Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/nrk.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import (
   8     compat_str,
   9     compat_urllib_parse_unquote,
  10 )
  11 from ..utils import (
  12     ExtractorError,
  13     int_or_none,
  14     js_to_json,
  15     NO_DEFAULT,
  16     parse_age_limit,
  17     parse_duration,
  18     try_get,
  19 )
  20
  21
  22 class NRKBaseIE(InfoExtractor):
  23     _GEO_COUNTRIES = ['NO']
  24
  25     _api_host = None
  26
  27     def _real_extract(self, url):
  28         video_id = self._match_id(url)
  29
  30         api_hosts = (self._api_host, ) if self._api_host else self._API_HOSTS
  31
  32         for api_host in api_hosts:
  33             data = self._download_json(
  34                 'http://%s/mediaelement/%s' % (api_host, video_id),
  35                 video_id, 'Downloading mediaelement JSON',
  36                 fatal=api_host == api_hosts[-1])
  37             if not data:
  38                 continue
  39             self._api_host = api_host
  40             break
  41
  42         title = data.get('fullTitle') or data.get('mainTitle') or data['title']
  43         video_id = data.get('id') or video_id
  44
  45         entries = []
  46
  47         conviva = data.get('convivaStatistics') or {}
  48         live = (data.get('mediaElementType') == 'Live'
  49                 or data.get('isLive') is True or conviva.get('isLive'))
  50
  51         def make_title(t):
  52             return self._live_title(t) if live else t
  53
  54         media_assets = data.get('mediaAssets')
  55         if media_assets and isinstance(media_assets, list):
  56             def video_id_and_title(idx):
  57                 return ((video_id, title) if len(media_assets) == 1
  58                         else ('%s-%d' % (video_id, idx), '%s (Part %d)' % (title, idx)))
  59             for num, asset in enumerate(media_assets, 1):
  60                 asset_url = asset.get('url')
  61                 if not asset_url:
  62                     continue
  63                 formats = self._extract_akamai_formats(asset_url, video_id)
  64                 if not formats:
  65                     continue
  66                 self._sort_formats(formats)
  67
  68                 # Some f4m streams may not work with hdcore in fragments' URLs
  69                 for f in formats:
  70                     extra_param = f.get('extra_param_to_segment_url')
  71                     if extra_param and 'hdcore' in extra_param:
  72                         del f['extra_param_to_segment_url']
  73
  74                 entry_id, entry_title = video_id_and_title(num)
  75                 duration = parse_duration(asset.get('duration'))
  76                 subtitles = {}
  77                 for subtitle in ('webVtt', 'timedText'):
  78                     subtitle_url = asset.get('%sSubtitlesUrl' % subtitle)
  79                     if subtitle_url:
  80                         subtitles.setdefault('no', []).append({
  81                             'url': compat_urllib_parse_unquote(subtitle_url)
  82                         })
  83                 entries.append({
  84                     'id': asset.get('carrierId') or entry_id,
  85                     'title': make_title(entry_title),
  86                     'duration': duration,
  87                     'subtitles': subtitles,
  88                     'formats': formats,
  89                 })
  90
  91         if not entries:
  92             media_url = data.get('mediaUrl')
  93             if media_url:
  94                 formats = self._extract_akamai_formats(media_url, video_id)
  95                 self._sort_formats(formats)
  96                 duration = parse_duration(data.get('duration'))
  97                 entries = [{
  98                     'id': video_id,
  99                     'title': make_title(title),
 100                     'duration': duration,
 101                     'formats': formats,
 102                 }]
 103
 104         if not entries:
 105             MESSAGES = {
 106                 'ProgramRightsAreNotReady': 'Du kan dessverre ikke se eller høre programmet',
