Skip to content
You're viewing an older version of this GitHub Action. Do you want to see the latest version instead?

setup-rye

Set up your GitHub Actions workflow with a specific version of rye.

  • Install a version of rye and add it to the path
  • Cache the installed version of rye to speed up consecutive runs on self-hosted runners
  • Register problem matchers for error output
  • Optional: Cache the virtual environment created by rye
  • Optional: Verify the checksum of the downloaded rye executable

Usage

Example workflow can be found in this repo

- name: Install the latest version of rye
  uses: eifinger/setup-rye@v3

Install latest version

By default this action installs the version defined as default in action.yml. This gets automatically updated in a new release of this action when a new version of rye is released. If you don't want to wait for a new release of this action you can use use version: latest.

Warning

Using the latest version means that the rye executable gets downloaded every single time instead of loaded from the tools cache. This can take up to 20s depending on the download speed. This does not affect the cached version of .venv when caching is enabled.

- name: Install a specific version
  uses: eifinger/setup-rye@v3
  with:
    version: 'latest'

Install specific version

You can also specify a specific version of rye

- name: Install a specific version
  uses: eifinger/setup-rye@v3
  with:
    version: '0.12.0'

Validate checksum

You can also specify a checksum to validate the downloaded file. Checksums of versions 0.12.0 and later are automatically verified by this action. The sha265 hashes can be found on the releases page of the rye repo.

- name: Install a specific version and validate the checksum
  uses: eifinger/setup-rye@v3
  with:
    version: '0.12.0'
    checksum: 'c48d850e90649d868d512f60af67c74aa844d80f951fdb38589220662e709da7'

Enable caching

If you enable caching the virtual environment which gets created by rye under .venv will be cached. This can speed up runs which can reuse the cache by several minutes.

You can optionally define a custom cache key prefix.

- name: Enable caching and define a custom cache key prefix
  id: setup-rye
  uses: eifinger/setup-rye@v3
  with:
    enable-cache: true
    cache-prefix: 'optional-prefix'

When the cache was successfully restored the output cache-hit will be set to true and you can use it in subsequent steps. For the example above you can use it like this:

- name: Do something if the cache was restored
  if: steps.setup-rye.outputs.cache-hit == 'true'
  run: echo "Cache was restored"

Working directory

If your rye project is not at the root of the repository you can specify the working directory relative to the repository root. This is useful for monorepos.

- name: Enable caching and define a working directory
  uses: eifinger/setup-rye@v3
  with:
    enable-cache: true
    working-directory: 'path/to/rye/project'

Local storage path

If you want to save the cache to a local path you can specify the path with the cache-local-storage-path input. This can be useful if you are on a self hosted runner and want to save time and network traffic.

- name: Enable caching and define a custom cache path
  uses: eifinger/setup-rye@v3
  with:
    enable-cache: true
    cache-local-storage-path: '/path/to/cache'

It is recommended to cleanup the storage path to avoid running out of disk space. One option is to use a cron job to delete files older than 7 days like below.

0 0 * * * find /home/ubuntu/setup-rye-cache -type d -mtime +7 -exec rm -rf {} \;

API rate limit

To avoid hitting the error API rate limit exceeded you can supply a GitHub token with the github-token input.

- name: Install rye and supply a GitHub token
  uses: eifinger/setup-rye@v3
  with:
    github-token: ${{ secrets.GITHUB_TOKEN }}

Limitations

  1. Currently only Linux and macOS is supported see issue #10

How it works

This action downloads rye from the releases of the rye repo and uses the GitHub Actions Toolkit to cache it as a tool to speed up consecutive runs especially on self-hosted runners.

The installed version of rye is then added to the runner path so other steps can just use it by calling rye. To prevent interfering the other rye processes on the same runner RYE_HOME gets set to the repository roots parent and is cleaned up after a workflow run.