Skip to content

Latest commit

 

History

History
63 lines (37 loc) · 1.35 KB

CONTRIBUTING.rst

File metadata and controls

63 lines (37 loc) · 1.35 KB

Contributing

Any contributions are welcome!

Open new issues at https://github.com/DataCrunch-io/datacrunch-python/issues.

You can open pull requests by following the steps:

Code Contribution:

  1. Fork the datacrunch-python repo on GitHub.

  2. Clone your fork locally:

    git clone git@github.com:{your_username}/datacrunch-python.git
    
    cd datacrunch-python
    
  3. Create virtual environment & install this local copy into the virtual environment:

    python3 -m venv datacrunch_env && source ./datacrunch_env/bin/activate
    
    python3 setup.py develop
    
  4. Create a new branch:

    if it's a feature:

    git checkout -b feature/new-feature-name
    

    or a bugfix:

    git checkout -b hotfix/some-bugfix
    
  5. Make your local changes

  6. Run tests:

    pytest
    
  7. Commit and push:

    git commit .am "Detailed commit message"
    
    git push origin {branch-name}
    
  8. Submit a pull request in GitHub.

Pull Request Guidelines

  1. The pull request should include tests.
  2. Please add documentation docstrings and type hinting to any new feature.

Release Guidelines

Some steps for releasing a new version:

  1. Update the version in __version__.py
  2. Add an entry to the CHANGELOG.rst file
  3. Push the changes to the master branch and add a new release tag.