korean parse

mtd_parse.py

#! /usr/bin/env python
# -*- coding: utf-8 -*-
# 
import lxml.html

f = open('in.htm', 'r')
html = f.read()
f.close()
doc = lxml.html.document_fromstring(html)
for tr in doc.xpath('//*[@style="border: 1px solid rgb(178, 178, 178); width: 70px; height: 40px; overflow: hidden; margin-left: 5px;"]'):
  url = tr.getchildren()[0].attrib['href']
	preview = tr.getchildren()[0].getchildren()[0].attrib['src']
	print url, preview