Skip to content

Latest commit

 

History

History
114 lines (77 loc) · 4.39 KB

CONTRIBUTING.md

File metadata and controls

114 lines (77 loc) · 4.39 KB

Contributing

Contributions to this project are welcome and encouraged! Please be sure to review our Code of Conduct before participating in any community discussions in this repo.

Issues

The easiest way to contribute to the radiant-mlhub library is by opening issues to report bugs or request features that might be useful to you or others. Before opening a new issue, please review the documentation and existing issues (both open and closed) to see if the topic is already covered by an existing issue. Issues in this repository should pertain directly to the radiant_mlhub Python client itself and not the Radiant MLHub API or the datasets contained therein. For questions and issues regarding those resources, please email [email protected]. If it is unclear whether the issue pertains to the API or the Python client, please start by emailing [email protected] and we will help to assess the scope of the issue.

Issues in this repository follow one of 2 templates: bug reports or feature requests. For more general questions or troubleshooting that does not fall into one of these two categories, please email us at [email protected]. When creating a new issue, please follow the appropriate template as closely as possible and provide as much detail as you can.

Pull Requests

We welcome and encourage pull requests from the community to address bugs or feature requests. Before submitting a pull request, please open an issue to discuss the feature or bug fix that you intend to implement. This will help us, and the rest of our user community, give input on the design for any changes and ensure that we are not already doing work on similar features.

All branches should originate from the main branch and all PRs should be made against the main branch.

New PRs will use our PR template, which includes the following checklist:

  • Unit tests have been added/updated to cover changes
  • Documentation has been added/updated
  • An entry has been added to the CHANGELOG (if necessary)

Development

Development Environment

Install development dependencies:

pip install -r requirements_dev.txt

Tests

This library uses pytest for unit testing, flake8 and mypy for type checking. To run them locally:

pytest
flake8
mypy radiant_mlhub/

VCR.py Cassettes

If you need to re-record any of the VCR.py cassettes, you will need to create a test/.env file with the following content:

MLHUB_API_KEY=<SUBSTITUTE_VALID_API_KEY_HERE>

The API key will be used to make new requests, but will not be recorded in any of the cassettes. All user-specific download links are also sanitized before being recorded to a cassette.

To re-record a cassette use:

pytest {path_to_test(s)} --record-mode rewrite --block-network

Releases

NOTE: This section is primarily for the maintainers and will almost never be necessary for community contributions.

When the code on the main branch has stabilized, we will cut a new release using the following procedure:

  1. Create a release/<version> branch from main, where <version> is the version to be released (e.g. v1.0.0).

  2. Run tbump to bump the package to the desired version:

    tbump --no-push --no-tag <NEW_VERSION>

  3. Commit changes to the CHANGELOG.md including the new version, and the compare links at the bottom.

  4. Test and make any final changes on the release/* branch

  5. Put in a PR against main

    This will trigger the CI to run unit tests and to publish a test package to TestPyPi.

  6. Approve and merge into main

    Once the PR has been approved, merge into main

  7. Tag and publish release. Use an annotated git tag.

    This will trigger the CI to publish the package to PyPi

After the release has been published to PyPi, you should also update the recipe in the conda-forge feedstock to use the new release. See the README in that repo for instructions on how to update the recipe.