yt_dlp/extractor/nrk.py

   1 import itertools
   2 import random
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import compat_str
   7 from ..utils import (
   8     compat_HTTPError,
   9     determine_ext,
  10     ExtractorError,
  11     int_or_none,
  12     parse_duration,
  13     parse_iso8601,
  14     str_or_none,
  15     try_get,
  16     urljoin,
  17     url_or_none,
  18 )
  19
  20
  21 class NRKBaseIE(InfoExtractor):
  22     _GEO_COUNTRIES = ['NO']
  23     _CDN_REPL_REGEX = r'''(?x)://
  24         (?:
  25             nrkod\d{1,2}-httpcache0-47115-cacheod0\.dna\.ip-only\.net/47115-cacheod0|
  26             nrk-od-no\.telenorcdn\.net|
  27             minicdn-od\.nrk\.no/od/nrkhd-osl-rr\.netwerk\.no/no
  28         )/'''
  29
  30     def _extract_nrk_formats(self, asset_url, video_id):
  31         if re.match(r'https?://[^/]+\.akamaihd\.net/i/', asset_url):
  32             return self._extract_akamai_formats(asset_url, video_id)
  33         asset_url = re.sub(r'(?:bw_(?:low|high)=\d+|no_audio_only)&?', '', asset_url)
  34         formats = self._extract_m3u8_formats(
  35             asset_url, video_id, 'mp4', 'm3u8_native', fatal=False)
  36         if not formats and re.search(self._CDN_REPL_REGEX, asset_url):
  37             formats = self._extract_m3u8_formats(
  38                 re.sub(self._CDN_REPL_REGEX, '://nrk-od-%02d.akamaized.net/no/' % random.randint(0, 99), asset_url),
  39                 video_id, 'mp4', 'm3u8_native', fatal=False)
  40         return formats
  41
  42     def _raise_error(self, data):
  43         MESSAGES = {
  44             'ProgramRightsAreNotReady': 'Du kan dessverre ikke se eller høre programmet',
  45             'ProgramRightsHasExpired': 'Programmet har gått ut',
  46             'NoProgramRights': 'Ikke tilgjengelig',
  47             'ProgramIsGeoBlocked': 'NRK har ikke rettigheter til å vise dette programmet utenfor Norge',
  48         }
  49         message_type = data.get('messageType', '')
  50         # Can be ProgramIsGeoBlocked or ChannelIsGeoBlocked*
  51         if 'IsGeoBlocked' in message_type or try_get(data, lambda x: x['usageRights']['isGeoBlocked']) is True:
  52             self.raise_geo_restricted(
  53                 msg=MESSAGES.get('ProgramIsGeoBlocked'),
  54                 countries=self._GEO_COUNTRIES)
  55         message = data.get('endUserMessage') or MESSAGES.get(message_type, message_type)
  56         raise ExtractorError('%s said: %s' % (self.IE_NAME, message), expected=True)
  57
  58     def _call_api(self, path, video_id, item=None, note=None, fatal=True, query=None):
  59         return self._download_json(
  60             urljoin('https://psapi.nrk.no/', path),
  61             video_id, note or 'Downloading %s JSON' % item,
  62             fatal=fatal, query=query,
  63             headers={'Accept-Encoding': 'gzip, deflate, br'})
  64
  65
  66 class NRKIE(NRKBaseIE):
  67     _VALID_URL = r'''(?x)
  68                         (?:
  69                             nrk:|
  70                             https?://
  71                                 (?:
  72                                     (?:www\.)?nrk\.no/video/(?:PS\*|[^_]+_)|
  73                                     v8[-.]psapi\.nrk\.no/mediaelement/
  74                                 )
  75                             )
  76                             (?P<id>[^?\#&]+)
  77                         '''
  78
  79     _TESTS = [{
  80         # video
  81         'url': 'http://www.nrk.no/video/PS*150533',
  82         'md5': 'f46be075326e23ad0e524edfcb06aeb6',
  83         'info_dict': {
  84             'id': '150533',
  85             'ext': 'mp4',
  86             'title': 'Dompap og andre fugler i Piip-Show',
  87             'description': 'md5:d9261ba34c43b61c812cb6b0269a5c8f',
  88             'duration': 262,
  89         }
  90     }, {
  91         # audio
  92         'url': 'http://www.nrk.no/video/PS*154915',
  93         # MD5 is unstable
  94         'info_dict': {
  95             'id': '154915',
  96             'ext': 'mp4',
  97             'title': 'Slik høres internett ut når du er blind',
  98             'description': 'md5:a621f5cc1bd75c8d5104cb048c6b8568',
  99             'duration': 20,
 100         }
 101     }, {
 102         'url': 'nrk:ecc1b952-96dc-4a98-81b9-5296dc7a98d9',
 103         'only_matching': True,
 104     }, {
 105         'url': 'nrk:clip/7707d5a3-ebe7-434a-87d5-a3ebe7a34a70',
 106         'only_matching': True,
 107     }, {
