Skip to content

Latest commit

 

History

History
221 lines (159 loc) · 9.05 KB

CONTRIBUTING.rst

File metadata and controls

221 lines (159 loc) · 9.05 KB

Contributing guide

Table of Contents

Contributing to Squidpy

Clone Squidpy from source as:

git clone https://github.com/scverse/squidpy
cd squidpy
git checkout main

Install the test and development mode:

pip install -e'.[dev,test]'

Optionally install pre-commit. This will ensure that the pushed code passes the linting steps:

pre-commit install

Although the last step is not necessary, it is highly recommended, since it will help you to pass the linting step (see Code style guide). If you did install pre-commit but are unable in deciphering some flags, you can still commit using the --no-verify.

Codebase structure

The Squidpy project:

  • squidpy: the root of the package.
    • squidpy/gr: the graph module, which deals with building a spatial graph, running statistical tests on graphs and features etc.
    • squidpy/im: the image module, which deals with image feature calculation, cropping, etc.
    • squidpy/pl: the plotting module, which contains all the plotting functions from the graph and image modules.
    • squidpy/constants: contains internal and (possibly in the near future) external constants.

Tests structure:

Code style guide

We rely on black and isort to do the most of the formatting - both of them are integrated as pre-commit hooks. You can use tox to check the changes:

tox -e lint

Furthermore, we also require that:

  • functions are fully type-annotated.
  • exception messages are capitalized and end with ..
  • warning messages are capitalized and do not end with ..
  • when referring to variable inside an error/warning message, enclose its name in `.
  • when referring to variable inside a docstrings, enclose its name in ``.

Testing

We use tox to automate our testing, as well as linting and documentation creation. To run the tests, run:

tox -e py{38,39,310}-{linux,macos}

depending on the Python version(s) in your PATH and your operating system. We use flake8 and mypy to further analyze the code. Use # noqa: <error1>,<error2> to ignore certain flake8 errors and # type: ignore[error1,error2] to ignore specific mypy errors.

To run only a subset of tests, run:

tox -e <environment> -- <name>

where <name> can be a path to a test file/directory or a name of a test function/class. For example, to run only the tests in the nhood module, use:

tox -e py39-linux -- tests/graph/test_nhood.py

If needed, a specific tox environment can be recreated as:

tox -e <environment> --recreate

Writing documentation

We use numpy-style docstrings for the documentation with the following additions and modifications:

  • no type hints in the docstring (applies also for the return statement) are allowed, since all functions are required to have the type hints in their signatures.
  • when referring to some argument within the same docstring, enclose that reference in ``.
  • prefer putting references in the references.bib instead under the References sections of the docstring.
  • use docrep for repeating documentation.

In order to build the documentation, run:

tox -e docs

Since the tutorials are hosted on a separate repository (see Writing tutorials/examples), we download the newest tutorials/examples from there and build the documentation here.

To validate the links inside the documentation, run:

tox -e check-docs

If you need to clean the artifacts from previous documentation builds, run:

tox -e clean-docs

Writing tutorials/examples

Tutorials and examples are hosted on a separate repository called squidpy_notebooks. Please refer to this guide for more information.

Submitting a PR

Before submitting a new pull request, please make sure you followed these instructions:

  • make sure that you've branched off main and are merging into main
  • make sure that your code follows the above specified conventions (see Code style guide and Writing documentation).
  • if applicable, make sure you've added/modified at least 1 test to account for the changes you've made
  • make sure that all tests pass locally (see Testing).
  • if there is no issue which this PR solves, create a new one briefly explaining what the problem is.
  • make sure that the section under ## Description is properly formatted if automatically generating release notes, see also Creating release notes.

Making a new release

New release is always created when a new tag is pushed to GitHub. When that happens, a new CI job starts the testing machinery. If all the tests pass, new release will be created on PyPI. Bioconda will automatically notice that a new release has been made and an automatic PR will be made to bioconda-recipes. Extra care has to be taken when updating runtime dependencies - this is not automatically picked up by Bioconda and a separate PR with the updated recipe.yaml will have to be made.

Easiest way to create a new release it to create a branch named release/vX.X.X and push it onto GitHub. The CI will take care of the following:

  • create the new release notes
  • bump the version and create a new tag
  • run tests on the release/vX.X.X branch
  • publish on PyPI after all the tests have passed
  • merge release/vX.X.X into main

Alternatively, it's possible to create a new release using bump2version, which can be installed as:

pip install bump2version

Depending on what part of the version you want to update, you can run on main:

bump2version {major,minor,patch}

By default, this will create a new tagged commit, automatically update the __version__ wherever necessary. Afterwards, you can just push the changes to upstream by running:

git push --atomic <branch> <tag>

or set push.followtags=true in your git config and do a regular git push. In this case, CI will not create any release notes, run tests or do any merges.

Creating release notes

By default, news fragments are automatically generated from successfully merged PRs using. Everything under ## Description section will be rendered as .rst files and automatically committed in the target branch in docs/source/release/changelog. When a new release happens, towncrier gathers all news fragments and creates the release notes under docs/source/release.

When submitting a PR, it should be tagged with one of the following tags, in order for towncrier to know under which section to render the news:

  • bugfix: the PR fixes some bug
  • feature: the PR introduces a new feature
  • deprecation: the PR deprecates something (e.g. a function)
  • doc: the PR is related to documentation
  • misc: the PR is not applicable to the above

If no label (or ignore-towncrier label) is specified, no news will be generated for the PR. If more than 1 is specified, the first one is used.

To manually create news fragment, make sure that the PR doesn't generate it from the description as described above. The command to run is towncrier <PR_NUMBER>.<LABEL>, where <LABEL> is one of the labels described above. This will create a new file in the appropriate location that needs to me modified and subsequently committed.

To locally create the news fragment from an already existing PR, just run:

tox -e news -- <PR_NUMBER> --add-author

You can modify the created fragment, then commit and run:

tox -e update-dev-notes

In order to inject the fragment in the release note.

Lastly, in order to see how the current news fragments would look like in the release notes, run:

towncrier build --draft

Troubleshooting

  • The enchant C library was not found This can happen during the documentation build and because of a missing dependency for spell checker. The installation instructions for the dependency can be found here.