youtube_dlc/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     determine_ext,
  13     ExtractorError,
  14     int_or_none,
  15     js_to_json,
  16     NO_DEFAULT,
  17     parse_age_limit,
  18     parse_duration,
  19     try_get,
  20     url_or_none,
  21 )
  22
  23
  24 class NRKBaseIE(InfoExtractor):
  25     _GEO_COUNTRIES = ['NO']
  26
  27
  28 class NRKIE(NRKBaseIE):
  29     _VALID_URL = r'''(?x)
  30                         (?:
  31                             nrk:|
  32                             https?://
  33                                 (?:
  34                                     (?:www\.)?nrk\.no/video/(?:PS\*|[^_]+_)|
  35                                     v8[-.]psapi\.nrk\.no/mediaelement/
  36                                 )
  37                             )
  38                             (?P<id>[^?\#&]+)
  39                         '''
  40
  41     _TESTS = [{
  42         # video
  43         'url': 'http://www.nrk.no/video/PS*150533',
  44         'md5': '706f34cdf1322577589e369e522b50ef',
  45         'info_dict': {
  46             'id': '150533',
  47             'ext': 'mp4',
  48             'title': 'Dompap og andre fugler i Piip-Show',
  49             'description': 'md5:d9261ba34c43b61c812cb6b0269a5c8f',
  50             'duration': 262,
  51         }
  52     }, {
  53         # audio
  54         'url': 'http://www.nrk.no/video/PS*154915',
  55         # MD5 is unstable
  56         'info_dict': {
  57             'id': '154915',
  58             'ext': 'flv',
  59             'title': 'Slik høres internett ut når du er blind',
  60             'description': 'md5:a621f5cc1bd75c8d5104cb048c6b8568',
  61             'duration': 20,
  62         }
  63     }, {
  64         'url': 'nrk:ecc1b952-96dc-4a98-81b9-5296dc7a98d9',
  65         'only_matching': True,
  66     }, {
  67         'url': 'nrk:clip/7707d5a3-ebe7-434a-87d5-a3ebe7a34a70',
  68         'only_matching': True,
  69     }, {
  70         'url': 'https://v8-psapi.nrk.no/mediaelement/ecc1b952-96dc-4a98-81b9-5296dc7a98d9',
  71         'only_matching': True,
  72     }, {
  73         'url': 'https://www.nrk.no/video/dompap-og-andre-fugler-i-piip-show_150533',
  74         'only_matching': True,
  75     }, {
  76         'url': 'https://www.nrk.no/video/humor/kommentatorboksen-reiser-til-sjos_d1fda11f-a4ad-437a-a374-0398bc84e999',
  77         'only_matching': True,
  78     }]
  79
  80     def _extract_from_playback(self, video_id):
  81         manifest = self._download_json(
  82             'http://psapi.nrk.no/playback/manifest/%s' % video_id,
  83             video_id, 'Downloading manifest JSON')
  84
  85         playable = manifest['playable']
  86
  87         formats = []
  88         for asset in playable['assets']:
  89             if not isinstance(asset, dict):
  90                 continue
  91             if asset.get('encrypted'):
  92                 continue
  93             format_url = url_or_none(asset.get('url'))
  94             if not format_url:
  95                 continue
  96             if asset.get('format') == 'HLS' or determine_ext(format_url) == 'm3u8':
  97                 formats.extend(self._extract_m3u8_formats(
  98                     format_url, video_id, 'mp4', entry_protocol='m3u8_native',
  99                     m3u8_id='hls', fatal=False))
 100         self._sort_formats(formats)
 101
 102         data = self._download_json(
 103             'http://psapi.nrk.no/playback/metadata/%s' % video_id,
 104             video_id, 'Downloading metadata JSON')
 105
 106         preplay = data['preplay']
 107         titles = preplay['titles']
 108         title = titles['title']
 109         alt_title = titles.get('subtitle')
 110
 111         description = preplay.get('description')
 112         duration = parse_duration(playable.get('duration')) or parse_duration(data.get('duration'))
 113
 114         thumbnails = []
 115         for image in try_get(
 116                 preplay, lambda x: x['poster']['images'], list) or []:
