在Python中,使用多线程multi-threading可以『同时』执行多个任务,比如你需要一个线程来复制读取信息,另一个线程来解析。为什么这里的同时要加引号呢,这是由于Python中GIL,也就是全局锁,看似同时执行多个任务,实际上是分布执行的,只不过各自完成不同的任务会提高工作效率。如果你不了解GIL,请看下图
实例1
import threadingdef job(): info = "this is an added thread, which is %s" % threading.current_thread print(info)def main(): print(threading.active_count()) print(threading.enumerate()) #print(threading.current_thread()) # 创建一个Thread对象,让他来复制job任务 new_thread = threading.Thread(target=job) # 执行线程任务 new_thread.start() print(threading.active_count())if __name__ == '__main__': main()
运行结果:
1[<_MainThread(MainThread, started 140637982934784)>]this is an added thread, which is <function current_thread at 0x7fe8cd92d378>2
解释一下:
实例2
import threadingimport timedef job(): print('T1 start\n') # 让任务延迟一下 for i in range(10): time.sleep(0.1) print('T1 finish\n')def job2(): print('T2 start\n') # 任务不延迟 print('T2 finish\n')def main(): thread1 = threading.Thread(target=job,name='T1') thread2 = threading.Thread(target=job2,name='T2') thread1.start() thread2.start() #thread1.join() print('all done')if __name__ == '__main__': main()
运行结果:
T1 startT2 startT2 finishall doneT1 finish
会发现all done并不是在最后执行的,怎么办呢?试试把join方法解除封(注)印(释),请看结果:
T1 startT2 startT2 finishT1 finishall done
解释一下:
实例3
import threadingfrom queue import Queueimport timedef job(l,q): for i in range(len(l)): l[i] = l[i]**2 q.put(l)def multithreading(data): q = Queue() threads = [] THREAD_NUM = 4 for i in range(THREAD_NUM): t = threading.Thread(target=job,args=(data[i],q)) t.start() threads.append(t) for t in threads: t.join() results = [] for _ in range(THREAD_NUM): results.append(q.get()) print(results)def unithreading(data): for i in range(len(data)): l = data[i] for j in range(len(l)): data[i][j] = data[i][j]**2 print(data)if __name__ == '__main__': data1 = [[1],[2,3],[4,5,6],[7,8,9,10]] data2 = [[1],[2,3],[4,5,6],[7,8,9,10]] time1 = time.clock() unithreading(data1) time2 = time.clock() multithreading(data2) time3 = time.clock() runtime_uni = time2 - time1 runtime_multi = time3 - time2 print("Runtime for unithreading is %s seconds." % runtime_uni) print("Runtime for multithreading is %s seconds." % runtime_multi)
运行结果:
[[1], [4, 9], [16, 25, 36], [49, 64, 81, 100]][[1], [4, 9], [16, 25, 36], [49, 64, 81, 100]]Runtime for unithreading is 4.300000000000137e-05 seconds.Runtime for multithreading is 0.001118000000000001 seconds.
解释一下:
实例4
import threadingdef job1(): global A #global A,lock #lock.acquire() for i in range(10): A = 1 print('job1',A) #lock.release()def job2(): global A #global A,lock #lock.acquire() for i in range(10): A = 10 print('job2',A) #lock.release()if __name__ == '__main__': lock = threading.Lock() A = 0 t1 = threading.Thread(target=job1) t2 = threading.Thread(target=job2) t1.start() t2.start() t1.join() t2.join()
运行结果:
job1job2 111job1job2 1222job1job2 2333job1job2 3444job1job2 4555job1job2 5666job1job2 6777job1job2 7888job1job2 8999job1job2 100110
你会发现这打印得错乱了,这也是多个线程交替工作得原因,如果先想要规定一个线程工作的时候防止别的线程干扰,需要怎么做呢?试试使用lock(注释部分)吧,再运行一下:
job1 1job1 2job1 3job1 4job1 5job1 6job1 7job1 8job1 9job1 10job2 20job2 30job2 40job2 50job2 60job2 70job2 80job2 90job2 100job2 110
解释一下:
多线程的简单应用就到这里,我们最开始说过由于GIL,Python得多线程并不是真正的去同时执行多个任务,那么有没有办法弥补呢?有,那就是多进程!你一定听说过多核吧,下次再讲:)
来源:https://www.icode9.com/content-1-605851.html联系客服