*Em;g=_c<$#L}qBbM86#`;mEmG1&IGi6!7O z`DNXZ^yf-G{`3}YrnNrmhg{M3f+3CXmATMk&9VmrHON6 z>hD{fM03_Ov#-mNv$E93t&=QxFY^0wxS$B#mHY3P$URnk9E)@rv636tzbR%TU*rl{ zCVs$2&xm~X23FLS`de3Xu1Xw!k&=?nNro0>2uF)S>Z;dxq5V8~u0Y0x;=6J*=3buA z$&`%>z7gPM^+xu8yJZx(s-1R@+xWvwB-~6dRZp>;bsHjE6JHcfucN4$a$q^A^*H>~ zv`RTnA0bW4`l7>h8p*y1^BJUhSay)N=!s${nl_wx+JGtv;jktpCnZ1CydgdRCNnBe<8s%u zfiLkqIw~PQZFpBt=`>R6W}L3L#H<0@`Fb9J=-MJtMwVFt;|+RPvc0ou>ZbVvVTZ7D ze(a?Eoo~qImw_jLhe9G2%~8|KCK`PBUpBuusQ+Jy>^=okX*QogAG_|J$77;#X{0e8Y5YO($LV=rv8axc2aV2$(7xx+ z*t6HK-u(JpXx|^d8=1ipKJ<^@AzFEgvrUyA6|$OV*2`p!+-I2b_YAW4HgzeGPGp0;%QfSW~1rMv3$&bf*^Mf1P04FvlzX>jsjyR823Qf zPuDM4qSaHE+mEs{OZ+CfyvJU#pWc6F{L;B$$KbaCjEusI_zjTkCYayyyX7JfYL~%u zGsssz|9W$C!)|78(rmX$PVzXmr}tudeDm?sb+GW)>G5_3mC?~`5q&~H+6knnklcXZ zFR^idgG9;3>%KQhyt;+=!(E^F$38#)aN~krnQV^1HC1 zo2Sod)ODoC%6C1tK-8uAZ{PjGmV-%lWYu_)B;?;pho5U76JP2udlvJ777xfIIk*EN zgl+kf!(#rgut^bnv<_hB3-wC`$b9FpE?nnUAhp((IfNy*5#qaCG!08 z&|O=Vc{I;L*ZoZ#)&lvk*Qof|nZigUf*36MwbLu;+&=rSl`m2@)P7$m=HP3;gxZH=lTOe!^rcBFjX7 zn9jfImHtaD=)8LWR`qM2n<2tkf*sQ1r}Q{oE=$+jl08kaqKeCD_JSWr!iB% zb4l^auym0;2=`edzxnv-hnv}(A5eaO*pOcFjmfw0z`xwO_X*Xv!Iy<8xccqlfd&vk z=u*~KJGOpv!+;;WvEsn`2h7*3biQ`k@o-}nmkm^W`=Vn~|fdDeKKmQO+?bCn=}$ z%31H^pPbW&DrcI+;tq@xC6r zc&B)rr{GSHxrk%c`3>IpVX(S?yXTYr(4E^JdYiM0!&V`7UWPg$wsu7rUE#fVb{6N{ z-<>CvdUUaPWxkS+66KWArECp$~F{yr! zl!Z7*LmK3Fx%Oi!za@mlH*Tr%*JT1+dEfTWdD;*7f+X}=6k`eQu&VM-AijR%ykuI= zr0cL~9f6E%7cZ$U3GscwBYyQJ?==zDTkL%cCS=ht+j zwQ<33*ACr(t~?lOC-RN=mrbg({$=-nC*0tjfxoR+c-wBBFJy21kaSb_-{LZecQ|Or zZ+I=bDb#c#@2~OBo}|<`u;QoV^Lthu((AfdrIiT7SBDhv0apl{-Ow9PC_;*4G%rMK0??Tx&9n^tCAEXf5~ z0(eqk6HRB?cOQJlx-~Mc;)1X0s`VSlU`iv}p&9oexU{A%YM=h#;pYNHZ=e(tjR- z%c&cSMzl6WYeN@>gZ;(D41G{eDE}2-AHpE@=Fl5buXQm)iI=2TP1Zb%i$496FZ$@Y z$kRVykw=xgAo8?+xTpwW;&`h$gZ@9S>ljS;x@TN26YT1Liv@!dJT@a19L0%u{O{C; z9+LEDrN--q`r~zjB>Ac3`XpW8)_~mHT9C(nhq5jD-Zl0;23o*yY#)03IOOK$ZgR7- zKrr6|z!2x>DM8wypDlUbD17kHi@rP|;eY!Jd&_$WdnbBGVPr0)(rHJpMWu~)Uah9e zIhj0ccMNoYGm#Z(Dy=MTEOF`zmDlFCf1gETQ6B&JeRR8E+4nM0<4HEp&0K&3c=50^ zfR`6$yr15`Wxksv`-{Yth&(4JOBGop3e+qU6bbrM3osB^Sp??cm6#cS6lq$2Oyi$@ z5~mIW-re~vHu$pFuU|iQuIJu^$DA;#tnAs5_pM8RHgN{v@~k-`5kFZk13 z=+iYGsEr5LOH7+H*G(_R0gYX_P8mrKshEo$BB~!N?YU#8jzdI*l^U@!Cae%v2rCC- zrOp#pZN*CYlG3)wD|;t2@u)r?HDRBbctku>lSlRLanS17MW;DW?GR@~Sws6-GcSx{ z>mILvct-zrytdFc@!ci1kWVqEP+<$%B;DWp_*lLp{3xFS*Gi04zHf0=&|&_oWb{|O zYngu(ACsVe38R|tKPN8%McXhDakmg&^0MY0r!p1(ALuXa9WTRAsMl|Ke_?;ccV2o9 zU{ZdXa@Xi*`y0a(yX;wdch>VOF-V1K>(wiN(do(H&*CC);Ib93m%N=)3C_!g?KKTT zY-JFqd#-`s@kHSjYX1Q z$B$UA5tNN(HJsgDd{s;hO8j`l52))wiB|7fx91n^4e6;c>Jo9FNZ=)U{j-eLA6rijWg(tT}yug?g8no`m|+LTzX&)LGP#P;UwKXmlQ-uE$+} zLZKdwB!wDkPGh0o;Z7l;zO+!Ax(am~btKeVLOnQ_P}gGQIH6DvWQBS#YJ!2IL}-O@ zS@yTIqEI#Km9Hve#uW3Mf^0*?L$xxAOWtV}d+!m-#svxb zpoKp!(Y7S$Q)$~)v_dc3(ls%bA(Mnw15lJRc-z_%e$FHZ`_f-UmbqM#IYLUDKPHu# znqxsNcxLXb)yNjHbI56yT}<(RRGsQ%hcM*^nkHExNRG(@ahx2J`JqaV2mRj31zB<1 z15|;WBx9YC?SVv@><&kXG8r8z9ckU}$wgUli-Tqp3s41;qHLNiu`Wn|NEKNYPOW2P zRyfG4P@|TSP2ofX%QqhUB4+Rhnn*Bvhe-1AEUPaZXNtk+b8;>0H_p7Dhf}ndS$Fm_ zy~KvR3R0?E5trmp*rmf|ORp6bsgW&V-_c{YMY7vUcY>O-JJCgU$%!Ci468$T6?Y#r z*ba7T{cWWYK}r2>mi=IVm;RFJpwbC;s)4WKPJ@A@(O}S`A#giw2Wm>Np@{^O#h`i! zJQ$xwgH<;cz=KN;g9qalscduH^U-Hffs5VNJKr&4Rg*s)L z+zOS81jhIrGOf6I0ct^V#lRR->I9N$@-fsZ(`0I>+)*2K&mq%_TO7=hq{jiPjgl#l zOq184Mwuq#!w%zr!#QMHeHVnROb@!0UV&tq%n-H8G&v(GW%?{@)2iDfOxd+*^U`b6 zXEQn+C|+cLID@j07vdlr3*23 z!ek8m3p5*-IuA>07TPPvf^KEG@PVPVPvVJxXeEop?qO)zV!;zts^JY1 z`;R}ntrY4#JQF0;DV~2QY-F|fC`ZH8yjuZ>V2#eq}61YI9994E3v~XF*vPOt8bGS$l4?Zt<>s0{1c?r zDkKBdG&Hl@U*OD~@q_REsDTEqgtZ@75-?P*%R1Wyz3{Z?a}o$ajBqCizy|pwS;m$v4y*$#+{5 zU8pPHhNkjOZjCDWHu`58ek*R;fO?P;aI7~{@V4f_P*=fCT@{=>9F+=goJqmewsJr{ zNGiC0(P{;+!P&8w43ekgJQ_*fj)S}%wF@PAJWeE$Z?C<@*T9=X=VNnddpos{kzu2u zm(?}PtWg?&WX;%pZm%p8MrB9V=5i3p{1--7Lz!)*u|iFmo#-L6WTB{%*}?E6GOKv@ zih-Qt4Qok1nQf&3Ld~m}yL!khSsf}zv%S%ONpx2A+!MW#(obpI=~YluX$>uu zmRt$dN;?>zLTOdc9089lskDQ!hDzH?Pl1|BYwDr22t9afAcx+jvHS|&1 z8utF5lq=fzKZTajF8@Kh{732;?dhM&_5_dU;|O6rhoA5ug96Mj$p&vHH%;vw0(hB(J%7YKX&qN`AfLFiiTzDn5MEtMe7J+{VJc!J?_70fV0~*`{Q?Gl=irPP6=b4 z)-qnnTaUm?B1(uAcK(&)EKX~;)>e*74aeW#8R}?p-FV_F@6T$ZTjK&|myK_TiB+s=znb<^0uRxPY8l~NBX%=g^TN-0pyvO&E|(^^du0sv%blo&dxn4wwgG@ai>xHT+D-Q65B{VjZ(*zrKC-=KWh1u3*4tejgPg z$Nh#K1>W+!qseM_Djj2vxE>*{0&z*d;bW&tWCd&yFh77u4I&H#U|iG^coDm zOc^8b!5@CU0chH{08?C`Omysz-_0xNSn#zFAvndpV)xLuZ(VmGqzpcEwM4+-dlJ!u zIR4}JZl|AhEw5DW$}m`bG%OK=cklZcqU=eD&%dyrMRV<^!=*7;sQWSwLQ}?eG95h& z3?1hdGHd*B^6#jxPA$}JCA=k4^_4R5qGe!qV&&#AjTIOy%+2!{^aak z_T-?5Fo%nwdivrZ5tk$AVl2cW!P7v~Vl!qTk(_j3qB22A;IkQjhM5Q52Su~6PR6A- zSY)h};#t*7-FG*ukQYQW8pB4*!|mDDBjU5MXzIdz)(cSib_MD4Q*f_uaPAF>du7~v z7a+&%`O=566jEx zn1!zUn=3`GY*BL8Hry7t9CykWcOy$|cUe%Pe0lR=j%jj?AnheRlKpDge-*Ll9{MM) znTc)xQOCGq-i%>nGmLEZ;EDym(4E700u}K$Q0*0jw4biCT`|#YEP^nECttv0=P~`j z8E}h>W&KWnFAG47CpRpU=y%z7i|>p|zqFRkr+xeoStQlI)Yg+;Yb#XXrS_brtNF?o z_%yrF^9%!{qj1u>4r$(z_}~d?PK^l#Z5E0gSrMqKYhj@+!;lhar13fD615~ID0Cwm zrxQgNFuq3Y4gLD@XZXR)=LLMh#9TGiW?o}yGm|`jUt%|=(A_W!ae&L?8I^VRsD;xZ zb=oh*_6P4w{O>q29xUM_<13KX1lu5&+vNt`fq<> z?>@hOV$W{getz|w{rKi%;>9}7@cX0{ZeJmXDE4`&_VK%|zE~bUlr?o@o8JiM1Ry3{ zl=Gdy;`eUA;LcfEK0K^%0)SB3{hkTfEhO{2B)1^MctrT{jsfHo$alA6^cchko8nvHB_c$i4Y|GZF2&D} zXscV`=l|zl|Bgx`wrnY9S262yOy6Y%vts-POBBu8wEb40WArpS#Ik)EHK5w2ZebdK zfDUS?&OIj!8#S+pLG9j)V&V3Np{2lFurMZ*E`)aK?>b`oaKLPV^D^^yLH#U!+~=8o zv2y*BAk5spgGO9A$QnO@sdH*=;op;4NmAeFK646a26yHOYsjvNML9`iW)g#^7W+l@ zFsvT7T8cv^%wFO-L+(8*G%qMkyb2$Gw{6{0J`jugW0LaH)g9jJ_>wvNO_-8-bAzrj zn?2kIxY}PHnaQfnJ@j1RzY224mC@*i;swODv(qBbl zhMH{`rC(@$h04Iwj@Z2)^~k-plX*2*sR4%9Z(%8vR~dJZvlwp#;p`dBDQp0saLqbTqidb zi4WkfcfU=2fNE4+Aorr=6Vo%Mv#MQ2D8D#M>T*NNeHg6n?(sYxuk<8;B0B727~I-i z(J{+PT3enlr$jcUV-FsfaF(DGEB__eUWmXrwQb$GTx}y<%tI}LzTwTI!;hh+X#kS0 zZUr;z&RN#W){Zd*u6}&;4rqXKKkPrhD()fg-_fo%Ah#I;X?0(*kA*0N=zGE5^H3wE zmoWY5`eb-$!!SDoHW+MwLo<<#{n+>Pm%ZH8c!y^wfG~T3pNu89ai20(oz3;c@1sqn zjE>#Sd7RXs1fKY;VM$5@cr8&=UuyEb{RR70v8zvzQ~{EWCEy>wr(3@Szb3)2yxp3| zy)74mS$$rn3V>MaM8Km67(_8m9EpiEaeRzTnkycS%S}S|k&wfGItfWLLYfh_i-$BN zq$y#oY@=DA%2?BOuGc?cuGgoz-f?Uz9*iiFSol}IrJfvo-Uun084?;>Ix~C^jrJmG zwMR##u-RyiCS)1JR{35oaCyRG@~qMXlO~vlv-60qM=sWj7*F7P%_dbwpSE+xeqVmh*l+4- z5Ow)S9$LaD%`j^sEV*jW>KYotmv?&Ps-B^@M zueo;J+CFx(YfwQ@yfiDlt1^J(qqTB{lmH{SDk=i-+U$&JQB(#e5a!Z2;k8`jU3a!K z_ZD7y1-Lb{r|~YXBzF|>KI`7bmgBwc0XLuyu9HhDoF#mxSl~;f0WaI) zwFTWhfGHFNGz7`?z6vZ{qToYd2=3>|z+C=h2nU7?F_AM-Zo_BS0JT5YIVq{E(iih#QexbUx z&UwWOJ(d{yh@pGdDjVaSA>&@h#6QN_9|jwLo1Gs1c620d3yt)crH`lpp=4L%6k_EC zd-q5?>R!4W?|*d@zVxy)uCtgFv8MCI(#~K9)3YB z>^;fD7q^ID>8J!Y6A(+t~g7v!B zJ%b2F1Un(Yj8+mX>sdrFbqSU=lIjFAC2n@W7q^ID#&ypmf)T+^NHDXV1WS7s5lmr% zrH!OI!3>EMUc6?MuohTgup>^hT1ppZrN}N4TOfBo>@K7KxSDr z(a;coYz=52ckC14b_5)rxRroxuu9orc_Ub&5iDo~Ya?8gKuPg-9WzB9q3Ks>B5E9VpJw0e%4F)FsIUK{Z=xW>@!JiWY|2~z0+8J49gP4)Cw7_xAZ1VS zJO)q?o>?A1ysFxmm!AMc0FpZ+q!WPXTu^fWr0Q9rEdasO9s(eLtfc1nyzxcM{l;F~ zi!XT6?(O2vhY(ZK>CFpU3|o8xD75eWbLBaO(1yBm<9(1+_4ouO%{|}P@Rw#=Mevcy z`*6BdR4H(5H0Dej8_2PK7ydLSJ@~WV#lH_s{BziA*EuE=Dn;Jr4R30|468n{ox$i$ zU~8S_yGZ~dk9U)Q0Mr&dD_(K-&Ch?~!ChuOsVkfo0 z*YCE9KWDIc6MtG~_-+z^$l={2{xrqo-Gk>pgg@}MR`KTyE^p#b>-^qL;tv_To5Y`{ zSiF0M>5mM5;B9T<&pAxq#GlsLy&J?I@^?3hKh1G?7jn#h$?yl>)++v-!Q)N*X`S1< zN&F#uca!+j7=t%F?IHZZN@~(oqdWX-_1y^Z;XHRhd`vD6a=d8i+iSLjTJa*((r1g+ z9QztZF6io?uKx6=Zl?It1>3PR%N@9rexbe%jaaAE0++Zs^i z`IElnH#FdXKkwXy3!!m+7kqWY#T@Ub+02;lEwZ~EdL8!m8wAo6OTB^!uou}y5cpXF z-L7-ZzawJ*_vep5-3VU|pYrk7U;h5~!>bQBZ##46G(NP`+}iNz-h~?Ze%Kj&f5kp* z-pP=J1f~#Z_g*~FS1t5(9e?Q#Fx&>gsdC|PB4vrhyrJIw=n-{cKu1$ z2-e2rt1sw z{zp>yGb%jZo_txq&zTi|IGji;Jg=lCDf|w<4k^5@3U4Z{@EDSb!v91He@=xTcP{Lx zIk&=(d-4j8mDK8#iOv$f72|)AcIZX00CBXVU}5|In%T}jR$QctCczDZXEWWQ!CUr! z%)Xxf*?E2h!|VviA~^L7o|)GV4|HCkKGxaugPb7fuS03!uX01IpW;hu=cLj5ey1Zf56mnDV`^A3dj!z40dzdBVTyEYPDTtD z0ntQ1L0rnYa4b%f2TN&+PrqkgMqW{WtxU^E@FaLD<*{ul#wlenZYtvnrsTK8xN|bD7B|Cjp5#k04jz4f#$DXdaH;f? z#5iJ{QV#2;GOpmsen*TuC*%5^(WO1vmtq_|`u>c=dTM0~OJW=`PAQLhQyEubaHao9 zj5{ymdY5*kUygA-8OHV6nI@AMM~qX-W#3T76@B?X5#!FwIP=oJ{L3-Ulwq9N&QzSl sIAWYq=M0+4xWdVVpNVnjWnA~tClfBmxULN2FiFt=2dg`U0Ms#n!T< znkzaFd@2vTfu9serdt~Gn}K@;G+qB((-$T2n|yjf3-f-h#n(~sF`+0Z>4g$gka`g9 z7)0h3Zlhp({tYJCjcDUNz>AgjJ$;jsam>bTX{bNjm$Pi;0^Fkb!_C$KcET5&t;Q89 zA`|uM-wr~NM{PG_fyv{OywGSGe9(^!;J9^!>WluxjienoIzXLaACnDj^vHLeeZepD zVV`02ECi!(_Xmw^YYw|FB5G=;e;R>%L(@&HRL>uVroiF%A>0g@*%3k=mk6CT>*?3j z9}n5@t}B=^hKx^lJwhr=T_DTfUOn8!%`~4qwh_ziB$SAfQI)sL3_CCHm^&c>g|m?~ z-*-$6*fB&P^bsmo72bYVs3s*Vrt)}mprYjjhM>$+CgwMZ;P>lS zKWfJ9_9$`x&*jiK-;LM_^;8QShmbm$a^tzDM-TPO28GyT*tf{7;Cmo;CWJOe>=DVOh*4dKP1XL#9|!FrbjQYf5Yv2UtwuLGowOAg!~<0|9*?CZi_LA223RV ztGl$-TW%J*&;<&zo4CkLR_vJL(E8}G1^ot`v0bon=(}|JS2<$xF}s99MbTXxLnQ&B z8IQP}dS|haruJKqzNM9+$|bH$g2BUM8u<=wD3X7$^j#xE{DDba1m9nje{9@xd`-6~ zg{ljH-Qpw2O{nAj;pC6mQXDq~h!t&ILkvN*5*5WWR@yPQ)o8gkVCB#Zw%G!KHE*f- zCl-6!#H%9r+Y~GVv3+^#42GakX|Z-JADTz!k~RiL+PygWqWS*(!@~u)5vYKke)dnrne&%Wit(1uo_QFDmWpb z(v3%D_Ams6{1yUZ4Yu$^oSW;I?OP^2BbXVAOYaBeGgG%^E~92wWr6W|= zyX(y0=7vTmbDbvG^k2&N9IhpL=Cl1Xn62nBVHO@%D zm)KH+p8wtLxR4BO(cL@n^P4%{9dT9D~DQ% z4;D(=CC)n4OI?{FhTtoZcS9K4QpJHZJW?KlpN9I5=&RTDWu0ZH=gLD! zI1q79j0+;TM$_2t^{X?)jtJ_pF#;F`q=1qXvV5z!)0B`Is4_W+HC4Wc!u)Axb};u6 z=T*ovHW^N;DqWjTAC3A&FH4eV(2$OK$a6A;%ZwgT`lGyG6JfkqXW#CE8UFp zPr%3{H4aV!{p6EZD0<|^lFB=2rDJ9a5G(2|@GVF~GTE z%GYb1s+d%nj>rq>Db|gr+MJQtepL4H^AO2A@9;VbmUT)CmU&nDdzy*o`TyH8a`H!H zw4?C2Er8LWDdpytKX;c*UjlL$nQsvhF8|$>Xe&#CPEnKetSC#tQ?OyF9Lz5re7qP` z>}Z1>Hb8lF6&8jMeaHvN)tw2MN(%g>_7mG}9Fq|SQ8F0x%TTDk4t$aJvW;jBabdwY zSpYH#5)`1^i^8@hIKItmY+6bjGKEcvw+F0Y{C>2T=g)v@FzcIQ} z+a)x-&zIs;W-2YU)iSKtr0`<4MfpTgy>=vrO{uB=?~SaN5UJ}>rRu1Y$@xLDo7z>A z*xRE+|_| zjumTokO;LZ#wlC+mr|CSoWKT`b{pf z;)Jn9n!~%0gQPl3UXKsjFyMjynq zk>*$dNK;C;6vzjqkcQkbND3|bnh5Q9Fag4i`+%G9j8{|o+c~enf`G%mQlxZwXe(*i zB7bC9N)f=9Hhe9>o<%`LH0w6aKXY0n2EtDzv<~n7E-F*L7x@M7<(oo9s6Q$xk}2dh zDHbZ|dg0jpjm(NYdRDEMtXemHM|Uk<1-Ow0RvCH-FR>pk*xnxy&Jn~m(xte7BTkns z{ksfBj3>WXXz?lAxmaI7*BM%)_{3zod952+=Fyt_D^8 z+u50(-dyrB(Ot11uV`U=IHj>#4cTcRL$2Nny+o(gVx`fgBJPY~T>bkTgK zDeH;~3sX%b5N*a{{(jbq>!eV)!$`9Dr4-0c^+@zmads*lLROT*} zv%eY07re8ng;vZa8K$R+hqIE-94)@W@#$EMlNwSlKB$b#qZQdl3CT<|qlvx?#N=sl zJz#0Epv9#$h4&OeA)k`<6|(584tN#u#@yNNj$upA=gH5+@>*4g!339~QeS`ZOcm=T zADHaR=-$>1d#w`fgB3*DpVV1jAsx*9b<0s3DzHZ6_D(z3!OBGR|MTTYGhJZ&~wZbFAr!~gA+@~VmV(A4Ev z4P?z?20{rCh8^e!D?Er&#i|K3rlx6=-?pS;*P%cOgRr097=q8GJ}OoLx40@r>ZWi1 zz$KgagT?)Ri9_{-_5b?)_@uf}dDyYH#mJHO9Y!b)zk}HrW96@oBd>3&3B@3kCECU( z#7ad+jT@EW8i2Lfy*7~Rnjr(LIlo$k#ESNzw?AD}Q?}WG##oS$?wA?-7m%^ow4@qxUnU%$YHOu3*FR2gCr~WX?8nEeE|%0J z^T6)g>>hiGcC$V59cVz7c;{v%y#>tMUF3axz5dhRPc~c!+x56CZ1Hz)lq+wu9umIg z!EQqDQIW0aA5J6_a67P(f4uEjSBubQSX?G3ci8IaS*qM#3Tbb@s{9wm#GLO~^~!1o@d zS`Q}=DV9e~pgRhb-4gJ{i#1I9v_2PRJHKAx%&^3-I+m6dVw4HgcP7K=(do%$R?CkZ zin#hi8-48fL4h{Llq0t~hVx@KnfGLvn1{>XUG`EVsg}EEg~6A}dP7n4K))~}I}gRX zlt!*ESu#orq4a-DYf(639ZltCQ>>7 zJ6qP}THgetZiBcePBko{#Kyz#k!*v>K~BDA|M|z%sF#qN8WIhIUsecKfN?u{odAD( z(|m3aOye3#bf0z`>WqfER3< zS>?k6lkw<2q}}6#%Gh$6+S-^%Yiz@xZtre+3b;l)bN)^0j(N~Uuu)pc%OyICoOlM! zoMubsRu8+Ov>I)}68f^rhtrbCe_UdBJ4U%phm*f;-p{prJ2&^ZN;x*>ASHi=Nz9ay zTNqz_h;QWhs1r2oa?Q*(_UmV77MkH?JKSRePm1s@R@>N4BwPBo!qlI%!|@y1f}wG5 z>hjx~69}g)^7seH>6e-E*Zm(O3MR%QYwsI#;7HLYwzvQK&1&f^AYBtR_^cCTKLr0i zlL`4Z6~D(HSi52+)8K3bM=l-SyD4C7gYk#OIHHy{4$bkxnSr51W;6vWoz$Kd%Gelv)dOGVQE=N#PzvkT)d-rz+etr#UvT{UUN2fUeqz^OFRkd!k z{5}fM6zTC?d>~1|cQlDpW4O_ZQjZ4Sx3kY$JCF=3tH){w;Rq!{azyhq+vxsfC|2y9 z>*+Ms!2YroG~33eBF?b|W?*qn&3Uc_L~>3y+_AMhM_U4r%;uShYUnCoK!MuvHgyK_ zjq&J{QLG3;x-S8$;V+Ct2MO~fb|{F^_&`~Y2U!(c>ro~+R_hE?!7vg7)&@$r}7*X({kz+v*I8ims0vnYXOI_o{4G-*Jtf~;SxG^(Z^2-W9Ge@+65{-cv zMP0ls#vTzibblGi(oa(oZY+(SP~~1}MddO`C`B(!M5ngOANckC%MLp+*lvz6Q2t}c zhOQ5g-72XAB8s^9dyt-e_*YS8tIB>}PcClQVbh~>$sF?Hw0)}+qwyt<3+X>tKk0Rs zPzkXv6dtw9+@VeP#Im{jxA53NL8J7s?ty&-qyxrj0+Y?h(cmB_^L;B#=D!Us4+muQ zD~BDKzqJTe=bM8J6-LdDSVwPR`IS}YjQ=2J2YG}`bZ0LxnKK<6Gmmk(r>yd0HT|jM zAF>7r=MS~6$2y|5QP|0D*wEOexgS>hSgfW_6I^tZyV`r2Y~|tep9u^D0r>ICx5Rn0 zSM2%;U-)_a5a8O5hXEc50JrX>yyqIRE%uX)qDuNN`zP~_A>-scVVyTZ7OeFge7ktu z4u0b}%F6)ctNR!uw|ngW1`3mZtQqRuKkXDurdtaM7C{~3_n6kdRno%O)o`qK=dMqW zBQ3Ke;#m6oHxi2n>fONE1u&yB96IY}FQ*}gNuR-eA$N-J)zzkaR{YayEruB*#2P;P zHf-Tx-)5=y+&gH@e3KZFJ6nr5T2QgBb`p}#kh69s8%W7 zz#3l(UMKWATIs(XJe{=&?@(xx9w3y}FexJ59EB}$gSdfgnEk+NNIBp7T+EM9UiwY^ z(qX73TCDkt-@4~_xg8AF1j<#5gbp8qex$I3Qi@#20mu!uVW8KP4Rg2!CY5yb>VJ_% zg{uj^nvMwvC%F?D3*yzP_J2npXU6ZJppNOy9E_h1nAfPgfW?nDU|%WzyQG~*n`<@z zQ^-*R+gS=}fxZL|JyE@e-n)9@P#8PYUL&4guK0RDfp(QA%x&XKllpB6ns)WD4YOi970G1uRzLW%9xi5-Zg2<|A+CYF_Xmz>Y6a`mqcs;<9P=!! z=Vgr;s;N>9{?jWL`XNi+jjVx`pc-Hm5DVnsXLKiKQk8YL>k)+=%4iDHgKqM0^K{JX zF&)-7hy`QgGaK+Q-tr5AYJa{EAFuAqz{AsmN!YMYabatoxT5Qb;Mo|W3d{MoK8zWZ zpV7#d+th&rD&hI~<=4hQ?G|C3qBuNZQhEN zEF=kLpa#hJi}WKyt50614^xA5qRzF#phu4JvFn{-VBSYo3e%qdH#FP7!z+Vs3d2{T z9F1t?g5{`)=6j(YUr8)u98m6liUB-rY;IS7Ch5a(+=_%D3fLg$gIns!pRTD-NF6Xu zY&0M_adKN_&71};HrL=0?WG({9k))AO4l`zKRc2JZMNi(^2SINLp?~ zxfejpxK^b?i*b<7Zdp@=6lq5ChxU?}Mii(TY?88F+SL^tHHm=J6zbUBldVoQMubaV z-K7Bk#{&`vmkAv<77YjA$BM~oOm1e(eK##~L*nJQQ(7$p!7wi@>E(|Nwij0j$1Nn_ zA=bt&?NA|ove=X<_aB?{*=cA~yf@Dfr>+;$v#}c$#xBocNtOT2g^+rcb3)`?nHn&s zZBn}NF!Umb?Y7ARw+ps_iqsD?v2xiF0!Dt`+}_}EWDJcJIlF;fK>1(b&Mp3?%97#= zq^diyj;VlLqH8hET}oDaY9>6`OhWPlk}tGBJlLT2CE``~&-m?6^kl?~aPl!&A!b1r z#OA|fLHaj2`YiT3$BizhM{G!($^tM?!)q<41Y-i&g*26`vN?V7F`g{U>KrL;pGibY zGQWt{o*2V!w=YMA;&>-?MS!DA2kzVq4Mr)F3+ltq?0Z1~y4XH<(&zc6e}izJ0CuwR z49&|(6@U{!=E{t9`z|*!xAEIFe{!_@qo$g7X*Y9ZXPMvsKHr%04Z=^}1rO*D=7q4# zcv`yC91nW$A}8%%t48}qq~xF)*t?4J9`I%IcYnwnEb-4vW)TBV{!=0yY6l~s#juH4ylaWF5;j@tz0GLDl$z%kV;yHF|Y0qrPNHjAEfa%CLvj zi?5ub2Y!;eVIG~J>4N@Bx_MAI6|vevT^0bto{5zXL+AKt*nT22G-c<#kiQgv@p-0< zDzT7eybO3a4=_7s`oDOoQAKlO`Ny4NN8qPIq9GOfR-6hU?zXnJRK!dJIX6HFkh4+L z|A;(5?>E`UhyS^kD+J>5c!E>Qo=_v81M;kOH~1r|`VBJzf8G@}b z&=PLNo~evIpf=RJNjW8E({MDG6XLd(6Lu!I;ZYQIGQcb^T;2rgu)~wgP?^#Lih9L5 z20+*?5EO}cK*seF3i4b#WpoELROjJ+)C;kQsfr$<5uFnE{eVhY4xSV(7{k!fzRGY2 z@cML_IZjmf-Y6;x4s@U@p5{u84e%`8I)&#1VT9d{O+el zs{nf5%7vz;9=&%xp2|U9Mpx(0;|OjuIO&eP1_4{BEnt{GUBf`RI^bxYB(Oyfw&BFo zwx8jg4-}9F;#&j3H&Z|^^1BX9MLS3~i81t}X3jXplKx$t)k$+(2F!e$k)K z5pw=Jb6|tbQ;-bx?uK56XcKpCt(MP5Z6-@6MdJrfO1g>;*Qsq=HqkvKE+et%uI`69 z_~lgNfI_<83ZzBmE=lG_4D46WKBIODi~9LA&h+A@^+wmi7cLs7MKI$l_fYaU;_ zIv%C(;arYoR0A+++{*5$88IOw=S>{@yBBT&I^C_!eWA*8+OEg5q>lr#45Orpt{s(7 z2gX*);o#myfixN63DU;sw}VB{&4W1vNsb>$g3_Y=9X}I{S?$?Fek;m}gXYe>ADa&| z2Rw~)3TjkqW59twjo4@gOd-zp!V|mjaa|V@c3KxZJ}SSTF^d+EipNM|yc;`ZusH>c zKv;laU{*rczinpE`6`4bSDq;XH@AIlb@o=LO}k3`T(|#OFQmHOvkJMN@B29d*zm~=vo)ISqJCh6Y@782J_H(V zsNZd01)T7=4hu1Do|-#<6~nAwB+`5pnaIZWF~t(qc`6h?SdKT=p=Zd00fws(=fZ84 zFZbPA%Wd`0Jd^AijtKMvEclP!|ITV}I7WTo7C-YGhqqUOSiF7zd8r`}v$$fBxg@Tx zwyi$PCL?1&4f*s;fuyqWK8V2s2Fa!r6`F#m{~aZ3)gXR^kL#1(=%hMYBbWA#-%SBA zv{ir)1T8qY1l%zGX=lRK2GD`67%Qv8EF*7vB8lfyktq$j(*>0;4xp=_7?0mL`!!+B z`*45Mq!Usi>b0>1&8P)vD7GM(K;sThYlSLn%Ma~!L~4j6SGmiXkJ!2aGs`&aBZ}A5 zzx0<6@k0au&zs12IF#WppKR0&nY%LM?FE%{(w!@4F^JjI}Pm2=sr7ZIaiR?_PL%1KHUuD*%)Q zCf{YLgu;aT=T`^5oX0-E*$iEzS)FOgf6)f_geWYzzNYEiQ=Yi~a2q4l^51+)L#T1l zvp7$EbUqcReQF*Z&cc?2eO>e=VhFEyb;{Zt*oKfdiGcdIo5**HvPHo^X=YT}0j-U0F@yREW$StYZWtD0Oq%%thEX=J4Xep* zYt&iY#%{B4T)}0rR)EB59dzMeDzlzgZQYWDe|xWzJIjQczPA=eoAF#aacO*FrFh*s zoE}WK*WX*z$NH5(f13rNpXaBTrS+UTe?ae0d$Ipxue)+vf#RAVco2lzi_A`Q{S9uu z=1WN~y}Pnt7xZiCc^CE0ea^Dr1?g0TjCdPiwV^f_7 z8sfepMSu|heQn8o($wj>Sz9x0L8@!%eWmo9 z;+6Pg?App0cz2H9KWD(A*NK!T1`3jj4#J(iWihCs`L(83wTxS-NB$b6pW(o}S*XD< z5LwFMZgLWEq#e_kGD|3I_AW}hl0g;tb%*|j(?%X>DDEaDJRJ+fKD70}dh{t?dv-d5y!{z9dwq4Hl zpwldlcTjj;I$fc?ppKx@phcMDIA|0y3-x*TEj@(hIa#o61p7O};C!X)Wa`G6`haFGy$SW4wP`WelO_%M+1Ce z@jLOTC+2=n@*;g^G58xT2s3_3Q!V!3(_BcFIwE10%!bofNR?6YLz7%S98L2!VQsTu z<}sJ1j3`A`bQT=JLWd&{OcD|T?}(Bvv6bV=_;cIw+Sd_`AKh_6n9gzi?X@$%-TJKR zNG)*)z-NIB`P=HoC)8dJJ8UmteMqM7x#%vhKkuVwm;4ClbO{}HvE0wMVy$v5)AaHa zV|_f6N5S*N0s3CrVPPRgZ*BkyjDgq07jF+yW(u<#$X!2PYUGV+oCjaoIL_VN{SH>LcqP9&yt@)PQP`@w1c{ zL2Qpfd7_t0t(awlK=*S?&%V2J`e&a{bG@3zcU$fKU?`fUw~z4~0erBgfyTBC0{lH9 zoZNfX-PHsJBfZ9GJo;`4>*)v_Sd98}FZDSFnf;;MZ?_8gO3hX5m41xB!Kq zKSR=Tlk>uTarH&w?>@qm#cW;|;UhC$EbNxmZZabYpoL}AY%-XW3vw0j-Czl zaowSSBXz?x1EX*bimuQa+ZbBRSicj@UgeBmEl;=&q_MC8n`|Qdg~-RxKjy@7)h1eU zx$xI9Itl$w#DP^Oa=F+BQ2@pu_qY~kKX9m(oPBH#^VCzE4J;@ji~EkFEFQZ|xiN*e zwgT+wY&4X;;d3n@$K<35cgKC_cCGYM2!wGRpp z{ai&sYpL|^DHfGW-E7|5hJ|>CemV==*so)V8*a#ec-G^HFMF|D@gJhlb0qafFu(p+ zyJk%-r{%;B9W5S82!3>Sm=D01=jV1locd(vS41lCW70GBW72o-OA9hJ+Z{v=Yht4T zX8MNQS2yr!I(VozNLYBmH$e!6X||-d)Ud*)`O+#_Hr3nt)N%xWd)=Prox?0<aQO?0|S z*en35WQHPhQ%%TyqH+JGfEqKFY&l@t?4O|7%TRdQ$NK9dJ>*hs%*#yhes}E&6b-?w z(Hra=s+q-E-bv>vfF&T~n{GJ0{X>}rlDs_lHrX7SD$cNDaJFMte+egXs@eCn)r-z2 zdF3q7VLW5B4ypqhzkfqV?ZMiH%RPL-X|xAC;D{`Rq7lv9BVs5E=k1_~6f-5*4EdsH;T5&+?9=Ng9mr7cC>V1`8^2) z=W|nkmhd(1UN^x`aLJ++wqTRr4Vg*iVsSjJQt+4-s(Ew^yTdkbAJ+6yW{2~53%}-f zCia!x2>D6HpDu}`l7glLXIB8t-!H&{ByZ&oee%N&L!PulKI#+*X-3k+0$h|_dX9A2 zjPw@8S`&i5I2Cg#>R?JR)5l;V={5kh)E%yz9VlSqPPCF*VQ*8`J}pTr`V1Du7|||& z2SyVL?{+8Mc5W{Ij#M-CJRqgB52|Q~jO?=KNhU4c(l4kHU8IsP-N!JE$w(t2)Pfb1 zGCyPW5K}%1*`1e+wnWJIBfQlU#TLVJr?PMSbg@&6re{=E7ZRiMzZCSSPl{(!l zOxuxrd|~i6@rz72|LcKL#yDlF^F<;l%xr{CGhMBiVD&%sjmDoHx;B_U`ro^f=Za>g z)ZIH~P_;}&OpQY2D#yFfG#1c$R4*?9gt#AUy-2$i9!x@I2AKwJ6QYj^^WKfNe8P;nrQ>P}OUrae zR#VDzq`_eU!B}+H6<{nW3nGgj?m2r-?3<9XF;az4EJ9+UpFUDPb_7gO)#l`I!Oj&GHA??vJMUY8m$FiigP^l4V0hN)LD%bwZ3YL$;sj|3LC403VmdjsWR+^< zRl3@+QJt^c93nZ-J zQ%{wcRb1i7UX+#_vbcV@=+QRIY%pLlq38pz!BoPp0rJU;{1U>YH;01#y^42382Eck z2}f{>SHM>gq^l$YDeM%x?1+*}d2Q5o(qG{|5T5qLa6QSRXKOVSyZNE5tKc9tZX%W+ z&2yb@TRmdXK}PvYY$C0g1q?~h+cXVt?tpeO?+P>C2`8hrGy!aar^Lcv$qGkc@@75r zm#YdhI#9EJBKg_!x9)1XMi|9x4=ZLthJf_M9KcsDlc9{+a<(b(>}Ukk(4h0j8by=L ztQl2p`q>XD8%dmJTNm(L9aaWL#k zUZvSsBaeZ*h*N?_XP&4ZqexwJI>Yod_{L3`K%BPF z_3Zt$gTV2w>j$jq6pnoQbOG>ZN_>TI=Wxui9wY)Guag@*Z@jF@WU2R?15A{J{VB04 z_ga*?i=JJp$zu{lAW^4KjLR?lzW!=F{On)JB3MIqxdpuTT7O_b@Inf;PGv*6K;PZ%!=3P zADR)DbsY(7#|y<29fd8x&CX3>R2UZKsS9!DdVly;lLFGcQU*Ztw%NfRR31qOwZwsDHGxF8Zf@11s|1~Y8#-ofkZuz)ybLhII zY;AMBzTB=9GMVn=juAquzdQRmPWH#O*;>ViN`bfF-zlfXC=$|CG$BV%>vA2qeq$l< zD|7xNAz0#wXuMjR!-wKR$|z?TYr2$@iA^z#t^P4dUyO=`$dsAabnOSL7^qOgUh3pC zQ;T{4g?13X92=@Qf*w0+d1$=`n-EH^%qLjEpSSk-?6yhK#fI=_huaL7(Xa4klg0c` zV!1|N0tl7M>(t`qzBB`+A$@!Yc5e6->aQrMl$G%QIwhw|z_O_KO_j%Fq=iBxMy*#4!(OaN``2 zJe~nzlncdASEQJ-EMxV#%d);o>o)6qM0Y&!y(8cW#>mgs%G(dCW%9Ut(GeC)HbV5f z5*nj5c){OpNAyI00X8>pYRQ^76tjAc_;zm6_W_wGwTlew2;JO|EUt!6W;dV*FJr7O z2JVYPC12S0M0^I{iou#ux~2Y8#+0rOZep)Pu|c-g#=mBw2te>-7Lp>q^8eYoVZCRf z#bKkp;)~QPcasf@Z)e?c@}Shh;11LNoI;kaamtm3lgsQaT<9 zOc=odWwZ^rkpKT1)mv8g74*Xxti{INFr^V+(#u80U_{_uS(pD;9cAmxsts;F`FMoa ze{9lykb^1~n8_z=L0W%e7BS&Ox;nAEYZ05G2~3zEA9NH?<3Oy70T*w(Xw0;FRcyxc z`xT20dW^Ijf0gzM?uh^ASR?t!%O@0H^&I(uLFh!D;ko+2>X%hw_6G{()UA!=B6ybr zi2uk4KV~Lo3$sH=s2}Vb+CAl8`H6mpiFLWHJm(4MYzHNYYwiH5i#4Taf;IAIl~uAU zOW**-ew{MyUsqzu`;L8TSsp32GBM+~EPHCF$zMIbsR=-q1j-ZKK}aCHKq|`Xr5=wr z*kuDr{pu#5GSFOw^_{<503YPr8N2&*q+xUfslnlOKuydj0bE<=>aHH}QuRX~D+J2C zJb&YNpuQy4gx>oQU2A$Z?12K8I-3E>0%SqHwbqvwo&fPcpuG{sN|iD|sre_%5)QG~ zCGgAu?k1|j4__XIjry1chklf>_rvX|4iEhkM5fO~qTi`qE6PlTeRSC`>0zPRvR6W2 zu-`jd z6TTF^DmS zrtH$oSC5n92;f;i)L7^cNjP8Zvt~&&T0k@of|P@w^Hazit+~s&*P<<)hDd@_gY7TH z^gW19Hm70dEA+(Yz=l}cw6GT%ARFliqB-b}61kD86z^V+uk6kdc#OYA%#~NmgKw;r z`V~N^f++W>qDQiZ&@f!6?JT*MPw*^RGwGCde(wghUhSeu{k;b0xcLEBfU|KFS%WS4wLqXi>=@@IQC3Sz3q^rOSk&12j=p2HUX#GGh=d=ybr*{=z7$ju;XJLiJo==FBLP$N%2w66USXv^0iKbh^=*!x5!5 z5mHJ1xc&`iF4;Zu7&efm@v!O?EDw@Adu06BTUwq2ss8vZkP;irb5*+*xtQpl$adt7B{_ltX9ARo^E#nX;D322bLXox zUq^5D(Q9F+((E4ynXXBL%OrUE!IakD7th_TZklVvr8_~;zf zbbq@TRK$M#;-_oPi(Eh-~F# zg?_a1My&Fq^F1dTn?wW4ZUACEKpV1BiWnRWZv=W zo5UhYa<3E#a=96;6qPEDFaHCTc_Ejoycq(ja7dA=7X&=B+_rx`kW)H}Z*XN(7-C02 z%&FHO3T;sKy`atf5DArMg!ZCdm(F{4!s!nfO~zW{fhTk_xQ#7Y9y%!v2UkHva06{CnrXgSOmpAl{?Dv#+5CbDrmrN~p zug<;?2!m?BDMI1v>2Q=qs8Ra3;xod^VIST&yWtcl1BClxGa-!sam2CRQhwy!r!j_b zynUBJ@g>)B$v_uE;$ISGHpVJBwsYLLAXdd=>ftLwFBkEyG7fXvnM)u}=T!jF-7l0{ z8d(Xo)jbQ9-Sh}?PbONfCpGf4HL@z^IgNRXw7BcO=D>&BxSuQWA2%SKk>|<<3Cg2g z)bNQA^6zO!}n3oLt11 zoENdO5YAKaHJeYR6R(Pm(Zt=($n+5_U9-wwdzH0`JF<<_B*lRzm0<&t^2F$RyQr$& z>c{Oag1~5;?{slK{m$R(HkW&HvAM-U_t2o|s(i-uFGR3MfTTUSqu8xQu&o1_*DSVP zs3atRCE|(mwU-c89&RW5{Nwy(*pz0w%3tJF zFYwczmv~;Lufzld+@>?GPEYp!7v(Oc&U(1dygD7Oq!s#PAIM=I#jEhEa%!W{Rx(bt ziMV_PXFw{31?9p7vcLaVR}%cd_`9OasM?kgExz~btr3x8x`Kc@W5ZxgEf1mHctfk$ zdN&xxt&Pvi6MU1$jN6iJ@bnC@pQ+Xz!r-xXMN)cM^4P2Ccf61I=nZ~ZQFS=%${KW) zZne->Q)rHgEHI~~wjyPUOOnXcY($dCp-Mu(M2uH;M6v?K-6$B+9*J|7BRr(2mY6-~ zb<2H>k2S6;zJ6Bmze9)8FIlR&+F|AChu8R8C)A2WkXTkFIE{SL-C%A$F8#yVlZ~ zrqpScH`qH&982rhsW(qVPIs=Zv63;V;pR~ul0B*_uU>zvK-+EtG~6rb*Y3Ybq4XRS+cQs*+Ym|}c)c4S!}27r5o;Zh8Pu2E2vKZU zcoZ8bN3$>E=IfVRUHk$}#Y?6V0~jl1s$N@5$as@}U2<%ht~OF?hHKz1wNkc9UDq*I za%LZH@ippJp8QGnIi-69#%~@G%iOLpR$^a%?lNj7y&wO*VNCTXlXYvY&k^^v+5(^D zfTGUsN(-KSx`=h}0<7WUXoY!SAR2BvX{&y{x@;iyg16X zF5*D1M?J}LLT6|Qlmi6N9r;R8&(KA$Ea?d$?9aEmICo6A2RebJayu?Ar&uXiGay8G z{s#ELURh-h2uhkX1^)%4`l$O4S=;Scc&^;98jQ()OHT@+-WL#Maz5l+-V51UGTXnG zS~`ERgOEh9IarOI!MHL90){(cT6haWF!88y)$cJ$|Lb1V*h9tG;;&qYFJF*zww5e? z(s18c$M{2p(*|^)$!4p0G1KS^&RMXbEUDGE?eC$A<^h7`9#tR#ug@utk`|NroPCASo;4T4GeI*Wz$M8;W5hfErsjfd%r(NVz zXhCO#tcZQ%6Fn&hDIMGZ&n|}H7W?`lS=~eRowXZUuL9mL*Yw%rU#K)fU5F4PLA{sV z|A7jIcF(8PS+GAFQk2tL4Wsm>@DFD57WIKEP%uo)K9-n-uk>{?LU`{o&dIL4Xe-Sc zXPE2sDcFZ%rGI(b+kmWOHaiT~z|};$~vWkw) z#dREvj$HysiGt6_?_lB4{Vz3Oxw8I5YQY*Y4Y0}{H~V+R)&icDrJR08fI)}o`Baya zYeI>#3Kf%#^1*-G)S7c?oJrl0tD;xe5fzEGt4#j1RS~l~A6)ncdEGNx5_5YFtBYYh zl;}ODI$V62-Uu^?@9q7h*5IyT#re*%S0+pUku#xJ`rjtVp2$P|-#(K&@1@D;&^E_} zZ8ab}=9gb6+9qyB{VB5U-)v6e1_g&D;`Of3rUxIAwS7Kydi!B2N>zCwhu|5Je>bo% z2gf5fWWXk1kb8H{0HUt=~UFHvY|{m%NXc;bB-GKE1X1Dc8$+sP`1! z6{YmdDjesmU2gL;Yh|sb-_DODnEbHgBadUD2=}9?*kdy0w^343tBD_JS8TUoc}<>u zWL$Rcp7)>Po=B>-y`aXUq0_bfHQ}=zLE4vZXU5I4jMy=P($L)q~f_;Df z{>cn^qZ!t!)(dBNtvQqNBqEU8ZsaW~{~9Nj;}`n{m-k85CQ(O0@}x8GO0;sOGI}<- zEjWMe(qe?h8LAkDyq44?$i%%EeBt;nlp~k%U%}r9gd`Tb&*ceoYuPMB+{K#UbMOV3 zTIlD3up)$VyTXAd8ceUS921yv)9ozZV�A&Zv_4^O^)s8S&`-3(3lwSzO0cAE)8E zQFg}0M`fr#2M;c35B_F+J+9EbdKp4qojk=i+tr*yfcJ$;|B`{DDA9?6_))4ugeYa9 z5aq=2Pwkh($sL|Q1(;Csl#&hoNEu)!1Zz<*xcF0+Vv3gC3Qbb<;rm8UQ9B!%cn><# zJp6dI8NY^AxB@m0ZcmL!a63fubHI;AMJHv~1C4dpnWQn<=MaiN!|$>8#2xQspqt5} ztcBvNQC4`b9eydaX>F_79{PAZ!XHkYGU9+Uajo1hjJT&DkZc^;sq*NCJS(ky>g88< zBIc$emCGe5HpHB-6nm1U8GpN7V|ND4*W-tcZBB_gk?=2D`6GG~B$5K;ej$!Xdyr`t z)r3Nzsrd}oSXL5Km2<7@0hF?adiyw_urYn~Qwc3Z(;d^kjOBS*S!;B{HBmMxp4JxE zqT}4G)BCnPEfJ#H(|<~xzj+QnEBqqV zx4@8$c>{mR*j=X2Iv_;7hxb@N-;}{o_9Z)K!jyXc3dAi&P9uM%vZG*kqQ^!Mi|e=F zn5d457}bDYjQx54E5D2Q>(n@lyxB3Irp_;h`#KePooR$JMPEn=wVt~&omVhkVzDg_ z(vAWPDjoEd6#&{7G}CwJ}#&I~!O_{wm6Kf!^glO|H-jZJi1{ZG8;j3Hp;GCioircBROQJ0d2 z8V=mGn-JH>JUeeQAW>?L&9kX?ez9ZcJ|!~yBc=L+KL3h)*AB|43fSbYRPWfmAM@HehUq*@*R_5KBjc7H=Z{Q@-DbWiz$*blGVu+xI zDszIc>Qv!?=tO5yodHoJ`^xI-SJOnGH3%OR@~0>$dh~K>mae z!9$Qu1NROVpL}e&*hEbEY{A3ad@{LbP<*SJKlQa0!`sOaQ=E-5;VLuAqn7=Y_lYeC-+PbSiXTC8er z<-xvFIC#)}ld`H2`dsrSdrcQ;4S<~Ec;;z(OG+{e1mq-IP_K=G+KV_?Pmu&H=rKEP==IG0pUL@0@3c|r zsxJF0BMC`lS zU7$}s#W%jEW^;kQzcFZ3RfT7qf6aA<^4FvZUn*g~3oJJInm3cqOT&^9{qn^`T!QsF zR2_T{p-G7?l>Y{rx;9l+Gvdvlm6Vi; zC-T+Q3mKL<;IYRbBNX{>fK6_u+INffQlmz>*TN7ii5U>3`fb z2e>H>aFZHo=j+s4P$i8*?a)ntcP0!-RATR{VuGf3=&r<(cGKrYI};1(JsB9`N!?Un zF#9mKL+2z=&?Sx9xcT9x(u%)>)uj&_fU8mu{%D^|4e`^e{$po7uW3r}j&O^z2=(>%LnehQBw`gIytZB;RbFmR$lbkjk*P^Ry z#l%*daS9py;rcqUcB`{2H?JhGyjrwe8rA!4Ti9h@KQL(YQha3RA8Hstp8(hZ&j8q@ zQ4Sd4pk;BYc>_O{7XmH;UF*W8Wlt9XRIDXSHC;^G!G*%7{SBsvEP7j6U7)vj z=Syx<$!z9R-R~;YU7A4}PZcu-<(@3tg(zrzFV>u_tE#S>A|+u}AhcxvgfGuDa-MUB zk^QugIOdl>7Wun-I5hqE&5$gY)O&Xhp^A*K3X=Iv!~a%<}%Nd+&1ZZ-Tb z!%a^`WS=Rh@_NL6|~pqzof$F5Ou*HI&-rv*Nj`TA5ooL+%-Ib&M^;MP>^5By-T@E**RhEowIr*_L6tGTFfO z+}fkUZyZwwJQ)Y+xfTy+Ek)wA2z5jV^acXH3pu7MoTbmg+3oat_`kESU2%eDG@LY? zzvW9s!#bcDi5|@w4tWah8QKL*c~CvBA`N^@|2J{qS|*(u<@NOJeori@R2c>*L99kE zGPLuJlnXQts-`5hMTZ&9sKYry1xhNz2KE^wn%U9$(3toyuQZ)F`h^I-YkV<$o}pcO zHX@p~;LQQoeW)W6Yh4R0|AjCQVxlyLk==KJg@yvI_X_K(*^z z1x%A?*C?>f1+C#0tv9K`QC3}PU5-h&r`HG_7!}nN9c5JLi%152=c;@j?>}1f%(TtU zf&00-FaN!0OnW-}@TGGh-;|@*LxS{WFj(i|Q}73_$Z`-BncB2yWWP;Ed2XZQL|ZdD zy74?&ELN*@wk{Wf%-kXxa?g*0KVfPaD|)H0OmgXv-0gvHq(vP9R>CfoSs8GxDFdhX zP)(P5l~;LlE=n~A>;IBK0~m>-85JKZaYH90BOh`$d!dS>OslM-slF)8R~ps2hf`yhA?(2i*VPc8 zu8zNSsGNl?FRGtAo*|%ds9ZK+W42@@4j!K7F$zS*O)P<)_#wF#ViKms7ELi}Zm#p+ zb-;w|AH5v%*y&-O&+(JM+>FUJUii@i5lLXBZ$*Awv%F=qQf+zmT}j|12+ud_)I~{3 zOKC^Il?W6FxR<~+l4^LiG(GGRN7WA^Q--iv=LqBiR?Fjsr6+-%z4d?bYE2qbO2f#C z)6*)U6H<}%oXtKl`UCtJDYij0$R;{dQ{BGho9!$bH0!)Ms(b$;2|Tq2c&>?^1g@k_ zZui1J>t9U*hb*$_Y`%Xy4l&s#aWXPm)6Vx(X4E_o{w@jE@kKLQJPt>L7`0U}2KJjI z)!Ah-uwOr<*b1_iG}oj3V=Pq=;3f<@SA+cm0iNRjCgpRFnphg9S5zd=vm_OXW50q1 zbsQAvPs@1!%7VUqop>k$7|fz`7v;~l3Ub-w(4d7(g-sHoShn2)g_VYvn~4_wXvy*H z9-Hhrco4>YGtsrp$Kemrk5V58@9iQVd^}+J2`Qgn2yms@&Q?L1(q`fVMQ;1cDFPeO zXq8Rv7wg)cz*PO5J#_AOOcXj-sS--X3-U!5#SHHi99P<8?-b~-cpwg2yrxZ&nNvM* z+iz$P=GU|um7+sd>V5KBFgNvtbY#Rz0Nfm=C&a%bJw;14`D!wasH|H;845K*fZe7a z=yP^d=#}z+lko90x_9Z|$xqFpl@$W!HqAcB*SBk~Z@h5`y7T&tPHZ3N-OksqZM)QE z9ridw0Iw1Mmp~H#OQ7C=3G`>Ive83TW6&yK*VK%sDz6uWab6x}PlaFi<~pTs<@Ygx zevcJ-N;Uzr$%xO#NK_)r(VAsR=MkcO8?UY$=xYs1_*l9;I6HkOBqAg5Lcm-3QYXD2 zLjk|vW0l1ost;v06Cb;^iWg2uOc_dL9OUO(P|do5s}cy^z_EHmj#MI!`3i&?GvaJ` znmJ{24!+Cc0|by8*Y-}4SlL(mJ=h@;0=BCT`$loR= z7-k~a34pwh@*MSh>Qz3b$V$fXht9cx6UffGRQXL3vhot=szx7tqpzz7aM0`3M8i3P z?}1;fwiru;8_}S}9ZDEy+AlePFC5aaMMjsTwH7s?u z3Z$YmQAO#1N6tB;Xpkalv*BjqwhYX7^9_~8I^xBWbhG1TqEoYv!$ZXbv>K}{kZFd% zEd#=`|DUjICW6lP7bINGXr>(nc7lE;I$@T3^EG;VL()0s9LVb8<>yIs%SxbCO>X!c zUnkwbyYXK@9yw8suJE4r-jcRl2jXmmx&kqiDA;ZDY?CAjiA%|V{< z+Nnc}r>5!foXo39<0&yavyi1xT^8OuFYpx0H?hbo;jKJ{MNcR&oq=XA59Ew*-_?LM zjiPtc(7<@kXqT#J`r`e9@En&#)`6(_fypzYY5rqH&QJUU@Pf1NavYz* zoeuxx#}e$lSu)g`^}e3t_b8(xlb!xx5e%PN+hIIW8L!!JJaIqRL3xu3)Q7%rWBRS> zDY$K{FxTmR$5XIZz*eq8Raha2iezve92sGATnLhhpU8mrS9N(|G?nOPwD^pW z!+Pk z_A{Gg&OuL-(R3crAiQ~pm=c8!Atce)9?N+Uym)>;T{*9vP^2Rubg@DyALnE*!? z=CU2;DhwN_jgNXNl}{;}p18iJMhALKfTs+FGY)cdt*B-#!Bt2?OMqR&l7=Dxrlc9O zVx@Y3IE$^J>OcY-0rq8LDB{V4S7RR_T@J65h7R*`V`A$7ZUK|~Ji9FcZpqOc=6S7Q z8?J<4rP$h;@=hE10kczzbQOo?qNH^wE&oHu<^RwT`-e_lZn0w{V(M37)+9Gq2FWW9 z_!_EWtYHG%$yFsR-}me{u)sTUpf>%>4)$*nk@+H9`3ftZP(ZpEY4-9#@%Z-AHV8ed z0Envjy!)DVsS%JJP=B=YTQV~kkb8dkwP3X+>B#D`5X3r0F+nwpYj@$P5Krm(=KUNbf+a78P$T_AP>E`I~d5k@*ELT9~5zF z!IHW0)ny~o0G+kk(KQJ)#R!h{n&`8U%Xl+bo&enR6|2_;^;c0%pq2r}75#q=3mRxx zKKR4{g_4!e5WZ&Jz~NE7U9bX?fpLca@<(fDC~uM+3x(AA(;l3YPFD_xD{Gp?#iCn` z(CjiW#~}S*)1t#j!rFg{5nw3di1?Qn(sut6LwapsnSR;y#7f4cmD6USi5zjP;~w() zF*KVn?l*vn88!e9U2?l0KJ~zAEjVPI#SfNbtQnvY6Bk`o?N>gg`A>CO0M#}7Pjz=m zk^$At5C|Z<-*}>$`nWW@HbW_Z3*KEOFIJHyZe~0svu3`J$i0`oECcjsJ>U;KkxT#e zXEH#4*0%&?63cR9>k|@Tm-@2w{d2UaS(&o2ynEI|Yw0Zn0Obc^^M`1Y{g9ka2he$8G4MmnNNl4&03Iw-PyATEGTo z<0aJd=B~2;sj`mSMbVT$G8yX520rQf9VGh4o+s!kU^Vzp{?T+5jCY-Lt~Zv@BINCIcj>9*}83 zt^pM)DT@ba8QO@TxxpRphkMTkdLZX=KbL3Xqgcp1C*DtvDlbuMz{>7#NkNYtvLed#CaF}z4i0S})U?}r_T$L-uY z9}t>c9Q38@{dVCF#pOh!K7X|6A$*4#I}16sJv)2*93mIp0mZ-_>uv*=mhaQay()f4 ztWIUX_4fGiPLUk%1|EekftHOJ^~f-WYasE%8T zW0}Of_SRSq5i8eIh*Ljq9o=F3_u3tQdxpgY?6hbBKnj{C*^DQKJUp} z&oV^sR((rDD#>ye{Hpz9LCo+^Z~s}k{)<~yi^bTKcbLt5uS@tgt6}f1W52K0v$|^! z#dW&dzUJnc$CXEvqc2^%lEQ0KruW#i+qN#Xl6k=7opwC$_YR|rhBn~uur41(Jc~0jeSs7qIQibdo3YY9lCE1wQ*UR)iT(awDr%ysss&Bw z7{r~|z&lV~^SzqH)gr)X`o*_Yh>^hPr#>oztDGhtN#L!BXJO$)Uz^LbVI=o34YDsn z>^$g#534N;d0z5_mkGDH3e|$~^B%=>@{c#j^0mt1GSFIa7oHiiZ$mA2!C^vp{f{eL z$&#yZBFR4$p@?d$sPhP-olT0j3svSCGU8sdaz&SQCX3@lX;7R?t^0VYQ`5c1!Y73| z>eNo4X36Qyg5n(dJ!Z|zGbUe4z|Ht4=FEy72pJCGjKwTVmuA(y3U|ORzn2p;+5~Jx zB@uc`rEOAV6Ap0=#}|3hE#EwVJK9sTOGYo5v-&gjg*d^WVP7AC=Y&o8^fZnRj*=Mv z#KJ@jC!bZ3R2a`}H0K|aq9j6GZE`$H8b^~NYV)G-L}I~-AWr)8en_i4E}S4EJ@r|| zYptPFUdg{$WFj<cU zM@7>*KBc7Au%2NO^%QafANY*C1(8=pM3Vn38pNrs#mmEwaW*Ns2yLH`GTOf!jnt}2 z8lW(}?X`I)SovFz*Tyf#1Y_EW0X2L$+NmKZN&ls7gaqpebx7~}j94Q`|E&~wTk+|n zrgl9SJGMGNvUMxkXc}|v>5MI2wr3RLcyV6+*tTK2&$XYGG@->dLG+W?cEyW7CeYOk z>rl66CSxN3Tv0Z!I{Ui<{g=E?Qx|j%85B{34wE9(AxTHC&kOVRxEniur~lsS+j6yh zc}?@u!qrAnrNj`OmJZ1vf(yf;t*J{ctWNKPgA8oCmNuAh^M%Eumy`7DZNW@WX;Q^l zD3OEHkd&}hTjZdOBo(6BpviAUIHPSeOEw-0>$L<{MVh@J!y5227RGEFP0pellT9!? zHY+HBB(Do4d4Qo;6VPtl4imguD@AFDP{W^W$4IxT9Jp*CMQ37vD@GN6K(solGIC4XRNhg>>hMUz~C*+HN*99^Rzf^#&C3T-{y3Yi|ePcgrY= zkUz2AA8Law&ti{pdQEHD{sx?h-g|wvvpPFbEMygpL`t*^4s=}voYjxJ*Y^_Rslrui zZbg5~S0+Q4lPAoQl%!hZ3U0d1xI8JnKQe4s)%d22YCd!^39S}*DMicwkyNS2M`Pi) zBO|QpsEcx^z`+Avb~48|Hmp{NvR*6B(eBP)t~nUZeIaWV;?KkxjXbg+a6!AfAKI|T z!A(P&dq)QdHblD9U@N%Kw4?KLk2coAQ1iFAG*&_{e|PonZpM1~{k_OP^19v|+Xvbq z#A01xPshPZLlCI9|7*T=(x=0Z!69!$xmuWR+!=4R$vENO$lSQ$R@$8VoR1_a1D|k) z;&W!hY~{2_C6+U%W1bT7APnbrRB}G#g3aw#UmbOwZNZk+5^Ry6xbDJxVahDWIjLKl zimNYJ8f70|nO6I!`SnY2uJi|~G&K%~U@p=C!+fRHSHHWw{`%zR4AnYtUQ%*=QWiQM z%BlQpp$)%g)aD=m9GblojX{8|K+NSvZY?eBv^J{_XxF4T2bs%7$?J|sJCbejX_cSj zSXmyZ%Zjs4TBu*+sPiX(w#g>?*o|oyvl#2#{!jn4E!ab0-C75h z$+&bZ*02Cq`Ai?|V=b&kS&A!D-fOg9OgCq32p!z?u*L+s>?;FnQ&nyUP@%hB5VK`% z!+q)%^oeEO_Dbc0r|!F0;?3{f&qB^O&VOHT&YynW3_rD5eby5i;uXXcnkfHUsCveZ z(PO*)c2;+Q;OgY+6yNL5x&%_rk{tKy>tH_RFaIJ3-0zpf{$&v6LOK{eU7;k+uI|V7 zv_EjU-5;kC_dnov_8Wh^SlOw_#}IhP&96x=BUt-WP5~kh%vdU&MS2x}xPKT**O=d1 z%=yXx<{8?`%Cd0ATND$H@M%2VS84WPHS$F{!T-F~k@vY;jcq#F_y73k-VS-adu(z; zFSurKz4>%^K7p2YP1MIwO@AQPn&e|szA!vppJ_|jFzjKGY^WYOz>2zvA8+JL7$bBg z{X`z+k5F=pO;#`?;O7SIM7N(ax8;Gb@Dk)5UT z3=y0#J-yhFO(HtUVqLzZa3(zI*^S~E@;CX_b;$qUX}JprSzn?PYS!hA8YgV{NzW-e zaQ@}5uFVPVf9C(rTmH{n&*`7}XYQf5YQ4_UGp2}#`Dz^9>zV;vk5465sMjJXlWR3f zjPBn$$$Wo2bac@{;=%GO#CtJeaNKIC(Av6?G`x533+idgr5d#*0aee9w!3ahY+~?t z9(c(YwN^$gn+dqn6X}~P_!T@h&mI=B^z82nT7i1@EJkx|MQ+-0TeZOO&UM+c9 zK_!~!97U(W{*=c%>*b-PnrSj!urq^-1oW{_$J_gj1_w`>Eei1u!h`WTBsQ_BJUyfi zYE#}9Z0R-`kMI`17lL?ZC`8l2Q~yoj=#W@Dru6g>K5!#@AH_2zy=Kys8XuK4H`CYS z|Hw|t;lQnQD~4bD*`E>$zV)R66~~{CJ{YG5gC9M^{*%47bSbHsBQh(^x)NuqSr*5; zveZS^$O!(^rtfPh>KW8Sl+IQUMJjwuWl}?LT>hcpI^fCKvojm~hvdKf0U%Oj3eC7asDHCwRASW6H8#v0&!8#XT<^Yn;HE%wws zjQXY6?ySV8YT26U&E-nUpa;@oaI)m6r9UhS6f|W0;c7p>)x@rTDm{@kV@@P~QDE*M zQGcQ2O{E$NXOW7eg!`~Od?R|eEk#i!nI_J)Rms+P?fxiv`rCh3P>EIYe@Sg3wu>#> zP%Vp`jy<|mlQ#TRz43hW!7xFC=A#KjMKF55N3A31@^WuhuK#zy++Bw!_O*!itprRi z2D!TNdsW2Z@}_8^PEf1TIA3XYY-F;ixdq!m!o#XZ8g2#uDlCUHx1l)CJP3Mu_OV8s zasVwXx5If0zWdE?@BZfFsM2$=?Llt^q~&z(5=5(GIsaK)Xc zv7b*=W9)02be(_XWf1|D{Py?U)Ap8kUyob}#{z4n@gD{>787w_@q^^zgDV z^uqf(3@bnlzxyl)-W^2$uqVjx*!9y~07qTj(n2rE;l>)~M)uHo(dCd3=gJGpJm_%?J{xq2j+Mqbuxp z=W(L9k)ay)yMKGg=)jqKsTq4YrT9$ohQ?@r-o2*Oq61Grjf7%e+~9C_4d@FPO}I2X3# zjiH9UA1(s#Z8qQgO&>i@|BHa7rWAhu&n91ImHQ8yTpG#4$j=0dtDteo$j_LxfDn8$ zKk?B)1HviP*ogdqXv)BHo>y99C`{c7SoX`*YjxKI0hBqG&^@r! z@j5|w_@pq-pSsE%EJ_|-Nak}wc*Y*B96fQw&l^)}293`}IDBzxd#a_S z==`|y<00^ZPli>3AY5EmiZAPAS}>y7AHn<7w!m$$$DI6i+beF(8r{67D1Czo64Crv z2XNu_y)n(Hz5UaJr2Xe}e8`puR$68e4u`V2fuBX^lY=G5UW<>GeBzCsWh+448^vl4 zd?~)RqVcIlamMwC&Iy>g{!9t6&U-9ToYsY!IWa8l z<)>QQ1=p=#m*BT;QY~x-K-=w$aits-x==l_4Z-3U^IWZUa4+jd0aU-{0R>b71i1WV+qC$Hn}+fo|4JmbzoClmT}b-<;~# zH$?osA6RW_IP-}W%-C%zeQnPOdAjr#_=*$3qq-UOI}r0oF7|cc0wnD7pBV8>i^HX` z;myo-b%M82ylJIm)TVao!7rpMH{1DcXtM!?f_3m+|MeGlHAAPy`-YV z&06F?Z@n*my8fL%Dt!SHFFjRYs=TT_hs!ToR9>;aZh*iSUN_)O-1_8(8@x+;svcrKC}~Pw`ZqvgBlJU%&y+Ry4Ia4rTQhTv^X95Z_q)@n=68cHFJD!j z-d%36Cfa>qEv&zMi@5VVJGpSajzlsG_L9H7`ttP@tDm-ga|U@vYn5dEci;H>*=TY7 zm0(B6mu=MH?H)R-q4@VviU))_^b&qm|13zZ;-lPdoX^?${-Ud3=?w53Z`bDa7(yW{MIN4BvayDp;1k)ecvC6;Ptm!TUpH-&k+#ph3_Hk3XDFXvy6ElT7n zdVQ#S_ffh_wx=h=kNmZLn1{7jjMwFpRL;?=daHNQ9@ORG!xekWck_2X;O@xYaZ$aE zYn8CWjm2vU&g?_ar<4^^iNLEw$0Ln*0e9(6r9nM(iV4dVqE;{@?dUl&-dCVC4tkM&u`Oc6 zolxVx*nDlFMctCcdDm6@irTEz?oG;RpAalW8coK`_|RF7z!|2r9|C?EWu8{FdsRLW zOL4v=fq7*A;tF->sV5lqZ5Ju@VBuCIabj#XC0uRojU~aG*0FCEPV>8G6lBl8T9U8e zd(GUCD%N;urMJHW;akDa?c^*_kEa%88YTF&2Jk%A)dq(b1wSob06=GgJosL0GHMy9 zdYXtkwN+J|_4w0=QY0LPl#;p&4C?O{x8=r|>@S@dza96|qGcHU@zN@>*E*v2)I+GE zyyIR-?ZJB@%C-|-mlM=t``C3|)v58m7}v66!tbP#p>hw#RaSVyx(B-OW`g99*m`p@gW6U{T#aV<2(>&$3tOFW7w1wwdv?07((?CZAP z!@>KE{}BUvN;xZgzgC`nz{aRtU1U>96R{fO4 zq&2PTf~ON%Xog|R(KTo-Ox4zHm zefbkJI8Y;=yXPL`R1Vj%6`uUV9aArRy*pDC>^d(q9&32-noQ-BtIDUvVTo@6-_tl1 zMw64+c_h~v3ed~s*jMOz$HrD}`mMdC63LRegS+!b>0Y8&7xtgA<=+urcICQy$rU*D zz7Fl|g(%#>RVv|<_ipI#AuxDih(*?PXBC9QO2MgFbMuBJP33pl3%x%te%H<6y2u{n zTIk$^FhK4$C4Z9F57c;f?F9aLa6i~E3izq?igaqq=4Xum-&e_Rw&*!8cEdAv!@hg$l@J_-Ba0+t7~1(Ht6` zPkgeRe}(x_DV<|)O3GIv!z@^6hY-MV!5(Au)T7o2A%@ayt5K&fbDJ>F<@Tdwr+bJOqk82neIj3SNsL*&)S&JV;G zYC~fJWG7AqpPpsaF`T8j#ek=8NrSECy!sg?%Xz)>ls*1B1Fat;WNCv-kyOfpD4VTSU;Yk{nNfYF=qNjnHcdD%SjH;`%T zbssZglTY>@o))cm7`?m0;=b4Khc{!SRmVq#yKZTWduGpN%EuwvY%s^uR%U6O^iuZG zmPH~LYsBu(8+q~>YIeqtOMZ9}Pc}PuG^&&K-t^yTA)^hfIx-$)u}U1O)4`(j(v#{T zer0+3mO4tqaBd}D?VLKAs7919jM1kKFQ^t|NW)hcf=p>dYYx1{s{i{yZ+tL520X&`tG*!XDdc*e*y5z0PyeEo6Q%* z993JcPzQA89BZbGI83UZjSoCNpH#QY_O;rE{2HB77;guwdG!VRy$K#YvI=jo?>1&} zeY%Q#)X{mW8Zru4 zot0p-^r@I38NtjawcXCYh9IKo4z;$`FqUp<)={MZJnF%ETC4DP#Psz+$(jPvu=nTF zT4m9mEiK2@q;;xmg}XoAoJJfIMMWPI<@5{&;Ehffchf`an9^*Xn`y3PrE3}?yC=uE z>r3>!nR7FBDbYwgFIUg+)F3qk($DIFdUwm|2nm{RMS7ouI6H&+EhoH%+Kh)ymch>D zfmLhYY``^GD*rwi3b_$Rv__q19Ejrk^eS-8h}_}?(OxK z-5+2?8)(xValGT1WX$4+hvEajBGz)rnawzvbkcofZ}lU<-fa7Ep>o=?BBTBN73%4U zFhTLMjDEg}cAOQAJlvWU^8ongQMWT1^;qA_ z5{-AoYO)p8#UI-bp!r?8ogG@GnSNidJ5$jn0;CuuS);yhJ`{l053%bOi9FUcwdJAa zbIW?sjQr3}yiEPtSm&Zn4=?kNC7hI?QybR$xLCabs_884 zfUy+wwsMgk?HFt=31I1xW<@Ez!XqAJ8>+(^UxymIym|Ow{x;2*brhp3DA45`Q=Q>8 zKDZ+)^rJ4!lR)T-0yORb_IzX%_4f^KK<;%v0LfFaXdV=#BMSATZg}Ki3aVz*B`avZ08@H zqBN>F84)k-mVSU(PU5)gHj;ksSWm;=G~fke^sj>imtn1ywHmumSV=8|`kJq;R{v3(Q-Nna|)5(X%QQdnVfe1+D5Z$#88wpC`Z{6 z_p_+SnXGccN+CF_q6f$I?Tppgr!$1TT>-i?KGcB zJ=shUo}Tf`9&fv^4e9-jtmg*w53*EdAq2RCvcGOe(C zsnzQjxO+;_CfRv-QuQ2|&2H~n$$CzWUo!RO4-^A2bG@RO#-x1p@>6A*#<+?)TpGbP z(LV7*ppMFu>IY`J1eWWQB55?4KmdD4)`0-32Lb^s?cRxVktxjInWVhv%rb_% zu>|I69#B#~bW#lTqmX!2+#PXJ&9hFu6LvC!eDL>kOs4QOZ(Fjf)0z?Ay>}!&;(Zmz zpZxul6JrmF=#$lsEU^o}Uev{eI9C}~l-caIK2&Wm*1D>rpG77UJev=oR>NPT<|baF z)(d`jOjKFhA0VbTDKLT*SC?VBTj=VDTX@;+oLnJf)fbQj>&9_;D`9okP2EaZH^={}W=omV z1djxg6;6fEM^{5O^Ahg}^0HyI5jUqN>qg1C7S@z0Th!gq7jK?1QouyyL+jwu*Mn9F zcF8y{V267_P43qRXv7-HxL^yymxZZ11)T$CQb$h?|OZ=>c z2C15jnF>cA?j_+w4DP&|kNqCEqV-5b60W1?@aKDO_fz_$7a7bDVvXG48)#bP<4j+~+ zZUmrzw2AOq>M%>n!cKfUZE|6Awrj}6Z(cSlEl`u`=T!5W&1J=jWY%RxQqH92Und41 zLYUP6t(X9W%|oWDe`iQ?U8A|#Y^J2qIl5&Et4V`_iy+O&M?HFoh!rWDM(x?vJyfk$ zw|$R?qo6fAlfB-B(!3RWi91I70ClVoY()F2xF_O#$FtS=lP}&J(#FYEX2sMAIqO{f zL&o~ud3Bn~iD>+u(s&$6aZf)EQ2xIZ`Wgh|T|$a_=(n_}Hr!oW2%9AW*Y3?!^f*WL zdQr^|Hmp)S?say9(XS$7z8GWe7_1`C=piw9seT4A_Nk}*XX&d*2M7{q1+42-z4~7N zYXyX?!yBSUfmT3lgHJA}en`XRJ;XJUobO%d7vtUs#~AiZ*JF&Rc0fIdY<1~LxABpc z*m{p5wcb*&SWSbGnt^61e+KV6E`siTQJFP_eTK<94!94hVVjv`7n<|gsC9#HIsL*p zJ$_Wey6hzC!dZbe<5MoIAg~S+Vz01PhB0Q3Ijf&QJbr8sH_I;25oZ+G>ZRXo#Lphc zU~Ec};lhc!Ci;%&r?5B_insH`mN8Gbs(vu38!Pw6a9UI>pnk5D z=d&?|HGDm;UVd) zO2f=K=@N<2Oq;eg4KTveJifm%h?*ecv%{jh8g(kAzHbr!&6(X3ZX&Kd_p3}Oo_+f= z5mJjK8PEFmqgM=tnOOTeNvY9*Dj${~Pbo%SAI?N057Ah!(VNqREv9QCAz@NXe(i&H z@#S6$PJXE}W1eQ=jnM^=m#$*E$YZqI=CHI|ZDT)(tADk^z19fq|B9>Q*V3lWFT!Sw z&B9jn9Fa4!{5ymje*{@O-Ja+N>n1Mc6_9bGqV+cy^HGwd-`->B_fkzYXQJG*X^-%M zuRYj<6m5CGLiz>QPa#eXcwoa$_KT}QNC*0$3buEP6AWtjAE_mY`>0gAx*aPR)>1X?=~pAK%U8$Qxb^yr5w9M1U3Qen3WX?{f-&-g_s`?ics}|< zwMyU39tqV3n{(}QkuTI@CvWU#mh+ei$s;(hQhFpvYd84^ zWutFl>z_^$aJK8`M6t2Q^NBz?0n=0x??iN_fr!6b{z$~I&G?P*RYa4N9`?LYgxH~g z_)KAmXygx(TGgDtArhqNHVL%ipJ>Aoggm#vn68tiL1Km*t0odK_S&8E);g8=9BEwvs$-!JZm+6{f6AYt(i-sS+5E`3s^@@Of+#n0l~>{ zl>bRTYT|<9qn2GOj?xgQEh8E;8KVfuQ+XC%N@as`cF(7GsEHXfdAVcjz%P~8$auz7 zWEH$wORJrb*9hfvs2Q%@%x^6D@De3GH0F@g`00nD1HZda+aqu`W1;L;qV497R3BZ_ zzO4L4DBC#jlsx@Lwx9cXP9xbGL~ztZ$WF{2r6|x4268cRx3c6Shp0+r?iZs@5K~n8x}x| z6n3qHhHn~KP$QIf;qBDqZyJL(=8`Yf&}L<=`Y7{^CuN-G4e42v(_2KXCP}VZYWmRS zhsh#QRK#B$Cuq&5oOo!C%p<9^+}@ikosrpk{D{SSTdx;iHkW3*IM#o#_Bk)=J{QDC z=DVk@*xqLx16m(jQbr>thQrGaW9hZbjQTQ8T$J;0#*Z@Ke>6*F&0S_kR?TW0UcIki z>lutpF9#_k`e`@B(h=7st5Mj6mk`G{;^#GqBDnharPu{B2pUF$8tKDbM2|XLlPDu; zh%=*Uh%0-H6Y#!I7k@Eu5y<9%ESXn`1)h{!31mBKCWyt^lM6st^&A_sqnc1Wjs+2Q zMhP5v0o@;H;9BHn{L_fkk@Wx9)_aG;^+j#NqZ^_#dhd*A(Ibc+LX?alA!bI28iuH0 z5JV3~??n9}j20~+MDJ~sh(zxqS|Sm7kL3BD?|QH6{b%NnS!d1Od)@cC*FI;Ty*;eh zG}FYVklQ*9a4ZK90MC$dWN2%_^>URN;M>3jN4mx~AxbCYnpQmbV7G-$62@PfE08#{ zpoKm}Hj>@nZWMBZnX;3d<^fRiFtefyY`M2 z1@y2M3GHmfP)OrzwS;!sQw@z+QSGu};x|xob?}7x{&x}_W#kHwv`LWESsB(HO504K zfY-CFRJJd{->Pb^GRK>5^*Tp%nn31>6U@RnDqGx~`*91YsL+a`0UT-Kj*&hRQ8hED zC!A_yo-ji$^U7pQ(1GuIKaLt*&x(dhHN|mI#?e*lCCI5;_;f|Iw0%DkKhyEVV#~w` zU0tyqW&Em(mUjiN1+M3=^?ZVlj6YRAI`@v3dCo~$5Wz{=@OCVLWC4ckv-)a{6%x#8 zvWlrSr*E=RNptT@VCq>p!g2%0G|VMbO-u$za$vnv=W&MY23KJ1YoRCTN+1EUNP)Xu zAK_ExCKOeF~iu0pCY31w((}jxONp?j*rgTy3q# zY4NCvT;C6Q|4pG92|kOwFg)o!*GRa}>d)I3#;xDo|OXk1*6F5fx7id+o(RIt4EyHYgNNK{ej~A`?qPT}t+bM&)30 zEBG*I1fMW0GCpbWN5DF~R8$FX1ofd&fbkhOn#|^?_&1mm`rKNW@~XC)^{<{8c=h65yw0s(4o`Aer{N%p zX?Vf@TW3JnsO8eED{Y&UUc=`Rp>rqfY>ef{)-*@Dt2u^?40K<`RCx3CCo7u zTuLzhhFqaUi6TQSG8I^$IY2_6s+%g1s0X2(LsXk3@sO-|U)L0q+k-oM@YhKUlT$o( z#(KKoE`S-&5Pe~syB_=Skf&5W&^NcXhIC=S-62e*AKDMvdR3=23@DC=P^>6F{H=Zyv_-R=rcCW45Go}_L_vS8@3ar}) zD)m%aCGEjSoVTRV^jdLh+C`lYbT*E2_n>@8bqzk%z|*JP|JI=&7czGn4drP9Ix_~; z10^N$c>V1X`aSVTMHq3%BKPeW;gZ(qgW|I1E54A51Gc<6oMYP{BmQDvveW*d0IOuw@dP~^M=ik!87 zMNWay&{dHGP%R#@syv}~kdVu}T%CADpcW+#`=lUWMSyEn8%-4$fJpEJlG5u!eQZy@ z`)FlVr;V};##x|Elwjg-{K{VC1;r{P0C0J}=o-mM87BecUDPfPMK#W3ozX^RpwE$Z ztBbr){5wjnYh6pK#m>l{LLZM<7ACJb)gBbxR^d)NU=tXTJ@f6&X0 zpqWXrE?_HW)Djh!8ad!kPe4XEMA)F;|8g1Z+oDI_f8q3NqoH1lvYBVwQn3E-ak z=|b`q?t#`*?YJD-i~ejMfZ67f{7P z%1M~M_#|El?_*-{KV^TG933;%lB{L^ngNin`jbg|@u&f|uir6N`*EUS&Ek#Hnw07~n@j=2>klPVG*++OUe+`)U_wX|Q z+#mzR1oGf26jJp9F6DDGjGL0ydiEsji*=4UcBkv*{ zuK)bqT&__YF61-CBHZPIZN`(ufS(b&{QYJZf>i+~`y;-tUkG5-B79`uv^##LmYvej zdN5v2F>Kl!&kEc((M?XZ??C0tFU!*UsI>JPj8 zpI{%?m0}#U{RWgvFG(H+AFWl+%w&K3s4$!UOI+v?|!nfM$=Uef}ewaO^B2`BjIQ}lwH zM>|{Ti!|%1pX&F$kxB#w$9!Y=OYd7H2f+u?)Lt@|&x6Ks{#4^8lFE4}>oXWu$E>%0 z2*)*&=gfls6BwiJuil?cX-Q}5X1Z#NEd6nr99+%4iZ{&-UsxP;$9T)9))jpQeN1P{ zo(*BrwdA9j*Ib*M5!m&9Nf{@o_==!USkJXTY-Dk{>K8&+Ijk;2rD z^=M$lB&4!J6A|M%SZ5=1*NsSrdvx&!x61W&%tS>GRF%%COOgXqkx&B1WaGL5R z772*{9NSi>7+WHj2p8PXNvwY@1~}F%&$#wAxLnWPOjtZc(WF6Gys3sjJRyjeLt5ts{56`G1aoNbx0MZhK4E(a9Q4NjFw_*SC~=CA~Pg=76j2Hr&QvH z-mo>Hhy3334kuwxOM{EP54$l2jvZMWr-ur4w+Tb(mv7+F(r)O*0p(8zKQ2}_PG?3A z7hm>mJ&UVtJZK&JeemTYN{%Pk@A9}l{YhTF+Mo0pvTm}kKV_Z`ceDj~wO!i2czJ~J zUGB#GGWfIftLE&zt}0{y@Ubw4brR`n7%OH~!1Iwar1ovs>5Aq6yONJ)J@fzMwT8VlJtg zpmUR`9j6Rk~m$y zOVEjzQ&jR{MtEv2O**7`v1fg>fA`r#A^N|z1b=Ox9TB%fy9^Z8OFUsph_^GY#d5NF zXP#fysU>}NmmH2|Qd#^kmTAC=;9=xytR*#e+Fm~Q~@?>Oh8}QS?Ye`?7 zyrbFl=nq#za&k2!0+MQ1eVjD*7j4AxT@%4gXvr?UJ?UJhzAj9ZA3vmJTVQW#l?=Kb zIwa_M&U)2ifzmQm1QI;|7*!b-$ChpnU_M8?HjG6gFnf8gIWr1}@ zf!r{$JraLN@Tj&Bqgq>A-BjV&Qrca6QU{2G5Qm!WM$zyE;V`s&xc`Vh4uVH_Ol)>t zsH|dIxWhIa>!bsHw?r|I5mO)>#s@kkI@)>t(L-9To%4^=ruA7 z+j30LXw!UBpOhJMtsfd^m=PQ@teteulxRts=@4V+SUd$ykp)U)HGW{5{CVw-gK;l; z8M7pU%!nlLN1DVI5g81nU7AK|F5eVfuoz=!kPOG>;ogVyTmZ&)TmS2q1@qUh3Qzqt z8EbDOp5kjeGYl@Cl5_k4nxlqgsBhEE#~|hDSQ9N-G7~RBnX!DwE?34%0S;iwGrNW< z8-v2rxc=}n$D{G&PlaQc$NGFSJoQpC%fAU^n6)<))E!vuD0No2nJ#sdD&lZ>EMEm@|1ck{G z-a$)3F95@C!nTtIGt$1^Kf4sI!~zra1%G z;HbWKRl`&K3w?aWxnFr_w;au?$ruSzSrUZfpe^J9(NcoG)qs$6JUd|8`@+QaX;z}J z+Xf0)!d8A_aXr5?tY|Ag<}p7{GIBZm>4F%i&_~v{$`JKMLQmgm>svHto}n++{m2zS ze(;bN#(AJ9?T{@Y^^?J>*oC8x2z^0w zL{Uyc;3Yw}IkJEgsUddTibyQ2U}dua$tpHjE}oQPr+Ejvd^sQDgLQ1_-l=q!qB_n_q3imsp@`g<1?_n<`D3~mMJC&_29y8A{8*!?yh%);Yb79DJf zbaJNNQI1}C-^Qqnr_A2@&<8CWb)<&CZ9Af?LHHW-K}Ihp%k`PA4%WiRz$w)>1OaEzEP}>M)q2#FJhz(;Kgdo_)fp zJ=-nyWT6QVqMmXJja+{m z%Iu%%^?#-mSEfrPg$H^{8VPZ>CGm2yllPuS@>8v9j1?o4CB|zxs3!7g*dPVVpVhnh zZeVV--AJ*{upUsVgRu)qEbOvsG8qd*gW|$>Sfsz zf?^z*D#MHIhLCQwqL$_@4>%#_MLy_-wnu-P;z_P&?}@BrdZ=W>+h#Q7%^ zi%dnsG7^roZr*@zq22U~7mSL~FwKrw&5vN0YFJ*k+KJtIT(3M0hr7pPUtLw^zuv># z*&3FgzYz5M!RiO^y~|%JY!G2L#XWLw!3r^*LxZ8`S&57aP~MwA?Q6GW3F>8bN2 z!a>o-p#aW`tp)P>N3EX3O%1}oM!gLRDB48wg+@Q9_GO72=-eE1noJ^yCBc!aP}RZ- zWAlk(?Q0dkO`VJ}(C5v!O4Tl%*$lJd$Tdeep;@0oKd)-MulS(e6J{P>LTV1x=#fbDRmqVBcg_Y(Lnv1ZwQiAq?IKmM(%+otQP#I#vz=KSl+j=S>+&5U78MH32ltW z94JPx;D=lRoHT|*W5ViH_Ydca0X>(EJ=5^s4QfW;z91TrQo^fbMJM!+P#6NnA@vE} zdnb1cdfc~s&0xb|?S9t-bx+?#QSO+l&Y|z6jL-k|D^UO^jxp8hbG@M=Si(DonBk4y zd|nhjwlS)d{mC=^eACr5^ge~QFP{=+j4F8c-WB^q!lG?g(xPuNGyJJ)AfTt@@f0Yc zytnhl#Ms%tkRCrlEMl=X&<`GXkQ=bN{utj|< z-(+_+^(Tm04~cRJ0^vZke@J=UAOuCxuqZ77DzLtx>7BY`f(cO6N2V3;>0^tiigwTk zw7OR1tq?dU3PVO43!2ao*E%S(&@`>L7!6e8rj-b3w&q>pd^-AwtF()dYG{I1^=APB z*a+93kOT6*Fjm2k1ji_l!dD${a=biFfPM0w1A|6}iiA7vrvMw~D6F!YkYa4+&QiTeGIp8)6;%lTSK33b2IrPSg33)8~tzc8g#$6aH*-ySrW5Rp7sS7-W5 z>5VbsbxVDaGV7UYlQGZ`4A2ttr&I-S!k9-_Qdz#bZ{}ESW+m=eb?@fpndqLJZ<8_5 z>b^6l|MC2f0OD7k83@_C@2FJ1Mcf@6_WO}yTMKw@mEHSA?n>_;8cmW({+ucU9v0m} z!Z0Ky?K9_BqQ^2YN?sC|JAuBFIJY>j`iT>BrYdjD@Xz|?KWieudO~nCFWG*X#6ThE zC_gD$YJO2Q@X9lXYWy)F1EmZ=7%3bvLkH=VXV|=-ae_pfjbL>|#&HVOLwR3(G#2q0 zRhU8kS8@CQDK3x`h_tyS8B2uR!liU9yX$=pG#xQsj)jWAV8CLc{~zr7Ktq3qXEC?~ zmMB5X-eS_c?dY|s-=)$$8hBGrG0CFDaP3t1&s)Pijb?|H4Y^YpIz3|gH+m;tyx+#| zJfOva=yZ9|wx%7nHt>T^%72+VFBAXz^=|qDmld#8aE$0Ctnm$d7GNQzl~_u^EGu>QFMRh8?0Pl8_%3 zARIaEDRERQwpS17RE?{8+jJ~JBL^^nYX^W2y)-AwUl>o3(NWTVd%}aH!>TzEsIe9(nc;J16s-N&1xU>@d;gp9@6GUDblG~5#i#&Dg`UClO1Q~GsS@FY0ORO~cqV%F z1Y~Ta*;gBIvS&zIMtR?;W|f^l9_a*1EvTLri>o?Nd}A&+qu`NI*KgUPBIwsLs_m9% za#YGZ<`$J~WdT}H&k6h!DEfZ_MGEMXbkm#&|HBN2T#r%ZnvI(XqWF>Arz>`I-{B%c z|BSi*8DEK+G-O^x@+PZrJEWIV)lr<(P~NxTif;vfxk;vt0b~CrpU;Fb+65vgY&=CJ zRdU`OE#-`w_v7m#cQUTj@81y56>6H~>IIR-RMI^0CAutkHw;TISi4BMsFxv3gn|Y@ zy?1rJz8}=r3pXY7@S1LVNWJHIbHQea9I!+v@LU zF@on5yr3UeAvG9+N3n-ly3aP>nS;i*L ztYK4?g}dyaa6kGWi;Fnl`&8pUgByB7v6p>J80YVymC|>V1ykPKro1bAt7x9^wUS3J zX&v`NGf%?T23<^e;?BkoxX;1`E_AJ}Z?|@;(-W#?i!kOZao$f0Tp-rMp?mtXS~&%~ z$=(sc1l|%Fv&QPPeBSFpTVhdN=)Ppx;iV|no@>`yH^!f`^#^<3?E$x5*SGNFP@kpu z{*7Vumbh(^Z)`E+Pr(hLepQ{Bn}ooH&Lw?C0ZVg#TG+NwYnS|!}c>UZYA^WHu+erY&?2OEwoi42H{{=y6v=GgWjJ-JIq--9CP=8h(tY%8<4EWn4 zVq0bssOw#jE4VWl+?hsDwFInB$XeEl^DvBY4Em2f?yhA>u23Dft(iYzZ7d;VQV7&U zY}gKZNV+kczcZ162BaWY{YM?6yFaG93Ck6z+oX_B5(c%2*z&p)+8Th)-hf6ac?#2uy0TMo$*LJmsXFK6}n0?#^x{3~<{kFcyu7WJ z9~7ZxkVoAidpdQW8Mf*D!)juWm`H`MV$neWjP6sKh$(%^s#8Q~iLA}ppBl$SzQ_-v zo6$R-0)eU~s<|R8Z2~AksFAHJxHp)x7lNo->|XKi%3Zm<7nih`lZr7K?uUcp&8Sgj zNqgFJ-gz%I=ju0Wu3sk-c>UsQ(9adu9B?qF{b}@#U#33CS-#Szr}sf~`;uk9Y2=FzqUTXc%0_LW@rN*@ z0FE40%<+*trC+wf0tT}@br{`oad}g}{nUOE^KI6G$gt_KlHEOfu}kP=o9X_aFX{O=&z+ePeSdBIL`!;S_UWWA6zqP-hd%U5+Lc_@b)0$R`lsnF?WLQb zvKy1W{-ldU3i;IhtFY~Z@4rVrIoZeL9fB{vU8GN0Fps^i`C0f5^HOc+5iI}GgYLhz zksP|~TiG7c5z)KaEJ-tVPKR(1Sgm_o5tuMX8R>d(rxo{d^JKCM>@qQ^qf>a!KNBRM zs~`RP_KG1iT=`Y*>7DQ2ecH`-JYT+F{N2(p$qE`Q_4LtRqUIOr#-H`Y%1)Gj0=?i03XSO&8f?fnah0RBOu zk(^v%6`xGT6XPavn!gfCS4FJf9mdI;hUm)j^>vaBk9F5gj1w`;*%5LiXB&oS7TztDV>Pl9 zPDwLdYZTxe_8Wf;yew9V>bi|wJ5zq2!t&!z%NHYouR{?dF116`e7t_|tc6~{zMI;x@DbK#=wg_Ykx(sQ zEH>SMhp%(}p4XeunUYRg(emyx)bXeh)DE}^rQDMzps)CZ5mGOjcAtu*l^OO}NVXlZ zvJ%>0_03L*p_)oM{^c%zx9@lVTFjiQGZSra?}!i~FgWs5G1K4lu#Hu4KgS$%w5xmm zllrUC8A>Y6eed;)%PEUmRVs8u$aY0F`);RJ%Sfr5kdao|fVhKhas0KX5+jQWzJ@t^ z@k@D^@qUUYp%kt+2xtf5g+N8)I#ZFzkr{{U=T%5eYy delta 42963 zcmZU4V{;~8vvh21Y}Ee%MnVIh!HV2i`{KxAh$7nbu(DmT4Hsh?kqzq&2B9|HuI|t z?`vu^&)+1PXmiLUPqkak0bA&xc{(lkmiRyumrVfcibg`)I$e;3PXW7ndFk*wZv8iW zGjf(N`o<*#_4&(WyJsRwAM*5=0Z81AWe5o}x|3>s&rSMq{j{9+^zy`ByUM{MnM=gb zE=?K|?oOp!?a;SqZ-j&RCQNF;SE4M9J<^Ux7<5jXle~B!b${Ew0q;O;U0eS1_N*X= zvMInhKo?Xet&W+r^x)hBz0MIXqYzROyHSduc$A>`hvVQOr}ORb6`=cuJE}YOS43AW zV8nc4jmqtPX6Eq-zf&k&Y>wcpTb$qyBnc=A!ZD|tOI^#kMU&7FYyW{GcebFo(GdLb zS1NuSh2PstXKx)gTy#}Pef!|Eh_8!ioagvHgffHsc5qjcM9y7dQj8ATn%bhGUrB>n6- zQ1gyVYddI|*W~}3togpJQj;F-D-uhjrawGVo4i)M{rO>pHQYTi;~6`meRGYmiiAr~tj6xZR)O zg(^YqPKUxHV*jd;cJ!SR`)|6r8EYTL6N}fv}>kXU-)@0Dd zjj$^X6F8$foQI|dbv{4P+0Ts8ICQ*_o8Vk1Fbr5FWn<4`DH`Q{-YV3-kDA+4Qqpqo zuMZlv##@D+l3%YSHu;dMl7trotBsf~-9Cw+V$Tq%w@m{s4XR2&&}ok%2#p zU7W|m!*Qgnugtm~WNsXi+j51zGV1$H~i;Ew0d7kiR2uF)Q{=9S~%=}Pc; zCbs;`7LcdF<6=CgbbYL^lzprW&GWHW`Hc*EIB+$!1?DkBH(w|LFwKuAX4hRcV;k?} zPZFC&qM5u|V|uI)1^}gsGlXRqr65Uoa<;Wb)^xq!ZP2UihMBN|emAG)0S-H0{>o9F z3S7wXKf&^2JS))WkbUJem}iNW&3Jd;Q1;h#-|`5d1GpcvC1H=qLVQ>SFP?Ni8PuQA z5?dt-o)asxnL^1uww9+0K&eyeT%izekYub7E(aOp(bq_7KY(_ZBCtitjh2+-OjqWz z0d_@(c`yR?1_e#(iT?~q=8`e>WH7U}lt@Iafkd>eYVb8kgnZUaS~sqPMD3Z>mLh^1 zK~i3wE$|OKjr<}=eFv{ZB^{i{Hjxt^&1G(3b4=DRmSQ7F!W}}wYW^+TQcw-?DsN0o z_9Di}0Z1ESfU^ZgNRX6AW0&AOe1UXN&~YF&9!oH~t;>Mv0qBdHWWt9x8EnOIY^h6U z_SN%cEica4(wVWp>ya1J?Q4Wjl98pN2a-gi@3b_B_|$;<S7Fx3WvvXibqZ z`W7L=kYoVL#cQBb$sfVId%q6^&yMPQtnGu_zoHn-M57tR#DZhq2o5wru8=a5OKl&D z=waTNZ-L>JnGOS~GDXh&eR-wCchE&=eS&-=tX7hyFzFbRM{A$L^bkA`xTkpggGr+v z#~(irAXF7^=}N!wCmG*=BU)<4ech9WgCJws!shnOi;Uc8mUBpi72Axc2Ivm-kJ7I% zp0zO(x6VIalVx_$@z17_-iKJ4wCgg_$3|#`>O3fZUxm-pm5?&iyg&?5!Cb1rRt(mz z&2hDqSy7jAQJt*%WC(QW3SzbCitN&nbMzw)0NjbBzm9SbDZdn_z0Kk*9t11})L7D5 zToAhco~UmK)`(y8h92kQi#4jpdSO^L6}Uf~i+!(;1&|OWU1@4zFc*O8Q@LF*FkUr_ zbG^ViQ(^$K59ZdeV%tb$v2|J~G^9ljxecc|ExMEfq{>2?Ny?^qBf?^^OI)bKHa$Zb zC1rRs&r{vgXN5z6++=)eFfN{g;^lggKC53`l1VW2MkR$41e}M&eED2&ExSYsjOZih z)mur6bkg=!H{w*Px8hW~ul$P5hYD7Y0E0ZZ%$UZigmK z?m%Y77v$^e!#MP3#z7%;>#U;0bd4xA(ou$o$PknLDwsKCw?_LvIca$i4;|^D05x%! zP4e7HM$&n3_%~CZJgb0)MN%g`Tjt^Ld-DMwOsoAQ$@D!c;_04jAaj{B(IZ)h9Nqa`WofhL=rrdQ!Q)&zfs~D{fZyJGfMqSjOjf~A zbp>3sxmd<1@pYyrha#Mq;4-m6WgH%5|2`rJCYo^t)O}zY2aD4ILyKu8W~mYM8-GHv zyMPDkj&^qpQ_|m#yi9a=6=lDuA>w69Yi@5TL%hWOlYAN6+nZtTRfBv` z{K&<2k3^somp~zJ6Rr`h0ikBdD{0*1_=h(us|=?1&uV!6fwc3k&L0Z8zbstWqb8?* zJ~oU^OQNupjZg`3)U2Lolj9sO8jWWg5TWGJ$DCvCH1Mt~nk*`SjhXeo;XJru2HU|h zPJ(1GOFWFIXzImwEU6fDh~WHy%oeu0A+o8?vs9XGl>&6qc3aU%0h4~<*rK;MBsa(( zAEM`%Mfu9Zj=jy=7TjO3JfY|vG)5TnAEk_`Z6g(M8a~WHHm*KKGAeS+$PA}{X$#$} z1GrA<6Ob!%D^>A~NMBm}(u6f-8y(5C_z38B8L`Hz>HFWq)7phe_2|(JYLL*KB(~&z z!ii^Of-g&XLQ`z50kRWq6Lj`mnL^DL8eZ#6AiilCt&dPudE+aSkG4O;rh$33cwXdV z+U^1Etv`nGQ&pP;1HC_`ydN+>&v!apnjLV@-sa}FuulmQe+w}Sy%s4j}jc^P18EGNpAyHOt&-QsjTzV$52MCg#|mu?%d zv3soTWOEHSU@3f(Us#2hZiO3F)!>rt6P5Ocbq2e?uxW`C8Z7w><_-D0t9jrDY>K!ATizL38hF6Efpr1Tb{Ei@+Yy$ z^Mw_&oBDlm0bsJQS#C`~%dRpRm*}Gv57d4jLg~@y!KGi!ixh&m-l~i`x%(nX8m-Ho zQW3-SJ)X#O*-gm9;^roMs~J_y**j0^!f3oDDRioypOKUc?^H}9-WDklC9LuQk7L|N zLcY!i__D;hE3(S_OMdvGvc)@)PL2rg#>y(CW6 zEg?sB6@U&qW1^hQux-2+l{fle9^Iv9oO(mue>)D_<6#JzTcD zj2)~hDzE`et2VWNxfF7&4zY~2;TEx4=G%Y`7$euBA-@MXoj68WvFTH)wBMv@Z~kW9 z0kVdkSv}D`0`K<8Cs*^x_lM>Mhw~pcTNel~CWiK&oSKGlZ?jordzqQCv&vzJd`mSU z%D7RMd8HyWB6<`jpnRl{5N4`OJ(czBU(=Vfm_I>7m%xDJYaE~B#UN!I!MB!d>p}?>h3?MmlpGCN)Q2t00(4&7~$29{Nw+f@I9H#c`^KofBI1H=%I9 z7^T7~3#7iRGPvy`pQpl$KQtca*gqbad7eh^7;)f{e!@ni%l_L&mR`f^IG#&64Z3a9 zvy1`ltW1LAjI7&7G*iMHQx&$RQ&AQGKr?LFMLiscfh{;9^Omn8+$Gax|MQ^yk?~OK z^O|f)tfYaRzLVI=B=w&lH$;z4YlLY}K^B+d5D(MSYdi@xD>BkGD#qa$lEW>VT*mGQ z-?X}tm2|NPj%Q8}+})MNYEYYQ5G01WvQ%Dc``qRP9gXPXDN(Y-n)FArH#{-`!ixG9 zudbMf%1Mfub=iT6QB75>CO^vqz$|TcDy;cYqMuLzBZcIa^3Hr_D=ztt@q+ zeWbkRuEpC$F|;xZMn+41aTZ9%2%%6zAW~r)H=Qv@m<{a#9cl7$T>KTlQuhHx_BWZZ zL^?i^(6ye>^m_3EuZB;#SqBE|)e#0tcR;GT{XNi@Ldqb&6gB2HkY^YEMWo?^g0IVq zt0QXQ*ocfK^Ed}}*Xqa!c7elU5)cKDUUdx;7vlo$Tyew^(tM92mAiWliNa@;KD5<$ zI19Z?HASem*)r-EXk!LAG*zhEX>7bZB&eD>>d4Gj#{M~77hotgO1@7&cn!`eqdcv< zhL{%M5iZn~wMb)5yL-Se&gvM4!G%4yS;gCDtt61&->@FxfYwHCA+=#cZJE%zRsG#) zHDM~}xV6~c!ozeu6NBfHYiJKYTJ{<@gZhG1Ga4E*gBJ)~#o-9x?-Z|W*dCMfRy)4J zdYqhFO8IX2V7A$39G}Un+CjjKx}1h*ABEe>sT)aj?S*uE7iHvrhh^mUgi7NRYk)fS zpdElob7vt=&#hwo8rE`JF3NjR4Z&<-ZSnFF*)WSIf~9wGFSIx>-!`)VW>l6*Xwu~6 zG;B4Q)_2I`M*aa@Tqw?_$2>06pqte}sO7M4M&<2yYnp4%tY%$p(S6*$fHYuG#dQ%A zNHcvBxsrB#<2J4d;+{<2S8ZjTjac^#X%*oLto$y-{dYDclQ`*QZ>hcdfJPAW_$R6c zOXl-pD{PGH!3ktU?k#rA=jD*{9lkYPNU#w@5A!*RAc?tA7b~8aK`ZE#i>*s<_95V#2p@? z4(U$r4uf`?SE#svM~&BDUrOJelg*(|*Xe;vWvK$}&ILDIT=%_De1zWHdEgT3+LB*F zpIk3~^Zc9t?L~3(o0y_sJy*F6{FPeE)ot}S<`+63INL!qAbjY2aiXjsNbjk85-d~F z7~9G5R?0gBzmm0a14&szw1>^dvck67^$1Uj@w%hN1whOefft+9#SHC|pcd4gd*n{K zxEqsrAjDY(y=#3ovVxaR*~FMCy~w9oq9r;kX+;63B}g{A%+1`Z`7U}ka0Ze@qU@jQ!DK@5;C> z=HMMg!#}wHw}AuYZ_rWye*H8@p{IT+xiw{8nu0nt%B9Lw1GJWN>il06M}ixQc`I4$z$(L*aEK!4BYe!g4@`rA-O#UIg;N!dV3D zPfg`>NC-bx*0)~i^CoiyzZLs%H@^wXhKCRbe*i3uaJcVyd2%y^D)QFLumFs`j3#l@C6;?lQyRR z&i0h@x&j@rsYpN0+Rt+eH=;goz-ho90bU%8VX=tJ3?s?mN~=uoyeulEB*Z|bZmegil#Os^Sn87dvQxac`{Egj9PMKXRQ8`D?(brHm3y`c_1OiP#niOB1nux;4 zz+R=9QDoX*4U##ay77+B4)y^R$D&;x2*9d^EhFy&NqcYQKP4p&PjWoPj7{(m^i-(# zA5X(ZNCwi=*2a;tfUASx@jnzqF|xw3_y*ExnVxsh)`MmRJ+<2-9se6_I8ebN3#(0$ zGO|A9QqEL70%*J0U!)C)O^g?1j1fAx*B#p-otgpCre-J({nL(IWc5+6=>1X8li>|q zNOh^SC!G0`?u;@c!iNYF2^g0p1G=c0qC$5AHsgS0-9C0J zXA*0r{-k)}zU6o;koa_72C`rIQH!I|0j$N%UfmGYqKAM2OyA#~8DgQ$ibMN*gr{<0 zuQf?pu0t7dkih@kiJ zlOLFH0o0Trv!dg~S8>fL3{m!G z?&eWnaDW@=9x#VC@z_DgL-(>9+^P*LG72;;wN!U$04xO?XPGH|Z13N@P?>$GSDaMy zX@Sl6oIgJXE2E+RUjA|MyiGohrDGj zn+Kb+FPXPUUkc6KjnrbZ{CNlg^oX@Ln8d&otvjSq1O1u#^5nK}>)2oiX;cSb4?W@q zhp^we>!zT51LR!F43Q{|kcJdz{AnI0kUel})qKCshj6@MIV4^}8=s{RJJC?c(@5P2 z0LBZ*)RQMZ(`t*tJ%mNaBxBmvy*38ATW8IH_|A7!LyaCUiN9lj>LZW9>n@9yW-v9f{dqt|fpxu36SZ=fw570!y zVfz&GElKc2cE)02Q6DbE4|k&kF%y^G2j^c%{Jh7;ul?vLUp3moVY(xVVYKX@Wr~Z;)()U8 z=eWEz7;VHx2ly(@SCgKSzLs}8{9kb1Fu7UVvp!{RYolbJU9>d-)O{^HWIzeO3T?#v zYRLTHbF8Jy_Z8=g>w%MGpOr>)|GI*$7hZPFi;J!F0evF8TXxXotc|%&UKbYZC!Hg| z@e+R7)}0z!AMGySGuO3vE*cNQ^2^xO)fP=xStOXuDE`N;p-+U4!AFsV(UKX}1q@bGDOuTM|Q`G?O`;*d$smER(|C`^VTeHxuK}oa1xz7UUMR@B4S1Gijc(PJ@O9P zTJjACBLPN2sadW(+9L*GF}d{GVacl^&sxhT=oe#wG2%O?n*fx?4Lzv+84etdae!fqJ_{whKHw(9U5NQRt&R zAK|at*>O~RNQyiE2e9U=cw1R$iCu7IG9|}yb6QoR?jFKU`9!eu3=O*rvcHF5*`9eD z`$yggd%MwA!mMyoJ?m>r`I8;i7|qhV2bHW_dk;$H=y38jab?!ufd(u>d)sK1s$9%U zf#kOhjYZtV%xKmVgVdFr+V3V5yKxrv79ZVu0|gj#c=wRnHh`m2p`M<})^p_FWcE?# zARrc%fCsp4rMtkenA9l%`h3-EhqPJx01I8Rxv0yHm%osm(8y%?vET>hU5x(H?-|AB z*PHK@Ym203wenNti|WmJ?bCa2au6xoRirS+vE8-M-G8k+^WUM|xj{6gEeP(7WKrX_Zz&Eyc% z@ zjb7_MX&t|U#egrTBsXa65t@if=PfPl{5_V@kWR@~ndSE|EiJY5*DD-aZik&!CSCxY zH&ZZc8QVVV=jNGC5+7K*?V(nwgGs=}X6;nd%r?5*sEux>b>h*sh-v^c;J_mA< z4$>lzOHrdXP=|?yznW=M4U14GSNZ_oXf&z&hLNUgI>SyLtoqccSxU58#{Zx$Rx34{ zLfZ#gM(+OMSiBHLidtJm{M>wN`Y5S@#ba$N3UK-xGU2r5xNZYd*qHZqG3R+UJheRU z)c#%yL%L0)VYc^EP-w>((~ZcnFMNvXD(Gi6|3761{hzXPUl5;_5(B_zlwJ7#5LKVz zdgoXn)7$5_!2Wx8%NbT}y>{~Vxx(o)iHX#T;M-xvKP(&DA-Ej4pOh_eC?Ef%Q-Z4& zY-~RL4dN~PZ{h=&oF_09Ry}+9SaA7u(djnq{%uIC?<%XZK^efJE4xwc<<(`AvRBn9 zZ=%NB7tAu4r7DfS$Os@P&m_>#?qK5r76i}kc02e<04-fh25<@@3&&^oQNF}p?^tCb z;)fQ2oz}t^ATgfHR`auB7`%S! z4U}e#KsPj=5yUpSkV_L_KFLe}BuKK_H$`|Uo@Wh$TQ;_XS0q@D&j62o{Ec^vtpe|r z7PXf?kFP-?wz@g>7zo%0b#M}g>=Hlv7)~mZI6u+tx@Hr`^u-&=m-Z6V$3cgn$zjm5 z9=it`6?i0wq5!)TT7r`$ro4{^XeOXcJ@^KzEm_kO3}dj>HfFf)85?~oHnpuE>| zm!R|jc#N|@w6@^u*;yU?1DD%D#MDSXHMLHi(|czO^ymCRa7$2Z|IH;{YNJE}YBB?~ zyo(c#8yTI9b_asCV%b(owH-!d%9i2;)G27m&Ikp-;ZR%gmFBLp_i8SQX$M2k@G?}N zTDSWKq-u(hP6Qu8PjtCqBn3{A=P7!-c#L5S@P)8}G7i*lnJPs%6K?+ktcxF3tfe5I z3VkZC95QS{=2x^MpNgs%s!AK^lGx%V`vX6dzk$nQ9dkjngapK8@ZNZkBjA)KGAZ|5 zpAI*af(^Yoe4veacv4AV)(LK)GRK0EYt+Vz!Z{<9$+(4JCXLkZV`51-k7O*hgUaFn z9yt*Zxwr-T<(KBChTDkzEfZQQKxO)O27BRHe$>Utn~ZpTQ*&|Z>zxJ1&f3sa9VMQf z{A6^Ri>E!{bbMQ**P8lim3idjfc+%e8B&@MGTNE26tVA)$b1~zzU{<%+)o8KJnHQY zHJgssK8>zn7H(h1L7%_$0^4PPm@%gTXhtFe(ZQBpzwd0)KL?$AAk#Kv>&&E+uX!b+ zQ2I9)%2|4(Gl47bpf?IS5;qE>DNiS7T1)V}l3CmWZwL$!^2vZGzqM>Y#mVl6=-BT8@6hid*ZD^kxWo+4KN*~vopzw% zb1WC#;Hx3u!G3QR@oA?BF<2VFVrgHFQL&N8ooRqdWoMx;%OAWu0m zbWc+wZ&d%$eI#~sCJOnhD651h)|gcpEmw2D)!!EQUTdYxXOc^iUU9CxjzR8v+wds1xkr1K%;XM z1n1~@60vn6;1|*y$xA=!_DyJcf4H$h-^9GCPSv}G(s6-3$q6;-V*b(gf(tN--No@-7!)dmhs37#x)a-JY3+&ExGLCa5k+fj#jjWRO$ZI;`hZTIjjzqUg(7*s! zxq7?biSe18jztHECS!NPA}U(KigeQnEZGhwz+o+)e3-yTvTIUy~%{wF=NaL+e+gUe7a&$h@59}S)gqxMz6Ye6x9r{ z2CmjZ>4*yHnwUyem(IUxwn)(IT=;;!y`i-s3|Hv>A`CS06G#{%$1F*Rgo}$c;jS)B zWX-t{BAWutP#B>&ivtOuR&vfq%8?k#*H`uE9+PW()Z1aSC6!5YiIVda44bOAv|(S$ zRWN6hgt`WqLBT%cHp0+IfkWjaTBcza*HhYeUJbUAnq^$_01Y<-!TwmQj*D5CZ0FR zyi;9FW^X6&-Ins-$iQtv4h;zXnylZ|9`- zYo9TtzS{G{i^c^S%zETq^eL{zlf>Q*C)x^W*IHe;+oJnU0k5BjuRnSaK-AFAYzDTn z_~(EqQGe?y(7GwUhFq)FqV08F9euvRzA<^BvfYf51_dtR!Fc6z?4=?6)OT8 zq^S-a)_eBJ#H01<;VM0cDVVE(3Ry87;*{nI_rur%K9Y@9Kw{GZi3)<_vH2T;G9SFV zn$a0&2zD8IY*Zn!;adibr|7`}#AZD;+9{P&eoo2$_)Y$oR@`+7!btO;sgB9qTm#@a zfyFCSzfr)DlHMF&E|1~{Le0qdE^;0?R5(n(qh35U5YvMcD8P5;X=#Lc=K37TO*!FM zpHoO`Wmi}_AI+zP%x;~hemYD%7<}!7L~@U3Q0=f`uq+Ak@i4AI#Uk;mdd1-u2$juk z;7oC5fBh18j;z~-lDC%;kD=bPfgYeIePf&msUyMk+7a7yxLiZtm7deO`~_ELBVx9d z(AKH8pUVyBChT>AAyk7Kg5+>Ej#IhhfFLhDD{) zH#5j9KIjiJ zQC3X6R?DgLoD1f_{tC53Z|0XJNEyxAYB&c*7VUbnu9i|K=3JPE^bKoS|G!vT6g<6& z{zM_T=;9eR1mo@E?Je(DVM|u2H@Eb(9v7*;L^Z8j-mVf254To-7+B+B8=Y=mbAkE3 zV3C4Nj}k+ciUbCyz#+r94_mV0OBm^MrLYAkCgnsH{% zR1BFf_%=s&tm|XHQoal$B*w*w^UEHgHX}KD-WTF3(M{TiDhm&#bq0X31RkFQF;qv^ zwE|WP2g$o)o7bw5TYRrme}q-Xux`K{%~Hj9;<|U&&*Hv1wm2q}%DUi=SnKZrrjBk@ zYaQaCdipFTO%G}L@i_Lp?y~@6K#}?z^dx$~;rnmPDaaqK5~gG4>E-3)zoUoQV(QrE z@mMpzzdl44VU4}^{4)W}7>>6(41JPvk}@kEDUA3%YH*8BK_gFRou^N%NK=t!(rGsM z0SqY9vCAULLOPEqH~wfsr77K{vYnS%85(*tMy8KuuQLAsJYAJ@TtY*_3_)s`_LdCt z1IEtn0b*N(#-crcH)=N4ib-5*1}5t)A|&S+yiA1pl4%{Oq%FWL1TEk;Wb!RUB!7fw zSBxmNbB0*TTiHHy%4ng*>DCQWmbU2+XWD09*EgFu~Eu z{cn;reONV$LN?x8Q5I&Bp)%zYc~i;FIy9Rau+7O1W{35hcP;kB0nY_azM?QUjJ!oZ zDP`;4+#XkKnHE5BWf7*#{)62KKl*xZmTb@hi{Zq3G=D72$>o9U%+Y{tMR)P)6^)3DrC{v zE(VGt9?Cu2jMRoW*MOD}MZQtagOt~E;aqV0ys5OcY!W~|pJKRn(?c|Au@CAx8%nhq z%3Jc_3&xl*s2-~a$!|n;ujM$@y~nB&`zFPjUd;a2)iqGNH4|aNS-n|WEIV+Sl$541 zQ^R2RI3UR-ot%fD%+^CbT%%S}+JZ0l=+|7Qjb6D(1CAK84>Ms(l&(*VkS1blBCm^5 z0#+9xV+0`mjHO-Z=t6Tuu-YdEHngTp{Od;UC^P8q?jyp}HK~t7kz_WATG>7O3X)Va znE)PmKA+L$SgrmQFIT~sr=HFqxuK^~agfh2mN~OPhMJ>JuOn*gn$Jhch*DA29;hYJl=FsHBpJ zp4uG3LXmi@&W2c#k0oY4XjaC!c8lz>B+mnOH|VqvsrYWG>jI3y#z!FALiQVGQSds!(r8uD9~25W89k$p+!&_chtkVuP+3*2G! z{WpRTdS5wcF=V|=b_zZi4;!?ZRN>9I9zS44!w!i))vuHj216x$Ba#`4;xm=xPjivt z_TwhvDrPg=5mYXTsz*ID)(kuoPR}CU6Vrv$ZgCA2l^3WQMmkHiI%bF@@(c?Wo#kS{ z`+J6exZc{`UXyNEe6q`teF1Ty$zM_531A$Ffy8Xmj65+OATpXtur{@dKrht*Dpf!~ zQH8)6uwP=I8fNkK!yqJf^O*F)0OA@VO1^64V`lU=H5Y`IwO%fIwTm6r%Uv>#4Iv8979E@p#j+_1=? zeP;9Cr%^T(rF*)YN$n7U$L1P+St0;;bdh8$TndnGe$S+DRId*&{wWl+v=v6F=%-E- zi9hEGdPYLNQ{GpV@9T&|dL%uT7SBMWFZ69H*rT>^6D`a)=i*Fk)qBsP*y^TqREJ+5 zR)KEF4@wT~?kh!<_eq&J3eGP%z?8^@2J=t7 z-yIl7(Fgs;Iu zB7Ivqt~bZZhYj4&qP7(R5CXGtC<(*BW@zn>2GBhSyU*5#zO z1A8Bj!@;U6n+2<@49B2ZC=n;HDhJwlw7<#NaOK3XIXt5|O&uWNz=P0M&93PrN#Eq@ z?X!|M1kzcPBOigpyV(vSpgkGUAQfGMXE0*noN&`kN15t(S^>|XZKV9jb85z}#?6DD zKXygqY2hJ#z8`U-ku$u+{OMhwoS=hYZ*|OKoXdxYpXlK1XJG9l7QGA^U|MKR@z5&V(j`Do2WaWlrau+ofS;qCK?^(Q!A}(3eE8fC7T^VaBPvKYhx?)zHJ8o44!a8gB(V z!pG5iLyI29m?GAtCsTxFBG>(vBkIV#*EIIWOH)q@2+_(!y-odF=G`x8hsRdxr4G$c zO$XC|w>W@gdwdJFPbQM{?4PwNJv%5F-0KUQNz-o-WAJh?hZI891dZRgsLMKG%J>hYE2!UTbVBy z$!1g+5njc;;VG$YWwFHKT<<{SGm58>FCz%`z6%%^2Ng8GKH$ZMkD|XhX5bFRv+TV1 z>!iCHgajUrctyrwx^_J4$3Mt{pJ(|NE4Atk}?v(HKF)U5x(4%Mt%R^&BB-$0pSto^@#v!KL* z7j~vdw`!-ks&+Ew>cikrc%B~D^bd?-0M|p70!5wTZzIMn)vBQsvXuTBf%=ZV$WhB2 z_At040$&k+Ez!aRRJGi?xas756_}}@rI==2fnOtOu&7~Z5JyhxE3XFs5)3z_Mf){e zVm!_uVX}Fs=C4it2n45AQM9GV&?BKIlU$wxbP=IgF&#q+>{>RzLy8ah5Ozj20rZS( zV0Y(%ZJ%v&pT%tbK~C|9o^_tLnH~2xRK<(3s*Y@fE)+oprt2tDkV~1-VxjDcw$2tb z#>2t0`^d&NlpO^K*YyRUMF@1Z+_|xc0v!ZQQmz=Gg-kFRA5V>8p1s$31XP|l11|}>MY=de>`dnk9o#_ks2HeW>Bi(tWWCje*lZSP}4_E;-E&8_I^pO(~>rcN6J0RZZ zM0}n{Gx1l%TPT}n3L+g3fNH-72zS6;vtt)6uWb5k?YgP~_5J6)=@2PYRGDS6C&T(} za;*(k@%SqT+gyT6vmiGR-I~U2U$m)6;RVI2Ve2=vF=zp!AmJNlu;5H7TgU`L^Q zPU$o2m*{q;VNeYk;8|omR3gQZFC`%7H}E|8ty|Q_A>B(dWYHN3jv|lHTvFswRpgO) zxkPNKgdcVdlwcxszvj|J zQdFYXsnt^~gh0O3?Mihy#`E7DI^~KqW$)JXx+3Cv+2-0a0H~>jF{K1k7{$GvsfKei zwQ)?1YM58@RoD{dLOa>B1lnop5=oh(U5 z6d!G+Yn_k{Mej^s0^QmN8!BQK=wxT&IEmRMfBqa0?4$5Q)#Ap){#PmWTe?U=aEu8;YmNMbO z*?y%1)`wPWSXLq|Ra|{+?E;rDnfHiO#D@K&RIp{C+lsY@u{C^MIdQRDm>3~TP4igx zoSv=BgEmy3XHM94H&BJJ9NL|pjk5XonBy;wkAN}Q6Hx#SRXwF3yi}z%nOf74Cf0pq z!Cq@!BVvG?)IsYy9}#K9p}4Slkf6jd6oB|{sJ_^VYmk6#%}SI+!zN^0oJ z*{s}cC9CIAS63pIrPy{;mw40*+^6t=1iV`9B)c_EU+u~xml!DA?F*V%WD6^Ow{bUw z7x67hnKeL!ay= zn-x0&(iJfyoNaPd>J?mV#Rpu?2N@oVR51d1P)7XP=Y$s0rz_Rx=$*hbbgCDrNj$CT zO$HEmb7z+uzlJ7RU5y$v<5+k9N37A8gZ)Sa@L^OU@%_7#)H(gSaq z-U>~VM89`}xEIO=s3SiDItrz(AM6EZM7MbQ9k*(Zb38q<7K#dmApavae{M%I#EmyZ zEmE3>y1J}lu%~qGrLWFacs|*%o1*9eZ7W!fhRF6<>vLL`D@yN^ULEK^NxsB;<^-*4 z5v^-L2+x2ykDOumOn26Q{%e5ANeWO3ovxPrgrk~+^$@{_5BvbR6ptkperk25wDqP<8L1VKt~q12kq>j+$e2Uu_^vfspBf!`2T<(OV1r7t#cZpQoOIMV_& z5B{8l>bM*XxLToQv^juBzy=kO(MyE5#SVcNEO=E|z0ld_uXPwz0+mHNRt$w$uUe56 zSLe3X)l*?)w-GjH%ILH`4Tbd2BV!bQ@c14GmHut|7ErXEoI@2|4qXx8Q9c-Wz8rv; zQeO$l@+S2SWOfwvg2|K7iO)P1iGT%^dn>3`_1IZ_^eLfjwy1uEN5HDzqHGbGspjpS z{2$74Rmv6rURScnThe$?N7KJF@!f&O_SDMn%!caOd*58aw|^J#gXW`X(&&xa#FFG-DJ*|E>ze+c`|aJarNTqQaQi8gxg5{yoCLWqc9 z^b(BTdyt79y^PKXQ4+@Jz4va^A$mmbElRZ9@%!KBe!U-NACI%!TJQU=wa=M-&ODBO zTLG~irM|)74X+40v}p_Zp@iPB#xTBsZN+y?(w?yB?^D5^^R~GM#v#n|yHEP~UBj9Q z`Ex4Y;eFc2-BKzBjA>O0!7Ti48^1dqj|`Ue>k|rWUt+u`_>23f1E-A```?L(U&QX; z1$MgA6GHMI?ROHKl=4oxCGY#4rLkYDocTsWBFH>LIHhBJt{T`z2FYx0Tk)yXWv7zE zh`hxC9wCa`fP{XDDp(0uMThxQspr?UobXR7mpLlAtVCv1HOmK{#$QZ+-+XB=tO_J( z>u>+Quc(7n`to$?*ZZa%PqGn3!232wbATA89PjfrQ&y+jE|$<)Z2N|La+9v?oc))t zAge;mX6--1{gzdi{(>YTU9+C;h_&>yoz~WsuAIs;=&B z&9VCBoJ*w5Cw{!$wg0PeoUGS8-k!gf7bM$sa2eJs&BaypWpafJ(^?}$|M;`;vx1D{ z(2UPUO!J>^Lw>{v!*Go_ z-d!bxy_md20(Pd0hMDbjCOZ*(&qI_4dG9wRz%KqZ`H%ha~O|YisCP_cah{DkJb2z$q!c5|M+3ALUv;uk9tR@ z%~yqT2UnsXsA>>Hcfs7r9cjRq{LjY&w9sJ%WD@$}4>(Nh8xo97=@lpQj_m&OWc+w; zaCDtS5l6?kN2l^<%ZN1dzsr9RQr&k0jQ`Fcc_%`Y3=DRa!XqF4R-+vWu9U{y$m&x! z6Xai*#?5>=?$j+5zH(mUP-~XMf+Wcl!pR;P$q-5X;W(OHmBbg{h(W88CK|&RpE#pl zdzzkCV~W&a$hY(x$(7a7p-73beoW%@Bd1zgrG4}?{ru$*Lx$=`pDJmHN+v;-v6B^qeoWiVBaU5(b=DiuM8Mj*96dw=7~YNjI2)K zXQ+Mw|IX=~hvfEV$|>g7NX&@QBOiAN7h9)ROH|_3Ouwq7@=X80bvnE(K9N$Lb=OkA z$<9nrW3(b7TtYy1iNS;-hex^P% zD?cRFBW(abMEM<_KBlBFB1lwtbTfHiv9z(Ojumx~a*jx57fN#}8+Zz^gQ@PVPqPb6 zN=kUd#&m)QenyoUWfpVuT%Xbc75W+`$b=jg-uZ=&T#F770ss+-6hf0$4JCvCJE81K zr9N|3tI{_6Ctl(>3sEBDBnM!1s~B2(W=J&>BjweAqiY;rV&Y5`x?mRqnXs_FNd-?2 z={r0idmon{p~f}<~3RP(_qHm1gmXJ+8X zgX3ajB1MF9Z5%zsEBH_}Kj+68a;jDOT1IIyYF|?8AQI;mr3_eE4<4Zl1SAZOKo8v2 z=wLN&KmO<5)!eqOsJO_EW~jqnJ9fm$kSxnzTV%)618AK%QfNR{rK(bt_ZnK5sEoEw z!Xq(eFxrZELC4dCemm8U+At)_YQ%^?$zUOIS6OJ!h&$Tq`HJyFUCG3mL3GsuFZX2@ z0F+{o7#G_}EZp@Gq~1xrx~52HFkQ@$y}FJu1#TsJ@h)|6Kr4n1nyAw4l`^QBzZzgI z#K8*{yvYQ2tc7380V8%t_5N>*e<{&cGk;ZP22|fG&kPLEWbBnCCSJT|wYSpHW6ifq zMJ$BmeM>JYqj8TLd1D~7AtD5qw$Lj}sO+iH(PR;!G_djj4Im?o=S}LvT%j8yB0?h{1aBbN zx}b=rq9+rTBDt9zH)Kfv`6ggDTi>sN)Y25bn97z^yTl@(ve zf#e2%hI>^%1c$GsN05!vbgwo@;)dX)4!Wq4-fSASENGqla!M_1GScQu-6T`tr3`x0 ztLKT1I_jw!c78F9L(eD>8SZ*XC_%Jhf1(wO5=1LjpFUsxY(oPjD5}^0IpIW{# z+qWXq)q4Fu>ka6u>+p0VHdFUurE!Vg!RTV93#HeO*6kun5F&ci@620*UZ(zx9ikR3 z5$|h5h3~_Sv|AZ(=Jkdq)762@R(3HXvMFdx)zR~F-<+hjDEl1sUf1adF)>OQ_oIis zrSpv81LZb{uJwj?(}%@CR$ehRnrO0}?o!*qvg;-Lz-yL{%Cr2V)8r^&-9kZ>;62E|@{7m5MHg?63N1c+e4oqG{s2BzM?cA&B#Zl;8Q-0=4CR}Z}bV&`fFaNv^?y#dEVSZTt8 z>(FDOUH1_37Ab^mMQlHsRUhPuXgT&WJbv@uK|AvQuh)^!WAPItOTrZItHZH-D{3H{ z@&OU>=qSt*4RQM0*(oUGG%I}|FyUJ7{#sk z%?KZDad*u-s%Xb=hlO=xRrFu8uophoP4a%D>jiVR)!}&m0mPBvqlSCPKsM;glz|`h_!Z>QeQAicHZGv&D98mf?!58+vVCFfOs()>gB~L@ zaFF#MKxm3-f+NI8&=x9>ay51m`azTG00r>izJXb;8A$f(i3^R^4GJ56Q=*dd&eHJQ z$bEy0BAmAx=kQo`LN6CfY@u$7SuyX1prB<5%4I3f!5W>Rf@Uj@G$rpe4aje81?ULo zv;cz$ai*14tCi>=wayD7k-@7rwLWicAvOMw!0*SJ&*#|I795i@zzGX%&uk(%0Mi7v zN%FO>HE)D|P^LO3?Z9!>U?fyw6Md3)B#9-(YDJ?wEJ(J?X*)WOOsmI;C^;xGGVhyJ zag`b_`*7DAA;sPs-18E z$T^2a2d^sC!bQ~9^b=UYaC^I?6NKoX#_51dQU-1$2VgNu9!gvizGj~!kE$?+j%W$m zPWgQ(4tqait9T%zUNrHPy$Pc#%jz319(ukHU2|ltIVZ^vYAZvaF^rj3OERxSf~V&^Ip*K z+{tyq#UZ)|kBfP%2LgW>mTYGM2&tC{7{QkOh(x4%CqpdoO$u-anXwej((cCLC79>7 zNlFM%U=Ha6BnxE30x8)?v}1w7qR51NIcnuTg74X;HE!XCPT6xL%#q8D@phuQB{2*5UTl2APtUWlHWO}egfAaL3de&o#NUvsTlrL;Ce zro(5OEguw&Rp`4$Oqd)-Yh&w}eImoxQJAQJzmoyC%g)oIwc!X_7ei`5mDY$42hrn% zbtj4-b6xc-+`wEkHHQV3g*DYPmEKiLeHkk70^O`nE8@3dUg)kr%5doPyaJ}un!>8$mO2Gz5^$K0VA{qRds~6v*k&ka}sM>-Fzqs z)_x?QQ{gpxfzZ5drQ=qNW3306GxWAE!_FH$^TN-4fN7%s1@VxN7AJifZO#cr)sqQ6 zeENophPo*6DKI@*M3O(Th$iG242TC(&Go>=d;hLhXHFud#Cit_knl$04^LL>nEENlNS?fkH&P*ZsFJTOOuqQ@ z41k=te*q#H5^QlYlhNjyu&AcalYF{NnwY5Tg{(1nrIfLrqYzuAv@+7EOSYZWS*erE zr+YD)Yh}>nq7pmW7Jf7O=YDM)0dV=MJ1LfL=yVQKN?O^K8hsiFe5F;H>w=3zNSORP z5XhxPd6@OiawR5d#&2~_YEWBSBoJ7FYRgb#`U?pJPO(jhg-8i)XFYfbT^sG8EI<_Q z8+!G^*o)wr_sH4>u_@b>rzV(lfxwBvkTw80AG-3&NaE#au6=R7iwa2~Fe$rshI+EJ z_9V0|qi86@b7f@x+kkVHnu&&OmPDw6N`8X;0`cqgmO22pRu>;iH2oPblL?hD}G$K#?Q+L%s2F6y{Cv_`v6oq&O>OeR}cUM=qnLMxD{^PUV* zqDb-NMav& z11DPa(U@sK8q!M=<<@c1IZSkJtj#SM;Ib1X2Hl6!(8VM+X%i6ol+oxyzoVI-Z; z8gWVLYZ{wBdfaNdp4=~rnn|VJRac?HG#Fh=Az&7A`)_KwOR%M%DAWy!O7LXw|ePa+c$wV;A2^H z1WFsS&R`n0Ca|k4t_ko5ZCt4Xu2*-k2~cq~w}FVr-07z+2fJ>!$^Z3^QDINoV|2$A zxen;+uh4y$6vjcyGFEdm`up(i^)IGn*?XisM}>8fY|Y?xq;jGIubP?qc7{Z#Wca)O zTpL^p4e(fu5>B zlJ6LC{BQ5xE>-^4%Nbjz68p?mJ=6hI>lAe`B|HHU{VyefUHKkTk`p=)q&jCAQ1NR< zcnMVs4jF8iC_4d1Q~7*YC;s9L3#)mK(#f|5mjRMFx!y+kjOI!kG&qF!+l;ERCZ7%^ zu)GifO^I|PWE{XxRqwmOMQY!iaOq{lOY=&de|s-_DS)4bB1l8KLtVBYvGK`4JpWD+HsA8kAx+(BEs7AJcf>tI=cw z{`k(yXSAS<13euDcFT8zXqSp;fM;VANUC*aW9XZ~HEgRC>Y{x_lhNLzKcNd=bkdb; zm^!P081e4mFda=0O2pFWWw0ax`#;iM;Wy}RjG2SV0CeigC~fOmG;Ir;!QtP+q~L=z zsgqp1P@p$E#IpbUsCXXsBwaLynBCq^RwY8l>A}pk2Qyu8#jMB09?ZP{&rGogGd+V5 zTBg3=W5CY!ZbcJ3>`hozQa0bz@en{R{c*{VnG9AJm^@1R)!v9Jh-0L(uj-ucIjK!( z-x_$H$7T`%Y^GBFw5758L1AlznH(bYIE+}*Myt!p8)=&?pzeiiIIb_1bjZfH(ta)8 z`sWO643%PZ!_Tl*V0l@C%+-_IBjyQTbArkDR`~8|m`SKkg-M$Frt0~zBHruPq9d$Y zPu{w-c#qvTugzbC1U1NCRy-xCJVrKJwR}sG)z1w>>onM4l=~DnQHC~|{15|gmdd{r z!@{pUk#uN@BLUN#KhPXN=E@i}c=W}>fJ%HikTo-L4wsUhY&Qg2tyqIc*%=&>NWdcW zAz?+Z?xC|#h56Qqmays)f!d727N6go`<6z&?rTq+bv^Er!Vu&`aLtzA8c9TVj1I0I zU@mjX?{i6Pgw|}@N$d=gQ+g86{p`FQWQH7`sYVXSlk$#_>iF!?ofbYOg5EuU$1w$LP3|LH-Kl;xo*n?3=XrmI2 z=&DH+_iJXg2fh7mXb{@pc$w;7N4}u{^RVVPBL0Y8bNJxus?)au6)~p(fSCcdnx{!! zQsdh&yOL}*^osID)c$hDG%9pC1N(```{Ejp%z z4p?sMNv*)O#ks|qSL6!dlmz@IXkJmEc@@0siw02u4I+^Iv9@P;WNe+v>kCbt^XUwg z|I&sBI&Ij(9?}K^D4u~%8$)P^@DtS|W0Z+}^s2|hTP}lJz0YfqHHK9X4VFRYi`aan ztr4>Spa~4VF7VwV-i2CA*@T~lD%sHcuO@>BCqYxSY;m<_aGG1CAJ_sxp5EOAsSPZK!& zI!ucH8ZReo15$OKgV1{WR=5c~Ukpw%0S5b&M~x(L1PZ2<*q@dpQ?-|J!rQPShf=?% ztz44Hg8fdUJ1pYoKisZ;Uvu;hr1%_oe=9t?2%qg%hFqUUR8NVet+|!1%5?2qLD}^g zz)Orbz?D0>3l~fOjKGFQmK&P7ewUpQE9-Xxf<8C z2zfi^GaJj_W@XEvZ`U@j+k7jpuZKC|U%QN}c4dV@|D=yx9>p*=YenL)5f(#-uPz~Z z5y9MR{P*A+N=9U1MJD0K`YzN$#rV#X?0E|#(yZ+g{!r8BM_D zW|~|Nq#Cu-m1Y$9$&DW4k9B&fjOpDU#ebU4%Po%`>K5E$-9P!wKzhph;!lR*UygAtKKlFY4!(XsIMA+ zwA6mLV5Zq^<=WWnNW2;VAuBuV?jPGd8x_Cvgpl#9cnMoHi8F%k#q!*V|NWYq+auMv zii~)&i8?<=z2$X~0`8+ODUAPh|AE|J-`*TkD)`;)-u~^K+ep8*|H(ydzmruVPAps7 zax=NL)!o8ie5bC*r|lXf%++>7j~RTapk;OT z7J?D@ME+|}Nolce29BS-e#+pnQEvhCd+(lpiW&wtW`SUhBM;&~%Q#AhDtLHnd9Z1z zOjFKvgbQ?aiGZ#Gy`3^ej)adhf`guz=XRZz7H?!=2ixoC6t`81#kAI=o{y1NBzH_w z8;YjS0I=UYQGS+cqo0#vm=hFB^kR(jBTrIIRoOW}E*)z%;k_tczfMpmp1GD~*vvHH zXo!MH>A2?@i&HkVGJu5>=e@}*TFfps{G04k3Go6#Vi}AOT;7@@J=?W?JYxPMFbNN3G( zg;Ci`OLr^|Hy!2%g>XN9EMe{8szE^)fPu&VUpH(5I^9^KkDY9*e&b}~q$io!y2s

n0x;7mH`UK6``nM>}3M!f+8im}31Oz)Z{0Y^4J62fNmOlsB=UGdj?^MV~>%%A2K^cc*``qJ6HN^jVKh;O<#*@Fu zZhF7Mb$R*=q4dJ_x1_>lM}q%aNAI+s8y&w*M?#((7d~4;(q9!sr7X1>m`(Q(SfejQ6z`&tDfAKU8HKXk|>RbLTA`fo>`DDWz8Ly53wL&nu-`W$GT`gFB5 z&&zV2x4QUT3xF(aS#UUY<8yqj7Uu-!eopYItFxNX&^;q8NC?Sk#gF{j#rLBU<||^GHnKD% z=OZzjNgc9*y}@Yrc?#hpQ30XHjsv!*@6F7J2Lc#H_+MLw8l&sm$mRz&Z2Z&mUv;LI zEr;?OKkyA?6d<&LRE)(ih8Pjqc-kmRMc~4sc@~@eY8RPsP1ijf-A1OHlp)wju4NYL zWK(n9l$kFe9G~xEsVK!R9LU&X9@>;vHZ=IE*VdqfawUM7gUeyX^FaMxpO(-VO{5Ix+ zuLJwNcE!HZAA2tien!0Er6OR&1@Al}q`I-KIf=kXhZt|dDBc>v5Kd-7HL-9P=7~iA zJ0SIAO7!j{{|?B`QzuftbJfjH>}p>_dFlzCE`W8Df7W+sK6(}M{HfdMrg1*M!moN` z`)?`{+PEvkHd+NPQRjy+p?2CF<8Qlqzri6YD)T#hh1V<+pnn&SzJ+{c6|R&pU=S1wo=wXREm7+ z2MuomL?I9kA<|IxQ!rP8g?CiLt3HwFZaAQX>ZW~dUV20R`!{OXs^p0fx?p-eK`Xxfi5Ii!Qq{P3UD-t%KzTC`_p#|D~d(x@qT^YAs8|@kEztGw94Rf;+i|iRdr`56Cv?uz z68WN1SK=OxTBlx_dxU|wG=;iFg+Xj900PhGlqaN+M|$i+wvNQVKB*`ofIu`p%#8yE z?TS@HggXUy71gbjK89UjsDJ}zH6+%XJ6(YW)iM_3AFQV6G$-Z%uG+{Fu_PIrk}Gj>tu4an>(Dtm*OAVZHGMg*2h*vgA^g(2PP(7; zGDLMmeS@g$WJbM>CFyAb*|Y%~{r70ikMF}f{?c*TrRae`AwF5XK#AYZ7Trl^L>f00 zewGVLLmvjt*h5oSh&Vib4+@#Sk<1YdMBkeNWgw_+(U`84;EW0vDv?D~YL#EJeW&BW z9od>x>(hx`pljfb1oLDiDWjqha-*1aPtF*rpNfdazTq55(+$K;637Xc)hq^;{6*#@|$RO7Cil*9~5*)xuOJC%ag zb^BZ@;?+;Z2BG8u(h>}J37RN3t@5|ePlqHb6Z z94p0Ppoe4UU_yvktgRgeKzVbEEyCQ6?9dbs2tJK_P~{$i)vmu(T1yLBWB))*b;N`q zP{uj|NFq?p_#930=mO(KDXQCIdsuvzUx@n!a_~AE_cx}& zpf;>b9yVs(zxz==RN+ds@P$@w_FAn?k5D6b;X~-Z`zcrg2n2J76QVR-1MOGIbr||Z zuGsV}7}YT<0xI?G14EpZZy42={&KS2l$qw@?Ck5_?D6D_WSqE4HQ1ArBXJV{7U#ZG z!f!$Kf{KFpcvrNjN@^@Iu0~ZysZnE9z%u{nIZE{c!4EiER3)zTKh}UcL1nv%IMGu~ z<&-u`AiuWBJ-%4;M*)jIsiD>Z%2jjA<_wh%LHC3dj;?@j&L?<9XO9t1{`cpsSpScm&eeD z%wnLCcn|Z-FKw6JvDB~}(|%c1w%Pj3s4q2)VH60$vr1&VDbDniNqo-c*3i6~IV(Nn zkBq7vD22Ua4O{wNJ&IkR_Qh!Rst_DS%%&|+@5`yX0qDU+Bt%(}X=#9Cmi95BAgF1Ie~?74ji3x_+DPMWAz!;#R3c ze5!gkK3^1Su?K14cvu!1G{O~JzA$fxf0D8gvX9 zRs=s+()=#(}M=T4N`0o%y9{5)DyGFY^cP`uVvAlS7zh@tJJ*F5+swk2``EBdR zcOzybql}l#+$}Dl@~`DGqDs##&?f64l!@t%JeC4-rY(GfV#92VhUs2^&sLrKb928| z$wsl5Sgup>pY%NXrB1%kdzF6$I1hA?TROBA%J}Lmz|Ssx;(9L1sws5-kW)+7$b`Q! zOZb4pW^Ie>(BYZFfy8S%V+|*#?Rn^h>ya^e%2h@BIaeE8=ughjz)%0$Gjy}5LhHQZ zxsg-1`de0EAs4dMV`#mQDH}i2YS926y9w3Ek#pAi<2sLVr&=|5-E`IHUiJ%bA$Q03 zzb)#2NbxTm5w9;wXN4WwFEr&?tWWt-4;}6_jkdh^#=RgW@2&KalzV;ZMmA5BCBG=; z8@9V=e9L#@>9$m(M5jl~(i1eaBDzqJKmJ<|H`zGoIcB0X#sk^b&fBB1}r@y+Tx}SG*KW z)_K%oU645){?+XA>%mQ5%u19@+N`;7GMw`dJg*3gm8o-{-otCO^Go;f&38&VIW1KW zhOD+{WGjrxJp8gw)nqlkPyD86Hkhxrtn0i<;XWT|`>MWwu1=5S`~5lQnKUP0w%WaQ zo7_T*-$q08VIyu$?zCIfA7>Uxe1zMy^Q1jAKdx_lR=Q1D?J1__J+7Jf1_zy|eVzQA zZY?--F*us_hZ0?FMmJDUu`)f&^W2RyYbL+z>9;q2V5BTy!cDn`1lYlRB z`WMYd3}!kE`qn~jv1WonA6a<@@0I?iGR&WCxA=lW2Ym72p~-#it>hK;(2|FyfAaGe zRMcWw<+A}*epAWjqaR>c^lXTZyTF6M%KYZfjJd1#ZOvkUKz!JkoS|lU!WmaP^mPO} zM?cFJ{kT@{2gFM4f)EJ>0$^+J1(Dk+B$zYY-Db+`=^ki!M-msV?Pu?`sQE!{rPg;N zLj&>$Z*(Fg7*9@RkGhmIZS)4IYeOu|xo(cuS@dX9cNIagfiJW1!d{(1sydKm?$kf> zQmH+x<72MX3Ui&mkmbi&)@e4Tt3bIuk|}k+Yn@`7y|B}l5lbRc7&(#OsvPX=_4nf$ z9ghD@jt(qP;B>`i_S+Ovn0D3{IkdNg8vQiY`AJv{;O%|(^^;#}vg4vBJuB`mzhy8} zwbRL5!k(T&C#qme_vZtTdXn_%-;VT2_a|{_<|aJh4B_^FU-f;~u#U(4mZr5kt!fFQ zD&8<0GIzEJi2s_ZIw^)yRAbXbqari%%d0zT>q^0=1<>BmA-W_1_2 zy9&xDDJ$!LOE)p>bA}g14mv|fEs#8gGPSg7uw5jN5=2c(HbI}}fh=2!`!-LC`wUs- z;|eKqoF+Vkp}57OG6>p*Uh-pV6_L5EfAkQU9-XtS>2K4t)tT&ZaBOmNP*MN_Tk*@C z8+S{f{+tqX7O+|sQTq=R)S)kXTr|}Ov*2;i{+ZP?3)$-N_6>t4{e7!(OvF>i2qXkp zMS?7tD0%l;93$j?pxVNuY2}z~>SG@}WaV~VC2q02+JE4CUL}H_mySB1_dDw0NBj8p z^-st)bq}#>CiYPOuV{xF3j6jrHpq)Hy)w7I{CvUxCfH$2P57m&0jt|x_p_Wb!szUbs)&j-lJ`T zrXUbLyn9fklk<2-Tq~JI?`7%w_3S;yWU6fmXkH%^M$E!#~xAt4zE(wtsDXU!h%duROJ=&qOqXmqdz5j`LW?9C(f|8s0yD z=chd^e7scccwd6*_$b7$k}QY%n67gcfZo*8K1Ryih)o{7r?;T_g_3EE2=<;1~&cl;g66FE${sg0L?LV0>{^;g)~jOX3G+iP20 zMbA>k$xi&F8(O0M6YiZJ$gXF{e|_WBg!gMA_iy+gqn{m~wqtQae|MGtxXsG_J)Czw z*wc$*3hw3L3!>3*+C6-I{^O^8BB7a$q&Et7l5+m@nO@uC8$j-|F}~FQ3Uqo_`mx7vR@E%&v

ekIk&K zQrRu3#8$g$bWW=e_V1??FOg>|OJDfJ(%|Wbu~lzBn&>3A;#V3ax|^nqEA85reBWe6 zmPGgneNLU&(thhNj3(ig*IYiW^@vb?TC262hmVhU$?~>=2nIQ8Rr(Dx%)j#Z^`z7ggCNGghvBIiRjI+7fVhDuT3G-w6J;xjrHY{{UyH&uJPcSgAef6CRF3G+8X1*i(kf?iVpWE z?0jPMdB5nDQqo?ruerEcza4DXfBSbz`fpFPm!H%}QVK4N*Y$Q4OIlqnrwe3IDk4{@ z=NTRqT>zFopp_T>Cp}%x_^^CA?}}wiV?1-) z&5&-A9&GZG`<%e2^LC0>>5$&7#v3W8-dz2BZLVJ6v`q6ETDYfdXr$`=ESFqM@mnGD zuHd~M(exzN`N<)FWNxrzUmvvF4x7)`_&d^t_VX zJUQu>TQL4lKN@NJpB^kTXAL)eq`zJ&ZS+@6(Wq;bY08QCv@1PacFTTeZAvf2DADW0 z;_5X|^_K0$YxIlQRX6c;UcMIz>(W?~_#Rw}-I4q8l_R-TJf)K|M}R|#(veNwKUYV@5uewMbz{dy*)*LeQfQ!Z@p}s^d$;?DF0>X z(QZy!-{M0%)s>6rBlSz?=W)X{Qbz6`ZI|{ zVgDOL9EW!OVkXX1>UdyQwsnOzk^3_(jCL$vbt)gE7NNG2z0-HPMv>VPE>(UoO`XmUEKdTi=os3I8<R;noD>qcMm3YwYb5J%R%yN_kJ^2;! z(yvAGlhY5I-{Pm9dKOkcKic^->dlHho!@2at3}Uiy*=rfN~OepOn^QUfq zebayWZPW4HGLyvR0YwCBtMBW3BTN2VHj&c(yBpqendzC2WO+c}GWk2DT<@2iR6BKS z@A>ecb006-F~5vj|KX5YWI62de1N1`$q?_;Ooor^%d^>;i&GM>4cxDb8qfEwk1m{J zdYzuWaO};T|Kq9akN!x{bozbnm;d}bL#+W_GskGDYNzY8d9?HEy51jmLJtj5f4I@! z^nDd`=@Da1vjW1b0$HoyS6p&3KF&(%aWYw&e9vn>Cpj<~c7{G-t`I8Ky!9lsC5b5BCcwp#BD6o=FZ zmC96DO@s5!pI4#ymQZ$3@? z#@}+*-KFOAT$OCE%oECJ1N#$l^eIQ_{r=uZHS?>g8jD#AT8>vlCFoPuLsifdk!`EW zT4lhO|7b;*KkBfGIGPXj3=y<}r15SafkvT6U6GL1jSExCO}|^ZGp~&(2hv2h!5mKj zWwN@pY_E2{eZPAR+_xI}WX48g;A^Iyn_cCv5DRai(2B*nn%`JW;pUa1Z z9qz5HaaX^@6ORuc>RTf2hTq-bP}A0$pz6}o)4Fyj*2^$`Tt|%laT)o+b$z>lR^MHx z{jMGliU-B@Ecl{<4#v0QEeVNh=0hE~gl|d5%=Yhf&pwLO3)kDADbZedihWs%GLk+S zO%mc7X%!65Ho@Px5~|zFeIv(H<<1+mf6wsbtdEpDLCioS$mk9jZ6A zly~Z{-69R#8vj&A?)>W^p;WH5cbU0VTsA*6d1S}eHXB+3@Yr)Py<30Pe=lgiN_FfKC2;&u{mmyjt7P;-naS@GKaUKIHy+m6`;l~T=I_xmL;9K7Xv60zc0qR`Cm zrfT#wVr8tP>&s3|Q>o(Ji8n80OpNZ-E*CZM@HQG%H#QZ!|FM?FT_wfataKl|>)hHp z?CCLSVJT=uce{A*uCLWcxA-y6{r$U>J?%2xTYG9+8R7ju>)FCz diff --git a/tests/test_signs.py b/tests/test_signs.py index c87261c80..15943acb8 100644 --- a/tests/test_signs.py +++ b/tests/test_signs.py @@ -74,6 +74,12 @@ def _row(metric, convention=None, *, country="UK", source="x", annotations=(), * ), "unknown", ), + # IFS's estimate of the Budget's tax change, which keeps revenue_change + # (the Autumn Budget 2026 harvest's README rule 10) + ( + _row("revenue_change", "positive = increase in net tax as a share of GDP"), + "same", + ), # borrowing conventions, verbatim from the Autumn Budget 2026 harvest, # whose totals of the Budget's decisions take this metric (its README # rule 10) @@ -283,7 +289,7 @@ def test_published_conventions_read_opposite_ways_for_revenue_and_borrowing(): if {revenue, borrowing} <= {"same", "opposite"}: read_by_both += 1 assert revenue != borrowing, text - # 31 with the feed and the Autumn Budget 2026 harvest alone, 32 with + # 32 with the feed and the Autumn Budget 2026 harvest alone, 33 with # every harvest and the PMD feed assert read_by_both >= 30 diff --git a/tests/test_uk_ab2026_harvest.py b/tests/test_uk_ab2026_harvest.py index 94ed56919..6f4c47f44 100644 --- a/tests/test_uk_ab2026_harvest.py +++ b/tests/test_uk_ab2026_harvest.py @@ -184,11 +184,12 @@ def test_table_62_rows_are_financial_transactions_not_revenue(): ) -# --- README rule 10: totals of the Budget's decisions ----------------------- +# --- README rule 10: stated effects of the Budget's decisions on PSNB -------- BORROWING = "decisions_effect_on_borrowing" -# keys that name only a total of the Budget's decisions or one of its splits +# keys that name only a stated effect of the Budget's decisions on PSNB or one +# of its splits TOTAL_KEYS = { K.package_key(s) for s in ( @@ -198,11 +199,13 @@ def test_table_62_rows_are_financial_transactions_not_revenue(): "package_tax_decisions_since_march_2026", "package_spending_decisions_since_march_2026", "package_non_scorecard_decisions", - "package_ifs_announced_tax_changes", ) } | {"ab2026_option__tax_decisions_with_continued_fuel_duty_freeze"} -# the OBR's total, which also keys its Table 1.1 lines by head +# the OBR's total, which also keys its Table 1.1 lines by head and its total +# of receipts measures GOVERNMENT_DECISIONS = K.package_key("package_government_decisions_since_march_2026") +RECEIPTS_TOTAL = "Total effects of receipts measures (£ millions)" +IFS_TAX_CHANGE = K.package_key("package_ifs_announced_tax_changes") def _key(r): @@ -222,15 +225,19 @@ def test_rows_keyed_to_a_total_take_the_borrowing_metric(): def test_a_totals_lines_by_head_keep_revenue_change(): - """The OBR keys its Table 1.1 lines by tax and spending head to the total - they add up to. Those lines keep revenue_change; every other row on that - key gives the total, its direct or indirect effects or a part of them, - and takes the borrowing metric or, for the current budget deficit, - proposes fiscal_balance_change.""" + """The OBR keys its Table 1.1 lines by tax and spending head, and the + total of its receipts heads, to the total they add up to. Those rows keep + revenue_change; every other row on that key gives the total, its direct + or indirect effects or a part of them, and takes the borrowing metric or, + for the current budget deficit, proposes fiscal_balance_change.""" on_key = [ r for fam in FAMILIES for r in rows(fam) if _key(r) == GOVERNMENT_DECISIONS ] - by_head = [r for r in on_key if r.get("metric") == "revenue_change"] + by_head = [ + r + for r in on_key + if r.get("metric") == "revenue_change" and _label(r) != RECEIPTS_TOTAL + ] assert len(by_head) == 288 for r in by_head: assert r["source_table"].startswith("Table 1.1: Breakdown of policy decisions") @@ -243,6 +250,18 @@ def test_a_totals_lines_by_head_keep_revenue_change(): ), _label(r) +def test_receipts_totals_and_tax_change_estimates_keep_revenue_change(): + """README rule 10 (delta review of 8b7b23db, findings 1 and 2): a receipts + total or a tax-change estimate keeps revenue_change unless its evidence + establishes PSNB scope. The OBR workbook's total of receipts measures sums + its receipts heads, and nothing in IFS's chart ties its estimate of the + Budget's tax change at the forecast horizon to PSNB.""" + receipts = [r for r in rows("uk_obr_tables") if _label(r) == RECEIPTS_TOTAL] + tax_change = [r for r in rows("uk_ifs_ab2026") if _key(r) == IFS_TAX_CHANGE] + assert (len(receipts), len(tax_change)) == (6, 1) + assert {r["metric"] for r in receipts + tax_change} == {"revenue_change"} + + def test_a_measure_takes_the_borrowing_metric_only_where_its_producer_says_so(): """README rule 10's exception: the OBR's paragraphs 3.20, 3.21 and 3.43 give four policies' effects on borrowing, and IFS gives the Budget's new From 4f424b918aff769a6eca78940ea649f6e317bae9 Mon Sep 17 00:00:00 2001 From: Max Ghenis Date: Wed, 4 Nov 2026 08:18:19 -0500 Subject: [PATCH 18/52] Correct who keys the OBR's Table 1.1 lines in a test's docstring The delta review of dfebe9c1 approved it and found that a test's docstring and comment said the OBR keys its Table 1.1 lines to the total effect of Government decisions. The OBR stager assigns those keys, so I corrected both. I changed no code or data. Co-Authored-By: Claude Opus 5.5 --- tests/test_uk_ab2026_harvest.py | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) diff --git a/tests/test_uk_ab2026_harvest.py b/tests/test_uk_ab2026_harvest.py index 6f4c47f44..5c3c12c18 100644 --- a/tests/test_uk_ab2026_harvest.py +++ b/tests/test_uk_ab2026_harvest.py @@ -201,8 +201,8 @@ def test_table_62_rows_are_financial_transactions_not_revenue(): "package_non_scorecard_decisions", ) } | {"ab2026_option__tax_decisions_with_continued_fuel_duty_freeze"} -# the OBR's total, which also keys its Table 1.1 lines by head and its total -# of receipts measures +# the key of the OBR's total, which the OBR stager also gives the workbook's +# Table 1.1 lines by head and its total of receipts measures GOVERNMENT_DECISIONS = K.package_key("package_government_decisions_since_march_2026") RECEIPTS_TOTAL = "Total effects of receipts measures (£ millions)" IFS_TAX_CHANGE = K.package_key("package_ifs_announced_tax_changes") @@ -225,11 +225,12 @@ def test_rows_keyed_to_a_total_take_the_borrowing_metric(): def test_a_totals_lines_by_head_keep_revenue_change(): - """The OBR keys its Table 1.1 lines by tax and spending head, and the - total of its receipts heads, to the total they add up to. Those rows keep - revenue_change; every other row on that key gives the total, its direct - or indirect effects or a part of them, and takes the borrowing metric or, - for the current budget deficit, proposes fiscal_balance_change.""" + """The OBR stager keys the workbook's Table 1.1 lines by tax and spending + head, and its total of receipts measures, to the total effect of + Government decisions. Those rows keep revenue_change; every other row on + that key gives the total, its direct or indirect effects or a part of + them, and takes the borrowing metric or, for the current budget deficit, + proposes fiscal_balance_change.""" on_key = [ r for fam in FAMILIES for r in rows(fam) if _key(r) == GOVERNMENT_DECISIONS ] From 6a1c06f5a5688a3bc0f95f17c7e231f8021568a0 Mon Sep 17 00:00:00 2001 From: Claude Date: Wed, 4 Nov 2026 18:22:18 +0000 Subject: [PATCH 19/52] uk_hmt_tables: apply review A's T-1 to T-9 - T-1: Table 6.2 line 6 (reprioritised FT budgets) takes the proposed key ab2026__your_first_home_ft_reprioritisation; line 5 keeps the scheme's key. - T-2: stage Table 6.2's total cash impact (E14:J14) as 6 total rows on ab2026__package_total_financial_transactions, written after Table 6.4 so earlier rows keep their indices. - T-3: stop emitting the hidden 'Fiscal Head check (hide)' column. - T-4: carry the shaded top-level section into each line's heading. - T-5: strip footnote marker 4 from lines 45 and 47's titles. - T-6: cite the cells' "+#,##0"m" number format; parse_confidence high. - T-7, T-8, T-9: NOTES' rounding gaps, footnote 2 on the totals, Table 6.2's sign convention. Read Table 6.4 by sheet name. The workbooks weren't re-fetched; the output comes from the stager run on their cells rebuilt from the pinned rows (staging branch, harvest-staging/ab2026/tables-reconstruction/), which reproduces the 29 October file byte for byte under the old stager. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_01LwQq6JvunwtJCAckL6L4aM --- pipeline/stage_ab2026_hmt_tables.py | 153 ++++++++++++++---- .../uk_hmt_tables/NOTES.md | 27 +++- .../uk_hmt_tables/claims_staged.jsonl.gz | Bin 38274 -> 38674 bytes 3 files changed, 139 insertions(+), 41 deletions(-) diff --git a/pipeline/stage_ab2026_hmt_tables.py b/pipeline/stage_ab2026_hmt_tables.py index e7a360dd6..cedaea600 100644 --- a/pipeline/stage_ab2026_hmt_tables.py +++ b/pipeline/stage_ab2026_hmt_tables.py @@ -58,10 +58,33 @@ "Energy Payment is -, as in the AB2025 Table 4.1 harvest)" ) FT_SIGN = ( - "positive = cash inflow to the public sector; negative = cash outflow " - "(financial transactions: the table prints no legend; convention observed " - "from the rows, e.g. the National Wealth Fund's investment is -)" + "positive = improves the public sector's cash position (cash inflow or " + "lower outflow); negative = worsens it (financial transactions: the table " + "prints no legend; convention observed from the rows, e.g. the National " + "Wealth Fund's investment is -, and line 6's reprioritisation of existing " + "budgets is +)" ) +# Table 6.1 footnotes 1 and 2, as the red book prints them (review A T-8). +T61_FOOTNOTES = ( + "Table footnote 1: totals may not sum due to rounding. Footnote 2 sets the " + "totals' scope: 'Measures with no net increase to borrowing are not " + "included, these are debt relief for councils in severe financial distress " + "through the Public Works Loan Board and a landlord registration fee for the " + "Private Rented Sector.' and 'AME and tax costings with an impact of less " + "than +-£10 million in every year are not included'." +) +# Table 6.2 line 6 reprioritises existing financial transactions budgets to +# fund line 5's new equity loan scheme. Its cells are not the scheme's, so it +# takes its own proposed key (README rule 2), as Table 6.1 line 34's +# reprioritisation does (review A T-1). +T62_PROPOSED = {6: "ab2026__your_first_home_ft_reprioritisation"} +# The workbook shades its five top-level section headings (B5, B27, B43, B58, +# B82) with this fill, as the red book shades them (review A T-4). +SECTION_FILL = "FFFFCCCC" +# A footnote marker HM Treasury prints as a superscript run glued to a title +# (lines 45 and 47 end "...4"; review A T-5). openpyxl's cached value joins the +# runs, so the trailing marker is stripped from the title. +FOOTNOTE_MARKER = re.compile(r"(?<=[a-z])\d$") def fy_end(fy): @@ -143,13 +166,27 @@ def header_years(ws, row): return cols +def title_of(text): + return FOOTNOTE_MARKER.sub("", " ".join(text.split())) + + +def is_section(cell): + fill = cell.fill + return bool( + fill is not None + and fill.fill_type + and getattr(fill.fgColor, "rgb", None) == SECTION_FILL + ) + + def stage_t61(wb, out): ws = wb["Table 6.1"] years = header_years(ws, 4) - # Headings sit in column B with nothing beside them. A run of headings - # with no measure line between them is recorded whole ("A > B"): the - # table marks no hierarchy beyond that adjacency. - run, heading = [], "" + # Headings sit in column B with nothing beside them. The five top-level + # sections are shaded (SECTION_FILL); a section's sub-headings follow it, + # so each line's heading is "section > sub-heading", carrying the last + # section down to lines whose sub-heading has none directly above it. + run, heading, section = [], "", None seen = set() for row in ws.iter_rows(min_row=5): b, c = row[1].value, row[2].value @@ -161,13 +198,17 @@ def stage_t61(wb, out): text = " ".join(b.split()) if text[0].isdigit(): break # footnotes - run.append(text) - heading = " > ".join(run) + if is_section(row[1]): + section, run = text, [] + else: + run.append(text) + heading = " > ".join(([section] if section else []) + run) continue run = [] if isinstance(b, int) and isinstance(c, str): - line, title = b, " ".join(c.split()) - fiscal_head = (row[3].value or "").strip() or None + # Column D ("Fiscal Head check (hide)") is a hidden check column + # the red book doesn't print, so it isn't staged (review A T-3). + line, title = b, title_of(c) head = row[4].value.strip() seen.add(line) for col, fy in years.items(): @@ -182,8 +223,6 @@ def stage_t61(wb, out): "heading": heading, "row_type": "measure", } - if fiscal_head: - cond["fiscal_head"] = fiscal_head out.append( { **base_row( @@ -229,22 +268,24 @@ def stage_t61(wb, out): metric=BORROWING, ), "measure_key": K.package_key(slug), - "note": "Table footnote 1: totals may not sum due to rounding.", + "note": T61_FOOTNOTES, } ) assert seen == set(K.TABLE_61), sorted(set(K.TABLE_61) ^ seen) -def stage_t62(wb, out): +def stage_t62(wb, out, totals): + """Table 6.2's lines go to `out`; its total row goes to `totals`, which + main() writes after Table 6.4, so the rows staged before the total was + added (review A T-2) keep their positions.""" ws = wb["Table 6.2"] years = header_years(ws, 4) seen = set() for row in ws.iter_rows(min_row=5): b, c = row[1].value, row[2].value if isinstance(b, int) and isinstance(c, str): - line, title = b, " ".join(c.split()) + line, title = b, title_of(c) seen.add(line) - fiscal_head = (row[3].value or "").strip() for col, fy in years.items(): v = ws.cell(row=row[0].row, column=col).value assert isinstance(v, (int, float)), (line, fy, v) @@ -254,29 +295,65 @@ def stage_t62(wb, out): "table_line": str(line), "row_type": "measure", } - if fiscal_head: - cond["fiscal_head"] = fiscal_head - out.append( + key = K.PREFIX + K.TABLE_62[line] + if line in T62_PROPOSED: + cond["proposed_measure_key"] = T62_PROPOSED[line] + key = None + r = { + **base_row( + T61, + title=title, + fy=fy, + value=v, + locator=f"Table 6.2 line {line}, {fy} (cell {ws.cell(row=row[0].row, column=col).coordinate})", + table="Table 6.2 Budget 2026 Financial Transactions (cash impact, £ million)", + column=fy, + conditions=cond, + metric=FT, + ), + "measure_key": key, + } + if key is None: + r["note"] = ( + "Table 6.2 line 6 reprioritises existing financial " + "transactions budgets to fund line 5's equity loan scheme; " + "it has no Table 6.1 line, so it takes its own proposed key " + "(README rule 2), never line 5's." + ) + out.append(r) + continue + if not b and isinstance(c, str) and c.strip().startswith("Total"): + label = " ".join(c.split()) + for col, fy in years.items(): + v = ws.cell(row=row[0].row, column=col).value + assert isinstance(v, (int, float)), (label, fy, v) + totals.append( { **base_row( T61, - title=title, + title=label, fy=fy, value=v, - locator=f"Table 6.2 line {line}, {fy} (cell {ws.cell(row=row[0].row, column=col).coordinate})", + locator=f"Table 6.2 '{label}', {fy} (cell {ws.cell(row=row[0].row, column=col).coordinate})", table="Table 6.2 Budget 2026 Financial Transactions (cash impact, £ million)", column=fy, - conditions=cond, + conditions={ + "sign_convention": FT_SIGN, + "table": "6.2", + "row_type": "total", + "proposed_measure_key": "ab2026__package_total_financial_transactions", + }, metric=FT, ), - "measure_key": K.PREFIX + K.TABLE_62[line], + "measure_key": None, + "note": "Table footnote 1: totals may not sum due to rounding.", } ) assert seen == set(K.TABLE_62), seen def stage_t64(wb, out): - ws = wb.active + ws = wb["Rounded Table 6.4"] years = header_years(ws, 3) group = None seen = set() @@ -319,16 +396,25 @@ def stage_t64(wb, out): "note": ( "A recosting at Budget 2026 of a measure announced at an " "earlier event; measure_key is that event's registry key. " - "The table's units header prints '(£)', but its " - "footnotes and magnitudes are those of Table 6.1 " - "(£ million), which this row assumes." + "The table's units header prints '(£)', and its footnotes " + "give no unit, but every value cell carries the number " + 'format \\+#,##0"m" (I9 = 7130 displays as "+7,130m"), so ' + "the cells are £ million." ), - "parse_confidence": "medium", } ) assert seen == set(K.TABLE_64), seen +def stage_all(wb61, wb64): + rows, totals = [], [] + stage_t61(wb61, rows) + stage_t62(wb61, rows, totals) + stage_t64(wb64, rows) + assert len(totals) == 6, len(totals) + return rows + totals + + def write_gz(rows, path): buf = io.BytesIO() with gzip.GzipFile(fileobj=buf, mode="wb", mtime=0, filename="") as gz: @@ -350,11 +436,10 @@ def main(argv=None): got = sha256(path) if got != src["sha256"]: raise SystemExit(f"{path}: sha256 {got} != pinned {src['sha256']}") - rows = [] - wb = openpyxl.load_workbook(args.t61, data_only=True) - stage_t61(wb, rows) - stage_t62(wb, rows) - stage_t64(openpyxl.load_workbook(args.t64, data_only=True), rows) + rows = stage_all( + openpyxl.load_workbook(args.t61, data_only=True), + openpyxl.load_workbook(args.t64, data_only=True), + ) FAMILY.mkdir(parents=True, exist_ok=True) write_gz(rows, FAMILY / "claims_staged.jsonl.gz") manifest = [ diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/NOTES.md index 8114761e6..fc81488ea 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/NOTES.md @@ -2,9 +2,11 @@ Staged 2026-10-29 by `pipeline/stage_ab2026_hmt_tables.py`, which is vendored, so the staging re-runs byte for byte. Source slug `hm_treasury`, source_model `hmt_costing_obr_certified`. -- **Rows:** 762. +On 4 November 2026 the stager took review A's fixes T-1 to T-9. The workbooks weren't re-fetched in that session, so its output was produced by running the stager on the two workbooks' cells rebuilt in memory from the 29 October rows (every row's locator names its cell), the five section fills and Table 6.2's total row (cells E14:J14, as review A read them). Under the 29 October stager, that rebuild reproduces the 29 October claims file byte for byte. The rebuild script is on the staging branch (`harvest-staging/ab2026/tables-reconstruction/`); a run of the stager on the pinned workbooks should give this file's sha256. + +- **Rows:** 768. - Table 6.1: 98 measure lines and 3 total rows, each across 6 fiscal years (606 rows). - - Table 6.2: 9 lines across 6 fiscal years (54 rows). + - Table 6.2: 9 lines and its total row, each across 6 fiscal years (60 rows). The total's 6 rows come last in the file, after Table 6.4's, so the rows staged before it keep their positions. - Table 6.4: 17 lines across 6 fiscal years (102 rows). - **Primaries:** 2 workbooks. @@ -20,28 +22,39 @@ The two XLSX attachments of [Supporting documents for Budget 2026](https://www.g - **Table 6.1:** every cell of every measure line, 2026-27 to 2031-32. - Fields: `metric: revenue_change`, `unit_concept: gbp`, value = cell × 1e6. Since 3 November, the three total rows take `decisions_effect_on_borrowing` (README rule 10), with no value or sign changed. - - Conditions carry the line (`table_line`), the Tax/Spend head (`head`), the fiscal head where printed (`fiscal_head`: RDEL, TDEL, RAME, SUBTOTAL, Tax, Tax/TME, …), and the run of headings above the line (`heading`). - - The table marks no hierarchy beyond adjacency. Two headings with no line between them are recorded as "A > B". + - Conditions carry the line (`table_line`), the Tax/Spend head (`head`) and the line's headings (`heading`). + - `heading` is "section > sub-heading". The workbook shades its five top-level section headings (cells B5, B27, B43, B58 and B82, fill FFFFCCCC), as the red book shades them, and each sub-heading belongs to the last section above it. So lines 12–19 read "Breathing space for families and businesses > Breathing space for businesses". + - Column D of both sheets, headed "Fiscal Head check (hide)", is a hidden check column the red book doesn't print (15 lines read "SUBTOTAL", line 30 has "Tax" on a Spend line and line 44 "RAME" on a Tax line). It isn't staged (review A, T-3); `head` is the printed Tax/Spend head. + - Titles are staged with whitespace normalised: sixteen titles carry tabs, trailing spaces or no-break spaces in the workbook, collapsed to single spaces. Lines 45 and 47 print footnote marker 4 as a superscript run glued to the title; the stager strips it (review A, T-5). - The three total rows carry the package keys `ab2026__package_total_policy_decisions`, `…_total_tax_policy_decisions` and `…_total_spending_policy_decisions`. - - The lines don't sum exactly to the totals (by £5m to £10m a year). Footnote 1 says totals may not sum due to rounding, and the rows are staged as printed. + - The lines don't sum exactly to the totals: by up to £10m a year for the grand total and up to £15m for the Tax and Spend totals (the Spend total in 2029-30). Footnote 1 says totals may not sum due to rounding, and the rows are staged as printed. + - The totals' note quotes footnotes 1 and 2. Footnote 2 sets their scope: it excludes the PWLB debt relief and the landlord registration fee ("no net increase to borrowing"), and "AME and tax costings with an impact of less than +-£10 million in every year". - **Table 6.2:** financial transactions, `proposed_metric: financial_transaction_cash_impact` (cash impact on the public sector, not borrowing). - Each line carries the measure key of the Table 6.1 measure it belongs to. - - The exception is line 4 (Sovereign AI), which has no Table 6.1 line and so has its own key. + - Line 4 (Sovereign AI) has no Table 6.1 line and so has its own registered key. + - Line 6, "Reprioritise financial transactions budgets to fund the new equity loan scheme", has `measure_key: null` and the proposed key `ab2026__your_first_home_ft_reprioritisation` (scope: Table 6.2 line 6; no Table 6.1 line), as Table 6.1 line 34's reprioritisation has its own line (review A, T-1). Line 5 is the scheme itself (its cells sum to −4,700, the red book's "£4.7 billion", para 5.5) and keeps `ab2026__your_first_home_equity_loan`. Both lines are whole rows, so they mustn't share a key: marking both as parts of the scheme would net it to an inflow. + - The total row, "Total cash impacts of financial transactions" (cells E14:J14: −230, −1,160, −2,290, −2,795, −3,675, −3,850), is staged as 6 `row_type: total` rows with `measure_key: null` and the proposed key `ab2026__package_total_financial_transactions` (scope: exactly Table 6.2 lines 1–9), with footnote 1's rounding note (review A, T-2). - **Table 6.4:** recostings at Budget 2026 of measures announced earlier. - Each row carries **that event's** registry key, with `conditions.recosting_of`: - lines 1–14: `autumn_budget_2025` (`data/uk/ab2025_measures.json`); - lines 15–16: `autumn_budget_2024`; - line 17: `spring_budget_2024` (`data/uk/events/*_measures.json`). - - `parse_confidence: medium`: the table's units header prints "(£)", but its magnitudes and footnotes match Table 6.1's £ million, which these rows assume. + - The cells are £ million. The table's units header prints "(£)", and its footnotes give no unit, but every value cell carries the number format `\+#,##0"m"` (I9 = 7130 displays as "+7,130m"), so the rows are `parse_confidence: high` (review A, T-6). The sheet is read by name ("Rounded Table 6.4"), not as the workbook's active sheet. ## Conventions (load-bearing) - **Sign.** The tables print no legend. `conditions.sign_convention` on every Table 6.1 and 6.4 row reads "positive = yield to the Exchequer (reduces borrowing); negative = cost to the Exchequer". The convention is observed from the rows themselves: the bank surcharge rise is positive, the Household Energy Payment negative. The AB2025 Table 4.1 harvest uses the same convention. Values are NOT re-signed. +- **Table 6.2's sign.** Every Table 6.2 row reads "positive = improves the public sector's cash position (cash inflow or lower outflow); negative = worsens it". The National Wealth Fund's investment is negative; line 6's positives are a lower outflow (budgets reprioritised away from other financial transactions), not an inflow (review A, T-9). - **scoring_method `final`.** Table 6.1 is the certified post-behavioural total. The static and behavioural legs are in the policy costings document (family `uk_hmt_costings`). - **period** is the FY end year (2026-27 → 2027), with `conditions.fy` holding the label. - **geography** is "UK" on every row: these are Exchequer impacts. A measure's own scope (for example England only for the Household Energy Payment) is in its title, verbatim in `reform_hint`, and in the registry. - **quote** is the line's title and the cell, as printed (Open Government Licence). +## For tranche 2 + +- Table 6.2's proposed `financial_transaction_cash_impact` and the registered `cash_requirement_change`, which `uk_obr_tables` uses for paragraph 3.20's £1.9bn rise in the Government's financing requirement from the National Wealth Fund and British Business Bank transactions, both measure the cash effect of the same transactions: one as a cash impact, one as the financing requirement. The ingest settles the metric. +- Table 6.3 of the red book prints Table 6.1 lines 45 and 46 again (`uk_hmt_redbook` rows staged from it); tranche 2 dedupes identical HM Treasury figures across the two families. + ## Not staged - Table 6.1's footnote 2 names two measures with "no net increase to borrowing" that the table excludes: debt relief for councils in severe financial distress through the PWLB, and a landlord registration fee for the private rented sector. They have no figures, so there's nothing to stage. The registry doesn't list them. diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/claims_staged.jsonl.gz index 4c6d32a83246aafaecda27b8d8d4304823892407..5f1f6d70924024e34fa12b615fd6d758f8e1aabe 100644 GIT binary patch literal 38674 zcmXVXV|Zpw(`_)p#F^N(ZQHgrv2AB!+qQkjnb^i1Pi&hf&-B!%Z`XQ9ZMDHz@rA@Ie!u=l4^z` zOVxwc7oP%ojJJnBFL-WB&S|)4SOTeq#|mLe^Mz}DD2&&r{lGG4?zBMibYHgpBPN&? z7~IS{=2PNV8s+xTm+p@6>crZ|S03 z0J68HD&l803Dx`SN5J0+<8jFh&-O2t6<#OI6G@)B$+v71&FLdmlz>wA@uOWmfR24( z)aEWT6^Z|`!eZ6F`*ufNgLSC2ru5lX0 z?LXt(gZom8oMep*?qsu$2sIhhlLPX?!+3Kowi|Yua@$&zo3F?$RxLWwjEGxmWQMJo zxD>O>i< zWPc3zKZK1@#A9_yp_+$4ZC^-V38JrtQu$u_NDAVY!&{t(NwF?Znq$@@4M#sI2H7al z;?QS;%9DNAo*6eWr<6IZu2s^PgP3Qs9<2q8F%9_K@OGpV@$TyrmeWDOu7nYa;cbr~ z{)C>YckR|hZYv;rl6}{DF9}w_tjB_g5pBB-d2FGvCh zuDHu{6met`bCjx~BGkc&ZVz9YW19>ycf(Z4>7ZO{-JaTOUHfI3fFH)<9SzXbaj9_8 zV(DZuX-T>ZV=L*O1LfD(cIf#@H^ylk?b6RtfBU^INmEkV1&A1;hi5e}&yrl3Th$@> zoO#vH?#_HA4T4esFS1+}Mpcs-$arl2C{6!XZ0s$EojH>j|-{ zszs7MB?zAfs%(6C^zcVn&#+Y%v7WR!D z9g%+w+%Wzuy(=M*=M~8Ilwr6r-}C6W59(bsSJWjtmF)S_>KaqK+AN=zli38iOHdYB z-o`NGR`k~)sJ<8e#=o&;R)A$on$>ZCk#5n^vWsf!{Y)XxFQ)CO=!p&+<6v%F;0m>O z`5 z?qoex?RLWq`CnGXb~C)r3=;IJyw8=d+a8`oiXV6~-qv(aA@SczY%5@)@@}UGL`R?!AFk%ONQ=H{v!*ZOIV+$jb z>iWjEoRr;E?76gN26s#mxwx=b z0m4W~GwUbMS#tfG7ag=Up3&8cop-dxaeL}nyC8ME6_YX#i=*$rAcS&_$mvIzU0ec; z^NN!*h9rctt$5`MnKI5ieY_Qpa@=UwhZr@f!8(}%A(~_hR)t@g(;L~dvLR1?4_>wMZd`l;HKqI#u(DjWZmV{8 zDh%w*ReAVC`dt`Qdf*ui;@10S>C8aiSB{7yZ7L1dIW+i$DCtBgZbr{Hox3P!gVY=G zl`#O0uyYSN4lWK;@DOBQnl3j?g{NKvXCLdj~)Zv)~G&*jDO6*eT0k#6`rP&Q-qA*U<5bWHN=dI z=H)YljL?&-f%sSmtk9EzeNWbNZpfIW%?G8dZkRh`J2y;_Dqms~>G*jg>!WdrQwP%W zCBRx*Xy>^|Vy5+cKjs!dft0t%^^@u+|2o<(S!kZ_DT+N6B<1HXt_oA2b-`yrnjZo#W!vy^I@my_ z5N6l9A4|LFiva1XiwreLbZ-e6N&n|(?JRK)o(L!UaQK+St+gd2tiral0-ip2;?Ws+ zl|1p5$=*I0JMqM`{^*4D$xf{oJlATdCJnzo4VDGf*dREUYSjkVwdV+Lz+X4?}wS-+;tsf(5vZ7Kgml!L05$q zhPO-j7J6QGr}}ewi9I`ZVKMtC;_cl-ANT?|)QiyPIPTM*+M>J|>q|tPMuSJbad+^~ z00Es&Ud9WjK7#8N{@1=Iy*~mNfArkUWS-y3aS@)uKclA}`dC*${re0$FMkB|IX*kj z?U`w`hWZB-9s~&N%xFX)%97z`vSXSt#l^@ zu~lCvDj!b?aM_t~@5&JQ(eErbh4BqC62!)iZ<7n$tW`h7LXS|CM#{J!xW&y}d@0I^ zGNb32OM4d?yUjLiwpRr1t3HnS@y*e^b7f8y?eMZ)Lz;}rvTeg3cdARWm)yEivJZBf z?)GU+GvlofPF1^igj9081l@58C8~T)m=?mNN zIExWIe*sn#i(x^E*^_21UbZRY{DogGTlT-65ONs*7$Jb?2vSgJd037A6)Yn=?lM5Q z#N#Fc7lDN24wRSwyjr~X50sagt?)IVnH?LtG>eeu>=vXjaq-eCY^t~c%W=zZT6P;@ z?@Nj+<*}=Pizzq(E9#@PXK}a61oV}YVb;iQ%)&%BGT9HLV!|73ffcD6vhs$_ZSofW zX{*v}fSRr1rP>z~b8@ey^fXY$IZimM55>tuzr>W}L#ybuvyeR$>-sfgxmbFD6(fmd zM=k#Oyy-8DOj!xtZL($?Qb|Q#w3V>Xg`beyV7vW+VAr;d0FD7kj|3B<{b#mbC4}8Z zmv9j##0tZGCx$Mc5(7)4)veQ7*&dDwLli=3Gr>hfWgbGQhVb+w(r(59MhGM5Kdn=W ztBA@ngm8`E@kgcI3=)hIM&bX>yNRf*K?vIlT9lkW5)|0D=;cC(7Q=>g1Se`k#vO6(6{%N=A{a-(2vkTDQxRF(2YzcS|Uj5w1lHbK~3^RtBQu8{rEzPQMDg| zgcx}#96$cGfCar1u4W8!vyiC^)=MCD8L0W9LJCWU#=|@Y`qG{+WLV|}`FJZUo0qb~JU)c`K1~$F{pF|3Po>GX-+dw^;JNi2Fge!w4HTX-T|Dy& zDW$=V;TkekJ%;IXGDY`h1IM1k5M`gO? zXpahwRfkH}NbIR~%yjI%`v=c8k^2HKW^x=ac&BYB6{V>MAAXk$tR* z76d9%hI4M~kDU}h4q=L@4k&0U+YvM0bGB*mcCcC7> zFs7Jq*;tvhwADYH@3)lnGxV8ljH}x?_kS#P3bLG^=W5>uSO34ku@5l*OlXLj936oG z(H{}vn-SqFmc;YA^8bxSDWpYXtHT6Q3TWG%qqyS336UFkkZ(HWM6s=T{u{lhMvQ*Q zsLYB|*2bo=Ns4lk5^pt#h7_JqOGl;a_mnC*iqT;YG5Z{U=^DuIWlUI>j`GV-@#m@x zMra%g9oBf#sXc%C zbG_6rTpwRoe3_BRt+`5qZZ+Y}csivSVq&_~TYNBfwr&_N*o(lHEPHR=swf`@EgJNfGu%*4`s2D~GxviE3-0m^BrpFW5 z?Dd?*F+q{hrU>e)nX)LW)Vu(dN0J#UOsn(Ch0=DjggMA=g`kf*M2-ht@TF{5r5hX| zhan_VyZ~Hjij_&Q0<35f1Ed_Vp+G|C(e70D~U*w{GGu3K6LFqaazif+d z>!bjuI?mf=8oTPkZW8iZJ{+|EYm%ANv-Kd(5&Tv ziF-1B zv%cV~W+5Xq9GH$R>{}@v9GElVb&a_VUm)=3IPcGQNiXLAunWq!(jqvpCPD`@4i-X2 z9J)+bWHQOQ%U>RgL48AFxe_l+)YNsi#qkp4c?Em=?X?ndaC~b7K|?tWn)0ocP(ym8 zl~f1Wg*If^S15H?is=+)+4HL2DQN4>#R0VTtguO{VZG0%4^INVbTTXUWO0d=*B{@z zv|6&XpL5NEUkDG0M6aCoobQF?HI;G(-Q4ZoEmxa89nmK$RHN3c{>*q;X4ZmVVV^Og z;1OXt5#p=1g$)crBy)m=H|D=$q`%JGK{f#laomVL>s+ z3M4tIMt+PE5brstpAfXr{pAxzvopmV<-|>VlL*_0iw4IU$s6OmHwG)lnXc6{1iX%BNvD9;e10bE?F_tZ(qg}B_+2XC=Gj#W6 zs$qp{16?UK{{-n4T>*yM#cB)Yqj|SbsXr=8k$&C?H%T76Usry3Oj^?W{O3wLZIC=WHo@_92VsBAL4&iL;$<=FF@1RJFTe45sCYY!o0maY zJ*0Bx7`H*}pmLg3ibIuR!)I%UE?CKcGc+GPDf4IjRIF+nw1jrY;A{aeOM;@~qu$FV zW}(7S48G{~<*513t}4!;1rb}GPIS9k2&OH;Bnk3v z>Rm>jCY!Mwv(XqcxGe3nWO}&Umrz#QHuAN(Gjer=&YIvAHtk;JU+5Wk+Wis%g$x52 zbDpCQz$0Q}9tQoZ-08Zy*LSpHxOAEu;$GtE2;dw+|ZMp4=hch4+zC332| z9;Ck*6t)d_57r$`dK5v$1>UVY`xbybv7C?L8vRcgyxUjw>K@bUh~}@P77Cj-*$T%} zEZi*=*^29NMTJP5(6-9So9;-qNA0fUyC{=8x&CS8*I49hSz6rpD5;QZ{&Y7BgqP2Cf}RbN-T? z^;+8?MR@t(g~^=LiWAk~1z9e=l%T^gOLn#*PiO2$HqNQ^1^u#YJDAlSRt5PLwo)yp zOa&&Z!|bq1c=@}7=54w%oK|+uspJK<8>`&x-zo>0FTVC3o0m@9yu+4ORTe@Y8P$bD zwOtX1o%604JgjDyAP)R!({)s&J!*m9OS~#jYSX?=P|N!%7oL0s{A%Sw`Jqz5_xd7! zC37i}rW7R-UloixUs^2 zB0D`^Lr8Xc@O;`VCxf0Q9B5#yZf4^xBg>v*;faFcMiV}kovvRtXfs|*_}e~03nTf0 zhUsN7#7tiHxQ>wA-uQL%SCi2|Q2GbiZ?(0Swn0;soRk+mAk=fikxtM6)Z{z;1p01b zyrTXXRa{RDSZKQ1^oeFf;`Y4dyveU5h9%HBk`+{fO1{2P3FBe`-r-T1(*?;V%jWOE zAG!xtEx-z~iWToc{cCCQCMTC$f1HsA)% z&cL2V^Ahs@CE)ke^OK%kc=}fsHI~LzNn3WB)6Y>eZVUx?)~tP2wOJ3pwE2}`h*m zzdylgtqBQ6-Ia=J(#ZL(0R5+wJEaleFM>T$`0e6;-{%gu zf$uOcE$;XrFO$4a`DC}j0a-@~``|D1k>_Z)IK({hi2Ey}b%QTI?ao_5tX*d z_Ps~q4H!70Rb$>L6{LN0=d@=Slzcd<+ApuY!v+MfiE%jDa6Ht-@;z$Yv^?}Ab~1AL zEJlZeu-O$zGfy3AT1_Q6?rd_(Rq77OuvbRL%EK;HP8*hjPxS;Y=lZk%a3Nj>b=t{1 z1J+h?7X!6^=B;9caXd*xc z_jlzu&p1r?8{)&&y4+BKLpCvWm(gMIhMvHGgOg+qGx2@>hUjoTH8|5dzgdg0_iC=h z@2Z#);J^n`jaRSG2i}a?87IL9T$yf}z~SzwbhlbNGC0DJ;o+Xp20oeKs!}kOx;Dr2 zi^kG^@7k)$4NlX{_EZRZ?_5@Fe9T1Wljv8sRmnni@+RE1p>hmi!H-^Sf))JMjO3<0 z*hyKppL7TKXjeb{EsSfQ!tO^XNZ zGB*5`zEA$^tyB_D_7Nv3g^2fpg&bJkyO5fmU0BM>_*2%&+1GdB0fPK3MvCs{5i3!5 zLpPN}+h`vwCVTeSHFCH#2$uB`J0?YE-GsG~>9=Pgaw_fYD0GZpn}v-AmH>Dw#3M6Iz^g`Ix3Ia&o37e0d@S~yRFmFdv)J{8&$@hTCmzmXa7=| z%?ZqSSW3G={r8Kd_B!$$e-r?2yi~1M=2r5D2JzX<$Q>`uA|`6@XS;np!n;WP`8!f} zi&jPV`p2)60c!|OPQMC03x4 zErfVDqFx`bD)8)fk@El*o%K2l=xbtib(B@IrSY`y6sb)7$JPB*ypb^QQ+;4 z>SnpIVo5v73a5VF*X&*`VU;|sPmZI#H!TWppV}hxRBEWz7b3eYUY*A%!#I)ly&d7b znepL_?;-LHt}v{uI&?S8+%~Mtk*zu9^e%|eG*em5$n4#`8ffgzw*YStq*YPiVdS$V zT$UfSI$0C%ak?q<)NlQp38{j`+`{9^&A4RAxFla{g}5eO>ZFwE$*qKx(XSdS`@n5x zweTBKrN;CWnGJ9x8)^AEuR@wO$Q2`%(xVRUG?hL9C_i zwnVZ$%qo?mH@-Vd2(@-4BuU{F6YIcW!GQCD@)cmNo3422&NMx+ zZ%JBI266!nmT!^oo{$mY(d(_z8RKD!xhdY{Wq}xyEd&)Q&u0oYWz;x7SM7C3$Mfow z0)#Y?mbq5kjXQb>t3N_&BIyY9wM|frfXPa~dnXS;r5T}N|8kpu%FEtigtX+6;{W8mR=&BlhfjeX}XEY{8ip&S_u800~!xw z?f(B9EQK-ZeQEi@Pl#it>dnR0H(n`1BpgRH$=u2_kDMdSkyB; zk}En=e0;j-ULDaIgVI<-1U(-WnYaX`7*9J$L|8?OjNhwZZfXJrxfjgigWxS`VSFgr zorz&F{I;$!UIfKhOiakI4D4~01x3iP7sGbu6bjwiKp^x)S6j?gWe^i199+|4dRgf( z9Nb;MLkWvQyDofFLnZ}XT^I#uBtG~+AEU~Or&Y&vA0rOj$}|fG+?`ThhDm!7_gEhz zidW?Pc!0wfAufKfL!9l%nUjZ7M@G-H3hP3{~%+v)+kMQv7M##d~5s%djEYZx8rj+yGSO_gjxz z{(C*?CTw5-I~naXt3$_Bq(-F&tY)YSz4|(lg&xQYp*L`~V({x+1HRMNP;oHTBFGmZ zKZL*T7BJg{0;BfadnECBd2akxQlClz)WGklnSLfjxe?rbs24doU%1;lmQByPH}MC% z_-(IS@0`Tmik9Ga!d}Z^Z$SZ(&MoT+Kl|vvyZ_Z1gi8_SGVrgV-sK3kaRx5-j_y_J zT0;e_a>2iR2fxlSKN(yE#9Rf$ZqrZvRT;z|rlEnut~36>y*^or!OKd)A8c+;*{7%X zCkxomMz@DvgoUq+MT|&-^9wNivxaU zPoqr>MAnmF2-5EyQU_!JU_byMPM{Wjl>me$gdr|15P5r>T;Bi;gxC#q(BqFa?fCYAHIT-LP)|y z7Yo)gUpwXnoJa?!ib2p-QH85P#*#u5{c9*TL^-1(_DUP9x~0~>wD2mEqM&m{j(YS|Hm$9?R5i_^xiW>*^(V@_eki#COZ|(0J#Llx4#^J_9Pld^6E= zyA3J>n4~pv2X%IKC&Yq*)?+s0;|FmUXhFDU1m7ck-vfy#+9c8Jw#JwDbijfkribxr zLsa}CVsVSrZp39Lf`2%UT?t?`JTXvPQwev>Aa7dN<2MPSY)U$ZYWx&aR! zWz*Kd%M~CV7S{7g1~(x#GdL+aZ0M;}PnX2Mp8UV^gn50*B29aWunxk8f4H7ek5!OF zL@oWMxZ$0N6t>S1?oX^aI4ifc5=AfZ`5+S;v2rExlt3J29rycC@@v(laZRYTP%kLx zrEn(Ldg<;RxvZX<^5t+(-xOrSQT?-7*g@nDzI3isb8#7X#AMEfF`)CPI^ny@tEvrE zBV?`rx6K@8JD<&4+j`G?Cv?(sO6)b2Zz*w^XwZHZyZP|I&%5*uUc0HRCAAl7E1jzb zaJ=1Ewsz?l%YBf~W-&DZ`Xmd5+k7a0CJjb#{hJNvr_XHSA}FeZ19tXrvq0}s2n6}2 zqvqv4fOBRPnOppj>8l|#%Vpz$kzvD9IkX>91u%(D~Cj$vmMJXXKV!N#jQz~t(m zW(0um4+%~~Dg`dlB3LZ#&ape}NSG`v=OaPvTG#~-E?~8qS1d=svcWIUHffz2DDxVr zOROY+dkaChV}k795*^^#+SwHusg`JCMo?TS+`;=}MQka7suFT%yW-uOZa^+un5+<7fyt(U^Nmb)-Ml6w1$>NJD}a`gJ>(s#p;Gp?MYlPj#?mVK z!6_ytx|bP<>a?>G#rjC|%X>j&LLC2#aFdQtmT$gs|LoDF zho`ukR(286;54c~*ni_%FQ^^g`0pHu_z_oBVp3FM>Dn`ei zu8Y;?FbctoGVzPEZV=S5H&rE}_xF#xydD%a!pa7mneE#B*`%x5BKPgDhxb3avAMUl z(~(_^%{A8)onY7u-X{sUE4Dnrm&YD_p zu-pRk=Nrs(v+#Dm7!eCS&_V-!rHMDEx$U?!8YxAD-G4yxe9}nvRoyQLq3u#7L>z6B zQpL+Eq>f(^*vG3tc}pJ^G17!C62y-;PZ8o~hgT-7`!m|v`+8PHjCi1LMXdf;}{+l7;jaE4jSk?wRx^E$oG+X};n{|Uo) zBZKEFPNlP`KUgcN=*JdE0PO?ue9Mit_K2(*7x@Ni0Fr#AIS80E-wIj2<7|3 zh%#!^wOP&CDWub_j%<7l#IrC}abB|Tgle(}l20lM1xn#Cah;(j!Na1aH$x=3dM*{< zrSGfoxvsW)og%$be(&kDtz#iexnN}1g2pNpE9ufrRiION$P2Q;lGKgo&rp-(*1x4c zl3bTmWEjWFDyh6>g;gNPZ3m0D>Bw=ys2MF`$*)$_-$o?ct5Z9xze|Dng=nhTxJA3J zy5e8bx6!-J#ZRL@V>{}@4wm$e05*;CvPo<|&;*dts(uQqFVWx0#_A%&VkHWP6&KIi zD{Pm0iiuMQNyfeK0%W5SOaF>Zl~8L^MK)T)BzMKL^iu<3EqRw_=A~$dWvH2cj9eY#zekFm)cO`?CsUNaR zwMuRM#0wM_&!@C-sYT(^jQX!oVb-l~Ub%8w-iJ2|(A7P+BOO{S-vB?;z_H49VJWMd zpP{3S)nHXQFwP(iY2m|oW%TrR>zmG;-KdCy`I}$Mw+P=^VmA;9>FTMh;XMmw4Y!V< zM-SsN;Wbz;4z~{mXpNXrQSpIv@+8)fnPG`t0qFW1X?^!4Xa$T93)071rycOJ(U}vb zQ8=~yoBRdP-T%EjDwbkOxI(<|l|$TC!3qwbP0%@+#7&qOHa(2dRn>)bx)7?}rgOsi zWR9Oe0j%l$j?krFW4^+fh)>lb;@}O_mDAxv-zP;X5j*dw!NqQMV*FvPU{*7{j1!S% z&OAR@byr5-cPzyA2h+ie_|~*D^2muT2Cgb^-mEaUcAIEhZ*ZjzH)|uo=COf!D$diU zeB@l4?{^={wJ8nWpD-6V52t`K_UM{F6c}0Q6ud^r;JNf?0|f<$;JLUv6?#Tz|Hg)3 z4}U5=_XU(q5Ib0e_XSWud03sT+*OZIV2q@76fVu=HJ%y1fE9yO;Cb0w9Cw3Ifyb-U z*L@J0nOr(SQEli2ESgzzuomkDB!f$~6r9C@tqYA0jg-_{Kqv!U8J$4qCBRP3F(oyY z!qY%vwn`3`f`U?w#R!;(hTZUoFo10E2Oeh~l{_qJdWa!>DfMr&Tqt!*J8Ga#7_o)e zLROP6{TNQquLBGSB62oM;Z;@p_GEGvidB^Ad2b_Edx8wo1%5?Wg`>6np5&kW3{9|^ z2&!kk&4Uj1hn^)0#2m*Jfk4x@^ee&TfsC#ghf_SepdC zTk>6K;EW+**|3sL{N2W3Gg6*KAM;5oJAkkM&|Jkwj5G31pd|w>QhmsOzqII#baAY*ct*J)E^;#v`#PCi|J{~&@8*YuaJIO(OBZSdl$fSo>Z4>F z2R79*u3c9LD(Nt4zszqrus4Gh*-V-Pcj=g@-$u3GurMLSxoGCl#Broq8LeQ&X)9nI14Ds*MKgpBe2>-X$hAC)1)Yx-q)w|)EhtNk z8?`_U%}!Vh14%TmT7v(uXG^{``IlRG7NfH<`60S}Rh=UfvuP)h$26&Xjq+|063A>c{X-czv?l(&4- zM4fU5<%XOrH>7*$-PSh&A@fr2vP>s~Lr<@1g(p+x1yGr&6$d6y+3J`ZeRmAeLdaxD z0eYwof=ZSfn&~sGfv!|}y8v73qe2&YO_)^XCTHJ;X$Tox6HBwxHCMH9aiG}?4i&53 zbrQ0`$9MTs(YasY?74zXZ{m4uXyxDpvwzMtG_T0yeWQEI1S{=|wC9#m{tmmMb{#;n zFr5>SU6y#KfoE^NJCq5Q%l#;RJX_@0_`iN#4445O%nPa3;2p!3kmbCSYI+`MZFE6J zIs{jAP4MM($tDt&Oo(NCSnD-kU))khWKB=~Tp=BZ5l#o2Wx5gUzto(Mnz##FCwmUj zQM}_xCW_+yOpv%|^5}ROjgp03l5;*v;_~e3xJ@%z zB+ZidGdWDE8#$T6B<-x4IG+1mHF(Dg88qt}_@MMc4cU~fK64Hg?swy!C)l_E;PUkm z7~|S*x=y*$$c7uJnB;|;k}MxdV)!63(zq|DEq-N8IIBT4sO^slX%~g-8<@zASQkL; zAo84&xGo!%EWMdkPc^v~8Xz&;cS$)b@s?o)@IRf!sp0u928$3WufQY(HRe~J$seq` zPzz}xBro}u3N+_6>Dnmt9)Lmmk50e+a1_#AxtM9N^bcys4X=_@Mi@NUEpei_1cy^Y z!WeYjRKd&2r4FA6Y+v&{5Y(lIq;uZoA1s0RPZqP*V7VC74i~Ni1i*z)3_l#mZ$N4G z64HmCb-eMgZyF$_fjh=|465It%QN#(yd2n(d{z9OtQ1If%&fCp^q5I9mid#eKfC=! z!KE&531O} z5DwE>0|WsM6ch6;8gn7Dmg8L)caM0%YoL|TS+EphOpw`qXsr*)TBvbGlDVeWOaD4U za#&2c1l?MKnPG>aBFL05qw}woA#@cX(3~`5S-M;689+bYzM9uKw$>oq%H;ip!U#UO z;VJ|^00A+{PcUg^;*9H8ddAFeOva|g)1>@OY5DaZWr-GCr0K^bKgnb>9M>&*-pP1$ zObcaJHQJ;q&Vb>M$@4(qKIUX&42%ZW_fMub4-)49Be_ zeN?Z*&EX<9n}L->L*eqvVV4Ex#${f4CF2K@f{<1G8XKOg@o70B8;i2_k?Vd_N8%yz za_oC*Th|BCpmscVv7205PI?EO1xp?+P?STYO%+xS4E)j ziGN!;K5)rLi;r*kbg0v|d2(-4Sg!RanX4*s=miF{|z0MYUm+kek@|Kl=_s)wtSXxv#-Fh#Y zquL)Pm4-lF!WF0e?=)g3(g*28CsvDIgP8hp%;x}jUZkT5mt6%e2GAK=bJ%n`mzR}_ zz^`h<=v+uE)O2+m6pAXW>#fU!+X=H8!mer~=+vpKva?zAEz7T}cQj$FO6ss{v#|Mh z64-;grR4VOTrWFqy>8yN60rogrQWtCrjD?0Ngt9>#OctJU?UPy1h=W*u=5VDdl84< zn&!+|ifL#y$jJm*2>GM_vg#xrr&Q-yoSI%KMv&9-`~i{&v*WL-GGr~*&+=LqJ%p{C zsC%n>&_{V+U$v{eF@YcX?OCdnR=^E$V6Vgxbk?}f%R+X<#KvSVkRL+C zIL>Vcn<{RX^s~^cYq7(Zrkfd5XJ9ei!bRu~14eQ57yyfwfAxAoWWb-3i`uiuF3q{$ zM>H-r9nJYBdB>VB4GWh3wYWgdx`o!I`?B*%jsK(gwYk`o*XEnl9hW~WS+0_BJ0&Pr zD6P*ig`~r9Ym1H;dPK+G>~e3FH{{qi=>~>|9lcTZj4qO1bfjU!$uY2j;3|afso=To zzUZwgIrL%}@cDiop0V7hO60X)nOU8Cwmlb4yvAWx=Doh{RQ^?iAO)4`B*b!q9Sf_3 zk$gl=e3Ge(Ly$VIn$-Z65=<=V))s0sssl^p3oU-s^!_PDFHEB#DQpTs`XbWE64w%B zI$#1p>J1}h+x++|QFl^3H9lkpMzRA$s!m0c-Bn$%F)yB1`T3!ND59}8UMEmGKe1_D zyP)x3D_9DjSV^mHw|-+{)a7YQa0wI;t+_cOb&E+`f5eOwxjelXfn|Krz!6!ErJOd7 zKIO&p=fe`q(#p?OPimgLBP}JkTr;jI(%r5|ZY_7+KRNhj+>&B!sg$|j9dVt~C|Yj1 ziv^VuW_bNh$zPB6Q-liiDw!b``bZc&ndOuSr)gfpwT4ePO?+Ik;jKuvpcWu&zXktn zf^`&Zg3We;)YxuPQ3yLX_>^5lBFGuN1uYV`WL?&ock~jnF(_M`JsX0O=txb8vDxW> z{A_XlFJj7p0$CtT<1;ZTTG}Qf7j=Eimy5%m;FLQhtm;&hIp{3nGtnx}D*K`(r^+fB zt(8SOzr!R+%QjsF&M$V!si;M@P-FSx-zxvoA2#hNw`-Hu1jJ8m8>#Ft#A?)mJOKVd z*M@bDp-USRt7tp zy9(!YhO1r=I&Jwfmw4#{DZf@puLik{h>;&K{52Q-sohvhHHh1Fo>|}1A8m)E1Y!Q5 zB7}hdL#W#Bsqh2;M^B_1^)J|Q2W}AHvA1YmLqBRpFn|j+9q8QS*KYs%uMZ%@x$^jI zcsR|*g#pNZt;U-+FSmft20zLsbJ|WG4RlU(s71RA#AqLIriXdWr%!k^%%9K?3qvRj zKK3zl-cFtpa`_XkzjEUicC=9$EYQyTj8w?2iIB>&Jm;$-0X3m#3`kvq%4O-zX>tTY z(Ve9sD)!-Ww!s_~{n_zJ7(7rDRm!AYXbc~B&%hQ`>M4pFUs65rdALoW0{qE;rzrFx zdRfgU9@-Hc&W~p*caT(x+6oL-)PShvw*5PwItej8yJ5F{JXTK$>t8KS6bl=R}@Q#$>-8z1S>DDVJ zv8}=^xJx~;?N@aTDFcp1CXc6)OmLT7!pc=@7kmWve=@|`0&f}p@F}D$A$^=t9)59s zEUm@_Sgj3%51P$U#8e3-(TB<1j(J}8ltZj^Z))c!^H>gVIPVG*-8j+8dR$2uRiw{}vnY_06+@?J|#31xHyKb-y2;Q6I(g*A&QJtMRva54FsPM!p7^MX7?X;`#zpUPm^*dhTt{lFQ?o2t- z$8!G%LpdyF;l^!`r;Q&Cy;IAO2~pivj_L0=M7lY=&vD}8?zQs{9zP-#jk)0~Ew@bF zY9jikP2Kk~E|lRP`%-u3jR`sGOKfC#sD$cBH%WD5Oe8HwH7vsBnFD-UtKI83dvWBP z{C~1ROBjs)<9*qUc%MlurFTr2P}6eKrfy|YYJ&L`cD${`cVSgM*w!ap29|bl4-TF5 z)beBF3wP)p?NtxS=d48DH8%Zg3clOn^^W3Wq;+%HlAE?}@pW?$kv+^l+U=atwu|9{ z-Rb;hbb+UO8){zebY*3)oOWLEjC#`Z^SbKOOixUQAtFu(-<{Jv_to$8dvb8qEqOpu z%?zcQYk42CaNZj7tWM8IBf_!Z+qfrQq|jgCuLCjfd4XHq9Eh#K7qqr7ef0tG95$)=I2R4QUEngU-yd3Nq-AtiEMIC#h(QJ{O zin+RiVK2%hWxltv!Js<{^7uPCpMVRzhg=uAZom1%l!L^dirRZb7(w>TR!6m{z;ph=LrC-;e3MF0k52ksIyde;FP%L2{}MWtx;GD&?!H z<}WS{Iw`*@oE|;pgUJOXhvO(O^n4k3T|AESDJiYE#0KoB*6Qbl=5keonUcZaL8M$0A@H3OfZ^;sWyyv7R9o^*Hq0$*x+S+BbFslPuOKP z$@d*xjBNK* z-Fs_0_lddZGW>-zYfh==fJn2}XwaRD76(+^AlgnO#0j(062ZoKrfh$*k+bf^W`57~wkNg)Do66X6tLws=@#CL;1RW>~!cTF9ziDtiM_O-5JJ_jR++Po5 zW^BV@;AH}i@tV|x6!jpF@diDOggkhKFij3r7wcMux)Vk5Guff-n+-C_7kkK}!Hlcj zYK!+JWW-A^j+lRs#iKV`|IYAo$2waL^m_!MeuPKyUGY0dJkqyF9XN^jR+C;vAPZ=O zC*pIpG$YAeG0;~ZCcx{^k{E#!+Pva*Mi+-Zo^~Ge-S9jeF5sSJT8>YfitTm_$ads| z8x$N{hmx*Q)0%cERYoUXy_tfE%3Hweb&ek=f?MDrajWw;U>mtdZxQUY&eztYBQt+i z$%DL^$(6T9fj~*%Z>Tfm@3*X=JlGW-kv5k@iCkXm1H&m$c0)khsnO$24)5W41g{_% zyD+s*-fHFSvpbc@i}k~2?s~?XEsz7)8hDOOeZvY$g17a z$iL=2L~d;}9+{K@5jweso@*u$y-aAg)lcsZ(udDQ(bFm+e&6`fDjzKI;2es@-Qkp4 z!u(rmkxl*9&t5w+`?D;&4`h*V-J4d{2r24j5nCeq&*nu-B8;o~`r^&b_tx5<707bA z0U%`UVF8S99-rkRWW^ii2=Q0@3)8DzRVC~O6%yQGg@qN)X*|THg2RvlR)iJkW95S} zi4(%#`7Kubq~+429_mC|%(vb( z=-Rt0Arc{kk?2u8AF=PB4rcirwbG+V{}L69sU2N6?Nr*ZN?%JLG7BwXvt8oqg&L)e zqp#Olhd*3;ED!Rq%%zKWSSVo4 zHT-HdXqvM*?8vFQo15RipsK-Z(=OOj94*(K`}yyCjj~UF5>xd({~+nFAu#d!^k>9rDhmso zHLw4P(y&n*{8;h(RWrH&AOyN83ic-#VWWGQNSsfNS=}n!JKYYD^`(ky4$S9Vey^V) zSHJ0hVEczZrs00~yYu(;2U?N34A=FG4pxMdA@83bAv+nCCOa(zp*oX#n;f#C%KwfI zc-gv7a>MBhKy~_6BW~B*g-!ZZ?_JKXIBo%Lf*G-nQd@HgyxvF(%(}5Iox9!dWe;#J z8}K?4udE*u8Syq#pAs3p%(|0#z20p9u1;k_=#Bm@H#yKZy#ZV2{~K&3YFl>mlmzDp z>9e_eRTQJB-SyYqL)O5#YLKnw#e9i?p!ZzcMM+t_mU1Ihd^<68)>0g%`}2E9kWllTouT*a1HbfR%tTY!@lJHRI|aX2J5sTt;3+{biFCFM;G(7GU0%!CacrB z{h+4s0l)4htH;28$r5?8b?>tC@iUnERnwR~DEh)ZCgxArwB~5fXIv~4*t=CTj(5*$CPkVev20J8DsSBSeTH?t;=TUkXiGV zS;eh7bbgLPbKuq)Li>@pcwLSzqgbOdiDQhTr&Yi=8K~z>w`Zv#94X}eS3;qEbm1V& zI;W0t)-*t!4s|m6n}zIiEHc(MTb~60N62L+fiS&@&2J2CXeE>I5ki#Av9=VJzFZRM zwKV6)x-|yMCQN^3+Zscg&@9=9H)9B6cjXeTg~@2?Y2hvQu;!WfW0mK!8co}&v}Raf zAJtm6uY@r65xVDrb^uku#Araa%v9n)rbhaVqYcGWK^VIT-M2wI^wz<|xI=1qF5>%- z_4n>md-8X8ax{fJdQwZLt5xdB9+nY4TNX2QjK>SNFB|h(jt2SYNHIPFucUWE5yLW+ zCIDbz9Q2^3?1Hc&EHMswBW_R5hyJF}eu?L`b+mZ4=&LCbU z*)!%XZ<_Wl`Rx0DEOrih0u8|7wB!4gs){oSKj!oFZ5g}@Y z5zNB%(sBP%4yeKo@uRn32YCgA=_QA0Y$|A}r#$c;rBId{hw6+~D!RG>&3J`Eg9j3zwL)bWcmK4oOXOr=OzZ@Zj>pF9Q?4Pz`U4J!aKlrU-1eV^d!pqdP z5yJ$IoGu(9itMP#+unT$3g%MTwconmqU-Inp!A zKyoN1#@|4H$%Mq+5Mj8xCK0<+Px+iXn1o4GA12t}AlYO;l$sT)m&5S!*Y29#XHlQv z*Uw;#ybfV%d34YlY^oal6y~_PFoPD8>&rrl?t0m(;R+Yj6R`uY8 z+O5$d=KO5|Smq(#td@C58JH@hUn~^XM_IywoNOhx?@a+O7^IhPjZV?PE5&SnS zlcZ*o-XAZF4bGat|6(!UQ%1jJx`Z0(ePw*}lFFC**_aec`NX?! zV+1Yw@F<9EF7hPH$=qWGwDqWrFF1K^lT==n$K>lv;S;qVMJlc=ZI?yuJV5tr`v+w%N1CXj zzjl*J<;t4meY0{&tacDx|7-VyeWgcW_?nYI{$)O-vyUI7p z!kBv0z_)aurwuy$6)#hVVfs4EJD~pyG0iVa0>xh2&=qxTh`js65)bD;REOk>Efpx( z4Xlsi5DTz=CB$Vm-!h%0=LR$ZC=J!$I=aH%cAzl&j}87nfgum&3om28Hy~y@18(09 zZ=-=qa~UBwNHLHFRnP0JvllRuo(MCf?0ZB61ylsUHXLGT$xQZCv!db8G4~h+56Qv4 zUNL!)?R$iFA)YnbUxY`>SAhT;BE-ZQfi9v%Eba!=9~g>j7YJciVb$)98x)0>yGT70 zG$o-6p*s`jPUc)~^&~m&G;Dh7!I?|F8V&QIpA*b@1o}4+vpIJ_8Rd`YpMGl78}R7ou*0GW-&75hWeCV^d0|Pcbg!jVda7kZIz8$Nd0ag(a#bbU550zk z>l5Num_Xb!XrkIo4)S-vQmE?K@joC^j@I~_5#C?HQ=T|f(UzP#`nehEynRdH1oIsH z5O1ABt~EgNr_@OBVr8ebmJuvbo?1pgopkfiR=Aofs;O@_mH0bj)l>zY`VW4@WrlCi zzm)A^wCP}-0u?*JpV^sM%N=sV8&j>qF0ezoINO1>lW$GV9GxwvjH_$E#1PdM^wY$X?Q;@Pd6zWOOLoNL!sZs*^F;SNlyi z;>v|KFrET8&A3TS*;?ge4r0G9{3M;B3Rsa9Ht&d`|Lpbwf+=vT-Kb*WN8kOP*`;#& zOrF&h$r#Uy+o<(a<8$Wfe0iK{h}%P3_WROE$@W14z>B4l3!&CM@6#N-{-+`L6tZfe zyP@pzbdv5bZ92>N-1b?OkqK6=B&#~&LM!d_bI~EEKgZ8-93`2N4e!jvYG4ZK>@BhWh1f@WYSewMywuT(#4emm(Fw|RanY)6r!sNA9(7_EG4^Q~R$Vrsk%&JnR^iVaax8`` zn$kQyL~__(264jb+LxTla_XH{tWj&VoWYPne3Oqs+cmECh?1R0+dXCHrbSCg>+0Gy zHVRXsZCYD%fK)@@{lxvWnrAeN3C-;(J0V4ZB+gGkiMO1f*+Rf=(j>$Q%`Gf<{^w{3 zT~2sHHy-p-5~!FxJlwCISh-fm^aOWL1J;MTeoz}M>Oye1Pt8c&;K9Cobd}?Ptgzlo z+WmdvDmPR^OljxvydGDd+Z&M4K~daj0^3tExVyVOJECgq*~Halq&{#zve(^3=7&m$dA z61tDF!5Z9FU1Fc5N?Rv9_G4mspiZ;EH7bubl>3THfjg~eH7)N{%Jpog6RQ%JxOR$Qd`S?VUo$%KGtgJ*+7s$Ma^<$@W=jg#P# zFtdI%3IgfD{0pFzA%TXRV7en5y`rjvFB{C1D{P^#VxZ<>**cTMhA6ctaQCa1z)oQ1 zuH=m1LJ{r^7aU&)v%a_}1?gXWa=Yk|&sBv_*c&?rqgEEnzV_|H7Hw1ng!xX26vJV` zCUOv^Me5u?27*o~T+P$gNd7%6lO{EaD5LtkXp<^z@C2Tg)3H3v1bzXK)Hr*2DnhICn!tfsR#Pl?D{Bq!#YE2*e? zp~6`IF=nw+MM1_e6rQ2q&&WM}fMDhUOVNN$$akVBX0Z91GgNez>_0O-%)^&>Z8p9A zl%KnZGS>qIi_{e9G>lBk5;Cf$wnWUs1ND1R-_;wt$*g3a*A46b&OpX@^cAQY@mKpo z0`aC3ahvw+;vv}7^k-OPCZY^kQZF1VrwVcuiR=)VJBQ9@a2QI;c8iGV#|~NI`1rx9&}VHl-i1Rxy3dvQkA98ZG)=v)K49 zh=2&yBE;Z(;$o`VBvjQuRA~jHyay#uG-JiIAHN4F23Nka2?@l+Av85TPmu$ug*1ff(~K(#z)Q7_{d0l%XrBD zsZ&&h@q&vE>#1%H)HgE7TEgQmK~uGq(ldz9izYPk5B*`!G205SN!o>u8up!G$~*>P z1S0+6n^_-9UX?3krFveKdVE71<#@UKO4}$M>OM}X)<3_ADKf6POL@1l(7X|DJ8ec2 zeP#P*9)nNT6uoQEM7F7xC+uJ*=`)sXI_S{6U|%WhtgsGo(wg_&>5NpyD|#n2?E-mh zZGE031Ke_eIudH-Rdo%z^)APgbcH!eggoc|F`F65u;DG;K#r{w(eJZY=ilC_3YQS_T{y2T4FZtVYl))tY9t6=Bf@) z*x#;1+JcEEcH2*@i5YS}ReACOiox{}=m;HoL-i3`DzL@Ph;l!$GF}{>dx*LO#}~Oy zVs73;pCK5y0q{cC0OKWk;&cvfz%s3n>V#`FU|4+26ln<7d>KFn7>0_ew}n_nuZI}Y zg}DAD@FwZ^b9d~U0}L^QB&Oe~oZ~&R`2@hY8c=GoYvB7hB!+FOQMf~r#?%bQaJ`kc zNR1_k64cW#@S}Fd{iRooV}xFwK>$nw!Riqs)V^)oGXyn2V|n(A(f8f|3fC@B$xxxI4)ZA1|?v3Tmq)ytVm&UK+CT1E=mKzS*ftJ2cHdE(FBGEKwOdSFr**_9L@35R2E5$bslW4y zYhrq>3$!fXlkky=Hv+(!#aY5C-|k-Mkh5KT^3!KDJnpf9v%-D5L4#$zOV5AVNrUx8 z!JLfF%2uPOS@ znuj(`8}mjt(t6CPp&bu7GkQ36hZ6?27Arky!{$7OLe_AhP3}~d^eU9dk}8ut>rtTu zRm&;8Ig*uiW3bpGq9z+&i|G`+XHYzY<=p}`8z7PHVkxuE8C-)blP!89?thQ_5jAb?SP4YpU?L!l%#e+c3}REOsmOM+b3YRe8?tKy5eyK4ph( z`3?=CO`7axmTHK+D)$owJ$bTAv0sMWc0?fYcTCN)#i@z()ANgcIAWC=N*Fnjqdqvw z1j?=Zsh**M4)?yZ#Mq;(Iwh9E>z8&W@c4*H7PN$P`cc#~rVrLifs4k4S*XmB`rL?R zxxTNGx7P(g?K4r=9s&y(vpxH)@_OQ*q#gFO9dK>3jA;~HGl&IPHZ$QP0>Bu(=7ToJ zcVUPs-s{hU*AW^`?&2T^i7bD2y&aC_a&o9ZiQh@caI3VHSpY~cZ9DVy|<@PIAk}jQV%jtDVT5}0&{jSXzOxd8XbbVq37OLLBQE>2!APcVQap7j+QyI!Zp?2YSDBCW+4=S!(T5-Fm`SrZF zIp#ATix~b;c3dzY@@(!oVypZ4^HdsMrQONHO&)uG*}KgNyG-p2#BZ2N06^N$y-gu@K(G#xT=RVo#3Esa z>A(>wM|(p`a5b_Xfg4Pgd)sjjgc^a{e(-^j7VHwDxiwYyz?%f?dTX1f3vm z66IhG-{iuony>C_UjD%Sn{e!dcxOKexVnh~IUof=`u^9AC6!%eSKlhesu` zp?c$Zx><)9TNuHaF|D(d8m6rRx z%9i2UNwMT!vF&2YT13>%yOwqiMUEr6&hvz`B)aq>(yL(3HyTK=_oPER2?V3e7kYS3 zF&AWq@oM&gR+Zk4A?#R)8I%=7`aVz#ehtnwSQ+bTycI3#TYPtQuuGeM?D{qY6&-R0Z~g~J>h&ScCO*-8 z`^~0Cfgt&YxN+l2mYY8iwtlNT_Al!JZ5Dn|Yp*?(nU~XC`pC8nNnCfzy!O#h%5bo) zFVW{m%uZNsXB?xU>C+el;ioXi-Nqng+H>v3zG7IqE1Yg`n&!!qU!ZAuSbGXxn}5P^ zs(srWuivt?XPg)$EbDUJqqEja2q_Y#+q*RSp7K_H?cAf4V|T9ZiD(8ADRv1yTy-C~ zj32bBKUr}NuruzQB(Eep}1j4otm5!x5~t4pJ5sq_$PgsOAQS4q&L^5W)gu!F}t zuxAy+MwtE6tWd?R&-`VwL*XcKdQhc-)CXCs+BQ53Wy^wMMQvFAba z1t$a^A=K&rhrf{U8OSzd^k$@H@C++&hU*Nv^5xKWx%klii0&g?@BE5 z_{$$JJAu)jUK5%FYn+}!Z{~m$O21xPh&84t2z^38)vJ-OxuDft!=Uo5YIFNq>%fy)hCM1s>1lW1DYL@y)e(>n!wMKf>-K2^*rrSXn1BGa4OUn{eQl0A z6gKAXbU>E(=Wffg94<{VErjiYU*-z*$t8vO6UbasgqNN)KAU7GHc6nH&>Nbmi9Qg* zOVA_3q>C1F;ww+Ff6LNz>cnv^&G!cqNP@i(CxO`-n8~Jfc zlAWY9j^(hkRKBfqej_$9+sOBF1=h1cONNKLX2ta7hLy2yHYOW_qF!6(b+$r(i~W=Z zWM^#qeKxEX9&1+}0vLtz9;?f=a&>-QV%5OKG55;E6C5jD#~-z3XKwt6`i6mvTkL8; zF!iq&raJXT6K`cB<}T8n;k=R4y2(w-g?7z`7;nhb(smL#WmSm7&R9*3)6TfQwxfj< zuaPrm9Cg{JX~?#9Lx{uC*a~qC&!LlkVgOidYZ+?&_*ckfGnOBZQ1m_w+2s39sHS^I z(vIj?&zAfaH@R`D5b1VjZM1A8p!v4IkT*?>Z<&ZZUl$_(Rh)=?RdCv<=&o?r%bdOox zQHNy0W7)t&eR4<6Rzs87NEq6ek&&{f+n&q>!t6wADN)u-$qS4qzg3l?VSpoEpkb%AO|(nd%< zP0Rbr_zz+FCI7F&)Qwfk*8(b6#5d$fFqi&0{tntZzp=+0GsjMnBN;sbuft!+Js+2~ ztq52U2(pqqhY5YJ+YVG3$rq)F?I_-_nB%oHql;V`?vMn;0G2WVKi4S$vi(@_emC%b zE}_LylmP9FJ&U$a`XdwS;*fryy3ar9FW5j%-6q8XxUvD}!Yf2PSrGQjwQoscdpZ-N*PjyhN+KTSla?_5!d|+f5g|Sy1-OW8PgR$XoU0?2-5YZohSIzeNk= zOMkN{{O&vaoG&XnUBK>v3^O01Es}jc2f!}Cn9obIL>lFn8ACsLthW55x*#O+G2EpN`g0F|~)r$isaBMA-NkjA>-W+{vZ3_^J8 z#Zz+f{xp$xx5k;|rv*%K5Vu&OmrB>!9Qa}DzcGXpQ z0Ky#$m6B7<e}`zod;I+=d!@S(GFzL@BI)rb9+Fqx z063*1AUKuhFVB0aF+9q#6(6ZThL$}Q<}W|-c-z)*HcJH~K;Kh*?^D@LiN%?GFun?l z-Lf>;c@RO>|KmR%=>#Z`9=I)V@0uUa?si|1PLdAU$EZ;|0+F0 zuHTW{yTIe+*NsPe#&qx8!IS0`pGT^0fu&-ocD(~$m3*H~dzoo-xQ5Y5sQu<7!CF)B zJ}Y5#edBq~o*ee6wvG^HOan%pK~s!PS*u?~X3YG4RbtHE_SJRpDHNUV|E%@4B&!K- z5E{HB<(V5jOX4FAv0mURu7L}_{j7J7hW&!lgp1ir6!KE%Rz+sU{B?C^#-1S*>%pmL zyS_bUMnhsfSVr+p2oc7ta@X1f9vG18onexXwCEa1E!fbGmOS&h84)n*eRdJfoW+$C5|4NKNBo=$Kxe6y5Xg+`@ zbgw)lP`)h(GZ2mLx?OH|6y3jQ<}`=yWGG8S>$xkw)}OD<$@<^^h;$zvQ@+o+J@*6L z=poS()_u%jScYKMTfx8pj z;zCwh+xbli`J;sB(j-8zds4%jkVqjL!e_C^_gkJz$jel1u2JA~j({kZr&o7B%$-j0 z`&(k#y1#&wP%k{}DTJrE$m63Q3i47L^2^JKR2!5hdPQ&wGB~pyQE__(K}%5Mihtka z(>D6sjB=#DJsT)XIA3qD{*fBeIP;_?lia|8f89xV%QuGz`&NE6g z{YgjpI_WBwS{Y7Nd?=fAo9*mU30pCG?qlD*o(=h4Y*%wtr5SZE6|xD&RYJ-I4HI>F z<_jr9P-B9#J%=UA2h#p=W)20jTRyHYWn7acGRJRBbJ%Y$?PCn)KYdHqh&JqZZop4Z z7T;v{gu!57%@{nMi}X?v8Ssy$l=@kt)^~LVaUS_Uf4;#Qq@uYj&>N(S@W5G78#G9q zd-_L4?0xkE-88it(6u4baG4{nTyy>jgc9d6x1zSWz-sZF)O=jLtOlzE ztMkHZtl?XSX^GM#6fP^(+F*{C-Fn_@6!7F0tF@Xsxtw%8O$jv=MM!wT-UGJ-~ zY=16;{vxtLM$#2*3`3M+nV&M^HZ<{GzUCA=?QZXnP}sVA&WB_oDg`9S1g#B2kYE*@ zGEz4<@?uXudk_rwS0u`=MJ3e9>n{f+5R*&Sx;O|`KvR24kE~=8!19xm6A8ND|0Xv9 zh+-CeqE3~njfYaAlU&wQQr$RlTT@f2`=M1DsukzJ_j4UV={_@I)~KL80{pju*#v<< z|E+*(l!225GCzQa+odz-{MPS`2Z~#s{NRRe)3oK7HtgcT4BYl69U;y-{&U@)(t=O) zLE$$w3>QO2z98ftk*PVV_m+GV?Tw+-P!X-ukj9*7$?7Be+nLj};8&#y*f`|pN^~-| z@I*@`(3LU@)jU3B2tR)p-H|M*Jcn6b)k1*G{rHMMC-*mt(a9-|STz^3(M==D7>|Ft zM)u#XGYMn7{v9@>g$)kDJ}!;YUjQQvnE#VS`ZdqTuh0JP(B4de6ia{C@@4{h%L06q zjp%e>BM)?29fFQQT;tvzg2rAxtZ;-#IUIt+dhaj^M;%S-}W+vJo?SneXXYPzo8=)_1nzpBc~O18bqjN4~iP5CSsy~*JuIP zaOh1}uQp!aMTHx^Fc_x|eWg0`x6gVNMVMI`N%IbDJY7D2_f8t6qunD7V4$5t#WEz= zRH#mvE3m3vBCkVp9GNoyb06uZ(Y1l~<><&#w9=JdX&JV!52#;$ru;S7#FT|s4X%c>3}iJr&GE?W1cc4Z45VS#MV^? zP5sHnLy+KfFOjE44}Gy;ujz}5$%yIY5 z*FkYQ8&RqGl;J@&+V!N=@iAH~*foWU*@ZR#TRoH6dvg6xTA~1z$EojUaBB&?4Hu%! zAihw%2BiV>UyNPu#)xmDoONoEI4`46nyn!f`*Rih_C}mxee!|JxokU^yzBIx4C$o= z4=TMZvCaNIIYfu9@Na+3#C<_Qu+O|=U6`}5?`|t3&n+d}WvMHpZ$H3 zrvzWd4(*N~gnWw+7g=7N*iMmOwbz_13sjeqBS-q=U~eJ3g*HhJU2|`Vf_!GLU~hSY zIyugv>=sDwTHr405;YC<*lE$A3D;mBdc2dHFT8cneQRO>YZq}Zi3@^P;_K`CX8Fsb zs+RrU^A$46p7umwPo3?~kJOe;?Ha(j>xg%Um%fzHNz{wB)4N6FTgc`M_!)p62Pp{-+0@pKU82m3JSPdp`C0aA!nE z5HnB(u$LdJ#jxW+07;;LyoW+19Jc%HWGrEg1}^bxxsFGUCbl$G3?VVUkcS^TWaVt& zz9B>bY&Io-iPYiCpiF>#GGPt1?M0I_f)k<<>VE+xfi-D4&K^_Rs(!ayHwCaNVjqf^ z+;vbKiT9-~nj5@ukaTQ9wtj1X76?`R-xEjz>K=Y;TlKpzcMFjD!)A{L{_^9#mQ_hD zJ;uNK;f9c6@Q69P%YNRs2s^q%Is{I};~hF96r5ad4l0is9#(dI0>IsbZ+b@G6 zgUF1hjlv&hH0o(!CfL#c6qswl-oR)TgyU>BV5STe+ zISjRQs*CcwLaiVTGaMlXmKZ2R=y&%CI0_}8F6GCUbHbOiF|5CKRSoIo!iwGWrlHT* zYYQ_3_0O|=#UJ(nVjP^dgW{Qlx()Lmwq7}V18|f>8%sCLK^4(0A}ZHBbDdi(jIah| zrM=#IiLh2?+;-A^BG^fsVk}xdp}jL})u^g-4JeX+29gM@CSLoSa$GXxe~)67 zMdKK6N)OnVTqZW#=3t*HpwTRZ@qFsuoO_iZ`iLgeomPI0YrZQa3;y;b@L&-Y<4sL* zYn?TftQ4$V6l4mJT?I=)iM3AI=#^A#_iA@{z468oQ4sN^5o((*e{C%gWvI?0Cs_X2 z+*m&6h|17-{LmFM2^M@8+I@$v$_B~>~X>$UE)|)M=IT3H_s<)?zdymOp!3}7gPv= z__P(GwA0D``*z7ezaqr+_JNShR4?=?3n9jv5kWx)Db2l)Y}5Ts!(S=RTR{BbyHxyP znkZ_|w4V;cVLVheiF8@`5*x)~R%Cstn^fbWnNa{!+1z(jE!xyoPU;0|!cmLL>v^81 z8cYAK1kJ6w+%WFDGy+9TR~1oGwd)Xc5d_zIO-?xkY130+j0jSpLRwox{GY(Iev+Wb zACaO6BtdOBQFxR=ISx#xBCwnS}_JyOA!~9?t5ggb0Ee<&x zX^M5CoCsW+orS$VyvhjJ22@U2oLX(}1k{ldeJPtXmL!{UH3jiEa09)R#)HG6h!FM# z4jCN&d7VzaLCI=*uwQe!Z2gZBbry$rIFmUAoI>W?HN@Ybfj~{!^r8s=G8W559*ni( zNRYT^I?F^@KH^2+G`ureVrq1F0!l;)p4dzkj-vOk-OjHq7Dd`Mw9Ja?YzdWN@7f8J zKqQm!C%*5F6wBw%u?c_UcQ-%kTGB9bGc~-L#yk>rU#U5d!--JnN^1&4h;pM>+CXP% zyco|ViZ-1Be~J;ZX>L)UvDuV}Mw3(y^Q)fKA5|;mB;+I6mOCt+mCwozvH$7? z88&J$SgEQ~6;*|4JZjji&V!g`T^g>16+tr05tj`j%vx0`s%RMXhChM~Ez44* zv&(0%8kA-7*Q>k~lu6@!V13l6Ojj{xZ3_&ODytP_bTH~KyLkMwTumZfdca|n$8rKy z)N@ljsf+4mnP0Z~Z7uj%J=~Ilyinbr!ElId^POe7r>_k#SxbX+v!(8aZ3>g#^EZ(d z(mb{r$)`y3d(fV}O}1Ncj)TcnjN>FeS>-Bsi=Z?oI^2kxm%5ll<0?LQllk{Oc@xLe zmL1!$*o^Zn{RH~a((+am|C1gzMCPCrjzyu9xzE2*Kxh5RJQkCjJ}0HINx0T!{AzF4 zvq2r0QV8TWHS-Il8aBBFE3`*5j^%9$xMv+I^Gc&ZH<*S9%r-TMi={30Bti4kSe=v-`MCnQ6jUhGlIj44JL|8=98C4LB7NW&ox-RMZK-dj@oh zP1l0rEUOEoVXHSLNb4ipM|g9(`w+?3CBCoQNTp5p%NhxMn&w)g632OMiw-N%y1BR~ z&U*UO95VE@Zm=C!Mbwy+z(oA0;HWdNlV7mc_$iH9w?r6Np~X5Rf_WF){|CN2k}mDk zja>`#uWJxpW!gF{NFw-&yK{;PR9Hs4vO25g;{88zWk( zIpbv7!XmbdP_9Fe9>47Mq;?>Q6v6v|n43=vvUBj=?RCQoh0r`f_Wfg(ME)A>e#^Zdn8}>xW34d} z!2r$s^7OXX7JZ@7nscyRtqV%@$}Q9ATzBQ+^?XaKUpxkUMn;AaOh)irbl9IgLLFvs zccZ1shQJ8({r>}{4!@

pOYSNow5Bg3C%p`g~D_5 z39jTwfQUsj<6_GDP{2t~FuhB!{z>$KecIasj6%wS(K}1G-aR1ADSGi5sxf8%?-%8+IPO@AOzcyFH=~hxgZ4&+UR1Cs8ZT%Hx!ZzgYB?=OZ zhHa`hj$4Z`N;C3Gp^#aq;Bt-9pOeSB{W;I!Gc)32Yxg|lneh-5*nzH7kKuY=&rc_Y z)L~V*afe&#iowfyTQ%*WR>5&*{aUez(!=VmVj*!wnuSeWL4!1GRlW-ynb01ViW9eJ zK1_=l<&p{GJHYDhZN8|up0$quz(NL$D}HO8T5nCn`k#xR%NvYz*6;~mTW8ts6QQbm z@NR?Zbp?f#?tDMOuvOhIJ(hz@;J#GW^a*>Lf7x#8jz-X~@bz!K;ubAVdy#8gdiR;8 zZ?ir;MT>|bY<=;@47;{51#IJLir8-KoX0xV=kvdF zU0;sZS53rb46112I&{=_pEvDIAvlXw`g_@}qX z#yP$QN^xCHdGqZI?WKEEF{8&*Hwx?YZIS!KBDUWPcS=ZXyV^ZrMjCXac(rT6f*Kx% z)iC-^c9i?Opc27!``^3ok<5fwMo>x`Nl}o=Vx0^MD88~a$#7_{PJbmR zMMMVgUs+ZU3MlZcMVfwHLxz`;vS`{miSDqwv^GrigX+a@K-E`Y zae^TzEs3E_4S$$fr?cq?TSDNo3g%D&B0;F#9T^~HnksqDJJp$4@e*>ObPbk@1&JPRA!AuN~-g;FB9jeGU?HgYii`%X~ zlf83^`pF0w8~f1%X6Zq;j+XtA7FpmUNiL3W`jkd{cK^{h{|owhG6ug^+4dfu4^C=^ z7{i^JhZw^LaDKJ2yy>pQ!_$E4P+~SlO1DHETYfOryutui9zLy4MpwGr|*tp%1;n!N6EJ_YSkJ^6%3SbiB&z%WxmDMO#)=g&2iQhA&p107pOHyuG*Z~QFMP& zQH4j%!8&Hg;rWV!=L4eQ*ges0ikrB%Jffon$bmVt;cvRTw;5T2PKG8mKbH47JovG$ z-DOw4gu_!I((5F~Q)TeB4k?pn8;OTVVoBTsc=bQTvMC=^7_!-OlfOUw$)BLI7q2~e zTwlVwJ>+>>&+Ym{%K7Tf&0VMP>R{wXeKyPI6<^RE#u(B?9~O;BWF!zcxSWvL;qCZz zH4puwA>-O;{mFVBZWl)>Bny%0ugPHcS6`c2qWt9m@!zK90v#PvyQV(+czcxmtw@vj5}B3J&1_m?fJR_Bx86U>oT3*t`` z(^TCpTTNEx)8Bh-p6OZ4X}ZTyw*&TrhBcIa`WzdFo;hueyscKc241;GbN z2odcxNSYeJjZ2s~LnjCky%He7a+XI;NWiuZ$;W~d-Pmv8C(%v16?^YyOn??W@>CGr zou9O&7^^pf7LA;)21$q$DWL_a1a?8al=S4Qiw=`0>|4(Em$2y}O_k(rf<%g34w22l z@sse+V@;3dF|?xPYt2IA1H_~uOWtz`ja>J~Ba)Xtt z;BmleICWMy97Gkj*A-6m#qOx+(Tq!4x~NA9-zyD!xuORgCTVo#1S2J`$a2Fk(K2r$ z^dcWvf4ueG#k~1XxBCc5Is}$ObR9B}HuGE-&+LRT^t$m5FU6QbHt4kkV#v`)5%(^v+b6y(7($c7DMYsNp9svR^;YO}v1YQwRaJLk}P=$rp(Q%`Bs#-O z^Jk7aVqYTa<*#Ha0FFHZg~M1hNR4UN*8F?WSqdPo16lv^@B_MH`zzd6jHKv~njH8a z_eZlKYKE2-O$c3Mk}(ecf{Sx@#mu}f_U>(lD}Y|Kw|L+=;9?`mV8THR9JQ0!o~ z4er%?A0FEK@`J)I6|xJ$67yiE7>52nx$B!G?^ zo`|whCXZi2fv-Y7#DZiw7(>+=elSn|ugEe!{f+f@_db69(^bE3x3uNLO+}wb)nfz9?aEpco>h|EuY|!`W*8KQ4j3_hP3zgRp>0B@iJb?>WZ5}Kven~8lBr=hMHBpxFXM<7> zSYG0fu8lhzk&lkgAlr-CXA!qvym(^olYI6#`CZKK{E=- zK0~U9j=jtSTX94Vf72%hQCw&e?xg+sWKC!RT_|-+Oa0p~j=tez;!!*&R_s!UVjbty zJ7puBN|cKo%#Hj4{5HDDS!VTE+=4Y(D03rS6D4F9BbR>YyuLzRm|Oud+SEYD+TUAS z*pCwTAHQ%CUM-xdoBfc~Rum_9Vd9gSm5w@KufT-xJh?l;f%?eQme0V#FMRRKAD5f% zKZ%ieRUpDo2-j9n#038eQGxf<6M-9m0|>d3L-5aI$AX9p><5@5?-}4ayyiE@dLYba zJ&<{ZdNq0v83Nv^WVKUhr0)T)?%WQME`0#l!wx>WvCBP3zr+I(7{uvvYUts-#A&BiUo>;Fe zN=-C%DUkyiFE8-y*A&!IlhL1jN|!6Cm*86Tg!fBttErWJxK?0F4-*}Oe_;0UkZ5&?hTq?ssev+NYs7s3ULp&(5ob$G-ghrv(K2mfK!s6?O=%lcB%V& z8+BC(e5H_gp)h2=_b~QLt1pBX3SP0W=H0r4iahA%Dn||F z?bcqi^u*}Ad0o`4oT2fv3l@J@*i1%op3+0C5>P5S4~Zf^_Y!`kzjUd0O2g4QVbZoE znO40SA=QfM-3SBM|0Vc@sAij4O%=Lm96hc|!QAEU{qXsWO$~n&QtT)_q9e`NTf#0C z*urco%bQTY4A!>xf^vPK9iKX;|Fh^bwb5-zBvv$-w&0#NbQwyF8s@HNDbjab9sZcr zVY9JSZ5n!x2I4+MNbmE~okHHK8~9@*lKT=WtK2}TEkL2iTaRW+6Ng;vP z2W%j1TdpJl5l!H1j6gw6c^nm4JdyiP+!nBH!Nrk9+H5B(LmM-VBZl(06`OE$MGHfA z!nyC2QEA0G$zqXic6!T_!vTKs*)-Zfzv*Z_aL4ugCbd_>8lv?ip6LB>{QfE+%xCTL zxrnRlk0pZRi;V);8OM=^2MS*mDxnW?uD=dtS081i3GL7LMJm5+CIRB^rGJaa_&d7s zCAx~)!qesX{zt{M&_MRadIbhoFYAXDdZ3hoPmk?Oo57i%;S!)Xl$L-hZ1w_m6~am( z3fASpQ?wA`s{f!~gHWSkQTeM1@&;%VUEO5?p4sWQ$oA!#rSohuTvX^^Y|^+=+hc)g zueRKo^P)~xegm7Ob0?h7R%Z17jrb^~GV}Jg1$fVVcF>&Q^O?T~)J#dPhmZ#lsw?+Z z$hb%0)*;=GY%SPbOhxNo<5jNlAantZsovvK#rbW5XL|Z6=#lujC397MMh0)--1^pgEEc}Zg;aa0XS}*mF zE?skyq#Hm!<8h18NJ`E&PD(JKjX##WDntXFAQ$uxo;09uavSxW<$TBRYRg-0zHIN?*-CaI>w$O3>wwNf?+g1t#mVGrr*CJk$gTGMJh45Z75W`_tn^3hpAJ!&&f2?c;xU0pF>*EixFYpbrPEfFHL@CH zrEF?UKdypjiwXZ%+?pL%Nw%KezCt$phi}XEng^T{Jf$HCAzM>=@ThQF;w=QZz~r(D zk);jytUP|xfx2lbBT+mi5h*r|Ce0^U+lFXbtpV-Kxd}zJtKad_mSApGU|m(%${5hG zL)q~h89Q=;sNAbASenv&cd+yD^O5+Ee6DP3m`^~k%saTa!HqT(Lh6B^?8^%5kOIe8 zA8VhA6Ro-8^b3{GnP=zQE6 ziqaJBcr0LQ&Yxh9kjzH0;cRlO(DAws)Rm%r=$GZ7Ub<-WE`&X7EzZc%T+k3)MQH)M zuJOK{?pya@ZX^c%*_y}{Q_*E+k05^;?JM=Y|C%D!V$xlqrf?xk2ZYYJ@n`!)HAz?= z9V$L*WfEo10FJ4!9yKG`;vbM1CQE08d2U-uHDx#1L=u(}r6g%CNDC%sCI*bse)KDG zWx0Q4d3T_bu_)3W(UCW})b>ZZknWx-Jl)%ozv3i<-^G)|$W5C?FZl-hmzI5Y^2Ua! z(I}C29N-9WM|d`-HE(pgc7G>ClDY6nOL1_@8LqG|^3k7NSYpm9owdt8kH{&d%`a-Z zO3yhit@Bagl;;ogr(1rV2Pkh$7pgFyHOP67Mog@0&ikm{C%;4MytUW$kRKI4(t}B$6sD)JpzEPg%>N)lxlq_07aw~o zM|e^+tB{c*{>_pM#b8C^KW!A{xuEhF8SRdN(IN^RCnG(CU8i$m#fbdPjkPRyM`N99 z-@jIXjTPh@L`JbzcAG(uk9(Zas7O{Xxc$`l9e0j%=rv;bGl`~--C6& ztaGB|bvvScv4OzVC|`)563eO?zxQ9QY&NH+J1G`v?fguy&s?akn;Q)(2TrNec!8!W zoW`OQ6JDkd<%Dttr#t{}@kv{_@EG}R$uZe<2O5~B~q zclbV5*4rwL848l9=cQ7EoU}=YTqva^J9D=WcS?A8yG6#*T^iTD ziwi~b!TZTH*A0U6Hx5H`#zPYt_M?cx`IE3*}m3n!1#DHO!MWHN2})X!ltlhLG_^F<1_s#a0#_GW*yj;Vnj~wskg?L5qeT z_!KVQEc4WS&5=(UGIx_-B*^t}LDMOz!8{4Ie`PK-W%nt3_FrtNS&oc({fA|`w;tjr zIr%kDYc!mq*P7b^2f<|Ja?+*igu>*ij2__AXks(-9Aawbih!oB?|oe~Trla;XA0*i zY2@Xen05`T8=~NAz$~xmdj4=XP|ySXC9D4L{fOvl8s{PltP52ZV35L@D7y=G&PH9u zUqUxC3_58|>@VCG*4(J4bdK1}#-pb6-@hF4vfrGb&MJfq64KBXT*X-Nqkt>JyR>~K z`8)SCGtwmUMEZufEmayWy+fDbzIktmHCfi)PemssN|v_NI++($YRt3V$2o5%%Bx|h11t7`*=GD z1-!YTNb35yQ$9BxqS8pc$e6c9)dpCM@&>0}0W1!_G{$p{dBGSM+Hek=31_aC7u_PV z$I@mrZ;$O@zZsE5hRp=GZ>!_gEMnB;I17Nu>q)89MF?y48!ldZ!4Ov1I9r&KJqmh6 z69(*n$A(dbiMXQGeef;j6H~}AO&s(nLD`+OIlP>`)GxzQ=67Yd*Yl6Q8k10p1Gl(K zK-T-|1K#4kDvCx2?$0f7^I&EEDe%|vju6+!&OL7b9Pf*km|nZ7ZKUlm^i}dD&MQ=t z^y3NAhnT4vyL1+Yo1*OKQ0}-}-JY=VQe^ICo+sC9!JAqOka)3q8k=SYi$;dfobm7b zeGztIOU8{1CvT?nITy>MON3RGH?*0xkls*B8c|?Q6TjT|nTN$)_05jYhA^kmrBN4X&zCtYrxhgm?AK3ZsHo}fuP`CV?{c#C(-0kC4~B02~? zb!qej(jDJ8*{^XPCJN2^XrZ5Jj3%|9jxnq!qvSa>e5EmGEu{cb-5oEB=DfDkNQ>^o zQBapQWliK%jMg}Jb;t^+yw1{&t<-iW&|#0L{%0tpavnCp`qxloqHwV*D%@Q+Ko#GJ z+_&Sbjd3xy_N(K{-W=+cik{%qa9=dYvfwSo*1jJXt}G&He}ZVgUuLsXg$Z&#bm4TJ z^c{QhDxx$^b~V@XAP;uIHq2(r9%@8)ce!50M5~KECZBxHo3~2&H;Q+LZ-x@wXtvYH zm;dyoO%~{I)c-~_ZTh zyVQxJDlcLyPZWHyOWcjA-PFT=Ao!5QXTH8ZrE2_*3%p8^=L(-ML>^4rq#|FQfd+!y zN7SX=Tz)gG@%8p0#NF?W3+;4rgLoA7`pExU4l()d@UeaL1$)Aw+nz$(OLZgvFik_D zhVHzJ?6Psu6q*Bh-j%eF?a4D~+{E+9>x%83uSAO}bxksu6(?%seXcD)7P_v?~h3Y5zX2-%(WW^gpTHg!Ne6^0VylZKFv1a`F4qt%%v=>1h4DUi08~Ho#~~vB<3M1A(c3;a3B$y9jKLu&5Wp+ zO;sIBULBs5Z7bW#VO=qp!}jK5rAu4YHqxvVd>U(d1f{%0o3JzLy78=fp(^E63)>6# zjw=S{Gkr4l$LDOdmsL|`I$A#IWXF`y<@$ z!%nYVuw;F>7Pg~{)2fc=Ewg)jS=Cg|HHF9e&d0w6PMO2mnfmAamsD-;(@OBSFEA8% zkA4f%es3pCSCrjv?N|*`jM^J+6OKQi=$Qt1j@>1nR-%I)@p7G<5kp7>&zBnqmaA5! z@8urQzQ`Wg2u!}qRFib(kI*Z6d!gvQmUG4=%bXu#ZLsZLom?6~0=cbSmo*kL84)8z z`WY*`9h5i}5o^ox5g6t%J+_u!b-omA-{hbpzDYEvhjCex%?O?%jra9l=uRQ6smFp1 zev>=kmAq^_NC~S?8H`b)8f-?0l!=0xYSA3R4>japexmlVcd(pdQ#v9A^iQdr0b4uf zCqjXeb71XegS-=SeTcPHV?a({J?xaOK7`Urber@wefg2Qm2#Cqt3&o literal 38274 zcmXt4B`~!@ofM`+a&pX+L#X8BD5p^sOK<6jyRpj!;gA_=2J|;Pm@Nd6&9%M8DWV ze1h<^NB~NdWB@LNajB)BuWU9F+A(a&T(*=9Ygv^&F%q#ns2St3Vce)x@e zn)hd0d3mFu-2#c}xpf)P`ZRKBC*OD9)`%;&*s?77G;p$?PGof)(5{Tw$k zk6Lw9>?6u<&8gZT(F3H_srk~1Z8LJ+o3qF8g(H1;3Ey}IZu@~N|5DEO;oRGLkC%Jh zUEw*;@r6qZ6~36ViXB7QT2AVUL*&@XJagKEu0CE3@^G?o(YmzI5r9l#*8a_YXSjD| z5@AL@Vb7Ke{)ntjzkJ7#!J{2NfW34fSZU9`miYR3Xyfo3S4LDL7gH*3UE z{ovbYjly7(Dm`Rf@IEpe?q0xY^)@xo(pe5X_y;txbiQK%sIZ2KCANP>mZ<;zsfmbaYPSV* zH6{AXR*L2Ew~e`l9PvtQlu|;TgjXff+pZcLA3lBL)e2*0l$I~q`oUv-DB!s5f4Cw$O`S@xmc!&{a0{`814tr;znmaCQ zJtSQEcCGwq_3tFm+xItfn%K{lt>QG)S^uy`3n7$=ZS%O&uR&`qLl`CCFau!L+v_-BcyJ~#bl{%czjlJpaYPJM*jgXn ze27Xy#=?shi>6{a7B1@|dOfeqqE`Fm3rI{72T5}w8&#}botu-8*G0l3d12=Ipl*HV z@;0B1*k-Qltj=45YYzgwlw613Upn7boU)@vsGtO(<+z~O4$z7m&=Lh+zY3hv$r*D1 z!vufoYmD_3^xnw*M&jIH;k2jdqPQ5s&x)71q!O18oyhW^Sld!`QMA2OAM5KX6nx0Y zaD&=FeK3lWwVkBm-=(C!e1v;Aq2o8^xW&ktlSc4L3zW6B+v574jZE@`QeIKtK0}wO z?qzXHapgl>dZ@o_;KgBv8~w#sFNcA`mH%_0xmeQl8n!v2<1bCcdN^U=pYZkdxLg6L zTN2)rJbm%2fhuZ6&b|x?5mJ&4Te%>o?s*TZO%Un~`N8%+U}BG}Qu`Z3U9lo(4>ftM z=87FFV1xg&DS1kjr!Dirun)rgy^TTD-dw+FDbZ(sd8vig^L;1qk~1=U$EV+Fc8r z4;H}tK3psqUHXDbt4EiML~YRL4%n1VKZXrUW-kh0kE`Yrv{z){c`vhF^bow{qTDh8 z5nNzHc6oQ*>F9SC+$csfOFU~rmtaq6O-_)>^xd~gtU73GKtcJ;4{OU`OlFoi*7rsI zulaOnMvCU1enAM!0S)9c^QQ?7D=^Ie#j?kL#A1ywCJAcBn;oC#oqC}%^ehFb9ar`q z7JS)`h6dX6D#n{N$OySlZqgQPj2Wzye(^+fcK>uj5Drd83)tfNvq2RphV4bo9_A+b zB>~>@<@W3C=@1Dq>sRcOfGs6pdDySpqgxV}YTkZ1yJpC+6fPJbn3%+Ij%^>*)D zs6VSp?7D3eCU*m^ze~jV_&uRbk0M;65>NQPjO($fS)-K(%BZ(na|h1hsc8>SZ0qUq zmM%R^&SKMHl0rj0O#WjqN-2Of9-0v_rCe1{`l`#ca>7|b{D_=T<0>ArI=y)?8C+eD z`t{4SoV^V}*8lZeEa(Wg!en4~xkrauqh%1A3Xj8NK$`N(4irvtJN3Fb*T=9}wNN?a>NaIV091>90<70! znt+g1UbIPBHV&6@J+c+@Ve@EEXuj1MO1}|)r5V6oE@V|8q+xP1i(;p4)rU#5f&R|& zIgD3g&|9CEUx}@Zf zqAjfSaN3O_Eg-LK?``3bbCm=@Es{g7tixIzWpK8H^_o>n{1jnT>#64t;JtaR-+gOZ zzZt+6cv6$-oFb-Jlehi$t^=mCSJ*z^9^wr*8-MC&!MyQD9RO--__C$w@E431H4Pdq{>7^saf)ll8fvuq|2m zcJJa3-iWVVkiXZkuvvrH9f(CEqN%JnCDllGJ0BzOOC>f=H97|g}P*=Il@fji{YC+kgA zT(6N2Vyok1xWU~&`qTM8LXGO=8K6(-?tkkpSp1OHl)38;neGf(L!-F|of*A+^&ivB z++|yVAfV&dCQ$i<(>}OiA6g58Xt2U66%zq43cm(yclWQk8-$w;e6IQQiNPCRblh8m zv5-HaR*PS`Z0BdLdJFvlVG@Hh78@pc)8O~0vF~e1kd&nFN^OJ2C)2SZ603ZUDt}D4 za*8gzNJ<`*3Uxq55fxcGtRnI93L-bjI0dc;=~k8Gi^i~C$J^B|kP!Rzeuyu*tb7-i zUwr~)E;ThjLA>CE>7JhAhJMUo?aB|L*6te%C@(VHAltDzSRX`C<@|PJ9|L9%nEEAG zg!jS{)`tgmSsajq|G4!}OvL_& z4{oRde@z3BQjE(rdl3b#6`0#pU=0Nd%58Eg9}{81{ki-Ltv9Lx6SVD3nK^wzV(tlR zfFnv{QJn!pcRA)j?6HcL%Qtzi_QW60bnpAN@IHPXOOA<&u1+0HM=bj%Gy7XffV@cZ zVWf`hO*E%ZXA?TZ4a=5ujWpaPaqP@E7!W@3Wm4Ps=Wo+56GYS@7R>U2GNBrd8G=f} zAx~8^t30_q`a!)2xdBBOW+U4L4G6h7LbJhayKCRkHJIy+zM+fMcZm>i0F1uPeap4D zUi^~la2a7uLVaTJEs;qcNFSp)MilTacjYmABY|+ccZT{e)!FQm zP|F5{oMum5!ZCMh({CNQOsTt5LcDqM>1c(lB*sm}SdYiD_g_V`DM@b#oL#?{6r?10 zh>yQ{h;T=Geaz0}REbH1R}YjK(zp%~R3RNKDc?XfHG|M8m%HCj6mBP{uNXR;QZ@3M^npsKF7O56<7rj71-2 zOwf-(H>#H=L`r*N9JKhPF6G|zZe&b~q$#R$&_h&;Z{*DL!xDfGqaI0AiV=)RM1yIb z!ls_p>4x~>n20QM7n>L~yrmME!rLZ2bMLOEFlfWzHhmM|t#_jq(Gk;4{-y}CQtZ~O zKJ<_>te7t`)lMFq4=2p#DqR^5iZGG>u02Hqb0z7!^2T@RI(cR36C-jS@);VSHUfUm0PBm zFNJd=WIg84DA*)Zr)Z&U{^$}X-%$wXl?b6jW8|Ns`>xlsbY5KS=QSkSL-;?1on#*% z+{xQ#7O=1l$yXkCr_-8co1j(bcMMUQ3@4rY7;ulsoR@KN^5F#+?r+*Jui{`DUgpl% zi(rAWYc#m0-WmDsZPSX%T%SHi`dtNzZ~dH3Z2GPcqS==CRBY<)h@^q=d@f1p#xbgG8#9t~R~ z!6V#NVr60_ZyPHF;CJqN11C*znTXUd-U~?`oH3m(p|IB0E%$ADo7_ z>$gijC?AuT0?cay?@`vzQT)5|D#zR11hJTrpylqTSB?UH;D#fbxn)CM*aYt^aD-{2{Wc@Ur-|H_lZoZYuYgHmnBjm zi~X0r<1utHOzZTtI(%Y2n1vX-MnQ>!-%i{=jIauL+Tb0I5PKXP8M5`N0^A}^j zf>^IZoOB+E2)Gd^!wD=vtylgmK@lV4CCGg#T8a@PtC#p$MUtpeK&RWf{r7?zgMZs6 zcTn;Ux^4|Wlsoi%U`nED#{1jNR%3YWg4(wp_G+vK30ftYSkY1~Xk3tFhkE5qS(P`} zU{y0-pP|&IVyjV_l=&rXD`p+DJG6)P%_A$x}eSQMf3HHs}d>IY6SbYvDGb{CgVPZA;6y^d(95L7QL#m79pW zk0);;{Yw_=3n>Czv^iqYZ5DJ(fK{e>%h$ZfVBX0+Cj~TQB8929$oz{ z&7fvXqf=I6!icsOa;VfoD17=H6B_L)GbXmSv;rn=wBV6}!O=8$D@EVY9zg-~FFm-B z-gJFwkEVVgo#=Js7*FEsxp!?0gn&n`de+@Llr5p3j?^3Hx1Zsf> zwLqurs>S7df)xQ;`OjLS3~z!J)6Nw=#>L;Rr3TiP+T{_bPo&x?*}2iVmkR9#b2CyJBcO_V-w2Je zl1FxN|6M8!I}{qlF3D2N+=sD0lpBBLGiGVH5w^&Z|9oe+N1-A9C-Hws$p54=m2XWQ zTtz#1Tdj1*tV1@;gy~fm{(#ba&rT{ZRJxeH;>t0__#ly$15>wgX``0sPXj-|m(8vM z@LS{hO{YH~zDoVnZ6ji=n>?!>rf5#nz;P5Z-7fyaLfNRZnB<>DE$zYi=e?KAp~3|I zdJicp7^s99Sf1yB@hQP$k1)!-D)?(a$9$5HO)DFZg+?s_#M9jTl~$1->i#{XtHzt4 zp)g-EeAwS=n#R~1NbQ6X4lnzL=H;DTsu7b(c%eR(gOZ^ET&{cDe9E%f3c8#-qdM2d zi~(*DOY4>RhDJAW{^6d$W<7dLDA!q5yKXJ|QM!@xRwU$unqqSq{OM_N2QMWySQme7 zdZF_os+&)(T2-1Y9Z?fdtyZjhM=?pRDc*$W&HmbY$>HC4E#SJl=rKF0{13B4-LY}cbO(F5d(u@GPaPcIS%0J7K!Cc_)SmTF%p_sojPJy!7zI*E>wxCR%p>c8F#J5%x=R>-U@2k+8w4Wg;A9&EDKgQE0k!kjk{KD-hXjY9RK2QCR8r4 z&@-2WxfWrKElTZ@`kW3ntrS;-W&rxFg7~b67KCK{Nqy&(WSOM`Jm}}M5a`SsXNTLb zoH|)@uvxSyLegJJbYE(6_~lI{h39jAVm4}Oo+TE(wCJ~OOn0wkQ8DE#k41p-M~=(K|l%kU9i$8I71He)$^B-HPn4vpq7@- z3N@$THuNj4i)&y(EBw=Kf)6q#hL6)zjtt|`=RlwAEd2tjNmgUQPM>L8+2-|u0>YB} zTn~l=*r(LJd>8-&`pq%bD*=7a^TCZUKznu;`hft|E8cq}9AhQ{NkEZ-BAZS~Tast| zOISEr=8DZ9nKYGyFL=C3fu`JNeU4$RikB8_iG|pu-*SjCO&t=}!<^{HjitjHjipg~ zjigmd(RKQhmx7+MiG$9B^kH>Fj~a?>;Bf+i?ox8%3$WpE?vjJ@@s4g1gRNfV3iAn% zviE2Qzc4d1i&6j-(Tq+Qut%n=+kcRzA62TveP5;=G{eGft(Bkvg#S+@srV;ym$Aov z|3x)uhKCK58B*hr7VKY(tEMhoR%N~_4zc)wos%9AQ4l+IniRqI{J+4Sf`5U_e^p{G zr7Yt!D?@R!Ag&5ofuS_HqV{yA!rb4&hnxn{%$0nKbW4mGGExm!M8wscDoHZ&UC&w&mOg&eFqbi_2@ zr5~`5IeA+z{e;Kyo}wL}#d51|(HNxId4Inxyw3A&MsY||OLTAXU})cMiz#KQh%4q? znY4rqI=w9mNj7|J3KbOA`0V&kUb1hLQ*YE2Du71gS4VjfWNGihVc~= z2t6v?Sy{4nqjz^A6#`;Y&On*Ii23>jWJ8802FT$|zwQ(Gy?h0+88s}ZRnVmc^JO}Qz_ zlY^|(yd`3;IQbeS^7yq&srN3z!^&Q+`(%KR+wB#h{S78+^9E|b%O1}ueA^mK$hqgg z?Ty}w2*n)ci5qFMcq_o7%PQL0csPpeeLXcQ=1SP`BI$YHD8QoRnw?6Q z)>^v(41)d-p{k0;&Q6W=4NQ;30qt%skA2C;6{S_}&KCs!il>x!#8g{(vi_cSR- z`cx%V34y9-ddy&>EA#4Enk;@pb=RBv$wIE9nJg64C)SXmkQBkjzOu(@B9qQg^{`2w ztkj`?yE{!CyC>R%uf5neMj{iEjW=afW7(#ihw3fE+x#=}k-vFuBqx~QJ%of{_#P)~ zd5#yYH3t3x#m-y&wejIUtY*L(P2EJ~6b9Ttlv;YIz?61!6Z6NrloYE~?b491x|n@N zBGQeT2J@e_ztyhy&F)!eg${uVccn*xJj6YD^`U8N{1zsc=sE><0nm9t5=x&ej zc}tBt`V@Csmi0tPRU3G_k~`=aCL&e$)v)r1m1UId45v3yBL{*0)FX2b#iySJ>J>lJ zOw23#vmag;th4U=z@Uk7U( zVx%W-U0FPu-BKd*8(qq9aB;j%sz$ACw~}H?OQnh3drDYMS3e873Vq7(pJFgqv(877 zpa{^nP(N8<*_UycH=~m*GcIGy=053x;BI+Lz_%^zV{28*O7T3QH z$g|=+av7OQ8-b5X>zN8JUwgAh^Ra#BdO~1xur@icBNQ&2x?+~1^ytC9-6zo=sVpR) zm@m$xwBqmz2J|jKNk;!TtffBLL6iRjbaNqeDcrKIk%0d`HTX(&P_$*0uS2>YG^aGq zWU+W% zhQM!GkR|uaT@IWeJua3cSa4072}a=Ve3e$|P*PP!_>vX+$R=~(?x-67VbU&J9yV=b z&~84;siu%4yG%X0=VxBwL_<3u$3kmuNid)kN2@XDI6sN6=J&Bx+nKynMKq@!z6?jx zc;vB6p;l!&c%y|*YdtUMI*)tR0aYET5|7rj-=R@6Y2n17Y85BxXIUYO5!Z_glq>r0 zbsN-?S8v+FrGNvML}_8Te9LBO@=!;T7w$B2koCu;taPqP-~0?1^8S)d2!C@;eb=|- zeF6+qEiL91v&i=hsOWLNC-eJ=gsvbdp1=x4W@xvy2UV7GobVX#=A-2?7h%Uh* zJL!i7P)1(!zQkXHRK8U_?u^SqpEo^318YKnPJFfnoZC>q2UTy9d#4}Vd|vb?X4Q9A zM}%eDwf%BO)ts~k0PAN#w2QZa1Q@qE8^<;*Ce`H?FCoD&;Kffd1bVnoe<7%N%-v=af*=*H z1vTLM&IX;5YGOmd4S&j3zr$7#Y@h^|Zcy|5aGO;M=vKaN)=&vvYiArFtEee{wi+S< z$~JrWaUxU%l%Bk-%g?++tB&nkr~{W?j+NP+lTX`p`1(*_RJ)a3YCd+r&jF?Vj`e81 z#n0Zg&BOgrN!n9!rL(Skj<*m$u(Y66mr~pF3YOq2(k7IxAE)axHy28+{^~BFpEglH zk9XCxo7(DlMTYeWSe{REuLaLrxxmhSyk-&(z(W)Vl**0qtG7wp-aOPTArNqkZ7}(H z(%P-RjyS>{MgN$*QhW-IU`HpNFcJuRUxbsmj0=!P<~79yUY%SUH78Zg^cokK4w(st zB*?!Uu1w5kxWy2!1`7YC8nt&mF6?{l>FFb)4;fJ)Cg4Cf_?C=pFO}$az@kr&o9DK8 zC3ZErFR;!)D}(S9&_OG@ZLi1*yfUMmXk&Z-io z1Nf&JrSt(nJ@(D6@VbIM0Za4s=ztSJW7{>aSdL+KBf|mWT@Hs1xlKw1x;f@hT#*0v zutX!+94Sr45n`G^Y8i8OVM!x0kYkFvB}}=hGZ#r`s5=#Y*ixs>efOD1^z0?8i{gwi{ER5#6FYQ)M=FfJDH!M@dM*z3ZjwQbCAhzzp4n02QkAh9%` zj}HH>h2~DW*(vb<7bJ>=#^YY%y>~iiU(r-v(S{8R-sRh|R5a24(yaV>l~i2JB0}Y4 zM=df5S9P^_p*BUTw-mc09yW<{O{Rgv#Xhkx8dL;Os||-5umM8iXUy;c@mnfdYtx=u z9?`2o!u=X-sC;y!TD*yQas9skN*xjWmp+lp^Y;m{PtTPO{8M%)LvbB{<#^ncf*JF; zgajXgC(NHS#?i0%h24Ekar~waB%ns6paN`W{L*HkSs}~?ND(JZI-^4V$q-pm z%znR4C16ISV93GbP;8roLlANPV$9j~zN@HuC2yLpvJgMEV_~q8VVoL~^rp37{r=CS z6w*%C?=TF4=lpAS0ACy=R6Jzij8S%n1jIiF*?514V!y!jPtwWooTsCL!6Rdkv$Nv- zdV2gDF6qd#D&L3*1^6ciVQ;uUVY#(}xh?slWt1)5GKy<)6gZa8c=lAbG%EI9;Z)h{ zUFnZsztwG0m0EqSUE{uT50k!Bt3I;Q%Vf*$C|_NirBSVZSNhc|$Ym_DXPh5S+#+Gf z89zsaI5Lq&vMna)79y;w&KQn+u^+;a0{$bxW>XvC8MytoPea zNBg^9JC{zvEW>n>UytTRGvvil?uY5bbIp_Xn%BC12mH_;UXWi=`Z5at_(T&H1=>=% ziXezEOL%=NeX(lJJY-g%EE5v;m1zT`FWV{J0{U&HgddoQ&c9 zF=c{=ie{@o(^FPH%gGeJ$a*vHErisWlyD0=&yrJ_4Y4UK4K_0?q)4EQxfVyMpy3%S z{oE1i8m&p5N>&|NwK`sq5F!$8uYRK@(gLl8T?$`tz1SGASBupm6R(27&g1?CWZ5)F zdwFu!>*$4Ak{6Gk4O#deRlDj^xoqt*M?dLBo2L6hoS6veO1)MbTe2-2HS9#c<7NND zsByBaN#bYU(vZd*gJ|5Du{@|SqAaT@7LIL6N^SJ zgIVL}6ZRi^7uD}xvlzDPhl8N*`om0WCoRS)2lNH&k|sTa^HjB#IzCepT+DfDa-m3Q z)YEwbZ~+E9`CV4%vgIicYO`B$MylJ}^Gx_U{dh-M{F7r#)cG&TiNlpzwU^ek_g(mp zWFa805jje|rMU&^S`@F+n2@X_qKWtOO1)BLasHFQcD0!oMK2&DdDBI;Av2dO*S?O* zAYfKYR9eO@#}KTecF8`BAxRb(AJk8xt-XteEA$q~sM5EjP%$cz%xWp-0qFnLulv2y zf$O3)5o^WR8+ch!PaN$Vb2opUPhhT&sej_Qd4}mxxk2AFBJO>gc0|)O64aVVe9X8> zPZ-}|U9PjP-wS5AG4Iy4OO94JW=qttOGb31xn{?@*q|pA<}fQgfU!Db@y#iOKMQE} zQqrz0J_=~HYSNtFMNH8;tD)*#X~6L5+BLYB64evHoZQ|1PM~{1uRPRE;TjBJGHV3S0n+85ira*!Tgj^Rt zwWv$)XO`TmU(#yn$YQD*e^%GEab3Jg6VJX_D7a`Xhd{d=5P=kKdwg6v&Jw{6u{aT% zxrgVv*=a330JA6H-^YY68g`6n%%!9?pko`W3b4Xp3yKx&T~?VB1V-?Y<++3eLv`93 z?T|nz&NPtZoxl}!5s?lK&LR@9gY{yngN98WL4*C`y~L|#ZedEaGj z&}O<#4?D#A#)}cKPkl4HnFansfcO=T@3H}|UGdUEm?4`h2FBKfMyf-+kIZ9GzMJ&w zS+*HjFXxln3J+Mce;m0>1AUHKONbwu9?)kzNk?zFMrCIXM?VtKP+<$hv@~1lykcX0 z`Y-JiM89E;&5XA>RCiPqlvO-XPt3sR{q*M0ql|td52203jQ`cFik5&gSJ)Q1==`SO8jj9=kM$`e*u05y>y((<02GAD*wn)9IREBh^;1s&_ zC5Zu*nn#FjEGw9F>&r-Sz&e)I-pJmxoXLm@`>)zaEB-!HnDlNFO{+cm8IzCorl$ES ztGOmMn$h}4I&xM1$6~aEq3SW~^wH|Dns;H-Wp`w%In}#vt>s{lyG&uo9C*I)lFFmc zpitK-x|Jek6H{0X3TM?=qBo9CSGfBu$B!}mE(uAYGwi1r9(3fV2_pO*B|uCnx5LVv zT#Z-uXwoiIo2wzlIzJMkw2?VTrK4qI^J8_kXm#uKp2i}>h%LWzdSZ++3X@gQ*nnU_ zq?L}?lOx8{rEC6ZF{uaZHUZMmQ4h}vemgA|5$pPQbwjnQiWWN%q$P5aG?yA#HnvK{ zWWf^Hw-@6!O&4nU#wIZu-?!YJY1c-*m#Zzitc@nCriGd}&LUxj5ZYQ>J^)z3A={HQ zu-2$0W5OI-7DLH1VU@7@77CP+)N(BqG)N+&Z?}jdGs#oZlr)&lf(acXu3q_YCjXyv zu8E`p{ipgw*TNjxS`z!J=suhtkC4G;V(F_v7h7&Kx0yGHFZ>wYX(pI)LbAqxLsj=p z0Tok$7D(od$DxB;w$IuzGxeA?~89X!!CXBX@(u&XVh6q7)5K0)WD#@|AdC5e}eEoLHa{bi3<#h z`%k!w#vgDiaj1?grqOtHnxTt;`{TX*VkU<&*Qy;OF=U)8&)*YyIPY#--ld?8cqMV7SZ%Tp`Zm<*OC zuFVXZGWHNA<<(fBN0d#Qu7bKCfM2u50EW3*T$DgGBa6#N5wY!Qz4LH+fxBH>)hXsJ z`jm0u%6=G`$(2&7WIZor3F}s-<@n^C=MK{UbSLpi$2j}}fGxS@*K}!a437FVO4)^A z1hr(j0^E_Dk%XyfCe?%=p*Ztnt7(21os~U}Iqk86sqwd}7{2~n%nP-&8e8sdb~d<| zE2SA`6*42x&^>rIUX;liX+f0EQ&Vr_y>->WG!6n4_*jHm&HDTo8h7`X${cEGG*%mD zbu@NN?7RVOX|mPdDn9nqCHy4q9?{5AydkNlnKV6|$p*HDA+yu%l`Y~@mFDYrY-?7E z5sL#FHbaHky-TO~A0WD>SG83Fq&s_0jENtgdPQ!tU(W2>LsT}s;-~IS^a;RnP}kEM-2Gr~*{9a0a1&6|?GQw@ipjhJDC7iGum z=Hou0-EzZm*b@0H$z`#D&Z|4a+ku3?L2w-PZc?F|ci*9+d~*5`qXP0W=+5Hu*urt~ z*#6|@N1h^(Ly>Y0_m?Kot;OZlL}S_Az=JL9(C~K4RFd(^i|00qsPO2}%}qB_y5-qx zYOO9C@SN2?pg*Snpc9@C=yCi%sC{>+D`KU^>a&r>g$r6DpAo-Er;;Fy2&DWtOij?o zIYSr`M1M156xkp#t?(=PfZ}dVA*aF3UpG?udID(Qc27Phe=Fze_A8?LMh9LST0{t` ziFNu9vLX@rzazQWO4?^c&K;ls&X@tw(CAnXFyR<$9_j9#fd!R3p8-6%#ah)0L75to zBzf{X4rlVMXc9$q7>zfbyK)|1LNseCSgh=2?`ga{>oe)uTw1}wLHw8@Ju}cEfMByp zIuXkndLoO@BK5_`+?kMQy9S5D^EQn`G27Lfl)Tz@vrbVmExTrbd1&#-jUskuU=Gj0 zPw{58mYRA^ov0Qoo1+gWam0Ny*P4o%1m|vfQqB^9cJqk)JJ+!2O#6yr2dLmmixW;6 z6OS!3Po@yB{JCYL=qnzbvG&04=N2nz?ZQ)U?cfP-?ZuD&OFZY+kHMDy8|=;*pW@*C zmJ(N8wZTOy3P#L^Jvgo91^bAgXVR=eT4nADgQU=;gdm%Jq}fJs{I+|nvJVhSg>RM{ zA?z$RmqDx~NA4b5EYq%ODij+xyxB;tL}=Yv&vb4T-r@=Igc#MQVSFzaR0BatrVTMG z;s*o%R7bw|DZ3xCj<)dl$%HropsdZqZY(US-IV|(O#$wyGui)C(BEBGR?NeG;*kI_ z|ETU@_0v{J=Nx}3fa$5@5nZsRZt*L&XDo+1#&IzoF6^>dd1}F{klsA`R3O$fdx%!x zEQ`s=hFFkaYdP;sh3mdhur5h_{3%=P)9fimHk%+O1RGMe#ylHBwpN+;I7GW&@d#n| z8o!^-nrSS;C~};xI@S+$kkSgicIDwgl;61F)MIN@qvEB3B&009Cymp=6z9Lw0N+@P zzrywM;8+(oP`r&_eJ%evqhWs6XRH2Qu$~jDLNFV8Vf3TKK(}CVMB8+yrZod|vMmr9 z-fuAnjkBZdH%c1=Xn|G-XhIJXz#A>|d*j8}%S)7h!tRsz8>NhKE{yO17^4w+(#Duy zU)u}xi>}vwHHu+}WbPH%bAqdBsrVYhhcbAlmJAF;xcVDP6v+EECKZ~wQ8Q+)PSWqB zTHvngb&+q7hzX|(ZsTMuya8of^yWJj3#?)9D;E5Av_oBR&D5@ipM`Q4tA_YY&Ayci z_Q$r)9w7rivR2ppglf2L*4~v`r~KpDQg*gNcXxq6Pivt8%we)8UN;@lB?xafBN0*^D?NocTdJAXokL*CJXVfOT4Q67bhT8l2;&g57bbR-_X>gUqCkf!|p+ zEJaFhppF@VY+m--_0ys!Ci5qQY>0{rt(7w5-FG4ncQrEh>rHVi#n^W=k4$s5{I~O6yDcYh zLS^>n?s51*p2u9U8&husJc)N2Qfqq`6p^a?oN<}w5!Iaggmf!apAJHlnoK0p`Dh$wZUXObaI zBB;TWIltE+7=##asTgz31#B;tIQv)&prB?#DyW8`#iBJbzTpD z;9Z4w+n`h-zTPl(8}|d6R`?IHpR)}q{SUg60!m(t(T_H{W!H=?Pf4@a2y#9%@1lO3 zJ3mPjdJRH@%kUP?P(IsZd{9|1@xqg`sC9|Uw=#;iyA$R9VF#13M5FbDxQW8~oWwb! z^LIa@V{Q%`H&DJoohGALjJ-;)i5vq@T%h9jsu|(VbRwJqcBOr_j)>x8{{gc)8K_*# z&#pNq>W58V^XcApkvTuNzl0%!wBH7iHSiw)Oxwy7XhUav_w5vYkZv6!ogajY=&}#sUBDEO10IojQrZTmnNs9ARSk%TUi+`+gu@ipla%$bS=?{E| zN$YIY=r1PywsaF7^rfXX^Z^z74Q=LU4oz=C4e&miedlLJYx0VF(4DE5t+Wz6CLmk( zdh-D~_}xhC3*1cR2W#|Bpv&NFto~vf%Nj_c082>phb-uNMqorVH&WYSHX+Q%Ln;J$1 zdWdqM=2=2?x=b@XKjFy!{<|BE&TxvN>4i{##&593o9MP+PnZYwbe>2Q9=0$Vv}h9j zY)vt2iVR~sOVC@kJ`#5rPry3XadQ4fh0c%4uqCH` zpiQVIOOK$IVrc2yIqCo#YO@xZ?Gj{pN~uSLI>6u~vt`3;iMPqooN0p8R?I2&J_SNc zJVk5K>$r?)kmo@*)(myg*;%sNwANM(0FH5**w7m_Dhblv|BQ&{5j3`T;6JYpb3-+D zm2WmVbAzn4<6vwMLDuHNJRPxZ#(8{-&yCOD7fe;%ar&U`5h*HG=uSVAnR?WKI( z&2|ut`E#$Tc7^gbA##~REYa)(z5B{d0vN-SsemeOz`f(YL=k&8;`G%GL%j(G&Z1x| zZ?=OYiTv(P4C)S< z7`}Y$7k#YzDrcyCj!Ql$Y9FbECFz$MdEiG_+D8Ym57_2X^E%UC*YL6wWGZ|)0`lAo zS$|E(CD0QxbcCLl$UM<@PuOvt0* zQZ@~R(Qk=H&!U*ioOF1!L4VHw>b4ygAQbpadFrbnJ)&dUOnNmb`??H=W4s;#ph z*nZP8uCccous*(z#?~?=Fjx;r)hoChb!s&>0W?Chc!H$MC!+!hnBv2+q#~E2QXzNL z7-!r}o&vo`BTP7^enmc50ZrgISj~QZka1mqJfbTRh~m7!NC2ORvChIhIWH#McV4joUEf_J(@11^`G z{@`w2tjEYfm7$w7hvEjxOsT6{_>cSKI?5KW42V&{EeTaY3qZ;}zkIwoX6wX&IHeM( zH$T3K%hpQJEW{VW|I(;e^oZbm$8KCQll+m~sFGr1(*@$CRWEuL-ACx-BXrQH#3z zIZiKA^FW;=8L|g$-<%T~ww3KsJhRAhs&Cmy5&U%tvxcDx(dxR_Ws}(|VvZtuMB;+If z3e#D>-MB*6X-UHQFc!XUoNprYVxoAs)bqSQIc~k**c!cl?3TDrx1d2gjqI=+cIwc) zk}xumeN&y_Qz{Ig`@>x-1ems!#8ZroD=Roum>>!#Ymvr~w5Kr>Lo-f5IqeJn+f*>6 zDH49Po9@h(Hg`hDMD?x)BJAVpCdHV;^ljPgRwvcTRav}sf?>z04cU?_TUbM#J7y&}JA#Zx zfSYT24-cp9_x)0jy?MDz&BU9)k-x}9BY_u0ZO*^iR+P2`-O|5Q{&_?lNL$}odbpvW zJ8sI#j^QDifVf&Cj>!Va+v6*o^k4rS?A;PB2~tk6KFDKc^H4Z9W@v0%YHP2>Ym{U$w4v#a>1=A&Y2jaOS(J=|_qn9&sA ztERo`zyXu1Cf4c59VXny)n5d?bgS ze#1gVf`vJ|i$w{7^~DNhz^7t-@Brh=$0%UDmR9g|`(9n( z+3I1%>3!@(`k5B5`%$5b(|!5v&br9C@#?SK@I&V`2^EEi#F>^s67&V4yS(o!>*7}K zA)8?0puAhwzb_U*lURZscHDEHUN=IXjhpHWA8J^TJGy=0RxU#0@2htswnU2ZWS?S8 z{*Mo&t>jM^?0tTw?L^^NeDhcz+T?vY3w@c$pl9zhCSULyGN;lN?;Bu#a1`x3wwRof zOZ4UGP%9|i^1y@~Six!IScJB%9Usu($>gp?pE5SpJR16UPi42kwpiyxEkKb4oEM{g zjO@|SaZqGd-7139mgaH)f5-^fR_*LRt{ z#yIhWRye!DyN`qX8jDEAlHJd$Hs$$q>j=DgNx%9#OEs3{#nFoDp0Lu*dyH-mfmvyX%eD^5$XD z+RSf}c1+*oeFMjW@4iKW?Y_k&nyD37FWTh!yBP|^xbQAR`U|>94w8BpqN8^P+j&?h zPo*JZ`AH;*?bg^Sl@{emmFsTAbs9pu67%c-qv@ODBWb>`<7~Ve8=GuwOgtOg_Qtkt zY;0p=+qP}nwzco<^ZUO4^wg*CsarEWJw4sG&N=0^UG(RxXQR2tUs>>nHT7?}FkD}( z=yL+B;j~$1_)R^&XshTwc}ikW6cV;AS!1N0>&2TQwc*zTk{9Ex&9v*r`7Jz5D;DgB znhwI!!euY(wMY)1jkPuV@L?+=3!GeFUW(0P0u~>qH7xcfr;{u!=o>j6)VMWJB3m1S zNsu-0q%4XQHI}69rdM$_lHHrx>WvM89E#YaY^PVXG(e|Xt8w7}b>DuRP7D8-NL-9u z(E6vtfC%RBh>CuyJYL7p5(Q{9d%qf-MG)Uzf4;Q|m!lfWs{X+v(#XPl9?va6uf z2?uD$fK=PY9x3{9#8xBYiES7bkI8X`wo5kFmEu{t4I29EKmjCId%vT6P1nUh^X|H% zYg)Qvoxb?U7E3A~JyyspsqK}_EgtLF315?`9d(J3j$jJB*ku`40Y(^}mF~F{^em?h z?CW!O9vac+sq6i{2Mu`H=T2V=EZ5eJ-a6yI5Lvt-CMD;OxLKC0n%vuPJxRPA0bbkh zF+a@3CMlKvCl~!LFr+QJfX~qtDd6MFJkGCJMgWmKw0ld$^(td{A6iHKir%^Zl|I>D zZo2q^oT%}a7gm`)CHJL^JCAp5Z+@d$&j;*8JS~F^*_TtIJ(NxlR|G-26Ryk32WI1p zkcL?XCk^hDHXvl0;0N@?M4{U>-GoIigQ+#fL<5kAP+iD@9>DCRL6bHn1Nd`j8_Vxd zX9i+S9kM~`|&dc)dJr)~(`vCLPe^*GAn+w@BF#k>Wb-({~2?DQMl{Rl97u4Hq1o{t5MHyIW0 z$IL!N&G!t&h~PUhm$(V*xBQdFH{X{UMxQcL*>@=d($C*j@l=`9Hp)<0vpqwSH8GGB zXfA$KK(Yb0OlE#_5<|-{{df>v=mq!Xr_aX;%B8;SWg6cfswL+SJ7*G#A2>eDCAGfhmy`wUVcNCVFN8#~{S2+MTL_^$YeXhU zIo;;qn4DGk!%F95yG|ZuA?|go;lMj7B2UyvyTWBCUyI?-B^VVAv*hLSr82)P3HyW?$baZF}WZ6`$${{GEMhU<^uhDr;q#N`vyuwkcP4?JVu87 zw!~-&@vMRE{PK52goDpcFNxL{cMIif=XxPGhicUF{XDFw0fD*Yr`Y}J!qk)^hOq@!-|kJ*!e!#f6~5R(%%w2&D~ zTfs0*5EmxJX>SFt)o3~Dn_sB(Cv-q&V)sKawTShnr8tFR_p|z^6nvhg*+!0ttUGDy z7_@@-Ykx)S-_JWs>MuWOB`(D2pU3XpJAfVDM{f@{7ct_{~K7k0W*)N5L7sw zLq%@`9b_0AUQ8*fnkDt;3(i+oZzBUxy-F4p>>&&K4RLkjOqS?uSk45;DXB;GdV#)S z(X6y#yX?kuVX@*U9Z8K|pmOO}0<$3oT*EghHQ|i4K+AX0b=v&b5)qz^20b-{lTce? zuE9JFj75)zF;~^xyjzdPo_d+|R?dlPgBIDg-q_e`4lEpYIkZJ;Xt#_a6pV$miIwtq zS5tt+%lrpHqv6GL2oQ16zo8vnzo_JEK*L~NBkkT9-lPVg!A+Cin&9t{8LT=in^%t-V^Xrm zIbg$panW`VB=@z845r1pSvG5#h}aY0g)#P1*Ey_NV~zGx?_<(*AL^V*F)|=%YBxB4 ztB&$h*G)=xJO>mQ=q}ohn;$H@z>*szh1*JPOe5D_g7q>nB8=DGMs5_^AdaU+uFL`e zZ{|sn7M2GKv|0wbn5A}*b!-dRDQx(48mx~L>B9wMfe zpt5AeJpKu-V9NI^5F1QJkc6jO2s??-^S&*vi7uC%=T~PK^qGous18aJ%CIw=E-J1G z8&r|t!b18!s!ez1k5IiCCNS9J8bq_fm8y$@sx%kwnyAbCdM&{ zx&mAL=fQxaNGJUI`Y6ahkoG9kQU6sIua+%bhxq6b_RU{hIT(prv=I08#wKoqAu)$} z`S@Y}6(;cwt9FPXTWDi6!-moBV#fwWvg+sa$?;=6en%nhL$o|T59{=a!q{^XHzmZ0 zM%@e>8IVV7U`FC{(#ClBru_$L9?|KOhOt*CRz(;ko$>@#MH58n*k5qmfV4)$O*X>> zjeORqlkBY~R*P{V(sV1}25tJ5wQ;eu#Y+gWlgh*ALaU1q_}^xHKZz%GLX*=I(;#YL z)l%f44oXMj4!9)EAjH^%w7TvFa&Jiq=;w`!$8;H%$2OD+0oq2@$rK(99iF?dh$mwM zCO|WF;g|lPDMaLezCncXzQS%qux(t$!NxdwBkG0>jEA$!aO{$lyr`(qBGkSEg&KC|@L?7*$ zayaXr<|;mrr~jm$M9) zmgc?2zZo;OE0gsaivrp?(@UHlhF3gf7|}Gng)ZH2SQdEr-Xd=LH+j0@IgYKRNg!=L zrZp9OyFhsj*97R-ZCN9HU!gkedjzY(sfM2;o^v$W`B)m^h#Hos^YDn8hyUvqqV z2Ghgwh1OkxXypF3Y7;gMW?2RPRz0x0z>G&g)U#C)3uOb0b8@^23nFoK~;Z^Ai;*L#s}wKgvT1R=Jz7$1!p7Gp)1aF&`ZS>Uz*x`o!IZxTekMPF4-Hz_Z)w(ARPn z$+xs`C2xb=XE=2~uAqVvqLfHTmFbIS_F@JHK@x6Sr&oJL2s;_q-^t1MIY@9T4$gMZ zYlJZ!_RJ05+}aY(So?<6xaG!ur`eh+OF~A~t70k^ek*)(4l`+1tuevlh@euncKAeI z;W8oz9^6q#&{0+&&UaQJ({3DF0LJ8o_)K0QI?{h3qZ3dFm5t2&UkGzjrpUGuVZpa9 z)OM3p;|Xu_Yc+{3-58kc0FfOh#njU0=K;k>gm7Gxo*^R*5*2TL;SbKu zmP6-4Uw|l}#us)J{#<1g0FGG^td4%Ca?0sr{wuQ@6tZ-c}X*#KwgY zQ#FyWv4Wb{u*TsU{1Cu_B%|4P0xHTa?OWNda>)FowSK~$jcG)Jx&vPXfLMTF)!39?E(&tHSpZTWMzq4Z>-W=!b5lW~4rBykT)5&Pk82^| zamaC`tF5;p>qh~M0xMCcu zrMX%)*}k%>A`ydn&4_W`w+7}!a}!7{0hOS=kY%_$6UzS^)^2vkG0kyaRp=k?#%xx; zvLzcRI@{x*4eMmpmTwTK@cu6!qYT%&b$rySIm-XW)sbb!Cq2<^H4bWR)ShJ;JjEa* za3&*&_YxOvCw^4;j6rWA$+XBAF{9P$XK%{6sw!%%Qc7i2Erk+ejFM*5S%=MhE`jH| zN%Ox{YS)lH5VwGsL#rV)75vNHaDx1CWLkZ)upIf zH8@C@Da8T0LanUsBHA`qWh~H;@pL6;9MnqmSR5g^0~aZM>$Yi8CP7ovmR7tF9BZnR zK&|P&H%#HJAJsR0s@n@f?an5fvT29B*Q=`SK&_;|H#cFYZ`YImVq)!FnY#^_DUemX z17}&guhCJe^V1=ENbL|+#=8p_m{n})xULs8Iz;rbf1h22e|VB!zdnVtGpxS$w>g)T z)ZK=CmDk!LL8bMi#grKM(8JU*o=+O)Cwl3Amx}etcQ3c!;wycmS7SRzp(JEcljO zzPb3j#~e?sB61&XXoKoU-7Xtem%5V*&49WK1ByG5E?2;2c9XKbz4(PJ&kg^BGNjJ1W|YyXbV%!KezlGnm*TKd?J zClYtSG?JsHk&=buQiOT-8iZrA{o}~=!RH$QZ$a}>ncERXo4m`3DAyZugRe=s(K+QHN+y7{`r@NCwB!H;PW5$H(^&#TyyuEO;z;K01fd1GxoKjehqS zEro~!q-N+p@m<_+`YCIe3EQu^60>~YV}>Q>oYT?(G0QGw=0mMpV*B4k*8=+)s9zDu zAVSO6O^Qx{kfC0p`c(AX+%^M`Az_yF1|WKZQpkYUJVWUkqJV!)IjupF8BRcq7lmQT z7nC<7F@XD3^~F_0XN6O;pXhdM;>a1Rw9a^f6$6o%<{3F;bT=uD z=AzkZZMVVJjaq4?OqGx0eqIR|9aaj5$mjd-w4n^}lS{~bt5&MYzB1_qVW<+d3HKRy zI?`z4Z)yBXZYfpDf0bqo>RL_#6UtcQJV~iX)ZOdB z&gjZrMchV@}~3Q_YOn$e6a2Y8DoQLyC1yygJ-eqst5dzg*5I`U@!jKP?? za`JN-L_do3xbFGgi%LEttW*>QbVwMn37C_7&6ve{HQ#ycm?ZCV_Qwe@Dm?iljdlXD z|3FGd&|;A=W-gEAYxO4J3I=?oY#M^}qAknym>Xllc3LrWw)R|^adq^I{uzZab*z>X z#1wcli{$pX0J>o_04L~SO5MSlU&dg#Q{TP`xsW{CeTq~y{OEV zKTr3^n{KDN`Y-H7i_-?eYgbl643B->7__i8F@j+Dm@Olyuq7j?)39T7ZUz-q1h>jR z4#r*Q1|+?!V9Y7iV5tL*WlQzi;I%{0>IKz6^3iA05~~U&Ou2wW_+oX-EO+TUbU6>L z;zX{KJb}o_@8%Tqy1yaJF*UPw<{+Tw39-yQ<|l0eS)bDm{SbS}`HbrlPS^^he9z_i z8>i~u0mK17v+u*(vJi%t+;KW{u~5T=ScV?`LpD&%%Mo5sIQ7Cw{kWH;Dj+={X%|cW zZ#bu_6?%H)^xfUB<6KGVRhI(Na>b#4l!`Mq6B{QH)CXUwQ^USz=~T@MyC)9MkiD77 zOH*gmQBD^JSIGN zKapalWbSm} z#=1V(8+w)giFthhxusC(-e2EF!f%HpJn)im7cYg1?P`f^*G=)M?#xUBNlx z%Ruy;1aWss?M>dB9C2O99^NMgEB)Z{0j&u=TbXqHu>9P;5d1Uhyh%*hdCejC@+y09k{f$cd zWcU|L(y2Oo$_TM^)NVSkQcNDxo`Ve3j`?P|PjM8LW=}u7?8BTiYjrg;JXp|>7!k5u z|Buc+l|jHZp&nE7WLO@t+!-GmeQZvLgCeKRFA@{`XOYX?yn@)^?}*xIsm_#-_P#+* z z+l`jT=kLRQB4VH7jnX7EIw32cnRWUgwNzHYm=or&p)xJx9jy6}6Xw*?x{l#ooXF@h z+Ng!$*l1o=5~-_Ic)b@@w%~f9BeMWRT?AzUX04LWHZjDV!$I({ z=@}5&XNmu-(Zi3ZThFv}yLtxyQU;2UU0ByP1vW%deHqVgn-MQ>*+P?e8qaQrF<;+y z*%p~Q)4sWqAfS(yL#t$jFyw94^WCWG!pkgr!pn?ue7kxFA2c6jD{T)G;%3{7o)ZEI zv()Eld9ssz@KAmJlT;OzKk+bZTbIcH?mLKGe8?OIVDlMOXBxG_$_*&|N&r?r1cKj5Y^ONf{GYP|hx^m+xQSKWp9PcC>J|^uc9ebdZ#s>o*1F zYnuSNPDt-(n|+7$?SRV03@0#|{t00}6If!WNh0SQ{DK%rgb1Ty1r95XfQ@+m1F~WM zteMx47)v-Aw}}8!8yPV$9RU>IhLMKE`Y$fGCJAFdV^Cp8+^)trZT-e(x>}{HON65~ zTJB{?jK{Vh1Jf%G4Osz@4JdSjhk1!Ci{UH zsPX}WoJc38iHgHB&u^w}=v-zJIcSX6t}r5FY|uADYD`BN{v4CC18uGyzYA-oWq)p= zH=QCjeR^V*@l$M-T6B#Sb9EtTx7_mwRl57(rXYlJ;E=4H9$#7v*!V~-z7~wLw+E}j z2jdP+XKMy8wkt5s>jozoSrIj_vijy)N~~~M(Tu*?D)lH{&s&N*%{qcw0d5+JMQo3y z39{lEcmRojEBGwj;kirCCGj~IXj!B~GiBpNwX9nzItvonok{;9=eD}|LT-etw6VE1 z`FIgTF=@R@Q1<>%26y=+;U*3U2j!8>P&Fx_W0u3j30nXVLj|L)ZPS^ZUHAAZOj zFULQOLw}r6#}>Wm^LvTeQU0rZmz1mMh)`!T4j|WT4-idl9{wx_gqC4#9b^5;FB_=^|z`(EpgRiO@{m*dfY@q7r*AKArz;SBCd5lwtKzoRazz3ja+w4GMLf| z{s-V$3PHIZWqOpA+Guy9`f`?I_=VLkoaIKCwSsj^=!O;7X}X+=pcTfP9?QIxjZ`#k zW{cf6L`w;wIwrVTv#!Q(SZN(}-E6r+cw^fyOlA}Gx zHTMvF9rl004eOc?D_6+1Mcuafo{Me(p3^$KWtYe2u9K$CyQeg(g&a4@J7ypun@5-f zFUtPzfpLKjr0q}4=1i|`9db8!r?yt4rLd7(^piieXUSt6O&jjYwVJ*sooD!;#*> zeVmtYrnf7GmN)Up6mL|ot@Ptx_?z&Hx2Q6lfCzR1z`KY z9`|$;U7LgupPzpX^y|{uk+o+(8Yr6Q?yrP?;N#_f*DwkBg>d>1>Rtf7 zMs!-kq+F4^32ska412ngQ^du6uOwHcIzNYM7F|3fe7`T%VPbMI1`*&cA`h}L>o8=a zAnUKYwvR1jF+cm^WL>%aO@xKO0i9mGU8b(rd2Mb(&%@n!yXLE>@?f?rwSgSZS{T2b z%$@@dT2<1SaCvYrVy0@Dk9FMn<1Yt0XRN$hKHxxc%o*kv)w0xAU;7?^ZuBlo%%;fN zE{aW4EXJ;zZz~ot3U^JkT|f2t)z$K|%XboxS_>?-siBP~@-5B&xOdtR8?F2q0*G7yc`x9t-a9$sng|E9Y~4kWB|ahrl7mRlI2*Z5B5dqPm2N-t@^!p_aJISzBf|;;s#uECCRd_+Kr7Z zGli6?EqC&jpZs=e6Sma;gw~%BI`2!e-2z9*Tpp)e&_#D3Eo+J{cdJPq(bdeKp-$i# zJ67y7t49_+Gi%nF&Z|sg;$3snw+WHqfl`ScIBh}ZyGes!{E0D@+fdvqSG)nMBz)I| z{<~M1qU7@=l-V{2aa~I&2-{0lnfqr^DvyrNY_>`TpWuc=|H`3xe3>)OQiA@{n(&e+ z2en6o{;7Y`&jEKWQ-nI~dbZR?{-*j0i-GKlmU_AceG^M+tCiz{d1yLz0UU=MIz)r% zMZ;B@)&eVS1{kY}0;>{i?!w-rNF&QL%&qANKaF^hTHoOO!WPtYkk@n{9L)V!Jf|_w z!GmlN$y+=@Z}`(Z$X8KQT;XZ3B{h31A4Rt;VT@(6gC9^w?yz`~ROhsVoxzq|r@;lZ zlY*whdPs54KKM})TFm?R<7B`F%cM$=q|W%lmdN6#9M}x;o#wm^&Fx%5MwyzBP4?7MoMI}B<)B48S^LQ_|H=L(%2*lA|uP=lc21^*!Hi~ z)+bc2f8N9}@Nn}G;A+^vA#pjyt2M|9gl2pWaH;2;L!^{36h!^{dT)A}!yvZ% z{_AY0Ml5&CFemHVw@BR;*71{Wf-L);G@g+q-j-{E3~2We zD^4758Y`Y~hLA0~+?3w~;DxZjiQ!90Nk zA7GKReKm%9m>vQAi7@sOLhfhllv$Sgm;BY%=nWpxX9zyx8*`umOJFkBY5!VVF%6vf z4ArIJ4IK2OV@>QGEU!d|IJne)XyGk5j@;jyyB2fEG-NYNlcw*PWkY|(W`iQIzG(}O zgayC-RKQg^E%^Ck-$i|}V1Y+V2s)_S6^VvfrC=eI#W zP2=W`JV!i9O|hkpLfUb2xkZlA)(wS5Yvq92b;`CX-S(CYVH~$UXG^m)`ed7x+-nu*}rRV=@^gDs$S#)lyX)AgRgJx;(-YPUlL7m~U_$xzoj>nPL;fP7)=}7* z33(tpVAj>-mlSENXx5rc6y1VjvK1pz*WDF}a(kM|rT$E8;J-lfI#fE`RZ^5~$h(~u zDtg^mNyqw?N6gy2j>Z`Mw92OiQ43v|x_9is@zxjlfmSNFlPrS_g_J zrfTLdwG13EXuinBq#^ElBd~cn-=u}4ArT#y+L96;Yjp*;@m=_P1Oz(BY_Q^)#v9iV zD@0#h_~}~vS`#w96(zAp4T$+YuxSRmOhjLhN<*@Dk06c_%$%H0@@`4Ww|2&j4*D2N zMHmLeq|Ec$yg~j05^$hr8>JesfR7fY)nI@Lvl+gZ9dEoAsE5DwcA{KhtC-KXz02_G}uI)+8oM-W%IMSXY?@>9;#WM@gL>;dx`~k*14U~# zb!+>uK?+BM758rppz~&eVsZ@RCTD{F$a15YWWA_QyW=PW>Cl+NlBr~%tbcB)@x_E0 zXLQI^CC58Fh_IO`$WSgvM0TLjl%zBz$p$1uHNTM-$g{R9qn<~5G8T}fLa+Nnb-K2+ zCvx)(;n6Kf9T8^j7(>?CbwHQV0mfpFjRSI16kCreul>U7njAI>1$!#VZ!ekg$8^Sufyo-INb<^u*5f=B_z6hA;)H^lqxk=x<4RZj1%}G&` zu&)p%nkaE7z6VIia4dJos#ctsEQ#|TH*|+2U9V^^!ikyNaG>yDSS?uU><9W0y|*Ir z-kbyP`z7IdTNq8Lr_|BTE}dA1z7Qn(=PID^iDt-C&?^1)nGF6s{q-dh2bX0V_L-?M zRq}bk%DB6#>Y4_Di2I|TC|zSvX3RVs*&GKl5G-ee+zzg{RUGl_BSxT68ai!hs&H37a-H-PI(#? zR)lHx{xTZ)(ZovNul3SFDVt-+lkNGqus8nPF518ON>xpgQ%q0u5(gI6+qLFv5%cxMs}GJLV&-D%J` zN2V3!!3tTiQr90Lu=RTfCuAW;8NO>w%_IP@5ZmzXHvleLlLWKE|6H`vkz6tRy^}-% z{viMjnFNIZVjJ##$sg;FG;V;)mIl?^kmn;1zjta)z{OeZ*th*5W~sI9?A006UsaZ1 z{3P z5W-@^QPP>(Hx*eXZRkI~eu1Oy>{l6cr*#|<1BVQ&FFmb|fC5y5$Ozm*l36WxZsVS} z{tp4Z_C>-zHecQ58Ilgyk~J!F183$EnH1Cvj9k?O_(y83>xVCYNyoOV*C2Ct`Q5LZ zhhs_W^b2iySY)IjG&07vMBSKGJCY@ePv|Xhi5V#pPvnDy1 zZWbF-m!eM-j(N`z76|hG;K%lfNIuI=Q?*)D?I3t9MPc=hETGy&;ZzMXLARe1x)Qb3 ziA+{>PkH6hgij-@iW?KEsEQe5zwCVoZ4OsmKzw3aTOK1`Zd8{g)R4JFiXPHvddNRm zgvSSD@^)K7^YhD9>X$VMf?5+~#gAH%WRzHE3nKo6+wK0@#FVNB3u`%`mNBXlUNwav z+P6{R(1Jyqp=lD>--N_N5V^jTs_ctvJA)9p#sW%ZwZ;Mo;oKHKgivZs!R7f9%7o4? zVn8iY5)G~fiqCEykVMq7EWIUUYr-fSogjDZzXWTGYkHs-D8X3MzXXw>1Um<~NHaUY z-wOEycY!IXfqKH+JP+&0 z{pzjPbbk?LgQ;Dkgz|mW;ef9r_8oG<>t*slcd@Gn=bGhZ#kwW*M-7DyylxDb1%4+3 z;hi01d2FSo%u|m91CMNEH=JAM807qPV|o;qlrg^63o)N&_-bAuQ8zv8d16j6>o6ei zdX8+seq;RmPTIg@4E{h&j$Ob$aSIjllWj;JQk-t>8GH_#GxOKcQLVzV}&L`<4 zk}wjGe))+!=Y?2H)Wy;lVsEraL*QMsn+eZ>K@Ot!xpq=QDvSdLDyHR2CX*q4NFIn4 z^?dAHD$6I0q|D<{6*Im5$6F8dp)S7KEZJEjjBZeAk^ZUL_X8G4r#kvUAi;TUMhR?y zvnB(!Jg)(E&^7_rW-Xl;SJO>?vVu1y6&6c3(omFG+G%NB3tn0)+Fy(O6td@SeO9-2 zWFZTJbMN&oTAql5DiAeso}qw9@kt*9v&`3KH3JXUHNZ9ivZj}dLBkDZ-&6TNV-R4vF?x&_zKAtyj7nhPUeS$C!lc(e{$ zooCuSXTjfNn@}*`xxOzS`ssD7BgAM0Xo~`*^;EM(SwHK7;9kw?_V@TW$ZFJ*#b3Ij zdTmZAuMO4Qd~n>Ug$U@*4t+f}4e5kOe8dV^mE*;cd3A>#9uQ2FUPlh7mEF$&2XVs% zL4rXL%z#8Qu6vI#ejwna`hxsikO)Y{_ysYoI3~)OGQV5bhe=H%w@>dE)9@fv0l9T( z&kh(pCF653HvU!LwsJy99a&fSWOXk=keqvi!VbpanUr(gf%UO#2p$(WX7s>r9jj(h z>M2dwRdms=mT)?JoFm4c{JR!fTtaDtkpkxPeic`E0{@n5H`#XwK@yIx#EiUY%(Wc4 zW70T0cZ3esjvSJo$#;GkuMe73VzUzxy<*3%qoMv`DfY{AcX$4q`<%6$dqyp^`t=t% zSsZ!VUjXlj6$?5>E{*RWEbZ# zmFCGS%@kkXzg3}&9?%#GeR$LBP>cx;E}i{P9FtWbG@3}3VVRzh^-Cx%1Lp$-9ZPpl2vyS~{8)RyL6J9$jHxA+wE<%;?m`WVz z(y|XGEfjmn#26`dw;H{eqgTOLf@`u2TxL@_LpHAKYA&`rt7*^Hmc&~fINw!Q^l0LJ zoMxY>1~iSkYGN!MI@*mD5jJ^YH8Ey^Ytj%LWuI{T*DdMV7+Mpc+@;?82sY!2mnf`} zzzRj#^_}vG&yGNPOH`E=-Eb`V26eKrXx@W*lNWywon`Jjnfqpg$xzI^38(Cw0Py>9 z8b5&@ebNlwg9N{HD|($YEjN=SySZ0ibo4py-?Jc|{z2B;D(-y)*kWGHhBmx56%LPc zS?3y?C+-V6F}m~;f(WI)wEjF^1$tK~!Utc`9`-|d?Rn^Eu7lV`W7+kcW>hq;WPBA0 zfr@Ail&7MI!mny;bF?(%Uy-QL%}-k(!@iJ*1Sjn-hGn$xVv0CzCDxs%il}dAFH~>j z#)j?vgx0A)W^p`GqI)qLI8Bq`m~5Sh~n7R^#ATR25JJtJ!h5DjssqFUR}* z?NP$TWpZN&=DDBIn7PEf zfKd~J2yv{96|ru%#z!g>)i!#@MqlHKJ(JcNIMYqpNtHY ztTr$#e|zB#V>v3vCMMw-{b`Swov z>h$HC^X

j&_&-Qq2+mOC|pL&|Cbs-iH%sZ(+|w;)aGo+Bw5?vcTu=@fOnK>8$Db z`SkA4r4)@|n~u(F&+O5y0ASyt2Tez(zBy0!xoxQ)Cfc?A%1$KX>U22LE*89tpHbH! zG=fnl7gwA{|KY?I`buQROlPygt?{#qrqpcC;=`?&&f0_QU8bdFq*mTj&_;W6(7-(; zmpgv5fPjAL*uEgoHP7`7jc7^6bDTkjCpFeH>94cpg6%<|>rsy0C9^LVekt-#Iq+H{ zA@DHNBIMUQl%uz8_sRVrnSqRl=n=0ZMHa0YHikJo^Xc#(v8MDC#Xecnb!Y%dxZT5e zeex^?+N$u8-?r959~2*(Y>=CP8aI%Ic132DI>ptZy@E1uO;&{{T#;QcYULC4rKZAVxfZ_d@BsxbbK4GL^EWtO>EwI21-WSBX9w!X}Cw z6OwZIYR}Y}8HJ%J8=SM`@TAI}*V9&eu1Q$&#R+9#F%Gf$Y|0X+e3|-_cVMv@r8#%g z^I3f{!GhK@n>>pXLQhrLYNw3Q?=-mFLr3}f)1|F*n04~v9hfxSBI}|w*QcP*`=$?jlwY8KdBJnG?_CTe*ntNq zEaUI2#S24@^JmT3O~uu+$M=FDyCo06m`s3c&50oDULT9P)||#V$9i=+_{Kr$8O9h?;TO5 z_G>Oc^tBVIa&{f`;`1FKE4P%LU+#Wzwk?;x^4uznd}-6bplx%V(o}hbtm64}Q;j1@ z+~L--U5VHX&5}YPGFC!z=*+X9t%#Z8VKkh%bD|h|&W9fKt@2Eu zo*z9oyF=4a=s`VjDRq3NOT=O-Q42qkCVt;SZyv9ku}`TLI=`T?h`(7i9JmH#79ef8 zW!=#Ie#^a0p*&&ftgk&g@Jhp~ouWU6mlhWGvXXneBScycN{K^~u9KL9$(*lm9s8PA zSWu`O+~+MTsLrzDxvaX;pSRUlZz+~Bh>GQ;uzMOTB$knc+T@|&Jj8}NMGjU4I@f$4RGbUB{9aOl>mY9D!MiFBR=vnlUGwXpusYW6xE(ZZ0Z9r#yM=gYFBSg0*Ld}%)+m80X1U6^fOVE$hmU)gon6Of%-_$~{%MGC@|Lvy*H#YB&)#B{Lw6!KpEe_L$c2&U%HA zPBcg(7=wtB&rbneqIVrY?4%J>%8S%d;X~&n6f3paHm&$hDy~Jv@tQ53xsYERpEiGe z%OB5GAzlVPq7(90g4vz2wunm`^^n(8A%v~Op|Qhztr1rGw4S^om?(r`kt`3EFr+jM zy*D4Uvzi=q;^!CgqNL3mNo!bLABk=(kmWty+>t6v`{Vq#*PgpX90`w*wPY^s{uyj9 z>Ig%4o*ARRY=t&jV@3DsV;*{iT?b0mzP{Qpb={- zNQr=2*+vCJJfx*k?9o^;_UL@hFfDPxy3$d}448~dilFD+*S|}~Y9b+e6SiSI=KnY2 z58CNtMpA(%Cv6FlTE)5Y5U$IS*v@mED(Vhcy&5WPW5$SVI<^+87bmwsoEd$$2rjBg8Sb(#XJsb90*7F z*GS>qSY-%7rhklftjoX6R*;mJnXQt9io)ZT>rDkvrQ`cY7tNB$(#|0pq)~U^8`MxY zWF++aNZ`d;Wl)J+63uBtZF5hlLa~S#c7%AfQ1i6oI9}U-m)sJ(MrlpkVL=D7+Fzr;cVV6 z)tx2vBT}V#a%fsohw$HguJ5Tk58@UYTv>9em`}o~lS&gsML?jM%*6s|W3|x9Ptj1d z!yfdAlj}8WWFGj4b57eJ{kBcfp!}9AWB#j83jTsc=q*2J{>uf+GQd3jv4_^YIUzHYjfs^A+E<5vx9@60!d`yg=yLg&CCrn_uDwq>se7tQA(PZF{jLVy& z$;5NaQFu&jnbUZhy(Ugz=_-&D5PYo5w$Wx1h@8gr(N6oS3E;)_mUCg=4)?{pqutjL9J=ML5t+7ktY2b zOpDw5vsI6u;zfChr#jv^n}vV13OPMui}HR{Or3M(9llC2eAJfo-Y;rqH6(3Fm?an{ z{~d~`{M6P^-99t?$kl-@-Mpn_c=0Zx+J7r^pX&8Q)XUADMYOOaXs9Sr~Z_ z2v;c#gfG3bf_QkqimCfBtwEyFrY#klBl!p9U02u4aeZpH9DO^kbaZ{a$UDcRjS-nm zu>v)hVw~f>ahq9+q(_8CdQ_dG2ju{j;Xq}g;uWNk4Ea4-3P)by)V~yEUe?|UW18*> z&ls_CS({Q}%w0$6Y(7)o+S)#^NWTE^gJT&wCArmCI`1z-V^=y%<{pDS+>;=ZswGL~ zA#F>}5v}O^A4esn5qqjF%G2;wuE#L1&P+s`6Fo;MrJs(-VF$Eyoj!=B&6e~gnpPb{ zLTJBoL+_@u-hi%wQk5Q^M3IBYH^V8#20Xa#^mLYcH%_avqR641DUb0MCLG6+_&%&* zxtqjDu%YbNFgwGNi6i|cYfC+4LDO9{s3A&@G35!K+9-*`2IBx;_vavs%mt{vh z#Z~K#ws`NFY!aTzVI4KPE*(^Gf8l58-iDlRM|a~kujs+TtbQ7PO+D+5w#2XJc!vD%ZVWUa#-8~e;qmE zSw~X|-U+_tYa=toKB;&s%}aFm&C>vGOX&rN_vNtJ4d*&?#d>7MRY zuOsO^Ngd=(Wb$*`o|`!-A)iH@fK6!7Uu3PG56$^cQF>3fcADG16m9298Bq%M-;;gz z2cDj(?5h^3Ge_4MaY|o;bE}Y-tgFOPL!I;dJl7flR@@GraC-DveE-8l1+Yv=8WBQX z+PQepW0onx$qYofPd$3K)S4W4K|2xzaO4eo`f}K#PR0DWPr&eWJCZFJQXYfaJLB&W z8B3`8V?U4Lw9w9xw|ca}E9rU{CTK{8cOMafkjYVb$_9T!^zZGlZs(Kuc zSTbiEPqN2MPs{I<6Mmc6gP8P(aDN<<`Xbtq2g?H14eI@&XT2JwD+BjQ9%pbFDCRQ!6=l%?MwnY`< zXy*6N(g!RHvSO1JrEB=a!$3e#tiNLAxha$lWm)V&L`)i~kX_b^#!`4?4H=@y5?S9Y*;)|IEA*Y--~0RTT=)6h_kFJO z&vRerxvukUZ>Fb`YT_PUi^Y5gNAAMD1vFL`G^}^6YcZ9pCEnrb4I+j5ZE=c9F`7S_ zB~m7e;12hZL>PJM>Qzw}dn+6KB3`Hd{#E#U`P;cCna(L3TfxF+W;Z?C_t#JN zT_fPvIn1(uo(^*WX;UbmCgODGwFz8o8>U=j1`@1`Y^qst1kUhx&&C<(kAGnTL7~K8 zL+zqmRULQw!)|a3+iW@|c%0LQ&Jl4TDr!E&Qs`#51mq@gmcU|xpBg-r^?4%LTi~!+ z+FFVJFki-pn?~Mmu@njWkS}d7X8WRP~QZY^uEnh++FSrV= zLRh(2LCmhJw-bwhVXR`xY%fI5@~h8^%o>@{-BO~j^35h#w*zKcjs+j7T^E~|-|Bd; zm*z~OFCw}Uy>w!R>-@>mHXRQlAFi^(5Fxy&G4 zuQg!hCeFTpCaXRl?)oA6<#vX}@3*frTNU2*ltcR?9^YJn+NWeKsF%Vq)FBVv#J${wzlTTAO0%T=k6$po0xOL@q9iA3;~=W?_a9@jLI zJo*(&dypPB{Tfjoj{#jgJ}C18`i;I7Ym^Or0I^^%!wM%=E-=kStSn2!b7hWnUSp-T_bnDR{Cc!++33W)VOidcS%7>_B_@4}nM zB)%(LT-(f3cS4;mceZj@v(#2>E>y=>lIs5!A zQfgn!<0ED818mM5uj|?nr)^XVc1tokxuD@jiR_g8y5oxdmKqK0JNq2Sucu+tDyktd zu~jTwys7Kf2NM6um*K0&Y0K2IH-d)RW%z7<3Mfh0-E|1t&sZDHa_lavDbY`mhy>AQ1)MLGF0&6w-74Xe)3hv zj{i!mp~86+BXl;`zpW;zUfix+a{spSHBRUtkrPi}BH8YzlHZ%o)x`_VlUvSd*dDob#)NyLx0M&kyTahg$$hbghWa(-`GRIoQ?eIPaD}xww z1GX4~o}omh!zFL!UC>e;e+))I(KBQTokzQ)*Dk%yWc%c3DDkJt!bzveRJ;GH5-zlU z*ekq^Js$P9!yrD1foF^PVN`IWwscYWiOVx>VAzC%`XW+B{PE0I;h|}o-a_9%DW)TT zEDT`q?zn+SX5J%6tXV))UfbnIxjNL=LVxh(6zRp^mEY0dfi+>>n`%O%2<-Sf%1=!} z1HSfeqZd|oo?9g7atqW5il3;*GW-y^;2yS5Qe~02o~xzL3Fpot(kXbI7UZ!+Y~<_E zv9F)coi}diM0vq$(}fT<3A&}=iMl(HIFzN^1X_uwnq(#H!ck3%m`JD|HOCVpgM8qu z!~Gsb^7jEZq$uDCwhgIAFGM!Z_+q`_X)YB5zH0+0hBq;i`Uacu<0R;q8MWqgT4B)Z-?TWZioZ*{8Utz_)NJRk-#T-M z{Hws~>g*Dz>r!8k#?aSM_eV9=0vPKR73RltmiP~A0TQ40tR6L7`0QqHalb{U#5p4dVNWG zPG?)-6ubDJ_)0PZ@`WG9!p|>efy=uD!ticzvtMrMRK5Q|#aqaiv46pxCrPin1QG-I z&e`WdSG?>`E|_bcOO5IhsM#W;(~P(6a(eSFZO2IW3BGo05S;T|kWn5Du+yzsA>Jz} zyr+iGOw{&ASDb>~xhJo&*t<-3R*w;FUte(IJqj_#_4;0 z?zQ%_BAaXX09{B5+l3vn(sxebsb1|yR%yZAY3)5PTsNhi=tk^B8Cttc50n;iO>rfR{)HJHighzYhJ>ls5t%# zS|+8U`C>8tomx_g1Mx~HGD@CS5>>*j;;Z(?QAd$p!SP1wa157IApPbCu(Yrf=>@B$ z@Uc}^!U@1PCkK(pi6zs=wY;w^A|DmMDeplZAZ4mSk4ElXVHV8Y>$cK(Y_+)kXgyWZ zRL}j92Hf4s=a+7X60KbzpE9puJodKv{%GLmhw%%8JBx!(XVs?M=2FR!CuIr~$Jb^L zSz~o|R7p+w5BA1jzki5{V(kavbw_9oip^h94!!iRSm4W@%X&lqGFE zK1Q1XBzGrRtXQ_J#x3Wz+~{Fk=(ANJ2Nj!F~vU*U-ZN7eTL)&YTho>ltq{)xDxghn~EQU@*~%Us_IyHBkx z-D#Ap+f3L0QFyCabFM6FCs&cal6q#s0WEg5slk8xX6ajTC7)mpguvy?vT~os37uj^ zoVFkZr0y!~4{X5%X~T@crhJ`rtRvaWjg%_h>z`Z6L7wh3S&+V-F;#rF0v^CFLQJEn zQ|h_E>g%+#Piae+MBpN>BC>k(^+{a)4=uYibCx8WYFlO)ezaVv;{~Y7~_nhft zg+!9da*0}}4lV5m5IFek!! zB3s7tgX6XPD(b*fZwN+)h$5=2fMye1gPj|jYJ_xNVam-28a=Ki8qwqsebQO{+c=it zPrMSW>jHt+3ntOd+g@Bo=-uqFMeW%&sgtZTz!&?MaL!ANUE z#Hqe@g&um<3GO@6hEcoFKzuToJ}T}xJJfyeM<71l0iQc6=xJbH9tdx4u@@! z2nB@%ecP)H>VDi*VZ4=6h}G1kdS}>&L;HA~(z$@EFhjhPCB5@+UnAuVX30|t@buo5 z52LNhKo5?;2ihMb+W8nlrMUTff1JtsU=z+}$9n;q^M3twiBufe82yPm1!&CL*VHh+ z3aTL}+(uYvl|1dHB)T~Ib{O`40cr&CR0A<=PqPci+aE6~)TVk@OAFd`P4r{oL?KYt zdhH2?^e*nx5T0hBJKM99g_!M+S7jGcvCZ~n)Wo8f&576T1f78#b#3zGJp`q)&^GV7kNTE-eyx{>i0=7+1>Z{?e zclxQrve@C>ja`6h)NR5+6sJGHKPJKduM?LCo?$M>zSXz#%e0TkJOxtHm>LM09gxB7 zB0bwDVx7j1sIKGz>19hFl$%w+A9OFbSjO6`GSdtx4)?Kcykgi~`qmjgI4lV`SPYJkgPZ!Hd2~h(??42@cMsZZ` zVj9F=ajp#2u&O5X$;nVj>ftOxxdUDgv}db|LOdJ{$~QS5NacprLhMaaHH~CSI|+_| z6dZfHBOlEEuNp_W`W<&b&>FR9B)c%fcyXIX ztqgoync}WRJMkjX3eHksT0H{$xhXTUNXd3wjdmY>Uj2EWebRy_iN-8*E>W^mhF2(c zkVXt79V3ls+>Q258=}1tCHOR4!X=f54o$c2Ts->;SI5vqNsf_0?$3`6zFCq}^--!G zDjL((*{e-i&UR0odd7cyM&;q6>bvd+3**( zsRGHVIYuKsgj2)c!~pL+#{Qxm|Dru4UE%WX;nM)as_OSniw)L`np4aT`IIc|)Gjj| zGp~8;iZnL)m=G6wkW)t~1g;^%IR^|OHu0loUc2oMBcgd)Uxc?le%qSFl#1TSF(NIf zeix4Bxzls4)#=iy#HLQSjT!F_yl5!mvWxt)giAy9y&GM^dnlh4{!Q>6Fb>hue79|V z#H3baG7xS{*(Z8$CP97j+gA6L>O|-qmXjOVFk{wDp3Nk}qVQe-NW9C#iylcCS=;3$ zsgs4ozW1`$oSS!2Yr!4qCov$_icjz~Hwx~i$DTOB1RuD5hV%8Pmx>ZS&9z)_d;Z`XFY&ZYdPdOj5{Y=b1y3#H82t@a%U^dLtsL|dSE)6rVAHT?nztnXw@{(46RQ% zz)6gnRhH}F7tbx9x2j%2buqZ;eVyvM4tKqE$5#Oz1(F?eAs;<}Lvd3-t@aLMDFn}yb#5Vdo;yr~b_I=evHbIsWV` z%6?9;?gU)ttjZF%q!*?%T06}d-E{gXYASy~NcdAW0SoV{J~}%X;Q1swt0Cb~IBnts=gBjXI~Bf^&Z{ zZ4&Io*Bc*~dK}80%GNt6`gWtKJErL5kfN#IW?oHush~G+N!VR6B<-toU%|A-yu?Fx z1^@>@Hp(sRe*uXbnub%W5`%)14I!b=A&ll!sl|`%$ax9V|CHb0{uc`DUsd9hz*0oI z-!l&ZTzsJ63ZePGc~Yx)Jfxx-fQ%&#Dp7 Date: Wed, 4 Nov 2026 18:22:18 +0000 Subject: [PATCH 20/52] registry: apply review A's R1 to R9 and regenerate - R1: sort the variable hits in ns(), so --check is byte-identical on any filesystem's glob order. - R2: the pre-event property income rise is AB2025's 22/42/47%, which Table 6.4 line 6 recosts; line 61 is the further 2ppts. - R3: package 'missing' counts the partial leg (line 4) apart. - R4: quote the bus cap's reference title and href at 2.151.0. - R5: name all six budget levers; line 85 cites budget.nhs, lines 9 and 92 budget.education. - R6: line 60 states capital_gains_badr's one-year lifetime-limit caveat. - R7: line 62's id is dashboard_registered_id; it didn't run on Budget day. - R8, R9: registry_rule and the keys docstring describe measure_key as the rows carry it; the seven's source_note names PolicyEngine's combined run. - Regenerated after T-3 to T-5: source_note drops the hidden fiscal head, table_61_heading carries the section, lines 45 and 47 lose marker 4. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_01LwQq6JvunwtJCAckL6L4aM --- data/uk/ab2026_measures.json | 396 +++++++++--------- pipeline/ab2026_measure_keys.py | 5 +- pipeline/build_ab2026_registry.py | 82 ++-- .../REGISTRY_KEYS.md | 4 +- 4 files changed, 262 insertions(+), 225 deletions(-) diff --git a/data/uk/ab2026_measures.json b/data/uk/ab2026_measures.json index 8efdc6ea0..31753bb1a 100644 --- a/data/uk/ab2026_measures.json +++ b/data/uk/ab2026_measures.json @@ -4,7 +4,7 @@ "event_date": "2026-10-28", "chancellor": "John Healey", "administration": "Burnham government (from 2026-07-20)", - "registry_rule": "One entry per measure ANNOUNCED at Autumn Budget 2026: every HM Treasury Table 6.1 line (including the lines a household model cannot touch), the one Table 6.2 financial transaction with no Table 6.1 line, and the packages producers score. It is the port's key space (#230): every staged AB2026 claim carries conditions.measure_key from this file, and every measure carries a computability verdict against the engine the Budget-day runs used. It stages NO value claims; values live in the harvest under their producers (sources/harvest-uk-ab2026-2026-10-29/).", + "registry_rule": "One entry per measure ANNOUNCED at Autumn Budget 2026: every HM Treasury Table 6.1 line (including the lines a household model cannot touch), the one Table 6.2 financial transaction with no Table 6.1 line, and the packages producers score. It is the port's key space (#230): every staged claim's measure_key is a key from this file, null with conditions.proposed_measure_key (rule 2), or for Table 6.4 recostings the earlier event's key named in conditions.recosting_of; and every measure carries a computability verdict against the engine the Budget-day runs used. It stages NO value claims; values live in the harvest under their producers (sources/harvest-uk-ab2026-2026-10-29/).", "measure_key_rule": "ab2026__, written out per Table 6.1 line in pipeline/ab2026_measure_keys.py so a reworded title cannot move a key; ab2026__package_ for packages; ab2026_option__ for options costed by others. The pre-event registry (data/uk/budget_2026_measures.json, #96) keeps its own keys; pre_event_outcomes records what became of each.", "verification_note": "Every pe_reform_delta and parameter_legs path was RESOLVED against an installed policyengine-uk 2.151.0 on 2026-10-29, and engine_baseline_2026 records what that engine returns for 2026-06-01. Every not_expressible verdict carries a name_search COMPUTED over the pinned engine's parameter tree (4292 leaves) and variable list (1271), never typed. Mechanisms named in a verdict were read from the engine's source at the pin or from the dashboard's source at PolicyEngine/uk-autumn-budget-dashboard-2026@1eca3182.", "engine_pin": { @@ -30,10 +30,10 @@ "reversal_on_certified_world": "the measure is already in the engine's current law; a counterpart executes the pre-measure world as baseline (#56 convention)", "package_of_registry_measures": "a composition of registry measures (`package_of`, a list of keys, as in AB2025); a PolicyEngine number covers the expressible legs, stated as constructed" }, - "exogenous_levers_note": "The engine carries exogenous incidence levers that spread a £ amount supplied as an input: gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'); gov.contrib.policyengine.budget.corporate_incident_tax_change ('Tax increase incident on owners of capital'); gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'). Feeding a Table 6.1 figure through one restates HM Treasury's costing rather than estimating it, so no lever makes a line expressible here.", + "exogenous_levers_note": "The engine carries exogenous incidence levers that spread a £ amount supplied as an input: gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'); gov.contrib.policyengine.budget.corporate_incident_tax_change ('Tax increase incident on owners of capital'); gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'); gov.contrib.policyengine.budget.nhs ('Increase in NHS spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'); gov.contrib.policyengine.budget.education ('Increase in education, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'); gov.contrib.policyengine.budget.high_income_incident_tax_change ('Tax rise for high income earners (functioning proportional to income over £100,000)'). Feeding a Table 6.1 figure through one restates HM Treasury's costing rather than estimating it, so no lever makes a line expressible here.", "action_link_rule": "Every not_expressible measure links to #230, which lists the upstream development items; the bus fare cap's partial verdict links to #106, the bus gap record.", "pre_event_outcomes": { - "ab2026__property_income_tax_rate_rise": "announced: Table 6.1 line 61 (ab2026__property_income_rates_plus_2pp)", + "ab2026__property_income_tax_rate_rise": "an AB2025 measure (ab2025__property_income_separate_rates: 22%/42%/47% from 6 April 2027); Table 6.4 line 6 recosts it, and Table 6.1 line 61 raises those rates by a further 2ppts to 24%/44%/49% (ab2026__property_income_rates_plus_2pp)", "ab2026__bank_profit_surcharge_increase": "announced: Table 6.1 line 55 (ab2026__bank_surcharge_6pc_temporary), temporary, 3% to 6% for 15 months", "ab2026__business_rates_warehouse_up_pub_down": "partly: Table 6.1 line 12 raises pubs' relief (ab2026__business_rates_pubs_relief_35pc); no warehouse surcharge line in Table 6.1 or 6.4", "ab2026__high_value_council_tax_surcharge": "an AB2025 measure; Table 6.1 line 93 changes its reliefs (ab2026__hvcts_deferrals_discounts_exemptions) and Table 6.4 line 11 recosts it", @@ -62,7 +62,7 @@ "measure_key": "ab2026__household_energy_payment_100_bands_a_to_d", "title": "Household Energy Payment: £100 council tax rebate for bands A-D from January 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 1 (Spend, RDEL); exchequer impact GBP m by FY: {'2026-27': '-2625', '2027-28': '0', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 1 (Spend); exchequer impact GBP m by FY: {'2026-27': '-2625', '2027-28': '0', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", "table_61_line": 1, "table_61_head": "Spend", "table_61_heading": "Breathing space for families and businesses > Breathing space for families", @@ -117,7 +117,7 @@ "measure_key": "ab2026__electricity_vat_zero_rate_to_july_2027", "title": "Support for Energy Bills: Cut VAT on domestic electricity from 5% to 0% from 1 October 2026 to 1 July 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 2 (Tax, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '-755', '2027-28': '-305', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 2 (Tax); exchequer impact GBP m by FY: {'2026-27': '-755', '2027-28': '-305', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", "table_61_line": 2, "table_61_head": "Tax", "table_61_heading": "Breathing space for families and businesses > Breathing space for families", @@ -164,7 +164,7 @@ "measure_key": "ab2026__fuel_duty_5p_cut_extended_and_rpi_rise_deferred", "title": "Fuel Duty: Extend the temporary 5p-per-litre duty cut and defer the planned Retail Prices Index (RPI) increase to 1 July 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 3 (Tax, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '-575', '2027-28': '-780', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 3 (Tax); exchequer impact GBP m by FY: {'2026-27': '-575', '2027-28': '-780', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", "table_61_line": 3, "table_61_head": "Tax", "table_61_heading": "Breathing space for families and businesses > Breathing space for families", @@ -209,7 +209,7 @@ "measure_key": "ab2026__carers_allowance_earnings_taper", "title": "Carer’s Allowance: Introduce a taper on earnings above the Weekly Earning Limit within 2030-31", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 7 (Spend, RAME); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '0', '2029-30': '0', '2030-31': '-40', '2031-32': '-80'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 7 (Spend); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '0', '2029-30': '0', '2030-31': '-40', '2031-32': '-80'}", "table_61_line": 7, "table_61_head": "Spend", "table_61_heading": "Breathing space for families and businesses > Breathing space for families", @@ -262,10 +262,10 @@ "measure_key": "ab2026__trading_allowance_3000", "title": "Trading Allowance: Increase the Income Tax Trading Allowance to £3,000 from £1,000 from 6 April 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 15 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '-225', '2029-30': '-165', '2030-31': '-175', '2031-32': '-180'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 15 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '-225', '2029-30': '-165', '2030-31': '-175', '2031-32': '-180'}", "table_61_line": 15, "table_61_head": "Tax", - "table_61_heading": "Breathing space for businesses", + "table_61_heading": "Breathing space for families and businesses > Breathing space for businesses", "program": "income_tax", "computability": "expressible", "construction": "executed_by_dashboard", @@ -313,7 +313,7 @@ "measure_key": "ab2026__pension_tax_allowance", "title": "Pension Tax Allowance: Remove the income tax liability for pensioners with total income up to the value of the new State Pension from 6 April 2027 until 5 April 2030", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 52 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-45', '2028-29': '-90', '2029-30': '-150', '2030-31': '50', '2031-32': '-15'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 52 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-45', '2028-29': '-90', '2029-30': '-150', '2030-31': '50', '2031-32': '-15'}", "table_61_line": 52, "table_61_head": "Tax", "table_61_heading": "Funding the Budget fairly and responsibly > Modernising the tax and welfare system", @@ -367,10 +367,10 @@ "measure_key": "ab2026__property_income_rates_plus_2pp", "title": "Property Income: Increase tax rates on property income by 2ppts at the basic, higher and additional rate from 6 April 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 61 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '35', '2028-29': '575', '2029-30': '435', '2030-31': '445', '2031-32': '455'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 61 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '35', '2028-29': '575', '2029-30': '435', '2030-31': '445', '2031-32': '455'}", "table_61_line": 61, "table_61_head": "Tax", - "table_61_heading": "Making fair choices on revenue raising", + "table_61_heading": "Funding the Budget fairly and responsibly > Making fair choices on revenue raising", "program": "income_tax", "computability": "expressible", "construction": "executed_by_dashboard", @@ -436,14 +436,14 @@ "measure_key": "ab2026__savings_income_rates_plus_2pp", "title": "Savings Income: Increase tax rates on savings income by 2ppts at the basic, higher and additional rate from 6 April 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 62 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '15', '2027-28': '125', '2028-29': '555', '2029-30': '520', '2030-31': '555', '2031-32': '590'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 62 (Tax); exchequer impact GBP m by FY: {'2026-27': '15', '2027-28': '125', '2028-29': '555', '2029-30': '520', '2030-31': '555', '2031-32': '590'}", "table_61_line": 62, "table_61_head": "Tax", - "table_61_heading": "Making fair choices on revenue raising", + "table_61_heading": "Funding the Budget fairly and responsibly > Making fair choices on revenue raising", "program": "income_tax", "computability": "expressible", "construction": "delta_on_current_law", - "dashboard_measure_id": "savings_income_rates_up", + "dashboard_registered_id": "savings_income_rates_up", "pe_reform_delta": { "gov.hmrc.income_tax.rates.savings.basic": { "2027": 0.24, @@ -473,16 +473,16 @@ "gov.hmrc.income_tax.rates.savings.higher": 0.4, "gov.hmrc.income_tax.rates.savings.additional": 0.45 }, - "note": "Registered by the dashboard but not in its published release, so there is no published receipt to cite; tranche 3 computes it from this delta." + "note": "Registered by the dashboard (dashboard_registered_id) but not in its published release, and not run on Budget day, so there is no published receipt to cite; tranche 3 computes it from this delta." }, { "measure_key": "ab2026__badr_lifetime_limit_500000", "title": "Business Asset Disposal Relief (BADR): Reduce the lifetime limit to £500,000 from 6 April 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 60 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '5', '2027-28': '60', '2028-29': '180', '2029-30': '255', '2030-31': '280', '2031-32': '305'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 60 (Tax); exchequer impact GBP m by FY: {'2026-27': '5', '2027-28': '60', '2028-29': '180', '2029-30': '255', '2030-31': '280', '2031-32': '305'}", "table_61_line": 60, "table_61_head": "Tax", - "table_61_heading": "Making fair choices on revenue raising", + "table_61_heading": "Funding the Budget fairly and responsibly > Making fair choices on revenue raising", "program": "capital_gains_tax", "computability": "expressible", "construction": "delta_on_current_law", @@ -503,13 +503,13 @@ "capital_gains_tax", "capital_gains_badr" ], - "note": "capital_gains_tax reads gov.hmrc.cgt.badr.lifetime_limit at the pin (variables/gov/hmrc/capital_gains_tax/capital_gains_tax.py). Not run on Budget day." + "note": "capital_gains_tax reads gov.hmrc.cgt.badr.lifetime_limit at the pin (variables/gov/hmrc/capital_gains_tax/capital_gains_tax.py). capital_gains_badr is an input, and capital_gains_tax applies the lifetime limit to one year's qualifying gains: the variable's documentation says it does so 'as if none of the limit had been used before', so a lower limit binds only on a year's gains above it, not on gains already relieved in earlier years. Not run on Budget day." }, { "measure_key": "ab2026__bus_fare_cap_2_england_outside_london_2027", "title": "Bus Fares: Cap single bus fares at £2 in England outside London for one year from 1 January 2027, funded by reprioritising DESNZ budgets", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 4 (Spend, TDEL); exchequer impact GBP m by FY: {'2026-27': '-50', '2027-28': '-180', '2028-29': '65', '2029-30': '170', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 4 (Spend); exchequer impact GBP m by FY: {'2026-27': '-50', '2027-28': '-180', '2028-29': '65', '2029-30': '170', '2030-31': '0', '2031-32': '0'}", "table_61_line": 4, "table_61_head": "Spend", "table_61_heading": "Breathing space for families and businesses > Breathing space for families", @@ -528,14 +528,14 @@ "2028-01-01": "inf" } }, - "missing": "The pre-measure 2027 cap. The engine already carries the measure: gov.dft.bus.fares.cap is £2 from 1 January 2027 to 1 January 2028 (its metadata cites DfT's '£2 bus fares from January 2027'). A counterpart reverses it, which needs the cap 2027 would have had without the measure; the parameter has one upstream revision (policyengine-uk c8d1f89a9d, 5 October 2026) and records no pre-announcement path. Supersedes data/uk/bus_fares_gap.json's not_expressible verdict at 2.89.2, which predates the bus model.", + "missing": "The pre-measure 2027 cap. The engine already carries the measure: gov.dft.bus.fares.cap is £2 from 1 January 2027 to 1 January 2028 (its metadata cites 'DfT two-pound bus fares from January to December 2027', https://www.gov.uk/government/speeches/2-bus-fares-from-january-2027). A counterpart reverses it, which needs the cap 2027 would have had without the measure; the parameter has one upstream revision (policyengine-uk c8d1f89a9d, 5 October 2026) and records no pre-announcement path. Supersedes data/uk/bus_fares_gap.json's not_expressible verdict at 2.89.2, which predates the bus model.", "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/106" }, { "measure_key": "ab2026__mileage_allowance_payments_55p", "title": "Mileage Rates: Increase the tax-free and tax-deductible car and van rates to 55p for the first 10,000 business miles from 6 April 2026 and the higher mileage rate to 35p from 6 April 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 5 (Tax, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '-30', '2027-28': '-200', '2028-29': '-220', '2029-30': '-225', '2030-31': '-235', '2031-32': '-245'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 5 (Tax); exchequer impact GBP m by FY: {'2026-27': '-30', '2027-28': '-200', '2028-29': '-220', '2029-30': '-225', '2030-31': '-235', '2031-32': '-245'}", "table_61_line": 5, "table_61_head": "Tax", "table_61_heading": "Breathing space for families and businesses > Breathing space for families", @@ -552,7 +552,7 @@ "measure_key": "ab2026__parental_allowance_fathers_partners_adopters", "title": "Parental Pay: Establish a Parental Allowance for fathers, partners and adopters in Great Britain from April 2029", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 6 (Spend, RAME); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '0', '2029-30': '-70', '2030-31': '-70', '2031-32': '-70'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 6 (Spend); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '0', '2029-30': '-70', '2030-31': '-70', '2031-32': '-70'}", "table_61_line": 6, "table_61_head": "Spend", "table_61_heading": "Breathing space for families and businesses > Breathing space for families", @@ -573,7 +573,7 @@ "measure_key": "ab2026__uc_surplus_earnings_threshold_2500_extended", "title": "Universal Credit: Extend the £2,500 surplus earnings threshold for one year from 1 April 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 8 (Spend, RAME); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-140', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 8 (Spend); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-140', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", "table_61_line": 8, "table_61_head": "Spend", "table_61_heading": "Breathing space for families and businesses > Breathing space for families", @@ -590,7 +590,7 @@ "measure_key": "ab2026__cash_isa_limit_held_at_20000", "title": "Cash ISAs: Maintain the cash ISA limit at £20,000 from 6 April 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 11 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-5', '2028-29': '-20', '2029-30': '-35', '2030-31': '-50', '2031-32': '-70'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 11 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-5', '2028-29': '-20', '2029-30': '-35', '2030-31': '-50', '2031-32': '-70'}", "table_61_line": 11, "table_61_head": "Tax", "table_61_heading": "Breathing space for families and businesses > Breathing space for families", @@ -598,7 +598,7 @@ "computability": "not_expressible", "pe_reform_delta": null, "why": "No ISA subscription or subscription limit: the engine holds ISA balances and interest, not subscriptions, so a limit has nothing to bind.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /(^|[._])isa([._]|$)|individual_savings|subscri/. Parameters: 2 node(s) ['gov.hmrc.isa.lifetime.charge_free_age', 'gov.hmrc.isa.lifetime.withdrawal_charge']. Variables: 7 ['lifetime_isa_balance', 'has_lifetime_isa', 'stocks_and_shares_isa', 'household_lifetime_isa_balance', 'cash_isa', 'lifetime_isa_countable_capital']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /(^|[._])isa([._]|$)|individual_savings|subscri/. Parameters: 2 node(s) ['gov.hmrc.isa.lifetime.charge_free_age', 'gov.hmrc.isa.lifetime.withdrawal_charge']. Variables: 7 ['cash_isa', 'has_lifetime_isa', 'household_lifetime_isa_balance', 'individual_savings_account_interest_income', 'lifetime_isa_balance', 'lifetime_isa_countable_capital']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "policyengine_uk_development_item": true, "upstream_item": "policyengine-uk: ISA subscriptions, a subscription limit, and Lifetime and first-time-buyer ISA withdrawals", "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230", @@ -611,15 +611,15 @@ "measure_key": "ab2026__founders_relief_cgt", "title": "Founders' Relief: Introduce a new CGT relief for qualifying founders of high-growth, innovative companies from 6 April 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 26 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-260', '2028-29': '-180', '2029-30': '-260', '2030-31': '-285', '2031-32': '-300'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 26 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-260', '2028-29': '-180', '2029-30': '-260', '2030-31': '-285', '2031-32': '-300'}", "table_61_line": 26, "table_61_head": "Tax", - "table_61_heading": "Backing British business", + "table_61_heading": "Good growth in every postcode > Backing British business", "program": "capital_gains_tax", "computability": "not_expressible", "pe_reform_delta": null, "why": "No founders' gains: the engine's relief-specific gains variable is capital_gains_badr (Business Asset Disposal Relief), and nothing identifies qualifying founders of high-growth companies.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /founder|capital_gains_badr|investors_relief/. Parameters: 0 node(s) []. Variables: 3 ['capital_gains_badr_behavioural_response', 'capital_gains_badr', 'capital_gains_badr_elasticity']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /founder|capital_gains_badr|investors_relief/. Parameters: 0 node(s) []. Variables: 3 ['capital_gains_badr', 'capital_gains_badr_behavioural_response', 'capital_gains_badr_elasticity']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "policyengine_uk_development_item": true, "upstream_item": "policyengine-uk: capital gains by relief (founders' and other business reliefs beyond BADR)", "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" @@ -628,7 +628,7 @@ "measure_key": "ab2026__first_time_buyer_isa", "title": "First Time Buyer ISA: Introduce a new ISA for first-time buyers, replacing the Lifetime ISA for new savers from 6 April 2029", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 38 (Spend, RAME); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '0', '2029-30': '110', '2030-31': '110', '2031-32': '235'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 38 (Spend); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '0', '2029-30': '110', '2030-31': '110', '2031-32': '235'}", "table_61_line": 38, "table_61_head": "Spend", "table_61_heading": "Hope for the future > Housing", @@ -636,7 +636,7 @@ "computability": "not_expressible", "pe_reform_delta": null, "why": "No first-time-buyer ISA, and no ISA subscriptions to move into one.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /first_time|(^|[._])isa([._]|$)/. Parameters: 40 node(s) ['gov.hmrc.isa.lifetime.charge_free_age', 'gov.hmrc.isa.lifetime.withdrawal_charge', 'gov.hmrc.stamp_duty.statistics.residential.household.first_time_buyers_relief[0].threshold', 'gov.hmrc.stamp_duty.statistics.residential.household.first_time_buyers_relief[0].amount', 'gov.hmrc.stamp_duty.statistics.residential.household.first_time_buyers_relief[1].threshold', 'gov.hmrc.stamp_duty.statistics.residential.household.first_time_buyers_relief[1].amount']. Variables: 6 ['lifetime_isa_balance', 'has_lifetime_isa', 'stocks_and_shares_isa', 'household_lifetime_isa_balance', 'cash_isa', 'lifetime_isa_countable_capital']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /first_time|(^|[._])isa([._]|$)/. Parameters: 40 node(s) ['gov.hmrc.isa.lifetime.charge_free_age', 'gov.hmrc.isa.lifetime.withdrawal_charge', 'gov.hmrc.stamp_duty.statistics.residential.household.first_time_buyers_relief[0].threshold', 'gov.hmrc.stamp_duty.statistics.residential.household.first_time_buyers_relief[0].amount', 'gov.hmrc.stamp_duty.statistics.residential.household.first_time_buyers_relief[1].threshold', 'gov.hmrc.stamp_duty.statistics.residential.household.first_time_buyers_relief[1].amount']. Variables: 6 ['cash_isa', 'has_lifetime_isa', 'household_lifetime_isa_balance', 'lifetime_isa_balance', 'lifetime_isa_countable_capital', 'stocks_and_shares_isa']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "policyengine_uk_development_item": true, "upstream_item": "policyengine-uk: ISA subscriptions, a subscription limit, and Lifetime and first-time-buyer ISA withdrawals", "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" @@ -645,7 +645,7 @@ "measure_key": "ab2026__lifetime_isa_withdrawal_charge_20pc_above_cap", "title": "Lifetime ISA: Reduce the withdrawal charge from 25% to 20% for first-time buyers purchasing properties above the £450,000 price cap from 29 October 2026", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 39 (Spend, RAME); exchequer impact GBP m by FY: {'2026-27': '-15', '2027-28': '-20', '2028-29': '-20', '2029-30': '-25', '2030-31': '-25', '2031-32': '-25'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 39 (Spend); exchequer impact GBP m by FY: {'2026-27': '-15', '2027-28': '-20', '2028-29': '-20', '2029-30': '-25', '2030-31': '-25', '2031-32': '-25'}", "table_61_line": 39, "table_61_head": "Spend", "table_61_heading": "Hope for the future > Housing", @@ -666,7 +666,7 @@ "measure_key": "ab2026__pension_death_benefits_two_year_rule_removed", "title": "Pensions Tax: Remove the two-year rule for taxation of pension and lump sum death benefits from 28 October 2026", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 53 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '-5', '2027-28': '-15', '2028-29': '-15', '2029-30': '-15', '2030-31': '-15', '2031-32': '-15'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 53 (Tax); exchequer impact GBP m by FY: {'2026-27': '-5', '2027-28': '-15', '2028-29': '-15', '2029-30': '-15', '2030-31': '-15', '2031-32': '-15'}", "table_61_line": 53, "table_61_head": "Tax", "table_61_heading": "Funding the Budget fairly and responsibly > Modernising the tax and welfare system", @@ -674,7 +674,7 @@ "computability": "not_expressible", "pe_reform_delta": null, "why": "The measure changes when pension and lump sum death benefits are taxed after a member dies (the two-year rule). The engine has no death event; its one death-benefit variable, pension_annual_allowance_taxable_lump_sum_death_benefits, is an input with no formula and no rule about timing after a death.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /death|bereave|lump_sum/. Parameters: 0 node(s) []. Variables: 2 ['pension_annual_allowance_taxable_lump_sum_death_benefits', 'lump_sum_income']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /death|bereave|lump_sum/. Parameters: 0 node(s) []. Variables: 2 ['lump_sum_income', 'pension_annual_allowance_taxable_lump_sum_death_benefits']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "policyengine_uk_development_item": true, "upstream_item": "policyengine-uk: estates, death benefits and their income tax treatment", "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" @@ -683,7 +683,7 @@ "measure_key": "ab2026__vat_5pc_family_days_out_summer_2026", "title": "Great British Summer Savings Scheme: Reduce VAT to 5% on family days out from 25 June 2026 to 1 September 2026", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 66 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '-310', '2027-28': '0', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 66 (Tax); exchequer impact GBP m by FY: {'2026-27': '-310', '2027-28': '0', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", "table_61_line": 66, "table_61_head": "Tax", "table_61_heading": "Other policy measures including pre-announced policies > Pre-announced breathing space measures and associated funding", @@ -704,7 +704,7 @@ "measure_key": "ab2026__free_bus_travel_children_august_2026", "title": "Bus Travel: Fund free bus travel for children aged 5-15 in England from 1 August 2026 until 31 August 2026", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 67 (Spend, RDEL); exchequer impact GBP m by FY: {'2026-27': '-145', '2027-28': '0', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 67 (Spend); exchequer impact GBP m by FY: {'2026-27': '-145', '2027-28': '0', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", "table_61_line": 67, "table_61_head": "Spend", "table_61_heading": "Other policy measures including pre-announced policies > Pre-announced breathing space measures and associated funding", @@ -712,7 +712,7 @@ "computability": "not_expressible", "pe_reform_delta": null, "why": "Free travel enters the bus model only as bus_pass_eligible, a dataset input whose documentation says it 'is not a pass-eligibility calculator'; no rule gives free travel to children aged 5 to 15, and the measure runs for one month (August 2026).", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /bus_fare|fare_allocation|bus_pass|concession/. Parameters: 19 node(s) ['gov.dft.bus.fare_allocation_weight_by_age[0].threshold', 'gov.dft.bus.fare_allocation_weight_by_age[0].amount', 'gov.dft.bus.fare_allocation_weight_by_age[1].threshold', 'gov.dft.bus.fare_allocation_weight_by_age[1].amount', 'gov.dft.bus.fare_allocation_weight_by_age[2].threshold', 'gov.dft.bus.fare_allocation_weight_by_age[2].amount']. Variables: 5 ['bus_fare_spending', 'person_bus_fare_spending', 'household_bus_fare_age_weight', 'bus_pass_eligible', 'bus_fare_spending_reported']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /bus_fare|fare_allocation|bus_pass|concession/. Parameters: 19 node(s) ['gov.dft.bus.fare_allocation_weight_by_age[0].threshold', 'gov.dft.bus.fare_allocation_weight_by_age[0].amount', 'gov.dft.bus.fare_allocation_weight_by_age[1].threshold', 'gov.dft.bus.fare_allocation_weight_by_age[1].amount', 'gov.dft.bus.fare_allocation_weight_by_age[2].threshold', 'gov.dft.bus.fare_allocation_weight_by_age[2].amount']. Variables: 5 ['bus_fare_spending', 'bus_fare_spending_reported', 'bus_pass_eligible', 'household_bus_fare_age_weight', 'person_bus_fare_spending']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "policyengine_uk_development_item": true, "upstream_item": "policyengine-uk: a free-fare rule for children by age and month within the bus fare model", "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" @@ -721,10 +721,10 @@ "measure_key": "ab2026__student_loan_interest_cap_6pc_plan_2_and_3", "title": "Student Loans: Cap interest rates for Plan 2 and Plan 3 student loans at 6% from 1 September 2026 to 31 August 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 83 (Spend, Tax/TME); exchequer impact GBP m by FY: {'2026-27': '-230', '2027-28': '-170', '2028-29': '-10', '2029-30': '-10', '2030-31': '-5', '2031-32': '-15'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 83 (Spend); exchequer impact GBP m by FY: {'2026-27': '-230', '2027-28': '-170', '2028-29': '-10', '2029-30': '-10', '2030-31': '-5', '2031-32': '-15'}", "table_61_line": 83, "table_61_head": "Spend", - "table_61_heading": "Priority Public Services Funding", + "table_61_heading": "Other policy measures including pre-announced policies > Priority Public Services Funding", "program": "student_loans", "computability": "not_expressible", "pe_reform_delta": null, @@ -738,10 +738,10 @@ "measure_key": "ab2026__company_car_tax_rates_2030_31", "title": "Company Car Tax: Set the Benefit-in-Kind rates (appropriate percentages) for electric vehicles, hybrids and internal combustion engine vehicles for 2030-31", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 86 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '0', '2029-30': '0', '2030-31': '310', '2031-32': '345'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 86 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '0', '2029-30': '0', '2030-31': '310', '2031-32': '345'}", "table_61_line": 86, "table_61_head": "Tax", - "table_61_heading": "Other measures", + "table_61_heading": "Other policy measures including pre-announced policies > Other measures", "program": "income_tax", "computability": "not_expressible", "pe_reform_delta": null, @@ -755,10 +755,10 @@ "measure_key": "ab2026__hvcts_deferrals_discounts_exemptions", "title": "High Value Council Tax Surcharge: Confirm eligibility for deferrals, discounts and exemptions from 1 April 2028", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 93 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '-5', '2027-28': '-5', '2028-29': '-45', '2029-30': '-45', '2030-31': '-50', '2031-32': '-50'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 93 (Tax); exchequer impact GBP m by FY: {'2026-27': '-5', '2027-28': '-5', '2028-29': '-45', '2029-30': '-45', '2030-31': '-50', '2031-32': '-50'}", "table_61_line": 93, "table_61_head": "Tax", - "table_61_heading": "Other measures", + "table_61_heading": "Other policy measures including pre-announced policies > Other measures", "program": "council_tax", "computability": "not_expressible", "pe_reform_delta": null, @@ -772,15 +772,15 @@ "measure_key": "ab2026__early_years_books_access", "title": "Early Years: Fund increased access to books for children aged 0 to 5", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 9 (Spend, RDEL); exchequer impact GBP m by FY: {'2026-27': '-5', '2027-28': '-5', '2028-29': '-5', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 9 (Spend); exchequer impact GBP m by FY: {'2026-27': '-5', '2027-28': '-5', '2028-29': '-5', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", "table_61_line": 9, "table_61_head": "Spend", "table_61_heading": "Breathing space for families and businesses > Breathing space for families", "program": "public_spending", "computability": "not_expressible", "pe_reform_delta": null, - "why": "Public spending (books for children aged 0 to 5), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /early_years|childcare|book/. Parameters: 73 node(s) ['gov.dfe.childcare_funding_rate[0].threshold', 'gov.dfe.childcare_funding_rate[0].amount', 'gov.dfe.childcare_funding_rate[1].threshold', 'gov.dfe.childcare_funding_rate[1].amount', 'gov.dfe.childcare_funding_rate[2].threshold', 'gov.dfe.childcare_funding_rate[2].amount']. Variables: 76 ['tax_free_childcare_uses_qualifying_provider', 'childcare_expenses', 'early_years_school_start_deferred', 'early_years_term_share_before_school', 'weekly_childcare_expenses', 'WTC_childcare_element']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "why": "Public spending (books for children aged 0 to 5), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.education ('Increase in education, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /early_years|childcare|book/. Parameters: 73 node(s) ['gov.dfe.childcare_funding_rate[0].threshold', 'gov.dfe.childcare_funding_rate[0].amount', 'gov.dfe.childcare_funding_rate[1].threshold', 'gov.dfe.childcare_funding_rate[1].amount', 'gov.dfe.childcare_funding_rate[2].threshold', 'gov.dfe.childcare_funding_rate[2].amount']. Variables: 76 ['WTC_childcare_element', 'childcare_expenses', 'childcare_grant', 'childcare_grant_child_eligible', 'childcare_grant_child_has_special_educational_needs', 'childcare_grant_eligible']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -788,7 +788,7 @@ "measure_key": "ab2026__food_and_heating_oil_tariff_suspensions", "title": "Tariff Suspension Packages: Suspend tariffs on 135 food products from 21 June 2026 until 31 December 2028; 101 food products and kerosene for heating oil from 26 November 2026 until 31 December 2028; and fertiliser from 26 November 2026 until 31 December 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 10 (Tax, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '-50', '2027-28': '-75', '2028-29': '-55', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 10 (Tax); exchequer impact GBP m by FY: {'2026-27': '-50', '2027-28': '-75', '2028-29': '-55', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", "table_61_line": 10, "table_61_head": "Tax", "table_61_heading": "Breathing space for families and businesses > Breathing space for families", @@ -796,7 +796,7 @@ "computability": "not_expressible", "pe_reform_delta": null, "why": "An indirect tax or duty collected from businesses (tariffs on food products, heating oil and fertiliser), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /tariff|customs|import/. Parameters: 26 node(s) ['gov.dwp.ESA.income.capital.tariff_income.amount', 'gov.dwp.ESA.income.capital.tariff_income.step', 'gov.dwp.ESA.income.capital.tariff_income.threshold', 'gov.dwp.JSA.income.capital.tariff_income.amount', 'gov.dwp.JSA.income.capital.tariff_income.step', 'gov.dwp.JSA.income.capital.tariff_income.threshold']. Variables: 5 ['jsa_income_tariff_income', 'esa_income_tariff_income', 'income_support_tariff_income', 'housing_benefit_tariff_income', 'uc_tariff_income']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /tariff|customs|import/. Parameters: 26 node(s) ['gov.dwp.ESA.income.capital.tariff_income.amount', 'gov.dwp.ESA.income.capital.tariff_income.step', 'gov.dwp.ESA.income.capital.tariff_income.threshold', 'gov.dwp.JSA.income.capital.tariff_income.amount', 'gov.dwp.JSA.income.capital.tariff_income.step', 'gov.dwp.JSA.income.capital.tariff_income.threshold']. Variables: 5 ['esa_income_tariff_income', 'housing_benefit_tariff_income', 'income_support_tariff_income', 'jsa_income_tariff_income', 'uc_tariff_income']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -804,15 +804,15 @@ "measure_key": "ab2026__business_rates_pubs_relief_35pc", "title": "Business Rates: Increase business rates relief for pubs, social clubs and live music venues by 20ppts to 35% from 1 April 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 12 (Tax, Tax/TME); exchequer impact GBP m by FY: {'2026-27': '5', '2027-28': '-160', '2028-29': '-115', '2029-30': '-5', '2030-31': '-5', '2031-32': '-5'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 12 (Tax); exchequer impact GBP m by FY: {'2026-27': '5', '2027-28': '-160', '2028-29': '-115', '2029-30': '-5', '2030-31': '-5', '2031-32': '-5'}", "table_61_line": 12, "table_61_head": "Tax", - "table_61_heading": "Breathing space for businesses", + "table_61_heading": "Breathing space for families and businesses > Breathing space for businesses", "program": "business_rates", "computability": "not_expressible", "pe_reform_delta": null, "why": "A business rates measure (business rates relief for pubs and venues), out of the household model's scope. The engine carries business rates only as a household incidence, business_rates ('Total incidence from exposure to business rates via corporate shareholdings'), and its business rates parameters are revenue statistics (gov.hmrc.business_rates.statistics.revenue.*): it has no properties, reliefs or thresholds for the measure to change, so its £ change would have to be supplied as an input, restating the costing.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /business_rates|non_domestic|nndr/. Parameters: 4 node(s) ['gov.hmrc.business_rates.statistics.revenue.ENGLAND', 'gov.hmrc.business_rates.statistics.revenue.NORTHERN_IRELAND', 'gov.hmrc.business_rates.statistics.revenue.SCOTLAND', 'gov.hmrc.business_rates.statistics.revenue.WALES']. Variables: 2 ['business_rates', 'baseline_business_rates']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /business_rates|non_domestic|nndr/. Parameters: 4 node(s) ['gov.hmrc.business_rates.statistics.revenue.ENGLAND', 'gov.hmrc.business_rates.statistics.revenue.NORTHERN_IRELAND', 'gov.hmrc.business_rates.statistics.revenue.SCOTLAND', 'gov.hmrc.business_rates.statistics.revenue.WALES']. Variables: 2 ['baseline_business_rates', 'business_rates']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -820,15 +820,15 @@ "measure_key": "ab2026__business_rates_sbrr_rhl_thresholds_raised", "title": "Business Rates: Increase valuation thresholds for Small Business Rate Relief for retail, hospitality and leisure (RHL) properties from £12,000 to £18,000 for 100% relief and from £15,000 to £21,000 for tapered relief from 1 April 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 13 (Tax, Tax/TME); exchequer impact GBP m by FY: {'2026-27': '5', '2027-28': '-315', '2028-29': '-320', '2029-30': '0', '2030-31': '-5', '2031-32': '-5'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 13 (Tax); exchequer impact GBP m by FY: {'2026-27': '5', '2027-28': '-315', '2028-29': '-320', '2029-30': '0', '2030-31': '-5', '2031-32': '-5'}", "table_61_line": 13, "table_61_head": "Tax", - "table_61_heading": "Breathing space for businesses", + "table_61_heading": "Breathing space for families and businesses > Breathing space for businesses", "program": "business_rates", "computability": "not_expressible", "pe_reform_delta": null, "why": "A business rates measure (small business rate relief thresholds), out of the household model's scope. The engine carries business rates only as a household incidence, business_rates ('Total incidence from exposure to business rates via corporate shareholdings'), and its business rates parameters are revenue statistics (gov.hmrc.business_rates.statistics.revenue.*): it has no properties, reliefs or thresholds for the measure to change, so its £ change would have to be supplied as an input, restating the costing.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /business_rates|small_business|non_domestic/. Parameters: 4 node(s) ['gov.hmrc.business_rates.statistics.revenue.ENGLAND', 'gov.hmrc.business_rates.statistics.revenue.NORTHERN_IRELAND', 'gov.hmrc.business_rates.statistics.revenue.SCOTLAND', 'gov.hmrc.business_rates.statistics.revenue.WALES']. Variables: 2 ['business_rates', 'baseline_business_rates']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /business_rates|small_business|non_domestic/. Parameters: 4 node(s) ['gov.hmrc.business_rates.statistics.revenue.ENGLAND', 'gov.hmrc.business_rates.statistics.revenue.NORTHERN_IRELAND', 'gov.hmrc.business_rates.statistics.revenue.SCOTLAND', 'gov.hmrc.business_rates.statistics.revenue.WALES']. Variables: 2 ['baseline_business_rates', 'business_rates']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -836,15 +836,15 @@ "measure_key": "ab2026__business_rates_rhl_transitional_support", "title": "Business Rates: Introduce a transitional support package, including a two-year real terms bills freeze for RHL businesses and a one-year extension of the 2023 Supporting Small Business support, part funded by extending the Transitional Relief Supplement for two years from 1 April 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 14 (Tax, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-10', '2028-29': '-230', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 14 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-10', '2028-29': '-230', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", "table_61_line": 14, "table_61_head": "Tax", - "table_61_heading": "Breathing space for businesses", + "table_61_heading": "Breathing space for families and businesses > Breathing space for businesses", "program": "business_rates", "computability": "not_expressible", "pe_reform_delta": null, "why": "A business rates measure (business rates transitional support), out of the household model's scope. The engine carries business rates only as a household incidence, business_rates ('Total incidence from exposure to business rates via corporate shareholdings'), and its business rates parameters are revenue statistics (gov.hmrc.business_rates.statistics.revenue.*): it has no properties, reliefs or thresholds for the measure to change, so its £ change would have to be supplied as an input, restating the costing.