chapter_21_asyncio

Chapter 21 — Asynchronous Programming

Theme: Modern Python concurrency relies on an Event Loop. By explicitly writing await, you cooperatively hand control of the single execution thread back to the orchestrator, eliminating race conditions while scaling to millions of I/O operations.

What You'll Learn

Concept	Explanation
Event Loop	The single-threaded engine that schedules and runs asynchronous tasks.
Coroutine	A function defined with async def. It does not execute until awaited.
await	The keyword that pauses a coroutine and yields control back to the loop.
asyncio.gather	Executes multiple awaitables concurrently, preserving order.
asyncio.Semaphore	A critical lock used to limit how many coroutines can access a resource at once.

Key Files

• examples.py — Native coroutines, gather, and responsive as_completed loops.
• exercises.py — The fatal Blocking Trap (proving that time.sleep freezes the entire application).
• mini_project.py — A high-speed web scraper architected with asyncio.Semaphore to throttle concurrency and prevent "Too Many Open Files" OS crashes.
• benchmarks.py — Proves that asyncio.to_thread() is strictly required when running legacy synchronous code (like requests.get()) inside an async application.
• notes.md — Clarifying the terminology: Awaitables, Coroutines, and Tasks.
• pitfalls.md — Forgetting to await, missing semaphores, and blocking the loop.
• interview_questions.md — L3 to L6 interview prep.
• architecture_notes.md — How the Event Loop relies on the OS selectors module (epoll/kqueue).

30-Second Rules

# Rule 1: NEVER use blocking synchronous libraries (like requests) in async code.
# Use aiohttp or httpx.AsyncClient instead.
async def get_data():
    requests.get("...") # FATAL! Freezes the whole server.

# Rule 2: If you MUST use a blocking library, wrap it in to_thread.
async def get_data_safe():
    await asyncio.to_thread(requests.get, "...") # Safe!

# Rule 3: Always throttle large async bursts with a Semaphore.
sem = asyncio.Semaphore(100) # Max 100 concurrent requests
async def limited_fetch(url):
    async with sem:
        await fetch(url)

Reference: Fluent Python 2nd ed., Chapter 21 — pages 803–852