yt_dlp/extractor/servus.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     determine_ext,
   4     float_or_none,
   5     int_or_none,
   6     unified_timestamp,
   7     urlencode_postdata,
   8     url_or_none,
   9 )
  10
  11
  12 class ServusIE(InfoExtractor):
  13     _VALID_URL = r'''(?x)
  14                     https?://
  15                         (?:www\.)?
  16                         (?:
  17                             servus\.com/(?:(?:at|de)/p/[^/]+|tv/videos)|
  18                             (?:servustv|pm-wissen)\.com/videos
  19                         )
  20                         /(?P<id>[aA]{2}-\w+|\d+-\d+)
  21                     '''
  22     _TESTS = [{
  23         # new URL schema
  24         'url': 'https://www.servustv.com/videos/aa-1t6vbu5pw1w12/',
  25         'md5': '60474d4c21f3eb148838f215c37f02b9',
  26         'info_dict': {
  27             'id': 'AA-1T6VBU5PW1W12',
  28             'ext': 'mp4',
  29             'title': 'Die Grünen aus Sicht des Volkes',
  30             'alt_title': 'Talk im Hangar-7 Voxpops Gruene',
  31             'description': 'md5:1247204d85783afe3682644398ff2ec4',
  32             'thumbnail': r're:^https?://.*\.jpg',
  33             'duration': 62.442,
  34             'timestamp': 1605193976,
  35             'upload_date': '20201112',
  36             'series': 'Talk im Hangar-7',
  37             'season': 'Season 9',
  38             'season_number': 9,
  39             'episode': 'Episode 31 - September 14',
  40             'episode_number': 31,
  41         }
  42     }, {
  43         # old URL schema
  44         'url': 'https://www.servus.com/de/p/Die-Gr%C3%BCnen-aus-Sicht-des-Volkes/AA-1T6VBU5PW1W12/',
  45         'only_matching': True,
  46     }, {
  47         'url': 'https://www.servus.com/at/p/Wie-das-Leben-beginnt/1309984137314-381415152/',
  48         'only_matching': True,
  49     }, {
  50         'url': 'https://www.servus.com/tv/videos/aa-1t6vbu5pw1w12/',
  51         'only_matching': True,
  52     }, {
  53         'url': 'https://www.servus.com/tv/videos/1380889096408-1235196658/',
  54         'only_matching': True,
  55     }, {
  56         'url': 'https://www.pm-wissen.com/videos/aa-24mus4g2w2112/',
  57         'only_matching': True,
  58     }]
  59
  60     def _real_extract(self, url):
  61         video_id = self._match_id(url).upper()
  62
  63         token = self._download_json(
  64             'https://auth.redbullmediahouse.com/token', video_id,
  65             'Downloading token', data=urlencode_postdata({
  66                 'grant_type': 'client_credentials',
  67             }), headers={
  68                 'Authorization': 'Basic SVgtMjJYNEhBNFdEM1cxMTpEdDRVSkFLd2ZOMG5IMjB1NGFBWTBmUFpDNlpoQ1EzNA==',
  69             })
  70         access_token = token['access_token']
  71         token_type = token.get('token_type', 'Bearer')
  72
  73         video = self._download_json(
  74             'https://sparkle-api.liiift.io/api/v1/stv/channels/international/assets/%s' % video_id,
  75             video_id, 'Downloading video JSON', headers={
  76                 'Authorization': '%s %s' % (token_type, access_token),
  77             })
  78
  79         formats = []
  80         thumbnail = None
  81         for resource in video['resources']:
  82             if not isinstance(resource, dict):
  83                 continue
  84             format_url = url_or_none(resource.get('url'))
  85             if not format_url:
  86                 continue
  87             extension = resource.get('extension')
  88             type_ = resource.get('type')
  89             if extension == 'jpg' or type_ == 'reference_keyframe':
  90                 thumbnail = format_url
  91                 continue
  92             ext = determine_ext(format_url)
  93             if type_ == 'dash' or ext == 'mpd':
  94                 formats.extend(self._extract_mpd_formats(
  95                     format_url, video_id, mpd_id='dash', fatal=False))
  96             elif type_ == 'hls' or ext == 'm3u8':
  97                 formats.extend(self._extract_m3u8_formats(
  98                     format_url, video_id, 'mp4', entry_protocol='m3u8_native',
  99                     m3u8_id='hls', fatal=False))
 100             elif extension == 'mp4' or ext == 'mp4':
 101                 formats.append({
 102                     'url': format_url,
 103                     'format_id': type_,
 104                     'width': int_or_none(resource.get('width')),
 105                     'height': int_or_none(resource.get('height')),
 106                 })
 107
 108         attrs = {}
 109         for attribute in video['attributes']:
 110             if not isinstance(attribute, dict):
 111                 continue
 112             key = attribute.get('fieldKey')
 113             value = attribute.get('fieldValue')
 114             if not key or not value:
 115                 continue
 116             attrs[key] = value
 117
 118         title = attrs.get('title_stv') or video_id
 119         alt_title = attrs.get('title')
 120         description = attrs.get('long_description') or attrs.get('short_description')
 121         series = attrs.get('label')
 122         season = attrs.get('season')
 123         episode = attrs.get('chapter')
 124         duration = float_or_none(attrs.get('duration'), scale=1000)
 125         season_number = int_or_none(self._search_regex(
 126             r'Season (\d+)', season or '', 'season number', default=None))
 127         episode_number = int_or_none(self._search_regex(
 128             r'Episode (\d+)', episode or '', 'episode number', default=None))
 129
 130         return {
 131             'id': video_id,
 132             'title': title,
 133             'alt_title': alt_title,
 134             'description': description,
 135             'thumbnail': thumbnail,
 136             'duration': duration,
 137             'timestamp': unified_timestamp(video.get('lastPublished')),
 138             'series': series,
 139             'season': season,
 140             'season_number': season_number,
 141             'episode': episode,
 142             'episode_number': episode_number,
 143             'formats': formats,
 144         }