多线程-共享全局变量

from threading import Thread
import time

g_num = 100

def work1():
    global g_num
    for i in range(3):
        g_num += 1

    print("----in work1, g_num is %d---"%g_num)


def work2():
    global g_num
    print("----in work2, g_num is %d---"%g_num)


print("---线程创建之前g_num is %d---"%g_num)

t1 = Thread(target=work1)
t1.start()

#延时一会，保证t1线程中的事情做完
time.sleep(1)

t2 = Thread(target=work2)
t2.start()

运行结果:

---线程创建之前g_num is 100---
----in work1, g_num is 103---
----in work2, g_num is 103---

列表当做实参传递到线程中

from threading import Thread
import time

def work1(nums):
    nums.append(44)
    print("----in work1---",nums)


def work2(nums):
    #延时一会，保证t1线程中的事情做完
    time.sleep(1)
    print("----in work2---",nums)

g_nums = [11,22,33]

t1 = Thread(target=work1, args=(g_nums,))
t1.start()

t2 = Thread(target=work2, args=(g_nums,))
t2.start()

运行结果:

----in work1--- [11, 22, 33, 44]
----in work2--- [11, 22, 33, 44]

总结：

• 在一个进程内的所有线程共享全局变量，能够在不适用其他方式的前提下完成多线程之间的数据共享（这点要比多进程要好）
• 缺点就是，线程是对全局变量随意遂改可能造成多线程之间对全局变量的混乱（即线程非安全）