3 from .common import InfoExtractor
9 class MDRIE(InfoExtractor):
10 _VALID_URL = r'^(?P<domain>(?:https?://)?(?:www\.)?mdr\.de)/mediathek/(?:.*)/(?P<type>video|audio)(?P<video_id>[^/_]+)_.*'
12 # No tests, MDR regularily deletes its videos
14 def _real_extract(self, url):
15 m = re.match(self._VALID_URL, url)
16 video_id = m.group('video_id')
17 domain = m.group('domain')
19 # determine title and media streams from webpage
20 html = self._download_webpage(url, video_id)
22 title = self._html_search_regex(r'<h2>(.*?)</h2>', html, u'title')
23 xmlurl = self._search_regex(
24 r'(/mediathek/(?:.+)/(?:video|audio)[0-9]+-avCustom.xml)', html, u'XML URL')
26 doc = self._download_xml(domain + xmlurl, video_id)
28 for a in doc.findall('./assets/asset'):
29 url_el = a.find('.//progressiveDownloadUrl')
32 abr = int(a.find('bitrateAudio').text) // 1000
33 media_type = a.find('mediaType').text
36 'filesize': int(a.find('fileSize').text),
40 vbr_el = a.find('bitrateVideo')
44 'format_id': u'%s-%d' % (media_type, abr),
47 vbr = int(vbr_el.text) // 1000
50 'width': int(a.find('frameWidth').text),
51 'height': int(a.find('frameHeight').text),
52 'format_id': u'%s-%d' % (media_type, vbr),
54 formats.append(format)
56 raise ExtractorError(u'Could not find any valid formats')
58 self._sort_formats(formats)