[英]Proper way to clear an asyncio queue in python3?
Errors in my producer/consumer coroutines may leave items and unfinished tasks in the asyncio queue.我的生产者/消费者协程中的错误可能会在 asyncio 队列中留下项目和未完成的任务。 If an error is encountered, I want to simply stop the loop, cancel the pending tasks and clear the queue.
如果遇到错误,我只想停止循环,取消挂起的任务并清除队列。 While I can finish the first two things, I cannot find an easy way to clear the queue.
虽然我可以完成前两件事,但我找不到一种简单的方法来清除队列。 After reading this answer , I come up with three methods:
看完这个答案后,我想出了三种方法:
import asyncio
q=asyncio.Queue()
for i in range(5):
q.put_nowait(i)
q.get_nowait()
loop=asyncio.get_event_loop()
#this will raise an error if q cannot join
loop.run_until_complete(asyncio.wait_for(q.join(),1))
#method 1
q._queue.clear()
q._finished.set()
q._unfinished_tasks = 0
#method 2
for _ in range(q.qsize()):
q.get_nowait()
for _ in range(q._unfinished_tasks):
q.task_done()
#method 3
del q
q=asyncio.Queue()
So which one is better?那么哪个更好呢?
Avoid using the "private" methods.避免使用“私有”方法。 From the documentation of
Queue.task_done
:从
Queue.task_done
的文档:
For each get() used to fetch a task, a subsequent call to task_done() tells the queue that the processing on the task is complete.
对于用于获取任务的每个 get(),对 task_done() 的后续调用告诉队列该任务的处理已完成。
def empty_queue(q: asyncio.Queue):
for _ in range(q.qsize()):
# Depending on your program, you may want to
# catch QueueEmpty
q.get_nowait()
q.task_done()
I did this in various DBussy example scripts :我在各种DBussy 示例脚本中做到了这一点:
for task in asyncio.Task.all_tasks(loop) :
task.cancel()
try :
loop.run_until_complete(task)
except asyncio.CancelledError :
pass
#end try
#end for
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.