Python——新浪微博爬虫之模拟登陆

　　在编写微博爬虫的过程中，免不了要进行模拟登录，因为新浪微博不登陆只能访问少量的微博信息。

　　然而，由于新浪微博的反爬虫功能在不断更新，例如改变了密码的加密算法（RSA），以前的一些模拟登陆方式已经不适用了。所以一开始试了好几种方法，均不能成功。后来受http://www.jb51.net/article/46053.htm启发，已经实现了。

　　目前，亲测能用的步骤是：①通过预登录，使用GET方法，获得登录所需的servertime, nonce, pubkey, rsakv；②使用encode64加密用户名，使用RSA算法加密密码；③登录。

　　步骤一：

　　response格式为（换行是我自己加上去的）：

sinaSSOController.preloginCallBack({"retcode":0,"servertime":1430743809,"pcid":"gz-9e4ab7356f6545655af2a97e1cad7aa86446","nonce":"VA2IUS","pubkey":"EB2A38568661887FA180BDDB5CABD5F21C7BFD59C090CB2D245A87AC253062882729293E5506350508E7F9AA3BB77F4333231490F915F6D63C55FE2F08A49B353F444AD3993CACC02DB784ABBB8E42A9B1BBFFFB38BE18D78E87A0E41B9B8F73A928EE0CCEE1F6739884B9777E4FE9E88A1BBE495927AC4A799B3181D6442443","rsakv":"1330428213","exectime":3})

　　python代码为：

prelogin_url = ‘http://login.sina.com.cn/sso/prelogin.php?entry=sso&callback=sinaSSOController.preloginCallBack&su=%s&rsakt=mod&client=ssologin.js(v1.4.4)‘ % username
response = urllib2.urlopen(prelogin_url)
p = re.compile(r‘\((.*?)\)‘) #用来解析括号中的json格式内容
strurl = p.search(response.read()).group(1)
dic = dict(eval(strurl)) #json格式的response
pubkey = str(dic.get(‘pubkey‘))
servertime = str(dic.get(‘servertime‘))
nonce = str(dic.get(‘nonce‘))
rsakv = str(dic.get(‘rsakv‘))

　　步骤二：

　　加密用户名：

username_ = urllib.quote(username)
username = base64.encodestring(username_)[:-1]

　　加密密码：

rsaPublickey = int(pubkey, 16)
key = rsa.PublicKey(rsaPublickey, 65537) #创建公钥
message = servertime + ‘\t‘ + nonce + ‘\n‘ + password #拼接明文js加密文件中得到
passwd = rsa.encrypt(message, key) #加密
passwd = binascii.b2a_hex(passwd) #将加密信息转换为16进制。

　　完整代码：

  1 #! /usr/bin/env python
  2 #coding=utf8
  3
  4 import urllib
  5 import urllib2
  6 import cookielib
  7 import base64
  8 import re
  9 import json
 10 import hashlib
 11 import rsa
 12 import binascii
 13
 14 cj = cookielib.LWPCookieJar()
 15 cookie_support = urllib2.HTTPCookieProcessor(cj)
 16 opener = urllib2.build_opener(cookie_support, urllib2.HTTPHandler)
 17 urllib2.install_opener(opener)
 18 postdata = {
 19     ‘entry‘: ‘weibo‘,
 20     ‘gateway‘: ‘1‘,
 21     ‘from‘: ‘‘,
 22     ‘savestate‘: ‘7‘,
 23     ‘userticket‘: ‘1‘,
 24     ‘ssosimplelogin‘: ‘1‘,
 25     ‘vsnf‘: ‘1‘,
 26     ‘vsnval‘: ‘‘,
 27     ‘su‘: ‘‘,
 28     ‘service‘: ‘miniblog‘,
 29     ‘servertime‘: ‘‘,
 30     ‘nonce‘: ‘‘,
 31     ‘pwencode‘: ‘rsa2‘, #加密算法
 32     ‘sp‘: ‘‘,
 33     ‘encoding‘: ‘UTF-8‘,
 34     ‘prelt‘: ‘401‘,
 35     ‘rsakv‘: ‘‘,
 36     ‘url‘: ‘http://weibo.com/ajaxlogin.php?framelogin=1&callback=parent.sinaSSOController.feedBackUrlCallBack‘,
 37     ‘returntype‘: ‘META‘
 38 }
 39
 40 class WeiboLogin:
 41     def __init__(self, username, password):
 42         self.username = username
 43         self.password = password
 44
 45     def __get_spwd(self):
 46         rsaPublickey = int(self.pubkey, 16)
 47         key = rsa.PublicKey(rsaPublickey, 65537) #创建公钥
 48         message = self.servertime + ‘\t‘ + self.nonce + ‘\n‘ + self.password #拼接明文js加密文件中得到
 49         passwd = rsa.encrypt(message, key) #加密
 50         passwd = binascii.b2a_hex(passwd) #将加密信息转换为16进制。
 51         return passwd
 52
 53     def __get_suser(self):
 54         username_ = urllib.quote(self.username)
 55         username = base64.encodestring(username_)[:-1]
 56         return username
 57
 58     def __prelogin(self):
 59         prelogin_url = ‘http://login.sina.com.cn/sso/prelogin.php?entry=sso&callback=sinaSSOController.preloginCallBack&su=%s&rsakt=mod&client=ssologin.js(v1.4.4)‘ % self.username
 60         response = urllib2.urlopen(prelogin_url)
 61         p = re.compile(r‘\((.*?)\)‘)
 62         strurl = p.search(response.read()).group(1)
 63         dic = dict(eval(strurl)) #json格式的response
 64         self.pubkey = str(dic.get(‘pubkey‘))
 65         self.servertime = str(dic.get(‘servertime‘))
 66         self.nonce = str(dic.get(‘nonce‘))
 67         self.rsakv = str(dic.get(‘rsakv‘))
 68
 69     def login(self):
 70         url = ‘http://login.sina.com.cn/sso/login.php?client=ssologin.js(v1.4.18)‘
 71         try:
 72             self.__prelogin() #预登录
 73         except:
 74             print ‘Prelogin Error‘
 75             return
 76         global postdata
 77         postdata[‘servertime‘] = self.servertime
 78         postdata[‘nonce‘] = self.nonce
 79         postdata[‘su‘] = self.__get_suser()
 80         postdata[‘sp‘] = self.__get_spwd()
 81         postdata[‘rsakv‘] = self.rsakv
 82         postdata = urllib.urlencode(postdata)
 83         headers = {‘User-Agent‘:‘Mozilla/5.0 (Macintosh; Intel Mac OS X 10.10; rv:37.0) Gecko/20100101 Firefox/37.0‘}
 84         req  = urllib2.Request(
 85             url = url,
 86             data = postdata,
 87             headers = headers
 88         )
 89         result = urllib2.urlopen(req)
 90         text = result.read()
 91         p = re.compile(‘location\.replace\(\‘(.*?)\‘\)‘)
 92         try:
 93             login_url = p.search(text).group(1)
 94             urllib2.urlopen(login_url)
 95             print "Login Succeed!"
 96         except:
 97             print ‘Login Error!‘
 98
 99 if __name__ == ‘__main__‘:
100     uid = ‘your username‘
101     psw = ‘your password‘
102     simLogin = WeiboLogin(uid, psw)
103     simLogin.login()