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /business_rates|transitional|non_domestic/. Parameters: 4 node(s) ['gov.hmrc.business_rates.statistics.revenue.ENGLAND', 'gov.hmrc.business_rates.statistics.revenue.NORTHERN_IRELAND', 'gov.hmrc.business_rates.statistics.revenue.SCOTLAND', 'gov.hmrc.business_rates.statistics.revenue.WALES']. Variables: 2 ['business_rates', 'baseline_business_rates']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /business_rates|transitional|non_domestic/. Parameters: 4 node(s) ['gov.hmrc.business_rates.statistics.revenue.ENGLAND', 'gov.hmrc.business_rates.statistics.revenue.NORTHERN_IRELAND', 'gov.hmrc.business_rates.statistics.revenue.SCOTLAND', 'gov.hmrc.business_rates.statistics.revenue.WALES']. Variables: 2 ['baseline_business_rates', 'business_rates']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -852,15 +852,15 @@ "measure_key": "ab2026__business_tariff_suspensions", "title": "Business Tariff Suspensions: Suspend tariffs on 72 products from 5 August 2026 and extend suspensions for 361 products in 2027-28", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 16 (Tax, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '-65', '2027-28': '-215', '2028-29': '-155', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 16 (Tax); exchequer impact GBP m by FY: {'2026-27': '-65', '2027-28': '-215', '2028-29': '-155', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", "table_61_line": 16, "table_61_head": "Tax", - "table_61_heading": "Breathing space for businesses", + "table_61_heading": "Breathing space for families and businesses > Breathing space for businesses", "program": "indirect_tax", "computability": "not_expressible", "pe_reform_delta": null, "why": "An indirect tax or duty collected from businesses (tariff suspensions on business inputs), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /tariff|customs|import/. Parameters: 26 node(s) ['gov.dwp.ESA.income.capital.tariff_income.amount', 'gov.dwp.ESA.income.capital.tariff_income.step', 'gov.dwp.ESA.income.capital.tariff_income.threshold', 'gov.dwp.JSA.income.capital.tariff_income.amount', 'gov.dwp.JSA.income.capital.tariff_income.step', 'gov.dwp.JSA.income.capital.tariff_income.threshold']. Variables: 5 ['jsa_income_tariff_income', 'esa_income_tariff_income', 'income_support_tariff_income', 'housing_benefit_tariff_income', 'uc_tariff_income']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /tariff|customs|import/. Parameters: 26 node(s) ['gov.dwp.ESA.income.capital.tariff_income.amount', 'gov.dwp.ESA.income.capital.tariff_income.step', 'gov.dwp.ESA.income.capital.tariff_income.threshold', 'gov.dwp.JSA.income.capital.tariff_income.amount', 'gov.dwp.JSA.income.capital.tariff_income.step', 'gov.dwp.JSA.income.capital.tariff_income.threshold']. Variables: 5 ['esa_income_tariff_income', 'housing_benefit_tariff_income', 'income_support_tariff_income', 'jsa_income_tariff_income', 'uc_tariff_income']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -868,10 +868,10 @@ "measure_key": "ab2026__climate_change_levy_rates_2028", "title": "Climate Change Levy: Freeze electricity & liquefied petroleum gas main rates, and uprate gas and solid fuels main rates by RPI from 1 April 2028", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 17 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '-25', '2029-30': '-25', '2030-31': '-25', '2031-32': '-25'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 17 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '-25', '2029-30': '-25', '2030-31': '-25', '2031-32': '-25'}", "table_61_line": 17, "table_61_head": "Tax", - "table_61_heading": "Breathing space for businesses", + "table_61_heading": "Breathing space for families and businesses > Breathing space for businesses", "program": "indirect_tax", "computability": "not_expressible", "pe_reform_delta": null, @@ -884,15 +884,15 @@ "measure_key": "ab2026__red_diesel_duty_cut", "title": "Red Diesel: Reduce duty rates by 3.7p-per-litre from 15 June 2026 to 1 July 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 18 (Tax, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '-60', '2027-28': '-30', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 18 (Tax); exchequer impact GBP m by FY: {'2026-27': '-60', '2027-28': '-30', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", "table_61_line": 18, "table_61_head": "Tax", - "table_61_heading": "Breathing space for businesses", + "table_61_heading": "Breathing space for families and businesses > Breathing space for businesses", "program": "indirect_tax", "computability": "not_expressible", "pe_reform_delta": null, "why": "An indirect tax or duty collected from businesses (red diesel duty), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /red_diesel|rebated|fuel_duty/. Parameters: 4 node(s) ['gov.hmrc.fuel_duty.lpg', 'gov.hmrc.fuel_duty.natural_gas', 'gov.hmrc.fuel_duty.petrol_and_diesel', 'gov.hmrc.fuel_duty.rural_fuel_duty_relief']. Variables: 2 ['in_rural_fuel_duty_relief_area', 'fuel_duty']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /red_diesel|rebated|fuel_duty/. Parameters: 4 node(s) ['gov.hmrc.fuel_duty.lpg', 'gov.hmrc.fuel_duty.natural_gas', 'gov.hmrc.fuel_duty.petrol_and_diesel', 'gov.hmrc.fuel_duty.rural_fuel_duty_relief']. Variables: 2 ['fuel_duty', 'in_rural_fuel_duty_relief_area']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -900,10 +900,10 @@ "measure_key": "ab2026__alcohol_duty_small_producer_relief", "title": "Alcohol Duty: Increase the generosity of Small Producer Relief from 1 February 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 19 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-10', '2028-29': '-10', '2029-30': '-10', '2030-31': '-10', '2031-32': '-10'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 19 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-10', '2028-29': '-10', '2029-30': '-10', '2030-31': '-10', '2031-32': '-10'}", "table_61_line": 19, "table_61_head": "Tax", - "table_61_heading": "Breathing space for businesses", + "table_61_heading": "Breathing space for families and businesses > Breathing space for businesses", "program": "indirect_tax", "computability": "not_expressible", "pe_reform_delta": null, @@ -916,7 +916,7 @@ "measure_key": "ab2026__good_growth_funds", "title": "Good Growth Funds: Seed funding for Mayors and local leaders in England and £1.5bn finance for Mayoral Strategic Authorities with a mechanism for subsidised lending (resource impacts of subsidy)", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 20 (Spend, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-340', '2028-29': '-30', '2029-30': '-35', '2030-31': '-55', '2031-32': '-60'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 20 (Spend); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-340', '2028-29': '-30', '2029-30': '-35', '2030-31': '-55', '2031-32': '-60'}", "table_61_line": 20, "table_61_head": "Spend", "table_61_heading": "Good growth in every postcode > Delivering Fiscal Devolution", @@ -932,7 +932,7 @@ "measure_key": "ab2026__business_rates_retention_mayoral_authorities", "title": "Business Rates Retention: Devolve a share of business rates income and growth to Mayoral Strategic Authorities from April 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 21 (Spend, RAME); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-10', '2028-29': '-45', '2029-30': '-75', '2030-31': '-110', '2031-32': '-140'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 21 (Spend); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-10', '2028-29': '-45', '2029-30': '-75', '2030-31': '-110', '2031-32': '-140'}", "table_61_line": 21, "table_61_head": "Spend", "table_61_heading": "Good growth in every postcode > Delivering Fiscal Devolution", @@ -940,7 +940,7 @@ "computability": "not_expressible", "pe_reform_delta": null, "why": "Public spending (business rates retention by mayoral authorities), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /business_rates|retention|mayor/. Parameters: 4 node(s) ['gov.hmrc.business_rates.statistics.revenue.ENGLAND', 'gov.hmrc.business_rates.statistics.revenue.NORTHERN_IRELAND', 'gov.hmrc.business_rates.statistics.revenue.SCOTLAND', 'gov.hmrc.business_rates.statistics.revenue.WALES']. Variables: 3 ['business_rates', 'baseline_business_rates', 'relative_capital_gains_retention_rate_change']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /business_rates|retention|mayor/. Parameters: 4 node(s) ['gov.hmrc.business_rates.statistics.revenue.ENGLAND', 'gov.hmrc.business_rates.statistics.revenue.NORTHERN_IRELAND', 'gov.hmrc.business_rates.statistics.revenue.SCOTLAND', 'gov.hmrc.business_rates.statistics.revenue.WALES']. Variables: 3 ['baseline_business_rates', 'business_rates', 'relative_capital_gains_retention_rate_change']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -948,7 +948,7 @@ "measure_key": "ab2026__local_apprenticeship_services", "title": "Youth Employment and Skills: Expand local apprenticeship services to all current mayors", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 22 (Spend, RDEL); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-50', '2028-29': '-70', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 22 (Spend); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-50', '2028-29': '-70', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", "table_61_line": 22, "table_61_head": "Spend", "table_61_heading": "Good growth in every postcode > Delivering Fiscal Devolution", @@ -964,7 +964,7 @@ "measure_key": "ab2026__regional_growth_investments", "title": "Regional Growth: Fund priority investments and local growth & regeneration projects across the UK", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 23 (Spend, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '-20', '2027-28': '-85', '2028-29': '-45', '2029-30': '-30', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 23 (Spend); exchequer impact GBP m by FY: {'2026-27': '-20', '2027-28': '-85', '2028-29': '-45', '2029-30': '-30', '2030-31': '0', '2031-32': '0'}", "table_61_line": 23, "table_61_head": "Spend", "table_61_heading": "Good growth in every postcode > Delivering Fiscal Devolution", @@ -980,7 +980,7 @@ "measure_key": "ab2026__cambridge_development_corporation", "title": "Cambridge Development Corporation: Pilot a new government financing model for land, housing and infrastructure development from April 2027 (capital grant element)", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 24 (Spend, CDEL); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '-50', '2029-30': '-60', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 24 (Spend); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '-50', '2029-30': '-60', '2030-31': '0', '2031-32': '0'}", "table_61_line": 24, "table_61_head": "Spend", "table_61_heading": "Good growth in every postcode > Delivering Fiscal Devolution", @@ -996,7 +996,7 @@ "measure_key": "ab2026__strategic_authorities_capacity_funding", "title": "Strategic Authorities: Increase funding to build capacity to deliver new powers and investment responsibilities", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 25 (Spend, RDEL); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-180', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 25 (Spend); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-180', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", "table_61_line": 25, "table_61_head": "Spend", "table_61_heading": "Good growth in every postcode > Delivering Fiscal Devolution", @@ -1012,10 +1012,10 @@ "measure_key": "ab2026__rd_relief_claim_notification_form_removed", "title": "R&D Tax Reliefs: Remove the Claim Notification Form from 1 April 2027 and simplify the Additional Information Form from Autumn 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 27 (Spend, RAME); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-45', '2028-29': '-60', '2029-30': '-65', '2030-31': '-65', '2031-32': '-70'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 27 (Spend); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-45', '2028-29': '-60', '2029-30': '-65', '2030-31': '-65', '2031-32': '-70'}", "table_61_line": 27, "table_61_head": "Spend", - "table_61_heading": "Backing British business", + "table_61_heading": "Good growth in every postcode > Backing British business", "program": "business_tax", "computability": "not_expressible", "pe_reform_delta": null, @@ -1028,15 +1028,15 @@ "measure_key": "ab2026__emi2_option_limit_120000", "title": "Enterprise Management Incentives 2 (EMI2): Increase the individual option limit to £120,000, extending the exercise period to 15 years and remove interaction with EMI option limit from 6 April 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 28 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-25', '2028-29': '-35', '2029-30': '-45', '2030-31': '-145', '2031-32': '-140'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 28 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-25', '2028-29': '-35', '2029-30': '-45', '2030-31': '-145', '2031-32': '-140'}", "table_61_line": 28, "table_61_head": "Tax", - "table_61_heading": "Backing British business", + "table_61_heading": "Good growth in every postcode > Backing British business", "program": "business_tax", "computability": "not_expressible", "pe_reform_delta": null, "why": "A tax on businesses (Enterprise Management Incentives share options), out of the household model's scope: it changes what companies owe. The engine's corporate channel, gov.contrib.policyengine.budget.corporate_incident_tax_change ('Tax increase incident on owners of capital'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /enterprise_management|share_option|emi/. Parameters: 27 node(s) ['gov.dft.vehicle_excise_duty.band_k_max_emissions', 'gov.dft.vehicle_excise_duty.zero_emission_exempt', 'gov.dft.vehicle_excise_duty.zero_emission_expensive_car_threshold', 'gov.dft.vehicle_excise_duty.zero_emission_supplement_registration_start', 'gov.dwp.carer_premium.single', 'gov.dwp.disability_premia.disability_couple']. Variables: 16 ['car_co2_emissions', 'severe_disability_premium_before_non_dependant_test', 'severe_disability_premium', 'enhanced_disability_premium', 'pension_age_severe_disability_premium', 'receives_severe_disability_premium_qualifying_benefit']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /enterprise_management|share_option|emi/. Parameters: 27 node(s) ['gov.dft.vehicle_excise_duty.band_k_max_emissions', 'gov.dft.vehicle_excise_duty.zero_emission_exempt', 'gov.dft.vehicle_excise_duty.zero_emission_expensive_car_threshold', 'gov.dft.vehicle_excise_duty.zero_emission_supplement_registration_start', 'gov.dwp.carer_premium.single', 'gov.dwp.disability_premia.disability_couple']. Variables: 16 ['benefits_premiums', 'car_co2_emissions', 'carer_premium', 'council_tax_reduction_premiums', 'council_tax_reduction_severe_disability_premium', 'disability_premium']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -1044,10 +1044,10 @@ "measure_key": "ab2026__british_business_bank_backing_innovation", "title": "Backing Innovation: Bring forward £1bn investment and increase financial capacity by £2bn for the British Business Bank to support additional investment in the UK (resource impacts of increased lending)", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 29 (Spend, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '5', '2028-29': '0', '2029-30': '5', '2030-31': '10', '2031-32': '25'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 29 (Spend); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '5', '2028-29': '0', '2029-30': '5', '2030-31': '10', '2031-32': '25'}", "table_61_line": 29, "table_61_head": "Spend", - "table_61_heading": "Backing British business", + "table_61_heading": "Good growth in every postcode > Backing British business", "program": "financial_transactions", "computability": "not_expressible", "pe_reform_delta": null, @@ -1060,10 +1060,10 @@ "measure_key": "ab2026__national_wealth_fund_capital", "title": "National Wealth Fund: Invest additional capital to support growth, including £2bn for the new Energy Wealth Fund (resource impacts of increased lending)", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 30 (Spend, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '35', '2029-30': '120', '2030-31': '250', '2031-32': '470'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 30 (Spend); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '35', '2029-30': '120', '2030-31': '250', '2031-32': '470'}", "table_61_line": 30, "table_61_head": "Spend", - "table_61_heading": "Backing British business", + "table_61_heading": "Good growth in every postcode > Backing British business", "program": "financial_transactions", "computability": "not_expressible", "pe_reform_delta": null, @@ -1079,7 +1079,7 @@ "source_note": "HM Treasury, Budget 2026 Table 6.1 line 31 (Spend); exchequer impact GBP m by FY: {'2026-27': '-35', '2027-28': '-115', '2028-29': '-185', '2029-30': '-145', '2030-31': '-5', '2031-32': '-5'}", "table_61_line": 31, "table_61_head": "Spend", - "table_61_heading": "Backing British business", + "table_61_heading": "Good growth in every postcode > Backing British business", "program": "public_spending", "computability": "not_expressible", "pe_reform_delta": null, @@ -1092,15 +1092,15 @@ "measure_key": "ab2026__ev_charging_public_chargers", "title": "EV Charging: Reduce the cost of using local, lower-powered public chargers in Great Britain", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 32 (Spend, RDEL); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-50', '2028-29': '-50', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 32 (Spend); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-50', '2028-29': '-50', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", "table_61_line": 32, "table_61_head": "Spend", - "table_61_heading": "Backing British business", + "table_61_heading": "Good growth in every postcode > Backing British business", "program": "public_spending", "computability": "not_expressible", "pe_reform_delta": null, "why": "Public spending (public EV charging costs), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /charg|electric_vehicle/. Parameters: 21 node(s) ['gov.dwp.housing_benefit.ineligible_charges.meals.at_least_three_a_day.aged_16_or_over', 'gov.dwp.housing_benefit.ineligible_charges.meals.at_least_three_a_day.under_16', 'gov.dwp.housing_benefit.ineligible_charges.meals.breakfast_only', 'gov.dwp.housing_benefit.ineligible_charges.meals.fewer_than_three_a_day.aged_16_or_over', 'gov.dwp.housing_benefit.ineligible_charges.meals.fewer_than_three_a_day.under_16', 'gov.hmrc.child_benefit.opt_out_charge_share']. Variables: 6 ['water_and_sewerage_charges', 'housing_service_charges', 'child_benefit_less_tax_charge', 'high_value_council_tax_surcharge', 'income_tax_pre_charges', 'pension_annual_allowance_chargeable_amount']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /charg|electric_vehicle/. Parameters: 21 node(s) ['gov.dwp.housing_benefit.ineligible_charges.meals.at_least_three_a_day.aged_16_or_over', 'gov.dwp.housing_benefit.ineligible_charges.meals.at_least_three_a_day.under_16', 'gov.dwp.housing_benefit.ineligible_charges.meals.breakfast_only', 'gov.dwp.housing_benefit.ineligible_charges.meals.fewer_than_three_a_day.aged_16_or_over', 'gov.dwp.housing_benefit.ineligible_charges.meals.fewer_than_three_a_day.under_16', 'gov.hmrc.child_benefit.opt_out_charge_share']. Variables: 6 ['child_benefit_less_tax_charge', 'high_value_council_tax_surcharge', 'housing_service_charges', 'income_tax_pre_charges', 'pension_annual_allowance_chargeable_amount', 'water_and_sewerage_charges']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -1108,7 +1108,7 @@ "measure_key": "ab2026__rough_sleeping_investment", "title": "Rough Sleeping: Invest in a route off the streets across the Parliament, alongside longer-term accommodation and intensive support", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 33 (Spend, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '-175', '2027-28': '-515', '2028-29': '-580', '2029-30': '-85', '2030-31': '-90', '2031-32': '-95'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 33 (Spend); exchequer impact GBP m by FY: {'2026-27': '-175', '2027-28': '-515', '2028-29': '-580', '2029-30': '-85', '2030-31': '-90', '2031-32': '-95'}", "table_61_line": 33, "table_61_head": "Spend", "table_61_heading": "Hope for the future > Housing", @@ -1124,7 +1124,7 @@ "measure_key": "ab2026__rough_sleeping_reprioritised_mhclg_budgets", "title": "Rough Sleeping: Reprioritise existing MHCLG budgets to tackle rough sleeping", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 34 (Spend, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '120', '2027-28': '405', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 34 (Spend); exchequer impact GBP m by FY: {'2026-27': '120', '2027-28': '405', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", "table_61_line": 34, "table_61_head": "Spend", "table_61_heading": "Hope for the future > Housing", @@ -1140,7 +1140,7 @@ "measure_key": "ab2026__social_and_affordable_homes_programme", "title": "Social and Affordable Homes Programme: Fund delivery of around 10,000 social and affordable homes", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 35 (Spend, CDEL); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-250', '2028-29': '-500', '2029-30': '-500', '2030-31': '-500', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 35 (Spend); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-250', '2028-29': '-500', '2029-30': '-500', '2030-31': '-500', '2031-32': '0'}", "table_61_line": 35, "table_61_head": "Spend", "table_61_heading": "Hope for the future > Housing", @@ -1156,7 +1156,7 @@ "measure_key": "ab2026__vat_zero_rate_land_for_social_housing", "title": "Social Housing: Introduce a zero rate of VAT for land sold to registered social housing providers (relevant housing associations) from 1 April 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 36 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-15', '2028-29': '-15', '2029-30': '-15', '2030-31': '-15', '2031-32': '-15'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 36 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-15', '2028-29': '-15', '2029-30': '-15', '2030-31': '-15', '2031-32': '-15'}", "table_61_line": 36, "table_61_head": "Tax", "table_61_heading": "Hope for the future > Housing", @@ -1164,7 +1164,7 @@ "computability": "not_expressible", "pe_reform_delta": null, "why": "An indirect tax or duty collected from businesses (VAT on land sold to social housing providers), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /vat|social_housing/. Parameters: 64 node(s) ['gov.contrib.conservatives.cb_hitc_household', 'gov.contrib.conservatives.pensioner_personal_allowance', 'gov.contrib.labour.private_school_vat', 'gov.economic_assumptions.yoy_growth.obr.private_pension_index', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.UNITED_KINGDOM', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.ENGLAND']. Variables: 13 ['private_pension_wealth', 'private_pension_income', 'private_transfer_income', 'private_school_vat', 'attends_private_school_random_draw', 'attends_private_school']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /vat|social_housing/. Parameters: 64 node(s) ['gov.contrib.conservatives.cb_hitc_household', 'gov.contrib.conservatives.pensioner_personal_allowance', 'gov.contrib.labour.private_school_vat', 'gov.economic_assumptions.yoy_growth.obr.private_pension_index', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.UNITED_KINGDOM', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.ENGLAND']. Variables: 13 ['attends_private_school', 'attends_private_school_random_draw', 'baseline_vat', 'full_rate_vat_consumption', 'full_rate_vat_expenditure_rate', 'in_social_housing']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -1172,7 +1172,7 @@ "measure_key": "ab2026__your_first_home_equity_loan", "title": "Your First Home: Introduce a new equity loan scheme to support around 70,000 first-time buyers from April 2027 (net income from developer fees)", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 37 (Spend, RAME); exchequer impact GBP m by FY: {'2026-27': '-15', '2027-28': '50', '2028-29': '95', '2029-30': '95', '2030-31': '95', '2031-32': '95'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 37 (Spend); exchequer impact GBP m by FY: {'2026-27': '-15', '2027-28': '50', '2028-29': '95', '2029-30': '95', '2030-31': '95', '2031-32': '95'}", "table_61_line": 37, "table_61_head": "Spend", "table_61_heading": "Hope for the future > Housing", @@ -1188,10 +1188,10 @@ "measure_key": "ab2026__defence_investment_plan_further_savings", "title": "Defence Investment Plan (DIP): Commitment to further savings made at announcement on 30 June 2026", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 40 (Spend, TDEL); exchequer impact GBP m by FY: {'2026-27': '-1755', '2027-28': '-1090', '2028-29': '-1010', '2029-30': '-905', '2030-31': '-940', '2031-32': '-970'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 40 (Spend); exchequer impact GBP m by FY: {'2026-27': '-1755', '2027-28': '-1090', '2028-29': '-1010', '2029-30': '-905', '2030-31': '-940', '2031-32': '-970'}", "table_61_line": 40, "table_61_head": "Spend", - "table_61_heading": "Investing in our national security", + "table_61_heading": "Hope for the future > Investing in our national security", "program": "public_spending", "computability": "not_expressible", "pe_reform_delta": null, @@ -1204,10 +1204,10 @@ "measure_key": "ab2026__defence_investment_plan_capital_reallocation", "title": "Defence Investment Plan: Reallocate capital spending to fully fund the Defence Investment Plan", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 41 (Spend, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '145', '2027-28': '2135', '2028-29': '1420', '2029-30': '1100', '2030-31': '1135', '2031-32': '1180'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 41 (Spend); exchequer impact GBP m by FY: {'2026-27': '145', '2027-28': '2135', '2028-29': '1420', '2029-30': '1100', '2030-31': '1135', '2031-32': '1180'}", "table_61_line": 41, "table_61_head": "Spend", - "table_61_heading": "Investing in our national security", + "table_61_heading": "Hope for the future > Investing in our national security", "program": "public_spending", "computability": "not_expressible", "pe_reform_delta": null, @@ -1220,10 +1220,10 @@ "measure_key": "ab2026__defence_admin_savings_reinvested", "title": "Defence: Reinvest defence and security administration savings into frontline defence and national security", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 42 (Spend, RDEL); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-25', '2028-29': '-70', '2029-30': '-95', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 42 (Spend); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-25', '2028-29': '-70', '2029-30': '-95', '2030-31': '0', '2031-32': '0'}", "table_61_line": 42, "table_61_head": "Spend", - "table_61_heading": "Investing in our national security", + "table_61_heading": "Hope for the future > Investing in our national security", "program": "public_spending", "computability": "not_expressible", "pe_reform_delta": null, @@ -1236,10 +1236,10 @@ "measure_key": "ab2026__ukraine_winter_air_defence", "title": "Ukraine: Emergency winter support for air defence", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 43 (Spend, TDEL); exchequer impact GBP m by FY: {'2026-27': '-500', '2027-28': '0', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 43 (Spend); exchequer impact GBP m by FY: {'2026-27': '-500', '2027-28': '0', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", "table_61_line": 43, "table_61_head": "Spend", - "table_61_heading": "Investing in our national security", + "table_61_heading": "Hope for the future > Investing in our national security", "program": "public_spending", "computability": "not_expressible", "pe_reform_delta": null, @@ -1252,10 +1252,10 @@ "measure_key": "ab2026__uk_export_finance_defence_capacity", "title": "UK Export Finance: Increase capacity to support UK defence exports from June 2026 (increased income from additional activity)", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 44 (Tax, RAME); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '200', '2028-29': '200', '2029-30': '200', '2030-31': '200', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 44 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '200', '2028-29': '200', '2029-30': '200', '2030-31': '200', '2031-32': '0'}", "table_61_line": 44, "table_61_head": "Tax", - "table_61_heading": "Investing in our national security", + "table_61_heading": "Hope for the future > Investing in our national security", "program": "financial_transactions", "computability": "not_expressible", "pe_reform_delta": null, @@ -1266,7 +1266,7 @@ }, { "measure_key": "ab2026__hmrc_tax_gap_package", - "title": "HMRC: Measures to close the tax gap4", + "title": "HMRC: Measures to close the tax gap", "reported_status": "announced", "source_note": "HM Treasury, Budget 2026 Table 6.1 line 45 (Tax); exchequer impact GBP m by FY: {'2026-27': '10', '2027-28': '100', '2028-29': '115', '2029-30': '1980', '2030-31': '2485', '2031-32': '3160'}", "table_61_line": 45, @@ -1284,7 +1284,7 @@ "measure_key": "ab2026__online_marketplace_liability_uk_businesses", "title": "Online Marketplace Liability: Extend to UK businesses with a minimum annual sales threshold per platform of £90,000 from 1 October 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 46 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '165', '2028-29': '320', '2029-30': '270', '2030-31': '250', '2031-32': '265'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 46 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '165', '2028-29': '320', '2029-30': '270', '2030-31': '250', '2031-32': '265'}", "table_61_line": 46, "table_61_head": "Tax", "table_61_heading": "Funding the Budget fairly and responsibly > Modernising the tax and welfare system", @@ -1298,7 +1298,7 @@ }, { "measure_key": "ab2026__dwp_fraud_and_error_prevention", - "title": "DWP Fraud and Error: Expand use of data and targeted controls to prevent fraud from entering the benefits system4", + "title": "DWP Fraud and Error: Expand use of data and targeted controls to prevent fraud from entering the benefits system", "reported_status": "announced", "source_note": "HM Treasury, Budget 2026 Table 6.1 line 47 (Spend); exchequer impact GBP m by FY: {'2026-27': '195', '2027-28': '-125', '2028-29': '60', '2029-30': '75', '2030-31': '155', '2031-32': '240'}", "table_61_line": 47, @@ -1332,7 +1332,7 @@ "measure_key": "ab2026__distributions_framework_modernised", "title": "Distributions: Modernise the distributions framework from 6 April 2028", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 49 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '55', '2029-30': '800', '2030-31': '775', '2031-32': '785'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 49 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '55', '2029-30': '800', '2030-31': '775', '2031-32': '785'}", "table_61_line": 49, "table_61_head": "Tax", "table_61_heading": "Funding the Budget fairly and responsibly > Modernising the tax and welfare system", @@ -1340,7 +1340,7 @@ "computability": "not_expressible", "pe_reform_delta": null, "why": "A tax on businesses (the distributions framework), out of the household model's scope: it changes what companies owe. The engine's corporate channel, gov.contrib.policyengine.budget.corporate_incident_tax_change ('Tax increase incident on owners of capital'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /distribution|dividend/. Parameters: 20 node(s) ['gov.hmrc.income_tax.allowances.dividend_allowance', 'gov.hmrc.income_tax.rates.dividends[0].threshold', 'gov.hmrc.income_tax.rates.dividends[0].rate', 'gov.hmrc.income_tax.rates.dividends[1].threshold', 'gov.hmrc.income_tax.rates.dividends[1].rate', 'gov.hmrc.income_tax.rates.dividends[2].threshold']. Variables: 6 ['dividend_income', 'received_allowances_dividend_income', 'dividend_income_tax', 'trade_loss_relief_against_savings_and_dividends', 'taxed_dividend_income', 'taxable_dividend_income']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /distribution|dividend/. Parameters: 20 node(s) ['gov.hmrc.income_tax.allowances.dividend_allowance', 'gov.hmrc.income_tax.rates.dividends[0].threshold', 'gov.hmrc.income_tax.rates.dividends[0].rate', 'gov.hmrc.income_tax.rates.dividends[1].threshold', 'gov.hmrc.income_tax.rates.dividends[1].rate', 'gov.hmrc.income_tax.rates.dividends[2].threshold']. Variables: 6 ['dividend_income', 'dividend_income_tax', 'received_allowances_dividend_income', 'taxable_dividend_income', 'taxed_dividend_income', 'trade_loss_relief_against_savings_and_dividends']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -1348,7 +1348,7 @@ "measure_key": "ab2026__advance_corporation_tax_abolished", "title": "Advance Corporation Tax (ACT): Abolish Advance Corporation Tax from 1 April 2031", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 50 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '0', '2029-30': '0', '2030-31': '5', '2031-32': '15'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 50 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '0', '2029-30': '0', '2030-31': '5', '2031-32': '15'}", "table_61_line": 50, "table_61_head": "Tax", "table_61_heading": "Funding the Budget fairly and responsibly > Modernising the tax and welfare system", @@ -1364,7 +1364,7 @@ "measure_key": "ab2026__vat_reverse_charge_overseas_services_restricted", "title": "Cross-Border Services: Restrict the reverse charge on supplies of services from overseas to VAT-registered businesses from 1 April 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 51 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '70', '2028-29': '75', '2029-30': '75', '2030-31': '75', '2031-32': '75'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 51 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '70', '2028-29': '75', '2029-30': '75', '2030-31': '75', '2031-32': '75'}", "table_61_line": 51, "table_61_head": "Tax", "table_61_heading": "Funding the Budget fairly and responsibly > Modernising the tax and welfare system", @@ -1372,7 +1372,7 @@ "computability": "not_expressible", "pe_reform_delta": null, "why": "An indirect tax or duty collected from businesses (the VAT reverse charge on imported services), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /reverse_charge|vat/. Parameters: 64 node(s) ['gov.contrib.conservatives.cb_hitc_household', 'gov.contrib.conservatives.pensioner_personal_allowance', 'gov.contrib.labour.private_school_vat', 'gov.economic_assumptions.yoy_growth.obr.private_pension_index', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.UNITED_KINGDOM', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.ENGLAND']. Variables: 12 ['private_pension_wealth', 'private_pension_income', 'private_transfer_income', 'private_school_vat', 'attends_private_school_random_draw', 'attends_private_school']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /reverse_charge|vat/. Parameters: 64 node(s) ['gov.contrib.conservatives.cb_hitc_household', 'gov.contrib.conservatives.pensioner_personal_allowance', 'gov.contrib.labour.private_school_vat', 'gov.economic_assumptions.yoy_growth.obr.private_pension_index', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.UNITED_KINGDOM', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.ENGLAND']. Variables: 12 ['attends_private_school', 'attends_private_school_random_draw', 'baseline_vat', 'full_rate_vat_consumption', 'full_rate_vat_expenditure_rate', 'private_pension_income']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -1380,7 +1380,7 @@ "measure_key": "ab2026__foreign_sovereign_investments_taxation", "title": "Modernising Taxation of Foreign Sovereign Investments: Introduce a new statutory regime for the direct taxation of foreign sovereigns from 1 January 2029", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 54 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '110', '2028-29': '495', '2029-30': '690', '2030-31': '675', '2031-32': '655'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 54 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '110', '2028-29': '495', '2029-30': '690', '2030-31': '675', '2031-32': '655'}", "table_61_line": 54, "table_61_head": "Tax", "table_61_heading": "Funding the Budget fairly and responsibly > Modernising the tax and welfare system", @@ -1396,10 +1396,10 @@ "measure_key": "ab2026__bank_surcharge_6pc_temporary", "title": "Bank Surcharge: Temporarily increase the Bank Corporation Tax Surcharge rate from 3% to 6%, to fund breathing space measures for households, for 15 months from 1 January 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 55 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '305', '2027-28': '950', '2028-29': '220', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 55 (Tax); exchequer impact GBP m by FY: {'2026-27': '305', '2027-28': '950', '2028-29': '220', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", "table_61_line": 55, "table_61_head": "Tax", - "table_61_heading": "Making fair choices on revenue raising", + "table_61_heading": "Funding the Budget fairly and responsibly > Making fair choices on revenue raising", "program": "business_tax", "computability": "not_expressible", "pe_reform_delta": null, @@ -1415,12 +1415,12 @@ "source_note": "HM Treasury, Budget 2026 Table 6.1 line 56 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '135', '2028-29': '150', '2029-30': '145', '2030-31': '145', '2031-32': '145'}", "table_61_line": 56, "table_61_head": "Tax", - "table_61_heading": "Making fair choices on revenue raising", + "table_61_heading": "Funding the Budget fairly and responsibly > Making fair choices on revenue raising", "program": "indirect_tax", "computability": "not_expressible", "pe_reform_delta": null, "why": "An indirect tax or duty collected from businesses (Machine Games Duty), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /gambling|machine_games|gaming/. Parameters: 23 node(s) ['gov.hmrc.gambling_duty.bingo.rate', 'gov.hmrc.gambling_duty.gaming.normal_six_month_bands[0].threshold', 'gov.hmrc.gambling_duty.gaming.normal_six_month_bands[0].rate', 'gov.hmrc.gambling_duty.gaming.normal_six_month_bands[1].threshold', 'gov.hmrc.gambling_duty.gaming.normal_six_month_bands[1].rate', 'gov.hmrc.gambling_duty.gaming.normal_six_month_bands[2].threshold']. Variables: 8 ['remote_gaming_net_losses', 'machine_games_higher_rate_net_takings', 'machine_games_lower_rate_net_takings', 'machine_games_standard_rate_net_takings', 'gaming_duty', 'remote_gaming_duty']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /gambling|machine_games|gaming/. Parameters: 23 node(s) ['gov.hmrc.gambling_duty.bingo.rate', 'gov.hmrc.gambling_duty.gaming.normal_six_month_bands[0].threshold', 'gov.hmrc.gambling_duty.gaming.normal_six_month_bands[0].rate', 'gov.hmrc.gambling_duty.gaming.normal_six_month_bands[1].threshold', 'gov.hmrc.gambling_duty.gaming.normal_six_month_bands[1].rate', 'gov.hmrc.gambling_duty.gaming.normal_six_month_bands[2].threshold']. Variables: 8 ['gambling_duty', 'gaming_duty', 'machine_games_duty', 'machine_games_higher_rate_net_takings', 'machine_games_lower_rate_net_takings', 'machine_games_standard_rate_net_takings']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -1428,15 +1428,15 @@ "measure_key": "ab2026__business_rates_vape_shops_and_agcs_reliefs_removed", "title": "Business Rates: Remove reliefs for vape shops and Adult Gaming Centres from 1 April 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 57 (Tax, Tax/TME); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '10', '2028-29': '10', '2029-30': '10', '2030-31': '10', '2031-32': '10'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 57 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '10', '2028-29': '10', '2029-30': '10', '2030-31': '10', '2031-32': '10'}", "table_61_line": 57, "table_61_head": "Tax", - "table_61_heading": "Making fair choices on revenue raising", + "table_61_heading": "Funding the Budget fairly and responsibly > Making fair choices on revenue raising", "program": "business_rates", "computability": "not_expressible", "pe_reform_delta": null, "why": "A business rates measure (business rates reliefs for vape shops and adult gaming centres), out of the household model's scope. The engine carries business rates only as a household incidence, business_rates ('Total incidence from exposure to business rates via corporate shareholdings'), and its business rates parameters are revenue statistics (gov.hmrc.business_rates.statistics.revenue.*): it has no properties, reliefs or thresholds for the measure to change, so its £ change would have to be supplied as an input, restating the costing.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /business_rates|vape/. Parameters: 4 node(s) ['gov.hmrc.business_rates.statistics.revenue.ENGLAND', 'gov.hmrc.business_rates.statistics.revenue.NORTHERN_IRELAND', 'gov.hmrc.business_rates.statistics.revenue.SCOTLAND', 'gov.hmrc.business_rates.statistics.revenue.WALES']. Variables: 2 ['business_rates', 'baseline_business_rates']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /business_rates|vape/. Parameters: 4 node(s) ['gov.hmrc.business_rates.statistics.revenue.ENGLAND', 'gov.hmrc.business_rates.statistics.revenue.NORTHERN_IRELAND', 'gov.hmrc.business_rates.statistics.revenue.SCOTLAND', 'gov.hmrc.business_rates.statistics.revenue.WALES']. Variables: 2 ['baseline_business_rates', 'business_rates']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -1444,10 +1444,10 @@ "measure_key": "ab2026__writing_down_allowance_12pc_and_fya_50pc", "title": "Writing-down Allowances: Reduce the main rate writing-down allowance to 12% from April 2027 and increase the main rate first-year allowance to 50% from 1 January 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 58 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '125', '2027-28': '765', '2028-29': '1040', '2029-30': '1075', '2030-31': '1100', '2031-32': '1120'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 58 (Tax); exchequer impact GBP m by FY: {'2026-27': '125', '2027-28': '765', '2028-29': '1040', '2029-30': '1075', '2030-31': '1100', '2031-32': '1120'}", "table_61_line": 58, "table_61_head": "Tax", - "table_61_heading": "Making fair choices on revenue raising", + "table_61_heading": "Funding the Budget fairly and responsibly > Making fair choices on revenue raising", "program": "business_tax", "computability": "not_expressible", "pe_reform_delta": null, @@ -1460,15 +1460,15 @@ "measure_key": "ab2026__private_hire_operators_vat_full_fare", "title": "Private Hire Vehicle Operators: Deem larger operators with annual fare turnover over £50 million liable for VAT on the full value of their fares from 2 January 2028", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 59 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '95', '2028-29': '385', '2029-30': '395', '2030-31': '410', '2031-32': '425'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 59 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '95', '2028-29': '385', '2029-30': '395', '2030-31': '410', '2031-32': '425'}", "table_61_line": 59, "table_61_head": "Tax", - "table_61_heading": "Making fair choices on revenue raising", + "table_61_heading": "Funding the Budget fairly and responsibly > Making fair choices on revenue raising", "program": "indirect_tax", "computability": "not_expressible", "pe_reform_delta": null, "why": "An indirect tax or duty collected from businesses (VAT on private hire operators' fares), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /private_hire|taxi|vat/. Parameters: 64 node(s) ['gov.contrib.conservatives.cb_hitc_household', 'gov.contrib.conservatives.pensioner_personal_allowance', 'gov.contrib.labour.private_school_vat', 'gov.economic_assumptions.yoy_growth.obr.private_pension_index', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.UNITED_KINGDOM', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.ENGLAND']. Variables: 12 ['private_pension_wealth', 'private_pension_income', 'private_transfer_income', 'private_school_vat', 'attends_private_school_random_draw', 'attends_private_school']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /private_hire|taxi|vat/. Parameters: 64 node(s) ['gov.contrib.conservatives.cb_hitc_household', 'gov.contrib.conservatives.pensioner_personal_allowance', 'gov.contrib.labour.private_school_vat', 'gov.economic_assumptions.yoy_growth.obr.private_pension_index', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.UNITED_KINGDOM', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.ENGLAND']. Variables: 12 ['attends_private_school', 'attends_private_school_random_draw', 'baseline_vat', 'full_rate_vat_consumption', 'full_rate_vat_expenditure_rate', 'private_pension_income']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -1476,10 +1476,10 @@ "measure_key": "ab2026__civil_service_admin_budgets_cut_20pc", "title": "Future Civil Service: Reduce administration budgets by 20% by 2029-30 compared to 16% reduction in Spending Review 2025 plans", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 63 (Spend, RDEL); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '185', '2028-29': '405', '2029-30': '595', '2030-31': '615', '2031-32': '635'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 63 (Spend); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '185', '2028-29': '405', '2029-30': '595', '2030-31': '615', '2031-32': '635'}", "table_61_line": 63, "table_61_head": "Spend", - "table_61_heading": "Transforming the Civil Service", + "table_61_heading": "Funding the Budget fairly and responsibly > Transforming the Civil Service", "program": "public_spending", "computability": "not_expressible", "pe_reform_delta": null, @@ -1492,10 +1492,10 @@ "measure_key": "ab2026__civil_service_ai_fund", "title": "Future Civil Service: Reinvest part of the administration budget savings in a new AI Fund to reduce administration burdens and improve outcomes", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 64 (Spend, RDEL); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-25', '2028-29': '-25', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 64 (Spend); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-25', '2028-29': '-25', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", "table_61_line": 64, "table_61_head": "Spend", - "table_61_heading": "Transforming the Civil Service", + "table_61_heading": "Funding the Budget fairly and responsibly > Transforming the Civil Service", "program": "public_spending", "computability": "not_expressible", "pe_reform_delta": null, @@ -1508,7 +1508,7 @@ "measure_key": "ab2026__hgv_ved_1_for_12_months", "title": "Heavy Goods Vehicles: Reduce Vehicle Excise Duty to £1 for 12 months from 1 July 2026", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 65 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '-135', '2027-28': '-50', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 65 (Tax); exchequer impact GBP m by FY: {'2026-27': '-135', '2027-28': '-50', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", "table_61_line": 65, "table_61_head": "Tax", "table_61_heading": "Other policy measures including pre-announced policies > Pre-announced breathing space measures and associated funding", @@ -1524,7 +1524,7 @@ "measure_key": "ab2026__foreign_permanent_establishment_exemption_mandatory", "title": "Corporation Tax: Make the foreign permanent establishment exemption mandatory, preventing their losses from reducing UK corporation tax receipts for accounting periods beginning on or after 1 January 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 68 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '120', '2027-28': '480', '2028-29': '495', '2029-30': '450', '2030-31': '460', '2031-32': '470'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 68 (Tax); exchequer impact GBP m by FY: {'2026-27': '120', '2027-28': '480', '2028-29': '495', '2029-30': '450', '2030-31': '460', '2031-32': '470'}", "table_61_line": 68, "table_61_head": "Tax", "table_61_heading": "Other policy measures including pre-announced policies > Pre-announced breathing space measures and associated funding", @@ -1540,7 +1540,7 @@ "measure_key": "ab2026__electricity_generator_levy_55pc", "title": "Electricity Generator Levy: Increase the rate from 45% to 55% until 31 March 2028", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 69 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '60', '2027-28': '35', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 69 (Tax); exchequer impact GBP m by FY: {'2026-27': '60', '2027-28': '35', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", "table_61_line": 69, "table_61_head": "Tax", "table_61_heading": "Other policy measures including pre-announced policies > Pre-announced breathing space measures and associated funding", @@ -1556,10 +1556,10 @@ "measure_key": "ab2026__british_industrial_competitiveness_scheme", "title": "Industrial Electricity: Fund the British Industrial Competitiveness Scheme starting from 1 April 2027, including a backdated payment for eligible businesses", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 70 (Spend, Tax/TME); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-1235', '2028-29': '-65', '2029-30': '-135', '2030-31': '-420', '2031-32': '-465'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 70 (Spend); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-1235', '2028-29': '-65', '2029-30': '-135', '2030-31': '-420', '2031-32': '-465'}", "table_61_line": 70, "table_61_head": "Spend", - "table_61_heading": "Pre-announced growth measures", + "table_61_heading": "Other policy measures including pre-announced policies > Pre-announced growth measures", "program": "public_spending", "computability": "not_expressible", "pe_reform_delta": null, @@ -1572,15 +1572,15 @@ "measure_key": "ab2026__carbon_price_support_removed", "title": "Carbon Price Support (CPS): Remove the CPS rates to reduce energy bills from 1 April 2028", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 71 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '-165', '2029-30': '-45', '2030-31': '-15', '2031-32': '-10'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 71 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '-165', '2029-30': '-45', '2030-31': '-15', '2031-32': '-10'}", "table_61_line": 71, "table_61_head": "Tax", - "table_61_heading": "Pre-announced growth measures", + "table_61_heading": "Other policy measures including pre-announced policies > Pre-announced growth measures", "program": "indirect_tax", "computability": "not_expressible", "pe_reform_delta": null, "why": "An indirect tax or duty collected from businesses (Carbon Price Support rates), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /carbon_price|carbon/. Parameters: 37 node(s) ['gov.contrib.ubi_center.carbon_tax.consumer_incidence', 'gov.contrib.ubi_center.carbon_tax.rate', 'gov.economic_assumptions.yoy_growth.carbon_intensity', 'gov.economic_assumptions.indices.carbon_intensity', 'household.consumption.carbon.consumption', 'household.consumption.carbon.emissions.food_and_non_alcoholic_beverages']. Variables: 2 ['carbon_tax', 'carbon_consumption']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /carbon_price|carbon/. Parameters: 37 node(s) ['gov.contrib.ubi_center.carbon_tax.consumer_incidence', 'gov.contrib.ubi_center.carbon_tax.rate', 'gov.economic_assumptions.yoy_growth.carbon_intensity', 'gov.economic_assumptions.indices.carbon_intensity', 'household.consumption.carbon.consumption', 'household.consumption.carbon.emissions.food_and_non_alcoholic_beverages']. Variables: 2 ['carbon_consumption', 'carbon_tax']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -1588,10 +1588,10 @@ "measure_key": "ab2026__youth_guarantee_and_apprenticeships_spring_2026", "title": "Spring 2026 Youth Employment and Skills Package: Expand Youth Guarantee and apprenticeship support", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 72 (Spend, TME); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-290', '2028-29': '-375', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 72 (Spend); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-290', '2028-29': '-375', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", "table_61_line": 72, "table_61_head": "Spend", - "table_61_heading": "Pre-announced growth measures", + "table_61_heading": "Other policy measures including pre-announced policies > Pre-announced growth measures", "program": "public_spending", "computability": "not_expressible", "pe_reform_delta": null, @@ -1604,10 +1604,10 @@ "measure_key": "ab2026__union_learning_fund", "title": "Skills: Re-establish the Union Learning Fund in England", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 73 (Spend, RDEL); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-20', '2028-29': '-20', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 73 (Spend); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-20', '2028-29': '-20', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", "table_61_line": 73, "table_61_head": "Spend", - "table_61_heading": "Pre-announced growth measures", + "table_61_heading": "Other policy measures including pre-announced policies > Pre-announced growth measures", "program": "public_spending", "computability": "not_expressible", "pe_reform_delta": null, @@ -1623,12 +1623,12 @@ "source_note": "HM Treasury, Budget 2026 Table 6.1 line 74 (Spend); exchequer impact GBP m by FY: {'2026-27': '-155', '2027-28': '-460', '2028-29': '-530', '2029-30': '-330', '2030-31': '-355', '2031-32': '-535'}", "table_61_line": 74, "table_61_head": "Spend", - "table_61_heading": "Pre-announced growth measures", + "table_61_heading": "Other policy measures including pre-announced policies > Pre-announced growth measures", "program": "financial_transactions", "computability": "not_expressible", "pe_reform_delta": null, "why": "Lending, guarantees, investment or a financial transaction (investment for growth): no household tax or benefit rule changes.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /investment|growth/. Parameters: 45 node(s) ['gov.economic_assumptions.yoy_growth.obr.rpi', 'gov.economic_assumptions.yoy_growth.obr.average_earnings', 'gov.economic_assumptions.yoy_growth.obr.consumer_price_index', 'gov.economic_assumptions.yoy_growth.obr.road_fuel_volume', 'gov.economic_assumptions.yoy_growth.obr.petrol_spending_litre_proxy', 'gov.economic_assumptions.yoy_growth.obr.diesel_spending_litre_proxy']. Variables: 3 ['unit_and_investment_trusts', 'other_investment_income', 'charitable_investment_gifts']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /investment|growth/. Parameters: 45 node(s) ['gov.economic_assumptions.yoy_growth.obr.rpi', 'gov.economic_assumptions.yoy_growth.obr.average_earnings', 'gov.economic_assumptions.yoy_growth.obr.consumer_price_index', 'gov.economic_assumptions.yoy_growth.obr.road_fuel_volume', 'gov.economic_assumptions.yoy_growth.obr.petrol_spending_litre_proxy', 'gov.economic_assumptions.yoy_growth.obr.diesel_spending_litre_proxy']. Variables: 3 ['charitable_investment_gifts', 'other_investment_income', 'unit_and_investment_trusts']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -1636,15 +1636,15 @@ "measure_key": "ab2026__growth_guarantee_scheme_expansion", "title": "British Business Bank: Expand the Growth Guarantee Scheme to support £6.5 billion of additional SME lending across this Parliament", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 75 (Spend, TME); exchequer impact GBP m by FY: {'2026-27': '-105', '2027-28': '-145', '2028-29': '-180', '2029-30': '-170', '2030-31': '-160', '2031-32': '-155'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 75 (Spend); exchequer impact GBP m by FY: {'2026-27': '-105', '2027-28': '-145', '2028-29': '-180', '2029-30': '-170', '2030-31': '-160', '2031-32': '-155'}", "table_61_line": 75, "table_61_head": "Spend", - "table_61_heading": "Pre-announced growth measures", + "table_61_heading": "Other policy measures including pre-announced policies > Pre-announced growth measures", "program": "financial_transactions", "computability": "not_expressible", "pe_reform_delta": null, "why": "Lending, guarantees, investment or a financial transaction (the Growth Guarantee Scheme): no household tax or benefit rule changes.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /guarantee/. Parameters: 25 node(s) ['gov.dwp.pension_credit.guarantee_credit.additions', 'gov.dwp.pension_credit.guarantee_credit.carer.addition', 'gov.dwp.pension_credit.guarantee_credit.child.addition', 'gov.dwp.pension_credit.guarantee_credit.child.disability.addition', 'gov.dwp.pension_credit.guarantee_credit.child.disability.eligibility', 'gov.dwp.pension_credit.guarantee_credit.child.disability.severe.addition']. Variables: 9 ['is_guarantee_credit_eligible', 'guarantee_credit', 'in_receipt_of_guarantee_credit', 'standard_minimum_guarantee', 'minimum_guarantee', 'child_minimum_guarantee_addition']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /guarantee/. Parameters: 25 node(s) ['gov.dwp.pension_credit.guarantee_credit.additions', 'gov.dwp.pension_credit.guarantee_credit.carer.addition', 'gov.dwp.pension_credit.guarantee_credit.child.addition', 'gov.dwp.pension_credit.guarantee_credit.child.disability.addition', 'gov.dwp.pension_credit.guarantee_credit.child.disability.eligibility', 'gov.dwp.pension_credit.guarantee_credit.child.disability.severe.addition']. Variables: 9 ['additional_minimum_guarantee', 'carer_minimum_guarantee_addition', 'child_minimum_guarantee_addition', 'guarantee_credit', 'in_receipt_of_guarantee_credit', 'is_guarantee_credit_eligible']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -1652,15 +1652,15 @@ "measure_key": "ab2026__leeds_city_fund", "title": "Leeds City Fund: Establish the Leeds City Fund with 100% retention of business rates growth", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 76 (Tax, Tax/TME); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '-25', '2029-30': '-25', '2030-31': '-25', '2031-32': '-30'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 76 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '-25', '2029-30': '-25', '2030-31': '-25', '2031-32': '-30'}", "table_61_line": 76, "table_61_head": "Tax", - "table_61_heading": "Pre-announced growth measures", + "table_61_heading": "Other policy measures including pre-announced policies > Pre-announced growth measures", "program": "business_rates", "computability": "not_expressible", "pe_reform_delta": null, "why": "A business rates measure (business rates retention in Leeds), out of the household model's scope. The engine carries business rates only as a household incidence, business_rates ('Total incidence from exposure to business rates via corporate shareholdings'), and its business rates parameters are revenue statistics (gov.hmrc.business_rates.statistics.revenue.*): it has no properties, reliefs or thresholds for the measure to change, so its £ change would have to be supplied as an input, restating the costing.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /business_rates|retention/. Parameters: 4 node(s) ['gov.hmrc.business_rates.statistics.revenue.ENGLAND', 'gov.hmrc.business_rates.statistics.revenue.NORTHERN_IRELAND', 'gov.hmrc.business_rates.statistics.revenue.SCOTLAND', 'gov.hmrc.business_rates.statistics.revenue.WALES']. Variables: 3 ['business_rates', 'baseline_business_rates', 'relative_capital_gains_retention_rate_change']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /business_rates|retention/. Parameters: 4 node(s) ['gov.hmrc.business_rates.statistics.revenue.ENGLAND', 'gov.hmrc.business_rates.statistics.revenue.NORTHERN_IRELAND', 'gov.hmrc.business_rates.statistics.revenue.SCOTLAND', 'gov.hmrc.business_rates.statistics.revenue.WALES']. Variables: 3 ['baseline_business_rates', 'business_rates', 'relative_capital_gains_retention_rate_change']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -1668,15 +1668,15 @@ "measure_key": "ab2026__uk_india_ceta_and_double_contributions_convention", "title": "UK-India: Bring into force the Comprehensive Economic and Trade Agreement and Double Contributions Convention from 15 July 2026", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 77 (Tax, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '-175', '2027-28': '-285', '2028-29': '-330', '2029-30': '-360', '2030-31': '-385', '2031-32': '-400'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 77 (Tax); exchequer impact GBP m by FY: {'2026-27': '-175', '2027-28': '-285', '2028-29': '-330', '2029-30': '-360', '2030-31': '-385', '2031-32': '-400'}", "table_61_line": 77, "table_61_head": "Tax", - "table_61_heading": "Pre-announced growth measures", + "table_61_heading": "Other policy measures including pre-announced policies > Pre-announced growth measures", "program": "indirect_tax", "computability": "not_expressible", "pe_reform_delta": null, "why": "An indirect tax or duty collected from businesses (tariffs and contributions under the UK-India agreement), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /tariff|customs|double_contribution/. Parameters: 26 node(s) ['gov.dwp.ESA.income.capital.tariff_income.amount', 'gov.dwp.ESA.income.capital.tariff_income.step', 'gov.dwp.ESA.income.capital.tariff_income.threshold', 'gov.dwp.JSA.income.capital.tariff_income.amount', 'gov.dwp.JSA.income.capital.tariff_income.step', 'gov.dwp.JSA.income.capital.tariff_income.threshold']. Variables: 5 ['jsa_income_tariff_income', 'esa_income_tariff_income', 'income_support_tariff_income', 'housing_benefit_tariff_income', 'uc_tariff_income']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /tariff|customs|double_contribution/. Parameters: 26 node(s) ['gov.dwp.ESA.income.capital.tariff_income.amount', 'gov.dwp.ESA.income.capital.tariff_income.step', 'gov.dwp.ESA.income.capital.tariff_income.threshold', 'gov.dwp.JSA.income.capital.tariff_income.amount', 'gov.dwp.JSA.income.capital.tariff_income.step', 'gov.dwp.JSA.income.capital.tariff_income.threshold']. Variables: 5 ['esa_income_tariff_income', 'housing_benefit_tariff_income', 'income_support_tariff_income', 'jsa_income_tariff_income', 'uc_tariff_income']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -1684,15 +1684,15 @@ "measure_key": "ab2026__steel_tariff_and_quotas", "title": "Steel Tariff: Change tariff rate and quotas on imported steel from 1 July 2026", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 78 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '85', '2027-28': '110', '2028-29': '120', '2029-30': '120', '2030-31': '120', '2031-32': '115'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 78 (Tax); exchequer impact GBP m by FY: {'2026-27': '85', '2027-28': '110', '2028-29': '120', '2029-30': '120', '2030-31': '120', '2031-32': '115'}", "table_61_line": 78, "table_61_head": "Tax", - "table_61_heading": "Pre-announced growth measures", + "table_61_heading": "Other policy measures including pre-announced policies > Pre-announced growth measures", "program": "indirect_tax", "computability": "not_expressible", "pe_reform_delta": null, "why": "An indirect tax or duty collected from businesses (steel tariffs and quotas), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /tariff|steel/. Parameters: 26 node(s) ['gov.dwp.ESA.income.capital.tariff_income.amount', 'gov.dwp.ESA.income.capital.tariff_income.step', 'gov.dwp.ESA.income.capital.tariff_income.threshold', 'gov.dwp.JSA.income.capital.tariff_income.amount', 'gov.dwp.JSA.income.capital.tariff_income.step', 'gov.dwp.JSA.income.capital.tariff_income.threshold']. Variables: 5 ['jsa_income_tariff_income', 'esa_income_tariff_income', 'income_support_tariff_income', 'housing_benefit_tariff_income', 'uc_tariff_income']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /tariff|steel/. Parameters: 26 node(s) ['gov.dwp.ESA.income.capital.tariff_income.amount', 'gov.dwp.ESA.income.capital.tariff_income.step', 'gov.dwp.ESA.income.capital.tariff_income.threshold', 'gov.dwp.JSA.income.capital.tariff_income.amount', 'gov.dwp.JSA.income.capital.tariff_income.step', 'gov.dwp.JSA.income.capital.tariff_income.threshold']. Variables: 5 ['esa_income_tariff_income', 'housing_benefit_tariff_income', 'income_support_tariff_income', 'jsa_income_tariff_income', 'uc_tariff_income']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -1700,10 +1700,10 @@ "measure_key": "ab2026__securities_transfer_tax", "title": "Stamp Taxes on Shares: Replace Stamp Duty and Stamp Duty Reserve Tax with the Securities Transfer Tax from late 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 79 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-15', '2028-29': '-30', '2029-30': '-30', '2030-31': '-30', '2031-32': '-30'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 79 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-15', '2028-29': '-30', '2029-30': '-30', '2030-31': '-30', '2031-32': '-30'}", "table_61_line": 79, "table_61_head": "Tax", - "table_61_heading": "Pre-announced growth measures", + "table_61_heading": "Other policy measures including pre-announced policies > Pre-announced growth measures", "program": "business_tax", "computability": "not_expressible", "pe_reform_delta": null, @@ -1716,15 +1716,15 @@ "measure_key": "ab2026__business_rates_ev_charging_relief", "title": "Business Rates: Introduce 100% relief for eligible Electric Vehicle Charging Points (EVCPs) and electric vehicle-only forecourts for ten years from 1 April 2026", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 80 (Tax, Tax/TME); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-15', '2028-29': '-5', '2029-30': '-10', '2030-31': '-15', '2031-32': '-15'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 80 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-15', '2028-29': '-5', '2029-30': '-10', '2030-31': '-15', '2031-32': '-15'}", "table_61_line": 80, "table_61_head": "Tax", - "table_61_heading": "Pre-announced growth measures", + "table_61_heading": "Other policy measures including pre-announced policies > Pre-announced growth measures", "program": "business_rates", "computability": "not_expressible", "pe_reform_delta": null, "why": "A business rates measure (business rates relief for EV charge points), out of the household model's scope. The engine carries business rates only as a household incidence, business_rates ('Total incidence from exposure to business rates via corporate shareholdings'), and its business rates parameters are revenue statistics (gov.hmrc.business_rates.statistics.revenue.*): it has no properties, reliefs or thresholds for the measure to change, so its £ change would have to be supplied as an input, restating the costing.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /business_rates|charg/. Parameters: 25 node(s) ['gov.dwp.housing_benefit.ineligible_charges.meals.at_least_three_a_day.aged_16_or_over', 'gov.dwp.housing_benefit.ineligible_charges.meals.at_least_three_a_day.under_16', 'gov.dwp.housing_benefit.ineligible_charges.meals.breakfast_only', 'gov.dwp.housing_benefit.ineligible_charges.meals.fewer_than_three_a_day.aged_16_or_over', 'gov.dwp.housing_benefit.ineligible_charges.meals.fewer_than_three_a_day.under_16', 'gov.hmrc.business_rates.statistics.revenue.ENGLAND']. Variables: 8 ['water_and_sewerage_charges', 'housing_service_charges', 'child_benefit_less_tax_charge', 'business_rates', 'high_value_council_tax_surcharge', 'baseline_business_rates']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /business_rates|charg/. Parameters: 25 node(s) ['gov.dwp.housing_benefit.ineligible_charges.meals.at_least_three_a_day.aged_16_or_over', 'gov.dwp.housing_benefit.ineligible_charges.meals.at_least_three_a_day.under_16', 'gov.dwp.housing_benefit.ineligible_charges.meals.breakfast_only', 'gov.dwp.housing_benefit.ineligible_charges.meals.fewer_than_three_a_day.aged_16_or_over', 'gov.dwp.housing_benefit.ineligible_charges.meals.fewer_than_three_a_day.under_16', 'gov.hmrc.business_rates.statistics.revenue.ENGLAND']. Variables: 8 ['baseline_business_rates', 'business_rates', 'child_benefit_less_tax_charge', 'high_value_council_tax_surcharge', 'housing_service_charges', 'income_tax_pre_charges']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -1732,10 +1732,10 @@ "measure_key": "ab2026__misogyny_prevention_programmes", "title": "Tackling Misogyny: Invest in grassroots misogyny prevention programmes", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 81 (Spend, RDEL); exchequer impact GBP m by FY: {'2026-27': '-5', '2027-28': '-5', '2028-29': '-5', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 81 (Spend); exchequer impact GBP m by FY: {'2026-27': '-5', '2027-28': '-5', '2028-29': '-5', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", "table_61_line": 81, "table_61_head": "Spend", - "table_61_heading": "Priority Public Services Funding", + "table_61_heading": "Other policy measures including pre-announced policies > Priority Public Services Funding", "program": "public_spending", "computability": "not_expressible", "pe_reform_delta": null, @@ -1748,10 +1748,10 @@ "measure_key": "ab2026__infected_blood_compensation_additional_report", "title": "Infected Blood Compensation Scheme: Implement the Inquiry's Additional Report recommendations from 24 July 2026", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 82 (Spend, CAME); exchequer impact GBP m by FY: {'2026-27': '-360', '2027-28': '-355', '2028-29': '-265', '2029-30': '-90', '2030-31': '-75', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 82 (Spend); exchequer impact GBP m by FY: {'2026-27': '-360', '2027-28': '-355', '2028-29': '-265', '2029-30': '-90', '2030-31': '-75', '2031-32': '0'}", "table_61_line": 82, "table_61_head": "Spend", - "table_61_heading": "Priority Public Services Funding", + "table_61_heading": "Other policy measures including pre-announced policies > Priority Public Services Funding", "program": "public_spending", "computability": "not_expressible", "pe_reform_delta": null, @@ -1764,10 +1764,10 @@ "measure_key": "ab2026__waste_crime_landfill_tax_clearances", "title": "Waste Crime: Fund Landfill Tax costs for six priority illegal waste site clearances until 31 March 2029 and a three-year pilot to fund Local Authorities' Landfill Tax liabilities arising from illegal waste site clearances from 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 84 (Spend, SUBTOTAL); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-5', '2028-29': '-5', '2029-30': '-5', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 84 (Spend); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-5', '2028-29': '-5', '2029-30': '-5', '2030-31': '0', '2031-32': '0'}", "table_61_line": 84, "table_61_head": "Spend", - "table_61_heading": "Priority Public Services Funding", + "table_61_heading": "Other policy measures including pre-announced policies > Priority Public Services Funding", "program": "public_spending", "computability": "not_expressible", "pe_reform_delta": null, @@ -1783,12 +1783,12 @@ "source_note": "HM Treasury, Budget 2026 Table 6.1 line 85 (Spend); exchequer impact GBP m by FY: {'2026-27': '-355', '2027-28': '-175', '2028-29': '-140', '2029-30': '-5', '2030-31': '0', '2031-32': '-15'}", "table_61_line": 85, "table_61_head": "Spend", - "table_61_heading": "Priority Public Services Funding", + "table_61_heading": "Other policy measures including pre-announced policies > Priority Public Services Funding", "program": "public_spending", "computability": "not_expressible", "pe_reform_delta": null, - "why": "Public spending (the NHS and the Violence against Women and Girls Strategy), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /nhs|violence/. Parameters: 1 node(s) ['gov.contrib.policyengine.budget.nhs']. Variables: 7 ['nhs_budget_change', 'nhs_spending', 'nhs_outpatient_spending', 'nhs_a_and_e_spending', 'nhs_admitted_patient_spending', 'travel_grant_receives_means_tested_nhs_bursary']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "why": "Public spending (the NHS and the Violence against Women and Girls Strategy), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.nhs ('Increase in NHS spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /nhs|violence/. Parameters: 1 node(s) ['gov.contrib.policyengine.budget.nhs']. Variables: 7 ['childcare_grant_receives_nhs_childcare_support', 'nhs_a_and_e_spending', 'nhs_admitted_patient_spending', 'nhs_budget_change', 'nhs_outpatient_spending', 'nhs_spending']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -1796,10 +1796,10 @@ "measure_key": "ab2026__supported_housing_licensing", "title": "Supported Housing: Introduce a licensing regime to enforce national standards for supported housing from 2027-28", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 87 (Spend, RAME); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-55', '2028-29': '-70', '2029-30': '-30', '2030-31': '-30', '2031-32': '-30'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 87 (Spend); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-55', '2028-29': '-70', '2029-30': '-30', '2030-31': '-30', '2031-32': '-30'}", "table_61_line": 87, "table_61_head": "Spend", - "table_61_heading": "Other measures", + "table_61_heading": "Other policy measures including pre-announced policies > Other measures", "program": "regulation", "computability": "not_expressible", "pe_reform_delta": null, @@ -1812,15 +1812,15 @@ "measure_key": "ab2026__vat_zero_rate_donated_medicines", "title": "Donated Medicines: Introduce a zero rate of VAT for the donation of medicines under qualifying circumstances from 23 June 2026", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 88 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '-15', '2027-28': '-20', '2028-29': '-20', '2029-30': '-25', '2030-31': '-25', '2031-32': '-25'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 88 (Tax); exchequer impact GBP m by FY: {'2026-27': '-15', '2027-28': '-20', '2028-29': '-20', '2029-30': '-25', '2030-31': '-25', '2031-32': '-25'}", "table_61_line": 88, "table_61_head": "Tax", - "table_61_heading": "Other measures", + "table_61_heading": "Other policy measures including pre-announced policies > Other measures", "program": "indirect_tax", "computability": "not_expressible", "pe_reform_delta": null, "why": "An indirect tax or duty collected from businesses (VAT on donated medicines), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /vat|medicine/. Parameters: 64 node(s) ['gov.contrib.conservatives.cb_hitc_household', 'gov.contrib.conservatives.pensioner_personal_allowance', 'gov.contrib.labour.private_school_vat', 'gov.economic_assumptions.yoy_growth.obr.private_pension_index', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.UNITED_KINGDOM', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.ENGLAND']. Variables: 12 ['private_pension_wealth', 'private_pension_income', 'private_transfer_income', 'private_school_vat', 'attends_private_school_random_draw', 'attends_private_school']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /vat|medicine/. Parameters: 64 node(s) ['gov.contrib.conservatives.cb_hitc_household', 'gov.contrib.conservatives.pensioner_personal_allowance', 'gov.contrib.labour.private_school_vat', 'gov.economic_assumptions.yoy_growth.obr.private_pension_index', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.UNITED_KINGDOM', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.ENGLAND']. Variables: 12 ['attends_private_school', 'attends_private_school_random_draw', 'baseline_vat', 'full_rate_vat_consumption', 'full_rate_vat_expenditure_rate', 'private_pension_income']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -1828,15 +1828,15 @@ "measure_key": "ab2026__vat_medical_deputies_staffing", "title": "Temporary Staffing: Amend Group 7 of Schedule 9 to the VAT Act 1994 to restore the tax treatment of medical deputies to its previously intended position", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 89 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '85', '2028-29': '95', '2029-30': '100', '2030-31': '100', '2031-32': '105'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 89 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '85', '2028-29': '95', '2029-30': '100', '2030-31': '100', '2031-32': '105'}", "table_61_line": 89, "table_61_head": "Tax", - "table_61_heading": "Other measures", + "table_61_heading": "Other policy measures including pre-announced policies > Other measures", "program": "indirect_tax", "computability": "not_expressible", "pe_reform_delta": null, "why": "An indirect tax or duty collected from businesses (VAT on medical deputies' staffing supplies), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /vat|staffing/. Parameters: 64 node(s) ['gov.contrib.conservatives.cb_hitc_household', 'gov.contrib.conservatives.pensioner_personal_allowance', 'gov.contrib.labour.private_school_vat', 'gov.economic_assumptions.yoy_growth.obr.private_pension_index', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.UNITED_KINGDOM', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.ENGLAND']. Variables: 12 ['private_pension_wealth', 'private_pension_income', 'private_transfer_income', 'private_school_vat', 'attends_private_school_random_draw', 'attends_private_school']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /vat|staffing/. Parameters: 64 node(s) ['gov.contrib.conservatives.cb_hitc_household', 'gov.contrib.conservatives.pensioner_personal_allowance', 'gov.contrib.labour.private_school_vat', 'gov.economic_assumptions.yoy_growth.obr.private_pension_index', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.UNITED_KINGDOM', 'gov.economic_assumptions.yoy_growth.ons.private_rental_prices.ENGLAND']. Variables: 12 ['attends_private_school', 'attends_private_school_random_draw', 'baseline_vat', 'full_rate_vat_consumption', 'full_rate_vat_expenditure_rate', 'private_pension_income']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -1844,10 +1844,10 @@ "measure_key": "ab2026__landfill_communities_fund_credit_cap", "title": "Landfill Communities Fund: Increase the Landfill Tax credit cap from 1 April 2027", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 90 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-10', '2028-29': '-15', '2029-30': '-15', '2030-31': '-15', '2031-32': '-15'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 90 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '-10', '2028-29': '-15', '2029-30': '-15', '2030-31': '-15', '2031-32': '-15'}", "table_61_line": 90, "table_61_head": "Tax", - "table_61_heading": "Other measures", + "table_61_heading": "Other policy measures including pre-announced policies > Other measures", "program": "indirect_tax", "computability": "not_expressible", "pe_reform_delta": null, @@ -1860,15 +1860,15 @@ "measure_key": "ab2026__franchised_higher_education_registration", "title": "Higher Education: Introduce registration requirements for franchise providers with 300 or more students from 1 September 2028, to strengthen quality and oversight of franchised provision", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 91 (Spend, Tax/TME); exchequer impact GBP m by FY: {'2026-27': '30', '2027-28': '70', '2028-29': '200', '2029-30': '220', '2030-31': '200', '2031-32': '225'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 91 (Spend); exchequer impact GBP m by FY: {'2026-27': '30', '2027-28': '70', '2028-29': '200', '2029-30': '220', '2030-31': '200', '2031-32': '225'}", "table_61_line": 91, "table_61_head": "Spend", - "table_61_heading": "Other measures", + "table_61_heading": "Other policy measures including pre-announced policies > Other measures", "program": "regulation", "computability": "not_expressible", "pe_reform_delta": null, "why": "A regulatory change (registration of franchised higher education providers) with no tax or benefit rule for households.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /higher_education|franchise|student/. Parameters: 13 node(s) ['gov.dfe.disabled_students_allowance.maximum', 'gov.dfe.maintenance_loans.independent_student_age', 'gov.hmrc.student_loans.interest_rates.plan_1.boe_margin', 'gov.hmrc.student_loans.interest_rates.plan_2.additional_rate', 'gov.hmrc.student_loans.interest_rates.plan_2.upper_threshold', 'gov.hmrc.student_loans.interest_rates.postgraduate_additional_rate']. Variables: 22 ['student_payments', 'student_loans', 'uc_is_student_with_no_work_related_requirements', 'is_dependent_child_for_student_support', 'maintenance_loan_in_higher_education', 'adult_dependants_grant_adult_receives_student_finance']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /higher_education|franchise|student/. Parameters: 13 node(s) ['gov.dfe.disabled_students_allowance.maximum', 'gov.dfe.maintenance_loans.independent_student_age', 'gov.hmrc.student_loans.interest_rates.plan_1.boe_margin', 'gov.hmrc.student_loans.interest_rates.plan_2.additional_rate', 'gov.hmrc.student_loans.interest_rates.plan_2.upper_threshold', 'gov.hmrc.student_loans.interest_rates.postgraduate_additional_rate']. Variables: 22 ['adult_dependants_grant_adult_receives_student_finance', 'childcare_grant_full_time_student', 'childcare_grant_student_finance_eligible', 'disabled_students_allowance', 'disabled_students_allowance_course_eligible', 'disabled_students_allowance_eligible']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -1876,14 +1876,14 @@ "measure_key": "ab2026__high_needs_local_authority_spending", "title": "Local Authority Spending: Impact of adjustments to high needs budget on local authority spending", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 92 (Spend, TME); exchequer impact GBP m by FY: {'2026-27': '105', '2027-28': '140', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 92 (Spend); exchequer impact GBP m by FY: {'2026-27': '105', '2027-28': '140', '2028-29': '0', '2029-30': '0', '2030-31': '0', '2031-32': '0'}", "table_61_line": 92, "table_61_head": "Spend", - "table_61_heading": "Other measures", + "table_61_heading": "Other policy measures including pre-announced policies > Other measures", "program": "public_spending", "computability": "not_expressible", "pe_reform_delta": null, - "why": "Public spending (local authority high needs budgets), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", + "why": "Public spending (local authority high needs budgets), out of the household model's scope. The engine's spending channel, gov.contrib.policyengine.budget.education ('Increase in education, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset'), distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it.", "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /high_needs|special_educational|send/. Parameters: 2 node(s) ['gov.dfe.childcare_grant.eligible_child.max_age.special_educational_needs', 'gov.local_authorities.council_tax.band_d.amounts.ROSSENDALE']. Variables: 1 ['childcare_grant_child_has_special_educational_needs']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" @@ -1892,10 +1892,10 @@ "measure_key": "ab2026__benefits_in_kind_payrolling_delay", "title": "Benefits in Kind (BiKs): Delay mandatory payrolling of company cars, vans, fuel and medical benefits from 6 April 2026 to 6 April 2027, with remaining BiKs implemented from 6 April 2028", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 94 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '-30', '2027-28': '-5', '2028-29': '0', '2029-30': '0', '2030-31': '-5', '2031-32': '-5'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 94 (Tax); exchequer impact GBP m by FY: {'2026-27': '-30', '2027-28': '-5', '2028-29': '0', '2029-30': '0', '2030-31': '-5', '2031-32': '-5'}", "table_61_line": 94, "table_61_head": "Tax", - "table_61_heading": "Other measures", + "table_61_heading": "Other policy measures including pre-announced policies > Other measures", "program": "compliance_and_administration", "computability": "not_expressible", "pe_reform_delta": null, @@ -1908,10 +1908,10 @@ "measure_key": "ab2026__low_value_imports_customs_relief_removed", "title": "Low Value Imports: Remove the £135 customs duty relief and introduce new customs arrangements from October 2028", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 95 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '215', '2029-30': '0', '2030-31': '-10', '2031-32': '-15'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 95 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '215', '2029-30': '0', '2030-31': '-10', '2031-32': '-15'}", "table_61_line": 95, "table_61_head": "Tax", - "table_61_heading": "Other measures", + "table_61_heading": "Other policy measures including pre-announced policies > Other measures", "program": "indirect_tax", "computability": "not_expressible", "pe_reform_delta": null, @@ -1924,15 +1924,15 @@ "measure_key": "ab2026__ets_energy_from_waste_delay", "title": "Emissions Trading Scheme (ETS): Delay the inclusion of energy from waste in the ETS", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 96 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '-125', '2029-30': '-490', '2030-31': '-450', '2031-32': '-440'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 96 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '-125', '2029-30': '-490', '2030-31': '-450', '2031-32': '-440'}", "table_61_line": 96, "table_61_head": "Tax", - "table_61_heading": "Other measures", + "table_61_heading": "Other policy measures including pre-announced policies > Other measures", "program": "indirect_tax", "computability": "not_expressible", "pe_reform_delta": null, "why": "An indirect tax or duty collected from businesses (the Emissions Trading Scheme's coverage of energy from waste), out of the household model's scope. The engine's consumer channel, gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers'), spreads a £ change supplied as an input, so it would restate the costing rather than estimate it. The name search records what the engine holds.", - "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /emissions_trading|\\bets\\b|carbon/. Parameters: 37 node(s) ['gov.contrib.ubi_center.carbon_tax.consumer_incidence', 'gov.contrib.ubi_center.carbon_tax.rate', 'gov.economic_assumptions.yoy_growth.carbon_intensity', 'gov.economic_assumptions.indices.carbon_intensity', 'household.consumption.carbon.consumption', 'household.consumption.carbon.emissions.food_and_non_alcoholic_beverages']. Variables: 2 ['carbon_tax', 'carbon_consumption']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", + "name_search": "Searched BOTH the parameter tree and the VARIABLE list for /emissions_trading|\\bets\\b|carbon/. Parameters: 37 node(s) ['gov.contrib.ubi_center.carbon_tax.consumer_incidence', 'gov.contrib.ubi_center.carbon_tax.rate', 'gov.economic_assumptions.yoy_growth.carbon_intensity', 'gov.economic_assumptions.indices.carbon_intensity', 'household.consumption.carbon.consumption', 'household.consumption.carbon.emissions.food_and_non_alcoholic_beverages']. Variables: 2 ['carbon_consumption', 'carbon_tax']. Both are searched because a parameter-only search misses a variable (#106). Re-run at policyengine-uk 2.151.0.", "out_of_model_scope": true, "action_link": "https://github.com/PolicyEngine/policyengine-scorecard/issues/230" }, @@ -1940,10 +1940,10 @@ "measure_key": "ab2026__gas_distribution_revenue_recognition_spread", "title": "Corporation Tax: Spread the one-off impact of revenue recognition changes for qualifying gas distribution activities over five years for accounting periods ending on or after 31 December 2026", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 97 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '-160', '2027-28': '40', '2028-29': '40', '2029-30': '40', '2030-31': '40', '2031-32': '0'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 97 (Tax); exchequer impact GBP m by FY: {'2026-27': '-160', '2027-28': '40', '2028-29': '40', '2029-30': '40', '2030-31': '40', '2031-32': '0'}", "table_61_line": 97, "table_61_head": "Tax", - "table_61_heading": "Other measures", + "table_61_heading": "Other policy measures including pre-announced policies > Other measures", "program": "business_tax", "computability": "not_expressible", "pe_reform_delta": null, @@ -1956,10 +1956,10 @@ "measure_key": "ab2026__us_llc_double_taxation_relief", "title": "International Investors: Relieve double taxation for UK investors in US Limited Liability Companies and similar entities from early 2028", "reported_status": "announced", - "source_note": "HM Treasury, Budget 2026 Table 6.1 line 98 (Tax, Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '-5', '2029-30': '-10', '2030-31': '-5', '2031-32': '-5'}", + "source_note": "HM Treasury, Budget 2026 Table 6.1 line 98 (Tax); exchequer impact GBP m by FY: {'2026-27': '0', '2027-28': '0', '2028-29': '-5', '2029-30': '-10', '2030-31': '-5', '2031-32': '-5'}", "table_61_line": 98, "table_61_head": "Tax", - "table_61_heading": "Other measures", + "table_61_heading": "Other policy measures including pre-announced policies > Other measures", "program": "business_tax", "computability": "not_expressible", "pe_reform_delta": null, @@ -2091,7 +2091,7 @@ ], "pe_reform_delta": null, "computability": "partial", - "missing": "89 of 98 legs are not expressible (see each leg's verdict); a PolicyEngine number covers only the 9 expressible legs: ['ab2026__household_energy_payment_100_bands_a_to_d', 'ab2026__electricity_vat_zero_rate_to_july_2027', 'ab2026__fuel_duty_5p_cut_extended_and_rpi_rise_deferred', 'ab2026__carers_allowance_earnings_taper', 'ab2026__trading_allowance_3000', 'ab2026__pension_tax_allowance', 'ab2026__badr_lifetime_limit_500000', 'ab2026__property_income_rates_plus_2pp', 'ab2026__savings_income_rates_plus_2pp']", + "missing": "88 of 98 legs are not expressible and 1 is partial (line 4) (see each leg's verdict); a PolicyEngine number covers only the 9 expressible legs: ['ab2026__household_energy_payment_100_bands_a_to_d', 'ab2026__electricity_vat_zero_rate_to_july_2027', 'ab2026__fuel_duty_5p_cut_extended_and_rpi_rise_deferred', 'ab2026__carers_allowance_earnings_taper', 'ab2026__trading_allowance_3000', 'ab2026__pension_tax_allowance', 'ab2026__badr_lifetime_limit_500000', 'ab2026__property_income_rates_plus_2pp', 'ab2026__savings_income_rates_plus_2pp']", "expressible_legs": [ "ab2026__household_energy_payment_100_bands_a_to_d", "ab2026__electricity_vat_zero_rate_to_july_2027", @@ -2235,7 +2235,7 @@ ], "pe_reform_delta": null, "computability": "partial", - "missing": "43 of 45 legs are not expressible (see each leg's verdict); a PolicyEngine number covers only the 2 expressible legs: ['ab2026__household_energy_payment_100_bands_a_to_d', 'ab2026__carers_allowance_earnings_taper']", + "missing": "42 of 45 legs are not expressible and 1 is partial (line 4) (see each leg's verdict); a PolicyEngine number covers only the 2 expressible legs: ['ab2026__household_energy_payment_100_bands_a_to_d', 'ab2026__carers_allowance_earnings_taper']", "expressible_legs": [ "ab2026__household_energy_payment_100_bands_a_to_d", "ab2026__carers_allowance_earnings_taper" @@ -2262,7 +2262,7 @@ "measure_key": "ab2026__package_policyengine_seven", "title": "The seven household measures PolicyEngine modelled on Budget day, together", "reported_status": "announced", - "source_note": "A composition of Table 6.1 lines that producers score together; see `package_of`.", + "source_note": "PolicyEngine's combined Budget-day run of Table 6.1 lines 1, 2, 3, 7, 15, 52 and 61 (receipt in pe_modifier); no producer row uses it", "program": "package", "construction": "package_of_registry_measures", "package_of": [ diff --git a/pipeline/ab2026_measure_keys.py b/pipeline/ab2026_measure_keys.py index ad842d218..06bb93655 100644 --- a/pipeline/ab2026_measure_keys.py +++ b/pipeline/ab2026_measure_keys.py @@ -1,6 +1,9 @@ """Autumn Budget 2026 measure keys, one per HM Treasury Table 6.1 line. -The key space every AB2026 claim carries in conditions.measure_key (#230). +The key space of the AB2026 harvest (#230): every staged claim's measure_key is +a key from data/uk/ab2026_measures.json, null with +conditions.proposed_measure_key (README rule 2), or for Table 6.4 recostings +the earlier event's key named in conditions.recosting_of. Keys are written out here rather than derived from titles, so a reworded title in a later edition of the table cannot move a key. Both the Table 6.1 stager (pipeline/stage_ab2026_hmt_tables.py) and the registry builder diff --git a/pipeline/build_ab2026_registry.py b/pipeline/build_ab2026_registry.py index 74c816631..3ac88d5b3 100644 --- a/pipeline/build_ab2026_registry.py +++ b/pipeline/build_ab2026_registry.py @@ -75,7 +75,6 @@ def _t61(): { "title": r["reform_hint"], "head": c["head"], - "fiscal_head": c.get("fiscal_head"), "heading": c["heading"], "values": {}, }, @@ -83,7 +82,7 @@ def _t61(): e["values"][c["fy"]] = r["value_raw"] elif c["table"] == "6.1" and c["row_type"] == "total": totals.setdefault(r["reform_hint"], {})[c["fy"]] = r["value_raw"] - elif c["table"] == "6.2" and c["table_line"] == "4": + elif c["table"] == "6.2" and c.get("table_line") == "4": e = lines.setdefault("t62_4", {"title": r["reform_hint"], "values": {}}) e["values"][c["fy"]] = r["value_raw"] return lines, totals @@ -96,7 +95,9 @@ def _t61(): def ns(pattern): rx = re.compile(pattern, re.I) ph = [p for p in P if rx.search(p) and not p.startswith("baseline.")] - vh = [v for v in V if rx.search(v)] + # The parameter tree loads in sorted order, but policyengine-core adds + # variables in glob order, which differs by filesystem: sort them. + vh = sorted(v for v in V if rx.search(v)) return ( f"Searched BOTH the parameter tree and the VARIABLE list for /{pattern}/. " f"Parameters: {len(ph)} node(s) {ph[:6]}. Variables: {len(vh)} {vh[:6]}. " @@ -126,6 +127,9 @@ def ns(pattern): "consumer": "gov.contrib.policyengine.budget.consumer_incident_tax_change ('Tax increase incident on consumers')", "corporate": "gov.contrib.policyengine.budget.corporate_incident_tax_change ('Tax increase incident on owners of capital')", "spending": "gov.contrib.policyengine.budget.other_public_spending ('Increase in non-NHS, non-education public spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset')", + "nhs": "gov.contrib.policyengine.budget.nhs ('Increase in NHS spending, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset')", + "education": "gov.contrib.policyengine.budget.education ('Increase in education, distributed by income decile as estimated by the ONS Effects of Taxes and Benefits dataset')", + "high_income": "gov.contrib.policyengine.budget.high_income_incident_tax_change ('Tax rise for high income earners (functioning proportional to income over £100,000)')", } measures = [] @@ -138,9 +142,8 @@ def entry(line, program, computability, **kw): "title": t["title"], "reported_status": "announced", "source_note": ( - f"HM Treasury, Budget 2026 Table 6.1 line {line} ({t['head']}" - + (f", {t['fiscal_head']}" if t.get("fiscal_head") else "") - + f"); exchequer impact GBP m by FY: {t['values']}" + f"HM Treasury, Budget 2026 Table 6.1 line {line} ({t['head']}); " + f"exchequer impact GBP m by FY: {t['values']}" ), "table_61_line": line, "table_61_head": t["head"], @@ -454,7 +457,7 @@ def rates_up(paths): "income_tax", "expressible", construction="delta_on_current_law", - dashboard_measure_id="savings_income_rates_up", + dashboard_registered_id="savings_income_rates_up", pe_reform_delta=rates_up( [ f"gov.hmrc.income_tax.rates.savings.{b}" @@ -468,7 +471,7 @@ def rates_up(paths): for b in ("basic", "higher", "additional") ] ), - note="Registered by the dashboard but not in its published release, so there is no published receipt to cite; tranche 3 computes it from this delta.", + note="Registered by the dashboard (dashboard_registered_id) but not in its published release, and not run on Budget day, so there is no published receipt to cite; tranche 3 computes it from this delta.", ) # --------------------------------------------------------------------------- @@ -486,7 +489,12 @@ def rates_up(paths): delta_kind="absolute_value_by_year", engine_baseline_2026=baseline("gov.hmrc.cgt.badr.lifetime_limit"), head_variables=["capital_gains_tax", "capital_gains_badr"], - note="capital_gains_tax reads gov.hmrc.cgt.badr.lifetime_limit at the pin (variables/gov/hmrc/capital_gains_tax/capital_gains_tax.py). Not run on Budget day.", + note=( + "capital_gains_tax reads gov.hmrc.cgt.badr.lifetime_limit at the pin (variables/gov/hmrc/capital_gains_tax/capital_gains_tax.py). " + "capital_gains_badr is an input, and capital_gains_tax applies the lifetime limit to one year's qualifying gains: " + "the variable's documentation says it does so 'as if none of the limit had been used before', so a lower limit " + "binds only on a year's gains above it, not on gains already relieved in earlier years. Not run on Budget day." + ), ) @@ -514,7 +522,8 @@ def _cap_history(): missing=( "The pre-measure 2027 cap. The engine already carries the measure: " "gov.dft.bus.fares.cap is £2 from 1 January 2027 to 1 January 2028 " - "(its metadata cites DfT's '£2 bus fares from January 2027'). A " + "(its metadata cites 'DfT two-pound bus fares from January to December 2027', " + "https://www.gov.uk/government/speeches/2-bus-fares-from-january-2027). A " "counterpart reverses it, which needs the cap 2027 would have had " "without the measure; the parameter has one upstream revision " "(policyengine-uk c8d1f89a9d, 5 October 2026) and records no " @@ -679,11 +688,17 @@ def _cap_history(): # Out of the household model's scope # --------------------------------------------------------------------------- -SPEND = ( - "Public spending ({what}), out of the household model's scope. The engine's spending channel, " - + LEVERS["spending"] - + ", distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it." -) +def _spend(lever): + return ( + "Public spending ({what}), out of the household model's scope. The engine's spending channel, " + + LEVERS[lever] + + ", distributes a £ amount supplied as an input, so scoring this line through it would restate HM Treasury's figure, not estimate it." + ) + + +SPEND = _spend("spending") +SPEND_NHS = _spend("nhs") +SPEND_EDUCATION = _spend("education") BUSINESS = ( "A tax on businesses ({what}), out of the household model's scope: it changes what companies owe. The engine's corporate channel, " + LEVERS["corporate"] @@ -706,7 +721,7 @@ def _cap_history(): DETAINED = "A change to the {what}: the search finds no parameter or variable for hospital detention, prison or offending, so the engine cannot identify the people it affects." SCOPE = { - 9: (SPEND, "books for children aged 0 to 5", r"early_years|childcare|book"), + 9: (SPEND_EDUCATION, "books for children aged 0 to 5", r"early_years|childcare|book"), 10: ( INDIRECT, "tariffs on food products, heating oil and fertiliser", @@ -866,7 +881,7 @@ def _cap_history(): 82: (SPEND, "infected blood compensation", r"infected_blood|compensation"), 84: (SPEND, "Landfill Tax costs of illegal waste clearances", r"landfill|waste"), 85: ( - SPEND, + SPEND_NHS, "the NHS and the Violence against Women and Girls Strategy", r"nhs|violence", ), @@ -884,7 +899,7 @@ def _cap_history(): r"higher_education|franchise|student", ), 92: ( - SPEND, + SPEND_EDUCATION, "local authority high needs budgets", r"high_needs|special_educational|send", ), @@ -917,6 +932,8 @@ def _cap_history(): PROGRAM = { SPEND: "public_spending", + SPEND_NHS: "public_spending", + SPEND_EDUCATION: "public_spending", BUSINESS: "business_tax", INDIRECT: "indirect_tax", FINANCE: "financial_transactions", @@ -971,7 +988,13 @@ def _cap_history(): "source_note": ( f"HM Treasury, Budget 2026 Table 6.1 row '{spec['table_61_row']}'; GBP m by FY: {TOTALS[spec['table_61_row']]}" if spec["table_61_row"] - else "A composition of Table 6.1 lines that producers score together; see `package_of`." + else ( + "PolicyEngine's combined Budget-day run of Table 6.1 lines " + + ", ".join(str(l) for l in lines[:-1]) + + f" and {lines[-1]} (receipt in pe_modifier); no producer row uses it" + if slug == "package_policyengine_seven" + else "A composition of Table 6.1 lines that producers score together; see `package_of`." + ) ), "program": "package", "construction": "package_of_registry_measures", @@ -1007,9 +1030,17 @@ def _cap_history(): ) else: m["computability"] = "partial" + not_expr = [l for l, c in zip(lines, comp) if c == "not_expressible"] + partial = [l for l, c in zip(lines, comp) if c == "partial"] m["missing"] = ( - f"{len(lines) - len(expressible)} of {len(lines)} legs are not expressible " - "(see each leg's verdict); a PolicyEngine number covers only " + f"{len(not_expr)} of {len(lines)} legs are not expressible" + + ( + f" and {len(partial)} {'is' if len(partial) == 1 else 'are'} partial " + f"(line{'s' if len(partial) > 1 else ''} {', '.join(map(str, partial))})" + if partial + else "" + ) + + " (see each leg's verdict); a PolicyEngine number covers only " f"the {len(expressible)} expressible legs: {expressible}" ) m["expressible_legs"] = expressible @@ -1032,8 +1063,9 @@ def registry(): "One entry per measure ANNOUNCED at Autumn Budget 2026: every HM Treasury Table 6.1 line " "(including the lines a household model cannot touch), the one Table 6.2 financial " "transaction with no Table 6.1 line, and the packages producers score. It is the port's " - "key space (#230): every staged AB2026 claim carries conditions.measure_key from this " - "file, and every measure carries a computability verdict against the engine the Budget-day " + "key space (#230): every staged claim's measure_key is a key from this file, null with " + "conditions.proposed_measure_key (rule 2), or for Table 6.4 recostings the earlier event's " + "key named in conditions.recosting_of; and every measure carries a computability verdict against the engine the Budget-day " "runs used. It stages NO value claims; values live in the harvest under their producers " "(sources/harvest-uk-ab2026-2026-10-29/)." ), @@ -1105,7 +1137,9 @@ def registry(): # consideration. "Not in Table 6.1 or 6.4" is a statement about those two # tables, read in full, nothing more. PRE_EVENT = { - "ab2026__property_income_tax_rate_rise": "announced: Table 6.1 line 61 (" + "ab2026__property_income_tax_rate_rise": "an AB2025 measure (ab2025__property_income_separate_rates: " + "22%/42%/47% from 6 April 2027); Table 6.4 line 6 recosts it, and Table 6.1 line 61 raises those " + "rates by a further 2ppts to 24%/44%/49% (" + K.key(61) + ")", "ab2026__bank_profit_surcharge_increase": "announced: Table 6.1 line 55 (" diff --git a/sources/harvest-uk-ab2026-2026-10-29/REGISTRY_KEYS.md b/sources/harvest-uk-ab2026-2026-10-29/REGISTRY_KEYS.md index 46f33ae28..b94ca9454 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/REGISTRY_KEYS.md +++ b/sources/harvest-uk-ab2026-2026-10-29/REGISTRY_KEYS.md @@ -60,9 +60,9 @@ earlier event's registry named in `conditions.recosting_of`. | `ab2026__defence_admin_savings_reinvested` | 42 | not_expressible | Defence: Reinvest defence and security administration savings into frontline defence and national security | | `ab2026__ukraine_winter_air_defence` | 43 | not_expressible | Ukraine: Emergency winter support for air defence | | `ab2026__uk_export_finance_defence_capacity` | 44 | not_expressible | UK Export Finance: Increase capacity to support UK defence exports from June 2026 (increased income from additional activity) | -| `ab2026__hmrc_tax_gap_package` | 45 | not_expressible | HMRC: Measures to close the tax gap4 | +| `ab2026__hmrc_tax_gap_package` | 45 | not_expressible | HMRC: Measures to close the tax gap | | `ab2026__online_marketplace_liability_uk_businesses` | 46 | not_expressible | Online Marketplace Liability: Extend to UK businesses with a minimum annual sales threshold per platform of £90,000 from 1 October 2027 | -| `ab2026__dwp_fraud_and_error_prevention` | 47 | not_expressible | DWP Fraud and Error: Expand use of data and targeted controls to prevent fraud from entering the benefits system4 | +| `ab2026__dwp_fraud_and_error_prevention` | 47 | not_expressible | DWP Fraud and Error: Expand use of data and targeted controls to prevent fraud from entering the benefits system | | `ab2026__benefits_removed_hospitalised_serious_offenders` | 48 | not_expressible | Welfare Reform: Remove benefit entitlements from serious offenders hospitalised in Great Britain from 6 April 2027 | | `ab2026__distributions_framework_modernised` | 49 | not_expressible | Distributions: Modernise the distributions framework from 6 April 2028 | | `ab2026__advance_corporation_tax_abolished` | 50 | not_expressible | Advance Corporation Tax (ACT): Abolish Advance Corporation Tax from 1 April 2031 | From cb3d41a6568565741ff21db46931ff8463097752 Mon Sep 17 00:00:00 2001 From: Claude Date: Wed, 4 Nov 2026 18:24:11 +0000 Subject: [PATCH 21/52] uk_hmt_redbook: apply review A's RB-1 to RB-21 and K-7(b) Applied as written, with the coordinator's calls: RB-7 drops rows 247 and 252 and their one-row keys; RB-17 keeps row 185 on its own key, records its three scopes and its sign. RB-2's drops extend to the four printings of 'up to 20 million households' (two kept). Barnett: row 258 is the total, 258-261 carry 'positive = increase in devolved government funding'. Geography uses the closed codes (GB, not Great Britain). Adds the republished explainer as a 0-row primary (RB-14). 432 -> 425 rows. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_01LwQq6JvunwtJCAckL6L4aM --- .../uk_hmt_redbook/NOTES.md | 147 +++++++++++------- .../uk_hmt_redbook/claims_staged.jsonl.gz | Bin 36385 -> 36306 bytes .../uk_hmt_redbook/manifest.jsonl | 1 + 3 files changed, 93 insertions(+), 55 deletions(-) diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/NOTES.md index f2ae6b9b6..38b6a153a 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/NOTES.md @@ -2,19 +2,20 @@ **Working files.** `stage.py`, which writes this family's two data files and is the file meant wherever these notes say `stage.py`, lives on branch `uk/ab2026-harvest-staging` at `harvest-staging/ab2026/lanes/hmt-redbook/uk_hmt_redbook/stage.py`; this branch carries only NOTES.md, `manifest.jsonl` and `claims_staged.jsonl.gz`. `.ab2026-primaries/uk_hmt_redbook/` was the lane's local primaries folder and was never committed: the repository carries no primary bytes, so fetch each primary from its `manifest.jsonl` URL and check its sha256. -Source `hm_treasury`. **432 rows from 3 cited primaries (5 hashed versions), 23 proposed measure keys.** `stage.py` embeds every row and manifest line as data; `python3 stage.py` rewrites `claims_staged.jsonl.gz` and `manifest.jsonl` byte for byte with the standard library alone. No primary bytes are vendored. +Source `hm_treasury`. **425 rows from 3 cited primaries (6 hashed versions), 21 proposed measure keys.** `stage.py` embeds every row and manifest line as data; `python3 stage.py` rewrites `claims_staged.jsonl.gz` and `manifest.jsonl` byte for byte with the standard library alone. No primary bytes are vendored. -The first staging (29 October, morning) produced 1,699 rows and 250 proposed keys. This rework applies README rules 8 and 9: it keeps estimates of effects, drops design (parameters, allocations, financing, contingent liabilities), and gives each printed part of one Table 6.1 measure that measure's key. The section "Not staged (rules 8 and 9)" accounts for all 1,267 dropped rows. +The first staging (29 October, morning) produced 1,699 rows and 250 proposed keys. This rework applies README rules 8 and 9: it keeps estimates of effects, drops design (parameters, allocations, financing, contingent liabilities), and gives each printed part of one Table 6.1 measure that measure's key. Review A at 4f424b9 (4 November) dropped 7 more (see "Review A fixes"). The section "Not staged (rules 8 and 9)" accounts for all 1,274 dropped rows. ## Primaries | Primary version | SHA256 | Bytes | Access | Rows | |---|---|---:|---|---:| | Budget 2026 (HC 726), 28 October version | `bed70546c026487e3141799968f6df08210c182b0695fed0e8adf7efd6697e03` | 4,657,025 | local copy | 0 | -| Budget 2026 (HC 726), current version | `19b337aced8fe47de1d6ee462dabf75155110d347c38123b6380d90f09e97b93` | 4,679,545 | direct | 363 | -| Impact on households: distributional analysis to accompany Budget 2026 | `36cd5afb4fceca2cecba0d065974ffe8280888e15ed852c129339439e539469c` | 197,606 | direct | 57 | +| Budget 2026 (HC 726), current version | `19b337aced8fe47de1d6ee462dabf75155110d347c38123b6380d90f09e97b93` | 4,679,545 | direct | 357 | +| Impact on households: distributional analysis to accompany Budget 2026 | `36cd5afb4fceca2cecba0d065974ffe8280888e15ed852c129339439e539469c` | 197,606 | direct | 56 | | Pension Tax Allowance Explainer (content API, full body) | `cc3ec22eb55f4c827c301bc67b06aa7482e4970ef7e1cec644bc382dac0dbb25` | 24,883 | govuk_api | 12 | | Pension Tax Allowance Explainer (content API, 29 October abbreviated body) | `791d2573bafc67fcfd89567dab55c0239dd26febc9feec44d10f2be4ae21a9e4` | 9,816 | local copy | 0 | +| Pension Tax Allowance Explainer (content API, republished; `updated_at` 2026-10-29T15:55:28Z) | `83f8c894ce76a8e9ce80854f9b7287442d5f6b521a4eb260e5cfacae1c98909b` | 24,356 | govuk_api (review A, 4 November) | 0 | ## Access recipe @@ -31,37 +32,38 @@ All fresh fetches returned HTTP 200, and the files sit under `.ab2026-primaries/ - **HTML discovery:** the publication content API `https://www.gov.uk/api/content/government/publications/budget-2026-document` (hash `4568690c1ca115ff4a7aa5aa116bcdb39cebee39d8af4490b8dc3ad544b3ec6d`) names the HTML attachment, fetched as a cross-check (hash `33ee5a00b6383104f83790719c0a67e396cfb52e7ea93986b76ab15a19fdaaec`). Neither supplies rows; the PDF is the primary. - **Impact on households:** fresh PDF fetched 2026-10-29T15:41:44Z, identical to the supplied 29 October copy, read with `pdftotext -layout`. Printed pages equal PDF pages. - **Pension Tax Allowance explainer:** fresh content API JSON fetched 2026-10-29T15:41:44Z, matching the supplied 28 October response. Rows quote its decoded `details.body`. **The supplied 29 October response differs:** its body is a 353-character introduction without the counts, worked cases or FAQ of the 19,727-character full body. Both carry publication date 28 October. No row cites the abbreviated version. +- **Pension Tax Allowance explainer, republished (README rule 7).** Review A fetched the content API on 4 November and got `83f8c894…` (24,356 bytes, `updated_at` 2026-10-29T15:55:28Z, a 19,206-character body): HM Treasury republished the explainer about 14 minutes after the lane's fetch. All twelve explainer rows' quotes and values are verbatim in it. It fixes "part-time wok" and no longer repeats an FAQ question; "£13.319" remains. The rows keep citing `cc3ec22e…`, the version they were read from. ## Coverage Rows run in document order: red book (printed page, then position), Impact on households, Pension Tax Allowance explainer. -- **Red book (363 rows):** - - 160 fiscal aggregates restated from the OBR, ONS and IMF: Chapter 1 text, Chart 1.6's six data labels, Tables 1.2 and 1.3; +- **Red book (357 rows):** + - 160 fiscal aggregates restated from the OBR, ONS and IMF, with two shares of HM Treasury's own arithmetic (Box 1.C footnotes 54 and 56): Chapter 1 text, Chart 1.6's six data labels, Tables 1.2 and 1.3; - 84 cells of Table 6.3, which breaks Table 6.1 line 45 into eleven items, with its subtotal, the line 46 memo and the subtotal including memo; - - 30 costs, yields and savings: Table 6.1 measures where the red book prints the line's own Exchequer impact, printed parts of those lines, and saving estimates with their own scopes; - - 89 effects on people and businesses: counts affected; gains, losses and bill changes; GDP, earnings, price and jobs effects; Barnett consequentials. -- **Impact on households (57 rows):** the distributional and context statements of paragraphs 1.7, 1.9, 1.10, 2.4, 2.5 and 2.10, and the 50 cells of Table 2.C. + - 28 costs, yields and savings: Table 6.1 measures where the red book prints the line's own Exchequer impact, printed parts of those lines, and saving estimates with their own scopes; + - 85 effects on people and businesses: counts affected; gains, losses and bill changes; GDP, earnings, price and jobs effects; Barnett consequentials. +- **Impact on households (56 rows):** the distributional and context statements of paragraphs 1.7, 1.9, 1.10 (first bullet), 2.4, 2.5 and 2.10, and the 50 cells of Table 2.C. - **Pension Tax Allowance explainer (12 rows):** three pensioner counts and nine worked-case outcomes. Rows by measure key: | measure_key | rows | of which components (rule 9) | |---|---:|---:| -| `null` with a proposed key | 67 | 18 | -| `null`, context (no proposed key) | 191 | 0 | +| `null` with a proposed key | 66 | 18 | +| `null`, context (no proposed key) | 189 | 0 | | `ab2026__hmrc_tax_gap_package` | 73 | 67 | -| `ab2026__household_energy_payment_100_bands_a_to_d` | 12 | 0 | | `ab2026__pension_tax_allowance` | 12 | 0 | | `ab2026__british_industrial_competitiveness_scheme` | 8 | 0 | -| `ab2026__business_rates_rhl_transitional_support` | 6 | 0 | +| `ab2026__household_energy_payment_100_bands_a_to_d` | 8 | 0 | +| `ab2026__business_rates_rhl_transitional_support` | 6 | 6 | | `ab2026__business_rates_sbrr_rhl_thresholds_raised` | 6 | 0 | | `ab2026__online_marketplace_liability_uk_businesses` | 6 | 0 | | `ab2026__electricity_vat_zero_rate_to_july_2027` | 5 | 0 | | `ab2026__alcohol_duty_small_producer_relief` | 4 | 0 | | `ab2026__food_and_heating_oil_tariff_suspensions` | 4 | 4 | | `ab2026__mileage_allowance_payments_55p` | 4 | 0 | -| `ab2026__uk_india_ceta_and_double_contributions_convention` | 4 | 3 | +| `ab2026__uk_india_ceta_and_double_contributions_convention` | 4 | 4 | | `ab2026__your_first_home_equity_loan` | 4 | 0 | | `ab2026__defence_investment_plan_capital_reallocation` | 3 | 2 | | `ab2026__ukraine_winter_air_defence` | 3 | 0 | @@ -73,41 +75,42 @@ Rows by measure key: | `ab2026__misogyny_prevention_programmes` | 2 | 0 | | `ab2026__badr_lifetime_limit_500000` | 1 | 0 | | `ab2026__benefits_removed_hospitalised_serious_offenders` | 1 | 0 | -| `ab2026__business_rates_pubs_relief_35pc` | 1 | 0 | +| `ab2026__british_business_bank_backing_innovation` | 1 | 1 | | `ab2026__carbon_price_support_removed` | 1 | 0 | | `ab2026__carers_allowance_earnings_taper` | 1 | 0 | | `ab2026__hgv_ved_1_for_12_months` | 1 | 0 | | `ab2026__national_wealth_fund_capital` | 1 | 0 | | `ab2026__parental_allowance_fathers_partners_adopters` | 1 | 0 | -The 191 context rows carry no proposed key (README rule 2): 136 restated fiscal aggregates about no measure, and 55 Impact on households rows (Table 2.C and the context statements of paragraphs 1.10, 2.4, 2.5 and 2.10). The other 24 Table 1.3 rows, the OBR's effect of policy decisions, carry `ab2026__package_government_decisions_since_march_2026`. +The 189 context rows carry no proposed key (README rule 2): 135 fiscal aggregates about no measure (133 restated, and HM Treasury's own two Box 1.C shares), and 54 Impact on households rows (Table 2.C and the context statements of paragraphs 1.10, 2.4, 2.5 and 2.10). The other 24 Table 1.3 rows, the OBR's effect of policy decisions, and Chart 1.6's "Effect of policy" carry `ab2026__package_government_decisions_since_march_2026`. Rows by metric: | metric | rows | |---|---:| -| `revenue_change` | 114 | +| `revenue_change` | 112 | | `income_statistic` | 52 | -| `affected_count` | 38 | +| `affected_count` | 32 | | proposed `government_borrowing` | 28 | | proposed `government_debt` | 25 | | proposed `fiscal_balance` | 21 | | proposed `government_borrowing_change` | 20 | | `decisions_effect_on_borrowing` | 18 | -| `fiscal_headroom` | 17 | -| proposed `fiscal_balance_change` | 13 | +| proposed `fiscal_balance_change` | 16 | +| `fiscal_headroom` | 14 | | proposed `government_net_worth` | 9 | | `energy_bill_change` | 7 | -| proposed `business_cost_saving` | 7 | | proposed `debt_interest_change` | 7 | | proposed `household_tax_change` | 7 | +| proposed `business_cost_saving` | 6 | | proposed `average_cumulative_gain` | 5 | -| `affected_share` | 4 | | `gdp_level_effect` | 4 | | `tax_liability` | 4 | | proposed `aggregate_household_saving` | 4 | | proposed `bill_change_percent` | 4 | | proposed `block_grant_effect` | 4 | +| proposed `jobs_created_or_supported` | 4 | +| `affected_share` | 3 | | `average_annual_gain` | 3 | | `gainer_count` | 2 | | proposed `fiscal_consolidation_share` | 2 | @@ -121,9 +124,8 @@ Rows by metric: | proposed `share_in_net_recipient_deciles` | 1 | | proposed `share_in_top_half_of_expenditure` | 1 | | proposed `share_with_zero_liability` | 1 | -| proposed `spending_to_tax_ratio` | 1 | -Attribution: 263 rows `own`, 169 `restated` (OBR, ONS, IMF, DWP, DBT, the British Business Bank). +Attribution: 257 rows `own`, 168 `restated` (OBR, ONS, IMF, DWP, DBT, the British Business Bank, the National Wealth Fund). ## Conventions @@ -133,22 +135,27 @@ Attribution: 263 rows `own`, 169 `restated` (OBR, ONS, IMF, DWP, DBT, the Britis - savings: "positive = saving to the Exchequer"; - yields, including Table 6.3: "positive = yield to the Exchequer (reduces borrowing)"; - Table 1.3: "positive = increase in PSNB" (the opposite of Table 6.1's yield convention), except on its two memo lines, which give the change in the current budget deficit: "positive = increase in current budget deficit" (README rule 10, since 3 November); - - Table 1.2: deficits and debt positive, surpluses negative, as printed. -- **Periods.** Fiscal-year figures carry `conditions.fy`, end-year `period` and `time_basis: fiscal_year`. A multi-year total keeps its whole horizon in `conditions.horizon`, sits at its final fiscal year and is never annualised. An undated figure ("by the late 2030s", "this winter") sits at the publication year 2026 with the printed horizon in conditions and, where the year matters, `parse_confidence: medium`. + - Table 1.2: deficits and debt positive, surpluses negative, as printed; + - Chart 1.6's three bridge steps (rows 21, 22 and 24 at 4f424b9): "positive = reduction in current budget deficit (increase in stability rule headroom)", the reverse of Table 1.3's memo, as printed (Chart 1.6's "Effect of policy" +2.0 is Table 1.3's policy memo −2.0 in 2029-30); + - Box 1.C: each row says what it measures and which way it runs ("positive = fall in borrowing (… as worded: fell by)", "as printed; positive = current budget deficit (% of GDP)" and so on; review A's RB-12); + - Barnett consequentials: "positive = increase in devolved government funding"; + - gains to a person or business: "positive = gain to the (…, as printed)", and row 185's CPI effect "positive = reduction in CPI inflation (as worded: reduce inflation by)" (RB-17). +- **Periods.** Fiscal-year figures carry `conditions.fy`, end-year `period` and `time_basis: fiscal_year`. A multi-year total, or an average over the forecast, keeps its whole horizon in `conditions.horizon`, sits at its final fiscal year and is never annualised; "over the forecast" means 2026-27 to 2031-32, as paragraphs 1.20 and 1.40 date it. A figure printed more than once keeps a row for each printing (AB2025 did the same), and the repeats share one dating, the most specific mention's (review A's RB-15); where more than two printings would share one claim identity, only two are kept (RB-2). An undated figure ("by the late 2030s", "this winter") sits at the publication year 2026 with the printed horizon in conditions and, where the year matters, `parse_confidence: medium`. +- **Geography** comes from the harvest's closed list (`UK`, `GB`, `England`, `Wales`, `Scotland`, `Northern Ireland`, `England and Wales`, `Scotland, Wales and Northern Ireland`); any detail goes in `conditions.geography_note` or `note`. The Ofgem price cap and BICS are `GB`. - **Prices.** Every money figure is cash, as printed; Table 1.3's rows carry `price_basis: current prices`. No basis is converted. (The first staging tagged the £595 million administration saving `real terms`; paragraph 2.14's "real terms" belongs to the Spending Review's 16% cut, and the £595 million equals Table 6.1 line 63's cash +595.) - **"Central".** HM Treasury prints single estimates; `point` means the printed figure. No row is an inferred scenario. The one OBR scenario row (borrowing £12 billion lower if energy prices follow a lower path) carries `conditions.scenario`. -- **Rule 9 components.** A printed part of one Table 6.1 measure carries the line's key, `conditions.component` naming the part and `conditions.row_type: measure_component`. Components are never summed with the line's total. This family has 94: the 66 Table 6.3 item cells and the 160,000 employees (line 45); the £3 billion and £1.7 billion parts of line 41's £4.7 billion; the two food-tariff consumer savings (line 10); the Scotland, Wales and Northern Ireland GDP effects of the India agreement (line 77); and, under proposed keys, the Tax and Spending parts of the OBR's policy effect, the three devolved Barnett consequentials and three parts of the regulatory administrative savings. +- **Rule 9 components.** A printed part of one Table 6.1 measure carries the line's key, `conditions.component` naming the part and `conditions.row_type: measure_component`. Components are never summed with the line's total. This family has 102: the 66 Table 6.3 item cells and the 160,000 employees (line 45); the £3 billion and £1.7 billion parts of line 41's £4.7 billion; the four food-tariff consumer savings (line 10; the June-and-November range contains the November-only one, so never add the two); the India agreement's UK, Scotland, Wales and Northern Ireland GDP effects (line 77; the UK figure is the trade deal alone, without the Double Contributions Convention); paragraph 3.22's six figures for line 14's real-terms freeze and SSB extension; the British Business Bank's 40,000 jobs (line 29's capacity half, as Policy Costings p.34); and, under proposed keys, the Tax and Spending parts of the OBR's policy effect, the three devolved Barnett consequentials and three parts of the regulatory administrative savings. - **`component` means a part only.** Where the first staging used `component` for a Chapter 6 paragraph title on a whole-measure row, the condition is now `paragraph_heading`; in the pension worked cases it is `quantity`. Chart 1.6's six headroom-bridge data labels name their bar in `chart_bar` (changed on 3 November; the rework had left them under `component`). - **Attribution.** OBR forecasts are `restated` with source model `obr_efo_forecast`; ONS, IMF, DWP, DBT and British Business Bank figures are `restated` and their notes name the original. HM Treasury's own costings, counts and worked cases are `own`. ### Which spending figures stay (rule 8 exception) -A spending figure is a measure's design unless the red book presents it as the measure's Exchequer impact. The test applied: the figure stays when it equals the printed Table 6.1 line (from `uk_hmt_tables`) for the years the red book states, within the red book's rounding. Tax costs and yields, savings estimates and Table 6.3 stay regardless. +A spending figure is a measure's design unless the red book presents it as the measure's Exchequer impact. The test applied: the figure stays when it equals the printed Table 6.1 line (from `uk_hmt_tables`), within the red book's rounding, as a total over the years the red book states or, for a per-year amount, as the line's per-year amount (review A's RB-20(b): the early-years books rows pass only the second form). Tax costs and yields, savings estimates and Table 6.3 stay regardless. | Table 6.1 line | Red book | Table 6.1 prints (£ million) | Rows | |---|---|---|---| -| 1 Household Energy Payment | £2.6 billion | -2,625 in 2026-27 | kept (4) | -| 9 Early years books | £5 million a year | -5 a year (2026-27 to 2028-29) | kept (2); years differ, see anomalies | +| 1 Household Energy Payment | £2.6 billion | -2,625 in 2026-27 | kept (2): the executive summary and para 2.48 (review A's RB-2 dropped Box 3.A's and para 3.5's repeats) | +| 9 Early years books | £5 million a year | -5 a year (2026-27 to 2028-29) | kept (2): the per-year amount equals the line's; the years differ, see anomalies | | 19 Small Producer Relief | £10 million per year | -10 a year from 2027-28 | kept (2) | | 32 EV charging | £100 million over two years | -50, -50 | kept (1) | | 41 DIP capital reallocation | £4.7 billion, 2027-28 to 2029-30 | +2,135, +1,420, +1,100 = 4,655 | kept (1), with its £3 billion and £1.7 billion parts as components (2) | @@ -168,29 +175,27 @@ A spending figure is a measure's design unless the red book presents it as the m | 37 Your First Home (Table 6.2 line 5) | £4.7 billion | -4,700 of financial transactions | dropped: a financial-transactions allocation | | 6.2 line 4 Sovereign AI | £60 million from 2027-28 | -45 a year | dropped | -## Proposed measure keys (23) +## Proposed measure keys (21) Each key below is the whole scope of its rows, stated as Table 6.1 lines or "no Table 6.1 line", with the red book's words. -- `ab2026__package_government_decisions_since_march_2026` (24 rows): the OBR's "Total effect of policy decisions" on PSNB in Table 1.3, with its Tax and Spending parts (components) and the memo effect on the current budget deficit, 2026-27 to 2031-32. Crosses Table 6.1 lines 1–98 and adds the OBR's indirect effects; Table 1.3 footnote 1 says the OBR's EFO Table 3.1 explains the "Differences between the direct impacts of the policy measures as set out in Table 6.1 of the Budget, and how the OBR incorporate these into their forecast". Not `ab2026__package_total_policy_decisions`. +- `ab2026__package_government_decisions_since_march_2026` (25 rows): the OBR's "Total effect of policy decisions" on PSNB in Table 1.3, with its Tax and Spending parts (components) and the memo effect on the current budget deficit, 2026-27 to 2031-32, and Chart 1.6's "Effect of policy" on stability rule headroom in 2029-30 (+2.0: the memo's −2.0 seen as headroom; proposed `fiscal_balance_change`). Crosses Table 6.1 lines 1–98 and adds the OBR's indirect effects; Table 1.3 footnote 1 says the OBR's EFO Table 3.1 explains the "Differences between the direct impacts of the policy measures as set out in Table 6.1 of the Budget, and how the OBR incorporate these into their forecast". Not `ab2026__package_total_policy_decisions`. - `ab2026__package_hmrc_tax_gap_and_online_marketplace` (7 rows): exactly Table 6.1 lines 45 and 46. Table 6.3's "Subtotal including memo" (six years) and footnote 73's "£2.3 billion from this Budget": Table 6.3 prints +2,250m for lines 45 and 46 in 2029-30, against +1,980m for line 45 alone. - `ab2026__package_parliament_tax_gap_measures` (1 row): lines 45 and 46 plus the tax-gap measures of Autumn Budget 2024, Spring Statement 2025 and Budget 2025 (footnote 73). Paragraph 2.46: "the total due to be raised from measures to close the tax gap announced during this Parliament to over £12 billion in 2029-30". - `ab2026__package_property_and_savings_income_rates` (2 rows): exactly Table 6.1 lines 61 and 62. Paragraph 2.50: "84% of income tax payers will be unaffected by these changes"; "Around half of the revenue raised from this change is expected to come from the top 10% of households". -- `ab2026__package_breathing_space_for_households` (1 row): paragraph 3.2's "Government action at this Budget to provide breathing space for households", which the OBR expects to reduce CPI inflation by 0.1 percentage points in 2027. Crosses Table 6.1 lines; the red book lists none for this figure. Table 6.1 groups lines 1–11 under "Breathing space for families", and Box 3.A's "breathing space for households" names line 2 (the VAT cut) and line 1 (the Household Energy Payment). +- `ab2026__package_breathing_space_for_households` (1 row): paragraph 3.2's "Government action at this Budget to provide breathing space for households", which the OBR expects to reduce CPI inflation by 0.1 percentage points in 2027. Crosses Table 6.1 lines; the red book lists none for this figure. Three scopes are in play, and the row keeps this key until tranche 2 settles them (the coordinator's call on review A's RB-17): Box 3.A's "breathing space for households" is lines 1 and 2 (the Household Energy Payment and the VAT cut); the OBR family keys its own CPI effect to lines 2 and 3 (`ab2026__package_fuel_duty_and_electricity_vat`); and Table 6.1 groups lines 1–11 under "Breathing space for families". Paragraph 1.38 repeats the 0.1 percentage points and is not staged. - `ab2026__energy_debt_circuit_breaker` (3 rows): no Table 6.1 line. Paragraph 3.8: "a new debt strategy to address this and act as a circuit breaker on energy debt … via reforms to prevent debt building up and a one-off targeted debt relief scheme for eligible households". - `ab2026__package_parliament_fuel_duty_decisions` (3 rows): Table 6.1 line 3 plus the fuel duty decisions of earlier events in this Parliament. Paragraph 3.13: "the government's decisions on fuel duty across this Parliament … compared with plans under the previous government"; footnote 98's baseline is the 5p cut reversed in April 2025 and annual RPI uprating. - `ab2026__minimum_wage_uprating_april_2027` (2 rows): no Table 6.1 line (a regulation). Paragraph 3.18: "raising the National Living Wage and National Minimum Wages" from 1 April 2027. - `ab2026__package_business_rates_rhl_sbrr_and_transitional_support` (2 rows): Table 6.1 lines 13 and 14, the three bullets of paragraph 3.22 (SBRR thresholds; the real-terms freeze; the SSB extension), to which paragraph 3.23's "As a result of this package" refers. The red book does not say whether paragraph 3.21's pubs relief (line 12) also counts. +- `ab2026__package_pubs_relief_35pc_total` (1 row): the whole 35% business rates relief for pubs, social clubs and live music venues in England: the 15% announced in January 2026 (paragraph 6.189; no AB2026 Table 6.1 line) plus Table 6.1 line 12's further 20ppt. Paragraph 3.21: "Taken together, this relief is a tax cut worth an average of over £3,000 per pub next year." Not line 12's key, since line 12 is only the 20ppt (rule 3; review A's RB-3). - `ab2026__package_regulation_tariff_and_sandbox_business_savings` (1 row): the executive summary's "almost £800 million per year through further cuts to the administrative burdens of regulation, extending suspensions on tariffs, and introducing new cross-economy powers so firms can safely test new technologies faster". Crosses Table 6.1 line 16 (business tariff suspensions) with measures that have no Table 6.1 line. - `ab2026__regulatory_administrative_savings` (5 rows): no Table 6.1 line. Chapter 6 decision "Regulatory administrative savings" (paragraph 6.36) and paragraph 4.52's "up to an estimated £640 million per annum of new gross administrative savings for business"; its £290 million, £116 million and £105 million parts are components. -- `ab2026__package_parliament_regulatory_burden_reduction` (1 row): no Table 6.1 line. Paragraph 4.51: the 25% administrative-burden target, "with over £2 billion of net progress secured" across this Parliament. -- `ab2026__package_bbb_equity_community_lender_and_venture_debt` (1 row): paragraph 4.46's "three measures": £1.4 billion of commercial equity commitments, £165 million for community lenders and £450 million for a venture debt product, which the British Business Bank's additional £5 billion of capacity enables. The red book does not map them to lines; the BBB's capacity sits in Table 6.1 line 29 (Table 6.2 line 3) and the Growth Guarantee Scheme in line 75. - `ab2026__rolls_royce_gas_turbine_programme` (1 row): no Table 6.1 line named. Paragraph 4.80: "co-investment in a joint, 10-year, £2.5‑billion programme for next-generation gas turbine technologies for civil and defence aerospace"; paragraph 4.83 says it "has not been contracted". -- `ab2026__package_defence_investment_plan` (1 row): the Defence Investment Plan published in June 2026, which no single line covers; lines 40–42 carry its savings and funding. Paragraph 5.34: "The Defence Investment Plan will create approximately 60,000 new jobs by 2029‑30 compared with 2023‑24." -- `ab2026__english_devolution_agreements` (1 row): no Table 6.1 line. Paragraphs 4.67–4.68: "the final stages of agreeing 10 new devolution agreements", after which "over 90% of England's population will have a devolution agreement in place" (the 90% includes existing agreements). -- `ab2026__package_dwp_fraud_and_error_measures` (1 row): Table 6.1 line 47 plus the DWP fraud and error decisions of paragraphs 6.19–6.22, which have no lines of their own. Paragraph 5.17: "new measures expected to save an additional £1.4 billion over the forecast". Line 47 alone prints +600m over 2026-27 to 2031-32 (and Table 6.2 line 9 +340m). -- `ab2026__package_parliament_welfare_fraud_and_error_savings` (1 row): line 47's measures plus the fraud and error measures of earlier events in this Parliament. Paragraph 5.17: "bringing total savings from fraud and error to over £20 billion over the forecast". -- `ab2026__package_barnett_consequentials` (4 rows): the Barnett consequentials of "decisions confirmed at Budget", 2026-27 to 2029-30, excluding the technical reduction from lower employer pension contributions (paragraph 4.150). Crosses Table 6.1's England spending lines; no line of its own. The Scottish, Welsh and Northern Ireland figures are components. +- `ab2026__package_defence_investment_plan` (1 row): the Defence Investment Plan published in June 2026, which no single line covers; lines 40–42 carry its savings and funding. Paragraph 5.34: "The Defence Investment Plan will create approximately 60,000 new jobs by 2029‑30 compared with 2023‑24." The OBR family's row on this key (paragraph 3.3's £1.8 billion a year) is the plan's RDEL funding part, a component (review A's K-2). +- `ab2026__package_dwp_fraud_and_error_measures` (1 row): the "new measures" of paragraph 5.17, "expected to save an additional £1.4 billion over the forecast" (2026-27 to 2031-32). The red book doesn't list them. Paragraphs 6.19–6.22 are line 47's own package (its footnote 4; Policy Costings pp.54–58, which `uk_hmt_costings` keys as line 47 components), but the £1.4 billion reconciles with no printed basis for line 47: the line prints +600m over the forecast, +940m with Table 6.2 line 9's +340m, and the costings sum to +570m. So the figure stays off line 47's key (rule 3; review A's RB-8). +- `ab2026__package_parliament_welfare_fraud_and_error_savings` (1 row): this Budget's new fraud and error measures plus those already in the forecast. Paragraph 5.17: "bringing total savings from fraud and error to over £20 billion over the forecast". It doesn't say which earlier measures count. (Review A suggested dropping "parliament" from the one-row key's name; tranche 2 can rename it when it registers the key.) +- `ab2026__package_barnett_consequentials` (4 rows): the Barnett consequentials of "decisions confirmed at Budget", 2026-27 to 2029-30, excluding the technical reduction from lower employer pension contributions (paragraph 4.150). Crosses Table 6.1's England spending lines; no line of its own. The three-government £1.3 billion is `row_type: total`, and the Scottish, Welsh and Northern Ireland figures are components, one per government; `uk_devolved_govts` types the Welsh Government's own figures as Welsh Government components of this key. All four rows say "positive = increase in devolved government funding" (review A's K-7(b)). - `ab2026__mod_counter_fraud_strategy` (1 row): no Table 6.1 line. Paragraph 5.32: "a new Counter Fraud Strategy in November, with investment in fraud and error analytics and closer working with the Public Sector Fraud Authority set to deliver £250 million of benefits". - `ab2026__state_pension_triple_lock_adjustment` (2 rows): no Table 6.1 line (its savings fall after the forecast). Paragraph 5.23: "adjusting the Triple Lock from April 2030 so that the State Pension rises each year by the highest of 2.5%, inflation, or the amount required to maintain its value relative to earnings as reached in April 2029"; DWP's estimate (footnote 61). - `ab2026__package_hmt_distributional_budget_2026` (1 row): HM Treasury's own grouping, the Budget 2026 policies modelled in Impact on households (paragraph 1.4): Table 6.1 lines 2, 3, 4, 5, 8, 11, 15, 26, 39, 49, 52, 59, 60, 61, 62, 72 and 95, plus the private landlord registration fee, which has no line. "Decisions announced at Budget 2026 (including those pre-announced where funding is allocated at Budget 2026)". It excludes line 1, the Household Energy Payment. @@ -201,15 +206,15 @@ The coordinator's shared spellings: no kept row has the scope of `ab2026__gas_st ## Quantity and baseline proposals - **Proposed metrics:** - - fiscal aggregates (restated): `government_borrowing`, `government_borrowing_change`, `government_debt`, `government_net_worth`, `fiscal_balance`, `fiscal_balance_change`, `debt_interest_change`, `fiscal_consolidation_share`; `block_grant_effect` (AB2025 list); - - household and business effects: `household_tax_change` (AB2025 list; worked cases), `average_cumulative_gain` (multi-year savings), `aggregate_household_saving` (AB2025 spelling; consumer savings), `bill_change_percent` (business electricity bills), `gross_earnings_change` (AB2025 spelling), `business_cost_saving` (administrative savings to business; replaces `operating_cost_change`, which the enum defines as an official operating-cost line), `housing_supply_change` (the 20,000 additional homes); - - shares with exact meanings: `unaffected_share` (the share printed as unaffected), `share_with_zero_liability` (half of RHL properties paying no rates), `share_in_non_gaining_deciles`, `share_in_net_recipient_deciles`, `spending_to_tax_ratio`, `income_mobility_share`, `share_in_top_half_of_expenditure` (Impact on households context). -- **Proposed units:** `businesses` (15 rows), `dwellings` (1), `ratio` (1). + - fiscal aggregates (restated): `government_borrowing`, `government_borrowing_change`, `government_debt`, `government_net_worth`, `fiscal_balance`, `fiscal_balance_change` (also Chart 1.6's three bridge steps, which change the stability rule's headroom, i.e. the 2029-30 current budget), `debt_interest_change`, `fiscal_consolidation_share`; `block_grant_effect` (AB2025 list); + - household and business effects: `household_tax_change` (AB2025 list; worked cases), `average_cumulative_gain` (multi-year savings), `aggregate_household_saving` (AB2025 spelling; consumer savings), `bill_change_percent` (business electricity bills), `gross_earnings_change` (AB2025 spelling), `business_cost_saving` (administrative savings to business; replaces `operating_cost_change`, which the enum defines as an official operating-cost line), `housing_supply_change` (the 20,000 additional homes), `jobs_created_or_supported` (the NWF's 100,000, the BBB's 40,000, Rolls-Royce's 3,300 and the DIP's 60,000; `affected_count` counts individuals affected, so jobs are not it, review A's RB-9); + - shares with exact meanings: `unaffected_share` (the share printed as unaffected), `share_with_zero_liability` (half of RHL properties paying no rates), `share_in_non_gaining_deciles`, `share_in_net_recipient_deciles`, `income_mobility_share`, `share_in_top_half_of_expenditure` (Impact on households context). +- **Proposed units:** `businesses` (15 rows), `dwellings` (1). - **Proposed baselines:** `hmt_no_budget_2026_modelled_policies` (paragraph 1.6, "a counterfactual in which the modelled policies were not introduced") and `hmt_no_policy_changes_from_ab2024_through_ab2026` (paragraph 1.8, "a counterfactual in which the modelled policies and spending settlements were not introduced"), one row each. Other rows have `baseline_policy: null`. ## Not staged (rules 8 and 9) -The rework dropped 1,267 of the first staging's 1,699 rows. 234 proposed keys named only dropped rows and are gone. +The rework dropped 1,267 of the first staging's 1,699 rows, and review A's fixes dropped 7 more (4 November), so 1,274 rows are not staged. The first staging's proposed keys that named only dropped rows are gone; their number can't be checked, because the first staging is in neither branch's history, so it is not given. | Category | Rows | Examples (locator) | |---|---:|---| @@ -217,10 +222,12 @@ The rework dropped 1,267 of the first staging's 1,699 rows. 234 proposed keys na | B. Spending allocations: programme funding, capitalisations, investment totals, spending levels | 221 | 166 programme amounts that are not a Table 6.1 line's printed impact (table above), e.g. "£120 million" for the AI Security Institute (para 4.130), "£30 million" for Cardiff Parkway (para 4.157), "£1.5 billion" for the Social and Affordable Homes Programme (para 5.4); 46 investment totals, e.g. NWF "£15.4 billion" (para 4.44), Your First Home "£4.7 billion" (para 5.5); public sector net investment (Table 1.2, 6); departmental spending levels (2); £200 million of private AI model credits (para 4.132). | | C. Financing: DMO remit, gilts, Treasury bills, NS&I, cash requirements, financial transactions | 177 | DMO net financing requirement "£258.6 billion" (Annex A.8); Table A.1 gilt sales by maturity; CGNCR "£144.1 billion" (para 1.47); Table 1.4 financial transactions by programme, e.g. Student Loans "15.9" (55 rows); Sovereign AI "£60 million" (para 4.135). | | D. Contingent liabilities | 14 | Table 1.5, e.g. Department for Energy Security and Net Zero "1.2-1.7"; Total reasonable worst case "4-4.5". | -| E. Programme inputs, scope counts and targets | 24 | "670" HMRC compliance officers to recruit (paras 2.42, 6.151); tariffs on "over 100" food products (para 3.17); "around 10,000" homes the Social and Affordable Homes Programme funds (para 6.114); the 25% administrative-burden target worth "£5.6 billion" (para 4.51). | -| F. Effects of policies outside this Budget | 2 | "more than 100" businesses using the Global Talent funder pathway since Budget 2025 (para 4.112); "over £5 billion" of services exports from the Switzerland and South Korea deals (para 4.124). | +| E. Programme inputs, scope counts and targets | 25 | "670" HMRC compliance officers to recruit (paras 2.42, 6.151); tariffs on "over 100" food products (para 3.17); "around 10,000" homes the Social and Affordable Homes Programme funds (para 6.114); the 25% administrative-burden target worth "£5.6 billion" (para 4.51); "over 90% of England's population will have a devolution agreement in place" (para 4.68), a coverage count that includes existing agreements (review A's RB-7, dropped 4 November). | +| F. Effects of policies outside this Budget | 3 | "more than 100" businesses using the Global Talent funder pathway since Budget 2025 (para 4.112); "over £5 billion" of services exports from the Switzerland and South Korea deals (para 4.124); "over £2 billion of net progress secured" against the administrative-burden target (para 4.51), progress before this Budget (footnote 137 cites a July 2026 annex; para 4.52 counts this Budget's £640 million separately; RB-7). | +| G. Repeats beyond two of one claim | 4 | The Household Energy Payment's "£2.6 billion" in para 3.5 and Box 3.A, and "up to 20 million households" in para 3.3 and Box 3.A. Kept: the £2.6 billion of the executive summary and para 2.48, and the 20 million of the executive summary and paras 2.48 and 3.5 (review A's RB-2; see Review A fixes). | +| H. The rework brief's drop list | 1 | Impact on households para 1.10's "over four times as much in public spending than they pay in tax" (`spending_to_tax_ratio`, which the brief's item 1 names for removal; RB-13). Row 365's 60% beside it stays as a printed value. | -Rule 9 moved 74 kept rows from proposed keys to a Table 6.1 line's key: the 66 Table 6.3 item cells and the 160,000 employees (proposed keys such as `ab2026__package_tax_gap_timely_payment`, now line 45 components); four food-tariff savings (now line 10 components); the £3 billion and £1.7 billion DIP parts (now line 41 components); and the Your First Home count (now line 37). One row moved the other way: the £1.4 billion of DWP fraud savings, whose scope is wider than line 47 (rule 3). +Rule 9 moved 74 kept rows from proposed keys to a Table 6.1 line's key: the 66 Table 6.3 item cells and the 160,000 employees (proposed keys such as `ab2026__package_tax_gap_timely_payment`, now line 45 components); four food-tariff savings (now line 10 components); the £3 billion and £1.7 billion DIP parts (now line 41 components); and the Your First Home count (now line 37). One row stays off a line's key: the £1.4 billion of DWP fraud savings, which reconciles with no printed basis for line 47 (rule 3; see its proposed key). ### Other material not staged @@ -230,15 +237,17 @@ Rule 9 moved 74 kept rows from proposed keys to a Table 6.1 line's key: the 66 T - **Footnote 73's earlier-event yields** (£6.5 billion, over £1 billion, £2.4 billion) are other fiscal events' measures; only this Budget's £2.3 billion and the £12 billion total are staged. - **Background macro forecasts** that score no measure (GDP growth, inflation, employment, Table 1.1), departmental efficiency histories, Annex B budgets and Annex C charts are not staged. - **The abbreviated 29 October explainer** has no quantitative content. +- **Other figures not staged (review A's RB-20(c)):** paragraph 5.17's Motability saving "over £1 billion over the next five years" (no Table 6.1 line); paragraph 3.7's £150 from Budget 2025 (an earlier event's measure); paragraph 4.36's reprioritisation savings (spending allocations); paragraph 4.42's National Wealth Fund and British Business Bank outcomes to date (effects before this Budget); and paragraph 6.9's £150 million (an allocation). +- **Paragraph 1.38's** repeat of the OBR's 0.1 percentage point CPI effect (row 185 stages paragraph 3.2's). ## Judgement calls for the coordinator -- **`spending_to_tax_ratio` is kept, against the drop list.** Its only row is Impact on households paragraph 1.10's "households in the lowest income decile will receive over four times as much in public spending than they pay in tax": a distributional statement and one of the Impact on households printed values the brief keeps. The 60% row beside it stays too. Drop both if the drop list was meant literally. +- **`spending_to_tax_ratio` is dropped (review A's RB-13).** The 3 November finish kept paragraph 1.10's "over four times" as a distributional statement. Review A read the rework brief's item 1 as a named removal list whose only exception is a Table 6.1 cost or yield: neither primary prints another spending-to-tax ratio, so the list can only mean this row, and HM Treasury frames paragraph 1.10 as "context beyond the impact of recent government policy decisions". The row is dropped (category H). The 60% row beside it, Table 2.C and the other Impact on households context rows stay as printed values. - **Mislabelled shares beyond the 84%.** Besides the 84% (`unaffected_share`, never 16%), BADR's "three-quarters … will still receive relief on their entire gain" is the unaffected share too; "half of all RHL properties … will not pay business rates at all" includes properties already paying none (`share_with_zero_liability`); and Impact on households' 20%, 60%, 59% and approximately 20% are not affected shares (proposals above). The registered `share_no_change` is a household-income measure, so the 84% of income tax payers follows the brief's `unaffected_share`. - **Registered metrics replace AB2025-era proposals.** The four "around £50 off the annual Ofgem price cap" rows and the three energy-debt "£30" rows now use the registered `energy_bill_change` (unit `gbp_per_household`), which the enum mints for HM Treasury bill figures; the India agreement's GDP effects use `gdp_level_effect` in GBP (the AB2025 IPPR precedent). - **Financial transactions and public sector net investment are dropped as financing and spending levels.** Table 1.4 prints forecast levels of lending by programme, not Budget effects. Public sector net investment is an OBR aggregate, but the brief lists `public_investment_share` for removal. -- **Jobs and induced homes stay; delivery targets go.** Estimated jobs (NWF 100,000; BBB 40,000; Rolls-Royce 3,300; DIP 60,000) and Your First Home's "estimated 20,000 new homes" are estimated effects. The 10,000 homes the Social and Affordable Homes Programme funds (in line 35's title), robotic procedures rising to 500,000, 2,000 learners and the 650,000 workforce requirement of the infrastructure pipeline are delivery targets of allocations. -- **The BBB's 40,000 jobs is now `restated`.** Footnote 135 cites the British Business Bank's press release. +- **Jobs and induced homes stay; delivery targets go.** Estimated jobs (NWF 100,000; BBB 40,000; Rolls-Royce 3,300; DIP 60,000; proposed `jobs_created_or_supported` since review A) and Your First Home's "estimated 20,000 new homes" are estimated effects. The 10,000 homes the Social and Affordable Homes Programme funds (in line 35's title), robotic procedures rising to 500,000, 2,000 learners and the 650,000 workforce requirement of the infrastructure pipeline are delivery targets of allocations. +- **The BBB's 40,000 jobs is now `restated`.** Footnote 135 cites the British Business Bank's press release. The NWF's 100,000 is `restated` too (footnote 134 cites the National Wealth Fund's own statement). ## Printed problems and differences @@ -246,19 +255,19 @@ Rule 9 moved 74 kept rows from proposed keys to a Table 6.1 line's key: the 66 T - Social and Affordable Homes Programme: £1.5 billion (paras 5.4, 6.114) against line 35's -1,750 over 2027-28 to 2030-31. - Early years books: "£5 million a year for three years from 2027-28" (paras 3.20, 6.122) against line 9's -5 in 2026-27, 2027-28 and 2028-29. The amount agrees; the years do not. - Local apprenticeship services £100 million (line 22: -120); Good Growth Funds £285 million in 2027-28 (line 20: -340); Strategic Authorities £150 million (line 25: -180); Union Learning Fund £15 million a year (line 73: -20 a year); Sovereign AI £60 million (Table 6.2 line 4: -45 a year). - - DWP fraud and error: "an additional £1.4 billion over the forecast" (para 5.17) against line 47's +600m in total. + - DWP fraud and error: "an additional £1.4 billion over the forecast" (para 5.17) against line 47's +600m in total (+940m with Table 6.2 line 9; the Policy Costings' +570m). - **Regulatory savings target:** paragraph 6.36 prints "£5.6 million"; paragraph 4.51 prints "£5.6 billion". The executive summary's "almost £800 million per year" and paragraph 4.52's "up to … £640 million per annum" have different printed scopes; both stay. - **Repeated amounts that disagree (not staged after rule 8):** local AI partnerships £30 million (para 4.135) and £15 million (para 6.101); Leeds College of Building £28 million (para 4.71) and £15 million (para 6.53); business tariff suspensions on 269 goods (para 3.28), 350 goods (para 4.147) and 361 extensions plus 72 new suspensions (line 16's title). - **Pubs relief:** the text prints an extra "20%"; Table 6.1 prints "20ppts". - **Table 1.3** labels the new forecast "November 2026 forecast" in an October Budget. Chart 1.6 prints 14.5 where the text says £15 billion. "Debt is lower at the end of the forecast" fits public sector net financial liabilities (83.2% to 81.7%), while public sector net debt goes from 95.5% to 95.6%. - **Paragraphs 6.225 and 6.226** repeat the Machine Games Duty measure. Annex A footnote 181 points to A.17 for green gilts, which A.19 covers. -- **Pension Tax Allowance explainer:** the FAQ prints "£13.319" for £13,319; the explainer prints both £13,036.40 and the rounded £13,037, and "up to £94" beside Steph's £93.28. The FAQ repeats a question and prints "part-time wok". +- **Pension Tax Allowance explainer:** the FAQ prints "£13.319" for £13,319; the explainer prints both £13,036.40 and the rounded £13,037, and "up to £94" beside Steph's £93.28. The 28 October version's FAQ repeats a question and prints "part-time wok"; the version HM Treasury republished on 29 October (`83f8c894…`) fixes both and still prints "£13.319". ## Verification -- **Checker:** `PYTHONPATH=. uv run -q python pipeline/ab2026_harvest_contract.py sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook` reports 432 rows, 0 problems. +- **Checker:** `PYTHONPATH=. uv run -q python pipeline/ab2026_harvest_contract.py sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook` reports 425 rows, 0 problems (4 November, after review A's fixes, with the coordinator's tightened checks). - **Determinism:** two runs of `stage.py` write the same `claims_staged.jsonl.gz` and an unchanged `manifest.jsonl`. COUNTS.json pins the gzip's sha256. -- **Spot check (the rework's, 29 October):** 30 rows drawn with `random.Random(20261029).sample(range(432), 30)` (0-based row numbers), each re-opened at its locator in the hashed primaries: red-book and Impact on households pages extracted with `pdftotext -layout -f N -l N`; table cells read against the printed row and column. All 30 match on value, sign, unit and period. +- **Spot check (the rework's, 29 October; row numbers are the rework's, which held at 4f424b9):** 30 rows drawn with `random.Random(20261029).sample(range(432), 30)` (0-based row numbers), each re-opened at its locator in the hashed primaries: red-book and Impact on households pages extracted with `pdftotext -layout -f N -l N`; table cells read against the printed row and column. All 30 match on value, sign, unit and period. | Row | Locator | Printed | Staged value and unit | Period | |---:|---|---|---|---| @@ -318,3 +327,31 @@ The OBR's Table 3.1 "Total effect of Government decisions" prints the same figur ## README rule 10, 3 November 2026 Table 1.3's total effect of policy decisions and its tax and spending parts (18 rows) already took `decisions_effect_on_borrowing`, and its two memo lines on the current budget deficit (11 rows) already proposed `fiscal_balance_change`. The memo rows' sign convention named PSNB, so `stage.py` now states which way the current budget deficit moves on rows 138–142 (underlying forecast differences) and 161–166 (policy decisions). No value or sign changed. The contract checker reports 432 rows, 0 problems, and a re-run of `stage.py` reproduces the gzip byte for byte. + +## Review A fixes, 4 November 2026 + +Review A (issue comment 6635870934 on #231, at 4f424b9) approved this family with fixes. `stage.py` applies each as an explicit, commented step after the rows, indexed by 4f424b9's row numbers, and drops rows only at the end. No primary was re-opened for this round; every new quote and value comes from the review or the row. 432 rows → 425; `own` 263 → 257, `restated` 169 → 168; 6 hashed primaries. + +- **RB-1:** paragraph 5.17's two totals "over the forecast" (rows 265 and 266) sit at 2031-32 with the forecast's horizon; row 265 is `cumulative`, row 266 keeps `range_low`; row 267 is `point_in_time`. +- **RB-2:** rows 191 and 200 dropped (the £2.6 billion's third and fourth printings); row 179 dated 2026-27. The same rule drops rows 188 and 201, which repeat "up to 20 million" in paragraph 3.3 and Box 3.A: once RB-11 put rows 180 and 201 in England, four rows shared one claim identity (the coordinator's new duplicate-claim check, H-3, rejects more than two). +- **RB-3:** row 226 moved to `ab2026__package_pubs_relief_35pc_total`. +- **RB-4:** Chart 1.6's bridge steps (rows 21, 22, 24) propose `fiscal_balance_change` with the stability rule's sign; row 24 carries the government-decisions key; all six rows name the stability rule, and rows 20, 23 and 25 their forecast vintage and policy scenario. +- **RB-5:** row 246 is a line 29 component (its proposed key is retired); rows 232–237 are line 14 components; row 254 is a line 77 component. +- **RB-6:** rows 219–220's component says it contains rows 217–218. +- **RB-7:** rows 247 and 252 dropped (the coordinator's call), with their one-row keys `ab2026__package_parliament_regulatory_burden_reduction` and `ab2026__english_devolution_agreements`; categories F and E list them. +- **RB-8:** the DWP fraud key's scope and row 265's note rewritten; rows 265–266 `hmt_published_estimate_method_unspecified`. +- **RB-9:** rows 245, 246, 253 and 268 propose `jobs_created_or_supported`; row 245 is `restated`; rows 253 and 268 `hmt_published_estimate_method_unspecified`. +- **RB-10:** rows 11 and 240 dated 2027-28 ("per year from 2027-28"); rows 6 and 11 `hmt_costing_obr_certified`. +- **RB-11:** geography corrected on rows 5, 8, 9, 14, 180, 190, 196–199, 201–205, 258, 269 and 275, and "Great Britain" written `GB` throughout (the harvest's closed list). +- **RB-12:** rows 30–39's sign conventions; rows 30, 31, 34, 35, 36 and 38 are ONS outturns (`administrative_data`, `administrative_fact`). +- **RB-13:** row 366 dropped (category H). +- **RB-14:** the republished explainer is a 0-row primary (manifest line 6). +- **RB-15:** repeats share a dating (rows 9, 10, 14, 196, 205, 248–251, 271, 273), and rows 0, 19, 215 and 244 take the multi-year convention. +- **RB-16:** rows 143–166's `reform_hint` is "Total effect of policy decisions". +- **RB-17:** rows 185, 216, 226 and 237 have conventions `signs.py` reads; row 185 keeps its key, with the three scopes in its bullet; row 182 is 0.84 `share`. +- **RB-18:** rows 33 and 37 are HM Treasury's own arithmetic; rows 6, 7, 8, 10, 11, 12 and 14 take their repeats' source models (row 5's and row 9's repeats are already `arithmetic`; row 13 has no repeat and keeps its label); row 222 is a worked example (`arithmetic`, with its subgroup); row 183 `decile_10`; row 184 dated 2027-28, medium confidence, `hmt_published_estimate_method_unspecified`; row 168 loses `presentation`; row 267 `hmt_published_estimate_method_unspecified`. **Not done:** row 194's note should give footnote 84's "fuel stressed" definition, which needs the primary. +- **RB-19:** rows 15, 29, 41 and 42 quote only their sentence; rows 103, 104 and 167 lose their leading paragraph numbers; locators of rows 172, 190, 227, 248 and 249 corrected. **Not done:** rows 365 and 368 still quote fragments, since the whole sentences need the primary (row 366 is dropped). +- **RB-20:** (a) the unverifiable "234 proposed keys" figure is removed; (b) the spending test reads "the per-year amount equals the line's", which rows 224 and 276 pass, so they stay; (c) five more figures listed under "Other material not staged". +- **RB-21:** this lane folder's gzip and NOTES now match #231's; the 40 proposed-metric rows set `metric: null` explicitly. Table 6.3's duplication of Table 6.1 lines 45 and 46 (`uk_hmt_tables`) is left to tranche 2's cross-family dedupe. +- **K-7(b):** row 258 is the Barnett `total`, and rows 258–261 carry the sign convention. +- **K-7(c):** row 222 already proposes `gross_earnings_change`, the AB2025 spelling; no change. diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/claims_staged.jsonl.gz index c9487174a0e7f6b6817eb11dfcbfce7fb65a9130..f488eb74c1a40768c121f709aa2772a330ad8ade 100644 GIT binary patch literal 36306 zcmXV%V{m5A^Y&xgwr$(CZQHxCZJQh0+~AI}adOADH}2J)y1a1qJPOuG8vzG3-fO$6-7d%f*v$Z_#HfcCS9 zoM%7wG5-or)IDUUhFgn27`$=LbAkAFpS==6nlC=7nYtCf!;cM&*qbXop7%S~1-!mw zNN22Z^gPh}6-!&SST5;P1-MLoZVVJCp57aO#}F%i47^6-FdJ!c`|{xZKy#)5&4~uWR zZpv0^jlo4g=L>2Z(}>~_i{T)iJp!|WTHzI#t$e0W#PfwC6ne&TcQa73`fd+)9zWS?zd@s#Fkb+FA;}x4!F=MB6OY4!(P}TO~6aEKZ1?m&0WuUv;p1| zJQ`O-skOFnuZ3WXym$^YQ-EM}iSElo&yC6REsevt{tzFB7?GcsRIlqrpwqOlf8@Pm z;A&gf8o?9)m6N+g>-tvicGB~kP(zpmzU((^;YCqcxf1WF_ZksVT#D^t`l@;j&f}6O$3xgBM5r7!PPkr_*42zrNMi0sb zY6W+DT!6lpL2!p_x^d+aK^)k=#Va&v!@c}nrCg^BQueZ#ov4l)NC5N((@Mz>d+J@MRro|Q7PdSjX>+F7Jrtl-UqTvZtpL0+kLph+#NRWwRE@k8hkLU)r1rxR&rvq{IzAxd zWsNI96xZmM0a#cp|Ce93R)K4FU$0}r`cjxI8RSRO2`2f7D7Eg*Fe`Q|`lDy0S;eRd zR`maHJ;HLF~>tj3NiIR|{hCLxXT~ zR>LSS-dqlNI=~M9mB1s;um=6Ue7R?VTn+rXGbYXLa)Ibq2;1CEvm-i9tDer_kOV{5 zkzpGaz>pAz`QT3~Oka@!!3#Y1fkXWr_*gH7e{_5-!Jz?1v5rQs4g5kg?O<~+Gku)X zBd=xMOGHz{Qu^RGg`XCufimGDzLQwURGhFlftphUy$1H*UtgN!g=|(->2Ct-<*Tdb zPlcsi!X1N)5=s(CoE8|v>zFO-1EG~$!?)*XdZiRyGX|d;-xjXz~Od!R0 znyoMCx(j)IYO_JA{zl&Xy&U#sVhC&N5m_e`|>jxJ0JZI!YKxB4}g>Lx5pYT{N8OqpA4+g4%=n zh9eBleusbIp79J3e_XvvL&XLU6r?DA-0z6u9W{p@3+9jr+%vPk#CvQXUdDUk!?znY z&)W}U)eAy@3>E`@cx*h(k3OLvnJ(LKUisgNNDbm8d47R?vqv!)I&kEkmZ2e0qeQxm?ZKizP}v7k6hGc_$O4YC;fb}%6!e?@D@n@lHvLOMh22siXs z2X8K<9gLr_@0nPs^FACJ@m@YwzL*Kpjkuvv*mN^yU2}dh^9MnL^}?Lir|d`NF^sj` z9$;3$!IO-{>H#uq+Bu4z4Z0z#g-ZR7 z5Mga7bUg}iD_TQH3}jegbjFk)K1@3NrNYE0k6-Y+te3GM#?8_hddUK?aZuiF6MLw( z$T1^tc=$^q;L%oQF}ggJy)@NQL5B|_z9QiyOx5IIrNtz3-B%dVQk3R6O*0Oy=-iQ( zVRCMAzm@pAiDoQ-dTtr7G<#Cf6#0@QjXz=GBQDxF+1r^*{WLa$nMWyZ*PmCm8YCK0 zDrJMHX&K&?wu0o9`-vt8+G5Wu#I`e`D8j`B$|IDLa8}ffz(@Xpa7Y#p5&(YMA;wb> zJ;EfZ@>-w2;8EtsM`Kf_IU^1oIRf~6an;MI(`sJE57P_LDcodegM3zu88VWs;QHDP zPvU*z%HVxsqj$dqpD>6A8&*w-w?f zU6Zi$I9sI)%~nK-lBE&_c6UxOXOqAwo<@lZ(WGIUrm3SJswfEeL>+`CyUXtVsrNNR zOuxvfab`@ZB1u3CeQq038C!*?l2A3OMdW^vUk0g=~kj`IKk9;jXV>T#ViouTk%Ylp=XLQ>UXgz<7%)R8@Qlcw|y zXL!OKZU9b|Vi21;a+*4YnHSqiM2oE(te8MtSN<@~ec$*z7%RVDI}Y1CEIQWRM(Esc z1bw%=a2=s#>n(nfwuva%WgnYm=X4Aziu5DObQ)bak2hJ7bku|G2&sfb{z3Ne2L{<} zqOis3s8NIPlyM3G-{Z&>MQ3ohtweS}oDX`CTWNd@dam$&tVp^D?{Rf^fZV8?6kfeI)D~lvd zU~$145qicBAcz;Q;4mQs-_!^5{Mjz`dH8O(xG&)GE0VYY+w0@6*BA;jjwI0xgU}sB z+w6=na6qpt+|BC?( zx-aC=@X$NUzIt**U~FD!HaN)Y5J}>*Sg0QZ-GuD5lvZ274%aj)BK!M89dZXZfE@b^X) zFOXD|!PgnMJz?>0c{2ypVk~PpMrfUGchTP0q~4>vn#1F3+*A-uPgfw!kf(t>#Xl+6KHN(nTous zt;^#R;7CmfUYbG9=dF_Vz0&+7f(6A8OA&B|)Iq``D>-d1DCaSIg}!~X!hy1#f28H2 zqv6vE>0SbTw{l}*A*0l#7;K|HBSsJILyDx zGGG{^4+T+T2veY0&?goDdb#oaH-Lk~sRjPS(Wj z&`KSTaLzl^t=AiOg(-Lc4gP_>vL0-#Q(G(@-?x|G9(LEXEZ4weGeIH{HHby!yxFW7 z6}4NB;t%A7oB}u#2$C`G2f^sm7tF7PwlQ;}C7ft67EI_=uTpI3A0_-C%p$v=$VrM% zP?o8mLCP4-y3%VFZsvQl=sz0Ze_#`gNU|al(=b#4=)4S_aE>Q4yYech>}sELvD$R9 zu;Ui=B3Bw4=LK%L8b`d(>6Mu&7b;Q(-UzYBK3~CoGii5&OAzQ`;r>ADCoc06RSkA+s0N+O%{&)$4Xs;gUOMv00lLurvl7 ziKmC+IPhf=A*>Y1p!mRN#yk&Hogp1!rg35ig|}Fh0xU4`YVQ=gmB&>Od`LiBsl`;y zZsQw97VMjbVOcY@Ne-W~(WwnBpX2*l-A&4m)m?|hw>(|Z2|)_vF`?E8xtvu>7kFOr z(p{_64NKf$;byF$;Gpr~Rz*!k(rGkV4c7)qR8bt`^pPa{n#13^xgfvG<+RHKdsMFe z;>3y*Nf-)L4DEC~)Q?ic%T@%-R_Ih}J+Cf&wp9W*nebq__qZ0A8Q z(p+2S9AbROCM*d7P9l+HG?$$YB=1S|vHNVndZSC9*IIjX;V71l@azv+!R>r@^-btZ zH#e9zbW!f4rZ?s_ezYNiIAq1(x0`luY=`YhnMk{UIC{gqml3Y$DQDL4TvV&FPZ5Qvw`4byU4)NM@pzpR-$kG$wz(Ei4t#<1sw5h^;Gl$9^U~zd1 zAI>tHD!Ei!q`Rk5dt1a}$3+8xLRLCaUvWU7+YJ&UaT5Xx)4+c4Oh`k$-+&BLxP;f5 zaR(+a9%iZ@*67TZc#v2P$&ES9sFkijrKarW^c>=;Oo7;4W0*N8m#Txt)NiJqgyu!8 z(MIQUbVx%LG|;Mo5hDBz_Ek@gJ;tgKI~?^pR0$}StZ`6uO+Ns&&3|9;yp}XwiBpL{ z#j-}$;+d^ZfyW)mHho;SHnn25m*JISpDYRSU9xK>meNby2FfJsK_F)Qx*W=+CxV63(Fo+pU4>oA-W_i+* z0~;!dMvuF|Kx^N3lI?knqP(BCZGFG&fPKyq%;6z9T0$5e{JqGQA9D_6vagx7DL(*p z$_M1?v*lU`Z*S)o5MYZSsJ1~T&aK=}wzWz%*H^q8bvlKJU{^31pNCLi$QoP-*doAS zbk0Z#oX^7aRz4qvZ6XwtR6v%vG@~9~m#tFme?_J;?$VeE3YKz;-^A-ej^4zFK<`j) zXA`Z_b;hlz*u}%6;#9E960qUggpvfG&c{Fm_6}j1#38TJvAldO_xP{$4Xul5ruR-V zqjjn#*Ce=_Y1U7-L*0bzev*d}1Yc_!w7|`bl2BVvDD4@mMIS|&^LO~JI6M%MkD3Q# zXH&&sYV9U4U1Fjg&Ap0sATw7lrPn&7N=awc41wzOjlD~YrxP=XNeg{^kfrYU_M%MG zyEJl^LnONe-F%o#k13}Y8Qq{tXPS5%N|X%E!GFV>EKaEps_||jq@0HY+Yi;T?eE^% zs7f^~c4n)8OvV@EG`4QpjLpT^UXsuyixY0_%CAGn;~u@l|I6Pp@M!4T8C&LUn-aYP z@>}iI#_v)TDiIPCPg{%V6D864!X${cT$McA)@9%c@t|`WZyQ7vCFcGR%pNSUC}o{D zngo4@2k;$}9BJSVdEsIqRp9NjuRaY)RUn?XhiYb2fQYD)&_~pr{j8ANqDtj^y6Kg- zz(mMKBWq#pXOJ(X%H!B81MgNlY)SE83Nhhmv9X$;Zf0yD_!NaaEG0=Uq981fQWdf; zn=#f*8tqQR%|7m8-W$MHmJjWy;dj-#BsbN`P_B>aK)?q&+PTs{oBwC zpC68flOm%OnsuL_pm_;(#CzUs@8BYAP|hgENdrmR33L37!`^L_UkfN-BmV0O`E7LY z;?KiSv_$Vj)}hDdmKgo_p~^4~XbA47qEEzPad12^*lW#_MZ(?bSP1*1pn;Q6OipA+ z%U?sFZ#@v7Zn)+-QT7o39v-;z6RBdEK@XGc8WDTKw$`5ANHk>Sa29#vmI}AkcdqaQ zImYT`N!~A!0y|ygDY6E<#l0ZK$%E^fn*5CuKlnbB9r_#QPg9|+Rv);(+1ySJe!=Zw zo8-@uR*BAQFeUYY7viSv{(AX_V7QCjkj&lX1Q6^Zkgn_`*L^m3nfx%z@dpBpuRMdfgXIJn9B64{@^)!qpKXF+2e!@rHSI%ajzg=$v7|MTAJohz3=Sh-OdRq z?7@CO;Ooa$&Bam@=zX4UR0X|e0zrlBYq9yN`#?BS)XvBy%sX!3#F-cHa&e^E+6wl4Cv zw?9&gZ+dYU@(aV0k^udRT1=jh2-WW3{R6F6DVSG7fNuv&jIp)PrDX0s-8Bo|r8Th? zCVx>;Kor>>;fHYCwpQP+xEDP79_hJ^O|*|cf*Z%N!sL(~l$b`Ivb&;(W`+@-1<$Wv zyq265NxP{60JVTJ#=(Fv#^4JCAcQ+rx?PIXZ?XDhQZ~sa2T}&WyN(0|)6$aZ{>G}3 z`)9+(Oz%%cNNIz5{qesE`enIyThnUxg6&J@DyBIg!!v?eQOWKHWyHON$fy!}NCOyt zepc!E$TEtKii$Zo9Gtdx3rnrsc5+!&Je-vDE+QRcQ=3JA{GY-gg45d>+sM3sZdg1h zmZenI%@PN`O`o35zh9z;5`ekEMjVGH%r5c2Rz%I+iMURb3Ep??j^`RTs|RS(^r93U zPVXj#E&KgOW%s=4co>S|a+s?f?aP(1^a4 zrpVSv;(UC|0|k0fi;;a+djz4!zkTlr37)u=X?RG1_yoe2Ri1N=+E`C6$1aJv{{!8F z1Y1}xv$3Ik?Q@t55ic<>^a6E&>K8?(4Gfoi(}I?dR;7lB*H!5}{+j?^S{OeLf9!BS zdi;%i%)hwOiGFu{Hiz`lejM(tQ$6V<#=U6P!G2r&icP~p+{aEbmm>a_RYDCvKrxj+Y;b@|??a4Ay;4ki?M=5jz`30^N#Z|@N%Hi{! zBw!!;8d~%{hhf!soPYn6lSx;)%0wWzBxG}&1m%zTwh?*ZuvCfL5ACL%$^H@pXB%!fpV_6C^Ki=vDxQDdq0EKJKl0!IsZ z**RxysL%HQqM(L<8$ApDov^(|^dJeNtXaGzjX$K0_Ml8r&A;KUp&h`J8}kKEUjxu0 zO3*Yo?a!^++gWNO#5zFF4gdFNhPZV5dYo0kE+CYOJD~3|n~5$JUDtvwV}XT{t@5o6j|M#z!oBP3-)}Bs$&uEqSE-KYzt5+L zVzo>YB`;FuRo`tBQ3%^^+R;2PzBaqB`vZn%vXPLI)@OXfK&lM;(?KjQk*@8%Z_z}a zV!vq7qR^+BQ1Tg5nG(46o#j*hIWy@kx5H5t7#X{Qk50^UrEOVkLivQ^aod zfwnc68767n&=6L{4*Th>r!iBXdlTgF2RXFWq~~Sv!BDz@fG0OsU6b&e1+B;4HqOxj zH9wAGK`{Y~0RTi|>sjY*YVnjkvB5uyi?(Z_WNR0{Rbe83qLsdPXQA%|I~ePvjAU-% z!H{R}dYe38U1UR8vS^v;aWY8DXBm|CZN26@JXKS#RON>a3;C(u1 z-R&Q3=+={i{8bTO3!=AFsC6*PpNS|oZIS(KMVLkbB5>lqubDwBG)S$Z;ow(90-qI# zt)sADbqN@i2sW0jB0IG-#%FOQkSX9a^`2{PNyWXY2BlR2V&%RS&NlNe2Euzd&Nkuz^`$hY zXVD7QIj3jJp5vlULw6U`%BsFg>I=(%hWyL_8Hz_b?ko#EP%d3!NzYu*7gHS=1K~H` z!_;5adpsSTYq!qV>wcrjxF1UTu5*v99Cc0e?952%^Y38qVjY*!Z0I+sO3{~>YcHWyg8ye=DbNF8;tk@|#uF6+hM^t)Qp=o&B` zbOS|(s#nYIa4a_1^?m=AyP9MR>%T2OXoCr< z^Gg07()2&%_kRdE)_+LhTwN{1)9Wy`M{MT6HIY8-r69<4jtvB2XAhhS57xjkD?C1U zeU!>+j%FLZKp_ldUTR5@JyG_Q!*fs%6OkzRb-fr)0)&g_($HuSy*Wm=l%`V4eYEPo z9R=4pI>I}%B1vI0u%(FtXacEE&*vh_&}q45_BF@d z$i3^B*~4N6(n!wZ$^NV5S2LC#r%1YG~muBKR)lgaFN6Rc0nz1Bl9G@BfaCi+{())gI$zlwzl# zl!9*fUl^wQvy!t;Uxq1hJ%An_?CJ9Gw0YSZ~pzKBoMuuq?hpuU8d&6L1c zAWG?4eYMtH6I5gx#JiU%eL@yL;uGGbGTJ3{1^5#mnWcOqwks`+&ec7V0xjax6`IM; z%Kvxud5$;t|&V@-sij(3?$IqzhGSd3nnqt>xA9e zMs3|{htafU$DKR%a}|s?ItNeuzk(}JFI{wzpTAL*NIw+PW+ABs&dsQj(?t8`Up~K| zvjwcjXEUH!ACoh)zofdY7>GDF-`vN^Irn*x`d80YZW$vZKJNS0@dL8$0E< z^*7xZ_tubQ>WI&i7M+q0%-9eW&9RZX?7uWHpo0agG@}@r6qn~oM;RAh4z4eUTkQWmO_+|HnLF|91U(s^T-QGfV@8~bC?&SiL4Au4;Ftr?Z;8MXs=p| zDiqg*Z2xN22)ux-3_s?N?; zR;VhM8qqHOE4No>PCtMelB6xuJ&FB<>JS|jdvr4;rgX29C=f!@?qoL09Qlrs#G=GN zrlTDj`8ES3o|R5cGyd?3y}xlnc_N|eetAd-GnY6J{(i~)SB-8W5S^mzzqSiitr4^p6c_>NmH>C-O zUl1kC%|t%cr`*KX^t$ddq|Q~%q{iPOH>A$)pVp<4=KHmCqI*kpO(f)6wC4T>%VE`v zW67ik6PQy0+1_2fUT#7^S>^eT+168Cmb2~}L z+j$!NnW%#Hu5whK(5`8MB%0W~iwz(W`V{F}14&UeQb?REp(*g1Tf4zCb#4`Ps48 z(q*+PR6guP6;zgGr;@rsjkg3nO-(pn;vLl*-R;zE0biD7q+jYDG?*ajSC;E|3coI@ z*k&I5{FJFHvVZRbqV}YQPHc@AgZ&iEbs**IhPLx>ZBgf#eno!Xp1n@;B=M~Ww_>zz zGZD82c==RrNm6b}N=HPl(`%q1TcRmzXC4_00~X?A;{aSt>##(270R*|v&nnKPgiNf z;%5`Ap5z^p0u08#fC9n-O98nl0lDM=@v8#b=*R_Jb35y}QSqyvYuk@*20oR743u{& zuja`@uDz zvdNIH(Z2wmrazcd@7Tdd>#SOo8SvJ+`tOrijlG5|nkzz;6_+@Je!^dE?vVF`l0jWh zYnrgfLy4=aPQ4>BWm^lFvTarYFgrWhA+L1byV=2APt;nl#|vXN8fkQ{Qc6+)8sndn zb3OqeCx>18U9PY`^@IhQ`#;p=8msb);C;3k5>Fx#@Lx}*+oyO8Lg@nm!{{b$(n>Pg z1i{wq;Yzlfeiq2GwTwyY%1?|PYOL6IFN}Ht_Ayh$iFUK7w23M6QPJ&w%+YW^S}9kA z&08^X_H4!jB4o3WevnuQf=Uqm$%|?+mP$_CDOc$o+#WYxK023o4T^R-F<(~(TR*}k zZRu+Y114=je5FMz0v&M}2wvI02>-a0w%&LVPybCCuqVCN~RlCsYxgDd_M z{${98`3g?`@^NO4Z(UuUcwdqDr8*z!b>XB;Je0jL#c!&iOl*B{e#I%@G2a)qIL39Y zuSk5NOdMKTf_x>jT_6!lGJ!%r&>By2? z;}3~-n&c*(;cAWNoI|)jYm};Uc8(}VXssE_eI<^BY>idY3@*e83W#eXHuBG2vx-@g z&}Jr&oPBV%s835u*|?9h>1PV@apD5ELrj=3%&pnezGcBw@1<$CwPM{TY(oD>mLEXl z1yHTy0D(hmbQ8^P?O9X($IC2&1iE^;)KH|}pGgEB4w|HHNi%-<H4L|;@-n$rxD*rTM8ro5dATeH4RyvD;mGL{5^aJ3BffMY+0Evz^+HwHZ-%H^Enjc=$5!qnr9687H? z0D}=9Fg*>b*bgHuDOi7>$mGc_s_<%gfGzz2`_HzJ!!%FoRy~5n146@f zD9JDuPmmB&lPa=An=i7&8#JuAwD=z(!=aJG;vw^oP;pFtavS0)Es*`CPh~g^D;_ER z&q?xaj--U~ znm^>7m$zr9D zyFCQ^+hvt6`C|1Uz5cedDO= zVxca)Vtq$deO0vN?N4I22hb#ai4q*%*LT8@uXJF@Xzh?PXTMKvmCJ%W0CRd|i9hJZ zlp`H<6M6j`4#IK@TA@=A$uF?m5$8#Cfp%vt81XOAsl2Ejl%!(N7L(@_;h@`9V3!cu z#d11QBx=Bmi-FC-1Yag;FbtBQ>G)my1_HUH=-(r}u)bo6%y5eY|KL6?H_5T6W~rOHBWwT%zXb`Chyv4f33cc;%_lC{|j*<#@YriHFUu z!r5jk@$AKkMAzd8BR=^uU~zJE89y`GtOi2HyW$;ijMB!1^3un+^7NeqLU>&%)#b}9*{+I^veE-?VkJJ+)Co7B>g!^DATUPI}=kq47NN}E}3fTPsJQoyW zrvwJyh*0IxP;Asr4R5k=a%a`um}1@XZAT|I zxHk%VwI#kHiZ!Qa)z)R(1MWw9by-r)%3UdNcVvT7IhB2#xZUzWKvLC_DU;g}b=Zm+ ztOX)D-=EsHPDi{u!SI@F+q2hBjY_A8xR~Kj)IoT?PCy{`{;Bei)K}S#Y@BfaWn#>S z>;&vc&pxMzGCGG{W0y?>$a=QqZ3{4yVC**8CXR-gh`}JoLFne94EJU*xBryIZDF9i z!=Ss2me=V(p&!uiq`#pPM6n#`l-S|HWsrFi9jVkUf>8p^NbloB#NC|;ijC{a z@>?sfs)UYhkZhDp*?E8GEX5Qc!d@j)F~vfr(2!4n)196%Tg}*b+FdeB)j7g_qI}tM z`!gc3}yslzYk&-p&3aT=f zQCcqHG0h?x^Axuy9q2Ed3XGS}(fqff%CkWSXxa-qmS;lSKp%kb< zh<5b9t#B=Y8O5Y?@0L3xgS7`01zPUx=eAe*KGlG`zLMMV*UQd~P2UW`0zSLxV#sPP z1V65I2>%M(hpXeFo8sVROz>axD>7YZw8`P96Jfyc*v3d|#y3JgsEjsW}=MU4^;q*z0ym_8DjdTBbd4lf1VS#>MpL6JI} z2X`A%rX7-=x(WR+?x{Q4N8DfhhI>|wETv(Wa2bT~oI&CTE}s7I_pY6#u0EiV{m;U} zf%G6NUgCv{jO-SzZW5im}O@;3OoHQFW2>tr(zbVU30kwRl=n&tMZC{IwqV5#@N!gpnQJ(*pMP@PlF0cTSp=SVUjF8ayfxnYD2u^oQ=W`IM$OQmAm!z0B06ZZfAHDBc;z- zI>(s%Q6HM!b% zSUs$mm4F~yFTz}fW;miF%ss%*+djaSYXX_VIjG!kxz*oHDU=4ubyj*gjz z?OY(u6Mg}}_8=V=__$5~XmXTbmQfxIMu>)6p|}@~f!s*5;W@TI@9<+2TL#Meh!oG@lYsLdjp%CR1he3vJ1mplcNv26Yl+G`Shb=BvYL`0hmzL}V>M+CfjZgP721#_^e zgidkMcL?=24+gIUNraNyEID!;t+shu9q9ko-6;_`JBM+TSB0@3|DPIq68mcP7PqgV zHkBM2mY~Z) z-uhgwp>%i+Q`P`W5h>D)N$NznQ0wfV;|-mTOf4s@AQbCyc`A*>B-n1kmgyVHHH{x}vFg=dEt~m7Sa15>L zpg(x>QN%?C!|o*Ut8X>T(oYCIVFc)h?rc!)N6rl*a}yPd-M(xD$`Do>z0N9O|Lqhi zMG$2Q$Z!YM4oRzuG{UEaHe%je>1`RU7R~%sw)CMfEBiTVZ{;zSIW5z->w2de5m1J6 z6~qDuRBnbkS1>5S7^kU!OQGMWT$$kZA$uk3c`UpC=M4m@|Enz^vLdk8aazU7eZ36W_Wus*+Bm~SK71_)Z1#IBnFs8uK48+Sbw0n>Szd(o6Rdx zA5ub|G&?redNf|#_Dew6w^bjORr9M+uOD58fO*b9F}1aOCm7YQgp(g+`G;X4+`fNJ zh0+W#0JCYzpVU>IzAcM+)dN2bsJ+7P@tVCN=gQ+%qsW|4g0=<8?I2K`1A-BXyYIUVL!pW~`8uWzZc>C3OsDwcew@&mZ0bk@i~_w)(zu`F5_ z?#1jw?}z5c80+mD&u7O;8KK~T5qDsf(m_Z=p=u))cixQ-&%?BAD+6fWVQVoyosts} z*}$3K=0(K6Nc(pJ;`CF7@W*vN?wN_R`39GC))ylBRd>dTjd8ceR(%BZ_ApQuV8#6) zN<62=G#_a4eearLh(2wgZv~@}@$<`}ZZLNp;b%iMzExYsRK!5geW3z3N~3x3iVQ%1 zGdo7{U8y$`y~DG@-9WhT*BYa6pkxVqf{i;k!W9E3RX=*2_B$o@^L95p_73y}BzOPOLWzC}TB>tc#t&ucF`+yE=p@HH*o7AkSm`Li@DKDMplTbX!>H z;dQ{!X1ZHPcTEk|sG=CUXYOT@OS|R%8t%+-%h4~q8h?XE_G>0HDI4E8x$gUg)mRdh zEN-q(Kgv~?R!@KKabX^>=jxN`ibOGpl<8TZQ4}`a$_o8s54p->-}syE+RT?N2Am$f z9&Wd?NsEusAK{z;b1&JWk^aUJ~GZ_)T3-t4p7(*j~{RGD( zVHV(viEv1RhL-|8-mq+6L9ThbWr`;!Wr;wSIGOZbyk8V2N_(Ss-ll}5qFTm)L0iLK zdQKn*oG`!-3V``LRn@G9a@!=O6<>>|OE13i(&R>Cxn(`I=*j?+n~J(dr)^%=_6okw2&729+pfBYqV08L2YnBfqwYJ`*Bn+@E4xIg2oN* zZR`tUCcBgaI*nl)VjIc&v_k$ea4m;mM4p+`^LQ-U zHW!A!D57l-BAsrS*JAJ-^(GSB$I7z*Zh*Ydc-cLFKeLDWaj|~ zme|)QFn#53#$N&@I)BlabxA8`8~k-6a=M7G0d@!qQx z?M<&ONSQe0ONeQre%BI;MmnK>`4*CW_v2fpg}s_3@)1x&qb*~EbGZX9L|nmB*Czu3 zr$KN{TygDqeTm3)MbivXUH!8v!nz8g;DT;G9Zm8osRen-1 zbCUDWZZWSuw|2}=oabdmU`NPV)BuI%^}OpSrihE0zTE9;)rmvD~fk2k&l01b~SIMvqkZW%j{d(*7s2 zRPZW7vSwqzd|$caPp6io2$ofCR;EdyzZMs-sx!0DnB-T$F2sl;6AB!a5FE?m8{cJM z!mxZS^VH2Muew~tA(7K^MkrkhJ(LC6$Zw2-UK}mT&YaTp(3SIo)YWNopc-0PF(}n- z3UKz4gOu?gBIecBNO*idi%1*vH{cTc^ntc-T>Y<#`Z5sk$VPy0e_%yMxeyj-w94`fm97c~z0=bY z>gv>?8f@1fVIARS=Mi7BjsmA`JF32sy8|qhX-yR$F)Vl4Raj$b=c5DK3g7{}L_zDK zn+;2>wW(+rRqY+Q{^WD6C|CYiw;YVDyE!$XgKm(=+X4j2 zmOA)~6vf}P{vQB;K!CrHj#&LxSv>&&J$}d5g@#d<+US!TJE2h2Yf1>R(wSH|leo7= zxlpvShR)Uz8ai9W4V}jwUZzVA=EF-bi4M@U2bY{TNkQs}PiKCBWmguRH`1@2stuSF z1YB1Pa7h8h4l*_@KimWz>8|aO)7&19S!X7+JKswe2+V1FD_gQHH~U4OO`I|tI$vOr zUNoTx7>QqKwJ!%okUUou8o)SS*bncifVJpEQx5WWQ7O>4yi$So?QOtsW2{NEmqgw` zB$DQm&p;Id34xOclZv9yd{Qb@Ht_<(yB%PMdP-PHJU4XJi@pUZC3V#}lUFI#eY0`Yg-_Rx%&KwiRgSNq)zp*f-Q|38DsYYOv6eaEk*ZK=?%smKA+)6RO#!< z?;7l9U;fSsMGbq5 zuxD<}8~HLyt!~os2ouaj!Gx5th)G!5X4uvjHPiIBE(j7<2=>Qe@~d{wWTHB5^1}(Y zKi|tAe&!Qr1m}UPAHM(FwYtm$!392qt80Y&0u1?B791vChNFQzLMm`fO+jPO&u!-u--;7}FBLaWo6Rp?9m%@02D82M!>ChF2mf+Kn= z;{?@Ps6)WmdDviYVk#Oi)vP$AcpRDqq#kEf$Mh|{7kP(CHjz-e6zKz`VRYzUQ>yxe zyOw$?<4=APZ_l2qw`cFE@JT6F>F<^&boZIn9U!QhuIeWiyZBQ>rNF9&}{! zvb3l>q$k}A6qRzSJdr|7^L%}8r5_J(RmVPHOm1F=@*kEA28+R=Lib*K>jvRDK0dQt zuf-#6wOAIs^JG=yxW;jf;~K})IG&k{qsMZi6qXzNkJcbmv(y?As?w!hncG((*l%ts z0=c2uz7#@?z)6Br5j&JKBb_-LEZd5bg3GuHN!YU{4ow^Y7A|3hBeH z0Bt^)d!EdfHPU6-Q0}BCqUbg0x}+>WMr)vm`ZElK$Ut@~b%sr)QRF{ zFw!$v;W7Ur!VnxR+Ug*;3pQYi*h-!i`gL>ga14~MswH^*OT?l(pB@)Jjdf=fYk4g5=duN1!hVsn+nP-cgTQcoGbU2RCh854HM=doKIo!{j_;VoX}zKcN^^){D~hE$Tu6U zX6w^$6Zu)=&_x0?u&)93m;O`~X2Lev6!st#@((kr^CXrelIXwGaw{F(`Oq$){?~as zAt_d3fO?=x+3XLprS&FxkGg!1%2fYug{r?%qI$E@Ihg8?%ZTqZRY?uhe1W2IeSeJF z1J=w7Va2G41D^HBWY}dt1Nl$m(%)CIp+xg~tCRk;^O791vJ(ItWP$z-5VjQkn56d-3`{U$R} zzUSUQ!=tTe_VJ0sp^AF~ac@wEKy|!SsVmgb0MT2tZmav9fXI$AgBjB4?2*av7wY@>FZ!WTWqpdwQCC7VTD%U>5P~yoxds9Y|nF zswk}iYhC^NSLR?%8_~~2fyBa|`SXu|30OgUQdCF48zV`1pJZ_x8$E49E^F1mH-tlT zM;5;sO?;pU^mX8!a+(fgw+fc2A@6G4=qL4bVkFFCGy=n9`_fb6tu0=eEo`LX#D*2c zWNMajC#nJ0$t00%`-dvZmgBo#^%)9=x$2y99milzfkSLV{1e33Ux`n(a=#Gl-#j?`%0rzb|ci^}Z*2f`lpE-~_p_5SlYJ zXKK#WoT)iebLRG(dGrTOn9fZOnuj9I>Qadws(6Me!_)cHy2I!Tm1juVMr>L%|7tb$Uq9F`x>T~ z+eFB)3Dn{1Gus)fl+K4+J_NQCCkN$<2kAZPi=V`

HrLbmELAivSNBpWirfP>9)R z#d?DrRUWXHdv+2{UJl{aw8$|=tD-|TtSHdmNp+Z3sp>GvbYCKBF9i4{(5Fz&TP^8V z-@Y2HhH2Dd7|cY`b*?KfQF-1Ru&mqPA{bdPYD_)}cbILY`C<)U{Ngbq)5US+j-a4$ z1+}Q7w;F@TB8!uD3Rx!Lm=@F+|9$QN5TRgV(;zRo49*Ht5XIgjAy6J1B@6hgw3I%) zT@ObhslD{$Z|X(FTg9)l+>vPGQ$fG#h;dkp!U2fz+e5}&luH}R>-(Fo7U?>-4yL+f zi)7r`Sc^5p`Hno&nXr8C*q=rDT_RIOlks(6SV;u|c`r-d%Zl$)_TaF6E8tzE*bHJ^ zHJch&SESg$lhb;z@aA_!mc5Xfq~B*vc6dXy%IWN;1O!rLMx%EiHQ@r5f?w}l8+*lUfl{>yk5Ri zlY=B!6@=zNLpJz|S2|#uKwPAgED71I+Tp<((H^6A%FJ5|uv}V<~ zv*;d81nkd0{t2RjU;DwW?a9^vAxPRM&R+R0pgv%D1b2EnwQn6ybPo8@8B%*t6tNc| z!5@Qp7}e6{UxWw{GvQXEC)|Vh-LlcdamN5Wbn%3{GfF6$;ww;wkQ2bhu*agj3tkqE zCcf|HvCH!ZAF@$u&&`~|<|)6pJ}NDI>9zf>Irg=EKkmsMMX+>dZzK6~QQ% zR1WuGS6G*6G>_KiljFlwQB63G4W%ss)B%`W=2F@${4YLXI>#D?q_~EKe2rTriocUW zXFLhnlWY^3Cyb;*_=rTkmsaO76I1q-J~iNPS?NKB79~BuyX=Z28g$|QfZAg{e?cE- zRpe_dB^nBC*)U-gW^*=+4z1Q=Ca&6#hQ{j$n-!tIkmN!*737vcV{J4+o3F~iRyhpc zzP5zb1OAPhh>C$wZn)9x<@b7r#gLCc)p2=4EpMKUI{v7k%aLtwrWNv2ago1rk&dmC z#q%|KV_bKZsybTXFbJ$*;#zog3Nbhgzri|22e7bayj-}NRD4zrSV(5IfQ5+l=O6#- z7zu2d8nz0Y#zBM~QIW{OShJ_>&p-Yf80r(9i-pSB69(F;56USnHj}k2lJ#U5QVP%( zS2`t;#6!G4znv`lypT9QOp*IvsNcV*-;27t?CQgd_wTP+BFECNCNy6v`PghK9%ff$ z9>5GB{Q8>oif>;{BmOa|w;&^nzwe0O<4=CaieESP_|ETt_>t;A;aQny6zpt|!d{-5 z{T}xl!{Mh#zxD|wK{_c7yM84WD!p#6l$)hNT%mQZM2tqSdCHpiqQa}uyRf>caeEDJ zzd_`QHN+a}lL-*CY8_N0LEdWc?*aT9Bq*}ypb$myky$gzUU2h2|DSA$P=g)GCbkod z16)t(*Bx>ji;9M}izZcJskQI%*pfx{L*pJPzA!SpB@FI}fUeakjB~VR=KymlQPXIr z(N3dX#b}p$xg!O=K?%?soCoL$@zQvw@lNAi#ds&*hjN;Ifh_U>>{QE-oF^75?6CEzp3ogG3Z*1lsgl(d0Q|SFJpuw@;BVE; zPLuitO2=g=J1eM$3N`kL&!SYp3(?j?7RV4uV8RFc;lm4Dvxtk#b$d3Wc=E(0TROKu zQ_C@>bV&iJ$$W!0EqeN3OtRpLLccFNcSo(U6k2N}x?#v=t{zmLBRBf@iUx0a}GWdad%G+_m zf2${mjGl})2WaKg%fO2$1Yl<0W269a97rXMbIwIMad;kRmMMXb1vxh{RqlcVGapWi zc-j0HJ(953UhxB&SV2v6$3t|5>l?&k$F7oXw6@T}UWjw6iU%jeo|+ytJ!*Q)Js66G z;N$G3JIaG$ai%=wN)06{@8p0!g-2sMqFnUfyt;a9femaSgQ%`fg->Eo7c&0sn@Ac`imHoI&CmAbR-ViCRjRX$49f4=AOj{!B{f~#B%ONLOsRSZc2bA zwtMriND{tND>?gqNh%|z*g>)e%p}pq!u{wbAy=gxr56IOC9%xtF^dJ(RD-02N2e7FwPyufVZ}W{}@yh0W=$(g7}Y1t--%E`0p`kMx&xe#bZT9>cywVMWZ|}V9dT| z?AtfW`1Xy1`u5{$-07O%>}sLrx3Q7`{W$8X+wRDp_^vgcM}R_jVzWdaQy$1#nLDYl zerOyv6C56B9A54sr)rifQv{X*U+Ni0I61XHsp@9$q~NH!X0prAb0K2v6(#lEhY_EO z+G@AlgJZ}%On1Kc_SI;{;uCK_|0cG{fZ&t8&PP=_ zy+-sAWz+*s&r@rN#an&vHpE%MyNIdGBZ0T7^jrR{h_?>qv8#0b^OS+K+WO~%s_h^~ zvstEHU3%pYh4;DAw|QIZ3UPNY8gFm^9$XE?8i-E}h*!L7mAdz!Ogp2vTt~UgLGuVL z&x>u_);SUbxO+IU;63Ls!4BLUeV#lbHJJJ*}V~?V_YEtn%HgKay*p64j`r?uFAv)M* z4Hg*W5s*h37l!w;JCI5B^#KiFWpD9{^lwqO3)#r^vE4nwFEIiKdGx2s4l>6xcODoP ze%*3lm_G%TVWAS|vRG)8ZQs^B0W^HjG7xZ}BH~PJJ+>%i)uU;-82XJ7G(m5C4{3U^ z;O8S5-=e`(gXwvIY4%<4RSyQGIX#Bdev38EPMPLrupEav*NbL0sk4aM?9;g$IU-Me7Htv3}1NUnt&gmM{(mNw}w*qnQYYBVY1 z14ZOw{G>gakVkef^gRo%1mTlA+Y0y;HurcZ#^X2e5@OA0T^p+1&bvq&URYZ zVY#7gB)IG)2{n6%Lal<@%7J-i2o;-al>}62t~Fb$8dk1NYD&Lk4!Torn^Xi3u0t;& zZaDH`Big~3iMZ$b%z*}VyD!Rq1y3pZyl3wV?$zp_0V z-bA33KSc4D0mLcB@dLoC8;-YC?8ER*E zb9)CT_skW=l!O-tWPF$$%Q!Wd4Bl?5Vz&#Z*yR2K;IQ!HJYdFwE<4hSMoctUnlg5W z#hDZwbvTM6zhBtir>88o#;4DVncqAKW!FhmxjoV8&z{+I90D;?ox4&3=YW125?sSS z;U~ZTm1WJWRggrID@{GRB@%~EV({KG9Xy!vQ^3v>2jvZ6-r+wZ?0*KM_k1z1Ex03h zY4gdWBgx}Q{xx^pg_q)S5Ey!&BnGNQpoYQlExUy&p$vv`3^L@Wfu&0|3(I(;p49Ti zu*QBFMg9<7TQFxQM_6V;SY1t6OCHsvg((jgw=|`Nt!FOsmqu`YW$lE*;sb`0Y~coo z_ZrJlR?=gbo6q3-k_QbqybFWYbc#%J&s zl}7u8i9iEnvG_o=m`#_8{QUI8%OAgIKLC8eQd96cWk+!ltoh!D$Jw*37)mnE7#V_= z(h~`cNf9Dn(3T1uqM{nlSPE~%@8_~7!9xMv@Zxw3^Z`p2CQIBpES8Hp^2N6(N%T%2 zP90NWyvVTRU-)QFwyJPGMf6ez9h=3<2rJmWTNV?E{4^0<-4YRG-$VX1yxt=JsOsRy z6^s(|HyRM+{O^KHS}~b)bOvc&2I+GzR&6l5*f6@8pcw*(xG${KU{yBhPeS(YQLp*I z#}&P;p&x-@obto~GJf>Y?)-f=ROX7{<@zCf$vay8%l=I4Ny1; zH8p8!($sWT)YRaY`&NBLTU+Hn*zM6yPD&F$FTOYTEAd@g< zv9pDtjs!0tX90YO%AW1r=MmL@Iq9LfIdWR+wpV3#t%?51h{|!Em);eX7JWF-nwVe^2{isr$PMjha-rZA72 zV{d*O!MhaC^jjct_bPN-K2g<5!y<=R?F}< zqNZkqK_@}vVFt?ZUg$RPxv13%>khcYy<7O$qr4JKFX6(m6Mmv36uir43Wk#gGoltM z2FfYqmwf(Ieotnbs1)g#67a)UJQ&{Bq_=2{C62_cUjzV+zMC7soO&17sZ8!vc zMoBqfINh&*t-pnDZOSkT_ymin0t3GmnMxxPhmV0h=Il9~b12S?UGXon>Lg0eD}Rpm z*_Viua{M9wCH3y>ssiOOOqvCA?-`r}0FUJ<#Y<*sWsqD76?0_uqMx|q0 zBY0V=csHiCutw>%NDnTPFLOJdSuVnd#VKulRmRLPeR3J(oLt^ZQBBPKuP~la?0?iq z<8Y){nx2I0-{bF~l%=wV1P-3);?y*`qe<76TkT~Fnc9ZBO0$Jv^w zdy1psBRhPcf;N(cIKeL+x+*5vK}@iQP6l0K*tec6ek&KOCTWaZMo3o(eOO$3jB#kd zhE%go$P%8jI6?s*i|Acgdf==-|M*vu;%Nr}mLV1lflbK2AlIZ%O_-wf zjw8Xk@aCbckiD>#kzeM-Q!KHm ztYohhp-RJSSAPg<_nT$wIi)7}fc*Jk{y9zI=h}XdR`MPArsWzCAPFgEl+O ze48>~znw6jjwH`LTP`p`n%uBP613f)v?3883&QU}{R$dj4ypEx~d+dUYtCEHSb#38_ zCNsUL1jojrMWyTCVGYLV%=NJa?M$}1&2WTl5G2BwMCG=2{L8p#_$D2D2lZBSrTJHG zyWE|%pbHFnbjLY$=mK}IJBC!jxJ|EDc?JmV?77V6To(6PujQMBZ z{R~T)ig}YkGus7FKIn!lQ$Vk|B7hRe>5~QP0~<8a1*+z{PQrB^9ai%|pRO?yosXM|*jLObtri z0UuYE>{72&DcQBFCcExtT$g^ov*kKrq&|D1%X4mOmOERP>!nz(r&%tqYDgKTTLg*q zG}rE=XbyO#0v1~uKt&Xm8fv8!)~lMr+MSIktlhZ)3R|?(fg&Z-lh2gQ9xzO5UOQV} zOIm{%U2U)|e0rObTPugyRzt^niZgqaQ{0-NR77^EL03t3-O7!VFwl)!l=My_5&N}w zt3bPWe25rqHcUj}!cq}=bV&PiOhW}&g%`lg8}5yTmvT+?#yu>xb>jqKWKDcnK2dSn za(&yYsJkk<&oN=77aRUXyFuXhTQqw(xP#-i0CkW>FSI33+b&@uUUa28L7AcxROebG z0Es9J5XD;z!oZD()elh5fX#J_rR8O{5horU^F8+-J8(Uj0M)=ug}7&D@LuEaN%^Ai z)#Cj?rG0lAx)W4hC%QS4WElR!9jsU6RGY}TpO>H9T_%au;nW}aq{Ro63j4A+A$=^mP)C$BkPo zw@BjRE6Bp{xD8pjLAya3f1UbG4CkoBre3=G<|tBBw~V|2LGhIDdyM3NqSJAgaZf2_B;ohg(7A`gS9^Z*#*SqNLt-)S{ z{q|s=JUxEw(SUrc-++dqdRXFjj2*7?V!${#rQy(ksggoA!_|rSs10HSKTPKhud*;|VXC|uw#I5H( z+TkZroB*%;XpcToV}qaMiy!&{}I^u`19Z5N51}jZi}*1dAzhR6>BzB=@_|Pb<0)TqT*EG(zV$;d$&U6x?AFY zhAPA4fGy86E(yAZT`i4IG(v}!44Kg|nHDf7YD>yPuXJem--jj#{Lk%J-ego2(Q>ngOPBm`ypvbTFrMU@@=Lly@9OSjpaiecr+!e0h$_?n)u zVA>lzfs0%4?X?)&JE1+j6c7P3K5KN^^-f+MZL*4q__tZPE9n2_Nau4AL#Zpl5c#LsEDJz;v*jH4On z(2Vmg-fb%xC!-A)<{R17{q(8EY&M(gYr*IBDNe7%=~se>2MYH-4~B< z6x{k5A0VzNhvRa7Xsl{>)IfHQ=B-_H_9$Js!R+immW=wQ1Wp)h=&ib_@LNU9a52!2 zEo4HptYiaNGuQSiZ~jq1;O#!!a1440*6QTRJP(Y*q_+ujz5~G#E_v0xXa>vpg)qXs zXWhg}POT<}_embV=xG#4KPZ0Q!fgNcIy78-vbu5U)_r&pU~)JQQ(JMj(6@c&<6;Ah zZVA-uo`|bG{>V}=5=UeB?AK2Ldd{CMXrAN=mzB7F{p*u32_aQR>(+H{*FNeXc1Ldg zXG%*19*6(o-q;SL6+ERf?%69p7w>C{AzHQ*DDd8iNB+<*@O}cox;R;$vWrEh5)IPH z*dM0Yh6uYgF7Fqr8V16{r0|3+CHT2*62x`#7wJpaJ+Z8>xU)XRd@E;hYljuW%9IG( zzoMCXJ-nM`dmd=~qqxmQRuTO`RSDoo^PYcpfJ~`CB5H$S#k#4=PqXk8j!9qS9Z90o zK%T{I_hto-QFa7}Dc6!aShQD_uL!#59N%*`IyBx21-TNujWY$i4YV%A@C(SvuHLq37?%h%!ic!Mf-DS&`EOx#Utk3=EHln#~Yafk|Gw#50gHp$za7=M}# zXo}%qocZ%;v?>614Z;*AsswFfq^Xv!;@w!ar5gmIEMX0fB)+M_bn=G?u_Vd$>*nBL z7pxpV{JsYetP0rSwd{&i6tiJ;t}AE8N>3jPyy#fCrFTMv85Bx^O9 zmNB5|c#c(&aZ5N9z=Wxc02t^C$A4$;jMl-siO(=_Xsi;|(oZ6e@_cHu3fB+08ms7r zAjdzm@5R{R((lnTgx${ZZVmPeP_~U=MMg*`FR)4^b^;cr`BfsziV9+tBFOQb4UOiG z*s0*raCu1%S#oSfiMpQ9+T-`XB~1r@hve!nD>JE(wao*o@(ApGkw-2$pI{G(lu?PC zDo;&@-HJ<}Ed?d9c_`iA54ii2iLe$Xd4q7_d-FH~K-L4m{fh)MpW(*%_O;v2oapeZRuGA5*!p}lNy-i4sqD@U#^St| zjDs9_ksaKMfU?m%TC$43UCHKLzR)5 z1QO>lZXw7k6_LS8z;_j84wCIuy{E9-ULRnrH1iC_=E&2DodApSrAlT_ovsl37i_*R@2r$ug5B1}Q#2UHC`({PxR!I-URa$CMT0%GisIId?q^iRUY7b?DryV2ox)*qQluE&hYG5V z?x6f!Y8zF+MKGD6EY5!dbw+c{oFpIu%27B7<>EdBJ&8mWhq$QQ7Cq0sfVC6FVn(3P zfk=9LK3&12xc6lE4lNX<(2^*=i*pr%QJ^F?wC}@}OY>~8c+Nmo%&B*fjJXXP9QLlP z*HFs~7ZUBr?J1IL+&@}$aSlZugm|tHu2NO(`a5Ai)6BA;k+5G+v)@YgTi!zjPG!lV z%zhXsC`e9q%aGy}+qQ=jk;hq~!eBX_SlHpE2xV1g0iXr}k4BNtB`O0_Qd!k0T z=i8}$>v%#%`_UN&6kP&xjW7B!nCDk7Sb-#En~|qpLB7{64^1HQhjq?X@0B|?i{a}j z4{l+Y*;<9G(py&NdeJ8NE}j0uvWz2xN5z38b;!u3`kieCn5xODO4Y(fLg-tll|cBp zbP_#^vsXJCiee_tF;hFoRT=VYN-9|U-M93o_?qlqVG{$1^S<`NhCf;`P&ntp)Axx zkshMs+i!m0!jZ-HJ~3A$%GThDR(c5PNHu078-|NELK?$gvE?_8=nZwe;&k|VmMrZlgI#fJTXQRhtqu9a{_Mb%+3r#&*4K5%`|LYpa*qpa zv>4@AU2kAz#C^GfmmlssI`XlaqAbNC&evX%@c(;zA@PstR! znte$`lw847#D813JVXNPNO&?0Mj)48tXq4gYBTM|U>UE~5Vqf^aH*s>8Y5)}r4#XK zC~IgBSUl$;Ob32FPA2`n-} z*k78{ykp${Mo^BNQnIrZ|HN|pl>MYC%TmmRUdr7T$-j7nACP@fQFWRht`& zQWUP)`4iI`_m?K7b0wz!fr&{8ohGJ(6O(ar%%p_p{FuoocZQ<1JAZ~+Q-ee^)VVU$ z;J^%}gibTm!5OO4IW>k-!gGEM)hQRNL2GyZ47H{dq-LlyWvJ?ERBp&nO6W909sUf} zKDk4blAbd>M77JFqS}{mjA}LOEkl#kg^`qTSdxl`PLtH(NveBlBo#}~8In}DEJ<}Q zf~3~8eAOg%t|Vn1n52}@X_7iTNp((+q?GiWAxU-0l2qp?N$Ncb6H*kP#-*aM4yQ*I z{<5@X>gc{)+A=k59X`6x)X{y#Dl+PZ+)KTTUl5c_B9*gs8MRnyk_ zh5>5Yy7+)VOp`(s6w-3c9waW=dVkM zGSvKa1W|^jjxyBzRXKm1Z@}Ss@Yhd%05b&B;NGjw;8dw{Q8*Tutfd1FbD6BAne6a^ zhbT+JuN^@+V@roK9y_n~jnn0|GcKgqH%jvw zt>p3X+A9JUHs(V|5E|Cr#7TVj{6AhVvRPY)H0H8dTeI2WLmJyUq_HTQrT2Kk8{;$- z*g=58h^!GV<3hlrdEjOELKfb|pLEkDfB~>)a9h}4z*YHms)7aAJsaD9m ze%8nO;9>;OTHs=I;9_h)&^nDXjh&2&FzYGGdSF}x%35GtbYNTr$}WpCea(wHpzQBq z5VMd+g=A|bU;WJmGyE)t^_XTZjEj{tb3wD~8y8&|7whjTyzBRG*$)9p?=)IV5nuuJ zhdZ7qq^+Xu=&Ql)CJd$XJz><{6Rs1a0mP@BRFw6D{o z7I+$6cp5uFGpqAB)7RK&DZW$M>Mhp#;A{leTHtJS;cV;#);zc4Sku_xsA|{udg}}6 zgUbIK2{>m1LCf4ee_^`tc$n@8u0x}g!#`O{{Agb|6CGSoZQhrg{lcrozd@d zk38=ohg!jR{@rZ}e78v0G$1<7i88ZdhhP#xI`%rGMxPOj=0Ozmk7fx=6M5LbZ0YZ3 z-jmjWD~FX*m;Faxnv^}TRf3c)uvL1nRf3dVBqeN`AZ1@Hn;>NiESnxIn;>NuNeK%m zNZA)VC`j1?JE#Xcs8*zOdxJ`S?WL}wz0?IVdtx;OGFxCZ^|qQ)6qat@wkE_sIJ2$ST3}LrDtW3n?gM;P+qtl}aht%LVThxiF9(Z$ici zH9G*)iRTtCaxxfEHG3lQT6F4xFBNoZfiKmEFZIfFGS5uLu8t>t-JhoF^vT5Zsb_Xj z0ICIcP#<>CD+8*7t^PH`3V827WfH$_zABJD7VG)C!M#mRGmP%?VLgV zMB(9oh@*gR^(_l#@;E^(;k5sZCG$D;kj(?b`w*VJN0TMPVtL-_q#O*YGLyNWh_y@J#zsxOVhw5?xWLBOpAUcn)}f~|&|ZFY9vrmxeG z8b$FrViPmBet8Z7w-$H~hwvP>d*!9sZrG6CR(w5aQpDncZ`yOyV>rsYs)BNGXxsGm5>m)E>cY1gm5v+(v-g^{tX4SS9Pi?L$Bkb}J;{ zyXZE`hy$-k9$r7958Q>HZ4%ytSinq+gn5-M9;yaE%6_Ze|0b2>XRFX3v8ArZzM)o% zMb@0y^>ZCs;7tg&{1FG=8fVc@&g3AlsHZ*{n#`aNL(HCR)cIwsxt0?f+2CwKL-e4AOG2Xt?fo^!k zcz@R;dtmGY)Y{UG=)L@5fC)C8o)TOOp9}DL$y;$C1mqkMy^Us8HvYLM;_}2pGqzUI_ zUH);6;0}8(66XYB@K9_AN1`G7z4&w|9MqDPguCo%;e9OaE@mkDLqv-KX0S>EX05PF zny^Z?24TIe_G(=##D9vZ=a*$?t;u?@TBgmvsDGIcVZU-g$oJ|K34KL2~B^r*Lni6ddoMEFwXpMa?7KKVpqVwo>QKHY7 zPax57m^2{K@R`(;XnSzNt^Z_+*4MpZtNQAT=)%8o4TwRQr1O6y^_*_(IGPq0lq=uZ@-R=0ZE;F zL$k^LHqDX=edpXerCW88Ft&7e=fvI_TEMC{TU?G=1$#@ESY>*yhP2Yy>f@;vlRXz! zA8-ripU1cb<4>2kWm`JVvh+1UQF|$_4q66GP~aA9QIBy8mZ&aq%kp*PmcHgNYA;3O zfe&XGxCO7+W88vItXtf2HRP7gjyYGATdo1S9JmDo-DBK>d9G93(y_j!v5nADt#4VL zetSTzHwm2}oQ+QaA-78(dZ?*+DoD{m*Z4SEP2=rMUU4NE^CpDr@G@5MqXvKxQQ0}Z z3!uz{M|PPy`Z~OM*mMdk01H~Nan%_X!07iR7FgazVW6+|%~E~9$nx~t$hI$;PuN-NlQp|LA(f{iT^yTO*S0EhD^Ve+udI`Cb@zaoG8Fa#~y_){hv3Q)@I53C8fVe$9+@1Wry8>x>OL~Qy=GANViR- z4_5cAgrR5A#~ixG-1Ho4IGek+?-`!u&%TxyM&r5h{=y#e=KZd+KPYvn0aNQy(&H+% zQ1sfAYulR&T9wgTB-Z{HYS7}>wY#9he#Vk*mQoiG*Pq|i4bRE^^HZRUZj)?T1-7bY zem1A5M^3{Me?;G`$tY@&4)N(sf%~d_qBBz~i!V9=REq#LdIV51uk+bjI_CP5H{&>4 z(4E{ayf4zs@9iVnb6CLSV79_$+-f29~XEKkGZ zxB+kN#T&Icb-y&bD&Wjr=H!1+nmHX1se~Z9u}~&3*W?wG`o_|AxDrbS^nG!Fdq%;L zAs64HDS6{8`j*ZRe$H4-!HwB=MSNh>?S{8e-X8}!fB3SaxKTopynuJ&E_1Y;U|8wC zpU#{rR8YfH9HVu_v+0tsFsYl&qfJVq35sl(33gau$8*?G<^|88Ns?uWT!C6fvsoN3NtX2vGuS0p7!1suVQKk-RPAkaUa2eKodf(Icv+iq6!D-@bdlu}0S^o2UU>E4*$ z6WC(k$jM0sTaFp>k1Yj8QY7MIrQIp3`N7jY|5DK(iX-sf2aL-ksdM>{#RWA@h3|g; zVZh$YjQs9?70}#+u5@xQ`pI`8HATjAk&#lfpOei;0goTTm7EB?q6~h`7L?uTg2t(02sV(t zjL$SEUZ2AIa8~L=1LltYu>VFrgsbR?MIXHjH-{ZFSMZAd`KAG=$Sb)i9bKHQ$n+{= z$h#big#*g%;F{oYM>WAgt#Hk?wD7Y)bHHX6oeK@>|) zic2S5;V~9^nalJ$R+6$C@u>);$XIHm(}2k{NS?Q1x>DR_0i#kV4wU0!bCBI-zhk*{ zvRk4(UjSo=0Xc~Zu0Gr4F)X}CCKGh$V}9r3WIit4^+mYhN^XD;&*4L`rsn)CMh)?O z#R}7*dLJwhvtSddX8b@R?vdLy-^8SRMTG*rz{Ef(is%`+qBOaj z^q}!{T3iwl9%uJ!sR%5XZR0GhvGgYH&GB@=Lx&A?DAC{(p?w2`yAc<(_RJ#EaFRn z`xK>yO(S7V5^{Q%(e)NB!zJ4jwrAfeu_W;=afM4Q>6{5PTAbZ{#5PTT{1-7SmU*O_{QnS8G{&JQ${n{<(S~6@O6563hP&SNiv|vs?1t z1gnp{kITt;Nb^xjl86S@Vtw<$fhWAH$*~Rpl?dlGpSLDN2@uefARl&dGUSffQu~7u zPi2H@SOdo}8qnENb12E^elYSubgDyf7DM@9T9owItKWXzF=$Yfv|nFVb}2?qZQU&b zs=Mrp6+Dwo7m*AWB)Gp4e4GoA<0s3p>o+UM$4#S~G`qfnH0Qy^0tWP0PdzRt07Ur7 zBJ7L?RwE)@ZV%m**BM#$l{b&5_R5PmGOs4s|a&4I|0*r#*b7gr1Rc?zak5JBQ zQiMHO^G7xBwB~vWCJL_~7K)c$wSfAzSADT$(UcO8+=fQO{V3#(-1-CIJhUj|L5g*h zVifTj(ch+5jCB-Ug+=0Ctr24^3;{;{uEF}9Wez;l?-c5Hmg&{6-{sLo{SNZnH+e?& zt|{ZOIq*@vQ#2o&zJmE!1{l@5=Zf+WqTCiy9-?~JqzJo4Lu(fLR_|P+{*px=PE_xp zzWq>NELb$=qwrC^Q}I;z^{aPzXi>d`6zeF(sNOZ9zrKSf@E#Lg->LrqULH?W?;yZ^ z5nv348?#b)4ed$Mlj7aTtG_2j9$S>|pv`^LW|Zz4D&0}2QMyws-I<<3=`Ig2N_Wo{ zWgnv47E$(5x@%H|9dj^3>F%uNI_9Xhf0*t|MV|6pv}G1X4LH(+E(7YgsnpBx}C4EYn$JXUE40N$PN^_EsE@*Zr7v&TV6w}75dig zEU*4zMIKJn?V!H>P+u%nH2CB*>UJtmK^rPhL8ER5Db`hrQMYSEe@6{373(OvBL&eN z^=Nq*V_^s|>UIr2{A>(7)a?`=gf_hT4?@e>qHYIm?wdBFZr7Nt!b9Co(OKcCI4eBV z?HW*I8;aZ(MYd75YqV}>HT1-@#&tWZ{u9o5I8nEQ`u0P8u~gBR5|2=~Q&DAwC1h)a zx*eogTPa4}u93Q(J#bOCd!%m1-4b={b~2u*+d+W)BET37H}>$e-SEpsb>Dr$uD>Tx z9$VDypv`^LX4LH(vQ=1@cpO^mx}9s)U$4l+i@M!&71@F!w?&aH)a{y7U~`DN-6M)? z4(r$L@^GSV2lefT`eLb~vAUg&_uHw|?dvoo} zAi91(L>FUW2r%k)O`9k@?31ZvsPGgXg!ZscCbYSi+KjqgL$(Uk?UbDr3UxaNb-QL1 z*@PmuMUhR^?HaAySpySwyYqEB%dB6wli@_&4(i(v^~F*}V@m9yZl|KkUi~HHJhZ6W zL5g*hV$|&#aM{)|fLyDxMaX`TKVd)7N8iM8au;x6c_^u01QUK$<8gitN{-LtVE2OqwGAjFQ<70? zy{(LYO|AWjE-3uGjspt*hRu>N6_?M|E{o>(;)27~EqzM2L7ieof1wUceJ|Xsmh>eS zlDaQR)|B*ecvdgz=-ZFuU>((x%#`k&$I}|H2_+3li928^T^hdtW>yR1bZUoF%L*s1 zJ9nl-%b(hYIhlI?#PEW^cZOEzjC>=U`=Lpy={e41Y7ebhFg0eAiRIQ7#>&=a-xndT zP`talyF%OcVg9-lt@P_WBW@@oZgv@Qvunr4K0nmH_bZm?``~nb%FLz&UD0;};xBao zu5csqZedpL22bq*1Nv#UD2~kc!Qn@a=!8i|@E|3i(M*9vj{^`O1Wwu?>ko)EQe{RIbdpcg=fO#R#)DHZ$JMXzI>N z3VFjnXUT-GI*32WB_~FCLg${bqRDc}$%9XP4EZnk0qH_k3H5$T0^W~hyNc7E(5fsO6yz~k(HD-XmJ4q#x~DkSPZ04HTN&4qX0;AQf*|9F zq%x9==rv$u8c(3nE{R@@N4xAFC)-Ta2soQ~T2x0(&OXz_jLI)d2J0@_e5BL~KL^Wo zETWbF>v$E&!~d9t>qWEED6wQZ&Z`{s*WZ3M?SWx?Zyck#(83r0<8a?ssi(}MPtj}} z#HA{U%%pS|lVG)?FEG(`JZT?v9YF?)B=2|>D!MO`6#v);gu3j0UoFYUx1tsiG_*3v zKRRfWnzOl|4_b`9;<<8TGv32CQ4qg=$ATpv0mzl^G`d~Uzwztd9W5s!v2R8{;VArP zxJh;i5>djCXmRka(rmNk%a>AP7`6X&8wb0g_?4%E-^&=J>{Yaz-@GP0P43v2f2tHq zmaO7?8f)y58Z14-(r;1;Q2L)_i{v3!PwH3uYkox?*c!OL@tV<1v%6#)&)6cMAN=;~ z$bLg0qTW~PNrWYuIBL$&bIMV{%wR)(lGH&KWM6uIb~{;@aVQ$CCJgf?1_S%R!0Lr< zVhm3v9+bT0*GWoEMm_1@_KOyLnWuE9uNj}O^6pi-k!K+_&nn$Hx=ARyG_KwcB}&Oc z!I>_C)ou6%klRJ%{yDiLaoG%bnu&=0{>#61?gHX)_z&Kl%P>3c;rxRNzzHz(2`=Sh z809QKEt@5xQcQOouabiPaJ*Jo4w!i#g<8jx5u1aTrmMl`CBRZgXVrG&be z1L+h0EI%$Lv6@U`-{-EfVlA>F11O?gpk>MsW{V65WgD2-5>AV$?VvQZ6(PTih-3LA zcW7XS+pQVy(5Owtp#dc7j6}?7F{cHAuQ<=;(Ubn4{XBOP(gB_{$>o#eIJ%9vOF1cv zy;i4<#S=bLcASB57|Vra`2jZ(^G_HxReCUsrl|98?dRKK{;fHLPfJ&u>B}3oCVd?T z^wpcbrM`6NSqFRpU%(ge?ZOQB?9(=2@6*^z64(TG{~FAq;02hQ@6*-l?%7lw^Fq#Ys`{sDz9>j&fR-8;hvt zTDHRhLkHx0tn~(4LL@*e5bN5+3PiW1lzb+#p0V#Hs0C_)T5YG+UMO0REt{IBKDb0{ P61)B%Z~TJTWZ?+_D#p&< literal 36385 zcmZ5{Q*>rsux)JHwr$(CjW4!s+ji1%$NplgV<#P(9dva2zvn!iJ;r{jS#!;&HELGX zB94Or`R@b!zUtdb+>*)U@*OB5GXc_2OR1gvH?e}h>Kvc*;k4JnyGg33(Tpm^YNTmY zuCxRd8xXhjS}=B|Rr9yB|_86JJjgqCXAz53ng*yVSb^GKWcFzFODm{i$d{V6V>u!CG= z)pbby4c%YpvSjSN-Feu|o#&FBp!(Et=*zy;{)LM4zNO98YLEYA_Wd5D8-}d&XKx+3 zQAnVK7@8pH@v=7Q>bl3+m}SWYt8#X^w)<{C0kL9;AnPsD&Cbm?i0g#`n(8ZpfV{<@ z%NW7*_xZ+C9-fe;qpuLRi7WLg&T2>@&!M>;Ynjt1ypS~y>ubm9oMBN?!rg3P{2L3) zoo?sqm<@ohBcCaukP6G5<2S&?bfy)@T)mfUw{NprJ1^j4+4F#HZJ2pC@PHhp(r@l# z6hXpbvTzdH{88^PvhYy%HR3>d)~YHFn-gkStJXU4wNv=udN-SU%y4dN=VYBg$+tkN zYRP7Kyx!Ak`GPyQ-n3(-F$wyQYyc3t@1Phnakt17Z!}4`WuB=17Ht5&4p?bAKvq)H zWjY_|KkGZoglK!P@%o1Y=H%<~b^to(B*ZwaBb@Wc+jdWfRIlgd?i!ZY-l2cqo|oco z9rAmDXNM4=b#q77X~OT5>-)>0Z{_+K`m^EwaK0!(JEa`YDaiMYIC7B#@OU3yes>E` z%TZL-v`>I&Q1(5B$4)Y^!*|H z?IRGr3m!-<9|IQ#s4a{fPa<#d{gs|)KJQ_z{2acgsKeXlSVb?h@q-xhD;Mvp9Y~yX z2+M%GzZCioL&}a6CPjdXHq;w|XpC}|W8W##960y&aTtWN3z05fO86CzS`B62DZz3X zJ_L?K711^n2_u{Dlc^kiHpHAad&I2&NBH~Ix>gk;v_j^sc~Bln^tzi!ylD;zW|dbT z!fvIue(a4@N1QxciY39M24>a%M7qi5YYuvFq*2A@PCC;KxlBTL8J`2u$!|DMtV*VC z^`K_MD1sV-P^agOtoV?M5dAJFOJ+Na^gQotJO zh~PYa6ZUB0dx&!+l(b`Jw#nt7iQo364xz0A58;8~0dY^Yqn7>VZ`Ff*C!v02%j*w{ zO;PyG(0c>b&k}-s#y1`SY|J%${jt$Q3ueb1<1q<>jp06G&JzoLsV7Xn{yVokgn{84 z)T{WlAB51G82C1EF&S8}{eT;a33D)!-(o%J>Uhv>1y?>*)gag(6E-IQT9Gidwl!UF zsNaD!M`J{@M@%wTs`*#D1U+Qir2X?xbUXVUq1sppPE~ z{XJ4E#c~4t2;~mFHP+hG5OK7fayg3QRf5Xrw{ksO>{DC6lzkrcO^rLIkXdlYXW7v= zFs_6~hVVvRFr*ko<;Bf3-yg?U!h{luMePsBzv0tO!eTAqr#wG)_xGp;TX$o}j)`zj z@Bd`WeHdT{nJpS^CfX<1?~8po0M_fn(B%7w%f_qVfn!XNz;Y`bmAa&c3HZzFgfM=j zjW#_xk|vM4N%B;ZPcJ{~x~hVE)tyH;<6oDXjQl@qti?dJ%v{o4)l z7fj&<{derh;3Z6c#uC!%tP@n+(eHHo3KK>q0e+&6^5A-+Ftvzl5Ce;$tuie9a`YMl zjyQBx>=+s}oq&(UkCL~;p;vQ4bUF76SuEIu% zQf}#lVFW4!7~F|s1W>$X*iJ&oVfl&zY_H_r#!6sxO{Nji8=`~GAoqyGiKb@FZue^4@YLM!w0nqfc-Mtk z{Sm{zzg#{z85Py=WiJTH7)Wd7Z_=k?^QguQoXQ{Zg6`xd2>ghr2>c4td7on}8pK%l zX~q+LA@{>9Q#?g+$}L#h0WFUlqF&}ImgyUCzB`Z&jSCq(P4GYPa?5&>?URMdUC0q{ zy?jTfLjqyDO2nzh2Sw{{biw=?8l`w(E|pV>PTkj{dB9abz9$}};Ik)589oSKh&oc< zPWs|u1{pD6s9!!b^;kq9&GxLKt*6d>WTpSy8pU{hCngx)q{Yl}v&Ke*vB4xHmyJmf zW$=SuaOllowZ|}LG8df6u&UsFbc~Did4bJSe)migd=-1%IW4e%=}mBokWRMoNXOsU zH>em;T*-9Fz}dU6+pJU^_7y^PB^l+T8%ee&ktTAE)!Xu(9%4(S;PPmsJV}a4dd6svoi}<*2--$;e-CpxMzsb8M3_JRr`9yNJ zC74{e6e+3QB^*EAx7`4@@WQ!6@BF9{J|uz*T>TQkaL^|asi2!RlW?=|goxm-z@x*Z z2cB7V=sar=*qXDnJu5Pr7!>{pA%#I&5V5Z%s(Pl%Za zSzX88%3b>ThYC3C=kM)4ETW@AP?>V^W~*?24$}WwPP)^0MunPLG^6B%a|{pB@fK5& z+eL?tA9w%U7$*+u)?Iw{@EA8?!I8u>uNS&Ed-?7oc>msYdHJTFe5sLrfh0goJUi#Q zPJs_~h{pycp#TK4V&k*#qoI8n8f;${Mqpxly&1se(ziRU2XWn^&_K!rCXB4Jq(_9uM;B+wNOu<&M~i0J=DLyN>ea#Dw|qctZca=w&4Nh0acsZ zSJ|yQQJla8s@Gp!b#^izad{$J8B+@Mf3nxbHff011LtQK3H3JiGbWxdhr~iy6sVx` zTM>xA*xd&ttWZr4(Xnn)_)lwqPXYp3gQw};=84vs=5NmZ<$H+OL?S(4Nj_It( ze$Y9f+jo&<8g)6Vs6D5&^MGXAGqF&%i9lJ!XUmbXYBJ&WP`<>>-8R=kSSmbldD>bF z6pSzlrBzwHc2^vw0iy}a)ieu{`2w5veuHtaAw-L1PT(fLlvK+#Zcv)c7=+Q1*dk`?NYp}Z=E&n_^#*5MW5 z5|=^1I%>IDsV%zvl;-^txZ)7>HX=kZvWC8fc*nOol`*|C zYcyDT_P34rZhx_w6V1}jV`p==^TQV~lAoo( zXAW^7SlxCeG_w0iTq>@nu`~Qo1BCu)K|>04EKpNx2y5f*rnO8vGog^1UD0bQR@75j zw$kYrLAYb%J+|s)xb$hhUDod_(3qyo+gWS2ud_l zzl|;T&RT2l$Td^JeDtP?cq2=qWDZ32GBN)>M)pm62{nDZ&O-eYYj&H0HX;WYE2{$- z1!q*k@Y|59_{}7JxUj8Rko&phX2vZV8=x4Qr0HQ+?9L*HLnPkJJ;ETvotkMvjRe7d zRA+R_qDva>OGXj}iW-XJtUTnZoU8zIe$b|a#hlJVI`#ku@gUSP^wC+qn{Ag1&fboo`K**13DkrseB5922xJw4v{wH)OyHmnsW5lQ>5#OwB}9ai)v&T2Su-XqSIb7v zz=Rxgk;Av_Sp5Zrftr3OsPgVAGRVMR)d= z9);bAt?479v=Dod43Nx#C?nL(g-7rhd|0;hmcL>b21de~reuhOf#mv|M9Fnh!SQ zDpM9+*_*voCV^=~Qdvz&Ym>wza@gW2!cMB#W-#IWg=&M1kz}@=GL`mC2wMxF5w$ zoT$qJ6Bmr3Nl(*Py{WVk!k55KkGyPecQ*&R+m1^i{wzyG=qA*32~=qCi?iLNGpb7DcvV9R^s45BTy=xl z=^8}nKhVF@&wlIbU_m49k{npHW98 z=^boOTD3Z8Ur>vKM{>0tyrqOXk!4Mv<3Lbnp8zIz>Zm*tenSK&d3Q}Q0sc2IJ(C)4l0jqqB&u@eEzd&i~ zaILLbRkl>{L(}>)fVYcsQxhS2bR_J=4dF z5_x5KkB9%2UXtf-TkgSsHbvyHu#gHvE}#nQ(poEea)0dNDPF7xI5$)=sZuaUCTA7o zSQtKWwAlbQq!@G6IjU@wHnP;Gq^V$qCF%=O#8l#^MsV)C=tJ4`5&*S6M(HVg6^qMf zjf=)*15!b_!Wx&>h;X zqA^;%!Q!deOD4YIdf?Ha)T~K967)hR)aie>N%kV;k16(zonz$RSPa@CF5neULoFP0 zjREtmm1b`RR{Sjcv{!NJSar;uo1yNNtirr?SzlHCx_aB`Do9MN=~m^8l;U@njCNO^ zAL1acwWm8-d2QblDk$O&qgq5iYuF5hJ9h~sd*S`%Mg2E^;G~Qnl89QkVCh1f&FZr1 zzlR^Q%C~jKpo{}QhCpX~!cSar#gCAu?-X~Sx%A^JYneg4iIJbJsr*>7TuC%OdU5-9q3teqB}>Pi4V6LGVC-!7c_F(T54p-XMw)PC_sH;edRx4>jIA2&iRR}I{_wmeU=dLMxqOt0f&ws5UP)r>W z<`{dM#Z99fdqL`MJ*Qt4%S+z4yy2q=bsVK&sBx+1S~t3vPot(*ykaix*OctzLHe&0 z@$~3#SAV9FogB@?pGD&FW(};dA&|bMHm=M&?2y8S4O3WX*^V?I}n1}RJ z;%_ueY!JEjVzFegBqcRYj3z%pys%eMK21-V@%6TL-l;ic{L(0|IUB03zKZNMvJy9H z3{R==;yGD-U1s6=bMH$UBhx#KiRknF$p74P88X+e3HEtO)}+dmdK@9%qu)_@>AWoL z0rw{lf|-+(+2N?7RSYL4RyI%vD{RJ@jY0$K<<2ohjUYIZu6WX3JO)7o+^O#jIy7j+ z&h*H;nu&UXz1<4^Fod|p)ZqtxchLrBzz@RS&<69TTPdhYID%@pNLn2UlhSWAwRNVm> zIVep}?MQF$2tu-hYsSr%Hr+$`7Vs!k8Rj$)U-@whpS@!ZEB6{cZ$>=QGGsWVn5h;> z)G$aE7`aoEx|g>nTo?blgBQzFl>U+G#%6h)nQ4ZOSXU@()V)9IJ8CiSM|0b}qbV7I z1R*RVXk(&i;#j_{?s^5XZwhBM6s(f1tg&s zx=PYu!3IW?`vEXMsV)XH=q@l^TE|`|IO^^`V&#_*DRet0GyrLtpqf|3ZU5NCL?UE^ zmqqUa-PhHqYD|o1xdfbLj9x+4UDP59=$`*ouqn3w$$Gt`Cds>uri+k5Z*8Ff&nR9C;fy5_EeO z|6F&#QbzLYvF6Q3y+WKq_5%=dTpdRU=JPJ!Td_7(Vf{ukY0SZW&eewU%hOvjLc5Al zYCnz?Y(v6(ei$N}+ji5cU)vnrFVu^992d+_iw8#T_upiY?%m|jS%>@2K9CpbK_~e8q##=-w)Wzn8+W2%hxiv< zO?b(X?`1xWZZ`VS4BUv~i`qjCGTPM}_WEJ|O?>m&lGOcT8lT^qvwyQIT7K|iW9YTA zc9LyYc@KCQ5Cz_di{B{+4WZ((inP}AR^Y`V)81X7NQT_fzm!$(MW>f+lZ-2uKX zN!-3J_wkzVa(9OQi<15q&Fj>)vthoc!{TByH$8-FzWYHqdORX_&HgV~aoDMPX*Y0B z^r_Z}3g5WlVW7saY;@Guq|%uRKRh&jPf-Mq^#vrn8O$GvT_q+$QNMGlHyg9l^%d0^ zE+#dh&aO5Se~&Pj|A9gZ^>{q_3lHAdDkck0)3|A|j9J$lK$e|NrZ8Bg1>$y<$zoM` zCWqqZL;GafXixfNmTJ@#zR8Mws^6xc?TSv1ccnH|%HjQ;EnpY*R#yDIjzQITQZW6T zdrF6}K1U$9EM$9^3>AR*_dO)=_Cv&28tMx9FP(si^Tq*am&icI2L38v4YxvAKh)=^ z$UsnFaiciNGjsqd^dcRg%6OG0I=^^t65+3JRio&)zi5#2KY+R5KLF<4V$W4@Qhorj zZ;rX3FWwX9SrPqN7DGET{Mlu3=BKE!R67U!)EA9|g}l;&lQq<5`+q8|;r}K)f(H=x zwFsZZp%j*i^~mE7rJ+B~Q~coHkkr%;?97h+qNlGN)*?#MG&rZpuRbtdYa`S;M8C}W z_pgOIcYA-wpel{q97*wLHVo9zbLCs_T$GXJH!KC2g)f1^cA+Yb5NljYa+-DLiy`n( zZ%kvhH$Wb_Ykr}J<$D58p?Gzp2AUk*ur;ED{d&3phX4^J?0sR@3EL<)$7FTN7ko*~ z!v5ma_k^p6D*P%&fdGZ%rM72f1lP^MT>K*ok54JllC&%h!+{ju;MB_7@Bqn7@@0-K zBR02>jtFJF(t!G`O~KFmmDR7casEUBp*WW!d&3t^;}@M)Ha`ZvwfMwMWqmRC<^C