 108         'url': 'https://v8-psapi.nrk.no/mediaelement/ecc1b952-96dc-4a98-81b9-5296dc7a98d9',
 109         'only_matching': True,
 110     }, {
 111         'url': 'https://www.nrk.no/video/dompap-og-andre-fugler-i-piip-show_150533',
 112         'only_matching': True,
 113     }, {
 114         'url': 'https://www.nrk.no/video/humor/kommentatorboksen-reiser-til-sjos_d1fda11f-a4ad-437a-a374-0398bc84e999',
 115         'only_matching': True,
 116     }, {
 117         # podcast
 118         'url': 'nrk:l_96f4f1b0-de54-4e6a-b4f1-b0de54fe6af8',
 119         'only_matching': True,
 120     }, {
 121         'url': 'nrk:podcast/l_96f4f1b0-de54-4e6a-b4f1-b0de54fe6af8',
 122         'only_matching': True,
 123     }, {
 124         # clip
 125         'url': 'nrk:150533',
 126         'only_matching': True,
 127     }, {
 128         'url': 'nrk:clip/150533',
 129         'only_matching': True,
 130     }, {
 131         # program
 132         'url': 'nrk:MDDP12000117',
 133         'only_matching': True,
 134     }, {
 135         'url': 'nrk:program/ENRK10100318',
 136         'only_matching': True,
 137     }, {
 138         # direkte
 139         'url': 'nrk:nrk1',
 140         'only_matching': True,
 141     }, {
 142         'url': 'nrk:channel/nrk1',
 143         'only_matching': True,
 144     }]
 145
 146     def _real_extract(self, url):
 147         video_id = self._match_id(url).split('/')[-1]
 148
 149         def call_playback_api(item, query=None):
 150             try:
 151                 return self._call_api(f'playback/{item}/program/{video_id}', video_id, item, query=query)
 152             except ExtractorError as e:
 153                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 400:
 154                     return self._call_api(f'playback/{item}/{video_id}', video_id, item, query=query)
 155                 raise
 156
 157         # known values for preferredCdn: akamai, iponly, minicdn and telenor
 158         manifest = call_playback_api('manifest', {'preferredCdn': 'akamai'})
 159
 160         video_id = try_get(manifest, lambda x: x['id'], compat_str) or video_id
 161
 162         if manifest.get('playability') == 'nonPlayable':
 163             self._raise_error(manifest['nonPlayable'])
 164
 165         playable = manifest['playable']
 166
 167         formats = []
 168         for asset in playable['assets']:
 169             if not isinstance(asset, dict):
 170                 continue
 171             if asset.get('encrypted'):
 172                 continue
 173             format_url = url_or_none(asset.get('url'))
 174             if not format_url:
 175                 continue
 176             asset_format = (asset.get('format') or '').lower()
 177             if asset_format == 'hls' or determine_ext(format_url) == 'm3u8':
 178                 formats.extend(self._extract_nrk_formats(format_url, video_id))
 179             elif asset_format == 'mp3':
 180                 formats.append({
 181                     'url': format_url,
 182                     'format_id': asset_format,
 183                     'vcodec': 'none',
 184                 })
 185         self._sort_formats(formats)
 186
 187         data = call_playback_api('metadata')
 188
 189         preplay = data['preplay']
 190         titles = preplay['titles']
 191         title = titles['title']
 192         alt_title = titles.get('subtitle')
 193
 194         description = try_get(preplay, lambda x: x['description'].replace('\r', '\n'))
 195         duration = parse_duration(playable.get('duration')) or parse_duration(data.get('duration'))
 196
 197         thumbnails = []
 198         for image in try_get(
 199                 preplay, lambda x: x['poster']['images'], list) or []:
 200             if not isinstance(image, dict):
 201                 continue
 202             image_url = url_or_none(image.get('url'))
 203             if not image_url:
 204                 continue
 205             thumbnails.append({
 206                 'url': image_url,
 207                 'width': int_or_none(image.get('pixelWidth')),
 208                 'height': int_or_none(image.get('pixelHeight')),
 209             })
 210
 211         subtitles = {}
 212         for sub in try_get(playable, lambda x: x['subtitles'], list) or []:
 213             if not isinstance(sub, dict):
 214                 continue
 215             sub_url = url_or_none(sub.get('webVtt'))
 216             if not sub_url:
 217                 continue
 218             sub_key = str_or_none(sub.get('language')) or 'nb'