 107                 'ProgramRightsHasExpired': 'Programmet har gått ut',
 108                 'NoProgramRights': 'Ikke tilgjengelig',
 109                 'ProgramIsGeoBlocked': 'NRK har ikke rettigheter til å vise dette programmet utenfor Norge',
 110             }
 111             message_type = data.get('messageType', '')
 112             # Can be ProgramIsGeoBlocked or ChannelIsGeoBlocked*
 113             if 'IsGeoBlocked' in message_type:
 114                 self.raise_geo_restricted(
 115                     msg=MESSAGES.get('ProgramIsGeoBlocked'),
 116                     countries=self._GEO_COUNTRIES)
 117             raise ExtractorError(
 118                 '%s said: %s' % (self.IE_NAME, MESSAGES.get(
 119                     message_type, message_type)),
 120                 expected=True)
 121
 122         series = conviva.get('seriesName') or data.get('seriesTitle')
 123         episode = conviva.get('episodeName') or data.get('episodeNumberOrDate')
 124
 125         season_number = None
 126         episode_number = None
 127         if data.get('mediaElementType') == 'Episode':
 128             _season_episode = data.get('scoresStatistics', {}).get('springStreamStream') or \
 129                 data.get('relativeOriginUrl', '')
 130             EPISODENUM_RE = [
 131                 r'/s(?P<season>\d{,2})e(?P<episode>\d{,2})\.',
 132                 r'/sesong-(?P<season>\d{,2})/episode-(?P<episode>\d{,2})',
 133             ]
 134             season_number = int_or_none(self._search_regex(
 135                 EPISODENUM_RE, _season_episode, 'season number',
 136                 default=None, group='season'))
 137             episode_number = int_or_none(self._search_regex(
 138                 EPISODENUM_RE, _season_episode, 'episode number',
 139                 default=None, group='episode'))
 140
 141         thumbnails = None
 142         images = data.get('images')
 143         if images and isinstance(images, dict):
 144             web_images = images.get('webImages')
 145             if isinstance(web_images, list):
 146                 thumbnails = [{
 147                     'url': image['imageUrl'],
 148                     'width': int_or_none(image.get('width')),
 149                     'height': int_or_none(image.get('height')),
 150                 } for image in web_images if image.get('imageUrl')]
 151
 152         description = data.get('description')
 153         category = data.get('mediaAnalytics', {}).get('category')
 154
 155         common_info = {
 156             'description': description,
 157             'series': series,
 158             'episode': episode,
 159             'season_number': season_number,
 160             'episode_number': episode_number,
 161             'categories': [category] if category else None,
 162             'age_limit': parse_age_limit(data.get('legalAge')),
 163             'thumbnails': thumbnails,
 164         }
 165
 166         vcodec = 'none' if data.get('mediaType') == 'Audio' else None
 167
 168         for entry in entries:
 169             entry.update(common_info)
 170             for f in entry['formats']:
 171                 f['vcodec'] = vcodec
 172
 173         points = data.get('shortIndexPoints')
 174         if isinstance(points, list):
 175             chapters = []
 176             for next_num, point in enumerate(points, start=1):
 177                 if not isinstance(point, dict):
 178                     continue
 179                 start_time = parse_duration(point.get('startPoint'))
 180                 if start_time is None:
 181                     continue
 182                 end_time = parse_duration(
 183                     data.get('duration')
 184                     if next_num == len(points)
 185                     else points[next_num].get('startPoint'))
 186                 if end_time is None:
 187                     continue
 188                 chapters.append({
 189                     'start_time': start_time,
 190                     'end_time': end_time,
 191                     'title': point.get('title'),
 192                 })
 193             if chapters and len(entries) == 1:
 194                 entries[0]['chapters'] = chapters
