python多线程中Lock()与RLock()锁

资源总是有限的,程序运行如果对同一个对象进行操作,则有可能造成资源的争用,甚至导致死锁
也可能导致读写混乱

锁提供如下方法:
1.Lock.acquire([blocking])
2.Lock.release()
3.threading.Lock() 加载线程的锁对象,是一个基本的锁对象,一次只能一个锁定,其余锁请求,需等待锁释放后才能获取

4.threading.RLock() 多重锁,在同一线程中可用被多次acquire。如果使用RLock,那么acquire和release必须成对出现,
调用了n次acquire锁请求,则必须调用n次的release才能在线程中释放锁对象

例如:
无锁:

#coding=utf8
import threading
import time

num = 0

def sum_num(i):
    global num
    time.sleep(1)
    num +=i
    print num

print '%s thread start!'%(time.ctime())

try:
   for i in range(6):
       t =threading.Thread(target=sum_num,args=(i,))
       t.start()

except KeyboardInterrupt,e:
    print "you stop the threading"

print '%s thread end!'%(time.ctime())

输出:

Sun May 28 20:54:59 2017 thread start!
Sun May 28 20:54:59 2017 thread end!
01
3
710
15

结果显示混乱

引入锁:

#coding=utf8
import threading
import time

num = 0

def sum_num(i):
    lock.acquire()
    global num
    time.sleep(1)
    num +=i
    print num
    lock.release()

print '%s thread start!'%(time.ctime())

try:
   lock=threading.Lock()
   list = []
   for i in range(6):
       t =threading.Thread(target=sum_num,args=(i,))
       list.append(t)
       t.start()

   for threadinglist in list:
        threadinglist.join()

except KeyboardInterrupt,e:
    print "you stop the threading"

print '%s thread end!'%(time.ctime())

结果:

Sun May 28 21:15:37 2017 thread start!
0
1
3
6
10
15
Sun May 28 21:15:43 2017 thread end!

其中:
lock=threading.Lock()加载锁的方法也可以换成lock=threading.RLock()

如果将上面的sum_num修改为:

    lock.acquire()
    global num
    lock.acquire()
    time.sleep(1)
    num +=i
    lock.release()
    print num
    lock.release()

那么:
lock=threading.Lock() 加载的锁,则一直处于等待中,锁等待
而lock=threading.RLock() 运行正常

你可能感兴趣的:(python)