 219             sub_type = str_or_none(sub.get('type'))
 220             if sub_type:
 221                 sub_key += '-%s' % sub_type
 222             subtitles.setdefault(sub_key, []).append({
 223                 'url': sub_url,
 224             })
 225
 226         legal_age = try_get(
 227             data, lambda x: x['legalAge']['body']['rating']['code'], compat_str)
 228         # https://en.wikipedia.org/wiki/Norwegian_Media_Authority
 229         age_limit = None
 230         if legal_age:
 231             if legal_age == 'A':
 232                 age_limit = 0
 233             elif legal_age.isdigit():
 234                 age_limit = int_or_none(legal_age)
 235
 236         is_series = try_get(data, lambda x: x['_links']['series']['name']) == 'series'
 237
 238         info = {
 239             'id': video_id,
 240             'title': title,
 241             'alt_title': alt_title,
 242             'description': description,
 243             'duration': duration,
 244             'thumbnails': thumbnails,
 245             'age_limit': age_limit,
 246             'formats': formats,
 247             'subtitles': subtitles,
 248             'timestamp': parse_iso8601(try_get(manifest, lambda x: x['availability']['onDemand']['from'], str))
 249         }
 250
 251         if is_series:
 252             series = season_id = season_number = episode = episode_number = None
 253             programs = self._call_api(
 254                 'programs/%s' % video_id, video_id, 'programs', fatal=False)
 255             if programs and isinstance(programs, dict):
 256                 series = str_or_none(programs.get('seriesTitle'))
 257                 season_id = str_or_none(programs.get('seasonId'))
 258                 season_number = int_or_none(programs.get('seasonNumber'))
 259                 episode = str_or_none(programs.get('episodeTitle'))
 260                 episode_number = int_or_none(programs.get('episodeNumber'))
 261             if not series:
 262                 series = title
 263             if alt_title:
 264                 title += ' - %s' % alt_title
 265             if not season_number:
 266                 season_number = int_or_none(self._search_regex(
 267                     r'Sesong\s+(\d+)', description or '', 'season number',
 268                     default=None))
 269             if not episode:
 270                 episode = alt_title if is_series else None
 271             if not episode_number:
 272                 episode_number = int_or_none(self._search_regex(
 273                     r'^(\d+)\.', episode or '', 'episode number',
 274                     default=None))
 275             if not episode_number:
 276                 episode_number = int_or_none(self._search_regex(
 277                     r'\((\d+)\s*:\s*\d+\)', description or '',
 278                     'episode number', default=None))
 279             info.update({
 280                 'title': title,
 281                 'series': series,
 282                 'season_id': season_id,
 283                 'season_number': season_number,
 284                 'episode': episode,
 285                 'episode_number': episode_number,
 286             })
 287
 288         return info
 289
 290
 291 class NRKTVIE(InfoExtractor):
 292     IE_DESC = 'NRK TV and NRK Radio'
 293     _EPISODE_RE = r'(?P<id>[a-zA-Z]{4}\d{8})'
 294     _VALID_URL = r'https?://(?:tv|radio)\.nrk(?:super)?\.no/(?:[^/]+/)*%s' % _EPISODE_RE
 295     _TESTS = [{
 296         'url': 'https://tv.nrk.no/program/MDDP12000117',
 297         'md5': 'c4a5960f1b00b40d47db65c1064e0ab1',
 298         'info_dict': {
 299             'id': 'MDDP12000117',
 300             'ext': 'mp4',
 301             'title': 'Alarm Trolltunga',
 302             'description': 'md5:46923a6e6510eefcce23d5ef2a58f2ce',
 303             'duration': 2223.44,
 304             'age_limit': 6,
 305             'subtitles': {
 306                 'nb-nor': [{
 307                     'ext': 'vtt',
 308                 }],
 309                 'nb-ttv': [{
 310                     'ext': 'vtt',
 311                 }]
 312             },
 313         },
 314     }, {
 315         'url': 'https://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
 316         'md5': '8d40dab61cea8ab0114e090b029a0565',
 317         'info_dict': {
 318             'id': 'MUHH48000314',
 319             'ext': 'mp4',
 320             'title': '20 spørsmål - 23. mai 2014',
 321             'alt_title': '23. mai 2014',
 322             'description': 'md5:bdea103bc35494c143c6a9acdd84887a',