 195
 196         return self.playlist_result(entries, video_id, title, description)
 197
 198
 199 class NRKIE(NRKBaseIE):
 200     _VALID_URL = r'''(?x)
 201                         (?:
 202                             nrk:|
 203                             https?://
 204                                 (?:
 205                                     (?:www\.)?nrk\.no/video/PS\*|
 206                                     v8[-.]psapi\.nrk\.no/mediaelement/
 207                                 )
 208                             )
 209                             (?P<id>[^?#&]+)
 210                         '''
 211     _API_HOSTS = ('psapi.nrk.no', 'v8-psapi.nrk.no')
 212     _TESTS = [{
 213         # video
 214         'url': 'http://www.nrk.no/video/PS*150533',
 215         'md5': '706f34cdf1322577589e369e522b50ef',
 216         'info_dict': {
 217             'id': '150533',
 218             'ext': 'mp4',
 219             'title': 'Dompap og andre fugler i Piip-Show',
 220             'description': 'md5:d9261ba34c43b61c812cb6b0269a5c8f',
 221             'duration': 262,
 222         }
 223     }, {
 224         # audio
 225         'url': 'http://www.nrk.no/video/PS*154915',
 226         # MD5 is unstable
 227         'info_dict': {
 228             'id': '154915',
 229             'ext': 'flv',
 230             'title': 'Slik høres internett ut når du er blind',
 231             'description': 'md5:a621f5cc1bd75c8d5104cb048c6b8568',
 232             'duration': 20,
 233         }
 234     }, {
 235         'url': 'nrk:ecc1b952-96dc-4a98-81b9-5296dc7a98d9',
 236         'only_matching': True,
 237     }, {
 238         'url': 'nrk:clip/7707d5a3-ebe7-434a-87d5-a3ebe7a34a70',
 239         'only_matching': True,
 240     }, {
 241         'url': 'https://v8-psapi.nrk.no/mediaelement/ecc1b952-96dc-4a98-81b9-5296dc7a98d9',
 242         'only_matching': True,
 243     }]
 244
 245
 246 class NRKTVIE(NRKBaseIE):
 247     IE_DESC = 'NRK TV and NRK Radio'
 248     _EPISODE_RE = r'(?P<id>[a-zA-Z]{4}\d{8})'
 249     _VALID_URL = r'''(?x)
 250                         https?://
 251                             (?:tv|radio)\.nrk(?:super)?\.no/
 252                             (?:serie(?:/[^/]+){1,2}|program)/
 253                             (?![Ee]pisodes)%s
 254                             (?:/\d{2}-\d{2}-\d{4})?
 255                             (?:\#del=(?P<part_id>\d+))?
 256                     ''' % _EPISODE_RE
 257     _API_HOSTS = ('psapi-ne.nrk.no', 'psapi-we.nrk.no')
 258     _TESTS = [{
 259         'url': 'https://tv.nrk.no/program/MDDP12000117',
 260         'md5': '8270824df46ec629b66aeaa5796b36fb',
 261         'info_dict': {
 262             'id': 'MDDP12000117AA',
 263             'ext': 'mp4',
 264             'title': 'Alarm Trolltunga',
 265             'description': 'md5:46923a6e6510eefcce23d5ef2a58f2ce',
 266             'duration': 2223,
 267             'age_limit': 6,
 268         },
 269     }, {
 270         'url': 'https://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
 271         'md5': '9a167e54d04671eb6317a37b7bc8a280',
 272         'info_dict': {
 273             'id': 'MUHH48000314AA',
 274             'ext': 'mp4',
 275             'title': '20 spørsmål 23.05.2014',
 276             'description': 'md5:bdea103bc35494c143c6a9acdd84887a',
 277             'duration': 1741,
 278             'series': '20 spørsmål',
 279             'episode': '23.05.2014',
 280         },
 281         'skip': 'NoProgramRights',
 282     }, {
 283         'url': 'https://tv.nrk.no/program/mdfp15000514',
 284         'info_dict': {
 285             'id': 'MDFP15000514CA',
 286             'ext': 'mp4',
 287             'title': 'Grunnlovsjubiléet - Stor ståhei for ingenting 24.05.2014',
 288             'description': 'md5:89290c5ccde1b3a24bb8050ab67fe1db',
 289             'duration': 4605,
 290             'series': 'Kunnskapskanalen',
 291             'episode': '24.05.2014',
 292         },
 293         'params': {
 294             'skip_download': True,
 295         },
 296     }, {
 297         # single playlist video
 298         'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2',
 299         'info_dict': {
