Fetches the feed from heise.de and modifies each link to the shorturl

heise-feed-cleaner.py

import requests
import re

from lxml import etree
from StringIO import StringIO

r=requests.get("http://www.heise.de/newsticker/heise.rdf")

parser=etree.XMLParser(encoding="utf8")
tree = etree.parse(StringIO(r.text.encode("utf8")), parser)

def get_plain_direct(link):
    d = {'A': '0', 'B': '.', 'C': '/', 'E': '-', 'L': 'http://', 'S': 'www.'}
    return re.sub('0([A-Z])', lambda match: d[match.group(1)], link)

def get_plain_head(link):
    r = requests.head(link)
    return r.url

def convertlink(link):
    plink = get_plain_direct(link)
    m = re.search("-([0-9]+).html", plink)
    if m:
        return "http://heise.de/-%s" % m.group(1)
    else:
        return link

for e in tree.xpath('//link'):
     e.text = convertlink(e.text)

print etree.tostring(tree)