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     JSON_LD_RE,
  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' or
  49                 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                 'ProgramIsGeoBlocked': 'NRK har ikke rettigheter til å vise dette programmet utenfor Norge',
 109             }
 110             message_type = data.get('messageType', '')
 111             # Can be ProgramIsGeoBlocked or ChannelIsGeoBlocked*
 112             if 'IsGeoBlocked' in message_type:
 113                 self.raise_geo_restricted(
 114                     msg=MESSAGES.get('ProgramIsGeoBlocked'),
 115                     countries=self._GEO_COUNTRIES)
 116             raise ExtractorError(
 117                 '%s said: %s' % (self.IE_NAME, MESSAGES.get(
 118                     message_type, message_type)),
 119                 expected=True)
 120
 121         series = conviva.get('seriesName') or data.get('seriesTitle')
 122         episode = conviva.get('episodeName') or data.get('episodeNumberOrDate')
 123
 124         season_number = None
 125         episode_number = None
 126         if data.get('mediaElementType') == 'Episode':
 127             _season_episode = data.get('scoresStatistics', {}).get('springStreamStream') or \
 128                 data.get('relativeOriginUrl', '')
 129             EPISODENUM_RE = [
 130                 r'/s(?P<season>\d{,2})e(?P<episode>\d{,2})\.',
 131                 r'/sesong-(?P<season>\d{,2})/episode-(?P<episode>\d{,2})',
 132             ]
 133             season_number = int_or_none(self._search_regex(
 134                 EPISODENUM_RE, _season_episode, 'season number',
 135                 default=None, group='season'))
 136             episode_number = int_or_none(self._search_regex(
 137                 EPISODENUM_RE, _season_episode, 'episode number',
 138                 default=None, group='episode'))
 139
 140         thumbnails = None
 141         images = data.get('images')
 142         if images and isinstance(images, dict):
 143             web_images = images.get('webImages')
 144             if isinstance(web_images, list):
 145                 thumbnails = [{
 146                     'url': image['imageUrl'],
 147                     'width': int_or_none(image.get('width')),
 148                     'height': int_or_none(image.get('height')),
 149                 } for image in web_images if image.get('imageUrl')]
 150
 151         description = data.get('description')
 152         category = data.get('mediaAnalytics', {}).get('category')
 153
 154         common_info = {
 155             'description': description,
 156             'series': series,
 157             'episode': episode,
 158             'season_number': season_number,
 159             'episode_number': episode_number,
 160             'categories': [category] if category else None,
 161             'age_limit': parse_age_limit(data.get('legalAge')),
 162             'thumbnails': thumbnails,
 163         }
 164
 165         vcodec = 'none' if data.get('mediaType') == 'Audio' else None
 166
 167         for entry in entries:
 168             entry.update(common_info)
 169             for f in entry['formats']:
 170                 f['vcodec'] = vcodec
 171
 172         points = data.get('shortIndexPoints')
 173         if isinstance(points, list):
 174             chapters = []
 175             for next_num, point in enumerate(points, start=1):
 176                 if not isinstance(point, dict):
 177                     continue
 178                 start_time = parse_duration(point.get('startPoint'))
 179                 if start_time is None:
 180                     continue
 181                 end_time = parse_duration(
 182                     data.get('duration')
