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 compat_urllib_parse_unquote
   8 from ..utils import (
   9     ExtractorError,
  10     int_or_none,
  11     parse_age_limit,
  12     parse_duration,
  13 )
  14
  15
  16 class NRKBaseIE(InfoExtractor):
  17     _GEO_COUNTRIES = ['NO']
  18
  19     _api_host = None
  20
  21     def _real_extract(self, url):
  22         video_id = self._match_id(url)
  23
  24         api_hosts = (self._api_host, ) if self._api_host else self._API_HOSTS
  25
  26         for api_host in api_hosts:
  27             data = self._download_json(
  28                 'http://%s/mediaelement/%s' % (api_host, video_id),
  29                 video_id, 'Downloading mediaelement JSON',
  30                 fatal=api_host == api_hosts[-1])
  31             if not data:
  32                 continue
  33             self._api_host = api_host
  34             break
  35
  36         title = data.get('fullTitle') or data.get('mainTitle') or data['title']
  37         video_id = data.get('id') or video_id
  38
  39         entries = []
  40
  41         conviva = data.get('convivaStatistics') or {}
  42         live = (data.get('mediaElementType') == 'Live' or
  43                 data.get('isLive') is True or conviva.get('isLive'))
  44
  45         def make_title(t):
  46             return self._live_title(t) if live else t
  47
  48         media_assets = data.get('mediaAssets')
  49         if media_assets and isinstance(media_assets, list):
  50             def video_id_and_title(idx):
  51                 return ((video_id, title) if len(media_assets) == 1
  52                         else ('%s-%d' % (video_id, idx), '%s (Part %d)' % (title, idx)))
  53             for num, asset in enumerate(media_assets, 1):
  54                 asset_url = asset.get('url')
  55                 if not asset_url:
  56                     continue
  57                 formats = self._extract_akamai_formats(asset_url, video_id)
  58                 if not formats:
  59                     continue
  60                 self._sort_formats(formats)
  61
  62                 # Some f4m streams may not work with hdcore in fragments' URLs
  63                 for f in formats:
  64                     extra_param = f.get('extra_param_to_segment_url')
  65                     if extra_param and 'hdcore' in extra_param:
  66                         del f['extra_param_to_segment_url']
  67
  68                 entry_id, entry_title = video_id_and_title(num)
  69                 duration = parse_duration(asset.get('duration'))
  70                 subtitles = {}
  71                 for subtitle in ('webVtt', 'timedText'):
  72                     subtitle_url = asset.get('%sSubtitlesUrl' % subtitle)
  73                     if subtitle_url:
  74                         subtitles.setdefault('no', []).append({
  75                             'url': compat_urllib_parse_unquote(subtitle_url)
  76                         })
  77                 entries.append({
  78                     'id': asset.get('carrierId') or entry_id,
  79                     'title': make_title(entry_title),
  80                     'duration': duration,
  81                     'subtitles': subtitles,
  82                     'formats': formats,
  83                 })
  84
  85         if not entries:
  86             media_url = data.get('mediaUrl')
  87             if media_url:
  88                 formats = self._extract_akamai_formats(media_url, video_id)
  89                 self._sort_formats(formats)
  90                 duration = parse_duration(data.get('duration'))
  91                 entries = [{
  92                     'id': video_id,
  93                     'title': make_title(title),
  94                     'duration': duration,
  95                     'formats': formats,
  96                 }]
  97
  98         if not entries:
  99             MESSAGES = {
 100                 'ProgramRightsAreNotReady': 'Du kan dessverre ikke se eller høre programmet',
 101                 'ProgramRightsHasExpired': 'Programmet har gått ut',
 102                 'ProgramIsGeoBlocked': 'NRK har ikke rettigheter til å vise dette programmet utenfor Norge',
 103             }
 104             message_type = data.get('messageType', '')
 105             # Can be ProgramIsGeoBlocked or ChannelIsGeoBlocked*
 106             if 'IsGeoBlocked' in message_type:
 107                 self.raise_geo_restricted(
 108                     msg=MESSAGES.get('ProgramIsGeoBlocked'),
 109                     countries=self._GEO_COUNTRIES)
 110             raise ExtractorError(
 111                 '%s said: %s' % (self.IE_NAME, MESSAGES.get(
 112                     message_type, message_type)),
 113                 expected=True)
 114
 115         series = conviva.get('seriesName') or data.get('seriesTitle')
 116         episode = conviva.get('episodeName') or data.get('episodeNumberOrDate')
 117
 118         season_number = None
 119         episode_number = None
 120         if data.get('mediaElementType') == 'Episode':
 121             _season_episode = data.get('scoresStatistics', {}).get('springStreamStream') or \
 122                 data.get('relativeOriginUrl', '')
 123             EPISODENUM_RE = [
