4 import xml.etree.ElementTree
7 from .common import InfoExtractor
14 class ORFIE(InfoExtractor):
15 _VALID_URL = r'https?://tvthek.orf.at/(programs/.+?/episodes|topics/.+?)/(?P<id>\d+)'
17 def _real_extract(self, url):
18 mobj = re.match(self._VALID_URL, url)
19 playlist_id = mobj.group('id')
20 webpage = self._download_webpage(url, playlist_id)
22 flash_xml = self._search_regex('ORF.flashXML = \'(.+?)\'', webpage, u'flash xml')
23 flash_xml = compat_urlparse.parse_qs('xml='+flash_xml)['xml'][0]
24 flash_config = xml.etree.ElementTree.fromstring(flash_xml.encode('utf-8'))
25 playlist_json = self._search_regex(r'playlist\': \'(\[.*?\])\'', webpage, u'playlist').replace(r'\"','"')
26 playlist = json.loads(playlist_json)
29 ns = '{http://tempuri.org/XMLSchema.xsd}'
30 xpath = '%(ns)sPlaylist/%(ns)sItems/%(ns)sItem' % {'ns': ns}
31 webpage_description = self._og_search_description(webpage)
32 for (i, (item, info)) in enumerate(zip(flash_config.findall(xpath), playlist), 1):
33 # Get best quality url
35 for q in ['Q6A', 'Q4A', 'Q1A']:
36 video_url = find_xpath_attr(item, '%sVideoUrl' % ns, 'quality', q)
37 if video_url is not None:
38 rtmp_url = video_url.text
41 raise ExtractorError(u'Couldn\'t get video url: %s' % info['id'])
42 description = self._html_search_regex(
43 r'id="playlist_entry_%s".*?<p>(.*?)</p>' % i, webpage,
44 u'description', default=webpage_description, flags=re.DOTALL)
48 'title': info['title'],
51 'description': description,