Testing

Informal Testing/Development

We utilize marimo notebooks under the notebooks/dev/ directory for testing the api. After setting up the envionment, you can run marimo edit to host notebook server on your local machine. For any examples worth including in the documentation, we add these notebooks to the notebooks/examples/ directory.

Feel free to go about your development process however you see fit, but for any major functionality changes, make sure any examples are updated accordingly. If you choose to develop with Jupyter notebooks, you can convert to marimo via marimo convert *.ipynb -o *.py and vice-versa.

Formal Testing

We utilize pytest for testing our codebase.

Unit Testing

Unit tests are under the tests/caml/ directory following the same structure of the caml/ prefixed by “test_”. For example, if we wanted to write tests for cate.py, we would create a new file to build these tests tests/caml/core/test_cate.py.

To run unit tests, simply run pytest in terminal:

This will run your unit tests (with respective output printed in terminal).

Additionally, an html report of the coverage will be exported to tests/reports/htmlcov, which can be a useful tool for interactively evaluating the coverage of your unit tests.

Advanced Testing

Unit tests are automatically run during PR process via GitHub Actions. Integration & regression testing forthcoming.