 300             'id': 'MSPO40010515-part2',
 301             'ext': 'flv',
 302             'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
 303             'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 304         },
 305         'params': {
 306             'skip_download': True,
 307         },
 308         'expected_warnings': ['Video is geo restricted'],
 309         'skip': 'particular part is not supported currently',
 310     }, {
 311         'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015',
 312         'playlist': [{
 313             'info_dict': {
 314                 'id': 'MSPO40010515AH',
 315                 'ext': 'mp4',
 316                 'title': 'Sprint fri teknikk, kvinner og menn 06.01.2015 (Part 1)',
 317                 'description': 'md5:1f97a41f05a9486ee00c56f35f82993d',
 318                 'duration': 772,
 319                 'series': 'Tour de Ski',
 320                 'episode': '06.01.2015',
 321             },
 322             'params': {
 323                 'skip_download': True,
 324             },
 325         }, {
 326             'info_dict': {
 327                 'id': 'MSPO40010515BH',
 328                 'ext': 'mp4',
 329                 'title': 'Sprint fri teknikk, kvinner og menn 06.01.2015 (Part 2)',
 330                 'description': 'md5:1f97a41f05a9486ee00c56f35f82993d',
 331                 'duration': 6175,
 332                 'series': 'Tour de Ski',
 333                 'episode': '06.01.2015',
 334             },
 335             'params': {
 336                 'skip_download': True,
 337             },
 338         }],
 339         'info_dict': {
 340             'id': 'MSPO40010515',
 341             'title': 'Sprint fri teknikk, kvinner og menn 06.01.2015',
 342             'description': 'md5:1f97a41f05a9486ee00c56f35f82993d',
 343         },
 344         'expected_warnings': ['Video is geo restricted'],
 345     }, {
 346         'url': 'https://tv.nrk.no/serie/anno/KMTE50001317/sesong-3/episode-13',
 347         'info_dict': {
 348             'id': 'KMTE50001317AA',
 349             'ext': 'mp4',
 350             'title': 'Anno 13:30',
 351             'description': 'md5:11d9613661a8dbe6f9bef54e3a4cbbfa',
 352             'duration': 2340,
 353             'series': 'Anno',
 354             'episode': '13:30',
 355             'season_number': 3,
 356             'episode_number': 13,
 357         },
 358         'params': {
 359             'skip_download': True,
 360         },
 361     }, {
 362         'url': 'https://tv.nrk.no/serie/nytt-paa-nytt/MUHH46000317/27-01-2017',
 363         'info_dict': {
 364             'id': 'MUHH46000317AA',
 365             'ext': 'mp4',
 366             'title': 'Nytt på Nytt 27.01.2017',
 367             'description': 'md5:5358d6388fba0ea6f0b6d11c48b9eb4b',
 368             'duration': 1796,
 369             'series': 'Nytt på nytt',
 370             'episode': '27.01.2017',
 371         },
 372         'params': {
 373             'skip_download': True,
 374         },
 375     }, {
 376         'url': 'https://radio.nrk.no/serie/dagsnytt/NPUB21019315/12-07-2015#',
 377         'only_matching': True,
 378     }, {
 379         'url': 'https://tv.nrk.no/serie/lindmo/2018/MUHU11006318/avspiller',
 380         'only_matching': True,
 381     }]
 382
 383
 384 class NRKTVEpisodeIE(InfoExtractor):
 385     _VALID_URL = r'https?://tv\.nrk\.no/serie/(?P<id>[^/]+/sesong/\d+/episode/\d+)'
 386     _TESTS = [{
 387         'url': 'https://tv.nrk.no/serie/hellums-kro/sesong/1/episode/2',
 388         'info_dict': {
 389             'id': 'MUHH36005220BA',
 390             'ext': 'mp4',
 391             'title': 'Kro, krig og kjærlighet 2:6',
 392             'description': 'md5:b32a7dc0b1ed27c8064f58b97bda4350',
 393             'duration': 1563,
 394             'series': 'Hellums kro',
 395             'season_number': 1,
 396             'episode_number': 2,
 397             'episode': '2:6',
 398             'age_limit': 6,
 399         },
 400         'params': {
 401             'skip_download': True,
 402         },
 403     }, {
 404         'url': 'https://tv.nrk.no/serie/backstage/sesong/1/episode/8',