 323             'duration': 1741,
 324             'series': '20 spørsmål',
 325             'episode': '23. mai 2014',
 326             'age_limit': 0,
 327         },
 328     }, {
 329         'url': 'https://tv.nrk.no/program/mdfp15000514',
 330         'info_dict': {
 331             'id': 'MDFP15000514',
 332             'ext': 'mp4',
 333             'title': 'Kunnskapskanalen - Grunnlovsjubiléet - Stor ståhei for ingenting',
 334             'description': 'md5:89290c5ccde1b3a24bb8050ab67fe1db',
 335             'duration': 4605.08,
 336             'series': 'Kunnskapskanalen',
 337             'episode': 'Grunnlovsjubiléet - Stor ståhei for ingenting',
 338             'age_limit': 0,
 339         },
 340         'params': {
 341             'skip_download': True,
 342         },
 343     }, {
 344         # single playlist video
 345         'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2',
 346         'info_dict': {
 347             'id': 'MSPO40010515',
 348             'ext': 'mp4',
 349             'title': 'Sprint fri teknikk, kvinner og menn 06.01.2015',
 350             'description': 'md5:c03aba1e917561eface5214020551b7a',
 351             'age_limit': 0,
 352         },
 353         'params': {
 354             'skip_download': True,
 355         },
 356         'expected_warnings': ['Failed to download m3u8 information'],
 357         'skip': 'particular part is not supported currently',
 358     }, {
 359         'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015',
 360         'info_dict': {
 361             'id': 'MSPO40010515',
 362             'ext': 'mp4',
 363             'title': 'Sprint fri teknikk, kvinner og menn 06.01.2015',
 364             'description': 'md5:c03aba1e917561eface5214020551b7a',
 365             'age_limit': 0,
 366         },
 367         'expected_warnings': ['Failed to download m3u8 information'],
 368         'skip': 'Ikke tilgjengelig utenfor Norge',
 369     }, {
 370         'url': 'https://tv.nrk.no/serie/anno/KMTE50001317/sesong-3/episode-13',
 371         'info_dict': {
 372             'id': 'KMTE50001317',
 373             'ext': 'mp4',
 374             'title': 'Anno - 13. episode',
 375             'description': 'md5:11d9613661a8dbe6f9bef54e3a4cbbfa',
 376             'duration': 2340,
 377             'series': 'Anno',
 378             'episode': '13. episode',
 379             'season_number': 3,
 380             'episode_number': 13,
 381             'age_limit': 0,
 382         },
 383         'params': {
 384             'skip_download': True,
 385         },
 386     }, {
 387         'url': 'https://tv.nrk.no/serie/nytt-paa-nytt/MUHH46000317/27-01-2017',
 388         'info_dict': {
 389             'id': 'MUHH46000317',
 390             'ext': 'mp4',
 391             'title': 'Nytt på Nytt 27.01.2017',
 392             'description': 'md5:5358d6388fba0ea6f0b6d11c48b9eb4b',
 393             'duration': 1796,
 394             'series': 'Nytt på nytt',
 395             'episode': '27.01.2017',
 396             'age_limit': 0,
 397         },
 398         'params': {
 399             'skip_download': True,
 400         },
 401         'skip': 'ProgramRightsHasExpired',
 402     }, {
 403         'url': 'https://radio.nrk.no/serie/dagsnytt/NPUB21019315/12-07-2015#',
 404         'only_matching': True,
 405     }, {
 406         'url': 'https://tv.nrk.no/serie/lindmo/2018/MUHU11006318/avspiller',
 407         'only_matching': True,
 408     }, {
 409         'url': 'https://radio.nrk.no/serie/dagsnytt/sesong/201507/NPUB21019315',
 410         'only_matching': True,
 411     }]
 412
 413     def _real_extract(self, url):
 414         video_id = self._match_id(url)
 415         return self.url_result(
 416             'nrk:%s' % video_id, ie=NRKIE.ie_key(), video_id=video_id)
 417
 418
 419 class NRKTVEpisodeIE(InfoExtractor):
 420     _VALID_URL = r'https?://tv\.nrk\.no/serie/(?P<id>[^/]+/sesong/(?P<season_number>\d+)/episode/(?P<episode_number>\d+))'
 421     _TESTS = [{
 422         'url': 'https://tv.nrk.no/serie/hellums-kro/sesong/1/episode/2',
 423         'info_dict': {
 424             'id': 'MUHH36005220',
 425             'ext': 'mp4',
 426             'title': 'Hellums kro - 2. Kro, krig og kjærlighet',
 427             'description': 'md5:ad92ddffc04cea8ce14b415deef81787',
 428             'duration': 1563.92,
 429             'series': 'Hellums kro',
 430             'season_number': 1,
 431             'episode_number': 2,
 432             'episode': '2. Kro, krig og kjærlighet',