+c@G)i_$Sf)H)+H$vK7jz#<+l)rtz9S#vfw4~uMpjz4ViuV|3 zjL$<$A)S?ef;l|4X7Emdv_>W2@HiDa6p}AoM28IpYe^73c`Og|Em$S-onK?v^q(DvqwMEfNGRs^Uy3ObvO2e;f|CL+Ktc>v)knY) zx+VW~^yYhm^tt}_M#ktSWFE0na{$uXmfiBn->DgJk$_=~noE^+4 zseJxfRG%gIk67bBVr0VQ{@Q>uyXuYE#N3@-0-q!?gaF+nbf_tBxMmT6-|% zv=60%*F9#Je+{VeZ)f%N`S-Nfu;yur>Vtq=F%A=GZm< z`cJ_{c34Xy?>a;WA3}*pXGRL_wVt6gRo$S2ApiENJY)xQ1Ga)PRuoIrOJrVn@Soux zWnp$a*$7>bE}|cSXrmlN)^LWwZJa}Gf_yYfX&$XZNGJ;I*hz|K?Xhgs21@o|bFxyv zit`>e;0{_$G3blRAiO200byek`+n>hgh+gQqLL2ywx=i1!5DwvF-CGB*suC7;wMN$ zqRu+5v!uURM-$q57JAXbtPxm`emmN_G{`pNA`G50GA2+33%J4x&kx=Zt#ZZ{2hjHB zMTO&`)&=d*Wy|_^b$aOwghFkLN3f(rID0G)%>~n&V{|8!=38ZgiOur&ZHO5o8acd@dHrlL ztB#cuRY3tYtY{xmaG!Egfaah%D~qiWL}jWsyBwRL#A9ao&GQb+3Rt;KKwfDt<^z#6 zYQN&X%_!frsl~gxh?)fUx%w72F2iv#E03JYLTGKeiCa>5LQcLK*R^EQIQYdd*zYA| z(n!w!4B>}bv{>B%M@Or>36hX+vNvU*q&D-kNAvtHIH@$eG2}}M`o$qgzZ8LqjNotq%ev2d`)1oxB9M*`#h+?ai9D2)lBp z9j<}##^vIPuRU%c5V%7=b5O7w^L{aE|q!X>I`+ph0vN+jnExN%h02VI+gGU)4Ic#Rh zuNfUEZm{G8dhM7D>=gpSiOLsW-a3y4Ue6SWb^QUBcPjsPK?Ng4v+n*3!}|$$->UUa z#U=-Z#}r0|3ob;2MmL(NNrin5T!P7#0EKNrg>EAZQy~YoLzGbj`1iSOQ58(xREHd` z!FB(JJfHH0Jow+PgngtKWpG_$Qt09L@G_p_K|}j&WIhp2Hc?{o462m~G4F>waHIzd z)QeV61+Bb?a1Nbjt)}*SI@2z z^GwVi>P2)+A;V2HskBRrOquu{p8&Vy@wc&fiX`l4nA}XnedEmG;V4b=L)_kt3ycuL zP~nIUc45DnNxaR+De?Ep)sBWvEXf%~qckYL%HIy{*ysd|r*X1ytyqg0K=Vj%AYsQ5py!YfX4i%8(r z14+5k@CGcYrRus?j;oVgNb-&rKebq4=iA@`imM;!dacyN*c(atoZ)|%uLd|takPG!b{t$=ZT(*3*qllA3P3%Jo(1{`%Vf7RyqN>cty zZ&x*6CetIT^*gXgtG1d7GVAcgwk_A^Q-~uWE4E|p_%AQh?(;mnQ`@(Vc5jB}AH3O8 zhH~sf@lf8H)8`g$X73B)$#@8{t{cYj;sBP3>z~B=%a%bp zO?^y7%c_Iz=|=U8Cbfk6J)!k?4uA9{F;V+)_Oy^8NQu{fQ5eyz8|C<2cA9MQ&^N6J z{WKiZ;^e==(kT3P-L}}dA#oxReJS_&iwxhc?}ra)+JU#vJB4iT?%!-Mp@*yWc!%Wd zrz>yJ;{~?wO~JQ+rg;K=$%ByOD-Kbn(zpTfGU-W?sVO*TAfeJJthsd(3nGoHORV6s z#9ED6YY`D?-m*5;=o0XE_eD5}ZGOVYcyd{R($*0T8Cp{XJVBeiZ@(8+sKc=KLxfrJ z0H=OL9O(r|dYcFRSjxHhi2B%(k$m?pAlPp)izCMfT&ByyXPw4K=wlC^7p05FuUTBTd86MPl0+-cH38S!U1kGQdiWq_ zex_~GCxw@9R=7nn6)?#B_!RIXuW|pvXlLvrI8EBk1SBm~c!&SX^hCc#to-G*O!q@= z;`lXD@6Si!)8Ems8c&*nGI3h)Yyk7pt_hu?a>TU)^Z1LqC;t++KMsoqg@55QXTii8 zN2DgpG9AGVEHO6dwVLlXGI3F3ICxDNp^TmA<@JZ?C3aNQreXS$p$414vQbICH%*#% zcQ?Z+9np!S!7`7-VQJWYiRa%V_rK`~$Zby4)I@U1O6uIM`lG`inz4CV|FWIZmYmf8 zW#8od%PwG5ypCV=t~O+T?=hN+h}s-ZMMusMuJ~u2f~MezxE>Bo55v7*?Lb0tedfaH zE+`7^lzz)%!(Q9?V9+d4Gu^nOj4U@~vCk}Tgl2DIw$Piw z6)Ch;p}T3Jwq$X$M^W9Yp{ny+&#{nO!udYCWJj(}AQke5$y5=|DH?|2N5kx>&^J(wUSSH5&%@XQ@1@-Oc8@qX#wN?M>J3{R~?G+{P%Qg%y zv?@Zk-FA`~lxN$%ER`)3n0K2q)eO}^VjreV!JptwD<6(@YcSjzLuT5I>!j!~7q})D0rjw55k5! z$@>VyhC10S(xJ$fCpIrD)`)?3W#htolUL^CC@cc<`qX&uc0viPAxuGWR z6(4h3b$%RFClaJQP~atG+YEb@Rwq;YnSH`b{Q$y92Ry?o4B1Zzl3t>cxCS{w=6~|R z8GLX>{^opE`uahqZ5#`(5Z;HACIoAfW&AqhK89u%z;(Zki!Qfb&zUXkdivp)f1u~M zcg>o8%o?&Lv?{&vyO}oLPphYGl}}P4?&-*vu&zE|B81)NO`thxP2+Lo_)%g&L!jid zX-P$R|vP=>?*P|h$_TKbz|)J$W{`8_AU zY~`5<_%SH%oH-`0b5tk50X9=2N0u2ryHK zY2Ln$ZtJJST(JgzlBthU^FktP*hhJRx*FZ4p1)>8xvm{Uybjd);4LCGr8yk%XUhsk)BL0GE@@B1?oLJ@tm2JghP)D( zgSt1qoFG+%#m40=4>yQH7My|y{P@i(alVZrop^~rIyH-+Er!6CErUMn1A$H%Eh>#+ zNezXnwOsQjjv_KHhVJ6Y4GmoB2oohNk@?q%Gts>ytl5I7)JBY`1YVnZ3z8?*KNLG) z&8-OMVod~rZ}&f@5&VBl(1ib(Hgv02=~Ho-s``|ey@Xm3`z%N{2tIs>DOdXTb zEfDxiR7dh54QAuCaV;&jGom zN@zWvE*KSIhq2`}niI9U)d{zu7v7v(Z?qSi_5P=uTq^3zvZZHCNu}=DMkyGS;BUSW zCtn6CB*x5656Ftgs>-k?zXTK7IjM`gT4`U>BC;*(k&y&%^g<~m|aDQ`#eTk~*icmfP-RtMtlA8it z%1|G3@sCwZ{9)30LA9_w>-LIve*}3(Nf`RX=vvx5RM!L}0VnjbCXrIZ6;+GfWp^WX zL@Pywsa$@5t91WRGCMhVzv=%If73->>9GEbvi)*+Uv_a~ZDAq{^-XA@m7UmX{m;HV zI?TQwRO-`7Byipsv>OiIfa`3zTN$+lWzA$hB01t)BUS~%q5?)%mIR){f5cn{qUa#p zVuU`o2OjjtH7&|1xJDk?lQ|nqx2Zs<5zC5=CRuAKCTZv2&q!T%9kF$D*pZ8NKz$XU zh3Z+)`Ny|)OBBee!)mT_8b8Z6&;zRNi_5@V%nB~DBitqf`_Nge1#0hb+Zb-l0TG4brP?BMAF{IrXZhjYv}zQ-|scydxt9_N%5@o*dZ)Vbg8Egp!d zke5R~|0h6Cu)?q1}(-<-=s@RGRU5$rSV1KuB+%@k4y*VTdQB(kPUb~)}1F;GMf z0eg^VK@Q&BsFdNLC0#Dyv?YbN!Yq|KS@onXy2Cx`9*Y=^G-dT(yb43NC7=-%+X>6B@w{e{;jn z!mnbTNquc;l4`}fj!FuMrvv^80OJ8-0If{N_0~lmWT%6(5^==9-RIy6GukV5H1i>(l z@OWCt#scIvl?)OW+P(?&NALPb@<{&xPTrV;pg!?=~) zZJm9XQ!#dh-?S1LGB4dze9Ms7zk~54$f?pUR&3wOspcplXJIPAM(X)RcXc2e7YMxv zUfFVTGWzLPyJfk72y3eV%6Lo{mdD1efwirO6(*6P$+=34864 znJ|hUFG}0_-*)rYu^(tt<#ms}72gTtK103;(Yjqa>a#3D=-@2~93iobItf{9yiDZT zvN&(Ftatc6&EYF{-EHoh1i?HF`gDQIF|3w5$$5xp&cO{YVa}4XVJ|Tf>Gbr`@Uy=c z1Kc%#MA@g(`<~5c64Wc~^n|pYH3pPW-quZOBFD#cKEKe*2D_!JBkO#C$!y+LaAD}) z)-0aBh$r8pR8zWfyuMq5Ye^c8omx8+@$e^Bs4n~1Za$u759yC~Jv>eA&!Tqtb;!6- zm6I|_+=!LNI;vvSo4FZ&q#hI~Co9mRvX)^8dx!~u`z5A|0__385#Bm6##k;Z3SDQd z!#yn3ab*q2!SPI=_J#uz_Im+j{ORYZGBc=ygCp}jIbH}3?T#R(Y!Foa6Tx;(5X-r$ zkkH$4G(;8(7oMbMBAL((h1IF}rVW^lKYptaqTQi#-zk(#XAY!ggo2Z_?tc$u`baCs zr59q~b$fg^`C>eT#izwqeNk9u+bvUNwFb+2hl2cx|4m2*l{R3Oovear92I zfFCS7@_aUW!5dT`(iFtdX!1EcNr~96%pB12=CnE7>2>e|Ugq=QiqQ$qy>}4)!!*L- zbYa@67_%LX305;@^#ZJIM)Z&a!qYub^MO~aK7xshI&gbW?Quzd8)~z2;-akfFQkk& z6}oCYPLAwOC9B~z@LRDpc1egg&bOYhAj;eJ92XusWY$;sm8hL#uOxx(UPC^pn8Ub_ zerx@whXDIV$&2%W;+KpH*CP_qdx3|ZUJpkH9J_>ih`R85x)rO6tt*N!b&C-x3ccTWcsN!~t)twp`(sUp)x6yIf zTp0}@PlK`&wg}DZ19?rL&m&Eu$M!%>W~+^oHL1vuwYX}IYnx`<1s_}*F5yvim7pT@ zK>%uk;};%Yvll(3P10Ok6$_wQCHS^RT1}+6?k{kr8%yL9OZT7fGTf`;2 zO5(;7EWE_HmbI&>xy36g5#RSW5mJ$#w8QuHzsI?EDe5*!q9QLBF83#VSUN-(=a!=% z=RCkaxG&pS8fQF)j@9pXVBJ8G1o@bUFgdI$dRs*omYY1VCn!KP1@D%jQ<&giXouy&nJ7l{~S0XT53- zKQ(nT4uc8)-#Zs#P(PHLV>+~ifiO#bv?`hJ0n!_o7EI?Lc61B(+~+N2)o+^q;JUVf zkH0xJkD?Ky4#uzee-b|KTclpj5si{OStER_6(7ix%k+hqN}ErIV38T0&Zwb*7oy%j z&z3pWoJ0WX+0VyD#2xoI)6f&9a|cdt9GIsd>GU9voae~qgJqBzlE@zmq$yDCw@12U z-p;i@ML?ppGEl-DiLu1ScS7MZRLdj?Q_Ux%hA$73#tM+Sn@}<4lt9q8Q6s${*CP!l z3>Guc|6bS43T`RSqu5bm?G6n(Fo_YJ)~Ql2$`hF;26tZi-J;a20(DB}M%X2((7!_^ z?G^~3lJR?(24*W7Q|>297|A#+0h{QUKd2!|4i3uuibZ%vR1p?Qp2VZ5%@nek^5}-! zjCusBaj-J9ZfVDE=3Lfg`{=PnB4aD|n~g3lM#k6avi9b4wnO`U-%iS4(Af@bdS?p- zIK~T2_0g4D0Q&6NLZ}G35U~M;zWLhU{VaVTIT(pl=~uy6{6z3OXrK@_WT2!yoQzdN z{#hj>_1C!B*FwRsOCsvr)sywM^}&CCNbg1i(gsfiaB&04T9(m&LawK$`{uBw86FW; zX3l0*>a7~YKn|01|3ak;KGnu{3)NXR{Q6fGhYO#oql2TEY79ptK`{^$ z0|~i?lf!f}KATYC)l=B#GPenQB`X>@9F5P!l+49>WK-S*uw-#(sf7kvg@BQmb7uT9e5@EPf3B8OwpA62hHjB=22D znawIB=*RGE#@W`A-Kp187r*n3jelT~`)!O&fRQNDf8wOjIrurEq~pp52V&(0A0HQv z(%-O#$FOqTXl%ppc*a8%nNZfmP|pIoWKpXs4rtp0*k*|By)UvGGd{K{u-Z2VxTfkR zErAR_1blWn16DUvNz?J;mVYnYm^W>IC{h{+;CzWla#9TQQ8mH-g@!a!#v%y&dGF)h z7$1VAE5`i`qzo9vZ zo3#ruX~?}j$?142xDZ%4YK5%#R5R-|WPVWoR-)S5!Ek6#wg_SGU}W?D2E&gU4KfsD z!u*D%4?yd+2o}wCc1Tl5-pabm>RW7}W*;K+_ z@IX$sOX;fo?6Ypb8)J~jYp`f8zT!N%IQ6R>xIwI9N^5UxYRoqBtLYUgt((ha+~~7J z`|8cK)hZG<_NCk|%c-r=r!ORrN?C8tTH(HD+|`R*%eg+~<8zYF@>sAW3*i4tl*`Wnqa4tDyQyHIYmnUtg&xT3G4Ocx5C$O?O>FDAXKE1547{ zBDco_W;ZlJhtoOX*3VKtRr_8G3vq#n zeP+Wi2HUHhhhEE(3ZUrIqS)!AO%hB_zqI_LD0HQvJ#GXBt!RtNv-@`O_0g)8+$ykM zcLq1@fwUjn(t;D|(|CqhKSU9CRGW<{6XpyDbLk39T1FKAckj0pbns0LJ4%8SePWet zQ}M=&Cf6Y^D2qz^uP-2iiHRh+X$pZ%0+kx!n_TS8M+j+YVA=H!Wp}GgNp^#Itp#@fl4*S*>pp z{d~I3OLN%iTru~G8ge3AC{k}M1wL;wygzT(afV52@G`na>g-fm$?dBnsD+bB0IuT( zNs@wvT-^;Qu~;qS>B#n8Vo$4Wj!O&t8ig5$JvIrGt0PJqDF1BTN{&V{oFw@~qy(8q z##3SmU(>(+tI1QkmcNQ8e7$n3Ag+~WEifMi3fV89B-Uipm_nUx3w?OCzkl6U>54!` z4h(MfVr`Ba!Cu1~vk*r&4!ryrb=Q8k{x*4@*J`+fFaWwCE6BXo|6X;H=)LyPGno~c zN64;%bY((7nuNqcbxL4ei9{I@chp$3D;r7Pj81tt5NR%{RjpV%SHCCyk@}GgKhMul z$Bmv4c(kB-o~kWYzW%OKzJ6IQ)2s}(RoLhI{dWm*r9Sj}I(M6_Zfc=%PQ6<&1K=<* z<`C|oX&^r?%^6Jz8m-q@f~hId1Y{P}K8L+3P{hFkd9|yHzSz7fu%3GjyWELEg8;&~ zU#!|h1?z)FS1l(ugd8nbQqMd5^quhkguXn^DmARP z%`H75w$4UeOFQP}00RuT#6DHRIM?l()=Z3rn)Z4TsE0qAE)C`a-4YI;K?U_D^lu(R z6W!?J&AKt4r#`{RKGe`>GF#1Zwhn}~e%W>U|Nke(>st~h4YB>}$+TEv!VIt<<%5BQ zkT=Qdhe4l8v8W$Dl4KpcT=lK#*`EZ6Pv*nrvu}?h>^ZqUkL;BthnfJ;M4gl!z2evIafa(;C?(8ru z2}Ql4gdi!Mk7e+YoD_-Ag2IAQB%;&$dkAQ+8aC5IBi2xlt55pV$^EnR<%}eipDpJZy%< zv#8YDP2-GSB~Me=d3c8yu7?GZGH+#JBS@8olAFuKeimi}8#<0)^2Qz>88`1~GwxY` z{Q2W+jO~MIh9enF>_|2y_t~?F>jwBcHt63}HmogtPJN97Iu8md!_X0g0Z?I7Tnw)A zF!4Tql8oijK;;Di~(eG;P7gzqy@XJd;}3tmQI|O&xWIpKb%k(VfFB>r zRzSqVa1@Y7Mg@kkDX0z7(v;tY(-7-Yo;o2p+diyMPYm*lN?gvuqr`xQ_nXoFXQa)CN3pA@<3#)=^YBtqu zs@YVt=@Hp9?Y+*7Rda^(%M~h_3v9BKDASV=lNN~;V+}r=+wPL{tsz!3%jSrErfs_q z=&(>3MZVGL%nS65{qA~qJVbt3Tr|IDj11xwk!UQ#+ z;%jIYkb1096%+LFUgYg1*+fF=QsfShMx#Ujno%83)LK{?Ylc7hS-d@asotJ_q*8Xd zSfzhhp3wbXH@9D;{s^j*cj&SU1)mnX5q=ITJE_VqVV1cIG8T`z8lebci51wkRG9CM zFp?_#I}C?Hn8q8=0Vs#`6^+H`)VH8n2l13!l(4}V+$S#T7VVRk0!5{qD$k-2(=^}R zd+C?MTUD_cIFriDQ2za#!CfIZ6zHzY&}ddd9Zpb2pItMT%dT0{M6A5vohGXq$2E>? z9M?D=$MM8mlpf0)xv;!({OAipHAAf-p(%K%=O-BMJp$VnuxmPtbFRZZHkqVt#EfI5ss9mX((qTJ_xBGS0p}MP^sA zD$4ug4}YzYXdw--w~ehu#cuD7d%(1Y0WxMZnubUx$N~C43p`ieA&KiRSm;S^)End` z_T!m)utSTDumalAg`Z)vYon=B8luXJwE{A?Pk`0B>G3L7zYMX@UgB_&qO70 zO%R7Fi35YBv&N>(m5HQnWMS82^ngD5>)SUsBtq2c2($op>)yg~WIk&nUPu?^PKqLm zUY)Ls8Yf}21&XLUr#g^vh>oT9kgL`P2!)YG4r!G`U#+V1DMN9A@f|;twPadMWnX2Uf zY+(nM_y)GkGb;$4_Q$L8%kL2vy<02`DqvCkEmOer*7Z#~5>o2=etvMPaNKgVx4Ms#6;b zVgy`G5KO8nRGs60aXm>mm{N^2^UY6b7rgx|Q|hHk3+GKOkG;pBZCDk>EZvpYw?o5<5 z0vP2KWFWT7pXO90Ni0bqk$$aZX*`zlF)g5~#%Y=mmHKGFWhXo}%aH9_;SJajx#~WlOEtHvfgf&M+kcZ(unx3fQG$3}JLT#=D2> z;kge69QEX`RbYKI87LGZcg_t3$jMiqBqqvr?BxqQc#10YcieM*?g+%4;C=x&-?<7! z;cg6I9kx)<67~Mn=VCv#_2hv=q!KMtZi?NDWrH4gAT5Si*)XqSr;|nznF3NSd8`nB zq}T`Knx5vGRVz{;m_+7Q16 zb!{y)`Dyag{R;xgfzISHPL+FRsZ`ljp@*DTkr{2>-o=daPft#vGJsjyfQ* zA>D$xh1QzwPPH=#k@V1Am>vRB6YYifK$wc~vevJZb- zHhcB{(`Tjt1GEQ}!!48=p-&SSca`Fs zelcZsz)vF0;CGa5k(nk;Hk~GW;QY={+_H0X@`U#oO!qZRueTsj?c`yHTAiELSfv|2 z-cGRpVNa%bB zuck$gG1wG4v14|7{~(o0TDdBfMAKamYOhQNwkJDN_{h%jK(n9krI zFn>j(_!boZ)nSc9-@c8AOhL;E0CDHsstulsJitb))@;ym9u7}lh4H^HEPw|TOl%Y5 zM3=!1UJ9ZkSb=y#dL@)B;IHCRssz&6IJnb6drX&p`dz(iSrbeZCOEQu=!-K>;C!?H86Az(8VEZ58$d5AMQ7-?lOd%77ZVb5Fy~4^nEeVJn$* zGpj{Rl(Pfq>I)YLl0pOv?6^g_uV}+@Gu=HHkzKdj%5w*<1KQC0>hv!Z_UXHC;t8M|v zld^+t_p{q{#T=iN6HZD&ftk#p0KO`w2nVR4JpE%w_C>&KO4x+|#kMJ9+ske%EU7(vArsOKA#7n!Ae<< zTcm6%4HP${Fle$Rhfh8A1{8BYmF6@wvOe0327Pd}{=y}IrK_;b(Sz#+Qxp@m*ea@C z34vOCDm6jb$c=h0y*D>3hI|BW^41k;U1@ge=%R)$N49@Jgpi+ti@fCv?rWVb@@B_9 zf9QFJH?<8s2 zJ!gOZ@jpObcXW0Y+rpkOqRw2{K;kkpSw|sSNrnNL0BT{4OA8h^^>F?7F z+5VOK{YUz}sJP0WeR}or;|+@>So*`n?v#rE)kCIb&*%t`*(mt+4ecbpeQh?vAJY~T zq+|8>9r1ho$=|V-)MSb8{PD-1sICy6XL({1oa{c=OWO*6L``Ej{P^g%J|QPOM@2!m zuf$4?&;2*f&D`*)r9n7FXl8#_b|-`yC=rxHqiOPBH?5R;7bmNRYP* z{CfcZ1_=tx1#E=^_^4TF(wTnpzyF^seqRBNWK+}f#vZPx^t}$biRC4I(?+wWu%?>ZFym2cuN>uiGZ%v$&7QfW|shSF;UZKr_oNMUBPGNj=a~P zUZx}{DIgUYjo+q4j~{6Xh0S}{8b?tSO4=y*HSEF=HDDl!Zwj9j0P%d&U4)!&xJxR? z1AfEQ+SP!2GYk#o{2wB@>61_=#N8Fg6+BFoC&&l)@j0M@i=ya-XEf~t>Fu0R`R!8% z<(CuKi;0#xASdKYraeO*SdaM{EoKW_XpxHh(zKvyp%k>BqOYR42vGvQX@*F)XDC*) z=uoU)n&&H0w3Jw^-ccQ3=XMGp`PKV>z5mxhf0;lp#$DE(e=|qKmRt!9<;Cy&55fX^ z$FGS%6G7>A{5bye*z32Not%4pn*VckhQ`o-T_jTQ{{=x?C=D$3h&Z;x$1VCLpp7-R zxVHD_Mt$wXEf>sV;>cT+d(l!RCZ1tQ%i&UJ##mLUq*~jXLG>93%njB(;4>ai#h^+j zuRNg! z&sD`ZkOmvj8?Gg>Ozc}$>o%zlDJzeTl^5A3qIG4Q2Mk+n?G5)Ks3^d%*E$*T9+z5! zeO9nPMV!AzMU9GOMMY}HrN%|0EEg~)Uo&(NH}W`$8z*%TfBV{~C)!2tQRn{Xm6%O? zMNDNHzQ0$6sM2Re zytOY+RJkiA$84R&R!ko3Q*?i|-Fv;Brwdhl(JutW3xwRG>|=Y!cXetE));(d7`)+C zi`-G?g5j`kK)yt(W(n=bTiQKs^9;gIKr#$?O1<%1w+&rMoGFcRaXXQE|Rkh4( z*P3k&sc9tuR$J!15(32DhXTZ2Ck2Ro60SLM;Fky}f((UWAnda=Go1zYyIiQ^_6aer zV722wIt49k&jbq$@(3U#tp~%U>?+diJQZpAu(G#!Mf|s-1NO*vv7t30F_8lYd30yW z(6PWnbs8>ae%*Yym_7xC!J-o9ys+5F+xDq`UcmUMsUT`Sj@^q^1+Fk)Jv4Ymz>VXU zwg&@#KH>tsHI8Z=y$l>pz6-wU$xt+%jqs-VUY_G9kIkMtU5J_c> zR}$6)Tp9ScH}`lDg%?CGzPOmFBq23!eJB~{k&_zbjmd=RtnJ9r5(Wq})*c5SsEU#}51Sq25rB9OdmsA*MvMcXUAGaU zU0d(2(>kams+H2pueQ}g>t?`F-qcPS&bqXV{u{_T@gX*HPdO(Om;GeCn-kw;6$dSA+kW*Dvby$Ng2kr39qyEI>ZJ01jcS9nnF! zOSU3Ct%101v1=qH7r4PBo5W%c*XP_V3oT}$7Biv6OlmMg4Q2|Ti{gy#0XU8D9l~j1ll@(LSu&KN7q_c@Qhc zkFm=|N`nJ;WhMw53W;ErSgn;j1TqiDXuyQi{X1MSG@r=+eo;jy6aAUxmBh}>F zS2E;MmS^6DHuP7CiEy)}BDYwCI+($&SU*HU?=rq6J3MW~1j zVKkFy-)+}Z(4GuC4{V!E5@ahWsK|md@Wt~+tIv>Dext{I``R4TH}Q$<+`6a=0_6Re z2f`GN2NFDwoXhYbZn{ionMNST<%-AVrmS?)ZErU`r3P6hWTc$uiw{}FL?0$J9cen! zbaYvCv_25|24&?TSc6@7D7JKE4k67^t%}1>F)0cT%dndGR0A-dQ-qYCD+(!AWAH3O zHoGB^cKaGg(*o2*B0e`AQFxV{uY>wi1VbrdWN(?fb82INj(WUeYThGWMH0nAiNXw| zxeYU#22XFl%uP|986FV3dgO}Ir}-c$EMXZR9mb!ju@neRLVyZPLK@_8%)*kCKc4;3 zt*dL>KKyq%!hSQp5iGnRjT$}c=4{3f8WH&+w$tsmfCHDl4C;|jSM^D;kT6zQ#kvh3 zTxhGKR#ZN0gNJ{EhtrPY!Su!z*f89>fkjUijw>6E5>#SgL|fea-Fk{ez+OE>z(nV!R?I4y}FGqH{UFTlF+u|xTr zn8wM43)H`(WKX=yJ%z|i6E;C770~9CfkV-YRKiwfb*h3S;ieNme8aurvLgMeV+2ES z>sKBioNK4%6UE*Iw2(NLvkE%3zJR-aE1^yIin}BnR56BMBhPV5sF7PIN^KI9)EKPSjjXWWO`H4cKkkRW zt&s5k;!>GQhrP#~mbBYC#IP5@=eg`>lHDVhZf z$zfX!KpNyO%)8x)eJGak~`}%pQuU`h{TNR5P zNigj;nh!M}YChC_sQGX|K2#%>13f=x?Br#}PALM6DPT5U14(IntTL74xEIv$_=^x* z^f^x~T>h2PBK*t9Uik9v^v9UmvdLN@zCc}5IC#lSFUp;=v1pcTyAN1Fcs94)CH!M9 z+ovZuLbk3G;fJC;X*>L7*!UvHSvPFOIH)0dPfFTbaIH!j1-D3yf&mMHfxU5f9R;Z-_z$rnoa;vOrrRVlJ9^IuY@7bmYh*?HtE^ zWcj2gw5r|=V~2u7h%yDc-NWZmta7XyBlbj&Djfm`)mot}sXThTi#eZ%+ZJjg$=NF* zwzzWv1Tk4$8~*&`pD}%azgnp+&gJ~7J0%rBrdh5HK5y~a*eewn3?hZI$e!kn)<&P( zXfa!e!bNE|$Jh&A62GkZ^~i`E)n>SG4q71_MKV;Fo30dP%45-mAd@!*5YMBB%ai5D zoR*ooF=o=SA(@#e4YBD-C z8TDdh)YD{?R`?|kA+752b~b8vVr&HPUjW&x4WJ;Vi4C<-rs);UH0{n#Ow;aM0n@D7 zpQ9Qb|z}}3TL7%Ln(-pVuLP{ zle&dlkdZD#j zv}XkxzC|N~g&I6j#Q>`yi;izfIyG&=M7-#kYPi=F4fi@XA_Yan0e~ppVi*7_9@fpk z?Eom(EjBZeb!n_{bj)?^CALibZn|*BFjGG2>>S=}?0BS5P7 z9E3$St(=NV#>@t6rg|??2OQ!3p&H0k^f*zBrYRna<3<%rEl0Wf3bHi`sv+}tXxB>P zuTtBi;R3gsv6n7BVONUkp1%KRh;d@*Y>6Nv*vERgk*t$uNPS!l(oxFu z?7OY%^4(2-NM7HoOXa0rvGy|74L1BOiX*N2gDCIqKhfZ>!M%*&9z8vLY-vC}(VHYA zY&S~}+p_r0#kbX-S-yL_be6`P+GbN7v9LCXf0)MFvi6RAg4o6G$3u(9Z8)4vnx@Oa&4b96_o9DbHB ze)8RaBP581t_%5iyD8K5U94VnZ;=pv``UvbU_P~g zNYLUl2ORYif*3Q^Od?@c;`yj((&MGgD65!O(5`GDP{_(eqT>oEKK}go@R4u-zA#03 zp)_7vnTjn6l{*mZP)*2{S`?gon!EV@dA1-AyY8wKh!-lY22^Nb**u<8k;kMzewZ5* z&s>Zt0|h9|_0%PPp2ETVkl0!oxoiaWXgIHkfmxfn8v6xCq>NGxVU%k3L_)%IQEHJs zlLeLkrCYDhs2oakzl3w8)u^_b@XnYXlQeY=*uS*eb*f$)+R0elPIhy0+G<9$rCsD{ zF7h{+m4#A126Uh%4vlSu_&ET zFqFKo&M`U{>7j)4q_p@5)ts@u`JRt-aZkKa?B>y?GG{hRj9^vn)|LYK`_W6U$gV2w z+@*bDYpJParCY9)t2L&H%fcBGjmB_|;itv$sFCwV3{ULpG9$Pc^?8AJ)Ll1cd)@l^ z0Qh4d3=b6kUhYZMB_krgUG4$4@`hC*TJXYrVqx*YFgLzH(0$bW^rN)KQ5k3v7dGYQ zIGJ1?iy}))EI)k?`JfwOcT1a%(y;qO_&{b7G`|*7$lj3@ev)F12ee-RD`!LQNgR)r zNro?1@z4bLUIx}k7WnJ~iu-hSn8Hu+`2MCD;y~3eS zsYn`Rzu^;3J301p-n+@XN@;I$B!bS zUMrsEhhdhjs;DknL{MiobPugJLWYb;(v7)oItMM)pF8}rV8HA}fTo8{oH76D7#*LY+=ysTgFN#>?Y@8?;eWU@ zHa$i&xN0~!qC0dU#%_oq8m8qbtUQP-^wb7`euRm-4pG`B66?!EJ2NYBgr8v_4{V?q zhK;yP%03kc0^wm&czhNU{M3r*;yU?@_@&#PnAb_y*`67@--hmh#eniu13Lb`F7|qO zHv_hAvIWZJe9BL8>0oj=X`W(ye)7K8fH#)Yj|o8;TMpTJ^S$H9sRR=Qyu_j z&whN(dQCB;GNk;ZiY>LbNF;Tp{))D*LoRamMEpdv{uHs?!!8;Cx3h5PT8-~pMFL!> z3iBc5f7OKB8Fb>@K!IRsxB2u@>3A(sD^3J~xPH`%G#acpno&OwWxqhJVV}fovzPCM zLl|j9yfB8MP$&m{HRhObD5fRKn@rvT1}5f&EgXlVV4Le*MTZLyv_v-lm|@%vXYL{x zZHj!{f-sqhDngqOX^N$*P#cTRX##=BOIU+W=?H6g<8nUESvwt&1jGE}>QjynmLU;G z&8$7B3Ox<1@RViX01rY;2~<$ZFDj{0esOXRC0e`?Dj%#ux5k-2WI*b-ci`sHhSah3#P4 z7hngvh4J5sJEKAGVd^prml><1p7@i9s~ne_Aj0)SuEr{AmCNzZ&7~MST$(_E8v7MH zrz38W-Ox%4ClX0;0E=;+l_0aB2vVU4QoiF{qwpFP`A&1|OLEkzN#Q%*(yqqiw^&8P zUEjFGg0YG{>V3@e$hA|SM4vdf(tZ&shWQ}&AO$0_gH#W*ul|-N>wuDl75jN&+4>8dfhpmU+^s z$4DlNIV!@puibXy6o==tjEGdhCd8v@R2>aZVh^S_7H70*9OPdMOz&32QH>TNXM5$Y zWD9~3$mh~Q5Trh)s(7~TJguNE0BYB|QQHrhWe41+GWDWR*)(>|1bw+;7+4M%FCw%d zQm5k2dfoQ+xLUOfPta+KJYA@XR3&-F@?}(F=}{T;vh&W%LX}1HU$2+szh20H&6eiB zjr_Mo>C%{wj&vgLoS#9~d8JHg=a1l9c3PZexlxfv%3kjhbI)_0Cy&B`t0i+jg=KP% zLw_f&n0Tqw1c$K%H#H-xXH#1}H*-_EGNjQGS_k2c-yQr_6pX{r$LTjD+eO!9Qj}c zlVUH)zU><*NWLLa{18Vf1S3yLtZy#;jZ5=lwQ$Nnl}V{~kc_zr3J$s}3#3%kb*j%@ zzuFv#p1|e|$u(@^Dq5-fq7blnt`M$V6;k>KVZUZG$$pK9{d$`HHnQKk4rMryMdvZ| zaiE~?5TY43mmMYGKEBGn44*$oof}PE*A9o0;7-@OH66JXL;MFBHifP7jceOWqDHvq z`@rKzSkkoOCa*6^}!YId@Gx=@a>Fy zw=m3P`L<1oC!4ddNJ+krC#J9@2iGS}5gd z>jT79v7L-KX-0mrruK_%W}kfewc$(_6{uVbjI1ZTW9vy|F@21g_clF0jXClo#7>pD zRL)vFS#5KMTA?h{uIps8V{naHVop|UYZPBr%_`T)Ds9awJFvU8jxM(A!G5iI42;+#}Q880aho8r(Ftpms zyy?^vV5}ZA`Q>`yE%EZ#pl6OdTw;^2Iq0@8AKi+3@6_}71TGOyH|jIsahOp+roe?4 zj?SV{n)kL$?lj2EPP6JR+qMUX;?}n2R*YC1@~L@m!IYWyTF=P0Rm-_|Ei$=>ZK_(# zb01jaW(O-H>g5Jr-rgY*S8C*9sX&apn-cUtp`MqI)9BLT_Be${g}L4b(eU8maO)f3 zQ!>SFW?vE!C08&N;ok->50St&5}wYy5y<5ilvT9;v&#Uad9`7&9PS$KZh@sWaduD%&o>vMF?kw35;Pe zq0@w&dJK$I{5kva=Z~*duSoJ;vA@3j%)}44FJk&Z4E%*kN(*NhVZi>SK1(~s9d88r$SEc}d+|>wx6j$ns%$C+yo`b2EoqqrRABzn2VIQn?S)?e zToi3?%t}$XW|vP)Tijoom@bu=`X?qPC3KpYPEJgXvtuSDJeS8zjeKV)TD!|Tl<-_0Lv`}SYS7wUK0|FO(x@5gLK&*K>XAD#loC44 zP^Ujbwa@MlrKINq4^i#Br>OQd9HUzGYOA41>dHu}aaxiJg-(;y=}D@4ZX^{-&jpfH zH!n$buY#mf>nCcGx@?kao|vSR&}ou7JxO)Wj--_ITp&qx@{&}ilqB_$gb68%&%(;k zSZ>pyVtz^5YU=2|RN884+B$u7UsFf-WiNiXMd_cjk5s6dl;GRfW(9+h-#b{?9VF@t z`^4=Ddx_!f9{nNBng4n;Nj{VcL zRWxl~ZWy4Zt*Z|R)U;KCfWW2>2rMscRr{UuZPJdVLIuBsi5K5@|s|2BhO&v;DV*YBL=K<`3qY9fjv$LAl zdH%YFC_~L(C5SR?>L^3aUxoA6u1Yrmt7SxH?Pr3UN^5@K{%u4wGxCgwsbgS*?FzsIA317;6jT1 zMs8lCl`J2xy&+&>V?MM5p<(S!oWu_=|Ksf{o3(XFV=9}qHJhD2q_M3-8nd!lT*o8c z7{{T&^gI+sWDS2E7d##;JSV{yvhXhcq#G{*41hg{+k$=pSEW1BPBk1I+IS((w8N9w z6$CVD5-UMKV_OF_Y7#pviRr+`YbCKO2xrtJR)TQGwhm|1Bz9O5(_xL*N@7mlo2r+@4kq8K%yl^}|-t)m#rPiSXNXBQmFn3K$| zl*+Cka8XlP2?7_}I&e`_*da)TEyfyzE|(`FoJ1z^h&t-t^O_Rep|apOwkuTn4& zcm`IFMN%ZC8SDa%?I4rvB*xkcUka2&wYWoxibzdQ=gRM}7mHl`O}ua7C)snVNQzQZ zYO?0*?{?$+%K07%>h2Ni1Ze>ADVK_(e$YRgg0$q`_ZzE! z^u3xFyCT>ZwRywl3(%|s4o1+d1rA0V4o1-IC1?g8V~1(h0V^YD)&eV|4J#vP_7XIM zp|Qg>>wu>bG;4vU(T1n-Drjb2J?buG|zCTIC(p!1T7W+nt|NP-Uzvc8#B|(JI z9sRFVH7Tky`d#Xg=N_`J6}A+SAQeGn|VbcUDyJFb{DO+IK zbYR&8DX)=~uz-S;U9p3Llr6A>IBy{l$jmyZdXPgk6*pk@o4tPY&4T2a$7yE4<%*cxlqeUi~KuVbzm5P0H5 zTB(DTNBB|yScNC=~K7-gaA|v{Ddz2gck;s-QxL^ z&f8hmoj}>`oI(9W;o(1ofk(IcR(LaY94}OG+I=RX`J8&l=7HgT2+v+ApoRd{QD;4|GZ`?eK=y3@F)nwU9|Z~@S*=w*7D6ZNCbPscUt1O{hfYH?h9Ksmrk!P za4GQnY4+svrO4I+zaPlf0>9r7e!o_e&FtIPiZ*?{g48H=UmAyLTerM|fLjZ^f~^0&d;1I|6PkusaT6cWgD>2FA74+%&dIQlr#^n{37Q z!R^CO-w8?v@qR^r_5~-^kA658-%#(0W0^*4pHd>Z7ZiJCsXc((2v*5bxQzg}%UdN! zuu4{i+Xs&%>{d#^cfoCt5C>k7JiLBHAJ~PTY$EPK!V`u`!n{fr4^@MYvfFC+zey$e zY$f_5x72mSH`Gd@%$g&6e6;p@;kwjhpO8zzu>)>Pz_A5x%Msj`*AK^=^68;Ibkx_l zc~f%|d4oA*$9$W>WD9(oBltEoz~n&lNu<6OP-?n*m>gg>*)0bsFxdhJ=m-u_4VGFp zpG<21YVXR=w*!%#@{t0OE%1?!;3GvfvgL@>*pZ5ABz&lVV_SWwBlu7O$F{>!Uq|b8 z!}0x+R9S3D$w>V||2Bw2DO`dM?A;%7(CEXnS_(p=4-#J9v}*L>TCIx0V#j|gzLUM3 zlRj9kLrDiE1JQg>6?j+Z(Axy_`&?34`PFk1^xdE+CzE?YDJ57(`G*C548=|9dg)~^ z3jWhw>9D53#_lTHI$${k+uGnc-Dew2r)_4N)z<)S8rw&$n!XsTvs1SYSVe)iHh4w% z@dmSKi}CiA4s^pS#=E;7*#ToGpwv+`&c;h^b%=@U1kk&$Ug(l z$#VV~;9m=zlLnlVRr$vn!47*CiBkg6dnmTO1JO|ZUVfSj2eo7=;m&(ncpnS9i}h>z zLu88qX0S>EX05PF8n8;X244W8-^f$rl{|nM43L0XTMUo}43KSsnc2UBWk!uHjNAbAU?y9v6)J0BvMRo3thRu$i^>qXp6snp6t)hE{5`D_11Br$g=lK#1M@~hFHhY6%qeEzo zeJ&=2s-8rr(Y;2AK4CtAM8jdyfJDP*Qca?*-VwL{lOUBg}?ne>i05uL!FjgJ5!wM*}NsHu5MNO6U(@o})4hTECC;!-lE zO$gWFd930G4FDrzZRdC%K$QoN>@s!mb$Iiz=@eK17PMmHsuL`L(eFttFrBl)Kws;d zS@!`W)6s92E8qY)7_|}y@XtL72LsDKCl2&=lnrXWlu$)lGe;S4(0LqOa%-*=IDjqf zX|sdgkxF}0K+xE#RlnxlyV|Pv3AYxQ045l1WCAQ-Pi2C8)=7fSHYnGsy>Xap>9=th zm;ff2tz-i1R8L}p)f<>+WrD^#l?_b0?oMSg*e}imzH5@BM(fFoh@EV=vJK@)CJT86 zN|%UYdb^3XYk8UeG$1J*$u>#s_3S?!9QZEa zuc+UC`T95z9*DjdXPUPs^K_z23i)I##Ije!*|p2K#;oKl-+bllAe?RpjmvH5@elI= zP%;Lef*HFC3ztmYf;C6{EfPaIh59xfi@Fj)a#p{foHb*XyjkyFMa6bfSsK#!WNrt_ zWL(IBjF|F*Na*SJD;hooE6%dC$-Za`J#>Sf%xGb}aS1Lfo?tr@A&>Z%M!=Z231Q*+!8s^BdMgwc)+w?DIG_A~z_e)B5nSQ&bdMozR zp1P11-cz4u?W5Z!vU{t0QNYf#Xk!fR{@ic|=5RK*E!XKgraSvuT^Nnu#=8r9$d~td zCBIkbLOoHh2Psdh(mvTsW7e`aC3GsHHcPDCFX};xuPeWx{eC8*Z4y%#kk_By)D7ok z`uQ=i1-DVMECX3tGe4V4)FY>Ufj6>mR%8-2$QAMEL_zzaJfbsGE6Xnq0Hg~7Y4iw? zR34|Zv3SgSlQ!EpS2||ylJ>t|${Surls87` zQ73#S?a6|$qWz!3} z$u8t3PvQMu(^^J2RT<%Qmk~8I`TNx^cQg8Y$&Z)bMVlGdB{uwpdMf=+{fLM3*;Im3 zjt>|s0jFS0)2Tu_#PS7=&^JjVvYh7gIEiFID!+mJyPR9HRFi*l4xNsDP7Mh4S%muG zfF_LY{5aX4VDAfXL_ZdD5-Y*V6AK!IEVt9eX-!^x_sxm%;c3zVU_DmZmt# zcC%7np%`p}m=Y-$3hJ)~{6kN?SKZ{h`<1s0rlbNNsCO*6XiSseSJl`-F)*$`F9I4{u~+S6~-moGFbg+%#dFxiywXVoli}Xa4wQqZ^X~h=A#^p z|KP9GLdq-3;MZb7*&Q!v3@4|sJ=LptrH115$-h@)j3c$UOMRt$5`rRD%0;oLCS9Ar!tT-W2upj10~I% zblCH7bd=p?2A@JG_LQTsHJto1--%Qz#VygC3kG6807uclYO`G)!ohiDF^2AR%y&AD z=HtR$o`n;(;ski`93FUUYR=Dcln~yRtjO=H_5S$Y+xT@eyeFyRfA*H^&~H$@)rFP! z!0nnhbt+y_rZ_J!!j}pndq%1tO)kGbXgr-3T4aRB$^BX>g79YBFo`QH!_53S&0w!xAeo;iFN|WAu2O)rzNN9r63@w2WTIh%b5}F7Gfg7)P=6-*l**(uZ@9*r) z?#!N-30oz#`a}M{zkNJPqwxBzRHqhD?&JNGTjvRU`@k&LK+^;^K`j3E1Vit00}d3ZOHCl2T%o8Gc`z;uY{M~)MV2} zodMbzR{3a{q zm@%Mv^iq>c0Fd{Wuo|vI6XW{F<36q0Y2zvx;FH*-nfRiz!Qa<^ojWC+mv%x>yLEEiVM|v4g9NIg6w4Q70S_s-_`!53=A9jK|qyG9J<2x#p$i7gzqqEOOH_Jua>+n%gCVXr( zk5f=;E-r`o99&WlT4$S{Rrb+cM-*ZO!}+N__e#}1SDK*=&ctG=7*pYrNKlGx3OS2& z@ij)EFdzu*&(vG-VWn(QXPu2W!@A$&333*XcJz(H+6HR(_*oK63$le+c|g`Y#X>H|l*LYfzAM z&{ZS_4&t{PC*{5GXX9*O zaT0|(VX#_Bhh>#}H5A_m$qXdIl;;kskC^lA5l|*6Xd!1Y6rbQr7uc#dbP*!hKu?5S zSdc$=jq!6FjzoDjhky_ez3!9i%m1jk`%5(O1-3>srvir^-;bi8YZH@AL!@4W&zi*w z?H>v1cL@l629?|hh|G_gnM~vTo@i<)s04pPQRTip;5JSosxuI3_c)b^ummUDQeGmZlNuJwMh2+HI`&QhTQXVi3S z5)fq#-IdGLex^vTyNPm=Pi;7-v)Z%CwVvgKMLZ1EO;0FiGqMOwFqX9oGsg-=@EcHwF+P0IFA7r)v9dUJi?eT zTy^rYRjKh0M_|ey4valgV{I*p_cZ$(_al9fbh`Z>t1})IaCp9lM;aGS$!X> z9uXS`BCIKwOPL5eGfv7fb2$?|%MDVY%4meT{bU&@8RJQY#c+ag_`hB*RUeA76*Pxc zDTq*42EyO3l3M>zpI0b?Gk3G5AoC1OpxRl#NP6DTMMfiUc_J+EkvCM3Cz~4D#+K0O zSd^{vWgsB1H7gC+5Wh;A&O7lN>#!;s-GL>^tWMka1lWZDVp>6>$@3j!k_>{m%rQm2yEJ;PIBY@l{@vnCQOCO@Kul5u&5M<|JJkQ& zgzTBNzoXB`-LTRLxH%U_C;L(~Zq_zA$iU*mPC7>7<}1@TXt(oc{<}R_X!a84o$r5v z25!ugghH3Tn>CNQYv&5zgIh)Yid47zSWmGZr$gI?Nf{nlh}Poo+~VlJ9;h-5-^gfm z<{0XoF>e!(7z}L}TH}>wFR&k+bK{68;%x8~BCA<&$sT!B+;or*3U}*Xfi%mR(xuRd zWfiYzivA4LIg_(J!?dv^YLwyiNq?OVD}Gpxvd)L<{sdsBgxXk31(`p!?9`P&(KWwA zPGwl;Ie@3~=Sfc)tm9@bw$+dEfYcC+K!Pn*BEIx)y|E|K)u;a$lA0s>i#h@pz?X zv^Mspg6=Iq$@8#)kiqMXvS>KpKP-v+;o4QD-PzG(#rQygPMg#FCF{}ZiNbpqT`na& zWc1`{r;Cj}0l23~U}AP8M^2tJ8o9*1n)>2lc3D~ZDE8onbh|89b*uYcBDn8#Qx0Dk z3!&sMrg=%B(1B~JRP+UIl|S!gVF!o_@2&V{cPvXkD%aG_*dCO&cS9o z%Zj85!|1&XYNvx>&Hl3Gl3;2Jc0S3Z*&WG;0ruCWPU`p11`Zq}tdrt>AIo)nxj#<0 zGwqce-O!SMU~Sm8^r)qt{-DliA-s0}obSL@nkK9u6HTAp)3SETOYam&-F_C$~C(%LD-~1RaO?9msWW)amUpiQP>W{&r4owOFWR^R|34 z!yB-nKQuk?nC#aI=3`%gDt_#=s`=gEKeL%=sZ8&4MO#ZrI?=o=Ny16Xw*M~WOMYoz zPlITbD6K{#&4mV(!eCv>;;4f5Gi`qTR(4S-`|Hp153KlhTVJWB8GA+s54zWS z>8fp7lr7b(ryZt-2LNMuxdLcr5^ z6XzeyGV`&XL|{z?7YUkrS--yK`uFb@Co=S_vF;LnARd{&o$p~|h{$(ReO?|f5H6t( zxXc2_DpqsH0(1Hg5wOlvV5@VS64oJ0~uo`bw*Y8doD~?#*=3G~%+G!#f(c zG?~0UT=cR0YOb4gzrup!evaU4tQ zo5(c_ZGXe>U2kkD&gfXja#&fao~(%+W0(xzbeW=TRbKb*7)HNMMeKroVL5cJw?3NA zz+=sr&A;&m^9Z+s6KdKW7$aUg+B#3>B*m`T7)HnHF}RdGc{uiATvA5^-^j?qdrNpL zae~iVKar;XZnXgbolHmr?`HPHdpj} Date: Wed, 4 Nov 2026 18:24:12 +0000 Subject: [PATCH 22/52] uk_rf_ab2026: apply review B's RF findings, K-4, K-6 and K-7 RF 1: Figure 8 and the two-year totals are cumulative. RF 2 (coordinator's call): the 4 November page is a 0-row primary; rows 56-57 are marked superseded. RF 3 option (b): the fuel-freeze totals return to revenue_change and the option leaves RULE10_KEYS (25 -> 14). RF 4-8 and nits; chart series labels move from component to series (H-1). K-4: the GBP 4.4bn rows propose the household-share key. K-7(c): gross_earnings_change. The tax-administration group takes OBR 3's _receipts key. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_01LwQq6JvunwtJCAckL6L4aM --- .../uk_rf_ab2026/NOTES.md | 88 ++++++++++++------ .../uk_rf_ab2026/claims_staged.jsonl.gz | Bin 52938 -> 53325 bytes .../uk_rf_ab2026/manifest.jsonl | 1 + 3 files changed, 63 insertions(+), 26 deletions(-) diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/NOTES.md index 67b5f9580..dd0604653 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/NOTES.md @@ -7,18 +7,18 @@ Staged 2026-10-29 by `stage.py`, which holds every row as data and rewrites `cla - **Rows:** 766 (767 before the review dropped row 397; see "Review fixes" at the end). - Text: 212 rows from the two press releases, the Spread thin page, both versions of the Spread thin PDF and one comment. - Workbook: 554 rows from the chart-data workbook RF links from the Spread thin page. -- **Primaries:** 7 (manifest.jsonl). +- **Primaries:** 8 (manifest.jsonl): the seven the lane fetched, plus the Spread thin page's 4 November version, which no row cites (README rule 7; see "The replaced PDF"). - **source_model:** - `landman_ttm`: the IPPR tax-benefit model behind RF's distributional and living-standards figures (488 rows). - `arithmetic`: RF's sums, averages and energy-bill construction on OBR, HMT and Ofgem inputs (148). - `survey_microdata`: LCF, NEED and HBAI analysis, including HBAI outturn years (42). - - Restated rows carry the original engine: `obr_efo_forecast` (42), `hmt_costing_obr_certified` (38), `administrative_data` (5), `hm_government_publication` (3, Government figures that aren't Table 6.1 cells). + - Restated rows carry the original engine: `obr_efo_forecast` (42), `hmt_costing_obr_certified` (38), `hm_government_publication` (5, Government figures that aren't Table 6.1 cells, including HM Treasury's Household Energy Payment count) and `administrative_data` (1, row 38's outturn). Two own rows also carry `administrative_data` (official counts RF computes from). `python3 stage.py --check-primaries ../../../.ab2026-primaries/uk_rf_ab2026` re-hashes every primary against the manifest. It also finds every text quote in its document (pdftotext `-layout`, `-raw` and default modes for the PDFs) and compares all 759 embedded workbook cells with the sheet XML. On 2026-10-29 it reported 0 problems. ## Primaries and access recipe -Every primary came from `curl -sSL` with the Safari User-Agent, HTTP 200, with no blocks or challenges. A second fetch at 17:02Z returned identical bytes for all seven. +Every primary came from `curl -sSL` with the Safari User-Agent, HTTP 200, with no blocks or challenges. A second fetch at 17:02Z returned identical bytes for all seven. The HTML pages embed per-request content (rotating nonces, tokens and related-article lists), so a later fetch's sha256 can't reproduce the manifest's; for those pages the check is that every staged quote still appears on the page (review B did this on 4 November). | primary | sha256 | bytes | fetched (UTC) | rows | |---|---|--:|---|--:| @@ -27,8 +27,9 @@ Every primary came from `curl -sSL` with the Safari User-Agent, HTTP 200, with n | Spread thin publication page | `66a999dc…69d792` | 152,177 | 16:16:38 | 18 | | Spread thin PDF, first version (`Spread-thin-1.pdf`) | `a669f39d…595603` | 1,410,169 | 16:16:41 | 3 | | Spread thin PDF, replaced version (`Spread-thin.pdf`) | `9536b6cd…5d9ff` | 1,410,388 | 16:17:29 | 147 | -| Chart-data workbook (`Final-clean-charts-all-sections-for-comms.xlsx`) | `a4e1c072…6e0d1` | 14,637,565 | 16:17:32 | 555 | +| Chart-data workbook (`Final-clean-charts-all-sections-for-comms.xlsx`) | `a4e1c072…6e0d1` | 14,637,565 | 16:17:32 | 554 | | Comment, *It's good to be flexible* (Nye Cominetti, 29 Oct) | `3510a245…62fcfc` | 131,349 | 16:20:03 | 1 | +| Spread thin publication page, 4 November version (review B's fetch) | `d81969ef…0b0259` | 154,801 | 4 Nov | 0 | - **HTML pages:** the stager quotes each page's own article text. Locators count body paragraphs after the dateline (press releases and comment) or name the bullet list ("measures bullet 2", "Key findings bullet 3"). - **PDFs:** read with `pdftotext -layout`, and with `-raw` for the two-column boxes. Locators give the PDF page, which matches the printed folio, and the paragraph, bullet, box, footnote or figure label. @@ -36,7 +37,8 @@ Every primary came from `curl -sSL` with the Safari User-Agent, HTTP 200, with n - `value_raw` is the cell's stored `` string from the sheet XML, at full precision, not the rounded display: F8's `_-"£"* #,##0_-` shows 145.33 as £145. - The quote is "series label | category | stored value". - Locators are `sheet F8, cell J8`. - - Each figure's plotted ranges come from the chart objects in the workbook. Figure 21 has no chart object; its column AD carries RF's own marker, "**USED IN CHART" (AD60). + - Each figure's plotted ranges come from the chart objects in the workbook. Figure 21's chart object is `chart34.xml` (grouped, plotting AD66:AD74); its column AD also carries RF's own marker, "**USED IN CHART" (AD60). Figure 8's column J (the net change) is staged though the chart doesn't plot it. + - Chart rows name their series in `conditions.series`. Only the rule 9 parts (Figure 8's two VAT-cut columns and p.17's extension) also carry `conditions.component`. - **Local copy:** the assignment's copy of `Spread-thin-1.pdf` (downloaded 08:46Z) hashes to the same `a669f39d…`, so v1 rows cite the fresh fetch. ## The replaced PDF (README rule 7) @@ -54,7 +56,8 @@ RF replaced the briefing on the morning of 29 October. - Elsewhere the lines only reflow. - **Which version rows cite:** v2 is the source for every PDF row except the three p.5 v1 rows (£158, £14, £144), which carry `conditions.document_version: "v1 (superseded)"`. - **Workbook:** uploaded at 12:16:43 GMT, it matches v2. Its Figure 8 net gains average about £145 over vigintiles 2–10. -- **Landing page:** its Key findings still printed v1's "gains £158 on average, against £144 for the richer half" at both fetches. Those two rows carry `document_version: "page text at fetch time; matches Spread thin v1, not v2"`. +- **Landing page:** its Key findings still printed v1's "gains £158 on average, against £144 for the richer half" at both 29 October fetches. Those two rows (56–57) cite the 29 October bytes and carry `document_version: "29 October page text (superseded by 4 November)"`. +- **The landing page changed by 4 November.** Review B's 4 November fetch (sha256 `d81969ef…0b0259`, 154,801 bytes, JSON-LD `dateModified` 2026-11-04T09:21:48Z) prints v2's £145 / £130 in Key findings bullet 3. It is manifest line 8 (README rule 7), with no rows: v2 rows 74 and 76 already carry £145 / £130, so they aren't staged again. "By", not "on": `dateModified` bounds the edit only from above. ## Harvest rules 8 and 9 @@ -97,8 +100,8 @@ The coordinator added two rules to the AB2026 README during this family's stagin | measure_key | rows | |---|--:| -| null (proposed key, or context) | 529 | -| `ab2026__package_household_energy_vat_and_fuel_duty` | 57 | +| null (proposed key, or context) | 532 | +| `ab2026__package_household_energy_vat_and_fuel_duty` | 54 | | `ab2026__electricity_vat_zero_rate_to_july_2027` | 42 (39 of them measure components) | | `ab2026__household_energy_payment_100_bands_a_to_d` | 31 | | `ab2026__fuel_duty_5p_cut_extended_and_rpi_rise_deferred` | 27 | @@ -113,11 +116,11 @@ The coordinator added two rules to the AB2026 README during this family's stagin | `ab2026__founders_relief_cgt` | 1 | | `ab2026__defence_investment_plan_further_savings` | 1 | -The 529 null rows split into 197 with a `proposed_measure_key` (listed below) and 332 context rows. The context rows cover fiscal aggregates, energy bill levels, living standards and poverty. They also include the three pre-AB26 component columns of F20 (57 rows: "pre-AB26 Direct tax changes", "pre-AB26 Benefit changes" and "pre-AB26 Indirect tax changes"), which score earlier events' measures. +The 532 null rows split into 200 with a `proposed_measure_key` (listed below) and 332 context rows. The context rows cover fiscal aggregates, energy bill levels, living standards and poverty. They also include the three pre-AB26 component columns of F20 (57 rows: "pre-AB26 Direct tax changes", "pre-AB26 Benefit changes" and "pre-AB26 Indirect tax changes"), which score earlier events' measures. ## Conventions -- **Sign.** Each row's `conditions.sign_convention` states it, and no value is re-signed from what the primary prints or stores. +- **Sign.** Each change row's `conditions.sign_convention` states it, and no value is re-signed from what the primary prints or stores. Figures 21 and 22 use "positive = growth"; the text growth rows and row 117 use "positive = gain to households"; row 10 uses "positive = saving (a reduction in the household's bills)". - Household rows use "positive = gain to households". Text printed as a fall ("£2,000 lower", "fall by 3.0 per cent", "worse off by an average of £61") carries a negative value, as in AB2025. - Exchequer rows say "positive = cost" or "positive = yield", matching how RF writes the figure. - Figure 10 rows keep RF's plotted sign, positive = raises revenue. RF flipped the OBR's borrowing sign (workbook F10 I15, "Signs flipped from source"). @@ -132,12 +135,13 @@ The 529 null rows split into 197 with a `proposed_measure_key` (listed below) an - `conditions.price_basis` records the basis on each row. - **Period:** - Fiscal-year figures carry `fy` and the end year. - - The temporary package (Figure 8, the 52/48 split, the halves' averages) covers 2026-27 and 2027-28 together. Those rows carry no `fy`, `period` 2028 and `conditions.horizon`. + - The temporary package (Figure 8, the 52/48 split, the halves' averages and gap, and the £4.4bn) covers 2026-27 and 2027-28 together. Those rows carry no `fy`, `period` 2028 (the £4.4bn: 2027, as printed or read), `time_basis: fiscal_year` and `conditions.horizon`. Figure 8 shows each measure's "total cash impact" over the two years (v2 p.20; workbook F8 N35), so its rows and the text figures that restate it are `cumulative`, and the F8 £ columns' normalization reads "£ per household as stored: total cash impact over 2026-27 and 2027-28 (Figure 8 note)". + - The gas standing-charge rows (115–125) carry no `fy`: RF prints no year. They sit at period 2028 with `conditions.horizon` "year not printed: the change lands by the end of 2027 or soon after (p.18)", at medium confidence. - Averages over a span (for example "£11 billion a year on average over the next five years") take the span's end year with `statistic: annual average`. - Calendar figures (the 2027 average bill, decade and Parliament RHDI growth) are `annual`; Q1-2027 cap levels and "currently" figures are `point_in_time`. - **Value kinds.** RF prints point estimates throughout. - "under", "up to" and "less than" give `range_high`; "over" and "just over" give `range_low`. - - Totals over several years and growth since a base year are `cumulative`. + - Totals over several years and growth since a base year are `cumulative`: 195 rows, including Figure 8's 114 and the 11 text rows that restate Figure 8 or the £4.4bn. - **Income groups.** `income_group` uses the AB2025 spellings `vigintile_N`, `decile_N`, `quintile_N`, `bottom_half` and `top_half`, the last two for RF's "poorer / richer half". The AB2025 NOTES flagged `bottom_half` / `top_half` for an ingest ruling. - **Geography.** - F8 rows are "UK", with `modelling_geography` quoting RF's note (Council Tax and VAT cuts modelled for GB, Fuel Duty for the UK). @@ -147,8 +151,8 @@ The 529 null rows split into 197 with a `proposed_measure_key` (listed below) an - F20's net and pre-AB26 columns, and the Parliament-wide text figures (99 rows), use `rf_start_of_parliament_policy` ("relative to the baseline inherited from Chancellor Hunt"). - Every AB26 measure row is null: pre-Budget policy, current law at the scoring date. That includes F20's AB26 component columns, increments within RF's stack. - **Attribution.** - - `restated` (84 rows): HM Treasury Table 6.1 costings, OBR forecasts and the Government's count of carers. `note` names the original on each. - - `own` (683): everything RF computes, including its sums and averages of official figures. + - `restated` (86 rows): HM Treasury Table 6.1 costings, OBR forecasts, the Government's count of carers and HM Treasury's count of Household Energy Payment households (rows 55 and 72, which RF repeats uncited). `note` names the original on each. + - `own` (680): everything RF computes, including its sums and averages of official figures. - `benchmark_class` is `administrative_fact` for outturns and official counts (53 rows), including the 2024-25 HBAI outturn year in F21–F23, and `different_model` otherwise. - **Worked examples (mode 3):** the computed outcomes only (rule 8). - The p.25 Pension Tax Allowance comparison: £34 Income Tax and £10 NICs for the worker, "nothing" for the pensioner. The £12,739 earnings input sits in `household_type`. @@ -159,11 +163,11 @@ The 529 null rows split into 197 with a `proposed_measure_key` (listed below) an ### Registered keys -- **`ab2026__package_household_energy_vat_and_fuel_duty` (Table 6.1 lines 1, 2 and 3).** This key covers RF's "cost of living package": Figure 8's net change and per cent change, the 52 / 48 per cent split, the halves' averages and gap, the £4.4bn costing and the landing page's £158 / £144. +- **`ab2026__package_household_energy_vat_and_fuel_duty` (Table 6.1 lines 1, 2 and 3).** This key covers RF's "cost of living package": Figure 8's net change and per cent change, the 52 / 48 per cent split, the halves' averages and gap, and the landing page's £158 / £144. The £4.4bn costing is not on it (review A, K-4; see the proposed keys). - RF's package is the Household Energy Payment, the electricity VAT cut ("original" six months plus the three-month "extension") and the Fuel Duty delay. - Table 6.1 line 2's title covers the whole cut ("from 1 October 2026 to 1 July 2027"), and its costs (£755m in 2026-27, £305m in 2027-28) include the original six months. - The registry's own modifier note gives the same reading: "about 57% of 2026-27 … about 22% of 2027-28". So RF's package equals lines 1, 2 and 3 exactly. - - The £4.4bn rows also carry `costing_basis`: RF counts Fuel Duty at the share accruing directly to households (fn 19). + - The £4.4bn rows (34, 54, 127) count Fuel Duty only at the share accruing directly to households (fn 19, `costing_basis`), so their scope is not exactly lines 1–3 (rule 3) and they carry the proposed key `ab2026__package_household_energy_vat_and_household_share_of_fuel_duty`. - **`ab2026__electricity_vat_zero_rate_to_july_2027` (line 2):** - **The whole cut:** the £40 "total nine-month benefit of the cut", and Figure 7's "Electricity VAT cut" series split by financial year (£26.77 in 2026-27, £12.80 in 2027-28). - **Measure components (rule 9, `row_type: measure_component`):** Figure 8's "Energy VAT cut (original)" (the six months to the end of Q1 2027, 19 vigintile rows), "Energy VAT cut extension" (April to 1 July 2027, 19 rows), and p.17's extension "worth £12". `conditions.component` and `measure_part` name the part. Never sum them with a whole-cut row. @@ -193,16 +197,17 @@ Each row below has `measure_key: null` and this key in `conditions.proposed_meas | `ab2026__package_rf_ab26_permanent_personal_measures` | 19 | F20 "AB26 Changes" = lines 52, 56, 61, 62 plus the gas standing charge (the workbook's column K equals columns F + G + I + J). | | `ab2026_option__lha_unfreeze_2027_28` | 17 | Unfreezing Local Housing Allowance in 2027-28 instead of prolonging the freeze (no Table 6.1 line). Figure 9 "Would benefit if LHA were unfrozen"; "over a million families lose out, 91 per cent … in the bottom half". | | `ab2026__package_tax_decisions_since_march_2026` | 13 | The OBR's Table 3.1 "Direct effect of tax measures" since March, sign flipped. F10 "Net direct effects", the p.23 labels, and p.4 "about £5 billion annually from next year". Differs from Table 6.1's tax total (-£1,749m vs -£1,645m in 2026-27). | -| `ab2026_option__tax_decisions_with_continued_fuel_duty_freeze` | 11 | The OBR's direct effect of tax measures plus keeping Fuel Duty at current rates: F10 "Net effect if Fuel Duty frozen" and its p.23 labels. | -| `ab2026__package_tax_administration_compliance_and_debt_collection` | 7 | The OBR's Table 3.1 group "Tax administration, compliance and debt collection measures" (it includes line 45 and timely payment for Self Assessment; RF doesn't list the lines). F10 row 52; p.4 "over £3 billion annually by 2030-31". | -| `ab2026_option__fuel_duty_continued_freeze` | 6 | Fuel Duty rates held at 52.95p a litre instead of rising to 60.59p in July 2027, relative to the announced policy: the OBR's constant-duty-rate alternative less its central forecast. F10 "… and if Fuel Duty frozen"; p.24 £4.3bn in 2029-30. | +| `ab2026_option__tax_decisions_with_continued_fuel_duty_freeze` | 11 | The OBR's direct effect of tax measures plus keeping Fuel Duty at current rates: F10 "Net effect if Fuel Duty frozen" and its p.23 labels. On `revenue_change` (README rule 10; see "README rule 10" below). | +| `ab2026__package_tax_administration_compliance_and_debt_collection_receipts` | 7 | The OBR's Table 3.1 group "Tax administration, compliance and debt collection measures", which is the receipts of Table 6.1 lines 45 and 46 (OBR review B, finding 3; it includes timely payment for Self Assessment; RF doesn't list the lines). F10 row 52; p.4 "over £3 billion annually by 2030-31". | +| `ab2026_option__fuel_duty_continued_freeze` | 6 | Fuel Duty rates held at 52.95p a litre instead of rising to 60.59p in July 2027, relative to the announced policy: the OBR's constant-duty-rate alternative less its central forecast. F10 "… and if Fuel Duty frozen"; p.24 £4.3bn in 2029-30. A receipts cost, on `revenue_change`. For tranche 2: the OBR family holds the same alternative's £4.3bn as an effect on borrowing (paragraph 3.43, `decisions_effect_on_borrowing`), and RF states a receipts cost; the ingest settles whether they compare. | | `ab2026__package_rf_other_tax_measures` | 6 | RF's residual in F10 "Other tax measures": OBR net direct effects less Fuel Duty, Bank Surcharge, capital allowances, savings and property, and tax administration. | +| `ab2026__package_household_energy_vat_and_household_share_of_fuel_duty` | 3 | Table 6.1 lines 1 and 2 plus the household-incidence share of line 3: RF's "£4.4 billion cost of living package", which counts Fuel Duty "at the share accruing directly to households" (fn 19; about half of line 3's £1.4bn). Lines 1 and 2 are £3,685m over 2026-27 and 2027-28. A two-year total, `cumulative` (review A, K-4). | | `ab2026__package_household_energy_payment_and_electricity_vat` | 5 | Lines 1 and 2: "A £3.7 billion package of household support"; p.17 "£3.4 billion in 2026-27", "£0.3 billion in 2027-28", 45 per cent. The sums show line 2 is the whole cut; see the discrepancies below. | | `ab2026__package_government_decisions_since_march_2026` | 3 | The OBR's total effect of the Budget's decisions on borrowing, direct and indirect, including decisions not on Table 6.1: "an average of £2 billion a year" (two documents); fn 47 "no overall impact". | | `ab2026__package_rdel_decisions_excluding_household_energy_payment` | 3 | All the Budget's RDEL decisions over the Spending Review period except the Household Energy Payment: "£3 billion a year", "£9 billion top up" (two documents). A fiscal aggregate, not a programme allocation. | | `ab2026__package_rf_absolute_policy_changes` | 2 | Every policy decision counted at absolute value (RF's fiscal-event size measure): "£7 billion". RF lists no lines. | | `ab2026__package_rf_energy_bill_support_announced_at_budget` | 2 | "Support of up to £116 announced today" and "less than a fifth" of the 2027 bill rise. RF doesn't itemise the £116; it includes the £100 Household Energy Payment. | -| `ab2026_option__fuel_duty_rpi_uprating_since_2011` | 2 | Fuel Duty uprated by RPI since January 2011: "around £20 billion" a year (two documents). | +| `ab2026_option__fuel_duty_rpi_uprating_since_2011` | 2 | Fuel Duty uprated by RPI since January 2011: "around £20 billion" a year (two documents). It mirrors the OBR family's `ab2026__package_fuel_duty_freezes_since_2010_11` (EFO paragraph 3.43): one pair of paths seen from opposite sides, kept as separate keys (review A, K-6). RF's +£20bn is the revenue the freezes forgo relative to RPI uprating, staged "positive = yield" from the option's side. | | `ab2026__minimum_wage_uprating_april_2027` | 2 | The April 2027 minimum wage rates the Chancellor confirmed (no Table 6.1 line): "worth £1,040 a year for someone working 40 hours a week" (PDF p.12 and the comment). The rates (3.9 per cent; £13.21; 50p) sit in `reform_hint`. | | `ab2026__gas_supplier_levy_funded_from_general_taxation` | 1 | Funding the Gas Supplier Levy from general taxation: "save the typical household £7 in 2027-28" (fn 12). No Table 6.1 line found. | @@ -235,7 +240,7 @@ None of these is a registered Metric. Each row carries one as `proposed_metric`, - `real_income_change` (£ change in a real income statistic). - `income_change_gap`. - `net_weekly_income`. - - `annual_earnings_change`. + - `gross_earnings_change` (the AB2025 spelling, which the red book family uses for the same minimum wage measure; review A, K-7(c)). ## Not staged, and why @@ -273,6 +278,7 @@ None of these is a registered Metric. Each row carries one as `proposed_metric`, - Within the staged sheets, the family leaves out: - Working and input blocks: F9's raw winner tables, F10's HMT and OBR input rows 5–40, F21's percentile columns other than AD, and F23's level, raw and scenario blocks. - F22's second chart (bottom 7 / middle 6 / top 7 vigintiles since 2019-20), which the briefing doesn't print. + - F23's second chart (AP72:AR73, 2029-30 and 2030-31), which the briefing doesn't print. - **Text numbers left out:** - **Macro forecasts and statistics:** - CPI paths (6.5 / 3.9 per cent, 43 / 22 per cent, 2.1 / 3.8 per cent, 16 / 13 per cent). @@ -297,6 +303,7 @@ None of these is a registered Metric. Each row carries one as `proposed_metric`, - The G7, OECD and IMF borrowing comparison (3 and 0.3 points). - The £72bn average of Reeves's two Budgets, which has no single period. - The 19th of 34 fiscal events (a rank). + - PR28's "mid-2029" date for living standards to recover, and its ranking of the Parliament as the "second worst … since the 1950s" (a date and a rank, not an estimate of a measure). - **Other design:** £100, bands A-D, 5p, 3p / 2p, 2 points, 3 to 6 per cent, the 31 per cent headline bank rate, 20 / 25 per cent, the £1.5bn Good Growth lending facility, the 20 per cent equity loan and 2.5 per cent deposit, the £110,000 / £140,000 income caps, £86.45, £204, 16 hours, £1 for £1, the 55 per cent UC taper, Business Rates retention shares (20 / 50 / 70 / 100 per cent), and the 3 and 3.5 per cent defence targets. - **Third-party figures:** - Anthropic's $65bn (£48bn) round and the "1.5 per cent" comparison. @@ -331,13 +338,13 @@ None of these is a registered Metric. Each row carries one as `proposed_metric`, ## Discrepancies in the primaries (reported, not fixed) -1. **v1 and v2 disagree on the income halves.** v1 prints £158 / £14 / £144; v2 prints £145 / £15 / £130, and v2 p.19 hedges them as "under £145 … over £130 … less than £15". The landing page still printed £158 / £144 when fetched. +1. **v1 and v2 disagree on the income halves.** v1 prints £158 / £14 / £144; v2 prints £145 / £15 / £130, and v2 p.19 hedges them as "under £145 … over £130 … less than £15". The landing page still printed £158 / £144 when fetched on 29 October; by 4 November it printed v2's £145 / £130 (manifest line 8). 2. **The £4.4bn package's years.** p.19 calls it "the Government’s £4.4 billion cost of living package for 2025-26 and 2026-27". The measures fall in 2026-27 and 2027-28 (Table 6.1 lines 1–3). The press release and p.47 print no years. The family stages the period as printed, at medium confidence. 3. **The £3.4bn's scope.** p.17 says "Together, these measures cost £3.4 billion in 2026-27" of the Household Energy Payment and "a three-month extension to the VAT cut". The extension falls in 2027-28, and £3.4bn equals lines 1 and 2 in 2026-27 (£2,625m + £755m), so the figure includes the original six months of the VAT cut. 4. **Your First Home's sign.** p.21 calls £95m in 2029-30 "a cost". Table 6.1 line 37 prints +95, a yield ("net income from developer fees"). 5. **SAHP top-up** (not staged, rule 8). p.21's "£1.5 billion top-up" differs from Table 6.1 line 35, which sums to £1,750m over 2027-28 to 2030-31. 6. **Good Growth Funds** (not staged, rule 8). p.33's "£285 million in 2027-28" differs from line 20, which prints -£340m that year (the line includes the lending subsidy). -7. **Tax gap measures.** p.4's "over £3 billion annually by 2030-31" doesn't match the workbook's OBR series (F10 row 52): 2.89 in 2030-31 and 3.64 in 2031-32. +7. **Tax gap measures.** p.4's "over £3 billion annually by 2030-31" doesn't match the workbook's OBR series (F10 row 52): 2.89 in 2030-31 and 3.64 in 2031-32. Row 67 records it in `conditions.discrepancy` (review A, K-7(h)). 8. **Figure 20's AB26 component.** The note says "‘AB26 Changes’ includes the Pension Tax Allowance, removal of standing charges on gas bills, and the Machine Games Levy". The workbook's AB26 Changes also includes "Property and savings tax rise" (column K = F + G + I + J). The workbook labels the Pension Tax Allowance "Pensioner Personal Allowance". 9. **Figure 10's Fuel Duty series.** Its note says the series shows "Fuel Duty changes announced in May and October", but the values (-0.575, -0.78) equal Table 6.1 line 3 alone. 10. **Primary balance years.** p.39 says the £107bn of higher tax revenues runs to the end of the Parliament (2028-29). Footnote 53 computes it "between 2023-24 and 2029-30, multiplied by forecasted GDP in 2029-30". @@ -348,14 +355,14 @@ None of these is a registered Metric. Each row carries one as `proposed_metric`, 12. **Footnote 50's figure reference.** It reads "From the forecasts shown in Figure 7", but the current-balance forecasts appear in Figure 18; Figure 7 is the energy bill chart. 13. **Capacity vs Table 6.2** (not staged, rule 8). The NWF's £15.4bn and BBB's £5bn "capacity" boosts don't match Table 6.2's financial transactions: NWF line 2 sums to £10.2bn over 2028-29 to 2031-32, BBB line 3 to £0.75bn. 14. *(Withdrawn on 3 November: the README's rule 3 example was corrected, so RF's package equals the registered one; see "Measure keys".)* -15. **The VAT element's description.** PR28 para 5 and p.17 bullet 2 describe the package's VAT element as the "three-month extension", and p.17 calls the six-month cut "support announced before this Budget". Yet Figure 8, the 52/48 split and the £4.4bn all include "Energy VAT cut (original)". +15. **The VAT element's description.** PR28 para 5 and p.17's heading and paragraph 1 describe the package's VAT element as the "three-month extension", and p.17 calls the six-month cut "support announced before this Budget". Yet Figure 8, the 52/48 split and the £4.4bn all include "Energy VAT cut (original)". ## Review fixes, 3 November 2026 The RF review (`harvest-staging/ab2026/reviews/verify-rf.md` on the staging branch) approved this family with fixes; all thirteen are applied, finding 1 as the `P_SAV_PROP` constant and the rest in an explicit block before `write()` in `stage.py`. No primary was re-opened (the session's network policy refused resolutionfoundation.org); no printed value changed. 1. The 25 rows for exactly lines 61 and 62 use `ab2026__package_property_and_savings_income_rates`, README rule 2's name (F10 row 51 equals lines 61 + 62 in every year). -2. The passage and discrepancy 14 that contradicted README rule 3's old example are gone; that example has been corrected. +2. The passage that contradicted README rule 3's old example is gone, and discrepancy 14 is withdrawn (kept as a one-line marker); that example has been corrected. 3. Row 168's note names the Government's Defence Investment Plan projection. 4. Row 38's note names the ONS/OBR outturn. 5. The v1→v2 record notes Figure 8's lower Council Tax rebate bars. @@ -384,4 +391,33 @@ RF's three OBR groupings are restatements of the OBR's own Table 3.1 rows, so th ## README rule 10, 3 November 2026 -Under README rule 10, `stage.py` moves the family's totals of the Budget's decisions from `revenue_change` to `decisions_effect_on_borrowing`: Table 6.1's total of policy decisions (1 row), the OBR's direct effect of tax measures that Figure 10 plots (13 rows) and that total with fuel duty frozen (11 rows). Counting from 0, the rows are 66, 102, 141–151, 385–390 and 432–437. The OBR's total effect of the Budget's decisions (3 rows) already took the metric, and single measures and groups of measures keep `revenue_change`. No value or sign changed. The contract checker reports 766 rows, 0 problems, and a re-run of `stage.py` reproduces the gzip byte for byte. +Under README rule 10, `stage.py` moves the family's totals of the Budget's decisions from `revenue_change` to `decisions_effect_on_borrowing`: Table 6.1's total of policy decisions (1 row) and the OBR's direct effect of tax measures that Figure 10 plots (13 rows), 14 rows in all. Counting from 0, the rows are 66, 102, 141–146 and 385–390. The OBR's total effect of the Budget's decisions (3 rows) already took the metric, so 17 rows carry it. Single measures and groups of measures keep `revenue_change`. No value or sign changed. + +**Not moved (4 November 2026, review B's RF finding 3, option (b), the coordinator's call):** the 11 rows of Figure 10's "Net effect if Fuel Duty frozen" (`ab2026_option__tax_decisions_with_continued_fuel_duty_freeze`; rows 147–151 and 432–437), which the 3 November pass had moved, are back on `revenue_change`. These totals add the freeze's receipts effect, which row 154's note derives from the OBR's constant-duty-rate Fuel Duty forecast less its central forecast, and RF's title for Figure 10 frames the chart as revenue ("Measures announced raise revenues"). Under rule 10 they are a tax-change estimate whose evidence doesn't establish PSNB scope. The OBR states the freeze's effect on borrowing for 2029-30 only (paragraph 3.43, from Table 7.1), and RF's values for the other years are receipts differences, so they are not recombinations of a stated effect on borrowing. Rows 154 and 397–401 (`ab2026_option__fuel_duty_continued_freeze`) stay on `revenue_change` too. + +## Review fixes, 4 November 2026 + +Two independent reviews of PR #231 at 4f424b9 approved this family with fixes: review B (RF findings 1–8 and nits) and review A (K-4, K-6, K-7(c), K-7(d), K-7(h) and H-1). All are applied in a block before `write()` in `stage.py`, with the coordinator's calls on findings 2 and 3. No printed or stored value changed. Indices are 4f424b9's. + +1. **Two-year totals (finding 1).** All 114 Figure 8 rows (245–358), the text rows that restate them (56, 57, 61, 63, 74, 76), the gap rows 62 and 75 and the £4.4bn rows 34, 54 and 127 are `cumulative`. The 95 F8 £ rows' normalization reads "£ per household as stored: total cash impact over 2026-27 and 2027-28 (Figure 8 note)"; the K rows keep theirs. +2. **The landing page changed by 4 November (finding 2, rule 7).** Manifest line 8 records the 4 November page (no rows cite it). Its £145 / £130 are not staged again (the coordinator's call; v2 rows 74 and 76 carry them). Rows 56–57 carry `document_version: "29 October page text (superseded by 4 November)"` and a "Superseded" sentence. +3. **Rule 10 (finding 3, option (b)).** Rows 147–151 and 432–437 are back on `revenue_change`; the stager's rule 10 count is 14 (see "README rule 10"). +4. **Restated counts (finding 4).** Rows 55 and 72 are `restated`, `hm_government_publication`, medium confidence, with a note naming HM Treasury. +5. **Tallies (finding 5).** The workbook has 554 rows, the restated source models are corrected, and the attribution split recounts from the rows (`own` 680, `restated` 86). +6. **Gas standing charge (finding 6).** Rows 115–125 carry `conditions.horizon` and medium confidence. +7. **Signs (finding 7, and K-7(d)).** The 122 Figure 21–22 rows (629–750) carry "positive = growth"; text rows 21, 23, 89, 117, 198, 200, 203 and 204 carry "positive = gain to households"; row 10 reads "positive = saving (a reduction in the household's bills)". +8. **Nits (finding 8):** + - chart rows name their series in `conditions.series` (310 rows), and `component` stays only on the 39 rule 9 parts of line 2 (review A, H-1). Seven text context rows that used `component` for a driver of a forecast change, or a scenario's spending type, use `driver` (rows 51, 80–82, 91, 92) and `spending_type` (row 170); + - F22's category labels read "2025-26", not "2025- 26"; + - stored F7 C/I/M and F9 cells say "as stored"; + - F7 rows 222, 226, 236 and 241 name their support series in `reform_hint`; + - gap rows 30, 62 and 75 carry `price_basis: cash terms (Figure 8)`; + - the Primaries notes, Figure 21's chart object, F23's second chart, F8 J, discrepancy 15's locator, Review fixes item 2 and the "Not staged" list are corrected. +9. **K-4.** The £4.4bn (34, 54, 127) moves to `ab2026__package_household_energy_vat_and_household_share_of_fuel_duty`. +10. **K-6.** `ab2026_option__fuel_duty_rpi_uprating_since_2011` stays separate from the OBR's `ab2026__package_fuel_duty_freezes_since_2010_11`; each family's NOTES names the other. +11. **K-7(c).** Rows 88 and 211 propose `gross_earnings_change`. +12. **K-7(h).** Row 67 carries `conditions.discrepancy`. +13. **OBR review B, finding 3.** The tax administration group's key is `ab2026__package_tax_administration_compliance_and_debt_collection_receipts`, as the OBR family's is. + +- **For tranche 2:** `ab2026_option__fuel_duty_continued_freeze` holds RF's receipts cost (rows 154, 397–401, `revenue_change`) and the OBR's £4.3bn effect on borrowing (paragraph 3.43). RF rows 385–390 restate the OBR workbook's receipts total exactly, on the tax-decisions key with the borrowing metric. +- **Verification:** the contract checker reports 766 rows, 0 problems, and a re-run of `stage.py` in a scratch directory reproduces the gzip byte for byte. diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/claims_staged.jsonl.gz index 13af90a5f1665918f1c5c990c0f4de587963ac88..fafd85c23f54d0320b908c4f045e89cc8e45d293 100644 GIT binary patch literal 53325 zcmV)kK%l=LiwFP!00002|LncpvfDd{RV;a#Qac1F&`uOu<7>qo;FJAE9csGw-I*({)M%{P;&*In3=5FN0d;vcX zlW;&IqA8k1?qc$z8x5!YUq65IfsDh54&6AxACA3v==u)5gU8}e-9@sP18h?Mi*ZXSSY4Q_^~$v_GGU*^(IsH9eXl-0z0STzoQWLgpQmU1?FcYPgm(= zyW$L|ZUEHB>lXtMfRxJr{mdN>Y0CESGscV~my?(AiK=w89)9YF@Ut*tT;RfHA|n!w z-81q)f*zcB>T0INc715-R-am9TOSO@qY?abG#>T)w(aVM+v`(B*9O#7b*j3WI#&Cd zt7s!tGw8VA>)E|*rSX(TVy8o&Nj7;KE`kw%0k|sh5}z&hJ&0hOZp91o4U-%)aD6uz z0ttvaObC33^EwKYj?C>Ra z(j|8I5<9TO4qIY}FR=q_>10dn@FjM#C3ey!K4%j5&xH_}Z&HpRUQ8BqGI#)qQq@e- zT*ih3{}KY*MjVBw&m3UkB$wA1p3ONke_dz1v^D$^;A0=4Z1tA{TjBTNAadXdP7w~p zHHkZ5o16X;gnhwyHT=Z)eny=Piv@(@3$Tp|yaZ)vNaqYuCxbbDeCPTLyrrrt>Hl?= z^mFHv7mPT};mc+pNA5j`q?}aG{>T6G->V7-XkNdrsiqmm4_sh%wWDi~s|wh6L>~%# z9*&ow?b$M>6p{LwJHW>|&fFb-0^V?ME#ahvOjqDINC z#U%%zc#su00{V=Oyv1x4(zS9(FDLT>@Z6Aieo1)#7FIANoGFNqJ>o``+=o#z1@^Wx zt_K4olRS9fEMAB9^4m2xNN4L=t!e=_4U>YuBIt~4iLjxL39zZob%yTZlBsmPIUnaGj%T;xjNe77Ih2j$Zl0xy_;aPma!en_Ll8+#lX50C7cwFj~n zs8o|_?~Z0(00wa6BDy+bm&@t_p#9rm0vtKYNHY)T3qM0rG$(NbP?eAFDZFAZi4l@s z!`==(A3*XmVOULGb>6fN0QW$$i3h~>k_RAoQV#eHCLCVSL2pQdp}o@W72*EG$Z1LB zyd>mwBU}+Fauq3v0`R{O8^YPsXNNr!FXnU5;3a74qN_wvfbIie8cxY6oYPXGZIY;( z&E#cNf#16aLRDb#h!$}b#IYirmeBra{t|$N5J;5DZw6Ux%TXf^!Xyc2V66G$^px)f z!*E8)2r5 zb^#&1qYuu^8%E)ARSU@GZpx*&5tpW1(yJpnxk^mE*N1TY+^H>CKh^V2?{Imi&B70(IEG-{v1#b=>?Y^i-NN&EcF(T<&B znwmmV4r{IGpsn(D;6U!_l!sPPbrcA9*(xKxW}Wd9^PnTswh90HSBiflQ~y^OeTt_5 zc|dF{@#%oCPOgW07^iH<*oEJZRT-@+ZkhzIWXNI3oO8=gg!!d&t& z=P{~>xrcd{qlN2t#tZ7_HpvbNI&RDZtUOAT!rF3-Rkg@`k??fSzo`t$$PnHi)DVVs z0X|5RC!BgQ9kpHxjmhM!5wmMFRl(dh{DS>lrJ-M90K&TsX#wBhGrelsNP#qGR$j9cw&k8v?ovdS z#$7I?7VS`BQv?@r2M5THeogqjn~0@8j-Ij0R?>7~S);|U=v2;D42NWfu3IW=3Tb_% zuvZa&KfDFurZMxpdI_Duft~PtRTWP8P4Oql%qCYXnfr^F#PBR%=!CQ^2 zbIy-R2z*$Q{!#sh7ikB>yR7Me2ob|c&8;OxP$0pn;&dExf>11@1Xkvf#U+00PSdz`1Yn zB>!fSJg}Vt&ZHQCkBYrB=GZqwHF=+ly3!YTQR5>Z8qOIbIOuM0Mhh%LnS0J+jw$rP z#EIdYM&Ws?`nV1K;8A~6r9yJeSgxt$aff^Y=S6xAe5@FbIU1lqI<&!ukhFP`)fs76(V*Z;-<1fwv1}&WNlxCVa=_w_Rp5 zsQMtd2>;EF-30Z$x(OmN&ashH3IMv z181@UtDsnFu_!qGxbS*(KY+EDBfz>yKQZHM%n%a^WG{vDKb7neJY&I7wy@zvES`9N z!h?2kTB|&#?k7BpU~V2T7+~)wIPDMULnvzI!7OZ`jNTR8Mpe?1M|6O^2Ye4kCW+@1 z$N{JSQkDmZtwk>@hzb3iMD7Z2V{gy+8nJbcNZcyT zE(%{$+cOC7vz8$x&F`_}2y1#@VJZyk&o6!-QP=OVUgMz8*E;?z0)VE3PgogIfQ|#uIYm;BY=o4G#dm1ooZXL4Y|VdJ zEf%Jj-xf?*`WmnZia;#-@>^Xyt8`%DRiFx-ezgJ5$j89!HNb00EeO1MMPx0~NP_B_ zf@BME5Uuc@Q7M*8| zl$Y?0Jiy7AjS&%j(3&6Kz4^Yym<#@c?@=8ug$0Upq+AeO-zUw6AKufbL7tNae!%`1 z0R%Q$FUio2A`dPP=!CWVCcuL5*v8g3YV2F(Y;6Hw(Y)<&&@1PXTom+v*jY-j>yq&5 z`Z#sa&=kU0{~}3Bz zg^_y-%+t=YNJR1!7`*pb`BJ6uB8M;WtHU&j=pO(TWsmraLPb`lwwCqu8=Sg`(zSAX zbW`@~rFVLOm#heSMRHsl_HDtjqjs(7pVFR_$aPW3)nvzG+qSaPaVyP@UgG!bYR?7l zyM|Z8#f%M4yQ5|EKW*y_r^7vlSd{qaa9l<$wJ%W1xCx7`ypGk7$anLt2&Oa;hgKj1 z^?fwpI!f2b?USps)A02w8fwdwIMd9`(%t}jwji8iSM8jxX&DgvI3~?vk^Nh50x(AG z)@s^OKv^dr(WRx0?By>;X&3(|NsMgiu7rCy*Dylor31`~1dcIVdp zy1#<1S#8(==mE6dkP2@o0~K!3vjzi7&C$Rek9CE*eQPjwtpOeDfW2dC+qPnwzs=3xRynM&SGf*p#GuUgj_Vu4q!9$7)c zoGWT-9Zf%H)N~nciT;=xvKcU0ai9x_w1dDVSj+?J^X&Z?WQBM9DuGf!8!UVeqfdh{ z{6yZ`@ihqA0cqGeuDfIP%_uvp1$f&&%56tveT6BaZTpU&|x6vP`1mjoj)T3a~W^*oOdYQ_G_@Vu|;_VU*>* z<0rxcB3N)kBa-rqUu--9wh6q(5ZTlNsSuKcUBkVJ7l82$YWvaq{29=!brlo)D}`(+@^USCMX+5 zKEi-ZLgw+~%|byoNIvRRYKjAD3Zv@038N=9#VsE#*8NsPH=Yt+T#;S9cX*Vs>O7pY za5Dc*;erEbs&pPv{=2m_j{~HcNM&_d%8KWZW3v?#$D^cSdiEAvewW!;YO6zb$=PWv zM!s`WU)|ctMk*_*tZq+PEe)!_@8B6mCEX9JuWvWD?hwn(^dIqov5-fwQCsM z3GPnq8)W#I7cAHumCV6gPy%&z8?ImTRn-AU!t+Bx^m-J0Njo?bcm7O5`9(pw%i0|x z5N897Fr`DH_F+-m+;6rN+nO^?TgE>+)vcK$01TOJLsT!m#kH=U1(-mU_&u^^6!87^N)Ws z-HOB~g>I==mKJ31)!@=T*9MnI04>pfk`0@lCvnX%Rm_eY2gKBICgXg8P1L=F&rnXE z{QSE&@3Vf!>6-}C7rS#3q!8-AOM`fu}n;nSm69O*eo_` zlE28{fsF#Ui)|^>0fsCKnC<;FHl&1?-m9NaPOFRZ6Q@m|e-RlMekX02F zz_GQ{4CN6XT=Su8pG8s5`cZc-3UaEq{!Kfr;Ys7Toqu6&vHEu7aC4THIV7Q>^yWImQim zfZ5Tk)M7-aZ|nr{8(o(97mPkmI9~Q*J~B`vsP-a}>OT5i5$%nU`M`FTzRWpzi!)dz4ARt9v!nUN#`(w5RgB)u@v{W!B%3-|6 zk|Z}Kv&C?V|I2yA=o%S^uJ~lrlF*yXfj_poGK_uIRbdCu)HlM~AXk{Ir02P{BRWR> zfdgQKJuP4(*p~C=o$z144Vld|5d)63@)lgh_Q|SF{`})#g(k{>{_*c(ALX;?H;-0) zrpG3u`QdDNT|(_4p!ObXiCOnoP~ogQoKfsn46A02vcq{^SEQ zQzx%2Is>f&*;8TT33Dq+lNBGs8=!2k5p*A1Fk*f3fx?GzBZi+xz=OO!&~6p$8)EhI zIPi`HQHF%5LqJsIVk=9&Ww>4-KdVgl;EWVAUC6v}-ToSE@#$-+AY8|A%M$Q{SFZ?O zMIxOucx|-8P6%Gz8tf$rMwbpoXex)#C>!ufgym>4PiDB*^0BvUvj?(vRStjB4?lB?BtPDx!*|(2;N@@;<7T8cKHWL${H61?M5N)k_t|&@k?kmwR zcWtD`hOBOEJ;pkvPAF*xVbGa#A1IAlWNXIvB{v>_6!1(5c!vNybRYo#qeqp^Ie@kM zNrd@1mZB8G&HzF&Eq{pv!{X5stl`CX^T)#vE*=P1(r}&zgiq}}eoo+d)&ssIE?;f% zR&-}g`1CXlPIeFpm2JgTNHae$wyqJKfh)}&@tG0kmJs@q0eI)B5$6xk5_cu; zULNk2O_|Vl0bcz|+yq;>c4H3vlQB5!&%Oq{XKcK_A`x;k#l#)vw4D|Dt)`-q%20}A zJ0>C?-{}3YNT%fVBEnJYgh54{UIGn6aAS9D_EbP+OMU@ApST z-85{=HI1R7=z~FjJaQE}7z}jF=nV&^W+?i|9@xF13E$EF*c_Wye`NQyJus+{Z2p40 z3S9poPRAs@Vw}Nqn8fI^W`Q2?IuqL38GP~^#Lr{)@D4tF1x@A)c!d#LMrO|8E|GMX z_&g9^8-oqm^5R4u1xoq$I)c(`ZR_?cOEk8YB_0lQmKy^tCCA;xVftPWrk@7EDgu$9pmPVS z$BLX4w1Zy-&IxYiWZ!M(_+%J_ece&oLKi1m@f*$*|UoPEr~=YK_W4U7MJlI zMv?>O47m5;FwJTt(17TgOHCjOY;~oEmbsUmm6PdczBCXk5XT_p|8#2#L;MPM)-UL3 zngD+a)4KWfi>7z2LC~tv>Z)!wHFy!>*c&I)xWV8@>00}iZEY}oQi|jme{(Fx4;QgC ze@+KFye7K1{|GEzk{%0K-OM%AUzNZ}|dFZ=pM>FNTF z_HI5Pn#iyQ1cj1jY)Wu}JqI}X^1cH)4rJWicc6QcrrZ($cLu;$jPY~#m0)K?!GPZ9 zft$42e?yi3rfPmsqfg6<`Y*d6U$^%y$S;svj&V5I*rC%*ifiDrQxnr9O%`4y17!Ht z68>T@AJ{9_fqpsm5%3)IPDIUkiv|_9mA;-t51j3I4*>rg_+t- z!?9WaayJNY2niiwrpY&pD44o4CUNhm>(d9>Y5VX)N_Uu$u)L8n;c(}K!wo5gOId_6 zb@1u9tV>+pj>{iK?$r|KE#(9T;XM(xGmLQnpY^Vimku^XveOmk0`6ud7pWtwW$Le% zc_Mbox)%we5=3PoO94Ey_ktZ>3Neo&SYDLM=31JIaFfP$kKAH-t1BH{tuXc_7)Go) zEvCPQ5$_kchMc0bMGO`@U3+g`EyuGg{GYu--LUo-`as4W^Km>j5RAhKGopt6-eZlC{_Ry-9p?rv*H47$+w(I)3ow1Q|pHEYP5<#Qy`BVB~oRSNXSQt%vrYpIY*?XBj$kzc$8tP z^oBj=4SQ_(Mw+MN#Vp2;a>+&^(2<KZdU2h_}!1>W5m|=&`t%GcffD(LQPE}>1_%c z4Ah5sGmK2&xFiqn7KDUzQ&-~V<>2PlT*OI;4wR5o^)r_jGY%6_L8KWkKrq4H;|Wm)`3Yxg!!@nC zV-44l5s8qDG{;`+&eDs&#h~=@=GRYFP0@OyCd9SNuRM>!<9yTav_I6AzFndzt zU!ulO!nsuU=c?|DKFvF=pwG)Sw$!{*^FCVP%=Bp@?6wjnlA86vJsgZ3HLGEqTgG$b z&M+Yk#Q;=|AL48W*jzRhX&!}s7)(-D|LeO~@HwTwqjAS`p7|Vd`0|0=hYMgF2HW14 zU3(7;B#QuU^*aq7pQTyw%{&s#3FnMS9DnLMcz;|HzLD+l+m>hvfDu4?9917+jEs!o zwSCaxqzS~OV^F;KBT!%!VmA1WG}_|OY`CfCgqH)dUW}}5<;J4G?@~%IF=4g@+k&Fj z5Zbd(nu1jAbPHN933l5agyARhwoiW4v~)1M+BR*it(w~Tz`7w}{gI$M7LH{sL^iT+ zDy$_a|9(YsLrfui#|sn4D<8NJ+dP#?^QA#@N{uC4e#GfnRBqN`8L*y0oM!uPG06*28yNx#qFTTm?h&XFBWm4m7+YYYY~xv zjhtLI=oa(-vjsRWwt?2g>=sAs};l=hVVP|@76GVCj6-?S9{@M>=WCdC*pc>m^YThLojxlIZi+)>#U z@2V8t`TY4}?uYIuevZ$5t}4&h?z$n&4J%=1g9(@Btjd#zwX?JNij%!qHZ5K9w;AkS z9#)a^4dY>|C@$ud3IMq@!+F?W|YhEh%MZ;?`FQEY;JY)Kg(z zyJ{4seTZ>;hJlWQ31A?f9ZG}W!Uvv8gI6TsTn3Df5is60UMx?CGSNyN4du~L>RzdP zrS84;WGxY(($$WxRV%VD*4XnR5Ky*hcC{GP3wQ=x*%0hVVduqpGwtNk*rmcg8awWQ z@kUA;0paWYDske3cvSim!Zv|TYRLr#nGuM2KY&NXsHz*WXp7Wg6a$|Aou(d^R?Xep z{tKzd01DKNpnZFpJh&PZEhPAEC8c<)vJ+T21#3Kp=>U~gyR`RB-5OZKQO~e@BkK0m zaj)m9iVm2s7{j4GFovTM)vUg2YUJh7kyuv?8}4)2)%SbNoxRLy8>_&~2t1?#kUY!@m*Gn+;D;Yr z;4_50IdwmAp7Ilx1Q4IsEZW^*@DvkeH-MliL9iweghMt@3TJIH-4T{1e)vXk-0~U%ypt&+HfJ}-U~>7%@X@ng}gTH&TA8$ zWN2XI?PW#lP_RTf0200e_LciNA0v~_bX(mIFD6FUGTWrVSdESv_^r89e^ISIU^d#1 zw^x^dyfYxfI;U`m8NeP-rUs;^)2NxWPL{0FiuuO&9=&)nUC+!h0$T(&K0 z?t#&(v^$>X9$pQY%MI9_o_%V_eI3P zDp6BwAwp=~UZ1y=W%po*54~#{;wTX&kWtBG9D>AE715vxk$yP@|oX(+{M6sqVM6 zZW~Ax!B#kMvBOtZzeB^Q5a5%DxHBSV=Wi_j6HYW1;B&!Z7D=3~aNYjYmX+0_-d+=srz758{MdhScLx+-t9XOm{<>zAG~ zD=>5GR6&&`^pdGQb)nZsRIYrIEz25qS%x>u;pydV=ox==w^=!&;h+T82#9;e(ynBG z{_(G&4^m1fy#Akm{F~*3gSK|zM9D<}SxdOH5jkJuhj3w>cN}?Eb7T9&Nhb~A$hRSl z5^wB@ES7^GFnhZv3H7!%vX{d-7R%(*jMBn~Tu7{SRBqsP(QbFlZ4GotX4QF>-w?fQvD9K7MD z(lwvd8m_b?m;WyBQYXRgtle@`?0#FEpUH~)MM===@V9INV>-uZD*`@HbiITL#v$Iy$!gddt6}fdEU?ZS)F3XsT%M|m8J`g^ zcWs<=ZK_m5d)eHNbuGeyQnG8&X=_u>U29X#OYK@D+Y!~*j%XR&4TkO4&;UpR{q8wR zg&qXvdJh6!kv#~`p*!8K&FSQ6MmF}5N10M>x39XBYD}szsm7k9#&l&zJ*F$?Fi{7Q zKMxPEXg%+!z)8j&uN)4@f59i@E$c1y4#4UPhU#RhZ41Ufz*{@4qn3klG>86k;o;1R zkWZ(meP+&or5ZmY68A(@o^d^tQ8wrH%l!e_G(G7SDkT|&Lb+r+vHnVE~wHWLuk>HC|o2yU~iigslQW3 zqgw8AO1CTTm{EUuTu;8`$pHB)>mC_*cOF@`j%D*W7#*qx zMPk@NFpTdGiD4tc2y}YFTW5d{XN~s20_|_8P|3LZI)JLAPNiS9Or>w1o5eVSBLQb& zQMS!ot8jiu8^G|Hin*U@F^^a^ciI)Pwx;)McH`1Mo}+R2k~AVZS`4un7=BVV-|-2^ zOPY=EgC(N@=_g+s%)=ag z??H)z+)rQ^lj-e3>MftQB|6rZTJ8;}WWHeYMHcgAbX@M}`1Om{HLo5mHFFQJEE48e zv8n2I>T)k+QOy_Eoj-SJE=nByFrw%NX6u4$(}W1^Vob)qJis630AC3X)x7WcfF}2p z24p#FX+u}T%Wur3?d5UQPGeH;=vXv6<*0X&78^PxA0LL0jcqs(GRc0bbs-b_?UmxR zg95`|dvIg7=!S9;VAJpD%2|UAX|je~%dNVW49Yh1aEr8wSMT4xB;#OL6G}M|N~wx8 zpbi03xDXfFQn6N#D1$i3*`WY*I}|xB+wkNiJ(v2I~s4$ zhnRw5bkz+xDA#3pmG(+iEB8uOFS}Rj32}(E;d4i$7%MUp$3O61WaFG-JTnGhwc*wgMa~>JP9wKMrR!3^@Pd!b|vf zrJ`(!?b$$XUd#9j8J3kVim9zkNbmv-R+`G3H1%%3P|cQvtkSuwfyMIxYI-W}23Hp^ ze0Je;f{Yywmbg0?)d>SP8m?60ecLpw4L8fb$I4*%J?k_tYK5C0-o5$0RcYwi%1E-j zN?0W3Rm9GpfBFSY$Yc=(qV_!iA8lIZi!vX{jF^E}KhN1|kS%E`a7Y;)!)|H{`*r3HY;ta+- z*VPa2DKEC0d6Ouk=aJ)&*(~r{N~U5!$YkvL{+6Ptw*=VO_i`6qqY#fZ$T#6T>+&1Cf@Ik*K%MAaW5At3U%x0_eGs&& zc=07c?{;?p2;_#<35 z-ZBksQx8+dRc?82(iEyiXVc4$DCpa4ysd{{{RD>Ia&qq9R3`Y@)}c>=ymsf9sQWjA zWMmUk?wJqlN2tuyb?HgOv}?NVE9q=_+s<=E5Aa5rA*ra-R$t!ux|f zjXVZNF=>tiFhO^UAMl}&>(*x4lUk@6e38T0o6@6KKaT#kn9VW9kmuvNL^K8^P0cPu zQd#`fOEMWE94jut_GEUW;{^_WTru#VRqfNH*~d3Ovo~ACxHjQ#P3$4qCIShMhjyg- z^LRpMEI_uy(%mf2z69&3Im#sZ+hH+Nt!&@8jN0* z{IKvJnDn;ln%%Pr)$*expC}|it~fdh-Yj0s<|$7^;eGb&6vgtleG~~}myNG#Vt{|N z;Ldt4$j*$dZ4-W0A}P-AGzuNeFL#pA`3-31q-TR1>B3OZUD_qffo!5RViM1B_FO)N%?2#By0*QPRqHR}HL#Z8d~M!E zUj54_Z(?l+3YfrCzyMfmmJNhRD$=hv4CkzHhs%atW9OXgb1e!+5#Nw3B-4G0={^NS zEM)%ZJ!RG=aVUn`^1LZ882XD5E!w=o0%_x9Xf{Hc=Yph$%I;B}L!hb5rmU?8ObF|m zVzE5@cbre8;@r*nWF=iYl|z{lv>_$P*dy(`K;~1;EY8(oI!X3WG%-QR zFf`rdE6hVT%&I+T%!}xiwGF50u?*K>Z3z!D++99xXCFJ1%|my9N0vroJgrsfyC2EN zh^_LWoeCin$lToIe9c!`&<@QGuzK61SJen0^H!USK~idGuaCdLy~8Hc6R@!X#ZWXXoLDlsH&l zq4RhV;mMg7o~;~`Q;E51UckB9{ghsPCZ}*I0XYREFn{8?OhG?+`y3Ep#YSJAY z;w-(o*Ur)jw4T%r0@!k7a{+aXakxZCsCx|Rq^(vL)WL#+37$mVB%dfc z0x{*mF-^6v5Xdyuu3nS^6p%o+ACR#i={exjfs5rNFdd*zAhsFq89l;lBB;m!W1m_| z>l!D*!rSi9f*GzpS76dQntlRc5;GLV0#bx(=fH&79jd92*N9l2Pd>Vzfz<~n;l~6i zU4oPZsfz_EYnMLpaP^6b=TgYHi&H>o6IOZ6%@^!KQ33-sWo&#KOnf)k5y`5b`0(n* zU=oFk`BIWJOG7EX)xgDeq_5DHe}mtp>$<@sqnw{z>#Po;to#QOC;0EViykmb<6U5I zVu$dSWe%}yRAa$FxH53Gsuyxt+UanS6>4!@Owp^3sOQN*FQ#G)@{r>1gq^1-r?BmM z=Ye*HM7x^ME@dv^@!@C{TrPLf&r{IaX-5UE4p3FPJ2=z?FsYC z(A1Uww)!$`;V{Y0pRD{27>n!8H8#f#+^w+2l6AiWcKJmxI{7OwI=K%QK1Q+#=Xk8i z3Kx?p5ll~(nEqldx$4nR`>H|=8~R*5J}YfK$dCmWxI0mF_|$GQv;deX#!1>N9-OdS zvR*K_Pc4amb+xQ#%*OfZ1q(A+J*!Te?b%Q8-mK9r=j(NLZWsM`I|13Vlt*5WkENJ> zpE?pP1QF$$6zA9VT6ID#-)n3EfC=x9Jim%Fk~b-txL&ZS zufS8frOfEaTg(pWmy}q15G*$BrJAl zs?*av!7wHX(}#d*1ZWxoUo28fc?G3s#bySU#Ab=j5}OZ>&C76>nbNwxTla+GK1rND zGEQ%g_ap9N^2stmHe^993s$`fAWdK50}YwMUPIP^)BzVeyI>8>+q`pwVWpUi?8Ui8 z;(Wc`wyOu!3`+v~75GQ|FqZBS3BP$;FkmS|&&}{=zuPZb+OX9bB)b_5BNu-CVyP02 z>+eciwVh94*Ca9@+te;Mn=^4afYr=;=fYZ~<3ysiMC~(lmh}6(`AP~~Ch_}nT_uOa zeBmj{a{|3!7|ynK0~VCV1S6M>UEe3oh9BP3r~zhu13zGYjDVJn)=QAV2(WEL0y<$6 zmI!9pN2IZp8V#viq;8SAzk08t}_d8*`&Cv2JRf)h+_cPyHy!_kf(-o7C7FpS{7&k67xmT^v9e_XtOrY{Czl7t`;@cmKJ z6|XcaM%L`r1@g#=q-iS4Lx9TVSa5mDyb|5(O@XPJ0i&GU-6LWC5tvWHxx{#h@n?eZ z%k2*hD>vzdaO9Qk^&ddrykSj?Hf58>{fKY}-t2)h%)_NA-j z5D5F*B2I|jTS{VljB6#z){3%v)k??0Hc6D-hO#*D8h_7vhDyAZcza&1XA~9ED?%8?~3SP((V|SH%CD@*$tN7IO>Px0F z7XUYHV8jL#=Zyku9$=4sl$_Bbd2r0RGh8GN{hUzNP7~)lM|01?Caunh!W&1NRJ9Lf z-erV6?h(=1$a>Q@`J84PU*A(^ZPrOgH{sJHgr5R3wN5s?xi4M02Whb>HkOEC@q~FE zc&-lYe%Ht*7O$f&h^qU*x@J;>oEyb}pUiTm#GDggew3WI*jVfh2n1uTSFmgTpMU&2 zCLi8^PZo24R($FLWX~r*Ec^#1n)Tx9bcXoHUe5^!ku#)G;*C8nd`Fm}V{Dlf9igB* zsqn0JP%4oBcbR|eu{ z2LwuhI0N)%UeH%5op0_AKP_j!iT-bQZo*JH|AL#Y;W&r&`c%f6om><8V-qoC&6xOM z_=(KCc877_CtKHe*ARWFj}vVw?-+Rtka-w&4GtKQoX$3j+r@qF^ zlzaz=y(qv=jSQ~8elc=1;yeye?|aO$781oe(v?W)hJMWDH+iwKV`Q69A9w?##RD|@ z$E%M_#WVpNCLYk3#NKBx^@C&@vxhUUO^1Czm=PAUJVX2{$26wHF1x+6u}!}W7qf-W z5^&2ZFHfLfYM!@t;(WVe>b8&YUp|qo^Jd$TVNDOsIlnuqqB!ghj*DKGmBRvT%EW{1 z17TpknA$xn8yl%g@IGho?%6ATY)UxSGOkWPDv^&8nEtq}f$x<#KYCSLMfL2npB8d% za9PY-lUOOS@*LxSw!Vg!z_w&Ni{2(w+Eq))ZT%vGa#ei9q$&Dy_u%|YXY()usUvK~ z9*tiT6wVi5|47dzyBpC5e77A+}eiVk9uF0TrGtt;+*neWC4>84B$Np;Ah|Di4qzE zticZ8#NmXP+8UbQ1D9>`Hr$qL0?ojg8#qZaJ7RFL>X_uh)=8ytGBaIj45=}$d!9YZ z=%XZLaZjU=WIYMln}pdJU_$|m)DSfnv{X&CHfm3`c1JTWKs04Oa%b#v`8=!_hLL0* z&Y5RE?XD5dF)WRFMx%QU!12c8XX{ihk_IqU)bWDO7^ubP(n%_H2PfF1(%7}6nIi2^ zRi(XNp#e?lPekFGh+ zPQoU@zaQYMAw~3LMh3DTOWKEn6&S&P4KW-4o%ryT?}zs;kh`_Peul3*f-eCa?vUf~ z!T8&;I|Hk>TB%BfW~k-U175>P4bEwqi{r2FUXd{z;eN+@u_hwjmW7 zXtV~9MIxSk-ATbpE%=CIurFumR7za`B}YQIB?`CzJTDQsxc!c(u8U6E*e~bIqUHsupvP@>D$YjR@JbMe?m6m=KNqLTdm)V01thONF9+eV!GAFPG zKW3kgd7Q{)7q@Bw#%gsI%T4dg>MT|{-bXeu)K}crXGcJezOtKBt1QiBREx! zU?o1f3doq!xY25pI3yq50@yb2bsKQ(;#3IdjBFr+K+rDv2QmxRxlVXP=Di9rBzm4M{qc)*Jhy{(PR{ zb@2YUBzz;=;kPXfcWpTSG=aDvf(a|UFHH;BeeP?+O?R6+HdR!s%t~pp2KHotT;D2U z`*$fNIORpMLZ37RRa!5xU-_p27TCS*lOHvWS4YNhbJNz^s;Ql-QhFJRW!^wtQ2f4< z^kI2r5c6Yy#&$V-rvhDq+12j<>LbiPG&Wmc;1m!sz~L3pYA@avbYAak(U5)vwxd zt*@SY9h|OqbZtZ8E|Lv)oP*0xY-e1byc!DoEpb0%)EsIcg+qEtkSe7aR)P6@(2-6SQZ}vl;b3~ki>r?{ISp!c@nTH*F z&hfEoJnP-;vRZbOH0;~L1TstcQ8m#&h#h1EFJBQpVJ!)N<04W4NvhTlI<1h{ie5LGg<9^QCe6^QIg&6SRZI`fUF1!ldA+28X-2wGUOrtxtE6G%J{%W$n zJ8IWvQLQ4WPnWTzw+jq{d3}8OZPNG&0B?yPaid-ALnXCJ4O*pxy$S6>*>QQD{zU=a z%BuUe(=B=C(w4948H`cEqT#Oa6E{vuw9T`OyIznV(}sC&5MaV8P@aKVie>Py!GgKd zBAWGvcgYkpdd7unt%%NC4?xLH$Z!#{v`8Q0zwxuWM|-OFmX2WMIxXALb5>)ej!>z_ zZdQDKbyQr5`R(+f*MH3jL~T z+z-ZR*8g6b(iC^SL*O$tW+m61tS+SbD3CMZ!M>mPxkg>+mX@T$E~n#q6kX6s{&D3% zXzZ)gVLDg)n(!iNwWqEo|H>``U7A;IP+07uq9K-tg6<)FA!AELcHAN~fpu+f3LhJU z&o0kpQOMxAv60^H6MPKh(w{^?>L#JJxO|~|i8suyMyf;>j_WvKh`4#Hje$jB|7s_W zk_uhsMLTDqvEAm4g;l-Hm{7XC-|s2)hSOXbQqFM&o0PXzYwUDU$M^q>**w)2r2h3nvt6Ee}5w3$(fRA@#qTs@z|P zm0O)6`iemYsxxvIYcIH}FtT48b(kLICY!F}{_Zgyn%h#UyP4LiJ0A- ztHRvNlcd%J-XqG=7sH(|*7z@1xjZ}74l4(z`}Lon`wlUh5RI|`+ho7d2#twFS!H7= z*&x3?GOJE8Kzs&JcS8u#^c4BU{xK7gg{^{MU^lUSKJ+${c<7!%yv!)UvcE+^#^G3^ ze2;a+M{kusH(R}}5+{ps=DX{+P-i@(gQZhZixmWurhpjeoc%1%$o(NFlKSJYH@3; Wc!4V|;(MN0$^EreTlDezh z{rQ0KG+;|^C2N64)B`(v)J1mMdIu`x?e7$;#2@UOGMwa^)!ant%x`)4=_&SSK~+j9 zc)x$>UUsP8G$+Nr^N|-E1jI@c;tGAB@pS9;m8dZi;Ehaj+$8yF$QC-gY;v9y?De$ay9 zk)r9mXsbmuysjOH*C8B*tdbM%XlPS1s$Je=SvTI8G3tl>4G>9~+7kA#TKyHvJ5?|g zyYz6^yPY(>88H22V0g?bWI!op!D(#x;H!xK0k7oL^eS;S29NjK3C@NH-jB94f0$*u z2R0Z-L=svmVgGl&?(pV=N`?67vF&mKbdHn$iLC0m(N9}&ff>vsS48HRv2k$aMLGy! zwOM%aV^ITnR3<9@yAHhtx*H|x$(4aaKRrx^NV#RssJZ04knbzjzIcbg1bj&e-6`3= zwwETf`}iCaXQ|N~gf|y|h+bdARHDv$ovwAmY!EQJWYrtaO8sCluy7K%nY#X->JO-i zG@x@~ZcMijeCs#(P?5plqUlDZ+XTJDIX9S&h^s6N*ONzADDbtnne~z_afC>5$YemB z-EyquwNcxH8RCj4D*%Sw&=3I~);^vqESe$+ehJYWZ6K?UF42Gtw${a-M*?;(+J)Mk zI?u+8cw~o68jkiMf-i%YBnRGDx2XL1gIewVrBQhLyOtDnqp6-jN)AC?kca4x!&Wp6 zmzP^p4;lTYqDilh`u-i?RtN_e)%ca-AOd9CSdtsD)7#nbl<_ZrQTe&F2C!3YwQ=MT zbgmF+V>9vI3F);3yLA|RR`TyjG-k#KM3Qn1!$m`-Fv&1P5g<4?zs$PmE9SzfT1B_u zN15A|xNLSiMWPw??xpt&_9&|<@}Q*{JF5TVVDpvuS9EsXGh7;!bwt__ZdzNA@o$c1 zbj^|G{DtXF1jRRSE}DoEo&Fu+{0?3eN5-pM&Dbuy<0lUA@97U&ZO^O&+$vZw4V&On zm0gjeS#R^UN+hVJ=Hm1QY{av}yC~D<;Tfao?Etgw-?(Z4E1ncSI8UCW@)+_%2$!$B ziO3QL{R-l{mIgaXsHgH1q^G)gTUV~W^zu}{?>y=56P@tx6YsO0za!9c+MyIN#$Qd4 zGp@&-+CVI^xFX!czP*TC2>2+q0um3xeex@aDhko4RSse@g()_-yB+&yN67RjKgH1w z4_?vKD4EthH3m*ZVZ?SgqB3s#;`~BNY2X#=!of!b>}x{RrF!jm}4$0t8fIZ6PLq6aE02PEr#6I8lc(vSO?e%!`}?l*+*cRC*qWIDY^!ZK*L z1p9~SWU`fX(76a<@JP6&m`rcAD6m6Nj)eRvw1TI5)^1;)dG740kweYadR(%*k3Gb# z8@bvJXA3caGl$^`?x@UK8CtRQBi8&5@%*V&l#x>ezW?RUi_NU0cCp#sLP|TLk8h_} z;_Wh}H})Y7Yg)o-umzE&AJ@T_yyx-FzvEInT-eY2gP!P~J5Q!}wBm!DC>41bbQFO~ zzhX^ie1-uW$}HCP0Hi-oZT)a;w}C?N^XfCbr8fUS3L&Hq4cQtq(n57j_Gllt8sKn7 zVHM*nY|=2wL14@x#V>c1(Hn(bE~B@zx{`gB$+u^2o~qv{Y2%aD5z!>=%D!Fm#Si~O zN)x!Z(x2MY3IH};TD&w9giGl62o+XUUvcf}MA}ZJ5`Nt2eY`C9mHJv$8iO8+B@K%6 zTMPLU)Yh*AO^#y~q&_-2oz64CT^XOvzPeHXtpOd8RSI6YjR(SrqV8vr*tQAUI56s` z4_BKpe$$h&*pf;3?XD;)Hc%2>%pwa80|`iYiw95U$dOAr^f0BkdYz27hpn=QZ1m!B zbyz-ALWi&FMPB_9ts8}tC$t=KP$)3Onk2E=qt?u12d89~nBHa{*URl;o5IF7e)k9d zWCS}OS5!^J)`L)C4ZFzuF(ce;N?8Sbs@pds?~v6vCQ>{0Lx+gWi_*E)Idqj^t4}ox z@~w_A_Wb?irCUN+9aM_2bDC~v|Bpi2On;6LL^e$Cv!fgLJ%PP zSd407ipCVY@%w`aW+l&Tq#}ihQe9QI!#*{ESDhnHYhtM!uOu~p!H|VkbL-Snwz>t6 zG|~y|Br5hVK`SiHB;6&X^0FNJgtPysYehxW#&EY^;DA zu;azXE_5b5l~GIjk=rdA96hPwk}0vr z$B5E7Ue0+|=4386V#BQ;_U>G9-ACnj`E8B_%jyhc2u;t{X0axAHwl zsWC)ydP$nofDBU?D}Savn#alpXVxhMUS#v|gN8~7OfAok+5xFN9yEmq|LpWg7@WjI;hS5bx%;iLYa)ycBCuQsr$`J+ydlzbDIGV( z=nd#xw>JHXA~3MjxwP4d+KsD2&ykHft+U)>s6XI@I24eeyJ`Ydd?opqoAr_;CfyeR zQ-!HXp_6M?OWt)q*iAb02W9-=v>CFvw8KDw-F|$UV@T){m7E*vP?9m#tUmFsUUxnp z+<-JhNYgCx((+ZgRZ+{Jj*h+Ha9g*0qx;r z6CF1%{0h@$lO;#7Q>qPWpGI+1S)zyzxfjMiebT%k>pNw2lv}o{KGgA=p~cK%S)8yW z^ljp7klL^C+cVdAY}%PFshLa-7H%HQt)$~a6uoIgE>qh?eXLNHu8iheD=wTW5lu`D z<`azI)0#&?e-LBA=U*4A*9ztKv0#uE{6Cb_ z*v);iU<*xFNSKBEjXPvnURLTz2*Lx{Ko*cR85coNig*n9BNobZ1>MjtSSuWg_@s4u zoiT3Ct48}8b0xKk4R07P1@{l1v)4si2g5JkarqR$oG0A(Krd)5Y}4-yVa^&>cT$MV za5hhWdT&>w=Mao(QUqZRqWhiiZkB#Smf;t(?W;{LE$$=z7|4=J%{Ypun-Y3`lAK2@ z^w|!~V8?NxslnL>lu}YNbjsqQ7%_y~JQ)`H+?+M|OrcFgTFPZskp!_|OnzFFQ4{<5 zNb?hdAyl5h7ntF(%*j#LsW~`iIdz`)L5p;)pB)N))aNU>sus7X`6 zDR3dl!A>z)YIdMx0bH-evIlP(8E|xc&+!NqhIruQE(XY}*|l_cyW`6@ zhgOE*%I}!6-;zYyPfwN@9;LJH`wizVr(vzHGcYuJ9#m&H6KzMryjG6*afM(~;@|Ta z4j1%6sVqaT6ERH*<{nxSaUWGtH9H zo<7HzQ3yw=A=Zmhz-R8raO0@a)k!H;p^#yq_RL=#d-f;XaSoCi7qXA+{__W8a=2Q; z&QAfos=H0QX}*0nWrSDB)~vD)4=2aaEw-Fz$-irCQ=@rcDuFVZmYj9Hk>gM_M zvR=SojS9xne1zc7?E36W$Uwd0)?D5MC}T4;zU$g8j<;w29|h?}QICUk*8 zfA}7!fPpGurza?y+dvdV-waLm+PFRY+VbJqZd*z5M)}VO+r-;OA9X&Ieu^g^WOqsN zLA76`Orwivr}2H|F>CYYF?vYfrPWs^vt<|~%{tYJ&09*5a#wN^);C*Nil^}Uq`osW zX3oPsX#b$tF*8y&>7UqMQgn!{ZD2grXlQ~;4dfX&Z!A)PJNRt{N$z7Z5HjyUB&Ir7 z*Ao``4RnQbO9y993NQB=WbQ$D_93aN3VmC*9N}04rbHd)xePKN?g0p?XA*BYfO288 zKspWlcO6v|ypGE5R`^lrQ+ZUH^gGb(WSp}orA?PBV=WMy|Bs&?^)*no+>7 z#jI0Qys2V*@1N_@F7?M~-jojxWgyp(lil9<7#~Jz{I|I(n62Lf5+ZE_jTF9AZ@uf_ zS=!`ZtU1IxQtIum0#R{~=uxk>JZm`kMF&C>5!dUwDsF)_b|NHfSBfx}@Mlr$q+~{7 z-JyQ)kLcGdh}TUZ4=cj!ppp{|l@2Rm40kgc$xkO}zGb#(cihJ4&da5k*KKj+L_ien z!E}VMeVXR%UaBv*=gL9>7Q}KB9Knch#9p*ymuGf7-jFA?q~90U)XM`cl-Zv6M+68y zyzNc3Oh=eRMGzTpT;>#q_7cO@#;@!MSD$OVb4Ck6u*^@N+n=k=Nz`L>-3C{i9S(S) zA|y?w!mzR;URdi(3;+1wfkbyMjF}{5fLe70WDSSjY8SX;gmP0<=0_}zz(Ko7N5I(o zdq|R#8Y0|#4!J%vseYSwwFnC}&fZCnl!v84qfs%T!FJd9Np1{RPGYyK=KRchpVOQg zGrqC|d>OhjRi3PfAZ;GF;y^^Bx*>-{o3C1T0?<-pHW0k%nA57Q@RHHnQG)o~ZV0 zPxZc9DvR?g<)o)DcDGHU)6VqQglCcAO{>9O(}9B2m5EsZc!82_K;+;(VhCE4QtDoU zLx*x~F&7$n#Qfp?=1MYM;T#`=9R z)8k(eEg&`&p^R0ec(*dWfg5{JQT~=ur(U&xe50+^PBk|xiL>^KdWP5H6wcSW^ecv@ zy3+(sVF?IJMRDB@03u`|G3C1gFzHv>0w%wN(ut`FVoSM~q2t;o)Oes}J};5W$xFnG z)g8*|K0IsK8hYllf?l*Vj;8`JpjcfWC52 zc^%R+nhWO${b!)*kyBBf=D}*Bm&YcN4@U=MwhGFng-5agbXWf1mCT=dMbn)0jcn9v znVKl>>FZU%vJZmb?+GIB-mY(b$cBw>X-rR0c;bCJnp=VS^4-wkC6FF5xNEbXP%5ud zqBmC@=Q??(Fa>1-pN;kt_>tM7D%J(~y?zbV?tHT?mmF8zwbf%wwiED#BFW&->EvZH z&nx2dvZW@B2=&2ObZa4CW3!-txcaoCI~zD5C$JIPr44UnScp}AW+`F2Y%W2JabOSc zihl43$(6UU+`t{NakT})K_qm;wD;IyIFJpE&fEF2_`%Q>k~d}d7tQ+F2V8G&CukR8 zbTZVWfuGLDj}&=w#r2|f0-YO*8^TL*gH~|<*GPJL7oUEWMc;O&4FX29!p>O{_Pm;4 zQT)H@F3Sd?+P!Pv>G>&meWMzB)Wfp9uiB{UW@ldvNjSGYfH9_)By<~QPAx9oq3qwP zfmus7u9af-1;26*VQvOxZu$Q?_RYGY$o zb;S(}CZ|sx1sWce_dk9xsyx<~F~-XmT=%JXXV0N0pNLufd|y-X_8+MEl$_w$(fe8I z_n+YXGujECrf-hyJ-sEd@vg!%{BfHYv5-Y=Id!di>mJ%&90lSic+?(;zw>)~M=;k4 z=L4k@;U0#>o2xa#MPq0KwS$FS{L!NpbGOv_r2?NY5FGHn3Ms#NwDYhk1ZuD`!pYB0 zd%--HvK-KUl0*8jUsDw~P$S&)b5#D-!8N`4fcS3O|2lhP)<@4TYn-VkO_mauAcM*B z7jY9Mz9qmGLtMuW?+zCf`OM=*V#PLMjL}<)xO=|GcUHK>a6cyM7_2hZiKVpRe2+VG zR8rBJVbWe`xK)3ORxD^60}CZiP~N=SlALea|Jyl&4S>0!{Cqbv6QuNH+}?@YQ~O>J zkQwC@KsA<|N{1^U!-~zt1upN8IJH6n+^y|iI}M1rQdXy|A)h>dM~6(?_7qlo!o8no zeohN{hnc+_wcdHkz6!?vb0*m{;FNA7y=m5m06uoI&(d3jnNR?g`8^K?Zsq9m_f_cB zj}h+g?+~o>Tv%a#Pwuj1D5(-W2T%D5Ppcys-29}DnH)DyulIIc7%G!RL>L>}L4#4e zif2EO#m#z)Ek3@6`HQW1n4xDPW}Q8kH$ZH&%JQ8KW)l4r?xd=n&keQT(SVtU{{daU zJZRE($<^-_Fj@mC9y;vva1Yv;dKe3@%k%S9`$I^lMNj}W6&!Yp9jz@@ zx=gSZ-v+DkHL@CcxXWdW@FZmiznMKRd_Q&}fzLyFZ3g$gs&0K=y1g2#GsbIo)>W~L z710Z&__bLx;{?loCqX*vcuxz+e5b*!P^I&qK7Ey}JL$!CL+#ug zPY@b!_o;zZab|!^52>_M2_KH)K+5=O7AyM!P+fR5eXNOO=T==tEz2cg8$EA>bgG8% zjoGWkK%Mw|ow(471j%ae21Ibq0!8Tp#g?4lXI8;$4AcxAC24iyRgU7}c>Fz6PFPRM z5=Z57z?8hr3vKf7PV#U)5>glvWja!pM5RDGZMjE{)S2}M=n6N7jWiDxrHeK8Vh`SC zyQopVW3&zRj>t{5C$8`y%!V(R%bKv9T9s^?hq~@Uz|jDiFfu_s)fn|ldm!g$!s(?1 zw-Jz)96eGwYmu4UadbP8o~l(=gxSf8Y%$18&NjPphR;@dzR2igI4R1*rzeFfd3^R& zm-d8F$TZDhUVZyL-B#(0;K1cYmu^AdiPAi|CYgfFj)0sq+|3>+sM=RYb^9ZMP;Uf0 zmlexY&Q>%WpTkrRIKHz7*S;|4R8!a`+k!vvs1Z9WDoU)WCYWP>Ix^5}USO8?bjYt7 zFlsNk?MSKCrH@}aPw(1Or#PVXFUB{EslmdM`g=1j=3#xhJv&%n<#ybWa*S7Q&o@!! zlD`t}1K=~I*_Y58$EXpw@Uh21*o6ss!VqpQ9h{%wfXd5Kp;Li1>J~&IXk7`BhsKd* z`Y9=HDy8373yIwK1v#LYsL0xV}h6q;T9XAvAU(6l2jVaDynj$4^aZA>fd@~ zh+K;vqGsrAr=g?PPVF*mZhZaqgw0K_nku6;Oz|~djZZR3vBHshWU^J5UdY{t+d_?0 zi%oBo4d(W&3-&YMfVVzA70qLNZJ-84jP5B zAD6XnoShictw&bexep!Z0^UBK$=b$jY9y<#N&%Uu_E@eP)yUq+-6s-uS;Q*K{HvE2 zeKLxVO4hiammmgMmTTCStaTPqnw#NN4SpO(E#NK|u&A14$=&R?(iiJqN*-r1%U|vJ7?$ zq@x9pPf@U%+;IX0Z&PfcKuVXcs+mj-*{iOq&BEC-NOn*EMOS)>evwd5^-Mo(jg9{W z>S(QOv-@RkfIXrb z4fL@D>hmZc9Gjj`v!i^|iq-b2YOyVEyHylCV#(WGOij(LXuG@`x6cTDno`|?PlKP< zkuL9~C-?{3c_iyze5F)(=lLH#$Y<+#jhGlMjV)Jy|G@%@AINNN*uM*H$!hZ&&PHnU znHmlepjeN*!yq;`=zG`&W!=!Cnn|RXnLPFjBy>_{+fzx`yuo3=p$sTC(sb?7X}+Iv z*^g?6XVMZo8>6~Ed8WMm3C!sKbx-m4fITFfQI*D@9G-zuu37zvyq=D`FN%ezE0hos zAG5cx-=G}f*MkV=zR^D3i%3U4i)N6Z5qw+JxGGPwW};q($AfYk_RA){Gz(lZK7^Q5 z378fm>P|eq3>36plJwNUv(;Q~MsE79E!b3g?+M?~r>W-oOHVT*m{gds zH?<;$DI8xD+suQugXMG!xxSI4TX zB*@roP#qI>5_mns20l?5CA+dXnGF@AP_+HARqoNnw7o9aZVf%L$lXW8z$(r zB0G(HPIp3~|M)Z=1~+)~;SNE*kA>266Z-J%u~Fftix>sD2N*E9%tM% zw))4`Br0@Fca$3atf8PkE4R4d8TRK-th8iXA3QsH-AuNHJ1L!1Zo74QABnz;*gQTP zeKti?mn?qzd^oI@XAX!;?TVt(t^L{dF|rS1Y6za02Hy4a5T5Jwz+?Y(qxnuS)7#L` z6;xT+P?vClQhJ0=5!Of|9D|rb3!Fk@E}q>Q*7 zCJ&RTgdHJs1sNkiSb{7yWfS%vL~$p7_5wkx*c}jIJF6^eFhO=D~AF0|v49EFw5 z6kEg8u#fD2C71nD3T-VD!jrsVwuXIJ?krV{Ig-UMg8Y@q0$N(1Gy!8$kS5sRvlOPb zYSuw<7#UNvEcS5ATt81*SRJ0&5+kw#lw|Pcwolt1d>F zOQc6*iN|j~1Y~^;s^t$A=SNQB?rTpYRpK~`mm2e+kUTW|U?WBRt1Bsh@=nOp(j;D< z*FfX?4=X&J7*@gENTsL(Md6!*MUhQ*-9ARxy-pUT|3y|pN@81c-GT~Jp?`gQktii{ z0Tb-3o((DybFw{#bJe=I)Ux_hyGid(*Y^{DJ*IJR$Ps@wPjS%CZ46)XFsCuI_UE7O z!_}?NEoN$LU`mqTSs;zhd-*lv2XlB-n4&d`k{9xQ<%Ty~Nk&&_?z(*S`cY9=Xr$hc zi$C~jqq~5?!$=LhqsO+{M9A?$U=_L1#eMEnX9L9*r=irG!{PxLo?8)`%UT&-X|W*i z;t=a-{xMxw$x$>4%1lYeToxaHZl=uM`h^rZCZyy?&F*kU0z}r#BIi5+Yf%|V{3sO< zFVp=L@tQet4q%jH)zK6Me|Zvsd6SOep&y+-lafQCcUnoP1mU&u)k;Zb%7Ef)qwNx& zF-}8CsDF_T#bgG(<>Y$@bBw}=uXG>bP|Mxfu&4aeKAH`i@WX6hu8vkWa-P5@=?^B= zc^yr(lD*%*qiU%9$bJ9wCI6KJlbFV)AuE0I8po!AaKh3X8gpqfwZ>Gpx%?K;QMw*( zano~Vy=>bjAeBV`uteZV#{)1h)@2H+jVa9)@iqYq$9E}o{3sQ2z-J`h7LU3X>{1X# z#rc3yt0+riY;dynzfBsFesaJjsXp&h9FjgEpmjj7QTwq=VPhhL7=l9rMBG~#lIBxt z)^N{6=30=~TaIa<7n*b8-0-ka@|b8b*ig~&bF-0(WE27oNzakCSA(0PX16|kTNQt; zDJX9@+4=Fi+obI25tC%2jI*(sf?r4cxw3evuqt5?bvS^{>oozurf3*w zehEogy&O{b-Q>`JV()wR=Sda{v?ihQYPV6y8cZDU!q0h3)q3w za7f{OWmJ5%+t#_ObC$R#|ImrF>)2UaNP7UjYTm11lNxk$QXX0`5?v@&3-1vlCkd5 z-pH{$EP+`Y;i~(fnT#dVVm_IqNvgvjP!}bsHl;jQ#@nP~Hikf{6GR!4Gp|8fK6l8g z0t1S`lF3W8Z`f*JTs&UY-8#>$haBykL?}uYJUmkmYV7_Gcq0FRCjkQjgJ%#FS`uF3 z`79{jz^TBgFt6;h#H&HNQF2t*O~b$cv*x0VQM4VvAellsicV>RwxJCVYW$|CA8dY& zPC3AZWuWETZh;@}Aw?x~P@O_reflGoN~ZFePA;B#3m-%!Qx6TsZCNdt(jj*~g*3ei zx_tR$wS9IOu-MErIX6`dJK8ymUzGg1QVnb4z+v%GHayQ94b%v5?Dz=O#V=(LW7k0g ztu6wfh?kdrGgh^nn*XcCRyDLuvUiaRMGQU|1%~*VR!E>Oqm@dKc|8K9LWBLr`7PEE zznE?g6}O3qlclI7n);JO*D*lXMh;(&i?s1`?PkH61}8&sZhG-tND+ zz_d~Crq_VH&SwDwa?iw)VeqZhgPD7Rl`Ry*$Et?}X5h{ZqGHL+4f43dpl9G`ENU=y zm6kU<+R)GZsiKc}oWIfDO3YSuT=rDgBsIuLCTEB@hXJGJSOp%T1eipounDVY8{w#e0vKNN>&)*Q#K3&**`%pViLt`R|FgxU6EE2D3`$GEw&-5bm4==w|42KIDFRg|5*NUWLug&3kT_r(W!@)+K-u zRrRE$%q=RJQGj;0U)&^29642jUPJNH5iG1`zD#rl!s3TWOpE7&?*|YEmi9I4_QXZM z748zFr6nvO@z6A)15wX&j=miAlIy@jzrAZa_xjapxs}2Ij<(@_2XS3Ml(siHC=ze z!V*g}8dYSMR)R1WG<+?|qWYx#THGo4P|Fl`z&tl*WX4a)nBFhY?`V-Zes2^}1vH88 zGP>^xHc9mb*()+s81JC@FO=?2apCxk*Ca785-LFjkZ%-8xU_7|e@)w~v_pBmLSYvQ z;}`=;uxmRdlxlYIh<2i+V2hcgi6~z)Jq$LbcDyp0hNY;OWYTIdY|#7+J%oDy}C%gRKF@X$uxAAqa6b+ zsEcOLLfT>7k>zDr$+BzV;U(qJs40)+Wv0?Z`k>J?$vz6maAI-H6BqR{)lGx|T!izB zMh)hmwd0LN#YPjSgMTSso7aJodn(r`lH`tWTMcK9FWR^~{b);dsOfpb{ z*t+rIw!0q@q?Kz)^B2n2gF{s*a$(9DII8R*>zZYwJ(pCvE<`IDYw$Kcj5EpX18guJRa?KR?)gKM(ec^7}FqM(maib2$v5(Yu-kcyP* zcxCJ!e@$Dqj6qYre_{2f;o&-@E0+AS7Vv}ekCxO~@aIp#G%PA)P&bwe#)Fqsv9P4X zluVzL@mCm(Sx!l2U$ZN0qWck}ZYK#yGS4TnDWf@+p;1QZi-Q(MugIOhqR4+5J&`_+ zBvFl_mE@981yO$*>cl9Yo3mp4AtpVite7Ra!-$YVsZNlr2U2$U)j)Zh9KIjk6=z1& zsU6WG)s$1DE0;M-R6CIbX#b+Tme&}>0AEw2=0TP3NY|gdkj&bwLZ-N(>03%=Jp|EQ zRDqCTV>Gv;P&bnTaF@=n88uu+wFlNx<)4pQ$z7NEeqIHYwNND}Q-a87WTRKAnkg$5 zQ`~6##q{QsMHjHj7>wn_l7r~YtiUE=BV|F8V2z7>7;lI%rwbL!Law6W2{)p?4BoSf zRv##>Roa^jt%Yt|Fs2VIi}>6DvOZZ@azH7)l8Q3pf@uSl>E_<$Q_A_TL5Z5k@dX8c zY^GcRDZgdKGQsaFS?ZzRHW#4|HhAdL6+^kQ{WB5(s(C(>4K2-7fipIY9(qO0m;9pxnoQ1@j1UtL`oe zqD1ruii__2gGx8$!#}97{6)nGwD|-=f>X)~p53L{%FP~0EvmewB0{KNNCm`Wd{yCz z4rYO0l-lnGQwGl;=yM`>Evg-iF`E)nzWnc{n-FsTz>sISVq{r@za!zZFhW00VF%yi zA1dDBB(@rd=5K}NxZVHZ%65tKxF1lh1rs6jG%O~F1-_dSQ|QqT2oa{lYy^tvB!Fk`?|59lyj ze;H%&moe>s8KXL|rfn5M&LCA}sP!Ml9LW9`W5CRvqwDy~m>pU6Z16GgUKiOE<4f%# z&Quy2Gx8lESneTqwZ9V6PVH-8W+EF*2k)-Lr3TUjdr>;b-cA-o?e#lZv!E%Z*INPD ztf)4p94c8-R2N*vpsdxK6iN@JF}wse6g_n1)m)7LbULiTc$KlJBrxsyD&y*CquQNE zK>ew1Zea-ju#vW6JTVpPKg60w%n+CKXM-Is(YM6qwrV;2W?FsBN@UR|fs z!UR7grKqQ>Wv%U~Zfwe+`RL(o_?@YIZp_jRjk6zq3;2z6Ae zQz;TI2Oi;7pHKaqVgJ<^y}9j@+W(C+A^**p(0-a!meHne%5^Vt@ipiR1>=kQCy;(N znD8H?xa9w%ykP47Ro=g5)gM^%g*&kEP6vBvN>^iP9Hd{(0>`spzS98#8{D7tmoxG} z)_*vY^p`VqHhQSuI9c(Z83`f32Frv0R7o5T^bmGCqpaiwhJ`qo20n*UUAS})D0Z8; zn)vI>xvI^5BRyMo8d?Vwc%0&D9-8fDa@7D3#4qXehw6?OPEbuz`_sD zY$a@o&9-1)Q$Gb(G}hy7Kt#!AUcH9Q!4@M6052)C>uKQKwC=$0*vM8$5J{H0V}S;v(PV+Oxco_ z69(P+K;h(!pDAOo{s~F3{{r*ZtP%l#_F(naeqK994fb4WFJJ_(=lVjx7BKp;^eU^S zfnsZvgXrK(z|kl<9i)u8&qd{6M_%W$s(6}(<`j%+E?dz|?n4`4ppAmOA4+osJx1mQ z+6IbUx$@;uVhX1Vg1e^y@>|HW#v;KuwBahDFF)lJ1(6~eB>|E!q*Zt#!E^`9OvzuF zQ2-hgoA_l_V961{Uu4(c`he)G?Zdfy+2!Ug&qWQuiA*dd4@hw}A+(?vNm|uYc0t=B zEtqF=>=~7i5U7#NF&|q54oMS<0C&0k+EK$5uYbm}lQ^AduKe)pn(}co(VVl~Si8Yh zG-2`>c-`a!`Xs4+2*juV)(+_oBLFO8SGG9)YwCSLep_EE#lGdS?`k)q=_S#OF0Tdg zqgt4toGYt({R7Ki3V>np{tL?;r8_4Y_a?}*s)@s}a?~STKc`5Rp&)@WioT%k|E@9j zPO|706bxY+MJ0FptsuW@Ct3WD^?w<|>w+6LK-&tA)9UB4VoeFN$qO3z!<>;_?4mGi zQ|!xjautn@V8(!bqwrt-#Z5Gp=d&>TtH05}4HcEF_-rc@o=wu-q@haCF}-z34Dc)W zkeJbb$qdCMRbOgoJ54UjXuoOWY;NKi3724>Z9EJ*2 zHd7@A#~Q2AWYdBfhEpkMWMQWcduU+>$dhSL+cVMq{?pA$r`~4v{{_W?z@rZr3&2aJQuE)K__zMwF~RgP z1JxPIJ4X`oR<@Fwsg^CK0piW`Ic#WYv%sLB_N7Bxp^xRxyHMx9jGoFPba4PPdof*( zs_UlKOtXObnpSfn@nKU%`B^;C{U(@M;8;ZONu9s;<52#X9vLBq^_geTG-Od~`9%I- zuKW)ySwLs%?0SyBuuT1hB`dblR&|M1;R=B<3#fqwn4kK`iPHadB7+rc7%TKFAg$ju zZz67(615db^UA5(rm8GYVP>pDhBkXXeL-nt@hmM`LR(c=p7tM4ivHJ={+iDOiS4%H zmBVax4VyNTqUtJ6;1NCHM%9-gdot2RgR{G`-ZW@4^vHs-XdrDQJ3-=8#y0{d9UW;h z=%GGY-r*u$hJql<@vJp7!M}No&&PlAnAfaWFbR|l!6bNzx9xORHf6M1%Etv~Y{}Cy zabYz@L1{eS`b9D2fwY|T+*J7sKdfZeQt%ddF0y7bAYlr{no5dn5icLz^?AL34JVuTmWbQGZ3S3mie>`x&F1+FnxNOQ2=%0kIrPGs>qLTmT@#I1`JGY+Hn;5<>5C7=>!CSa^Ltom~NJuyr_To14d`kGsxW2}X-y!9HIX30Q zt}AQI)}MHP!SP|t0{^WJ%YEvD7uV3f79$;Zt{cYg!APaop8uD}YXv`O3jEv==`w%f z_4J+VgJuI_zvsJBnW3mjqutbXKNoilvKx!6N};uv2Y+9x7N@O2$xVZvvNk{eSs~Nq zSF@Mvr;+i8(iV>G$^A+m4Y%nF$M`oq`?G_puRo6^?H0aiI@Nq9Ms-A~Ff@xlsx$Y| zZ9LTi2@mm~y-50&4h@;VY@Y3Jy}A+8ge-rR8=;-@)Vews@_DlFUO}FZ)2e&epvSO^ z8V@Ml=E9DhyPJ z!}|iK_L#7mtnQeTfd{pduO$8ft7pikEtLHpas?j&0Z@RNqXYZ62)^X-h8Hr-#m2Bd z*|~2t6ImDCvZsv!g03KBB8 zmWQKyohR1+a)@~fsSF@m4DLt&zxv5CaY?au`P)xclQQL*yG|vCvvXufwqgU@N+kk< zP;fBzNkhhx6$C21iNB>DrVdRXR)`B8!(>GDB=?F9)l3t%0{V&-8&>s? zv0zh)$Em8OYWd_E`JixJS*CR*zhFu7+Xk4goV~n_#(Y%Z9cE1|(*|6gll26YmsTZ&+4K!yK3U$osr;K0zqhQr49t z06|DsMejBR>tUThrLA{~Xg+$1Rd++~QE;jiv3`bI?h(v*gQ1D@QJsGOHPU#4`Zz}* z+n?(y$%4eZ*0Oy(RvtX$xnqDZL6I63t{*fk`09!jD_Bf=6L0FX>*%2q(^M-+W(B5c z-N^%%X}++&;*rnr;~b3Bs!Y=#h9|}wup`3VXisjCaI=tD9o;i>jeeFRAPGT+K>X7Y z_TP@M*Q&)X6zVD)J}@er;Y*DkIN>)xLCVI3p)Xl#XBO*b21TYXTQZ@#fRWJW>_L(F!r4$N z{1of*6l-I%b5E5PbxcrwiuH#bHz~~pn*a$-23vZh;iJ7LjZU4G=|_I=tG)Z>tB4FW zyKGqPp04hWj##CIp>dr3?@!DHW7Frmi$f6)2iIwR>xv&dQ@`}A|7>iSb$I!DmPl+p zJtUSAF2g+40eC6$72rPG!+Tfc6s^PB3)UKo@Y34`ZidL!moFG6z60Tk2tdiYSY@2S?S&ad>g_ac!F5;EUpInWj}-!jl*( zb33zHRy{|iO@7w+{!%59xp^KbI?2YV!ILyI zQFPFFqb-yoB40(spvNB^vAPtCVqJzmt&8ZpC;6jZhKz;e!~5oBPri8rPK;UgY(N9F z_7#MLlJ)1<4Z|0=j>qk%uVphs)Uo5^9`su=GjL&5m#FW_o?)Jz{+?I$4utN8i44x7 z%Lb(F7A0#^zM}`+v5r)*9mIXifS8?78K>JhNozlg^wj>|Op|ljFV_`o{cIzo*L=@B2aFWRFls#~arY z&KXhNqxkOtR%8dE)BL8Ruk{>4`Y-+s8rxjQZ^{#rRX0i#vZ{L80k(Lx`iLo8gZVn-R z2V_*=>mPl5Ge_VK+&kl~t{^T-`Zwggo>e6F44wmrP~31|n{S`}wpc#(coC3{OmA+! zHvZlm+BjHU&&2+enR><0+4)+E_OQ|7L_Bv9eOI`3Ga#H8g3+ElF*sU`llGCCm>0u> zi8=t5y2Q$qU(Q|HX;_ygC~3)%V4MpP!_(E$LH;L7Km*b1YN<8Fs;pFjJq-jYnv4Jy zWq4Ek18;m9al{$pqGdeim@nD>o{wbI-l+fLZ9;1%N2T~~z)fbr!*6yMu@+D|?o>1+ zM&BTY+5cket;3>9ErAxX&7(u$b8;M~EVL-aOyGv38NojBfB!+IJ8ziJb8YBb+ zY2HDfUwrTT{o(SuxYq2m&sl5T_r1>9XP-hXd+t+gE^cw6B?HLVXQwvo{z#!pLu>x` z<6nDMattVmR{s62Uo*)P@H|!_mmG;Ov{y%#p)#@;j2JtVGC!t$>b)$)LQh1M^2bCO zm*KcXUMBjU6*`%J@4pe8gTIi%KY*c!tg-iC92GG1LaZf`m$%Y1JqpMckS5Hu!WuBPHy>{yP#EaiKX z|E2~^lCOq3?3itr6WGo~x7m90`X{}4z+lGvsgat7Pe_qcKlY1U`Xqq)TqIbHN1xdd z(bi`D4RO$Rr!}eV-?I&(pXG{O&mAhDUTdc~FGXa91)LmkpcR1?UVx6@^&55_N6+7OZ zf3X&3_Ijn#w5!sVT7JEB75h++S!R`5P7)S8kLO9uBOzI}i7MAV;k3AX8lT~vAz1ZF zL>k=C-OrJl$iPg&@^G4b+ z39z?UH+!bL#9dy||Cw`;Li}^)>b#WdW<|--{aj$!W?NsGZP_2hID+Xuaa>{}!twRZ zv3iU?%b$nX2tF6QW!0f}k|oWqc8BF9{^DM?_wDC}^0u;Z^*EL;C%KL<%SPM(9x8jH zJtZd6?c`M2W}T)fBh?=u@F~Rqx~b7oF;=IvnpHbgu1+Am9RIi-8oQm=+?Y|gU;pvx zH!)%(bCcIceLp*M2ZE<0U~9&C=h@b~-WgS^Z`OKkaaXdF4pu$Bq$$-PSXK5_Xl}T} zuJ4bVg$GG=611WXfI1_vo>#y0d2l$95zX&a-=*)wt1CzOa7%qQd{H^iiur;$^4kLt z*3De<`ohel`*R;L7Os{_mqz=%L5(KCymDNwv58)hr}m*PZC=TR+o4SY6^Ch=;bFxB z3&tT+-?;x*$ecOuTPVH}>Ldc6!R;ABR$9IRTEt4BCW(_u5s`&RG+b_McD<4qw(XPIr3gWx zt6c0uqbouG*71Jq{7XILQQqK}ZeaM5NffFNg=%z=U>8=kVf`PRY85=F<0B>iuvLN1 zMW9RZ;WvdQjhB31LlxJjAnLB`T7EQ+l$#sGyXS{Hz%?dC*R2fXAyJRfP$jx=Ei3MH z+HMR-B75cZAA$w(Dn08m%nQ_*|gUd}>Cc1!B62__pz7lYO&5V9YAVcE}d7w^i+w?0KqOq8Zl_Jen7JhGeU z>h3)~me{8b(~=k7Wq@#8=LUe1aq7rwXHwh z9;%;Z$@jRGh`X>-=d=~NM& zLsJOBULVH#LUK!u_+@K-@sB9GJ?uoVa4qGM=g?1vR4D7{^iV`X`upZe-`d2hyv`QC zKu>?~=@s~fyx?|U-&y{i#u@*hYe&(!SKv#$LEoHA&Z^br(10?9j-!vkb=P9-e)k6l z1_O-`>Hn8RxIEjTCsMcAh^D>!v2gMp96l;>z&5> zcM+1OH0o01uO$RtRLwtNz92*24)ibgbnWi7sCZ7m{UB6u_r+;iC;H5!y);A9ir2T| zHr6{YY{>MihTFHVRFA7!h~)%Kf7ENB?b^pwF*vD3cx@~WE7RC&_fmhg>OhAlFL~-@ zu>M>~f`_VVAqJa@EPB(i=!;W@)lUwsH!OWa3oS)kr>!>unz;jhRVLFT*4Bi+GJlAi z&rnlCx6EbF;_iE-+-#LrFUWo>gw?Bvo|?d{SqEaayair<5=(`jNPB=7&+E^vsf z@=Mh#lRnyY0k>Bw*Jd+2Ugm*>S zxgd*hJ)M!6Q!?L|=PaDN?KmT3f*|j;0rba%h0jxm?=F$oeltpDSnFhwVAZkO=%7s^ z)Om(9yyfK%NvE=bf1yZ7*v>Ni%r)WUWA3Se&*c}BHJzP>pylkK_1#T}a{U56|1dQ0 z7ibSlX(jCC&= zb)pLf^l7l71^A@rSlcu$h<- zd2Fs=A&w(I-@q>Gh|ywmG5@Gt(4j)7+iqAI=G|bE$2E4jx%wMflG2K?8!K{wsLm3} zDr7-~%c#bekV_R87Fbcisv7z-+H}cJa^%}#?pOlrGuVBT;C&ep?A=d_#clZksPK;qxm*YPhFug~HC;iavL38+dTN_M2(5VXlsy&d25S;U*gOdn5~Z*%`FM^V8988S|S1 ziNWNI(9GFZ7KJqLD+--rs#c(>3gjb{~=aS;R6F+R*7k~+b!N^u&SC&SX5 zS@TN~!3*SJn)q50Ba+bCiWf@_9C@FkG6xOe$I<50yXbiFarXO!PXk%;K1rs(mB8oi zA$~q2>#hYx*9NOH2A6aa>)S7VLMX`{{nj~2P1`iowDXFaA$1u_m zEtBLI;vD0o3^dYK8ehh_%w#&;K|ws%8&M7~pfw*{|IR1df{|YH%Lf7`S`vR^ z>DmgWrAxp8OZtfd`EG7ZBXOJvC$u_A>*T`5PudEdB`M-17ng6C$r^a=&3#d4i?Qan zm5b2Jf%t;oZ$B6j@b8W!G3$zWm6z=T|fP zc?&w438u6^J{e!i`C*+Q=nZ=H<>vaTw}CYGAc)@%rP|sV+#cvVHNN&CsM)X7C^2ADcgVWw_0P8P>GHgu6Qve&M=5h4;z% zdvN}&$j4{@y#2K46^z&33{UoF_trFbq2%!`@wA-IhDoarI=f4sZti~tz^JQ&D_*|c!!-!Ex0Fy`4-akN|rY@74x zsU!Qv9&{vm?39Y*xRG&vX(Rb3&?}U$z&O2qEy{0y$D2cK zVjP%Eo%p!CI)S;8Sap8xBYe|3^8KiewsZD`)7hzNGrZ;cYJ21-?b3JIlDGST6>7oU z_e&2nCi2@Kk5ct7*Nz^*Uf?2HnEGvLRlD2R&gS>*#;PUt`_`+XI_JAMiq2z}M7$Qv zpVL8hSL+{zZc&9=DXiB7@!sh}Stf5_L_>_fZdT(Vcr`jpeJ2<(XtE&dz2$dxmlN<- z>3*iY(5XVp%H8Ez^ruv3uFiv3KkiyD3H=l`MT@|z zb^Nny8+bdbVm(3U%3sVok<>up`&UCWL^J%vO09G(<82#5ngqY8b$S`AgL5ke`y6-@(XVpxvJR$YO<+PIB>{bG4SvG9M2jvqZVYkUHEsn;ge%9guRnfPUkU{Y^@X>xHQ!zb2l^B2<_4v=tOV3t zg9SR{9&gznPQKq8INzzzE=$Y{I0dKQwr}ohK$BY79J!`fyxi4pGh;24;M@Ixn{U56 zmyXX~mW>VQBHp2qblH(~^$44PHJ?dlmS-@@VX3rW7k1%05(_&(&(JE0x8a+FIm(&G zy6pHOvurKnDZ|Bev6XovneZtS&v`>Gu+bKMi)`W7 z&CbF~dhzbc3vbEsdlYUN`e#kx8qt4EgZ z6it-MteRgsuus?XQBI?nomAAuDZl3cWDo0p8y8D8^AbJRm!fNYxy6u!)iy@0fx5bb zQ)Xv2;y~e*SYQ48+xUX^*Uty}hcGCPBx+4*(2 z{kL6ACh8q^toDXL>QU-_uJ_GEH=)h;ykus27%&L^C{{8j-RSUd$C0dhZk4d_8Z|hj zZCn5yRr7PHxa+_%Mw}XC<2hWu+KOr%W4UyaK_jVjprXkM#nF-kwW5L=m)y3c0Z?W& zh1h*aw1A#;X#ZE-3qY}>#k*Q0Rd;!VOo?F?gOWb?JIk~w;qVGI4r#q;SqaGjJ35jV znf?HBHobMbf+xexrblf9WPJOh#UiNVxm`7;{9%AcRp#=zg3%tV~}&pSVs^+v+L7R8p)^19ntl zcHDNkPB9yg_6@@yO#_bkpd#fWRDTxC^vv5sX5K5Zf2|0aSBy45tdWl}$UJt$)0uUu)oEX0f^JJiTiy`(^(lZ;@Bq#Y?O(+y0py%5-qj(g!7L3I zBoZ?#c#n*nl)}6gn{z(vHKea^|Hz-vG@%(n)!$jwRj4KTWoYwcH0ppOMoTDV;bbGl z%+*3i`HfZi!-&>FKO~@ zkr_xTP34r(#L*N5c7*xbrl}}HCBE$khHqHEMnBrl%IF%+`jk1Y=O+M%b&l1bBKaZ| zf10H!%-g$W-dnTtRCdiPS{sJeSVkCn+RsG|D$_yhvHNYz<8bP0XCn*~GS=E15*WC60JQXUXqso2lpd8Xm#rqpg~d{Nu$L`BD;|O2}XxQYrReom)FVA4ZfS z#bV=ZS?en7Uityr2M|ks+X3_`rf(=1S&ULBCb=veiK#>K!;Yytm{hf(AZ&2$zbsUH zt!q*YWZt7z<<4k_$2&I%w|cNnW=~;U4KT?EDxLMZu&dWLe51P%>+Q>12{9;fXh&m4d2J)J64(HA>nu;UxwnAs@!TWO9J1 zLV#pmaL#Ocz8uqXhowee?Bde0tKKY~C)e?LrU>z~*W$kA3=>$%fdP3WbVhP{@GNp_ zv%+*}JJYbgxf@N-wAzL!!?Y7NWVNgceNE=5= zKz?Hg9~8IULf1U)e4Kn_T$p576$U|Rmu1_DHb~&_-sMoRfWkAJmF|byKQd_VNqJq704<>WE_GUlr-Rb zY!ozr^ke3a(0rm40L73Qvkpn@9M#ZMbY`)>HIL=$1q^RIGFfyAl4^X=-8JF$TLS|P zk|zdCu|XY{keI?s<6xMAAUoN#!AwLM3CmnbwAehm$qfXUXb3{r( zZ<+j2j0&2sufi*zDdmXH4|{j(N$_c43ae@%mAw2w;;;?CY@itC68&yhL7Qwu>Xu@U zx@t91kK}y5V_083n@2G+j-r{qQ0v0h3@TfCVj}>Dyh~$|9k_CdR zfvQWB)k3=lDW@jQT@7`na0(FpX2#;d`WxBfdlaWzWwJd`{tL+qW2rA6PAa<<0I#B$ zY>sr_SbWH^-?ry|MhpKm_-&*j;aj4X3U!-aLo?Qvclv`#j%BW^Ng-Q&+n@9`83|0-Luxw?>w>JfP)RM+ULuh@&{<2%S;}mv4O!nw- zr=3DK8IEe6KS4Y)l~UieW#Q(nOr_znThY6CZ0Z6j4fMj-Ugbb)F&YTZA0P&iU08of z&M|!40zHiS1DVR#D8+d72ez6UgF*ui5#p`~YEZ*p;?H*$u4snCk=Hg6LTi(h5^-R5o*Xau3v2IcOt9MjU`RXhgdAVxKTch07y&T}KC?)FYCgIt&S0tJ`=%)F=RQWg*0VpVXHdQH*-XTkE} z(B9^7h)m=0MXA^YjWpVSs4whcR$|Up2*C$>ifI8&9Usae>5X5sEZ&`sswwDH zm&k)=rB_m(@fL`t;|&^9He)o%R0B!6^-MhwyY*HIKT?wP>NkAVBgd3TBXa-_o%z6n zLb{8IQ-f_h$H^B~SxsWBP)#yvBvlQJHYpTxg~2#AR)}*z+9uWiII#7^?j-|U#vsJo z03w{X-ZdbpyUQCnf~rI~1Ly6j$)qG=8ib4f9~xYEqQNJ_y`Wbx(!fx)U~AvojQ?HC zQqd?=ltRNdS-SBkp|rMWh%5wJFo z2JiRs#qS6?p(}CRHpV5a1wG~3%C3ok1wCp-2{efb6wuG=dIfN4XF1GJ+Ca=I$)dK7Qcp2*w8Ca}alljg|(T7F`pA?w>;?0hfN zL?V)tB)J?Hy0QF2uejwyRc!)eiRTifY94F5Jo$9x4Q5qXj<3^1$$3SnNT)byK7pty|Vn;ZBBfiRx*dh1WTbr4cRT*Z}FKG&14IoEk^=XkWC z2=m%HaPw9hdNaMqg6}_asPDTAd~=yb{RTK@dn}dFY~4!ahcwiebIH(#^#h1eb?_DX z8nYO$YLSL$sUMYA5eW17`Ho4m3bKRm!}kq_7dk-k1OG}1Bk7KP3&cEBg(P#n4RUmB zTNRM7F&BH^7!`Pl0JVG9+LQ9hahX~btu{}>fWQp(n}ozytYQv^uXe3Lj*IQ;C2HBf zDBECP=AS`zN=`juW~E3)C422uUcbKn{8K(K6k#Ew;j7S#SD5QnB?)l02c0B&pMG_ z!y>_+#GVOrc?n$6m$_g1w{SWk5g9#NpH}L{=*e6%1Kr7gfDjx$KN;a4AP(2hHCuz# zAO-uMADez1LPe6r(k)q1S%0IX#0wwGUy{^TB?r!0f(!5mlLO_AmrGdau}(AGy5qe? zd@o6Wgr>q2`wIl=<_Rv(0+=EcpZrrTz=#Fpu&S0LyH^kbYwErNZvh*UT5B+{)>546 zdLd{Hb4Y4#niWu6CABzU4Gyidv;be^>ZzMLeqb1gO3<^z&b9t2~$&K(f?U*5CwG6i@6T&f2Pw z<--M#l+(^3G`+EdyW0oqL3NMcrq%KhevVS2#HK7(tRAwedNOGY*H}ZKf%gHZGo;A@ za9ZJW2xF}6$MkhYWdYBIuGHZ3c8eC!b+wB45Clftdqz(uRt6Fkxv=f}BZ%N2^ zGuz!z72sz_DL(i1Quvp?{J-!W{taIs>RozrV0~W|U+>pB;i^ijQY2Uvr2FJ`{Ph*} z=a_9bnSO>_tSKsCSQWyI!vd&fqT_5vL_8cv!sWG$0j$b{V*wPd_32X}e=2sUj( z0^@O)t9DDJY|5AGP?atbDqnHYE4!EEgMK#_^1n!wZm-q>$qnz;p&euJP%FJt;!01r z3Jb6DQ11rG8tkBh)iJTyCHX(#gjG$R2HbUn5V^)_0xVl*Uw!WPsnYjo1+nE}VR!I+ zsECXf88yL9`WL>ir|@+@g>M}&&Fm@!RbKKcu&DGN?Lz@)st^iAA&qLRKWB?htcc)} z|4mX@mB@O7Rn0+n13~2!oavBBPpBVn>PnA`?|>H|BfXWIw~UmFx99QSSOb~Bh%?Mg zU{9l|VoJqCVtelXS{N$Y-vFk*z2yF-B(lHAn-@MOvm@P^r``MH(r13 zR*q4vd~JI8;WO*yK>=wS(}f)!L;5f#LHrCL!fi{f9Z3^bV&aLWO8Pp=}i#&0XZSY{#Qs3^_X91X7H4Vs zrf%G2?St(y$0Tl5Ez5v3ByzW^2BTRVwx8j=(jN`^-B`DQS7yA)-7 zo=}Ee`8NrE7WWNA_Os1*6m15>!f$@Ho9vEfKxPYSoVz#Gi6qD%%}-E_4kT+@+5pt? zbbwqOx zaI9mRfZDLp5mgN_X@DgFvx8!fOA?+Rbmd0RJqZF(w{~zlklftS!JciN5N`JIxxyhO zQ(SE|_$-R4UvN>@5e*lx0_fFK0wTAdMamsWj;cx>cCkMxRfp}I>tL>{ESqtvPcby7 zw@USvkq+_p{KasCH3j0A1BcGY%Yi-P%?nwoNzeL)$=YRHCnh}R*{DL8mw~&^hiz1eY#bv6ok+u2){DkVf_HHeXRe589-p{l@9xLnZ5@9aH?LPHS=I7v*;| zx?Y2wGT+ZncfkavIpUAQ^>ny#ly9y1qx0L6FC(|X%_5{dZ~*<|3}>nXQU&8JFk80= z4VW*-;~OM=Ql1d6ejk`?Yqyh=u5I+21@V>%B`?E^NMaZAme(`+oU-*rMz zI~h5wHoYqt(x2$>3Q+96_3nxO`oPVOcOQOS`f$aVPnq!~FdE-|fU-Osn9dx9T|o=p zbXi`mSi+amRhRU`&2}%izPF@5?k%C7qgAT*TkNDx2=b7*ffa+Vj3$2b$8(cjF#QA8 zDaiPa1S|B}Dd=9K-m#4lYgH85UT?w*_qL?3!7DT(m!psb~{a7ipeMaZfxZ5NM7GwT>x67h<81J#vCk8$Rm2__$?t_-2C5Z&nNsJupYRC7! zeyeIqf3a79dtQ{Uh=tS|X-0MmGIhGV0&oes-vQCzm{Wd}85cU*ctc(M@i$aONtDo( z(`I0}005v@O9_lTgx^W)Kytty?AZ7sY=};)Gzb&a3#c;I`T=^QEc*L(%)-$mBP!zR zY&ZNu_kw{0!Km69_`$Qz3aUhMHPtntVR7arrguON%5U(XI4dOqd!sze;M;U{e*NM> z4v)7pJl^6#xqMmmQ&d!_L}kZF@Qd;P7N77_@j0A=mIuJxfNu=_Y*5+two}j>)w|gr z<+sQ6-CN3hOdzW;90j-Q8)woo+$nQ#&3nf zNT+BB{>4+O*f{K_ty9p43TQN=_!6F&9!7MR;#cnoF#R37@)Zd}7WW-Q@@!UtKhD6P zOji^XefR@ZPxUKNA?OnEM4vbR;`dbCFF^j~=?{>UBV1Fp;dHvmEZPZBJr=&qpbY8* zMuMq_H^_>?>KK11Q2blf4HXeW@((0u zZ~}b9YT}@K9QIzB8fOP^`ndL!mw38fgE~@!_6qP}w*~OTXoH56U4sTYAq{-ExFuk9#4p&P!m38aCw0?7hxXzYFH%dQf4H?WDls7JaZQ{-;8NBd?(B8&4{9 zqpD@LMj*k;K2}N0|ESa;SNH*v>IrWK7p%L@Notsl=hOK%s;XZZ!=dz-5gdDipC1Mk zEeg)ZdKV)BjGrnN{VTj;KViTt)=ik+oFF*TCo$&~gkaZj1$~HVq_6#|cf0=1t$VNH z6rTJ)aTNnEYmW8*ao`!8{EBY;JwyXQacS&$#vPUAYD6ZN@A-s=Ei7F25&qL$Npzw+VDNMuG6#h8RBsMWXm9|xwO2HGaXrBI#ldRIF5#GcWoT9=@&Q$%oVs6 zf7Hcbpx^tdT`j|*J|RoJBPgCkjN*me!}+$lYx$5{-_&X6c?E8hhG3#&^&F~m*zRkM zXL`)#)GyYp8&$QtR%WHF#(v*{1K4?|rfO%9tS1o7wPx>w{T<82IOkc9)$eFs!V z- z4w@9g`tu(-8PTz?IOx%#lDG|(w=;J3;*>;>lvtzii=pIG+;?}mmlCJX$!hV^tr~wi zg%=%LYAii`R%=E;0iD?b$KcZBrS(|%zfQ2-F!!$tZGA`e60Sm@Ob%7W!@G=1?>U<_ z1rGh~Fn|~uG@WQo*4XDV^h95iuK6P}Sk9Cl37bIQ7=@HQ^cjKvy|Swfu~CWIn}*CT z>2Hdrvu?OzRQn49UGRs<@Ln)P5s8$rI+e*ZG@qQP`bHK!F{5uUM^39G{7DiKn$Ea7 zH8>%$L2Zv#mnUJE5`sD-A%Tcp&cTQXa3HGYGZD~X8Dbh z_UZYsAw44Y6A^~=nYtdMRQTN1QC+G~RJ|kJ>}%aG+I)wpe;)zrjwe+R8D%-qmo{zW zI%Gh^9v#aILEYI=?>Y#zmn{m5f8Q{>(cajQhEQSn#D-o)zmeB+K_c21nSUP!?bZAXBLXQ>H3Xvz35t!;wdknW+8<#$ zPkEFRfjHn%{s3_boLzn7WsMFNi&H{L8|g|5T1Q!xV<=)*st0GuBmltXoK& zFlP?9G6><|e=;oh5LTz3D8LFOxtO}|3d0(mkjNo(W2+_qfjFe$r+)Ae>j@qY&Z=Hx z7f*C9I_bqbr4eGfUNH#~XygV1$Mk;N6k9vXC-Hw(e!@~jeT*&@!VuJm9qg_HEjte` zoW@sPWU?;{nY|J8IG)t#VO_$1+WE7mc77M$&dYCs9luv45LN-J3&YOnt2qQ74|HwC za)Y1h@L@JBPZMSzczQQG`xCm<^!4!=8{t{*?Wb_M~HNh%`8lhNg=16WF*d2tC>a+$rG?JD(bRtCeG5lDDQ4cuTE_ zaqu{cd=}tCK<+vv{If*VArx>R$B)>aIj5rRt%-BFlpoHMD>h}%(3Y#5)Tsx&(ZpFnhhWG znyU1vrCfa6Ao-Y@%J%no6bxr^Iau^BctYnahMfpMN3j+CF#eim>xg9;=RkYLthNw{ zruO+Go0)tm{(t>QG2!`lR8YV4xk)yEN^Q0*M0(i9BZJOTJIlN-t3CWh>qN+l3&QPv zqhT6+`S2HPvEL*nsI|gS$||(g=KM=Q#wCX=DRId>&sQy#GQX9ol|J{hvZ2?x{Di2( zZiQ7jNQ_*{WTEm$><=`pYuvRWEL`yGYUN2gtFlg|-e_kY)@`1kqeg*BqhE09QWGrQJhF=0?7578urdPl1 zf*MoASB$jtzFOVbv^f^x?3geMUFj-J;Jtc$INV;^&6s$!JeU3xvf7q7-QD=z`J{nJ zDqyQK;eGFl=23kk_$VA)chxpuK9-$9HX#zaVf;;GWl^`d{J=Y70=s@|k3;s+=B2=` zp84MWQ3>9i(2R|IVw!$Uii?K0FZ{KplRJx2s+AUrfy?nDkLFzRjn2-y+iW3`Q{|-h z7mXQsE4j3eL2YrtJKGmY8G_bJ@Mr(6_0bKyhftxOFX3}Px+XlrGyrIj(h>+FPko=MT%4rvew>V-62A zFaGj-Z7Fqu06|A~-j#(fF_Plv&=t?eDyKozm3(-vt1bhS%!iyiv~6_h&vfsJ;RDZdc?eKmlFW-$**sJRgXk zW`V3LiLCox>cdPvvOYPzvk+(CGlR~;N|qpN(~EUu9CT#KR%9O;E>nJPJxqO6y(C2F zCuFE?wTf&jWIjK)T4xxg!B74RVUyoF2!6lx+G`fG4`Hs?RK9_*%a7Rf9?4W4Sx_B=`bb#MGTfbr#?ZIM>U=p{O&4cw@zwkMeI;q0CV!ii@_T}GCz(%G^XmXLEp=x z)xxCd-*LvO5iun_mFv-rn&;kOtSHDJB6ku@)%61{mkU)FejHFVxoV-T>Gpi~C3TaN z{4P7a#eN99OPYXni1^LJ&TRfr)#&5P9uG{IVK%*ZQGJ6`orMCu;qu&nC#qS@t0(+z_(j|ZEsRi~-z|eC({aEs?@g}_d?0)Q1ROYzq<-8?%Y_4cxUr~@cwCbl91@|wV&G!8~Z*2v!CXT_^ z0R=W6h|VDKPiqmux0$t)p1C6`&y2|0NzQ6HT}*yHk|&!J-)@kUg*J2eS(xT`^L$P^ z_PkoOI*KSlHmceAM3)tz!=kK1mI50Aw37R6+H6o?Un#19ze6H=Ag4gtHj==!8Bq;R zmzDJV9d>mlO14weVqF}`XFB=^H@h*i5g}9e#ziXJo4eVav^QN=A`T2`i1RNovV&0t z-kZLUo@Zhl23)2mB)Ff&Why*#IIi{;+6$84FZ(y;SnR)3zA`uE#T2-w7x>JbD2~dQ z-kaP2csq@pEiNz+E|HNv?G~E7-xMo_>H=eO3PN&X`BD0;`7>KkAnR%}_o!E2k2<2D z_cDAE0Ztl2s7|GuEN6sWW2sHoY)zRF*=B?4XLE7d^eEfHA4$URkeZq|cU&fS-Q86< zU3kqV@d*hK##6hl$-6RHG?l1RbtvqG$NTh9so&A3UxeW)A?CJ@A`4SpE-2J*V!~r!A6&*wK7((J`0#-CH3z`iR1pFx5z#S}Pykk<4dC z<%X}*EycW@+G>BUK8N1OAeur|XI(~(d;7*q0~^hY#e%gqylXsl{+jv8+3Puwa@XwW z-ML;$$@<(fOg>;TFZbAG-Zx19D9uF5TE>Gx5@YWrW-b^p*LHG0%BbNRS~U9H*M55w z_t{vKf?MjP?(&JXYPkMiSHsX?!a&Sn4n6;lU`^q2<}yZ(u2bi37@;(hwJsMmd7tXe zb#R`ML*#~-gV`O6Pagtfa(O^oS3@3`j(cr0Xn@skRJrh+NDqZoJK}Qv8T=sy_PBC0 z!g$2GTpWIv0ijSZwo57E?hA#%YFxvfCP5mW`YckSU?f3P?Huxud-6QvRf z%g|XXZ{1OBsmYFOB_~_cFCE{r%?jjJJZF+qB3yj$E_84{C0s%_nTeZAp)4b1t-N=K zwx<@RQK@h*J(q0B%))6V#$6e$k-$`NpDm1D)R8Y~BlAUOXvlhTg2CFG_i}4Fu=I@^ zQVzoITbG;qu4_eej;72Ic;YSXvptOS9eUh;W2dH~{mq)6l5xKi|E7zDQ5LqNb%z8%%1e zL@-Yo>FbM+`(6K+^XflmEVwfZ-p9{j-{)*6vT`u0eDa3=?>Wr;b%j&mq#BZXFoL?> zAbE+<^CgWz6FrYyA1PS)j?bvIjJx4WDPC$m0(S(Pbd)hogfXVcG-R8oe&5|iO#Lud z!{2dL_(X8$QoFA|^0PB1b4al#)zHC-|ze5v6sXRmnTSJUNP zG-P!Enh#3^YicOMMC$rA>Ut(xUi5<05wu(v-;8#E--Xhpg+VL@*8@+}Tj&_u1SHP< zd_FTdR|x}2RQ2s{RTn(&Vxt%$MNzrcdaHI-8(~p;%Qf;!lQg+}&o8c}kN=bSAL*VF zKL`_^%VpHZ^Jpjay#2oP-1HU78gi&dL!Z%mB!WA=R9@xt(u<^!qIjf7j8}z>^X-X^ zVvE2MFh#1p;`G~6`G;b=^2LRZOlNje-L*MGcjwQF|6%0=QMP>2M)b3MC?t`LicRb&0x{^9_4XEk>gl12LDm;M_i+7e}NA zKet=ovDX_-y2no4gF$JJCO5iA<9{ko>wi~Xl)8*w+ti{~a#O?<)6W7OM%BFz)jgN_ zi#yH&wstJ}GRbKk?A(==`6Bl}nQl{-O>Hj!m{BuTqau$5!NSNYP%!g@P8T-Ab6j{I<`b4{6M>y3ul=iS7;fIWc zRbPXi7|ge6G0ltAtGE(nQx@Z=+56D36}$7Nb%G^3X?lP0;u_%zFKC0G*|3C`(mGDD z6ILpb%g%{d&xMzw^ZhZZ3zr60%Zt*JHDxAd$A}{cs&4GCu3Hg04g1u5*2Vvc11PHm z*Cgq6jotR`{K$W0;GDRPBslF#l!?A>rM!GcysY+hjBCs4*D9pTa4`dLJBQvxDv135 z4jLWxtU`hzf~prT?NZ(wZ4#c$=`;=uXq)qHjCYO6*xz)?FebN$e=HO9nky=8dLx&o z|1{3FJ_4cO_I|NP3tVjskI7(Wa?T(=lCUbnuCvrQ-`4NV2gA2o(y^DhKg{Ia807b) zgOA{|u&tsVL3Lwcg0r90csZ9YP7d?$=~3qM zVv5vLw2zBayQ}rTfYzS^I@shHRrVeN=g^Bdu}!SS=|KfsX-Dplu&qO<{=A&li8NIW zIF!D5HeO9C1aBO}voV_%%wf5Ch7Kw{3jWR;<#xE;0YO%CVOA5PD{cC(c!7-3<`Nm1 zI)IM zy;JvQ-AcrX;R0d#Wo-7RkZk29wxTa|RU7N6@b``g3~B&MU1NcuNz+L(4^;KRzeTD9nY{cbdcXzHm>?Yx-YUTo9grXx@E3e-@TUUEeQXI9VTWAoo zAVeTPSMiagIAok!B)UlA|Fm(PVM(xE7*GODh=d$uh$GS5>8nWO3QfxuUm0KnM~*bb zG&8MmK;oX611&YNw_G{O%mNit8){PKXrZN+Bh5D(j$X9y$M@s=f3EY~&v~A6?)!S4 za~;*9|NI|~Ir(A--u0MrXYA0(7K!|vfEP*PmfoYYo}VNdqK_(1-~-l*dU~I7oVx6( z@WOBQHHj{y4R6ZU?pU8ii9{kHia0P3VZZIfIti&jAh%DuBuCiG)q7-eq3~v_&qS4f zz_=psM!_@IYf>J>xDM;DyA9Qbhjn)2#iS6&jg@O8I` zt^Ls9&6^1wx*;iEfgp&TZYlDUjf7Q{zJcF!-PUy_3-Uj@63QhRC~cl-Inz?A%SNb% z_daX4IS%N#O?&Ta8?3$?x^@VB;2MDR;dqL&-7!|<4f??ATXq-wE=2c9>PQ2`c@iA1 zMZ$qtF+Ev*KRwpceXoq|BOO&g@0u1r)u#n3HXKCr@vyZ*QW|FahlykyN-_-;?jmu_+f!km^e4DLLbfEhWR}yB<_x%I8YzGA!l6J3 zx7kUk@V~V(ZClZ}(+Fhp_4&xN+H;1FFX>QZb93AKtOdyK}LTWG4h37rS{C08%8q=f?;fq0{5YJao^f zl<*J^T|FnA4yLxf53m4=xd{6VY3vaM>$fZv7i(k~TN`E{UFJkTak+fUJEfkOITh$8 zFm;1i=a20UF!l&Jd!h3hLtb{t4h2_fP+W9rtoe8(vkikPwi8QceCiO@aN6ALHfR%o zy}xu0w`Tf8URJlX>Acrh7fpPl*CIQnElit;@=<12e2`{`k(H>**BFkdRTRp_R%~>& zjVFUrRY)v5W!EAKo%eAsWdPXB;KSBYlF3@7N&3cOrh1 zZZmuA1h72}o*>y}q#4m2iR#(sN%D(r+kW83MmIc+t;q8GsfF*xXWo+Y*aFkY3GA3iRgoRR5@ga_zum&(7vmBP{$eCThNHc?5GQ} zAeQ5-xm$z|y&%qh(EJrZC>k-c9cwn-+=KbZPO1D=YOU%23^ebSc>-J9kHq%Kda?Y0 z!tJp?bYt)^j$()Z0smB9=;jkCP1wMP(-;fdg1b=4PNAA3>D%Mf$0^y+0HKc9)xWH{ z8-&TF4F?(STm3-Ek-pT;!mj>;2~~`CU7r=PWE})lv<@N!Flvfj+^-^(wDS;?PbX14 z9(dI#j5dr@Igl*9Fof$T8H69$_-g}TG|PI~V-3fN?qoQu-c$Wc7iD63!Y@J}uHhD_ zkT4dbvafF1IRP0J*G{!aA?Y)kG~rB`B-(Rv4psGL8F(S^)m!kJRsPY?Ky(Ss#DS>R zEicdCbD$QFOa6=8C>YJ6EW;>VR!5Z0&4GMV`{8j+MMw0&v8pbwR7#bF9%f%1M<|N2 zge9)et=Og(D!TAxK~ZtVX|DJAU~$0RoJaS2Taq7yRBH%kFA_qSwX41}2~*scKP9t0 z4lC9$F{$R6QM>el_;1H~MGHL7S3O(j$vrfcBShxX*Ino!n)w}W}9ROypBbqyK)hmNm1|g$k3L*E@N+NFFVa`2qwOl=k%v}F1P!GRXVSi<) zWa{j+O?6|}w~E#0Lk;F5{4XIa=DQn@>eQ+nA%T|rbNR`*=u3^)<(5K|QP_p8Av}EO1CBpG0X#aC7YjSRQ3$_aWQ<){5`(`VDL-q@9FUgtJxq#L7JGAP~9+q*)4C3FskQ6<2-x~=T=D^)+ zTBwbM1k5=X^5Cyg#s>xTi%omCMG6*Cb8{y5-2{7@kKTwTW(di-Lrn(Y z+a3j>?B`uWCpt&QLJd(H z&+}HudefHiRpn@)`0#6*EPzy3RlcY*tE6V18Br?CMsCAZo;iCZqP{bL?uCzhVq}bo zqJun~1j61RV<7U=oxH`o+Ts?o(kk;-!fe=qfe~+lzLVSrSHqy(ZLr-&vCdio&(qm5 z42Op9hVAY7!8x~SoVb_QbEdYYx)1vhw(B{iye@_F5@QW>Hm*87Asa~09Oa>9)TU;3 zo}+Y#W_C8ZhTc67wKa$g_P*D0h@|`7XvQ~PM6r~EgbaQ4B>p3ij>wNm{QhMp5Givu z|4tr0EVY(rtu@dvJ@^C1>=gNzeadkN<)J@r&}i~^G~T@`07n!8690aqWv%`)Sd@k! z^e}QA6Sv%=I{0+F;bnWZGd{HSKjVp;u2P^HW`!rgBuZ#~B7oyl=zKKiMp8dhO zp1`u`u%)O%=ik5S<(&*_KTAGx{JKGa?WSMPPkM#d1~e34jRs9=&V3=ZGBoE&^klvg z4ede6+lns8Ljy+|FPXkIH|Ktq2`bq;`g3bui52Zo3fwf+9YKK<*y}RRkd<*YdT# zwS4{l*9$jEB5$-zyfAq2ioA$uoVW>{yy%h_BR8hL7f@#r`rh~v{v0fQAOApu@oer! zpPaGp#xZ_x;!USCqCw)!!wL2A>*Fw(cz9pD;J@&09=miA(cYZ8@e-cJubVI2$cy;` z{yI*=5siqZXcoDT$xm)Hp7G!Q`ppM24I?^s;{?Ar_2RMXJMRz09g4p6|nN zUr(k}T9mK)B!F>!+l;Y20jGV#OkQ64G}aYTJah(+WDNlb!}h+TP2 z*2qfkfOauyZQGkxA72kxD5gmGsW$=kWUo5&?q=+7?8)#6?3{|fMT~ zeK!~b35YvR2>cDnKXvDx@4*kF$Q{G}0>MBQmUlA(#7bai81sb7sE3pIhwt>KpdANv4h>)#aE3f~V$kpoX~ zif|~dN!$_J-1M6u>`TV0;U~WLbLwPREFctLfNf0RB`8Bzl1)X%U9wZcGG?yIfBc zHA;3ZE;#_jgRHm}(C2jGE$8cyu9ZW2Ihl`u=Z3`dOTzPau!0%kOhJU~5jUdbA&in4 zu(zFYJs2RF$tyIy*0FRRdr)Co|x9VEB_L0Q(U9w|Bw+@y;9j=>TTj zZ_xp}P5N;Nwk@0r5Zb=`Jf6{iE*z?T)AATYqBeufFZi*BPl0PE<5UZ~e+zbNeoZi3!Qd0^9@1?mQr_mplT&lXAdsG2!rn9(qGs4DFTvpa}P8 zMovp2=OrPh8{vvbk*i2S6oCJQ*bvU1K0EA@c)3`B1}{NV7hNTa0(2h%({M`8;GC8c zZJR{ZY$mUw3Vh!^5~>1=N3@8mAdVH`w1oC2^OpcDgg~NHely5oTaFrW6edYH2V>2j zrl))_7>9F8CP>$GeU+W^m9d#W#c@A#@@Fe6dwiF~3&2a_CTxM|EN}z#hrEL0Sr~!f zz9fLI02%B>I)(e+0r3~i$-`dM!M`&WS&p(l`6vxQ4}?yTFyz&uaI`lW`8$Rb?kK#c z9rokg0}Kk62|5dLiCayJ`t^wmJb;dQCu7k~Z4w3)S&zaCCaE@2HB;&nsZXAzPl|j+ zo&zq8)Gi>T_w>=3d*diPt!e?;+)ud_H{#NiOL}!g@2*nSi1&g&`~~QT(WGF)ZvG1> zj2ZbZjq;Req@9|F{qz8*Gy)h7)EiU$R`awEXbv*QXvIr{GL71&aB(BFk1e&&C260( zanl*2k}I%uc79XXJDKINfRR2>DvUAD@IuUTjO%slAGv|Yk~{s+aMk*WVP zj6TIP06g{^dgp&w&POzo>O`s&sZJ`Z6P~9CubhObA)Pz`hdq+ES;aYAvEv zJ%N~~*CN?#efRCV43hvsD0%BHvdj~fOoQ2_D~XOjhb%=hB;Uau!-xm&4@fxu{u`b} zl)_x{xZp9W$AyP^mXoFH_ohqg=Qhb62|8}f1FSqsl)~C-j8(PBVwv!C&;L*vl#wyK zKd2!L>jHd`HcvS9Vmj%(78;YuSu1AOXsUv_Z}#E?-r`u<&-(!>>q-lcDKqntl-n zFioGKW+DIeL{Rx;b`E#evnX}^Y+$Vc?*xg8cz3+SL5zJ9Ub;L?WI z*9QgrQiP7y1@3w<5Z<~n1j4lv03`rQ0F(fD!2k&FHl_u9gU|G;X(I*FG_1U4FKo*_ zr`)B8ERDNdNG;lzXXbMXz$UWH===c->K2 zQ^@Np#l4F7`{5l(H;tL=)l29cPV9sytg3L#Z;Ky6Xtw!c$--a8B!*}CLMfzmL0G6D z2!s+Q0BYhscKYN!gLPibGmXMI-x;_a;(v>u!}!fWJ<|QEcaGs$m9T$Nu>Xq?GJyQ$ z0+0KgopXLnLg2%a{EzBiyhuYB-ep4xM2HwpYHzJ*f&vN76{zC?gjyScAO*@4;~7Hm zvJU%MCh`$9L3PWx<&vp4O=f9=oL0p^-T{4^CNf(r63_{!(-=sR`Q0T>-ZDy@KLdO#sV8>s3vc8QCI!~uWEb(M8gGR1PA>M&SZ&IC=1V7 zE-;5axN~ASs8M*Hsy=N*KYG-kR4I|%FqUhodEBEQ(I7Dv?Y2b1^)1N5nEHMg#f0OE|5$*J z6ks^4YzLFeEkxIp{VUN+sN7_aNu8~#_A#38@Wug4!kW;r7vnn!xSwM$u57iZ>~xi@ zgnUhuC4o7>vMcvLSVN=^jaQ7S*>gy%D5^=A;7KNH3AAk&$REG|Z|oH02Kk1eM^jr9 z@D^!5%siiR_Yx~wNW>Xw*wdBH{=Kd0RWPb5HOxh&PCU(WTZ-OuozjYP?!Oi_u}@X| z$+bh?8ZnHMs=X0cu3hVt*q|U?7%u?2&Bn$qiDREh7s@F1sT!gfzp6^x_FC4xB-^RvGy7< z;0&3&zhY~eSvdX_SLouG!mG%8pi^(`C1*4yKQZHM%n%a^WUqzuKb7nWJY&gFwy@zv zEWY#nga_^7v{-pT-A{NH!Q?z(Fu>kVaM~Zy$50f_gIU-@8NDyKjjH4&Pv{7F5BMI8 zOcF0BkONNtr@`IwcG6pr7gy?}D?Q@bUgGX$*+5EJ@2iQF~b#^#>!HDc=l zk+^l5T@=2iwr3DNWIaPly5D2V5!Us-!dMvApkI7HqORX#&Bj5WZ?ybb1^`V7AF(o_ z02>KL>SEHS{kuN-(-PgKv`aolU>9SuTbKMJj6Sh-zNZ1h=FTb0K(;mK$G3D+xIZGA z%HI5!)nj3X`E9`+OJ4&PK@o^WV}7e|XQd7-yb4r-)2}w-8TlBPy#{zKDF%TzuZXNg z8c9%XD5%!$Dv%H>Ayz`{Gu*BRr3oNa&QyDhAbB5?iRf#(9jgZSpOnmm_a8pSdknzgkvXdK{Gt3Z6=ZHqL8c2j=HvOWhdQEn&-U2@7L99 z3*L7Fug1$cn}l{x%Z7Q{Rs%=9J%w15_~<-bMJ;tGP|LVD>#MxJ)Q~84^DP2qG!TbL zAOrP%G{$;L-^gu^tFyWA^(q=&tCVPH#$;*D{(-##7ufDPr)yfC!#<8lyIB1G&btE` zBX(y!9Vnoz!;a{l(na?2my@(X*B)K}5AR}~uI&|Ry_uo_^NCNd&^g_;jX{fvxOls! zwWS`ep=(yNH3E77ZMUST8_Gz9TMVqxh*EPha;H;Wq3+NcOg$Cox4HUzxwLY9c)cHu+rG{((!=P} zC=5T5cQ*OS1{-i-_0{2UxK}Z{rb}eKQe@4}QvzCt4RZ7V#~rw-id2<7FL$wbGlhrgJS`@}_;-#MwTZtY|vm6cRhx2LRDCdl7&^wSmg7OBiiWp%RNlUtOPA3kj8s(iD~RPmSE zH4N?qcc=CZGW^^NmTbOB=HM+Tfx5Z?*RT0bgww}@-p%Ot3hFy|?&5`n=!=4Am$er} zfXxO5VXB5iSdDtu`wlxh+NE!$keiBplURFPoe&>mX>2pAyd&=|JlkkD%ECV>; zY|k_Hq1aY#lY*4!Rr%bKn{{AJ+)vbfz@bet>*uKXl&m$cU1%OzU_&wK@o#_r6-|X7Q0~**l+D$u~;Ys7Pop$%B&PDD%K?kvaC465L)L3#<;)xxp=X zfZ5Zm)M7-aZ&ip>F;7v25y;PuMAbDinYE>|`YJ&NiHX|NNxYonD}4Q8b`?XJa{(@K z3cn^F-tjGu32*hxFiXsK=^LYf6j=kidajb&?27$9)qrKGFV2)|_<)r|ZcOIO@eKdV zdBo@&8HcX9SkubQn@m+dwe~TL1MK?3<~7vI!5Roxm`SAP*{>5iMd*Q}T!dXGU=P@C z^R}1pFJOesdbty$fwl66SH;G|s!sm+{l5$ChyVEfU&Ri@jp)>l_BhknPB?Y$CX3@) z<+?=KW1#EEye9LgXKz>#`?!j>?X1b91)B3{=*y3}`QbD+mv5+O;1+QKayoy9>GI(vZ z!p;a@{Tl2g2}YL=MrbODVmGZ{I0P`LIhBQUPN>l&X#fhF0JO}G3d4zfV_DupojtMKLOg#AIT5PU{5Q_kic}Ez|_|I<{BI^);URM zvQ?Rc@4KgjB(pLknT>Do_?#pg3X4ne+n3zt`$@nvCEy(c@X&z({Er@0HdO%D?k5pu+E|KG z2p|Ip#kBk-jtq+@PnCuj-!BjhU$}T6TuZ2V9uPiv)>V_hv(clmBracV+*b5YO*@UR z_t9mXn9flUsgYg9R7g8NFm|pHoq;ROJ@J_l=9Uopk^y+{xem<-Xo~HFpJ0lvCP=h6Z z&kGV&Nv=Lsgjy-bq6r=0RNjOl*`2Hss4hH;mvdJAn`6+H>}kt#hr{7yteb{yxu!8z z6n!)rPA9HHN28H$8H4f2)C@(R*du!|HsNn{I5nrHHJsQ(?EnlaE9dmTaPjmC0^6&M;j3k;8kIjgOER+8iHWAc+5jc5Efj1(71Srb@QF9TV4|B_C`8vqLndCUQrDw#^mC9F-$)Q!u0bXSVbTb6m;%k^;nFvhIVia z3FD-fcCM<5UD1+ra9S3cb`h!IloRWo0rN#+IDc^wpe2#$EJ!3~#o{u)$4GL(oB%$EjY1>zW_{GV=3Va#5^&iVyCO?SYb!nAIG z{i5l8YZP>9w7ROBO)XwOIQ6EceF$&7bpu4ljo; z?mq&Hm!!u6RyT7E^;ad6EUb!%vPq-M?+?D#PQb=-n>wq6$IHHcQTn<-qpE#kwzY#{ zy#Wf<%-AoWzV-wqVbJ5W8{p*2TMpkEd~DD3i(BOJ}s-~zwBCk-9EGvOo7~T(8AdU1D$7%nbl0_&} z2cM72y2Rz(xcpJ%UafH6N={%9J`hnm!x#tfS??=(>0nzVJ6&-u;C@zekvg(krv7S~ zXJV(Udyyb2K~xs96u>ikFWBLg5Hn`qz+Q4lEzNWhp3=DPxmXNxb)~1P6=t{u(#BR<&c(T0A#OFH?sYiK9Z@&+#FBMH;~wtV%rQDhKQWP zUzt*&Q<0tmo^+#D>ocfPk;QpN$b`#^`Bsx@+EyNE>fBIjjaCt83gnSq9}#G>xMSBBg)p zDgV|}lQh!I8!u*2dlWM^+<;ExP>mZHs&SMas|b7>DEnP-hwjT=xVAN!5sW*ab1;b= z*GWQWvI=YPvUeVAl*9B$DXh(ln09E6wS=;>LfOuQzq=4!6rre%)fAHMc)?--b)7Qd zH8o0Ijk;yMQn#baj+yG~Vx~yl*E&Fw(xqk6)pVch3{=db1V8* z(Kvhm?K{Hwg>gIV&Zq2X!$7dVm`uCC)t&T|fu1!pBX}XxQq&Nrx|031z#y;l0}PzQ zd_Ozd+^a2iQ*!_ug*P|I!;A)8;OOK$DfLo89;hIT{2+SflzcKNw|=Tx2cMEg?Y7tW-Zs9QU_j6Wi8Wei7;eJ_EPG&Bliw4_UIU3 zG%d+Ca_5+Ug%SW@?}s=}0QQv);#owY9|m_Rm;dF5H}E;7zoCW2(`@-0C+PJUza#;Hh|`7C2lm{&4)6>)`!yN%%&#!*5%H(|{m=wK(TI zz?ck~!fX4WMM)coOQ)dv@JpaVDwHwcJJM)}@2%yg-UMEO#@hd}PKjIV(7sP8!6bxf zbG8LV+!p$4NZNu_-E<2&uL(AI9);m2@@`0e(zJ98x!N^tt*e?^a}eE-i2g*77i*O= z79v|&Hx8&+Nr}jy9+O=Pv9KwzJ()O>- z&gvdKp7OxC?c6%ivFV47B z+>b;E5#=`Tc;@*Npi31HOqYkZ+vQ!@s6X4RF*|!@Z5b$<5)^lXB4d_}tGrmmOxEha zw7x||!Y{f@nS=u;jH57)c~IxdhS23N>+qLdsTI>odVB3C6)UCQS)JbZDrv_{SQD66 zbkdQ|Y#B(t5Px>Yw6DmTZ+ktK+=0{7!e_HKfX<*nr;KhfZ+Xgt%PbBo{-Y~ACl^2r z8#M>8>wWSIO6-KhJZE+4MOaQLfPX@SL?LX3IG<}AMrR5mFR(RAUJbqFEZ|z!aHW@2eEu#p301;fL-deu>Y0sVXlw?z$z`3@c%0gF7zGSydsA zEoW!*HUE0CQdfH8Z!Y4zIvOEm8O}n}jl7n*7c9}9alj_HO{d#PGRb|Y7}Nhh@owUfxdt{F!1;UN*eqcKJuP)ctsK}Wcc?K;om)? zzw)dn6PM&cPagE7+LdZos@)sTx)PBlUG3>wwd(9*aW=1P06m-bo)JS{0m^_OTY?iQ zY`Qo-rY%5P`wUu$rj9#eY?0DNp!a64L!4G19+iHDuuWi7K5|(>J_KSujNlP5YUx%i z`swr-#ekiEqp7Q;Q*+y||3oS>YywRqXy08W53UBK3JJbnzizx$*}1Enf;Aq#bU?|f z9lnRAZjG$*WMJ5X33Z3+bTDvLMF+T7jPckW8RN->YSz#-HFaufiZP)2*wl1aH7CRA z^wioQQ|Dz0;TY(z`f136q4k9s?6h4AaVI5Nv9ME%&+XI`4yKmR{zhhIL?O z1Rm1>NFL^d%kU)@@Wan4@EJnhg1VnLPx%Rp=Eu#NW4a#`!8FXh!~4CW{Y2U<-i5v! zW;E?J$t!U`I}O%g|R9^47F_Z%y=%p@EV2B^5mKW~2LbYi%;Tx{DCBHGRuzK4m*%)iSFH`rBl zLZw@M2q>R!qT&k70%YS1t~Nj_X))*h%G-8H{<`c4#w?fobz2=KXIg(SsE6q{UGA>z8bJd5iEqJWS}sm3#UXR0+ln2jdYoh{Xl1?|u%tm^qsZ>>A*@ zgjFRm6;H>rfK6w}6g^1Nm=*pYJ^uLpzY&-3+{GWi|LelXI_^T|C(5)6G(J-TPqFED zcDyI|*d#*qh5kVAJ^Fwbm;_ifi0f?Cxzee#&Eg8n2$#R{d=4d@4Wk%mvh;X(@x!;@ z{e_ieV4y7wp4~rvzh;@cX2wLY6%JhN@Ril?H!v!M&?F-6jfmMX7mMYDuZo4xT(B6% z-U7!5@Rm#XN*^P}Y>%K#LEac%PCP!$c8!~_aYBQ#x9W7;J?tDn0=#1FC&$G?Un^qNLA?i>xMwl=~nfhk0& zqs(Yni|?xvoN5B62P|uvJbL8=+`+21%y_$O1lty%mI!uvG3ixAN+BzEMqDXI z!An;jD`W4(sHV)Y+6h$kPEZ(wgh#RCrfmoupgU$f#;|J(9yQt56}Q@Ayu=lkP15?| zr_9RlrgFt(>?3jvvHB|Zfu+PI?cUcf17%)d=FYid7c1x`GivHWuaBr)`LI!zm+P_| zYL+T9$eWlm{^oA8azw)(_^fdb_l%`oy#Dz8zsEjEDWUNCfBgP0mURu<+JzG(7Xf4~ zOug_H3_0K8hj3w>cj&0su{N70PC5w&M{D31d(41{MFeFK%UcLpe;HVPeCPs-@P?mc@x*D@h=oU9grF;yC8Q87~uu zrTGn^A;DP=I37s$-8pYP zrnyHwrg^D-cVt7Ieev30aD5FeY$SH?p9oYar8hTA>2*bx(l?=H+8!O#{FrjZAe zQbl*Drjr^>YA~t6o~6O`>cLI8wh6;s5cG3DKMREOo&cO$%!A62fcz(XLf)~)NIw8h zUBP^pOsnmH_eXeZhqZ%pFcjv{e=a>7pAhmv3AK-J`Oj3@XGG#Ig~~&>e%2JC^leK! zmjxX8zq=d(zdNE*Q}E{6}E|bi6;2Cl7-4rQU3V- zzt^N5Q&}dISyLiPd<8ROY^x${7f5{CbiVB6fJ1skSK^Lxu=|+&gk22yZ&-Pn7Kqj5 zL0au-!0Qf?>sVR$6s>LCgqR@}I{@(p0cri=ZZjRgI%3Vlz_3|qdT>xvg@a0Xv%Epk z4WUcEgXMxM9W#U$U3J1`;sf?}Ns;<{B@3$Mj;3_G@}3X%SDp=IYV<>DQH$sv&?}(}=I< zR%AQ_y1Hb9{FU`TjQe{}&)USYMI20yRo)>n>?jzgv`K!>wdabSTC zw-kS5TzwrtRZ;>ntXcsww3}!#hJipQQ54M_t8j!6`RFm9@Rxg;mWznhN~f*;YHQlQ zVLGmCWfSeg*Q6EE$#RSxt?-kwX@k!IU(#;<7_8V0K#xx$R>-tE7im;V4?t~3CFi^B zyp@}e%xW~YH+a1g6R~F;Vj>z1vk+GR_yJTYi2faHWHP&5RK4X>w*<)gGR3{|j4YOH zM8k}wN&Yw|`AUeW<}I{GGa1^52M*DN4<+Q z+R!`sbYK5$0>P1xN%kzQ3z^V?uM`{i7Z~>1qa6E%Ih2b4n_*8^8Vxq2&6-}Vw!>O6 zD7(znE%GPcynpwaOoM$*DCJBjr7F^ZItEOY54gc02SNCd2XEoX05=))8ISfFtfiuH zov>ycG;OIsrPIRNk<(_L#KT5SRyvi#?L0G?Bqkosbjv%;=b=IXJ0?EY3w8QINT=V3 z{Z@NA$`K5&d>@$rrqk-&=Xn0)+IaLsr!jF{%|8!73V*uMm;Rf0q11tC)n7-n=j|&0fR1BwMC;YF%&G! ziDKfC2TDIx8Js5WGuI)%XTEmK8FJ38YL)@B)lhn#zK- z_5QF>*Or8=+_|rT$@2($dM54$R~IjQapA*wj6F`4xH}is4g)tDu2kaEX<&JgBG&r? zsIb=Tq5`=6;fHU3>{OZ(wl>2nul*H?b`^o}uRs5S_F%G%0#VGK4#q8?y_P0jmw!s1a3x+wgWPg+4gKPJeJ9@&f)qV`vfxg<>R6p?0MPJlBe^dj;mxX!L<J| zddAB+>!5N^H=s&w?!cZs@M^6GO`0IXD=3Z%Xh)dEqwE-ZxY3P9!c(K+sg0c}K;RIA zn3ZHGitvCHIJ9%&FtlbV86{&2cR&y$%1L`ZX1S|=_&|9T-Q2s2GLoJ+o|qPR)l&!+ z14t%Q*Y|gnLcJrv$-bAn+**bBszJU9KNF*Ga-sxD>Vbpz$o7XdLo*5MOC1xBad+$M zm$LYJtBznnmgu@UeW)%7w7TrYW#)fF!Z`QyzVV|9W_?oGe|xhYgu? z9;|$JMw-RX&p?3#F&E~TP3Zgf9rw(TF}y#>)5K$N6qEKe026etC<33^SQ3d9t)_*l z!52A>y(#^A_2cNT%lQH$4S9;KOGML8($?%kY?MV*y(E(%!m;8KY)@u4I$q%5#}&g5 zTGbv*ntginGkddjjO!Ba+{FHXT_TY1bZAG~f1TdZIr;eQ+rPge!>$Rs^wU3PVUy_w_)E zoY~?-15dqf15es(w_X=zDFf2h2L;PaFqp93SRNhx<9jr^N>~B{RF6C;1BpozlO!fd zOgj3BQLhS11wx%C99qSHN!mPn8jN0*{AuYwGU@HqHJ4`_s^v#XJ}pRqTyY{4yji@Q zFH)X}!iVhJDT?K9`y>*^E}Kf##u)!($({9HkewMjyQcT7MN*vKXcRh_G43Rx^DEHI zNzVp3(v>4zMWn?cD*1+uG2juCY@hzL4ERyCPyUgO?6JuYszzqwAo~EMrEGiMgs577 zkS1~Q=+jy@*VzImwgl>S_(bsv;om02V9e@OA3K@+*Vt3~jr_>(f424P+Co6_a>@ zbLH|0Yc^o1)wk`H+**GbZ-BJ~=WFvW@`7GIP!lUWP{0J90tUchvuq$lQjvtcW#(pu zJ6tyG8hdAEH?__eMSM%LkWBX}ru!5Sv5@(r50rVA#G#l~%k!qZVC*j^v}p4R3#5w! zpV?$+o(qy1D*H!uj)A5!o3ggHE+MRIhV}38?>L_}#krgD$y&yCDu*&9XiG|vaX{vG zfz0QcN1Us}bP%;*+6f(dF-wLiAWQq}7rSqXbjyY#`PuT$w|Q~1ymLlxv6~K({e?u# zrXyxy>ia|3n|f0mu--Pz&9T2+nwX$u7@BVK73QHEX4SSb=0&u>+J#f~ScV(0wt@#4 z?yjD;vyUChW}Q30BP-J`p4WZ!ho8vDh^_Lmn+hQl$lToIe9c!`&<^b$uzJ@Vn)8C9 z=tzU{%8uVVfUVSMNU(8W3L{5`z1cirY82ZrVTY|TjS_F_aV$B$>1SI9n0^H!USNwY zdGy{qdLy~8Hc6R_!X#ZW=gn|KN*t`O)_J;&@Z`)3&(;pfsl;4aS+jlsz7>dbQLe?O z@`+Zmh<5g);YB1m*0`TQO_|Zw5dh<0$pbv331kLZf+{ zcW}nAGxyqo&9*X7Lp0g~U1sSgHR(a(pDhNonk>$T2JZ*0c<(4xqv#x zI9wtm)I9}t(pIYr>R>^^9iBwpBp(tw0WsylF-^6v5Xdyuu3nS^6p%o67?80bX%q12 z$i;FJn2yjb5E};%i~->_5maP=u}>|fbB#k@;cfS5sti{*6_~W1rk??r#8^bJfE1y+ zIWS>%hiWS1Eh3iZlaKCaVD%A7_$fh3mmnoU>S964+Nbk7Tz#hExfC+);}j6OgjHU1 z^98$5l)yku85`dPcfK3!iDcDJe0cS8bQguo#Y&PiOG7EX)yTz;q;Jrc{}%t2uImPm zjB+G)tE>#=F4a3=ZKf%N%0a^u>aKaAn|VRWIbQwDaL2E7an+ zn4(u5QO~o1UQESg<1xkW2|G_w&SBdPnt^tPM7x^ME@dv^%N**sr*mz0V%nV>JLo*aRvV5_?;o8h>@!@C z{TrP-1Or%a;(m6IR&=|Oo%XCk+B4>tp{XnVZS`f?!eNq~KUw)5Fc#OF8*GjlxI1Bu zCF_0R!v?Y` z7SHOUTIukdxGZTLPpZ)!Zu>oaAqgEuO*RjH53p;60qrHQ&KaL#sVXZ|oVU>`n3ODbEtlvt0sAn^FSZ z(u1@I>%#RW_=NIvcNipLaVhN$>Fd;cb_M~ur;pCu8%H60**#^uR<){Iabq{m)W`%j zx(8PTd`f&Ci)aKQ@-4wieauT2&6SX9Q{FBRlv^BvBr}Cu#b?NSiQKtfu&w;R?xrRK z37Aa>Ow(SOjJZuBNL$tWI+^q8%uenY9MJ^rz4GC25>lVnC&jGlnKaKZG3lnU6j78$ zuzkza>Rf};d9g8oB>_|ds07d>1L!K|WJam3@7IT*xK9#H%fWO@3ePGp*NeDm?hT*D z$IRFZ2zg7;JyP5K4fO2trUQnRf*>*%=NSpC^)`yG9t$!oiJ+%D+WTQFPxcZc)3#H< zP==lxr|n^XShPQ3qchxeGps}&{QAXGC1%#&^|Wfcn8L0}_}t3?&A9pq_9tth3Tu&P zehIA-S{rxH9}f3B<`=e1g6tNM-I>OYCvdE(SY7uSfLS;O^876?7>Dznom+*60Fyyn zGIf2Qv|E1oK%*8I$SwSU{V)MqwmPptR3d<&2?^*On}$O$8#*Gboz!Sa1t1lGQ~<~K zNAxVUdY!p?F0e|MJgUicoHREiIu(}{<( z@;p3VvQG#NV+@-p@f{Dot7_x+KyS$oHCdM8aBdfYh3)^-D7#S`T z35*gL8+vFic*aW1#&zD6h}cB;<+(>vmkeH=g^~B$Y8QAmgEa32R`ZN%Y?b0nmdT@I zE}Zc)ap>oSvW|H;!ZliW4)$SmCKTQ{;-sp5?&>~Ms&S8q&ZbhEw#i4a;)LpfGViix z8M+CdCLw$kkg3ye1PDvdWo=#b#Ksaa51OzuJWug}-R~ROP|{7*1yOY$Sz|m(kaMFL z@ROOhl$Z+w%#V`u4tocE3j)Dd>kaIh|HtqDiphufKa%AFpcS9G1ljY+pO*e36U`Ub zH11k8c8r6KP|%%Jc-EVw6v%(N4$7QR*~{kp$C5uWafnB4m>wY@&(RXMkc%^H;7TG8 zZ}2|s92ie$;G?-tgAWF-k8MQ(>s0go?$il1bjdv1-wgkhLljd2XerCDkt%fNZf8b zZV&7=w=pH0YZ+IkAC<^Q2||C|)ot@i+!%wZEn^0D<6HYT4K9o6RuU^ERyHxrX6qYx z32aMtv*>L?rCqfi*VZp023Ez}OWLBD?f{O?biN29kUGLv?9uo&LE(HwT7`DG0hT>| zPFkG?7^sX`KDuUBUcAWExL$FNiV43SWD-wt--F~NXkKM`K3lL3NmjY+%a``;27lx5!5iXW~p}j)5zyLF53{4(50LNR8pPh5L zNLs*DQO^r{Q=k?fCZY5P4RDa)d`_8ecOBGa3=D35Kv5RDgjV; zWlpiM3oyG`py@8RON}wQRoOqW*LlCe4@KIas|I+p{y5=}L37Y_aeuO`bDWWGcGLZFH`aK-1zzF_J zhmFJ1Uk&0BS50A z0#I0iNs%pt`>f7klV>~aXTn$PJIV6Ywj?zJjWz(XNW{ihkQA)cf{!=``*My>rNs4L zb0maYqJRs)^AeGUJM8JoE*tO>Hp!WKQJjD|2s=Ulxx^zo@MW-S^?^ty-q>5XL9#cT ztyg;3t(Y*mB(eJ#la^3fHs@=$njZ2687&_{-dg)*1MAAMz)Y5W)CFdk-XNIRiwXU_ z;4N5_FjMKv1^$+@L_zLLyr+Ns{@?!zd`#XU=YW=<(?5Rym$-l3WgO03mc;B7ne23c zXYatf(jtx`DVqp*nLWtBY6k)yP*i~@a{_DdWA^!!$BA5aaiQOl7w_7;UTo?e;2u@WKSc#9W0y3pEZgsjO4#|gi0JbfB-4-0X zI2FPLBO8by5VTMJj?97;t9QI3??E*f5_K=&jGhS2=Z$cIu~N1y7z^iz*t1s34b8m^ zA96Z2?VugGX`^rSPx|ODL7yQ*i$xUrVQ`l+aDK!DW&jiwsRyt5*_zLLVm-FKDZBOo z7D$!>-0C-84xgcTTGYG^#h>W;5H1++IR4Ca@cy_Yd?VZ8w=E5KZ8`q5fw&-o2`iy5 zO$*q6uw=_kcUz1pHhNTUL-3FNn22*^BP-Le;Q$J-Mb&&?T5%?fx&W;QX&X zQP;fH1>l~0{KQt+dYFZ4V9mNG^DPCo&3E&FVAxae-0XhlJGH|;F4^a0as8tR*Xocc z{h?w>jJ__69`Q{=@WwNxMU3yr=<*tsU-)+0xIj^2|1UTaJXfB~Q3>sIb9+qW(mSXL?m)L~Ck z2-6SQcMe0ICL&J3bt-|U(ZCZ^=3&R~aeT5C&w4letd<=m4O_M_fy`2VR86!FVh0() z%U6U?SWCj+xQJ9hlB#t|WyEWAn5Hyshojb0{an-H8-#!f9ec>fu>(SYJLg8@S@yKu zv7X2U7L19Ay^wZ9CrdyIu*;(;d;pQ@bP@YLyd%Kgz>C7=rYfY3Jo(peKC}uKvp5Ku z`rcxJEv<5299;jSYLeD3@b@^qfHN{@s8%JPV*aL-9X}o5>FnWqHRXEEms$?ZS<6N7 zie2&@5HOs3W0p6{cNu*O*u<+#{yI*=5siw?81=p?dZ-b;Qg*m;KW81jx+|nY4EXS_ zPgpb;UIp%uR8z&{& z=0(O`ugFhnuRJ#hFkuxa&%i9jGI-c%$=qoX&04X$WQG|%(?YdYMCYytpyVcGyo^{{ zq>u66_(k2*9aVepW?ljxoGW!tJ-P?@rRlmS-SV5$J?bIKhi~Y` zrCCemBb85ukdAUz<&$!*R6g}6AG1Gby7Ec4{MM9@IXFc5@C|LS^3j!^u1e)2mCqr{ zM?I(V$v9UkpL&#!)wde1e6lUSHRWSjhbSMup-omkJMyih@{!8t>14LEDW8;crShpq z`PhB4>B=Wd^t>_UW1EL4AHJatZFF^bszF-sw5oQn#T8O8@n8cB9jl$ZDokpjy|vKC zYH|*Yy;VS5O%pAO1PD&>K|*j3GK0IjyW8OI?iSo-ut0Ek2<{GpYjA>V5+qo1Cf|SV z!@W=UrRJrjdUe&R-g{Tq(CC>*=V!zDx{TcBc$Q&;pI1&BSMyO7f(Q- z8PV>u*x_FrP7pVj_z0fbdWU6^Zx8!?Bv#Kx-gLwlip9Mb!`w=IG>r+@h7trbd%j5O z^P5dy0Nm2^<1b-;1PM*l%}*?`RtDSy1k5?%xD^?Ws7p$7!02)uSt0i!zoUp#M3qwj zL|(OCLTPJHRUoclJ(Jg^$4^>l+_j8*&GkOs0lw1ONpH47X&%UFU+5(SbVx5)&!Ki1 z@w4*DE&y=r%#sxlf4G}wJqOfs_({{fU8=TjiF%xAD?>D_FPOiOxQ$;~8b$2N4fx@z zJ@0cQ?TwFvrK*Q<<7izBZ)|y~V4po*#72M8AlB`)zhHc2R3N z;)+z;Uui9{uFfb*a-kTMHNugD&MWu>LSchghFix7;JJAm{+-Do2XBRG8l0CZ_c0n+ zDA&Q#`?aPm>0G+1DVR_VRc!WCws|m)RDTIi>z_G8*m2(WoEHG84sIAzel*fb!w|*TEED6X7O1K|Wh)+OYDcijL0*s3=j` z{S)NmH}6x9&g{`)j?axQ>z%9QlgA)e>{TD(=hWr|s0e6*K$;vj66BFNB3H!?EZ z^wC3G62Jd<@`JCJ7#j`v3TZolj~;t%inxNbkiCCh^nAk)4dl@DE^k|_r@nF+(?!>N ziS`T*3+FfWJxg%m?V^3y+=via%roQ%2}RS7cR|%~V<%}3LPTBz3hy7Xb>bF{LOf5c zy?bv?neY8xEDdTJpX_ya!;l#k9-qgq1W6H^`WxHU$VhkI6X*Tp*uH-BUW<5gKz>|D zLr`3&ekJ}iC0IWG$oH+-9U*!RmV8UffXGMGEHdDU7Ni*ZZBRyjVpgO#p}t7s$xno~ znA7JhYIwSn9cW$cAv>G|UsyI(9*F{oJXjQCvdt2wWUm zj*%cf;E13{CCjAc4$wduj~cYjDJoEieR%)Zk|LqnTt_|+{W4B3sn3NXr0PgMkH{!S zn>Exkp~W=-5qT+hmk4B;eoVYZh%9XN^XmbG*CAVeRPnLxGtpEIF3!%-sm_Ok!_p6C z{m8Z&9MgPxoa+SUnWSS;7T^yv0BiMQERf^$0k7cq6F4LhH5nk$K@?n> z)J7E+JCZ;4U(V?l698l`_VQM*&1u-`0Ex2rKwEP}Ze!|^!*w~8`+&pcZi~~L`d#zz zrJxY{n&(I{Bbwm&WFW@PWF3-Gv|kGQov(i>?xpK5U(*UBv0w?mJ;^x%(r z#0fuc3ex%X4xR(I?DVd`+3X5?_VK-;-RHp^R<*WYA6S*&I4hJxe}A+^t{J4ai9VCJ zi{|{2$0o3o>2;z8T|?~*4&e*Q{G=Bj3L$;#C2YZL=6R9h{JNjN`yF-;C3@`7eyrNP zO2+XBjmY9xZL|716_^4loZl*LcA+@_O;`2Bbmf`G%P>z#$W}VDYm!~{TtA2Jx?wKS zsgKFZKK}La%@q$)P$I`+^sA%zwZz}|5vN!hjR*mKza5&jbX*pbzv}7m(!LSE1HOIM z9`%0NEI`wk=y;7zf1dD;f1dD%zUm;Q=Xk=Y6OOMnyq&tD_SX-&n+hB32^A5U^G_1S z`8ZS3Q)*mc*L-dhCYSE(dTc#Ry1DAe`vd~V4ez0jWb+1%+{v+Me` z)w0LA-Z##$lcjjKOZeSSX3(wJjE{}%>E_ZqSS*~N#$cd#ux5sMaVwMeGb$Ms|5U^A z^*34wTr;Rd_x4#N8_~V_icWdt`n)e|mO257Kq=<`^ zl)gI$2kfv4X6|8s7Vg)c^CN{Gzc0oM(ol_g%QUc0m)?-Z%vNHxIw0W@`C=Q>i>|9@ zud-7|H_-E=Hf>Ld$`kXxTh%ImK^cHrI=euk3f=*IgSWR zzd=XC>B7~C1~uVGC#4njI&I~}nzEwTXkMHZ_xwvpEV;;bSITwA`uvwq>K5L6luVAH z741l$IvsgaDT_A@Ki7!O(o06ve6yG z8OM90x@&f~x73l#LNg3{Z%&uOV@V%eYjX06C_VeI)5`{?&Vw^x!tfWs9qg~xk<6w_ z`FM#RB~$OYYB1kxW*7D7=AGB@H9Wp$(aK-?c)c)VOaAtOCD1(JQ_|^oFPt*gNejK# z4^kp*rg#if)k;Y=|6Z#x1OicN!vtb~JSlrGX=O{_JQ0Le%T~nIC{`|mo`k0eZAT39 zli(=&5LSzHThS3DRj08TaM#}ygVa;@=2H~QwKP-9G>j@D)*mGu%&*?UeKIGY*Ppkg zxnRs(Zm-@1qqD7CB6$DwU@RX8)3JVN?RMQ!YCH~U~ZT>Tt0e`rlN<_PE3aEpL0hdt4H_cui>CJN8ujh7$cjB(Q&AR z8|#ugh&7LGP;^(BlD&-({afFBp(JCBdIHOAerhy?XB1;cu#e|*!h{M==dJPz_lJhw`+cEF8d-Pka z=k_g1b

a%MbZ8o;TENI1BEmMl1B3F6>;M&|dZQmnHUE$x!3o@nU(EuudX(o1Qfb zA@1R=V5|r|;_d1c%E_KsCzUq7>y7$K?W(^2-BX8z2FM)~Gs1jlKkHvXlD1S&jG?zE zXnfM)#nj_Y1V%$GX^rE#MdV%n(K@w zwzfr?zYY@!>7IDU^hH|1)HT^GIzhyhD(B{z-~E9j-gK+MP?12c>l#!@jdqP(XqE8j zqf-gwA|J8Z-ii2pSdq9P%EBy+%zbo$!K}g;F2kTLa9fJoOvrg>JE$y*giQCD!6;d` zWo_g+HeK)QJdZfeIEpZv5(Z*Vov|8<6yL`-jdaNwr6o|#yo~&y)#b0VZe4J8gV9HG znqI^j0%g(~emPMSVF0rwGU}IQk}L9%FQdxDLX&)g{z7n_C0SUJ$3$fZwA?ByBlZ)4 zia$fp8IO!tQ@px2mLrIof3GH@qCNw%C`A8*78H zBZUD9gZ~W=~ za&*py7%e!(uvs=+?5{kAb;30*h7h+!tmf=y?1hYVUj8DSN&^)zX9^GixkeF; zc4LtV`y5@7_>w>-Ql+1+_6Ts2b|=LzJZo|4&bqg4CJsL+1;6hQ&Q;NezZh<^?$3i& zc=)_?ZwRMV4Jr$$>*(yv`|Fdpcp^_t&N^7us(y5e ze4l?Db6JgGVNJ%8gSuN6>ktha3xi*?XOEJ?BoE$?RqKNMXwIZfbB740N@+iGDn)YE z-=-Lf(|vev!NzB?@|Un!u?AS`;x}FUZVwrEFL5zGLTMN;e#Cm;b_F#lf?kDRs6rrr z<@VlVou-o$eU#|_q0=^m6aDdcU!Tu6=Ue0 z{MUawv^PHj+X)n-(SHWo9_1IWYm0amrqZR&o>)~z8sVN=eMlsUl(M4 z7LpHVXpNJ?Cx+Grt7uv!LYSW_@jMRMI9_HYdI!I06?N2A{6*0#-=?a34j3nVh;4~A z2JR|xBw>%KbG%S;h$<0V1~qta4NDzP!?DUokY*~;Oj?(5_}xHkFEYBoy|;MId;T|M zykYnSIXd~fk+mPaZXWbR#8XW%11^L|8I2xDH`_72@gAC+*LCE~or*cUhy6R&nR|mj z(&`o&IgopJ9#~xr9nGwv44lW=ob~@Kj*xl)?uhygJQgD~<8(>G-gzls>|MoQMi9Ym zs*k-+$B@C1#?7TfzZU$k_aJ=A#!Vo?{Y@%buhK;|AhDTs^L)fJ{q(EM#6=zZ_t%DU zTUx|tUT9Q~9oc2IHb}CXrX22yw{J{r9D&y}g zVHRIzFAPo1Kn-~hyyNz!)eBVRAN0IOQrs3rHsD@@#->d-)In)A&VY8B7%oAk52kNuk9QsH8t^tRZ}uOzOO=s&lFKkH#Lps z*P;zdA=oklZdG1OwVsK+l!;f8;H^-<3ho+O`2YPD>MN)J8~)i0p~cSAIIWRX%EwLC zXjt$`MPIV+PlE$eA zoFy_qn00Z85^ZOV**3qGnh66;;_=`Qu@)0AGJYdElYTqYOg-@Niy&1y#s0-2e5Al~ zU%h8zVw0Tnj6*wlZ6`I4*y#ipI6N#skWJ9S<`fo`h| zld|L`PY~VNigGO33iZ9}GS!83Kh18P3gdTo3Mp6-D|$IzY0ffnvqy*{y~~72w|{z4 zFnbd15nuo*7+o+O=l8h*uP?%R!-x9ib?wR!OI401fpS4oxIeB$UB|->Bg0q?PQUxr z#Z8kCb!9JZ>a`!L{=#avz_TsRUIpxFEJ!!u_56yhwYV7a!a+-)$wFdh$G)~Tlo1(+ z^TJ@f5+O{NHo~d-A;%u|W~Wo|o(0iEOjQ8AJcj7qEd&kkAZh0>kFy3k%HL2b!;c>f zJ9KNs*yxB3&ikaitdv@IO2~|Md#27mCvX>}^}1^>vaby}bE6xFJgo&~VvYqK6NkiYbVcyCgOUx72=6b7AOTo75>-rv(> zq+A{7IrMkY264kDvSXvoSEbCIHyxZVL#`Ki$@d`*@%9;!4-A}?&zP3*Tgud?YSMf= zx&Crn2Z(WjR&f_TEx-KZZ8R>lg)83Y9hCHqZX`KgZDf4x$$oo*Y>-zdMtn;M5)Fl? z(s7Mw8o2lUIVQm=pln5|qE&}_<^gec!RSAzkh?xfN)3!JF;HGQA;KL&(v5{}T<> zX$-!7nNH#3L?pL!((F^3uIXy)FGh|qYXKd28+$%A?U?yi#A#y3j|)BRNFxi^hOK*+ z3J!Kb$D7lm^44RB_>e$aY)ofHA;EdF>gH^Ht6(+{rquI&rP=_i>BaIP30x;Awu2J1 zA|-k`j<{l1iWz0`%jZN6{5*v_R^rL;3Z^zprmLTdgaSoymYwk>KMU7$BUrNGNFDxo zWJ;Q~J<-t|I@0UHV~fI+(Ntj7_o~wFimfy}0FeBi+x&$;W9wQCk1-Q*kf5sd#r)EX z8c?R#F1!`0&rA~T3nDuv$lg1}iJ)2z*QyLTs?tj-2!b0*QmX#`SsK}&n`7&7vP4M2E zqdNW3j@0L|7eTL!SG+SV5rFu^XUn}q`B%p9)R#BrrrlZ0otC)f-Lx?o{}=X1@7*Jc zUf*BT%F;@iYW-tzhc8>Ps8@&-oNYRxk(58`PJLFbO5f#dwYv*oFZbxNn=7X{BTNUW zP~>-@;Gmm$sqhr2HebfB_cCjC%1&J^yi*`Czzz7g5X&?~u2CB+aI~DqAhqgly2wF` z^jsxoJsHg&cKP+ZDD+G@GYEIpuM?kjOCirD*i~kU-aVPBfG5Nv+#C+XJy$SuO!Ku( z4faj{8kFC%)7`FRS`sQYoP%2P$mRQs;>K0wlyprutOM@HuUD3K{sBDYx3(I}`GrpO z!v2f*bfp5g`u1Tf!HB)0#?^R23X;WLoNQKXx`8Piuo|XbtCd|fQq%LwEt5f91ONyv z69rKQQmUM$Fh7&Cy8ah8ql)-v69Fb?r?urG&ILybUFgXsi|%&9U`>0>N$*c|hqE_) z5C5w(!XUbl3yFAc<%?ea?d34bGXN3U8qI_OV8;^&#J3iBGt&=w;s7N>pH_UY^ z=8aEBN}(2SDn#aIGhI)p54FnU5EtL*5)8TZ;}6B_H0=C#3*B!R3^BZ#e0qQJd^eR3 z25oYU*;p&@Hk_vIUlee!?Hw38ri^G(4Xn-md)-+gz4w$vB?d`GzqZD*0 z`bZ6px^G4zlbnI~#2Zmvi@PVy{1>*lmo0sN%gBy*`#x+4LqaZKB^31)BZx2&`YhVW zeUmb1V0qGT#fR@YTyAB*zJ!yInc4UEDPN!D1-#r}?!a4jmVHT9`J4UjY>4(9G&1NS z)$IO9er`UFH|CM=^PjN+Oq^KYQe%wSQczODMTVYC(mw-S9Wd(UAbXaB@!5rDxoYE@h!6saXDB@Vov2( z;CY@Yw={B)>uPuxs(2WX1CGX9E+;hebu}Rsvs`9;MLIlnuTaFHHlF)R(%Us8S8Qmx z5LtPw+iE!L38mlI>uk_V2Vt>ipG;jg2lnQn1ryDSzi?cTI@rrAgCSc~1~nh~bEV#CUW=vt5f}U#D$Z zGv9iqnw0vQOd`SGUh=+je>2tZI|y2f$CyNLoVi4-ZXvsqsY%lM+*Yz+^U3*qNxwAJ zS)nSwT2;VKWwFHme5EDMD`+*7wsmU4$bjL3McO<&a8YyjucBum7UnNdT`j(S9S=#9 z^6KP#asp-wlf+Y73039(x7670lD)NyHc48r@d4_Vvas1mS_RrmR%~M3&xh?wbKDiq z@~h_r7*9*Q&oh_ef`YCyS(m3KdSJ8Mq}ACD^;%YEqLs9w9PNTEWB3$JDks+p?WGdb znkdSbi2J4E*m2Ixk^1^fI#ba5$DWjAN7Ofj8A*MAUz@&H%2X88U~L$vspTbi7;9)1 zU+qG`zqa#xxI=I9n)X<*51X0}?)zBl`NrtBc1cQRQR>jA=}V%Y#K;wCSy6Qq>e32I zm43DGMpHzts$6bMZAIE7x72$J@UYZ-rz%Oq5^+AYVW|onPt!0YM=Y)xP*iwPEd?<5 z_lD;vKSx4UV*QK~3~^+`l7$LlmlXM8M))kvYF)(YdaP)PA#UWO{+0`OWWze+O9$1i zs_pg7GKE}m2E>m#b;j|(Pz;gPd-4Q#2c_NiRWg@XHACn}mb5gg^->%#zn18_*D_h8 z)oGPCI693V${5yZG>?Cp>R4Xkp!%o|$fw<>x>ZwIzWk$#V!y7f=~=XiGEtXpY~!|8PTV6q&V}znA-7^b8%~oWfLW-Yr!X zUp8!@T&+p-IZv3aUu<;?2(pkUnM@91t(2|$-C9D$mtt}b+OSw|q}7Vj1T3`Dwn|^tu~u^B z>~HWl>Q^3pOZ7Wux=ma^ppd;@VXY)Of7xc$N_jf`e#M&ZRVL@rTAwT7Z?l0o145?H z>C6b5xjT#Z?9^TXkdjh)MtL~PDXEY2&~i;wz8I9NQB?n7tpt`plVLyHb`Q>St@Fev zt1)--Tqs$Zg5}zRc+UK#?@sW-tM8=x8FHL3we2+k>NC^Q?JgKt?qG=`Ry>q zShL#Ujd#pL64}p)`c~@XrSp~|Wk0^3rvzT@H$_MyzHe%a2$dkL!t}ld<2=7&3mzv@ zMf@Yb*ikuL>A)IdPVrkF@z>W6%RBGdKftrRlG0`2ho zyo=1i{P4K5hE&N0-Vf6RXyFo-Oxnori&tq}ES{v>#h~lddt69lBnC3dk6qa!LcY4# zcUsPc3?nI>Wv0=8ENnSY$#J@MG#Kg>^RGB7^UWEK#CND`8~TfERPFV9XEQlY+3>4! zEfpS7-3M1QmzH6q$~kDt*AK9yeyy3_QcQdU9S4G4O%COZuP&GRwGl1LctqUXhhca` zCE95*=vf6{*?IM0ngy4u7>=iZUZG&9as=djso0n!v9Xpn{2FBRRqS-K)5SfhL`9YE zqpZpvte1Lo9)l~TCw5ir@+zhhg>IGkx3($@4%fZ2z%bI`nlrr3&HwL$2V*?tTVcw1 z#v)%e~J2YZ(SEfbaB%@EKgdu+g|PJDqEwKik1np4 z|6j^tPxv+Rt^2}R2|pGw)a(apTMfX1!gK_uZSHm28z+u4jpMOX^Qtuv7s{b`s z+wd>jtgnj^(vq9mD8gvo-LQlj1ZPO#keSaI!})=_wybIiZJi<6S}5Q89kBvD^6Pt3 zCd!= zx)C>iqL{0nzbIrcru7^%7}9+|Ct7BoC9NbKciooZ!Qyr$^$14RkPP2J_6q*Q8W>;X zw2{NXpkTHc_BUD=nU>j4&rl5lfD2=XbcMoRSde?c*~C#Grdk+f!DaWaq*V$$1YqEW zy)V(=t*BW?wE5+3J5e`HL)Q4dboAls-4oH(@3g5O<;u^df!I}vi_Gg?&DkhOBlv1h zq_lM@8{?8wRTdV0jEwg}uO{{tWRX1&ckx zgvjfXMi-EUnLw;FF=h^Otjwxs!@d|iBVtbtr+>RP@37BWqqwc4!$9|5Gs)SJ~dQR$F&*_&<_TyYXBw4|v)UX|mRt&&SOdtc&Uln+u| zc__Ap25blH$T2QjM~_AN2BkjD1OtG0a(?ac}Az(Y#bDN9s&L6(}6RKvht2 zAipmsdxE3h8Y7w=r`q}u4N~O>6xxy2C_(w+B7gpMy0Rn9c0iUy_GB9pTF~VyYiuH> zAN=S_`~#juSfB&aZB`Ky8<)*Qn41Nm|L^ZJn9%LP)I3Wj zePO_&&na`6WiBkRqI5dyj^1e{J?97Lu=Cks(2 zha3xjmD=xt=_^x^Ned5_B*1NSY}J?4#_6IW%w_k?v6FVa_szX1T2mI>56YQivOqDI zu-Ma|vVfV;)lcGl>)fi&nX*0Gvf6X$;z^5JgBc#9Z)s$GKQx+d*J|!P=a3X2x zvna77r4UX*rYy1tI4_uUsz?d3ev*?!Rq>Py9!dXikLw(qN^6oj0_>$$#$*p9N$AV# zl2S$|e9tZBxYd@0@7vTdB$}gh%&P1(&Mcf3IoH>csvb#lpBc_Az+}>}KDm2UXBH(h zF1=60R&3dRuq2VTI@`{Kl;Cr#a|@>>SjEGTD#A&++~5Kn*+Voc%}yk?=HkuL6o6FV$aqEZ3_=AwW^l=f z+Z8UrtbySSj#)IijB75cY&{PblFOyD;#Jbw)6|>MBzg36hCq*sx)#qiKd)0#TA$v! zE`?(a17I-ksQ~1Xk}?UhkRB)I;TN}Z6ktk9*4vTRqoh7yZxWZz!O+p7q*jUoW{wj( zM8q%aToz%Kn;JACcNT3q#KgN6Q-fCkg+JdyFrg{5oD2Fe0h_K$0bDY%5KjVVF|PR| z3iMb+KmzCLmR3%ZKn$8JWFx+fr(|@Jbq;$YBkEUzK*QBJ}D~yl;wEn@C!qF&= zd>ikGg_2pSk7z);G3357p53lT-Q0!}Ud6?k>rZbFF3JA@6Rj?-T(Eh0y%R9>F>|&l zurOv;K+7N{34pFrRZ52vv$?xU6l=af(y^iR9A09ZS_l?5`h`x zn-v{Ry_8zD83_sLNSs;D)#C^N+Z@@&@86L)7eIW@p&0sfRSP^VMMcC>+^P;!cGLKf zZ!HSkZN+v^r=#4HKZya6|F{SIweBaeEoWZ6U=##BYHO8i`TYqR1K?YOQyYvFF4OK_ z>kZ#GLwTA6rE{2IB6G~p+c&;NheYjtcv;-~DyGCLN?YAYRpnaj$0-Y^iBv|`Bn}Azv!qkHrl2A_*sJd2FVP=FD<&26tD?3il=rhBtZOc$^M_!NI z{qZQaCNX6trbSuVn4G#c?U(l0nnOT*?MY`TB^f3$c0l*9g;ny=sV6ZhoV zKg?iy!7zjA1;dPg)M}Z-Z-*QIo4Tiz2CKbK*CEG!LgrWvQ1tLA8|J65z0ijV7-e;$ z|MSyN2rxg5|04=u$aG{b42$1$VTbKrJ9sLcaGbbWkfY{*o|+8vRI3lQ#4u{O%!(!8 zjA}7~YElCG(EvU545Okz6l5C!`5$*E8f3sxd%WgaY#9wQ)00NP3BwI0V7UDwy1XuN zY=FN;>rG|INfo> zHZdvW6?0YZmO&nwXcFQS6c&?aa*oOH*0(M8WL4taJfjBF`FT-$g~Os zj+}xm6U3#0zK4+7n}T`TMRmrst5k>~?z$0Yi~_0>CyMHj&l0;$s@PrXI*o;D56cqG z!809TRnAt(byyR_Z2iZ}U zMhAeYH4RH!fyy~XsxHXUmR41mlr5)PlE?)OO@rh?(_~@;VTphx`U;g;@99Ee zy*3Lg1{GNkmN${h?Wr^qPj&^&i#D#fD5*O@pc1YS@GE_I82}$kW(5>JXx32%X|$PZ zYlcQP-Y0TJD-^HSLA0yWPteAla$`P{x6BlpitS)LStWKmBWo!T1Og@Civ{+E9BN8| zsXh2B#ja(QZgg9G&HifNd7v z-eYrs(YnFgOe!99Fbxs}K?^lS4`8SqIOd5BH254Sn()eQaWEH0v|iL$e5oR_-e95m zqUxqnG-~$6`Dpj-?d{C2AyaH7)3cRo3(FRd#JQEpIt5r$S&n<9D)z~$VqJS@OZML8 zE*~5-3ZaH&Uh{?{AR)>B7zN-npz}cI)_4yBbwd=WBXJfC43z#;28c@2=W3NhLt^cs zf}8+kAWN)qMUET_O8`pD7kbjKGnKCMwdGiLB>)q&8RrR-w|Mc4_-$bNhK&MX3-&$0 zsLJFgv3^33_S6{v#avHGGWjcxC0->DD48K;C=@`ZD7CC^c??siRKWWC8Xt9D#9w^l7I6lL7z3rDcrg919P^3+xY+`eakvjXC; zIbj{vp^z=n`;$BfO$-#UnJ#vWffh}Do+q}~V0s*H=ndta&FnGD-0)j=I*xZ}bGLdx zTPt?Aayu<_M@E8;KF8<&Js()1e|D0WB5@r5PMpE=X%WQ?6e)GAsBx&+32UmBeF23m zEk{5h5*1*4CV&UDFlyo=??;7M=2@QFo^$?ni|dMBb^2mYu=zX`u>5^R24CDNBl#_M zEs%j=jz2tX&&R-8m7Qgqe~}1l!D}G}t+bm*5)9VL2Rg32Q={uBS1o3l=W%z@dh=xE z!a_i+kP!%E(968aRkLWWZ+x}#`|ElOjc=c2I?Yw9ZF^FtnsC0$hg}8!&vyE(6I8;YqwCcDGKBrBt7;4%oH#L%ccIk$d#}4;UzV|39qptkK z_Q!AB#oruqAyLWSxbdQ++W_fku+{@EEsj72;!NkJ>*>pyUKVc;YHZSwf8Mq%J& zfS@Nuu2F+v2Yg$z8rN+G5EP^ZEg`k0BZ`Dx<;fW-o#gRF$d4AL!t}Aro`wR%F`G4y z<4*;wJ&r%&%N^!xZ1XqU9LL)>yIbLkUqNN41Yt&|30p9XbuRy7o#S}G-Q^~agFfMk z4eOcTZoKXCPH(PUZ*O2K?S>UrU99Lg4CT@<;L0XPWZJ0(N8owv2*5czqS2wJz4uES z#0ot#U)C^afHMKt2q>f4P-u^QhkBN_Zv?DRA-u}rIQV7N;v%m%u_55KptHzjp&z(C zt^9%MuPboz!#r2`Gsj7e*p>?z)|`~^A_riHY}wTSw=zoNC?~hlgvuj=zkzW|Y*^-cFN= zlUt~b6-4 zA7ktrFg(ocM*%<}_FD#lwB2csx3s;6)?&P))|N0Tt;v?K5}qRvMMt*K5qV5qtyOcK zJG{#Bv<>1pKgeH`HbGLn#yCsRIs-Cj+Dwo6=y57qJfZ zYP6{yqi=wsZtM7Fz$)isS;{qvz7Xd% z^Hsac%1dJWItLOJT?g|jSmObgZX=ji5nkuG9>8nv1;QnnTj(*ZXnDdKClxA-Pl!bD zE8!XahmHOJ#fIx2Hg%99o5uXb@9`N64D?bz<0sUOfC-r_YBU|R3VF1!C<}9{o&TI_ z1Qy;3SvGkNM$2a!=HP7|q(X8%?rvcdeCCFQtACb5GtU#A%XyM3w&wzdF&Y`+MJ^x~ z*^=KZU)9)0TiW28(}7~K7R++M-cZRbEPZR3U3wzd0%f-O!>dXYS0H~Cct~oaO zd|$;8YJt3v*s9XW6_}qZv_os+2#fM*Q)?pcCZA&Ut#erQyr^6s38p~uuNIapTfmww^t(4yF-`Iq<7xsHgCd7gJfVdyx()7D z3?wC25j4kK&D3lLS9!KA@-cNGn@qI59-&a5r)u^@qjK)Wk(N82~}fiX0F-Y_Ee`;UmP|7SD) z|A=@~;9zVUP}yB7%2l4ht*^5vaA0tm4|$WiBLF-8^#H#zhHw9O{1bp3{|2^U7WrYP zvGpSchkbo%>g7EYq+cqjBkGiaYr=n$4+DK0a0@%)T>m@b_*q9GlmR(wQsiuCX(GJ~ z=LPR`cWtR6=6Fv5p1*_YB-m~#g; z<`cdm&JMb*`#R^Vs;GcEtu`vHV01iqWlR8}ReCaK(pvtm#mE+TBN4{ zp?=mRvcvrMixcwc>IZwE?Uv;Vu|B<<9G4XaNC;mkX)5!Q>1UQpX@2iXeM>_!#!gr8 zb9-N2)Woiv_d4<5;~y%+t>~oB%g4E{F!eNh&us=hT!rgioW&1`{|q!YuToar;9*lt zSD&9(f7XV2TCG)IGk^UPefsz1?(t!dHR9*f^5d_#<|&T5X&L~eu#Ih=dG zLw6Th0^$HTWK3-vrP{gCb!=jNzZk1Q=XHZJI~^NjQ~40vaTZcVi$S0P0 ziuH$s1@xAJm<1oLRaTG-yKiR6EaP8>iDV|;-*8y^EOQ(Bh40ukhY)6>cKC}k>byd& z;LF}*M@=`uvSBgUh42-13(Z2c9YT>gKX!{Y4oqO|8ou6~)u4Py{jR804OC_C;6HbtG|9*OH<0h2 z^8TMSwTBS+=qexH){{uBmE{)~J9zHW3#~w#$NMTu^M=y_he0aLxEAv`P`~SGxyFMo zc)yd-_Jf&z7J=JjyX9R`=DumRaGh~voaZZFFg<(F(;id67q{5jgAeE4I_>Q`dw8Dr zKc!yJdz+JgHp49kQS}Oo*!;W)_Fq2EwpX7A@pq#BZTMT^zY*4$IzkB=o%)C+0ucUO z{pT0pkFtlc?#!eQeT11;=U|tw_qN-wC6CZ6A>%=5laH?j)#Mnghv&2$Q)!;8l;U zXeWj2n^%ts@Z#q;t2cR0a-_9QX9&i^!3FTZm8V#;ksiPv?X4-yBe6dyM`fa{&#bfY zRAbi#VZ6()9qjMEIQ}_<6R0{tf=$<8x$4ud({9c7d{8_p$4T(YdFHL!B1*aG}G zCbRku(HJ{~C^7N9#`$3TF4TOg0@GDoX{I%A<_&n8)$B@RV`xw?B1cqloTJivph2Fo|b+7G&IH;(bqdITO)kI z78d;a9h+kFaj5!FZmn^5x6kX-%VS@6K94YV@E>N*YV1G6e_m&fu;ZJbd?7_xi8#e22;F^WVqd+CO8#e|JWC0w1Eg zzt;L2-9`UE=+&#G{_S%IdJ}MWw!c1C#BLl}UCqRNsq^BpobO^9OV* zT9K$tkvONHAi>unlK0xJPw#@~)<=3|T;q{6Estxj8&t(=%16Y~hbhtHF1LAWLuOrh z(}j1IXXRY_bAx$iLZ)EUhCePcOf4|Q_7Hlv9N)(dRz zogP9t;_Obb-a`~tjLe1H936y=e*1$+Xv}VU+tqu19PR4)`!V;%_2pvb5_THQ->NI< z>A2c!@B{P?PUV^;^k4V+2WAD)w~)U&k^ND`qI0W;#7wjjWxu6+&x_~#kuC_CF4Njv zL;;g$Cw(PK2$eBkbK^`Z`zr)!V6pFY({*=Zy(1H6UJZ!pE_f3n`Qb-)^2xFIzQ%_} zLELA(D(@7}K|Z>@K|iUc#)-h?#}rt{T#0?J+b-AN?vU*9S}VFIJlcLq_8;Tf6~l&# zFJy`=v%^7|XOYYKKw|(!nU-Tq3k{Lkz7^CVA;wDJX^Z!zcRU2F+oUDiVUQ62+2k^(YC{xq{XcBI zbzBu+)ILf`cS}h(NH<6~BA^@&DR2l0>28oN0V(MYIm98Q`w-G_KvKFJK@o&Ie0}eG z@9%!@U-QSTy=L})*0a_#duI0b?WSL{xK;Yx(OrIRi&n~*5TM?E8_Xiv^2)bHo&AI3 zRRW^ReQJaLG*^uLP^+ss>6gsm_e)XCuc)oP`dv5c2W43Z3@4eTO|c(2jJ+Q(0Rs2U znmepCn+U@oC#ki@dN$$L>5F#YU4R(2~RPV;9=MJ|G`j!4t*6ns9 zvhL#8tLxX31A-dI$b{NXLFnLi{pI0OECt1;klPy-oxaGoxLtKw{K-1Bd;D*LiSwlb zwa2|p$@00;My~h~e76$MW#L{q3hyK1))hKsC~4^q&R^#fvP%V$j(cm7v2%y_UY)5KaGB?&z z?${S(j6Zguw#XM9m(0Lvv#qw>N%qIqkEc!kwbr(MY~9a3l&;cg7d~&0L!70;(!QEMoK1Rm*&2j`44yZKJXqZphK9~4=8+02KnQ3Jcy87bx`nVj~!c~O&*t*Qz zyM4tkukUkfuC+^7exaXAFS>dLxZEma&^*tjT`?eUi7%wC8bJ0&i z{LorNcace*oy1S7J{D*1Y1QeEa26Evft-1k{pRJ;n09hkm8b9wejA^#$U{H*b@e7b zwqj`!-8?68cH%#Ik6#?dF(a`Fu1kywOI@7r>F|065w|{hU$hs&t@+|tAtBfHYM;`( zwP#=Tll=sry+Hnzwz>L8Og%$jL4t8T+CaS|nsa@%4If4$_^V0B9YTJy94jAo;Q&L; z`H^FCzbSMVdoOtV_ogD?@6pfeJH%n-A>Y{g&z%oiPE@aBbw@uKxbYu;I%_fb>vGw$ za2^w_6>XxgXXn#&i}p;tba*M)%)viu3fN-(jpbx4vig|3E!-_RSXYPrJ)Q93OC4-s z-pdQ+qn8X$`4=djiv#0vyQCpGh2Ix`*GX{1T9oB9N=8dpFzd^i8FI$IrUdWA5$i}i8{oAmQ*IAL-;=_XFdkMDVM5Aj z$-mrRIE-DYPp>>4+_iSijJG`4lD(^oL}5QicTT7Zx-?(=)tNdI8UD#0i8xs3kV6Ei zLMpL|Z6SMwrivDixhNj_eVL2FOh_B5Pm|~ z>M#0bLUR1x5S6tlXyitB(S?*DT^!EqY$I#Yni~aIQsuxPdhrh=ZuG=aLZ|!UgcV{* zB5Zr*Ot3r0X9;st|6M>SQo+lN3@m(zELHnOs&X#!?2n4Xo9-_=uXc%sF>2MHP3DyO zjURWYwD|t{bF?7S`_p^zp~V^bPC)Zs)J7(va%1P`hJf*(zxm76yc*^DSqckdBh&fy z4amBpPNp3aL8jmMWl$D(s~I*$T~lx=hLTgTk88@Nl{>US%x-za>{~jK)ZJ?JBjnt zFvV0bm-R_U%#6KXX5yCLz{X`gt__ZwwB^Hj@RLn~pnl#-^oFsRCUF8*l=lNuP5jp$ zt4AlMJoQ@k_ZfwvJIIR{w(c0d)Z{EybpOUprgea~^9G z@SzwZ%@?j^o&zgW>CqT3T_(sM4*a5aSxsu1gBG z%%RBncAj-~sLVytXET1VKRtKuc`X~*m^QXKQmzr}8;ic+@W$p&!3v>$)e>2hR6=hhUtGv^nDP zX7$ZXe!NrVX2->*Ll57Tvc$pGS6RPeFkY~=ZJbdw*CE}pe2-7ucPi26cl0!E>1Qkd ziyW*C#S@jlUKJ+7!yaUh;Jw!$ab+IaP@Nk&D5{+g~r4#~=m_d(?IyDK| zEM5iwJ}CPJ5fcjdyYchOM~C2ZYu6_>AH(YNf+lU;-OlgE7QXR450PT}0@__y_a`^~ zEkNk-JDK;7G4F(FINchnGq2)}`OuDa$qFCrU66&PuuV2uqBw=XLt`E;$Xwt%NL|t< z<5l@v1fjwEZ2IF8)hew`UH})C)$tCS#EEKN#~@6)yI7k+<$97ue1Fjd|K}6BpY^aC zRK$F6^O4_OBE5%y=tVMg$*&?ni5%9!J`TsR_lH#JI1=S{9MYWfUph#htcac6h-I;Y zIL4l>pFMwvo;jvaPdTv|YCT}vB(0$#Z^uyHbqf41#Y@q%(5BV*jCZ_gbM-xX;K!Xw zTg%joBcgct#yXuVesr?f@(`@s$)0{Cq6G@!|{cR%01CijA&Mg0jU z#+!)4bqlJe*Z$J4SF`U)tHT2oop|N*Ctke`n*>DGq}E&Gyu63YyYKYBt}*D?e$DVP zE;+3dURDPh+ET9`WhELv_*HM;6{0up6wWp8W+hvD`Qp!imzm%#s*lU;&@%_6ta=Y} zhSc76ttq%%=Qb~#P!7JNUy(?X9v9Q9mueME2&P`UR&~xg=Vn@KVqX^7rLtlRbA@xI z#SCkhkS$S3ow7IkxI^e$-o<{BYQ0RLryjaiwahvlQc(Tt;C>N zeq*BNV1IhHIV_UssLA{b-ALeDk>p>@))2>!3eDoL#6ulpr{DW+@W)jrjyz=%Hez}S z*q_Ws}r8{N;^rwfKs zWG!Pw)TqPgMxd@EX9@4AvY=YG`+y$XwzQiDu@#t)%sL(05bZ(#=H$& z0GD~Gn+DCGAVeIwp02u@_g=iSZS1^ssWRt?AYFawMfmCX^9L8Id!+}p`}@1u&OlFJ zzx#tLgh$5RecvYCg$&_`i}WARdv|Hv@4R$1FDR>1+Hx(IbRLw)_2K^Qtv;6rz?_of zRoPj*<3vy=o)AvG2Ep3B>&=?O@1?V%*ES+WmvXe_w^i4&E5_xw%>K2Ypk@DBl*xN{ z55H^I4vLDuD{rxCk30Vq&}X$r{)jte2(O|uy?XudlD(%(<`@q?F<#@uYib;Kd7kuM zii-d4K=Dbf4N!)BeC8&83%;o_%xz(FOjwt$E=qQ`RMO80~g-eZ1npTQm>M zdmftmkLAcD!bT-;&+l5Dkd`|S-2e71PfW}onm@((^3ZnQ8yGv&ikz8&-+9Q^bct{_ zk2rO$|MU0y;BCO~!KM1x!@Ti8O1Mn>$HU6Arn#9|()Esr!~7uBn%?>QOV(Sms^0XQ z6DygPs-U?sg0FifGA+_o!`%nQ;0~K)9EX*0g33TFpF$HA!#{yL18Qr zOi0nsqqJN`e{$EKGwMv_Sk+_#rZDJ9HW2*O@v$lT=0d-~^UulR=8^lI+s9Z|S1ylT z7mSiFnXaSVEwuIBU-)}h?^W+_!y$H8D)YI)i^d~Y`b)L zMTr0En+d@>y371uJb9>#Msjl=CA#S~M8+7-FtZ{F6XnwALvz|z85~~W?OwS7MmdGj ziR<+)jfSqWGg!p7siN$v!;31AF!nT{2+xscQ0J(i>2n15c=bX*eZob9f_n73nQ2_- z+tAKSC%&@MmL~yKC&kz#O&41B7x0zlvAgPuOZ{VEwjb|ef5sOMl{(REFve~>jpzQI zq}#elMJuN{=_Tk$`Mt{HW-;Gc%DYpzwJI*KmR4VfxKXwZv3qY8ey0mR#0i= zgm<-+e^ck_;OWNpH);pIAI&{95&Q0fzGa7LS%Io~d8?*5_~`^q={@r*Ylluf^9#Q- z`EiuHQ)xBkl98-*R?YBH6r8UNV|^d2_qBXxH;U*YGlBsmfJ=E)ttQn!Or_sX#7 zl=o~BtK4WpJTr&bHg)#+4+7-d@uhdLMs}*bqt!yO!o`z^>u1R>Lbo?o{m8O;Sby3+ zJGjVX+jjZ~wswN#{wkFRi9PvC^&|&@CY-avm~5w%W4)ezX;@#sV?tg&D|Ms&>-Ep- zMO1sKlh4L4pMa=A54&P(1T}q7*R}(nUqqOGZ^D4*El<3e8S}LWuiu^tQF~OZH-{mg z4nuyB8DQ~8L3<}|kM1tWGXiKv15roEBNl#*dJ?(lvE?)`K4l9_wB(VIPQu!=9xWtT zKfPUbL}l+3Y;ka=2~YN$T%T@?IX{Rc^9AUV66I7L|$X-Xk5qa_5of|^W z#Zg;dIP!IQWq$RDu0C?(x5`T2<<@Z=jt~A! z{#<+}cxN7^pQ>e=AgZv7nDQ4hdQ4|P=!ZKoDnwNqE~Q88eKDCLb8b!9Q;#p?DhRi< z>EmTdG1he}F`P0!#`o#mwj(B^(U1IEk2^>iBa6DlJPKRlp!JnHp8o7dB*<|0MiHql4DvSKsfViBs>UF2vgG0~4VlX|vi z9LL5OdUy~eOE6EyOw+aMffo9FxAj2nOZ-t(?}J8VxUOjnVEaDUL*iuNOQjfZc{LKa zyjpg_?~Q)G1KAfYVW|{3oKw}OHeuSy@iicil1v?wmkPr{>dYRAb5g0SQ z9GtFoPxzkHTunZ&M_rtg5OiV=RdjSV5gvT~wR){}#MLCEk!+~icQyHh`PWIn1$0=| z*6x!@W6z3Mq~;kbO4#w2=NG-6XiJEm#cJ6DDtG-p(}sgMkvT%HK6SNk)h|}`m~N+B zNLy~Fq6ydAdKsyH4wSl&c%j|R(YI5beVf&JIQ!1$M_tq2RQh6FkLfo6S_uFeW{u%e zcEs5ilXWuh){JxYk0aL8%2yM84A;}DkTiGGebCnEjNC-Cpo&Y6pGTD6=$UCqtyQ}| zy@T{d<+GO7OJ(1|#biq-r8<{p#$!Ed&D_%Dl{xA%tZjtc@=Evx_DE$~ihemNxw9#r zN^7~GVe(AGkR)%)^Qv~cts#)mc#BHYMlgN51mmcJHyYVs4TWh)9u8HFZ-P3HgiZ1F zoJL8wb3?=V7rbia&D_K9QJ10T*kR%4*clyz-e{w91?^OzYT6|8SBC1FsmbbkmF?4G z>@~pGeRQ8wv`EG)(OevyXDL;3Dm_ezRNEUaM{viv)+LTvaMDVh*h^Yd^h=4h5klb( zn|h>lVfskVgK^fh!u63(>Ut}4o77_x8psSen53v7_C%(QJqzADSGT9nCV=ooPIIo*r*q~8W}__NX3yZ_TPIwD$i3cr6h7wdq2~wA@%r>0XmBn; zPaF;VjD=4Z)3#dAtW3XqTN&;%22 z{du=-M@kU~P;C%*lMK^n$Q!K}1ZfxeaIS7(!9m9LKJ|5-<9e#CqPH*>i;=f59!Y2A zhp){zKgj&`U*1$}qow>Z!NX|Pv}t50P_4C8x0A3&me)3-@5!WBHx za0Z!%s*xtPE%oiAh>4c&liV3c5n8-FZFUh~KDkT`9tiP$uMP%(ukK#BbAgj-7-xZl$rr9AxNVhK#PW<`G!jD_id&B{pBE=^Oiwkr>Snp7^-N zv`;c=lv~{YZNnwHvAou5mr&+x(&Lf3+lpeJ?q!14u9ri|>`NdcZqJ>yXsjRaEIPtK z(e4uLY=TfSePpJD@jRKTNywqy_t>3r%pP^|Lt&q}tphg`;W|3$c*bUZj#LI?eXs&* z=l4%qt<|zWu#NRm1Jt6mur}j#7O5a%H4`2P%z!eTt;*^R+=5oS_fAw_DkJpt*Y>48F5Zgo62S_{q2$)5WI;O&U&0OYm zU8M-&vyW~kuR}J2^8x5a6}-^!2ATG%yC=4RWCCgw`g>h=@x1@2oqv`0xXsS8%~#-T zA*VCCd#w!xvfhUB?)1_Gxk2xDHTuqNxvQuF6_vJC%*59x%_a2 zJ%r{OvRnZiP3C+NmmrCQ2H_FO={dV^P2`7q)cNs)G9(SrYxx~^_~nzPa(=009Viaz z^QKqbS2k174oe%xfjNG ztxH9eW_V>#W6TxbFR8P4_hN%xUbUt+6x|QTl9OVJ+)1Wdv&`O6I1WzBL=&%JLF~4> zR*BWHs%=TuloHANAQn4?QuOh-YGh|hQuI;fU$UwBnTNZ@3{kd}Uwk1l)WR^?o)$1fTHv2uzdQw5IC6r}xQJKOrF@ww3qsAQbg1Xb-*k z!m&+u5PUIGN%XrvB87jBvnj#Bc2ke*Ngl47NKe^$x0i=vT;wS>Rf5Wy33BeYo9c)F zyNWkuCFX~X#LBh$8n@L{3zZ2NFP6N?RM0oc*l{+=wEgnVvCFG^y@j>sJyu~lreb6A zzmy;B;`OUpHdB`X%x*gOrOew66NlpJ$HfkftIpL?;TqI6-h8G$^J?4Nl0{Y5hOMn} z`pZ%0oG}bWK4ch}+hiFyJE1MFfnyu=;tNLgbPmZ;d(xcIz=$)?NTrCXB<;g;TTAq^ zSvx>EKGPoWRGW5ya-ybJ?PQHu)r;NGJw;Y98ZIa`gcP0YDHMO+4DAeLR=WX1&FDz| ztMInUrY=P)zrD66#%LdBHC}$_7n%yCym>ea|5_~1PH`{x&nj7S;mLV73TykGj&V6Z zd!=yvO65Jtc*a})P?s=gaAVU9ZlqoxKG&K%Mvr1sV8ZV$dQ}( zgX)#p_LzG_QI9%1@A#CuzcymkS3QAt7oQ9YQ89zOWM+qG~h!P|CNVY}w zaSetdn|=w#qerAkR!I9yn8R4&erotZsgMud*Xo5q(f#=_6w3k_itXtPIHW=E9Ta^} z*Cqv|i&=q3&1jpOJ0u|tT26p-YO&lFgbYg+zyQ3S?dG_wwiw2zBFClAfp5&O4&ShG z8`7v3;~ zRHsJGhLNP_rW!tk%~&awMCRPqKJ-PUdjAOrOLk*g3UiG;iBTi+Dtm1_PyX4!v>n}-47eX;d3KP+|1s%O8RA&RZ?z(O65r^Jpb!YR2%(f zohX*(od9rxZvo&kYT$0595;CFJ{|6%Wk$bV;||x4m{HNUB^O~gc5;lLx^35;VV{!T2zCkNr**TEo@=(IfKT!J{p@O zr`agout{t#2-4HrQQIOiOyiI#J7(t~u!!d3RtaAW^ZuVWP!*C`3}bqiO9Y{zaPu~9 zkzSuN7m^1@5kggmmRuI3whh%@RJU14R<8sFi_>~E4?Ey_9Ud8)c4FdV*e_Y%N=>yDzt}>JjG4d*O@Nc9o^GAa?YT%Bs2M zI1ih4oFYyh$@iobUPi)h3Sw0IN7TJ8%85p_BuD1Wp>CT*N1*&YSI6674$ALF`OPUGryWO?Aj`M=SI^3&%dl?gkPGG{A?O3=A1pC zvRYi#tgt=*0XH+`;{d$nZC0QiKohqbp+)lEp$JAo?q0t5DGxQ}h#+06)X80a< zVWrd*nYU5M-WQwaUdj=TNr+5KFqgP(rQqAfo>g1Ima;Fz$ntS|lzt5G(5Hva17I@` zJPU3r>X=3w#9>MtiG!+6xCOUe$(VPtQec(~DPubrtootrd$yJz3~7z39W2yD>0-li zr`xZoZfjKVG}h2oG1EHfM-*uZrZC7YYgQW~427e-rg#m(qZiIe8Ru-8`!x})XJ;S` zVj9X$@Vyq_Q8dqmaI#y(c~%!qJEPA87_8Z ze3b#i-_+tA1@l3Kt>E%JwZW9J=yavbtQYI)e5)|Qy04V6RQ$%uOgYD2;{-z~4>QaN z1j#LFKbpZ&lqiHCc(-#2N`uPO*a}V!>Ak#;a68LWUx|x_fbQraz{JKJj^c1C3_))4 z$bi8==;t)+;>f`{D^uuAYA9n31t2dihSTNPP{S$h)z~#?KEV0+?8d;j&j3fsA1E~r z6TXwc+;ba$wF(HPIrPT~k1;I80aZOt~kUPrX?Fk`xM$^m^3mW%m9Gg&&-Z$(($BZP#8E>AZ5*dE1AC9O&XuEtYUf;|QvBlONxdJ$4YY7Vz^%%ZY}MvQ^Y4lJy>#WP_vSG4)2 ze!Eo^nz5NMOD=2ZaQSW)tnjNDTw-XgPznZ zm?hH$@Kz!DuT(Glf1rW>576wvwzuQ8Y052u+i*i|7k&^^?BTn7H;Q=|`AG{}lKp~# z(27l>2C(d2s38=&WPES@UD~*9h2K0eZ}os(Sk+8Y`hp=;OyC`?+u$7xY00k<9Q>Zv z5PCkVH>ga+WnI&tcf>hoF`U)_o&jGpf-7VT1`LTuux;{P&|3|8(l9@At_oKm5CDj)8hKXwfpc>bPnCv^4tcfhE{)lv%XoT`(!O%a zqC$C#2B>V?G&-(?kjB9`Q80S=2ni&lnJfzCyR(?^M`hfwq)`WM^b!#bTQmb!`eXGa zLGFzOn6^F;N^$7i@yIk&sQIf*OPis@+T_&fxjpPgaJ++tijL(E;sk*(Qk-w^CMk?& zY&po!j6RTRye|f?Ug29feUC&-uhEVvSxmB68S6dS?9U9tWCOWL&N^ENAN8Z?Qn*>* z(df}qpZ;mlm)K0vYNAMWvk;aa-7mLkCTH0QB{`%Q&0OzCf#8S(@cCwgrQMZO;C92S z8oa7cKM6_9GY{CIesg*L8LWkd%{BL!D9La@&{8P49G$$@qUsk^l6cXlNXp25VPwM) zYVBEE#?&$RA-R-NF3v-b%F{ZB%}~028NzlCzQk1!!6Ft^9iGN#1}>tVlQ;o z;t)^rfgC3#oE)dTLq8U6cCKJR*PbVh1+;2ZChT)jYRZ#lryeUDZAHKXV$^;IOAD)m zbG*wAt=5ZWLGy0^O0Gd*$UT>nIFjsaNTra*HDj@_+yMSW0|4z3JZTZqn*Y395&;}F z>|y9zR8t&8tPW|#-;EAisCYq4wK>d*V~dKeXBLeZdfu?LfsqRfyhH?o( zgzLZw^+TOJG-a5geuci7d$R|(3-8t;UwD%6 zI*g{W<-OeyY%vuQQt@pVhMst!rDkCcMU1GjrJBA;Kwt_K=}fR#Gd>@J;k(vTx&lm# zqrr)*QcVz+)%WZTZ)y8=!F8?dvh4^}E~gnt*4P$%RSqOb5brEl8`4<(m>|QbG-?pw z)kOSQk%A6F2L~77G)LulVak2!S@29;K3szJyFxtL+rb)r05Xj)usI>JWrLT@xPIF? zTLUZLu}Ke!iyQZNfe*Ae`QfxU4IL8+Xp10=LBms%PpmB2O@=YGZ-|>rCDR-S6UYWO zE<|fyhx8^iGN<5;%WcxQH+piTe57cGe9j#({hk1wn)cB=iA+^`>w3S5P;aFS10w%O~;A3;LKQ6v$`LEX@q}1%gH=CJ|yzi-!d9li(Rd39(Ov#6JO5yV0P}fXkj}6mX?L z8s*p6V;xWiIph-%%g=z8Aps3eo(u_rT@3-@rCVkIw&3zDN+yJIr9&F!w#CvkIoU44Xb$-tujCWZ z#6hoJxo1&uhG1|W`xS8o;EzN@iK-m#=Yw`u3KD=E!P^&O8iFNYT_-~1=Fc%*xwYNp z#OG9J_%ys$lij(avd^qt#6eu{@yahAY{PV{6CubAiI9+{Un4NtC5iv;*wvho4HOZd zH1gYb8(hjEKBvX)NgxVheDxKUW?47D|BmKE^;|ql$R+^3C#mA6;|4O221~ezI%o%8 zSG8AICcbr8SoDNp5XCdZ#7s09g4~Al4TiqMxsqflFrrFQYCnPa3h?|rsqAz6%`G#y z-Ny<(NKag<4IIUZix0A}{bOK{j)ERR$pJjOPkX>W+}K zElHlEzTg~PB^!YmGB~FwiG*yDG7LVJHCJfF_G4Ld<(3?QLCKgurxF?ZI;E#V7C>9z zwxW}ZZLp4UA|OPp1-BlBi1?!rp|NjY(wG*fos-l}f>;~1&{U@Sbnyh0uP(+bxMF&y4qSNYRlWIXD7WCy`9g(AT|ERl({7=;m)e7)pESd<+iu z!1Ls8ez7+S7poZ}!72il6inF86vvhxw)ogLIYVkb*sDv01V{H+Oh7}KHyfln#g&}m zAe%Fqwo{W=`!bn;7+G0ok3DMZbBU>i2sgjPlLYm7a`SocOQsnu!Q6RSXSa$|)9&g# zx##0(4^PU!NpFE?8t4ZlhB~Z=V8P$zUAgI-fWAxs+Nq>OWKM0rg!t%@Z^+xW+%50B zqQ%zX5^>ZW&XAsXL<%)dNjHikbc_c=wnh)UIuqltjBSH0wlGKmDbf-1l_1b7KtxKC z`vW3!5D<|KkI1Fg<+vJ(O@LrCt}c=wE>XXtq>0W2Y8!M(a#QP{SGoJpoyhUg1r1Nlm=14l79s5|IU*DaIR4E zcy2wmsyW5}aS7Y$ucqJx#E_jhpd&j09l7;SN9JpNd(@EtIpfVYC8lIf5^@;Us}dh6 z*CI}1l@#D52Cx&l(^LU451fEj`kZx;%ITM^Z~L5mZ|0~7$rAq;{yNoJDKhNq!MllGz)vw-UtM7Vcbk*~#4%L-c!5F}^e%s9!= zQf7m4O!7p__~wN7yB#TMDL)svw335N)c`mmiju}V`DtvPsLk>BJWJ+&C;3b+87<~H z`!JOgs{M|Z7)xquXkVSZ&!cg`BX;`$c4`22YLD141K24f3qOZQs%8EuL-uB1;ZV9# zulvfPHpiUiuL9zHB~}L{%R%#ZG!fNvNmjv|1GEN#QrJ8JHmYFaRY5;qE;q{?P=eem zw;>sHS{6(-@>-lDl(r;hobM*!WMRJfp>gnWqsS`9DYQl<8G-qp4eRR>>a2^avn8NO z#Ol_DM%*lszmkd7`R>Jx8$jYxBZhrEw+=p80wqo?TJujSV;D=Nz+G=Ln#SC^VO{Vc z6}Lejqe1wgR(+x+DOTU&XKWsWc<%H~ZdZ!YXj1dxUJHP8_#TVNkDLQW0q?MrOHZRJ zN00MvUjN;&-S#z*E$>72*_FU`nM>snES2UO9~Mx9~HZCh6CSmzD(E|wgC z++jF9PAIhWi2*x0IiY2|S9|~)lw3Z1AnRhz%|{Nz^((B5DJMWfECDkR3bByB1M_)G zh0@^=noNEaUPbaulc*xY*_OiEO#20CbZrx*x;a|vaIroEs@&OQmHYEpO&mb%pVRZI#tQsMZNilSjD!OkdgBrK6-tC#9?^q|<+R3R9 zo9X5xil319OEBsLB+7Lt!YP~n4&415Ku_7xb_P01#Vr3_3W$)T25@$w@A)4zz>@)~PjOnwi> z*~j~6OABm={Bns!sV$36X~b*UsAp1l%gQP-|=-rtN=xq!DGj|eesA|Y8*|Ik$Bgq|yZvKY1*v7*5tdsyMAS;L!#% z>~YQ3@LY1rWQnla({zXzeD^PcSfB;tNeGRrf_K+#8%hhXgcC4L>5VO@P@Fo()w+AX|z7*%Fv7 zp~b1K0#qX~TN0W>YFdBR>*M6n=m~|Oxh^IMN3;A$NFBL!X{wpX?P~qn{|tzB!l}reoCl-FM4V)e4~5IESnn-O!P&vx&A`I8;h#oVE>Vg0wlZ?d5H zJnMevM2g=^`M3Xr_x-p1%2=Cu{lD5jF!2ARBjzdo@`Rpy_AOjpX05xy1l^#f*&VIk zK2twJF>6||<_$B-O~a6vJPRY~=#{d>*Z3(ny`_I0ja8as(=0UheRO|OauRqFjqfIK z{~=%!=7Tq>!z+>da~lyvX_MLEZEmxL_JM-MXDUdm@6`Q=E+IOi4$)g%Xy60;O!Mq3 zTukbXW~BZ}UyQp*YF*(Eq@{GStodLvl_6|VCN3$kPBK~Fj%QaBp}J9BXbx-7^lxzk zpkb!@fV=l`;9<-gpjB^9oEP0M^MM^c6+lY$^LD#D8nR9s(?Le_qtc-j{;X5SsdWEq~9MD_%fVyW~Xj@MO`kV?s&Xg-?Hc|Y^i*i-v*Bl%t(&E?nx(=qOx-@hg&mN)>w7RN^h0Gj(CiZ2qCsr|e z!dQ;86YHxvfr}}WXW*U&maAI5X5&f`(`jJiYN|ozN=PIVV&WnN>m-u(?YbN5dE3X~ z=~a5gR0#^!SxAaEG(dq?xrZ@|VyV5MS}DsE=e|O!((5TbW zz#KiHLK9fAV_TEN+^l_VY}~}_LUXSC?_HONC00&;-)-z%A;!*@KMwBs$JI8Tv`+jz zSzCe#(+7IrU)HBP<>tveq$4o;Fi!sn_>XqA`MR~;n?-)S@LKN+y484Cy{)G&O`j9-NvwlLh?Ugo`xJjcqr;=>E_3Ty_xjYPbn zSf6MI{*k@=WBIq0=goY4?BC@Z&e5DpSvQG`&(bmbL*3u%+TX0Pc(>5=KA@PAua_)t zm+TLLnRI1oip1jcJth?(NX8caqKqPtrwtS!@|-Ie3jT!cubx(#gLr z4{VPQ9r_!y68~q5{GYA=#glgKUJcRh5*LuTc?ZIwSXeOk+GTAUpVv!e)%sE(^{KZ( zS~{+2Xu8J51<6?*Ca<{RTM2RyHNygCvIV>iICWCZ>!qA~F`pjY|2QB&9|we)SLU&Y z)31Cu8expM?64_C!PSnR-P`!j*--Q1zbTsKTNxN8BRvoK#^Q2A`FO+3rDc$C3)hhd zL=5AxwF4{G5326v#`DuS|LgtliWW%L=4QmfSZ2?$M$3?_Da~|e1BVwi`sOeT+7YhG z7Q|64OF=D?KW~oHqlbb=dxye)U2nP35OljU&;u-D~LEu3{h6&h6WI<`?WYSGU;B^x6=kO$~C4$5jc7Y%q z%QU&t9L5k0RnTSWMB4k zXUlvdsc%!r3$hR)qD(Y+`Z@6)>2oaOrTwF^xUV%9&b4L3jHyP!Roqb6wavw3@=C!( z&^t9o0s4AAf%U)00u)L+qO#VMmdYD+WcSMA6>+e+iG6dnj3Cy+_LZawitT}D0_!e8b0V0!q1qa76k-wyW z9W=Mw@2>mQN{HHA={JScrn)N+*V9g~m>Ax^@SB69i=Wrmh$K`OiGvn`E`Jr{HFNQ2 zy*fzG$k*22cK0z#y&-uvGOnfsZSa_jRGX`~@hajhkPc!5I$!myiqUfMT+0(;|a&h|IO_40b>E-#5=*f{bejxQT$_h)>#ydS9=n6S=)m>_>`PBHM7Ez0{AYXh-UIl?{7=B+HG|--Q~b6#R~fKG|YD%wv(0 zmUW4;LxNm(8O6nG*5n1jYnPD7ZEs9bByVtVQxnB1*+&Vqhe06g9We z7}J;-C2r;Vi52fi6}jaV>4=iD=eU9DBX)?0%M>GEbc`0Gz4B))y+9$t8F8F!BVbOG z-LR3hpe&2e;7gLMdY@t0cs9+Yl(uU>#fHr8*mmjTwOsGSQi_U7iX)$dq$CSNBF#gI zvC3DS+NWBO5t;mSO!18K&PJJ?IR9&SVRnMwJ#y9?3t$BmPI*IA&b0iE-Fyp5&XIDi zq!dd9^-t^f3&EbsoV`YuX)$&ZCpQ7n75%aUp>M)k(cf^(4e-hi2)xN2v3DQrtz+H= zt4)t*QC493jTvzs?0?%8U#| zo4`GWp@fw+w#3G!6%p`}L?pVM)2dK9>3M9lH0HeKa(k&MD@wHN26?pObLM*kTRcW` z`VCA)>WtYR`fqZC>w+#4f5JI+g^E!9S70*!hovr034B@cXKX_vw%;XR;G zkrxzXPEQ(v#;FDu?FWi?iwaQiM1`{PeMRtI|1*95&lLO7bhV`5L{(faAdL=93-9gFbO@gbU4f*R45y1QIQghCKepCMQ z`-!dFoQS5?ghp3I8Oyw5`mg4pH{28lgC;>N4nvwGa=#(C&s9uO;e|68B{B~wQptI! zP3&im3;d@bm!B2$H%huE1KsOU3R5>LI{Kf*G|SLv()t*D&qMl%#qLPOxI-1XVilIK z?5jYzeKJ2Sx8vX<1T>D$@4Qr`DgJXF>PXeF{vndZ`!}UG82*x}luaYta)xe9d(qH* zHk%-bn0_N7%2(oO!A9e?4{|hGlBz6YA{Z(9RR|Db#U{+TL*s@{aWf;})2NOB-@;uB z#*k>EDzB#z!`-_hw$n+-v1l+%NxWJxN^Bl#tW~Yp`OMD=65`y2R=(QRYtvCm7^cD) zEqCTh$d66Ap8w%p$^VOY-ga8$^$}I1S{iJKX%%dtR?WqgYBq4kxlMUdtSGF^^qy)lQjQ_|Gj*KKMzDuHvAB zkR(e9!-P44A$s>iVc5jZxOmozOZv^OgKmf(wvrd4I8KZ-*1Yy3?bt)3%{J1ar6@=q zpbf~xIP@;eg*e&=%{s5%*zjNh+fS_)SG9A|UA((lfI=`Y6{jNICC;Bp5e&wUdiNdJIqLr@ausfd@5~!m6mp+T z`OD5CZBRv>m+w+L==vZ>Z3jnL2Oznrua#gAqXbC4|usP303XrqOS((OVEPx>|&= zhuIvb0BfF!Zc@HRCpzUO^im1P6Sl~vhynv&nU7--X%L>jCWK)U1t=<>W|Oa~pXAvc z{%LAziQ1Fv=OF?n;a>H>F{QVVGspA24CyCUy~9=IW`Yv&lQatVz2~rPci!58Fk;D` z^N_Ojbz+c`gdoSD6{rR?=7W)9ULgZx9H_S4g`s1Y)AR3E>y$tzS`#M!{lo|hv+=8nA zA{R#%hD)l~9x##+o;*@lr+X{j{K~w#r9MD{@>;s-)xZ0|^uWspqr|*@thVKw$FFUw zUK?`RK((cIuQ`r{y_#MD_dEy`u78gIQG)ByF$oHFpSomahxyBqQSaX=X0?FhR>7lN zOb240kZ82em>=*mz@gDyXe8hG_&vXSm#97+9>*++%b9XjPXsEW^a|~@w8$%xf5spG z8KVNmQ!L}T$yV!V8lu@3c}dZH3(L}ekB(`i<1bN2a0CDb@xh*ubfG^w1}%8MO5kZT zWvL3!xJy_*q_mff;OVTgV+`3|W=+Hcp`)a+h zBYY)5r22%oS@owI!})}|5g`nTNQK_jOt;HSYJHq@KO4J`v+CEuesm^IBokib>9Po( z(qP+%T#ddrU^8=Z@lc?5Da8(Q2Dj2k*>5_*)7~saU10ZyW`Dvg27r`?a zL^y=!PLI)^iAD6jtg*3|lX{MJ*-vnnsSuf{S+Sw5@5C)6WBfnoLi_(Y_j+`G{3NPc zaH}}YA%(2AJ+hzM*6%~VCM1e!2@`9t3qK3{}sMvbq81F$C>l#gAv$|jiBJO?EGev)m{^9adIY1%-tb2=R_`j^%tRx~LLDTqV zStT$U+K@S=&xACpw2*XF%vGN9u*$>tpR(L= z|9)AJ@lVh_wf|4h(L6>DrNWe-+?F42hNt>CmeMVIXO}on1)mj!4o&d zihK5}j4l#KH=3INDF5>x<$L@Y=|7;Heq z5=8tRh**Y*E=&q1{BLc zF}Bz!BV3qfop8=NujV{q1C7m{b>d{$_gE+<0s|BaLJ=MtK(PcAzX24>Krt5CC@(x2 zC%hKM3q5Bm223`0pg0-!J)oEf3{WfwMG2D)pjZNm-vWw7peSRLjS8y`lAhQQ3)Me# zXf7y9m~8GqaWd?CKrs>6p+ixfNPP#2s}FB?1OBLPOoEc2#tC3n zAI6i^M?F7HCzKGvv)4vSXL-gGIemM7&iiT`tx+QJChW5}uZCIn$Mcx4=|I8PCHV>= z0Qg#fuYUnwi||$D^Hp}vupmX2h*{`4MXL&UeLr5~K=z?)I!?fKNw~r-0Js)_>t6xa zB5>9DaMhhumZfvd+G=fPT)Jwo_5Ey(^4JHh$uL3ICCLgw0mxc_tbYSpi;&gila)Fr z!elI1coKbA{ET;14bb|2w8nw#!`5`1fa{WQh3WusEdbZQ1Fl8jYWIPwF-kfml(#|) zO*1~C*#fRPaGeISA6!#$0aCZKtuvy3RN{0oSUz^f$VwjHBJ+BU6ig+HG!@r==wS6 zT86HB$eLW|jj@(Bs}Q|lo^N-U;A+a+9(I#N_Q7j1PT+M>yu#E3yq3W07r<*7ysCW` zuS8+EwG_s&V$w67945$`vR1FtMD|<0rt$<}7X>VAO#o~Oz=JD0v6IH0Ja2RzXD*(0Cq2J6ZyVTGAw$OV&i$vXNkkx zl(&MNB(moUHjyU)yC`5GZUSIS0QR>4Y#G4rrEU`5XsL|ymf2a$GXfUwro8^wNg{jp zzb5hoU>5}}C+M55t$`Uz4N*m*zp6I#w3V~A=U&DaxMzY@%=@NN@uZ!{( z1}E^f1Yf@aU(4`y4~5f7FRb-pz@;-@_W!SE+#-g=Dd&mmaU^>^Q9Yd~7`r56B`i*0 zYzfAG3&s{?jZ+R|r;+SM*mRzcL)f$DuP!bFB(8_= zefr{ToC$s%W%Y?P`%-rEns+hv^;H1a?JuYO4>n$`<8Yl?ar)}~`Ni3IrCxl0*=Nze z(ebjrv(y;62jqp}ZdK{cRfAjJ-3wWD1+JE8G@QJ_t z{ya&%^<~n3@z+-aOt$c=8*ZE2!Y8hO*+0*(!^bULa#KQWQtkiXm!CZOY88-o`{w#` zFLas5Mwi;4>-W}yTyv4@FXKmN)n9D?w!IMkZ6!Kw6-i0bYb}%r)7hhf>F)Ra&E08b zik|QZ{_*^m(^nVIuCMx`zK=iZUsj*~a1pBH&72V~=$mjcg^BA$7?IUvY4BJ82|vT> zI5X^Sh)zdvAC{!XA0U|#=p|rMdoDu2m63y^BQbq9ScJCA%&fBzluSMV*6G7r+91` zH&%s#RxG`iGjCRHrj7^RC#h_Gu{&~mHg@lhI`Dw)Dd6EfyHLPGxm~EYcpwg?thLH` zFG$b43w30Gn2g=N_R7h~0}|ML0veKgOh7}SJ*G7mL}y6LUahNyA}>k#8Ml~@i3-uk z>HThR$Kzfp7VL?1`(k198f-1K2naXh){+Mxift|RpW=ZLtOQBJf*l!_VZxescWED3}_)59pP9*Qk zh0SZQ%hVzvc)ndGhgR`&yG;G3cvwDoCgYv$HCj#!qM6SyZDx<-SUyY#Z@=Y3JoLZ> zcA0>Q`*)duiGsUKYfPM&j;$n5D%7;rSmhr5*+|hw(=n0aM0CPTMsTksMSA>!3+y%l z5~T)9EzR6FcTAi}jozOAjd=8d3T!k15Xp@u0HV-F(}oVRvm(q)!t&q< h3csImqv@!~5RcrRJ5ABR{~!O;{|^yqW3AM96aaio-?{(* diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/manifest.jsonl b/sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/manifest.jsonl index 9a5912ece..4119c562c 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/manifest.jsonl +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_rf_ab2026/manifest.jsonl @@ -5,3 +5,4 @@ {"url": "https://www.resolutionfoundation.org/app/uploads/2026/10/Spread-thin.pdf", "title": "Spread thin: Analysing the Autumn 2026 Budget (briefing PDF, replaced version, Spread-thin.pdf)", "date": "2026-10-29", "publisher": "Resolution Foundation", "sha256": "9536b6cd8478dea91f88a103c9004cc7b4cdde269a521f62048e3c523a15d9ff", "bytes": 1410388, "doc_type": "pdf", "access": "direct", "retrieved": "2026-10-29", "note": "Fetched with curl -sSL and a browser User-Agent (HTTP 200). Fetched 2026-10-29T16:17:29Z from the landing page's PDF link. PDF CreationDate 08:13 EDT; server Last-Modified 12:14:43 GMT. Differences from v1: p.5 and p.19 income-half figures, Figure 8's axis top (+£160 / +3.2% for +£180 / +3.6%), Figure 4's x-axis adds 2031, a p.30 typo fix and reflowed lines. Re-fetched 2026-10-29T17:02Z: identical bytes."} {"url": "https://www.resolutionfoundation.org/app/uploads/2026/10/Final-clean-charts-all-sections-for-comms.xlsx", "title": "Spread thin: chart data workbook (Final-clean-charts-all-sections-for-comms.xlsx)", "date": "2026-10-29", "publisher": "Resolution Foundation", "sha256": "a4e1c072bf8f95f364a470be46ad1d1fad2c4f86fa83d371b45b577df1d6e0d1", "bytes": 14637565, "doc_type": "xlsx", "access": "direct", "retrieved": "2026-10-29", "note": "Fetched with curl -sSL and a browser User-Agent (HTTP 200). Fetched 2026-10-29T16:17:32Z from the landing page's data link. Server Last-Modified 12:16:43 GMT; core properties modified 11:51:06Z. One sheet per figure (F1-F23) plus 'Extra spending stats'. Read from the stored cell values in the sheet XML. Its Figure 8 values match v2. Re-fetched 2026-10-29T17:02Z: identical bytes."} {"url": "https://www.resolutionfoundation.org/comment/its-good-to-be-flexible/", "title": "It's good to be flexible (Nye Cominetti, comment)", "date": "2026-10-29", "publisher": "Resolution Foundation", "sha256": "3510a245ad89667dee8e6dbb42d175970c05667a244c4034197c43000962fcfc", "bytes": 131349, "doc_type": "html", "access": "direct", "retrieved": "2026-10-29", "note": "Fetched with curl -sSL and a browser User-Agent (HTTP 200). HTML bytes fetched 2026-10-29T16:20:03Z; the page's HTML can change between fetches. A Budget-day comment on the April 2027 minimum wage rates, listed on RF's comment page; not linked from the press releases or the Spread thin page. Re-fetched 2026-10-29T17:02Z: identical bytes."} +{"url": "https://www.resolutionfoundation.org/publications/spread-thin/", "title": "Spread thin: Analysing the Autumn 2026 Budget (publication page, 4 November version)", "date": "2026-10-29", "publisher": "Resolution Foundation", "sha256": "d81969ef4b4d280e978ff75d8ac529c4fb73009ebaafb52e2e146124cb5b0259", "bytes": 154801, "doc_type": "html", "access": "direct", "retrieved": "2026-11-04", "note": "Fetched by review B of PR #231 on 4 November 2026 (hash and length from that review). JSON-LD dateModified 2026-11-04T09:21:48Z, so the page changed by that time. Its Key findings bullet 3 now prints v2's figures, 'gains £145 on average, against £130 for the richer half', where the 29 October bytes (line 3) printed v1's £158 / £144. No rows cite this version: v2 rows 74 and 76 already carry £145 / £130, and rows 56-57 cite the 29 October bytes they were read from."} From ab2cf6336a93ddbcd2d02c6ab1fba173e268f815 Mon Sep 17 00:00:00 2001 From: Claude Date: Wed, 4 Nov 2026 18:24:12 +0000 Subject: [PATCH 23/52] uk_tpa: apply review B's TPA 1 to 6 and K-5 TPA 1 (first fix): rows 98, 99 and 130 take the total-tax key. TPA 2 is recorded in the access recipe only. Restated rows name whose figure; row 278 and (K-5) row 30 are components; four by-tax scopes carry F6's caveat; geography is UK_excl_scotland. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_01LwQq6JvunwtJCAckL6L4aM --- .../uk_tpa/NOTES.md | 50 ++++++++++++------ .../uk_tpa/claims_staged.jsonl.gz | Bin 20792 -> 21107 bytes 2 files changed, 34 insertions(+), 16 deletions(-) diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_tpa/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_tpa/NOTES.md index 7576e339a..7a9ab6e8a 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_tpa/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_tpa/NOTES.md @@ -14,9 +14,11 @@ Dan Neidle's "Budget 2026 – what it means" (taxpolicy.org.uk, 28 October 2026) Every primary was fetched directly on 2026-10-29 between 16:17Z and 16:19Z, with `curl -sSL` and the browser User-Agent in the brief (HTTP 200). The bytes are under `.ab2026-primaries/uk_tpa/`; nothing came from a local copy. - **The article** carries `article:published_time` 2026-10-28T15:51Z and `article:modified_time` 2026-10-29T09:59Z, so TPA edited it once after publication. The copy hashed here is the 16:17Z fetch. +- **The article changed again on 4 November.** Review B's fetch that day (217,180 bytes, sha256 `d95a0e8f8a0e6465…`) carries `article:modified_time` 2026-11-04T10:10:31Z. Five reader comments account for 15,211 of the 15,247 added bytes; all 49 quotes and footnotes 1–15 are unchanged, and it holds no figure the rows and these notes don't know. The article's own text and figures didn't change, so it is not a second primary under README rule 7 (the coordinator's call), and the rows keep citing the 29 October bytes. Like every HTML page, the article embeds per-request content, so a later fetch's sha256 differs; the check is that the quotes are still there. - **The charts** are ECharts embeds. Each chart's `data-json` attribute names a JSON file under `/wp-content/assets/`, and those files are the primaries, URLs as named (three carry a `?v=20261028-n` cache-buster). As in AB2025, `doc_type` is `flourish_json`, the nearest enum for a chart-data file. - **Values come from the data files, not from bar heights.** Where the chart prints a data label (`label.formatter`), the label is kept in `conditions.chart_label`. Each chart file also has an `aria.label.description` sentence, and the net totals are staged from it. -- **Quotes** from a JSON file are the element re-serialised with `json.dumps`, with `£` decoded to £ (the AB2025 `uk_tpa` precedent); the file itself may print the same element across several lines with `\u00a3` escapes. +- **Quotes** from a JSON file are the element re-serialised with `json.dumps`, with `£` decoded to £ (the AB2025 `uk_tpa` precedent); the file itself may print the same element across several lines. Three of the chart files print £ as the JSON unicode escape `\u00a3`. +- **Chart descriptions:** only four of the seven chart files carry an `aria.label.description`; the net totals are staged from those that do. - **The PNG** (`child_benefit_budget_2026.png`, linked from footnote 1) is a calculator screenshot. Its values are printed text in the tooltip panel, read off the image (`parse_confidence: medium`). - **The "Download as PDF" button** builds a PDF in the browser (`tpa-pdf-exporter-loader.js`). It is not a separate document. @@ -39,8 +41,8 @@ Every primary was fetched directly on 2026-10-29 between 16:17Z and 16:19Z, with | measure_key | rows | |---|--:| -| `null` (65 with a proposed key, 142 context) | 207 | -| `ab2026__package_total_tax_policy_decisions` | 37 | +| `null` (65 with a proposed key, 139 context) | 204 | +| `ab2026__package_total_tax_policy_decisions` | 40 | | `ab2026__writing_down_allowance_12pc_and_fya_50pc` | 15 | | `ab2026__bank_surcharge_6pc_temporary` | 11 | | `ab2026__hmrc_tax_gap_package` | 8 | @@ -50,9 +52,9 @@ Every primary was fetched directly on 2026-10-29 between 16:17Z and 16:19Z, with | `ab2026__electricity_vat_zero_rate_to_july_2027` | 2 | | 14 other single measures (Table 6.1 lines 3, 5, 15, 26, 44, 46, 49, 56, 59, 60, 61, 62, 86, 96), one row each | 14 | -The 142 context rows have no measure key and no proposal: +The 139 context rows have no measure key and no proposal: -- the earlier fiscal events in the two history charts (`conditions.policy_event`), and the 2026 Budget series' 0.0 fills for 2024-25 and 2025-26, years before Table 6.1 starts; +- the earlier fiscal events in the two history charts (`conditions.policy_event`). The 2026 Budget series' 0.0 fills for 2024-25 and 2025-26, years Table 6.1 has no column for, carry the series' key (rows 98, 99 and 130; review B, TPA finding 1); - the other countries, the pre-Budget UK, and the after-Budget UK's bank levy and VAT components (which the surcharge leaves unchanged) in the bank chart; - the fiscal-context levels in footnote 2; - capital-allowance present values under pre-Budget regimes; @@ -62,7 +64,7 @@ The 142 context rows have no measure key and no proposal: TPA builds no yield estimates of its own. Every £ yield in the article and its charts is HM Treasury's Table 6.1 costing (or the OBR's), and TPA uses it in one of two ways: -- **Repeats it,** for example a chart group that is exactly one Table 6.1 line, or "about £0.7bn" for the £695m total. These rows are `restated`, with `source_model: hmt_costing_obr_certified` (or `obr_efo_forecast` for the OBR's paragraph 3.17 averages) and `conditions.originator`. +- **Repeats it,** for example a chart group that is exactly one Table 6.1 line, or "about £0.7bn" for the £695m total. These rows are `restated`, with `source_model: hmt_costing_obr_certified` (or `obr_efo_forecast` for the OBR's paragraph 3.17 averages) and `conditions.originator`. Every restated row's `note` names whose figure it is ("Restates …"), and says "TPA names no source" where TPA cites none. - **Re-expresses it.** TPA groups lines, sums them over years, deflates them to 2026-27 prices, or divides them by GDP. These rows are `own`, with `source_model: arithmetic` and a `conditions.derived_from` naming the HMT or OBR input, and every note says "arithmetic on HM Treasury's costings, not an independent estimate". This matches the IFS lane's treatment of IFS's own aggregate of HMT costings. The ingest can separate these rows on `derived_from` if it treats them as restatements. TPA's own calculations from statutory parameters are `own` with no `derived_from`: @@ -88,8 +90,10 @@ Under README rule 8, a measure's own parameters are its design, not rows. The al - every other £ figure is cash. - **Cumulative rows:** the six-year totals sit at `fy` 2031-32 (`value_kind: cumulative`, `horizon` "2026-27 to 2031-32 (cumulative)"). - **Present values and the bank chart:** the capital-allowance present values are properties of a regime, so they are `point_in_time` 2027, with the regime in `conditions.regime`. The bank chart is under "2027 rules" (`annual` 2027). -- **Rows with no stated year:** "about £1bn a year in the medium term" and similar figures sit at a stated year, with a `year_note` and `parse_confidence: medium`. Footnote 15's "£450m to £495m a year" (a range across 2028-29 to 2031-32) is staged at the years Table 6.1 line 68 shows those values: £450m in 2029-30 and £495m in 2028-29. -- **"Central":** TPA states no ranges for its own estimates. The one `central` row is "About two-fifths". +- **Rows with no stated year:** "about £1bn a year in the medium term", HM Treasury's 84% (row 9, staged at 2027-28 as uk_hmt_redbook stages para 2.50) and similar figures sit at a stated year, with a `year_note` and `parse_confidence: medium`. The OBR's paragraph 3.17 averages that footnote 8 quotes (rows 29–30, "a year on average from 2028-29") sit at 2031-32, the forecast's final year, as the uk_obr_tables family indexes the same sentences. Footnote 15's "£450m to £495m a year" (a range across 2028-29 to 2031-32) is staged at the years Table 6.1 line 68 shows those values: £450m in 2029-30 and £495m in 2028-29. +- **Value kinds:** TPA states no ranges for its own estimates, so "About two-fifths" is a `point`, like the family's other "about" figures. +- **Geography:** the calculator rows use AB2025's code `UK_excl_scotland`; the bank chart's other countries are named. +- **Parts (README rule 9):** the bank chart's "Corporate income taxes" for the after-Budget UK (row 278) is a printed part of the 33.7% total (row 268), so it is a `measure_component`. The OBR's £0.4bn (row 30) is Table 6.1 line 61's receipts head, so it is a `measure_component` of that line ("Tax head (receipts), annual average 2028-29 to 2031-32 (para 3.17)"). ## How each TPA group's Table 6.1 lines were found @@ -109,18 +113,18 @@ A chart group that is exactly one Table 6.1 line carries that line's registry ke | `ab2026__package_tpa_savings_property_income_and_cash_isa` | 11, 61, 62 | "Savings and property income" (by-year), "Savings and property income tax" (by-tax). The group includes line 11, the cash ISA limit held at £20,000, which TPA doesn't mention | 7 | | `ab2026__package_tpa_temporary_vat_and_fuel_cuts` | 2, 3, 18, 65, 66 | "Temporary VAT and fuel cuts" (by-year) | 6 | | `ab2026__package_tpa_fuel_duty_red_diesel_and_hgv_ved` | 3, 18, 65 | "Fuel duty cut, red diesel and HGV VED (temporary)" (by-tax) | 1 | -| `ab2026__package_tpa_business_rates` | 12, 13, 14, 57, 76, 80 | "Business rates" (all three charts) | 8 | +| `ab2026__package_tpa_business_rates` | 12, 13, 14, 57, 76, 80 | "Business rates" (all three charts). Registry caveat (review B, TPA finding 5): on the by-tax bar, line 76, zero in 2027-28, is placed by assumption; the values do not depend on it. | 8 | | `ab2026__package_tpa_energy_and_environment` | 17, 69, 71, 90, 96 | "Energy and environment" (by-year) | 6 | -| `ab2026__package_tpa_other_income_tax` | 5, 15, 28, 52, 53, 86, 94 | "Other income tax" (by-year); "Income tax: mileage rates, Pension Tax Allowance and other" (by-tax) | 7 | -| `ab2026__package_tpa_other_vat` | 36, 46, 51, 59, 88, 89 (the by-tax bar may also hold line 66, zero in 2027-28) | "Other VAT" (by-year, by-tax) | 7 Registry caveat (thinktanks-a review, finding 6): the by-tax lines that are zero in 2027-28 are placed by assumption and line 66 is undetermined; the values do not depend on it. | -| `ab2026__package_tpa_other_corporate_tax` | 49, 50, 54, 68, 97 | "Other corporate tax" (by-year, by-tax) | 7 | +| `ab2026__package_tpa_other_income_tax` | 5, 15, 28, 52, 53, 86, 94 | "Other income tax" (by-year); "Income tax: mileage rates, Pension Tax Allowance and other" (by-tax). Registry caveat (review B, TPA finding 5): on the by-tax bar, lines 15 and 86, zero in 2027-28, are placed by assumption; the values do not depend on them. | 7 | +| `ab2026__package_tpa_other_vat` | 36, 46, 51, 59, 88, 89 (the by-tax bar may also hold line 66, zero in 2027-28) | "Other VAT" (by-year, by-tax). Registry caveat (thinktanks-a review, finding 6): the by-tax lines that are zero in 2027-28 are placed by assumption and line 66 is undetermined; the values do not depend on it. | 7 | +| `ab2026__package_tpa_other_corporate_tax` | 49, 50, 54, 68, 97 | "Other corporate tax" (by-year, by-tax). Registry caveat (review B, TPA finding 5): on the by-tax bar, lines 49 and 50, zero in 2027-28, are placed by assumption; the values do not depend on them. | 7 | | `ab2026__package_tpa_other_by_year` | 10, 16, 19, 26, 44, 56, 60, 77, 78, 79, 93, 95, 98 | "Other" (by-year) | 6 | -| `ab2026__package_tpa_tariffs` | 10, 16, 77, 78, 95 | "Tariffs (incl. UK-India)" (by-tax, by-measure) | 2 | +| `ab2026__package_tpa_tariffs` | 10, 16, 77, 78, 95 | "Tariffs (incl. UK-India)" (by-tax, by-measure). Registry caveat (review B, TPA finding 5): on the by-tax bar, line 95, zero in 2027-28, is placed by assumption (the by-measure bar settles it); the values do not depend on it. | 2 | | `ab2026__package_tpa_cgt_founders_relief_and_badr` | 26, 60 | "CGT (Founders' Relief, BADR)" (by-tax) | 1 | -| `ab2026__package_tpa_other_by_tax_2027_28` | 17, 19, 44, 56, 69, 71, 79, 90, 93, 96, 98 (line 66 undetermined) | "Other" (by-tax). The non-zero 2027-28 lines are 19, 44, 56, 69, 79, 90 and 93; the zero lines come from the by-year mapping | 1 Registry caveat (thinktanks-a review, finding 6): the by-tax lines that are zero in 2027-28 are placed by assumption and line 66 is undetermined; the values do not depend on it. | +| `ab2026__package_tpa_other_by_tax_2027_28` | 17, 19, 44, 56, 69, 71, 79, 90, 93, 96, 98 (line 66 undetermined) | "Other" (by-tax). The non-zero 2027-28 lines are 19, 44, 56, 69, 79, 90 and 93; the zero lines come from the by-year mapping. Registry caveat (thinktanks-a review, finding 6): the by-tax lines that are zero in 2027-28 are placed by assumption and line 66 is undetermined; the values do not depend on it. | 1 | | `ab2026__package_tpa_other_by_measure` | 11, 17, 18, 19, 28, 36, 50, 51, 53, 65, 66, 69, 71, 79, 88, 89, 90, 93, 94, 97, 98 | "Other" (by-measure) | 1 | -`ab2026__package_total_tax_policy_decisions` is registered. It carries TPA's rows for "today's Budget" and "the tax measures", the net totals, and the 2026 Budget series of the history charts from 2026-27, because their scope is all Table 6.1 tax lines. The totals come in two kinds: +`ab2026__package_total_tax_policy_decisions` is registered. It carries TPA's rows for "today's Budget" and "the tax measures", the net totals, and the 2026 Budget series of the history charts, because their scope is all Table 6.1 tax lines. The series' 0.0 cells for 2024-25 and 2025-26 (rows 98, 99 and 130) carry the key too; Table 6.1 has no column for those years. The totals come in two kinds: - **Sums of the 53 tax lines** (`own`, `arithmetic`): footnote 5's £700m, the by-tax chart's "Net £0.7bn", footnote 7's £19.1bn and £21.0bn, and the by-measure chart's net. These differ from HM Treasury's printed totals by rounding only (£700m against £695m in 2027-28). - **HM Treasury's printed totals** (`restated`): the by-year chart's "Net total" (it prints £-1.65bn for 2026-27, HMT's −£1,645m; the 53 lines sum to −£1,635m), the GDP chart's £ figures, footnote 5's £695m and the article's "about £0.7bn". @@ -160,6 +164,7 @@ The family uses the registered Metrics `revenue_change`, `caseload` and, for the - "the UK's approximately £3,000bn GDP": an approximate OBR level with no year. - "over twice its pre-pandemic share of GDP": a ratio with no printed base. - **Not numeric:** the OBR's "High" uncertainty rating for the property costing (footnote 8) is categorical. It belongs to the OBR family. +- **Out of scope:** the calculator page's 69.5% marginal rate under Reform UK's £15,000 allowance is an opposition proposal, not the Budget's. - **Not TPA's:** - image-6.png reproduces Red Book paragraph 2.54, HM Treasury's text, which holds only parameters. - Reader comments under the article, for example "the cost is £1.2bn" for Founders' Relief, are not TPA's. @@ -195,4 +200,17 @@ The thinktanks-a review (`harvest-staging/ab2026/reviews/verify-thinktanks-a.md` ## README rule 10, 3 November 2026 -Under README rule 10, `stage.py` moves TPA's figures for all the Budget's tax measures from `revenue_change` to `decisions_effect_on_borrowing`: the 37 rows keyed to `ab2026__package_total_tax_policy_decisions`, in pounds or as a share of GDP, and the three unkeyed 0.0 cells that TPA's charts print for years before Table 6.1 starts. Counting from 0, the rows are 2–4, 25–28, 98–104, 130–135, 143–154, 168, 235–240 and 264. Single measures and groups of measures keep `revenue_change`. No value or sign changed. +Under README rule 10, `stage.py` moves TPA's figures for all the Budget's tax measures from `revenue_change` to `decisions_effect_on_borrowing`: the 37 rows keyed to `ab2026__package_total_tax_policy_decisions`, in pounds or as a share of GDP, and the three 0.0 cells that TPA's charts print for years before Table 6.1 starts, which now carry the key too (40 rows). Counting from 0, the rows are 2–4, 25–28, 98–104, 130–135, 143–154, 168, 235–240 and 264. Single measures and groups of measures keep `revenue_change`. No value or sign changed. + +## Review fixes, 4 November 2026 + +Two independent reviews of PR #231 at 4f424b9 approved this family with fixes, all low: review B (TPA findings 1–6) and review A (K-5, H-6(d) and H-6(e)). They sit in an explicit block after the rule 10 block in `stage.py`. No printed value changed. Indices are 4f424b9's. + +- **Finding 1 (the first fix, the coordinator's call):** rows 98, 99 and 130, the 2026 Budget series' 0.0 cells for 2024-25 and 2025-26, carry `ab2026__package_total_tax_policy_decisions`, with the note "Table 6.1 has no column for this year; TPA's chart prints 0.0". The key has 40 rows and the context rows go from 142 to 139. +- **Finding 2 (NOTES only, the coordinator's call):** the access recipe records the article's 4 November edit (reader comments only); it is not a second primary. +- **Finding 3 (and review A's H-6(e)):** the 21 restated rows 18–24, 26, 43–48 and 136–142 each name their originator in `note`, and rows 18, 20, 22 and 142 say "TPA names no source". Row 21's originator is the OBR's *Fiscal risks and sustainability*, July 2025, which footnote 2 links. +- **Finding 4:** row 278 (corporate income taxes, 30.5%) is a `measure_component` of row 268's 33.7%. +- **Finding 5:** the by-tax caveat is on the four further scopes (`…tpa_other_corporate_tax`, `…tpa_other_income_tax`, `…tpa_tariffs`, `…tpa_business_rates`), and the two existing caveats moved into the scope column. +- **Finding 6, nits:** rows 299–304 use `UK_excl_scotland` and record the tooltip's "Smoothed tapers" (`conditions.calculator_setting`); row 9 has a `year_note` and medium confidence; row 1 is a `point`; rows 10–12's locator names the third paragraph after the Red Book excerpt; the chart notes, the £ escapes and the Reform UK rate are recorded. Also applied: row 141 carries its printed `chart_label` (38.5%), which review B's sample noted. +- **Review A, K-5:** row 30 (the OBR's £0.4bn, line 61's receipts) is a `measure_component` with `component: "Tax head (receipts), annual average 2028-29 to 2031-32 (para 3.17)"`. The optional alignment is applied: rows 29 and 30 sit at 2031-32, as the OBR family's paragraph 3.17 rows do. +- **Verification:** the contract checker reports 305 rows, 0 problems, and a re-run of `stage.py` in a scratch directory reproduces both files byte for byte. diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_tpa/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_tpa/claims_staged.jsonl.gz index 880354a3c0a883fe4184dfa2181c281b33351e1e..064e102ec280bbc690c651d5ad8e850163f5db54 100644 GIT binary patch delta 20309 zcmXVX19T7II?n(6K{ zGu>4+t@&X0g<#r5kucx>+rU3Bv({twC7O$`k7>ZaW3ZwZsNMS1)|nTKhc>&%>PZ0F z@B-ZebIEBF^w7#E%f`p;*k`Y6nK+?R+wL+vwhD7hmLiBLXkDdLNX1ojp)!-=w3C3y zs^sFhwH`b<1F{8@PrD_{7hbp_(kPUKs=MA}zT9nn&hqab7t{FH=(-fy6MB>ro;y*= z34BbBJ3pa5F}9B3#Z`)Q0K-^Uoi~h5SkG^dIlx$^BCLJ#Sa@WEM?tE%38r~u%@;sO z%b)??c1K?pRt=LTA@lR}V<5Moa4Ybb%p+a=AXdN}vxPjfHCBTqBlP$4X!UKtKi{Db zy`7)6x7~B#xr9|YEuVkQ?;=ry|EHXkH*j1?jME;FJ7Huu6N@k0=YY1#p2tHdUl=Wfoo5)2L_H%5F_Amj$AGvo<5hrh8OZ z$Iy>G>aHdBL1guNh#6%Fl%H>a`al5x?NmDyx<9_FC_!`R`J@&?Po^K&@fRM*KOA2O|O zW*3r6V6oS=IaAmX0VA1mQf+?3H|{G5u_4Lay{T^&=f#=b^n(6TDLDVe1RCuvw$9zr z-3MjKUKREzW%UsUV)JW{cp!^ld>=N#QWSTnG&e`K0!DW28hcZO+8QVb(E?~p6R03_ zfB_ypxO;AncO674pdbdvIkK8bg#@E3voq5Pf7Ic8yi5F^{d7 z2YC0OP?yYLzK@eF;lsWTH>XD@*JUP7-i?YB0_1VyOm315TWs-N2x`TDA6y8*JCu1p z&02m|@O*fk`|AG`!ch#Ua@J#B?s>OD0J7`kuu*TkW^!)$j9qgr+X`-2>$}H40er+b1Qr~Wqiu*%c0*v3ljk_EemLtjtjMXMrr z;<4UxrV6m81R~H9sKsUSSH7<8jk`$?bfsK-!zM}NJ-oT*UKbM!J#pG;)r(x@W%(|V!GD$bVNh95 zKBX6L-|Hv*gEQ_9mT>2UK7^DG1=gpt^bY#BCBRO*1bM>9iaKL7ertMmh9pbzqQV!S zUWr5lp252x!Ii z^AT13Bzu4<^5Heu3J`sF&sC2lj_1^U<ekVZ6+xx_zDH&nU=3g1*WXze7d>1 zUrdi^iFV$YqU51@po4vaDLkx*dSYaDNLDROysZql{hWQC(^A3L>AQiIbeD^|a)^i< z`Y;b^jg0N*iPF0QQ%C1^8ZpnleD6BpP;r|QT;CMs8R2F#`FQA6D!&XqJ~BXNMHPZ} zLVP4k<*s{$-Q5F$#at4%1GQ{qN3kKyCx-I-49EreArpk8}}hX&5Od+S9>mLni%;GA0&IqQzowK2h4 zi1etq!4f66QU5+98Q}5ZkQz<;0fq99X#$%sxSnK6>klAHfIXWLNCtvF>oTlzlKH-0?^?0)esuSH^-{n_Z%TUQ!S@!Gl$aTa3~wA7c^oDd~w^yhk?qHL4JsJCiUfv^GB1g@5U zL?l;s4{Dos9t<>G?z?H?Js4Apat6rzUQa)C<-m!SLy6#1UcA8JUW`AVfki-Y-!D$- zH=!u57b_J5>85&9>Nv6raz1%~dLPbZu?YS}q)LtDpKbPi2WH|P^XA&eK$b@CQY?hZ$cWUu4dD@-m~i z9&rqM4ZnI2U# zp>y&jw@ffXh=xLqrV)w}T+4zY*XQ)%UYnbCx3?~Y?<0kEUj1*2+w=pHLi^7iw64#` z01BIJLf)OI@Ge2tn^=k^{urI5tE>+H$kgoP1?%KKX~P|?TMB%9UCd3VZQ~yvYVXGe zz3twCkk2&l=Q*Fy=TvHO478OmoaZ9%%R-;_6Xub*GJMui*(1(&m>gN6g)>Xr|pQgJTtV-8tcjX8AG6bo83*@}bO; z#hTg!<9C*6yD+jEb^Q7&?C&GU>6lg_ECd#S#p|$xjL%PxOYt=uMwz|d{^QTEYRMYF z?%aSHpyuOLPUhuoqr4XwlD>b#O9&ZVB#TDbX~Fr)oKa^(hTUlz>B&0hJzeWo$7`cI z;qkJ?__#s`9J1SFHG^+yH$Jm09xW&q{4>!T%IeC@OFN{kLoRX^RX&$>@YB&e&u#{I zd%Ur8z4Ni1d*_)mqThJZLd4%tB<7E5A2I%k0-l7U-lD6DS9>V(Fc>Gf&RtV|+f-^j z^Bcie0@{Fa+_j5(*>%s0q}m#~BDuN8D!MkptJ1DB<)VKX%x23K3)OhtJ*(<^W@%y% z94Nw1(v82zMZIn)sM^zEBsj6857_~1VAc!-0rzy(gA_l4Q1?68)md@r#EkPDo!N)= zBS#&;tNcTuD<}mqo!FS|8?m|x*=$)-WKD~xlQ8q0SZH&WSbjL?wmfU9d-b)3QzH)7YuZwMiFkmG2~1#Ty4Qy~=Fq zO$PTzjl}izI@c?M+LkBZIR|rpwuPIYSxS1qvB(iI8AZ#Al!`O8=q!BSTVjXb`>`j| z$`G_pcAPTPziLI9J7ps{urRKihAkV#(6@r+o6g7?sUVr!Q9`zsWiZ$cD2lj?Ec`I8 zh~XzwoFddHQx<8(27zx^Fs%ZG;bD5_BSv+yyqV-x>oLV1Y$hP26P73iOk1R{i!+Br zlg_AEo@Wwx0z25u+4@xc^BwjOHT@y7GOg+97B6|bAiBJ3`Gyk zy6q2l`K$?7hGnet0Vz++CW8OQ7Vj~D;-gw^vHx%9I}nd59+3imrbsko{wF;9}L2Vm3SY% zDKk~{Xb3ayUU0|1DARi@Zb%q=*8IFJzA-@Zp^=gy5b>&03R?ox?g2j=E!dg(LAXeA zdc_q?TI!Jb0lP0*qE&987bd@lGkmiE^-Q%RNo*eiJdDlF?1$PfB){84ZMm%Tz@FB2 zyg)2qUVq8niuPPZxc)-FPtJ*k6$NNO$hbNO=7QN?9l#aSH}j43CKA{CMA956zJQD*7!%yam-hObhzix9 z=}b=Vuh|nA0VAcBgs^A!1gBcL2>*z_>CUw08hhV30{yc6cBl8_)5tS@R|IP;vAcWi zw&y7>kn5Mjjz%koxR=9LQty1p@^lA%fhnXC?i~k@M>Ic=e<^!Wd-vQ`HPg~1xyuzL z`~~P`clZuSU8&iY?T@Z-8<^x8qoJ(s)x&>uyzLx2+}_)kG=K?j25?RC$vjWD=}{?k z=GWY?DNW_;CN(m(P>Fn^T10xHLcTPw9V0UN=|+G6b zy#UVpZ)<}HB1sI+JX3C*K_bh8?B~^Df=2bqOVnE#!KbF#%bvTlvsa1@@-}2CKK7e1 zWY#yJ4D?Jko*?>I`K6Dkt4UL%cRfJV-^$WlZb`G8;VP>|My$`Trc~$qkW|WNgOU&sI+~bHJ@o@4`YabM4CVw07 z+Wf){_4?Gbgp6>pw}4ATOWzRof>BrS@oBxLp5FBFI9|C?flrt6)`5}J{8&kS9tep) zJ(P@C?;BqLOb{^myrtSQ7ukQ~mgsS~`t5%b6js*Il%qpbK9|h&NfITPkBa2ojV`3% z&4qBao*_aJ+7UvDjlY)4W3L}d9dZRUIi8yh_EEV5LagfLYs)dz zfj-4>MNOpB(YnF+ZqzlLVqjpKRy3QxR~te+O8p$$?KMK!#Ti8#bSt&{SjEYboBT~9 zBflUg`j|W;UH`Yq>I2cblWh z4ZdaMn^S8>u87({4t#8hl`Ev}p|3R=MWGGYfBn{zp0D}*!ZYfzPE71r8>Mo^j(M#k zH5}Kz*@xu}&M>ZrlE6_oj_(7P9C9$Z#0)ad?>cfC&h5K>Gn!A9MgN)gCgwe$v6<7~ zdIqMxVRd8?wb1ZAu`GMnx41Pt!fzOKNpZENG5lSAU3(@RAzhRX#B~JV9pa_mz!N>=u zQ=0kUiL%5UBg|~lZ?e`v*WN=(LF&CC!V)~?I%(>&zev`UZXeq*3CzA%f39qUQyj(v zE*9@g=^Oe;4zAavC;J*HW9l{4I^1~m_V>5@O!-%LD-0oZy4=Oh@KkU0DsolA*B-x| z8brEY)|GG^@-2Zqmlm2G?!#+sbW5I`{pFU@)6NMPMaMqQBzCC46b-G|`tnGfnL2Bj zqGP6bkXdi`PtGbRJc3mYHQ2{jWn_V)#rpw-h4Y&nS$K6FIZk>gol}m3N(WB`#Ce&H z&-KS>m%`nwZ-XfGo40=k1Q$^lD&Q%^jIIgWQ*zPPP-*iBRKnS)>WaIXXg77NTi0g9 z*X#1&#XKh^T?}u4Lsxi%Nteuo+51*x)a9%D{+EgsSx9w`WKZ-3TkM0#x%-(Vq((JN zZri_guO9XHUnEOsNX7Zojk^yl4dg#WE?@>78ggMqAMY?5i@^3fz85i_f%2WkUTrY9 z=Dv%H`(K?FH?NS++1f7taL8EhDewQc_1V|=+B5_jg=_Q{#|MY=7M*JKd=H~7 zlRUzm5BpXjg`Mg75i!5Sr~LUtX2T%#nB`TQeWD&e0Zkz)nHzu%Ll|Mr~fqP0G%oqPDla`t%7PgHDp`wY&K|MVn>t#*Nu`-hzAa zNor@Ygb(UBt%hKT zKo#8lzLE3j_Tb0b6bu~+J=su%&apr{?QC6Ri9k%vuR1eG?|Oq)=ZK~DC5g|n|0Tj`}G2<6P8^2)1`a2V?5c$QnjbD7Ox>Nq%8try^7|(J_YO zN#T|y42>iEv2UXKp_DzSf}_c(ZBT+`fvF!~W?q)QU|9bLV<#iIPQ@DbkHmpZI|_E6 zks3Wb4mM54Vr)`Lr`EluJ<7-6ApBCZ9jFVh*utZJrI+GhVv3zdzA+ zOKWaLn(EO;0{`!dE&myt|IgU-*A*xDlEvW9@qTGl8;U=CH`HS6V8n%x(G=|akkG82 z|E_^CVG_azkvt`!&mCi$Q0wz;!pxICBj^zQ%^~QJ*Va+IPVTC8t*+9>n1^EoEE%)^ zj)b1BsIC$vS&?gW*LAcdf)xdiGM?p14IfB{6O$vBjNo|OUt4*7LSP=k?bASraBU&I z*;iAUGQ-n2s`F?Ou#Aa2|0wodkv0qxrSmS|{t#MgJy$%eL}37x0^6 z+fPBTYuaDt{pB#g1>3U@Y8vOIX8sejtr~6HHB#eh_7)#kh5BYD!P(EW+aeCtyan7} z7G(1qt965VJM%`g22t+1L;F46fr`XusqhwGVz$Q-t!8#pqAbmOuls#w=oE{VY3WA>T7(`z6VYZ^`cb@iWgQACnK#>IG5__yQ?>BApVJ4iJ zY5#4A|MXQNd@%+^{C;B|Zg+xtUmfB5)e*Oqc5k?%K8hW;OZ1i?<4kfHZ-%=^chsGlSP8!5IN+vSgneew_Tsjbxu1orN{m z;!sE%N~Aej1D{)pu83QyJt*qGYwXIL3%Ex&IH%}ZKCTBWzI)Dw&^o%O02wqHY~j^m zdrSD{ypVt0UwosK)P5&0aIPVdFE>PwlKtkeD$h*522lXVrhWa-IxDNkM4}?2=Ou>! zA`MHvh6M4I;zolhBUG=1{Q`jf_shoXmkl!QOE#3V7@Dv}&P>`V;fu6DY5~~1IeIZn ziNWu{@mn;STDMKI3MP20Nd1!)(eg7S>yaV0BM~Qj z51zxKNcX!SN58|NLK7)AWsUYbw)&IK;+ePLvMWN$7EfCC3}9D{3_;@xKo?->RIwOq zXk4QuU(Muf3S4HfNgQT2t*-2?kif3qs8WXbYnaNsf!4C-p3Qep*kI#5 zoW?M~w1G;v?D*pfRNxJr#4NWX=-lMD7yX`^(hF6o!Nfw$*+VfA6 z+e=+Eq|JDI!*nlC%NL*lwp&}7)!)vOh3 zg`TT5C^KL=7keNMOMs$J8Bgx|5U%JCSq-73k|M__cL!7)W*sb|whpRvpj065jv6_l zm3m-0@Igh8iS(ebQb)|rU7#67eS94P-He=U$mF2;VQ`IMsxs@Som~7hCSTKB1H&np zGmekbIrwUdK-Nn{+HC!VfP&>TCUwOe3;nKnDcjd9G}UBFJ$Ue7ftcu6aLwnxggfTRJqF;a*12s;j{g{bm|R8JBl_RpWLl$4rnG zu-<$rT1$3~#Wc&beUJ;bq3t2r)I68PKWG)bGx>eE(v@;D-n`Cka_tET%g0RnKghZ6 zm7A6_?Mdfw~~N)fluopB3hzKYCP;Z0XdY zp2oN!;y#HDZj&(gb719ExRBY!S8jnC1h|91Fe-?&d~P6Yii0gEUlumKvs8}T7t;@?OZOtDnmjSwN> zag{R3maFNNF8BA+$b;r&OZR1;y!L*p$QYy;#6rIonTk|6M76{4**NRa%emKa8p z2(Jqgrga4BCsFTzKh5T7CHaXCp02u$8oNn5MAHtS$iw-SS|f=E1sc-x5_92 zaHwz*r>bvRPFaAi�o}^Nb9;U?Azx!oC1fI(Bs_A%DgIB$fT;0}ZWY9D`a z8n92~71C6eL>wVDCSgajX5L`x8eJUUv8AfWV6fLK=qHlB=SP^}w8aP#pbpX7){ux< zzTUmwMZCAr(a1%}rWq9J_Y7nl1vWU6}Hx`DWiBg-d?1^G-n(!-`lj?FIc3WJ_gB2CKFq zi`djZabbgnMMm+%Rev&XfHU#Ajmz3%0V)-_npLP6PK$aI=cb zT%e_LvOWlk`bAE4nUt_sfbMyG0vP~p{sX~6SGD|JFE|d^B8X@;4|JU-dNBR*sS_nM z;2qK;MzV(#{Y^9(cI@Ejz?XPC9D5``qT3VtSUZakemdsd`=~y~owLabS?Z+9z*kBK zye-E{jBx~vRfFSCTTf?d8<~9w)7v|iO@T9UAlB}7PiJpz=i8xp5#dJVANRlP_~Ntn zuvz@H>1)G(sW0sNV}bg(s>qvHwyvPF;rzX)u!XL^2d}iZO3Au6~iyu8aO1TsfY516hOn% zpLu^x?_RiBucdgsT(hOKnFjT1ChNz&%EbzkDn2mW7e}<)_p;0~I_){Zt&j=wuQ=19 zX40ijktWs;PmEUQFhA^@lY1F00>(}MuJ_w_x|sjFa*Q0qp;V6ZCEq#g!{m1J_Noiq+lQ$Z z`*{r6n$Ujs*M+r=2It}!Zo^s~Z}A$z3yw&A*cNJFw&nf;T|u?FV|s$D<6d{Se-Qax z--SAKCrq_%R8<)V;3DoqI=jrO+zf0MQ>P9QpsuF199@^^z2w2_O5t1RiB{@wUdrhr zptcXQ02B(1S74>GyejuZHiYNFE1dzT>WeC=`1*CiIebp`JwGM!fVa|tdH=w9@LZVZ zTIK>ezpmHZ&0>G&>gerRwp(wUYSxX*N-h+mNhm5r{1MU~Wx#`m_%z!0Q3TP6FznE~ zgT$BroGI^6Mhw4c@@(HzA19R2XK?ZPj+G$bZOrb!`3QpTc!n9Rn z_PF}VsHWs-`sTg+HkjB8G59yTK2=RzUJ%`QsG9c(XuTH(N*f_#w@Lq`MMaTaM?*nq z_h14t5s%LnPk(H&1GU5Xf!fc!ufbT8O6EP7S-!7tiN(``buS`(uMv9fEi#6nC(Wi5)8-LqTW)hYZ}m-lEEI`*GV$4|&rz@jZbZzm(!&}L2U z#?d5@9$l{4Py4&7y9-+9;{#qO?1@YGI$BwbU&cN3zApR zB@DJ-+L7Y`*yO+y0~7#az+!7n{04edSW4;brna znYz0&TtWljld;<)RR5^DA>=m`md>`d`M{j0l6oLJE59s`bh24nz1c}I2QE%>xu&r# zdsdWWLYdiUwYgV$x%@&MJC5{xU4D7JEn$KVg8>m2br@b(dk!>xSI@f_G48@iTmzoa zCAIocdF?aC=f~;x4FoHf10f3BYnx9J4V0hfZ7c!YZ%?qH&3iouk zI5%n6@0&fFc~omLh(-8p3FgJ^?0Dojm?*+`o1Eh;f+^-+LphZHml`SghxyMBVwXL{ zRcLNLA$EKzj0QZ<8?92OIIcdo?>Tte`{6^bH(oJpx^4IRpP;7!uIteIv8fb(A|S^M z&m|R$k9%JY&QSmR!3*05X6qx2C=r#0EjpJegPna{L>&NAwEItcmewXhlYc8t@Q|%W z9ASkWUku>muY{_XS*fWcPskcC&%8nc+afkI;>WTnh{(boOrO5PQu_6PauIGdd!3x6 z!-qvRCt{UIA4m+#*AP^u4hB0f1C+u3abVg1UGgMGw@h8DO{HpSEXgZfckIZ2tk)h9 zik$am!+%@;S3B(I8VWXEy9{TiosPWhikWel`jm{=Z;Y!uOauD||eSj@Kob{OhGb}5oyP!$e-#re>?vQfZ!<0>q>GQ~HZ z@|=M~bM%r~)8g29#}b;rku^toLN5>Ww?m%43V=_YR|L55z#bsq>ocxVpDaaCi7x zFNs~d2}CcXA6I{``XHb$ryY*fX*$@5A-jFg-1Zviu#G2fgjk{7%YX}uU=24U(Mt~n z z$;vIT;)eE+8x*Ktjm7Uls&3LMmv`}2E%tCMOqKF8B(&xkB5Fue#9nOvOr^Ow{Q0PJ z|1z+GV{V>9rwx)DB!OGCmI}AEZ!6aS{tcWX+jJgGoKt-Z!FEPC43L}f&RD<$%vE4! zvKIMP_9t>nnp2V6=b4PtBKQ>f4sgDUIYVvqAPoGFQ?%H_rS59G)dtUsJ7uq439a)D zWEbpsyG_fSaK9B^pmV!~-e3Lm{`R_lombUH9%`N$$>IF39+0bbmvD)s2aI?55 z#3@RWXJb@a#rW0(_gN5)zl%)zivt9*ZH&5YHxgVa(>`$9ri0Z(4q{BH*q7HH8uURK zTVFk#s#f33PCWiglgWuRPeDN^1Yf#(iO%{uC_dTi`YrtL&ZtfMgW0r68b7bUra3ru zK+j)3C*bqgzeHfnMTO3EXS*RKh@L4F`!BUm7&N{9^f7o6kIKm6 z*?#aLvq70c)ZYQR&r1!}_@lNdzK4v?2tb3fxGNmE#t{&S$2|m{iBR$bJeKR15+eE! zZ>jD&Nh_rm+&38a;2Kd6uI`{<%#|#JX4VAkvl1rC3F8Y<#+XKQ2BMQq^%2bNl zi)1iWf}2wQs~4#N)XOZOZh9py=Bj8DRN_3hKGDYFhdv#CoU-c$esM-g}Twr0m z`sE_+KXrgqg;YiPB5jdf)m*T;;Uz^CX_Zw)iwNCy)oeR&eYw?NTX8!fZ5P?(ZdFss zkNYJVxx1$<1K>}?1L5ZJA0I50h?{YZ4>?A6^VLO7=9uJdHneWiS8dqJP%vl($=pv* zk*HeFQc>+x5_*%jv#o@HgJm)SQlP57ub0>r|6Bpm8LC)yC1Ib27o!wr+nv;(Q7?@g z;yCD7$g;qI1+ofJaVNE;$C!h#H`m88RGdqxn09Uh4gHwL>tda>xfFrGtC6K)kHFpy#F_u;J^)<){M#mKy3jPuE8~%5aoqk zL~B=QHBCgVAvb*#g)A-Zb0Yc$X3vR+!n(9#B+`GX3C@wAO@QCJmljvurmznYvo31)dbs_o4T!P;eYc|mq@Gy+wO{SN`U`~L zY|8j(*uk&G+Sh+a1$K8tT%%cocqK1I842^<_{tb^qSp?xNvzPDbfhjJ%&`7DSCtvP zX2MEh^_;jyaKo1lVYUKcCc-ow($N%sbjRuW8*5YGj^U_lCxIxqZv(xqdiK86x97mU zvvxzCR_ZTTC*!Srx3g>bB-<2-Q2gvCg*-X+TTDC6$HF{NEg)S_D&q;`q7Jv|)nW5S z7`Moo7f0YnS;QSroRqRc$$c0GPmok$JClmcDWU{M*9a{@Ihjg@X8gg0Fo~25qfG4? zaHh_X<0uO7u2jXO0Fq|qLn?i}i=2Go6!iQRgkdJmFno%7L;wJ7YBX9IL|4K*(gZx% zMGBZCHGN=dp&&d!9JWBJl-hE^GbqBxms|~tSWRFp67nnz;d8&ihQK3Fl%PIWeO22z zTd2;vmMCg~M?Gi;)4St&i}`dhIPIg+Cg&F=ExJn3h2NA74n2-YG)CNsk|H_*?w&Z0 zuPRfWxkI;-;+i10dv5iGyo%szUS5JAz~q_CmlW7zaQPypAT(DpeU_l&_f|LyDx)nR zIz{Jsw_2vU1H}ots^>i1dGnq9K+G+70>%A|8>(^pP3lYptL9DxDx8C$E_l9EOR9-I z2#@k@nn2fjk;-o8>!C^^6zbEE%B8)#|DxpmN1^+V@?D0BNy$s{J6x_m-V=Cr9S{l! zk@8%}RM=Ip4YnmiF=cBKhbbY#oo1`uRHYcc;p+E`uJ33$`s>%nX+t_{G2GCV6w3ge zlmeLn(^ZOaJmiBs+~!uUJ~ZM{GmY0_GKM-CA&-GdtCQV}P7BlXpH%(ml)L^IsWFlC zQm|CAO=2?r>Pd|Wz~q{2j5KiD6G#!+=|`@GMWTh(<0|=`77#ZQs}`Z_cMPCmeT$Qc zlX(;^DzAU}Dy?33(u#KrsnDpDWtXan`+b$1-m)}`M&Bf9#c3D6C#gh(&F=sSUK-j3 ze*NfOug_4}g60KOv0k>x&y~R6zvIwJ@_q}|^+VM9Yg#Q*9xTKV{W%CEDqi`yr6TJw zTZgU4mNS6bZWKnHlU!3#kx!eDCF6V-ck#!)hIf^ugm6p&+C;Kx{OVt|wg6U^4Ekl< z$)DG1-oGR@1mOzMVk2|o5iv!Nx-PWx*^>F9X6h zORZ<@p-~QnJOs1&FpAQ^so(Ywx~{bH&XVP#>Mdt#p;5Mlyn5KDrA3mwR+7)s&~X2o zy_qw(36W?Jyv1Nq#A;aqT(k$?H~IW7)H`)vUaFll(JVYS4 zn=jMV@5JfrMd+`}td?jr&k41ak7r6ciR*VYOklP>OI0&UR6oo)Ee@>qmh1YGYna^e zb?C-j#vYs*0T0IMUmxJQ>tt(4k(nkkGNl{K@1MS$KF82FuukLz_)0>>&Ow95AU6kS znWPP8>OR-+SCxAV3snR%4JrfBr79k<=qJqFS2jNcTa}h#D(Qzw+s^;X+z-%bO?N%b z=waN{>ePt33F$V`60lqVn1h}4IAQtiO|Q_JH6q&p*?dwkV;tvqd{Wpy<}qrIg%U+P zww1F-RD55RI)3>HN!H#WC8NiXnj_pimvsO9MVI|RU6chY1HJ0k3nOV`FSK6OP@k1) zKKLa2tHUx;U)sv$&n2{+@!-=_!=@nJ`dE>;O;4C#M`4nyg|znbieK+9aT|Mqb-ET# z0y91yV0zePO5amF`W63GR^TfMA6j46yF;sM(;C2s`lUYmwfL}Fnahy7W|^WI~t?}j5&@m2Tz zP8O>b8_^XJK{v!5ilSQho_i9ahifK|w)G4jKviyjTXCki5yu88#fg(v$I2KRbgesO zuvlQMYuOwDsdNgpmb!1dm^#H=(ezooGVgJ1-jilB?WSnFNw|~IB9-hxJ*n1NCH)vm zUw!O0L)S>Z+Nodyxm@B~xu79wBsQ$7d+=EExS&CO)yJ7ilecm_qjGY^i%a;1ZJ{gy zaEYSp!hdYZoBvn2TZar9<+wQ@hdGWbHmGc%zcBF!WiYFTr`-I^QFDkYfrHNWOaaST zy7Mt3>tz8wfX8P7?FKI#7R+e*zm-M>-d!a}V!$7B;sc=r%+{p55DbmbdgZWn#fUpi z6MMqqf(l|;`&%i`vapS;F=1E#je|2FsguFCHrWG=!b(-Kt7(%|-k)}?f~a3Va9qv3 zmQU0xXcdKRW{nZM`fnZ}lDfI4Z35Z?y^`4@RyLC$m7!PUKX8oPLszRTC{plmlvK5; z1KiD3eiHa|dCI|T5rnkR;vMpCVek88$i<_~`x?@OuN$M=UyQstS@-?}v#GfNCVB`f zRIl0T&f@O<=N>i^RLDdD{~@6a`iAv}Ki_iu4zUK;Rq_V)0rriORA$38l|P7I%=zt= z_c%C8t21+-%I+{P@2nIj?XUnxCj#%^mBUd?=c>cr(Eg~w&Mnl%nWRm!hp0vnh!>?N z;u=c}@mJ4#M3V@SFlfGH>?R(_&R+C@=_3PHLoGTXXVBNi*m-X!CV0`7h5Jo{;iCxafAv>C93#Qk#SL|v@tF*7M-I(v0`ryKPEnL&`;V% z{3WHHubgF-nqmECEh!<5`KH>23F?2oWZ~^wAL~R!Fm;AshPxC!aE;=};;OF{{eR*X z8C;!`9fC&j*S-ZMTzq0sgtCtKOGR1zr;R2h6KUMREyLrlu9SKa%~`t3mD4^WX*k4!M^iO^jT%Lc z&B^Vf`J`it7VkVb>XJh1@qf$Vqilx+^w}3>dMD?mkt^%+c0oX6980~fXx#9AU7ke* zxn&4P+9t>$%xxd#JSQSyv9l3WU2d)SuAYdl zlY~Bzgbr(IECQFRnrI%JI8dAWhmi5qIO%>znU5w(DIw`jWo)9tRi&)ip>|_H%WQy_ zG>I7AnXA`G5(~RzKlAf(HNzP>8p2+&7N1C(3-7PItqzxD9QD09l~h^5H=gMfWR# zASI&On}XihV46Q-4;Q56HR2EFJ~(v%oHz?k>huVcA1G6M%_Y93K9y2eJo;SXr~$rF z)cwoalo5Bc7Q2-;$l}UE)DaHDK*d({CxP6}z7c#)wK|9z9@SE^>4oS~;*BVJTx+d2 z=|r>v6(gX>`|pQZ7J zZUFZCPC#lLo{!58B7F}ZXwT?^d<(zDwuEOf#AQbJd<&17pqZB7t@1V=s8<6QmOQ{- z7MN~@fzrPhOP0fMVH|wh2H~zsP4YGZL*Y~9DziF#s#*{_UouQmgaDz>Qj6pYPz2^3Clgb#l9Oj}^tOBXu*)kbXq61YRtylR&eD4<&mz z)v6zyYMr~d#@}=9+_@J!lV$SgpbB`A1GW02=O{X57-~ucb18;?5XHoJF)=JAQo#mQ z&@kl!Drgo~!5zhWtPE}+s*`^@2oXL;2jOe2)4>*+Q>`BApmk6OF;Ar%Ji@>);z?_H z(hUbQP&;Ci5)#1k3t%n<&sU;+7?clBijw69=jB9N*q|2bwo*U~iA|8tdIOUK2koXf zLfnrMNBCab#8Do~?Gt~;9u5%?ju5>WFdj~d%QSFV|KGeQly>oF!e_mX?3X!d)emIHk=Lq#e9uA1RNh<;EG{GhyUR<1K!8A zQzrb2rT`26wc&F%dZf!1Q%7V@ldh&S6+Vx)N!DWL$2eH%VhKxYclm0 z8tpHbdXfG@@!iElnI4_J55L{j-hrtf*E=wE>Kzp59jN`A(pC@7XfM5)$g`j?obeKO z*i2B5CXO3SoM_=YkspghM5sM9+uexXDY*nh-$IG;h4kFiLe{93XrQVp)oS4Waz1+g zd*2Z|s+E5b@}sm8zSlmjY~h|Mb`PKb+6SNi0yi8l@S?j%9!%!}N(IGn=2B|DhLkav zmFSzfAIE$v#BA-vdnoI<$9MseJT@wiIKA_{JQm+G<(a}hRwEA|wyQ#E+Q$i{X}3=( zTbRE{hQP|kfSLH5(;6^azr-oy1Zw(pTJ>Y%t^bd=>`mY~gD>HX=^PYQX9Oy15aFEg!QJNfzs~VSgMhgCu%P0}f4K z<_42kus;arevma$7QtIrB1X zn@s$*ccih!NuV0QB1i>jbAZ5ufmPLI`)&a~1y$9Hhm0qFjC^Py3`U`g$h%Fe*9`6r zP92AEkgS?9xTrD+Rm{yHU8V~{4WNdsa>>4^m$hFq-ev!j{O2L>#pC0vt6YKX?XiEm zmZobHLsmwHrx_zxcSjRlofso+qDiu5OeVVNNv0-ix-vBsQ?}HxDchRtP83&>j==Bt znH#$0po#Y z8qOA13P^t3E}>BwN=4DggVg{Uzm2jqq337g`rHJ^2H}Wv!&jm#h>o?=Q3QXPsYcag z6bJ4Q8ZiU~(I}$T02+V(J#(jt7v<(m&d*3O!@7%)SRfEqDA{o1YH_M3#nv^Rm;tP? zo!oiMczJVpr)j9>t4}+|V(hHkZ7=Es?MY=Z>Gtl?K_VGwz1C2tyKrYcoWf~H6A5+9Yq6QCko4e;^( z!u5k@=(svHv9UomVw&hHUKB*eTFJ;`3|+>dRxKC#al1rCh(J*)@?d{8GQe;Dn0|y6 z$U8KsS9cETiKDe})DWY5f@s}Okr>o{RjRso`Sb6)gixt?$5LvjWBK#@h7kImx|9v0 z(Bvq5PbjP(h3}c7R^6fe`F)2_h__!5g(6Z-n2aJ1y;}|0jMgV7C^iU1iyMtZVGtQ> zC1V<8CL2S!VxN&Xa)*ETh%qS0M-i?j<{AC}3RcQ6NRRwg7@?UW!KL@`h9@w0GFZRL z&ViLeQ1&odpVXis`>{(9&8_R-9nfr1KP?A3D@lLBqH}RA{-1wT2_d3MI4D&W{!<$2 z1AORV@ZX^`rkMCUj7BkuvxdJTPySrrilIHNuNxDccJYZ$n^q7JhFgSTT3gg#0H-UG zx{F>B!eUF5;G?uciQwBR_|n>?9@95l_0-{>uTtD4rJNPN&m)b0tCfqT!UrdLT<{{z zn7h8AYFAU8BmsX_FT6TgBrZf0Xerjks6#u~ZgqO9MK3hZS>?zGCQ&t3l2 z@~{pgn1`(B$_Rs&ixGZXgkM@n)?fIhYtTB~`<09Leq~ydcFLMGf_3AtrqyaaB{i;d z>n^56go>^9Fq`Qiw{P7W;kI$y(u%mg;`V)~ajO>Y72|(LX>A>Ww`Jg^Rd>BM;bqJ9 zx{F{Dnxd;x3|hWN7;YhkX<=W_=ZDwp0XvQ2G7AK0ePD4Mzrji3nfJ-Z4SItKn|OR* zN4O{;AvanRnU5i}3}k*bwzO)nX-vy9LRnO3X{h3M>T344___50{;YAy;PVZ#3&55O zTiWoWw1j_!Fx>%6)54a%5;|MeauIOmQ7pt}_#|!eQ|GdjE?^zETwUMEB=qG9{9VHYOH3nN-?2GLzgtVdXG7|WDuuXjLBGwhS^OgIIBprH*2lPpc6 ze>vwN+}U_J!)E9}{Ym`bj!)Xp!w|wefvCWp#OxX(gHyIi7|s%Z!4_`d%~*UNKEyZ0 zIXit^F+eeQ%pyD9=kd&m$6=E7iSv`kUo9iN69M7@i=g;;q&?oKgXMe#l@Sif-+=4mjyMbqG~iP;05eI>yxaS58GH93ocAG&K0={* zpIx{yk6mM-GW7;8(4C0QKk?S$apZXcTttG>Ac{=5_fn5qoO$-3C3)!CgT4#@fA;3> zEf>IeyzA>JR0q5&Cd0|_4qf(v_kWC`oT*G1$KTwW_{(`g*?}s32VHn)i!KJQ#DO-4Oz1!d1Tq3 zLAI^A@)6|skFIcZ)_f1WmMgr&Vh zJ9u!`ff+o#kF}@a2zA*>m$piYU0h1+mo!%4oXyyrJT7=SeS`?FDH?d6u`$WKEdk+$ z7|MoGI)bbFIDv?;csa)lf&lB5ic$02@%^gguX*PSPJs82;`3JE6Pq(XJ|;kl z8MNvdx_@Ea~wj>+$z1a-jH1Qnj{;HI~|`wAdLasD1!}z|kB+;a}q`0KG zy;0mBzlYmUsukKFzi+3&D#i6xA$cWvees=tchAyp<5&C;Z_-12#d%I*OJYl6OJZBt nOSG$+@_JXTd1%LJ7uu=FaOZ_38Kd-%|Mvd@yESq-d>2pqX z&rDT!J$>dbAFMeKOp71_>f3+E_s@&0wdlR!hQg~Os_#G0m{9+!+;|@kW=ECsHG6rr zmMNHQhypXKe3A9s?3ZV<7L7$o;v%JrBc`J3%hHU~ zN&-eu#S_D8v}aE36wHvkUe8&(^1}C#LZQZ#TzBmAWv*|x6#DYGnnyoESE5WE&?Ox3 z*o%#gWTkM}69xcbuI_?}OJ*ym_h8-h+%a1sygpv0x`t9_BOKy~f(TyH!zF8{3l+OPDl>sa{7ff|=EX>%co?o(baHq3pVFZ4|lnp|aG8zHTp*CHJ1U zPNPpcd%|UReM>O9UM%9z5#!H6d>=bDL+H$-P2(+Z{4Fn_uIEeyaNl z5ydb+Ih%wIVG1p9W&ZEedkHfIyPeoGna!f%UVY||1yJe5uc=zrLHl3dpm zE$l(E`@;k#?MvaJ*pf>v_siNzSL6-`hV>bASNi@#(c^i#T!0U_ z^@}j=dZz}APUHWu$5T{W7uGYd1)_OrpeX~t%Lal;>C27QwhEzRrI09^%t*?&Q=)hl+_`1Vu(ErqHoR)cNxP1GX)?jf5KJv9NZ_h z%=JgO*L3w4{%flOqPT(7Xf3*8I?F1R7a1h(XZ@?^eD zdrkMjL!cX`&n5oP1m5*F=sQ%JNk5@-lcVrs_JQ%H$-; zr~}2+V~iwdsO3IE+c#4k*k1bSlBOml0Fk^5u*gBg_wpMZ6T^qNK(;1NE{8ebf)u*( z?+Shx&)Detc|4EyvYb8!pTOW)}Dm%;7 zeOnm!Zb6(6g+l^aRq&3uqQ)JrzfvHdr)~cj#WiaArLkSlo;|zP_84e=1qBhhC$Tnx z0s;i2<%!Ov6Th^9jg=2rr&IEM1b;i6hm~BcJ z3IRIg(x%0kT1)cXiN21FuCpH^*x)Yab6=9HG^@OP84yeHy06_6^oT^I2yBD@OX>aX zf@%vqqz7r86m9q%!<}^qxakSPoK_x8Fn9&(V2deLTm*n z=R}v-w*O5CLj6c~G9UUNry|udVioLy;r8l{(Au1n5Y9CMmOh)&8_?|3-yQ?yOX&Qs zcl!e&7&qYwn&NTHnoF~1Z^962aw>(kv_O zMZGwbU)GCaW;x@Ub-M4Am+T=R@)-m8YhwjB5tS5g)N1e>D)^xsImZ(6ArTGygyQt~ z`OPWhNS)Ixdo)gCda1V(`ejvqj|q|MczDmibDXyW z4)zu!JtN@9+QLQ+H^CA7O>c@|pwzZ3#2X~Xf-&NOqWvX)gA)CnG2$AS;Fk1}q=U_m zJ__6oM#k1YNimF95cVz|AIuShYYU`??DjAo$gA=3F>gs?h4s_M4ejE5AU$Z8}2ace2l?v|0M-5Fk%mqqcPyvtlY+29*ulh%sU zxJP!YSaz#A|E{q_xtscU%#x1<*5KV)YRUrmdY!?>CAu zplt(=+xGkABCgp(GsyC%hBeaubT2Udsk(_+fgBVPZoZ0%earx0a*y`|uhDlR^q?O89PY*6I-_Fr(j z?L8U=Y-2`O6uV9+5~%)&Uz{;Ff$=fS{nq6FvT9&y3w$iJQ~LRVXrlg;&T*2l9j)K) z$Mh%9RD->yvaWGZ3o{w@@F*c(3C>U9`?2&pPd&PDCXrj7@;z7$X%Stfzi<9y&)BC^ z94WT73qv zs0JLBK~f@YkE^o-y%vOcerE`!RkDEV4=y@K`+l5#gHM)r8+;OB&otn>6}UzUR({sxRQKMd}*d&z}?(!b(WcTy6IwD-M+)^(UP8P zvZ^pKDXS3$|2$P!`$fTr(;%5m!ZphUPH|!hXCK~!D5d=PSxkZZ@o55JnF@PkjQBY5}&RX#(<1dz{ z3qq$(;<2Y<-kHchA(15nHQL|6g$!>3H>K)DLu3_Ma_GQ(9$ltBlaxS!M~b&v2J4WJ}aXaatM=>Wv)2 zjqB=ms#SbySsH(5>(3Er38UDs#B_&Ylp%OAOwfhH1ezyI^{!YIW=z;1^v#bdreR4( z()KK4`lJ)HMaYSywowqTX5II7suqOT`sP;}ltrHt$Y(jLmrw~U_dV6FmC>&hc3p-= zMQ>S(?pI7Ir;sDAzz}nl9Q?_bv_8wMUE^|7p5!i)x=%=LTOA4S;Z3MZGoTVseA@d- z)BQ(o5;&HWbn2S5`Dc@N8PAxhHI~0kSaa(cA{Yc>r_=THP<$4WbC@ld;n$|{j7n~K zT>;UzM0vI#O{_1f;aMHe-9p%IGD_AVuAEihA=G5j3)SNkt#D(MxJ1&Ljir`Qe|iN| zSufs0Qu-5ltqSV7gB30ckRTJ_X&(kEM0_J3vk*`kNUA1r z!U=Dbn|QK^IlA1u$xRXH(|uui?jlY;(J+`I*ndrtESv6)1^OY_KM)z#Jc>*ZkbiXx z9F12aI=uBroM4dWNz`(Ezsy+D|Gb8eNX}tcg#YXuJ5s}d4mQ{24U%R+mEBf-LQ>yB z55(}tgN`c-Lc{|k)HQn{gcAYbJ@)|Gl#u zXN`!NgC(;jC4t;sRvmT}6V0*;6)s$_)*(4}ut)I6rn2wv=;N%aLv3fahniJ91vEHH zy+f<_n1G<+8>Sej%lB5;9th!o?$OAhhCVG01Z6t4=?z*VT~aQX+h?9^K$V>q zYf<{b{o3qP$Kv(_`~#af{+GcnbO*D)P{S_rVddd8mx)iia&G#Yi6>$bE2V}0t1>)| zrImRx3Hs?iipGfV$2vhTJv~D;3~m%o{MUP=yVjCfC4kj&zMQ@R(lJWPDW{cSDb>)U zI3E;A^bD?27XE*KMFu)w(!EJF~{F4n0qP@88Xy%R)f5ZGRKNF4_2En`Ram+w#aC zW{&3vH8RWX=Xys;pHJ&g~76s}9&P~DafFu}!8yAwYxO+=k2}@=fkg9KhI3o)K>*XRl zU=@-VIj!Z$BpEADrvU)UF0jA_ZVGkQHS-oAP~Hr zIAhl}TAP^kK^9;F%!+N4ky_QUnUsqwFZ_gJ$cQlcsP&02^G?9k|Gc5tloz~v;1HX# zvQ_$JdtjVt+fny`b|$R4dK=6iI3vztYGD+e7$}Y+grdTU`T!O|6^ez<2H^n^)cI+MFOjO4K>iebsUU33Y5t5W%n+l7mjtY=tmPM24;D_Q8?|@GW ztBXo)hU4|iryTEy5j2Om_>RpwgUuYWg}AF`Z6Mh5pM=|7QILO40BUQzwzjVwH3QaW z)&9g5WOWYxk_(LSwW+qG{6&by{g?0mU zbDVGCK7$Su&BTagHJk)2+pih;5B#Rue6D>bbeFv;1ZqtX*&26rNNEK9Pd?25j7Ymq z8K`Fc?A-bZdYvF;8e!Y9KFuH6-|;>sj7w%SDP_Gv=VJlMcamQC2V9>h}}xxgA2=u}st3^x!8jXY4yPO5pkZ zz#;e4f%heAB{1-XMoHM1^)N!^VPT)XAD~&2 zWS+anr2NMY9O0`>M9cQ?mGG-EF<{n-aj14`LzL~X7y~&L2xl;Fvd1Im9a^d-@4m}2X{beyVQWJk%?IOg=Z52IbJ z0;2sE)p`nv{n3drF~5}8%*kvNLof8unN8zOjD8n>IS932D%f^50@=JBW+oI}|0ZHP zvAf=B&&%#jk>lUD-W~569p2Q{V$Qc&9bX5{iG$>d(bmVdJ5c<3y^yVY-S@1SXMb%p zEL4!lZv2|Ty~>?q+e5sj3Q&kaj+6+~Y!`JwFEo7fr>qBsAAqGBF(+NbC}!M}5?=@! zJF*wDmwD%T_eRGVh=fm&=fscaohO-pw`(;-DW`IYm^9MWl&z#5AzxA^@(|IGP;aLe zl5P1AxcT$Z(3p&!bl<&j!iXuVkjG+l&nSy@(DNcs&>AzWRb->J1Za(>7h%EDhEa1d z-{FlKW`X&A(u9a;J{!KFgDD0})Pd$AP1DwG*8Np2sZ{FM6za?IY8rF@-oze}@iKqPa!$?1u7=|8C?&iQRs zpZn4`sr*zQA(lz_P~e?0Tl+9ps_pj-Rg!c2eDvrFY8@IDMYGSHpVE{lzJi#^;(F9s z;++B77R6u5R8;u0{9G9p(qFDw!Dc7AtTG(|EIhH%Sr!`5@aTC&Gu3A-^U`w)OJ1EF z=@LHHtRljM)9dgIdRX?Gg-35yQ=8s@%pJEcpf@LtjV;Oz*@3K@6fDHQSgkuiM_(OP z$@!+gRN(vxo4Abz>8S$`g4 zM@{QrPjvmpIhT+IgBxy;15IN#HiJcMPL~a~K}ib|J(Bv{vh12Y%vPVYt0SU%e>Ue6 zoYmfBqusmoI*{WX0Y1y9O6cRw^sH3Vn|{&!xyJp}g$k<5_%H!U4yp=a=Y=ouF8E@+ zH$)d}sV($|sZASc5PGlSvOo}vB2qo{rXvFvHqzYXg=$EcuMjqRXm z&nbXVWKvG!{O`>7aAe!m*JQP}vu%20ZgS_`b~%ftaA2o6#VmC#*Q{IWw{M+!o7BI$ z7H^bj%vNe#7&AitJZkLZi)mpE*=jj=bQWjeabqR*{MAFpH|8kD{FHoQZd=7(EdvWM zZhrc6R-bR{>NKxF9(=eOUt#&O%xoeWC+lIA8qD4xvw$y4qw1(SHO=F}{Ceebr|sni@n< zO@3%EtD*+Bl{VU-A1q8?o5+qXnp*uVTCfI6kn|H>Xz#riMYxg%I@(}7G0c*%;Q&7M ze^4iwy)P&pnwm`kO%P+U$PUUBVp#7z9~(f*P*wp;I-F7mCn$;*#<-gq21jHk3ym#0 zr&me9xOJi}<2F!J?I2BI%n3X|Ew>Il;KH*J zt&tIHT&=OWv{!&4z-g+n3_N((U>jSdWcUx%Zu14A{rv@+Y;3J5KjEf9?_Ca-zP7Sg z(LqRN?I^GM4lbp^*Rw9BQ8~A#hBjv6Qie0uTEBGiuvYa%Zs{3FL%n1BXY~A)t zOxEs8Y_n;#hUQ8Gg|^(z&$J~NeFihowB@|j0*=z~8R8Ykutg42N0BB0`w~ur?a_UQ zBU!fh7K|q2pvDFzS^l~YIr*%90DXX|WzTA;wl?Zu9v4zW4Qec;x;$~PR!DFPRHuSFIjK7nVftRf$;h z=H3zqv+u~08n^ER-dF+i9!{s^)5c(xrfUVu&-iPv(w&OOC)TM*eWvDxX#F$49{4TN zH`~1>>a(0~0Tf?(*2a7MuEn$)X|YDJ>kRA_sII^*4D4mk7=7UD_!C**D)x2R6>IrU zM&K6S)56$Z3WgxaM&*p#7r2h6xNc*3Tdxd^r`SwZ^?P;n$sn-2Vl_!=StgG8m|F!xWWqxhO z+rsDutj|Z-QLi1_#r!V$#Orc5{eOq*?)vlXa;iDl4#n4PTOAOiKU(j4TbbK2ROMmz z7JpOj`bN#ocC3LtmDu`lbj;?PG@F13Bg>cBscUg{W1WTFWx~S^?u&@^)!qnm!Qq>9 z`$V?4GV&VufZz4-^&kf8T5qi3Dms9j%D~Eljpt);;T^M%$vvgBYx$qd$sxoS;Dno|p=K4{p6 z-A5AY#5|vSKqb7FKlmgS;AsMxU$HwSkv1Ft)~c{w(Fey9di#S#eW89_e4(H#kfF@! zSo|3Tn`zZpzC_LHzeItq^{Yf*UwmlX^USrZ|1}-dm}!cU0VQ-VMd~b6@BQ*@b_fIZ z+~hQB;{(^+e~8KKYSm5(naORKbIXzaH(jL6_ zE?O7J9?>2^9BngTb!JxEzEs&3M)~Db0Gmo!TboS}(8^M~TIw6$M%<#;w+wSg z2F}^jtEY4?cuOVt3oXh9e6Sy-b;L#G%`frPgpx?;jpH< z{mmE;Zk%cj2qfF=s12xei88~7No7pJy-NG4eYebJ5tt_m|A8MuWG(?5ltdxWH z`=auK_VXXJUNu74`@^N;DD(2rOp|niCL%UM{)=KF@r3a?2v2AkNjaF8P%&p5ut2xB znT~+gkYhJTRiB^_%oHkBSuQFNCi1P~A}w)K8@?74$-%{+6Q=k$BPK^}PpymO!v(oJ zo%lRwamDgF0E~xF#xQP%htQ*0G#PI$THSBw==fA;ahdbF2xv}i>nYwkVHqa#G9e>J ztMK$%Y%2*_Ii~LTbZ0SH>ZW79Dj>PZbOnFO%g67id!I8>30W93eRU~w{h|Z3q9=Dr z+19`0w<`;Q-h=QqGc~u^R~vZ&^MMby76o|E;`Ey66lCK(NmCftoCJzYhuOdHLpn%I zy${C+*a2Tjg!%(#i-{0j&nAQnlmU=Q%LThx6tiLScu|9!?*zCd!;Nh3Ik1)9I{9)u zeLjPCL~Ye>Jl(vPW`+4|u$=57wRZAWXE!5oMaOZhmU$XWp!XbFZ_L+tkqgyqq)vm~ z?6+44EFUxNe-IaPx3=;EJaCCr-kzXkyYba8nU-fOP)1TcUZ$>TX}=qEeG5yE)=4H0 zmXOc|G)9Ns-$uu|H9p{c{6dB6{ODq?P2v+>u4|Om3U!ng98vBA#HUu5A^-&4un%5_ z*AVX54L!I;Hh)-K2`~i(-ypF?trBLw3N9Y@6EHt~ZK_j$H^N`RjcPJ@3&822KpLTW z+c(6gWFHA9cB_QfBRRdio&z((LD$zLSdw49XMsZ%K~=UpWd*Xf_gGzAET!LHT!`Od zxNc~e4Be8;<96y-mm#`e#XSzM4?Cq1)Im0Tg)WR|ke};+zR8bgvkHk{%_F+c+*%0o znDdFSw@sEsAtg-}p$8EPDx6!h_WaBYoZK)0bHSW^CH`cz!D^59n#Rig9<2Pedmrv1 z@Ewm*4`1Hf5)uf9N}`!WGgr&1nWU@Ds5_#h>c%p|IA-52uc>Uwl{Q)l=b0NR zqkJRtdp}xs4F@k5r$>VSTWPQQXhU zSgGr&c&SWHGAVbamm-wIppwA1@~r?Bf0f-gY9eAa8M5oTd=zuFFqGu@kCH$aL+V@} zYjvU|!XyhJvU(Ki+&HOBZ89kiqc8Hq7x|@pE%=L^2c0x6T^uM!?O>ZkX?S>&B?SdO z^`itB_)xAG`d@ooO`xI?jbW@*raIZ^Q|uS}_KW>dJ{0y}D{@k`bTNcSrq;Q`3yjf` zC`jj$W5Vh{U>$KDONME_THd8;ANRtF3-Q|YoA00Lc=;3D{FIPXDA~cxv)gI5C;s0b z%I!8P5l?xtrBODRxzx>yt$C;`Yrr$O zgiY*VS0birGyZExktGqM$_|kl@H%`B8jR1tb?clc>cBfP<50o-CnBKSB zHo!U)jDw%#=-l$WlA_aqdea@yQ!j!_go{aCZGg|RD% z{|pS( z5gfklVq!2`Qr={p#e4A|it=^OTVmlf)3dpfK=A4emtG3k2h*~@7ZXaia0xDE--RN= zG1%=C@pVfTT!Sz^65&6bQeN7JFKv_l!aeE)Ep0|8*hA26Th8$b6}sZ*xu_A8RL_YM zQ(0Bb6yZ~O?nDExyZ@in14M}{R|#JpxF>o*9u_(|{sVsB2A`bQTPh4ibx?FgS+`j!QP#l);aGK<#2^k4Zf8JoU z$BqIgwg=7hYiag}v=g)%LP(sOs`&C%)6S(?nG%}J0}HJPS%4mlPU4$z`cJR)%fpA$ z^&LQ{&yRy2zWWFC^wjs$WOiiA{x38{ z4MsYIaMz+7iQrJJTnqSvqZ}xL69Afrb2v~Cna|TwfyaQdnT67puhxzi?WD$cLSa5T zByUll(rO^LSQiTokr$(jgK{*>pO`nZ5EShGpYG$ zcyc{rrgsZEQeMg zz6aMxGEomuU@)3ZyzqEz*t>gFC2Mx^j7zyj@pjmC`eYnld+?&UjujP7XCgD)D?wW? zu9~{v8&*wROFhBe$lw&|Y4S9LqnZG9!b9%vf8^K3A{>fxUBhiOMAfi16o&K0^Nuj9 z$}Z4+Yc&Ei=8lF_pS3cwIZyqPFRaXuVUhoiehs$}Y#V5W`{(8ANMTm;P;gwARsEHA z-zJAoUDg9$Q1bryhPwNm8FFjw>Hun6ey(ONUJxYC>*M6Z-1OynLyS_iuY(1?c}1nm z;;@XNuUMYYCNe$Ku3jH^&o@1CfSGoQbZQ-%d5w3eik|;1@E9<;*+wz z!1XxC5FH|0f(-xx*p*i-KbE742mDH5w!D9UV=$aPHc{-wIx_@qfMsBf44*LhL&P?z zKm^cjYHy9&bx8>)aJ_peBaM(bB8_aE-C{n>DW!Cw>-wI(B^J{0*SrY5zM73}`f7dq zeRi&$vbMr+@Vju**Xy=N5^^pAznVY6PlcE;xsi+z%lSqRs3(w^E1UJ+0liy~g1uXR z>GA+ug_=9*3*dTvc&3ug@-ey(a(zH=bG8Ge2#k5Y);hFwyn|c27B_8WJ7>-IoKHb6 z-pBf@2ZZv?mo#Lw(LW&;@*9fAbrM&x8!@ z^`(0>3IC(7_huX-PTxWi(S+J9tH?Mpl$!-3HBHjGvkn|LC8fq_J%a|EJ-y=1_hc!8T)QXOsQ*tnW!D4zATbhI)SaN|ANJUZlt9q^b zGx1Ob;L1-Fg!dMWNfr}L5GhQl=1mFV<_o-fs?n|queeHAIlWbt(l49$Qd}Z*MonH| zCqKoJWnjzEuhrYXNmJ73&IW857hI(6xr+ZY={Qed(X=b+5wZ>G5#k1c%X1%hw}Hfs zcY<6n)4Jo+cB@&itSz4|L)`TLEKg!G6A&cd74qGPL0k5=u;%&^4+o(e+ST9#egIz) zmNlJ%Xp-iukq1ZU&Dwqo%rI`+w$o14_A#!A_#75)!Q4$! zD_LMf*B8o7s36@QEbZXWSk7kY0$?RVY7nAjeIbEr0kcn18|32_!^!5sS$4x>$CbJ= zi+*(r|6R@tV|~N`_(KBbcDO-s=|8SIiMvg2iJ017ilS|yG#J2C7oimbK>84*8P@V6&|BIg=fo}Xor|^)2 zA-%CF0zII?9+|!#KNv%Qsxc*|iPq*?{$_E!Ud47I)rB#7VLvAqH<1S{nX;PUHsAuM z@q{9sdHcf@tI@zy2z5!|WFl$_H{wV@D8wAS)ozLlJEE)40-h=P$gDXL-mELg#fPjLj zd7^^)$XO{%>_vhDbT;qnA1-mqBGF9qyyAI&ojeS)76ET1Kr*V5btW$5nyqE> z@Ww_Nr93w#(u_r|xoIyJOv&pr^9e;mWvJg3RmxIz?@&+&0%Z3J{g`%+GUcUoOB4Cu zdTyZgh!OZc0@)axb+-FFE!<}ndK{a%qy5Z=*D;MEOHFx{6Xpe|UeXR!!|020%3Q z3xWbbDRT{1J<*Yfm)k3SS2EoyMUh08R4q^{kaCeuqRxl0rvI-lT~J!CGy}K&Cu+Gs zTa~-9PiTJA+Ewt@PW?$F&rCx-g=EDk@8 z3RuO*OYDY!q6m2pMXatCvkl z4MAZEu|O2BrAHicv-=mkkU7oZBsI8Q5zq~+d^Na`u$4%|2A!rhm(}04vk1ge|12-M zgMWMHyEfp$-!;-A`gTd(^}C}y zWBHP>6A`NtYU`rE$%oB5VJ7T)L)9A~rTKL}Ui)-si#*wqbpo1xTnw?V_(N?0bqCzS zSiQ6UJ`iCdDz;_KCr^p}cAP?L2wSDie*mk4{qLUIq}Xq!G^7SEsSAWBya_Pob1>#y zO!K~tRWT=zjP~w`OZ;byJH6|Xm|la+Fb(x%kM&OdTW&p#OLFx5V;sGdC-SY9R*}Or zGw&E;mz;QV6f|G)4K(1XB{fiRf-Yay3+6#1YSpLN_K7fRmNP37KSD+HIa4B^vS`jj zBpg?WRA~c~s^H&GlJHIAq*BT;)DkpPPp%lLIOK4Z(pPGivgE1e+*01flJI!`I61}d z#dpVOoShOybOJ?$fu=4nowNG6rPQ=&(C7uhTrmo9BT<2;D1lPc^eR9lP*7=JpdzVC zBFia{kZ30#N+~Q-DgK4%cNdZ1&uh&_g!buN1Wko9n+nb;&@%5b;@}Q0p!X=eW5dfX z_1|f)3{b04o&hxpno7vE%d8PD87@~WT*A4yC=3DqCNHzIBxRYIL%*2fH$f)1g8BnF z73h3=S`tKE`x?g=w*{pg%uhMEUXX%G+41-L4~1@cFR&$z0AS?Lg!_ux94^p$;MzoywB2E3 zisV2Wn%xEy)m)I~qkvB(pUFbl`_4V&CF$U4e>GQJL=|*gd`NOQ7b&8tFfd-E(Vcun zc)+t<29Lu`98EfSIyr$x?^j1sd3d`wspkG6ca=$8O$_Oo5UDhqlytJqKlNJb+Q&GP zxSp%_c%Y~re|&jFY)-E5rNevT)S`4 zf}TGjlSuIVgz371%a57Vi57Z6$%DXMM4++_pf~sQD zh+F~ZvxMt#n|juHta6M=5qRxiEz?JD^@{8Wxq`3<3Fl!S^{nIArD%~N@H}HnQ_-PW zPdct7@+p!TB4%BeJmE!7lq|Z?XcaBzmkQ1Wvwo5ZB5NI&bYU@$*-QirgutTwLtr<% zXC+rsVGGFuQJJ<&sqi41QWjmL^ZYCcUc=wleDEOnH+jp+I*c|@Y#D;@P(=I?O#?O8 zm!<0xkKYxrN$caYl+TT6X2>k)wi)qj0IyE7LaNRpzN8|pVm1hAgsYGp-{>_VxUm;_ z%WhMrVYVPl=2gz|^{UwSrRccDfT!hG7o;tv!)IcSqKi#+W9Vr+5|wn~m7}Lg<&TP0 z)U_jVdU4j5=boZ`T(e~nGXPCGcDzusY2l92VMB3G(creaGjz>si7GnLDw~4yIe_*F zNJrIllB9*WYIEfveBFyw9g{@e)3W7s$HqXlk}n>>^qi$zIs72*=)&j-G)?yL{IRuM zpdJ^Jav(cNsdffW`jEl2n4?t(4_1}bk$|cPvK+IiOzqUM^<+QZ! z+duslSbhg{8+7JX$VO~n=NQaTv!#8f^si4#nB_YHDWdi($|>WLPM>03hAt3^h6m&r zw0Ioz&!<;qeZp#VdC$bzNr-{~8-9a8lEx;`#iHhl+$6KnD_Qjhi&PaUE7$M?cp;;a z*XxQcDCuH0t}Kn&#+IMLk5*2~VGc z&F!$c6!QkN8Z@r_`x%=9*X&^7@3QUG;9tFGHnQkp$XS1`dO4^nD&!u!QIty)&ExPl z*YS64GusNrClwGkyD=|ic&}MAzsS^iokLB$F^uN9*l(z7s8!_zAueA!k*17Km$(e& zM9JJ`gfh!%S;31&I_Z)~vQ~t%<}`uKQ8SLlMJ?GR)L>=kDtXh`u+$kOi9#;rs|sR> z6O9ZkY3nu7HU-hIr~zF{Ry!yqQz*w&eOg5>*=7q6&VccHuH7bQy!Z^NZMtO8s0VeW z@~Oia;=S{RhDuVf2|{UQJcTFbPTGAX2+b8&M>AM16D>}dX%C9YrMR7@&`(kQf!}J* zu6*4e7vMb>e@O`rGb7y++`_DnDhP*{39nH0Ur>m8(A2Uc&dsRC7Bso%x2uR;PVVP$ z3D`QiAOltu|c8hy5w#rJ`1LrMdeEUmy;&3T-~;}Katzo z#xG0P1$m`^iJo6d`4@wqlYhe1wDp~<)yGRCxEE8=qV=>fQ{^CZXYi0iTp{#nCq+5p z*+n{Z$&yQko^)0x_FvS5u|F95wy)XCQmFrCr?Y^+Ss06kWF(-Fg93y7 z`f1J&);IcO?db#TchHxO8~pRzH&kS@OC}k7A^bef51*{NuNO}af7Amoh~sz2};QU0@f%XdI4H-_--WX{B0 z@)s~&=>>+bz)w9f*R-%|dmXbq5Apt=wztTSDlB2%cZ?Lvk8t>{Uzf2kHr`ls^&;3DHOsu2sau1b`)rhoSbr^-;yo9cr0^GqEL<8)%bV!qucLq&~7#PEJiT3Bm_ zsiImEAn_GezJ3O2rUOEWvRxD0q!<@Kl!()FK2nVTqHREENm^bwEdIDjJG4K~)QBiS zBlauiGxN3SdEv=+H|^X|!bHdVg@|ehzoJ-^#-^@aW7(@PcQOS|ZeC=wa6mLo*c2_4 z#bHi!@p`M%W2p1x#4Fh+K^J&;v4-|s6>URf#&69B@z&pQ3)2uk?Gzpe3YkJXlqPS`g#V)^8x1jzMmS^Yg5Z z{RsaT>q8j#8&0Oo5Y4E94+8gT4n>RX!3 z=k#VxyzUq_!b;?|woZNrHnUa=Gds~Tj`4svIDHCOP%^L!Q1 zb|Qfwdx~=}fwq7c&gx<#NIj=Q1@X)hoH)T|Y)zKM+NT6cvfzUL|FSG6hQ^+hh%RTE zocq@R$`EZ+uW|%FUImID3X6}IBMt)JY814JfWN(|SNeW1zV!THBw>F#?#9vZb-NxI zSe9((v|W>PDuqAEZdvSLSKzWy<$BWGBLlYSp(D|TSSmrXE>IE%4iJd*X)h1K>^VVR z7Agom#UM*PX`Ln3M$Hw8Ar{F;2}@!BY&SB%yN(|i4=PK4raR68@oSMRgXAZw&Dy&F#d8KBfk5UUOqK;5Vt+Mi!$Hf zQ38N7FID2FgrU!=RNNecZ;HX;&W^4eADtebq*a*mmHFG@s0%q0sSh|4F`zsFGtD_? z=p!&w_j}3-JiJtSNI5~Vy00Q=zFy7z1C{1WcA1WwGl|}|tlzyeZ+^wK-yu+OC zyPn_!!nMUu&C-e9LrtLDHMA)Hn86ZAJ=mA3@^dV3xj9629j&)G(YyYa4iH4YCWf@S z7oSpe9%YzMggR7$dnXJJyD>7-kjeLlFU>I={h=~VQ|`|l{-}Yy&Zj~FbU`Bajzs8P z(kcp|Whv6)ofkaIK`_deVs{Vxqu-yfc$<%dV3yCqSsrE`%wsC{5{Mo2xn@}GTiF|JP6 z&X49tsFWd9SE=|xvwX%Bpy&~WA|2eoiHjQD{|N;R`tl@+jY;Ay@JxMf9b<;QqANFc zmhsWcH+CJZlOlKvkuI--qJ-KfgDspfm3z31Lq52SBZ=Gz=4D@}WLL>x)0htCvf_LY z1(VC1&HWH6Icu7hL&LI)$YJsM(t|~P^adse4%$s|gt#9ij_|#pXqgu(_EZd`fIr*b>RrXBdIN39e_Q_rg zpE)W$wB;zq;q>H7kcB9Qra$w&sWZMyNbQg<8z8{9;lw@_kyAw4&>kTt3$8mOvDwHmm;oR6OW-gm@~Y9)V!{3xx2@3l`WTexS6 z-NWa<_QB`BzzxR>yy)(c2h(|gQbBQ?xs;l(A!W>ECHiLW$1&duFd zVm^8$?Vba6RL%7x{ai7Q}xzbhj)JyyK^sX2B59$UR5*t0&UnG zavlvQUObNc1s-p}X4o~1J%pjiFGG@XF3I@OOVV2cEwFL5%W#JjU{%5Ac-E+fI}0QxxpkB><_}ZA7o7w#vN0i7bc=I z;okscz=3}ts6GHRPH;UT!%d2dBY%VNPc9%@&b$oUCKG?{9cgTF5~v2S2vPys93b#u zU{!V5zFUA#K~?qQA>)Z3BOe+FgHh-r@@~`WHG_MDQ^z43B&%i&E~*Sd6?1b)m+68~ z1E?XZT(U3fW$l-YciI0W|9QxJ@%Z@aDpw$Td+dL%rRmzlkd=|)X~xLa-O)r>C&oyd zXp*cMlZkG6lBvm>u1pQZlr42^%C;uE6U9}eBk=qEW=Va)0R+YU5=&H6(x4QGoh1tdRim(ZvTrJ`u$!D;}F z-$q%Q(DSo#eQttdgK)&T;VV%VM8{g`D1v{?RHJG#iUW5DjTnN0XcW{3LJ_GZOh%E1-mQjgM(dLk6dQ!1#f?UyFo=w`k}-`klZ~NVvCl{xxkG<^#26Ih zqX<_MbC7;;=PtsiA+yoO>4}jIGScRAky&vNCF`A!%;L;sZD?0KCz&0)OQeJ-6eT4O z7%lqQH&clf1uJD3q(}ZLjL=Mx;L>||!xNZ08LVGr=fFxKD0>*KPioMR{n#al=GOJ^ z4rsQhpOyoim88F5(Yd%5|4)Cagb>jr9F(dG|0xai0Y3CF_`jhurkMC|7>!~QXAS?2 zJo$5dD~9&8zHUr(+Qlb2ZCXJ@7;X`UX>Cz|0i3Q#>MnXk2#YOIf{)S)C4z6O;7etN$9S6!PLoczYv(y5~_=#dOO!#b$Wk_L@z!?qSG2I z>Z!L3?zB3q&t3J^vaSvzn1`(BiUxz0c@chFgkM?!)?fIhYsWg>`<09Leq~xqcFI~Z zf_3Atrd4J=B{i;3>n^56go>@wFq`Qiw{N`~;kI$y(n_|z;`V)~ajO>Y72`)~O&o!@ zW#FaNa=kXsWy|cki(r2dnxd;o3|e+a7;YhkX+d7k=ZDwp_&SZ^G7I}?U0-n=zrji3 znfJ-ZjdgX-vy9LRnO3X{h2h=xX-2___50 z{;YAS-}4Q!3&55OTiWoWwDg29-2qI~f|R}zI$Mo$5pd>FEX03i_#`b$sdHIM7qE_7 zrLON}5_q^dp$tl4Wl)+@hD&zhQ769z za3_CWKY30Fr-|}IGceyrfSvG@7h&15KfM0H>qzXv=dn}?%}1eG0W?1oS(+qooY2S= z{@jxPi$B?vhYFM2O&Jb|M_6tc%akIocR)@vL-IG^`naJjzK2lXoQ!Zx1Ud6=@55#6-G^}AhcNo+O(0e- z+?dC%u}qkHgBR#d#NM5FYw>af4Z%ap9dteTw>NKZ zxd24tU0+YbFu|*0GMo(W&}AQZ_r(~>nMjl|?9IK2znm9T4XBZScc9?2>O~`s4O_T@ z2jw#LBd8xjNp@?4XV_y!2hN+*(L}aKs%^O=do;0(v2NPdMDh;3`Ge1tZDbaS*oBO5Pgc;&hB{eoq|VLVHyo2)DE1JG-;KnHA`Z$m6?HtbXO zBn8Xhm13dRz}K9A^lSOHJX{2KM+(cqV{U3kfu_!MYpA}gR3C~0=7~&%3omdsUfiJ; zKje$)d0j}M`sle)L2#1Mf4o z1(`Pm0JLB^jWD_|D*U7)xVn!MhzN_9bG#r3ux_asHP0R2uS))!cYWXlcn>N4ZUz0Y zHS*(Q_Mn(Sqr8JSEQ1LIe}j`l_Bs(um9mulc&v@PN_|LB(%FgO(>k9Ra2d8_9-a*p z$9T!Ip)wzT;?A=fjiBnlYwAETq*d`}U#NUw_0eZb_TMd~NEi!iuo@f43@@Iuy`tB|6S zqECW+iB+2O|%jx$ka-eofQni+WHI~{bl&IxzMTHV#Lqgegq5RQJ9R%N3mfx5J zlLV6lvp0hIn4c#|IDD~@B*O43Tw qO47>0Hke)2l-IjzeJ?vs(@sH!J1?xL7^Q#wxBm|}9o+uAAO`^TNiz2U From 1e162115dd0242630462a56a500b0db9e86df7bb Mon Sep 17 00:00:00 2001 From: Claude Date: Wed, 4 Nov 2026 18:25:57 +0000 Subject: [PATCH 24/52] uk_hmt_costings: apply review A's C-1 to C-7 and T-3's NOTES C-1: the landlord fee's receipts (rows 585-590) are a component of the fiscally neutral measure. C-2, C-3, C-6: notes corrected. C-4, C-5, C-7: NOTES follows rules 1 to 10, tallies the 2 context rows and points to the staging branch. NOTES no longer reasons from Table 6.1's hidden fiscal-head check column. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_01LwQq6JvunwtJCAckL6L4aM --- .../uk_hmt_costings/NOTES.md | 36 ++++++++++++------ .../uk_hmt_costings/claims_staged.jsonl.gz | Bin 63405 -> 63483 bytes 2 files changed, 25 insertions(+), 11 deletions(-) diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_costings/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_costings/NOTES.md index 77ce1fab2..34a7ad8fd 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_costings/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_costings/NOTES.md @@ -2,7 +2,7 @@ **Working files.** `stage.py`, the file meant wherever these notes say `stage.py`, lives on branch `uk/ab2026-harvest-staging` at `harvest-staging/ab2026/lanes/hmt-costings/uk_hmt_costings/stage.py`; this branch carries only NOTES.md, `manifest.jsonl` and `claims_staged.jsonl.gz`. The repository carries no primary bytes: fetch the costings PDF from its `manifest.jsonl` URL and check its sha256. -This family dates from 2026-10-29 and follows README rules 1 to 9, including rule 8 ("stage estimates, not design") and rule 9 ("a component of one measure keeps that measure's key"). `stage.py` in this directory holds the rows as data and rewrites `claims_staged.jsonl.gz` byte for byte. +This family dates from 2026-10-29 and follows README rules 1 to 10, including rule 8 ("stage estimates, not design"), rule 9 ("a component of one measure keeps that measure's key") and rule 10 (below, under Conventions). `stage.py` (on the staging branch; see 'Working files') holds the rows as data and rewrites `claims_staged.jsonl.gz` byte for byte. Source slug `hm_treasury` and source_model `hmt_costing_obr_certified`, as in the AB2025 family. The exception is the two tax gap rows, which use `administrative_data`. @@ -19,7 +19,8 @@ Source slug `hm_treasury` and source_model `hmt_costing_obr_certified`, as in th - **Measure keys:** 76 registry keys on 573 rows. - 416 rows cost a whole Table 6.1 line: `conditions.row_type: measure`, 68 keys. - 157 rows cost a printed part of one line: `row_type: measure_component`, 8 lines' keys. - - 22 rows carry `measure_key: null` and one of 3 proposed keys, for measures with no Table 6.1 line or none this family can identify. + - 22 rows carry `measure_key: null` and one of 3 proposed keys, for measures with no Table 6.1 line or none this family can identify. 6 of them (p.108's receipts) are `measure_component`, so 163 rows in all are components. + - 2 rows (p.50's UK tax gap) are context: `measure_key: null` and no proposed key. ## Access recipe @@ -104,7 +105,7 @@ Source slug `hm_treasury` and source_model `hmt_costing_obr_certified`, as in th - **Metric choices that could go either way:** - p.15's population is the people a new benefit would cover, so it takes the registered `eligible_count`. - p.108's 4.9 million is the base the fee falls on, so it takes `tax_base`, as AB2025's 165,000 HVCTS properties did. - - p.84's £20 million is the council's estimated response to the measure ("Additional business rates income is assumed to facilitate borrowing", and "none is assumed without the zone"). So it is a `behavioural_parameter` in pounds, as uk_obr_tables stages costing assumptions, not a financing remit. + - p.84's £20 million is the council's estimated response to the measure ("Additional business rates income is assumed to facilitate borrowing", and "none is assumed without the zone"). So it is a `behavioural_parameter` in pounds, as the AB2025 uk_obr_tables family staged costing assumptions (its £1,849 row), not a financing remit. - **PSND and PSNFL figures.** - They sum over "the period to 2031-32", so they carry `value_kind: cumulative`, `period: 2032`, `conditions.horizon` and no `fy`. - The document prints them unsigned, with words. The family gives all seven one convention, matching the Exchequer rows: positive = lower PSND or PSNFL (an improvement); negative = higher (a cost). @@ -127,6 +128,7 @@ Source slug `hm_treasury` and source_model `hmt_costing_obr_certified`, as in th - Table 6.1 prints the same values with the same signs. - Values are NOT re-signed. - Text rows carry their own `sign_convention`: as printed for p.58's "+£130m", and for the PSND and PSNFL rows as described above. +- **Rule 10 (borrowing metric).** pp.34, 35 and 41 say their tables' figures "impact public sector net borrowing (PSNB)". They are single measures' Exchequer impacts, which Table 6.1 frames the same way (red book para 6.1), and README rule 10 keeps single measures on `revenue_change` unless the producer states the Budget's effect on borrowing. So rows 134–145 and 179–184 keep `revenue_change`; p.35 equals line 30. No row in this family takes `decisions_effect_on_borrowing`. - **period** is the FY end year (2026-27 → 2027), with `conditions.fy` holding the label. - **geography** is "UK" on every row, because these are Exchequer impacts. `conditions.measure_scope` names the territory where the note or its title says the measure applies: - England: pp.20, 21, 22, 36, 69, 96, 102 and 108. @@ -164,12 +166,12 @@ The document orders its costing notes by Table 6.1 line. Mapping each note to it | 84 | `ab2026__waste_crime_landfill_tax_clearances` | p.92 six priority site clearances; p.93 local authority pilot | | 85 | `ab2026__further_priorities_nhs_and_vawg` | p.94 Erasmus+ (by position; see below) | -- **Proposed keys (3 keys, 22 rows, `row_type: measure`).** Rule 9 keeps proposed keys "for scopes that cross Table 6.1 lines, or have none". The registry adds them before ingest; the table gives each key's scope as the producer's own words. +- **Proposed keys (3 keys, 22 rows; `row_type: measure`, except p.108's six receipts rows, which are `measure_component`).** Rule 9 keeps proposed keys "for scopes that cross Table 6.1 lines, or have none". The registry adds them before ingest; the table gives each key's scope as the producer's own words. | proposed key | rows | page | scope | |---|--:|---|---| | `ab2026__british_business_bank_defence_commitments_accelerated` | 9 | 41 | No identifiable Table 6.1 line (see below). "British Business Bank: Accelerate £450m of commitments to support defence-related and dual-use businesses" By Table 6.1's footnote 2 and its £10m bar, this costing is very probably inside a Table 6.1 line, most plausibly line 41 (lines 29 and 44 are fully accounted for by their own notes); tranche 2 must never add it to Table 6.1 lines or totals. | -| `ab2026__private_landlord_register_fee_65` | 7 | 108 | Not on Table 6.1 (footnote 2). "Renters' Rights: Roll out a £65 fee for the 'Register your rental property' service for private landlords and private properties in England from 15 December 2026" | +| `ab2026__private_landlord_register_fee_65` | 7 | 108 | Not on Table 6.1 (footnote 2). The six Exchequer rows are the fee's receipts, a component (`component: "Receipts (fee income; offset by increased DEL spending, fiscally neutral)"`): p.108's note says "this is offset by increased DEL spending so is fiscally neutral", and the OBR stages the measure's total as 0 (review A, C-1). The seventh row is the 4.9 million tax base. "Renters' Rights: Roll out a £65 fee for the 'Register your rental property' service for private landlords and private properties in England from 15 December 2026" | | `ab2026__pwlb_debt_relief_councils_severe_distress` | 6 | 109 | Not on Table 6.1 (footnote 2). "Public Works Loan Board (PWLB) Debt Relief for Councils in Severe Financial Distress" | How I settled each line: @@ -189,7 +191,7 @@ How I settled each line: - Table 6.1's footnote 2 leaves out only two measures, neither of them Erasmus+, and the note's figures exceed the footnote's £10m bar in four of the six years. - `conditions.document_position` records the position. - **p.41.** Position narrows the note to lines 40-43, the national-security block. - - Lines 40, 42 and 43 carry DEL heads (TDEL, RDEL, TDEL), outside the document's tax-and-AME scope. Line 41 alone has a SUBTOTAL head. + - Table 6.1 prints lines 40-43 under its Spend head and prints no finer fiscal head, so nothing printed singles out one line. (The workbook's hidden column D, "Fiscal Head check (hide)", is not printed, and this family no longer reasons from it; review A, T-3.) - But line 41's title, "Defence Investment Plan: Reallocate capital spending to fully fund the Defence Investment Plan", does not name the British Business Bank, and no figure ties the two. - So the row carries a proposed key and `document_position`, not line 41's key. By Table 6.1's footnote 2 and its £10m bar, this costing is very probably inside a Table 6.1 line, most plausibly line 41 (lines 29 and 44 are fully accounted for by their own notes); tranche 2 must never add it to Table 6.1 lines or totals. - **pp.108-109.** Table 6.1's footnote 2 reads "Measures with no net increase to borrowing are not included, these are debt relief for councils in severe financial distress through the Public Works Loan Board and a landlord registration fee for the Private Rented Sector." @@ -212,7 +214,7 @@ No `row_type: measure` row carries a registry key whose title covers more or les - **Table 6.2:** p.58's "+£130m estimated in 2029-30 in cash terms" equals Table 6.2 line 9 (DWP Fraud and Error) for 2029-30. - **Lines without a costing note:** 22 Table 6.1 lines have none: 1, 4, 9, 20, 22, 23, 24, 25, 31, 32, 34, 35, 40, 41, 42, 43, 63, 64, 67, 73, 74 and 81. - The introduction limits the document to "tax and annually managed expenditure (AME) policy decisions". - - Table 6.1 gives these lines an RDEL, CDEL or TDEL head, a SUBTOTAL head, or none. + - Table 6.1 prints all 22 under its Spend head. It prints no finer fiscal head, so whether each is departmental spending (outside the document's scope) is not printed; no note costs them. ## Things in the primary that look wrong or contradictory (reported, not fixed) @@ -223,10 +225,10 @@ No `row_type: measure` row carries a registry key whose title covers more or les - p.33 and p.109 (+0m in all six years; p.33's PSND and PSNFL figures are not zero). Table 6.1's footnote 2 sets the same bar for its lines: "AME and tax costings with an impact of less than +-£10 million in every year are not included". Yet two such lines appear: - - line 48 (0, then +5 a year; Spend, no fiscal head); - - line 84 (0, -5, -5, -5, 0, 0; Spend SUBTOTAL). + - line 48 (0, then +5 a year; Spend head); + - line 84 (0, -5, -5, -5, 0, 0; Spend head). - Lines 9 and 81 are also below £10m every year, but their RDEL heads fall outside the footnote's AME-and-tax bar. + Lines 9 and 81 are also below £10m every year. The footnote's bar covers AME and tax costings only, and Table 6.1 prints no head finer than Tax or Spend to say whether these two are AME or departmental spending. - **The lines 45 and 47 differences.** Table 6.1's footnote 4 points to this document for the packages' full costings, but the components do not sum to the lines (above), and no note explains the gap. - For line 45, the gap has the sign the HMRC spending the notes describe would give: £70 million on debt collection (p.49) and £485 million on compliance staff (p.50), "over the next five years". The document does not say that this spending is the difference. - For line 47, the gap changes sign: +205 in 2026-27, -135 in 2027-28. @@ -378,7 +380,7 @@ D marks design, C a contingent liability, S a spending allocation, input or meas ## Distinct condition values - `row_type`: measure | measure_component -- `component` (25 values): each component note's title, and p.82's fiscal head +- `component` (26 values): each component note's title, p.82's fiscal head, and p.108's receipts - `basis`: forecast | outturn (the tax gap and the two present-tense counts) - `scoring_method`: final (every Exchequer and parameter row) - `table_row`: Exchequer impact | Exchequer impact [AME savings only] @@ -402,3 +404,15 @@ The costings review (`harvest-staging/ab2026/reviews/verify-costings.md` on the - **Finding 5:** no change (p.84's £20m and p.58's +£130m are accepted as staged). - **Verification:** the contract checker reports 597 rows, 0 problems, and a re-run of `stage.py` reproduces the gzip byte for byte. COUNTS.json pins its sha256. +## Review fixes, 4 November 2026 + +Review A (PR #231, issue comment 6635870934) approved this family with fixes C-1 to C-7. They sit in an explicit block before `build()` in `stage.py`. No primary was re-opened; no printed value changed. + +- **C-1:** p.108's six Exchequer rows (the landlord fee's receipts, 585–590) are `row_type: measure_component` with `component: "Receipts (fee income; offset by increased DEL spending, fiscally neutral)"`. Value, metric and proposed key stay. Components: 157 → 163. +- **C-2:** p.50's two context rows (233–234) no longer say they are "a tax-base input of the measure"; their note says they are context figures with null key and reform_hint (AB2025 rule 3). +- **C-3:** p.41's three text rows (176–178) carry finding 2's double-counting sentence. +- **C-4:** the family follows rules 1 to 10; the Conventions bullet on rule 10 says why pp.34, 35 and 41's PSNB wording keeps `revenue_change`. +- **C-5:** the key tally names the 2 context rows. +- **C-6:** p.84's precedent is the AB2025 uk_obr_tables family, in NOTES and the row note. +- **C-7:** `stage.py` is on the staging branch. +- **T-3 (uk_hmt_tables):** `uk_hmt_tables` no longer emits `conditions.fiscal_head`, read from the workbook's hidden check column. The sentences here that reasoned from it (the p.41 bullet, lines 9, 48, 81 and 84, and the lines without a costing note) now rely only on Table 6.1's printed Tax and Spend heads. diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_costings/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_costings/claims_staged.jsonl.gz index 6126493f739f8894497a7bd1e75041d6d8fb018b..b76df060f6bb933ab2b71f80c7d4423eb03f924a 100644 GIT binary patch delta 43204 zcmZTuV{~0>yKR~@wrv}Y)!0d6+qSV9+ji5&w$Y|B8{2AZ>+YV@bH4B1n>j|tob#Q} zeD>Hs_Sk!^RdEXbcnS^>MMA&%{rC3usAI{ARI|3yyDOKpe8h`p3by1ODv!y+s+ni$ z!BM4%mf5Ub<|8B=@uy;VXdXl*C1=LEdqnG%ktJ1orC8Qnlh+gD9xzc_v4~G^UK$w- zARMuwh)!PjAOv2X@i?VOyD};LxbAn;FXcRV?+Z74rd*Ag&HySsb6iM2`w(|ZOXLB( z)qGCewuyduCWkEz`TfU0e=N_$d^W^KJ>sI666o-UGpTQPCMhfVzg5) zx_`oj#`6;FA9lTn=&`DwT#28U1Q2ds_dHMIc-v)(=ioCt7Bo#J>M_YuNi7dPKJh{0 zFI~+$l8vW}>qTMa0UJGz*}2ChW6s{CUyFDY3aLhwj{ynvRkwtN>BO?+6P(*_$v)#F zim%mGKMuEe9#RCeT$o`doq21fnERrZJ;q*-M|_xm@k;$7CmW{;IhESJp2`V*3`MaE zml1p(OZO0&%c0EHaGE7|VJ2J6kBzxsWB|VOU;WUdRes1;uJ3_qtHD}X|15+ab6xFY zbJmZo<_>K2j%V!;E+nJ1)ymOZ2<6Uc1=$|~XuUay8W~O|+Xm;Ic4uqh>=RD_Gw2}LqSK+lGFTis;1+7eoz)y8y z%+!^`z4_|<@_oM`T-Jza^aGVw43-B!eNj&CNGBjSrd5a3ZQ==! zQ*xq7TM!3>X$Yc^=RQ1ZzQw}LguAo=v&twXrQ(zQi#Zu*(wKz8@%*v$El;)*951Kc zn);XEIHplA>L~_q&AR#++w>JKN#IMz6N{QM02R zxB`$5wRlPHC9!|9i=9Lb%&0c*vVb8BG8AJ_s4G zuNa8Q&B?uCluKmSKRZ>Vj5QVJ!ro= zZ;o6;!)a&myP3vqpXFT+kDngHq!WhPCI#S`N2Mu2ytn-697&Ff>KxWRI-SIhQQ@%+d5Bq^HtoopLvki4xb7 zLE9IQT@HWr!R=1T3Mjyj93OZnmkXD8A!eyEMx9_5q#gf!jw6EIbVbwi=)hWnHPF&z zHywK#HGK)~n_^v%n1e;Zn9_5Ueq#wlt;tkHl$17IZMC<>V}(U8l0-bO)v4|La3wu7 zJrJ*<67VBVS(s{Iy9g!o9gp;hhOSN7Ii>DV@pzWUoQ?Fz=}Kn0?d6;d&rQ;kZ$pMy zXnno4XXe(U1UE3nT`+9YBUH{TI;+31Lf))rp44!ewBc4rfwFV;N_%e+rAY-4wp82% zhcpiF26H+*o+Nv(3U=7&UbUN*dGj=P9_1p)KiZvc)-Lc`yJxla6m_bfc)4;g4armZSCHcT(d@A+8f?B?iO z5iLum`t~b%df)iVgL*LuQ@8?fBQTsjkJafA!2{k*JvMRo$v%>w?Akg$z#5)8a>$dVhuW{q4iu z5&CORHeoLN&f4y|CYQQW)XQMvTDf#NH9Cq8lW4WYvB>O zCrq}db5NdLQ`Is|27{nPsTk)i`} zY2lQDKjg6`)F^Q3k&8M)Ao@mYA-vE}qIGfKTVpcbF|*+-L>3&*FRJ)>xf3b^?R^}` zpK_utF)IS-PY8Cu%);U#8aYz#4+Z73niTKwDB&2|5Wcbx0XzddcBb=orX6 z&Wu)wGk)qze&>Fd45uLsEpE^+;a8!i}cL_~ihQ3NvEe@P6Hf3l?xQk6|gL zSj10E$&F}C+fq?ca12v@%|zbmi#YGD)jJk3%wwayS){qSDWJX?E}IIazB%pH?~v|2 z&@a>Hyg7^}F=KI+_x6Hkmz|6b$S@%MUtWTG_2U{7l?g6=y`E+ zoLPkGZkhnJx6a`%zF3rBTK6&s&2*~Kr-+1)Ux*~#{e#S;JN4fjis`(J5v^Vym58$1 zD0qU->|B#O3^n%;fI8II3=n@^r|_G>@Vk-Tyxhmfe9o&Z`h%*X$|`FL!UL~^2ZV!% z(#Xq?bINmddH|9u{~!2Xw{}YDD(C{gT01XVd8V(Yujv~^K*rzD@2Ip_1kDh&S5~!> zBmN#L+x|Fr8d7O5C_NYZ{LML3uOA`cY&IBeUv?#ZL}}Twsi-08e9IDfjJX7X3XvtG z*-pRy^H?<=1=!mgxIc0ngZ-`2{=h&A`v?x zny#ieeSaxfomQrXRkKKz+dJ6WnX{*&J^KoHJL0zxX1z~Iy1TAXgLl z?c29FyIs#$Ck0zK-?7pH2VBI5JP!t2+8&S%JsOghy{91gA;=QgS?2nBk83auU?!EV zLlGjWUz#(8Q6ovLlBZzlLNgTOdD%}0O8dHbXO~PCUrfbSa1ZJvUs2;b1#}vs1dX=d zwvmXF>1%mb?Y2cBpT;HwCKbNoEvgR-U(QUd_|)-Byd%#F8%_KNuD&gw$=t%qil(lb zNe^;-yV^NH4VRY5U!D?{HE`_Tc#5;gQ}Ldn9|M6s7`TtX&dL6M)|WS zG1T}HReCHA+`!7<8VUt+EOHAl=mGD0aV$|pq;RV-v!s(rnQKJuZ6qCa`jzbuk)#7(Do*0*v4CIO?yy4ba z7;~wOaas1`>fPD$oa~N5G=1lyQbJEr5l}NUeKos&=17{QM14vnK_rTA0kncy)2Qc4 z5PRWBaPxyJp0&-_{!BLBtbkTaE))x!774o4kr-8@f*3Z^^NCm)_<|As(Fl>iz4+pS zdicognHmR!XGOLp%s3ubeGZt=b&HwWm)7@cD=w0?;8hSvLP$GU7jUg!_R8Mv4fc8| zU=)s~>VDYgO+B7^*E*1r4Jh5dIYfiBQ7PMnL$!yFy4p;n=w7eTof*?s=YS6HrTV<5 z0dvnA+?n#hRC`XtLTmPIc@}q~`J@4_EtRzA_V*?3O`MOTux93bbvntf39n23$Y4fuVvI=K%t zV&p_UD(^B#dlXKi6PQ7@o{Wtd@&vn-+87fn2z`w3MUvK+EHLtgD@+lYoVf-kLmkPl zjO84p3Snius22@+E75Ci#o`lQg_K;Z&mQ}Jz_H12+CAptkPs%3DA~?grAwF9*!N86 zhZf;}C%)!AGX_!N;?Ex{SU{{-+QFeSEFMr_I`A$fw;}^7|MA(Ej7-{7Lsn67b}d`h zh^9q@9GyKkjBUFG=8o=^%~LE>mOkeJHF7Q%yAixS66s=@N@N#Nc8449zK0BeEMYou zf=^B04^tY-TQpA`p@rM|&dT)a05LkLpgNsVCp*0fuFJI6NdsfY571<1#B!S=NWzD_ zWN|m;+@f3>Ew&xh5fzy8+jQlk$M#~xzHH}iV3+v7SFs}S+0xEF_ZJ~tp2749y( z4@Uaoi_7{>Z~G=fA*x-W6GzVTZC~u@mT~L8+Z&>YlL3p=+|eJAA4T{duV#szlrQn4 zp^)&?odLf~C9k*UYe4KP?X4hW`LB6{nOlLr$7(XeOojaECGScWigf}@;1MJZjYS^ai) zZ{*4TIflhCeZ}6%h!Pg({vclBrA)&La-7>Ff%w12VL_v{?omo}1m2*!Bx#(2gJV~> zY(?g(vgF$q^73f|V}fp2!{BOR4U%$XwHqY0j3A8ep^fj-JG?Q-YS47r^VUIs4x#CS zlAi;UF+pjkPt{7-l<+~xB5M<8=MAj`$R=S3EIm74vA;mUq00`y0B;4v-IgLYXp0KV z%^9k};MQjZ*H%|@Dydd!Tq-7MZ{U(FJ!pKVlwkpe&}SYLFf8Nnj-c?za7?9+>uN?_ zvdjv2H5NL~!5iw~;fh`BD~VTeuPX)jUkpnfhyq1a(hD~F<_jcCwj;-Q{nW=Alao&7 z)6zwi3y<=dt!8#hYrN#Y*SIhjH!=r4X=bGGs&;2DV4SKj@t#+T|kHu3o zrR45h?bpG5R_A@!A*=I#LewTmH&V88r8Vl_+Enf_u1t;`VTUSC5^SX)NK!6u8I5z* zkr=N68@QIA<8PYCx;PUKi~Mr}@g&TK^FQ)j7+g}1i?UEV1G=n_Y&i7siuKC%SBPed z%Z_1~k!!6OUZVrB{ zXvCXY1&LbMx0*A(E6^sPZ%tw=yZ;jc{-g?y|4Ahu{heaumiWI_u227jOdp|5pg>hj zW&bbK01E@!$NpAxrvSzsVSw4QO!bzFD!rmDgm^H@M4Xi@J$^*ejGQfD!4BPw8~~|V zDKphM!)0#PHm&%P$^#v8zku%o58PRo1B54&A|I!5hjhm?ivzRME_yjkB8FI#qk{eX z5(_G>Ca!}R%X9g;jW8sf2)V1)+sh;z^%f7;cP!)jD7x_sKS~(>Jgr|WDk3;l4p?{( z`ppu&UXbEx$4tz2eeHhL@-D>_FshXw8CkeJnGdZ3_IlM&jYI6)(=CnmC%Ltx{oOU} zx%#$K=)N)qwwJ(gMtCU6E=RL=)3r|CZW8h}7t5f1T1V(5!C%^^J;~y}Mk`{SYc+GV zJEozFsZwXqV8Gjy;yIC~XYArscLeYwVte#`HD6#xMx~4gY>ILZK5g0qiq4MZPe#*M z0!h%v5q*dExZ&3qXv{DakOREgg;hf@oxSX{q-vl{Y0J8TY) zR!Ve4vgNWnII?v;S8#v(DV=%$HOxc)&#*d4-k3nt^OPDbW3YD}@t~@Hw?fd?yl1jem~9=b zc7XJy-RC;dq;8zS_gCwQM!bxAU0@qCP06QGu)Vs`^*tp1oU_4}5qP_{*m>$+VHJ2Z z35{66HI`b8##3Bq-y;_n(NU6=qObuQ^DvyBn8Autx-#&e-X)8=m0%N@v?a9;S2EEq zYH-DjUDwxhUW9Za2k91iVqx~&gO-<%`oiCZoqx z4rD_Xf4eqq>k=i$O_90ivhdBfn9xbAM2EA)hyHx~4pnHYdF3TR?ZrUt`=bwl!DYX) zD&bBoG}b=fL?0&}Eai$a7JL zOI@c&0x4%GvpCnOeB8YUAstTtQ;YpYna|6nQ;9#ch347>E!A9AAIzS~vp%wI)!p+? zj`J-Rc9k%q9;8jE^T4|rsyoNTE=(TksaTBjuX0ZwvgrhW9SA)U`yD~~O*fn%f?|#q zcZ#mA#A7gBi6mHYfI8MEe+(mC3R=&d)@pKs0&F|UBL!TBw}ROpH{6U-|Ujq;aP)r(HDVu66_+5j~>4 zXJm~;@S5)M-bF?W#0at*-x`DVTfcR1Lb|c=h-@dHQ7RWp1s-!+gWD)<)N6}d*w~R_(y0CxpJAiaVr0=lV&T`4{%oKdJ$o6xYuAh}-eAL1`%&1oyhOOV zPoutcF^A%;l0Ueu@IXvvOLl%98lP6;&aFsu5-=W157znb#2fg(64J%f_nHWHyu(K1 zDq||Z80a4ecpt|YXw7N&d=4bQ8L%kWz{wKG!g14*UB%tZaGgZ%RSdk3w989Y{};CIuf8K`pp8W-4PKb`B1xy-U4!v%ZgRMN(i@kZ7n%9|UiBInk$!n=aChn}MX zIT{|2H+vz2@lmNVtqI7td>(51(DWf?AFf!V3JvoUVbGrQs|*$8R>FU2S?=T24FI*A zGr}DP4oA^NLR^W9k)R)y9V>F!%V~ujr@`TO!y%UnoN;-nXaFMFMo##pxaeiRzA5{6 zn~$Rp>@s5Zf!r4m$jc;*5iA%RQM@D5g)R>0mb-EcV_qjaXRI2f*IWF>V3Rxr^HtgN zSe|R`$i8bfjTL8{uM8-UOtroD@&M)@o8}yA*6L z=z>{xS?II@*DN%48UN&kZKjK28!2Tz+~m9c02iTFVqFypEcL(SmjflbM=>lT$TD9) z5;u{aV(ktrQb?4*Frgk|BV?9RktplGq@t^8hObD}ZiN0YMkEPfG+n`r_-`rv)Fejeaa6NY7e*b~`|Czfr2?vj84h`XUf`g%fj z9G`+&Jkg1PL(!g87+oDpYNIjmJYQ$!5#Sq4yRw`0ozBQKG9m(Enr$hJLHiYT!5jW(!M&mLr$4cJLsnRQiWp{BPm{ejHehom_;|BU} zh3$VxCqdOOP6gF;_Y1K`?K4<_ipdvzPdviNq7d7yk3lyq^-)9+0f;%bZRa$fBbG5V zs5Xu6J$7YJl{Yu+NL5BkMPtXOv$rapQr{KV99w=CvX~*xzKYhLIlN-)e+Wj-%Q@tz z$7$0o_Ze8&42-3Zmoy<~Ma97mj}jFpC5x}w?Zx?vZrw<8J2jR*CPER)fFUI+6PkgI zC$D68I@5y8oXz^+ADZz*NO-y zDkl*fhi|E83&eJ!!E@RBnZQ53)biq}xQCjLcf|3%0CN2zjWi1aD**0$@Vc-ANi(v} z&1V-XkWj*CR?&q{TzAp{MPTFpR|L5EY)Af2CfJelyI7%XuTw|`2acJqEA$|<2{tP{ zP6)h62+5Blp=)7~D5MQ3X~kTRVTGr3>3wcz9ALJlkCFhL#~BX%9XCT7+K&eb3>Mj) z-hH=ZLq>JeG#S=2fe&^7NXe?gE4wz->zbyOVKnv4Xj1*Cev5-+Hix}y z#E(&c#@XF{++)edZi=(j`YDQ`n?9qAO5GP`4O!T+Q<|l`Q8V~?`#Q-G-r)o575`n{ zJWh-FtLb~<>*KZ@&Uhi#8Iq0{;pAT^4_`t*kUX2&o!oq3=0F%~PbCwv*j@&hm{y-} zffx0xE>qF==}^RtgHXfy*d4dH-1Ei=KEYhHiQppD_u!;M80B+AWvt^eyHVtqFyC<& z#KHO^vJx73;i_Sb?FV40?rRA@UH95C*4+p1RhQfJg!V1 zM560hBYFu4sus1zhtq6`N^O8_<)S&cBBC;8dv?{drX{4xo~Q5hKW=-!c$mHZTB0ab z?Pt=ituL@@8>Okw-rGk;6k4i;n^T#~*4C@_8fVPtEuqvH1Bf~(XPORJNDq^4$^;C1 zp~^M#BD-_`pi|2b5HnN&p)34A<&o;h5T4HOwN1)$99IoUy7>pm(vQ~C-XJ{6P1`RD zS|YJ&oI7T56!6RNDlRJjfjZ6cp;l+uWX`1D)MV#Dt++SUamxzCFuvlr@*m8JjB}Rk z{RXGvW~f(=R}KYG{Fnai(M?=39C(6oFPGGomje{#nBU_a+!IV*rp8WQc6=!Kn!K+e z%gNRp)Dvkcq%%{LM8#J$u=TN?6upv~nL$qjzIu79zezBx{zwK6vZSbi9ZlENaS=m| z5$*M2Cl;>NS6&3q^+c}Sf@=z|Xse`V=DpM;^wc3`<@wPr_Ot6ze_xaKzWtMr*>n{j z%}=1XR@3Lz+IiEvDpg!_3J7`k3h1G)+Abnb@v!u=?=c$%UBB=yUQ=XY)B10VTL#8l zZilMd=T>T{ULlH^xYC>&Lm%g$8@rM@lXRJ^YMNLMB(xFScf^mxc2vvi@I>nzifwZ_ z@3uABW5R}}SckOAP;gepR0lC)1T#J>Q2@Mx8M#NETiq*{hw~?p8w_Tmp`u=hfvM8D z<5YD!?x`W+o?kAZlg@&CLGQ0lDZ_V7ilFVp#EF?s%yY{`=>u(>x{f zT$6vZ;F+QUM&RGXLMpBnNDQkj6kJ67W^i=2@wx+ye*(2k*WDnIdR!|rGxOuOt%gC? zb_CX@!ylo`JeVzD5{KoIe`E!vTTE6h3A6A`fi#eQ)1=_2GOgHH>lHoF7TrK^2~${A z)!s3#gq|+vuE#TxUIa+NhnJ?hetV~*J~Xi^8FnEp`^6_8Yg9fdvUp7!D$oNvzyV%l zC^os1#065#d`n5|m(mVzh&Oj(IW%=_mFQE03d-=qsMifeJ@x2M^FiS~i4xLPZ0sZJ zmP$Bbq2~x_IFzjHF;%1&*>@Z68jMvgEqu9pmF5<75w$*&>1!??=S4eK@fZ?98s_Jb+#8 z#}Vz|; z_BEmbRXiN|1840=w?px>A(RKy=DG{f2ct2KjhE@sZw8wyDWrndO?*nX`bxKRHJ`}} zt@mRoCIy{WtCVUUI4MfqLXNwd9P6Ss>}g(0Wzw- zWG*Y<*ug1n?h0z#_7 zrbAnOry#D=HUW0X61OZ!JB8PPrRTxy;7!#vs?CYeFv9UE(V@feHUU&~x*_%#9x*?O zzJ`q3lKRS|%Rr)I3kQE@nJB-LC_FhbUb~5U?hZ4my%=%2=>?-+(fbt~;iR6snZyTb zT7aODJA&S<>%qZwhp$E;{d~Xk&IH!9lpx#=uf@Xhktw~Frb4!jST*#(d*O->Dk*jU zagPPvxsy;KCFl#5+dbE5z0Xpc%{`y9o+l>>GTL%>Tc}ONf|k9eh^)|vc3v&h9pu(* z?hnX-sxU2#))EOZA>IP{O<(EvH!}twi2yXp4RbF&nwRi?OF?GwZj-|d`oKpw6CH>) zw8ODj*wV0#DIYn-Vn88ldMJ4bx!5^L|5mbkuh=E?>jTE#aKsGjd0)>GCaQhK?pIWM zeTMx;4bdgo&UfZHR2^3r9&3;9d(SebGdbZ=>T1#q={|h^X;Hz{_DF>~m&@ajz6yL( zpDVyrhK+!$7=Zey%p`cn=@hd{)=7oD@^A*5mKQ!j>ZFWko6UmT<yXVTNAi$y4+{k@$uK&{-8$yUgZY?3N&bwX4 zQ5}y^Uy5&kPAZdW%4m#v5WZjN<_l2svy<_ro5FFY;@N*wGN2Ax}ow>f50g z)Jd@OvnMC}lmgW!S{_SVo3`4GI}}l-UWzl9RWYm{M1Zk0O<`P%#Qt2FR$69XdVg3GjKiA`U2rT#HY^RDB8P^bMJWO4pLmC zbHU3~@-P4ID5AgpftG3lXmEjfO+`1rY5Z>2NvyMld~N#kFAC83KaEh|IpzNJb!PyN zH$IC)NX6$c4Lss^qI^x~Lhi#^EIuwI%K9O&DIS||ZyQnszYh18r zrnB!zaif_T^2>Vs^Dh6o_(DtYh*rge6lT}(2?}gh;3NN!5D6sU-_o1r6#z>Mx3%E_C2^7qTW2Wj#&z@^3>Q(_rAuAT}66Zg-G~M)aUt_rc!gy|_F!9)wv&q_;zEWM8#_$FmYx=D=@f|&Y)DCj4^nW?@#-?dn_b)Ky^y#{mzWL&M% zJ2qihlHfRDNs@eIB}81MN@$uObm1XL+RS5={DJAJo4km5SkIk9IKHyQidgjWTwX*DIA_1p{(Hvrw4*!AfpEGM5J#H16b3O zrgwREc6k<+Xt42839i~+&!uIsB8jBsEc?5O;C_C)FSch%6&2Zm|5`h@vESvmY;y5ENSViIiV;11q%S}|}CELPK`)(n?MyQckJC5OC40Fx|4rCay ziPhZ1GpIdRa(7w16i;gW<^CO;&SJ3A>WBc1%)ixE(fww{1o})LXJEW*=`|kGh-VnU z{1aAmZ?s7GEh9(KGf6@9gqlvj$9``b06=zYomk{w+NO|sTA;RfCt_Sw_aeQxKPl9R zTid8NrF!!~QDY`raQA|g-qeN|=jq-jf zGi8;ai;D$PG^yx`ABL*m$+HOvbY3PItF`Csq382BbJ+DfPc3*bKb-GQ3}UtfV@*6^ zUIr7X8)k6;19MO0V|xK+hi7`tA@uSHPKw*_dq1=^?Kz!363vy|URi0Q_}qH!^-F$x zTzT0D>kdUa%aSeE$wPQ6{T_t+2I=~KF8+wd{_w`(o3ElvroKdFakfh)<=bBul|GLP z)S*0HUjMN94&Y}l@|8o-?ooH0#DmlE)2{6P+Lw>$)w0!74h5oFupsDj8Mz0FWGacS zT?W-He~k**UJUp6VEf|jRw}F}^pfCS-W@%a_(|(Wl|)5PwSw+>rulK%U-;2K)PpXf zx(z>aSb^6mT*<_g?-`Ms61K5rO?7iGbf(Ja7GY~-mD zM4mSD0={u&6p=|FEc{f~X%Evf-;8m!r6d!HncVwoV{7o)Zkqo zVd99VXXoKb$rf`?Y?zbl?m^fkl+%3q_7RiQ`=VONi4Ljb*KpX3Z_oxLV%X)pjQ~cH zn9(Nq=G`}>_9vqNT7Q!+Q^8>ez8kbhzA)d6FHg8;4q;;hDg#qU@%QuRV?W-viddTT z(}NyX6%-#MhPEGVNJgve)#lq$v6BZZFKo-|ND?&}#^U#!Ym1zY{Hew+--H(Yh{rAm z55|$)y9Pttk9DncX0-l9*T)laqln~wKy@)uUaF{{BiTYAMK&Trd3J5)b2&kclspK) zE3LG#XW~p_(ZlJ+!Fe_-d~q`-a!ZVqGJ$>aFKR1X9M@HGSsx{c zTI*M8@7?CMG{_l)HM;@N8u#%_?^0iF495AN?hPP#1B189Ou9R!Yp6%D0D=`Wqr*%e@b&KH;>!N zL0ue2Qe7&r4iU8qjEcDQUT^u%yYELS2+y2L6;r$(`=*$JYts;D1(+alNEs2DeD#b= z<_I++4Dv+Y6-t9L{BIIRure3*+doKzD)A~~{X1Y{>s`r$)XfJCToR82GXG}(kbxLG ztbyP1dN5F>1DyD8zqYXgKK*|;|Jh|C;l5!4+8Fn+EmXRyk@)v(&|E^y`#;5hHq?I2 zTw(<+l(fAdPCfyq#6sILjd6p>FMmZe5o`3$e2dINjH^ zIcPIir0fH|338pufDS0j=%!mOpfQfBgZNlmUX`u3B??v9V5U{K9Mcu5iBPhKy`L^g zQ)ghD-64)zc=gwe1@PY-@e zgjuY_&E1lkm4e*=<9a&f?f8%6sf?#>agO7qS%XSv@6O>Wj

