yt_dlp/extractor/crunchyroll.py

   1 import base64
   2 import uuid
   3
   4 from .common import InfoExtractor
   5 from ..networking import Request
   6 from ..networking.exceptions import HTTPError
   7 from ..utils import (
   8     ExtractorError,
   9     float_or_none,
  10     format_field,
  11     int_or_none,
  12     jwt_decode_hs256,
  13     parse_age_limit,
  14     parse_count,
  15     parse_iso8601,
  16     qualities,
  17     time_seconds,
  18     traverse_obj,
  19     url_or_none,
  20     urlencode_postdata,
  21 )
  22
  23
  24 class CrunchyrollBaseIE(InfoExtractor):
  25     _BASE_URL = 'https://www.crunchyroll.com'
  26     _API_BASE = 'https://api.crunchyroll.com'
  27     _NETRC_MACHINE = 'crunchyroll'
  28     _SWITCH_USER_AGENT = 'Crunchyroll/1.8.0 Nintendo Switch/12.3.12.0 UE4/4.27'
  29     _REFRESH_TOKEN = None
  30     _AUTH_HEADERS = None
  31     _AUTH_EXPIRY = None
  32     _API_ENDPOINT = None
  33     _BASIC_AUTH = 'Basic ' + base64.b64encode(':'.join((
  34         't-kdgp2h8c3jub8fn0fq',
  35         'yfLDfMfrYvKXh4JXS1LEI2cCqu1v5Wan',
  36     )).encode()).decode()
  37     _IS_PREMIUM = None
  38     _LOCALE_LOOKUP = {
  39         'ar': 'ar-SA',
  40         'de': 'de-DE',
  41         '': 'en-US',
  42         'es': 'es-419',
  43         'es-es': 'es-ES',
  44         'fr': 'fr-FR',
  45         'it': 'it-IT',
  46         'pt-br': 'pt-BR',
  47         'pt-pt': 'pt-PT',
  48         'ru': 'ru-RU',
  49         'hi': 'hi-IN',
  50     }
  51
  52     def _set_auth_info(self, response):
  53         CrunchyrollBaseIE._IS_PREMIUM = 'cr_premium' in traverse_obj(response, ('access_token', {jwt_decode_hs256}, 'benefits', ...))
  54         CrunchyrollBaseIE._AUTH_HEADERS = {'Authorization': response['token_type'] + ' ' + response['access_token']}
  55         CrunchyrollBaseIE._AUTH_EXPIRY = time_seconds(seconds=traverse_obj(response, ('expires_in', {float_or_none}), default=300) - 10)
  56
  57     def _request_token(self, headers, data, note='Requesting token', errnote='Failed to request token'):
  58         try:
  59             return self._download_json(
  60                 f'{self._BASE_URL}/auth/v1/token', None, note=note, errnote=errnote,
  61                 headers=headers, data=urlencode_postdata(data), impersonate=True)
  62         except ExtractorError as error:
  63             if not isinstance(error.cause, HTTPError) or error.cause.status != 403:
  64                 raise
  65             if target := error.cause.response.extensions.get('impersonate'):
  66                 raise ExtractorError(f'Got HTTP Error 403 when using impersonate target "{target}"')
  67             raise ExtractorError(
  68                 'Request blocked by Cloudflare. '
  69                 'Install the required impersonation dependency if possible, '
  70                 'or else navigate to Crunchyroll in your browser, '
  71                 'then pass the fresh cookies (with --cookies-from-browser or --cookies) '
  72                 'and your browser\'s User-Agent (with --user-agent)', expected=True)
  73
  74     def _perform_login(self, username, password):
  75         if not CrunchyrollBaseIE._REFRESH_TOKEN:
  76             CrunchyrollBaseIE._REFRESH_TOKEN = self.cache.load(self._NETRC_MACHINE, username)
  77         if CrunchyrollBaseIE._REFRESH_TOKEN:
  78             return
  79
  80         try:
  81             login_response = self._request_token(
  82                 headers={'Authorization': self._BASIC_AUTH}, data={
  83                     'username': username,
  84                     'password': password,
  85                     'grant_type': 'password',
  86                     'scope': 'offline_access',
  87                 }, note='Logging in', errnote='Failed to log in')
  88         except ExtractorError as error:
  89             if isinstance(error.cause, HTTPError) and error.cause.status == 401:
  90                 raise ExtractorError('Invalid username and/or password', expected=True)
  91             raise
  92
  93         CrunchyrollBaseIE._REFRESH_TOKEN = login_response['refresh_token']
  94         self.cache.store(self._NETRC_MACHINE, username, CrunchyrollBaseIE._REFRESH_TOKEN)
  95         self._set_auth_info(login_response)
  96
  97     def _update_auth(self):
  98         if CrunchyrollBaseIE._AUTH_HEADERS and CrunchyrollBaseIE._AUTH_EXPIRY > time_seconds():
  99             return
 100
 101         auth_headers = {'Authorization': self._BASIC_AUTH}
 102         if CrunchyrollBaseIE._REFRESH_TOKEN:
 103             data = {