 433             'age_limit': 6,
 434         },
 435         'params': {
 436             'skip_download': True,
 437         },
 438     }, {
 439         'url': 'https://tv.nrk.no/serie/backstage/sesong/1/episode/8',
 440         'info_dict': {
 441             'id': 'MSUI14000816',
 442             'ext': 'mp4',
 443             'title': 'Backstage - 8. episode',
 444             'description': 'md5:de6ca5d5a2d56849e4021f2bf2850df4',
 445             'duration': 1320,
 446             'series': 'Backstage',
 447             'season_number': 1,
 448             'episode_number': 8,
 449             'episode': '8. episode',
 450             'age_limit': 0,
 451         },
 452         'params': {
 453             'skip_download': True,
 454         },
 455         'skip': 'ProgramRightsHasExpired',
 456     }]
 457
 458     def _real_extract(self, url):
 459         display_id, season_number, episode_number = self._match_valid_url(url).groups()
 460
 461         webpage = self._download_webpage(url, display_id)
 462
 463         info = self._search_json_ld(webpage, display_id, default={})
 464         nrk_id = info.get('@id') or self._html_search_meta(
 465             'nrk:program-id', webpage, default=None) or self._search_regex(
 466             r'data-program-id=["\'](%s)' % NRKTVIE._EPISODE_RE, webpage,
 467             'nrk id')
 468         assert re.match(NRKTVIE._EPISODE_RE, nrk_id)
 469
 470         info.update({
 471             '_type': 'url',
 472             'id': nrk_id,
 473             'url': 'nrk:%s' % nrk_id,
 474             'ie_key': NRKIE.ie_key(),
 475             'season_number': int(season_number),
 476             'episode_number': int(episode_number),
 477         })
 478         return info
 479
 480
 481 class NRKTVSerieBaseIE(NRKBaseIE):
 482     def _extract_entries(self, entry_list):
 483         if not isinstance(entry_list, list):
 484             return []
 485         entries = []
 486         for episode in entry_list:
 487             nrk_id = episode.get('prfId') or episode.get('episodeId')
 488             if not nrk_id or not isinstance(nrk_id, compat_str):
 489                 continue
 490             entries.append(self.url_result(
 491                 'nrk:%s' % nrk_id, ie=NRKIE.ie_key(), video_id=nrk_id))
 492         return entries
 493
 494     _ASSETS_KEYS = ('episodes', 'instalments',)
 495
 496     def _extract_assets_key(self, embedded):
 497         for asset_key in self._ASSETS_KEYS:
 498             if embedded.get(asset_key):
 499                 return asset_key
 500
 501     @staticmethod
 502     def _catalog_name(serie_kind):
 503         return 'podcast' if serie_kind in ('podcast', 'podkast') else 'series'
 504
 505     def _entries(self, data, display_id):
 506         for page_num in itertools.count(1):
 507             embedded = data.get('_embedded') or data
 508             if not isinstance(embedded, dict):
 509                 break
 510             assets_key = self._extract_assets_key(embedded)
 511             if not assets_key:
 512                 break
 513             # Extract entries
 514             entries = try_get(
 515                 embedded,
 516                 (lambda x: x[assets_key]['_embedded'][assets_key],
 517                  lambda x: x[assets_key]),
 518                 list)
 519             for e in self._extract_entries(entries):
 520                 yield e
 521             # Find next URL
 522             next_url_path = try_get(
 523                 data,
 524                 (lambda x: x['_links']['next']['href'],
 525                  lambda x: x['_embedded'][assets_key]['_links']['next']['href']),
 526                 compat_str)
 527             if not next_url_path:
 528                 break
 529             data = self._call_api(
 530                 next_url_path, display_id,
 531                 note='Downloading %s JSON page %d' % (assets_key, page_num),
 532                 fatal=False)
 533             if not data:
 534                 break
 535
 536
 537 class NRKTVSeasonIE(NRKTVSerieBaseIE):
 538     _VALID_URL = r'''(?x)
 539                     https?://
 540                         (?P<domain>tv|radio)\.nrk\.no/
 541                         (?P<serie_kind>serie|pod[ck]ast)/
 542                         (?P<serie>[^/]+)/
 543                         (?:
 544                             (?:sesong/)?(?P<id>\d+)|
 545                             sesong/(?P<id_2>[^/?#&]+)
 546                         )
 547                     '''
