python的asyncio中的Joinable PriorityQueue

Question

根据文档,有几种Queues实现.我感兴趣的是JoinableQueue和PriorityQueue,因为我想拥有一个具有优先级的可连接队列.

似乎我只能获得较低版本中的一个功能.即3.5我可以使用Queue(可连接)和PriorityQueue进行区分,但是在3.5以下的python中有JoinableQueue和PriorityQueue(参见下面的例子).

是否可以将它们组合起来以获得在3.4中获得可连接的PriorityQueue的一般方法？

try:
    # Python 3.4.
    from asyncio import JoinableQueue as Queue # joinable
    from asyncio import PriorityQueue          # I assume this one is not joinable
except ImportError:
    # Python 3.5.
    from asyncio import Queue                  # standard joinable
    from asyncio import PriorityQueue          # I assume this is the one I want

另一种方法是以某种方式影响队列？

Answer 1

由于方式JoinableQueue和PriorityQueue实现方式JoinablePriorityQueue,只要JoinableQueue先列出,就可以通过多重继承获得子类化.

这样做的原因PriorityQueue是非常简单的实现:

class PriorityQueue(Queue):
    """A subclass of Queue; retrieves entries in priority order (lowest first).

    Entries are typically tuples of the form: (priority number, data).
    """

    def _init(self, maxsize):
        self._queue = []

    def _put(self, item, heappush=heapq.heappush):
        heappush(self._queue, item)

    def _get(self, heappop=heapq.heappop):
        return heappop(self._queue)

虽然JoinableQueue更复杂,但它和PriorityQueue实现的唯一方法是_put,并且至关重要的是,它在自己的实现中JoinableQUeue调用,这意味着它将正确地合作.super()._put(..)putPriorityQueue

这是一个证明它有效的例子:

from asyncio import PriorityQueue, JoinableQueue
import asyncio
import random

class JoinablePriorityQueue(JoinableQueue, PriorityQueue):
    pass


@asyncio.coroutine
def consume(q):
    while True:
        a = yield from q.get()
        print("got a {}".format(a))
        if a[1] is None:
            q.task_done()
            return
        asyncio.sleep(1)
        q.task_done()

@asyncio.coroutine
def produce(q):
    for i in range(10):
        yield from q.put((random.randint(0,10), i))
    yield from q.put((100, None)) # Will be last
    asyncio.async(consume(q))
    print("waiting...")
    yield from q.join()
    print("waited")

loop = asyncio.get_event_loop()
q = JoinablePriorityQueue()
loop.run_until_complete(produce(q))

输出:

waiting...
got a (1, 2)
got a (2, 1)
got a (4, 4)
got a (5, 0)
got a (6, 8)
got a (6, 9)
got a (8, 3)
got a (9, 5)
got a (9, 7)
got a (10, 6)
got a (100, None)
waited