 117             if not isinstance(image, dict):
 118                 continue
 119             image_url = url_or_none(image.get('url'))
 120             if not image_url:
 121                 continue
 122             thumbnails.append({
 123                 'url': image_url,
 124                 'width': int_or_none(image.get('pixelWidth')),
 125                 'height': int_or_none(image.get('pixelHeight')),
 126             })
 127
 128         return {
 129             'id': video_id,
 130             'title': title,
 131             'alt_title': alt_title,
 132             'description': description,
 133             'duration': duration,
 134             'thumbnails': thumbnails,
 135             'formats': formats,
 136         }
 137
 138     def _real_extract(self, url):
 139         video_id = self._match_id(url)
 140         return self._extract_from_playback(video_id)
 141
 142
 143 class NRKTVIE(NRKBaseIE):
 144     IE_DESC = 'NRK TV and NRK Radio'
 145     _EPISODE_RE = r'(?P<id>[a-zA-Z]{4}\d{8})'
 146     _VALID_URL = r'''(?x)
 147                         https?://
 148                             (?:tv|radio)\.nrk(?:super)?\.no/
 149                             (?:serie(?:/[^/]+){1,2}|program)/
 150                             (?![Ee]pisodes)%s
 151                             (?:/\d{2}-\d{2}-\d{4})?
 152                             (?:\#del=(?P<part_id>\d+))?
 153                     ''' % _EPISODE_RE
 154     _API_HOSTS = ('psapi-ne.nrk.no', 'psapi-we.nrk.no')
 155     _TESTS = [{
 156         'url': 'https://tv.nrk.no/program/MDDP12000117',
 157         'md5': '8270824df46ec629b66aeaa5796b36fb',
 158         'info_dict': {
 159             'id': 'MDDP12000117AA',
 160             'ext': 'mp4',
 161             'title': 'Alarm Trolltunga',
 162             'description': 'md5:46923a6e6510eefcce23d5ef2a58f2ce',
 163             'duration': 2223,
 164             'age_limit': 6,
 165         },
 166     }, {
 167         'url': 'https://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
 168         'md5': '9a167e54d04671eb6317a37b7bc8a280',
 169         'info_dict': {
 170             'id': 'MUHH48000314AA',
 171             'ext': 'mp4',
 172             'title': '20 spørsmål 23.05.2014',
 173             'description': 'md5:bdea103bc35494c143c6a9acdd84887a',
 174             'duration': 1741,
 175             'series': '20 spørsmål',
 176             'episode': '23.05.2014',
 177         },
 178         'skip': 'NoProgramRights',
 179     }, {
 180         'url': 'https://tv.nrk.no/program/mdfp15000514',
 181         'info_dict': {
 182             'id': 'MDFP15000514CA',
 183             'ext': 'mp4',
 184             'title': 'Grunnlovsjubiléet - Stor ståhei for ingenting 24.05.2014',
 185             'description': 'md5:89290c5ccde1b3a24bb8050ab67fe1db',
 186             'duration': 4605,
 187             'series': 'Kunnskapskanalen',
 188             'episode': '24.05.2014',
 189         },
 190         'params': {
 191             'skip_download': True,
 192         },
 193     }, {
 194         # single playlist video
 195         'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2',
 196         'info_dict': {
 197             'id': 'MSPO40010515-part2',
 198             'ext': 'flv',
 199             'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
 200             'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 201         },
 202         'params': {
 203             'skip_download': True,
 204         },
 205         'expected_warnings': ['Video is geo restricted'],
 206         'skip': 'particular part is not supported currently',
 207     }, {
 208         'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015',
 209         'playlist': [{
 210             'info_dict': {
 211                 'id': 'MSPO40010515AH',
 212                 'ext': 'mp4',
 213                 'title': 'Sprint fri teknikk, kvinner og menn 06.01.2015 (Part 1)',
 214                 'description': 'md5:1f97a41f05a9486ee00c56f35f82993d',
