Python模拟登录 —— 征服验证码 9 微博weibo.com
抓包分析能够使用Http Analyzer，Filders，可是看起来很复杂，仍是使用火狐好（chrome远远没有火狐好用）。javascript
首先，在输入用户名后，会进行预登陆，网址为：http://login.sina.com.cn/sso/prelogin.php?entry=weibo&callback=sinaSSOController.preloginCallBack&su=ZW5nbGFuZHNldSU0MDE2My5jb20%3D&rsakt=mod&checkpin=1&client=ssologin.js(v1.4.18)&_=1443156845536，经过响应（sinaSSOController.preloginCallBack({"retcode":0,"servertime":1443156842,"pcid":"gz-e88b75a929252baec7c12c741985eaa45627","nonce":"2L4IZ3","pubkey":"EB2A38568661887FA180BDDB5CABD5F21C7BFD59C090CB2D245A87AC253062882729293E5506350508E7F9AA3BB77F4333231490F915F6D63C55FE2F08A49B353F444AD3993CACC02DB784ABBB8E42A9B1BBFFFB38BE18D78E87A0E41B9B8F73A928EE0CCEE1F6739884B9777E4FE9E88A1BBE495927AC4A799B3181D6442443","rsakv":"1330428213","showpin":0,"exectime":16})），咱们能够得到四个有用的变量，servertime、nonce、pubkey和rsakv。php
新浪微博的用户名加密目前采用Base64加密算法，而新浪微博登陆密码的加密算法使用RSA2，这是模拟登录的重点，须要先建立一个rsa公钥，公钥的两个参数新浪微博都给了固定值，第一个参数是登陆第一步中的pubkey，第二个参数是js加密文件中的‘10001’(针对网友的提问进行更新：这个其实就是在ssologin.js的响应中). 这两个值须要先从16进制转换成10进制，把10001转成十进制为65537，随后加入servertime和nonce再次加密。java
要提交的数据是：python
postdata = {
'entry': 'weibo',
'gateway': '1',
'from': '',
'savestate': '7',
'useticket': '1',
'pagerefer': "http://login.sina.com.cn/sso/logout.php?entry=miniblog&r=http%3A%2F%2Fweibo.com%2Flogout.php%3Fbackurl",
'vsnf': '1',
'su': su,
'service': 'miniblog',
'servertime': servertime,
'nonce': nonce,
'pwencode': 'rsa2',
'rsakv': rsakv,
'sp': password_secret,
'sr': '1366*768',
'encoding': 'UTF-8',
'prelt': '115',
'url': 'http://weibo.com/ajaxlogin.php?framelogin=1&callback=parent.sinaSSOController.feedBackUrlCallBack',
'returntype': 'META'
}
提交以后，由于重定向，还要获取重定向的url。git
使用python3。github
import time
import base64
import rsa
import binascii
import requests
import re
import random
try:
from PIL import Image
except:
pass
try:
from urllib.parse import quote_plus
except:
from urllib import quote_plus
'''
若是没有开启登陆保护，不用输入验证码就能够登陆
若是开启登陆保护，须要输入验证码
'''
# 构造 Request headers
agent = 'Mozilla/5.0 (Windows NT 6.3; WOW64; rv:41.0) Gecko/20100101 Firefox/41.0'
headers = {
'User-Agent': agent
}
session = requests.session()
# 访问 初始页面带上 cookie
index_url = "http://weibo.com/login.php"
try:
session.get(index_url, headers=headers, timeout=2)
except:
session.get(index_url, headers=headers)
try:
input = raw_input
except:
pass
def get_su(username):
"""
对 email 地址和手机号码 先 javascript 中 encodeURIComponent
对应 Python 3 中的是 urllib.parse.quote_plus
而后在 base64 加密后decode
"""
username_quote = quote_plus(username)
username_base64 = base64.b64encode(username_quote.encode("utf-8"))
return username_base64.decode("utf-8")
# 预登录得到 servertime, nonce, pubkey, rsakv
def get_server_data(su):
pre_url = "http://login.sina.com.cn/sso/prelogin.php?entry=weibo&callback=sinaSSOController.preloginCallBack&su="
pre_url = pre_url + su + "&rsakt=mod&checkpin=1&client=ssologin.js(v1.4.18)&_="
pre_url = pre_url + str(int(time.time() * 1000))
pre_data_res = session.get(pre_url, headers=headers)
