python爬虫-全局的资源访问问题

python爬虫-全局的资源访问问题

概述

为了解决全局的变量的问题，线程不同步。

线程锁的使用

代码1

#在使用多线程的情况下，会使用到全局变量
import threading

gmoney=0
lock=threading.Lock()
def creater():
    global gmoney
    lock.acquire()
    for i in range(1000000):
        gmoney+=1
    lock.release()
    print(gmoney)
    return gmoney
def consumer():
    global gmoney
    lock.acquire()
    for i in range(1000000):
        gmoney+=1
    print(gmoney)
    lock.release()
    return gmoney


if __name__ == '__main__':
    threading.Thread(target=creater).start()
    threading.Thread(target=consumer).start()

代码2：

import threading
import time
import random

money=0
lock=threading.Lock()
gtime=0

class Product(threading.Thread):
    def run(self) -> None:
        global money
        global gtime
        while True:
            lock.acquire()
            num=random.randint(0,100)
            money+=num
            print(threading.current_thread().name,'生产了','%d元'%num,'当前的总额为',money)
            lock.release()
            time.sleep(1)
            gtime+=1
            if gtime>=100:
                return



class consumer(threading.Thread):
    def run(self) -> None:
        global money
        while True:
            lock.acquire()
            num=random.randint(1,100)
            if money<num:
                print('余额不足无法消费')
                lock.release()
                time.sleep(1)
                continue
            money-=num
            print(threading.current_thread().name,'消费了','%d元'%num,'当前的总额为',money)
            lock.release()
            time.sleep(1)



def main():
    for i in range(5):
        th=Product(name="生产者%d"%i)
        th.start()
    for i in range(5):
        th=consumer(name="消费者%d"%i)
        th.start()

if __name__ == '__main__':
    print(threading.enumerate())
    main()

使用conditoin

提供了休眠的方式，减少浪费cpu的资源

import threading
import time
import random

money=0
gdondition=threading.Condition()
gtime=0

class Product(threading.Thread):
    def run(self) -> None:
        global money
        global gtime
        while True:
            gdondition.acquire()
            num=random.randint(0,100)
            money+=num
            print(threading.current_thread().name,'生产了','%d元'%num,'当前的总额为',money)
            gdondition.notify_all()
            gdondition.release()
            time.sleep(1)
            gtime+=1
            if gtime>=100:
                return



class consumer(threading.Thread):
    def run(self) -> None:
        global money
        while True:
            gdondition.acquire()
            num=random.randint(1,100)
            while money<num:
                print('余额不足无法消费,%s正在等待资源中'%threading.current_thread().name)
                gdondition.wait()
            money-=num
            print(threading.current_thread().name,'消费了','%d元'%num,'当前的总额为',money)
            gdondition.release()
            time.sleep(1)



def main():
    for i in range(5):
        th=Product(name="生产者%d"%i)
        th.start()
    for i in range(5):
        th=consumer(name="消费者%d"%i)
        th.start()

if __name__ == '__main__':
    print(threading.enumerate())
    main()

使用队列的方式

from queue import Queue
import time

q=Queue(3)
# for i in range(2):
#     q.put(i)
q.empty()  #判断是否为空

# 超过队列的容量，对阻塞，如果不需要
for i in range(4):
    try:
        q.put(i,block=False)
    except:
        print('error')

print("done")
print(q.qsize())  # 取决与你放进去的元素的多少

def addvalue(q):
    for i in range(100):
        q.put(i,block=False)
        print("生产者生产了%d"%i)
        time.sleep(1)

def getValue(q):
    for i in range(100):
      a=q.get()
      print("消费者获取到了%d"%a)

import threading
if __name__ == '__main__':
    q=Queue(100)
    threading.Thread(target=addvalue,args=[q]).start()
    threading.Thread(target=getValue,args=[q]).start()