这里以抓取 http://www.proxy.com.ru 站点的代理服务器为例,代码如下:

  1#!/usr/bin/env python
  2#coding:utf-8
  3import urllib2
  4import re
  5import threading
  6import time
  7import MySQLdb
  8rawProxyList = []
  9checkedProxyList = []
 10#抓取代理网站
 11targets = []
 12for i in xrange(1,42):
 13        target = r"http://www.proxy.com.ru/list_%d.html" % i
 14        targets.append(target)
 15#抓取代理服务器正则
 16p = re.compile(r'''<tr><b><td>(\d+)</td><td>(.+?)</td><td>(\d+)</td><td>(.+?)</td><td>(.+?)</td></b></tr>''')
 17#获取代理的类
 18class ProxyGet(threading.Thread):
 19    def __init__(self,target):
 20        threading.Thread.__init__(self)
 21        self.target = target
 22    def getProxy(self):
 23        print "代理服务器目标网站: " + self.target
 24        req = urllib2.urlopen(self.target)
 25        result = req.read()
 26        #print chardet.detect(result)
 27        matchs = p.findall(result)
 28#       print matchs
 29        for row in matchs:
 30            ip=row[1]
 31            port =row[2]
 32            addr = row[4].decode("cp936").encode("utf-8")
 33            proxy = [ip,port,addr]
 34            print proxy
 35            rawProxyList.append(proxy)
 36    def run(self):
 37        self.getProxy()
 38#检验代理的类
 39class ProxyCheck(threading.Thread):
 40    def __init__(self,proxyList):
 41        threading.Thread.__init__(self)
 42        self.proxyList = proxyList
 43        self.timeout = 5
 44        self.testUrl = "http://www.baidu.com/"
 45        self.testStr = "030173"
 46    def checkProxy(self):
 47        cookies = urllib2.HTTPCookieProcessor()
 48        for proxy in self.proxyList:
 49            proxyHandler = urllib2.ProxyHandler({"http" : r'http://%s:%s' %(proxy[0],proxy[1])})
 50            #print r'http://%s:%s' %(proxy[0],proxy[1])
 51            opener = urllib2.build_opener(cookies,proxyHandler)
 52            opener.addheaders = [('User-agent', 'Mozilla/5.0 (Windows NT 6.2; WOW64; rv:22.0) Gecko/20100101 Firefox/22.0')]
 53            #urllib2.install_opener(opener)
 54            t1 = time.time()
 55            try:
 56                #req = urllib2.urlopen("http://www.baidu.com", timeout=self.timeout)
 57                req = opener.open(self.testUrl, timeout=self.timeout)
 58                #print "urlopen is ok...."
 59                result = req.read()
 60                #print "read html...."
 61                timeused = time.time() - t1
 62                pos = result.find(self.testStr)
 63                #print "pos is %s" %pos
 64                if pos > 1:
 65                    checkedProxyList.append((proxy[0],proxy[1],proxy[2],timeused))
 66                    #print "ok ip: %s %s %s %s" %(proxy[0],proxy[1],proxy[2],timeused)
 67                else:
 68                     continue
 69            except Exception,e:
 70                #print e.message
 71                continue
 72    def run(self):
 73        self.checkProxy()
 74if __name__ == "__main__":
 75    getThreads = []
 76    checkThreads = []
 77#对每个目标网站开启一个线程负责抓取代理
 78for i in range(len(targets)):
 79    t = ProxyGet(targets[i])
 80    getThreads.append(t)
 81for i in range(len(getThreads)):
 82    getThreads[i].start()
 83for i in range(len(getThreads)):
 84    getThreads[i].join()
 85print '.'*10+"总共抓取了%s个代理" %len(rawProxyList) +'.'*10
 86#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
 87for i in range(20):
 88    t = ProxyCheck(rawProxyList[((len(rawProxyList)+19)/20) * i:((len(rawProxyList)+19)/20) * (i+1)])
 89    checkThreads.append(t)
 90for i in range(len(checkThreads)):
 91    checkThreads[i].start()
 92for i in range(len(checkThreads)):
 93    checkThreads[i].join()
 94print '.'*10+"总共有%s个代理通过校验" %len(checkedProxyList) +'.'*10
 95#插入数据库,表结构自己创建,四个字段ip,port,speed,address
 96def db_insert(insert_list):
 97    try:
 98        conn = MySQLdb.connect(host="localhost", user="root", passwd="admin",db="m_common",charset='utf8')
 99        cursor = conn.cursor()
100        cursor.execute('delete from proxy')
101        cursor.execute('alter table proxy AUTO_INCREMENT=1')
102        cursor.executemany("INSERT INTO proxy(ip,port,speed,address) VALUES (%s,%s,%s,%s)",insert_list)
103        conn.commit()
104        cursor.close()
105        conn.close()
106    except MySQLdb.Error,e:
107        print "Mysql Error %d: %s" % (e.args[0], e.args[1])
108#代理排序持久化
109proxy_ok = []
110f= open("proxy_list.txt",'w+')
111for proxy in sorted(checkedProxyList,cmp=lambda x,y:cmp(x[3],y[3])):
112    if proxy[3] < 8:
113        #print "checked proxy is: %s:%s\t%s\t%s" %(proxy[0],proxy[1],proxy[2],proxy[3])
114        proxy_ok.append((proxy[0],proxy[1],proxy[3],proxy[2]))
115        f.write("%s:%s\t%s\t%s\n"%(proxy[0],proxy[1],proxy[2],proxy[3]))
116f.close()
117db_insert(proxy_ok)

