【python，threading】python多线程

使用多线程的方式

1、  函数式：使用threading模块threading.Thread(e.g target name parameters)

import time,threading
def loop():
    print("thread %s is running..." % threading.current_thread().name) 
    n = 0
    while n < 5:
        n += 1
        print("thread %s is running... n = %s" % (threading.current_thread().name,str(n))) 
        time.sleep(1)
    print("thread %s is over..." % threading.current_thread().name) 
    
print("thread %s is running..." % threading.current_thread().name) 

ts = []
for i in range(5):
    t = threading.Thread(target = loop, name = 'loopThread '+ str(i))
    t.start()
    ts.append(t)
for t in ts:
    t.join()
print("thread %s is over..." % threading.current_thread().name) 

多线程的输出：

thread MainThread is running...
thread loopThread 0 is running...
thread loopThread 0 is running... n = 1
thread loopThread 1 is running...
thread loopThread 1 is running... n = 1
thread loopThread 2 is running...
thread loopThread 2 is running... n = 1
thread loopThread 0 is running... n = 2
thread loopThread 1 is running... n = 2
thread loopThread 2 is running... n = 2
thread loopThread 0 is running... n = 3
thread loopThread 1 is running... n = 3
thread loopThread 2 is running... n = 3
thread loopThread 0 is running... n = 4
thread loopThread 1 is running... n = 4
thread loopThread 2 is running... n = 4
thread loopThread 0 is running... n = 5
thread loopThread 1 is running... n = 5
thread loopThread 2 is running... n = 5
thread loopThread 0 is over...
thread loopThread 1 is over...
thread loopThread 2 is over...
thread MainThread is over...

python中得thread的一些机制和C/C++不同：在C/C++中，主线程结束后，其子线程会默认被主线程kill掉。而在python中，主线程结束后，会默认等待子线程结束后，主线程才退出。

python对于thread的管理中有两个函数：join和setDaemon

join：如在一个线程B中调用threada.join()，则threada结束后，线程B才会接着threada.join()往后运行。

setDaemon：主线程A启动了子线程B，调用b.setDaemaon(True)，则主线程结束时，会把子线程B也杀死。【此段内容摘录自junshao90的博客】

2. 使用面向对象方式。创建子类继承自threading.Thread，需overwrite run方法

import time,threading
class threadTest(threading.Thread):    
    def __init__(self,tname):
        threading.Thread.__init__(self)
        self.name = tname    
    def run(self):
        print("thread %s is running..." % threading.current_thread().name) 
        n = 0
        while n < 5:
            n += 1
            print("thread %s is running... n = %s" % (threading.current_thread().name,str(n))) 
            time.sleep(1)
        print("thread %s is over..." % threading.current_thread().name)         
print("thread %s is running..." % threading.current_thread().name)   

for i in range(3):
    t = threadTest('t' + str(i))
    t.start()
    t.join()
print("thread %s is over..." % threading.current_thread().name) 

运行输出：

thread MainThread is running...
thread t0 is running...
thread t0 is running... n = 1
thread t0 is running... n = 2
thread t0 is running... n = 3
thread t0 is running... n = 4
thread t0 is running... n = 5
thread t0 is over...
thread t1 is running...
thread t1 is running... n = 1
thread t1 is running... n = 2
thread t1 is running... n = 3
thread t1 is running... n = 4
thread t1 is running... n = 5
thread t1 is over...
thread t2 is running...
thread t2 is running... n = 1
thread t2 is running... n = 2
thread t2 is running... n = 3
thread t2 is running... n = 4
thread t2 is running... n = 5
thread t2 is over...
thread MainThread is over...

3. lock

 多线程和多进程最大的不同在于，多进程中，同一个变量，各自有一份拷贝存在于每个进程中，互不影响。

 而多线程中，所有变量都由所有线程共享，所以，任何一个变量都可以被任何一个线程修改，因此，线程之间共享数据最大的危险在于多个线程同时改一个变量，把  内容给改乱了。

lock 对象：

acquire()：负责取得一个锁。如果没有线程正持有锁，acquire方法会立刻得到锁。否则，它闲意态等锁被释放。一旦acquire()返回，调用它的线程就持有锁。

release(): 释放锁。如果有其他线程正等待这个锁(通过acquire())，当release()被效用的时候，它们中的一个线程就会

被唤醒

以下内容摘自“廖雪峰的官方网站”

http://www.liaoxuefeng.com/wiki/001374738125095c955c1e6d8bb493182103fac9270762a000/001386832360548a6491f20c62d427287739fcfa5d5be1f000

balance为共享资源，多进程同时执行，一定概率结果为balance != 0[详细描述见原文]

def change_it(n):
    # 先存后取，结果应该为0:
    global balance
    balance = balance + n
    balance = balance - n

使用threading.Lock()

import threading

total = 0
lock = threading.Lock()
def change(n):
    global total
    total += n
    total -= n

def run_thread(n):
    lock.acquire()
    for i in range(100000):
        change(n)
    lock.release()
    
t1 = threading.Thread(target = run_thread, args=(5,))
t2 = threading.Thread(target = run_thread, args=(8,))
t1.start()
t2.start()
t1.join()
t2.join()
print(total)

 4. 其他详细关于对进程的资料可参考

 解决共享资源问题的：条件变量，同步队列

 Vamei的博客Python标准库08 多线程与同步 (threading包)

 片片灵感的博客Python多线程学习