Collection of GitHub workflows

poetry.yml workflow

A Poetry-based workflow that runs jobs in a custom Docker image. This workflow provides a job scripts, that checks out a project, installs its dependencies and optionally runs list of scripts, provided in poetry_scripts input using poetry run command in a Python environment with all project dependencies.

It can be conveniently complemented by Poe the Poet scripts.

It has the following inputs to control its behaviour:

• image_name - Docker image for the job container
• image_options - Options to pass to Docker
• poetry_scripts - List of scripts to run, e.G. '["poe lint-all", "poe test-unit"]'
• debug - Use this boolean input to troubleshoot poetry dependencies installation with richer log output
• home - Allows to override the value of $HOME environment variable, which is set by GitHub when running jobs in a custom Docker image
• github_app_id (and github_app_private_key secret) to provide GitHub app credentials used to generate GitHub token that will be used to access GitHub for git+https:// dependencies

If you specify image_name, don't forget to provide DOCKER_USERNAME and DOCKER_PASSWORD secrets.

In addition, if you provide SSH_KEY secret, it will be added as a private SSH key to the home folder. You can use this to install dependencies from private GitHub repositories or other private servers.

Codecov support in poetry.yml workflow

You can add Codecov.io support by specifying CODECOV_TOKEN secret and making sure you run pytest --cov ... somewhere in your poetry_scripts.