 215                 'duration': 772,
 216                 'series': 'Tour de Ski',
 217                 'episode': '06.01.2015',
 218             },
 219             'params': {
 220                 'skip_download': True,
 221             },
 222         }, {
 223             'info_dict': {
 224                 'id': 'MSPO40010515BH',
 225                 'ext': 'mp4',
 226                 'title': 'Sprint fri teknikk, kvinner og menn 06.01.2015 (Part 2)',
 227                 'description': 'md5:1f97a41f05a9486ee00c56f35f82993d',
 228                 'duration': 6175,
 229                 'series': 'Tour de Ski',
 230                 'episode': '06.01.2015',
 231             },
 232             'params': {
 233                 'skip_download': True,
 234             },
 235         }],
 236         'info_dict': {
 237             'id': 'MSPO40010515',
 238             'title': 'Sprint fri teknikk, kvinner og menn 06.01.2015',
 239             'description': 'md5:1f97a41f05a9486ee00c56f35f82993d',
 240         },
 241         'expected_warnings': ['Video is geo restricted'],
 242     }, {
 243         'url': 'https://tv.nrk.no/serie/anno/KMTE50001317/sesong-3/episode-13',
 244         'info_dict': {
 245             'id': 'KMTE50001317AA',
 246             'ext': 'mp4',
 247             'title': 'Anno 13:30',
 248             'description': 'md5:11d9613661a8dbe6f9bef54e3a4cbbfa',
 249             'duration': 2340,
 250             'series': 'Anno',
 251             'episode': '13:30',
 252             'season_number': 3,
 253             'episode_number': 13,
 254         },
 255         'params': {
 256             'skip_download': True,
 257         },
 258     }, {
 259         'url': 'https://tv.nrk.no/serie/nytt-paa-nytt/MUHH46000317/27-01-2017',
 260         'info_dict': {
 261             'id': 'MUHH46000317AA',
 262             'ext': 'mp4',
 263             'title': 'Nytt på Nytt 27.01.2017',
 264             'description': 'md5:5358d6388fba0ea6f0b6d11c48b9eb4b',
 265             'duration': 1796,
 266             'series': 'Nytt på nytt',
 267             'episode': '27.01.2017',
 268         },
 269         'params': {
 270             'skip_download': True,
 271         },
 272     }, {
 273         'url': 'https://radio.nrk.no/serie/dagsnytt/NPUB21019315/12-07-2015#',
 274         'only_matching': True,
 275     }, {
 276         'url': 'https://tv.nrk.no/serie/lindmo/2018/MUHU11006318/avspiller',
 277         'only_matching': True,
 278     }]
 279
 280     _api_host = None
 281
 282     def _extract_from_mediaelement(self, video_id):
 283         api_hosts = (self._api_host, ) if self._api_host else self._API_HOSTS
 284
 285         for api_host in api_hosts:
 286             data = self._download_json(
 287                 'http://%s/mediaelement/%s' % (api_host, video_id),
 288                 video_id, 'Downloading mediaelement JSON',
 289                 fatal=api_host == api_hosts[-1])
 290             if not data:
 291                 continue
 292             self._api_host = api_host
 293             break
 294
 295         title = data.get('fullTitle') or data.get('mainTitle') or data['title']
 296         video_id = data.get('id') or video_id
 297
 298         entries = []
 299
 300         conviva = data.get('convivaStatistics') or {}
 301         live = (data.get('mediaElementType') == 'Live'
 302                 or data.get('isLive') is True or conviva.get('isLive'))
 303
 304         def make_title(t):
 305             return self._live_title(t) if live else t
 306
 307         media_assets = data.get('mediaAssets')
 308         if media_assets and isinstance(media_assets, list):
 309             def video_id_and_title(idx):
 310                 return ((video_id, title) if len(media_assets) == 1
 311                         else ('%s-%d' % (video_id, idx), '%s (Part %d)' % (title, idx)))
 312             for num, asset in enumerate(media_assets, 1):
 313                 asset_url = asset.get('url')
