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：
1.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）。
2.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）
