yt_dlp/extractor/nrk.py

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