Python抓取页面中超链接(URL)的3中方法比较(HTMLParser、pyquery、正则表达式)
HTMLParser版：#!/usr/bin/python
# -*- coding: UTF-8 -*-
import HTMLParser
class UrlParser(HTMLParser.HTMLParser):
????def __init__(self):
????????HTMLParser.HTMLParser.__init__(self)
????????self.urls = []
????def handle_starttag(self, tag, attrs):
????????if tag == 'a':
????????????for name,value in attrs:
????????????????if name == 'href':
????????????????????self.urls.append(value)
????def geturls(self):
????????return self.urls
if __name__ == '__main__':
????urls = []
????url = UrlParser()
????url.feed('1111111111<a href="http://www.bccn.net">BCCN</a>2222222<a href="http://bbs.bccn.net">BCCN.BBS</a>333333333')
????urls += url.geturls()
????print urls?pyquery版：#!/usr/bin/python
# -*- coding: UTF-8 -*-
from pyquery import PyQuery as pq
class UrlParser():
????def __init__(self):
????????self.urls = []
????def feed(self,data):
????????d = pq(data)
????????if d.find('a'):
????????????#关于下面一行，我用d('a').attr('href')只能得到第一个URL，暂时只会用map，不知道有没有别的够pythonic的代码
????????????url = d('a').map(lambda i, e: pq(e)('a').attr('href'))
????????????for u in url:
????????????????self.urls.append(u)
????def geturls(self):
????????return self.urls
if __name__ == '__main__':
????urls = []
????url = UrlParser()
????url.feed('1111111111<a href="http://www.bccn.net">BCCN</a>2222222<a href="http://bbs.bccn.net">BCCN.BBS</a>333333333')
????urls += url.geturls()
????print urls?正则表达式版：#!/usr/bin/python
# -*- coding: UTF-8 -*-
import re
class UrlParser():
????def __init__(self):
????????self.urls = []
????def feed(self,data):
????????url = re.findall(r'''<a(\s*)(.*?)(\s*)href(\s*)=(\s*)(["\s]*)([^"\']+?)(["\s]+)(.*?)>''',data,re.S|re.I)
????????for u in url:
????????????self.urls.append(u[6])
????def geturls(self):
????????return self.urls
if __name__ == '__main__':
????urls = []
????url = UrlParser()
????url.feed('1111111111<a href="http://www.bccn.net">BCCN</a>2222222<a href="http://bbs.bccn.net">BCCN.BBS</a>333333333')
????urls += url.geturls()
????print urls??速度比较：正则表达式 > pyquery > HTMLParser测试的时候遍历大约1000个页面，正则表达式占绝对优势，这3个速度比例大约是 8:2:1HTMLParser最慢，pyquery速度大约是它的2倍，正则的速度是它的8倍，看来以后如非必要不再考虑HTMLParser了，用起来也不如pyquery方便，正则速度倒是很快，功能也强大，前两者能提取的内容用正则全部都能实现，而正则能实现的功能前两者就不一定能实现了。只是正则的可读性不好。以后遇到数据量大的用正则表达式，数据量不大不考虑时间因素但逻辑复杂的的用pyquery，以后维护起来方便
