主页 / user-516938

Maksim Alekseev's questions

Martin Hope
Maksim Alekseev
Asked: 2024-02-27 19:53:16 +0800 CST
  • 5

我试图timeout在到达方法后取消正在运行的线程concurrent.futures.wait，但程序仍运行 10 秒中指定的时间if url == "url_3":

import concurrent.futures
from time import sleep, time

sources = ["url_1", "url_2", "url_3"]

def get_request_header(url: str) -> str:
    if url == "url_1":
        sleep(0.8)
    if url == "url_2":
        sleep(0.9)
    if url == "url_3":
        sleep(10)
    return f"{url}-headers"


start = time()
with concurrent.futures.ThreadPoolExecutor(max_workers=5) as executor:

    futures = [executor.submit(get_request_header, url) for url in sources]
    completed, uncompleted = concurrent.futures.wait(futures, timeout=1.5)

    for future in uncompleted:
        future.cancel()

for future in completed:
    print(future.result())

print(time() - start)

如果您在cancel()之后打印并检查状态，那么它仍然存在running并且实际上返回False

for future in uncompleted:
    print(future)
    print(future.cancel())
    print(future)

需要通过timeout参数限制线程的工作，让未完成的线程停止工作，这有什么问题吗？

我也尝试过as_completed，但map不提供通过。

python