fix(enterprise): fail closed on unsupported anonymization config; reject non-finite trust weights

[kevincostner17]

Summary

#247: EnterpriseConfig.anonymization is accepted but never applied.

• The bug: EnterpriseConfig.__post_init__ type-checks the field, but no stage of clean_enterprise reads it. Masking is driven only by masking and privacy. A caller who set anonymization=(AnonymizationConfig(strategy="redact"),) got raw PII back, with mask_report=None, privacy_report=None and no warning.
• The fix fails closed: clean_enterprise raises ValueError when anonymization is non-empty. The message points to masking= (MaskingRule) or privacy= (PIIDetectionConfig with enable_privacy_detection=True).
  • EnterpriseConfig still constructs with the field, so existing config objects keep loading.
  • The field is documented as unsupported in the EnterpriseConfig, AnonymizationConfig and clean_enterprise docstrings.
  • The check runs before any cleaning, whether or not enable_masking is set.
  • The CLI and polars inputs go through clean_enterprise, so they fail the same way. The CLI prints a one-line error and exits 1.
• Why fail closed rather than warn or leave it:
  • A privacy control that silently does nothing is worse than an explicit error. A user who configured anonymization believes the output is safe to share, and a warning is easy to miss in a pipeline. An error isn't, and the fix is a one-line config change.
  • The change is easy to reverse. Once anonymization is implemented, the check comes out and the same configs start working, with no API break.

#277: TrustScoreWeights accepts NaN and inf.
value < 0 and sum <= 0 are both false for NaN, and +inf passes both, so the trust score came out nan with grade F. Each weight must now be finite (math.isfinite), otherwise ValueError.

Tests

• tests/test_enterprise_config.py:
  • NaN, +inf and -inf are rejected for all four weights.
  • EnterpriseConfig(anonymization=...) still constructs and type-checks.
• tests/test_enterprise_interface.py:
  • clean_enterprise raises with enable_masking both True and False.
  • It raises for polars input (skipped if polars isn't installed).
  • The CLI clean path exits non-zero, prints the error to stderr and writes no output file.

Verification

• ruff check .: all checks passed
• mypy src/freshdata: no issues in 202 source files
• pytest -m "not online and not large", Python 3.12: 4256 passed, 6 skipped
• pytest -m "not online and not large", Python 3.9 / pandas 1.5: 4252 passed, 10 skipped
• The repro scripts from both issues now raise ValueError.

Closes #247
Closes #277

[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: c7be8f48-e2fe-4c5b-aa3f-7e6c7c535cf8

---

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.}

[github-actions]

FreshData benchmark report — performance

• freshdata: ?
• python: ?
• platform: ?

fixture	n_rows	n_cols	p50 s	p95 s	peak MB	repair %	false-repair %	preserve %	trust	monotonic	export %

Authored-code reduction (Metric 6)