 405         'info_dict': {
 406             'id': 'MSUI14000816AA',
 407             'ext': 'mp4',
 408             'title': 'Backstage 8:30',
 409             'description': 'md5:de6ca5d5a2d56849e4021f2bf2850df4',
 410             'duration': 1320,
 411             'series': 'Backstage',
 412             'season_number': 1,
 413             'episode_number': 8,
 414             'episode': '8:30',
 415         },
 416         'params': {
 417             'skip_download': True,
 418         },
 419         'skip': 'ProgramRightsHasExpired',
 420     }]
 421
 422     def _real_extract(self, url):
 423         display_id = self._match_id(url)
 424
 425         webpage = self._download_webpage(url, display_id)
 426
 427         info = self._search_json_ld(webpage, display_id, default={})
 428         nrk_id = info.get('@id') or self._html_search_meta(
 429             'nrk:program-id', webpage, default=None) or self._search_regex(
 430             r'data-program-id=["\'](%s)' % NRKTVIE._EPISODE_RE, webpage,
 431             'nrk id')
 432         assert re.match(NRKTVIE._EPISODE_RE, nrk_id)
 433
 434         info.update({
 435             '_type': 'url_transparent',
 436             'id': nrk_id,
 437             'url': 'nrk:%s' % nrk_id,
 438             'ie_key': NRKIE.ie_key(),
 439         })
 440         return info
 441
 442
 443 class NRKTVSerieBaseIE(InfoExtractor):
 444     def _extract_series(self, webpage, display_id, fatal=True):
 445         config = self._parse_json(
 446             self._search_regex(
 447                 (r'INITIAL_DATA(?:_V\d)?_*\s*=\s*({.+?})\s*;',
 448                  r'({.+?})\s*,\s*"[^"]+"\s*\)\s*</script>'),
 449                 webpage, 'config', default='{}' if not fatal else NO_DEFAULT),
 450             display_id, fatal=False, transform_source=js_to_json)
 451         if not config:
 452             return
 453         return try_get(
 454             config,
 455             (lambda x: x['initialState']['series'], lambda x: x['series']),
 456             dict)
 457
 458     def _extract_seasons(self, seasons):
 459         if not isinstance(seasons, list):
 460             return []
 461         entries = []
 462         for season in seasons:
 463             entries.extend(self._extract_episodes(season))
 464         return entries
 465
 466     def _extract_episodes(self, season):
 467         if not isinstance(season, dict):
 468             return []
 469         return self._extract_entries(season.get('episodes'))
 470
 471     def _extract_entries(self, entry_list):
 472         if not isinstance(entry_list, list):
 473             return []
 474         entries = []
 475         for episode in entry_list:
 476             nrk_id = episode.get('prfId')
 477             if not nrk_id or not isinstance(nrk_id, compat_str):
 478                 continue
 479             entries.append(self.url_result(
 480                 'nrk:%s' % nrk_id, ie=NRKIE.ie_key(), video_id=nrk_id))
 481         return entries
 482
 483
 484 class NRKTVSeasonIE(NRKTVSerieBaseIE):
 485     _VALID_URL = r'https?://tv\.nrk\.no/serie/[^/]+/sesong/(?P<id>\d+)'
 486     _TEST = {
 487         'url': 'https://tv.nrk.no/serie/backstage/sesong/1',
 488         'info_dict': {
 489             'id': '1',
 490             'title': 'Sesong 1',
 491         },
 492         'playlist_mincount': 30,
 493     }
 494
 495     @classmethod
 496     def suitable(cls, url):
 497         return (False if NRKTVIE.suitable(url) or NRKTVEpisodeIE.suitable(url)
 498                 else super(NRKTVSeasonIE, cls).suitable(url))
 499
 500     def _real_extract(self, url):
 501         display_id = self._match_id(url)
 502
 503         webpage = self._download_webpage(url, display_id)
 504
 505         series = self._extract_series(webpage, display_id)
 506
 507         season = next(
 508             s for s in series['seasons']
 509             if int(display_id) == s.get('seasonNumber'))
 510
 511         title = try_get(season, lambda x: x['titles']['title'], compat_str)
 512         return self.playlist_result(
 513             self._extract_episodes(season), display_id, title)
 514
 515
 516 class NRKTVSeriesIE(NRKTVSerieBaseIE):
