这几天学习python多线程的时候，试了几次thread模块和threading模块，发现thread模块非常的不好用。强烈不建议大家使用thread，建议使用threading模块，此模块对thread进行了封装，而且还加入了其他的一些方法，比如同步机制，程序员不用考虑主线程退出去的时候其他子线程也强制退出去，不说了上代码

# *_* coding:utf-8 *_*

import time

import threading

def a():

    print ('now a start running:'),time.ctime()

    time.sleep(5)

    print ('a is ending:'),time.ctime()

def b():

    print ('now b start running:'),time.ctime()

    time.sleep(10)

    print ('b is ending:'),time.ctime()

def test():

    a1 = threading.Thread(target=a,args=()) #实例化线程

    b1 = threading.Thread(target=b, args=())

    p = [a1,b1]

    for i in range(len(p)):  #启动多线程

        p[i].start()

    for i in range(len(p)):  #join()方法等待每一个线程结束

        p[i].join()

test()

执行结果：

now a start running: Wed May 17 09:59:17 2017

now b start running: Wed May 17 09:59:17 2017

a is ending: Wed May 17 09:59:22 2017

b is ending: Wed May 17 09:59:27 2017

[Finished in 10.3s]

方法二，写一个类继承threading，重写run方法，举例：

# *_* coding:utf-8 *_*

import threading as th

from time import *

class test(th.Thread):

    def __init__(self,args):

        super(test,self).__init__()

        self.args = args

    def run(self):

        print ('the %s is running')%th.currentThread().getName()

        sleep(2)

        print self.args

        print ('the %s has done at %s \n')%(th.currentThread().getName(),ctime()) #返回当前线程变量的名字

tread = []

for i in range(4):

    t = test(i)

    tread.append(t)

for p in tread:

    p.setDaemon(True)

    p.start()

for n in tread:

    n.join()

print ('all has done!')

#执行结果

# the Thread-1 is running

# the Thread-2 is running

# the Thread-3 is running

# the Thread-4 is running

# 3

# the Thread-4 has done at Wed May 24 16:48:00 2017 

# 2

# 10

# the Thread-1 has done at Wed May 24 16:48:00 2017

# the Thread-2 has done at Wed May 24 16:48:00 2017 

# the Thread-3 has done at Wed May 24 16:48:00 2017 

# all has done!

# [Finished in 2.4s]