pysofra 0.1.0a1__tar.gz → 0.1.0a3__tar.gz

pysofra-0.1.0a3/.gitignore

@@ -0,0 +1,45 @@
+__pycache__/
+*.py[cod]
+*.egg-info/
+dist/
+build/
+.venv/
+venv/
+.env
+
+.pytest_cache/
+.mypy_cache/
+.ruff_cache/
+.coverage
+htmlcov/
+
+.DS_Store
+.idea/
+.vscode/
+.claude/
+
+site/
+docs/_build/
+paper/
+
+# Tutorial / example export artefacts — produced on demand by
+# scripts/render_tutorial.py, not part of the source tree.
+*.docx
+*.pptx
+*.xlsx
+examples/tutorial_*.docx
+examples/tutorial_*.pptx
+examples/tutorial_*.xlsx
+examples/tutorial_*.png
+!tests/fixtures/**/*.docx
+!tests/fixtures/**/*.pptx
+!tests/fixtures/**/*.xlsx
+
+# Hypothesis example database (auto-managed)
+.hypothesis/
+
+# uv writes a lockfile when ``uv add`` / ``uv pip install`` is used
+# locally; the project pins its runtime versions explicitly in
+# pyproject.toml, so the uv lockfile is a local convenience that
+# should not ship.
+uv.lock

{pysofra-0.1.0a1 → pysofra-0.1.0a3}/CHANGELOG.md

@@ -5,6 +5,21 @@ All notable changes to PySofra will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [0.1.0a3] — 2026-05-24
+
+### Changed
+- Documentation polish: scrubbed internal-process references from test
+  docstrings and inline comments. No public API or behavioural changes.
+
+## [0.1.0a2] — 2026-05-23
+
+### Fixed
+- Theme styling now survives notebook viewers that strip `<style>` blocks
+  (e.g. GitHub's notebook viewer). Critical theme properties (font, border,
+  padding) are emitted as inline `style` attributes on each table element, so
+  `jama` vs `nejm` vs `clinical` vs `minimal` stay visibly distinct everywhere.
+- README image and link URLs are now absolute so they render on PyPI.
+
 ## [0.1.0a1] — 2026-05-20
 
 ### Added

{pysofra-0.1.0a1 → pysofra-0.1.0a3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pysofra
-Version: 0.1.0a1
+Version: 0.1.0a3
 Summary: Statistical reporting and table preparation framework for Python — the missing reporting layer.
 Project-URL: Homepage, https://github.com/jturner-uofl/pysofra
 Project-URL: Documentation, https://github.com/jturner-uofl/pysofra
@@ -72,7 +72,7 @@ Description-Content-Type: text/markdown
 
 [![Coverage](https://img.shields.io/badge/coverage-100%25-brightgreen.svg)](https://github.com/jturner-uofl/pysofra)
 [![Python](https://img.shields.io/badge/python-3.11%20%7C%203.12%20%7C%203.13-blue.svg)](https://www.python.org/downloads/)
-[![License: GPL-3.0+](https://img.shields.io/badge/license-GPL--3.0--or--later-blue.svg)](LICENSE)
+[![License: GPL-3.0+](https://img.shields.io/badge/license-GPL--3.0--or--later-blue.svg)](https://github.com/jturner-uofl/pysofra/blob/main/LICENSE)
 [![Style: ruff](https://img.shields.io/badge/style-ruff-purple.svg)](https://github.com/astral-sh/ruff)
 [![Types: mypy strict](https://img.shields.io/badge/types-mypy%20strict-blue.svg)](http://mypy-lang.org/)
 [![Tests: 886](https://img.shields.io/badge/tests-886%20passing-brightgreen.svg)](#status)
@@ -111,14 +111,14 @@ Description-Content-Type: text/markdown
 - **One immutable object, seven output formats** — build a `SofraTable` once, render to HTML / Markdown / LaTeX / DOCX / PPTX / XLSX / PNG, all byte-deterministic across processes
 - **Auto-dispatched statistical tests** — Welch, Wilcoxon, ANOVA, Kruskal–Wallis, Fisher, χ², Rao–Scott, design-adjusted *t* — picked by variable kind, overridable per-row
 - **Inline forest plots and KM curves** — embed matplotlib figures directly into the table; the same `SofraTable` renders them across every backend
-- **Statistically correct** — every numeric output validated against `scipy` / `statsmodels` / `lifelines` at machine precision (and cross-checked against R's `gtsummary` for the JSS paper)
+- **Statistically correct** — every numeric output validated against `scipy` / `statsmodels` / `lifelines` at machine precision, with cross-checks against R's `gtsummary`
 - **Method-chainable and immutable** — every modifier returns a new table; no in-place mutation, no global state, fully reproducible
 