 314                 if not asset_url:
 315                     continue
 316                 formats = self._extract_akamai_formats(asset_url, video_id)
 317                 if not formats:
 318                     continue
 319                 self._sort_formats(formats)
 320
 321                 # Some f4m streams may not work with hdcore in fragments' URLs
 322                 for f in formats:
 323                     extra_param = f.get('extra_param_to_segment_url')
 324                     if extra_param and 'hdcore' in extra_param:
 325                         del f['extra_param_to_segment_url']
 326
 327                 entry_id, entry_title = video_id_and_title(num)
 328                 duration = parse_duration(asset.get('duration'))
 329                 subtitles = {}
 330                 for subtitle in ('webVtt', 'timedText'):
 331                     subtitle_url = asset.get('%sSubtitlesUrl' % subtitle)
 332                     if subtitle_url:
 333                         subtitles.setdefault('no', []).append({
 334                             'url': compat_urllib_parse_unquote(subtitle_url)
 335                         })
 336                 entries.append({
 337                     'id': asset.get('carrierId') or entry_id,
 338                     'title': make_title(entry_title),
 339                     'duration': duration,
 340                     'subtitles': subtitles,
 341                     'formats': formats,
 342                 })
 343
 344         if not entries:
 345             media_url = data.get('mediaUrl')
 346             if media_url:
 347                 formats = self._extract_akamai_formats(media_url, video_id)
 348                 self._sort_formats(formats)
 349                 duration = parse_duration(data.get('duration'))
 350                 entries = [{
 351                     'id': video_id,
 352                     'title': make_title(title),
 353                     'duration': duration,
 354                     'formats': formats,
 355                 }]
 356
 357         if not entries:
 358             MESSAGES = {
 359                 'ProgramRightsAreNotReady': 'Du kan dessverre ikke se eller høre programmet',
 360                 'ProgramRightsHasExpired': 'Programmet har gått ut',
 361                 'NoProgramRights': 'Ikke tilgjengelig',
 362                 'ProgramIsGeoBlocked': 'NRK har ikke rettigheter til å vise dette programmet utenfor Norge',
 363             }
 364             message_type = data.get('messageType', '')
 365             # Can be ProgramIsGeoBlocked or ChannelIsGeoBlocked*
 366             if 'IsGeoBlocked' in message_type:
 367                 self.raise_geo_restricted(
 368                     msg=MESSAGES.get('ProgramIsGeoBlocked'),
 369                     countries=self._GEO_COUNTRIES)
 370             raise ExtractorError(
 371                 '%s said: %s' % (self.IE_NAME, MESSAGES.get(
 372                     message_type, message_type)),
 373                 expected=True)
 374
 375         series = conviva.get('seriesName') or data.get('seriesTitle')
 376         episode = conviva.get('episodeName') or data.get('episodeNumberOrDate')
 377
 378         season_number = None
 379         episode_number = None
 380         if data.get('mediaElementType') == 'Episode':
 381             _season_episode = data.get('scoresStatistics', {}).get('springStreamStream') or \
 382                 data.get('relativeOriginUrl', '')
 383             EPISODENUM_RE = [
 384                 r'/s(?P<season>\d{,2})e(?P<episode>\d{,2})\.',
 385                 r'/sesong-(?P<season>\d{,2})/episode-(?P<episode>\d{,2})',
 386             ]
 387             season_number = int_or_none(self._search_regex(
 388                 EPISODENUM_RE, _season_episode, 'season number',
 389                 default=None, group='season'))
 390             episode_number = int_or_none(self._search_regex(
 391                 EPISODENUM_RE, _season_episode, 'episode number',
 392                 default=None, group='episode'))
