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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import random
   5 import re
   6
   7 from .common import InfoExtractor
   8 from ..compat import compat_urllib_parse_unquote
   9 from ..utils import (
  10     ExtractorError,
  11     int_or_none,
  12     parse_age_limit,
  13     parse_duration,
  14 )
  15
  16
  17 class NRKBaseIE(InfoExtractor):
  18     _faked_ip = None
  19
  20     def _download_webpage_handle(self, *args, **kwargs):
  21         # NRK checks X-Forwarded-For HTTP header in order to figure out the
  22         # origin of the client behind proxy. This allows to bypass geo
  23         # restriction by faking this header's value to some Norway IP.
  24         # We will do so once we encounter any geo restriction error.
  25         if self._faked_ip:
  26             # NB: str is intentional
  27             kwargs.setdefault(str('headers'), {})['X-Forwarded-For'] = self._faked_ip
  28         return super(NRKBaseIE, self)._download_webpage_handle(*args, **kwargs)
  29
  30     def _fake_ip(self):
  31         # Use fake IP from 37.191.128.0/17 in order to workaround geo
  32         # restriction
  33         def octet(lb=0, ub=255):
  34             return random.randint(lb, ub)
  35         self._faked_ip = '37.191.%d.%d' % (octet(128), octet())
  36
  37     def _real_extract(self, url):
  38         video_id = self._match_id(url)
  39
  40         data = self._download_json(
  41             'http://%s/mediaelement/%s' % (self._API_HOST, video_id),
  42             video_id, 'Downloading mediaelement JSON')
  43
  44         title = data.get('fullTitle') or data.get('mainTitle') or data['title']
  45         video_id = data.get('id') or video_id
  46
  47         http_headers = {'X-Forwarded-For': self._faked_ip} if self._faked_ip else {}
  48
  49         entries = []
  50
  51         conviva = data.get('convivaStatistics') or {}
  52         live = (data.get('mediaElementType') == 'Live' or
  53                 data.get('isLive') is True or conviva.get('isLive'))
  54
  55         def make_title(t):
  56             return self._live_title(t) if live else t
  57
  58         media_assets = data.get('mediaAssets')
  59         if media_assets and isinstance(media_assets, list):
  60             def video_id_and_title(idx):
  61                 return ((video_id, title) if len(media_assets) == 1
  62                         else ('%s-%d' % (video_id, idx), '%s (Part %d)' % (title, idx)))
  63             for num, asset in enumerate(media_assets, 1):
  64                 asset_url = asset.get('url')
  65                 if not asset_url:
  66                     continue
  67                 formats = self._extract_akamai_formats(asset_url, video_id)
  68                 if not formats:
  69                     continue
  70                 self._sort_formats(formats)
  71
  72                 # Some f4m streams may not work with hdcore in fragments' URLs
  73                 for f in formats:
  74                     extra_param = f.get('extra_param_to_segment_url')
  75                     if extra_param and 'hdcore' in extra_param:
  76                         del f['extra_param_to_segment_url']
  77
  78                 entry_id, entry_title = video_id_and_title(num)
  79                 duration = parse_duration(asset.get('duration'))
  80                 subtitles = {}
  81                 for subtitle in ('webVtt', 'timedText'):
  82                     subtitle_url = asset.get('%sSubtitlesUrl' % subtitle)
  83                     if subtitle_url:
  84                         subtitles.setdefault('no', []).append({
  85                             'url': compat_urllib_parse_unquote(subtitle_url)
  86                         })
  87                 entries.append({
  88                     'id': asset.get('carrierId') or entry_id,
  89                     'title': make_title(entry_title),
  90                     'duration': duration,
  91                     'subtitles': subtitles,
  92                     'formats': formats,
  93                     'http_headers': http_headers,
  94                 })
  95
  96         if not entries:
  97             media_url = data.get('mediaUrl')
  98             if media_url:
  99                 formats = self._extract_akamai_formats(media_url, video_id)
 100                 self._sort_formats(formats)
 101                 duration = parse_duration(data.get('duration'))
 102                 entries = [{
 103                     'id': video_id,
 104                     'title': make_title(title),
 105                     'duration': duration,
 106                     'formats': formats,
 107                 }]
 108
 109         if not entries:
 110             message_type = data.get('messageType', '')
 111             # Can be ProgramIsGeoBlocked or ChannelIsGeoBlocked*
 112             if 'IsGeoBlocked' in message_type and not self._faked_ip:
 113                 self.report_warning(
 114                     'Video is geo restricted, trying to fake IP')
 115                 self._fake_ip()
 116                 return self._real_extract(url)
 117
 118             MESSAGES = {
