Repository navigation
Restore test coverage the fixture cleanup dropped - #99
Merged
Merged
Conversation
The wire behavior check replaced the recorded payload snapshot but only looked at selected fields, so a key added to a body, its metadata or its output (a leaked client_* key, say) passed. Assert each body's exact top-level keys and compare metadata and output whole, and seed those four faults in the release contract test. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: Manush Murali <manushm@nvidia.com>
The PostgreSQL branch expected 1.0.0's bare OverflowError at from_source, but the client reads keys past int64 as exact text there and succeeds, so the test failed whenever a Postgres DSN was set. CI never set one for contract/tests, which hid it. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: Manush Murali <manushm@nvidia.com>
The maintained-fixture README still described label-expectations.json and eight sdm-shared-fixture/1 files, none of which remain, and rust/testdata/fixture_tables.rs no longer has an includer. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: Manush Murali <manushm@nvidia.com>
Only current_user was checked once the expression vectors went, so dropping session_user, current_role or localtimestamp from the session words would let an untrusted view read session identity or the clock without failing a test. List all thirteen, in both dialects. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: Manush Murali <manushm@nvidia.com>
contract/tests runs only in contract-python (pandas3, duckdb), which had no Postgres DSN, so every @postgres case in it skipped. With the fixture cleanup those cases became the only check of their PostgreSQL expectations, and one of them was wrong without CI noticing. Give the job the shared service and require it for that step. Locally, against PostgreSQL 16: 2,483 passed; the one failure is the service-version check, which wants CI's 17. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: Manush Murali <manushm@nvidia.com>
The keyed-read check looked only for "posexplode(from_json(?, 'array<", so rendering int64 keys as array<int> (read as 32-bit) passed every test once the goldens went; the Python Databricks fakes replace that CTE before running the read. Assert the whole CTE for bigint, double and string keys and for a windowed neighbor's struct. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: Manush Murali <manushm@nvidia.com>
With view-expr-vectors.json gone, only five of the expression refusal reasons kept their user-facing wording under test; the session-value, SQL-keyword, parenthesis, qualified, quoted-name, IDENTIFIER() and Databricks-quote reasons could be reworded silently. Pin one example of each, with the wording 1.0.0 gave it. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: Manush Murali <manushm@nvidia.com>
The fragment list dropped the typo diagnostic's "row 1, column 12" position and the rest of each sentence, so either could change for an unusual column name without failing. Compare every scenario's message with its kind's full wording, the query text substituted, as the ordinary-name twins gave it. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: Manush Murali <manushm@nvidia.com>
The rewrite kept a count and a few substrings of each labelled warning, so the context-before-data, anchor-after-data, missing-index and guessed-time-column sentences could be reworded without a failure; no other test holds them. Assert each one whole, as the 1.0.0 recordings did. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: Manush Murali <manushm@nvidia.com>
The lookup and neighbor checks only looked for the JOIN target, so a LEFT JOIN, which returns an all-null row for a seed key the table lacks, passed every test once the recorded statements went. Require "FROM TMP JOIN" in every keyed read and compare the plain lookup's whole statement. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: Manush Murali <manushm@nvidia.com>
The rewrite checked only orders.d's dtype and stype, so the other columns' types, the keys, the time column and the link could change on any database without failing. Assert the whole schema on DuckDB, SQLite and PostgreSQL, as the recorded expectation did. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: Manush Murali <manushm@nvidia.com>
|
Important Review skippedAuto reviews are disabled on base/target branches other than the default branch. Please check the settings in the CodeRabbit UI or the
|
Manushpm8
marked this pull request as ready for review
November 10, 2026 22:45
adi-kumo
approved these changes
Nov 11, 2026
Signed-off-by: Adi Wadaskar <awadaskar@nvidia.com>
Manushpm8
added a commit
that referenced
this pull request
Nov 11, 2026
The rest of the coverage #96 dropped, which #99 merged without. Each of the six open items from the #96 review was decided with Codex, and each new assertion was proved by seeding the fault it guards against, watching it fail, and reverting. ### Decisions | Item | Decision | What this PR does | | --- | --- | --- | | Executed SQL (`l4-goldens` removed) | Restore smaller | Fixed read budgets per journey: DuckDB (10 describe + 8 materialize, 18 logged) and a new SQLite journey (24 fetches, 6 refreshes). One whole lookup and one whole neighbor statement with parameters for DuckDB, SQLite and PostgreSQL (rendered, no server). Cast, window-bound, decimal and key projections in the four SQL family tests. No observation-wide pins. | | Batching, query options, ranking | Restore smaller | `contract/support/shop_requests.py` derives each call's context, anchors, labels (window, forecast step, candidate membership), scored entities and related rows from the case's own inputs; query-options and ranking cases assert them. A seeded three-call fixture is compared as three whole requests. | | Java error messages | Restore | All 23 rows of `BindingTest` assert the exact message, with real fields for the three parameterized sentences. | | PQL | Restore smaller | 11 exact `(code, span)` assertions in `pql_corpus.rs`; fuzz proportions stay unpinned, plus an explicit packed-date refusal. | | Write guard | Accept | Protects only our live harness and fails loudly there; no change. | | Time-zone guard | Restore smaller | Also scans the 378 Python-held DuckDB setups, `PROBES` included; zoned setups need a global pin. | Also: PostgreSQL keys above `2**63 - 1` are now asserted as exact strings on the wire, and CHANGELOG says so, since 1.0.0 raised `OverflowError` there. ### Seeded faults that each new check catches Re-reading unchanged tables; an uncached time scan; a dropped timestamp cast, window bound, per-seed limit or wide-key `TRY_CAST`; constant or swapped labels; a dropped related table; a short batch context; a ranking part reusing another's labels; Java printing values without quotes; collapsed PQL spans; a session-only time-zone pin; float64 wide keys. ### Verified locally Java `BindingTest` 84/0 on JDK 21; `cargo test` for the touched sdm-core and sdm-testkit tests; the three contract files on DuckDB, SQLite and Postgres (620 passed); `test_exact_numeric_keys.py` on Postgres; `test_contract_cases.py`; fmt, clippy `-D warnings`, ruff and repository hygiene clean. 🤖 Generated with [Claude Code](https://claude.com/claude-code) --------- Signed-off-by: Manush Murali <manushm@nvidia.com> Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Follow-up to #96. Restores coverage the fixture cleanup dropped where a realistic regression would now pass CI. Each fix was checked by making the test fail first (seeded fault or broken payload), then pass.
Draft while the remaining open items are decided; description will list each decision.
🤖 Generated with Claude Code