sever_data = eval(pre_data_res.content.decode("utf-8").replace("sinaSSOController.preloginCallBack", ''))
return sever_data
# print(sever_data)
def get_password(password, servertime, nonce, pubkey):
rsaPublickey = int(pubkey, 16)
key = rsa.PublicKey(rsaPublickey, 65537) # 建立公钥
message = str(servertime) + '\t' + str(nonce) + '\n' + str(password) # 拼接明文js加密文件中获得
message = message.encode("utf-8")
passwd = rsa.encrypt(message, key) # 加密
passwd = binascii.b2a_hex(passwd) # 将加密信息转换为16进制。
return passwd
def get_cha(pcid):
cha_url = "http://login.sina.com.cn/cgi/pin.php?r="
cha_url = cha_url + str(int(random.random() * 100000000)) + "&s=0&p="
cha_url = cha_url + pcid
cha_page = session.get(cha_url, headers=headers)
with open("cha.jpg", 'wb') as f:
f.write(cha_page.content)
f.close()
try:
im = Image.open("cha.jpg")
im.show()
im.close()
except:
print(u"请到当前目录下，找到验证码后输入")
def login(username, password):
# su 是加密后的用户名
su = get_su(username)
sever_data = get_server_data(su)
servertime = sever_data["servertime"]
nonce = sever_data['nonce']
rsakv = sever_data["rsakv"]
pubkey = sever_data["pubkey"]
showpin = sever_data["showpin"]
password_secret = get_password(password, servertime, nonce, pubkey)
postdata = {
'entry': 'weibo',
'gateway': '1',
'from': '',
'savestate': '7',
'useticket': '1',
'pagerefer': "http://login.sina.com.cn/sso/logout.php?entry=miniblog&r=http%3A%2F%2Fweibo.com%2Flogout.php%3Fbackurl",
'vsnf': '1',
'su': su,
'service': 'miniblog',
'servertime': servertime,
'nonce': nonce,
'pwencode': 'rsa2',
'rsakv': rsakv,
'sp': password_secret,
'sr': '1366*768',
'encoding': 'UTF-8',
'prelt': '115',
'url': 'http://weibo.com/ajaxlogin.php?framelogin=1&callback=parent.sinaSSOController.feedBackUrlCallBack',
'returntype': 'META'
}
login_url = 'http://login.sina.com.cn/sso/login.php?client=ssologin.js(v1.4.18)'
if showpin == 0:
login_page = session.post(login_url, data=postdata, headers=headers)
else:
pcid = sever_data["pcid"]
get_cha(pcid)
postdata['door'] = input(u"请输入验证码")
login_page = session.post(login_url, data=postdata, headers=headers)
login_loop = (login_page.content.decode("GBK"))
# print(login_loop)
pa = r'location\.replace\([\'"](.*?)[\'"]\)'
loop_url = re.findall(pa, login_loop)[0]
# print(loop_url)
# 此出还能够加上一个是否登陆成功的判断，下次改进的时候写上
login_index = session.get(loop_url, headers=headers)
uuid = login_index.text
uuid_pa = r'"uniqueid":"(.*?)"'
uuid_res = re.findall(uuid_pa, uuid, re.S)[0]
web_weibo_url = "http://weibo.com/%s/profile?topnav=1&wvr=6&is_all=1" % uuid_res
weibo_page = session.get(web_weibo_url, headers=headers)
weibo_pa = r'<title>(.*?)</title>'
# print(weibo_page.content.decode("utf-8"))
userID = re.findall(weibo_pa, weibo_page.content.decode("utf-8", 'ignore'), re.S)[0]
print(u"欢迎你 %s, 登录成功" % userID)
if __name__ == "__main__":
username = input(u'用户名：')
password = input(u'密码：')
login(username, password)
另可参考：
http://blog.csdn.net/andrewseu/article/details/48730735
http://www.jianshu.com/p/816594c83c74
https://github.com/ResolveWangweb