 <div align="center">
 
-**[Showcase notebook](examples/pysofra_showcase.ipynb)** · [rendered HTML](examples/pysofra_showcase.html) — *47 cells, every section a side-by-side numeric proof. Start here if you have 60 seconds.*
+**[Showcase notebook](https://github.com/jturner-uofl/pysofra/blob/main/examples/pysofra_showcase.ipynb)** — *47 cells, every section a side-by-side numeric proof. Start here if you have 60 seconds.*
 
-**[End-to-end tutorial](examples/pysofra_tutorial.ipynb)** · [rendered HTML](examples/pysofra_tutorial.html) — *126 cells walking every public feature on a synthetic two-arm trial.*
+**[End-to-end tutorial](https://github.com/jturner-uofl/pysofra/blob/main/examples/pysofra_tutorial.ipynb)** — *126 cells walking every public feature on a synthetic two-arm trial.*
 
 </div>
 
@@ -171,10 +171,10 @@ fit = sm.Logit(df["event"], X).fit(disp=False)
 )
 ```
 
-The full worked example from the JSS paper — baseline table by
-treatment arm, regression table with forest plot, and Kaplan-Meier
-survival summary — is in
-[`paper/replication/example_trial.py`](paper/replication/example_trial.py).
+The end-to-end worked example — baseline table by treatment arm,
+regression table with forest plot, and Kaplan-Meier survival summary —
+is in the
+[showcase notebook](https://github.com/jturner-uofl/pysofra/blob/main/examples/pysofra_showcase.ipynb).
 
 ---
 
@@ -255,47 +255,40 @@ pip install "pysofra[dev]"        # testing + linting (pytest, ruff, mypy, hypot
 
 ## Status
 
-PySofra is in **alpha** (`0.1.0a1`). The public API surface is pinned
+PySofra is in **alpha** (`0.1.0a3`). The public API surface is pinned
 by an explicit
-[API-stability test](tests/test_joss_api_stability.py) so that any
-unintended rename, removal, or signature change surfaces as a failed
-test. Quality bar at this release:
+[API-stability test](https://github.com/jturner-uofl/pysofra/blob/main/tests/test_joss_api_stability.py)
+so that any unintended rename, removal, or signature change surfaces as
+a failed test. Quality bar at this release:
 
 * **More than 800 tests passing**, **100% line coverage**, mypy strict, ruff clean.
 * Every numeric output is validated against `scipy`, `lifelines`,
   `statsmodels`, or a hand-computed textbook formula
-  ([test_joss_statistical_correctness.py](tests/test_joss_statistical_correctness.py)).
+  ([test_joss_statistical_correctness.py](https://github.com/jturner-uofl/pysofra/blob/main/tests/test_joss_statistical_correctness.py)).
 * Universal invariants enforced via Hypothesis on 720 randomized
   examples per CI run
-  ([test_joss_property_invariants.py](tests/test_joss_property_invariants.py)).
+  ([test_joss_property_invariants.py](https://github.com/jturner-uofl/pysofra/blob/main/tests/test_joss_property_invariants.py)).
 * Renderer output is byte-deterministic — identical input always
   produces identical HTML/Markdown/LaTeX, required for reproducible
   publication artifacts
-  ([test_joss_renderer_consistency.py](tests/test_joss_renderer_consistency.py)).
+  ([test_joss_renderer_consistency.py](https://github.com/jturner-uofl/pysofra/blob/main/tests/test_joss_renderer_consistency.py)).
 
 Bug reports and use-case feedback are very welcome.
 
-A **Journal of Statistical Software** paper ([`paper/paper.tex`](paper/paper.tex),
-bibliography in [`paper/paper.bib`](paper/paper.bib)) is in
-preparation. The [`paper/replication/`](paper/replication/) directory
-regenerates every numeric output and figure shown in the paper and
-includes an R cross-check script (`example_trial.R`) using
-`gtsummary` for digit-level verification. The
-[`paper/figures/`](paper/figures/) directory holds the rendered PDF
