Python线程同步的实现代码_Python

本文介绍Python中的线程同步对象，主要涉及 thread 和 threading 模块。

threading 模块提供的线程同步原语包括：Lock、RLock、Condition、Event、Semaphore等对象。

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()