Async CMs, async generators, bounded queues

Async context managers, iterators, and backpressure

Async context managers (async with) allow awaits inside __aenter__ / __aexit__ — connections, locks, transactions:

class AsyncConnection:
    async def __aenter__(self):
        await self._connect()
        return self
    async def __aexit__(self, exc_type, exc, tb):
        await self._close()

contextlib.asynccontextmanager turns a generator into one, exactly like its sync sibling.

Async iterators (async for) pull items across awaits; async generators (async def with yield) produce them:

import asyncio

async def ticker(n, delay=0.01):
    for i in range(n):
        await asyncio.sleep(delay)
        yield i

async def consume():
    async for value in ticker(3):
        print(value)

Bounded queues are async backpressure made visible: queue = asyncio.Queue(maxsize=10). await queue.put(x) suspends the producer when full — the producer slows to the consumer's pace instead of growing memory without limit. await queue.get() suspends the consumer when empty. Workers + queue.join() + task_done() work exactly like the threaded version from the previous module, but on the loop.