今天简单使用了一下python的re模块和lxml模块,分别利用的它们提供的正则表达式和xpath来解析页面源码从中提取所需的title,xpath在完成这样的小任务上效率非常好,在这里之所以又使用了一下正则表达式是因为xpath在处理一些特殊的页面的时候会出现乱码的情况,当然这不是xpath的原因,而是页面本身编码,跟utf-8转码之间有冲突所致,这里看代码:
# !/usr/bin/python
#-*-coding:utf-8-*-
'''
功能:抽取指定url的页面内容中的title
'''
import re
import chardet
import urllib
from lxml import etree
def utf8_transfer(strs):
'''
utf8编码转换
'''
try:
if isinstance(strs, unicode):
strs = strs.encode('utf-8')
elif chardet.detect(strs)['encoding'] == 'GB2312':
strs = strs.decode("gb2312", 'ignore').encode('utf-8')
elif chardet.detect(strs)['encoding'] == 'utf-8':
strs = strs.decode('utf-8', 'ignore').encode('utf-8')
except Exception, e:
print 'utf8_transfer error', strs, e
return strs
def get_title_xpath(Html):
'''
用xpath抽取网页Title
'''
Html = utf8_transfer(Html)
Html_encoding = chardet.detect(Html)['encoding']
page = etree.HTML(Html, parser=etree.HTMLParser(encoding=Html_encoding))
title = page.xpath('/html/head/title/text()')
try:
title = title[0].strip()
except IndexError:
print 'Nothing'
print title
def get_title(Html):
'''
用re抽取网页Title
'''
Html = utf8_transfer(Html)
compile_rule = ur'<title>.*</title>'
title_list = re.findall(compile_rule, Html)
if title_list == []:
title = ''
else:
title = title_list[0][7:-8]
print title
if __name__ == '__main__':
url = 'http://www.baidu.com'
html = urllib.urlopen(url).read()
new_html = utf8_transfer(html)
try:
get_title_xpath(new_html)
get_title(new_html)
except Exception, e:
print e
下面是结果:
百度一下,你就知道
百度一下,你就知道
简单的小实践,继续学习,欢迎交流