test(metamorphic): assert what cleaning must and must not change

[kevincostner17]

Why

The suite had no invariance tests on the public API. Worse, tests/test_execution/test_action_parity.py::_normalize sorts rows and columns before comparing engines, so an order-sensitivity bug in fd.clean would pass that gate silently.

Invariances

Reordering rows (4 seeds) or columns, adding an unrelated column, renaming a column whose meaning is pinned by semantic_context, and cleaning twice must change no surviving cell.

Row order is checked against imputation specifically — impute reads the whole column and is the most plausible place for a positional accident to leak in. Both mean and missforest are covered.

Required changes

Without these, a cleaner that returned its input untouched would satisfy every invariance above and look perfect:

Property
valid amount → "apple" changes the result
valid country → "N/A" changes the result
outliers="flag" flags and leaves the value at 9999.0

The outlier case asserts both halves deliberately: silence would be a missed detection, a changed value would be an unrequested repair.

Honest framing

Every property already holds on main, so these are guards against regression rather than bug reports. The engine-parity case is the exception in spirit — it covers ground test_action_parity cannot, because it compares the frames as returned instead of sorting them first. Both polars and duckdb run it; neither is skipped.

Two things that looked like defects and were not

Recorded in the test docstrings so the next reader doesn't re-derive them:

• assert_frame_equal failed with an empty message while every column, dtype, index and shape matched. fd.clean returns CleanResult, whose _constructor is pd.DataFrame by design (result.py:22), so .reindex() yields a plain frame and the comparison failed on frame type. That design is right — a derived or filtered frame should not carry a .report() that no longer describes it — so the helper normalises both sides instead.
• A 2-vs-2 categorical variant tie ('USA' vs 'usa') is not normalised in either input order, so there is no order-dependent tie-break to guard against.

Verification

• 20 tests, all passing; polars and duckdb parity cases confirmed running (not skipped).
• Full suite py3.12: 6656 passed, 22 skipped, 0 failed, coverage 93.91%.
• ruff check . clean.
• No library code changed.

[coderabbitai]

Important

• 🔍 Trigger review

This repository does not receive automatic reviews because it has fewer than 10 stars.

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Advanced

Run ID: b4e738a7-07e2-495b-839d-2c4ac67fe0d9

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}