Generates a m3u playlist from the 4players video newsfeed. Already processed videos are cached. New videos are labeled with "NEW" prefix.

4pl.py

#!/usr/bin/env python
import feedparser, re, urllib2, os, sys

CACHEDIR = os.path.join(os.path.split(os.path.realpath(__file__))[0], 'cache')
VIDEOFEEDURL = 'http://feeds.4players.de/Allgemein/videos/-/rss.xml'

def formatted(s):
	return re.sub(u'[^\w]+', u'-', s.strip(), 0, re.UNICODE).lower()

print '#EXTM3U'

f = feedparser.parse(VIDEOFEEDURL)

for e in f['entries']:
	title = e['title']
	summary = e['summary']

	m = re.search(r'href=[\'"]?[^\'" >]+redir=([^\'" >]+)[\'" >]+Weiter zum Video', summary)
	if not m:
		print "#error: no video link for '%s' found" % title.encode('utf-8')
		continue
	url = urllib2.unquote(m.group(1))

	# cache
	if not os.path.isdir(CACHEDIR):
		os.mkdir(CACHEDIR)
	
	filename = formatted(title)
	fullname = os.path.join(CACHEDIR, filename)
	try:
		with open(fullname) as f:
			page = ''.join(f.readlines())
	except IOError:
		title = "NEW: %s" % title
		page = urllib2.urlopen(url).read()
		with open(fullname, 'w') as f:
			f.write(page)

	# find video url
	m = re.search(r'<meta.*swf\?file=([^&]+)', page)
	if not m:
		print "#error: video url for '%s' is not found" % title.encode('utf-8')
		continue

	video_url = m.group(1)
	
	title = re.sub(u'[,-]', u' ', title, 0, re.UNICODE)
	print '#EXTINF:0,%s' % title.encode('utf-8')
	print video_url.encode('utf-8')