 124                 r'/s(?P<season>\d{,2})e(?P<episode>\d{,2})\.',
 125                 r'/sesong-(?P<season>\d{,2})/episode-(?P<episode>\d{,2})',
 126             ]
 127             season_number = int_or_none(self._search_regex(
 128                 EPISODENUM_RE, _season_episode, 'season number',
 129                 default=None, group='season'))
 130             episode_number = int_or_none(self._search_regex(
 131                 EPISODENUM_RE, _season_episode, 'episode number',
 132                 default=None, group='episode'))
 133
 134         thumbnails = None
 135         images = data.get('images')
 136         if images and isinstance(images, dict):
 137             web_images = images.get('webImages')
 138             if isinstance(web_images, list):
 139                 thumbnails = [{
 140                     'url': image['imageUrl'],
 141                     'width': int_or_none(image.get('width')),
 142                     'height': int_or_none(image.get('height')),
 143                 } for image in web_images if image.get('imageUrl')]
 144
 145         description = data.get('description')
 146         category = data.get('mediaAnalytics', {}).get('category')
 147
 148         common_info = {
 149             'description': description,
 150             'series': series,
 151             'episode': episode,
 152             'season_number': season_number,
 153             'episode_number': episode_number,
 154             'categories': [category] if category else None,
 155             'age_limit': parse_age_limit(data.get('legalAge')),
 156             'thumbnails': thumbnails,
 157         }
 158
 159         vcodec = 'none' if data.get('mediaType') == 'Audio' else None
 160
 161         for entry in entries:
 162             entry.update(common_info)
 163             for f in entry['formats']:
 164                 f['vcodec'] = vcodec
 165
 166         points = data.get('shortIndexPoints')
 167         if isinstance(points, list):
 168             chapters = []
 169             for next_num, point in enumerate(points, start=1):
 170                 if not isinstance(point, dict):
 171                     continue
 172                 start_time = parse_duration(point.get('startPoint'))
 173                 if start_time is None:
 174                     continue
 175                 end_time = parse_duration(
 176                     data.get('duration')
 177                     if next_num == len(points)
 178                     else points[next_num].get('startPoint'))
 179                 if end_time is None:
 180                     continue
 181                 chapters.append({
 182                     'start_time': start_time,
 183                     'end_time': end_time,
 184                     'title': point.get('title'),
 185                 })
 186             if chapters and len(entries) == 1:
 187                 entries[0]['chapters'] = chapters
 188
 189         return self.playlist_result(entries, video_id, title, description)
 190
 191
 192 class NRKIE(NRKBaseIE):
 193     _VALID_URL = r'''(?x)
 194                         (?:
 195                             nrk:|
 196                             https?://
 197                                 (?:
 198                                     (?:www\.)?nrk\.no/video/PS\*|
 199                                     v8[-.]psapi\.nrk\.no/mediaelement/
 200                                 )
 201                             )
 202                             (?P<id>[^?#&]+)
 203                         '''
 204     _API_HOSTS = ('psapi.nrk.no', 'v8-psapi.nrk.no')
 205     _TESTS = [{
 206         # video
 207         'url': 'http://www.nrk.no/video/PS*150533',
 208         'md5': '2f7f6eeb2aacdd99885f355428715cfa',
 209         'info_dict': {
 210             'id': '150533',
 211             'ext': 'mp4',
 212             'title': 'Dompap og andre fugler i Piip-Show',
 213             'description': 'md5:d9261ba34c43b61c812cb6b0269a5c8f',
 214             'duration': 263,
 215         }
 216     }, {
 217         # audio
 218         'url': 'http://www.nrk.no/video/PS*154915',
 219         # MD5 is unstable
 220         'info_dict': {
 221             'id': '154915',
 222             'ext': 'flv',
 223             'title': 'Slik høres internett ut når du er blind',
 224             'description': 'md5:a621f5cc1bd75c8d5104cb048c6b8568',
 225             'duration': 20,
 226         }
 227     }, {
 228         'url': 'nrk:ecc1b952-96dc-4a98-81b9-5296dc7a98d9',
 229         'only_matching': True,
 230     }, {
 231         'url': 'nrk:clip/7707d5a3-ebe7-434a-87d5-a3ebe7a34a70',
 232         'only_matching': True,
 233     }, {
 234         'url': 'https://v8-psapi.nrk.no/mediaelement/ecc1b952-96dc-4a98-81b9-5296dc7a98d9',
 235         'only_matching': True,
 236     }]
 237
 238
 239 class NRKTVIE(NRKBaseIE):
 240     IE_DESC = 'NRK TV and NRK Radio'
 241     _EPISODE_RE = r'(?P<id>[a-zA-Z]{4}\d{8})'
 242     _VALID_URL = r'''(?x)
 243                         https?://
 244                             (?:tv|radio)\.nrk(?:super)?\.no/
 245                             (?:serie/[^/]+|program)/
 246                             (?![Ee]pisodes)%s
 247                             (?:/\d{2}-\d{2}-\d{4})?
