Question

我已经定义了具有特定队列大小（阻塞）的ThreadPoolExecutor
执行完任务后，它不会退出，因为它在某种情况下会卡住。

from concurrent.futures import ThreadPoolExecutor
import queue


class ThreadPoolExecutorWithQueueSizeLimit(ThreadPoolExecutor):
    ''' 
    Sub-classing ThreadPoolExecutor to implement
    blocking queue feature, not currently available.
    We used queue.Queue module here, which is blocking in nature when maxsize != 0
    Reference: https://stackoverflow.com/questions/48263704/threadpoolexecutor-how-to-limit-the-queue-maxsize
    '''
    def __init__(self, maxsize=0, *args, **kwargs):
        super(ThreadPoolExecutorWithQueueSizeLimit, self).__init__(*args, **kwargs)
        self._work_queue = queue.Queue(maxsize=maxsize)

exe = ThreadPoolExecutorWithQueueSizeLimit(maxsize=2, max_workers=3)

l = []

def f(args):
    a = args[0]
    import time
    time.sleep(2)
    b = args[1]
    print ((a, b), exe._work_queue.qsize())
    return (a, b)
for i in range(5):
    for j in range(4):
        l.append((j, j+1))
    print (list(exe.map(f, l, timeout=1)))
print (exe._work_queue.qsize())
exe.shutdown()
print ("done")

按Command + C（mac）后，它会给出以下跟踪（以及输出）：

Traceback (most recent call last):
  File "test.py", line 30, in <module>
    print (list(exe.map(f, l, timeout=1)))
  File "/export/apps/python/3.6/lib/python3.6/concurrent/futures/_base.py", line 558, in result_iterator
    yield future.result(end_time - time.time())
  File "/export/apps/python/3.6/lib/python3.6/concurrent/futures/_base.py", line 407, in result
    raise TimeoutError()
concurrent.futures._base.TimeoutError
(0, 1) 2
(2, 3) 2
(1, 2) 2
^CError in atexit._run_exitfuncs:
Traceback (most recent call last):
  File "/export/apps/python/3.6/lib/python3.6/queue.py", line 133, in put
    self.not_full.wait()
  File "/export/apps/python/3.6/lib/python3.6/threading.py", line 295, in wait
    waiter.acquire()
KeyboardInterrupt

所以，我想知道为什么程序会卡在acquire（）上，以及由于队列大小在那时也为0来释放它，该怎么做。
谢谢

队列为空后，Python3 ThreadPoolExecutor卡住了

0 个答案: