Python3.4实现从HTTP代理网站批量获取代理并筛选的方法示例

__author__ = 'multiangle'
__edition__='python3.4'
import threading
import urllib.request as request
import time
rawProxyList=[]
checkedProxyList=[]
class proxycheck(threading.Thread):
 def __init__(self,proxy_list):
  threading.Thread.__init__(self)
  self.proxy_list=proxy_list
  self.timeout=3
  self.testurl='http://www.sina.cn/'
  self.testStr='手机新浪网'
 def checkproxy(self):
  cookies=request.HTTPCookieProcessor()
  for proxy in self.proxy_list:
   handler=request.ProxyHandler({'http':'http://%s'%(proxy)})
   opener=request.build_opener(cookies,handler)
   t1=time.time()
   try:
    req=opener.open(self.testurl,timeout=self.timeout)
    res=req.read()
    res=str(res,encoding='utf8')
    usetime=time.time()-t1
    if self.testStr in res:
     checkedProxyList.append((proxy,usetime))
   except Exception as e :
    print(e)
 def run(self):
  self.checkproxy()
if __name__=='__main__':
 num=20
 thread_num=10
 checkThrends=[]
 url='YOUR PROXY URL' #提取代理的网站。
 req=request.urlopen(url).read()
 req=str(req,encoding='utf-8')
 list=req.split('\r\n') #网站返回的是字符串格式，用'\r\n'进行分割
 rawProxyList=list
 print('get raw proxy')
 for i in rawProxyList:
  print(i)
 # s=proxycheck_test(rawProxyList)
 batch_size=int((len(rawProxyList)+thread_num-1)/thread_num)
 print(batch_size)
 for i in range(thread_num):
  t=proxycheck(rawProxyList[batch_size*i:batch_size*(i+1)])
  checkThrends.append(t)
 for i in range(checkThrends.__len__()):
  checkThrends[i].start()
 for i in range(checkThrends.__len__()):
  checkThrends[i].join()
 print(checkedProxyList.__len__(),' useful proxy is find')
 for i in checkedProxyList:
  print(i)