yt_dlp/extractor/roosterteeth.py

   1 from .common import InfoExtractor
   2 from ..networking.exceptions import HTTPError
   3 from ..utils import (
   4     ExtractorError,
   5     LazyList,
   6     int_or_none,
   7     join_nonempty,
   8     parse_iso8601,
   9     parse_qs,
  10     smuggle_url,
  11     str_or_none,
  12     url_or_none,
  13     urlencode_postdata,
  14     urljoin,
  15 )
  16 from ..utils.traversal import traverse_obj
  17
  18
  19 class RoosterTeethBaseIE(InfoExtractor):
  20     _NETRC_MACHINE = 'roosterteeth'
  21     _API_BASE = 'https://svod-be.roosterteeth.com'
  22     _API_BASE_URL = f'{_API_BASE}/api/v1'
  23
  24     def _perform_login(self, username, password):
  25         if self._get_cookies(self._API_BASE_URL).get('rt_access_token'):
  26             return
  27
  28         try:
  29             self._download_json(
  30                 'https://auth.roosterteeth.com/oauth/token',
  31                 None, 'Logging in', data=urlencode_postdata({
  32                     'client_id': '4338d2b4bdc8db1239360f28e72f0d9ddb1fd01e7a38fbb07b4b1f4ba4564cc5',
  33                     'grant_type': 'password',
  34                     'username': username,
  35                     'password': password,
  36                 }))
  37         except ExtractorError as e:
  38             msg = 'Unable to login'
  39             if isinstance(e.cause, HTTPError) and e.cause.status == 401:
  40                 resp = self._parse_json(e.cause.response.read().decode(), None, fatal=False)
  41                 if resp:
  42                     error = resp.get('extra_info') or resp.get('error_description') or resp.get('error')
  43                     if error:
  44                         msg += ': ' + error
  45             self.report_warning(msg)
  46
  47     def _extract_video_info(self, data):
  48         thumbnails = []
  49         for image in traverse_obj(data, ('included', 'images')):
  50             if image.get('type') not in ('episode_image', 'bonus_feature_image'):
  51                 continue
  52             thumbnails.extend([{
  53                 'id': name,
  54                 'url': url,
  55             } for name, url in (image.get('attributes') or {}).items() if url_or_none(url)])
  56
  57         attributes = data.get('attributes') or {}
  58         title = traverse_obj(attributes, 'title', 'display_title')
  59         sub_only = attributes.get('is_sponsors_only')
  60
  61         episode_id = str_or_none(data.get('uuid'))
  62         video_id = str_or_none(data.get('id'))
  63         if video_id and 'parent_content_id' in attributes:  # parent_content_id is a bonus-only key
  64             video_id += '-bonus'  # there are collisions with bonus ids and regular ids
  65         elif not video_id:
  66             video_id = episode_id
  67
  68         return {
  69             'id': video_id,
  70             'display_id': attributes.get('slug'),
  71             'title': title,
  72             'description': traverse_obj(attributes, 'description', 'caption'),
  73             'series': traverse_obj(attributes, 'show_title', 'parent_content_title'),
  74             'season_number': int_or_none(attributes.get('season_number')),
  75             'season_id': str_or_none(attributes.get('season_id')),
  76             'episode': title,
  77             'episode_number': int_or_none(attributes.get('number')),
  78             'episode_id': episode_id,
  79             'channel_id': attributes.get('channel_id'),
  80             'duration': int_or_none(attributes.get('length')),
  81             'release_timestamp': parse_iso8601(attributes.get('original_air_date')),
  82             'thumbnails': thumbnails,
  83             'availability': self._availability(
  84                 needs_premium=sub_only, needs_subscription=sub_only, needs_auth=sub_only,
  85                 is_private=False, is_unlisted=False),
  86             'tags': attributes.get('genres'),
  87         }
  88
  89
  90 class RoosterTeethIE(RoosterTeethBaseIE):
  91     _VALID_URL = r'https?://(?:.+?\.)?roosterteeth\.com/(?:bonus-feature|episode|watch)/(?P<id>[^/?#&]+)'
  92     _TESTS = [{
  93         'url': 'http://roosterteeth.com/episode/million-dollars-but-season-2-million-dollars-but-the-game-announcement',
  94         'info_dict': {
  95             'id': '9156',
