• 欢迎访问开心洋葱网站,在线教程,推荐使用最新版火狐浏览器和Chrome浏览器访问本网站,欢迎加入开心洋葱 QQ群
  • 为方便开心洋葱网用户,开心洋葱官网已经开启复制功能!
  • 欢迎访问开心洋葱网站,手机也能访问哦~欢迎加入开心洋葱多维思维学习平台 QQ群
  • 如果您觉得本站非常有看点,那么赶紧使用Ctrl+D 收藏开心洋葱吧~~~~~~~~~~~~~!
  • 由于近期流量激增,小站的ECS没能经的起亲们的访问,本站依然没有盈利,如果各位看如果觉着文字不错,还请看官给小站打个赏~~~~~~~~~~~~~!

Python多线程抓取代理服务器保存数据库

其他 开心洋葱 1746次浏览 0个评论

Python多线程抓取代理服务器保存数据库

Python作为一门功能强大的脚本语言来说,经常被用来写爬虫程序,下面是Python爬虫多线程抓取代理服务器。
年前是用 //blog.linuxeye.com/340.html 来抓取代理服务器的,谁知道过完年回来http://www.cnproxy.com/被墙,导致某网站不能抓取东西,停止更新。
现在改抓取 http://www.proxy.com.ru 的代理服务器,这里的代理服务器更多,更简单。
下面代码可以直接用,代码如下:

#!/usr/bin/env python
#coding:utf-8
#BLOG: blog.linuxeye.com
import urllib2
import re
import threading
import time
import MySQLdb
rawProxyList = []
checkedProxyList = []
#抓取代理网站
targets = []
for i in xrange(1,42):
        target = r"http://www.proxy.com.ru/list_%d.html" % i
        targets.append(target)
#抓取代理服务器正则
p = re.compile(r'''<tr><b><td>(\d+)</td><td>(.+?)</td><td>(\d+)</td><td>(.+?)</td><td>(.+?)</td></b></tr>''')
#获取代理的类
class ProxyGet(threading.Thread):
    def __init__(self,target):
        threading.Thread.__init__(self)
        self.target = target
    def getProxy(self):
        print "代理服务器目标网站: " + self.target
        req = urllib2.urlopen(self.target)
        result = req.read()
        #print chardet.detect(result)
        matchs = p.findall(result)
#       print matchs
        for row in matchs:
            ip=row[1]
            port =row[2]
            addr = row[4].decode("cp936").encode("utf-8")
            proxy = [ip,port,addr]
            print proxy
            rawProxyList.append(proxy)
    def run(self):
        self.getProxy()
#检验代理的类
class ProxyCheck(threading.Thread):
    def __init__(self,proxyList):
        threading.Thread.__init__(self)
        self.proxyList = proxyList
        self.timeout = 5
        self.testUrl = "http://www.baidu.com/"
        self.testStr = "030173"
    def checkProxy(self):
        cookies = urllib2.HTTPCookieProcessor()
        for proxy in self.proxyList:
            proxyHandler = urllib2.ProxyHandler({"http" : r'http://%s:%s' %(proxy[0],proxy[1])})
            #print r'http://%s:%s' %(proxy[0],proxy[1])
            opener = urllib2.build_opener(cookies,proxyHandler)
            opener.addheaders = [('User-agent', 'Mozilla/5.0 (Windows NT 6.2; WOW64; rv:22.0) Gecko/20100101 Firefox/22.0')]
            #urllib2.install_opener(opener)
            t1 = time.time()
            try:
                #req = urllib2.urlopen("http://www.baidu.com", timeout=self.timeout)
                req = opener.open(self.testUrl, timeout=self.timeout)
                #print "urlopen is ok...."
                result = req.read()
                #print "read html...."
                timeused = time.time() - t1
                pos = result.find(self.testStr)
                #print "pos is %s" %pos
                if pos > 1:
                    checkedProxyList.append((proxy[0],proxy[1],proxy[2],timeused))
                    #print "ok ip: %s %s %s %s" %(proxy[0],proxy[1],proxy[2],timeused)
                else:
                     continue
            except Exception,e:
                #print e.message
                continue
    def run(self):
        self.checkProxy()
if __name__ == "__main__":
    getThreads = []
    checkThreads = []
#对每个目标网站开启一个线程负责抓取代理
for i in range(len(targets)):
    t = ProxyGet(targets[i])
    getThreads.append(t)
for i in range(len(getThreads)):
    getThreads[i].start()
for i in range(len(getThreads)):
    getThreads[i].join()
print '.'*10+"总共抓取了%s个代理" %len(rawProxyList) +'.'*10
#开启20个线程负责校验,将抓取到的代理分成20份,每个线程校验一份
for i in range(20):
    t = ProxyCheck(rawProxyList[((len(rawProxyList)+19)/20) * i:((len(rawProxyList)+19)/20) * (i+1)])
    checkThreads.append(t)
for i in range(len(checkThreads)):
    checkThreads[i].start()
for i in range(len(checkThreads)):
    checkThreads[i].join()
print '.'*10+"总共有%s个代理通过校验" %len(checkedProxyList) +'.'*10
#插入数据库,表结构自己创建,四个字段ip,port,speed,address
def db_insert(insert_list):
    try:
        conn = MySQLdb.connect(host="localhost", user="root", passwd="admin",db="m_common",charset='utf8')
        cursor = conn.cursor()
        cursor.execute('delete from proxy')
        cursor.execute('alter table proxy AUTO_INCREMENT=1')
        cursor.executemany("INSERT INTO proxy(ip,port,speed,address) VALUES (%s,%s,%s,%s)",insert_list)
        conn.commit()
        cursor.close()
        conn.close()
    except MySQLdb.Error,e:
        print "Mysql Error %d: %s" % (e.args[0], e.args[1])
#代理排序持久化
proxy_ok = []
f= open("proxy_list.txt",'w+')
for proxy in sorted(checkedProxyList,cmp=lambda x,y:cmp(x[3],y[3])):
    if proxy[3] < 8:
        #print "checked proxy is: %s:%s\t%s\t%s" %(proxy[0],proxy[1],proxy[2],proxy[3])
        proxy_ok.append((proxy[0],proxy[1],proxy[3],proxy[2]))
        f.write("%s:%s\t%s\t%s\n"%(proxy[0],proxy[1],proxy[2],proxy[3]))
f.close()
db_insert(proxy_ok)

使用方法如下:

# python -V
Python 2.6.6
# python proxy.py
# more proxy_list.txt


开心洋葱 , 版权所有丨如未注明 , 均为原创丨未经授权请勿修改 , 转载请注明Python多线程抓取代理服务器保存数据库
喜欢 (0)

您必须 登录 才能发表评论!

加载中……