Python线程同步

线程执行

join与setDaemon

子线程在主线程运行结束后，会继续执行完，如果给子线程设置为守护线程(setDaemon=True)，主线程运行结束子线程即结束；

如果join()线程，那么主线程会等待子线程执行完再执行。

import threading
import time


def get_thread_a():
    print("get thread A started")
    time.sleep(3)
    print("get thread A end")


def get_thread_b():
    print("get thread B started")
    time.sleep(5)
    print("get thread B end")


if  __name__ == "__main__":
    thread_a = threading.Thread(target=get_thread_a)
    thread_b = threading.Thread(target=get_thread_b)
    start_time = time.time()
    thread_b.setDaemon(True)
    thread_a.start()
    thread_b.start()
    thread_a.join()
    
    end_time = time.time()
    print("execution time: {}".format(end_time - start_time))

thread_a是join，首先子线程thread_a执行，thread_b是守护线程，当主线程执行完后，thread_b不会再执行
执行结果如下：

get thread A started
get thread B started
get thread A end
execution time: 3.003199815750122

线程同步

当线程间共享全局变量，多个线程对该变量执行不同的操作时，该变量最终的结果可能是不确定的(每次线程执行后的结果不同)，如：对count变量执行加减操作 ，count的值是不确定的，要想count的值是一个确定的需对线程执行的代码段加锁。

python对线程加锁主要有Lock和Rlock模块

Lock：

from threading import Lock

lock = Lock()
lock.acquire()
lock.release()

Lock有acquire()和release()方法，这两个方法必须是成对出现的，acquire()后面必须release()后才能再acquire()，否则会造成死锁

Rlock：

鉴于Lock可能会造成死锁的情况，RLock(可重入锁)对Lock进行了改进，RLock可以在同一个线程里面连续调用多次acquire()，但必须再执行相同次数的release()

from threading import RLock

lock = RLock()
lock.acquire()
lock.acquire()
lock.release()
lock.release()

condition(条件变量)，线程在执行时，当满足了特定的条件后，才可以访问相关的数据

import threading


def get_thread_a(condition):
    with condition:
        condition.wait()
        print("A : Hello B,that's ok")
        condition.notify()
        condition.wait()
        print("A : I'm fine,and you?")
        condition.notify()
        condition.wait()
        print("A : Nice to meet you")
        condition.notify()
        condition.wait()
        print("A : That's all for today")
        condition.notify()


def get_thread_b(condition):
    with condition:
        print("B : Hi A, Let's start the conversation")
        condition.notify()
        condition.wait()
        print("B : How are you")
        condition.notify()
        condition.wait()
        print("B : I'm fine too")
        condition.notify()
        condition.wait()
        print("B : Nice to meet you,too")
        condition.notify()
        condition.wait()
        print("B : Oh,goodbye")


if __name__ == "__main__":
    condition = threading.Condition()
    thread_a = threading.Thread(target=get_thread_a, args=(condition,))
    thread_b = threading.Thread(target=get_thread_b, args=(condition,))
    thread_a.start()
    thread_b.start()

Condition内部有一把锁，默认是RLock，在调用wait()和notify()之前必须先调用acquire()获取这个锁，才能继续执行；当wait()和notify()执行完后，需调用release()释放这个锁，在执行with condition时，会先执行acquire()，with结束时，执行了release()；所以condition有两层锁，最底层锁在调用wait()时会释放，同时会加一把锁到等待队列，等待notify()唤醒释放锁

wait() ：允许等待某个条件变量的通知，notify()可唤醒

notify()： 唤醒等待队列wait()

执行结果：

B : Hi A, Let's start the conversation
A : Hello B,that's ok
B : How are you
A : I'm fine,and you?
B : I'm fine too
A : Nice to meet you
B : Nice to meet you,too
A : That's all for today
B : Oh,goodbye

Semaphore(信号量)

 用于控制线程的并发数，如爬虫中请求次数过于频繁会被禁止ip，每次控制爬取网页的线程数量可在一定程度上防止ip被禁；文件读写中，控制写线程每次只有一个，读线程可多个。

import time
import threading


def get_thread_a(semaphore,i):
    time.sleep(1)
    print("get thread : {}".format(i))
    semaphore.release()


def get_thread_b(semaphore):
    for i in range(10):
        semaphore.acquire()
        thread_a = threading.Thread(target=get_thread_a, args=(semaphore,i))
        thread_a.start()


if __name__ == "__main__":
    semaphore = threading.Semaphore(2)
    thread_b = threading.Thread(target=get_thread_b, args=(semaphore,))
    thread_b.start()

上述示例了每隔1秒并发两个线程执行的情况，当调用一次semaphore.acquire()时，Semaphore的数量就减1，直至Semaphore数量为0时被锁上，当release()后Semaphore数量加1。Semaphore在本质上是调用的Condition，semaphore.acquire()在Semaphore的值为0的条件下会调用Condition.wait(), 否则将值减1，semaphore.release()会将Semaphore的值加1，并调用Condition.notify()

Semaphore源码

def acquire(self, blocking=True, timeout=None):
        if not blocking and timeout is not None:
            raise ValueError("can't specify timeout for non-blocking acquire")
        rc = False
        endtime = None
        with self._cond:
            while self._value == 0:
                if not blocking:
                    break
                if timeout is not None:
                    if endtime is None:
                        endtime = _time() + timeout
                    else:
                        timeout = endtime - _time()
                        if timeout <= 0:
                            break
                self._cond.wait(timeout)
            else:
                self._value -= 1
                rc = True
        return rc

def release(self):
        with self._cond:
            self._value += 1
            self._cond.notify()