-and PNG embedded in the manuscript via `\includegraphics{}`.
-
 ## Contributing
 
 Bug reports, feature requests, and pull requests are all very welcome.
-Please read [`CONTRIBUTING.md`](CONTRIBUTING.md) for the workflow, the
-quality gates, and the
-[Code of Conduct](CODE_OF_CONDUCT.md).
+Please read
+[`CONTRIBUTING.md`](https://github.com/jturner-uofl/pysofra/blob/main/CONTRIBUTING.md)
+for the workflow, the quality gates, and the
+[Code of Conduct](https://github.com/jturner-uofl/pysofra/blob/main/CODE_OF_CONDUCT.md).
 
 ## License
 
-GPL-3.0-or-later. See [`LICENSE`](LICENSE).
+GPL-3.0-or-later. See
+[`LICENSE`](https://github.com/jturner-uofl/pysofra/blob/main/LICENSE).
 
 ## Citation
 
 If you use PySofra in academic work, please cite the project — see
-[`CITATION.cff`](CITATION.cff).
+[`CITATION.cff`](https://github.com/jturner-uofl/pysofra/blob/main/CITATION.cff).

{pysofra-0.1.0a1 → pysofra-0.1.0a3}/README.md

@@ -6,7 +6,7 @@
 
 [![Coverage](https://img.shields.io/badge/coverage-100%25-brightgreen.svg)](https://github.com/jturner-uofl/pysofra)
 [![Python](https://img.shields.io/badge/python-3.11%20%7C%203.12%20%7C%203.13-blue.svg)](https://www.python.org/downloads/)
-[![License: GPL-3.0+](https://img.shields.io/badge/license-GPL--3.0--or--later-blue.svg)](LICENSE)
+[![License: GPL-3.0+](https://img.shields.io/badge/license-GPL--3.0--or--later-blue.svg)](https://github.com/jturner-uofl/pysofra/blob/main/LICENSE)
 [![Style: ruff](https://img.shields.io/badge/style-ruff-purple.svg)](https://github.com/astral-sh/ruff)
 [![Types: mypy strict](https://img.shields.io/badge/types-mypy%20strict-blue.svg)](http://mypy-lang.org/)
 [![Tests: 886](https://img.shields.io/badge/tests-886%20passing-brightgreen.svg)](#status)
@@ -45,14 +45,14 @@
 - **One immutable object, seven output formats** — build a `SofraTable` once, render to HTML / Markdown / LaTeX / DOCX / PPTX / XLSX / PNG, all byte-deterministic across processes
 - **Auto-dispatched statistical tests** — Welch, Wilcoxon, ANOVA, Kruskal–Wallis, Fisher, χ², Rao–Scott, design-adjusted *t* — picked by variable kind, overridable per-row
 - **Inline forest plots and KM curves** — embed matplotlib figures directly into the table; the same `SofraTable` renders them across every backend
-- **Statistically correct** — every numeric output validated against `scipy` / `statsmodels` / `lifelines` at machine precision (and cross-checked against R's `gtsummary` for the JSS paper)
+- **Statistically correct** — every numeric output validated against `scipy` / `statsmodels` / `lifelines` at machine precision, with cross-checks against R's `gtsummary`
 - **Method-chainable and immutable** — every modifier returns a new table; no in-place mutation, no global state, fully reproducible
 
 <div align="center">
 
-**[Showcase notebook](examples/pysofra_showcase.ipynb)** · [rendered HTML](examples/pysofra_showcase.html) — *47 cells, every section a side-by-side numeric proof. Start here if you have 60 seconds.*
+**[Showcase notebook](https://github.com/jturner-uofl/pysofra/blob/main/examples/pysofra_showcase.ipynb)** — *47 cells, every section a side-by-side numeric proof. Start here if you have 60 seconds.*
 
-**[End-to-end tutorial](examples/pysofra_tutorial.ipynb)** · [rendered HTML](examples/pysofra_tutorial.html) — *126 cells walking every public feature on a synthetic two-arm trial.*
+**[End-to-end tutorial](https://github.com/jturner-uofl/pysofra/blob/main/examples/pysofra_tutorial.ipynb)** — *126 cells walking every public feature on a synthetic two-arm trial.*
 
 </div>
 
@@ -105,10 +105,10 @@ fit = sm.Logit(df["event"], X).fit(disp=False)
 )
 ```
 
