#!/usr/bin/python
import urllib, re, cgi

# _pattern changed 17/01/2003 JMH

_pattern = re.compile(r"""\<a href\="/cgi/NGoto/(.*?)" target\="_blank" onClick\="return wopen\(this\)"\>(.*?)\</a\>.*?\<span class\="src"\>(.*?)\</span>""", re.DOTALL)
data = urllib.urlopen("http://www.newsnow.co.uk/newsfeed/?name=Southampton").read()
# data = open("nn.txt", "r").read()

links = {}
for link, desc, src in _pattern.findall( data ):
    src = src.replace('<span class="hl">', '', 1)
    src = src.replace('&nbsp;', ' ')
    src = src.replace('&#63;', '?')
    src = src.replace('&#58;', ':')
    src = src.replace('&#47;', '/')
    src = src.replace("&#39;", "'")
    src = src.replace('&#33;', '!')
    src = src.replace('  ', ' ')
    desc = desc.replace('&#63;', '?')
    desc = desc.replace('&#58;', ':')
    desc = desc.replace('&#47;', '/')
    desc = desc.replace("&#39;", "'")
    desc = desc.replace('&#33;', '!')
    desc = desc.replace('  ', ' ')
    links[ link ] = "%s - %s" % (src, desc)

sortedLinks = links.keys()
sortedLinks.sort()

print """<?xml version="1.0" encoding="iso-8859-1"?>

<rdf:RDF
  xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
  xmlns="http://purl.org/rss/1.0/"
  xmlns:dc="http://purl.org/dc/elements/1.1/"
  xmlns:dcterms="http://purl.org/dc/terms/">
<channel rdf:about="http://www.cix.co.uk/~jimh/weblog/2002_10_01_archive.html#83671998">
<title>NewsNow - Southampton FC</title>
<description>Southampton FC news - pulled from http://www.NewsNow.co.uk/ </description>
<link>http://www.cix.co.uk/~jimh/weblog/NewsNow.xml</link>
<language>en-GB</language>
<items>
<rdf:Seq>"""

for link in sortedLinks:
    print '<rdf:li rdf:resource="http://www.newsnow.co.uk/cgi/NGoto/%s" />' % link

print """</rdf:Seq>
</items>
</channel>"""

for link in sortedLinks:
    print '<item rdf:about="http://www.newsnow.co.uk/cgi/NGoto/%s">' % link
    print '<title>%s</title>' % links[ link ]
    print '<description>%s</description>' % links[ link ]
    print '<link>http://www.newsnow.co.uk/cgi/NGoto/%s</link>' % link
    print '</item>'

print '</rdf:RDF>'