Skip to content

CONTRIBUTING.md

Latest commit

 

History

History
82 lines (60 loc) · 3.64 KB

CONTRIBUTING.md

File metadata and controls

82 lines (60 loc) · 3.64 KB

Contributing to Flare AI Kit

First off, thank you for considering contributing to Flare AI Kit!

We welcome contributions from the community to help make this SDK robust, feature-rich, and easy to use.

How to Contribute

  1. Search existing issues/PRs to avoid duplicates.
  2. For opening an issue, use labels bug, feature, docs, or question.
  3. Fork the repo and create a feature branch: 
    git checkout -b feature/your-feature-name
  4. Implement, test, then submit a PR referencing the issue.

✨ Write High-Quality Code

We strive for high-quality, maintainable code. Please adhere to the following principles:

  1. Clean Interfaces & Simplicity:

    • Follow the Zen of Python principle: "There should be one-- and preferably only one --obvious way to do it."
    • Design clear, minimal, and predictable interfaces for functions and classes.
    • Use descriptive names for variables, functions, and classes.
    • Keep code focused and avoid unnecessary complexity.
    • Use Google-style docstrings.

  2. Use dependencies sparingly:

    • Every dependency added is a potential security risk.
    • If the dependency is required, it should live in under [project.optional-dependencies] with a key marking the high level function, e.g. rag, consensus.

  3. Formatting, Linting & Type-Checking:

    • Use ruff for formatting and linting, both settings are defined in pyproject.toml.
    • Use pyright for type checking, all new code MUST include accurate type hints. Avoid using typing.Any unless absolutely necessary and clearly justified in comments.
    • You can also install Pylance into VSCode for an easier experience.
    • Ensure your code passes all checks locally without errors: 
      uv run ruff format
uv run ruff check --fix
uv run pyright

  4. Test Extensively:

    • New features must include appropriate unit and/or integration tests using pytest.
    • Bug fixes should include tests demonstrating the fix.
    • All tests must pass locally before submitting a pull request. Run tests via: 
      uv run pytest tests/unit/your-test.py -v
uv run pytest tests/integration/your-test.py -v

  5. Use Conventional Commits:

    • All commit messages must adhere to the Conventional Commits specification. This helps automate changelog generation and provides a clear commit history.
    • Read the spec: https://www.conventionalcommits.org/
    • Format: <type>[optional scope]: <description>
    • Examples:
      • feat(rag): add support for Neo4j AuraDB connection
      • fix(ecosystem): correct decimal calculation in FTSO price feed
      • docs(readme): update README with installation instructions
      • test(rag): add unit tests for GraphDbSettingsModel
      • chore(deps): update ruff version in pyproject.toml

🚨 PR Checklist

Before you mark your PR as ready-to-review:

  • Branch is up-to-date with main.
  • All linting, formatting, and type checks pass.
  • Tests cover new code (unit + integration).
  • Commit messages follow Conventional Commits.
  • PR description includes purpose and implementation details.

📜 License

By contributing to Flare AI Kit, you agree that your contributions will be licensed under the Apache License 2.0. See the LICENSE file for more information.

Thank you for contributing!