 393
 394         thumbnails = None
 395         images = data.get('images')
 396         if images and isinstance(images, dict):
 397             web_images = images.get('webImages')
 398             if isinstance(web_images, list):
 399                 thumbnails = [{
 400                     'url': image['imageUrl'],
 401                     'width': int_or_none(image.get('width')),
 402                     'height': int_or_none(image.get('height')),
 403                 } for image in web_images if image.get('imageUrl')]
 404
 405         description = data.get('description')
 406         category = data.get('mediaAnalytics', {}).get('category')
 407
 408         common_info = {
 409             'description': description,
 410             'series': series,
 411             'episode': episode,
 412             'season_number': season_number,
 413             'episode_number': episode_number,
 414             'categories': [category] if category else None,
 415             'age_limit': parse_age_limit(data.get('legalAge')),
 416             'thumbnails': thumbnails,
 417         }
 418
 419         vcodec = 'none' if data.get('mediaType') == 'Audio' else None
 420
 421         for entry in entries:
 422             entry.update(common_info)
 423             for f in entry['formats']:
 424                 f['vcodec'] = vcodec
 425
 426         points = data.get('shortIndexPoints')
 427         if isinstance(points, list):
 428             chapters = []
 429             for next_num, point in enumerate(points, start=1):
 430                 if not isinstance(point, dict):
 431                     continue
 432                 start_time = parse_duration(point.get('startPoint'))
 433                 if start_time is None:
 434                     continue
 435                 end_time = parse_duration(
 436                     data.get('duration')
 437                     if next_num == len(points)
 438                     else points[next_num].get('startPoint'))
 439                 if end_time is None:
 440                     continue
 441                 chapters.append({
 442                     'start_time': start_time,
 443                     'end_time': end_time,
 444                     'title': point.get('title'),
 445                 })
 446             if chapters and len(entries) == 1:
 447                 entries[0]['chapters'] = chapters
 448
 449         return self.playlist_result(entries, video_id, title, description)
 450
 451     def _real_extract(self, url):
 452         video_id = self._match_id(url)
 453         return self._extract_from_mediaelement(video_id)
 454
 455
 456 class NRKTVEpisodeIE(InfoExtractor):
 457     _VALID_URL = r'https?://tv\.nrk\.no/serie/(?P<id>[^/]+/sesong/\d+/episode/\d+)'
 458     _TESTS = [{
 459         'url': 'https://tv.nrk.no/serie/hellums-kro/sesong/1/episode/2',
 460         'info_dict': {
 461             'id': 'MUHH36005220BA',
 462             'ext': 'mp4',
 463             'title': 'Kro, krig og kjærlighet 2:6',
 464             'description': 'md5:b32a7dc0b1ed27c8064f58b97bda4350',
 465             'duration': 1563,
 466             'series': 'Hellums kro',
 467             'season_number': 1,
 468             'episode_number': 2,
 469             'episode': '2:6',
 470             'age_limit': 6,
 471         },
 472         'params': {
 473             'skip_download': True,
 474         },
 475     }, {
 476         'url': 'https://tv.nrk.no/serie/backstage/sesong/1/episode/8',
 477         'info_dict': {
 478             'id': 'MSUI14000816AA',
 479             'ext': 'mp4',
 480             'title': 'Backstage 8:30',
 481             'description': 'md5:de6ca5d5a2d56849e4021f2bf2850df4',
 482             'duration': 1320,
 483             'series': 'Backstage',
 484             'season_number': 1,
 485             'episode_number': 8,
 486             'episode': '8:30',
 487         },
 488         'params': {
 489             'skip_download': True,
 490         },
 491         'skip': 'ProgramRightsHasExpired',
