buildstream-plugins-community 2.2.0

A collection of community-maintained plugins.

A collection of community-maintained plugins for BuildStream 2. You can find the docs here.

This repo collects plugins which don’t provide the strong API guarantees required by the buildstream-plugins project.

How to use this repo

The plugins in this repo do not provide strong API guarantees or backwards compatibility. You should use a specific commit in your project and update it manually as needed.

You are recommended to import these plugins into your project using the junction plugins feature of BuildStream 2.0 so you can control exactly what Git commit is used. Please note that using it this way requires at least BuildStream 2.2. This is done in several stages documented below:

Using via a junction

First, make sure you have the git source from buildstream-plugins available and declared in your project.conf file. If you installed the PyPI package then you can import it as a pip plugin:

plugins:
- origin: pip
  package-name: buildstream-plugins
  sources:
  - git

Now, add a junction element referencing this repo. Here’s an example you could save as buildstream-plugins-community.bst in your elements directory:

kind: junction

sources:
- kind: git
  url: https://gitlab.com/BuildStream/buildstream-plugins-community.git
  track: master

You can then run bst source track buildstream-plugins-community.bst to set the ref field appropriately.

Finally you can define specific plugins you want to use in project.conf:

plugins:
- origin: junction
  junction: buildstream-plugins-community.bst
  sources:
  - pypi
  elements:
  - pep517

Alternative methods

You can use Git’s ‘submodules’ feature to import this repo into your project’s repo, then declare the plugins as local plugins.

BuildStream also supports pip plugins which are imported from the host Python environment. While buildstream-plugins-community is available on PyPI, the project does not provide any backwards compatibility or “semantic versioning” guarantees. Make sure you can control exactly what of the package version is used if you consume it via Pip.