-The full worked example from the JSS paper — baseline table by
-treatment arm, regression table with forest plot, and Kaplan-Meier
-survival summary — is in
-[`paper/replication/example_trial.py`](paper/replication/example_trial.py).
+The end-to-end worked example — baseline table by treatment arm,
+regression table with forest plot, and Kaplan-Meier survival summary —
+is in the
+[showcase notebook](https://github.com/jturner-uofl/pysofra/blob/main/examples/pysofra_showcase.ipynb).
 
 ---
 
@@ -189,47 +189,40 @@ pip install "pysofra[dev]"        # testing + linting (pytest, ruff, mypy, hypot
 
 ## Status
 
-PySofra is in **alpha** (`0.1.0a1`). The public API surface is pinned
+PySofra is in **alpha** (`0.1.0a3`). The public API surface is pinned
 by an explicit
-[API-stability test](tests/test_joss_api_stability.py) so that any
-unintended rename, removal, or signature change surfaces as a failed
-test. Quality bar at this release:
+[API-stability test](https://github.com/jturner-uofl/pysofra/blob/main/tests/test_joss_api_stability.py)
+so that any unintended rename, removal, or signature change surfaces as
+a failed test. Quality bar at this release:
 
 * **More than 800 tests passing**, **100% line coverage**, mypy strict, ruff clean.
 * Every numeric output is validated against `scipy`, `lifelines`,
   `statsmodels`, or a hand-computed textbook formula
-  ([test_joss_statistical_correctness.py](tests/test_joss_statistical_correctness.py)).
+  ([test_joss_statistical_correctness.py](https://github.com/jturner-uofl/pysofra/blob/main/tests/test_joss_statistical_correctness.py)).
 * Universal invariants enforced via Hypothesis on 720 randomized
   examples per CI run
-  ([test_joss_property_invariants.py](tests/test_joss_property_invariants.py)).
+  ([test_joss_property_invariants.py](https://github.com/jturner-uofl/pysofra/blob/main/tests/test_joss_property_invariants.py)).
 * Renderer output is byte-deterministic — identical input always
   produces identical HTML/Markdown/LaTeX, required for reproducible
   publication artifacts
-  ([test_joss_renderer_consistency.py](tests/test_joss_renderer_consistency.py)).
+  ([test_joss_renderer_consistency.py](https://github.com/jturner-uofl/pysofra/blob/main/tests/test_joss_renderer_consistency.py)).
 
