（一）基于python的requests库，模拟登陆
1 爬取公司信息，不用保持cookie的情况，代码如下：
# -*- coding: utf-8 -*-importrequestsimportrequestsimportreimportMySQLdbimportmysql.connectorimportMySQLdbasmdbConn = MySQLdb.connect(user=wenrui, passwd=wenrui, host=localhost, db=companydata, charset="utf8")cur = Conn.cursor()cur.execute("CREATE TABLE CompanyData(nameVARCHAR(70),realnamVARCHAR(20),phoneVARCHAR(20),mailVARCHAR(20),Bigaddressvarchar(10),Smladdressvarchar(70))")mit()classCompany:def__init__(self):self.headers = {User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.1.6) Gecko/1201 Firefox/3.5.6, Cookie: Hm_lvt_3ac4a19eb5f266a8046f9f5b29c52a00=1447818297; Hm_lpvt_3ac4a19eb5f266a8046f9f5b29c52a00=1447818297; so=296ef4827c1eabe98da0650c74e0bc7e; vr_1447818247=296ef4827c1eabe98da0650c74e0bc7e; un=2af45194a1467fca68d2ab4b6b0c083a58ebf139; zh_CN=zh_CN; PLAY_SESSION="74b2f9223df0e04bd8d69cc881072bb46fce9e63-userId=418920" #直接把cookie写入请求头即可 }self.rname = ur arget="_blank">( .*?)</a> self.rrealname = ur联系人：(.{1,6})& # 匹配联系人：后面任意一个1-4次，遇到&后结束 self.rphone = r<label>(.*?)</label> self.rmail = urhref="mailto.*>(.*?)</a> self.raddress = ur"<font color= ed>(.*?)</font>(.*?)</p>" # self.raddress=ur地.*址：(.*)defgetPage(