 183                     if next_num == len(points)
 184                     else points[next_num].get('startPoint'))
 185                 if end_time is None:
 186                     continue
 187                 chapters.append({
 188                     'start_time': start_time,
 189                     'end_time': end_time,
 190                     'title': point.get('title'),
 191                 })
 192             if chapters and len(entries) == 1:
 193                 entries[0]['chapters'] = chapters
 194
 195         return self.playlist_result(entries, video_id, title, description)
 196
 197
 198 class NRKIE(NRKBaseIE):
 199     _VALID_URL = r'''(?x)
 200                         (?:
 201                             nrk:|
 202                             https?://
 203                                 (?:
 204                                     (?:www\.)?nrk\.no/video/PS\*|
 205                                     v8[-.]psapi\.nrk\.no/mediaelement/
 206                                 )
 207                             )
 208                             (?P<id>[^?#&]+)
 209                         '''
 210     _API_HOSTS = ('psapi.nrk.no', 'v8-psapi.nrk.no')
 211     _TESTS = [{
 212         # video
 213         'url': 'http://www.nrk.no/video/PS*150533',
 214         'md5': '2f7f6eeb2aacdd99885f355428715cfa',
 215         'info_dict': {
 216             'id': '150533',
 217             'ext': 'mp4',
 218             'title': 'Dompap og andre fugler i Piip-Show',
 219             'description': 'md5:d9261ba34c43b61c812cb6b0269a5c8f',
 220             'duration': 263,
 221         }
 222     }, {
 223         # audio
 224         'url': 'http://www.nrk.no/video/PS*154915',
 225         # MD5 is unstable
 226         'info_dict': {
 227             'id': '154915',
 228             'ext': 'flv',
 229             'title': 'Slik høres internett ut når du er blind',
 230             'description': 'md5:a621f5cc1bd75c8d5104cb048c6b8568',
 231             'duration': 20,
 232         }
 233     }, {
 234         'url': 'nrk:ecc1b952-96dc-4a98-81b9-5296dc7a98d9',
 235         'only_matching': True,
 236     }, {
 237         'url': 'nrk:clip/7707d5a3-ebe7-434a-87d5-a3ebe7a34a70',
 238         'only_matching': True,
 239     }, {
 240         'url': 'https://v8-psapi.nrk.no/mediaelement/ecc1b952-96dc-4a98-81b9-5296dc7a98d9',
 241         'only_matching': True,
 242     }]
 243
 244
 245 class NRKTVIE(NRKBaseIE):
 246     IE_DESC = 'NRK TV and NRK Radio'
 247     _EPISODE_RE = r'(?P<id>[a-zA-Z]{4}\d{8})'
 248     _VALID_URL = r'''(?x)
 249                         https?://
 250                             (?:tv|radio)\.nrk(?:super)?\.no/
 251                             (?:serie/[^/]+|program)/
 252                             (?![Ee]pisodes)%s
 253                             (?:/\d{2}-\d{2}-\d{4})?
 254                             (?:\#del=(?P<part_id>\d+))?
 255                     ''' % _EPISODE_RE
 256     _API_HOSTS = ('psapi-ne.nrk.no', 'psapi-we.nrk.no')
 257     _TESTS = [{
 258         'url': 'https://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
 259         'md5': '4e9ca6629f09e588ed240fb11619922a',
 260         'info_dict': {
 261             'id': 'MUHH48000314AA',
 262             'ext': 'mp4',
 263             'title': '20 spørsmål 23.05.2014',
 264             'description': 'md5:bdea103bc35494c143c6a9acdd84887a',
 265             'duration': 1741,
 266             'series': '20 spørsmål - TV',
 267             'episode': '23.05.2014',
 268         },
 269     }, {
 270         'url': 'https://tv.nrk.no/program/mdfp15000514',
 271         'info_dict': {
 272             'id': 'MDFP15000514CA',
 273             'ext': 'mp4',