 119                 'ProgramRightsAreNotReady': 'Du kan dessverre ikke se eller høre programmet',
 120                 'ProgramRightsHasExpired': 'Programmet har gått ut',
 121                 'ProgramIsGeoBlocked': 'NRK har ikke rettigheter til å vise dette programmet utenfor Norge',
 122             }
 123             raise ExtractorError(
 124                 '%s said: %s' % (self.IE_NAME, MESSAGES.get(
 125                     message_type, message_type)),
 126                 expected=True)
 127
 128         series = conviva.get('seriesName') or data.get('seriesTitle')
 129         episode = conviva.get('episodeName') or data.get('episodeNumberOrDate')
 130
 131         season_number = None
 132         episode_number = None
 133         if data.get('mediaElementType') == 'Episode':
 134             _season_episode = data.get('scoresStatistics', {}).get('springStreamStream') or \
 135                 data.get('relativeOriginUrl', '')
 136             EPISODENUM_RE = [
 137                 r'/s(?P<season>\d{,2})e(?P<episode>\d{,2})\.',
 138                 r'/sesong-(?P<season>\d{,2})/episode-(?P<episode>\d{,2})',
 139             ]
 140             season_number = int_or_none(self._search_regex(
 141                 EPISODENUM_RE, _season_episode, 'season number',
 142                 default=None, group='season'))
 143             episode_number = int_or_none(self._search_regex(
 144                 EPISODENUM_RE, _season_episode, 'episode number',
 145                 default=None, group='episode'))
 146
 147         thumbnails = None
 148         images = data.get('images')
 149         if images and isinstance(images, dict):
 150             web_images = images.get('webImages')
 151             if isinstance(web_images, list):
 152                 thumbnails = [{
 153                     'url': image['imageUrl'],
 154                     'width': int_or_none(image.get('width')),
 155                     'height': int_or_none(image.get('height')),
 156                 } for image in web_images if image.get('imageUrl')]
 157
 158         description = data.get('description')
 159         category = data.get('mediaAnalytics', {}).get('category')
 160
 161         common_info = {
 162             'description': description,
 163             'series': series,
 164             'episode': episode,
 165             'season_number': season_number,
 166             'episode_number': episode_number,
 167             'categories': [category] if category else None,
 168             'age_limit': parse_age_limit(data.get('legalAge')),
 169             'thumbnails': thumbnails,
 170         }
 171
 172         vcodec = 'none' if data.get('mediaType') == 'Audio' else None
 173
 174         # TODO: extract chapters when https://github.com/rg3/youtube-dl/pull/9409 is merged
 175
 176         for entry in entries:
 177             entry.update(common_info)
 178             for f in entry['formats']:
 179                 f['vcodec'] = vcodec
 180
 181         return self.playlist_result(entries, video_id, title, description)
 182
 183
 184 class NRKIE(NRKBaseIE):
 185     _VALID_URL = r'''(?x)
 186                         (?:
 187                             nrk:|
 188                             https?://
 189                                 (?:
 190                                     (?:www\.)?nrk\.no/video/PS\*|
 191                                     v8-psapi\.nrk\.no/mediaelement/
 192                                 )
 193                             )
 194                             (?P<id>[^/?#&]+)
 195                         '''
 196     _API_HOST = 'v8.psapi.nrk.no'
 197     _TESTS = [{
 198         # video
 199         'url': 'http://www.nrk.no/video/PS*150533',
 200         'md5': '2f7f6eeb2aacdd99885f355428715cfa',
 201         'info_dict': {
 202             'id': '150533',
 203             'ext': 'mp4',
 204             'title': 'Dompap og andre fugler i Piip-Show',
 205             'description': 'md5:d9261ba34c43b61c812cb6b0269a5c8f',
 206             'duration': 263,
 207         }
 208     }, {
 209         # audio
 210         'url': 'http://www.nrk.no/video/PS*154915',
 211         # MD5 is unstable
 212         'info_dict': {
 213             'id': '154915',
 214             'ext': 'flv',
 215             'title': 'Slik høres internett ut når du er blind',
 216             'description': 'md5:a621f5cc1bd75c8d5104cb048c6b8568',
 217             'duration': 20,
 218         }
 219     }, {
 220         'url': 'nrk:ecc1b952-96dc-4a98-81b9-5296dc7a98d9',
 221         'only_matching': True,
 222     }, {
 223         'url': 'https://v8-psapi.nrk.no/mediaelement/ecc1b952-96dc-4a98-81b9-5296dc7a98d9',
 224         'only_matching': True,
 225     }]
 226
 227
 228 class NRKTVIE(NRKBaseIE):
 229     IE_DESC = 'NRK TV and NRK Radio'
 230     _EPISODE_RE = r'(?P<id>[a-zA-Z]{4}\d{8})'
 231     _VALID_URL = r'''(?x)
 232                         https?://
 233                             (?:tv|radio)\.nrk(?:super)?\.no/
 234                             (?:serie/[^/]+|program)/
 235                             (?![Ee]pisodes)%s
 236                             (?:/\d{2}-\d{2}-\d{4})?