7)mEC;8*!K89>O7C z(N9vIq=e)@r28`q_cgsTm(h@P1bH~P6KjrNSK1Xs1c;@&EpI{+ak*&#zX5CgZkE>d z&fy~NM{-Nl#R7{dM=J)Z%y~ewb(#e+7Pj^V_TnX|1+T|2kNT2bkKQKpTO4<;Lej2` zZhMNNkMsTIcty62r))~m4qIO9lOCc`sSjnlLV_KeKx)LPf5hpgmFxE+%F+Hvt44^H z4OSxZ%|KC~xG*Ht4wX9( zLhVA-y6yIsjBn4r2B9NuPHnd-e-mc{*0w*1F-tIR!^?YFwDi_x<1*#%)WZcPjiYD^ zO=LvswKX?(tE{n@L*z``Q%kR=p{smBBxq)6GxzNgsbfz~>GP!L(eIjwBe9&;@0Hor zR7S2AJx>EZLImjmx&lwfxm+Vd>Iup2h{3Are(QrXJG=I*wEvKW{(9-izrBTr*+uU-HOq>-YU#2ug)-T8^RbDn)`&bC_Q=hov0;^mL6qAO z&4eTa+D!vxl-=;IDtWQ(6tBY;A=U#RO!hznr=$1k2VZ-{bkC<4RR_HFr~BNsvaf!K z9~qZjc3rojT@}4LH8l^^A55F+-H3ArQXIi+tk1vKhzWE_o|zqTVeKFpFdqh$=a1sr z;FZnwNC#NHM?YH`7{($gf0!RjO}sY`jP*=_fQ+EqonvnG%f|`c2me^62rt_LI4%(H z@Z9%ke&?(zpO|4w632NTO*1A=asxo^$W73+&sEE4DB^CvNS%=M(e5#1PmN&GJ01Nn zL$xiv)+%+qKb0-;8n$q~_w@5@m95=7f6hH{$UZgvK`LYaUQqdqdFBu`MN!BZQ-T&f zD(|J#M7aB5dydD1gIcG@F-OciFa~|rcFRlL?r!J)4%Yq;POqU)=|KEc1Y=b-6}#xH z@N6{|b{T#39~`2r&i&@0wC!b#B5jx5k2bv$@?&=cgPn6W1#gP}KtJ+Ll{Je9tOM-X z!8?zcU^sWZPsSHiy1p>SsD8If zL2=iq9{2@x@;CHjA?y_clmm!LAwUlo4~@wd564qAp`@BxoF`2h=LKe`%B| zG$;0sM)%{#fvilo8L}Tbl^Z;*lNhwGz*`6hAJ5FlDANQhS_CXkoqFZ5BKPd`P}Z5+ z!>N+r<~|`Psh?aNiVi9EYVa@_wqA@?G38p>3OFi|nzEF+#jnzsaTlFcP_x605AFSa z{$u~Ukm9jI*7KTPns=AWI?1!6O+_)pSO>jH}y(am8UXvCr_u zS~gl&^BlSeIncnek=n+Fvi{}m*B(q5F^f;4r$rxsY2w!D;$_|RD&Kp>1Nk2t#lpOiR1aNFsTekN?ne7J=DetReF*O9VTnVRD4Q#AbAfgs)T{#K0WJ3F}& z(Ts*Se99S+Pw2t~YrJK>+N!wTRybcmC&L*+u~ru%XSvcNl0@^QeH_s{Z6Y}f&0>T_ zrx=BcN@La$Y$VB&kF5kx6M`-h!Oa)ZLfRV#vhd2<>q61NBQ>IpdV(Zb4l7zkB1rx^ z0fk346i3+q8d%W4`RbdWRyKt}vl{^@BvrJzp!{|4cNR*5#g*Yn6f|iQY*bqLt|1iN zxnWa+Mg57~Y^5Rhziw4=^MzO@{JwXU7L75rC=$}5cOIB#?T1)*vZ?v+{Vb9n_=G>| z|C$=dHm@YWPa2tM&9y*0ZT(Jo2S-1IZbGq@Px@Yvl{oz*6e+W@5H!==MwHy~Fx5-rMd^78vk?kkkJ7G4U!1Y<1 zDGucmE4v;`*9T39w^jyna1R=;E%-2xGlXp&K!IO*4A)^&)cKR9#k25{Su8on_~!MoC{l`C2g z#!}Rut$IjgVxd}QSv=cOZIuuCD4NFfYpbfxeZUzlYICjGiSF)Xb)92*r2{?gbg-w-g{Z{=%bGEQ?6fKn<^UlsqUzyR zx_2WU!v&}%hev{AEqRq9HLzI=xlIgT7wliSH5RK5<1uKX3$fd#QYK>wuSSQ!msAHo zWnN?09(ae)z%DFlNHalep2431h%TS4?T{ah1s%w!#6u0sDU7}kQAb+owxnx^3O$N+ zMWzP{iDZAR>pYZf8}%Qu@DkhqQgunv3kPm*!v|y;e1DUA+;&6C2x&@r!qAqg^H$_U ztlw9`7d;L^kJu&HZ@Gus*%=8(BTAdcHELIEYL4N?R&$Op1X7vMFiSKz@+OXDm`c!Y zp0z>z*%cSZ%gxa=75MC15sdCu%`O0`twtxZ|&UjpxdY=~EB z_GIcintxG-@uZIag#HvteOJc#wSFm}t7CE1*$8v`QD6Tu%q#a_bZK<}avcgCK$ut8 z0joq?#jW~s(eV;sy?aZfocONtNImY$kzh@!mq)r zOX_#sC$FvxR-x&~P*7p%z6i+r2l^GO_KK&?t6HOwt_9$(K7!(jBue>U)*-g1Y_JL^cX2mi{u@Zds3s+UA#Fz( zU@xWethO)OY8B;$yq#SnFzMfpO*_rUjQ4u83x=v^RK35?1;VN!_hwrXJ zt}{UfJzOFTK@xy06h_>pS`!8m3*-SRx?R7SAqE*K^uGZsebMlLj+Mfp0noW2#s4~= z`}cguG8Y-5u5&R}xJMS6to%U~n(;fd!jqA?TeV!*|j}Mfj8>6g-pAD59 z=3*A2%?C`zGBmmmgnS(Qqy>sidzZInEnmLsKGXyu4Y|FJcorQ>+uoc^va-}aW7QnY z?(3&mZ;u=gk=}jTYIvE|yI;I7NDWFD8w}E_IMAN)&YuM&F`4s{2ZUu~48GH9K)?mO z`yOB*R6S2n-At;1Ob?BzU7_X5WiX+2d{4vjKH#0THVVl6D`({XZ8HOuCDyQMz{CKJ zeUDK6PXSbQu_X!(sP2n^fud`03HaMc45DDGb46;k+`pSa*8PZ%IAx7oEG3IBLcVE0qJPyTwsq%GK7+cnV?Ar#6i%t>9 z%?v2}jrA32F%kW-leMCrACWip2ze0g~i8M=2}DQ01kKBH@; z7Un$Kgk>vnrlFnC)xbb7r;kp9WIU3KNyN&&EWGR~$GIbF(+IML`gH@)D?6~}VQZ01 zJq@6EmPI0&fuVj2NDYbcfBBUa#VLxrNS|2O%<7mQ6rTDc(*4B)V5tk}~73pQ^fq9%PjFS-B*H%q> z$OB!4?KY&}XBCz|b8%jPKGZWzdSiU=HYazN)ix`2WZ@!d7WY z$7!rfF1T1R-1Z!&Uma2nO^|`>G0d1-mxO01P9^%ny})~XS<}JS9zsQe*chaD-2u({ ziQz^d!>w6~-4D7fVE@Nhaq3)UDwL4K$Y*2Ww++s=4VNa2Fo0idKuRn!75{=C$=OOJ z(OpKQ*O$X%8kMlZTIW3O#v0U5W4A6w4{z zVQPlb5%1`m(nW2uj^tKpZt3o5^=SVVoUFL6v~DM{zUpcxaSN;NEW8Vq`sJbn5l{}e zhI=+ztaHMVI0nRz#g7{)k8doRWJXi08*lCpGUONvD`T6b&4eK(31jHQ?&&4NqwgMM zNsfeKMmY$Q9QN^d=J*rdII6UcPR#Msw8?SbyJk}52@${q^gZ!zM4)<=kN{)tMwV$7?Bt3FyXx5WtJ|@>A8Mgrn&fHVN}5X?HgX zAM3Q)BXndqtbpx?gW?MPO#uka0w9f1b2521b26Lx8UyXZh>#eyZ64|ed}Ab zV0KmQ+THWVfiq{itEqKbU#@|k09}6GmkCrGD^l+hRda%=qFI3~T3EdgUrov{`-y*A zBe-aMoz?H?tm!Uz5GLwwLBy9MNG91IN&ttL_ros-edSZH&)uKl5mO$Z0Me(fx!)_0t>ax7~j8E(Q zyiJIZ`xHG0S%izQBJqAHnD@nHSJ6rmyfG?m?oH&W?3{{na)310b5u^GuO<{n7m8_zf8)~IGSIJ`~mS+u^0qxCT!;NQ0Fx8iji6TYo&$(`eu17F$ zJ`|X=s=e%VJ#i8SA9t~%?0EL+8O~p`VTI4ufT5ak;Erv5uYVDhaIf5tWHy5(2vY5> zy!^eZY?r%(PvaNY_gF5j6Qf&|zx{5(YMxt@b99F2Vt0Q3TL79W{@jJNGphpf3g%-h zcOKy&cS&{GyO;oZDokgQ<@LODRhs5=tCS>AhPR~poF=9ll&Kq9Ssqq)s5vg{K}&R7 zNL%_P<>n9ruJDfdtJB~Tx`TW`oZBEuYC~SnlClNpC!_&X&bQ6;%n}1A;7|G;|WgYko0gUpa%pTM742h zQ=<&+H4%TMB=y{!+8cp}IZ5yq*BEY=cw{0O(dzd4i%IVi@ z(_dkh6)=s4y#;ewKGI+Ml_g-%wxWQ%ggjtf@DZj`N-qrk)Txwb5Mu##3_3w_xIajs z&UD-3C+1XoHIecx42P`#y*j|@&E)C@zJvFJnMcu!ob$}km_UO#2j7_w+I5zkh;*b2 zo0NZek1m6->DQw>rH7+=?TlYxr0Xv#*{T?Nw1}nrtR;CJf}zOuvTo{${0MJ~)=> zNM*`rw!O*Jb<9+H`&$J2p(qY2XFTs8TP+{AQQA}9J1rV7Yo;Oad+vb$9d}Ts2nzje zHtKtIR9Dnjx2k(;+TH=WF_%g$;2|*;${)dPVs4cM!RceTK5+t`78#%Yc}9xxLI|l5 zH;rUcGkExz`M-o#u!{2rV2H8%LzPW_rxrz_%QZWPA-LJvhG7*sZP8ofOv05d)0i#T&tRnVzb?V2@9Hf6Zk?(|5uYE|lE&#; z`le**Y8&R)#_k)SBu~;kZ{Ib#6a{^k95jZKbk5a5E)Oo-!CRr5`hg+i++l zJ{X{(C)GY#6{Tdm?Md6r^cU?(CCyo~$sC(-_Ow)DuWerdlD)QF%H6#%2zIsITwmwd z$r&(fDO-|`jmOdwJ6Aq`URWzatAl|H=8t~cN_;ndjzv8+ zecz5xSg0bj5F)GKlKU{_bQ;mDF6}=jIZN36{PeIC3v^*$evr!xFKx^6uaq(8g|^I3 zxUj9!$r8%P@jR5Np>wx0?q zzk8(A!jbEcII904V4vWBfGOdAD(!2GyZx)&v?Gf%SK2SFbLuCGUmpN=sYB4hRJmNo z#2Nk10bdCJ2bdCmNdqkrOq}6vua=uuiE`#D`DI-eKjX#CRHB?xJ<2d`VDkSd`y(_H zIa6SfDnCAD|&_kT=kc zgw7L?FjwF{7Nae9q*a_e*Pv&RCl-{ZyCCC6}Uzefhw!jWRI$y zc`JO*utq`LNXp{zsFx-bU2 zG;X|8mpNc6xV+B}XNRY7Oq|kJ%Bz(m{3h`2MwA(jA~GAb{@NN$#q#z|x#(;{uj#jX z94_gBl)2AlbUqe1nk9C6Gr>0W__IR{45yu51mfueC%I8Z@bPTh-S;amfN$Lc8p^PL zpVW1Go#FOk^N9cV5vT< z_+iIf_Is99^@^x^fLS)azy7|msk$b#%zkk+1?uNMT-ItVgE+GAxpATGBK1wYO>C?O%I`e1f1c%#F(yh? z$~Naz!>`{MrhlHiSsYgWP@%cRz^V01|LeMoB>}=S02!D>4t&?O<1s9&{>G%ZXLx{F zwB=xd^oC1G+KzQ7assnC@MYmPfPX2N%)=i%nMCXK%}Vs+_;I&D3`fxJBrt(K=wNh!bdI#-&`}YFT?klOQq-mRwCT)3$7gW3%_^sHazHcQ|&6McMl$yo_AY zzTnZLZY0831Bf1(Fm2(8UC?W5NSXrQ6|IlLa3CEHVxsI!h5Z*OeotxN^BRMRMK5qy zqu3EmL|624FC|* zkhae5d6k2S#V>Go!`l&aBIdk^ftp9gVXR-NqsSek!v+63Hicmar$f}16gD0@d%If= zwsc+*>VZ@>yOiS}6fGhj%aqd*`$icrMgEIBf4IKyxb~dy((2<@3Y_QTkV78!y)9L% z#jT?U@eAXMWb*CJeOpWIV-MgT4n8i#Ed;SyjSC0a5tz+=xH?~6mr6|KgzqU4O$*d+p+oUtqqb8gs5H{$ZKiZ)`bz^>07joF3&F21Y z$2f12OT;y)i4I#Y!8Qu1KN{Rbd?%;>%PB$~mLIO5IjMf)8|zq$SW=**T#TZ%%Hc|D zsPb#%yaPMrt|e4gLu>MqH^ZSu!{%f8kVMK4O|^0M$4ZXT;nL!;5ar8L6$k1wYj|q& zn(O|VH(u5GZAX>(u}Qjhgs> zdA7jvEy~|C8#=+iXurVzhqg@H^QxR8vQ{|McM1u?$-G|VK|41G>r(H;JA81lrGLTX zlE^FbjsiI*V4$m}3`q_IZ9)TR8F5>&c|(JA;sSa&`n#?-^;Iz_|8V|O0)T2t-9DOu z%D*}AYB?tDwXEsOoE7pGb5wNk(p?CIO!fHU%KO_+U!(T_l-P(2;zAP|+HL1MTp`^K znuIeyAjv^noyCK+)d=*{>)n_zGGJj&N|ceISYboASc)byPFuY5Q!sTR`Ha*&#Nnp| zkck!Zun#~VSqqy}oznqw&30YZpK{DfnCNbq%d>|~8z0b0q)9`**id*t*@2C@(fzKt zJ+sP2xLT?Ti7iEI`hqdj-pN%Q+!0B?|E=&Kn*?93BPNk-qujLiE-16f)dZk1tXQG^ zW(*g`@*UA@jje)rrFZ=I{(Ex99(~RE&Idr<@k$5ZS_B_!u7_|_l+RkL)mu6>7EAQ7P^#q6|}eM z!F#o}l-i_9n3FjrQFfTZ1idO~zT48k={j+l9ONKlRneB+w=gEQkQ$*^90t$PJ;-)b z7@$`K&52r)a-gOJYI7UZ1U1?0bPZz+1IfAE` zYu23i7w*jJ5p%|I5d%cpMIh}&e08`w!0ez5gyWhJ>q>)E(4)G&M87nc>%1)Ut8}nz zWLqk6P^ZMo4wfcYhIy=^&3k?{K9ZyKK#szZn~Xw1n6>&Hp;fvo=i%}N)-3AezM^MS z_eF65>SLNb_gqJ>r?j2zs+v={ zqj!%*r>;=MT27Qei_dtGUjDrGu)j;_XYNT;Hxd0M(ZH-_55{wu88DwS#RYpv!7je7 zFbRL|MF}ap5AX^f7&a}*N+h|s#q?0F*zwKcII^QOTdN%|eMS)%U~6_Z`3vy7xr<28 z5W<(;cax!!kjnvt9q~+?PuL7@kkE(@flRdu?`-7`G>mv?xs(u;dD~_+9IsMR#5hJD zKc7BEb7@%+SPvlkbW6K@8xmD4N?u)j2k+^U%sV`|`ajoklS)wiOkX7)dS3jR?n^&qi)y~>qAS2^<$-^TjP zzPT=$byUdVP!C<8ECU60OvpgbmVrHUL!y?}%iuH!p8MC=UmQ%kb@vomqdRsyd}iM< zP_6hEQ--0`18DdHs`{p^rR6xV&$qF0Q)ry$oD!lt!0=XJn@0(>1wsCkpBYB))*63f zfNvDpf}`Vv$)Ps^&@otSjH@G9Qtfq!IatwY@9B;8_($N=ir#~TScC10Ewy7nV2W6#25o-wukV# zxZ-gl1o3w=o7Z577uV1Qhaw)X3N6k`sy>R%n(UG8WU3!w7R7#jh}^tWuah#B1ky4Z z4F+!)U~9t5jYuAOt3M<}OuKaaQ9o31cq)N|@4|mP^1KEAqpW@?tf^Tqi>MV2k2gFM zzE!4VMR!4GRxhc~nm^Mx(^|y4EhI6ha7sAdrZ1mNXmHTxrOXYKA%hrTqo$QBE^037 zd6L%EPh%_(vl_w#HyIINoD8;uiN^AT&l)AMfZ>rvevFP*((GBh$wjFHtIR7_7yDa= z+*8=KsY0!r(oozs4%sEsd#eq8yNr7ICCU@@+n~qp-`efJY5#(A?EZZd z%)YwV{k+`gIpa?WTF)BvyW7)1pdw>V{%Do$hVzdL@JUS*5#Q9QmMqU2oVR*_0a$I6 zcj!-W&rpE3Ku5-!{MjngT^;PJ3-Fnr6A|BKlr#=JYFLc%p97E|LR8KM0XCxLY0Ruf zRMz)e7Q<-hdkLoLDmQQkP>$XogwlSr+9@P(W@D;`_dA)bk#S=~{u=X%UDGYw^CG&A z@S;+GDbwMP7@2FIoFpD#0Da)JjFcl!H*TuBCTto>AsilKeKNiM2~T-HINs~1-A#lvVp^4BU9~Kh7#wozZniS`f}ZE&bEgNld^@*~ zFA>q7nL7Z-Wi$LysO8NI9eK^L3TgqH%&?%kuj=d_7M4mXAV)4hqRr9NtDjYEdf&oY zqzrf`IT-CcF)B48X5XI7a7OHeL2og1aWQNpHBf0r(-K;m1&hQU0qO~1Lb5*B<`8C) zW^W}?*WZ(sJj|abC$^W2@D5o0mdvd-#&vGdNIb>gvse+*)EiZaUFptsMJ$5YF;xtI z3t6woBgK*jNyW{Dl;m<3T|v~>vra)2D=m!*{-|*HlDzM|kOTkqrNxAH2uG`y-SEeN z&(Mtz%bP=zpRNhJx_~lKL(`MjIoHryOy9ufmfZ!HODeD2JIV+m@(A51P{?g~6Td7& zSKysZ=>wcgOh%-QPrpV%x~Lo}WmHVR}p0w9h z?`e+xVun7kE>60cp>o&GXlP(tj5z0I!!i9mLsuB5b~Oa)0*MD>N?`}gf?UIarQ;{F znk)H^H$bw9MWxK3C;tf%dy#6RxAqC3v-!Cg)k~Ik@VemdgjGu+__L=LsDVWavHpf} zW&edzDjc*Iu&4lSMxlyQu{=N?D>3h^DE(V@5oS8ARNsD?>`W zuNZiXEUsC!zOr=OZ;`uKx6#4$PpEUa1a%=z3+1Ks0o$C@uo|_n-qkrL4`mmLHv+uW z_{fmw9$xBVg{y^jtW2$h;tr zooew7mZ*D^=IcPMf?Rj_o>JVI-)v>j*B+2&_Cy$o$h>D}6Y$^Z8&?4kv_*96YfITK z$2zC;H0dWV?P;f@aHIo{Dayh<6Mx7iHF^p>4o@XMlvw&558=Eoz}FMpiVQeayhG3}3)K7J zCrrPvt)MMh3qx}LdLn)8viq3Ban5~W%^WY`W}Hor)heB~sQ58IbluKaWEK6~^(9#< zljJUjcBubDB~N{^4VVMozURRixz4km;Hp|IPEv?;mU#M9ph#+>k&fQhVfy~Xa6tGL zsBzG}OGB=TTe7hA;QOxuCim*^+S9-USkYy#7WBPkD$e$UOZ8M`AnMD{i?RfpAS^Cy&es_q6Yy~xF^b?wI! ziJE%WUJcMYSpNs))79DbEI~NH2c&A|yIRzyjnV7%sy%&p`U00S?+4m3ZCk_{llp~J zW@Ks96q{+b$QZkuD^YvS>j;9g+|bEu9@+#uQCsi7K)+<(FZm>_?+Bs_9)FPGQ;W;W zi9nNWg*o*Lya_gF>vsF`UCdi^QZ(B3}lT#Mx_LBftB45EajcTlK?2l}NbiAJV|RTag1^_H5tG41AT*q;JCXmPT@ zE2o?L>A%^35tM((;hEn=j)Auj2HM7qnGj}&&_3lmqiUO=WlJqi82(-+xu?TBaCO*K z#|c5=KymX&pf7=8;%LZKM4F`GR^%uW$A6A%L~UjhWJPvA->`IHVKv?CJkG>@9)y~XsMqmY7Z}rxa zWrzWz53#&Dw24F2bP@`-2K~z?j_xR)pyl-Gw}~1m=JF4!x%z{B%D9(#Xs*38VH@G? zSCSRRIXc7HVfK7xh>BRP@I?Fc2L%kB`^SwjSQRf#YPIRS6}fVrOh8Bu>QE)A{ZKz61Ra zaN=OJ?P&(Pgcydmw4(I)8Gt7!yW47KCL)+yh63lb)ARlg2SEXwNc<0&n}_OPM~9L@ zVG=5@8NooJ6tZpoybe*pLILZS9e{Rr%D(0*b9ll-4p`^*ii-ZJVR+z3O=oRHo$Z|G+X9rt&D->)ZVQ=aI1aI#_6$g{-*drOaeLNV$R-Ub z*j+M}ds~MpWG#h(@0UiS>=MwWW1M)vUBhvOugk2C(IYX=HXJ8UQ9 z#5=B#nvuZqxe0f>_)0|(3fBWymtB?LHcA%LXfdsg(`txYB5Kkh!)MQn>49+d+!dS< z6Mx%Pdy1b$lK$JZF!J~xY(QPb4;#)MwG=<%Dki9?%R6}zB>(!FQszJ6mZ&DX`_|3- z`R&v2XlPIoOherGEc-j`m^CXplJc??^y)-}d`%W$P5lDQG|vTTEnTbqCMaJOALX?N zk_Me#kAb&Ur$@7;AZyz2?p?Y6;pp8id~nQ{gFH0ajN=XMqp`t7ypG3^F+guQYZ^u(9xigivo9djTF zTxYu~@fD6@mH4A$i>-d`&0(BD_GTbEgc$W-`=ozF>%FC5(O z3_+~yKqLI9rFu)y%8TLCWnn(o-TF%Z6;15#F&ojB{3Iu;-kK?n#tfJBc))!z!9e(EKV3!o zqqU`3hm2q?R3Pq>K3H?mjl(AOwUHHo#w>uaK1A83!$ zH0{tF79;Lm*WI0K<`<@)B`9zb21hKD)-iG;5w{mO<_axjk6D$32s!5-Y5^A#e2cW2 zt=pNLXUjwR=2j?VPv)5?W&M%Si22X1{+i_Jb{=idUDV!wLAs@6LGgCMR7^?P(W4XZ zqeJy}exhWFK(H|+^&=X?{l4$J{<3*>HPyZSpMLpVI~F(7h0ViGC2#!u6(Eh~awca) z-;hE^dG!k#afuq$ld_-yxPb^M3@c4#{a}yk1)*hl80-3`@VjX5V3B=cIw9BBQP$U! z&jh~_^IA5t?X`A7(iyUULs2H>J+kzW1LeK`(OEB;_NlWHPuVZRiXp^_aCD3{*sYk+@zCI zlH^{Fi58>h^h$>NJ*!e(IViGQd31_Vg$)um^f44AiZF{*3e%Qu`(~%I#Qol(RTvI( zBETlb%UkUe?s-$~mk)B|JSTI=xngeAO~J*{!D%UZx*8;W=;>gbhI7xH9>H^Y@M!}g z*U#)-0R7QAFs(Ztkw#f-Uf~TCc(>7zI?BzHLJ5fns2_w}{PyNW_fNtXms*qJ(-FV^i_9FD!9T3`nQJM6<{A~6jo((-!CyEt{!S6EP5`ZZ7 z6$u%LTuL|%2t8r^3b{8;=`cu8%JCeocMwIAb?5JVhA zTqi}W@b%b)61Bo#HkF6bdZnmN15M)>M&*YDqWs5%ePTPY`F(+pIf6Oms5#>~N0amnHt8fh?qQu2n@keos!K034a*Ez?@cKzl>M{rz)F`q zoQ5{gSAghsrrC%zS=|rLEAC&HyGhfv-2=6|^%Q8h$35)MzuX!~`0KC;`eU~^;4ZH{ zUsFHt%1Ot~ycIwxu^4+%V}9)6#W0TZG1q#tnG=1+vG%!ROGy^g+*|IFIGNeZ_g@ z`B{68cX53Gn1``3AZILZ+_pW~dg3$tIs1s&c=}1Ck@vX~ystoUL-`+5i@THMTMPl6 zC;%9vwVhNoQk{ePyzqIwM*g<8%I;GIPz7L9*IoAe8>N)H)RQZ+RaST2KUPL}-t>BS zfhNQGAwO!qEZR%{Mn|#u$ZtA+8rew1nj<%{HBBr2bYh~E-KX16Aamv>UM@W3T+bTycu9HhKvo-=P?hls%G0p3`V1j$5uKM@`I`4pI?6K#}3pm zNq%&a8tkaz&bs4cJ2>W`xk}?DFM_(rb+&c7_+&+B>Oh*3{Ktay!>3b(`YKxST)WkBE#vucdG+I|gR#*~^o&!J2~o=$SW zZ5la?w<;Yqj5Wb7Oyfd=9*41+_O6hWZMwVr6}dWLP#q$N6koK? z{U*wXmUPU;X$QlmUHGx!<@_n@*!f8laf6&)pfg>S5BsKCT*s1zv36iC zs98dXoG}`EAY|*~#RkQ<^Uyzm(vD6Gw{NH<3g_s)l$VC2E8R^eJT0X|xSmzrWDmuJmj2!XUi(A2x|D7$ViyeCN-1|QxS)2(hlj`u6 zJ-AF=Isxx$^irHjNc(%*hiUV)VkBJ|a_=qBirDm#rvNMu#etwLQF~HuJLOb${*}(k zU^;H@BM2>Q*@~G|yRT$q?)y1)!VVOVhvoYS{StRR7Q76MvbG{=C2R``oAz(^{0sIH zbnY`}mG)D^tA6YZLl@#Ae95H)bD68X8abRD>(SKKC^LH@Pd4(>K??zle*Ah@<%JC< ziHgV_BL%P!e17_BgHpf1*Bj-c{%?>=7u zMu?r@{tPzKGfG&Xlxu^yew@Z92WSm>%}m!DY= z9RV7sz$di|$o8zs^9c7`sL`9s8Lzuz8;GMQf5h+YAyv&W7~vCyaDZQ}CAeoUK3n;K z?w>i|;5lze+VuxnzfB#o7Thg2ZDW{2pM#ZSAAEFTB`fdb`)-a8HWv62)&(pqp}hpX zbORd#=NWhMPn6H-FDa)MG9WM;@#3iY5x4H$3-cN zddPM{uoEy_nD0DbBhH08ud#zw@chNp#@{?(Dtra<^^oKBgkYZ)W7=A=fcurr*Fr&1 z&T%oU4uDx}&azm#CvgVDtA!MFu=`8t-{RHnvgxk_=6k^WaDu$QIR(=*FR}tRqq9Z+ zq2UI`7P|Wm>=vvZrkxM$1l$(+yBpXN#?SmvaS?r+!@fb5X>Md@x%%%BZ)+Y?W{V;)W5MUofC1( z3A$VU3FFwn85QBww1DkwRtRCu+U=ekN0RV$x5IQV>{9$pE|QjEfROnv1{RY4Rf`0DoG8ue-6UrWks{&1kxLwi4sGMrpV^aHz((Dlp=VHq77I zLI$MjE-|6YF`X<=Mxa%UI4U!!c2^4tf3ZScu1$~Iv$WpXa@x)}?~*i?WoHw3-^Uhi zII0kz;rJooqULeU!jxo4YY*Z+d!9s~+=5?&I8Zh?k)Fs%vZ_23kh3W_bU|F!B=JjC zxgrxF^RAvlO8sZQjA*1PNk1-Ve@tZ!tWuqYeHAK{gG!i?kv$wqi6c3MFN!cM$)8`@ zDk%1HAZ$tRJ7u-BU<KDmClHHS=|6h`CHJos=IlUx^R}3b6a1&1M3s`ddRu3jb z{UhWXsVXE90CMM5VGEX9!%-@aQ6?!2x&9sqr1lw_L>uH9S&(E~zZB=b$V!wQtopLu zBLDhG*}oQ8IdK&dxtg*))XnW5}5vT0fUN&2V0mPxA{vx<-LBtz=) z+<TJvk0xpiCDK{DZVlR+6Ym`OqL_qAf1L>qj;Wcl~gD zq83Y1R(r(AX*WI($E16{Yy}GH=ven#ebBdxA*h$1O4YfH5=J}}-?7W4bki2-&$t1E zLOjunYj%!eiDm>c`P^mSF5eNZyi}r)VP#ih_qW8fL+>U`R&?K!%{QoLVOMFr48&7= zc+c2qOHNk3an-XjLs1PRUkxP#6exFBCqVN#N)Y+zB(?IQ>jdr2JB^(;V;au?oElHnnfqkXoGl8_+( zi#zFwOb045LKIbM6UTZj7GU^CBvRtDdEl$alsvfw(=cgNq=eb%s~`Cv0ZHlQ3XoDX zJBsA*6OC6z`dSHc9#L}B0BHbqASci$(7$yKo}oNUp?{=JE+jg%g)ggr4-Gu*5n5!<>BlInrXDV{I;JXOgT zF!0bvVU$8_n7hVl6XG_`=+K)&gW#O|-&F-7MJIn?Df(rSx z1A1*AbZ=G%ddUlLx<>$vv2s*^vRP^M*5#-QCPDe)W?Xtn^YDQm8cgWfD=KMl&=&uc`wPztL2A!1Etl`iO|+MAeo zP7~!Jf3+BfWq?X3T(g(^({GsLbh^f2O({wpb~Dp;bV;=5mg$$z0pi>O=K;9I&wiTnnwXM%uKdFVBqFC;c)|W9=%NY zM3`A;^M58EDLz!S#|LN*90VaklNnC6up#4tMLQqX*IDzjU5h^{mvHYgDY>F;qdr=H zr?T~#YK-Ndq=^M4GqK06g7*EA;|uG@?c7^uxi+heHVUjLHA0GO;ARdOdOmwn#n)sF zc~jxte6%r~us|RMPjdI;>Xi^7ut{uU+lUnnedsIfuh<3;8kxH24xzv~%*qP-X|?uX zn75?X{2{!OmUZj&D3PnoNQSsYYOHCz_*1`;Go8gJYgT7~T{d{-%*)Guh5y$Aj8$_> zIMcb19_Ws8(1g|Y^g)7hhszF6#6ug5wgV>sej{Xjxh3*1KE322{ea&dE0 z5H4O1&!Pkm)?PBh(xVTvU0&shP(ew_haPf@oUe$Or^{r`xa@HsG)~fBvph0OEv(`# zlnk@q&E_%yTa3@HD5ZGv*kHK~7-|>jI)yrmU{t3x!4pA?7GrOE*FFTJwS-zT?YS4g z+&uZXG{^cF!ly2eA`e8#Zua~9M0X56f#Y?O-#Tn~+Q3-l(=;RceZBog3s1h-xmn}K z%ka~;w@JV4HJKXNqF3M+aPu zS_q9>8e6~@M_m)Q%!?@%g&$q@*JkAJ;CmB3v@_M z7VYgwOp@(KT}il-^F_atER&l&j08DNlCCv83bkoha^`@OfuT`MQ zb~rlRd(WJPvlr~nVXPt4vRrTF2_q&lwkezd!}K2A>d~>hn3i8P$}gdFZG|9DFL~N?AP3|;guNcPPYscFGq%VVzcr;trTj*ZRAD} zYk55jVdxO*G4(oxhR}{(lk2t*aLgGv1?q-%m7PSrE3kD{Pc5}g%%~|p^Yv?9d!h+> z^1H;N{U%iCu0DM7KHyWdm*Zm_7}dANOcKKAYcGzr;nE-VH5OqGK=0BNPVrUQUO0_s zwGxD%pqQ@YpOb$6KKbYPrj#6XPS6avk>jG8HDsiIFKxPvvlQ{qdos|eL4UKLDr@m& z(E4GG4#6rL`6%ij`U@{`>_VSIyO}pYuUujTWZ|fN3n$=`AW#C?cTnp^4hooM0TO>Q zfrfvG#eQO8ATWn#TkKTUYt>#6>=ee=iQ|C;XXwk^iK8FNB!)v4TC*e_1p=Bdq0?4* z^_Zt&D%llsF?Mv0MwX(f1=I91BP(>L6zS9^T@vUX!IcJxXYpnz`hHZBrfPOw~^U7K5wmPpz9|+b+@5=z$8XVp@2~+s9Il7 z3u2wG^zast3ynx9?l3R=p>XBRZbI-LjhT4=sIU6*tr>&r;O!Bz(eE3`Ci|$1(}$|- zdLiK_KDNPU4w~o$Uh+eTg90~O&#O-{LzY~|>Zw(RrZ%!cfTY`Ya7ODcVZM6ColrKs zVK%4VN>|}%IfWghu7O=Cq1S2QanoHKq=(9?g@R|p`;SW}L97iQ!b83~2s;^Z3nDKgrR`V ze)zRW4!Y_ocEG!XEoLPaU*BlD_qfZ(L@%+A`wK=kp5Jz zTv?LoDF`$e7pO6&ab#Sq>^$fby1h&bgx!kK84P;2Xh%bcX#-?!Z4dbh6kD3S@M<*) z1E?OQ357B5zEhp>@oaZ2>CIsMieUrZirms52X_uXd-m_m2(H9`bQY7}-CK=3&Cx4J z^&$4|PUmAD+$#4nV~d7F6!gt;z}R7=)8myqrs$&O6ic2-OR&SL7d#Usy!x#X##8|H zJGrps`KXd5v-Xtpd7PmCtQWt>+V~B-vNPF%x-M8ns8mKY(u9h^CeBvg z_zW^6)qatAl!2|acp=% zwBi6)Bd)lStoU=CqN%jT^$*Q+pcsnuP?f~uFjP3GSX7miHY%=P%IJON2|;B#x<*hj zk9V=7W)d0*(TmQ^Bo(}(e`w3>MplK8bfA((d!v%YlbVu#*H4HV5rM^vzQxC^sqZWQ z553W$e}M!PaE8>dsO#P)s)$xu>@EaKE;s*~AX&V$`R=Kd)(WWV9I$3l%y()i;bGI ze1gYB!ayiE%rEM_7ahmUo8dw7RxNZn1qAcT$-W*1vO*lyfZaylHnv$Z)4l{oN~xj) z%ao6w-sdVr{x0$Em0kmzJi!&4-W9|cp|0gXPQDMKW~5iHD9NL}Ai##23@htPA&j0` z2a!1zIR2UwlZ~!}&D@v%7@fP$BAHAs8%-e@lPx((zE4Q9?-0!f3biPn?pTrT%dtnF zL}oh!k=gr@|FFJ*7_RsdjRhtXiKW#AjgJWLp`cgYN>jR0}oA)b_ zpFHuK2DqftTWnYS3q@beIE-1a%tJc{VX|sflD~GEcU!MhtJUn`y zjVKn`P>qHD&PiBt)zPv22+j9tG)Q7Mx;Eb1*@+axb>*inXOydWWFzO}HC6Crs@O}P z&7GL2F}DNM8r9DeTX=G+VF2_9>Y!m)I_jP(WMZz~W*!6c2ayh?$JZKQb4AB0au5iS z^3mQIGM6ZS%mLs&_j63(3m7UIKd)U6AN(Uq9hiIdHo;hV8m)Flzi$ zIBO;XLUAkH4SJ>B`x(8$qW7Ls010iE6!W4n`DB0zyq*qcS4oz7D! zEkhCAV^L=SOBy65<7rd|-LH3-QqK>@dF{r+Gm!4J8_DqgZbuTIK+`&wm+=MLks&qP zI(SfA4%8fCgxX^+Xl*Fq_ug0LuE@yf!$4a8X12h<4t}W2NCtrRxhstuqbwW5ZqvY= z7!x#fUBF+6>Gu-{8{u6$)$(WNdJB&55jA{YvZKgL=Q-`v(1s0w{+XS7SV6vnU!t6V z6Zez`kG)yY^e)##itXUJQ+DNn?H9g;$|V0`d|#{Fi4_Kr9`;kzJ$n(nz!X9xyQIdz#=@q%YUt-yUMd;7I zEf`I9Qp08GqjArCi;h>DaV8LY;O0cL2-|me^2787<7)Qh*F6KXqYYV>$%%Qasmrp5 z_CArvEzR?^b3_yL68#T$ZfE#<&Gu!86D-@h@2{aXz*^|g8GiJyXz^-#%f}|@ynr|k zcyiHtjrI|U#xjSh?zdZRcopL33faX?rBzQ*lHceK%)`R9H9<;`6HME(?{3gbv}-!x zuc7u9@0(N1ude{&1he>s%?s-ob9^o`1&WvUdkWMN`5FZ{CtST`yDWq?nr)0;00&&- zU%?qb2OI^!+`mFn9z;*%V!=ApY z>y)alx%BE$y)~@uN9#f0pF+>>$%Q1o7b^ejV(_-9! z??QHrY8h?c=5CNe5YCYV^^2eYTUn`4+*SMD^(tB8;$ztl^fy$v_7_X9YnL&%jALw(!WefvdKntZVXZ*U4LRTn4GAptH0@L&H8$qJuyYI}ruwG%ir&m14LVXeamQcrK2E zIw=q+o~exe>%gwpB3UvqAq-U2Yo3Pm?-9QKQR@{OS$2YDMq!-} zPm3WYvvhAW#MAdF7n&osMB#wN#}5c&_~XvyOHMI#h)uJWM&Nn$JJR%q6>8f9J1rV= z@-m8k-O98bQv)NY|GgJp0o@A&N9<$8=21fqrD)*jZZNv4X?&E*ojlgU$L zvT%M)wS|9D-0bYnzgYHybcATCw-DivA$!j!bfXG!cxbMWV*fmXvHG^zEbxZfY!4er6Hwo z=~xWRQ~F-QJ^$hRof z`oY#hw9g|fs^lUh1;V=9K*V;uSwXAG%;V6PFQFHi^|RA$n4Iav!{fMYL&`sOG^#Hi zA1m1UecEkjkB}-q1msUBHTp-X*6k^Y!NM)wnBza?$pwc&?1eIKKUcFDuB^r{vDGw#ct_fY@mSeZu5xtr!P;p@M+)@@f~< zBW6OuIRNu_0J-{}ll9_kH*$Vj_F zot5&T*E*0-%0z&QgSU1p`bP>x1EmSu$4NlCPN+EMocg37#IZK^Kl%5sc{H2@4y8YN zy>L+xwTpKDZ)=v+@9OP-J-0P;DN)WoqeLYXp5SCnXC|;sOisgEghoHU;R9wA9QRzlW#)cEoyCoP5ka4MG%5E3(i=M&Y`uT$5UJJJ z5Qd3%ORraC?fyK!x?*Jhw*RR`J7jjq5tTA=nX4y;e^jeyhFGIZt|p*Jb~*7Q6NT zSJqiZMcI8}Ujzy17LksjJBE&-yOEYoL6EMY8wZB&?ruRs>5%S5LX_@~ckuE5yz715 zwZB~Z+P|~+nNPE3)_tF|Tmb7`cvQLC7u1>7nN~@QxAKLdbu@`!CG{Sw9OM9ohNByh zMH1XNYzT*&_j5kR zLA9Wm$``*|b&-F*Q@`MilA;RZ0loMkiV&o$c4Pctvugx2Xh7L$^ejwQo#7pa9L*Ei zgcGD%!v3-aH^y&))Gd}gWrHv8k-;W)Lf zNG?%_HxtjS{kTM5lcP_`!;}0;s&%wUpG1gIKHT5x+$(fvS$J6b)uJN6H5AcP)7mc! z&n7ccg{7_D$+oPsOjIuC?iS1(Aqfr|5h&645pyN#{gz5Nos1-m zxX|jnD?cpv9NjvX#&iOH5PsJ`p5Ruj>NXMwAzKlWJVU_NLgK2#UIY)=ZM7a}`zv0@ zm35#y+yP&1ZToEZ(Q#}Xsd7}FI&nGqU=76auF$z=RSr^21w{qMTq7o_z+0K;)()uY z9WE=Pd1fOZChj;bLw2j+86)P84%CP=1kW!Pkhp%@B39w>g3O4tuSdA1@!HZOLKYU7 z*?9N?*<9EvYzC&b`e6P`bGVGV)X@E1ol-YniBiHWrJ*C7mY!4LVP@+%EB+Dr?=IQH z_UR<5LPGfvET145xcp~o&*=T|L|?u|eUTpfTn27Y7S4qjhdc8*2LH4$6?L8b8>`d= zEJr_}=o@kjETk~_hK;X$u(Ub@sL}*bMg5SyE^A8SD^=`a4;Jr+TYvT!or;E6DH{ad18b=z(@v(Mzj zHjbBQXk4Qxbi9{&d~^-hwDlgA^zSjOL(IW@@`{8xYdt(SK7bLZM3gsz$E?V^GOzDxMfSo3JKMNIpp^Tmy&&;iE%9i$vS8YFvb7E<x{LfRkC<4t@U=~lz5w%`I5s}G zpY-I2CO%IhMz!93y~|$C>sQ)tnn8jwjLi=er}1K z*}s;~#ZGrBpsTW|GmO;NR&uFbVf$FCDv&6sLM_c0bI7Vvj9D!b1 znktYNl-;ObRXeyXYB_(w! z{spD13qXI;#G-_S_ti3t-VWv~E%$5wQs&&9T$!u$p&YZQb` z$s^wb;R)`o_FsOMTw|5Pvd%GCu=LDYgi8n6*5bK;G(ICnqs^us^_%k~$?!y!aYCP) z4>_QU=#?a$xY!TQ)NhVC2@)tK@NhHwh@o-F${KKa7tiV8jj3y0u9mLBLe3-UMjUr) zQZooW3Gv7cQ!Y-?!RJSKKj?ki7=q<1nwq!`JAQ>bLfWD^m`3Ht`$5hZh^*g@X^Hfu zv012m?lmn&J*+L25u80qq~{aDb5U;<7}r^92dE>|w*f?qi-yf%kj9cvvu|)-j8ALm z<+=7Wx@?Kz%h5N8g@2_7$9&dsIcz@PIoCH6A3)C(z>2v#d_3nOL4Rj{kG{r0oj!z!iMYI6{~3yU4N8e~XV_S^8e57uKX zTfH4na{j}hd7i40f$i_3m(#))b*T$Uzl#=%b_u41|Az1G)B_PrKPVO7x3pF27F|FX zeQ{)IGi|ckH8pPk7ZvKa+fP~`=VI42(k5-;(IP=PTDb{eSdW+V8G{EI>a6)KSHfB=FqyxQ51tTQ>JpOz){#<%!;3f?&_v0^PK2yzg0VZ-> zaLo!ycY@AX+VuHbw_Zj($H0V`QW@N7%HPXZYzqaJhz@ii>5tcLvqZOI*4j9?Ua>R{~!qlJ_>8+nfj)s&vxEzY%FM0;w4Iap#D(Fmdm%GS_MG(qt{%rF-Fj;l+2a|usd6H zMv#f2BVl#C;~MDG2#&)s*Q^6>Db)v(oENjIY=X$tN%I3-D%+7=t6Irt&L~KVfr0Sx zBKL0k-mww%Mx-^^(|7B`TjjYL?O&@B()PtC$EM%tPNm;d9;Xib>--oXVKt(}dcH5e zA?U?3{)nI3CB}`E6-bI=g5p4m14cjzKoAQc7mFcZkixR0M#kx8!I4-N7NrhbrVW$x z6@fYbL+Yc+M`W?6!wjgyTH1mal(72nvDcH)YNE2Rhyu{WLdZce6#ZDtreVFQ_^~k6 zz#*}`B54*zVi{)RD`|$zf$$frkEY;+{a8$A`$9;sThid zUMvGxYC=eqYkx??fgqI9z8V(}`*HS`5b%bz2j%V8739E$OOPVAk5vKirNm$$>2tRQ z37M~~E_a6Wy+h9TB++iry%Sb$h3M?L&Kx>u3Fq$k$87_K2^ZuSnh49IhhUL**IXvW zy1DvVl{TlcS?Emq$k2R?H=)TJ1L_sNDrLY?v~M7n_dzRD^NELBYJ0kO{+1z%Mn;Ok zyjjdGlTBq!5OZ@3LDJ{8PZA3dfIonj9W9$AZh^`yU<)4zD0q2_g_;d#UQK;0C<>p) z&P?Pz>*t+8#MOFYzkZOz32t^h54q&Om^+HXdp5ps>sal;q*SmxQa~+vyQjxX`da+B zZwJZ`9V%I@&}Tf<9cVmf9IL6;vGR$+#^(?f@g_-GTF~-v!5iA?R*daHc)N#{Y=<2z z^9J4x`jR5Y-{o9AgI0-@+RA6ZUb70qrdj^u61jx^1z=;6GR z31-$KSd~;aF}t7l+HYuItv%gWjZeQ(4T6)wI`gmVKk0_g?$NAEoW)rG4C%KR1 z@H&)NU=)27h^sRs0Op+gsWtC@!r`6f^Vlg6AI0URn$|edOlD;4SU7e;=o=*xs!rZo zjzdK1s57iy-|IsUcQ{bWcu$g3!7Yq#wme9)NgFY>oj76`XHYcbMNK%MCxGu^8`|0+ zio|$A;1Ss>xkBj>+tjazxn1G&$C!Q#VL2gKZr3`zkOVJKzK?B`(aUF^Ah6#3QST!9 z$9TPKDYdZP`-zcDg31EBE=;GJ$%KGCGWq=7cbjr;Vwh^cNY-rH`lG~0fq#mU-zopF zn0@6KFhUZBGLo`Ku9E1teId1ute5=XtcuXd8J<6e4Y(Xk{-nsnCuTNsDh~AP@hOrO zD8JUk`07+a3X-DgNaJ;0p&fUy=Wk@kj=5Nb%0C1k)qsg;21MhSNu|3+5r6kJ1)F++ zQ{^H83|qHA_F1dM3hcHPSZ8jYTdHjCJP>|dyq|P3|8j=|S8K6D?4Y|L z_^AdAhx#*fxgQ>6~=*!)1l(DU)+YB#B) zc3qnT%Mr(crdW5&vGSLOZxd_tG3%(l$#t667K`TKz7Kp%j2b07jR;{aRQ?KHJ?+C$ zLrI_2I<47EY~utI5^c`&sAT>I5RFx0eA+88g6YsslLeexMWDF_YaEm2=e%3kfmilF zGEK8dvZARyMD>__RD*mTNYH=80`pGruWHzqRUR zx{KCfVQn?7(0<)b%MH8J17{j2nHtvV@BaBamBl36ak@E3stla7>E0W@vx6(;gRRopVY)yaScUVMthb^n^K&6^7Ij z#lhR9`7~5Fl{a2uZ}RiXGNU99amx6us=+9nzU*>2v!gfz1XTJ=uR?VSzp_D*98Yw+ zoUUZ6Dlut}v3KBb+@ao34Gqqj88oKYCNjjak8w*Smkk(kZa97kW>Q<Fvh(l?}>W*Ec$kBoSUF+hsh+Ed^#BtSW z@NEASZbDtvyGx4id|QBJJIW37~0NOJ=N*jSU33=mmi! z@NZo4U-PR5sRyx21kJu*mB3cQR>}iJeE1tjprk<%l)FG@AklFp7It)2jXXr+L(NEQ7Y_&2Uj&aQZ1+CA;ueVLZgHmA?6Z0C5HcJ~;JK??FBxagZnWfmr@**E zVa-dF+61)FgQ+m^yR?Z`wVhK1PtM-F$5hHTbe9*agUmolcWegU($yW6+5jtoT^$>( zg4XFEVKq>>oO;AN8+lC92ew~GrnbFJv%ttWo~Ui{4vBW{>0lp7Zu$c~-qt^XPOx?R zkBY?Fa?#6Fi8sRC!oS^Ao~JEJd5O;l3oswPM;!h&+@Iu{63=7#0idy|v00|RS+AZ$ zS@b8Wv}1W^l7b@ABfN)U1P0v`!Kl8;rSUJJ7wcBsUjhrl?0(xz(I?+1gR-dLpDC~* z!x7$Tum-Re*b%Mvgm?@mjqfK*?7`gd6~q4QLU)-X^mwy)vr_tpqcoL8I;;^lZDJe= zxa3hjj5T>|uXt3I7Imc7k*6LpbI&*QTKPp;NXhl|W?LYD6;&X{7U`$#Y{%c>bXY}| zTc0@YgiqAZzr}rtk4fL0SmagX9TR|XvTsKM#uEx}6oTlX`G6iJ77G6JC2(KNmv@g> zGghm2*cQcSMzBjXPm}_ze%&LC$C=wi(ect^J)4Kb+e_XKb4Ig`*U3wM`wPzlcPhYeq!dn}(#_Q+H3Vah2m6&zGt-(V%$Lhmi2-?MIug3`|igw zg0&pF;yBR+xL*0?;V*-!S;=+%vqx(zwor&I{U(e3 zDfYOYx&D#?%~9-OxgrY$k#s7xRo5y1NfXVPBw%7}FSPpKdWL!D2%Pbgp_6E-u4imn z2G)BY8o-hg>MCbO%7K{@`W^I>;*=@Y{a;v0*-SF0~)(Rra*fk=&s; zkMWcd$|qGDRVyh1{YewI1YyN1Sxm)m{eu5C(t$g*i}QdPndTl}Vm-9_1P7csQVMJM z%%U``tvBa1*8nQu^2Yi*EA@5_UF{mgl^?rQtOC5G?rE7s6lk#L(Q`C%8H!C;7wQ}m zsH!CF9quB6?>gfH*=~O-X2%a0^B#7)^NxN603^kjLQD+%@{o^Dmd`ar)T=1B4aq-V zZeG#JA_>J35Op?(8t?AA96rl8WS`;rDX#j<34zuFPp*CutnO8k!`_UmV^qkspchYe z-e|z=oRpPQ^ZxTvxlWS==Zp+>BvVpnPr^{d+*>PU_2xGHp!avhZWMh$=rg|or}=mt zc5W?N>2n=bf^)3ciuNK~P??C9VZmYkp=QNG#DLS9n|$%|?Mhs&y+u!ZmC0xXGcHen z_;zyH=*?>_@Z~`r{AjS9_5Nn0wOsdSdY;HeDSAyx`u*1uM z!j&0_Q%l_E>f@nJ=g`8XTC+D@$s;I$=h>>)!NP6!>E?WTen)TE^{He*)SHa ze>G)@b7|L9F?*e8SJLp~%Nxb+i{m_NGR!!$b+av##08M&81MNpl=%BKmXtp@$G}K1 z*;Y4KLHKmV>Tnz_p=m$-o-g-j24D9l_VsB*$Zvog=h%zZO4s?~Z!~rpHowyfCT;{X z+zW>*!J3*gz9LceHWm}15_BeT>)iQ>=klRwy|)d+#p<_d4CaUG(1%zL>2NYHR$LL1 zU+=CwTpC|?@i!;gJ~%X*kfoV92fUM?uq3YB5~#aCR3L%xt-qnN$7V4F4G)bHp$Ks^~{u=vTa0%(xLaWHKdPpa+U%>8z##<&Y z<-tRbd1od5U#xHUWW-g;x4QICFR8H$ zS!le#M$gh*ZQQ(ZgPfx2DeRSx)Eixj`WAg7lMs?6P}<@^U*23l;&#uV$~axmIE(wo zZC;Lms;mbqw5{V1#~z4UdzEtJ7tK@e{&nyy|6{PsFfj(Dz8dd(U19BA%J*|L?5lgu zeKn?=xTyKt&3H@qy+?{j6{+2&$|iLkAithdA@Q!Gx(I5z6mF85_;)If2*FrZC#kCZ zkNJ62r>bAo>BF{O8tamhr2*b&r>?X5U_@OwQ<7c&R`uC@XwNFtj&Ku;LOmNU@ID3m zCyPLTrN>u#dL+7o@V=jH=^1mkjY8~RU&cmzE1cHXO(uw7Zf?KF9#C#_JG1n-UtL*u zwi>!sls{^)2kjOw0HpCB!^b9F?f~L@7|W9+rDjXR2UuS~dl)qW-Tx64fxIG5m4-F} zjZ3L#kZJ*N{ba=)>B8??k*U%Lm4Y^mIWiM)*geHX=zm0x%iKuNlfSS+R_b*aFFi-b zUwm_(1MdqiHn`7`yX~wN>3+bv1?^$h{739opW{CH4Q(PCm%{3hDv=<+@)EFFuk?`P| z!#+w*<7Md1O4}LxMQAB(uiqX(?AZ^&O~+_g?=W0kwFFK-NL zp^P1qz%FYQhWKc+&HxpqA8XxsksluiXK1Dv1P=?RzOW5#1@9z>+t6vh5R*mRF*7+O zZc3Xr>B3X`5M^9Y(|RF>dtrong#>$j@;m%xUK4o!ng8t`M??0|nYt@;iB}TzR@ypS z-ome3B-zGw81zj#M?=>cQe(ZVP@KlG8!*)y1#Wg{pJ>z^&)-A;wwnX})PPB@ZeGcE zfyCd|R7UyryDRtL==4v@4zweOt`G$ziIra14ek5;j}~jc4X)3_3sF=s6OSBYzZxar zT!qW7iEN|XRZ~M>H!Q2Z$7~iwxLf>nnZ71{WQ%b%+*XD24Fc0tSJ-#Rz6c#Iu+z2I zz|u!)n0GrX13`Z()9}6@9g@ZW$A|=weRIg*Q6+~h`3c|A(nR;9kbTenR+j`Qu%6mH zAx~ZP0B+7twD3{cqSw<%s;lYezNdQWxc@J;X!=C49F;9=B*Q3H7zUf^Mfe{~1a>p# VKh2oyebz3)Jw0$q@El~%{tp;n>2&}A delta 43141 zcmZU)Wn7ir{yZ!pBHi67UD74p-Q8W%&82jAOGtNjOK(Du2I-OpH%QmB_4YpJ{QeK~ zf|*(Koy99Y*IsL1_h&FwXD|S16#R?dKhV#|T}v*6I*HZ3Kn?}tS6j7uSRGjD_sZXS zEi>8WGETpCrw;RHwbu&kr+!I($pMKM9W~f6@4_j>VbNYZ zH8U5(57fdLoj&iO|GGR|Z z33}K4lcsBGF-FM4=Of{;i|jfnQYYvWRkriuH8gJ*uHs#{zADO%XIo}X7)l@LsLeUQ z<)`Fn3~6x!q1(!nlTB8%GcxaJ?ytbcv&(GSx;@?`R&Y*c)KG*lz6Z3^BEGjH>2qp< z5pWpO2^1Ob+-#RE19Dlkz!Td%CS1Tu|6muPuK~BhG`Mg^;a-g8pixl|*C+YcDdFu& zt2-SjQrCODwwZ*0rz7DVO4nUQryWI_eR3hC6{YODy&;>O5hzS89|pmL%RaZnf$Yx7 z$)GQF3US2T=UFy8gVyMGxQavzw@W#P0xczcKIIqYKGOnoK!EtB18s(aV4Lnijx$TN zxxkPiJHkC*r%-i)(QUuXt0XeP7EFF*vvE+&-oyID_;p}+Qm`WXd6wx-j<})liFYf2 zx0%}dYvwE~g4^xHW;%k}New(`Pqc|Uz-{<2o{^Jfz+1;3MNA6A{xfXf^_WN^hghJIq1WGMbCE;q&@46ReIy*R z0E>{zN~K`kg~PoYGyrkn7IXC=V1xHm*bx96kG_!uVflAlCaHk)=!&)oQnz8M>xg z*KL~Y^{zL^;*MMHiM1Dc1morg`|KZdHhFlt6S0$hY@fu;PEFo=z&v|$H&l}vcDfJ6 z67F&R1jKzl3o218NG*p=Ox=Lm!1>II1j6n@yb5Q1yiWj8mMwFJ^|g>Od~K^adbopp zEf%Zp<`=M#f*_cFf7WVPK(S=z%dmByH?lvzvlkJuhS{>eR6D&^|1l@gwsoPyv+PuM zuly4)!ptIJh)B;SXlJ_~Ig9w-zs{I(Ca2N_I7mMlzzhE3JX~+To4@*Yh-$E|nV(q6 z!c8%&jl{%6u3@+uDUnFXE#UUHOAv}f`_y7xVWp!4mF-Ad3Qy)l`-ke(Dd?FP#*CTX z=<$4O6?onQ1_#O6@m%;Ak*|ba4{x!gt`p+YBOgaxkUr-j*G0K^6Ru`#o324-6)z3(g1uG2pZ564UZSrbnJL$DNm^1s>t1&)1b2x9_t- zJ{j#Q#}RR);TBhJC(#d^(mfXxMwEe-7Po>3Y}kf9Ng=CqM?(47w2>Fu8Im50RcXVsKj;Li{0M@1n4;g}WWUv$NZS3VmrX*0Okg$-L`XSS z_~+%F)O&9tuXOwF=3FbW=-~?1EFI>cRI}V$1#O!V-5V#+?MJuzJn9aMB`5Y6Fcla)-mt(K0oOI6uCqJk2agMj6 zwR>K&ID!77N}~L>74XO~8=p1+&|cB-Jcv^eN1vJFtz{LGKJ#7Ob>A-_F4!3wWFvQ8 zG(^eOtacpcp=98bf;wE)@wkL{WS(f30dGLgL1hemR%h6DeoK{38)#bZ@unJ_#Z8bxs zPHP@l!yAzgi;oy0DuLU3S&hskS4IPUO3$VNc6>d*5^JNHrGJ5Qqm$yvfdKK+*3wo1n3e>Q`gLrwwx?Ncp(_eA|?!D+ zq^_lXGo^jF$S_+p*UB*o#LUvvX%Eu<>MWa>uH{dO(=glc%-B+U>^bhAkOHQ;@04k{ zo_Sm}bW0WdE`*t;gOd3>Akr(s{3qbv`;n30{|0pbw*h9Jm4b)} zjVMbH0Y!Yy?T+r-XTcNT^DXh2T3}lc{5uBzso67WEq;GgIX-sB29M+quH-O}RP6O5 zkA-Fog8eVrBi=03Er$j6^dhe!p{q{K53vJ}!x0x;2tdDiUI;gawm<_H=5+x#oxNHO zT|o+>y-iPhn|rAT>MFwU;GoGMt8~I~2F?>XLN7rL`@LvEoj4{ytSAoJ?F+C<_RIS= zd^*l1*(CnSZ2DYn3dc1o@{kUt?hwGtU(se0(w?vIf5Ye|IcUf~xYms|7~t)Ghc@`s zbyJ~Lru}Glc{CKz#WuiLJYl00XA(L*hGfzt%z!hNH(+&3F6;Lf4+^pm3=sYhly;+$ z_WCUEe--!%ZcTc!bhl^yowY3J?zb#OOIh5cZ`8-+G`#Jt;4!{RQOLLK6n6 zyP*3coV)+rpm*s}*%6>q6V{&d7u`Mv>xI=N)xvHDM&1U*hdzfsso(P;#y_YrTxj_{ zgKD^-Wq|Q#Ov9!0$Wv^gg|z5sIAEhqtY#Zd#OtUp!mapJ-jeRLad}2}^alJE-YR`I z6f+D@OuvOiSJ`SYCh>Iu{_V`7wGW_jCRobws{VW^8_z!@tvRbWVV}a4eVjJrYW8FN z$VePPSVT1=WvbcS2c0SfM$Q#-#SZ9~OSWE5GZ!tIaNKB6&<&;mqs$lW-&T;N5(sW3V{e|BhqT! zA<{bMkE>_}2`p8eS{DYA;n%>hZ2i1Y!3zDOy^0q*XAVQ`Y?p+WnqK`ob01^Jn=>B>t_kHZ?YKm`Ei7THuxR04<}P^>E5sK&E)Q#SV(78rFho=%=z6s;~m|3;p<^S$chsM^V$cJ&3z`lRMV|22mTubf`O*vRgViwkdC zmcAl0ZCV2n5rlPL>l?Z&@Va>S?pDoU_v#;~nomT!QMo4aj%1lDveRZHjm!L~+>(Z{ zGu{j>1B66#LyC*}a`QEcjD}*W{|Wt`IRa`EA$eb=9~!u>)EQ|bon7=9t_7BwA=H10 z{?ltkZ2{TAZ+)cm+-02V0wW{0O7@2pmFv7a&Hoc*a{C=T8|gd@Jp2=^k^Ql^Om@Ml z@L7o#b@&H$au)fkFSB0`Q7MqJaDfCCs2!oi!4LP)%6#3`u+857x#UF@rJ*Ubu7`Pf zvObBh9JIJt^(kY8JTH3aP1_W?5iBln{5GVN-^BZ>$Y5>dydD@g=k+sK8P-%w5RRaN za%rv4J}@UfXnlMiIW~Mz7=lkcim+d*sw1oW(k!LypsIIsTb!pzc}*!+C>-GV<_0sb zS}hd!?iYKkiw|7>hG(5c9>ef}9(pCcLKIqA82nyaY-W`reClB54OT~j1OrCi01@A* z^6sum>e9)b3Ol`9dxi`8Fa>9I^>cp5F;U4(%G3Tyehz!yiNLKgmvMQ}{cNwxk$WaC z?%{1(J3MC&{87o7ezOwpJ_VrXA-f7wsf1UnQkBUf-AT?m9i@~CAJpQZ!S7_SL`H>^ z`yt1SdS@N$P8T}G!N4rfMhCIh<;O#Qgorn6yCGBN!ib)btfIEwdyF%+dSZ`GhxTG$ZsX69x zN#9oPCY+8Xq_C)gEy|I* z2vurgl!U0tMFfx7+NTda=R!$ZjUxkFRV;O0{sXnd8wP#}dqVi^p2qZNOCZd+2jar+ z(lIuD*((ZJx-i>y1@;QYWZbdNhfVm*A|;L4%m(?{EkpyB^=`V?j)HVRcII$aJscIB z7%-8gH+diK&30mNxf(a!LDIag6uGb!sRf){@kr8>_E*TmD1C}_e+Jw1clrf7$A7q* za^CEJpBpOw5x1>tDB~)>x4%61D$0h~6fyqEd58B6nr*4o`Wk__qwQJ& zc;h3b>*=JEp{&xz@jVzY+I<@SfK5zTdABy~Me0Fie-V@1-j#696li(XvpsemmtByC zY){(i`nmq+#IRU{<@$V!{iY+ws4PERq6+AZITkwten@mZ5bYu%NnSuiR-oRMLh##Z z>zet`vZ&;euw)xeLA(k|yh{0+dSo=jSz%T+2mTosg^V5(7tFd3lR)Z)SH;W@i*G811dco0imx^10$EltKlE)chpD~Y`pvi*>Rj%a)9S@KPq3it+n=F57W8(^b z({!2LRd)hFdLQ5?^1pE^HlO!4h7y+8>5^jw76nE*RFFpX(}3|Oc{n+9tP?DzuAmGq zPon(6tro`+LymgU&!G~RYAd1#-xQOG0u=*n0+q6^b8hca?CEX|IhBqF4%C}JB}ah# zYMa}s!mBc((BA4>we-KOqeAN2lxM11j8Tax-0Ki6A9681n;%hc6JAqY?mh4cZ6w-* z#VdI`9bgu1tOg8?W%Ep4P~(mh9lvgZA;h2!HcB=#3XbOyg4j8H2%lvcamh3 z@h6f^+z?@({Z7-}nHzd34-%@V{htu-xM|fKPkD1|4IE5A{PEPv?K(*^zp^g z_SoV7FH|M^I~4Oy)7z%c0ES(D&2~<3LdNo_np~QlG9zJJ+LDSR5q~7wyqPcrU&a!k zYH&}~nRCbUj2*Amq-v<(!;Q!_u8j*o*%PP_>-dF+)fDQoa6xluOk3<%hn$gLFHdcj zm%DCm!{F~Gn1^^|r4*|@n2;qk!N7BNsFIa-#dqUHQ0N@-Qxb1h1@9Ms)zyK`@O`o| zR}5yg47pdCMGhenNZ$XE*x&J^MNT?UiOJ!5)y03cd9?0bmokvPN$6ymsom}to2t4& zJWm+Q*0wvkKSt(c+ZRKTgN5l?5dAZ<`d3%$D^%vgX;?BZF=zN;kM>A@E5*lwvhk*B zooqed(a7y14MfecP_M+;4veU=^{*N$q+wf>U_GN$0NT}c8jpSXpoH)9jlnFSPsGrJ z$JvESeaTGMe)Oh2?$aLP^!;*CzV8&FVE^twT=~@VtyBfn!lCwp9_ZW3CB2uJEH4Fl zOpP5*N)L;jma-Gf>TZXPZ%KnrRvh1_Ke&J4c7^TE<$l$DbX-ZO^!!p3B9Ymb^r>$sc^_kj2#xxf$K1=<-@XoG#NRzW z@%ehvFDP57S=A_5X&JVJ{`QNU{&yJV^}oX!(xnZkX3vc&ks(y4w2`KfTTJ0wKMo4i z0WcZTJ#nLSYP@qyzzc3X6Lc|V#Gixhjx818b)Dz!@${{qU432V(tcuY@^}~);h8R1 z_N9KF?j+o2L@#(XJDzd+Pki;~G%dLzc&By%g&n^1B;g~aqo zGP3YQ!o)r^b!_qNg!s?=W-bZXA}mU{WdPq4u2I57U3CEi##n?p!GM0~)e>{jY5h@D zBo|I>fvX_XNla%R3#>_+&05Gywty{xT+Htw&nbj)tKA7lnkH7*Xd2j(miVaVn)mQg zgR>h*!3kfu-3f`!)D%aGxeU18la`0+tI#hf(38pu@OF!T<8f^1DNGSg*Z3)9^b(lR zYq_2xmW}5iy*iPdQmy&g&`e449mB)a6{@QE;u;G>RnFT~rZ4JqOigD_NibdZ&4f)3 z7w!Z$E_f=aqNZGHBg-&&JV_%L?Z{UY%#s@Mf)t z`)^#}Z8lEzaMA(Ptykcnn>yMXXNja99Nsqv!;v>rGywv|<0uec0p%S@rW#cM};tS+ zm_){jjmx_+NTThz%QO)ku)4CEuYGn|e7@a1fChvazbCyuvFd69M$GjJ{Fj_4`3SZ%K2 zYGsL{>~z3f1QpEyM@kv;(OdLwW<1~f7Qe|&uLSX ziftaEp?e<3QG~ObR@9p48*eqCWI9?MQ@(V^I@arq3HiJl__oSN)2nA z5hUI#{}-KrTHtMDU`C*~5v=Q6QT*YX^f;|YJ;OswO^kfU;=fpHKoFn(I{WpqVpC0S zbw$@1@dFOW1Y@AsuKZtgIt{v(W|d_fX$2Igj^5f2CSTyYPB89>{!iki@jnUCqKPxx z_scpslzA@2fJ^ZotbK5CRN+@ltQ? z#zPs<>e?FA4Pj^sDRAsBJtNv(5Omq-h?8yME%!Hwl}~$zMVR;C!;bVB@680=1h)=d z?d2Q%RqWScUZ_S%Ue2%TP@aSN!Eo6PSB+Y@)IhdI35UWb%z0g2R;|`)WMVyHsoS-| zz$8b`ES}?AZS8lD)pS!>;nZ<#(1KF@p=Lui@i%HK;vu*CZM_c_ph&&@Hm}F3lpDR) zqLoP=YqbhgazlG%5V>=G+q?+-Lq21bruofLU7!>bn|pyTTM^UU z$m*mg{mpWv>t#Fv(l7y(S(BD-w?h)Mt=2%Wn@BsEK|HfGZei5n{#N84scv9v(X%f) zZrrqoYsxJi-9G0_Zi>bhjwg;p0f%0|H_%G+M`~ylEpZQ6SKnB*nX&kk7vfPCSJn8Ve41%Rbu+-Ls$+GbT`x9q0qa_I+kq|J?E@ z_#y@ASAFHzEy!_gQ76Wv}tG=77L+d>(LCW2B5Qcr?cm4oF~yWvU7 zulEjFnnR;%0VfX67Ci+pj%2l6WO;^A#0fBpq%XdSE(w6PioVQGT)CL55cIxoAbLx)7wmFJC)c$7q= zv^EtZ2F~lKV2M7{JSGS`GJb7m7t$rIjr!f!cdqALc-tuHJ&`<=LQ)q|9UvHflK2LT z4SGQRRE&W%khd2ZEZS6S+y7(V=5XHLrq??vyWG^jSu?J z%hEaqhb%|Tc-fA!54l6(>jmq))l#ay!{{at-ttQ^)plq=xF0#9>&M&pV7~ABQ>>1| zCFW_&`rYBOr=Sx>ka@Y7&tnMv6y+}o)Q_TfUsqOsKRpm6d$UM|egm}uMuwgDXU`8c zJkO)yUS+7neRHV&wWMPwWWw7H*xms`wE2ii71nU#$qd>xDOwgeY2)}ZKQIG1>SLgz zv3aQ(=u_AwTQf^9z8!t9*f7UuiVpldJbn;~ne?Ku4~>{l3p zl?7H5Kc8V!Q%)M&l92imz*nMJ^GQ)y(`3!IoYkp?R6pW2NGJ3xN&-CoG~FVnUJ+~5 zY;Ger=kZC+ntyVJoG7r_0JplVn!lyf{4&SjZIqfqZPv5zuT1sz^5O7GweK}|q1Z;X zrXkD-zJGCV8}f)9A_&1FO}W}hZfzJ_<<0fFf~@a+Es)<>@)tGg-;LD1|Ate3z|513 zZ+dTJ&2}T*iO%72jlmbEcjIG0GiO1}D(dZm%Z`JLZp@`OhFN4QVUUv!iBbjt`T zLC_O{Av}9Ad%<`s_DBErj3RQJR3_CRaLQjnz61m=1;*BuC(q(@di;`e&R)_JSIG~X zwrBKGQyDTMZj}oa32gF8kc8!9=e?KsXanvu{jf_S-2Et0XknzwDlJ%5m9bABHBN@44Wo5JMOvMk8Fw5P{y|9gmuD^hegD(fub2ulc9Jl zhgG-ly=C5e>o&{M4t1C9LwLD;Rh9i|77d{OqTqHUhnsmtuqXD!5pB&>(oMQ>kJYQ} z*R3DBUVx`fdO9s%+`HFGi`{?E(*pB9@pt)C~lGGZPKA z`Z1E)vAKf9mq{xJVsPFfQlr%#-$MAHqmtv1AE?+t6pEUor;!nS#g|HKi;_@3njFa0 zH7;>pWsKHuMJQ!)kFW#gBfX#9R9QYux)%|h#00;o8K4@K@|ijoZwtTca%T+>BZnVm zEqu^EAQZJScIRR)3SrjhHfv2Y8;$&Mej__zB!^vBxn5~yBpNl1dV_uds;K9HXquUz|J0o-;uB`D4vj5=TS8p*yHcR$_O}6u} zo>6_PvRIT_k-A%uI^gjQ^(kD}dn{di5;2ok#0Be}@brQ|yLxd#Ze3@1wpHrjAXh@b zumVoxZF&5sx_#cJ#tdeLw}$HaM%Wu!;h(e%9xq8_y_i{SHjG;LgX_@Ja6Qi4VtZ zz>Vp#PW;Yl;$DMZ^SM}jZbI31Kk`LJghEeR>2_xmB09ZpzM|&)pT0GOMd^otkLLTb z@;KNR^N!>%&|bj5fPDe|0&3$0!{Bcb_v%CW(ctlm7sei%4EQrtuWQYbWIj;6!ytPd z4~r58TP*`k_0gaDWBJ%@Jgjzo=-(A-h-*Oeu^k6Q4#|-yL2IOqkV9rs{xd5L&94Bh zkoZCdFjkWJe9kUN1(kaII}90xg4o6-GM^lbRy%gw`4Z&QEv?HPaY2-c`NKgKGT%hdFh5xa$&Zia_QIm+Bw{ec$B z8Qgx*B&rWO|&=pmauaGz{_T+#bk2%wSgoBY+O`hXcX6JQ)2JTlLz19yD_ zYK8ZpozEu4R!8qIxy{U1e$L@3OJQgsSN?NUZ#?+yTlywvry~V>m9%+%A{9Qc;u%V|4-LGW3K)F=ILb ztyHTIYtoW0m{6l5v`ID0*}6hFld-jG)H$ko_0nqK`RSX>iZSM@&G_=;$S9whj z!9@?&nfj}@)8?zp=+kfrFHU)ukBzFntPW+S$f97_g>w!Nm?GDOYP+Ioy8;rGFpDQ-jk#Jb7XLWBeU@b4> z2TQyqFwvFPihvQ@`*ZxF!3Q~haWGd5jCOH<(Wz8Rr!#REk-4m7vY2~d2`yQBe=~pg zx@~3gM2w^ZJnnt|c#&%;JjZd-e$LV`&jTRNEb1*jzo<$gF*Y!%+)Dw<26EQfJ`{gl z@_@TtP?c+q|5Zz3*C5D{YFF3sv7@mvpxf#l;}{TFCdjNKBjoSmhXM7eP|(@BHQ%1r zKvS(s~lm5`gtclsx7KWZuE*5}?d|1R(>u6uPz^GTgCc_YEQQCaM9 zc)DOLQ3Udkcu{FyYw0$MLNrx)I>J)>W~q1zwVv-{*mc~jR~ z-Ot{;nVq7BA3~^XeQFKjghRD87nOpM+=(HlzIPYNke9V} z&317mb|}M&ju}I_TYdH_DvBCJEh=s_+l+f@z2P_8k}j8@=Mt*2bojICQBh(!mv<}7 z41BtkAuqTFf2i-*-1Z^rK;Jl`^6TWf-x*~(`b=Oivwp8Xa`xhAyM(Ln>41k{cZ}zf zMMC96gnThR?L*QyGXYdZ(jIdb(7BaWqa|3pg0mE8FVak=7taL3akq-HtA7bG>N4`R z??E^fPgT6nvkxjobs!vb#6K${&8570vyDF3zf&)pOWQxp8P*R*PBJmytZcDaRQ_*} zU+?Y6@3}$CNQDGQeHl#h3$rGG+4xJSu@*U$^OM9QsV;+a4FaDAr%w;;Y^ViJ{VSIvTJ=2Q zW+b`U!gD3R2Gd6ud8+L8_CpTBE@;k%^Of!|;w)P_oeH>Eee;c6%mr~N=`I$tH?wdi zMy|T_5BUbaY^nR}-DUs>LLM3qcc?nnG?mK-(_uG*axXtRzg3E(AgIqPY4t*HniwEH z`YtaSN>nr{we+wjC015@<|QI z^jki(2bpU|x$Zn>9ajJ?WL<2&|6G6e zU4%fU2It`zAzcflO|bU}Z1-BUlMumf9{X*aNV#u|C5M@h5I4ULj#-+{kQ+JhrT*K? zw)n7nDqHR=3+}db@!~I}UhQ_}{-^nBFixk&4??M1Of-`+Fc%)0tPQB|g)SawT-n>? zSk_eQX;@^6$3X)$<~ny2CDAj}bS#XM%@h)ggLB3aOE|4jb{iq`(K>3ZifwP0X&Rj7 zc~KUUCb;l{2%D!OYJD+Dd4H?!Z?aq~t&@%lug*`RXfK@DM_Jm1zRI}`UT*s@dLBeO ziOzu)N~xch?-|G7npOE4VMj>p69)u9k{4tQ4k&{H@E2s`L9D2K|A+9m*Wmy5XK7JY zFEuUx34p#|>6?)Rd!y<75J`V!@|*oj;)z$8*;?v;RNMf2(3o z`$m56csUZJ-c?8Q&)-nrOe~s|z>SB;-C_Z0kN>_SLcd;<`p-5x%gr50z1(;Jza$5w zQ`-NWOv4HKtAMn^1sUpy)VRY~*RS;Fr^)B~QP&Kw;fK!>-N8ANWiCX}3OSqcxUSbf z$~~Mkk?~MY*OSkVno$+lczPXAzqHt^b9E%G!CmtS?MeEeP7^qtH9gsrEBT0gZ^BvB+9Uu?9za z{ZOW7DQb5a+uMi~^YpNdQ9J*J9r2|QqjHgpiwg@c6M1adQ5n=ZURg?z4gGwK(Ei?~d~YCf@tS#+z4TVsK>uK)DlYDcR_lC*1cD__dolNOXtKN);2 zGwm?6rwSHS1{>&CG2hG<-ukF~O(o;j{}Cnbmk`c^0hY*r^hp4$4{ipKU$ooh-lD2; zyO@ZtWiO}>gh(Em{p^tc@~s=!EGa*rvLRrrlMbV5lFa)lPTp_(q46a8M2?p|`~3Sy z+yg(eN05D*q4*6m|4Ba2YeCR?*~pt|Tz|tYq$w(ny%ewr-|~6|xi_mg-g{F+LlCpq zn$TAkZFWK!^;j=J*57ydh{kix*`;KL?9*5HHAxbqLk7j+^GG#tDmL>`TUUY~84D7A z6y0p;Wr;~9Xm*!GFB?Hphnc#;Va=>CK3ze5 z?}FZt?_%B3G-`kwXC;u)5cF}+X-->(ZN_e|3sDpu&BYz?zF8J_PEVo2SFugS8{+dP3J z;Usk%jxrZ(8Ei-BuZ0Ln^z!AbWqto8qe3(q)fwt0sv$BdRcHKPJ=N0{m$uyyt`jZ) ze_NRFS)k6)xvx%V{@%MeNpWfVYR*pzGU=~W{@%=%MkD=C5%zxwK%;TRUN&tGX@{82 z-MCjIR2rEj6&4zqepmjsJKlRZdl1uWVB0BUvgM z4 z#->c^$qSG7S4b-$?Q6emfttvDzU|^W+t^qt7!mo!I4y^O1jXCCW65)$%H@}vTAjbQWmz~#VCP+pqNdv$V~c{J(lUSr}7$B zcAqD5M=*Ag$Z75liD{+F+BEN1`(BD+A$yvsV%?e|-L&WIGIb~XDrVtH7NspZ>yEuXqD83x&j>ki1L3jE<8M<%wW#ew7?_xivb`KKWbZgxdzyc+( z|G+2?)=4!`Es(Snwx)A@ZbI(o@g%d>O^2N9PyZS?iPwsM(M|GtEgv!?tp3{2T_!m@ zZ6keYj;b|NRDrscsgJtuJI70l+*5IfB>BmAR|Isw?=ST^mXg<#*FPn$Tg4@=>vGp) zzRKa|~q`wWMm{X z_uX$RM|{*&^Ms;HZQMn1Q%}VfCnsex$!#TNIw&ardoIs?PZ#TTz`+->yX{3Rju4Gv zk?KzJ(GAlfv$DBTPL8z0Fsh+~5m2$-S#FYn$KT>C8cSw5W~c4!)W^;jVP(yIE~8KzsSwR|SvI=wdzwdYyAn(Lkr2 zjqx^z(_Z$7h9qBpbKO>pv6x1zUNRiLLp#EBHeQ z6k;uD?`8`Y;%N`qVo}nFN9s7u5vwBoS@|7Ci8HO1`7Vsv*+wFuC{(Ix(>V-5z_2&wTkfFFK z49T@?D=6JgC``u_%=q_~>-Q}dN}P%ApKC8mvpLXJf{CUdiffnJ^3J7L)w>M9JnudS z)RiH4p-FF%+M28VCS9NbIj7AFYjUMkX3vuafj_^JkK&SF+bf$Vxvzt`!m(wXOIzP_ zs(73Z&()@_9VsES__B>lD~|=!Z6&0FGsR+;zW&a~hANgRJ>nMBb1yk+W)YS@DFF(j z*Nh3CT&^$la-M*$<9dhemzM!dw?9?kg5cj}5Y(g_eP-l=;;b!x>0jBd9~=B=8@IaF zqbjT^mE|rwg^bMgcymYl>2?|DzY%S|gse;m7^9DqAl9g|cO;0>No=bu1HK^~sk7iKV#Hg;#;4{FO~KGo^ zKAAjkDMaSU^;3C9Uvk^IU>jmZxxsf5dL}ZuMu62Y-9JsZYRi2lt}!sFZI7KI@ZRdR zh*gR}|5Q#b8BHdMMj4I{48WTXWU_8W;)y^U`x5pADB}Tec}jbxA7fozx#7<1b6D>{ z0g)UsrOF7}>JdzcG2mfqopFZHCt`Q=O@wZv6dP zO)<~rxk?smfBg~C!yCmHFW>X=pjMs0=JA&)Kj_wx%k_( z`(-oP&Yeq+;YcVeEEmb_hQxaokc-jVlRtmGi{!c#?8gc@c3>>R4QBL$yAgiH_l=V~ z{IQO-H374rpLHIi&e$SZqN#diysH26W8V6peBbB1i7B0R_rld5?=0SS<-1KS+{SN( zeB|aL0TgSEKa5=&_@d(mv14ARuF5fgB)`tz?k*8R0Ku`KwTlWnnkKikg&{ExF=g{f zUgRGf;oLV@+M$XC(Z#VU=J!eXg3@=BNV)AZbPM{;W@hz_ipROC1oEhtSHqwcwWz}G zQv4*4X5N*kI6e(yxZQB`ptj12LrHJfao0py05X}i#sYB${@N5ez1G};J^TD=CCDH} zse+L@Y3-6v6XKW^V`a|H#Y}Fwfjx?WKvAYF*28}>hw#LJ5PnUC-?OHGVUH7ib$UB4V@Ub6KIQJ_LV*QI6&9j!$ z>#}Q9l`}1ap)jA^mp;Fxr|6lHpzIl8nbS&)R|!unX@`5|vQ6lgB$L?AG;s+%Hixff zt)hFrDR>^5%~Jkev-A0O*jrEVBI1HZhbI0pVn9?a4D&h3R6-ugGpR;`8R}Egngj`y zR?;JcUnen2z5$*yziwF_+0!}Vl|72gk0V|dYV&$Zv{dTqD0un-Vi@cnm&g1jvxX<4 z_~lYeD!Nz@F8vKUY(?})qdD_ILHNOd=y;-tI2r{!dN8~wEP3Fu<{$+`m_xoV;{N!X z1#3iyll>=9=8sPO@3IUv1prwON%=qP2DJaKrcQy$2^=d+FhYcR6054af2DsQ`uMdiYd##LrHio- zq}T_#EHI|3?Un_|3f17h#2&m!9co-~?of`&&f%}3EW&bVo^UrrEXT^;4$o4LR4s|U zQV!#nN;$Txm?U<4W+uh56=+$PK#S*^cp1!3&u>WUY2A5yNh;^vk1Wc__-10uT6|5$ z$l_R^aJ!N48V!ln&?p$qlMmTa$F-Iu?yG4(KL(6VjtOiPW(JLw76Xi7*_XYh6-}kT$22SdTOjm7LLo{T9a47*6;;k4PmHErc|)%r zRU=w=hxMoVcZmn^nlbZl$6WrOj>P|V7=Dn5`m>vT(>Z2uezS3c%tWk8$B!!mmbOcQ z0vQQ|^bJ*bb{n;0>5qJ1b0toatlT4~Dr{rZ*Zhn7Or+M8>QavOy_L(f){X?FUz&5+ zuh-=U^|(cAw$L4kN`$*Ng_wWOyv!=)EVQAMbqG)S<^=sYF zw<#9`*&xwy<3yLoP-%gM+pFqj@UH92Xm780V@-Ye{%tkPJB}i2@os&xZ0U~9Cr?S# zs1I0qq_~-K#b9_{mT1_49A2p(k?_+}6x0p(mWX}0&dk|kXhQ>nw18yzjH=yhbs1bc z{mkpecoUC4GU>*jnP5iGrY6qq!Ss84XMLQ`^^CRM$9iIy9KT{Qr7?sy)zzP0-{$<{ z$Du99MN`;|h)GmOx~2<7mF2Dn1@gF!=OZJW zJNxN8ncLoc>ZmnJ_x+OW20bJfODYiGbX&)6<;}h~f#k@GAiN3e(-Zm!$o>0KVWVvR z7e8h(+-@o&c4;9Uiey7P(&NrrA#JsnlT^ND#MKcCW;S*slyU&amK*(vYh}L>*ZXe^ zbr>|fsnGoyPS;y=$&-A3%Vru$O?tFoA?orJE&=T$EMcpv$?#kX%XO7^a`s7q)~GM1 zejm!BAI!}&N9>4kPK@*7R->aB2}Bh7SV6 zbg4rq;cR^F+bm$iGX=f1hNM}+{_`L-dA5CM^T5>Od{VfYd5E1ivrcR{?yJME@*$e{ zt!hYO7!+=q_{|L|)Xl3VJKmFw$0HhHi8=@aCYh5P($OB8EW+Zh^-kaS%zVAQkd;J< z>|b{s`l36&&^z{jd-A?0^bSq)K&+NFPLU?5<4(mPsuReIH2aCOSpQsAqR#Y=t>B-aKu z4C|r0yb1vV9JnaqRK%!%qk)F{!kSk6WfDXB)$JjIBvWzq88kxU=!+3O@uKv!E*)~N zt)nrm+E=_>%h?KnVXnzW*PxHo#3_!hzuarE9M$ywSD87im8Akdvpp;F94w}d@O>R! z*I8*z-X4p8b-}*VzwV>2UiisB1iV)}tEQng0SwG(ZHZGZrV_w`#$mc;*L9L3CWx}Z z4BwzH90!q#3}s`#ycuC{e3C}_u*_S(p3)f?;2IEsrAIp(uRwY-=9eMG6Y8WunZdo? zGochWIjNXsBwl8$!p-`%cigb-fT)YTV%}R#;L(?`71Nu02w8pB=QgPSc-N+`7s}Eq z6v+LKJb62A5u26JH-B$rGFA`Y!7@@&hjf}aJhDKAP?Vkk?0Oh;?6QQ$(#F?rN>t18-TVEX&*Rp(#Yw!ejC%C&4+}+*X9R_!ImjJ;bxJ!Zu4Fq?0cmIamd-Lvl z-*44gXLjx0Ro(N)fwRtZS9$26wyLZ!t3(Fy2lA3cNeK^xwb_$g*dt>vFdAAjMh#sm zqFDD)f1ZELF&t~H1|B+X!Jr$)reYrJDmHKDz_h9NX+jWX#RS19EIRN~Vb8-*Ft1yS zr#RKR8AY!ITztE=?;Ci9a@lH^BjtOy?%Zp*!;En8c`4gXlZ(}8<0F=I)B5US=WK&2fe4sdw+;`C# zi5|wjHqe{CMO?L$c5*`Dit6F-_}&e+Ev~Xkf|MI-Rk~Lh)hyO$$xxx)?I0tqXGe8y z_XxT9)AoQVUKL(8pQ=R59j|M(-d#Od`WuZIfwV`wz-W`%DRmCrSul~j`Do5W@QoL= z^Pw-oBK>(p_N1Cv!(;3RvVG|pYWMKev9|`yHRP*e)!U{$HuHw+ZkRI!7l-rv?xCov zICD7GZp`vXD|~lJoW*!QI4kQxu@WQXC_gksFReG_sFul}*<_{zM|djhg~<{JI>^a z^TqvN6Z}18<`{278gP>Rua0t90nE4cXkNNhGUAIr)F=PLD!C}@XId~cL<#5GFM+U0UH5N`uL4l z^U1P1@>q0i$%V;V47ld60;;eT4jap3E@J6I+#4-xOyO9-t>iZq?RND63yC*I2M^s} zsAvcdMUh468M?K>SxhND#%)3QRz#7&7dmfN6$_kZYzY^7+zIuDVr6gF8Qt>co)%Wi zj=J*%d+cBRLhc&rfE+xs0yr#+YGF8PCs5T`O`=nb1(FQ>xXT+WPZHLQK$~Ql5{HC` z3@B7?e3mSPZoZ)eH*I6cRjiz8^2rw;w3t3yfsG%f+#pPu#eJ zG_%AN&k3wpMiidDio=1O)?KNtPej`0{bSD|+s_}ic>&wcD?c8~*qur`+Q(l!J%ai5 zzD!ixZbHEM{IO>Q_-8Ai_uTQ}L1lLA%LKoBewSjh5ti|@3B)|QC;1t?5XK~h!J{S3nqM#yq={5a{m}R&6U>shxZR& z)K}$n-+K5HuM|5#LlbM2K7r~b-YE%yQYUgMvjbkH`5NB2WvH>Db1P!^52bPNyLlRV zPhuLG#keCWqNBv=DfwehR4fHQ0e4^j!be0OavdQh_Vu91F0A`&vp{yDVoF@Vfbskb~8J-fElh%q)$^DzB_X`ZlI9fGcK^Pu{srV8CqWC=qi_MBIz5U`WJ2& zP*=98oQpBFH^=&jhr>ff83#?h*NoaNQcxy--m4vCe2Z>xaqLMaN+Lgvk$B&qHPR~fe16$P9=`>X z&vS1PM&^y-LgX@P@eH= zCf!o)CXZ_YONqHbdeppSs}Wt6Z+4ab%FN5c(zK^ryS&vR^Mref6FO&XzR1%?dY2)) z#~E>o^MAk?*CeOf8|UP=pH+5fd}7a4+_Ir{C-cPxV3{;CUv!WqeaKMRQ;#^s^*;a@ z$~F7OIr&*+mz5Kr+G`ds>(W+&JLzi{vsq;8l=Y4LA7Q@}RS;Dw%}Uy?d0U60f-7qr@4T7Y)rKILg}wsLHnQaN5>Kw_K3bit6u27zjW^7{hId6Kymj{0?!Iv z`hcOdr%=`=k?YL&>?qt2Rhm$O#Qty@q-^d4K|j2TK6?Djp3y!ffxV_-LLF+Bp;2?Q zyW+#DCR8E1b5nb;&Kb`-tZ?Mo+);d<%F{YmXaXDj)}wSj?LZtdWp1?yNB+ zaR3-)(8N1jnz&h5Vd-r*X0yYJPN9i(QOWql+lwffF*ebd%C8o=eY@F#XWr6Yicz~y& z@VF_x*x9hDS;w~dOV8f-%Q+6rr&k0*(*yvBZ{KlL$K=L0k^I0o*aoWC7 z8evhjGvXJkRe}e5>X<7mDDx2e4jgx>6X7PbSLp@P9I41Ku|ck5Xl6#t#JJ$s>`w-S zo%!q1>kqZ_wdLA~jNs;zA?mNLu?F7xN;McbQrXN(io?>dFK*ss;R#a9r)@UAz#Ur& z)-imC4f(&I9!EzD5jiNnM;E4{8OR24qU?zL7ci#fjHiC%IPbT|D2S<|X?NEr{f?^BZy9DDzG7`XW=lNqO-1xxAv3#T=n{NzgrDaf3*RId$>rgVzA6~N1{kZv3YQ_z2 zGTMGIC^+tgqkdsYo79&yx}=I8A7G3^ZiUZ`d-TfnX8vXe@Hn|5CP3_SJMOGP+IsEd zmbhHudu^I^?E=hAFoJMVQi$tkv5^=MV#c$WJ#*vSq@Y3c+4dkwpl<06{b&)6&0SP< zuUbumo=0^}(LqIgnsYb9^~u_PM^m(!AuJxnk-6Ny?L*(c=IX%P6d5w!(|)9kZ6;3d zl)4Regi1FFs=?HKG}!*0^Ei8b<}(AhxD4sLsqSdUXhlHbhnc}%H`kaPD~%LHCa)^f zwXA4;j#;%ttqBueYBz5JS1b!O=R>$wQ@xq~^!ell2M5a+uW4<#_C=uxO&i|Y20{H% z8Da)@G9oZE|L|@RePRZbG#F`!m1r5Vux^pHQ6l?kZW(0Ykd~yU~L`5qs@(P>f&+1;E~27jZco0 z7&_FmwZ5PRO#y`L{XNO4?^faQ`$ejM^cdo7tv{SIP)GdM82k+pKSgV=4HCub z4iT`=P|rd%2_%y$<(OysZzTz$8i-m2Cp4iqnFl*epny9dn>B*&e~$n?{DC5P_LiH+X2WE)CS z8cl2G(0DTd3j_}fGP}rLFxoq0F+y_XmPPx+40To%30(1{l=7YH)nly>G$gV=$W0P! zCFTo{%BU9Spy}d|)s*{BoF{LQ!`DgiV87Wf&%6$9d!GDCjScZb-_Ie?;(56yD3X!? zb)OW7XMNQ49{SeDWLJH*BP(6D4smDwicXy0c$MxGz{i5AVr%@V{DTdL5W5XP=b1O{;C$j*e%;q-0b{&OL@`qR(vOkt%6x~AHl}g=0@GB zG}g{Q0K@ve?;k{6)@#}2Cm2FujNrFHcMEeqHP!`?^fB~ZdS$DKb+uR+Z?;!ochZOV zH1w~hJTZv3Cxtdizi%kJ?>MemBallB)ws|6&rSSxgSt0KI=|hmnH%e6!@p6(h5`EF z-?#Z+x3DyR6~ITu=1r}{qwQkyhMHA*MjjTn{t^78Z=es70p9I)5uxRDAsu2oOI$j~ z<6&A`8xM|*@P>(zLw#F@j*Km7+R}977_(iDtU%aKL|4`Bd5!OIUbZNr^R-hox+P7e z;=h!MkyZdb+YDgT%6kko9Y@dD=H?E|`Tgac4Yk_H?{fDs@5i%d(Nu`AgBdxSV^4?- zw!&O>bn53@2?uhk{6z7!dXsk@1=&RRTD}H?YF2iWw_)lR+8>uv_!HFe`7NbX6GkH! zUeyBR&Re9J4wVvfhsL+eE|~8DtivwN@^7S>?8SukAiwCvyvJtBrQXTy~^vkRJK`Lfo$gk2=E}lr^%#g6wy35aYEv z^1U(??jo1TyD%git8JG_5*3|>3t>0I?8R|Md(31pGYBn_-~#S-Y|O4%eh-%*+!O={ zISydR!jh8(hu!j-b3`75a|7-SFnZ5?T_h9RPQ+sDeJZM;gh}!pE2HL}0#1L-8rX=a zmboI0L)q7onnA*5qIWFs?h2q=0#W=oCBVjQv6Iu`LPQ!9GUk2>e1;b?zE-aAuH-q9 zHRYIjBwcJZwSM)tKO^MfohfveXf| z&ke@W@G-T6+NhsDll-LaiOnImS#F8q1noBbvDatgg>LQT!*8lujMaXF&jC>FkUQP? zdyMaols)Qk7AWR(=)&(C1CBj+^!uR8{Ye6hdP}-Wb28f?#0`wn<6o+7O?d>}@b&7; zIJGf+Ju%|9>Op%9f9~})uF0KI+}UA1Y5jJ)@Eh)ske|j?m$#lP8qDOLR{h2dKePjwM#bD*f=g_B zrD12W>v|y`w$sg$eFC@yv-4!-DN5{ce{I79%g~Y3;$h#3kM(t5T@sSr%~E@=i9H(; z!p*#vr;Zyjvwf{+5wDF*VdKClL<={YKe6^1u)Q>ezPx<*I{&=)U80kz-a!Y$9*(Lr zY??%E(;GkOtB%oZhMx= zp?n%^qo)AWC}zpWR6;pW{w_dD9IM+-cRKwE|xVj|s5>3*baNkZdfY#@^XxMxg^OoHIfRx zgV_FxNf_A&kwlCUHpTBWJ{oXvXdZ&zo>Ua+=KJ-~W!jqOx(~JtsGk491eIztifMka zR{NY6_04mzWM%svF#AjFgbtw3Gsbkc<#55_k;Nm6PZ66qFtiChw5bN9?(i?~`VV(! zV+S(J`H1W;VDH(&vXT!Zwn!aOIm5F=WQi%_69%;&z?`7A~(h&xSv&DtB#V9%Sx1%((tF z`=7<|OXJ%m0GIPPyqQT4r^kJRX$3wrA6q7uJbnezt+MpPHYzsB)p`zaH?l? z`HAhSAHv$7)6KT*OTtA)RBF8W^xdGU&cSnqoxSa)IIZQ;36mgJ9s&WP9U7zcy4N)2W3$3WmN(q2x-BvE0Qcu2>q; zb-k>t*KNV7BBHVz;NRt03R>t~$l5dqpC%lVoNyG+pL6DI_Y78}>+1^VC@XdnggW*o z->HX%O7Ik}Hk3a5D{1?Y3fgik<)}@S zSkDB|(aHVH1U2QG91*_fZ5aqsq1nq(i%3-@zutQC7gQNp6EL(SX_^-kks7LZp~&EC z(@o$rnlt`wOOkUmoXvrVVUbxL*&nF9C@se1v~&o-vtL1cE(JicV$k= z+j}}YIL3i2qXJ>_7fT_ZXH=8AoEnAs)Hr= zQ{?Cvqwf98{LS#>C+A-$O7>R1JDE1=RpVav-$}EToEfnXhHY1BJ3&V*3LkTO65a0u zpE|248JRFrA!6D!P(nEvhuXUbXL-lA@euHa(01=HE*)jcZ&-qtlBD+HN)pA^W=h!| z#lQg;DUx8*(rNB7-mA_efZ!{_yE1JWlUg4nR}8e3U9|lS-gq{gxZDG%Et%)r;FtEK z+}+V4q@t$4d6xxW>svGg>P`dQO^lIrtuX-Wmi_H08;9+WeD>hS%G#!0c|9Qy7CJQWS+NCiw42Icq6O3MtG@UvB zBQW+XqqosNvcG^q<$Q3)iOIkbSRFjh4?5+J`Q?+YHv1aO{bzVn4B(0D?&HfuTuWlq z!_MV#I_;_qon~WK1BzPsCAdU+OA=6`L}yD@s%Jxuc-DkB@31t1Z!WDJWeRW$RUAKC z)5WqqQM-`1Tlnd79Ut;tPWDFiz;pZxN8^J!UWs{HF&A_i1j(2VVJj3g6ctYp%Mq8< z3*n1)x%}DIbX2U6;>kc`5wfz>`q~G01Uq`nkuf#VB4`CoNpYC%KakD<$#*4={t?3A zNKvy`rwPhjE_p1m`h!LK-uGcX#(3M0#zPqiHbbR8S z{>0It<2Ffr1GxP$lSrZZem8Y)h|hXw# z`OB+tCXLA5-gGU=!(V3<-X0oe2Qm#qU)RqA^dI1#JqoYQNDus#>k>?BF??PB0&IY~ zlhb|#1KO2cuv-QcB@38Myy| zuWj@GF;uuV?JOWVDZ zg*pz*VBh)IbI;`ND4y)VUe=2Zmpg43MFs7ncjLAiV>a425VR$`a)}=3BX!4- zf;l?efuJP^KS+58;7H-=p0~&WZ>GTb>O)*M4F$o98RLQuuU?a;pJ}gW)w~-67UcDv zcRj~mu9;#9?F9DL3;b3MNIm=a_)NFK9APcxR;t6EMU9^vM&B+W`MJxrQ>>t!hTryj zZSbD^+MT^*04aX^_G(Wgzvu5>o$*~1%HzZ@g`SGMrGOk|xiX6B!~==zN1ln2-v38bnV{;tBI!TocT)slM1MNUXk~S1dj3B zmsPiF^OgFlmE?Td>!MmtJo8RF^FF6^N2F-2G6BGP4YAm-lh*l^m}Xy*s6yA5Q)C8M zqm*J?-zcik*JW4YO`id)Jg_=9fQlxUeHoA~9bn8u&iS^d!aE$_k(z>ALOT&`$@k#{ zH2)G#x^6J=7~$PVFR8loq|XMFHy%-}T-P&WNPXyrhCz4VFN*`H;6dV}HKPt!xHjL* zByIu*P!sF@<9a(-P-hXoC_>U2ec?^FdsLJT<6W+^EFFwSpiP4vbk)Zp`3jfaTYInt zXO5sM#*SD}mHn|Gf4MUpzc~-zunB!Cg%vjoM$D|J!--nf(Dsp0k3Q?PicPC+fnW$? zQ?VtTmO1PD``Uc%MJO$o%2FTEk0qN>TB~dTdSj!?x)k6s+EV_-2ee%JZ-IJItgp)I zb(cnsDLL!t;bz*a)vxvu`UV@RhvJ`2vWVfN$x92JoWn^;((WmP_ToUIrN|wHHG}{! z4n7MUyu7RqE&ETCQs3!f?lXKEKGN6el=~Ag{Q|*qEv;vb_a786CS*d!LPWu;3VSC4 zfFkbs5_x+<&D71O`f;-y>XZB2{P&h1=2k_qf35APn&1=4d~Fk1sCYP6bCvfG zX|p0`yp%eoR&Fj#WLQb@KoEFniHbRl$J**gTfN!LUj{l=9QoUi;>!GK7NxQFY`_vcdq0$BIZ@=mAf4>e7g?goiAx zB0Y_Hp2YNS7NhYOcAto}BKXK5;p>;1!_n-jsV3DNOgnG(T)sbV8*-JEu6&u)hzkX_4O$iMz9x8SKflzuPOUnJ)uJ3 z<(_sSiHXWae=Wp2)%^har+ZKrP#3}H*1QKZ<5KvNzwKZOQtu&+8aVU~8~AHutKZ+2 zC)m0MPf+}sE{{$m;Ss4 zoZ$@bt?zc41@m3mZpU5LU?6!{#3>+uatOfRxG!-it?Xc1slO6(Gdy1FazHQa^mcA7 z(Wx7j)SSJas8Op)e_~E-pb}Q_wpTsjYcn|?vL3QeI~*8GqR+;#;{!O_8lsHT@p;o3 zGEgSt?sYcbC3!NE3e#f|_C&6VSc=1UP2j$;q zf*9kdz=C-cW8?nvpypyx2JVh*bJtcTDlKJaLC%20ezoLBl| zYN|J8$~OSz8?CCNLq4e3A2~ldQ<#rSj!OR+HT9#@N3V~Mv}!jl1ZgwGCb3lIJC8rN zYo-x6prTQ8|5#U9uiZhqcG6wa9d+WJOK#04VB-M2(;?h!YQymNbXBm#$66&t#GFx-zSDdKyC#MF(=8$mvN-mRz`Dd@%R>$v zZ7kB=t6%fOPBZYF#Inj~G(&@gYJh4?qm8)%A}p~r z8Ne!<)yFNLJ3SXP%p`d$CwbaWN#jxK4JA13H#zvjraK|(sDux>kQ4(hw>B43?N+*6 zHr~1M06q?n@fkD8?+vw-Ag{Ds(CM*96F$L{eypzroW7r%rNsNm;jm&HcBZ|CA%(TE z*yix_y+OMMPSI zrR+v#c9|Ae&DE_M!8OOi6OmKQ$GJ_y3-H~K5g`koDMOO9@gO9yx2(ywE59MZM6K8p($>K-hg*`Rqwur7iE;MVN(J_88>8(WQTi@k-Q}D>5Du zLwY8}XIoH==-YvKyUkaHgZQfqxBkP@$?TlH1>T>h*HN=oIkyYEdhY#)qD8I+BPnd$ zsMNgU*YyV;RmHp)hDshaGJt;Btxp%k9`Wtsx}@|880_C7&3+RO2yMGq+kI^F!SNVFk$?0{5cD~TZ$uv}W{(XTz3Ovh9a?ACoVu=c;w|o!bTUlXl?iKCnWg$MkqXPSfLM3FKlsJc=M4l&LP~c82t%6XU2}qIwam zS6<^@q@*kWed*fDZOF7TM```O>W% z{P6trNb!>t{whKJi3B|RCJmt-L_ETLFkLxeePW~7E3kXIzeSky7C0kgOGvLdut0B zJi&r5u;{+;&dmQkB0yVVVD?jbisI*@@dK16hO?(|YvK`v1O;(a+aYybcB@yC5j`&V z0(D!DHaLN%-G+DZ(eb^Zt~AJGHO?wkb&gSpb?0QZ{#MD*oHSoR(}(ZPLbCWW&@vJVqTvM-y&4cdRf?vL|!-)buKQm4=lgLE~%Q`EqX}f}Rh4MbSe9P$;wJ%dx zY0AX%|8N4Tya?hh`ecXuGPRR(C-+40g7zBz+6&m=KhM2e_0}^2f-c{u>ct<;*LSh^ zRd=u;Cm_H#u+V<}eM_Kgh#+nt&EVa5K++!Cj~iqO)Vx!P^^KE%MHE=x`E|SsT5=%a z^BvEx{}^NpwdfKPnI1IdVV`YZLn;sC(u z@*r(5-jWF%bAvSC3^oCvL6o#?;028Qy+B46!1D}vZb6!%yLmxQK<$zIyg-(~*YH7Z zK{5bBz-u$K{42fy@aT^{QXZZcNd3lZ^(61ma@)|dnUsTDf~W2T`)(?wtCWIcBElrA z9F!LL!;|@Sow+xHzqbH#4nPG7p)S;^4}XDLruP(|w`8~@+`E2py0ckW>4SYH(au!G zyOu(3BtBugEdB%nezp-f>pi>|c|N9*tFL=mu)7|{KN21HN0@26XP2tg-K&@HQ2Dze z44ub4IB`O6b1K03vAE@QyFBk-EIJeQZ%|I~?l+${LelL698ObwlK^2hRhf=8#9t6l zC9NX&8m}uv%$%4hnyAH8aAdix%JT22IKMdHI)&6eRZMXvRWC#9al>ou2w@CTj3VCq z>s$_j30f=mTHg~&Wdxfzdj{eSTkWmDQEgRxCb-8!DeQAy`*UiBD4-rra13}%Ca&^R zT&qrEW;9_Mpxp8uaiSEAMI{L%Q4q2a%wLzvfH)^0Hnt)vwB4Ku5f*Ry7~~8~M9cPx z>_4JH8RSW2D~UH9B(h?Y8!3_TFj1jOXViZrLG^G#TtIiQQAqsw#w94uCNz!>s|V$u zd4d4QlOJLBQGvb|Jdw+z8SAs1kf+Kkd0`76jgCtYN-Qb*i(b&v3XL9lj>KjU%#dmgmja?pT+4kYF8dcA8s ztDmNzJ!8=AfUU70UD&obZxN{XkQmZ(1IO>Z@2FUJ-6#eBnkY088W%0NPM@XQR=U}O ze0>08^W+=6-d^6_bwRcw!=YIVV@zgX%@Y{cqDpr?N-j69AJsrQGZR7ENJaE*PXJ(y z{h|c%fzwv*!D05kI;o}^kv@eoxP+6W7ZHR zcLp?0KgKga!h{C^Z|z&~gTou~J4ukp zq@pklFiW8?u|}W)dU%kCkP{uy2Y&v}pm`AU2?#3u#Sh%Z-)1+j`9wTVM9j&p9U)Hl ziBy>7s!x~U+s2+123Z_ohPT}>md&+vuqIkHquyTFr-aX5kFm1^2|hmgbks%wO>VEs zb`diHu|;$K5)X5#`)&0o&GSSNBtl(_(@Fg5l=^l&u%Amg=ntd#JlirUn;B)= zFhwll@(+ilU?FAq8Ul^Sg%%3v&-?x^L67;pbu zObO&}95Ya1=~_j&7?@9|XAH}D*#!owC)lK;eGZiE_OY{&v(l|35|H%gq21%3ZEF!y zGXx8tPiO8P$gksMeS4~HN z7NB8dH~(GTDi~PVN`t>T>#c6pR`O`3{V{g+#Mj6Woqv@SR)$P<`dweR)}y*A8$2~) z2_R&hUye}|JSSvL)QGI*imR1zF!>3HS>Uw}*sc~jZnZwX7)XT@yf3)N_`5JMvbHOp zHrMa8CYPOhPkEU9l%lZ=e+m~PYisV`e*#u8W`Wl}XoMzduE)^MBMr1Jne1bpySVDh z^4~Ta&r}}`0l|a^OeZP14b3~O&Ocr*An_bxMWnZg;!PXJPUb>3@+?WJ%2`ulVilYe zc}#%BwQCJ@xNPQ^@^9o?j&g>^PbN~v&bfh$_~R7L6kWxu<|iR`xPOt8rvFv1+!q|P zdD+pAl}m;uXTv?DX|zIp5t6pjS8`539fmg(;`D|Ud)h)DU%Yd2-k zlr@9x1Rmhw0I>EOkm<}imOz6WHr`&A8JVD5nHOEC5twyeUm=JJ=T3G(lC-gz zrXpvgDzYkZKgC}kEh+sqUup zTdVOR)*Ii(D84(H7kb>Qre|N9bT{nS90hwe+Zj+AJW8#rxK}tV4Dw{@XAX6p$4PfM zdP*?xV41A2L3`FKwYZW{ncYhiuEp1=&alAMh^-KtAvuh{9`rQs;LB-(!kp_&52>0FZ#djr+5}Fb|4NIk zQJsGhX$%Z{u25^rH(b;>ZQl^%7{7;=P0;mm1zg48Xz@Tc`$-@L-2)CrCiUWnSq|%cNgc9L z`)z4fUW5ETEGYbOe4o`~i4(o7xrPJ_G4^Qw&!39sNrm(BGj7N&26{X2;1Or86wir6 zdHk?vi?#%`Cj_h^k=`A|sL6m>9%|t$&`Rf4PlM+hafr$pXo}PL1IDeK3Cvb%fwUZM ze#2zKO|Db29*cq(0c_U`?=P4)Z0)G=@;`*-bF191w5snm*suxi;~`?A%1HWzjC_of zc?#DmfGvVYl;{ig>Z0Z*dKOp*-aWl61{@IqocypLR;zap0lm8!>L#01WUgY4(GoRzehGYciUOHUN6s!?oeaL(rXJ_n;zNNE znPF#2H#e^E9pRd2YLEr6XwWqLO?q{%<8ol$y>olaCW(mtGa;-RD7fs;nCfm@YeWM9 zw~4S&LGyKiM{aiGZ9?&rz-sNgtx&J%9%+8Q+Q%68M7<$(+1g(p6WI%zB)$2;x#F-_ zJ{O7Jk9;E3MQskTjc86gU%yp7mPBUvfmnyTLi&7m020WUW2CJO@Tj7=2a|35hF&N9 zWQptR5_=^cJk_(|9NhC%5gbfGvjw@Urm;nzHLLy&UU9p}(3PIEFyURr;wz-Uws$i7M)t1RZ7Fa7#OfKP79=LN~Aln_i0lBoag17y|YMh0vM)QCt}zS!xA zECk_$0UhD z>4z~2V*HK?FsRtpjR2xKF8vBy?!HQQy9(lHmy9T*QW7JQ4e3hT8rvIF5B76uMjX+M zJ*A`%MoQi|eZpqd%_lj55{uN!rl+MAnbX>i<@`S}$33E@AwixPdw#t{+~b9pUsBwT zWxxz?QM_dDad$^42Y@j6*E*J0L)C`e((Oo)!CC@DBbZ|M>4LgBJMb~q3;=Av4Yb*i zeVauB3TpZQ!owzlx#Y+sz-(^i00Qa;@SV?1lR|tu@SP8ayxel&WkGR$j@uZFoU=aD)gCcXWWGtT_7n{aYj{7aWbUr=og(>9c31LjalQU>r zs=l7>o;3db^R!qn6Vd91Msrz*bj+SphjR86h={}KJQdOoR9$0QPN4#3tD1-Wd()d+ zVbLb2i$6lKzqGG1a?08Dw)8k+y?!J}~ z#u^oOr^^USI9~0LfUUXckXtrPW!Is48KVU=p|3iRr{d!f?pxY^tUvBWTNJA5R(YWP zDq5Y`(u`GVKWZdbNUD;RC#bl^sEiMm8BgOGOt(gJY5gdJUXCy$L)K`5)6g<|FKJ9x zi>eV`A)-o59q+OmcbzpTX}pYLT)Fro35aZnY8%I-lLsXM|Iq8W`Fx#-n%?s0sWbE>c z6ZauqJkEx0aBRH!RE1|qG+;1GClRAMSZsT4;7EW}yhOZ*DFw+k7RycA+w#zaN@ja7 zXOi?t$_B1BKN^jnu1jDlNJ8xMoO5(WVm>|%#8igO85SS$4I8=_H{^vc^NnrVJM0tD z_WhqkLTLGb*}w^BHgX^epAa!8WG%NSBO3HZv0Emze}a+G`Rhq1d9TbGOUs05{ZTEB9n(r zNjn3r;S=2c6IVrl%?bEtQaqUDxAlZi;50P94Y`5mTcpbTUdG=m^Q1;@mHKO42kkOJ zw&+5kX=xpDpM5SIGcmgDs~(czdycTjZ}Iak7UyL8H7<^VFWHpI9tKwNNX-fT_!%jP z=Y&%EQQqYu6nki(eE67X@}&~+_A&r#a`iUQE=JQw*u(UQzT;}<@*;rx<6gd_tO!?X zeZ%|qI<)+WOZ?~7@L${kzdGit#T1fuSRVL&d)SFT4yTiVh+(;eY7q(=kOSXIjq~b7 zjEHGO_4oS>e<=ZZKI7UiD1XMey|D~be$i6KK4Nm)*&l^@<>AVv#D@d8N@k5J@vxkD zzsxrQPac*9H&m@_usLYduS(&cfR-@KhRWTbW>!hRlL2$f^Mct~&#AjDz*5^4W~^$I zhCd&JYO9zxksxj56Fnge?O65goOnccqQm)s+Ti;>jM#07`V0#FDmP4rd5r(wA&vv1 zG;VgK#w9wmpgpY%!3=J=PVoVBfqUo$G-{mo^qYftDSC12aKcauq*EnLb=R=0A0RAG z?gt^aVwgFvmpm}W00%-GG?-+pIEhpiqlsF z$`Wogns{zp9PJA3F0FCNHb|Bc#d*;%AFg)QpIdu~I>axubL@~M8NG}pk9@=3QavTl znDAS7&pW=?0-H?mci~GtV&ChJ+L|0T>@ieOK_}okRQDO+07ibqw&YHHeGYmxP+qN5 zniu8hcZ24wE6(eB$sKbGFZIGfVRcd~;2AzVIh1_eYg0R~?(|xsU9$zP9BR?(UIsfs zw{P-mgZxdo!E1JD+4uiN>ObEc(Js((@K)4o5E6YK1+zt;@X5}ET%uVs1!aeC*60RO z0CfAhyyHyB-M`tHQ-RgbqXCz z086B68lb=nIVd}9vqZNHm@TS(qF*KicfG{K_Fv@JmrGQ0>*n}nFr4VmDZD+Sa zx(#TcTe!6dOw1AUZf>HYtU8HRQp+UiU?Lu~2kq7Mnk@@jc{>~h)GH%>Qq7#(Nvye3h zsVl1S=6FS;FpngUZho3?K5KEf)gdUg&}rMF6BaD*)ttf0@8FbzN-(}=aOcS&PTyk) z4b&sn@2qY!WD-o@UL(^WZlf)3+1C#NNYi%2JJJ#q zc_$Bl%ofa!B+z?8W-|y3#R1DW|E$?{y8UsPi_Y(M4r6Bd@b3SxYd`9*z7uJsnT&li%}X-{-HFmgD9&Xb4g zpncuX-1KhDwU=>43J&V=*ZC<0n9H-l48}3 zk!QBZdh@~{>DvOIH--y54cmXFU5dQ75S)McshxwrI2nxhH(Mb!H=jaD~wSHA>TAx-l_HdFVcBq)V{PYFD``<*chUs$){{l zEX;t=C8nwN`^%mYe^uY}Ity1_8r)Oa*|LlRLv#Qt=V5z<&(EAkSvtpvk`cX|#w#X^ zGpnHrLxghX;E#@x2lkclGsc2)sZp8;+^*Z@I_KaagUia>w(MXGMz{1eRyo*#yZ#F& zl1W@>vuZymcY14_h@-&b(BkP42EhVJzbOOaWHkw+TVe{14zL+__3lx%RrNT@jnSZ$ zs4%+Nb%A@^`~A_S7$c2(q(57OL(ktWWGG`u`c_>^ged#vAt*V+Yo6Ps)DUCANql(t z2%u>ak-BEBThWu1GD!^CEY(9rD2pP-M$icq;@i}>QHMU3)A0nj9)GcnBQzROY zGRYcM-})()Nl1LHQEhRJ-$m%11#t-ORd2c~kLCI2$y7*OM}^m16 z-UfV3+4AS*z#vF}gH2oczcm8-gbcqx-pR1K6Eh%VT0(`i+GJ`Xy%YQ3zaG{-0b)n- zJJXvA8S?-n1QD3_Yh<^48!Xxz0tJx&GYp;;gZWtjqju9S-Y0qQ9|Dm)v*? z?5S-Z-YK11JI>mI@PCP9{}!zxywyqVU$&zJ7E%4n*lF7{E#JS%J9O7|n(RgzVNW9n zSobdq$Pobc@%d+8{`$AtBu%NH<9L(9$I!4bsg=cf){mNT-xEC?FjILx^;D45@&e!R`K^efIgSo3+-v zp7qXL&Bc84jQcbBP=#!XV>(%tTg!*;GrGtUyx@2s29A6)+!r@oP&U5Rg*v~9Rc+;D zN+#0)kAX%dcwB}}o-ENP_`-*ZRSz$7-W04~#-Ax9lz;2`JF<{M{iI}NnHmkPuehy( zz4*w)KRf+I1J?@g_iSf9VoFO8*=-Nab*i5j8-x8)K{@WAo_ZQhsM`zqof*lOPqON} z&3H%wiAE9@)F(7lkmM=sa~*?+p1lec;^rCdt=RgmsDRLkaW$pJZz2cJDbl*w39j-I z3kr$xw+!cz?s~ih(=>kJbItb&f(JW4JA{)&4|r2d*BcwJhv%H1oBt%Q$4BU8ePf8! zW?L`0_ojHe)mT%V1?>mEI~YlGNjDJvHNjjMup}T3mc@pko2?p9uaby6ft|%rbVOH7&-s^DwqoL{u1w$b$%%^`8ax0%3B#>c?o! z$Isp}L;{Q$-@?QHN$@tj(o}IAM6d|j{`hJMMWDS0A*a_iHdeY_hh}7^la_!9kX9zB>7K?z*FJ``YbfD*!p+#5fx;LF0PDm zJomAHMpcrC^BPfiyqI9nNd;>OILSo6xUjpff{04;1$*u_L3=<%1@{nwxxdwnFMF-H zPhxM-IlNSSO@~xmn`^G$-RUuNA!Fo6p?v4KdC*(dowz5M_QQFG_YPhg1L{oTC&J>g zNhA>o$<8_HACeA2zQq_0@#C22d!`TPua`H^iRE*eaC?DUNN7e{hCNHlOLn)kcp$TmL3NXY@WoMq!F!|z z8&jq-KCWV z(5a9U5#-^E;50t#Gp)uFyh5J`p_lzgEIT83_pSVjggV1${nd0;Xe44E{5qU-punr= zW#muBqY(Roj{1VyM5FWr5eZdE4^Fuy$6WDguyCIIP3Aj3x32ht3!C`Jh6Qm@l73Ew zRrLiF2`ixbS1M@1IUkUMDSLV>$noER>VE~49?uzJ1SD|c*ZMAlSouczStgH5>=G8mS@WV9r89N%cA!cX8S~(Mv*Y_!ATbo)yK; zQF{oK4E7sF%k$e^Nt7q$DEt#08q;Ge-LK9r`qPc=A5veNl=0P}LDTuvYt4^Ln-b1z z^>``RNPoWgh`j%t0??S@2hC7M$?cWkO1~b`D>k@H+&iYWu_UQw zoI4x5`Pc(e?LU7lYYhK~|Rk2m*YHtmJ=my+}30 zF+@gx&6kzTW%zMap0$V&c4YaXaf<$rWX%V+8MVo?HeA2<(OxmIsTX~>+xhNFOreQj z369wiw9oexVvd`jUw7)4kZVRqil-7~-KK;aC;uS;6-)jXjWkZ39!o$YGfz0?AvZqNU?{`I(6SJkg3ekz1p2?G-(k4~+?u?%lxNhGiucdRi4W<1zZ#bDYgBRX*iK}MtSMO_ZDRdY&5B*2rdW=t zhy)7iE3WXJsMe|7p}|*lJY5V5>;oQZAC9QjQ-wMnKXF^qa1@MN@NyDZzOE~cnvG@y zMsf<&6ftKtvW+}8c=8^ULO;-M6$aG~@s zoN^Y#ANdMelS#9I)(^;U{z^Yvzcx|z4b8u3R^8yA?TsSMBL$jByB!ngZD9c)y8jLE z%WZg;ot2+?N^Hy>D@(_z#;F2U)Q_n@xZAeAW9Jp){SC6dYrFR=JM-LvG%1P+mHX`LdH_+|%jqUODNjqX9U>JTym0M z00cS@{yf1TU{a2P?*{!0KMi7iT*b0=VB%G`*MCwgw*8@2-3V%W-?mgCoPjjG7^`{w z;2eCBm&qq{Wn(V@N|+ICOs0VZLeh)l2y6t|22g)Jgvi1wWnS3*SwK$@W1UcVCVq z>fND>`&?z#`Q3tamm4n zbw`RtI#6zAi}aNUOD0J#YY|xq3xb0>< zKD*V9v$0EaHty{&6y$V4Rtw#XrI2=+5~7{Tr1}9j~)&)ge+RAyT;Mm4}~?h z^c}g=O$CUciL^$U!5zJaP*CHu)qWOa)_xzM3x~-RLLapQPf8qTGql={vUxC?zXtZi z_r%T_YO3V#n5!EIw<>~*oLx=`@i%v7Ir3nqXI`*cX*>Id6D|Ha9g<`ST5{ad_nuSy zR1KH}#fOba9+=0LCBpw&I~d%Y{WM(pLH9MIWAq*U=Qk0bEn=9Y*NIzFv{2D7FQg_d zg>Wk0ruIFgr~Be@>~9B%aeGNbuS9&m!3zr3F+`p$CPN|)M3p$%x@yjqEs=006mIBa zKPy=dsSh`0qRCS6sQ0S9hZqZh5r|c{b$BhXGbM{TYM$ z%jKq7!j>D~D^HECj+9R1xil^d`5baE#Jl>}l?#qI?umi;x39eP=($AJ>oZHO<8A>B zgynQ?onOL|26ld=gKc1?K^21Xi5EHdeC?p>YkBtMW)C~)EW5W|5Z3F>7f>Sam=2j` z*sq+MDJBe_{4pwHA%bhxG;?4#7QP|%5deCCa z6DSSUKN)mVAlxmnNn-=dk}~=8G6;H(23=B=hi(>v6mIyZNq0C$e4RKNymBr_$ZisIL zzZHCmMrhgxDJN%t!ou$i>QF`7g0&gANfOdQ|CD~^o*m+(lIW`{N&~ZBn_v{Z(BPqK#j&W zI9U)w@C1E@b2JZb7~{t zBN_Wr+y`JgLeb7P71kA|yS)4Ge=XuZ`uH#}6c+PF$_1X|!5mdQ)l9xj15c4JY;(Vo z)5DP^!a&YQgu&<%Vfm73)Mac>we;<7WMPa4gGI%l>CBZtsOrC~+@fGNmPRth8Dv)r z=bE}4+?)So7Ytjr*d%a47Yxgm1W}i=C2M*NGH<3b4l%1_g;a}|0uY`{+q1dOd>s)I zvPmJXUxg+@GWBDL8Isx&gMBWCY}^fQ;3BEol)PrzH-(yQ$`RY9edO!$UvutT-#z(S zN7*MK)O9?fdr|wbB7`XG<;JrO0-t-4Bzw;OufYoMEzzHvd*d+H^0LWw@6TKo(1}Tk zzCut6x$kLCER?7s9ah1XSht2KVdy?#t7k-1wOE$sZ$8G}Xm zdG<;jkj5m|7JrWxyva1me#eaE!=cwq!56rS65x~8C+q)A_syK8vmCmQ&d^go+R3&JKXL~(=6R+*#b%XcQ(;5MHs1hKBX3~e<43gf!@1d? zh@7m|r&a$Vr&~FurX-Z8n-!k)f zl;Y`Ai7bPq;ri*f@7|J+s4(y!eE!-!6iTUTU}Zl8;f5-yWYPB61j*on{-O)i1tM4@ zSi`_wQvrVPtA}caPVYw1WdFq-C(T_VjQgd?cuh0N*irRZ_Sgn+T1KoW7v$ky3jD>* z)PpCYN+z7l%xw5N%x*ZJP93CHD%EsWtNDN4 zKahzCVAfjFT9S1K#MAe~?f0+?Y>dKph27&Uvt!MT*6*M89yU)=yAV@x7L!^C{q!1= ztI=4IQn%QW<5uJ(vMD7ch+#qZ-+B5|NPz>76&Mo#m3K0?K`FZdr^=fhT9>`Z_N30y zuw3d*&V`;5n#szCkcJ3DWd0s6i#%wP1Ju2}oddz6c0EDy%*Rf>`vSz9VpAm1 zNtl;rzJ#YX^_NJ>a_rfzxRtyUt9t2z=UR{_MoRrRUPy|5bg2@P@)(p;=hx?`U*tsEAe{RbUm(RYb|m>+I*hs^^%iwg^MhS_DITrElS5BMtw4Gxqf zWk51rZ=-<@ncMuc%|~FUZN~ZLKl-(xq-l+RZZVVTl7cnBI#6Yn z8!^xcf0j5*mpk%GRgVR6ir%G-GZV}a%*h&3j?&jy>4`_9+?#Qw;8Vr=v(|lhN!+9k z`k^PgjtY4IunRAzg*$|#*(qoY4CdNYOP8Gm(6%Vz3-X=!%9(IVYqsLK?!=&)mtRua zQxmhdr+)Bj@P89d;O5+kDV=O8`K27noLt2C6~s<0blEeI*lzl8F>bwbhs&b%OZ7z6 z*tb-I2O1E#JH*8+n@Wa{8h9wD*j#vjmC^rxcbg!>>y^Z8c9T;%;spEgWnQ*b75Lf zAU7`G>|jV4wO~jm?Gkw7LxPRRE8vxDJOljG!m;K*;SeLXvd!Kj_3BCoV#;J6vz?Al zS@4Mg>1-C1(CPLU!o-?UY~sL(zhQ z3p*oXRBD_0oGr;)|Da3;#;O9nGd;d8=x*=LE~UlgU3{yL1|U-V)R&1q@}fn##Tu&V ze_gZ!fd5&{{RLX}FDPy05lwm=q%Xap6dUYp?~ES58DjIU^&~APpVyXv7hCQCFeH zw(<%!9Cc2^X`P%u)tzQ%m^~-a@9j#V9-^(Z*0|kCY=MR)L^|x1+U8E8O0Z4SL6DzW z%VEzqw8U2BCB2oldm!Q&crzfGp(!aqTH{x4fA#CPRnht@&sJ&;)CtYDK0BqU>0ozQ zS*ncK{!-(uj(&m97tH+ZedWY2(%FmU7^#Ucdz+)p7@7S3>f~pHgy+&}I_Sn?67nj= zRCMZ_)Rxpl57#UAx!)Rz5YfJmM;Vzc^C~8=t7zf-HXX!UMMx%i9?a2)j^eg<0s5WU zO}cKr=Ql;{Ou(o?^by2odi7ij<1{n&Jfv99RZ~3n7V4ty+?By{UX6)$kh<&Jn>99E zY!fmd*t$M>tFYL7UlDLUK_~O!#Wceu`i~mbk|QC-kRlDl?WUfWpC{o>e5=s3xy0T6m1ZY&e% zB353t9_{O+LJqDFFj4(o#s*C;8R4i8Q6)!wCHafkr-(tNvv zy^^fVvvzC20d8$3;>ls9g#bEZ(lB*X{kE@cnTToyZx>2qLs-R!9j~$rBIvz!8HDHD zvn8{lcp56c4ZX;=8@b490x+=HdzS2nnwqys<^Ui2vZD0Gb(m9Y1${nw* zF-L%7nAu)DTx&EhEj9Xk&=oKxLd|NtJyC}ze0^U!Hk~)vm*l~>L?l0vN8FRP%GG%I z+#w+*%jB6F(^_ZA*wAY$UjJ>rL-B1{N!M%CYIgqg_mh{$1!G5>8-Th!T*&zw%d-5G zb?avxt>*?LU*=R*c*3Dxy)El<%q#q*UzaMp#VpRq$j>ETZI8hvv6hpaRg|*1j>NsLgL z@m;W6yqlYa82U$MWT4KFT6tu{{JaQLOY^SyNzi1j%9c5osBPe)`04b8{v>Opb5Kxg zs*{*JbO0}BdP=)?9N#H4Sg-3&R+oqJ^h5ltKZKgnN#x}CYI8#Br?;Fm^gPGtUeEZU zxmkd`ZLKq({zk${(x7yik)8L!X3X{XbSBu%(e}GWgKO_NCLqjzxcKpr{RcU^%(B`E z&&g6bS>$nWd_WTBZ|}^k&NBLq|3&-4(YE!l!LxT6)0h8n>EnNqqb0GBPO>XUzE02S z3uAKCkDji&*LM67cgtfeKKxA{7-lEokLUC4e+lppTWH6`d91YeIVsr_jmR}VL4nt+ z^`yleT^^hA<-fcVdGlYh|1p|D@@LfRq~t7TI@k2%ZS(fEBb~?hcGxzulG$a4|Nlz= z<){F@{FWB5Uf6qG;^{=()I3~*+%K!d+&iw*^=s_k2`F)jmmey2&~tr>jALrKP$wo3 z{}>#QQnfjJ!Z!YnG!@;FHsd2zs-}Pd=hyX~?oaM$P3RNt^q1_TF5U$~Z!1WLwI$S& zgM>2X@A{9z^Y}RmgvkooKV)YLu?l;W;Q~>177ZBc1SJIx7jGd0Tr{tXD9PV)6I%r> za+h#-@VKB%(GfbVqK@+mN0@IrA6~Eonr0cBy?>X_lv1}UUB!FHMlzi2b@jV+ZUC4= zJ|C$asO>VkxH718Ff@{}@{dwzZ}wYRd$;oO-3snYVckR1x{*c_qY}z`GKHn~FRKJhaWbhF#ua^nJ`TS_cC{hv8+M4hoh%>03B1?d?o_ zplsT=nFV3gUgyJ|VV2=$I{fdfkAf4_NzLwqtQNl4-RL5n9Nl8mJn;Kp@Whz$ZVogM z#Cg%~21J4=!O|L)pSo-gsh>p#0wlnF7wrx1-2!5O>&ddz=_5xCKoSq8{Nj{zbp`v6 zn!9bp?h?=OM}`_O7Pzu#0O0&PlCUawYET*n51Z)z`MK8Shd*pCVRi25%~+=s9E{2R zyRh5cg^=-YnP@xO;|(UXtNh)2Ac6lItEXdlj3IRQV*>w*aj)_>n{)o(6qK9(Z`qP0 SKla^kY(~uU)Ol7)xc>nP;IUf( From 0ff32abd72e30b0a4155978e3c461580252bb99c Mon Sep 17 00:00:00 2001 From: Claude Date: Wed, 4 Nov 2026 18:25:57 +0000 Subject: [PATCH 25/52] uk_centax: apply review B's CenTax 1 to 6 Row 2 restates the OBR. The 4 November reaction page is a second primary, and its GBP 800m (restated, line 49) and 'over half of CGT payers' (own, CGT package option, range_low) are staged, per the coordinator's call. Bunching counts propose parents_bunching; per-income-type shares are components; Table 1 shares carry their denominator. 110 -> 112 rows. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_01LwQq6JvunwtJCAckL6L4aM --- .../uk_centax/NOTES.md | 43 +++++++++++++----- .../uk_centax/claims_staged.jsonl.gz | Bin 10529 -> 11179 bytes .../uk_centax/manifest.jsonl | 1 + 3 files changed, 32 insertions(+), 12 deletions(-) diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_centax/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_centax/NOTES.md index 332ee6117..912f00ed7 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_centax/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_centax/NOTES.md @@ -11,14 +11,15 @@ The source slug is `centax`, as in the AB2025 `uk_centax` family, and `source_mo `stage.py` holds the rows as data and rewrites `manifest.jsonl` and `claims_staged.jsonl.gz` byte for byte. `stage.py --verify` checks each primary's sha256 and finds every quote in its primary's text (HTML text, or `pdftotext -layout`, whitespace-normalised). Result: 0 problems. -- **Rows:** 110 (own 108, restated 2). -- **Primaries:** 5. +- **Rows:** 112 (own 108, restated 4). +- **Primaries:** 6 (the reaction page in two versions, rule 7). ## Access recipe Every primary was fetched directly on 2026-10-29 with `curl -sSL` and the browser User-Agent in the brief (HTTP 200). Nothing came from a local copy, and LinkedIn was not used. - **The reaction page** was fetched at 16:17Z. It says "This page is being updated live" and carries no published or modified time, so a later fetch may hash differently. +- **The reaction page's 4 November version** (sha256 `5103e06ae3a5831a…`, 271,039 bytes, retrieved 2026-11-04 by review B and matched by the coordinator's own fetch that day) says "This page has finished updating". It adds seven sections the 29 October copy lacks: Big picture; Tax reliefs for founders and employees (with its CGT, BADR, Founders' Relief, EMI2/CSOP and evaluation parts); Company distributions framework. Their markup accounts for the growth from 248,542 bytes. It is a second primary (README rule 7): the 3 earlier rows keep citing the 29 October copy, whose quotes it still prints, and the 2 rows from the new sections cite it. This session did not re-open either version; the two new rows' quotes are the coordinator's, checked against its local copy. - **The two briefing landing pages** were fetched at 16:30Z. Each prints no figures (0 rows) and links its PDF; they are manifested as the route to the PDFs. - **The two PDFs** were fetched at 16:31Z and read with `pdftotext -layout`. Locators are printed page numbers; in "Taxes at the top" the Key findings are p. 1. @@ -26,7 +27,8 @@ Every primary was fetched directly on 2026-10-29 with `curl -sSL` and the browse | primary | sha256 | rows | |---|---|--:| -| Autumn Budget 2026: Initial reaction (HTML, 28 Oct) | `92c6008835547d41…` | 3 | +| Autumn Budget 2026: Initial reaction (HTML, 28 Oct; fetched 29 Oct) | `92c6008835547d41…` | 3 | +| Autumn Budget 2026: Initial reaction (HTML, 4 Nov version, finished updating) | `5103e06ae3a5831a…` | 2 | | Taxes at the top, landing page (HTML) | `c954e08f2aa6de7a…` | 0 | | Taxes at the top (PDF, 35 pp., 23 Sep) | `af4cab614220c286…` | 104 | | Unicorn Relief, landing page (HTML) | `91a6c8dc749273b1…` | 0 | @@ -36,9 +38,10 @@ Every primary was fetched directly on 2026-10-29 with `curl -sSL` and the browse | measure_key | rows | |---|--:| -| `null`: 52 with a proposed key, 56 context (descriptive HMRC statistics, current-policy childcare behaviour, OBR CGT forecasts) | 108 | +| `null`: 53 with a proposed key, 56 context (descriptive HMRC statistics, current-policy childcare behaviour, OBR CGT forecasts) | 109 | | `ab2026__pension_tax_allowance` (CenTax's 53% marginal rate) | 1 | -| `ab2026__hmrc_tax_gap_package` (the timely-payment component, `row_type: measure_component`, README rule 9) | 1 | +| `ab2026__hmrc_tax_gap_package` (the timely-payment component, restated from the OBR, `row_type: measure_component`, README rule 9) | 1 | +| `ab2026__distributions_framework_modernised` (the £800m, restated from Table 6.1 line 49) | 1 | ## Proposed measure keys @@ -46,7 +49,7 @@ The Budget took none of the options below. Founders' Relief (Table 6.1 line 26) | proposed key | scope | rows | |---|---|--:| -| `ab2026_option__cgt_reform_package` | "Structural reform of CGT and equalising rates with Income Tax": equalise CGT rates with Income Tax rates; an investment allowance for the risk-free rate; remove uplift at death; rebasing on arrival with deemed disposal on departure. "We use administrative data from HMRC to cost the first four elements"; the fifth, more generous loss relief, isn't costed. No Table 6.1 line | 9 | +| `ab2026_option__cgt_reform_package` | "Structural reform of CGT and equalising rates with Income Tax": equalise CGT rates with Income Tax rates; an investment allowance for the risk-free rate; remove uplift at death; rebasing on arrival with deemed disposal on departure. "We use administrative data from HMRC to cost the first four elements"; the fifth, more generous loss relief, isn't costed. No Table 6.1 line | 10 | | `ab2026_option__nics_on_partnership_income` | "Applying an equivalent of Employer NICs to partnership profits": a top-up rate ("6.9% on partnership profits taxed at the additional rate, 7.6% … higher rate, and 9.6% … basic rate"), a £10,500 Partnership allowance and a £5,000 annual exempt amount per partner. No Table 6.1 line | 24 | | `ab2026_option__nics_on_investment_income` | "Applying an equivalent of Class 1 NICs to investment income": rental income, savings income outside ISAs and other non-dividend investment income (`conditions.income_type`). No Table 6.1 line | 10 | | `ab2026_option__childcare_100k_threshold_removed` | "Removing the £100,000 childcare cliff-edge": make the free childcare entitlement in England available to parents with incomes over £100,000. No Table 6.1 line | 3 | @@ -54,9 +57,9 @@ The Budget took none of the options below. Founders' Relief (Table 6.1 line 26) | `ab2026_option__childcare_100k_threshold_raised` | "Increase the threshold at which entitlement to free childcare is withdrawn", to £125,000 or £150,000 (`conditions.threshold`) | 2 | | `ab2026_option__childcare_100k_entitlement_taper` | "'Taper' the withdrawal of free childcare", at 8p, 18p or 28p per £1 over £100,000 (`conditions.taper_rate`) | 3 | -Every childcare option is "costed on an actual ANI basis" (`conditions.costing_basis`). CenTax's actual-ANI threshold itself ("could be set higher (£111,000) … at the same overall fiscal cost") is that option's design and has no row (README rule 8). +Only the options short of full removal (rows 101–106) are "costed on an actual ANI basis" (`conditions.costing_basis`); the full removal's £640m and £520m (rows 1, 25 and 99) carry no costing basis. CenTax's actual-ANI threshold itself ("could be set higher (£111,000) … at the same overall fiscal cost") is that option's design and has no row (README rule 8). -The £1.6bn timely-payment figure carries `ab2026__hmrc_tax_gap_package` under README rule 9, not a proposed key. More timely payment for Self Assessment is HM Treasury Table 6.3 line 3, a sub-measure of Table 6.1 line 45. The row has `conditions.row_type: measure_component`, with `conditions.component` naming the part. +The £1.6bn timely-payment figure carries `ab2026__hmrc_tax_gap_package` under README rule 9, not a proposed key. It repeats the OBR's printed average (EFO paras 3.4 and 3.13), so it is `restated` (review B, finding 1). More timely payment for Self Assessment is HM Treasury Table 6.3 line 3, a sub-measure of Table 6.1 line 45. The row has `conditions.row_type: measure_component`, with `conditions.component` naming the part. ## Conventions @@ -67,19 +70,22 @@ The £1.6bn timely-payment figure carries `ab2026__hmrc_tax_gap_package` under R - **"Central":** CenTax's central scenario (`conditions.scenario: central`). It uses "elasticities drawn from the best available international evidence" for CGT, its "central revenue estimate, after accounting for behavioural responses" for partnership NICs, and its central bunching scenario for childcare. The investment-income NICs figures are static upper bounds ("naturally at the upper bound", footnote 13), so they are `range_high` with `basis: static`. - **Units:** - Shares are fractions (`share`). - - Effective and marginal rates are `percent`; rate gaps and top-up rates are `percentage_points`. + - Effective and marginal rates are `percent`; rate gaps are `percentage_points`. No top-up rate is staged: the partnership top-up rates are the option's design (below). - The CGT package's "increase revenue by 62%" is `revenue_change` in `percent` of baseline CGT revenue. - - Income groups follow the AB2025 names (`top_10pct`, `top_1pct`, `top_0.1pct`, `top_0.01pct`, `decile_10`, `bottom_50pct`); percentile bands are `p50_p60` … `p99.5_p99.9`. + - Income groups follow the AB2025 names (`top_10pct`, `top_1pct`, `top_0.1pct`, `top_0.01pct`, `decile_10`, `bottom_50pct`); percentile bands are `p50_p60` … `p99.5_p99.9`. Single percentiles are `median`, `p99` and `p99.99`, and p. 11's "by around 2.8 percentage points on average" is `top groups, on average`. - **Attribution:** - CenTax's model estimates are `own` / `different_model`. - Descriptive HMRC statistics are `own` / `administrative_fact`, as in AB2025. - The OBR's £32bn and £16bn CGT forecasts are `restated`. - - CenTax's 53% marginal rate and its £1.6bn three-year average of HM Treasury's timely-payment costing are `own` with `source_model: arithmetic`. The £1.6bn row carries `conditions.derived_from` naming Table 6.3 line 3, the same rule as uk_tpa. + - The £1.6bn timely-payment average repeats the OBR's printed figure, so it is `restated` with `source_model: obr_efo_forecast` and `conditions.originator` naming EFO paras 3.4 and 3.13 (review B, finding 1). + - The £800m for the distributions framework (4 November version) restates Table 6.1 line 49: `restated`, `hmt_costing_obr_certified`. + - CenTax's 53% marginal rate is `own` with `source_model: arithmetic`. + - The "over half of CGT payers" claim (4 November version) is CenTax's own, about its own CGT package: `own`, `share_with_tax_cut`, value 0.5, `range_low`, medium confidence (no year printed; staged with the package's 2030 costing). - **Context rows:** the descriptive 2022 tax statistics and the current-policy childcare behaviour (bunching counts, response shares, partners leaving work) have no measure key and no proposal. The option keys carry only the options' costings and the distribution of their revenue or affected taxpayers. ## Proposed metrics (registered Metrics used where they fit) -Registered: `revenue_change`, `revenue_share` (also for "paid 60% of Income Tax and NICs"), `income_share`, `income_statistic`, `average_tax_rate`, `affected_count`, `affected_share`, `participation_rate`, `tax_collections`. +Registered: `revenue_change`, `revenue_share` (also for "paid 60% of Income Tax and NICs"), `income_share`, `income_statistic`, `average_tax_rate`, `affected_share`, `share_with_tax_cut`, `participation_rate`, `tax_collections`. Proposed, with CenTax's own words: @@ -89,6 +95,7 @@ Proposed, with CenTax's own words: - **`behavioural_response_share` (share):** "fewer than one in ten employees with eligible children responded to the cliff-edge". - **`workforce_exit_share` and `female_share` (share):** "3% of lower paid partners … leaving the workforce, with 90% of those partners being women". - **`partner_not_in_work_share` (share):** "the share of parents whose partner is not in work is about 6%". +- **`parents_bunching` (count):** "from 1,100 in 2022 to 11,980 by 2030" (rows 26–27; the label of the repository's childcare adapter, which drops these counts as "a behavioural-model count of parents" with no registered metric; review B, finding 3). ## Not staged, and why @@ -118,6 +125,8 @@ Proposed, with CenTax's own words: - **coverage:** #207 holds only Table 4.2's full removal; the cheaper options (rows 101–106) come only from "Taxes at the top"; - **bunching counts:** row 27 has "11,980 by 2030", where #207's 2030 central "Number Bunching, Before Removal" is 16,280. This may be a different concept, or CenTax may differ between its two publications; row 27's note says so. The £520m and £640m staged here are CenTax repeating its own central net fiscal cost in AB2026 publications; the ingest should treat them as the same estimate as #207's. - **LinkedIn:** CenTax's 28 October LinkedIn post (seed inventory) needs a login; not used. +- **The 4 November version's design figures** (README rule 8): the BADR lifetime limit's cut from £1 million to £500,000, Founders' Relief's 10% rate on up to £10 million of gains, and the EMI, EMI2 and CSOP limits. They are the measures' parameters, not estimates. Its "Big picture" section repeats the Budget's headline figures and holds no CenTax estimate. +- **"Taxes at the top", pp. 2–3:** "around two-thirds of total tax revenue" (a description of the tax base, not an estimate about the Budget or an option) and the "over 100%" marginal rates at the childcare cliff edge (the current regime's design, restated). ## Review fixes, 3 November 2026 @@ -132,3 +141,13 @@ The thinktanks-a review (`harvest-staging/ab2026/reviews/verify-thinktanks-a.md` - **Finding 7:** rows 20, 58, 108 and 109 are CenTax's 2024-era estimates at fy 2025-26 under the CGT option key; their `scenario` condition says so, for tranche 2. - **Verification:** the contract checker reports 110 rows, 0 problems, and a re-run of `stage.py` reproduces both files byte for byte. COUNTS.json pins the gzip's sha256. +## Review fixes, 4 November 2026 + +Review B (PR #231, issue comment 6635498758) approved this family with fixes, and the coordinator's brief settled finding 2. They sit in an explicit block after the 3 November fixes in `stage.py`. This session re-opened no primary. + +- **Finding 1:** row 2 (£1.6bn) is `restated`, `source_model: obr_efo_forecast`, with `conditions.originator` "OBR, Economic and fiscal outlook October 2026, paras 3.4 and 3.13" in place of `derived_from`. +- **Finding 2 (coordinator's call):** the reaction page's 4 November version is a second primary (`5103e06a…`). Two rows are staged from it (rows 110–111): the £800m in 2029-30 for the distributions framework, `restated` on `ab2026__distributions_framework_modernised`; and "over half of CGT payers" would pay less tax under CenTax's package, `own` on `ab2026_option__cgt_reform_package`, 0.5 `range_low`. The new sections' design figures are listed under "Not staged". +- **Finding 3:** rows 26–27 (the bunching counts) drop `affected_count` for `proposed_metric: parents_bunching`. +- **Finding 4:** rows 87–92 are `measure_component`s of the investment-income option, `component` = the row's `income_type`. +- **Finding 5:** Table 1's share-of-affected-population rows (67, 69, …, 83) carry `conditions.denominator: "taxpayers affected by the reform"`. +- **Finding 6:** the costing-basis sentence names rows 101–106 only; no top-up rate is staged; the income-group list adds `median`, `p99`, `p99.99` and `top groups, on average`; row 0 has a `year_note`; pp. 2–3's two-thirds and "over 100%" rates are listed under "Not staged". diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_centax/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_centax/claims_staged.jsonl.gz index d4f855416f313180cd19e16e908aca442adeb225..eca254b968ab2e76446a243baf9de57aaf540c55 100644 GIT binary patch literal 11179 zcmXxqV|-mr+c)53C)u%Wqp@wfv5m$?!^UcC+h}a=G`1Vtw)M8x{k-c~>+AgH%gj3G ze-3dZ4A{R1@OkcQ=ddbWfARQAho_B`0xq-NQH1bdU{=q=7POwOO`*LO3y^50kNKj~ zSF+=Q4ZY-7_vkYzH1d)7PT5k?@k}P?Oe59&r5y{}!sI8@`hxn@ocg-{eJu8m9E%y- zY$>ao9Zc_d18Sz1vw?AJ{unet{bDDKH4xJXZ_G7x4rF9;uD(Lh5;lMsw(vR2f`6oR zjD3$qkf=xoXPfRi)UPv5itK9l%O?!iHu?vzAyeY9Euq1Uk2P*e81lq58`D1_q2&OihgYM)ROdE7oW}4DJr?cg&^CUZwM<{>$Oa|c5gC%y}8B2ZExYj+u9}g zzr79xA!hU>VSQ*Gg;;#p1T9v@Ue7wGe0nvsllJ&KL7Z!jLYY+MFJ1|(ZylE1V9Ehy z&)=V-<%n2?EOgj?J^T1a#jA+|n6{F8PQLm0Gxad?&Ae_+nM`)(k2nga+jN=S((ALv zc7+c1-~7U6`oLdl*z4A^@bMZb?tjOP8F#Jd;0hs9klQwTO=Q=7ak~57qm)zOcVe;7 zz{MXx)yw>Z3*$R-OXER_k06T$B`aO&!Ek}oO!^AevkZ&RyKp{z@SZI|g}FvGPlzq}9zoI6lz3b|0W#B@<+xz9#kO#uo0NhiykS z8yw+WmS^Fznf)^VUci(T`KFUaH&;f~`gD^3_1!lgEa$Pb*l%l2Qs!a(fffre4McBT zQ%UPv(w)FBYcMGMG&FCF-b|Q6u{T=o_f3$1hb~A!Qu*W0_axBL7K)3ZJ3O(LMIAv1 zzhQ(XlOgbm@N~E-@#P6-w9t)VfQ#DFkoz)*F>HHb=u=#mP94qQ!?gvJ2a~#F=7KR(sB>HhhQqYWw>fb9Kox*{(_a;5PfN+%(iPE%@&c7Wv5D zMMO!20c8--ro0zBzavl74-yebG^$!I?;r#P|M8k(p?l|!)M|sS9C3b}VbwRWHKq}P3Fl>fNsccF)fPv%!rCB71UG+TdZv!U z0R>IBa?@<`gYt=)0{Dqb`E)M%YGlv zLMGI86TKeciw`$C8jH2F^twpnG#z?%d8+4UtwQVR6>QK0lEHQZjnoytEs z%<>h$FSp3!NLihq{EHy_ARb7Z-KLzO1e5UAbGM;GXTu#pf9&^+wzay4tR8GgLDl3XaIan?8xG@!aEAk zd#OAgMXvUda0{E=s!|L}2 zqxsBVHt~&Vk#VDJVgFLgLH~yD_0( zB_V9~iP-PQW2hZcx+_(8-FDn+^0MV{P09AXutw-R!0#YE#vC_*BeYz@D0nv>zUt^{t%aY@k&q~AErJj3fZ!v? zdwU#tJf6p0%H`GL$lDuaSl2^I3R1lo`vE4~>p;T?{7GiI1f#$mt8|f0Mjl<-wYV)M z6!CI4biXfJM4R{gqf76%%e!vO;eaN~yB9}XP}UyPVkP&z*fkolkCnVZ#pO}W5%TbJ z@Yy723VJ#l+|{BL?y%+B@Jq8*mPi8x5*{XNaVAdghG|%4Sx-O(^)6~siS||b)qFF}8CqHispeRNH-pse!G@g{(0bmcWxps&lq(G%oTSEvk3-OtAZ!p;%u` z0aJ2`Jm-D5yFid%Q?hNz=t_PZ9ZLjr&8@?R%0qyEcJ&uoY)#ABBHYr5TDv~K1NvwS;{JVAd4?3` zEp^jC)P()K44PF_v2i_5tZ8u~l?RwGYRQ7RD03#zK^|@cg>K>MgAy{5 zKNtemm?IqpKh9g&_Ug{ou`App-sVZuS@zn|#sL1Q0f9*ds?iSiy?OC?OKI-!l2qsh z6A|J8K)yr169#;g92O}du%8NHyo3dRuwPT(IY7KTK3$%|kSI`P%5X6dHUP1g?bcXb z^2R+ZwXvI|<9?BeG%LLFS{YJ10X!l5Vf$gH0?`$}i1)wqXO4=3(_G^S@Mg!(1k~v` z3MDNwz?<=&6BaRmeK1&lMOPmTpgy6fUM33OoP0E#wsR|cGPVMd;#uM%&d+0dku3qu zY(Y?Ph|EQOLv9nH+=e4n`i>69CxuZ1mbJm>$o}VoufFFNPb{$KDRN7SOB*kFe#K2o zxw<4X$}}w*R0cF5x0d(nNWta~M$s#-mX#Xc2678k@;QuW33#N0>4t^1d?wZG-FwA@ zCHzoIn?;3EB4tJ@TahFK937^#jzq{>(M&vH&gGUL*sc;MXufK_u>cCO z+OigHQ&N)Is-*SGz2csPbB|wiND?e_c85cZpy4}XaA4L&+bnPydCR_r5|qf3!jpr0 z1SJ1NYlWmr*o7?9#QZq`XI=k;KrzJ(DoGBBJf}MTHJFAfOAAVI-R-&n-RFQqAZ1HC z`+)6x9wR?RE+*JD_L0TWP-hGZw`ogOWIeq}?!=Y8_0JylisAUKg^j71CuMJRMfJ?< z6-^ZC$W*Hvo(Mp)p;K`lrO~9e!pdm!ZUrZC3VDzSgogl)^Z{z2U|&5W3DT2L`jaTo zr0#I{T3e!Ev2(sbiOG>qym-BE_PUicInbS-D!HO1nQ~E$P)g(Gn|1_J)XSK~Dy_{_ z{!ZQ>tF=Wt@1^#MJrSaO_S=t)BsW=OK-{B`yJ}K)N?{Z>6kVh;^0N00V+1_D^Y+)a zqDno=jgOnD)i)mR7tlN6VX2lp{Ej8%nV--6nJ!?I1%|w))78 z<|DRtI?am?QIDIV?lu^=_0y)rHNJz3RpK&1bmU64O&hDLA3l#+*jT1iR9Wco!JxP# zYpj-vY-JkCCT1I+h>$~{CQ~#;cAL-=w=ErYg!X>uco*%6W6*8vJ%HEZe)gx*<$;PT zU>F`xbSXBem>8$#|#=$YaCm${&9<;+U1=?Ax0ry z8-On3boN%F7)=ig^gGT^A{Cw2FgFc;k8L)0m}M8|!WHVZPF-ukB3Qhgt~Ho14suHE z@R$~pAZu-fZr8MTTjwEV3hv@&(tz-3;^QlOCLm+bxO3;jV+DiuMXTga;NxS);gegE zd#Yi%8};L9C^8ou3Fj^|C%6OlhNTzFslW#x6Sj*?kb)1!ZS$699deQsMSu0jK-m_~ zT{sn473pIctP)DCb^*5@{o(bgqYjeKKgX4uHA9gkRu1Spm(1zz*v9c*k48ZKrh{|F zklS}`KyYT~jJu0k0HN_6OPK&DyuB5KqdDh%d8{U3-RB+F&pf(<$*A+EZMf29hObKk zMW{1EwkoM=;~wNiT*P*S@JkR#F)eIx^L1LxUn3vYMoEwZbdB=?EStCRtBsl&h%>H+ z97|5Ii@v+)PjOHrK(}d%YTC%xH=U~OJ8Ht2r7Umc2nQbm5-644l|d>d8pXkhEs5xR z8C%>IXQWmjp3~5^;i>r|^+F7avY}OyE?G4s91`F`RKmBy!&;q)wdJUfnELEV_bQP# z5=o|OV1m!NJEq&v&S(D0*LB>Ku!&8$Sc7~#mA^VOZNnTSMony}_Y4fFM=n;SP+_=g30PC}pN4EcNB4C8{Y z1|i*h7*nxRJ*wB&lz9tLY`Mul2OA0RuHRZyxHpRJ{WJfa^+HkW-+0YAexQgwz{i?t zfo~8L;od+XBR}$nr~i||hJM%Zg+nIabvY<*s0h0LM4?VNj5B!fods^uzjBsnTJ(qR zZpvwF-XF@>#$;#PE)@Hv?@BSWpQT|Mh;0nQpl`oW1$OW;qP|m|`#HNKl4j4VC577D)R{-O$9;~0|Pt@5?nXcDeq z~V|P%cX=YAa8+;o#)_GB4V9TI->jqxyz);$LKq^-* zOA)-X8M=ZVy#o24Fmo43WJ_#~O!5w2SsW{9qRk%OMUt(oqUN+Kf7)EW!^y{+}T}4Jcm$$a%SK zeh95i#WP`4$3UyVf&u|y$mtD?JkSm3n ze_5a}Js^z@lov^oD9l&`{y7=q%eo|L1e8Nb2rGzK23r4hG9*uh;446r=t!Dn5%Gu5 zT9EA-`iSqOhc@o5AT&c_a#YSwo#;(+Rrhbnl5ArwK|JkaM%!L5%Dw^XAnT6pV}Eb_ zQ3~~Ve#?DgNqwTF-0cNt>O7vvv&)OV)DOvE8i?CSB#+t9fjA_Q?o_Kmm9Mx~zlo&& zV=&>oA{HMcWSb%BQQB(hGP0g2=gvGZ&#qD7Esiu5LqOFJ&9DoB5*nVO97Sb>NWF^k zU0L?41A_-uC$rT!Jo-*Z;T2%!A5oj;K1(jJkev%xOfSbEFus=14SOk3?dO385`>mI z@@+v?sRCt!2rd8rY?-YCYwA|F;#VCJT6%HN%u)VoG%w~cFQLT$b^+G9oRkFaoS}if|Nc@dC9PfV=WAQRxP)nrd9dS5p)xj>t4lOa(3d zQTVL+TdXh3-Dl(|eC@rz!XB@N-aU=NR2B!F2bja17136+@bo3%HG5OZWk`)^o|fja zH#?BLn0?(|;_ljzyoWW1pwgoLj{{fL>ILaNNJR8&%<0+`+fki${ zV|0Fc2=SEqGE&P#9qth?80HjFo^_bE5WyA-s0CrGIIyIG(V{CMfuAjuji&bkGAicJ zl_UP3iINXP&?5o-TR1`0#7Z4+jckd1G^O$9b)Aqxtpga$sl2kk1c5NYv?x*;T}9PK zk<;gf^sKT>&n!XDf=8X$l>zhLq&n| zB*?ZX1iThr;FyUe_~4quDSMorXlTI+dh+OO@9_bB{GUy>Jan`BA#?NFAux&rO zYeT0-4yq_Wl&G)=upN|H^w}2T;~H#bdS`cC3x}yg4h?>9{!I+k4yAR@NoGWwD0i=$ z(i%%9qyV0PXWxgN`EQ3L>AHEs9d-weG9Ugs)pA6VKSaxKb(^OMzXoUupP`|&FvwS@ zjCK2P4!y@4*XGn0Mpmb!;>9ABGg6v6n=sq*Ld^M&KG)G&)_>+D=*7*j2`LklGf(|o z@O*adrfD{w*)A^hZ$BnNpZ{{}yXxpOxtkVo#2RtH{97hQ2B=hiHJ?C#V%?7O)-3t* zbPK~CAD!_npWs~lG0HSDX`2XeZOoo~ zi-1>1=?vW&wC1@L`9>s?Gy>WPyR^d6my|Is@A>grJ+&bUuYvUpT#Y>mLG||s$yUX`5pBH6>@Pb>Z6u1;I@;3p~ z#uyrQ{+EC~7a>iMFlt0`(y8S^Zi<2a2+}UaR6K<%G(oLu>Bd2wB247G^FrUL&%PMw zFnJFR!0}=RtVlc~o(3f7tLbQDpoZW6t2CR?H}lRfmvFKt>%Vk5>qzB=Ay*E+1fKXs z2hk<-D09xiY`nsD9ViRPLxhq1 zH`HnWFO3`?jE#lwZOR6^VR(aUM6?%U{NP1oye%jCNG8@hE{Fq!u0c!3)7-n|eFM|- zhohDCC&G(Ca~Byy#Y2-Z{ER)L+Hul46ZU=mqII>o$iK=1JEWQF@NKc1#_i$QVh1+{ zg+`?oBrhB$)?@wIMgyC7M13HwUyPgAb7mLE2>wV82HXML@y{lhhQP@DTSM1V9=qlG zVuXJxB7Qzs?~ltO&+IOMUHuGD7*#6kte8w6<_B+@U%s&>-UD~xsqOWN=nfzrI&R8^ zT<{1uF+e=P)yy1MZcF{L1?DQ0amX!8`gjXvkXvqMO0!EJ30AT|EjF2I1aAx}7%JQ& z7$9c^RTk<&X+9tDz!U(2amFlZPZ(dYf+7J9mpt+1@BQ zx30m|LA||BeIl>F-+60=#M}-uQQvDER6nUM*0$;>)Uyh2_hDlWh_5c0nSa1~F}Uf) zXWL0nh?9p;Gctz@?iL8>Y1;3t6!6L;(QxTx;8Q%BZ5OeyXk_A^YFZ}i;KHRrhIPv` zm17URDFuY-H6&(~{(ghG3Nb#}K+DhO&Ocwmx90pc>fH%B`m3J|kDjHks+F==`1x>h7>Ag%Ou78^ysbwk4x_ zC@9C(s|B+Ha^;pxXWP||Xn%)d63QW0V`&T_rZm0xy1nJnN~IJx(z3*yJ<$mC6gpC9 zDxlO44%XNX%bY(qvSzOdHtd^fTni2$&K z_{|IZ>iF*9@1~801?8QR0WR zhReF=^Hrchck)DV_P14A7wX5x8*_URwU()qq|&coY36?R(VK2 zUhELwpvZj`7j%rro`{FI#Y~xp*a>Le)tMlQD>Oz~v||p=k}H0}(QJ!Y+iPO7Cv_L3 zw|kc5-)$~z7>%e}!A=yj)}Yw?9QkM|u(W3)fu#vwa5J$&F-j25ONXjWD+~F{oT=$7qDTPy2Y95lG_x#hP9lE2 zbM)~jTl;B1n!)kkkIxezVhQ3eh9L0QmH(}>8HkO)zD6G|9BF02*e)5gk6sl#G=OaxAw(^>HqRvV$i>h zlR{rklP<^}DcPH5E|!^`E>h53ci}%fup#37<>{KCipNca2FYH|T_RT``cKXf<2R@A zlIk#(e^ZJIprXol6Od2S#Gs#{MW)CYw^(4~Y4J)DBQGp%4 zJYvpgzv9CH9sx)WEXDu-c#|hY{>Tb#5>~wp&H_}!7>SeZLa;iw$d8BE4nG1ZTYopL7M*xMF1e4zT`y|A;$fMVYiw02Pj-{cD{1a#6%r&PK zZ6jyJkf(B@eG$0_~O2Mrh zxEJG^sfp)7ZoqIIvG}K-x}UC}ZZClU`le%%`#IvVY~<^xpm8kt;~wFddWPi=C=C3Z z4n~l-VULTh`&JH`RWiwEkIc}-qVhW!SPC(Xf$BXnG;h1S<5?}A&g9h$%zeyz{JJC6YwI@!Z!m!!H+V)$Z6i*7SVgQ9y)w(2M|S~tg!AVop$dNsr`j*lGQPoEk>eA( zP0Aax|HCWj>G`-@y|UF9+LeK{6f~nX2Smw5{Xc}cQ(3P!>2th--cOFkZJikZ3wLzT z<<{DGF)7hhA*Ilfy6fo&3bI1Kh9lepe+ILy?*J~g*L&v6F?7NnKt zr&$L2%3oOtr*rHL z0WOl>sCdu9zgEhw5JoK3YG=Zi{y9oovYbzfmK2wo=pr+raE=X+`$rKZ(7tfkRzvSi zOx?S*cS_#|-9vI^cRlvERU3+$0x+184ctD#;W{c5H#-_P4=zm)lE4Vd<7Zf2?F{Jz z@h0ec6rk5FfIm&}iggi}i5RnWgEc|_`b}+CsI<^+I@>R_?>qgyq-S)iGR#&>a`;z+ zjR@gBweb>YDS$#ES}F$kr*DxG}*(Io~~7XzNl4&NIyHdS{o02VMA0(aF{@B_-6UuQ3_4p z=d8Zqs-LWBqNcm2gGf#^Y z-Q{14jNyhaS*tBLv^NmjrE&>~QPiM&Z$a44b!25Y2s8x~i017LSAl^gUcukiux#ElDSP_Gog7)dhM0d)0* zgcG`a=>T_+l&j3-iSsKJ8E>9S$^Q9^{ZX829xIRJqB21PLaa+)m8TeLQ@lG)awb-8 zMP%`qNF*S|+*Xu{XWS#ppv69>11f)1ciH#qG{2;7RT(Cm4kLsuI^j#)@&W`TDU7jh zOl?*KS%a;W3cOM>St>mqco_^CTm{UB_K`!_O!M_}wF=f0g7v-`*E`IM6PcqfCK`1s zWvee)E0nEKI9~{+j6*z~H|VHXtH>#wFWE7Hu43t`Xikfd2gW9MM;IN)MerBH>KcTj z6oicr_7>-wBsKOXAZ{B&>Csw;cH-{^0RP=9_eSt!QLd+pJC|04YP#jal>0Lm{JGQb zJ_GN1P44JuT3@sYIgJI9y*F_I1$}4G(L0oJ5K~}_!J&~N&+}kIrcarmag%@1qe$Cv z)isL@h!V|`4c4D|=ji7Um!&=1zk&4d#eC1Z+PcbDb z7d@oPdAtcCv+D1wbuVU@MYm%COAmXgs|w`!ffp;CgqD!iOvM6D=a%kF== zXA%+sS?|ZMg&y+*`Z6;vT)_!Z=QwnB=CSq^*NzqWm#wlCiZ0oBd~AX5yFA-JPA3nN zM#_q{lhA*Rb#iRCM3iH3eW64vWHanDW~-Ge?Up7f|V&L_XAq*r4eC(-_D=Ys&Am`CDMA#W8{b z>04w5ZLeju>%jf)EL68f$kTSjwd|ko&Y zB#MhXWq7tnjbI#MP!cB{TtbEVmelPK$t_keG@sg#k}?E6?w6axnRDWp9~PI_%n%}J zg8nj%WaaiBn4MAw;)Y;cM_1RwHf$qa4097%CRrg}4HvEm_OcfcCaPB1$B*`Ai6%J% z1jA*bME|NEpdm$G3>crWj_DQNdxq)d^pMn9N}by6*wJ{jKyJ`IV-ZXe7-TSw-QDN6 zA4N!gR(?@n9SaQhEFX*Y_+8RTo1W^A3Z&k%5mRd zy`izj4nLm{Ki_h#^*-nh7JWTVH&ajP@qCco;VYVZ;d+3cFXvfCcp2zdL7BOgbA6e_ zf(Wx_wKH`A_MgH{em%&6Qu}H!l~+RJBaUhn=N|&<)@qYbK~rpYH#N-bELvcl&^0?% z3J%_%GTGH2mN?Bz+jdHf%D!n=8KLf&(q2i{Q7GaZ{MyK7-f%0!r_SHaZy6tterNI``&q(q2Bphf7XSg$XcUTk%4ON$Tl zb%nsi82av1D6(B#dS0&}7hmanB=WbBQ8g@3meR z#9s9^7lC+4g-^c{pAx~zU5YCeWGor1V=$x1(b-B==seVab~GoJBkj}v9F;bDe@ zj_)y@u^h8Y7z<`7cVA7nN9V7l#17)Y;cB8;a#>hmjp` zPw7i+t@3QGNQ(7+cK0;`npg55THrLsO~;;#*Ua<{A87eI>zw%55Ka~xq3_?y1+!|F zrb0NRx6g5HlDk(-YmBvaS~9~8nFzrKcL zMF|{NdybTx1}H|!olX6C@?K6?BSe)dH_fC^gjh%+$gMjOXL6r2`fn0DrLKdMdT*o% zT>RE*;!`)54_<7{IuSFu9mc3&Cphfo0j@5~v|!pUgrip#uO+MEmAmCiws{pfG@R6( zNVhkyrg%TU5$H+FBa-Y!FBfxO8YPH+s$ECC`}QgMv`Y)OXNz8K+d(gYG9QvM)|L^6 zbuu`M`BkZI5>>f3%d>=qkR7qKlCna>B)MLESR?RB~j z_R(7RB9BIb`FCL7PmPY_Rz?%weydHV8*>neCwZlrQ@39@bqFGkjDT$So5d4ZBgvv> zDH-FxQ;4rqWo`Il@2gzV7|gA8Skc6m)qaH(GU=XNs&}~R==gonar2nlU|VHT3LarX z**e&wZI5s$ie_t1Z(+m;iHJ{gBJlk7K&u}{`PNPC(B+t%Rg2$d9Tor3<2s5>X0a`E zat$WG-x|ia`0m-a!B#AB)Vi4-5S(=Z^7vkfH|<$>Whp{ToVf{=(tH_*f8jkKlUN-n9sBDKyWF_-HKMnNf&El^ryhKLF{~#0@(~{EN zAGM&!)C2c~bKP_I@o&2nFY&Wiw~ebTae)eBo)_=~!ev=nvZ}B?bZr(!pTWs+l2zQt zZsKki@-7}QsN&fZyv$9=RrnUX=(+Ep#eZnYE7-7y?KHe#x%eEl*ISB^Ydgnf`15|a z?9Yd82i(@k@L<#i(@~*9u?3bz|Byfb8RKz+x!%0{(g@?^%FusZpPe3WyuPH;RwUGK zx{rQeRV%~3=z0cM_^&~=ey-1m4!~XT8AhTJ|n0$iuslF!68*%W+2fM?W1l9h3AyugVD_!5&2{= z3yzT=D@*(<1q+77QHQyh3t1`nRLKFrJOv zzc=CY+X+g+!tbgSI@2O5e}h`ag?r7H07+74k+J$%LI8SQqdJk!>SqECd+{c($Vx?l z8&}R;L@ZhNxJe~XwOioS**eeW#jZ?86xRcVt}A#VwA=Y07^33oKGxbFCk8Bz2>+6H z3K_K1^KGsHKbQ}K#nJmwgRC!^5*6783_k<2kR)bkpUTIuET`X4fYl-?t=OE4kK7lX z-=|(4WRH4-<1s<~yzmp>&ge{6CGe-PL~9HqYe9U-0Z%&+X5UJ?y?_{p;9-1^0f}9* znjV9WsF1t+_otO)4&`Zbm~dN>O_zI3m>I``PQ;l%rIt1ZH`jcbVNpU%;0BbgZCbu> zB6oWd2R>{sRo|8;1hVKwa)t`TR3BhnB|4bkH+u;NAE6Zp)Q~L(2&jYS5}G++8e^nl@ua*NfEK2PM#*y1^_aM(!|l9-SGwPhC9nu@36 z2TR@xpo+Pp{0h_a3n>G@z>TJB5mMUZpMs$iU#OD-7A@=#mI(yiQss`dIvuUyQfu$& zR6H+(QwR%w-R5+>k`1zeYVrqEZ$#6s&2aVnGSb-Pb#a&E_m}N2hpMJo!I$9b)7{1& zAhk^G7Hw<`9p9bKvSOE z>yIhuEsA6@Zxay`jI?0<^&6+EK+L%l#K@(--(Xyw%VfQ6BMqKF^;Yz#y(^lt5jMZ z`Mk7yWy1w%cg50B;L~ZI#XO_#+bK_T&yaV>pJyQaePi=HDg!U*Tngn`IR&G4?chV2DgpWvJ-3J8=eG4&=Ew{6q--q(=Z2NT zWZ1r*@|X`Vfm8qsDOIV1Bjc~*~HOXXxpt&{H z>HBDc9-BB0sUIW3nXba(muI#uz0Z*xp@lu67$9AYoWTDPan%xCF!i-X1uklj(=&~~ zgXGgo+&*XlTvZF-B^2Qe>jRcq`bIFM4x{#8PNT4LnecLfmR{mpV(HbSxyh0DlAkMK zb)D^ZX^e~IDWxSTCg(P-?^<}4Ts#zkN^hTA>ybEvvw-U>ADg`TI=$`l6dQ4BXqQy* zQshW{>!6LS*v%f~qFnH3>BTPM9pJ7Mpx0-=T+AVRTUAa!Q> z3{(xV`Jp}Zhti5fALW<5P6M2a^2huQC-$sa`l`-#qTxjrI=9Kb1Yl1%q=|A4YuuSO zqM9wJeSwCsH%y~ElHTIp0ywKN*A83tFxvhLIlD=FyJ5Rh8klcXvT7P~av%L+ADnQv zv`M$_06iReU=6e0^JGD-rNco1VZv>MK}m3~L7fRsW%Az{VKRI3{Jlf&upeTj&VB2C zlXsD-1T`07x?{VGp}6lW;ZJcjP(g%1Gp*k&s`RZN8DS?z)0hmRpsFe_2pTy6Z+hQa zSQ&ieZlNq)Hl6eya=04iiVEp(hg<|yz-vZxrk#Ch<98xu?IuE{DhCzlvCZWdE zrorTg-gszTlUI^m+>vTs7!6V}Ir@J-``f!D15xH$N(LqQ`=g43qEo3D1slLACjMvI>%{ktD~klt2PFwFt8o(2CfP@aGRg^EHJsH~jlYRZg;R6HI`B zR48phU@>FPe%vbu=OIa9R#0 zxRx3oMV^MhQWyl87MC$9pl4hOj6rJ=F91Lq9$KnCWA7_v@S>cdDf^|=fYDNGUNLLd zEPhw5!imx4uhoE7Kr6)LH|&jg&n}2G`pCPoYyBu;j=3zyZ$ULDH!%{)=yuvsJuRAA zObv9(nbdnoR`4LCXtj)S%)C0D?iuL_O@jjRzCqU=3q!;raf3ySVFpLT6q>6qwOq+v zNvb+Tu5tHFhA9^o+2-*iwTxFw0sYh1)*YMI&4F5VL=cLi_Rf<)(C2!=4vmyW!E89k z=5-sVGz5gu3(IsH#D2`d=3dlmA!bj;mY#-dj@Y~q_j)*6Ba&AxebQ3SBLth6Vw_jl z~@84mzz?nHfH@x!2-bQURn3`(izR zo0nK6G1OrS#5D}le}2GiK<{w9lZ6qOgMEPC+ z&BI?%Fp&{0jPSSl5%+Gk-Luq|jZW%3z-7F@<7@rauXGdm!`NT?kr=KS<**1;pkl(f zYz%jfq2q|4Db^I9BxqsC%a1=MqdVMU@&I-y)ytX7ugy^G{J2sLyzangA%{XLuxhIG zX)<9rX1va@pLM31KrKl^$MI7kuJ7KgXeyMTa*w;6o?V14Zx`lL?x5rx(h&s~FysGq zT-c(OrgCBA9Q9lK$1ql>_WGac2MsUmPkItp^WVYria!~z1S7dJHynS~!05tMvixK< z5@cgWGKFU&;0as^2{uRPJR(eW12}cRNA}}Wfulvq(b<)-I(NfzA8mqGsF>`?zoUoq zmf`Qa(#@{M_vKKEa%EHg5Oy(MDY0cSBH(}y1gk!LPWGRuMP z9RlsU>7^}$>zze6s@_U~F^h1_(H?jC2A&zD7e|h8%>Q$Bn}3?WNnrLVFIq(cAQQZc?GPLavhfc5w$`~JljfZI;KRtS6BhjG# z`BYV{<)X=rKF?-S%JNZIpqVIosivnEqiV8og0s}3gprMj<@L{~v{AHE2Ig8KEr0YL z;GiTYRt3ePb$yZgq8UQBRei!L4!55;Mhs>sJP;Bo& zWf4jJ6|@#SXVr`|SF{@CR+H7)s+qry`jHwlfLn|~!ea@83y*Hk*WUA15n&uZ&>VZx z@Dl^2Cg02eaMy?k=|Nm;Gt!3MwwGM)ffceauFbpl`;*%(U7Ml6*ZcPcaP z9V1z8G#9&H6~2ZF8}E{ae60kF_VOYRw&L80wTZ|wWv&Rc0@{JX&^(E>JO)a%9*^y# z=;&RKjB04X-iYKGUSg?at$y~XY(uCceSeSjONISdsV7-G>xwv%!x2cNt6F@VHpmn;O@TQPQHLI=$|_iFXA5H za@gEc_X#q^^OMB^0c^7ws=w~qv=z-;ztKotsz-}GV`>N(T_~HyLe}an$7)^2v+5Q! z{eR*i2UFKlU?Wfz__ZCgGtf1bBjm_RIUiS^mI2lVSrS|8A2Ew%6z9-Jf3xsLzH`(yZvxtHeK-ZofnB$1GXCryOt<~Q*du2 z+vGb)*)$l!*#oDKQ5@y9`qbBvFQChet)8}CyMZ_7Ik-E$gO}qjO5kqSOu^@6U^;}VP5uXcm`Sh zQ^2`ml0qCPS?ke;Oxi*TFUf0lns)vQs{R z9X^siyQwE>!hclstm885BOr&uFjeif=^dDQy{w*Fg0+3yEQ^x-M_*TXBM8ZQQHeBw z|I>Gd*8YhewnDk900OOIuR9?EpS0cV%5u@~N0o zw2P~|?ZDW_ID^ssBi;gmk*NaOe_c2OV<87w-)Gc_R`|hw4JFdH1RM56PZ4zh>|g(f zpLky7Ibn2zqQG*$*FV^Wd;H-x)1WetUZwQ|#BA!Z=|fe7K85br^OyN1tf=&0jY zavN`I=gUO3BJ^=h(Jnr#rnAFM5saa41MYWUPV5O|czF;nljBhy^ zvbI~Iv8nkxi1mp4^#TM6Q8xUJy7iHK-iSa1zwu+e)5;M`^*+e;cv@jlpi_!c`duf! zxs!!j51t*JtUY%0M%52pt8nt7%&DAW{Z|9ADxnW}in`kZr(>s!nX{`}xT7bv0+Ab5 zzA9$zU@)AuLH(h!DMG(Zo>wn%JVE_@0H^A{f|808p&VZzPT$C5E}BM#XZ6o0k0+IO?H=I`g3Tz0;HecznFHjnviV#%`V#``SHESgaZBB=qiy0Nn+3M$(s`J zCqPy89gn@rs1viFYxfZ>;LN7}s6()3a?)RJQL^AnB3SOv1GsYH%coGcOdqZ44dIb5 zMB!{Fih@t`W8%}b3~C}9;WL^Qr%smGHO0R$R+5L_)_I5+NtVE>w^JC@qcbeIm05%l z1_}K3=ZuvUzC;^+i;JDUq#^8M9@2*MeyNEnWoLNT5pMZ7x_u+Qd^UpOCf=qMf;(9g zl;_g;PR%poA@zCAXFh1bU<39jS#Em)6IPl)T$o?_#;Ds?0B+n8I8N@UvjW)TS$hzmW`^%ajNq`b2DI=US%njQ%53$MKs5?dnAf*l@{O zU8WYLMsb$4>|(o?e{9g^A+5@zC`z)JV~0&VO1#Vg$P4U-)k@HSy_s@MsX(U`4wMCsju z4D*?*MgbEUS7%ocJCQy=NyxzPG+4x#4UsRMb&B7ykE6?tx|)PteJ$&6V(gqO$aGJI zKoWR{_#dj=$i6+_&@py~EviK}4WPT>?+Eh zHTAW}uhQ=;E3?0&h50>MqGfl;xkPY5HT1_IXdF>~gFPw@J*IuH@M?gHi+rSbvJg)}JT{7LBE2!|M~ zuYG!Xv$Pexs+I5^GC?pFbI;YY?BmmyRX zI_PCxvwqabtD6o{RzO~~V_y@=zP>HevN=dI> zd~}luj9YSJ2w)htWLh=KMYy0*TV*g=$&IcXQJ$9AE#^gFe2tOGYit{s^Gz(w(Qmpb zkw`x`L$RtQ$xB{ff7D)v;RPkYEbEb?QjU5RCrrILCp>as-L9&MCJT&h4J^(NCq90- z8%aB9#JohmoNV^A++2s60-jqGxj*Su{HZl1WY;NZ7jBY%xsrb=oHJq~&(SH37T6N# zVH=l(&?8d0=DD;@*>0h1MSJ@rcj;1~TY9*H9;NK;G4m~B6fLgYjrpJ%H^Iv_^4**oD zAajYJ%ap?&nBpi?klmOc$+F7fFA#2A1Fca8TP4#|vgmV$njH6o|Jji{DFg+Ew<*|% zVP)g02#8~c@I7M|&BPP%CexAtWPqlTd$rP>Jjym$Frl*k($6mu8h&D?(s0vUVv@|S zNL2shOzqx|dD|JmZP1PoOu!z_dPgzP+R6lW0G~N0DBWC zU#Rj8tG8G?3=}IL{LgBn=q_7Bn{+`6v?IbaRC>C31)Q|ob3T42#zYWs7u9g zTBM2rFeQXtBZWJIgCe{9`{L3MkzK{H>)+nGL*{ z>GkhX8XI-ay##unG(l23v~pU7;#$~@MO-4vWhwUE<|!}HO_&JA(nh^tZ}HG+K76Ug zMpINNVoWpeUkQ0mVs-?L-v}64^OM%>oViaVHu$chmVql)bk@G=yjhX#nS=O`!yG^5 z;{@`Rx@0QkVS@_nPwfj3q_Iw&S+f<1FiOg-MIvY-Xohn<_q6Rx+?g=;6*RqY{CdfN zXOY5qV*(%hgD`;0X<%R1uOstHUDry8T(V%lE$gQ4$<%Wg7?uwSzl%@E(Oo&B9Zzb6 z6(=vIVQ%AIQ!9aFPcSmDiR=D%d&ZEDkNoMSjPRPBf}fODN3_2Du2BF(bid#H+l6gJ z1b5<_y(byTn5+R!_i84y?JwE3PU+n)5Vb!6?+IBJ+8;nV>)m(-Th>lZ1Fu&ne}~x5?6;= zYK|bd%~FK?dsWw^5f%IYSG}E_cFl2Cj;d<5SaB98N)$V12XjgI|1DiUj(aoY>8U{L zi!?QB19x25ksfsya+&m7T)Y?~;23_8*iAhTwqc=g?Oo4?1VIm(=W*ED_T>X|%@=KC zqf828W&cAcM`7E0z{2TX#h6xbwOFp zE}_UbMo>f~MKlS5?W|#w5j)dm;0rLu{f$gO-zGq1B;Xd`Ay9OE!w|tmMvI_5+e$S^ z@s|KCV$s=>_c1DIOo^d^nA9@CP6sE&`ac4+>jbmAu|aORxd={(tt8T<|NJ?3UOl{? z{hrzxFKTRCr;Q~V_yPRTbg=z5>M;oo3)D#t7HbiKMSWrb6AE#oT+saD$B-mXDWQBe zZBF2iYk}gCmBkop!}!-Ls-OEft1pKBkvNs8xByKvXf^66r$6lfLV22;)gkn~O_-=c zYSq3PFDrVWK_w=5#*%IBrIrKw)+0aQ0sQ-50!AGKkrCC~3adA`1nF=G_XpVA*h3X8 zNT3J<{GN{K1c|&Ndf06WCoKC>=lm z;Q1en2e08%i1i7jTx zcTm;S(*PF8$=(cxzNWrxwdc+W-UbeshHTe2D@3$D-@| z;S*q}&fpZG11(C!s?Y(zvIMIajkJou6Jp2M8yLXa0k5nqMe3<8`Y&v`5z|al)h15XkA0*`P z5m}tJ+wZj`kIba;O5VQDlPGXA*3CyGJnqLfuKw1d@<>k0LXKzC-|$C%!@~4o5$P(!=oPd=`F>lzhgR{NWHJ7IE4Q z3XqwS#=@Y8tg|`NXw82JCZS!cAwJ}3fY3rQrZfs8zs^XTz_8#|*fbTpXV_sQE?O2i zY<=xx1K1;mMBF?E)7%ZgDav9h+%XS&F54D-3$K%rY3s+Eus|`OZld_h3eF&9&~vLR z%&*451lxdQ91mAJc-pYsW^L#vif<+AWgm9m&r&q>OeR3L!`|LcyqO5pDXeIoBpLVV zFg-m;?yq~9pP^nL!OhWp3`Zm|w%WNfM7U>B+X9W+Y!oN$p%pKM@q_2>5>Et2ZX(Cs zg4L=yY6!zWX3<;%fdjo0<_Lt5IGg&~R-sE52p%*f7(7uth@Lwd_Ibw0({Uq=vhDJ_ z#;TNxf&g-9dRfZ9mP~Q-o|)BltmeXvDYks2NBKNubr%=d`<}NxVOU~x8l7;8q=-V$ z%;L9y@>GqX*4zGSus-zWuj+53ZpW`=0_2hZShVT!3#8WAyF}?a)>k98uMb9N&l+@s z_!$?i@o;25NyW#*OsR}zqZ>}T<-;2x#%%4)lIUVD5Tz#i%adKF7xw!D*q7nmoajJMQ}Iq^B=Ldb9N2r zfI|Oz(7@GUz}sc@JLAx9fRz7ASo51(tzz&xhH-Pg4e_sw=(K^2KcKc^0f1P`V=I$r z$Kz}GXOde!p?S^I^O3vrdx?uu!$wJw{HIbehFSObTqZ##9mbG9+!j~toNJuAFdc~1 zTfbE7eBKputGYSCI_(tIHfN?6fm;UnUd-g{H1ZDK!JZdEY??kT;^I9#QZAQ0-g};; zLEv-@G+G6YrMkaUso>e4`w*NOcrTV@z@7;H!D?;2xIH0ZD|Eh@43CfQutFl`K&e^Vt;NzFt-geNZq#wM>kqbvSpfjPE^K zS2C|DY6S<|-c$Nw=$O%Ms|tq#8Y8gtR=jaL0#|JhR)mOjYZ^-C+Zp~|np#)Jhiwq< zXQxV@^_2T!lnG4;JSC0efoUn*aa+ diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_centax/manifest.jsonl b/sources/harvest-uk-ab2026-2026-10-29/uk_centax/manifest.jsonl index 383a016c6..0da39f47c 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_centax/manifest.jsonl +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_centax/manifest.jsonl @@ -3,3 +3,4 @@ {"url": "https://centax.org.uk/wp-content/uploads/2026/09/AdvaniHughsonSummers2026_Taxes-at-the-top.pdf", "title": "Taxes at the top: Understanding what high earners pay and options for reform (Advani, Hughson and Summers, September 2026)", "date": "2026-09-23", "publisher": "CenTax (Centre for the Analysis of Taxation)", "sha256": "af4cab614220c28640c0d875ab24aea263c54ec75957cfe263f6dbac171857f3", "bytes": 963857, "doc_type": "pdf", "access": "direct", "retrieved": "2026-10-29", "note": "35 pp.; PDF CreationDate 2026-09-23. Text read with pdftotext -layout; locators are printed page numbers (Key findings = p. 1). Fetched 2026-10-29T16:31Z."} {"url": "https://centax.org.uk/capital-gains-tax-should-the-uk-have-a-unicorn-relief/", "title": "Capital Gains Tax on Entrepreneurs: Should the UK have a ‘Unicorn Relief’? (landing page)", "date": "2026-10-08", "publisher": "CenTax (Centre for the Analysis of Taxation)", "sha256": "91a6c8dc749273b1c6b6e330a2acaef4010a379d5e82e1bab22c4a820c5cb65a", "bytes": 267896, "doc_type": "html", "access": "direct", "retrieved": "2026-10-29", "note": "Linked from the reaction page's related-publications panel; 0 rows, the landing page prints no figures; links the PDF. Fetched 2026-10-29T16:30Z."} {"url": "https://centax.org.uk/wp-content/uploads/2026/10/AdvaniFlewGazmuri-BarkerSummers2026_CGT-on-Entrepreneurs-Unicorn-Relief.pdf", "title": "Capital Gains Tax on Entrepreneurs: Should the UK have a ‘Unicorn Relief’? (Advani, Flew, Gazmuri-Barker and Summers, October 2026)", "date": "2026-10-08", "publisher": "CenTax (Centre for the Analysis of Taxation)", "sha256": "9fd8ce97f3343f7aee1cab11dede3f374ba473b8db4056268a0becc154e7a71b", "bytes": 568966, "doc_type": "pdf", "access": "direct", "retrieved": "2026-10-29", "note": "74 pp.; PDF CreationDate 2026-10-08. Text read with pdftotext -layout; locators are printed page numbers. Fetched 2026-10-29T16:31Z."} +{"url": "https://centax.org.uk/autumn-budget-2026-reaction/", "title": "Autumn Budget 2026: Initial reaction (4 November version, finished updating)", "date": "2026-10-28", "publisher": "CenTax (Centre for the Analysis of Taxation)", "sha256": "5103e06ae3a5831ac866ded44b6aa68187adae2b2fa5dc8b68700938a02b3446", "bytes": 271039, "doc_type": "html", "access": "direct", "retrieved": "2026-11-04", "note": "The reaction page after it finished updating ('This page has finished updating'), hashed by review B on 4 November and matched by the coordinator's own fetch that day. It adds seven sections the 29 October copy lacks: Big picture; Tax reliefs for founders and employees (CGT, BADR, Founders' Relief, EMI2/CSOP and evaluation parts); Company distributions framework. 2 rows cite it; the 3 earlier rows keep citing the 29 October copy, whose quotes this version still prints."} From 5c3fc6dd405dee8bfa9612b2fa406067ce89ef6b Mon Sep 17 00:00:00 2001 From: Claude Date: Wed, 4 Nov 2026 18:25:58 +0000 Subject: [PATCH 26/52] uk_ippr: apply review B's IPPR 1 and 2 and K-7(f) Row 0 keeps IPPR's unsigned 0.2 with an as-worded convention; NOTES no longer cites dropped red book rows or reads dateModified as an edit. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_01LwQq6JvunwtJCAckL6L4aM --- .../uk_ippr/NOTES.md | 14 ++++++++++---- .../uk_ippr/claims_staged.jsonl.gz | Bin 1588 -> 1607 bytes 2 files changed, 10 insertions(+), 4 deletions(-) diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_ippr/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_ippr/NOTES.md index dcbce2ee0..7930900a0 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_ippr/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_ippr/NOTES.md @@ -19,7 +19,7 @@ The listing shows nothing on 29 October. The source slug is `ippr`, as in the AB - **The listing:** the media-office listing (https://www.ippr.org/media-office, pages 1 and 2) was fetched on 2026-10-29 to find the responses. `/news-and-media` returns 404, and `/articles` lists no Budget piece. - **The releases:** all three were fetched directly (`curl -sSL`, browser User-Agent, HTTP 200), "Unfinished business" at 16:17Z and the North and Scotland releases at 16:35Z. -- **Edits after publication:** each page's JSON-LD shows a `dateModified` on 29 October (16:04Z, 16:21Z, 16:34Z), after its 28 October `datePublished`. The copies hashed are the fetches above. Nothing came from a local copy. +- **`dateModified` is a render time, not an edit:** each page's JSON-LD showed a `dateModified` on 29 October (16:04Z, 16:21Z, 16:34Z), after its 28 October `datePublished`. Review B's fetches on 4 November show 14:39–14:46Z that day with byte-identical length, and two fetches of the same page 30 minutes apart differ only in that field and Cloudflare's email tokens. So the field records when the page was served, and no edit after publication is shown. The copies hashed are the fetches above. Nothing came from a local copy. ## Rows by primary and measure key @@ -36,7 +36,7 @@ The listing shows nothing on 29 October. The source slug is `ippr`, as in the AB ## The three rows 1. **"It will lower inflation by at least 0.2 percentage points"** (Carsten Jung, on "the cost of living support announced today"). - - Staged as `cpi_inflation_effect`, −0.2, `range_high`: "lower by at least 0.2" makes the signed −0.2 an upper bound. Context, with no proposed key: IPPR's words do not define the scope (changed on 3 November, thinktanks-a review finding 1). + - Staged as `cpi_inflation_effect`, 0.2, `range_low`, with `sign_convention: "as worded: positive = reduction in CPI inflation"`: IPPR prints the magnitude unsigned, README rule 10 keeps the producer's sign, and "at least" makes it a lower bound (changed on 4 November, review B finding 1). Context, with no proposed key: IPPR's words do not define the scope (changed on 3 November, thinktanks-a review finding 1). - IPPR names no source and no period. The OBR's EFO paragraph 3.10 says the fuel duty changes and the temporary VAT cut on domestic electricity (Table 6.1 lines 2 and 3) "reduce CPI inflation by 0.22 percentage points in the second quarter of 2027". So the row is `restated` with `parse_confidence: medium`, at `period` 2027 (`point_in_time`). 2. **"the £18 billion annual taxpayer losses at the Bank of England"** (Carsten Jung). - Staged as `apf_indemnity_losses`, the AB2025 uk_ippr proposal, which staged IPPR's "£22 billion" the same way. `restated`, source not stated. @@ -52,8 +52,8 @@ The listing shows nothing on 29 October. The source slug is `ippr`, as in the AB ## Not staged, and why - **Design, not estimates** (README rule 8: spending allocations, financing remits and payment amounts are a measure's design): - - IPPR North's "£285million of good growth funds" is HM Treasury's allocation ("£285 million in 2027-28, comprising £165 million to support places with a mayor … to establish Good Growth Funds and £120 million of incubator capital funding for areas transitioning towards a devolution deal"), restated. It is a spending allocation, and part of Table 6.1 line 20, whose costing (−£340m in 2027-28) also includes the subsidy on the £1.5bn finance. The uk_hmt_redbook family stages HM Treasury's own £285m as a row keyed to `ab2026__good_growth_funds`, which rule 8 may also exclude. - - IPPR North's "£1.5billion borrowing facilities for building infrastructure" is the size of HM Treasury's lending facility for Mayoral Strategic Authorities, a financing remit (uk_hmt_redbook stages it as `policy_parameter`). + - IPPR North's "£285million of good growth funds" is HM Treasury's allocation ("£285 million in 2027-28, comprising £165 million to support places with a mayor … to establish Good Growth Funds and £120 million of incubator capital funding for areas transitioning towards a devolution deal"), restated. It is a spending allocation, and part of Table 6.1 line 20, whose costing (−£340m in 2027-28) also includes the subsidy on the £1.5bn finance. The reworked uk_hmt_redbook family does not stage HM Treasury's £285m either (rule 8). + - IPPR North's "£1.5billion borrowing facilities for building infrastructure" is the size of HM Treasury's lending facility for Mayoral Strategic Authorities, a financing remit; the reworked uk_hmt_redbook family stages no `policy_parameter` row and does not stage it either. - IPPR Scotland's "Bill support of £100 for households in lower council tax bands" is the Household Energy Payment's amount. - **IPPR Scotland** prints no other number, so it gives 0 rows; it is manifested for the tally. - **Pre-Budget IPPR releases about this Budget,** outside the assigned "responses of 28–29 October". Each was fetched and hashed on 2026-10-29, and each is a candidate for a follow-up if the family is to include pre-Budget options, as AB2025's did: @@ -80,3 +80,9 @@ The thinktanks-a review (`harvest-staging/ab2026/reviews/verify-thinktanks-a.md` - **Finding 1:** row 0 (−0.2pp CPI) is context. Jung says only "The cost of living support announced today", and Quilter-Pinner "action to support families with the cost of living". The lines 2 and 3 scope came from matching the OBR's paragraph 3.10, which is our inference, and README rule 3 says the scope must be the producer's. So the proposed key `ab2026__package_cost_of_living_support` is withdrawn, the reform hint and note stay, and `value_kind` is `range_high`. - **Verification:** no primary was re-opened (the session's network policy refused ippr.org); the contract checker reports 3 rows, 0 problems, and a re-run of `stage.py` reproduces both files byte for byte. +## Review fixes, 4 November 2026 + +Review B (PR #231, issue comment 6635498758) approved this family with fixes; review A's K-7(f) asked for the same NOTES correction as its second item. This session re-opened no primary. + +- **Finding 1 (coordinator's call: take the fix):** row 0 stores 0.2, `range_low`, `sign_convention: "as worded: positive = reduction in CPI inflation"`, as the OBR and red book CPI rows keep the magnitude (`uk_obr_tables` row 2653). +- **Finding 2 and K-7(f):** "Not staged" no longer says the red book stages HM Treasury's £285m or the £1.5bn as `policy_parameter`; the reworked red book stages neither. The access recipe reads JSON-LD `dateModified` as a render time, not an edit. diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_ippr/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_ippr/claims_staged.jsonl.gz index afa3134524389a26273ce21c8928d96817bea8b5..0e1df28c107a9008e8d3d7d60c0ab8c99ff72654 100644 GIT binary patch literal 1607 zcmV-N2DtejiwFP!00002|Ls@JZrer>zV|5xy2u7hTcRXOve5!bngDHqCJmBai6wVP zEktse-K7-P2#~jEd-j{SZ;~hJ%us&uZBh8~x7p7R4NlSLuSGD>n`E!# zP>KTOB^1IX3$)-C2u~kA?^~!)ECDH!#KF*v8L(m6p%~eaW90N)XQbKJn?W3F!3h)Ov5|&;o_64Mt1cD@=h@ z%!J(}m5R1OL>s#kc5e!itKUSIP(Rme0%*Gz;)X-X>3GL59ir5S;5+~1s1RAHTrCl{ zuv|jFv-bPi(_>wi^0@Wi!tlehXU}~dEf6AB*W~0BUhLAPT6rS)f`JuHTj#XK9Axbp zNEV{Z$Xt}tKXF)8+6dFYuMb{8Qahkzq$UqeC4e~rK-egeG$2=yE0qodJNXeEr6)3>H^3pRgxMD zPZ=l5-(s{|WX#%~{;6jWt+kBlRdq}q+ST-`Cz1razBHT4OVJ|Ftq=?cK{(^pTqj*; zy{f4;kQ!aUBl7OBgdgiNJAqWuuU$oa07g;&H<4Y4aB|Mp>1;HPQ3yGm4Z`s>l*uFw z@jRW(r=uB;()b*4J{d_l9!;j{G)=9oXdVwy49C4L4ZA?oF@zVLa=8!u-P=E2 z5*#J(kPz3FYiHlR{gVsLpzDGPZ7?N@gGd>C%e;e!USdvK5_QMcCg(#J&Q}aEY9^BL$e=o|*E!@xTXyu+Z?w1E8ue5h|&53f(mQQ`9( zKDqHY?ABYfP_2kE;Sx;-M6+D2RGERTE9wJBlrAZlgrIU-ipJ}iZ!@;@QEgXB7$lV^g=fDPC8iL31#v`oy4GF|`ACz+b<^5a;`2;*l;LzUHo2Az4!(n?r#Pz4Y!MJS3kJsFx9w z8d6cHyy-&ZZl`I|zquv>@7#15cR+0Ncqqyx_tE1r%V{3;+vI3O!CEw*Bg#e-=qRJz zR2EfA?3Y*&?VG4|-}}e)doz*#7w|76;6h5$jP^cqb3L4GiPbjDNAl}X&2NeCXQ(ta zO|+A;&(Y_NU~wnkrOsJg?d$PyHt7BiblJ=8fhO;xb(a+Db@_Fw@LtVYDc;^;`R3@q z{7gf4aRUACaYc3V6WuMS^b{1yuCW*J#0Nr?i03{Po~WMDHd@MzUb{sfjwtIT^@eX^ z=8QqU58!d{ww~#6^Y>W@EN^v65N{~N?yzuj>X F004$v9%29h literal 1588 zcmV-42Fv*$iwFP!00002|Ls@XZrer>eeYKc^dcKDZHbaB$wmvrX#%tbnlwoIN-ViU zY9W%#>@KCSMu7Z9+h_lY`zQI5&MZYqjvVBI6itE353-g!JC}3j%<{S?oHHt}ozi9R z0(u6m6AoqX1bVTsn5z<_O6MwR=y6%+IX}QMSro!7qa+vBvYk|EiUvy;6=7}z) z;&XY!buU#m5qX3w+Qb&3c6CukaV;}+Q8)k9i_(YSI|%sEQ6aKYxmw~KljGJ=o9_=# zpFZ>WSfJFcuIaQ>c)rVvFz^6-Ai#>I#V@Tf2U)uYl7%QUG8d)vPaGDNHo`RU>x1Wz z)D9?WsmX&=31Ci{QV|qc9z7Rvj&OcD1g5Qp&>u2727|Ir@GexG`Fm(U*^y^uS`>&B z9EU|FG`y^Zap=)v{C0R+=z0g%$}NB>s_FvBC{>ag22Z(8l)uGjx5zbXclxKEDzw&8 zqgT~2bzTSFbx$M-cKv2H5KGZko+%y-2SGUF)m$fC$Gofw8%T{V;30W;Si+BWnVmqY z=+_PrAAnKR*G*&>BAlGFbvhf3V-!M8XM=D&4P`P(Lp)C>^XX`YqclE8oKHqljz^Pe zI!%*sG{(tjKAOiv6vJ_^OT$vLbPVA|r(Es>fBWW-7gUatcSwk9%h=hsZ~kPUx#+qe zpbe(PWe_PBpLj)3TzTm@X-U)_TbrB@T{u_Si?g%!dVQ*@%A9JGopGT>pwm<(cvf$H z2A$79mqF(+XdMRLVc;DGtyTr>FW^IcXZrH`#2f*i=kUoL#$mVKqJ?V3kqMV*G9a4e zYNg5yY+X?wIHGh(!6Z~Fr=@7TPWd)tJ0I0{g?N^e#_j}Tv)oEFZP&3OvCt9u#)RVA zGj-fcdzQrP+A~itTl5NyXh102sv^Zr0u=GGM}c&LS)O3!^V#Jb?Q&j;yk-|HF5S;) zsY>Yw5vxGN8oS*Z8L{R^_ZRW?H`ia=FiV?lV}{d@nr?0Jfrs2iOLVI~6=nM2k(6s| zc_8XmImN3=G-wWnQlB_fi@bly8trG#;c)Vjmr!&>J-?+( zFvR)(pm?ZDhp)IVR!G(q$mUR>LodDi2Up~C0`)RtT0<%dl{a08-0w6``nNYE;hmc< zel1 z|0VnjNw|=bG^4$d+};dlTVk~h^O5{I)bl&?`x!b-O%v^;>~j=)BUs$Y_vv#MSNn=Q zoDI6qfiQcyJ<#O+YTYNrdR=~>7Ikk9{CCiFoo8@I>{Dw$V~%^x7@@a70-zsW*HRGiMC)jR22(clBgnS9>Dns3TqZ zTRrJZLf2lx{Z%xn5^iwXDmkbX**YB+qS3~FcuW|sSdFm2T=Ml~XZP;A%fB=nc4a4t zC+YxS_UUyGxywWeZABo)MsMB{S|RE4D;yVzI~C*(FIp>9-qx$1d7ya>0^8W@}pFWvO5&!@WRR5v? From c6c19c4804a8a1761e99603679d8c9f11f3ffe53 Mon Sep 17 00:00:00 2001 From: Claude Date: Wed, 4 Nov 2026 18:25:58 +0000 Subject: [PATCH 27/52] uk_devolved_govts: apply review B's devolved 1 and 2 The Welsh rows are Welsh Government parts of the shared Barnett key (the red book's row 258 is the total); row 0 is dated 2029-30; locators, source_column and the Scottish manifest note corrected. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_01LwQq6JvunwtJCAckL6L4aM --- .../uk_devolved_govts/NOTES.md | 31 +++++++++++------- .../uk_devolved_govts/claims_staged.jsonl.gz | Bin 2566 -> 2528 bytes .../uk_devolved_govts/manifest.jsonl | 2 +- 3 files changed, 21 insertions(+), 12 deletions(-) diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_devolved_govts/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_devolved_govts/NOTES.md index 4e1b51211..5e5886de7 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_devolved_govts/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_devolved_govts/NOTES.md @@ -6,7 +6,7 @@ ## Changes on 3 November 2026 (thinktanks-b review) -- **Finding 2:** the Welsh Table 1 Resource and Capital cells (rows 1–7) carry `conditions.row_type: "measure_component"` with their `component`; the four Total cells and the £448m headline (rows 0 and 8–11) carry `row_type: "total"` and no `component`. An ingest can no longer add parts to their total. +- **Finding 2:** the Welsh Table 1 Resource and Capital cells (rows 1–7) carry `conditions.row_type: "measure_component"` with their `component`; the four Total cells and the £448m headline (rows 0 and 8–11) carried `row_type: "total"` and no `component` (superseded on 4 November, above: all twelve are now parts of the three-government package). An ingest can no longer add parts to their total. - **Finding 3:** the Welsh package key names its scope, not the nation: `ab2026__package_barnett_consequentials`, shared with HM Treasury's figures for the same scope in `uk_hmt_redbook`. Nation is `conditions.geography`. - **Finding 4:** the seven real-terms settlement rows (Welsh rows 12–13, Scottish rows 14–18) use `proposed_metric: devolved_funding_real_terms_change`. AB2025's `block_grant_effect` means a policy's effect on the grant, positive for an increase; these rows are a year-on-year real-terms change in the whole settlement, positive for a reduction. - **Finding 7:** the Scottish Government's rows moved here (above). Both slugs are now in the harvest README's slug table. @@ -15,22 +15,29 @@ No primary was re-opened on 3 November: the session's network policy refused gov.wales and gov.scot. Every change relabels fields; no printed value changed. +## Changes on 4 November 2026 (review B) + +Review B (PR #231, issue comment 6635498758) approved this family with fixes; the coordinator's brief (section 3.3) settled the Barnett key with review A's K-7(b). The fixes sit in `review_fixes()` in `stage.py`. No primary was re-opened, and no value changed. + +- **Finding 1 (Barnett):** the shared key takes the red book's scope. HM Treasury's £1.3bn (red book row 258) is the total, and each government is a part (rule 9: "a part can be ... a nation"). So rows 0 and 8–11 are `measure_component` with `component: "Welsh Government"`; rows 1–7 name `"Welsh Government: Resource"` or `"Welsh Government: Capital"`; row 0 has `fy: 2029-30`. The proposed-key scope below is the red book's. +- **Finding 2:** row 13's locator says "third sentence", and its note no longer repeats row 12's "next financial year" sentence. Rows 14–18's `source_column` is null, as on rows 12–13: the labels were ours, not printed. The Scottish manifest note no longer names the old `uk_fai_sfc` folder. The sentence on the Welsh package mapping being "unresolved" is corrected. + ## Coverage | Primary | Position | Rows | Coverage | |---|---|---:|---| -| Welsh written statement | Paragraph before Table 1 | 1 | £448m cumulative headline (`total`) | -| | Table 1, Resource row | 3 | 199 / 165 / 146 (£m); 2029-30 is N/A (components) | -| | Table 1, Capital row | 4 | -27 / -19 / 2 / -18 (£m) (components) | -| | Table 1, Total row | 4 | 172 / 146 / 148 / -18 (£m) (`total`) | +| Welsh written statement | Paragraph before Table 1 | 1 | £448m cumulative headline (component "Welsh Government") | +| | Table 1, Resource row | 3 | 199 / 165 / 146 (£m); 2029-30 is N/A (component "Welsh Government: Resource") | +| | Table 1, Capital row | 4 | -27 / -19 / 2 / -18 (£m) (component "Welsh Government: Capital") | +| | Table 1, Total row | 4 | 172 / 146 / 148 / -18 (£m) (component "Welsh Government") | | | Paragraphs after Table 1 | 2 | 0.5% resource and almost 10% capital real-terms reductions (context) | | Scottish Government news release | Opening paragraph and paragraph 2 | 5 | £650 million; 1.3% and £550 million (resource); 1.5% and £100 million (capital), all 2027-28 real-terms reductions (context) | -By `measure_key`: **null 19**. Twelve rows carry the proposed key `ab2026__package_barnett_consequentials` (5 `total`, 7 `measure_component`); the seven real-terms rows are context with no proposed key. By metric: `block_grant_effect` 12, `devolved_funding_real_terms_change` 7. By unit: `gbp` 15, `percent` 4. Attribution: `own` 19; benchmark `different_model` 19. +By `measure_key`: **null 19**. Twelve rows carry the proposed key `ab2026__package_barnett_consequentials`, all `measure_component` (Wales is a part of the three-government package); the seven real-terms rows are context with no proposed key. By metric: `block_grant_effect` 12, `devolved_funding_real_terms_change` 7. By unit: `gbp` 15, `percent` 4. Attribution: `own` 19; benchmark `different_model` 19. ## Proposed measure key -`ab2026__package_barnett_consequentials` (12 rows): the change in a devolved government's block grant from UK Government decisions at AB2026, by resource and capital, as the producer prints it; nation in `conditions.geography`. The Welsh rows exclude employer-pension-contribution savings ("Changes to the Welsh Government Settlement"; "an additional £448m of funding over the current spending review period", 2026-27 to 2029-30), as HM Treasury's do. No Table 6.1 line covers it: the consequentials follow from England's spending lines. +`ab2026__package_barnett_consequentials` (12 rows here): the scope `uk_hmt_redbook` gives it, the Barnett consequentials of "decisions confirmed at Budget" for the Scottish, Welsh and Northern Ireland governments together, 2026-27 to 2029-30, excluding the technical reduction from lower employer pension contributions (red book paragraph 4.150). It crosses Table 6.1's England spending lines and has no line of its own. HM Treasury's £1.3bn three-government figure (`uk_hmt_redbook` row 258) is the key's total, and each government's figure is a component. So every Welsh row here is `row_type: measure_component`: the £448m ("an additional £448m of funding over the current spending review period", `fy` 2029-30, cumulative) and the four Table 1 Totals carry `component: "Welsh Government"`, and the Resource and Capital cells `"Welsh Government: Resource"` and `"Welsh Government: Capital"`. The Welsh rows exclude employer-pension-contribution savings, as HM Treasury's do. Compare the Welsh £448m with HM Treasury's £450m for Wales (red book row 260), component with component; never add either to the £1.3bn. ## Welsh Government @@ -67,8 +74,8 @@ These are the Welsh Government's own published settlement accounting and real-te funding. No totals or differences are derived. - Table columns are fiscal years; `period` is the end year and `fy` retains the printed year. The £448m headline is cumulative over the four displayed - years, `period: 2030`, with that horizon stated explicitly; it is not an - annual 2029-30 amount. + years, `period: 2030` and `fy: 2029-30` (the horizon's final year), with + that horizon stated explicitly; it is not an annual 2029-30 amount. - Resource real-terms reduction “next financial year” is 2027-28 given the publication date. Capital reduction is explicitly by 2029-30. Percentage values retain the printed magnitude; conditions say positive means a @@ -129,6 +136,7 @@ The HTML's JSON-LD `dateModified` is `2026-10-29T08:22:59Z`, while the visible p ## Verification +- **4 November:** the contract checker reports 19 rows, 0 problems, and a re-run of `stage.py` reproduces both files byte for byte. - **3 November:** the contract checker reports 19 rows, 0 problems; a re-run of `stage.py` reproduces `manifest.jsonl` and `claims_staged.jsonl.gz` byte for byte (COUNTS.json pins the gzip's sha256). The fixes relabel fields only, so no spot check was repeated; the lanes' checks below stand. - **Welsh rows (29 October).** Ten rows selected with `random.Random(2302026).sample(range(14), 10)` were checked against the saved, hash-verified HTML, reselecting the table @@ -150,8 +158,9 @@ confirmed the number, sign/magnitude convention, unit and end-year period. The headline and all remaining numeric cells were also read directly when transcribing. Every row's quote contains its raw value. Scope review found -no assigned registry key: the proposed Welsh package mapping remains -explicitly unresolved, and the context rows carry no proposal. +no assigned registry key. Since 3 November the Welsh package rows carry the +shared proposed key `ab2026__package_barnett_consequentials` (scope above), +and the context rows carry no proposal. - **Scottish rows (29 October, in `uk_fai_sfc`).** Because the family contains fewer than 10 rows, all 5 were checked. A reproducibly shuffled order (`random.Random(20261029)`) gave rows **3, 2, 1, 4, 5** (now rows 17, 16, 15, 18, 19 of this family, one-based). The saved primary was re-opened at every locator; its hash was also checked against each row's publication hash. | Row | Primary locator | Verbatim figure | Checked normalization | Sign and period | diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_devolved_govts/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_devolved_govts/claims_staged.jsonl.gz index 0cf67caeb647877e068d2758e4ce1f33b613c3e9..5618000aabfc7f14eb36af05fb42f1a0ccc340ed 100644 GIT binary patch delta 2502 zcmV;%2|4zL6yOt(907`v9UOn=Ee3ntH|f5~KFOX#N|qEmnYeXRJ7y7}i9}L|=kL=Y zW%+%F=DFmvBIhFQyd)j5OgmlDnNh`Jo-#iZF%MR7CN1I^Pp~vtBvfAeK}?mxE+LO1 zCRv*Mi3nMY-2;(^9EZdP@8QoR6DhR7=0CAmEy!ze!(^J^F=)?z3F3c3l_Ibti^Ymb zlCczz5g2VPBT@8*r*nn9BCY}&`|PG1nHG7Gr2ec3=PdW#fjic%Q2Gx1HSL`b@W7mj zxun@bAODHQOb@mYlK*N1z@9zVBe@`DH}c|&&SFNK@3CpZf(1>vO0e-Il_|?}6jHHw z1&G3FtjHO`Hgl$M6f=Ki1Y#6%jmad#1W)e`uQqj*sB_( zwiYx{e4b)I)NyGaxFp#2hLJxJo(3SVV(<&OZC4T%-~{Qs^PzuRk8K1jeJK?(p)voJ zRwIA~$Xy*U^P8>ESy+4!Jmv1fzib1F%ptTef(b2tgw#}cOE6`ykBsIfW_I5}^i4 z%XEo$XPi5EkS~AtiS|MeJfkuvBGMPHj7v%@R>zG{(0IG&^|fPAsnC#V#6t*Jt)~T_ zFZ4Mkxd`EiJ2)S-od_fgu&LIQul3|J2=2gmY%zmeP#NoehXw(wGJq`Ggwz;ChmbSq zxf47J!_izwJn@D~s-utaBw-W_8PD!QBNR^Tb8EcE=d8S^T zoh_Hk{#<|D^q1%bXG?RXo~o|YTjGU!^Yw*#)rES+bpePPOC7x=56{jQ>-Z<& zt`~o#y8ktamc01%G5Ovy_zM{9Os5Bk&t@LpiJ_n4W1T%i7gTx&bTh%tVX~i!GV9k_LWI)b`{yO*0T2WR*}o_w~NSM1+`_a zvx~_4LoLX?6KZjuCyH9ThHO!5*N_huwRQ=4uc&pVb_sdUsC9PRuSFA_C~EBrvPG?3 zK|Wa2+6Cl&qSm#`$NL4XyAx*JQ-$pP9|p3JwUD)+@cD+2)gQXEklkB7J2_CyR%w6u z2kZXDwkZVNY*!_Da~T*{}8#l;_&5**!l621Td`D!;|_siFb-G<#aUne{A_0-mf?GCKv ztKES;P`=vzSG)ge_g_Dgul1A!#4LYRS{maVc(nOS3giIjS{Z3-dXX3~8efq5JVC_( zymH8V891McS0%CbzIJ>In@CHdX?a2Psalj;fOiE@CgLWF+Yq@QZwZ})rHfGK-ZfMj zgsxsxDp^KNmEDlti1>d74wY)~QRBYWxNkJRH&XsZzh^jGzmu3oaGB~JL#siG z()}VKa`c6+)^%)OPP`8=4*H{KWk5m@)rkgNYs0Cxyn(p&0^mb`YjBy1nh=BOKM-H2 zs6*jb<5Asjgs=Nc$1lfMoCpQC#5E9R44jlWgJ+g!U!7-<&Y`<1v`2sB&|8Kfco&Ge zpja+Li=2HRr(9Q9hvAZgm(Lh6L60%M6-cvHRTe)%4mq1ultlmbt7vJlsMge>RaDJ| zBVtagHl~_bC{y!BKCZBZ!P!nVRc0)MZ(oJi+d$+oh;HJ#Tp%;PfOnsNz%u$kkkl+yYrh>p!!Sho~N!*4zrfi}g2P zMC6A{v{bz~*(X;Z43!Xtkbpx0V8z}kP0+`HU3{*K00tu@=qFu5o206|A1O!6yN6ra~dX8cl9p0X?!Z@5hX#Xz7^zKF)xC)w|rEICr0Hh^Bt8sfe zx?tR1E_$jQSM|z4&u)F`j80tS)5q~>uucNtag7Z42Lx5+#E;`EV5;}yK6Ev*rI>@G z=%wDmd5v2WM@`SxD&#Q|bX&P+*=gDNnC!f4F_XBrznp)qTW{Mow|-04#XZ#)IVYB! z_mWfl9m8_d>7PG4eo$_!Hv8ufeJBQ1VCiUUOIur7J{~O}al!>H=%2%z`SX)loa?;|`;DT`S3^ QHa>j!f7jr>`6h}0016b%-2eap delta 2540 zcmV@jZ zCRvhsu?Sd%m-`|KIChB*-gnqJaNSZED#?CJ^V*1*pko@;zJZ#t-x+D_>KF)gQ z1M=dE&Lc*gAF*l7{3T7eim~w}l?lr-lu5C7IgHF{q{tb;HVdY(mtR<th{WUGQ27i8P&E>e34*VRHHEg3u=;nMcCECSY)sPKucP^DlLQ z1Lkw|DbRLCvLz#z+F80}zJ@N9$feH+O`uzva?nq=CJ}1DSffj{$Ku?{z1+1=wC91~ zDU}%!p&q<6C^3JQD~}gKLF3)t*4NHJrDABdkOvTQT2D*9Sn6|3auL9h;oyAGbR3Y( z$EI3Op4OAcAUyrTOY=EIb7`#i9qRk6jA3co1hg=U4jX6C8&2^kbVoBG@x&V{sg5|p zW*fE@p$N;Pl9O}24i>X z26JcX3|)6HVxx&0%+ANN85_HU>162AU_J?`8~7n}T;KIKQ&LMNcR3O^$T)*O-h7Rj zNIg}K)LVbyK)uCgpk6spFTd^;d97FEwVoaU1MXdJ!h=HG1VuWf5&86SG@8cfN+vo9 z_Nt}zb18Lt0aw*$B)j2kMGAk>ZY9x7vA5(!N_|yD7k03vkGGp72gxdm#0q=;tYV57 z!@l*Ly+y+q#aVFc2rjMCh&f!~17{2hT!J%mzjl97hOSk+29%p?-xb$|o#R>pHtbk4 zDz&*1uJfhOQa$g*JjKH|G|I6{(C9$_E01~CJPC9n&*MCzm@7B!JxNzZszXC`KK$_h zQ5for7(B+H@E<0ZVY`QfYq+iv^lsXoC_>9lD%D$@o;FU0?xW+>=;$T6e=fctW&9Ix z*9(7A-r*K$mAv@$G5*mK_Zx`o%w`7&%X%K(iIW1W@}3o5+>v8mw2a1FYm37!U8 zU6T_mS}j^n8LiVJMJvXqMQdwlos16@tz{lQKD188`=J#~@HEgm?G0xZtro4Pj@H?c zq7~!QqO~=&jwT0+)-n$tA6iF~{m=>~_%45FeO06~yB=)GYRUTa)`Ly>+vVT~0jtS% zb~(5`u!7usffeWZB*1D{gDqI?YVg5=)h-6N3RY)k7lT^{tFzxeE1KX*fYq)ATd>-- z;DZILT?%d!tV6pH+%8s!dqH*hG?0qlrLjo0NVVUwIY^}Huew>JwpOc74-~3p8a{u* zT6MZ-tvY=)u-aWz`^8iX&C}mGy%Kic^uZoTa))-`wB-kq-2Ez1G{H$B^GWWHT4dTi z(L+V1-4T5n&+s1a*JBR&8D6{dX_0A>dHTDZO=Q}=&UTSGwY!~dBXhbZGEZz__DSw+ zT4dVY%tJ+{-OIFlnRYMpE0I}ENkD(hQl+ISoa6Ru8StqV(6uV0sfa})KxcG8s`CUD z1Hj54^QGW?Dq0&#wZ5u*3zbMnq6v9H^r=deN`QAcP$uL>6gMIAdA!AR0hZ1KoqIP> zsZqIn?WSZw?&}Wj4?D=iK5DZP=z>5X0V^P7LvX69*KSP`$ubJw&_LiTfC_*64N^`` zv;uvICG`FfgxI?E=xROTX)t1>Zz_sxzlECGJ4elw`{eI}n28d&6WP$JJO-18HXs!V zc|sE(*i_gC8USI-)xrg6Iw3W*5cq@RKl_=OgD%iL`s7VzTFu{Eo=0IkZSoWd4u%Fz zCJd4UT+bD#0Ikb{Hr8JQb|ZhH8#rXt;GxDnt#Pl=cx$BmlYVP%x_QGc4Plt_9z&}^ z3gdpE5E=SHSL-^mui9+`jD!C8nF&Y;qAJmVYi&98j`!&{UT1sg@4z*=s17lh{yp)9 ziaHd2)gF8OPWbwKx%ln)iejPQmbe1KoPm=PXYkDO%<^pCO79VQ_LhHPDBlINF4$3@ zx9nIuGd~H>5Z+($PAbJtkW~Ht01-y;@0vu7(H8z&A?u|S% zdz>Q*l<|jKvfK#K5CMNRTO)6Y*&6(I$W>p`+(B!|>%TGsR#Xp7DggT6#rlUKLh{ok zTB=&*?2{`HhDwM+$iRUBLgUp5jnT&df_$N?2?jmH=qFu5>!>QdA8|>|?aRha4W5Du z?_w$f4I%_=2tXrmN}ZLfsvF%Y4aM?ip$i!;TN-7SHZ53$pWuIgA%nt|t02SqB0h_0 z!c-_E*g}&#mr{?cUi>*_30itCf^|jN?gprk|G}QR$@8lMREAKHjzH&9Jtr^zeD$}# z5^dD;k$cC&YIq(Doas3mxr5MkocUxrbl7M%o-k)H^~ZtZ4xIr#pAFfBPN0>;d_Nq{ zTz@)n>4a`Gc!z(-H!!>{X#p)i&yc!hX8GU0{w0%*4O67;4tPtH^I8j_>`8Ddj6? z@4MXDP}srI^-}K7PGq5unxEhO0=g8R`4tbXTp~KfmweHmE4eP+OVW%G6Ty zXeo*lCNMn?xedKtN5m#&muj>u{hI1@Tr0L7mZ-J;om&Pygfex!cbPgqyfXDwW8K#l zsJ1}0XtiKyt)unUHfndXRb-3OuLLPnW8-J_ZPG>rx5U%U3!9FmKpo$^Ks~spgHC~- zwluZ;wETP=es0yK{w7)5()3$OQ)loHrRffo9(HBw4EC)|>x%pE!~X%S0;06Ji2wk7 CM&(QZ diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_devolved_govts/manifest.jsonl b/sources/harvest-uk-ab2026-2026-10-29/uk_devolved_govts/manifest.jsonl index 99d59358a..f5338cbcc 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_devolved_govts/manifest.jsonl +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_devolved_govts/manifest.jsonl @@ -1,2 +1,2 @@ {"access": "direct", "bytes": 26908, "date": "2026-10-28", "doc_type": "html", "note": "Fresh curl -sSL fetch with the assigned Safari User-Agent; HTTP 200, completed 2026-10-29T15:22:18Z (HTTP Date header). SHA256 is of the unaltered HTML bytes saved as .ab2026-primaries/uk_devolved_govts/welsh-response.html. Publication and last-updated dates both print 28 October 2026. No assignment local copy was supplied.", "publisher": "Welsh Government", "retrieved": "2026-10-29", "sha256": "87490552c36d8ffd5787db18127704e467d93599e5708627adb6da7dcfe17c7c", "title": "Written Statement: Welsh Government response to the UK Autumn Budget 2026", "url": "https://www.gov.wales/written-statement-welsh-government-response-uk-autumn-budget-2026"} -{"access": "direct", "bytes": 27662, "date": "2026-10-28", "doc_type": "html", "note": "Fresh exact prescribed browser-UA curl fetched HTTP 200 at 2026-10-29T15:22:47Z. SHA256 covers the fetched HTML bytes. Visible publication timestamp is 28 October 2026 15:45. JSON-LD dateModified is 2026-10-29T08:22:59Z, while dc.date.modified repeats 2026-10-28 15:45; no earlier copy was supplied, so a numerical change cannot be established. Five rows, read from opening body paragraphs 1–2. Local file .ab2026-primaries/uk_fai_sfc/scottish_government_budget_response.html.", "publisher": "Scottish Government", "retrieved": "2026-10-29", "sha256": "f3d0183e470f711b6821e4956e108c5d17210a392e6a6e471eccf297c807a6ac", "title": "UK Budget delivers real cut to Scotland’s funding", "url": "https://www.gov.scot/news/uk-budget-delivers-real-cut-to-scotlands-funding/"} +{"access": "direct", "bytes": 27662, "date": "2026-10-28", "doc_type": "html", "note": "Fresh exact prescribed browser-UA curl fetched HTTP 200 at 2026-10-29T15:22:47Z. SHA256 covers the fetched HTML bytes. Visible publication timestamp is 28 October 2026 15:45. JSON-LD dateModified is 2026-10-29T08:22:59Z, while dc.date.modified repeats 2026-10-28 15:45; no earlier copy was supplied, so a numerical change cannot be established. Five rows, read from opening body paragraphs 1–2. First staged in the uk_fai_sfc lane on 29 October; moved to uk_devolved_govts on 3 November 2026.", "publisher": "Scottish Government", "retrieved": "2026-10-29", "sha256": "f3d0183e470f711b6821e4956e108c5d17210a392e6a6e471eccf297c807a6ac", "title": "UK Budget delivers real cut to Scotland’s funding", "url": "https://www.gov.scot/news/uk-budget-delivers-real-cut-to-scotlands-funding/"} From a7fe8fb70140d20cf466c6fd4e2f0007a73182bb Mon Sep 17 00:00:00 2001 From: Claude Date: Wed, 4 Nov 2026 18:25:58 +0000 Subject: [PATCH 28/52] uk_thinktank_misc, uk_trussell_wpi: apply review B's misc 1-3, Trussell 1 Misc row 8 restates HM Treasury (England). Registered metrics move from proposed_metric to metric (misc 2, 8, 10; Trussell 0; review A H-6(a)). Row 9 is per person. The two stagers' shared block stays identical. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_01LwQq6JvunwtJCAckL6L4aM --- .../uk_thinktank_misc/NOTES.md | 37 ++++++++++++------ .../uk_thinktank_misc/claims_staged.jsonl.gz | Bin 3963 -> 4014 bytes .../uk_trussell_wpi/NOTES.md | 3 +- .../uk_trussell_wpi/claims_staged.jsonl.gz | Bin 927 -> 926 bytes 4 files changed, 27 insertions(+), 13 deletions(-) diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_thinktank_misc/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_thinktank_misc/NOTES.md index a9edd248b..5311c6ab4 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_thinktank_misc/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_thinktank_misc/NOTES.md @@ -12,6 +12,15 @@ Staged from five primary HTML responses published on 28 October 2026. Sources ar No primary was re-opened on 3 November (the session's network policy refused every producer's site); every change relabels fields, and no printed value changed. The 29 October record below is updated for the new row set. +## Changes on 4 November 2026 (reviews A and B) + +Review B (PR #231, issue comment 6635498758) approved this family with fixes, and review A's H-6(a) asked for its finding 2. The fixes sit in a block after the rows in `stage.py`, shared word for word with `uk_trussell_wpi`'s (judgement call 10). No primary was re-opened, and no value changed. + +- **Finding 1:** IfG's "around 80%" (row 8) restates HM Treasury's paragraph 3.5, which `uk_hmt_redbook` row 193 stages: `restated`, `source_model: administrative_data`, `conditions.originator: "HM Treasury, Budget 2026 para 3.5"`, geography England, denominator "households in England (HM Treasury)", and a rewritten note. Own 4, restated 8. +- **Finding 2 and H-6(a):** rows 2, 8 and 10 carry the registered `fiscal_headroom`, `affected_share` and `affected_count` in `metric`, with `proposed_metric` null. +- **Finding 3:** row 9's `measure_type` is "real household disposable income per person". +- **H-6(d):** row 11 (Citizens Advice) carries `geography: "not stated"`. + ## Access recipe and primary hashes Every primary was fetched afresh on 29 October 2026 using `curl -sSL -A "Mozilla/5.0 (Macintosh; Intel Mac OS X 14_0) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/17.0 Safari/605.1.15"`; every response was HTTP 200. The lane kept the HTML bytes, headers and text extracts in its local primaries folder. Hashes were checked using `shasum -a 256`. BeautifulSoup extracts were used to read the page's own body text; hashes refer to HTML, not extracts. There are no PDFs or spreadsheets in this family. @@ -35,18 +44,18 @@ Reading order is CPS, IfG, Age UK, LWI, Citizens Advice, then top to bottom with | Source | Rows | Own | Restated | |---|---:|---:|---:| | `cps` | 4 | 1 | 3 | -| `institute_for_government` | 6 | 3 | 3 | +| `institute_for_government` | 6 | 2 | 4 | | `learning_and_work_institute` | 1 | 0 | 1 | | `citizens_advice` | 1 | 1 | 0 | | `age_uk` (manifest only) | 0 | 0 | 0 | -| Total | 12 | 5 | 7 | +| Total | 12 | 4 | 8 | | Measure key | Rows | Scope | |---|---:|---| -| `ab2026__household_energy_payment_100_bands_a_to_d` | 1 | IfG share receiving the £100 council-tax rebate for bands A–D; Table 6.1 line 1 | +| `ab2026__household_energy_payment_100_bands_a_to_d` | 1 | IfG's share of households in England receiving the £100 council-tax rebate for bands A–D (restating HM Treasury's paragraph 3.5); Table 6.1 line 1 | | `null` | 11 | context and document-tally observations; none carries a proposed key | -The sole registered key is attached to the specific rebate's coverage, not to the entire cost-of-living package. IfG's exact identification of the £100 council-tax rebate for bands A–D identifies line 1; it gives no contrary scope. Its household denominator's geography is unspecified and is not inferred. Context forecasts, outturns and document counts, including CPS's PuFins assertion, have null measure keys and null reform hints. +The sole registered key is attached to the specific rebate's coverage, not to the entire cost-of-living package. IfG's exact identification of the £100 council-tax rebate for bands A–D identifies line 1; it gives no contrary scope. Its "around 80%" restates HM Treasury's paragraph 3.5 ("around 80% of households in England"), so the row is `restated`, `administrative_data`, geography England, as `uk_hmt_redbook` row 193 stages it (changed on 4 November, review B finding 1). Context forecasts, outturns and document counts, including CPS's PuFins assertion, have null measure keys and null reform hints. - GBP values preserve printed signs and cash magnitudes. “Tops £3 trillion” is the printed strict lower bound of debt stock. The six-month £61 billion debt-interest total is not multiplied to a year. Fiscal headroom is positive headroom; PuFins' £3.1 billion is a positive debt increase. No sign is reversed. - Percentages remain percentages. “Almost 100” retains the printed upper threshold 100; no exact policy count is inferred. “Around 80%” retains the qualifier in conditions. @@ -54,40 +63,44 @@ The sole registered key is attached to the specific rebate's coverage, not to th - Only IfG's fiscal-rule probabilities name a fiscal year: “2029/30” is normalised to `fy: 2029-30`, period 2030, `fiscal_year`. The income projection prints calendar 2030 and uses `annual` without `fy`. Other observations use the publication-year point-in-time anchor 2026 and preserve “last six months”, “this winter”, “each year”, “last 12 months”, or unspecified horizon explicitly. They do not claim a fiscal/calendar scoring year that the producer did not print. The PuFins annual-rate assertion has medium confidence because no scoring year or fiscal/calendar basis is supplied. - Real income remains real; all other price bases are left unspecified where the page does not identify one. No baseline proposal is needed for these current observations and explicitly named context comparisons. - CPS's own PuFins assertion uses `source_model: unstated` because it does not identify its calculation. Official debt/outturn context uses `administrative_data` and `administrative_fact`. CPS headroom and IfG fiscal probabilities/income forecasts are `restated`, `obr_efo_forecast`, `different_model`. The precise original release behind the unsourced CPS figures is not identified, so these originator assignments are explicit in rows. -- IfG's two counts are its own `arithmetic`, `different_model` observations. Its unsourced rebate share uses `source_model: unstated`, `own`, `different_model`. LWI's rounded NEET figure is treated as restated official labour-market context, `administrative_data`, `administrative_fact`; its response gives neither a quarter, age range nor a citation. Citizens Advice's unsourced bill forecast is treated as its own assertion with `source_model: unstated`, with missing method/geography acknowledged. These two method-unspecified estimates and CPS's PuFins assertion have medium confidence; the other nine rows have high confidence. +- IfG's two counts are its own `arithmetic`, `different_model` observations. Its rebate share restates HM Treasury (paragraph 3.5): `restated`, `administrative_data`, `conditions.originator` naming the paragraph (changed on 4 November). LWI's rounded NEET figure is treated as restated official labour-market context, `administrative_data`, `administrative_fact`; its response gives neither a quarter, age range nor a citation. Citizens Advice's unsourced bill forecast is treated as its own assertion with `source_model: unstated`, with missing method/geography acknowledged. Citizens Advice's estimate, CPS's PuFins assertion and IfG's rebate share have medium confidence; the other nine rows have high confidence. ## Proposed keys, metrics and units No measure key is proposed. CPS's PuFins row is context (see the 3 November changes above). +Registered metrics (since 4 November, review A's H-6(a) and review B's finding 2; `models.py` registers them, so they go in `metric`, not `proposed_metric`): + +- `fiscal_headroom` (1; GBP): CPS's current post-Budget buffer against the fiscal rule; target year unspecified. +- `affected_share` (1; percent): IfG's share of households in England receiving the rebate, restating HM Treasury. +- `affected_count` (1; persons): LWI's rounded current count of young people not in education, employment or training. + Proposed metrics (counts and exact meaning): - `public_debt_level` (1; GBP): CPS's current national-debt stock lower bound; no series definition is supplied. - `debt_interest_cost` (1; GBP): CPS's total debt interest over the last six months, not the change caused by a measure. -- `fiscal_headroom` (1; GBP; existing README proposal): CPS's current post-Budget buffer against the fiscal rule; target year unspecified. - `public_debt_change` (1; GBP): CPS's asserted annual debt increase through PuFins, a context row; period limitations above. - `measure_count` (2; proposed unit `measures`; AB2025 family precedent): IfG's approximate full scorecard count and exact count of tax measures costing/saving more than £1bn in the last forecast year. - `fiscal_rule_probability` (2; percent): current and last-budget OBR probabilities of a current-budget surplus in 2029/30, as reported by IfG. -- `affected_share` (1; existing README proposal): IfG share of all households receiving the rebate (percent). -- `affected_count` (1; persons; existing README proposal): LWI's rounded current count of young people not in education, employment or training. - `average_energy_bill_change` (1; GBP/month; AB2025 IFS/RF proposal spelling): Citizens Advice's forecast monthly household energy-bill increase, excluding any assertion that this is a Budget measure effect. -Registered `real_income_growth` accounts for the remaining row (IfG's restated OBR 2% cumulative income growth). Proposed unit `measures` is used only for the two counts. Other units are registered. There are no proposed baselines. +Registered `real_income_growth` accounts for IfG's restated OBR 2% cumulative growth in real household disposable income per person (EFO para 1.2; `conditions.measure_type` says "per person" since 4 November, review B finding 3). Proposed unit `measures` is used only for the two counts. Other units are registered. There are no proposed baselines. ## Not staged and primary limitations - **Age UK: zero rows.** Its own response has no numeric Budget estimate. The exact “On Income Tax threshold.” subsection was read; it welcomes protecting low-income pensioners from cash tax demands but prints no number. Dates, contact numbers, navigation “Nearly 4,000 flu mortalities” and the unrelated £2 donation link are not Budget claims. - IfG's £100 rebate is a bare policy parameter; its 80% coverage is staged separately. The £1bn cutoff defines its count of one tax measure and is preserved in conditions rather than misrepresented as that measure's costing. Other dates and the 1991 council-tax valuation year are context labels, not model outcomes. - LWI mentions extra Mayoral apprenticeship funding and OBR employment effects but prints no amount or employment effect. Its “one million” NEET level is retained as context. Navigation cards and unrelated articles are not harvested. -- Citizens Advice gives no source, exact winter dates, geography or method for £60/month. Those are explicit limitations, not supplied from secondary material. +- Citizens Advice gives no source, exact winter dates, geography or method for £60/month (`geography: "not stated"` since 4 November, for the contract's closed geography list). Those are explicit limitations, not supplied from secondary material. - CPS does not link its “paper released last week”. It is outside the assigned 28–29 October post-Budget window, and no unseen figures from it are staged. The comparison between six-month debt interest and an annual defence budget prints no second amount or exact ratio, so none is derived. -- No restated figure is silently removed: the seven official/OBR context rows are staged with `restated` and an originator note. No PolicyEngine restatement was identified. +- No restated figure is silently removed: the seven official/OBR context rows and IfG's restatement of HM Treasury's 80% are staged with `restated` and an originator. No PolicyEngine restatement was identified. - No pages were unreached. No charts contain unstaged numeric labels in these primaries. -Potentially ambiguous source wording is retained: IfG describes the £100 rebate as reaching “around 80%” of households but does not identify the denominator's geography. No UK/England assumption or correction is made; the row identifies the precise rebate and flags the missing denominator scope. CPS's PuFins claim lacks components, annual dates and method, so it stays context; its £15 billion headroom lacks a target year. Age UK's Pension Tax Allowance paragraph contains no number despite the assignment's pointer. These gaps have not been repaired using other sources. +Potentially ambiguous source wording is retained: IfG describes the £100 rebate as reaching “around 80%” of households without naming the denominator's geography; HM Treasury's paragraph 3.5, which it restates, says households in England, so the row carries England (4 November). CPS's PuFins claim lacks components, annual dates and method, so it stays context; its £15 billion headroom lacks a target year. Age UK's Pension Tax Allowance paragraph contains no number despite the assignment's pointer. These gaps have not been repaired using other sources. ## Verification +- **4 November:** the contract checker reports 12 rows, 0 problems, and a re-run of `stage.py` reproduces both files byte for byte. - **3 November:** the contract checker reports 12 rows, 0 problems; a re-run of `stage.py` reproduces `manifest.jsonl` and `claims_staged.jsonl.gz` byte for byte (COUNTS.json pins the gzip's sha256). The fixes relabel fields only, so the lane's checks below stand for the rows that remain. - **29 October (13 rows, including Trussell's, now in `uk_trussell_wpi`):** diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_thinktank_misc/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_thinktank_misc/claims_staged.jsonl.gz index 0aa281cf2bd37508a44f6dc50d8c12fa18421480..af68d8592e9a436e2767f181548dfb6d6ced39c8 100644 GIT binary patch literal 4014 zcmV;f4^i+RiwFP!00002|Lt7clH0}+ea}}+u}Zae6|u|vC8|8gvSTN)W0mbx9#I2e zKu#6}7|dWP+Lfw2mTn0To-RJaWMn80< zwWW?KtBbtzYtbo{vC^uf(-oaa8kOl>h3g{I@fNr9D$D4G%H!oqmX~3iNn_|enXGiK zjV-0sS1L?pY-xzN$P-Ot(?dVdh?*ZSDr>7Ur$CFY%7%m9 zU_wtVRIw=KdP!1r{`3dBv@A;fy^}gGHeDgDSXU7wl{^_dfd>{km$oR~TcSu)9c!7@ z4~i6b^H>R*GlbW%5g+A+73$?0x3$f-0)MLHV;arqMIK^8BxtRzbzwBkSNsM;6!9fK zAaedZzLBWNcK(JJS4x_yRN2Rf>_jvbvUX&}D>F=c<`}He2Kcs4|F(Do?~;D-4lpRd#qpt7xF-MQ+tg>m=+7nx7P(uzH1C zS)tNY#r9l$w^U*+7fK|BGW=vNS4vo1u63F1bXckk6QeskRr~>U@?t;5t727@Q0!X8 z5GWQ#l^R3h*&@QvS3FITz|?3mI!{$us^pyHUdz%bQgo^l7#h!EsTWJWhhd5YSI92t z_p-nms3dIE*>Ou7l1O0&tI9g<=PF3JQ+nH8$UbEpD`(G%E|oK3l98^^7lVE;7|iHu zR>a<}zSMFhzK~d{h0NKM)Cwljg(geMMk^u(-Q_x8h#&v{?@vl?rOto+`+v;7vEYi6 zRl~Qop;^koc*38KXTyQ&_4~8QL`Cy-GE4`(@qDVLQ92q%aj&0@M$shdN8?D1hDs?s zD3eJVk5!xuMjby@vs5hdbI6C)NwFL+oCSPqs{{)~Pg>YL&-m;a7ARk+vZ#!}dRU!T zZrxz!UrLxdyBaiz+Hj1=#X5jm@)k5pyG>uT-V z{QBY|UYqlxT%1>z7b}%$d4bik7o_qF88ku!Z6Sgp4KfzO1bZ?BwG06U4Qc@bJP!c^ zhbkBBmVd2k2RXVRFyoCNj}xz74I6JH@k<5Wt-__wTh?SCi!)bzxzk*LDQ7^m>(7O& zB2n1^?1aI|*4tJ_yd3^z*jME+^wO=?N`(!st_w&M>O7>&tY_Fe`BJRbjIN7lO_#4^ zR?(otUh5x!J-pO;!el^-4*@XED`6=&4xC#0_}j03ct2>QfVHrlLLA_wehCoE?b19n zm$2<&Ar+|+Frt&UrsLBuhDhhV#1D<7O9dWcJ|tE44? z$2cZEa%VvH(59)u0> zqMxRaMhcz#1n2GRS;xr&GQsi5F&ath)Pr^6vJ3(V^7#b?QaR}X?`)O@Naf9{sFm0i z7K$g=lME$m37~gsreQZqT>?zWB3jV$RA^TPBU7`iLIBWDtm^FFQe`P6>Cb z)g8)qv|bDKb0=n>%3P~Ve4$F7$_bg-_U`WZfr;M@6T7Ln8Z#prtsG+tknr`Z;rT9! zYoPny$+*gQQU8Nf_4>Ew;FcmDjD+ulZwZJ%;AKKA=lvidn;gu<*Rd@kK)fI7I^X8i zif+T=c|P2rXdAO=Mt=+5t1Dd?7MU!_DiK&6kN@VWIFs254sj+*MRgy=GeN`;q(7~% z6LW2j?A*3xPl~v5aOOHbJeLI67GF|_r`_^#r8+l`cIlVz$T)fg_&F&7V^3Lr?MTH> zj)mP#@jfbbUfV3zAujMwn9hAM!qDz!AuSZfb?E88PEEI(QGswj%TAmI+xLSIF;jGe?E>U^BchK>BM3C z=S>i_hwHyDmN_U5JiX-#BCJQ1o2ANDJh8fVTTx6A=|vOB7)N2y%=7?)Z#J8A?Z1wX z?*cFn>*vuVp8T4 zF7}jkYW1u{38?0d!^t76 zMi|h5rl7h#nF;*B$SeA3<)U`rCi2NBckyvuX>IChp4?3pEf908c`Cw1n`CmaTx(11 zjw~xS)XXvWfcu*QrhSGu{!IM%kAI53pG6zWf0usF(N>miNjuFv)yr6^gdL&muuCWi%)C(54Z{5Bp?i@oSZt#2_Qx(i39Vt`h+nq+?C{PtS z5L8{|ZtWAF;;a!);xX>Dq`Kr1e$4_OK>n^E|8-7Qol_zAW!w1QMmZtGx*JJ?7L=yRow%fOWaB13%>UQ%)>4IS4gAc7lzRePqsVA?h}#vlt=(MG4_JIhES6$CdBPD- zth&Pa4N~6qUYNvA^+3V)ef247)AteH^#Tc;Csr*W#!wL@qo(-&UPJc$RLGt`Fl58yi`5S3mN0xpxjAF)XZa$d ze%?o4d?mgs*}1kK-?8Cbk9DoK!mYp!?>oB&*QXW2Tk>K}mA>*Nhu5#D{zl1rtTRsF zN)-{N+`#tM+`8=`pD{rU&&ST~YX@@K8P&EAV0=eL^kx_~VudPKY{UXZ*2_ZO_oC!TxvhE32fkpb8z3T>O=6ezUQgb+=JsLT zC@6Jn4F)kiOQko1C1)LcAjrF>H#=`+r*mMYS{7VeZ3{4Ol4eI9#XYJLAMv@)y;rXu za2M|ZvYnmUJchP^0r`=OFSRN1utAMZYd6mt>!en$5_li&42tCG2plgCnZ=QBU8h>$ zL$)&^k!SWI&C$we0k*6Lt$DNx?UNO0ZLU zA^gSAJs&Q>Hg4!-QA9P;CoWViuR=J7MdJ5p_8BPunKP%^Dm}i@Z?SB zRSxbsCJM4T-BP#QA6!Jp68QDJ_3q&#$=+ioEw}sl1692(RTb0~fsNMQ$0Zpc3b!zi zK*{4t?VK~7NHl{RJQFuSaD!|HPu`#7hQ2!IcDe}?jmvF6agMz?c<9~>+MnVg zHm~()?*~(%*-O(X$ivI~^Y;yIZg*ch=H5D`=u%K6DB$Kz78Hkbz!59jXH{Co#}}LP z?*5oNHO3u3@L>~j&SuAy1%TRm=a%rsxH}>G$Im|d&dK#Sj-GsL)q1Irz+TCmDguZE zasx%JD2F8kt4+kSz_o3%6&qt+ECo|vLP2mFw`bP!uK0yJ*?4sGR6I>pZ#tE|WS)+u zvyqgec;1iay-5_SRQ6`mIG)Q9!uuIm1@lM-qv2>859BNwCvkd%LB0fYvjmio|G6kH zo87r;k3_!V$tLw4q{BVSYZQ^yLt|u^^QGs7vKc9`Md0wEW`ck>zh38q#s$RPJoBd+ z!b=)L)R1Hibz2F)%)G>Ve`~kYEq1oH)cWWSm+T)GY!QdV9SRyq6zU?|+tmI}Kliaq z23+}#Hwc@zIs!83aYKR5gYR+xMtt1){F4`){qX%Z;QuN2%SPP?-=>6w&ms2r3K;b) zxkBbsA0gQRRtuM@kVZEF&Hx)kl6(?ln`O4mhD|l;T*0kiXr_uhGHx z@x|wU4Z985xaHx$?!?el8kDWXo%4`c%<&EutQN?h_h_}C^3(Qa-RjVVbd@JxUZwSY zu)`Nfb)RQ6&#z$Epxte2-RgD{gJW`ZC$8qdRe&v%RZ9Ei)*N{dcAG+$NBLzU`@P|C zA}7-{?u}>T{s06MC-bDIW<5EWj;6g-fm`HwIvr2uy)^2laU_THB$`Zzv+>PS7yFjD z$0^tD)$>H$11X<#D?ZDLvMDql$wK0K6v=~lmf>yRL1o<0iXi609~ZDYV}d$zF&F;T zz`YxI0f?C5MgVj81cgkP4flMPx93vQxFv(k*vA|=oN#oa7ABO9rhY-kD5!W8HUQL| UYQM!{cG{KuKOv!9PghF-06Esmk^lez literal 3963 zcmV->4}|a^iwFP!00002|Lt7eZrsKdzVA~E7BJKROKJaqD)ho}(j;xt7)gp=5<_yh zI~AAQa7am;AV9AA0KI8@?Kg4XBu~=soSETrC9S=Z*LLjK2pow&!!u{j`M&cr;~zV+ zs!AQzl`itmA4I2ArjnIPI$hC;q*0m9Rk$uP9dB?uud|G9s61Y-WO)(BnKXvplgUcw z+Ek^i^rZ?@8CNt!T;z$SvFV{7X++JB7j;$DWloQz+Qc#o)uqZS9;vSCRUSrlvQSkx z=nW?H)It@DQm&UIMdxpSp-ao6)IV6M^Wv&2WF^*h1W6@N22bFDh0bMFl=dxAq^XXz z%-n+_#oatsg60h2b!^1%^P&>!dX3w<$~FRjtK=gZ&FDoQVnQUS)7H8$n&vD1h#`vj z0v`}Le;(gRR8)0-!iy^deAr{cS%5^K3oA}N&NCv&+{qQYgT%WSK|Qe~JJ-QlU?FQ}6j`y8)|RZ&8* zYZXJFSQu4m42frp2tQx)G(`eaqsizzRb{D?Q<8fvOQT59sZL;MJcp%TEcqUWDH2>E zyP)680&Ad>uu*5nE^SC6g&C|W>$IP%A>mf(ZF?d6lyR)AJtw+U)`Uq$xwF=8`T0K|m%5TV|K;cZnq6bT z6)UTTZ*4-el!NhvKON781J&#IXOoGF=ILaZ4tnGHR86CFG>qb2KN*dpNz{+Vks1w^ zQg~1%lQbTyI2nvOeyV1vSmY;=537@6Ii6Yz__nGOED$|e!R~p+Cy%i}`9hUNZ3Nb% z(s^yy4QBqOgvpgBMpW`Tka-edsbPVs%qwX&Cn69}o_;GXRhf{H;vF&~YjyOIy3BZ8 zt7>ijaCR22&1q3CPV0-al}fZc!)jG$r1CQvG(rSzA%Y?eG8V!FJ2C`Lh5&;GPJjT< zLx8}d${D-m_qB77y$b>}z7phd;`y^-^~YZiFLa(T8Ia;b08H~rSjsC4PAz@>^t&J54;m?8Eo`R{2Y9Kk0b+Tz zG;f+q*!HlHiqr@g(aEdvNYtj0d7B%))z<;fk^P%kqADtxIhM%rAG|FSas_&+2)*|) z$6pI!S3qo(Ea^#gjm0A+x(~oJ**Za8fF6JxzL%T}i|F`JjD;|){RTpBQ2H_0foZ|3 z7Q>k8U809B{}42$0l%q@`al|Qfi#{!oAevUq9It`o1VL&rG~|4MUFt}1@Q8|5Xz*# zpH6tJJE&x<$YWATq5(}M)bI520vdtpRV*zO9D_&f7(rQQh@E9gaLK4hpt_yRE~9+? z@i!hpy*Ll%kEN1HSrjWF7eEpWL`LTZC!t;!n>>a81&$xuI%wo^B;yhd3t8J-gE@g^ zbxLb7N>`Wy59ORLn0gJJFnf?m0S^N*PtPTIqJ`Z2`Z1`(S%#m?TjPzC)3814@fdI# zftJf6!Icjg#oHkVXSi4oZ!K5}5!Hl9u=gv-ULdFFj~n1SHU$(>8DhM!4Q+1%|NGndqUh`B@v zuAm&aw+7ssiUVUPvJg*g;_)X=JSY?S0$8sr)*OR@c+tx# zq>)1BKEZkQde(mCfJ|_Fa*Re&mwK>HY?eVFK|a5rKq@Ca;GL^w0fu?8Dx4C#qJrYd z^&~^dS_1X0nrYZ8r7i%bWDzZBITzYi!HA=L@{?c@4TkQD@5q_N=<`&3rK%<8LkZqlt-*LcW#Z_;F&`~UwIrgm(0&Tk zy@LtcS(VjiP*TawDw&pIH^b_(e8{TV)(9CF(3L|oh>HKglF5u?m^^O2kXs3N>(v9w zcC=m#^|}f66Pat3iO*HZQ#l|r+dSMIKQQszVPZQKTVrNKqm^S!0TMocHay)XaSe1I zos8>z8}+}Ds$T!z9NbdG8zbRk@GSum2)s;)<-8vxWRruL_&TnN2oUdwy2-bBy`tN& zc%Bb8DB8v>n$cf@_gtk5!y=OfStSCiEY_Ix*Mg z$j+;_>`4*V7S3$Phv$+2+u{rA@U&Y#wp3@w(KZ+J9T`iH06!}wVC;cX5G#!gP+s2wwwY2p^&%+3X{Mzf+*V1{(r~OehN@u;< zY&1!u*0CzR;Z%-${rNba%x?j^rxT0qpEW_y z4zB;XSmvNK@brc&h_D`2UM*F&;)%K1ZA~#nq!&#fV=RS1Gt)Z|{OaoJRQs>vRuwNvP>Rvm;Mmx_cOC%DjhK}Ago_;| z9a=pr5d*;vVTP{>uiYR`&a}7n;@xEjthb7Jr;wV9p0+t)!+4*1oo|a{4o>dEYJ>p| zXbP&^lbOH|jJ%|uRyJw}ZX%zIavLAJN^4V3^W<*oXn~kx%~KIJ+9Z>Ug$DfE_{`oJ__p@k2`4{PD9BpOUhP2bnQ(ebOCF~F_u(dO6z}5r2 zz3fzTDn4=eA8-@ANkAA*IXShK5xCsS^TLt4@oxdesS4(ij+80$%~qpu6sU?E2&%4g zyY`7san=YY@fdenQeAQh-?6|4kbfx1f1Q(6=Tz0bZX5smh~x*x+t9{4)9P?fao>P4 zpSq+kc`L(WuCtmfi|lHCHtg{`?kgd%xdN)ti;sD9ADuR%Z=5gB1N-V7ByvZi#El;S zrwRng(lkD7I}SjZNFtxn)#eo-T$(l`d9eOQZ5!0Um$p3Z?|N9~CCMP)wumJOR$!HN z2Hta`-3M%ky%%;UNCRP81FyM|awNbtii~!PxLx6F?e>y>z~TpDu@vLUBaV1t)fLun zkn*l~!X$R8Hxz6it54xf-$!^i3nXx!Shau{*T$0r!5SL-&4D$?&Gs$sA0*1en7nNu zAX@Dd$HPL9CCQEgxxDj%?*R+dRa2c zS?D|$4Nk(`cKX4GVSKBm!Y#ln-fwmTrVlHCZv_SL#Y$iLQp59SRClAKJ=Pf~Zl#I{ zOW1dv7uXbw3ek5Qk{#cRI;)}pex&(u5)it z6wbqu_(g8%x{-aS=Tv;wEV3(iHF^*&i}B@lH>aIlPE8qxYxH@&QVmP_V-;L5y0u=x zcD;D2tYq;x_@pa{nkt=cXq~sMYwlV1Z*Ah$jjoT=;+32L;Sy{34KUlMY;fE(FG>z@ zTdqlSgC|v*T@#C)U>WTeP2PQGw@=-w19hEdgRQ+lrL2z+y!Ozngh52idXL~tS>~`v zwJh6Na|`cob6rR7g)J)XwpJeTZ{ri4W7FenZ}#3njI%=HFb_TD063u(`|3rrJY0V zqzM2Kw-%eeaKL_Eq!GZSmy7fAe2Kpp+ULVXSzJ}i&fz@!q+Zodi}u~^o_0WaQpbvR zjck{fd^Mr7nf5pdZ|L$breg5uMJGz`+#>Axi~8CAXd&W}z^~%HH~a2M$2<|ei~MBw z2Y+CtS7oJw`WCQJ=Y3o>0giB!aSwpJJIR-I#slePaEo2y1_*9Z%;3@cb0E;y&+J|_ zL85U%>?J47o1=mDy}-VA;K{&NZ+e43W+B$9SZB)Z$`J}E>-K7@)jBISwhsYRD5a)N zDo1Kr8}3<)mE9BOde>drQr4p#8LBj-{B~P}VmQ3KJ7?c^<|g!w!{@EDgEpr`+J<!r%Yr2J}5@=2rJ9#T|b9VjJ<>0vhkrHLr9MW76ZgJ{`;62Ko%~xbxY^PYK`c z{Wik?tzG_Vx`mMN3B>+E0f-(amq>5iNscWl-BC(=N%hR{ZM8EFIBRN@Vrhh+zTHxH z5a8SC+X7s-VHdkR{I8A}x=MqxDcIu{GK)E=!Gbw~{IQExCn!H`i`4xJW44my3#7EZ zZ|n@jN{J^K&GRc5Hh6Z^S~u6#VFXO}Ziv-%Rn z5XX&sZk`BqAmtP8h-XLaU+1a`yhi%nDTafmv_lhX1F1P%-F{)I2>?-!3h(}MpM6{(+pG>3L5}w V4wv8KOgU|i{XY!YQhwe_008x*p~nCK diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_trussell_wpi/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_trussell_wpi/NOTES.md index 7cbc84c91..0ed3687d2 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_trussell_wpi/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_trussell_wpi/NOTES.md @@ -14,7 +14,7 @@ The thinktanks-b lane found [Our response to the Autumn Budget 2026](https://www |---|---:|---| | `null` | 1 | Helen Barnard's “over half of private renters on low incomes” in receipt of Local Housing Allowance ran out of food and could not afford more, in the last 12 months | -- `proposed_metric: affected_share`, unit `share`: “over half” is the strict lower bound 0.5, `value_kind: range_low`. +- `metric: affected_share`, the registered metric (`models.py` registers it, so it goes in `metric`, not `proposed_metric`; changed on 4 November, review B finding 1 and review A's H-6(a)), unit `share`: “over half” is the strict lower bound 0.5, `value_kind: range_low`. - Trussell calls the finding its own research but gives no method, research dates or exact percentage: `source_model: unstated`, `own`, `administrative_fact` for a retrospective reported observation, `parse_confidence: medium`. - Context, not the modelled effect of the LHA freeze or of any Budget measure: no measure key and no proposed key. The 12-month window is `conditions.horizon`; the period is the publication-year anchor 2026, `point_in_time`. @@ -25,5 +25,6 @@ The thinktanks-b lane found [Our response to the Autumn Budget 2026](https://www ## Verification +- **4 November:** review B (PR #231, issue comment 6635498758) finding 1: the row carries the registered `affected_share` in `metric`, with `proposed_metric` null. The edit sits in the review-fixes block `stage.py` shares word for word with `uk_thinktank_misc`. The contract checker reports 1 row, 0 problems, and a re-run of `stage.py` reproduces both files byte for byte. No primary was re-opened. - **3 November:** the contract checker reports 1 row, 0 problems; a re-run of `stage.py` reproduces `manifest.jsonl` and `claims_staged.jsonl.gz` byte for byte. No primary was re-opened (the session's network policy refused trussell.org.uk). - **29 October (in `uk_thinktank_misc`):** the lane re-opened the saved page at the locator and checked the quote and value against it, and checked the quote is contiguous text under 15 words containing `value_raw`. diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_trussell_wpi/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_trussell_wpi/claims_staged.jsonl.gz index 6b177800811cced4f4e4bfa4d65afe226e5d233f..e3242e3873e904fba2e743123849cf25c27e603f 100644 GIT binary patch literal 926 zcmV;P17Z9hiwFP!00002|9w=sj^joU&HWVxCXV!lziiQHhX~ZbXVR*J7fQjp13Y)^shmbh+t|Ri z3QwFTH<)7TT~#M@z*wyot0(a~ARW9KS^-!5`hz^RLiRH0R4B}cwjoqbi)V7Ue-H}KG^u6cm>KmUTod;>Y) zdw8G|`Ufr?6;GW}1W*tpeUurXFKbGS?Dxb;8;gFd5i6K#{;k9_?9%8BlrM&Vc>E5W5OxmQ@jLbIE=Tu_ZIR`GQfBl05UwZLkBbnp}t+VCL`>b46ibmRO^CQcQk8oxF_~pgN#GWv*u9+{PRo zQ-PV=JYSFadY#H#2pE(PSh3I>XrTC3hYmg|?-c8hX&iVVaRv;X z@;pgI!9Y}A2B`61ofDAnU;q9M;plh+8&5rp+~3CV{`H@X&|5wQ7g616^Cab-v{@~m zR%W^0t(wIe_v@|VziPd&&3dOl?6+#SHtWr{Hv6Z&S?S$ov)^q$EX;DN_M0r-j61aG z@3kd{$jZuSe;ixMP()F(sn2=I;Zt66n3g>GP6FB(hw$<7@q9kd<2B=){9&Gsj}FhF zFb$3$_ zS;CTSYP32JwzyeXZXGxAZ9tXvDq)l6??R%gGjUS{+tYZ9k0Y-=R=4{+5cgAASIKkc zkX0!n*Er;Jdb=?v)g?*GcVzmij=bmBvv|OY0pA=eug?D=srd5Uzxdl3>3;?Q0J84P AG5`Po literal 927 zcmV;Q17Q3giwFP!00002|9w=+j@w2Mz2_?myv#+^Xt73;eHxzzh=BlEkR#FTE{YA2 zU34!}Gy?-ZhytuJyfoRWF5#(c3yjORjhi1)W_ngQ8XeJL|9- zsI&UQ7MHqC4Cr**D}StXr$UerWqRvuh+ai|!m3d^N(xQRSV=2RFOtz_cS?>cr%M0d35^QwZ`&XX=OQZw*V_=2Y*j-?f63;5OHb7%p z0SqD%*vR8N%K!r^uiF{09~fd+p~;S@N>Y+%Zn+%G3kLL>Bh&%s z@CzmWV2TZPz*UnAkps*eerA3s45&qyc{)sZK%Kme7oa+zKV`0F$=UdA zoI|Ra*=d+Bh0+?wjWd=Nl9x=VkrydfE*Ir$C$Amv#@kLkVTTSrD(@8QkSQH_A#nyw z1LfJ02!uhYybMs|!8#{^U%&qSJHpZN1~#607P-HULYqNjYo0Z;eHv8T7-NG!lYQGuj5Zs|f zf2%VwL{?cw`_tG;h9c^cO?}Ku4j=Q9!?fhdcM{OXIE452_viC@9^W#~$sgwFc<=BW zO5WOX0+(Oz8E!e%C?l0zqZ|)OIUbTS*O2ARB<9n`3k9-kqMZM~X40HTJ`R1=T6Z(` zkmW4drbestV2i7TrPpy2-v?A#yAn3(|0X1=IukcVusw~p_&D;~V|BgH193l BzU%-1 From 451c303c499115de37eb001b909326a731fb435c Mon Sep 17 00:00:00 2001 From: Claude Date: Wed, 4 Nov 2026 18:25:58 +0000 Subject: [PATCH 29/52] uk_ifs_ab2026: apply review A's K-7(a), K-7(e) and H-6(a) Row 78 records the GBP 1.1bn against Table 6.1's 1,045m; row 42's note states its scope is exactly lines 1-3 and it is cumulative over two years; rows 39-40 and 47-48 move registered metrics to metric. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_01LwQq6JvunwtJCAckL6L4aM --- .../uk_ifs_ab2026/NOTES.md | 16 +++++++++++----- .../uk_ifs_ab2026/claims_staged.jsonl.gz | Bin 17266 -> 17391 bytes 2 files changed, 11 insertions(+), 5 deletions(-) diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/NOTES.md index d90103ddc..82484ecce 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/NOTES.md @@ -4,6 +4,14 @@ **144 rows from 13 primaries.** `source: ifs`. The manifest lists 14 primaries: four current briefing decks, the superseded TW_0 deck (no rows), two IFS pages read as browser text, and the public data of seven Flourish visualisations. The first lane staged 106 rows from the decks on 29 October. On 3 November the harvest applied the IFS addendum (`briefs/ifs-addendum.md` on the staging branch). Its sections 3 and 4 cut the deck rows to 87, through an explicit block that follows the deck rows in `stage.py`. Its sections 1 and 2 then appended 57 rows in a second JSON block: 11 from the two pages and 46 from the seven visualisations. Attribution: 48 `own`, 96 `restated`. No primary mentions PolicyEngine. From "Family conventions" on, the sections keep the first lane's record of 29 October; where they differ from the sections above them, the sections above win. +## Review A fixes, 4 November 2026 + +Review A (PR #231, issue comment 6635870934) asked for three changes to this family. They sit in a block before `main()` in `stage.py`; its indices are 0-based, as review A cites them. The coordinator read Adam's deck (slide 2) and Wernham's (slide 3) for K-7(a) and K-7(e); this session re-opened no primary. No printed value, sign or year changed. + +- **H-6(a):** rows 39 and 40 (`cpi_inflation_rate`) and 47 and 48 (`affected_share`) carry the registered metric in `metric`, with `proposed_metric` null. +- **K-7(a):** row 78 (£1.1bn, lines 61 and 62) keeps 2031-32 ("Biggest revenue-raisers in 2031–32") and adds `conditions.discrepancy`: "IFS prints £1.1bn for 2031-32, where Table 6.1 gives lines 61 and 62 a combined £1,045m." +- **K-7(e):** row 42 (£5bn): slide 3 gives it as the cost of the three measures "over next nine months", and Table 6.1 line 2 costs the whole VAT cut from 1 October 2026, so the re-key to lines 1–3 stands (£5,040m over 2026-27 and 2027-28). Its note says so, and it is `cumulative` with the time basis of RF's two-year totals: period 2028 (the final fiscal year), `fiscal_year`, no `fy`, and `conditions.horizon` naming both years. + ## Coverage | Primary | Access | Rows | own | restated | @@ -46,7 +54,7 @@ Rows 0 to 86 come from the decks, by deck, slide and position. Rows 87 to 143 fo Four proposed keys name seven rows. README rule 2 sets their spelling, and each takes the producer's printed scope (rule 3). -- `ab2026__package_property_and_savings_income_rates` (1 row): exactly Table 6.1 lines 61 and 62, Adam slide 2's "Increased income tax on savings & property income", £1.1bn in 2031–32 (lines 61 and 62 print 455 + 590 = 1,045). +- `ab2026__package_property_and_savings_income_rates` (1 row): exactly Table 6.1 lines 61 and 62, Adam slide 2's "Increased income tax on savings & property income", £1.1bn in 2031–32 (lines 61 and 62 print 455 + 590 = 1,045). Slide 2's list opens "Biggest revenue-raisers in 2031–32", so the year stands, and the row's `conditions.discrepancy` records the £55m gap (review A, K-7(a)). - `ab2026__gas_standing_charge_moved_to_unit_rates` (3 rows): no Table 6.1 line. Wernham slide 8's "Shifting gas standing charge onto unit price": "This charge will be scrapped, but per-unit price will be increased to absorb this". The spelling is the RF family's. - `ab2026__package_ifs_new_giveaways` (2 rows, from the initial response): IFS's new giveaways, which push up borrowing by £8 billion this year and £4 billion next. The page names no OBR line and doesn't say the giveaways cover every Government decision since March, so the rows don't take the key the OBR, HM Treasury red book and RF families share for those decisions. To the nearest billion they match the OBR's direct effects of Government decisions in EFO Table 3.1, 7.6 in 2026-27 and 3.5 in 2027-28 (uk_obr_tables). We infer from that match that IFS restates the OBR, so both rows have medium confidence. Neither the total effect including indirect effects (6.3 and 7.0) nor Table 6.1's total of policy decisions (7,595 and 2,475 £m, uk_hmt_tables) matches both years. - `ab2026__package_ifs_announced_tax_changes` (1 row, from Flourish 30726047): the tax change that Budget 2026 announced, at the forecast horizon, as a share of GDP as forecast at the event. The chart's footer credits IFS calculations from the OBR's October 2026 EFO, policy measures database and historical official forecasts database. The chart lists no Table 6.1 line and names no OBR grouping, and one rounded figure cannot tie it to the scope of `ab2026__package_total_tax_policy_decisions` or of the OBR's tax decisions since March, so the row proposes a key in IFS's own terms. @@ -55,7 +63,7 @@ Four proposed keys name seven rows. README rule 2 sets their spelling, and each | Rows (first lane, 0-based) | Was | Now | Why | |---|---|---|---| -| 45, 46, 52 (£5bn, £140, 35%) | proposed `ab2026__package_ifs_cost_of_living_support` | `ab2026__package_household_energy_vat_and_fuel_duty` | IFS's three measures are the energy payment, the whole electricity VAT cut and the fuel duty cut, exactly lines 1, 2 and 3 (README rule 3, corrected). Lines 1+2+3 print 5,040 over 2026-27 and 2027-28, IFS's £5bn. | +| 45, 46, 52 (£5bn, £140, 35%) | proposed `ab2026__package_ifs_cost_of_living_support` | `ab2026__package_household_energy_vat_and_fuel_duty` | IFS's three measures are the energy payment, the whole electricity VAT cut and the fuel duty cut, exactly lines 1, 2 and 3 (README rule 3, corrected). Lines 1+2+3 print 5,040 over 2026-27 and 2027-28, IFS's £5bn. Since 4 November the £5bn row (0-based row 42) is `cumulative`, period 2028, `fiscal_year`, with the two years in `horizon`, as RF dates its two-year totals, and its note no longer says the scope differs from lines 1–3 (review A, K-7(e)). | | 48 (£36) | proposed `ab2026__package_ifs_electricity_vat_oct_2026_june_2027` | `ab2026__electricity_vat_zero_rate_to_july_2027` | Line 2 costs the whole cut, 1 October 2026 to 1 July 2027, including the weeks before the Budget. | | 89 (£1.8bn) | proposed `ab2026__package_ifs_self_assessment_payment_timing` | `ab2026__hmrc_tax_gap_package`, component | Table 6.3 row 3, "More timely payment" for Self Assessment, +1,815m in 2031–32: a part of line 45. | | 92 (£1bn) | proposed `ab2026__package_ifs_hmrc_compliance_staff_and_debt_collection` | `ab2026__hmrc_tax_gap_package`, component | Table 6.3 rows 7 (debt management, +575m) and 8 (compliance officers, +415m), 990 in 2031–32: a part of line 45. | @@ -259,11 +267,9 @@ The table lists every proposed metric in the 144 rows. | Proposed metric | Rows | Unit(s) | Where and what | |---|--:|---|---| -| `affected_share` | 2 | share | Wernham slide 5: households receiving the £100 payment, 80% of all households in England and 91% of the poorest 30% | | `average_energy_bill_change` | 1 | gbp | Wernham slide 2: EDF's forecast £470 rise in annual household energy prices | | `average_household_support` | 4 | gbp | Wernham slide 4: nine-month cost-of-living support per household, £140 and the £24, £36 and £80 labels | | `business_rates_bill_change` | 1 | gbp | Adam slide 4: the maximum annual bill saving, £7,146 in 2027-28 | -| `cpi_inflation_rate` | 2 | percent | Wernham slide 2: the OBR's CPI inflation forecasts for Q1 2027, 3.8% and March's 1.9% | | `current_budget_deficit` | 15 | percent, gbp | Stockton slide 7, % of GDP (4); Flourish 30726471, £ (11) | | `current_budget_deficit_revision_risk` | 1 | share | Stockton slide 8: roughly 1 in 4 revisions | | `debt_interest_share_of_revenue` | 4 | percent | Stockton slide 7 | @@ -286,7 +292,7 @@ The table lists every proposed metric in the 144 rows. | `taxpayer_gain` | 1 | gbp | Adam slide 3: the maximum Founders' Relief gain, £1.4m | | `ten_year_bond_spread_other_g7` | 1 | percentage_points | Stockton slide 7: 134bp | -The proposed units are `index` (16 rows, the two index charts) and `measures` (the measure count); no row proposes a baseline. The first lane's rows propose `affected_share` and `cpi_inflation_rate`, which the Metric registry now carries. They keep their proposals, because this extension leaves the 87 deck rows byte for byte as they were; the new rows use the registered `affected_share`. The new rows also use the registered metrics `taxpayer_count` (15), `fiscal_headroom` (4), `decisions_effect_on_borrowing` (2), `real_income_growth` (1) and `revenue_change` (1). +The proposed units are `index` (16 rows, the two index charts) and `measures` (the measure count); no row proposes a baseline. The first lane's rows proposed `affected_share` and `cpi_inflation_rate`, which the Metric registry carries; since 4 November (review A's H-6(a)) those four rows (Wernham slides 2 and 5) carry them in `metric`, and the new rows use the registered `affected_share` too. The new rows also use the registered metrics `taxpayer_count` (15), `fiscal_headroom` (4), `decisions_effect_on_borrowing` (2), `real_income_growth` (1) and `revenue_change` (1). ## Family conventions diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/claims_staged.jsonl.gz index 487eaedf483e66e9173eb13fbbf635dc26e2939c..ba5746a19f925fceaa0f8d64c8c890fb53d63867 100644 GIT binary patch literal 17391 zcmV)kK%l=LiwFP!00002|LuL-lG{j<<@tJ_|x_FQQrC-ND!4ayG*+c<4>%PW;98 zW=@je@BCmg;V}sLZUJ37iDK?KNs8C_JdEZ+=%i7MU;Xgz z1AaOQ63?00{43lKKX#Vsavs|5(!b%UZD_`Tu9Ek$zXoCI+;F_|S1*lR9<%%!eCrGR zt7{DLgBu>*#Li+$pa0vx;zvnv6WZ{y_%tbnMU=psf934IGjAE=u$Xu_^Z6w3g7lh& z{Kl!jBw5Cb*)pM9gDit24LrQ*ceQt7S1(vJVKDNS7yki2nR5qj!0j)*B=9+hN%D%n z!zi9Rv*33pdtoPGix|WpCN7LplEw$|5&UY-;rGnrvl$CQ=7fyTUGDqvT`&jHvD9&A z{F;4*%O(MQ`ZY+F&TMwqW1pv-C9}Zi>;^`gurOjtw2VE@;N|K0*X-A~A6ZF{SdhR| zKpkSZ(YMb$LX8ET|$q>J?DO~oYS{3pvZ^MKm~^QcO1doa^KE$yO3%ARW9rc zIkB1I$r-ySUyScjy2sLmm6FZMKQ0{4RQ(rgTFfLUIpv0(n?@9Es~EDQe4*L=>$#%M4Y^@omQ4aS;#kx3w# zavI4`A?SX(OgT<=Hk?oC(hoQ-5BPFMviKW!Kt*C46RL_>|11H4%>NlLT*foJY?`Kv zIHMHXJn%Y^; zz+*|u6MYfVnsd8AQGtqM&tpvplIX85*o^=>GWI{b@a=>9WWCxN+d+5Q`yQZP&q1E(YD0*O} zm7wUMOb>_o4!KZIogVtl>7iBL!`{0NV#Tur7B{GP2+PjRmggD)`~u$=i&x?P$!;fz zNwvkK+AyhhWl~QP{5L6=+DfR$e5+@)U@HMe*%0wj9%?H=0e&m@2+j`h)&9U5;G2hr zfS$$vfY~?#WVR{*wEzBq0{k-!{GT1*tDyfH;2S-?1u=gZWkbM6d8mlL0{m8n^3M(M zRT#f6J9N~7C_ap`adzlP#qgT|zF0!Fd%M|pn>KYnRDY`|;Dn?45-Cr88+WrxsaXCd z;MZDlEI?}v{F;j8E8thr{j&mo71OWD`8QgT^KUfH`8QNdUje_0@}Cv(tJwZ}#^%8% zfuBc0fAgcj-FBS?J|MiWu)h@n|Fgh;73QzS{$U%k^n&!y-qSpUQ3YZdOFZj_&<*v_ZTUYX*P6`#|8pk;wIi@b2IB*LgXgg$FmCJ`jN9cI?7btP$l&aE1J0Zb)Gomq84M5I9)1>^ z1jEML!^2i>5AVN4pvd6tw+GIf4AizkU6`m1rvpZfPX~3~LL;Bz-cE%m;vZwG(+ zLuf^T9c_`8V5*_*0|(J5(b*o-HMs0hEp_%;^p zdHxu$qW0?;Z?z(~ABWi#<1LljuQ0x)b$XsX#;ag_Eyimth{oeEn`6AD;_(eIJ_tR4 zq=X}?A~tl>h~O#4#lIc)apt;F*u{S`Z|a0Myu%b1Jj=MfE#~jpP{MZIQ!)JqD%$)k zLi(y`^RWZ^iX>Fv|2#=TZ3@(se``Z!f4wnD=&G{6A_=u~&|;EMy9%lWuxbH(tX&1g z6KeP2EO|ohOw{s((T-lgMsuDp)Xv0?JRwi%fU3;`#|>t14NvYRj`x36l=190De|ee zOH=;HuXC7{&LUBfG_bkqzGt|CCP4|L9E4mj98h0-ksC^5? z8EOkdafaH)*c_f3wqqM(*gQNnRNEMeHq?%Wq7AjHv6(go?by{AG^dS$+SS;RHUhXN zCOGD4_cn^t>AjrsH+Gf%s>2j{JoCK`nsYmw(C5J#J7}QV-PjbVYQye^)_8YAQ@b0A zJJcSB;tsXT(SSSjR_t=XLasH) z4Wxo`kbGJFvR0+k{>7%QpsnZ((Y!IlDW3d(U=}#6IfP(?V*#}_P{#so=mtG*-bU`Y8Lf^FYZ{;r_gBAid*X=Yu!#wxWtuMIu!acICDMldjU@|!X`fxW3s}=%!8Bff>qBUT1SZfwVXv63X{`G{n4EER z0mN7PbI$uutdKgp4bo}N2(CdM0N}GbT*Y$ggm!n>Eti0X5VlQ@YXrAUz%juoSYbDM zK>Xkgk@AqNG8^a77V~J+*zKF)NwdSw3aDt}-b#R`_24WK!{P$<1MPc!asVi>3#ce6 z1DcK7Q<&D{v&1yO84J9UJ&R6qkkF+^A_Fk6P~$$K8GLg=N&yfIh$UiL(eh?GyE_2F zbu`f}gl!RS(t1X}jMoUk#X!5+_)4(cVf>pnTT&Vvr(d6tD%LLwXqU3*0UPQO{=4ZMG82;W#kmd4k%6V|(zdA8#3x zn#kGoIeI6loe9WBFnk=G27ep_YA#~&%n5w9m^mT9?MHZnq;P(69R(jON^`45!E6q* zay2ozCMs)rPwd?=2fHKi8!(U1q7sR^_<|ougV-lZNoK!*(epNnW3e=n026-=P1lY9 z(z!!g985F%O@Wz2c>AHR-l64$chK?=hZhLy_ak6$3m z-4(h7h(#K99(biUj-sL{o(*1AkOwj_JgZM7zqvl6B<*laC zGU3x`=Cd2z)CzsZXRuOZbnEUAtY8HneGBgUckdYwd3t-AendGFDLCGX z8NMF)EOar;WQTbfk?hry7DwBla)LkV~+*1t6NT&$P)i=;;i{nJ^u%GAXwRl_Q-31hDirq7%u<{n$-E zgj=8o`~5d>%70tJB3wu23&+KGvh9WAeL)b))iqO^uTE;iDDBCT&uo90l7<%GB{QB% zfVGC|fBr!fgznJ)`#((QV~nK!@b<$>2o*&4=O2AH6sVlgtEbbL^8x_y3;Mhy-HJGS zD}rQ(OwYm}-NLXq`(JN9GjEv+LewiW)4}yCqqO*g1%_ZT61W5v2a>^M$>|#bI31P< z2`65(phOJ*&GC?FDwB8x!|G_3HW%+-qdG5lVwlm%n&GBmlD(MZSF`}YNb}_Df=(rw z@&i;)DFxKIb?%baVvVe`#KmX28s20P#c1QL{;uF5v9LkA@8*CV9fm8L;mW3UfoeH{ zTm8<(hVbt^j%d-~hHy0Jlr~hNu!{q@X=^yQ5gFZ+o43M9eJ%S>Z=IUF<$O7#{Er%b zLYbd1KYbtG%wV!sNNOYN;B^3XFq^jXO>WfX>X9tC$Qkyz2RdE4_9l*&3wop3Os;gV zG#44LNjKnLj=FJ>*`h)-lLsKE{j&Y0-0^WjuZd~8#562fE?{k?D|vds@7$-M^w>Mh zAd=dFId~F3SnPyflF~5IUCE@}W8X!vWJ5{}!_CEn0}lY)&hD}oB#=XRDDH4ACx^Mf z8Egt31m=g!baQz0%yIcFc_hE3m=P(V$8lRMMK*hp4OpkK12m#r8e+Pf#gc8cV!qaJ z&B{pgg#m!mMUd$*2}rNk$R_6V2(T4@ElH)sC1oxW97GfZwA6@YF5U=S+HKdc@nG`g z5Q%UvFOKIc!DKY{mEhV(<03BuC%eQeSr}=RQWSC35m(Mjc7I`mD%-HW;6(u%k+tGQ zfN2mMlOre-`v~kMaz`+URS-c9fz+lF{t}gF1vrp0nAj~xrw2GJ@j%_SBz`MaX)&w( z`G--He|PbI1D>;N5wI&j6V}BgPWtk)G16bNLvH@8*uk(GlY?O?Kjlf`zGTjjYl8kW zZSmS^WdD6RyR);A9U0|oA^knvI*W0UzJdWZ3;0Avy10mpEn?o4u#sK+2A(|2gI{R7 zT?@ZQDi%B?iDu}ffMiyKZi*rlMbuG*SyKq+;^Q`tfL(z?iWl9FBM!Rzvmyy|(3m95 zfg*`Dk`NxMoq@pWUiqVv$Xbvqdpljz;2lCgu^r@-V|WWN{%#?#My!%0+U6uuifw`B z%z~Q$2?1dX=>=R?P0K<6iW?L+)N#W|tKkNrIX~vw`MroHeGsZ7%==v#VUK%0D;^j% zk0gyAFOVca0o$Gu(1{IgDJMWi!Zh{<5%S&Pp`?8wK3Xhi1!`6TCdWtX3x_hx0gdv* z18c}FGun!CqoqrWvq#xJL2+p1bx|0n1kHeDTu5B(ZVNmXNmI^To<1TcdV%LU;D;h< z>IUvjVaHeKU%o>7;T0+mn*=umCNW=dM{cQ09HiUeE1=WhFBb%`4;?&uw2kh*g898* zex$vq>g2;?{>B@X5NYMO*Xe-%KIe{_Cg{UlJQFYAUhBRN^A}M>m8DG0F69O9Mgj63 z%v@93XT=5r;&Bjo0yC6BpJeHhun4IGkt{30o)Q(sDfDQWqFfbjw;H#{`U{zt#5S%8 zX1?Fz)X#w+L%NGPD0e)Qq9pEhv4Fxl4Jp8#C0I=PZ)-84*T) zk}T)&`>MZ$=PrN}kS3TH#Q+(LNJQdL$LWW#HpD0BcjU*mlOFrGkU~#0%(-=9Mp|9w zrw0xp<+Ug)Y4h3S+hXz`sl1HhtG&_Y6$cu$S4iC<08*sj=5m zNFG3P=OcK|Bn7<_YIe*}s^W||6&_kCS61~yO0b%_2}nLSqJ&HZM8z{Ms>oK$ zOYmzu`?W284HEhaKLF>1rV@9pn;hqph>2Sq(iz*+b%EK<-4{$WpW}HwXC6_+3Im4O z=!Hikfw3r%Tqjt-v+M|9{n0-`7DDp_s?=HFx2$AVr2G=s`Ku2?G=J!7XOHG1jTGnX z#W{mjhX~JhMQCS{M>OMzy|CyyHH`CV8cb3gT-}n6LI7m4nh3I=_Q1tHE=$vm>4By& z|CwR_44kLOg8UEn19iQe(fX-@e%hWW&4D;*nbV2_^p*oU>VgP@S^(|jKe(LTMzM$h z^je4GO|wv~cH{WUc12wg3Lg{EXm#$ebd10*gn-F!*rEQ>Dr?AT z5=ZbpNS7cl{~=~Z6bRkm9^hAaL=TthS6Eq;%o-sV0L#!MHF6Sq?59+O+F+Q|Y;I=6 zVJr#e!3?Qr9xX9aAWr?uT^FbB&>tNo7qKtJL@j5c4;%Mfh?Qs+{`rS)qMAyz&JXsT zxx2A?j?Qd;jLtl}lABFhuUCthSY4x!0AB93d`~B?qQVXQOS!&{A0%)gp{HOpQb+h- zBKO_OCHQywZ39Zd+t~$=-lo$XYlK#IjlO^X>y{!@+5So{mAP{RTDb&((_e^6?*xn# z8Op?kz6RA^j)7)7RHi>#Vr>`rgqEVi=0UiG`-SZ}mw>~ys?07<4RN)XWp^xz;w1?=@ z)pR5M5%57T?jMFFl*rli39`LF-{zYN6UNtY{OOAK^n%iE8@zgrCgr@JJqQkz;QF_`yyUKTLhY z_O}$djdX=#zBqD`D0cPX-Fvpc2%Nd0i^8?UM@hK{<04tWG~i@!mfC+ZuVkSxj%caF1}Rv?c3Tj7WsORC_=60 zObsAbEb6_xr({r*LlORQNZOUumDenvkYpC!`Vw`ISrlrtttiySlM|UFPl*};1-vOU z$w&O*-G^*;(a21(!w`H7OAYk@-A^A_iq0O{Hb!sw;=zaj%UXoei`)wOlC|XZ1zS8c z-s5^LR8bVVKxLMvB#QEXQeXkP$2AOvThiE-C81UV^LaqL_bLO?VhedjhzK{_zsT%U zLAOYVvZE~zeibM{L@NUBkxrR- z0Xjy2?UumV$k*ma{x#sY*ST7KfvLo+kc1>}t;LG65@j;dk1yJ#7iB^?@_9Q@g;VO8 zdu(4o<|c^^gFu|M;Gv)A&u6B*7*tovin@7$W?qYJhkR!MJp+p9%Ios3z(=493Yz7d zQH0HonZR0NjG&=F}7)+p31c{onk>Oa)P#hkBE;FsybTNWKT zACO)4cSJI|mFec1L{EL#WwSrx1JUC#DcVNn6hbk6jGE@tO6K5ma4-~IJXvgLT#60V zuN?fMcD3(GAD_jFJv>VE2n@2~X^i)AG{(x$*vck93Zg(W`cvcuaMgD*1?`Moj4ZO> zu;Nmd>?B5?*KE$5ii`rX&0^PyyP}S3o?NRO2gL;2m>`P=?+NPli@2oT&&y2!X*9%l zlA8kdI$(cS+5=EXu8>?I`J*7Yei)Kx(fa2K-Fjo_))jPbg6=m@9pMT2jUyvGRd@89 zfdAnFHVWU*7vH-SnLc2Xw{C4eTe$BwkLY(*NdFj{y)Gz^3_WKbC#QqTAE_oOrs+#~ zrjd0ykjo6abxQ5TQ&-pPO|bGlYRpVR`0bmOu3`gaGC}rF$P=EY1mz$XL@Y$OBD%s_Pk^0%jX%uEF%syqz ze&+F+=LFiTIKlj+9CyVy}721tTxq=^~~ z{@)!cC0+K!35jADy|02diZiwZZ&qt11e0RXMZA8^JwW}0^!Y|~R9yv?7XQ#2w?B~Q zOi;0LP|810SpNtK@6S7kqtf10+WQFsyMk;5*^i#X{=5-fWwM_#hVK)oQE0BvycK9} z^z`GS`7;mDqfnZ{Imz!#;JgwgZ*ZJwF>{tlfK36i7%X-Nm@R3otktxq2I4|H^XrrS zO3UXV-^Zx%g`KOiQH!(vxa6?|3F=LZMZU5E{WwH%_&;>!YxRODOkOHn3_pU2P2;6# zJQ5^(G)sS~=(w(x5L?$2WItwn+$gHTAE|oqhkPZkudzL$Ti0H*9|5|(nEP~R>Zdbv z!sMTi9ql&uDhO{pVMX{Un2!-GRcFRZJFJbtysk-9e#eqIF4~9~nyxj4chDak0!(K; z5Y(E!F&NxQmTN1_;L0I7VD8%~xOZX{37angD_vd`RzJRR{G${{aZYq1_j3|fNQDqt zz9^Mw4Dy?Gz{Xf930#8DE=q?bp|eP)QCcY^Pu)UrT~?|FbO2=VsnXiOUW%lb$swHr zx0~p8HZ2PJ&jtG56w0iZREd!)q3xA7daGZChcE=%H{XGBduZFoNxrbes(KQZxVP2d z+G^XYnbex3+iLGmS3$K#35zy0n^kaUga2_W`QjW>yWbK^OYV@B;$`IJHy9s=TVl}K zXc-bw#5{Fi{;4%Nt=ntDoNzhEUWKwqyi9|^Fcj5}pE~hZew+&N%J4s1^#5D5ocYue z`7Tp9uwu1xhbVT$4pH!&fDRiz8NTo61!#MD0opb%VBg-djz14VK#!P$$o9nX(i(V| z_AmUsDTHo}IsU_xc zEJKs`o-JuokNv!yrGe-QQapv-f^NdvSYeGqNv=x{NC1A5oPbhI`OZo=@_ZKpi%qhh z$%SAAm+a_lj5j$OBriLkCYxk&Y@*T2M58OBc|j_T?oQkdxn8t$=5Y@Q?8! z$vkTh*_(9q*R+7kw>k0_fz^~@tojNIPfjf3uBkl;v1B*iPAs0Y2-0e|Sj8%>U=>qS ztkQB;G5W+R83t^m6l^3ez+~T=Yt<*5FrOIxrboj1N9B_pVoZ_0mL!pd3)@(ahsr`q$hn}bBW)CTT!qga#xyx1wyRKpC3T#T=N{F!}@>5cE0cp>T_j-9|e%;Rw? z(sw8@M4QY#K#MTzAO-7}e*?GN1Gu0JCLLFZf%uM>iw0Ao4Y@lN4aNblofPDiu+ON} zY|zt1T}=iu$zCBJJiGeyk3nZV)Yx2(jN%O3f4Ng^%GejQN4200x06^_7%3tAg(HZA zgx7=kU^q5>&$31nOY==@q7N;1;F#XX9FGR3+xNzUu`@JH%NPK}`vbRcj$J-59L+LD zt~=Tv#6QD(NwrM&fLi!@H~V?_)0=f9Esq_6^v^$rE^P%*3RQuR%R%ERfL?4B=k`0@ zkGl8H)hY&0=LnCq3hc;@?R*_J}6yrW}6iOWa8IR4OH~=OV z&K((~sI;H>Y{K3o39Ll=>?#8V6Bha{Gdt`rdS|5f=P!?o?r?KEy<14DPylxE!Vdud z(|PO>>fbnvj0>uOALFoKI>m(6dk_=)01nIo9=DAw_7)V{ z0(x)chA(RfODJNTF!`=DClVk#m>3DBdSUh500<^z zXC-W)Q%3o~z`Vj}?ouol?$Xh${=9cwOfE%ax5iWOI#7%|0@efu4zOkmdbTk>oE#D0 zIFKF8o<7jr?3@qEza}>0#*yO#r^NgQQv>c! zAUs0>NK)aFj)E8`6F`)u2wSfyHyE&<1nD;M#38aG%BUxdL#N<=?l{g9PK-u&-v`iD zn5Ge8V~jf!P-0JFXU=aU7z4ueF9N(E@v!1N<0I+Tdmmxn50(4h^|at|_b|4m!q^S*(RKXTo&^(56Eq9vFcWm};J)6&!fea)OJDk>q(UJ0=<)8Hmjed%ujS)`^FEt5FeH#ht1Y_i$nUL2OrV z-@JPX);Nx?ZCJdAh}c!)V9*n>!mJ17Jq3S1E+2H`gP_?wHZ+@iK(pCOXvUe^37IDn zvTUINws(_0IV69`#zUhiC#qw7Am-gHKp@6hL>&sTJY|f)!~k+CqTwu)FD5m&6E?TW zf?eSh=lGb#f`}69BgNb0+9W94Xm8*ZU8XJs@n2rw#9VvOEyc1L;A)zde0v+K2DFC{r; zC-@x(QyP%uF1M)ZS#;5m&6!8goo#i?c0R615JamM+-cW zE))cn@3RvE6D56e~0VVSe*A5%b zoo~T>ITsh=Ocrx{w+BSl8jbU^(Kx6v8hF1k zg?P7J48%>PsWREw*#diDeA_xDEkiky!E%8rjJw^|oN>pSJ`T#=HxCblb@$eUTHWul zcFh^E=E`Q=)n?4sezaK5a&+6yjC;WIAx5c0Nn5;ej#j(arcMycV;Wh3ENY0hW^2?4 zt|&Yx^Tf{8{LCby;F(=qWy1LLk3K*Uz|hOb!RGrW;nD5T6;ST)hRITRCkD%TMU0_b z9~GSbKD?PZp}!7_AAkI75r&qzcjzaGDN3nS1Ma)X`r2~gr&Ca}X*Bb9fX(bF8?9Oi z-Wau%`G8xuujK7S8nB(XacsxanO(dZe3R|+-vg^nZO&1>&AIX_IDk@U_J{~)by&sv z7OJ1WNy=7tp=9f{xB3M_ zQ0e21tmk5$I7%O;suA~5-m(t(RbJ0nDvH)(C)18qg}Baamb|VCNoYg9?|5KLhUSo4 z{=^+>mOhva2hMQfnYw08#-0V_s9Ah6=nn=?f5bK3?@#=(57&-$b1<3gPnGNClU{>* zz(N$2gcdLqzMVoHFxfn$Pl1qKp=wgNw$*_7ttg>yu#~(ur{)1c>-4H(<;&D3L^@N{^|WEcp>bLw8E%9g%VQH{wn}q zf?d6z!5YVkBc?Vq2G&sT$X-W)DYNMIWyY*sdaO7gE5}qzzkNl&n_*h0&8ns-mMvIz z#wnUnT4`xTs~(U;H&4##Q)PcIs_Z{Ms@#S123=!#;F)^RB|GG|HYTn(x6WMyGA#Fc zDasT!V7?>PIV|yDVfp1RwsD^XNaGSl%kt4)iairi#}<7*0KyJ7QuXmBQpEI4R>D|r zdPwZ>sVwd1j3x5Yx=47Uuoe6i_r1eQa79JcpyKI36JKFoK_%il_Cry>Hxra>7xl9f zds8aROWQ4YvQ|zn@C(QqC!%0PY5)uG6iFZ0K|L=THB909IhJKCnFAvB$i%E9)Tnm- zWib1megFPfna*iqpUPoxz^Xa5)iBBpksaQyg?F+VFER!ipZMi6k(7gkN8kLr0 zNpLH{>R&1%uDjdC4yM5#6M6EM!Cw6rN9S!9)b9EzyFl3m$}V_LcERW@>;l?_QFg&< zBV{XJdKCek0RLCs=j8d{+K2zGr)$3qmJX@ zI5pd(tcUv4NwMv^o*?ARW^5WQ6P~C0Y-!p&+FC*>yNivLBX90(St{JtH4O$5`ibq> zX=Ja|g5#F6$wQ2n2O;1gxXW;t$TQI2*~cE-H4z=5Or0w{&zbfcFgdqGrUgxR8A&-t_z z9eQNKJjL(fizvjtjBMk)OZxyPKZ?`RoiUpWzr@xo$^I$#r$f~m8 zvQqF5F|h1_<@c=PM|lmsblLLS#AM?JUhMsC+}0^b93Yk4lkc0d3eR4%_xB_UfxJG6O;x!K{N^oO)F&s2{&m2><-=w9C3V6t_8ink>;3s0{j|hCeULSq#T!QFvcW&s%X9iuGEVvU=w1kHbRoERn=?&jDTnWKTHSwqKiEz28wzGaT}N#E@E;X}jpj>(+t3d0I3y74+6Vm8_C zT}2!2X<1j?u8v`=g!-8(Zr|enIaqmfvB2lT;>s7(q1(SZ&)5O1w<5R0Nly+nW4v)w z!W)ks-;YC0$~%Z=4ewc=W3-=7w?>Va4AM39ot!-KQ8JIN+*woJtb-nd!+=2Zz+rJT z7ZfEshfwatjomn@dVi^JD$RE(r^@*n1$eLz%t?whVqv3%{7qTXP^Rhy;Y4_pqCOXV z2vi(?4dPM<*WZ>wysf?Jy*%9{q`dXfC3gS3k~NfOCCn&xOq0N_lEA}k zCYpy1n`6SzEck+5vt*He4Rerm!Dl!}!duLK4|b~yFF!{^St<<4ldy*VO-?1vE84@a zG5I;+!u^}viA<^$M{=VNS)?}QBgb!hY_l_+MvLxp(J3cvgG-?2QpREWV02Nk7#R9V zaip-Cxea1Z_Bl>JZFk6(wo+E$?f&#;yc{ zJ5gT-oTo~Gn`r@%F?Na{hn)MORNMpuK@~#CVy5}FA2I=6-TWJie8Uy`49S*{?`z1pIG)|a$#h!2}<{R?e zIG?&h7NLR6bpQ^?Vro-Iri)PPfOc|pHd#_#3fw7LA{W+>V(fbvyRsa%p~Hjm_RceU z(k!m;#p3!yWJozaa3EIVcaTf$i6b1%O>|yVizxs#IhJQlE(V_Y7Oq*+h~$&dStQda zy%yY#ZMce8W1ty60B*UKKtb|Z)%6l<0tkKO_lBPDwF>f0OL<3lB}`Ph%U6iE-dpi__sGgW-N0mw<2I*>25YO^~KphHo?B(PtSQQkv)!7)ojeJD^z8d_VBc#-Fa6I+s+lL*nWRJ_LXK@hizI( zRn1aNdoR<96=?KdWVUoB*<8&zR(Hl_l~v|jMa^TxtM{1LlOkN1a8f2+qu-EeLHK7& zuu+;78WWqjxwrDrRO>r;-8Aa@&Yfag#kPuV4`SObi%AjgUW99DIpLOsTLq}2@aaxd zcj}aCHFaLCN6db@C#x5pbQ6Fd9i&8&<wW>KeOgU< zNy>SjSyt-ytY1Z54U&QH%>e!}m-_d#*x83IDZx1|?w0_(*TkO`^*xik-`b_W_u?WuVB4#U5<~cN1a~`{*oC+Uhxwdj z5>nyrqWuh&P+IjuXx90e3>blm?an?tUMTvBSi$)h}u%crn-3eN?(pEG6G5-FB6`v zoDFdQebI4KY)xl7z#I$bU(y=^F*8D@PVqjM3|@tBOIG&Gl&JEc8v7$xA9Hh{n`lLu{ek08Jb!Fx1J5%CE_cnLHL_gO z&`r;sIF6+ceBT^;MTPR|V&S>u zmC&p%Wqsw(>lV-JisyCld0p|WEQ^DGI6iU`86Ald&AMtGAwg=K=7b${&usA z?opisO7yR$g%<;Enjc;gkus3lXdn%hfusy1WgtD7fi!HXfizSG(qkG(!#V>=$IQ1e z2uvHd!$2wxxS@em7LhWL+Grq+l!2rSBxN8ynSnHFsev?72GV01NTW>#(#V4Um@P7p z3dd9eK493H@*hu++ z8{0^Qh}zbVaQmU!`Yx{5EaW#%{Uu;yEN08(0F&d-KkAEJt(qHxo{9oX@)Rr14i=h{ zRsvzYM4^z*!A@m^9>II~b#2$(M9>71ZsmKTph`E|j%ZOU%Xdlv6bEmU;CG%?Ul#3q z8u!>OcQuwh=`X)qX;mMANr}LfJ=eXR zCjj6+jDwa`A`RT%c^s|1t9->v60G9_vejQF)$Eeq*C~{ZL|+jy-bftR!g)en-s!Z| z?%6L=+q2au-*m4dz>R5m)R=~48|9pa@fM(>Mv?6Ts+_~aIsyBxX?m97dt7sdqls>C zKC!(1$QloQ*B$$YYfd!RbWP3D^wGqh7=wwS^^NhkKLOGn?N7k<8qjIBQX6xwoW1NU zm3&Hfjt>!i7vLe?NIXP*P&}kUxFyMEBE_Syy})9etbV5Gd*_fWD6ITV4yutEDU$oeOMu(GjwDX!zzYV467JcIkFGJ zu@y(wFx#~i3j^MDGgipx9DP|+>zcY^TE(=AX{$`DpFPvk(pR4B(OFi?wdBVddT-bw zKi1HPJM$|_==}Ux#jc876}u{SRbK4z`L((ktF>)6R@;3$Ry))ESX1wshGJR8vWjJ^ zENh%U%i_{kKJ4*1R>-yN!)garop^RWtTx`+ht*D^>I5J{G!MdTHU04J!!bSE8j6iv zERUS?+s{}|cW^)V=DdKFh5=|f5`=u)N&)ENLP-~5EG%Bmuud^>L@=ix=Tf+zoHKc3 z_r5r!@H5sdUVBNre(gn(=j6+B2c(+4aR-Hf<&}wX5bd$fD3)`6%`Shucl@Q>rlrck zm$D$S90bw}Tn~bt#EwT=Bt>4Xcud?$6@ghk{tgNNSB#jb>!LI=uxhxRBWNKiI9RMW zMFN9ahkJxn!dk~A6ophR7BG3n7Bj~ahKs{q0M|_>!8h@y>z0TY0jMPl(0i&cTadZd zvlwt4TqT|Y!>Gu9#cq@41&l*l539yQk9{EK6!n@p)lps*XldXj!%W^9f6!yUR7xtt zN3#H4l9nA92AP)Zl)Sui1NYT0o|LhNqfv>m1vun{{bLCOoHZ0!r*Myt%(2f+XD}XG z#@NBKxdYcShW=Fds^CG7HJHYZmM@2;`_lw<2sDCFA*c zd8x+_MwUe!KbxdnYeeXIV-ujfH&M32JpRB{k*OUJK=LEC{n>>-$NF%kAaZwya>$P6 z^tn6Ge8aW0$#CGh{%~ZD#;)s*_{8jw4eoMoSwn6NtpOjn-1GVa&(tUW0DkK2Pnfw% z$;r*T`joM-5pz-lADEo3Z0JM%dMdVlrKWLrwW2XL=rcxr?9q3!o47E`hX$Ez8X3w_ z@6gCFlT+K8^eCcKM5&1KjEQm;6cklv0xC;J_P;%BC$=Kz=RuS^_4;jDX<|iKz)rW^!sA952_ zuKJKh&uWc#|F$f<2TY7|i_qEtkA#zd)n$W2ta>O)#Rqh&rMjCco%G+I`fS5c#)#@12e9wm97 z6E)&;S3cxMVie-3`jCU3u8%FfWx1^Rs2dYxA*Z%AiBUwUh*A;d855=QAvaOwst*Yy z&@vwqM!X9}wyY|yqDDoH?W0EhoT+iQO1P~f#+q`rx;E?$O>1Nf2d(lU^HDb@$U;tS zYtEvGQW2#h$}=WP0OcI zrjqRyavUGpLi^c~ zg{v&=J1}Mothx!LBj#j#6ejtM+aqfVf%PS(eEE@=t zzws~mEHnzr+9=+zTKN`?g``r@_PYq?3x?&m(K5vmNL8Rm!JnWJ51S=>QgvX*=D510 zDHdzaxj*E}zs|Pffi@aA#@Mwit*`aH0g$e)@!`m_Oha?~{zxB=`$NMsCS!ltAN7%m zNAQD@GxUwVxj);!*>~-Cu$531h#HZy1e>mSU`EYE)%nx9#nZatY2DJ3R(D1-IST<} z?bJigW){=5O4a}f=@fJ?@OGwYb9_Hd3xU*ChlMrb@e==1eORk;WmxOUDf-a5?b~@V z$4&VeVp;x~lUEa`XIwLY+&O_X)xQR!VTB`KLYOkAT{tejiES?&?+fUz4Z35)bVW;; zA|F)Xn+=Xqcq5k9b~*-i23=36Cs#;iyT5HHR9t3gyh=QR2DCEZW;6`%16m>Oawr80 zb}hWiick9e`(Hau=o8D^m>~k7vJp{%YYA!);H*RwY_sId3Z^P9y%O%X;7M8PodDL4 zh%r-!e3Ax}J5eD%o7r4GIK@Opf@J{HyeO8$C$yqVS|LlD7ao89fHdHucbFF~$SnL! z5IE@RmsqYng{R@%VBz;f)SHj1RotuJf~D$7hd$@yOzDs=OWn%kREQ_PX1ThtLKl)2 zRaw`pvOg;=A2BM)ub;QB5q9+=?**_0A70k;v*|mdiKP#${$Sz_bUq&ImgPE5vj9v|k2(R=>H^Ync_fVq zd6qq)E1uBJZ{LO6cjfK7<@Hjf;L2odp9LhYnnN$d7xklpn^f-4k=zikv%U~(VAzP| zkEblhvl2kXk5Ed!)>tqa8w-zo=teri`rt3}-cMUjJ^|6e1V;2p=%eowbN)NOpVy_U zg#9#}Zwu#+$4Ol%JNO#-Fz=w(H=x2WY%rnk(9cA~#^1;2W|4rf0Sfw-(szg3HsSf~ z>F@!N18@AMOEJM~JWnVpnq^`D9E@%Un~z{O)sCiV*iRLdC_%ou!WD)tB{12guJuN2 zE<1c?T}0%r(n?gCghtQkFkKsI+U^}yw3lTobjU4ms7R$_YcYEi91g%fjoWFhk-O%mJ~bQy8CYosw|LKY1q_eUc}wy?&<`IzXG>vd ztoq@VGP&hd&8|vYE1AqUSOBhNYG4JJj%f~caKIf`^FT?R!DL|RV^148d@wY}`otf5 z=8#W3-}S6<-_a(0ojU_}e+%H399a&=kESMpSfAa}^Sb!Fu0H6B$92Wyx>4LM9@oXk zW&S#v4%&u-y!X$+Yc?8WzoEmDoO5qNPM4S_lm|9 zjVl^&4~^HRL?|MEWFoJh1)#6-@OV_dd8(fanS(TTAaj(RBJ+&Q|Azal-+lOrEf=7p zpudXI6{9OgKL2fNX1>XAu6?o4b z_5NTdN4?*QL+Ogi6_YC_S4^$~@2A4$n*;A7O$FXdd_K~4;`5QF_*@0v6{#yyKXX!7 mf%lVP_MHOn!+{FCuLRyH^4>V`4g&m-fBOF*-O|f9xB~z=@}-{u literal 17266 zcmV)rK$*WEiwFP!00002|LuL-lG{j<<@TQa%kM?#|lnW?u5e z^`~B%;_rfRJmv|HGItgSe2PE!<0uI6zIeeu@NOYpI*t?WdufK(1U!moVdQ0Tf?xgc z?gM^04pZNoy8J8L4?p%6*w(=7Dyrr*`xiCw*5@tDELUtatN{A9*GxB+*+@RGo1945&x z{*K~g=1s%jJ@vv~%H|1(K}=i}XC#de;v@LgjKlAl&!2@L0{Ht8p z7jj~i;>j7iC|`{4QM$)s1S>_&%0DhV-{)$5z-Wcadt!1(DNUno4e`@y?5jl}qz!NT zo|Xa25^oF-X4CLylBF&7HU>f7q8PmhKer78n8NLu=6M4Z3~Lb9zH#bfF}W zPB@L^rwDXETVxz3TMg$^wg^H_%LBfgk}UqlJy4MZ$AqdP);~)@AhUnQ3m3^0FPmi9 zJbnG@RX9#NadOjHe0i1Lfo9BJjl(HVUrCX^DsIG^Ik>@hW)gMg9S3+!Qcr30>rO0s`Tyc$BzsQ@0Fx@)}xkcunmr zEAUv5^2A?+v}W8bP*h;#9rqHF5a=PU0-w(bGUsA3;a6Y1=>newqf~#nU&1I5Ko^!S zaue@X$>Tr%=^y994gIz?(A;p=+%VAGz)CAYb3@JzMyIYPNp)dT-O8jMCHSvXF1?jdkNHl=I)kkQ7-d7mM|tS21P%CSxkqqvfUoxlRsi4L zHw5%7_6O|75g_}l0zkX(4`{$Y!NC8?0lp6UuK>Q)G0!0852I`d_$Ut@@z;QVmZAJp z1AHCEugeY{oIw;HM%g$!bf9DSO#ojkq1wG&_1(Hn-4E5@Y6>{wsJ=wXW8cPIuTnae zzYh3~vp5!DGzNY{$MQAs>*)ST0l$vv*W~s6!7cV{%XeN z-Y0>dM?!z|qrmO+It#oqI9pm3%#eT1`{yCNN!8vvLH+K~Cr$+mCs9p&4DBO`Z1#AZ)PdLw7tX{wLKERsL zkF{@whi9=7fWvGW{ISlf#>38A0UF~^uY%9>$9NsyU&VNPPb>bJ#P{vSW;{tmNBA|y zpJUOU=a2C^YQKu{&ROL4<1m|IyrXmbHO8N5ot|fp@j4h^i}A)8MB{Oo%`x84@%RQ9 zA4WbvQpyok5gXcBOz@Q8;@=GWICJedYU4kdKk=d)-eQ^yo@LzL74!FOC}Fei>6rcl z6>WYNA$?u6`OpDpXe^Tg`dG(mNAd@`OC41FAL+y-_%YYj}DualHSlri^F5 zNs&*zTblakK_2((pWdi=nBFtbr3HcRJ@arI6`C{7cF*GUIYaMT)O0WCokjP8UgPcs zJ-u(CIYVz@XwJ~v80*7R{qxwy=r<2f_4PJ}rVYKLp=m?!YOJS?-g)e5^qSK~Pw#4M zNgE+tlMo#9tbH3N+2mf%_#3-YziKf}9?yJlgXY{-6Z$+@V+#$`yBq5wRp+p~VKm;| zF!b(*<_^8bp}9lvax~x$^DK5b%*Nbd>Rk@a9eS@rbBEsTxIb()JdfRuVe_!nQ15nV z`p|nGnm+Wd$Nin6t@G#{g0F*uWmIj^YQ}g1|Ji77Nl`nD> zl#pu;as#Pg5~g34zpPa$y??Q;E9hBthG^ax;uueUKQIgI*BnB#!11!cK(oLq3+SDQ z8W!lCMQ`X{V;1P09SiK&K0>p=(Xzk}+YD#G0(wuOh6T>y0L8HJ`HA5%9-nwv7TB}F zgl2)`WdTdGzzPfKt${igIEQY~!{%+|4x7>H1+b<8`fz{sD|g49SPtu$AYNqIB8h5< zAd54Q6zY6)DyTvANt}egiy?(_7fnX>u30=vdGeLkF~jxB7EwCqKH*?JipRGRVB6>p zFe++;h!B+WrAKf4a8OQg-6mBcIB(Tq_D^c3kTez)jpHN=rlR?8gA`D?DuV}NK%vO} zTV#m{?w3P$hy9X6KdkK)v4#12gG^DeL-=u35d?Ul_>y!I&RH(nnvG{r*N|Yc1y2ia zKqaPd|M)8iJ1I}mmcgRlZUjM7?vCzUm~uCtazPiz&k+q2wv{W*m0TGwHDy$LQCNE= zBh`v`nLUq+na_WS3x7aP0fQAsZNscQ7w@AS; z!6{f`H+n$w;0%%Rh^#Uf=g}4O=+fBToBmO=!`2F@Y2w~WfVT7CERn$C0`&v!dwg;L zC~ym?C@KTmjoVY$&f~MhB*YmD{gOS4PI8#ir3WGdFtAYLKA{n9QOsb@jBebYQqAtGR2ht$+iBgi;FJSb%jgv$yjWoo> zU&}C!1AuhykQN8itZq|aCQ%;2X22d1|F<96a@4;DvjLZrOz8XnGzn4;9@2M56K-@Ru%;>pcjo1U=Nsw*c=1Tp1IWZ-zu zr}%o{v(Uv%$qw@}V&O47Eoge;C0sE^7pLn{k4kFr+zmWYFLPD=^AD$Q?7C#|a=A7S z4!b-UE6t*ZJs|$xjOJzR1%WigUJ~A;1t$*?a*WV(jNp~n43J}zzOF{uvY7qLBH|5! z7Bh(OC+L2D|IMH9-xjc9R$=tQ8{yquckcOL5HWJyQZxM3%Un<~cf8S1?!=699{jePGq|7pmJl@bYu zGZV0gi(ocjhP%NOZ|8bpiXlu zV@0&qW*EzyDgZzoNtxNUnKzB22gdE?FCGk^ZH>=6;q$M^tDuU$d|gmKK;~?ca8On+ zols(>ckA83>izSN?kHNtUU2g9YIqaSNVJ@me^;QpSb?Chcc_yshD(s)5@h~D?Ej77 zR=;zx&H6h}Vp=1(%^J@*CF7Kc;bITO-xwr4EW~#W^`B02tVc;7G1h+{-ArMMmH>Yv zyaN=$9J<~&x#^WFMLOpq|2E)0XmaV=8vxKbz0q_k*SBAq0l5vSOzL|n{0=vvH($U_P5f4VHM(m$WoYH2sy7 zJM6m{7HULER=81@azK9o+3B5nL4q~}h2jp^a&p)Rn1W5f8pm8~nYs;+o_ZrbO&`ez zC}u=T=y6;CM^nsB6a!XZ;sJS>j)6!nXOW>>N!HG?nOMUxOC!w}1^`YQL8iqdAiZ9p z7%+cgz*hXVq>>VYlo?1c4-pTLQ6p}-cq4FWyB)a3gUJ&uB*MMCI2x-$c)teEnOX$w3ebeL zafy?@yljm0*KD7gKPz^yoW|r}IoeNoQn)XfGvu0}*G#*LZWgBVGHR6Tvo$SApp${nj7l4 zVPMp7gV3BGbM5?IM3X)URTAd?wv4dDJ)ad144Oxh29Fm=5}<%BL<#7!39LgUCqPER zH1-A&^4ByVO(>vrimv*^R%qI;*jQAD$l&{ccc!dhX#^DVCNW$mb%VgCSfqirvd81KJ>miTu7>%YV18h{sOsm#WB%G3l@Mv^=+yCm{w`;DnkKu$Ts#vm z;9l##4)f=6OjVj}!!705Zbt#~9?V`*+oNX#0rB|ac!DvKfu3|RA^{Onha%Zjf;}br ziBsJ1B12hf+}60QN?@5UWHu68yC$Tm!OMn*hWCTR`><*AK$x>5!}q!qOGTnQTS}C# z&HmpvP_#KnlN-?aOH^Jh&yip}4E{7-%;5J`e+kbp06idOFh5QJGUl;}#G#JU4`FSH zPtXU+k3gp#_HPk|o~C$6!b=!wb(yOkn3(dqj+GR2HTkZX{6{L)qWNm)bFLz3Hh;t0 zkkV=~%c)ojFqE|kw0`A@`HeQ^&z9E>?l#E1HcF&j+N*xmG^uHqgxS2RfP^RU1{*Og zE$xK}&@8d=j)AEf_OhE67=Xz+HuhQw$pc7keVol3XP|dN&8}HEn@`0l06K3?MSPY@ ze^vdE609n#8Oi5Hlw+xYsCdRj71?Te34ZOWU%TShAfd1D18`1gDsk7k$7>-spE9@$LQpV57PXQs$LfO zEvteRDZj*Z!SaI;%^$jQ*rOfY7--JfiF10(4pEEUPGYpP$RnC^#9ml*ts2JpGzrIe z5UF-a#}NQBSxp4lPkYQ_7nh~!#&plnnE%W$e+tgiV?q9h`+>S%&S?GAKtF9&l;!~U zp|XNjG@ze3prbB`fT#n|PX5D)v)eck5rEFw;ds+5ROh^Ld}+I)uAoDJiDhMXpG1n+}%3Gz}JVrE3K%{A@;euYQ$aH)QU zl|{*{F>(R03{6raC!xcB%0#FQhB;AlGbIjVX*dg~NJX=Ffsuk+Afk!df${1N{qb)e z_~W#$G!vaU6Ma~_=R&MRtMJc1OdHiys&#&_@5J4W<#Tj)^J8@O$(65c(#E_*aB!j1 zwz>%5<-W!DbhRp~#lXLmYsCa%3KtT33PvM!g#V>-->qEEew*JmpcK5FTkz;@Ho4fY z{$^#@===A-ZYbEK_E&PL%)A@W$^`(N`9f4pCt#$=kP;XA8dS$P0h(=5$@_SLmCN80 zT8bWhRYOdjKD%R0ycfBq?boo-`LF-{ z-vuJzAN_lF&zm%{`q+H2;|cMpXaGbnVGi*njECrW)pR5M5%w zXR_W=IazBn$9RE|N$E%Vc-ND`3fTYf_CtMLsA~dQR!;`7j)5Z6|N77W%Rnr?bh%Qt zz(AK;-a+gyD9Z%V9@yd<-udqftfz|;00PD-2Zzv?Kc+4v6iSer+@hk6D6kK35=R^s z`*?ZJQEuNAlv6B4_{sDRkL1x7Io8>SA8b|e!#3A!e@l_ONLMK4ivyR2VpkvDy=QZb zz}ag$g7h83=Du(A9CKvZ*r6U&F3Dq}20#IS!ff&pe|Yyn%`O_58FuS|k722S{=fU_ z1Iy6aBiqL6^j|y}5nx%1NP3YQL0__#yuM(IhsJwcuZ1d#LKmpa^29_@@J|XXMEAIc zp>RtYJDj9crhhgIiT7S95G}Tlr-X=b!`%zEJ{5F3CFpiEbUz++e-FkFAP`y{pnZS- zVVb$7&*G$5Er$a1H-&z$EXW<9+NA zlofTx0?oV@+Yb580D1-#(UsTb9d-{u7ZfzhIim=h9WsHn#29_e7>~yowab;12>)qd z4+W1YT-}i+zN53F>0%@Vx+5|4!Nv#EUMYC`+XC}rQQ@eaQM=lAq>s;H#qJ*^dISbp@-&9K zI2uFkXKZAX9|cjM8T~2tL%8ZYnSyr0E{2NiH>|j%lAXlp^P0`dsZbQ4Hj77I(iU}G z^W<9XIA|u=!~`lDyd$XBFXEDVKQA`{q|p%HNp2e0>wx`TX%9dnxkhr0-ey3NMWZEEOV2iHlSqc#!Ysk)=51pE&du+jK_y7=Cv$n+kYymf2)*}{Fh zc|^ahL;8o<>>Yvf$k4O;I6WRz{z&ynF->2>GmWgxfn288ms4sdp1QhTZGx4@(rfZ6 zo(V1wOSnALxU9EcSCM+Cw_eXarh=E!)=rT(I=U;r(X%qL|DEHmDCNY<)EoN+Ce-7w z@=P`3Ng6sgM(4&nlFffDn%2m1QZ>#w5dgLj!f;d`h|}MXIc}}3cT&|!f^mC6rdbDUCJYE=n zV^D{G%M!1qTF$Y@aPk9Bg{MnJZGa@GMw+O>5d7VvQqpBloRBDi(FZzsqd8+k@MgJI zLO3oKUCis(+ym54$ewRRN7q%*Y4H!ear*;#&IA=}2c`W3jrEU^@cz7mI6CcJr@bE$ zuxrTHkp1X6?9Us)btd~UWB4wC8ja=}&CdeOt&VwkG=Js+dK5}iI4Aj?2%J}<-(Y+gZb?9E<3&V7k?_oe`KQ+8v~I5nbE3rzdlkwe z@iGkt!%$Q^e&Qux`C%%=Ys3F!(f@DpVj56O@r08x`{5sv97V;b4 zLkr$rJBz{Ey(?!bQcKkPft>o$myG;n_d$FQwv4W}FfTM1BRMH0IFf=B{veLf+mn}Yix$^dmZ8ae&z3Z)!+u^&vru#eDW1Y^K{sJ- zEU`wRB-bSeBm}=nPC%)qd~2y2dA`Xf5Yg)kN+Z=d{z-h`bPJM-i zCnuI++cqAASk#R-6N~T7!>rmZRK2vpnn(gDb>BfCGh<>f5LxT zz{*+$@(XW-Pj%h7=YPStAJveA3H>Ie;lcu-s5og%0YCZ2E(eMDnG4)GiW5-+@M5b- zQw=kuaxuR8@n`lqp*Ox`;)SS#ICcVWF`p-yNZ+Br5N$Gd04*ZbK?>F{{|0Wk4{$*l zOeU@n3-KK<7Y(LF8}iXaG#CfG_A-!H%08n~vtGv(bu}5tM7=^jcy{&YAHCMFZ?Ksh z8O0fR@N%ozl(j2p4{AXhZYQyzT*tWjuF_-*z7x_o@e_5 zdpPLXqpm;f4ZXf?J5~=MKIo0Q_He{|mS;HDU^E)+4&tBTy`);EJ3uY`ysduT{`6)Q zNy}qLApP@?{)o1M$AzlE$K{}L6+kaGigWXw?nm8w=V}>)Co_acS_N)AO5A)MtcPUl z(9xrQ9z1FsIs%~|5ftM-avVt<{uz(Wp*R30=H49{q^Pu?_;k$Pq$#XK`s_-9f(;A( zmf0=#7qc}myR(;vMR&Nlo88T&RVVgTUfaCD+jCzP)P9;iHt0W zx!8Tc7baN4=H~nxxRBykg%5Ep9Do#O1w+8_qPLS6-V3j5J6#6SequsEO>TRICWG)W zClIC_^qnV^?+gEi?;a+EK7a$$kSFIx7CQ?HZ2_}07)38@2xKqu-%lY2xcza_i#gbu zUP$)Z)^7mzx8TF!|7Z!|X`%G5@LP7nBM!EH z%5wRoqDCinPR9$v2)C$blCRGRLARBU08XPo^vkdqhF+)W@bFVY*~er>tjEqJGa( z**tNufIEu7e#)duUBhw-!mnX~k?(js4ypbW8Oo%ADWx)?ot!R45~+eGHj00e8Gu2_ zV8TeUR0{Urn9>z_!MQ>C)&kmy@VGZcSe8w#nC~R$7PE*gxfsB_~ug31;Zz|6QEu; z2ECF*LgGMnusdeY7^$N-DF2#p$ta1v05~P#H@I7WcLWg=iV2bmmvj`wI35F{EJP%E zMY+MI^d(526Hn|TE24~g!q|5T?&prfJmJV_Wc$#(sUs(i5F5kJNWQ-(F zJ=rq%@4!H8x7hndLghz%-0qBvSSM~CWw@;h^ABRXdi&p^)(!QYR|d);_1Xtoax&Grt^Y@a1G<4kRZ%wq{zuFwG2zsa5)l0T^N&}hoJ z;}{=Ecv}Ss#5jwnLm`%@j1kxvK+Z%ooGJNYp7D9Y<|bLNE1cpSA5|=fD6u|LyjiYI zg2GMiaI*0CAjj)>FSZgV;qUTG6QXkZXvRHDQfz1{SVQkSa-V(4bqEA>}Oa#DxtJqA-B&fronlZR%?@(;u! zwR*c$Tsf@Cx?*kfh!dA27(RrD=e_3732%Fb0b`Hnct~6*2rA`+faw-9u*-Ro20$v7 z%9_I=sXU#KH*{`{pm62$LeC|zyx&knogf2BrUI^<|C@Q=!r5XbF2tEE=Jsv}h^#dl zXJw43qMpp29-B+cpN`x@u8McCK1r4~%ac zr{v5~j%2WyqYC40w=-kh^Cypka`(-{17Y30Wtx`vdz@`^2Ar9;8Mn0=^R*w(7twPVC3s`hQsx70+`f{x6KTM1I!Y2Z znM~c{)!-Yo%YP57Hnlki^)~0ytKa}iq1gi>oYP_z>s!FCn((tqVuJgV2>$$|-!h%v zk|>riEzGk{k-YdfE_*)bxYsL1WgVsNLi>`UTj%JG=Ho7oo3z*DAod0a$|9zRHZIm zrmQT){BXUf?~;N5s8TerOq8M9*qepZJJGxVB~Z+4 z@pe?Dz_dE0+n1)&UnHpds$)*(CHN145vW(}YVn13F$e?FmogqqRcG@<8@9SN=Jw_- z8K$r_rbXKZ+L?ow@$w_o+ibX1N*AfsT`1Wq?Ja+S5LEhjE$fXiPaLHWQ`Lz3C~w&W z{3@?!tlmUxv8A+QSs|{K&61ZoAqlO?_Z{}Eao_H9Cm4_VhGX`|{hrq!`?hH~R-%4@@>6=~EzN zTc}!_1S;>;&hJ#A(S)O=AFxZM%$N7UkX%Z+Lhez)=mF++nQU=g!ClZ01 zVn1Vrr=(W!2hI1jcWPd2KSUwdbE9QY63xq1UPiho2C7JLJ;}BdaS=`8!z;Wl2x~2S zmM37DKAjSME-|8@=%3zyf)~PuMN5ntP$(f29l1jACD_&bDXeiU0%99|tLOC1mTW%+ zm@i-ad?O~Frb z-#fYlS5y@8DV`2A@fGG3R3f=!KNKZ(m7rt?r=PvVpHQt_+HS#Uq&SVFK6Buu@{l91yWbCT1n2-m$AMgW31&`}e=fbWR&vNA^2Cb|t0s zlE!KShh9lpy<8>lYQ0~OC2n3NO9o1bcFUqm;|}|Uhmgq%GTiu~NwfyHFR?F#kXD@~5FeDyl6H=RW|MK(wBd7a8 ztF*{Ud{S9S#X{0-U$Fgrkxo!*i+M?eqdQQjL1{^r1h*8d{-q+~y1QL$VH)f(kwu*aN2f3?XI7;3$$IJ?SkiI7Yt6qE}&f)Z5J%}Hg@u*R}s(=@PFlfj-LOWUHIR5 zx{kfNqs7%-41NL|m(Vt1)7hez05?wPV;8_1>;>=o8i zL9?KAX_@em6V+r;*JRa|hT`D)AO#(PvU4;l1Od^Twp4Gzf$>66{hxo}6J--+Ic{W8 zj&_qeV_sU|z|njTltOX3!ON1pfQymr1!}*XPfO7>M<&cu{2soDBW#t(*3P@M4{-A1 zBrDw+v$^m~?3R-5o^pSDoo0(&hZz)ZocMbVONTSmToo-32Vq2Nnv0xiA* zewH}v*NC^{@mMYgRh2)Rz|C$Zbk5BWMOO}Nqu%Th4H<5)S>G->YfqB9_pCsUT8ik5 z&Y&T`4eDF59qloJ)F)#iA>T@^5^78ge$}#*tN3K8;2&aO*#XP%S;vp^8hYun<+YB< zh7G*f`#Y+uiMZ+{vMHbuWu;GAR;hyR%#1lc`>NPumPvZ#4>yy}2 zmD|8?-oiy)a+2s@dkkDYT$W!lFZ$xnKyAd1Fy*UFUDGiP@=G~q3L14wZ+8XCi{HMIL<%SOZ99(n_N(DMza?>QsK z@rQok*h6#NwYy#T(6aqQGAFyju)>mVyv~QHCVO;O(T01v$Q8G%W7sO8f@*@aZt?#- zth_m&<8xtg<%?QrL%94gMRWFRj!lM-Rx!^;f;^-@%f<83}GnZ-*mzt{nwg{6= z9Z&D&=_VoNjgKy|`{$*sp-h!9qu4P`0=r5>AG4Wg9$IXM2}9HH3pTP6IY_dzz2GyP zBjGKo--F#cf|sA6p)3`KgAa^e1s??fimk|Vj+hb&T?@{!}W z9k$+?&fjR>U&PuH;GVGEhI!zYqO>dkN|>txq~vOgj?@%FMS{nwCrho+%a_0_$a8(AoZ^BMXGoPRlPkfB z3<7qA#z41AhlB9!e#y`PjT5F`u_s)L`9^#@&ZmZrMQA{|4!{8_rZ(|px(KxnI8TmN zlO@%qz@4H?abXQ<#=e)aE6ZUMIy@L}Z#|<&&En=xEN(tThLjfo2Vy0D54pr0d&1FN zN9RSgm;zAKLwVNZV&Iu?;hLq5NIs3cc{+)+Yr*Zzu9sL7 zKlDj8<*!nuU*Ze(b|?r@Ju%OY zDwRYGv*n~9uJVCE9SU{qs>KZ4@>JeW4NKe(*TUt?mnDODoCdYL4CQQ(RW6 zGT$<49wJ`7!^9pH;mU-QGU-~~hD-~>KUso}GgW9zZ06?P%0p9a;N7*exE*+Rnr$`P zYPQ{rZ8t0?O}IM|u4CkcTM}*+pbo;PTTR`uQ>xR{d37E!`{|ynUU!5k;0$ z-zdeEy=lKY)$HG)&i>WezfYC@3$OQSIprlS=Y6WI)XiDHn!Fk$1K*ng{9`Wl?`g5K z4;xZ~b6ngl0eGj0KPu{bCV9WLM_BV8 zyrp|_ksYw<)kTRR{5gWV8;{*W+`avLPBICpaCgyuhDswZ(q@s__;3YHA<28-z;;1hjMy4Pf>xnOwqKR385yHxY z(RflsaE+)fRcxw@hcETzxFRE<xGPR2L zxn%Gvgj=$*r&6NI54+6gQw%`PxhJClVl+!v$-CcDJ}qs=-NrIsea5k&Cmq_3X<}r> zu`tV1>2MwsT*q+54g3vSoRA%ugI1dyXYR(*`q}Na$0yX;HLTEB@t-@=^PEDzBZ7w zfus$jCo_=xXKEnzwSn}Q22#JyKr%7&Z3qH0hUZ}*l?L3kOoU1OKtl$Uu?<-oijqBGLxZIT}bqZ6Ij_NgGH{W*`mE z)Ib_)1L-jhq+z{*WOi(O*zH>Oxfn?KfEyb~g^09)bdCm+X=n>cTS(eMdMXQP`>uf} zVj-D^Hjy6FL_&Gj*+^E$8g{MTuzxN#Qa<3uHc}y?bL&UA`A}_r7uRbR@f)xH60kAm z(?z<6$?@kO^~J81%?&|MMS&%GiWO%E3r$HYfv{epP(` z`pepOJ|z=@)VlOu`OpOymW?eO+hs8`mu2$=06c(k(2`1}f%`j8;+1!muXss<^+tef z_18%?yQKH^3S}eFSA>i=634Z0o=}&!IxV$*_KVbZwHoD{?sNpWF%1tI)39u#oYOGg z0#wv0vRy!xb9h)K;9z9izGDSGH@yB}Y+9U;9ltwphW%hP8V1(L9vdTjWE+lQ4#vUQ z>WwX8!dCmk|XrrcAO|P0>HN7s;YZV|g!5)NQ zi&uQjY`e2hE<|;Ho&m|)_Hav*B?+Bd_gqc0nr6?IW(TKFvwft3f%ak#PqIp~g%@k~ zJKgi~V$FVcOM)f&oS+x0=~dIKrdLg`+KYV;)s~J`^tY%p4$UX?iRvcN&KCi7<81S~6u|h_t=*!wh+cq`R zYNpjpTV-1F7f&azUjB|p|OJN+~AV=c44HNT>SPS20k?5f#Sv#VxT?ZqCR zU#pw3#<}gr8ryHj8Yj9RYnyG`(k!c4RxnV z1ZMg8TPOfrF=C>wi_*xzs?lPGpoOU5VX@*22@GZ(?h#Q5YZI4H98tAcz~m{LPd#54 zE*^UUTsIzv-^81)S|WZ7pq9=-@2S3QLFPusVZe28m3RsaqaynyyG@!GFb-)wEE^9U z_JNpF)NAHcM|oMGrGb|WGkI(LL5KZPDX9z}O+$D|T6SOhF;+N-nc*N zT9#*z`av(~jmALh!FXWpPT@apK&wXVtfszAzi!_I^D5^m$*dR2YtVVa==IIvKHIgM zMSf^%hS|TLnu$-{FmmzZI&!ABEYou?h=$x;2=4Uwxj`R@C59$jOf$4*XwA@?p_dt2 zfP-e}RfblQxsRb8SWc#9Xeo#-7+MJKyow`Nn=Qe$-8KMg-ed_Dm|vO)>dc`O;lBP$ zq9GDX_*8@!MSG6w{nvL|LbMhgl^&;a(uQ#NHAO$Mt&VJNq#6ZXUE=@qZ*%HFBpcPh zFjEZCqI|+J;AoUh?Y!rhCJr!HO zQq#D-T+tXC^jU*0_UJp>OMqKXNhg?gHLOfL;(&`Ki3)~~; ztbE9P)LS%qRdPDF7CD+IHBoA!JYk~LKIA&8T=pTYj&o){B#d|qigeDbJg}xlO^xSA zjl+|sMqKXNhg?gHLOfL;(&==qUUy&&&dP_(N8Okp3pt%zj~Y#snkY35T@*(q4HzvqJPUqH~MH8haN==j}OqAM(%+_~SsrZon z&fv^^NEq=J6gfD%s_&W{H94LiIa()8j?0zaH94*;*#n~3@{)}VFY@(bQP*$#J^?viH-U7-oNTT_OS~CI zAOkK})C$FVur8<^Y7~KjsdWeBM8BQHisXnX zL0D9ZgJG(HyFi`F6aHGF0alM?e!>If%_XG(4ol)0DaFSh-~Cil2uVA}fW^_RmjI1j z{rbx%0hMBwfsLicfLh_BmQ@7{+gA1EHq0inAZ`<_5p*X|N z8+uJdH+XyYR`$d2?h5;X43gBl19mR2i1e~-AV~hkzvQ#fC@5>Ac*APtTQC;VNb)h$i2SaZ((Ay@uYwjK72LC>>>BgZki zM%V8F>6!-b4;;s~j8QijnEhe5Z~4}E81%b?E;8`|elYO*fz`ElXWKWsuKgCa5~>1G zBT|-N(-sd@)J#;JKdoIntu3C`E~q2pDUp?sGP?n5Jd220+LrpnIXeHBH;Y z`)OJTq^>$FtP!7=_?PO#I*lvCI!{i~`_^sW%!@g0%FhtX^3S}ynm9e^aq~E{)wZ()!vAm5LA^<8I z5f!+Wpavn%N<79kOWw3#s^Zcs;eHFAl%?JYV1t+#Gh@gnSvbBE72?&*=JLTQCNdH% z1DNK=i6lOu6>ZWAS=zkt`11#3As4;F{CG}g;b(%tUdOz|a_t#B4d(_6zo(+!d{V9A zUi}sc$FPNYALsy5W@lS!wx*QAvLNym5`Ns~34M zfDQQYvYwyqz#EJmv*&bsW4~we;m~xP(b%%PJ$vMLyEf0aW||rQ~ak1*@^K@W_X5 zq$6zh{vz-FwB_Ve5FJclOrJyn`aTKgzw`TfUAjuxPs95*_wIO-)`haeuVDc54tjk9 zDh$I06Z#JQOhjz_eS&Tl2?!gYpl>OCcerg6o>xzYH(iJcUg3E{QPC_D1K?nEd)Ryg zyQ#Je!@z#3phPM1)fKKVbSZ(!E={8|U^Cg_Q*{xMyGlz@X%ZSeqs2^PU>MtXR54zb zt&opW11n*k*QIVHDjWcqR)pfo5-Go#@RP|s4m@(IQI%7MYFk(!)t80W%V37YFPH~n z%O)^B**YcVlB@sZXL71eS6}E=O4o~EhD_XTD|ct+<8MMa)Mf# z@RX^0R;jKL-of0ix-6q{Z>9IA`@n{keY*F;eBIv7G6N`kM&` zCf`0hn(h0!@$knLRn_G|Q!hk6%b()<*=?1|&dZ7)W4rEVOI)wj)P?rZv&<#xSio#bB; ziQ3f6?5KLJsC*mI^)rpPe1G_Dt=iw20ybSsEDug*@V9%iZ{f;@Z&yt>XQ#=}5!)QT zi|fUN-@!qvIin_1%#7^Rd)sr~hoy{KGICd_ESjavZYT-Pf zg|iG^LM?Pt{Jw9c^Se{t?`Pkxc&V||_`@aBUPuVUwBPA(IdQ%14m-1Kkv_uPuP5wI zD>?f@{p<^?Cv_%U_W7RJd;eB{UN`@*2>jo%k1ep}1*-d46Eo^Q|hZCrc4``LfkxAOe~>>4U$Vnz4San-5fPU(Wn~=i+;BbDyp#KmRCTS6b28Cm;)7ANako;ODOc zzYkn!