Crate pyo3

source ·
Expand description

Rust bindings to the Python interpreter.

PyO3 can be used to write native Python modules or run Python code and modules from Rust.

See the guide for a detailed introduction.

§PyO3’s object types

PyO3 has several core types that you should familiarize yourself with:

§The Python<'py> object, and the 'py lifetime

Holding the global interpreter lock (GIL) is modeled with the Python<'py> token. Many Python APIs require that the GIL is held, and PyO3 uses this token as proof that these APIs can be called safely. It can be explicitly acquired and is also implicitly acquired by PyO3 as it wraps Rust functions and structs into Python functions and objects.

The Python<'py> token’s lifetime 'py is common to many PyO3 APIs:

  • Types that also have the 'py lifetime, such as the Bound<'py, T> smart pointer, are bound to the Python GIL and rely on this to offer their functionality. These types often have a .py() method to get the associated Python<'py> token.
  • Functions which depend on the 'py lifetime, such as PyList::new_bound, require a Python<'py> token as an input. Sometimes the token is passed implicitly by taking a Bound<'py, T> or other type which is bound to the 'py lifetime.
  • Traits which depend on the 'py lifetime, such as FromPyObject<'py>, usually have inputs or outputs which depend on the lifetime. Adding the lifetime to the trait allows these inputs and outputs to express their binding to the GIL in the Rust type system.

§Python object smart pointers

PyO3 has two core smart pointers to refer to Python objects, Py<T> and its GIL-bound form Bound<'py, T> which carries the 'py lifetime. (There is also Borrowed<'a, 'py, T>, but it is used much more rarely).

The type parameter T in these smart pointers can be filled by:

  • PyAny, e.g. Py<PyAny> or Bound<'py, PyAny>, where the Python object type is not known. Py<PyAny> is so common it has a type alias PyObject.
  • Concrete Python types like PyList or PyTuple.
  • Rust types which are exposed to Python using the #[pyclass] macro.

See the guide for an explanation of the different Python object types.

§PyErr

The vast majority of operations in this library will return PyResult<...>. This is an alias for the type Result<..., PyErr>.

A PyErr represents a Python exception. A PyErr returned to Python code will be raised as a Python exception. Errors from PyO3 itself are also exposed as Python exceptions.

§Feature flags

PyO3 uses feature flags to enable you to opt-in to additional functionality. For a detailed description, see the Features chapter of the guide.

§Default feature flags

The following features are turned on by default:

  • macros: Enables various macros, including all the attribute macros.

§Optional feature flags

The following features customize PyO3’s behavior:

  • abi3: Restricts PyO3’s API to a subset of the full Python API which is guaranteed by PEP 384 to be forward-compatible with future Python versions.
  • auto-initialize: Changes Python::with_gil to automatically initialize the Python interpreter if needed.
  • extension-module: This will tell the linker to keep the Python symbols unresolved, so that your module can also be used with statically linked Python interpreters. Use this feature when building an extension module.
  • multiple-pymethods: Enables the use of multiple #[pymethods] blocks per #[pyclass]. This adds a dependency on the inventory crate, which is not supported on all platforms.

The following features enable interactions with other crates in the Rust ecosystem:

§Unstable features

  • nightly: Uses #![feature(auto_traits, negative_impls)] to define Ungil as an auto trait.

§rustc environment flags

PyO3 uses rustc’s --cfg flags to enable or disable code used for different Python versions. If you want to do this for your own crate, you can do so with the pyo3-build-config crate.

  • Py_3_7, Py_3_8, Py_3_9, Py_3_10: Marks code that is only enabled when compiling for a given minimum Python version.
  • Py_LIMITED_API: Marks code enabled when the abi3 feature flag is enabled.
  • PyPy - Marks code enabled when compiling for PyPy.

§Minimum supported Rust and Python versions

PyO3 supports the following software versions:

  • Python 3.7 and up (CPython and PyPy)
  • Rust 1.56 and up

§Example: Building a native Python module

PyO3 can be used to generate a native Python module. The easiest way to try this out for the first time is to use maturin. maturin is a tool for building and publishing Rust-based Python packages with minimal configuration. The following steps set up some files for an example Python module, install maturin, and then show how to build and import the Python module.

