Python批量查询关键词是否是指数词

简单的查询一个关键词是否为指数词，暂不能查询指数，过段时间研究下

# -*- coding: utf-8 -*-

import urllib2

import urllib

import cookielib

import re

import bs4

url_baidu_index = u'';

# 也可以用这个

url_baidu_token = '';

url_baidu_login = '';

#设置用户名、密码

username = '';

password = '';

#设置cookie，这里cookiejar可自动管理，无需手动指定

cj = cookielib.cookiejar();

opener = urllib2.build_opener(urllib2.httpcookieprocessor(cj));

urllib2.install_opener(opener);

reqreturn = urllib2.urlopen(url_baidu_index);

#获取token,

tokenreturn = urllib2.urlopen(url_baidu_token);

matchval = re.search(u'"token" : "(?p.*?)"',tokenreturn.read());

tokenval = matchval.group('tokenval');

#构造登录请求参数，该请求数据是通过抓包获得，对应请求

postdata = {

'username' : username,

'password' : password,

'u' : '',

'tpl' : 'pp',

'token' : tokenval,

'staticpage' : '',

'isphone' : 'false',

'charset' : 'utf-8',

'callback' : 'parent.bd__pcbs__ra48vi'

};

postdata = urllib.urlencode(postdata);

#发送登录请求

loginrequest = urllib2.request(url_baidu_login,postdata);

loginrequest.add_header('accept','text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8');

loginrequest.add_header('accept-encoding','gzip,deflate,sdch');

loginrequest.add_header('accept-language','zh-cn,zh;q=0.8');

loginrequest.add_header('user-agent','mozilla/5.0 (windows nt 6.1; wow64) applewebkit/537.36 (khtml, like gecko) chrome/28.0.1500.72 safari/537.36');

loginrequest.add_header('content-type','application/x-www-form-urlencoded');

sendpost = urllib2.urlopen(loginrequest);

#获取指数页面

keylist = open('key.txt','r') #查询关键词的文件

i = '如要查看相关数据'

for key in keylist:

key = key.strip('\n')

teibaurl = '{0}'.format(key)

content = urllib2.urlopen(teibaurl).read();

content = content.decode('gbk').encode('utf8');

if i in content:

print key,'不是指数关键词'

else:

print key,'指数关键词'

喜欢python，seo的同学可以加qq群472779600