PyCon APAC 2025

Prerequisite

Please note that in this workshop, besides assuming that attendees have experience programming async in Python, attendees are required to have basic knowledge of Rust. Attendees are highly advised to go over the Rust Book to make sure they understand coding in Rust. Attendees are also advised to be already familiar with PyO3. If they are new to PyO3, first doing the PyO3 101 workshop would be ideal.

Preflight checklist

• Install/ Update Rust
• Make sure having Python 3.13, both the "normal" version (with GIL) and free-threaded Python)
• Make sure using virtual environment (recommend using uv)

Windows checklist

In this workshop we recommend using Unix OS (Mac or Linux). If you use Windows, you may encounter problems with Rust and Maturin. To minimise issues that you may encounter, please go through the extra checklist below:

• Install the c++ build tools
• Check the dll files are linked correctly

Setting up

1. Create a new working directory

mkdir pyo3_202
cd pyo3_202

1. Set up virtual environment and install maturin

uv venv .venv
source .venv/bin/activate
uv pip install maturin
python -m ensurepip --default-pip

Note: the last command is needed as maturin develop cannot find pip otherwise

Target audience

Developers who would like to create powerful Python libraries that can support async in Python and speed up tasks with multithreading.

Goal

Deepend attendees' knowledge in PyO3, and strengthen understanding of parallelism in Python and Rust.

Outline

Part 1 - introduction and revision of PyO3 (30 mins)
- a recap of what we learned in PyO3 101
- about async and multithreading in Python and in Rust

15 mins break + Q&A

Part 2 - Support async and await in Python (60 mins)
- creating Python async functions with PyO3 (some coding exercises)
- troubleshooting for async (some coding exercises)

15 mins break + Q&A

Part 3 - Multithreading (90 mins)
- introduction of multithreading in Rust (some coding exercises)
- multithreading in PyO3 (some coding exercises)
- Free-threaded CPython and benchmarking (some coding exercises)
- Thread safety (discussion)