 548     _TESTS = [{
 549         'url': 'https://tv.nrk.no/serie/backstage/sesong/1',
 550         'info_dict': {
 551             'id': 'backstage/1',
 552             'title': 'Sesong 1',
 553         },
 554         'playlist_mincount': 30,
 555     }, {
 556         # no /sesong/ in path
 557         'url': 'https://tv.nrk.no/serie/lindmo/2016',
 558         'info_dict': {
 559             'id': 'lindmo/2016',
 560             'title': '2016',
 561         },
 562         'playlist_mincount': 29,
 563     }, {
 564         # weird nested _embedded in catalog JSON response
 565         'url': 'https://radio.nrk.no/serie/dickie-dick-dickens/sesong/1',
 566         'info_dict': {
 567             'id': 'dickie-dick-dickens/1',
 568             'title': 'Sesong 1',
 569         },
 570         'playlist_mincount': 11,
 571     }, {
 572         # 841 entries, multi page
 573         'url': 'https://radio.nrk.no/serie/dagsnytt/sesong/201509',
 574         'info_dict': {
 575             'id': 'dagsnytt/201509',
 576             'title': 'September 2015',
 577         },
 578         'playlist_mincount': 841,
 579     }, {
 580         # 180 entries, single page
 581         'url': 'https://tv.nrk.no/serie/spangas/sesong/1',
 582         'only_matching': True,
 583     }, {
 584         'url': 'https://radio.nrk.no/podkast/hele_historien/sesong/diagnose-kverulant',
 585         'info_dict': {
 586             'id': 'hele_historien/diagnose-kverulant',
 587             'title': 'Diagnose kverulant',
 588         },
 589         'playlist_mincount': 3,
 590     }, {
 591         'url': 'https://radio.nrk.no/podkast/loerdagsraadet/sesong/202101',
 592         'only_matching': True,
 593     }]
 594
 595     @classmethod
 596     def suitable(cls, url):
 597         return (False if NRKTVIE.suitable(url) or NRKTVEpisodeIE.suitable(url) or NRKRadioPodkastIE.suitable(url)
 598                 else super(NRKTVSeasonIE, cls).suitable(url))
 599
 600     def _real_extract(self, url):
 601         mobj = self._match_valid_url(url)
 602         domain = mobj.group('domain')
 603         serie_kind = mobj.group('serie_kind')
 604         serie = mobj.group('serie')
 605         season_id = mobj.group('id') or mobj.group('id_2')
 606         display_id = '%s/%s' % (serie, season_id)
 607
 608         data = self._call_api(
 609             '%s/catalog/%s/%s/seasons/%s'
 610             % (domain, self._catalog_name(serie_kind), serie, season_id),
 611             display_id, 'season', query={'pageSize': 50})
 612
 613         title = try_get(data, lambda x: x['titles']['title'], compat_str) or display_id
 614         return self.playlist_result(
 615             self._entries(data, display_id),
 616             display_id, title)
 617
 618
 619 class NRKTVSeriesIE(NRKTVSerieBaseIE):
 620     _VALID_URL = r'https?://(?P<domain>(?:tv|radio)\.nrk|(?:tv\.)?nrksuper)\.no/(?P<serie_kind>serie|pod[ck]ast)/(?P<id>[^/]+)'
 621     _TESTS = [{
 622         # new layout, instalments
 623         'url': 'https://tv.nrk.no/serie/groenn-glede',
 624         'info_dict': {
 625             'id': 'groenn-glede',
 626             'title': 'Grønn glede',
 627             'description': 'md5:7576e92ae7f65da6993cf90ee29e4608',
 628         },
 629         'playlist_mincount': 90,
 630     }, {
 631         # new layout, instalments, more entries
 632         'url': 'https://tv.nrk.no/serie/lindmo',
 633         'only_matching': True,
 634     }, {
 635         'url': 'https://tv.nrk.no/serie/blank',
 636         'info_dict': {
 637             'id': 'blank',
 638             'title': 'Blank',
 639             'description': 'md5:7664b4e7e77dc6810cd3bca367c25b6e',
 640         },
 641         'playlist_mincount': 30,
 642     }, {
 643         # new layout, seasons
 644         'url': 'https://tv.nrk.no/serie/backstage',
 645         'info_dict': {
 646             'id': 'backstage',
 647             'title': 'Backstage',
 648             'description': 'md5:63692ceb96813d9a207e9910483d948b',
 649         },
 650         'playlist_mincount': 60,
 651     }, {
 652         # old layout
 653         'url': 'https://tv.nrksuper.no/serie/labyrint',
 654         'info_dict': {
 655             'id': 'labyrint',
 656             'title': 'Labyrint',
 657             'description': 'I Daidalos sin undersjøiske Labyrint venter spennende oppgaver, skumle robotskapninger og slim.',
 658         },
 659         'playlist_mincount': 3,
 660     }, {
 661         'url': 'https://tv.nrk.no/serie/broedrene-dal-og-spektralsteinene',
