python模拟浏览器登陆

#! /usr/bin/env python
# -*-coding:utf-8 -*-

import urllib
import urllib2
import cookielib

class NetRobot:
    def __init__(self, baseurl):
        self.cj = cookielib.CookieJar()
        self.baseurl = baseurl
        self.opener = urllib2.build_opener(urllib2.HTTPCookieProcessor(self.cj))
        urllib2.install_opener(self.opener)
    
    def getData(self,username,password,**data):
        print ‘origin data:‘,data
        self.opener.open(self.baseurl)
        try:
            data[‘csrfmiddlewaretoken‘] = [x.value for x in self.cj if "csrftoken"==x.name][0]　＃用于Django开启csrf中间件时
        except:
            pass
        data[‘username‘]=username
        data[‘password‘]=password
        return data

    def login(self,follow_up_url, username,password,**others):
        print
        print ‘-‘*25,‘login...‘,‘-‘*25
        login_url = self.baseurl + follow_up_url
        data = self.getData(username, password, **others)
        post_data = urllib.urlencode(data)
        print post_data
        header = {‘User-Agent‘:‘Mozilla/5.0 (compatible; MSIE 9.0; Windows NT 6.1; WOW64; Trident/5.0)‘}　　＃假装是浏览器
        request = urllib2.Request(login_url, post_data, header)　　　　　　　　＃构造post请求
        try:
            self.opener.open(request)　　　　　　　　　　　　　　　　　　　　　　　　＃提交请求
            print ‘^‘*25,‘login success‘,‘^‘*25
        except:
            print ‘!‘*25,‘login failure‘,‘!‘*25
        print

    def visit(self, follow_up_url):
        request = urllib2.Request(self.baseurl + follow_up_url)
        content = self.opener.open(request)
        print content.read()

if __name__ == ‘__main__‘:
    nr = NetRobot(‘http://192.168.1.100‘)　　　　　　　　　 #指定网站根路径
    nr.login(‘/login/‘, ‘user‘, ‘password‘, hidRem=‘‘)　　#此处假设需要额外的的hiRem参数
    nr.visit(‘/api/folders/15/?format=json‘)