 237                             (?:\#del=(?P<part_id>\d+))?
 238                     ''' % _EPISODE_RE
 239     _API_HOST = 'psapi-we.nrk.no'
 240
 241     _TESTS = [{
 242         'url': 'https://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
 243         'md5': '4e9ca6629f09e588ed240fb11619922a',
 244         'info_dict': {
 245             'id': 'MUHH48000314AA',
 246             'ext': 'mp4',
 247             'title': '20 spørsmål 23.05.2014',
 248             'description': 'md5:bdea103bc35494c143c6a9acdd84887a',
 249             'duration': 1741,
 250             'series': '20 spørsmål - TV',
 251             'episode': '23.05.2014',
 252         },
 253     }, {
 254         'url': 'https://tv.nrk.no/program/mdfp15000514',
 255         'info_dict': {
 256             'id': 'MDFP15000514CA',
 257             'ext': 'mp4',
 258             'title': 'Grunnlovsjubiléet - Stor ståhei for ingenting 24.05.2014',
 259             'description': 'md5:89290c5ccde1b3a24bb8050ab67fe1db',
 260             'duration': 4605,
 261             'series': 'Kunnskapskanalen',
 262             'episode': '24.05.2014',
 263         },
 264         'params': {
 265             'skip_download': True,
 266         },
 267     }, {
 268         # single playlist video
 269         'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2',
 270         'info_dict': {
 271             'id': 'MSPO40010515-part2',
 272             'ext': 'flv',
 273             'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
 274             'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 275         },
 276         'params': {
 277             'skip_download': True,
 278         },
 279         'expected_warnings': ['Video is geo restricted'],
 280         'skip': 'particular part is not supported currently',
 281     }, {
 282         'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015',
 283         'playlist': [{
 284             'info_dict': {
 285                 'id': 'MSPO40010515AH',
 286                 'ext': 'mp4',
 287                 'title': 'Sprint fri teknikk, kvinner og menn 06.01.2015 (Part 1)',
 288                 'description': 'md5:c03aba1e917561eface5214020551b7a',
 289                 'duration': 772,
 290                 'series': 'Tour de Ski',
 291                 'episode': '06.01.2015',
 292             },
 293             'params': {
 294                 'skip_download': True,
 295             },
 296         }, {
 297             'info_dict': {
 298                 'id': 'MSPO40010515BH',
 299                 'ext': 'mp4',
 300                 'title': 'Sprint fri teknikk, kvinner og menn 06.01.2015 (Part 2)',
 301                 'description': 'md5:c03aba1e917561eface5214020551b7a',
 302                 'duration': 6175,
 303                 'series': 'Tour de Ski',
 304                 'episode': '06.01.2015',
 305             },
 306             'params': {
 307                 'skip_download': True,
 308             },
 309         }],
 310         'info_dict': {
 311             'id': 'MSPO40010515',
 312             'title': 'Sprint fri teknikk, kvinner og menn 06.01.2015',
 313             'description': 'md5:c03aba1e917561eface5214020551b7a',
 314         },
 315         'expected_warnings': ['Video is geo restricted'],
 316     }, {
 317         'url': 'https://tv.nrk.no/serie/anno/KMTE50001317/sesong-3/episode-13',
 318         'info_dict': {
 319             'id': 'KMTE50001317AA',
 320             'ext': 'mp4',
 321             'title': 'Anno 13:30',
 322             'description': 'md5:11d9613661a8dbe6f9bef54e3a4cbbfa',
 323             'duration': 2340,
 324             'series': 'Anno',
 325             'episode': '13:30',
 326             'season_number': 3,