{4O+~zh~`_nuH9y>y72UuV!8GpK}FhM#hZV)d#Bg15NO5EbrQ5&bOzT;s4|L P|LpbZm9C0yXJ!BZ9jW1S From 80b23f4dd38bb43c900a863b197e65919760fd54 Mon Sep 17 00:00:00 2001 From: Claude Date: Wed, 4 Nov 2026 18:27:59 +0000 Subject: [PATCH 30/52] harness: apply review A's H-1 to H-4, H-6 and H-7 Contract (pipeline/ab2026_harvest_contract.py): - H-1: a row carrying conditions.component is a part (measure_component, or the OBR's measure_total marker), with a non-blank component. - H-2: value is finite and is value_raw's number at its word's scale, failing that the normalization's factor, failing that a fixed set; its sign is value_raw's unless the normalization documents a worded sign. - H-3: no more than two non-restated rows share the AB2025 claim identity, and a pair agrees within 0.5%. - H-6: registered metrics never in proposed_metric; recosting_of only on recosting rows; fy is one fiscal year and a fiscal_year row has fy or a horizon; geography is a closed code; restated rows have a note; stricter proposed-key slugs; whole-token NOTES scopes; malformed rows reported, not raised; rule 4 case-insensitive and on Figure 1.A/1.B locators. signs.py (H-4): a deciding fragment found only in a negative leg reads unknown (negative_leg), and fragments from both lists read conflicting. None of the harvest's 76 (metric, convention) readings changes. Tests (H-7, T-1, T-2, T-7, RF 3): a mutated-row case for each new check; the Table 6.1 sum bound is GBP 25m and HMT cells are below GBP 10bn; Table 6.2's reprioritisation and total take their proposed keys; RF's fuel-freeze totals keep revenue_change; fiscal_balance_change rows are selected on the metric. Ruff-format the registry builder. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_01LwQq6JvunwtJCAckL6L4aM --- pipeline/ab2026_harvest_contract.py | 245 ++++++++++++++++++++++++-- pipeline/build_ab2026_registry.py | 11 +- scorecard_db/signs.py | 25 ++- tests/test_ab2026_harvest_contract.py | 135 +++++++++++++- tests/test_uk_ab2026_harvest.py | 81 +++++++-- 5 files changed, 461 insertions(+), 36 deletions(-) diff --git a/pipeline/ab2026_harvest_contract.py b/pipeline/ab2026_harvest_contract.py index bbcf7d573..5078cae4c 100644 --- a/pipeline/ab2026_harvest_contract.py +++ b/pipeline/ab2026_harvest_contract.py @@ -10,8 +10,10 @@ import gzip import json +import math import re import sys +from collections import defaultdict from pathlib import Path ROOT = Path(__file__).resolve().parent.parent @@ -65,8 +67,140 @@ # README rule 6: HM Treasury and OBR material is under the Open Government # Licence and quoted whole; every other producer's quote stays under 15 words. OGL = {"hm_treasury", "obr_efo"} -PROPOSED_KEY = re.compile(r"ab2026(_option)?__[a-z0-9_]+") +# A slug of lower-case words joined by single underscores (review A H-6(f)). +PROPOSED_KEY = re.compile(r"ab2026(_option)?__[a-z0-9]+(_[a-z0-9]+)*") +BARE_PACKAGE = re.compile(r"ab2026__package") SHA256 = re.compile(r"[0-9a-f]{64}") +# Rule 9: a component names its part on a part row; OBR workbook totals mark +# themselves `measure_total` (review A H-1). +COMPONENT_ROW_TYPES = {"measure_component", "measure_total"} +# conditions.geography is a closed code (review A H-6(d)); prose goes in +# conditions.geography_note. +GEOGRAPHY = { + "UK", + "GB", + "England", + "Wales", + "Scotland", + "Northern Ireland", + "England and Wales", + "UK_excl_scotland", + "Scotland, Wales and Northern Ireland", + "not stated", + # countries a producer compares the UK with + "France", + "Germany", + "Ireland", + "Italy", + "Netherlands", + "Singapore", + "Spain", + "US", +} +# A `normalization` that documents a sign taken from the wording, not from a +# printed minus (review A H-2(3)). +WORDED_SIGN = re.compile(r"sign from the wording|carried negative|-> -\d|→ negative") +SCALES = (1, 1e3, 1e6, 1e9, 0.01, 100) +WORD_SCALE = ( + (re.compile(r"\b(bn|billion)\b|\dbn\b", re.I), 1e9), + (re.compile(r"\b(m|million|mn)\b|\dm\b", re.I), 1e6), + (re.compile(r"\b(k|thousand)\b|\dk\b", re.I), 1e3), + (re.compile(r"%|per ?cent|percentage point|ppts?\b|pp\b", re.I), 0.01), +) +NUMBER = re.compile(r"(? 0.3333) + return math.isclose(a, b, rel_tol=1e-3, abs_tol=1e-9) + + magnitude = [(n, s) for n in numbers for s in scales if close(abs(v), abs(n) * s)] + if not magnitude: + return [f"value {v!r} is not value_raw {raw!r} at a stated scale"] + if worded or any(close(v, n * s) for n, s in magnitude): + return [] + return [f"value {v!r} has the opposite sign to value_raw {raw!r}"] def _keys(path): @@ -112,7 +246,17 @@ def proposed_keys(rows): def row_problems(r, keys, earlier, metrics, units): - """Every way one staged row breaks the contract, as strings.""" + """Every way one staged row breaks the contract, as strings. A malformed + row is reported, never raised (review A H-6(h)).""" + if not isinstance(r, dict): + return ["row is not an object"] + try: + return _row_problems(r, keys, earlier, metrics, units) + except (TypeError, AttributeError, KeyError, ValueError) as e: + return [f"row is malformed ({type(e).__name__}: {e})"] + + +def _row_problems(r, keys, earlier, metrics, units): out = [] fields = set(r) if not REQUIRED <= fields: @@ -130,13 +274,22 @@ def row_problems(r, keys, earlier, metrics, units): out.append(f"{field} {r[field]!r} not in {sorted(allowed)}") if r["status"] != "ok": out.append("status is not ok") - if not isinstance(r["value"], (int, float)) or isinstance(r["value"], bool): - out.append("value is not a number") for field in ("value_raw", "quote"): if not (isinstance(r[field], str) and r[field].strip()): out.append(f"{field} is empty") + if not all(isinstance(r[f], str) for f in ("value_raw", "quote")): + return out + if not isinstance(r["publication"], dict): + return out + ["publication is not an object"] + if r["measure_key"] is not None and not isinstance(r["measure_key"], str): + return out + ["measure_key is not a string or null"] + out += value_problems(r) if bool(r.get("metric")) == bool(r.get("proposed_metric")): out.append("exactly one of metric and proposed_metric") + if r.get("proposed_metric") in metrics: + out.append( + f"proposed_metric {r['proposed_metric']!r} is a registered Metric; use metric" + ) if bool(r.get("unit_concept")) == bool(r.get("proposed_unit")): out.append("exactly one of unit_concept and proposed_unit") if r.get("metric") and r["metric"] not in metrics: @@ -154,18 +307,26 @@ def row_problems(r, keys, earlier, metrics, units): out.append(f"conditions.fiscal_event is not {FISCAL_EVENT}") if "fy" in c: m = re.fullmatch(r"(\d{4})-(\d{2})", str(c["fy"])) - if not m: - out.append(f"conditions.fy {c['fy']!r} is not YYYY-YY") + if not m or (int(m.group(1)) + 1) % 100 != int(m.group(2)): + out.append(f"conditions.fy {c['fy']!r} is not one fiscal year YYYY-YY") else: if r["period"] != int(m.group(1)) + 1: out.append(f"period {r['period']} is not the end year of {c['fy']}") if r["time_basis"] != "fiscal_year": out.append("an fy row has time_basis fiscal_year") + elif r["time_basis"] == "fiscal_year" and not c.get("horizon"): + out.append("a fiscal_year row has conditions.fy or conditions.horizon") + if "geography" in c and c["geography"] not in GEOGRAPHY: + out.append( + f"conditions.geography {c['geography']!r} is not a code; put prose in geography_note" + ) if not (isinstance(r["period"], int) and 1945 <= r["period"] <= 2075): out.append(f"period {r['period']!r} is not a year") k = r["measure_key"] if "recosting_of" in c: + if c.get("row_type") != "recosting": + out.append("a recosting_of row has row_type recosting (review A H-6(b))") if k not in earlier.get(c["recosting_of"], ()): out.append(f"{k} is not a {c['recosting_of']} key") elif k is not None: @@ -175,14 +336,20 @@ def row_problems(r, keys, earlier, metrics, units): out.append("a row with a measure_key carries no proposed_measure_key") elif c.get("proposed_measure_key"): pk = c["proposed_measure_key"] - if not PROPOSED_KEY.fullmatch(pk): + if not PROPOSED_KEY.fullmatch(pk) or BARE_PACKAGE.fullmatch(pk): out.append(f"proposed_measure_key {pk!r} is not ab2026[_option]__slug") if pk in keys: out.append(f"{pk} is registered; put it in measure_key") if c.get("row_type") == "measure_component" and not ( - c.get("component") and (k or c.get("proposed_measure_key")) + c.get("component", "").strip() and (k or c.get("proposed_measure_key")) ): out.append("a measure_component row names its component and measure (rule 9)") + if "component" in c and ( + not c["component"].strip() or c.get("row_type") not in COMPONENT_ROW_TYPES + ): + out.append("a component row has row_type measure_component (rule 9)") + if r["attribution"] == "restated" and not (r.get("note") or "").strip(): + out.append("a restated row's note names whose figure it restates") pub = r["publication"] for field in ("title", "url", "date", "publisher", "primary_sha256"): @@ -204,8 +371,11 @@ def row_problems(r, keys, earlier, metrics, units): out.append("same_assumptions rows are PolicyEngine-computed") if ( r["source"] == "hm_treasury" - and "Impact on households" in pub.get("title", "") - and r["value_kind"] == "approx_chart_reading" + and "impact on households" in pub.get("title", "").lower() + and ( + r["value_kind"] == "approx_chart_reading" + or re.search(r"figure 1\.[ab]\b", pub.get("locator") or "", re.I) + ) ): out.append("no HMT Impact on households bar readings (README rule 4)") return out @@ -237,16 +407,67 @@ def family_problems(family_dir): for i, r in enumerate(rows): for p in row_problems(r, keys, earlier, metrics, units): out.append(f"row {i}: {p}") - sha = (r.get("publication") or {}).get("primary_sha256") + pub = r.get("publication") if isinstance(r, dict) else None + sha = pub.get("primary_sha256") if isinstance(pub, dict) else None if sha not in set(shas): out.append(f"row {i}: primary_sha256 is not in manifest.jsonl") + out += duplicate_problems(rows) notes = (family_dir / "NOTES.md").read_text() for pk in sorted(proposed_keys(rows)): - if pk not in notes: + if not re.search(rf"(? 2: + out.append(f"rows {idx} share one claim identity (more than 2)") + elif len(idx) == 2: + a, b = (rows[i]["value"] for i in idx) + if not math.isclose(a, b, rel_tol=0.005, abs_tol=1e-9): + out.append( + f"rows {idx} share one claim identity but differ by over 0.5%" + ) + return out + + def main(argv): bad = 0 for d in argv: diff --git a/pipeline/build_ab2026_registry.py b/pipeline/build_ab2026_registry.py index 3ac88d5b3..836ecb7a4 100644 --- a/pipeline/build_ab2026_registry.py +++ b/pipeline/build_ab2026_registry.py @@ -688,6 +688,7 @@ def _cap_history(): # Out of the household model's scope # --------------------------------------------------------------------------- + def _spend(lever): return ( "Public spending ({what}), out of the household model's scope. The engine's spending channel, " @@ -721,7 +722,11 @@ def _spend(lever): DETAINED = "A change to the {what}: the search finds no parameter or variable for hospital detention, prison or offending, so the engine cannot identify the people it affects." SCOPE = { - 9: (SPEND_EDUCATION, "books for children aged 0 to 5", r"early_years|childcare|book"), + 9: ( + SPEND_EDUCATION, + "books for children aged 0 to 5", + r"early_years|childcare|book", + ), 10: ( INDIRECT, "tariffs on food products, heating oil and fertiliser", @@ -1139,9 +1144,7 @@ def registry(): PRE_EVENT = { "ab2026__property_income_tax_rate_rise": "an AB2025 measure (ab2025__property_income_separate_rates: " "22%/42%/47% from 6 April 2027); Table 6.4 line 6 recosts it, and Table 6.1 line 61 raises those " - "rates by a further 2ppts to 24%/44%/49% (" - + K.key(61) - + ")", + "rates by a further 2ppts to 24%/44%/49% (" + K.key(61) + ")", "ab2026__bank_profit_surcharge_increase": "announced: Table 6.1 line 55 (" + K.key(55) + "), temporary, 3% to 6% for 15 months", diff --git a/scorecard_db/signs.py b/scorecard_db/signs.py index d40f0891c..1432d0856 100644 --- a/scorecard_db/signs.py +++ b/scorecard_db/signs.py @@ -32,6 +32,8 @@ from __future__ import annotations +import re + SAME = "same" OPPOSITE = "opposite" MAGNITUDE = "magnitude" @@ -101,6 +103,9 @@ "reduction in public spending", "reduction in welfare spending", "positive = saving to the exchequer", + # a cost saving is a yield, though "positive = cost" also matches: + # the two lists conflict, so such a phrase reads unknown + "cost saving", "positive = saving against the current budget rule", "positive = fiscal tightening", # the OBR's Table 1.1 workbook legend @@ -206,6 +211,10 @@ } _ORIENTED_TO_CLAIM = ("PE oriented to it", "PE's own orientation") +# A phrase's negative leg ("...; negative = loss to households") states the +# same direction from the other side, so a deciding fragment found only +# there says nothing about what a positive value means (review A H-4). +_NEGATIVE_LEG = re.compile(r"\bnegative\s*(?:=|implies|means)[^;]*") MAGNITUDE_CONVENTION = "magnitude_with_direction_in_label" @@ -233,10 +242,22 @@ def classify(row: dict) -> tuple[str, str]: unanswerable, opposite, same = _PHRASES[family] if any(p in text for p in unanswerable): return UNKNOWN, f"stated:{family}:not_this_quantity" - if any(p in text for p in opposite): + positive = _NEGATIVE_LEG.sub(" ", text) + + def hits(phrases, where): + # a phrase that names the negative side itself ("a negative sign + # implies a loss ...") is read where it stands + return [p for p in phrases if p in (text if "negative" in p else where)] + + opp, sam = hits(opposite, positive), hits(same, positive) + if opp and sam: + return UNKNOWN, f"stated:{family}:conflicting" + if opp: return OPPOSITE, f"stated:{family}:opposite" - if any(p in text for p in same): + if sam: return SAME, f"stated:{family}:same" + if hits(opposite, text) or hits(same, text): + return UNKNOWN, f"stated:{family}:negative_leg" return UNKNOWN, f"stated:{family}:unrecognised" # US producers publish signed changes, after minus before, which is # PolicyEngine's reform-minus-baseline orientation; no US adapter diff --git a/tests/test_ab2026_harvest_contract.py b/tests/test_ab2026_harvest_contract.py index 304b1858d..e1d37be09 100644 --- a/tests/test_ab2026_harvest_contract.py +++ b/tests/test_ab2026_harvest_contract.py @@ -74,7 +74,8 @@ def test_a_staged_row_has_no_problems(): ({"metric": "not_a_metric"}, "is not a Metric"), ({"conditions__fiscal_event": "autumn_budget_2025"}, "fiscal_event"), ({"conditions__geography": 1}, "is not a string"), - ({"conditions__fy": "2026/27"}, "is not YYYY-YY"), + ({"conditions__fy": "2026/27"}, "is not one fiscal year YYYY-YY"), + ({"conditions__fy": "2026-29"}, "is not one fiscal year YYYY-YY"), ({"period": 2026}, "is not the end year"), ({"time_basis": "annual"}, "time_basis fiscal_year"), ({"measure_key": "ab2026__not_a_measure"}, "is not in data/uk"), @@ -108,6 +109,47 @@ def test_a_staged_row_has_no_problems(): ({"publication__date": "28 October 2026"}, "is not YYYY-MM-DD"), ({"publication__url": ""}, "publication.url is empty"), ({"attribution": "same_assumptions"}, "PolicyEngine-computed"), + # review A H-1: a component is a part (rule 9) + ({"conditions__component": "Receipts: VAT"}, "a component row has row_type"), + ( + {"conditions__component": " ", "conditions__row_type": "measure_component"}, + "names its component and measure", + ), + # review A H-2: value against value_raw + ({"value": float("nan")}, "value is not finite"), + ({"value": float("inf")}, "value is not finite"), + ({"value": -2625}, "at a stated scale"), + ({"value": 2625000000}, "opposite sign"), + # review A H-6 + ( + {"metric": ..., "proposed_metric": "affected_share"}, + "is a registered Metric", + ), + ({"conditions__recosting_of": "autumn_budget_2025"}, "row_type recosting"), + ({"conditions__fy": ...}, "conditions.fy or conditions.horizon"), + ({"conditions__geography": "England for count; UK funding"}, "is not a code"), + ({"attribution": "restated", "note": ""}, "names whose figure"), + ( + {"measure_key": None, "conditions__proposed_measure_key": "ab2026___x"}, + "is not ab2026", + ), + ( + { + "measure_key": None, + "conditions__proposed_measure_key": "ab2026__package_", + }, + "is not ab2026", + ), + ( + { + "measure_key": None, + "conditions__proposed_measure_key": "ab2026__package", + }, + "is not ab2026", + ), + ({"measure_key": ["ab2026__x"]}, "measure_key is not a string"), + ({"publication": None}, "publication is not an object"), + ({"value_raw": None}, "value_raw is empty"), ], ) def test_each_rule_broken_on_a_valid_row_is_a_problem(changes, expected): @@ -115,6 +157,45 @@ def test_each_rule_broken_on_a_valid_row_is_a_problem(changes, expected): assert any(expected in p for p in found), found +def test_value_raw_scale_words_set_the_scale(): + """review A H-2: "£15 billion" is 1.5e10, never 1.5e7; "N in M" is N/M.""" + bn = mutated(value_raw="£15 billion", quote="£15 billion", value=1.5e10) + wrong = mutated(value_raw="£15 billion", quote="£15 billion", value=1.5e7) + share = mutated( + value_raw="1 in 4", quote="1 in 4", value=0.25, normalization="N in M -> N/M" + ) + worded = mutated( + value_raw="£2bn", + quote="cut of £2bn", + value=-2e9, + normalization="£bn x 1e9; sign from the wording (cut)", + ) + assert problems(bn) == [] + assert any("at a stated scale" in p for p in problems(wrong)) + assert problems(share) == [] + assert problems(worded) == [] + + +def test_malformed_rows_are_reported_not_raised(): + """review A H-6(h).""" + for row in ( + mutated(conditions=["fy"]), + mutated(publication=None), + mutated(measure_key=["x"]), + mutated(quote=None), + "not a row", + ): + assert problems(row), row + + +def test_rule_4_holds_whatever_the_title_case(): + row = mutated( + publication__title="Budget 2026: IMPACT ON HOUSEHOLDS", + value_kind="approx_chart_reading", + ) + assert any("rule 4" in p for p in problems(row)) + + def test_a_component_with_its_measure_or_a_proposed_key_is_valid(): """README rule 9: a part of one measure keeps that measure's key; a part of a measure with no Table 6.1 line keeps that measure's proposed key.""" @@ -152,13 +233,23 @@ def test_only_ogl_quotes_may_run_to_15_words(words, source): assert long_quote == (source not in OGL and words >= 15) -def _family_with_a_proposed_key(tmp_path, notes_extra): +def _write(fam, rows): + buf = io.BytesIO() + with gzip.GzipFile(fileobj=buf, mode="wb", mtime=0, filename="") as gz: + for r in rows: + gz.write((json.dumps(r, sort_keys=True) + "\n").encode()) + (fam / "claims_staged.jsonl.gz").write_bytes(buf.getvalue()) + + +def _family_with_a_proposed_key( + tmp_path, notes_extra, key="ab2026__package_test_scope" +): fam = tmp_path / "uk_test_family" shutil.copytree(FAMILY, fam) rows = read_rows(fam) i = rows.index(BASE) rows[i]["measure_key"] = None - rows[i]["conditions"]["proposed_measure_key"] = "ab2026__package_test_scope" + rows[i]["conditions"]["proposed_measure_key"] = key buf = io.BytesIO() with gzip.GzipFile(fileobj=buf, mode="wb", mtime=0, filename="") as gz: for r in rows: @@ -178,3 +269,41 @@ def test_a_family_scopes_every_proposed_key_it_uses_in_its_notes(tmp_path): "NOTES.md does not give the scope of ab2026__package_test_scope" ] assert family_problems(scoped) == [] + + +def test_a_scope_for_a_longer_key_does_not_scope_its_prefix(tmp_path): + """review A H-6(g): NOTES scopes a key as a whole token.""" + fam = _family_with_a_proposed_key( + tmp_path, + "\n- `ab2026__package_test_scope_wider`: Table 6.1 lines 1-2.\n", + key="ab2026__package_test_scope", + ) + assert family_problems(fam) == [ + "NOTES.md does not give the scope of ab2026__package_test_scope" + ] + + +def _family_with_copies(tmp_path, n, factor=1.0): + fam = tmp_path / "uk_test_family" + shutil.copytree(FAMILY, fam) + rows = read_rows(fam) + i = rows.index(BASE) + extra = copy.deepcopy(BASE) + extra["value"] = BASE["value"] * factor + rows[i + 1 : i + 1] = [copy.deepcopy(extra) for _ in range(n)] + _write(fam, rows) + return fam + + +def test_three_printings_of_one_claim_are_a_problem(tmp_path): + """review A H-3: the AB2025 ingest merges two rows that share a claim + identity and raises on three (red book rows 179, 191 and 200).""" + assert family_problems(_family_with_copies(tmp_path / "a", 1)) == [] + triple = family_problems(_family_with_copies(tmp_path / "b", 2)) + assert len(triple) == 1 and "more than 2" in triple[0], triple + + +def test_a_pair_that_disagrees_is_a_problem(tmp_path): + # a value off by 1% (value_raw unchanged) is also a value_raw mismatch + found = family_problems(_family_with_copies(tmp_path, 1, factor=1.01)) + assert any("differ by over 0.5%" in p for p in found), found diff --git a/tests/test_uk_ab2026_harvest.py b/tests/test_uk_ab2026_harvest.py index 5c3c12c18..2de9349df 100644 --- a/tests/test_uk_ab2026_harvest.py +++ b/tests/test_uk_ab2026_harvest.py @@ -132,15 +132,20 @@ def test_table_61_rows_carry_the_written_out_key_for_their_line(): def test_every_hmt_cell_is_a_multiple_of_5m(): + """Every cell is a multiple of £5m and below £10bn (the largest is Table + 6.1's total of 7,595), so a cell staged at the wrong scale fails.""" for r in rows(HMT): assert r["value"] % 5e6 == 0, (r["conditions"]["table"], r["value_raw"]) + assert abs(r["value"]) < 10e9, (r["conditions"]["table"], r["value_raw"]) def test_table_61_lines_sum_to_the_printed_totals_within_rounding(): """Footnote 1: totals may not sum due to rounding. Every staged cell is a multiple of £5m, so if HM Treasury rounds each line to the nearest £5m, 98 - lines drift at most 98 × £2.5m from the exact total; the observed gaps are - £5m to £10m. Tax plus spend equals the grand total within the same rule.""" + lines could drift up to 98 × £2.5m from the exact total; the observed gaps + are up to £10m for the grand total and £15m for the Tax and Spend totals + (Spend, 2029-30), so the bound is £25m. Tax plus spend equals the grand + total within the same rule.""" t61 = _hmt("6.1") for fy in {r["conditions"]["fy"] for r in t61}: year = [r for r in t61 if r["conditions"]["fy"] == fy] @@ -156,8 +161,9 @@ def test_table_61_lines_sum_to_the_printed_totals_within_rounding(): tax = total[K.package_key("package_total_tax_policy_decisions")] spend = total[K.package_key("package_total_spending_policy_decisions")] grand = total[K.package_key("package_total_policy_decisions")] - assert abs(by_head["Tax"] - tax) <= 98 * 2.5e6, fy - assert abs(by_head["Spend"] - spend) <= 98 * 2.5e6, fy + assert abs(by_head["Tax"] - tax) <= 25e6, fy + assert abs(by_head["Spend"] - spend) <= 25e6, fy + assert abs(by_head["Tax"] + by_head["Spend"] - grand) <= 25e6, fy assert abs(tax + spend - grand) <= 3 * 2.5e6, fy @@ -172,16 +178,36 @@ def test_table_64_recostings_name_their_event(): assert c["row_type"] == "recosting" +# Table 6.2 line 6 reprioritises FT budgets to fund line 5's scheme, and the +# table's total spans lines 1-9: neither is a Table 6.1 measure (review A T-1, +# T-2), so each takes its own proposed key. +T62_PROPOSED = {"6": "ab2026__your_first_home_ft_reprioritisation"} +T62_TOTAL = "ab2026__package_total_financial_transactions" + + def test_table_62_rows_are_financial_transactions_not_revenue(): t62 = _hmt("6.2") - assert len(t62) == 9 * 6 + lines = [r for r in t62 if r["conditions"]["row_type"] == "measure"] + totals = [r for r in t62 if r["conditions"]["row_type"] == "total"] + assert (len(lines), len(totals)) == (9 * 6, 6) for r in t62: assert r["proposed_metric"] == "financial_transaction_cash_impact" assert "metric" not in r - assert ( - r["measure_key"] - == K.PREFIX + K.TABLE_62[int(r["conditions"]["table_line"])] - ) + for r in lines: + line = r["conditions"]["table_line"] + if line in T62_PROPOSED: + assert r["measure_key"] is None + assert r["conditions"]["proposed_measure_key"] == T62_PROPOSED[line] + else: + assert r["measure_key"] == K.PREFIX + K.TABLE_62[int(line)] + for r in totals: + assert r["measure_key"] is None + assert r["conditions"]["proposed_measure_key"] == T62_TOTAL + # the total is the lines' sum within rounding (footnote 1) + for fy in {r["conditions"]["fy"] for r in totals}: + total = next(r["value"] for r in totals if r["conditions"]["fy"] == fy) + summed = sum(r["value"] for r in lines if r["conditions"]["fy"] == fy) + assert abs(total - summed) <= 25e6, fy # --- README rule 10: stated effects of the Budget's decisions on PSNB -------- @@ -200,7 +226,12 @@ def test_table_62_rows_are_financial_transactions_not_revenue(): "package_spending_decisions_since_march_2026", "package_non_scorecard_decisions", ) -} | {"ab2026_option__tax_decisions_with_continued_fuel_duty_freeze"} +} +# RF's totals that add a continued fuel duty freeze's receipts effect keep +# revenue_change: the freeze's effect on borrowing is stated for 2029-30 only +# (OBR para 3.43), and RF's other years are receipts differences +# (coordinator's call on review B's RF finding 3, option (b)). +FREEZE_TOTALS = "ab2026_option__tax_decisions_with_continued_fuel_duty_freeze" # the key of the OBR's total, which the OBR stager also gives the workbook's # Table 1.1 lines by head and its total of receipts measures GOVERNMENT_DECISIONS = K.package_key("package_government_decisions_since_march_2026") @@ -251,6 +282,12 @@ def test_a_totals_lines_by_head_keep_revenue_change(): ), _label(r) +def test_the_fuel_freeze_totals_keep_revenue_change(): + freeze = [r for r in rows("uk_rf_ab2026") if _key(r) == FREEZE_TOTALS] + assert len(freeze) == 11 + assert {r["metric"] for r in freeze} == {"revenue_change"} + + def test_receipts_totals_and_tax_change_estimates_keep_revenue_change(): """README rule 10 (delta review of 8b7b23db, findings 1 and 2): a receipts total or a tax-change estimate keeps revenue_change unless its evidence @@ -283,20 +320,34 @@ def test_a_measure_takes_the_borrowing_metric_only_where_its_producer_says_so(): def test_the_current_budget_memos_name_that_deficit(): + """Every row that proposes fiscal_balance_change names, in its sign, the + deficit it measures (selected on the metric, not the label: review A + H-7). The memos by label are among them.""" + balance = [ + (fam, r) + for fam in FAMILIES + for r in rows(fam) + if r.get("proposed_metric") == "fiscal_balance_change" + ] memos = [ (fam, r) for fam in FAMILIES for r in rows(fam) if re.search(r"memo:.*current budget deficit", _label(r), re.I) ] - assert sorted(Counter(fam for fam, _ in memos).items()) == [ - ("uk_hmt_redbook", 11), + assert sorted(Counter(fam for fam, _ in balance).items()) == [ + ("uk_hmt_redbook", 16), ("uk_obr_tables", 6), ] - for fam, r in memos: + ids = {json.dumps(r, sort_keys=True) for _, r in balance} + assert all(json.dumps(r, sort_keys=True) in ids for _, r in memos) + for fam, r in balance: assert r.get("metric") is None, fam - assert r["proposed_metric"] == "fiscal_balance_change", fam - assert "current budget deficit" in r["conditions"]["sign_convention"], fam + sign = r["conditions"]["sign_convention"] + assert "current budget deficit" in sign or "primary deficit" in sign, ( + fam, + sign, + ) def test_every_borrowing_row_says_which_way_it_runs(): From 6d1b12a3f0b0993ec26c0fbbdd6fd0a3fb93634a Mon Sep 17 00:00:00 2001 From: Claude Date: Wed, 4 Nov 2026 18:28:35 +0000 Subject: [PATCH 31/52] uk_obr_tables: apply review B's OBR 1 to 16 and review A's K-1, K-2, K-5 With the coordinator's calls: OBR 4's estimates listed as not staged with reasons; OBR 5's old row 2771 (GBP 1.9bn financing requirement) restored on cash_requirement_change and the lending/financing rule stated; OBR 10's whole sentence; OBR 11's two Box 3.1 counts staged; OBR 12 option (a); H-6(b): row 2734 is a context row with no recosting_of. Section 3.3: totals 2490-2495 and 2534-2539 are measure_total; their parts are measure_component with the label their primary prints; 2352-2357 and 2527 move whole to the spending- and tax-decisions keys; 2668 (with TPA 30) and 2522 take the brief's components. The corrected 30 October EFO is a 0-row primary. Receipts-only groups take _receipts keys and the OBR's own residual groups package_obr_ keys. 2,797 -> 2,800 rows. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_01LwQq6JvunwtJCAckL6L4aM --- .../uk_obr_tables/NOTES.md | 145 +++++++++++------- .../uk_obr_tables/claims_staged.jsonl.gz | Bin 111493 -> 112860 bytes .../uk_obr_tables/manifest.jsonl | 3 +- 3 files changed, 94 insertions(+), 54 deletions(-) diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/NOTES.md index 2fc483356..13381161f 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/NOTES.md @@ -2,7 +2,7 @@ **Working files.** `stage.py`, the file meant wherever these notes say `stage.py`, lives on branch `uk/ab2026-harvest-staging` at `harvest-staging/ab2026/lanes/obr-tables/uk_obr_tables/stage.py`; this branch carries only NOTES.md, `manifest.jsonl` and `claims_staged.jsonl.gz`. The repository carries no primary bytes: fetch each primary from its `manifest.jsonl` URL (or, while obr.uk refuses non-browser clients, the coordinator's local copies) and check its sha256. -Staged 2026-10-29; reworked 3 November 2026 under the OBR review (see "Review fixes" at the end). Source `obr_efo`. **2,797 rows (2,891 before the review), 3 primaries, 98 individual registered measures plus the registered total-policy package.** All outputs are confined to this family. `stage.py` embeds all reviewed rows as metadata templates and cell data; rerunning needs only the Python standard library and reproduces the gzip without primary files or network access. +Staged 2026-10-29; reworked 3 November 2026 under the OBR review, and on 4 November under reviews A and B (see "Review fixes" at the end). Source `obr_efo`. **2,800 rows (2,891 before the 3 November review, 2,797 at 4f424b9), 4 primaries (rows cite 3 of them), 98 individual registered measures plus the registered total-policy package.** All outputs are confined to this family. `stage.py` embeds all reviewed rows as metadata templates and cell data; rerunning needs only the Python standard library and reproduces the gzip without primary files or network access. ## Access recipe and primaries @@ -10,37 +10,37 @@ Each resolved URL in the supplied OBR `PROVENANCE.md` was attempted once using ` The coordinator supplied the local copies. The lane kept exact byte copies, failed-response headers and bodies and fetch timestamps in its local primaries folder, never committed. Workbooks were read with openpyxl `data_only=True`; the PDF with `pdftotext -layout`. `shasum -a 256` verified all three primaries. Both workbook member hashes match the browser-download provenance from 28 October, including the ZIP's recorded Last-Modified of Wed, 28 Oct 2026 14:05:59 GMT. -The resolved ZIP URL is the publication URL for both members, with the member name in each row's note and the manifest. The ZIP container itself was not supplied locally and is not represented as if it had been read. The PDF attachment URL is absent from PROVENANCE; the blocked landing page prevented discovering it. Its manifest and rows therefore use the EFO landing URL and identify the supplied local PDF filename. The cover/imprint say October 2026; publication date 2026-10-28 follows the EFO-page date recorded in PROVENANCE. No replacement version could be compared after the refusals, and no changed-primary claim is made. +The resolved ZIP URL is the publication URL for both members, with the member name in each row's note and the manifest. The ZIP container itself was not supplied locally and is not represented as if it had been read. The PDF attachment URL is absent from PROVENANCE; the blocked landing page prevented discovering it. Its manifest and rows therefore use the EFO landing URL and identify the supplied local PDF filename. The cover/imprint say October 2026; publication date 2026-10-28 follows the EFO-page date recorded in PROVENANCE. **The PDF changed after the lane read it (README rule 7).** GOV.UK replaced the attachment on 30 October 2026 (change history 2026-10-30T14:21:10Z) with a corrected version (sha256 `555e0c26…`, 1,771,863 bytes), and the OBR's download link (`obr.uk/download/economic-and-fiscal-outlook-october-2026/`) redirects to it. It adds a correction slip at physical p.4, and paragraph 3.19's first bullet now dates the abolition of carbon price support April 2028, where the 28 October text said April 2027. Review B found every other staged PDF quote unchanged in it. Its physical page is printed page + 4; the 28 October file's is printed + 3. The manifest lists it as a fourth primary; every PDF row cites the 28 October version (`9e0e9c34…`), which the lane read. | Primary file | Rows | Bytes | SHA-256 | |---|---:|---:|---| | efo-october-2026-detailed-forecast-tables-policy.xlsx | 2,520 | 106,602 | `cdcc2a06b100241bf3010bd159a4e5d43daef2348a3850797b1e88a308fd45fd` | -| economic-and-fiscal-outlook-october-2026-web-accessible.pdf | 264 | 1,754,344 | `9e0e9c340e35e603066f3124a614fcd68feadb3d5d10d1f76fb536e150e6225e` | +| economic-and-fiscal-outlook-october-2026-web-accessible.pdf (28 October) | 267 | 1,754,344 | `9e0e9c340e35e603066f3124a614fcd68feadb3d5d10d1f76fb536e150e6225e` | | efo-october-2026-detailed-forecast-tables-aggregates.xlsx | 13 | 218,000 | `3ffa3f1e63667b2ae9fb9626dc1da881673227d43aeeb52f8c3db168a5c75d6b` | +| economic-and-fiscal-outlook-october-2026-web-accessible.pdf, corrected 30 October 2026 | 0 | 1,771,863 | `555e0c266a5c2a9a5fdeaee9a36425c6bf3b899fde456160945b5323f610c3f4` | ## Coverage - Policy workbook Table 1.1: **2,520** cells, every numeric row series (420) across all six fiscal years, 2026-27 through 2031-32. This includes 2,112 numbered-measure component/total cells, 48 fiscally neutral cells, 30 non-scorecard cells and 330 summary cells. -- EFO chapter 3: **222** rows (316 before the review): Table 3.1 114 (its "Spending measures" and "Tax measures" rows repeat its "Direct effect" rows and were dropped); Table 3.3 35 costings plus 7 categorical uncertainty ratings; policy/behavioural prose and Box 3.1 66. Table 3.2's 47 funding cells are design (rule 8) and are not staged. +- EFO chapter 3: **225** rows (316 before the 3 November review): Table 3.1 114 (its "Spending measures" and "Tax measures" rows repeat its "Direct effect" rows and were dropped); Table 3.3 35 costings plus 7 categorical uncertainty ratings; policy/behavioural prose and Box 3.1 69 (paragraph 3.20's financing requirement restored and Box 3.1's two measure counts added on 4 November). Table 3.2's 47 funding cells are design (rule 8) and are not staged. - Headline fiscal context: **55** rows: 13 workbook cells (PSNB cash six years; ex-BoE PSND/GDP seven years) and 42 PDF cells (Table 7.1 headroom 13; Annex A Table A.9 PSNB, PSNFL and ex-BoE PSND 29). The current-budget and PSNFL rules include current pre-measures, post-measures and the printed memo excluding fuel-duty rises; welfare-cap margin is included. Forecast duplicates between the PDF and aggregate workbook are represented once. ## Conventions - **Values and precision:** policy workbook values use their displayed integer £million formats (`0`, `#,##0`; eight exact General-format zeros), multiplied by 1e6. Unprinted cached decimal precision is not a claim. Decimal rounding is half away from zero; a stored negative fraction that displays zero is `0`. A LibreOffice PDF rendering independently confirmed these display conventions. Aggregate workbook values use displayed one-decimal precision. PDF £billion values multiply by 1e9 and £million by 1e6; no ratios, sums or inferred annual paths are staged. -- **Signs:** policy workbook prints “This table uses the convention that a negative sign implies a loss to the Exchequer (and is therefore an increase in PSNB).” PDF Tables 3.1/3.3 print “A positive sign implies an increase in borrowing.” Both are retained. The indirect-effect workbook row's conflict with that workbook legend is explicitly noted in those rows. Prose uses positive magnitudes for costs, yields or behavioural erosion exactly as worded; per-row conditions state the meaning. Funding amounts are positive cost provision, except the printed negative change. Headroom is positive margin, negative breach. -- **Period:** FY end year, with the printed `fy`; quarterly CPI peaks use calendar-year `point_in_time` and quarter detail without `fy`. An annual average or recurring prose amount appears once at the terminal forecast FY, with its interval in `aggregation`, rather than manufacturing six yearly values. The 5,500 previously announced HMRC recruits have no separately printed delivery year and are an announcement input at point-in-time 2026; the separate 670 figure retains its printed 2029-30 horizon. Debt stocks preserve end-March timing and end-March-centred GDP denominators. +- **Signs:** policy workbook prints “This table uses the convention that a negative sign implies a loss to the Exchequer (and is therefore an increase in PSNB).” PDF Tables 3.1/3.3 print “A positive sign implies an increase in borrowing.” Both are retained. The indirect-effect workbook row's conflict with that workbook legend is explicitly noted in those rows. Prose uses positive magnitudes for costs, yields or behavioural erosion exactly as worded; per-row conditions state the meaning. Headroom is positive margin, negative breach. +- **Period:** FY end year, with the printed `fy`; quarterly CPI peaks use calendar-year `point_in_time` and quarter detail without `fy`. An annual average or recurring prose amount appears once at the terminal forecast FY, with its interval in `aggregation`, rather than manufacturing six yearly values. Staffing figures are design (rule 8) and sit in `conditions.staffing`, not in rows. Box 3.1's undated share and counts sit at the publication-year anchor, 2026. Debt stocks preserve end-March timing and end-March-centred GDP denominators. - **Central:** ordinary policy tables/prose are `point` estimates of the single final forecast; aggregate forecasts are `central`, meaning the OBR central forecast, and observed 2025-26 outturns are `point`. No statistical confidence interval is inferred. “High” ratings use categorical `value: 0` as the precedent's schema placeholder, not a numerical score. Health Foundation range endpoints span alternative reform options, not uncertainty for one policy. -- **Attribution:** 2,774 `own`, 23 `restated` (2,821 and 70 before Table 3.2's 47 rows went). OBR-certified/HMRC/DWP-scrutinised costings follow the AB2025 `own` convention; OBR fiscal judgments use `obr_efo_forecast`. Twelve workbook fuel cells and two PDF fiscal fuel costings are restated `hmt_costing_uncertified`: EFO p.59 footnote 1 says the final Government costing arrived after certification. Table 3.2's 47 funding cells name HM Treasury as original source; six latest ONS outturn aggregate cells name ONS. Two social-care options name Health Foundation; one FTA effect names the Government/DBT impact assessment. OBR's CPI judgments remain its own estimates. Funding/ONS facts use `administrative_fact`; modeled estimates use `different_model`. -- **Scope:** all 98 numbered workbook titles match registry titles after whitespace/footnote normalization. Fiscal-head components retain their full measure's key, with component conditions; totals and components must not be summed as independent estimates. Only workbook row552 and the PDF's explicit HMT-policy-table memo use the registered total-policy package. All other groupings use null plus a proposed key. Context levels use null without a proposed measure. The lone prior-policy salary-sacrifice recosting stays null as assigned; it is not recast as an AB2026 intervention. The AB2025 OBR family actually contains no headline fiscal aggregates; the live `Metric` enum supplies `fiscal_headroom`, and cross-family proposals supply the closest aggregate names. -- **Quotes/order:** full numeric/rating cells or full OGL prose sentences, preserving the printed number; PDF whitespace/line wraps normalized, footnote superscripts excluded from sentence text. Reading order is policy workbook row/column, PDF page/position, then aggregate workbook table/row/column. PDF printed page + 3 is the physical PDF page. +- **Attribution:** 2,777 `own`, 23 `restated` (2,821 and 70 before Table 3.2's 47 rows went on 3 November). OBR-certified/HMRC/DWP-scrutinised costings follow the AB2025 `own` convention; OBR fiscal judgments use `obr_efo_forecast`. Twelve workbook fuel cells and two PDF fiscal fuel costings are restated `hmt_costing_uncertified`: EFO p.59 footnote 1 says the final Government costing arrived after certification. Six latest ONS outturn aggregate cells name ONS. Two social-care options name Health Foundation; one FTA effect names the Government/DBT impact assessment. OBR's CPI judgments remain its own estimates. ONS facts use `administrative_fact`; modeled estimates use `different_model`. +- **Scope:** all 98 numbered workbook titles match registry titles after whitespace/footnote normalization. Fiscal-head components retain their full measure's key, with component conditions; totals and components must not be summed as independent estimates. Only workbook row 552 and the PDF's explicit HMT-policy-table memo use the registered total-policy package. All other groupings use null plus a proposed key. Context levels use null without a proposed measure. Paragraph 3.28's salary-sacrifice revision is a context row with a null key (4 November, the coordinator's call on review A's H-6(b)); it is not recast as an AB2026 intervention. The AB2025 OBR family actually contains no headline fiscal aggregates; the live `Metric` enum supplies `fiscal_headroom`, and cross-family proposals supply the closest aggregate names. +- **Quotes/order:** full numeric/rating cells or full OGL prose sentences, preserving the printed number; PDF whitespace/line wraps normalized, footnote superscripts excluded from sentence text. Reading order is policy workbook row/column, PDF page/position, then aggregate workbook table/row/column. PDF printed page + 3 is the physical page of the 28 October file that rows cite (+ 4 in the 30 October correction). ## Coverage by measure_key, including null -100 keys on 2,207 rows, 1,621 of them `row_type: measure_component`. Of the 590 null rows, 535 carry a proposed key and 55 are context (fiscal aggregates, headroom and ratings). Row 2828's recosting key is AB2025's (`conditions.recosting_of: autumn_budget_2025`). +99 keys on 2,207 rows, 1,577 of them `row_type: measure_component`. Of the 593 null rows, 535 carry a proposed key (374 of them `measure_component`), and 58 are context: 42 fiscal aggregates, 13 headroom rows, Box 3.1's two measure counts and paragraph 3.28's salary-sacrifice revision. Across the family, 1,951 rows are `measure_component`, 649 `measure_total` and 200 carry no row type. The uncertainty ratings are keyed to their measures. | measure_key | Rows | |---|---:| -| `ab2025__salary_sacrifice_pension_nics_cap_2000` | 1 | | `ab2026__advance_corporation_tax_abolished` | 12 | | `ab2026__alcohol_duty_small_producer_relief` | 12 | | `ab2026__artificial_intelligence_package` | 18 | @@ -49,7 +49,7 @@ The resolved ZIP URL is the publication URL for both members, with the member na | `ab2026__benefits_in_kind_payrolling_delay` | 36 | | `ab2026__benefits_removed_hospitalised_serious_offenders` | 18 | | `ab2026__british_business_bank_backing_innovation` | 18 | -| `ab2026__british_industrial_competitiveness_scheme` | 24 | +| `ab2026__british_industrial_competitiveness_scheme` | 25 | | `ab2026__bus_fare_cap_2_england_outside_london_2027` | 18 | | `ab2026__business_rates_ev_charging_relief` | 48 | | `ab2026__business_rates_pubs_relief_35pc` | 48 | @@ -140,43 +140,43 @@ The resolved ZIP URL is the publication URL for both members, with the member na | `ab2026__writing_down_allowance_12pc_and_fya_50pc` | 43 | | `ab2026__your_first_home_equity_loan` | 19 | | `ab2026__youth_guarantee_and_apprenticeships_spring_2026` | 18 | -| `null` | 590 | +| `null` | 593 | ## Proposed measure keys and exact producer scopes -All of these rows have `measure_key: null`. Names follow README rule 2 as harmonised on 3 November (review finding 9): a single measure is `ab2026__`, and a scope package's slug names the scope and carries no `obr`. Where the OBR publishes only a named residual group, its line decomposition stays unresolved; no guessed membership or registry key is attached. +All of these rows have `measure_key: null`. Names follow README rule 2 as harmonised on 3 November (review finding 9): a single measure is `ab2026__`, and a scope package's slug names the scope and carries no `obr`, except the OBR's own residual groups, which keep it (4 November, review B finding 15). Where the OBR publishes only a named residual group, its line decomposition stays unresolved; no guessed membership or registry key is attached. | Proposed key | Rows | Scope | |---|---:|---| | `ab2026__departmental_spending_growth_2031_32` | 13 | No Table 6.1 lines. “Departmental spending: Grow day-to-day spending by 0.7% in real terms in 2031-32, matching 0.7% average annual real terms growth from 2029-30 to 2030-31”. PDF describes the same baseline change as a £5.5bn reduction in the 2031-32 pre-measures RDEL assumption. (Was `ab2026__package_obr_departmental_spending_growth_2031_32`.) | -| `ab2026__package_carbon_price_support_and_cfd_levies` | 3 | Table 6.1 line 71 plus funding for Low Carbon Contracts Company to reduce CfD levies (no separately identifiable line in Table 6.1); related support for BICS/Supercharger includes line 70. OBR paragraph 3.19 combines these measures and dates CPS abolition April 2027, while HMT line 71 says April 2028. (Was `ab2026__package_obr_carbon_price_support_and_cfd_levies`.) | -| `ab2026__package_civil_service_transformation_programme` | 1 | Table 6.1 lines 63 and 64 net programme: “civil service transformation programme ... reduce resource spending”. (Was `ab2026__package_obr_civil_service_transformation_programme`.) | +| `ab2026__package_carbon_price_support_and_cfd_levies` | 2 | Table 6.1 lines 70 and 71: row 2673's £0.4bn is their receipts (`measure_component`, excluding the backdated AME support of footnote 9), and row 2677's £0.5bn their totals. The CfD-levy funding sits in line 70: its "Environmental levies" receipts (−480 / −63 / −130 / −408 / −450) are the only environmental-levies effect of any decision (review B finding 7). Footnote 9's £0.8bn of backdated BICS and Supercharger support is line 70's current AME in 2027-28 and is keyed to line 70 as a component. The 28 October PDF dated CPS abolition April 2027; the OBR's 30 October correction moved it to April 2028, matching line 71. (Was `ab2026__package_obr_carbon_price_support_and_cfd_levies`.) | +| `ab2026__package_civil_service_transformation_programme` | 1 | The OBR's “civil service transformation programme ... reduce resource spending”. Membership of Table 6.1 lines 63 and 64 is probable but not established: the OBR names no lines, and lines 63 and 64's totals average £0.56bn from 2028-29 against its £0.5bn (review B finding 14). (Was `ab2026__package_obr_civil_service_transformation_programme`.) | | `ab2026__package_compliance_measures_announced` | 1 | No Table 6.1 line list. Box 3.1 (p.66): "Of the compliance measures announced in this Budget". The OBR does not list them; they are probably Table 6.1 line 45's items and line 46, but that is not stated, so the scope is the OBR's words. | -| `ab2026__package_corporation_and_bank_tax_changes` | 7 | OBR “Changes to corporation and bank taxation”; paragraph3.16 explicitly discusses line58 WDA/FYA, line55 bank surcharge, and line54 overseas sovereign immunity. Complete Table6.1 line decomposition and fiscal-head allocation are not published; unresolved, no inferred additional membership. (Was `ab2026__package_obr_corporation_and_bank_tax_changes`.) | -| `ab2026__package_defence_investment_plan` | 1 | OBR “funding for the defence investment plan” / “the defence investment plan, which increases departmental resource spending (RDEL)”; related to Table6.1 Defence Investment Plan entries, but complete line decomposition is not published and is unresolved. No inferred40–43 membership. (Was `ab2026__package_obr_defence_investment_plan`.) Shared with the red book's Defence Investment Plan row. | +| `ab2026__package_corporation_and_bank_tax_changes` | 7 | OBR “Changes to corporation and bank taxation”; paragraph 3.16 explicitly discusses line 58 WDA/FYA, line 55 bank surcharge, and line 54 overseas sovereign immunity. Complete Table 6.1 line decomposition and fiscal-head allocation are not published; unresolved, no inferred additional membership. (Was `ab2026__package_obr_corporation_and_bank_tax_changes`.) | +| `ab2026__package_defence_investment_plan` | 1 | OBR “funding for the defence investment plan” / “the defence investment plan, which increases departmental resource spending (RDEL)”; related to Table 6.1 Defence Investment Plan entries, but complete line decomposition is not published and is unresolved. No inferred40–43 membership. (Was `ab2026__package_obr_defence_investment_plan`.) Shared with the red book's Defence Investment Plan row. Row 2522's £1.8bn a year is the package's RDEL part (`measure_component`, review A's K-2): it matches gross RDEL (line 40's averages £1,766m), not the lines' net effect. | | `ab2026__package_dividend_and_capital_gains_tax_changes` | 6 | Table 6.1 lines 26,49 and 60: OBR “Changes to dividend and capital gains taxation”. (Was `ab2026__package_obr_dividend_and_capital_gains_tax_changes`.) | -| `ab2026__package_fuel_duty_and_electricity_vat` | 11 | Table 6.1 lines 2 and 3 only; OBR “Fuel duty freeze and VAT off domestic energy bills”. Excludes household council-tax rebate line 1. (Was `ab2026__package_obr_fuel_duty_and_electricity_vat`.) | -| `ab2026__package_fuel_duty_freezes_since_2010_11` | 1 | OBR cumulative 2010-11 to 2027-28 fuel duty freezes and 5p cut relative to stated RPI uprating; includes Table6.1 line3 and earlier freezes since2010-11. Not the cost of the AB2026 measure alone. (Was `ab2026__package_obr_historical_fuel_duty_freezes_and_5p_cut`.) | -| `ab2026__package_government_decisions_since_march_2026` | 347 | Table 6.1 lines 1–98, plus the three fiscally neutral measures (PWLB debt relief, Scottish APD devolution, landlord registration fee) and the two non-scorecard policies listed above. OBR “Total effect of Government decisions”, with direct, indirect, receipt, expenditure and fiscal-head projections distinguished in conditions. Includes indirect effects where stated; not the HMT scorecard-only package. (Was `ab2026__package_obr_government_decisions_since_march_2026`.) Shared with the red book's Table 1.3 "Total effect of policy decisions" and RF's restatement. | -| `ab2026__package_national_wealth_fund_and_british_business_bank` | 4 | Table 6.1 lines 29 and 30 together, plus their associated financial transactions in Table 6.2; OBR “funding available for the National Wealth Fund and the British Business Bank ... to invest in financial assets”. (Was `ab2026__package_obr_national_wealth_fund_and_british_business_bank`.) | +| `ab2026__package_fuel_duty_and_electricity_vat` | 11 | The receipts of Table 6.1 lines 2 and 3 only (Table 3.1's tax groups are receipts groupings, review B finding 3); OBR “Fuel duty freeze and VAT off domestic energy bills”. Excludes household council-tax rebate line 1. (Was `ab2026__package_obr_fuel_duty_and_electricity_vat`.) | +| `ab2026__package_fuel_duty_freezes_since_2010_11` | 1 | OBR cumulative 2010-11 to 2027-28 fuel duty freezes and 5p cut relative to stated RPI uprating; includes Table 6.1 line 3 and earlier freezes since2010-11. Not the cost of the AB2026 measure alone. (Was `ab2026__package_obr_historical_fuel_duty_freezes_and_5p_cut`.) RF's `ab2026_option__fuel_duty_rpi_uprating_since_2011` is the same pair of paths seen from the other side (RPI uprating since January 2011 against the freezes); the two keys stay separate, and each family's NOTES names the other (review A's K-6). | +| `ab2026__package_government_decisions_since_march_2026` | 340 | Table 6.1 lines 1–98, plus the three fiscally neutral measures (PWLB debt relief, Scottish APD devolution, landlord registration fee) and the two non-scorecard policies listed above. OBR “Total effect of Government decisions”, with direct, indirect, receipt, expenditure and fiscal-head projections distinguished in conditions. Includes indirect effects where stated; not the HMT scorecard-only package. (Was `ab2026__package_obr_government_decisions_since_march_2026`.) Shared with the red book's Table 1.3 "Total effect of policy decisions" and RF's restatement. **Parts and wholes (rule 9, 4 November):** the totals (workbook row 549 and Table 3.1's total) are `measure_total` with no `component`; every part is `measure_component`, with `component` the label its primary prints: the workbook's direct and indirect effects, its total of receipts measures ("Total effects of receipts measures", which keeps `revenue_change`) and its 288 lines by head ("Receipts: " or ": "), Table 3.1's direct and indirect effects, and paragraphs 3.2 and 3.5. Never sum a part with the total. The workbook's total of expenditure measures moved to the spending-decisions key and paragraph 3.4's tax line to the tax-decisions key (review B finding 6). | +| `ab2026__package_national_wealth_fund_and_british_business_bank` | 5 | Table 6.1 lines 29 and 30 together, plus their associated financial transactions in Table 6.2; OBR “funding available for the National Wealth Fund and the British Business Bank ... to invest in financial assets”. Paragraph 3.20's effects: the financing requirement (`cash_requirement_change`, restored 4 November), the net debt effect (£3.4bn) and the borrowing effect (£0.5bn) are wholes; the gross debt (£3.9bn) and the additional income (£0.6bn) are their parts (`measure_component`). (Was `ab2026__package_obr_national_wealth_fund_and_british_business_bank`.) | | `ab2026__package_non_scorecard_decisions` | 12 | No Table 6.1 lines. “Direct effect of non-scorecard decisions”: exactly the departmental spending assumption and public-pension valuation changes above. (Was `ab2026__package_obr_non_scorecard_decisions`.) | -| `ab2026__package_other_ame_spending_decisions` | 6 | OBR Table 3.1 “Other AME spending decisions”; excludes net unfunded pensions valuation. Producer does not supply a full Table 6.1 line-number decomposition in PDF. (Was `ab2026__package_obr_other_ame_spending_decisions`.) | -| `ab2026__package_other_departmental_spending_changes` | 7 | OBR Table 3.1 “Other departmental spending changes”, excluding council tax rebate line 1; includes defence, capital spending and the £5.5bn 2031-32 pre-measures RDEL reduction, which has no Table 6.1 line. Producer does not supply a full line-number decomposition in PDF. (Was `ab2026__package_obr_other_departmental_spending_changes`.) | -| `ab2026__package_other_hmrc_compliance_and_debt` | 1 | OBR paragraph 3.13 residual “Other measures”: direct debt recovery, e-invoicing, debt collection agencies (components of Table 6.1 line 45), and online marketplace liability (line 46). (Was `ab2026__package_obr_other_hmrc_compliance_and_debt`.) | -| `ab2026__package_other_tax_measures` | 6 | OBR Table 3.1 “Other tax measures”; residual of Table 6.1 tax measures after tax administration/compliance/debt, corporation/bank, savings/property, dividend/CGT and fuel/electricity VAT groups. PDF does not enumerate a full line-number decomposition. (Was `ab2026__package_obr_other_tax_measures`.) | +| `ab2026__package_obr_other_ame_spending_decisions` | 6 | OBR Table 3.1 “Other AME spending decisions”; excludes net unfunded pensions valuation, and is net of the council tax rebate shown on its own line (−£2.6bn in 2026-27): Table 6.2 prints "Other spending decisions" at 0.4 in 2026-27, the rebate's 2.6 plus this group's −2.2 (review B finding 9). (Was `ab2026__package_other_ame_spending_decisions` until 4 November, review B finding 15.) Producer does not supply a full Table 6.1 line-number decomposition in PDF. (Was `ab2026__package_obr_other_ame_spending_decisions`.) | +| `ab2026__package_obr_other_departmental_spending_changes` | 7 | The OBR's Table 3.1 group: departmental spending changes excluding the pensions valuation, including the DEL funding for line 1's council tax rebate (the group, 5.4 / 1.2 / 1.5 / −0.1 / −0.2 / −6.0, equals the workbook's DEL total less the pensions valuation's DEL, which includes line 1's −2,627; review B finding 9). It includes defence, capital spending and the £5.5bn 2031-32 pre-measures RDEL reduction, which has no Table 6.1 line. (Was `ab2026__package_other_departmental_spending_changes` until 4 November, review B finding 15.) Producer does not supply a full line-number decomposition in PDF. (Was `ab2026__package_obr_other_departmental_spending_changes`.) | +| `ab2026__package_obr_other_hmrc_compliance_and_debt` | 1 | OBR paragraph 3.13 residual “Other measures”: direct debt recovery, e-invoicing, debt collection agencies (components of Table 6.1 line 45), and online marketplace liability (line 46). (Was `ab2026__package_obr_other_hmrc_compliance_and_debt` until 3 November, and `ab2026__package_other_hmrc_compliance_and_debt` until 4 November.) | +| `ab2026__package_obr_other_tax_measures` | 6 | OBR Table 3.1 “Other tax measures”; residual of Table 6.1 tax measures after tax administration/compliance/debt, corporation/bank, savings/property, dividend/CGT and fuel/electricity VAT groups. PDF does not enumerate a full line-number decomposition. (Was `ab2026__package_obr_other_tax_measures` until 3 November, and `ab2026__package_other_tax_measures` until 4 November.) | | `ab2026__package_planned_capital_spending_reductions` | 1 | OBR “Reductions in planned capital spending” / “changes to departmental capital spending”; PDF does not enumerate the exact Table 6.1 lines. (Was `ab2026__package_obr_planned_capital_spending_reductions`.) | | `ab2026__package_property_and_savings_income_rates_receipts` | 7 | The receipts of Table 6.1 lines 61 and 62: Table 3.1's “Changes to savings and property income tax rates”. (Was `ab2026__package_obr_savings_and_property_income_tax_rates`.) **Receipts only (3 November):** Table 3.1 prints 0.1 in 2027-28 and 0.9 in 2029-30, against lines 61 and 62's totals of 160 and 954 (£m), but in line with their receipts (Table 3.3 gives line 61's as 2, 543 and 400 from 2027-28). So it is not `ab2026__package_property_and_savings_income_rates`, which is exactly lines 61 and 62. | -| `ab2026__package_spending_decisions_since_march_2026` | 7 | OBR “Direct effect of spending measures” / “Spending measures”: expenditure-side effects of decisions since March, including unfunded public service pensions valuation and the £5.5bn 2031-32 RDEL assumption reduction. This is not the HMT total-spending package. Table 6.1 lines 1–98 can affect expenditure fiscal heads; a complete narrower line decomposition is not published and remains unresolved. (Was `ab2026__package_obr_spending_decisions_since_march`.) | -| `ab2026__package_tax_administration_compliance_and_debt_collection` | 6 | OBR “Tax administration, compliance and debt collection measures”, including Table6.1 line45 (HMRC tax-gap package) and line46 (online marketplace liability), with funding/receipt interactions. PDF group yield (£3.6bn2031-32) differs from standalone line45 (£3.160bn) and must not be attached to line45 alone. (Was `ab2026__package_obr_tax_administration_compliance_and_debt`.) Shared with RF's restatement. | -| `ab2026__package_tax_decisions_since_march_2026` | 6 | OBR Table 3.1 "Direct effect of tax measures": the direct effect of tax decisions since March 2026, as the OBR classifies them. Shared with RF's restatement of the same row. Before 3 November these rows sat under the total package with `component: tax measures (OBR classification)`. | -| `ab2026__package_temporary_business_rates_reliefs` | 1 | Table 6.1 lines 12,13,14: “package of temporary business rates measures” including real-terms freeze for certain sectors, pubs/live-music reliefs, small-business thresholds and extension of transitional support, partly offset by transitional relief supplement. (Was `ab2026__package_obr_temporary_business_rates_reliefs`.) | +| `ab2026__package_spending_decisions_since_march_2026` | 13 | OBR “Direct effect of spending measures” / “Spending measures”: expenditure-side effects of decisions since March, including unfunded public service pensions valuation and the £5.5bn 2031-32 RDEL assumption reduction. This is not the HMT total-spending package. Table 6.1 lines 1–98 can affect expenditure fiscal heads; a complete narrower line decomposition is not published and remains unresolved. (Was `ab2026__package_obr_spending_decisions_since_march`.) Since 4 November it also holds the workbook's total of expenditure measures (I520:N520), Table 3.1's direct effect of spending measures reversed (review B finding 6). | +| `ab2026__package_tax_administration_compliance_and_debt_collection_receipts` | 6 | The receipts of Table 6.1 lines 45 and 46: OBR “Tax administration, compliance and debt collection measures”. The group (−0.1 / −0.5 / −0.6 / −2.4 / −2.9 / −3.6) equals lines 45 and 46's receipts (107 / 455 / 567 / 2,411 / 2,881 / 3,630) in all six years; their totals (8 / 262 / 434 / 2,248 / 2,733 / 3,423) round differently in every year (review B finding 3). It must not be attached to line 45 alone. (Was `ab2026__package_obr_tax_administration_compliance_and_debt`, then `ab2026__package_tax_administration_compliance_and_debt_collection` until 4 November.) Shared with RF's restatement. | +| `ab2026__package_tax_decisions_since_march_2026` | 6 | OBR Table 3.1 "Direct effect of tax measures": the direct effect of tax decisions since March 2026, as the OBR classifies them. Shared with RF's restatement of the same row. It equals −1 × the workbook's total of receipts measures, so it is a receipts grouping. Before 3 November these rows sat under the total package with `component: tax measures (OBR classification)`. Since 4 November it also holds paragraph 3.4's "reducing borrowing by an average of £6 billion over the final four years", the mean of this line from 2028-29 (review B finding 6). | +| `ab2026__package_temporary_business_rates_reliefs_receipts` | 1 | The receipts of Table 6.1 lines 12, 13 and 14 (paragraph 3.19's "net cost of around £0.5 billion in 2028-29" is their receipts, 536, against totals of 663; review B finding 3 and review A's K-7(g)): “package of temporary business rates measures” including real-terms freeze for certain sectors, pubs/live-music reliefs, small-business thresholds and extension of transitional support, partly offset by transitional relief supplement. (Was `ab2026__package_obr_temporary_business_rates_reliefs`, then `ab2026__package_temporary_business_rates_reliefs` until 4 November.) | | `ab2026__private_landlord_register_fee_65` | 18 | No Table 6.1 lines. “Landlord registration fee”; receipts, current DEL and total. (Was `ab2026__package_obr_landlord_registration_fee`.) The costings family's spelling for the same measure. | | `ab2026__public_sector_pensions_2024_valuations` | 27 | No Table 6.1 lines. “Public Sector Pensions: Net impact of the 2024 valuations”; departmental reductions, net public service pension payments and their net total. PDF calls this “Net impact of unfunded pensions valuation”. (Was `ab2026__package_obr_public_sector_pensions_2024_valuations`.) | | `ab2026__pwlb_debt_relief_councils_severe_distress` | 12 | No Table 6.1 lines. “Provide Public Works Loan Board (PWLB) debt relief for councils in severe financial distress”; capital AME and total. (Was `ab2026__package_obr_pwlb_distress_debt_relief`.) The costings family's spelling for the same measure. | | `ab2026__scottish_air_passenger_duty_devolution` | 18 | No Table 6.1 lines. “Devolving Air Passenger Duty to the Scottish Government in April 2027”; Scottish BGA, self-financed expenditure and total. (Was `ab2026__package_obr_scottish_air_passenger_duty_devolution`.) | -| `ab2026__state_pension_triple_lock_adjustment` | 2 | No Table6.1 line: OBR paragraph3.49 adjusted state pension triple lock expected April2030, retaining CPI/2.5percent floor but linking catch-up to2029-30 earnings ratio. Figures3.51 are historical simulations1990–2025, not the medium-term policy costing. (Was `ab2026__package_obr_adjusted_state_pension_triple_lock`.) The red book's spelling for the same measure. | -| `ab2026_option__fuel_duty_continued_freeze` | 1 | No Table6.1 line as an announced measure: OBR paragraph3.43 counterfactual “if fuel duty were to remain frozen until 2029-30”, extending line3 beyond July2027. (Was `ab2026_option__obr_fuel_duty_frozen_until_2029_30`.) Shared with RF, whose series for the OBR's constant-duty-rate alternative prints the same £4.3bn in 2029-30. | -| `ab2026_option__health_foundation_social_care_reforms` | 2 | No Table6.1 line: Health Foundation illustrative future social care reforms cited by OBR paragraph3.41, from limited home-care reform to wider schemes including residential care; costs additional to current spending in2035-36 per footnote13. This is not a settled AB2026 policy. (Was `ab2026_option__obr_health_foundation_social_care_reforms`.) | +| `ab2026__state_pension_triple_lock_adjustment` | 2 | No Table 6.1 line: OBR paragraph 3.49 adjusted state pension triple lock expected April 2030, retaining CPI/2.5percent floor but linking catch-up to2029-30 earnings ratio. Figures 3.51 are historical simulations1990–2025, not the medium-term policy costing. (Was `ab2026__package_obr_adjusted_state_pension_triple_lock`.) The red book's spelling for the same measure. | +| `ab2026_option__fuel_duty_continued_freeze` | 1 | No Table 6.1 line as an announced measure: OBR paragraph 3.43 counterfactual “if fuel duty were to remain frozen until 2029-30”, extending line 3 beyond July2027. (Was `ab2026_option__obr_fuel_duty_frozen_until_2029_30`.) Shared with RF, whose series for the OBR's constant-duty-rate alternative prints the same £4.3bn in 2029-30. | +| `ab2026_option__health_foundation_social_care_reforms` | 2 | No Table 6.1 line: Health Foundation illustrative future social care reforms cited by OBR paragraph 3.41, from limited home-care reform to wider schemes including residential care; costs additional to current spending in 2035-36 per footnote 13. This is not a settled AB2026 policy. (Was `ab2026_option__obr_health_foundation_social_care_reforms`.) | ## Metric, unit and baseline proposals @@ -185,6 +185,7 @@ No unregistered enum literals are used. `fiscal_headroom` is registered in the l | proposed_metric | Rows | Scope | |---|---:|---| +| `measure_count` | 2 | Box 3.1: the 15 measures expected to reduce the tax gap, and the four of them that target the small business tax gap (context counts, as the uk_thinktank_misc and uk_ifs_ab2026 families stage measure counts; review B finding 11). | | `cash_receipts_change` | 1 | Self-assessment overlapping transition payments, £17bn in 2029-30; assumed extraordinary accruals adjustment excludes the temporary cash effect from PSNB. | | `fiscal_balance_change` | 6 | Table 3.1's memo line, the total effect of Government decisions on the current budget deficit (README rule 10, 3 November). | | `tax_gap_share_of_compliance_yield` | 1 | Box 3.1: the share of this Budget's compliance measures' yield the OBR expects to reduce the tax gap ("a third"). | @@ -197,34 +198,44 @@ No unregistered enum literals are used. `fiscal_headroom` is registered in the l | `public_sector_net_debt_change` | 3 | Marginal stock debt effects of NWF/BBB investment and Your First Home lending, paragraphs 3.20/3.22. | | `public_sector_net_financial_liabilities` | 7 | End-March PSNFL stock, GBP. | -Proposed units: `percent_of_gdp` (28 rows, nominal GDP denominator, including one prose package effect and fiscal shares/headroom); `rating_category` (7 High uncertainty ratings). +Proposed units: `percent_of_gdp` (28 rows, nominal GDP denominator, including one prose package effect and fiscal shares/headroom); `rating_category` (7 High uncertainty ratings); `measures` (Box 3.1's 2 counts). + +Registered metrics beyond `revenue_change`: `decisions_effect_on_borrowing` (91 rows, README rule 10), `fiscal_headroom` (13), `cpi_inflation_effect` (6), `benefit_cost_change` (2, paragraph 3.51's simulations), `cash_requirement_change` (1, paragraph 3.20's financing requirement) and `gdp_level_effect` (1). + +**For tranche 2:** HM Treasury's Table 6.2 (in `uk_hmt_tables`) stages the cash impacts of the same NWF and BBB financial transactions on the proposed `financial_transaction_cash_impact`, while paragraph 3.20's £1.9bn uses the registered `cash_requirement_change`. Both measure the cash effect of the same transactions, one as a cash impact and one as the financing requirement; the ingest settles the metric. Every `baseline_policy` is null; effect estimates use the following explicit unregistered baseline proposals. Rating, funding-level and aggregate context rows have no proposed baseline. | proposed_baseline, verbatim | Rows | |---|---:| -| OBR pre-measures Autumn Budget 2026 policy parameters | 2,729 | +| OBR pre-measures Autumn Budget 2026 policy parameters | 2,730 | | Fuel duty policy reflected in OBR October 2026 economy forecast, confirmed before the final Budget parameters | 1 | -| Current social care spending (Health Foundation2024 scenarios, additional spending in2035-36) | 2 | +| Current social care spending (Health Foundation 2024 scenarios, additional spending in 2035-36) | 2 | | Fuel duty rates increase in line with RPI inflation: stated policy since 2010-11 | 1 | -| Current triple lock uprating applied to historical outturn1990–2025 | 1 | -| Earnings uprating applied to historical outturn1990–2025 | 1 | +| Current triple lock uprating applied to historical outturn 1990–2025 | 1 | +| Earnings uprating applied to historical outturn 1990–2025 | 1 | ## Not staged and unreachable material -- **Design (rule 8), dropped on 3 November (review finding 1), 58 rows:** +- **Design (rule 8), dropped on 3 November (review finding 1), 57 rows (58 until paragraph 3.20's financing requirement was restored on 4 November):** - Table 3.2's 47 funding cells: HMRC and DWP baseline, Budget 2025, Budget 2026 additional and total compliance, debt and fraud-and-error funding ("the funding provision", Source: HM Treasury). The table prints no yield, and its Budget 2026 lines are not the scored costs (its DWP line prints −16, 188, 178, 188, 210, 214; the workbook's line-47 departmental spending prints 218, −194, −184, −195, −217, −221). - Six staffing inputs: the 2,000, 200 and 150 DWP staff (para 3.12); the 5,500 and 670 HMRC staff and the "23 per cent increase" (para 3.13). They now sit in `conditions.staffing` on the £0.5bn and £0.4bn yield rows. - Guarantee capacity, a contingent liability: UK Export Finance's £40bn and the Growth Guarantee Scheme's £7bn (para 3.21). - - A lending allocation: Your First Home's £0.7bn and £1bn of loans (para 3.22). - - A financing allocation: the NWF and BBB's "£1.9 billion a year" financing requirement (para 3.20), which the review marked probable; applied. + - A lending allocation: Your First Home's £0.7bn and £1bn of loans (para 3.22). Paragraph 3.22's effect on debt (£0.7bn in the first year) is staged. +- **The rule for lending and financing (4 November, the coordinator's call on review B's finding 5; README rule 8):** the harvest stages HM Treasury's Tables 6.1, 6.2 and 6.4 and the OBR's policy workbook whole, as the decisions' costings. Elsewhere, an amount a scheme will lend or spend is design (rule 8), and an effect on borrowing, debt or the financing requirement is an estimate. So paragraph 3.22's loan volumes stay dropped, its effect on debt stays, and paragraph 3.20's "£1.9 billion a year" increase in the Government's financing requirement, dropped on 3 November as a financing allocation, is restored as a fiscal-aggregate estimate (`cash_requirement_change`). EFO paragraphs 6.19–6.20 say the CGNCR "forms the basis of" the DMO's remit, so the requirement is the aggregate and the remit is the design. - **Repeats, dropped on 3 November (review finding 5, the coordinator's call), 37 rows.** The table cells are the staged rows, as in the AB2025 OBR family: text that repeats a Table 3.1 cell (paras 3.2, 3.3, 3.4, 3.11, 3.13, 3.16, 3.17 and 3.19: 19 rows) or earlier text (paras 3.11, 3.12, 3.13 and 3.16: 6 rows), and Table 3.1's "Spending measures" and "Tax measures" rows, which repeat its "Direct effect of spending measures" and "Direct effect of tax measures" rows (12 rows). +- **Chapters 1, 2 and 4–6 and paragraph 7.20 (4 November, review B finding 4; the coordinator's call: not staged).** Most of these chapters repeat Table 3.1 and the workbook's lines by head (Tables 4.5–4.7, 5.4, 5.6 and 6.2 among them). These estimates are new and are not staged; tranche 2 may add any of them as components: + - Tables 4.2 (p.87, indirect effects on receipts), 5.2 (p.108, indirect effects on spending) and 5.8 (p.128, "Effects of Government decisions" on debt interest, −0.9 / 1.2 / 1.0 / 0.7 / 0.4 / 0.3): parts of the decisions' total effect, which rows 2490–2495 and 2534–2539 (4f424b9 order) stage whole. + - Tables 4.3 to 4.7's indirect rows by tax and their PAYE / self-assessment split, including Table 4.4's self-assessment income tax by measure (e.g. "Timely payment for self assessment" 3.5 / 6.5 in 2029-30 / 2030-31): splits of lines the family already stages, by tax or by cash timing. + - Paragraph 4.30: a continued fuel duty freeze "would reduce receipts, on average, by £4 billion a year between 2027-28 and 2031-32", the receipts average of the option whose effect on borrowing row 2737 stages. + - Table 6.3 (p.141, "Effects of policy on financial transactions", −0.1 / 0.8 / 2.1 / −12.4 / 3.9 / 4.0): the OBR's whole-package effect on financial transactions; tranche 2 can stage it beside HM Treasury's Table 6.2 total, which `uk_hmt_tables` stages. + - Paragraph 7.20's decomposition of the changes in the fiscal-target margins. - AnnexA has no policy-costing narrative beyond introductionA.1; current receipts/spending/economy forecast TablesA.1–A.8/A.11–A.12 are outside the policy-family scope. Headline fiscal aggregates inA.9 are owned by aggregate lane. - No printed numeric Pension Tax Allowance or Carer’s Allowance taper claim appears in chapter3 or AnnexA. Whole-PDF search finds no PTA phrase; Carer’s Allowance appears only in a generic welfare footnote, not taper costing. These measures are covered by workbook rows. -- Charts3.1–3.4 and Box3.1 ChartsA/B contain no printed data labels for the policy values; no bars or paths read. Tick labels and dates are not claims. Box3.1 tax-gap context restates HMRC and several-Budget historical compliance yields and is not staged; its one AB2026 estimate, "a third of the yield", is staged (review finding 10). +- Charts3.1–3.4 and Box3.1 ChartsA/B contain no printed data labels for the policy values; no bars or paths read. Tick labels and dates are not claims. Box 3.1's tax-gap context restates HMRC and several-Budget historical compliance yields and is not staged. Its three AB2026 figures are staged: the estimate "a third of the yield" (review finding 10, the sentence completed on 4 November) and the two measure counts, 15 measures expected to reduce the tax gap and four that target the small business tax gap (review B finding 11). - Policy design parameters (tax rates/thresholds, dates, litres/pence) without costing or behavioural-estimate content were not separately staged. -- Older-policy employment/housing/childcare monitoring examples in3.29–3.36 do not cost an AB2026 Table6.1 measure; no new AB2026 measure key invented. Paragraph3.28’s salary-sacrifice recosting carries AB2025's key with `recosting_of` (review finding 6). +- Older-policy employment/housing/childcare monitoring examples in 3.29–3.36 do not cost an AB2026 Table 6.1 measure; no new AB2026 measure key invented. Paragraph 3.28’s salary-sacrifice figure is the OBR's +£0.4bn revision to its Autumn Budget 2025 costing: a context row with a null key since 4 November, since a revision is not a level and doesn't share a key with Table 6.4's recosted level (review A's H-6(b), the coordinator's call). - Paragraph3.48 FRS long-run pension/debt scenarios are current-policy/earlier-publication context, outside requested headline medium-term aggregate scope. Paragraph3.52 adjusted-triple-lock debt effects are explicitly2075-76=>2076; checker caps periods at2075, so neither25pp nor10pp long-run debt effect was staged or misdated. - No additional primary access attempts owned by text lane; root records direct/local recipe and unreachable EFO landing page in manifest/NOTES. @@ -235,21 +246,22 @@ Every `baseline_policy` is null; effect estimates use the following explicit unr ## Contradictions and limitations retained -- Workbook row549 “Total effect of Government decisions” displays **-8,905 / -50 / 1,322 / 5,484 / 6,425 / 14,093 £m**, while PDF Table 3.1 prints **6.3 / 7.0 / 2.1 / -2.6 / -3.5 / -9.3 £bn**. This exceeds a sign flip or rounding difference. Workbook direct row551 follows its negative-loss legend; indirect row555 has the same signs as the PDF borrowing effects. The workbook appears to mix conventions in its total. Both sources remain unchanged; row notes call out the inconsistency. +- Workbook row 549 “Total effect of Government decisions” displays **-8,905 / -50 / 1,322 / 5,484 / 6,425 / 14,093 £m**, while PDF Table 3.1 prints **6.3 / 7.0 / 2.1 / -2.6 / -3.5 / -9.3 £bn**. This exceeds a sign flip or rounding difference. Workbook direct row 551 follows its negative-loss legend; indirect row 555 has the same signs as the PDF borrowing effects. The workbook appears to mix conventions in its total. Both sources remain unchanged; row notes call out the inconsistency. - Workbook/HMT precision differs: Household Energy Payment is -2,627 £m here versus -2,625 £m in HMT Table 6.1. Rounded components need not sum to the separately rounded totals. -- Paragraph3.19 dates carbon-price-support abolition April2027; HMT line71 says April2028. Its combined CPS/CfD/BICS/Supercharger scope uses a proposal, not CPS key. -- Paragraph3.16 dates first-year-allowance increase April2027; HMT line58 dates1January2027. The review found the date difference a description difference, not a scope difference, so the WDA/FYA row takes line 58's key. -- Paragraph3.21 says£7bn Growth Guarantee expansion; HMT line75 says£6.5bn additional SME lending. The £7bn and £6.5bn are different quantities of the same measure (guarantee capacity against lending supported); the borrowing rows take line 75's key and the £7bn is not staged. -- Table3.3/paragraph3.24 call reform Company Share Option Plan (CSOP); HMT line28 says Enterprise Management Incentives2 (EMI2). The workbook's line 28 total (−26 / −33 / −47 / −144 / −142) matches Table 3.3, so the review re-keyed it to line 28 as its tax head. -- Paragraph3.11 says departmental spending+£5bn2026-27; Table3.1 “Other departmental spending changes”+£5.4bn and separate rebate+£2.6bn. Preserved without inventing reconciliation. +- The 28 October text of paragraph 3.19 dated carbon-price-support abolition April 2027, and the OBR's 30 October correction moved it to April 2028, matching line 71. Its combined CPS/CfD scope (lines 70 and 71) uses a proposal, not the CPS key. +- Paragraph3.16 dates first-year-allowance increase April 2027; HMT line 58 dates1January2027. The review found the date difference a description difference, not a scope difference, so the WDA/FYA row takes line 58's key. +- Paragraph3.21 says£7bn Growth Guarantee expansion; HMT line 75 says£6.5bn additional SME lending. The £7bn and £6.5bn are different quantities of the same measure (guarantee capacity against lending supported); the borrowing rows take line 75's key and the £7bn is not staged. +- Table 3.3/paragraph 3.24 call reform Company Share Option Plan (CSOP); HMT line 28 says Enterprise Management Incentives2 (EMI2). The workbook's line 28 total (−26 / −33 / −47 / −144 / −142) matches Table 3.3, so the review re-keyed it to line 28 as its tax head. +- Paragraph 3.11's "increase borrowing by £5 billion in 2026-27" is Table 3.1's "Other departmental spending changes" (5.4), rounded; that group includes the DEL funding for the council tax rebate, and Table 3.1 nets the rebate out of its "Other AME spending decisions" group instead (review B finding 9). So the rebate row (2.6), the departmental group and the AME group are not additive in 2026-27. - OBR tax aggregates include fiscal-head/classification differences from HMT:2031-32 OBR -£7.5bn against HMT+£6.970bn. No total-tax registry key used. -- Paragraph3.2 prose rounds2031-32 borrowing reduction to£12bn; Table3.1 prints-11.7bn. Both preserved. -- Paragraph3.41 says social-care options; footnote13 calls cited amounts “extra costs of healthcare”. Reported as printed, no correction. +- Paragraph3.2 prose rounds2031-32 borrowing reduction to£12bn; Table 3.1 prints-11.7bn. Both preserved. +- Paragraph3.41 says social-care options; footnote 13 calls cited amounts “extra costs of healthcare”. Reported as printed, no correction. - Health Foundation range is across different reform options, not statistical uncertainty about one fixed policy. -- Whole-Scheme UKEF exports in paragraph3.21 are not proven identical to defence-export capacity in HMT line44. India-UK FTA GDP claim does not include Double Contributions Convention component in HMT line77. The UKEF borrowing row matches line 44 and takes its key; the India FTA GDP row is a component of line 77. +- Whole-Scheme UKEF exports in paragraph 3.21 are not proven identical to defence-export capacity in HMT line 44. India-UK FTA GDP claim does not include Double Contributions Convention component in HMT line 77. The UKEF borrowing row matches line 44 and takes its key; the India FTA GDP row is a component of line 77. ## Validation +- **4 November:** the contract checker reports 2,800 rows and 0 problems, and a re-run of `stage.py` in a scratch directory reproduces the gzip byte for byte. No primary was re-opened in this session; the new quotes (Box 3.1's sentences) are the coordinator's reading of its local copy of the 28 October PDF, and the corrected PDF's hash and change history are review B's. - **3 November:** the contract checker reports 2,797 rows, 0 problems, and a re-run of `stage.py` reproduces the gzip byte for byte (COUNTS.json pins its sha256). No primary was re-opened: the session's network policy refused obr.uk, and this lane's primaries are the coordinator's local copies. The fixes relabel, re-key or drop rows; no printed value changed. - **29 October:** the lane ran the full-family checker, a deterministic gzip digest and 40 seeded random row checks, plus ten aggregate checks. Independent audits verified all 2,520 workbook cells and all 316 chapter-3 rows. Its REPORT.md, the only report file any family carried, was deleted under review finding 7. @@ -262,7 +274,7 @@ The OBR review (`harvest-staging/ab2026/reviews/verify-obr.md` on the staging br 3. **Row types:** Table 3.3's tax-head rows for lines 54, 61 and 46, the three behavioural-erosion rows and paragraph 3.12's welfare saving are `row_type: measure_component`; the bank surcharge's erosion row is `cumulative`. 4. **Sign conventions:** the workbook's indirect-effect rows say they follow PSNB; its total rows say they mix two conventions. 5. **Repeats:** 37 rows dropped (the coordinator's call; see "Not staged"). -6. **Field fixes:** row 2774 uses `decisions_effect_on_borrowing` with `component: gross additional investment income`; row 2828 carries `ab2025__salary_sacrifice_pension_nics_cap_2000` with `recosting_of: autumn_budget_2025`, which the contract accepts. +6. **Field fixes:** row 2774 uses `decisions_effect_on_borrowing` with `component: gross additional investment income`; row 2828 carries `ab2025__salary_sacrifice_pension_nics_cap_2000` with `recosting_of: autumn_budget_2025`, which the contract accepted. (Overridden on 4 November: row 2734 in the 4f424b9 order is now a context row with a null key; see the end.) 7. **Files and paths:** REPORT.md is deleted; the manifest notes and these notes name no local machine paths; the PDF's manifest note says its origin and hash are recorded only by this lane. 8. **NOTES.md:** tallies, keys and the not-staged list are regenerated from the rows. 9. **Key names:** renamed under rule 2, as the proposed-key table records. The Table 3.1 savings and property grouping is receipts-only, so it has its own key rather than the lines 61 and 62 package. @@ -281,3 +293,30 @@ The IFS review of d5e323dc (finding 3) found this family's totals of the Budget' The OBR sums its receipts heads, from air passenger duty to visa fees, into the workbook's total of receipts measures (6 rows, cells I490:N490, rows 2190–2195 counting from 0), so that total keeps `revenue_change`, the receipts line `models.py` keeps apart from PSNB (delta review of 8b7b23db, finding 1). Table 3.1 gives the same figures, in £ billion with the opposite sign, as the direct effect of tax measures, and those rows take `decisions_effect_on_borrowing`. Table 3.1's memo on the current budget deficit (6 rows) now proposes `fiscal_balance_change`, and its sign convention adds that, on this line, a positive value increases that deficit. The workbook's 288 lines by tax and spending head, Table 3.1's groups of measures, Table 3.3's costings and the prose costings of measures keep `revenue_change`. Paragraphs 3.20, 3.21 and 3.43 give four policies' effects on borrowing, and their seven rows already took `decisions_effect_on_borrowing`. No value or printed sign changed. Counting from 0 in the 3 November order, the moved rows are 2352–2357, 2490–2519, 2521, 2527, 2533–2563 and 2630–2641, and the memo rows 2642–2647. The contract checker reports 2,797 rows, 0 problems, and a re-run of `stage.py` reproduces the gzip byte for byte. `scorecard_db/signs.py` now reads the workbook's legend on both metrics: on `revenue_change` it runs the way PolicyEngine's sign does, and on `decisions_effect_on_borrowing` the other way. It reads the six workbook total rows whose sign convention starts "mixed:", and paragraph 3.6's two absolute averages, as having no single sign. + +Review B also noted that rows 2520, 2648 and 2649 (paragraphs 3.2 and 3.6) carry `decisions_effect_on_borrowing` as well, so the family has 91 rows on that metric: the 81 rule 10 moved, the seven of paragraphs 3.20, 3.21 and 3.43, and these three. + +**Text that restates a Table 3.1 group (4 November, review B finding 12, option (a)).** Text restating a Table 3.1 group or measure as an effect on borrowing keeps that row's metric, and Table 3.1's "A positive sign implies an increase in borrowing" is a sign convention, not a statement that a group is the decisions' effect on PSNB. So paragraph 3.11's "increase borrowing by £5 billion in 2026-27" (Table 3.1's "Other departmental spending changes", rounded) keeps `revenue_change`, as that group's rows do. + +## Reviews A and B, 4 November 2026 + +Review B (issue comments 6635496734 to 6635499259 on #231) approved this family with fixes, and review A's key harmonisation (6635870934) touched its rows. `review_fixes_nov4()` in `stage.py` applies them after `rule10()`, with indices in the 4f424b9 order; no printed value changed. Rows: 2,797 + 1 restored + 2 added = 2,800. + +1. **The corrected PDF (finding 1):** a fourth manifest line; row 2673's `reform_hint` and note give the corrected date; the rows stay on the 28 October file. +2. **Parts and wholes on the total-decisions key (finding 2, review A's K-1 and H-1):** the totals are `measure_total` with no `component`; the workbook's direct and indirect effects, its total of receipts measures and its 288 lines by head, Table 3.1's direct and indirect effects, and paragraphs 3.2 and 3.5 are `measure_component`, each labelled as its primary prints it. The workbook's scorecard and non-scorecard rows are `measure_total` on their own keys. Table 3.1's total, indirect and current budget memo rows take `source_model: obr_efo_forecast`. +3. **Receipts-only figures (finding 3, review A's K-5 and K-7(g)):** paragraph 3.17's £0.4bn and paragraph 3.19's electricity VAT figures are receipts components of their lines; the tax administration and temporary business rates groupings take `_receipts` keys; Table 3.3's online marketplace rows are line 46's whole costing. +4. **Chapters 1, 2 and 4–6 and paragraph 7.20 (finding 4):** not staged, each listed with its reason (the coordinator's call). +5. **Paragraph 3.20's financing requirement (finding 5):** restored on `cash_requirement_change`, under the rule for lending and financing above. +6. **Splits on the split keys (finding 6):** the workbook's total of expenditure measures and paragraph 3.4's tax line are whole rows on the spending- and tax-decisions keys. +7. **Lines 70 and 71 (finding 7):** the carbon price package's scope; footnote 9's £0.8bn keyed to line 70 as its current AME. +8. **More parts (finding 8):** paragraph 3.20's gross debt and additional income, paragraph 3.12's two pension heads and paragraph 3.19's AME to local authorities (now `revenue_change`, as worded) are components. Table 3.3's four rating rows that name a part are components too (H-1). +9. **Two spending scopes (finding 9)** are restated. +10. **Box 3.1 (finding 10):** the whole sentence, from the coordinator's reading of the 28 October PDF. +11. **Box 3.1's counts (finding 11):** staged. +12. **Paragraph 3.11 (finding 12):** option (a), above. +13. **Averaging intervals (finding 13):** row 2529 averages from 2027-28; row 2522's interval is not stated. +14. **Civil service programme (finding 14):** membership probable, not established. +15. **The OBR's residual groups (finding 15)** keep `obr` in their keys. +16. **Stale and run-together text (finding 16):** notes and these NOTES corrected; `review_fixes_nov4()` inserts the missing spaces in every field but quotes, values and locators. + +Review A's K-2 makes row 2522 the defence investment plan's RDEL part, with the sign convention "positive = cost to the Exchequer (higher RDEL spending)"; K-6 keeps RF's RPI-uprating option apart from this family's fuel duty freezes key; K-7(d) gives paragraph 3.51's two simulations conventions `signs.py` reads; and H-6(b), the coordinator's call, makes paragraph 3.28's revision a context row with a null key and no `recosting_of`. diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/claims_staged.jsonl.gz index 5dd582647a3398d726c21b8414f468c6f8b69a25..2c09e17b98ea4c2533ac7cb596fe79b667e7f428 100644 GIT binary patch delta 48361 zcmY(q1yoyKusuw0C%C)21$UR?#oetGD3TTm1b26eODS4h3KZ9(K?;T9ZpF387y5he zTkpTl0*1Z!%sB~{duQgHgzX~#K0~hMxq*C1s*39DN{hy?+w>Yr`aO%`_Gf{)_xSo? zoAf<*#IX_k1aokD;y5Xq6Wq6BPcctsPrKa_c!qiwU=&;7dHM5GtwSj%ZK|H_6XI2k zDX`?2DNzuP6)rTc3=S2}>PuNMTz|!TIOW%HMIyBTnn^@dav=Z^u6IIJ2@sjK4+lvs zg_HMGMUc)z0}6`5)4+x1M#jSf*+&30Us~G_(H?rJWW3ux`%C=p36O3@cwcKZsr}oE zS*W#`^p7PwU4G%&zjJOV zm`QX;LK17Kt#7DrXi}|ZZ>Myqt;dJJzBSySv5T>bWv7IGuV0!&U>AG&B9Hv1>{D7` zZ_mZAtxv%}Jrfmkt1W0KXb@10#PD4g^(XTBI^!VYFp5s(MdWpzis~WL!X4-!vkU)$ z=+Y+QeNDuB8y|^JHgB)tljc;-gDuo+Y>1|o_U&U^9>NW}x_n5FFCfbiNWKYDVG^`? zO5wI4m~PAl=h;6%S21HY8PGSRB~aUR&1b06U6v4i7Sa~$$TMHouIBdfR@a}QnUng8 z@4g)ElCghWzg!HvV9}>^Z}C9_xAS;Zh#r?7uWp!Is>rzWf2?U<=$>yj`Lcx!MY`~e zLkMOqxua(A6^H8u#UKh&5+Z2njR4l4XMVXdiz=$Eo<~^;v6k;tf+?_CV(`Y+`;;R{ zqu#Rf;3-aj()FXDz%6p+ zcb1YmziUTE$D)*Yv%dkq*bpC0|5mq5KF;p8oeG+YnW~&9B>p%EyC!?g`Mm+5W-Pv+?Zvb#BAN{J=6@S0hOq?w-aj? zV=^)ykexV2XLtV^bxL}_ud}_YMYyO9nZqYF#XLdIZI216noN=5OZ>vC%!`(dlDi_h z51E8IyG952T3@q|sEB?Tz8Ro!`W3>$ef}e7SzJ$Ay2W&dvKPQlpVXkj7myNFNlt=stAS)jqMmM!>5S_2}!>j zf+|oNZuQeOBL^$gMt%1$PLGOjssi#4G?&iF)Y3i0IJ@ zVdfY_J`UPsR57|jCUu)wHX8&)B6X2(11)hmtf>nM3~eJEYf*^=sueLZ6{zPEp8u%h z3UxOApaeQi6yRJIiiipLA8`V!S!+u84udJX7HD}Z3B7;c*GrE5vcv|Ejs=MzLTm*b z;&WqrC^BbC%rSSwGel-Bc~Ew=LgsQrz|TmY;{43TPZju{zqvdnEGLjX%i1lm#|r;c zh}HJ>{tOo3i7>9*8DqUP;u+%i+#vGN&&_G*7+CL|O6rU7ik_zJ zmK9ZwfUDe@u1jGr+4Qf$h_3WKhIp;FPKK>i%bB&8m%33#>reeC{1_3wA}Dc0Xc}^m-Cs>P8>`?WM8k*r zAJ>mI7v%53w&X{*85F;?FCoVE%h1t1x0~|?9F(J(QxjWr%NR6E!6*B4E7In>8|5dX zyi4y(gQw+&x(5jHyvGF2wpzZz&H15OcE&^BXj#4{g`Cgy40P{zhiFmzSyNR2jM|n0 zS=iftySEMqe0X~P@+kgwsL#)1LdjlqVt(7N{@0`0Ik!)==v@mogli0l`87jI*FN4Y zWC*2K_^8p-SIq7wQ^+Cyp?&_eTOn)LC$^9LrK}CCGHQQ>T5cdZM}zT_3s~(|Ejw?i zFGH8B*ihSjx-d+rI|RPVMK23-ME^aF%Z*+m_X}|wUn1H^>ogJ}^1GO5;qrUq3m|&G zeET!Z?Wh}Z6{dItSpjQmo{p=ouVugQY+5&&ITRaWUu}g2`1~(``tLz!q17Wy+|Iil z6O8Wy-~E1Pyycu7$9K53_yC)Ze+16)dhm{2m86beY=1n^^H#aJ>vEe9pMk3RnkI~?FC@{ZY#MQ z&W~TI^bqwh4=wB;0&MEb>fdB@(YN;~%3kkN9E=E&>D?l)u9>8e%`xBFJ5-!vbtaYf z3QaDRwrnX<4UzSs`Ueue5yz(QI-Sd~YR2bKg#rW?Jr0VnFs+D^F>zP%p7Sn>|#J@x87k_H}Y-jFG7Uq((9?CHEx0gB>sM><69yvv( z)#8a^hdVQhW%jAAxOsm?e)TD(dErcdt6&68{fG+nlQjOvzhTi#Vnqn*Us@H#)j&z# z=62ZrOilGm`WK*q>SD~x9P9f;rw~6WuS%y_fa>U}9O2cE42ao2q1REu%u9%S!3h7S zg$Qg=5ED2GgHRnDg#rs(XFBXyF?8Kw`wn)4Pa2quRA?8Pia~w;+g^1Qf z*}}n#LWK35-cL@>G4hmC4eQruaEB_$yXY={GM@YttBB|m#KO9p2fi}~l6W)$3F=wp1ChfDfB<72 z0yY8%tzoe+V+M!y-?$zS^rix>AX!^GEmZWRG$c9NcKQ(6PS)Aa^cd2z@%agJdoy(j z7k^(}cyx7AY6zPAo$qTu`+c-@^EW#KI+#JlR5%n-!XO8cRgO%q`ike9Q!X$2^T;4M z*}DK}Qc?HW_=8EsoTBus4Js-?J0XmoMvPicIuB`fA3tF(UplvQCR@fst57RRw#4RT z1@U396;OuA=CP(;u?`(*7LLWqk`gSI0GCM!f@QsI*0qMQhYAuM(4lc*(zHV^RT^`K zW-RhTke64tY$!~?7HXq4EEKlrWdqPEJ60?dBv^zI|Bv&DdzHE{k+krRV&TQ>*N$dl z`L4SdAPU6np%Y0Sxa=A#xQT93VVifQGD-T938`C#W`R8u<^)Doe~z|ONaNE8#{x-t zHt)vvoUFo^#d|PMZs3m~_PUz1zcFhMhcP`ZulV}vEQ3jT5bVYCN&Pv;YCqCtuIG{( zsc93y^VcsVk$U`-VE-q{CM_fuPZ+B;q0C%=l*2d91xoV0d+3(u<}uAC4+QYQ4PFG{ zvI7(g0|6v?K}o>w*;V2wd~UpfHuo{ZDWo=O;>mdhNRa|DST~rf1EMOchu0s$5O1WC zi?`H*k2|ES@z;&~{+6eWB<^I#!}phk7f?|jXOKyLl3&H4kD8WHK%_q8BP5bMKi^;F*g$=O2KM4Cw*NU_ zQ_Q^yiFv~x=XdwC>*Xgh$_&?#slj{)?|HHN$Uz+f-D(j37U4EbZGVL_fc@E&Bkx)o zJEo(%3+voLh-X#U@PRQ{dm(kN@p}J(qPnzt?@N;LA>ceQ-|_kbBITmtAs0)&?>v9gkJlt4$2U7;mt!7Hi-GBTh{M9J<* z&kxF;NOcC-5GCStb^5ee70utgzSv88SQ;(6O+(30M+O9lcRZMK&8PRprmjF6Xc-)87L@#5I{sm z3<`i7BB9=N@z?3{l(4hcVBGbxStE(@yLoie?Qnv*=A2T%Ixr?4$6AI%RMM`$K3g&p zsm}49=BHDn!XZi>4N-MIpY^Y0>i%FuF9`4u)L(=WY6y&>(ju)p5(L7$7NsM-@vNbk zvK)~L{AIC*VF5sEpVg33O=*%HCXoM-nIq)hJ*E-p+QZp8SJLP%)N(WibiLcr@6Cf) zd9c6xeaS6y^%ZjX;oeKk!hMqlPa@zZh;q&E3Qih!iz>_oI%`#COtLcq%)@1GP2=FW z7bfTYe)RDs$WaX=_~>#c&}C>@N_@i@j0voTxLNvJWb0FR7?s0p%d=&-r01D-d5vMXT-F$kXpWK;FL>gR!e_7h&gimtem99}kS&m_FEi zLmKB-o}%|0yPZCg#ZeCn82AZa>7d`lnMv~pDg&-hBQ1y$NX*pT-1Lc2{=jEMp#fq6 zD;Y%S07Zz)5<^)GQDkal>cSlgEiAX5njQCRTN1jEL2Z<5loW9FU#LdmEI<-m6GeRW}2 zbKFUbJ2DR;nt~^vKAH|*mMK7Llt#f-{Bt*HkPi`rJG1iQC0iYL;4a?oqC$+CE3OS0 zyAX#io%Yw9>j-fFSXIi{#JH=^dX#;BR-9S#n_<0}PNdK5N*%{<>oDt>7>9Hj`MY^Z zeM|jvYjIQGXyGqZa~5-Ux$!hOhDdC0QvN`gulNj>BHga}Y}R%wnu9ZzF!{cX+>|SC zcMY=cK#O^uD)QabS4ye!8|(c@hqUTOtYqk!T4w9f^eFzMOFtRLRDV3N^ z99Dr9bJl&(psNa8oZA$`yM+R%X*xsv$rQu5e?&N?^^0IK6DF82gNu7V#b6bp==qS$ z5MQ7|hCy~63;+d&FIyd;khKp3Kr0ZLlyEGvaes|u(O{;xY>o?8NKP}j~Ry-5i&dYM;*F^k@j5B#w(DxBT@bI;$d%jQ4N zcr8s*WmVgD+F-!ITF zu--3l#1Xx zu=pW^ud=*K+hVBYrs449&V%js;OCA9`#aW~RQdD-5krX`%VxiO;ZVxA8YkILGh&PV zXj5kxx%z0H&SI-cgc_uq-t*QD@qdK)zODYFO))ksa(v#I zKw;C|*Yk33u2t7qezeIDza46T+&e!O3!m+`-+fuNzoB@5-Q{uG4`x@Sz#l^)et*Aw znUjhxE(w?%my!@%`p+3nbNTUg+2_*)hv)ukL)JjhdtHy&97NlAzNHIk8dSufh2}4R z5bN=|QMVwu(pmcOtnpo2Wg>WySpLc*0PJq7;(VP*m!&=C0qRtPR0FGY1DC%cB0MNd zfVmY4@O7xjaNcm9>}1{oBdT8@4j89GLh@(_)o+RdN)fg~QCAMUg`;*6)cxqU(1UYJ zb4z=RCUI+pLw5?0T*xl_bIUR-xE8V)zNjekwh!`E;T8c>o2`BPX!V8bHe@zzRzc>V zUjJ`rN=95x?EFsZLee8qS8Q&AhC_P5n?4Slh~2B>@i-B)@yVy9$E|4>zh*zJ3q-2U zKPhT(=d0VgQC#8gVsIPPb#ZTa;o_ zx>^spU%nElVkmJc&fbgw&1sqv@D%=$@bZYRCC4QQqB~;pf{$MC$09iU{?^M>Dwc)I zW+H=n>gYSZ7v`%42GsO!Ji$8=)>lX8$mt&(RCL7BSzPUHTPcl_yYa671Tofzet)kq zg4aG0#E4}Zwebl#z4kLUM-Zb+6h6@_^d@W$$k{}4qbmnA=hv{iSUp1+ql+iPD@(Mc z*_>oXa6iJUjs}KG%2x#QLVUmDIm)&DFP=6Jh3xO2PY<)-lXLj^!^$J7k*@Et(H1_5 zYCoxxkWq31@yT**)@94bayG}P6BepPrIaT;ePZZ^5V9L(Aymt&fhlXua-sJWb-sAaFwVDs) z>LvBRKJg(k+^Q@0gu}2boDg9$v|YrlO`UXN%+(-no7z7$!m(z>#W==>J+T|bz7WNj zozbKZe*ZNcs_y-v;j?O6;;l6Cp6rIFbK624j@OwwvEYgblNp7Zd%kpr#GWh~qFDh3 z<5c;+wUErzXCE(UhX{^Wc^(&{+!O?<4Tzchi7U(twG9-{RWU=mw-Lm(B@au56Y*g< z*bvd|lM?Z`mA0jqk>0)XY7l-(qJHzlzFH?|weA4B|F6%_MT>Ji zfH8Y}PyE;2UWZV4uDgkGJ-i*j=&~8;>79g)n##rxrnkqb!C|X^d-I_>Fv%nZ@srv~ z@5!Qc?P&vS93Bu*5k1L9cE3b9(V~M%PK3E0C|{z;bpIFSknMRWY_&5s|0T@ow&;@$ zgzLdt@9EQJ2zM?goaUr#&N>6#JyT1sc^J{y9!mI9Oz!ii!SA}`6ZZBj z`6y1>+%U!7KF9qV@n^ix)%pz20XBI{5SIYRx7+vu@iT36j=TEr9meVb5_f3DBMx`o zO!W`g;{-YR;w^mfd^8s4UVqjd%(wzFeI!njM>pqjI=8ykEF15UP!78YeXSglXDXGS zZsrh)A20g}+OlLBIJdkW20njO-QIYcbP(AU0VW zg2?H2R;2CU|05^EU?ii$3kl;q&|WyPU;l9l-(Vb)#s6?$VG@|%A*2ggw}rg6g6)WGe(M+`YL_PJx~v_i zrlw9_#PgNm-uwKu*))4|Oe5v+!d}IKG=#Hj!MFoU-m+ZH#OcTj0aFA+%Rt+#KeY_Q zyO0Zucl-50{e{ZxwvpL#Xn;;(Tbrv%j5q#lQz4hJ%4}GhRz|q~Iji?CZ-#^Os;{l9 z#Rvl??VlUpWGq~Ae(u5lAex~}04Rq1t0wH9|5X#el2Q6{ReG|NB%N5J1di+oa_Gu-Oc;l#g{}0K-ou$I19RV2gBq*Q`@WcmFj&0-IdQlc zV4#~Es&YFoQOam=e78n-XGRn=ak~D_est*90O5sn$U_kFT?-CXr~m2cv2qyatOgei z?j&T9f#A`0{0-winL}G#P9O@`B8>(ov z6zD|PNH0WOJB(=e_>aK*kC=gn5m+@Zgw21%Pt;lpRHEy`7b33c04`m&9MO;rn`m4B z*HVOl19G&AXb6Wv&M^nz0LLI2e!?ai_^uUbIieYNc(Q9Bmtk?p*JjxHNon;wL@dML zPJehyO#9DY)Jb>^s}R)+xoB{0Qzhmzio3zPqKXh)oRjRvrwf#MBUUa?L zPAf(V&WoT?WdtGyv!orw(zIK`%XH$VUOQ%LL_i8=-+2Z%q6Ftf(HtvD5Qu(UwWAa~tn-n=_vF=p}A2~oyZO8uFZ)=>H*nez5D--VF+L@Dn?xI6s2*Zz=m z=6p7H%oHsEn>5m=%|a!@1qmQq0ZI`{g*MLCYoz$k*f63@+Ee|a3|=fz1__Z5kuREY zE%Pf4fFTeOw9A-)R~v)NAk8bwTVw$Ll@kR|8eUo2)9s8VNLDroQJF=Vh2T{;uPY3N zLvzFiQZZTzAt?uz29_2`@@}s}wuSR|H8sqiZ&G5;nsUZh>3k?hwW`Tt=LwG={Z~k*@-y1I#2XZmb-rEJ#791k{;&di$5ZA>m#DK zld`w1Jkhv+>I=3{SMlty%j3~>y?pECnUU`fox8kdiY_+b`);wDk8vVTH8dY{=W=>l z8xo;JDK1HOV2P78^M>hB#CTj8s;QX)PXg#dh_S=S=qCXrLtc7jrsl}`5t~s)4PD1@ z>ihFLeCuSUeR}-9s-qAD#msY3r4Mu1 zD}q^tG1FV5?a{sbA4*xZ3G!$|VHPADsG8z%go!@6mQ|{&^&$KmZE}tD2eU#RtC!{n zU17UDY{N7K9f4ZDm~fpe0v{fgfIDUzq5>NZ~lL8Goe{cN+oWv z=;59Fmw86nsiC053saTV)HPCZxz$hyuy1(h%QW|;q7dJ~KiMy9gUQ<^UdU5S|JRff zHaF_9Ij4lyvzGsHuu$dkd3Ar8#6ukRvQNh}UbvTNx=ND-_y3E!{vUL~A0PHUT&g!+ z#d4fpeCvoTvIzJ{_~>5u!S1j5KtC>^{K49| z`IXX{4JIfnKd7c`uYB?PiheBoCgSIdz_`-)%@-}RLdr%t?=bN1Cb_6E@Dv1_QU80B zjgXNe%JG!#+$*ue_&ZqhgJq4TTpZT?V4QdtIIMy~m9ro8WJjV%pbtuN8IqqR9o_wk zXumH*C0x_iKEu3OKbCAF)sm8Di#+Oj^aLtlb~q$4Y&qsAg)0^*x>H88G%O=_Z*} zH}LSo#Kd$mIlR*PUE7jp;LPg8u-jw}_Sp7p?Z13I3Pm=%`>D5RXzCRIPDjY(^Hg;8 zhXTUJV126=1u-YP7R9L_f5{EvKBw(CvO?bZEukCNo#4ReDYhf{H_X*XbQs1m>=upF zIbfMeImZDexha^!AW6fRuuZ@bsahPW>J5idja@`2Mkd4VI54Oz>SQU1=Uk=K&M4pLU96}eZx}%afAyZNGt-( zBD1zXA?S3{625+B4Ncw5CEP%ki{8Ld2Czm;y-ln)k)lB=aeNF+RZ@oD09a`hR=;5B zfPbNs1QOU$C!V9QI!nR?Wp|A-6;o#~{18fsxNJ1|CBUKtEI_5i3xE9yvHXVGqKkKT zIl^MKV+js2cldvH4?|pSlwB8N{&!aF9t2bk}=I1tQg!J;MF03sm@vQVU0PoNsOUi%;w^{f%ujalbMLl@f>vxi8p(Kyku(z8(KJC~wuHHyt*qnQ^VE2%aAlMv4%VjMg z+GuJPF;ZGn7u_)$%SMzXIJ&ljn5BfUKJd{l%K41RRFjP1zKS1N)^ z%7P*l|1Jm@{Ze3%4{;<&0#?V%7w3gK0^@w^-^n**t41P29ocA&v7!_qXiyrmN4w2P z{W3mUGWp^kp^n}RWJt^}emQaUpO(E}g3bS8DEi+LaqMg}k7MbU>Rz1%d7yygamO{< z2p_U*-;mX`M%?a&|5ng3;}4lwtT=weI?wr-qVSk1-#ZewkR;o4!3C8&1z)tTHR63c z`dhp0^izz0B$M8nl=GM>ynr6=`X7p?qpGay#1C|zLoL2+5sxrD1Y0nq#I0rIs`PT- zDlnmSY5d^!P`*X#8m1-h;bspCwK$DI)Zv75{j|E;2X|VI$awia{6Xj&xH|yA;akG1 zO(MPy+35A#VJ%0o+1mIrRwEgD!0d%c zlc0LS_lD+0O5hE;cGF;n(!z)EMBon!tOxS|k$ZE3dFt?b2}IdYH?w(IY13qTOI$-$ z2=NDbL;jbKf(7Tq3k~AwS3W>%_!-*OnIUqnXg~jVpmJhkJOVoKQuw{}6u^n^^>K#& zyqUDo{1_r};`$u@?wzg}!hbuFedaXPL;_h$nv|S{6=OrG-^t^<3^-S(>O?NVZy`g77+Sr z1}6;efD;Y}^{XW!PXqbq(m~)EM5m4vrq50lxK27s)3c7xLz*p+A35*-&ORj!a2LM` zh6vOm|71D*^z+NNypIP`0p5HQj zkdu?(qNX%i5z>=myWf!i~)2Rsn75j?aKE zCWB|z4RJ=zasEms_D)t^psZ9S>~(F0kGKCjcK);16|4cI{`eI#0%eLutBsQ5X_x9| zx_u~`uw?CVZGS8CC7vZL;e%bQf0uPcT#!`;G9!QyLS&B)TPtjJBQVbB z#j;h0tp3Bav}=eBxgY&OTc@FDV-A%=3WITksmSXKyac^6!u?vP5g>FJH_d|8TZh3K z8Uk@=DFI@Ce<3e*&_iDjmjbbSX|rX8<#ka})`8ki!1!eyT8dD)S9%uf|7FS~|7FT= z|DQ~`{}U`zE|e7<4iB$jsyE8cLKHeoF;kJ6Q-b|B1)$Fz>0>Puz}_gY*G+nH%W8#W zx^nzs&Uso8tm+rIQRf9#d4ZWH zM2;Au$G<;3#xTvAe=Uen4#v2!q(WyM+jOO>U>ytbDJzBNNtn~Lq8gKv3iH9{#Zf2R z=@AF%0WyawDtT~r3cvP>M`T89pu^mvE8)g$nhY{gE~d<_@({-462g3Va3u8McIrks z#x|G4cX^%k-%ioje0_L6cA^#mukAWLBxZ;*1DGJpq2K_iFk%*Xg#zXG>rLM19Od7w z1N)h+XU(sk2(9$0YUvI5X()gBkcrAF#g)(WPD9(gC6_0NMTM1lUb>nhg0QZpYY!*Z`&S_cU;S2CRGRUo zx|P%!Os?+BMwp2yLfBg2h}VL}h<%M)Ei=^IgG|;y%y#Ii{_F&~LC-zes=P!24#D}L)K*pdmuTX}VP(-Cs_SV!QrrNWVL>G85z1Hy2tuuH0 zWY<4cbR)sy|NaJKUzaV!4Z>KFDpW3$cfQ0`*tjKCD9Xf+r$q4z#GdNO#f~BpT>v0? zooYk~qAA4fO@T|~+9i|rv!5eZ{O(DxPA)z`=-Zsg76(*l$SY{w$S!Qy2R$mt$-wQH&c^rW@7Ps(33%3dy8aY;u2m`VLmN3qm!QnCg>T z=@CFf&wKr&M+)7z2j4SHSpj4K=8X5O0D3@gM)u1Oa7Gs^fEMs0W1SVi23fj(?bL9Z z^4pz}{bOnEbOUU#qZ&8A4eM3aZ0J>;LfXiEpzrYNzr2f&Mv?_m&BGL;*wr z%~WTtr%Q8)7*cbHbN&03Xkc=c^?mCYOU)vW> zeRtB)L(2rdg*?3N1)Xr#3OK?d9I@Z+vPtXiwxI(Mtr%1tnnb`D<9 z#SIzeAy$cmNiG%zFUX+Hpxu^I)^z!BEbqmiOg!8|x&Q4^;b;*o`+O<*aG{^xyKf6= zjSX1+;mQ9hkD>|3gK9)3n#=8E{34^cCDU^3@ImUyEMV%KeLUeIApzQcHom~ob~4yE zfJyN^{ftWE+(IGmOzr8>S2%90CU27Z&NcqLxAiH{VyYH{@1~*W>T1C9Obxlxfi5=5 z;RBM{g8T6$!-3JmA$>9D<^p=PCfA(IS1mN?MNEPz6-3Fmw{A*cjyQb`h`fo5377U zKZ~das?#VKhs2%-S|hO9-Esa{)l}2`wm~s%9curm!*~QWwMN^^nN10%5*&wzNEAdo ze(jY`AbcN0H6>R{Na4HCgWexVBA8!+CfMvEITzJ~l!2b2P{+=+q-ce%))TLZlGQtr zz=KpuBh^cAv=GY5gOtrlX^$>;(L-P>X)9T1CwcS<5q6D{Q_W!$gEbJihMk6;;K!U6 zI#4avFs>;pvboWJp?9vKYt$1$zOLCK(84aKg}=L5t+f|Gm^my5?`r*QmEONMcj8t= z4cp$d@kx7bT~C;gf$&&d@hzTXp+IeqgKpH-sCK3}M^~0D<{{WzN|1>R_T}y)4R7Ag zJTwhN(&B+XAMA^&L_5qtQRhE0WB=G)wFU%gQl$E&_xnhFe`I4fzJy3x*cCxW@I+!( zmVXF6d%V%T>H7=${&#z5IM%C?WZy5@D6ZLOYamd`zJIH0Bv2glk;x=|SD48MOqQK5 zKe~T9I`v?3d9zxqIpf#b#XWyB5hKi(GS)}uhAsD`>_Kr}zAF9;+A?tc&S19zWqkA* zEW~H=&6l=5Zpy0d{0;*DSHw8ms5t+}&QUVimzeE%$adAdPjhsy{oUITZ)ta2mlRMl z{e+B-BW7G--P&l8sN>k5%qhbZf#zy;SsV5Fh4?D8xI$h~_+YSG{Xqe5JDi-2`p<>t zGK@F@GaE5c9vCx4LWUt}T^TZXWQ!dqU}UpICk*=O9(P=^1A$R;HjW&(TQ$gJ?1?2=1(?MB&1Sk}iF|kG4 zpOG&9YJO#i1D#@|`&8<38|DbaDnio+c|lARY+C#y!<%uSAp3S+Sm_H?k1Q(8U1B3^ zPmlv$Lf%D|@_x*LmB%5D@B~h zDL<20{Tmis21FYjj_q9Oc=QtuIweZ`DJ6ZEn%i~P)2C4Wzf)El0nSuk86+(p`sgHA zpClr#b`uiJ^aKPQ^Y#VLP}IL6ufKytCUPIgzr)GW3+#*}hz{fOfO1Q4#~s`hIJ zXZ9YQ35qT32n?jQ*L81?^#fu?;zZ)lBWrARx+ZF}57fQ(Ds-8Fv4ktk^2@>MdpI$}dh=v5{ED zFI^=;7p%`|X?-&6CyfU`H*KW-x3gbVZGJSZDAjqdXJ8tn>V?C9j)r*p4B#A zleU}^mjuSo#(%-fG?S#`S%s(fzhvT>zdn1oJS!ej&(q9cs_)|$>{Xmx`0=ghZmlnG zJdBA4zyt~c1zD+X*$zh07C*al1urf@eM(rJzCJv*$VuCS=!=4Oy^hvU^ zq4J+-aKj#fuwc-T{|Sc2prh=Rs7%jLR8x%RP8{>zL)2pR2ZJDl(w}P5R&acNvJz_F z`0)Rqx5{1h1Ul-&uEH>S<}HIeMwIzc2T?=u*H&@jqcUIA^E+OHmT-7f3q@S-!4@VK zl&}+0J{^9FAjL`m1qF!EZ}#Kkh|klzc{Ej^xEgBlX-^R&+kNI@a?3o4b(pX&Ij zL7OK0PPhkg{P!D_q?|&NC9DO%5I~a|U-$uZ5Y3PsQZw9wmlIL`dCnMQVxsV%JeidK zr6XmTV44hyDp$<{TPB3NXU-XsUb2xduwznTWz7Y3_$)6nr+IS6%P@65W7mQtXf>AOzYii*F}{2d0pfb7AN$`v zr{tTNfjOH*{vhI0*$Dy~^Pi=BXQTH=hIqV9v@yF_!17GfYmnDi=EUJF?_4Mc=v5+8 z{S?POZHyV44AF@#t)2&8j7_DNd8J`FlQXTFEN595k4l2^5d~V)2LlX=

9ykdg~y zLKOjRj4NDc8O~WH-&u&YkW$$|i*aOp&g_OUW$rwiSVe@fa;z4HGcE6Qak{KwsBuSS zSpBdHXS@^Cr+TsIQ(0I%YEuD`D4m3J-68Glh8gFUoQCteqlgJjYM{^W>N5gX;9rn6 z&b9e$?xFb67ef|LQTMgmzpT^SwI&zMSI>xMScg)u(e>wg8}jb4g`fE6v*kR*KMv#Y zSp}W&C*)-;6OL@&yP?omJ@7RrUNb+CIh(U9DZ*zUD<8p>k9X%yCq@)Sd(OTe8+B}4NcMTHBac&-1(OPXJoUU3o10^=e`lk5rU^|jmnmx~mFxmoclUcT)Iw|`EL zx}Uc9Y|LpE8O7mV%F&DM3UT#@e+Z;LwaI8&M0h)zfhGiqf_Pl-G6(s<|LVqkaOWla z&1;jrxw4Bu;y;GM6R|+sZb@I3-Vrfw!AOS^b`TVY z%;$@?)L~w%;(bw@pE#wHZ9E*P!lgwKCMm!rXLByrE-VdE>~xw{sZaU-ri905bZ90{ z@$O~5w>qf*=A)K@O=lwXGTrf?eaDg-MF>okq_~4R8?Lwm$6X5Bzb&;Gue2`5$g=6o zzJJ%O_j0_sL0Ink1MIV~ssX?&vP5_t>1bbdCb^(lsu@?6*`GZI@^9teIz}U6s~56@ z=57%u)L(;1BTB{J9Ekt1eI#-+wJR~kESveOH);b{mnX-eBN~ZRr@T+|Go`;Cf*}^U zoo~48s%mY(SZ6D37G)O?==a60Q$D95>ZcnCG1SBx<+`Fte{bz==cjzd#*j{`rDVN5 z1ZOP~Vp!r2#ITmBBC(cGsFFQHse8a@KOSO$=;6IWo=r2{A(5`y1;)s~9GJ$FiM(Q{ zJMsWp-!Z3Gos@+9k0v8t1i%f@%;*#WFah3XY>U83+L;U#QCNZNl))|vpn)`s9BCnW z-U=y2XnfZ3^fKiq^h9*F$Ie=<*_xUh3;Z!3H~9X%tEQ&=V{l2duHA>Td_D^t$<-el zlDrB_YwHWx=Qd)kyhGRP5&C3Dk`iQxUuKSmfZiJqV!?P27T(dn&JPeWe|*Q$Q3GVr zWT)GA8sq^_;sU$jwOh^--DwbLUhJ4or!Q1Vk@Vth?&iUWbLgaQ(x3M5_qkzqio@eB zr_IVkXv*`9xY7j>^87gSDR!sNO=|r7Bne=ghX*|;6q_+0ZtZjxjL950zCVl8T+Q)r z;9H{r)d{tPizKvXRlYe-GQiH>CE$bAY;feqwfh2yg8GpNE`EJX+qEc_`tgb{=IBEw|_W)24S(6^i;#Y9{cGU)Lr>wQR7( zS=8ZJ!^3ut1%jC>$``FsVHDf$F(=Nf18aUG&p8^XPXp%+ISDGxsz3cP5|h0CD1eS% z_BFOB#hGZ+oO-UR^jikB@=TX0`j-occjjjMabxR?|MX^7GJPzp-R=E)3y1q$O;siB>}_H1y)O zd#}{tl?JZ{s@3|v7|DF<0$hTSY71EEcG{EE8A1p;Z+cnz6|8WbFR6s$$sxZTb&+V% zDx?V+7}hI#ae~=ob;5cH5cPl| z1?ajm$asxxM{tOGxB3e0st~=sH&PQzt`&~+I)q)A%+I7F>=rxC`__zw@X*cZUZ?`m za4>cW&u&^O&0ClCc_ODzX=~P=oX+l;baGo;82lw{K{7yi9ynwb<%RCrVqoApDG8rK z6sEwzOD+-?*wG87J zTu)5)N|5GI8v}{F=&fdYxs(bN3>&91;>!vK93?aM!>r}ASkog2)dJCUlIA!JrX#(y z`p<-DOiaAUa>7HuLk<_#-3t(5r;CFSVdqB|j#1oC{h{*ddh#e?BT75^X`kw5!%$^$ zfB-rcaap<0TyP-jYUCaRA+4;%46V&(cw~HY0#BrfG-71I1}YXVxD!C%?!=(x4qX)?^ zs31q8BIDjWV+us_%zW*zE-vqob_gZ?!C^&h2HOJO<45;2S1E=4S(t}Gg8FpMaawQU zc8h`?5p)C57vpNRt(YkadT&;Wow^LF*)Km+*vI|`(+GrAOp|~pNf2xj!J%(4+fTbm zI^&=!GtP04!kSASn=N}x{xY^Wne!lC9^`GMg~3_`VC?!5iAq!NR7Oxv`{$C~C{?qm z-~tRU89cmZE_6H8w$mqU#m{Y83lTZ*DOSHvm1EGo><&39^!^OXjV-`fZQC=b#PDAq z_E}XYel$MVxH@%p(f>8sch=7ly~?8G1+ck*v%!SK%81vCT%=vnOJuHRk=lLC$@r>* z<*Gcfd@UlCFsB-lgXlzq6^(~AU0aM-vi@itt6%ByhUG{#-N07ZGzIEheueDxa)xM@ zk*vcCh3P|IyHEw>jhvNrvT$ch(w@I($RMIiAQm4!)X1AMGK7DSeUJ}J=+u)csO2+UF!o5C(SHs!2EJF2O0TK?@XjDems>Zp9%u!QE-G;#ykV-QBIYyF;pPNzP`^?Ci|M)RR7M7wL$uL*VFUI-gZJ%6w5ElsUXH_2^;$_hl&kXqIU_J!C@G z(>=auCFkDGgjy?p?ewoP_>yUm#oK9^WUB_7gls2*=1#K>{?Vf(s}l0}FVIJ`TGaG2 zc3bCe-pM(F9o7~U36ex_P{?{H6yl}xkv2497@^FWkE)gN2NF8rKv-wsJrW{1XcL;G zeW*wl*Lk`*DDo7YGK!K{#R&>=OQaeLodghv;CzK1!LZ>ny<`EY#o(pR+;GY3hJwuY zBT69eQZ_u;`3fzfKH8py6NFo~%Y|;MSmn~3il4w395X=GO6LEnWST#YMH4}w)eQ)a zsaqX23m!XJ6#l(%fpgALX z0}-b4j;CSc?)qVEND9%;eX7F;_X-z1;w0Om*G0jw!2-^RvXOY6<@l7%yGR!@>KGR) zEwC~tN9csQ4HrO)9!SHG0*M%b@@_a!k{;5EA|KBmYj8#T=|3}*exITPaMpqZkksm% z%{!EPNZ;0Hrao7Tn^w+xwtw^ND*1cenMQagb+GfymmSURP)@E7Ip9LlD1WRhQi%E! z_E#V5k!m!gsuP~sE*%-0c&aGSM&uxx8O&4-uPE2tSZ$8;W!rzTMyt3{r8x-#?4nyR zeQV5<>cLwIMx`;(UwTDuyn5@Dd?cfaSP?V#+G$^wG4v$%&S9ccs)la_av?9`fy{%M z{8kHA%aZW`tI3W-ig@gKPbQ*jkn{+6Ca`RCHJt=?Hzl2pO{{bv{KR{zL_eizUvH&E zV3?E6t)6DDm?>H9+PDSepA+xPDW1jJ+R-z{k8Zc#oP>%KpWFs3X#v?arz(BuOOWr9 zJvXDREH=)LHbr(b9EEbZ4jQ^Vf6WJ33-`O2&_TE)Ahrayo$K{inKKypBriO$?8+ZH zY=1tvllx=&727}=I(1BBN>S?ka0cb5{tGS45oB(gUjXVSqSR_0yI**e7L1Uk4k}6I zS86HHOqyAQ+9dMmlTjejG74mg6rp|r93V&H6c7NhnI&-W*hNAW5F#OK8+HKK4UWWF zBtj${Dwe;b;udCxAE`2=U<+(Ksl%7Wu81{F_d#ZvEu3sR#I?lz?USsP+PDjh+ZWyM zL|LPOXe$KO(A?cpiK^q=UzWH~m##X=<)>nJI4{O_e7X|LOIOG=SeIaZ6T6OTBGx-h zFgEXxiiD2xomW$pPG~U*bq22Te4xp?L)IiQKA%m(dwFfrG2dO6MD&6U4R_}MC}MEb z`<#m&1w-m!!eYpLac1$(wqIe>w5CD?IUQdpsrDa&Vo)p0e5V^tFwR8h3x>!*6TyP* ztA*#g{upt-eZw+*D}{!0e+A#Dd~S}-IoqeScOwW>swD}98%eC-f<GP_CfJx~YaIEfGJT&BJ@a0Wkc_P_FWvAuku!3y!4aEHGld z1q$Ob?K>^0^1aGZ>#zx*kxQogSAozi_=qh`5^Rf`+CB=sXPNZZF;D~!HUY`bDuu1^{Vrx4ZY*l*cX9Ce(Fq8N0u{Z7f|q(2 z&niOvj07RV%TtEs&mLx&B!npxw{Sl_O6*!6v-o@Z`XYc9%k2G`Os71E3sTr;-I_dE z9LLXg1e4seDjdbA^rD;O!Xl}CfPw}_%8S8;W~V?!nn2}O zk=Zf=Z6u)J0`maKSeRkxa3Y~FNMXy;ZqTA@&_F0SKIzE7YFKnAEL7m_9g!*Q&Lz7* z*5Q=8QNe%tvl&*D-hak!xeXa&;{_On;Cw2}27wnFB*HGUc5$-8Bs5vZdxU*xDiU@AkcX;vw#R# zAr^tA`|+h0E^K`c3b2_MIoHguO1Bm+2_77Z!?6{%a3=q(U#>o*)%DHR94YQIxgx%TlNzP<8m3 zC;slL?ocae;;dk6Y(G0h*A;F4RKNuYBr_hj8RoVSsCxVa61_v8Ajd)_ZC3*U(0JX!?9}c$IMrD~Q@(8`nlo(}Yp6L2-;<{4m z)d5MRcvZAcER^O4AP#4a;W$=62eMQT9RbbJ%bslE5*Gk0^{10hw)Aq5j);fxnD^q% z5upW3ioy`Wz!$<9)_%6d5%|(WmlIdZ>;4v6z~V0Jqa>6B?HHc9BK-cAuTyLP>~Cd<`yddIkyya?^Sz?Q&q`@@t{#Ha3O+JikTto=Js| z|3Uj|)ea63OMq2@0*D|BOnGC|fhu5CKEoDg7Y76tuxut#xWPmd?(oCxXOkH4(N`|7 zN|H;Lo1+MIBPPXjZ935MxuDp2Z5(vM5e;zt=(&p{X3$6u9z{?S0c2)6%bMbHCie2s zKWQe588hx`joxx42h+U|Vy8m};aL)UpAo^7@1yIppywlTDtbCyAbe4tK5`n4Rxu2H z@Y9yKaZ)cLdHqL7R6->)R36awDG2AyiByY*0Y{6PT_nBtU{Q-Ab)q0zyu~ZfGz+zZj}63P z#dhqiC8YvQ!ASXS&VRoDI~mytZl`f+$L5Oq-Nhkx`9EE}0n5=zV#u&ke&*E2&%Kd@ z6PnmfzQGwYzvHop2w5SJOmojTD0b-%mEccY((J3vub#-xue2^}(cERnTJ_*|f5F1c znf>^ebTQOOQDDlH6AhC>Fzx52W?5uftr*3SI($OqQW&0@A_M>!e?E{XQnyz%WW6E} zVbjiL2;zns7f{0P9v~3PNIK_9y>E@Ks7ZL2;*j5c#egmV@6nj7g~{61^rd|;>^_n+ z6uJcEc6ZN(Xkf@+-|*CphB00Arxgqi_IOW}ietrA^g7slL{K+6C+YT5exj2J;$^Ou z1@5%yE&{Csc<4E{i9^vDI0^-^^F|4=QwINTr(7SN1jkV&xkVG#3V4bX0amaw9#6h~ z`rr`8EL2Xa`@SJ#TsVL{e<;JSbDSzO>gDPFTl6fcZQIrr)x*_E>4ur1;PZf;NFQ)O z!O(PmAv=uccfY8@2G_-W9Md$Ir8Ke`z&7kE@IUj$k<@WPxj_u=zX~?1uHpb;s{z9g zygz$mfADK!etF+DH)54AoO!`&K6F~{>$d6VGvHioQta-6zK{AzG?WE@h1JXh1jn;# zst)t3gn4SZe|mwF*p(tN98wm+qxz|Y>rVE*A@>a}1GGJg;J#@lHT+Z$>m>zqwd#&( z;_3=I@&0iUYHx{QzQix(c#I&^guF+MAckS#_xJchC64=X$*RvNVCf6h)dY7=G4cuq zXArH8aPNHRdO#lXROZqB&$-aftHM>c*-+FHe#L3OJY-JM8>|xAvjbQ~uT7P9b<&Tw zxa91g;{_r;Z-au@-oN&)-~?7t(?hbr9qP6aG9*A4F^F!N>Xt7r?-sIh1kX|_sNl$9 zh-{PM{5BNIcO}LcI{mov#GD7bejgsDQc+n-Ld?5Dj|O?kr}??JvDlsY&*I4LJ$~ct#T%@l4!I=$GSmUs%VXI{ba82`1FfltBDA7S zLlF&giHU|7+6E$Ql51r(JY@rx@>r%L!Ima#LO+Hh)DlRBITB%A^G1;*(Lu36Brjw6 zX_`Ppe^Mz%*j)xhLOCwiyj~IX;h+af96p>(l)WuObeCEGP7<60yc*QlLXvTnlO)(0 zcvlg0PQ3i#9;Hu-A)b8H*?K z(w@7K&On!}7)G^cjm1c(wy80tbuapDk?;ExHH|@h-U_-X$F;2^1xc2puC&CJK@q8Wc3} zhy&uF8e0C(K^;JX!yIB^)KdQq&6)lk_idR`C)ea|cq`G^4MY4hSgo1CKF8EOxqgf%6Mm0@i zpSxp-@1Hd0%|LZ5htKgqF;g6z1K%nFDNBYgjAzzj0mB-;Tnd2~j{ic``Fl@zGg2Kl zi~0*aK+hCMCnNkvf(zrAL1nOp@gixUzi^kWbf~_Q=7h4^t?4}&QOvlUpCLv?@JGo# z+-ry+b;{qFmvz^ol!5sElB%?@$8c!_(~G6b0q-^8n6FO<*&()g#+I~8wL{ZQcQIh4 zzvbjSvzAce);JX%)N)VxIO$%O5vebqhKs8EIa^WeSUjsej)KSYxcf+aKruZe>9B5` zKum_rK6Ef^gtc3LlmfbL*ugaZZ9irF_z20Es6#m4?HAVf#sCL4BFpNW5r$_b=$i*J zQ+hF~mBS=Ya#_6Juv|HY%v8RcqJsr}XCYkzgZPLN_i!dmOS#kgn)qt#jJihu^4-#K zy3kVXS;Y_Jarcr$H`e^mcwRAmdg1i23HrZ-qoMPY@%(E4C=NmjZMXFQOrQDvzoy|J zrb(Mj(5DV$^sKWPswgydDyS>sce#qjL@SE(pW?u}ltL;soXWt3o(LdjXzV~H zraXTtYDm@(bNWh+P)0d1yeT9KduR$^XeEtJ=#nW45})RhE2|it{)u6umlRSq3KBF$ zD6=S4KeM68aG6|M2$tB1+U5KJhZvGEy13*ugg zi6GG{OoNf=>sf(&8cgH)NEG2^-l)5T6h&-6NWYUoxHwFKcxZ~!XmfNaBoBqS)AaW~ zsN4ct38EMuw36)OiBCjG_gH6Ew=f6)tR(0e{oRHo;$#tl zwMlG;Lq%)zf+p&$TW$7T*zoRR({_MF4J=|;!11{?<>3v4XL+tcq}3tAG$x>>IFvJ9 z7y&)qQZdkxy@5rnMP$~p>3@vUwZ$?V4dYzyfcXq28SPS$t4y; zmtWT7Ol7Gi9Ya?OKdIGkn6hsL6*OJh4GEXDVuM))&RpR^BtfLFjs_G0x8FXoOR-B# zp0W1n$%EQsjm97`@y(kL4-1TDNsNZqUL3WlB~ccq8G*`I$Vp@OWwq4a#|5}udi3jH z$3?w`tv5^KpYJ)Zg$2G5MB8%768 z9^7m@p;Wz;k7n`*E%ka#i2*vMcW>|x@U_**-&LVHaT^K`!u7_vf`SK>+@WGTs~>zL zBu&fbT>K)6`ve%_K0|giP^p1#Iff?m{epo50yr7RyOW6uWe-#w-Mg zXEY7xkU=dUoJXd_7l2vsk6efVXhPt}R*(q_L}rDiH3f+Xn3ze6boc)wrpqAzPNxuP zQEsp)JVJDV0{~_Pt2@i3KTe^SE}dg2A>W|k_%CBP<*7VDeK3Jh;ItMJFo7KQ3w2W~ zq#up>1|myhKvZ^1xJ}lP(uSoXro$nB2qVN`&0m8^;nHm;L7?t_M^K<_I_pf@DI7+M z=RvXW>+2fh#I^jWM&rw3Ls^VPsj0&9X%Q4tURa=Y6>>fG;juVDS7?;}Id=b9+Zq?| zWm1q^Yw!XpXEVh!G|v_oVeKe7@P05Z8!m`=Q&weVB0iruEGRKHOCM+vPPCbbvQ6w4 z;^aOc?)yBb6m=E%@B7+){5El%KH$!`(zE@1oA@Uc1h|T;MX+_+T$04@{1YIbMlBfb zgmWQ6O?BuMf=_7YC*B_FbcNPA^Ns!d^t;$ns8joM$vS_q6DQbFXWr1rzpl$^#lS!T zHtau2OXB>TxGkZVXTyK)oTQJ3=r7L+h{T^t+)#bQSK zauDN-=RD`enVV7`vAZ~b!_;kyt`d{bG!)NJT-RTN<0V}BHS8Ch^!#s-dH{rRI+^18 zfmltsAi_)V<$Qm3T{#$amhMaOPCRP%Y8%(s_8=q~p8#>!gha7rzH_j~jy~26b=SS^ z7%R50IxSj&-F^=NS|?Nl(xw8qo|Z>hapvJ|C4-Lxa?{j423v%EVNx*4TX<-Q!@xALOM;6@_kI=kOh+9V|f%kBbG<0Z^ z&EKwg9?6D}o^SZrwr$H$xWGWX{Gw1o&U2CIY?dILO&Qz2uRuxT3TpBF@UNh* zZJB(f{~^ni5X6w|9XYlNNGx!EE&^BX-pj6+FUwFc{>Q*&VmV05Q4qn<=OoQ4nX$;I zrX!80@KFq)I@BeJVZy0FwkpUR|CWPr1n!~fJCkPbZ!rg-(df^gx>6{owold= z7p~>?QwTDY)q84Rze2HWh3#1&Ipq> zXCWZBLC3JI*uP9EdKusun>D7geXk}yj_Ky`Sfs@*B7;I!`jvAc0hV5_4a2(UpXqi6 zh0fMRj4ta4?1AxpIo3ys`5h)M1l4G(!4lTwu0Ld;IW%vvg(xUPhbSoG zgD5EFjRZq9?Cj4Q(aYX`<6&1&qGjvTbNf0>DDO4r{H;6}wi+%~W#g-20?n}^V%md; ztbK$W;FEU9*mr-vSvz(qBu7Z}2bxMykf0`on01GcN_RCx>K2zLV|Bu!4wRxP<`^QU zg3NGysv1SQmWMsXa6Cw+QBc)<+`0~EB<9hj0aNq*fcxewC@Y^JcXovE%}Yymb?76{ z)6ehw`ZUi!i`Ms)orfJEDnw~0JbB>1;q($~>^yfQE8Q`u>TI~LSGt~aKT>MX5^Ds@ zsmKnWE2`3f{6ggX4!aF87-sM{Z`&LNotu;x2sh#;?4BXrnId3|t_U=^O`$Aao1rnb zz@mm9-u>>MnPgj6DDg(i0o-Bx#XWqAQx4?cjb<@zRHmWO@(gu z3j$N@30?8r~HCdE>^k+tPa*a9>^{zg<;pNvE?Zx4|7TppQt4ODARLBY9N_19sdJP9D z>n<^e%W<~Or{A%i?nAMe#MV1L>zQ{&Ir-;399{@hHpl@WpTWw&t5~P0TlLMX8GMA1 zycXM%++f013RYOF?pSbbx(&V+APg)0Ph5tD^^8H76^@qlVU+nn76wNSg?#V3nj^b! z5H;xO?`kYaFLP$_xiVDUZej@!GJOtShc-7dJ@&YsFFB`}7kY;1dBe0uGBJ7Dips-e zKVNpMH|`s|e6zd8`|RlZde<8F6|n>BT7;}#Qgp2EBodq$@sG)Smhl%hIpFVJ5|-;) z%22HB2Y4~+P!UWdQI%gmA$kbe6$*k(vX*L@EQsS$j=-On+nrN~%2{QkklFwDDh?o1a%nehuaM zApct88bWgwK`%(NIj2#bZKsax^NjdzCBVP74>PwH=V=j0lCsKakHjNmxP`Em=Z>6C z3_610@pCIwr{}?OjGuGamnvmj%d>wEI+W9ZO&lS6iM6ZAIIVAMD$TxesEJfOdWzc! zSSV|O#GH)KZhcBRTj75()N4Uu5TH%rAbt9PY6ZzrrMw%hS+p9Pl?TuPcvk=lyL7Z! zdy_;MP9i=JZ*LRLc@JTD$PKSm@8UFXXA@T#g{@6D3&{UE97ZFY(R1BJYMTJhR93H& z$cfVHg`^s`P2e)5;k0^{$ge=|9Cs<3f6{-q+3K@6NeHB$N6w7q_}Oi2_RJRHWe$CZ zs%~+P>e50{JP*rQ1mTp*CZsD!h_yQnpMapbC%@NVlXY^u=M-J6r^Wo*XY+8vsl>nk zBe;YiM1;sY0rv6V%u*#lynAxYOcF)l=og3hhk=BKk~7TXIP6sK6MTJ~!-}&++HmOl z$M+z@S?d4hZLN{*N2{|)K@@k@ z3>TLIhi^7cB*M%z_c>Qk>a8N-&p*F1YXxEd8iLP=NO;dbW#Z*ge1YK%M3|Bar5*m}h{_Lx|Ilpthpys!k7T7L?@BmH z{lp*sbFx+Ogdy&;KwvM=oB!OqAFids?poP-yhE%9i2=WR>ST7tn>d+te!H7I;0~$> zDRTlEs>74-WStZcpZft}lTNtD!dMMX>PPs0G(T0o13}Gq;D7v`ut{h?;{W)g z<=`Z-Zdj?_zQ|r3#MtRwh!z+r_XD0)RCf&zivZT)eYvg@?_{1*kk5`8z!VUWk`AO1}5E z@$1i9q?a4?9UrprFX`4Q=s+FB1_qmLJ=x;|p<)nfi`4LLI%lIMg|dCu-+HjfjS+%% z$Vg4Ocn%hYc-kagp>OmE{DdtQ>@e4s_ZsT*d>XojxU8`jw+0<^LjrqB5GCL((%)$F z0UPw8s$N5Fp0noO(z_Q@>^|`uvVYJeaUlx4q&=+~NxNAy#Gt z&Q}6*?py+#4n%YrBT@*lwE9%c$^G|&4d^P)=xeA;PkKLeeI6NIxrz(>*)yPV(70;0 zU{@j}Q!tJAl;;aDAP235f`vYxQaNgmVfaI*#3rT)k4JK213BfO-~urz;n6YMxM6@f z(8=|?a?k};bjet*Fp-2aQZS)5NP$q8qj}xXW{1L1{X8NJZ%|UkJ1M*YFe#k3R0IrU zP%t7g9x#Y9`9sjaPUzZZlR}0-x-X6vp=(f-R4X!piNB#Rp~M6}fO$pEL=XZW5M zaDY#g1cVekkiuJ~fWI1&hNItFbl-B#5!|9*?ypeElv9{wnhC+ghMlo{WfJEglGm&1 zr+=3b@c&F}{;{>Tb>=d4HIlvp3)zeU&IhBrYPK$KNDBRdSG*-Lh{S87Gws=<2}#?% z@Nw>jMhPRgqxdvZ`)39ypg3*jiJ^YL> z4Gmc=&+S%wvg6~6$_IO>MOIYo&5wfKZfhuYHD_@;5g#%>*#0awl=!Lt(H;I zcOxu85u40`p&C2Qs+VzzOm4D!a*JhMacYlG=gbKCo5Y03&l2ZBC$*>z`P}p2D>y%x z`lXzl8kUdA>~6{%?{yc(;tRO_RgLw-CuJRCdeCJYK2H5nA+{5D_@NlGV%A~z={Eky zMBNI%wXW;em=&DVL{dWjj}1%+0@MxJt)>zV@4pHG=0>xV$N4o8FQNMVQdy?1PD@x2 zuzGt}=t66Lcu#n2)_m%QA7gxD>w~camztyHItGxw{{2LT`%G$+j`U$!p&~M^!${ye z&HfJF($T;!{Ns8ZsOyYd-JL!p9Y8%*VeCHd-2S0C(s(kox?oN;5j81I?r@WaPUQ z$yfb`YW}fwjMfZ=ztxNG{M5j00(aFV9ZPoicy};cNJTC)11K%zJ#<}ty7uN?BSv+!Bgh#& z#otlbeqO-F8EtvCCRRA${AszO5?$0&$AUG464A`3MIDtmaSYMJTMo4x>pi0Y%4TTn zSJ6LN5lO?R+}g*)myhi)rLBpG>LvgM(NuL`^(9U8k}HSkJ;u|<1XDnmsuHC*q-{&| zfX%1nQp?C&*wnW);@_9Ch8dc6Cp`{Egag|TCO{XDq}teprN3>MRT zH5Jy6Z2|10<-&D=R34mg${N&MeY#o#7n*#e_cDo)R0$N}GjI-Ez{8<5N$Jo|L@%0- z8KIDNKsuK8Y-%^g1GMn4a{X1GlFHT))FrL=gEEYyjpj zx@~C+ogi}QEolKEY+)R&PrV7Cv+)E_d9nU~L}H`Oc0_hlH^$}n>f+;>+k6(65sO=r zzos0!hPhz>rH#==-{q)^XinPZ?4`XXqC=9$P2%#%fJ823AbZTl%Q^YjG8e=HWr)W2vXg{&;(IN6*&hxNTWw18Gbxj!;GK0?&nFeM~<#7 zUHS*d=C1v`^)j7uBp+qTJO14{C6u=No%iv9-mbvx-hLF>BhVxn|LBB81GZKY5*Pda z|4Z2mJg{rn;?wKC-}0`eLr(v=->XFclf0Zz$7I2;&G4tE}lJf(zQb zs$kmxIK(Fr<4D~?l&3TBPpn&SY2#E2X|deS61LPHH;G)r3xvF?Xl(3vjsKBsWw1{G zCi4Vz~Fy?ADQ_WQjfMa{lFMhEG9{_KbL&)dOrw1$<# zv)Oxwt1GA-HFfiu@@XIRMq?dRj$b!(dn(5~b~&QG8qvd18rCy+_4M@e$D{GH$PC!n z+~xLlZ4Jfp_I&IrAT?o!ifw%POXnS^ru&6a9>htQS(8Y}{%5g+Oa{gOEQ{(3&_J|`PdK{@K1qdHO=Lez4HWA z`SztWTXoGbj$d|`jMNEtwNqrI{S`cC^C)w8EpsS8#Y{e;mlIH%Mz(uDR(g>8BFU@g zwzG3D(D#YYVkUQUzkD*bSBoWj(TMtx#@NDnI{CjU9`xjWaI!CAb z-Y9orQ~5_-cxWfzaMNfx9V;hKm88$ENp$nBL-u*-l90T6sOW-NG} z?8j?L{_&5n^i7vk;Yc6K;)~cXLOm;wp8k_!CiRPdFHL-mV>ed{ojhHSE9(v3b$i!; z4u!lN8Xs}+#)h9h=rEE#GsA+PUv|E=L%;10DlA;jTXwQyxUgkT4r%WWe7DkH2(7=? zp!YDsUWivwTnHr}_qCj>is6aEyToGJ5c7YkD(GV3%0c(Y{@6FY;U}-6GEIAB`$>htx1q*S*lAkAK>q7EW4NAaw)lrLN)5DBN5N zPgK3I!U$i~kz_QD(z>E&cnS(9#Mix1mkU!=2@l{OTw5fNDc}3)vVuv{szfkYW|Ado z17>`Am&21X7{;`M1#`t{AlWo#Y>~bfVDiuL7K(Tw+XM4{1l2;jxlfa=kL^j|6)yqO zVcV9(mKSqw);_+h%LlM~%?n0IM9OLw$9VU6cg{w=H?QDCySH#lGNEe8^0)At5Bz%B z(-`WPXHhLY%;O!u(#kW;wX3FUA0Kkjm8E8iNXS{~hqyAnl5grv39A^kt%@sNfI461 zT%Fn4iDRzFH+#HaKFrygH44dkJ-8(yXCk*gAwTu)wgFvhD>lH?x;Y~WZq7odjYhXI zbr(;cYQMCuUvA$i)cYTLgB(=sd8R61PBX{r@OP`sGYqBYs?6OOOZ!H?xL~DzJ`DOC z`T1gE;^?Xjw3|(MU~pidOgA~TZ{y7@ecwVNm^DUxsld3IzTt<#1`cVxE9ii<4r1A_ zMfUl3SYVy(NX`nDZAuW5&Gni!1W!Gw*5{TsfhQ&+5^X>0keO>Vh z=~-{=F#KqMeRvSv`vL;HH9<|k&KvMtJM)dCJ~qDcGM)8bb_#3H;G7B#H7c_X33+Ek z>T;*0MU@Y>(Guifbdfqo?d~es_ZV;du;-r|ci*sO%T5C;9#-9a17lgEc2%IuEPGHw zzEu%zVV_>cxu`JkgZDnVz73piYWYR9++1NEMDGl&j?%s)Q?%tC)aPWd_??^|%vuN3 z+6w*h$-HbfMqMh~q;_J4TDMMaVt#s{?gKekgIG#`AUR(={SKVZjNV`NGC;b_Ra z4S$MZKa~dlek|E}`8kwy(~0RoUa<$2qo-b=y}z&PGqSy|`D>b5j~V zRjpV4T{^Wr%E(CR_qH-3Ud2v86OZ`oRCIfhb+^$0O}2!a(`)mI7YT0lhika0-0hR`1U&2K zU9dr`5WQ17)}_zuPF>E2ej&jc&vgw@{{$B*wXv2^dGh+5?Bhg6HsIVFs$9uc3yI(Q?8#QJC>S99*lTIh1bo@7r@AF zOcJYGJm`?c{kl%&N7Xs|f3?A0qeHFbX4+*OpAn8p9+)wn2FH&xKEi1x|SwD^p% za)hOmqu~-A_L-UJ*B{{*vN6}CPhH@j^LGJeR@Gnql)CMD>{NT~X4G5sARlZwUfAz~ z=K_z|-#dI@GcSFn)=IE>dJn+1LC3$K2ZC-}Hyr&kVA850){5!1OJbxo-TXCR_hR3m zM_uVD#u0)$Yge;E{{46G@7^&xQ|>X9=KoLa&S&o#zCqU-ey9dLF!SVx#d#+3%ESvS zjh<8*Mgm3x(~BL~hu}X~t)@M6%miZ?$$~5dxL*Jl<~=fomwsO{w8b46hJt&7vF<&3 zr*v2$uK+K0CwNA9G-4h13*DJ{&rw^zN;ZaeA9MVQr+UA76%z3?Sjb@JZ#694Y0!mz zZjKOq^R?}w-E_bVc-7i`eF9$#cTtdUfMjDKwDFWo&uW|q1da0vW|8Mu539M=b}Txv z^XSNcsRFSl@|Y+tcXj`!32aZ2@JppTt(ZC2e@l#b~wQD%EEj zD4oB}rs7fr%-p>L5@lqc#MV3vu*TE5&g5vcz;MhGY7mBS_8vEX*I#nU2Te^;D>VKl zdfIlSSf(7wz_!Rqc6e2QtK+5GBdJ*~RxVbm8}`(yMNzrGk`psz zq@C|$Jmirxc+umv>nJs6U_i;y;#alBcw+~$Z#Llbb<3HOg>l_2YoZr)WvQW*t_x>v7%bmVJkBTrIx5L30GtpHl?& z*|1L;wP}P}Cl{%}p4C`)ydEj2sNGzRDVkK1;jEqFlm33HgP0Q8C1|Sxys@E27DM#G zelj@sq|u)c48mkR+Xz0mYg%GH_^ZGE*DEEeQW$!FC_c=|15V|kufn5TKt<+F9%Ft% zPrS-iAVoiZ&0X5X>FHLt;70Fw&)H2mCWk(A!N2GS(MwWp{M91G%Y@G~T@l^!RvQkQ&w0dH!=m}GQxz0)ik&44-}D?&)aAcDB|=Ai%~Z?d z5-%G8s~4-uJ2u$e|w#5TG8(8j$s8B8+!>~b=4^x3~ma3k4e z#%*S|bxKk@cRgB0B>byZVju}a2J-Q5fDEw7Qc@M=W3l$?jv)m$CbFKX*N8lfYqv$R zi2*e7USmJqZZ9=K(*UtJH4wYLS0yD4Zm)WTpNn5+f<%KGK&?!y!D~OAcTq0-yYwFg zUZQ0w8Y2*63Ji^>Q7uC43q#t#i!$H%;2R0wAFP!gd|ex8Ag0ThUbGTAY8AAXVL=;~ zQl`jnL9w&0eIxRNL_35IQ@497jSB6PFCtOS!d%yJa9Nbp`sK_k^Rjygn0A_);bMni;a2k$~^r*~5FTnuJ8CYn@O00o=+ zFG0gA=87yHl^D!pj9xgLE1_&4TuX&ig>agIT_xqwUd^)mUvv(>vyv*I_P98gFmRVG)1lE=ou#8$NvC;^kqMqr@D9b zb=70Ox72ZMeDYM}(OOu!Ze+e}Q_1}?$glAep$4H99Ov<6Rjh}TuEN%#(bZ(@*IV{E zUf0mh^^H-)zO~Dv?hkv3s77yZB_xE}Cz2%dr>~QXlpj@A-IUWz;V8#3tyaK0Kz>{f z3)a!W@n)&_xUc1pzwmC=%)(0_3mZTg{IIdyOD0f@`o>H$=E;2N9OM3%Si9!sbSZ ztodE0^Q{3ZzliwPC{4)T8+~y#SgveU*aIX(Qfq)kJq2^2N0bwudd!ADCq z5^RUfuR?oilro8nWnqaxd3A1TtHX%Uy-6BNEfVay%Hk0i(`DWuF1*+3#iu7?ZLk`} zSJwxVDhb`C1=FbO>y5B|OBJKmiTZCBMeR(=WW`k1340e>Ory5HpDd_o_Ub8V6WPj` z6o9^J_s6ZKnhJ@Zp}+An27Gl~YCzS1gJ-8{fCpG8GJx*xbn}4Ttv~LKLkz__>4ISh zlUHEiYiVSm;FnVb8b|?_)Si@diD2k58WkXMGMq}H&a8d9C#NJhMuk&HxK`_a)(fP3^xiEx3`F1n|t zW}0UKuk)W&(?ZU8;|jQX{&RYW8i$b``y!}^>s#UL2!}_qHG?t+)sO}vi8>t8wOj)} z=HAy)xKKSN;9jFFuHbhb8y)NfYy~wqK&^BGK8{5^|B9MEH@!uZRKvQ8elSP0j%d!0 z``Ta<)#dVZ%`w2pt4iHYdAi30hGaSLqlUe%s1Tj44n-Tv&pU8AT@M^Y`mFA&D|+w| zB5k=VJ1?i}d#t}z)aQ8VX0u<(vjP@juC8P=4_TI6=7pll52sezzFA#wy$=1V8rNmc z*e+Lytb^xREb+bR*32}ht>6&EO@L+>A{7fu&eXe#%vm72Nm1h*%A2jm3887uL&CEN zv}0l%hWmyLMk)5u8FmO|$a)YzHfp5{hR?{;LqTB|!{t|VvCT=|^H0 zZb-$l2uzN`a`nyA8;xYoQiccx8@^##_(6opzPT%-k=P{fLVpwDQrFB6gcn+C( z0Q%LX1(|yuo_^~JIc8ukL2`idj}tf}%(g3Tq(H-Ok*~P4emHpk-f$nAQKo2Tk11Js zU_?M;*v6+ITK)pft}Uq^&Iz-OHzQzA-G%+mg1LTtOxvYJQQ)Sx?Bc+X<1#Ta<&(4! z*2hJ~l1SxIe3uqOu9RL%q2tW9aEYuh9J5ip%VxEb?JsvYBMvNmkdd>@S8$j_mLJn~ z6z`)JV4cCaAad+n-u8I{8u5kx8M~cbTge0P0^NBI&%rp)Yb9pc4bHMu>svTG#7OcH zyaDlxEak!|Ms3oVSs9Q0-KzIsea-Uf!t3}KLE4cJP#y(ihpyrAAhY)+N&jt%&}1)( z4=&mEBzaO-z|VJ$d#>P@IC##g0-e-#6*DDLw&MOcD_eH&VZ7fy;I-pl9CZCRnGxcfG2g>EH$cF_GH9fx0?DH9_*3Q%VVsZ=@I>$ zXc|{7p<~zZ`}|CYa?SuRwWMsp+P0ed(a>wIfZNV!lT;q70f9aq$3eTR_#?4%o<-|< zOKp^a1L!!LKxA+JQ_jz;v+Pzu$Bsr?Zbk|kJk^*Vzjeut3Fj$UjPpri)&8HJ-YP1u zt!Wx1xVyVM!QI{6U4sON;IeQhxVuB};OoK6GmKf^!p|Uz8|~01Tl$R{+vHMxXX7Rh zVnttyuvJ?n=;-5g5w$!YsDSyYqz&85SZQ8u(hb6is7Z zbs>fIb}-E8NPaw5D_E`y7SwI@GULF-T_76(!Pi*O?L;(fJ~nPgNb*33E8Ob2 zLwwTe^xn^qPYn{}YzOI^`HsYe8gXr(R2ju*myRAY%=X2nUOr~YSe$4mKJJwR z%Rr1_$4`f4lo|E6>S)0SRS}vF&<*em$_RcJ8A1s&oz#=L$(k0#9jARW#f`&x!!!ik zwvx8h-Q~420>TqCpZlr`yYF-0?E-%W&x@ee4iF=Xik%pVtQWWj^;fs-2!v6(yj!FG zG!7@v=aoMnUXI^M+VPON3RWVf$_ten2l4>-TA>d2Vw>ZvX3NJl7$h?zc6(Gw94iNu z)tv4>iRN+Z{57Wd%sVBeJuFKczBMDQg)Vnbg0EIK2vFk%lhLsyl8&(RZ2_b`uGviQ zIkCn8kfi9~>%E8OkG`{bLK|$vH+C~8Ot(EHr&XStSY036Y!?GNL`xv%P4?hvAhsm^ zxCSB4fyEE*BZQ9mhZSsGb1@ala~a##Ut#iN<)ru8mV@93Usvvfl{*Nz*%&B0sg-^D z!OxB3-4oG7zf?{tXT1Lf!w#lf)F20^-6@{n#Igy48yeYYwC+TE2Qbmh%GQbKyEZu+ zSX(xOS{_e<&gUq3fv^U)E+!3kK#YpVqjOvdPLkE@nuDx{NpXG=r2{cz-%3V~z~dYY zO|@7nYBGXcBEdm*FPWWpLACfup|}r|Zo5^<3scs?5JLh#uPTdF43!r~LWvhlok&9J ziHz+#wZz}_z7$#Z$!docvYU?nYk@fffet&x9N&qB>;2Y8n$`AqJ6_pi})Rm)WaVC zY-4pJNUz}gSEtn4fm#-CQPwB%M*C=eGU{*abYx2vS2!Li<$0t`!27E?A=FdRH|igZ z6>^&M&hVA+k&^cOLfA30loPZUVGtjzmir4(psvV)``-1J~Lt zU{2Z@Qj4JwMZs=hDV1BTW0fe5#!mcMGm!BD56Car930OAqCCWP=Uq{PwVf>Zh>;@J zjrokk{zEbY!!s@0zz_`T`s%%@d6c;S&cF@~6L=LU@}d1(%FOpARv1aDuY~^`O;lN2 zNo>4vZn0YqqVj|}^Pv`VpDN=7J*-%Ks0+bZ2h{n(+eJ>oVhwL%Oh%b*-0|xZp z?*1B!D=)wNw#`jb6{F#V}9$C?7XF1N$Ef8|9S1^bmkd;(Ek@((uRFhYbw^&Ms6XuhV>Mb z*YNEfdc>vkmifs|X$H@^+7qfBzDs8lo2At{UlRgi|0avEvf^Qtn&1~3k))XYa2mS1V+loX zUYHW(-s9oB28O)8Z7LkRLM==jkuMY5eZ)dM$>N(vbaG?^=H)T-Eh3nE9ZAb=fCL&N zZv;(e0`Qk3-=kmFTY-p(-c?{aJLJdo`n7&SH8DlFx6zo^o`p*%0Nib#OU9kEv~@m( z!RiRf7)Mk0Y+IQWnH_ofE@+kGD#R@RjY~C}$TB+c$``g9(O#9^fD? zLRElny6i~bgdO%M7pw{Q3WAR5iM{%geL-Mg1_T)A%LfYLSdfH8y*~Z$k4Wo*OpZP5 zo8`tvT>ok~H$1Q8lV@7JJByFG2<{@S_`?~tdH79gmxOu}@zHcOOgXiGq=Z+ahIc+a zBGo%WkS`axA*bYW$DclW74d+u-L$nloDMCPHnM#uxP?=@HOJb2GooHu0A}s9)>q$Z z1|T{7xBO&ahNjOUjE|E`MD@#nJc8=}-2WXkvA`aOVq_N`%`y7H{V$$tf=btm#k)=c z_x7!|yy>5d@ANQKDCfmr@oBt=d>b!AN?Y>5lo|LfdgTk) zK1Zly73#TovzeYp9uy;5Lf&-z4wwNvHvrd4D1ytWV3dRf2cg>uGuXxxRiTB!;|uG% zVaF`D9(g3>_7NN*5}b6K0SxZ@$NQV{u^FlzT0vF#jXyHsES^UP)c8huYv|V2!;*A@ z=?W$uk!XTy686ftf)$p}RO|atL0|Y`n3o|vzx8H_#66+=?mAjyyc$~6(X8CuIFREz zqQLmLBO-)8hUB0Kdx~W&+f35x1wt0=d&j<&>m+^)UzYDovx5?jTABrUGF-G6=$-P9 z?0l{BeBZL;`q?2#B!0=h_#1c2Jj2s*Ft1wRHK3`XJH}^Up7uKD*giOJZJF<4{q*bR z+BO9x3KaI*Y(PO$!y`6gO|uHS9Wd4)IKVKMD^gW?)f2e9*0;lJJ&2FA$~?qD`HSZD ztEuflhhn+^ffgo=9DPtoTQ}3p%|p*`+s$Ac@7zdt*jMVD>Z+!uYwbr&eP5$T73*$& zn823M)GQRIdWzWZg?%KkG{D}sxI&S}Y8D-fw0g;Z>Qd6h5rmw+9>iyL^X*yHWKjW>sTD7<@En=j zwA*jZ5^2XnVbG7B6QH*0fJ?c;!7OYVa2k9(}|ACfJEO@vEXvJ#dI>O zw#PTZA&aQi8%ark2=TRWNARYQyDOk3ObK5AYruFlIvZ-16ZZL;%-0WP+bh_Os&@;R z+j3pnG|7Hc)E8}V$_UKUlP9Gt`XxDokB3pd-o zUN--WAc7_Wr8}}1K%ntyYm(N9Iv{#`@ra%@7%p-H@+x%hydpc0@3cFK&aF$!=UE|! z%p4a@Izm(aF>u?x7JPVT=ouT0GO6q%hSONoCi{4Ssdi@l?EGBY0Y(?W1>5Wj|6U}} zKh9J3mtps7q=2^K(OTEwFhPs1AQnMq;Oi+tr&Q!F=bPrI{vOOcCD=rQkQ5meCIX_& ziPZ6)sVbvgsgChAB~GZ1-)?6D%UY{7X5#!10_r{Pd{R9ZDXb4G(Q1mwx4SFOn&||; z1k3*mJ~Uc|A{PSFIGO4T_79B2`UqRW1Io;b!bl(N0z0$3>kw=`fx}2#h+EXcR@kK9 z4}|}6Xpl*U0spMV{p4EJgCx@^UY5y0s%rd6N?tzjLb%4D>$_{X6OOjI&Dz)jze!W|Iu9BF6%?kV}D~Vj@&W zu>g-M&Dksvm-NzA&|KB*wWC-2M`OYFOYb#_?Vrk{04$b;9zl5&$x2^%bs|kAr82dR zu91n%<4+}t6vecB9p z3%;}}(A7-aof!<+Yo2vM__lkf;>^Gq#p&*N!iDcPlME?*qaeUY^EWKS3TU-`+3fj< z)~&>XmyUli)+tqCSM;ys(q4puXY*foY zZIDlp`c(tI)klH63fb422vwdc?XZdkxikk^($g%bG@IkIyV8Icf9y~8RbcM zLnNgVQaOy41*w_yn|}6A;H;C=8Rv{0;d5y%Qi0N7q}Y#ym~q{*`;~bDLwR~5zpqn( z&5D{Juh=gdba_sni$L(Ye6wA`TAZiFQ1mZ^Jil1+-FB6X!#k35+Z5aVHUY2w9Iut5 zMpw^6T`q4I)d*|xM&8r5QwtIV993@xjpT2_l;@fJU3@&SOjClz_ZP%Bz5;}Y^$+=C z9{q1ml7s9%O{iSKMs;teRCXol3*$pTF<-XE358mYRB^sA71`K}Mzl^baUXe8IxH?vM)$P&DkE@FD4i~v4az2jTY&RlyEbGIG6o|sg=wvU|pPVt=N2Yy1>fei5 zrOXCYcUb-fbDFcu0)TIvaWpQ>r+|~=o6A5Mp^a;1(KDcoJA*;IXpI~^Ua^3Oi+M78l7zu+*Wr z^%`5++0&`EqkkLf>cb(4oW)59%G(V}gjVjEr*=tMq-`>JhZTzyu~0h@!@S0sOe=_O zTKgCdKTpm|OiVUBFcszeYvHlZ55w`!MNl!Mqq%h@@iFl+x7+fI-<(^&i3A{^@MBk? zRa@9^|HkKC!c#k<$+s6PiiPls2p2DnngDqFr`}|*|x!@JzANCf0_BbgJ1>=!IjFMmGaezxxJT$5c)m zr~4z-4-L_#A|$kmT9nHuZ=(LQ`YEk!>Wei!;DdOu5v`x_^M(PyzzI}-j%r=D`*o|| zEFnzuk7zI_O=fENvVN_t;_Ekk?#o!W+QK@hJGAxmD+j_~6C3cV=@rE0H;&rc+L2Wl z+7sN27w30XdDA7EyKaAYB+GmoLO?%n)h*%hP7J2(UBAgbuX*GxTMEQ&S9HQUysvh6 zFkttH)FG8JxA+nQBNS|{p`&0}{lmUv>~YWe&102}15ldGqk@BlPRQL{uy ze}wl4)-Mk1yQB}{25-S%D4BOI;5#3w=wAzRW_U8wK3`<1r}6^*k5Fizma`0$_nL7X zaeqi1h+j6Z0ME^%Vyf2uXun3t=tc}A++e+qN8tCcui37Fr}%hnsy%je7j&1%5CLWA zu=bG7zL&BNkK(7V{@^wHoidE)o&LaGXv;ayBsIlf>(=cbRVR?bUb2Cws;a-@#C?8$ z?HuS>$nW$0_Y!{C%goX88tjy(aPwKLjTaX?wo&pLJyW>(i{%U?SGAw>MiOb%`y=4D z{at$P#VVZx26JO-krr>~it(q-wvjf8gL_-+odSO@NiF+0A3K)8yb!GEISeiK8M-iW z8`x~){DEFtg{+g=?K`DOCV$KgF_};Dpux*tm-yV^2KBdQU%o?QvpvN*@$vQeFS+Jn zsd#Q<5_4Mk%I)Kfn2v6uHdc7_!OtwKKd&b@X(Y9&fTO>})0?xj_YgPqWzc2Xkf@&ReEvt^DryzB))y3I&xFlhw?AHbU zY==p;CNXdyT9R*UYXVD)>lWYW5Qqk{5s0+kxmr~c@TTC?RcwYUBH7k?qz2IJPb5R(S8>w-|J^p^S? zQ2Hv|w!e&gKo4J$yTo|aQ5ngr98xyKa;MO^dpX|a_$=t4{_~|Gn{p%0Jx8+|!+0of z`w3udEVTb+;`kx$K@GDeM2lQqOy51$j=UiDy8!45q@eDvWYTnQj^{@_>371r`a0trg|{q}?Z_RN z_m22u76*p;yl6UO-2DyakBwVtz=Ha#^$c+fuP;AOTcS$lo!tF8>6jx$G==dmU-f1C z50J#|K+4$dI*{-dF9p7#J0}AYQfdb`MwJk}Y1RoMsQMLB2&&!Wt;SDfKjNJ90R_-7 zUAu7yf>gL zz$t!r@*}9NdbZ3YP|h~?zS*8**9b~!CAlP4=g{?p>l{USU& z)ks^Zf7vHFP76L5)M_tFMF>IG!~&hW-&wks>I%{0X$DcvU|D(ZHKQmjX%WmaKYQb1 zc~TvOH-rNQI>gGJbSnA0DqmC`xb%1_uBMtUgM#Y^n{r>9SJxqv$@@eIfs&;aTT_0z zNDpPx?Nm9&QSx<1<|(h^s&klw6;um-0u)UqRgdkg!W<8R;RIOlP0Y&K=gf8SC3fbk zrcUVX0(<|>-B^@Z6j!r31{{RyK@oj-$NM0L)nBSGIO|!0XByJD&7yx$4I{ov;C!x< z`e8Hdmw29Mz9n=t6@*?*0L-?lRy)n~WTY?Sj>GL=LY?-b<}n;0Z5PyAoNzLX%8^us zQ{#k4Ss9~#KLnpSihv^3t(l2xAlnV12#GX(Vt{@FG#9-#`u0^Sm;<{%6odQVazC)5 z6UdWL-aQJ$QQ)2%G=%m5?|-l{NulQIf~R=FfE$*h_7J#0hj16#1Krqw-wZu zIQ1mNZNDX4*#b>Ihh)g6FvJ5I#f8^EA?c4>HT!-(V*MZQ+V`mssP{^@@q?qC@p7MS z9?TT2b}xh)#$)E)x<5&1(iRVkg{^ScoB7u#w#>_JLLI~J3%;nXrU@pVujC2skJG}X zOSbm{^kfJNA3+m?z%Je;;x3L=N1qg2=pQtm9W985kut*#^4SzkJ} z&cgshQ+oZJL-@dy+`&os-nY#<#0)=}*5&{9d==QX zJOE0klUsyqr=6bOSU={`XYtzU%~>FQJvrhpD#3rXK@4>Q1Yy7LZZrl_kju{@E=7yR zEtMun<(Si#HWy9rQB;ie|ECZTFz1X+fBKJt`Q_u)R&&hE|7LJYUaqIx>70f$)5N ztl!V4aKyD`fd1J3qT%qW=CpvB^rzElsOq1G-r%%= z!gJh=R`W4V?q`Sx-I59rAP$&bdgA#@b@@-cAiNNCH(Wdg$*ny9HLNziyF_FU-Rf1N zbZ9?<(nd#q8E~QEo0I|H%$SaDcQUa1hFYdy6DSM<_;uLk-zli&y!@@&m+}=WqjAnO ztDYZSe)unAE5rB9p%khr8*}gvCm}07=0~2vKduND;gfX2t#H;$5MStC_fQE$qQ-9C zop0DtSy3lyCXy!Immq#h>*~qj>9}g!uo3P8EdnU#-4O0NtHp?aTz)ho>N1DwO}~YwpN9N?KVkeiD_O= zK9nzt5B)^2AXljp?Zk!f-3s#rIBv{W5&zSqQ;x!;D^q@o{k`e`OO~q=SBZY27DM>S za#k%MR?afnlIgGwKu9YEP#EAf8w{St*(@1ZWcfz9Fkh*9Dh37_;_UoPH~6>)|J6q~ zCn4^`k0+Ql!-{{EoK14{Stuqm< zJ#e@+khw~bIlmKXSX%#tYT8=zmZ+9$jtCZ4;ym}RE<0&dbcbzP7Ez~SfE;7hB~?l? zXA`ypBk!|d9VG^jvpSqLS&4-LeJ=Ix+tm%%XiXA=;MLTzkA_bwBb5&gKLIJw zGqxpx5qGA;n=8uJ_V4o}=gzOFYsALD9wrFEemSf-FnY^Qpn&B(`5pdCk_>2=geP>8SoWv^tY3Mit_596TEGCB)1?(dzl~*+OCARs8 z+vZHZz74f+ex?0mzWPu80fMIK4~fRV*dM^QsrBmKdnCR6_CD0pNM5iKyc=>f7#t@r zj1u;DHY5wUyC3eB1*!kOvM?QCr%&AM@U52Oue8@1BncQyqJE7o)% zPq(sU+LDDi^N+~}VRb4)rRbM+ih&_7$%Ia9X)h)@45{C`*>CqXZ{!V?X>>COopUL?6Maem!IGMq?CL|u9f=Jtc9u#zxhVhBT3db zg_pY5!9%&aV@@LLK9y;o@Ftdni9ag_R44d)+!4cg?fYzKJyCN@h54d{V~w@^D~&O& zzl|Co8}qQL4UE_7&f5P&OTM``I~spE+erSDx>muOiJj&w8D;`rO% zVyuPjm-Qo4n-9cEL41IDS3;1--7}mQmVtd`P*MhkdS6Nurh$Vp%)GcLRN5FFV3^;v z=j^_hD0=-vqbOnToM(JZwOu7~WLE+G>He?YChsCRABn!`+G&Ab3?hMIS6i2=W$gST zh3)V*gs%-9D;vLytwS4aI;=QeqlzWJO;x7g}?r#+fqiG@D=t;TUVKV_bPu`c} zP_{K$h>+~159C#{1`D@Q!avAmfYq$-N|Fj<8PrI_fXz#CbCwYlZ;F;p=-)&JGJTto z2G{?V2~cj;Bha6FE{c_$7ng}Bm1wRNH0}|PYU%AW$}xC>!+M?-%2hT>+f=f^2vk^E zAo7GSCm?On67)s}NoTzux-|Dwtn!3->hvjl#*2PQO%H|O`%90zpVDFpcn=1kkNZU{ zqK%4;*P$2gZa5>&dG>*Z5OWtAx%8Rw#d9U4of$oI2oa{#KfOaY{dVvawOE4KdmIgk z#^L^63bDk~FD=>Qp5lMpb}`lZFB$N9fX1ugICpj`A8#VUt*AL3z*k4zw9anQ;%c2O z2-^g*V;V+JiG=!E+%Gfhu@kA#5Bu5z%vgF|%$T_jC=B(OZmqzv|c zNgsn+7uCI{+K~&_DiGIGu-PsD7NqgnV->^Q$XqVREd_l?L1^C-jv)XhgL(t>iFdk9iLxMdN$cE`+{{YhmyKR-nbE++~0cNU&VopFu|I<{`9h@813r~ z3Q%GMx>=}ZPb1-w&2jEJC}{uRF35sNJB;rCTcoM1W28k-?gu?W}1@P3UU$$@6$>lMzuNk(&@f5-^?D^!atwYHEN zDz=8#HdG8iQvj71P^O>jbq2gWjpPfr8;!WV3m`78K#f8)G}FQ2Hq=$*d%U5*oa@Zx zw>WPv;0UbY4E}5zj+8M1bEb6as9{&BT5211Rkqhr%NgB3O5r=uFAjhT!e*m zApV0AjEyT|yV@6-VOe}2w1IsRtSUF$W~A4JY5@|zQ_!iY?wRKP36;B`-r z8~ekBxY7=Z-oeC5^4yqj3fp)zc^l!L9ryv%rDPy*8Dja{CoaiVd^|TR0p1@j&uZFz9DFtzU{Le zuG3H8YTsTChVrTyHueA)99fnwR2q2MW&g5)tsUcR%c?rRt~7*=EkrPqKEVb-iUyI2#*@mR+R;LyRL8)>hmos( zTwBAD@;6eYVOYp-HgXPZ3Yk=k`8X{ zWR4~N^E)+L2-asT-a=Ua-q>rhBlJBIgk?K8*ZtRgG11h5J;u)6F{(s($pGk!0LQxG zREF2IN2F9XFmRD@( zABAg!k#2>;4pYn>u(T0&`qOJwgmCebI^(EV)QybCO=k;%jb3R7AJXc+GO@#bii4c6 zLv!y()(etV_uLt@EUudUzJ~$n{(L7MVGD)Us5}P$nEbL*SOn%CS|5N;m*6{=e|hFv zg8RQM8;bMRPnn;S_bTTDu~Dt@`&;eV`jYn9)52P9?bl^xMN($jW&FKscwUM6 zKk_LfP=~-Aqc5$!se_w;LiTn*x|8b92gsz&4v1g^^P($3ccz*kLBg|49)ubiMWZ@y zYCR8f$7x#QcR5jrz9Nsbneop2{=kvwu=6)#TVEHe6xUdX2Phe*J z9GN*R!8~UDy$N`DwfR9@To(_F-8S`LmLk-m*vdV-(VMSx(d_*hFn&9+cF4ltyPkLItmfRoM<*^MaL_F@;w zhPgF_mg(LPX;GCtiE3Td2^Wo_CsxvKqRgadq+c$6lZ4z1cFr|MIjI3>noy<<&)JE>Sf8k;*^AGOu_VlHqEk0d4kT(eLziZ~7=)FMCotQ;FUr*dtH-evVn zUJm4vX4}SBx!z7+y=jWAEoH3FjwV469IMg!3&c6a{gy?%W@za`su=+bWa`j? z&=seB9tSdl0)9B87C3_^wg+KW_BIe~od5?!t%gM){2g zc-3|c*sVP5Y+hP6L9!T=yDMH|OB6V^Oj5aU>M8FrvC?~Hg3iypbf1c7RNb*9InwhZ z#GA9^VjC;=FDr{rWxpojr-kbmfyl%?zY;iFMUhhxhCjR$jYx_aQPP_eCDz)OsdM1$ z6LhReuk|4^nR=-s!kLQWG2M$(c+y2XuEVP8%E6-(!q+qCLN;N#=%!K+-JD@N4cCmPdUmGu2?z#V@Z0y^@OdVVCY zo)6@^VO8rJBh#2$sJMf!wb9YChLBEOA-)<4U+Q5YO{=fL}W31y1wFEs>x*Z_y;UQv9so&$dJyW|Z`(Fl3Ho9Ft-2Z%=44;N+#KGD{!gv!s3zTCP$Q%#d85zBV{lf#MR0VYgl`?vbvCb(n2qloL zm52f&>CVw((-)T+DmNuO4>}lljhN=$eBcNP0`fW;pIt{INLMMk04sz*S|McwhhOqp zMNsDh64FxLkfwik1?477ReXmK1m5Rzx``!x%-E>BPb%L9oIgl@tTr}#kNp6L@<;3e%086>aGtLZlmoNb^Gm8B*zo9 z?0H%eYY}*_$mpll0r2rOpo^2Ra#&UW#xufaBe>Um zHOGD{sLNO-*>NnR`O7sc6Q}ze0go!@&80-9WmM-8*I0B`GNJn}pA9Oxn`)((+}%^l zLC1%Sg__lYG@WxGXFSslVTnY;=K6az*Z2Jqswv(lF`W4V73k}iKaDhzsAEU7TT~1R z6^w?ZX6@^jO3X89AcxFyi!f=BcT<9-2+Bz?xp6@MOe4V(35U&vj#gC<8C``d!9u>Q z3)_79eu>=IpDPKOlx|{5F{aE?0^65-fojLoN=W}71rM(DN1}=9Go% zmL-g@b2tE7YJjXsj*E5{1F7C2TUTd=)<;RLvn=x)LxC>7KYA$!)BVjH>EXqRE-J=$ z1d1_$t03tYHJx24-+L5gE}O@|gV2aTz``^!^nW6;-G6JudvI+M~+ga<`o1~4sFXfzzC z_{h&~Ymv}wE)$VqVV*7Fbgp5H9Eb#5wh^66A<$w2qD4RWXGM^0cxPu&)f=M<(zw$M z0bt|P)MAub+WLqXADJiqGyXU6fAOuc3*eUKO%AnJ^$w@i2kV3+Ob64C811EpG6c&)ZFQda|5Zb-U3Pgpy6 zesqT{PdHn1|BF=HiQeJMFNUKGegK8P?x4cZT4=dM92y-W4GU+zpG&Oavs(`jC|R;F zo&14Te{`+c+C03WBe;j2V|}fNp~(TyhEHLmmEkc;GVJ^pFuF+fJH=V9SP>E^Od|V3 z_URxfc@KhcW2N=aq0=(OxuhlqLkoz?k+i`GIC_(i_0-3y#r{9o z^FKIQ{y%suO%SZ_hU9;+BlQ2lmH)wvivPh+nSNk@;xfPVt+l5w=YdIp(Yt$>NrDw( z0Q`Tx#5b0C5<^(SP8$%4bpC2jPy0WDRJkHpp{wX8US$<}4+g31gBVbkn7Mm1jKg}v z?^9jdvn|`va!NoTrl3^D&1}bJcLkR}(|lLyh!;5%*3RDI6`y?f2w9Tq(_O!PYN;;; zN*H_bZ2zV~=VZsXtB^%ET0aX0ylWhaKe}@C_QYwvy0V=u@={~bm$d#N*q9H%)ABN@ z>^)L6InCbhkOd=;^GE`l$4%p&A5zs`^ed_B(I?T~#Xi}px*oWo3dU@@&{ zXzb<2{}CmOwcHsgoWt(R9VxsAazF<%cdm&0Z%776JoU>S)dMH89_M(0p3gXO=u7IW zGCJ%_Q?qCF5MfaB*Q+x4B<`y+SNBWktMJ!##T{i=!i`8mSII{R$5Tvz9^{Du@EzWN zCLP^ar+nua-a`7k9^mtOek>e)`Ox#?7W1t-K#rL1+ZPSnx3_x>w9nUKL}|KxS~WH` z9vFo(e2d!vUPtl)VK68cH`MPq^LoY&`V(MxAxgi&$m(}J9w40C5p%Ix>sW{Ek;<=M zWsR>1@IiUp2dg17t)0#n3Lahk{NLT#5A;uu9KucWy55ij8++AWP`^C$wZno7%jd6g ze0}opJp1VUK%Bn=Er96qfT!EI77}B>PVY5iyu+)4oWn&RjJbt+ezZXrc&!={y4V;) z4Y7zbPUx#iR&8%YNt5?$)tDS}V#_e9Z09vL`V?;u>d=M>Wzi)at7w)DTbE}Y-02uj zVT7qA>se=GB80a)rP>3;s#O%J~q9t6<(E#8P*yre6zGN2mkYm8hQ0{reL)`OvTK_xQ zM|(hpiSfEHGn4WfM!}5o+8*mEW1c*Y3@_r90R4$xl9>6pVrfLPX_VpnRYGTZDRt!Q z7aHtHUwujcuT7)oGSK87sS@2{2_yW=CwFq1Rad~xQ)}JejokOdei9I0LVt_K_Ki>K zcU4csF_G=fgnsREkY}G|mqT$BmOo!MmX@cHERx!zxC6P`V>5jxc|-qAJR;^hjoKqT z7O|sp&C;cMeZMmGO+WlJLxn6n&|51Gzo`|fU842BgaCDDt=hKLTP6{wCc~?pj=?ri zpU9WPHoM&0V)j})fZXN2Ncyw=u*z-C-=zIYIWJn{FJ-Ga$gY%s-fJTFG%dOZ7l(#L z;UR!ANPT)c!|M$sa)|ZaJjaj_s)<|#159pJKp5MmAHAN+l~~Hrp425jGN%OE^h5mG zjh5+Evo!iH$yfF`TO({5nNMxEH%1L$2@9iglL4zn%x}c0(2-I1ea`m6^OAI9W1{Fw+M-JtX|9NuZ*|#3-TU=#Li?qtU9pW54Vgsn!(C`m zokm?4eK*2cO*bZMyfuXGiHWsFZa8JflIKs70nYB$>YZo0F!`0VlF*vD1;#yU5ZvSNRjU!{iGRXsLMk;mnLN^eH$;kK8lPs*v~8Je+AQ zDHx#*ONif*sgZ(faS7l+)CUl6NU-50Hk>`W6d#HJ+@@4;tvmiRQRM|AvasQRXQrI* zA2W^n%*Z`6%Ghwi{Y%E0t2kGv6*Dgm|EytOA@oqRgkkw7u49XS?WP!aLwUWonyBtm z>x_NWMRDNh3Gp#h#yYZ#7M*vGq_9j3P>m4vyqt+lOZYOMjZ)z8ej9QLtAz3<-8eoWu5sk z$&+NP;A)kukoHRAL6K_9b~)aPFu9!m$(m=fV5D^zSV}5iv3}|QU(?Z1+(lY=8xSbc;!s-$=1%$ zSRlv`ipOq`pi{MRBNEn-_Rei*o|lTGdp#9VnAC-cA+t9s<%$ble3m4V!z4#9xl&!}81klI66chXH5+=*Yp(A> z3Kr)$@!=`+eDp$e3tKn}XU?T2fFu_!OJew@SqqvnHl@@r45b4d_+&V|)*M8XQW<5W z+&n=4mryF^b$d zU%};;LXN*aoW znVE3m6+ZjfL@_%cMQE>)V=)+)wt-3bwKf_n#0;0+Lb0&WP;a-$?2{OZE};o5Zt;=Z9;`5Tn7jd?3Ym0hnp zGq?;toqeXlTF)_m8b^UwZ=^l=ne$I?F4i_UxtlywR!Sf6+@2pM9t#mx-|sh@C>kq5 z4in~zBCXmUtGnaO+{h9pxirYj+<>NdB6|vC$}S;!nqhkzjAvQ?eZz&a&G%_LR2YRP z$_Qh-i`?OJHgEZqZ8!PkSj_9qSKb~IXM|C$GK$z#-&s_m;4t>6Dd4NJRR=YylN0~8 zb1+x$H_k)=8xll*&Dh?DW!W9x%tq2|cSjVYev22iK6XVG`)er2{??7CSq@DG`lH@8 zRp}P;LN?0G@bZPx;snz(#i*iqGPujG_utib!}q;+GO9O3aks(yFT*V_!)Ub-W9M~o z17F0}l95B!jB(KjG}^&y4yfbdZtM=!h~dHDJ6u}eVt*?I3|)bzX~Qp;U(?}~5emx0 zWqV}rZPX6|gW+Wkl%oyNXwVN)!%x&vo31`(P?Niw7qDK1St5i|ogf6w&)~!mO3uq8 z2c0|;%H`0Pn7B$eP^^U7%`r&^a@HXE<{6Kw-Y=%-!Nr;CFLVG-K^qf^PGJ;JsdS>T z%R@kws2|D0p$X;PT5up~P}~?=Qzo3vhN}%px_J(ev{|#IY^n%st%#G2K7DOTpS#yG z*{k0*32Jxr9FfdbcnDZ09!bkjRM^=*h&~#a`*J&6dzqumP!Uf23wU|6=WeIm+QBt99)Xq%o9lm>&gITRv zl@qAa8BuocP;oGxSvFRN$o+NY_wti+9hvB@nczX>Qg(CT6fd;Cg8E?lfx{{AJab)!dy>2`z`|F z;=O`w5n-;&H$Nj?)oYH)YXUZU1fl8w1#l;;JPfGkmAl@~aeZ;|^Wx0ips7)dtktSW z;78=I>}G~JUWfX1O_;r3B+&oWKG@-IV)d6{@e;)|AI_3>gQMo& zy$!n5Y>Vf?ljdYsk4N-01DAV^r2+|Si)cp=5i8k^Pex{c$tnvrhBVyC6MNm&M>GoD zbB*)Yg`Qe}!G9fa4aq32zT1`g3csbKRYd+S%tKv8tDq>?xJX9$_}vEYO95Inavxlk zYp+HnVGliKl&!)yX&ygpnQe(aX-(Wspo%jOMkb37E)n}_O_Y(_vjVUXTGiV2`_K`B zeLZHy9jQE(9WBj{>5n~v*T^f`KR!zI3X1ARxSk}a5p?)2Yb@~#Ip0F}sWHQm>U{in zwRd}$4!9V9O&Kh;x9P9_^Z6xyc_w9Nz5-%KK8qMaRh=e z%pHfQ1cyc)0aw~V2v9G)rQCdP@mu`LGMstZ=*db zDG|ctH~134h{jHs3rp#@E?9hOB{j*EygK@BjK-+Ce+ zEq2jpTH9HBOUoG#EG@?~-ef$A~8gbimbNKsmycwf)i z^7XkymmBB01nTYkjst@iZL1`#dpg zrtJBhP;@z#oKLtYW(h$W-_F9dQ)}8v(W`I|G8Y{c!^qS8>rFoeY4&_M4qpf(bQ*&| z9cqY@Erq{FW_K3H2<<~qFN}HEmyy?{X?iv(u>n3`1^m0I5XPxK4)>zsVUANNd^_57 z3=;Lr`TPuaY}j1?rG>Q4?r!^Q1a@^oSb*rqyp)~@>Y}ENbcXG&;-bIdu6C$cxE&Fn zXj;8wWjpxaS)qM?4M&=~u~|mrX*NCOA1lBY9?ATx=SOm0Z4u4!5>TVsM<{c%@<*iU z$pd&+=5>}(GtaSqrE!t)DIQsJQw2f~?$_gRYuvsCD<#)BAGUZJ4ZeIR{Z5z+j?$Qe;i zJTe#zzBCkq|8mZaxcN#dPvX?@0qw`0Xgdi#3IRF+i<}5@L?L+3Pzds6k(+rlOXgL9 zN7W-_=^}OX=>%ySW;sDZ?1%$5*zn81ONOWrkFkM$y@c5a881sM02QW*kS&rC zLXHT~{irTT2p93o1>zuEg!TV8-(U~^3bzD>Te?6*B>5wLHlFW6JEVF|YJKmI#1U$5 z_Cc$dxgorSwxkgIt#c*i#91Wbg9G^#g6#sFbUU&o6SWdSMO1bS)`AR6}P-L>e!HlqRmjq>?r1- z2&zpNW~MtAuMFo~$}IC{5w705D8>d6ngFZy>yQ`|vlanCI0Q4e_}&m)HjqM5Fo-xm zBn8~{V}&RNpA)bDo7ZH;HYcu}B}qK8*#+0< zc0AumCnpEf&VdXfE31iLS@##B^@Qq!zJkK3=Z+To4GT#(h@#)pwR!*inVZfy?QaCK z6gmQ)JViQkxA22ssgnRJF}wTp5AC=7(F?`Ji!+Lg?c?{Wvx-&%UTnr13&M0?!$oH(GwhXlE-ezYeje6 ztJQ+bY|N)sDhHl>S?NgN1m{uvO|tm4Z97G;{#vj-8GaysJUuE1 zl3^yK4Zd}`3%pE2mW}2U#Y6?se6OvgZlH_+(Hw(gU!!RoM&qR-?8wo07CKrMNlJA1 zIua)1bhZU{nimWH)ji)$xnCOnc>Uf#LlYSkB$gY*k>NOMuAp+fw~&mdQ*1j!!yxJ` ztD+S1MLpwPIQIG1T;J)*&W^;O7DF`@NeTP1xz6_YM5}2lp2-<#D4;M8xyID(ZWCLlfQvOkdz#;&uiYw@?bo0#3t2S3&&EH8i$C>j>aVx# znR@ISKm9ztfIaoA119u-u%}+M3-;8*CMbBB*EBy0^*IX+RN&h%BTl}Of=p@)DsSFJ zSyD!cC8L~uG@%rKaaP;Pw;3X~`#o7PON_jd8)hdfUY_kbrbOjA(%=A_($x`SBb#f) zx->X2jx|GZw|k~U>>w7+ml}taaZ{TXXKnx$W^Zb@i&-$7G53iN_124h7os^ro2K+w z&8-&TM=`3AA)g_Ggvw|sU7{;DF!#BbXhpk5Sh0+&CZiAQ2u2KrRI6Lu*Gyl|x;YSz zW~n(&_n)}<>%4i?xje}CHX8L3WIsQBJUW8H*v^%0jIGh#B`b;dH>w(#zSum28fTZD zBDd@{f#(4V_&0S#jFiUm#h$(zJQ8vBL02cyri4kv#_F!F`a~!$2pMr`pm@+k1~EEF z0T63c)*^ZVX2E5tJI9fcLPah5lyl=%usnuBGj%(4`=jO3yqU5gm~cLLN6x4=jif<} zN0vu-!zz*jIt6cJHj|5Q-_4;p1;2&1rJOK|5AB!6unS3W(I|UyJ5AXonx^*Q@+n(N_nL48 zBTvU0&G+^xV>#`vv+o6;w1u1n zM8%?Sg3AtGz1Bd@na})E+%B{us)H%FBP!2w3NhP^GwYXKPb}a#h3Jmut}f(`B|2J& z#@B7a+c+KmvZ)Y_1M!0s+K!ACG#{2xg|^%67o&-?J)2dC>b1OBu?O9H7M+X4isg5tK%ytV~L$G8g!t$NzXmxkHA-aYBq@W1EZ z|8sp61bq-U)vz9Ec{k&l-np5dHgv{~QaMOImaa(^Ra-K?7dQh1Ddq@)M1qQDSj{z@ zcSL#KkxRAql=nXh)S_HWzstHrxnk0 z&7^VYf6hHHXNUvK{l|0v7#I~^n!4%mlW5369|pWH6QEW0 zUMhBjQ`jL9(L$C=$0HcQZGN{LLaXU~k+zOA!ZuRLHOkiX04ocG9R@qNVvG9skXvb5 zX%^XP9^fHLbfL7O%u-T8JrO1B673RYTodQ@Py>Q-1js7HB@R|m0>Jm+`*6zpS`^%C zII076J)Z!G1KEkRv#fJTVUzhBKzkn|IY0jK>akXr=w5nXX1|!)`&DJ&%BVdZ=YZpR z>U`=0VP||^l7?-1kYg`9PSno$;dp}Y*X)9a=DXn!3)fCJh1+Oi0ZU9-aGR4`AMlJZ zj|u%V`==eVe%G9jJ~Wwc_}x%%8&(C)uc`#io1I1@_{xoSPslSht94pm0?;adn9XFC{C77Y=7O}$@bdHOyNHmwJ5uI#|Ha*wAY$z|1G$`%g$uq+yTs|W z%3$^o2B$Mmo%H!FEcVlh)KVzB_V2MFH_Qng4A3+LRnj56ets-= zb5#mMc-474CiZlQ1BKx_kjG%ZWLPX*pRxwEQbt#u-$qXF!fW~@%%FM@jgtP>R#{&( zow?f8s)gc}PRZ?dQj+U(_TE!3r|XJq?dJHy4>3P_;4cqr(3T0$D}OmsayNJlJGaD; zv_^ZyfG-kh@2!L=0%Rg(BWK>Gfj7+X)MVm4Nz>-Vu^H`il_7INK&F*xn%4R=5%WxN z|3{PyJ>$u1&v@iB&hQ`Z1H(1brdFTvFc=QuAy->}@H1e|-B?j^-l;V1Rt7P>6Iw;g z)mqO*W`f2eXk&%V&86-t@)lgKz{xV@R1o5mrEOS>Lq$~wCRnwvRG$~$+cYlkd# zci;%SWb!=KYKI!=%vcI$R7CPPyozSl>M8qlXbFE&z+TvPJ{m6wTi2!86a4ExpylkJ z*r>kuqB#xWn2PAz`QcS(NwvG4!{OBCsrF51c~v5WLv$(v4;7Zysub$Z^t*%>s-Eu_ z`+8FD>WCYoMiycKUNpoa{Kn{eU@v=eW4&_OseT0Gp_oDpFB&Eoeo%}0jW3BKySKP& z_e)}`X7hP}Y6p=hWvBlnxm|}JTrEg|&$2{%SVMy49iOnceqc6vHo1IUyQOlk#U!lN zR6>2<9319G^e_8Qq)ye7*W0?^D+X2FnBF8(pkqj40FoG;)%u(6Yt5Em!k}XeAcu4_ z69ZUuUvp(Mxu#&lCJcIR#vl4OdR)sP?Dxw=kMZg2Fv)hyiLD<|&>V7TL;Z zo)F0({)-ZYw+538y00BG!=hlo|4$Ty8%d7)E9?7zVL0YasXT$TV-b$~rCYh^W2@{m z607{qBhEq7@pvzi;km>CpYsJ|dvRv%g6b!!#Z*+*B- zV;%MdU=3H6g;>{)kpZKSX71WCoYDEi{~bpz?6_tOQ8Vv=djmh?l-^&?)#PGJaV<^zgFjg$)nB0%O6Tel8I7r<9_BT1sMc_>2 z-)hUkbFLz_@%>1i>>fR$3dyj2%ix{_uS3(e*&@pea@=( z=OS9I9~vAp<}Y4+=*E95lA!_tl>k6M!1>RyN2(y-yP$f8&Vp>kq|IX!h5nk~oD4vK z9Mnz}Q)U=Be0e7>f?dSaN@|n+@vUGgQSWBG8jGM$Z`{2crO9LG-})O1yp$cwpMt*M zOGaNc2)U))bHvHpUY`inZ*4ej#QGLWKT3t&FQVYANW8sh%f9e~FmiH0gk3#Hmfmo26jGq{vM_n~4~T|WNEmma&%}T+ zj4&4dkBIq?uzw~Z4Pe9)_kV;Vj6lS=(|RTpw7ua}i}c{xu~6c+vr%ggFbIJwS9tc| z2(0vdIKozWcy@J^xOEd|soG7>gg?hSHVGM~e|Wwb=8u=S-gM`UzIvp$I?H4B2@DSl z)}1d*tw3FcBd;jlfVZs+gMFFC%FaKKCGwB27aeBE@{T&%J~u4}`fk~U6!wDUJ|Tum zfj{Wv;aU-?Q4a}*ey7Bm*#iUbL|inIucW3zP>X|L`9bvUrwP0`T_F$YwqQ;En7aYc-Rb{U3Xo~Y;IzDWdwJ6p!z&1Fcc1NikwW#Sv9f zb?0+OvCu7PMG9L@OPgyzg(FLr8&>sN1&KJr%m80A3Ev9|<6uEb zUwnMN*+{`UGFOYYvdf6Bt609eTy1r%Om7ifN%e%g$+Djf^O9)p z06m*9pC+w`_idyaE8OIK3M9XqSKYq!1k?E0^3u*0voz3ti#*B45Zl8Wt_8CIu#Ej# ze4g|giP+3Lbo5Cr3792&Q#!;OafCLxnpI|{`8M(g%p%cYQqQlGWfBXs&HgTow43#> zM)5QJUyULp>;I}znA;rU?&X|d<`=n&wA=nhC_<(IVR4^6wlSE3MYJCs8$x!KpEP_hW<6t*+um-`%c zI|L@U(7F7XXsl60(I6`v0sY-5tNtwDf7#w>>GKliHTANH!uD{==vm zmP$A3N?5#p0C{*V9W3*%?0Vf@gqaDcy&M)oI^H;i1Y-OdfuQVPKi4of}O zfO6Vwj-{|$l=RyL%cXSK1{qe?x)14_JG?2_5>_^iG#P~tLW?uCN`WQLlXP2Hv=bef zZ^0*ILke3#2S+k~fF#+~zD9fI107;u{Bd+-P$28H4?azwlF9EC0ijcKFK*J6aln?7 z?k?`?d;fm-{O4^hD=V_0YhOMRxx7PFo}UC^!RY_fDi@628*$_a=rB>aH5D zg$FMivgeBnewGcVvX_R%h~+THsH#DZ4wTHlOufWw-BU3^-i=*ypCb4sV<64M``X)@ ziEnqLx%&ISx{7Cf+P4MP2z{r1gwKTp4d|^eeyOmv@`vg#ue3O@f96wEjoAPe^`~M3 z0T>;sZ%$)0bDgjBu?F*gg*wFzZchHsS` z;Vt!pgrL)Gy-g}J38Rrm08mSVAy|b1N(+}%$!A!fGH*3$O39C_z=Fg&5I_oDN&DVD z?A_8Wh=p4I4@#++Ol${^5{Tti!a4cNs01}qDf34PpaTnCzkmh=y=+=wLGn3`4#UI1ICCL{{YVJQP?WAWwl>n~2D6Bpr-Y zYlj37AXFw}J;VGk7>S)!<{3sMKp34sgXI&&j81U1a^$dbY9bG>iD@2L))1$y%~orQ zFLjn!zSZ2HmF13j=-p!&7o%jB;a#E9jz4J1FtpszxJE?{bnUHpsk3o2Djx$tYJzH_a^c?6>Mf zK7CM)4-58ff)V#{$JA6&@(sI9B9)KT!oc4tR~c?*zSB+`Rc_mJnuCimgD8xqfO)?_08X`n2Y5jSiEBc7O3-l9XK1;0`V3-Lff`rzI+~e{jQ8X1V=00>t;?JfB5{IL3PdHRO95A~62gMX@+)!j)*6a|{3~St<&{T* zhncp%H^n|YV!EDv67eoBTML;~cal#O<%hc#hv1P{9Gb_h1uUBcBJW8wC(?&^YocmM zHOdumw#Q5-C>&=F^gq8D)qdUfdoXT?$53M1>{NE^Da#g$W;8E)ctPyY8JodK3*rKp zxO*eb%#g=w^9;nfEs3m#*{k#*o@m ztQ-xyYVBO!ehR^#URqe9hDD@$y^p$=`sy_AW%xT@rXiP+jz{PoKz-fDwz{1#BOJ7_ z7DZ-_`Jv(w4gr;A6xj|CMNRibb$;OR_TzRzjiht;*qN(xE*EBGH`fRrQ`VBoNEzlg zH*PL#U1v^Yi7$q`nXv7JBFl>Rh6e#?b<~7s41F?Hu&o1(Ivt9B3qh6W{(3_~SHooH zS=d}Qx3@tf3|0StA&lo0!m+8;SR7|eBTcmO7wJfwW_AbKA>O=rbCs<^lnv1B!W-Kq zI>=|pI3OAa%rH#YCG1;f2zmC#k++xaM>L_f7ZUv{Hrnggh@Fb_Y4@^=btuh9EHQEL zLdo>O&$`t$Vb!xExK-EVp$*of@yicj5cun7#tIr=5TS7y4a+=gO*RD{hR~IeQL_D zQ)Lj_SmZTU`EdJo;Kdh-<8_j649`hX`u={FOh>qFn<*T58t3dT{ObHE{3`KxFt0SJ zlI)n?>+kdG_@$>B+3(6eDU>QxDC?FyP)RO5vUm<0dfrLGN_*94z|SsH!;&gTqk?VN z700e!OEnuQ`(=U`;4!8p7(&7&yQHT%21~axL9pb>EOG!1wi!xV-o<5UpaHvS_!vFl zc@dv2u#kOse4Y@@VM<@1pFrWGz0jZ@f$(}rW9T+>hUbTr`He<4|7w>1!H&P_ecvNYNCf* z`e>q}tWAm2kh3zwxQ~T=sDp_`U~Hg=CQ?SIR2xPnE7*e*jg@G~!)0kA+;%^Jv3rU# zbf|M-TlQ{X>;*^?%u;r%msogzp&O_d=CE6?H4dGfQU_p9o~+=@*R%8EkP2yjKHZPoB(qa0 z&nVk7n)i%)>HI?#pHZr3)cF~0(Ee!C>R?w4`!q*w0~aI{E+IIL!SeDY2IWz;+G;NyHz-QIr)ab2!lVs42dE%vM&kVNB7uX=r|> z)c86R`GGm5ANyFC?`|DDSCX-fIpxZ>a1yR^y+atKaO>IA zfgP1Fbalr!L!1f34A6x)1xZE_F~iFjD*3EkZbA%o{QKczH2Xfb??1&D3UL&}KOcxe zXXheXkSFHok%ySKMC)J@t!zik;hJaeq>U{nTV6?dcQKTC^zIp^~VPbAeXkN8mH z{U_jC9;a7rhS_rs)58f?!-VZ^k{+ca&l=yfX{S0d6ow>aD>dz3YNu&(2qC|8195aA z`EXn`xOH=6t-?2V*=vkIe!XxD_}0xW)PvOm5^QNLs_Dmq?uC(wwbI0hHb^vvgpp%0 z2?LoWeybv*i+qE*hN0PNnW*<#JgQgkUp;QoA4g~mmwIi_${6^`aKacbnrxN zfJ};3%#U5V*uGQVSUGMAa#8GEIEunBQ3|{lUbP75yQ_KFuwo7AbkU(w6f}@_ruhlC z{o#U}bb05`mo|D1h38|Wvc#}sq^CUUzs@N9_cnTO5o!}Amr?hes5cMC-!=(RNvKC< z*dX0AOYSKI;RpwxvEB2ezUKna&v>hk&->LT%BHRO9_|C(JDwOgv+kL8H1)9W1Tr{T zK*XrN1fOkkcFW5%G+99GAlr-p77!h1Afxd4S5rnG3y21`uer?vVg)o0%fH*KF@0TA z(Z5nN|6mJi2H*ye(eU~c<3~mfQ}xu6f^jvz}s{s~G|6o0FX9aRQ&gN*o#*@8^o1feW*V=WFS{pySID%$R9`H~B{v?6# zkSyM2J!z4AK3}Y4JAW5RVCJMPhVk^yyp1`}_UYEIJ9C^OvV< z@}$GE@tq^hLmA5>#KTrBonM6230GKEM-CtQa!02vUgQA7n5mNjHXIULa{{2Q#5EE~3#gDP~Gu5(rAw=0O_su3o`aN`e9 z&2wI5-E@N4zDOC{F4%5$G&nGq;L0_BkTX(79xi+}{%^7tFJ$~ETId^JnWPrrLm39~MJRF(&r$t>+ z0}z>+>@Q@ec}ffz1l>j_7J$kiI|XB*k<^;sahf4G&HJ*ynTP1SfSF2-!B~qqFpu5| z2T<>TMsYBfGDP{^5xv6;aX=7vDSi|2rinqkxii{$QDe7YITBbN))C4BFf*o)MSWdt zhSN%}q)ik(l8Z+d{m<{Wc3&_Od*(k4c8=n}mMk|ZPp8~|k{u*D@_!%Rwq%Bs2=XlV zDnC{Z@#YV&nBtVD+{~We(5;MXYsCLKzCprTr1q2~dt+*fNjPa$>&Mtb>>Ft?6@fLU zp;?^&GC^|UZO;4IVz^KxV30w_xlGuIrK(ro-Vo(a?P!+5AH!jSpWPy*x`t84BemE+ z&BM5@f7Y?GS8Y5T31<-=Fi(7%;B9`rYE>AO*CU0qYo1oaHDyU)R5SddzV>gU zVQr7LbJKYL`C5#>tCUy2(0jz8T@n6mtxPX_D90JngC~^5S$555*nueozxiWU%nL(z zqFs=3ut?FA8}KWF^#`cOQimj^6gyTra5+?#X{@dpWDgqt=-8Rj}K&-OP@Ff=6Jhf zs4W}jDJIq4y^Ly(W7M9AAKo?5Px^y$-6uV4tDGxY)@*7m9|t@R8D#C6evz9f{@Z*u zF1bQF`*1SYx%_l*X{8`y6z?T5T&1umc4pB<$7^wB%R`ML+c})R7qK|9A|Eo63*@^bi_S$J&8>C**0v z8U7{)Ub*K^i|SU%hDGegIC!QF)A+CStYQY&r5U@T{BTbK_?SQ4QjHDZ$U{C;-7 z(SHal2&()~K@fHZc1;CXKkqkXUq3JJwC#MnKdq~*D6<*06(UY74-+N%oWfo+)4la@ z-h)BnmiYzg-&#QXC%Uluag-~5Lbd9=& z*RsOa{ibdsg5_W6KO=p-@HXN3^7&p;v}c4I&F)g{Gq&B}zk<=h@1I&muLwvR&NM2k zj4_P%Fea?5mh2c5;f=&KD-mb~?2tYe5-is8zth*pM)}+}Q}H?}33U11mFJ5~!~R*w zDt3cNX&J0^glZ|&ndH6r@?w((j@%RH2c?vFyH1QqI$OeI1_2semNRnf-z zA;>{`2%e zIBCldF&BcT6O%JT#SFovq@+#f0Z{_5KcSV;U2PcghSND+;C;Lka@vK3t6|Fj07FYM?lIVd7Vc7cf-H$EZSWF(=EHFqgdwWEzS28u}r z-tW4n?aX|LIXH<=sGSDA78Vqu3Yf3|I)%vC70#H<$fXvOTQx6BtArRsP79<^;<1!^ zD3#*kP0tO}q~NDJ!f@&oS^Dpav=LUAF?=$#w|?C89zCHLG1R#99>H>?Z(fP3^l6kQ zE78#x*W}31GO=eRqvA=^Mk&!N&_;x##_&ngqTkWLe2P*2`9#a%(d#JL|L0JG=0+6v z_spZC|(s*19;0$tH0*>o#|{0(Qp zp4q8d1pgU3N}fRAnf|m`V6Zm&xcf}=o!{BPTY}Hqb-U%xrj04PR>6auZ`XNFos0)a ze=qg9{OygK9^=RsF5HynwL>1SP1ASgx3rb88)&xq{sT;Ad(kh-W8{*rtSeQIV27Pi z#S5YZCd0Oat5K)mp3f?2HXx;liaP}`7x*zRkH!m|t~8-bPSpR5%;OUMU>N3shDV?6XjNa(0l zRv5+NTlQSg?Xi+HQ?(_oC_S%t9ksrn6BWT+(fi@s8NpF6fb&zd*K zs$UUfgqo&>;XtGIy@XIyQRc<+tW45oDd8aXUYowWj`|Ooz;GiijLQ_tMqqv%*WvUt zVKE(3V~f7xPUA@8ei0!a-H%WsrTB}FAHak~gvSm=>>fICmPTGCl}#TR9g65=UVMy5 zR7x^B@I!5UYk+Z+oMuFRS9CsuB1h@tC?=2ICy_88)=uQTBZ70sLO7C(O)m5p6ze5? z@bmu1Ef88Ad07B!rapIPK+EG|7RDc-_D^TWK*Z$}eFXX+Nm%PVirsX7Xb!{u`D8W+ z1U91nc~UmQ`3XF?CGmlcTi0@;s;BPir`vBBKcCX{@T;Dl+qcoksLM_;ckN@;w?{U1 zV#6oR?^Lyq^|1EG1(9+N5JfS2z=1x@&B2F%d93^;1#Euz(mN;f)w>a{Z%w&>lnI{v z!Zp!}IX3E4`!?&P5yxYY=&PKM2E6vs9@o~+*Tu6Z?eRhHI}n$Rz0sF8##Va`9UmaxIN(J*I#yQZ7=@z z_E+sSjzO?PTb-D}7}HiWe>p$>8QyCORN(~K4=NftGeJpl3=Ch0vI)q@gLdU|u1jPVn`;(@K+h|q@yeGZ=c5Z|5<>Y@AC4#311K>&|(-E$D zZjH;{++@Yci#)xuQ7dWWMl$;Sy?Af2_9*a5=qT-1r12my>G+m6!Pdr7g`n2)0zXsmh9e9J@>~L@vpxt+~@AR!=Af4 zg8hHgtpm^yAVy(tl_=$GXN7%Yahojcd^?*P% ziK1NtVxdMV{WxdE15rBUI!u>yrU;P#q3|QJOlfrrk}ac~pt<^+&pN~1Zlu0!VngGn zYs$b&g^K=Qk#^HYF=X%`utmyuY4&J&6_mPshIA+6d1>~izI~`JUMp)11H!%vyC){~ zPntg3;V{^tWO^J6aaH!#A7O^eF|+SaX=W>v4Q-?-V%}3BM0>;4Dcw;M_Jta9_xU9n zMy%sl+%W0?N>YBTpijsA$W}MT9c&l{j*|mhNVDTfM_;mM1Pg;?K$980!XQSFYsS7X zY`uD(fhhu8y*g%aiGZjA(6scTi%DK?OHP{0Hwgy`a^!EnDdESA@WTK z>F-R0%M;l-Ztm7j?(<+ay0fqHgjH-eC0V`^L!1(%E5iVgDKlq+TTdm67!$)s0o)d_IsUlueUxStc{-fNzdkIUXaG;2DxWZ`r^@QrpVBHP`hMyK!WsrhfN z`z<>uyoeJ$l7$q=3IDONR##^Z(XT}l=w=jg{_Ys9UPX?a zbKRtz-I#=hABYu?S&XkbRtQ8{rOL(wd_G-mb(9R3QIeS!C);uE=|sme^cwpZw|#czbr zVQFW~e^Ms3b(^j}Y}eoz*G#X8UwNBDT-x)CURIDivoFMdlg&H;B`E~8U^K32w6szh zuDX44vvo4YrFv+H>ZsXI~Uj2fdh5oNm|E%&tl#9Y| zF#?nEsnv`~M1w2rWmXdwM452VkfK!B*=W!Uy92ZQsZir@g%KwWg*8^wpO9_|8iEO# z^<;jAE*1Vmff_CtT#N^UhZJBiMdUx2^G@p%k_rwCwG)>4DO2=Q3ZvUz+B~jE7h{tM z78z_9UWFuDB!#hNFKtR)be@Kq9er!RcDt_sQ!O_b|09jQhM_2vibymj&GoWrX*}v5 z{FgI-vD|<2g+t)!gAtRZ5Z7?gppS3>UI1N@0R!H}dz>AYzatXI+EOqU4$qECjJCU) zQYzGy-m5|`v}1LxZ6h)qdksMe4vjiwvzrd@U>-z|7=aj=7v?}eCHfq44@`=Oo`9f) ziiU=Mj=ZEq?fITw?|MqNeG9&SDG^5&uv~xo+@*JLyMwb+X2=#CFyRAtw^#* zigEDG-~(l?3J4e?WXYJXBV@8|7<8D&^v4s)t)g>l%_858fnF3MHAwBQ8el;8g9*hP zrOTqBMxTi;F0t$4Q}BriOH?l4NT?C4sQph=kwpW7G4qF$t@0eY3u^+CX*|wCNp$v6 zU>9`?{vpit#XJ(C5peQ=EZe9*l_Ax#jYm6UJcpY8KZNVXT!THJ(6W6vPbFJQe31hE}MnLzB1jb$fe3r_iaG1d$E^3R9 zq{z^AILv_F!ZbZZgusQ+_yjUc5wrz(E|`C?Ns$Nfd=Wg?304@ihL)Jw3G&NDR3f{s zRQEYLF_>8oGfXB64zsPdP!|gmqJYLDkjQ1`iEQ>9U0&vrt5z4$42+3+@e*~!Tq;IO zmN*k2&|#djWA)s_q*_G?MLZ4h)_>=PW}N93%F&%)IfOri4~CbKLe)d}28U9xQ6sMS ziCvmVl>>uyKs68QiVBDNNYRSHQ2rBjc26I-kM&L88wjFr zX~^a?_fhI*H47aI2OcL$G9xJr zFK|t90)7d7#4ZSV(u{c#%g=Mf0gCN@)I@jTt40Xh7eKHnj2DU}IS_&ec;9w<70p(> z!9u@_-nzuVHsES7y1*R*2~(rs#gq(nSJNkl`3UPhLBo19Xx#%kX<|Onn;|#_V};`> znujf^bRi#8>EwQR2-@<{%2ey)(MT+KxhbnE41!(H0P1`pcq;S<#DJ5@sFu8!<0uIv z^6VR%BPhHLKOnGkUR@|aXsqQ)w+riPFYdLLpWsmMbs^X1Xv;ZK8IH*G9(W1zo~Q!T zr$}Njf_6l-^h88zVlfV$7+ulD0}?MczgV7IfFuoF9vlMfE6`Haiw21%!yzoMaPPh> z0!1jx9o#;`ktwW##DY=ms4Fgl^Tby{$CSGUYoLe^vIl;R76rif|BtJ$jEb{|x?GFY*}-Jw9CSSb|O;;kR*?H!f zH`i2$ZLPLReO*#$VCV z#h+l)1sVciyr*!t65*7!F^D4TONjP_bJ%ul7@7lzD4LrHBCZ)I)z*2*yG{DL#fKAf zC#P_fz3HW5;NV|+x_E*sFK5**YB04t z=bj^>XK0AzOL=H`ZFlV|2V8{ZWvnr_&GyO6a=fFa%V^w={G`NgQcSd(ZoGN@$O2 z(UMzOC5KK=Tr8n;dWfbe905Uf^(v0j9 zeU0{23vtqn-ZbH{?16wHae}DOcbnQpwTGzE8wLdS1I0|2r34jAktFt!u?QLOuULMSZ7!N4`hUddZPq3c$R>H#v4ME@hkzuv;!fA(O;|LUguKHi`r zy$M9{Cla?T-bEs#u=GbT4$LXyn3Kohgx?`Z-mK-f?uOuyHVs1)-B-$5C}%I&8=Hwe=+RKo_1qf|9xw`3jzQN(g@2cgtUG7($Nt z@=nBqH)Md`?jbk<=!L`ssi+bg;6`F@5obxxEd`H%w>=cppHc%_E5(6~QQX&CAsw-j zwh;OX`^v428!NN0h0$pdE3Au8|)bcKvDP4=`dw zxHyp~`uF2B>;yR@z|Rz{B4kp#KMAYJ^q?QiUd-j<4rUm_SlzEyedaF%2_yO`Ue~8> z$!1^DzS;s=wq*~J@^af!XNlvd(9{$S@D!Dx?r$Qmov1V@E!VzL;tPY1BlkA?I;ix~n;Koby{ zWT@NGn}we6= zP{}c5;FdZEqoQz0beM}!Wf2SMcA?Q>DY5F8$tdiOBVr76%6YjpT2!Jpw!p!DayoUv zX+%`-12{YcDTI4n9D!f}%x9F7KTvn8Qiw~M!8E4uo-_hXh{B;lu*_jH_P^3};c{t& z@Rw36SZU$7P%xwx3NGOE9*+jxy||qy@*e3>V_0z)w1C5F{MHT8OA0d)gnuIc8|C6L zeXGW%r8D40h6~XKE@_AsCN)=4ca#_aPel zg1353go)9zu}QcGn%MoPsbJOubPFi{VUM%ac(|=`!`JB=_RAkM+*YbBoKnhfLfNsv z<3B#!z7+)5peoe?5(j(m4tM)(@bH?TVY@5em6pxMw8pB%mSdrlfw18lj8iTmg4XRI zgcIsQeOXBm+7ZzEWd*bh>m%^gQ2QH#P(o`E;3WU8HZvd)Np|Vj1NjaeRtpE=Teuvx zz7vup@IVVWkRp6jjl_-#=0giIz(5tCH&qDE!7rm*HAE{m5z{v=^O}gdk+#@4V%XfAcsp83NddJZ0k37)HWP=5ikov2F*hvCh<#3 ze4}t7yZ0jB^5CRz31*L;sGXZv)+=P-L+EP(X`o1fX!@f!2ui-4wpQQ+O#C}dRFc;N zOERCf@mvv9yHFRSlU;kTnAagLswt~<^`xqMhb$HY_fq|$e;Z#S0972roCthxFL=g) zH4mY`*MA@#Kw7u9F79Ylu78iMO~SY#O+68e-yxWI8R#at^@lKeDYtNRZ_*JqLBlgf z@vC;Dwr;=J!5S@VHhekb>0Wg4MotXoPTv$}HN;dOp8jAOnNXvkVI_tM5-VE70qA^x ziYDXHZsTixZN(r_D;wk~9DzfG$v&tVO!l=l!Xb8WjmKAx2Y1v;BKOB3azi16zHr#< z@T4CPM9g;ae7l-Z?jB*%t|k^5;&vh4sYVilata5g1cjE5BYK|%qqMD&)nBfv{sfKB zy~#*)Yh$(2Z*nhkhK;UZXyG3^CSG)C*S#FF0TRvl_gv|vvo{mRu_ExW7DQm71WY~d z!C~M06`%dmgNW7?H@JLe=bo=Je3feY)Te90}LWQ zwH}G;y#R;*z(eh*GXMw|X7c=+#)@B0l!&M-070eN?agl8J2$bRd|@MhamRG-gJwOb zZ?yHZOXhV$oCskrC>VtR%WFNvSfC5>m&o4V&#;NDt#HX>GXw+7V2${qqJ!(Zv9XWp z;VFIBzBX&+Vqbb;a*M=g^ZG=Q!p4a>P#QB2VkU8v~CIAO)_K5#CE0KqnSt zh-(zi+Z8Y@KkVqWY!}cILy5L3AD^CHij3a3W5gmFx61GH#uDH66|+R>3lT(ocwGbN zJsON7f?jA`igP)G;e0~VU{CVz`KB=_!_BTSfZbnl{B_VIOn`2o4m7;4em2xcJD6!{C#$ldSy!d+d9~A0>)3d+4MV zO`BY*gCycT)bcmGj&GcLkhL)^5CsVpH^Z^$PICl?`=DvY`F5W^MVxfwj0}E9ijV0o zF@r-YfFrwhKEHKb_Ab$gL4^z;n%20 zbH3f>q6rk#C`MVEEx+X)j7$=M#|0rsizln60*?sbac_tvI4Z>fcPtQn2=j>pR%M``EWQci z)>hzgx4?uh4Ej=d8UblQF~=6VA5F0fcLwceX^6OAEnd&>3JpyfPu#E*JnGl&qXkx* z4*uy-Z}P8@2C0;ns3r)cDUd~VWCt(iyH-`V6M}t56LDcZ_5Hp&?llcEEaK`y)aac! zHIu_y@8+S@#P%E~7AaA6m1m|fLxe@K;1p~*%P?cfF61?S#nUe2Kpn`EF(JM?5SA}t zhUnKrvcgKKG?|WAdBO4$`Wj+yVr_f(9C))r zkZQ$MsrB`9ap-YnuxW1SCHLu}*gV=_TceE?#*0#6{sazBUEY^EWJN|i6l^A6P!m3` zfPe;sP0V|gJ=7t_!s4MXrEaj&17ujK9qCIc_kX3P>Iw+MNH18z9%_j~&0s(wRNOqf zAQ_;I3L_q;fk?q}+k*k!P;pbAf{U=Q@tClyA)N%Yzqfoor@dPn1|pa1x+brhv~CNz zpIpHRs(Y1$e+81?v__O5i3F@I-&UnxDPGo{K-SieX&1SZ;(daDqU5t6G2E=_x+K=%_v$P+2nF`YZ6WR9_w+a)mgUt9OGbX%;axpirveeOK9LuB?z(^^Q zSSN(zc-}}J1Vm}FbE3yP9Tiw!2_)L0Sp|^GYOzF^mJ8R9v%avq^Y2jPUmjSQS{OAss{vs@9NY-nv&WOerr++w?Fd zX7sGnFVRO;b2aDfsDa>U#hf8vv%LjZxdX36S162%UePs+iGabbr<38%tyM^ciA`=( zS$0qd-$wn)F0>k;QhoTY@VC3Vz0f#BP37U= z*@8JL*n|u^7!<4vQccmOAjuQZm@!w;wW4_W3r0{M2YW$~(S(oS2+Gie{~{w|%VrTUCS(Hq)LE%+rAmfmj6IGw@9Da36JND{u(zK4X#4lF-DR&-INX_RkX|F)At5$Gu>$v zj-I9VCT93m)L}SD7W*)WtQ%&Jl1}G8Dgn6oUPp9Z^Aq&bQw&`53u)8)BD$<&k|z86 z7pv#2W1g(;-Gk2~`u~oUm^PVB%h9(M4ZnkHJ)pX`(hd|v@`0V{|W`m3i;CTn-Ok`SfSn0VGsJw5i z<(9z?>0Op(nO_#%qA_x&X7X(chLqt@LGX0ajPN0Ss_JmjCjk0mn*p5Y9*~Ue&)4!? zQYz3;1x{23YBkhBI`67bg@-kz1|N)MyJh~;HIMc>yM!d9>7 z06?<(B)}T4wen?0X!4l6myYz>Zr__+g+@xnGfW{MYuhKYg(%sNmrSKIOF<=QEa0q- zEQnu9Wf@(;&KW`{oWqhmG`6`CB<;RD>62~{H}-CR6e$FR1CX>TRtD)>hKYmQIvMeN ze%3#6kr&n0M+6Tp zS)}_tsI&-$!Im(o~3MYy0nzeEw^r@gH2CZh}eR z%n#1UvkyNVJDIkF$su%P+rCc$%oR#_j_qFs@@Qu^pw<35QS z=0E051{gC5mcUtj^3OWgA~&P-cv@9EkFIFboPf6PxxcJUVuGHo_y|I1QUneOn8kz@ zTe1m!eIw**`Y6o?FxgGo5Uf=iu8nYLLtdwF;k6^_D5(6$IjSQu?G3h7nb#m!7UNOZD?5fyYh)AO17l*dw^rR>lala;%8mou zl{rc4mPmLvjcE-IOli@s9TVL=rlT-;nK4PvDiXm%3l3@db;|G3Z8Tb>sP@7kR3&#x#2JKnZ}`_hivvk?;ouOW4*SFuWKNVtN%PaB z@5mge-T#e`uF|(HwPF9phkd8N(cXxjW)tlCm^W74gx?YLpr5Dv3B9vISw&ill66ne z0;v@Tjx}Rc6(OOp6~ts7Dld|HSvk&$0^`HXG+(o!e>*W@n z8?%n~t2!zj-@mkWTrYBvQ*fvUgVhPxAb|YZ>Y$rwK2*SJipa4{R)Jy?o;b*DVo@^Y z%Aj(=ZLs>2T_OawEFZ#TCwV}NCX6OQkSCM)O%y6=il5ZH@|(wTDh&MJ*OWYvLT?3z z5aroV#EFhen&M4*wc6UIuP7HFH3Sp!Qu7ssQ-q77|7x(izX3^`64nPyS?C!y7->_F zBf#dRdg$VK(KrmSc4FxHxh%Ygv*ghG)}~47)yrJgP=HZ^{rZR(V0fdd+Dd6x$bq@< zvywIC9M$j&yc#$5`2lwR6GHv1?8n_iC;>DSL}w^WuP6znOgUAGHd$b&4wotIMw0*K zA1yJ7O%#8H_f&Tz@; zk0ucnZ#s5o4Bj!*U&>Ir9mz-kfF2YAN>NgR@dzlv-5NdY!UYhpV#yaY=0c8xsI7(_ z#ps68NW2sZbp%CqPk>ysEi8W-ECswVTY-&vuKdL{N6T@+eBq(>e?58>D3%-QK5YmS0Gk5H4AOZ7p%Gw0`koz z>c?q*smmWM+;+}r`(-ICAOIv7jB}m}hofm{C?4ekp))$1XCWx6^$=7ePhJSf7<%`Q zqJaZmpbyK!HOeK@isA%of_Mnlh({EO4Jirv<`gwJ;NdFw_K^RwF(ZtUfq&pmux_El z>UF?|A2_-G%Vbsb=MA!)@5hO=hm*|(K$w%XGgPerR+>uSJ~at|6~0y}$}=r*Ny`qp zCE)XX0j0EqdcTK4#*P6r(-;!siNixHV7vuTWJ3Yh__8Hjv%k@0YE_&{tbuMK|KLGB8tXOx&K|V zs2B|ob~71*JdQie+PVdYw@N+n84nNb8Vj5V?+%ex!l#RRS;j8d0Ak^J_rTM&fg2LP zu|gm{D`MOOF_+ThO6?jqu+Z zRHCvxuuh5u$2|*Oo7E?(NB`3XaRuIp85mDFRG|mc#`{e)#h4l1Ge7P=OlG3>l8^|_ zkwtp~ZDC5co34+CIin>sAhPES;5Mhd-gBZ%Xl-q zIE{Z+e=EL!*2yfWvc)c6MqjH!%fR<)9ZRcZ_BDou6}lCoLQDqmnL`pd!-aEz<<8u1 z_z5Th0veB1mTylp4cRS5ZQYQlEL6$xN%4g-;UE=(l2DZiyz#@B5D3CMo=T&O3j8Dp zqC)%VeKqD~(2IcT1a}9i3?D-+kTgJO2&wBb=GRP7G~yb2j%uV#X#V)pAxwixp~NzU z3=#wz(URtZlhR-f^_Z8PTXMWMrXwPN27n1Ia77MZ?5g0ZyoQNgj@#(hSl6N-5lmPB zK-r`J#F?bfeDC!e7q@ddG4G7cEaRAr36JG-m56M2K>j*@fTSv3f!a|&Am8^|VZXh# z?M8ZIm9_5hvDHH5`lLvzoACyaUmm69eHm?BwGPHVx$mLRwY-(yc+1A^AOE43{-`%p z?G*JbwHN9nBTSpBJ=~Xo+47l|lVxOo%`gU%k_5ciH*WXJtY!aDru1S0*8pHWGRl+Y zu!M=(twp}KhPS>RV?OV57R8d4^HPWP`-y+~QU#v`x#*6pkmX^gbeF4j@$Twhkx(-# zX5i9)?Ebgc+L1Gj=e-(AFJeSvKagrDY1%_WyKRNaQv_hsM^G|DWl2po>`~RIg}5kb zKcGf-EYrC?M9~g`=b=b2E1QIdyP*<6JuchE5?)LU+2?vl4=>E$jp1<@L8$)`93;FD zh`bQ=Ju1E~d#XROu>WD7UP|Y)h#||pEXVFjL7Fn zPdg7xU}NjwtYN?Pt?f>jSh;$9fBWqimK*_&JUe&f53bz^GtRSYh#{kt^EfjLGZKu4 zt&eeWzvC=QjEyp>$@{$t9Nrr`JaZOE^wSACt`HQ$zYyYx%^3BxBni8{J@>Nj@FNAaQh`~p8GHCy|zxc$xb~|!xR197=@$A~u zX7+F8kCR85yT>=fyZ$^j+>aTFiIz+?DBY>Iczz2^9M;|>XGy`I61GIjTH3z`RT@h+ zucW0$oLEgI*Lf&;5%+#s^@A_t7r=()pNZt?A?{ug2><$%?V9JZoBCH}hE{FETReVc z>b9b_%<_hPog8g#LpUJv#9_U)!(we=jHKgYr6#{r!JS}ZzjDHYM5w7jl_~j5NMNR!QN&MFznJdX_%KC8nUyf<@~N6|`Q0`tZTT3_VTa)0KI zX?wc;avI&_qTb81IQ#99S8%uujPM z(P}XH)jNy)2OHYRfOh}SF}9eES(mimmkAk2U(JZZ>DT_))M>sQ-CSd9pVjCqs@dj& zg}!MfmGJNERfkaY*3sT+%$FShBxF?#W9?{8%KavErqNc!p*Ui-m(y1N3jF@c%w9be zSsxDFOk-Z}7=uW-7J==ld-9}o04}DdizT_x0}w;<{?W`{)g0+R6xuNs5i1b_+p1;Z zP1Tg}un`-nWubom&(}nT7(Q`h7P|holl!GLx-tez@q{l)f+i{8$kIGMc0UBEUQ_ms%XEMUx0lAHPJ>*Q%wr$Pxsf zM4~Zr$s~x1zvO|86e<6~oK#`3GC`a^9*LILZ+($>Ql=SGUKr7>SJhFKg#&3UDE{Ku zdlI%rwHkb0o z;JQHcS>|7%)PaclB;B-3>*i-LEzUU71XL5~iNaeq`?v$DEx2e`fpyxo{8$zeY*FFr zWgQlEmK1&NfI8S&cs**Oz#s8ZKfHD~enf%!CD?S{$HNk^(JiQtE57KX`C(fKhUzb# zCH587p5fLtSlZS#Y}rJ?5~18+7zwsM@hid@3IH|H$z1qw0;Cg-r)Ih`Tc_oGxG6c4Fr8Z~qf!k(k`tj@b})b|%;^)=GW5+s5Xs)9_)H zqiD^gIG}~!k|8+(CQc&G{c$ofITF+3&PyBy{63eIJ6MXIoTv_!riv$+SMeO)?CdrpP`1cDW;vccLi{u zgHt9UT8?s6U}9M@H>V;hd>rPVS3#WbWi*snUdC}0V||)a-t_7ZgL+NqXY7N2*eu0% zoML(6g9R13=?m3%wxM`w(Lfn1wy<4!QJ!VY0uw?(y6G@~b?gJaH2*q>140YU^4)y@ zg~ZhvuvSJ=Keuo7^a$x$wFw~w^Z&<1J0C0%vQJ+qw{Nq>B*=&c;#Rl)U4ChrbdASK zFt1XuBDV1M^V1Z zuIYKH$!zd~-JMry;?DIp>t1gif8=7Yoo`^^lkXq40kJzrwl^L5*&kA3I|4~of1PWL z=0|9R`GsViv>8{C)L0S5Gf43eaPRMc1(Q)dMH2OPwSR5eQ0ihy;yt=vgNs(4>jKOV z5)6xDlL4^{VC=8^81R*V}9pn;Q9Rz1$L8(~?^$wq$Qh&}3As zn$#dSV3d((;$g+|WZw`_;^9|vL9daXVD&{nY@sr`@MeccYj~5K{L6%kn_kXFDtZ0h zuaonlpBPoTyM0zy25XrQD&VE3i-(o4ho*18iiZ^OIJ}#?rH68SQuv&%K7M|{jM~X+ zxUX+`>n~@NtNRQzR9Q|QYHu@%^#YcWylXY=?jfzGka=9&cy&3EJ>dmCTidp{;Ji9ca-JVU{WTIU-`_X5$+ zf?3aeq$-rE?=H~4*w|&x{J^W!GWxXqeCy(&pH|b$bG!PGXXi8b`!Qi;ApgGW+rn=x zeKy)dHM2U;qZW*#^BBEz)xYUvJO(3L>9;*1eLw7ks7&LCPr^j&7)nz0WGXI_YP}&2 z^$ZvspMJm2jReatIrce#nX4`0%Gas5(u-MnoUr=2AAkM66|)&VT(URz6~G-mv`xS2 zYJ22O46|ERcn}f#i2$vIgS5dtIGz0(eb~7;$$xj~>QyobUjwH4pVx#KZ z6K$K1NX)Dp{WCQK+;uC+#%=H?+e){)YW58Fo$EpRgItz+t(C^F-az%Nr`@@@kyoR#`y}csR z#s;1}eYrm}z%KDp5Ud_5dARBKexlx4%*A!++1TkkMkao_(=h0=*N7?Mx!)rzb|U&>UShj?XCyRI<@Z++pb*k)W2_8Io`NZZV24<^LA2q zafTt*m$Gy_zN?q?Y|Bf*eBMl({be!0q=Q6Fdq{*BK>Ca``1{ zBKs!sRyP;?<+C#Pe*X7!%K>N^{b4(}vfoB)y|T^ML<2;;8So3QhK=5ZH+aKQiJ|6< z{c-t21lN+Eh0ae|z3Xey;UuPtZgkdq_)CZPN6W9eWBk#&UnXVdX=BE3!@gFFo1C`;1dD&lCZy|dY{6$LeQK&que|+i`>SKGhGQ#+gY2Y; z&Ez4Qip4w6|GG~Zi$hV1X3n<6b}>QtL%^r`jrvx^)#r}f)fH7@7oU-RdKBmL{9EO= z3_Dvn22t>o$kfAAgy{#aUA2L>49o;M>p3axfv*$&0l8Ji*vHNT@Gn5{wr+d?VGvrHbw$QAL3ho_p3kw zyQV+v4|Z><@Oj?H@CT1_SpL?&#{wjTT^?*OU^q6-`F0yll!Ni~YtD8?_=oRrBGXUa zU+_%+aByQBcrtn_o4X~G{PAtmTT`ehgUlngM(&oBn`$NFhxy&g=lRjK?GZ?u?{*0k z+)!l<{`S2^V))#+yyt`ZFB68ZJJ;Hm9Fd}=KXJr@RvN)uxS#GH@=>sBCM4u+ZX<~= zht_H19(#3?Uyf;dzUyxeE1qx;Q}FsN&%5^X;vbcdZtTYSDg3L+KJ9sD z1-pPcHMRGL7c+H%m2W%-ng=TJXs4W2E=I}^I-2i-h!v_IgSfwmnN-yJJ4L#!sK8CS z@Qk>mcEI0Xv7KzzoN(-b^?ANtWd$}oX_YzfK8TR3+1!d64NUF1-#i8cePDk+Ir}I- zg$f>+W*WJD z%MZl2iCnY1EkRS;vNIB?F==I^_|PG2`F2lq5utWxb~)Rz_vH6%r`Vv>GqcC?SOQ1! zT!N!?P^*Q9)at|Ef#mah$M*T}`F~Zf@85z~{4LTDJS>+p0}lrto?)lSZ%12!NR4(~_9Fch&yup^wkbW`!{&=;5dt%HgbGQtuHud21;M}p&=Km=@CYy`n zPvcSJ?)#>Wn5C_DzOe(MXWz!Dqd;hVK7XVN_vsKY!f~PBr&(sO2WmjrbEY%cUEk`1Z|dBe$pV^Rb!i z>ylLSD|-Z1MAqI_Rw>h+;aVX+eG)WQsrn4tz#5Fovzpyu==ar-(O5%y)?NowTU}(` zh(NsvAs0t**Z%|e`kfbf#Fm0LDnxe=>v(e*tLY8iI#aJi?q=L~eHCCbCbx)->FurP z-c)}_o|b>?Z$kcgVX1d*6WX$>z(my)v0M`OBjiWn-}Fh!T7fE6Orprm&2h0|l!=*l z+XpUgd#youxd}$S?3#;Sa-CWnm}J>{L`^QX=Ho#jlNTKD>}rdRYz?O1OYa3j?~A@k z4>!#vuV-c+8_ZULXu=TmWX~_acWIY>AE6z4_olc9ff||zEzB__Un7iD`aX6cDYWPU z*ya4)W$gZ0xsony(Lw@C)PMYtuL8b{Dshxq=PqH?h1|%@J%3a`afEJGZ+$fDuQJyI z)zP?KC4MFhaZJcF0pkj->OrzheYe}%CCKObDhS_s!|3VdUg!n<4?>HNWH!MH<+nSs znTN7yx8;0C?zCU2RqPH!nkDFzLfQ$o@A)!9hD<}<8(%ZPZqy?EeJd0Bo0TePTlJ3LwGB7(i(s<}RBcXj=~C+!CRa(;eb$LWpUrz0F4SsV*LcFBecfv1 zmE%%4Z@;?Z@WRDt5RPYDi9$<$<5EzK%f3c zQZj#|YGYI2yQW4juP7xB()nAqDG$sgzoP(n3GgUUC>z1@7o{o5?ip)UAG@=9)#-Gv$%JEp26vFSQRtHhxd92TVCzJV$MK3DFh^xs7ehjeI!XIV5Z zgafR9R!a9%hE?G$?<S5?q$@L_XLoklw&r7LrgS6?`tOc znqRdlZsM#jRKFXn7um|aIgQIbT3o|KqXCPfasfHvm@++5=j3Oes(-8e);OnIGh7jX zEro^xhGZ>EH7SiY3n$v|o)3FF^*%7@QUIx+t${@czT>u`vnBYn;8_!2GEp$31oL;5 z#!N9?u}40gAm*~+&K}M83e2XuB3D|qg{P*V#A0Pxxdr?6Sr%H-ZWX8RDg@zV2H*|F z_1V|&G6I&~(7v0|8FCiZr8lz9r)^{&YeX*7p+#KlAU?4x3(4M&p1m`?8u4i$I{jkY zFSePz7%(-xO=f`Iw2gh72XUz{b5+t~V5P%5ZVYkxQqP9BHvK!hnrUs2*C6Z4HuC;p z(JTw{@s-ZcyQzla79(<~t=tG-`2~0y_m&g!o@mDmpO#rjf=cRs)#r8Z-=lSBBsOEW z;2vIb$*!-Gy}UGQ>k=s%j9powL^-S3tGO!CE89djV^C1cx;IKdqD*apWrZcmbYyD# zA1NK~Z;4h~AHPM&zad#uY=>XDqc?I+Uix$YY^+mXny9Ij^)-Ttulc)b#k4zExKUf@ z;B^|Srh_8iEKVtd+9Y$Tq4RxWkz)Iph_s^0en(c}^mesisiu;1iH1%^)4Z8{1@n%s zR7zF=O_{bsGX?eM>Y!;T@(9W?5Mw0b4{0H9Pg5j%wvtvAxR9Xd8Fggne8);b&_MGi zMcc=1Qv0L^*xeOYvyO}UIFxBpUYBF#)cb=LXldS=9W(*|p*7l}-bJ{9~=WL_CrcD}Ax_m*Fp55*z#sPINpS?eiiWFQhQF%4GFmlCYZKylAn?IeX+uC+SKgmsDx+R|Ex#)J@Xn z(V0O85zV62Ut!Dx`?<~Xr+0|vrOn=Tw#6au%x5@amkJXmRNT-)&0`c|yTnyFkh;Yq zsy#v_oR^3~%!$FoAtB0!XrUo(`n!Oz<#<+vUNHMC8v^2OB>@0&p?vraO@jrQPP-L! z`E{v|x03y?w-N;f-V%l`0M%Wg0tJ;@$|`%w918HuC4c}7 zvWBU--oB-y^I^B&I)J`vdc8pi9O+4*sIv=e=WBO)rx30_oVapom=Z4Yb|f|Znw^a! z417NMSghER?Cs=t&UBn;WVP#Ds-DHNgC9K&{q!+Iyq``dL;P>axs+GC7MH~Nwx)m0 z9cqioQoRx`!C*=W8o^+e_HoZavAD!}j{1SAiLsO>Qz!ySl6DBfLK0nY=ut(EcBCXl zMGg?nN?a3wW+jybhc?M*rb&s$M`uRE1cq+N6EZ=M_Z~i=)n_M6uW6Nc2S94VejB*a zKz<`O7}y0%-+EM~4Ps$3Ol?Bnhr4|aB( zY&eMAY-n7_0+KZZPUujQS^2W$zgJd`!yJlJOFl@BQ&vw%TiCy*_%+Qg8cVMA+Ep1BG{{=Y4jN?EK?6QE1Vfh? zbgtf;G?^=iIB+W9G7VCevNH{me`_jjbk-bnj?-LGHwxqVs4tPn2mbov?MBR&5Z(z36xgg^E9CVd>`<3MDLo=q^8|y|} z6d%^=$?YVapO@OA$b&Zxn1qJr4&9>uSwg=Vzc|E8d@*N6c661%WgCPmWn&uz=%Aq= zU)DmGfI5O_k+U718US=|31%=lI=7^bkfxiJ{@`)A=9Rt-CKi(gATvM38bDHDWdT{H zWRfA#h`fk?CGm+-heV>EG4mC5^RzMJB*x-|cWA1zd6Pd<6&}&pn>1`S1>09joFXXjT-`SY>DRaG znF&wL9}(D_Vr=vcx+0xCrl7&FU0N1+Ms(%o;JMBSn&uy+dInt|oW8ceEMOUSct)^d z=@KFqfot<4YE&n0Xa zqR8KARWUELN{_yGxJAyOp_&?vz9!qh2kTGWUk*et6>Lc9FeXT|6E*KVXTZw_WjKLI z2-$bYzx7sbRgvfz6O!^d>eyv}jmVHz)?7@7t>q{7mO&K!ZNVU(WV%jyC-p=H{iTn- z;ZX{Gbfvkna<+JWX~R5b304tF+dhqPpLa(nD=dO(Sa4HQ8r3Lr6|MSMiYE(PspLsn z6oQ183l8$|YFuRA8$;-Ok|saZT3SXQS(oaQ66MvWAt}|HO7m9*^j_5(Fg7-2&H$Al z^rc^%d084nmBQ$UPTI(HzNuf_IUBGrVV4N#GO;hV{&w5#}kwdP)Gt<{dY%&+}&@l^~`_#=3v_m}+d(yn0J zMTpQUxl&ez7cdKWY>y24U6j;DdH!xDqud=rFghd5+dKX598z}oxw{-}iavsJY<;`k zcfr=E*v(RJ=iGu+slY2=jUMWRIXRyA4REoMjEFji>}dXwH+^!n@U}?-@Ewnp>Mv>IiMJ{OIlQXkoP{uKx49p$D^56&1dg#O#BI3- zC{Q(%M)C^P6(GPI7-_cJ7OF4+c2ujFEr zC#4p!`V(jBMZRsxBli#G^|)z^hHvfvG3Pq@gK*D2BYyD)mh#A*k=F!4<}C!z1WSyD zHNt~3#xs8Np7ug_GUo-bzSF7wAMr!Q;HG!FF;2co{~Jj ziR;x!qx8ayPv&RXt)g=+D`rPUinj8bL;IZ(*BX`yFQ=^u%zg#pjo-3AY*H!Rj5?Q9 zZlU!0_T!U2{=GnowQo@6h#ml^=OTo3E}jpaG9Q*!$_#Bak@)%fv8*dt%>?CCekKe0 zf}a8Z#1MuiOe$e0r@qYN)IH7pxt}ir(8^v5C90zVx?=>@+M$^NNkmx z8r^vO0C=~?cDmbk@^Q7ZW%RDrci{DdBgx{~u8rpCC@5I2DMK&}+_OyF!UKt985-Ss zs_<&QyjHl2DS19@Z^r{Cy3`Dav>F!@#2NgbmcBA5kG5$RcY?b+!7aE23GN=;-CY-V zCund85ZooWySux)-}vEu&i8w#d#ko;w|1`S?in-trCTZDv(I|b+O`ydfYpKEJA#86 zfSbg?VtBQ)j^(G+c}Fn?uPzp1-SH20>Bcz=3K4VpHiSf&z3!N;NB`sPgue1#^L_0L zMafyFw8KSkLndQs7gB}y6$e=vqv+fM$|(}wmAwmJlLloUCja0G-}hGgKb9k!{TKuP+W*D>1;NcCs!=3W zzvbeTDl)b=8gfIsS4me*(x8Uzi5~qqO54@m2eNkvFqnG9U>FL}gM*xyI_XuQ^x!%ldD6GN6a6Y5z zBPSw}2lj<$;&giPvYGZSg8vV^8wl+P;&z!_rH?5_6aQH%q(z)2{OhmP9? z4r#;~Ex?)UKU%~{_EzXM3YH~`bU_mC_h@cx3MX@~YcYuInD70epPq}7dQSh)#yzWKlEJG&aG#N(`IUQ; zme9~OSwKHD))m5k!bJ4N2L(2gKO+DO`}n=e^V+ZI*TT$z1iGor=&z~oTHPcUWT4(8 z#P)~#dlZY9+lkSQqX0cJM zCg3teAG5KYT_(irQg(=56)|5I3o`C(9R!Vyh<)`BD}23 zc}nf5-;5GTa%WEnDMgm64{JC^Z9p3FdoP<#oA*gXV0LU&8z+;PaemCbOs0{(>Pk3H zG~f=l+h+nEkkd1=S{d7s!zzQUFP7=F)Am^8=Rf*V65XO6x%xJ({}0>Y2t003YpmfZ z0N{lZH@y2ZxM*cExZRpE9Fj8bhT!6|ujZP&xrg)+VpFB}mb&K!Xk;&ky%ODNL z1T%0r_aPO$2qy5F$Vvg#C^Cy_*WTC%%$$jliM5M)>d(>?LM@7@_`46a(8Gv6#K?fk zP#Pj^s9ox(6iB@=X40X}rXD85)ql+rqkRoV`_$3!3I${J17tW5sF0F4_=jgs9&j!J zN2|D6a`(z5FypH;^>2-5uw_qkPD{bcCB_!x$frDyEMuuXY(_K(^uucuIAK?hs(iQ# zSMko8Np4yO$muwQ@Kv*k_Dglh!}l9o{hqE)@8ifw+M2)MSjKGQC<)Ut8V z0yiTt!jUB4`*VS7N8&a`4^5WIJ7^^HJG~To()Gt+a-i#s3e@8VQ?#MZvi{kY_JQ1Y zCvJftr%pK+^2+!-a;fVWmnq@8wxU-u>(lK~XGyIxJg%cmN^rswAK+hKuZ zm0jfBW3FVURFsV?wLdx&W2LW_=2c#c>E^-+@(MdyfIujZyYC%?iE1$UN=BrM6Qdjp zMpQd@xB|3Oz$$uAMW05(I@aT=KyHv5UVwVUVu2}&w-hL2d|P4}AD(Sw!em}CPxChG@y%VS1Ol?>48+>zd zyPyEIA8U0w&@f5p4I8DLyabPPS1NHkSzLj1VGgt@e6;SdO?cNW%6Dtc3N3XC-zk&k z|0It=mqQe>BP=BzUu|%VaQ1r6eCCGNgV;OuBM!nZY4{-aa4Z#Y_KAec!4bsFxZIFf z5}$HPVq$W^M3dtHCDrUHw4GUEhBsnWhTX8y_3>A9N{ICP^}45AymdXv{ladN6??y+ zyBbAfh9{!Ps;89O%N-`#L{21FlF}cI$xCc{#~qdL_5$yKe8HDx{4(_Cy&xju6yj7u zOs{D+=z~?Sg#QJoWLRSg-B>jy`CfBC+i?uQP2pE z@61{UX)U=oZf_$0t`{oZaX#dOp5mB_i}bs%XhEu-6TC4nGqI}UN^Pk?9or_rvTw}P z=%G2s%Qo~AQ9EJ)T6{Nf#xFxV)nTU?$|Ia2>W z2h@O_p}&SfbZXf)DhODX67PNAe3>v-U3m0&Vku>H>bv&|Rq63KoC;(#fMQ6v*;GPm> z+)I_bDW0$#3CxRR#Dl9kIq@=`YiOJvAhY5OXB?uVs2JTFnlsph3|A1JBKr>-bEpXZ z3-he@&hIC184opuQT6D-;?6q3n97wy=zVRvbSP{<{>y|PEfvkQ&p}qds%b<=!=>EX zFH`}ZD{~JA_X}eyYJN4?YicMDTMQas480a~kRiXKJgoEm@%ybd7lFItK0*^vi&;Xt zswJ%U+Se#}_SKVZ^43Jt1p$MaVjKOR4>xk|Rr;}6<_6UL=SygRWAX8{_~D(^9$W7V5+cVzrKDA#eu+7fY{~Ps#x;wX}AMrh@54dg;kahtF~mm zLt#KNaQn%Gm+u1@QR7EjD301|zk0LhsagQr`ZbZ|=3S!2gIP`ExW?n1Mlid8u1OIB z+m@X@gB`ZL_xO+3)6KE;NhR)`woIcSm~XQO8k>%$S^1>9Ve9F!IFUF9KVC7iY0>oJ z_&S$a2UpdM7`_UwJGBMoyocTwL_Ao;jk{~baZpFl#wHJsWCG(J9+DPVlsYO?BKig0PN$Yta&neOg`DN1dvK-`Dmey}&-$XEy(rs(hS z%p%%vhraC836ap_A&NMAB#|Z(#ImLb#WM2(E)|jB}2S!9Xt=%=#2`&G+oWD*$ z+^(fNTgPG0$w%fgx%`ZIBSh>!hzd7%`_^w_3nH(9B|PAaxPImDEX_`m21H7G!!aL= z`6)efl*7fJi5X@rR~+^%AxkNPdn12k{(9BC4c&XyPFC_nPhBlS_K7$gR*C>lW<%ZY zf60^mu^Tv!bRNUc&0;`v-uQil z-!^UD*lssK`Moayc9%JdGuj_7th^`i`lfEsJTZ0EjCFy zt^anT3*NC zvIlK5Ai>cSwm1AJ``QO^%KjjYT{1DAD(?Ax=2}$y`|nvO1kUgU+KA*&SqNgIwVtki_Qne#WyJUutp+Rm2jL?l?!dd~czE8$h*v*?T{ zrkerWUSMb=Q6B>dYUgU1@c;#pUy-k@&)@zrtv=YF{G_{kTiD^Kjr-PL;;6!H-$T<) z`q|r1Atq-mVmmowUFI)vs7!$iF&nRA5!?`ku}6Z`u!Gd0x>3<4TQ#$3enguWE=Nto zya9h2<~--tv(A)C$z{bC-@LTcLqdR8uN79`j;M;_3HcZl;o>99 zE6>Z65PO+ev8BVivGq3Y`k2S_W_HbZ#dTTLd3Ep_IIJ#~5zIOJ?WJsktwY|}obXVFw_icTgBr+! zWTUa7I0o9930fBdQV|!^#f?}$!0Q$$-vqR@ws8dezZMzuW1?-1vx}es3i8YoNy*eC zndg|kKU;-*SK6KEw<)h%X)t8evR1)7rAPOkO|z}-yzD1!4Vi}Z@lqHPU%HvB_ZD+H zUtI*eUob{46|86IJ6PYbi9v_jkAnHnoS<43ctE4mmVXM298Pz{gN0(3i|^{Gn}Fgb z2I%q`mprgOBtqdbbJU^;Sxd2F%~8sV+Vx_p3~Y$lh_%L89Hs5Z3f6ij)EFpYz*#ry zV;|!Gs^~wK=hU4z-&4TMlVg8djmPsZqjqBrbDuEDDbY!~D)V0ceVC`c%86P^_DfoC zF%Z!ev&S#&lc){KFO%D16{zb=^6(WoS@8f|?L_Z4#x@s?4=yp|YBAS_S#cc8d0V?* z1g}!vAAM>`mYiq5+dF$XF0pk@j4IOfEmXo4AhtbD$nIvowAKYHD==X*k89kti15%K zJLSVVozy;CgYT$4PQkW(zv-1bx6Eg`0URWdm?NVXKSHL2Bd&@B+A5t+r+bUZNh0hy z4h|4@roP?e)D{_U=PNF~Qt#gIeB&;y!WUqUUNh|KIlW%p&b!_ZX9Duv7&bkC zG`9r(wJp^G~2C1wtCba21)*EDO%G3lV;<(Qw76IaxLc%ewFo2+~T*yy~C8# zU6%j?NGD(8@5C=eBaKJisY{Qz7T~=#zFSwj@u$CAug6Cqx_uDA0MX2xVkpZnzs4=k zf~#&wn%9Tu;E&HvA@mXTLH;g&ZKy3#nue$yJ9%Nnvmt}w(4@9{ooz}^T)prw$4oP8 z-(PxFy`;@|U-TGp&)40@>59RbOVj*;U7PIF7BJYW9Yb?4=E-r7PM>DON+OyU#Qx-P zei|AXOf5o*>NZTSfA95)eL6?rm4-qum{BJ;8EEZT-Yv?Wa2BwYIcbsj#g~G4o5wQaFA=xn# zpKW2+TfX05wT`Hg=*LzfE^d3uExk)2(-VjT#7f$aW2xKsK z{K!7ECr~UXo6ath_06+sh7{Ldc`e5r@sChLNOlPReOlM<@kN$3fA$`)_v1Xx<@UwV zy!!tm#NR_hKra2A23J+OoRCoJ2#)xfq+AZW7lp~@xE%&@H!AAz5;h%E@gqWrpxsNa`S5@nEnAfI&Z>%X~ zNx{mHioe)BR!30%Cb(f=dUQiWJ2c~Y@)Nm%SbBEW!N8waV{4XW35-Uk@hZ&!(so*S z)rP<1X%JYPn1qn>P$TvkI>{b{vGEY!c}w7!4BMh8LeEp;okrWBP>z_mb4&gjK2ki9 zJavm>8cEb9EHy*fbNzu0i*8yAa1xS_m{`%`n9A`r(vI4h`_q-sjwqU&n3F0|6ZC%1 zmxT#cR$7Sb$YyAD1uxCbr#gLtncy$kAo7gKo&K415TYQ`R-XQ1l#e`Ei>hL$r_C z<5faj)l}2?p-t)5X1yf%_X-%>VnYehi5xRGTW!w%JTsK5@^s1w%cOr&Z4f`}7| zO(&<(60_o-E}i;cK8nz{NcSJY4NCb?iI!}%b6;oo)M<4orv{3Fc+25&L`6g_4WxS3 zKk}sM)Fh%6Y;r}RGeiwakC<$XE&bm~)b6+q8kE?e6D@Ce=V)h5e3p!SxI<(|F9S>H zW`mgiXnexh#XeyZ;mYZP!z=Y?obc@@EA@UC4aZ4%L=TwEwL+P~%R`7YsW|z1t`x?7 z2{1W8*o8pU3xE?esS88RPkrkN2eX|4e>U0^ipyuj2J;Rb86#6M7{ZD;_QzDXiOmMo z)unm$y_gWEiED)BTDCq>+~|d+-lAqJolflvCSrXsXI&lP{!2DIx|NeOA;X+WA}z+r zG-ybKu|>d0b()A3ze%~o-9av1F-c>`AdmXDSj({{A5g0-a}ocU3wgq|uu?MLs!1bR zHq7E!qMoE-6s2M0pi_~uN&T4(mV{fcT;_hI5U-wOy<<>8-7Drv7crdB@bmc>dQ$0E zDr(!A6MTLgti4zt?7?}FTFo=lni4sbVO>WTYx|$ytMI;Pc-XWXEnXSTYS%D7niBt(+*iYeaQn{zQblpQq}HAGPT+^7`E3n z3Y6Z18EQkWc_{SACV2~z9jh75B6?5RC3Kr6J+ujzdY581j7>NN8jKC5RT4MrE5X<^ zBkpw!cxrpU%kijg)=9D?JN7v(?j(};2&`iET7ilProCi)Ii)A+rXT4()^8Zdom5=5 z!UC%34&_|RSB(c<@fzK(YFL^4gntk!vn0Fs8=XF{sybB6Bj^Ry8PniVDhyh3B(@QG zd-7S~NZ)u6N|D<$5nXn>7fn4a5w2{=o*A1gSsa3_LK?lli^-b2E04t)&FgZU2CaOq8XzS`9I1 z5lKtbwNER#s6VV>v@7*r+t?Ro&os`KkrvNbNz*d6ppPWGu?U9FjAZjb5Vz*fd>KHX zv!xe+cA~4O44p~z&egGOD4?y_^Ca1-0m|qqiehILsvkefb*4K$SnH$Ro*q~Bg%l?( z6h1T?X*|SmB;d+;UV~xrFbzVbICCk+cfafS%H(@Totm;UYpU>dFLZ2!i4 zDH>K9!~PHna7sZw(an$_wFTAPT&~JDxFGLM+Uze?_WK1&SEs04B#$uzzXI(Ha8b{7 zKx1DOilA-Hy)|P(4`sjm1^7L8(9mZ0Z&g|DVo8S5E}8zU787N{(%UdHqg{Q1bEnGggD9$C~mVg%WtJ2Q=|o;?=g6vt`1ekh4nSX zM;*N1srNBoOVtOazqv_{>t9#|h%I}EAWz6O7fuCZg974Pd0%TFEVI9k&h~xKsmj(G z&+DjX2^)&oT2BWMIU{uY2B%+guUDfFN~&`|P6{ILA!G_XIgSbv(}~TEssW3FQh=0; z8cz!q^O*^V7=fY+m^8yR`{F`^Iljh=m#ViZqooyLZc6Veq{9rUhdOP6M8JsHghLP) zJ~`0y7x`9eP-1@-yO24qyS-zeg^Th6%keTRj*$Hj5oZ8UmTVs)RBSw`hI z5QX~CHcc38FtveI1cgPHP`D_{+(Jw$C>>7%0 z$3AyO2F{X8%e>Vfe_0dG5KiHu+t{4-Z3xdHD$4Xto*7v9(0IUgRA5BK*7H+(_&LLzsRqik z8j)Dh3c`u0uag!y|2cd6-u!swb$eqE7+!Yh)!;00=~wC1=4|@Kwrhsr`Q0 z_A@chGx;^`bM@&ntYIDM-RMzPY~wf?ym_Dy?=MUCAj-G8J7M5A<`Ft>l0O% zso!JZlg+-Wrp0=~WgyM6GlqxeJd&2=YC2_%CUJ%U6VOBbWFkj%uH4Kh^k^78!Rqe= z@pR!bHrwa~D}G=yvwFKM}^BvN_XT z$m%hGMex}$0W-wR)W4mGC+6Ebf|GZ@6OQ$vd&b83Ec*V0lso#S@zN|{1MdQVlPRxb zF4&6w&8p2~t3z#mYhor9j>sBatr>Ne5Qon9rXV^F1*aqOn-8^{w;VwKiAd?5ZLH89 zfi+SBWR>-r|Mhwk>9d7(!mF@-?%z@~s~z+v(RhiXVApGC3CpH`)$luLp5c9yoe}@F zdT^J`JhwXrj$>60LM(>P1A^|SCxK8+gpp1xL3sDJ9*(idyN^`zZ{6^TNB>ZhPA607 zMWK0BHL4*G@F6SA41~4MGfcP_$%CV2cgZiH=&p7SEH9}fQ7PfjXGHQl*(sk7{rkuW z5O&lO@VXRkMYXPZ=L)ntO5r2_Ii%*LABa-g$cHdD3Tu$Z$Kw6m3K<3Tn z_I&z1KiJ}8tV#4<1f)kSH?|vz#x|2gcHom1o9BWNtx#e^+NpSC{SP&5%X%I54Q&J9 zs0{2TeX&7tSIi!Nfsl7I8!V?bwH>)@?N+stnGQT1fNXLg48AOpRWEl*D-dPZ98I$+YH6)0bRma5}T22s%0 z=xXlVfb7VM(g61y%8v3V6q->89^U$Idor5OT8UYkZw0j(cz!zP%LN~KtNyql?+k16 zmT1)sTJwUVocHY(Jj_SUzr=d!5{*gI>$*EUP1?>wmb*k+GN^_h*JN`u-n`X+@Y3Er z&&dPpIm7Z}OIhpl9T9U~Si}I37L(cdUAPV{m4hHWcIsfDzanonh{?JB9e#WecxX4# z$RYg-Nl73BbP4$>H2T5&$pz{=wtkEACSZbTe{GY+1FKnx@P|d4LZQguM#r$+0b%~; z(BQ{SJqL~v6Pln>G`?*qfR)zrszzks&i1MO2N6kcPT{(SBbRd zs>()2KoHc1t<1qr$$vjVmJA>fLHfDN2eTCUMLK!C`Xi>{^=)tA!C*yaaJP7 zxeYZbpq2%ktTf-eVpwO|559OlZ}Dkhr4`f zM{CU6Lo>Lm1D?LJA>$TbKZHI5vd1!Z730x`pllv#E}3M}qp86*Iy4D{r~{WM%)v?A z!e$#0GgfwTSPpm)@EShJ*Kz3y*Z(01cxc_F2BSIF7kf<>K`8lTW9$FW;E;$~4U344 z_KE1-3w)$y{Vj5h#SHhPu=wmnc-}Jn>*%jWmXudFNw2*2oO+d(7M#lxjS{LyJs)dO za6a%13aJPa=)40nPn<=SZiVx)<{7uc^&wN;!|J$(0B&j#VDGn|#p#COjL^IQcAQ~| z^WWOf{O2b}9+~DRVlwFnSSY&z>mshkS+{}vUJR^v@S^VJYMcXnq66)*u|V{!5C`W1 zpTO(+k?_g^{_$6t97S-EVTQr4eIxV_7ZGlA?JhhUZzt3~$DZaoG~&Bg$|Pv`@G)4^ zn$U&?ZsZKdTEQ$4Xek^s@3Y;&>=mMw;h_|9N3%GTjNKoaK(y{yv1&&?_>?;a^0YoE z8GI?fz?d>rn>F&l80LF;v}iatu|T;Msp>*=v~Egj%^69t>H&0y1|!t&a03alrGGD- z0#Gtu*l4mwfmAV+Lf!RrP_rwd)q0P$#c2O4p-fwl4*Vk!1xL-?Jp=qm%rfu_`Au}y z&q$u;wmchQAXZI*E^ANIotP~Sm6T>6MKLVTQXzuCtjDgrbpQBm&A#Uk9h==7ST)#8 zGbHuQVl7Tn>cOZpn9(R?W)jwA;waa>}}hOKi%SF*a?Oa=~WkM z%CI1%Xv(y(2zPJQU@@l^6|FL?3!U?pvmoWL`6gW!0plb&8UcicrscaLwd*=ByHdx9 z^rpNz1&2Q58Ad^B$#KNrYszzQ^x2kcun?RcKHy)TVpDUV;y3>EP8Ngo^x78_Rk3ON z8Itr%uI1UiH&o_-%=crZ5io3{BN3sYAY*G`$QI)q_$J7$tFa0wlc3}gm{Edm)EEO{ z3S8@La%kxrpy@Wg-&NnwqTR0;7T7^Ul$+(#ZlCb-II=hON z`(wm@^($xFjkL|}Ih-v&d^7#jky}t*Aib^Iu2i$6M{vBq`sMNWHzSs*?a6K2B5|DE z3%K`83((`RNuL+a`ZU=4989k|xO|xOu{tGT)i9?XU?*&1jj?>fO_h%xaAYFs3`cot znSlH%tm{&yZyu?(G|&nUCk&NqWv-+Oftn>``NiKvAgLCdv?RnA49ebmcOC+Y6hg59 zgP9IIT)u?^R=76XXBiBY|3eTgu&LqN3=9Glykw;XKN(ES+%f}_RBLk42Ew9t5eaT% zVW96F_zykHhwzC$`$W%V?|35Q67xH|f?bzd8SDA|2gf7(gdYa}2S>04gUfa!0W+lf zHh37rZ1mv%FmTtgE!9eJIT3=IGJ+r@Z2137cmF_*W>pXj_aJqa|68Rw3Y4UoJ8U8_I)b1<6QB7`b# zy}3R`rB`<7K!(xv#eR{H8VJ(f`g29B;SKeB{lvvO>3e+VCxc~dFJG#0B?7QzTd^|< z&%e=^v(VE1yN{&S{gWVu)uVUzqL-w+kXrZKzy413%@|Y?1!vg4L(mQBC_GMOixFZ8 z)~7DjcboBW30P!t`mXWea$PLs%BQQC+Xa>PMYuvPZk-RigvKTDD_q3y_EwCjS<~0@ zu8gKpc>LX)_tzre9XP=&Mr(ZY~3eJIK${~Dk^wU zsvcB84%L*ccIqpi$OEtmz?_7FtU$jeJN=$t#s|ydfJLsX7FR-&6-PT;kdwV=%q#a9 zl3$wv$>OC+xaduwenNL;gI`VkqZxZ0*X)_0JvZU{Ns&PMx_s#KREAXw53zNxlER{d zdZ+oy!?|!yOtHYJIzjAsw+lygXj5_0gN7x2(UG~M^Nx(UxQKc*#Ar^aP)J8Z+wG_I zQrCS3n14;?^qeZ{M7DwI$46qC-ID6yjLje2%gbdo;anOsx+i-U}pUm6I z+YNrbsVW{(zSKn)Xs;cLva(7XvS*D*7%HQ?G5#w~EXs!c)I@nxjOp1{c-SUbsS?%p zX?z_1|8 z`N^pcaxc81QT1puxNWm5#GrpI;i@dZYD>f~yjoooE4(VDYoo5~xs8v;dN)@FVbJk! zWc%oKPDWt)ac8Qk{~xCxk+O#xmR{=5ZB|O-kdF||_g@Hr&_(95lCxC^$zv%!zq)dn z3iVG!ka{%0d_0wX8Ni*9cbQ(YgyL41nAXEOPv+cW=1hc!-LR* zD%R^_zkS{gn4q8Q=k;#EnJWd81EaOSebxK}Z{3$N=7}phDY7Sw@xl(-^B(qks`wLB zb_RAz^N|Rw@cg~^TRV{)hhUfMPQy*PBL=@S0V9gDP9rDENJ8rs7v@+{KutAbB8;2` zb)#qO%wDFYCTOVA&$6NO!)Ggd+Wban$)tZ|{C@Gv`^l>quv586hYO4AcySY4P@z{5 z`p>bU75#gpIrHmND6YHo!2FIG_gIn-(?EB@B{0!kM}R+(toNUHW5jS>bzq-=`7+$G z*hHt3owteRMJGB)Vl&H8Q=vlEP%4^YKx&`Wo^IEv@-8RBvDj4Sa;@_2mbFqN#IfD2 zXqa*UZMV>And_IMr~W&0U-K-RyigPKhfnlu8H>EVIi0)_A*+0;GmM<6&V@;jx6vBD eCElXI<8bgNg?5Nz8s*Uk Date: Wed, 4 Nov 2026 18:29:16 +0000 Subject: [PATCH 32/52] tests: pin the harvest's reviewed sign readings (review A H-5) tests/fixtures/ab2026_sign_readings.json pins the relation signs.py gives each of the 76 (metric, sign convention) pairs staged rows carry on an oriented metric, and a test holds every row to it. test_declared_conventions adds H-4's negative-leg and two-leg cases, and the property test's filler can now write letters and a negative leg. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_01LwQq6JvunwtJCAckL6L4aM --- tests/fixtures/ab2026_sign_readings.json | 382 +++++++++++++++++++++++ tests/test_signs.py | 47 ++- tests/test_uk_ab2026_harvest.py | 29 ++ 3 files changed, 455 insertions(+), 3 deletions(-) create mode 100644 tests/fixtures/ab2026_sign_readings.json diff --git a/tests/fixtures/ab2026_sign_readings.json b/tests/fixtures/ab2026_sign_readings.json new file mode 100644 index 000000000..a0bd6c717 --- /dev/null +++ b/tests/fixtures/ab2026_sign_readings.json @@ -0,0 +1,382 @@ +[ + { + "metric": "average_annual_gain", + "sign_convention": "positive = gain to households", + "relation": "same" + }, + { + "metric": "average_annual_gain", + "sign_convention": "positive = gain to taxpayer", + "relation": "same" + }, + { + "metric": "average_annual_gain", + "sign_convention": "positive = gain to the EV driver (saving, as printed)", + "relation": "same" + }, + { + "metric": "average_annual_gain", + "sign_convention": "positive = gain to the pub (tax cut, as printed)", + "relation": "same" + }, + { + "metric": "average_annual_gain", + "sign_convention": "positive = gain to the ratepayer (protection, as printed)", + "relation": "same" + }, + { + "metric": "average_household_income_change", + "sign_convention": "positive = gain to households", + "relation": "same" + }, + { + "metric": "benefit_cost_change", + "sign_convention": "positive = increase in state pension payments relative to earnings uprating", + "relation": "same" + }, + { + "metric": "benefit_cost_change", + "sign_convention": "positive = saving (state pension payments lower than under the current triple lock)", + "relation": "opposite" + }, + { + "metric": "cash_requirement_change", + "sign_convention": "as worded: positive = increase in the financing requirement", + "relation": "unknown" + }, + { + "metric": "cpi_inflation_effect", + "sign_convention": "absolute magnitude bound, no direction given", + "relation": "unknown" + }, + { + "metric": "cpi_inflation_effect", + "sign_convention": "as worded: positive = increase in CPI inflation", + "relation": "same" + }, + { + "metric": "cpi_inflation_effect", + "sign_convention": "as worded: positive = reduction in CPI inflation", + "relation": "opposite" + }, + { + "metric": "cpi_inflation_effect", + "sign_convention": "positive = reduction in CPI inflation (as worded: reduce inflation by)", + "relation": "opposite" + }, + { + "metric": "decisions_effect_on_borrowing", + "sign_convention": "Note (verbatim): A positive sign implies an increase in borrowing.", + "relation": "same" + }, + { + "metric": "decisions_effect_on_borrowing", + "sign_convention": "Note (verbatim): This table uses the convention that a negative sign implies a loss to the Exchequer (and is therefore an increase in PSNB).", + "relation": "opposite" + }, + { + "metric": "decisions_effect_on_borrowing", + "sign_convention": "absolute average, no directional sign", + "relation": "unknown" + }, + { + "metric": "decisions_effect_on_borrowing", + "sign_convention": "as worded: positive = additional investment income", + "relation": "opposite" + }, + { + "metric": "decisions_effect_on_borrowing", + "sign_convention": "as worded: positive = borrowing increase", + "relation": "same" + }, + { + "metric": "decisions_effect_on_borrowing", + "sign_convention": "as worded: positive = borrowing reduction", + "relation": "opposite" + }, + { + "metric": "decisions_effect_on_borrowing", + "sign_convention": "as worded: positive = increase in borrowing", + "relation": "same" + }, + { + "metric": "decisions_effect_on_borrowing", + "sign_convention": "as worded: positive = reduction in underspends / borrowing increase", + "relation": "same" + }, + { + "metric": "decisions_effect_on_borrowing", + "sign_convention": "as worded: positive = yield / reduction in borrowing", + "relation": "opposite" + }, + { + "metric": "decisions_effect_on_borrowing", + "sign_convention": "mixed: row 551 (negative = loss to Exchequer) plus row 555 (positive = increase in PSNB)", + "relation": "unknown" + }, + { + "metric": "decisions_effect_on_borrowing", + "sign_convention": "positive = cost to the Exchequer", + "relation": "same" + }, + { + "metric": "decisions_effect_on_borrowing", + "sign_convention": "positive = increase in PSNB (matches EFO Table 3.1; contradicts the workbook legend)", + "relation": "same" + }, + { + "metric": "decisions_effect_on_borrowing", + "sign_convention": "positive = increase in PSNB; negative = reduction in PSNB", + "relation": "same" + }, + { + "metric": "decisions_effect_on_borrowing", + "sign_convention": "positive = increase in borrowing", + "relation": "same" + }, + { + "metric": "decisions_effect_on_borrowing", + "sign_convention": "positive = raises revenue (reduces borrowing), as RF plots Figure 10 (the OBR's borrowing sign flipped: workbook F10 I15 'Signs flipped from source')", + "relation": "opposite" + }, + { + "metric": "decisions_effect_on_borrowing", + "sign_convention": "positive = yield to the Exchequer", + "relation": "opposite" + }, + { + "metric": "decisions_effect_on_borrowing", + "sign_convention": "positive = yield to the Exchequer (reduces borrowing); negative = cost to the Exchequer (the tables print no legend; convention observed from the rows themselves, e.g. the bank surcharge rise is + and the Household Energy Payment is -, as in the AB2025 Table 4.1 harvest)", + "relation": "opposite" + }, + { + "metric": "energy_bill_change", + "sign_convention": "negative = reduction in the typical household's annual bill", + "relation": "unknown" + }, + { + "metric": "energy_bill_change", + "sign_convention": "positive = reduction in the annual Ofgem price cap (as worded: 'take around £50 off')", + "relation": "opposite" + }, + { + "metric": "energy_bill_change", + "sign_convention": "positive = saving (a reduction in the bill)", + "relation": "opposite" + }, + { + "metric": "energy_bill_change", + "sign_convention": "positive = saving (a reduction in the household's bills)", + "relation": "opposite" + }, + { + "metric": "energy_bill_change", + "sign_convention": "positive = saving to billpayers", + "relation": "opposite" + }, + { + "metric": "gdp_level_effect", + "sign_convention": "as worded: positive = real GDP increase", + "relation": "unknown" + }, + { + "metric": "gdp_level_effect", + "sign_convention": "positive = higher GDP (as worded: 'increasing UK GDP' / 'boost ... economy')", + "relation": "same" + }, + { + "metric": "pct_change_after_tax_income", + "sign_convention": "positive = gain to households", + "relation": "same" + }, + { + "metric": "poverty_rate_change", + "sign_convention": "positive = rise in the poverty rate", + "relation": "same" + }, + { + "metric": "real_income_growth", + "sign_convention": "positive = decrease in real household disposable income per person", + "relation": "unknown" + }, + { + "metric": "real_income_growth", + "sign_convention": "positive = gain to households", + "relation": "same" + }, + { + "metric": "real_income_growth", + "sign_convention": "positive = growth", + "relation": "same" + }, + { + "metric": "real_income_growth", + "sign_convention": "positive = higher real household disposable incomes", + "relation": "unknown" + }, + { + "metric": "real_income_growth", + "sign_convention": "positive = real income growth", + "relation": "unknown" + }, + { + "metric": "revenue_change", + "sign_convention": "Note (verbatim): A positive sign implies an increase in borrowing.", + "relation": "opposite" + }, + { + "metric": "revenue_change", + "sign_convention": "Note (verbatim): This table uses the convention that a negative sign implies a loss to the Exchequer (and is therefore an increase in PSNB).", + "relation": "same" + }, + { + "metric": "revenue_change", + "sign_convention": "as worded: positive = AME increase", + "relation": "unknown" + }, + { + "metric": "revenue_change", + "sign_convention": "as worded: positive = additional spending", + "relation": "unknown" + }, + { + "metric": "revenue_change", + "sign_convention": "as worded: positive = cost", + "relation": "opposite" + }, + { + "metric": "revenue_change", + "sign_convention": "as worded: positive = increase in borrowing", + "relation": "opposite" + }, + { + "metric": "revenue_change", + "sign_convention": "as worded: positive = increase in net business rates revenues flowing to local government", + "relation": "unknown" + }, + { + "metric": "revenue_change", + "sign_convention": "as worded: positive = net cost", + "relation": "unknown" + }, + { + "metric": "revenue_change", + "sign_convention": "as worded: positive = pension spending increase", + "relation": "unknown" + }, + { + "metric": "revenue_change", + "sign_convention": "as worded: positive = reduction in receipts", + "relation": "unknown" + }, + { + "metric": "revenue_change", + "sign_convention": "as worded: positive = reduction in receipts / behavioural erosion of yield", + "relation": "unknown" + }, + { + "metric": "revenue_change", + "sign_convention": "as worded: positive = reduction in resource spending", + "relation": "unknown" + }, + { + "metric": "revenue_change", + "sign_convention": "as worded: positive = reduction in spending", + "relation": "unknown" + }, + { + "metric": "revenue_change", + "sign_convention": "as worded: positive = reduction in yield", + "relation": "unknown" + }, + { + "metric": "revenue_change", + "sign_convention": "as worded: positive = spending cost", + "relation": "unknown" + }, + { + "metric": "revenue_change", + "sign_convention": "as worded: positive = spending reduction", + "relation": "unknown" + }, + { + "metric": "revenue_change", + "sign_convention": "as worded: positive = welfare spending reduction", + "relation": "unknown" + }, + { + "metric": "revenue_change", + "sign_convention": "as worded: positive = yield / reduction in borrowing", + "relation": "same" + }, + { + "metric": "revenue_change", + "sign_convention": "positive = cost to the Exchequer", + "relation": "opposite" + }, + { + "metric": "revenue_change", + "sign_convention": "positive = cost to the Exchequer (RF calls the figure a cost)", + "relation": "opposite" + }, + { + "metric": "revenue_change", + "sign_convention": "positive = cost to the Exchequer (as printed; not re-signed)", + "relation": "opposite" + }, + { + "metric": "revenue_change", + "sign_convention": "positive = cost to the Exchequer (higher RDEL spending)", + "relation": "opposite" + }, + { + "metric": "revenue_change", + "sign_convention": "positive = increase in net tax as a share of GDP", + "relation": "same" + }, + { + "metric": "revenue_change", + "sign_convention": "positive = raises revenue (reduces borrowing), as RF plots Figure 10 (the OBR's borrowing sign flipped: workbook F10 I15 'Signs flipped from source')", + "relation": "same" + }, + { + "metric": "revenue_change", + "sign_convention": "positive = saving to the Exchequer (as printed; not re-signed)", + "relation": "same" + }, + { + "metric": "revenue_change", + "sign_convention": "positive = saving to the Exchequer (counter-fraud benefits, as worded)", + "relation": "same" + }, + { + "metric": "revenue_change", + "sign_convention": "positive = saving to the Exchequer (reduction in State Pension spending)", + "relation": "same" + }, + { + "metric": "revenue_change", + "sign_convention": "positive = yield to the Exchequer", + "relation": "same" + }, + { + "metric": "revenue_change", + "sign_convention": "positive = yield to the Exchequer (reduces borrowing)", + "relation": "same" + }, + { + "metric": "revenue_change", + "sign_convention": "positive = yield to the Exchequer (reduces borrowing); negative = cost to the Exchequer", + "relation": "same" + }, + { + "metric": "revenue_change", + "sign_convention": "positive = yield to the Exchequer (reduces borrowing); negative = cost to the Exchequer (no legend printed in the document; convention observed from the costings themselves: p.38 says its figures 'represent net income from developer fees' and prints +50m to +95m from 2027-28 after -15m in 2026-27, the Bank Surcharge rise prints +, the VAT cut on domestic electricity prints -, and Table 6.1 prints the same values with the same signs)", + "relation": "same" + }, + { + "metric": "revenue_change", + "sign_convention": "positive = yield to the Exchequer (reduces borrowing); negative = cost to the Exchequer (the tables print no legend; convention observed from the rows themselves, e.g. the bank surcharge rise is + and the Household Energy Payment is -, as in the AB2025 Table 4.1 harvest)", + "relation": "same" + } +] diff --git a/tests/test_signs.py b/tests/test_signs.py index 15943acb8..59cccf0e9 100644 --- a/tests/test_signs.py +++ b/tests/test_signs.py @@ -200,6 +200,38 @@ def _row(metric, convention=None, *, country="UK", source="x", annotations=(), * ), (_row("energy_bill_change", "positive = saving"), "opposite"), (_row("energy_bill_change", "positive = higher council tax bill"), "same"), + # review A H-4: a phrase's negative leg doesn't decide its sign, and a + # phrase that matches both lists reads unknown + ( + _row( + "decisions_effect_on_borrowing", + "positive = increase in borrowing; negative = reduction in borrowing", + ), + "same", + ), + ( + _row( + "revenue_change", + "positive = yield to the Exchequer; negative = increase in borrowing", + ), + "same", + ), + ( + _row( + "average_household_income_change", + "positive = gain to households; negative = loss to households", + ), + "same", + ), + (_row("revenue_change", "positive = cost saving to the Exchequer"), "unknown"), + (_row("revenue_change", "negative = increase in borrowing"), "unknown"), + ( + _row( + "decisions_effect_on_borrowing", + "positive = increase in PSNB; negative = reduction in PSNB", + ), + "same", + ), # levels, and changes with no stated convention (_row("poverty_rate"), "same"), (_row("revenue_change", country="US"), "same"), @@ -256,11 +288,20 @@ def test_a_result_oriented_to_its_claim_is_same_whatever_the_text(): for phrase in phrases } ) -_FILLER = st.text(alphabet=" 0123456789;,()=-", max_size=12) +_FILLER = st.text(alphabet=" 0123456789;,()=-abcdefghijklmnopqrstuvwxyz", max_size=12) +# a negative leg that names any definite phrase (review A H-5) +# (a phrase that itself describes the negative side can't head one) +_NEGATIVE_LEG = st.sampled_from([p for p in _DEFINITE if "negative" not in p]).map( + lambda p: "; negative = " + p +) -@settings(max_examples=200, deadline=None) -@given(phrase=st.sampled_from(_DEFINITE), before=_FILLER, after=_FILLER) +@settings(max_examples=300, deadline=None) +@given( + phrase=st.sampled_from(_DEFINITE), + before=_FILLER, + after=st.one_of(_FILLER, _NEGATIVE_LEG), +) def test_one_phrase_reads_opposite_ways_for_revenue_and_borrowing( phrase, before, after ): diff --git a/tests/test_uk_ab2026_harvest.py b/tests/test_uk_ab2026_harvest.py index 2de9349df..e8b7e1821 100644 --- a/tests/test_uk_ab2026_harvest.py +++ b/tests/test_uk_ab2026_harvest.py @@ -364,3 +364,32 @@ def test_every_borrowing_row_says_which_way_it_runs(): assert relation in ("same", "opposite") or basis.endswith( ":not_this_quantity" ), (fam, r["conditions"].get("sign_convention")) + + +# review A H-5: every (metric, sign convention) pair a staged row carries, for +# a metric signs.py orients, with the relation a reviewer checked it reads +# (4 November 2026). A new phrase, or a signs.py edit that changes a reading, +# fails here until the pair is reviewed and pinned. +SIGN_READINGS = { + (p["metric"], p["sign_convention"]): p["relation"] + for p in json.loads( + (ROOT / "tests" / "fixtures" / "ab2026_sign_readings.json").read_text() + ) +} + + +def test_every_staged_sign_convention_reads_as_reviewed(): + seen = set() + for fam in FAMILIES: + for r in rows(fam): + metric = r.get("metric") + if metric not in signs.ORIENTATION: + continue + pair = (metric, r["conditions"].get("sign_convention")) + seen.add(pair) + assert pair in SIGN_READINGS, (fam, pair) + relation, _ = signs.classify( + {"metric": metric, "conditions": r["conditions"], "country": "UK"} + ) + assert relation == SIGN_READINGS[pair], (fam, pair, relation) + assert seen == set(SIGN_READINGS), set(SIGN_READINGS) - seen From 1ecc1e5f5dc8215be4e046ba7cee3ca008a60aa9 Mon Sep 17 00:00:00 2001 From: Claude Date: Wed, 4 Nov 2026 18:30:13 +0000 Subject: [PATCH 33/52] harvest: regenerate PROPOSED_KEYS.md and COUNTS.json; README rules and counts - PROPOSED_KEYS.md: 86 keys (15 measures, 58 packages, 13 options) on 973 rows, from 85 on 958. - COUNTS.json: 5,952 rows (450 restated), from 5,948 (446). - README: rule 8 states the lending and financing rule (OBR 5); rule 10 keeps restatements' metrics (OBR 12 (a)) and says why RF's fuel-freeze totals keep revenue_change (RF 3 (b)); a section lists the contract's additions (H-1 to H-6, categorical value_kind); the family table, FAI's 29 October post, and a slug table that no longer lists uncarried families as directories. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_01LwQq6JvunwtJCAckL6L4aM --- .../harvest-uk-ab2026-2026-10-29/COUNTS.json | 44 +++++++++---------- .../PROPOSED_KEYS.md | 33 +++++++------- .../harvest-uk-ab2026-2026-10-29/README.md | 40 ++++++++++++----- 3 files changed, 68 insertions(+), 49 deletions(-) diff --git a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json index 33432f8dd..4ac536a8b 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json +++ b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json @@ -1,8 +1,8 @@ { "uk_centax": { - "rows": 110, - "primaries": 5, - "claims_sha256": "ccac3eb6af8eff5eebba81f24ab7b8082a13526ac308943f72db18b9e445bdb6", + "rows": 112, + "primaries": 6, + "claims_sha256": "2529e86fc898493e4e045072f6aebff2c757470cd9d578acbce8ca0790ce348c", "sources": [ "centax" ] @@ -10,7 +10,7 @@ "uk_devolved_govts": { "rows": 19, "primaries": 2, - "claims_sha256": "745a5df6a7d0250658bafb2ad71ca49f90ec383a806633ef82adf381164d4440", + "claims_sha256": "8f47eb3c1d0d67c76b19cdb02c09eb7211e2ad26e8957215b755009a64ba5dfa", "sources": [ "scottish_government", "welsh_government" @@ -19,23 +19,23 @@ "uk_hmt_costings": { "rows": 597, "primaries": 1, - "claims_sha256": "d84725f64c8ba4d14060c90b01df55f92f10490f0e042e5a2ad5fcbcc835bfb3", + "claims_sha256": "66a8e7fe77952438747bca6dde96be71a2f1deb8cf88716bf4a7bbb8ada7242e", "sources": [ "hm_treasury" ] }, "uk_hmt_redbook": { - "rows": 432, - "primaries": 5, - "claims_sha256": "fbb4c6e185201f8904007545f17b750bbae35182833fe2b77887dbc385d23b69", + "rows": 425, + "primaries": 6, + "claims_sha256": "82e3bcdd48ed7295b2d2c293273e7c5ae9711f4646d4b83d06144011dafc0342", "sources": [ "hm_treasury" ] }, "uk_hmt_tables": { - "rows": 762, + "rows": 768, "primaries": 2, - "claims_sha256": "b962aabb0064fe51860db293bee21aaff644894d472e409d05bd96dd8f2276e8", + "claims_sha256": "e44ba6ae818aa4f4c759ba135ea5a33a3dbbe0753f81e64759a8b0d25cc02819", "sources": [ "hm_treasury" ] @@ -43,7 +43,7 @@ "uk_ifs_ab2026": { "rows": 144, "primaries": 14, - "claims_sha256": "bb7efa637a6f748fe94d43592c0847ef9d759f710c513e5ac4a311dd769d8ba8", + "claims_sha256": "b76375ec9ccba03011181edcb47601e81f281636fda880f09688446bb582363d", "sources": [ "ifs" ] @@ -51,23 +51,23 @@ "uk_ippr": { "rows": 3, "primaries": 3, - "claims_sha256": "457ac8cabd815ca698b375f3271b3f6861f99bd958c6b7db779ede0a8e993c4b", + "claims_sha256": "dcabaf8fb3157e4fa6d18c410add5a430eadbdc02784dd9d7831b2cea1bc2ab9", "sources": [ "ippr" ] }, "uk_obr_tables": { - "rows": 2797, - "primaries": 3, - "claims_sha256": "6ae80eb0cbc011fa8dddd6aa5fd8eb8f6cce3b675790d6c79be04907ad92fbeb", + "rows": 2800, + "primaries": 4, + "claims_sha256": "ceae9897d7c680f5c1be6e04a7dba3eb7114e1f542fcb5ae2236cd775d9fa199", "sources": [ "obr_efo" ] }, "uk_rf_ab2026": { "rows": 766, - "primaries": 7, - "claims_sha256": "48e9f2889097817a177bc98f9b0a903020222be6c034b2f74fb61c4bf35dee6b", + "primaries": 8, + "claims_sha256": "e8f75767a7f871c0f2030aaf2d47348e35f6caa3418ab773e53b77205a682325", "sources": [ "resolution_foundation" ] @@ -75,7 +75,7 @@ "uk_thinktank_misc": { "rows": 12, "primaries": 5, - "claims_sha256": "15a0efea0765f575d08a7184f1f7ca60edc20d53b3866edbe596638283ba3613", + "claims_sha256": "20695e34914c2b80fc58b92aa87b7a096bc974a004939b35a5fae5a135aa685b", "sources": [ "citizens_advice", "cps", @@ -86,7 +86,7 @@ "uk_tpa": { "rows": 305, "primaries": 10, - "claims_sha256": "4225c7cddab4e4d85a94efd4e624319d4457ed040f7b2656aae1ee6a5b5826df", + "claims_sha256": "241f625fb725640835bde53dd5a2d1217ddd758f9df260abf778ac4bf96ba7cc", "sources": [ "tax_policy_associates" ] @@ -94,14 +94,14 @@ "uk_trussell_wpi": { "rows": 1, "primaries": 1, - "claims_sha256": "8589f64e0c03ebaf6ab7f81e888577e89f723dbe26301537b57c5968a6679476", + "claims_sha256": "e97bfd502e5dcf591f53a95c7acfef169be6a554f267b7e2c11006db9286ec79", "sources": [ "trussell_wpi" ] }, "_totals": { - "rows": 5948, - "restated": 446, + "rows": 5952, + "restated": 450, "families": 12, "pinned_on": "2026-10-29" } diff --git a/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md b/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md index 8552c3b3e..ef940b7c4 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md +++ b/sources/harvest-uk-ab2026-2026-10-29/PROPOSED_KEYS.md @@ -4,23 +4,21 @@ Generated by `pipeline/ab2026_proposed_keys.py` from the staged families; don't A row whose scope matches no registered measure has `measure_key: null` and names one of these keys in `conditions.proposed_measure_key` (README rule 2). A key is a measure with no Table 6.1 line (`ab2026__`), a package (`ab2026__package_`) or an option (`ab2026_option__`). Each family's NOTES.md gives the key's scope as Table 6.1 lines or the producer's own words. Tranche 2 registers each key in `data/uk/ab2026_measures.json`, or folds it into a registered key, before the ingest. -85 proposed keys (15 measures, 57 packages, 13 options) on 958 rows. +86 proposed keys (15 measures, 58 packages, 13 options) on 973 rows. | proposed key | kind | rows | families | producers | |---|---|--:|---|---| | `ab2026__british_business_bank_defence_commitments_accelerated` | measure | 9 | uk_hmt_costings | hm_treasury | | `ab2026__departmental_spending_growth_2031_32` | measure | 13 | uk_obr_tables | obr_efo | | `ab2026__energy_debt_circuit_breaker` | measure | 3 | uk_hmt_redbook | hm_treasury | -| `ab2026__english_devolution_agreements` | measure | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__gas_standing_charge_moved_to_unit_rates` | measure | 34 | uk_ifs_ab2026, uk_rf_ab2026 | ifs, resolution_foundation | | `ab2026__gas_supplier_levy_funded_from_general_taxation` | measure | 1 | uk_rf_ab2026 | resolution_foundation | | `ab2026__minimum_wage_uprating_april_2027` | measure | 4 | uk_hmt_redbook, uk_rf_ab2026 | hm_treasury, resolution_foundation | | `ab2026__mod_counter_fraud_strategy` | measure | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_barnett_consequentials` | package | 16 | uk_devolved_govts, uk_hmt_redbook | hm_treasury, welsh_government | -| `ab2026__package_bbb_equity_community_lender_and_venture_debt` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_breathing_space_for_households` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_business_rates_rhl_sbrr_and_transitional_support` | package | 2 | uk_hmt_redbook | hm_treasury | -| `ab2026__package_carbon_price_support_and_cfd_levies` | package | 3 | uk_obr_tables | obr_efo | +| `ab2026__package_carbon_price_support_and_cfd_levies` | package | 2 | uk_obr_tables | obr_efo | | `ab2026__package_civil_service_transformation_programme` | package | 1 | uk_obr_tables | obr_efo | | `ab2026__package_compliance_measures_announced` | package | 1 | uk_obr_tables | obr_efo | | `ab2026__package_corporation_and_bank_tax_changes` | package | 7 | uk_obr_tables | obr_efo | @@ -29,26 +27,27 @@ A row whose scope matches no registered measure has `measure_key: null` and name | `ab2026__package_dwp_fraud_and_error_measures` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_fuel_duty_and_electricity_vat` | package | 11 | uk_obr_tables | obr_efo | | `ab2026__package_fuel_duty_freezes_since_2010_11` | package | 1 | uk_obr_tables | obr_efo | -| `ab2026__package_government_decisions_since_march_2026` | package | 374 | uk_hmt_redbook, uk_obr_tables, uk_rf_ab2026 | hm_treasury, obr_efo, resolution_foundation | +| `ab2026__package_government_decisions_since_march_2026` | package | 368 | uk_hmt_redbook, uk_obr_tables, uk_rf_ab2026 | hm_treasury, obr_efo, resolution_foundation | | `ab2026__package_hmrc_tax_gap_and_online_marketplace` | package | 7 | uk_hmt_redbook | hm_treasury | | `ab2026__package_hmt_distributional_budget_2026` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_hmt_distributional_since_ab2024` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_household_energy_payment_and_electricity_vat` | package | 5 | uk_rf_ab2026 | resolution_foundation | +| `ab2026__package_household_energy_vat_and_household_share_of_fuel_duty` | package | 3 | uk_rf_ab2026 | resolution_foundation | | `ab2026__package_ifs_announced_tax_changes` | package | 1 | uk_ifs_ab2026 | ifs | | `ab2026__package_ifs_new_giveaways` | package | 2 | uk_ifs_ab2026 | ifs | -| `ab2026__package_national_wealth_fund_and_british_business_bank` | package | 4 | uk_obr_tables | obr_efo | +| `ab2026__package_national_wealth_fund_and_british_business_bank` | package | 5 | uk_obr_tables | obr_efo | | `ab2026__package_non_scorecard_decisions` | package | 12 | uk_obr_tables | obr_efo | -| `ab2026__package_other_ame_spending_decisions` | package | 6 | uk_obr_tables | obr_efo | -| `ab2026__package_other_departmental_spending_changes` | package | 7 | uk_obr_tables | obr_efo | -| `ab2026__package_other_hmrc_compliance_and_debt` | package | 1 | uk_obr_tables | obr_efo | -| `ab2026__package_other_tax_measures` | package | 6 | uk_obr_tables | obr_efo | +| `ab2026__package_obr_other_ame_spending_decisions` | package | 6 | uk_obr_tables | obr_efo | +| `ab2026__package_obr_other_departmental_spending_changes` | package | 7 | uk_obr_tables | obr_efo | +| `ab2026__package_obr_other_hmrc_compliance_and_debt` | package | 1 | uk_obr_tables | obr_efo | +| `ab2026__package_obr_other_tax_measures` | package | 6 | uk_obr_tables | obr_efo | | `ab2026__package_parliament_fuel_duty_decisions` | package | 3 | uk_hmt_redbook | hm_treasury | -| `ab2026__package_parliament_regulatory_burden_reduction` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_parliament_tax_gap_measures` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_parliament_welfare_fraud_and_error_savings` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_planned_capital_spending_reductions` | package | 1 | uk_obr_tables | obr_efo | | `ab2026__package_property_and_savings_income_rates` | package | 33 | uk_hmt_redbook, uk_ifs_ab2026, uk_rf_ab2026, uk_tpa | hm_treasury, ifs, resolution_foundation, tax_policy_associates | | `ab2026__package_property_and_savings_income_rates_receipts` | package | 7 | uk_obr_tables | obr_efo | +| `ab2026__package_pubs_relief_35pc_total` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_rdel_decisions_excluding_household_energy_payment` | package | 3 | uk_rf_ab2026 | resolution_foundation | | `ab2026__package_regulation_tariff_and_sandbox_business_savings` | package | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__package_rf_ab26_permanent_personal_measures` | package | 19 | uk_rf_ab2026 | resolution_foundation | @@ -56,10 +55,11 @@ A row whose scope matches no registered measure has `measure_key: null` and name | `ab2026__package_rf_energy_bill_support_announced_at_budget` | package | 2 | uk_rf_ab2026 | resolution_foundation | | `ab2026__package_rf_other_tax_measures` | package | 6 | uk_rf_ab2026 | resolution_foundation | | `ab2026__package_rf_parliament_permanent_tax_and_benefit_measures` | package | 42 | uk_rf_ab2026 | resolution_foundation | -| `ab2026__package_spending_decisions_since_march_2026` | package | 7 | uk_obr_tables | obr_efo | -| `ab2026__package_tax_administration_compliance_and_debt_collection` | package | 13 | uk_obr_tables, uk_rf_ab2026 | obr_efo, resolution_foundation | -| `ab2026__package_tax_decisions_since_march_2026` | package | 19 | uk_obr_tables, uk_rf_ab2026 | obr_efo, resolution_foundation | -| `ab2026__package_temporary_business_rates_reliefs` | package | 1 | uk_obr_tables | obr_efo | +| `ab2026__package_spending_decisions_since_march_2026` | package | 13 | uk_obr_tables | obr_efo | +| `ab2026__package_tax_administration_compliance_and_debt_collection_receipts` | package | 13 | uk_obr_tables, uk_rf_ab2026 | obr_efo, resolution_foundation | +| `ab2026__package_tax_decisions_since_march_2026` | package | 20 | uk_obr_tables, uk_rf_ab2026 | obr_efo, resolution_foundation | +| `ab2026__package_temporary_business_rates_reliefs_receipts` | package | 1 | uk_obr_tables | obr_efo | +| `ab2026__package_total_financial_transactions` | package | 6 | uk_hmt_tables | hm_treasury | | `ab2026__package_tpa_business_rates` | package | 8 | uk_tpa | tax_policy_associates | | `ab2026__package_tpa_cgt_founders_relief_and_badr` | package | 1 | uk_tpa | tax_policy_associates | | `ab2026__package_tpa_energy_and_environment` | package | 6 | uk_tpa | tax_policy_associates | @@ -80,7 +80,8 @@ A row whose scope matches no registered measure has `measure_key: null` and name | `ab2026__rolls_royce_gas_turbine_programme` | measure | 1 | uk_hmt_redbook | hm_treasury | | `ab2026__scottish_air_passenger_duty_devolution` | measure | 18 | uk_obr_tables | obr_efo | | `ab2026__state_pension_triple_lock_adjustment` | measure | 4 | uk_hmt_redbook, uk_obr_tables | hm_treasury, obr_efo | -| `ab2026_option__cgt_reform_package` | option | 9 | uk_centax | centax | +| `ab2026__your_first_home_ft_reprioritisation` | measure | 6 | uk_hmt_tables | hm_treasury | +| `ab2026_option__cgt_reform_package` | option | 10 | uk_centax | centax | | `ab2026_option__childcare_100k_entitlement_taper` | option | 3 | uk_centax | centax | | `ab2026_option__childcare_100k_threshold_raised` | option | 2 | uk_centax | centax | | `ab2026_option__childcare_100k_threshold_removed` | option | 3 | uk_centax | centax | diff --git a/sources/harvest-uk-ab2026-2026-10-29/README.md b/sources/harvest-uk-ab2026-2026-10-29/README.md index 4e734ea42..227182455 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/README.md +++ b/sources/harvest-uk-ab2026-2026-10-29/README.md @@ -44,36 +44,53 @@ The AB2025 harvest README's rules 1–10 and its row contract apply, with the AB - a consequential for a devolved government. A measure's own parameters are its design: rates, thresholds, payment amounts, dates and durations. So are spending allocations to programmes, financing remits and contingent liabilities. They go in `reform_hint` or `conditions`, never as rows (the AB2025 red book family kept "£9.90" in `reform_hint` the same way). A worked example's inputs follow this rule too, and its computed outcome is a row. + + Lending and financing follow one rule across families. The harvest stages HM Treasury's Tables 6.1, 6.2 and 6.4 and the OBR's policy workbook whole, as the decisions' costings. Elsewhere, an amount a scheme will lend or spend is design, and an effect on borrowing, debt or the financing requirement is an estimate. So the OBR's paragraph 3.22 loan volumes for Your First Home are not staged, while its effect on debt is, and paragraph 3.20's rise in the Government's financing requirement is staged on `cash_requirement_change`. 9. **A component of one measure keeps that measure's key.** A printed part of one Table 6.1 measure carries that measure's key, with `conditions.component` naming the part and `conditions.row_type: measure_component`. A part can be a fiscal head, a sub-measure within a line (the tax-gap package's items, for example) or a nation. Components are never summed with the measure's total. Proposed package keys are for scopes that cross Table 6.1 lines, or have none. -10. **A producer's stated effect of the Budget's decisions on PSNB takes `decisions_effect_on_borrowing`.