 104                 'refresh_token': CrunchyrollBaseIE._REFRESH_TOKEN,
 105                 'grant_type': 'refresh_token',
 106                 'scope': 'offline_access',
 107             }
 108         else:
 109             data = {'grant_type': 'client_id'}
 110             auth_headers['ETP-Anonymous-ID'] = uuid.uuid4()
 111         try:
 112             auth_response = self._request_token(auth_headers, data)
 113         except ExtractorError as error:
 114             username, password = self._get_login_info()
 115             if not username or not isinstance(error.cause, HTTPError) or error.cause.status != 400:
 116                 raise
 117             self.to_screen('Refresh token has expired. Re-logging in')
 118             CrunchyrollBaseIE._REFRESH_TOKEN = None
 119             self.cache.store(self._NETRC_MACHINE, username, None)
 120             self._perform_login(username, password)
 121             return
 122
 123         self._set_auth_info(auth_response)
 124
 125     def _locale_from_language(self, language):
 126         config_locale = self._configuration_arg('metadata', ie_key=CrunchyrollBetaIE, casesense=True)
 127         return config_locale[0] if config_locale else self._LOCALE_LOOKUP.get(language)
 128
 129     def _call_base_api(self, endpoint, internal_id, lang, note=None, query={}):
 130         self._update_auth()
 131
 132         if not endpoint.startswith('/'):
 133             endpoint = f'/{endpoint}'
 134
 135         query = query.copy()
 136         locale = self._locale_from_language(lang)
 137         if locale:
 138             query['locale'] = locale
 139
 140         return self._download_json(
 141             f'{self._BASE_URL}{endpoint}', internal_id, note or f'Calling API: {endpoint}',
 142             headers=CrunchyrollBaseIE._AUTH_HEADERS, query=query)
 143
 144     def _call_api(self, path, internal_id, lang, note='api', query={}):
 145         if not path.startswith(f'/content/v2/{self._API_ENDPOINT}/'):
 146             path = f'/content/v2/{self._API_ENDPOINT}/{path}'
 147
 148         try:
 149             result = self._call_base_api(
 150                 path, internal_id, lang, f'Downloading {note} JSON ({self._API_ENDPOINT})', query=query)
 151         except ExtractorError as error:
 152             if isinstance(error.cause, HTTPError) and error.cause.status == 404:
 153                 return None
 154             raise
 155
 156         if not result:
 157             raise ExtractorError(f'Unexpected response when downloading {note} JSON')
 158         return result
 159
 160     def _extract_chapters(self, internal_id):
 161         # if no skip events are available, a 403 xml error is returned
 162         skip_events = self._download_json(
 163             f'https://static.crunchyroll.com/skip-events/production/{internal_id}.json',
 164             internal_id, note='Downloading chapter info', fatal=False, errnote=False)
 165         if not skip_events:
 166             return None
 167
 168         chapters = []
 169         for event in ('recap', 'intro', 'credits', 'preview'):
 170             start = traverse_obj(skip_events, (event, 'start', {float_or_none}))
 171             end = traverse_obj(skip_events, (event, 'end', {float_or_none}))
 172             # some chapters have no start and/or ending time, they will just be ignored
 173             if start is None or end is None:
 174                 continue
 175             chapters.append({'title': event.capitalize(), 'start_time': start, 'end_time': end})
 176
 177         return chapters
 178
 179     def _extract_stream(self, identifier, display_id=None):
 180         if not display_id:
 181             display_id = identifier
 182
 183         self._update_auth()
 184         headers = {**CrunchyrollBaseIE._AUTH_HEADERS, 'User-Agent': self._SWITCH_USER_AGENT}
 185         try:
 186             stream_response = self._download_json(
 187                 f'https://cr-play-service.prd.crunchyrollsvc.com/v1/{identifier}/console/switch/play',
 188                 display_id, note='Downloading stream info', errnote='Failed to download stream info', headers=headers)
 189         except ExtractorError as error:
 190             if self.get_param('ignore_no_formats_error'):
 191                 self.report_warning(error.orig_msg)
 192                 return [], {}
 193             elif isinstance(error.cause, HTTPError) and error.cause.status == 420:
 194                 raise ExtractorError(
 195                     'You have reached the rate-limit for active streams; try again later', expected=True)