 662         'only_matching': True,
 663     }, {
 664         'url': 'https://tv.nrk.no/serie/saving-the-human-race',
 665         'only_matching': True,
 666     }, {
 667         'url': 'https://tv.nrk.no/serie/postmann-pat',
 668         'only_matching': True,
 669     }, {
 670         'url': 'https://radio.nrk.no/serie/dickie-dick-dickens',
 671         'info_dict': {
 672             'id': 'dickie-dick-dickens',
 673             'title': 'Dickie Dick Dickens',
 674             'description': 'md5:19e67411ffe57f7dce08a943d7a0b91f',
 675         },
 676         'playlist_mincount': 8,
 677     }, {
 678         'url': 'https://nrksuper.no/serie/labyrint',
 679         'only_matching': True,
 680     }, {
 681         'url': 'https://radio.nrk.no/podkast/ulrikkes_univers',
 682         'info_dict': {
 683             'id': 'ulrikkes_univers',
 684         },
 685         'playlist_mincount': 10,
 686     }, {
 687         'url': 'https://radio.nrk.no/podkast/ulrikkes_univers/nrkno-poddkast-26588-134079-05042018030000',
 688         'only_matching': True,
 689     }]
 690
 691     @classmethod
 692     def suitable(cls, url):
 693         return (
 694             False if any(ie.suitable(url)
 695                          for ie in (NRKTVIE, NRKTVEpisodeIE, NRKRadioPodkastIE, NRKTVSeasonIE))
 696             else super(NRKTVSeriesIE, cls).suitable(url))
 697
 698     def _real_extract(self, url):
 699         site, serie_kind, series_id = self._match_valid_url(url).groups()
 700         is_radio = site == 'radio.nrk'
 701         domain = 'radio' if is_radio else 'tv'
 702
 703         size_prefix = 'p' if is_radio else 'embeddedInstalmentsP'
 704         series = self._call_api(
 705             '%s/catalog/%s/%s'
 706             % (domain, self._catalog_name(serie_kind), series_id),
 707             series_id, 'serie', query={size_prefix + 'ageSize': 50})
 708         titles = try_get(series, [
 709             lambda x: x['titles'],
 710             lambda x: x[x['type']]['titles'],
 711             lambda x: x[x['seriesType']]['titles'],
 712         ]) or {}
 713
 714         entries = []
 715         entries.extend(self._entries(series, series_id))
 716         embedded = series.get('_embedded') or {}
 717         linked_seasons = try_get(series, lambda x: x['_links']['seasons']) or []
 718         embedded_seasons = embedded.get('seasons') or []
 719         if len(linked_seasons) > len(embedded_seasons):
 720             for season in linked_seasons:
 721                 season_url = urljoin(url, season.get('href'))
 722                 if not season_url:
 723                     season_name = season.get('name')
 724                     if season_name and isinstance(season_name, compat_str):
 725                         season_url = 'https://%s.nrk.no/serie/%s/sesong/%s' % (domain, series_id, season_name)
 726                 if season_url:
 727                     entries.append(self.url_result(
 728                         season_url, ie=NRKTVSeasonIE.ie_key(),
 729                         video_title=season.get('title')))
 730         else:
 731             for season in embedded_seasons:
 732                 entries.extend(self._entries(season, series_id))
 733         entries.extend(self._entries(
 734             embedded.get('extraMaterial') or {}, series_id))
 735
 736         return self.playlist_result(
 737             entries, series_id, titles.get('title'), titles.get('subtitle'))
 738
 739
 740 class NRKTVDirekteIE(NRKTVIE):
 741     IE_DESC = 'NRK TV Direkte and NRK Radio Direkte'
 742     _VALID_URL = r'https?://(?:tv|radio)\.nrk\.no/direkte/(?P<id>[^/?#&]+)'
 743
 744     _TESTS = [{
 745         'url': 'https://tv.nrk.no/direkte/nrk1',
 746         'only_matching': True,
 747     }, {
 748         'url': 'https://radio.nrk.no/direkte/p1_oslo_akershus',
 749         'only_matching': True,
 750     }]
 751
 752
 753 class NRKRadioPodkastIE(InfoExtractor):
 754     _VALID_URL = r'https?://radio\.nrk\.no/pod[ck]ast/(?:[^/]+/)+(?P<id>l_[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'
 755
 756     _TESTS = [{
 757         'url': 'https://radio.nrk.no/podkast/ulrikkes_univers/l_96f4f1b0-de54-4e6a-b4f1-b0de54fe6af8',
 758         'md5': '8d40dab61cea8ab0114e090b029a0565',
 759         'info_dict': {
 760             'id': 'MUHH48000314AA',
 761             'ext': 'mp4',
 762             'title': '20 spørsmål 23.05.2014',
 763             'description': 'md5:bdea103bc35494c143c6a9acdd84887a',
 764             'duration': 1741,
