Conversation
Rename pyMathTools/pyMathToolsPlotHelpers packages to snake_case (math_tools/math_plot_helpers) and re-parent QuaternionABC onto the new foundation_abc layout per the template conformance spec.
Add the math_tools exception hierarchy (MathToolsError and three domain subtypes) and an AST-based package layering test enforcing math_tools never imports math_plot_helpers/matplotlib, and math_tools.otg (once it exists) never imports numpy.
Add .claude/CLAUDE.md and a real README, correct pyproject dependency names (numpy/scipy were missing), and fix dead MathTypes import paths in the two spherical plotting examples so they import-run cleanly.
Add PrecisionTimeInterval: an immutable, hashable, attosecond-exact subclass of PrecisionTimeIntervalABC backed by a single signed total- attosecond int, with full arithmetic/ordering/wire-dict surface per precisionTimeMath.md.
Add PrecisionTimestamp: an immutable, hashable, epoch-referenced timestamp composing a PrecisionTimeInterval offset plus optional timescale/reference_frame/uncertainty metadata, with Swift-parity can_compare/compare_validated semantics per precisionTimeMath.md.
Add Position: a numpy-backed 3D point type subclassing PositionABC with cylindrical/spherical/spherical-iso constructors and inverse accessors, full vector arithmetic, and the shared API idioms (normalize/normalized, isclose, __array__, unhashable).
Add dot, rotation_matrix_elements (delegating to to_rotation_matrix), rotate_position, __array__, and snake_case aliases for the two existing camelCase methods to Quaternion. Purely additive: legacy tests/test_quaternion.py is untouched.
Add SpatialPose: an SE(3) pose composing Position + Quaternion, subclassing SpatialTransformABC. Constructors (components, homogeneous matrix, Denavit-Hartenberg, identity), composition/transform via *, translated, inverse, relative_pose, interpolate (lerp + slerp), position/angular distance (double-cover safe), and normalize/normalized per spatialMath.md.
Add UnivariatePolynomial: an immutable, numpy-backed general polynomial with ascending-degree float64 coefficient storage, derivative/integrate, companion-matrix real_roots (raises PolynomialSolveError on the zero polynomial), and value evaluation via numpy's polyval.
Faithful port of the Swift OTG analytic root kernel (Roots.swift/Utils.swift): solve_cubic, solve_resolvent, solve_quartic_monic, evaluate_polynomial, polynomial_derivative(_monic), shrink_interval, integrate_jerk. Corrects a latent 1/a^4 vs 1/a^3 scaling bug in solve_cubic's halfq term (inert in the Swift source's monic-only call sites, not inert for general coefficients per Compliance 2b), documented in polynomials.md. Re-homes POLYNOMIAL_ZERO_THRESHOLD from polynomial.py into roots.py. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Adds the scalar waveform container: attosecond-exact PrecisionTime axis, 22 signal generators, None-on-empty statistics (population variance pinned), attosecond-exact slicing, and the list-vocabulary mutation API. No DSP mixins yet (composed in a later chunk); no arithmetic operators (chunk 12). Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Adds the elementwise operator surface: arithmetic (+ - * / // %, forward/ reflected/in-place), bitwise (integer-dtype only), unary, and comparison producers (elements_equal/less_than/greater_than, isclose_elementwise, isclose). Waveform-Waveform ops require equal dt+length (WaveformCompatibilityError otherwise); results carry dt/t0 from the left operand. Fixes a bug where Waveform1D's __array__ let numpy's ufunc machinery bypass Python's operator protocol for numpy-scalar/array operands (e.g. np.float64(2.0) + w silently returned a bare ndarray, discarding dt/t0) by setting __array_ufunc__ = None, with a regression test. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
The 22 generators themselves landed early in chunk 11 (umbrella spec groups them under Constructors). This chunk adds the dedicated compliance-4 test file, and fixes a real gap: sine's frequency parameter had no default, so the spec's own pinned example (Waveform1D.sine(n=1000)) raised TypeError. frequency now defaults to 1.0 Hz on sine only, matching the literal acceptance criterion. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Adds the shared substrate for the upcoming DSP mixin chunks: all 12 support enums + 15 frozen descriptor dataclasses (eq=False on the 5 ndarray-bearing ones), the WaveformProtocol structural contract mixins will depend on, and Waveform1D._with_values (the replace-values factory). WaveformTriggerType (undefined in the Python spec text) is flattened from Swift's associated-value enum into a 4-member discriminator, with the payload moved onto WaveformTrigger's optional fields. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Adds the position time-series aggregate container: (n,3) float64 storage, on-demand Position materialization, from_components with length/dt validation, vectorized normalize/component_waveforms (never loops over materialized elements), and the full mutation/extend/concatenate verb set. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Adds the quaternion time-series aggregate container, mirroring WaveformPosition's structure: (n,4) float64 storage in w-first order (pinned repo-wide, not Swift's x-first), on-demand Quaternion materialization, w-first component_waveforms/from_components, vectorized are_all_unit/normalize, and the full mutation/extend/concatenate verb set. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Adds the 6-DOF pose time-series container: parallel (n,3) position + (n,4) w-first quaternion arrays, from_poses/from_waveforms constructors, nested component_waveforms reusing chunks 15/16's per-axis NamedTuples, and the is_valid/sample_count=min(...) semantics for the (deliberately constructible) unequal-array-length edge case. Completes Track C (chunks 11-17). Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Add CalcMixin (integrate/derivative). Fixed a runtime bug in the planned mixin pattern: nominally subclassing WaveformProtocol shadows Waveform1D's real properties under C3 linearization, so mixins now self-type instead (mypy's documented "mixin classes" idiom). Spec and protocol docs updated accordingly for chunks 19-29.
