concurrent futures处理并发

油腻的胖子

于 2023-04-20 17:30:10 发布

阅读量119

点赞数

文章标签： python

本文链接：https://blog.csdn.net/u010575789/article/details/130271824

版权

文章展示了如何在Python中利用`concurrent.futures`模块的`ThreadPoolExecutor`进行并发任务处理。通过`map`方法和`submit`配合`as_completed`，分别演示了保持原顺序的结果收集与不按提交顺序获取结果的场景。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

from concurrent import futures
import time


def test(n):
    time.sleep(7-n)
    return n


# map方法
with futures.ThreadPoolExecutor(3) as pool:
    result = pool.map(test, range(3))

print(list(result))   # 输出[0, 1, 2] 这个函数返回结果的顺序与调用开始的顺序一致


# as_completed 和 summit方法
with futures.ThreadPoolExecutor(3) as pool:
    to_do = []
    for i in range(4,7):
        future = pool.submit(test, i)
        to_do.append(future)
    
    result = []
    for future in futures.as_completed(to_do):
        result.append(future.result())

print(list(result))  # 输出[6, 5, 4] 不管提交的顺序,只要有结果就获取


with futures.ThreadPoolExecutor(3) as pool:
    to_do_map = {}  # 构建一个字典,把各个期物映射到其他数据
    for i in range(4,7):
        future = pool.submit(test, i)
        to_do_map[future] = i
    
    result = []
    for future in futures.as_completed(to_do_map):
        result.append(future.result())
        
print(list(result))