 492     }]
 493
 494     def _real_extract(self, url):
 495         display_id = self._match_id(url)
 496
 497         webpage = self._download_webpage(url, display_id)
 498
 499         info = self._search_json_ld(webpage, display_id, default={})
 500         nrk_id = info.get('@id') or self._html_search_meta(
 501             'nrk:program-id', webpage, default=None) or self._search_regex(
 502             r'data-program-id=["\'](%s)' % NRKTVIE._EPISODE_RE, webpage,
 503             'nrk id')
 504         assert re.match(NRKTVIE._EPISODE_RE, nrk_id)
 505
 506         info.update({
 507             '_type': 'url_transparent',
 508             'id': nrk_id,
 509             'url': 'nrk:%s' % nrk_id,
 510             'ie_key': NRKIE.ie_key(),
 511         })
 512         return info
 513
 514
 515 class NRKTVSerieBaseIE(InfoExtractor):
 516     def _extract_series(self, webpage, display_id, fatal=True):
 517         config = self._parse_json(
 518             self._search_regex(
 519                 (r'INITIAL_DATA(?:_V\d)?_*\s*=\s*({.+?})\s*;',
 520                  r'({.+?})\s*,\s*"[^"]+"\s*\)\s*</script>'),
 521                 webpage, 'config', default='{}' if not fatal else NO_DEFAULT),
 522             display_id, fatal=False, transform_source=js_to_json)
 523         if not config:
 524             return
 525         return try_get(
 526             config,
 527             (lambda x: x['initialState']['series'], lambda x: x['series']),
 528             dict)
 529
 530     def _extract_seasons(self, seasons):
 531         if not isinstance(seasons, list):
 532             return []
 533         entries = []
 534         for season in seasons:
 535             entries.extend(self._extract_episodes(season))
 536         return entries
 537
 538     def _extract_episodes(self, season):
 539         if not isinstance(season, dict):
 540             return []
 541         return self._extract_entries(season.get('episodes'))
 542
 543     def _extract_entries(self, entry_list):
 544         if not isinstance(entry_list, list):
 545             return []
 546         entries = []
 547         for episode in entry_list:
 548             nrk_id = episode.get('prfId')
 549             if not nrk_id or not isinstance(nrk_id, compat_str):
 550                 continue
 551             entries.append(self.url_result(
 552                 'nrk:%s' % nrk_id, ie=NRKIE.ie_key(), video_id=nrk_id))
 553         return entries
 554
 555
 556 class NRKTVSeasonIE(NRKTVSerieBaseIE):
 557     _VALID_URL = r'https?://tv\.nrk\.no/serie/[^/]+/sesong/(?P<id>\d+)'
 558     _TEST = {
 559         'url': 'https://tv.nrk.no/serie/backstage/sesong/1',
 560         'info_dict': {
 561             'id': '1',
 562             'title': 'Sesong 1',
 563         },
 564         'playlist_mincount': 30,
 565     }
 566
 567     @classmethod
 568     def suitable(cls, url):
 569         return (False if NRKTVIE.suitable(url) or NRKTVEpisodeIE.suitable(url)
 570                 else super(NRKTVSeasonIE, cls).suitable(url))
 571
 572     def _real_extract(self, url):
 573         display_id = self._match_id(url)
 574
 575         webpage = self._download_webpage(url, display_id)
 576
 577         series = self._extract_series(webpage, display_id)
 578
 579         season = next(
 580             s for s in series['seasons']
 581             if int(display_id) == s.get('seasonNumber'))
 582
 583         title = try_get(season, lambda x: x['titles']['title'], compat_str)
 584         return self.playlist_result(
 585             self._extract_episodes(season), display_id, title)
 586
 587
 588 class NRKTVSeriesIE(NRKTVSerieBaseIE):
 589     _VALID_URL = r'https?://(?:tv|radio)\.nrk(?:super)?\.no/serie/(?P<id>[^/]+)'
 590     _ITEM_RE = r'(?:data-season=["\']|id=["\']season-)(?P<id>\d+)'
 591     _TESTS = [{
 592         'url': 'https://tv.nrk.no/serie/blank',
 593         'info_dict': {
 594             'id': 'blank',
 595             'title': 'Blank',
