gistfile1.txt

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

from urllib2 import urlopen

from lxml import etree

import os,codecs

regexpNS = "http://exslt.org/regular-expressions"

url = urlopen('http://www.ddmap.com/map/21/busline-overview.htm')

parser = etree.HTMLParser(encoding="gb2312")

tree = etree.parse(url,parser)

xpath = etree.XPath("//a[re:test(attribute::href, '^/map/21/bus-key-')]",

                    namespaces={'re':regexpNS})

#xpath2 = etree.XPath("//a[@href]")

matches = xpath(tree)

#matches2 = xpath2(tree)

f = codecs.open('shanghai.txt','w',encoding='utf-8')

print len(matches)

f.write(str(len(matches))+os.linesep)

for m in matches:

    f.write(m.text+os.linesep)