 248                             (?:\#del=(?P<part_id>\d+))?
 249                     ''' % _EPISODE_RE
 250     _API_HOSTS = ('psapi-ne.nrk.no', 'psapi-we.nrk.no')
 251     _TESTS = [{
 252         'url': 'https://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
 253         'md5': '4e9ca6629f09e588ed240fb11619922a',
 254         'info_dict': {
 255             'id': 'MUHH48000314AA',
 256             'ext': 'mp4',
 257             'title': '20 spørsmål 23.05.2014',
 258             'description': 'md5:bdea103bc35494c143c6a9acdd84887a',
 259             'duration': 1741,
 260             'series': '20 spørsmål - TV',
 261             'episode': '23.05.2014',
 262         },
 263     }, {
 264         'url': 'https://tv.nrk.no/program/mdfp15000514',
 265         'info_dict': {
 266             'id': 'MDFP15000514CA',
 267             'ext': 'mp4',
 268             'title': 'Grunnlovsjubiléet - Stor ståhei for ingenting 24.05.2014',
 269             'description': 'md5:89290c5ccde1b3a24bb8050ab67fe1db',
 270             'duration': 4605,
 271             'series': 'Kunnskapskanalen',
 272             'episode': '24.05.2014',
 273         },
 274         'params': {
 275             'skip_download': True,
 276         },
 277     }, {
 278         # single playlist video
 279         'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2',
 280         'info_dict': {
 281             'id': 'MSPO40010515-part2',
 282             'ext': 'flv',
 283             'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
 284             'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 285         },
 286         'params': {
 287             'skip_download': True,
 288         },
 289         'expected_warnings': ['Video is geo restricted'],
 290         'skip': 'particular part is not supported currently',
 291     }, {
 292         'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015',
 293         'playlist': [{
 294             'info_dict': {
 295                 'id': 'MSPO40010515AH',
 296                 'ext': 'mp4',
 297                 'title': 'Sprint fri teknikk, kvinner og menn 06.01.2015 (Part 1)',
 298                 'description': 'md5:c03aba1e917561eface5214020551b7a',
 299                 'duration': 772,
 300                 'series': 'Tour de Ski',
 301                 'episode': '06.01.2015',
 302             },
 303             'params': {
 304                 'skip_download': True,
 305             },
 306         }, {
 307             'info_dict': {
 308                 'id': 'MSPO40010515BH',
 309                 'ext': 'mp4',
 310                 'title': 'Sprint fri teknikk, kvinner og menn 06.01.2015 (Part 2)',
 311                 'description': 'md5:c03aba1e917561eface5214020551b7a',
 312                 'duration': 6175,
 313                 'series': 'Tour de Ski',
 314                 'episode': '06.01.2015',
 315             },
 316             'params': {
 317                 'skip_download': True,
 318             },
 319         }],
 320         'info_dict': {
 321             'id': 'MSPO40010515',
 322             'title': 'Sprint fri teknikk, kvinner og menn 06.01.2015',
 323             'description': 'md5:c03aba1e917561eface5214020551b7a',
 324         },
 325         'expected_warnings': ['Video is geo restricted'],
 326     }, {
 327         'url': 'https://tv.nrk.no/serie/anno/KMTE50001317/sesong-3/episode-13',
 328         'info_dict': {
 329             'id': 'KMTE50001317AA',
 330             'ext': 'mp4',
 331             'title': 'Anno 13:30',
 332             'description': 'md5:11d9613661a8dbe6f9bef54e3a4cbbfa',
 333             'duration': 2340,
 334             'series': 'Anno',
 335             'episode': '13:30',
 336             'season_number': 3,