 196             raise
 197
 198         available_formats = {'': ('', '', stream_response['url'])}
 199         for hardsub_lang, stream in traverse_obj(stream_response, ('hardSubs', {dict.items}, lambda _, v: v[1]['url'])):
 200             available_formats[hardsub_lang] = (f'hardsub-{hardsub_lang}', hardsub_lang, stream['url'])
 201
 202         requested_hardsubs = [('' if val == 'none' else val) for val in (self._configuration_arg('hardsub') or ['none'])]
 203         hardsub_langs = [lang for lang in available_formats if lang]
 204         if hardsub_langs and 'all' not in requested_hardsubs:
 205             full_format_langs = set(requested_hardsubs)
 206             self.to_screen(f'Available hardsub languages: {", ".join(hardsub_langs)}')
 207             self.to_screen(
 208                 'To extract formats of a hardsub language, use '
 209                 '"--extractor-args crunchyrollbeta:hardsub=<language_code or all>". '
 210                 'See https://github.com/yt-dlp/yt-dlp#crunchyrollbeta-crunchyroll for more info',
 211                 only_once=True)
 212         else:
 213             full_format_langs = set(map(str.lower, available_formats))
 214
 215         audio_locale = traverse_obj(stream_response, ('audioLocale', {str}))
 216         hardsub_preference = qualities(requested_hardsubs[::-1])
 217         formats, subtitles = [], {}
 218         for format_id, hardsub_lang, stream_url in available_formats.values():
 219             if hardsub_lang.lower() in full_format_langs:
 220                 adaptive_formats, dash_subs = self._extract_mpd_formats_and_subtitles(
 221                     stream_url, display_id, mpd_id=format_id, headers=CrunchyrollBaseIE._AUTH_HEADERS,
 222                     fatal=False, note=f'Downloading {f"{format_id} " if hardsub_lang else ""}MPD manifest')
 223                 self._merge_subtitles(dash_subs, target=subtitles)
 224             else:
 225                 continue  # XXX: Update this if meta mpd formats work; will be tricky with token invalidation
 226             for f in adaptive_formats:
 227                 if f.get('acodec') != 'none':
 228                     f['language'] = audio_locale
 229                 f['quality'] = hardsub_preference(hardsub_lang.lower())
 230             formats.extend(adaptive_formats)
 231
 232         for locale, subtitle in traverse_obj(stream_response, (('subtitles', 'captions'), {dict.items}, ...)):
 233             subtitles.setdefault(locale, []).append(traverse_obj(subtitle, {'url': 'url', 'ext': 'format'}))
 234
 235         # Invalidate stream token to avoid rate-limit
 236         error_msg = 'Unable to invalidate stream token; you may experience rate-limiting'
 237         if stream_token := stream_response.get('token'):
 238             self._request_webpage(Request(
 239                 f'https://cr-play-service.prd.crunchyrollsvc.com/v1/token/{identifier}/{stream_token}/inactive',
 240                 headers=headers, method='PATCH'), display_id, 'Invalidating stream token', error_msg, fatal=False)
 241         else:
 242             self.report_warning(error_msg)
 243
 244         return formats, subtitles
 245
 246
 247 class CrunchyrollCmsBaseIE(CrunchyrollBaseIE):
 248     _API_ENDPOINT = 'cms'
 249     _CMS_EXPIRY = None
 250
 251     def _call_cms_api_signed(self, path, internal_id, lang, note='api'):
 252         if not CrunchyrollCmsBaseIE._CMS_EXPIRY or CrunchyrollCmsBaseIE._CMS_EXPIRY <= time_seconds():
 253             response = self._call_base_api('index/v2', None, lang, 'Retrieving signed policy')['cms_web']
 254             CrunchyrollCmsBaseIE._CMS_QUERY = {
 255                 'Policy': response['policy'],
 256                 'Signature': response['signature'],
 257                 'Key-Pair-Id': response['key_pair_id'],
 258             }
 259             CrunchyrollCmsBaseIE._CMS_BUCKET = response['bucket']
 260             CrunchyrollCmsBaseIE._CMS_EXPIRY = parse_iso8601(response['expires']) - 10
 261
 262         if not path.startswith('/cms/v2'):
 263             path = f'/cms/v2{CrunchyrollCmsBaseIE._CMS_BUCKET}/{path}'
 264
 265         return self._call_base_api(
 266             path, internal_id, lang, f'Downloading {note} JSON (signed cms)', query=CrunchyrollCmsBaseIE._CMS_QUERY)
 267
 268
 269 class CrunchyrollBetaIE(CrunchyrollCmsBaseIE):
 270     IE_NAME = 'crunchyroll'
 271     _VALID_URL = r'''(?x)
 272         https?://(?:beta\.|www\.)?crunchyroll\.com/
 273         (?:(?P<lang>\w{2}(?:-\w{2})?)/)?