 517     _VALID_URL = r'https?://(?:tv|radio)\.nrk(?:super)?\.no/serie/(?P<id>[^/]+)'
 518     _ITEM_RE = r'(?:data-season=["\']|id=["\']season-)(?P<id>\d+)'
 519     _TESTS = [{
 520         'url': 'https://tv.nrk.no/serie/blank',
 521         'info_dict': {
 522             'id': 'blank',
 523             'title': 'Blank',
 524             'description': 'md5:7664b4e7e77dc6810cd3bca367c25b6e',
 525         },
 526         'playlist_mincount': 30,
 527     }, {
 528         # new layout, seasons
 529         'url': 'https://tv.nrk.no/serie/backstage',
 530         'info_dict': {
 531             'id': 'backstage',
 532             'title': 'Backstage',
 533             'description': 'md5:c3ec3a35736fca0f9e1207b5511143d3',
 534         },
 535         'playlist_mincount': 60,
 536     }, {
 537         # new layout, instalments
 538         'url': 'https://tv.nrk.no/serie/groenn-glede',
 539         'info_dict': {
 540             'id': 'groenn-glede',
 541             'title': 'Grønn glede',
 542             'description': 'md5:7576e92ae7f65da6993cf90ee29e4608',
 543         },
 544         'playlist_mincount': 10,
 545     }, {
 546         # old layout
 547         'url': 'https://tv.nrksuper.no/serie/labyrint',
 548         'info_dict': {
 549             'id': 'labyrint',
 550             'title': 'Labyrint',
 551             'description': 'md5:318b597330fdac5959247c9b69fdb1ec',
 552         },
 553         'playlist_mincount': 3,
 554     }, {
 555         'url': 'https://tv.nrk.no/serie/broedrene-dal-og-spektralsteinene',
 556         'only_matching': True,
 557     }, {
 558         'url': 'https://tv.nrk.no/serie/saving-the-human-race',
 559         'only_matching': True,
 560     }, {
 561         'url': 'https://tv.nrk.no/serie/postmann-pat',
 562         'only_matching': True,
 563     }]
 564
 565     @classmethod
 566     def suitable(cls, url):
 567         return (
 568             False if any(ie.suitable(url)
 569                          for ie in (NRKTVIE, NRKTVEpisodeIE, NRKTVSeasonIE))
 570             else super(NRKTVSeriesIE, cls).suitable(url))
 571
 572     def _real_extract(self, url):
 573         series_id = self._match_id(url)
 574
 575         webpage = self._download_webpage(url, series_id)
 576
 577         # New layout (e.g. https://tv.nrk.no/serie/backstage)
 578         series = self._extract_series(webpage, series_id, fatal=False)
 579         if series:
 580             title = try_get(series, lambda x: x['titles']['title'], compat_str)
 581             description = try_get(
 582                 series, lambda x: x['titles']['subtitle'], compat_str)
 583             entries = []
 584             entries.extend(self._extract_seasons(series.get('seasons')))
 585             entries.extend(self._extract_entries(series.get('instalments')))
 586             entries.extend(self._extract_episodes(series.get('extraMaterial')))
 587             return self.playlist_result(entries, series_id, title, description)
 588
 589         # Old layout (e.g. https://tv.nrksuper.no/serie/labyrint)
 590         entries = [
 591             self.url_result(
 592                 'https://tv.nrk.no/program/Episodes/{series}/{season}'.format(
 593                     series=series_id, season=season_id))
 594             for season_id in re.findall(self._ITEM_RE, webpage)
 595         ]
 596
 597         title = self._html_search_meta(
 598             'seriestitle', webpage,
 599             'title', default=None) or self._og_search_title(
 600             webpage, fatal=False)
 601         if title:
 602             title = self._search_regex(
 603                 r'NRK (?:Super )?TV\s*[-–]\s*(.+)', title, 'title', default=title)
 604
 605         description = self._html_search_meta(
 606             'series_description', webpage,
 607             'description', default=None) or self._og_search_description(webpage)
 608
 609         return self.playlist_result(entries, series_id, title, description)
 610
 611
 612 class NRKTVDirekteIE(NRKTVIE):
 613     IE_DESC = 'NRK TV Direkte and NRK Radio Direkte'
 614     _VALID_URL = r'https?://(?:tv|radio)\.nrk\.no/direkte/(?P<id>[^/?#&]+)'
 615
 616     _TESTS = [{