 337             'episode_number': 13,
 338         },
 339         'params': {
 340             'skip_download': True,
 341         },
 342     }, {
 343         'url': 'https://tv.nrk.no/serie/nytt-paa-nytt/MUHH46000317/27-01-2017',
 344         'info_dict': {
 345             'id': 'MUHH46000317AA',
 346             'ext': 'mp4',
 347             'title': 'Nytt på Nytt 27.01.2017',
 348             'description': 'md5:5358d6388fba0ea6f0b6d11c48b9eb4b',
 349             'duration': 1796,
 350             'series': 'Nytt på nytt',
 351             'episode': '27.01.2017',
 352         },
 353         'params': {
 354             'skip_download': True,
 355         },
 356     }, {
 357         'url': 'https://radio.nrk.no/serie/dagsnytt/NPUB21019315/12-07-2015#',
 358         'only_matching': True,
 359     }]
 360
 361
 362 class NRKTVDirekteIE(NRKTVIE):
 363     IE_DESC = 'NRK TV Direkte and NRK Radio Direkte'
 364     _VALID_URL = r'https?://(?:tv|radio)\.nrk\.no/direkte/(?P<id>[^/?#&]+)'
 365
 366     _TESTS = [{
 367         'url': 'https://tv.nrk.no/direkte/nrk1',
 368         'only_matching': True,
 369     }, {
 370         'url': 'https://radio.nrk.no/direkte/p1_oslo_akershus',
 371         'only_matching': True,
 372     }]
 373
 374
 375 class NRKPlaylistBaseIE(InfoExtractor):
 376     def _extract_description(self, webpage):
 377         pass
 378
 379     def _real_extract(self, url):
 380         playlist_id = self._match_id(url)
 381
 382         webpage = self._download_webpage(url, playlist_id)
 383
 384         entries = [
 385             self.url_result('nrk:%s' % video_id, NRKIE.ie_key())
 386             for video_id in re.findall(self._ITEM_RE, webpage)
 387         ]
 388
 389         playlist_title = self. _extract_title(webpage)
 390         playlist_description = self._extract_description(webpage)
 391
 392         return self.playlist_result(
 393             entries, playlist_id, playlist_title, playlist_description)
 394
 395
 396 class NRKPlaylistIE(NRKPlaylistBaseIE):
 397     _VALID_URL = r'https?://(?:www\.)?nrk\.no/(?!video|skole)(?:[^/]+/)+(?P<id>[^/]+)'
 398     _ITEM_RE = r'class="[^"]*\brich\b[^"]*"[^>]+data-video-id="([^"]+)"'
 399     _TESTS = [{
 400         'url': 'http://www.nrk.no/troms/gjenopplev-den-historiske-solformorkelsen-1.12270763',
 401         'info_dict': {
 402             'id': 'gjenopplev-den-historiske-solformorkelsen-1.12270763',
 403             'title': 'Gjenopplev den historiske solformørkelsen',
 404             'description': 'md5:c2df8ea3bac5654a26fc2834a542feed',
 405         },
 406         'playlist_count': 2,
 407     }, {
 408         'url': 'http://www.nrk.no/kultur/bok/rivertonprisen-til-karin-fossum-1.12266449',
 409         'info_dict': {
 410             'id': 'rivertonprisen-til-karin-fossum-1.12266449',
 411             'title': 'Rivertonprisen til Karin Fossum',
 412             'description': 'Første kvinne på 15 år til å vinne krimlitteraturprisen.',
 413         },
 414         'playlist_count': 5,
 415     }]
 416
 417     def _extract_title(self, webpage):
 418         return self._og_search_title(webpage, fatal=False)
 419
 420     def _extract_description(self, webpage):
 421         return self._og_search_description(webpage)
 422
 423
 424 class NRKTVEpisodesIE(NRKPlaylistBaseIE):
 425     _VALID_URL = r'https?://tv\.nrk\.no/program/[Ee]pisodes/[^/]+/(?P<id>\d+)'
 426     _ITEM_RE = r'data-episode=["\']%s' % NRKTVIE._EPISODE_RE
 427     _TESTS = [{
 428         'url': 'https://tv.nrk.no/program/episodes/nytt-paa-nytt/69031',
 429         'info_dict': {
 430             'id': '69031',
 431             'title': 'Nytt på nytt, sesong: 201210',
 432         },
 433         'playlist_count': 4,
 434     }]
 435
