Python代码的执行由Python虚拟机(也叫解释器主循环)来控制。Python在设计之初就考虑到要在主循环中,同时只有一个线程在执行。虽然 Python 解释器中能够“运行”多个线程,但在任意时刻只有一个线程在解释器中运行。
对Python虚拟机的访问由全局解释器锁(GIL)来控制,正是这个锁能保证同一时刻只有一个线程在运行。python
在多线程环境中,Python 虚拟机按如下方式执行:git
设置 GIL程序员
切换到一个线程去运行github
运行指定数量的字节码指令或者线程主动让出控制(能够调用 time.sleep(0))编程
把线程设置为睡眠状态json
解锁 GIL安全
再次重复以上全部步骤服务器
在调用外部代码(如 C/C++扩展函数)的时候,GIL将会被锁定,直到这个函数结束为止(因为在这期间没有Python的字节码被运行,因此不会作线程切换)编写扩展的程序员能够主动解锁GIL。数据结构
Python提供了几个用于多线程编程的模块,包括thread、threading和Queue等。thread和threading模块容许程序员建立和管理线程。thread模块提供了基本的线程和锁的支持,threading提供了更高级别、功能更强的线程管理的功能。Queue模块容许用户建立一个能够用于多个线程之间共享数据的队列数据结构。多线程
避免使用thread模块,由于更高级别的threading模块更为先进,对线程的支持更为完善,并且使用thread模块里的属性有可能会与threading出现冲突;其次低级别的thread模块的同步原语不多(实际上只有一个),而threading模块则有不少;再者,thread模块中当主线程结束时,全部的线程都会被强制结束掉,没有警告也不会有正常的清除工做,至少threading模块能确保重要的子线程退出后进程才退出。
thread模块不支持守护线程,当主线程退出时,全部的子线程不论它们是否还在工做,都会被强行退出。而threading模块支持守护线程,守护线程通常是一个等待客户请求的服务器,若是没有客户提出请求它就在那等着,若是设定一个线程为守护线程,就表示这个线程是不重要的,在进程退出的时候,不用等待这个线程退出。
multiprocess模块的彻底模仿了threading模块的接口,两者在使用层面,有很大的类似性,于是再也不详细介绍。
建立线程的方式1:
from threading import Thread import time def sayhi(name): time.sleep(2) print('%s say hello' %name) if __name__ == '__main__': t=Thread(target=sayhi,args=('egon',)) t.start() print('主线程')
建立线程的方式2:
from threading import Thread import time class Sayhi(Thread): def __init__(self,name): super().__init__() self.name=name def run(self): time.sleep(2) print('%s say hello' % self.name) if __name__ == '__main__': t = Sayhi('egon') t.start() print('主线程')
pid的比较:
from threading import Thread from multiprocessing import Process import os def work(): print('hello',os.getpid()) if __name__ == '__main__': #part1:在主进程下开启多个线程,每一个线程都跟主进程的pid同样 t1=Thread(target=work) t2=Thread(target=work) t1.start() t2.start() print('主线程/主进程pid',os.getpid()) #part2:开多个进程,每一个进程都有不一样的pid p1=Process(target=work) p2=Process(target=work) p1.start() p2.start() print('主线程/主进程pid',os.getpid())
开启效率的较量:
from threading import Thread from multiprocessing import Process import os def work(): print('hello') if __name__ == '__main__': #在主进程下开启线程 t=Thread(target=work) t.start() print('主线程/主进程') ''' 打印结果: hello 主线程/主进程 ''' #在主进程下开启子进程 t=Process(target=work) t.start() print('主线程/主进程') ''' 打印结果: 主线程/主进程 hello '''
内存数据的共享问题:
from threading import Thread from multiprocessing import Process import os def work(): global n n=0 if __name__ == '__main__': # n=100 # p=Process(target=work) # p.start() # p.join() # print('主',n) #毫无疑问子进程p已经将本身的全局的n改为了0,但改的仅仅是它本身的,查看父进程的n仍然为100 n=1 t=Thread(target=work) t.start() t.join() print('主',n) #查看结果为0,由于同一进程内的线程之间共享进程内的数据 同一进程内的线程共享该进程的数据?