 Bug reports and use-case feedback are very welcome.
 
-A **Journal of Statistical Software** paper ([`paper/paper.tex`](paper/paper.tex),
-bibliography in [`paper/paper.bib`](paper/paper.bib)) is in
-preparation. The [`paper/replication/`](paper/replication/) directory
-regenerates every numeric output and figure shown in the paper and
-includes an R cross-check script (`example_trial.R`) using
-`gtsummary` for digit-level verification. The
-[`paper/figures/`](paper/figures/) directory holds the rendered PDF
-and PNG embedded in the manuscript via `\includegraphics{}`.
-
 ## Contributing
 
 Bug reports, feature requests, and pull requests are all very welcome.
-Please read [`CONTRIBUTING.md`](CONTRIBUTING.md) for the workflow, the
-quality gates, and the
-[Code of Conduct](CODE_OF_CONDUCT.md).
+Please read
+[`CONTRIBUTING.md`](https://github.com/jturner-uofl/pysofra/blob/main/CONTRIBUTING.md)
+for the workflow, the quality gates, and the
+[Code of Conduct](https://github.com/jturner-uofl/pysofra/blob/main/CODE_OF_CONDUCT.md).
 
 ## License
 
-GPL-3.0-or-later. See [`LICENSE`](LICENSE).
+GPL-3.0-or-later. See
+[`LICENSE`](https://github.com/jturner-uofl/pysofra/blob/main/LICENSE).
 
 ## Citation
 
 If you use PySofra in academic work, please cite the project — see
-[`CITATION.cff`](CITATION.cff).
+[`CITATION.cff`](https://github.com/jturner-uofl/pysofra/blob/main/CITATION.cff).

{pysofra-0.1.0a1 → pysofra-0.1.0a3}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "pysofra"
-version = "0.1.0a1"
+version = "0.1.0a3"
 description = "Statistical reporting and table preparation framework for Python — the missing reporting layer."
 readme = "README.md"
 license = { text = "GPL-3.0-or-later" }
@@ -101,9 +101,8 @@ packages = ["src/pysofra"]
 
 [tool.hatch.build.targets.sdist]
 # The sdist that goes on PyPI ships the library + tests + license +
-# changelog. The JSS paper bundle (paper/) is intentionally NOT
-# included: it lives in git for source consumers and reviewers
-# download it from the submission tarball, not from PyPI.
+# changelog. Docs, examples, build scripts, and CI config are excluded
+# — users get the importable package, not the development surround.
 include = [
     "src/pysofra",
     "tests",
@@ -114,13 +113,13 @@ include = [
     "pyproject.toml",
 ]
 exclude = [
-    "paper",
     "docs",
     "examples",
     "scripts",
     "site",
     ".github",
     ".claude",
+    "paper",
     "uv.lock",
 ]
 

{pysofra-0.1.0a1 → pysofra-0.1.0a3}/src/pysofra/__init__.py

@@ -50,7 +50,7 @@ from .summary.tbl_summary import tbl_summary
 from .summary.tests import available_tests
 from .themes.registry import available_themes, register_theme
 
-__version__ = "0.1.0a1"
+__version__ = "0.1.0a3"
 