 274         watch/(?!concert|musicvideo)(?P<id>\w+)'''
 275     _TESTS = [{
 276         # Premium only
 277         'url': 'https://www.crunchyroll.com/watch/GY2P1Q98Y/to-the-future',
 278         'info_dict': {
 279             'id': 'GY2P1Q98Y',
 280             'ext': 'mp4',
 281             'duration': 1380.241,
 282             'timestamp': 1459632600,
 283             'description': 'md5:a022fbec4fbb023d43631032c91ed64b',
 284             'title': 'World Trigger Episode 73 – To the Future',
 285             'upload_date': '20160402',
 286             'series': 'World Trigger',
 287             'series_id': 'GR757DMKY',
 288             'season': 'World Trigger',
 289             'season_id': 'GR9P39NJ6',
 290             'season_number': 1,
 291             'episode': 'To the Future',
 292             'episode_number': 73,
 293             'thumbnail': r're:^https://www.crunchyroll.com/imgsrv/.*\.jpeg?$',
 294             'chapters': 'count:2',
 295             'age_limit': 14,
 296             'like_count': int,
 297             'dislike_count': int,
 298         },
 299         'params': {
 300             'skip_download': 'm3u8',
 301             'extractor_args': {'crunchyrollbeta': {'hardsub': ['de-DE']}},
 302             'format': 'bv[format_id~=hardsub]',
 303         },
 304     }, {
 305         # Premium only
 306         'url': 'https://www.crunchyroll.com/watch/GYE5WKQGR',
 307         'info_dict': {
 308             'id': 'GYE5WKQGR',
 309             'ext': 'mp4',
 310             'duration': 366.459,
 311             'timestamp': 1476788400,
 312             'description': 'md5:74b67283ffddd75f6e224ca7dc031e76',
 313             'title': 'SHELTER – Porter Robinson presents Shelter the Animation',
 314             'upload_date': '20161018',
 315             'series': 'SHELTER',
 316             'series_id': 'GYGG09WWY',
 317             'season': 'SHELTER',
 318             'season_id': 'GR09MGK4R',
 319             'season_number': 1,
 320             'episode': 'Porter Robinson presents Shelter the Animation',
 321             'episode_number': 0,
 322             'thumbnail': r're:^https://www.crunchyroll.com/imgsrv/.*\.jpeg?$',
 323             'age_limit': 14,
 324             'like_count': int,
 325             'dislike_count': int,
 326         },
 327         'params': {'skip_download': True},
 328     }, {
 329         'url': 'https://www.crunchyroll.com/watch/GJWU2VKK3/cherry-blossom-meeting-and-a-coming-blizzard',
 330         'info_dict': {
 331             'id': 'GJWU2VKK3',
 332             'ext': 'mp4',
 333             'duration': 1420.054,
 334             'description': 'md5:2d1c67c0ec6ae514d9c30b0b99a625cd',
 335             'title': 'The Ice Guy and His Cool Female Colleague Episode 1 – Cherry Blossom Meeting and a Coming Blizzard',
 336             'series': 'The Ice Guy and His Cool Female Colleague',
 337             'series_id': 'GW4HM75NP',
 338             'season': 'The Ice Guy and His Cool Female Colleague',
 339             'season_id': 'GY9PC21VE',
 340             'season_number': 1,
 341             'episode': 'Cherry Blossom Meeting and a Coming Blizzard',
 342             'episode_number': 1,
 343             'chapters': 'count:2',
 344             'thumbnail': r're:^https://www.crunchyroll.com/imgsrv/.*\.jpeg?$',
 345             'timestamp': 1672839000,
 346             'upload_date': '20230104',
 347             'age_limit': 14,
 348             'like_count': int,
 349             'dislike_count': int,
 350         },
 351         'params': {'skip_download': 'm3u8'},
 352     }, {