 617         'url': 'https://tv.nrk.no/direkte/nrk1',
 618         'only_matching': True,
 619     }, {
 620         'url': 'https://radio.nrk.no/direkte/p1_oslo_akershus',
 621         'only_matching': True,
 622     }]
 623
 624
 625 class NRKPlaylistBaseIE(InfoExtractor):
 626     def _extract_description(self, webpage):
 627         pass
 628
 629     def _real_extract(self, url):
 630         playlist_id = self._match_id(url)
 631
 632         webpage = self._download_webpage(url, playlist_id)
 633
 634         entries = [
 635             self.url_result('nrk:%s' % video_id, NRKIE.ie_key())
 636             for video_id in re.findall(self._ITEM_RE, webpage)
 637         ]
 638
 639         playlist_title = self. _extract_title(webpage)
 640         playlist_description = self._extract_description(webpage)
 641
 642         return self.playlist_result(
 643             entries, playlist_id, playlist_title, playlist_description)
 644
 645
 646 class NRKPlaylistIE(NRKPlaylistBaseIE):
 647     _VALID_URL = r'https?://(?:www\.)?nrk\.no/(?!video|skole)(?:[^/]+/)+(?P<id>[^/]+)'
 648     _ITEM_RE = r'class="[^"]*\brich\b[^"]*"[^>]+data-video-id="([^"]+)"'
 649     _TESTS = [{
 650         'url': 'http://www.nrk.no/troms/gjenopplev-den-historiske-solformorkelsen-1.12270763',
 651         'info_dict': {
 652             'id': 'gjenopplev-den-historiske-solformorkelsen-1.12270763',
 653             'title': 'Gjenopplev den historiske solformørkelsen',
 654             'description': 'md5:c2df8ea3bac5654a26fc2834a542feed',
 655         },
 656         'playlist_count': 2,
 657     }, {
 658         'url': 'http://www.nrk.no/kultur/bok/rivertonprisen-til-karin-fossum-1.12266449',
 659         'info_dict': {
 660             'id': 'rivertonprisen-til-karin-fossum-1.12266449',
 661             'title': 'Rivertonprisen til Karin Fossum',
 662             'description': 'Første kvinne på 15 år til å vinne krimlitteraturprisen.',
 663         },
 664         'playlist_count': 2,
 665     }]
 666
 667     def _extract_title(self, webpage):
 668         return self._og_search_title(webpage, fatal=False)
 669
 670     def _extract_description(self, webpage):
 671         return self._og_search_description(webpage)
 672
 673
 674 class NRKTVEpisodesIE(NRKPlaylistBaseIE):
 675     _VALID_URL = r'https?://tv\.nrk\.no/program/[Ee]pisodes/[^/]+/(?P<id>\d+)'
 676     _ITEM_RE = r'data-episode=["\']%s' % NRKTVIE._EPISODE_RE
 677     _TESTS = [{
 678         'url': 'https://tv.nrk.no/program/episodes/nytt-paa-nytt/69031',
 679         'info_dict': {
 680             'id': '69031',
 681             'title': 'Nytt på nytt, sesong: 201210',
 682         },
 683         'playlist_count': 4,
 684     }]
 685
 686     def _extract_title(self, webpage):
 687         return self._html_search_regex(
 688             r'<h1>([^<]+)</h1>', webpage, 'title', fatal=False)
 689
 690
 691 class NRKSkoleIE(InfoExtractor):
 692     IE_DESC = 'NRK Skole'
 693     _VALID_URL = r'https?://(?:www\.)?nrk\.no/skole/?\?.*\bmediaId=(?P<id>\d+)'
 694
 695     _TESTS = [{
 696         'url': 'https://www.nrk.no/skole/?page=search&q=&mediaId=14099',
 697         'md5': '18c12c3d071953c3bf8d54ef6b2587b7',
 698         'info_dict': {
 699             'id': '6021',
 700             'ext': 'mp4',
 701             'title': 'Genetikk og eneggede tvillinger',
 702             'description': 'md5:3aca25dcf38ec30f0363428d2b265f8d',
 703             'duration': 399,
 704         },
 705     }, {
 706         'url': 'https://www.nrk.no/skole/?page=objectives&subject=naturfag&objective=K15114&mediaId=19355',
 707         'only_matching': True,
 708     }]
 709
 710     def _real_extract(self, url):
 711         video_id = self._match_id(url)
 712
 713         webpage = self._download_webpage(
 714             'https://mimir.nrk.no/plugin/1.0/static?mediaId=%s' % video_id,
 715             video_id)
 716
 717         nrk_id = self._parse_json(
 718             self._search_regex(
 719                 r'<script[^>]+type=["\']application/json["\'][^>]*>({.+?})</script>',
 720                 webpage, 'application json'),
 721             video_id)['activeMedia']['psId']
 722
 723         return self.url_result('nrk:%s' % nrk_id)