 __all__ = [
     "CellPart",

{pysofra-0.1.0a1 → pysofra-0.1.0a3}/src/pysofra/summary/stats.py

@@ -45,12 +45,12 @@ def continuous_stats(series: pd.Series) -> ContinuousStats:
     # subtract`` (and similar). Under ``filterwarnings = error`` — which
     # this project's own pyproject.toml sets and which is a common
     # user-side ``-W error`` posture — those warnings escalate to
-    # exceptions and crash ``tbl_one`` on perfectly legal data. The R6
-    # audit fixed the ``int(np.inf) → OverflowError`` path in
-    # ``infer_kind`` but didn't reach this downstream stats site.
-    # Wrap arithmetic in ``np.errstate`` + ``catch_warnings`` so the
-    # stats compute cleanly to ``nan`` / ``inf`` (which the formatters
-    # then render as em-dash).
+    # exceptions and crash ``tbl_one`` on perfectly legal data. An
+    # earlier fix in ``infer_kind`` handled the ``int(np.inf) →
+    # OverflowError`` path but did not reach this downstream stats
+    # site. Wrap arithmetic in ``np.errstate`` + ``catch_warnings`` so
+    # the stats compute cleanly to ``nan`` / ``inf`` (which the
+    # formatters then render as em-dash).
     with np.errstate(invalid="ignore", over="ignore"), warnings.catch_warnings():
         warnings.simplefilter("ignore", RuntimeWarning)
         mean = float(np.mean(arr))

{pysofra-0.1.0a1 → pysofra-0.1.0a3}/tests/test_extract_edges.py

@@ -236,7 +236,7 @@ class TestExtrasFinal:
     def test_add_global_p_metadata_missing_f_test(self):
         # A SofraTable whose metadata['model'] doesn't have .f_test now
         # raises NotImplementedError rather than silently inserting an
-        # em-dash column (which a JOSS reviewer flagged as misleading).
+        # em-dash column (which would be misleading).
         from pysofra.core.schema import Cell, HeaderCell, HeaderRow, Row
         from pysofra.core.table import SofraTable
 

{pysofra-0.1.0a1 → pysofra-0.1.0a3}/tests/test_extras_edges.py

@@ -353,9 +353,9 @@ class TestTypingCorners:
     def test_datetime_falls_through_to_categorical(self):
         # Backwards-compatible fallback: PySofra doesn't natively
         # summarise datetimes, so it returns ``"categorical"`` (so the
-        # caller's table doesn't crash). The warning emitted by the
-        # Fix is verified in
-        # tests/test_audit_regressions.py::TestInferKindWarnsOnTemporal.
+        # caller's table doesn't crash). The warning emitted alongside
+        # the fallback is verified in
+        # tests/test_regressions.py::TestInferKindWarnsOnTemporal.
         import pytest
 
         from pysofra.summary.typing import infer_kind

{pysofra-0.1.0a1 → pysofra-0.1.0a3}/tests/test_joss_api_stability.py

@@ -1,12 +1,11 @@
-"""JOSS-grade API-stability snapshot.
+"""Public API-stability snapshot.
 
 Freezes the public surface of ``pysofra`` so any unintended rename,
 removal, or signature change surfaces as a failed test instead of a
 silent breakage downstream.
 
 If you intentionally change the API, update the constants in this file
-**and** bump the project version per semver. This is the contract with
-downstream users that JOSS will scrutinise.
+**and** bump the project version per semver.
 """
 
 from __future__ import annotations

{pysofra-0.1.0a1 → pysofra-0.1.0a3}/tests/test_joss_property_invariants.py

@@ -1,4 +1,4 @@
-"""JOSS-grade property-based invariant testing.
+"""Property-based invariant testing via Hypothesis.
 
 Uses Hypothesis to generate hundreds of random DataFrames and exercise
 PySofra against invariants that must hold *for every input*:

{pysofra-0.1.0a1 → pysofra-0.1.0a3}/tests/test_joss_renderer_consistency.py

@@ -1,4 +1,4 @@
-"""JOSS-grade cross-renderer consistency validation.
+"""Cross-renderer consistency validation.
 
 A single ``SofraTable`` must render to HTML / Markdown / LaTeX / DOCX /
 PPTX / XLSX with **consistent content**: the same cell texts, the

{pysofra-0.1.0a1 → pysofra-0.1.0a3}/tests/test_joss_statistical_correctness.py

@@ -1,13 +1,13 @@
-"""JOSS-grade statistical correctness validation.
+"""Statistical correctness validation against independent references.
 
