Source Distribution

Maturin supports building through pyproject.toml. To use it, create a pyproject.toml next to your Cargo.toml with the following content:

requires = ["maturin>=1.0,<2.0"]
build-backend = "maturin"

If a pyproject.toml with a [build-system] entry is present, maturin can build a source distribution of your package when --sdist is specified. The source distribution will contain the same files as cargo package. To only build a source distribution, use the maturin sdist command.

You can then e.g. install your package with pip install .. With pip install . -v you can see the output of cargo and maturin.

You can use the options compatibility, skip-auditwheel, bindings, strip and common Cargo build options such as features under [tool.maturin] the same way you would when running maturin directly. The bindings key is required for cffi and bin projects as those can't be automatically detected. Currently, all builds are in release mode (see this thread for details).

For a non-manylinux build with cffi bindings you could use the following:

requires = ["maturin>=1.0,<2.0"]
build-backend = "maturin"

bindings = "cffi"
compatibility = "linux"

manylinux option is also accepted as an alias of compatibility for backward compatibility with old version of maturin.

To include arbitrary files in the sdist for use during compilation specify include as an array of path globs with format set to sdist:

include = [{ path = "path/**/*", format = "sdist" }]

Build Wheels

For portability reasons, native python modules on linux must only dynamically link a set of very few libraries which are installed basically everywhere, hence the name manylinux. The pypa offers special docker images and a tool called auditwheel to ensure compliance with the manylinux rules). If you want to publish widely usable wheels for linux pypi, you need to use a manylinux docker image or build with zig.

The Rust compiler since version 1.64 requires at least glibc 2.17, so you need to use at least manylinux2014. For publishing, we recommend enforcing the same manylinux version as the image with the manylinux flag, e.g. use --manylinux 2014 if you are building in quay.io/pypa/manylinux2014_x86_64. The PyO3/maturin-action github action already takes care of this if you set e.g. manylinux: 2014.

maturin contains a reimplementation of auditwheel automatically checks the generated library and gives the wheel the proper platform tag.

  • If your system's glibc is too new, it will assign the linux tag.
  • If you link other shared libraries, maturin will try to bundle them within the wheel, note that this requires patchelf, it can be installed along with maturin from PyPI: pip install maturin[patchelf].

You can also manually disable those checks and directly use native linux target with --manylinux off.

For full manylinux compliance you need to compile in a CentOS docker container. The pyo3/maturin image is based on the manylinux2014 image, and passes arguments to the maturin binary. You can use it like this:

docker run --rm -v $(pwd):/io ghcr.io/pyo3/maturin build --release  # or other maturin arguments

Note that this image is very basic and only contains python, maturin and stable Rust. If you need additional tools, you can run commands inside the manylinux container. See konstin/complex-manylinux-maturin-docker for a small educational example or nanoporetech/fast-ctc-decode for a real world setup.

Usage: maturin build [OPTIONS] [ARGS]...

          Rustc flags

          Strip the library for minimum file size

          Build a source distribution

      --compatibility [<compatibility>...]
          Control the platform tag on linux.

          Options are `manylinux` tags (for example `manylinux2014`/`manylinux_2_24`) or `musllinux` tags (for example `musllinux_1_2`) and `linux` for the native linux tag.

          Note that `manylinux1` and `manylinux2010` is unsupported by the rust compiler. Wheels with the native `linux` tag will be rejected by pypi, unless they are separately validated by `auditwheel`.

          The default is the lowest compatible `manylinux` tag, or plain `linux` if nothing matched

          This option is ignored on all non-linux platforms

  -i, --interpreter [<INTERPRETER>...]
          The python versions to build wheels for, given as the executables of interpreters such as `python3.9` or `/usr/bin/python3.8`

  -f, --find-interpreter
          Find interpreters from the host machine

  -b, --bindings <BINDINGS>
          Which kind of bindings to use

          [possible values: pyo3, pyo3-ffi, cffi, uniffi, bin]

  -o, --out <OUT>
          The directory to store the built wheels in. Defaults to a new "wheels" directory in the project's target directory

          Don't check for manylinux compliance

          For manylinux targets, use zig to ensure compliance for the chosen manylinux version

          Default to manylinux2014/manylinux_2_17 if you do not specify an `--compatibility`

          Make sure you installed zig with `pip install maturin[zig]`

  -q, --quiet
          Do not print cargo log messages

          Ignore `rust-version` specification in packages

  -v, --verbose...
          Use verbose output (-vv very verbose/build.rs output)