Add CorrelationMixin (auto_correlation, cross_correlation, find_max_correlation), backed by scipy.signal.correlate. Lag sign kept as scipy's native convention (spec-permitted divergence from Swift), documented in the module docstring.
Add EnvelopeMixin: amplitude_envelope (Hilbert), upper_lower_envelopes (peak/valley interpolation), instantaneous_amplitude (HILBERT/RMS/PEAK).
Add SpectralMixin: fft, power_spectral_density (Welch), spectrogram, mel_spectrogram, spectral_features (centroid/spread/rolloff/flatness).
Add FilteringMixin: low/high/band-pass, dispatcher (incl. band-stop), moving-average, exponential, Savitzky-Golay, Whittaker-Henderson filters, and frequency_response.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Corrects otg.md's numeric-truth-table atol from 1e-8 to 1e-6 (semver 0.0.4 -> 0.0.5): the Swift source literals are recorded to only 6 decimal places, so 1e-8 was unsatisfiable by any correct port. Also corrects "32-case" to "31-case" (the Swift array has 31 entries). Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Final chunk of the math-tools-port plan: curates the flat `import math_tools as mt` root surface (19 re-exports, no logic), pinned by test_public_surface.py. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Fix j·tf⁴→j·tf³ transcription defect in UDUD-T0246 discriminant (tf_p3→tf_p2) at position_third_order_step2.py:1731,1752,1763; add regression test. Scaffold Track F corrective plan overview. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Trivial branch builds fixed-length 8-element a/v/p arrays and sets pf to current position, per otg.md fidelity requirement 1; add regression tests. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Replace 8 bare math.sqrt sites in position_third_order_step1 with the IEEE-754 _ieee754_sqrt (NaN on negative radicand) so out-of-domain candidates are rejected by guards instead of raising; add regression test. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Position.is_unit uses rtol=0.0 so only spec atol=1e-12 applies; days_since_epoch/seconds_of_day derive from signed total_attoseconds (floor div/mod) for correct pre-epoch behavior. Add regression tests. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
solve_cubic uses a scale-relative leading-coefficient degeneracy threshold; shrink_interval guards f==0 before the Newton step to avoid a ZeroDivisionError trap. Update polynomials.md (semver 0.0.4) and add regression tests. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Route __getitem__/get/pop through a shared _resolve_index helper that bounds against sample_count (not either raw array's length), per waveformCore.md compliance 9; add regression tests. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Add __array__ and isclose to the three aggregate waveform containers, restore WaveformSpatialPose.from_components, and fix Waveform1D.__array__ copy=False contract (C-7). Update waveformCore.md (semver 0.0.3) and add tests. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
chirp/heaviside/sigmoid derive total duration from an endpoint-inclusive (n-1)*dt span via _span_seconds (0.0 at n<=1), fixing off-by-one phase conventions. Update waveformCore.md (semver 0.0.4) and add tests. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Give math_tools.spherical and math_plot_helpers curated __all__ re-exports; update README/examples to import at subpackage level; pin the surface with tests. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Reconcile window symmetric/periodic conventions: add periodic param to windowing helpers (default symmetric), share DEFAULT_KAISER_BETA across windowing/spectral. Document per-family rule in waveformDsp.md (semver 0.0.8) and add tests. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Close 11 Track B/C test-coverage gaps (hashability, double-cover isclose, operator surface, generator analytics, bulk smoke, indexing); set Quaternion.__hash__ = None explicitly (the one permitted src edit). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Close 23 DSP test-coverage gaps with value-level known-answer/analytic assertions across all DSP mixin families; tests only, no src changes. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Close 8 OTG oracle-coverage gaps: add a reproducible 450-case multi-DOF corpus that exercises Step2 (0->900 invocations), plus continuity, retarget, invariant, and the missing testBugFix_NegativeTimeInterval_Case3 cases. Correct chunk 42's stale 'all 4 testBugFix' claim. Tests/data only. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Reconcile docs against post-fix repo: fix dead CI/workflow refs, stale OTG tolerance (1e-6) and case-count claims, install/keyword drift; wire examples/ into the gate (lint+typecheck clean); delete hints.py dead __main__; add governance tests. Bump templateConformance (0.0.3) and overview (0.1.1). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
pyproject declares dependency NAMES ONLY per the BKM; requirements.txt carries the pins and git pointers. testInEnvInstallFromSetup ran `pip install ".[dev]"` alone, so pip resolved the bare name pyFoundationTools against PyPI and failed with "No matching distribution found". Every other install path here already installs -r requirements.txt first. make testInEnv now green: 1550 passed. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Back-port of the py-cookiecut rename. testInEnv green end-to-end after the rename (1550 passed in the clean room). Co-Authored-By: Claude Opus 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
1 participant
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.
No description provided.