 353         'url': 'https://www.crunchyroll.com/watch/GM8F313NQ',
 354         'info_dict': {
 355             'id': 'GM8F313NQ',
 356             'ext': 'mp4',
 357             'title': 'Garakowa -Restore the World-',
 358             'description': 'md5:8d2f8b6b9dd77d87810882e7d2ee5608',
 359             'duration': 3996.104,
 360             'age_limit': 13,
 361             'thumbnail': r're:^https://www.crunchyroll.com/imgsrv/.*\.jpeg?$',
 362         },
 363         'params': {'skip_download': 'm3u8'},
 364         'skip': 'no longer exists',
 365     }, {
 366         'url': 'https://www.crunchyroll.com/watch/G62PEZ2E6',
 367         'info_dict': {
 368             'id': 'G62PEZ2E6',
 369             'description': 'md5:8d2f8b6b9dd77d87810882e7d2ee5608',
 370             'age_limit': 13,
 371             'duration': 65.138,
 372             'title': 'Garakowa -Restore the World-',
 373         },
 374         'playlist_mincount': 5,
 375     }, {
 376         'url': 'https://www.crunchyroll.com/de/watch/GY2P1Q98Y',
 377         'only_matching': True,
 378     }, {
 379         'url': 'https://beta.crunchyroll.com/pt-br/watch/G8WUN8VKP/the-ruler-of-conspiracy',
 380         'only_matching': True,
 381     }]
 382     # We want to support lazy playlist filtering and movie listings cannot be inside a playlist
 383     _RETURN_TYPE = 'video'
 384
 385     def _real_extract(self, url):
 386         lang, internal_id = self._match_valid_url(url).group('lang', 'id')
 387
 388         # We need to use unsigned API call to allow ratings query string
 389         response = traverse_obj(self._call_api(
 390             f'objects/{internal_id}', internal_id, lang, 'object info', {'ratings': 'true'}), ('data', 0, {dict}))
 391         if not response:
 392             raise ExtractorError(f'No video with id {internal_id} could be found (possibly region locked?)', expected=True)
 393
 394         object_type = response.get('type')
 395         if object_type == 'episode':
 396             result = self._transform_episode_response(response)
 397
 398         elif object_type == 'movie':
 399             result = self._transform_movie_response(response)
 400
 401         elif object_type == 'movie_listing':
 402             first_movie_id = traverse_obj(response, ('movie_listing_metadata', 'first_movie_id'))
 403             if not self._yes_playlist(internal_id, first_movie_id):
 404                 return self.url_result(f'{self._BASE_URL}/{lang}watch/{first_movie_id}', CrunchyrollBetaIE, first_movie_id)
 405
 406             def entries():
 407                 movies = self._call_api(f'movie_listings/{internal_id}/movies', internal_id, lang, 'movie list')
 408                 for movie_response in traverse_obj(movies, ('data', ...)):
 409                     yield self.url_result(
 410                         f'{self._BASE_URL}/{lang}watch/{movie_response["id"]}',
 411                         CrunchyrollBetaIE, **self._transform_movie_response(movie_response))
 412
 413             return self.playlist_result(entries(), **self._transform_movie_response(response))
 414
 415         else:
 416             raise ExtractorError(f'Unknown object type {object_type}')
 417
 418         if not self._IS_PREMIUM and traverse_obj(response, (f'{object_type}_metadata', 'is_premium_only')):
 419             message = f'This {object_type} is for premium members only'
 420             if CrunchyrollBaseIE._REFRESH_TOKEN:
 421                 self.raise_no_formats(message, expected=True, video_id=internal_id)
 422             else:
 423                 self.raise_login_required(message, method='password', metadata_available=True)
 424         else:
 425             result['formats'], result['subtitles'] = self._extract_stream(internal_id)
 426
 427         result['chapters'] = self._extract_chapters(internal_id)
 428
 429         def calculate_count(item):
 430             return parse_count(''.join((item['displayed'], item.get('unit') or '')))
 431
 432         result.update(traverse_obj(response, ('rating', {
 433             'like_count': ('up', {calculate_count}),
 434             'dislike_count': ('down', {calculate_count}),
 435         })))
 436
 437         return result
 438
 439     @staticmethod
 440     def _transform_episode_response(data):
 441         metadata = traverse_obj(data, (('episode_metadata', None), {dict}), get_all=False) or {}
 442         return {
 443             'id': data['id'],
 444             'title': ' \u2013 '.join((
 445                 ('%s%s' % (
 446                     format_field(metadata, 'season_title'),
 447                     format_field(metadata, 'episode', ' Episode %s'))),
 448                 format_field(data, 'title'))),
 449             **traverse_obj(data, {
 450                 'episode': ('title', {str}),
 451                 'description': ('description', {str}, {lambda x: x.replace(r'\r\n', '\n')}),
 452                 'thumbnails': ('images', 'thumbnail', ..., ..., {
 453                     'url': ('source', {url_or_none}),
 454                     'width': ('width', {int_or_none}),
 455                     'height': ('height', {int_or_none}),
 456                 }),
 457             }),
 458             **traverse_obj(metadata, {
 459                 'duration': ('duration_ms', {lambda x: float_or_none(x, 1000)}),
 460                 'timestamp': ('upload_date', {parse_iso8601}),
 461                 'series': ('series_title', {str}),
 462                 'series_id': ('series_id', {str}),
 463                 'season': ('season_title', {str}),
 464                 'season_id': ('season_id', {str}),
 465                 'season_number': ('season_number', ({int}, {float_or_none})),
 466                 'episode_number': ('sequence_number', ({int}, {float_or_none})),
 467                 'age_limit': ('maturity_ratings', -1, {parse_age_limit}),
 468                 'language': ('audio_locale', {str}),
 469             }, get_all=False),
 470         }
 471
 472     @staticmethod
 473     def _transform_movie_response(data):
 474         metadata = traverse_obj(data, (('movie_metadata', 'movie_listing_metadata', None), {dict}), get_all=False) or {}
 475         return {
 476             'id': data['id'],
 477             **traverse_obj(data, {
 478                 'title': ('title', {str}),
 479                 'description': ('description', {str}, {lambda x: x.replace(r'\r\n', '\n')}),
 480                 'thumbnails': ('images', 'thumbnail', ..., ..., {
 481                     'url': ('source', {url_or_none}),
 482                     'width': ('width', {int_or_none}),
 483                     'height': ('height', {int_or_none}),
 484                 }),
 485             }),
 486             **traverse_obj(metadata, {
 487                 'duration': ('duration_ms', {lambda x: float_or_none(x, 1000)}),
 488                 'age_limit': ('maturity_ratings', -1, {parse_age_limit}),
 489             }),
 490         }
 491
 492
 493 class CrunchyrollBetaShowIE(CrunchyrollCmsBaseIE):
 494     IE_NAME = 'crunchyroll:playlist'
 495     _VALID_URL = r'''(?x)
 496         https?://(?:beta\.|www\.)?crunchyroll\.com/
 497         (?P<lang>(?:\w{2}(?:-\w{2})?/)?)