 327             'episode_number': 13,
 328         },
 329         'params': {
 330             'skip_download': True,
 331         },
 332     }, {
 333         'url': 'https://tv.nrk.no/serie/nytt-paa-nytt/MUHH46000317/27-01-2017',
 334         'info_dict': {
 335             'id': 'MUHH46000317AA',
 336             'ext': 'mp4',
 337             'title': 'Nytt på Nytt 27.01.2017',
 338             'description': 'md5:5358d6388fba0ea6f0b6d11c48b9eb4b',
 339             'duration': 1796,
 340             'series': 'Nytt på nytt',
 341             'episode': '27.01.2017',
 342         },
 343         'params': {
 344             'skip_download': True,
 345         },
 346     }, {
 347         'url': 'https://radio.nrk.no/serie/dagsnytt/NPUB21019315/12-07-2015#',
 348         'only_matching': True,
 349     }]
 350
 351
 352 class NRKTVDirekteIE(NRKTVIE):
 353     IE_DESC = 'NRK TV Direkte and NRK Radio Direkte'
 354     _VALID_URL = r'https?://(?:tv|radio)\.nrk\.no/direkte/(?P<id>[^/?#&]+)'
 355
 356     _TESTS = [{
 357         'url': 'https://tv.nrk.no/direkte/nrk1',
 358         'only_matching': True,
 359     }, {
 360         'url': 'https://radio.nrk.no/direkte/p1_oslo_akershus',
 361         'only_matching': True,
 362     }]
 363
 364
 365 class NRKPlaylistBaseIE(InfoExtractor):
 366     def _extract_description(self, webpage):
 367         pass
 368
 369     def _real_extract(self, url):
 370         playlist_id = self._match_id(url)
 371
 372         webpage = self._download_webpage(url, playlist_id)
 373
 374         entries = [
 375             self.url_result('nrk:%s' % video_id, NRKIE.ie_key())
 376             for video_id in re.findall(self._ITEM_RE, webpage)
 377         ]
 378
 379         playlist_title = self. _extract_title(webpage)
 380         playlist_description = self._extract_description(webpage)
 381
 382         return self.playlist_result(
 383             entries, playlist_id, playlist_title, playlist_description)
 384
 385
 386 class NRKPlaylistIE(NRKPlaylistBaseIE):
 387     _VALID_URL = r'https?://(?:www\.)?nrk\.no/(?!video|skole)(?:[^/]+/)+(?P<id>[^/]+)'
 388     _ITEM_RE = r'class="[^"]*\brich\b[^"]*"[^>]+data-video-id="([^"]+)"'
 389     _TESTS = [{
 390         'url': 'http://www.nrk.no/troms/gjenopplev-den-historiske-solformorkelsen-1.12270763',
 391         'info_dict': {
 392             'id': 'gjenopplev-den-historiske-solformorkelsen-1.12270763',
 393             'title': 'Gjenopplev den historiske solformørkelsen',
 394             'description': 'md5:c2df8ea3bac5654a26fc2834a542feed',
 395         },
 396         'playlist_count': 2,
 397     }, {
 398         'url': 'http://www.nrk.no/kultur/bok/rivertonprisen-til-karin-fossum-1.12266449',
 399         'info_dict': {
 400             'id': 'rivertonprisen-til-karin-fossum-1.12266449',
 401             'title': 'Rivertonprisen til Karin Fossum',
 402             'description': 'Første kvinne på 15 år til å vinne krimlitteraturprisen.',
 403         },
 404         'playlist_count': 5,
 405     }]
 406
 407     def _extract_title(self, webpage):
 408         return self._og_search_title(webpage, fatal=False)
 409
 410     def _extract_description(self, webpage):
 411         return self._og_search_description(webpage)
 412
 413
 414 class NRKTVEpisodesIE(NRKPlaylistBaseIE):
 415     _VALID_URL = r'https?://tv\.nrk\.no/program/[Ee]pisodes/[^/]+/(?P<id>\d+)'
 416     _ITEM_RE = r'data-episode=["\']%s' % NRKTVIE._EPISODE_RE
 417     _TESTS = [{
 418         'url': 'https://tv.nrk.no/program/episodes/nytt-paa-nytt/69031',
 419         'info_dict': {
 420             'id': '69031',
 421             'title': 'Nytt på nytt, sesong: 201210',
 422         },
 423         'playlist_count': 4,
 424     }]
 425
