Atom feed script

feed.cgi

#!/usr/bin/env python
'Atom Script, by Sean B. Palmer'

import os, re, cgi
r_title = re.compile('(?s)<title>(.*?), by (.*?)</title>')
r_entry = re.compile('(?s)<div([^>]+)>(.*?)</div>')
r_attrs = re.compile('(?s)([a-z-]+)="([^"]+)"')

with open('index.html') as f: bytes = f.read()
title, author = r_title.search(bytes).groups()
site = os.environ['SERVER_NAME'] + os.path.dirname(os.environ['SCRIPT_NAME'])

print 'Content-Type: application/atom+xml; charset=utf-8\n'
print '<feed xmlns="http://www.w3.org/2005/Atom">'
print ' <title>%s</title>' % title
print ' <link href="http://%s/"/>' % site
print ' <id>tag:sbp.so,2011:%s</id>' % site
print ' <author><name>%s</name></author>' % author
entries = [e for e in r_entry.findall(bytes) if 'datetime=' in e[0]]
for i, (attrs, content) in enumerate(entries[:20]):
   attr = dict(r_attrs.findall(attrs)).get
   if not i: print ' <updated>%s</updated>' % attr('datetime')
   print ' <entry xml:base="http://%s/">' % site
   print '  <id>tag:sbp.so,2011:%s:%s</id>' % (site, attr('datetime'))
   print '  <title>%s</title>' % attr('title', attr('datetime')[:10])
   print '  <updated>%s</updated>' % attr('datetime')
   print '  <content type="html">%s</content>' % cgi.escape(content)
   print ' </entry>'
print '</feed>'