youtube_dl/extractor/mdr.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     ExtractorError,
   6 )
   7
   8
   9 class MDRIE(InfoExtractor):
  10     _VALID_URL = r'^(?P<domain>(?:https?://)?(?:www\.)?mdr\.de)/mediathek/(?:.*)/(?P<type>video|audio)(?P<video_id>[^/_]+)_.*'
  11
  12     _TESTS = [{
  13         u'url': u'http://www.mdr.de/mediathek/themen/nachrichten/video165624_zc-c5c7de76_zs-3795826d.html',
  14         u'file': u'165624.mp4',
  15         u'md5': u'ae785f36ecbf2f19b42edf1bc9c85815',
  16         u'info_dict': {
  17             u"title": u"MDR aktuell Eins30 09.12.2013, 22:48 Uhr"
  18         },
  19     },
  20     {
  21         u'url': u'http://www.mdr.de/mediathek/radio/mdr1-radio-sachsen/audio718370_zc-67b21197_zs-1b9b2483.html',
  22         u'file': u'718370.mp3',
  23         u'md5': u'a9d21345a234c7b45dee612f290fd8d7',
  24         u'info_dict': {
  25             u"title": u"MDR 1 RADIO SACHSEN 10.12.2013, 05:00 Uhr"
  26         },
  27     }]
  28
  29     def _real_extract(self, url):
  30         m = re.match(self._VALID_URL, url)
  31         video_id = m.group('video_id')
  32         domain = m.group('domain')
  33         mediatype = m.group('type')
  34
  35         # determine title and media streams from webpage
  36         html = self._download_webpage(url, video_id)
  37
  38         title = self._html_search_regex(r'<h2>(.*?)</h2>', html, u'title')
  39         xmlurl = self._search_regex(
  40             r'(/mediathek/(?:.+)/(?:video|audio)[0-9]+-avCustom.xml)', html, u'XML URL')
  41
  42         doc = self._download_xml(domain + xmlurl, video_id)
  43         formats = []
  44         for a in doc.findall('./assets/asset'):
  45             url_el = a.find('.//progressiveDownloadUrl')
  46             if url_el is None:
  47                 continue
  48             abr = int(a.find('bitrateAudio').text) // 1000
  49             media_type = a.find('mediaType').text
  50             format = {
  51                 'abr': abr,
  52                 'filesize': int(a.find('fileSize').text),
  53                 'url': url_el.text,
  54             }
  55
  56             vbr_el = a.find('bitrateVideo')
  57             if vbr_el is None:
  58                 format.update({
  59                     'vcodec': 'none',
  60                     'format_id': u'%s-%d' % (media_type, abr),
  61                 })
  62             else:
  63                 vbr = int(vbr_el.text) // 1000
  64                 format.update({
  65                     'vbr': vbr,
  66                     'width': int(a.find('frameWidth').text),
  67                     'height': int(a.find('frameHeight').text),
  68                     'format_id': u'%s-%d' % (media_type, vbr),
  69                 })
  70             formats.append(format)
  71         formats.sort(key=lambda f: (f.get('vbr'), f['abr']))
  72         if not formats:
  73             raise ValueError('Could not find any valid formats')
  74
  75         return {
  76             'id': video_id,
  77             'title': title,
  78             'formats': formats,
  79         }