 Every numeric routine in PySofra is verified against an independent
 reference: scipy / lifelines / statsmodels direct calls, hand-computed
-textbook formulas, or published reference values. A JOSS reviewer
-should be able to point at any number rendered by ``tbl_one`` /
-``tbl_regression`` / ``tbl_survival`` / ``tbl_uvregression`` and have
-this file demonstrate where it came from.
+textbook formulas, or published reference values. Any number rendered
+by ``tbl_one`` / ``tbl_regression`` / ``tbl_survival`` /
+``tbl_uvregression`` can be traced back through this file to where it
+came from.
 
-Each test in this file names its reference source in the docstring.
+Each test names its reference source in the docstring.
 """
 
 from __future__ import annotations

{pysofra-0.1.0a1 → pysofra-0.1.0a3}/tests/test_misc_fixes.py

@@ -1,8 +1,7 @@
-"""Audit-discovered defects pinned by regression tests.
+"""Targeted regression tests for previously-fixed defects.
 
-Each test pins a finding from the most recent god-mode audit pass.
-The names are deliberately literal so that any regression is easy to
-locate.
+Each test pins a specific bug so that any regression is easy to
+locate by name.
 """
 
 from __future__ import annotations
@@ -15,7 +14,7 @@ import pysofra as ps
 
 
 # ----------------------------------------------------------------------
-# Audit P4 — predictor / adjust_for overlap raises cleanly
+# Predictor / adjust_for overlap raises cleanly
 # ----------------------------------------------------------------------
 class TestUvregressionOverlapErrors:
     def test_predictor_also_in_adjust_for_raises(self):
@@ -46,7 +45,7 @@ class TestUvregressionOverlapErrors:
 
 
 # ----------------------------------------------------------------------
-# Audit P2 — formula-API model works through the design refit
+# Formula-API model works through the design refit
 # ----------------------------------------------------------------------
 class TestFormulaAPIRoundTrip:
     def test_design_refit_with_smf_ols(self):
@@ -69,7 +68,7 @@ class TestFormulaAPIRoundTrip:
 
 
 # ----------------------------------------------------------------------
-# Audit P3 — high-cardinality factor scales reasonably
+# High-cardinality factor scales reasonably
 # ----------------------------------------------------------------------
 class TestHighCardinalityFactor:
     def test_50_level_factor_completes_quickly(self):
@@ -91,7 +90,7 @@ class TestHighCardinalityFactor:
 
 
 # ----------------------------------------------------------------------
-# Audit P1 — Newcombe handles imbalanced n
+# Newcombe handles imbalanced n
 # ----------------------------------------------------------------------
 class TestNewcombeImbalanced:
     def test_imbalanced_n_produces_valid_ci(self):

{pysofra-0.1.0a1 → pysofra-0.1.0a3}/tests/test_regressions.py

@@ -1,7 +1,7 @@
-"""Regression tests for issues caught by the full-system audit.
+"""Targeted regression tests for previously-fixed defects.
 
-Every test here corresponds to a real bug the audit surfaced and the
-patch that closed it. Each test should fail if the patch is reverted.
+Each test corresponds to a real bug and the patch that closed it.
+Each test should fail if the patch is reverted.
 """
 
 from __future__ import annotations
@@ -630,9 +630,9 @@ class TestInferKindInfSafe:
         through. Under ``filterwarnings = error`` (the project's own
         pyproject.toml gate, and a common user ``-W error`` posture), that
         warning escalates to an exception and the table build crashes
-        despite the R6 fix. ``test_inf_does_not_leak_runtime_warning``
-        below pins the strict-warning behaviour so the inf path is
-        end-to-end clean.
+        despite the earlier ``int(np.inf)`` fix in ``infer_kind``.
+        ``test_inf_does_not_leak_runtime_warning`` below pins the
+        strict-warning behaviour so the inf path is end-to-end clean.
         """
 
     def test_inf_in_numeric_column_does_not_crash(self):
@@ -685,7 +685,7 @@ class TestInferKindInfSafe:
 
