Skip to content

Latest commit

 

History

History
33 lines (26 loc) · 1.53 KB

CONTRIBUTING.md

File metadata and controls

33 lines (26 loc) · 1.53 KB
Raw

Contributing to ansible-tailscale-inventory

Contributions are welcomed! Please feel free to submit issues and propose fixes and changes to the script.

Development dependencies

  • make
  • python 3.8 or higher

Running make dev will install more pip dependencies.

Formatting & Linting

Code is expected to formatted and type annotated to conform with mypy, and ruff. Formatting can be tested with make test. Some formatting can be automatically applied by running make fmt.

Testing

We use pytest to ensure the script works as expected as we change things. Simply running pytest with no arguments will test the script. Or, you can run make test to run the full test suite with linting.

Pytests can be found in the tests/ dir. The main existing test simply checks that a given Tailscale status output will produce an Ansible inventory in an expected format. The expected inputs and outputs are defined in tests/mock_data.py.

For most inventory structure changes it should be adequate to simply update the input and output data structures in tests/mock_data.py. A recommended development workflow is to update the mock data and then start updating the script to produce the expected output, running pytest along the way until it passes.

A GitHub Actions workflow will test changes on Pull Request. It can also be run on-demand against branches. The GH Actions workflow will run tests using multiple versions of python to ensure the script remains compatible.