** That covers the total effect of Government decisions, its direct and indirect effects and any part of them the producer identifies, its splits into tax and spending and into scorecard and non-scorecard decisions, and figures that restate or recombine those effects without ambiguity. `models.py` defines the metric as PSNB and keeps it apart from `revenue_change`, a receipts line, and `ingest_obr_policy_effects.py` maps March 2026 Table B.1's totals, its direct effects and their tax and spending split to it. HM Treasury sets out in Table 6.1 the cost or yield of the decisions "which have a direct effect on Public Sector Net Borrowing", so Table 6.1's three totals, and the figures that restate them, take the metric. A receipts total or a tax-change estimate keeps `revenue_change` unless its evidence establishes PSNB scope, and a total key or a package label alone doesn't establish it. The OBR sums its receipts heads, from air passenger duty to visa fees, into its policy workbook's total of receipts measures, so that total keeps `revenue_change`. IFS doesn't tie its own estimate of the Budget's tax change at the forecast horizon to PSNB, so that estimate keeps `revenue_change` too. `ingest_uk_ab2025.py` maps AB2025's `package_tax_rise_size` rows to the same metric. Single measures, groups of measures and a total's lines by tax or spending head keep `revenue_change`, unless their producer states the effect on borrowing, as the OBR does for four policies in paragraphs 3.20, 3.21 and 3.43 and IFS does for the Budget's new giveaways. A row giving the effect on the current budget deficit proposes `fiscal_balance_change`, and its sign convention names that deficit. Each row keeps the sign its producer prints, and `conditions.sign_convention` says which way it runs. +10. **A producer's stated effect of the Budget's decisions on PSNB takes `decisions_effect_on_borrowing`.** That covers the total effect of Government decisions, its direct and indirect effects and any part of them the producer identifies, its splits into tax and spending and into scorecard and non-scorecard decisions, and figures that restate or recombine those effects without ambiguity. `models.py` defines the metric as PSNB and keeps it apart from `revenue_change`, a receipts line, and `ingest_obr_policy_effects.py` maps March 2026 Table B.1's totals, its direct effects and their tax and spending split to it. HM Treasury sets out in Table 6.1 the cost or yield of the decisions "which have a direct effect on Public Sector Net Borrowing", so Table 6.1's three totals, and the figures that restate them, take the metric. A receipts total or a tax-change estimate keeps `revenue_change` unless its evidence establishes PSNB scope, and a total key or a package label alone doesn't establish it. The OBR sums its receipts heads, from air passenger duty to visa fees, into its policy workbook's total of receipts measures, so that total keeps `revenue_change`. IFS doesn't tie its own estimate of the Budget's tax change at the forecast horizon to PSNB, so that estimate keeps `revenue_change` too. `ingest_uk_ab2025.py` maps AB2025's `package_tax_rise_size` rows to the same metric. Single measures, groups of measures and a total's lines by tax or spending head keep `revenue_change`, unless their producer states the effect on borrowing, as the OBR does for four policies in paragraphs 3.20, 3.21 and 3.43 and IFS does for the Budget's new giveaways. Text that restates a Table 3.1 group or a single measure as an effect on borrowing keeps that row's metric (the OBR's paragraph 3.11 "increase borrowing by £5 billion in 2026-27" restates Table 3.1's other departmental spending changes, which keep `revenue_change`), and Table 3.1's "A positive sign implies an increase in borrowing" is a sign convention, not a statement of PSNB scope. A total that adds the receipts effect of an option the Budget didn't adopt is not a restatement or recombination of the decisions' effects, so it keeps `revenue_change` unless its evidence establishes PSNB scope: Resolution Foundation's Figure 10 "Net effect if Fuel Duty frozen" adds a continued freeze's receipts effect, which its note derives from the OBR's constant-duty-rate fuel duty forecast, and the OBR states the freeze's effect on borrowing for 2029-30 only (paragraph 3.43). A row giving the effect on the current budget deficit (or, for an outturn, the primary deficit) proposes `fiscal_balance_change`, and its sign convention names that deficit. Each row keeps the sign its producer prints, and `conditions.sign_convention` says which way it runs. + +## Row contract additions + +`pipeline/ab2026_harvest_contract.py` checks the AB2025 row contract with these additions (review A, H-1 to H-6): + +- `value_kind` may also be `categorical` (a rating, such as the OBR's Table 3.3 uncertainty ratings). +- `value` is finite, and it is `value_raw`'s number at the scale `value_raw`'s word gives (bn, m, k, %), failing that at the factor `normalization` states, failing that at 1, 10³, 10⁶, 10⁹, 0.01 or 100. Its sign is `value_raw`'s unless `normalization` says the sign comes from the wording. +- A row that carries `conditions.component` is a part: `row_type` is `measure_component` (or the OBR workbook's `measure_total` marker), and the component isn't blank (rule 9). +- No more than two non-restated rows share the AB2025 ingest's claim identity, and two that do agree within 0.5%. +- A registered metric goes in `metric`, never `proposed_metric`. +- `conditions.recosting_of` appears only on `row_type: recosting` rows (Table 6.4). +- `conditions.fy` is one fiscal year, and a `fiscal_year` row without `fy` gives `conditions.horizon`. +- `conditions.geography` is a code: UK, GB, England, Wales, Scotland, Northern Ireland, England and Wales, UK_excl_scotland, "Scotland, Wales and Northern Ireland", "not stated", or a country a producer compares the UK with. Prose goes in `conditions.geography_note`. +- A restated row's `note` says whose figure it restates. +- A proposed key is `ab2026__` or `ab2026_option__` followed by lower-case words joined by single underscores, and NOTES scopes it as a whole token. ## Families staged -12 families, 5,948 rows (446 `restated`), as `COUNTS.json` pins them. +12 families, 5,952 rows (450 `restated`), as `COUNTS.json` pins them. | family | rows | primaries | producer slugs | |---|--:|--:|---| -| `uk_centax` | 110 | 5 | `centax` | +| `uk_centax` | 112 | 6 | `centax` | | `uk_devolved_govts` | 19 | 2 | `scottish_government`, `welsh_government` | | `uk_hmt_costings` | 597 | 1 | `hm_treasury` | -| `uk_hmt_redbook` | 432 | 5 | `hm_treasury` | -| `uk_hmt_tables` | 762 | 2 | `hm_treasury` | +| `uk_hmt_redbook` | 425 | 6 | `hm_treasury` | +| `uk_hmt_tables` | 768 | 2 | `hm_treasury` | | `uk_ifs_ab2026` | 144 | 14 | `ifs` | | `uk_ippr` | 3 | 3 | `ippr` | -| `uk_obr_tables` | 2,797 | 3 | `obr_efo` | -| `uk_rf_ab2026` | 766 | 7 | `resolution_foundation` | +| `uk_obr_tables` | 2,800 | 4 | `obr_efo` | +| `uk_rf_ab2026` | 766 | 8 | `resolution_foundation` | | `uk_thinktank_misc` | 12 | 5 | `citizens_advice`, `cps`, `institute_for_government`, `learning_and_work_institute` | | `uk_tpa` | 305 | 10 | `tax_policy_associates` | | `uk_trussell_wpi` | 1 | 1 | `trussell_wpi` | -Each family's `stage.py`, which writes its claims file, is kept on branch `uk/ab2026-harvest-staging` under `harvest-staging/ab2026/lanes///`, with the lane briefs and the five reviews; this branch carries only each family's three files, as AB2025 does. On 3 November 2026 we applied the review fixes, all four sections of the IFS addendum and the cross-family key harmonisation, and each family's NOTES.md records them. +Each family's `stage.py`, which writes its claims file, is kept on branch `uk/ab2026-harvest-staging` under `harvest-staging/ab2026/lanes///`, with the lane briefs and the five reviews; this branch carries only each family's three files, as AB2025 does. On 3 November 2026 we applied the review fixes, all four sections of the IFS addendum and the cross-family key harmonisation, and each family's NOTES.md records them. On 4 November we applied reviews A and B of 4f424b9 (the registry, the HM Treasury families and the harness; the OBR, Resolution Foundation and think-tank families), and each NOTES.md records those too. **Families with no rows** are not carried here, as AB2025 has no empty family. Their working files stay on the staging branch: - `uk_cpag`: both CPAG pages (its Budget response and a pre-Budget call) returned Cloudflare's challenge (HTTP 403) to every fetch on 29 October. - `uk_smf`: the Social Market Foundation page returned a challenge to every fetch on 29 October. - `uk_niesr`: NIESR's Budget-day reaction ("Breathing Space, Not a Strategy") was fetched, but it holds macro and fiscal-stance commentary for the #55 lane and restated OBR figures, which the AB2025 NIESR family did not stage either. -- `uk_fai_sfc`: the Fraser of Allander Institute's reaction refused the fetch (HTTP 403), and the Scottish Fiscal Commission published nothing in scope on 28 or 29 October. The Scottish Government's five rows, first staged here, moved to `uk_devolved_govts` on 3 November. +- `uk_fai_sfc`: the Fraser of Allander Institute's reaction refused the fetch (HTTP 403), and the Scottish Fiscal Commission published nothing in scope on 28 or 29 October. The reaction's feed text holds only restated OBR figures. A second FAI post, "Rising or falling? Why both claims about Scotland's funding can be true" (29 October), was found on 4 November through FAI's RSS feed and sits behind the same 403. Its table of Scotland's Barnett consequentials is not staged: FAI says the split emerged late on 28 October, so it probably restates HM Treasury's allocations, which `uk_hmt_redbook` rows 258–261 stage. The Scottish Government's five rows, first staged here, moved to `uk_devolved_govts` on 3 November. ## Producer slugs and families @@ -85,9 +102,10 @@ Each family's `stage.py`, which writes its claims file, is kept on branch `uk/ab | obr_efo | OBR: Economic and fiscal outlook, October 2026, and its detailed tables | uk_obr_tables | | resolution_foundation | Resolution Foundation | uk_rf_ab2026 | | ifs | Institute for Fiscal Studies | uk_ifs_ab2026 | -| tax_policy_associates, centax, cpag, ippr, niesr, fraser_of_allander, scottish_fiscal_commission, cps, institute_for_government, age_uk, … | other producers, each in the AB2025 family of the same name where one exists | uk_tpa, uk_centax, uk_cpag, uk_ippr, uk_niesr, uk_fai_sfc, uk_thinktank_misc, … | +| tax_policy_associates, centax, ippr, cps, institute_for_government, learning_and_work_institute, citizens_advice | other producers, each in the AB2025 family of the same name where one exists | uk_tpa, uk_centax, uk_ippr, uk_thinktank_misc | +| cpag, smf, niesr, fraser_of_allander, scottish_fiscal_commission | producers with no rows (see "Families with no rows"); their working files are on the staging branch, not here | none | | scottish_government, welsh_government | the devolved governments' responses | uk_devolved_govts | | trussell_wpi | Trussell, as in AB2025 | uk_trussell_wpi | -| commercial firms and platforms | worked examples (mode 3, #63) | uk_cases_commercial | +| commercial firms and platforms | worked examples (mode 3, #63); none was staged for AB2026 | none (AB2025: uk_cases_commercial) | Macro calls by banks and City forecasters belong to the #55 lane and are not staged here (AB2025 rule 10). From bbf4f4ba9e631ab139dec8380f5fe31f69414d1a Mon Sep 17 00:00:00 2001 From: Claude Date: Wed, 4 Nov 2026 18:41:55 +0000 Subject: [PATCH 34/52] signs.py: read a conflict only where one stretch of text reads both ways The H-4 guard read any convention matching both lists as unknown, which changed 77 committed feed rows (an OBR AB2025 convention that quotes the Box 3.3 note on higher taxes after its own 'positive = increase in borrowing'). A conflict is now two overlapping fragments ('positive = cost saving'); otherwise the opposite list still decides first. A fresh export equals data/populations.json, and the harvest's 76 readings are unchanged. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_01LwQq6JvunwtJCAckL6L4aM --- scorecard_db/signs.py | 24 +++++++++++++++++------- 1 file changed, 17 insertions(+), 7 deletions(-) diff --git a/scorecard_db/signs.py b/scorecard_db/signs.py index 1432d0856..d2b1a4a54 100644 --- a/scorecard_db/signs.py +++ b/scorecard_db/signs.py @@ -103,8 +103,8 @@ "reduction in public spending", "reduction in welfare spending", "positive = saving to the exchequer", - # a cost saving is a yield, though "positive = cost" also matches: - # the two lists conflict, so such a phrase reads unknown + # a cost saving is a yield, though "positive = cost" also matches + # the same words: the reading conflicts, so it is unknown "cost saving", "positive = saving against the current budget rule", "positive = fiscal tightening", @@ -244,19 +244,29 @@ def classify(row: dict) -> tuple[str, str]: return UNKNOWN, f"stated:{family}:not_this_quantity" positive = _NEGATIVE_LEG.sub(" ", text) - def hits(phrases, where): + def spans(phrases): # a phrase that names the negative side itself ("a negative sign # implies a loss ...") is read where it stands - return [p for p in phrases if p in (text if "negative" in p else where)] + out = [] + for p in phrases: + where = text if "negative" in p else positive + i = where.find(p) + while i >= 0: + out.append((i, i + len(p), where is text)) + i = where.find(p, i + 1) + return out - opp, sam = hits(opposite, positive), hits(same, positive) - if opp and sam: + opp, sam = spans(opposite), spans(same) + # one stretch of text read both ways ("positive = cost saving ...") + if any( + a0 < b1 and b0 < a1 and wa == wb for a0, a1, wa in opp for b0, b1, wb in sam + ): return UNKNOWN, f"stated:{family}:conflicting" if opp: return OPPOSITE, f"stated:{family}:opposite" if sam: return SAME, f"stated:{family}:same" - if hits(opposite, text) or hits(same, text): + if any(p in text for p in opposite + same): return UNKNOWN, f"stated:{family}:negative_leg" return UNKNOWN, f"stated:{family}:unrecognised" # US producers publish signed changes, after minus before, which is From 15c077368706a86d11aac98a4637bf9a6244fd92 Mon Sep 17 00:00:00 2001 From: Claude Date: Wed, 4 Nov 2026 18:43:11 +0000 Subject: [PATCH 35/52] tests: reach the contract's manifest branches (review A H-7) A bad sha256, a manifest line without retrieved, a primary listed twice, a row whose primary isn't in the manifest and a missing NOTES.md each give their problem. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_01LwQq6JvunwtJCAckL6L4aM --- tests/test_ab2026_harvest_contract.py | 37 +++++++++++++++++++++++++++ 1 file changed, 37 insertions(+) diff --git a/tests/test_ab2026_harvest_contract.py b/tests/test_ab2026_harvest_contract.py index e1d37be09..f90b523ae 100644 --- a/tests/test_ab2026_harvest_contract.py +++ b/tests/test_ab2026_harvest_contract.py @@ -307,3 +307,40 @@ def test_a_pair_that_disagrees_is_a_problem(tmp_path): # a value off by 1% (value_raw unchanged) is also a value_raw mismatch found = family_problems(_family_with_copies(tmp_path, 1, factor=1.01)) assert any("differ by over 0.5%" in p for p in found), found + + +def _family(tmp_path): + fam = tmp_path / "uk_test_family" + shutil.copytree(FAMILY, fam) + return fam + + +def _rewrite_manifest(fam, edit): + lines = [ + json.loads(x) for x in (fam / "manifest.jsonl").read_text().splitlines() if x + ] + lines = edit(lines) + (fam / "manifest.jsonl").write_text("".join(json.dumps(m) + "\n" for m in lines)) + + +@pytest.mark.parametrize( + ("edit", "expected"), + [ + (lambda m: [{**m[0], "sha256": "abc"}, *m[1:]], "sha256 is not a sha256"), + (lambda m: [{**m[0], "retrieved": ""}, *m[1:]], "needs url, title"), + (lambda m: [*m, m[0]], "lists a primary twice"), + (lambda m: m[1:], "primary_sha256 is not in manifest.jsonl"), + ], +) +def test_each_manifest_rule_broken_is_a_problem(tmp_path, edit, expected): + """review A H-7: the manifest checks, reached by a test.""" + fam = _family(tmp_path) + _rewrite_manifest(fam, edit) + found = family_problems(fam) + assert any(expected in p for p in found), found[:5] + + +def test_a_family_missing_a_file_is_a_problem(tmp_path): + fam = _family(tmp_path) + (fam / "NOTES.md").unlink() + assert family_problems(fam) == ["uk_test_family: missing ['NOTES.md']"] From 93d2407394339dab41aab6f37387ae2c9bc5d3d0 Mon Sep 17 00:00:00 2001 From: Max Ghenis Date: Wed, 4 Nov 2026 14:00:13 -0500 Subject: [PATCH 36/52] uk_hmt_redbook: finish RB-18 and RB-19 from the pinned primaries Brief 82's hand-off (comment 6639062906) left three red book edits open because each needed a primary. The staging stager now applies them (uk/ab2026-harvest-staging fdf31ccb), and this commit carries its output: - RB-18: row 192 (4f424b9's 194) quotes footnote 84's definition of "fuel stressed" in its note (red book 19b337ac...). - RB-19: rows 359 and 361 (4f424b9's 365 and 368) quote their whole sentences (Impact on households 36cd5afb...). Three fields change. Row counts and values stay the same; COUNTS.json takes the family's new claims_sha256. Co-Authored-By: Claude Opus 5.5 --- .../harvest-uk-ab2026-2026-10-29/COUNTS.json | 2 +- .../uk_hmt_redbook/NOTES.md | 6 +++--- .../uk_hmt_redbook/claims_staged.jsonl.gz | Bin 36306 -> 36551 bytes 3 files changed, 4 insertions(+), 4 deletions(-) diff --git a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json index 4ac536a8b..622f9666b 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json +++ b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json @@ -27,7 +27,7 @@ "uk_hmt_redbook": { "rows": 425, "primaries": 6, - "claims_sha256": "82e3bcdd48ed7295b2d2c293273e7c5ae9711f4646d4b83d06144011dafc0342", + "claims_sha256": "04e5a88f9d20f8761049040460eef001880f471f74018e9d16c23e9b8c4a485d", "sources": [ "hm_treasury" ] diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/NOTES.md index 38b6a153a..f914716e7 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/NOTES.md @@ -330,7 +330,7 @@ Table 1.3's total effect of policy decisions and its tax and spending parts (18 ## Review A fixes, 4 November 2026 -Review A (issue comment 6635870934 on #231, at 4f424b9) approved this family with fixes. `stage.py` applies each as an explicit, commented step after the rows, indexed by 4f424b9's row numbers, and drops rows only at the end. No primary was re-opened for this round; every new quote and value comes from the review or the row. 432 rows → 425; `own` 263 → 257, `restated` 169 → 168; 6 hashed primaries. +Review A (issue comment 6635870934 on #231, at 4f424b9) approved this family with fixes. `stage.py` applies each as an explicit, commented step after the rows, indexed by 4f424b9's row numbers, and drops rows only at the end. No primary was re-opened for this round; every new quote and value comes from the review or the row. The coordinator then finished RB-18 and RB-19 from two pinned primaries (below). 432 rows → 425; `own` 263 → 257, `restated` 169 → 168; 6 hashed primaries. - **RB-1:** paragraph 5.17's two totals "over the forecast" (rows 265 and 266) sit at 2031-32 with the forecast's horizon; row 265 is `cumulative`, row 266 keeps `range_low`; row 267 is `point_in_time`. - **RB-2:** rows 191 and 200 dropped (the £2.6 billion's third and fourth printings); row 179 dated 2026-27. The same rule drops rows 188 and 201, which repeat "up to 20 million" in paragraph 3.3 and Box 3.A: once RB-11 put rows 180 and 201 in England, four rows shared one claim identity (the coordinator's new duplicate-claim check, H-3, rejects more than two). @@ -349,8 +349,8 @@ Review A (issue comment 6635870934 on #231, at 4f424b9) approved this family wit - **RB-15:** repeats share a dating (rows 9, 10, 14, 196, 205, 248–251, 271, 273), and rows 0, 19, 215 and 244 take the multi-year convention. - **RB-16:** rows 143–166's `reform_hint` is "Total effect of policy decisions". - **RB-17:** rows 185, 216, 226 and 237 have conventions `signs.py` reads; row 185 keeps its key, with the three scopes in its bullet; row 182 is 0.84 `share`. -- **RB-18:** rows 33 and 37 are HM Treasury's own arithmetic; rows 6, 7, 8, 10, 11, 12 and 14 take their repeats' source models (row 5's and row 9's repeats are already `arithmetic`; row 13 has no repeat and keeps its label); row 222 is a worked example (`arithmetic`, with its subgroup); row 183 `decile_10`; row 184 dated 2027-28, medium confidence, `hmt_published_estimate_method_unspecified`; row 168 loses `presentation`; row 267 `hmt_published_estimate_method_unspecified`. **Not done:** row 194's note should give footnote 84's "fuel stressed" definition, which needs the primary. -- **RB-19:** rows 15, 29, 41 and 42 quote only their sentence; rows 103, 104 and 167 lose their leading paragraph numbers; locators of rows 172, 190, 227, 248 and 249 corrected. **Not done:** rows 365 and 368 still quote fragments, since the whole sentences need the primary (row 366 is dropped). +- **RB-18:** rows 33 and 37 are HM Treasury's own arithmetic; rows 6, 7, 8, 10, 11, 12 and 14 take their repeats' source models (row 5's and row 9's repeats are already `arithmetic`; row 13 has no repeat and keeps its label); row 222 is a worked example (`arithmetic`, with its subgroup); row 183 `decile_10`; row 184 dated 2027-28, medium confidence, `hmt_published_estimate_method_unspecified`; row 168 loses `presentation`; row 267 `hmt_published_estimate_method_unspecified`. Row 194's note gives footnote 84's "fuel stressed" definition, which the coordinator read from the red book (19b337ac…). +- **RB-19:** rows 15, 29, 41 and 42 quote only their sentence; rows 103, 104 and 167 lose their leading paragraph numbers; locators of rows 172, 190, 227, 248 and 249 corrected. Rows 365 and 368 quote their whole sentences, which the coordinator read from Impact on households (36cd5afb…); row 365's sentence includes the lead-in that dates para 1.10's bullets to 2028-29 (row 366 is dropped). - **RB-20:** (a) the unverifiable "234 proposed keys" figure is removed; (b) the spending test reads "the per-year amount equals the line's", which rows 224 and 276 pass, so they stay; (c) five more figures listed under "Other material not staged". - **RB-21:** this lane folder's gzip and NOTES now match #231's; the 40 proposed-metric rows set `metric: null` explicitly. Table 6.3's duplication of Table 6.1 lines 45 and 46 (`uk_hmt_tables`) is left to tranche 2's cross-family dedupe. - **K-7(b):** row 258 is the Barnett `total`, and rows 258–261 carry the sign convention. diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_redbook/claims_staged.jsonl.gz index f488eb74c1a40768c121f709aa2772a330ad8ade..6a9d6e2626604bc44d6402504ad2f59f4f3b7ae3 100644 GIT binary patch literal 36551 zcmV*EKx@AriwFP!00002|Lnc%j@w9Oq$4&Mk*opSf?{apuAQ zvT(xL*z@_&3c_d1c1QdRAFN5OL{Kn*;VL0@t zc{rcC=3qX$6F7d4CB9D z&!1Y3{r70XzI|l_+i~D8_CNoZHF%#@|BZeB;@u3W=E4)a+O<2aUJ%SadCrtUso=@pp?%Uo7K7;8n@NeIn zc>y1p(VILP@j+Qyv5a;HFKYA$&1Tmc@=yKvjJxedyWMEiN6l7u*z7fo=AhH;)kppMxZdae?x5d{MjuQ#jpB!&*hin% zbAjK5cIeQsUy!gd96h`F;U()D9U9Wyckt^;7|w#H-+c$WhlfE;-V)ZK7I6QwJ><1J z?{jVb={uZn>${G{jryS1ZguLzMzhr!cL%)NYw_=1*PESAuixvKKm7F3d=X2y`TED_ z|6!VjS(}ZfrI37_*Z~6>bOVxyU;o13?_kDZx4L(13gW>)#>6=A6Wn~bl2Tys znqS6a7$b@jlP~(72Q%2|Hi)Pl!cu@F-bSK^cae{}2N8xHvLFdnqB&;YzBXEsc!OjK zVYFSwzl6TU?ofK&DePw&2~#q~hTasURDc9N;RA}m9Z9dx;qg-wCN+$ARp8BiA;^=d z873A^d;!Al090&}UHLGIF?StcPVBqMGF;XVf`Nv{ zzzrD4lb`}i1|T)(0piG%!=Q;Ng|6Yx9r%K<&tii)iuF_W=O6#g@FuW1B;?`G7$IxO z1`9;Ab|gFiWKviwY=F4LnQsRKiyr*aFw(zl*8lwDAI(M?FqR?ZCyrq4EfB1Iwg}d& zSDTIVL$HQMFpXdu!7>rdFp48sJb&kdVB%LOH25Ri{S2c-S=0)ehPR#JO`%gPR;%!2 zw3gusK?l}uHV^5{NoPg))*_IDtKls9H&{9k*5Q6)!v_*JC62)!vC&LIwi19pqaHl~ z&ZyQ%>>L2p#Hd0%O~12z$F?L8%M)pISf2U(v+d1;1YH05$A1NE=D-pBIuUYt;K zQ>rps(i90yU;)Vxb5RgI-}ml8e%4we0cgCb?J7k0{?)q@ZIs)a6VXs9kZ4D&Tx0u- zh22cek%rY{YmKnRgg;71SupYLg)J}(p%{SKe8^)-Uo|C@of|+m0k|XY-o3Z{Q37(G zCT=md9EViZMy>Z4!4@y(Gof>hcDoQ)bnqm>u%u-3;2Mp zbG>`~%ILsXM)$T1tX(J8CkD8%L5Dnnqj<2c^>TtOu1OZi!jkS$1I70R=)n5`&L><% zzl`tr6u}CFaA?igO}LoZc&wQKsq%^EjBb;l@f?2X;O6V*$JrFDe*c1|-@bEkM)5-V z{s3_{toDQ@K3yI2xAVd zFL)#P{(1FP1>}~d97jH%2?sNq%Av>i%3!ZY`4{U^A(oRMLDOx*bZ_u~Sw1`v;6*@p znG{FlWZJHYAFnx8m;fne^!7=*Yv(;$B!upmq*u`Kg|P^0f)+M27m0=RmcITnobcc0 zfaW(ztOqy;c(3Ap_~>>M0Yx)!Dv1%B#)J)Il7yQUa?vQwK|a1EtiT5T$z#}gGZ)BX zyF+I_;^?LR>^YzLsG2a00_MZv@EN_Y*viq)4lN(%Hc&gXrs+%5m!_`^pfA8=J|?b6 z9lRpWnoPxaFk}-CYC{gg2d#E=WV2YOxK%6`lVg&XLb;?YZF%~V^0s^`%9Oo5P?H63 zkDJ=0Es{x1((_j-p4dCDm=mtwVnuFxn@RsO1gp)K1 zh6~Vd25>wRi(QDIvRylc=}7r>P|!%Lng=!xiwB4C<1{r}t!1k>2(<*U9=Ta6OYWNl zg}staZT!ItA#?0)By;R#TISdXlDP^6h@%`g>J@+wmPgtJbI>M(u~m)s;}q|WU@N1{ zR?r(S0Dg(L#6ezg*1#HmidYAEL>fJ^fPbr3yg&|+*c*9#3LGWGGM^>VCz5o82eJMr z?UK&+orqHbh4~L{e>jJk9DoFT;{L-%w7~i-Lq<@k0x9Kwu*?q@?1jX)IRc(?x@NNB z{mBdQ?oZgr;{n<50t>@jIP40sHRPy4*NieOU|$XA)44-3TQKRTu36v&f~4NMq3@A{ z-eNAlS33PGbe+Exat)YBaxG34!%q%Za4yn%q+Fy0Ha>nd>(Vn{&%)jZFD{6)A}7cc z2Oaayvfaaa9mInRyBnH?-3`riayNA4uFYer8P9!!9P$vum?vH^vqOtQjOk>`ZA3v9 z>yyayC`7IoeaO&_y;*;jza)n@=*k1%H~ooYh7+F$bPj@V?0}Dspg;FcL4R&(0f1{^kjHzWK|T_d_#`)0mU9$x3ehzPnn^mD z6j-LD6tH&adx0tekOjG!&nbqxt73m9uGmK56x4IImJ8_IyS4lQDa=e|*cxrG$40K< zSHtgR8Lf@=&PvC-UqRk@rjM=v?xf!4f#jdP6znF;H9cbqs1e`HrRC!xJNV2RenNvA zyb_s#JyrYJAtN&20n2@{R;_KY8{s(r_O;WBcJA#n%>P8N zgPX*pQ_R$Ebh2~oqE$f+N-oV|c&DRzCvRm2G7@1JP-N>6HX<_2NXoE5&H7C16B)ZJud%Tc&cj%`;-L@fn#tOFJx)S(SZ0D&|a+0xNKhFX5l zh`*{F#MhewKVphrfCeJr3LaW6yvdk53l?IbF1#Y-!-?xT-kl9EHMW3g{S6+42PU8x z6T;@jIeb(vw9YhUJiqAts^ z-fvsfVrgD5N^28Z!@`Dh-`&C%->$Aby%b2Fj`QgETpai35A0p4fCq{)tRyHnA7DB8 zbijQHo@37o@jBM56^7k>wJ^#qhx-nXA&puiRRvdvz<)TR1zDiDP(qIds;>(EzvhTCTAuBnqH|S(f`T13*-Q=en zau?1SULuN|QaP|Bivx2a>z&Ljk|&r6tjdGsJbWyGo-8P`5oE<(4u-CUD+i^cWR8QD zofKU5G+aJZIf%iggXzIw4T=PTw{WO<51g9rL0&;wa{yR4tJm60R-wTf-=zBl48?%^ z5?n9nf|Fng^Ol7oLv_2e7=!s$7ub_4M z&LV2h5w&uRin$B!2@YiCE9R>n&V8RehGIvJ_!w)Oup607xk*rNpD~y8bGmEtf^|17 zl$y2|HFv*=ysFU3l^j?O`sK^kXxF26ZlToRUIg58PwP`wZ5XwB3E*D{UM`0ls>K~T zV#ZI59e$w@=6Sq`niVuF9E%km6h_SkMcClM>3quSwSEa4kPAdE2mAX4fgdON#9;p& zfS3fXfR*g#&#;C3G7LQx>zS7h#7vmU*sWP&)Y@QD5L`2tTmxzmK+T*&jP0qlOF+@J zd-8ISvthIgVkgb*iGk;9N|ZncgM8q@ROGz5#fcEP5g*!=Uue=Qn^RonXh2%$zIL5^ zS(r5=6k&wr*@G3%vZ`c4Mt){kfAvL}*^3%dPSK3Atl*xs8tnz4)}yyJQY02ZgJmIm zL}KPJqM#J5d3jGx&fGY z6~y-vumQRbwoR`awN{^P>EzgE9lNnNbOOT#CNN}Zf(dC% zVQH^w^z)WCOUS=y%!<1wapRcSj-etZ2O@T{9boOG#Hio=2a~T{sAfQ4BezCwjoimX z?k$3eTW4FDpjodr8|SCsHZ)9XnA9+N)NHsdiUt^aDH^bSg)6`q53W^U@cpZIhpc;r zwS~$md6iuxaH}F$?Z%l^xiVTBymNy0j|g%Ex;IhHDl{8|oREVknA z!mU!`;PP-;MsJh2nNj@8iJ%y;v{h?G(28m7AS$6v9oyhb+5{^Z<@imb)_W{9P*iN2 zMy=gsAMd>?Y1gEX<0Yw3862C?X+c*Wi6(Rii9EdLoa$<|kQUlB4m4rmU;T=Rpq<&S z%c)_UFx6qf0P73=bi_f*sWOD+Ff1HSHN)vJfyHA1QECg~I&>R{@2vo7no_ddi->w45O}1VbT&l^rq9 z{qv9i_sMPXP?oj1ByfEIvN94)2I=KA<7mdwjI)Kac+FngvYTCbj1!F{wpw+T{x2u} z#By+C%T+p7;WV8lsVx$&hHE4f886*+vfs=_O*GrOpEro5qMF7E(MbKS)_yDqqaAwz z@qW@7z2>%Zd#1hA?KaDZXR?D7GbsWir}q?lf0tSJdQ~*f#0AmFdvsV?$LqmxJRR(8 zQd$S!=kJba=K*bv)*7u(4Xsb!tPj(kJ7K;3SS!^Ukn?NN8jv+0Yd|h+YH#N@wYLvv zYR`+q4_e~owN)ttDU(#W_V)*HIRtRMl=!C2%LELY&suKmR|zMA<0c-4Y4L_nd&;z|crwD{N#fs-ahf|B3AP6&iiq!0eX(=3j=+ ze6&&_{0gk|86?%uo5}^UcjPmJ61)WepMfQQdAn#6yk?jqoFT;e+p@J^Iu|TUOk?kF zFOhfy=;`t=iF`8Ako%U|D`x<5`>x3uGurAj(M?f`BC{zVeSpC1Md`eQ1WmF(|M(}+ zf(=xYN&z0*K0HO;Y&8+MS~uuY>N>uN-Od%IkR>~}fr!PqZ%g;AwR_?ev>2i|JI;bx zqrc4q62|5I{_^>|SGS2F1LoPY08dw{963&t@Ko z7e+>>9&6jpZ2;^(K7m3)yHL!7VY5-pY{5g+{%U2s{pk(+>W;zE%SL%RF>YF49$X zBLiKhUg&t#J9cvSs}wI8nwQq{((~aE4&{hf65~KPQiXy70nEStMHuB|=&5j{V7Wp7 zdl#Ug3TQG-+rpJ4;Cny=l5BiRY!O57k9jQ$Xpgqb5i+3d>DavI60n6k*QqX&Bz~EpA&oV z^_Ss<|2~J=y-88R2YeQLC!Q5)DS1vGAQSN5dGZvyRROzKzr9-;#ji6w_M**s9d|Tp z>}L;V6+57yeuyIR9sECx-cr5b19k@!;UeltXbE|B1adUuOv=Jq20~PvPcuz)WtLv* zr#PjlIb}Vkkl>R%!%}{vEx~?-ErII{*aUMXm1>f1zDBN*AyP&5EFmm|MI<5u7b@y7 z@$GB7zWLX!dRD!roFgvQEo61aEz`~DerBq*9+~Pfs4z_c;^r<)#Lynw$p`PoLj;kg z4KM{CUsz@5BT`dKQf z0v}D^RUzNn#&#Ele2tSwzGxjZ@)blrnmO9#ol=l5T8WE5zD|DR>s%S~m5|nP`p9RT zKJuw`(8yO5`EcfFmzQJWi&o+ykguH|`HpGi+fYZyqqoC&z&vWzeX&)X1DN(VxjgB?}UZIO>EQ}+Dp8Q zg^k=;*tkM0ylO}BX=0)F8SjLJ(sA6V^|ar3hkmEmh=teeIzCY>)L!JBuu!;?OT)s; z@g$!M7JkoN?!%lU3cl|j5Hs*^rI5u-SrI`Lsh}ZX%ayp0LhG4a&#Dk12J_|1iWl=h zW{ICPsy<`G2Vb(r)I=~bMVNw5W%Os*X<)=GQPDr~@>1xdhSx&q#Uwprscpx6oNv{t zV8SgdG_Xx&iIc(mhvZp1%pU~(ho&&?HE&`GLNU_1GQsRlrC0vI9N}c1h^wj`iqj(AS`UuAqNnv3+?!UkA19 z4*D{(#V9HJ<76XS3W9zsH|V!C=xfkFThMQuAL#4oo!voSgz%IV+I_MiJOx4jOrpBe zJx6urhl{D`zqGPCY~|rt#ijncsT%%uUE5(fKCzxVKCyl+2&aqp9~t3vj7DLEQ|0;% zU9sUj>sB8W;dGt;BO{y+NGObOBL1LSFRKyLpWG8RWDBzA(ti zI{c+|o4cSo{Q29>RU20Wa=HNjkwH#7oC|}TD!Jb%tMK#%mE3RSFFf5iZsF-ef}F0k ze`Ju;4%WgTC#&l>N(!C4r0V*O+yRs{eMjsovKZn{bDRaUMt_?Jv`e(F^-{a1KfK!7 z|N3ypk%wn7>65MNzUr$Ui)L<*Me~}`QTu665FNGewiI*}{@kkWzkbHenT`&S+Rb}{ z0I6NSr2u5)4(@4p@Crco@&jb=ngLR~h))n8wcEH9fQ($p9qme9F+g_m8z4JZ43OHr ze1ZU}UCyNdWbB64o!>P;YS;8Rw1LkFklI~c3P5TX_W8T8S6erX0rK3T4KwA~*1&e) zN*@GV-V zWX@#L72a#=-t)^qi*u3_XleSc9BB9N5kGIB?VTggYL9wpKr6iJfD}FLRbK(xUVfnM zT|3b3-?M(+K-)h@pw*uF(tuWY=gR}y%kj=@pgl1U{aFKTjuO2m2(;S6UmDN~FFznf zLwosGfVPnzXd71!vq1TK%RgIReuSFZb-yiY3Qv7`y5{A0 z>b0(^9rG9Im{+K&{qWnOrgXda+qJ%SyRX4Z zZ2OM(p|7B3CqHU-P7XC0|H58a?kAkW*Y2HTxg+gX*KYN-K$;;h+S9%*wyw@)%K2{8 zTH5)(1h+6JdEpkP@9N=J`{U2rAHNQ6wTr$)xRqXeoyETjxA~Jx`=^#%dK9?Tj{LKB z1uw}8)p z+h%^aZC*Xx>d=An4jotnw>p}jM7UJ}2Cb56uU=AsK`VFdRhqu5hg%(UaMm#g>)=+0 zA(RNWGBTl6tLw;w6}YYEhuiu|!R@{0hm)hlAt;U*DaegEdePmD%VVKzXnk6S;D zVj7Qgg)YF_n$g-G9F*zFI$B{12&S1!#wK){WtkH$CpH1+Bq!j~^j#U?ZiqzCfV;2Y zg;N1s9j~x8;EL!3K#Z1-PFMllR(`;3T^rzTh(*wVyRYztQvqBZv9LAZiui<1X(g~P zCqAK*zXW#Y)Jk9<0B|=%BWS?gR{+DQ0IrT%*cxy}ghIzCt6kycL@3~#6~NVT z3tI!Oh*D_xOUf|2peO~LlbnD{(|0_;{hqtrhgpetCcE>!AlP5=nJkngE^cfpI_>wiICW476 zV(=5Zp%1(ynFl32@$ypU6#z<^7nAgirFI?jalT!=g%z?1w?>F5>rfuSWj*$CFfqvusC1WHSb>q^5BrMmRTTV`>r}@Z<^r57_I-j2*#V z#Ah_>&5}w=o@{_dL9lP;E+;AGPJ_J$`?Cc5-uZyNjtbck>_u!yquwg3R?w*ihZF?+ zR&Fbfm^%&j8tl&!?EB{f_BtYEN3a)hDF|4t%MMH_4EC)8VBgYUufhH-!Txm0Cp`%E zIvQq2u$M702v`>$6jK=N>jl8RuEAb|{dt0Y<9xtgN8;=V_97k!?amk8(4?@47Bw{np6`42u(e8qrgxTafjl<*OVFetXC36cP;0<$-hKG_#Vs3KjuS9k zB-+PHVk>OydsART<1ub=3{ACNdZR!8_=nMDp$GHcdn`U|HsHf1@Ph(~=54J6v|1-R z+d$WQ5RHN7ePS;s)(p6uHTDrBn7?iuPV9w%(%M+-vxr&Nprk?RvO%el-{S?7cLJc4 z6Mn?6{JEG*7~#Nn>~N7Q(fMI$QuoIrn(ukjM_D-a?JI+aI`9|!pa07myw9rt#=d{? zu8dq_4MoR*qK^PK!?}+_%Y?v*@2n0gOLUe(JnsOc@a!HebkU@r0$?olN!YsHA z?g9j?Z(kd2B`_6+l_B9O7w$g9S>gc#*tf5S5ebz5vN}Er4*mBNdpLo+Iu09fdSGbz zI6^W)SRsAn@-HD1?^*`jW$c0zLAtxaEQNCIy86GIXqQ}#i$cA%*~X`hdJTEmFH`yz5#{A(n#D4}>zNjNV4 zmHB)$ABIwX@!K8FzxzShqnPNT(^mqF5sm-+-2-q33mP+Vqv}^6t zN$6@mMFe#3IB`lOV=RCT#gh_n6AS)r=?WqUMnS9;(|C~}xv|FWRfA;jj6hPSL~R9< zvW;%1EQ~B859%Z^vfn8Rku-@HiI5wM(q1(}_Rk0*bso`H2r06MdYzI&3`guSlJ38Wu8jAe@ba8CxMXdeo-BgCh>R(`7`Xw>45v%U$Z+%K4T=5BEB(T z*;C6#Jeb)bXTJZCtXc{jS zZ?DVsb6C9XofqC}|IxO1D{JXBjIu1G7vn>MIjI-+vl|zz-(Hs^>9Ba)KQFx1-l%Qy zRyd^k{jyy47vqrX_Y1r1`$u)z?-y*Z%O`bEusuny)&{WEj;(FMR`|BM&9X{|UW{+6 z3xZG#Y-tiN6l|}|y>(cyZJZU@Y9HCQV5?djc1kMBe>rZlPN!cqkf3wa>iqi!+l?I! zHQ4S1w#IpZt*)xJE!c{3YMow5?XF6xrgj=&+iMmD+ul+8UcHh(ho#bS2&a=bY7R-1 z4~CMl80Y|k2az$JtW?lIx7O7$18E;tw_DVOg_C$mXs8oJOYrd@84b02zX%#imwsDk zg|3uVDx5ff+`Ren(NHIYmf(RuG8$^9Nmf^!bE*fh0brCd_F6*xLSTBbM6?Is5kLs`f z)g5u)S>QCq9@S{61MOG2V!z4*vZ0+V4}#_cGng*dS(aXTLDf17>y?e8>XioqQ|*>H zL$^#GFx9@AoxxN%aC+K*vs|;TsKK&#RD0!tkW{;L&d{Zk2T8S`XJ;gp&Y#lM(#vuF zXth*(gf7J+lm|?;n`mb+mA<29S(QRA$9L4M7x$#sFCI*_L+KnHN;$w(dzp3yQ|Wpt zO*6e5-_zM?rmy%QWVn&=dD_=>Bwy1AR>4%OaBwGlQUtV1>0v*7a=uayq_%b-tpu&N ziU+M<93EJ0O~3q!Praw??Q~`hLk6G#Yt9sS2^eG}KDLK; z$ZqDF^Q&&p%T<*9mg|8x`}b$#PLgXdupG<9-FH|JE>q03`{_aGE->8*msDxif{XD; z6$NvR<;QViT`HFMt^&)oW2{Up7tXSl_LVJRd8??ataVgR*(rkgVf<^C3g-Q*0CVlL zD-+CxAFrp~c*_+@JB2HfYR&u@zP}3v^L75g3j=fQPb?G6rH`>uQmVuSIT;&`;-#k= z8qbg5oV-*#Z(IYOYlmly#!(@UcEtK; zjJ5(J91YmdFbUHE_l2y-fXIWH9dh=|F!b=-1d7+#8~LlMTzx4HFcHptMEglVyZj_T3(WSsoI2?PGnAEgqV}q-V8LYpzm6F4VbF zziYD~f8YdPK7H)sOx5&p`Sj5{Df-ZE-QDR!ICHyMCyesx<6>OB-Fo3Ry4P^}*u{OT z>ErV0qkmHLp&iM)(}!>$XVzvapFS?exm>t#({bxI9g047aTsg*xOn;yF7cD05AC|% zoj!!ax@u&FAC=D^7vs*Z8ikX6|D(hY8@tb?{cD_TFv5J{4Uq?a9Tl6njdsj)IDZ=%{Ln7W5ADn4 zhm#6v()^%fo*v`}9oCeKA7tcGOGhqU_W9w%jKh}#jUFb6U6ehq}7gP$~EoJa6eiZFBp zRDQycU5jd69R*b+hoW0JhoXB0grTFMG+~@cG*pr>bVyWw!cc9I4syb{pq@#ETP9t_ z2}6fRX~H;@@Te4F=oqQ|gb@Ww^>v_Bp+KpA@j$8mnRHEEUgp~?Z^3s?eOE6iSn8ZR z!mex~9gS6bOK;WQuc70$inRA@6mIX=xOUz;)82oYw~i9Cb$Ps{gTl(>EfF@>tF?64 zSOMPZwF>eUPVklU)|oaP)Vy_+(6P(oEghRyCU2>RhHV|aR;1c*yKuGN_OpF_AP$$KD@lJ~ME2pk=ZL_GU>nO2w7e`$>Sg%a#Qf1DJ zgV+i$sLYwsDo$P3ZY@05{*IcujuO>(dDNxD2Fs)_8CcjoNbb@F1r~OTCNI$xAC$Uw zAsg{4e=erp27C{~sSMzL(EgVrC|f>Y!mW|-dDC6Pu#b{X`Ofk!kjGX%0d+M1ISo$h z!0Bfg*VuM}cWl6w5gz~{0^3=z$gqJeAVCHb=5xfF?_a$WJcxq!=O6!Q)`=n!&tmNa zFQ0NA;^p{FET{}+7Xc!^e6WD~!>ZiDr|^JpUyYU$u-Yp4Ns-MDZy6%J4a~g*6bXP~=O#S3kodAZx_V%VAMsgp5q?D*eZ?9m2{7M|y5uw&7ZYiaF5I!m! z#~K>Pa^l!Knsyez40}GDU_yaqH}5|D{OXqQELg&>3JX2fZ`V4H5rz<|IDi?T5dKri zb{Qy?AtWb)LLlKoo&Xp43@0Qf1kv1G2=;7Fu;2mQBm-0`nDO=-QJjzOdwif;7yqA- zLD(DSz2}D$jaDawRAmzbARSw-PeWV;07F_wN2Y)bbf2`E0DA?*##^p4wg0xf9w| zd}B88eEaX7`wYF1{0n>sGGT`S*=8Lk=7y=cNmZ%&6@EaQ7%qZwpy3t9yg*sPXQ?gq zBw#~pX5&cSbKDs`N>IaRCDg<6v`YC>iG-@9N9|adCa^w}nc1T^ef?!P;lIy8+HPVP ze#QWVna`_uCqBB(!1l=1KH_79WHUBdn={9P=ZkqWVJ5{)<497z6CWpQp7h*q|XI585$ zGS7zqfBx}bD;SVL901-@=3a^Yz~06#oOiGvqI=*NWWlEVG_y2-(<2=FH+ zGS3h+G<$3d{c0`kBt`8&^)ZKYKywP4$?(-AR`g;>=zuhFT`Uks8A%>cK~Mi;R2t28 zh5bvT(lUg82>)!<>l9Zpm2nlQ8)h(jh22l=;RHDXKW$Yy?RJg5OY#T{W)3blPGg#H z;PD=O)QmnL+=0WwV>XXYUHQSLai-X zsC9uXlsXCl9HjzQH-VBq{piz==9S~~isA`tII&?FM`-LC&wW@xUuUf3_V{ z?mg2QO>H&izywtr20j~ZgTB)w`gStV_t-$$jDWJy(nz@$DUpQ{$%NPi%dm$hE&3gN z516`SSZep%kszd84XAZkz9P>Vh?MQkL4E=A)~MrvSUe5OB&&81`a-XlLR$a|QFr$R*3zlsYHGpH|aCv3PePa5m^rH zdL#!436BvRWgP#4^<6Qah~pV$S9w#e{K1M2Bt4Lp8EMbi$ee}i;{9+4d0wU%wiml& zS!mxb9qqT$f8!ND7~rQYBkyUQQ5g7i=6DM_((o@c?gl(yH_GJPWWIOLjNANxo`K(+ z%^Dj(b~5#cu#{iSJ8P!$p3@j>QQjjT!TsifS#w6<92-?T5vn$u8daD5pN;x1juT1>J`|SQo=dNHLyQy7<F$|G`V;I*-$hORokfod z@X61IB)EjI@hGZt6v<(xF}oPj#uGr}TeIu4ur~SB&)ftFxALH@km=1RYq{>+a@Iul z1Tu3u^0Rh|w3HI-eukEXytN--njPCBldOVAc}4KB9=x!*Gs;keNj*TxSLxcfd9s8{ zvh^O@+R%&exTWo~tJVfUHkr>4+ZGTDW`U8rFLneX5Jnh>qJ`iUEP{|v1-!W^Mhq?U z3t)bj?gbq*ZNO{`{v)!ALy4tKoJU7<{1=*orLmOE8%ov%o_mSs!A?kwa5VP?e8VW; zMOy~1q9~egUmHfkqpdtKwQe)3&Vsyq^z4ToQ`F-akEL{_w^e@J%Z7?}J_?HIc=uUx zFUG0ceK+L&9nicm@j%a*c+SY=F8A*ifcQk#W}{w*HFHOSX_=vCj`resdn3FVaV4A) z|Li%R`KX#uzG%M&&**(+og6tx*oi(lVqUv<4ve+NmLgc90G6sz=@{*Fw2BObu43k* zh)@wh1yZC*wWFQ??JI_;B)vC=6+n35l~n*{UO%=5#@wt~<3L2X13X55HMd#?z~_q+OiRQUiL ze*H`J9#8*x+lYhc{*s3hU=lo$Or#k=Gk|7*QW+qAs}gNPd4PYtZ781I6!dRCizoz0 zH-|X91^)1cqLc+4fTI`IhwZ}#NFd?2|ff}bg$dQTE@_eQ%uRe|-mPV5HfHDIkap-l^;5JlV~ zeBhA1Ltfjc^n3Mb?F0;7<9G@+Uu6Z@VC6as^bo*!a?^Hd?R3JzbK1Zx9)7D5nU>PO z{4gUM(Fe?*C%lIrw=2Xzt@JPBo3&-0gi&iQ=Sf7eUCTCARa2&_jk}{@wwA0@8b|-f zcZO}2va*E&!go40EXio8=Y{4L=Y^AxTin!O)tiw+S_~Gj4dVq+j{qHn`-O^6A&}($a>%QQycP z-!wXpZ}EDx!YLr66NdN*MC*=^%IHx#ccWt!J^Ile;JY6kHrjO!-fO_y5}pOA{-b7( zp@aY?)(rl(h7)l#K)C{ZU)c~*D0;l0vhqR@`9res_l-PWg!f)m3NwgSl|Y4VAJtq( z1wuvYs8Qm!l4B_b&8y>&$DNJCpc@KbE8q(%i4g=nq{9oD+X;-LouS~kUhB2;_9JaQ zK12AG@OTGjxfwm1!#R$djryU1H&Vf)HCV-9i8N#163J8i3s%cP4;;X3;=1C|G(C-m z2acu@(J&Ls8LQ|(DF7*_Dyf75)et)|DufUZeXNKpQHE~l&48H+GUL;Dt=l>sB#a7F ztM?RR_hvZkv<2bhL9v|^6BDa^_NwWN=da$2P$dhF%$v;K18d})jy>iy&5k{V`4j<} zk8-J+#bEB^#OJ*F``q$Fq*pk?Pyz;8_IM(kOawKcDFvkP&a&MK3p_G|uP74C@$O-N z(7-db*0jF5Ry#%5m`_hrKh?&C_1QZzC`RVar=$zuE4>e79#4LuyPtr5p={b@Yxjop zfXvpi5=EY&ZA80^PhckQSoD7IEAMV8@Y%~auH$0Idh55HyW~2wE$kedkgY zI`^JG;-jbHXFT!cV3EtRJ)GHQbdf1q@;usqaqn`GyQLD;n>9lN_iDP2@?c7ZVKoR+ zI2!qiKNA5q=^IR_*isij0l>`kRq+D^Xs}p;X0-PM87kGyf>$}7Xb@E~h-d(@aacYs zwH*p3@}zcBa+Bfcb}kv2l*S!G%H`adV(!xL+Q^Bnnagm>kt3PSQX*VUIL*z^R%rg6 z`yN>lFjC}ib7x_y1uQ9IjF3`teH$pAAveYML4D&tf(CT@csG6KS@7TswiK682&&- zO{Bv{#-Jblor228bB+P#_&mH<&OnYlp&?m-{93(sGExQK9j2jRR7St_s0_IuN)F2C z7xJzpUh8ZD6-CNKkuuves?MQ=oy@g9=!YP`oGL5%J|Agh)yS%mwKQan9xI1Z)P1%( z9j5}Y85xnyVl_sZhc7UKffDg+v0uh_d@A}}W4$YO6E0>p77U4`VcRV8?_U%=`jee? z&8QM~fG3oUCO0Cb?0e(!liRX2Y8kMN)3qrmZudO){o=kZ(v*q&gpmOWUfQXaC0v^% z6Xf>6#q*OE1-R=Es?7n+Q9ghO}A{_UOVy8%t@5Np=>>HYo~OY|zR++ZI95OPOr& zG4c~tA~nrgPGDHFZ@P?xk3FxxN@{s4c@Aflz92>IDzg`~K6eHBoW|4*n2Nq^08RKQ zo!&(tXzcF>8FIxT-g3w*ZaK~pBt|E80%n(w?Jy3kiq7xSEQt7=I=qW?_%iiKaX^<> z_m+dYlDJJnMh|+FJr|8z@IYC{&~@V$(P2eXm}OAbPI_uv>!~N8r>+lH)kvz5^jeX0 z-EWf*q|F6tcrol$S0`Qq2TgHgINsT^N`73`I-?zAFE6^na-9RX;_S=^QT_m$yQ?tC zB3?Cdv^XIUj=jC>Tyogq&*49d0MkBk z_2L*Ag)P7Q{DJ)_q6=Sofox<2?>J%~=Kg2CU?}P`Ex6II8g*(aV^oc%D&yO%8ZCA+ zj;H0eb~;3*E8jg(wnD{*1{uB0;iYDeg;ORZ(0ceHtyOdO&M6k8EHWoP;J zwOMC!gH*^3C~NPF%m&(b47URt%x#B`V=`6Df@2o1lEs9ozhYfv{GH6g3IHIdA_)RM zxU;d^QX+e}*UPdJ3k-jSw|OacAA2EoU-W{-Kz#MRm#xa%+K795vk^B7`2}$SPg!Fv z?AuFv2oUi$rvp1>Ui zE#8fK$@;y~E)+!4ethCwQ99NirejnJ_i@<>^`jlE!Cr&C2K&>0?!D!YGCQ&gRC^`5R}@pDD-`*MP`FMro?7mFYz;%vat{@EndX8sWx?30 zs9-1YXaw5@1+*ecn~A<=sldHht+m*|8h%Qpw8&m$wZo~QQgz#2;Uz+N_>p zmsM?SK9aFnqozj9W1?o(;AM5y0V=VF#nU{d+9Jn!u|gVaHP&jZ)mVFySUaE1P8Vy7 zxy8-|Y)>?%4Rf^;-|B!i4e2{UI%4%(W%UFAboo8o6dFdEe8W#pNPIGHCV(ppG?mRbHL@}c^uxwkm+7KCiHc?t} zV1I!@y5WQ#(1`s)>wP&eg7~?j7z4)f!hU#9MZQJ1o??)no>d-QD^8F2b5v zdx_-@L?Uj$`3zJckPtYDFsUdC%_pHkr4uhOyxRtLs3wG!#B)nmz3AN}q+ z6FrpOkb*?6aKCq3S+mygIkip+>3k@pPWAeNEsYx)WA}Yb!%7}4MTc4G4aOZlpW#MS zDgN>AYV2oE{>~0Woqvqvhl@$3Y*#R5p=l@vE<=ZVnow+P5d}tpXC)+O+)^2m*oSVF*K_lnN?Re zv}RY$u9{snyB>pGGbZ3nSQb{P7|X&1nq{wrWnpV-cGc{v*;TXa5!p5Cz0Qnf3zlh@ zD|Ch(*^?~7r!&V}aK14_Yic_jvCncG?;af-s)ATxw%fB3eQCdW{wE$GzbwT>ovcZ4 zMBizgplTC!2pBsLYAmWm+3z#e&^e%Z9GV5B9(!2D^ewy>d51|hkx)7m=>w$EXw$!@ zRP_mWE%lwopZp}=o;_D@&)!qvlR~W0+b>V(4n6C;WKhLn)pIR!@kN49i`|Gmhb=p< z$}eG-xd$?qjM5sS2x0;*s4W#1ctec5ivEtmrx2#`#xnrQ0evM}ymP7qY}7$KrHX>? zL0c9tONzQpdeTCmsFYLXi4*EpWU@zh+LJeD^KVR_^5(HewmrdmTnRl2kzbNfmJ z`|V9dAU9OomqLgUI7x6yVuxa8qzh+*Ra;S9a2Z!24tv(bp^4+Hh(m5!6<-~8qRmu- z4QF@qShCZDVG&l=i zmC>fsaLx9t!f2aN*g1%^=kiGSq|7Z{FA0Lm`wTg5U{Pic6E*4^c!Qoh@WM&tHVbO~ z{-LRYuF}W6E+VSGTq z?AC1um$CbKgoVxmwRMpZvz(t37g)1E9I7M^43^FwSuz(VmbRgdeGAcJdh^%UuWm_% zsE|JF3ee{Lx$DY&*+#M~8_Jy&MHIa{T^E<-$7l@{QE!HU5Glw`qz;j5*8xa{k;V>e zm7^2XK=QL%q`PjNMO$s;cEJWr5nI8tLa%D{AC7_YRJ8<`e+gN5@6qFer;+X~ zd)(pO*v&^=VAiQM7)A6njoXLGhpNHKOuN@#h7vMia zob~TODop{i;%}L<9%D1LwE#)jkb3^@D@!{vSoPo7_b=Whsp#XqSCz^btY(Eqs=|(N z*UMgVKc3oBERkIj>BqlR{ezAUm{XI}jFQypXIxD-Swu9d7Y&#Ii|C`7C?i*tP8T3X-rOAuzGw)QnS^o8mQ0lUq!P`Z0sdXjK3wK1}cML%X;@cOSzZ7vqfXEV<)T=6sU9_u=1;ox!mdIR4aazDhw`Y0EP zbV|1y)$40q4jZFvlP`JZCj&YUC~#Kcr0hoUekE~IxLp^88egTFd`-}16A9A&#FLVU zQF+ukj)$c{yb8>eRyGxsS?rJj%Q#o=FRAXN4jKl`K`@`f;Co5)kT{{k-tW{}Gx!rf zCXlb!n)T+V-zM_2qc?)pz`g<4UwTtfm=I zdW2`~gt%Bq1Jnam%I0v8Ev+~4dsO9nl&1Q3N>u$D1*$hU+DB9UaT)QQrYfm{nlH%S z*(1~*ux4HWD@IKm@T^BB!w&ly$bTA@{=Sh7CF-|3r923XF)5yZ=JPNfXACw%%atu4 zyaDb#`utgRY*9%YcmwL?tt8!Yqj-Caz>n6$_wG$NOv<850_)?+K%p3VGj1wCPQUs% zHBp}HES}-fRy6zg#NklIU4gh`)FDtEFI4IZH8g;AP^)fI6%5tqLdV*87QrD>k(mj% zghpdKun&BY7E`Qjl=<Nk!DW^Qq8aUqQd*q(3=ALD{RVbK6{5q?mj6??# zSduDAYrtApzy5{USkp%Ib5S6%v}gYO`TWopQ~T5a^=dO8sjW-%IpVZ43msnOOJugsP2oU*>U?8;hbGSpPQGYFdz~c+jju1>;^xK7(jj~3*D0Z+~%fI@`p-p zSfbOYbz z60U5?C0Np7g_1J$jLW1Wfld zOs}?ykYN+3!&hgPJyI#154U{qEjvmM$`lXMd(;;{iCshAU{dJB9!{149yUH7+fh)6 zQENtegB(>BuoruF98F#f;nlRrF^21+L$<6a(A!IOm}a5sF!6L>BWf=N_{GqtP|jU1 z={MiLHkvh~QHfwM6-C#+ExknLWplu)ZhwbhWWuO1`6Sq5wvpzG4Sey7%NnUJj%#-W z1%)f9MIF6a>pvD*oU~KOG6Balzry(Ma~psN1rwVFx$$LiR*-@y_8tj=^57_0z+WY$ z^x^G#JQ7Llr5}G&FCyM5dY#pdL|dNAHK!o2OGv=aL+E7;C-(M z)g@gdV{Bn9)&S=_bV+Bz^1UN(7G`&eOcYJV*MVWh6$IqHOm#0ax=+!A!|tts_mN`L zk8stfYg}EEVgpZ3>cPUB-xgW+LT2KApB35R4bdv2vzrnW5K>U;upk5mV5_>h!2S2$ z+(DneGCaz_BY7Uk!pL`0YOz%>S)y}qi-aiW2g~IbE>YbUB3NRtHOl%g;$4uddMt!g z6&2#>44Km*91K_!fwHa+S)`@hOi9#liPi!WS03i%oN`tm?%@8Y>aD2{G`(ZHX-^a^Rs zs&QxDJsJzxpMU%VL)z~?3YoSJCF zRnDRlc_d~9{DI=}a%|<=ic3JA3=Ou`%Wo+bb9`1tVkrfMIvULff$3AJJG0rI4NWS7 zQ7ow(?!j)bF4J%xZpGhU?)lweHzd)Z3-|ie9_#rF z`Z%p3-(o4zKxoSi14d!gXS49wYAt5ss{Po|c>Q3rBJ>xMTnMLv+!AQ44JT;xRTNzcO_qILNSq(0$o((W@88q!McrL?^Wnw&_qQyTW9e5DnlF`ntk)F} zvl}uGUIR^}-MJKdwO zo26#I$Nk1|_{q_4d_qBxj!VOCUWuhjue&SdMqvnZ)Z18!ka(ZF)hq$(`6)&h?$X;eQp?vdgPBg0$5;7$nWn(f>;M{9NoFqaTD zjdmLCG}@Jnc8QleQPArb0KNWYfSwR9jdvRFG~Shrck)e666y>if35*8%OybFWzG4Xl!)zH+IFnevbqF-|Ms;@Kp+hK z&5F^kQ@=pzxC~@x1=UcY!ang?m?(H5+Iq+W86pWx_+USLc!6sca*?@i&1Mu&9=l`< z=N4#cIi-{?DIgV@Z_uJePadp^3#`wcjZsvLldD}x8|4f~Jv-ud9Yi#S@QeW+%+}pS z$myE9WKs5jo^M%lq> z-FDH2Q{96)12tuBb0VT$q|ZY7ETrk?GU-N)`&3UsMu8YQxe}UCF7*tw7cSDLA5BM^ zj!y3MgX2Gsb5GM~7d-dS{2z>_Qk?UP%LV5|%cBw#C_``kat6Z*7fqnyhfRVjzy*6| zV=0%is)=N?&Z^{P@X#|FQv6nlM-)2mWWAj}{nT`j!THh+CN{msjUn1hCB!o-5<6Ik ztus<}Umrk&8xWXVg$apNA9i z6(#&PyMoB*$#`>sR!+SP+>k;5X4V2D1&HH7Dq);+F3O3+^FXsq2y`UKnTe@#7aW=S za9YI6X20l(guQn2AIQWCYN9(GqRU<1AQC%v6K|u9g$@ovoI6!KI3xDd^r-1k(_`kr zP%H!=r8nJ49t`s{vSU%iOu{w_dV0Gt%2>>;bPB6v!>>rH?HwOEnLkSTVMtN}Agw6SzQx=FxQX-DaWfNM!CQ+mvDfi=}2Y3b34 ziaoo8hcAtDk1^oQo#8(M6-5B`T01BH<5FwzFAV+%Oq$WCs8R7$QIUG_sd3RLPYW2+ zuNnFF8%2EkjidVZqiWpAn&0eZspfZMEC2gx)K#~fp*Qgyb2JYDh492?iaw^?m$fqY zQepkrIBdi?+}Aj~+C@&)ELWxoECs&OGmda_YJF1G&E84DQFYB^m!IcC#MmoJ>RSXM zpNiUQcie?z$UI1PzWDaFQIEtY+J4?RvdMtplfBMJnnC0fl;o~7yF-u|sTM-<8^{8| ziQ&CQ^buv$eNN9)Yly{LJ@+oaS;4!AsmvmQx2p78_N<7v4&%$?m;x(?%_SS8i+LzpBWIZdDSv??|zYXMp3zrVwr=+ z30j_)+qP|TB>HgoU}D01&S8RWxH0b(!H0>nNDPoA&`NZ1pBo5F4o^-G&t?tE+w!?|iw@jW(hqeob_TfzF`q4Xiz z*k%nD7~~O=y=%f9^sc5frC7HQ)LI41iE~jbY!q$ZRzCwYe9$uBbD$#POl&>2EM?`RX|Wjktr0XvZ+s7F zdNASVLmA(q!Bm6kWq@h=UGP*uE0=zxe;Tlk+w zo4^3DpWVbyxs6QwOOse(_Y-?K0jQeNr_FYe5DdJ2pS0B0w6vz>vWR5+_7xT^lB6ix zX=#ULhPIL5vX>;(>>UcV3Ti6`=BXi+Y_3%jP^P)oXs&BmxiP6J`H~svPPuJT5j?sM zy@I&G(1VR=`6DLwE&$_nYKLf&3ZAhMyyOVziJ`p-B?xAB>|27>#RVpAPJia}FjjtL zdo;X>Kq+s4;x7Y;6O7~efLCLVw^i)J@J{4G4mhrK$eqO|>=i9vVH)na<%AOxyQ2ai z&urP*ZcA1DXL(!p8{L+@L2o&%B;ZZRpR%#%h4@njyOx2rB3s5~UG*ejID;evI3tDP z*gDy#fWN_nnrPZXJC5fo&JnLrGU#?15$Lw`Ss<&yQmU9S3Epx?Pq9=e?n>{6<2mX4 zxRRN@gX4QT69I z`@`Z?3XVD)MUmexEceq>7Fpwy=f%`-o&>V%B&yu5==5jJEIJN>7^%)(34yagzYPek z;h*r6U;o0=X4W!DBF>ej9^Dd&gC`MqAD9jv%=pP+=ZS*y1~BjNpCR@?gVDR57}yTn zk-4zYIfffUYf=v$G>CBPOhlCae{YYTUw<2g_TU(!w?~7x_y=IKQ%1OkvRh!;ZIb z4a9qcmCii$Qgz|BF(6sMHMbDcf#T&tkb zzE!5terY1m0GTg75G`iYl_Ebs{qXX~@7WIkU$E2^yiVCsTm);r_uz5XY$t}2j59)p zprv$00%Kf+$P=`s0*9!m#xs_{8}a+O>`CxYKsUTN9s@nVlBLNKw+`~KNl%r0&;I;o%;0*9zCtkhstHtA17_7A^L-2C@5WZv`t$ybBN{Na$^qP%OjE8c7T}kZI zHcDWh<5~fG&N~zx+rR@w=<}Tyk98SZz7w^OMsjCFl5i?A(|n=1yK$m1zBM}bH|&3xNi$B zhh-%M!KD$zBRzFas<#V=(6>*HlP-)#G)tTlOBk9ZHYD7JBFR?#dL^x3Vg(ErS9qmm zP?eQtl}@Xk!76Z|aRJdc$tZ1JB?T2(CIh~B+GzF|($}x_xNl#LetjKpdG4KuTSJgZ zkg(WU!ca$o=aaJlK15~Dau-=dby!Y%Xl_oNmO8CTJz1F79ncIYiE~Hhc0HI3>HGSfzL&)PFQ!qCGOtA$1dfSV0sA`j-B8WC86M5K2tE9G?)>! zP%%(WA;0AFr}BF;+eD>EN0fjczT*C1QIX!F5tcX-w|?ORGNgO`<)`+v`aL%DPGj_wj$f}blIj_7q z+Gk%PPRj8I_?OhXZ_5gl!!T*)&)sKm4gfrsrxZ7yrL{qF2~^C8krQ{bL3zl!>5kQ} z5!gY!`u^2B%6|c^Xv9ZP*%R3i3-{r#f5ESxu#+sMKbv2<8!qT2YneSs{CADZ7c=0#aI)cP=zPZFg7*aqo=`O;3M+qhoKj$nx$K?T=McIuA_C7Rywgzn)wdQd=J{} zH1loCe7#o8d^(am^K7}q1W9tk8dYSWh^>Zb$)-)DAcaU$?j}J=js9xQtqS|ia(J-% z<6iXJ3JLGe9+e1l*;}7y-@YAPQERk21vm~ZfgS|3O*Vg&P&-W zSW1lgazsYIa8bj}R=@+1Ac@eH3>aOj=}*(2raw)8n*Mg9zu&lXj@;KPN`1XkV7_Is z*pdXZZln27^P%QL&4-!~cjH4fQaRA`W5#w-W^A8CBxLDT@yP2K$h$`_$fzo5AW+v5 zu4ppTi%M{8BwAED-aXb}oX#8%ThPvAtJ@Su$Ob_oj7d~(Yem0|iiU5~vA17s*4LVU z<+jV*SxdUWfQR>-Q-|J8)#$1wzUub5`(|0b%R>^o0PFJsEn3b>tY*$&O66%lN5_bN z_MFeKl!=%(88ow70Of;j$RY*w>T3chft)^>us*OsV_l$Zu4~6!*VbV*5A^98@u3~i z)}&SSrx=J8rlKn|L@v>03!7cWHGP2gYRyQNRO&PZL}P)^L1hVR)Uo8= z`}2=~Bz7PC)y(W=E9Y0;DZSlf#@ou^i*m{71~B=8mqckR zu0*n1$LaVNPJ2t#rYMcdq|%k5Xlx=>6J+wXe45+hJbMtzD>G-{xRlpPP@ax+$xV3? z(aB>sVonB32(nWx*p~^fQ;*Mi9|^%?s|5V<`V01Vx2c%Z&Bs@-f4nV7i&?anHOR!E z)E)3~MaeGlI;E0bt8B9CY{zxU_q$lG6GrNbC%P=>re?W|Ww~yG<+_^XvZ{s@VY+3I z*i3V+c7o=BS4v>9l>wAQVTqxZN@3lyDXi7rio#m$E1M(h~8MhQk!Gj4?=U|!SacU z)28EDZb{u$(S42yBi+dGFWL5Vz-Fr25%(hyjxeeKOeKF)B}EMJSR6NQ zvCJZg%da2{zvDJ!!4~ZXY5Y~{H!+x_4x4!C@|&YbQQa}}A|7Ir7`g?bFS;k+>h=7I zpjf4kM7JN3ecCBec;N30Ug*l`CuK2jPEL5|^XzF35oq$K%6MuRSd%&elduRKyWX~Ks=ged^fT;D{ zM=SUwiWA^SPBDVGFY!){i_CmNx{N(QV^GwLK^jasYD5F=s0&s5eG?CG)X@MOwOA#cX$ zerAq+Yd)e3B_LJLSC<4u3fAsItIO_-oRs-s?Dfa&wXd-RlCzP=1NTYD0HTBntKMFu zl+cC3Am?L}#?Z!)ctSzfi*yy*Q4)euec4;xr=rRZ{vitk|4XM{pNe5+$-=+%TkthK zVac@De*zab;oDm=ws%5%dMO|RW_(s_x2o-)YT*7a?QVTqV;@k$fdud^VZw#X%h9nL ze-4jm-WJQY+DEVVz7bQpj#J2rQ=p=DxZkKlH!`>4J2;Auv5_O~ptqt8m4QYBEc7V0dKeYxmi}UTCQkItiSw@n4-XXXeeO#zCL1xoTkL?r z@`hC*Z}7r=Y-6R*s1QYoSp1-M@ds&arE2nmb0^#yr7Jlk7&@zp!|@9?^g9utH?>tT z3(((14`deM^-Cdz>gn+1GDup0>5I@F3cKCVf!*Yj z#g4cFuSbcYrw`c&SXoa!GHSWVU`9mlDY=jk%oD-W-Z*8ct$}Se2iZw_( zWq+7r8zSu1xVm4cY8VI)lfo0Qgy3hkNf6h`UnDQx^u(gR;`Zhg^PQZ<%{^8KD^eot z@QP;Q_3&<{<+`Bp52H30X+`t{RV9EU&3pdY1~R1riKq<*CF`atKh4BbI3|6OcO;HZ z19=v=J(v|ZLfIi4rW{l5V9{Pxz9Q(Jb9~R)@Yr}O6y#FyHp&$2)X=&R!7m^uyLtEO z4gIrodpZEjZhm;lx`r6iDP*E$iY@iLNhL97{)%+v0T(%{B8m+jB(emltx6pxs7n-A~R9@2uz z*atD&?D<0I}Svr{euuEh| zpecrbapuj#;kp3a4G5E)sA9B zU9fWe@cS-6uqt4O*RmT@QH)xneOo#+mU{YF;zirUExi{aOrcN$Y@!0sOJFMT=^6Q| zN6L{+b4nix!65{?@x6Q2=ynRm1Bt${nr}7VYQ8-%->$qUy_uSlr5e#EO}|@3G__Cf z$XG^@()TaclHG42PT?u6@znc_Cb-BohP@>bp+vD+qJscIdX96dbS%Fm|u@kT`$*&Sw)>IHH6+uq#Y}lyp ziJfvD4Of@sm?g(%l&I?ptv!1GJJNLEcSx?@sxp%ZS=&CaDvQA0=XvCc^9lBlNEwyL zsq)lh*sZwq*-B6nn+MYU{eZhao(OYklD7!AjT7Y9jhN@`H@)3+F|T*43AZqNRofLM zFvuQ2(+0CECejXMu_ixRWJapRc#B}0Ft3anVRD9Oqsu9Nv=6pzE}bHqG28Yv+pcBX z2RA8k=u)C7wl|L=0AxJ?+`mXL^BHc8Z(lpD)QJwyYB`apg00VolepXHf*{q=A&btsIS1!pt?fkl*%hSfN%L}ahm5wWgauT-Ak;oknfDS2M$aYfy*Mo z<@!(`qMA^L;lQJo3%yi+f)iPapp4iYYG`xldV|+%6Mpt0->Wa51AQ_OjIHN>N+5?Gz4^lRU-fK9o>x zbPwg{65FT*E`rGnWO4p8s56>j<|F|TP>#YuC>Ql1=t?9iImAWXHt%`v1+1MY7Bd8W z4n)#D@aYOB#aWQyJ1|j@0#l;+KF(DLM!u5Rz*+=rm*&}W@tnS@m{V;d8FLFZIP6_n zuc49^E+pEKTT>+0sDHHR;v9%P2=QDYT!pIG_4dMkhLL8!M$CR)&3plIF-eR zGV5WWpddNbkwXO5!F>XiiyA$DhI?)}@jNFQN{Tx@|ITuyDu(zEGHeNhvqx_{+L=+T*vwh zdyCG^7Lz4QkThgHv_@PH4XgIdsiF3-f5}39dHSJJh@MMv73hD$CXx6YiqqleVJeKQ z_A)y=^8}bEeooH0Zgfk${3Yx&*BdOb!`%$_b|4?!OuF#Y^Y|Dp5uP_{s^N2(Q9!29 zgBOm^qH!|$rhM@%$jnc(>MonqgMD#pOLHrRtqu6Z`fS6LSPA2`K}}F`BhF3e zvwx{ivyO3x8$mI0O32Pm{1eITQ}&apEK5-@BVc%2T4n(iSg;6SFUHjeqb~q1$~HIV zr6^pp%O|D{?k`PDmr6{%BNLMnI!#PRC#J^PF_RLW%VVZSu`?8{-Q_dXh8iTAp)Qr7 z`bTCcC3Kphj?Pf+_PH^X5}wOrsCKbf4O+X)XQ&OOAT>i>C_|N3qjF1zQbMO0>iB1< z*4Z7Rl=NKSA*xmM6xF(hV^p(VZ8kJXT^UI=j!ROJ&}ou7K1p@Xjie&!xj>TY6eX$7 zRglz%mam$mE|sK=Ba@U8I!#i?C#m+?k(8323nZy_QIcw(BuTv|VM2=H)2LK5*5P!i z!e5%U3?1EZ`c89KTze+9%1oBk<#PZg|b6MXw>R4^F%t&5f3L888}58N5E z=UBrjNP`8Y8wV55k=+^+^@n?FV*Cz14jt)tA(4Ljr>!f9{nNB{0pXy6LnxuK&1A<9 zJZ$Q~!`!VWSGPaXhKN+(z!f0H@N?`k1!4sM0xb5qZ0TtGASf2Lio`CAAZenwwrY5? zuD;+ghN8m{FC>oRkTiA$fry&MP9PAmsRI!;jqQiVbU31>u`38n)HHSifr(8Wn5b!N zKQwl^(TSSIbaRn%7Ps zY_X}s7EhhmI`y;Vwac!GSeVynC9j*;t{|LI^V2l5BjTA>#Yp%(j)Q@ZLq_ButG>H|^c<)UP z;>WlD`Tai4Tv!@wY371v$G0@PurxNMS#^wWxiPCsLAY9RGBP!?+q$rd^7U%5!;50$ zUH*z+-2#{ddr60-@kOkvFQgr-Ij}YwQcOEM#Cl+81jIBjG`cV}0%C`S7>td5g;)>l zjDVO1c19O=MnLSa5QD9;uMq2jnGq1vz|82v%&0%aEZ;k6j2Ucc^xIa*{C?KQ`ru*& z&@^x{x^OY-544lUnZZuRmN4rn%6ed21j;lpF1j!-0%eCqnYrdg6HxZ|C`?4eQ6bq{ zs8@fpU`Ah*u0d~KNYZQir&Kj5sx$gs;gRPNva1!m z^Y3O$;OluRrXkVkf+#a9cL*jCq+>55>hu}0=pu|0{%DqpELDg7%bNav&wJ8NpyjYq znzH}MbCa?Mwn~sv16!pJTO~+&ill^16Qt~mWfP>-z_RJXvI$b2A}L`31u6St2L&lL zu!H)rgBnFjuQ#YP*Iw$iw3m87W>2i9K&A#(Qy*4SqmXIWn|BOkrorA-yRFN|hRvrh zPF7G;11GBwC#zA^bnL#&G!3@KI&Gh1bnMfZYsLhgESFa5c%hPyGmIrD;r(DqsQ2I* zz$04=&j9c|zDG8IN46C_KUAdPy_AAd27dqLpP7<^tz7V^D1?FZcohjI)NBV#cRaUv zkdwiPYS;AObPM<7HpL%8o1)wyrg9fmJ z9vM&?;|Z3_Zvdr!IWGYv4V;$&oR>$2QlRmgp1JPEpyirgfHl2-nH)hU4NQ&!OpZsU zlkMp{X)@YK*miSg57Vc9`3V6i4g7=w{Demal&A6h$>i-U+fJZ7J!eoqQh4|eNgUFp zz82w39Vbi_oVK5d^x}d>$m*8ieFV>5EucmK)bTxgBY5_j0%~6hzW=ghDZPEiD&bKP zhU;|oiQq&3rmW?wb({y9PHFiZv4xpi zzdVP4n+Bf45j=-_&)oczg`2@|`}W&9LVUojUzSC{O#{o~2$n^?;TD`K+)TDR25sRM z0B-%VI|6PR*d0f(J8BI#*E-djo55B|>XdeHlf9@P+&=vHPEaz4_Y3-SBsi&lh>{EO zmPS{W$uwFAPAN_hl{P*^qz^CF+$Rt-MNcB|e0B9-K`)##7hQr8!6sh1L&HFxasp>2EN zx-?{;kYmBI2X0HiQ3JQ-7;ekchvQlO^w1nSnrqxVYdMKL!yK|_zD;0K1K;KtzD)x# z=^8$XG}i)3UDpniE@qSca)1Jp8aP15aDW=H)N1%-()d?>X& z6xB%05oxd^71c=iPyt6>eW+vjPyt8X;b^X-^|aym{)$vttVqd7{gUW1&JrnHf({(s z>~qi>!n0ZnLTd;T9^bTT4dGgCio)u^e=Xk0(auR9Uo8_!2PFgP#f~cQuFy%iiZ5<9 zlFG`jo|~ZW##K3)+z3i3!8*!6tngzb&T7|7kJ~8tPq(GRmIfQIuWajquGPYds3Wp}Y*OMi&$F~AH~Nx)1CtE2_1q&6@cp90KG z-YRq47G`6>48};nOc!IM1!JT(Fmt>UfSJJ?T2cqJgqh<3W-vqoW||lxEf^xTfth~_ zGc$87jobk(Vdi6I1|uY3ri&5Mf)P?1n0Y4vGlQ*+n>X^bcqI>D1_LBuri%g6f&o$& znAyV&TuVtpw93vHB{%pc+~CPv$>8Si$c{> zpjC8FP@?a%=|H04#d*3!!;#ZaqV0h@>hu=cV4sUkp=u}5MRZS5qVF-EK%(I==|H04 zGifH#&fpHW{-Y(@T=$04)>mIf7ygZ7NHj~?ee#?xWaK%>QNBz_5$LDkBu?V|=6wB$ zB9qbQ{OVgcI=UBaZZq5*jbATK(RwZZCNoxi*k-Kw_cCK&)6mGHg&bc0@o!?Ayy7~} z=OUzpT*OyvmdeGNls^j-*3>yE>pZ1VN$DWykU2wuHz@q=*YR*bQm5X~Y>B^3^K?Sr zx$m9QS~U{Jmebuiv3G_Ruu5l(%RZ}MZ#gAaS-xi=tqivM_-%{Hz6YxhxCQgiL)?P# z=ajhR*e1@h%r!w#f3;j4w1+T3fm^UeJ;W_oqE3lhc3>j6%r%Emf3-9o1aO9dTkwiK z#4Y&5PK#Tff!s3LG3T}AmN$f54%~u)?jdf$Ja#M^Y(yRu2MQd zG&{co2!&mG*F#O+Q$vaqbd8_Ki)pf+sTS9gvFJj$4i94$-|GMv5nDUQHvv?6@W3up zdtZk)51USb1z>>|8&}=K0vP=s#RA(uC=ATCzS(UbFtUB~cDVu$fP=A?IDmidQ8;iN z?~ph!*HPxSd?}%dv|)}i;Gp+7IOf(|_uv4ww8za326t52y8?p2R;~6m@4?Adz3*{r zfeB!Or6Uty`Fbo9f`d*HOtwJ`|K!@vYE!Pb%quv0yX3C_T^56T3CcPevj zukB7{GT6_~1iosLqfYC|^OT)zm$DD#NhWi721=KdVtTns*Gt*VXd080j%A+||6vW= zu_QmA#n~oa)c(Ep%InK=rMNWkR}nkqxP=jW=tvS5G7(hUidL}6F3iO&6(q8P0vuh$ zxgZe^KSwJ%uXyowm?VPEsgfvN(#Zb9-hr=U{)+nT*Do6Yq4@7x5D2qU19xPfPFzYO zpPYqRj*_gr78&P^m7V3BuY?_i(-ooeYMq4q!%aXaIpfdqj9rG6OQ2!G8YKRjiV>Yc zeVdL&LyMp}YhPe4yR}T_ap)t-{NS95d4;MG0f}tnT z#~OLV3(I%y(d@!=0)Ob+!R%YrFuI>R-!|+vZ{BYz`-94m8i;lyN_kwB7RphZv9`Uc zp;Z;VWnyiA(GFVNu1$k>$C*gic}7D(w!gTjGtSB4^SeM7U#9t09oXu+`N^E35jl-2 z{E=g`A)}~6PKZzU6u6JdCpt6rvizb8KxqW1@dJQTd0oubvUAp-q8rciIbF%+T>Yo7 z3#`TrD{BZ^J)Kr8CQqkTKIdsi6J~w4mIHuSCT=539qg@?ze(@D)~*R*D{ z$hARJIOAtZDGB7w#rZ|3q~b0-5%lx*YN5VDp;^Tl<@`p$NLnoK%<>;WFXv26e`@Hp zPpO~R6d1)Rng5$p))R`;z4%f&9?B!|??XArNdK#UEF07{ci#Qu3N{cN3WONSz9{ZImN#ni|YA6=fEBnZP;qP_DKt={jOQ{Gug}poEf$+wE*nXq9 zNvs)&QIc zy6B5UbIwY-QF1xI43mrIyl8-jauTyuHo8&H!#Gi#luajH;TTK3EM)qPs7TqF{8R=~ zW-N8m-GC`FNS(K4LR4O538P9Vc9i2_vz$${--tpw#U;_6m)OK^K<-2ZtIzss9~S-t z(-pe%bH4KP^y0j7)d%5*r?>$=JcSQG%{hfBN!79`O@H%MO850N30_5s?5)i`N+Fbt zhf3cLieK~CxZ=y&Spm!EO`KO$6G$6|MWusC3r#?x(wm`4u~0%)AV84bq^WdCB1I6D z7CJ1wO6V;VX`v|ui2{+{Vn9F;2nk3BW%=Xof3df7<~!dr&%Ec%T+A5_n^m9zy1K{X z?E)V_?g$+Ud*!ZRePo8e*L!1Sjwi@S`&lu8*{j2I)>=4o__690dX9+E%UR0`qFqG@jP)D96cq zopkg@OdxwNF`u*C@H-`Lj%K% z5RQwAjuJi{SuBwvZDVIn*hhTMtkhf06?jooI>qw_ijbSa7GO$7Hg%@6mAiq>JCDfI ztgq?ZSXWk@-7;OiD^C$YMP-9^d1uxxhMZscuD304bQUrvFY669@B~T?{IY+PqZp#e z{8;u?94*UQLw2y;gL5(4DDaFR+orKj+njKj1X{f%SIUu$l5oI%5+&k1fA z7I~&%lk}CF7S=paAynRUDZ{U-)ubo$#2iwqJml!K-c)0&igx0m<9>fC_@SZ3ff`1D zq9@|Jsno+Rs}%J05(`a@Q@ZNr1LOIaWp-W0DZbR@NhW$F`Q&u~Th{NPH zI-gzoZRGRfV0s!#+_`1h}mTn*s$OLJdUymi;>P-B9T{G}Zk zrRsye5RBCM_%Wh6SoNhi3HX`D5Mht3f6Dq%oJg#0uT%1oZl_{k{>88v&5m_Zp&I>r>QcnT%xe%?)_6%cD|Ws;&(<{=GJl5`huyxREnOw$PwfoI3=+^jG*g z+xNwIKu$pE#UW$)alt+Ce+Y1v&caLno+uK$fEe0U%pkV`H-WkeAZy})1S(MF`zZ!k z2R+v`zY?0jy{JRHv_W9o?>7T`z|`rtTA!M?KFjG@sAG`%!@>Cv$DYcXemH9#VOEoQ zl=Vae^M4J^vAqL#vTIqm!>A%&Y^E`KQ%aSoi%+RN59^r1BLOgV9$1GD*X%1GsTc>? zfagzC|wot{BjLvX#TEe-YU zeBR`bT>^`s)+TqUObS}}oA zJKeZ=yg)6;|F%g#EbqEmVa8$tSDD6dtql;X#LF7!u?C0mqzN`@Z@}UV@2|vM$*YAba^(n70boOOq?QJYttPPDU(KOCnw`UZSevI?z#{0Ki|>2 zXc=gmEwSq8K3Jw>sCM8jHy6v9gDd{*WJahpzOGbwCu)8-i{(_FuRGfBzQfO9qsccyep<(9t$)}Tpp zrzDqsPmwQzCd+pC>;+T2$@pY^G~WXv7%cMT=vj*K^GVIWHDy#8KXa-b-v7(qQs0<8 zbm^aW zQueo*(0ed^r->LZnN#`-mS1-VvZ3SHTkQr!L4T8uffmtcB1sv$KIM>rPkWse0V9s# zza|6dvuFItHRG(c;s)t??m0O6} zMsA&(Fhl?xW!Dzkh`A_ko$Cj~Yhv-e;X2iUX;*LFV2K61Mx?A%E6F)ZCa_POjl(#6 zeEO3#xPLqJU*(ir8vea{WBqZkO!wdBwN}W$l&>RQA4MxajD=X8D4U-vJ(Fk=7itV1 zYny91re$>s`9|B9`AXKtuAJm5x+-$|t<`Y=FLjK8R%enn}B<)Nae zfuM!vLHiGn%rrOd=*DsW-ug8c`Fb{%8pj#YkA5J3Q)Gc-86S)=I3>Dc zLO)RL)&Q9%7OHny4D?+G#c%B<840{Laf7Ag%WrFcyZbJz+1PG5JGT`Z*9UZ_I@U$d z&zIc?S?Jx=vt~Gq2P?@%pyqUup2XNk{nrtSSeC{8C>b*Z`jvv;hF?{pZ^4|dKtYl-#h^;eHPXr}m1L3c_);TMn-=yD; zI5-aTaHWC2>^gry-a|XM#r~M2p(|S$2t2SMdP^5<-=Brik2%)>GA8N0fU71KVbB(Z zxU!1f)|Opw)idQQy$GXB{N75Ux4z>v?Tg_uUpHxT@Dz>Xd!B9DP!ngm(5-Q0r>*4~ zBaRtb_!&QbX`3(0c5{4+)(6|drfiBD`SdE-34&m|7Ne0yO>=xNu73@*mY%vd$$1}#-Di5rD6 qHDa3`j87tLiZUn@b4-y3GXWJLIj@ZlL>owFG#8B2_M?{>&;1Yd;pTe) literal 36306 zcmXV%V{m5A^Y&xgwr$(CZQHxCZJQh0+~AI}adOADH}2J)y1a1qJPOuG8vzG3-fO$6-7d%f*v$Z_#HfcCS9 zoM%7wG5-or)IDUUhFgn27`$=LbAkAFpS==6nlC=7nYtCf!;cM&*qbXop7%S~1-!mw zNN22Z^gPh}6-!&SST5;P1-MLoZVVJCp57aO#}F%i47^6-FdJ!c`|{xZKy#)5&4~uWR zZpv0^jlo4g=L>2Z(}>~_i{T)iJp!|WTHzI#t$e0W#PfwC6ne&TcQa73`fd+)9zWS?zd@s#Fkb+FA;}x4!F=MB6OY4!(P}TO~6aEKZ1?m&0WuUv;p1| zJQ`O-skOFnuZ3WXym$^YQ-EM}iSElo&yC6REsevt{tzFB7?GcsRIlqrpwqOlf8@Pm z;A&gf8o?9)m6N+g>-tvicGB~kP(zpmzU((^;YCqcxf1WF_ZksVT#D^t`l@;j&f}6O$3xgBM5r7!PPkr_*42zrNMi0sb zY6W+DT!6lpL2!p_x^d+aK^)k=#Va&v!@c}nrCg^BQueZ#ov4l)NC5N((@Mz>d+J@MRro|Q7PdSjX>+F7Jrtl-UqTvZtpL0+kLph+#NRWwRE@k8hkLU)r1rxR&rvq{IzAxd zWsNI96xZmM0a#cp|Ce93R)K4FU$0}r`cjxI8RSRO2`2f7D7Eg*Fe`Q|`lDy0S;eRd zR`maHJ;HLF~>tj3NiIR|{hCLxXT~ zR>LSS-dqlNI=~M9mB1s;um=6Ue7R?VTn+rXGbYXLa)Ibq2;1CEvm-i9tDer_kOV{5 zkzpGaz>pAz`QT3~Oka@!!3#Y1fkXWr_*gH7e{_5-!Jz?1v5rQs4g5kg?O<~+Gku)X zBd=xMOGHz{Qu^RGg`XCufimGDzLQwURGhFlftphUy$1H*UtgN!g=|(->2Ct-<*Tdb zPlcsi!X1N)5=s(CoE8|v>zFO-1EG~$!?)*XdZiRyGX|d;-xjXz~Od!R0 znyoMCx(j)IYO_JA{zl&Xy&U#sVhC&N5m_e`|>jxJ0JZI!YKxB4}g>Lx5pYT{N8OqpA4+g4%=n zh9eBleusbIp79J3e_XvvL&XLU6r?DA-0z6u9W{p@3+9jr+%vPk#CvQXUdDUk!?znY z&)W}U)eAy@3>E`@cx*h(k3OLvnJ(LKUisgNNDbm8d47R?vqv!)I&kEkmZ2e0qeQxm?ZKizP}v7k6hGc_$O4YC;fb}%6!e?@D@n@lHvLOMh22siXs z2X8K<9gLr_@0nPs^FACJ@m@YwzL*Kpjkuvv*mN^yU2}dh^9MnL^}?Lir|d`NF^sj` z9$;3$!IO-{>H#uq+Bu4z4Z0z#g-ZR7 z5Mga7bUg}iD_TQH3}jegbjFk)K1@3NrNYE0k6-Y+te3GM#?8_hddUK?aZuiF6MLw( z$T1^tc=$^q;L%oQF}ggJy)@NQL5B|_z9QiyOx5IIrNtz3-B%dVQk3R6O*0Oy=-iQ( zVRCMAzm@pAiDoQ-dTtr7G<#Cf6#0@QjXz=GBQDxF+1r^*{WLa$nMWyZ*PmCm8YCK0 zDrJMHX&K&?wu0o9`-vt8+G5Wu#I`e`D8j`B$|IDLa8}ffz(@Xpa7Y#p5&(YMA;wb> zJ;EfZ@>-w2;8EtsM`Kf_IU^1oIRf~6an;MI(`sJE57P_LDcodegM3zu88VWs;QHDP zPvU*z%HVxsqj$dqpD>6A8&*w-w?f zU6Zi$I9sI)%~nK-lBE&_c6UxOXOqAwo<@lZ(WGIUrm3SJswfEeL>+`CyUXtVsrNNR zOuxvfab`@ZB1u3CeQq038C!*?l2A3OMdW^vUk0g=~kj`IKk9;jXV>T#ViouTk%Ylp=XLQ>UXgz<7%)R8@Qlcw|y zXL!OKZU9b|Vi21;a+*4YnHSqiM2oE(te8MtSN<@~ec$*z7%RVDI}Y1CEIQWRM(Esc z1bw%=a2=s#>n(nfwuva%WgnYm=X4Aziu5DObQ)bak2hJ7bku|G2&sfb{z3Ne2L{<} zqOis3s8NIPlyM3G-{Z&>MQ3ohtweS}oDX`CTWNd@dam$&tVp^D?{Rf^fZV8?6kfeI)D~lvd zU~$145qicBAcz;Q;4mQs-_!^5{Mjz`dH8O(xG&)GE0VYY+w0@6*BA;jjwI0xgU}sB z+w6=na6qpt+|BC?( zx-aC=@X$NUzIt**U~FD!HaN)Y5J}>*Sg0QZ-GuD5lvZ274%aj)BK!M89dZXZfE@b^X) zFOXD|!PgnMJz?>0c{2ypVk~PpMrfUGchTP0q~4>vn#1F3+*A-uPgfw!kf(t>#Xl+6KHN(nTous zt;^#R;7CmfUYbG9=dF_Vz0&+7f(6A8OA&B|)Iq``D>-d1DCaSIg}!~X!hy1#f28H2 zqv6vE>0SbTw{l}*A*0l#7;K|HBSsJILyDx zGGG{^4+T+T2veY0&?goDdb#oaH-Lk~sRjPS(Wj z&`KSTaLzl^t=AiOg(-Lc4gP_>vL0-#Q(G(@-?x|G9(LEXEZ4weGeIH{HHby!yxFW7 z6}4NB;t%A7oB}u#2$C`G2f^sm7tF7PwlQ;}C7ft67EI_=uTpI3A0_-C%p$v=$VrM% zP?o8mLCP4-y3%VFZsvQl=sz0Ze_#`gNU|al(=b#4=)4S_aE>Q4yYech>}sELvD$R9 zu;Ui=B3Bw4=LK%L8b`d(>6Mu&7b;Q(-UzYBK3~CoGii5&OAzQ`;r>ADCoc06RSkA+s0N+O%{&)$4Xs;gUOMv00lLurvl7 ziKmC+IPhf=A*>Y1p!mRN#yk&Hogp1!rg35ig|}Fh0xU4`YVQ=gmB&>Od`LiBsl`;y zZsQw97VMjbVOcY@Ne-W~(WwnBpX2*l-A&4m)m?|hw>(|Z2|)_vF`?E8xtvu>7kFOr z(p{_64NKf$;byF$;Gpr~Rz*!k(rGkV4c7)qR8bt`^pPa{n#13^xgfvG<+RHKdsMFe z;>3y*Nf-)L4DEC~)Q?ic%T@%-R_Ih}J+Cf&wp9W*nebq__qZ0A8Q z(p+2S9AbROCM*d7P9l+HG?$$YB=1S|vHNVndZSC9*IIjX;V71l@azv+!R>r@^-btZ zH#e9zbW!f4rZ?s_ezYNiIAq1(x0`luY=`YhnMk{UIC{gqml3Y$DQDL4TvV&FPZ5Qvw`4byU4)NM@pzpR-$kG$wz(Ei4t#<1sw5h^;Gl$9^U~zd1 zAI>tHD!Ei!q`Rk5dt1a}$3+8xLRLCaUvWU7+YJ&UaT5Xx)4+c4Oh`k$-+&BLxP;f5 zaR(+a9%iZ@*67TZc#v2P$&ES9sFkijrKarW^c>=;Oo7;4W0*N8m#Txt)NiJqgyu!8 z(MIQUbVx%LG|;Mo5hDBz_Ek@gJ;tgKI~?^pR0$}StZ`6uO+Ns&&3|9;yp}XwiBpL{ z#j-}$;+d^ZfyW)mHho;SHnn25m*JISpDYRSU9xK>meNby2FfJsK_F)Qx*W=+CxV63(Fo+pU4>oA-W_i+* z0~;!dMvuF|Kx^N3lI?knqP(BCZGFG&fPKyq%;6z9T0$5e{JqGQA9D_6vagx7DL(*p z$_M1?v*lU`Z*S)o5MYZSsJ1~T&aK=}wzWz%*H^q8bvlKJU{^31pNCLi$QoP-*doAS zbk0Z#oX^7aRz4qvZ6XwtR6v%vG@~9~m#tFme?_J;?$VeE3YKz;-^A-ej^4zFK<`j) zXA`Z_b;hlz*u}%6;#9E960qUggpvfG&c{Fm_6}j1#38TJvAldO_xP{$4Xul5ruR-V zqjjn#*Ce=_Y1U7-L*0bzev*d}1Yc_!w7|`bl2BVvDD4@mMIS|&^LO~JI6M%MkD3Q# zXH&&sYV9U4U1Fjg&Ap0sATw7lrPn&7N=awc41wzOjlD~YrxP=XNeg{^kfrYU_M%MG zyEJl^LnONe-F%o#k13}Y8Qq{tXPS5%N|X%E!GFV>EKaEps_||jq@0HY+Yi;T?eE^% zs7f^~c4n)8OvV@EG`4QpjLpT^UXsuyixY0_%CAGn;~u@l|I6Pp@M!4T8C&LUn-aYP z@>}iI#_v)TDiIPCPg{%V6D864!X${cT$McA)@9%c@t|`WZyQ7vCFcGR%pNSUC}o{D zngo4@2k;$}9BJSVdEsIqRp9NjuRaY)RUn?XhiYb2fQYD)&_~pr{j8ANqDtj^y6Kg- zz(mMKBWq#pXOJ(X%H!B81MgNlY)SE83Nhhmv9X$;Zf0yD_!NaaEG0=Uq981fQWdf; zn=#f*8tqQR%|7m8-W$MHmJjWy;dj-#BsbN`P_B>aK)?q&+PTs{oBwC zpC68flOm%OnsuL_pm_;(#CzUs@8BYAP|hgENdrmR33L37!`^L_UkfN-BmV0O`E7LY z;?KiSv_$Vj)}hDdmKgo_p~^4~XbA47qEEzPad12^*lW#_MZ(?bSP1*1pn;Q6OipA+ z%U?sFZ#@v7Zn)+-QT7o39v-;z6RBdEK@XGc8WDTKw$`5ANHk>Sa29#vmI}AkcdqaQ zImYT`N!~A!0y|ygDY6E<#l0ZK$%E^fn*5CuKlnbB9r_#QPg9|+Rv);(+1ySJe!=Zw zo8-@uR*BAQFeUYY7viSv{(AX_V7QCjkj&lX1Q6^Zkgn_`*L^m3nfx%z@dpBpuRMdfgXIJn9B64{@^)!qpKXF+2e!@rHSI%ajzg=$v7|MTAJohz3=Sh-OdRq z?7@CO;Ooa$&Bam@=zX4UR0X|e0zrlBYq9yN`#?BS)XvBy%sX!3#F-cHa&e^E+6wl4Cv zw?9&gZ+dYU@(aV0k^udRT1=jh2-WW3{R6F6DVSG7fNuv&jIp)PrDX0s-8Bo|r8Th? zCVx>;Kor>>;fHYCwpQP+xEDP79_hJ^O|*|cf*Z%N!sL(~l$b`Ivb&;(W`+@-1<$Wv zyq265NxP{60JVTJ#=(Fv#^4JCAcQ+rx?PIXZ?XDhQZ~sa2T}&WyN(0|)6$aZ{>G}3 z`)9+(Oz%%cNNIz5{qesE`enIyThnUxg6&J@DyBIg!!v?eQOWKHWyHON$fy!}NCOyt zepc!E$TEtKii$Zo9Gtdx3rnrsc5+!&Je-vDE+QRcQ=3JA{GY-gg45d>+sM3sZdg1h zmZenI%@PN`O`o35zh9z;5`ekEMjVGH%r5c2Rz%I+iMURb3Ep??j^`RTs|RS(^r93U zPVXj#E&KgOW%s=4co>S|a+s?f?aP(1^a4 zrpVSv;(UC|0|k0fi;;a+djz4!zkTlr37)u=X?RG1_yoe2Ri1N=+E`C6$1aJv{{!8F z1Y1}xv$3Ik?Q@t55ic<>^a6E&>K8?(4Gfoi(}I?dR;7lB*H!5}{+j?^S{OeLf9!BS zdi;%i%)hwOiGFu{Hiz`lejM(tQ$6V<#=U6P!G2r&icP~p+{aEbmm>a_RYDCvKrxj+Y;b@|??a4Ay;4ki?M=5jz`30^N#Z|@N%Hi{! zBw!!;8d~%{hhf!soPYn6lSx;)%0wWzBxG}&1m%zTwh?*ZuvCfL5ACL%$^H@pXB%!fpV_6C^Ki=vDxQDdq0EKJKl0!IsZ z**RxysL%HQqM(L<8$ApDov^(|^dJeNtXaGzjX$K0_Ml8r&A;KUp&h`J8}kKEUjxu0 zO3*Yo?a!^++gWNO#5zFF4gdFNhPZV5dYo0kE+CYOJD~3|n~5$JUDtvwV}XT{t@5o6j|M#z!oBP3-)}Bs$&uEqSE-KYzt5+L zVzo>YB`;FuRo`tBQ3%^^+R;2PzBaqB`vZn%vXPLI)@OXfK&lM;(?KjQk*@8%Z_z}a zV!vq7qR^+BQ1Tg5nG(46o#j*hIWy@kx5H5t7#X{Qk50^UrEOVkLivQ^aod zfwnc68767n&=6L{4*Th>r!iBXdlTgF2RXFWq~~Sv!BDz@fG0OsU6b&e1+B;4HqOxj zH9wAGK`{Y~0RTi|>sjY*YVnjkvB5uyi?(Z_WNR0{Rbe83qLsdPXQA%|I~ePvjAU-% z!H{R}dYe38U1UR8vS^v;aWY8DXBm|CZN26@JXKS#RON>a3;C(u1 z-R&Q3=+={i{8bTO3!=AFsC6*PpNS|oZIS(KMVLkbB5>lqubDwBG)S$Z;ow(90-qI# zt)sADbqN@i2sW0jB0IG-#%FOQkSX9a^`2{PNyWXY2BlR2V&%RS&NlNe2Euzd&Nkuz^`$hY zXVD7QIj3jJp5vlULw6U`%BsFg>I=(%hWyL_8Hz_b?ko#EP%d3!NzYu*7gHS=1K~H` z!_;5adpsSTYq!qV>wcrjxF1UTu5*v99Cc0e?952%^Y38qVjY*!Z0I+sO3{~>YcHWyg8ye=DbNF8;tk@|#uF6+hM^t)Qp=o&B` zbOS|(s#nYIa4a_1^?m=AyP9MR>%T2OXoCr< z^Gg07()2&%_kRdE)_+LhTwN{1)9Wy`M{MT6HIY8-r69<4jtvB2XAhhS57xjkD?C1U zeU!>+j%FLZKp_ldUTR5@JyG_Q!*fs%6OkzRb-fr)0)&g_($HuSy*Wm=l%`V4eYEPo z9R=4pI>I}%B1vI0u%(FtXacEE&*vh_&}q45_BF@d z$i3^B*~4N6(n!wZ$^NV5S2LC#r%1YG~muBKR)lgaFN6Rc0nz1Bl9G@BfaCi+{())gI$zlwzl# zl!9*fUl^wQvy!t;Uxq1hJ%An_?CJ9Gw0YSZ~pzKBoMuuq?hpuU8d&6L1c zAWG?4eYMtH6I5gx#JiU%eL@yL;uGGbGTJ3{1^5#mnWcOqwks`+&ec7V0xjax6`IM; z%Kvxud5$;t|&V@-sij(3?$IqzhGSd3nnqt>xA9e zMs3|{htafU$DKR%a}|s?ItNeuzk(}JFI{wzpTAL*NIw+PW+ABs&dsQj(?t8`Up~K| zvjwcjXEUH!ACoh)zofdY7>GDF-`vN^Irn*x`d80YZW$vZKJNS0@dL8$0E< z^*7xZ_tubQ>WI&i7M+q0%-9eW&9RZX?7uWHpo0agG@}@r6qn~oM;RAh4z4eUTkQWmO_+|HnLF|91U(s^T-QGfV@8~bC?&SiL4Au4;Ftr?Z;8MXs=p| zDiqg*Z2xN22)ux-3_s?N?; zR;VhM8qqHOE4No>PCtMelB6xuJ&FB<>JS|jdvr4;rgX29C=f!@?qoL09Qlrs#G=GN zrlTDj`8ES3o|R5cGyd?3y}xlnc_N|eetAd-GnY6J{(i~)SB-8W5S^mzzqSiitr4^p6c_>NmH>C-O zUl1kC%|t%cr`*KX^t$ddq|Q~%q{iPOH>A$)pVp<4=KHmCqI*kpO(f)6wC4T>%VE`v zW67ik6PQy0+1_2fUT#7^S>^eT+168Cmb2~}L z+j$!NnW%#Hu5whK(5`8MB%0W~iwz(W`V{F}14&UeQb?REp(*g1Tf4zCb#4`Ps48 z(q*+PR6guP6;zgGr;@rsjkg3nO-(pn;vLl*-R;zE0biD7q+jYDG?*ajSC;E|3coI@ z*k&I5{FJFHvVZRbqV}YQPHc@AgZ&iEbs**IhPLx>ZBgf#eno!Xp1n@;B=M~Ww_>zz zGZD82c==RrNm6b}N=HPl(`%q1TcRmzXC4_00~X?A;{aSt>##(270R*|v&nnKPgiNf z;%5`Ap5z^p0u08#fC9n-O98nl0lDM=@v8#b=*R_Jb35y}QSqyvYuk@*20oR743u{& zuja`@uDz zvdNIH(Z2wmrazcd@7Tdd>#SOo8SvJ+`tOrijlG5|nkzz;6_+@Je!^dE?vVF`l0jWh zYnrgfLy4=aPQ4>BWm^lFvTarYFgrWhA+L1byV=2APt;nl#|vXN8fkQ{Qc6+)8sndn zb3OqeCx>18U9PY`^@IhQ`#;p=8msb);C;3k5>Fx#@Lx}*+oyO8Lg@nm!{{b$(n>Pg z1i{wq;Yzlfeiq2GwTwyY%1?|PYOL6IFN}Ht_Ayh$iFUK7w23M6QPJ&w%+YW^S}9kA z&08^X_H4!jB4o3WevnuQf=Uqm$%|?+mP$_CDOc$o+#WYxK023o4T^R-F<(~(TR*}k zZRu+Y114=je5FMz0v&M}2wvI02>-a0w%&LVPybCCuqVCN~RlCsYxgDd_M z{${98`3g?`@^NO4Z(UuUcwdqDr8*z!b>XB;Je0jL#c!&iOl*B{e#I%@G2a)qIL39Y zuSk5NOdMKTf_x>jT_6!lGJ!%r&>By2? z;}3~-n&c*(;cAWNoI|)jYm};Uc8(}VXssE_eI<^BY>idY3@*e83W#eXHuBG2vx-@g z&}Jr&oPBV%s835u*|?9h>1PV@apD5ELrj=3%&pnezGcBw@1<$CwPM{TY(oD>mLEXl z1yHTy0D(hmbQ8^P?O9X($IC2&1iE^;)KH|}pGgEB4w|HHNi%-<H4L|;@-n$rxD*rTM8ro5dATeH4RyvD;mGL{5^aJ3BffMY+0Evz^+HwHZ-%H^Enjc=$5!qnr9687H? z0D}=9Fg*>b*bgHuDOi7>$mGc_s_<%gfGzz2`_HzJ!!%FoRy~5n146@f zD9JDuPmmB&lPa=An=i7&8#JuAwD=z(!=aJG;vw^oP;pFtavS0)Es*`CPh~g^D;_ER z&q?xaj--U~ znm^>7m$zr9D zyFCQ^+hvt6`C|1Uz5cedDO= zVxca)Vtq$deO0vN?N4I22hb#ai4q*%*LT8@uXJF@Xzh?PXTMKvmCJ%W0CRd|i9hJZ zlp`H<6M6j`4#IK@TA@=A$uF?m5$8#Cfp%vt81XOAsl2Ejl%!(N7L(@_;h@`9V3!cu z#d11QBx=Bmi-FC-1Yag;FbtBQ>G)my1_HUH=-(r}u)bo6%y5eY|KL6?H_5T6W~rOHBWwT%zXb`Chyv4f33cc;%_lC{|j*<#@YriHFUu z!r5jk@$AKkMAzd8BR=^uU~zJE89y`GtOi2HyW$;ijMB!1^3un+^7NeqLU>&%)#b}9*{+I^veE-?VkJJ+)Co7B>g!^DATUPI}=kq47NN}E}3fTPsJQoyW zrvwJyh*0IxP;Asr4R5k=a%a`um}1@XZAT|I zxHk%VwI#kHiZ!Qa)z)R(1MWw9by-r)%3UdNcVvT7IhB2#xZUzWKvLC_DU;g}b=Zm+ ztOX)D-=EsHPDi{u!SI@F+q2hBjY_A8xR~Kj)IoT?PCy{`{;Bei)K}S#Y@BfaWn#>S z>;&vc&pxMzGCGG{W0y?>$a=QqZ3{4yVC**8CXR-gh`}JoLFne94EJU*xBryIZDF9i z!=Ss2me=V(p&!uiq`#pPM6n#`l-S|HWsrFi9jVkUf>8p^NbloB#NC|;ijC{a z@>?sfs)UYhkZhDp*?E8GEX5Qc!d@j)F~vfr(2!4n)196%Tg}*b+FdeB)j7g_qI}tM z`!gc3}yslzYk&-p&3aT=f zQCcqHG0h?x^Axuy9q2Ed3XGS}(fqff%CkWSXxa-qmS;lSKp%kb< zh<5b9t#B=Y8O5Y?@0L3xgS7`01zPUx=eAe*KGlG`zLMMV*UQd~P2UW`0zSLxV#sPP z1V65I2>%M(hpXeFo8sVROz>axD>7YZw8`P96Jfyc*v3d|#y3JgsEjsW}=MU4^;q*z0ym_8DjdTBbd4lf1VS#>MpL6JI} z2X`A%rX7-=x(WR+?x{Q4N8DfhhI>|wETv(Wa2bT~oI&CTE}s7I_pY6#u0EiV{m;U} zf%G6NUgCv{jO-SzZW5im}O@;3OoHQFW2>tr(zbVU30kwRl=n&tMZC{IwqV5#@N!gpnQJ(*pMP@PlF0cTSp=SVUjF8ayfxnYD2u^oQ=W`IM$OQmAm!z0B06ZZfAHDBc;z- zI>(s%Q6HM!b% zSUs$mm4F~yFTz}fW;miF%ss%*+djaSYXX_VIjG!kxz*oHDU=4ubyj*gjz z?OY(u6Mg}}_8=V=__$5~XmXTbmQfxIMu>)6p|}@~f!s*5;W@TI@9<+2TL#Meh!oG@lYsLdjp%CR1he3vJ1mplcNv26Yl+G`Shb=BvYL`0hmzL}V>M+CfjZgP721#_^e zgidkMcL?=24+gIUNraNyEID!;t+shu9q9ko-6;_`JBM+TSB0@3|DPIq68mcP7PqgV zHkBM2mY~Z) z-uhgwp>%i+Q`P`W5h>D)N$NznQ0wfV;|-mTOf4s@AQbCyc`A*>B-n1kmgyVHHH{x}vFg=dEt~m7Sa15>L zpg(x>QN%?C!|o*Ut8X>T(oYCIVFc)h?rc!)N6rl*a}yPd-M(xD$`Do>z0N9O|Lqhi zMG$2Q$Z!YM4oRzuG{UEaHe%je>1`RU7R~%sw)CMfEBiTVZ{;zSIW5z->w2de5m1J6 z6~qDuRBnbkS1>5S7^kU!OQGMWT$$kZA$uk3c`UpC=M4m@|Enz^vLdk8aazU7eZ36W_Wus*+Bm~SK71_)Z1#IBnFs8uK48+Sbw0n>Szd(o6Rdx zA5ub|G&?redNf|#_Dew6w^bjORr9M+uOD58fO*b9F}1aOCm7YQgp(g+`G;X4+`fNJ zh0+W#0JCYzpVU>IzAcM+)dN2bsJ+7P@tVCN=gQ+%qsW|4g0=<8?I2K`1A-BXyYIUVL!pW~`8uWzZc>C3OsDwcew@&mZ0bk@i~_w)(zu`F5_ z?#1jw?}z5c80+mD&u7O;8KK~T5qDsf(m_Z=p=u))cixQ-&%?BAD+6fWVQVoyosts} z*}$3K=0(K6Nc(pJ;`CF7@W*vN?wN_R`39GC))ylBRd>dTjd8ceR(%BZ_ApQuV8#6) zN<62=G#_a4eearLh(2wgZv~@}@$<`}ZZLNp;b%iMzExYsRK!5geW3z3N~3x3iVQ%1 zGdo7{U8y$`y~DG@-9WhT*BYa6pkxVqf{i;k!W9E3RX=*2_B$o@^L95p_73y}BzOPOLWzC}TB>tc#t&ucF`+yE=p@HH*o7AkSm`Li@DKDMplTbX!>H z;dQ{!X1ZHPcTEk|sG=CUXYOT@OS|R%8t%+-%h4~q8h?XE_G>0HDI4E8x$gUg)mRdh zEN-q(Kgv~?R!@KKabX^>=jxN`ibOGpl<8TZQ4}`a$_o8s54p->-}syE+RT?N2Am$f z9&Wd?NsEusAK{z;b1&JWk^aUJ~GZ_)T3-t4p7(*j~{RGD( zVHV(viEv1RhL-|8-mq+6L9ThbWr`;!Wr;wSIGOZbyk8V2N_(Ss-ll}5qFTm)L0iLK zdQKn*oG`!-3V``LRn@G9a@!=O6<>>|OE13i(&R>Cxn(`I=*j?+n~J(dr)^%=_6okw2&729+pfBYqV08L2YnBfqwYJ`*Bn+@E4xIg2oN* zZR`tUCcBgaI*nl)VjIc&v_k$ea4m;mM4p+`^LQ-U zHW!A!D57l-BAsrS*JAJ-^(GSB$I7z*Zh*Ydc-cLFKeLDWaj|~ zme|)QFn#53#$N&@I)BlabxA8`8~k-6a=M7G0d@!qQx z?M<&ONSQe0ONeQre%BI;MmnK>`4*CW_v2fpg}s_3@)1x&qb*~EbGZX9L|nmB*Czu3 zr$KN{TygDqeTm3)MbivXUH!8v!nz8g;DT;G9Zm8osRen-1 zbCUDWZZWSuw|2}=oabdmU`NPV)BuI%^}OpSrihE0zTE9;)rmvD~fk2k&l01b~SIMvqkZW%j{d(*7s2 zRPZW7vSwqzd|$caPp6io2$ofCR;EdyzZMs-sx!0DnB-T$F2sl;6AB!a5FE?m8{cJM z!mxZS^VH2Muew~tA(7K^MkrkhJ(LC6$Zw2-UK}mT&YaTp(3SIo)YWNopc-0PF(}n- z3UKz4gOu?gBIecBNO*idi%1*vH{cTc^ntc-T>Y<#`Z5sk$VPy0e_%yMxeyj-w94`fm97c~z0=bY z>gv>?8f@1fVIARS=Mi7BjsmA`JF32sy8|qhX-yR$F)Vl4Raj$b=c5DK3g7{}L_zDK zn+;2>wW(+rRqY+Q{^WD6C|CYiw;YVDyE!$XgKm(=+X4j2 zmOA)~6vf}P{vQB;K!CrHj#&LxSv>&&J$}d5g@#d<+US!TJE2h2Yf1>R(wSH|leo7= zxlpvShR)Uz8ai9W4V}jwUZzVA=EF-bi4M@U2bY{TNkQs}PiKCBWmguRH`1@2stuSF z1YB1Pa7h8h4l*_@KimWz>8|aO)7&19S!X7+JKswe2+V1FD_gQHH~U4OO`I|tI$vOr zUNoTx7>QqKwJ!%okUUou8o)SS*bncifVJpEQx5WWQ7O>4yi$So?QOtsW2{NEmqgw` zB$DQm&p;Id34xOclZv9yd{Qb@Ht_<(yB%PMdP-PHJU4XJi@pUZC3V#}lUFI#eY0`Yg-_Rx%&KwiRgSNq)zp*f-Q|38DsYYOv6eaEk*ZK=?%smKA+)6RO#!< z?;7l9U;fSsMGbq5 zuxD<}8~HLyt!~os2ouaj!Gx5th)G!5X4uvjHPiIBE(j7<2=>Qe@~d{wWTHB5^1}(Y zKi|tAe&!Qr1m}UPAHM(FwYtm$!392qt80Y&0u1?B791vChNFQzLMm`fO+jPO&u!-u--;7}FBLaWo6Rp?9m%@02D82M!>ChF2mf+Kn= z;{?@Ps6)WmdDviYVk#Oi)vP$AcpRDqq#kEf$Mh|{7kP(CHjz-e6zKz`VRYzUQ>yxe zyOw$?<4=APZ_l2qw`cFE@JT6F>F<^&boZIn9U!QhuIeWiyZBQ>rNF9&}{! zvb3l>q$k}A6qRzSJdr|7^L%}8r5_J(RmVPHOm1F=@*kEA28+R=Lib*K>jvRDK0dQt zuf-#6wOAIs^JG=yxW;jf;~K})IG&k{qsMZi6qXzNkJcbmv(y?As?w!hncG((*l%ts z0=c2uz7#@?z)6Br5j&JKBb_-LEZd5bg3GuHN!YU{4ow^Y7A|3hBeH z0Bt^)d!EdfHPU6-Q0}BCqUbg0x}+>WMr)vm`ZElK$Ut@~b%sr)QRF{ zFw!$v;W7Ur!VnxR+Ug*;3pQYi*h-!i`gL>ga14~MswH^*OT?l(pB@)Jjdf=fYk4g5=duN1!hVsn+nP-cgTQcoGbU2RCh854HM=doKIo!{j_;VoX}zKcN^^){D~hE$Tu6U zX6w^$6Zu)=&_x0?u&)93m;O`~X2Lev6!st#@((kr^CXrelIXwGaw{F(`Oq$){?~as zAt_d3fO?=x+3XLprS&FxkGg!1%2fYug{r?%qI$E@Ihg8?%ZTqZRY?uhe1W2IeSeJF z1J=w7Va2G41D^HBWY}dt1Nl$m(%)CIp+xg~tCRk;^O791vJ(ItWP$z-5VjQkn56d-3`{U$R} zzUSUQ!=tTe_VJ0sp^AF~ac@wEKy|!SsVmgb0MT2tZmav9fXI$AgBjB4?2*av7wY@>FZ!WTWqpdwQCC7VTD%U>5P~yoxds9Y|nF zswk}iYhC^NSLR?%8_~~2fyBa|`SXu|30OgUQdCF48zV`1pJZ_x8$E49E^F1mH-tlT zM;5;sO?;pU^mX8!a+(fgw+fc2A@6G4=qL4bVkFFCGy=n9`_fb6tu0=eEo`LX#D*2c zWNMajC#nJ0$t00%`-dvZmgBo#^%)9=x$2y99milzfkSLV{1e33Ux`n(a=#Gl-#j?`%0rzb|ci^}Z*2f`lpE-~_p_5SlYJ zXKK#WoT)iebLRG(dGrTOn9fZOnuj9I>Qadws(6Me!_)cHy2I!Tm1juVMr>L%|7tb$Uq9F`x>T~ z+eFB)3Dn{1Gus)fl+K4+J_NQCCkN$<2kAZPi=V`

HrLbmELAivSNBpWirfP>9)R z#d?DrRUWXHdv+2{UJl{aw8$|=tD-|TtSHdmNp+Z3sp>GvbYCKBF9i4{(5Fz&TP^8V z-@Y2HhH2Dd7|cY`b*?KfQF-1Ru&mqPA{bdPYD_)}cbILY`C<)U{Ngbq)5US+j-a4$ z1+}Q7w;F@TB8!uD3Rx!Lm=@F+|9$QN5TRgV(;zRo49*Ht5XIgjAy6J1B@6hgw3I%) zT@ObhslD{$Z|X(FTg9)l+>vPGQ$fG#h;dkp!U2fz+e5}&luH}R>-(Fo7U?>-4yL+f zi)7r`Sc^5p`Hno&nXr8C*q=rDT_RIOlks(6SV;u|c`r-d%Zl$)_TaF6E8tzE*bHJ^ zHJch&SESg$lhb;z@aA_!mc5Xfq~B*vc6dXy%IWN;1O!rLMx%EiHQ@r5f?w}l8+*lUfl{>yk5Ri zlY=B!6@=zNLpJz|S2|#uKwPAgED71I+Tp<((H^6A%FJ5|uv}V<~ zv*;d81nkd0{t2RjU;DwW?a9^vAxPRM&R+R0pgv%D1b2EnwQn6ybPo8@8B%*t6tNc| z!5@Qp7}e6{UxWw{GvQXEC)|Vh-LlcdamN5Wbn%3{GfF6$;ww;wkQ2bhu*agj3tkqE zCcf|HvCH!ZAF@$u&&`~|<|)6pJ}NDI>9zf>Irg=EKkmsMMX+>dZzK6~QQ% zR1WuGS6G*6G>_KiljFlwQB63G4W%ss)B%`W=2F@${4YLXI>#D?q_~EKe2rTriocUW zXFLhnlWY^3Cyb;*_=rTkmsaO76I1q-J~iNPS?NKB79~BuyX=Z28g$|QfZAg{e?cE- zRpe_dB^nBC*)U-gW^*=+4z1Q=Ca&6#hQ{j$n-!tIkmN!*737vcV{J4+o3F~iRyhpc zzP5zb1OAPhh>C$wZn)9x<@b7r#gLCc)p2=4EpMKUI{v7k%aLtwrWNv2ago1rk&dmC z#q%|KV_bKZsybTXFbJ$*;#zog3Nbhgzri|22e7bayj-}NRD4zrSV(5IfQ5+l=O6#- z7zu2d8nz0Y#zBM~QIW{OShJ_>&p-Yf80r(9i-pSB69(F;56USnHj}k2lJ#U5QVP%( zS2`t;#6!G4znv`lypT9QOp*IvsNcV*-;27t?CQgd_wTP+BFECNCNy6v`PghK9%ff$ z9>5GB{Q8>oif>;{BmOa|w;&^nzwe0O<4=CaieESP_|ETt_>t;A;aQny6zpt|!d{-5 z{T}xl!{Mh#zxD|wK{_c7yM84WD!p#6l$)hNT%mQZM2tqSdCHpiqQa}uyRf>caeEDJ zzd_`QHN+a}lL-*CY8_N0LEdWc?*aT9Bq*}ypb$myky$gzUU2h2|DSA$P=g)GCbkod z16)t(*Bx>ji;9M}izZcJskQI%*pfx{L*pJPzA!SpB@FI}fUeakjB~VR=KymlQPXIr z(N3dX#b}p$xg!O=K?%?soCoL$@zQvw@lNAi#ds&*hjN;Ifh_U>>{QE-oF^75?6CEzp3ogG3Z*1lsgl(d0Q|SFJpuw@;BVE; zPLuitO2=g=J1eM$3N`kL&!SYp3(?j?7RV4uV8RFc;lm4Dvxtk#b$d3Wc=E(0TROKu zQ_C@>bV&iJ$$W!0EqeN3OtRpLLccFNcSo(U6k2N}x?#v=t{zmLBRBf@iUx0a}GWdad%G+_m zf2${mjGl})2WaKg%fO2$1Yl<0W269a97rXMbIwIMad;kRmMMXb1vxh{RqlcVGapWi zc-j0HJ(953UhxB&SV2v6$3t|5>l?&k$F7oXw6@T}UWjw6iU%jeo|+ytJ!*Q)Js66G z;N$G3JIaG$ai%=wN)06{@8p0!g-2sMqFnUfyt;a9femaSgQ%`fg->Eo7c&0sn@Ac`imHoI&CmAbR-ViCRjRX$49f4=AOj{!B{f~#B%ONLOsRSZc2bA zwtMriND{tND>?gqNh%|z*g>)e%p}pq!u{wbAy=gxr56IOC9%xtF^dJ(RD-02N2e7FwPyufVZ}W{}@yh0W=$(g7}Y1t--%E`0p`kMx&xe#bZT9>cywVMWZ|}V9dT| z?AtfW`1Xy1`u5{$-07O%>}sLrx3Q7`{W$8X+wRDp_^vgcM}R_jVzWdaQy$1#nLDYl zerOyv6C56B9A54sr)rifQv{X*U+Ni0I61XHsp@9$q~NH!X0prAb0K2v6(#lEhY_EO z+G@AlgJZ}%On1Kc_SI;{;uCK_|0cG{fZ&t8&PP=_ zy+-sAWz+*s&r@rN#an&vHpE%MyNIdGBZ0T7^jrR{h_?>qv8#0b^OS+K+WO~%s_h^~ zvstEHU3%pYh4;DAw|QIZ3UPNY8gFm^9$XE?8i-E}h*!L7mAdz!Ogp2vTt~UgLGuVL z&x>u_);SUbxO+IU;63Ls!4BLUeV#lbHJJJ*}V~?V_YEtn%HgKay*p64j`r?uFAv)M* z4Hg*W5s*h37l!w;JCI5B^#KiFWpD9{^lwqO3)#r^vE4nwFEIiKdGx2s4l>6xcODoP ze%*3lm_G%TVWAS|vRG)8ZQs^B0W^HjG7xZ}BH~PJJ+>%i)uU;-82XJ7G(m5C4{3U^ z;O8S5-=e`(gXwvIY4%<4RSyQGIX#Bdev38EPMPLrupEav*NbL0sk4aM?9;g$IU-Me7Htv3}1NUnt&gmM{(mNw}w*qnQYYBVY1 z14ZOw{G>gakVkef^gRo%1mTlA+Y0y;HurcZ#^X2e5@OA0T^p+1&bvq&URYZ zVY#7gB)IG)2{n6%Lal<@%7J-i2o;-al>}62t~Fb$8dk1NYD&Lk4!Torn^Xi3u0t;& zZaDH`Big~3iMZ$b%z*}VyD!Rq1y3pZyl3wV?$zp_0V z-bA33KSc4D0mLcB@dLoC8;-YC?8ER*E zb9)CT_skW=l!O-tWPF$$%Q!Wd4Bl?5Vz&#Z*yR2K;IQ!HJYdFwE<4hSMoctUnlg5W z#hDZwbvTM6zhBtir>88o#;4DVncqAKW!FhmxjoV8&z{+I90D;?ox4&3=YW125?sSS z;U~ZTm1WJWRggrID@{GRB@%~EV({KG9Xy!vQ^3v>2jvZ6-r+wZ?0*KM_k1z1Ex03h zY4gdWBgx}Q{xx^pg_q)S5Ey!&BnGNQpoYQlExUy&p$vv`3^L@Wfu&0|3(I(;p49Ti zu*QBFMg9<7TQFxQM_6V;SY1t6OCHsvg((jgw=|`Nt!FOsmqu`YW$lE*;sb`0Y~coo z_ZrJlR?=gbo6q3-k_QbqybFWYbc#%J&s zl}7u8i9iEnvG_o=m`#_8{QUI8%OAgIKLC8eQd96cWk+!ltoh!D$Jw*37)mnE7#V_= z(h~`cNf9Dn(3T1uqM{nlSPE~%@8_~7!9xMv@Zxw3^Z`p2CQIBpES8Hp^2N6(N%T%2 zP90NWyvVTRU-)QFwyJPGMf6ez9h=3<2rJmWTNV?E{4^0<-4YRG-$VX1yxt=JsOsRy z6^s(|HyRM+{O^KHS}~b)bOvc&2I+GzR&6l5*f6@8pcw*(xG${KU{yBhPeS(YQLp*I z#}&P;p&x-@obto~GJf>Y?)-f=ROX7{<@zCf$vay8%l=I4Ny1; zH8p8!($sWT)YRaY`&NBLTU+Hn*zM6yPD&F$FTOYTEAd@g< zv9pDtjs!0tX90YO%AW1r=MmL@Iq9LfIdWR+wpV3#t%?51h{|!Em);eX7JWF-nwVe^2{isr$PMjha-rZA72 zV{d*O!MhaC^jjct_bPN-K2g<5!y<=R?F}< zqNZkqK_@}vVFt?ZUg$RPxv13%>khcYy<7O$qr4JKFX6(m6Mmv36uir43Wk#gGoltM z2FfYqmwf(Ieotnbs1)g#67a)UJQ&{Bq_=2{C62_cUjzV+zMC7soO&17sZ8!vc zMoBqfINh&*t-pnDZOSkT_ymin0t3GmnMxxPhmV0h=Il9~b12S?UGXon>Lg0eD}Rpm z*_Viua{M9wCH3y>ssiOOOqvCA?-`r}0FUJ<#Y<*sWsqD76?0_uqMx|q0 zBY0V=csHiCutw>%NDnTPFLOJdSuVnd#VKulRmRLPeR3J(oLt^ZQBBPKuP~la?0?iq z<8Y){nx2I0-{bF~l%=wV1P-3);?y*`qe<76TkT~Fnc9ZBO0$Jv^w zdy1psBRhPcf;N(cIKeL+x+*5vK}@iQP6l0K*tec6ek&KOCTWaZMo3o(eOO$3jB#kd zhE%go$P%8jI6?s*i|Acgdf==-|M*vu;%Nr}mLV1lflbK2AlIZ%O_-wf zjw8Xk@aCbckiD>#kzeM-Q!KHm ztYohhp-RJSSAPg<_nT$wIi)7}fc*Jk{y9zI=h}XdR`MPArsWzCAPFgEl+O ze48>~znw6jjwH`LTP`p`n%uBP613f)v?3883&QU}{R$dj4ypEx~d+dUYtCEHSb#38_ zCNsUL1jojrMWyTCVGYLV%=NJa?M$}1&2WTl5G2BwMCG=2{L8p#_$D2D2lZBSrTJHG zyWE|%pbHFnbjLY$=mK}IJBC!jxJ|EDc?JmV?77V6To(6PujQMBZ z{R~T)ig}YkGus7FKIn!lQ$Vk|B7hRe>5~QP0~<8a1*+z{PQrB^9ai%|pRO?yosXM|*jLObtri z0UuYE>{72&DcQBFCcExtT$g^ov*kKrq&|D1%X4mOmOERP>!nz(r&%tqYDgKTTLg*q zG}rE=XbyO#0v1~uKt&Xm8fv8!)~lMr+MSIktlhZ)3R|?(fg&Z-lh2gQ9xzO5UOQV} zOIm{%U2U)|e0rObTPugyRzt^niZgqaQ{0-NR77^EL03t3-O7!VFwl)!l=My_5&N}w zt3bPWe25rqHcUj}!cq}=bV&PiOhW}&g%`lg8}5yTmvT+?#yu>xb>jqKWKDcnK2dSn za(&yYsJkk<&oN=77aRUXyFuXhTQqw(xP#-i0CkW>FSI33+b&@uUUa28L7AcxROebG z0Es9J5XD;z!oZD()elh5fX#J_rR8O{5horU^F8+-J8(Uj0M)=ug}7&D@LuEaN%^Ai z)#Cj?rG0lAx)W4hC%QS4WElR!9jsU6RGY}TpO>H9T_%au;nW}aq{Ro63j4A+A$=^mP)C$BkPo zw@BjRE6Bp{xD8pjLAya3f1UbG4CkoBre3=G<|tBBw~V|2LGhIDdyM3NqSJAgaZf2_B;ohg(7A`gS9^Z*#*SqNLt-)S{ z{q|s=JUxEw(SUrc-++dqdRXFjj2*7?V!${#rQy(ksggoA!_|rSs10HSKTPKhud*;|VXC|uw#I5H( z+TkZroB*%;XpcToV}qaMiy!&{}I^u`19Z5N51}jZi}*1dAzhR6>BzB=@_|Pb<0)TqT*EG(zV$;d$&U6x?AFY zhAPA4fGy86E(yAZT`i4IG(v}!44Kg|nHDf7YD>yPuXJem--jj#{Lk%J-ego2(Q>ngOPBm`ypvbTFrMU@@=Lly@9OSjpaiecr+!e0h$_?n)u zVA>lzfs0%4?X?)&JE1+j6c7P3K5KN^^-f+MZL*4q__tZPE9n2_Nau4AL#Zpl5c#LsEDJz;v*jH4On z(2Vmg-fb%xC!-A)<{R17{q(8EY&M(gYr*IBDNe7%=~se>2MYH-4~B< z6x{k5A0VzNhvRa7Xsl{>)IfHQ=B-_H_9$Js!R+immW=wQ1Wp)h=&ib_@LNU9a52!2 zEo4HptYiaNGuQSiZ~jq1;O#!!a1440*6QTRJP(Y*q_+ujz5~G#E_v0xXa>vpg)qXs zXWhg}POT<}_embV=xG#4KPZ0Q!fgNcIy78-vbu5U)_r&pU~)JQQ(JMj(6@c&<6;Ah zZVA-uo`|bG{>V}=5=UeB?AK2Ldd{CMXrAN=mzB7F{p*u32_aQR>(+H{*FNeXc1Ldg zXG%*19*6(o-q;SL6+ERf?%69p7w>C{AzHQ*DDd8iNB+<*@O}cox;R;$vWrEh5)IPH z*dM0Yh6uYgF7Fqr8V16{r0|3+CHT2*62x`#7wJpaJ+Z8>xU)XRd@E;hYljuW%9IG( zzoMCXJ-nM`dmd=~qqxmQRuTO`RSDoo^PYcpfJ~`CB5H$S#k#4=PqXk8j!9qS9Z90o zK%T{I_hto-QFa7}Dc6!aShQD_uL!#59N%*`IyBx21-TNujWY$i4YV%A@C(SvuHLq37?%h%!ic!Mf-DS&`EOx#Utk3=EHln#~Yafk|Gw#50gHp$za7=M}# zXo}%qocZ%;v?>614Z;*AsswFfq^Xv!;@w!ar5gmIEMX0fB)+M_bn=G?u_Vd$>*nBL z7pxpV{JsYetP0rSwd{&i6tiJ;t}AE8N>3jPyy#fCrFTMv85Bx^O9 zmNB5|c#c(&aZ5N9z=Wxc02t^C$A4$;jMl-siO(=_Xsi;|(oZ6e@_cHu3fB+08ms7r zAjdzm@5R{R((lnTgx${ZZVmPeP_~U=MMg*`FR)4^b^;cr`BfsziV9+tBFOQb4UOiG z*s0*raCu1%S#oSfiMpQ9+T-`XB~1r@hve!nD>JE(wao*o@(ApGkw-2$pI{G(lu?PC zDo;&@-HJ<}Ed?d9c_`iA54ii2iLe$Xd4q7_d-FH~K-L4m{fh)MpW(*%_O;v2oapeZRuGA5*!p}lNy-i4sqD@U#^St| zjDs9_ksaKMfU?m%TC$43UCHKLzR)5 z1QO>lZXw7k6_LS8z;_j84wCIuy{E9-ULRnrH1iC_=E&2DodApSrAlT_ovsl37i_*R@2r$ug5B1}Q#2UHC`({PxR!I-URa$CMT0%GisIId?q^iRUY7b?DryV2ox)*qQluE&hYG5V z?x6f!Y8zF+MKGD6EY5!dbw+c{oFpIu%27B7<>EdBJ&8mWhq$QQ7Cq0sfVC6FVn(3P zfk=9LK3&12xc6lE4lNX<(2^*=i*pr%QJ^F?wC}@}OY>~8c+Nmo%&B*fjJXXP9QLlP z*HFs~7ZUBr?J1IL+&@}$aSlZugm|tHu2NO(`a5Ai)6BA;k+5G+v)@YgTi!zjPG!lV z%zhXsC`e9q%aGy}+qQ=jk;hq~!eBX_SlHpE2xV1g0iXr}k4BNtB`O0_Qd!k0T z=i8}$>v%#%`_UN&6kP&xjW7B!nCDk7Sb-#En~|qpLB7{64^1HQhjq?X@0B|?i{a}j z4{l+Y*;<9G(py&NdeJ8NE}j0uvWz2xN5z38b;!u3`kieCn5xODO4Y(fLg-tll|cBp zbP_#^vsXJCiee_tF;hFoRT=VYN-9|U-M93o_?qlqVG{$1^S<`NhCf;`P&ntp)Axx zkshMs+i!m0!jZ-HJ~3A$%GThDR(c5PNHu078-|NELK?$gvE?_8=nZwe;&k|VmMrZlgI#fJTXQRhtqu9a{_Mb%+3r#&*4K5%`|LYpa*qpa zv>4@AU2kAz#C^GfmmlssI`XlaqAbNC&evX%@c(;zA@PstR! znte$`lw847#D813JVXNPNO&?0Mj)48tXq4gYBTM|U>UE~5Vqf^aH*s>8Y5)}r4#XK zC~IgBSUl$;Ob32FPA2`n-} z*k78{ykp${Mo^BNQnIrZ|HN|pl>MYC%TmmRUdr7T$-j7nACP@fQFWRht`& zQWUP)`4iI`_m?K7b0wz!fr&{8ohGJ(6O(ar%%p_p{FuoocZQ<1JAZ~+Q-ee^)VVU$ z;J^%}gibTm!5OO4IW>k-!gGEM)hQRNL2GyZ47H{dq-LlyWvJ?ERBp&nO6W909sUf} zKDk4blAbd>M77JFqS}{mjA}LOEkl#kg^`qTSdxl`PLtH(NveBlBo#}~8In}DEJ<}Q zf~3~8eAOg%t|Vn1n52}@X_7iTNp((+q?GiWAxU-0l2qp?N$Ncb6H*kP#-*aM4yQ*I z{<5@X>gc{)+A=k59X`6x)X{y#Dl+PZ+)KTTUl5c_B9*gs8MRnyk_ zh5>5Yy7+)VOp`(s6w-3c9waW=dVkM zGSvKa1W|^jjxyBzRXKm1Z@}Ss@Yhd%05b&B;NGjw;8dw{Q8*Tutfd1FbD6BAne6a^ zhbT+JuN^@+V@roK9y_n~jnn0|GcKgqH%jvw zt>p3X+A9JUHs(V|5E|Cr#7TVj{6AhVvRPY)H0H8dTeI2WLmJyUq_HTQrT2Kk8{;$- z*g=58h^!GV<3hlrdEjOELKfb|pLEkDfB~>)a9h}4z*YHms)7aAJsaD9m ze%8nO;9>;OTHs=I;9_h)&^nDXjh&2&FzYGGdSF}x%35GtbYNTr$}WpCea(wHpzQBq z5VMd+g=A|bU;WJmGyE)t^_XTZjEj{tb3wD~8y8&|7whjTyzBRG*$)9p?=)IV5nuuJ zhdZ7qq^+Xu=&Ql)CJd$XJz><{6Rs1a0mP@BRFw6D{o z7I+$6cp5uFGpqAB)7RK&DZW$M>Mhp#;A{leTHtJS;cV;#);zc4Sku_xsA|{udg}}6 zgUbIK2{>m1LCf4ee_^`tc$n@8u0x}g!#`O{{Agb|6CGSoZQhrg{lcrozd@d zk38=ohg!jR{@rZ}e78v0G$1<7i88ZdhhP#xI`%rGMxPOj=0Ozmk7fx=6M5LbZ0YZ3 z-jmjWD~FX*m;Faxnv^}TRf3c)uvL1nRf3dVBqeN`AZ1@Hn;>NiESnxIn;>NuNeK%m zNZA)VC`j1?JE#Xcs8*zOdxJ`S?WL}wz0?IVdtx;OGFxCZ^|qQ)6qat@wkE_sIJ2$ST3}LrDtW3n?gM;P+qtl}aht%LVThxiF9(Z$ici zH9G*)iRTtCaxxfEHG3lQT6F4xFBNoZfiKmEFZIfFGS5uLu8t>t-JhoF^vT5Zsb_Xj z0ICIcP#<>CD+8*7t^PH`3V827WfH$_zABJD7VG)C!M#mRGmP%?VLgV zMB(9oh@*gR^(_l#@;E^(;k5sZCG$D;kj(?b`w*VJN0TMPVtL-_q#O*YGLyNWh_y@J#zsxOVhw5?xWLBOpAUcn)}f~|&|ZFY9vrmxeG z8b$FrViPmBet8Z7w-$H~hwvP>d*!9sZrG6CR(w5aQpDncZ`yOyV>rsYs)BNGXxsGm5>m)E>cY1gm5v+(v-g^{tX4SS9Pi?L$Bkb}J;{ zyXZE`hy$-k9$r7958Q>HZ4%ytSinq+gn5-M9;yaE%6_Ze|0b2>XRFX3v8ArZzM)o% zMb@0y^>ZCs;7tg&{1FG=8fVc@&g3AlsHZ*{n#`aNL(HCR)cIwsxt0?f+2CwKL-e4AOG2Xt?fo^!k zcz@R;dtmGY)Y{UG=)L@5fC)C8o)TOOp9}DL$y;$C1mqkMy^Us8HvYLM;_}2pGqzUI_ zUH);6;0}8(66XYB@K9_AN1`G7z4&w|9MqDPguCo%;e9OaE@mkDLqv-KX0S>EX05PF zny^Z?24TIe_G(=##D9vZ=a*$?t;u?@TBgmvsDGIcVZU-g$oJ|K34KL2~B^r*Lni6ddoMEFwXpMa?7KKVpqVwo>QKHY7 zPax57m^2{K@R`(;XnSzNt^Z_+*4MpZtNQAT=)%8o4TwRQr1O6y^_*_(IGPq0lq=uZ@-R=0ZE;F zL$k^LHqDX=edpXerCW88Ft&7e=fvI_TEMC{TU?G=1$#@ESY>*yhP2Yy>f@;vlRXz! zA8-ripU1cb<4>2kWm`JVvh+1UQF|$_4q66GP~aA9QIBy8mZ&aq%kp*PmcHgNYA;3O zfe&XGxCO7+W88vItXtf2HRP7gjyYGATdo1S9JmDo-DBK>d9G93(y_j!v5nADt#4VL zetSTzHwm2}oQ+QaA-78(dZ?*+DoD{m*Z4SEP2=rMUU4NE^CpDr@G@5MqXvKxQQ0}Z z3!uz{M|PPy`Z~OM*mMdk01H~Nan%_X!07iR7FgazVW6+|%~E~9$nx~t$hI$;PuN-NlQp|LA(f{iT^yTO*S0EhD^Ve+udI`Cb@zaoG8Fa#~y_){hv3Q)@I53C8fVe$9+@1Wry8>x>OL~Qy=GANViR- z4_5cAgrR5A#~ixG-1Ho4IGek+?-`!u&%TxyM&r5h{=y#e=KZd+KPYvn0aNQy(&H+% zQ1sfAYulR&T9wgTB-Z{HYS7}>wY#9he#Vk*mQoiG*Pq|i4bRE^^HZRUZj)?T1-7bY zem1A5M^3{Me?;G`$tY@&4)N(sf%~d_qBBz~i!V9=REq#LdIV51uk+bjI_CP5H{&>4 z(4E{ayf4zs@9iVnb6CLSV79_$+-f29~XEKkGZ zxB+kN#T&Icb-y&bD&Wjr=H!1+nmHX1se~Z9u}~&3*W?wG`o_|AxDrbS^nG!Fdq%;L zAs64HDS6{8`j*ZRe$H4-!HwB=MSNh>?S{8e-X8}!fB3SaxKTopynuJ&E_1Y;U|8wC zpU#{rR8YfH9HVu_v+0tsFsYl&qfJVq35sl(33gau$8*?G<^|88Ns?uWT!C6fvsoN3NtX2vGuS0p7!1suVQKk-RPAkaUa2eKodf(Icv+iq6!D-@bdlu}0S^o2UU>E4*$ z6WC(k$jM0sTaFp>k1Yj8QY7MIrQIp3`N7jY|5DK(iX-sf2aL-ksdM>{#RWA@h3|g; zVZh$YjQs9?70}#+u5@xQ`pI`8HATjAk&#lfpOei;0goTTm7EB?q6~h`7L?uTg2t(02sV(t zjL$SEUZ2AIa8~L=1LltYu>VFrgsbR?MIXHjH-{ZFSMZAd`KAG=$Sb)i9bKHQ$n+{= z$h#big#*g%;F{oYM>WAgt#Hk?wD7Y)bHHX6oeK@>|) zic2S5;V~9^nalJ$R+6$C@u>);$XIHm(}2k{NS?Q1x>DR_0i#kV4wU0!bCBI-zhk*{ zvRk4(UjSo=0Xc~Zu0Gr4F)X}CCKGh$V}9r3WIit4^+mYhN^XD;&*4L`rsn)CMh)?O z#R}7*dLJwhvtSddX8b@R?vdLy-^8SRMTG*rz{Ef(is%`+qBOaj z^q}!{T3iwl9%uJ!sR%5XZR0GhvGgYH&GB@=Lx&A?DAC{(p?w2`yAc<(_RJ#EaFRn z`xK>yO(S7V5^{Q%(e)NB!zJ4jwrAfeu_W;=afM4Q>6{5PTAbZ{#5PTT{1-7SmU*O_{QnS8G{&JQ${n{<(S~6@O6563hP&SNiv|vs?1t z1gnp{kITt;Nb^xjl86S@Vtw<$fhWAH$*~Rpl?dlGpSLDN2@uefARl&dGUSffQu~7u zPi2H@SOdo}8qnENb12E^elYSubgDyf7DM@9T9owItKWXzF=$Yfv|nFVb}2?qZQU&b zs=Mrp6+Dwo7m*AWB)Gp4e4GoA<0s3p>o+UM$4#S~G`qfnH0Qy^0tWP0PdzRt07Ur7 zBJ7L?RwE)@ZV%m**BM#$l{b&5_R5PmGOs4s|a&4I|0*r#*b7gr1Rc?zak5JBQ zQiMHO^G7xBwB~vWCJL_~7K)c$wSfAzSADT$(UcO8+=fQO{V3#(-1-CIJhUj|L5g*h zVifTj(ch+5jCB-Ug+=0Ctr24^3;{;{uEF}9Wez;l?-c5Hmg&{6-{sLo{SNZnH+e?& zt|{ZOIq*@vQ#2o&zJmE!1{l@5=Zf+WqTCiy9-?~JqzJo4Lu(fLR_|P+{*px=PE_xp zzWq>NELb$=qwrC^Q}I;z^{aPzXi>d`6zeF(sNOZ9zrKSf@E#Lg->LrqULH?W?;yZ^ z5nv348?#b)4ed$Mlj7aTtG_2j9$S>|pv`^LW|Zz4D&0}2QMyws-I<<3=`Ig2N_Wo{ zWgnv47E$(5x@%H|9dj^3>F%uNI_9Xhf0*t|MV|6pv}G1X4LH(+E(7YgsnpBx}C4EYn$JXUE40N$PN^_EsE@*Zr7v&TV6w}75dig zEU*4zMIKJn?V!H>P+u%nH2CB*>UJtmK^rPhL8ER5Db`hrQMYSEe@6{373(OvBL&eN z^=Nq*V_^s|>UIr2{A>(7)a?`=gf_hT4?@e>qHYIm?wdBFZr7Nt!b9Co(OKcCI4eBV z?HW*I8;aZ(MYd75YqV}>HT1-@#&tWZ{u9o5I8nEQ`u0P8u~gBR5|2=~Q&DAwC1h)a zx*eogTPa4}u93Q(J#bOCd!%m1-4b={b~2u*+d+W)BET37H}>$e-SEpsb>Dr$uD>Tx z9$VDypv`^LX4LH(vQ=1@cpO^mx}9s)U$4l+i@M!&71@F!w?&aH)a{y7U~`DN-6M)? z4(r$L@^GSV2lefT`eLb~vAUg&_uHw|?dvoo} zAi91(L>FUW2r%k)O`9k@?31ZvsPGgXg!ZscCbYSi+KjqgL$(Uk?UbDr3UxaNb-QL1 z*@PmuMUhR^?HaAySpySwyYqEB%dB6wli@_&4(i(v^~F*}V@m9yZl|KkUi~HHJhZ6W zL5g*hV$|&#aM{)|fLyDxMaX`TKVd)7N8iM8au;x6c_^u01QUK$<8gitN{-LtVE2OqwGAjFQ<70? zy{(LYO|AWjE-3uGjspt*hRu>N6_?M|E{o>(;)27~EqzM2L7ieof1wUceJ|Xsmh>eS zlDaQR)|B*ecvdgz=-ZFuU>((x%#`k&$I}|H2_+3li928^T^hdtW>yR1bZUoF%L*s1 zJ9nl-%b(hYIhlI?#PEW^cZOEzjC>=U`=Lpy={e41Y7ebhFg0eAiRIQ7#>&=a-xndT zP`talyF%OcVg9-lt@P_WBW@@oZgv@Qvunr4K0nmH_bZm?``~nb%FLz&UD0;};xBao zu5csqZedpL22bq*1Nv#UD2~kc!Qn@a=!8i|@E|3i(M*9vj{^`O1Wwu?>ko)EQe{RIbdpcg=fO#R#)DHZ$JMXzI>N z3VFjnXUT-GI*32WB_~FCLg${bqRDc}$%9XP4EZnk0qH_k3H5$T0^W~hyNc7E(5fsO6yz~k(HD-XmJ4q#x~DkSPZ04HTN&4qX0;AQf*|9F zq%x9==rv$u8c(3nE{R@@N4xAFC)-Ta2soQ~T2x0(&OXz_jLI)d2J0@_e5BL~KL^Wo zETWbF>v$E&!~d9t>qWEED6wQZ&Z`{s*WZ3M?SWx?Zyck#(83r0<8a?ssi(}MPtj}} z#HA{U%%pS|lVG)?FEG(`JZT?v9YF?)B=2|>D!MO`6#v);gu3j0UoFYUx1tsiG_*3v zKRRfWnzOl|4_b`9;<<8TGv32CQ4qg=$ATpv0mzl^G`d~Uzwztd9W5s!v2R8{;VArP zxJh;i5>djCXmRka(rmNk%a>AP7`6X&8wb0g_?4%E-^&=J>{Yaz-@GP0P43v2f2tHq zmaO7?8f)y58Z14-(r;1;Q2L)_i{v3!PwH3uYkox?*c!OL@tV<1v%6#)&)6cMAN=;~ z$bLg0qTW~PNrWYuIBL$&bIMV{%wR)(lGH&KWM6uIb~{;@aVQ$CCJgf?1_S%R!0Lr< zVhm3v9+bT0*GWoEMm_1@_KOyLnWuE9uNj}O^6pi-k!K+_&nn$Hx=ARyG_KwcB}&Oc z!I>_C)ou6%klRJ%{yDiLaoG%bnu&=0{>#61?gHX)_z&Kl%P>3c;rxRNzzHz(2`=Sh z809QKEt@5xQcQOouabiPaJ*Jo4w!i#g<8jx5u1aTrmMl`CBRZgXVrG&be z1L+h0EI%$Lv6@U`-{-EfVlA>F11O?gpk>MsW{V65WgD2-5>AV$?VvQZ6(PTih-3LA zcW7XS+pQVy(5Owtp#dc7j6}?7F{cHAuQ<=;(Ubn4{XBOP(gB_{$>o#eIJ%9vOF1cv zy;i4<#S=bLcASB57|Vra`2jZ(^G_HxReCUsrl|98?dRKK{;fHLPfJ&u>B}3oCVd?T z^wpcbrM`6NSqFRpU%(ge?ZOQB?9(=2@6*^z64(TG{~FAq;02hQ@6*-l?%7lw^Fq#Ys`{sDz9>j&fR-8;hvt zTDHRhLkHx0tn~(4LL@*e5bN5+3PiW1lzb+#p0V#Hs0C_)T5YG+UMO0REt{IBKDb0{ P61)B%Z~TJTWZ?+_D#p&< From 3b1abdaacac6e667323384f68f94e6d4f9aabba1 Mon Sep 17 00:00:00 2001 From: Max Ghenis Date: Wed, 4 Nov 2026 14:00:13 -0500 Subject: [PATCH 37/52] uk_hmt_tables NOTES: quote Table 6.4's footnote 2 (review A T-6) Footnote 2 (cell F25 of the pinned workbook b0449df1...) sets the table's scope: it leaves out measures that interact with this Budget's new measures and measures that extend an existing policy. Brief 82's hand-off left it open because neither review quoted it. Co-Authored-By: Claude Opus 5.5 --- sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/NOTES.md | 1 + 1 file changed, 1 insertion(+) diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/NOTES.md index fc81488ea..2173ce099 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_tables/NOTES.md @@ -40,6 +40,7 @@ The two XLSX attachments of [Supporting documents for Budget 2026](https://www.g - lines 15–16: `autumn_budget_2024`; - line 17: `spring_budget_2024` (`data/uk/events/*_measures.json`). - The cells are £ million. The table's units header prints "(£)", and its footnotes give no unit, but every value cell carries the number format `\+#,##0"m"` (I9 = 7130 displays as "+7,130m"), so the rows are `parse_confidence: high` (review A, T-6). The sheet is read by name ("Rounded Table 6.4"), not as the workbook's active sheet. + - Footnote 2 (cell F25) sets the table's scope: "Measures that interact with new measures announced at this Budget, for example on savings income tax changes, as well as measures that extend an existing policy such as personal tax threshold measures have not been included." (review A, T-6). ## Conventions (load-bearing) From 6e06907cbb148ae2ac36a6ea99b0e0054f624b88 Mon Sep 17 00:00:00 2001 From: Max Ghenis Date: Wed, 4 Nov 2026 14:00:13 -0500 Subject: [PATCH 38/52] tests: pin each receipt's policyengine version (review A) Review A noted that test_each_receipt_matches_the_dashboard_commit doesn't compare the registry's policyengine version with the receipt. It now checks execution_receipt.wrapper against {"name": "policyengine", "version": receipt.policyengine}. Run against the dashboard checkout at 1eca3182, all eight receipts pass. Co-Authored-By: Claude Opus 5.5 --- tests/test_ab2026_registry.py | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/test_ab2026_registry.py b/tests/test_ab2026_registry.py index 57d3eb128..223996351 100644 --- a/tests/test_ab2026_registry.py +++ b/tests/test_ab2026_registry.py @@ -261,3 +261,4 @@ def test_each_receipt_matches_the_dashboard_commit(key): ) assert ex["data"]["dataset_revision"] == rc["dataset_revision"] assert ex["data"]["dataset_sha256"] == rc["dataset_sha256"] + assert ex["wrapper"] == {"name": "policyengine", "version": rc["policyengine"]} From ac4d14a1fd4bc833a58dece831d17a554e3e0190 Mon Sep 17 00:00:00 2001 From: Max Ghenis Date: Wed, 4 Nov 2026 14:54:45 -0500 Subject: [PATCH 39/52] Apply delta reviews D2 and D3: costings row 584 and red book row references D2-1: C-1's receipts marker also reached costings row 584, p.108's count of about 4.9 million properties in scope (proposed tax_base), and gave it a receipts component. The stager (uk/ab2026-harvest-staging e928465d) now keeps the marker off that text row, so the row matches its 4f424b9 version again and the family holds 163 components, as its NOTES says. D3-1 and D3-2: misc row 8's note and the misc and devolved notes cite red book rows 191 and 252-255, the rows the staged file holds (83d5a32e). D3-3: TPA's NOTES drops a stale sentence on the charts' aria descriptions. The red book NOTES says its review section numbers rows as 4f424b9 did. COUNTS.json moves with the two claims files. Co-Authored-By: Claude Opus 5.5 --- .../harvest-uk-ab2026-2026-10-29/COUNTS.json | 4 ++-- .../uk_devolved_govts/NOTES.md | 4 ++-- .../uk_hmt_costings/NOTES.md | 1 + .../uk_hmt_costings/claims_staged.jsonl.gz | Bin 63483 -> 63475 bytes .../uk_hmt_redbook/NOTES.md | 2 +- .../uk_thinktank_misc/NOTES.md | 4 ++-- .../uk_thinktank_misc/claims_staged.jsonl.gz | Bin 4014 -> 4014 bytes .../uk_tpa/NOTES.md | 2 +- 8 files changed, 9 insertions(+), 8 deletions(-) diff --git a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json index 622f9666b..313b84e9a 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json +++ b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json @@ -19,7 +19,7 @@ "uk_hmt_costings": { "rows": 597, "primaries": 1, - "claims_sha256": "66a8e7fe77952438747bca6dde96be71a2f1deb8cf88716bf4a7bbb8ada7242e", + "claims_sha256": "4769087faa0876ca168f1976ac16ce23129617cd7f1ae35c406f75761fc9ea32", "sources": [ "hm_treasury" ] @@ -75,7 +75,7 @@ "uk_thinktank_misc": { "rows": 12, "primaries": 5, - "claims_sha256": "20695e34914c2b80fc58b92aa87b7a096bc974a004939b35a5fae5a135aa685b", + "claims_sha256": "ba9ecc122125e3eefecc0c1433ef5cd9b24e91f59d3fd2c2e63e18a2b51c2069", "sources": [ "citizens_advice", "cps", diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_devolved_govts/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_devolved_govts/NOTES.md index 5e5886de7..00288945e 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_devolved_govts/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_devolved_govts/NOTES.md @@ -19,7 +19,7 @@ No primary was re-opened on 3 November: the session's network policy refused gov Review B (PR #231, issue comment 6635498758) approved this family with fixes; the coordinator's brief (section 3.3) settled the Barnett key with review A's K-7(b). The fixes sit in `review_fixes()` in `stage.py`. No primary was re-opened, and no value changed. -- **Finding 1 (Barnett):** the shared key takes the red book's scope. HM Treasury's £1.3bn (red book row 258) is the total, and each government is a part (rule 9: "a part can be ... a nation"). So rows 0 and 8–11 are `measure_component` with `component: "Welsh Government"`; rows 1–7 name `"Welsh Government: Resource"` or `"Welsh Government: Capital"`; row 0 has `fy: 2029-30`. The proposed-key scope below is the red book's. +- **Finding 1 (Barnett):** the shared key takes the red book's scope. HM Treasury's £1.3bn (red book row 252) is the total, and each government is a part (rule 9: "a part can be ... a nation"). So rows 0 and 8–11 are `measure_component` with `component: "Welsh Government"`; rows 1–7 name `"Welsh Government: Resource"` or `"Welsh Government: Capital"`; row 0 has `fy: 2029-30`. The proposed-key scope below is the red book's. - **Finding 2:** row 13's locator says "third sentence", and its note no longer repeats row 12's "next financial year" sentence. Rows 14–18's `source_column` is null, as on rows 12–13: the labels were ours, not printed. The Scottish manifest note no longer names the old `uk_fai_sfc` folder. The sentence on the Welsh package mapping being "unresolved" is corrected. ## Coverage @@ -37,7 +37,7 @@ By `measure_key`: **null 19**. Twelve rows carry the proposed key `ab2026__packa ## Proposed measure key -`ab2026__package_barnett_consequentials` (12 rows here): the scope `uk_hmt_redbook` gives it, the Barnett consequentials of "decisions confirmed at Budget" for the Scottish, Welsh and Northern Ireland governments together, 2026-27 to 2029-30, excluding the technical reduction from lower employer pension contributions (red book paragraph 4.150). It crosses Table 6.1's England spending lines and has no line of its own. HM Treasury's £1.3bn three-government figure (`uk_hmt_redbook` row 258) is the key's total, and each government's figure is a component. So every Welsh row here is `row_type: measure_component`: the £448m ("an additional £448m of funding over the current spending review period", `fy` 2029-30, cumulative) and the four Table 1 Totals carry `component: "Welsh Government"`, and the Resource and Capital cells `"Welsh Government: Resource"` and `"Welsh Government: Capital"`. The Welsh rows exclude employer-pension-contribution savings, as HM Treasury's do. Compare the Welsh £448m with HM Treasury's £450m for Wales (red book row 260), component with component; never add either to the £1.3bn. +`ab2026__package_barnett_consequentials` (12 rows here): the scope `uk_hmt_redbook` gives it, the Barnett consequentials of "decisions confirmed at Budget" for the Scottish, Welsh and Northern Ireland governments together, 2026-27 to 2029-30, excluding the technical reduction from lower employer pension contributions (red book paragraph 4.150). It crosses Table 6.1's England spending lines and has no line of its own. HM Treasury's £1.3bn three-government figure (`uk_hmt_redbook` row 252) is the key's total, and each government's figure is a component. So every Welsh row here is `row_type: measure_component`: the £448m ("an additional £448m of funding over the current spending review period", `fy` 2029-30, cumulative) and the four Table 1 Totals carry `component: "Welsh Government"`, and the Resource and Capital cells `"Welsh Government: Resource"` and `"Welsh Government: Capital"`. The Welsh rows exclude employer-pension-contribution savings, as HM Treasury's do. Compare the Welsh £448m with HM Treasury's £450m for Wales (red book row 254), component with component; never add either to the £1.3bn. ## Welsh Government diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_costings/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_costings/NOTES.md index 34a7ad8fd..1a7800f55 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_costings/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_costings/NOTES.md @@ -416,3 +416,4 @@ Review A (PR #231, issue comment 6635870934) approved this family with fixes C-1 - **C-6:** p.84's precedent is the AB2025 uk_obr_tables family, in NOTES and the row note. - **C-7:** `stage.py` is on the staging branch. - **T-3 (uk_hmt_tables):** `uk_hmt_tables` no longer emits `conditions.fiscal_head`, read from the workbook's hidden check column. The sentences here that reasoned from it (the p.41 bullet, lines 9, 48, 81 and 84, and the lines without a costing note) now rely only on Table 6.1's printed Tax and Spend heads. +- **D2-1 (GPT-6.1 Sol's delta review of these fixes):** C-1's step had marked p.108's whole costing record, which also feeds the note's text row, so row 584 (the 4.9 million properties in scope, `tax_base`) had become a receipts component. `stage.py` now marks only the six Exchequer rows, and row 584 matches its 4f424b9 version again. diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_costings/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_hmt_costings/claims_staged.jsonl.gz index b76df060f6bb933ab2b71f80c7d4423eb03f924a..2bd03d8b07ccb318ed92e4567e67edb3e84a5325 100644 GIT binary patch delta 1895 zcmZXPc|6qlAII}kj+9~9O~j0=btgX?JIFCf2QeXMMC6zq)|IQ~8#yA^HX}zON8^}8 zF=iRWQWJ(2F%0V*6J|6T$4rc|`2Bt!k6-)W>-~N|U$4*Wv&j-%VF}h=Sq4H5LHbKe za2_YD6)p88UY?T<=B3W9|Dm#^TQ_8$`dTwckhQ$_Th3EAVzbt79pekevC!FCW5LjZ z8wZ_ahNI*&mB@>c6e+f~V+uJcjz>k$ma53Fo;%7$`96Qx8=^=I`Gt}bnK)2-S!$@$V=1ts4pvMet}(u|pg(Z; zeA-fgcMfk_;g;#B0c5w-rH?V_sg-kL%lT<5Eze!T>`$e|ve_?Y?wB~z-Q-R-i?-MTsVa)bUkS?D%VO&2%pnwGA7E1XjQ!yh zZ<~yYXg~2+H`AiQXeyITRR=iX|O+MmYW|mA5a`&O-H+NH|9(^hcjzD`Jn?Ht4MsV zZ=s9TnK)qPy*5V(ZG_Au2${sd?!cmgUX<2$H>{xlDgvPQ8h3v;L9N zw)*skF8adHuG!#^$qMJ7FxOcKSlDHD?RX@XVVKC5`s$Zmjt)1ZU>8CV-h^*XjbOc}6QSV3T>l0uGt$a9>%M zz!Y)lH^#Z|AwAfQpYO-2Mc>oU3W}r$_!S3FK){X|;xPK*ijuUC(5JWEcKvE^`*oti zJ{jo6*A$V_?JLs8P3*)Ay>jA8?pFaeeQ(OqjtB)MDvUb?EaVV^V_JA(0xVs0xw zv^&oje@U=+F@(B>V&xbi_NL`e(2}bEu;oLhy(}jr^-co2zHTtv6h!-klisN1gfYcm8Fl`j0GQ&{_k+*wp71S34up%_ z%1rU4aLtC~+f7UmzRJ`g0Q|4@` z`NQ3`Eugq+r)=%E-H{4l8p$)ZIc-|WyxpR%*Zv?~?e}H4jI?QOJSyh}t(&Fa_p2=C zu&;7nEd*n4Xn6YV%=nwv5u*6RABgH}I^@vkV(6_B*=|R3yGL=*yb<2JrSq>Tr;3ek zmFScfQZnfcKvYt{OaKRKxzNLFbu+h)jFUpH{r0k*wwN0`h z_;wkIJUi0md!YSHW2f%48T_Pjdx|rH8=#k^`i1t_shSLau^63TYjx_4=PDh6Lwa35 zUv84pRcJm_*fCw`HC?DYwYgs4E6nH*uO0CXG|yQmXS!bH=%{is2h4KF2V4N%A!a_1 z9G_`XSwF9Eb8tA%&cJv0P-Xh@urJ84bYPp%%30-T^1f zqfh82?jr|H<0qlW60;pguP`bi7{03{d5Z#X8Y3+rdnr4}ZYbf(=lNaw&Xf~@WB8x2 zLp(ac@W*`$wud#4b}{!UCt{`U{+sJMC-<>E3Wfb&n4AA*8*BRuTOz&aH2A>Z2~_Rd delta 1813 zcmZ9Gc{mh^8^?*HvIuD%HI9m>Mag}vS%#GeV@7h8$u+3im9xP}jwNd{$T~asJY!O%7!qf6tUOULeT#ZL>K)Nf2s_RQs$bAZpif~dz ze9#7$uMU!Qjg6!p*SR)m=y;{FnS&!;&ZiyT8llh8Vz)lk%}?5v>YX&Pt86EC=uf%w zbo7|b?@=mOdnEY2LcHsvnoZBj{N@`@b_Z2z7=WFMLuAx`_}D{)#spg5jk_SSA6GS0#bxm+tLGgjAnuF|5f7;32AE z0K8Xl>^2FAS8$YD^!tUV=E$d+XC*%6)y)G-Uez_qO zEHlFA{D@$(GG$!OX1y|MPIh~>^tQ1mcOp0gLHVX0b#`JmFXV!q<8e=OUyr^W$IZTL zDaZQ%((vh(k8#l%j1G`@V}nDfVWQpY9W6mqFG$~{*miTdjxTNdX58=vXwpPv*|~Ke z2#Ixr$jT2y$a=RE9`b=Se`?ZPLuaL(!tGqA#Z ziyLD{JUy4SaBqC~j#5_v+c^QL8WE%YJoWJc+?RluQ~)HV8-2q*Whhy-3oW*F+V`rq z^=gDaTGYE9OU~u_iOg>%{N9Fks3+S!t94{Vw$FYwJcz&1Ehnnv=AbwGuAqYZ;xkj^ z%%)Aj-AfNp4`^ZVl9KC=6LnF1y0FfrYWblyWd1Ip5TC(rR;Rl zzb!9Q^$3 zceEd!1Qa3Z>6XI)Ln50=5Qs(*-S-JMIn8CjR-QNqv?^BqouHL5C6D%1R3KEEWdcEy zfpIYFl`PSOKGe#)XSl-Af(;>~ z0r7_;UW~gk*%9~8VNf9Y5dvrSF9M7OG$C?y!3u|t1%ZYS6xAao9L Date: Wed, 4 Nov 2026 14:56:14 -0500 Subject: [PATCH 40/52] harness: apply delta review D1's checker and test findings D1-1: value_raw "1 in 4" let 0.01, 1 and 4 pass, because the checker compared a ratio's two numbers at every scale, and the 0.1% allowance meant for a rounded share also let a GBP-million cell drift (-2,626,312,500 against "-2625"). A ratio now gives only its share, and a year after "in" is no denominator. Only the share of a ratio or a worded fraction may round; every other printed number matches to floating-point error. One staged row used the old allowance, OBR row 2662 ("a third" as 0.3333), and it still passes as a rounded share. D1-2: the balance test accepted either deficit's name in any row's sign. It now reads the deficit from the row's label (its aggregate, table row or the stability rule) and requires the sign to name that deficit and not the other. 21 rows measure the current budget deficit, and the red book's primary balance outturn measures the primary deficit. A new test swaps the two names in every balance row and expects the check to fail. D1-3: family_problems raised on a pair whose value is a string, on a non-object row in proposed_keys, on an unhashable primary_sha256 and on a manifest line that isn't an object. Each now gives a problem. D1-5: the README's FAI bullet cites red book rows 252-255 with their paragraphs. Also from D1: a test pins one row per line, or total, and year in all three tables. New cases reach the blank-component arm, the exception wrapper, the worded-fraction parser and the rule 4 locator arm. That arm now reads a locator naming Figure 1.A or 1.B as a bar reading only when it names no paragraph, table, box or data label that prints the number, so red book row 357 (paragraph 1.7) stays valid if its locator mentions the figure. Removing any of 14 checker arms, D1's fixes among them, fails a test (a local run with a hypothesis stub drawing 200 seeded examples; CI runs the real library). All 12 families report 0 problems. Co-Authored-By: Claude Opus 5.5 --- pipeline/ab2026_harvest_contract.py | 113 +++++++++++----- .../harvest-uk-ab2026-2026-10-29/README.md | 8 +- tests/test_ab2026_harvest_contract.py | 124 ++++++++++++++++++ tests/test_uk_ab2026_harvest.py | 119 +++++++++++++++-- 4 files changed, 322 insertions(+), 42 deletions(-) diff --git a/pipeline/ab2026_harvest_contract.py b/pipeline/ab2026_harvest_contract.py index 5078cae4c..ebd6727a4 100644 --- a/pipeline/ab2026_harvest_contract.py +++ b/pipeline/ab2026_harvest_contract.py @@ -108,6 +108,17 @@ (re.compile(r"%|per ?cent|percentage point|ppts?\b|pp\b", re.I), 0.01), ) NUMBER = re.compile(r"(? 0.25), whose two + numbers count for nothing on their own, and its worded fractions ("a + third"), which a stored share may round (delta review D1-1).""" text = re.sub(r"([-−–])\s*[£$€]", r"\1", text) # "-£1.7bn" + fractions = [] + for (start, end), share in reversed(_ratios(text)): + fractions.append(share) + text = f"{text[:start]} {text[end:]}" + exact = [] for m in NUMBER.findall(text): t = m.replace(",", "").replace("−", "-").replace("–", "-") try: - out.append(float(t)) + exact.append(float(t)) except ValueError: pass - for a, b in re.findall(r"(\d[\d,]*(?:\.\d+)?)\s+in\s+(\d[\d,]*(?:\.\d+)?)", text): - b = float(b.replace(",", "")) - if b: - out.append(float(a.replace(",", "")) / b) low = text.lower() - out += [v for w, v in WORDS.items() if re.search(rf"\b{w}\b", low)] - return out + fractions += [v for w, v in WORDS.items() if re.search(rf"\b{w}\b", low)] + return exact, fractions def _factors(normalization): @@ -175,30 +201,46 @@ def _scales(raw, normalization): return set(SCALES) +def _number(v): + return isinstance(v, (int, float)) and not isinstance(v, bool) and math.isfinite(v) + + +def _same(a, b): + """A printed number at its scale, equal up to floating-point error.""" + return math.isclose(a, b, rel_tol=1e-9, abs_tol=1e-12) + + +def _rounded(a, b): + """A stored share of a printed fraction, rounded ("a third" -> 0.3333).""" + return math.isclose(a, b, rel_tol=1e-3) + + def value_problems(r): """review A H-2: `value` is finite and is value_raw's number at its scale, - with value_raw's sign unless the normalization documents a worded sign.""" + with value_raw's sign unless the normalization documents a worded sign. + Delta review D1-1: a printed number matches exactly, and only the share + of a printed ratio or worded fraction, the fraction itself, may round.""" v = r["value"] if not isinstance(v, (int, float)) or isinstance(v, bool): return ["value is not a number"] if not math.isfinite(v): return ["value is not finite"] raw = r["value_raw"] if isinstance(r["value_raw"], str) else "" - numbers = _numbers(raw) - if not numbers: + exact, fractions = _numbers(raw) + if not (exact or fractions): return [] # nothing printed to compare (a categorical label) norm = r.get("normalization") or "" - scales = _scales(raw, norm) - worded = bool(WORDED_SIGN.search(norm)) - - def close(a, b): - # stored shares round printed fractions ("a third" -> 0.3333) - return math.isclose(a, b, rel_tol=1e-3, abs_tol=1e-9) - - magnitude = [(n, s) for n in numbers for s in scales if close(abs(v), abs(n) * s)] - if not magnitude: + printed = [n * s for n in exact for s in _scales(raw, norm)] + if not ( + any(_same(abs(v), abs(p)) for p in printed) + or any(_rounded(abs(v), f) for f in fractions) + ): return [f"value {v!r} is not value_raw {raw!r} at a stated scale"] - if worded or any(close(v, n * s) for n, s in magnitude): + if ( + WORDED_SIGN.search(norm) + or any(_same(v, p) for p in printed) + or any(_rounded(v, f) for f in fractions) + ): return [] return [f"value {v!r} has the opposite sign to value_raw {raw!r}"] @@ -236,12 +278,15 @@ def read_manifest(family_dir): def proposed_keys(rows): - """The proposed measure keys a family's rows name (README rule 2).""" + """The proposed measure keys a family's rows name (README rule 2). A + malformed row names none; row_problems reports it (delta review D1-3).""" return { r["conditions"]["proposed_measure_key"] for r in rows - if isinstance(r.get("conditions"), dict) - and r["conditions"].get("proposed_measure_key") + if isinstance(r, dict) + and isinstance(r.get("conditions"), dict) + and isinstance(r["conditions"].get("proposed_measure_key"), str) + and r["conditions"]["proposed_measure_key"] } @@ -369,12 +414,13 @@ def _row_problems(r, keys, earlier, metrics, units): or r["source_model"] != "policyengine_uk" ): out.append("same_assumptions rows are PolicyEngine-computed") + locator = pub.get("locator") or "" if ( r["source"] == "hm_treasury" - and "impact on households" in pub.get("title", "").lower() + and "impact on households" in (pub.get("title") or "").lower() and ( r["value_kind"] == "approx_chart_reading" - or re.search(r"figure 1\.[ab]\b", pub.get("locator") or "", re.I) + or (FIGURE_1.search(locator) and not PRINTED_AT.search(locator)) ) ): out.append("no HMT Impact on households bar readings (README rule 4)") @@ -393,9 +439,12 @@ def family_problems(family_dir): return [f"{family_dir.name}: missing {missing}"] out = [] manifest = read_manifest(family_dir) - shas = [m.get("sha256", "") for m in manifest] - for m in manifest: - if not SHA256.fullmatch(m.get("sha256", "")): + entries = [m for m in manifest if isinstance(m, dict)] + if len(entries) < len(manifest): + out.append("manifest has a line that is not an object") + shas = [m["sha256"] for m in entries if isinstance(m.get("sha256"), str)] + for m in entries: + if not (isinstance(m.get("sha256"), str) and SHA256.fullmatch(m["sha256"])): out.append(f"manifest {m.get('url')}: sha256 is not a sha256") if not (m.get("url") and m.get("title") and m.get("retrieved")): out.append(f"manifest {m.get('url')}: needs url, title and retrieved") @@ -409,7 +458,7 @@ def family_problems(family_dir): out.append(f"row {i}: {p}") pub = r.get("publication") if isinstance(r, dict) else None sha = pub.get("primary_sha256") if isinstance(pub, dict) else None - if sha not in set(shas): + if not (isinstance(sha, str) and sha in shas): out.append(f"row {i}: primary_sha256 is not in manifest.jsonl") out += duplicate_problems(rows) notes = (family_dir / "NOTES.md").read_text() @@ -461,6 +510,8 @@ def duplicate_problems(rows): out.append(f"rows {idx} share one claim identity (more than 2)") elif len(idx) == 2: a, b = (rows[i]["value"] for i in idx) + if not (_number(a) and _number(b)): + continue # row_problems reports the value (delta review D1-3) if not math.isclose(a, b, rel_tol=0.005, abs_tol=1e-9): out.append( f"rows {idx} share one claim identity but differ by over 0.5%" diff --git a/sources/harvest-uk-ab2026-2026-10-29/README.md b/sources/harvest-uk-ab2026-2026-10-29/README.md index 227182455..62fc7cc40 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/README.md +++ b/sources/harvest-uk-ab2026-2026-10-29/README.md @@ -51,10 +51,12 @@ The AB2025 harvest README's rules 1–10 and its row contract apply, with the AB ## Row contract additions -`pipeline/ab2026_harvest_contract.py` checks the AB2025 row contract with these additions (review A, H-1 to H-6): +`pipeline/ab2026_harvest_contract.py` checks the AB2025 row contract with these additions (review A, H-1 to H-6, and delta review D1): - `value_kind` may also be `categorical` (a rating, such as the OBR's Table 3.3 uncertainty ratings). -- `value` is finite, and it is `value_raw`'s number at the scale `value_raw`'s word gives (bn, m, k, %), failing that at the factor `normalization` states, failing that at 1, 10³, 10⁶, 10⁹, 0.01 or 100. Its sign is `value_raw`'s unless `normalization` says the sign comes from the wording. +- `value` is finite and equals `value_raw`'s number, to floating-point error, at the scale `value_raw`'s word gives (bn, m, k, %), failing that at the factor `normalization` states, failing that at 1, 10³, 10⁶, 10⁹, 0.01 or 100. A ratio ("1 in 4") or a worded fraction ("a third") gives a share, the fraction itself, which `value` may round within 0.1%; a ratio's two numbers don't count on their own. Its sign is `value_raw`'s unless `normalization` says the sign comes from the wording. +- An HM Treasury Impact on households row reads a bar (rule 4) if its `value_kind` is `approx_chart_reading`, or if its locator names Figure 1.A or 1.B and no paragraph, table, box or data label that prints the number. +- A malformed row (not an object, or a field of the wrong type) gives a problem, never an exception, and the checks across rows skip what they can't compare. - A row that carries `conditions.component` is a part: `row_type` is `measure_component` (or the OBR workbook's `measure_total` marker), and the component isn't blank (rule 9). - No more than two non-restated rows share the AB2025 ingest's claim identity, and two that do agree within 0.5%. - A registered metric goes in `metric`, never `proposed_metric`. @@ -90,7 +92,7 @@ Each family's `stage.py`, which writes its claims file, is kept on branch `uk/ab - `uk_cpag`: both CPAG pages (its Budget response and a pre-Budget call) returned Cloudflare's challenge (HTTP 403) to every fetch on 29 October. - `uk_smf`: the Social Market Foundation page returned a challenge to every fetch on 29 October. - `uk_niesr`: NIESR's Budget-day reaction ("Breathing Space, Not a Strategy") was fetched, but it holds macro and fiscal-stance commentary for the #55 lane and restated OBR figures, which the AB2025 NIESR family did not stage either. -- `uk_fai_sfc`: the Fraser of Allander Institute's reaction refused the fetch (HTTP 403), and the Scottish Fiscal Commission published nothing in scope on 28 or 29 October. The reaction's feed text holds only restated OBR figures. A second FAI post, "Rising or falling? Why both claims about Scotland's funding can be true" (29 October), was found on 4 November through FAI's RSS feed and sits behind the same 403. Its table of Scotland's Barnett consequentials is not staged: FAI says the split emerged late on 28 October, so it probably restates HM Treasury's allocations, which `uk_hmt_redbook` rows 258–261 stage. The Scottish Government's five rows, first staged here, moved to `uk_devolved_govts` on 3 November. +- `uk_fai_sfc`: the Fraser of Allander Institute's reaction refused the fetch (HTTP 403), and the Scottish Fiscal Commission published nothing in scope on 28 or 29 October. The reaction's feed text holds only restated OBR figures. On 4 November, review B found a second FAI post through FAI's RSS feed, "Rising or falling? Why both claims about Scotland's funding can be true" (29 October), behind the same 403. The coordinator left its table of Scotland's Barnett consequentials for tranche 2: FAI says the split emerged late on 28 October, so the table probably restates HM Treasury's allocations, which `uk_hmt_redbook` stages in rows 252–255 (paragraphs 4.150, 4.154, 4.157 and 4.160). The Scottish Government's five rows moved from this family to `uk_devolved_govts` on 3 November. ## Producer slugs and families diff --git a/tests/test_ab2026_harvest_contract.py b/tests/test_ab2026_harvest_contract.py index f90b523ae..1950d6e06 100644 --- a/tests/test_ab2026_harvest_contract.py +++ b/tests/test_ab2026_harvest_contract.py @@ -10,6 +10,7 @@ import gzip import io import json +import math import shutil from pathlib import Path @@ -150,6 +151,19 @@ def test_a_staged_row_has_no_problems(): ({"measure_key": ["ab2026__x"]}, "measure_key is not a string"), ({"publication": None}, "publication is not an object"), ({"value_raw": None}, "value_raw is empty"), + # delta review D1-1: a printed £m cell is that many millions exactly + ({"value": -2626312500}, "at a stated scale"), + ({"value": -2625100000}, "at a stated scale"), + # delta review D1: each arm the dedicated tests didn't reach + ( + {"conditions__component": " ", "conditions__row_type": "measure_total"}, + "a component row has row_type", + ), + ({"conditions__component": 1}, "row is malformed"), + ( + {"measure_key": None, "conditions__proposed_measure_key": ["ab2026__x"]}, + "row is malformed", + ), ], ) def test_each_rule_broken_on_a_valid_row_is_a_problem(changes, expected): @@ -176,6 +190,66 @@ def test_value_raw_scale_words_set_the_scale(): assert problems(worded) == [] +def _share(raw, value, normalization="N in M -> N/M"): + return mutated(value_raw=raw, quote=raw, value=value, normalization=normalization) + + +@pytest.mark.parametrize("wrong", [0.01, 0.04, 1, 4, 25]) +def test_a_ratio_is_its_share_not_its_numbers(wrong): + """Delta review D1-1: IFS's "1 in 4" (its row 38) is the share 0.25. + Neither number on its own, at any scale, is.""" + assert problems(_share("1 in 4", 0.25)) == [] + assert any("at a stated scale" in p for p in problems(_share("1 in 4", wrong))) + + +def test_a_worded_fraction_is_its_share(): + """Delta review D1: "a third" is 1/3, which a stored share may round to + 0.3333 (OBR row 2662), and nothing else. A row whose fraction went + unparsed would compare nothing and pass whatever its value.""" + note = "'a third' -> 0.3333 share" + assert problems(_share("a third", 0.3333, note)) == [] + assert any("at a stated scale" in p for p in problems(_share("a third", 0.5, note))) + assert any("opposite sign" in p for p in problems(_share("a third", -0.3333, note))) + + +def test_a_year_after_in_is_no_denominator(): + """A year after "in" prints a year: "5 in 2027" gives 5 (here £5m) and + 2027, not the ratio 5/2027.""" + assert problems(mutated(value_raw="5 in 2027", quote="5 in 2027", value=5e6)) == [] + + +@settings(max_examples=200, deadline=None) +@given(n=st.integers(1, 50), extra=st.integers(1, 1000)) +def test_a_ratio_passes_only_at_its_share(n, extra): + """Delta review D1-1, for every "N in M": N/M passes, and N, M, N% and M% + fail unless they round to N/M.""" + m = n + extra + raw = f"{n} in {m:,}" + assert problems(_share(raw, n / m)) == [] + for wrong in (n, m, n / 100, m / 100, 100 * n / m): + if not math.isclose(wrong, n / m, rel_tol=1e-3): + found = problems(_share(raw, wrong)) + assert any("at a stated scale" in p for p in found), (raw, wrong) + + +@settings(max_examples=200, deadline=None) +@given(cell=st.integers(-9999, 9999), off=st.integers(1_000, 10**8), up=st.booleans()) +def test_a_money_cell_passes_only_at_its_exact_value(cell, off, up): + """Delta review D1-1: a cell printed in £ million is that many millions, + to floating-point error. £1,000 off any cell under £10bn fails: the 0.1% + rounding allowance covers a share of a printed fraction, nothing else.""" + raw = str(cell) + quote = f"Line 1 | 2026-27: {raw}" + exact = mutated(value_raw=raw, quote=quote, value=cell * 1e6) + moved = mutated( + value_raw=raw, quote=quote, value=cell * 1e6 + (off if up else -off) + ) + assert problems(exact) == [] + assert any( + "at a stated scale" in p or "opposite sign" in p for p in problems(moved) + ) + + def test_malformed_rows_are_reported_not_raised(): """review A H-6(h).""" for row in ( @@ -196,6 +270,28 @@ def test_rule_4_holds_whatever_the_title_case(): assert any("rule 4" in p for p in problems(row)) +@pytest.mark.parametrize( + ("locator", "bar_reading"), + [ + ("Figure 1.A, decile 3 bar", True), + ("p.5, Figure 1.B, top bar", True), + ("paragraph 1.7, p.8; discussion of Figure 1.A", False), + ("Figure 1.A data label, decile 3", False), + ("Table 2.C, decile 3", False), + ], +) +def test_rule_4_reads_where_the_locator_places_the_number(locator, bar_reading): + """README rule 4: a locator that places the number on Figure 1.A's or + 1.B's bars marks a bar reading; one that names the paragraph, table or + data label printing it doesn't (delta review D1: red book row 357's 20% + sits in paragraph 1.7).""" + row = mutated( + publication__title="Budget 2026: Impact on households", + publication__locator=locator, + ) + assert any("rule 4" in p for p in problems(row)) == bar_reading + + def test_a_component_with_its_measure_or_a_proposed_key_is_valid(): """README rule 9: a part of one measure keeps that measure's key; a part of a measure with no Table 6.1 line keeps that measure's proposed key.""" @@ -330,6 +426,7 @@ def _rewrite_manifest(fam, edit): (lambda m: [{**m[0], "retrieved": ""}, *m[1:]], "needs url, title"), (lambda m: [*m, m[0]], "lists a primary twice"), (lambda m: m[1:], "primary_sha256 is not in manifest.jsonl"), + (lambda m: [*m, "not an object"], "a line that is not an object"), ], ) def test_each_manifest_rule_broken_is_a_problem(tmp_path, edit, expected): @@ -344,3 +441,30 @@ def test_a_family_missing_a_file_is_a_problem(tmp_path): fam = _family(tmp_path) (fam / "NOTES.md").unlink() assert family_problems(fam) == ["uk_test_family: missing ['NOTES.md']"] + + +@pytest.mark.parametrize( + ("extra", "expected"), + [ + # a copy of the base row whose value is a string: the pair check + # compares the two values + ([{**BASE, "value": "-2625000000"}], "value is not a number"), + (["not a row", None], "row is not an object"), + ( + [mutated(measure_key=None, conditions__proposed_measure_key=["ab2026__x"])], + "row is malformed", + ), + ( + [mutated(publication__primary_sha256=["abc"])], + "primary_sha256 is not in manifest.jsonl", + ), + ], +) +def test_a_family_reports_malformed_rows_without_raising(tmp_path, extra, expected): + """Delta review D1-3: family_problems reports a malformed row as + row_problems does. The pair check, the proposed-key scope check and the + manifest lookup skip a value, key or hash they can't compare.""" + fam = _family(tmp_path) + _write(fam, read_rows(fam) + extra) + found = family_problems(fam) + assert any(expected in p for p in found), found[:5] diff --git a/tests/test_uk_ab2026_harvest.py b/tests/test_uk_ab2026_harvest.py index e8b7e1821..3818f4e1f 100644 --- a/tests/test_uk_ab2026_harvest.py +++ b/tests/test_uk_ab2026_harvest.py @@ -9,6 +9,7 @@ manifests, and per-family pins that the ingest's exact accounting will read. """ +import copy import json import re from collections import Counter @@ -210,6 +211,52 @@ def test_table_62_rows_are_financial_transactions_not_revenue(): assert abs(total - summed) <= 25e6, fy +YEARS = {f"{y}-{(y + 1) % 100:02d}" for y in range(2026, 2032)} +# every line or total each table prints, by (table, row_type) +CELLS = { + ("6.1", "measure"): {str(n) for n in range(1, 99)}, + ("6.1", "total"): { + K.package_key(s) + for s in ( + "package_total_policy_decisions", + "package_total_tax_policy_decisions", + "package_total_spending_policy_decisions", + ) + }, + ("6.2", "measure"): {str(n) for n in K.TABLE_62}, + ("6.2", "total"): {T62_TOTAL}, + ("6.4", "recosting"): {str(n) for n in K.TABLE_64}, +} + + +def _cell_problems(hmt_rows): + """The (table, row_type) groups whose rows aren't exactly one per line, + or total, and year, plus any group the tables don't print.""" + out = [] + for (table, row_type), ids in CELLS.items(): + staged = Counter( + (r["conditions"].get("table_line") or _key(r), r["conditions"]["fy"]) + for r in hmt_rows + if (r["conditions"]["table"], r["conditions"]["row_type"]) + == (table, row_type) + ) + if staged != Counter({(i, fy): 1 for i in ids for fy in YEARS}): + out.append((table, row_type)) + groups = {(r["conditions"]["table"], r["conditions"]["row_type"]) for r in hmt_rows} + return out + sorted(groups - set(CELLS)) + + +def test_each_table_cell_is_staged_once(): + """Delta review D1: one row for each line, or total, and year in every + table. A copy of line 1's 2027-28 cell over its 2028-29 cell (both zero) + leaves every row valid and every sum intact, and fails here.""" + hmt = rows(HMT) + assert _cell_problems(hmt) == [] + copied = list(hmt) + copied[2] = copied[1] + assert _cell_problems(copied) == [("6.1", "measure")] + + # --- README rule 10: stated effects of the Budget's decisions on PSNB -------- BORROWING = "decisions_effect_on_borrowing" @@ -319,16 +366,50 @@ def test_a_measure_takes_the_borrowing_metric_only_where_its_producer_says_so(): } -def test_the_current_budget_memos_name_that_deficit(): - """Every row that proposes fiscal_balance_change names, in its sign, the - deficit it measures (selected on the metric, not the label: review A - H-7). The memos by label are among them.""" - balance = [ +DEFICITS = ("current budget deficit", "primary deficit") + + +def _deficit_measured(r): + """The deficit a fiscal_balance_change row measures, read from its + producer's label (its aggregate, table row or fiscal rule), never from + its sign convention (delta review D1-2). The stability rule targets the + current budget.""" + c = r["conditions"] + label = (c.get("aggregate") or c.get("table_row") or "").lower() + if "primary" in label: + return "primary deficit" + if "current budget" in label or c.get("fiscal_rule") == "stability": + return "current budget deficit" + return None + + +def _names_its_deficit(r): + """The row's sign convention names the deficit it measures, and not the + other one.""" + measured = _deficit_measured(r) + sign = r["conditions"].get("sign_convention") or "" + return measured is not None and all( + (d in sign) == (d == measured) for d in DEFICITS + ) + + +def _balance_rows(): + return [ (fam, r) for fam in FAMILIES for r in rows(fam) if r.get("proposed_metric") == "fiscal_balance_change" ] + + +def test_each_balance_row_names_the_deficit_it_measures(): + """Every row that proposes fiscal_balance_change names, in its sign, the + deficit it measures and only that one (rows selected on the metric, not + the label: review A H-7; the deficit read from the label, not the sign: + delta review D1-2). The red book's primary balance outturn measures the + primary deficit. Its current budget outturn, Chart 1.6's stability rule + steps and both producers' memo lines measure the current budget deficit.""" + balance = _balance_rows() memos = [ (fam, r) for fam in FAMILIES @@ -341,13 +422,35 @@ def test_the_current_budget_memos_name_that_deficit(): ] ids = {json.dumps(r, sort_keys=True) for _, r in balance} assert all(json.dumps(r, sort_keys=True) in ids for _, r in memos) + assert Counter(_deficit_measured(r) for _, r in balance) == { + "current budget deficit": 21, + "primary deficit": 1, + } for fam, r in balance: assert r.get("metric") is None, fam - sign = r["conditions"]["sign_convention"] - assert "current budget deficit" in sign or "primary deficit" in sign, ( + assert _names_its_deficit(r), ( fam, - sign, + _label(r), + r["conditions"]["sign_convention"], + ) + + +def test_a_balance_row_naming_the_other_deficit_fails(): + """Delta review D1-2: swap the two deficits' names in any balance row's + sign convention, the red book's current budget and primary balance + outturns among them, and the check above fails.""" + balance = _balance_rows() + assert {"current budget", "primary balance"} <= { + r["conditions"].get("aggregate") for _, r in balance + } + current, primary = DEFICITS + for fam, r in balance: + swapped = copy.deepcopy(r) + sign = swapped["conditions"]["sign_convention"] + swapped["conditions"]["sign_convention"] = ( + sign.replace(current, "\0").replace(primary, current).replace("\0", primary) ) + assert not _names_its_deficit(swapped), (fam, _label(r)) def test_every_borrowing_row_says_which_way_it_runs(): From fe58796c683759616f4d60e8890167df6662ff3e Mon Sep 17 00:00:00 2001 From: Max Ghenis Date: Wed, 4 Nov 2026 19:25:53 -0500 Subject: [PATCH 41/52] Answer the final review of f27b927c: ratios, tolerance, malformed input, rule 4 The contract checker now reads a ratio in any case, hyphenated or in words, gives the share the ratio's sign, and reads a year after "in" as a denominator when we write the ratio in `normalization` (F1). It compares printed numbers within four units in the last place, where f27b927c allowed one part in a billion (F2). It reports a row without `value`, an integer too large for a float, a line it can't parse as JSON, a file it can't read and a field nested too deep to print as problems, never as exceptions (F3). Under rule 4, it reads only the locator's first reference, and treats a negation or a stated bar reading as a bar reading (F4). We added tests for each fix. On f27b927c's checker, 37 of the 123 contract tests fail, and each of 33 one-at-a-time mutations of the fixed checker fails a test. No staged row changes, and all 12 families pass the contract with no problems. We also rewrote the harvest README in active voice and cut its review history. Co-Authored-By: Claude Opus 5.5 --- pipeline/ab2026_harvest_contract.py | 229 +++++++++++-- .../harvest-uk-ab2026-2026-10-29/README.md | 122 +++---- tests/test_ab2026_harvest_contract.py | 312 +++++++++++++++++- 3 files changed, 554 insertions(+), 109 deletions(-) diff --git a/pipeline/ab2026_harvest_contract.py b/pipeline/ab2026_harvest_contract.py index ebd6727a4..0a0f06a79 100644 --- a/pipeline/ab2026_harvest_contract.py +++ b/pipeline/ab2026_harvest_contract.py @@ -13,6 +13,7 @@ import math import re import sys +import zlib from collections import defaultdict from pathlib import Path @@ -64,6 +65,18 @@ } TIME_BASIS = {"fiscal_year", "annual", "point_in_time"} CONFIDENCE = {"high", "medium", "low"} +# What a malformed row raises when the checks reach it: a field of the wrong +# type, a missing field, an integer with too many digits to print, or nesting +# too deep to print or encode. The checker reports these as problems (review A +# H-6(h); delta review D1-3; final review F3). _number turns away an integer +# too large for a float before any arithmetic, so no check overflows. +MALFORMED = ( + TypeError, + AttributeError, + KeyError, + ValueError, + RecursionError, +) # README rule 6: HM Treasury and OBR material is under the Open Government # Licence and quoted whole; every other producer's quote stays under 15 words. OGL = {"hm_treasury", "obr_efo"} @@ -109,16 +122,72 @@ ) NUMBER = re.compile(r"(? 0.0005", "N in M -> N/M") +# (final review F1). +NUMBER_WORDS = { + "one": 1, + "two": 2, + "three": 3, + "four": 4, + "five": 5, + "six": 6, + "seven": 7, + "eight": 8, + "nine": 9, + "ten": 10, + "eleven": 11, + "twelve": 12, + "thirteen": 13, + "fourteen": 14, + "fifteen": 15, + "sixteen": 16, + "seventeen": 17, + "eighteen": 18, + "nineteen": 19, + "twenty": 20, + "thirty": 30, + "forty": 40, + "fifty": 50, + "sixty": 60, + "seventy": 70, + "eighty": 80, + "ninety": 90, + "a hundred": 100, + "one hundred": 100, + "a thousand": 1000, + "one thousand": 1000, +} +_OPERAND = r"\d[\d,]*(?:\.\d+)?|" + "|".join( + re.escape(w).replace(r"\ ", " ").replace(" ", r"\s+") + for w in sorted(NUMBER_WORDS, key=len, reverse=True) +) RATIO = re.compile( - r"(?|→)") +# README rule 4: a locator's leading reference, up to its first comma or +# semicolon, says where HM Treasury prints the number: a data label when it +# names Figure 1.A or 1.B, else a paragraph, table, box or data label. A +# locator that names Figure 1.A or 1.B anywhere needs such a reference, free +# of any negation ("no data labels"), and one that calls the number a bar or +# chart reading reads a bar whatever else it names (delta review D1; final +# review F4). FIGURE_1 = re.compile(r"\bfigure 1\.[ab]\b", re.I) -PRINTED_AT = re.compile(r"\b(?:paragraph|para|table|box|data label)s?\b", re.I) +PRINTED_AT = re.compile(r"\b(?:paragraph|para|table|box)s?\b", re.I) +DATA_LABEL = re.compile(r"\bdata labels?\b", re.I) +NEGATION = re.compile(r"\b(?:no|not|none|without|unlabell?ed)\b", re.I) +BAR_READING = re.compile( + r"\b(?:bar|chart|graph) readings?\b" + r"|\bread (?:off|from) (?:the |a )?(?:bars?|charts?|graphs?|figures?)\b", + re.I, +) WORDS = { "half": 0.5, "a half": 0.5, @@ -136,26 +205,49 @@ } -def _ratios(text): - """The ratios a string prints, as ((start, end), N/M).""" +def _operand(text): + """A ratio's number, from digits or number words ("1,000", "a hundred").""" + word = re.sub(r"\s+", " ", text.lower()) + if word in NUMBER_WORDS: + return float(NUMBER_WORDS[word]) + return float(text.replace(",", "")) + + +def _gives_ratio(normalization, n, m): + """Whether a normalization gives "n in m" as a ratio: the generic "N in M" + or the printed ratio, followed by an arrow to the share.""" + printed = ( + rf"(? 0.25), whose two numbers count for nothing on their own, and its worded fractions ("a third"), which a stored share may round (delta review D1-1).""" text = re.sub(r"([-−–])\s*[£$€]", r"\1", text) # "-£1.7bn" fractions = [] - for (start, end), share in reversed(_ratios(text)): + for (start, end), share in reversed(_ratios(text, normalization)): fractions.append(share) text = f"{text[:start]} {text[end:]}" exact = [] @@ -182,7 +274,7 @@ def _factors(normalization): f = float(num.replace(",", "")) except ValueError: continue - if f: + if f and math.isfinite(f): out.append(1 / f if op in "/÷" else f) return out @@ -202,17 +294,32 @@ def _scales(raw, normalization): def _number(v): - return isinstance(v, (int, float)) and not isinstance(v, bool) and math.isfinite(v) + """A finite number. A bool isn't one, and neither is an integer too large + for a float (final review F3).""" + if isinstance(v, bool) or not isinstance(v, (int, float)): + return False + try: + return math.isfinite(v) + except OverflowError: + return False + + +# Floating-point error: a printed number at its scale and the stored value +# differ by at most four units in the last place (final review F2). Every +# staged row matches within one. +ULPS = 4 def _same(a, b): """A printed number at its scale, equal up to floating-point error.""" - return math.isclose(a, b, rel_tol=1e-9, abs_tol=1e-12) + if not (math.isfinite(a) and math.isfinite(b)): + return False + return abs(a - b) <= ULPS * math.ulp(max(abs(a), abs(b))) def _rounded(a, b): """A stored share of a printed fraction, rounded ("a third" -> 0.3333).""" - return math.isclose(a, b, rel_tol=1e-3) + return math.isfinite(b) and math.isclose(a, b, rel_tol=1e-3) def value_problems(r): @@ -221,19 +328,19 @@ def value_problems(r): Delta review D1-1: a printed number matches exactly, and only the share of a printed ratio or worded fraction, the fraction itself, may round.""" v = r["value"] - if not isinstance(v, (int, float)) or isinstance(v, bool): + if isinstance(v, bool) or not isinstance(v, (int, float)): return ["value is not a number"] - if not math.isfinite(v): + if not _number(v): return ["value is not finite"] raw = r["value_raw"] if isinstance(r["value_raw"], str) else "" - exact, fractions = _numbers(raw) + norm = r.get("normalization") or "" + exact, fractions = _numbers(raw, norm) if not (exact or fractions): return [] # nothing printed to compare (a categorical label) - norm = r.get("normalization") or "" printed = [n * s for n in exact for s in _scales(raw, norm)] if not ( any(_same(abs(v), abs(p)) for p in printed) - or any(_rounded(abs(v), f) for f in fractions) + or any(_rounded(abs(v), abs(f)) for f in fractions) ): return [f"value {v!r} is not value_raw {raw!r} at a stated scale"] if ( @@ -277,6 +384,35 @@ def read_manifest(family_dir): return [json.loads(line) for line in text.splitlines() if line.strip()] +# A line of a JSON-lines file that isn't JSON keeps its place as UNREADABLE. +UNREADABLE = object() + + +def _json_lines(path, label): + """A JSON-lines file's values, one per non-blank line, and a problem for + each line that isn't JSON or for a file that can't be read, never an + exception (final review F3). read_rows and read_manifest raise instead, + for the generators and tests that read checked families.""" + try: + if path.suffix == ".gz": + with gzip.open(path, "rt", encoding="utf-8") as fh: + text = fh.read() + else: + text = path.read_text(encoding="utf-8") + except (OSError, EOFError, ValueError, zlib.error) as e: + return [], [f"{path.name} cannot be read ({type(e).__name__})"] + values, out = [], [] + for line in text.split("\n"): # as read_rows splits a file's lines + if not line.strip(): + continue + try: + values.append(json.loads(line)) + except (ValueError, RecursionError): + out.append(f"{label} {len(values)}: is not JSON") + values.append(UNREADABLE) + return values, out + + def proposed_keys(rows): """The proposed measure keys a family's rows name (README rule 2). A malformed row names none; row_problems reports it (delta review D1-3).""" @@ -297,7 +433,7 @@ def row_problems(r, keys, earlier, metrics, units): return ["row is not an object"] try: return _row_problems(r, keys, earlier, metrics, units) - except (TypeError, AttributeError, KeyError, ValueError) as e: + except MALFORMED as e: return [f"row is malformed ({type(e).__name__}: {e})"] @@ -414,19 +550,33 @@ def _row_problems(r, keys, earlier, metrics, units): or r["source_model"] != "policyengine_uk" ): out.append("same_assumptions rows are PolicyEngine-computed") - locator = pub.get("locator") or "" if ( r["source"] == "hm_treasury" and "impact on households" in (pub.get("title") or "").lower() and ( r["value_kind"] == "approx_chart_reading" - or (FIGURE_1.search(locator) and not PRINTED_AT.search(locator)) + or reads_a_bar(pub.get("locator") or "") ) ): out.append("no HMT Impact on households bar readings (README rule 4)") return out +def reads_a_bar(locator): + """README rule 4: whether a locator places the number on a bar of Figure + 1.A or 1.B, rather than naming where HM Treasury prints it.""" + if BAR_READING.search(locator): + return True + if not FIGURE_1.search(locator): + return False + lead = re.split(r"[,;]", locator, maxsplit=1)[0] + if NEGATION.search(lead): + return True + if FIGURE_1.search(lead): + return not DATA_LABEL.search(lead) + return not (PRINTED_AT.search(lead) or DATA_LABEL.search(lead)) + + def family_problems(family_dir): """Every contract breach in one family directory, as strings.""" family_dir = Path(family_dir) @@ -437,10 +587,16 @@ def family_problems(family_dir): ] if missing: return [f"{family_dir.name}: missing {missing}"] - out = [] - manifest = read_manifest(family_dir) + manifest, out = _json_lines(family_dir / "manifest.jsonl", "manifest line") + rows, unread = _json_lines(family_dir / "claims_staged.jsonl.gz", "row") + out += unread + try: + notes = (family_dir / "NOTES.md").read_text(encoding="utf-8") + except (OSError, ValueError) as e: + notes = "" + out.append(f"NOTES.md cannot be read ({type(e).__name__})") entries = [m for m in manifest if isinstance(m, dict)] - if len(entries) < len(manifest): + if len(entries) < len([m for m in manifest if m is not UNREADABLE]): out.append("manifest has a line that is not an object") shas = [m["sha256"] for m in entries if isinstance(m.get("sha256"), str)] for m in entries: @@ -452,8 +608,9 @@ def family_problems(family_dir): out.append("manifest lists a primary twice") keys, earlier = registry_keys() metrics, units = enums() - rows = read_rows(family_dir) for i, r in enumerate(rows): + if r is UNREADABLE: + continue # reported by _json_lines for p in row_problems(r, keys, earlier, metrics, units): out.append(f"row {i}: {p}") pub = r.get("publication") if isinstance(r, dict) else None @@ -461,7 +618,6 @@ def family_problems(family_dir): if not (isinstance(sha, str) and sha in shas): out.append(f"row {i}: primary_sha256 is not in manifest.jsonl") out += duplicate_problems(rows) - notes = (family_dir / "NOTES.md").read_text() for pk in sorted(proposed_keys(rows)): if not re.search(rf"(? 2: out.append(f"rows {idx} share one claim identity (more than 2)") elif len(idx) == 2: - a, b = (rows[i]["value"] for i in idx) + a, b = (rows[i].get("value") for i in idx) if not (_number(a) and _number(b)): - continue # row_problems reports the value (delta review D1-3) + continue # row_problems reports the value (D1-3, final review F3) if not math.isclose(a, b, rel_tol=0.005, abs_tol=1e-9): out.append( f"rows {idx} share one claim identity but differ by over 0.5%" @@ -523,7 +679,8 @@ def main(argv): bad = 0 for d in argv: problems = family_problems(d) - rows = len(read_rows(d)) if (Path(d) / "claims_staged.jsonl.gz").exists() else 0 + gz = Path(d) / "claims_staged.jsonl.gz" + rows = len(_json_lines(gz, "row")[0]) if gz.exists() else 0 print(f"{d}: {rows} rows, {len(problems)} problems") for p in problems[:200]: print(" " + p) diff --git a/sources/harvest-uk-ab2026-2026-10-29/README.md b/sources/harvest-uk-ab2026-2026-10-29/README.md index 62fc7cc40..81935b23e 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/README.md +++ b/sources/harvest-uk-ab2026-2026-10-29/README.md @@ -1,74 +1,84 @@ # Harvest: Autumn Budget 2026 scores from every producer (#230) -Staged from 2026-10-29, the day after the Budget, from the primary documents. The layout follows the AB2025 harvest (`sources/harvest-uk-ab2025-2026-09-24/`): +We began staging this harvest from the primary documents on 29 October 2026, the day after the Budget. We lay it out as in the AB2025 harvest (`sources/harvest-uk-ab2025-2026-09-24/`): - One family directory per producer group. -- Each family holds: - - `NOTES.md`: access recipe, coverage tally, and what was **not** staged and why; +- Each family's directory contains: + - `NOTES.md`: how we fetched each primary, a tally of what we staged, and what we left out and why; - `manifest.jsonl`: one line per primary document, with its sha256; - `claims_staged.jsonl.gz`: one line per claim. -- `COUNTS.json` pins each family's rows, primaries, producer slugs and the sha256 of its claims file. `pipeline/harvest_counts.py` writes and checks it. -- `REGISTRY_KEYS.md` lists every measure key. `pipeline/build_ab2026_registry.py` generates it. -- `PROPOSED_KEYS.md` lists every proposed key a row names (rule 2), for tranche 2 to register. `pipeline/ab2026_proposed_keys.py` generates it. -- `SEED_INVENTORY/` lists the publications the harvest starts from, compiled while monitoring reactions on 28 and 29 October. +- In `COUNTS.json`, we pin each family's rows, primaries, producer slugs and the sha256 of its claims file. `pipeline/harvest_counts.py` writes and checks it. +- `pipeline/build_ab2026_registry.py` writes `REGISTRY_KEYS.md`, a list of every measure key. +- `pipeline/ab2026_proposed_keys.py` writes `PROPOSED_KEYS.md`, a list of every proposed key on a row (rule 2). We register those keys in tranche 2. +- In `SEED_INVENTORY/`, we list the publications we started from, which we gathered while we monitored reactions on 28 and 29 October. -Nothing reads this harvest yet. Tranche 2 of #230 adds the ingest. That ingest's exact accounting will read `read = ingested + dropped` against the rows pinned in `COUNTS.json`. `tests/test_uk_ab2026_harvest.py` checks every row against the contract below. +No code reads this harvest yet; we add the ingest in tranche 2 of #230. In its accounting, the ingest will require `read = ingested + dropped` exactly, against the rows we pin in `COUNTS.json`. `tests/test_uk_ab2026_harvest.py` checks every row against the contract below. ## Rules -The AB2025 harvest README's rules 1–10 and its row contract apply, with the AB2026 exceptions and additions below: - -1. **`fiscal_event` is the slug `autumn_budget_2026`** on every row, whatever the publication calls it. -2. **Measure keys come from `data/uk/ab2026_measures.json`.** - - `ab2026__…` keys cover announced measures: every HM Treasury Table 6.1 line, Table 6.2 line 4, and the `ab2026__package_…` packages. - - `ab2026_option__…` keys cover options a producer costed. None is registered yet. A row on an unregistered option, or on a package whose scope is not one of the registered packages, has `measure_key: null`. Its proposed key goes in `conditions.proposed_measure_key`, and its NOTES says what the scope is. The registry adds it before ingest. - - Table 6.4 recostings are the one exception. They carry the earlier event's key, named in `conditions.recosting_of`. - - **Naming a proposed key:** - - An announced measure with no Table 6.1 line is `ab2026__`, naming the measure. That covers a measure below Table 6.1's £10m threshold, a fiscally neutral one and one costed only elsewhere. - - A package is `ab2026__package_`, and the slug names the scope, not the producer, so that producers with the same scope share one key. Lines 61 and 62 together are `ab2026__package_property_and_savings_income_rates`. A producer's or the OBR's name goes in the slug only for a grouping that is its own, such as `ab2026__package_tpa_other_vat`. - - An option a producer costed is `ab2026_option__`. - - The coordinator gives each scope one key across families before merging. `PROPOSED_KEYS.md` lists the result. -3. **A measure's scope is the producer's, not ours.** A producer's figure for "the cost-of-living package" joins `ab2026__package_household_energy_vat_and_fuel_duty` only when its scope is exactly Table 6.1 lines 1, 2 and 3. Otherwise it is a proposal (rule 2). - - Line 2 costs the whole electricity VAT cut, from 1 October 2026 to 1 July 2027 (£755m in 2026-27 and £305m in 2027-28), including the weeks before the Budget. A package of the energy payment, the VAT cut and the fuel duty cut is therefore exactly lines 1, 2 and 3. A package that also counts the gas standing-charge change, which has no Table 6.1 line, is a proposal. -4. **HMT Impact on households, Figures 1.A and 1.B: no bar readings** (AB2025 rule 9, #61). Values HM Treasury prints as numbers, in text, tables or data labels, are staged as printed. -5. **PolicyEngine's own published figures are not harvested here.** They are tranche 3's counterparts, computed from the dashboard's receipts. A producer that restates a PolicyEngine figure is staged with `attribution: same_assumptions`, as the Fabian Society's row was for AB2025. -6. **Quotes from producers outside the Open Government Licence stay short.** `quote` is the sentence or cell carrying the number, cut to the words that carry it: under 15 words for think tanks, firms and the press. HM Treasury and OBR material is OGL, and its cells and sentences are quoted whole. -7. **A primary that changes is a second primary.** If a producer replaces a document, both versions are hashed in `manifest.jsonl` and rows cite the version they were read from. IFS replaced its cost-of-living deck on 29 October, and HM Treasury replaced the Budget 2026 document PDF after 28 October. NOTES says what changed. -8. **Stage estimates, not design.** A row is a number someone estimated, forecast or counted about the effect of a measure or of the Budget. That covers: +We follow the AB2025 harvest README's rules 1–10 and its row contract, with the AB2026 exceptions and additions below: + +1. **We set `fiscal_event` to the slug `autumn_budget_2026`** on every row, whatever the publication calls it. +2. **We take measure keys from `data/uk/ab2026_measures.json`.** + - We give `ab2026__…` keys to announced measures: every HM Treasury Table 6.1 line, Table 6.2 line 4, and the `ab2026__package_…` packages. + - We give `ab2026_option__…` keys to options a producer costed, and we haven't registered any yet. On a row for an unregistered option, or for a package whose scope matches no registered package, we set `measure_key: null`. We put its proposed key in `conditions.proposed_measure_key`, state the key's scope in the family's NOTES, and add the key to the registry before the ingest. + - We make one exception, for Table 6.4's recostings: we give each the earlier event's key, and name that event in `conditions.recosting_of`. + - **We name proposed keys this way:** + - We key an announced measure with no Table 6.1 line `ab2026__`, and name the measure in the slug. We use this form for a measure below Table 6.1's £10m threshold, a fiscally neutral measure and one whose only costing appears elsewhere. + - We key a package `ab2026__package_` and name its scope in the slug, not its producer, so that producers with the same scope share one key. For example, we key lines 61 and 62 together as `ab2026__package_property_and_savings_income_rates`. We put a producer's or the OBR's name in the slug only for a grouping of its own, such as `ab2026__package_tpa_other_vat`. + - We key an option a producer costed `ab2026_option__`. + - Before we merge, we give each scope one key across families, and `pipeline/ab2026_proposed_keys.py` lists the result in `PROPOSED_KEYS.md`. +3. **We use the producer's scope for a measure, not ours.** We key a producer's figure for "the cost-of-living package" to `ab2026__package_household_energy_vat_and_fuel_duty` only when its scope matches Table 6.1 lines 1, 2 and 3. Otherwise, we give it a proposed key (rule 2). + - In line 2, HM Treasury costs the whole electricity VAT cut, from 1 October 2026 to 1 July 2027 (£755m in 2026-27 and £305m in 2027-28), including the weeks before the Budget. So a package of the energy payment, the VAT cut and the fuel duty cut matches lines 1, 2 and 3. If a producer's package also counts the gas standing-charge change, which has no Table 6.1 line, we give it a proposed key. +4. **We don't read bars off HM Treasury's Impact on households Figures 1.A and 1.B** (AB2025 rule 9, #61). We stage the numbers HM Treasury prints in text, tables or data labels, at the values it prints. +5. **We don't harvest PolicyEngine's own published figures here.** In tranche 3, we compute them from the dashboard's receipts and add them as counterparts. When a producer restates a PolicyEngine figure, we stage its row with `attribution: same_assumptions`, as we did for the Fabian Society's row in AB2025. +6. **We keep quotes from producers outside the Open Government Licence short.** In `quote`, we cut the sentence or cell with the number down to the phrase around it: under 15 words for think tanks, firms and the press. HM Treasury and the OBR publish under the OGL, so we quote their cells and sentences whole. +7. **We treat a changed primary as a second primary.** If a producer replaces a document, we hash both versions in `manifest.jsonl` and cite, on each row, the version we read it from. IFS replaced its cost-of-living deck on 29 October, and HM Treasury replaced the Budget 2026 document PDF after 28 October. We record what changed in the family's NOTES. +8. **Stage estimates, not design.** We stage as a row each number someone estimated, forecast or counted about the effect of a measure or of the Budget, including: - a cost or yield; - - people or households affected; + - people or households a measure affects; - a gain, loss or bill change; - a distributional or poverty effect; - a behavioural adjustment or tax base; - a fiscal aggregate; - a consequential for a devolved government. - A measure's own parameters are its design: rates, thresholds, payment amounts, dates and durations. So are spending allocations to programmes, financing remits and contingent liabilities. They go in `reform_hint` or `conditions`, never as rows (the AB2025 red book family kept "£9.90" in `reform_hint` the same way). A worked example's inputs follow this rule too, and its computed outcome is a row. + We treat as design a measure's own parameters (rates, thresholds, payment amounts, dates and durations), spending allocations to programmes, financing remits and contingent liabilities. We put design in `reform_hint` or `conditions`, never in rows, as we kept "£9.90" in `reform_hint` in the AB2025 red book family. We treat a worked example's inputs as design too, and stage the outcome its producer computes as a row. - Lending and financing follow one rule across families. The harvest stages HM Treasury's Tables 6.1, 6.2 and 6.4 and the OBR's policy workbook whole, as the decisions' costings. Elsewhere, an amount a scheme will lend or spend is design, and an effect on borrowing, debt or the financing requirement is an estimate. So the OBR's paragraph 3.22 loan volumes for Your First Home are not staged, while its effect on debt is, and paragraph 3.20's rise in the Government's financing requirement is staged on `cash_requirement_change`. -9. **A component of one measure keeps that measure's key.** A printed part of one Table 6.1 measure carries that measure's key, with `conditions.component` naming the part and `conditions.row_type: measure_component`. A part can be a fiscal head, a sub-measure within a line (the tax-gap package's items, for example) or a nation. Components are never summed with the measure's total. Proposed package keys are for scopes that cross Table 6.1 lines, or have none. -10. **A producer's stated effect of the Budget's decisions on PSNB takes `decisions_effect_on_borrowing`.** That covers the total effect of Government decisions, its direct and indirect effects and any part of them the producer identifies, its splits into tax and spending and into scorecard and non-scorecard decisions, and figures that restate or recombine those effects without ambiguity. `models.py` defines the metric as PSNB and keeps it apart from `revenue_change`, a receipts line, and `ingest_obr_policy_effects.py` maps March 2026 Table B.1's totals, its direct effects and their tax and spending split to it. HM Treasury sets out in Table 6.1 the cost or yield of the decisions "which have a direct effect on Public Sector Net Borrowing", so Table 6.1's three totals, and the figures that restate them, take the metric. A receipts total or a tax-change estimate keeps `revenue_change` unless its evidence establishes PSNB scope, and a total key or a package label alone doesn't establish it. The OBR sums its receipts heads, from air passenger duty to visa fees, into its policy workbook's total of receipts measures, so that total keeps `revenue_change`. IFS doesn't tie its own estimate of the Budget's tax change at the forecast horizon to PSNB, so that estimate keeps `revenue_change` too. `ingest_uk_ab2025.py` maps AB2025's `package_tax_rise_size` rows to the same metric. Single measures, groups of measures and a total's lines by tax or spending head keep `revenue_change`, unless their producer states the effect on borrowing, as the OBR does for four policies in paragraphs 3.20, 3.21 and 3.43 and IFS does for the Budget's new giveaways. Text that restates a Table 3.1 group or a single measure as an effect on borrowing keeps that row's metric (the OBR's paragraph 3.11 "increase borrowing by £5 billion in 2026-27" restates Table 3.1's other departmental spending changes, which keep `revenue_change`), and Table 3.1's "A positive sign implies an increase in borrowing" is a sign convention, not a statement of PSNB scope. A total that adds the receipts effect of an option the Budget didn't adopt is not a restatement or recombination of the decisions' effects, so it keeps `revenue_change` unless its evidence establishes PSNB scope: Resolution Foundation's Figure 10 "Net effect if Fuel Duty frozen" adds a continued freeze's receipts effect, which its note derives from the OBR's constant-duty-rate fuel duty forecast, and the OBR states the freeze's effect on borrowing for 2029-30 only (paragraph 3.43). A row giving the effect on the current budget deficit (or, for an outturn, the primary deficit) proposes `fiscal_balance_change`, and its sign convention names that deficit. Each row keeps the sign its producer prints, and `conditions.sign_convention` says which way it runs. + We apply one rule to lending and financing across families. We stage HM Treasury's Tables 6.1, 6.2 and 6.4 and the OBR's policy workbook whole, as the decisions' costings. Elsewhere, we treat an amount a scheme will lend or spend as design, and an effect on borrowing, debt or the financing requirement as an estimate. So we leave out the OBR's paragraph 3.22 loan volumes for Your First Home but stage its effect on debt, and we stage paragraph 3.20's rise in the Government's financing requirement on `cash_requirement_change`. +9. **We give a component of one measure that measure's key.** On a row with a printed part of one Table 6.1 measure, we set that measure's key, name the part in `conditions.component` and set `conditions.row_type: measure_component`. We count a fiscal head, a sub-measure within a line (the tax-gap package's items, for example) or a nation as a part. We never sum components with the measure's total. We propose package keys for scopes that cross Table 6.1 lines, or have none. +10. **We give a producer's stated effect of the Budget's decisions on PSNB `decisions_effect_on_borrowing`.** We give it to the total effect of Government decisions, its direct and indirect effects and any part of them the producer identifies, and its splits into tax and spending and into scorecard and non-scorecard decisions. We also give it to any figure a producer derives, without ambiguity, by restating or recombining those effects. `models.py` defines the metric as PSNB and keeps it apart from `revenue_change`, a receipts line. `ingest_obr_policy_effects.py` maps March 2026 Table B.1's totals, its direct effects and their tax and spending split to it. HM Treasury sets out in Table 6.1 the cost or yield of the decisions "which have a direct effect on Public Sector Net Borrowing". So we give the metric to Table 6.1's three totals and to any producer's restatement of them. + + We keep `revenue_change` on a receipts total or a tax-change estimate unless its evidence establishes PSNB scope, and we don't take a total key or a package label alone as that evidence. The OBR sums its receipts heads, from air passenger duty to visa fees, into its policy workbook's total of receipts measures, so we keep `revenue_change` on that total. IFS doesn't tie its own estimate of the Budget's tax change at the forecast horizon to PSNB, so we keep `revenue_change` on that estimate too. `ingest_uk_ab2025.py` maps AB2025's `package_tax_rise_size` rows to `revenue_change` as well. + + We keep `revenue_change` on single measures, groups of measures and a total's lines by tax or spending head, unless their producer states the effect on borrowing. The OBR states it for four policies in paragraphs 3.20, 3.21 and 3.43, and IFS for the Budget's new giveaways. When a producer restates a Table 3.1 group or a single measure as an effect on borrowing, we give the restatement the metric of the row it restates. For example, in paragraph 3.11 the OBR restates Table 3.1's other departmental spending changes with "increase borrowing by £5 billion in 2026-27", and we keep `revenue_change` on both. We read Table 3.1's "A positive sign implies an increase in borrowing" as a sign convention, not a statement of PSNB scope. + + When a producer adds to a total the receipts effect of an option the Government didn't adopt, we don't count the result as a restatement or recombination of the decisions' effects. We keep `revenue_change` on it unless its evidence establishes PSNB scope. For example, in Figure 10's "Net effect if Fuel Duty frozen", Resolution Foundation adds a continued freeze's receipts effect, which it derives in a note from the OBR's constant-duty-rate fuel duty forecast. The OBR states the freeze's effect on borrowing for 2029-30 only (paragraph 3.43). + + On a row with the effect on the current budget deficit (or, for an outturn, the primary deficit), we propose `fiscal_balance_change` and name that deficit in the sign convention. We keep the sign each producer prints, and state in `conditions.sign_convention` which way it runs. ## Row contract additions -`pipeline/ab2026_harvest_contract.py` checks the AB2025 row contract with these additions (review A, H-1 to H-6, and delta review D1): - -- `value_kind` may also be `categorical` (a rating, such as the OBR's Table 3.3 uncertainty ratings). -- `value` is finite and equals `value_raw`'s number, to floating-point error, at the scale `value_raw`'s word gives (bn, m, k, %), failing that at the factor `normalization` states, failing that at 1, 10³, 10⁶, 10⁹, 0.01 or 100. A ratio ("1 in 4") or a worded fraction ("a third") gives a share, the fraction itself, which `value` may round within 0.1%; a ratio's two numbers don't count on their own. Its sign is `value_raw`'s unless `normalization` says the sign comes from the wording. -- An HM Treasury Impact on households row reads a bar (rule 4) if its `value_kind` is `approx_chart_reading`, or if its locator names Figure 1.A or 1.B and no paragraph, table, box or data label that prints the number. -- A malformed row (not an object, or a field of the wrong type) gives a problem, never an exception, and the checks across rows skip what they can't compare. -- A row that carries `conditions.component` is a part: `row_type` is `measure_component` (or the OBR workbook's `measure_total` marker), and the component isn't blank (rule 9). -- No more than two non-restated rows share the AB2025 ingest's claim identity, and two that do agree within 0.5%. -- A registered metric goes in `metric`, never `proposed_metric`. -- `conditions.recosting_of` appears only on `row_type: recosting` rows (Table 6.4). -- `conditions.fy` is one fiscal year, and a `fiscal_year` row without `fy` gives `conditions.horizon`. -- `conditions.geography` is a code: UK, GB, England, Wales, Scotland, Northern Ireland, England and Wales, UK_excl_scotland, "Scotland, Wales and Northern Ireland", "not stated", or a country a producer compares the UK with. Prose goes in `conditions.geography_note`. -- A restated row's `note` says whose figure it restates. -- A proposed key is `ab2026__` or `ab2026_option__` followed by lower-case words joined by single underscores, and NOTES scopes it as a whole token. +`pipeline/ab2026_harvest_contract.py` checks the AB2025 row contract with these additions: + +- The checker accepts `value_kind: categorical` for a rating, such as the OBR's Table 3.3 uncertainty ratings. +- It requires a finite number in `value` that matches `value_raw`'s number within four units in the last place. It takes the scale from `value_raw`'s word (bn, m, k, %), failing that from the factor in `normalization`, failing that from 1, 10³, 10⁶, 10⁹, 0.01 or 100. +- It reads a ratio ("1 in 4", "1-in-4", "one in ten") or a worded fraction ("a third") as a share, accepts a `value` within 0.1% of that share, and won't match `value` to either of the ratio's numbers alone. It reads "-1 in 4" as −0.25. It reads a year after "in" ("£5 in 2027") as a year, not a denominator, unless we write the ratio in `normalization` ("1 in 2000 -> 0.0005", or "N in M"). +- It requires `value` to share `value_raw`'s sign unless we note in `normalization` that we took the sign from the wording. +- It treats an HM Treasury Impact on households row as a bar reading (rule 4) if the row has `value_kind: approx_chart_reading`, or if we call the number a bar, chart or graph reading in its locator, or say there that we read it off one. It also treats the row as a bar reading if we name Figure 1.A or 1.B in its locator, unless we say, in the locator's first reference (up to its first comma or semicolon), where HM Treasury prints the number. The checker requires that reference to name a data label, or a paragraph, table or box without the figure, and to negate nothing ("no data labels"). It ignores later references, so it counts "Figure 1.A, decile 3 bar; see paragraph 1.7" as a bar reading. +- It reports a malformed row as a problem, never as an exception. It counts a row as malformed if the row isn't an object, lacks a field, or has a field of the wrong type, an integer too large for a float or nesting too deep to print. It reports a line it can't parse as JSON, or a file it can't read, the same way, and when it compares rows, it skips any value it can't compare. +- It treats a row with `conditions.component` as a part of one measure (rule 9), and requires `row_type: measure_component`, or the OBR workbook's `measure_total` marker, and a non-blank component. +- It allows at most two non-restated, non-categorical rows to share the AB2025 ingest's claim identity, and requires two that do to agree within 0.5%. +- It requires a registered metric in `metric`, never in `proposed_metric`. +- It accepts `conditions.recosting_of` only on `row_type: recosting` rows (Table 6.4). +- It requires `conditions.fy` to name one fiscal year, and `conditions.horizon` on a `fiscal_year` row without `fy`. +- It accepts only a code in `conditions.geography`: UK, GB, England, Wales, Scotland, Northern Ireland, England and Wales, UK_excl_scotland, "Scotland, Wales and Northern Ireland", "not stated", or a country a producer compares the UK with. Put prose in `conditions.geography_note`. +- It requires a `note` on each restated row with the name of the original figure's producer, and rejects a blank one. +- It requires a proposed key of the form `ab2026__…` or `ab2026_option__…`, with lower-case words of letters and digits after the prefix and single underscores between them. It looks for the whole key in the family's NOTES, where we state its scope. ## Families staged -12 families, 5,952 rows (450 `restated`), as `COUNTS.json` pins them. +We pin 12 families and 5,952 rows (450 `restated`) in `COUNTS.json`. | family | rows | primaries | producer slugs | |---|--:|--:|---| @@ -85,14 +95,14 @@ The AB2025 harvest README's rules 1–10 and its row contract apply, with the AB | `uk_tpa` | 305 | 10 | `tax_policy_associates` | | `uk_trussell_wpi` | 1 | 1 | `trussell_wpi` | -Each family's `stage.py`, which writes its claims file, is kept on branch `uk/ab2026-harvest-staging` under `harvest-staging/ab2026/lanes///`, with the lane briefs and the five reviews; this branch carries only each family's three files, as AB2025 does. On 3 November 2026 we applied the review fixes, all four sections of the IFS addendum and the cross-family key harmonisation, and each family's NOTES.md records them. On 4 November we applied reviews A and B of 4f424b9 (the registry, the HM Treasury families and the harness; the OBR, Resolution Foundation and think-tank families), and each NOTES.md records those too. +Each family's `stage.py` writes its claims file, and we keep it on branch `uk/ab2026-harvest-staging` under `harvest-staging/ab2026/lanes///`, with the lane briefs and the five reviews. We record each family's review changes in its NOTES.md, and keep only its three files here, as in AB2025. -**Families with no rows** are not carried here, as AB2025 has no empty family. Their working files stay on the staging branch: +We leave the **families with no rows** out, since AB2025 has no empty family, and keep their working files on the staging branch: -- `uk_cpag`: both CPAG pages (its Budget response and a pre-Budget call) returned Cloudflare's challenge (HTTP 403) to every fetch on 29 October. -- `uk_smf`: the Social Market Foundation page returned a challenge to every fetch on 29 October. -- `uk_niesr`: NIESR's Budget-day reaction ("Breathing Space, Not a Strategy") was fetched, but it holds macro and fiscal-stance commentary for the #55 lane and restated OBR figures, which the AB2025 NIESR family did not stage either. -- `uk_fai_sfc`: the Fraser of Allander Institute's reaction refused the fetch (HTTP 403), and the Scottish Fiscal Commission published nothing in scope on 28 or 29 October. The reaction's feed text holds only restated OBR figures. On 4 November, review B found a second FAI post through FAI's RSS feed, "Rising or falling? Why both claims about Scotland's funding can be true" (29 October), behind the same 403. The coordinator left its table of Scotland's Barnett consequentials for tranche 2: FAI says the split emerged late on 28 October, so the table probably restates HM Treasury's allocations, which `uk_hmt_redbook` stages in rows 252–255 (paragraphs 4.150, 4.154, 4.157 and 4.160). The Scottish Government's five rows moved from this family to `uk_devolved_govts` on 3 November. +- `uk_cpag`: Cloudflare answered every fetch of both CPAG pages (its Budget response and a pre-Budget call) with its challenge (HTTP 403) on 29 October. +- `uk_smf`: the Social Market Foundation's site answered every fetch of its page with a challenge on 29 October. +- `uk_niesr`: we fetched NIESR's Budget-day reaction ("Breathing Space, Not a Strategy"), but in it NIESR gives macro and fiscal-stance commentary for the #55 lane and restates OBR figures, and we staged neither in the AB2025 NIESR family. +- `uk_fai_sfc`: the Fraser of Allander Institute's site refused our fetch of its reaction (HTTP 403), and the Scottish Fiscal Commission published nothing in scope on 28 or 29 October. In the reaction's feed text, FAI only restates OBR figures. On 4 November, we found a second FAI post through FAI's RSS feed, "Rising or falling? Why both claims about Scotland's funding can be true" (29 October), behind the same 403. We leave its table of Scotland's Barnett consequentials for tranche 2. FAI says the split emerged late on 28 October, so in the table FAI probably restates HM Treasury's allocations, which we stage in `uk_hmt_redbook` rows 252–255 (paragraphs 4.150, 4.154, 4.157 and 4.160). We stage the Scottish Government's five rows in `uk_devolved_govts`. ## Producer slugs and families @@ -105,9 +115,9 @@ Each family's `stage.py`, which writes its claims file, is kept on branch `uk/ab | resolution_foundation | Resolution Foundation | uk_rf_ab2026 | | ifs | Institute for Fiscal Studies | uk_ifs_ab2026 | | tax_policy_associates, centax, ippr, cps, institute_for_government, learning_and_work_institute, citizens_advice | other producers, each in the AB2025 family of the same name where one exists | uk_tpa, uk_centax, uk_ippr, uk_thinktank_misc | -| cpag, smf, niesr, fraser_of_allander, scottish_fiscal_commission | producers with no rows (see "Families with no rows"); their working files are on the staging branch, not here | none | +| cpag, smf, niesr, fraser_of_allander, scottish_fiscal_commission | producers with no rows (see "families with no rows"); we keep their working files on the staging branch, not here | none | | scottish_government, welsh_government | the devolved governments' responses | uk_devolved_govts | | trussell_wpi | Trussell, as in AB2025 | uk_trussell_wpi | -| commercial firms and platforms | worked examples (mode 3, #63); none was staged for AB2026 | none (AB2025: uk_cases_commercial) | +| commercial firms and platforms | worked examples (mode 3, #63); we staged none for AB2026 | none (AB2025: uk_cases_commercial) | -Macro calls by banks and City forecasters belong to the #55 lane and are not staged here (AB2025 rule 10). +We leave macro calls by banks and City forecasters to the #55 lane and don't stage them here (AB2025 rule 10). diff --git a/tests/test_ab2026_harvest_contract.py b/tests/test_ab2026_harvest_contract.py index 1950d6e06..4d67bff99 100644 --- a/tests/test_ab2026_harvest_contract.py +++ b/tests/test_ab2026_harvest_contract.py @@ -12,6 +12,7 @@ import json import math import shutil +import tempfile from pathlib import Path import pytest @@ -19,10 +20,14 @@ from hypothesis import strategies as st from pipeline.ab2026_harvest_contract import ( + NUMBER_WORDS, OGL, + duplicate_problems, enums, family_problems, + proposed_keys, read_rows, + reads_a_bar, registry_keys, row_problems, ) @@ -164,6 +169,13 @@ def test_a_staged_row_has_no_problems(): {"measure_key": None, "conditions__proposed_measure_key": ["ab2026__x"]}, "row is malformed", ), + # final review F2: £1 or £2 off a £m cell is no floating-point error + ({"value": -2625000001}, "at a stated scale"), + ({"value": -2625000002}, "at a stated scale"), + ({"value": -2624999999.5}, "at a stated scale"), + # final review F3: an integer too large for a float + ({"value": 10**400}, "value is not finite"), + ({"value": -(10**400)}, "value is not finite"), ], ) def test_each_rule_broken_on_a_valid_row_is_a_problem(changes, expected): @@ -218,26 +230,114 @@ def test_a_year_after_in_is_no_denominator(): assert problems(mutated(value_raw="5 in 2027", quote="5 in 2027", value=5e6)) == [] +def test_a_year_after_in_is_a_denominator_when_the_normalization_says_so(): + """Final review F1: "1 in 2000" stored as the share 0.0005 passes only + when its normalization gives the ratio, as "1 in 2,000" always does.""" + stated = "1 in 2000 -> 0.0005" + assert problems(_share("1 in 2000", 0.0005)) == [] + assert problems(_share("1 in 2000", 0.0005, stated)) == [] + assert problems(_share("1 in 2,000", 0.0005, "share as printed")) == [] + found = problems(_share("1 in 2000", 0.0005, "£ million x 1,000,000")) + assert any("at a stated scale" in p for p in found), found + for wrong in (1, 2000): + found = problems(_share("1 in 2000", wrong, stated)) + assert any("at a stated scale" in p for p in found), (wrong, found) + + +@pytest.mark.parametrize( + ("raw", "share", "normalization"), + [ + ("1 IN 4", 0.25, "N in M -> N/M"), + ("1 In 4", 0.25, "N in M -> N/M"), + ("1-in-4", 0.25, "N in M -> N/M"), + ("-1 in 4", -0.25, "N in M -> N/M"), + ("−1 in 4", -0.25, "N in M -> N/M"), + ("one in ten", 0.1, "one in ten -> 0.1"), + ("fewer than one in ten", 0.1, "one in ten -> 0.1"), + ("about three in ten", 0.3, "three in ten -> 0.3"), + ("Eight in Ten", 0.8, "eight in ten -> 0.8"), + ("one in four", 0.25, "one in four -> 0.25"), + ("one in a hundred", 0.01, "one in a hundred -> 0.01"), + ("1 in 2000", 0.0005, "1 in 2000 -> 0.0005"), + ], +) +def test_a_ratio_in_any_form_is_its_signed_share(raw, share, normalization): + """Final review F1: "in" in any case or hyphenated, a minus on N, and + number words (CenTax's "one in ten", rows 16, 93 and 95). The share + passes; the share with the other sign, N, M and N/M as a percentage + don't. Before F1, a worded ratio printed no number, so any value passed.""" + assert problems(_share(raw, share, normalization)) == [] + flipped = problems(_share(raw, -share, normalization)) + assert any("opposite sign" in p for p in flipped), flipped + for wrong in (1, 3, 4, 8, 10, 100, 2000, 100 * share, 0.5): + if not math.isclose(abs(wrong), abs(share), rel_tol=1e-3): + found = problems(_share(raw, wrong, normalization)) + assert any( + "at a stated scale" in p or "opposite sign" in p for p in found + ), (raw, wrong, found) + + @settings(max_examples=200, deadline=None) -@given(n=st.integers(1, 50), extra=st.integers(1, 1000)) -def test_a_ratio_passes_only_at_its_share(n, extra): - """Delta review D1-1, for every "N in M": N/M passes, and N, M, N% and M% - fail unless they round to N/M.""" +@given( + n=st.integers(1, 50), + extra=st.integers(1, 1000), + form=st.sampled_from(["{n} in {m:,}", "{n} IN {m:,}", "{n}-in-{m}", "{n} In {m}"]), + minus=st.sampled_from(["", "-", "−"]), +) +def test_a_ratio_passes_only_at_its_share(n, extra, form, minus): + """Delta review D1-1 and final review F1, for every "N in M" in any case + and with any minus: N/M, signed as N is, passes, and N, M, N% and M% fail + unless they round to N/M.""" m = n + extra - raw = f"{n} in {m:,}" - assert problems(_share(raw, n / m)) == [] + raw = minus + form.format(n=n, m=m) + share = (-1 if minus else 1) * n / m + assert problems(_share(raw, share)) == [] + assert any("opposite sign" in p for p in problems(_share(raw, -share))) for wrong in (n, m, n / 100, m / 100, 100 * n / m): if not math.isclose(wrong, n / m, rel_tol=1e-3): found = problems(_share(raw, wrong)) - assert any("at a stated scale" in p for p in found), (raw, wrong) + assert any( + "at a stated scale" in p or "opposite sign" in p for p in found + ), (raw, wrong) @settings(max_examples=200, deadline=None) -@given(cell=st.integers(-9999, 9999), off=st.integers(1_000, 10**8), up=st.booleans()) +@given( + n=st.sampled_from(sorted(NUMBER_WORDS)), + m=st.sampled_from(sorted(NUMBER_WORDS)), + wrong=st.floats(0.001, 1000), +) +def test_a_worded_ratio_passes_only_at_its_share(n, m, wrong): + """Final review F1: number words give a ratio as numerals do.""" + share = NUMBER_WORDS[n] / NUMBER_WORDS[m] + raw = f"{n} in {m}" + note = f"{raw} -> {share}" + assert problems(_share(raw, share, note)) == [] + if not math.isclose(wrong, share, rel_tol=1e-3): + found = problems(_share(raw, wrong, note)) + assert any("at a stated scale" in p for p in found), (raw, wrong) + + +@settings(max_examples=300, deadline=None) +@given(tenths=st.integers(-99_999, 99_999), how=st.integers(0, 2)) +def test_a_printed_percentage_is_its_share_to_floating_point_error(tenths, how): + """Final review F2: a share computed from a printed percentage in any + usual way passes, and any value 1e-12 or more away fails.""" + raw = f"{tenths / 10}%" + share = (tenths / 10 / 100, tenths / 1000, float(raw[:-1]) * 0.01)[how] + note = "per cent / 100 -> share" + assert problems(_share(raw, share, note)) == [] + moved = share + (1e-12 if share >= 0 else -1e-12) + assert any("at a stated scale" in p for p in problems(_share(raw, moved, note))) + + +@settings(max_examples=200, deadline=None) +@given(cell=st.integers(-9999, 9999), off=st.integers(1, 10**8), up=st.booleans()) def test_a_money_cell_passes_only_at_its_exact_value(cell, off, up): - """Delta review D1-1: a cell printed in £ million is that many millions, - to floating-point error. £1,000 off any cell under £10bn fails: the 0.1% - rounding allowance covers a share of a printed fraction, nothing else.""" + """Delta review D1-1 and final review F2: a cell printed in £ million is + that many millions, to floating-point error (four units in the last + place). £1 off any cell under £10bn fails: the 0.1% rounding allowance + covers a share of a printed fraction, nothing else.""" raw = str(cell) quote = f"Line 1 | 2026-27: {raw}" exact = mutated(value_raw=raw, quote=quote, value=cell * 1e6) @@ -276,15 +376,28 @@ def test_rule_4_holds_whatever_the_title_case(): ("Figure 1.A, decile 3 bar", True), ("p.5, Figure 1.B, top bar", True), ("paragraph 1.7, p.8; discussion of Figure 1.A", False), + ("paragraph 1.7; discussion of Figure 1.A", False), ("Figure 1.A data label, decile 3", False), ("Table 2.C, decile 3", False), + ("Table 2.C, p.16, decile 1, column 1 adult", False), + ("paragraph 2.4, p.14; source Table 2.B, p.13", False), + # final review F4: a secondary or negated reference names nothing + ("Figure 1.A, decile 3 bar; no data labels", True), + ("Figure 1.A, decile 3 bar; see paragraph 1.7", True), + ("Figure 1.A decile 3 bar, paragraph 1.7", True), + ("Figure 1.A decile 3 bar beside paragraph 1.7", True), + ("Figure 1.A (no data labels), decile 3", True), + ("no data label; paragraph 1.7, Figure 1.A", True), + ("paragraph 1.7; bar reading from Figure 1.A", True), + ("Figure 1.B; data label", True), + ("data label for decile 3, Figure 1.A", False), ], ) def test_rule_4_reads_where_the_locator_places_the_number(locator, bar_reading): """README rule 4: a locator that places the number on Figure 1.A's or - 1.B's bars marks a bar reading; one that names the paragraph, table or - data label printing it doesn't (delta review D1: red book row 357's 20% - sits in paragraph 1.7).""" + 1.B's bars marks a bar reading; one whose leading reference names the + paragraph, table or data label printing it doesn't (delta review D1: + red book row 357's 20% sits in paragraph 1.7; final review F4).""" row = mutated( publication__title="Budget 2026: Impact on households", publication__locator=locator, @@ -292,6 +405,52 @@ def test_rule_4_reads_where_the_locator_places_the_number(locator, bar_reading): assert any("rule 4" in p for p in problems(row)) == bar_reading +@settings(max_examples=300, deadline=None) +@given( + lead=st.sampled_from( + [ + "paragraph 1.7", + "Table 2.C", + "Box 1.A", + "Figure 1.A data label", + "data label", + "Figure 1.B", + "Figure 1.A, decile 3 bar", + "p.8", + "no data labels", + ] + ), + rest=st.lists( + st.sampled_from( + [ + "p.8", + "see paragraph 1.7", + "no data labels", + "discussion of Figure 1.A", + "decile 3", + "data label", + "Table 2.C", + ] + ), + max_size=3, + ), +) +def test_rule_4_follows_the_leading_reference(lead, rest): + """Final review F4: where a locator names Figure 1.A or 1.B, only its + leading reference, positively naming a paragraph, table, box or data + label, places the number where HM Treasury prints it; anything later in + the locator places nothing.""" + locator = "; ".join([lead, *rest]) + printed = lead in ( + "paragraph 1.7", + "Table 2.C", + "Box 1.A", + "Figure 1.A data label", + "data label", + ) + assert reads_a_bar(locator) == ("Figure 1." in locator and not printed) + + def test_a_component_with_its_measure_or_a_proposed_key_is_valid(): """README rule 9: a part of one measure keeps that measure's key; a part of a measure with no Table 6.1 line keeps that measure's proposed key.""" @@ -458,13 +617,132 @@ def test_a_family_missing_a_file_is_a_problem(tmp_path): [mutated(publication__primary_sha256=["abc"])], "primary_sha256 is not in manifest.jsonl", ), + # final review F3: a copy of the base row without its value, and one + # whose value is an integer too large for a float, pair with it + ([{k: v for k, v in BASE.items() if k != "value"}], "missing fields"), + ([{**BASE, "value": 10**400}], "value is not finite"), ], ) def test_a_family_reports_malformed_rows_without_raising(tmp_path, extra, expected): - """Delta review D1-3: family_problems reports a malformed row as - row_problems does. The pair check, the proposed-key scope check and the - manifest lookup skip a value, key or hash they can't compare.""" + """Delta review D1-3 and final review F3: family_problems reports a + malformed row as row_problems does. The pair check, the proposed-key + scope check and the manifest lookup skip a value, key or hash they can't + compare.""" fam = _family(tmp_path) _write(fam, read_rows(fam) + extra) found = family_problems(fam) assert any(expected in p for p in found), found[:5] + + +def test_a_family_reports_unreadable_lines_and_files(tmp_path): + """Final review F3: a line that isn't JSON, or a file that can't be read, + is a problem, never an exception.""" + fam = _family(tmp_path) + gz = fam / "claims_staged.jsonl.gz" + text = gzip.decompress(gz.read_bytes()).decode() + gz.write_bytes(gzip.compress((text + "{not json\n").encode(), mtime=0)) + (fam / "manifest.jsonl").write_text((fam / "manifest.jsonl").read_text() + "[1,\n") + found = family_problems(fam) + assert [p for p in found if p.startswith("row 768")] == ["row 768: is not JSON"] + assert [p for p in found if p.startswith("manifest")] == [ + "manifest line 2: is not JSON" + ] + gz.write_bytes(b"not gzip") + found = family_problems(fam) + assert "claims_staged.jsonl.gz cannot be read (BadGzipFile)" in found, found[:5] + + +def test_a_field_nested_too_deep_to_print_is_a_problem(): + """Final review F3: a field nested too deep to print, or to encode as a + claim identity, is a problem, never an exception.""" + deep = [] + for _ in range(100_000): + deep = [deep] + found = problems(mutated(period=deep)) + assert [p for p in found if p.startswith("row is malformed (RecursionError")] + nested = mutated(conditions__geography_note=deep) + assert duplicate_problems([nested, nested]) == [] + + +# Any JSON value, with the integers JSON allows but a float can't hold. +JSON = st.recursive( + st.none() + | st.booleans() + | st.integers() + | st.floats() + | st.text(max_size=12) + | st.sampled_from([10**400, -(10**400), "1 in 4", "-2,625", ""]), + lambda inner: ( + st.lists(inner, max_size=3) + | st.dictionaries(st.text(max_size=8), inner, max_size=3) + ), + max_leaves=8, +) +FIELDS = sorted( + [*BASE] + + [f"conditions__{k}" for k in BASE["conditions"]] + + [f"publication__{k}" for k in BASE["publication"]] +) + + +def _changed(changes): + """BASE with each field set to a value, or deleted (...), nested fields + first, so a later change to their parent can't strand them.""" + row = copy.deepcopy(BASE) + for path in sorted(changes, key=lambda p: -p.count("__")): + *parents, leaf = path.split("__") + target = row + for p in parents: + target = target[p] + if changes[path] is ...: + target.pop(leaf, None) + else: + target[leaf] = changes[path] + return row + + +@settings(max_examples=400, deadline=None) +@given( + changes=st.dictionaries( + st.sampled_from(FIELDS), JSON | st.just(...), min_size=1, max_size=4 + ) +) +def test_any_malformed_row_gives_problems_never_an_exception(changes): + """Final review F3, for every input: row_problems, the pair check and the + proposed-key scan return strings for a row with any field changed to any + JSON value or deleted, alone and next to copies of itself and BASE.""" + row = _changed(changes) + found = problems(row) + assert isinstance(found, list) and all(isinstance(p, str) for p in found) + pairs = duplicate_problems([BASE, row]) + duplicate_problems([row, row]) + assert all(isinstance(p, str) for p in pairs) + assert isinstance(proposed_keys([row, BASE]), set) + + +@settings(max_examples=200, deadline=None) +@given(row=JSON) +def test_any_json_value_as_a_row_gives_problems_never_an_exception(row): + found = problems(row) + assert found and all(isinstance(p, str) for p in found) + assert all(isinstance(p, str) for p in duplicate_problems([row, row, BASE])) + assert isinstance(proposed_keys([row]), set) + + +@settings(max_examples=60, deadline=None) +@given(lines=st.lists(JSON.map(json.dumps) | st.text(max_size=30), max_size=4)) +def test_any_family_file_gives_problems_never_an_exception(lines): + """Final review F3: family_problems reads any line, JSON or not.""" + with tempfile.TemporaryDirectory() as d: + fam = Path(d) / "uk_fuzz" + fam.mkdir() + (fam / "NOTES.md").write_text("Fuzzed family.\n") + shutil.copy(FAMILY / "manifest.jsonl", fam / "manifest.jsonl") + body = "".join( + line.replace("\n", " ").replace("\r", " ") + "\n" + for line in [json.dumps(BASE), *lines] + ) + (fam / "claims_staged.jsonl.gz").write_bytes( + gzip.compress(body.encode(), mtime=0) + ) + found = family_problems(fam) + assert all(isinstance(p, str) for p in found) From 1bf45071ac42ca55162aeeeffafc54ed8faa17d5 Mon Sep 17 00:00:00 2001 From: Max Ghenis Date: Wed, 4 Nov 2026 19:48:14 -0500 Subject: [PATCH 42/52] Close rule 4 and manifest gaps found before the delta review The checker now reads Figure 1.A or 1.B in the spellings "Fig. 1A" and "Figures 1.A and 1.B", and treats a leading reference that negates its data label with "never", "missing", "absent", "lacks", "lacking" or "n't" as a bar reading under rule 4. Before, "Fig. 1.A, decile 3 bar" and "Figure 1.A data label missing" both passed. It now requires a manifest line's url, title and retrieved date to be text with more than whitespace, and names a line by its place when its url isn't text. Before, a list or a number passed as a url or a title. We narrowed the ULPS comment. Of the 5,952 staged rows, 5,907 match a printed number, each within one unit in the last place; 27 match only a ratio or a worded fraction, and 18 print no number. On fe58796c's checker, 20 of the 152 contract tests fail, and each of 15 single mutations of the new arms fails a test. No staged row changes, and all 12 families pass the contract. Co-Authored-By: Claude Opus 5.5 --- pipeline/ab2026_harvest_contract.py | 40 ++++-- .../harvest-uk-ab2026-2026-10-29/README.md | 2 +- tests/test_ab2026_harvest_contract.py | 114 ++++++++++++++---- 3 files changed, 118 insertions(+), 38 deletions(-) diff --git a/pipeline/ab2026_harvest_contract.py b/pipeline/ab2026_harvest_contract.py index 0a0f06a79..bf5a4b805 100644 --- a/pipeline/ab2026_harvest_contract.py +++ b/pipeline/ab2026_harvest_contract.py @@ -175,14 +175,19 @@ # README rule 4: a locator's leading reference, up to its first comma or # semicolon, says where HM Treasury prints the number: a data label when it # names Figure 1.A or 1.B, else a paragraph, table, box or data label. A -# locator that names Figure 1.A or 1.B anywhere needs such a reference, free -# of any negation ("no data labels"), and one that calls the number a bar or -# chart reading reads a bar whatever else it names (delta review D1; final -# review F4). -FIGURE_1 = re.compile(r"\bfigure 1\.[ab]\b", re.I) +# locator that names Figure 1.A or 1.B anywhere, in any of the spellings +# "Figure 1.A", "Fig. 1A" or "Figures 1.A and 1.B", needs such a reference, +# free of any negation ("no data labels", "data label missing", "isn't +# labelled"), and one that calls the number a bar or chart reading reads a +# bar whatever else it names (delta review D1; final review F4). +FIGURE_1 = re.compile(r"\bfig(?:ure)?s?\.?\s*1\.?\s*[ab]\b", re.I) PRINTED_AT = re.compile(r"\b(?:paragraph|para|table|box)s?\b", re.I) DATA_LABEL = re.compile(r"\bdata labels?\b", re.I) -NEGATION = re.compile(r"\b(?:no|not|none|without|unlabell?ed)\b", re.I) +NEGATION = re.compile( + r"\b(?:no|not|none|never|without|missing|absent|lacks?|lacking|unlabell?ed)\b" + r"|n['’]t\b", + re.I, +) BAR_READING = re.compile( r"\b(?:bar|chart|graph) readings?\b" r"|\bread (?:off|from) (?:the |a )?(?:bars?|charts?|graphs?|figures?)\b", @@ -304,9 +309,15 @@ def _number(v): return False +def _text(v): + """A string with more than whitespace in it.""" + return isinstance(v, str) and bool(v.strip()) + + # Floating-point error: a printed number at its scale and the stored value -# differ by at most four units in the last place (final review F2). Every -# staged row matches within one. +# differ by at most four units in the last place (final review F2). Each +# staged value that matches a printed number, rather than a ratio or worded +# fraction, differs from it by at most one. ULPS = 4 @@ -599,11 +610,16 @@ def family_problems(family_dir): if len(entries) < len([m for m in manifest if m is not UNREADABLE]): out.append("manifest has a line that is not an object") shas = [m["sha256"] for m in entries if isinstance(m.get("sha256"), str)] - for m in entries: + for j, m in enumerate(manifest): + if not isinstance(m, dict): + continue # reported above, or by _json_lines + # a problem names the line by its url, or by its place when the url + # isn't text + name = m["url"] if _text(m.get("url")) else f"line {j}" if not (isinstance(m.get("sha256"), str) and SHA256.fullmatch(m["sha256"])): - out.append(f"manifest {m.get('url')}: sha256 is not a sha256") - if not (m.get("url") and m.get("title") and m.get("retrieved")): - out.append(f"manifest {m.get('url')}: needs url, title and retrieved") + out.append(f"manifest {name}: sha256 is not a sha256") + if not all(_text(m.get(f)) for f in ("url", "title", "retrieved")): + out.append(f"manifest {name}: needs url, title and retrieved as text") if len(set(shas)) != len(shas): out.append("manifest lists a primary twice") keys, earlier = registry_keys() diff --git a/sources/harvest-uk-ab2026-2026-10-29/README.md b/sources/harvest-uk-ab2026-2026-10-29/README.md index 81935b23e..698f0b9ba 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/README.md +++ b/sources/harvest-uk-ab2026-2026-10-29/README.md @@ -65,7 +65,7 @@ We follow the AB2025 harvest README's rules 1–10 and its row contract, with th - It requires a finite number in `value` that matches `value_raw`'s number within four units in the last place. It takes the scale from `value_raw`'s word (bn, m, k, %), failing that from the factor in `normalization`, failing that from 1, 10³, 10⁶, 10⁹, 0.01 or 100. - It reads a ratio ("1 in 4", "1-in-4", "one in ten") or a worded fraction ("a third") as a share, accepts a `value` within 0.1% of that share, and won't match `value` to either of the ratio's numbers alone. It reads "-1 in 4" as −0.25. It reads a year after "in" ("£5 in 2027") as a year, not a denominator, unless we write the ratio in `normalization` ("1 in 2000 -> 0.0005", or "N in M"). - It requires `value` to share `value_raw`'s sign unless we note in `normalization` that we took the sign from the wording. -- It treats an HM Treasury Impact on households row as a bar reading (rule 4) if the row has `value_kind: approx_chart_reading`, or if we call the number a bar, chart or graph reading in its locator, or say there that we read it off one. It also treats the row as a bar reading if we name Figure 1.A or 1.B in its locator, unless we say, in the locator's first reference (up to its first comma or semicolon), where HM Treasury prints the number. The checker requires that reference to name a data label, or a paragraph, table or box without the figure, and to negate nothing ("no data labels"). It ignores later references, so it counts "Figure 1.A, decile 3 bar; see paragraph 1.7" as a bar reading. +- It treats an HM Treasury Impact on households row as a bar reading (rule 4) if the row has `value_kind: approx_chart_reading`, or if we call the number a bar, chart or graph reading in its locator, or say there that we read it off one. It also treats the row as a bar reading if we name Figure 1.A or 1.B in its locator ("Figure 1.A", "Fig. 1A" or "Figures 1.A and 1.B"), unless we say, in the locator's first reference (up to its first comma or semicolon), where HM Treasury prints the number. The checker requires that reference to name a data label, or a paragraph, table or box without the figure, and to negate nothing ("no data labels", "data label missing"). It ignores later references, so it counts "Figure 1.A, decile 3 bar; see paragraph 1.7" as a bar reading. - It reports a malformed row as a problem, never as an exception. It counts a row as malformed if the row isn't an object, lacks a field, or has a field of the wrong type, an integer too large for a float or nesting too deep to print. It reports a line it can't parse as JSON, or a file it can't read, the same way, and when it compares rows, it skips any value it can't compare. - It treats a row with `conditions.component` as a part of one measure (rule 9), and requires `row_type: measure_component`, or the OBR workbook's `measure_total` marker, and a non-blank component. - It allows at most two non-restated, non-categorical rows to share the AB2025 ingest's claim identity, and requires two that do to agree within 0.5%. diff --git a/tests/test_ab2026_harvest_contract.py b/tests/test_ab2026_harvest_contract.py index 4d67bff99..dcdd8f110 100644 --- a/tests/test_ab2026_harvest_contract.py +++ b/tests/test_ab2026_harvest_contract.py @@ -391,6 +391,19 @@ def test_rule_4_holds_whatever_the_title_case(): ("paragraph 1.7; bar reading from Figure 1.A", True), ("Figure 1.B; data label", True), ("data label for decile 3, Figure 1.A", False), + # the figure in other spellings, and negations in other words + ("Figures 1.A and 1.B, decile 3 bars", True), + ("Fig. 1.A, decile 3 bar", True), + ("Figure 1A, decile 3 bar", True), + ("fig 1b; top bar", True), + ("Figure 1.A data label missing, decile 3", True), + ("Figure 1.A data label isn't printed", True), + ("Figure 1.A data label absent", True), + ("Figure 1.B lacks a data label", True), + ("Fig. 1A data label, decile 3", False), + ("Figures 1.A and 1.B data labels", False), + ("Figure 1 below; paragraph 1.7", False), + ("Table 2.C; config 1a", False), ], ) def test_rule_4_reads_where_the_locator_places_the_number(locator, bar_reading): @@ -405,21 +418,65 @@ def test_rule_4_reads_where_the_locator_places_the_number(locator, bar_reading): assert any("rule 4" in p for p in problems(row)) == bar_reading +@pytest.mark.parametrize( + "negated", + [ + "no data labels", + "not a data label", + "none of its data labels", + "never a data label", + "without data labels", + "data label missing", + "data label absent", + "lacks a data label", + "lacking data labels", + "unlabelled data label", + "data label: unlabeled bars", + "data label isn't printed", + "data label isn’t printed", + ], +) +def test_rule_4_reads_a_negated_data_label_as_a_bar_reading(negated): + """Final review F4: a leading reference that negates its data label, in + any of these words, places the number on a bar.""" + assert reads_a_bar(f"Figure 1.A {negated}") + assert not reads_a_bar("Figure 1.A data label") + + +# Leading references that say where HM Treasury prints a number, and ones +# that don't; and the parts of a locator that name Figure 1.A or 1.B. +PRINTING = [ + "paragraph 1.7", + "Table 2.C", + "Box 1.A", + "Figure 1.A data label", + "Fig. 1B data labels", + "data label", +] +NOT_PRINTING = [ + "Figure 1.B", + "Figures 1.A and 1.B", + "Figure 1.A, decile 3 bar", + "p.8", + "no data labels", + "data label missing", + "Figure 1.A data label isn't shown", +] +NAMES_FIGURE = { + "Figure 1.A data label", + "Fig. 1B data labels", + "Figure 1.B", + "Figures 1.A and 1.B", + "Figure 1.A, decile 3 bar", + "Figure 1.A data label isn't shown", + "discussion of Figure 1.A", + "fig 1a", +} + + @settings(max_examples=300, deadline=None) @given( - lead=st.sampled_from( - [ - "paragraph 1.7", - "Table 2.C", - "Box 1.A", - "Figure 1.A data label", - "data label", - "Figure 1.B", - "Figure 1.A, decile 3 bar", - "p.8", - "no data labels", - ] - ), + lead=st.sampled_from(PRINTING + NOT_PRINTING), rest=st.lists( st.sampled_from( [ @@ -427,6 +484,7 @@ def test_rule_4_reads_where_the_locator_places_the_number(locator, bar_reading): "see paragraph 1.7", "no data labels", "discussion of Figure 1.A", + "fig 1a", "decile 3", "data label", "Table 2.C", @@ -436,19 +494,13 @@ def test_rule_4_reads_where_the_locator_places_the_number(locator, bar_reading): ), ) def test_rule_4_follows_the_leading_reference(lead, rest): - """Final review F4: where a locator names Figure 1.A or 1.B, only its - leading reference, positively naming a paragraph, table, box or data - label, places the number where HM Treasury prints it; anything later in - the locator places nothing.""" + """Final review F4: where a locator names Figure 1.A or 1.B, in any + spelling, only its leading reference, positively naming a paragraph, + table, box or data label, places the number where HM Treasury prints it; + anything later in the locator places nothing.""" locator = "; ".join([lead, *rest]) - printed = lead in ( - "paragraph 1.7", - "Table 2.C", - "Box 1.A", - "Figure 1.A data label", - "data label", - ) - assert reads_a_bar(locator) == ("Figure 1." in locator and not printed) + names_figure = any(part in NAMES_FIGURE for part in [lead, *rest]) + assert reads_a_bar(locator) == (names_figure and lead not in PRINTING) def test_a_component_with_its_measure_or_a_proposed_key_is_valid(): @@ -583,6 +635,18 @@ def _rewrite_manifest(fam, edit): [ (lambda m: [{**m[0], "sha256": "abc"}, *m[1:]], "sha256 is not a sha256"), (lambda m: [{**m[0], "retrieved": ""}, *m[1:]], "needs url, title"), + # a field that isn't text, and a problem that names its line when the + # url isn't text + (lambda m: [{**m[0], "title": 5}, *m[1:]], "needs url, title"), + (lambda m: [{**m[0], "retrieved": " "}, *m[1:]], "needs url, title"), + ( + lambda m: [{**m[0], "url": ["x"]}, *m[1:]], + "manifest line 0: needs url, title and retrieved as text", + ), + ( + lambda m: [*m[:1], {**m[1], "url": 7, "sha256": "abc"}, *m[2:]], + "manifest line 1: sha256 is not a sha256", + ), (lambda m: [*m, m[0]], "lists a primary twice"), (lambda m: m[1:], "primary_sha256 is not in manifest.jsonl"), (lambda m: [*m, "not an object"], "a line that is not an object"), From 72cdfdcc60911cbf9c19680804dad71e34775d13 Mon Sep 17 00:00:00 2001 From: Max Ghenis Date: Wed, 4 Nov 2026 19:53:19 -0500 Subject: [PATCH 43/52] Split the README's 44-word rule 4 sentence We split the approx_chart_reading case off the opening sentence of the README's rule 4 bullet. No rule, check or row changes. Co-Authored-By: Claude Opus 5.5 --- sources/harvest-uk-ab2026-2026-10-29/README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sources/harvest-uk-ab2026-2026-10-29/README.md b/sources/harvest-uk-ab2026-2026-10-29/README.md index 698f0b9ba..2bf6b7e6d 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/README.md +++ b/sources/harvest-uk-ab2026-2026-10-29/README.md @@ -65,7 +65,7 @@ We follow the AB2025 harvest README's rules 1–10 and its row contract, with th - It requires a finite number in `value` that matches `value_raw`'s number within four units in the last place. It takes the scale from `value_raw`'s word (bn, m, k, %), failing that from the factor in `normalization`, failing that from 1, 10³, 10⁶, 10⁹, 0.01 or 100. - It reads a ratio ("1 in 4", "1-in-4", "one in ten") or a worded fraction ("a third") as a share, accepts a `value` within 0.1% of that share, and won't match `value` to either of the ratio's numbers alone. It reads "-1 in 4" as −0.25. It reads a year after "in" ("£5 in 2027") as a year, not a denominator, unless we write the ratio in `normalization` ("1 in 2000 -> 0.0005", or "N in M"). - It requires `value` to share `value_raw`'s sign unless we note in `normalization` that we took the sign from the wording. -- It treats an HM Treasury Impact on households row as a bar reading (rule 4) if the row has `value_kind: approx_chart_reading`, or if we call the number a bar, chart or graph reading in its locator, or say there that we read it off one. It also treats the row as a bar reading if we name Figure 1.A or 1.B in its locator ("Figure 1.A", "Fig. 1A" or "Figures 1.A and 1.B"), unless we say, in the locator's first reference (up to its first comma or semicolon), where HM Treasury prints the number. The checker requires that reference to name a data label, or a paragraph, table or box without the figure, and to negate nothing ("no data labels", "data label missing"). It ignores later references, so it counts "Figure 1.A, decile 3 bar; see paragraph 1.7" as a bar reading. +- It treats an HM Treasury Impact on households row as a bar reading (rule 4) if the row has `value_kind: approx_chart_reading`. It does the same if we call the number a bar, chart or graph reading in its locator, or say there that we read it off one. It also treats the row as a bar reading if we name Figure 1.A or 1.B in its locator ("Figure 1.A", "Fig. 1A" or "Figures 1.A and 1.B"), unless we say, in the locator's first reference (up to its first comma or semicolon), where HM Treasury prints the number. The checker requires that reference to name a data label, or a paragraph, table or box without the figure, and to negate nothing ("no data labels", "data label missing"). It ignores later references, so it counts "Figure 1.A, decile 3 bar; see paragraph 1.7" as a bar reading. - It reports a malformed row as a problem, never as an exception. It counts a row as malformed if the row isn't an object, lacks a field, or has a field of the wrong type, an integer too large for a float or nesting too deep to print. It reports a line it can't parse as JSON, or a file it can't read, the same way, and when it compares rows, it skips any value it can't compare. - It treats a row with `conditions.component` as a part of one measure (rule 9), and requires `row_type: measure_component`, or the OBR workbook's `measure_total` marker, and a non-blank component. - It allows at most two non-restated, non-categorical rows to share the AB2025 ingest's claim identity, and requires two that do to agree within 0.5%. From 615d6a87f55191ff7815be2c3ba345fbc476a572 Mon Sep 17 00:00:00 2001 From: Max Ghenis Date: Wed, 4 Nov 2026 21:17:26 -0500 Subject: [PATCH 44/52] Read more ratio forms, worded fractions and number words MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The contract checker now reads "1 in 2 million" as 5e-7, and "one in every ten" and "3 out of 10" as shares. It reads number words up to the hundreds in a ratio ("one in twenty-five"), and worded fractions from "half" to "tenth" ("a third", "three-fifths", "another quarter"). On a percent row, it requires 100 times the share. When value_raw has no number in digits, the checker reads its number words ("four", "over a million"), except on a categorical row. It reads no ratio after a currency sign ("£5 in 10 years"), or where it can't read the multiplier ("1 in 2 millionths"). The tests check three properties: - Hypothesis writes each number from 1 to 999 in words, in any style, and the checker reads it back as that number. - Hypothesis builds "N in M" from numerals or words, six joints and four multipliers, and the checker passes only N/(M x multiplier), within 0.1%. - Across the 5,952 staged rows, the new checker reads 5,943 as before. It now compares 9 rows it skipped before, and all 9 pass. No staged row changes. 47 of the 208 contract tests fail against 72cdfdcc's checker, and the tests kill all 27 mutants of the new grammar. Co-Authored-By: Claude Opus 5.5 --- pipeline/ab2026_harvest_contract.py | 167 +++++++++---- .../harvest-uk-ab2026-2026-10-29/README.md | 6 +- tests/test_ab2026_harvest_contract.py | 235 +++++++++++++++++- 3 files changed, 344 insertions(+), 64 deletions(-) diff --git a/pipeline/ab2026_harvest_contract.py b/pipeline/ab2026_harvest_contract.py index bf5a4b805..b4206e84b 100644 --- a/pipeline/ab2026_harvest_contract.py +++ b/pipeline/ab2026_harvest_contract.py @@ -121,13 +121,18 @@ (re.compile(r"%|per ?cent|percentage point|ppts?\b|pp\b", re.I), 0.01), ) NUMBER = re.compile(r"(? 0.0005", "N in M -> N/M") # (final review F1). -NUMBER_WORDS = { +# Number words: one to nineteen, and the tens +ONES = { "one": 1, "two": 2, "three": 3, @@ -147,6 +152,8 @@ "seventeen": 17, "eighteen": 18, "nineteen": 19, +} +TENS = { "twenty": 20, "thirty": 30, "forty": 40, @@ -155,20 +162,43 @@ "seventy": 70, "eighty": 80, "ninety": 90, - "a hundred": 100, - "one hundred": 100, - "a thousand": 1000, - "one thousand": 1000, } -_OPERAND = r"\d[\d,]*(?:\.\d+)?|" + "|".join( - re.escape(w).replace(r"\ ", " ").replace(" ", r"\s+") - for w in sorted(NUMBER_WORDS, key=len, reverse=True) +MULTIPLIERS = { + "hundred": 100, + "thousand": 1e3, + "k": 1e3, + "million": 1e6, + "mn": 1e6, + "m": 1e6, + "billion": 1e9, + "bn": 1e9, +} + + +def _either(words): + """A regex alternation of words, the longest first.""" + return "|".join(re.escape(w) for w in sorted(words, key=len, reverse=True)) + + +_SEP = r"(?:\s+|\s*-\s*)" +_SMALL = ( + rf"(?:(?:{_either(TENS)}){_SEP}(?:{_either(w for w in ONES if ONES[w] < 10)})" + rf"|{_either(TENS)}|{_either(ONES)})" ) +# A number in words, up to the hundreds ("twenty-five", "a hundred and five") +WORDED = rf"(?:(?:a|{_SMALL})\s+hundred(?:\s+(?:and\s+)?{_SMALL})?|{_SMALL})" +# "a" counts as one before a thousand, million or billion ("over a million") +_A = r"a(?=\s+(?:thousand|million|billion)\b)" +_NUMERAL = r"\d[\d,]*(?:\.\d+)?" RATIO = re.compile( - rf"(?|→)") @@ -193,29 +223,48 @@ r"|\bread (?:off|from) (?:the |a )?(?:bars?|charts?|graphs?|figures?)\b", re.I, ) -WORDS = { - "half": 0.5, - "a half": 0.5, - "one half": 0.5, - "a third": 1 / 3, - "one third": 1 / 3, - "two thirds": 2 / 3, - "two-thirds": 2 / 3, - "a quarter": 0.25, - "one quarter": 0.25, - "three quarters": 0.75, - "three-quarters": 0.75, - "two-fifths": 0.4, - "two fifths": 0.4, +# A worded fraction: "half", or a numerator and a denominator ("a third", +# "three-fifths", "another quarter"). +NUMERATORS = {"a": 1, "an": 1, "another": 1} | { + w: v for w, v in ONES.items() if v <= 10 +} +DENOMINATORS = { + "half": 2, + "halves": 2, + "third": 3, + "thirds": 3, + "quarter": 4, + "quarters": 4, + "fifth": 5, + "fifths": 5, + "sixth": 6, + "sixths": 6, + "seventh": 7, + "sevenths": 7, + "eighth": 8, + "eighths": 8, + "ninth": 9, + "ninths": 9, + "tenth": 10, + "tenths": 10, } +FRACTION = re.compile( + rf"\b(?:({_either(NUMERATORS)}){_SEP})?({_either(DENOMINATORS)})(?!