 274             'title': 'Grunnlovsjubiléet - Stor ståhei for ingenting 24.05.2014',
 275             'description': 'md5:89290c5ccde1b3a24bb8050ab67fe1db',
 276             'duration': 4605,
 277             'series': 'Kunnskapskanalen',
 278             'episode': '24.05.2014',
 279         },
 280         'params': {
 281             'skip_download': True,
 282         },
 283     }, {
 284         # single playlist video
 285         'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2',
 286         'info_dict': {
 287             'id': 'MSPO40010515-part2',
 288             'ext': 'flv',
 289             'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
 290             'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 291         },
 292         'params': {
 293             'skip_download': True,
 294         },
 295         'expected_warnings': ['Video is geo restricted'],
 296         'skip': 'particular part is not supported currently',
 297     }, {
 298         'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015',
 299         'playlist': [{
 300             'info_dict': {
 301                 'id': 'MSPO40010515AH',
 302                 'ext': 'mp4',
 303                 'title': 'Sprint fri teknikk, kvinner og menn 06.01.2015 (Part 1)',
 304                 'description': 'md5:c03aba1e917561eface5214020551b7a',
 305                 'duration': 772,
 306                 'series': 'Tour de Ski',
 307                 'episode': '06.01.2015',
 308             },
 309             'params': {
 310                 'skip_download': True,
 311             },
 312         }, {
 313             'info_dict': {
 314                 'id': 'MSPO40010515BH',
 315                 'ext': 'mp4',
 316                 'title': 'Sprint fri teknikk, kvinner og menn 06.01.2015 (Part 2)',
 317                 'description': 'md5:c03aba1e917561eface5214020551b7a',
 318                 'duration': 6175,
 319                 'series': 'Tour de Ski',
 320                 'episode': '06.01.2015',
 321             },
 322             'params': {
 323                 'skip_download': True,
 324             },
 325         }],
 326         'info_dict': {
 327             'id': 'MSPO40010515',
 328             'title': 'Sprint fri teknikk, kvinner og menn 06.01.2015',
 329             'description': 'md5:c03aba1e917561eface5214020551b7a',
 330         },
 331         'expected_warnings': ['Video is geo restricted'],
 332     }, {
 333         'url': 'https://tv.nrk.no/serie/anno/KMTE50001317/sesong-3/episode-13',
 334         'info_dict': {
 335             'id': 'KMTE50001317AA',
 336             'ext': 'mp4',
 337             'title': 'Anno 13:30',
 338             'description': 'md5:11d9613661a8dbe6f9bef54e3a4cbbfa',
 339             'duration': 2340,
 340             'series': 'Anno',
 341             'episode': '13:30',
 342             'season_number': 3,
 343             'episode_number': 13,
 344         },
 345         'params': {
 346             'skip_download': True,
 347         },
 348     }, {
 349         'url': 'https://tv.nrk.no/serie/nytt-paa-nytt/MUHH46000317/27-01-2017',
 350         'info_dict': {
 351             'id': 'MUHH46000317AA',
 352             'ext': 'mp4',
 353             'title': 'Nytt på Nytt 27.01.2017',
 354             'description': 'md5:5358d6388fba0ea6f0b6d11c48b9eb4b',
 355             'duration': 1796,
 356             'series': 'Nytt på nytt',
 357             'episode': '27.01.2017',
 358         },
 359         'params': {
 360             'skip_download': True,
 361         },
 362     }, {
 363         'url': 'https://radio.nrk.no/serie/dagsnytt/NPUB21019315/12-07-2015#',
 364         'only_matching': True,
 365     }]
 366
 367
 368 class NRKTVEpisodeIE(InfoExtractor):