 765             'series': '20 spørsmål',
 766             'episode': '23.05.2014',
 767         },
 768     }, {
 769         'url': 'https://radio.nrk.no/podcast/ulrikkes_univers/l_96f4f1b0-de54-4e6a-b4f1-b0de54fe6af8',
 770         'only_matching': True,
 771     }, {
 772         'url': 'https://radio.nrk.no/podkast/ulrikkes_univers/sesong/1/l_96f4f1b0-de54-4e6a-b4f1-b0de54fe6af8',
 773         'only_matching': True,
 774     }, {
 775         'url': 'https://radio.nrk.no/podkast/hele_historien/sesong/bortfoert-i-bergen/l_774d1a2c-7aa7-4965-8d1a-2c7aa7d9652c',
 776         'only_matching': True,
 777     }]
 778
 779     def _real_extract(self, url):
 780         video_id = self._match_id(url)
 781         return self.url_result(
 782             'nrk:%s' % video_id, ie=NRKIE.ie_key(), video_id=video_id)
 783
 784
 785 class NRKPlaylistBaseIE(InfoExtractor):
 786     def _extract_description(self, webpage):
 787         pass
 788
 789     def _real_extract(self, url):
 790         playlist_id = self._match_id(url)
 791
 792         webpage = self._download_webpage(url, playlist_id)
 793
 794         entries = [
 795             self.url_result('nrk:%s' % video_id, NRKIE.ie_key())
 796             for video_id in re.findall(self._ITEM_RE, webpage)
 797         ]
 798
 799         playlist_title = self._extract_title(webpage)
 800         playlist_description = self._extract_description(webpage)
 801
 802         return self.playlist_result(
 803             entries, playlist_id, playlist_title, playlist_description)
 804
 805
 806 class NRKPlaylistIE(NRKPlaylistBaseIE):
 807     _VALID_URL = r'https?://(?:www\.)?nrk\.no/(?!video|skole)(?:[^/]+/)+(?P<id>[^/]+)'
 808     _ITEM_RE = r'class="[^"]*\brich\b[^"]*"[^>]+data-video-id="([^"]+)"'
 809     _TESTS = [{
 810         'url': 'http://www.nrk.no/troms/gjenopplev-den-historiske-solformorkelsen-1.12270763',
 811         'info_dict': {
 812             'id': 'gjenopplev-den-historiske-solformorkelsen-1.12270763',
 813             'title': 'Gjenopplev den historiske solformørkelsen',
 814             'description': 'md5:c2df8ea3bac5654a26fc2834a542feed',
 815         },
 816         'playlist_count': 2,
 817     }, {
 818         'url': 'http://www.nrk.no/kultur/bok/rivertonprisen-til-karin-fossum-1.12266449',
 819         'info_dict': {
 820             'id': 'rivertonprisen-til-karin-fossum-1.12266449',
 821             'title': 'Rivertonprisen til Karin Fossum',
 822             'description': 'Første kvinne på 15 år til å vinne krimlitteraturprisen.',
 823         },
 824         'playlist_count': 2,
 825     }]
 826
 827     def _extract_title(self, webpage):
 828         return self._og_search_title(webpage, fatal=False)
 829
 830     def _extract_description(self, webpage):
 831         return self._og_search_description(webpage)
 832
 833
 834 class NRKTVEpisodesIE(NRKPlaylistBaseIE):
 835     _VALID_URL = r'https?://tv\.nrk\.no/program/[Ee]pisodes/[^/]+/(?P<id>\d+)'
 836     _ITEM_RE = r'data-episode=["\']%s' % NRKTVIE._EPISODE_RE
 837     _TESTS = [{
 838         'url': 'https://tv.nrk.no/program/episodes/nytt-paa-nytt/69031',
 839         'info_dict': {
 840             'id': '69031',
 841             'title': 'Nytt på nytt, sesong: 201210',
 842         },
 843         'playlist_count': 4,
 844     }]
 845
 846     def _extract_title(self, webpage):
 847         return self._html_search_regex(
 848             r'<h1>([^<]+)</h1>', webpage, 'title', fatal=False)
 849
 850
 851 class NRKSkoleIE(InfoExtractor):
 852     IE_DESC = 'NRK Skole'
 853     _VALID_URL = r'https?://(?:www\.)?nrk\.no/skole/?\?.*\bmediaId=(?P<id>\d+)'
 854
 855     _TESTS = [{
 856         'url': 'https://www.nrk.no/skole/?page=search&q=&mediaId=14099',
 857         'md5': '18c12c3d071953c3bf8d54ef6b2587b7',
 858         'info_dict': {
 859             'id': '6021',
 860             'ext': 'mp4',
 861             'title': 'Genetikk og eneggede tvillinger',
 862             'description': 'md5:3aca25dcf38ec30f0363428d2b265f8d',
 863             'duration': 399,
 864         },
 865     }, {
 866         'url': 'https://www.nrk.no/skole/?page=objectives&subject=naturfag&objective=K15114&mediaId=19355',
 867         'only_matching': True,
 868     }]
 869
 870     def _real_extract(self, url):
 871         video_id = self._match_id(url)
 872
 873         nrk_id = self._download_json(
 874             'https://nrkno-skole-prod.kube.nrk.no/skole/api/media/%s' % video_id,
 875             video_id)['psId']
 876
 877         return self.url_result('nrk:%s' % nrk_id)