//python/extensions:python.bzl

Python toolchain module extensions for use with bzlmod

module ext python

Bzlmod extension that is used to register Python toolchains.

python.rules_python_private_testing

rules_python_private_testing(register_all_versions=False)

Attributes:
python.toolchain

toolchain(python_version, configure_coverage_tool=False, ignore_root_user_error=False, is_default=False)

Tag class used to register Python toolchains. Use this tag class to register one or more Python toolchains. This class is also potentially called by sub modules. The following covers different business rules and use cases.

Toolchains in the Root Module

This class registers all toolchains in the root module.

Toolchains in Sub Modules

It will create a toolchain that is in a sub module, if the toolchain of the same name does not exist in the root module. The extension stops name clashing between toolchains in the root module and toolchains in sub modules. You cannot configure more than one toolchain as the default toolchain.

Toolchain set as the default version

This extension will not create a toolchain that exists in a sub module, if the sub module toolchain is marked as the default version. If you have more than one toolchain in your root module, you need to set one of the toolchains as the default version. If there is only one toolchain it is set as the default toolchain.

Toolchain repository name

A toolchain’s repository name uses the format python_{major}_{minor}, e.g. python_3_10. The major and minor components are major and minor are the Python version from the python_version attribute.

Attributes:

  • python_version – (str)

    The Python version, in major.minor format, e.g ‘3.12’, to create a toolchain for. Patch level granularity (e.g. ‘3.12.1’) is not supported.

    mandatory

  • configure_coverage_tool – (bool) (default False)

    Whether or not to configure the default coverage tool for the toolchains.

    optional

  • ignore_root_user_error – (bool) (default False)

    If False, the Python runtime installation will be made read only. This improves the ability for Bazel to cache it, but prevents the interpreter from creating pyc files for the standard library dynamically at runtime as they are loaded.

    If True, the Python runtime installation is read-write. This allows the interpreter to create pyc files for the standard library, but, because they are created as needed, it adversely affects Bazel’s ability to cache the runtime and can result in spurious build failures.

    optional

  • is_default – (bool) (default False)

    Whether the toolchain is the default version

    optional