Async iteration

Streaming results as they arrive

Sometimes you cannot fetch everything up front — the pages are unbounded, or you want to start work on the first result before the last arrives. That is what async for is for.

An async iterator defines __aiter__ and __anext__, and signals the end by raising StopAsyncIteration:

class Pages:
    def __init__(self, last):
        self.page = 0
        self.last = last

    def __aiter__(self):
        return self

    async def __anext__(self):
        if self.page >= self.last:
            raise StopAsyncIteration
        self.page += 1
        await asyncio.sleep(0)
        return self.page

async def main():
    async for page in Pages(3):
        print(page)

An async generator is the shorter way to write the same thing — an async def with yield in it:

async def pages(last):
    for n in range(1, last + 1):
        await asyncio.sleep(0)
        yield n

Your turn: write an async generator countdown(n) yielding n down to 1, and collect(n) that gathers its values into a list with async for.