\w|-\w)", re.I +) + +def _worded(text): + """A number in words ("twenty-five", "a hundred and five"), "a" as one.""" + n = 0 + for w in re.split(r"[\s-]+", text.lower()): + n = n * 100 if w == "hundred" else n + ({"a": 1} | ONES | TENS).get(w, 0) + return n -def _operand(text): - """A ratio's number, from digits or number words ("1,000", "a hundred").""" - word = re.sub(r"\s+", " ", text.lower()) - if word in NUMBER_WORDS: - return float(NUMBER_WORDS[word]) - return float(text.replace(",", "")) + +def _operand(text, multiplier=None): + """A ratio's number, from digits or number words, times its multiplier.""" + n = float(text.replace(",", "")) if text[0].isdigit() else float(_worded(text)) + return n * MULTIPLIERS.get((multiplier or "").lower(), 1) def _gives_ratio(normalization, n, m): @@ -236,25 +285,38 @@ def _ratios(text, normalization=""): as N is.""" out = [] for m in RATIO.finditer(text): - sign, n, d = m.groups() - if YEAR.fullmatch(d) and not _gives_ratio(normalization, n, d): + sign, n, joint, d, multiplier = m.groups() + bare = joint.lower() == "in" and not multiplier + if bare and YEAR.fullmatch(d) and not _gives_ratio(normalization, n, d): continue # "£5 in 2027" - a, b = _operand(n), _operand(d) + a, b = _operand(n), _operand(d, multiplier) if b and math.isfinite(a) and math.isfinite(b): out.append((m.span(), (-a if sign else a) / b)) return out -def _numbers(text, normalization=""): +def _cut(text, spans): + """The string with each span cut out.""" + for start, end in sorted(spans, reverse=True): + text = f"{text[:start]} {text[end:]}" + return text + + +def _numbers(text, normalization="", words=False): """The numbers a string prints, as (exact, fractions). Exact numbers are its signed digits. Fractions are its ratios ("1 in 4" -> 0.25), whose two numbers count for nothing on their own, and its worded fractions ("a - third"), which a stored share may round (delta review D1-1).""" + third"), which a stored share may round (delta review D1-1). With words, + a string that prints none of these gives its number words ("four", "over + a million") as exact numbers.""" + ratios = _ratios(text, normalization) + fractions = [share for _, share in ratios] + text = _cut(text, [span for span, _ in ratios]) + for m in FRACTION.finditer(text): + top, bottom = (g.lower() if g else g for g in m.groups()) + if top or bottom == "half": + fractions.append(NUMERATORS[top or "a"] / DENOMINATORS[bottom]) text = re.sub(r"([-−–])\s*[£$€]", r"\1", text) # "-£1.7bn" - fractions = [] - for (start, end), share in reversed(_ratios(text, normalization)): - fractions.append(share) - text = f"{text[:start]} {text[end:]}" exact = [] for m in NUMBER.findall(text): t = m.replace(",", "").replace("−", "-").replace("–", "-") @@ -262,8 +324,8 @@ def _numbers(text, normalization=""): exact.append(float(t)) except ValueError: pass - low = text.lower() - fractions += [v for w, v in WORDS.items() if re.search(rf"\b{w}\b", low)] + if words and not (exact or fractions): + exact = [float(_worded(m)) for m in WORDS_ALONE.findall(text)] return exact, fractions @@ -337,7 +399,10 @@ def value_problems(r): """review A H-2: `value` is finite and is value_raw's number at its scale, with value_raw's sign unless the normalization documents a worded sign. Delta review D1-1: a printed number matches exactly, and only the share - of a printed ratio or worded fraction, the fraction itself, may round.""" + of a printed ratio or worded fraction, the fraction itself, may round. A + `unit_concept: percent` row stores that share in percent units, 100 times + the fraction. A non-categorical row that prints its number only in words + gives those words.""" v = r["value"] if isinstance(v, bool) or not isinstance(v, (int, float)): return ["value is not a number"] @@ -345,19 +410,23 @@ def value_problems(r): return ["value is not finite"] raw = r["value_raw"] if isinstance(r["value_raw"], str) else "" norm = r.get("normalization") or "" - exact, fractions = _numbers(raw, norm) + exact, fractions = _numbers(raw, norm, words=r["value_kind"] != "categorical") if not (exact or fractions): return [] # nothing printed to compare (a categorical label) - printed = [n * s for n in exact for s in _scales(raw, norm)] + # a ratio's multiplier ("1 in 2 million") scales the ratio, not the string + rest = _cut(raw, [span for span, _ in _ratios(raw, norm)]) + printed = [n * s for n in exact for s in _scales(rest, norm)] + per = 100 if r.get("unit_concept") == "percent" else 1 + shares = [per * f for f in fractions] if not ( any(_same(abs(v), abs(p)) for p in printed) - or any(_rounded(abs(v), abs(f)) for f in fractions) + or any(_rounded(abs(v), abs(f)) for f in shares) ): return [f"value {v!r} is not value_raw {raw!r} at a stated scale"] if ( WORDED_SIGN.search(norm) or any(_same(v, p) for p in printed) - or any(_rounded(v, f) for f in fractions) + or any(_rounded(v, f) for f in shares) ): return [] return [f"value {v!r} has the opposite sign to value_raw {raw!r}"] diff --git a/sources/harvest-uk-ab2026-2026-10-29/README.md b/sources/harvest-uk-ab2026-2026-10-29/README.md index 2bf6b7e6d..f2c9bdfac 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/README.md +++ b/sources/harvest-uk-ab2026-2026-10-29/README.md @@ -62,8 +62,10 @@ We follow the AB2025 harvest README's rules 1–10 and its row contract, with th `pipeline/ab2026_harvest_contract.py` checks the AB2025 row contract with these additions: - The checker accepts `value_kind: categorical` for a rating, such as the OBR's Table 3.3 uncertainty ratings. -- It requires a finite number in `value` that matches `value_raw`'s number within four units in the last place. It takes the scale from `value_raw`'s word (bn, m, k, %), failing that from the factor in `normalization`, failing that from 1, 10³, 10⁶, 10⁹, 0.01 or 100. -- It reads a ratio ("1 in 4", "1-in-4", "one in ten") or a worded fraction ("a third") as a share, accepts a `value` within 0.1% of that share, and won't match `value` to either of the ratio's numbers alone. It reads "-1 in 4" as −0.25. It reads a year after "in" ("£5 in 2027") as a year, not a denominator, unless we write the ratio in `normalization` ("1 in 2000 -> 0.0005", or "N in M"). +- It requires a finite number in `value` that matches `value_raw`'s number within four units in the last place. It takes the scale from `value_raw`'s word (bn, m, k, %) outside any ratio, failing that from the factor in `normalization`, failing that from 1, 10³, 10⁶, 10⁹, 0.01 or 100. +- It reads a ratio ("1 in 4", "1-in-4", "one in every ten", "3 out of 10") as a share, accepts a `value` within 0.1% of that share, and won't match `value` to either of the ratio's numbers alone. It reads "-1 in 4" as −0.25 and "1 in 2 million" as 5e-7. It reads no ratio after a currency sign ("£5 in 10 years"). It reads a year after a bare "in" ("£5 in 2027") as a year, not a denominator, unless we write the ratio in `normalization` ("1 in 2000 -> 0.0005", or "N in M"). +- It reads worded fractions as shares too: "half" alone, or a numerator ("a", "an", "another", or one to ten) and a denominator from "half" to "tenth" ("a third", "three-fifths"). On a `unit_concept: percent` row, it requires 100 times a ratio's or a fraction's share ("1 in 4" as 25). +- It reads number words up to the hundreds ("twenty-five", "a hundred and five") in a ratio. Unless the row has `value_kind: categorical`, it also reads them in a `value_raw` with no other number ("four", "over a million"). - It requires `value` to share `value_raw`'s sign unless we note in `normalization` that we took the sign from the wording. - It treats an HM Treasury Impact on households row as a bar reading (rule 4) if the row has `value_kind: approx_chart_reading`. It does the same if we call the number a bar, chart or graph reading in its locator, or say there that we read it off one. It also treats the row as a bar reading if we name Figure 1.A or 1.B in its locator ("Figure 1.A", "Fig. 1A" or "Figures 1.A and 1.B"), unless we say, in the locator's first reference (up to its first comma or semicolon), where HM Treasury prints the number. The checker requires that reference to name a data label, or a paragraph, table or box without the figure, and to negate nothing ("no data labels", "data label missing"). It ignores later references, so it counts "Figure 1.A, decile 3 bar; see paragraph 1.7" as a bar reading. - It reports a malformed row as a problem, never as an exception. It counts a row as malformed if the row isn't an object, lacks a field, or has a field of the wrong type, an integer too large for a float or nesting too deep to print. It reports a line it can't parse as JSON, or a file it can't read, the same way, and when it compares rows, it skips any value it can't compare. diff --git a/tests/test_ab2026_harvest_contract.py b/tests/test_ab2026_harvest_contract.py index dcdd8f110..f51864212 100644 --- a/tests/test_ab2026_harvest_contract.py +++ b/tests/test_ab2026_harvest_contract.py @@ -20,8 +20,9 @@ from hypothesis import strategies as st from pipeline.ab2026_harvest_contract import ( - NUMBER_WORDS, OGL, + ONES, + TENS, duplicate_problems, enums, family_problems, @@ -301,21 +302,229 @@ def test_a_ratio_passes_only_at_its_share(n, extra, form, minus): ), (raw, wrong) -@settings(max_examples=200, deadline=None) +SAY = {v: w for w, v in (ONES | TENS).items()} + + +def _say(n, a=False, hyphen=True, joined=True): + """n, from 1 to 999, in words as a producer might print it: "a" or "one" + hundred, "twenty-five" or "twenty five", "and" after the hundreds or not.""" + if n >= 100: + head, rest = divmod(n, 100) + out = ("a" if a and head == 1 else SAY[head]) + " hundred" + if rest: + out += " and " if joined else " " + out += _say(rest, a, hyphen, joined) + return out + if n in SAY: + return SAY[n] + tens, unit = divmod(n, 10) + return SAY[10 * tens] + ("-" if hyphen else " ") + SAY[unit] + + +STYLES = st.tuples(st.booleans(), st.booleans(), st.booleans()) + + +@settings(max_examples=300, deadline=None) @given( - n=st.sampled_from(sorted(NUMBER_WORDS)), - m=st.sampled_from(sorted(NUMBER_WORDS)), + n=st.integers(1, 999), + m=st.integers(1, 999), + styles=STYLES, + joint=st.sampled_from(["in", "IN", "in every", "In Every", "out of", "OUT OF"]), + multiplier=st.sampled_from( + [("", 1), (" thousand", 1e3), (" million", 1e6), (" billion", 1e9)] + ), wrong=st.floats(0.001, 1000), ) -def test_a_worded_ratio_passes_only_at_its_share(n, m, wrong): - """Final review F1: number words give a ratio as numerals do.""" - share = NUMBER_WORDS[n] / NUMBER_WORDS[m] - raw = f"{n} in {m}" - note = f"{raw} -> {share}" - assert problems(_share(raw, share, note)) == [] - if not math.isclose(wrong, share, rel_tol=1e-3): - found = problems(_share(raw, wrong, note)) - assert any("at a stated scale" in p for p in found), (raw, wrong) +def test_a_worded_ratio_passes_only_at_its_share( + n, m, styles, joint, multiplier, wrong +): + """Final review F1: number words up to the hundreds give a ratio as + numerals do, after "in", "in every" or "out of" and with a multiplier on + M. The share passes, and so does nothing else, N and M included.""" + word, factor = multiplier + raw = f"{_say(n, *styles)} {joint} {_say(m, *styles)}{word}" + share = n / (m * factor) + assert problems(_share(raw, share, "share as printed")) == [], raw + for bad in (wrong, n, m, m * factor, n / m): + if not math.isclose(bad, share, rel_tol=1e-3): + found = problems(_share(raw, bad, "share as printed")) + assert any("at a stated scale" in p for p in found), (raw, bad) + + +@pytest.mark.parametrize( + ("raw", "share", "wrongs"), + [ + ("1 in 2 million", 5e-7, (0.5, 1, 2, 2e6)), + ("1 in 2m", 5e-7, (0.5, 2, 2e6)), + ("1 in 2.5 million", 4e-7, (0.4, 2.5, 2.5e6)), + ("1 in 4bn", 2.5e-10, (0.25, 4, 4e9)), + ("one in a million", 1e-6, (1, 1e6)), + ("one in a thousand", 1e-3, (1, 1e3)), + ("one in ten thousand", 1e-4, (0.1, 10, 1e4)), + ("one in two hundred", 5e-3, (0.5, 2, 200)), + ("1 in 2 hundred", 5e-3, (0.5, 2, 200)), + ("one in a hundred thousand", 1e-5, (0.01, 100, 1e5)), + ("one in twenty-five", 0.04, (0.05, 0.2, 1, 25)), + ("one in a hundred and twenty", 1 / 120, (0.01, 0.05, 120)), + ("3 out of 10", 0.3, (3, 10, 30)), + ("3-out-of-10", 0.3, (3, 10)), + ("three out of ten", 0.3, (3, 10)), + ("1 in every 4", 0.25, (1, 4, 25)), + ("one in every five", 0.2, (1, 5)), + ("1 in every 2000", 0.0005, (1, 2000)), + ("1 out of 2000", 0.0005, (1, 2000)), + ("1 in 2000 million", 5e-10, (0.0005, 2000, 2e9)), + ], +) +def test_a_ratio_with_a_multiplier_or_another_joint_is_its_share(raw, share, wrongs): + """A multiplier on M scales M, so "1 in 2 million" is 5e-7, never 0.5. + Number words run to the hundreds, and "in every" and "out of" print a + ratio as "in" does. A year after "in every" or "out of", or before a + multiplier, is a denominator. The share passes; N, M, and the share + without the multiplier don't.""" + assert problems(_share(raw, share, "share as printed")) == [] + for wrong in wrongs: + found = problems(_share(raw, wrong, "share as printed")) + assert any("at a stated scale" in p for p in found), (raw, wrong, found) + + +@pytest.mark.parametrize( + ("raw", "value", "share"), + [ + ("£5 in 10 years", 5, 0.5), + ("-£5 in 10", -5, -0.5), + ("$3 in 4", 3, 0.75), + ("1 in 2 millionths", 1, 0.5), + ("1 in 2 million-strong", 2e6, 0.5), + ("1 in 2 thousand million", 2e6, 0.5), + ], +) +def test_a_number_after_a_currency_sign_or_before_an_unread_multiplier_is_no_ratio( + raw, value, share +): + """A number after a currency sign starts no ratio, and a ratio whose + multiplier won't read is none: each number counts at its scale, and the + ratio's share without its multiplier fails.""" + assert problems(_share(raw, value, "as printed")) == [] + found = problems(_share(raw, share, "as printed")) + assert any("at a stated scale" in p for p in found), (raw, found) + + +def test_a_ratio_s_multiplier_scales_only_the_ratio(): + """ "million" in "1 in 2 million" scales M, not the string's other + numbers: at the stated factor of 1, the 3 in "3 (1 in 2 million)" is 3.""" + note = "count as printed x 1" + assert problems(_share("3 (1 in 2 million)", 3, note)) == [] + found = problems(_share("3 (1 in 2 million)", 3e6, note)) + assert any("at a stated scale" in p for p in found), found + + +@pytest.mark.parametrize( + ("raw", "share", "wrongs"), + [ + ("a fifth", 0.2, (5, 0.5, 20)), + ("less than a fifth", 0.2, (5, 20)), + ("about three-fifths", 0.6, (3, 5, 0.35, 60)), + ("three fifths", 0.6, (3, 5)), + ("another quarter", 0.25, (4, 0.75, 25)), + ("around one-tenth", 0.1, (1, 10)), + ("seven eighths", 0.875, (7, 8, 0.78)), + ("an eighth", 0.125, (8, 0.18)), + ("two-thirds", 2 / 3, (2, 3, 0.23)), + ("three quarters", 0.75, (3, 4, 0.34)), + ("half", 0.5, (2, 50)), + ("a half", 0.5, (2,)), + ], +) +def test_a_worded_fraction_in_any_form_is_its_share(raw, share, wrongs): + """A worded fraction is a numerator ("a", "an", "another", or one to + ten) and a denominator from a half to a tenth, with a space or a hyphen + between them, or "half" alone. The share passes; its two numbers, and 100 + times the share on a row not in percent units, don't.""" + assert problems(_share(raw, share, "share as printed")) == [] + for wrong in wrongs: + found = problems(_share(raw, wrong, "share as printed")) + assert any("at a stated scale" in p for p in found), (raw, wrong, found) + + +@pytest.mark.parametrize("raw", ["a third-party", "the third year", "per quarter"]) +def test_an_ordinal_or_a_compound_is_no_fraction(raw): + """ "third" in "a third-party" or "the third year", and "quarter" without + a numerator, print no fraction, so the row compares nothing.""" + assert problems(_share(raw, 7, "as printed")) == [] + + +@pytest.mark.parametrize( + ("raw", "share"), [("around one-tenth", 0.1), ("1 in 4", 0.25), ("a third", 1 / 3)] +) +def test_a_percent_row_stores_its_share_in_percent_units(raw, share): + """Resolution Foundation's "around one-tenth" (row 114) is 10 on its + `unit_concept: percent` row. A percent row stores 100 times the share, + and every other row the share itself.""" + + def row(value, unit): + return mutated( + value_raw=raw, + quote=raw, + value=value, + normalization="as printed", + unit_concept=unit, + ) + + assert problems(row(100 * share, "percent")) == [] + assert problems(row(share, "share")) == [] + for value, unit in ((share, "percent"), (100 * share, "share")): + found = problems(row(value, unit)) + assert any("at a stated scale" in p for p in found), (raw, value, unit) + + +@pytest.mark.parametrize( + ("raw", "value", "wrongs"), + [ + ("four", 4, (5, 40)), + ("one", 1, (2,)), + ("twenty-five", 25, (20, 5)), + ("a hundred and five", 105, (100, 5)), + ("one million", 1e6, (1, 2e6)), + ("over a million", 1e6, (1, 1e3)), + ("a hundred thousand", 1e5, (100, 1e3)), + ], +) +def test_a_number_printed_only_in_words_is_that_number(raw, value, wrongs): + """A row that prints its number only in words (OBR's "four", row 2664) + gives those words, at value_raw's scale ("over a million" is 1e6). Before, + it compared nothing and passed whatever its value.""" + assert problems(_share(raw, value, "printed in words")) == [] + for wrong in wrongs: + found = problems(_share(raw, wrong, "printed in words")) + assert any("at a stated scale" in p for p in found), (raw, wrong, found) + + +@settings(max_examples=300, deadline=None) +@given(n=st.integers(1, 999), styles=STYLES, off=st.integers(1, 50)) +def test_any_number_in_words_reads_as_that_number(n, styles, off): + """Every number from one to nine hundred and ninety-nine, in any of the + ways _say prints it, reads back as itself, and n + off doesn't.""" + raw = _say(n, *styles) + assert problems(_share(raw, n, "printed in words")) == [], raw + found = problems(_share(raw, n + off, "printed in words")) + assert any("at a stated scale" in p for p in found), (raw, n + off) + + +@pytest.mark.parametrize("raw", ["two-child limit", "one-off", "four"]) +def test_number_words_in_a_compound_or_on_a_categorical_row_compare_nothing(raw): + """A number word inside a hyphenated word ("two-child") prints no number, + and a categorical row's words are its label, never its value.""" + if raw != "four": + assert problems(_share(raw, 7, "as printed")) == [] + label = mutated( + value_raw=raw, + quote=raw, + value=7, + normalization="rating as printed", + value_kind="categorical", + ) + assert problems(label) == [] @settings(max_examples=300, deadline=None) From 92f364c793f1602ccab262c8686ff36db88361ed Mon Sep 17 00:00:00 2001 From: Max Ghenis Date: Wed, 4 Nov 2026 22:23:05 -0500 Subject: [PATCH 45/52] Read each ratio number whole and check every field's type Answers the delta review of 72cdfdcc (R1 to R5). R1: the checker now reads each number in a ratio as one run of tokens, takes a minus on either number, and reads the generic normalization in any case. It reports a ratio or a number in words it can't read as a problem. R2: rule 4 reads the locator's first reference through a closed grammar, and a negation cancels a bar-reading phrase only right before it. R3: a test pins the match at four units in the last place. R4: type_problems reports each field of a type the row contract doesn't allow before any check reads the field. We rename the malformed-input property to say what it tests. R5: the README rewrites rule 1, the ratio bullets, the rule-4 bullets and the malformed-row bullet. All 5,952 staged rows read as before, with 0 contract problems. The old checker fails 168 of the new tests, and the tests kill all 41 mutants. Co-Authored-By: Claude Opus 5.5 --- pipeline/ab2026_harvest_contract.py | 574 ++++++++++++++---- .../harvest-uk-ab2026-2026-10-29/README.md | 20 +- tests/test_ab2026_harvest_contract.py | 429 +++++++++++-- 3 files changed, 868 insertions(+), 155 deletions(-) diff --git a/pipeline/ab2026_harvest_contract.py b/pipeline/ab2026_harvest_contract.py index b4206e84b..ec8bee44d 100644 --- a/pipeline/ab2026_harvest_contract.py +++ b/pipeline/ab2026_harvest_contract.py @@ -65,11 +65,12 @@ } TIME_BASIS = {"fiscal_year", "annual", "point_in_time"} CONFIDENCE = {"high", "medium", "low"} -# What a malformed row raises when the checks reach it: a field of the wrong -# type, a missing field, an integer with too many digits to print, or nesting -# too deep to print or encode. The checker reports these as problems (review A -# H-6(h); delta review D1-3; final review F3). _number turns away an integer -# too large for a float before any arithmetic, so no check overflows. +# What a malformed row could still raise when the checks reach it, after +# type_problems has turned away each field of the wrong type: an integer with +# too many digits to print, or nesting too deep to print or encode. The +# checker reports these as problems (review A H-6(h); delta review D1-3; +# final review F3). _number turns away an integer too large for a float +# before any arithmetic, so no check overflows. MALFORMED = ( TypeError, AttributeError, @@ -121,16 +122,6 @@ (re.compile(r"%|per ?cent|percentage point|ppts?\b|pp\b", re.I), 0.01), ) NUMBER = re.compile(r"(? 0.0005", "N in M -> N/M") -# (final review F1). # Number words: one to nineteen, and the tens ONES = { "one": 1, @@ -163,16 +154,22 @@ "eighty": 80, "ninety": 90, } -MULTIPLIERS = { - "hundred": 100, - "thousand": 1e3, - "k": 1e3, - "million": 1e6, - "mn": 1e6, - "m": 1e6, - "billion": 1e9, - "bn": 1e9, -} +# A multiplier after a number: a word ("2 million", "a thousand") or, after +# digits, a short form ("2m", "4bn") +LONG = {"hundred": 100, "thousand": 1e3, "million": 1e6, "billion": 1e9} +SHORT = {"bn": 1e9, "mn": 1e6, "m": 1e6, "k": 1e3} +MULTIPLIERS = LONG | SHORT +STEM = re.compile(r"(?:hundred|thousand|million|billion)(?:th|ths|s)") +# A string as tokens: digits, words, whitespace and single characters +TOKEN = re.compile( + r"(?P\d+(?:,\d+)*(?:\.\d+)?)|(?P[^\W\d_]+)|(?P\s+)|(?P.)", + re.S, +) +CURRENCY = "£$€" +DASHES = "-−–" +YEAR = re.compile(r"(?:19|20)\d\d") +GENERIC_RATIO = re.compile(r"\bN\s+in\s+M\b", re.I) +ARROW = re.compile(r"\s*(?:->|→)") def _either(words): @@ -181,41 +178,22 @@ def _either(words): _SEP = r"(?:\s+|\s*-\s*)" -_SMALL = ( - rf"(?:(?:{_either(TENS)}){_SEP}(?:{_either(w for w in ONES if ONES[w] < 10)})" - rf"|{_either(TENS)}|{_either(ONES)})" -) -# A number in words, up to the hundreds ("twenty-five", "a hundred and five") -WORDED = rf"(?:(?:a|{_SMALL})\s+hundred(?:\s+(?:and\s+)?{_SMALL})?|{_SMALL})" -# "a" counts as one before a thousand, million or billion ("over a million") -_A = r"a(?=\s+(?:thousand|million|billion)\b)" -_NUMERAL = r"\d[\d,]*(?:\.\d+)?" -RATIO = re.compile( - rf"(?|→)") -# README rule 4: a locator's leading reference, up to its first comma or -# semicolon, says where HM Treasury prints the number: a data label when it -# names Figure 1.A or 1.B, else a paragraph, table, box or data label. A -# locator that names Figure 1.A or 1.B anywhere, in any of the spellings -# "Figure 1.A", "Fig. 1A" or "Figures 1.A and 1.B", needs such a reference, -# free of any negation ("no data labels", "data label missing", "isn't -# labelled"), and one that calls the number a bar or chart reading reads a -# bar whatever else it names (delta review D1; final review F4). +# README rule 4. A locator that names Figure 1.A or 1.B ("Figure 1.A", "Fig. +# 1A", "Figures 1.A and 1.B") says where HM Treasury prints the number only +# if its first reference, up to its first comma or semicolon, takes one of +# the forms PRINTED reads and nothing more: a paragraph, table or box by its +# number, or a data label. The checker reads any other first reference as a +# bar reading, "Figure 1.A data label unavailable" included (delta review +# R2). A locator that says we read the number off a bar, chart or graph marks +# a bar reading wherever it says so, unless a negation comes just before +# ("not a bar reading", "isn't read off the chart"). FIGURE_1 = re.compile(r"\bfig(?:ure)?s?\.?\s*1\.?\s*[ab]\b", re.I) -PRINTED_AT = re.compile(r"\b(?:paragraph|para|table|box)s?\b", re.I) -DATA_LABEL = re.compile(r"\bdata labels?\b", re.I) -NEGATION = re.compile( - r"\b(?:no|not|none|never|without|missing|absent|lacks?|lacking|unlabell?ed)\b" - r"|n['’]t\b", +_FIG = r"fig(?:ure)?s?\.?\s*1\.?\s*[ab](?:\s+and\s+1\.?\s*[ab])?" +_ID = r"\d+(?:\.[0-9a-z]+)*" +PRINTED = re.compile( + rf"(?:paragraph|para|table|box)(?:s|es)?\.?\s+{_ID}(?:\s+(?:and|to)\s+{_ID})?" + rf"|(?:{_FIG}\s+)?data\s+labels?" + rf"(?:\s+(?:for|on)\s+(?:(?:decile|quintile)\s+\d+|{_FIG}))?", re.I, ) BAR_READING = re.compile( @@ -223,6 +201,9 @@ def _either(words): r"|\bread (?:off|from) (?:the |a )?(?:bars?|charts?|graphs?|figures?)\b", re.I, ) +DENIED = re.compile( + r"(?:\b(?:no|not|never|without)|n['’]t)\s+(?:(?:a|an|the|been)\s+)?\Z", re.I +) # A worded fraction: "half", or a numerator and a denominator ("a third", # "three-fifths", "another quarter"). NUMERATORS = {"a": 1, "an": 1, "another": 1} | { @@ -252,19 +233,245 @@ def _either(words): rf"\b(?:({_either(NUMERATORS)}){_SEP})?({_either(DENOMINATORS)})(?!\w|-\w)", re.I ) +# A run is the longest stretch of number tokens joined by single spaces or +# dashes: digits with one multiplier or none ("2 million", "2.5bn"), or +# number words ("twenty-five", "a hundred and five", "two hundred +# thousand"). The checker reads a run whole or not at all, so it never takes +# part of a number for the number (delta review R1: "twenty-one in a +# hundred" is 0.21, and "4-5" is no one number). +# +# A ratio is a run N, "in", "in every" or "out of", and a run M, joined all +# by spaces or all by hyphens ("1 in 4", "1-in-4", "one in every ten", "3 +# out of 10", "1 in 2 million"). Its share is N/M, negative when a minus +# comes right before N, or after the spaces before M ("-1 in 4", "1 in -4"), +# and N and M count for nothing on their own (delta review D1-1). The +# checker reads no ratio after a currency sign ("£5 in 10 years") or in a +# name ("COVID-19 in 10"). It reads a year after a bare "in" ("5 in 2027", +# "5 in 2026-27") as a year, unless the normalization gives the ratio ("1 in +# 2000 -> 0.0005", or "N in M -> N/M" in any case). It reports any other +# ratio it can't read: a run it can't parse ("a thousand and one in a +# million", "1 in 2 thousand million"), a multiplier's ordinal or plural +# after M ("1 in 2 millionths"), an M joined to a word ("1 in 4-year-olds"), +# or separators it can't place ("1 in - 4", "1 in-4"). -def _worded(text): - """A number in words ("twenty-five", "a hundred and five"), "a" as one.""" - n = 0 - for w in re.split(r"[\s-]+", text.lower()): - n = n * 100 if w == "hundred" else n + ({"a": 1} | ONES | TENS).get(w, 0) - return n +def _tokens(text): + """A string's tokens, as (kind, text, start).""" + return [(m.lastgroup, m.group(), m.start()) for m in TOKEN.finditer(text)] -def _operand(text, multiplier=None): - """A ratio's number, from digits or number words, times its multiplier.""" - n = float(text.replace(",", "")) if text[0].isdigit() else float(_worded(text)) - return n * MULTIPLIERS.get((multiplier or "").lower(), 1) + +def _kind(tokens, i): + """The kind of token i, or None past either end.""" + return tokens[i][0] if 0 <= i < len(tokens) else None + + +def _word(tokens, i): + """The lower-case word at token i, or None.""" + return tokens[i][1].lower() if _kind(tokens, i) == "word" else None + + +def _is(tokens, i, chars): + """Whether token i is one of these characters.""" + return _kind(tokens, i) == "char" and tokens[i][1] in chars + + +def _number_at(tokens, i): + """Whether token i is digits or a number word below a hundred.""" + w = _word(tokens, i) + return _kind(tokens, i) == "num" or w in ONES or w in TENS + + +def _starts_run(tokens, i): + """Whether a number starts at token i: digits, a number word, or "a" or + "an" before a multiplier word ("a thousand", "a-million").""" + if _number_at(tokens, i): + return True + return ( + _word(tokens, i) in ("a", "an") + and (_kind(tokens, i + 1) == "space" or _is(tokens, i + 1, "-")) + and _word(tokens, i + 2) in LONG + ) + + +def _continues(tokens, j, k): + """Whether token k carries on the run whose last token is j.""" + w = _word(tokens, k) + if _number_at(tokens, k) or w in LONG: + return True + if w in SHORT: + return _kind(tokens, j) == "num" + # "and" after a multiplier word, before more of the number ("a hundred + # and five", and so "a thousand and 21", which the checker can't read) + return ( + w == "and" + and _word(tokens, j) in LONG + and _kind(tokens, k + 1) == "space" + and _number_at(tokens, k + 2) + ) + + +def _run_end(tokens, i): + """The last token of the run that starts at token i.""" + j = i + while True: + if _kind(tokens, j) == "num" and _word(tokens, j + 1) in MULTIPLIERS: + j += 1 # a multiplier written on to the digits ("2bn") + elif ( + (_kind(tokens, j + 1) == "space" or _is(tokens, j + 1, DASHES)) + and j + 2 < len(tokens) + and _continues(tokens, j, j + 2) + ): + j += 2 + else: + return j + + +def _small(parts, k): + """The number from one to ninety-nine that the words from k on start + with ("eleven", "twenty", "twenty-five"), as (number, next word), or + None.""" + w = parts[k] if k < len(parts) else None + if w in TENS: + unit = parts[k + 1] if k + 1 < len(parts) else None + if unit in ONES and ONES[unit] < 10: + return TENS[w] + ONES[unit], k + 2 + return TENS[w], k + 1 + if w in ONES: + return ONES[w], k + 1 + return None + + +def _run_value(tokens, i, j): + """The number the run from token i to token j prints, as (number, + multiplier), or None if it doesn't read as one number: digits with one + multiplier or none ("2.5bn", "2 million"), or words up to nine hundred + and ninety-nine with one multiplier or none ("twenty-five", "a hundred + and five", "two hundred thousand", "a million").""" + parts = [t[1].lower() for t in tokens[i : j + 1] if t[0] in ("num", "word")] + if tokens[i][0] == "num": + n = float(parts[0].replace(",", "")) + if not math.isfinite(n): + return None + if len(parts) == 1: + return n, 1 + if len(parts) == 2 and parts[1] in MULTIPLIERS: + return n, MULTIPLIERS[parts[1]] + return None + if parts[0] in ("a", "an"): + n, k = 1, 1 + else: + head = _small(parts, 0) + if head is None: + return None + n, k = head + if k < len(parts) and parts[k] == "hundred": + n, k = n * 100, k + 1 + if k < len(parts) and parts[k] == "and": + k += 1 + if _small(parts, k) is None: + return None + rest = _small(parts, k) + if rest: + n, k = n + rest[0], rest[1] + if k + 1 == len(parts) and parts[k] in ("thousand", "million", "billion"): + return n, LONG[parts[k]] + if parts[0] in ("a", "an") and n == 1: + return None # "a" needs a multiplier + return (n, 1) if k == len(parts) else None + + +def _glued_left(tokens, i): + """Whether the token before i ties a run that starts at i into a name or + a number: "Q1", "COVID-19", ".5".""" + if _kind(tokens, i - 1) in ("num", "word") or _is(tokens, i - 1, "."): + return True + return _is(tokens, i - 1, DASHES) and _kind(tokens, i - 2) == "word" + + +def _joined_right(tokens, j): + """The last token of a word or number joined to the run that ends at j + ("4th", "4-year", "two-child", "4–5"), or None.""" + if _kind(tokens, j + 1) in ("num", "word"): + return j + 1 + if _is(tokens, j + 1, DASHES) and _kind(tokens, j + 2) in ("num", "word"): + return j + 2 + return None + + +def _spoiler(tokens, j): + """The last token of what stops the checker reading a ratio's M, which + ends at token j: a word or number joined to it ("1 in 4-year-olds"), or + a multiplier's ordinal or plural after it ("1 in 2 millionths"); or + None.""" + joined = _joined_right(tokens, j) + if joined is not None: + return joined + if _kind(tokens, j + 1) == "space" and STEM.fullmatch(_word(tokens, j + 2) or ""): + return j + 2 + return None + + +def _seps(tokens, p): + """The spaces and dashes from token p on, as (text, next token).""" + text = "" + while _kind(tokens, p) == "space" or _is(tokens, p, DASHES): + text += tokens[p][1] + p += 1 + return text, p + + +def _sep_kind(sep): + """A ratio separator's kind: "space" for whitespace, "hyphen" for one + hyphen, or None for anything else ("", " - ", "–").""" + if sep == "-": + return "hyphen" + return "space" if sep.isspace() else None + + +def _joint(tokens, p): + """The ratio joint at token p, as (its words, the separators inside it, + the next token), or None.""" + w = _word(tokens, p) + if w == "in": + sep, q = _seps(tokens, p + 1) + if _word(tokens, q) == "every": + return "in every", [sep], q + 1 + return "in", [], p + 1 + if w == "out": + sep, q = _seps(tokens, p + 1) + if _word(tokens, q) == "of": + return "out of", [sep], q + 1 + return None + + +def _sign(tokens, i): + """Whether a minus comes right before token i, after a space, a bracket + or the start of the string, rather than after a word or a number (a name + or a range).""" + return _is(tokens, i - 1, DASHES) and _kind(tokens, i - 2) not in ("num", "word") + + +def _money(tokens, i): + """Whether a currency sign comes before token i, perhaps across a minus + or a space ("£5", "-£5", "£ 5").""" + k = i - 1 + if _is(tokens, k, DASHES): + k -= 1 + if _kind(tokens, k) == "space": + k -= 1 + return _is(tokens, k, CURRENCY) + + +def _year(tokens, k, m): + """Whether the run from token k to token m prints a year: "2027", + "2026-27" or "2026-2027".""" + if not (tokens[k][0] == "num" and YEAR.fullmatch(tokens[k][1])): + return False + return m == k or ( + m == k + 2 + and _is(tokens, k + 1, DASHES) + and re.fullmatch(r"\d\d|(?:19|20)\d\d", tokens[m][1]) is not None + ) def _gives_ratio(normalization, n, m): @@ -281,18 +488,62 @@ def _gives_ratio(normalization, n, m): def _ratios(text, normalization=""): - """The ratios a string prints, as ((start, end), N/M), the share signed - as N is.""" - out = [] - for m in RATIO.finditer(text): - sign, n, joint, d, multiplier = m.groups() - bare = joint.lower() == "in" and not multiplier - if bare and YEAR.fullmatch(d) and not _gives_ratio(normalization, n, d): - continue # "£5 in 2027" - a, b = _operand(n), _operand(d, multiplier) - if b and math.isfinite(a) and math.isfinite(b): - out.append((m.span(), (-a if sign else a) / b)) - return out + """The ratios a string prints, as ((start, end), share), and the text of + each ratio it prints that the checker can't read.""" + tokens = _tokens(text) + found, unread = [], [] + i = 0 + while i < len(tokens): + if not _starts_run(tokens, i): + i += 1 + continue + j = _run_end(tokens, i) + sep1, p = _seps(tokens, j + 1) + joint = _joint(tokens, p) + sep2, k = _seps(tokens, joint[2]) if joint else ("", p) + if not (joint and k < len(tokens) and _starts_run(tokens, k)): + i = j + 1 + continue + words, inner = joint[0], joint[1] + m = _run_end(tokens, k) + n_text = text[tokens[i][2] : tokens[j][2] + len(tokens[j][1])] + m_text = text[tokens[k][2] : tokens[m][2] + len(tokens[m][1])] + if _money(tokens, i) or _glued_left(tokens, i): + i = j + 1 + continue # "£5 in 10 years", "COVID-19 in 10" + m_sign = re.fullmatch(r"\s+[-−–]", sep2) is not None + if ( + words == "in" + and not m_sign + and _year(tokens, k, m) + and not _gives_ratio(normalization, n_text, m_text) + ): + i = j + 1 + continue # "5 in 2027" + n_sign = _sign(tokens, i) + start = tokens[i - 1][2] if n_sign else tokens[i][2] + end = tokens[m][2] + len(tokens[m][1]) + # one kind of separator throughout, so a hyphen is never a sign + kinds = {_sep_kind(s) for s in (sep1, *inner, sep2[:-1] if m_sign else sep2)} + a, b = _run_value(tokens, i, j), _run_value(tokens, k, m) + spoiler = _spoiler(tokens, m) + share = None + if ( + len(kinds) == 1 + and None not in kinds + and spoiler is None + and a is not None + and b is not None + and b[0] * b[1] + ): + share = a[0] * a[1] / (b[0] * b[1]) + if share is not None and math.isfinite(share): + found.append(((start, end), -share if n_sign != m_sign else share)) + else: + last = tokens[m if spoiler is None else spoiler] + unread.append(text[start : last[2] + len(last[1])]) + i = m + 1 + return found, unread def _cut(text, spans): @@ -302,16 +553,40 @@ def _cut(text, spans): return text -def _numbers(text, normalization="", words=False): - """The numbers a string prints, as (exact, fractions). Exact numbers are - its signed digits. Fractions are its ratios ("1 in 4" -> 0.25), whose two - numbers count for nothing on their own, and its worded fractions ("a - third"), which a stored share may round (delta review D1-1). With words, - a string that prints none of these gives its number words ("four", "over - a million") as exact numbers.""" - ratios = _ratios(text, normalization) +def _words_alone(text): + """The numbers a string prints in words, outside names and hyphenated + words ("two-child"), each without its multiplier word, which value_raw's + scale supplies; and the text of each one the checker can't read.""" + tokens = _tokens(text) + out, unread = [], [] + i = 0 + while i < len(tokens): + if tokens[i][0] == "num" or not _starts_run(tokens, i): + i += 1 + continue + j = _run_end(tokens, i) + if not (_glued_left(tokens, i) or _joined_right(tokens, j) is not None): + v = _run_value(tokens, i, j) + if v is None: + unread.append(text[tokens[i][2] : tokens[j][2] + len(tokens[j][1])]) + else: + out.append(float(v[0])) + i = j + 1 + return out, unread + + +def _read(text, normalization="", words=False): + """What a string prints, as (exact, fractions, spans, unread). Exact + numbers are its digits, signed, outside its ratios. Fractions are its + ratios' shares ("1 in 4" -> 0.25) and its worded fractions ("a third"), + which a stored share may round (delta review D1-1). Spans are its ratios' + places, and unread lists each ratio or number in words that the checker + can't read. With words, a string that prints none of these gives its + number words ("four", "over a million").""" + ratios, unread = _ratios(text, normalization) fractions = [share for _, share in ratios] - text = _cut(text, [span for span, _ in ratios]) + spans = [span for span, _ in ratios] + text = _cut(text, spans) for m in FRACTION.finditer(text): top, bottom = (g.lower() if g else g for g in m.groups()) if top or bottom == "half": @@ -325,8 +600,9 @@ def _numbers(text, normalization="", words=False): except ValueError: pass if words and not (exact or fractions): - exact = [float(_worded(m)) for m in WORDS_ALONE.findall(text)] - return exact, fractions + exact, more = _words_alone(text) + unread += more + return exact, fractions, spans, unread def _factors(normalization): @@ -402,7 +678,8 @@ def value_problems(r): of a printed ratio or worded fraction, the fraction itself, may round. A `unit_concept: percent` row stores that share in percent units, 100 times the fraction. A non-categorical row that prints its number only in words - gives those words.""" + gives those words. Delta review R1: a ratio, or a number in words, that + the checker can't read is a problem.""" v = r["value"] if isinstance(v, bool) or not isinstance(v, (int, float)): return ["value is not a number"] @@ -410,11 +687,15 @@ def value_problems(r): return ["value is not finite"] raw = r["value_raw"] if isinstance(r["value_raw"], str) else "" norm = r.get("normalization") or "" - exact, fractions = _numbers(raw, norm, words=r["value_kind"] != "categorical") + exact, fractions, spans, unread = _read( + raw, norm, words=r["value_kind"] != "categorical" + ) + if unread: + return [f"value_raw {raw!r} prints {unread[0]!r}, which the checker can't read"] if not (exact or fractions): return [] # nothing printed to compare (a categorical label) # a ratio's multiplier ("1 in 2 million") scales the ratio, not the string - rest = _cut(raw, [span for span, _ in _ratios(raw, norm)]) + rest = _cut(raw, spans) printed = [n * s for n in exact for s in _scales(rest, norm)] per = 100 if r.get("unit_concept") == "percent" else 1 shares = [per * f for f in fractions] @@ -506,6 +787,81 @@ def proposed_keys(rows): } +# The AB2025 row contract's types (delta review R4): a string in each TEXT +# field, a string or null in each TEXT_OR_NULL field and PUBLICATION field, +# an integer in `period`, and an object of strings in `conditions`. +# value_problems checks `value`. +TEXT = { + "source", + "source_model", + "value_raw", + "normalization", + "value_kind", + "time_basis", + "attribution", + "benchmark_class", + "quote", + "parse_confidence", + "status", +} +TEXT_OR_NULL = { + "reform_hint", + "measure_key", + "baseline_policy", + "source_table", + "source_column", + "metric", + "proposed_metric", + "unit_concept", + "proposed_unit", + "proposed_baseline", + "note", + "local_artifact", +} +PUBLICATION = ( + "title", + "url", + "date", + "publisher", + "primary_sha256", + "locator", + "access", +) + + +def type_problems(r): + """Delta review R4: each field of r whose type the AB2025 row contract + doesn't allow, as strings.""" + out = [ + f"{f} is not a string" + for f in sorted(TEXT & set(r)) + if not isinstance(r[f], str) + ] + out += [ + f"{f} is not a string or null" + for f in sorted(TEXT_OR_NULL & set(r)) + if not (r[f] is None or isinstance(r[f], str)) + ] + period = r.get("period") + if "period" in r and (isinstance(period, bool) or not isinstance(period, int)): + out.append("period is not an integer") + for f in ("conditions", "publication"): + if f in r and not isinstance(r[f], dict): + out.append(f"{f} is not an object") + if isinstance(r.get("conditions"), dict): + for k, v in r["conditions"].items(): + if not (isinstance(k, str) and isinstance(v, str)): + out.append(f"condition {k!r} is not a string") + if isinstance(r.get("publication"), dict): + pub = r["publication"] + out += [ + f"publication.{f} is not a string" + for f in PUBLICATION + if not (pub.get(f) is None or isinstance(pub[f], str)) + ] + return out + + def row_problems(r, keys, earlier, metrics, units): """Every way one staged row breaks the contract, as strings. A malformed row is reported, never raised (review A H-6(h)).""" @@ -524,6 +880,9 @@ def _row_problems(r, keys, earlier, metrics, units): return [f"missing fields {sorted(REQUIRED - fields)}"] if not fields <= REQUIRED | OPTIONAL: out.append(f"unknown fields {sorted(fields - REQUIRED - OPTIONAL)}") + wrong = type_problems(r) + if wrong: + return out + wrong for field, allowed in ( ("attribution", ATTRIBUTION), ("benchmark_class", BENCHMARK_CLASS), @@ -536,14 +895,8 @@ def _row_problems(r, keys, earlier, metrics, units): if r["status"] != "ok": out.append("status is not ok") for field in ("value_raw", "quote"): - if not (isinstance(r[field], str) and r[field].strip()): + if not r[field].strip(): out.append(f"{field} is empty") - if not all(isinstance(r[f], str) for f in ("value_raw", "quote")): - return out - if not isinstance(r["publication"], dict): - return out + ["publication is not an object"] - if r["measure_key"] is not None and not isinstance(r["measure_key"], str): - return out + ["measure_key is not a string or null"] out += value_problems(r) if bool(r.get("metric")) == bool(r.get("proposed_metric")): out.append("exactly one of metric and proposed_metric") @@ -559,11 +912,8 @@ def _row_problems(r, keys, earlier, metrics, units): out.append(f"unit_concept {r['unit_concept']!r} is not a UnitConcept") c = r["conditions"] - if not (isinstance(c, dict) and c): + if not c: return out + ["conditions is not a non-empty object"] - for k, v in c.items(): - if not (isinstance(k, str) and isinstance(v, str)): - out.append(f"condition {k!r} is not a string") if c.get("fiscal_event") != FISCAL_EVENT: out.append(f"conditions.fiscal_event is not {FISCAL_EVENT}") if "fy" in c: @@ -645,16 +995,14 @@ def _row_problems(r, keys, earlier, metrics, units): def reads_a_bar(locator): """README rule 4: whether a locator places the number on a bar of Figure 1.A or 1.B, rather than naming where HM Treasury prints it.""" - if BAR_READING.search(locator): + if any( + not DENIED.search(locator[: m.start()]) for m in BAR_READING.finditer(locator) + ): return True if not FIGURE_1.search(locator): return False - lead = re.split(r"[,;]", locator, maxsplit=1)[0] - if NEGATION.search(lead): - return True - if FIGURE_1.search(lead): - return not DATA_LABEL.search(lead) - return not (PRINTED_AT.search(lead) or DATA_LABEL.search(lead)) + lead = " ".join(re.split(r"[,;]", locator, maxsplit=1)[0].split()) + return not PRINTED.fullmatch(lead) def family_problems(family_dir): diff --git a/sources/harvest-uk-ab2026-2026-10-29/README.md b/sources/harvest-uk-ab2026-2026-10-29/README.md index f2c9bdfac..18f07f300 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/README.md +++ b/sources/harvest-uk-ab2026-2026-10-29/README.md @@ -18,7 +18,7 @@ No code reads this harvest yet; we add the ingest in tranche 2 of #230. In its a We follow the AB2025 harvest README's rules 1–10 and its row contract, with the AB2026 exceptions and additions below: -1. **We set `fiscal_event` to the slug `autumn_budget_2026`** on every row, whatever the publication calls it. +1. **We set `fiscal_event` to the slug `autumn_budget_2026`** on every row, whatever name its publisher uses. 2. **We take measure keys from `data/uk/ab2026_measures.json`.** - We give `ab2026__…` keys to announced measures: every HM Treasury Table 6.1 line, Table 6.2 line 4, and the `ab2026__package_…` packages. - We give `ab2026_option__…` keys to options a producer costed, and we haven't registered any yet. On a row for an unregistered option, or for a package whose scope matches no registered package, we set `measure_key: null`. We put its proposed key in `conditions.proposed_measure_key`, state the key's scope in the family's NOTES, and add the key to the registry before the ingest. @@ -63,12 +63,22 @@ We follow the AB2025 harvest README's rules 1–10 and its row contract, with th - The checker accepts `value_kind: categorical` for a rating, such as the OBR's Table 3.3 uncertainty ratings. - It requires a finite number in `value` that matches `value_raw`'s number within four units in the last place. It takes the scale from `value_raw`'s word (bn, m, k, %) outside any ratio, failing that from the factor in `normalization`, failing that from 1, 10³, 10⁶, 10⁹, 0.01 or 100. -- It reads a ratio ("1 in 4", "1-in-4", "one in every ten", "3 out of 10") as a share, accepts a `value` within 0.1% of that share, and won't match `value` to either of the ratio's numbers alone. It reads "-1 in 4" as −0.25 and "1 in 2 million" as 5e-7. It reads no ratio after a currency sign ("£5 in 10 years"). It reads a year after a bare "in" ("£5 in 2027") as a year, not a denominator, unless we write the ratio in `normalization` ("1 in 2000 -> 0.0005", or "N in M"). +- It reads a ratio ("1 in 4", "1-in-4", "one in every ten", "3 out of 10") as a share, accepts a `value` within 0.1% of that share, and won't match `value` to either of the ratio's numbers alone. It reads "-1 in 4" and "1 in -4" as −0.25, and "1 in 2 million" as 5e-7. It reads no ratio after a currency sign ("£5 in 10 years") or inside a name ("COVID-19 in 10"). It reads a year after a bare "in" ("£5 in 2027") as a year, not a denominator, unless we write the ratio in `normalization` ("1 in 2000 -> 0.0005", or "N in M" in any case). +- In a ratio, it reads digits with one multiplier or none ("2.5bn", "2 million"), and number words up to nine hundred and ninety-nine with one multiplier or none ("twenty-one", "a hundred and five", "two hundred thousand"). It reports a ratio as a problem if it finds: + - a number it can't read that way ("a thousand and one in a million"); + - a word joined to the second number ("1 in 4-year-olds", "1 in 4th"), or a multiplier's plural or ordinal after it ("1 in 2 millionths"); + - separators other than all spaces or all hyphens ("1 in-4", "1 in - 4"). - It reads worded fractions as shares too: "half" alone, or a numerator ("a", "an", "another", or one to ten) and a denominator from "half" to "tenth" ("a third", "three-fifths"). On a `unit_concept: percent` row, it requires 100 times a ratio's or a fraction's share ("1 in 4" as 25). -- It reads number words up to the hundreds ("twenty-five", "a hundred and five") in a ratio. Unless the row has `value_kind: categorical`, it also reads them in a `value_raw` with no other number ("four", "over a million"). +- Unless the row has `value_kind: categorical`, it also reads number words in a `value_raw` with no other number ("four", "over a million"), outside hyphenated words ("two-child"). It reports words it can't read that way ("twenty eleven") as a problem. - It requires `value` to share `value_raw`'s sign unless we note in `normalization` that we took the sign from the wording. -- It treats an HM Treasury Impact on households row as a bar reading (rule 4) if the row has `value_kind: approx_chart_reading`. It does the same if we call the number a bar, chart or graph reading in its locator, or say there that we read it off one. It also treats the row as a bar reading if we name Figure 1.A or 1.B in its locator ("Figure 1.A", "Fig. 1A" or "Figures 1.A and 1.B"), unless we say, in the locator's first reference (up to its first comma or semicolon), where HM Treasury prints the number. The checker requires that reference to name a data label, or a paragraph, table or box without the figure, and to negate nothing ("no data labels", "data label missing"). It ignores later references, so it counts "Figure 1.A, decile 3 bar; see paragraph 1.7" as a bar reading. -- It reports a malformed row as a problem, never as an exception. It counts a row as malformed if the row isn't an object, lacks a field, or has a field of the wrong type, an integer too large for a float or nesting too deep to print. It reports a line it can't parse as JSON, or a file it can't read, the same way, and when it compares rows, it skips any value it can't compare. +- It treats an HM Treasury Impact on households row as a bar reading (rule 4) if the row has `value_kind: approx_chart_reading`. It does the same if we call the number a bar, chart or graph reading in the locator, or say there that we read it off one. It ignores such a phrase right after "no", "not", "never", "without" or "n't" ("not a bar reading"). +- It also treats the row as a bar reading if we name Figure 1.A or 1.B in the locator, unless we say in its first reference where HM Treasury prints the number. It finds the figure in any of the spellings "Figure 1.A", "Fig. 1A" and "Figures 1.A and 1.B", and reads the first reference up to the first comma or semicolon. There it accepts only a paragraph, table or box by number ("paragraph 1.7", "Tables 2.B and 2.C"), or a data label ("Figure 1.A data label", "data label for decile 3"). It treats any other wording as a bar reading, so it rejects "Figure 1.A data label unavailable" and "Figure 1.A, decile 3 bar; see paragraph 1.7". +- It reports a malformed row as a problem, never as an exception: a row that isn't an object, lacks a field, or has a field of the wrong type. It requires: + - a string in `source`, `source_model`, `value_raw`, `normalization`, `quote`, `status`, `value_kind`, `time_basis`, `attribution`, `benchmark_class` and `parse_confidence`; + - a string or null in the other text fields, such as `measure_key`, `reform_hint` and `source_table`, and in each `publication` field; + - an integer in `period`, a number in `value`, and an object in `conditions` and `publication`, with a string in each condition. + + It reports a line it can't parse as JSON, or a file it can't read, the same way. When it compares rows, it skips any value it can't compare, such as an integer too large for a float. - It treats a row with `conditions.component` as a part of one measure (rule 9), and requires `row_type: measure_component`, or the OBR workbook's `measure_total` marker, and a non-blank component. - It allows at most two non-restated, non-categorical rows to share the AB2025 ingest's claim identity, and requires two that do to agree within 0.5%. - It requires a registered metric in `metric`, never in `proposed_metric`. diff --git a/tests/test_ab2026_harvest_contract.py b/tests/test_ab2026_harvest_contract.py index f51864212..e334ae5fc 100644 --- a/tests/test_ab2026_harvest_contract.py +++ b/tests/test_ab2026_harvest_contract.py @@ -21,8 +21,6 @@ from pipeline.ab2026_harvest_contract import ( OGL, - ONES, - TENS, duplicate_problems, enums, family_problems, @@ -31,6 +29,7 @@ reads_a_bar, registry_keys, row_problems, + type_problems, ) ROOT = Path(__file__).resolve().parent.parent @@ -154,9 +153,9 @@ def test_a_staged_row_has_no_problems(): }, "is not ab2026", ), - ({"measure_key": ["ab2026__x"]}, "measure_key is not a string"), + ({"measure_key": ["ab2026__x"]}, "measure_key is not a string or null"), ({"publication": None}, "publication is not an object"), - ({"value_raw": None}, "value_raw is empty"), + ({"value_raw": None}, "value_raw is not a string"), # delta review D1-1: a printed £m cell is that many millions exactly ({"value": -2626312500}, "at a stated scale"), ({"value": -2625100000}, "at a stated scale"), @@ -165,10 +164,10 @@ def test_a_staged_row_has_no_problems(): {"conditions__component": " ", "conditions__row_type": "measure_total"}, "a component row has row_type", ), - ({"conditions__component": 1}, "row is malformed"), + ({"conditions__component": 1}, "condition 'component' is not a string"), ( {"measure_key": None, "conditions__proposed_measure_key": ["ab2026__x"]}, - "row is malformed", + "condition 'proposed_measure_key' is not a string", ), # final review F2: £1 or £2 off a £m cell is no floating-point error ({"value": -2625000001}, "at a stated scale"), @@ -177,6 +176,17 @@ def test_a_staged_row_has_no_problems(): # final review F3: an integer too large for a float ({"value": 10**400}, "value is not finite"), ({"value": -(10**400)}, "value is not finite"), + # delta review R4: a field of a type the AB2025 row contract doesn't + # allow + ({"source_model": []}, "source_model is not a string"), + ({"reform_hint": []}, "reform_hint is not a string or null"), + ({"baseline_policy": []}, "baseline_policy is not a string or null"), + ({"source_table": []}, "source_table is not a string or null"), + ({"source_column": []}, "source_column is not a string or null"), + ({"attribution": ["own"]}, "attribution is not a string"), + ({"period": "2027"}, "period is not an integer"), + ({"period": True}, "period is not an integer"), + ({"publication__locator": 7}, "publication.locator is not a string"), ], ) def test_each_rule_broken_on_a_valid_row_is_a_problem(changes, expected): @@ -225,10 +235,14 @@ def test_a_worded_fraction_is_its_share(): assert any("opposite sign" in p for p in problems(_share("a third", -0.3333, note))) -def test_a_year_after_in_is_no_denominator(): +@pytest.mark.parametrize( + "raw", ["5 in 2027", "5 in 2026-27", "5 in 2026–27", "5 in 2026-2027"] +) +def test_a_year_after_in_is_no_denominator(raw): """A year after "in" prints a year: "5 in 2027" gives 5 (here £5m) and - 2027, not the ratio 5/2027.""" - assert problems(mutated(value_raw="5 in 2027", quote="5 in 2027", value=5e6)) == [] + 2027, not the ratio 5/2027, and "5 in 2026-27" gives a fiscal year with + any dash.""" + assert problems(mutated(value_raw=raw, quote=raw, value=5e6)) == [] def test_a_year_after_in_is_a_denominator_when_the_normalization_says_so(): @@ -302,7 +316,23 @@ def test_a_ratio_passes_only_at_its_share(n, extra, form, minus): ), (raw, wrong) -SAY = {v: w for w, v in (ONES | TENS).items()} +# Number words as a producer prints them, written out here apart from the +# checker's own lists (delta review R1). +SAY = dict( + enumerate( + "zero one two three four five six seven eight nine ten eleven twelve" + " thirteen fourteen fifteen sixteen seventeen eighteen nineteen".split() + ) +) | { + 20: "twenty", + 30: "thirty", + 40: "forty", + 50: "fifty", + 60: "sixty", + 70: "seventy", + 80: "eighty", + 90: "ninety", +} def _say(n, a=False, hyphen=True, joined=True): @@ -392,24 +422,190 @@ def test_a_ratio_with_a_multiplier_or_another_joint_is_its_share(raw, share, wro ("raw", "value", "share"), [ ("£5 in 10 years", 5, 0.5), + ("£ 5 in 10 years", 5, 0.5), ("-£5 in 10", -5, -0.5), + ("£-5 in 10", -5, -0.5), ("$3 in 4", 3, 0.75), - ("1 in 2 millionths", 1, 0.5), - ("1 in 2 million-strong", 2e6, 0.5), - ("1 in 2 thousand million", 2e6, 0.5), + ("COVID-19 in 10", 19, 1.9), + ("Q1 in 4", 4, 0.25), ], ) -def test_a_number_after_a_currency_sign_or_before_an_unread_multiplier_is_no_ratio( - raw, value, share -): - """A number after a currency sign starts no ratio, and a ratio whose - multiplier won't read is none: each number counts at its scale, and the - ratio's share without its multiplier fails.""" +def test_a_number_after_a_currency_sign_or_in_a_name_starts_no_ratio(raw, value, share): + """A number after a currency sign, or in a name, starts no ratio: each + number counts at its scale, and the would-be share fails.""" assert problems(_share(raw, value, "as printed")) == [] found = problems(_share(raw, share, "as printed")) assert any("at a stated scale" in p for p in found), (raw, found) +@pytest.mark.parametrize( + ("raw", "share", "wrongs"), + [ + ("twenty-one in a hundred", 0.21, (0.01, 1, 21, 100)), + ("twenty one in a hundred", 0.21, (0.01, 21, 100)), + ("one in twenty-five", 0.04, (42, 0.2, 1, 25)), + ("a one-in-a-million chance", 1e-6, (1, 1e6)), + ("2025, 1 in 4", 0.25, (1, 4)), + ], +) +def test_a_compound_number_reads_whole(raw, share, wrongs): + """Delta review R1: the checker reads "twenty-one" as 21, never as its + last word, so "twenty-one in a hundred" passes at 0.21 and at nothing + else; and a comma ends a number, so "2025, 1 in 4" is 0.25.""" + assert problems(_share(raw, share, "share as printed")) == [] + for wrong in wrongs: + found = problems(_share(raw, wrong, "share as printed")) + assert any("at a stated scale" in p for p in found), (raw, wrong, found) + + +@pytest.mark.parametrize( + ("raw", "share"), + [ + ("1 in -4", -0.25), + ("-1 in -4", 0.25), + ("1 in −4", -0.25), + ("one in -four", -0.25), + ("-1-in-4", -0.25), + ], +) +def test_a_minus_on_either_number_signs_the_share(raw, share): + """Delta review R1: a minus after the spaces before M signs the share as + a minus before N does, so "1 in -4" passes at -0.25 and fails at 0.25.""" + assert problems(_share(raw, share)) == [] + flipped = problems(_share(raw, -share)) + assert any("opposite sign" in p for p in flipped), (raw, flipped) + + +@pytest.mark.parametrize("normalization", ["N IN M -> N/M", "n in m -> n/m", "N In M"]) +def test_the_generic_normalization_reads_in_any_case(normalization): + """Delta review R1: "N IN M" or "n in m" gives "1 in 2000" as the ratio + 0.0005, as "N in M" does.""" + assert problems(_share("1 in 2000", 0.0005, normalization)) == [] + for wrong in (1, 2000): + found = problems(_share("1 in 2000", wrong, normalization)) + assert any("at a stated scale" in p for p in found), (normalization, wrong) + + +# Ratios the checker can't read whole (delta review R1). "and" after a +# multiplier word joins what follows to the number, so the checker can't +# split "one in a thousand and one in a million" into two ratios either. +CANT_READ = [ + "a thousand and one in a million", + "a thousand and 21 in 100", + "one in a thousand and one in a million", + "twenty eleven in a hundred", + "1 in 2 thousand million", + "1 in 2 millionths", + "1 in 2 thousands", + "1 in 2 million-strong", + "1 in 4-year-olds", + "1 in 4th", + "1 in 4–5", + "4-5 in 10", + "4–5 in 10", + "1 in - 4", + "1 - in 4", + "1in 4", + "1 in4", + "1 in-4", + "1-in 4", + "1 in 0", +] + + +@pytest.mark.parametrize("raw", CANT_READ) +@pytest.mark.parametrize("value", [0.25, 0.5, 1, 4, 2e6, 1e-6]) +def test_a_ratio_the_checker_cant_read_is_a_problem(raw, value): + """Delta review R1: a ratio whose numbers or separators the checker can't + read gives a problem at any value. Before, the checker read part of a + number, or no ratio, and could pass a value the string doesn't print.""" + found = problems(_share(raw, value, "as printed")) + assert any("which the checker can't read" in p for p in found), (raw, found) + + +MULTIPLIER = st.sampled_from( + [("", 1), (" thousand", 1e3), (" million", 1e6), (" billion", 1e9)] + + [("bn", 1e9), ("m", 1e6), ("k", 1e3)] +) + + +def _operand(x, word, styles, multiplier): + """x in words or digits, with a multiplier; a short multiplier ("bn") + only after digits.""" + text, factor = multiplier + if word and not text.startswith(" "): + text, factor = "", 1 + return (_say(x, *styles) if word else str(x)) + text, factor + + +@settings(max_examples=500, deadline=None) +@given( + n=st.integers(1, 999), + m=st.integers(1, 999), + words=st.tuples(st.booleans(), st.booleans()), + styles=STYLES, + multipliers=st.tuples(MULTIPLIER, MULTIPLIER), + signs=st.tuples(st.sampled_from(["", "-", "−"]), st.sampled_from(["", "-", "−"])), + joint=st.sampled_from(["in", "IN", "in every", "In Every", "out of", "OUT OF"]), + hyphens=st.booleans(), +) +def test_any_ratio_reads_whole_and_signed( + n, m, words, styles, multipliers, signs, joint, hyphens +): + """Delta review R1, for every ratio of numbers up to 999, in words or + digits, with a multiplier or none and a minus or none on each number, + joined all by spaces or all by hyphens: the signed share passes, and its + opposite, N, M and the share without its multipliers fail.""" + top, f_top = _operand(n, words[0], styles, multipliers[0]) + bottom, f_bottom = _operand(m, words[1], styles, multipliers[1]) + sign_m = "" if hyphens else signs[1] + if hyphens: + raw = f"{signs[0]}{top}-{joint.replace(' ', '-')}-{bottom}" + else: + raw = f"{signs[0]}{top} {joint} {sign_m}{bottom}" + share = (n * f_top) / (m * f_bottom) * (-1 if signs[0] else 1) + share *= -1 if sign_m else 1 + assert problems(_share(raw, share, "share as printed")) == [], raw + flipped = problems(_share(raw, -share, "share as printed")) + assert any("opposite sign" in p for p in flipped), (raw, flipped) + for bad in (n, m, n * f_top, m * f_bottom, n / m): + if not math.isclose(bad, abs(share), rel_tol=1e-3): + found = problems(_share(raw, bad, "share as printed")) + assert any( + "at a stated scale" in p or "opposite sign" in p for p in found + ), (raw, bad, found) + + +@settings(max_examples=400, deadline=None) +@given( + n=st.sampled_from( + ["21", "twenty-one", "twenty one", "five", "105", "one hundred and five"] + ), + m=st.sampled_from( + ["100", "a hundred", "twenty-five", "2 million", "a thousand", "4bn"] + ), + joint=st.sampled_from(["in", "in every", "out of"]), + spoiler=st.sampled_from( + ["one ", "3 ", "two million ", "a thousand and "] + + [" thousand thousand", " millionths", "-year-olds", " 7"] + ), + value=st.floats(-1e12, 1e12, allow_nan=False), +) +def test_a_ratio_whose_number_runs_on_is_a_problem_at_any_value( + n, m, joint, spoiler, value +): + """Delta review R1, for ratios whose N or M runs on into more of a number + ("one 21 in 100", "1 in 4 7") or into a word ("1 in 2 millionths", "1 in + 4-year-olds"): the checker reports each at any value, rather than read + part of it.""" + if spoiler.endswith(" "): + raw = f"{spoiler}{n} {joint} {m}" + else: + raw = f"{n} {joint} {m}{spoiler}" + found = problems(_share(raw, value, "share as printed")) + assert any("which the checker can't read" in p for p in found), (raw, found) + + def test_a_ratio_s_multiplier_scales_only_the_ratio(): """ "million" in "1 in 2 million" scales M, not the string's other numbers: at the stated factor of 1, the 3 in "3 (1 in 2 million)" is 3.""" @@ -511,6 +707,18 @@ def test_any_number_in_words_reads_as_that_number(n, styles, off): assert any("at a stated scale" in p for p in found), (raw, n + off) +@pytest.mark.parametrize( + "raw", ["a thousand and one", "twenty eleven", "one thousand two hundred"] +) +def test_a_number_in_words_the_checker_cant_read_is_a_problem(raw): + """Delta review R1: words alone that make no one number the checker + reads ("twenty eleven"), or one it doesn't ("one thousand two hundred"), + give a problem, rather than count their parts.""" + for value in (1, 20, 1000, 1001, 1200): + found = problems(_share(raw, value, "printed in words")) + assert any("which the checker can't read" in p for p in found), (raw, found) + + @pytest.mark.parametrize("raw", ["two-child limit", "one-off", "four"]) def test_number_words_in_a_compound_or_on_a_categorical_row_compare_nothing(raw): """A number word inside a hyphenated word ("two-child") prints no number, @@ -559,6 +767,26 @@ def test_a_money_cell_passes_only_at_its_exact_value(cell, off, up): ) +@pytest.mark.parametrize("sign", [1, -1]) +@pytest.mark.parametrize("toward_zero", [False, True]) +def test_a_printed_number_matches_to_four_units_in_the_last_place(sign, toward_zero): + """Delta review R3: a value one to four units in the last place from + £1.7bn, on either side and with either sign, passes, and one five units + off fails.""" + raw = "-£1.7bn" if sign < 0 else "£1.7bn" + exact = (-1.7 if sign < 0 else 1.7) * 1e9 + target = 0.0 if toward_zero else math.copysign(math.inf, exact) + value = exact + for units in range(1, 6): + value = math.nextafter(value, target) + row = mutated(value_raw=raw, quote=raw, value=value, normalization="£bn x 1e9") + found = problems(row) + if units <= 4: + assert found == [], (units, value, found) + else: + assert any("at a stated scale" in p for p in found), (units, value, found) + + def test_malformed_rows_are_reported_not_raised(): """review A H-6(h).""" for row in ( @@ -613,13 +841,31 @@ def test_rule_4_holds_whatever_the_title_case(): ("Figures 1.A and 1.B data labels", False), ("Figure 1 below; paragraph 1.7", False), ("Table 2.C; config 1a", False), + # delta review R2: the first reference takes one of a closed list of + # forms, so a qualifier there, even "(not missing)", reads as a bar; + # and a negation right before a bar-reading phrase cancels it + ("Figure 1.A data label unavailable, decile 3", True), + ("Figure 1.A data label (not missing), decile 3", True), + ("paragraph 1.7 approximately; Figure 1.A", True), + ("Table 2.C and Figure 1.A", True), + ("paragraph 1.7; not a bar reading from Figure 1.A", False), + ("paragraph 1.7; isn't read off Figure 1.A", False), + ("paragraph 1.7; never a chart reading, Figure 1.A", False), + ("paragraph 1.7; a bar reading from Figure 1.A", True), + ("paragraph 1.7; read off Figure 1.A", True), + ("paragraph 1.7; not a bar reading but read off the chart, Figure 1.A", True), + ("paragraphs 1.7 and 1.8; Figure 1.A", False), + ("data labels on Figure 1.B", False), + ("Box 1.A; Figure 1.A", False), + (" paragraph 1.7 ; Figure 1.A", False), ], ) def test_rule_4_reads_where_the_locator_places_the_number(locator, bar_reading): """README rule 4: a locator that places the number on Figure 1.A's or - 1.B's bars marks a bar reading; one whose leading reference names the + 1.B's bars marks a bar reading; one whose first reference names the paragraph, table or data label printing it doesn't (delta review D1: - red book row 357's 20% sits in paragraph 1.7; final review F4).""" + red book row 357's 20% sits in paragraph 1.7; final review F4; delta + review R2).""" row = mutated( publication__title="Budget 2026: Impact on households", publication__locator=locator, @@ -646,8 +892,8 @@ def test_rule_4_reads_where_the_locator_places_the_number(locator, bar_reading): ], ) def test_rule_4_reads_a_negated_data_label_as_a_bar_reading(negated): - """Final review F4: a leading reference that negates its data label, in - any of these words, places the number on a bar.""" + """Final review F4 and delta review R2: a first reference with any word + after its data label, a negation included, places the number on a bar.""" assert reads_a_bar(f"Figure 1.A {negated}") assert not reads_a_bar("Figure 1.A data label") @@ -656,11 +902,14 @@ def test_rule_4_reads_a_negated_data_label_as_a_bar_reading(negated): # that don't; and the parts of a locator that name Figure 1.A or 1.B. PRINTING = [ "paragraph 1.7", + "paragraphs 1.7 and 1.8", "Table 2.C", "Box 1.A", "Figure 1.A data label", "Fig. 1B data labels", "data label", + "data label for decile 3", + "data labels for Figure 1.B", ] NOT_PRINTING = [ "Figure 1.B", @@ -670,14 +919,22 @@ def test_rule_4_reads_a_negated_data_label_as_a_bar_reading(negated): "no data labels", "data label missing", "Figure 1.A data label isn't shown", + "Figure 1.A data label unavailable", + "Figure 1.A data label (not missing)", + "paragraph 1.7 approximately", + "Table 2.C and Figure 1.A", ] NAMES_FIGURE = { "Figure 1.A data label", "Fig. 1B data labels", + "data labels for Figure 1.B", "Figure 1.B", "Figures 1.A and 1.B", "Figure 1.A, decile 3 bar", "Figure 1.A data label isn't shown", + "Figure 1.A data label unavailable", + "Figure 1.A data label (not missing)", + "Table 2.C and Figure 1.A", "discussion of Figure 1.A", "fig 1a", } @@ -697,19 +954,24 @@ def test_rule_4_reads_a_negated_data_label_as_a_bar_reading(negated): "decile 3", "data label", "Table 2.C", + "not a bar reading", + "never read off the chart", + "isn't a chart reading", ] ), max_size=3, ), + affirm=st.booleans(), ) -def test_rule_4_follows_the_leading_reference(lead, rest): - """Final review F4: where a locator names Figure 1.A or 1.B, in any - spelling, only its leading reference, positively naming a paragraph, - table, box or data label, places the number where HM Treasury prints it; - anything later in the locator places nothing.""" - locator = "; ".join([lead, *rest]) +def test_rule_4_follows_the_leading_reference(lead, rest, affirm): + """Final review F4 and delta review R2: where a locator names Figure 1.A + or 1.B, in any spelling, only a first reference in one of the forms + PRINTING lists places the number where HM Treasury prints it, and + anything later in the locator places nothing. A bar-reading phrase + anywhere marks a bar reading, unless a negation comes right before it.""" + locator = "; ".join([lead, *rest, *(["a bar reading"] if affirm else [])]) names_figure = any(part in NAMES_FIGURE for part in [lead, *rest]) - assert reads_a_bar(locator) == (names_figure and lead not in PRINTING) + assert reads_a_bar(locator) == (affirm or (names_figure and lead not in PRINTING)) def test_a_component_with_its_measure_or_a_proposed_key_is_valid(): @@ -884,7 +1146,7 @@ def test_a_family_missing_a_file_is_a_problem(tmp_path): (["not a row", None], "row is not an object"), ( [mutated(measure_key=None, conditions__proposed_measure_key=["ab2026__x"])], - "row is malformed", + "condition 'proposed_measure_key' is not a string", ), ( [mutated(publication__primary_sha256=["abc"])], @@ -931,9 +1193,12 @@ def test_a_field_nested_too_deep_to_print_is_a_problem(): deep = [] for _ in range(100_000): deep = [deep] - found = problems(mutated(period=deep)) - assert [p for p in found if p.startswith("row is malformed (RecursionError")] + assert problems(mutated(period=deep)) == ["period is not an integer"] + locator = mutated(publication__locator=deep) + assert problems(locator) == ["publication.locator is not a string"] + assert problems(mutated(invented=deep)) == ["unknown fields ['invented']"] nested = mutated(conditions__geography_note=deep) + assert problems(nested) == ["condition 'geography_note' is not a string"] assert duplicate_problems([nested, nested]) == [] @@ -974,16 +1239,106 @@ def _changed(changes): return row +# The AB2025 row contract's types, from its README, written out here apart +# from the checker's own lists (delta review R4): each field with the types +# it allows. +STRING_FIELDS = [ + "source", + "source_model", + "value_raw", + "normalization", + "value_kind", + "time_basis", + "attribution", + "benchmark_class", + "quote", + "parse_confidence", + "status", +] +NULLABLE_FIELDS = [ + "reform_hint", + "measure_key", + "baseline_policy", + "source_table", + "source_column", + "metric", + "proposed_metric", + "unit_concept", + "proposed_unit", + "proposed_baseline", + "note", + "local_artifact", +] +PUBLICATION_FIELDS = [ + "title", + "url", + "date", + "publisher", + "primary_sha256", + "locator", + "access", +] +TYPED = ( + [(f, (str,)) for f in STRING_FIELDS] + + [(f, (str, type(None))) for f in NULLABLE_FIELDS] + + [(f"publication__{f}", (str, type(None))) for f in PUBLICATION_FIELDS] + + [(f"conditions__{k}", (str,)) for k in BASE["conditions"]] + + [("period", (int,)), ("value", (int, float))] + + [("conditions", (dict,)), ("publication", (dict,))] +) + + +def _named(path): + """How a problem names a field: "quote", "publication.url" or "condition + 'fy'".""" + if path.startswith("publication__"): + return "publication." + path.split("__")[1] + if path.startswith("conditions__"): + return f"condition {path.split('__')[1]!r}" + return path + + +@pytest.mark.parametrize("field", NULLABLE_FIELDS) +def test_a_nullable_field_takes_null_or_a_string(field): + """Delta review R4: null or a string passes in each field the contract + lets us leave empty.""" + assert type_problems(mutated(**{field: None})) == [] + assert type_problems(mutated(**{field: "x"})) == [] + + +@pytest.mark.parametrize("field", STRING_FIELDS) +def test_a_string_field_takes_no_null(field): + """Delta review R4: null fails in each field that needs a string.""" + assert type_problems(mutated(**{field: None})) == [f"{field} is not a string"] + + +@settings(max_examples=500, deadline=None) +@given(data=st.data(), typed=st.sampled_from(TYPED)) +def test_a_field_of_a_type_the_contract_doesnt_allow_is_a_problem(data, typed): + """Delta review R4, for every field the AB2025 row contract types: any + JSON value of another type gives a problem that names the field.""" + path, allowed = typed + value = data.draw( + JSON.filter(lambda v: isinstance(v, bool) or not isinstance(v, allowed)) + ) + found = problems(_changed({path: value})) + assert any(p.startswith(f"{_named(path)} is not") for p in found), ( + path, + value, + found, + ) + + @settings(max_examples=400, deadline=None) @given( changes=st.dictionaries( st.sampled_from(FIELDS), JSON | st.just(...), min_size=1, max_size=4 ) ) -def test_any_malformed_row_gives_problems_never_an_exception(changes): - """Final review F3, for every input: row_problems, the pair check and the - proposed-key scan return strings for a row with any field changed to any - JSON value or deleted, alone and next to copies of itself and BASE.""" +def test_changing_or_deleting_fields_never_raises(changes): + """Final review F3, for every input: changing or deleting fields does not + make row validation, duplicate checking or proposed-key discovery raise + an exception, for a row alone or next to copies of itself and BASE.""" row = _changed(changes) found = problems(row) assert isinstance(found, list) and all(isinstance(p, str) for p in found) From a9f81e87cb8dbeed1e2f65753da1749a83ed866a Mon Sep 17 00:00:00 2001 From: Max Ghenis Date: Wed, 4 Nov 2026 23:55:35 -0500 Subject: [PATCH 46/52] Read value_raw whole or report it, in answer to the 92f364c7 reviews GPT-6.1 Sol and a cloud session each found more strings that fooled the checker's scanner, which accepted a value that matched any number in the string. We replace the scanner with a closed list of forms: the checker reads value_raw only if the whole string matches a form the README lists, and reports any other string, so we narrow value_raw to the number. The checker reads a bracket only if a percentage restates a ratio or worded fraction to the nearest percentage point. It reads a range only if the range runs upward in value or in size and doesn't join a percentage to a multiplier. Under rule 4, it looks for a bar-reading phrase in any spacing and with hyphens or dashes, and its problem message says how the locator should start. We change no staged row, and all 5,952 pass. Co-Authored-By: Claude Opus 5.5 --- pipeline/ab2026_harvest_contract.py | 756 +++++++---------- .../harvest-uk-ab2026-2026-10-29/README.md | 32 +- tests/test_ab2026_harvest_contract.py | 762 +++++++++++++++--- 3 files changed, 984 insertions(+), 566 deletions(-) diff --git a/pipeline/ab2026_harvest_contract.py b/pipeline/ab2026_harvest_contract.py index ec8bee44d..157e62b5c 100644 --- a/pipeline/ab2026_harvest_contract.py +++ b/pipeline/ab2026_harvest_contract.py @@ -115,14 +115,14 @@ # printed minus (review A H-2(3)). WORDED_SIGN = re.compile(r"sign from the wording|carried negative|-> -\d|→ negative") SCALES = (1, 1e3, 1e6, 1e9, 0.01, 100) -WORD_SCALE = ( - (re.compile(r"\b(bn|billion)\b|\dbn\b", re.I), 1e9), - (re.compile(r"\b(m|million|mn)\b|\dm\b", re.I), 1e6), - (re.compile(r"\b(k|thousand)\b|\dk\b", re.I), 1e3), - (re.compile(r"%|per ?cent|percentage point|ppts?\b|pp\b", re.I), 0.01), -) -NUMBER = re.compile(r"(?\d+(?:,\d+)*(?:\.\d+)?)|(?P[^\W\d_]+)|(?P\s+)|(?P.)", - re.S, -) -CURRENCY = "£$€" -DASHES = "-−–" -YEAR = re.compile(r"(?:19|20)\d\d") -GENERIC_RATIO = re.compile(r"\bN\s+in\s+M\b", re.I) -ARROW = re.compile(r"\s*(?:->|→)") - - -def _either(words): - """A regex alternation of words, the longest first.""" - return "|".join(re.escape(w) for w in sorted(words, key=len, reverse=True)) - - -_SEP = r"(?:\s+|\s*-\s*)" -# README rule 4. A locator that names Figure 1.A or 1.B ("Figure 1.A", "Fig. -# 1A", "Figures 1.A and 1.B") says where HM Treasury prints the number only -# if its first reference, up to its first comma or semicolon, takes one of -# the forms PRINTED reads and nothing more: a paragraph, table or box by its -# number, or a data label. The checker reads any other first reference as a -# bar reading, "Figure 1.A data label unavailable" included (delta review -# R2). A locator that says we read the number off a bar, chart or graph marks -# a bar reading wherever it says so, unless a negation comes just before -# ("not a bar reading", "isn't read off the chart"). -FIGURE_1 = re.compile(r"\bfig(?:ure)?s?\.?\s*1\.?\s*[ab]\b", re.I) -_FIG = r"fig(?:ure)?s?\.?\s*1\.?\s*[ab](?:\s+and\s+1\.?\s*[ab])?" -_ID = r"\d+(?:\.[0-9a-z]+)*" -PRINTED = re.compile( - rf"(?:paragraph|para|table|box)(?:s|es)?\.?\s+{_ID}(?:\s+(?:and|to)\s+{_ID})?" - rf"|(?:{_FIG}\s+)?data\s+labels?" - rf"(?:\s+(?:for|on)\s+(?:(?:decile|quintile)\s+\d+|{_FIG}))?", - re.I, -) -BAR_READING = re.compile( - r"\b(?:bar|chart|graph) readings?\b" - r"|\bread (?:off|from) (?:the |a )?(?:bars?|charts?|graphs?|figures?)\b", - re.I, -) -DENIED = re.compile( - r"(?:\b(?:no|not|never|without)|n['’]t)\s+(?:(?:a|an|the|been)\s+)?\Z", re.I -) -# A worded fraction: "half", or a numerator and a denominator ("a third", -# "three-fifths", "another quarter"). +COUNTS = ONES | TENS | {"a": 1, "an": 1, "zero": 0} +# A multiplier after a number: a word, or after digits a short form +LARGE = {"thousand": 1e3, "million": 1e6, "billion": 1e9, "trillion": 1e12} +SHORT = {"bn": 1e9, "mn": 1e6, "m": 1e6, "k": 1e3, "tn": 1e12} +MULTIPLIERS = LARGE | SHORT | {"hundred": 100} NUMERATORS = {"a": 1, "an": 1, "another": 1} | { w: v for w, v in ONES.items() if v <= 10 } @@ -216,6 +169,8 @@ def _either(words): "thirds": 3, "quarter": 4, "quarters": 4, + "fourth": 4, + "fourths": 4, "fifth": 5, "fifths": 5, "sixth": 6, @@ -229,380 +184,234 @@ def _either(words): "tenth": 10, "tenths": 10, } -FRACTION = re.compile( - rf"\b(?:({_either(NUMERATORS)}){_SEP})?({_either(DENOMINATORS)})(?!\w|-\w)", re.I -) - -# A run is the longest stretch of number tokens joined by single spaces or -# dashes: digits with one multiplier or none ("2 million", "2.5bn"), or -# number words ("twenty-five", "a hundred and five", "two hundred -# thousand"). The checker reads a run whole or not at all, so it never takes -# part of a number for the number (delta review R1: "twenty-one in a -# hundred" is 0.21, and "4-5" is no one number). -# -# A ratio is a run N, "in", "in every" or "out of", and a run M, joined all -# by spaces or all by hyphens ("1 in 4", "1-in-4", "one in every ten", "3 -# out of 10", "1 in 2 million"). Its share is N/M, negative when a minus -# comes right before N, or after the spaces before M ("-1 in 4", "1 in -4"), -# and N and M count for nothing on their own (delta review D1-1). The -# checker reads no ratio after a currency sign ("£5 in 10 years") or in a -# name ("COVID-19 in 10"). It reads a year after a bare "in" ("5 in 2027", -# "5 in 2026-27") as a year, unless the normalization gives the ratio ("1 in -# 2000 -> 0.0005", or "N in M -> N/M" in any case). It reports any other -# ratio it can't read: a run it can't parse ("a thousand and one in a -# million", "1 in 2 thousand million"), a multiplier's ordinal or plural -# after M ("1 in 2 millionths"), an M joined to a word ("1 in 4-year-olds"), -# or separators it can't place ("1 in - 4", "1 in-4"). - - -def _tokens(text): - """A string's tokens, as (kind, text, start).""" - return [(m.lastgroup, m.group(), m.start()) for m in TOKEN.finditer(text)] - - -def _kind(tokens, i): - """The kind of token i, or None past either end.""" - return tokens[i][0] if 0 <= i < len(tokens) else None - - -def _word(tokens, i): - """The lower-case word at token i, or None.""" - return tokens[i][1].lower() if _kind(tokens, i) == "word" else None - - -def _is(tokens, i, chars): - """Whether token i is one of these characters.""" - return _kind(tokens, i) == "char" and tokens[i][1] in chars - - -def _number_at(tokens, i): - """Whether token i is digits or a number word below a hundred.""" - w = _word(tokens, i) - return _kind(tokens, i) == "num" or w in ONES or w in TENS - - -def _starts_run(tokens, i): - """Whether a number starts at token i: digits, a number word, or "a" or - "an" before a multiplier word ("a thousand", "a-million").""" - if _number_at(tokens, i): - return True - return ( - _word(tokens, i) in ("a", "an") - and (_kind(tokens, i + 1) == "space" or _is(tokens, i + 1, "-")) - and _word(tokens, i + 2) in LONG - ) +QUALIFIERS = { + "about", + "almost", + "approx.", + "approximately", + "around", + "at least", + "at most", + "fewer than", + "just over", + "just under", + "less than", + "more than", + "nearly", + "over", + "roughly", + "some", + "under", + "up to", +} +# Words after the number that leave it as it is +TAILS = {"a year", "of gdp"} +# Words that print a zero +ZERO = {"nil", "no change", "no impact", "no overall impact", "none", "nothing"} -def _continues(tokens, j, k): - """Whether token k carries on the run whose last token is j.""" - w = _word(tokens, k) - if _number_at(tokens, k) or w in LONG: - return True - if w in SHORT: - return _kind(tokens, j) == "num" - # "and" after a multiplier word, before more of the number ("a hundred - # and five", and so "a thousand and 21", which the checker can't read) - return ( - w == "and" - and _word(tokens, j) in LONG - and _kind(tokens, k + 1) == "space" - and _number_at(tokens, k + 2) - ) +def _either(words): + """A regex alternation of words, the longest first.""" + return "|".join(re.escape(w) for w in sorted(words, key=len, reverse=True)) -def _run_end(tokens, i): - """The last token of the run that starts at token i.""" - j = i - while True: - if _kind(tokens, j) == "num" and _word(tokens, j + 1) in MULTIPLIERS: - j += 1 # a multiplier written on to the digits ("2bn") - elif ( - (_kind(tokens, j + 1) == "space" or _is(tokens, j + 1, DASHES)) - and j + 2 < len(tokens) - and _continues(tokens, j, j + 2) - ): - j += 2 - else: - return j - - -def _small(parts, k): - """The number from one to ninety-nine that the words from k on start - with ("eleven", "twenty", "twenty-five"), as (number, next word), or - None.""" - w = parts[k] if k < len(parts) else None - if w in TENS: - unit = parts[k + 1] if k + 1 < len(parts) else None - if unit in ONES and ONES[unit] < 10: - return TENS[w] + ONES[unit], k + 2 - return TENS[w], k + 1 - if w in ONES: - return ONES[w], k + 1 - return None +_SIGN = "[-+−–]" +_DIGITS = ( + r"(?:(?:[0-9]{1,3}(?:,[0-9]{3})+|[0-9]+)(?:\.[0-9]+)?|\.[0-9]+)(?:e[-+]?[0-9]+)?" +) +_LARGE = _either(LARGE) +_UNITS = _either(w for w, n in ONES.items() if n < 10) +_TEENS = _either(w for w, n in ONES.items() if n >= 10) +_BELOW_100 = rf"(?:(?:{_either(TENS)})(?:[ -](?:{_UNITS}))?|{_TEENS}|{_UNITS})" +_BELOW_1000 = ( + rf"(?:(?:{_UNITS}|an?)[ -]hundred(?:(?:[ -]and)?[ -]{_BELOW_100})?|{_BELOW_100})" +) +# Number words with one multiplier or none: "twenty-five", "a hundred and +# five", "two hundred thousand", "a million" +_WORDS = rf"(?:zero|{_BELOW_1000}(?:[ -](?:{_LARGE}))?|an?[ -](?:{_LARGE}))" +# After digits, one multiplier: "2.5bn", "2 m", "2 million", "2-million" +_TIMES = rf"(?: ?(?:{_either(SHORT)})|[ -]?(?:hundred|{_LARGE}))" +_PERCENT = r"(?: ?%| per ?cent| percentage points?| ?ppts?| ?pp)" +# Digits with an optional sign and currency sign: "-£1.7bn", "£-1.7bn", "+434m" +_FIGURE = ( + rf"(?P(?:{_SIGN}[£$€]?|[£$€]{_SIGN}?)?)(?P{_DIGITS})" + rf"(?:(?P{_TIMES})|(?P{_PERCENT}))?" +) +FIGURE = re.compile(_FIGURE) +WORDED = re.compile( + rf"(?P{_SIGN})?(?P{_WORDS})(?P{_PERCENT})?", +) +_PART = rf"(?:{_either(NUMERATORS)})[ -](?:{_either(DENOMINATORS)})" +# A worded fraction, perhaps of a multiplier: "half", "three-fifths", "half a +# million", "a quarter of a percentage point" +FRACTION = re.compile( + rf"(?P{_SIGN})?(?Phalf|{_PART})" + rf"(?:(?: of)? an? (?:(?Phundred|{_LARGE})" + rf"|(?Pper ?cent|percentage point)))?" +) +# A whole number and a fraction: "two and a half million", "2 and a third" +MIXED = re.compile( + rf"(?P{_SIGN})?(?P{_DIGITS}|{_BELOW_100}) and (?P{_PART})" + rf"(?:[ -](?P{_LARGE})|(?P{_PERCENT}))?" +) +_COUNT = rf"(?:{_DIGITS}{_TIMES}?|{_WORDS})" +# Two numbers joined all by spaces or all by hyphens. A sign comes before N or, +# with spaces, before M. +RATIO = re.compile( + rf"(?P{_SIGN})?(?P{_COUNT})" + r"(?: (?:in(?: every)?|out of(?: every)?) " + rf"(?P{_SIGN})?|-(?:in(?:-every)?|out-of(?:-every)?)-)" + rf"(?P{_COUNT})" +) +_FIGURE_END = _FIGURE.replace("?P", "?:").replace("?P", "?:") +_FIGURE_END = _FIGURE_END.replace("?P", "?:").replace("?P", "?:") +DASHED = re.compile(rf"(?P{_FIGURE_END})(?:[-–—]| [-–—] )(?P{_FIGURE_END})") +WHOLE = re.compile( + rf"(?:(?:{_either(QUALIFIERS)}) )?(?:an? (?=[-+−–£$€.0-9]))?" + rf"(?P.*?)(?: (?:{_either(TAILS)}))?" +) -def _run_value(tokens, i, j): - """The number the run from token i to token j prints, as (number, - multiplier), or None if it doesn't read as one number: digits with one - multiplier or none ("2.5bn", "2 million"), or words up to nine hundred - and ninety-nine with one multiplier or none ("twenty-five", "a hundred - and five", "two hundred thousand", "a million").""" - parts = [t[1].lower() for t in tokens[i : j + 1] if t[0] in ("num", "word")] - if tokens[i][0] == "num": - n = float(parts[0].replace(",", "")) - if not math.isfinite(n): +def _signed(lead, number): + """The number, negative if its lead holds a minus.""" + return -number if lead and set(lead) & set("-−–") else number + + +def _own(times, percent): + """The scales a number's own word gives it, or None.""" + if percent: + return PERCENT + return (MULTIPLIERS[times.strip(" -")],) if times else None + + +def _in_words(words): + """The number and the multiplier that WORDED's words print.""" + number, large = 0, 1 + for w in re.split("[ -]", words): + if w in LARGE: + large = LARGE[w] + elif w == "hundred": + number *= 100 + elif w != "and": + number += COUNTS[w] + return number, large + + +def _part(text): + """The share that a worded fraction prints.""" + if text == "half": + return 1 / 2 + top, bottom = re.split("[ -]", text) + return NUMERATORS[top] / DENOMINATORS[bottom] + + +def _one_number(text): + """One number, as (number, its own scales or None, its kind), or None. Its + kind says how value_problems compares it: "digits" and "words" exactly, + "rounded" and "share" within 0.1%, and "share" in percent units on a percent + row.""" + m = FIGURE.fullmatch(text) + if m: + number = float(m["digits"].replace(",", "")) + if not math.isfinite(number): return None - if len(parts) == 1: - return n, 1 - if len(parts) == 2 and parts[1] in MULTIPLIERS: - return n, MULTIPLIERS[parts[1]] - return None - if parts[0] in ("a", "an"): - n, k = 1, 1 - else: - head = _small(parts, 0) - if head is None: + own = _own(m["times"], m["percent"]) + return _signed(m["lead"], number), own, "digits" + m = WORDED.fullmatch(text) + if m: + number, large = _in_words(m["words"]) + if m["percent"] and large != 1: return None - n, k = head - if k < len(parts) and parts[k] == "hundred": - n, k = n * 100, k + 1 - if k < len(parts) and parts[k] == "and": - k += 1 - if _small(parts, k) is None: - return None - rest = _small(parts, k) - if rest: - n, k = n + rest[0], rest[1] - if k + 1 == len(parts) and parts[k] in ("thousand", "million", "billion"): - return n, LONG[parts[k]] - if parts[0] in ("a", "an") and n == 1: - return None # "a" needs a multiplier - return (n, 1) if k == len(parts) else None - - -def _glued_left(tokens, i): - """Whether the token before i ties a run that starts at i into a name or - a number: "Q1", "COVID-19", ".5".""" - if _kind(tokens, i - 1) in ("num", "word") or _is(tokens, i - 1, "."): - return True - return _is(tokens, i - 1, DASHES) and _kind(tokens, i - 2) == "word" - - -def _joined_right(tokens, j): - """The last token of a word or number joined to the run that ends at j - ("4th", "4-year", "two-child", "4–5"), or None.""" - if _kind(tokens, j + 1) in ("num", "word"): - return j + 1 - if _is(tokens, j + 1, DASHES) and _kind(tokens, j + 2) in ("num", "word"): - return j + 2 - return None - - -def _spoiler(tokens, j): - """The last token of what stops the checker reading a ratio's M, which - ends at token j: a word or number joined to it ("1 in 4-year-olds"), or - a multiplier's ordinal or plural after it ("1 in 2 millionths"); or - None.""" - joined = _joined_right(tokens, j) - if joined is not None: - return joined - if _kind(tokens, j + 1) == "space" and STEM.fullmatch(_word(tokens, j + 2) or ""): - return j + 2 + own = _own(None, m["percent"]) or ((large,) if large != 1 else None) + return _signed(m["sign"], float(number)), own, "words" + m = FRACTION.fullmatch(text) + if m: + number = _signed(m["sign"], _part(m["fraction"])) + own = _own(m["times"], m["percent"]) + return number, own, "rounded" if own else "share" + m = MIXED.fullmatch(text) + if m: + whole = m["whole"] + digits = whole[0] in "0123456789." + count = float(whole.replace(",", "")) if digits else _in_words(whole)[0] + if not math.isfinite(count): + return None + number = _signed(m["sign"], count + _part(m["fraction"])) + return number, _own(m["times"], m["percent"]), "rounded" return None -def _seps(tokens, p): - """The spaces and dashes from token p on, as (text, next token).""" - text = "" - while _kind(tokens, p) == "space" or _is(tokens, p, DASHES): - text += tokens[p][1] - p += 1 - return text, p - - -def _sep_kind(sep): - """A ratio separator's kind: "space" for whitespace, "hyphen" for one - hyphen, or None for anything else ("", " - ", "–").""" - if sep == "-": - return "hyphen" - return "space" if sep.isspace() else None - - -def _joint(tokens, p): - """The ratio joint at token p, as (its words, the separators inside it, - the next token), or None.""" - w = _word(tokens, p) - if w == "in": - sep, q = _seps(tokens, p + 1) - if _word(tokens, q) == "every": - return "in every", [sep], q + 1 - return "in", [], p + 1 - if w == "out": - sep, q = _seps(tokens, p + 1) - if _word(tokens, q) == "of": - return "out of", [sep], q + 1 +def _count(text): + """A ratio's N or M at its multiplier, or None if it isn't finite.""" + one = _one_number(text) + if one is None: + return None + number, own, _ = one + count = number * (own[0] if own else 1) + return count if math.isfinite(count) else None + + +def _printed(text): + """What one number, ratio or range prints, as a list of _one_number's tuples, + or None.""" + one = _one_number(text) + if one: + return [one] + m = RATIO.fullmatch(text) + if m: + top, bottom = _count(m["n"]), _count(m["m"]) + if top is None or not bottom: + return None + share = _signed(m["sign_n"], _signed(m["sign_m"], top / bottom)) + return [(share, None, "share")] + m = DASHED.fullmatch(text) or re.fullmatch(r"(?P.+) to (?P.+)", text) + if m: + low, high = _one_number(m["a"]), _one_number(m["b"]) + if not (low and high) or (low[1] and not high[1]): + return None + # The first number takes the second's word: "£4-5 million" + if high[1] and not low[1]: + kind = "rounded" if low[2] == "share" else low[2] + low = (low[0], high[1], kind) + # A range joins like with like, and runs upward in value or in size, so + # neither "4% to 2 million" nor "2025-26" is one + if (low[1] == PERCENT) != (high[1] == PERCENT): + return None + a, b = (number * (own[0] if own else 1) for number, own, _ in (low, high)) + if a > b and abs(a) > abs(b): + return None + return [low, high] return None -def _sign(tokens, i): - """Whether a minus comes right before token i, after a space, a bracket - or the start of the string, rather than after a word or a number (a name - or a range).""" - return _is(tokens, i - 1, DASHES) and _kind(tokens, i - 2) not in ("num", "word") - - -def _money(tokens, i): - """Whether a currency sign comes before token i, perhaps across a minus - or a space ("£5", "-£5", "£ 5").""" - k = i - 1 - if _is(tokens, k, DASHES): - k -= 1 - if _kind(tokens, k) == "space": - k -= 1 - return _is(tokens, k, CURRENCY) - - -def _year(tokens, k, m): - """Whether the run from token k to token m prints a year: "2027", - "2026-27" or "2026-2027".""" - if not (tokens[k][0] == "num" and YEAR.fullmatch(tokens[k][1])): +def _restates(first, second): + """Whether a ratio or worded fraction and a percentage, in either order, agree + to the nearest percentage point: "one third (33%)", "25% (1 in 4)".""" + if len(first) != 1 or len(second) != 1: return False - return m == k or ( - m == k + 2 - and _is(tokens, k + 1, DASHES) - and re.fullmatch(r"\d\d|(?:19|20)\d\d", tokens[m][1]) is not None - ) - - -def _gives_ratio(normalization, n, m): - """Whether a normalization gives "n in m" as a ratio: the generic "N in M" - or the printed ratio, followed by an arrow to the share.""" - printed = ( - rf"(? 0.25) and its worded fractions ("a third"), - which a stored share may round (delta review D1-1). Spans are its ratios' - places, and unread lists each ratio or number in words that the checker - can't read. With words, a string that prints none of these gives its - number words ("four", "over a million").""" - ratios, unread = _ratios(text, normalization) - fractions = [share for _, share in ratios] - spans = [span for span, _ in ratios] - text = _cut(text, spans) - for m in FRACTION.finditer(text): - top, bottom = (g.lower() if g else g for g in m.groups()) - if top or bottom == "half": - fractions.append(NUMERATORS[top or "a"] / DENOMINATORS[bottom]) - text = re.sub(r"([-−–])\s*[£$€]", r"\1", text) # "-£1.7bn" - exact = [] - for m in NUMBER.findall(text): - t = m.replace(",", "").replace("−", "-").replace("–", "-") - try: - exact.append(float(t)) - except ValueError: - pass - if words and not (exact or fractions): - exact, more = _words_alone(text) - unread += more - return exact, fractions, spans, unread +def _read(raw): + """What value_raw prints, as a list of _one_number's tuples, or None if + it matches no form the checker reads.""" + text = " ".join(raw.lower().split()) + body = WHOLE.fullmatch(text)["body"] + if body in ZERO: + return [(0.0, (1,), "words")] + both = re.fullmatch(r"(.+) \((.+)\)", body) + if not both: + return _printed(body) + first, second = _printed(both[1]), _printed(both[2]) + return first + second if first and second and _restates(first, second) else None def _factors(normalization): @@ -622,20 +431,6 @@ def _factors(normalization): return out -def _scales(raw, normalization): - """README rule contract, review A H-2(2): the scale comes from value_raw's - word (bn, m, k, %), failing that from the normalization's stated factor, - failing that from a fixed set.""" - for rx, scale in WORD_SCALE: - if rx.search(raw): - # a percentage is stored as a share or in percent units - return {scale, 1.0} if scale == 0.01 else {scale} - factors = _factors(normalization) - if factors: - return set(factors) - return set(SCALES) - - def _number(v): """A finite number. A bool isn't one, and neither is an integer too large for a float (final review F3).""" @@ -673,46 +468,86 @@ def _rounded(a, b): def value_problems(r): """review A H-2: `value` is finite and is value_raw's number at its scale, - with value_raw's sign unless the normalization documents a worded sign. - Delta review D1-1: a printed number matches exactly, and only the share - of a printed ratio or worded fraction, the fraction itself, may round. A - `unit_concept: percent` row stores that share in percent units, 100 times - the fraction. A non-categorical row that prints its number only in words - gives those words. Delta review R1: a ratio, or a number in words, that - the checker can't read is a problem.""" + with value_raw's sign unless the normalization documents a worded sign. A + number's own word sets its scale. Failing that, the normalization's factor + does, failing that a fixed set for digits and 1 for number words. Delta + review D1-1: a printed number matches exactly, and only a fraction or a + ratio's share may round. A `unit_concept: percent` row stores a share in + percent units, 100 times the fraction. A categorical row prints a label + with no digits. Any other row matches a form the checker reads.""" v = r["value"] if isinstance(v, bool) or not isinstance(v, (int, float)): return ["value is not a number"] if not _number(v): return ["value is not finite"] raw = r["value_raw"] if isinstance(r["value_raw"], str) else "" - norm = r.get("normalization") or "" - exact, fractions, spans, unread = _read( - raw, norm, words=r["value_kind"] != "categorical" - ) - if unread: - return [f"value_raw {raw!r} prints {unread[0]!r}, which the checker can't read"] - if not (exact or fractions): - return [] # nothing printed to compare (a categorical label) - # a ratio's multiplier ("1 in 2 million") scales the ratio, not the string - rest = _cut(raw, spans) - printed = [n * s for n in exact for s in _scales(rest, norm)] + if r["value_kind"] == "categorical": + if re.search(r"\d", raw): + return [f"categorical value_raw {raw!r} prints a number"] + return [] + numbers = _read(raw) + if numbers is None: + return [ + f"the checker can't read value_raw {raw!r} as one number; " + "narrow it to the number the row stores" + ] + factors = _factors(r.get("normalization") or "") per = 100 if r.get("unit_concept") == "percent" else 1 - shares = [per * f for f in fractions] + exact, near = [], [] + for number, own, kind in numbers: + if kind == "share": + near.append(per * number) + elif kind == "rounded": + near += [number * s for s in own or (1,)] + else: + fallback = SCALES if kind == "digits" else (1,) + exact += [number * s for s in own or factors or fallback] if not ( - any(_same(abs(v), abs(p)) for p in printed) - or any(_rounded(abs(v), abs(f)) for f in shares) + any(_same(abs(v), abs(p)) for p in exact) + or any(_rounded(abs(v), abs(f)) for f in near) ): return [f"value {v!r} is not value_raw {raw!r} at a stated scale"] if ( - WORDED_SIGN.search(norm) - or any(_same(v, p) for p in printed) - or any(_rounded(v, f) for f in shares) + WORDED_SIGN.search(r.get("normalization") or "") + or any(_same(v, p) for p in exact) + or any(_rounded(v, f) for f in near) ): return [] return [f"value {v!r} has the opposite sign to value_raw {raw!r}"] +# README rule 4. A locator that names Figure 1.A or 1.B ("Figure 1.A", "Fig. +# 1A", "Figures 1.A and 1.B") says where HM Treasury prints the number only +# if its first reference, up to its first comma or semicolon, takes one of +# the forms PRINTED reads and nothing more: a paragraph, table or box by its +# number, or a data label. The checker reads any other first reference as a +# bar reading, "Figure 1.A data label unavailable" included. A locator that +# says we read the number off a bar, chart or graph marks a bar reading +# wherever it says so, unless a negation comes just before ("not a bar +# reading", "isn't read off the chart"). The checker looks for these phrases +# with hyphens and dashes as spaces and with runs of spaces collapsed. +FIGURE_1 = re.compile(r"\bfig(?:ure)?s?\.?\s*1\.?\s*[ab]\b", re.I) +_FIG = r"fig(?:ure)?s?\.?\s*1\.?\s*[ab](?:\s+and\s+1\.?\s*[ab])?" +_ID = r"\d+(?:\.[0-9a-z]+)*" +PRINTED = re.compile( + rf"(?:paragraph|para|table|box)(?:s|es)?\.?\s+{_ID}" + rf"(?:(?:\s+(?:and|to)\s+|\s*[-–]\s*){_ID})?(?:\s+and\s+{_FIG})?" + rf"|(?:{_FIG}\s*:?\s+)?data\s+labels?" + rf"(?:\s+(?:(?:for|on)\s+)?(?:decile|quintile)\s+\d+|\s+(?:for|on)\s+{_FIG})?", + re.I, +) +_VISUAL = r"(?:the |a )?(?:bars?|charts?|graphs?|figures?)" +BAR_READING = re.compile( + r"\b(?:bar|chart|graph)s? read(?:ing)?s?\b" + rf"|\bread(?:s|ing|ings)? (?:off|from) {_VISUAL}\b" + rf"|\bestimated from {_VISUAL}\b", + re.I, +) +DENIED = re.compile( + r"(?:\b(?:no|not|never|without)|n['’]t)\s+(?:(?:a|an|the|been)\s+)?\Z", re.I +) + + def _keys(path): return {m["measure_key"] for m in json.loads(path.read_text())["measures"]} @@ -988,18 +823,21 @@ def _row_problems(r, keys, earlier, metrics, units): or reads_a_bar(pub.get("locator") or "") ) ): - out.append("no HMT Impact on households bar readings (README rule 4)") + out.append( + "no HMT Impact on households bar readings: start the locator with " + "the paragraph, table, box or data label where HM Treasury prints " + "the number (README rule 4)" + ) return out def reads_a_bar(locator): """README rule 4: whether a locator places the number on a bar of Figure 1.A or 1.B, rather than naming where HM Treasury prints it.""" - if any( - not DENIED.search(locator[: m.start()]) for m in BAR_READING.finditer(locator) - ): + plain = " ".join(re.sub(r"[-−–—‐‑]", " ", locator).split()) + if any(not DENIED.search(plain[: m.start()]) for m in BAR_READING.finditer(plain)): return True - if not FIGURE_1.search(locator): + if not FIGURE_1.search(plain): return False lead = " ".join(re.split(r"[,;]", locator, maxsplit=1)[0].split()) return not PRINTED.fullmatch(lead) diff --git a/sources/harvest-uk-ab2026-2026-10-29/README.md b/sources/harvest-uk-ab2026-2026-10-29/README.md index 18f07f300..fe230d826 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/README.md +++ b/sources/harvest-uk-ab2026-2026-10-29/README.md @@ -18,7 +18,7 @@ No code reads this harvest yet; we add the ingest in tranche 2 of #230. In its a We follow the AB2025 harvest README's rules 1–10 and its row contract, with the AB2026 exceptions and additions below: -1. **We set `fiscal_event` to the slug `autumn_budget_2026`** on every row, whatever name its publisher uses. +1. **We set `fiscal_event` to the slug `autumn_budget_2026`** on every row, whatever each producer calls the Budget. 2. **We take measure keys from `data/uk/ab2026_measures.json`.** - We give `ab2026__…` keys to announced measures: every HM Treasury Table 6.1 line, Table 6.2 line 4, and the `ab2026__package_…` packages. - We give `ab2026_option__…` keys to options a producer costed, and we haven't registered any yet. On a row for an unregistered option, or for a package whose scope matches no registered package, we set `measure_key: null`. We put its proposed key in `conditions.proposed_measure_key`, state the key's scope in the family's NOTES, and add the key to the registry before the ingest. @@ -61,24 +61,28 @@ We follow the AB2025 harvest README's rules 1–10 and its row contract, with th `pipeline/ab2026_harvest_contract.py` checks the AB2025 row contract with these additions: -- The checker accepts `value_kind: categorical` for a rating, such as the OBR's Table 3.3 uncertainty ratings. -- It requires a finite number in `value` that matches `value_raw`'s number within four units in the last place. It takes the scale from `value_raw`'s word (bn, m, k, %) outside any ratio, failing that from the factor in `normalization`, failing that from 1, 10³, 10⁶, 10⁹, 0.01 or 100. -- It reads a ratio ("1 in 4", "1-in-4", "one in every ten", "3 out of 10") as a share, accepts a `value` within 0.1% of that share, and won't match `value` to either of the ratio's numbers alone. It reads "-1 in 4" and "1 in -4" as −0.25, and "1 in 2 million" as 5e-7. It reads no ratio after a currency sign ("£5 in 10 years") or inside a name ("COVID-19 in 10"). It reads a year after a bare "in" ("£5 in 2027") as a year, not a denominator, unless we write the ratio in `normalization` ("1 in 2000 -> 0.0005", or "N in M" in any case). -- In a ratio, it reads digits with one multiplier or none ("2.5bn", "2 million"), and number words up to nine hundred and ninety-nine with one multiplier or none ("twenty-one", "a hundred and five", "two hundred thousand"). It reports a ratio as a problem if it finds: - - a number it can't read that way ("a thousand and one in a million"); - - a word joined to the second number ("1 in 4-year-olds", "1 in 4th"), or a multiplier's plural or ordinal after it ("1 in 2 millionths"); - - separators other than all spaces or all hyphens ("1 in-4", "1 in - 4"). -- It reads worded fractions as shares too: "half" alone, or a numerator ("a", "an", "another", or one to ten) and a denominator from "half" to "tenth" ("a third", "three-fifths"). On a `unit_concept: percent` row, it requires 100 times a ratio's or a fraction's share ("1 in 4" as 25). -- Unless the row has `value_kind: categorical`, it also reads number words in a `value_raw` with no other number ("four", "over a million"), outside hyphenated words ("two-child"). It reports words it can't read that way ("twenty eleven") as a problem. +- The checker accepts `value_kind: categorical` for a rating with no digits, such as the OBR's Table 3.3 uncertainty ratings. +- It requires a finite number in `value`. Unless the row has `value_kind: categorical`, it reads `value_raw` only if the whole string, in any case and spacing, matches one of the forms listed here. It reports any other string as a problem, and we narrow `value_raw` to the number, which `quote` still contains: "£5" from "£5 in 10 years", or "1 in 4" from "2025, 1 in 4". It reads: + - digits, with a sign or currency sign and one multiplier or percentage word: "-£1.7bn", "+434m", "2 million", "4%", "0.1 percentage points"; + - number words up to nine hundred and ninety-nine, with a sign and one multiplier or percentage word: "four", "-four", "a hundred and five", "two hundred thousand", "a million", "four per cent"; + - a worded fraction: "half" alone, or a numerator ("a", "an", "another", or one to ten) and a denominator from "half" to "tenth" ("a third", "three-fifths"); + - a worded fraction of a multiplier or percentage: "half a million", "a quarter of a percentage point"; + - a whole number and a fraction, with a multiplier or percentage word if printed: "two and a half", "one and a half million"; + - a ratio of digits or number words with one multiplier or none, joined all by spaces or all by hyphens: "1 in 4", "1-in-4", "3 out of 10", "one out of every five", "1 in 2 million"; + - a range: two figures joined by a dash, with no spaces or a space on each side ("£4-5 million", "4 – 5%"), or two numbers joined by "to" ("four to five", "4bn to 5bn"); + - "nil", "none", "nothing", "no change", "no impact" or "no overall impact", as zero. +- It also reads a qualifier before the number ("about", "up to" and the others in `QUALIFIERS`), "a" before digits ("less than a 0.1 percentage point"), and "a year" or "of GDP" after the number. It reads a ratio or worded fraction with its percentage in brackets, in either order, if the two agree to the nearest percentage point: "one third (33%)", "25% (1 in 4)". It matches `value` to either number there or in a range. It reports a range that joins a percentage to a multiplier ("4% to 2 million"), or whose first number exceeds its second in both value and size ("2025-26", "5 to 4"). +- It takes a number's scale from its own word (bn, million, %), and gives a range's first number the second's word. For digits with no word, it takes the factor in `normalization`, failing that 1, 10³, 10⁶, 10⁹, 0.01 or 100, and for number words with no word, that factor or 1. It reads a percentage as a share or in percent units, and requires `value` to match within four units in the last place. +- It reads a ratio or a worded fraction on its own as a share, and accepts a `value` within 0.1% of that share, or of 100 times it on a `unit_concept: percent` row ("1 in 4" as 25). It won't match `value` to either of a ratio's numbers alone, and it reads "-1 in 4" and "1 in -4" as −0.25. It also allows 0.1% on a whole number with a fraction, and on a fraction of a multiplier or percentage. - It requires `value` to share `value_raw`'s sign unless we note in `normalization` that we took the sign from the wording. -- It treats an HM Treasury Impact on households row as a bar reading (rule 4) if the row has `value_kind: approx_chart_reading`. It does the same if we call the number a bar, chart or graph reading in the locator, or say there that we read it off one. It ignores such a phrase right after "no", "not", "never", "without" or "n't" ("not a bar reading"). -- It also treats the row as a bar reading if we name Figure 1.A or 1.B in the locator, unless we say in its first reference where HM Treasury prints the number. It finds the figure in any of the spellings "Figure 1.A", "Fig. 1A" and "Figures 1.A and 1.B", and reads the first reference up to the first comma or semicolon. There it accepts only a paragraph, table or box by number ("paragraph 1.7", "Tables 2.B and 2.C"), or a data label ("Figure 1.A data label", "data label for decile 3"). It treats any other wording as a bar reading, so it rejects "Figure 1.A data label unavailable" and "Figure 1.A, decile 3 bar; see paragraph 1.7". +- It treats an HM Treasury Impact on households row as a bar reading (rule 4) if the row has `value_kind: approx_chart_reading`. It does the same if we call the number a bar, chart or graph reading in the locator, or say there that we read it off or estimated it from one, in any spacing and with or without hyphens. It ignores such a phrase right after "no", "not", "never", "without" or "n't" ("not a bar reading"). +- It also treats the row as a bar reading if we name Figure 1.A or 1.B in the locator ("Fig. 1A", "Figures 1.A and 1.B"), unless we say in its first reference where HM Treasury prints the number. It reads that reference up to the first comma or semicolon. There we name a paragraph, table or box by number ("paragraph 1.7", "paragraphs 1.7–1.8", "Tables 2.B and 2.C", "Table 2.C and Figure 1.A"), or a data label ("Figure 1.A data label", "data label for decile 3"). It treats any other wording as a bar reading, including "Figure 1.A data label unavailable" and "Figure 1.A, decile 3 bar; see paragraph 1.7". - It reports a malformed row as a problem, never as an exception: a row that isn't an object, lacks a field, or has a field of the wrong type. It requires: - a string in `source`, `source_model`, `value_raw`, `normalization`, `quote`, `status`, `value_kind`, `time_basis`, `attribution`, `benchmark_class` and `parse_confidence`; - a string or null in the other text fields, such as `measure_key`, `reform_hint` and `source_table`, and in each `publication` field; - an integer in `period`, a number in `value`, and an object in `conditions` and `publication`, with a string in each condition. - It reports a line it can't parse as JSON, or a file it can't read, the same way. When it compares rows, it skips any value it can't compare, such as an integer too large for a float. + It also reports a line it can't parse as JSON, or a file it can't read, as a problem. When it compares rows, it skips any value it can't compare, such as an integer too large for a float. - It treats a row with `conditions.component` as a part of one measure (rule 9), and requires `row_type: measure_component`, or the OBR workbook's `measure_total` marker, and a non-blank component. - It allows at most two non-restated, non-categorical rows to share the AB2025 ingest's claim identity, and requires two that do to agree within 0.5%. - It requires a registered metric in `metric`, never in `proposed_metric`. @@ -107,7 +111,7 @@ We pin 12 families and 5,952 rows (450 `restated`) in `COUNTS.json`. | `uk_tpa` | 305 | 10 | `tax_policy_associates` | | `uk_trussell_wpi` | 1 | 1 | `trussell_wpi` | -Each family's `stage.py` writes its claims file, and we keep it on branch `uk/ab2026-harvest-staging` under `harvest-staging/ab2026/lanes///`, with the lane briefs and the five reviews. We record each family's review changes in its NOTES.md, and keep only its three files here, as in AB2025. +Each family's `stage.py` writes its claims file, and we keep it on branch `uk/ab2026-harvest-staging` under `harvest-staging/ab2026/lanes///`, with the lane briefs and the review records. We record each family's review changes in its NOTES.md, and keep only its three files here, as in AB2025. We leave the **families with no rows** out, since AB2025 has no empty family, and keep their working files on the staging branch: diff --git a/tests/test_ab2026_harvest_contract.py b/tests/test_ab2026_harvest_contract.py index e334ae5fc..7cc17ac97 100644 --- a/tests/test_ab2026_harvest_contract.py +++ b/tests/test_ab2026_harvest_contract.py @@ -13,6 +13,7 @@ import math import shutil import tempfile +from itertools import product from pathlib import Path import pytest @@ -217,6 +218,15 @@ def _share(raw, value, normalization="N in M -> N/M"): return mutated(value_raw=raw, quote=raw, value=value, normalization=normalization) +def _cant_read(raw, normalization="as printed"): + """The checker reports raw as unreadable at every value we try.""" + values = (0, 0.25, 0.5, 1, 4, 5, 19, 25, 2e6, 1e-6, -1, -0.25) + return all( + any("can't read" in p for p in problems(_share(raw, v, normalization))) + for v in values + ) + + @pytest.mark.parametrize("wrong", [0.01, 0.04, 1, 4, 25]) def test_a_ratio_is_its_share_not_its_numbers(wrong): """Delta review D1-1: IFS's "1 in 4" (its row 38) is the share 0.25. @@ -235,27 +245,29 @@ def test_a_worded_fraction_is_its_share(): assert any("opposite sign" in p for p in problems(_share("a third", -0.3333, note))) +@pytest.mark.parametrize("raw", ["5 in 2026-27", "5 in 2026–27", "5 in 2026-2027"]) +def test_a_fiscal_year_after_in_is_a_problem(raw): + """The checker gives years no special case: it can't read "5 in 2026-27", + and we narrow value_raw to the number the row stores.""" + assert _cant_read(raw) + + @pytest.mark.parametrize( - "raw", ["5 in 2027", "5 in 2026-27", "5 in 2026–27", "5 in 2026-2027"] + "normalization", + [ + "as printed", + "N in M -> N/M", + "1 in 2000 -> 0.0005", + "£ million x 1,000,000", + "/ 100", + ], ) -def test_a_year_after_in_is_no_denominator(raw): - """A year after "in" prints a year: "5 in 2027" gives 5 (here £5m) and - 2027, not the ratio 5/2027, and "5 in 2026-27" gives a fiscal year with - any dash.""" - assert problems(mutated(value_raw=raw, quote=raw, value=5e6)) == [] - - -def test_a_year_after_in_is_a_denominator_when_the_normalization_says_so(): - """Final review F1: "1 in 2000" stored as the share 0.0005 passes only - when its normalization gives the ratio, as "1 in 2,000" always does.""" - stated = "1 in 2000 -> 0.0005" - assert problems(_share("1 in 2000", 0.0005)) == [] - assert problems(_share("1 in 2000", 0.0005, stated)) == [] - assert problems(_share("1 in 2,000", 0.0005, "share as printed")) == [] - found = problems(_share("1 in 2000", 0.0005, "£ million x 1,000,000")) - assert any("at a stated scale" in p for p in found), found - for wrong in (1, 2000): - found = problems(_share("1 in 2000", wrong, stated)) +def test_a_ratio_reads_the_same_under_any_normalization(normalization): + """A ratio's share takes no factor from the normalization: "1 in 2000" + passes at 0.0005 and fails at 1, at 2000 and at the share at that factor.""" + assert problems(_share("1 in 2000", 0.0005, normalization)) == [] + for wrong in (1, 2000, 500, 0.05): + found = problems(_share("1 in 2000", wrong, normalization)) assert any("at a stated scale" in p for p in found), (wrong, found) @@ -352,6 +364,16 @@ def _say(n, a=False, hyphen=True, joined=True): STYLES = st.tuples(st.booleans(), st.booleans(), st.booleans()) +JOINTS = [ + "in", + "IN", + "in every", + "In Every", + "out of", + "OUT OF", + "out of every", + "Out Of Every", +] @settings(max_examples=300, deadline=None) @@ -359,7 +381,7 @@ def _say(n, a=False, hyphen=True, joined=True): n=st.integers(1, 999), m=st.integers(1, 999), styles=STYLES, - joint=st.sampled_from(["in", "IN", "in every", "In Every", "out of", "OUT OF"]), + joint=st.sampled_from(JOINTS), multiplier=st.sampled_from( [("", 1), (" thousand", 1e3), (" million", 1e6), (" billion", 1e9)] ), @@ -369,8 +391,9 @@ def test_a_worded_ratio_passes_only_at_its_share( n, m, styles, joint, multiplier, wrong ): """Final review F1: number words up to the hundreds give a ratio as - numerals do, after "in", "in every" or "out of" and with a multiplier on - M. The share passes, and so does nothing else, N and M included.""" + numerals do, after "in", "in every", "out of" or "out of every" and with + a multiplier on M. The share passes, and so does nothing else, N and M + included.""" word, factor = multiplier raw = f"{_say(n, *styles)} {joint} {_say(m, *styles)}{word}" share = n / (m * factor) @@ -404,13 +427,20 @@ def test_a_worded_ratio_passes_only_at_its_share( ("1 in every 2000", 0.0005, (1, 2000)), ("1 out of 2000", 0.0005, (1, 2000)), ("1 in 2000 million", 5e-10, (0.0005, 2000, 2e9)), + ("one out of every five", 0.2, (1, 5)), + ("1 out of every 4", 0.25, (1, 4, 25)), + ("1-out-of-every-4", 0.25, (1, 4)), + (".5 in 4", 0.125, (0.5, 4)), + ("1e3 in 4", 250, (1000, 4)), + ("1 in .5", 2, (1, 0.5)), + ("1 in +4", 0.25, (1, 4)), + ("+1 in 4", 0.25, (1, 4)), ], ) def test_a_ratio_with_a_multiplier_or_another_joint_is_its_share(raw, share, wrongs): """A multiplier on M scales M, so "1 in 2 million" is 5e-7, never 0.5. - Number words run to the hundreds, and "in every" and "out of" print a - ratio as "in" does. A year after "in every" or "out of", or before a - multiplier, is a denominator. The share passes; N, M, and the share + Number words run to the hundreds, and "in every", "out of" and "out of + every" print a ratio as "in" does. The share passes; N, M, and the share without the multiplier don't.""" assert problems(_share(raw, share, "share as printed")) == [] for wrong in wrongs: @@ -419,23 +449,23 @@ def test_a_ratio_with_a_multiplier_or_another_joint_is_its_share(raw, share, wro @pytest.mark.parametrize( - ("raw", "value", "share"), + "raw", [ - ("£5 in 10 years", 5, 0.5), - ("£ 5 in 10 years", 5, 0.5), - ("-£5 in 10", -5, -0.5), - ("£-5 in 10", -5, -0.5), - ("$3 in 4", 3, 0.75), - ("COVID-19 in 10", 19, 1.9), - ("Q1 in 4", 4, 0.25), + "£5 in 10 years", + "£ 5 in 10 years", + "£5 in 10", + "-£5 in 10", + "£-5 in 10", + "$3 in 4", + "COVID-19 in 10", + "Q1 in 4", + "£5 in 2027", ], ) -def test_a_number_after_a_currency_sign_or_in_a_name_starts_no_ratio(raw, value, share): - """A number after a currency sign, or in a name, starts no ratio: each - number counts at its scale, and the would-be share fails.""" - assert problems(_share(raw, value, "as printed")) == [] - found = problems(_share(raw, share, "as printed")) - assert any("at a stated scale" in p for p in found), (raw, found) +def test_a_ratio_with_a_currency_sign_or_in_a_name_is_a_problem(raw): + """A ratio takes no currency sign, and "COVID-19 in 10" names something, + so the checker reads neither the numbers nor a share at any value.""" + assert _cant_read(raw) @pytest.mark.parametrize( @@ -444,14 +474,12 @@ def test_a_number_after_a_currency_sign_or_in_a_name_starts_no_ratio(raw, value, ("twenty-one in a hundred", 0.21, (0.01, 1, 21, 100)), ("twenty one in a hundred", 0.21, (0.01, 21, 100)), ("one in twenty-five", 0.04, (42, 0.2, 1, 25)), - ("a one-in-a-million chance", 1e-6, (1, 1e6)), - ("2025, 1 in 4", 0.25, (1, 4)), ], ) def test_a_compound_number_reads_whole(raw, share, wrongs): """Delta review R1: the checker reads "twenty-one" as 21, never as its last word, so "twenty-one in a hundred" passes at 0.21 and at nothing - else; and a comma ends a number, so "2025, 1 in 4" is 0.25.""" + else.""" assert problems(_share(raw, share, "share as printed")) == [] for wrong in wrongs: found = problems(_share(raw, wrong, "share as printed")) @@ -476,20 +504,70 @@ def test_a_minus_on_either_number_signs_the_share(raw, share): assert any("opposite sign" in p for p in flipped), (raw, flipped) -@pytest.mark.parametrize("normalization", ["N IN M -> N/M", "n in m -> n/m", "N In M"]) -def test_the_generic_normalization_reads_in_any_case(normalization): - """Delta review R1: "N IN M" or "n in m" gives "1 in 2000" as the ratio - 0.0005, as "N in M" does.""" - assert problems(_share("1 in 2000", 0.0005, normalization)) == [] - for wrong in (1, 2000): - found = problems(_share("1 in 2000", wrong, normalization)) - assert any("at a stated scale" in p for p in found), (normalization, wrong) - - -# Ratios the checker can't read whole (delta review R1). "and" after a -# multiplier word joins what follows to the number, so the checker can't -# split "one in a thousand and one in a million" into two ratios either. +# Strings that print no one number the checker can read. Each would give a +# wrong value or none if the checker read part of it. CANT_READ = [ + # operands (delta review D1) + "one in zero", + "1 in 0", + "1 in 1e999", + "1e999 in 4", + "1 in 1e308bn", + "1e999", + "1e999 and a half", + "four million per cent", + "zero million", + "hundred", + "million", + # ranges on either side of a ratio (delta review D2) + "4 - 5 in 10", + "4—5 in 10", + "1 in 4 - 5", + "1 in 4—5", + "1 in 4 to 5", + "1 to 4 in 5", + # numbers in words we don't list (delta reviews D3 and D4) + "twenty-one fifths", + "four million; five thousand", + "one thousandth", + "a millionth", + "- four", + # a ratio with more after it (cloud review A1, A3) + "2 million in 5 years", + "5bn in 10 years", + "200,000 in five years", + "one in a hundred and one in ten", + "2025, 1 in 4", + "a one-in-a-million chance", + # separators (cloud review A5) + "1 in-every 4", + "1-in every-4", + "1 out-of 4", + "one in four m", + # commas, signs and characters that print no number + "1,2", + "1,23", + "1234,567", + "12,34,567", + "4.", + "--4", + "٣", + "4", + "4\u200b", + "", + " ", + # ranges the checker can't place + "4 -5", + "4- 5", + "four-five", + "4% to 5", + "4bn-5", + "1 to 2 to 3", + # brackets + "(25%)", + "4 (five", + "4 (five) (six)", + "one third (about 33%)", "a thousand and one in a million", "a thousand and 21 in 100", "one in a thousand and one in a million", @@ -510,17 +588,53 @@ def test_the_generic_normalization_reads_in_any_case(normalization): "1 in-4", "1-in 4", "1 in 0", + # numerators and denominators past the list + "eleven fifths", + "twelve thirds", + "one eleventh", + # a range that runs downward in value and in size + "2025-26", + "5-4", + "5 to 4", + "4bn-5m", + "1.5bn to 200m", + "a half to a third", + # a range with a word on its first number alone + "4k to 5000", + "1m-2000000", + # a range that joins a percentage to a multiplier + "4% to 2 million", + "2 million to 4%", + "4%-5bn", + "0.1 percentage points to 2.5bn", + # brackets that don't restate a share as a percentage + "4 (five)", + "4 (2025)", + "£5bn (2027)", + "one third (50%)", + "1 in 4 (26%)", + "25% (1 in 3)", + "-1 in 4 (25%)", + "half a million (50%)", + "a third (33)", + "a third (1 in 3)", + "25% (25%)", + "25% (0.25)", + "0% (0%)", + "1 in 4 (25-26%)", + "25-26% (1 in 4)", + "3 (1 in 2 million)", ] @pytest.mark.parametrize("raw", CANT_READ) -@pytest.mark.parametrize("value", [0.25, 0.5, 1, 4, 2e6, 1e-6]) -def test_a_ratio_the_checker_cant_read_is_a_problem(raw, value): - """Delta review R1: a ratio whose numbers or separators the checker can't - read gives a problem at any value. Before, the checker read part of a - number, or no ratio, and could pass a value the string doesn't print.""" - found = problems(_share(raw, value, "as printed")) - assert any("which the checker can't read" in p for p in found), (raw, found) +def test_a_string_the_checker_cant_read_is_a_problem(raw): + """Delta reviews R1 and D1 to D4, and cloud review A1 to A5: a string that + prints no one number in a form the checker knows gives a problem at any + value, and the problem says so and asks us to narrow value_raw.""" + assert _cant_read(raw), raw + found = problems(_share(raw, 1, "as printed")) + assert any("narrow it to the number" in p for p in found), found MULTIPLIER = st.sampled_from( @@ -546,16 +660,17 @@ def _operand(x, word, styles, multiplier): styles=STYLES, multipliers=st.tuples(MULTIPLIER, MULTIPLIER), signs=st.tuples(st.sampled_from(["", "-", "−"]), st.sampled_from(["", "-", "−"])), - joint=st.sampled_from(["in", "IN", "in every", "In Every", "out of", "OUT OF"]), + joint=st.sampled_from(JOINTS), hyphens=st.booleans(), ) def test_any_ratio_reads_whole_and_signed( n, m, words, styles, multipliers, signs, joint, hyphens ): """Delta review R1, for every ratio of numbers up to 999, in words or - digits, with a multiplier or none and a minus or none on each number, - joined all by spaces or all by hyphens: the signed share passes, and its - opposite, N, M and the share without its multipliers fail.""" + digits, with a multiplier or none on each number, a minus or none on N, + and a minus or none on M when spaces join the ratio, joined all by spaces + or all by hyphens: the signed share passes, and its opposite, N, M and the + share without its multipliers fail.""" top, f_top = _operand(n, words[0], styles, multipliers[0]) bottom, f_bottom = _operand(m, words[1], styles, multipliers[1]) sign_m = "" if hyphens else signs[1] @@ -603,16 +718,7 @@ def test_a_ratio_whose_number_runs_on_is_a_problem_at_any_value( else: raw = f"{n} {joint} {m}{spoiler}" found = problems(_share(raw, value, "share as printed")) - assert any("which the checker can't read" in p for p in found), (raw, found) - - -def test_a_ratio_s_multiplier_scales_only_the_ratio(): - """ "million" in "1 in 2 million" scales M, not the string's other - numbers: at the stated factor of 1, the 3 in "3 (1 in 2 million)" is 3.""" - note = "count as printed x 1" - assert problems(_share("3 (1 in 2 million)", 3, note)) == [] - found = problems(_share("3 (1 in 2 million)", 3e6, note)) - assert any("at a stated scale" in p for p in found), found + assert any("can't read" in p for p in found), (raw, found) @pytest.mark.parametrize( @@ -643,11 +749,93 @@ def test_a_worded_fraction_in_any_form_is_its_share(raw, share, wrongs): assert any("at a stated scale" in p for p in found), (raw, wrong, found) -@pytest.mark.parametrize("raw", ["a third-party", "the third year", "per quarter"]) -def test_an_ordinal_or_a_compound_is_no_fraction(raw): +# Each numerator and denominator the README lists, written out here apart from +# the checker's tables. +NUMERATOR_WORDS = { + "a": 1, + "an": 1, + "another": 1, + "one": 1, + "two": 2, + "three": 3, + "four": 4, + "five": 5, + "six": 6, + "seven": 7, + "eight": 8, + "nine": 9, + "ten": 10, +} +DENOMINATOR_WORDS = { + "half": 2, + "halves": 2, + "third": 3, + "thirds": 3, + "quarter": 4, + "quarters": 4, + "fourth": 4, + "fourths": 4, + "fifth": 5, + "fifths": 5, + "sixth": 6, + "sixths": 6, + "seventh": 7, + "sevenths": 7, + "eighth": 8, + "eighths": 8, + "ninth": 9, + "ninths": 9, + "tenth": 10, + "tenths": 10, +} + + +def test_every_worded_fraction_reads_as_its_share(): + """Every numerator over every denominator, with a space or a hyphen, passes + at its share and fails at twice that.""" + for top, n in NUMERATOR_WORDS.items(): + for bottom, m in DENOMINATOR_WORDS.items(): + for joint in " -": + raw = f"{top}{joint}{bottom}" + assert problems(_share(raw, n / m, "as printed")) == [], raw + found = problems(_share(raw, 2 * n / m, "as printed")) + assert any("at a stated scale" in p for p in found), (raw, found) + + +@pytest.mark.parametrize( + ("raw", "near", "far"), + [ + ("a third", 0.3331, 0.331), + ("two-thirds", 0.667, 0.663), + ("1 in 3", 0.3331, 0.331), + ("half a million", 500_400, 503_000), + ("a third of a million", 333_100, 331_000), + ("two and a half", 2.502, 2.51), + ], +) +def test_a_rounded_number_passes_within_a_tenth_of_a_percent(raw, near, far): + """A stored value may differ from a share, a fraction of a multiplier or a + whole number with a fraction by 0.1%, and no further.""" + assert problems(_share(raw, near, "as printed")) == [], (raw, near) + found = problems(_share(raw, far, "as printed")) + assert any("at a stated scale" in p for p in found), (raw, far, found) + + +@pytest.mark.parametrize( + "raw", + [ + "a third-party", + "the third year", + "per quarter", + "third", + "quarter", + "2 and half", + ], +) +def test_an_ordinal_or_a_compound_is_a_problem(raw): """ "third" in "a third-party" or "the third year", and "quarter" without - a numerator, print no fraction, so the row compares nothing.""" - assert problems(_share(raw, 7, "as printed")) == [] + a numerator, print no fraction the checker can read.""" + assert _cant_read(raw) @pytest.mark.parametrize( @@ -674,6 +862,21 @@ def row(value, unit): assert any("at a stated scale" in p for p in found), (raw, value, unit) +def test_a_percent_row_scales_only_a_share(): + """A percent row stores a share in percent units. "Two and a half" prints a + number, so it stays 2.5 there.""" + row = mutated( + value_raw="two and a half", + quote="two and a half", + value=2.5, + normalization="as printed", + unit_concept="percent", + ) + assert problems(row) == [] + row["value"] = 250 + assert any("at a stated scale" in p for p in problems(row)) + + @pytest.mark.parametrize( ("raw", "value", "wrongs"), [ @@ -716,23 +919,368 @@ def test_a_number_in_words_the_checker_cant_read_is_a_problem(raw): give a problem, rather than count their parts.""" for value in (1, 20, 1000, 1001, 1200): found = problems(_share(raw, value, "printed in words")) - assert any("which the checker can't read" in p for p in found), (raw, found) + assert any("can't read" in p for p in found), (raw, found) + + +@pytest.mark.parametrize("raw", ["two-child limit", "one-off"]) +def test_a_number_word_inside_a_compound_is_a_problem(raw): + """A number word inside a hyphenated word ("two-child") prints no number.""" + assert _cant_read(raw) -@pytest.mark.parametrize("raw", ["two-child limit", "one-off", "four"]) -def test_number_words_in_a_compound_or_on_a_categorical_row_compare_nothing(raw): - """A number word inside a hyphenated word ("two-child") prints no number, - and a categorical row's words are its label, never its value.""" - if raw != "four": - assert problems(_share(raw, 7, "as printed")) == [] - label = mutated( +def _label(raw, value): + return mutated( value_raw=raw, quote=raw, - value=7, + value=value, normalization="rating as printed", value_kind="categorical", ) - assert problems(label) == [] + + +@pytest.mark.parametrize("raw", ["High", "four", "two-child limit", "one-off"]) +def test_a_categorical_row_prints_a_label_and_compares_nothing(raw): + """A categorical row's words name its label, whatever its value.""" + assert problems(_label(raw, 7)) == [] + + +@pytest.mark.parametrize("raw", ["Tier 2", "3", "1 in 4", "٣"]) +def test_a_categorical_row_that_prints_a_digit_is_a_problem(raw): + """A number on a categorical row would compare nothing, so the checker + takes no digits there.""" + found = problems(_label(raw, 7)) + assert any("prints a number" in p for p in found), (raw, found) + + +@pytest.mark.parametrize( + ("raw", "passes", "fails"), + [ + # a minus on a number in words + ("-four", (-4,), (4,)), + ("−four", (-4,), (4,)), + ("+four", (4,), (-4,)), + ("-three-fifths", (-0.6,), (0.6,)), + ("-half", (-0.5,), (0.5,)), + ("-1 in 4", (-0.25,), (0.25,)), + # a whole number and a fraction + ("two and a half", (2.5,), (0.5, 2, 250)), + ("2 and a third", (7 / 3, 2.3333), (1 / 3, 2)), + ("-two and a half", (-2.5,), (2.5,)), + ("one and a half million", (1.5e6,), (0.5, 1.5, 5e5)), + ("two and a half per cent", (0.025, 2.5), (0.5, 2)), + ("twenty-one and a half", (21.5,), (0.5, 1)), + # a fraction of a multiplier + ("half a million", (5e5,), (0.5, 1e6)), + ("half a hundred", (50,), (0.5, 100)), + ("a quarter of a million", (2.5e5,), (0.25, 1e6)), + ("three quarters of a billion", (7.5e8,), (0.75, 1e9)), + ("half a percentage point", (0.005, 0.5), (1, 2)), + ("a third of a million", (333_333,), (1 / 3, 1e6)), + # a range: the first number takes the second's word + ("£4-5 million", (4e6, 5e6), (4, 5, 4e3)), + ("4-5%", (0.04, 0.05, 4, 5), (40,)), + ("4–5bn", (4e9, 5e9), (4, 5)), + ("4 to 5bn", (4e9, 5e9), (4, 5)), + ("4m to 5bn", (4e6, 5e9), (4e9, 5e6)), + ("4 - 5%", (0.04, 0.05), (45,)), + ("4—5%", (0.04, 0.05), (45,)), + ("four to five", (4, 5), (45, 0.8)), + ("four to five million", (4e6, 5e6), (4, 5)), + ("a third to a half", (1 / 3, 0.5), (0.1,)), + ("a third to half a million", (333_333, 5e5), (1 / 3, 0.5)), + ("-4 to -5", (-4, -5), (4, 5)), + ("-5 to -4", (-5, -4), (5, 4)), + ("200m to 1.5bn", (2e8, 1.5e9), (200, 1.5)), + # a second number in brackets restates the first + ("one third (33%)", (1 / 3, 0.33, 33), (0.5, 3)), + ("1 in 4 (25%)", (0.25, 25), (1, 4)), + ("25% (1 in 4)", (0.25, 25), (1, 4)), + ("half (50%)", (0.5, 50), (2,)), + ("-1 in 4 (-25%)", (-0.25, -25), (0.25, 25)), + ("1 in 8 (13%)", (0.125, 0.13, 13), (1, 8)), + ("1 in 8 (12%)", (0.125, 0.12, 12), (1, 8)), + # words that print a zero + ("nil", (0,), (1,)), + ("nothing", (0,), (1,)), + ("none", (0,), (1,)), + ("no change", (0,), (1,)), + ("no impact", (0,), (1,)), + ("no overall impact", (0,), (1,)), + ("zero", (0,), (1,)), + ("zero in five", (0,), (5, 0.2)), + # words around a number + ("about £2bn a year", (2e9,), (2,)), + ("4 per cent of GDP", (0.04, 4), (40,)), + ("less than a 0.1 percentage point", (0.001, 0.1), (1,)), + ("over a million", (1e6,), (1,)), + # each spelling of a percentage and of a short multiplier + ("4 %", (0.04, 4), (40,)), + ("4 percent", (0.04, 4), (40,)), + ("4 ppts", (0.04, 4), (40,)), + ("+4ppt", (0.04, 4), (40,)), + ("4pp", (0.04, 4), (40,)), + ("2 m", (2e6,), (2,)), + ("2mn", (2e6,), (2,)), + ("2k", (2e3,), (2,)), + ("2tn", (2e12,), (2,)), + ("£3 trillion", (3e12,), (3,)), + ("2-million", (2e6,), (2,)), + ("2 hundred", (200,), (2,)), + # signs and currency in any order + ("-£1.7bn", (-1.7e9,), (1.7e9,)), + ("£-1.7bn", (-1.7e9,), (1.7e9,)), + ("+$2m", (2e6,), (-2e6,)), + ("£5", (5,), (6,)), + ("$5", (5,), (6,)), + ("€5", (5,), (6,)), + ("-$5", (-5,), (5,)), + ("-€2m", (-2e6,), (2e6,)), + ("$-5", (-5,), (5,)), + ("€-2m", (-2e6,), (2e6,)), + ("−2", (-2,), (2,)), + ("–2", (-2,), (2,)), + ("1E-05", (1e-5,), (1,)), + ("1,234.5", (1234.5,), (1.2345,)), + ], +) +def test_each_form_reads_at_its_values_and_no_others(raw, passes, fails): + """The README's closed list of forms. Each string passes at every value + it prints and at no other.""" + for value in passes: + assert problems(_share(raw, value, "as printed")) == [], (raw, value) + for value in fails: + found = problems(_share(raw, value, "as printed")) + assert any("at a stated scale" in p or "opposite sign" in p for p in found), ( + raw, + value, + found, + ) + + +def test_every_number_up_to_999_reads_in_every_style(): + """Exhaustively: each number from 0 to 999, however _say prints it, passes + at itself and fails at itself plus one, alone and as a ratio's N and M.""" + for n in range(1000): + for raw in {_say(n, *style) for style in product([False, True], repeat=3)}: + assert problems(_share(raw, n, "printed in words")) == [], raw + found = problems(_share(raw, n + 1, "printed in words")) + assert any("at a stated scale" in p for p in found), raw + ratio = f"{raw} in a thousand" + assert problems(_share(ratio, n / 1000, "as printed")) == [], raw + if n: + ratio = f"one in {raw}" + assert problems(_share(ratio, 1 / n, "as printed")) == [], raw + + +# Numbers in every form the README lists, each with its value and with no sign, +# written out here apart from the checker's tables. +READABLE = [ + ("5", 5), + ("1,234", 1234), + ("2.5", 2.5), + (".5", 0.5), + ("£5", 5), + ("2m", 2e6), + ("2.5bn", 2.5e9), + ("£2 billion", 2e9), + ("4%", 0.04), + ("4 per cent", 0.04), + ("0.1 percentage points", 0.001), + ("four", 4), + ("twenty-five", 25), + ("a hundred and five", 105), + ("one million", 1e6), + ("a million", 1e6), + ("half", 0.5), + ("a third", 1 / 3), + ("three-fifths", 0.6), + ("half a million", 5e5), + ("two and a half", 2.5), + ("1 in 4", 0.25), + ("one-in-eight", 0.125), + ("3 out of 10", 0.3), + ("one in every five", 0.2), +] +SPOILED = st.sampled_from(READABLE).map(lambda pair: pair[0]) +AFTER = [ + " years", + " people", + " or more", + "th", + "-year-olds", + ", 2025", + "; 2025", + ".", + "/yr", + " per household", + " banana", + " in", + " and", + " (", + ")", + "%%", +] +BEFORE = ["the ", "year ", "£5 ", "and ", "Q", "x", "COVID-", "(", "), ", "per ", "9 "] + + +@settings(max_examples=500, deadline=None) +@given( + number=SPOILED, + before=st.sampled_from(["", *BEFORE]), + after=st.sampled_from(["", *AFTER]), + value=st.floats(-1e12, 1e12, allow_nan=False), +) +def test_a_readable_number_with_more_around_it_is_a_problem_at_any_value( + number, before, after, value +): + """Any number in a listed form, with words or characters around it that + the README doesn't list, gives a problem at any value, so the checker + never reads part of a string.""" + if before or after: + found = problems(_share(f"{before}{number}{after}", value, "as printed")) + assert any("can't read" in p for p in found), (before, number, after, found) + + +@settings(max_examples=300, deadline=None) +@given(number=SPOILED, value=st.floats(-1e12, 1e12, allow_nan=False)) +def test_a_word_inside_a_readable_number_is_a_problem_at_any_value(number, value): + """A word at any space inside a number in a listed form gives a problem.""" + for i, ch in enumerate(number): + if ch == " ": + spoiled = f"{number[:i]} banana{number[i:]}" + found = problems(_share(spoiled, value, "as printed")) + assert any("can't read" in p for p in found), (spoiled, found) + + +@settings(max_examples=300, deadline=None) +@given( + pair=st.sampled_from(READABLE), + qualifier=st.sampled_from( + ["about", "almost", "Approx.", "approximately", "around", "at least"] + + ["at most", "fewer than", "just over", "just under", "less than"] + + ["more than", "nearly", "over", "roughly", "some", "under", "up to"] + ), + tail=st.sampled_from(["", " a year", " of GDP"]), + case=st.sampled_from([str, str.upper, str.title]), + gap=st.sampled_from([" ", " ", "\u00a0", "\t"]), + off=st.sampled_from([1.5, 2, 0.5, 7]), +) +def test_a_number_reads_the_same_with_any_qualifier_case_or_spacing( + pair, qualifier, tail, case, gap, off +): + """A qualifier before a number, "a year" or "of GDP" after it, its case and + its spacing change nothing: it passes at its value and fails elsewhere.""" + number, value = pair + raw = case(f"{qualifier} {number}{tail}").replace(" ", gap) + assert problems(_share(raw, value, "as printed")) == [], raw + found = problems(_share(raw, value * off, "as printed")) + assert any("at a stated scale" in p for p in found), (raw, found) + + +@settings(max_examples=300, deadline=None) +@given(pair=st.sampled_from(READABLE), minus=st.sampled_from(["-", "−", "–"])) +def test_a_minus_flips_a_number_in_any_form(pair, minus): + """A minus before a number gives its opposite: the negative passes, the + positive doesn't, and the unsigned number mirrors both.""" + number, value = pair + raw = minus + number + assert problems(_share(raw, -value, "as printed")) == [], raw + flipped = problems(_share(raw, value, "as printed")) + assert any("opposite sign" in p for p in flipped), (raw, flipped) + assert problems(_share(number, value, "as printed")) == [], number + mirror = problems(_share(number, -value, "as printed")) + assert any("opposite sign" in p for p in mirror), (number, mirror) + + +@settings(max_examples=300, deadline=None) +@given( + n=st.integers(1, 9999), + word=st.sampled_from( + [ + (" thousand", 1e3), + (" million", 1e6), + (" billion", 1e9), + (" trillion", 1e12), + ("k", 1e3), + ("m", 1e6), + ("mn", 1e6), + ("bn", 1e9), + ("tn", 1e12), + ] + ), + other=st.sampled_from([1, 1e3, 1e6, 1e9, 1e12, 0.01, 100]), + normalization=st.sampled_from(["as printed", "£ x 1,000,000", "/ 100"]), +) +def test_a_number_s_own_multiplier_sets_its_scale_under_any_normalization( + n, word, other, normalization +): + """A multiplier scales its own number, and no normalization or fixed scale + overrides it.""" + text, factor = word + raw = f"{n}{text}" + assert problems(_share(raw, n * factor, normalization)) == [], raw + if other != factor: + found = problems(_share(raw, n * other, normalization)) + assert any("at a stated scale" in p for p in found), (raw, other, found) + + +@settings(max_examples=200, deadline=None) +@given(n=st.integers(1, 9999)) +def test_digits_alone_take_the_normalization_s_factor_failing_that_a_fixed_set(n): + """Digits with no word of their own take the normalization's factor and + nothing else. Failing that, they take 1, a thousand, a million, a billion, + 0.01 or 100.""" + raw = str(n) + for scale in (1, 1e3, 1e6, 1e9, 0.01, 100): + assert problems(_share(raw, n * scale, "as printed")) == [], (raw, scale) + assert any( + "at a stated scale" in p for p in problems(_share(raw, n * 7, "as printed")) + ) + stated = "£ million x 1,000,000" + assert problems(_share(raw, n * 1e6, stated)) == [] + for scale in (1, 1e3, 1e9, 0.01, 100): + found = problems(_share(raw, n * scale, stated)) + assert any("at a stated scale" in p for p in found), (raw, scale, found) + + +@pytest.mark.parametrize( + ("raw", "value", "wrong"), + [("four", 4, 40), ("twenty-five", 25, 250), ("a hundred and five", 105, 1050)], +) +def test_words_alone_take_the_normalization_s_factor_failing_that_one( + raw, value, wrong +): + """Number words with no word of their own take the normalization's factor, + failing that 1, and never the fixed set digits fall back on.""" + assert problems(_share(raw, value, "as printed")) == [] + for bad in (wrong, value * 1e3, value * 1e6, value * 0.01, value * 100): + found = problems(_share(raw, bad, "as printed")) + assert any("at a stated scale" in p for p in found), (raw, bad, found) + assert problems(_share(raw, value * 1e6, "measures x 1,000,000")) == [] + found = problems(_share(raw, value, "measures x 1,000,000")) + assert any("at a stated scale" in p for p in found), (raw, found) + + +@settings(max_examples=500, deadline=None) +@given( + text=st.lists( + st.sampled_from( + ["1", "4", "2.5", "1,000", "1e3", "1e999", ".5", "£", "$", "-", "−", "–"] + + ["—", "+", "%", "(", ")", "in", "every", "out", "of", "to", "and", "a"] + + ["an", "half", "third", "quarters", "one", "four", "twenty", "hundred"] + + ["thousand", "million", "bn", "m", "k", "per", "cent", "percentage"] + + ["point", "about", "over", "year", "gdp", "nil", "no", "change", ","] + + [" ", " ", "\u00a0", "\u200b", "٣", "th", "s"] + ), + max_size=8, + ).map("".join), + value=st.floats(-1e12, 1e12, allow_nan=False), +) +def test_any_string_of_number_parts_gives_a_verdict_never_an_exception(text, value): + """The checker reads any arrangement of these parts whole or reports it, + and never raises.""" + found = problems(_share(text, value, "as printed")) + assert isinstance(found, list) and all(isinstance(p, str) for p in found) @settings(max_examples=300, deadline=None) @@ -804,7 +1352,10 @@ def test_rule_4_holds_whatever_the_title_case(): publication__title="Budget 2026: IMPACT ON HOUSEHOLDS", value_kind="approx_chart_reading", ) - assert any("rule 4" in p for p in problems(row)) + found = [p for p in problems(row) if "rule 4" in p] + assert found + # the problem says what the locator should start with (cloud review A6) + assert all("paragraph, table, box or data label" in p for p in found) @pytest.mark.parametrize( @@ -847,7 +1398,6 @@ def test_rule_4_holds_whatever_the_title_case(): ("Figure 1.A data label unavailable, decile 3", True), ("Figure 1.A data label (not missing), decile 3", True), ("paragraph 1.7 approximately; Figure 1.A", True), - ("Table 2.C and Figure 1.A", True), ("paragraph 1.7; not a bar reading from Figure 1.A", False), ("paragraph 1.7; isn't read off Figure 1.A", False), ("paragraph 1.7; never a chart reading, Figure 1.A", False), @@ -858,6 +1408,32 @@ def test_rule_4_holds_whatever_the_title_case(): ("data labels on Figure 1.B", False), ("Box 1.A; Figure 1.A", False), (" paragraph 1.7 ; Figure 1.A", False), + # delta review D5: any spacing, hyphen or case in a bar-reading phrase + ("paragraph 1.7; bar-reading from Figure 1.A", True), + ("paragraph 1.7; BAR READING from Figure 1.A", True), + ("paragraph 1.7; read off the chart from Figure 1.A", True), + ("paragraph 1.7; chart–reading of Figure 1.A", True), + ("paragraph 1.7; bars read by eye, Figure 1.A", True), + ("paragraph 1.7; reading from the chart, Figure 1.A", True), + ("paragraph 1.7; estimated from the figure", True), + ("paragraph 1.7; estimated from Figure 1.A", True), + ("paragraph 1.7; not a bar-reading from Figure 1.A", False), + # a word between the negation and the phrase, or a comma, ends the + # negation (delta review D5) + ("paragraph 1.7; not only a bar reading", True), + ("paragraph 1.7; no, a bar reading", True), + # the figure with a hyphen or a dash in its name + ("Figure 1-A, decile 3 bar", True), + ("Figure 1–B, decile 3 bar", True), + # cloud review A6: forms that name where HM Treasury prints the number + ("Table 2.C and Figure 1.A", False), + ("paragraph 1.7 and Figure 1.A", False), + ("paragraphs 1.7–1.8; Figure 1.A", False), + ("paragraphs 1.7-1.8; Figure 1.A", False), + ("paragraphs 1.7 to 1.8; Figure 1.A", False), + ("Figure 1.A: data label for decile 3", False), + ("Figure 1.A data label decile 3", False), + ("Figure 1.A, data label decile 3", True), ], ) def test_rule_4_reads_where_the_locator_places_the_number(locator, bar_reading): @@ -904,6 +1480,7 @@ def test_rule_4_reads_a_negated_data_label_as_a_bar_reading(negated): "paragraph 1.7", "paragraphs 1.7 and 1.8", "Table 2.C", + "Table 2.C and Figure 1.A", "Box 1.A", "Figure 1.A data label", "Fig. 1B data labels", @@ -922,7 +1499,6 @@ def test_rule_4_reads_a_negated_data_label_as_a_bar_reading(negated): "Figure 1.A data label unavailable", "Figure 1.A data label (not missing)", "paragraph 1.7 approximately", - "Table 2.C and Figure 1.A", ] NAMES_FIGURE = { "Figure 1.A data label", From 9fd73b11d440a2c5c53f470c4e85d49689f15bc2 Mon Sep 17 00:00:00 2001 From: Max Ghenis Date: Thu, 5 Nov 2026 00:49:28 -0500 Subject: [PATCH 47/52] Tighten the checker's scales, ranges, years and rule 4, in answer to the a9f81e87 reviews GPT-6.1 Sol and a cloud session each requested changes at a9f81e87. Digits with no word of their own now take normalization's factor, failing that 1, where they could match at any of six fixed scales. The checker reports a year after a bare "in", a range whose first number exceeds its second unless neither is positive, and a fraction joined to a percentage. Under rule 4, the checker now marks a bar reading wherever we mention a bar, chart or graph in the locator, or say there that we took a number from a figure. It cancels such a phrase after one negation and not after two, and accepts any dash in a printed place. It also reads a fraction with a direct percentage word, a whole number in words up to the hundreds with a fraction, and a lone dash under a nil note. We change no staged row, and all 5,952 pass. Co-Authored-By: Claude Opus 5.5 --- pipeline/ab2026_harvest_contract.py | 106 ++++-- .../harvest-uk-ab2026-2026-10-29/README.md | 22 +- tests/test_ab2026_harvest_contract.py | 357 ++++++++++++++++-- 3 files changed, 410 insertions(+), 75 deletions(-) diff --git a/pipeline/ab2026_harvest_contract.py b/pipeline/ab2026_harvest_contract.py index 157e62b5c..62f119bb9 100644 --- a/pipeline/ab2026_harvest_contract.py +++ b/pipeline/ab2026_harvest_contract.py @@ -114,7 +114,6 @@ # A `normalization` that documents a sign taken from the wording, not from a # printed minus (review A H-2(3)). WORDED_SIGN = re.compile(r"sign from the wording|carried negative|-> -\d|→ negative") -SCALES = (1, 1e3, 1e6, 1e9, 0.01, 100) # A percentage is stored as a share or in percent units. PERCENT = (0.01, 1.0) @@ -208,6 +207,12 @@ TAILS = {"a year", "of gdp"} # Words that print a zero ZERO = {"nil", "no change", "no impact", "no overall impact", "none", "nothing"} +# A dash prints a zero only if the normalization says the producer prints it for +# nil ("dash for nil"), since a table may also print one for "not available". +DASHES = set("-−–—‐‑") +NIL = re.compile(r"\bnil\b", re.I) +# After a bare "in", a year gives no denominator: "5 in 2027" +YEAR = re.compile(r"20[1-9][0-9]") def _either(words): @@ -242,16 +247,18 @@ def _either(words): rf"(?P{_SIGN})?(?P{_WORDS})(?P{_PERCENT})?", ) _PART = rf"(?:{_either(NUMERATORS)})[ -](?:{_either(DENOMINATORS)})" -# A worded fraction, perhaps of a multiplier: "half", "three-fifths", "half a -# million", "a quarter of a percentage point" +# A worded fraction, perhaps of a multiplier or a percentage: "half", +# "three-fifths", "half a million", "half million", "a quarter of a percentage +# point", "half of one per cent", "half per cent" FRACTION = re.compile( rf"(?P{_SIGN})?(?Phalf|{_PART})" - rf"(?:(?: of)? an? (?:(?Phundred|{_LARGE})" - rf"|(?Pper ?cent|percentage point)))?" + r"(?:(?:(?: of)? (?:an?|one))?" + rf"(?: (?Phundred|{_LARGE})|(?P{_PERCENT})))?" ) -# A whole number and a fraction: "two and a half million", "2 and a third" +# A whole number and a fraction: "two and a half million", "2 and a third", "one +# hundred and a half" MIXED = re.compile( - rf"(?P{_SIGN})?(?P{_DIGITS}|{_BELOW_100}) and (?P{_PART})" + rf"(?P{_SIGN})?(?P{_DIGITS}|{_BELOW_1000}) and (?P{_PART})" rf"(?:[ -](?P{_LARGE})|(?P{_PERCENT}))?" ) _COUNT = rf"(?:{_DIGITS}{_TIMES}?|{_WORDS})" @@ -359,6 +366,9 @@ def _printed(text): return [one] m = RATIO.fullmatch(text) if m: + joint = text[m.end("n") : m.start("m")].strip(" -+−–") + if joint == "in" and YEAR.fullmatch(m["m"]): + return None top, bottom = _count(m["n"]), _count(m["m"]) if top is None or not bottom: return None @@ -369,16 +379,21 @@ def _printed(text): low, high = _one_number(m["a"]), _one_number(m["b"]) if not (low and high) or (low[1] and not high[1]): return None + # A fraction before a percentage could read as a share or as part of a + # percentage point: "a third to 40%" + if low[2] == "share" and high[1] == PERCENT: + return None # The first number takes the second's word: "£4-5 million" if high[1] and not low[1]: kind = "rounded" if low[2] == "share" else low[2] low = (low[0], high[1], kind) - # A range joins like with like, and runs upward in value or in size, so - # neither "4% to 2 million" nor "2025-26" is one + # A range joins like with like, and only a range of no positive number + # may run downward, so "4% to 2 million", "2025-26" and "2 to -3" aren't + # ranges, and "-4 to -5" is if (low[1] == PERCENT) != (high[1] == PERCENT): return None a, b = (number * (own[0] if own else 1) for number, own, _ in (low, high)) - if a > b and abs(a) > abs(b): + if a > 0 and a > b: return None return [low, high] return None @@ -400,10 +415,12 @@ def _restates(first, second): ) -def _read(raw): +def _read(raw, normalization=""): """What value_raw prints, as a list of _one_number's tuples, or None if it matches no form the checker reads.""" text = " ".join(raw.lower().split()) + if text in DASHES and NIL.search(normalization): + return [(0.0, (1,), "words")] body = WHOLE.fullmatch(text)["body"] if body in ZERO: return [(0.0, (1,), "words")] @@ -470,11 +487,11 @@ def value_problems(r): """review A H-2: `value` is finite and is value_raw's number at its scale, with value_raw's sign unless the normalization documents a worded sign. A number's own word sets its scale. Failing that, the normalization's factor - does, failing that a fixed set for digits and 1 for number words. Delta - review D1-1: a printed number matches exactly, and only a fraction or a - ratio's share may round. A `unit_concept: percent` row stores a share in - percent units, 100 times the fraction. A categorical row prints a label - with no digits. Any other row matches a form the checker reads.""" + does, failing that 1. Delta review D1-1: a printed number matches exactly, + and only a fraction or a ratio's share may round. A `unit_concept: percent` + row stores a share in percent units, 100 times the fraction. A categorical + row prints a label with no digits. Any other row matches a form the checker + reads.""" v = r["value"] if isinstance(v, bool) or not isinstance(v, (int, float)): return ["value is not a number"] @@ -485,7 +502,7 @@ def value_problems(r): if re.search(r"\d", raw): return [f"categorical value_raw {raw!r} prints a number"] return [] - numbers = _read(raw) + numbers = _read(raw, r.get("normalization") or "") if numbers is None: return [ f"the checker can't read value_raw {raw!r} as one number; " @@ -500,8 +517,7 @@ def value_problems(r): elif kind == "rounded": near += [number * s for s in own or (1,)] else: - fallback = SCALES if kind == "digits" else (1,) - exact += [number * s for s in own or factors or fallback] + exact += [number * s for s in own or factors or (1,)] if not ( any(_same(abs(v), abs(p)) for p in exact) or any(_rounded(abs(v), abs(f)) for f in near) @@ -521,26 +537,37 @@ def value_problems(r): # if its first reference, up to its first comma or semicolon, takes one of # the forms PRINTED reads and nothing more: a paragraph, table or box by its # number, or a data label. The checker reads any other first reference as a -# bar reading, "Figure 1.A data label unavailable" included. A locator that -# says we read the number off a bar, chart or graph marks a bar reading -# wherever it says so, unless a negation comes just before ("not a bar -# reading", "isn't read off the chart"). The checker looks for these phrases -# with hyphens and dashes as spaces and with runs of spaces collapsed. +# bar reading, "Figure 1.A data label unavailable" included. Anywhere in a +# locator, a bar, a chart or a graph marks a bar reading, and so does a number +# taken from a figure. A single negation just before the phrase cancels it ("not +# a bar reading", "didn't read it off the chart"), and a double negation doesn't +# ("not without a bar reading"). The checker looks for these phrases with hyphens +# and dashes as spaces and with runs of spaces collapsed, and reads the first +# reference with each dash as a hyphen. +_DASH = "[-−–—‐‑]" FIGURE_1 = re.compile(r"\bfig(?:ure)?s?\.?\s*1\.?\s*[ab]\b", re.I) -_FIG = r"fig(?:ure)?s?\.?\s*1\.?\s*[ab](?:\s+and\s+1\.?\s*[ab])?" +_FIG = r"fig(?:ure)?s?\.?\s*1[.-]?\s*[ab](?:\s+and\s+1[.-]?\s*[ab])?" _ID = r"\d+(?:\.[0-9a-z]+)*" PRINTED = re.compile( rf"(?:paragraph|para|table|box)(?:s|es)?\.?\s+{_ID}" - rf"(?:(?:\s+(?:and|to)\s+|\s*[-–]\s*){_ID})?(?:\s+and\s+{_FIG})?" - rf"|(?:{_FIG}\s*:?\s+)?data\s+labels?" + rf"(?:(?:\s+(?:and|to)\s+|\s*-\s*){_ID})?(?:\s+and\s+{_FIG})?" + rf"|(?:{_FIG}\s*:?\s+)?data[\s-]+labels?" rf"(?:\s+(?:(?:for|on)\s+)?(?:decile|quintile)\s+\d+|\s+(?:for|on)\s+{_FIG})?", re.I, ) -_VISUAL = r"(?:the |a )?(?:bars?|charts?|graphs?|figures?)" +_PICTURE = r"(?:bars?|charts?|graphs?)" +_OFF = r"(?:off|from)" +_UP_TO = r"(?:[\w.]+ ){0,4}?" +# A verb's whole phrase comes first, so that a negation before the verb cancels +# it: "read it off the decile 3 bar", "reading of the chart". Then any bar, chart +# or graph, with a reading if it names one ("bar reading from Figure 1.A"), and +# any number from a figure. BAR_READING = re.compile( - r"\b(?:bar|chart|graph)s? read(?:ing)?s?\b" - rf"|\bread(?:s|ing|ings)? (?:off|from) {_VISUAL}\b" - rf"|\bestimated from {_VISUAL}\b", + r"\b(?:(?:read|estimat|tak|measur)\w*|took)(?: [\w.]+){0,3}?" + rf" {_OFF} {_UP_TO}(?:{_PICTURE}|figures?)\b" + rf"|\breadings? of {_UP_TO}(?:{_PICTURE}|figures?)\b" + rf"|\b{_PICTURE}(?: read\w*(?: {_OFF} {_UP_TO}figures?)?)?\b" + rf"|\b{_OFF} {_UP_TO}figures?\b", re.I, ) DENIED = re.compile( @@ -826,21 +853,28 @@ def _row_problems(r, keys, earlier, metrics, units): out.append( "no HMT Impact on households bar readings: start the locator with " "the paragraph, table, box or data label where HM Treasury prints " - "the number (README rule 4)" + "the number, and call nothing there a bar, chart or graph (README " + "rule 4)" ) return out +def _denied(before): + """Whether one negation, and no more, comes right before a phrase.""" + m = DENIED.search(before) + return bool(m) and not DENIED.search(before[: m.start()]) + + def reads_a_bar(locator): """README rule 4: whether a locator places the number on a bar of Figure 1.A or 1.B, rather than naming where HM Treasury prints it.""" - plain = " ".join(re.sub(r"[-−–—‐‑]", " ", locator).split()) - if any(not DENIED.search(plain[: m.start()]) for m in BAR_READING.finditer(plain)): + plain = " ".join(re.sub(_DASH, " ", locator).split()) + if any(not _denied(plain[: m.start()]) for m in BAR_READING.finditer(plain)): return True if not FIGURE_1.search(plain): return False - lead = " ".join(re.split(r"[,;]", locator, maxsplit=1)[0].split()) - return not PRINTED.fullmatch(lead) + first = re.split(r"[,;]", re.sub(_DASH, "-", locator), maxsplit=1)[0] + return not PRINTED.fullmatch(" ".join(first.split())) def family_problems(family_dir): diff --git a/sources/harvest-uk-ab2026-2026-10-29/README.md b/sources/harvest-uk-ab2026-2026-10-29/README.md index fe230d826..8f4b5077d 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/README.md +++ b/sources/harvest-uk-ab2026-2026-10-29/README.md @@ -63,20 +63,20 @@ We follow the AB2025 harvest README's rules 1–10 and its row contract, with th - The checker accepts `value_kind: categorical` for a rating with no digits, such as the OBR's Table 3.3 uncertainty ratings. - It requires a finite number in `value`. Unless the row has `value_kind: categorical`, it reads `value_raw` only if the whole string, in any case and spacing, matches one of the forms listed here. It reports any other string as a problem, and we narrow `value_raw` to the number, which `quote` still contains: "£5" from "£5 in 10 years", or "1 in 4" from "2025, 1 in 4". It reads: - - digits, with a sign or currency sign and one multiplier or percentage word: "-£1.7bn", "+434m", "2 million", "4%", "0.1 percentage points"; + - digits, with a sign, a currency sign or both, and one multiplier (k, m, mn, bn, tn, hundred, thousand, million, billion, trillion) or percentage word (%, per cent, percent, percentage points, pp, ppts): "-£1.7bn", "+434m", ".5", "4%"; - number words up to nine hundred and ninety-nine, with a sign and one multiplier or percentage word: "four", "-four", "a hundred and five", "two hundred thousand", "a million", "four per cent"; - - a worded fraction: "half" alone, or a numerator ("a", "an", "another", or one to ten) and a denominator from "half" to "tenth" ("a third", "three-fifths"); - - a worded fraction of a multiplier or percentage: "half a million", "a quarter of a percentage point"; - - a whole number and a fraction, with a multiplier or percentage word if printed: "two and a half", "one and a half million"; - - a ratio of digits or number words with one multiplier or none, joined all by spaces or all by hyphens: "1 in 4", "1-in-4", "3 out of 10", "one out of every five", "1 in 2 million"; + - a worded fraction: "half" alone, or a numerator ("a", "an", "another", or one to ten) and a denominator from "half" to "tenth" ("a third", "three-fifths"), with a sign if printed ("-three-fifths"); + - a worded fraction of a multiplier or percentage: "half a million", "a quarter of a percentage point", "half per cent"; + - a whole number, in digits or in words up to nine hundred and ninety-nine, and a fraction, with a multiplier or percentage word if printed: "two and a half", "one and a half million", "one hundred and a half"; + - a ratio of digits or number words with one multiplier or none, joined all by spaces or all by hyphens: "1 in 4", "1-in-4", "3 out of 10", "one out of every five", "one in every ten", "1 in 2 million"; - a range: two figures joined by a dash, with no spaces or a space on each side ("£4-5 million", "4 – 5%"), or two numbers joined by "to" ("four to five", "4bn to 5bn"); - - "nil", "none", "nothing", "no change", "no impact" or "no overall impact", as zero. -- It also reads a qualifier before the number ("about", "up to" and the others in `QUALIFIERS`), "a" before digits ("less than a 0.1 percentage point"), and "a year" or "of GDP" after the number. It reads a ratio or worded fraction with its percentage in brackets, in either order, if the two agree to the nearest percentage point: "one third (33%)", "25% (1 in 4)". It matches `value` to either number there or in a range. It reports a range that joins a percentage to a multiplier ("4% to 2 million"), or whose first number exceeds its second in both value and size ("2025-26", "5 to 4"). -- It takes a number's scale from its own word (bn, million, %), and gives a range's first number the second's word. For digits with no word, it takes the factor in `normalization`, failing that 1, 10³, 10⁶, 10⁹, 0.01 or 100, and for number words with no word, that factor or 1. It reads a percentage as a share or in percent units, and requires `value` to match within four units in the last place. -- It reads a ratio or a worded fraction on its own as a share, and accepts a `value` within 0.1% of that share, or of 100 times it on a `unit_concept: percent` row ("1 in 4" as 25). It won't match `value` to either of a ratio's numbers alone, and it reads "-1 in 4" and "1 in -4" as −0.25. It also allows 0.1% on a whole number with a fraction, and on a fraction of a multiplier or percentage. + - "nil", "none", "nothing", "no change", "no impact" or "no overall impact", as zero, and a dash where we say in `normalization` that the producer prints it for nil ("dash for nil"). +- It also reads a qualifier before the number ("about", "up to" and the others in `QUALIFIERS`), "a" before digits ("less than a 0.1 percentage point"), and "a year" or "of GDP" after the number. It reads a ratio or worded fraction with its percentage in brackets, in either order, if the two agree to the nearest percentage point: "one third (33%)", "25% (1 in 4)". It matches `value` to either number there or in a range. It reports a range that joins a percentage to a multiplier ("4% to 2 million") or a fraction to a percentage ("a third to 40%"). It also reports a range whose first number exceeds its second, unless neither is positive: it reports "2025-26" and "2 to -3", and reads "-4 to -5". After a bare "in", it reports a year from 2010 to 2099 ("5 in 2027"). +- It takes a number's scale from its own word (bn, million, %), and gives a range's first number the second's word. For a number with no word of its own, it uses the factor in `normalization`, failing that 1. It reads a percentage as a share or in percent units, and requires `value` to match within four units in the last place. +- It reads a ratio or a worded fraction on its own as a share, and accepts a `value` within 0.1% of that share. On a `unit_concept: percent` row, it requires 100 times the share instead ("1 in 4" as 25). It won't match `value` to either of a ratio's numbers alone, and it reads "-1 in 4" and "1 in -4" as −0.25. It also allows 0.1% on a whole number with a fraction, and on a fraction of a multiplier or percentage. - It requires `value` to share `value_raw`'s sign unless we note in `normalization` that we took the sign from the wording. -- It treats an HM Treasury Impact on households row as a bar reading (rule 4) if the row has `value_kind: approx_chart_reading`. It does the same if we call the number a bar, chart or graph reading in the locator, or say there that we read it off or estimated it from one, in any spacing and with or without hyphens. It ignores such a phrase right after "no", "not", "never", "without" or "n't" ("not a bar reading"). -- It also treats the row as a bar reading if we name Figure 1.A or 1.B in the locator ("Fig. 1A", "Figures 1.A and 1.B"), unless we say in its first reference where HM Treasury prints the number. It reads that reference up to the first comma or semicolon. There we name a paragraph, table or box by number ("paragraph 1.7", "paragraphs 1.7–1.8", "Tables 2.B and 2.C", "Table 2.C and Figure 1.A"), or a data label ("Figure 1.A data label", "data label for decile 3"). It treats any other wording as a bar reading, including "Figure 1.A data label unavailable" and "Figure 1.A, decile 3 bar; see paragraph 1.7". +- It treats an HM Treasury Impact on households row as a bar reading (rule 4) if the row has `value_kind: approx_chart_reading`. It does the same if we mention a bar, chart or graph in the locator, or say there that we took a number from a figure, in any spacing and with or without hyphens. It ignores such a phrase after a single "no", "not", "never", "without" or "n't" ("not a bar reading", "did not read it off the chart"), and rejects it after a double negation ("not without a bar reading"). +- When we cite Figure 1.A or 1.B on those rows ("Fig. 1A", "Figures 1.A and 1.B"), it reads the locator up to the first comma or semicolon. Before that point we must name a paragraph, table or box by number, or a data label. It accepts "paragraph 1.7", "paragraphs 1.7–1.8", "Tables 2.B and 2.C", "Table 2.C and Figure 1.A", "Figure 1.A data label" and "data label for decile 3". It treats any other wording as a bar reading, including "Figure 1.A data label unavailable" and "Figure 1.A, decile 3 bar; see paragraph 1.7". - It reports a malformed row as a problem, never as an exception: a row that isn't an object, lacks a field, or has a field of the wrong type. It requires: - a string in `source`, `source_model`, `value_raw`, `normalization`, `quote`, `status`, `value_kind`, `time_basis`, `attribution`, `benchmark_class` and `parse_confidence`; - a string or null in the other text fields, such as `measure_key`, `reform_hint` and `source_table`, and in each `publication` field; diff --git a/tests/test_ab2026_harvest_contract.py b/tests/test_ab2026_harvest_contract.py index 7cc17ac97..481cf46e6 100644 --- a/tests/test_ab2026_harvest_contract.py +++ b/tests/test_ab2026_harvest_contract.py @@ -245,13 +245,65 @@ def test_a_worded_fraction_is_its_share(): assert any("opposite sign" in p for p in problems(_share("a third", -0.3333, note))) -@pytest.mark.parametrize("raw", ["5 in 2026-27", "5 in 2026–27", "5 in 2026-2027"]) -def test_a_fiscal_year_after_in_is_a_problem(raw): - """The checker gives years no special case: it can't read "5 in 2026-27", - and we narrow value_raw to the number the row stores.""" +@pytest.mark.parametrize( + "raw", + [ + "5 in 2027", + "40,000 in 2030", + "two million in 2030", + "1-in-2030", + "5 in -2027", + "5 in −2027", + "5 in +2027", + "5 in 2010", + "5 in 2099", + "5 in 2026-27", + "5 in 2026–27", + "5 in 2026-2027", + ], +) +def test_a_year_after_in_is_a_problem(raw): + """The checker reads no ratio with a year from 2010 to 2099 after a bare + "in", and can't read "5 in 2026-27": we narrow value_raw to the number the + row stores.""" assert _cant_read(raw) +@pytest.mark.parametrize( + ("raw", "quotient"), + [ + ("5 in 2027", 5 / 2027), + ("40,000 in 2030", 40_000 / 2030), + ("two million in 2030", 2e6 / 2030), + ("1-in-2030", 1 / 2030), + ], +) +def test_a_year_gives_no_quotient(raw, quotient): + """A year after "in" counts no denominator, so its quotient is no value.""" + found = problems(_share(raw, quotient, "as printed")) + assert any("can't read" in p for p in found), (raw, found) + + +@pytest.mark.parametrize( + ("raw", "share"), + [ + ("1 in 2000", 1 / 2000), + ("1 in 2009", 1 / 2009), + ("1 in 2100", 1 / 2100), + ("1 in 2,030", 1 / 2030), + ("one in every 2027", 1 / 2027), + ("1 out of 2027", 1 / 2027), + ("1 in 2030 million", 1 / 2.03e9), + ], +) +def test_a_denominator_no_year_could_print_stays_a_ratio(raw, share): + """A number outside 2010 to 2099, a comma, a multiplier, "every" or "out of" + marks a denominator: the share passes, and the numbers alone fail.""" + assert problems(_share(raw, share, "as printed")) == [] + found = problems(_share(raw, 1, "as printed")) + assert any("at a stated scale" in p for p in found), (raw, found) + + @pytest.mark.parametrize( "normalization", [ @@ -602,6 +654,14 @@ def test_a_minus_on_either_number_signs_the_share(raw, share): # a range with a word on its first number alone "4k to 5000", "1m-2000000", + # a range that runs downward from a positive number (cloud review A-L2) + "2 to -3", + "4--5", + "4 - -5", + # a range from a fraction to a percentage, which could read either way + "a third to 40%", + "half to 40%", + "a third to half a per cent", # a range that joins a percentage to a multiplier "4% to 2 million", "2 million to 4%", @@ -624,6 +684,10 @@ def test_a_minus_on_either_number_signs_the_share(raw, share): "1 in 4 (25-26%)", "25-26% (1 in 4)", "3 (1 in 2 million)", + "1 in 3 (33.9%)", + # a hyphen-joined ratio takes no minus on M, and a slash joins nothing + "1-in--4", + "4/5", ] @@ -808,7 +872,9 @@ def test_every_worded_fraction_reads_as_its_share(): ("a third", 0.3331, 0.331), ("two-thirds", 0.667, 0.663), ("1 in 3", 0.3331, 0.331), + ("a third", 0.3334, 0.3337), ("half a million", 500_400, 503_000), + ("half a million", 500_400, 500_600), ("a third of a million", 333_100, 331_000), ("two and a half", 2.502, 2.51), ], @@ -821,6 +887,36 @@ def test_a_rounded_number_passes_within_a_tenth_of_a_percent(raw, near, far): assert any("at a stated scale" in p for p in found), (raw, far, found) +@pytest.mark.parametrize("minus", ["", "-"]) +@pytest.mark.parametrize( + ("raw", "value"), + [ + ("half", 0.5), + ("a third", 1 / 3), + ("1 in 3", 1 / 3), + ("half a million", 5e5), + ("a third of a million", 1e6 / 3), + ("half a per cent", 0.005), + ("half a per cent", 0.5), + ("two and a half", 2.5), + ("two and a half million", 2.5e6), + ("two and a half per cent", 0.025), + ("two and a half per cent", 2.5), + ], +) +def test_a_rounded_number_stops_at_a_tenth_of_a_percent(raw, value, minus): + """Each kind of number that may round, with either sign, passes 0.09% from + its value and fails 0.11% from it, on either side.""" + sign = -1 if minus else 1 + for off in (0.0009, -0.0009): + stored = sign * value * (1 + off) + assert problems(_share(minus + raw, stored, "as printed")) == [], (raw, off) + for off in (0.0011, -0.0011): + stored = sign * value * (1 + off) + found = problems(_share(minus + raw, stored, "as printed")) + assert any("at a stated scale" in p for p in found), (raw, off, found) + + @pytest.mark.parametrize( "raw", [ @@ -877,6 +973,31 @@ def test_a_percent_row_scales_only_a_share(): assert any("at a stated scale" in p for p in problems(row)) +@pytest.mark.parametrize( + "raw", + [ + "half per cent", + "one half percent", + "half a per cent", + "half of one per cent", + "a half percentage point", + ], +) +def test_a_fraction_of_a_percentage_keeps_its_value_on_a_percent_row(raw): + """A fraction of a percentage prints its own unit, so "half per cent" stays + 0.5 on a `unit_concept: percent` row, where "half" alone would read as 50.""" + row = mutated( + value_raw=raw, + quote=raw, + value=0.5, + normalization="as printed", + unit_concept="percent", + ) + assert problems(row) == [] + row["value"] = 50 + assert any("at a stated scale" in p for p in problems(row)) + + @pytest.mark.parametrize( ("raw", "value", "wrongs"), [ @@ -976,6 +1097,29 @@ def test_a_categorical_row_that_prints_a_digit_is_a_problem(raw): ("three quarters of a billion", (7.5e8,), (0.75, 1e9)), ("half a percentage point", (0.005, 0.5), (1, 2)), ("a third of a million", (333_333,), (1 / 3, 1e6)), + ("half of one million", (5e5,), (0.5, 1e6)), + ("half million", (5e5,), (0.5, 1e6)), + ("a half million", (5e5,), (0.5, 1e6)), + # a fraction of a percentage, with or without "a" or "of" + ("half per cent", (0.005, 0.5), (0.05, 5, 50)), + ("half percent", (0.005, 0.5), (0.05, 5, 50)), + ("half%", (0.005, 0.5), (0.05, 5, 50)), + ("one half percent", (0.005, 0.5), (0.05, 5, 50)), + ("a half percentage point", (0.005, 0.5), (0.05, 5, 50)), + ("half a per cent", (0.005, 0.5), (0.05, 5, 50)), + ("half a percent", (0.005, 0.5), (0.05, 5, 50)), + ("half of a per cent", (0.005, 0.5), (0.05, 5, 50)), + ("half of one per cent", (0.005, 0.5), (0.05, 5, 50)), + ("a quarter of a percentage point", (0.0025, 0.25), (1, 0.025)), + ("three-quarters per cent", (0.0075, 0.75), (75, 0.075)), + ("-half per cent", (-0.005, -0.5), (0.005, 0.5)), + # a whole number up to the hundreds and a fraction + ("one hundred and a half", (100.5,), (0.5, 100, 1.5, 150)), + ("a hundred and five and a half", (105.5,), (0.5, 105, 5.5)), + ("nine hundred and ninety-nine and a half", (999.5,), (0.5, 990)), + ("two hundred and a half million", (200.5e6,), (200.5, 0.5, 2e8)), + ("one hundred and a half per cent", (1.005, 100.5), (0.5, 100)), + ("-one hundred and a half", (-100.5,), (100.5,)), # a range: the first number takes the second's word ("£4-5 million", (4e6, 5e6), (4, 5, 4e3)), ("4-5%", (0.04, 0.05, 4, 5), (40,)), @@ -1030,7 +1174,8 @@ def test_a_categorical_row_that_prints_a_digit_is_a_problem(raw): ("-£1.7bn", (-1.7e9,), (1.7e9,)), ("£-1.7bn", (-1.7e9,), (1.7e9,)), ("+$2m", (2e6,), (-2e6,)), - ("£5", (5,), (6,)), + ("£5", (5,), (6, 5e3, 5e6, 5e9, 0.05, 500)), + ("4", (4,), (0.04, 400, 4e3, 4e6, 4e9)), ("$5", (5,), (6,)), ("€5", (5,), (6,)), ("-$5", (-5,), (5,)), @@ -1044,8 +1189,8 @@ def test_a_categorical_row_that_prints_a_digit_is_a_problem(raw): ], ) def test_each_form_reads_at_its_values_and_no_others(raw, passes, fails): - """The README's closed list of forms. Each string passes at every value - it prints and at no other.""" + """Each listed string passes at the listed accepted values and rejects the + listed invalid values.""" for value in passes: assert problems(_share(raw, value, "as printed")) == [], (raw, value) for value in fails: @@ -1057,9 +1202,28 @@ def test_each_form_reads_at_its_values_and_no_others(raw, passes, fails): ) +@pytest.mark.parametrize("dash", ["-", "−", "–", "—", "‐", "‑", " - "]) +def test_a_dash_is_zero_only_where_the_normalization_says_it_prints_nil(dash): + """A table may print a dash for nil or for "not available", so the checker + reads one only if the normalization records nil, and then only as zero.""" + for note in ("dash for nil", "the table prints NIL as a dash"): + assert problems(_share(dash, 0, note)) == [], (dash, note) + for value in (1, -1, 0.5, 1e6): + found = problems(_share(dash, value, note)) + assert any("at a stated scale" in p for p in found), (dash, value, found) + for note in ("as printed", "dash for not available", "vanilla"): + assert _cant_read(dash, note), (dash, note) + + def test_every_number_up_to_999_reads_in_every_style(): - """Exhaustively: each number from 0 to 999, however _say prints it, passes - at itself and fails at itself plus one, alone and as a ratio's N and M.""" + """Exhaustively: each number from 0 to 999, however `_say` prints it, passes + alone and rejects itself plus one. Its ratios pass at their shares and reject a + different share.""" + + def rejects(raw, value): + found = problems(_share(raw, value, "as printed")) + return any("at a stated scale" in p for p in found) + for n in range(1000): for raw in {_say(n, *style) for style in product([False, True], repeat=3)}: assert problems(_share(raw, n, "printed in words")) == [], raw @@ -1067,13 +1231,16 @@ def test_every_number_up_to_999_reads_in_every_style(): assert any("at a stated scale" in p for p in found), raw ratio = f"{raw} in a thousand" assert problems(_share(ratio, n / 1000, "as printed")) == [], raw + assert rejects(ratio, 1.01 * (n or 1) / 1000), raw if n: ratio = f"one in {raw}" assert problems(_share(ratio, 1 / n, "as printed")) == [], raw + assert rejects(ratio, 1.01 / n), raw -# Numbers in every form the README lists, each with its value and with no sign, -# written out here apart from the checker's tables. +# Numbers in the forms the README lists, each with its value and with no sign, +# written out here apart from the checker's tables. Ranges, brackets and the words +# for zero have tests of their own. READABLE = [ ("5", 5), ("1,234", 1234), @@ -1095,7 +1262,9 @@ def test_every_number_up_to_999_reads_in_every_style(): ("a third", 1 / 3), ("three-fifths", 0.6), ("half a million", 5e5), + ("half per cent", 0.005), ("two and a half", 2.5), + ("one hundred and a half", 100.5), ("1 in 4", 0.25), ("one-in-eight", 0.125), ("3 out of 10", 0.3), @@ -1133,9 +1302,9 @@ def test_every_number_up_to_999_reads_in_every_style(): def test_a_readable_number_with_more_around_it_is_a_problem_at_any_value( number, before, after, value ): - """Any number in a listed form, with words or characters around it that - the README doesn't list, gives a problem at any value, so the checker - never reads part of a string.""" + """Each of READABLE's numbers, with words or characters around it that the + README doesn't list, gives a problem at any value, so the checker never reads + part of a string.""" if before or after: found = problems(_share(f"{before}{number}{after}", value, "as printed")) assert any("can't read" in p for p in found), (before, number, after, found) @@ -1144,7 +1313,7 @@ def test_a_readable_number_with_more_around_it_is_a_problem_at_any_value( @settings(max_examples=300, deadline=None) @given(number=SPOILED, value=st.floats(-1e12, 1e12, allow_nan=False)) def test_a_word_inside_a_readable_number_is_a_problem_at_any_value(number, value): - """A word at any space inside a number in a listed form gives a problem.""" + """A word at any space inside one of READABLE's numbers gives a problem.""" for i, ch in enumerate(number): if ch == " ": spoiled = f"{number[:i]} banana{number[i:]}" @@ -1214,8 +1383,7 @@ def test_a_minus_flips_a_number_in_any_form(pair, minus): def test_a_number_s_own_multiplier_sets_its_scale_under_any_normalization( n, word, other, normalization ): - """A multiplier scales its own number, and no normalization or fixed scale - overrides it.""" + """A multiplier scales its own number, and no normalization overrides it.""" text, factor = word raw = f"{n}{text}" assert problems(_share(raw, n * factor, normalization)) == [], raw @@ -1224,18 +1392,41 @@ def test_a_number_s_own_multiplier_sets_its_scale_under_any_normalization( assert any("at a stated scale" in p for p in found), (raw, other, found) +@pytest.mark.parametrize("normalization", ["as printed", "£ x 1,000,000", "/ 100"]) +@pytest.mark.parametrize( + ("raw", "value", "number"), + [ + ("one million", 1e6, 1), + ("two million", 2e6, 2), + ("a thousand", 1e3, 1), + ("three hundred thousand", 3e5, 300), + ("five billion", 5e9, 5), + ("half a million", 5e5, 0.5), + ("one and a half million", 1.5e6, 1.5), + ], +) +def test_a_worded_multiplier_sets_its_scale_under_any_normalization( + raw, value, number, normalization +): + """A multiplier in words scales its number as a short one does, so "one + million" under "/ 100" passes at a million and fails at 0.01.""" + assert problems(_share(raw, value, normalization)) == [], raw + for scale in (1, 0.01, 1e6): + if number * scale != value: + found = problems(_share(raw, number * scale, normalization)) + assert any("at a stated scale" in p for p in found), (raw, scale, found) + + @settings(max_examples=200, deadline=None) @given(n=st.integers(1, 9999)) -def test_digits_alone_take_the_normalization_s_factor_failing_that_a_fixed_set(n): - """Digits with no word of their own take the normalization's factor and - nothing else. Failing that, they take 1, a thousand, a million, a billion, - 0.01 or 100.""" +def test_digits_alone_take_the_normalization_s_factor_failing_that_one(n): + """Digits with no word of their own take the normalization's factor, failing + that 1, and nothing else: "4" as printed never passes at 4,000,000,000.""" raw = str(n) - for scale in (1, 1e3, 1e6, 1e9, 0.01, 100): - assert problems(_share(raw, n * scale, "as printed")) == [], (raw, scale) - assert any( - "at a stated scale" in p for p in problems(_share(raw, n * 7, "as printed")) - ) + assert problems(_share(raw, n, "as printed")) == [] + for scale in (7, 1e3, 1e6, 1e9, 0.01, 100): + found = problems(_share(raw, n * scale, "as printed")) + assert any("at a stated scale" in p for p in found), (raw, scale, found) stated = "£ million x 1,000,000" assert problems(_share(raw, n * 1e6, stated)) == [] for scale in (1, 1e3, 1e9, 0.01, 100): @@ -1250,8 +1441,8 @@ def test_digits_alone_take_the_normalization_s_factor_failing_that_a_fixed_set(n def test_words_alone_take_the_normalization_s_factor_failing_that_one( raw, value, wrong ): - """Number words with no word of their own take the normalization's factor, - failing that 1, and never the fixed set digits fall back on.""" + """Number words with no multiplier take the normalization's factor, failing + that 1, as digits do.""" assert problems(_share(raw, value, "as printed")) == [] for bad in (wrong, value * 1e3, value * 1e6, value * 0.01, value * 100): found = problems(_share(raw, bad, "as printed")) @@ -1434,6 +1625,116 @@ def test_rule_4_holds_whatever_the_title_case(): ("Figure 1.A: data label for decile 3", False), ("Figure 1.A data label decile 3", False), ("Figure 1.A, data label decile 3", True), + # Sol N2: a double negation affirms, so one negation alone cancels + ("paragraph 1.7; not without a bar reading from Figure 1.A", True), + ("paragraph 1.7; never without reading off the chart", True), + ("paragraph 1.7; isn't not a chart reading", True), + ("paragraph 1.7; no never a graph", True), + ("paragraph 1.7; without a bar reading from Figure 1.A", False), + ("paragraph 1.7; never reading off the chart", False), + ("paragraph 1.7; no, not a bar reading", False), + # Sol N3 and cloud review M2: how we say we took a number off a picture, + # and the same with a negation before the verb + ("paragraph 1.7; read off of the chart in Figure 1.A", True), + ("paragraph 1.7; not read off of the chart in Figure 1.A", False), + ("paragraph 1.7; read directly off the chart", True), + ("paragraph 1.7; not read directly off the chart", False), + ("paragraph 1.7; read off this chart (Figure 1.A)", True), + ("paragraph 1.7; never read off this chart (Figure 1.A)", False), + ("paragraph 1.7; read from that bar", True), + ("paragraph 1.7; not read from that bar", False), + ("paragraph 1.7; we read it off the bar in Figure 1.A", True), + ("paragraph 1.7; we did not read it off the bar", False), + ("paragraph 1.7; we estimated it from the bar", True), + ("paragraph 1.7; we haven't estimated it from the bar", False), + ("paragraph 1.7; read off the decile 3 bar of Figure 1.A", True), + ("paragraph 1.7; not read off the decile 3 bar of Figure 1.A", False), + ("paragraph 1.7; reading of the decile 3 bar, Figure 1.A", True), + ("paragraph 1.7; not a reading of the decile 3 bar, Figure 1.A", False), + ("paragraph 1.7; taken from the chart in Figure 1.A", True), + ("paragraph 1.7; not taken from the chart, Figure 1.A", False), + ("paragraph 1.7; took it from the chart", True), + ("paragraph 1.7; measured from the bar", True), + ("paragraph 1.7; not measured from the bar", False), + ("decile 3: we read it off the chart", True), + ("paragraph 1.7; reads from the chart", True), + ("paragraph 1.7; never reads from the chart", False), + ("paragraph 1.7; reads off the chart", True), + ("paragraph 1.7; readings from the graph", True), + ("paragraph 1.7; read the number off the bar", True), + ("paragraph 1.7; estimated from Figure 1.A", True), + ("paragraph 1.7; isn't estimated from Figure 1.A", False), + ("paragraph 1.7; read off of Figure 1.A", True), + ("paragraph 1.7; isn't read off of Figure 1.A", False), + ("paragraph 1.7; not a bar reading from the figure", False), + ("paragraph 1.7; not a bar reading from the figures", False), + ("paragraph 1.7; not a bar reading off the figure", False), + ("paragraph 1.7; not a bar reading off of the figure", False), + ("paragraph 1.7; reading of Figure 1.A", True), + ("paragraph 1.7; not a reading of Figure 1.A", False), + ("paragraph 1.7; no readings of the chart", False), + ("paragraph 1.7; no readings from the graph", False), + ("paragraph 1.7; not an estimate from the chart", False), + ("paragraph 1.7; never took it from the chart", False), + ("paragraph 1.7; not read the decile 3 off the bar", False), + ("paragraph 1.7; not read off the top decile 3 bar", False), + ("paragraph 1.7; off of Figure 1.A", True), + # any other bar, chart or graph, and any number from a figure, whatever + # the verb; a negation before a verb we don't list cancels nothing + ("paragraph 1.7; decile 3 bar", True), + ("Table 2.C; the chart shows it", True), + ("paragraph 1.7; graph", True), + ("paragraph 1.7; eyeballed from Figure 1.A", True), + ("paragraph 1.7; interpolated from the chart", True), + ("paragraph 1.7; not interpolated from the chart", True), + ("paragraph 1.7; from Figure 1.A", True), + ("paragraph 1.7; off Figure 1.A", True), + ("paragraph 1.7; no bar", False), + ("paragraph 1.7; without a chart", False), + ("paragraph 1.7; not the graph", False), + ("paragraph 1.7; no bars", False), + # words that only look like one, and readings of something else + ("paragraph 1.7; Barnett formula", False), + ("paragraph 1.7; Charter for Budget Responsibility", False), + ("Table 2.C; read with paragraph 1.7", False), + ("paragraph 1.7; taken from Table 2.C", False), + ("paragraph 1.7; read from the text", False), + ("paragraph 1.7; discussion of Figure 1.A", False), + # Sol N6: any dash in a printed place, as in the figure's name + ("Figure 1-A data label", False), + ("Figure 1–A data label", False), + ("Figure 1—B data label", False), + ("Figure 1‑A data label", False), + ("Figure 1‐A data label", False), + ("Figures 1-A and 1-B data labels", False), + ("Figure 1.A data-label", False), + ("paragraphs 1.7—1.8; Figure 1.A", False), + ("paragraphs 1.7−1.8; Figure 1.A", False), + ("paragraphs 1.7 — 1.8; Figure 1.A", False), + ("Figure 1-A data label unavailable", True), + # cloud review A-L1: each arm of the negation and the printed forms + ("paragraph 1.7; graph reading, Figure 1.A", True), + ("paragraph 1.7; without a graph reading, Figure 1.A", False), + ("paragraph 1.7; it has not been read off the chart, Figure 1.A", False), + ("paragraph 1.7; isn’t read off the chart, Figure 1.A", False), + ("para 1.7; Figure 1.A", False), + ("para. 1.7; Figure 1.A", False), + ("Boxes 1.1 and 1.2; Figure 1.A", False), + ("Figure 1.B data label on quintile 2", False), + # each dash, in a phrase and in a figure's name + ("paragraph 1.7; not a bar–reading from Figure 1.A", False), + ("Figure 1–B, decile 3", True), + ("Figure 1‐A, decile 3", True), + ("Figure 1‑A, decile 3", True), + # plural figures, each kind of picture, and any spacing + ("paragraph 1.7; not read off figures 1.A and 1.B", False), + ("paragraph 1.7; from figures 1.A and 1.B", True), + ("paragraph 1.7; graphs", True), + ("paragraph 1.7; charts", True), + ("paragraph 1.7; not a bar reading from Figure 1.A", False), + # any phrase that stands decides, and a verb starts at a word's start + ("paragraph 1.7; a bar reading but not a chart reading", True), + ("paragraph 1.7; spread from no chart", False), ], ) def test_rule_4_reads_where_the_locator_places_the_number(locator, bar_reading): From 3c65fd7a091a4d00352239e40c6880529822890f Mon Sep 17 00:00:00 2001 From: Max Ghenis Date: Thu, 5 Nov 2026 00:49:46 -0500 Subject: [PATCH 48/52] Cut one example from the README's ratio form The list ran to 42 words, and "one out of every five" already shows "out of". Co-Authored-By: Claude Opus 5.5 --- sources/harvest-uk-ab2026-2026-10-29/README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sources/harvest-uk-ab2026-2026-10-29/README.md b/sources/harvest-uk-ab2026-2026-10-29/README.md index 8f4b5077d..4c6dd8f3c 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/README.md +++ b/sources/harvest-uk-ab2026-2026-10-29/README.md @@ -68,7 +68,7 @@ We follow the AB2025 harvest README's rules 1–10 and its row contract, with th - a worded fraction: "half" alone, or a numerator ("a", "an", "another", or one to ten) and a denominator from "half" to "tenth" ("a third", "three-fifths"), with a sign if printed ("-three-fifths"); - a worded fraction of a multiplier or percentage: "half a million", "a quarter of a percentage point", "half per cent"; - a whole number, in digits or in words up to nine hundred and ninety-nine, and a fraction, with a multiplier or percentage word if printed: "two and a half", "one and a half million", "one hundred and a half"; - - a ratio of digits or number words with one multiplier or none, joined all by spaces or all by hyphens: "1 in 4", "1-in-4", "3 out of 10", "one out of every five", "one in every ten", "1 in 2 million"; + - a ratio of digits or number words with one multiplier or none, joined all by spaces or all by hyphens: "1 in 4", "1-in-4", "one out of every five", "one in every ten", "1 in 2 million"; - a range: two figures joined by a dash, with no spaces or a space on each side ("£4-5 million", "4 – 5%"), or two numbers joined by "to" ("four to five", "4bn to 5bn"); - "nil", "none", "nothing", "no change", "no impact" or "no overall impact", as zero, and a dash where we say in `normalization` that the producer prints it for nil ("dash for nil"). - It also reads a qualifier before the number ("about", "up to" and the others in `QUALIFIERS`), "a" before digits ("less than a 0.1 percentage point"), and "a year" or "of GDP" after the number. It reads a ratio or worded fraction with its percentage in brackets, in either order, if the two agree to the nearest percentage point: "one third (33%)", "25% (1 in 4)". It matches `value` to either number there or in a range. It reports a range that joins a percentage to a multiplier ("4% to 2 million") or a fraction to a percentage ("a third to 40%"). It also reports a range whose first number exceeds its second, unless neither is positive: it reports "2025-26" and "2 to -3", and reads "-4 to -5". After a bare "in", it reports a year from 2010 to 2099 ("5 in 2027"). From 820688216771d5c7fd3aa4c86cc212f751968e2e Mon Sep 17 00:00:00 2001 From: Max Ghenis Date: Thu, 5 Nov 2026 01:46:40 -0500 Subject: [PATCH 49/52] Read each factor whole, close the nil note and widen rule 4 GPT-6.1 Sol and a cloud session each reviewed 3c65fd7a and requested changes, and this commit answers both. - `_factors` reads each operand whole, multiplies operations in a row, reads a factor in brackets, and returns None for one it can't read, so the checker reports that row at any value (R2). - A whole number with a fraction takes normalization's factor and reads a short multiplier (A3, R4). - The checker reads a dash as zero only under one whole clause that declares it for nil, and it reads the figure dash and the horizontal bar (A4, R3). - Rule 4 reads "Fig." and "Figs.", "on", "in", "using" and "of", plots, diagrams, visuals, a figure's column, readings by eye and "Figure 1(a)". It also counts a second negation anywhere in a clause (A1, A2, R1, R8). - The tests add each case (A5, R7). - The README's contract bullets state what the checker now does (B1 to B4, R5, R6, R9). We change no staged row, and the checker finds 0 problems in the 12 families. Co-Authored-By: Claude Opus 5.5 --- pipeline/ab2026_harvest_contract.py | 111 ++++++--- .../harvest-uk-ab2026-2026-10-29/README.md | 12 +- tests/test_ab2026_harvest_contract.py | 234 +++++++++++++++++- 3 files changed, 307 insertions(+), 50 deletions(-) diff --git a/pipeline/ab2026_harvest_contract.py b/pipeline/ab2026_harvest_contract.py index 62f119bb9..aae9b2c32 100644 --- a/pipeline/ab2026_harvest_contract.py +++ b/pipeline/ab2026_harvest_contract.py @@ -207,10 +207,15 @@ TAILS = {"a year", "of gdp"} # Words that print a zero ZERO = {"nil", "no change", "no impact", "no overall impact", "none", "nothing"} -# A dash prints a zero only if the normalization says the producer prints it for -# nil ("dash for nil"), since a table may also print one for "not available". -DASHES = set("-−–—‐‑") -NIL = re.compile(r"\bnil\b", re.I) +# A dash prints a zero only if one whole clause of the normalization says the +# producer prints it for nil ("dash for nil", "the table prints nil as a dash"), +# since a table may also print one for "not available". +_DASHES = "-−–—‐‑‒―" +DASHES = set(_DASHES) +NIL = re.compile( + r"(?:(?:(?:the )?(?:table|producer|source|workbook) )?prints? )?" + r"(?:nil (?:as|=) (?:a )?dash|(?:a |the )?dash(?:es)? (?:for|means?|=) nil)" +) # After a bare "in", a year gives no denominator: "5 in 2027" YEAR = re.compile(r"20[1-9][0-9]") @@ -259,7 +264,7 @@ def _either(words): # hundred and a half" MIXED = re.compile( rf"(?P{_SIGN})?(?P{_DIGITS}|{_BELOW_1000}) and (?P{_PART})" - rf"(?:[ -](?P{_LARGE})|(?P{_PERCENT}))?" + rf"(?:[ -](?P{_either(SHORT)}|{_LARGE})|(?P{_PERCENT}))?" ) _COUNT = rf"(?:{_DIGITS}{_TIMES}?|{_WORDS})" # Two numbers joined all by spaces or all by hyphens. A sign comes before N or, @@ -415,11 +420,17 @@ def _restates(first, second): ) +def _says_nil(normalization): + """Whether one whole clause of a normalization declares a dash for nil.""" + clauses = re.split(r"[;,.()]", normalization.lower()) + return any(NIL.fullmatch(" ".join(c.split())) for c in clauses) + + def _read(raw, normalization=""): """What value_raw prints, as a list of _one_number's tuples, or None if it matches no form the checker reads.""" text = " ".join(raw.lower().split()) - if text in DASHES and NIL.search(normalization): + if text in DASHES and _says_nil(normalization): return [(0.0, (1,), "words")] body = WHOLE.fullmatch(text)["body"] if body in ZERO: @@ -431,20 +442,33 @@ def _read(raw, normalization=""): return first + second if first and second and _restates(first, second) else None +_STEP = r"[x×*/÷]\s*\.?[0-9][^\s;]*" +_CHAIN = re.compile(rf"(?:^|(?<=[\s(\[]))({_STEP}(?:\s+{_STEP})*)") +_OPERAND = re.compile(rf"10(?:\^|\*\*)(?P[0-9]+)|{_DIGITS}") +# What would carry on a number the chain seems to end: "x 1 000 000", "x 1 million" +_CONTINUES = re.compile(rf"\s+(?:[0-9]|(?:{_either(LARGE)}|{_either(SHORT)})\b)") + + def _factors(normalization): - """The multipliers a normalization states ("x 1,000,000", "/ 100", ...).""" + """The multipliers a normalization states ("x 1,000,000", "/ 100", "(x 10^6)"), + or None if it states one the checker can't read whole. Operations in a row + make one multiplier: "x 1,000,000 / 100" is 10,000.""" + text = re.sub(r"\s*(\^|\*\*)\s*", r"\1", (normalization or "").lower()) out = [] - for op, num in re.findall( - r"(?:^|\s)(x|×|\*|/|÷)\s*(\d[\d,]*(?:\.\d+)?(?:e[+-]?\d+)?)", - normalization or "", - re.I, - ): - try: - f = float(num.replace(",", "")) - except ValueError: - continue - if f and math.isfinite(f): - out.append(1 / f if op in "/÷" else f) + for chain in _CHAIN.finditer(text): + if _CONTINUES.match(text, chain.end()): + return None + factor = 1.0 + for step in re.findall(_STEP, chain[0]): + operand = _OPERAND.fullmatch(step[1:].strip().rstrip(",:.)]")) + if not operand: + return None + power = operand["power"] + f = float(f"1e{power}" if power else operand[0].replace(",", "")) + if not f or not math.isfinite(f): + return None + factor = factor / f if step[0] in "/÷" else factor * f + out.append(factor) return out @@ -509,13 +533,18 @@ def value_problems(r): "narrow it to the number the row stores" ] factors = _factors(r.get("normalization") or "") + if factors is None: + return [ + "the checker can't read a factor in normalization; state one number " + "or power of 10" + ] per = 100 if r.get("unit_concept") == "percent" else 1 exact, near = [], [] for number, own, kind in numbers: if kind == "share": near.append(per * number) elif kind == "rounded": - near += [number * s for s in own or (1,)] + near += [number * s for s in own or factors or (1,)] else: exact += [number * s for s in own or factors or (1,)] if not ( @@ -538,14 +567,14 @@ def value_problems(r): # the forms PRINTED reads and nothing more: a paragraph, table or box by its # number, or a data label. The checker reads any other first reference as a # bar reading, "Figure 1.A data label unavailable" included. Anywhere in a -# locator, a bar, a chart or a graph marks a bar reading, and so does a number -# taken from a figure. A single negation just before the phrase cancels it ("not +# locator, a bar, chart, graph, plot, diagram or visual marks a bar reading, and +# so does a number taken from a figure or read by eye. A single negation just before the phrase cancels it ("not # a bar reading", "didn't read it off the chart"), and a double negation doesn't # ("not without a bar reading"). The checker looks for these phrases with hyphens # and dashes as spaces and with runs of spaces collapsed, and reads the first # reference with each dash as a hyphen. -_DASH = "[-−–—‐‑]" -FIGURE_1 = re.compile(r"\bfig(?:ure)?s?\.?\s*1\.?\s*[ab]\b", re.I) +_DASH = f"[{_DASHES}]" +FIGURE_1 = re.compile(r"\bfig(?:ure)?s?\.?\s*1\.?\s*\(?[ab]\b", re.I) _FIG = r"fig(?:ure)?s?\.?\s*1[.-]?\s*[ab](?:\s+and\s+1[.-]?\s*[ab])?" _ID = r"\d+(?:\.[0-9a-z]+)*" PRINTED = re.compile( @@ -555,24 +584,28 @@ def value_problems(r): rf"(?:\s+(?:(?:for|on)\s+)?(?:decile|quintile)\s+\d+|\s+(?:for|on)\s+{_FIG})?", re.I, ) -_PICTURE = r"(?:bars?|charts?|graphs?)" +_PICTURE = r"(?:bars?|charts?|graphs?|plots?|diagrams?|visuals?)" +_FIGS = r"fig(?:ure)?s?" _OFF = r"(?:off|from)" _UP_TO = r"(?:[\w.]+ ){0,4}?" # A verb's whole phrase comes first, so that a negation before the verb cancels -# it: "read it off the decile 3 bar", "reading of the chart". Then any bar, chart -# or graph, with a reading if it names one ("bar reading from Figure 1.A"), and -# any number from a figure. +# it: "read it off the decile 3 bar", "reading of the chart", "visual estimate of +# Figure 1.A". Then any picture, with a reading if it names one ("bar reading from +# Figure 1.A"), any number from a figure or a figure's column, and any reading by +# eye. A column of a table prints its number, so a column counts only in a figure. BAR_READING = re.compile( - r"\b(?:(?:read|estimat|tak|measur)\w*|took)(?: [\w.]+){0,3}?" - rf" {_OFF} {_UP_TO}(?:{_PICTURE}|figures?)\b" - rf"|\breadings? of {_UP_TO}(?:{_PICTURE}|figures?)\b" - rf"|\b{_PICTURE}(?: read\w*(?: {_OFF} {_UP_TO}figures?)?)?\b" - rf"|\b{_OFF} {_UP_TO}figures?\b", + r"\b(?:visual )?(?:(?:read|estimat|tak|measur)\w*|took)(?: [\w.]+){0,3}?" + rf" (?:off|from|on|in|using|of) {_UP_TO}(?:{_PICTURE}|{_FIGS})\b" + rf"|\b{_PICTURE}(?: read\w*(?: {_OFF} {_UP_TO}{_FIGS})?)?\b" + rf"|\b{_OFF} {_UP_TO}{_FIGS}\b" + rf"|\bcolumns?(?: [\w.]+){{0,2}}? (?:of|in|on) {_UP_TO}{_FIGS}\b" + r"|\b(?:read\w* )?by eye\b", re.I, ) DENIED = re.compile( - r"(?:\b(?:no|not|never|without)|n['’]t)\s+(?:(?:a|an|the|been)\s+)?\Z", re.I + r"(?:\b(?:no|not|never|without|cannot)|n['’]t)\s+(?:(?:a|an|the|been)\s+)?