python多线程中join()的理解

在 Python 的多线程编程中，经常碰到 thread.join()这样的代码。那么今天咱们用实际代码来解释一下 join 函数的作用。

第一，当一个进程启动之后，会默认产生一个主线程，因为线程是程序执行流的最小单元，当设置多线程时，主线程会创建多个子线程，在python中，默认情况下（其实就是setDaemon(False)），主线程执行完自己的任务以后，就退出了，此时子线程会继续执行自己的任务，直到自己的任务结束，例子见下面。

# coding: utf-8
# 测试多线程中join的功能
import threading, time
def doWaiting():
    print 'start waiting1: ' + time.strftime('%H:%M:%S') + "
"
    time.sleep(3)
    print 'stop waiting1: ' + time.strftime('%H:%M:%S') + "
"
def doWaiting1():
    print 'start waiting2: ' + time.strftime('%H:%M:%S') + "
"
    time.sleep(8)
    print 'stop waiting2: ', time.strftime('%H:%M:%S') + "
"
tsk = []
thread1 = threading.Thread(target = doWaiting)
thread1.start()
tsk.append(thread1)
thread2 = threading.Thread(target = doWaiting1)
thread2.start()
tsk.append(thread2)
print 'start join: ' + time.strftime('%H:%M:%S') + "
"

print 'end join: ' + time.strftime('%H:%M:%S') + "
"

代码执行结果如下：

start waiting1: 22:24:20

start join: 22:24:20
start waiting2: 22:24:20


end join: 22:24:20

stop waiting1: 22:24:23

stop waiting2:  22:24:28

第二，当我们使用setDaemon(True)方法，设置子线程为守护线程时，主线程一旦执行结束，则全部线程全部被终止执行，可能出现的情况就是，子线程的任务还没有完全执行结束，就被迫停止，例子见下面。

# coding: utf-8
# 测试多线程中join的功能
import threading, time
def doWaiting():
    print 'start waiting1: ' + time.strftime('%H:%M:%S') + "
"
    time.sleep(3)
    print 'stop waiting1: ' + time.strftime('%H:%M:%S') + "
"
def doWaiting1():
    print 'start waiting2: ' + time.strftime('%H:%M:%S') + "
"
    time.sleep(8)
    print 'stop waiting2: ', time.strftime('%H:%M:%S') + "
"
tsk = []
thread1 = threading.Thread(target = doWaiting)
thread1.setDaemon(True)
thread1.start()
tsk.append(thread1)
thread2 = threading.Thread(target = doWaiting1)
thread2.setDaemon(True)
thread2.start()
tsk.append(thread2)
print 'start join: ' + time.strftime('%H:%M:%S') + "
"

print 'end join: ' + time.strftime('%H:%M:%S') + "
"

代码执行结果如下：

start waiting1: 22:34:46

start waiting2: 22:34:46
start join: 22:34:46


end join: 22:34:46

第三，没有设置守护线程且没有设置join函数的timeout参数时，主线程将会一直等待，直到子线程全部结束，主线程才结束，程序退出。代码如下：

# coding: utf-8
# 测试多线程中join的功能
import threading, time
def doWaiting():
    print 'start waiting1: ' + time.strftime('%H:%M:%S') + "
"
    time.sleep(3)
    print 'stop waiting1: ' + time.strftime('%H:%M:%S') + "
"
def doWaiting1():
    print 'start waiting2: ' + time.strftime('%H:%M:%S') + "
"
    time.sleep(8)
    print 'stop waiting2: ', time.strftime('%H:%M:%S') + "
"
tsk = []
thread1 = threading.Thread(target = doWaiting)
thread1.start()
tsk.append(thread1)
thread2 = threading.Thread(target = doWaiting1)
thread2.start()
tsk.append(thread2)
print 'start join: ' + time.strftime('%H:%M:%S') + "
"
for tt in tsk:
    tt.join()
print 'end join: ' + time.strftime('%H:%M:%S') + "
"

代码执行结果如下：

start waiting1: 22:41:50

start join: 22:41:50
start waiting2: 22:41:50


stop waiting1: 22:41:53

stop waiting2:  22:41:58

end join: 22:41:58

第四，当没有设置守护线程且join函数的参数timeout=2时，主线程将会等待多个子线程timeout的累加和这样的一段时间，时间一到，主线程结束，但是并没有杀死子线程，子线程依然可以继续执行，直到子线程全部结束，程序退出。代码如下：

# coding: utf-8
# 测试多线程中join的功能
import threading, time
def doWaiting():
    print 'start waiting1: ' + time.strftime('%H:%M:%S') + "
"
    time.sleep(3)
    print 'stop waiting1: ' + time.strftime('%H:%M:%S') + "
"
def doWaiting1():
    print 'start waiting2: ' + time.strftime('%H:%M:%S') + "
"
    time.sleep(8)
    print 'stop waiting2: ', time.strftime('%H:%M:%S') + "
"
tsk = []
thread1 = threading.Thread(target = doWaiting)
thread1.start()
tsk.append(thread1)
thread2 = threading.Thread(target = doWaiting1)
thread2.start()
tsk.append(thread2)
print 'start join: ' + time.strftime('%H:%M:%S') + "
"
for tt in tsk:
    tt.join(2)
print 'end join: ' + time.strftime('%H:%M:%S') + "
"

代码执行结果如下：

start waiting1: 23:02:34

start waiting2: 23:02:34
start join: 23:02:34


stop waiting1: 23:02:37

end join: 23:02:38

stop waiting2:  23:02:42

第五，当设置守护线程join函数的参数timeout=2时，主线程将会等待多个子线程timeout的累加和这样的一段时间，时间一到，主线程结束，杀死未执行完的子线程，程序退出。代码如下：

# coding: utf-8
# 测试多线程中join的功能
import threading, time
def doWaiting():
    print 'start waiting1: ' + time.strftime('%H:%M:%S') + "
"
    time.sleep(3)
    print 'stop waiting1: ' + time.strftime('%H:%M:%S') + "
"
def doWaiting1():
    print 'start waiting2: ' + time.strftime('%H:%M:%S') + "
"
    time.sleep(8)
    print 'stop waiting2: ', time.strftime('%H:%M:%S') + "
"
tsk = []
thread1 = threading.Thread(target = doWaiting)
thread1.setDaemon(True)
thread1.start()
tsk.append(thread1)
thread2 = threading.Thread(target = doWaiting1)
thread2.setDaemon(True)
thread2.start()
tsk.append(thread2)
print 'start join: ' + time.strftime('%H:%M:%S') + "
"
for tt in tsk:
    tt.join(2)
print 'end join: ' + time.strftime('%H:%M:%S') + "
"

代码执行结果如下：

start waiting1: 23:23:57

start waiting2: 23:23:57
start join: 23:23:57


stop waiting1: 23:24:00

end join: 23:24:01