pytest-cldf

This pytest plugin can be used to run validation of CLDF datsets via the pytest test runner. In particular, this allows hooking up continuous validation with CI services like travis-ci easily.

Continuous validation via Travis-CI

To make sure a dataset - curated in a git repository on GitHub - is continuously, i.e. after each commit, validated, you have to

• hook up the repository with Travis-CI
• add a Travis configuration file .travis.yml with the following content:

language: python
python: "3.6"
cache: pip
before_cache: rm -f $HOME/.cache/pip/log/debug.log
install: pip install pytest-cldf
script: pytest --cldf-metadata=cldf/cldf-metadata.json test.py

• add the python file implementing the tests test.py:

def test_valid(cldf_dataset, cldf_logger):
    assert cldf_dataset.validate(log=cldf_logger)

Extended validation

TODO