 
             # ----------------------------------------------------------------------
-            # SofraTable must round-trip through pickle ( audit, §F.2)
+            # SofraTable must round-trip through pickle
             # ----------------------------------------------------------------------
 class TestSofraTablePicklability:
     """A SofraTable produced by a builder must survive ``pickle.dumps``/

{pysofra-0.1.0a1 → pysofra-0.1.0a3}/tests/test_validation_fixes.py

@@ -1,8 +1,7 @@
 """Regression tests for input-validation edge cases.
 
-Each test pins a specific defect the external auditor flagged, so any
-future regression surfaces with a clear name pointing back at the
-finding.
+Each test pins a specific defect so any future regression surfaces
+with a clear name pointing back at the finding.
 """
 
 from __future__ import annotations

pysofra-0.1.0a1/.gitignore

@@ -1,90 +0,0 @@
-__pycache__/
-*.py[cod]
-*.egg-info/
-dist/
-build/
-.venv/
-venv/
-.env
-
-.pytest_cache/
-.mypy_cache/
-.ruff_cache/
-.coverage
-htmlcov/
-
-.DS_Store
-.idea/
-.vscode/
-.claude/
-
-site/
-docs/_build/
-
-# Tutorial / example export artefacts — produced on demand by
-# scripts/render_tutorial.py, not part of the source tree.
-*.docx
-*.pptx
-*.xlsx
-examples/tutorial_*.docx
-examples/tutorial_*.pptx
-examples/tutorial_*.xlsx
-examples/tutorial_*.png
-!tests/fixtures/**/*.docx
-!tests/fixtures/**/*.pptx
-!tests/fixtures/**/*.xlsx
-
-# Hypothesis example database (auto-managed)
-.hypothesis/
-
-# uv writes a lockfile when ``uv add`` / ``uv pip install`` is used
-# locally; the project pins its runtime versions explicitly in
-# pyproject.toml + paper/replication/requirements.txt, so the uv
-# lockfile is a local convenience that should not ship.
-uv.lock
-
-# JSS paper bundle — paper/ contains the manuscript, bibliography,
-# embedded figures, and replication archive. The .tex / .bib / .py /
-# .R sources are ALWAYS tracked. Replication outputs are tracked in
-# the text formats reviewers diff (json / tex / md / html / png /
-# pdf / svg) so the archive ships with a working baseline; only the
-# heavy regenerable Office binaries (docx / pptx / xlsx) and the
-# fontconfig / mpl cache directories are ignored.
-paper/replication/trial.csv
-paper/replication/table*.docx
-paper/replication/table*.pptx
-paper/replication/table*.xlsx
-paper/replication/.mplconfig/
-
-# Submission-required text outputs and figures: explicit re-include.
-# The blanket repo-wide ``*.docx`` rule at the top of this file would
-# otherwise eat ``paper/replication/table*.html`` etc. — these
-# negations keep them tracked.
-!paper/replication/paper_outputs*.json
-!paper/replication/table*.html
-!paper/replication/table*.tex
-!paper/replication/table*.md
-!paper/replication/figures/
-!paper/replication/figures/*
-
-# The `paper/figures/` directory IS tracked because paper.tex embeds
-# its PDFs/PNGs via \includegraphics. To regenerate, run
-# paper/replication/example_trial.py then copy the desired files in.
-!paper/figures/
-!paper/figures/*
-
-# LaTeX build artefacts (in case anyone compiles paper.tex locally).
-paper/*.aux
-paper/*.log
-paper/*.out
-paper/*.toc
-paper/*.bbl
-paper/*.blg
-paper/*.synctex.gz
-paper/paper.pdf
-
-
-# paper/ is excluded from the public repo until JSS submission lands.
-# It stays in the local working tree for editing; the replication
-# archive moves into the repo at submission time.
-paper/