 369     _VALID_URL = r'https?://tv\.nrk\.no/serie/(?P<id>[^/]+/sesong/\d+/episode/\d+)'
 370     _TEST = {
 371         'url': 'https://tv.nrk.no/serie/backstage/sesong/1/episode/8',
 372         'info_dict': {
 373             'id': 'MSUI14000816AA',
 374             'ext': 'mp4',
 375             'title': 'Backstage 8:30',
 376             'description': 'md5:de6ca5d5a2d56849e4021f2bf2850df4',
 377             'duration': 1320,
 378             'series': 'Backstage',
 379             'season_number': 1,
 380             'episode_number': 8,
 381             'episode': '8:30',
 382         },
 383         'params': {
 384             'skip_download': True,
 385         },
 386     }
 387
 388     def _real_extract(self, url):
 389         display_id = self._match_id(url)
 390
 391         webpage = self._download_webpage(url, display_id)
 392
 393         nrk_id = self._parse_json(
 394             self._search_regex(JSON_LD_RE, webpage, 'JSON-LD', group='json_ld'),
 395             display_id)['@id']
 396
 397         assert re.match(NRKTVIE._EPISODE_RE, nrk_id)
 398         return self.url_result(
 399             'nrk:%s' % nrk_id, ie=NRKIE.ie_key(), video_id=nrk_id)
 400
 401
 402 class NRKTVSerieBaseIE(InfoExtractor):
 403     def _extract_series(self, webpage, display_id, fatal=True):
 404         config = self._parse_json(
 405             self._search_regex(
 406                 r'({.+?})\s*,\s*"[^"]+"\s*\)\s*</script>', webpage, 'config',
 407                 default='{}' if not fatal else NO_DEFAULT),
 408             display_id, fatal=False)
 409         if not config:
 410             return
 411         return try_get(config, lambda x: x['series'], dict)
 412
 413     def _extract_episodes(self, season):
 414         entries = []
 415         if not isinstance(season, dict):
 416             return entries
 417         episodes = season.get('episodes')
 418         if not isinstance(episodes, list):
 419             return entries
 420         for episode in episodes:
 421             nrk_id = episode.get('prfId')
 422             if not nrk_id or not isinstance(nrk_id, compat_str):
 423                 continue
 424             entries.append(self.url_result(
 425                 'nrk:%s' % nrk_id, ie=NRKIE.ie_key(), video_id=nrk_id))
 426         return entries
 427
 428
 429 class NRKTVSeasonIE(NRKTVSerieBaseIE):
 430     _VALID_URL = r'https?://tv\.nrk\.no/serie/[^/]+/sesong/(?P<id>\d+)'
 431     _TEST = {
 432         'url': 'https://tv.nrk.no/serie/backstage/sesong/1',
 433         'info_dict': {
 434             'id': '1',
 435             'title': 'Sesong 1',
 436         },
 437         'playlist_mincount': 30,
 438     }
 439
 440     @classmethod
 441     def suitable(cls, url):
 442         return (False if NRKTVIE.suitable(url) or NRKTVEpisodeIE.suitable(url)
 443                 else super(NRKTVSeasonIE, cls).suitable(url))
 444
 445     def _real_extract(self, url):
 446         display_id = self._match_id(url)
 447
 448         webpage = self._download_webpage(url, display_id)
 449
 450         series = self._extract_series(webpage, display_id)
 451
 452         season = next(
 453             s for s in series['seasons']
 454             if int(display_id) == s.get('seasonNumber'))
 455
 456         title = try_get(season, lambda x: x['titles']['title'], compat_str)
 457         return self.playlist_result(
 458             self._extract_episodes(season), display_id, title)
 459
 460
 461 class NRKTVSeriesIE(NRKTVSerieBaseIE):
 462     _VALID_URL = r'https?://(?:tv|radio)\.nrk(?:super)?\.no/serie/(?P<id>[^/]+)'
 463     _ITEM_RE = r'(?:data-season=["\']|id=["\']season-)(?P<id>\d+)'
 464     _TESTS = [{
 465         # new layout
 466         'url': 'https://tv.nrk.no/serie/backstage',