  96             'display_id': 'million-dollars-but-season-2-million-dollars-but-the-game-announcement',
  97             'ext': 'mp4',
  98             'title': 'Million Dollars, But... The Game Announcement',
  99             'description': 'md5:168a54b40e228e79f4ddb141e89fe4f5',
 100             'thumbnail': r're:^https?://.*\.png$',
 101             'series': 'Million Dollars, But...',
 102             'episode': 'Million Dollars, But... The Game Announcement',
 103             'tags': ['Game Show', 'Sketch'],
 104             'season_number': 2,
 105             'availability': 'public',
 106             'episode_number': 10,
 107             'episode_id': '00374575-464e-11e7-a302-065410f210c4',
 108             'season': 'Season 2',
 109             'season_id': 'ffa27d48-464d-11e7-a302-065410f210c4',
 110             'channel_id': '92b6bb21-91d2-4b1b-bf95-3268fa0d9939',
 111             'duration': 145,
 112             'release_timestamp': 1462982400,
 113             'release_date': '20160511',
 114         },
 115         'params': {'skip_download': True},
 116     }, {
 117         'url': 'https://roosterteeth.com/watch/rwby-bonus-25',
 118         'info_dict': {
 119             'id': '40432',
 120             'display_id': 'rwby-bonus-25',
 121             'title': 'Grimm',
 122             'description': 'md5:f30ff570741213418a8d2c19868b93ab',
 123             'episode': 'Grimm',
 124             'channel_id': '92f780eb-ebfe-4bf5-a3b5-c6ad5460a5f1',
 125             'thumbnail': r're:^https?://.*\.(png|jpe?g)$',
 126             'ext': 'mp4',
 127             'availability': 'public',
 128             'episode_id': 'f8117b13-f068-499e-803e-eec9ea2dec8c',
 129             'episode_number': 3,
 130             'tags': ['Animation'],
 131             'season_id': '4b8f0a9e-12c4-41ed-8caa-fed15a85bab8',
 132             'season': 'Season 1',
 133             'series': 'RWBY: World of Remnant',
 134             'season_number': 1,
 135             'duration': 216,
 136             'release_timestamp': 1413489600,
 137             'release_date': '20141016',
 138         },
 139         'params': {'skip_download': True},
 140     }, {
 141         # bonus feature with /watch/ url
 142         'url': 'https://roosterteeth.com/watch/rwby-bonus-21',
 143         'info_dict': {
 144             'id': '33-bonus',
 145             'display_id': 'rwby-bonus-21',
 146             'title': 'Volume 5 Yang Character Short',
 147             'description': 'md5:8c2440bc763ea90c52cfe0a68093e1f7',
 148             'episode': 'Volume 5 Yang Character Short',
 149             'channel_id': '92f780eb-ebfe-4bf5-a3b5-c6ad5460a5f1',
 150             'thumbnail': r're:^https?://.*\.(png|jpe?g)$',
 151             'ext': 'mp4',
 152             'availability': 'public',
 153             'episode_id': 'f2a9f132-1fe2-44ad-8956-63d7c0267720',
 154             'episode_number': 55,
 155             'series': 'RWBY',
 156             'duration': 255,
 157             'release_timestamp': 1507993200,
 158             'release_date': '20171014',
 159         },
 160         'params': {'skip_download': True},
 161     }, {
 162         # only works with video_data['attributes']['url'] m3u8 url
 163         'url': 'https://www.roosterteeth.com/watch/achievement-hunter-achievement-hunter-fatality-walkthrough-deathstroke-lex-luthor-captain-marvel-green-lantern-and-wonder-woman',
 164         'info_dict': {
 165             'id': '25394',
 166             'ext': 'mp4',
 167             'title': 'Fatality Walkthrough: Deathstroke, Lex Luthor, Captain Marvel, Green Lantern, and Wonder Woman',
 168             'description': 'md5:91bb934698344fb9647b1c7351f16964',
 169             'availability': 'public',
 170             'thumbnail': r're:^https?://.*\.(png|jpe?g)$',
 171             'episode': 'Fatality Walkthrough: Deathstroke, Lex Luthor, Captain Marvel, Green Lantern, and Wonder Woman',
 172             'episode_number': 71,
 173             'episode_id': 'ffaec998-464d-11e7-a302-065410f210c4',
 174             'season': 'Season 2008',
 175             'tags': ['Gaming'],
 176             'series': 'Achievement Hunter',
 177             'display_id': 'md5:4465ce4f001735f9d7a2ae529a543d31',
 178             'season_id': 'ffa13340-464d-11e7-a302-065410f210c4',
 179             'season_number': 2008,
