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