LinuxEye - Linux系统教程

LinuxEye - Linux系统教程

当前位置: 主页 > 脚本编程 >

Python多线程抓取代理服务器

时间:2015-02-26 11:42来源:Linux运维笔记 编辑:yeho 点击:
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 http://blog.linuxeye.com/340.html 来抓取代理服务器的,谁知道过完年回来
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 http://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
#BLOG: blog.linuxeye.com
#coding:utf-8

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)
测试:
python proxy.py
结果如下:
['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']
['200.84.116.99', '9064', '\xe5\xa7\x94\xe5\x86\x85\xe7\x91\x9e\xe6\x8b\x89 ']
['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']
..........总共抓取了1921个代理..........
..........总共有524个代理通过校验..........
# more proxy_list.txt
202.106.169.142:80      北京市 联通ADSL 0.291432857513
111.13.136.59:80        北京市 移动     0.297957897186
111.13.136.56:80        北京市 移动     0.373070955276
111.206.81.248:80       北京市 联通     0.403017997742
111.13.136.58:80        北京市 移动     0.414332151413
124.202.217.134:8118    北京市 电信通   0.416817903519
124.202.183.218:8118    北京市 电信通   0.426618099213
120.132.71.232:80       北京市 联通     0.440200090408
61.232.6.164:8081       北京市 铁通     0.469615936279
118.144.96.253:80       北京市 电信通   0.485229969025
203.192.10.66:80        北京市 新华社   0.51485991478
124.202.182.22:8118     北京市 电信通   0.553130865097
数据库:
mysql> select * from m_common.proxy limit 10;
+----------+-----------------+------+----------+----------------------+---------------------+
| proxy_id | ip              | port | speed    | address              | create_time         |
+----------+-----------------+------+----------+----------------------+---------------------+
|        1 | 202.106.169.142 |   80 | 0.291433 | 北京市 联通ADSL      | 2015-02-26 11:29:24 |
|        2 | 111.13.136.59   |   80 | 0.297958 | 北京市 移动          | 2015-02-26 11:29:24 |
|        3 | 111.13.136.56   |   80 | 0.373071 | 北京市 移动          | 2015-02-26 11:29:24 |
|        4 | 111.206.81.248  |   80 | 0.403018 | 北京市 联通          | 2015-02-26 11:29:24 |
|        5 | 111.13.136.58   |   80 | 0.414332 | 北京市 移动          | 2015-02-26 11:29:24 |
|        6 | 124.202.217.134 | 8118 | 0.416818 | 北京市 电信通        | 2015-02-26 11:29:24 |
|        7 | 124.202.183.218 | 8118 | 0.426618 | 北京市 电信通        | 2015-02-26 11:29:24 |
|        8 | 120.132.71.232  |   80 |   0.4402 | 北京市 联通          | 2015-02-26 11:29:24 |
|        9 | 61.232.6.164    | 8081 | 0.469616 | 北京市 铁通          | 2015-02-26 11:29:24 |
|       10 | 118.144.96.253  |   80 |  0.48523 | 北京市 电信通        | 2015-02-26 11:29:24 |
+----------+-----------------+------+----------+----------------------+---------------------+
10 rows in set (0.00 sec)
原文:http://blog.linuxeye.com/410.html
------分隔线----------------------------
标签:Python爬虫多线程
栏目列表
推荐内容