 180             'channel_id': '2cb2a70c-be50-46f5-93d7-84a1baabb4f7',
 181             'duration': 189,
 182             'release_timestamp': 1228317300,
 183             'release_date': '20081203',
 184         },
 185         'params': {'skip_download': True},
 186     }, {
 187         # brightcove fallback extraction needed
 188         'url': 'https://roosterteeth.com/watch/lets-play-2013-126',
 189         'info_dict': {
 190             'id': '17845',
 191             'ext': 'mp4',
 192             'title': 'WWE \'13',
 193             'availability': 'public',
 194             'series': 'Let\'s Play',
 195             'episode_number': 10,
 196             'season_id': 'ffa23d9c-464d-11e7-a302-065410f210c4',
 197             'channel_id': '75ba87e8-06fd-4482-bad9-52a4da2c6181',
 198             'episode': 'WWE \'13',
 199             'episode_id': 'ffdbe55e-464d-11e7-a302-065410f210c4',
 200             'thumbnail': r're:^https?://.*\.(png|jpe?g)$',
 201             'tags': ['Gaming', 'Our Favorites'],
 202             'description': 'md5:b4a5226d2bbcf0dafbde11a2ba27262d',
 203             'display_id': 'lets-play-2013-126',
 204             'season_number': 3,
 205             'season': 'Season 3',
 206             'release_timestamp': 1359999840,
 207             'release_date': '20130204',
 208         },
 209         'expected_warnings': ['Direct m3u8 URL returned HTTP Error 403'],
 210         'params': {'skip_download': True},
 211     }, {
 212         'url': 'http://achievementhunter.roosterteeth.com/episode/off-topic-the-achievement-hunter-podcast-2016-i-didn-t-think-it-would-pass-31',
 213         'only_matching': True,
 214     }, {
 215         'url': 'http://funhaus.roosterteeth.com/episode/funhaus-shorts-2016-austin-sucks-funhaus-shorts',
 216         'only_matching': True,
 217     }, {
 218         'url': 'http://screwattack.roosterteeth.com/episode/death-battle-season-3-mewtwo-vs-shadow',
 219         'only_matching': True,
 220     }, {
 221         'url': 'http://theknow.roosterteeth.com/episode/the-know-game-news-season-1-boring-steam-sales-are-better',
 222         'only_matching': True,
 223     }, {
 224         # only available for FIRST members
 225         'url': 'http://roosterteeth.com/episode/rt-docs-the-world-s-greatest-head-massage-the-world-s-greatest-head-massage-an-asmr-journey-part-one',
 226         'only_matching': True,
 227     }, {
 228         'url': 'https://roosterteeth.com/watch/million-dollars-but-season-2-million-dollars-but-the-game-announcement',
 229         'only_matching': True,
 230     }, {
 231         'url': 'https://roosterteeth.com/bonus-feature/camp-camp-soundtrack-another-rap-song-about-foreign-cars-richie-branson',
 232         'only_matching': True,
 233     }]
 234
 235     _BRIGHTCOVE_ACCOUNT_ID = '6203312018001'
 236
 237     def _extract_brightcove_formats_and_subtitles(self, bc_id, url, m3u8_url):
 238         account_id = self._search_regex(
 239             r'/accounts/(\d+)/videos/', m3u8_url, 'account id', default=self._BRIGHTCOVE_ACCOUNT_ID)
 240         info = self._downloader.get_info_extractor('BrightcoveNew').extract(smuggle_url(
 241             f'https://players.brightcove.net/{account_id}/default_default/index.html?videoId={bc_id}',
 242             {'referrer': url}))
 243         return info['formats'], info['subtitles']
 244
 245     def _real_extract(self, url):
 246         display_id = self._match_id(url)
 247         api_episode_url = f'{self._API_BASE_URL}/watch/{display_id}'
 248
 249         try:
 250             video_data = self._download_json(
 251                 api_episode_url + '/videos', display_id, 'Downloading video JSON metadata',
 252                 headers={'Client-Type': 'web'})['data'][0]  # web client-type yields ad-free streams
 253         except ExtractorError as e:
 254             if isinstance(e.cause, HTTPError) and e.cause.status == 403:
 255                 if self._parse_json(e.cause.response.read().decode(), display_id).get('access') is False:
 256                     self.raise_login_required(
 257                         f'{display_id} is only available for FIRST members')
 258             raise
 259
 260         # XXX: additional ad-free URL at video_data['links']['download'] but often gives 403 errors
 261         m3u8_url = video_data['attributes']['url']
