Screen scape heise.de newsticker (german)

tsch () — Sun, 13 Nov 2005 03:04:25 GMT



#!/usr/bin/env python

# -*- encoding: latin1 -*-



import BeautifulSoup

from PyRSS2Gen import RSSItem, Guid

import ScrapeNFeed

import urllib2

import re



debug = 0



def fetch(url):

    response = urllib2.urlopen(urllib2.Request(url))

    return response.read(),response.info()



class HeiFeed(ScrapeNFeed.ScrapedFeed):    

    def HTML2RSS(self, headers, body):

        items = []

        soup = BeautifulSoup.BeautifulSoup(body)

        for item in soup('a', {'href' : re.compile('^meldung.*')}):

            link = 'http://www.heise.de/newsticker/' + item['href']

            if not self.hasSeen(link):

                title = item.contents[0].strip()

                if debug:

                    print "title: " + title

                    print "link : " + link

                response, headers = fetch(link)

                s = BeautifulSoup.BeautifulSoup(response)

                desc = s.fetch('div',{'class':'meldung_wrapper'})[0].prettify()

                items.append(RSSItem(title=title, description=desc, link=link))

            self.addRSSItems(items)



HeiFeed.load("heise.de newsticker", 'http://www.heise.de/newsticker/',

             "heise.de newsticker", 'heise_rss.xml', 'heise_rss.pickle',

             managingEditor = 'tsch')

DZone Snippets: scrape code

Screen scape heise.de newsticker (german)