 467         'info_dict': {
 468             'id': 'backstage',
 469             'title': 'Backstage',
 470             'description': 'md5:c3ec3a35736fca0f9e1207b5511143d3',
 471         },
 472         'playlist_mincount': 60,
 473     }, {
 474         # old layout
 475         'url': 'https://tv.nrk.no/serie/groenn-glede',
 476         'info_dict': {
 477             'id': 'groenn-glede',
 478             'title': 'Grønn glede',
 479             'description': 'md5:7576e92ae7f65da6993cf90ee29e4608',
 480         },
 481         'playlist_mincount': 9,
 482     }, {
 483         'url': 'http://tv.nrksuper.no/serie/labyrint',
 484         'info_dict': {
 485             'id': 'labyrint',
 486             'title': 'Labyrint',
 487             'description': 'md5:58afd450974c89e27d5a19212eee7115',
 488         },
 489         'playlist_mincount': 3,
 490     }, {
 491         'url': 'https://tv.nrk.no/serie/broedrene-dal-og-spektralsteinene',
 492         'only_matching': True,
 493     }, {
 494         'url': 'https://tv.nrk.no/serie/saving-the-human-race',
 495         'only_matching': True,
 496     }, {
 497         'url': 'https://tv.nrk.no/serie/postmann-pat',
 498         'only_matching': True,
 499     }]
 500
 501     @classmethod
 502     def suitable(cls, url):
 503         return (
 504             False if any(ie.suitable(url)
 505                          for ie in (NRKTVIE, NRKTVEpisodeIE, NRKTVSeasonIE))
 506             else super(NRKTVSeriesIE, cls).suitable(url))
 507
 508     def _real_extract(self, url):
 509         series_id = self._match_id(url)
 510
 511         webpage = self._download_webpage(url, series_id)
 512
 513         # New layout (e.g. https://tv.nrk.no/serie/backstage)
 514         series = self._extract_series(webpage, series_id, fatal=False)
 515         if series:
 516             title = try_get(series, lambda x: x['titles']['title'], compat_str)
 517             description = try_get(
 518                 series, lambda x: x['titles']['subtitle'], compat_str)
 519             entries = []
 520             for season in series['seasons']:
 521                 entries.extend(self._extract_episodes(season))
 522             return self.playlist_result(entries, series_id, title, description)
 523
 524         # Old layout (e.g. https://tv.nrk.no/serie/groenn-glede)
 525         entries = [
 526             self.url_result(
 527                 'https://tv.nrk.no/program/Episodes/{series}/{season}'.format(
 528                     series=series_id, season=season_id))
 529             for season_id in re.findall(self._ITEM_RE, webpage)
 530         ]
 531
 532         title = self._html_search_meta(
 533             'seriestitle', webpage,
 534             'title', default=None) or self._og_search_title(
 535             webpage, fatal=False)
 536
 537         description = self._html_search_meta(
 538             'series_description', webpage,
 539             'description', default=None) or self._og_search_description(webpage)
 540
 541         return self.playlist_result(entries, series_id, title, description)
 542
 543
 544 class NRKTVDirekteIE(NRKTVIE):
 545     IE_DESC = 'NRK TV Direkte and NRK Radio Direkte'
 546     _VALID_URL = r'https?://(?:tv|radio)\.nrk\.no/direkte/(?P<id>[^/?#&]+)'
 547
 548     _TESTS = [{
 549         'url': 'https://tv.nrk.no/direkte/nrk1',
 550         'only_matching': True,
 551     }, {
 552         'url': 'https://radio.nrk.no/direkte/p1_oslo_akershus',
 553         'only_matching': True,
 554     }]
 555
 556
 557 class NRKPlaylistBaseIE(InfoExtractor):
 558     def _extract_description(self, webpage):
 559         pass
 560
 561     def _real_extract(self, url):
 562         playlist_id = self._match_id(url)
 563
