yt_dlp/extractor/discoverygo.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     ExtractorError,
   6     determine_ext,
   7     extract_attributes,
   8     int_or_none,
   9     parse_age_limit,
  10     remove_end,
  11     unescapeHTML,
  12     url_or_none,
  13 )
  14
  15
  16 class DiscoveryGoBaseIE(InfoExtractor):
  17     _VALID_URL_TEMPLATE = r'''(?x)https?://(?:www\.)?(?:
  18             discovery|
  19             investigationdiscovery|
  20             discoverylife|
  21             animalplanet|
  22             ahctv|
  23             destinationamerica|
  24             sciencechannel|
  25             tlc|
  26             velocitychannel
  27         )go\.com/%s(?P<id>[^/?#&]+)'''
  28
  29     def _extract_video_info(self, video, stream, display_id):
  30         title = video['name']
  31
  32         if not stream:
  33             if video.get('authenticated') is True:
  34                 raise ExtractorError(
  35                     'This video is only available via cable service provider subscription that'
  36                     ' is not currently supported. You may want to use --cookies.', expected=True)
  37             else:
  38                 raise ExtractorError('Unable to find stream')
  39         STREAM_URL_SUFFIX = 'streamUrl'
  40         formats = []
  41         for stream_kind in ('', 'hds'):
  42             suffix = STREAM_URL_SUFFIX.capitalize() if stream_kind else STREAM_URL_SUFFIX
  43             stream_url = stream.get('%s%s' % (stream_kind, suffix))
  44             if not stream_url:
  45                 continue
  46             if stream_kind == '':
  47                 formats.extend(self._extract_m3u8_formats(
  48                     stream_url, display_id, 'mp4', entry_protocol='m3u8_native',
  49                     m3u8_id='hls', fatal=False))
  50             elif stream_kind == 'hds':
  51                 formats.extend(self._extract_f4m_formats(
  52                     stream_url, display_id, f4m_id=stream_kind, fatal=False))
  53
  54         video_id = video.get('id') or display_id
  55         description = video.get('description', {}).get('detailed')
  56         duration = int_or_none(video.get('duration'))
  57
  58         series = video.get('show', {}).get('name')
  59         season_number = int_or_none(video.get('season', {}).get('number'))
  60         episode_number = int_or_none(video.get('episodeNumber'))
  61
  62         tags = video.get('tags')
  63         age_limit = parse_age_limit(video.get('parental', {}).get('rating'))
  64
  65         subtitles = {}
  66         captions = stream.get('captions')
  67         if isinstance(captions, list):
  68             for caption in captions:
  69                 subtitle_url = url_or_none(caption.get('fileUrl'))
  70                 if not subtitle_url or not subtitle_url.startswith('http'):
  71                     continue
  72                 lang = caption.get('fileLang', 'en')
  73                 ext = determine_ext(subtitle_url)
  74                 subtitles.setdefault(lang, []).append({
  75                     'url': subtitle_url,
  76                     'ext': 'ttml' if ext == 'xml' else ext,
  77                 })
  78
  79         return {
  80             'id': video_id,
  81             'display_id': display_id,
  82             'title': title,
  83             'description': description,
  84             'duration': duration,
  85             'series': series,
  86             'season_number': season_number,
  87             'episode_number': episode_number,
  88             'tags': tags,
  89             'age_limit': age_limit,
  90             'formats': formats,
  91             'subtitles': subtitles,
  92         }
  93
  94
  95 class DiscoveryGoIE(DiscoveryGoBaseIE):
  96     _VALID_URL = DiscoveryGoBaseIE._VALID_URL_TEMPLATE % r'(?:[^/]+/)+'
  97     _GEO_COUNTRIES = ['US']
  98     _TEST = {
  99         'url': 'https://www.discoverygo.com/bering-sea-gold/reaper-madness/',
 100         'info_dict': {
 101             'id': '58c167d86b66d12f2addeb01',
 102             'ext': 'mp4',
 103             'title': 'Reaper Madness',
 104             'description': 'md5:09f2c625c99afb8946ed4fb7865f6e78',
 105             'duration': 2519,
 106             'series': 'Bering Sea Gold',
 107             'season_number': 8,
 108             'episode_number': 6,
 109             'age_limit': 14,
 110         },
 111     }
 112
 113     def _real_extract(self, url):
 114         display_id = self._match_id(url)
 115
 116         webpage = self._download_webpage(url, display_id)
 117
 118         container = extract_attributes(
 119             self._search_regex(
 120                 r'(<div[^>]+class=["\']video-player-container[^>]+>)',
 121                 webpage, 'video container'))
 122
 123         video = self._parse_json(
 124             container.get('data-video') or container.get('data-json'),
 125             display_id)
 126
 127         stream = video.get('stream')
 128
 129         return self._extract_video_info(video, stream, display_id)
 130
 131
 132 class DiscoveryGoPlaylistIE(DiscoveryGoBaseIE):
 133     _VALID_URL = DiscoveryGoBaseIE._VALID_URL_TEMPLATE % ''
 134     _TEST = {
 135         'url': 'https://www.discoverygo.com/bering-sea-gold/',
 136         'info_dict': {
 137             'id': 'bering-sea-gold',
 138             'title': 'Bering Sea Gold',
 139             'description': 'md5:cc5c6489835949043c0cc3ad66c2fa0e',
 140         },
 141         'playlist_mincount': 6,
 142     }
 143
 144     @classmethod
 145     def suitable(cls, url):
 146         return False if DiscoveryGoIE.suitable(url) else super(
 147             DiscoveryGoPlaylistIE, cls).suitable(url)
 148
 149     def _real_extract(self, url):
 150         display_id = self._match_id(url)
 151
 152         webpage = self._download_webpage(url, display_id)
 153
 154         entries = []
 155         for mobj in re.finditer(r'data-json=(["\'])(?P<json>{.+?})\1', webpage):
 156             data = self._parse_json(
 157                 mobj.group('json'), display_id,
 158                 transform_source=unescapeHTML, fatal=False)
 159             if not isinstance(data, dict) or data.get('type') != 'episode':
 160                 continue
 161             episode_url = data.get('socialUrl')
 162             if not episode_url:
 163                 continue
 164             entries.append(self.url_result(
 165                 episode_url, ie=DiscoveryGoIE.ie_key(),
 166                 video_id=data.get('id')))
 167
 168         return self.playlist_result(
 169             entries, display_id,
 170             remove_end(self._og_search_title(
 171                 webpage, fatal=False), ' | Discovery GO'),
 172             self._og_search_description(webpage))