 436     def _extract_title(self, webpage):
 437         return self._html_search_regex(
 438             r'<h1>([^<]+)</h1>', webpage, 'title', fatal=False)
 439
 440
 441 class NRKTVSeriesIE(InfoExtractor):
 442     _VALID_URL = r'https?://(?:tv|radio)\.nrk(?:super)?\.no/serie/(?P<id>[^/]+)'
 443     _ITEM_RE = r'(?:data-season=["\']|id=["\']season-)(?P<id>\d+)'
 444     _TESTS = [{
 445         'url': 'https://tv.nrk.no/serie/groenn-glede',
 446         'info_dict': {
 447             'id': 'groenn-glede',
 448             'title': 'Grønn glede',
 449             'description': 'md5:7576e92ae7f65da6993cf90ee29e4608',
 450         },
 451         'playlist_mincount': 9,
 452     }, {
 453         'url': 'http://tv.nrksuper.no/serie/labyrint',
 454         'info_dict': {
 455             'id': 'labyrint',
 456             'title': 'Labyrint',
 457             'description': 'md5:58afd450974c89e27d5a19212eee7115',
 458         },
 459         'playlist_mincount': 3,
 460     }, {
 461         'url': 'https://tv.nrk.no/serie/broedrene-dal-og-spektralsteinene',
 462         'only_matching': True,
 463     }, {
 464         'url': 'https://tv.nrk.no/serie/saving-the-human-race',
 465         'only_matching': True,
 466     }, {
 467         'url': 'https://tv.nrk.no/serie/postmann-pat',
 468         'only_matching': True,
 469     }]
 470
 471     @classmethod
 472     def suitable(cls, url):
 473         return False if NRKTVIE.suitable(url) else super(NRKTVSeriesIE, cls).suitable(url)
 474
 475     def _real_extract(self, url):
 476         series_id = self._match_id(url)
 477
 478         webpage = self._download_webpage(url, series_id)
 479
 480         entries = [
 481             self.url_result(
 482                 'https://tv.nrk.no/program/Episodes/{series}/{season}'.format(
 483                     series=series_id, season=season_id))
 484             for season_id in re.findall(self._ITEM_RE, webpage)
 485         ]
 486
 487         title = self._html_search_meta(
 488             'seriestitle', webpage,
 489             'title', default=None) or self._og_search_title(
 490             webpage, fatal=False)
 491
 492         description = self._html_search_meta(
 493             'series_description', webpage,
 494             'description', default=None) or self._og_search_description(webpage)
 495
 496         return self.playlist_result(entries, series_id, title, description)
 497
 498
 499 class NRKSkoleIE(InfoExtractor):
 500     IE_DESC = 'NRK Skole'
 501     _VALID_URL = r'https?://(?:www\.)?nrk\.no/skole/?\?.*\bmediaId=(?P<id>\d+)'
 502
 503     _TESTS = [{
 504         'url': 'https://www.nrk.no/skole/?page=search&q=&mediaId=14099',
 505         'md5': '6bc936b01f9dd8ed45bc58b252b2d9b6',
 506         'info_dict': {
 507             'id': '6021',
 508             'ext': 'mp4',
 509             'title': 'Genetikk og eneggede tvillinger',
 510             'description': 'md5:3aca25dcf38ec30f0363428d2b265f8d',
 511             'duration': 399,
 512         },
 513     }, {
 514         'url': 'https://www.nrk.no/skole/?page=objectives&subject=naturfag&objective=K15114&mediaId=19355',
 515         'only_matching': True,
 516     }]
 517
 518     def _real_extract(self, url):
 519         video_id = self._match_id(url)
 520
 521         webpage = self._download_webpage(
 522             'https://mimir.nrk.no/plugin/1.0/static?mediaId=%s' % video_id,
 523             video_id)
 524
 525         nrk_id = self._parse_json(
 526             self._search_regex(
 527                 r'<script[^>]+type=["\']application/json["\'][^>]*>({.+?})</script>',
 528                 webpage, 'application json'),
 529             video_id)['activeMedia']['psId']
 530
 531         return self.url_result('nrk:%s' % nrk_id)