\Z", re.I ) +NEGATION = re.compile(r"\b(?:no|not|never|without|cannot)\b|n['’]t\b", re.I) def _keys(path): @@ -853,16 +886,20 @@ def _row_problems(r, keys, earlier, metrics, units): out.append( "no HMT Impact on households bar readings: start the locator with " "the paragraph, table, box or data label where HM Treasury prints " - "the number, and call nothing there a bar, chart or graph (README " - "rule 4)" + "the number, and call nothing there a bar, chart, graph, plot, diagram " + "or visual (README rule 4)" ) return out def _denied(before): - """Whether one negation, and no more, comes right before a phrase.""" + """Whether a negation comes right before a phrase, with no other earlier in + its clause.""" m = DENIED.search(before) - return bool(m) and not DENIED.search(before[: m.start()]) + if not m: + return False + clause = re.split(r"[,;:()]", before[: m.start()])[-1] + return not NEGATION.search(clause) def reads_a_bar(locator): diff --git a/sources/harvest-uk-ab2026-2026-10-29/README.md b/sources/harvest-uk-ab2026-2026-10-29/README.md index 4c6dd8f3c..181f727c5 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/README.md +++ b/sources/harvest-uk-ab2026-2026-10-29/README.md @@ -63,20 +63,20 @@ We follow the AB2025 harvest README's rules 1–10 and its row contract, with th - The checker accepts `value_kind: categorical` for a rating with no digits, such as the OBR's Table 3.3 uncertainty ratings. - It requires a finite number in `value`. Unless the row has `value_kind: categorical`, it reads `value_raw` only if the whole string, in any case and spacing, matches one of the forms listed here. It reports any other string as a problem, and we narrow `value_raw` to the number, which `quote` still contains: "£5" from "£5 in 10 years", or "1 in 4" from "2025, 1 in 4". It reads: - - digits, with a sign, a currency sign or both, and one multiplier (k, m, mn, bn, tn, hundred, thousand, million, billion, trillion) or percentage word (%, per cent, percent, percentage points, pp, ppts): "-£1.7bn", "+434m", ".5", "4%"; + - digits, with a sign, a currency sign or both, and one multiplier (k, m, mn, bn, tn, hundred, thousand, million, billion, trillion) or percentage word (%, per cent, percent, percentage point or points, pp, ppt, ppts): "-£1.7bn", ".5", "4%"; - number words up to nine hundred and ninety-nine, with a sign and one multiplier or percentage word: "four", "-four", "a hundred and five", "two hundred thousand", "a million", "four per cent"; - a worded fraction: "half" alone, or a numerator ("a", "an", "another", or one to ten) and a denominator from "half" to "tenth" ("a third", "three-fifths"), with a sign if printed ("-three-fifths"); - a worded fraction of a multiplier or percentage: "half a million", "a quarter of a percentage point", "half per cent"; - a whole number, in digits or in words up to nine hundred and ninety-nine, and a fraction, with a multiplier or percentage word if printed: "two and a half", "one and a half million", "one hundred and a half"; - - a ratio of digits or number words with one multiplier or none, joined all by spaces or all by hyphens: "1 in 4", "1-in-4", "one out of every five", "one in every ten", "1 in 2 million"; + - a ratio of two numbers, in digits or number words, with at most one multiplier each: "1 in 4", "1 in 2 million". It takes "in", "in every", "out of" or "out of every" between them, with spaces or hyphens throughout: "3 out of 10", "one-in-every-ten"; - a range: two figures joined by a dash, with no spaces or a space on each side ("£4-5 million", "4 – 5%"), or two numbers joined by "to" ("four to five", "4bn to 5bn"); - - "nil", "none", "nothing", "no change", "no impact" or "no overall impact", as zero, and a dash where we say in `normalization` that the producer prints it for nil ("dash for nil"). + - "nil", "none", "nothing", "no change", "no impact" or "no overall impact", as zero. It also reads a dash as zero where we say, in one whole clause of `normalization`, that the producer prints it for nil ("dash for nil", "the table prints nil as a dash"). - It also reads a qualifier before the number ("about", "up to" and the others in `QUALIFIERS`), "a" before digits ("less than a 0.1 percentage point"), and "a year" or "of GDP" after the number. It reads a ratio or worded fraction with its percentage in brackets, in either order, if the two agree to the nearest percentage point: "one third (33%)", "25% (1 in 4)". It matches `value` to either number there or in a range. It reports a range that joins a percentage to a multiplier ("4% to 2 million") or a fraction to a percentage ("a third to 40%"). It also reports a range whose first number exceeds its second, unless neither is positive: it reports "2025-26" and "2 to -3", and reads "-4 to -5". After a bare "in", it reports a year from 2010 to 2099 ("5 in 2027"). -- It takes a number's scale from its own word (bn, million, %), and gives a range's first number the second's word. For a number with no word of its own, it uses the factor in `normalization`, failing that 1. It reads a percentage as a share or in percent units, and requires `value` to match within four units in the last place. +- It takes each number's scale from its own word (bn, million, %). When only a range's second number has one, it gives that word to the first number too. When digits, number words or a whole number with a fraction have no word of their own, it uses the factor in `normalization`, failing that 1. It reads a factor whole ("x 1,000,000", "/ 100", "(x 10^6)"), multiplies operations in a row ("x 1,000,000 / 100" as 10,000), and takes either factor where a semicolon separates two. It reports a factor it can't read ("x 1 million") at any value. It reads a percentage as a share or in percent units, and requires digits and number words to match `value` within four units in the last place. - It reads a ratio or a worded fraction on its own as a share, and accepts a `value` within 0.1% of that share. On a `unit_concept: percent` row, it requires 100 times the share instead ("1 in 4" as 25). It won't match `value` to either of a ratio's numbers alone, and it reads "-1 in 4" and "1 in -4" as −0.25. It also allows 0.1% on a whole number with a fraction, and on a fraction of a multiplier or percentage. - It requires `value` to share `value_raw`'s sign unless we note in `normalization` that we took the sign from the wording. -- It treats an HM Treasury Impact on households row as a bar reading (rule 4) if the row has `value_kind: approx_chart_reading`. It does the same if we mention a bar, chart or graph in the locator, or say there that we took a number from a figure, in any spacing and with or without hyphens. It ignores such a phrase after a single "no", "not", "never", "without" or "n't" ("not a bar reading", "did not read it off the chart"), and rejects it after a double negation ("not without a bar reading"). -- When we cite Figure 1.A or 1.B on those rows ("Fig. 1A", "Figures 1.A and 1.B"), it reads the locator up to the first comma or semicolon. Before that point we must name a paragraph, table or box by number, or a data label. It accepts "paragraph 1.7", "paragraphs 1.7–1.8", "Tables 2.B and 2.C", "Table 2.C and Figure 1.A", "Figure 1.A data label" and "data label for decile 3". It treats any other wording as a bar reading, including "Figure 1.A data label unavailable" and "Figure 1.A, decile 3 bar; see paragraph 1.7". +- It treats an HM Treasury Impact on households row as a bar reading (rule 4) if the row has `value_kind: approx_chart_reading`. It does the same if we mention a bar, chart, graph, plot, diagram or visual in the locator, in any spacing and with or without hyphens. It also does so if we write there that we read, took, estimated or measured the number off, from, of, on, in or using a figure, or read it by eye. It reads "Fig." and "Figs." as figures, so it counts "read off Fig. 1.A". It counts "column 3 of Figure 1.A" and leaves "Table 2.1, column 3" alone. It ignores such a phrase right after one "no", "not", "never", "without", "cannot" or "n't" ("not a bar reading"), but not after two in the same clause ("not possible without reading off the chart"). +- When we cite Figure 1.A or 1.B on those rows ("Fig. 1A", "Figure 1(a)", "Figures 1.A and 1.B"), it reads the locator up to the first comma or semicolon. Before that point we must name a paragraph, table or box by number, or a data label. It accepts "paragraph 1.7", "paragraphs 1.7–1.8", "Tables 2.B and 2.C", "Table 2.C and Figure 1.A", "Figure 1.A data label" and "data label for decile 3". It treats any other wording as a bar reading, including "Figure 1.A data label unavailable" and "Figure 1.A, decile 3 bar; see paragraph 1.7". - It reports a malformed row as a problem, never as an exception: a row that isn't an object, lacks a field, or has a field of the wrong type. It requires: - a string in `source`, `source_model`, `value_raw`, `normalization`, `quote`, `status`, `value_kind`, `time_basis`, `attribution`, `benchmark_class` and `parse_confidence`; - a string or null in the other text fields, such as `measure_key`, `reform_hint` and `source_table`, and in each `publication` field; diff --git a/tests/test_ab2026_harvest_contract.py b/tests/test_ab2026_harvest_contract.py index 481cf46e6..c2d93136e 100644 --- a/tests/test_ab2026_harvest_contract.py +++ b/tests/test_ab2026_harvest_contract.py @@ -300,8 +300,9 @@ def test_a_denominator_no_year_could_print_stays_a_ratio(raw, share): """A number outside 2010 to 2099, a comma, a multiplier, "every" or "out of" marks a denominator: the share passes, and the numbers alone fail.""" assert problems(_share(raw, share, "as printed")) == [] - found = problems(_share(raw, 1, "as printed")) - assert any("at a stated scale" in p for p in found), (raw, found) + for alone in (1, 1 / share): + found = problems(_share(raw, alone, "as printed")) + assert any("at a stated scale" in p for p in found), (raw, alone, found) @pytest.mark.parametrize( @@ -906,12 +907,12 @@ def test_a_rounded_number_passes_within_a_tenth_of_a_percent(raw, near, far): ) def test_a_rounded_number_stops_at_a_tenth_of_a_percent(raw, value, minus): """Each kind of number that may round, with either sign, passes 0.09% from - its value and fails 0.11% from it, on either side.""" + its value and fails 0.105% from it, on either side.""" sign = -1 if minus else 1 for off in (0.0009, -0.0009): stored = sign * value * (1 + off) assert problems(_share(minus + raw, stored, "as printed")) == [], (raw, off) - for off in (0.0011, -0.0011): + for off in (0.00105, -0.00105): stored = sign * value * (1 + off) found = problems(_share(minus + raw, stored, "as printed")) assert any("at a stated scale" in p for p in found), (raw, off, found) @@ -1088,6 +1089,8 @@ def test_a_categorical_row_that_prints_a_digit_is_a_problem(raw): ("2 and a third", (7 / 3, 2.3333), (1 / 3, 2)), ("-two and a half", (-2.5,), (2.5,)), ("one and a half million", (1.5e6,), (0.5, 1.5, 5e5)), + ("2 and a half bn", (2.5e9,), (2.5, 2.5e6)), + ("two and a half m", (2.5e6,), (2.5, 2.5e9)), ("two and a half per cent", (0.025, 2.5), (0.5, 2)), ("twenty-one and a half", (21.5,), (0.5, 1)), # a fraction of a multiplier @@ -1186,6 +1189,8 @@ def test_a_categorical_row_that_prints_a_digit_is_a_problem(raw): ("–2", (-2,), (2,)), ("1E-05", (1e-5,), (1,)), ("1,234.5", (1234.5,), (1.2345,)), + # a range may fall only if neither number is positive + ("0 to -3", (0, -3), (3,)), ], ) def test_each_form_reads_at_its_values_and_no_others(raw, passes, fails): @@ -1202,16 +1207,65 @@ def test_each_form_reads_at_its_values_and_no_others(raw, passes, fails): ) -@pytest.mark.parametrize("dash", ["-", "−", "–", "—", "‐", "‑", " - "]) +@pytest.mark.parametrize( + ("raw", "whole"), + [("two and a half", 2.5), ("2 and a half", 2.5), ("one hundred and a half", 100.5)], +) +def test_a_whole_number_with_a_fraction_takes_the_normalization_s_factor(raw, whole): + """With no word of its own, a whole number with a fraction takes the factor, + within 0.1%, and fails at 1.""" + note = "£ million x 1,000,000" + scaled = whole * 1e6 + assert problems(_share(raw, scaled, note)) == [] + assert problems(_share(raw, scaled * 1.0009, note)) == [] + for value in (whole, scaled * 1.00105, whole * 1e9): + found = problems(_share(raw, value, note)) + assert any("at a stated scale" in p for p in found), (raw, value, found) + + +@pytest.mark.parametrize("dash", ["-", "−", "–", "—", "‐", "‑", "‒", "―", " - "]) def test_a_dash_is_zero_only_where_the_normalization_says_it_prints_nil(dash): """A table may print a dash for nil or for "not available", so the checker reads one only if the normalization records nil, and then only as zero.""" - for note in ("dash for nil", "the table prints NIL as a dash"): + for note in ( + "dash for nil", + "dashes for nil", + "dash means nil", + "dash = nil", + "nil = dash", + "nil as dash", + "the table prints NIL as a dash", + "prints a dash for nil", + "print nil as a dash", + "the dash means nil", + "dashes mean nil", + "dash for nil", + "the producer prints a dash for nil", + "source prints nil as a dash", + "the workbook prints dashes for nil", + "£ million x 1e6; dash for nil", + "as printed, dash for nil", + "as printed. dash for nil", + "as printed (dash for nil)", + ): assert problems(_share(dash, 0, note)) == [], (dash, note) for value in (1, -1, 0.5, 1e6): found = problems(_share(dash, value, note)) assert any("at a stated scale" in p for p in found), (dash, value, found) - for note in ("as printed", "dash for not available", "vanilla"): + for note in ( + "as printed", + "dash for not available", + "dash means not available, not nil", + "dash means missing, not nil", + "dash does not mean nil", + "not available; nil is printed as 0", + "no dash for nil", + "not a dash for nil", + "never prints nil as a dash", + "nil-rate band", + "nil", + "vanilla", + ): assert _cant_read(dash, note), (dash, note) @@ -1434,6 +1488,79 @@ def test_digits_alone_take_the_normalization_s_factor_failing_that_one(n): assert any("at a stated scale" in p for p in found), (raw, scale, found) +@pytest.mark.parametrize( + ("note", "passes", "fails"), + [ + ("£ million x 1e6", (4e6,), (4, 40, 4e9)), + ("£ million X 1E6", (4e6,), (4, 40, 4e9)), + ("£ million x 10^6", (4e6,), (4, 40, 4e9)), + ("£ million x 10**6", (4e6,), (4, 40, 4e9)), + ("£ million x 10 ^ 6", (4e6,), (4, 40, 4e9)), + ("£ million × 1,000,000", (4e6,), (4, 4e3)), + ("£ million (x 1e6)", (4e6,), (4,)), + ("£ million [x 1e6]", (4e6,), (4,)), + ("(x 1e6), sign as printed", (4e6,), (4,)), + ("£ million x 1,000,000, sign as printed", (4e6,), (4,)), + ("£ million x 1e6: sign as printed", (4e6,), (4,)), + ("£ million x 1e6.", (4e6,), (4,)), + ("£ x 1,000,000 monthly", (4e6,), (4,)), + ("thousands * 1000", (4e3,), (4, 4e6)), + ("÷ 100", (0.04,), (4, 400)), + ("percent / 100", (0.04,), (4, 400)), + ("x .5", (2,), (4, 8)), + ("x 1,000,000 / 100", (4e4,), (4, 4e6, 0.04)), + ("x 1,000,000 ÷ 100 * 2", (8e4,), (4, 4e6, 4e4, 0.04)), + # two statements, so two readings; a semicolon ends a run of operations + ("x 1e6; percent / 100", (4e6, 0.04), (4, 4e4)), + ("x 1e6; / 100", (4e6, 0.04), (4, 4e4)), + # no factor here + ("x-axis", (4,), (40,)), + ("see box 5", (4,), (20,)), + ("printed as a fall ('lower' / 'fall' / 'drop')", (4,), (40,)), + ("4x", (4,), (40,)), + ], +) +def test_a_bare_number_takes_each_whole_factor_the_normalization_states( + note, passes, fails +): + """Sol R2 and R7: the checker reads a factor's whole operand, and composes + operations in a row, so "4" under "x 10^6" never passes at 40.""" + for value in passes: + assert problems(_share("4", value, note)) == [], (note, value) + for value in fails: + found = problems(_share("4", value, note)) + assert any("at a stated scale" in p for p in found), (note, value, found) + + +@pytest.mark.parametrize( + "note", + [ + "x 1 million", + "x 2 bn", + "x 1 000 000", + "x 10^", + "x 5^2", + "x 5**2", + "x 1,00", + "x 0", + "/ 0", + "x 1e999", + "x 10^999", + "x 1e6a", + "x 100%", + "x 1e6->GBP", + "x 1,000,000 / 100 million", + "x 1,000,000 / 10^", + ], +) +def test_a_factor_the_checker_cant_read_whole_is_a_problem_at_any_value(note): + """Sol R2: a numeric prefix doesn't count as the factor, and the checker never + falls back to 1 past a factor it can't read.""" + for value in (4, 40, 4e6, 0.04, 0): + found = problems(_share("4", value, note)) + assert any("a factor in normalization" in p for p in found), (note, value) + + @pytest.mark.parametrize( ("raw", "value", "wrong"), [("four", 4, 40), ("twenty-five", 25, 250), ("a hundred and five", 105, 1050)], @@ -1706,6 +1833,8 @@ def test_rule_4_holds_whatever_the_title_case(): ("Figure 1—B data label", False), ("Figure 1‑A data label", False), ("Figure 1‐A data label", False), + ("Figure 1‒A data label", False), + ("Figure 1―A data label", False), ("Figures 1-A and 1-B data labels", False), ("Figure 1.A data-label", False), ("paragraphs 1.7—1.8; Figure 1.A", False), @@ -1732,6 +1861,90 @@ def test_rule_4_holds_whatever_the_title_case(): ("paragraph 1.7; graphs", True), ("paragraph 1.7; charts", True), ("paragraph 1.7; not a bar reading from Figure 1.A", False), + # the figure in any spelling, and the number read on, in or using it + ("paragraph 1.7; read off Fig. 1.A", True), + ("paragraph 1.7; estimated from Fig 1A", True), + ("paragraph 1.7; read off Figs. 1.A and 1.B", True), + ("paragraph 1.7; taken from Fig. 1.B, decile 3", True), + ("paragraph 1.7; eyeballed off Fig. 1.A", True), + ("paragraph 1.7; reading of Fig. 1.A", True), + ("paragraph 1.7; bar reading from Fig. 1.A", True), + ("paragraph 1.7; read on Figure 1.A", True), + ("paragraph 1.7; read in Figure 1.A", True), + ("paragraph 1.7; estimated using Figure 1.A", True), + ("paragraph 1.7; measured on Figure 1.A", True), + ("paragraph 1.7; not read off Fig. 1.A", False), + ("paragraph 1.7; not read on Figure 1.A", False), + ("paragraph 1.7; not estimated using Figure 1.A", False), + ("paragraph 1.7; read in the text", False), + ("paragraph 1.7; read on page 8", False), + ("paragraph 1.7; figure in the text", False), + ("paragraph 1.7; configured using the table", False), + # a second negation anywhere earlier in the clause undoes the first, and + # one in an earlier clause doesn't + ("paragraph 1.7; not possible without reading off the chart", True), + ("paragraph 1.7; cannot be had without reading off the chart", True), + ("paragraph 1.7; can't be had without reading off the chart", True), + ("paragraph 1.7; no, not a bar reading", False), + ("paragraph 1.7; no rounding; not a bar reading", False), + ("paragraph 1.7: no rounding: not a bar reading", False), + ("paragraph 1.7 (no rounding) not a bar reading", False), + ("paragraph 1.7; cannot read off the chart", False), + # Sol R1: an abbreviated figure counts as a figure + ("paragraph 1.7; taken from Fig. 1.A", True), + ("paragraph 1.7; taken from Figs. 1.A and 1.B", True), + ("paragraph 1.7; estimated using fig 1.A", True), + ("paragraph 1.7; not taken from Fig. 1.A", False), + ("Fig. 1.A data label", False), + # Sol R8: the rarer ways to say we read a number off a picture + ("paragraph 1.7; estimated from the plot", True), + ("paragraph 1.7; estimated from the diagram", True), + ("paragraph 1.7; taken from the visual", True), + ("paragraph 1.7; visual estimate of Figure 1.A", True), + ("paragraph 1.7; estimate of Figure 1.A", True), + ("paragraph 1.7; Figure 1.A read by eye", True), + ("paragraph 1.7; read by eye", True), + ("paragraph 1.7; by eye", True), + ("paragraph 1.7; estimated from the plots", True), + ("paragraph 1.7; estimated from the diagrams", True), + ("paragraph 1.7; taken from the visuals", True), + ("paragraph 1.7; column 3 of Figure 1.A", True), + ("paragraph 1.7; columns 3-4 of Figure 1.A", True), + ("paragraph 1.7; the column on Fig. 1.A", True), + ("paragraph 1.7; the third column in Fig. 1.A", True), + ("paragraph 1.7; not estimated from the plot", False), + ("paragraph 1.7; no diagram", False), + ("paragraph 1.7; no visual estimate of Figure 1.A", False), + ("paragraph 1.7; not read by eye", False), + ("paragraph 1.7; Figure 1.A not read by eye", False), + ("paragraph 1.7; no column of Figure 1.A", False), + # a column of a table prints its number + ("Table 2.1, column 3", False), + ("paragraph 1.7; column 3 of Table 2.1", False), + ("paragraph 1.7; subcolumn of Figure 1.A", False), + ("paragraph 1.7; nearby eye", False), + ("paragraph 1.7; by eyewitnesses", False), + ("paragraph 1.7; Table 2.1 column 3", False), + # Sol R7: a negation in capitals reads the same + ("paragraph 1.7; NOT a bar reading", False), + ("paragraph 1.7; NEVER read off the chart", False), + ("paragraph 1.7; DIDN'T read it off the chart", False), + ("paragraph 1.7; NOT WITHOUT a bar reading", True), + ("paragraph 1.7; NOT possible WITHOUT reading off the chart", True), + # the cloud review of 3c65fd7a: Figure 1(a) names Figure 1.A + ("Figure 1(a), decile 3", True), + ("Figure 1 (a), decile 3", True), + ("Fig. 1(b)", True), + ("paragraph 1.7; Figure 1(a)", False), + # plurals count + ("paragraph 1.7; the decile bars", True), + # a word that holds a negation isn't one, and a bracket ends a clause + ("paragraph 1.7; casino not a bar reading", False), + ("paragraph 1.7; note not a bar reading", False), + ("paragraph 1.7; no rounding (not a bar reading)", False), + # each earlier negation counts + ("paragraph 1.7; without doubt not a bar reading", True), + ("paragraph 1.7; didn’t say not a bar reading", True), # any phrase that stands decides, and a verb starts at a word's start ("paragraph 1.7; a bar reading but not a chart reading", True), ("paragraph 1.7; spread from no chart", False), @@ -2251,3 +2464,10 @@ def test_any_family_file_gives_problems_never_an_exception(lines): ) found = family_problems(fam) assert all(isinstance(p, str) for p in found) + + +@pytest.mark.parametrize("dash", "-−–—‐‑‒―") +def test_rule_4_reads_each_dash_as_a_space(dash): + """The checker reads a phrase and its negation through any dash.""" + assert reads_a_bar(f"paragraph 1.7; read by{dash}eye") + assert not reads_a_bar(f"paragraph 1.7; not{dash}a bar reading") From 4707e9177933c4ab45508b41341961575f6dbbec Mon Sep 17 00:00:00 2001 From: Max Ghenis Date: Thu, 5 Nov 2026 02:32:54 -0500 Subject: [PATCH 50/52] Report unread operations and look past punctuation in rule 4 GPT-6.1 Sol and a cloud session each reviewed 82068821 and requested changes, and this commit answers both. - `_factors` returns None for an operation on a bracket, sign, currency sign or number words, and for an operation in words, so the checker reports that row at any value (Sol F1). - `_CONTINUES` reads "hundred" too, so "x 1 hundred" no longer reads as 1 (Sol F1). - Rule 4 looks past a colon, bracket or quote before a figure (Sol F2), reads "fig1a" and "figure1.A" (cloud F4), accepts "Figure 1(a) data label" as a printed place (cloud F6) and ends a clause at a full stop (cloud F14). - The tests add each case, each short multiplier on a whole number with a fraction, and ordinary notes that state no factor (Sol F3). - The README's contract bullets state what the checker now does (cloud F1, F2, F7 to F10; Sol F4). We change no staged row, and the checker finds 0 problems in the 12 families. Co-Authored-By: Claude Opus 5.5 --- pipeline/ab2026_harvest_contract.py | 57 ++++-- .../harvest-uk-ab2026-2026-10-29/README.md | 10 +- tests/test_ab2026_harvest_contract.py | 175 ++++++++++++++++++ 3 files changed, 218 insertions(+), 24 deletions(-) diff --git a/pipeline/ab2026_harvest_contract.py b/pipeline/ab2026_harvest_contract.py index aae9b2c32..28e5a4526 100644 --- a/pipeline/ab2026_harvest_contract.py +++ b/pipeline/ab2026_harvest_contract.py @@ -445,8 +445,20 @@ def _read(raw, normalization=""): _STEP = r"[x×*/÷]\s*\.?[0-9][^\s;]*" _CHAIN = re.compile(rf"(?:^|(?<=[\s(\[]))({_STEP}(?:\s+{_STEP})*)") _OPERAND = re.compile(rf"10(?:\^|\*\*)(?P[0-9]+)|{_DIGITS}") -# What would carry on a number the chain seems to end: "x 1 000 000", "x 1 million" -_CONTINUES = re.compile(rf"\s+(?:[0-9]|(?:{_either(LARGE)}|{_either(SHORT)})\b)") +# What would carry on a number the chain seems to end: "x 1 000 000", "x 1 million", +# "x 1 hundred" +_CONTINUES = re.compile(rf"\s+(?:[0-9]|(?:{_either(MULTIPLIERS)})\b)") +# An operation on what the checker can't read as a number: a sign, bracket or +# currency sign first ("x (1e6)", "x +1e6", "x £100"), number words ("x one +# million", "x hundred"), or an operation in words ("times 1,000", "divided by 100"). +_OPENERS = rf"[{_DASHES}+(\[£$€]\s*" +_NUMBER_WORDS = rf"(?:{_WORDS}|hundred|{_LARGE})\b" +_UNREAD = re.compile( + r"(?:^|(?<=[\s(\[]))(?:" + rf"[x×*/÷]\s*(?:(?:{_OPENERS})+\.?[0-9]|{_NUMBER_WORDS})" + rf"|(?:times|multiplied by|divided by)\s+(?:(?:{_OPENERS})*\.?[0-9]|{_NUMBER_WORDS})" + r")" +) def _factors(normalization): @@ -454,6 +466,8 @@ def _factors(normalization): or None if it states one the checker can't read whole. Operations in a row make one multiplier: "x 1,000,000 / 100" is 10,000.""" text = re.sub(r"\s*(\^|\*\*)\s*", r"\1", (normalization or "").lower()) + if _UNREAD.search(text): + return None out = [] for chain in _CHAIN.finditer(text): if _CONTINUES.match(text, chain.end()): @@ -562,20 +576,21 @@ def value_problems(r): # README rule 4. A locator that names Figure 1.A or 1.B ("Figure 1.A", "Fig. -# 1A", "Figures 1.A and 1.B") says where HM Treasury prints the number only -# if its first reference, up to its first comma or semicolon, takes one of -# the forms PRINTED reads and nothing more: a paragraph, table or box by its -# number, or a data label. The checker reads any other first reference as a +# 1A", "Figure 1(a)", "Figures 1.A and 1.B") says where HM Treasury prints the +# number only if its first reference, up to its first comma or semicolon, takes +# one of the forms PRINTED reads and nothing more: a paragraph, table or box by +# its number, or a data label. The checker reads any other first reference as a # bar reading, "Figure 1.A data label unavailable" included. Anywhere in a # locator, a bar, chart, graph, plot, diagram or visual marks a bar reading, and -# so does a number taken from a figure or read by eye. A single negation just before the phrase cancels it ("not -# a bar reading", "didn't read it off the chart"), and a double negation doesn't -# ("not without a bar reading"). The checker looks for these phrases with hyphens -# and dashes as spaces and with runs of spaces collapsed, and reads the first -# reference with each dash as a hyphen. +# so does a number taken from a figure or read by eye. A single negation just +# before the phrase cancels it ("not a bar reading", "didn't read it off the +# chart"), and a double negation in the same clause doesn't ("not without a bar +# reading"). The checker looks for these phrases with hyphens and dashes as +# spaces and with runs of spaces collapsed, and reads the first reference with +# each dash as a hyphen. _DASH = f"[{_DASHES}]" FIGURE_1 = re.compile(r"\bfig(?:ure)?s?\.?\s*1\.?\s*\(?[ab]\b", re.I) -_FIG = r"fig(?:ure)?s?\.?\s*1[.-]?\s*[ab](?:\s+and\s+1[.-]?\s*[ab])?" +_FIG = r"fig(?:ure)?s?\.?\s*1[.-]?\s*\(?[ab]\)?(?:\s+and\s+1[.-]?\s*\(?[ab]\)?)?" _ID = r"\d+(?:\.[0-9a-z]+)*" PRINTED = re.compile( rf"(?:paragraph|para|table|box)(?:s|es)?\.?\s+{_ID}" @@ -585,20 +600,24 @@ def value_problems(r): re.I, ) _PICTURE = r"(?:bars?|charts?|graphs?|plots?|diagrams?|visuals?)" -_FIGS = r"fig(?:ure)?s?" +_FIGS = r"fig(?:ure)?s?(?![a-z])" _OFF = r"(?:off|from)" -_UP_TO = r"(?:[\w.]+ ){0,4}?" +_OPEN = r"""[\[("'“‘]""" +_UP_TO = rf"(?:{_OPEN}?[\w.]+ ){{0,4}}?(?:{_OPEN} ?)?" +_AFTER = rf"(?::? |:|(?={_OPEN}))" # A verb's whole phrase comes first, so that a negation before the verb cancels # it: "read it off the decile 3 bar", "reading of the chart", "visual estimate of # Figure 1.A". Then any picture, with a reading if it names one ("bar reading from # Figure 1.A"), any number from a figure or a figure's column, and any reading by # eye. A column of a table prints its number, so a column counts only in a figure. +# A colon, bracket or quote may come before the figure: "read off: Fig. 1.A", "read +# from [Figure 1.A]". BAR_READING = re.compile( r"\b(?:visual )?(?:(?:read|estimat|tak|measur)\w*|took)(?: [\w.]+){0,3}?" - rf" (?:off|from|on|in|using|of) {_UP_TO}(?:{_PICTURE}|{_FIGS})\b" - rf"|\b{_PICTURE}(?: read\w*(?: {_OFF} {_UP_TO}{_FIGS})?)?\b" - rf"|\b{_OFF} {_UP_TO}{_FIGS}\b" - rf"|\bcolumns?(?: [\w.]+){{0,2}}? (?:of|in|on) {_UP_TO}{_FIGS}\b" + rf" (?:off|from|on|in|using|of){_AFTER}{_UP_TO}(?:{_PICTURE}\b|{_FIGS})" + rf"|\b{_PICTURE}(?: read\w*(?: {_OFF}{_AFTER}{_UP_TO}{_FIGS})?)?\b" + rf"|\b{_OFF}{_AFTER}{_UP_TO}{_FIGS}" + rf"|\bcolumns?(?: [\w.]+){{0,2}}? (?:of|in|on){_AFTER}{_UP_TO}{_FIGS}" r"|\b(?:read\w* )?by eye\b", re.I, ) @@ -898,7 +917,7 @@ def _denied(before): m = DENIED.search(before) if not m: return False - clause = re.split(r"[,;:()]", before[: m.start()])[-1] + clause = re.split(r"[,;:().]", before[: m.start()])[-1] return not NEGATION.search(clause) diff --git a/sources/harvest-uk-ab2026-2026-10-29/README.md b/sources/harvest-uk-ab2026-2026-10-29/README.md index 181f727c5..d8d04b222 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/README.md +++ b/sources/harvest-uk-ab2026-2026-10-29/README.md @@ -67,15 +67,15 @@ We follow the AB2025 harvest README's rules 1–10 and its row contract, with th - number words up to nine hundred and ninety-nine, with a sign and one multiplier or percentage word: "four", "-four", "a hundred and five", "two hundred thousand", "a million", "four per cent"; - a worded fraction: "half" alone, or a numerator ("a", "an", "another", or one to ten) and a denominator from "half" to "tenth" ("a third", "three-fifths"), with a sign if printed ("-three-fifths"); - a worded fraction of a multiplier or percentage: "half a million", "a quarter of a percentage point", "half per cent"; - - a whole number, in digits or in words up to nine hundred and ninety-nine, and a fraction, with a multiplier or percentage word if printed: "two and a half", "one and a half million", "one hundred and a half"; - - a ratio of two numbers, in digits or number words, with at most one multiplier each: "1 in 4", "1 in 2 million". It takes "in", "in every", "out of" or "out of every" between them, with spaces or hyphens throughout: "3 out of 10", "one-in-every-ten"; + - a whole number, in digits or in words up to nine hundred and ninety-nine, and a fraction, with a percentage word or, after a space or hyphen, a multiplier other than "hundred", if printed: "two and a half", "one and a half million", "2 and a half bn", "one hundred and a half"; + - a ratio of two numbers, in digits or number words, with at most one multiplier each: "1 in 4", "1 in 2 million". The checker takes "in", "in every", "out of" or "out of every" between them, with spaces or hyphens throughout: "3 out of 10", "one-in-every-ten"; - a range: two figures joined by a dash, with no spaces or a space on each side ("£4-5 million", "4 – 5%"), or two numbers joined by "to" ("four to five", "4bn to 5bn"); - - "nil", "none", "nothing", "no change", "no impact" or "no overall impact", as zero. It also reads a dash as zero where we say, in one whole clause of `normalization`, that the producer prints it for nil ("dash for nil", "the table prints nil as a dash"). + - "nil", "none", "nothing", "no change", "no impact" or "no overall impact", as zero. It also reads a dash as zero where we say, in one whole clause of `normalization`, that the producer prints it for nil. It accepts "dash for nil" and "the table prints nil as a dash", and it ends a clause at a comma, semicolon, full stop or bracket. - It also reads a qualifier before the number ("about", "up to" and the others in `QUALIFIERS`), "a" before digits ("less than a 0.1 percentage point"), and "a year" or "of GDP" after the number. It reads a ratio or worded fraction with its percentage in brackets, in either order, if the two agree to the nearest percentage point: "one third (33%)", "25% (1 in 4)". It matches `value` to either number there or in a range. It reports a range that joins a percentage to a multiplier ("4% to 2 million") or a fraction to a percentage ("a third to 40%"). It also reports a range whose first number exceeds its second, unless neither is positive: it reports "2025-26" and "2 to -3", and reads "-4 to -5". After a bare "in", it reports a year from 2010 to 2099 ("5 in 2027"). -- It takes each number's scale from its own word (bn, million, %). When only a range's second number has one, it gives that word to the first number too. When digits, number words or a whole number with a fraction have no word of their own, it uses the factor in `normalization`, failing that 1. It reads a factor whole ("x 1,000,000", "/ 100", "(x 10^6)"), multiplies operations in a row ("x 1,000,000 / 100" as 10,000), and takes either factor where a semicolon separates two. It reports a factor it can't read ("x 1 million") at any value. It reads a percentage as a share or in percent units, and requires digits and number words to match `value` within four units in the last place. +- It takes each number's scale from its own word (bn, million, %). When only a range's second number has one, it gives that word to the first number too. When digits, number words or a whole number with a fraction have no word of their own, it uses the factor in `normalization`, failing that 1. It reads a factor whole ("x 1,000,000", "/ 100", "(x 10^6)"), multiplies operations in a row ("x 1,000,000 / 100" as 10,000), and takes either factor where a semicolon or a word separates two. It reads x, ×, *, / or ÷ as a factor wherever a number follows, at the note's start or after a space or bracket, so we keep other slashes and products out of `normalization`. It reports a factor it can't read ("x 1 million", "x (1e6)", "x one million", "times 1,000") at any value. It reads a percentage as a share or in percent units, and requires digits and number words to match `value` within four units in the last place. - It reads a ratio or a worded fraction on its own as a share, and accepts a `value` within 0.1% of that share. On a `unit_concept: percent` row, it requires 100 times the share instead ("1 in 4" as 25). It won't match `value` to either of a ratio's numbers alone, and it reads "-1 in 4" and "1 in -4" as −0.25. It also allows 0.1% on a whole number with a fraction, and on a fraction of a multiplier or percentage. - It requires `value` to share `value_raw`'s sign unless we note in `normalization` that we took the sign from the wording. -- It treats an HM Treasury Impact on households row as a bar reading (rule 4) if the row has `value_kind: approx_chart_reading`. It does the same if we mention a bar, chart, graph, plot, diagram or visual in the locator, in any spacing and with or without hyphens. It also does so if we write there that we read, took, estimated or measured the number off, from, of, on, in or using a figure, or read it by eye. It reads "Fig." and "Figs." as figures, so it counts "read off Fig. 1.A". It counts "column 3 of Figure 1.A" and leaves "Table 2.1, column 3" alone. It ignores such a phrase right after one "no", "not", "never", "without", "cannot" or "n't" ("not a bar reading"), but not after two in the same clause ("not possible without reading off the chart"). +- It treats an HM Treasury Impact on households row as a bar reading (rule 4) if the row has `value_kind: approx_chart_reading`. It does the same if we mention a bar, chart, graph, plot, diagram or visual in the locator, in any spacing and with or without hyphens. It also does so if "off" or "from" comes up to four words before a figure there. It does the same if we write that we read, took, estimated or measured the number of, on, in or using a figure, or read it by eye. It reads "Fig." and "Figs." as figures, and it looks past a colon, bracket or quote before a figure, so it counts "read off Fig. 1.A", "read off: Fig. 1.A" and "read from [Figure 1.A]". It counts "column 3 of Figure 1.A" and leaves "Table 2.1, column 3" alone. It ignores such a phrase right after one "no", "not", "never", "without", "cannot" or "n't" ("not a bar reading"). It counts the phrase after two negations in the same clause ("not possible without reading off the chart"). - When we cite Figure 1.A or 1.B on those rows ("Fig. 1A", "Figure 1(a)", "Figures 1.A and 1.B"), it reads the locator up to the first comma or semicolon. Before that point we must name a paragraph, table or box by number, or a data label. It accepts "paragraph 1.7", "paragraphs 1.7–1.8", "Tables 2.B and 2.C", "Table 2.C and Figure 1.A", "Figure 1.A data label" and "data label for decile 3". It treats any other wording as a bar reading, including "Figure 1.A data label unavailable" and "Figure 1.A, decile 3 bar; see paragraph 1.7". - It reports a malformed row as a problem, never as an exception: a row that isn't an object, lacks a field, or has a field of the wrong type. It requires: - a string in `source`, `source_model`, `value_raw`, `normalization`, `quote`, `status`, `value_kind`, `time_basis`, `attribution`, `benchmark_class` and `parse_confidence`; diff --git a/tests/test_ab2026_harvest_contract.py b/tests/test_ab2026_harvest_contract.py index c2d93136e..0abeedd0a 100644 --- a/tests/test_ab2026_harvest_contract.py +++ b/tests/test_ab2026_harvest_contract.py @@ -1508,6 +1508,8 @@ def test_digits_alone_take_the_normalization_s_factor_failing_that_one(n): ("÷ 100", (0.04,), (4, 400)), ("percent / 100", (0.04,), (4, 400)), ("x .5", (2,), (4, 8)), + ("x1e6", (4e6,), (4,)), + ("×1e6", (4e6,), (4,)), ("x 1,000,000 / 100", (4e4,), (4, 4e6, 0.04)), ("x 1,000,000 ÷ 100 * 2", (8e4,), (4, 4e6, 4e4, 0.04)), # two statements, so two readings; a semicolon ends a run of operations @@ -1550,6 +1552,15 @@ def test_a_bare_number_takes_each_whole_factor_the_normalization_states( "x 100%", "x 1e6->GBP", "x 1,000,000 / 100 million", + "x 1 hundred", + "x 1e6 / (100)", + "times 1,000", + "times 1,000", + "times .5", + "times (1,000)", + "times one", + "multiplied by 1e6", + "divided by 100", "x 1,000,000 / 10^", ], ) @@ -1936,6 +1947,58 @@ def test_rule_4_holds_whatever_the_title_case(): ("Figure 1 (a), decile 3", True), ("Fig. 1(b)", True), ("paragraph 1.7; Figure 1(a)", False), + # the cloud review of 82068821: a bracketed letter names the figure in a + # data label too + ("Figure 1(a) data label", False), + ("Figure 1 (a) data label", False), + ("Figures 1(a) and 1(b) data labels", False), + ("Figures 1.A and 1(b) data labels", False), + ("data label for Figure 1(b)", False), + ("paragraph 1.7 and Figure 1(a)", False), + ("Figure 1(a)", True), + ("Figure 1(a) bar", True), + # a number right after the word still names a figure + ("paragraph 1.7; read off fig1a", True), + ("paragraph 1.7; read off figure1.A", True), + ("paragraph 1.7; read off Fig1A", True), + ("paragraph 1.7; column 3 of fig1a", True), + ("paragraph 1.7; estimated using figure1.B", True), + # a word that starts like a figure or a picture doesn't count + ("paragraph 1.7; taken from the figurehead", False), + ("paragraph 1.7; read off the barometer", False), + ("paragraph 1.7; figures taken from Table 2.C", False), + ("paragraph 1.7; configured using the table", False), + # Sol's review of 82068821: punctuation before the figure doesn't hide it + ("paragraph 1.7; estimated using (Figure 1.A)", True), + ("paragraph 1.7; estimated using: Figure 1.A", True), + ("paragraph 1.7; estimated using(Figure 1.A)", True), + ("paragraph 1.7; read from [Figure 1.A]", True), + ("paragraph 1.7; read off: Fig. 1.A", True), + ("paragraph 1.7; read off:Fig. 1.A", True), + ("paragraph 1.7; read from(Figure 1.A)", True), + ("paragraph 1.7; read from ( Figure 1.A)", True), + ("paragraph 1.7; read from: [Figure 1.A]", True), + ("paragraph 1.7; read from (the Figure 1.A)", True), + ('paragraph 1.7; read from "Figure 1.A"', True), + ("paragraph 1.7; read from 'Figure 1.A'", True), + ("paragraph 1.7; read from “Figure 1.A”", True), + ("paragraph 1.7; read from ‘Figure 1.A’", True), + ("paragraph 1.7; from [Figure 1.A]", True), + ("paragraph 1.7; from: Figure 1.A", True), + ("paragraph 1.7; column 3 of (Figure 1.A)", True), + ("paragraph 1.7; column 3 of: Figure 1.A", True), + # and one negation still cancels the phrase + ("paragraph 1.7; not read from [Figure 1.A]", False), + ("paragraph 1.7; not read off: Fig. 1.A", False), + ("paragraph 1.7; no bar reading from: Figure 1.A", False), + ("paragraph 1.7; no bar reading from [Figure 1.A]", False), + # and a table in brackets still prints its number + ("paragraph 1.7; figures from (Table 2.C)", False), + ("paragraph 1.7; read from [Table 2.C]", False), + ("paragraph 1.7; read off: Table 2.C", False), + # a full stop ends a clause + ("paragraph 1.7. No rounding. Not a bar reading", False), + ("paragraph 1.7; no. 5 not a bar reading", False), # plurals count ("paragraph 1.7; the decile bars", True), # a word that holds a negation isn't one, and a bracket ends a clause @@ -2471,3 +2534,115 @@ def test_rule_4_reads_each_dash_as_a_space(dash): """The checker reads a phrase and its negation through any dash.""" assert reads_a_bar(f"paragraph 1.7; read by{dash}eye") assert not reads_a_bar(f"paragraph 1.7; not{dash}a bar reading") + + +@pytest.mark.parametrize( + "op", ["x", "×", "*", "/", "÷", "times", "multiplied by", "divided by"] +) +@pytest.mark.parametrize( + "operand", + [ + "(1e6)", + "( 1e6)", + "(.5)", + "[1e6]", + "+1e6", + "-1e6", + "−1e6", + "–1e6", + "—1e6", + "‐1e6", + "‑1e6", + "‒1e6", + "―1e6", + "(-1)", + "£1,000", + "$1,000", + "€1,000", + "one", + "twenty-five", + "one million", + "a million", + "a hundred", + "hundred", + "thousand", + "million", + "billion", + "trillion", + "zero", + ], +) +def test_an_operand_the_checker_cant_read_is_a_problem_at_any_value(op, operand): + """Sol's review of 82068821: the checker never scales by 1, or by part of a + chain, past an operation on a bracket, a sign, a currency sign or words.""" + notes = [ + f"{op} {operand}", + f"as printed; {op} {operand}", + f"({op} {operand})", + f"[{op} {operand}]", + f"x 1e6 {op} {operand}", + ] + if len(op) == 1: + notes.append(f"{op}{operand}") + for note in notes: + for value in (4, 4e6): + found = problems(_share("4", value, note)) + assert any("a factor in normalization" in p for p in found), (note, value) + + +@pytest.mark.parametrize( + "note", + [ + "as printed", + "x-axis", + "4x", + "box 5", + "box (5)", + "box one", + "cell 2/3", + "and/or 5", + "4x (5)", + "x onerous", + "x millions", + "/ thousands", + "x (see note)", + "* see note", + "* (note 5)", + "x + y", + "£x million", + "printed as a fall ('lower' / 'fall' / 'drop')", + "sometimes 5", + "3 times the figure", + "times onerous", + "divided by region", + "multiplied by the rate", + ], +) +def test_an_ordinary_note_states_no_factor(note): + """A note that holds an operator or a number beside other words scales nothing, + and the checker reports nothing about a factor.""" + assert problems(_share("4", 4, note)) == [], note + found = problems(_share("4", 4e6, note)) + assert any("at a stated scale" in p for p in found), (note, found) + + +@pytest.mark.parametrize( + ("raw", "value", "wrong"), + [ + ("2 and a half k", 2500, 2.5), + ("2 and a half m", 2.5e6, 2500), + ("2 and a half mn", 2.5e6, 2500), + ("2 and a half bn", 2.5e9, 2.5e6), + ("2 and a half tn", 2.5e12, 2.5e9), + ("2 and a half-bn", 2.5e9, 2.5e6), + ("two and a half thousand", 2500, 2.5), + ("two and a half million", 2.5e6, 2500), + ("two and a half billion", 2.5e9, 2.5e6), + ("two and a half trillion", 2.5e12, 2.5e9), + ], +) +def test_a_whole_number_with_a_fraction_reads_each_multiplier(raw, value, wrong): + """Sol's review of 82068821: its own multiplier scales it.""" + assert problems(_share(raw, value, "as printed")) == [] + found = problems(_share(raw, wrong, "as printed")) + assert any("at a stated scale" in p for p in found), (raw, found) From 2f60cdb80af333789a19ae737f65740213ee46bb Mon Sep 17 00:00:00 2001 From: Max Ghenis Date: Thu, 5 Nov 2026 07:21:31 -0500 Subject: [PATCH 51/52] Report more unread operations and read nested wrappers in rule 4 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit GPT-6.1 Sol and a cloud session each reviewed 4707e917 and requested changes. - The checker's `_UNREAD` now reports an operation in words with any spacing ("multiplied by 1e6", "x by 1e6"), an operation on a wrapped or quoted number word ("x (-one million)"), and an operator with no space before it ("£m×1e6", "£m;x 1e6", "%/100", "£m/100"). It still ignores a slash between digits ("2/3", "2026/27"). - Rule 4 now reads nested wrappers ('read from ("Figure 1.A")'), a word in a bracket or quotes before the figure ("read from [the Figure 1.A]"), and a word that holds an apostrophe or ends a bracket ("HMT's Figure 1.A", "(HMT) Figure 1.A"). A word's first character excludes quotes, so reading time stays flat on a long run of quotes, and a test holds it under two seconds. - New tests fail on 4707e917's checker where it missed these forms, and cover an unread factor on a number that owns a multiplier, a share, a fraction or nil. - The README states the forms the checker reports and tells stagers to write "page 3 of 4" for a slash that divides nothing. No staged row changed, and the checker reports 0 problems on 5,952 rows in 12 families. Co-Authored-By: Claude Opus 5.5 --- pipeline/ab2026_harvest_contract.py | 25 +-- .../harvest-uk-ab2026-2026-10-29/README.md | 8 +- tests/test_ab2026_harvest_contract.py | 142 +++++++++++++++++- 3 files changed, 159 insertions(+), 16 deletions(-) diff --git a/pipeline/ab2026_harvest_contract.py b/pipeline/ab2026_harvest_contract.py index 28e5a4526..a615bf1ef 100644 --- a/pipeline/ab2026_harvest_contract.py +++ b/pipeline/ab2026_harvest_contract.py @@ -448,16 +448,22 @@ def _read(raw, normalization=""): # What would carry on a number the chain seems to end: "x 1 000 000", "x 1 million", # "x 1 hundred" _CONTINUES = re.compile(rf"\s+(?:[0-9]|(?:{_either(MULTIPLIERS)})\b)") -# An operation on what the checker can't read as a number: a sign, bracket or -# currency sign first ("x (1e6)", "x +1e6", "x £100"), number words ("x one -# million", "x hundred"), or an operation in words ("times 1,000", "divided by 100"). -_OPENERS = rf"[{_DASHES}+(\[£$€]\s*" +# An operation on what the checker can't read as a number: a sign, bracket, quote +# or currency sign first ("x (1e6)", "x +1e6", "x '1e6'", "x £100"), or number words +# ("x one million", "x (a million)"). Also an operation in words ("times 1,000", +# "multiply by 1e6", "divided by 100", "scaled by 1e6", "x by 1e6"), and an operator +# with no space before it ("£m×1e6", "£m;x 1e6", "%/100", "£m/100"). +_OPENERS = rf"[{_DASHES}+(\[£$€'\"‘“]\s*" _NUMBER_WORDS = rf"(?:{_WORDS}|hundred|{_LARGE})\b" +_ANY_OPERAND = rf"(?:{_OPENERS})*(?:\.?[0-9]|{_NUMBER_WORDS})" _UNREAD = re.compile( r"(?:^|(?<=[\s(\[]))(?:" - rf"[x×*/÷]\s*(?:(?:{_OPENERS})+\.?[0-9]|{_NUMBER_WORDS})" - rf"|(?:times|multiplied by|divided by)\s+(?:(?:{_OPENERS})*\.?[0-9]|{_NUMBER_WORDS})" + rf"[x×*/÷]\s*(?:(?:{_OPENERS})+\.?[0-9]|(?:{_OPENERS})*{_NUMBER_WORDS})" + rf"|(?:times|(?:multipl|divid|scal)\w*\s+by|x\s+by)\s+{_ANY_OPERAND}" r")" + rf"|(?<=[^\s(\[*])[×*÷]\s*{_ANY_OPERAND}" + rf"|(?<=[;:,=>%)\]])[x/]\s*{_ANY_OPERAND}" + rf"|(?<=[a-z])/\s*(?:{_OPENERS})*\.?[0-9]" ) @@ -603,15 +609,16 @@ def value_problems(r): _FIGS = r"fig(?:ure)?s?(?![a-z])" _OFF = r"(?:off|from)" _OPEN = r"""[\[("'“‘]""" -_UP_TO = rf"(?:{_OPEN}?[\w.]+ ){{0,4}}?(?:{_OPEN} ?)?" +_UP_TO = rf"(?:(?:{_OPEN} *)*[\w.][\w.'’]*[)\]]? ){{0,4}}?(?:{_OPEN} *)*" _AFTER = rf"(?::? |:|(?={_OPEN}))" # A verb's whole phrase comes first, so that a negation before the verb cancels # it: "read it off the decile 3 bar", "reading of the chart", "visual estimate of # Figure 1.A". Then any picture, with a reading if it names one ("bar reading from # Figure 1.A"), any number from a figure or a figure's column, and any reading by # eye. A column of a table prints its number, so a column counts only in a figure. -# A colon, bracket or quote may come before the figure: "read off: Fig. 1.A", "read -# from [Figure 1.A]". +# A colon, brackets or quotes may come before the figure: "read off: Fig. 1.A", "read +# from ("Figure 1.A")". A word there may hold an apostrophe or end a bracket: "taken +# from HMT's Figure 1.A", "read from (HMT) Figure 1.A". BAR_READING = re.compile( r"\b(?:visual )?(?:(?:read|estimat|tak|measur)\w*|took)(?: [\w.]+){0,3}?" rf" (?:off|from|on|in|using|of){_AFTER}{_UP_TO}(?:{_PICTURE}\b|{_FIGS})" diff --git a/sources/harvest-uk-ab2026-2026-10-29/README.md b/sources/harvest-uk-ab2026-2026-10-29/README.md index d8d04b222..1f6fb7dfd 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/README.md +++ b/sources/harvest-uk-ab2026-2026-10-29/README.md @@ -67,15 +67,15 @@ We follow the AB2025 harvest README's rules 1–10 and its row contract, with th - number words up to nine hundred and ninety-nine, with a sign and one multiplier or percentage word: "four", "-four", "a hundred and five", "two hundred thousand", "a million", "four per cent"; - a worded fraction: "half" alone, or a numerator ("a", "an", "another", or one to ten) and a denominator from "half" to "tenth" ("a third", "three-fifths"), with a sign if printed ("-three-fifths"); - a worded fraction of a multiplier or percentage: "half a million", "a quarter of a percentage point", "half per cent"; - - a whole number, in digits or in words up to nine hundred and ninety-nine, and a fraction, with a percentage word or, after a space or hyphen, a multiplier other than "hundred", if printed: "two and a half", "one and a half million", "2 and a half bn", "one hundred and a half"; + - a whole number and a fraction, with the whole number in digits or in words up to nine hundred and ninety-nine. A percentage word may follow, and so may a multiplier other than "hundred" after a space or hyphen: "two and a half", "one and a half million", "2 and a half bn", "one hundred and a half"; - a ratio of two numbers, in digits or number words, with at most one multiplier each: "1 in 4", "1 in 2 million". The checker takes "in", "in every", "out of" or "out of every" between them, with spaces or hyphens throughout: "3 out of 10", "one-in-every-ten"; - a range: two figures joined by a dash, with no spaces or a space on each side ("£4-5 million", "4 – 5%"), or two numbers joined by "to" ("four to five", "4bn to 5bn"); - - "nil", "none", "nothing", "no change", "no impact" or "no overall impact", as zero. It also reads a dash as zero where we say, in one whole clause of `normalization`, that the producer prints it for nil. It accepts "dash for nil" and "the table prints nil as a dash", and it ends a clause at a comma, semicolon, full stop or bracket. + - "nil", "none", "nothing", "no change", "no impact" or "no overall impact", as zero. It also reads a dash as zero where we say, in one whole clause of `normalization`, that the producer prints it for nil. It accepts forms such as "dash for nil" and "the table prints nil as a dash". A comma, semicolon, full stop or bracket ends a clause. - It also reads a qualifier before the number ("about", "up to" and the others in `QUALIFIERS`), "a" before digits ("less than a 0.1 percentage point"), and "a year" or "of GDP" after the number. It reads a ratio or worded fraction with its percentage in brackets, in either order, if the two agree to the nearest percentage point: "one third (33%)", "25% (1 in 4)". It matches `value` to either number there or in a range. It reports a range that joins a percentage to a multiplier ("4% to 2 million") or a fraction to a percentage ("a third to 40%"). It also reports a range whose first number exceeds its second, unless neither is positive: it reports "2025-26" and "2 to -3", and reads "-4 to -5". After a bare "in", it reports a year from 2010 to 2099 ("5 in 2027"). -- It takes each number's scale from its own word (bn, million, %). When only a range's second number has one, it gives that word to the first number too. When digits, number words or a whole number with a fraction have no word of their own, it uses the factor in `normalization`, failing that 1. It reads a factor whole ("x 1,000,000", "/ 100", "(x 10^6)"), multiplies operations in a row ("x 1,000,000 / 100" as 10,000), and takes either factor where a semicolon or a word separates two. It reads x, ×, *, / or ÷ as a factor wherever a number follows, at the note's start or after a space or bracket, so we keep other slashes and products out of `normalization`. It reports a factor it can't read ("x 1 million", "x (1e6)", "x one million", "times 1,000") at any value. It reads a percentage as a share or in percent units, and requires digits and number words to match `value` within four units in the last place. +- It takes each number's scale from its own word (bn, million, %). When only a range's second number has one, it gives that word to the first number too. When digits, number words or a whole number with a fraction have no word of their own, it uses the factor in `normalization`, failing that 1. It reads a factor whole ("x 1,000,000", "/ 100", "(x 10^6)"), multiplies operations in a row ("x 1,000,000 / 100" as 10,000), and takes either factor where a semicolon or a word separates two. It reads x, ×, *, / or ÷ as a factor wherever a number follows, at the note's start or after a space or bracket. It ignores a slash between digits ("2/3", "2026/27"), and we write "page 3 of 4" rather than "page 3 / 4". At any value, it reports "x 1 million", "x (1e6)", "x one million", "times 1,000", "multiply by 1,000,000" and "£m×1e6". It reads a percentage as a share or in percent units, and requires digits and number words to match `value` within four units in the last place. - It reads a ratio or a worded fraction on its own as a share, and accepts a `value` within 0.1% of that share. On a `unit_concept: percent` row, it requires 100 times the share instead ("1 in 4" as 25). It won't match `value` to either of a ratio's numbers alone, and it reads "-1 in 4" and "1 in -4" as −0.25. It also allows 0.1% on a whole number with a fraction, and on a fraction of a multiplier or percentage. - It requires `value` to share `value_raw`'s sign unless we note in `normalization` that we took the sign from the wording. -- It treats an HM Treasury Impact on households row as a bar reading (rule 4) if the row has `value_kind: approx_chart_reading`. It does the same if we mention a bar, chart, graph, plot, diagram or visual in the locator, in any spacing and with or without hyphens. It also does so if "off" or "from" comes up to four words before a figure there. It does the same if we write that we read, took, estimated or measured the number of, on, in or using a figure, or read it by eye. It reads "Fig." and "Figs." as figures, and it looks past a colon, bracket or quote before a figure, so it counts "read off Fig. 1.A", "read off: Fig. 1.A" and "read from [Figure 1.A]". It counts "column 3 of Figure 1.A" and leaves "Table 2.1, column 3" alone. It ignores such a phrase right after one "no", "not", "never", "without", "cannot" or "n't" ("not a bar reading"). It counts the phrase after two negations in the same clause ("not possible without reading off the chart"). +- It treats an HM Treasury Impact on households row as a bar reading (rule 4) if the row has `value_kind: approx_chart_reading`. It does the same if we mention a bar, chart, graph, plot, diagram or visual in the locator, in any spacing and with or without hyphens. "Off" or "from" up to four words before a figure there counts too. So does a note that we read, took, estimated or measured the number using, on, in or of a figure, or read it by eye. It reads "Fig." and "Figs." as figures and looks past a colon, brackets or quotes before them. It counts "read off Fig. 1.A", "read off: Fig. 1.A" and "read from [Figure 1.A]". It counts "column 3 of Figure 1.A" and leaves "Table 2.1, column 3" alone. It ignores such a phrase right after one "no", "not", "never", "without", "cannot" or "n't" ("not a bar reading"). It counts the phrase after two negations in the same clause ("not possible without reading off the chart"). - When we cite Figure 1.A or 1.B on those rows ("Fig. 1A", "Figure 1(a)", "Figures 1.A and 1.B"), it reads the locator up to the first comma or semicolon. Before that point we must name a paragraph, table or box by number, or a data label. It accepts "paragraph 1.7", "paragraphs 1.7–1.8", "Tables 2.B and 2.C", "Table 2.C and Figure 1.A", "Figure 1.A data label" and "data label for decile 3". It treats any other wording as a bar reading, including "Figure 1.A data label unavailable" and "Figure 1.A, decile 3 bar; see paragraph 1.7". - It reports a malformed row as a problem, never as an exception: a row that isn't an object, lacks a field, or has a field of the wrong type. It requires: - a string in `source`, `source_model`, `value_raw`, `normalization`, `quote`, `status`, `value_kind`, `time_basis`, `attribution`, `benchmark_class` and `parse_confidence`; diff --git a/tests/test_ab2026_harvest_contract.py b/tests/test_ab2026_harvest_contract.py index 0abeedd0a..d856cd723 100644 --- a/tests/test_ab2026_harvest_contract.py +++ b/tests/test_ab2026_harvest_contract.py @@ -13,6 +13,7 @@ import math import shutil import tempfile +import time from itertools import product from pathlib import Path @@ -1510,6 +1511,10 @@ def test_digits_alone_take_the_normalization_s_factor_failing_that_one(n): ("x .5", (2,), (4, 8)), ("x1e6", (4e6,), (4,)), ("×1e6", (4e6,), (4,)), + ("(×1e6)", (4e6,), (4,)), + ("[×1e6]", (4e6,), (4,)), + ("£ billion × 1e9 → £", (4e9,), (4,)), + ("per cent / 100 → share", (0.04,), (4,)), ("x 1,000,000 / 100", (4e4,), (4, 4e6, 0.04)), ("x 1,000,000 ÷ 100 * 2", (8e4,), (4, 4e6, 4e4, 0.04)), # two statements, so two readings; a semicolon ends a run of operations @@ -1561,6 +1566,39 @@ def test_a_bare_number_takes_each_whole_factor_the_normalization_states( "times one", "multiplied by 1e6", "divided by 100", + "multiplied by 1e6", + "divided by 100", + "x\tby 1e6", + "x 1e6 / (one hundred)", + "£m×1e6", + "£m*1e6", + "£m÷100", + "(£m)×1e6", + "£m× 1e6", + "£m×(-1e6)", + "£m×one million", + "£m;x 1e6", + "£m:x 1e6", + "£m,x 1e6", + "£m=x 1e6", + "->x 1e6", + "(£m)x 1e6", + "[£m]x 1e6", + "£m;x one million", + "%/100", + "£m;/100", + "£m:/100", + "£m,/100", + "£m=/100", + "->/100", + "(£m)/100", + "[£m]/100", + "£m/100", + "per cent/100", + "per cent/ 100", + "per cent/(100)", + "per cent/(£100)", + "per cent/.01", "x 1,000,000 / 10^", ], ) @@ -1987,7 +2025,38 @@ def test_rule_4_holds_whatever_the_title_case(): ("paragraph 1.7; from: Figure 1.A", True), ("paragraph 1.7; column 3 of (Figure 1.A)", True), ("paragraph 1.7; column 3 of: Figure 1.A", True), + # brackets and quotes in a row, and before a word + ('paragraph 1.7; read from ("Figure 1.A")', True), + ("paragraph 1.7; read off ([Figure 1.A])", True), + ('paragraph 1.7; read off ( "Figure 1.A")', True), + ('paragraph 1.7; read from ("the Figure 1.A")', True), + ("paragraph 1.7; read from ( the Figure 1.A", True), + ("paragraph 1.7; read from [the Figure 1.A]", True), + ('paragraph 1.7; read from "the Figure 1.A"', True), + ("paragraph 1.7; read from 'the Figure 1.A'", True), + ("paragraph 1.7; read from “the Figure 1.A”", True), + ("paragraph 1.7; read from ‘the Figure 1.A’", True), + # a word that holds an apostrophe or a dot, or ends a bracket + ("paragraph 1.7; taken from ... Figure 1.A", True), + ("paragraph 1.7; taken from p.3 of Figure 1.A", True), + ("paragraph 1.7; taken from HMT's Figure 1.A", True), + ("paragraph 1.7; taken from HM Treasury’s Figure 1.A", True), + ("paragraph 1.7; read from (HMT) Figure 1.A", True), + ("paragraph 1.7; read from [HMT] Figure 1.A", True), # and one negation still cancels the phrase + ('paragraph 1.7; not read from ("Figure 1.A")', False), + ("paragraph 1.7; not read off ([Figure 1.A])", False), + ("paragraph 1.7; not read from [the Figure 1.A]", False), + ('paragraph 1.7; not read from "the Figure 1.A"', False), + ("paragraph 1.7; not read from 'the Figure 1.A'", False), + ("paragraph 1.7; not read from “the Figure 1.A”", False), + ("paragraph 1.7; not read from ‘the Figure 1.A’", False), + ("paragraph 1.7; not taken from HMT's Figure 1.A", False), + ("paragraph 1.7; not taken from HM Treasury’s Figure 1.A", False), + ("paragraph 1.7; not read from (HMT) Figure 1.A", False), + ("paragraph 1.7; not read from [HMT] Figure 1.A", False), + # a comma ends the words before a figure + ("paragraph 1.7; figures from Table 2.C, Figure 1.A data label", False), ("paragraph 1.7; not read from [Figure 1.A]", False), ("paragraph 1.7; not read off: Fig. 1.A", False), ("paragraph 1.7; no bar reading from: Figure 1.A", False), @@ -2537,7 +2606,25 @@ def test_rule_4_reads_each_dash_as_a_space(dash): @pytest.mark.parametrize( - "op", ["x", "×", "*", "/", "÷", "times", "multiplied by", "divided by"] + "op", + [ + "x", + "×", + "*", + "/", + "÷", + "times", + "multiplied by", + "multiply by", + "multiplying by", + "divided by", + "divide by", + "dividing by", + "scaled by", + "scale by", + "scaling by", + "x by", + ], ) @pytest.mark.parametrize( "operand", @@ -2546,6 +2633,10 @@ def test_rule_4_reads_each_dash_as_a_space(dash): "( 1e6)", "(.5)", "[1e6]", + "'1e6'", + '"1e6"', + "‘1e6’", + "“1e6”", "+1e6", "-1e6", "−1e6", @@ -2562,6 +2653,17 @@ def test_rule_4_reads_each_dash_as_a_space(dash): "one", "twenty-five", "one million", + "(one million)", + "(-one million)", + "(a million)", + "[one million]", + "'one million'", + '"one million"', + "‘one million’", + "“one million”", + "-one million", + "+one million", + "£one million", "a million", "a hundred", "hundred", @@ -2573,8 +2675,9 @@ def test_rule_4_reads_each_dash_as_a_space(dash): ], ) def test_an_operand_the_checker_cant_read_is_a_problem_at_any_value(op, operand): - """Sol's review of 82068821: the checker never scales by 1, or by part of a - chain, past an operation on a bracket, a sign, a currency sign or words.""" + """Sol's review of 82068821 and both of 4707e917: the checker never scales by 1, + or by part of a chain, past an operation on a bracket, a quote, a sign, a currency + sign or number words, or an operation in words.""" notes = [ f"{op} {operand}", f"as printed; {op} {operand}", @@ -2590,6 +2693,25 @@ def test_an_operand_the_checker_cant_read_is_a_problem_at_any_value(op, operand) assert any("a factor in normalization" in p for p in found), (note, value) +def test_a_long_run_of_quotes_stays_quick_to_read(): + """If a quote could both open a word and stand as a bracket, the reading's time + would grow with about the fifth power of the locator's length.""" + started = time.monotonic() + assert not reads_a_bar("paragraph 1.7; read from " + "' " * 120) + assert time.monotonic() - started < 2 + + +@pytest.mark.parametrize( + ("raw", "value"), + [("4m", 4e6), ("1 in 4", 0.25), ("a half", 0.5), ("nil", 0), ("4%", 0.04)], +) +def test_an_unread_factor_is_a_problem_whatever_the_number_owns(raw, value): + """Sol's review of 4707e917: the checker reports a factor it can't read before it + looks at the number's own multiplier, its share or its nil.""" + found = problems(_share(raw, value, "times 1,000")) + assert any("a factor in normalization" in p for p in found), (raw, found) + + @pytest.mark.parametrize( "note", [ @@ -2616,6 +2738,20 @@ def test_an_operand_the_checker_cant_read_is_a_problem_at_any_value(op, operand) "times onerous", "divided by region", "multiplied by the rate", + "scaled by region", + "multiply by the rate", + "x by y", + "printed fraction two-thirds -> share 2/3", + "£m×rate", + "£m*rate", + "x**2", + "2026/27", + "2/3", + "and/or", + "a;x y", + "(b)/c", + "tax 5", + "lax: y", ], ) def test_an_ordinary_note_states_no_factor(note): From 05265c2dc23394e2fa18894fee5eac79c250414a Mon Sep 17 00:00:00 2001 From: Max Ghenis Date: Thu, 5 Nov 2026 08:39:15 -0500 Subject: [PATCH 52/52] Report fractions as operands and more spellings in factors and rule 4 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit GPT-6.1 Sol approved 2f60cdb8 with four low findings, and a cloud session requested changes with low findings only. - The checker's `_UNREAD` now reports a fraction as an operand ("times half", "/ a quarter", "x 1e6 / half"), a "by" after an operator ("× by 1e6"), "times by 1,000", and an operator right after an arrow ("£m→x 1e6", "£m→/100"). - Rule 4 now reads a word before a figure that ends in closing quotes or several brackets (`read off the "Impact on households" Figure 1.A`, `read from "HMT" Figure 1.A`, "read from ((HMT)) Figure 1.A"). A test holds reading time flat on long runs of quotes and brackets. - New tests cover each form and its negation, and "two and a half" and "half a million" under an unread factor. - The README lists "x half" among the forms the checker reports, says that rule 4 reads the locator, and states that the checker counts "figures" where it means numbers. No staged row changed, and the checker reports 0 problems on 5,952 rows in 12 families. Co-Authored-By: Claude Opus 5.5 --- pipeline/ab2026_harvest_contract.py | 20 +++++---- .../harvest-uk-ab2026-2026-10-29/README.md | 6 +-- tests/test_ab2026_harvest_contract.py | 41 ++++++++++++++++++- 3 files changed, 53 insertions(+), 14 deletions(-) diff --git a/pipeline/ab2026_harvest_contract.py b/pipeline/ab2026_harvest_contract.py index a615bf1ef..10c8b77e7 100644 --- a/pipeline/ab2026_harvest_contract.py +++ b/pipeline/ab2026_harvest_contract.py @@ -450,19 +450,20 @@ def _read(raw, normalization=""): _CONTINUES = re.compile(rf"\s+(?:[0-9]|(?:{_either(MULTIPLIERS)})\b)") # An operation on what the checker can't read as a number: a sign, bracket, quote # or currency sign first ("x (1e6)", "x +1e6", "x '1e6'", "x £100"), or number words -# ("x one million", "x (a million)"). Also an operation in words ("times 1,000", -# "multiply by 1e6", "divided by 100", "scaled by 1e6", "x by 1e6"), and an operator -# with no space before it ("£m×1e6", "£m;x 1e6", "%/100", "£m/100"). +# ("x one million", "x (a million)", "x half", "/ a quarter"). Also an operation in words ("times 1,000", +# "times by 1,000", "multiply by 1e6", "divided by 100", "scaled by 1e6", "x by 1e6", +# "× by 1e6"), and an operator with no space before it ("£m×1e6", "£m;x 1e6", +# "£m→x 1e6", "%/100", "£m/100"). _OPENERS = rf"[{_DASHES}+(\[£$€'\"‘“]\s*" -_NUMBER_WORDS = rf"(?:{_WORDS}|hundred|{_LARGE})\b" +_NUMBER_WORDS = rf"(?:{_WORDS}|hundred|{_LARGE}|half|{_PART})\b" _ANY_OPERAND = rf"(?:{_OPENERS})*(?:\.?[0-9]|{_NUMBER_WORDS})" _UNREAD = re.compile( r"(?:^|(?<=[\s(\[]))(?:" rf"[x×*/÷]\s*(?:(?:{_OPENERS})+\.?[0-9]|(?:{_OPENERS})*{_NUMBER_WORDS})" - rf"|(?:times|(?:multipl|divid|scal)\w*\s+by|x\s+by)\s+{_ANY_OPERAND}" + rf"|(?:times(?:\s+by)?|(?:multipl|divid|scal)\w*\s+by|[x×*/÷]\s+by)\s+{_ANY_OPERAND}" r")" rf"|(?<=[^\s(\[*])[×*÷]\s*{_ANY_OPERAND}" - rf"|(?<=[;:,=>%)\]])[x/]\s*{_ANY_OPERAND}" + rf"|(?<=[;:,=>%)\]→])[x/]\s*{_ANY_OPERAND}" rf"|(?<=[a-z])/\s*(?:{_OPENERS})*\.?[0-9]" ) @@ -609,7 +610,7 @@ def value_problems(r): _FIGS = r"fig(?:ure)?s?(?![a-z])" _OFF = r"(?:off|from)" _OPEN = r"""[\[("'“‘]""" -_UP_TO = rf"(?:(?:{_OPEN} *)*[\w.][\w.'’]*[)\]]? ){{0,4}}?(?:{_OPEN} *)*" +_UP_TO = rf"(?:(?:{_OPEN} *)*[\w.][\w.'’]*[)\]\"”]* ){{0,4}}?(?:{_OPEN} *)*" _AFTER = rf"(?::? |:|(?={_OPEN}))" # A verb's whole phrase comes first, so that a negation before the verb cancels # it: "read it off the decile 3 bar", "reading of the chart", "visual estimate of @@ -617,8 +618,9 @@ def value_problems(r): # Figure 1.A"), any number from a figure or a figure's column, and any reading by # eye. A column of a table prints its number, so a column counts only in a figure. # A colon, brackets or quotes may come before the figure: "read off: Fig. 1.A", "read -# from ("Figure 1.A")". A word there may hold an apostrophe or end a bracket: "taken -# from HMT's Figure 1.A", "read from (HMT) Figure 1.A". +# from ("Figure 1.A")". A word there may hold an apostrophe or end a bracket or double +# quote: "taken from HMT's Figure 1.A", "read from (HMT) Figure 1.A", "read off the +# "Impact on households" Figure 1.A". BAR_READING = re.compile( r"\b(?:visual )?(?:(?:read|estimat|tak|measur)\w*|took)(?: [\w.]+){0,3}?" rf" (?:off|from|on|in|using|of){_AFTER}{_UP_TO}(?:{_PICTURE}\b|{_FIGS})" diff --git a/sources/harvest-uk-ab2026-2026-10-29/README.md b/sources/harvest-uk-ab2026-2026-10-29/README.md index 1f6fb7dfd..d3868bce5 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/README.md +++ b/sources/harvest-uk-ab2026-2026-10-29/README.md @@ -70,12 +70,12 @@ We follow the AB2025 harvest README's rules 1–10 and its row contract, with th - a whole number and a fraction, with the whole number in digits or in words up to nine hundred and ninety-nine. A percentage word may follow, and so may a multiplier other than "hundred" after a space or hyphen: "two and a half", "one and a half million", "2 and a half bn", "one hundred and a half"; - a ratio of two numbers, in digits or number words, with at most one multiplier each: "1 in 4", "1 in 2 million". The checker takes "in", "in every", "out of" or "out of every" between them, with spaces or hyphens throughout: "3 out of 10", "one-in-every-ten"; - a range: two figures joined by a dash, with no spaces or a space on each side ("£4-5 million", "4 – 5%"), or two numbers joined by "to" ("four to five", "4bn to 5bn"); - - "nil", "none", "nothing", "no change", "no impact" or "no overall impact", as zero. It also reads a dash as zero where we say, in one whole clause of `normalization`, that the producer prints it for nil. It accepts forms such as "dash for nil" and "the table prints nil as a dash". A comma, semicolon, full stop or bracket ends a clause. + - "nil", "none", "nothing", "no change", "no impact" or "no overall impact", as zero. It also reads a dash as zero where we say, in one whole clause of `normalization`, that the producer prints it for nil. It accepts forms such as "dash for nil" and "the table prints nil as a dash". It ends a clause at a comma, semicolon, full stop or bracket. - It also reads a qualifier before the number ("about", "up to" and the others in `QUALIFIERS`), "a" before digits ("less than a 0.1 percentage point"), and "a year" or "of GDP" after the number. It reads a ratio or worded fraction with its percentage in brackets, in either order, if the two agree to the nearest percentage point: "one third (33%)", "25% (1 in 4)". It matches `value` to either number there or in a range. It reports a range that joins a percentage to a multiplier ("4% to 2 million") or a fraction to a percentage ("a third to 40%"). It also reports a range whose first number exceeds its second, unless neither is positive: it reports "2025-26" and "2 to -3", and reads "-4 to -5". After a bare "in", it reports a year from 2010 to 2099 ("5 in 2027"). -- It takes each number's scale from its own word (bn, million, %). When only a range's second number has one, it gives that word to the first number too. When digits, number words or a whole number with a fraction have no word of their own, it uses the factor in `normalization`, failing that 1. It reads a factor whole ("x 1,000,000", "/ 100", "(x 10^6)"), multiplies operations in a row ("x 1,000,000 / 100" as 10,000), and takes either factor where a semicolon or a word separates two. It reads x, ×, *, / or ÷ as a factor wherever a number follows, at the note's start or after a space or bracket. It ignores a slash between digits ("2/3", "2026/27"), and we write "page 3 of 4" rather than "page 3 / 4". At any value, it reports "x 1 million", "x (1e6)", "x one million", "times 1,000", "multiply by 1,000,000" and "£m×1e6". It reads a percentage as a share or in percent units, and requires digits and number words to match `value` within four units in the last place. +- It takes each number's scale from its own word (bn, million, %). When only a range's second number has one, it gives that word to the first number too. When digits, number words or a whole number with a fraction have no word of their own, it uses the factor in `normalization`, failing that 1. It reads a factor whole ("x 1,000,000", "/ 100", "(x 10^6)"), multiplies operations in a row ("x 1,000,000 / 100" as 10,000), and takes either factor where a semicolon or a word separates two. It reads x, ×, *, / or ÷ as a factor wherever a number follows, at the note's start or after a space or bracket. It ignores a slash between digits ("2/3", "2026/27") but reads "page 3 / 4" as a factor of 0.25, so we write "page 3 of 4". At any value, it reports "x 1 million", "x (1e6)", "x one million", "x half", "times 1,000", "multiply by 1,000,000" and "£m×1e6". It reads a percentage as a share or in percent units, and requires digits and number words to match `value` within four units in the last place. - It reads a ratio or a worded fraction on its own as a share, and accepts a `value` within 0.1% of that share. On a `unit_concept: percent` row, it requires 100 times the share instead ("1 in 4" as 25). It won't match `value` to either of a ratio's numbers alone, and it reads "-1 in 4" and "1 in -4" as −0.25. It also allows 0.1% on a whole number with a fraction, and on a fraction of a multiplier or percentage. - It requires `value` to share `value_raw`'s sign unless we note in `normalization` that we took the sign from the wording. -- It treats an HM Treasury Impact on households row as a bar reading (rule 4) if the row has `value_kind: approx_chart_reading`. It does the same if we mention a bar, chart, graph, plot, diagram or visual in the locator, in any spacing and with or without hyphens. "Off" or "from" up to four words before a figure there counts too. So does a note that we read, took, estimated or measured the number using, on, in or of a figure, or read it by eye. It reads "Fig." and "Figs." as figures and looks past a colon, brackets or quotes before them. It counts "read off Fig. 1.A", "read off: Fig. 1.A" and "read from [Figure 1.A]". It counts "column 3 of Figure 1.A" and leaves "Table 2.1, column 3" alone. It ignores such a phrase right after one "no", "not", "never", "without", "cannot" or "n't" ("not a bar reading"). It counts the phrase after two negations in the same clause ("not possible without reading off the chart"). +- It treats an HM Treasury Impact on households row as a bar reading (rule 4) if the row has `value_kind: approx_chart_reading`. It does the same if we mention a bar, chart, graph, plot, diagram or visual in the locator, in any spacing and with or without hyphens. It also counts "off" or "from" up to four words before a figure there. It does the same where we write in the locator that we read, took, estimated or measured the number using, on, in or of a figure, or read it by eye. It reads "Fig.", "Figs." and "figures" as figures, including where "figures" means numbers ("from HMT's figures"), and looks past a colon, brackets or quotes before them. It counts "read off Fig. 1.A", "read off: Fig. 1.A", "read from [Figure 1.A]" and "column 3 of Figure 1.A", and leaves "Table 2.1, column 3" alone. It ignores such a phrase right after one "no", "not", "never", "without", "cannot" or "n't" ("not a bar reading"). It counts the phrase after two negations in the same clause ("not possible without reading off the chart"). - When we cite Figure 1.A or 1.B on those rows ("Fig. 1A", "Figure 1(a)", "Figures 1.A and 1.B"), it reads the locator up to the first comma or semicolon. Before that point we must name a paragraph, table or box by number, or a data label. It accepts "paragraph 1.7", "paragraphs 1.7–1.8", "Tables 2.B and 2.C", "Table 2.C and Figure 1.A", "Figure 1.A data label" and "data label for decile 3". It treats any other wording as a bar reading, including "Figure 1.A data label unavailable" and "Figure 1.A, decile 3 bar; see paragraph 1.7". - It reports a malformed row as a problem, never as an exception: a row that isn't an object, lacks a field, or has a field of the wrong type. It requires: - a string in `source`, `source_model`, `value_raw`, `normalization`, `quote`, `status`, `value_kind`, `time_basis`, `attribution`, `benchmark_class` and `parse_confidence`; diff --git a/tests/test_ab2026_harvest_contract.py b/tests/test_ab2026_harvest_contract.py index d856cd723..2439f430f 100644 --- a/tests/test_ab2026_harvest_contract.py +++ b/tests/test_ab2026_harvest_contract.py @@ -1582,6 +1582,7 @@ def test_a_bare_number_takes_each_whole_factor_the_normalization_states( "£m,x 1e6", "£m=x 1e6", "->x 1e6", + "£m→x 1e6", "(£m)x 1e6", "[£m]x 1e6", "£m;x one million", @@ -1591,6 +1592,7 @@ def test_a_bare_number_takes_each_whole_factor_the_normalization_states( "£m,/100", "£m=/100", "->/100", + "£m→/100", "(£m)/100", "[£m]/100", "£m/100", @@ -2055,6 +2057,21 @@ def test_rule_4_holds_whatever_the_title_case(): ("paragraph 1.7; not taken from HM Treasury’s Figure 1.A", False), ("paragraph 1.7; not read from (HMT) Figure 1.A", False), ("paragraph 1.7; not read from [HMT] Figure 1.A", False), + ('paragraph 1.7; not read from ("the Figure 1.A")', False), + ("paragraph 1.7; not read from ( the Figure 1.A", False), + # A word before the figure may end a double quote or several brackets. + ('paragraph 1.7; read off the "Impact on households" Figure 1.A', True), + ("paragraph 1.7; read off the “Impact on households” Figure 1.A", True), + ("paragraph 1.7; read from ((HMT)) Figure 1.A", True), + ('paragraph 1.7; read from "HMT" Figure 1.A', True), + ("paragraph 1.7; read from (“HMT”) Figure 1.A", True), + # Up to four words stand between "from" and the figure. + ("paragraph 1.7; read from a b c d Figure 1.A", True), + ("paragraph 1.7; read from a b c d e Figure 1.A", False), + ('paragraph 1.7; not read off the "Impact on households" Figure 1.A', False), + ('paragraph 1.7; not read from "HMT" Figure 1.A', False), + ("paragraph 1.7; not read from (“HMT”) Figure 1.A", False), + ("paragraph 1.7; not read from ((HMT)) Figure 1.A", False), # a comma ends the words before a figure ("paragraph 1.7; figures from Table 2.C, Figure 1.A data label", False), ("paragraph 1.7; not read from [Figure 1.A]", False), @@ -2614,6 +2631,8 @@ def test_rule_4_reads_each_dash_as_a_space(dash): "/", "÷", "times", + "times by", + "times by", "multiplied by", "multiply by", "multiplying by", @@ -2624,6 +2643,10 @@ def test_rule_4_reads_each_dash_as_a_space(dash): "scale by", "scaling by", "x by", + "× by", + "* by", + "/ by", + "÷ by", ], ) @pytest.mark.parametrize( @@ -2672,6 +2695,11 @@ def test_rule_4_reads_each_dash_as_a_space(dash): "billion", "trillion", "zero", + "half", + "a half", + "a quarter", + "another half", + "three-quarters", ], ) def test_an_operand_the_checker_cant_read_is_a_problem_at_any_value(op, operand): @@ -2697,13 +2725,22 @@ def test_a_long_run_of_quotes_stays_quick_to_read(): """If a quote could both open a word and stand as a bracket, the reading's time would grow with about the fifth power of the locator's length.""" started = time.monotonic() - assert not reads_a_bar("paragraph 1.7; read from " + "' " * 120) + for run in ("' ", '"a" ', 'a" ', "“a” ", "(a)) "): + assert not reads_a_bar("paragraph 1.7; read from " + run * 120), run assert time.monotonic() - started < 2 @pytest.mark.parametrize( ("raw", "value"), - [("4m", 4e6), ("1 in 4", 0.25), ("a half", 0.5), ("nil", 0), ("4%", 0.04)], + [ + ("4m", 4e6), + ("1 in 4", 0.25), + ("a half", 0.5), + ("nil", 0), + ("4%", 0.04), + ("two and a half", 2.5), + ("half a million", 5e5), + ], ) def test_an_unread_factor_is_a_problem_whatever_the_number_owns(raw, value): """Sol's review of 4707e917: the checker reports a factor it can't read before it

O_%e2--RFe}KRZzb7m_|Ik z25>2{F&q=2)7gpVI&K*>GlTU8TTz5>JfPjk7E$6IxY=1a-DmNI53*7*My2zSE2ToR zTYtCo=Pt3nv9H$*_B*wzq7oEbr}uYTJ673JAM?%s!;7_LHuMPaW%f)nSM!0-w2KP<(9>W17U zJlb6H8y;13mwM-f58zB!8`k+F(sefHZ(o()vpK0gnU9yh(bRFp-_h{1%QZ@=xRjmU z04f9#g!CAy9*HChPs+S3t}t;qiwhd1&gwc+`5{!95ywC@K$8%OvUGsIl5?v>tBM!dyCGy{YeLe}`tyqufQGCHU2cHomocLUcxIk0rT#$CEn8fBOn^-te*OudmMi-@IeQbd-U&mkWr@L zECO^VX4LBIcXECH_YC4Eic9_`nc=r5}N8Q#VJyKj#@Iz4>r(b}g9R4fx4J ze@&14WQvHm<#-e**4p~ZmJtxgHsg7#1Kbv;qHiZrZ1qrE{KH50z3{mC*>0(KPT?tJ zp7$VWOgy%|Pq8uGSFGJ&DTD31bj$VOwQx(?v30it(%Jbah$RwY5Ln$d(v7z%)RJhk ziY+q|m{okVxz^FOaF0!CD&Tt7+%~!u;U;nQH|yB#c7OEXh0!={sVt|p@zAc~BcQEo zILmCb<*401kjjU%hW^f(?q1`V-5m}A1P}I7GU4Mh&^%f)CcPyl<$Kck$&kjmoEs5) z##T_HZ8sAw=L%G-!|goy<+C?2*7lyVR}YB2^zkVMc4tQjIQMIG(A&o;1I~TQ8Py>taQqAXw6WkZ zkwAqwc7z@GZJbp1x~iX|6@tQ<;F0szzzpfYLg!K6#F&zKw&Ccbf@kdVa-_$<2)MRb zex)&)`(dF6yMi4M`@J!|5vW>*ROwssdP=F)h08Ugy+VDzsB6wrhy9Z3W5@cu`?Yd| ztLE1TqOIu`nZS$9i$Awp3unVynEMvLWm-Q4x_zYo6!ZC5pk&sPu49SP$Lv$}_0+jQ zMSDPt#8x1lJzegMSp#qF^>L3ru;aK#vhuLspVseRbsWqughMKqH{d)A!dUjH$sq9f zqWWdr*%CL)t@ujzolwA%h>k4vR%G$hT>)8!-a;<=cE< z%s77^M+aQOC?8I*ajwLC*N|^hSYM!f&M(*I!$6Nd9TRYw{S@;5&a3~gInRI1sd4_E z4uBA;l5mj^XL2F+#so)fR`(n&9|Br1n^@wogCX;ez3_zsK0WWI#rOMaeEIRXI<0sbDMkismcvWqM8``YdoeJi#z0tUR0SNF<57? zRkPCPzF>I%3j5-7YtqlG6MVvodb*352yIOro0PLYYBjc(+*MGyFe=Dm1nyywKTf0A z7ACo?i^L%!z94&BQ?i_#H?Kf|&zOJFe^x$~bu(xAq30{(Lw1EysDFD`&_gTsMzOV% zXRa=Shfa?RLb;)cN=->-3WC)6+0IrU21^O>p;P{LIG={he}+PIQl^lA$lX0)jWJA@xk3B#6c>j%` zI(~6IwtuxJhvI=;D(vLUu*sM$Hq1iEH?4V@=QCN~$n8xxH)4Ng zDJ;uv;Oy5#!vJI4SW?QB)hE_o+LXPTXQzDJm;G_c$F`VKqaSRLmcbpG@Aem2;hFIg z<9hN&IO7k2wx842wyWgwtGv?W_-99p}Mw=luV}0+4$Ni{1joCTpVlX^6gOT5D z%3q-KVp$#I*Q`N1(WGK1_Q@~FWv;O(%Vsdo9S3-QxOjJ3bx{KMjDe@{n8(0U}_ zz|2P%iXGpPZ}6ly=+%Z_jSMwak$Cfnv*}sHI#T$$j2Tz6l47RViuAd&_$?AE$=a0|2qhFD%Nl(mEq;sX-US==& z$Ge|H{+!QMLqc~>YW{hwVqICM-?|__5rcY1cY3WG{@lnnnwX_(m5WAHWl>D!RK{*DG7CP#b=S0`YUR5B37wTTor7;0Ch1*_cr?TQ+e*H!#QYt_kg0tkkkFF^Hptj zezM%A@jyVwx!?php}GoDY8Sz3sX7!ADF@*wgt;TR{nPnA&CX3)+MI~CI>(Gz=Zhon z2F5b5dgV@na((-ta5*ojruO=-gygfoyu|#C zmS_Cx_@DFWDY#4(W?$AIS@z4P5ag$O`L7eBy=P)JFS;8ZoyPnyghlY482GLZ%T=tZ zw{3g0O8ZlM`kd9#O1crjoRyVQ*WL5&E;137?WXtO)!W%mLi%BY=0@BH*%mT+pL7*_ z=Jn<0)e{Y2eP-)HZq$6bgah}ILTN47GB6HWM{4=onG;AH#C4kqhCO5N~`!%PWP!g=bANB7zB}ZYs&~c z86zojc0JBfhBYp3Tinc1#cIIYPr=|qyNe6I%kUfMMO2DtmxG%6_&$si7@rPaa69S? z8Vffd*Dn8g9S(bHVcH`u2}9{=pwV^r!dy?_b$5%tqFr zMfBD8=;~-2!MpQzpISQo>5ggR^y`a&_uc!*{#BZrt-scuzVD>Qcr-#LXrKosx{-gQQ*E_ zSVjmY$+1mdkZux8Fi!1NV_1q*$tC9U=NbY_m*6M!M^&9y_CA{sd}ephCCn8{pxcTd z1Rua`q~~xaVf$CsCPdu6+;II84@vO4~7S}p1WZiel?rp;2n0MMzqLVr})ezjb zK}`rJ_OfD?q)U zrzO)ITXe8=Q+^B+K=nc^Y1CJRjLj(7YVRznifE(x>5%?fHfkN-WHR$hLZxvupB8)Y zg6);O-nN?N_+A8jFB%ee z+nOFjS0KItTJ&GGu7+LoAWaOo`)To@uf(7slPp1>>Gs646hX#v^qrZHUiU3VciaK_ z3uj+lut&T0C$Ssajb+4X??rF7r!z*k1GMmVKfzyWiBav!5HOaa7{wS%As98XCQCW3 z#z&&|kV_DjV{`c@t^tmtRk1aHRf|3J#nm+Anf>9>TmP}+>a)FtXez3;fB-9_Xcgji zNixj+xh#;2Y0mkXkGqXq(!9)G=lv-RIWFEqRH{%!Qm!5IwqWRJD#&Lwx? z!y-SsIJegaIW3Z|9rww1Az3f_20v2w_BYR7bfdDe?iF@kq_4*2+zIGNQKhZkWq-}N zlg6DA!vsfK(+7RM!?g?Kx^o0THlkvhQ$}AIe4UP4KLeG5$(tJw_Psn zzb`Ew2QEPd4|p}jC@*Y$%0fvBBOn3Q*T+*Ukn@QfiMu<4x7qM|hzt6yGl15%k1vg% ztT`Os2E}+S&Bbrc24*c^=zia0-2C*Z&GbpPxYpguIn{+x_6*5*kLR(8Y$dEboCp|@ zmW4X12mRaamR-&r2sg@l|ZD+?Jwk#OKA$ zpwnHo^QR8wqrJg{NQgRKDKZS@&(YX1b8uU+o0h;i?_E%MWaB*H@@$RSbX?s z)r`75=0&E%rgr%+BF<{37r^e0;ZOS!!{Sy={+t)!UWl;qF`ulY=HVyILuCKbP0@(2 zQHG=gjLkZfXqe+b{9L2;QnUNXhm%&9noh)qc+|@P;VDx1lF|L{sf6>+hY$emNc6;| z`nby|hTJ8(pte^Rr>dv`*Tn3 z~Nl+?V*3Q?UnD^<~h(&nsLMag7?@{ zG^fwQkoetHBBx!;dy7-AR7b*ZJA)_8>l!VExNZ{kfA(wbSrE#$e z&Mj^{(CvM9t=i4-WE|@wtCD5 zd$)g(>Xs!XO~#gM-*glQbZJyz^iR1OpUO2u?{P{5?aacPe)&;D7;)BCG;qkdvWjCeTrGE!6+16|?&W68Wf0lXn z#|Jp~`;+)oz~5WXKt%9$wAMdY{A&?>NjQ-`M6y|#4HYGM5L!FqaKM-4W00$TwywDJ zcB0w-N$E+0jk^e$zLWml*R{{GEsv{*4obF`UNP;zBfB`SUJ%iE8xcgiyt*L4^35TA zy&O-o4DS=jioJ)aict2b`fS+9{K8G(hD1P$7Q6ynQ0~!%)#TI)>ftj-&b@q@NU#ka zfwnBBF^KhVSCTWeYMcg|iqL^&iV(AX?W%9}lZ`OLFQ*N-Qn)CB35x z1e++mbZK*A`(!~T6kE7nx*pgh^h9m(XgKzm8v(P4t}1 z#fF%lHAmj&HkCv-YnjMQ(120fx=J7H?9&HEmm4>lKev3>}t~Kg-Y5Z?3MGIMac!useMm* zIxkUJqZ_B*Tt?IBD?Q#w{Z{GGEQ&0Mp^R5m{8Q^5;Ug2I7YeFFJ4Y7wFRe68+jOqb z1myJDOY(581TOyKM;V6xk21MAe;p9iK7~-sMUC1aRBIAohS+^AlHZgWp7c*ETqj+? zgdg8B))FbcX$;?qdL-9Uv-|qVO}~}osGj;*c(NN5p27mzDsY=LbWfFDFf;}^C3s8T zH%3%GZA4F*?(J$*@E*^w7tV|xQ}eE$dK0wF=LqVP8IxQjI;Ts!80#_c$c07jXW1^x z3rA3e34Q4SVL_R^kI23<9BJDfo-L%wJ&2t7-fV9rD;dY)BP&V95&+Uy3TgY9togU> z^9oHU2pjT=-fLP`T6AQ&kGFhonzP0JQTsdc#y@YY@K<#mvZhE)Yk6C+-ckWgcizou z%>TIe6JKu;o=j(lftn~J&OFD@4o2P?ADhzuW&pN;m>dB|6!=E*S0%5?4qFnJB}ds{ zHi?dl2vvU+F7$h`RsDINor0eisgF-yOYmt4`v!z!_cKm5s5|lbZcM6GE%WINQK33- zIgLaWMXbkRB0`)AzdR2AjMP-V_hc^&^lK})51x~+9gU4qAjl&R+l2^xp}ELe0`RAi_k5WU^&IcP)PT{4Yo8`;6>2kMP!X zJ>amDYZT_qEhi;-&;KH;IgbW>pBg@$1-IB^x41oSycu%)EMSUKHl;2%Z;>=! zqfj*oIvI*t>Pm|BKtfL^?bD=(S}IP4Y9BU#$R#D_?`cuVsz%^Y8IhJkWjaJt|@ zkT#_f)_zHi`5A%@926Rv&^9;n#;Z4>_-hZy2$+gEwqR+c37H~hD+~#lVFcW?HltO2 zaIhYHN`B0&OvSC!*XxV_$b(bDV$j%#^zj0z%gLcoZHYny1NJ>i(+rhR5LBxw3^XOM z+AR6Szxmjt(+8D&M5rt0w*o=A<_>$7k`O3&sl8Hr(_h*IS3x?wiIBK#&?A3h1$x!v zJ5Xl6>icMp;#Xt#zCFo17WYNjB+byq`Q1Wm(5RM|(BWTw7%uJt1u4-2g}DFM1Hx=X zmp6(vO^SEUPft&umHp+*39qIy4PY7X1i>;1JG8Hu&!WkYv^NHb6jxj`M(n%}?o!Vx zYmm*NNPPnnCBOgBCC-!A@Ek=^?`Fsw$9!3%$09dS)?q9_H3Azd>sYk52EBA0{2SLo zwhGBnm4Yf~aOoR}?rS)e!BrY)Hxh`J#pN0v6k`l>HK8xtLo24k3$2)DB>rl0)xL<| z_raO1CbVKc(kMHZYx5{JlvRY|22^swJWYbyC>qLeualm_p7zn?3)4b-GZ-NZx*$D5 zp`rKGO3L%6NhQ3_%3ifn+s{`#DGG6>P|Sw{-JUodltLz$6{z$wUjY+>HKl*lbfQf@ zyg!D?ee^SRt67gs3)4GFb2J8EYreE_Axo{0KwjCinE929fQE{4ZTsg zU2AU0Kps~*QcuvUP`~Z6+*OG0qa})~#_U7;pCze9EB89f;pqsmzMH~8P^=^9Ev`)S zlo%Ox0LYv#0EBT6*v9+0j&KFlVZ36Hj}RVe6o6Jg+uVsPdGFUd%2lqZwB=(tDC3fX zd^QiqiF9;Z1oXq;C{ykZU%P`8`Wn!TkBYcZrotly+Y)tuYuX9ZHxSE#I@8|2ogZ=5 zu^QzLr_smf4z0U>y!>97v8SoR)!_bG{qlJXC$xNW(egQXkk19-JY4$VK|U8Hcx6XF z0hR^{ReIE8O`-?7w7+8A2iBcgWcAB;v0jvdDop9i&IxnNIzC$Wr&_Rwc|~QXO@67E zE*bl1ZDhv=S&mfLx+Hr=rtmS0LilGi|4zXl!efNoN9$KR6kg%5Y0ZRlL+wB$??Tj> zeUeh!nLQ12G6->?uIPuVI3*M|dF>~{H0_yGL&aCBt28s9oC1m+cbDAeNlT5a$!(GB zIk4s`$M~r*#EVf!v!VOe_%Kw+qH%2inwT>%i)gWjU=L2|xPMJgP<=Lp2C$N7gCZI9 zhPxaMAjmYfgVRP8&|OMr^n0=ou7@uJ*mMkjyB2h``-S?+3WYFT{I6y#>9jKwlACGm zVomk6pAPIXuBQA4=#X75$#CVkrb`Ei0p;?=faDGW`*?Hf2-i?@CJTOhF{fSjoSrU! z${{np)<#06b4&VFUyrho&-&1~?YqtZY+!H^Q#&+nYbmrbpk;4u1MeBz8B=wmw38u` z4=qt22J}pOF%|r9HiH|@4|UMT=kS)@Ah@SZSoN~%MQOHB8Rdp9ZNbB8eQ8n2869o< z-#-MiA)skc?J#Zm$K|N`M<6FYb3IR%XQf!U@F6+E5DH?w2!{$RUtR(I#!Bp;eyMkg zW^+;>UNdN!{xsh(gc8E&37emf%ygYg;vHl_1z~KEN~>&A(?)iBy@OI+A6%28N9zSGf+Fe3GZZ*?cjyiP8c}JYF7`qm7 zQSy9W!5`-UwQ67fd55qD)IFIlwF;XorFY27u5Xh$&}EHYwyN?@T;r;Wx9DY72tdsvg?m6G(d8J^g*~7> zky2tJraJ9#>x6U7UG`qkXsO=!keqho*x$i_cS$ePE!wawh%(CMe%ghttP(Zzgq2w{KI&1TQ2SOE#Pql$tr?S`>&jM z(mwMBYy48s^snrY+CK3qS?ZY~fk?RhD5#D748=x~;nMbcgWwDmti%xn91}k?MqEe9 zuN9&|Y~knS<1l7sYlvYgA17@brKef!nLhULs@mhZe%Lcig-$D=@$eC-p zhNv<6WIX0=aj-z$CAH^`mwwOux*WZz#XjcM1<#jo*HT0Uo!Z0C`7M_KF5N#~ zd+X6|L^UHa8=2{1JugUB#aRasaRe6?MxBi(5PWZ5El-JNR8#BhF7jh>vUQbVrHQc% zN<;C;bG@4O=`l>|@ z!Nc#R_(6n5E%7(uw28hOiWTIo(paH0scFB5IIg1^)TJ$r#U8M7Vv#^!OIY$Kfk;g@ zBFj1!2$euDJxpw@^l2++*t!h|UzZ^RP`Liz<=1mcUO&PY)U)0XqbB-L+D%~@gcA8u zY3s-}vs`TY-Kx|zoLrZy@0;Fy45tBPy&mw2xhpy>O!z8#bf<=j2&1esTG=uyaWB2= zg)%ZxJCm2(@*Qui>J;B8ZQ?ibygqwpD2;t-I7;7q_k(>YWr zBacR~dQ6b>cQDN!(`9B}AC@Iiab?HPkJI`uJ&UKTR4XZ7zjjJ*a0lwq>cl}CzR^Qo zt#`k$Xz`jE_zmW>?{Njhde*H8A@RYbmccJa^b@?Kg8@Y3CgU+8`$k5bk~TGY0)JHe zxo>}iOI0q)zO#N^!pr_b4CZsC1oL%Vm;x!yjUoUJ(!MCaR6~5BQhjduCiFjRt@<*1 zl3s*p;t~R9lC)CAabqpVQJIx=zL1s%fcj%FA41AKXFzQac3}zv_a()qR~AWFWmW#h zu;|0`k~h&36S53yBFd`#xIVAxXI-0M@u#)^^??0|zEg!ajU0#Fv&PP76vMmK-e4-; zv1pumyX8)J<*4F&88 zIG=aL-~jdBgZZYhh|(*=Q|gXlrY5--$``XKb&O+B1HUC{wikQ7vPo<7`_cP5E6sO( zAMsO?Hsw+}$Q6q;0q4~mYNuNr2Kmc+?L!e$iD`3&rQ$-il^xLW0#e9U49oZIv)9$` zbszXs9o3>mV%g22QwYfoOWE^EU2$n_a^a{Ce#gFU{HZiE9c+@;2y@PxH|jRULj0-r zR?5<&%>K%Bc=-I3{}U4>5n4>qg}7=F6G;V!Y3hbX-Nvz$>LM&#>9`5}TVIkwsNpY* z9w5-mp(biUJduk*0n?;*86}2aJLJdMK%EOYjLdhECmlV*3?Hz_=c;lccm+_13bPj8 zjz;X%kP}@Y4ayio{!25+Rd7q6u8{(xB= zEm4UZSAqn%+%$p=EmNRJg&f%7wVqJwB7Vrn{1!`ae^mn|JcE8`NA>6=x4=$?)RO+$ zK676%e2uI}KX*I$LupWvW+oEqs@arKqruR@v)60?sBEHRt4c(Z*pN?)(vUCr!jzv< z%P4|U9Z{>Vq3JiK=Wx3Irqz%Sl}zdIbeO@gR4I4-13!Uv?E7(X91}RDI^lyuA*l>C z`kFopJRBCa!r?!4i+U$xZ7Vq>)*aB~&+Gk<{5~bR|NoB}jy;adbZm}I_Uzc4MCp*x z;SkxI#6ie7WOYJT63P+B2sz5$qq0S^vL!M?d~dx!>-Tv4{yG1gm+OAruj}<3*Kys~ z?WSs~V_a7B3AHBV^m^B!w&lG25C6CQTY&ao$A|&#&zx1d3*)hJ!u}kiy7io+EVaWo zM|a<24tSwDp)!Lqf}ok}?-<#9v!0+wTD`_A>42m^LMP(`6iO+PTD1^lRP?o zMPU)S*-L01Sz1Jr83dFFD5ElKWj`CeWPjqQ5Y_K8;uoJ42lmB3oOX}p24f}oyGR|G z#M2zOcGK-kxiH@P?ltHiF#!S| zPJyuD&a~j{tNT80;1QJopA<-za!nGnW@0vxWWHG{J+0zyL|GhOesqGs8R?OuCF#L3 z+46Xl*K)2a>XP`-V_xms7I#3+d5({##a)Pk4&sO+CAA0$-Pk zjWWc*_91s4_Fgh(kO1ku$EWeoADUPNKQQYU$av7*AxaXYq7wP}2}Uk-%plch3gcA=qH1rLySzXmiMZYPf%?tbK^44X47ld#i|{N_R7jDhhc8ykD6D_6Nc z2@LFiBc=dR^wql03EwAK718b{h&_~AM5Z|Myx|5{zR^xr*fMGyRBcIqvcIT%aTlm= z`G$$U75azCc{Bbc?akF1Zh*9bQ09qxMAeP7rppzr8@A86%h*VPNERgolP~`J+gCt| zk;xRr7)hb7`RmCYFawnm?8nccc9zK<<=uop{h9B5gu;4YV!H>NyMeOm_`;;$@+yx7 z*F7z&Rlx@pLDzE@dVvI{k8?ZI!hR_BviRcL>PNxKkVT|v55Wu!k9X zy5lRc;*wqW&;jZ^s~Z|~N|#Fs$<%vh7)R}~pH%l?(*D{CW5LI_#6D3wyC$*!@!w>> z>X;z%E}2RTlDVvs;1%{V&p1Vww^$|~YKIv~Kz%|duEf>etC znmR6E>L!wasq0Hl@{YAKmdbc>e-WAVF)&-}ZJaC<&jxbyI$I!Yg0QoJkf%wK3lIOq}J_p7@HUJC< z02mPfFp%ngH@l3CJtmA+UosEg0;YpZxOW)o`z4&kr#0!^&(3v5ag~WGLX09QrjgHX zcZxM@Q+aK8o5!U?%LS_I+MinqTfl3qr?T=0o##YZmy zdqn?l+Q%#+>G}ZP?^r}GY<^8PP#ybx|3&mqBjDJ!drOcXfG73(+3C}g9FX#C03cpM z%m(s~yLup=V`H3#H-!C8#^1>EMgxTQxb347t!+7fAE+x)wdy38)%s{K# z)#WC9f9le0Wl8`SXBUAE7O-lT1UOAE4R$98WjNMs!>`&l<>o=W`L(J-KW&uCLzsGE zgKWgd*Bf1X;igj6x9(m)G)X<1Qq}zxU{kbA>`NbX6InHSJ)Ky=k5RqcBf{M2>Lyl$ zR;aQ`uI{##7B-I-1SbMTR~>ORJ}ys2-lhnG zvE^N93SWO61UM%Pbl}!WiwK7%2inX(hWW$bcWy~ zmu29}=r%s*%`vgTGzb2%z5r4@I{boORTd4nwwTfH$SnUeII?{txYyZ=#i1G?y zD&2;_j3H0IiAb+DPvf(A##_d8sAt%{!SdNB`lI8`So;-10ifaP5O3Q^Ubku~50#}R zhq5A>!BG2r-m=`L)kdqslcZG<9l#r5bMx7HUl^aj zfaBG;F-DpnpQ3d9uCFA{O`Qer4Q%&-GbU}oXK!V5uN%2+a#stdeHHSlJ{?hCG5dUI zit!7|%arAR@E)Ufw)AVXqk6q8yz~dzg+G03>xrk3mf*&@jicWK+YJ+0%9c`tcD6(! zX2nZ8dVD(CO~T}cRBMEX>Zdk_l)?e-?M$o>F)wug`=~TU^2XGhzDmA%kmdpWkut|w zi^j?Xk#eP!o#)!o&e{j8+<+bN8%M}ob7C5CMYCis=$)4+)x+N8T4}^5JaNK=p25bE zn^&y3P5c0*tMbgq>XTQ*W?#a9TXOqU;ph6_hzJoq7eTlCWUYO|$LlQ#hkUr6P?_EMGO=Y2g_|cH-VeVR7H;rs-sFXoS zb~@X$p648v@U}(<@z0YtdTHv?D{D9~gQ9XszP17@;MC31iGI3)0^oGuFtb!_vAOwl zsP6_(5pYjMN*d;Br%Yw4MJar`hUUNWha*|PhVa)gnoL<=zH58aUFcD)nGDC*KW3@D z#Ry9gpF)>wmhSdsmfV-cUGywPDhv6hhj3Poc`Nm2PXrRU6z&diDRopT zRS5x?=$gKO7+VeOT4qPv;y@XGWiW*}I_nAOrIOeChOy{n5c!Of7@P8X@>CbJ@7mP#zsg%&%j5{MShq z^FOC2AulBaeFr)<#gk0yCMOQ}cha7Pr;UCz)5&E;8z^o1YdvLp9Ikb^Rd+hDfBS6e zE%W37e`nI8^>}Y}#b^xXba!1^W3HOFc<#sGkA0qS;?XMfBC1 zAz#%#rooTb1);A|tuMpF5p0EfN<$=@wd2-rNLZsm!#N& z8~noWtL$VEP4p@4a7!;KrCRzZLVn-cuA-9K%Xm6v-3RZzx6ZA3I=DXLMQ8VQX=$r_ zc|rEcMQk^(F$8zy`IK6q1pB z6h6Nfse`Sep9c0CJ6c(_Z=xSsij}3xSgX?L0vsPbKjR1)a=~LtR9l38+RFZ}!*Sctyy2a?r*mS1` zCrp^%D?~K7EzDv#es2sD;@d&}t<3@;mS!8~%xfh5vdu-Me?GfPB{}j(@8FlP?E;K` zb!WaoE4yE$SW-!{Vp6F;Fr+ky$ZB4Y?c=jR9H7hhHSjlb*cWrR{Ul|dA5P9Z$)yk5 zf9ozLCtWQqH+M)P$L{o<{f;w-vmWjes^(7MAH96K-5*%YayV6XeoyDd%=40yut~TB zpEJs#b%)GB(x_ibyI-~)tEP##Kh#=z^UqcN0c>t2`KKXCnV?5`)u~wtwB0bQNoIIN zzi#SpwiGMMEawChx1v#m2yl|d;1LIKIR22NY|x{xPW*Cg94S^8vnv09vJ#=Ybz06G z=pELq*e5BN&oYL-*}t%KLgaHmu<+IVr2ZRdEoT(E^^goLnb1E=lI4`*`hukzM%=Tj zyaO>B3~_)A=J&);&O9H|lzA5SHwgpG-y{^ylaRdc9M3YC{>d^_B4$hu$}*MTbJaNT zk;d6`!T9co{?_j3gA3g@tXZ=lU}P{yJm$tBm(xqx*vWL`20={pBZ%gU8fk$TW>!ns zJt4GFJ!T8K3-?#!I5BZ~&tKcs@_jdvU23qRE!1{-r!3VK(TUl%L(}gL$Mp}mt?3HI zg)FvDU%;xdUp&Ltkg|ilecB7D(Q89TIhhEW7uZ0(N~*Kgf%-I?KAPAcxrOt6-l+E~ zjcQFzVF8O;ITp$Hx7ws;nb?Do$&b8QBOXx%AF^*(QZbB1KdP!WPqoOFIF|0QR=C8# z7F_=wrMA|fWwK1cha-csl7k#b_)gihV5D$$lESrUZ-jAqhJ}u?YP!xlesYKaaW~!d z9Y3ar zZNm|LvvZ)Y?vfW$Fpg_s_0^xotrE4`ec|wR%{XLKTz*8HZrlnKYD6>JtW;(o>a4zC z3cmpLj922>djKuoA!X`gV#@u)=Rge&$Iw6_mkS@q8CbdrM<3WB_wVVM=3HQ^gXbjQ z*I=>;iHxIgQe2t2z--L-Q;v^-9o93+Tu~`kHmR}>UVQ=$+^t!gBowFn^Ae9{D`0)tPhoRgYYKC&Eu8Q4zc_*tpVrgg3(y424iOvPl(cnExK zm&-}mjLm-fnGtB&oJO7)7^z$}O85CwO%jq`(#Y?)Os;A;rXpx~+snBKx}o6pstks@O3|BD!w7`KwY|X%jQwZsrL9k8{^7=ll9n<<<@NB9M}>POOuDP@lm@}~nljb8>Uzo;*I#uJ;{qfDavC{183 zn0Bw`c3Bgw(-j=gYB)b4?sTLRE6Wu%@LhtXq%Du3YDWRbU{-R| zeJW?*gcQu#Yqde;Z&+~N5A3ZXAVSNzSAV;dYPHeP0skK}ZM$JzfKuixG`=Jc(`x$= z6|3t~8su84YHpb#$e;5$b^t3sQZQ%G7W`4pH4_0mORImiX8B<(h0CGh$4V+X%zz}F z6T9=RDqLo*&tgpO<5)_0Zr6(p_l4+&T_a)H#(3j50KlpMt<{dN2#r3 zDVdZ|aO2u8bM8d8F)LwC$dzDGSL+h5&T_b!XjZFKVEuhNFK8-wo1!PxQmBL z>HFpwn+}^%)ZF1T+>n;%vmKX%qOxjtBu}nhgWZ*X8}iNj;nlA^dp3WA!S7r@#xn5Y zzRlhJhT>36%FKhYea<>{iT&K)TB>~MR6whpj%m+-qIw920e}g#S!x8Fj{suMCyjSX zF=#YUOXaCn?7ZhRaoaa9KX0WW)8Nj$eLyrEwG#T)n>Z^C)mENc@=xYm0Z zS6$@8kJ#f9#!+Gn)HGJzelxclYygXcm)?tLXEr!tHh?{Z^K9ndcYArg7pad3(G+HR zUKY{^SLQf4nHfHBPi>J_B$)^KUvySV)wSmGQd+?W|N)-?A_=W<8LRHV4%8y2 z^ky(JU9PLwl{Y6C!FOBF$~c}=MGn}CY$YG_nwgOQ4duRa9vV(3f{QFnEa-Okzi2O^ zz|rG8D5>k>zNO_8G<^`^ASSIv;A2Hqk+2$T{p8ZZ5NKJ$$aO1fZVb}@@l5E*F^lg1 zPR|))@ZT2>^#Xp0Om@ZHtE-g?(>IR~Jd zz7xMIubRbT3e{63!3pZr<7VgJ6N!qyGmne+loSHMM!f3kGWcKoY0x9^q2R24yn5s*}bqD){q(mON-aFb0r}6E+7iDG-61 z@zZMtJ_u-jmZTl3wdfMFa{GBPN>-jvS}N6Y7VD=tWXW=V>{0(in5Kr3(9Ia>b( z!A`E?s2o;-3}|`M_1(FTr0JC{+?aRmq}Qo_)ZBz&M*=(Lis|9RYOQIWy|A%3A3Ht= zVC$&rB{)pG{hD?=6#k2esj1dF821MAx_tA?)OTp%|0_~43M&Ku5vfYOc9RkJ*e~Xv zgHb($U2x`z`P(9h*f0hWA=3Zq6XxFng|g6YO_fI=#SMRAADOl(USL%$F$E|WGxaao zkRJOh8_MUhA-mQ55jAib@4Q^^74(cLA!TgZwIvf^T#1YE&7h7bv*@>!W-yWtDNu*3 z*^Qt3uc{FuuO>usQ$r15G=2lQo5MYoKL}yjRJ0f~7?p0R+g~&KhJ@W0s7vHvf&gvhUSih5XuF35|@EzDU&wR|7kUmdLc!Cz^( zNFfA4Lus$vaiA7Dg_ywF+9yy8PoSuxYCRFBV7)$3Z3n-*2mV2riGv9>(|@AX{)wgs zqT$b!LX9?}*%4E78E06N=$`Lelf${F+2s}29QI6P3ZQKf@q(F$3Yo}QK-)S)tCDj! zZ+^CpQ@%FY?BsVhSBYa^9g^^`V)XwlhSzy9e*M(Be($p$l_M*cqT8~DKjumhBq1F& z>-aL0>i+|;6F$^Xr?jYGO-AqF;SE~q+<{zS>L0X?Ulz1m~x~VW!r#1q>CZrT-ALDlaJAd2w8qmgC z?H?imSLPxFUqrY$w{8tFm3`28ksrx7wi5jf3dzu9fypL3)F$ny8b8Q9CJQ%qMX#z#TA*&e%X!(Q!kr+iB z-YizVdQ=YZjU5uK|M5Ei5tHR}G3mXS@D&W54Q;8p>5CB2x$Sr1ym!d{-DmIq3p$f- zH{?bBT=`y6XE}bB%M}fN#y&wK2lX#5c2Phf0?=uNzs{E$1I1k~OpYrpGK^@@6qFc_z0pJt?q4~e|kn|sW2nOsS zw97z1)=v2@b~N_xMhBmW`xfJQ23ii>>~YAHIO&OqIKj9>j4C66mdWEmb8HMVI@fVE z`j+2TPAkf9Ibr}kaiDZ3{Tvlkx(5IFJx+ijIwl-mU@KACwXN1~?FK?xDLJ31uD74_ z$-jk9?H}O-cp}mPR(0JjSk^LF0?U4X(2_RuT=--+LbYU=0543Rz>E>%(jt(vexP7Y z)Vkg&#KmUGC7`Z+6;XQn{Hh-yXXyWyn4TwL4ZfyoW)l|N$hEB>;K-aoF1TE>nE!++ z0!?B-DUASCA4%sGX{`NrRe2uG_%+0%sAj`iEp@o=V3}C)_7kTdvsbG|=a@?Ax_e_BS=`o$e;F6vf5ou-r#{dQECK`*>U_C2S=-nZ?_d z`HeTQC57VK(XHlMDw_Zen*d=pRo8OA?Aj}GlC}yHN8xYvLN*?bp)>i`)6g*mAhd+m z*oR2GZuh;N9-~|~lq>`h%p7*>1RHdd?&DE-O=o1fW;YwA_qciVqF_05Z}mr=VeDfU z=VpZqZ-%w@AG(Plnga~$TeCs+IHmn#a(@Y(TaE0t>Y=VYxA~9s1qTgo0mssP3SjQe z^EYj$ekW;oYlU5FoZ0`+-_SY+TZ-$*EorNT97P8`nTKSyek&?;Xw7%U^W?Tqi^SFF z3*4yR0n4vGp6_Q+QDvfT zp%Vz^$qN-dN|EEbKWQvjwyjAk>_HVy+pqVnh|Do_W9ApvGx*gPdQee#QMWx6Q}p>| zuYm%uH@A+FH{(DXPeJY+&1+$xvI$yYhvAEGovFW}m|N$u zTQ5DJvx-Ksx{u8VgB6#n+vhxbZ?^zx_)$FTW{I$6tz=XrsT2eq-v+&no$NbOZ~ zLoXz7P_~hS1RP{xceXB&s(x(G5<*&;NDA2z0=*@_gSpU*Bf-y&Zy&brNdeET1{~JM7<(F( zn!f&X>0~Ee*uI=}$8*-kek(?DcVfky%l}WFM7IKa!>MezzvZ8feXlRzxmnA?VZ4D+ znw{Hoc;+B4n@P}zA6{Dsaf@Hk*R~%of5>_NE;ul3lwl7Oc5?l!>gd;_L^DCJ?^ti` z+2}-36C2h&7Aokfze?0`rGUaIsArnbsv;AV_JX|Uj6S9RLfY)( zOSx16BOD%+;xn^fepNh<~J+7}TXBC4}zGtKT+hhBrtQ zO9<5nia$x8604wl9G>=jKQkH3lHQUY%8N^i<;`psV)>RbjV9%$@sKU=@rJ5CuknZ5 z88f$HFC;l*rqC)vFabT!3Ko<2$chZ`KKxx+1CUQ(mbF0iGq0yTg*S=Qk``y8yf}uh zsXuX{Nx0D{m+x{W=lS3mhFiXP{boE*$QDTG6!<9oCOOQ>B3$eT>q)0kkl?c1|CFD^ z>LJg=B8q?u_oK-Vtj&wMF4Z@Cvhbx&VLV6$j5e=$P+e}LEdiE82;;!TX>6mZen;$T z_wO*pXT=NQXkT6TE44IKnhVY>B#a$bo>A(q1sBCniV71IEX97RK6Kt9vX=lK*4FFx_Z8E|o|S!XwRZDlH@ayF^|;Zpv&LdG`X zLduuMWeS2nPb#M4L3cs>O`>MLQPF(J_j$2e+RgZQI#1Q~Q}fxSv*9y_V0X$<+aOlZ F{{hX`x(EOO diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_tpa/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_tpa/NOTES.md index 3d73324e8..7576e339a 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_tpa/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_tpa/NOTES.md @@ -141,7 +141,7 @@ New proposals, each with TPA's own words: - **`revenue_share_by_fiscal_event` (share):** "About two-fifths of that comes from Conservative Budgets", the share of the £123bn of extra tax in 2030-31 coming from a group of Budgets. The registered `revenue_share` is a group's share of one measure's yield, so it does not fit. -Registered Metrics used: `revenue_change`, `caseload`. No baselines are proposed (`baseline_policy` is null throughout). +The family uses the registered Metrics `revenue_change`, `caseload` and, for the totals of the Budget's tax measures, `decisions_effect_on_borrowing` (README rule 10, below). No row proposes a baseline (`baseline_policy` is null throughout). ## Not staged, and why @@ -193,3 +193,6 @@ The thinktanks-a review (`harvest-staging/ab2026/reviews/verify-thinktanks-a.md` - **Finding 6:** the two by-tax scopes carry the registry caveat in the proposed-key table. - **Verification:** the contract checker reports 305 rows, 0 problems, and a re-run of `stage.py` reproduces both files byte for byte. COUNTS.json pins the gzip's sha256. +## README rule 10, 3 November 2026 + +Under README rule 10, `stage.py` moves TPA's figures for all the Budget's tax measures from `revenue_change` to `decisions_effect_on_borrowing`: the 37 rows keyed to `ab2026__package_total_tax_policy_decisions`, in pounds or as a share of GDP, and the three unkeyed 0.0 cells that TPA's charts print for years before Table 6.1 starts. Counting from 0, the rows are 2–4, 25–28, 98–104, 130–135, 143–154, 168, 235–240 and 264. Single measures and groups of measures keep `revenue_change`. No value or sign changed. diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_tpa/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_tpa/claims_staged.jsonl.gz index c14eee2decd5498acb213d3e8ee1e853d0e65758..880354a3c0a883fe4184dfa2181c281b33351e1e 100644 GIT binary patch delta 19615 zcmXVX1yCJL+iVCVxO;GygS!*l-QC^YIRtltI|m4QaCdiicXtmixxC+h>#3>Q+3D_` z+1;w0?Rlm*5286Aq8j@fa6mlO=e~AN$SWG1D!2{t54G=?3z{v+upX3ka*W|~EKl|! z;HEbSds<~M(ePCld;H#^>j&h-;t>Meqd+kdr@`&(gj*K!W9HI^so#QB=L=FL1nIkb zmO6|QccpVNZtYT8=+HKG3mEG0_sG?kP*~VwdS9nqdcG%!gwq}$po`_fJoHtlVmqRF z-1q@;kTkR!ITgy;W@B@$-mc@-fscW(#q9Tv4`q+G3Go$q7lx#^{V&2$8b@-I1+WLX z6=_ycs}L7Vw^wh(HWu8(2<}nvj5*A{fM%Ee_Ba@SV%NXD+aE}w_=!)jRF7jeJX$?_ z6Gq4-T_bXZ$Huf`K*{e!aQ$2rp{UOn8Y&5gX2)R7%lMRkcaY=TuZ2jBy@-qO4DEGH>Xgebx*S$F}O__rQb>! zm(>KlCdhgKy-N)_TuA_ZjK8Ino35d7zf>?a_Z_y{gT97N0=2nrNpCu;{9kei6ui_r z)sp#>x}XjJ$puF(xebUO7HkbeGq?6hiDSKjaCaF5 z;EtHwTc9=NfZM|aAfLv^$NVM9750C&9>@pyd*i>K?;u4=8?kdTfAJ5CzTpqE&&L_= z_Fqy3N9{p#H3l7pdL1%5Q?GfQIDo8M^I2}qX!s3X3umXur`|1@LRuzg%AZNI z$Hco54vecc!=E=Te3q_<7e4(_>g-kZO}suz(X6$T<^4t}4zz8^b=!WwT+B0jXij~d zS9X*qLfuWZ{uUoH{?Uh+GVSOU7pv!>2i~^G;@{GpaJ+We%nCGU*P?o?kGo0#!0B*( zT1N9&jSF@wo7wNRg_>}x#qqAX^8 z0&1vNqVrS4ejJ0&Q?D+(Nz|670xxzWdStihubY24Gxr&khQSBYh}E!$?`CB$0PWfr zG@woLO#<>%W89jN-E>?!!U_z>eFyK{qdzS|=?@sEwCXc}Wi{Za44N8gdt8IlXF-_H zE>k$Yk`2;eaM3l^|KscnVv3B%;FBn4mZ8Ay&$Wm9e@@m}X91nuNR9&wcPSt10l7xk zT^_$|jZ=)i3z{3TT;3q9Ef#Q@*$uy5wC~ zH3kn$e+)T@H{;!++WAbXo5nIh@wecJW$4Ac=QgT!3b+Pmbtbd&+WScm^zX+1%2Mm&zCz&Cxud^cO{({5ujE$si1 z`c+1Ix;0uYYKvVRT+?ij98|W$gw{(Nm@7O~a$o^cJYG zb%-gYu3rm1ZCWDNgdVPUd2v=Tpw(i=`s?va@)ddsN!5q%38X=3^{I#;_gdab^53uc z?O!bJ5hc{IOe!hoB$j3h2Rto2R%coHr<*ReH5@xU9<3O8Caa2)lCvAp2+x6OdO9zP ze%yvB9IDr;U%N*6%h)_v`RV)C^@;}0!zmSUcmFllDLy_mZ1rS!2@Tu{#C2HCI1IU) z<41fU-o(KCM*>C2T5j1Q&S5wfatt7gXcW@dtFuu$%KU@l?S|B;n{@20lz%4rS6Flj zNuAyfxbVXW^Imhqka5MW3UUG1KE^;g*m%XG69lTT8W4Me@(A=%YIeA8=rb+qbRRHH zspzhRo^q0+_a?e zO*UOsXGk5ARw4xS?-}z6RIt*S^Eq7AGzxcna%9$I%Ex$v{On5%qnLnMHSxUYscs#w zzRU)}?ka(^Md`|`Y#xgkPS_&#&5tUj zt7uJL%NO!5{>-zTQEt&T+Y@F&)#8`At# zdfNL=*ZoU=ax6Ld)IEFi*CzimfeB0NZ^1SZt*vLMP!NcdLC+iLrSvQ;?=)LDBd9~= z9h1`Xx&mTsiScehnOI*`C$K)AyM?mfWR|K!UOB72L#oMQ6t2fBUg5b~_1c|cO=W!*X z-Ro&26UKD)R2>TFsHNmfG)ZX#Kx1Rbit>+Zw-Agt!zrNBh{>ZYl<{MtK_)czGY|EQgWFV5&r=@$BxvoU_&kR_(Nov(B-z( zo=`M)FoUrI31H)kLy!poiMjRT&#U;_qv_a$zKGC8408*qbbzl)J&tMQKj(O==X5w`|`CFz6VO=uV*Z3xRGCr6H%FNZAOFkNSCx5_V$@K zhw4s?jTmFmer?XFb4mLF;el3;S~C#bklW$i)(&WjZJkO`NN+(#HOHPkT;n%8%u$~ zm9suBF$otMk*={}!I?2gb#)q|AOu$Hna;e`p3a>-0!R+6K%H*WBcmms5oIFDWn}xn z_b{rKr>flre-n*mceKYp9s`wXHYKS#Sv5e1I?Q~9y}vhmZVN#<_WeynyAi-aROqb87ZcWAVVvKoek)>J zQL~RY>c^wduzmj~N(OW?E8N9au}@znM1w8hV)733ar+uhmP5_?L4K&Zwx`uCIhqN4 z-CK}8hV@bYI}ECaCbywegLs_OIndbF%a?x)dXNR(EWE!_a^^r|`k8kDwDs_#2}m`R zm?9RDwsNX6Pg^es#s)~E)%hYo*;MEccsGT^15yw~9y};25}qw(rEFPcG<^f4nb}x4 zFBdsMtI+(Y=`Ba5DLDDM4FFJ1nIljumMh*C;ci^24Q<>N$4Y@RWzom|GEFB^(KcJ_ zsXC-S=Gqn2?Bzn=!Sdd(JW{i^?A7#~684JDhzb!|nyvH~rNa|l-3OdZz+-H!WgX7r z!q3p{q#1{<(b}Zs4~ihu?B9*D(yO|*lk)N9MgL%!GNVjCYW)(*d=v2v{@nnnHWh^K z9=OD(tnHP5*dLf=*>}`EV4R7lt=@(T2G2;anHU{8?G76Z3(n>9kad`Kak&0^<-_0$LwpGVryTh`8G5PfF~vSPEUj zaD37o$Y~J`G3m`nf_{b6;~jCL<}f$^u~}D$nL~~+Pqpj~BuBxMNZTta%Fl0t`WnBz z{cA_ffQ@-|AgQG}I{4ncK1tGL2&;X(!{z-85#*t>1&q$urZ(Y`$QYoXPqR&P1tw+duo6)aK*||KDh5;cf(tfx;Nl`Ef%PJ0BGG+#MVm2al=dOIyCL90T)bdA_WM zql(@US|}-8XABRM7_U-9O&9DB&`WLlP42vB@asLTlBo^pa>Yx$VUL7EWI5)hEF%?} zy?R9s0}ONAZ;^h3P7}apQe=u6ZX&kr*G$3(K{Fiz_r4Q`%idHX^(Lqs%{vB^bfW$z zKi0p-0{kFW{EK9Y!>3jV5OPa!Z z>AzW+qt>4q+WEc)-$z0b-bZYnsT5)MiZW08+1$UrRt)|9Bks=54LiDhx$D}Cj26H4 zZJy--7vLg@2{+(vL=B<^F?L^ws{uPwZ&H=EI=1u|*Q?l!)SrqMFoD)#Fy4pzE%k8< zg|ANt8oZL0yp6vDsNlAh>$eC23j&tm@Bj`yt_*Cp5h9buL4>oj*F4&gVtiU0UbNJxkhuEw6#!D-} z{2XOI$jVrBp0~%e{M!x!@vCf9%eLK0(nxBy4~@+|FUd4dhuh#+K^ZC!92YN zK#7mpp;i&uaf@a>mCW(z#Dr8(+GpluHioGeb_jT8+jtXa&_!4dLNA;OwV#bbwP=T% z3CA?JiP}!;u6Nn^l(1eeZ0}k3 zHGAecPzM7C9VEJ&ux5C#dgt8skf5b1%p^}vf^EKwzMvl-x%pkriz)!X){UH-A!-~q zZUv+!6o$o(?uG7U-Fe=<(RBr)5EA9P2om__OBLMhS`Sgnt6m}}k90NVC~HJ1l$MD; zL^UMVJE(`{SUm)9et$GFq2MIn_bi$)W{D}{vmD(s&L$u9zQ`A{!H#GZ-DoYf!O)Mg zWNX8!xtQq-FXYJpc z*ds9+$Ls&skYg9iOj8^%fQlPSD+OOAy`Um<#@_IOtsc+l8T+^}Nd5qbW^Fe;J&^(b zl}_fK-&XUx2fs-dr1=T6O(KThnQ*iZi| z%$(wqPiIGlq@NADs0i`&IwF%kj^k$0(OcEjrtdEc=j{vF%}Emz%W@-5b}cG4(jV+L z9iXGneyZkvF<2^e{ewL5owAbZ5d=KElf&hV)U}MXRkK-M&-S!cbAx(vUS?&4^FbL% zSng{!n1|X?*Z$iRTfcG6Bdp2ffgj>T*O-ILWcfR{%NEzLv;~C`MdNK*Zp{&Ht53$= z8Cj!0hx-Y^dT+AP;az5(>m3O(+qg>j%{HEztAVZVTnnmYyJ+Bie@+O8d)C7A@gILSSK^f3udxJt19NxiVP zt>UhhK?IpJKm9tZFR*uanOCF?JzPzwuzFc$H5H4O^RiA0UXKwY8i@;O-0HaCCUfa`Nk+B%liquL^qP?As}VV zt__Ca$#(#$W(ETwj~$jvj`3{VYbPCR&Kh6GM3Ky*5>(vL+_R zU;-)Jl8Dg&A?<%q7r4DoC;^7LZ6RFnUvvAuLugdxIfs<8?_c-Q0@Tcu|D57ch^38J_A1Wh)!)|8*{ z(qZ;4hss=AJF4m;rLcFDSAB()))WBttcz<_&h4pVjG4NX;f=M{FP%KBRXtG}c`>HA zmC~jB?~3d8pJH+jpJJO$t2K02nrQUp4gqE@p_nt+!DcPzt(FMXM$b^Mct$Pq*t$w| ziMW>tnjDXwJ6tJpy|)l_nFlqtXekQUeW)pC^#hm#EG>K1L$$Rr2lGIDXiasPv9Rj$ zq`_KYO@R&KuqBc$+R>l3ZxVyeDo*95T>fZUEdAN=eYKr`JhE_iuP3~j+W%0vL3h>a zV|C$3_Fk2WH*fAOak2W3JgM{gPY{e1vhLw^NsV5j-u7?WJM~foxUJc>MvKCH!FqEJfK6U9T}#Pn0_z(Eby*PlXZ= zRCu$W?B`E5r1r&ALz9bXC4%LD0K6{u6{z(8b=)T$_<`f!D?_rUrS_y&_cu*}RA#WQ zZb#$C>T8U}wK;zavqyaa(vC*$*e>>0sV9E7yXpTwR8RNsZp2tWG`4s~eka+%8i-UPymr zAkHVC5$=M^Kl%2FVsB;SHTZ$B>*4c84A!;Y*dSDNbh3KJZMS_h=%XFchat;*oSzVS zjoloTvzW?Dm9>z|<-E&6^@i=RTS2XVJWig9)@R9Dh|#|b*cvau-6A+&p%dxVl_U*g3a#>EuN~26Sm5`iH&Sm5R7@79t|= zdvJ}|i+`yca(v(m7;#qT3`iEq^rfXF@uUh&o?Y{yt?xLpDq;kGPK}){T2U_ku}$FMse$E+o(bvcBSWN}+5v{HaytxMB=VAoc|YLd1N+ zzFmC6U@K5zEEw1VnS`3@)!9B(&Feo^-RoCLKHvE8_~)5xxxi}%m@%_dV?%1#UaGWN znBM#4*_<#Y+_}kV^u`CCx&IK8In`>NRI-!XaOYN~#Y$8@+F~e+ODO(mI<>{w@V&FK z!6@nca^N0<_AYuiAasvd4LQNyKP^(YzwRWaw>>JHKMJ}w#V8^r&{`p zz()L{&zDRKXeRF2)2pWpA4Dr<#7slouK{EBfh048IBID#-ctr==Wi6O+8b{R9{Wa9 z?Uvq%R?CdPy1ZBT&Up%rs(QmeHP${gaA&MRzZ5-Wh01}AJE~Sl##8vBCsHuU<|JUI z{Ck4lWW1UGMm1w?kX3#W;Ak$|#|hiphDBv#MiQMz4g<0mJ>w5Ps1r$xwzL>V$vbDy z-^qm9$;Ks3y2T|OM%4X{Yjdo11Nh!?zaQ8RfSYqn2f8cuBRAeC0eZcko&-9GAo} zC=mi2&|a6Kh5|gmm_1Bh3#UWXwOe5It-F>=ph@%J$&tUlm2kyBjGCRRv{4Zfbt?Q3 z@O&n$o&C}QZLvmw*g|Ps%)aTf_c|WfH<02`c46Cg{QWO0$Ko8gGQi|Amb+}tJMPN_dfSNXsz9qruppKnFazL0P z0L_{opmyWTsJ;%WV?R*nG!cF`c)f&vAlc=-hmn|)_D|!*sz=_2xr0m*Ab3fLkU@p& z`ws6QBNEY6x4#+Z#fw+1358;x6SD!cG8xEJCW}DIWjh;Vv0F65YlG}{qn(X&`Ar>+ zNJmn%%}zb&xG$z4r*3)yFUz;0HVz+z_Ls-2$_o73I{HVizW~!_kL&Ol5%k71{%^9 z2E_*$#ZR76a;<+TZ&wzAeFqV3XKHS7uQu|7=7S$@Eei>rB^b3ZsVK(zlBck)xrvlm z4s-18!#c>!d=JM5I02vLL;Zm(&~hS7&$|gJ6Kw!w+H%2Z9>Z#sGG5%^;XeUs#dIUr zd)~@ulX5wpF`vmlqQ2@do?+2Tx59chSWa<~Ry%pCyPFxjqU$_X%Q}tov-ccUf6U)x zkq6y;q)wC5+|DNqUVxSUKZqNpM_YLz0i^m^*oQ15Xoz&~h8^6Zm_Mwo1ek$BZ%{a5R*AD-g%%I{iC7=ns(`d#jR{xqW10-# zg7A8%P)6w9_KgUsft(|eq#l)s`edh<*K-i&c$fxSL`w?G_iPC0qUb6Pr|j(QJ=PZ& zOBweU7ZSHv?i-q>L${O)_?-sTWyqda@sGpn!!GGWb+4Za-FT4UZrX?FqKz8-q_o49uxuTSy>MI&Zospkc^7xtXYgil;~B ztG;+>m5$UmQlQgzw*uzZDyaKy1u##25`;=a9{AQEpz7`$EeWZ*EX8$Q0h)z+1X@bM zM`^H|5p5oyjRr|Fak8Z_MLimAUc7Xc4uv$A@hABK_{n}LUkm+Y=ffsX%ajDm(>mEF zQyU$gWJ|+9PJJs4ekfN8|F1c|CRoXs&gi#vmIlS>)9+9A?I-!8d?@0-R@CHbnGz_k zEbVir7dYc1F_7*h*M#-K&voQ^99fq6Y6Z8Zef$e+9^`AYFaE!#6BJJH3sS?n*a^M(S>^$0LrPh4(m9;aOQ^=Kx#l-A(G0}&V-y3uY?upHt z`cz6p&GrpBXZzntE1QDP8kKO)IVP5ymfz=~Yu75Mc)2~p&+SfX)&|L-;tm05s@)PZ1y^@)DFRzv(1 zsbo z_%DDx8&?FklJwWG@klsO$GA)-+mv#A!zJ~lefZKo`47UQZpadFGd9r?ihkQ_PC&TG zoiN``ov5^WPJ)!ix@xAFkj8r__PRT^&T)ungSP}kjW1t`SRTA5c0^uwfb@soSXh7* zEiolh>PM5XbFZLG;H{bbuxMzyR)zW zO+w~0f+w~I%?)bl_J?#5wHw07T$`$Z0tIU6=Q8XpiA@&4Mb^Y@ zKrdz&iA@B9r&q@1;lt_r4w!!*2jBen4;UF~@2AZ}+Tp`?r~_jqj`!ERRo0k(gdj>k z&Bz4pLO@Or2+f>hV2p&S05B%etTo|Gg0EU-qq!6lrx0NEFA$2nw zwY@;C13$(|jsJw=d`?*Yq5-w_KwgO+4hAwmRu>obXm%hee^w=K zpdsglC?s|Nd_&v)&I-M?c69)=tuR+Jmmma^;P-R!V{Phsr$vrYa;$@Y^NGoj!($u6 zT(LT#PhxqdU%fu=U>H)06%FEduHJH;95Y0KM!2jd^PXCPN`L}TK&anMs!%~dhE1KVQHL&Rkwl(%9~G1l3TKp&jk8DZLSVLQ0mVy@7G$V zmX3Ew8~2i?tsK|v*`D($=*9crfvS1Lwf6QjFyHk|3&#QB0`sL!U5q^w*bGaR6!h<#qV!F>b6KQA{^(ejH3*5NStCcf3ECs-z%M90}GN zCSw$!ePeFI2!_@pN1Oxz!-YED-#h&3L>;#N6MP!={}jw`Vg!L^=hvvz4J;kDsJGTQ zOPYX-o1MO8wDW6>RA|yR-mg~PKj=hHFP2J)Q(6m&asJwUV-7>Re|WSn%1>FA6Z>5F z`u9b)v}vn%v(wrC|N3^)_`ciWr~j3H`z?_{u##{P-UjHtgSE*zchBP(M-g-6gs{>x zcOB4OiZt9kj3~f!{XGv7 z=kwWYHh=Q33C-+UkjOo$VW)DrFF15)3G|+(*YS&~Rt*Q@ixz@I<-^}?s=ZRW7=$Th z@q|LH%5orXt5uA6Xd&&QCY~JC-i#?2xArVO2W*Y*Mkgme>snp{w#^*cS07s%LWJAJ z%DX&*khyZ`A`Jh|?!A9LsK6VkRzrKvKhOny;d~1iN6dl_?)7JQG>!aap#NqPChYrk z;-+o#{80)m^PpL}3>;1@%(q0PYcY6G|}C0QllSJWl$DE zLdnZY1w-JzhdiD#IP-^s+-llxPVM`oFE+oJ;bJrIkAlrI_zihPe=dC=|1!O7-tdwz z#_9|7%(vL#Az{KgmBujwlJ)$!ZPBJ*ajNsFV_z!AA_WOJfTkrK9orSSYKy-;zb`}c zjdbF5f9B)$1`!GM{?rbbj^H#U+CG1qb|h9Bx8Luv$Qu>ggj&P&wfBvq!dmJ%6Q)qST@* zz&N)d^B3c%<|W_yC#dP{+amvJx~(~t-^Oud%(9Bm^NuplFOHejR;(TZ1afyj~0KUY=H~-Emv9*6!uNlWDM4Cxt6; zOs#TUMGql!<85li#LX=|n7?2%JNpezCJZjUzze3K0{^I9>;9K?s0wfwBncsSi^Zmh zizSK{p;q^$hVt;gdaBW>iLAKFP(20Ss!1D^&HE@V5xb(NEO1hu;>j^_?f2$pMxz zeTHmsTy6A$@3GMbN0`kz0Sl~f9y<2ZF4c~4?#P5(mL8$JP0}mb5G21ARdV zrv5Y&YHU;O&9#Efk_7#V?IfBD6U?H1ZXR9|FL(+Sb)#*-1zzI`RR-(!hZ##l z($ABLm?gC1XxUsRgSe@E-OecFeEk|5L7b`3^P{aTe`LD}3U0nOd~0zM z?s@pJoy~lH_16c2Y7w9b)dY2gZHCjY(V{O;YEQ^^`!GNhh_ap`hd*Y~OPF^(@(ZL* zuliEQ-o&mgx&#ihw3LTIlADtZ4(<)74+VU2@;miTX$&-4&sskx>_NX*s2GgDS3*ig zf0r>On&Nb7-<1mAjYVD(9n1U`Z}m|S(Y z`@AhZXBB%Jx~>6!O3KfSVsro2A5<}+$BMqg`ZSXw$&`Ztzx#Etm+zZ%TXR!g(1b#Mu~ z9GGq7kGj#+kWdjyGWb5P1L5U9 z>T%lue(ZusT`zJ;(~dHQ-@`!?f%8H_r|}DUq}Z1v9yVmpM#_i@Jkn{-2eB08MX zzz@5;fbKE-z9tXsPWQXE+YQ}_Exok?H^Huv7O}TW+Ah0}^33H+;!b3oN|>#S{w6;T z-^7`S>kTzufV9?UdA#=P&JlgGr|1MUeY+UqTnU8P(i`xIVE4`b^FV?PL{@6co==$) zw{xCCX^2>*FL;2cga7ZG`lPrWOFBw}kMsr76Tt)o>p28#9=1i_#;UlBS7v*6)Ft6F z)}8+KNL;VsWrU{2vDbR1!7Z=8<|QR&!7-kG>J#NwORMN%y18#0satLWB^rjm#0CcN z)QVPUf}udp2kt>LW)3vO;h;yJ6YtW zNivvv480V?%$p}pIvzDr75qx;R+ci=oCoG#ER9SMjF(sXT5@-c!QClY%nu{gvB}q(7^C0V8RQr zuWq8D|E{$fk=m#8khB!bY%92@K+F8g$b&ohfZn6Xjtw7&w7;jlvOw)d1t#=lSQ=sX zF7rl&6ofqSNJ-a{;s_+foBXWK($r;EE`t&(JEAOJMU4ka8qoRlv=oT8_BCE0eyhse zkF=QRM8)D>3P>a9uWb3InmiNtHzLcw;SAmDO%lXYrr`bDLEskkT~UquO)__QSEKD` zd`lb@xDDM``2-?q#>Wdhx+HNcI@q~-5JBfhuCS54l_>L>_6{Id#C(4Kq)h)uN&b)W zwJ;T%y5rB+Z;IWBJ`hWqLDAn6?<;C^c|hyIz_p2D8HdB@RH=bB42KOCnz<0I$De+c z0;UTQ?>qO6An| z{U064<&o{a(lRNwe>G}pYaioH<9n{!6U6id6Uw80mq%^2 z0{?tX1`qE^(26m1-GMQ1KO_q!3t#e-)wJFv3y04=^pi=qh)OX^RTWghq+Op!+#i!= z%BCM=$)w4rfztms^+I~r+8|_t|4f{bqq>632CJuk&b?0NLY_aOlgS7IL>Rh4%a2(! zNEUiQDTBaWWEERthf)f@W6kJdVEWV!j2^RSlET>1f|^qDhO?p4Sr2<}T zb0bzEkIkqb0|a$i71DK<38fY36|*5IBRoZ%gvPIlp^d%3TTa_LP4fj23ZHVW&qu|% zFU`O!ep-HYLEZu$J`;BaiY+$PjbWzmNLDgPRF0k|mp>|1(bkT{>&M$%o_mW4@XVG) z%>c9*I0?cjrbRly!$uO`VxetyXP8>ol2r_1Rknraa{!$akgl59Bv}h-)#l1UJ z1~!?7w^hsOj;*0ur9c9JOwo`E3m%j!r@PbYjwwSD|(2cmA zV{k*wR*s!AKmGx&;Fj-*q)FPZsHaRyJN-)Zn7Tk@nqJUjuoCgC-=AJp42Y{S6}*#X zC!q?1Yy}O2$r_tL7mHde@{_E_uN2iAY|>Ta>^#E{kVVYKKCdhGpyZ4FS<*%W(JsB& zQJ#A83ZG9_HECm0&PB3LCQ@BCzFhyqn4!-ItOw!aB(UfGH$lYqrhoI8MF{UR5atpa z!b6L;qdwoqQqXXNq`ryhV)Sz%P9InJ)ht(!CgvfQL1g+IVs3}St%N_6-LP@x@Au!i z2+dBGfo|I_4T05r<|B(l@CrLTo2T3U_B(1@+xR7TU5H=ihuHb0bYKbO zIpsfunzp`k_4))EB+n8WdW@bnR+?O-?o2*Ps4JvCo#Ys20*7d)E;&l6@RQE!r2dPV z2+ju+{}%S`w=azv?$YhVR2p`i43>yD3&62hXl5cRc^C-z&nBv-QQ9s~;J#TpU>9f)pK(Q;S-Vb62Le zH~X_cI8}yr-c%Q|pKs=D6tDaHGv?d-M2+s(WJI)QSSqTe0Fs|!?H^#6ZZ;sC zB-b^;OOADc7Ik_qK#uia0NVzHmt+(~A`*_9bi(`d&5TJBHGhA`d{%)DBR?X=?xuqW zT7=kGzc5J+@n@81(b?9uYcBf~4Ls++KYs6}R6BIc9KE1FWHjmL|Tm6Eup<2K5WxjKAQD`24wS1>;X<0RS zWLMJz(XI{yakMJ)B=>HX71_iR*!nVL_an+m!Db4wBDdluW^5tAG2+}%z{ss9ZJfjq zfWpBOcd<*-$7zR(anb+Dj;>U0tJ9eb+A8+@xcmb9nu5hO)4Cfp33yp`{e&bA%EHy@Uf+r-Ng4@b zAsWBV`Z%>|i@0^EonW}K6l{5Y`wDdVQomlSVMudoJ=Cdu-1!?9{Uko;!l(@eYQk8e zpAMh<*d&!3(4vUdL4q9}FRN$ury)@KTAy~V62vnMWm&rHytqKfkPZ3k)u?fNf5Cl> zRR>Q!eM`WiOh`kgY7Y_M)l7bNgfIYZK)DmTTc*y8|BG2kgX3}e{bwQg!DAw|U20su z{bnGL#VUX3xQTOA5E4Dz-2$tpw$%DkK@Fs8q@_v$P@&BZ24gT4Q%}Lsgc%8u@axVF z(H`~uaaX3zqXZljCZuYd6)4;77`BAfjfbj<(h6Z6+6N58GBC*xasNB3<2)k#0lk@H z$Ssgy&yku25{_R<_}I0aCx%_gMqYX zLnUC^vkH|34sXyo;b##6^W5yI#YB+gT(0}TRWWyYxBhDDGQ+OSH(u`ksX}6Cgg?hzN3a)C9#W={D>D5{xjLi2}#Mp@jIZ~bb z{A>$~spg1E|Av z%zVm`1o#!Hz9}v~UXD2Z{8FQ+T@1AIrCsU!#{AOrjhT$|>9`wD)8FHIU|?CQo!fp* z-UW2hoFHCDPB#^S!N*aV3b)pk!DL2duh5YOAK)7Kbs@)@;k6OPB;PH%StJH!ocp zpp0d}tz6O^M`(t{<;jVuk`SAbkgQ#l`stnSaMgvKiPi_5i5gO$K$zv8GxZUfYXm&y zhI5&C?8MIbQ=J?>Cj$0au-@N%s7>j}A1bKHl}83um49Ts+yPfGxZdGT_uWqj0g*Zq zr{)jJ7&ZYZR_(MY{Z+QmWiY{lCA z^Fn92NX9wRKu*u#Uk3e&i?;=MNanfj{gf2s0Y})~x^OvFqR&j!L^&7~V~CSu5%twV z^z(Lw0a#d^rGyM8CT2I=O1!1NQ65b&&^i}sL)r&cXM{CkEB`u{l;G=j?R;x~gh?Gz zbC*sSG%sLI6+5C*Vn7%;aZ_g?kBE`lrN5T-vnMbK1$N{EINeijX2v+(2`UmvJ>laS zssi(aSE+*NvzU{8nxjPonxn}g)?p9L42n%PMhxfDkPd5ieNnz8j%7`}qtJ5OWTc4n z)kY2*-8|0*iTVBgP6>^)D(4j3OhgfT-O(8Fhc9W3DwZFgln|f9o7Ipn3nSHPyY{2Y zOKI3b83=(jX^>JU5&cU_c_mu)(0MDji4ZXeWk!l?$OorRcW0(`$ep^FFUj@U>tNrO zt|ZG$`a_e1-YM4p6pS$799(_7OKwEZ^1RyDvJ}KX06g4jq&{Ol>h;4|8F%v&3S5oO zJ*+)^BZdnnlxY0PBKJpoWSws9k01B*A6K;pS>PVm3Tw>6h(0`so0h-+fO>G}@4n3M z5bDq(!Ml!9)=nA8iwq@RRuE7(Y4p?cj0%COlIlscs|eqwp-_$yq}kkaiF6<=`BMuG zAt3#*?L+P(mc#sY+7$C6$h$lfGL(QzrnR1)J~5uJ&P6{{b6vd@i zB5+acM;!;_P9&)YU`GM&@mLfRFYKhmSHv+c!Qn~&b-yhn(|SbF!wdy~CGeju7v9^T zZwwC)J2zioEN&7lSr$~SS9fNx!PPg#hd9{X~{p_&Srqz|SZv1A)QRl1Ml zR0ucxlD*$&T>C7z9RnG%?;-y$)e9>0M*9nmuOVg3WhMG%?#D6T3Nc$d@gB;0?lE3KB#({CBTnx; zFOS9dOnIiTkJZS-hwZ9Rn)Y!*Y1-`*$`S>=d=dQ)-Q3&IDwizomTyr zcq?=?(@QhGIepJ1UHTv@AiMU1@odO`{{8KL_ka+W2EL6dyI#!iRXpR;-eNv_CGDOA zc2v#vBmG}`)BdmW@`|akqVORG7^HG~dBe~{DULGr<*hMS;V0hM=e^X;QRnJq$AbbI zCr$`a2_uLgsK9e7xvhV%S8V|`V^j6kT8DQNyK^sX2B59$UR5*t0&UnGavlvQUObL} z`~@Cwz-HJrj6H;*$S*^ZaW2XD(M!@>0xhs{waaja6kt`tCZB}?J6re~kBx}au^OpBE;gGU49~s*}hwV zPeE1n;vwUSA0rMEGYw~pD+MG!ZkN!ge+;FfXyn0a0FB>9S(?!EvvGZHf@6bl#JS-sQ5HnUTInc) z%v7UlGKvFt2#pwmf@l=cY5Upae`&2aeG+Y^szY-J zS6^(k0(KpQ4@`Y|!?ODA&mIR1pQtdI{f7%6iynDp!`VnxWB z5Y?in<^|9Is(o)Kej;vW7hDiv@nRgwURfDGE?f123XAR94F^c61(B4e#&43I8X4fX ze@s8Z3gjIc)T=uO^~BLyIBJMdK0&nZr$`LyzA9DSyZrh0T|%f-ykjXf)Uo{eeM1O+ zPhHA}QD|}$z9$sckHYs%QLFAy{`|f}D8$<@h(ZymCQL?=e}~?!hHOUblM@shgrdcb zMxro?jJ1+6jWUyspJvMc*7Hz zI~lBBW#_<3e<3J)7_Cog(2)JuC5YzM_3sX7wy2+$1D%zmzhKe1xEB9Us)P{HBpj5g z3jZk$^#MNgF!;ZrGp3mMZy1eY5@!wnjXe2teJh6cw7zakblSxyI&E4(L>O)nhG}h4 ze*v7XNa`+nMF@*6QG$=s3MGPXtKds(mwHU!Y}HeTe|x@4ahH_R41O-aj2;xY^SXkA}%9KXRy;+gl!$BlJ^37dF)KSsDHAR#wM6Pb@8vkYW@ zHnz0duW3xnGD2BYXlbb8Ht1^hxA?jB0{*OVf2rT|4YCWsmJ3_j@T0W!gfQI!Ow)pt zz7jfHjdBri=20xfX80s6OQ~~NN*Az>Tcxh=WDuXZjwx?m?_iv!%ex2^YlTs)kyo#3uO7j>VR(~IuWvIPG^s&Z^a#H# z!7t_2>luFEUkH9%zV@+eS@j6S4TG5S?Df%sm!;Rch>48CsX{1I$n!vci)MK?e_F;6 zia4h1dM8j=1q$y9C%Nq!ByWBlpp2Y-x-f9x&w5;UJDXc4s9QLr-Wlg1iV<@9xN5zM zxnus9d7sC#<-`x}*!yvqWL?_)-0xS*2roo%big8rK86~;dyU6^rI@_N$IH=O6fPH> zH}l0TeDpATBA+nMzJve9?}f`Ge+H1PPI<$^fVu3qUrlKsNfL@OumR6C_WVzHapQY` zHenDbkGIKSIUhk?f1wcx&-E^1J{pB0;i;;VkhN!hDxn%;L;)2MxhP%N=w*fB3gIZ*RE( zMB`mwPs1?5t70;o4DZloA9(l07|NMQlrikhy@|h^7gP z^&_YsLP>UOgJ;-dMF-BC)6qn>N2+bPBYQNljInOo)R z>3Lm9q59~#Q9*IQa1>Mr5xJJ?gPFF{^7@e0;oGbhlNfYb46KI5e<1d-B9j_4uLe&W zHnPs%fqovGbzlZh?_=#KI6@qD(xt6RU>8>c`{iYf6f(3Xj|(2=j}V|W1Ox9gwgs6t z1pu^QIgK#7FDm?`Be=Sc6Nm_lmvg)z2(WIc7&Xrw->*vkns1I zWA>n!L8H8bI4pw+e*}MnlSB485lfY_l>B(Cjk`*HNKn$*iQ&^apBQi%wqzcj4HU;he`xU(k7Dpb)|;!4qLQLk zHC5@(@qr}C=_SkQ_bYOsc1lvUmVq^v+9{N%uj${<|Ot04t=EWdHyG delta 19515 zcmXVX19T|y=gf3HPfhjA z-0JS?>XtmPzFe>xEJ#4CP}dgwy+>CFX2KRs!RhU&^3elK*O(87bHqhT=ifs)%O7i> zsT}r8CrA5Ritb(=`Dc3xj}4|p_0HY$%WYiZ2D@QP1mwmYTDLS3uTCpV_~`g@UVEg} zO+*8oSihb=EUL}E0HT~+!VazwGr7zQTk8HbgNG5FpNu8W3Sxj`c+G`g1l?NUw@1CM zc8Wj>o8D6jwiG)Mlp;h(uGyf%4YBlqrHoddCQ>^qKd+Kr=2uAo6 z-qdYt@s#2!c6JFNxMfd42~|2RrrV{tNF6*a14-Zzy?N>Cmf4_76jZWBJ?bK4yh_ zBg2Gh-gzC#KYe!p!hDnD_BA$o3%(7ee-LEb*Y-y5HgtAGa!%`s;}yuXzPzNW#;N-y zy>da=&eIH}lH5vw2P@eK@^Y1WBKtq_{4p;f>KzE%kcY+^MzX*AHZy58&hmAr%Gn6Mxzs#!{S}3UE0YNb_)>VE#??^KcT3e}jegDLHv(u5 zEyE=uSWfv@oSfxeJLB{+mB2-akw))9okHws*0F^rk!La6)io!$b4m-I1lT{sj4xIi zJ*JOC9acZqxNyjkAemsS=C~{Q#nFqYrH^=WW)B*=)}auojdPcbO5cJkuma~fln!nl zU?KqvtlTkYpM$-Ko**0kB}qYfU(Zuk117aoLrPR}f>dsmfA*O$DZap@+-XbJub7M64VQ{~)z zI_rx;L}{|o!IjloQXXKmSDB=KeC>Dn!;T{YHk$LEn>|b7EEG_BRKTV8b z0Co7+%wy67VlNl4ljQDVkEZ!Y1N^ef%j_(T44}!}Jc%dv^-V3+BTp47J*P1t0 z!CHhTIglO4cATTSd9kYFAY22qo|kd8d-VpBdx}NcJqn^M8=~g^)f^%=oH`Wi6EDE6 z=<;0!TZsdCm)USh{f~jVnHrJ;eotz(?L%3Rdk1F_tFGm@uE@$`{P(vdFi*z#Gc;AUW3*l*=>@xqDX|_Fk z8C+L^HfQxc|9ujZ4iS7~z~CrwY7Y4QyM(QpWRY!i__PHZ|CBgx`keUpA@BwP>!1Qx6-yt!mHLlFSVUII3ooT_8a zD>7>3P*Vw|1yYHp8T7Wl^o_XaMdxnfDT(;L0gSctTc3}I#|0twJum9N)JI{{pCbF32^rdkux z7T_R~j~HG!ml(2Fg7s%S6~4>c&q zt*SS@Utv|%QJ1%;Y5}lp%Gn^YE+E3}&f}v<_RP#S3%VcoxmltuvCn zky~-;4cW*-`S9q?r47_HTA2TBrE<`uo{*^13+CsS|2u`>Pz054gYsvkB!x3JJ4?0x3aINVNj>{nHcjg&Jdrt%wx3C`a2g;*0fb?Iqm;+Yjsy zZDa{%%6sF)Mo(ztl#^7E)BeP}Mnj@H2ca=Svu+-yQsu}_&wOKMDO7~g4DG&;lc&|d zQ?Qtf0`_Szuyx=>7xA^fzACGaA|;ISiq<={{0bHjhU<+El$_`bi3Uk2XdQanATk_G zz#(-7LntR7pU5O({4OzQDLYgs-yq@D%(*txHwT@;ojDv>`^ra0R5DC@C;3VmTqvwf zYFYSoa<&lq%KGWuK?9}knSbQ&gaqpg5RAa|7o*N&aqw}cmdB;gI*Q6658%QZmT10; zF|7dH;|hD*IpCKx-JBA4Q|h-2-|=2G&Jn3$zo9cZHGT~D3h%q8-gXOkISKfHZ3G$k zlX&tCLl4w+t+&LeHfc<1RqpE_G{IfD@Z*(Toi*fPDLZP)yYQ%NI>tCQjS%_lZH05! zD}LQ?D;8A~!*rAOb|`LVvKw%Lwubbh2OWg_cT)4 z6IzL$98j+t3#R5VY|M!M>=?zHq9-1;T8k9Y4mh2ue!Iw^VneH#=y801m~2ek)XoOYU&B9Wg#arJ)mCs@C#Ut7V|(pmF{z0R~7EtPT1k26dy zHv_OU#SSk|*KH?4@?5*Tu-B+&rwBmV)EB(7{T`eQZ?E!rc6$ji%>b^g=&A7pim{w8 z67Y(-7vo~HB%cEe>tbc36*t%ik23JND2%2VDYoxHR zO8D6Py`A5|22kV^@c>k(vh! z(*ACNUyL?hXen98`hp7X)I#EpQQW%>9Q?cMw9t7#R_MgK6RQloV$~{0epKo`dBdt?I@n#`#Uw;M3QAV%FZMen^VyejVA7RvhWTFuq)BA_Nlzp~)N>Y5Pf zmQ<4v3`LPP|099RJ4q7=qUiCG&}`6*e6=DLu^;=}tZ%A0CjBp7sP3nvHYJ4(Bd2+> zq}m+tHDywa6<84%Rqc|BuKxFgc3GU`;f0@X&cQ{4(0s%4^Kn(vD;&7brM299!!aizZ$}2mlI4Kd|`5g*72TBnlV7VLV2eWo+T)t zOft(F?IgmpbAUfq9>w%yqUprP?ay7sAa0p{hV4iqm~`BD4GmVgTCCJes4OVqzvBX8 zt#A<{h(8uZ^{I{o$fq?WPKZz)MrA}2MU1-w`UiiIR!m`cJ z!bxAGqwM1BUo(-7*uyOvn}Gs&IRXKf+s)?k0Ed%`&_w!*i&f_qz3J3??I@$urH)|F zk#DOBPG3mRd!<61(dj0kZJ$(0$H_@HgS#)PMknxqvUpr(n3ppshV$b1M z3shLUWc9i7GeiE$J=a#3TGH+@NcBI0V%ppLnT7{y2R>atJjoo z<-8=7AxB-Z!Sk|tyy*+@!YU8Q?zc;}a_8AR~lO3%tuk4&mudkWr+Bs>9;(S~T z3ZAg+SVp@hygaR|>{W&ewhWWW^*?UzpNc@Ps&jsap7#UG!S$gLJHl;TxKM^;>I~&@ z6kpoh0QCtO;u6=dLX9F{`JKRZRwIgMJKclJb%`4Y)_(j*9QM9L`KEu?C8XmEXWYzF zb?ujz#*euCk=Su;kKC*_Rp%fl^7iF>kOwZ>!;6tq($Nj)Bkq!*un*>OW-ab*5ap}Y z4VDoRO0agM=#Q%8>Kq0H>@8z*AH`uTkj!RgF>vLIj7W*cAVPaLBOO*^8Epi&cSn+6^?~muVl`aLCY;{8rnhfP`R%#3X+MN<($0^0vN*=~>)JgDcJ~(UP z-{#h!3C=rTLF=NVe+I7iZ(T#Wge}%UZq!<57<=k zv2bkHVOC6{Vs-&V9cVomxZ6+)a-_A>2khubxq)@)?Dp$G4G`p9uUBGKiZ9E>TBj21EORBgs zA^U+t?N3zWPS=^7#t2c}Jllr;dLy_A;RuIt!d8Tlpy7vywZWoG8O*D20i6d zz!2lv-`qj}oWrQQkGo%{D64DWvHD0MK>dE9{Dv-0`2l8zn~QiVX61_1*q|-$ms(U! zf1Db3Q<;73AJPAtwMsrg>S`3ZmpT_cAzp;%N;j9uvR5Pov(Cg?EV}@>X@!rI%=d`1 zS3f>n@(d~1RYHSB`>W;@nB+usFQ20=N$)EckP&|WaN_E5H|+69&-!mLatoO?`n#FR zoi`@AK|iem4HqRUt#LE&7Gns9R>80ORHF zQ`vWLZ=SPMtRVhSBPrm_aGlA@-=2=)#Xw@!)|G&SK7}{Bj~i2a8XlWn@{fKKe9gR0 zC{`$)?^%%zchKby1#fEzozXPD_qRp|^R?y=3R_>_kl>%ScqG6_oy!7!7uIS0n+D%~ z8+zap%dHeNRiJu^y$?~JN58AtPCsMhsqTP#>IMV!Uf{LX<19cNi8e$x@U&?Y2{}~% z{XKDyVR_u$iA`>OntN?6i8uXb?YdP6nbe$${$a)fY!I4j!bhfj!{sVDlmM+wL653I zTWl*I%@RYC;FwqT%NKn`JB%p(Q+FE7`m^PXl*#@cuL^r5GKNTFE`~Ne{rSmgg2=IL z#cJ3X7X0~$f8@Y>s_r@b`GQ|Bi_Tk*iN%Kx%GRu)W z{9XMqL|)h3^}nbKsCV4h3@qhDp|6`W6}5(gfw2DVB4{9h2(ww?Cye(I=<2LR(VaBv3iw6_SmPeZpdH=3D@Is4ElXGt`mjC< zZ!FOr&q_2gnhIE0V=0uN)7%;(B+!3Qo}lYbC?YIflTuP2woIOV*ipzQmM9&gu!fR4 zl0poL)Cogu*g0ERf8+_o?PS=M%>Ge=tjN;Y}8%NH8omRMttj z%dEKA@oBhmb|xJw5VK;-rfRnsjkP`OZ9FOLlDOeLO^`EeNV~hC|`%Oo;hlvI+&Jb(QlS1HVvnwNI`H8}ou{H!i+Mf)(*%tUO;)1V~NKG)!M)PWXU3On*}o;$c>BLc z)#wz%RZ%C6!n=K=PxtDu#~-IOeEyO&d85GzGfwri13m7tx*Kkky8h`>HCKilt)e05 z=>iW#G}NSxHA(2RZDadSW9=|Zi8(*WjMqy6ZZ%eX!7zH9-Kii$&8->UDR_q`Gk)yN znfSHS43s;L? zZ)xU3OgQUS!__W5+rf6IzRlOaKrRXf#4z2(xNDv<`ll056lB_DPm__>f%<)R5%AVP z=Ww-;*;~BfIsDVYBktmrT;C<)+sVPTOhk@6XW2bLHwbS&tDgyHZpv@#eTdyV;i?^? z+sxx_HorLm!Sa*s{mBNabbVW0<6%|*_CG+Be!I25?0<&O^NRxTf;)w=*A=vHWtyH7 zRB;r0D=SxY2O{QSqsPu{Hu4hjF3pt6 zTQ2RAf7JS6DgYQdJ5RAAEwANzi#6fm#puHAadWdCwRdge*&+xI0c}-$>+b((n$P9g znhl1!;RP~GxX8XN8FPCP_L%ULMibx5%L{R~0pB(;u-!q>u%A z{r*Gw|Az`(gbrlG!WBv(UdN=v{Zz`f3Vdqi)jCK1{Ne{j-YlLdh1{_M59(xCm=nM5 zpvhW(-SM=xoEAxmv^4(ayNC>ZNU<1%?GH7gI z(4hE|eIa?u?>AWwk%Dh|@8&((8O&Z9EXT|U)?>=m8*70SNjqWDc~tG<`Vo&)Y5Q@) zdsAbs$Nq^rOYYI{v?g@ToUKte6#dk3{;4CmXatykF1r5UX3N`2|GJ|akxlJmcsf=7{lhDX`|R@D=V2?6fU8ys{zxl@@^VNqLEg}-nnSMaaum=Rrm;>gV&+dP?@1R& zuUgMk3+&sMFVF5<1Ow8<&^M(F%h`i0Ep;FfIr8ie2pdV!DU-F{$Jq_%!T%(BFpV(RlUX5$m;@GT(0-vB?2?k$%W)*1IXD_pEwdHX!;Wbfr zj)P^26Krr3el}%`#+0@vXxBLnM;(gYjSzGTcFM;V%&KDz4mR9}ieiLjceXdY%e2zg z%KWtICRl@n(j6he?Ws23zh0Xo^o2WxsOHY$PqM5|IPe3rNx@wkSGvQH;?Z{kZ_#g% z)JXBEXUH~?*<2`Pf$$p%9;anDn?v&7%!Qx{x5}xnkKl9*-$G(#u5i(fxo`4{8UB?Z z6h1lzDh1pf|0@b)CQ5D)8wr>0($eSZ;5Nrkoc!SP+O?R>@4051MvwU(G%i2H< z^JL^37tXkMj$48l!BEZ~hk5m%Ks756iO@TbFCymy0!z<&KW z)Cz#N%wuf&Z=>Dm+bEPx?TmEtRg7L8?awh|587;oZ6|>|lWhLAJa{h&%Et-82s^bd z`aR3><3CF9?RS*8oCkJ-qu6V96S(Gm^gQCfF5l9i>XdiW16N%17wI=rIZC;_;;|aG zQUH$WK+QUh#VAdgRThcN!t4PqnEKX?1k>US7Qet1w72;8p$ZquiFkxs+lkc&L@KXC zk^dm4Ur${%wZDIhW%lx~t=URx{8V#ux*)(sXUWUmE=?SEiE81r;?+1?>c|`zxQZt0 z-YIT#z}W0lLSd(UUIMBVks`g;6+Q=W8|zeY8?l9j&tZZaee=}>(5yzG)Of8h4*bv} zyJboyDMZxA1Q7N}tZ^HKxE}*5Cc^~H&fN6>seB#atK-47>wSXa@|P!$*L)i7-F!Rrv8-`FXJkOimhsuQTnt~s^DrVOPl+MNGon6`y0#Xx&u|8`)k<0( z$YpVRcN?e^U2T$G#;D*U;9|b>Q%y<<0FkWVMr+Ca^5r%{>e(J@-IQ9GbU7>?7 z>uDAdn^L@vRD8lTjNGUs!fwk@qXDlBRCgYE!blXx5n(q>FPWjFVclo4Pte zklw`gU6|quO#x2l0TmBr9&m+&Bw%%8y;ie|nWS|P^ptn@PyX5w39sKA4=cLc^E;dN zs~*frZ?=ky1r0=zm5Ay=H~Y1W{6E$Krq^kZm1vlnrk2nchMXnB=m05zphE)M&ZYoy z*Gnf}ncT@!Ou#9_JZpACm3}Q9cvck_tw@pHHkYE9sz9Znr+8O}Iqg#s_E@YErWK-{ z36!>=P!T2raW_jR+7En^&pyfbn#~BGWKlHQ+++oqOqJbrTE)JJIbIO*w=6g%m~+j1 z{LZEfsA#Gvo zcMDMoONvQ}VfKmbD9`GGGvW65mpAZSitJe)NhTx~l(b0Ag@at1D`54l*nOo0^I9xj z1#g8>Sk?ZYqZn0P^HTCWR9$){6|YA`=oteE(5{JnM(NPm`7=+ze#44?X%sWo_2DIl^tlGO2SgI+HNgXD)wEx-Q`~cw`f^4lCP~ z>puZwxp;|@-rY$i6w_cjhN28zp3r8>jWwX| zcOK*~bCKfoM;CXM)<{S3s@SZ!K>c&Y571ip48_!2k!^|9z+MvwN_*Q?3lcLN&xLv;nvrA{k3OUg&$Vu2kPE-HR_DfZ)*G@K*fp2F21KO3 zzhe|=HLnp@Ubg?h`{7R@X}tK|%B;tt3QqhUWTH2lRoK;D_|MmmyZg1y0?gR|=j}~G z;51b+Lx*Y(7>O{Psnuad#+?Wof1?~Qyv;*tl%jvk^<~%f0vGY=SSuTq@$oy9L>i{n$S|57Bn$Y@T^{E1`Tz6y%m3fw-=hWfk2d}W*>_lUi#M)| zG)VA|(1@PQ0800gduoz`Pv%lWtEU8{kfmxu6^qo+9&wUFtGpiy;YdG5yjG3uP>&@V zUksbY1xHRrUuN{)Yr;>V`1vhsU&HS5mEJA-%o`;HN94p_X0+a8qxxP*lrO?D+sh>d z%NCGYIbMWp8D?W1P9(0cLpq$^z^<9;Eefi~f&MBMD= zA@Yw3Lf|12li9^c`VJiM4-#($KdlLCXREuTC#K`=K&MP^{mrdT5btHl2~Q8!hE=hBGArEP3Vc}>!6Ux3n}Kwt;>dff0SGD zh!yVXKz*VrJd^ol>Ww?>DM{>e8HT}<2)o}jo<2XvigE0rQmLXoGy`s(M^q5r{m!GX zBFgP-ZQ`MzX4xll1%wwgZqTQ4f1q0G;AGwwsv-D6HBaboAcoA>)6sBPe=4@W!MJX) z3w*9iAbb%jwWE#GHK*{dpNp=C`3i5GV4@N(Pj57l#^HL?wsUo(#{ly6NCBPUdUKf#h{_c*qP+3Q-p;8pp3r`t{OxKvy`sshmGMZ=9_#%p3n!Sl-@4?q|m$ znlXL9`m^2`Rm;&>6wV*dKgy;mw?I$ak`PCO37mNaTZwcFaYP%uJ~~malCu#UR;7)< zVj6QUXO$;+NaqBXPW}?XFi}8Fpd@-{HUR4Qk^S8~gjvv3X zrMsFW(bdg4Dp8xpD>dwBdg)^3zW=v1S8r)&JFuxnNgjHY9+^Sy9`(E!2j`z*)nHr}WaDXDvUTqpItrW8l!r{|Z6L zbh5&gf+hJqMf1Y%Ho|$&d83Dt$!_%9H$b3RC5QS_6)_gUEM&TRC>6CWzTBe|Ze9?t>faPjIfTaMEdy*6jts=b4 zni4#Gn>$lA_Sm1Ii?_`V8?CSb8?DE-+aN4A1-Gu$iw+-OdS$n)ZSUL-k3q(bY~YCE zPu)j{wT>93sr)^~^|{9pFFul%Tx4;QVYDYF;w;u<^iRn5`=iGLp}EQ3 zus_cAUhmI+r>LcZTatiY>}!Wc(FfnBnXk8PdToN!GC4Ny&*Pcdf(_J(FSiIqEJ=uk zy@Cx&iRAgHqs-0Z0Z_Uor;{?lfYAH%`@2WID6n~^2XsBfdw!ZQmz_KvV|p@;mci5@ zXqR|7jW?ey_zTeMQpGr#7R-m0{`GQed5A^NaWcI`gOk)ofrEW#Gs+x-`f1*Ua$eGM ziRif=@bQKJ)41S|4#&?X1D{Pket6z}_s|1p`0|?YlfPNFo?d%Hh2I1DViec z$N_$(ckarsrxej1*#AaE;0_@I`NV~3Om%%2!3-XpDePSm@5}fP$9xg|$*e3Ksgx|At?Tr))p+4|n?0WyLZ$!=T>lYK2j@H#V)9lFNI(Ak*862r3v(`GZ zzeOClI7ww1#y0GkK%_)Mso6-Cxo26K+~8k)M|reHIL~3K90ApAXu9<36Wu6Tlu1CzWI7w#}L3??-VatMjfQb z5Yn$07aIK|kxvTG6VqvwT=wLN>M^p){~;+-P62M1DNkG>0)StC$(Qa3lzKwu{x40X zMF9q16RglaAdKz*l1ilL5hH>+Fe9qhgJK{uRoRVE!`LDME%~ZZGaw;VjU?g-%MsT? zPAX_qR_guw%D!w_EI56sfE=Rcw$;lZ~LV<%^eQO z_bUv9C73YPW~@)o`QPh5_QZ-4w)?&2xnp8R{hm+jPlWRHRRK#*89QT{k1xZ`L|q<& z8S=JW1M*>SlfJ0^Zdh+=iuCL9Rf2o^zqMnS?;HTf1x^R$#oY(Ya%^o?mVLO=wbX@4 z1-Q_3MU8b=jN{sqrag^=Y*Y3^^I$F4MB|dm1h#W_9_1*oxott+;D3G(nA?MO=}8Up zwn~+;=dE(6_2XD<_oXTCJNHoWHEc2|Rv59&l?8!@LRpkL_JJ28AS)1vZow*3r?hw! zWDOvuz`rdT{K$itMtEw5k>%QHM)+NuTW->S+O*IRXv3LgwNN37F1f>~ztcbkD{q5u+LwUS(&2ns!M3s&oT6v8NTKT66he7l2e>9V&YG=N}u)I$MOwxn|g4iJn6;BG# zpRVBAxY-CpdyAuV{-8XuORe-VuS)1IB|T7XCp*uTW?=oJ+oW@3kFBuAk6*oOSbB|Z z_&oaWsOJimW!}iN(Uv^=%sQtHC}H)YKosYSQu|#;-L#TR`p+Lmt%?91a0L z_Sy$hccn3z!ZV@4|H?f-neYG1v*IA{RK|VO`DhuQ;DH1lf9nG2YBh8Z(M!yK7dt*) z2yn)y>Fk@ol$)bpG_B_x8SD_A6Y|;rxqx@G1=V?732m^G@=K8;Pbd3 zp8R27M#*qVGP^w#|3U-6-#1epuYH8{Lv*C^n9ihZ!H$ofA(^fYUvPs^mQp~0aipQ6 zS*lZ!qEoAvMtOCZRyNGZ*;V)kQ@W_4uSb2A%X&_p-Mr_0oY}hfYanIS@7Z*`YKlKN zsXzC^eX!TKB=_paiLysqy|IWyJB^lnnX?d9rn|e}30Ko(9Nk@c<|Bn2OR$pqJH@;z z9?p3ULWX{(DQc~-?}3Blz^;JuE27mAln-KXwSbr{@p0^Gy2F6XGBMl7&iFs~KmIzl ze%~xmm7Rc!>&VpIJhNIjm+~OY?6cfs)>W#zsrO6U5Fn|S*BOAZ3Iy^9lLQ1MTzPQaqrnZ z@Mcsy%y~t|V%>az;og6B1-K4053J{-R1zF#=onVlG3`_k%1gSc2UQfebfpmN?mhzq0f7 z8x@&Y>fbE#81XiK$H`K@St@1KS$a)^Ur84y%Ka(JMYQsKD@qeUsjWqkKw2b)8~gqK z@%2gI5A+3Ev!#v!(QAXb7MdK{C!V7OrDWOzB*6ZMe{yJklwj^990TT)Ga+skd!l_U z)0?z@qRNNN%$;B}Ki6PyU1@K<-U~_GY{Pr*f|aU^CxsH7Qn}~h7U#S)rd3d5@_hJw zS)o;7Qud}20%|$XX#@$Hr>rHdEmNyFYa5hCG5W*|gb$Qu8mWGNjw_gJvFcV?X)Y60 z5>rwwRVh{4P%4Bi?OT=kyk1Z$r7c%shj2G2cR5xANtH^LmZ{RV+7`|ZOkHVs{dc|k z9YVtsTtP{y{J{$3T~1q|TA;dM1yX8OI~Fdld58zp;?x=!v~xj^Y8JW#nyQUsO{C34 zm30&{Tcvem|2%#jQ+T{iRSV`Y-w_=sAbXOYhAbsj zCxih|H7`%l$Zxjeh+Pz7JxyFnB%{mfHskQ)x~l|uNb?Cs6|s&^$HwCpop{PdbR`@` zG234DLpz+wC3F+XIy>->8J2NszVog-to=KFS$X(3cQ4|hAYpqo^P8(iJm3#bL~Xrv z>)Nr+zp|!WnowJ4Ol`;rYm2Cybh}+D1e*^)S+NDm^F1HCmTUal!pIK4sBH3a3ihh* z_2PWPwR+>N%KKGq*KMiEW@FmU(H5n6EgGH}n!Bn6A%__u3{KVavps)Y}bX zD_jY4D}0vUo2Z5grAs^Pa?D2P^;)ipxX(@fTGt3c|ER=gWKd+@ma6ffUNL>0P1FX+ zbA>DQ={XbruM+fGDN8AOo~6_d;Va?P1Mk%%`*(1~zPV{eXOoiYDz9CltGENp`zD_` z#fx+i!;`T2GWjhUE2&)ylIA6NiK46At%@_(JU zEH!H?U&$np6m-%hy-{|+W?Nzv<5HHlFJGXX_~nG}yI3^SY$@Cy8fZh!O#kXT^^!7a zI5A?4xE(?7 zq)G@T=MoXjd0OpXQ@2#ENDl!H+{58ltwwf_n|PF+ZOn7oN#|%2qNc*(u^ZK?3;`4( zC2^#v1SghxA|f5s#daqY5Aai1CsJ!rM~L&FPN^{}UUui%;_x%y0{Q*Na^38aV~Lk} zO&_JPm~3=hK~>W`G9S`nK>z$jlu&TCc>-BIf66g1t_>|L^Oi^*|xZWPmrt|EXDqS-UY6H!(W6p+){0iCSNSZE2 zfcyJz-mu#5EEPZZcYg^N?B^EdA&!cy{K+PdW5|+vP^cxOQ-+FOK^crEX*eK{NEWxF zC?{nilEQ3#o}$w8{9 zy?KiJOwYRPpasN&1(j%&N_|UIRMkGuSYI>M3r8L$XvJt3&_HD)m!D_R-M3}vssAJk zB@5efl*(8PaNmkW0JUbQ32=99*SXQxAH`*=Mf}ZDFV2Qtg2o4uP}{NN)g*Ln6RMJp zSH*?b*-!oGcq7_YFi77q_bd;cxP{f>Q4Z#ATwGKE71KIp%bx=A1UDC|RJ{B1WGlyh zXC-PY{9CH!M|sMYdHkpZj$5i!-}f8IW{&;-YG`TB)HY;A{45;3m2@7FRnPnz>k=(m z43|d+!p!vj^|KM@TGD9*S3UDE7A2a!7$OZ~ho+S0Dk`h|XQAbVq|3;OTINY?ZInte z92r3)a3?8o{#W4mhatp2<+R(1!3(PWvVg7YynPm#%J84RTDrZ4V>ZIwWc0T$O`w8iPf-JhR1ikWf8mygGl19hGhr#|!L;nt1?>icQD?Rs%B(RU|C zchX7Ya!^JvcM8?Cj>~a+aXS}ho?@7Mv*j`~blNnpcVlWVRO+g!w~|#OqH7 z3xgvdJ*_HFZafW8dCY#etgm@8UEO<}j&7t*s%f@fFPH1F>i&+YM*>@zn#*Pv^UdWO zKujhST@V0qGVB~3>khw#c=4k&-V2eUrK%@WwIislQN&Ws>gkIrER!KP!MEYe$v!Bo zvDG-xpa!|y|0XC`uC362+D_=dysKaI$*<#R4@N%xwh6wh z>tnIS&A5yo0@A=A-a?e1FL}p4Q`!JKA}g$k%OhyD&0fd7B;X6 z9TTr1Vef>VpPRS4s5K?rMlSopSaU_q(n&=j*1h{#MlNsqHm#YP^0qdFo@33}B{$R| zGDw{gezGb;&nwNn4i=00t*Io?YTsFx^&I;$hdVz|Q<2Eph}_F@^{!Hk_Sc3|^pi2h})aLX$^J>IgFVb&C?6bowf}ra=HUq`2}g#SOPD z=CLIX0ax@woDHdr0PXfQa9PW!+9pDNj7l+HCZ9Y~CfYWysmm&TgwHuoIIdLaB^ghz z__N~0Buqz*TqYAlfkUeCaNS<`sF6auVIChC7O%fRC2N90W_VJ?yrQ;pB-6B@uUqEI zBt1(p=BS#@7}p-)hGH~ae_v9a~Q`AxM-v#&h`~39MZ;E zzO1Y-+LuAtp1yjpi|aaFtcw--Bo@+!*lB|MlChBdID2U)+VpV|$!ZK%LcW0yd<*FS zZ)Tm3-91s@gmrk8QY4Hxo9uUX#xOZ%Af#c!h-+u#DM5@)*U=TdfFpJb%kWQh2_M-= zg;$0eksG2SbxR%vbUpQ5G#6>g(QDh;a0(D>(mR9(E*R_EY26g2lS8AsDI>UIz>vjoJs77r}^b`(W*1J*f zDf+eu+=71RpiESzkx|vPD7f#KyK5aEU_ODH$_5ZuIfL?ms76aAvj?g#uif@ zZjbLP#;+g&|EuDdm>mjB3+ZFJCk=2t#o(mWF8>7Gt8jt&IlF5T^bRBV`#Wu&L{$7;cz?+7L#)b(MXh)4x`r@vGw%1o%Q<5lc5h~*D_h-pH zL|84TJ0XTG-?GI)4JjJzrF0P)I4|}oS*((qiG+iYL{wuEK0aeLa zny3^rcEd=eB6BCi3EyKs%Y25WHWPn6Ik3BFCxh~zAm_j8PeBD@lcpZQ89$rMk<1FE z6{+3aZr(&LE37DM^D+IeC%df_ujwzD$dlIt==B6o~4 zc+Wg>gJEuB!>nl}s6yvf8e0hM9F;4xHUd80S|QjB&| zC4c<}@Lbe!Jb2!!*^U*)2C!yqQ0lQ`#j9G9wo<+~uIoie?tbCWtE3yIQyne->goNC z!!~0&G~Vzd5De|YqbudP1(Irg|7(M9;=0Z`%RyRMl{|)xNhJM}Lcz7WHFX-PRGOh2 z<;}o?%0eAg5s62BEo)j|H(-ja&<`^4-~$ck?}6gEI^a+B3pSdr60pu&?!o&~7qn;A z`e2okgD-gIlb0aOcd_8(gL#p52u}nzkIZOJU9+_~UfY0y`Ytsh!pl?~> zX=NO=}RfX`NMIs?Y+EU`j$fmz z)=)Z~tXOdm%ZH?-W0I1_OHi`0FjQMGFGwV-%}Er{|D2rut(Ukcuon(e9fzZ0YF>mX zo)_3*-v?t$tO_-s;X8Wb`3yU>&VKe{=t#pzIb2~Jr)qIpnae@#v-A>*+z=yqF?mj(bSB!F&|M#sxtluDcIYu5WU z&H8CcLBw#MIgdsXYgom58UOBA(w&f^AnfF8@QI|j@Mh#}w7DeW;LejP$Do5Z&|)fK z!_garAXQB}{rt-cG45Jf1yyCTd3?E4RYCX0DxZhNQY*Ypc~_)P`3Dl!%%|(Jp#fxh zw+sG2K!K>{D!(Tth$dF_?u@joTI}xB8>dzkC(eSCIxXA;6M1sCso2NVyF&7UM~_Pk z#os5Ax^GE~GW_O`#ZHAal9-|(b-4YIe2e})f6hklFusOLEkresatYbgeAEc>dL%vW zFU>cpMAUwz|BdeMYFPLaL)z{u@ExBp68nM5I({{%Y{HTRaOzVpw<&I7KMm^2xo%DY z(AkbZ_WQPfN*tcI^ELu~H}BK#e;XI^xzvk!7J{9pbxt?&s0o^A3EnEM<8`ZHLzDX1 zO9RrZ(2@IgW5}`@&WwYuTOr((sY%|3p((s8U8Glrj+OI6=8A_%3gIF2SZWa6M**Y5 zKGjKYth*_B!psASP|--J@Xi$HdX$ZZ#kS|9WQ=hT1|sxwS%)MS-n?9^ACVd(OZJPk zAyIkyG(k&x8x7H(*FoXp_YLd4F|5)MGs5F5XeZQO^J2p4Mm7L(L7DG~GSga=>U{m_ z=6>l?Lya>sS+1$>vROYj_M3uq0RTd$Aga#o|pcz4@C1>mCQu_>zA_{oNGXO8To4O|UQW;D9>|3q7S$?>E&y%m?7K z7MNtWiJ`sFc;0L7-j zYlMf2+H(tMzz8|5p6897hy=)o#~*wz z{2J4}NT4~IrC)CbgvUGn^!EU-J$g|Jr6Wi(SzSm`_&lK(b1G$wUk4Z>A<`L&6TSF) zTUV`IS8h^cDGTMK=^zm5KWaL4=nD)z{ z?U-Pp(7)svd;F1_Rabz;hB0rVGNE+aTZf|~C#K32D+N#yo5U@_<)RvF#3QWX;SgVg znd*JMvwkX+94&Zjubp_G;gQXMpn4O(Pr4Pb8 z_XdUimkPE=o76ZQsve7TEc?P+V<`q{0z7SD(UGc|vl1^iTS=7P)6XIW-1nWo4 zpwL}SN*e9crKl=3MjGExEZ}FblpoIsJwko#kUAxe(o9db7mF(gT~P&jIp_V3>UUP@ zB59Er z$A62{dn$O^fNL>$Z+VeI8c5mB8{;6P#r!um?)M#W-eJTA-!V0E(@3u9Lpc8(QkiL1 zpSr6(g?cW({&A;7K=ISa=0 z5`8oGz6oXoIp*VPOE-QycIf{>7|+8oWAFhE`1OckX`%1cs67| z|Ni!SK!{5N-$s>PFXs0uo^fe!F(18>cFzGjs^{AgDeNGfr>;A;VpXi!*)$=}#^oTF$(T(~s*}hwVPeE1n;vwUi9wQ$b2!mngBJyt2>NSIZdxKNQAsi&DW(+Q>3_=xia}1XW zfd}O_#u~Cp9{ZwR)_%!&AN^1ApNG8Lj?b&Eas{%ti|$&Qu1ySC85y2tj9lFvO>}i) zjI@a+$(k{l=%y!`nyl%{)KE;>QpcujYqC30Ttzwpzu#|`)E68;P~0!EL`5}yf?N31 zEqv&ImM7(f0$hH)74`5IqjU;p`5xTxxmQKb@$aAh4gW1migl}Q(iVr2xoG2nkEhKC z`EwJdsYC1KHht5oekEwR51=)G+l&#?o$~^ zMaUYU==TuI$7cAsIyI5AL2}~rj4NIg#LQZMnF$077!NelaC*2>K=R{u36087DvCxP ztOn5dZItB&JwF@Q=O#Ed2uGX@z7l0YbgY$*BFIcNswSg2aEH){At;DO5v>N$`19|X zJ59VOH)nEwMv5ubU3|m>fv`f!h8tIlQ#~oRuJObSV1@1E&SS>Qo5Nd6Lp5K0+A$V? zV`pV&dr>ENFM*<%to`aql_8wgiqj|2W~w?gcX0K^Rx4oFLHNMbmp3e{-~Q}z!0-_Y zli7c`0J7+jS2moDRAu&q3tz?;tP^&w%wKDlPz_NnifUc}4WRmW9*`+$Xo#OXIrU>3 zBrv84ze)>($XhFU(wb#FpzftQbQffpWiox(D&4(Y#4JH`4 z?>mG-y#0bG6p?DeWE6Sm-D=2Yv_3gOu|X(W+-M{UgUDDb8Pg~;*%-a5XUp=?8c2B8(a`8*QAP7}+2rZ9W&76$eqW-ucKZ&P>*ZcExj&*|EDsN{B*H zQu2V&qMv;;l~_@*Qief#58Q8qF02l*b*gw_$aMVBKWon zzO;6!$MnrsJ$1O}s}y%hDb3*bGFL&Jgzl;rOq~q(3xPQ;p}H8Vw{y)^r>98t;!`9# zt-+$6dduKWtF!vtRbMUZ>M(+N$cnCLFld<<;kQNjr3GO9g>Slctkb<;xp?narnO|J ztR*8@Hx6rBW!6(tDF@BWR z#1VK~23}e%*K6}!w#=@(2o|9!x|+nGWp{+(7GjtdhAQrRu4aFW zpIa~B&l;EdJ>MX^0BpIir42tyOHT;X9l$g#Na-t~v(+dU0cRe?LTrXl(z28~m!)(8 z>$p|w`c5XH$B)3omg#WQK5liMo}_zejgHX)TFcUVLD*V;auLd)6jlbMDP_21Cmwb3 zTL5?R=k=54gm9XFC_gj<^L+%^2|sxemM#0k>kqt+#4dauOO?=k6q*%4^D~j9N%F=C zjZERsE&0FrlTCT3;14bRQxYH1Yo8u}t{HY=a=b92r85{w&$F+HK|AC0qTlhiY&CzI7ZnQT9Q%6su#vIP6-sbr5uVC#A0 z1*k#sodX)$IG}4;@u+*=Frq0>UJuc{N{@FD9dxB|2VJ2od4%Iu>YZ^Oq8K5kkE_5Z~R`kOkx1p z>XbJu44BJ)`_+^Nk|d!h0~_#MW6%GD7dO85XA=g2@_3sJmh%zRB{(F11Fnx7+Twc% z_07o$$3&1b@Af`i#@>Ah=Y0sHkKP1g<-(15>>A62sW*6m?nLa}iMJMyBhL%qA`&Ef z7|s%Ze<93wsl_bLEO*clJha?F*Mon1^Y)etKs4U<^)w6%~ zL>a^0+?)8zc|p~H8hHl_KC50d(%7(t8+cGIQ$K?GA(UjdHh6|TR&?OJIUP-8d!*Wy zJF-U;%NXmXZB10qR8-rr6;0JN#WiHjlH`$pWrKv+w&u!5Xah($M+-Ev@p6V&o-5xk zSQZ?{vxK_Iy7E2%y*3MUz{dGD#L{NNK4ni*unb-)7HSQA%}KwOZ_C3)aCfAz96aWx zb`)spOt*&W%S!d3C}5t*M7Zz*XXC{kYVkw9n4Z^#6snJ&8x<4>3`aqA5Rq%CKA35L zD=n`NX&t`JYB7mHr^Uc(SPWthD>A7;^J?(4VI%A89q8x5SqEnD^gh;(f+NIXCtcdA z1a@&HuwP!rNFhUO^0?q({s;kDLoo0@V_T4UQvg5!E#MVRHY*)e9Xwx4MD%Wdw8#r6 zgrJ~)if~<|eqy@_gNqhV@hAo_WWBixDJm&yRa2Gj93M!MoL;h=e!n6IYNsSsYZ+K$ zshvWJT3%ErAvPqGT^Gt9-PA$weP#KLNiaz;NicgOm_L3Gx1m(4lRtjnPJL90%cDZ- z*nM^U&cC~7X{YTgeuy{eA->`*I3}$mtt71^tt@PV*;P$>y{p#uvg0)E6jZqL!itJf Q`p19!|1E{c>0=%T01QxWb^rhX diff --git a/tests/test_signs.py b/tests/test_signs.py index eeef21348..c87261c80 100644 --- a/tests/test_signs.py +++ b/tests/test_signs.py @@ -3,11 +3,14 @@ rule, and carries the value in that orientation.""" import glob +import gzip import json from collections import Counter from pathlib import Path import pytest +from hypothesis import given, settings +from hypothesis import strategies as st from scorecard_db import signs @@ -71,6 +74,93 @@ def _row(metric, convention=None, *, country="UK", source="x", annotations=(), * ), "unknown", ), + # borrowing conventions, verbatim from the Autumn Budget 2026 harvest, + # whose totals of the Budget's decisions take this metric (its README + # rule 10) + ( + _row( + "decisions_effect_on_borrowing", + "Note (verbatim): A positive sign implies an increase in borrowing.", + ), + "same", + ), + ( + _row( + "decisions_effect_on_borrowing", + "positive = increase in PSNB; negative = reduction in PSNB", + ), + "same", + ), + ( + _row( + "decisions_effect_on_borrowing", + "as worded: positive = reduction in underspends / borrowing increase", + ), + "same", + ), + ( + _row("decisions_effect_on_borrowing", "positive = cost to the Exchequer"), + "same", + ), + ( + _row( + "decisions_effect_on_borrowing", + "Note (verbatim): This table uses the convention that a negative " + "sign implies a loss to the Exchequer (and is therefore an " + "increase in PSNB).", + ), + "opposite", + ), + ( + _row( + "decisions_effect_on_borrowing", + "positive = yield to the Exchequer (reduces borrowing); negative = " + "cost to the Exchequer", + ), + "opposite", + ), + ( + _row( + "decisions_effect_on_borrowing", + "positive = raises revenue (reduces borrowing), as RF plots Figure 10", + ), + "opposite", + ), + ( + _row( + "decisions_effect_on_borrowing", + "as worded: positive = borrowing reduction", + ), + "opposite", + ), + ( + _row( + "decisions_effect_on_borrowing", + "positive = increase in net tax as a share of GDP", + ), + "opposite", + ), + ( + _row( + "decisions_effect_on_borrowing", + "as worded: positive = additional investment income", + ), + "opposite", + ), + ( + _row( + "decisions_effect_on_borrowing", + "mixed: row 551 (negative = loss to Exchequer) plus row 555 " + "(positive = increase in PSNB)", + ), + "unknown", + ), + ( + _row( + "decisions_effect_on_borrowing", "absolute average, no directional sign" + ), + "unknown", + ), # households, poverty, bills ( _row("average_household_income_change", "positive = gain to households"), @@ -114,6 +204,18 @@ def test_declared_conventions(row, relation): assert signs.classify(row)[0] == relation +def test_a_borrowing_total_without_one_direction_is_declared_unknown(): + # a total adding rows under two conventions, and an average of absolute + # values, state why no sign applies rather than going unrecognised + for convention in ( + "mixed: row 551 (negative = loss to Exchequer) plus row 555 " + "(positive = increase in PSNB)", + "absolute average, no directional sign", + ): + row = _row("decisions_effect_on_borrowing", convention) + assert signs.classify(row) == ("unknown", "stated:borrowing:not_this_quantity") + + def test_an_unsigned_magnitude_is_magnitude(): # the issue's example: HMRC "Change personal allowance by 1%" row = _row( @@ -136,6 +238,56 @@ def test_a_result_oriented_to_its_claim_is_same_whatever_the_text(): assert signs.classify(row) == ("same", "result_oriented_to_claim") +# revenue_change counts a gain to the Exchequer as positive and +# decisions_effect_on_borrowing counts more borrowing as positive, so a +# convention that both phrase families read must read same for one metric +# and opposite for the other +_DEFINITE = sorted( + { + phrase + for family in ("exchequer", "borrowing") + for phrases in signs._PHRASES[family][1:] + for phrase in phrases + } +) +_FILLER = st.text(alphabet=" 0123456789;,()=-", max_size=12) + + +@settings(max_examples=200, deadline=None) +@given(phrase=st.sampled_from(_DEFINITE), before=_FILLER, after=_FILLER) +def test_one_phrase_reads_opposite_ways_for_revenue_and_borrowing( + phrase, before, after +): + text = before + phrase + after + revenue = signs.classify(_row("revenue_change", text))[0] + borrowing = signs.classify(_row("decisions_effect_on_borrowing", text))[0] + if {revenue, borrowing} <= {"same", "opposite"}: + assert revenue != borrowing, text + + +def test_published_conventions_read_opposite_ways_for_revenue_and_borrowing(): + texts = {r["conditions"].get("sign_convention") for r in _feed_rows()} + for path in glob.glob( + str(REPO / "sources" / "harvest-uk-*" / "*" / "claims_staged.jsonl.gz") + ): + with gzip.open(path, "rt") as f: + texts |= { + (json.loads(line).get("conditions") or {}).get("sign_convention") + for line in f + } + texts.discard(None) + read_by_both = 0 + for text in texts: + revenue = signs.classify(_row("revenue_change", text))[0] + borrowing = signs.classify(_row("decisions_effect_on_borrowing", text))[0] + if {revenue, borrowing} <= {"same", "opposite"}: + read_by_both += 1 + assert revenue != borrowing, text + # 31 with the feed and the Autumn Budget 2026 harvest alone, 32 with + # every harvest and the PMD feed + assert read_by_both >= 30 + + def test_external_value_pe_sign(): assert signs.external_value_pe_sign(2.0, "same") == 2.0 assert signs.external_value_pe_sign(2.0, "opposite") == -2.0 diff --git a/tests/test_uk_ab2026_harvest.py b/tests/test_uk_ab2026_harvest.py index 44d1cba32..94ed56919 100644 --- a/tests/test_uk_ab2026_harvest.py +++ b/tests/test_uk_ab2026_harvest.py @@ -11,6 +11,7 @@ import json import re +from collections import Counter from pathlib import Path import pytest @@ -24,6 +25,7 @@ ) from pipeline.ab2026_proposed_keys import render from pipeline.harvest_counts import counts +from scorecard_db import signs ROOT = Path(__file__).resolve().parent.parent HARVEST = ROOT / "sources" / "harvest-uk-ab2026-2026-10-29" @@ -116,6 +118,10 @@ def test_table_61_stages_every_line_and_total_for_six_years(): ) } assert len(totals) == 3 * 6 + # README rule 10: the totals take the borrowing metric, the lines keep + # revenue_change + assert {r["metric"] for r in totals} == {"decisions_effect_on_borrowing"} + assert {r["metric"] for r in measures} == {"revenue_change"} def test_table_61_rows_carry_the_written_out_key_for_their_line(): @@ -176,3 +182,114 @@ def test_table_62_rows_are_financial_transactions_not_revenue(): r["measure_key"] == K.PREFIX + K.TABLE_62[int(r["conditions"]["table_line"])] ) + + +# --- README rule 10: totals of the Budget's decisions ----------------------- + +BORROWING = "decisions_effect_on_borrowing" + +# keys that name only a total of the Budget's decisions or one of its splits +TOTAL_KEYS = { + K.package_key(s) + for s in ( + "package_total_policy_decisions", + "package_total_tax_policy_decisions", + "package_total_spending_policy_decisions", + "package_tax_decisions_since_march_2026", + "package_spending_decisions_since_march_2026", + "package_non_scorecard_decisions", + "package_ifs_announced_tax_changes", + ) +} | {"ab2026_option__tax_decisions_with_continued_fuel_duty_freeze"} +# the OBR's total, which also keys its Table 1.1 lines by head +GOVERNMENT_DECISIONS = K.package_key("package_government_decisions_since_march_2026") + + +def _key(r): + return r.get("measure_key") or r["conditions"].get("proposed_measure_key") + + +def _label(r): + c = r["conditions"] + return c.get("table_row") or c.get("aggregate") or r.get("reform_hint") or "" + + +def test_rows_keyed_to_a_total_take_the_borrowing_metric(): + keyed = [(fam, r) for fam in FAMILIES for r in rows(fam) if _key(r) in TOTAL_KEYS] + assert {_key(r) for _, r in keyed} == TOTAL_KEYS + for fam, r in keyed: + assert r["metric"] == BORROWING, (fam, _key(r), r["value_raw"]) + + +def test_a_totals_lines_by_head_keep_revenue_change(): + """The OBR keys its Table 1.1 lines by tax and spending head to the total + they add up to. Those lines keep revenue_change; every other row on that + key gives the total, its direct or indirect effects or a part of them, + and takes the borrowing metric or, for the current budget deficit, + proposes fiscal_balance_change.""" + on_key = [ + r for fam in FAMILIES for r in rows(fam) if _key(r) == GOVERNMENT_DECISIONS + ] + by_head = [r for r in on_key if r.get("metric") == "revenue_change"] + assert len(by_head) == 288 + for r in by_head: + assert r["source_table"].startswith("Table 1.1: Breakdown of policy decisions") + assert not re.search(r"\b(total|effects? of)\b", _label(r), re.I), _label(r) + for r in on_key: + if r.get("metric") != "revenue_change": + assert ( + r.get("metric") == BORROWING + or r.get("proposed_metric") == "fiscal_balance_change" + ), _label(r) + + +def test_a_measure_takes_the_borrowing_metric_only_where_its_producer_says_so(): + """README rule 10's exception: the OBR's paragraphs 3.20, 3.21 and 3.43 + give four policies' effects on borrowing, and IFS gives the Budget's new + giveaways' effect on it.""" + outside = Counter( + (fam, r["source_table"] if fam == "uk_obr_tables" else _key(r)) + for fam in FAMILIES + for r in rows(fam) + if r.get("metric") == BORROWING + and _key(r) not in TOTAL_KEYS | {GOVERNMENT_DECISIONS, None} + ) + assert outside == { + ("uk_obr_tables", "para 3.20"): 2, + ("uk_obr_tables", "para 3.21"): 4, + ("uk_obr_tables", "para 3.43"): 1, + ("uk_ifs_ab2026", "ab2026__package_ifs_new_giveaways"): 2, + } + + +def test_the_current_budget_memos_name_that_deficit(): + memos = [ + (fam, r) + for fam in FAMILIES + for r in rows(fam) + if re.search(r"memo:.*current budget deficit", _label(r), re.I) + ] + assert sorted(Counter(fam for fam, _ in memos).items()) == [ + ("uk_hmt_redbook", 11), + ("uk_obr_tables", 6), + ] + for fam, r in memos: + assert r.get("metric") is None, fam + assert r["proposed_metric"] == "fiscal_balance_change", fam + assert "current budget deficit" in r["conditions"]["sign_convention"], fam + + +def test_every_borrowing_row_says_which_way_it_runs(): + """signs.py reads each row's sign_convention as same or opposite, or as a + stated reason no sign applies (a total that adds rows under two + conventions, or an average of absolute values), never as unrecognised.""" + for fam in FAMILIES: + for r in rows(fam): + if r.get("metric") != BORROWING: + continue + relation, basis = signs.classify( + {"metric": BORROWING, "conditions": r["conditions"], "country": "UK"} + ) + assert relation in ("same", "opposite") or basis.endswith( + ":not_this_quantity" + ), (fam, r["conditions"].get("sign_convention")) From dfebe9c1bfcbea9cf8927f88a1bf02dc03fd21b6 Mon Sep 17 00:00:00 2001 From: Max Ghenis Date: Wed, 4 Nov 2026 07:49:25 -0500 Subject: [PATCH 17/52] Answer the delta review of 8b7b23db: rule 10 by PSNB scope MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit - README rule 10 gives decisions_effect_on_borrowing to a producer's stated effect of the Budget's decisions on PSNB, to any part of it the producer identifies, and to figures that restate or recombine those effects. A receipts total or a tax-change estimate keeps revenue_change unless its evidence establishes PSNB scope; a total key or a package label alone doesn't establish it. - Finding 1: the OBR sums its receipts heads into the policy workbook's total of receipts measures (6 rows, cells I490:N490), so that total returns to revenue_change. Table 3.1's direct effect of tax measures, the same figures in £ billion with the opposite sign, keeps the borrowing metric. - Finding 2: IFS row 129, IFS's own estimate of the Budget's tax change at the forecast horizon (0.2% of GDP), returns to revenue_change, and its key leaves the test's total keys. Row 76, which restates Table 6.1's £6,970m, keeps the borrowing metric. - scorecard_db/signs.py also reads "positive = increase in net tax" as an exchequer convention, so row 129 reads as same on revenue_change. No exported row changes class: 29,894 rows, including 85 PMD files. - Finding 3: against the f4c7c94a classifier, the current one recognises the conventions of 2,527 more harvest revenue rows at this commit: OBR 2,484, RF 42 and IFS 1. At 8b7b23db it recognised 2,520 more, not the 2,576 the PR body gave. - Tests: receipts totals and tax-change estimates keep revenue_change, the OBR's lines by head leave out its receipts total, a revenue case reads the net-tax convention, and the cross-metric test's comment gives the new counts, 32 and 33. The stagers now move 165 rows to decisions_effect_on_borrowing: OBR 81, HMT tables 18, TPA 40, RF 25 and IFS 1. I changed no value, and no sign a producer printed. Co-Authored-By: Claude Opus 5.5 --- scorecard_db/signs.py | 2 + .../harvest-uk-ab2026-2026-10-29/COUNTS.json | 4 +- .../harvest-uk-ab2026-2026-10-29/README.md | 2 +- .../uk_ifs_ab2026/NOTES.md | 4 +- .../uk_ifs_ab2026/claims_staged.jsonl.gz | Bin 17278 -> 17266 bytes .../uk_obr_tables/NOTES.md | 8 ++-- .../uk_obr_tables/claims_staged.jsonl.gz | Bin 111510 -> 111493 bytes tests/test_signs.py | 8 +++- tests/test_uk_ab2026_harvest.py | 39 +++++++++++++----- 9 files changed, 47 insertions(+), 20 deletions(-) diff --git a/scorecard_db/signs.py b/scorecard_db/signs.py index daeff9aee..d40f0891c 100644 --- a/scorecard_db/signs.py +++ b/scorecard_db/signs.py @@ -105,6 +105,8 @@ "positive = fiscal tightening", # the OBR's Table 1.1 workbook legend "a negative sign implies a loss to the exchequer", + # IFS's chart of the Budget's tax change at the forecast horizon + "positive = increase in net tax", ), ), "spending": ( diff --git a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json index 260f396db..33432f8dd 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json +++ b/sources/harvest-uk-ab2026-2026-10-29/COUNTS.json @@ -43,7 +43,7 @@ "uk_ifs_ab2026": { "rows": 144, "primaries": 14, - "claims_sha256": "bb1ce115de1d141db8af91dd84ec8f94417c051f6994a85fe2d91220c19704b5", + "claims_sha256": "bb7efa637a6f748fe94d43592c0847ef9d759f710c513e5ac4a311dd769d8ba8", "sources": [ "ifs" ] @@ -59,7 +59,7 @@ "uk_obr_tables": { "rows": 2797, "primaries": 3, - "claims_sha256": "abe826475c2e5da312fb55a69bbe08b5e5dc71ab12e3b655b8923ff782cdd23a", + "claims_sha256": "6ae80eb0cbc011fa8dddd6aa5fd8eb8f6cce3b675790d6c79be04907ad92fbeb", "sources": [ "obr_efo" ] diff --git a/sources/harvest-uk-ab2026-2026-10-29/README.md b/sources/harvest-uk-ab2026-2026-10-29/README.md index 592b218ea..4e734ea42 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/README.md +++ b/sources/harvest-uk-ab2026-2026-10-29/README.md @@ -45,7 +45,7 @@ The AB2025 harvest README's rules 1–10 and its row contract apply, with the AB A measure's own parameters are its design: rates, thresholds, payment amounts, dates and durations. So are spending allocations to programmes, financing remits and contingent liabilities. They go in `reform_hint` or `conditions`, never as rows (the AB2025 red book family kept "£9.90" in `reform_hint` the same way). A worked example's inputs follow this rule too, and its computed outcome is a row. 9. **A component of one measure keeps that measure's key.** A printed part of one Table 6.1 measure carries that measure's key, with `conditions.component` naming the part and `conditions.row_type: measure_component`. A part can be a fiscal head, a sub-measure within a line (the tax-gap package's items, for example) or a nation. Components are never summed with the measure's total. Proposed package keys are for scopes that cross Table 6.1 lines, or have none. -10. **Totals of the Budget's decisions take `decisions_effect_on_borrowing`,** whoever prints them. That covers the total effect of Government decisions, its direct and indirect effects and any part of the indirect effects, and its splits into tax and spending and into scorecard and non-scorecard decisions. `models.py` defines the metric as PSNB and keeps it apart from `revenue_change`, a receipts line, and `ingest_obr_policy_effects.py` maps March 2026 Table B.1's totals, its direct effects and their tax and spending split to it. The AB2025 harvest kept several producers' tax-package totals on `revenue_change`, the metric its README gives exchequer costings; here they move. Single measures, groups of measures and a total's lines by tax or spending head keep `revenue_change`, unless their producer states the effect on borrowing, as the OBR does for four policies in paragraphs 3.20, 3.21 and 3.43 and IFS does for the Budget's new giveaways. A row giving the effect on the current budget deficit proposes `fiscal_balance_change`, and its sign convention names that deficit. Each row keeps the sign its producer prints, and `conditions.sign_convention` says which way it runs. +10. **A producer's stated effect of the Budget's decisions on PSNB takes `decisions_effect_on_borrowing`.** That covers the total effect of Government decisions, its direct and indirect effects and any part of them the producer identifies, its splits into tax and spending and into scorecard and non-scorecard decisions, and figures that restate or recombine those effects without ambiguity. `models.py` defines the metric as PSNB and keeps it apart from `revenue_change`, a receipts line, and `ingest_obr_policy_effects.py` maps March 2026 Table B.1's totals, its direct effects and their tax and spending split to it. HM Treasury sets out in Table 6.1 the cost or yield of the decisions "which have a direct effect on Public Sector Net Borrowing", so Table 6.1's three totals, and the figures that restate them, take the metric. A receipts total or a tax-change estimate keeps `revenue_change` unless its evidence establishes PSNB scope, and a total key or a package label alone doesn't establish it. The OBR sums its receipts heads, from air passenger duty to visa fees, into its policy workbook's total of receipts measures, so that total keeps `revenue_change`. IFS doesn't tie its own estimate of the Budget's tax change at the forecast horizon to PSNB, so that estimate keeps `revenue_change` too. `ingest_uk_ab2025.py` maps AB2025's `package_tax_rise_size` rows to the same metric. Single measures, groups of measures and a total's lines by tax or spending head keep `revenue_change`, unless their producer states the effect on borrowing, as the OBR does for four policies in paragraphs 3.20, 3.21 and 3.43 and IFS does for the Budget's new giveaways. A row giving the effect on the current budget deficit proposes `fiscal_balance_change`, and its sign convention names that deficit. Each row keeps the sign its producer prints, and `conditions.sign_convention` says which way it runs. ## Families staged diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/NOTES.md index a2af4f9a8..d90103ddc 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/NOTES.md @@ -97,7 +97,7 @@ The opening remarks (rows 92 to 97; the saved text holds the whole speech on lin | 30726788 | 2 | `fiscal_headroom`, `share` | headroom at the March 2026 and October 2026 forecasts, as fractions of GDP | `restated` OBR | | 30726917 | 11 | `real_gdp_index` (proposed), `index` (proposed) | the March 2026 forecast for 2026-27 to 2030-31 and the October 2026 forecast for 2026-27 to 2031-32, with 2025-26 = 100 | `restated` OBR; the addendum classes this chart as restated | | 30726939 | 1 | `real_income_growth`, `percent` | 0.4% a year over the 2024-2029 parliament | `own`: the footer credits IFS calculations | -| 30726047 | 1 | `decisions_effect_on_borrowing` (README rule 10; `revenue_change` until 3 November), `percent` | Budget 2026's tax change at the forecast horizon, 0.2% of GDP, with the proposed key above | `own`: the footer credits IFS calculations | +| 30726047 | 1 | `revenue_change`, `percent` | Budget 2026's tax change at the forecast horizon, 0.2% of GDP, with the proposed key above | `own`: the footer credits IFS calculations | | 30727239 | 15 | `taxpayer_count`, `persons` | the three forecast series (the freeze at the March and October 2026 forecasts, and CPI uprating at the October forecast), 2026-27 to 2030-31 | `restated` OBR: the footer cites the OBR EFO, October 2026 | Only the tax-change row carries a key. The other 45 hold forecast levels or IFS indicators, not the effect of one measure, so they have `measure_key` null and no proposed key. @@ -175,7 +175,7 @@ We couldn't fetch two of the sources the addendum names, and two others held not - **Flourish bytes:** fetches on 29 October and 3 November returned byte-identical embeds. - **Tests:** `PYTHONPATH=. python -m pytest -n 4 --noconftest tests/test_ab2026_harvest_contract.py tests/test_uk_ab2026_harvest.py`: 52 passed. `--noconftest` skips the conftest's build of the derived database, which neither file uses. - **Generated files:** `pipeline/ab2026_proposed_keys.py` and `pipeline/harvest_counts.py --pinned-on 2026-10-29` regenerate PROPOSED_KEYS.md and COUNTS.json, and both `--check` runs pass. -- **After the review:** commit f4c7c94a's fixes and README rule 10's two moves changed no value, sign or year. COUNTS.json pins the gzip's current sha256, and `stage.py` still writes it byte for byte. +- **After the review:** commit f4c7c94a's fixes and README rule 10's move of row 76 changed no value, sign or year. COUNTS.json pins the gzip's current sha256, and `stage.py` still writes it byte for byte. The scripts and their outputs sit in the ignored `.ab2026-primaries/uk_ifs_ab2026/a3/`. diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_ifs_ab2026/claims_staged.jsonl.gz index 7d900fbc785fcfc50748d8db0a1bf5fe53fb4275..487eaedf483e66e9173eb13fbbf635dc26e2939c 100644 GIT binary patch delta 16764 zcmY(qQ*fY77d09i6Wg|J+cqb5GI8?6$;7s8+qP|UCbs8)zjOYoQ|GELR@Yj!`=WdA z-D`h`f!>FK#xoHG3N!!4gslfruBdKh$pY5HTl{K;D!!855`Vo!^oJ%W;sB`>2UZE8 z&Te;^2ITxR-OE4KGtDt5d4vrCdWUWgN(jzufhF!k<|9GIzQ>Q zk%gWyH%u`*iX@zd*w$F92>}C&nb?`*5~jG zgNBo*un{8i`skMAG02BxuPf0*HyLwfYM7GKQP_DgoP$LsOP9KkKKKvlbHYF^xtKSTC0OK+Bo1Ann3U%FEc$xC#$~;S2hnrE{>1m^jLPtv z;p{?$5ito<6ym)HW&(oFFU>=OPVIE#(Ebb%i$UQ(r_EsdKuKJ>^#*&a@p?_8sHb_k zS3A;*@yi?$JbwG)zC=SGm`d&q1{{J2f@7Q??a3gxPJwG(xRWe>R_U)}xtOLLPT+J= z8%5u!bdH%q&`Hcp3NiBUo!R*HTZK;gn7T;QC=}Tv-Yr>QRsj;k#fXj{xv8L2sE!E8 z7E8Ali6soroKPaWh1fkX$M98D_Xl;DU0MUW)y=5FM#ivv9T&q*odUY)fh@;6znG=^>t}wE{4ric{dd*@^VuqXNna5Mv>gq^m-5BE(I=C_L5#I#ZVU2EX6e9}R9r zK4+ID9bd1M-9U?J!FranhYSXRjSt*bA!B}4Isx4HX%3^G^j2z$4 z^;B`NoV`SJ8vm1lCh6QI_H;B}9y>+Gp;MNfCD9e+P; z6w++5&7iDqd`$3yn)u_Kq!ui14wm)Il;uFnsOJJK528pN=r9_j{vnK&%IJc2qd9RT z^nF2eHi-Du$}zv~w7ShazuhlOC*mqLizr#vMZ`s&LCC??PmSjSK6w<~ZnEBrID>dL zjo4by0R*OK2Azgw<84ZgZN%XxE5~KxJEbFT1d_W5J{o7)Z!JDLtv(uCdmERe5O&U798`CQ=XekdM_SU*J1F15r0z&F%wd4NLiRFX){TA@?I`|8`V%Sh z%S3f&vY3$KwN8z(Cq+5#O>vAjxCR3Ir`WKR7jRw!f%t0KA?4%}Q1=3@39`4#1@TD# z41GC@Li|w;Cfs#{Y}GF`(Eo1&v=`~U4y1R5Wsdk1xyTOYl^U7gZY=;vE%;Oe_azvu zn?fSU7C|SH+~DSO%Qt&f05PI3*z7TQ+UZo-L0;Ljf#W_b?-idk+Z#2gQ{_$VLwb#A z0=O}a*`fT!#_l<8|DJ!#fcC{+tQ~*qStdL1lq2K;p|F#)D-K7f5 zlM)3Uiu$h}c;UOHJdMo$=|Frcfqnc+@5B7~IE$}A9WSVe`ODb)qdJ&bu0>>7=D`Y6$TXAHkwjau-_h)cb4jOLjwNHAT;DO-=Z@dHcs@w&aN%eR=By%6Kn^ArO z`aFGv9?>>aT1iX?Hu}6hF)Yj9R$+`PqRcW+kpG<;Mk6z~cnF!&_-@M&& z&-7ikxf000Yk>zRJd8}XS@*faM;@^oy=6D1^BjqnAj6|;NP!;50v<6(6DweLehd5W zklNS7`sg_qveMHNXILp>Pf>xj*o{qfS&#N{r$zoZ&x^<5_HyF1K}*I}nfJ!1tOVmB`A z`$!M;?mW|@+a`YukY)(H>%;)p@ZGC+Av+HKxSd#&hx!%e8+WVsc^|?|!N=OMubje~ zhjVm$jvI8p5KA6Z^XSm|oFdz=0PW~Q6aQ=r8u0OF;}mb$u@x=<2e{Dgs<#Si_%eSN z(QXxz7*><&_>nHt!vyAJdWfGO$pWvbP&sF64~L8Vop3wVVwYN=6b zU$q8J571Cccdp|WX0?lbh^^!=#0-DAhA-OGf7jSA+|*ACf1VS8i}~LvI-VPv{kIbQ z2MhiOFEsWeZ6R3pgLx|zgh9?8fQ;1Nz5PC5Aq=WL@+6ZT z(icS%>Vl;p`BbaJFY9%mvdS{_G>NIgdnR~bED$-?hGM*Rj>JrB6CCj)`MyNT@zY6J z(P@WLgOuBG#pGGZrYxQv6FugL9ImwR%`q5Vxq~(s+Ef@Q82{2U{|$4UCgvXgl_(}u zeTC5w`sNcXuhJm)>a2kR;aTvS=!QD0!?>9lW2w9rZ`=r}1n($JQR?*Xl@5lFMrk2+ zpm5g#LPYI4_F6yPg)7fR1Z!*#1rfb9g>EKAABTzKr@D`dwhsl_N~A}IK&lg(`Vb#+ z;QS)A-q;8bm-Sp)k9$U$5MUz>(R>5i1M^8TH|a!Pn+r-AQsE1pZS^dc=iClIKgAh4 zBGw37?f(7*GdjA+)9=<_E9t7eWHi8T==aypDinfpUg&3Cx;1w7mr#1Pvm5BJR+sDS zhf;MsZx>>d%nVGe2qDP+5}*Ev%byVUwsnTIfUqAxt4u5%M4R*g%K(kGDVMkK{XRm% zT|_l2*4YBa!};R+PS&?d)6gnhx%J_$Y(VS7K7B&}A=nf6m5;9r2(=S@IW;Z#X3Un_ z`s`8DF!?YJ{8yqtl%h-eVhJ=UICH)7kzfIiud9eG=nsSl3RB&?LkrfPE$~4F_633v zdlUzNl~=PZdJSe^M0m@_mqOct<$ETNks2M`42+v)f$oQ5Zh!@D5b_Hal+YL6Hi;tG z<0N;O6~C}Ci*y1T+&b7ii2%YIq(fR@dD+sdX3ek*H z*j<@ZTF?%du(;r*{WvpW`w(sZr@?@kGD8T!=>uh#m?1k)Uupn4s%g7&fjMiaT3jt=Bi|*95>nUavwBhvu zP$$0`{(jBZE3N==>>G6mBjqpdh}pS$k(YHnT%M$xHcUn;;U{R2DX;TWN2JaK0o4FA zg^12E@TQw+dfQ`&)RT=^A?z5iGTU@9Oa36GnZOIgV23&`)~oONW~poshCSD?Toc4_ zAL4BfXW9F_Frx`QltrbD6-(iWzB_Ut)3eF;-e)&HsM4C$|AExe8@lt$1)6>sy6wnW zkH5W!aQpj{YfWE5^f>8$B06~^X8`_rCp=FU4u($8`h6Ber^BHtqV4@e(hlulbd-S& zI!4a|$&lYi{$oS#a{`Sl(=&V$hge_u-nnX=(piB3dJ(&n#f~8?Bj*6#yr&BgZ3AhL zK2ij)n;unBM;~b#sDeX1;APrvIL4ZaDOTamfqQs2;(sOXn0hOJQ&IG)lIfOyBhQ7p zgyL!`47j7IUWVA|oGW|v-fx;8 zk)J?%U^P8p;b*Y&KBUlRk;DSZ{;I^u+T3m<@~bepYJ5H*mpNW`n%}x~a3Ujh@nh4? z6ov~BfA2=0{=9??WFIy}-7SKoLPl|ko_CGjPgzHZH%siTmE34d|C%B_L=41ai@2pa z_xJxdlXts^S#PRo@;Hchuw&qH{buxo#brm8?Q>;HUzx?DdtZ%9+q@W1gj^hdEJp3D z7QW2`ei-H7-2eMVW_$ZFlq=NzRgZsDJk1)-zq?KT3FV9-1zJ@)v^%t2|L^IP4Xr*R z%Cr+kc*@z=H5z-s798I!+8%Tgje4VHFCa`&r!#%-2%?@lCIg>F?#mM?U(uD$Orm+! z0(&Z(7DSjpCS#*DgDVCof*!p(VIoAxvyYkMJ9wxYugZ_?^)*ly5(hkh8_b5+d2$PFHKsEmsz5Ew*pxNiz zF!Ro*WU5QnK7%{QO>vYQabSu(yoXILKtJq(_>`yUS}i3*#IgziL_6{(k*Od6SvNTP z0Dk8#SI0#PodoY0=|9eWK`WPwO+1i!o}L9a+O;;LOwbO*3i&@iG{>XhXkdnPfZIFB zXM^79c8CwC;yK6N&#c$R{#V&qy;nYDD{6vgq$cNLcVodUal};5g|udNdG@cAWWLKT zY)l(*8O?T;G#5Y^0NJ?6@fHcV6I39YfTp$( zmpymZ2a!^nRzwHaG(H6NN{M}sHO7y$CYo^Eb@ngMh6Q_}Y5h4yi+ z{o_C?n(8f#&;eoIPK`=Hj z#heT`Y3<^4{ej7d!ww>1%wp8Ce#&i#GRv77v*{Y&pF@=sKf65ppyo1kAui$OLt31F zC-rvQMFWp(c6nJsbnf-pqsPWPn@V1)?MxYL$C>$S@?DV*m|cL8{Ea?GBY2T7L$R8^ zTxg|qEvyv?Q5{74CMsv82?NHp=+>3)tAJ|5K8-cM1QR6rL&jo{4?$9)&h#RB0Zm~n z>1AdiEfx`%*MGw3s@>-f_zi12c^ES5J)`6%zze%fb$rOJvH>L6)W>X`EGeA`r zfoCX``Hg+6^(BtUE}7-V;t6s%#Zmf`i_@9(o3`BnIZXv5VE?V5Vf`jnDT?98fI!Si3$% zmyHT|sj2-Ew6hA!S1o+l_Oi;V*hmq+*$t-mUo-&NIc5+6;-AbyiXg0~4H3fyu}jD% z_7O_+&=YaOWdzTTD51Hfp3;MI0Jin7j4HzP^!v`(6m&O^x7zeiDcY67H=8rpT2gCH zU}!6T(S~KJLK}`bxdpb_|K#^u4Cl#Q>KfuT=XejNY_Y*+mH!HRg>_oUMt5aE2-O`F z9t#7*RVd6neZy^RU`_6(k7zK7^ctqykc)eOmS{NO9RSO%IC6U$J3s!7y!aVk zX^8pCP!)th(>MHU;6C^VFhkVovR?EbiPh}JZ&&XrB$cQgMbqY{ak5O8VMA@V8r;!0 zcGxk`9~JI0V<;f3ItmaQNHnsg0l3*EGm-D`kHAjDS*x#QVIg~s zEXNa3%H`sesTvH42h8d1)Q2$V= zN2dz^&=iS@Y8DsJdISYkGOB_h|4H$3#4sKeS2hYHGiz@6GHHn>&%lHrJ>tUOn}K(K zkiPC5315$zYuV9&93C`3?XEM?YVoSuIU#wx74^zr{`EJ*J25A&Ztf3eZqE}bb#{xR z-QrG81xu}+F#9LHBQI_(w3Q?MeaW^X5GHU~sNQgrWWPfGmALHH=+F9R?Q!O#cjxHc zp4)SFtk-^9{9HT#c%I*~uy++!AmU6WNOu&`;pd(!Ze(95FAKIjg`0Wg zQD;{p_{ERF7U?jhG%-bYfK&z53eFn;coX$GZKdRt>E9cS+LPji=pc0=@gn;y+j2+{ zFu}|eALoH6%f?+59vxe@G0(twnG2~m&$Xlhxb=Ux&tiTC<+fltno3ahv4NyZ|K8{R zO$pQYx^<|w_6w)!?uuM8QBOfE5UwBM?!Mi?SC>TrXdqytG|=JmX+oup6zH-^Ab1Dvd7|Ck%60Z}FNO8vtm^`Z+;Y4;CcSD?Oc7R+$m$Vl3X%u)` zn5h=gZjYsFbpM7eZ_p#b&}pyZBfhKo#?iIsCEoWqZnTfz*fms9KR#j*OMcG@!uzb} z<((k#>9Kv|f;YAD;tu_(Bzqt{Y&ws?gNNpAm{j&UnirtxtQrq{wggycIr3OKnguvI zIx}GrD*_v#KbE6fj>;Jft7{CSEBC8jY>wnw_Ihgs%o6ILf0Hhc&TxX5juh^8p#2WY z#E-m<^INPfuI;vX3n^ZD<%}=y!osM|PFn{G2-O)U%nG4|L4iTLdlbO)MP!B%Hv`B0 zZ(y8BMc{Ye#|8M*{07<9=$}U;eryzDohWJV1UwO=#?T#RAaFGaDE;JH9nnMz59`Dbm63>inGzmq`GThvr^VM-3sNyhRxFbch7NaOjw1@>U3mI{~}j}N7< zia`L;qY9YO91581Z1X7AD67=V41U~->*qqt!z0~w&e+KpgGT5(3`Fq;?jGXF+mmjt z3Wa64r;W~pcY)_=>6i@Iu5bm|LXrv())cO{JlxosjJfoj?DQjHf3%SNRk}7kisR-? zN3Mw|ydj`;?%Bh-xN*gKLRgN$L3mt58D#fdq?=MR}P zlNHM7`J7QwhKrcL^dF9Ouehx!*M}i!al6UN*O(3yy@i;#K#LcC?8!LbuKx}AGO$J# zChL%Ni8+o7IzIK6t}T3;!|0qDowv6+H-z4C%;gs&Iy(Vr7f;1CUEg~()BGgu{Qq8 zIjW^T8J=3k`+%feJ_4)1Pol@=1uaan#3jPwK>=y~G3caD7_m6o1NAhfAg$Orh45>drQdIo*RRg_g-l*awi8OJ>&4 zJga6IzLeRk!inV*hLTebv+zzj{d%7MtD^9zR7KQoX+Hid{m!|n91h6;GeC5o(%m`g z9^$a%IpYsNoIj%x&gWF*%`QhR0)lSMTbmWc-A^GH2UNy_qxI@NG;~+>oKO#)lqzoo za;NQ9WTrT!3+THdfD;UJ&XcgR?K`EhQyJ}=X8J&im|OpL8=BM2FtsYD9*AbFr;7I1 z?dV;cD*g!1Lc{i)j1CEgsQFu=dLhfS<0=@QSp`7u1S_~jO^_Qxebr3x432 zP6h5fSo}bltSz_?O`>7xha23sU167vF8gC1++!b2@&s4jp+ihFGA$u(&)65+K}Oj2NJ3mZ;dVOst!(wU*`W^BwQvN%2V;`-jujDVNLiiF z|B%Ay&4hN?cIcn^!av75Q(R@aW4;n6;}P5wB$dkmg2H_s5&}+NR^PovcxLx|uxJIt z{@R|OJ;s=l0fexS2|EpEmK?W?S4OWs1~9-5?)b0tsp4}7^-tcNT+DmCJu&!Q1HPz% z%^o^ST2;B_kHGBrg|z6;pD!Z5$gR=cWcrZ?ch`iAOt0m-L%YTY=@Fd)=y6(L7|cxU z*kn936x@kHXy3%K^IJ)2PEnX^+NlK-K69z;6+`}p@=eNakTki24kBD8_Vf*s1?0dQ z1vEKCpNyh#MX9fC1)f(geBcP>#@w-JeUMzM8)w+xY{bO0ETts5MsD_n^q+++91Hw~ zSqWVHuHql}s0iZ|jtpA;O;E36PI>0JDdCs$yr;Ih{U=-ao>BeQlgx^Kp`lb%sU};T zgdy6?F=w=P)fC_^K)#Cn}b4qwP$~AGFwXrD}&I zGLxO)`w2`IYVGRGLKtk1y58NutvV|Lt2q{F;}o?_inK4FH>?i&By??(#^nJF_V z;{Q4o-w~}rW1*G(V<-8}7{W}Pk**Ip*HSw^XEr#&RBRzqx{wb1vcR@4L3OzK-}!|i ze?op6d|t2uA{6u^K^J@({BNANnF4MQ%F|r}oZ|B4yjQpB-C~+*@%KyFz#v5Uuhb2O zXRIa;=B}o+a0!q0j;x)T5)EpQF3$t#fPWM8cd$>6-5rNPwEI-t^y!KXCLhHx|Kz5@ z9*sza=nB|1B*n5zjDTX(GLTdBJA+@umO?-6ppYISmNxXX+aKwmt6`O;@6din(?@iy z`}Jc{Y7d>Be&V&qA8LgnJ2T)1pj$$U_1_k@C1LNv*V*zJKrQ;4iV7<@_=Hjh-!fDE zD8+xwp-R4i{}*;cNEsp+0KODYbtZ|)2M=ct+Qfz}!DI_#&IH`;DUm?8e>pEi@vLVW zNfew+kRj|#76TaVFBBwLgZ7Wva@qkJS5%B#=0RIXjV!?aj?_XV! z+!!I09wyxj#y+o8A*rLfT1JGZd&#up?;#0%jx30zjs1+`3fbZxdOcO0ha946L(4bh zgz8BY7d;equjjLv!>%aeqW+iZG(r8l2ZWpAfU8{(@PI%;i-b*tOLIdX<|YI$q!%am zlIR#H1s*A+^U#zo$><8By$9Dl;I7YNhxiZ1t;|2>yDP}3Vt(Ka&B_W_@F{lV^u?GQ zb5E1QsgckUu<3v<#bqax;qU!!I;TxL7=HyMN=GVZf zL5QD_`nCVsQP9{G*;H68J=gG90v7o-JHBgXK48rqoN=C5PlSyn7w`%vhjbFOyDm%8zjNj{l&a#Yq}csaX8TP2^8}M0K*qy=TE2P=7$E*`Lo#c2zWw zCRY;JV12pi7q)9#|Bw(pk@DTXj*il%av=QKUIX`lY})1hak?W{KkDpu7B<2c5n)iW z3uwje#x6Ac74ec9*o&L;+3{quYDGqzc}AMI%cEdFlD7~2&yb)~I}ftfH<62I<|0M1c)cCAV`>&J3lC16kKxP<(GlcgQV?aT z>_$AvJE5`Uiiq4@v@-r-tX?>b-Lvx;tD05?ZF+98M)#y8>!KR|M*vEVXm($f2}laJ zn7l_GCT;QVQrG9Hp6E2Se~3U!BgYmD2|w#Oor87s2?I5YJM)y*r-q?D2@bfxWUHU@ zN`MlkO`XgfBB|idiRW^2h|YI9c@UkWsPEoW7SACDmn{XW&Hk6=wKuq7O365!L4W$} z03A`yfou7*s6oq&{afE3woY9=5(uDGg;eKt)2EY?>8wOZy3fxY6z$&PggWfBQIU18 z)ip+rcEl??o>q=EuTL+6J921aGuwWkiVJpmnHd5%+3ES2nfl_BZSM+|ozaTZ)`IiU zLNtf9gulbdc7$}>9j8L8vWVfBU9`g4T7gA*e;iV&BOx%UX%~Y#fYAc-6Ube!$jp;v z#m(IqNkJXW1KdD^K*9_KkvR9bVB;LZ83SEB_~cUu`B=4QqtqWy-pltNUlwMTqk=S) zKVOfGqL`bfd&Go9(g|Y$?6%MWIA?+&5s#|O*Jk6F>Pr)6wdlTEx ztot2!Wl^~@vzJUU{_6GDX5Kc*ng)F& z!F+!f?Xl+UCQ_#|GfX!rCUn1YG5QFeb&3uRlsJV-?DCC=)7VG%*`;ST`)j&`a-Ho{ z1&D~x%zYOM=UL@hgq!9cR+^JQNW(^&lw895kJLtDFsZ7X1nc+X0Q7gi`@UA@Osma1 zGlxbVh_eG#pE(x0h}5fK z>Mmx^`RXN%2Bp>X!=Fw)y`;aHw`41=bkT^s{Jqjl@s#2fIS{2>>(b^f)SU$ed@+rcP+r6G zK@6|8ljHRHlmVx-p1o9J3u-eXIPLD9Y8 zFF{sYiTI#)+<^F`t+TA&6}N3DX*Sa#sZE(lOyebit(kZsc&>KdyLP3*j(ZO?dmjDD zT_gHvwdOu5xB=n8MIw6X?8#jta0>S5Ix@!YZxLI%jv4DxJF$@{TFWP_{~_U=$Ab{+ zSgMRH&q;J4phj%2SFW1{L3^`2d!t^k5K>e^R4-RDo%S*iuP@sndaW{W4tn}4!_L`Y zz+iiYlnCSx1JE#(?4*XOly(-FXhnBMQ8*l;>t=+pW$;O*GR_dp zLG@!~Da=35+XaHK2E!!eqH&d}Nze#7L~kk9S3YQ}o$0|$@%*aJ9%oYRgdoycV;qfW zE$o-fQ1TvFRu&z#mTtV$F>RWvX)=Gpr{=gsS|n@gYA#elAJ0D}xqhl|06V{FyXN9( z+YY)7xBlFCl@hv5#Jtu-{ifTk@mLE$n7Y!6oD4)vzX>RTmgCpf_J512*`WemY!6vz z^leTvofA(^mDocLs1NN?0Pp^d&f-^82{)yH{F?w&Bl4?So3ze7Dul{LVL-`wkFOI# zOC9~5CKrkQ4&xLz zrv2Ve_{VF3&jf63u3pu20{=&Y6r}}4$!B?z5wh6+A4%IxH+<; zt|z`xx%BK}Y3UyM6N5j0#Gdtb3iw+-&xkaqE<7xmLeH7W2`$l7v)umOKYWpueTY;w zkIVY1>d1_UF4`H@fB7aTD%(VP$%GW~DVK{_%cMfIp;v#GkREn+8Q7WNUNR?0IpSO{ zqdGfnx{F-Z=C#5%8{p?(1)Z5=At(+q<{Mz!W6{5Gy5E;*q0FP)0w+m~w0Kb2BV?>p zQAtX)SkP{hly{^jUP<<}I1JR%26Ho}9u-hx>0yO|+DN?tcG*FH*9obLYTRVoyL+!X>0lU$&Cm0=pFDwAv?OZv>fL z%J}MRS2MEH{Si6c4o*|Yt=wg5;&rYI;Cmc$%me1bK-MomP%W{LXs;*P>zhkgrP7rA z!?|Ez&K8WJ7tBAE`|nyr+tHmiBeX_}P*(qp1Et&S46=k4l3fw7dx?UG<4~uw$H-%= zl-TuBEgY*Gc^*BRPtu3@N3Vh4iY0KF%8Zm`PqqftMjpX`r7@}M)x2V#$kccPbm~TD zEbboH`=arrC3YeVv3vc4uc}5ho?DoN7kg7r#NkS)_Qft1YZUFg7tht^s#2 zIPp~C=R6k2ro*?-Q(vAb~p{dF0psM+D}AAfZ>Pu`KMiStR6_1@K9jinv(s$SyOcQI9XCAQ;w@@-h~*vpY} z(tT-eCA7elYqa~rg=RTQlE`LiM@o{EVC4PzP6g@gvg#OgAFP49JPL~%X z$N|HaX7eL0v+}zYCP{%n>E`5ulH?>NjRdt$`npXJt zH%FxU1ws21-`jXz7uWxQqrGODCcZ~Xg%7S)<8F9g2u~_?dT{_BsFt*sz~o3 zC?1xK1p;kK;g8%gOyD!Ppf7bM(UJ~;DKCk6tsa~1XE+{Ymc&R4#&mn}{%*1zxLOaZ z^5lTZL;=}PwS`3|udA$0(?%0Mr|aW-^sUo~chsHa8>BNST+H!q>9GrXzkOPi{Q2`Q z`V{|3oob2EA^ej&HH?Y?%UmpNia0{kNilKleV{M_+^@^yV_Izt^tNNSGD3GCr88z^ zYhdxY)Kg762b=n38N5!sja14U{MSzeoM^=K$`OR^d-YuxJwwBU7zdLfNvs#TJ7MzhgF-q%F>D%E z@^+`EyOUyS_WL_6P3;7p9+~zWTQ;pbQTAN)lnU*9D79wCA$q6|82LzpWqJxb4QiE3 z%U-_-Sw1dus|`ID6=3|`(orN*)%K`IQhKcW_*O3W3TD4oHu{uOzToSC7i*pq3WhVk z%If7AoxaR&TypTeEYMFg#-|(Wh9+b@)gV{>Z3a-!vfo0qe|#vt`dXxPK&85c8I;Hb zE-zMb9o5`IndFe$*5#TQvdRyTG#5k&&ByyttP;A}m?p&AjD5XsZ{-&ros#gR+uAW9 zLn5q%SI&vaTN<#4(2B}|EV*Qf3wo{Y5+COTg8GXUiQgojJV9M}aXX^hJzu_xtgsbM z9t-|qveXPd1?)8v3q>7Ns7a6|i^3t6-CpDjnsjdN5?-qIH+YHEqy@64Ir}R|+@tp# zgMK5f%_*dsdwu^}M*P1{&^}&|y7I=GvO@#&2L>Z(-r~4zbl0tc@E(-xnAc37f@6V6 z@OtZkfqTCw!larYgC$i)bx@Lh*YG9zpJi$!^fbNq`=x}`G2@+-H?Q}5C#TJx`xI>R zM;tMK56hVdBuk0$|Ndr5Dg3Pygfci8yx^%S?i5qbpkY@T70&c@(oRa2e|*P!urav- z@*?zufNNy6c@y*$(B}S^xC;}Jfx_An`O;y2OBFrz*%`Ke{u@YUDTCP(Ven8Eb87KGMKErFP*&h^9a0G=lhvsGA*6kbaCPN4PsN zhU?UUqhF1p6@*(;x*2kFnAM`Wq`ktgr#tloT9^2gj3xI0;FiJ%{8q2V+dAKxoj-3A zH{nR8rJ;`2laJ0Wq2s7xTEj2Gg$hpKId-3GUWjb$5@5pq7?@}E?sa^U^fiCU31n+z zSbCN+n{%Bu{IM=Z6kT&SzqOP9E+Wq+yxqesXL$qP@4(`^&Y{}iOI}seR<0~)PtTv$ zi<-GoT3z=EcwHTO+23_@3f3{?u9Ql(jJVRSrlGzxW*apf)GrY|5==FG_&^~4Id$ag zIwxG&C=&nYqU*`BU;1`Hfnp7Qfsa_+%NgGsSd3Uq#TNf=X+!+f68M-7rP?au~3QDR`4=<(wm~*wD8s+wSiQoBpG3x34}}HqF!qo(wbwq?lrsh03)ZW z;65}>t(Kx+ji({?YQqFjP!FV|e2a>7>v5*gfP{mQ4ck|yzy zJ1XElWf1dI>3!-ZRd;2~Se4YN;w<)BOnvEnYSUei0xfX4^U?5a9ug}xh5>);CktEJ z&j$z){s8CsMA?$S4?d;C6pxfTPnn=6$pE{jH5rbs;+H45pJWSKNw6GbK=u;zE|~f< zZKHv3imRtK#lErmR<9YPAXoRdv7K&q96=wSQ`pyk?NzANaY8&YA9K7|6Qt3B^*=ET zMYf3HJ1p_f#vWj`YXO=~DVyo8CzlmX>i0BPID+Dj1_9yCW_LgCMDF6@4pPah#3Q9l z_<97mTlu-VnqNAkn2`4NfwD3?)!KHfoj5PkSr5~hX~8jLlezPfJl9mDQ&PAYD!@KLJAuhaKhXci z-N3wU|6&pvqBsvEtRPh3XtGsXXkU#B!i;I+>r4sO)o5eiwq`YmN|v2ZZAxx&axB@D zuE^{C)O|S4NKNg1HbLjxL-v|eNm5yeUUCPgzB}*Os_e~&NfPW0GBh^patgaGmZZ@1F zOz)%qKkH=Qe)+)GkZVtD7vh+t@5gT>vX+~E@)QzSo?QK}EI#?vSR zI#{X}A+0LDApQ_wmA-K*n}#hFS2 z!s6{!j~Yau#k8hPAq#))gqw8*rP0g@D>*e(q>UzqEpT_J3 zIvCXq-f0~6V+<9_la9Jx_CSzmeJEhd&~e?Px*+_gs}I@OAL!ukbrycX$1ubFkKJ~m zKe{BTmT~VQOy@D9!hZ)mjjt3BgdBvcBrmH=dlmW|_8=Z;lSa7-QJ7usj(A;4+vt~E z86(2B^bfZk@Khp*12fg2Ng(RqQuj(OK51|UViN6py17}VpF&*?lmahfdH3Dn-VaYnDFKJvh}-W{LSmf ziDr-gTs3<`siZ>8nN&=0QmLf=^=7!fT(l^^1`)9M2LTa&o9wLGcieYbkKJ3tg|`ip zCzG1vKTdv=!ezl@-b(#vlA2go3qkT|Pxwf;*=qj!|4TOX|0i3#aH}!&Fh-c244;|JATaY_qY190qi38al zNKA&el#eqsMpl-~EABqa7x;hLY}cY*Z=jed@aQd_d%U-@)xY;N-IF}1a8ZtnDxR%o zOQ4nf#|58EC7-+*KE3LpS&>x3M-CX0j2-kcH0KfUAo0z-5QjfqH63l8gL)-QMtG|M zEEtK}#U^fI^I2rVv&0qc3kb}_@<-!6p`as=_k9c?o5nVh+)decG-Z(m?G_O!qb3mE z?QR&ACc}L{-%gMXecb=orWdVrN(uIZ4zr8Z;0kmb&bF%+mFvclBnD^H0+!)Myz3du z@omrpgDQSOAB*?p3`Zp^9nugxrVqxbFr6S~Ef-Kkqo;L^Jp9$pNEgvoj|?`c4D1rl znMC{6Ed&~FXf|0K)K<0D~NM>`aEks z58YnFeFP#TE_Z>uGuA~sHrf$Oe~yfzk6poxtal*(%(w4>D=i+DTU=eWUzn|^m0&7ls;S0Hn8@nO z^6EK4H3s}O47sa}0@hxZ>JvIJ*tTWYuAEKjr9T^?gvTlKeG2+G6t;#wtaV|EYkBFb zQS#DW)BX(c#?TG%0!oStAFK9z(@{JQlXE|&?&rF@Gr0whGC_}LVTvohIesA0;lH5{ zCi7hV=Yh{^zutc63%Ksg=VO%D>JFfR6Nc>V*r%4@GPSlZHB+{ne@gatkx*~_*vb?g z`587^HB4qDEC=o&Idm(Sk8#NOp;~XeHrQ4DTKxai3L%0(Lw5`gDF`hd3PyWsJ`~BU zEdMB{+-|dR<>4aTfj@k1a>*{Nad0{|o7p&(4tFma#^K8J>HPo=B`YdDc>3wsZ*su` z4Esemoz3ps*9T~G60Dw~PftaQ1IwU(!?EsRmn$58#4{-PzOuqyU)7pJ!!JyF4I|!1 z9wNFDu}*s67hRGXN$zLn$)8a;1L?W)$@c!4gtq;dm@O%ww?y-*VraL zp?@5)4x8m0^bkqQWr`pzodIt8gfeELxs_DSBWRnz+trrSM{BoVSGQp$x_(CVOjE$H zhULEYZaPr)Pmq#h+Zf2IhKAemaED?tyXSAx zo3G}RW3?Xy^2YLc;-K$>uxEV_A6^EXvg=AiADl1Oo{6iWjYD-NAP-31e&DnSLr`e% zNMG`RxZS5;kOL+x3|@Y|BaNR&yFFU2mQpyEK?JPfftRSt0zDe|zOkFnx-UdmMJvKk zzkg-EIL-YEMlG`&P2dEhLJ^xW+s1*~FO-BXSveJJR9u0g5qUFX>@5#<7Q@t<_&8N? zI+NVB5(f%ee2{d_ki!X7n!b$NqzM5eewWJszHimSZ1l8=SJFb)ri{aenbM(6rzn{P zH+U&d5`lX~_HhX{$KintuodYl;*_qfgxj+_O&T6;*(csh1?P@+dYyPm3$)~v?n;Vk zw07`tic+h<9xu83z5kLRAxpUeM~07s6xNOtjhR|mJv!M_Tl#;Wsfp9H0$^EXKQ&92kuN5!0E!_RC5Om#!F*Ez~?O~R}L*UR|xwnsp zYRbIcsbjLlgJdOn#}dh%LFA4t?i#ybt@Wp`RW~PvDB2qtn%5USh#%rdc+vfn+L!(~ zesKe2M$sjE1|91^yz3_amjkNS^x)W_ZyU+A=vtgFbnK*f5i+16;CY3k$<~4>8?#BP z6Yf0zLrt|aJ(7~dZ`U zInv^K(6ows=Jc;|1GY|!$0Q54|90T&?ol2KVz2Wa8zOcDOSi&w6oDuS9`_bNC;eZ+ z-2VhK0p^SUt0+nG(pw>RJt=pSwdHw1-NAOd$Px3uDF0Ip-;N6zUklrY1}AV~`u`&P z8{cK&HgirjbIyD0&)(dtF#dl6{{R60z2DVY?n@ze5XlxHcNE~shuk-`tkk}`X7;mZ z_U&6(9yhaBdRFR?`%_Z7n2IJLcYo6`Egf^utG$CTn$x(D-5EIQn#nbjYbMuBuA}Zp z!sY8cbJMU*9d}BO`oR0Z(1G_7pAU?!_Xqv2EM7oi{cowrwX9+fHU;b7D-4J>UNCbI!#+SG{_5_piI^ zqSvaYs_w!;Kf^&2nTYmK<%-0?=t0zItL>+ci5`vZ^UoD+zRY^fw)jaw5|b2gfV}b~ z!MS4cUauKOmAoT^*`XNBuA81%#iER^W;0X>&l`8Xe7PGqLa2kv^Z@6OsC4t^E@s>~ zugQbWTuPGs-MJ8QM6zxsgBzr7U9t4|A1P1SrLKgVCp^#mcH9ROcTdv zcC2%!zlf)(r;0%wU-d6@S6ATDVzFc2k4Aug#j?#1Jf|nPdT75%;~&9yz%{pY`8N6M zmMYVQxm{W?_VDnCw8G*uu|Tk)JkNU`gdeN1@^p8`ZROk$a&Jr-@%J>}&rrkWdZ#t& zV8uR(wasj~tM3>08O?U6VL4mj+{W#rx{9jogua9KarzkP3Ua%FSu%5W3 z6DdbtXxIU)P^poQr%*Pp!0=Wx;#>b-9Wl{!E3y$TPfJK>|_vlPkplLQ;6!kNgbaY319{ zgwvD@Gu(4J7xrMm-Qp}hEaO-jiunD83?B8tty<<3fxZ2xy{^rTmhQeS%%Y~_ZxTDM z@BKHBn>QDUym|b0;o%tY!Nl-#SKs2nzc6u0PDE|z&k=t5&$;@D4)rOaj*sR@+^NF< z%%5|OPp%EWr^$~1P|A-$4Hcv?fdj${sc2IXb(HgbbK=A^!Go}Sx8U@+d%v`vfWNf3 z=DFlOXVZP%lUG7)W%UvZp_cU}QiiGRwnKs&`yTc`_20@BbG5mPN*_1$>d=}^4^&kU z>v`PAl=U)I3NO4c`2H3>gRnb*WweAc-tmG4-Kg)EGj1*ewIX@l;Tre`s=QOMe8(kx z6Idj*G@i2P7}(xpJt^ehd?KrfciDocGqvQNS2P`2$EXIVR}8PxsSzRO#ZzN8m%G&O zo)duAu}&w^K6?aqt{e_7cy_LaNtyXgC8v;uv#T)#nKH>41m?-`&A>ZuG6(E7dr{XR z9@h|iD+ePWHMO9#a5ubd!TFI4iekB#5(?i!)Pp!G2Z?{xVxx_nKZk=~D~s@-_pN*? z7KZ|QMM;)T%Of0IeYokajZgHU8uzs!WAN1Y!p%>rk2*y_`uf|C_E$Q9mJJva6@8XU zM!Bjk=1NJJ&+`;nM~Y?B-lfJ_BRWutHRM}mLLY!m82nTF3MB`B_xvLWqrd$L0hB-H z1GLNe+qghA$o$nCIPQ%i^9|q!?0fFzVW=ML}=y*`4NkcGY2=zXy zjSs*vi^7UpJ;OV3rhaM z8NnftT4dP&O7zRH+WE_?_?>;kfyMr65k1)e!V~DsFHVCr)MyV8{J4w1b!;%{`>pkr zGx@Tc^YcCXoa()D;gg^W3ydH3_Y-B{CGjE4iFeG(dHOb#7TRm$W9`iiJtvq`hJ1+N zFYgtvRc|^W)Auv=8}Co-`k}XKY{zTs5y9l=-mTSGs;E=wzR_n_`t|#F76ZvkO_pt* zm`xx6-$FIk)6L=E5?k%k{RN8w)PXI_Hg|{>A7V4lDBgF%9gu@!ZV;(S?jl+fQC>yp zJQcn7Y-1wJy4+#=cPhA11HU0%IoE;a*o`ie&pZSQyI=eSOF%)<%p7S+eGDhj_kwz9 z5vWhaWb%hHA^Def*NO{APD&V6&Bx#5g@7mGYzFy5U9baajV=h0ji&H@=(az%blyBK zS}UJ%t|XIDHIJf}kEMxhUSheM&_s=n)k)e;(cJ0 z1ixaQGf5w^nzYS>&-LGC^b0X;?{B?*ZL+UvxJQKb4+1#O7 z|DK?*pmDFTAnBDxCO(=0x5e}G7gFR1H6bRsb{jLf%W`)Idx8V2%TWwI1O>Vd5V9#mn<255 ze#B&}E^}>Ia}Mi+nM;zz70x}Hyn{3i_Qnx0Pw&)C&pc1xFbU>ERTK{4W4VC$^LNI- zaSZIOHR0cK3w^^7=MYXq;I?XoagZyw5R08(fQeV^7!_d}#G@T^q4utqcK;%M!yuTC zZ9zQ5Y$4u5gEwC1kOk1Hrg*Y)k8RpBj$+^9cZ&{#HNII3K4SbD$=gFtNW1%Mkm7yLNNC|oEuV)` zB&efrzwK2hg6}m5EWS!SV5Lok7q}IQjA*mIIHf>SZH;e);Hd%hP9TOJW#|KT{Hnna z_B5ikhqSH2%Q2CL*5jL?f!a4v1L|IOJV`c<$To`uoMjHu47-7)B0Qxim`%YK*g=JA z)fV%HigrBV`IN5DTsYM$OjX9y{b!z{puslS(#{`r2c>1PBL?Wd3?`&n_w_9jC^o;X z+#pI3w>%8|a=rx)%6e#OMmwd+3vf|EFMWlWz<}|4TKuT$@<146${i4@tdDt1{421B zra05vL_4tR+5Cw#Vqq@PYNOkO>?~EFr&E z-jnXHVVTP24Qis_vAej3pdbRBL+bS93Ra#mTP`qG(8fUaVTXv*Jv=^IPB4W>F;ofC zEHY2R1g)zxp*~9jW87BH84(AsU2R|%$5P@>_k2Tr69tF*(4OilPff657t4O;x=KyyH_w~P&` zCEMX??qMBjVvNxuV>Ng%M1a097*_KyA?RzT|qmPTCo8kohA zjfsmqea#U;qvs61T|&QOKc7g94EdjZxBy}VlYDB5%nvY$xeappk;QRRx%XE1rGNh; z7NWH`^UNPk2n*CB6g$WZlw05}i9zzimKo29XC|!oOEH=FX!jP^FPWN9J`_TC4?n84 z9|HjS{d*eRE^H?NoICJVLF@)%B_ZTA7@xo`p{F#7?Ipl=&jnr$0BOh(cep(+z8xi$ve>hBwB8pA!>OzRDRt55N_Q5`(D0FkljB2z7E4{cAl+f z&j~5?jd4QiGx?Oxn@;2vr|T6S`j#ti)D&>^kq?-|Q&HlOA9ggjkiQaYH@pfAd*1b6 zwTXdhUkTaB|2od9mWuRK;m=HB>0u7YI znif7t%M8JqBU}o2Tr_$gj-=c+-h)SGB!1~oH%sm)iwC!cU&3y~9GafJfoR&+uLN`h zwMv}JgETD;sA;43*LGLHQ1`ww?KJLVO~sK5X32$me=rgHC2nK*Ex%Kjcdu|~TKeV4 ziFkx$>o|`;f77sucGw~a#;Vy30pESP-ex1be>kp_SJbjyQBS5E40WM&aq6~2^W(qK zJ~b&bi}1#Yy1~xFWaqy{t-~&kSpxo6NK$ZkT|wqmq=(UdzCdFZoxb?Rk9Md%a8 zr&ufu5h4SQ@7fr?!o&!T>7h?n0~4a5_@)o|rd|}B5Qo^7^;pGr-h-d!g!kb>F*st+ zXl{aiUzo8kj;>lP)yzJ7FJ7L%bG-+!#9~F-m*IiGE?8>Q1T!zg2q!$1sz4MwD}QA$ zBOUVl3DGdtY1ez-*N~jRaKj0bqaghlFHJK%SppXqO`ovdh;o0`S;G@E*9Fk?GX0Hy zzbIo`blwGj2cAsgW*uY<+cZ1mE>y+2Hf_UVJ`Qy$YuwV1VEVh8vZ(bayS|hDC>r{)p(_G~xuqsuxV35%bZ9viZl8 zu}vE|u@8Xip^pg2N& zV+;zzRsDhCuxyS$4-g4?xRe_t79a>0b`Q1Y;2(plHn34rp zNR)&UjzKuFk6ehc7(tW~CPtiLM;^Lh;`~fCyQI&K;(%{vUo-KPg}-$)dz)jo>VKAID($X)*aIWGPQe1I+|fe|pxNs<6jsXOXgt zFDr-YtEwICA5bO)>IaELV{OHeJ^p-aCWYzxok4z)2T(>~{6HqGsunWx-g!}$Hlr9X z-JWHK_ln@2)&!KVFI&?)RH zO27pli|c@#b#GdjyAxxz2vsOE*u9g6;?Z%^ydNZdFSWRP3nmSQ-VjC&sxJ`kUUB$s zTq?R8G;|}M^MY~nph-eGa z5R@RyNq|hr|2YtaqHjnt!K=Tw+BkqJg9w`m6Zm_-#{yi&(MGnox&84H0@B&|Xiu4Q+lOpW|Y)yn|ri(XPQ=UP#}S{nev-AaR6qH8HAtMyn$F6l+_|Ca=0*l z3CY+#QgI%NCSJ&r;L(vKERWJlYETvjuy#+aCd|l){uvKPb82&^&+?L>RwaJFGJB#Y zzG!zF-i}hbVv?rXiEcz~i)IxPMuo|>o+7BL&(*O({IpJ;<9CBRjLKLr*UZK+33s>{GMP_O^jR`VT~`~f2XWEMuA;!1~$ z2bT&3t)07st*sajyjiz{HN!w)gbjC_z3>zR5`N`0%#$prk)BhJN~@pEJ*VyyJD2ZP zIEzSmWeT;=T)lzyD7}MGF95j+coEe^^lkYEnc?AeR&ZP+v1#0J4)(zG7DJ0)g^Q_E zAL7m*U4TX+U12;UhZLf$jv{`uPSC#CZ(`rdxWd7`&&N4d?DzJeAOCU^RQf%IQ=5-_ zT3#DHiN7nLX0R9y`M?I{whV!S><+R%0o?F($*Z7%p5_q5bIpjj&*A}N>1}j>V84fE zL`qb<#?;1y>!LxG1Pz&S)G9?xE}(!>_U#pUZ!faij$5Nj-Uvvg7f_dySV*96=p7L5 zYI=zwrT@{?(oUJqBE|xlnkVOQLH&MFt{1v)lwr!jg|^R}Ccb19ti~>Xj7`%4L5f>V zfL(tzPxN&hB&cvM!?plqvMofk`Z5m7w7`D7549-Em+D|ay(w&NX$eSLatTUU=Igyo zQrb_`eS^aKZeJY(VsNSJr>I%oqyzsh+o~alZSF);SZ>6wBKd4c`LtDabkY7|4=yvX z(Mgf8eS8pe`Nxs*0@wfR^EU!S_VT9;Nou7hxDF3gAe&)i)!8>779B@rU&Juv4|cgw zGXt%6{&b@Yj-QC^g6R@<-jK_;+k-+UWYzMyD=oBZ z@j~l24tlrGbHOjqDClR&ue_~$O8FWAPQwc?5kFE~kOkQcQ9qailJ5o_TT;NV|AS7v z)`j8^s^&*ONR?N>p3oSLJQ3BP|6n@#PpmEBp*E%)hI+Ai{2T62etPMGI{+ z9Lr!JphWW3JsrCvuG-fXSv9Shrrk>8C3nRp-RBcV5Ep>gTVXHbGDi@7H5Itlm{8Ih z@Nx2GPMjw6Qx7NGv_O{Oni=Le#XQ&w$<7(!`u&OL*U|_80|I)G5+aH{PNbTS3QI9A z@1U$KiM|TOEp37rk^NUqXj| zKN2J>+E|Trz1@r>v3tQmAp9C{d{4~m6Mv@rSiy@6A+f6)qFjHyx;2j^4l9;XfkDkb0pixg{;!D|gkg^+o?+F;qx~gLj0wLKdLgGn8z?dz=HJ;| zDfDG@rBL7^*@MFUh=Y!jJe$0|P;9HG!sw`b7!mke04q(1#sLER&D0_};d!VF$;M-0 ze7aE=$q!x@;o)mbkw`0kmT-+zqx@@sAQEIsXLvDY6Z-W>>)~GtT|V~h10M!;W%Qpy z=4n)vlfi(B5z5ao4)`$L#nnCXV%Co(Yq9PpnINmqFg_BoP`oRHhQE~s8q4Ja{}rA>%QS#1G^ zG$O(QAaB>s|VdUiAwiFsAGAY1>lwhw_wm}Na-uQ{VFor?t*F~W}Xl6F~ z`FOM|79V2FySMpdjBsmdO$8gK)p6^p2ek*}JX=9zU3cASgD=R>f&%RU9P`zSM3@?k z(No(|FTT+V{kjsdXGL&dYA?CHKk@Xp1i1t=$Gym>@x${_ctmo(rMqj6uG0xKTH!zt z36iM}63ROZE|)g*FLv$uX(7oHajjyt_KQ7kB6RM+G*JTA>)Bs z%w)q!S1zI)xK97r)57ql19PHv-FaTFAqqx~-Ahub-f$S>Z-9XVm{c)IZ}KL4!*9^H zu1Uf`375D-+zBFxiJ6Z?J@NB=I=Ae&!j0wm9>kwJeD0xCqZ0O{gUzS$d8ojRpkBYk z!Q0usZ#LRFDvl5yZ5YrtBx%FeW=cdrbnpZv35~&&tx-zfITrh4k6?uH5Ip&dm}-XkTwue7< zmoJLu)tGi(kQio_E2ilV`P@Y|`hFf%@asW-V0Tunu9i6j3vmI|btitH2d#K~)eTYW z?7petbuaFKsFZuswTP@UKeyi&|mZLqfcit_9kb`chHhms-J>eC4!-cq(=!>bik=ar>)z z2a;%k|2+YyAT}@kVmUOMutKQ|uEo;RuYU8ysiyl}9`x`+A z3d0;xnE+~1E>2{;Sp&9b%6{q7j@dOm%R6uW)Rqz zjAdkZ2j+3fa2~cq4f!BN6L}HeOcBHHsQFIdnU(J5;x&L!t>@e!;h`NSfgB-tUKG5uMZvx(D1(7+cdQ z4jBrdvyPh>zlJskE6z@5C;#d($LK@5e}C)MyQA zTxHZT%2sUAUMAUkx~7qkLV#^2SRO~eVE7yPeGQ+P4X^dmr)|UnPMxQb-6mPEUV<$6g6y3dbf<`boQuU$_pd@UNaI`AS=OdSE*M+~|8d zPfd+Fao>%fU)A|+_b%M%M@2Zf+$k3g0v0BS_=vQr%rlLE2ago_xc;t!ptLwle@rbv zD$g;UI5G14XDom1BxU2(%x< zoH?^YgnrnwTaP^h)qeb=(TW+->+*=WKGt8N)S@Sz=8p|)Vw6vq$<4)|IYFOV;~gVa z9@c2Ko0}tzMjBhqdBWvj&P@dqZR-+X_y#MUIF8i-X}{-l9u|BICl;TqB{yl!^F7ZJ z3f!0bbS;1j`!U>Lr;VT*3YqBfy^jCEd+rW?4EOod$P@Vs{+eQibpzC9@5KJ^{bM;- zwUiD4FW6Z=@Zi4=OPE>(Mx+@s8Uy0fsGO61?I!qVn^$uRn=g-)Z;{psP0$^Pac>~} zQ7bB$@G5>&;d}}m8X%?=`!+63@BD1Z#~HB?l9A!D7_gs=;+t&VfO^(U2La1T z|DbEpzG^uAtN%j5kPvfg@7N)LD_Nru zZDQUtc>*=eFha=AzIWUei8f3X7;bd}Va>sS2OWypjz{xi>aM$V!;yN6f&t~|yiCo4 z9~{aN%5a(Y%v0%QddKVaO+kGHL6jrS%fr*Ypr>wun4axVF#Lr)ZmO>_fF@l6NAO1X2aJ};=nuKT6xS<9CS9AcM^}8J<4l$W;gqr#5_;}yVUf*|l7AGRr(m?`I_ zgl}6_LcfnhiSnZ!1i?)MvzQ*|d?^8v;W6EzkGE9tbN{qp>Y!-^Z_qx^)X$aXGe2!<*1;7D1>z6)Cu^ zk#V&8T<7faHjCw55z07*C8V3AIwLW*Fk3I@* z&g7WJd30iB_`z=FX%+JUsGKgb9KZXGUJmAf581wv1c>7FlZdRF54RP}mesv<#*yp& zBKgC9SbO1nGwIO%fj=q(=}x}Nw%e_~=SRai zQ%-gEY=Dy5#*JF9SY51O#|dg04pyV5k`@5b>Y%BJ1ixvOe$-VBmhe@SHy~fP#RYZL z>7@L~t;SphKiC+h;bBW5!$5Uvdp`6f!p<68BWX>RzNdd4lGPh2$^}3A1vM|lr7lYLQF2biHA>%A;s|T6u@+PeS5ap zAK9a&4tT$xybDgUfZ z!7}h6{SEkBaLiP2lt!1$z{Jolmoj+AOXtsr(=XiHRp}Tcy(Tyuisc+NnAQcwgi+56k8OVMOl=b)xTqbm3$kqD!hOT7^$1YU{X;&3eoGwp$BfG`ab^E z9o3k1s&5e~dm9(J3r%jQZ|%S*XiX8zgBbJ6%pyOk&p>D4Yb7=*wRf*JMOxizjPT0$ z;3?JcR2+`(3uVga|D4uRj~v*5V{8wq3pQ&T60+gR~ofhCgb43JlGF3SE^>!@)E6=urHY2w|` zz3b!I63M&^-5?Z7vvw~YYNh=x-%~d<%UCnYD_*Ni6|PF3k%CYvG%svE!#r4^!JE=q z3FXu+V?^=VesUZ?Vlv=RHn3BgZ$PeRhNRy{lPz!9bugqbs#JHWH391Dx}RVq6?eBf zzwTKS@k^0ZS0dl29oN3SSJhe8Y>U}8mbRE`lhmcosHA)+fvKH%B6zG(-nnw6!hi7} zV(~2wm^)7Z+hEUGV0H?`M~F!9*)o_rN9-8r+EV6*-&-%Vev&BOs&r{BTCi0@(uyqV zn1c@=>t?~)tevG4PvVHpUedoqH6*Em`R%r~{9)mPl z%G%a1!5aL;5q0@F2j!Dj3TkWq0cxEHCx6#9fH}6m1+os)j{0(Xw+=mBkY-x;eJUD= zcKv=(^!!?T1Hj@Poq^q@08HX2MQj>IaSysLWl;>p6_xHdFp{5@g%0O$yuLrq zbg$)3--y^Vrk%4!I-1gSb`3o42bsw~N3`Bdzuq9U_m9SMs&I3Gi*@Qv655KH(r>es)bRb?k>pldNZyW6Lo*Je zZnF%LVw#9{#lZFj3IdKpz4i_xkFDa5uBRHIAf4#b*x3S-K7>EIjRco0LG;9?q!c?c zwWv07@Bu4L$<@zhmAgbHMk}CnSK4Cb+CU@8 zWMRRD^j7T6#n@mD6>6hzV=djMZHBq@FvseR5Zt%%)x%2g;muH8jZ~+282e0zeo5Eex2V+gf*kLYcW5A{@ioC5FX(-|4 ziv+oZ>^O(FoI=xc>k;jt5~kxX|XL)?M!mu8nw16EHF zE-0vlR~82E<4#?bbW2s6H*Xz|AbE=H$@`wu3yV)at5O{29p&Cfio zfEt#L$3p@TOSGGeYLQVAg5>iacyy%{D(E@59`Tx9^zL zG%dg(0dZ(e6Oz8G`UCicae#*Q-ua5oJ?TJPV)clr(38UM_Cin#}S1z1uwG^fN1ASeo`GmZ3jf#G=Cwu+mv#9MEF~EXdZ-s-dRRXx(yixt9yM%<6i!|~jhPx+m$k0`Ol_&q&a%EX z*J%W+y@E5BrJKOCiQl?p)6=M8-C?yMBsJ(68YaZpn+#2+J=56=mxCJ=)DD(+QMZz_ zJI36e6jil9-2z(gO9*^Cv+OxGY}&SB?78SEP*o?by2QJg$P0w+p~U}maCOl z-v8ks5}X^IZz?tXk_6&Hbou2J&g&unex3Tqb2TX5Q4a zE3=yeIJ`NeAbYvMpT|W0>@1mB5^YR6>VFO$%L42q`EGYFv*>J(k}J`{nE@6W zWg)O?L0HPhlu3Y6LYXU_I%ZM3&P(=bpOn{VsX7K#^x5^BCl^jqtus|5$)-q72`5t;1U=_6{C0STLk4S=qkeeucI1zihFD6X(9|Uf7Q(YNUP( z3aF_k24A^08IKv-P9=)az&QeWm(*N^u!fRnpeiKWtAL&KJgfU)3P{gh+7C*f7E1Wd zUN=?T>xP{f>{LV5s}@aZ0V4rTha_%Xb9!R5?MwpJx}Ji;aVC}67# z)ue@uH%&|q>GdnM{X37bSD5jij}tfefh7sL$*ae1C-hL!sZ)rcmHSS|r`(`}jWlbh z&+&WwW~}f>W)M?b_6HBHiU8y#cw|qe*~PUt)CS#Ye2+LedZG!Na|YTJMLf7U^ZHPW zR+{r_FM}>W16{uQxH;Px%hoex``a9dbJ-|g?b-S@r`0p0*Z8B=5B^{%_xtu-K;JFl z&xLXkz{Aju@3i!O6CK$S*CZuphwMYz_}td!}dJY_&rpd6b|J?pBNVjlV9 zd@nS*aHsMf(svkIs}%c7#CBEY{$8ssy-uxr2wGR#UU%oA`A1)RPO&EH)n?QQhhbBFU_hmlotHO1C( z;Co04t%q#EbQpl^%xKt{&Zah(&#Fw6x0F94FL;6yE-DO*4nrM@VnVl*YE!82!ET&g z4m7TGu9f!8@%uPGLka@?>it!>Fh+@Uu9T}PcpN=8J}VDZXYNr3SfC0xAF8@fuR0j=KM7>c!0#I@#FYo_sy zJV}r@o`4t$z7@{bD_s2tgZ^rOrIP@m9ne70Rph2H53Fofb*j5oA8TCp2|#ygeEe?P zYMh%0vfp@2!K!G-peP3E6#VQIohp^V-z5&?Z0Al4Z@Y#K>Cf zeJ3X?Yv|hqWvNZ`JDSKgGxrncr7>-7C(BS-r2J&e#$)B8YE8Yl`q2bwu0AHaakTSt z6nDdM443LBk!`m9#v*;Swnp;>>6A|STvNwGD}dzd6_4Rk``R#SpiY&Si)0Jm{%9Om zT=5+=GEuHrtG0~jU*Ua;rLJgWRKigbT)-R7)Y;?J$#yVS!y+>XgJf``W0w-h&u*o~ zFl;#46SNvtczhK3v#~JrD=JG0B=_qWe^^1&nf=uC;d4y`bZ>=T`X`~PKXJO$-{|Y* zVu28=1b2zXoO@f3mxG>z`c2MyHN@T&h^QmtM1x%) zN+2C3!(#gBrAa~R2QTQtaLNNiCL=tk*wf!P$;wcFQ8E^im)SH1PM~pSdj&D;@jZfd zIIu_eMsex4iUofYV#p5KR`p-|F&!5_6bPK}(%zzDj|N|)UvOF*^mCW|ZLs5WTZJ-F zP_EBZB{(t<(ttEznLez&g(7&I5|K84wn80{U2Fm$60)Aqn3&S#ax(nfm&WN5#8$)B z4HOk#71Utv*A)&@hedksCk)?w^cm_kMc}8;-nb{ICPI8A8@qX)$MZ2p_A2h(X(geb zdweryx_BWg`0K>p;8A`<#8se9=AogwQ)bxiIJicdJjP9kLhW&T@ZF`Xy=KvsBr<$c z4>tLcn-yT<)D5=nvIqH9G=WOrP{4|`oGD5(=>L%}CQ6=m7X{4}{fHm)WVmAuv!LUq z=94okjW@XU8~tOr?R!&ak`4O5c9n&$xaP?Xqs6p<}OMiu-I9kGr484&rdtN zr;#jPjE8B9)@hr0dQ-*tmNG8c>YXFV`obioaSBMVQK$<*ZC=*NnNCkPF$w@9O)N&; zif(XDi9IOnJsC_no{M14c;Uz1K+3<5!3bSFTsw{tkb039n!HpnvxCb4nT` z1pv4g3K?NC6B7(SU)35LP+V2Oi86B9>`p~Ld@$`k!m>s$kD^Cc3+8a$8YOoY9QFt4 zukp<4tj(w;d&J2^Dau`K7mK4*&`K#M(I_X)mFCqdzp16hUI1M?3KzjS`B2IZE9Xzs zGHA9o%gbH?%iP@7Hp54C{^PB~)WglEqJU0rp>dG!VNNQNcC1f~M7@l{$uCmLe>@ra z6e_ut^$0!5)G2|I73bOMh)@(q;iTEjRH$Fe6_vN|70zCZ%aK$1Z;7^1%8`}}5pLi0 ztA7(bvQxeVhNW(khRY!$A|~=_0bRET`7` zf!BWnN9$u-`z3Lt0uRlM=AYMQi~hIQ;QqNy=2GBTz?mHSeLNf66_^;nXq{=BZ#2+f$E?k`}$Za^s zu1-X@8)t+VoKX{4h8^)~V64Em!4Cm#7Lk4;F1A$Cj|h*M@dLdae&q=-dN z?>f3a*UHQg)>4ZOF|G>g63VSY`_w4{)krWGo)kTtp(q7Gluf3g85twm`A}1amgXG| zq`0k32aD7Hc%7mGO*KgiiYpxr)Y&UeQ-^@tVEi*>pOOHW3vW{Dkew>T3uz$)2qnJ90TY&m%1SE|x8X6V zjssfbn6T>s>HU5K4Sa%`C){9_un*MFXzcsO&1MNzh!^vOj1Jb(BRWo3NMU#NO_vRp z@IHAQ4+Oy2DehJQREaHax1baL2J5>6H}l(&!wziC0`r9@l)l+M=)2MGU(ssY>)AF( z5B)a=JBk))u{f!JA&V}bk}{jJ`)!NIOG}%VPS;(@QLZwDC5cf@?$5ckX_u_D30x2o zqL3@`fnV(QNY~}oB#A%&G&D$n8SEKWA)T|=!Jh+sJmS~7?fz^`m>o2NqV#kE_t7@{ z53#n8iM$kvWv{=5*lXMl>7Z4QTHxl!xSY2#vHU={+=V^$GV06_RndcC{ago>qaBAA z*X70vW(&@7{vJkG+C8;%ghf}w$7x1M&IhJJZia?8vU`W4K|CHqYmGygjw`Atc^Gy2 z<7fkk_GfmC9Toqy!q$;cpT?qU95ApygtwSpSp^s<3M#CFb+X$fnW)_ zFpfi^(Cn5{Oy2B2(I5_1#IW`Uai|p6Uq97(> zYli1SBYWb#qgFOty>J!ON$lKYv7&!61;7E}w!|NA!Fj2Oq`n$i=@XU6iUBjQm!_pQ zW|nGhTTWG1wd|q8735n1+Z|@BD=|eJ{g54A+KRtDP7x&H*Ia-0D9^LKI9Yv<7i_w4 zn0pLy__upIoZ4uwJ~vNF!lE?2rzcKW$q6EPtsuhKG~4^c??l|wZH>|z z7?mgixXpQT7C>kN0w?cU+d@`WJ-iOtxEGl?eSxH>fb~CjIFX2y%!P{k$ohpT_d72i z>`eSrcJ+Jy+1{RfGk-`o7JgHQdP4y{y`UK3#v$VV5e`(OoSy>X^)cTxLT}`Kl`w!N2^0)x`G6=r>C*Tfk|k zw0o}1iXN9QD<$TDiOX9sDPv)dy~L}*W~$a0m#hqeQ_laLdQ3t49VA0T;&=i;t?bQo z%AU|gWxZ=p`QyHtedp5X$x0 zcgo>KWM%VDm4#+%E+)l22WB8}y$Y!X+Ge@okM{07#|5UOQqT2==a`xKmGJuSWg!o2 zM9|hrG4TVqkNsP5QhA>rpru55HQR)ucU0>$)MPwEsh6S;^{*D7xY zOC4p-@&_lP2dy}$S>r^RiQ^c?@y*L8^pak7oHqV;Aqh6V zz`LnIHgOc&O8k!f+{8%^6DAWG3w&JRXtuQ=%E4@AV#k>$c`7e)W5JLYxf<8~e#j`L z=j5e|N;Q!ilwN18_dj=jKMWuj>z!pF`p!?_!ZeYrBZ3F`GdH-8`?DM(lCF^+#)I*_ zB))~G$*pVtp03OE_tKf}ollmI*v$X2IHJZ5+xa9<;G${Zv$HuWEBXI}<~$U@K27wM z{{L)<*!=I+g4PnhNuG)e|JsXOebmjm7{YdjNVP$C7J~qa0*AeYA*KJ>lF#4lxU+EA zk?CI;R*KjuNKqP7X@%;-g|M{9@>KN93th|;2e+0R^uGeV07Czxu9;jjxn^?B diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/NOTES.md b/sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/NOTES.md index 39a58d3b9..2fc483356 100644 --- a/sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/NOTES.md +++ b/sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/NOTES.md @@ -270,14 +270,14 @@ The OBR review (`harvest-staging/ab2026/reviews/verify-obr.md` on the staging br Row indices above are the 29 October order. 2,797 rows: 2,891 − 58 − 37 + 1. -## README rule 10, 3 November 2026 +## README rule 10, 3 and 4 November 2026 -The IFS review of d5e323dc (finding 3) found this family's totals of the Budget's decisions on `revenue_change`. `rule10()` in `stage.py` applies README rule 10 after `review_fixes()`, and moves 87 rows to `decisions_effect_on_borrowing`: +The IFS review of d5e323dc (finding 3) found this family's totals of the Budget's decisions on `revenue_change`. `rule10()` in `stage.py` applies README rule 10 after `review_fixes()`, and moves 81 rows, the OBR's stated effects of the Budget's decisions on PSNB, to `decisions_effect_on_borrowing`: - Table 3.1 (42 rows): the total, direct and indirect effects of Government decisions, the direct effects of tax and of spending measures, and the two memo lines that split the direct effect between decisions on and off HM Treasury's policy decisions table. -- The policy workbook's Table 1.1 (42 rows): the total, direct and indirect effects of Government decisions, the direct effects of scorecard and of non-scorecard decisions, and the totals of receipts and of expenditure measures. +- The policy workbook's Table 1.1 (36 rows): the total, direct and indirect effects of Government decisions, the direct effects of scorecard and of non-scorecard decisions, and the total of expenditure measures. - Chapter 3's text (3 rows): paragraph 3.3's spending policies, paragraph 3.4's tax policy changes and paragraph 3.5's fall in departmental underspends, a part of the indirect effects. -Table 3.1's memo on the current budget deficit (6 rows) now proposes `fiscal_balance_change`, and its sign convention adds that, on this line, a positive value increases that deficit. The workbook's 288 lines by tax and spending head, Table 3.1's groups of measures, Table 3.3's costings and the prose costings of measures keep `revenue_change`. Paragraphs 3.20, 3.21 and 3.43 give four policies' effects on borrowing, and their seven rows already took `decisions_effect_on_borrowing`. No value or printed sign changed. Counting from 0 in the 3 November order, the moved rows are 2190–2195, 2352–2357, 2490–2519, 2521, 2527, 2533–2563 and 2630–2641, and the memo rows 2642–2647. The contract checker reports 2,797 rows, 0 problems, and a re-run of `stage.py` reproduces the gzip byte for byte. +The OBR sums its receipts heads, from air passenger duty to visa fees, into the workbook's total of receipts measures (6 rows, cells I490:N490, rows 2190–2195 counting from 0), so that total keeps `revenue_change`, the receipts line `models.py` keeps apart from PSNB (delta review of 8b7b23db, finding 1). Table 3.1 gives the same figures, in £ billion with the opposite sign, as the direct effect of tax measures, and those rows take `decisions_effect_on_borrowing`. Table 3.1's memo on the current budget deficit (6 rows) now proposes `fiscal_balance_change`, and its sign convention adds that, on this line, a positive value increases that deficit. The workbook's 288 lines by tax and spending head, Table 3.1's groups of measures, Table 3.3's costings and the prose costings of measures keep `revenue_change`. Paragraphs 3.20, 3.21 and 3.43 give four policies' effects on borrowing, and their seven rows already took `decisions_effect_on_borrowing`. No value or printed sign changed. Counting from 0 in the 3 November order, the moved rows are 2352–2357, 2490–2519, 2521, 2527, 2533–2563 and 2630–2641, and the memo rows 2642–2647. The contract checker reports 2,797 rows, 0 problems, and a re-run of `stage.py` reproduces the gzip byte for byte. `scorecard_db/signs.py` now reads the workbook's legend on both metrics: on `revenue_change` it runs the way PolicyEngine's sign does, and on `decisions_effect_on_borrowing` the other way. It reads the six workbook total rows whose sign convention starts "mixed:", and paragraph 3.6's two absolute averages, as having no single sign. diff --git a/sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/claims_staged.jsonl.gz b/sources/harvest-uk-ab2026-2026-10-29/uk_obr_tables/claims_staged.jsonl.gz index ef0a63666186bc78a126c1e79244140f89a135df..5dd582647a3398d726c21b8414f468c6f8b69a25 100644 GIT binary patch delta 15229 zcmY*p~%{Np^&ojbF$Gk5MYv+3)2;p=!I4S*_XWEAh!=gz9p%7a^%W9=$Fr$Qoc zcG1n12C=Iv|EHGptTB}*5ymPnMMNqOwp!wdpublt_s-v&#V^>|Uc%GguGQTa%s(k; z-0?V~&b`t-Vi-}px^C2c_9=_B?OJT8O%>1iJUJ?TeN*e@>TkU&fkL>x<0 zOrbek7-^rZ_=-vkr7je$lX6UN3*pV^V4v6aoY?vs{|*wAro~gT(VHANtB4w4nI;qh zaJDOPXu+R9|BH(L`2&k$LCB@RXGZmwLDfp*dpc6z8#Q1%i-Q_actP|J%cZsK*xpUz z78322>HYOr-fF*){?v@OJv7g{p5`>s+EYy)M=Jy@GzLy7Wf7{|Vz6GO=bU?w2;A}2 zf~rVa&?<9f^7ZC%;%72orA)3=3~#=lcD6PCHS{v8tPuaWuqpcfx7g%rPF(7~1->^( z`>(dtnOKO>qGSa^X8^zRj&=Sh%A%<-e7~Wv#(t~<_kyw^go@h;K7pDm{4Xkmi55dJ z(S8+7lriF8RPa)-0#}0+6Lpq?Pr!>NR= z(NPL0JLJfW2vLQ&3QtYa#}r6CN0rg-%Il-Ofd3=PF*4*@31 z1DD9bh#M?OAX||M2jn=7blvrI@Zp}mJSK?+ab}if?WkswN4c?kzg0rDE-tmKM4(BQ zaMZBCu+Y^GHptEtfE^eP49W|8%04P{$GAy`q=OK|;i!SYV1gsOIW?Hq9k`uEjD5-l zM{Pe2jYv})H4)EkoJ$;ipM}G- z9XSlRhmA+#8SJBSYpo+vhXMSBxDE0fizbAKtq>~NX9{nk5N7YFk55SIVv{InkW?ia z$5^y59DCTyeN8->z!0MRfkgYfryYegL5VD$hbUTuEqu(Zj$&VjS@BQd0?qcX@f1H8 zk7me!nsdL$8PX1x1aKwFpZUO9Q0?H>g-7K-zy!Qe-MfsqbRf*cx?@wKow%kLfjXce z@C6*yqXP>oFQQO6`W%ZdkA(yslVYQLHuWG6y%ReQB#>q!IwOA=z5kpktzqG{WBDu6 z(9|~L2`ygMhw)WL;ShL54D%z!?9_*_h(Z8DFxzrEJj53)qQI`g6eA!`&@?JGj-E&w zJdG`cX&z!(G$ba1 zs(?lA#|v3|;aT6&E5lEOY7K$bWJE;Ak3fSq^3i&VG?|nJoRiKxUYkVBi#RCYL8y;$ zsyEmo(;X$ykzd(Q(N95!4VMQQ5%`dy7PjZ*Vx3VY5hjS=xoW8cw- z26PD0Aa|kwLT7l2T-uemVXc`Ru}Sna!-dJB|AtpicZKSRr^gC&6Jr1AsbID2{wMum zJ{OQg55OajD~;b~#Jx(Q*KzrhFwH>0=GcASBhV^z^)6Q~f4-PSfWQJ!1{xMw@lzf< zkgW$v@Ek|O3OiYh{6t&ma~2Md++~H4x?A*S@`T=GHN|lhCG6oF0K9M3+%v=hUV~2& zBp!c0vK{+tQH9b9;W!q?O1`4qk;E4DYx6Gp`SuAh!DZCi2_cDzP=nbqc|Sl3h7XA@ z>F=m!PYm;wGP*^Cb?UHs1h%n6SFl^-JP#p8LsV_T<}}*z_LvQFUtb5WdFo{%wYo)g z;tL+FDr!mrFcTUG%a_DvCb$L$?thE?l=o-^KaN(JZ)J55U$o%|bRY1r9Ut||Zok8o zXR(X?OfNv9zw4|}=v|cEqzqgjBE1v)Awy?mfz@pU4H=Gpc$C7|@Z|VnQiv>D(eX8| zsIa(fo_nAi3tJu*4(2P+(=DLhISw8Tp^sPHU>v_aRvwJLODt5%tvIBE7F07k6Cv<#! z{GvNjOs?vb#`PlsgtB4b;mtIcD6OnElqz+uo(%_bH=iPbGciQtGKZ zu{Wg(@w0U$bv^Pf+`YzUd|zg5g5Ewot45NDz^f%(cOb@?MTwYgb^4;f z46~@0(n&?&@#gX$n%F7-EqXJ81fyCug7U>{i||7W*x%Qv-hfVj8PMOb5CRM!QQPk| zYEDxgcfn`_@irCJsZ%`T5r!h-P6O&-gep|FGlzn}0wYC(V7ix}HJw6a#9f$(Rg7!u z8b@b;7j0Ro!NpC)oIazmvxByb*#~ejETbDI4~GWeJV!dm{$jM+Bka;8`DkKqGR^zX zrIa=5qkzh1kCeCs6uLjUR87YzaUod4V+DeM#q!W7)RSeS;@WMn($Zt7OJ521@oQ+s zOeC#KWGuEmj6g{N`N%^=e<}&LSh8cn_)#d_%Du=dhL@EP24Hs<@K)&55eiXiPaK`J zD@Ow<@i51X5^WPq-&rg~=~JQ}&GQX3FzYUAMBfCnQDu;4wN~5UmG%y9Q0UW_Q`2gV zFKavx8QkM>%D*B5hQdISN@N`CD(at;;E@!gMbGD-A`x{(--%M`G6ABA&S+P>=8<-y zdEbv4{x)H~i~>*gn38Z2m!NSX5P-#51kQzKw37!dpf^e%0*l;NVsy6fK(wZ>i2f~> zd%Xu&))F*l9hdQsD>52k&_9vnW$|tbnMJRALdYr z&0~mex}NFNerqW+xJrfrjwXY3blcoe)A7)CF7d}voEH%rOUNZOcZKjTzmCh1dqXYz zlIr3z9YS3z0lI@|2AsP&kq=o?1|d8{$pQ(BsQc2u3QHfUMPyy!LM!TCp?9-dSijE# zY_60CaYc!ou7tM6!kr@2r2$|VpiW7Wz4i#CMn(ia{z=B8E)Teijwnj9a~qK)!k`fD zSURM4a!4WoB9A>D?yhFtN011kfowdZ@D6sDPtcQrTv>1)9$j&8=q?3&3b)?pg|oNi zRVf)hK25?lfEh;R%jjxsg_DFxK7{4xeVZ~uX>+*aR<~1j6Bm8zbY*f^-l(>21OLK* z63Xk2tKNOj!X^Xqsq=C!tEttY$ers|!#2_{<&A(aZf`pcO;~R+UT^Jp!q}x-lEl`y zBKOEw4eJ4HYFk9Hp~6CVW4)Vk`p!~95x0!Ivh<4UUn#5UO%h*P-(}0?pIQ^ikaW<@ z`%WDPQAYGK^4F)YsN@{8(X7DiT5|g7#6+A}vgIjKS!;^>M2kzoo6C3$d)f_}$l2Qa zn&N?~F%7$C0!ld$;J-OPT2cog=I-X4MQ3Fzs?c4Nkd|43Efc`8jfI&)b~5)f1ytVC z5a2hp{^ejFl~HPk{(VXE4k9jnJj~e-3ccz~Aln(mX7Qtf2Tbf4pXrCSmb0cEc)&Yt zl{R^-u`vc33boACb4GGlqxK{qUz39?ML4;4SnEso5M#@>O6qL?KuTo6nbzwhLLD*` z7jvc3oEw_-BDBn^oSEdWT|>b5!V0Z6vdr1k(#Gus>?E4Jre(_N>!UbCeQl~f9v^Kh z2|qUd8?Tt^lh|N< z?0?Y=nQ|rU@VlrzCR#Kq9fGb+EEGbFMguSIWG9NDQwe2(iMA&e3i^zmd*Jve5+}hZ z_SAmkA6;jx)HpV^1)eDj<2b?)LXK02^wb8@Akfk|$yfDX_a1dmY*rzc=;it|KhOP~ zCK7P~Vk{X}nBH&#+<$=nJ3tO$$Pz70jN*C% z^#xsQ={lN_*4m5+WAqDgU3y(yY%=IZJ4N}I8@4B@F2;#zMYx`1vT!cTQ*U5fjC$u9 zQHOWqSok4Vkez!n)C4)Cv?6_t0s%jmG3i1=+eC`(PFKaza3_m~_djX`j(b=|>~)%i z6t(W71V~1EUYwro9osjMIZbt~DF)?g??vv)M^@!m}wcKoM;&>2d_lA&_ z%gPdTJe4Lxxv?BqLQ$=OUGL^#vzRpkU`cd2gCbFs4lv+e!9$~az#v?enO!fu3qnjY zGVBlGsw}22aN(dWL%bly@MT?EXYRZ0o(a+hA(%tSt`?-;YOrxvvDk+&>_O5D6c>RQ zw~eVnq&s*R-BCy``$y6WR+BKC)eUrPHRc%GLT}GhsK-1^aH?Nk(_MNhhD}@;itM)u zjTomWE%l2cN3UPXe#pmCI)_f|+_1QQo?EO?2843d0@GoUlF=MH_pvqo+MMjSh$$`% zz_k2*GsjKL+dQ^?mx{d){(iix0CdRAaw8}Z zJ^-%31uyBpr+?6`AU2Pdc9Cc#Soq(#jcB=+lvu_>C|aeGj)OelD}N}XXYz`MH^!e~ z4rl;Zl!Q(PxOw$1uiv!bD4si-;lmzp{({ry#Rv5XeL0=0n?JD-e#-)V{qelc zSJ>#}{mdkf7G8VPM$ckb%BweyNK07?*YQQ=Zby!Tqt_SXlt316fQZ zfpL!;Sft3azw#6BI&laZU{ib+r$YNCnI}2TP^z} zmYW|NBRE2#tR18jI|2YsKPPyb)si`wz}FR-c;I1=qSc4sW@1rTV;7;AIj25jYwZW& zbiYS5!a2ETlN5UUcHcyvMK37*M4@)zp4Nu(Sy5PT9#vC3voxPdYBqoq zgOLBe7zsv9V^Lg*vOIg#2Mm|t_I{tvOHogigrg;i1~adB`S~@b_ks%rG^qAeV z-?|fg-^etlbjmpN1tn&NLhzv^K8v9ilI=L(WHGlr_LpLR?g7K;g*WCalJWr< zfe(mCMC)zPlA(NAkVd{>)LR#ia3!J$ac;nzB38%YpCXAs9q-8k9uX*5eOpYI`Rn|@lbGBo zG5K4JYUCGfBNTyVc>k0^JWP%S8i0o;OC*E-CCUz$AG*^EaYbi%?Muw4FEx5fHCQ(Y z>1-R9g?0LA>Mmn2xk4aDMY9Vy}a2sQ(jg zd`w5FH5P6m7L-iDC>*=l1PixyubSbMhb**JPocMG_v4^2y<{IE7|bYx5eK%6xIA!PInkTP zFG1eHe$*o$0;wF;G_Erfk?_Ue){9U1M!0JsNwg+M1|1w-? zzx*^HBN4&y9^IgrdRNp0!we*|nRlE4a()X{$hx&q(hJszBZu#wF014CS@AF%yd6is zYi!?Im1MC)l*xe9dz4zXRL5QI9-}o5ri@J`RkaJVfVbl{hsS(>Nq>%i~uoB#Gp3sF(oS|%51>B!z+eVQ!v$+yC3V+h1_^5 zmK};)E3d<1YMMvJL9R^7dd30Y(b3`ewY~aG|c~!=V#&O*qdSP*w?UzV$yl==?`~n{AkL^@baD8$C z^QyntAN`y(117c+y6sfqGsdMN$*#{I6FpqRLp~4XuMOW~d-BmV2=rVE+r0koz4;eo z87aACBFWXyg~N0$u+6OrpBHI?Gx&|Wu+6XbV8Iicrx>1~fgyiBw#`oieh<%}LbbS< z?zXb5r)_=j*RFw{yZGwvyT)|#t;?n-c{{|=2J$dY-7$shetKSK2Af*W;h%my8;#U0 zqzYQoB6+F1%Kta7IOONor4dkYw*yMN1|PuvF?%TomV+ z@+C>R*h-ZLh$$q8agG9CXuD_E<{)Xsmdq(OAM>j7{y?`I#Xh&hFI5ga|xuKlSoo$>K9?jC)+-@poKe+4?Lshi(xRA6B(rnpd(PyE$7O?+LX=dD>Xj=pdM?tVXkG2A zgM$(Cqe*7WfehQ~m#PcaNi|}m&6m{wX`^jy&v^G2f~_~|gT^C}*};OA5y{t?GG8)ZECP7|ce%_K7J`Oj2z? zXrziUF-k6FQOx{_{rtp;F4tGmQi}Ze5rr~*)CXg(6eh?SSa#mx!2K3Dzr38751{5X zQQfGNwvYW4U*J&{@raarLmDG=FyuCf1d8|x;Eu<7Wjs+q2EBMoLd!ZF%x^Am@l zi3IZz`lV4g;TAqvuh>FZHxazR@Yc7Ay=oEtv`|OWpCPDf96c<8`uIh8{c>rn9^r?F z8|*(r;Wj^Q9v+%nL~I@=MG{pqVWtf;9e2ax#tci30RX*reZsq9UU7wNR)_ zV-mpKf`U&JF0HHUx2FK2tNF*_Ut}7lc6tO2ER$ndQz&)%e(#I=UFJJ+hZ>RSQ(3= zGqE#TCJsM|+7749MhXnV=sPfgD5VUEol>Bi-)ThKsRXo_gR$>aLfVMK565khm@XwC zphPcsk$7)@NknBU)xJEVCHapTr9e9{cfhPEtC z)4Oq%Y_9Jm4I55+-k-gdqPH4M0MJ{9!PXub1tU^_a3dd-dR)=1MSqTghC^}9 zA|dSjJ0M)MGb9DuQ3+u<_*HD$_DdRCa)NPrI<+wD*RJEWM--*xzjJ6V22(g&qZQn+ zF|NNZ8e%r}8r%R%p}JsA!5c?405Nn;5RL|JC5};6H8buV^=vgl9*8r5UrjA_wDXT@ zp>ig&VKgIN>;Q)xl_WjB{3R=_rmY|+G|?kf+{OQFBG{r3Al9#Hsob2)%OR}Jj5u>p zoqd|oZdaD|ZrUT7DEIr%0>{DuEi5n$z^=pv2<_I|);C^;Q6-xbTL9@Y^D8+eJ;x%%?>r0(;LNacOHAD+O2}xGp8N^NTYkQ=c z%+$qBaB7^CAQQJFAJurLKH>-&%#b1nGyDeK?J9q>cc+f&0 zETO19OQ`Pbie{I|<;fWsS)L14@|+#_&3GI)@?>flHxxz&gxi;B!Hn&~irakSzA6pt+vuqeXUv}eX5*5{+vc+Hsg=nvUroaH>KkLYCnT) zy3I}vkJkwb|7ypsA1c&{Tas9|>$5iQ)er_WJN>HPP2jx>O$dYM)rdJFe_^U4obk!O71_Y`RKXViUP!m^pX{DK z*67ZWCn9|_NC)9G0!&G@i1y}CLgr3AAe20aOIbC&um7na#(ddRX9=a12o_;^2{zzF z5Qb?54wVM+WpU4rzTV|bx$i?HqQKdYw&#@3f=bibFJJd=PA54RdunD3AAH=T^&tMo zyv(VTJ4H((c@UrSwa&fR!#pFNO~=`*J=!Wau;o+UHwVj@;F}Y1Xeg33RdO48PfkXo zDF^E37ol3y&8VwoNxU%mdOaeR=`(Tf-ObRaH?F<4hIFw@mn4oZGo&KN;i4i&t`fqU zGiXsE9LIJdk*p@+u6SGP2sg)u%izqSn$V;oPdk-nU0fVdYrNq?{|b81olacGGe;GP zqb{Ow`c`j7iNMYJB_!c+2(3_9$ZOYO?o}S4E>Y26>O0}I1S?Ql9xlhsA`YE@Zy$-p zKVoSSe%PXcq9wcQb41};(L`zVqQ7Xfz{8RxqN_dLKCTgpD&NiGVEC&rsmE-6DNY*d z!O0z@ZC6uf1+ovy#G92o2=zDg{#E{B_c~zRe+=cLPLb~{&9ab@&a-Q;*ZfR`L`PS7 zoPlwL^pdb4b6cd}XGVjSxM=OR>83ldO z5<=3|&CrT1Td+OevSK4ReU{|m^&^T%>>e@QK}8aBVESo#L7mo<3`hMYMdpn#@86Fa z{2fzEMP4*4nbWI-?`t@|i*WlsuEX#miL4(Z-T;1(?G~I+CM1-q)g>tvPYC}P+-5)d z6P_Trs4;nEghj@%eBF=%9bY;fPcgYpx$N#ffdy_AS-6t8S^afyN7js_Zqxk9XlNmoSk5+s4E_jg_e*La$6YiV4#`m*ev_N!yE?Ubu*J|5wD zF^dG>s-rR}0$$gSnnd;sNey%o@xA~xNaFF^Z*@>jp(O3UQ#~tFQD+=S0;q#M#(u$L zP99fId-PYoaZW;G?&U+do#9(-1Tq8)(0t{j4-XRIRusvf=YESm`yTe+$AoVw5;lWE zsq$UM;vS5`tsvt*%}!1kb6SPC4IxxwEaJuCjN$Tx-x{1R&tOVcl=Xq%ZB5J?%#Bz^ zu`%aTed4dt;?ZzWaxM}i z-t?Wz{_YypKr6{kbuM`?yl7$Fu$h<9^2q zkyhThjg3(&1K@YlOMRbbf||a~pqW8H%rW&ekql_d#R}?fly8~`7uV1wDR_yhHEN4s z!{tGsA~rC91@QzM9>~Tn@P)(aH34#rIyY0_%TA5q_XFuoZFdk!2MZ5d>N}FUAt=N0 z49oQo3B}_c9ua3(g3Yp29vBcELUx#jMP}e^CLiU7W;frS;(->|dP(Urq)Z294m|nC z&>#qq>?W~wk8+E&XWSzhqaMKY;T1*eNJq=C{GA0Ecy*F@zAbV5F%uca#3|8tAyqfs zYR@0IYwg{dS2y_tS3D#P;~-LJ%~R)!9s^0#s?GKz`FGss6r7LydcOMM*Xd3ryVluD7y8xUlbK^}>%rsoC32h04%g0!-XRX`yG5=@() z;Kd~bkW>bFY&9}ZOtyIi3o_3#_7(ymI3-B}c%qdwDcGa#uCWW3KqL@ECxE>L*^u;s z7meM#wI~Lel7`&`6?0Huhhy=96G^SV#=Q zlTB7_Y6&lKpCsD22Cn8j8(Q+y3yqHD{)v}p%_ZcmFi771aRafC1Sk2nswNY?7c@qx z!YuOCl|;Bp>o`;q^OY9?!R;&vCX`}}OzAP6X_~Z>jcBs;Jd~H&F`JwLg@LfBW0BNt zL+;j;dSA05l#}Zp0kTWW^Da$NqiMj$Zgz>2;^-f|(#QkW-5o^e;1ouFiy9Fy<3zetDD4GC>ojKonQkAr;LlutyFLF7QI3nq(9IgYo6 z#OeP_2iS}MOOIzKYT2%fXLdKMH}!q0eq?8#BCxH9Cc`a;C4;q~2#PRnY-o7v&FkpmLVpEl z6snuUPk&08uo$F=gkEf=alS&?;_AZ)u+tWmZBDtPSmpIMp6SwY69iBwg>v_6qNtUN zI2E8v(Qq57aV34dn7fx-21pC{4i!)z$W0J5r%EArQg~8c!!4jHV*YKU{g*!###aDg zIm7?V8C-g*>-4^x$055MB6B(0A|`XpYxYn(BF7U6ELfxngzG>Gb$5D!1%9XMn_rq+ z&XktsdF!^Xn{BmE_lgxexXyqD%IC~wy7 zQto>mO=?{=V)gYzQ~7VR8E5QWXIdSekNx|(RN)&bVfGz+JT<=vrP*p@$VEL3Een@M zCOOBej(<=kwrUR+Ee7(LVryE1dD2qMI(77G(QcYX zSK!FjS$2=BD7JyAL|jagPeI?)OcPp9&bKrJiHV_qTaWG(!|2@zM5Gwa@~{0?!d-jW zyY}7J8vI3qD~aTu@09|V`;ToLc?7B*`H(?K(<$b@?*JIVulN@wdbY1}4jRYShSc(v zez^9Hk&>F8Es#F+tL;b_n>)FFe*R$>^W%ufgTrSMA+<|k7WsA!G4xETUI*4;=2G#P z>oH-GCql(Zu~C*a`M;MzgBt_e2dQZ-=u3?{iIJW$qQ6d#*E^_{Asi53WjRV z-IE6dMf{VO4+Eyxa+A1(MXltrtcF;Vzg4{7cpC81?iO#Xpk&Xpb~SQjSPr0)ba3kJ zIPtf#XYZQz;`;vJdVr{-$aQ8?k{x#qKC&b29O6ICE$D!tJxC6DldvLN*7W6laHR$O z^IUpb#Gd{4lsYd>AL_1m^ZtM`2}#oRCD~jPob?l^%(p)|?)h%(X@9kUvgs}%WwOyDVp=~{yt0{P`?g?GmtQ@nYm2PZ!!2JBApAh-U z5XT=zx*FQ@olUTN%gt$ZS{T-U6x;Im@x)L(N;W9V+C1Vb*gvLYBEBUN2cC$3D9_Yi z2{^R<($8y@4f`f~DDCvvGnMcV`ZtG>R+?*-AJ6>3A+~^w0#iSRxdyLa6Uz4OtCK?h zUc$5Y?~UaiygEEi`{2Q8xBfSD*B1F{d!@zd&?UPSf!z3ck3LuC zMvS~YzP%V;7G~MV|8VKxOE%*#{LGW*S6ZFdHi)Y0-2&f;i?T4GV+S8WW#8P0twTa) zPxIsP_lRtU9zKGhKBL~RBYZJS6K?6hj}tO+-&<3Kb1eLEtTT8xyu85oWkMg>U0ky& ziYa~5Eh_;}TyzHTO%@5B=uegIex>Bq2;*)2oSgST`ar*>SWsigej~S~p62<9wToUF zo+%cgwf>aJE)kVXEjHgx$N1jsKCGC|cAk_XFJKJq^J{At9UI*LPz1XqRJ>H!eDiii zXI0;22F>~C?21f#zw%Qd&4_&CMr@4(&c`=j*BFCrb_8PU!|K4X#JeLSXM9!)IM0;s zjML4uXFyuF92l=c^XjO3S?RLm?R7;2%zT^~gpE;PVcj^eZKEt%w}e9S`>L6TcnZ)r zve86&^a@lZ@A6^h8chFSFLW@lGAXidQH7@F58aU$ib$)>?>z__iXrJxgOi@iQl3s<%^*e5DC%`d}AuVp2 z(6VXrt0E5*R$8e(i|Ap~%G17dM?543r+{l{hnylOy4k-T8+zO{mEPYdcg!JVK`Yg$ zYjs$Mtge&eY)tXmBOk7wtCr2`=t`+GF`#)bW0x=ern^ZfT<7=WH?kfVsSP6cj1Dx@ z^8U;>Ie?$J^SrOrGaDII$2DImr0_=Rs5x&{Qpl7jaJ7okWj*-y#g1IrMuTE#|MP3H z7;-nXDbExQejSXe8}4CfRX6pM*!Dm$T~a5314$L$yxc7kSZyoJHV>*Zs+AzIRp5&X z*DGtat+Qk7_5{}Lq!XC1PzCkGNA>uuFZD!8DDXXc9v@~(Mrg~jTk+16z@OAs3T681 zAgQ~!_JF*u!Op3!VZ|{D^Ft^P3`T*kTmFPHh7rg@wKsG2dULG3Sdd`Q{+Pk{P;|ut z$z0u%Qgd3LVMVE}2YY!wuqX1B{IE?zi2NGHo40RA_2qg-F5S!a>2D=};N#x{ebh%; zA4pUt>&@IQ z*M~A*pZ|WG+O|<57koamqG{RscS#&^gLKc9%~$(BYUlYd_KhdQ@KAs3^^@Qc{L>+f z*g>u^Vz*RvU)R>LHCUuAJ)FJ3$Z727@!#GX;$k_~g*EE|(wr=$bF97IYKNoaL7Umt zykPNyTS;J(gdJx}0>+#~90rhOW(g)`#2uDC?el+ITK?4T!5(r?XP`XQ`WpbCJs-lI zKIRqVHAv(~(O{xr7yfsY>6hZmxSy7~jemFG#Jm0zmI3J}<7_s%Rt20jLX9zBsJrHX zTaU=f6dsi0pA=f!RZLE5KaklC3&^jaF7PoQNGdNA+=+46&n<7H`NOGKgL+H4^$(k^ z@rA5Jje2Vusm=c5NaqXRK)lj}AZ2^LuywTuqO-(>mXuQLKf(g^NVmk(1L~e`QQ8`m zuNMSNC(Zvf%uMbT@vHtZM0ZeaNy&Kc|HGnPx27otejI6iS>=mKP<{|ZUfuF{_O5HX z1yOsc{7R|P-@Uw{lR{Qp0weVej*G|t ziY39`H%QgPvhADmg4_6%$QrX*m=dQ4ZK-p>%YvGrnhE0^1}ssp_u14INA&%gFVDXS z78kg!TOGc(oCukAzVLaSw0646yU|rA5&0{`*)J&Q#_tbbpWMYWzWc2OIWJRVTZ53a z^WP5jhYKR~!~8?D_gXBfXlv{#<2e;Yp(2}WNU0RCw``Khy3w~~Mw0Q2e zF3@Hx!K~y)Is1c(%gY~0$T@_xiA%11mrkyQ#S=t;$#RR__o2TIvrVxzO?j1aGcHch zM)MM#vPStt-%w9(2b>cfUZ|b-10ZXg13oDhKEbI zmy*%9QrWlSblOZhPmT!RIXY+k?15Aon!lO7J$LgmO|R(^J)ghIclMq9eVs7WS8&<> zVfweBDIeRmu63RFP7~42VT{S44)SjXz3Ag1!;JGz+3p@^X=bZ9>b+5&M4s(eL>PIa{cZ1xt!zh_Kb^#pCtM4 zwo}GQd&{+0Qke6+`jxEoSM0=EEOZOj)r*5~!&jTx$RGi$#<7*{`vn;zayrPNjF3ob zmjOQJwv7iZQxQp7m4k1V84KVXLNz_l8BS+IZzPjXa^tpzPJSP$+<@nIyb=bB%Fx>PlCp!Djj7jeU~_A8J4Y3I^f+b$k;SM$M>;NvcH?u6T3 zuk6fszK6^=UwVNfK#ObO_0qu$kbm8~`PZhc$;|7EffFyjY*0JRoru;yZ<^a(I?-wf zTK7l1(0eBFt-5|Sd-y%|N{wBXh3fYjJ1@Snu7T$sHYu{}p|Vl3e?~@}k+&7boE29N zow-A~+w?u>&!o&+)7dURFg$2Umzd8oZ53@f(6DGxoDCQYeu}zZ<~a@ga_KcDs@`~- zSvsRyP^v4tX&G;Sb|g_gp?v8da5*^}NRZhZwnnbO$7bO{pe%xuqUdDM{7HBN z0*hG=oICQz?GF`JQ$aSN1d^$G-Osl5g`5td`9bscw?|?E2nO;etoa*j16pO7Q6c*gtQK**Hs3 z;yaHQ~|!+AAOs^qImLdGM(6+o8eo5sqF%Ja8Q-&!|n z1Xp4N>Gz8HEU$8yc@X)?fYWX*5ehJ{r2eMU?Kn93;B{C`SL}6VV2s_ zrdq26a_U%l)<10_zn#dqnR+B*!D+0Gc{1r4dz>e=LI3m25j zm-;-w^CU>px$%#{mGeVp3eo2=5+TEacE61-Nq`Aq$6HIBM9&&0{W{FX%8?M$nuE0= ziNVXW$c(+`N223BPmvy6eK+PeWs~Rh@SYFL2m|THOnR@_8r5?;5$3tf9-E7~w^PFl zt3&7(ztz%2WJ8t3hfhrkgNGL7oiFviS#rK#J2g5MjC?@%l}s*p?jv%A{LST60X}KX zn1ZU~c_j7mz(Q@sQS0;Jufty_#6@Io3N;YEb`jNo7YN0hBgT+{c?@HFQlFfOs!sh9W3%e&a$F5 z;6t<4F}`e-v3t1j0Qn2Mc5Pxd=ULa@?};|K ze#P5+Ub799};AGWDt8ug>Y^!?^(4bZjrX*{q=LzN?#< z%4X&2KSH?+kVBL6@6BovCttNnfA2w%8{dm4RUm-F!VxKiwmZ&t4k8Dxh`hEPqTl5WcaTnH8TT6qW${3_xfc-jg zylAc_tUQq0SfC>9Yda)-XyNwBK}CDu@^2MU6;a3S`}L4C8=t4OB2eRqKQR-+#t|A1 z@9%#U@wT|#wN~LTO|v<1!RE!`?V9IRv|1ail@>RpCE!)8M`k(&)eu!4)T|FCewq&* zjx|%`?RskEWQ=DV5o8h}?e+}W{{PTj{)fe0v9FgZ9;|QIiRE%tb3A_N$}$(Zo_X2+ zUXt5_!8R_YtE-}8S?>u$dckghCBxh4nXZN9#HM+5Zsx{_+0wY4(4L~d8RJZ~l2tmy zRFTWe$WggL{IQ=;R=3@4N8RII9tJZ$xA#OHiJK*(DMJZUyx)PIq#t*`N^IS@v?AXM(l^*@x)($H zKEfik`&B!xdXq6wKsCTq+4)~qIQhtlAQTx?s@Kz_Rt0(zRr*YMQKXd17=5NZdHYIl z?^)t<^~x*j-YOdtSRJeTNzz-&&}RwxmgLg&Cg^M{ztxs63Jg=?6_ig97ijyj*<8lkGi=w)1xAX9af6&@a%{OYzLm0V|Z}M}AJ^?~!Y| z-{;Dazj>L1M>WrvhQ9mPQASL++EC^BS~c9}^0_TNZ}mA}zcpD`ZkxF<4A!T0J9k(c zy_yW_s`fTRf<2QAaotQFuQ%jx=QOk#KI%Eq=2PHP;ChS-XiA4wzYW;3QhYalRoon=N%5fXp$4XWT( z@G{eI>^9VMJIdwdx-#P;#+nF@pV3^*ZFbv)#BIQ_wn*w;A4(YVHuY^9cxxPw6PhFisz|O6@s>OA3ne;fxai6 zRve#Xw+9}v%1rR+AIStdAiq{B^)iK3L1s5KM=-4n%d9xIf+q!cE0s#_)7z&l{;~Q~ z3CxgqP^ZD(XU?$|RAji^q`_Wh$syqCw(20x^O@G(Jl#}RlINlxF4^C&UzZ(A2ZwqwOp+CxNjPYT+ zox8dH`}M`=7puQ6YsSSt(==-=lPykHKk2WRUCBG!kIUQnwLnb3Di0P031M+(d8JLN z{k*CEt^Hg7kbS{yP7-q!iUJO(*u6GjGG8tl`*Lx+-PLCDva!Mku=W77sFwC4@9GKG F{{Sz;_8wjI85H>)VNQX3R1L;;2>6TECkcPqN9^EM@3epJDB_SyyA>Bww zH|RV3UC(=6&vWq?_kExHe0O%vC(d^v>%=MR!~r$rFggbdAOFH+zZ)A=<(^O_WRpuP;d1ilhemK_j7;yXPmdeLGM2fV3wrs4erW|sd9OqqFJcOUVUtwpZ(jy1y=Ybywa>wMU zmct%bS}+zbWpdk%%YJ%vL@Oi|uBj>*#6qP*{>vB&RtSX0Kvi2|3Y9-^hc9KrpMP}R zomBpoI#wtY>8RS*Bg65Id9-(dCJXZ@V<8E0nbn+ywm23S|5Eqp=kwWu4JpV=$DnmTE}-K_o-VzXS7qy9&j z5Gwv<)Bm4v_~(A*yp~SdrQ3gw9`>wRE&-2B;@%0p8@729@yw2Y*BYHuKpn~gkp8xo zu9L7)xUh6KJ(4u=Y@46D?T-*Z%od+Fk0|EWAhkioea0F{4&9NSfMhM=5<@paAfeW4 zP!UA?4_$@h(7+p1M+JvgQ~pCA!nkV@BsZv?Qm+BLi^&!y&I?x;cC!z>O^7olh{B?4 z&A4c6VbZ)Cq#SPkR7h$aIbI{vnXefU)UOCcyS{=P))iDp;$eV!!{j7NbH4)n3p$S2 z%Q!450D#4#3WrLwwGXi(Fispt9~3Yx2a7X1-9Oi5z(5>{AlL~~;+XtFsJs^wteLDl zLNA@=F@qg}`Qm|~_#9YiPX!G6j#NN!9yYJsRXYaYO@bQ*zk`xT^I103Gnom3_M=`pyXEId3n`H_eh` zOv@ZBMbL$y)8cUi3zX0*fXs@*zt1d;0^2Ukm{9tNm*J?ELP$Z=L_l8kIsmt-26^bc z5~9Wc1Zm5>S@Agdf(ls2#Pn@%Gb3aIz%QoG8A&`3-_F5Vx?J zd6>OzKx0rF2yHHbGzI2Qf~+8-ZXLF=srz_9yud%mfC}+DsvTkELgIhcv`VJlPc;V( z-N`LxZo^}MIspj@-6!yfeNEl5*0jzD^C9f#)i@g-JH`Nf0X$*_GQG~WGWvi(CkD5K zNV7bCLu{pU5RwaKh)0YYK63vtIib{JJmMDfk>7e~0AWZBYWii5l5mO}7)Rp~5`;vW zYs0An0zDYqJ_H7Kz>(eEl*BKF@W^oXdCJ8s#O{%JWSGFf7BEtR3{GheB&15qJ@x5I|s{YCbut!jdR>!N;kmX1;_DbWwmN z5>|ASlFmThHT>tV&3{G|bVm@6c;0NI(5}{OgkCsJ(1kFF@v+Oz!{ZU^1(^y=Iun}h z-3l!<;M-7R9-1BPISg*Y=SC0}Bi0vE!67sF+@5Hsd|hsMG|P}w?CyQ6G>*fgg|kXO zYA1*0VVt;+N{TQoWJ^uEXvGj&O2A|BV6aoDhvlK6KT9mvnu?!1scj(nl3tCoMN%O| zrfq?d%(@Zjr1@-c2sSk99xNNz$0k5%(?*s=sU#S5N6CzMH9+CsjCNcf~jalL0RRZdC-LwU^*8DhxrB3)k|I z)iv|wM&#hc2>tH9{pv_x+k0FKguVe3UDQeo^7**|6o1Ykb=4ov^90|7TW1-an}w!i z5XiQXoFTNG$z&%rha~ zYme$2Au-Q6l3H5PW}I(B-%-cza#58G-7To113wtU@WE-vn{v=JH*F1D||uZ(A@p|1lglK!dD_U2zp zcvdn6>nO9brG(<$Y?ckmSmMxQX?y_yh}iEVMpHl=j>8cZkOvIomqcx^qH`*@(CuWm zeL_tpL|<*|iY$;^1+W5{W7oot=kVp2+bQfe+bOtDos{ofL)q2v16lV7{h{IF39O~E z1UK}@LT|a^nYR2@2JMXaqO>-OGkiISHgiA4mJE^q-fo!xS|A1ZFtTNNHsg}fi^x+P zNO4+qHrrd}C8u<&IIygv*c`o+1qX3k)Gu7S#}uVXm1^nTn*V`R>^<&|G4254HfigSp1@dHjp?Pp9`zx>59saY6Br9kd_qr zGoZ1X_=QCVuK^m-ZJ?Iy5eL0rfRM#b7sH5b+&`4E3&HYPdnX3|63>*a=OSd|IN!Wa zS9m^_SP@8W(HF}5PFaO6J`dO$xzTkueDXnzXk+|Fmu4aMTTvW>CY(SAtCpoo<1P~j zmyTjX?|9OLv?Q^;z7$~J!Rq|W;lD=0j?NFQ@HqkigCTs% zlOSR&kb&yLn=Bcd*MUccz`*6{IlwVEl%5J4bSoPR{eFYmBRg7Oc#`?}_PdydscH8Z;c_d*GH4s7okpMobkm zGH6!EnAUzdL?^@lVx+z^WNVhas?CuVnSl?30yGpif-5Y^fQy-s?6EK?jic{3m03I( zp<4WZo8^mm4(m2Cp`7Qu;QiwRy5Np$fu*s1UZ8MI4FiM#7ZlNK(QhaVv`rDfzn8A3 z(h$`06GkE3vP-wqVgV>1H~_Ljuf!2NYm5Ny2HReG(xaC7ZVTzg2HSo&MTB<(_gTM9 zYq7pr^!sBdcNGdJc!)I%XZxxZmjGrbq~aNPBvfR3cxaYl0k?&Aa>2HsQ__2hv0QTSXhEwA=JFzS$rT2D0yQ&GWb_-uj##0J z;M0CsLJpT2d|T*75U%X)#`sS#8f|GY&%*WIhKs?*(`Z*E9c%e-O@IPTz;;bO2*wF^v{WwygU_6xLVGlTHqv%{ z|5o)eoouj~2Q6?BptI&LJ(8-kF(9DA1;yuprU1hL?kxBq3@otaHtH@24Ts|3JTDv^ z@)e&d7)21G6QoH65ke3sBN`y&plyA8Du6tJ-%HF81^EW0y%MIiU%@}8wQ%L$j22(_ zgw4b7mAb@aTkb7cT?E(){LOV;MWuNJAJRk8Rei{qsao+KP}W9UkEw^zJe?dQECbHc zZi^u=1Ma^fDuYI##f$E5>k>TmY?XR6QO85`TOr>4j??Ja-M4KF41|^=`eSVj?%sid zSM&Izb%Hd%wc}5WoB%4>fr0|W-O!)n&4OTFlJ31$%aG@m8#A9fUN3`PD57szTIE~O z<~=Zm&P(#lmhz5~`Zo{9U1n?({Q;KDR?!~R&w>=!cofjowiw+Ny(3S>BoUbL%)(`J zgnI1++p6(oB9udGE+922NT@Xo(O{EAUTf7_k_woGSG!RNw#txt`LO= zj=kt&Lul*|_OT_#}^ zu4|OcM81v2!__=y5Y;R?26r)hfhH%)Z@(^yLc3qGOU5!8;RGN46Fj>SxZ(>)6IwcN zTz^po2AKR>FsU%QOM!WVz&MXI zBQ|lf0PC4Yy6a?g=C_VAKg0>n#vjIA5aL)jB$C zoA5rxkuUfSp8WG%>${@n_fnu#&-5*>vnJJ1OLQ)KO-vpUOr>5QrjElhDRmnta1jkc zkQ}B2)7KzD?ePYq{gY86sZRB^Our?`vN98||xmgtv%rL6n9%AJqPv9fJ8QrKD#E2oVh?!!8>a=-LCb z0&vwft_$JPFCx_3+YK&N<=WV~Z56( znR3J4ta>)~CK(ZE;u0%T%9%=Y5=NA@HyAORsqmx|j2t7sHpF+ zI?!c&3&FcSnmvE!vPbC6Ib!McIR#UiL5|Xo5BelGdm`rNo6Y<0uOQlCE&}T) zmD2bC!0_kix)#B{7WVF`u;pJgUT97?#(!lWRWz>bo1imm6C?iVOnZBT{z-3Jy)%M1 zp@M({SHB_Tb+SpejR+{tP7tl8AtXo1ltKcIEE}iy*m1>M8=(S(*1PTZoxOllANiEI z={`ol#ZO{kp;(U)vaKhYW4CaJ3nDa@4G6Kpr}F&U3oStaZr3H+N}=aC8?xiM{#A%@ zN?iGP=gtEDJWFqbo+o1Q6J@%gsTpxX3mygYLLge*#&qCm>v=F(8$Y?dWMdx}5pL`?uKmu`eeWX|hi(bU#;Oajs}|rssE0e*q0Cc60R%U2EZw7 zZ(j+t^D&uW(41BuH`DT7k%TyoyI7;f+0ZTlpfG{OWnWKMdgHnN0YaQxy!=f#Rv>3o z7jF**J0#Lfb60#Mz}44#0Ui0|(iNm6o(t(DvnH=`nN7^iM^hTxUibpX;<0EyI9m)f ze$Ym^>+`_l!(m`c88WoTIdP3#hY)v6fPuyBiIDALd{~xMT(?X}-Ya%u^sJQVs4x@| zj2MO_K)^DDs6D>mO+KXdXThLlQQ%H+(#@d^MtogR0lI$;fHaZj8)rjk9Dj^BN`dn= zz63zXDLdXx%UQ=Ze*rQC#+us&AG;P5TG_q<40}A|5yXgUiTXAY*D>)>3E_MKAj*u= z;PzNX2llvczX>R)JexhPVAOB`gn$auk%O#;*fRb@ff_(}9O`TOKJBpzzz-?E{Y1JO zBtfVw4gmf29ek^So(zaYzJR_%sWgoQ0&=L8(iujpwx|ck#`a_wEEpN<+MET`Xvp&L zXaWd2LOy7lor3e)PG^&sO^Db==(&6A>t~-x*enwX;1Rds#h_j@RD1^rEL_S5RX=Hc zgims$2dX5Co+cqa*QVlv5-0=!ZK*y`1!l8B3I4KeSks215vC!K{N>0$AxV=nn4!$!@0TC+R(;Y1gZk7q>m;Ybzgi-5hXFsJ-}YmhE1A zY0qx?%|P-COxPHR{9X~^zc%(@(e_1!8gO#qQW3;5nO7e9s5D+#qAAJGEjBKEgX@-@9iolDtGT~Vn6bK3kH2qqOOs%CD4HE1Kxq8hLF@bqLf@h5>~YKC{s7;sJmz(E)5_H>jKf{Kkc^Ruj;{C7*i25k8>=KCoE5g|#>yH=-69&F zeSD$UB9_mu6Cun96Vu=!){Ef;>fbmi;G9A_02*f#3UZE5-+@9_m9zSQpY1!yKCozv zBldV=vmb9R=)mubIm`jL5k)NHX@`N65*xcscyJAjBnP zg*y^l?+F2j=Ch2K7 zSU6a4E_mSEaRvb$vC?>7-m}FuBay55c?X`?(}vt}W*IuJHG6;@=t20nW<)~w8$vE= zSBvRv!PF?#phNPW`&^h7`vPTP133^M)+>o&4>o>{>m;WOJ%)c8?pmyF0?~eYVbMC% z`@*q5DmL+nj7e~j&gxEkU`LrG-WN*5x37}47?kM`m?ek&3K>|q)Ej}ExcLarlwRdk zl#nBw0ook1jv;v?Pha;k^g$$oI><_HYmdCf10#^*l+19j9!Z{fw-QTsp`*t`o~#(PM`&qA z7L6p<)v}z_0e0_QhboRAzaG z;w+P&>D$&sij$m<;~-zTOoFl4ymX_jF_~~@{X0r z^_5J5*QUWxDEdKA)gim(}ffR3;-&mA!r#DYCGgZ0^U3#Pz0616=_L(j(LLgi_m zJaASD9oq4WUwxJM6OS5KLBd~R0wZXD;w4(P6;5l!;ZcXY*O-aVcMXkTbVPeI;S>R& z&l&;(mG(?#ORSu;!i;(BZ3^a2)E?Fw%;r@Ljs^c|1~uRUm5EGdE@Wwi*OPMUB8p