Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/ard.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     determine_ext,
   9     ExtractorError,
  10 )
  11
  12
  13 class ARDIE(InfoExtractor):
  14     _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P<video_id>[^/\?]+)(?:\?.*)?'
  15
  16     _TEST = {
  17         'url': 'http://www.ardmediathek.de/das-erste/guenther-jauch/edward-snowden-im-interview-held-oder-verraeter?documentId=19288786',
  18         'file': '19288786.mp4',
  19         'md5': '515bf47ce209fb3f5a61b7aad364634c',
  20         'info_dict': {
  21             'title': 'Edward Snowden im Interview - Held oder Verräter?',
  22             'description': 'Edward Snowden hat alles aufs Spiel gesetzt, um die weltweite \xdcberwachung durch die Geheimdienste zu enttarnen. Nun stellt sich der ehemalige NSA-Mitarbeiter erstmals weltweit in einem TV-Interview den Fragen eines NDR-Journalisten. Die Sendung vom Sonntagabend.',
  23             'thumbnail': 'http://www.ardmediathek.de/ard/servlet/contentblob/19/28/87/90/19288790/bild/2250037',
  24         },
  25         'skip': 'Blocked outside of Germany',
  26     }
  27
  28     def _real_extract(self, url):
  29         # determine video id from url
  30         m = re.match(self._VALID_URL, url)
  31
  32         numid = re.search(r'documentId=([0-9]+)', url)
  33         if numid:
  34             video_id = numid.group(1)
  35         else:
  36             video_id = m.group('video_id')
  37
  38         webpage = self._download_webpage(url, video_id)
  39
  40         title = self._html_search_regex(
  41             [r'<h1(?:\s+class="boxTopHeadline")?>(.*?)</h1>',
  42              r'<meta name="dcterms.title" content="(.*?)"/>',
  43              r'<h4 class="headline">(.*?)</h4>'],
  44             webpage, 'title')
  45         description = self._html_search_meta(
  46             'dcterms.abstract', webpage, 'description')
  47         thumbnail = self._og_search_thumbnail(webpage)
  48
  49
  50         media_info = self._download_json(
  51             'http://www.ardmediathek.de/play/media/%s' % video_id, video_id)
  52         # The second element of the _mediaArray contains the standard http urls
  53         streams = media_info['_mediaArray'][1]['_mediaStreamArray']
  54         if not streams:
  55             if '"fsk"' in webpage:
  56                 raise ExtractorError('This video is only available after 20:00')
  57
  58         formats = []
  59
  60         for s in streams:
  61             if type(s['_stream']) == list:
  62                 for index, url in enumerate(s['_stream'][::-1]):
  63                     quality = s['_quality'] + index
  64                     formats.append({
  65                         'quality': quality,
  66                         'url': url,
  67                         'format_id': '%s-%s' % (determine_ext(url), quality)
  68                         })
  69                 continue
  70
  71             format = {
  72                 'quality': s['_quality'],
  73                 'url': s['_stream'],
  74             }
  75
  76             format['format_id'] = '%s-%s' % (
  77                 determine_ext(format['url']), format['quality'])
  78
  79             formats.append(format)
  80
  81         self._sort_formats(formats)
  82
  83         return {
  84             'id': video_id,
  85             'title': title,
  86             'description': description,
  87             'formats': formats,
  88             'thumbnail': thumbnail,
  89         }