Thread实例对象的方法 # isAlive(): 返回线程是否活动的。 # getName(): 返回线程名。 # setName(): 设置线程名。 threading模块提供的一些方法: # threading.currentThread(): 返回当前的线程变量。 # threading.enumerate(): 返回一个包含正在运行的线程的list。正在运行指线程启动后、结束前,不包括启动前和终止后的线程。 # threading.activeCount(): 返回正在运行的线程数量,与len(threading.enumerate())有相同的结果。
代码示例:
from threading import Thread import threading from multiprocessing import Process import os def work(): import time time.sleep(3) print(threading.current_thread().getName()) if __name__ == '__main__': #在主进程下开启线程 t=Thread(target=work) t.start() print(threading.current_thread().getName()) print(threading.current_thread()) #主线程 print(threading.enumerate()) #连同主线程在内有两个运行的线程 print(threading.active_count()) print('主线程/主进程') ''' 打印结果: MainThread <_MainThread(MainThread, started 140735268892672)> [<_MainThread(MainThread, started 140735268892672)>, <Thread(Thread-1, started 123145307557888)>] 主线程/主进程 Thread-1 '''
join方法:
from threading import Thread import time def sayhi(name): time.sleep(2) print('%s say hello' %name) if __name__ == '__main__': t=Thread(target=sayhi,args=('egon',)) t.start() t.join() print('主线程') print(t.is_alive()) ''' egon say hello 主线程 False '''
不管是进程仍是线程,都遵循:守护xx会等待主xx运行完毕后被销毁。须要强调的是:运行完毕并不是终止运行。
对主进程来讲,运行完毕指的是主进程代码运行完毕。
对主线程来讲,运行完毕指的是主线程所在的进程内全部非守护线程通通运行完毕,主线程才算运行完毕。
主进程在其代码结束后就已经算运行完毕了(守护进程在此时就被回收),而后主进程会一直等非守护的子进程都运行完毕后回收子进程的资源(不然会产生僵尸进程),才会结束。
主线程在其余非守护线程运行完毕后才算运行完毕(守护线程在此时就被回收)。由于主线程的结束意味着进程的结束,进程总体的资源都将被回收,而进程必须保证非守护线程都运行完毕后才能结束。
守护线程实例1:
from threading import Thread import time def sayhi(name): time.sleep(2) print('%s say hello' %name) if __name__ == '__main__': t=Thread(target=sayhi,args=('egon',)) t.setDaemon(True) #必须在t.start()以前设置 t.start() print('主线程') print(t.is_alive()) ''' 主线程 True '''
守护线程实例2:
import time def foo(): print(123) time.sleep(1) print("end123") def bar(): print(456) time.sleep(3) print("end456") t1=Thread(target=foo) t2=Thread(target=bar) t1.daemon=True t1.start() t2.start() print("main-------")
多个线程抢占资源的状况:
from threading import Thread import os,time def work(): global n temp=n time.sleep(0.1) n=temp-1 if __name__ == '__main__': n=100 l=[] for i in range(100): p=Thread(target=work) l.append(p) p.start() for p in l: p.join() print(n) #结果可能为99
对公共数据的操做:
import threading R=threading.Lock() R.acquire() R.release()
同步锁的引用:
from threading import Thread,Lock import os,time def work(): global n lock.acquire() temp=n time.sleep(0.1) n=temp-1 lock.release() if __name__ == '__main__': lock=Lock() n=100 l=[] for i in range(100): p=Thread(target=work) l.append(p) p.start() for p in l: p.join() print(n) #结果确定为0,由原来的并发执行变成串行,牺牲了执行效率保证了数据安全
互斥锁与join的区别:
#不加锁:并发执行,速度快,数据不安全 from threading import current_thread,Thread,Lock import os,time def task(): global n print('%s is running' %current_thread().getName()) temp=n time.sleep(0.5) n=temp-1 if __name__ == '__main__': n=100 lock=Lock() threads=[] start_time=time.time() for i in range(100): t=Thread(target=task) threads.append(t) t.start() for t in threads: t.join() stop_time=time.time() print('主:%s n:%s' %(stop_time-start_time,n)) ''' Thread-1 is running Thread-2 is running ...... Thread-100 is running 主:0.5216062068939209 n:99 ''' #不加锁:未加锁部分并发执行,加锁部分串行执行,速度慢,数据安全 from threading import current_thread,Thread,Lock import os,time def task(): #未加锁的代码并发运行 time.sleep(3) print('%s start to run' %current_thread().getName()) global n #加锁的代码串行运行 lock.acquire() temp=n time.sleep(0.5) n=temp-1 lock.release() if __name__ == '__main__': n=100 lock=Lock() threads=[] start_time=time.time() for i in range(100): t=Thread(target=task) threads.append(t) t.start() for t in threads: t.join() stop_time=time.time() print('主:%s n:%s' %(stop_time-start_time,n)) ''' Thread-1 is running Thread-2 is running ...... Thread-100 is running 主:53.294203758239746 n:0 ''' #有的同窗可能有疑问:既然加锁会让运行变成串行,那么我在start以后当即使用join,就不用加锁了啊,也是串行的效果啊 #没错:在start以后马上使用jion,确定会将100个任务的执行变成串行,毫无疑问,最终n的结果也确定是0,是安全的,但问题是 #start后当即join:任务内的全部代码都是串行执行的,而加锁,只是加锁的部分即修改共享数据的部分是串行的 #单从保证数据安全方面,两者均可以实现,但很明显是加锁的效率更高. from threading import current_thread,Thread,Lock import os,time def task(): time.sleep(3) print('%s start to run' %current_thread().getName()) global n temp=n time.sleep(0.5) n=temp-1 if __name__ == '__main__': n=100 lock=Lock() start_time=time.time() for i in range(100): t=Thread(target=task) t.start() t.join() stop_time=time.time() print('主:%s n:%s' %(stop_time-start_time,n)) ''' Thread-1 start to run Thread-2 start to run ...... Thread-100 start to run 主:350.6937336921692 n:0 #耗时是多么的恐怖 ''' )
进程也有死锁与递归锁,在进程那里忘记说了,放到这里一切说了额。
所谓死锁: 是指两个或两个以上的进程或线程在执行过程当中,因争夺资源而形成的一种互相等待的现象,若无外力做用,它们都将没法推动下去。此时称系统处于死锁状态或系统产生了死锁,这些永远在互相等待的进程称为死锁进程,以下就是死锁。
from threading import Lock as Lock import time mutexA=Lock() mutexA.acquire() mutexA.acquire() print(123) mutexA.release() mutexA.release()
解决方法,递归锁,在Python中为了支持在同一线程中屡次请求同一资源,python提供了可重入锁RLock。
这个RLock内部维护着一个Lock和一个counter变量,counter记录了acquire的次数,从而使得资源能够被屡次require。直到一个线程全部的acquire都被release,其余的线程才能得到资源。上面的例子若是使用RLock代替Lock,则不会发生死锁:
递归锁RLock:
from threading import RLock as Lock import time mutexA=Lock() mutexA.acquire() mutexA.acquire() print(123) mutexA.release() mutexA.release()
典型问题:科学家吃面
死锁问题:
import time from threading import Thread,Lock noodle_lock = Lock() fork_lock = Lock() def eat1(name): noodle_lock.acquire() print('%s 抢到了面条'%name) fork_lock.acquire() print('%s 抢到了叉子'%name) print('%s 吃面'%name) fork_lock.release() noodle_lock.release() def eat2(name): fork_lock.acquire() print('%s 抢到了叉子' % name) time.sleep(1) noodle_lock.acquire() print('%s 抢到了面条' % name) print('%s 吃面' % name) noodle_lock.release() fork_lock.release() for name in ['哪吒','egon','yuan']: t1 = Thread(target=eat1,args=(name,)) t2 = Thread(target=eat2,args=(name,)) t1.start() t2.start()
递归锁解决死锁问题:
import time from threading import Thread,RLock fork_lock = noodle_lock = RLock() def eat1(name): noodle_lock.acquire() print('%s 抢到了面条'%name) fork_lock.acquire() print('%s 抢到了叉子'%name) print('%s 吃面'%name) fork_lock.release() noodle_lock.release() def eat2(name): fork_lock.acquire() print('%s 抢到了叉子' % name) time.sleep(1) noodle_lock.acquire() print('%s 抢到了面条' % name) print('%s 吃面' % name) noodle_lock.release() fork_lock.release() for name in ['哪吒','egon','yuan']: t1 = Thread(target=eat1,args=(name,)) t2 = Thread(target=eat2,args=(name,)) t1.start() t2.start()
queue队列 :使用import queue,用法与进程Queue同样
queue is especially useful in threaded programming when information must be exchanged safely between multiple threads.
queue.
Queue
(maxsize=0) #先进先出
先进先出:
import queue q=queue.Queue() q.put('first') q.put('second') q.put('third') print(q.get()) print(q.get()) print(q.get()) ''' 结果(先进先出): first second third '''
class queue.
LifoQueue
(maxsize=0) #last in fisrt out
后进先出:
import queue q=queue.LifoQueue() q.put('first') q.put('second') q.put('third') print(q.get()) print(q.get()) print(q.get()) ''' 结果(后进先出): third second first '''
class queue.
