Python 在真实条件下退出所有正在运行的线程

Question

我正在使用线程从 API url 检查 header 状态代码。如果条件为真，我如何打破循环/停止所有其他线程。请检查以下代码..

import logging, time, threading, requests

#: Log items
logging.basicConfig(format='%(asctime)s %(levelname)s : %(message)s', level=logging.INFO)

class EppThread(threading.Thread):
    def __init__(self, group=None, target=None, name=None, args=(), kwargs=None, verbose=None):
        threading.Thread.__init__(self, group=group, target=target, name=name, verbose=verbose)
        self.args = args

    def run(self):
        startTime = time.time()
        url = self.args[0]
        limit = self.args[1]

        for i in range(limit):
            response = requests.get(url)
            if response.status_code != 200:
                break
                #Exit other similar threads (with same url)
            else:
                print('Thread {0} - success'.format(thread.getName()))

         print('process completed')
         # Send Email


number_of_threads = 5
number_of_requests = 100

urls = ['https://v1.api.com/example', 'https://v2.api.com/example']

if __name__ == '__main__':
    startTime = time.time()

    for url in urls:
        threads = []
        for i in range(number_of_threads):
            et = EppThread(name = "{0}-Thread-{1}".format(name, i + 1), args=(url, number_of_requests))
            threads.append(et)
            et.start()

    # Check if execution time is not greater than 1 minute  
    while len(threads) > 0 and (time.time() - startTime) < 60:
        time.sleep(0.5)
        for thread in threads:
            if not thread.isAlive():
                threads.remove(thread)
                print('Thread {0} terminated'.format(thread.getName()))

    os._exit(1)

如果条件在任何正在运行的线程中为真，请提出一些停止代码执行的更好方法。

感谢您的帮助。

Answer 1

这里需要注意的一点是，当一个Thread的run方法执行完成后，这个Thread就会被置为dead和garbage集。所以我们真正需要的是一个 boolean 类变量来打破这个循环。类变量对于从该类和子类实例化的所有对象都是相同的；所以一旦我们设置了它，我们类中的所有对象都会以相同的方式运行:

import logging, time, threading, requests

#: Log items
logging.basicConfig(format='%(asctime)s %(levelname)s : %(message)s', level=logging.INFO)


class EppThread(threading.Thread):
    kill = False  # new Boolean class variable
    url = 'https://v1.api.com/example'  # keep this in mind for later

    def __init__(self, group=None, target=None, name=None, args=(), kwargs=None, verbose=None):
        threading.Thread.__init__(self, group=group, target=target, name=name, verbose=verbose)
        self.args = args

    def run(self):
        limit = self.args[0]

        for i in range(limit):
            response = requests.get(self.url)
            if response.status_code != 200:
                self.kill = True  # ends this loop on all Threads since it's changing a class variable

            else:
                print('Thread {0} - success'.format(self.getName()))  # changed to self.getName()

            if self.kill:  # if kill is True, break the loop, send the email, and finish the Thread
                break

        print('process completed')
        # Send Email


number_of_threads = 5
number_of_requests = 100

if __name__ == '__main__':
    startTime = time.time()

    threads = []
    for i in range(number_of_threads):
        et = EppThread(name="{0}-Thread-{1}".format(name, i + 1), args=(number_of_requests))
        threads.append(et)
        et.start()

    # Check if execution time is not greater than 1 minute
    while threads and time.time() - startTime < 60:  # removed len() due to implicit Falsiness of empty containers in Python
        time.sleep(0.5)
        for thread in threads:
            if not thread.isAlive():
                threads.remove(thread)
                print('Thread {0} terminated'.format(thread.getName()))

    EppThread.kill = True

现在，当任何 EppThreads 有一个错误的连接时，它会将类变量设置为 True，这使得所有其他 EppThread也打破循环。我还在末尾添加了 EppThread.kill = True，因此如果运行时间超过 1 分钟，它会更干净地中断请求循环。

最后，我添加了 url 类变量。这是因为您表示有兴趣同时运行不同的 url，并且只终止那些连接特别糟糕的 url。此时您所要做的就是继承 EppThread 并覆盖 kill 和 url。

class EppThread2(EppThread):
    kill = False
    url = 'https://v2.example.com/api?$awesome=True'

然后您可以实例化 EppThread2 并将其添加到 threads 列表中，一切都应该按您希望的方式运行。