 596             'description': 'md5:7664b4e7e77dc6810cd3bca367c25b6e',
 597         },
 598         'playlist_mincount': 30,
 599     }, {
 600         # new layout, seasons
 601         'url': 'https://tv.nrk.no/serie/backstage',
 602         'info_dict': {
 603             'id': 'backstage',
 604             'title': 'Backstage',
 605             'description': 'md5:c3ec3a35736fca0f9e1207b5511143d3',
 606         },
 607         'playlist_mincount': 60,
 608     }, {
 609         # new layout, instalments
 610         'url': 'https://tv.nrk.no/serie/groenn-glede',
 611         'info_dict': {
 612             'id': 'groenn-glede',
 613             'title': 'Grønn glede',
 614             'description': 'md5:7576e92ae7f65da6993cf90ee29e4608',
 615         },
 616         'playlist_mincount': 10,
 617     }, {
 618         # old layout
 619         'url': 'https://tv.nrksuper.no/serie/labyrint',
 620         'info_dict': {
 621             'id': 'labyrint',
 622             'title': 'Labyrint',
 623             'description': 'md5:318b597330fdac5959247c9b69fdb1ec',
 624         },
 625         'playlist_mincount': 3,
 626     }, {
 627         'url': 'https://tv.nrk.no/serie/broedrene-dal-og-spektralsteinene',
 628         'only_matching': True,
 629     }, {
 630         'url': 'https://tv.nrk.no/serie/saving-the-human-race',
 631         'only_matching': True,
 632     }, {
 633         'url': 'https://tv.nrk.no/serie/postmann-pat',
 634         'only_matching': True,
 635     }]
 636
 637     @classmethod
 638     def suitable(cls, url):
 639         return (
 640             False if any(ie.suitable(url)
 641                          for ie in (NRKTVIE, NRKTVEpisodeIE, NRKTVSeasonIE))
 642             else super(NRKTVSeriesIE, cls).suitable(url))
 643
 644     def _real_extract(self, url):
 645         series_id = self._match_id(url)
 646
 647         webpage = self._download_webpage(url, series_id)
 648
 649         # New layout (e.g. https://tv.nrk.no/serie/backstage)
 650         series = self._extract_series(webpage, series_id, fatal=False)
 651         if series:
 652             title = try_get(series, lambda x: x['titles']['title'], compat_str)
 653             description = try_get(
 654                 series, lambda x: x['titles']['subtitle'], compat_str)
 655             entries = []
 656             entries.extend(self._extract_seasons(series.get('seasons')))
 657             entries.extend(self._extract_entries(series.get('instalments')))
 658             entries.extend(self._extract_episodes(series.get('extraMaterial')))
 659             return self.playlist_result(entries, series_id, title, description)
 660
 661         # Old layout (e.g. https://tv.nrksuper.no/serie/labyrint)
 662         entries = [
 663             self.url_result(
 664                 'https://tv.nrk.no/program/Episodes/{series}/{season}'.format(
 665                     series=series_id, season=season_id))
 666             for season_id in re.findall(self._ITEM_RE, webpage)
 667         ]
 668
 669         title = self._html_search_meta(
 670             'seriestitle', webpage,
 671             'title', default=None) or self._og_search_title(
 672             webpage, fatal=False)
 673         if title:
 674             title = self._search_regex(
 675                 r'NRK (?:Super )?TV\s*[-–]\s*(.+)', title, 'title', default=title)
 676
 677         description = self._html_search_meta(
 678             'series_description', webpage,
 679             'description', default=None) or self._og_search_description(webpage)
 680
 681         return self.playlist_result(entries, series_id, title, description)
 682
 683
 684 class NRKTVDirekteIE(NRKTVIE):
 685     IE_DESC = 'NRK TV Direkte and NRK Radio Direkte'
 686     _VALID_URL = r'https?://(?:tv|radio)\.nrk\.no/direkte/(?P<id>[^/?#&]+)'
 687
 688     _TESTS = [{
 689         'url': 'https://tv.nrk.no/direkte/nrk1',
 690         'only_matching': True,
 691     }, {
 692         'url': 'https://radio.nrk.no/direkte/p1_oslo_akershus',
 693         'only_matching': True,