测试: python proxy.py

结果如下:

 1['61.58.94.179', '8088', '\xe5\x8f\xb0\xe6\xb9\xbe\xe7\x9c\x81 \xe5\x8f\xb0\xe6\xb9\xbe\xe5\xae\xbd\xe9\xa2\x91\xe9\x80\x9a\xe8\xae\xaf\xe9\xa1\xbe\xe9\x97\xae\xe8\x82\xa1\xe4\xbb\xbd\xe6\x9c\x89\xe9\x99\x90\xe5\x85\xac\xe5\x8f\xb8']
 2['200.84.116.99', '9064', '\xe5\xa7\x94\xe5\x86\x85\xe7\x91\x9e\xe6\x8b\x89 ']
 3['183.223.204.8', '8123', '\xe5\x9b\x9b\xe5\xb7\x9d\xe7\x9c\x81\xe8\x87\xaa\xe8\xb4\xa1\xe5\xb8\x82 \xe7\xa7\xbb\xe5\x8a\xa8']
 4..........总共抓取了1921个代理..........
 5..........总共有524个代理通过校验..........
 6
 7# more proxy_list.txt
 8
 9202.106.169.142:80      北京市 联通ADSL 0.291432857513
10111.13.136.59:80        北京市 移动     0.297957897186
11111.13.136.56:80        北京市 移动     0.373070955276
12111.206.81.248:80       北京市 联通     0.403017997742
13111.13.136.58:80        北京市 移动     0.414332151413
14124.202.217.134:8118    北京市 电信通   0.416817903519
15124.202.183.218:8118    北京市 电信通   0.426618099213
16120.132.71.232:80       北京市 联通     0.440200090408
1761.232.6.164:8081       北京市 铁通     0.469615936279
18118.144.96.253:80       北京市 电信通   0.485229969025
19203.192.10.66:80        北京市 新华社   0.51485991478
20124.202.182.22:8118     北京市 电信通   0.553130865097

数据库:

 1mysql> select * from m_common.proxy limit 10;
 2+----------+-----------------+------+----------+----------------------+---------------------+
 3| proxy_id | ip              | port | speed    | address              | create_time         |
 4+----------+-----------------+------+----------+----------------------+---------------------+
 5|        1 | 202.106.169.142 |   80 | 0.291433 | 北京市 联通ADSL      | 2015-02-26 11:29:24 |
 6|        2 | 111.13.136.59   |   80 | 0.297958 | 北京市 移动          | 2015-02-26 11:29:24 |
 7|        3 | 111.13.136.56   |   80 | 0.373071 | 北京市 移动          | 2015-02-26 11:29:24 |
 8|        4 | 111.206.81.248  |   80 | 0.403018 | 北京市 联通          | 2015-02-26 11:29:24 |
 9|        5 | 111.13.136.58   |   80 | 0.414332 | 北京市 移动          | 2015-02-26 11:29:24 |
10|        6 | 124.202.217.134 | 8118 | 0.416818 | 北京市 电信通        | 2015-02-26 11:29:24 |
11|        7 | 124.202.183.218 | 8118 | 0.426618 | 北京市 电信通        | 2015-02-26 11:29:24 |
12|        8 | 120.132.71.232  |   80 |   0.4402 | 北京市 联通          | 2015-02-26 11:29:24 |
13|        9 | 61.232.6.164    | 8081 | 0.469616 | 北京市 铁通          | 2015-02-26 11:29:24 |
14|       10 | 118.144.96.253  |   80 |  0.48523 | 北京市 电信通        | 2015-02-26 11:29:24 |
15+----------+-----------------+------+----------+----------------------+---------------------+
1610 rows in set (0.00 sec)