线程池的简单实现

池：存任务的空间，存入多个线程就叫线程池。（每个线程开启关闭耗费资源，线程池统一管理，线程可重复使用）

重复利用的线程池，代码实现：

# -*- coding: utf-8 -*-
""" 重复利用的线程池 """
from threading import Thread, current_thread
from queue import Queue
import time


class MyThreadPool:
    def __init__(self, n):
        self.queue = Queue()
        # 生成线程池
        for i in range(n):
            Thread(target=self.run, args=(self.queue, ), daemon=True).start()

    def run(self, queue):
        while True:
            task, args, kwargs = self.queue.get()
            task(*args, **kwargs)
            print(current_thread(), task) # 打印当前线程
            self.queue.task_done() # 计数器 -1

    def join(self):
        # queue计数为0时终止程序
        self.queue.join() # 重写thread join方法，调用queue join

    def apply_async(self, func, args=(), kwargs={}):
        self.queue.put((func, args, kwargs)) # 计数器 +1


def apply1(n, m):
    # print('apply1: n+m=', n+m)
    time.sleep(3)

def apply2(n, m):
    # print('apply2: n+m=', n+m)
    time.sleep(3)


if __name__ == '__main__':
    my_thread_pool = MyThreadPool(3)
    my_thread_pool.apply_async(apply1, args=(1, 1))
    my_thread_pool.apply_async(apply2, args=(2, 2))
    my_thread_pool.apply_async(apply1, args=(1, 1))
    my_thread_pool.apply_async(apply2, args=(2, 2))
    my_thread_pool.apply_async(apply1, args=(1, 1))
    my_thread_pool.apply_async(apply2, args=(2, 2))
    # 主线程先执行到这儿，此时queue计数器 +1 *6
    my_thread_pool.join() # 等待计数器为0时结束队列queue阻塞程序