 694     }]
 695
 696
 697 class NRKPlaylistBaseIE(InfoExtractor):
 698     def _extract_description(self, webpage):
 699         pass
 700
 701     def _real_extract(self, url):
 702         playlist_id = self._match_id(url)
 703
 704         webpage = self._download_webpage(url, playlist_id)
 705
 706         entries = [
 707             self.url_result('nrk:%s' % video_id, NRKIE.ie_key())
 708             for video_id in re.findall(self._ITEM_RE, webpage)
 709         ]
 710
 711         playlist_title = self. _extract_title(webpage)
 712         playlist_description = self._extract_description(webpage)
 713
 714         return self.playlist_result(
 715             entries, playlist_id, playlist_title, playlist_description)
 716
 717
 718 class NRKPlaylistIE(NRKPlaylistBaseIE):
 719     _VALID_URL = r'https?://(?:www\.)?nrk\.no/(?!video|skole)(?:[^/]+/)+(?P<id>[^/]+)'
 720     _ITEM_RE = r'class="[^"]*\brich\b[^"]*"[^>]+data-video-id="([^"]+)"'
 721     _TESTS = [{
 722         'url': 'http://www.nrk.no/troms/gjenopplev-den-historiske-solformorkelsen-1.12270763',
 723         'info_dict': {
 724             'id': 'gjenopplev-den-historiske-solformorkelsen-1.12270763',
 725             'title': 'Gjenopplev den historiske solformørkelsen',
 726             'description': 'md5:c2df8ea3bac5654a26fc2834a542feed',
 727         },
 728         'playlist_count': 2,
 729     }, {
 730         'url': 'http://www.nrk.no/kultur/bok/rivertonprisen-til-karin-fossum-1.12266449',
 731         'info_dict': {
 732             'id': 'rivertonprisen-til-karin-fossum-1.12266449',
 733             'title': 'Rivertonprisen til Karin Fossum',
 734             'description': 'Første kvinne på 15 år til å vinne krimlitteraturprisen.',
 735         },
 736         'playlist_count': 2,
 737     }]
 738
 739     def _extract_title(self, webpage):
 740         return self._og_search_title(webpage, fatal=False)
 741
 742     def _extract_description(self, webpage):
 743         return self._og_search_description(webpage)
 744
 745
 746 class NRKTVEpisodesIE(NRKPlaylistBaseIE):
 747     _VALID_URL = r'https?://tv\.nrk\.no/program/[Ee]pisodes/[^/]+/(?P<id>\d+)'
 748     _ITEM_RE = r'data-episode=["\']%s' % NRKTVIE._EPISODE_RE
 749     _TESTS = [{
 750         'url': 'https://tv.nrk.no/program/episodes/nytt-paa-nytt/69031',
 751         'info_dict': {
 752             'id': '69031',
 753             'title': 'Nytt på nytt, sesong: 201210',
 754         },
 755         'playlist_count': 4,
 756     }]
 757
 758     def _extract_title(self, webpage):
 759         return self._html_search_regex(
 760             r'<h1>([^<]+)</h1>', webpage, 'title', fatal=False)
 761
 762
 763 class NRKSkoleIE(InfoExtractor):
 764     IE_DESC = 'NRK Skole'
 765     _VALID_URL = r'https?://(?:www\.)?nrk\.no/skole/?\?.*\bmediaId=(?P<id>\d+)'
 766
 767     _TESTS = [{
 768         'url': 'https://www.nrk.no/skole/?page=search&q=&mediaId=14099',
 769         'md5': '18c12c3d071953c3bf8d54ef6b2587b7',
 770         'info_dict': {
 771             'id': '6021',
 772             'ext': 'mp4',
 773             'title': 'Genetikk og eneggede tvillinger',
 774             'description': 'md5:3aca25dcf38ec30f0363428d2b265f8d',
 775             'duration': 399,
 776         },
 777     }, {
 778         'url': 'https://www.nrk.no/skole/?page=objectives&subject=naturfag&objective=K15114&mediaId=19355',
 779         'only_matching': True,
 780     }]
 781
 782     def _real_extract(self, url):
 783         video_id = self._match_id(url)
 784
 785         webpage = self._download_webpage(
 786             'https://mimir.nrk.no/plugin/1.0/static?mediaId=%s' % video_id,
 787             video_id)
 788
 789         nrk_id = self._parse_json(
 790             self._search_regex(
 791                 r'<script[^>]+type=["\']application/json["\'][^>]*>({.+?})</script>',
 792                 webpage, 'application json'),
 793             video_id)['activeMedia']['psId']
 794
 795         return self.url_result('nrk:%s' % nrk_id)