Skip to content

CONTRIBUTING.md

Latest commit

 

History

History
108 lines (76 loc) · 3.19 KB

CONTRIBUTING.md

File metadata and controls

108 lines (76 loc) · 3.19 KB
Raw

Contributing to privacy-mask

Thanks for your interest in contributing! This guide will help you get started.

Quick Setup

git clone https://github.com/fullstackcrew-alpha/privacy-mask.git
cd privacy-mask
pip install -e .

Running Tests

PYTEST_DISABLE_PLUGIN_AUTOLOAD=1 python -m pytest tests/ -v

All tests must pass before submitting a PR. We currently have 208+ tests covering all 47 detection rules.

How to Add a New Detection Rule

Adding a new regex rule is the easiest way to contribute. Here's the step-by-step process:

1. Edit mask_engine/data/config.json

Add your rule to the rules array:

{
  "name": "MY_NEW_RULE",
  "pattern": "MY-\\d{6}-[A-Z]{2}",
  "flags": ["IGNORECASE"],
  "enabled": true
}

Important: Since patterns are JSON strings, backslashes must be double-escaped (\\d not \d).

2. Add Tests in tests/test_detector.py

Every rule needs both positive (should match) and negative (should NOT match) test cases:

def test_my_new_rule_positive(detector):
    """MY_NEW_RULE should match valid patterns."""
    assert_detects(detector, "MY-123456-AB", "MY_NEW_RULE")
    assert_detects(detector, "MY-999999-ZZ", "MY_NEW_RULE")

def test_my_new_rule_negative(detector):
    """MY_NEW_RULE should not match common words or partial patterns."""
    assert_not_detects(detector, "MY-12-AB", "MY_NEW_RULE")
    assert_not_detects(detector, "MYSTERY", "MY_NEW_RULE")

3. Test for False Positives

This is critical. OCR can read common English words as uppercase text, so make sure your pattern doesn't match words like:

  • ORGANIZATION, REQUIRED, CONTINUE, INFORMATION
  • Common abbreviations in your target language

4. Run Tests

PYTEST_DISABLE_PLUGIN_AUTOLOAD=1 python -m pytest tests/test_detector.py -v -k "my_new_rule"

Other Ways to Contribute

Improve OCR Accuracy

  • Better image preprocessing strategies
  • Support for additional OCR engines
  • Multi-language OCR improvements

Report False Positives

If you find that privacy-mask incorrectly detects normal text as sensitive data, please open an issue with:

  • The text that was incorrectly detected
  • Which rule triggered the false positive
  • A screenshot if possible

Improve Documentation

  • Fix typos, clarify instructions
  • Add examples for your country's ID formats
  • Translate documentation

Submitting a Pull Request

  1. Fork the repo and create a feature branch from main
  2. Make your changes and add tests
  3. Run the full test suite and ensure all tests pass
  4. Submit a PR with a clear description of the change

PR Guidelines

  • Keep PRs focused — one feature or fix per PR
  • Include test cases for new detection rules (both positive and negative)
  • Never commit real secrets, API keys, or personal data — use constructed test strings
  • Update the README if adding user-facing features

Issues

  • Use the issue tracker for bug reports and feature requests
  • Tag issues with good first issue if they're suitable for newcomers
  • Check existing issues before opening a new one

License

By contributing, you agree that your contributions will be licensed under the MIT License.