Python 3.6:islice的异步版本? [英] Python 3.6: async version of islice?

问题描述

我正在尝试做这样的事情:

I'm trying to do something like this:

import asyncio
from itertools import islice

async def generate_numbers(n):
    for x in range(n):
        yield x


async def consume_numbers(n):
    async for x in generate_numbers(n):
        print(x)

async def consume_some_numbers(n,m):
    async for x in islice(generate_numbers(n),m): #<-- This doesn't work.  islice doesn't recognize async iterators as iterators.
        print(x)


loop = asyncio.get_event_loop()
loop.run_until_complete(consume_numbers(10))
loop.run_until_complete(consume_some_numbers(10,5))

有没有办法使这项工作奏效，或者至少获得类似的功能?

Is there a way to make this work, or at least get similar functionality?

推荐答案

此处尝试实现异步友好的

Here is an attempt to implement asyncio friendly islice (and enumerate):

import asyncio
import sys

import random


async def aenumerate(aiterable):
    i = 0
    async for x in aiterable:
        yield i, x
        i += 1


async def aislice(aiterable, *args):
    s = slice(*args)
    it = iter(range(s.start or 0, s.stop or sys.maxsize, s.step or 1))
    try:
        nexti = next(it)
    except StopIteration:
        return
    async for i, element in aenumerate(aiterable):
        if i == nexti:
            yield element
            try:
                nexti = next(it)
            except StopIteration:
                return


async def generate_numbers(n):
    for x in range(n):
        await asyncio.sleep(random.uniform(0.1, 0.4))
        yield x


async def consume_numbers(tag, n):
    print(tag, "start")
    async for x in generate_numbers(n):
        print(tag, x)
    print(tag, "done")


async def consume_some_numbers(tag, n, a, b, step=1):
    print(tag, "start")
    async for x in aislice(generate_numbers(n), a, b, step):
        print(tag, x)
    print(tag, "done")


loop = asyncio.get_event_loop()
loop.run_until_complete(asyncio.wait([
    consume_numbers("A", 5),
    consume_numbers("B", 10),
    consume_some_numbers("C", 10, 0, 5),
    consume_some_numbers("D", 30, 3, 20, 4),
    consume_some_numbers("E", 10, 3, 8, 2),
]))
loop.close()

未经在实际应用中进行了测试，欢迎发表评论:-)

This was not tested in a real world application, comments welcome :-)

这篇关于Python 3.6:islice的异步版本?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！