 262         is_brightcove = traverse_obj(video_data, ('attributes', 'encoding_pipeline')) == 'brightcove'
 263         bc_id = traverse_obj(video_data, ('attributes', 'uid', {str}))
 264
 265         try:
 266             formats, subtitles = self._extract_m3u8_formats_and_subtitles(
 267                 m3u8_url, display_id, 'mp4', 'm3u8_native', m3u8_id='hls')
 268         except ExtractorError as e:
 269             if is_brightcove and bc_id and isinstance(e.cause, HTTPError) and e.cause.status == 403:
 270                 self.report_warning(
 271                     'Direct m3u8 URL returned HTTP Error 403; retrying with Brightcove extraction')
 272                 formats, subtitles = self._extract_brightcove_formats_and_subtitles(bc_id, url, m3u8_url)
 273             else:
 274                 raise
 275
 276         episode = self._download_json(
 277             api_episode_url, display_id,
 278             'Downloading episode JSON metadata')['data'][0]
 279
 280         return {
 281             'display_id': display_id,
 282             'formats': formats,
 283             'subtitles': subtitles,
 284             **self._extract_video_info(episode),
 285         }
 286
 287
 288 class RoosterTeethSeriesIE(RoosterTeethBaseIE):
 289     _VALID_URL = r'https?://(?:.+?\.)?roosterteeth\.com/series/(?P<id>[^/?#&]+)'
 290     _TESTS = [{
 291         'url': 'https://roosterteeth.com/series/rwby?season=7',
 292         'playlist_count': 13,
 293         'info_dict': {
 294             'id': 'rwby-7',
 295             'title': 'RWBY - Season 7',
 296         },
 297     }, {
 298         'url': 'https://roosterteeth.com/series/the-weird-place',
 299         'playlist_count': 7,
 300         'info_dict': {
 301             'id': 'the-weird-place',
 302             'title': 'The Weird Place',
 303         },
 304     }, {
 305         'url': 'https://roosterteeth.com/series/role-initiative',
 306         'playlist_mincount': 16,
 307         'info_dict': {
 308             'id': 'role-initiative',
 309             'title': 'Role Initiative',
 310         },
 311     }, {
 312         'url': 'https://roosterteeth.com/series/let-s-play-minecraft?season=9',
 313         'playlist_mincount': 50,
 314         'info_dict': {
 315             'id': 'let-s-play-minecraft-9',
 316             'title': 'Let\'s Play Minecraft - Season 9',
 317         },
 318     }]
 319
 320     def _entries(self, series_id, season_number):
 321         display_id = join_nonempty(series_id, season_number)
 322
 323         def yield_episodes(data):
 324             for episode in traverse_obj(data, ('data', lambda _, v: v['canonical_links']['self'])):
 325                 yield self.url_result(
 326                     urljoin('https://www.roosterteeth.com', episode['canonical_links']['self']),
 327                     RoosterTeethIE, **self._extract_video_info(episode))
 328
 329         series_data = self._download_json(
 330             f'{self._API_BASE_URL}/shows/{series_id}/seasons?order=asc&order_by', display_id)
 331         for season_data in traverse_obj(series_data, ('data', lambda _, v: v['links']['episodes'])):
 332             idx = traverse_obj(season_data, ('attributes', 'number'))
 333             if season_number is not None and idx != season_number:
 334                 continue
 335             yield from yield_episodes(self._download_json(
 336                 urljoin(self._API_BASE, season_data['links']['episodes']), display_id,
 337                 f'Downloading season {idx} JSON metadata', query={'per_page': 1000}))
 338
 339         if season_number is None:  # extract series-level bonus features
 340             yield from yield_episodes(self._download_json(
 341                 f'{self._API_BASE_URL}/shows/{series_id}/bonus_features?order=asc&order_by&per_page=1000',
 342                 display_id, 'Downloading bonus features JSON metadata', fatal=False))
 343
 344     def _real_extract(self, url):
 345         series_id = self._match_id(url)
 346         season_number = traverse_obj(parse_qs(url), ('season', 0), expected_type=int_or_none)
 347
 348         entries = LazyList(self._entries(series_id, season_number))
 349         return self.playlist_result(
 350             entries,
 351             join_nonempty(series_id, season_number),
 352             join_nonempty(entries[0].get('series'), season_number, delim=' - Season '))