build-conda-action

A GitHub action that builds a Conda recipe

Action Marketplace · Report Bug · Request Feature

A Github Action for building a Conda package stored in the project using conda-build. It is meant to be simple to use with reasonable defaults that should work without any inputs when your project Conda configuration is setup according to the guidelines. However it also supports powerful customization of the conda build configuration through the inputs. It supports both GitHub hosted and self-hosted runners by setting the "base_env_prefix" input.

Getting Started

All inputs are optional. Just Follow the project configuration guidelines below

Standard Conda Configuration for Basic Operation

• Name the conda recipe folder "conda-recipe"
• Place any test result, lint, or other build outputs you'd like uploaded as artifacts in a "test_results" folder
• The build environment will contain Conda, Conda-build, and Conda-verify by default.

  The build environment can be customized using the conda_build_env_filepath input. Include just the packages needed to build the package. Packages required to test or run the project are specified in the recipe's meta.yml.

Basic Usage

job-name:
  runs-on: 'ubuntu-latest'
  steps:
    - uses: actions/checkout@v4
    # Builds the package using the standard configuration
    - uses: cascode-labs/build-conda-action/action.yml@v0

Custom Configuration

Inputs

• recipe_path: The path to the Conda recipe from the repo root.
  Default: 'conda-recipe'
• conda_build_env_filepath: Path to a custom build yml env definition file

  Best Practice: If you need to customize the build environment, place a customized build.yml file at 'envs/build.yml'

• base_env_prefix: The prefix of the base Conda environment.
  Default: '/usr/share/miniconda'

  Use this input for initializing the base environment of a self-hosted runner by providing the base environment prefix of the runner

• package_artifact_name: The display name of the uploaded package artifact.
  Default: 'conda_package'
• test_results_artifact_name: The display name of the uploaded test results artifact.
  Default: 'test_results'
• build_options: Options to be passed to conda-build to customize the build configuration. For example, you can provide a different set of channels.
  Default: '-c defaults -c conda-forge'

Outputs

• package-filepath: The file path of the generated package. It will return "None" if no package was created.

  The folder "built_package_outputs" is reserved for the build process. It is used to assemble the package to be uploaded as artifacts

Artifacts Uploaded

The following are uploaded to the run as artifacts.

• Conda packages: The built Conda packages
• Test Results: Any additional testing or build artifacts that are copied into the test_results folder. Examples include test results, test reports, lint reports or coverage reports.

Example customized workflow

job-name:
  runs-on: 'ubuntu-latest'
  steps:
    - uses: actions/checkout@v4
    # Initializes Conda on a GitHub hosted Runner
    - uses: conda-incubator/setup-miniconda@v2
      with:
        python-version: 3.7
    # Runs the action with the following inputs or defaults if not specified.
    - uses: cascode-labs/build-conda-action/action.yml@v0
      with:
        recipe_path: '{NEW_RECIPE_PATH}'
        base_env_prefix: '{NEW_BASE_PREFIX}'
        package_artifact_name: '{NEW_PACKAGE_NAME}'
        test_results_artifact_name: '{NEW_TEST_RESULTS_NAME}'
        build_options: '{NEW_CHANNELS_TO_USE}'
        conda_build_env_filepath: '{NEW_BUILD_ENV_PATH}'

Roadmap

See the open issues for a list of proposed features (and known issues). Milestones outline the release version of each issue.

Contributing

Contributions are what make the open source community such an amazing place to learn, inspire, and create. Any contributions you make are greatly appreciated.

1. Fork the Project
2. Create your Feature Branch (git checkout -b feature/AmazingFeature)
3. Commit your Changes (git commit -m 'Add some AmazingFeature')
4. Push to the Branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

Testing

This repo contains a test workflow with each job of the workflow as a different test case.