/rules_python

Experimental Bazel Python Rules

Primary LanguagePythonApache License 2.0Apache-2.0

Python Rules for Bazel

  • Postsubmit Build status
  • Postsubmit + Current Bazel Incompatible Flags Build status

Recent updates

  • 2019-11-15: Added support for pip3_import (and more generally, a python_interpreter attribute to pip_import). The canonical naming for wheel repositories has changed to accomodate loading wheels for both pip_import and pip3_import in the same build. To avoid breakage, please use requirement() instead of depending directly on wheel repo labels.

  • 2019-07-26: The canonical name of this repo has been changed from @io_bazel_rules_python to just @rules_python, in accordance with convention. Please update your WORKSPACE file and labels that reference this repo accordingly.

Overview

This repository is the home of the core Python rules -- py_library, py_binary, py_test, and related symbols that provide the basis for Python support in Bazel. It also contains packaging rules for integrating with PyPI (pip). Documentation lives in the docs/ directory and in the Bazel Build Encyclopedia.

Currently the core rules are bundled with Bazel itself, and the symbols in this repository are simple aliases. However, in the future the rules will be migrated to Starlark and debundled from Bazel. Therefore, the future-proof way to depend on Python rules is via this repository. SeeMigrating from the Bundled Rules below.

The core rules are stable. Their implementation in Bazel is subject to Bazel's backward compatibility policy. Once they are fully migrated to rules_python, they may evolve at a different rate, but this repository will still follow semantic versioning.

The packaging rules (pip_import, etc.) are less stable. We may make breaking changes as they evolve. There are no guarantees for rules underneath the experimental/ directory.

See the How to contribute page for information on our devlopment workflow.

Getting started

To import rules_python in your project, you first need to add it to your WORKSPACE file. If you are using the Bazel Federation, you just need to import the Federation and call the rules_python setup methods:

load("@bazel_tools//tools/build_defs/repo:http.bzl", "http_archive")

http_archive(
    name = "bazel_federation",
    url = "https://github.com/bazelbuild/bazel-federation/releases/download/0.0.1/bazel_federation-0.0.1.tar.gz",
    sha256 = "506dfbfd74ade486ac077113f48d16835fdf6e343e1d4741552b450cfc2efb53",
)

load("@bazel_federation//:repositories.bzl", "rules_python_deps")

rules_python_deps()
load("@bazel_federation//setup:rules_python.bzl",  "rules_python_setup")
rules_python_setup(use_pip=True)

Note the use_pip argument: rules_python may be imported either with or without support for the packaging rules.

If you are not using the Federation, you can simply import rules_python directly and call its initialization methods as follows:

load("@bazel_tools//tools/build_defs/repo:http.bzl", "http_archive")
http_archive(
    name = "rules_python",
    url = "https://github.com/bazelbuild/rules_python/releases/download/0.0.1/rules_python-0.0.1.tar.gz",
    sha256 = "aa96a691d3a8177f3215b14b0edc9641787abaaa30363a080165d06ab65e1161",
)
load("@rules_python//python:repositories.bzl", "py_repositories")
py_repositories()
# Only needed if using the packaging rules.
load("@rules_python//python:pip.bzl", "pip_repositories")
pip_repositories()

To depend on a particular unreleased version (not recommended), you can use git_repository instead of http_archive:

load("@bazel_tools//tools/build_defs/repo:git.bzl", "git_repository")

git_repository(
    name = "rules_python",
    remote = "https://github.com/bazelbuild/rules_python.git",
    # NOT VALID: Replace with actual Git commit SHA.
    commit = "{HEAD}",
)

# Then load and call py_repositories() and possibly pip_repositories() as
# above.

Once you've imported the rule set into your WORKSPACE using any of these methods, you can then load the core rules in your BUILD files with:

load("@rules_python//python:defs.bzl", "py_binary")

py_binary(
  name = "main",
  srcs = ["main.py"],
)

Using the packaging rules

The packaging rules create two kinds of repositories: A central repo that holds downloaded wheel files, and individual repos for each wheel's extracted contents. Users only need to interact with the central repo; the wheel repos are essentially an implementation detail. The central repo provides a WORKSPACE macro to create the wheel repos, as well as a function to call in BUILD files to translate a pip package name into the label of a py_library target in the appropriate wheel repo.

Importing pip dependencies

Adding pip dependencies to your WORKSPACE is a two-step process. First you declare the central repository using pip_import, which invokes pip to read a requirements.txt file and download the appropriate wheels. Then you load the pip_install function from the central repo, and call it to create the individual wheel repos.

Important: If you are using Python 3, load and call pip3_import instead.

load("@rules_python//python:pip.bzl", "pip_import")

# Create a central repo that knows about the dependencies needed for
# requirements.txt.
pip_import(   # or pip3_import
   name = "my_deps",
   requirements = "//path/to:requirements.txt",
)

# Load the central repo's install function from its `//:requirements.bzl` file,
# and call it.
load("@my_deps//:requirements.bzl", "pip_install")
pip_install()

Note that since pip is executed at WORKSPACE-evaluation time, Bazel has no information about the Python toolchain and cannot enforce that the interpreter used to invoke pip matches the interpreter used to run py_binary targets. By default, pip_import uses the system command "python", which on most platforms is a Python 2 interpreter. This can be overridden by passing the python_interpreter attribute to pip_import. pip3_import just acts as a wrapper that sets python_interpreter to "python3".

You can have multiple pip_imports in the same workspace, e.g. for Python 2 and Python 3. This will create multiple central repos that have no relation to one another, and may result in downloading the same wheels multiple times.

As with any repository rule, if you would like to ensure that pip_import is reexecuted in order to pick up a non-hermetic change to your environment (e.g., updating your system python interpreter), you can completely flush out your repo cache with bazel clean --expunge.

Consuming pip dependencies

Each extracted wheel repo contains a py_library target representing the wheel's contents. Rather than depend on this target's label directly -- which would require hardcoding the wheel repo's mangled name into your BUILD files -- you should instead use the requirement() function defined in the central repo's //:requirements.bzl file. This function maps a pip package name to a label. ("Extras" can be referenced using the pkg[extra] syntax.)

load("@my_deps//:requirements.bzl", "requirement")

py_library(
    name = "mylib",
    srcs = ["mylib.py"],
    deps = [
        ":myotherlib",
        requirement("some_pip_dep"),
        requirement("anohter_pip_dep[some_extra]"),
    ]
)

For reference, the wheel repos are canonically named following the pattern: @{central_repo_name}_pypi__{distribution}_{version}. Characters in the distribution and version that are illegal in Bazel label names (e.g. -, .) are replaced with _. While this naming pattern doesn't change often, it is not guaranted to remain stable, so use of the requirement() function is recommended.

Migrating from the bundled rules

The core rules are currently available in Bazel as built-in symbols, but this form is deprecated. Instead, you should depend on rules_python in your WORKSPACE file and load the Python rules from @rules_python//python:defs.bzl.

A buildifier fix is available to automatically migrate BUILD and .bzl files to add the appropriate load() statements and rewrite uses of native.py_*.

# Also consider using the -r flag to modify an entire workspace.
buildifier --lint=fix --warnings=native-py <files>

Currently the WORKSPACE file needs to be updated manually as per Getting started above.

Note that Starlark-defined bundled symbols underneath @bazel_tools//tools/python are also deprecated. These are not yet rewritten by buildifier.