IP地址爬取

ip_spider.py= = =

#!/usr/bin/python
# coding: utf-8
import os
import sys
import requests
import re
import urllib

import sys
reload(sys)
sys.setdefaultencoding( "utf-8" )

def getUrl(urlIP):
    url = ‘http://www.123cha.com/ip/?q=%s‘ % urlIP
    r = requests.get(url)
    reg = r‘<td class="tg-data">(.+?.)</td>‘
    gre = re.compile(reg)
    number = re.findall(gre,r.text)
    print number[0]
    print number[2]
    fsock = open(‘ipaddress.txt‘, ‘a+‘)
    fsock.write("%s|%s\n" % (str(number[0]),str(number[2])))

if __name__ == ‘__main__‘:

    file_object = open(‘ipfile3‘)
    list_of_all_the_lines = file_object.readlines( )
    # print list_of_all_the_lines
    for dd in list_of_all_the_lines:
        getUrl(‘%s‘ % dd)