First, create a new folder (let’s call it string_sum) containing the following two files:

Cargo.toml

[package]
name = "string-sum"
version = "0.1.0"
edition = "2021"

[lib]
name = "string_sum"
# "cdylib" is necessary to produce a shared library for Python to import from.
#
# Downstream Rust code (including code in `bin/`, `examples/`, and `tests/`) will not be able
# to `use string_sum;` unless the "rlib" or "lib" crate type is also included, e.g.:
# crate-type = ["cdylib", "rlib"]
crate-type = ["cdylib"]

[dependencies.pyo3]
version = "0.21.0"
features = ["extension-module"]

src/lib.rs

use pyo3::prelude::*;

/// Formats the sum of two numbers as string.
#[pyfunction]
fn sum_as_string(a: usize, b: usize) -> PyResult<String> {
    Ok((a + b).to_string())
}

/// A Python module implemented in Rust.
#[pymodule]
fn string_sum(m: &Bound<'_, PyModule>) -> PyResult<()> {
    m.add_function(wrap_pyfunction!(sum_as_string, m)?)?;

    Ok(())
}

With those two files in place, now maturin needs to be installed. This can be done using Python’s package manager pip. First, load up a new Python virtualenv, and install maturin into it:

$ cd string_sum
$ python -m venv .env
$ source .env/bin/activate
$ pip install maturin

Now build and execute the module:

$ maturin develop
# lots of progress output as maturin runs the compilation...
$ python
>>> import string_sum
>>> string_sum.sum_as_string(5, 20)
'25'

As well as with maturin, it is possible to build using setuptools-rust or manually. Both offer more flexibility than maturin but require further configuration.

§Example: Using Python from Rust

To embed Python into a Rust binary, you need to ensure that your Python installation contains a shared library. The following steps demonstrate how to ensure this (for Ubuntu), and then give some example code which runs an embedded Python interpreter.

To install the Python shared library on Ubuntu:

sudo apt install python3-dev

Start a new project with cargo new and add pyo3 to the Cargo.toml like this:

[dependencies.pyo3]
version = "0.21.0"
# this is necessary to automatically initialize the Python interpreter
features = ["auto-initialize"]

Example program displaying the value of sys.version and the current user name:

use pyo3::prelude::*;
use pyo3::types::IntoPyDict;

fn main() -> PyResult<()> {
    Python::with_gil(|py| {
        let sys = py.import_bound("sys")?;
        let version: String = sys.getattr("version")?.extract()?;

        let locals = [("os", py.import_bound("os")?)].into_py_dict_bound(py);
        let code = "os.getenv('USER') or os.getenv('USERNAME') or 'Unknown'";
        let user: String = py.eval_bound(code, None, Some(&locals))?.extract()?;

        println!("Hello {}, I'm Python {}", user, version);
        Ok(())
    })
}

The guide has a section with lots of examples about this topic.

§Other Examples

The PyO3 README contains quick-start examples for both using Rust from Python and Python from Rust.

The PyO3 repository’s examples subdirectory contains some basic packages to demonstrate usage of PyO3.

There are many projects using PyO3 - see a list of some at https://github.com/PyO3/pyo3#examples.

Re-exports§

Modules§

  • bufferNon-Py_LIMITED_API or Py_3_11
    PyBuffer implementation
  • Utilities for a Python callable object that invokes a Rust function.
  • Old module which contained some implementation details of the #[pyproto] module.
  • Defines conversions between Rust and Python types.
  • This module contains conversions between various Rust object and their representation in Python.
  • coroutineexperimental-async
    Python coroutine implementation, used notably when wrapping async fn with #[pyfunction]/#[pymethods].
  • Ths module only contains re-exports of pyo3 deprecation warnings and exists purely to make compiler error messages nicer.
  • Functionality for the code generated by the derive backend
  • err 🔒
  • Exception and warning types defined by Python.
  • Raw FFI declarations for Python’s C API.
  • gil 🔒
    Interaction with Python’s global interpreter lock
  • Internals of PyO3 which are accessed by code expanded from PyO3’s procedural macros.
  • inspectexperimental-inspect
    Runtime inspection of objects exposed to Python.
  • instance 🔒
  • macros 🔒 macros
  • Fundamental properties of objects tied to the Python interpreter.
  • marshalNon-Py_LIMITED_API
    Support for the Python marshal format.
  • Helper to convert Rust panics to Python exceptions.
  • PyO3’s prelude.
  • Contains types for working with Python objects that own the underlying data.
  • PyO3’s interior mutability primitive.
  • PyClass and related traits.
  • Contains initialization utilities for #[pyclass].
  • sealed 🔒
  • Synchronization mechanisms based on the Python GIL.
  • Python type object information
  • Various types defined by the Python interpreter such as int, str and tuple.
  • version 🔒

Macros§

Attribute Macros§

  • pyclassmacros
    A proc macro used to expose Rust structs and fieldless enums as Python objects.
  • A proc macro used to expose Rust functions to Python.
  • pymethodsmacros
    A proc macro used to expose methods to Python.
  • pymodulemacros
    A proc macro used to implement Python modules.

Derive Macros§

⚠️ Internal Docs ⚠️ Not Public API 👉 Official Docs Here