 564         webpage = self._download_webpage(url, playlist_id)
 565
 566         entries = [
 567             self.url_result('nrk:%s' % video_id, NRKIE.ie_key())
 568             for video_id in re.findall(self._ITEM_RE, webpage)
 569         ]
 570
 571         playlist_title = self. _extract_title(webpage)
 572         playlist_description = self._extract_description(webpage)
 573
 574         return self.playlist_result(
 575             entries, playlist_id, playlist_title, playlist_description)
 576
 577
 578 class NRKPlaylistIE(NRKPlaylistBaseIE):
 579     _VALID_URL = r'https?://(?:www\.)?nrk\.no/(?!video|skole)(?:[^/]+/)+(?P<id>[^/]+)'
 580     _ITEM_RE = r'class="[^"]*\brich\b[^"]*"[^>]+data-video-id="([^"]+)"'
 581     _TESTS = [{
 582         'url': 'http://www.nrk.no/troms/gjenopplev-den-historiske-solformorkelsen-1.12270763',
 583         'info_dict': {
 584             'id': 'gjenopplev-den-historiske-solformorkelsen-1.12270763',
 585             'title': 'Gjenopplev den historiske solformørkelsen',
 586             'description': 'md5:c2df8ea3bac5654a26fc2834a542feed',
 587         },
 588         'playlist_count': 2,
 589     }, {
 590         'url': 'http://www.nrk.no/kultur/bok/rivertonprisen-til-karin-fossum-1.12266449',
 591         'info_dict': {
 592             'id': 'rivertonprisen-til-karin-fossum-1.12266449',
 593             'title': 'Rivertonprisen til Karin Fossum',
 594             'description': 'Første kvinne på 15 år til å vinne krimlitteraturprisen.',
 595         },
 596         'playlist_count': 5,
 597     }]
 598
 599     def _extract_title(self, webpage):
 600         return self._og_search_title(webpage, fatal=False)
 601
 602     def _extract_description(self, webpage):
 603         return self._og_search_description(webpage)
 604
 605
 606 class NRKTVEpisodesIE(NRKPlaylistBaseIE):
 607     _VALID_URL = r'https?://tv\.nrk\.no/program/[Ee]pisodes/[^/]+/(?P<id>\d+)'
 608     _ITEM_RE = r'data-episode=["\']%s' % NRKTVIE._EPISODE_RE
 609     _TESTS = [{
 610         'url': 'https://tv.nrk.no/program/episodes/nytt-paa-nytt/69031',
 611         'info_dict': {
 612             'id': '69031',
 613             'title': 'Nytt på nytt, sesong: 201210',
 614         },
 615         'playlist_count': 4,
 616     }]
 617
 618     def _extract_title(self, webpage):
 619         return self._html_search_regex(
 620             r'<h1>([^<]+)</h1>', webpage, 'title', fatal=False)
 621
 622
 623 class NRKSkoleIE(InfoExtractor):
 624     IE_DESC = 'NRK Skole'
 625     _VALID_URL = r'https?://(?:www\.)?nrk\.no/skole/?\?.*\bmediaId=(?P<id>\d+)'
 626
 627     _TESTS = [{
 628         'url': 'https://www.nrk.no/skole/?page=search&q=&mediaId=14099',
 629         'md5': '6bc936b01f9dd8ed45bc58b252b2d9b6',
 630         'info_dict': {
 631             'id': '6021',
 632             'ext': 'mp4',
 633             'title': 'Genetikk og eneggede tvillinger',
 634             'description': 'md5:3aca25dcf38ec30f0363428d2b265f8d',
 635             'duration': 399,
 636         },
 637     }, {
 638         'url': 'https://www.nrk.no/skole/?page=objectives&subject=naturfag&objective=K15114&mediaId=19355',
 639         'only_matching': True,
 640     }]
 641
 642     def _real_extract(self, url):
 643         video_id = self._match_id(url)
 644
 645         webpage = self._download_webpage(
 646             'https://mimir.nrk.no/plugin/1.0/static?mediaId=%s' % video_id,
 647             video_id)
 648
 649         nrk_id = self._parse_json(
 650             self._search_regex(
 651                 r'<script[^>]+type=["\']application/json["\'][^>]*>({.+?})</script>',
 652                 webpage, 'application json'),
 653             video_id)['activeMedia']['psId']
 654
 655         return self.url_result('nrk:%s' % nrk_id)