PriorityQueue
(maxsize=0) #存储数据时可设置优先级的队列
优先级队列:
import queue q=queue.PriorityQueue() #put进入一个元组,元组的第一个元素是优先级(一般是数字,也能够是非数字之间的比较),数字越小优先级越高 q.put((20,'a')) q.put((10,'b')) q.put((30,'c')) print(q.get()) print(q.get()) print(q.get()) ''' 结果(数字越小优先级越高,优先级高的优先出队): (10, 'b') (20, 'a') (30, 'c') '''
#1 介绍 concurrent.futures模块提供了高度封装的异步调用接口 ThreadPoolExecutor:线程池,提供异步调用 ProcessPoolExecutor: 进程池,提供异步调用 Both implement the same interface, which is defined by the abstract Executor class. #2 基本方法 #submit(fn, *args, **kwargs) 异步提交任务 #map(func, *iterables, timeout=None, chunksize=1) 取代for循环submit的操做 #shutdown(wait=True) 至关于进程池的pool.close()+pool.join()操做 wait=True,等待池内全部任务执行完毕回收完资源后才继续 wait=False,当即返回,并不会等待池内的任务执行完毕 但无论wait参数为什么值,整个程序都会等到全部任务执行完毕 submit和map必须在shutdown以前 #result(timeout=None) 取得结果 #add_done_callback(fn) 回调函数 # done() 判断某一个线程是否完成 # cancle() 取消某个任务
ProcessPoolExecutor:
#介绍 The ProcessPoolExecutor class is an Executor subclass that uses a pool of processes to execute calls asynchronously. ProcessPoolExecutor uses the multiprocessing module, which allows it to side-step the Global Interpreter Lock but also means that only picklable objects can be executed and returned. class concurrent.futures.ProcessPoolExecutor(max_workers=None, mp_context=None) An Executor subclass that executes calls asynchronously using a pool of at most max_workers processes. If max_workers is None or not given, it will default to the number of processors on the machine. If max_workers is lower or equal to 0, then a ValueError will be raised. #用法 from concurrent.futures import ThreadPoolExecutor,ProcessPoolExecutor import os,time,random def task(n): print('%s is runing' %os.getpid()) time.sleep(random.randint(1,3)) return n**2 if __name__ == '__main__': executor=ProcessPoolExecutor(max_workers=3) futures=[] for i in range(11): future=executor.submit(task,i) futures.append(future) executor.shutdown(True) print('+++>') for future in futures: print(future.result())
ThreadPoolExecutor:
#介绍 ThreadPoolExecutor is an Executor subclass that uses a pool of threads to execute calls asynchronously. class concurrent.futures.ThreadPoolExecutor(max_workers=None, thread_name_prefix='') An Executor subclass that uses a pool of at most max_workers threads to execute calls asynchronously. Changed in version 3.5: If max_workers is None or not given, it will default to the number of processors on the machine, multiplied by 5, assuming that ThreadPoolExecutor is often used to overlap I/O instead of CPU work and the number of workers should be higher than the number of workers for ProcessPoolExecutor. New in version 3.6: The thread_name_prefix argument was added to allow users to control the threading.Thread names for worker threads created by the pool for easier debugging. #用法 与ProcessPoolExecutor相同
map的用法:
from concurrent.futures import ThreadPoolExecutor,ProcessPoolExecutor import os,time,random def task(n): print('%s is runing' %os.getpid()) time.sleep(random.randint(1,3)) return n**2 if __name__ == '__main__': executor=ThreadPoolExecutor(max_workers=3) # for i in range(11): # future=executor.submit(task,i) executor.map(task,range(1,12)) #map取代了for+submit
回调函数:
from concurrent.futures import ThreadPoolExecutor,ProcessPoolExecutor from multiprocessing import Pool import requests import json import os def get_page(url): print('<进程%s> get %s' %(os.getpid(),url)) respone=requests.get(url) if respone.status_code == 200: return {'url':url,'text':respone.text} def parse_page(res): res=res.result() print('<进程%s> parse %s' %(os.getpid(),res['url'])) parse_res='url:<%s> size:[%s]\n' %(res['url'],len(res['text'])) with open('db.txt','a') as f: f.write(parse_res) if __name__ == '__main__': urls=[ 'https://www.baidu.com', 'https://www.python.org', 'https://www.openstack.org', 'https://help.github.com/', 'http://www.sina.com.cn/' ] # p=Pool(3) # for url in urls: # p.apply_async(get_page,args=(url,),callback=pasrse_page) # p.close() # p.join() p=ProcessPoolExecutor(3) for url in urls: p.submit(get_page,url).add_done_callback(parse_page) #parse_page拿到的是一个future对象obj,须要用obj.result()拿到结果