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

import sys

import urllib2

import re

import chardet

import sys

print sys.getdefaultencoding()  

keyword = u'来源：.+[\u4e00-\u9fa5]+'.encode('CP936')

html = 'http://finance.people.com.cn/money/n/2014/1009/c42877-25798373.html'

src = urllib2.urlopen(html).read()

print chardet.detect(keyword)

print chardet.detect(src)

match = re.compile(keyword)

list = match.findall( src)

for line in list:

    print line

ascii

{'confidence': 0.73, 'encoding': 'windows-1252'}

{'confidence': 0.99, 'encoding': 'GB2312'}

来源：人民网-理财频道 原创稿

来源：<a href="http://finance.people.com.cn/money/" target="_blank">人民网-理财频道</a></span>    <a href="http://www.people.com.cn/GB/123231/365208/index.html" target="_blank"><img src="/img/2012wbn/images/peopleclienticon.jpg" />手机看新闻

来源：<a href="http://news.baidu.com/view.html?from=people" target="_blank">百度新闻