      --color <WHEN>
          Coloring: auto, always, never

      --config <KEY=VALUE>
          Override a configuration value (unstable)

  -Z <FLAG>
          Unstable (nightly-only) flags to Cargo, see 'cargo -Z help' for details

          Outputs a future incompatibility report at the end of the build (unstable)

  -h, --help
          Print help (see a summary with '-h')

Compilation Options:
  -r, --release
          Build artifacts in release mode, with optimizations

  -j, --jobs <N>
          Number of parallel jobs, defaults to # of CPUs

      --profile <PROFILE-NAME>
          Build artifacts with the specified Cargo profile

      --target <TRIPLE>
          Build for the target triple

          [env: CARGO_BUILD_TARGET=]

      --target-dir <DIRECTORY>
          Directory for all generated artifacts

          Timing output formats (unstable) (comma separated): html, json

Feature Selection:
  -F, --features <FEATURES>
          Space or comma separated list of features to activate

          Activate all available features

          Do not activate the `default` feature

Manifest Options:
  -m, --manifest-path <PATH>
          Path to Cargo.toml

          Require Cargo.lock and cache are up to date

          Require Cargo.lock is up to date

          Run without accessing the network

Cross Compiling

Maturin has decent cross compilation support for pyo3 and bin bindings, other kind of bindings may work but aren't tested regularly.

Cross-compile to Linux/macOS

Use Docker

For manylinux support the manylinux-cross docker images can be used. And maturin-action makes it easy to do cross compilation on GitHub Actions.

Use Zig

Since v0.12.7 maturin added support for linking with zig cc, compile for Linux works and is regularly tested on CI, other platforms may also work but aren't tested regularly.

You can install zig following the official documentation, or install it from PyPI via pip install ziglang. Then pass --zig to maturin build or publish commands to use it, for example

maturin build --release --target aarch64-unknown-linux-gnu --zig

Cross-compile to Windows

Pyo3 0.16.5 added an experimental feature generate-import-lib enables the user to cross compile extension modules for Windows targets without setting the PYO3_CROSS_LIB_DIR environment variable or providing any Windows Python library files.

pyo3 = { version = "0.21.0", features = ["extension-module", "generate-import-lib"] }

It uses an external python3-dll-a crate to generate import libraries for the Python DLL for MinGW-w64 and MSVC compile targets. Note: MSVC targets require LLVM binutils or MSVC build tools to be available on the host system. More specifically, python3-dll-a requires llvm-dlltool or lib.exe executable to be present in PATH when targeting *-pc-windows-msvc.

maturin integrates cargo-xwin to enable MSVC targets cross compilation support, it will download and unpack the Microsoft CRT headers and import libraries, and Windows SDK headers and import libraries needed for compiling and linking automatically.

By using this to cross compiling to Windows MSVC targets you are consented to accept the license at https://go.microsoft.com/fwlink/?LinkId=2086102. (Building on Windows natively does not apply.)

GitHub Actions

If your project uses GitHub Actions, you can use the maturin generate-ci command to generate a GitHub Actions workflow file.

mkdir -p .github/workflows
maturin generate-ci github > .github/workflows/CI.yml

There are some options to customize the generated workflow file:

Generate CI configuration

Usage: maturin generate-ci [OPTIONS] <CI>

          CI provider

          Possible values:
          - github: GitHub

  -m, --manifest-path <PATH>
          Path to Cargo.toml

  -o, --output <PATH>
          Output path

          [default: -]

      --platform <platform>...
          Platform support

          [default: linux windows macos]

          Possible values:
          - all:        All
          - linux:      Linux
          - windows:    Windows
          - macos:      macOS
          - macosarm64: macOS(Arm64)
          - emscripten: Emscripten

          Enable pytest

          Use zig to do cross compilation

  -h, --help
          Print help information (use `-h` for a summary)

Using PyPI's trusted publishing

By default, the workflow provided by generate-ci will publish the release artifacts to PyPI using API token authentication. However, maturin also supports trusted publishing (OpenID Connect).

To enable it, modify the release action in the generated GitHub workflow file:

  • remove MATURIN_PYPI_TOKEN from the env section to make maturin use trusted publishing
  • add id-token: write to the action's permissions (see Configuring OpenID Connect in PyPI from GitHub's documentation).
  • if Environment name: release was set in PyPI, add environment: release

Make sure to follow the steps listed in PyPI's documentation to set up your GitHub repository as a trusted publisher in the PyPI project settings before attempting to run the workflow.