 498         series/(?P<id>\w+)'''
 499     _TESTS = [{
 500         'url': 'https://www.crunchyroll.com/series/GY19NQ2QR/Girl-Friend-BETA',
 501         'info_dict': {
 502             'id': 'GY19NQ2QR',
 503             'title': 'Girl Friend BETA',
 504             'description': 'md5:99c1b22ee30a74b536a8277ced8eb750',
 505             # XXX: `thumbnail` does not get set from `thumbnails` in playlist
 506             #  'thumbnail': r're:^https://www.crunchyroll.com/imgsrv/.*\.jpeg?$',
 507             'age_limit': 14,
 508         },
 509         'playlist_mincount': 10,
 510     }, {
 511         'url': 'https://beta.crunchyroll.com/it/series/GY19NQ2QR',
 512         'only_matching': True,
 513     }]
 514
 515     def _real_extract(self, url):
 516         lang, internal_id = self._match_valid_url(url).group('lang', 'id')
 517
 518         def entries():
 519             seasons_response = self._call_cms_api_signed(f'seasons?series_id={internal_id}', internal_id, lang, 'seasons')
 520             for season in traverse_obj(seasons_response, ('items', ..., {dict})):
 521                 episodes_response = self._call_cms_api_signed(
 522                     f'episodes?season_id={season["id"]}', season["id"], lang, 'episode list')
 523                 for episode_response in traverse_obj(episodes_response, ('items', ..., {dict})):
 524                     yield self.url_result(
 525                         f'{self._BASE_URL}/{lang}watch/{episode_response["id"]}',
 526                         CrunchyrollBetaIE, **CrunchyrollBetaIE._transform_episode_response(episode_response))
 527
 528         return self.playlist_result(
 529             entries(), internal_id,
 530             **traverse_obj(self._call_api(f'series/{internal_id}', internal_id, lang, 'series'), ('data', 0, {
 531                 'title': ('title', {str}),
 532                 'description': ('description', {lambda x: x.replace(r'\r\n', '\n')}),
 533                 'age_limit': ('maturity_ratings', -1, {parse_age_limit}),
 534                 'thumbnails': ('images', ..., ..., ..., {
 535                     'url': ('source', {url_or_none}),
 536                     'width': ('width', {int_or_none}),
 537                     'height': ('height', {int_or_none}),
 538                 })
 539             })))
 540
 541
 542 class CrunchyrollMusicIE(CrunchyrollBaseIE):
 543     IE_NAME = 'crunchyroll:music'
 544     _VALID_URL = r'''(?x)
 545         https?://(?:www\.)?crunchyroll\.com/
 546         (?P<lang>(?:\w{2}(?:-\w{2})?/)?)
 547         watch/(?P<type>concert|musicvideo)/(?P<id>\w+)'''
 548     _TESTS = [{
 549         'url': 'https://www.crunchyroll.com/de/watch/musicvideo/MV5B02C79',
 550         'info_dict': {
 551             'ext': 'mp4',
 552             'id': 'MV5B02C79',
 553             'display_id': 'egaono-hana',
 554             'title': 'Egaono Hana',
 555             'track': 'Egaono Hana',
 556             'artists': ['Goose house'],
 557             'thumbnail': r're:(?i)^https://www.crunchyroll.com/imgsrv/.*\.jpeg?$',
 558             'genres': ['J-Pop'],
 559         },
 560         'params': {'skip_download': 'm3u8'},
 561     }, {
 562         'url': 'https://www.crunchyroll.com/watch/musicvideo/MV88BB7F2C',
 563         'info_dict': {
 564             'ext': 'mp4',
 565             'id': 'MV88BB7F2C',
 566             'display_id': 'crossing-field',
 567             'title': 'Crossing Field',
 568             'track': 'Crossing Field',
 569             'artists': ['LiSA'],
 570             'thumbnail': r're:(?i)^https://www.crunchyroll.com/imgsrv/.*\.jpeg?$',
 571             'genres': ['Anime'],
 572         },
 573         'params': {'skip_download': 'm3u8'},
 574         'skip': 'no longer exists',
 575     }, {
 576         'url': 'https://www.crunchyroll.com/watch/concert/MC2E2AC135',
 577         'info_dict': {
 578             'ext': 'mp4',
 579             'id': 'MC2E2AC135',
 580             'display_id': 'live-is-smile-always-364joker-at-yokohama-arena',
 581             'title': 'LiVE is Smile Always-364+JOKER- at YOKOHAMA ARENA',
 582             'track': 'LiVE is Smile Always-364+JOKER- at YOKOHAMA ARENA',
 583             'artists': ['LiSA'],
 584             'thumbnail': r're:(?i)^https://www.crunchyroll.com/imgsrv/.*\.jpeg?$',
 585             'description': 'md5:747444e7e6300907b7a43f0a0503072e',
 586             'genres': ['J-Pop'],
 587         },
 588         'params': {'skip_download': 'm3u8'},
 589     }, {
 590         'url': 'https://www.crunchyroll.com/de/watch/musicvideo/MV5B02C79/egaono-hana',
 591         'only_matching': True,
 592     }, {