 426     def _extract_title(self, webpage):
 427         return self._html_search_regex(
 428             r'<h1>([^<]+)</h1>', webpage, 'title', fatal=False)
 429
 430
 431 class NRKTVSeriesIE(InfoExtractor):
 432     _VALID_URL = r'https?://(?:tv|radio)\.nrk(?:super)?\.no/serie/(?P<id>[^/]+)'
 433     _ITEM_RE = r'(?:data-season=["\']|id=["\']season-)(?P<id>\d+)'
 434     _TESTS = [{
 435         'url': 'https://tv.nrk.no/serie/groenn-glede',
 436         'info_dict': {
 437             'id': 'groenn-glede',
 438             'title': 'Grønn glede',
 439             'description': 'md5:7576e92ae7f65da6993cf90ee29e4608',
 440         },
 441         'playlist_mincount': 9,
 442     }, {
 443         'url': 'http://tv.nrksuper.no/serie/labyrint',
 444         'info_dict': {
 445             'id': 'labyrint',
 446             'title': 'Labyrint',
 447             'description': 'md5:58afd450974c89e27d5a19212eee7115',
 448         },
 449         'playlist_mincount': 3,
 450     }, {
 451         'url': 'https://tv.nrk.no/serie/broedrene-dal-og-spektralsteinene',
 452         'only_matching': True,
 453     }, {
 454         'url': 'https://tv.nrk.no/serie/saving-the-human-race',
 455         'only_matching': True,
 456     }, {
 457         'url': 'https://tv.nrk.no/serie/postmann-pat',
 458         'only_matching': True,
 459     }]
 460
 461     @classmethod
 462     def suitable(cls, url):
 463         return False if NRKTVIE.suitable(url) else super(NRKTVSeriesIE, cls).suitable(url)
 464
 465     def _real_extract(self, url):
 466         series_id = self._match_id(url)
 467
 468         webpage = self._download_webpage(url, series_id)
 469
 470         entries = [
 471             self.url_result(
 472                 'https://tv.nrk.no/program/Episodes/{series}/{season}'.format(
 473                     series=series_id, season=season_id))
 474             for season_id in re.findall(self._ITEM_RE, webpage)
 475         ]
 476
 477         title = self._html_search_meta(
 478             'seriestitle', webpage,
 479             'title', default=None) or self._og_search_title(
 480             webpage, fatal=False)
 481
 482         description = self._html_search_meta(
 483             'series_description', webpage,
 484             'description', default=None) or self._og_search_description(webpage)
 485
 486         return self.playlist_result(entries, series_id, title, description)
 487
 488
 489 class NRKSkoleIE(InfoExtractor):
 490     IE_DESC = 'NRK Skole'
 491     _VALID_URL = r'https?://(?:www\.)?nrk\.no/skole/?\?.*\bmediaId=(?P<id>\d+)'
 492
 493     _TESTS = [{
 494         'url': 'https://www.nrk.no/skole/?page=search&q=&mediaId=14099',
 495         'md5': '6bc936b01f9dd8ed45bc58b252b2d9b6',
 496         'info_dict': {
 497             'id': '6021',
 498             'ext': 'mp4',
 499             'title': 'Genetikk og eneggede tvillinger',
 500             'description': 'md5:3aca25dcf38ec30f0363428d2b265f8d',
 501             'duration': 399,
 502         },
 503     }, {
 504         'url': 'https://www.nrk.no/skole/?page=objectives&subject=naturfag&objective=K15114&mediaId=19355',
 505         'only_matching': True,
 506     }]
 507
 508     def _real_extract(self, url):
 509         video_id = self._match_id(url)
 510
 511         webpage = self._download_webpage(
 512             'https://mimir.nrk.no/plugin/1.0/static?mediaId=%s' % video_id,
 513             video_id)
 514
 515         nrk_id = self._parse_json(
 516             self._search_regex(
 517                 r'<script[^>]+type=["\']application/json["\'][^>]*>({.+?})</script>',
 518                 webpage, 'application json'),
 519             video_id)['activeMedia']['psId']
 520
 521         return self.url_result('nrk:%s' % nrk_id)