Python Async & Concurrency

When to use this skill

• Writing high-concurrency I/O bound applications (FastAPI, bots, scrapers).
• Using asyncio, aiohttp, or httpx.
• Managing background tasks or task queues.

1. Asyncio Basics

• Entry Point: Use asyncio.run(main()) for the entry point.
• TaskGroups: Use async with asyncio.TaskGroup() as tg (Python 3.11+) instead of gather() for better error handling/cancellation.
• Blocking Code: Run blocking (CPU-bound) code in await asyncio.to_thread(...).

2. HTTP Clients

• HTPX/AIOHTTP: Use httpx.AsyncClient or aiohttp.ClientSession as a context manager.
• Singleton: Reuse the client session across requests; do not create a new one per request.

3. Pitfalls

• Sync in Async: Never call blocking sync functions (requests, time.sleep) inside async functions.
• Fire and Forget: Avoid unreferenced background tasks; keep a reference to prevent garbage collection (or use TaskGroup).