 593         'url': 'https://www.crunchyroll.com/watch/concert/MC2E2AC135/live-is-smile-always-364joker-at-yokohama-arena',
 594         'only_matching': True,
 595     }, {
 596         'url': 'https://www.crunchyroll.com/watch/musicvideo/MV88BB7F2C/crossing-field',
 597         'only_matching': True,
 598     }]
 599     _API_ENDPOINT = 'music'
 600
 601     def _real_extract(self, url):
 602         lang, internal_id, object_type = self._match_valid_url(url).group('lang', 'id', 'type')
 603         path, name = {
 604             'concert': ('concerts', 'concert info'),
 605             'musicvideo': ('music_videos', 'music video info'),
 606         }[object_type]
 607         response = traverse_obj(self._call_api(f'{path}/{internal_id}', internal_id, lang, name), ('data', 0, {dict}))
 608         if not response:
 609             raise ExtractorError(f'No video with id {internal_id} could be found (possibly region locked?)', expected=True)
 610
 611         result = self._transform_music_response(response)
 612
 613         if not self._IS_PREMIUM and response.get('isPremiumOnly'):
 614             message = f'This {response.get("type") or "media"} is for premium members only'
 615             if CrunchyrollBaseIE._REFRESH_TOKEN:
 616                 self.raise_no_formats(message, expected=True, video_id=internal_id)
 617             else:
 618                 self.raise_login_required(message, method='password', metadata_available=True)
 619         else:
 620             result['formats'], _ = self._extract_stream(f'music/{internal_id}', internal_id)
 621
 622         return result
 623
 624     @staticmethod
 625     def _transform_music_response(data):
 626         return {
 627             'id': data['id'],
 628             **traverse_obj(data, {
 629                 'display_id': 'slug',
 630                 'title': 'title',
 631                 'track': 'title',
 632                 'artists': ('artist', 'name', all),
 633                 'description': ('description', {str}, {lambda x: x.replace(r'\r\n', '\n') or None}),
 634                 'thumbnails': ('images', ..., ..., {
 635                     'url': ('source', {url_or_none}),
 636                     'width': ('width', {int_or_none}),
 637                     'height': ('height', {int_or_none}),
 638                 }),
 639                 'genres': ('genres', ..., 'displayValue'),
 640                 'age_limit': ('maturity_ratings', -1, {parse_age_limit}),
 641             }),
 642         }
 643
 644
 645 class CrunchyrollArtistIE(CrunchyrollBaseIE):
 646     IE_NAME = 'crunchyroll:artist'
 647     _VALID_URL = r'''(?x)
 648         https?://(?:www\.)?crunchyroll\.com/
 649         (?P<lang>(?:\w{2}(?:-\w{2})?/)?)
 650         artist/(?P<id>\w{10})'''
 651     _TESTS = [{
 652         'url': 'https://www.crunchyroll.com/artist/MA179CB50D',
 653         'info_dict': {
 654             'id': 'MA179CB50D',
 655             'title': 'LiSA',
 656             'genres': ['Anime', 'J-Pop', 'Rock'],
 657             'description': 'md5:16d87de61a55c3f7d6c454b73285938e',
 658         },
 659         'playlist_mincount': 83,
 660     }, {
 661         'url': 'https://www.crunchyroll.com/artist/MA179CB50D/lisa',
 662         'only_matching': True,
 663     }]
 664     _API_ENDPOINT = 'music'
 665
 666     def _real_extract(self, url):
 667         lang, internal_id = self._match_valid_url(url).group('lang', 'id')
 668         response = traverse_obj(self._call_api(
 669             f'artists/{internal_id}', internal_id, lang, 'artist info'), ('data', 0))
 670
 671         def entries():
 672             for attribute, path in [('concerts', 'concert'), ('videos', 'musicvideo')]:
 673                 for internal_id in traverse_obj(response, (attribute, ...)):
 674                     yield self.url_result(f'{self._BASE_URL}/watch/{path}/{internal_id}', CrunchyrollMusicIE, internal_id)
 675
 676         return self.playlist_result(entries(), **self._transform_artist_response(response))
 677
 678     @staticmethod
 679     def _transform_artist_response(data):
 680         return {
 681             'id': data['id'],
 682             **traverse_obj(data, {
 683                 'title': 'name',
 684                 'description': ('description', {str}, {lambda x: x.replace(r'\r\n', '\n')}),
 685                 'thumbnails': ('images', ..., ..., {
 686                     'url': ('source', {url_or_none}),
 687                     'width': ('width', {int_or_none}),
 688                     'height': ('height', {int_or_none}),
 689                 }),
 690                 'genres': ('genres', ..., 'displayValue'),
 691             }),
 692         }