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         media_assets = data.get('mediaAssets')
  52         if media_assets and isinstance(media_assets, list):
  53             def video_id_and_title(idx):
  54                 return ((video_id, title) if len(media_assets) == 1
  55                         else ('%s-%d' % (video_id, idx), '%s (Part %d)' % (title, idx)))
  56             for num, asset in enumerate(media_assets, 1):
  57                 asset_url = asset.get('url')
  58                 if not asset_url:
  59                     continue
  60                 formats = self._extract_akamai_formats(asset_url, video_id)
  61                 if not formats:
  62                     continue
  63                 self._sort_formats(formats)
  64                 entry_id, entry_title = video_id_and_title(num)
  65                 duration = parse_duration(asset.get('duration'))
  66                 subtitles = {}
  67                 for subtitle in ('webVtt', 'timedText'):
  68                     subtitle_url = asset.get('%sSubtitlesUrl' % subtitle)
  69                     if subtitle_url:
  70                         subtitles.setdefault('no', []).append({
  71                             'url': compat_urllib_parse_unquote(subtitle_url)
  72                         })
  73                 entries.append({
  74                     'id': asset.get('carrierId') or entry_id,
  75                     'title': entry_title,
  76                     'duration': duration,
  77                     'subtitles': subtitles,
  78                     'formats': formats,
  79                     'http_headers': http_headers,
  80                 })
  81
  82         if not entries:
  83             media_url = data.get('mediaUrl')
  84             if media_url:
  85                 formats = self._extract_akamai_formats(media_url, video_id)
  86                 self._sort_formats(formats)
  87                 duration = parse_duration(data.get('duration'))
  88                 entries = [{
  89                     'id': video_id,
  90                     'title': title,
  91                     'duration': duration,
  92                     'formats': formats,
  93                 }]
  94
  95         if not entries:
  96             message_type = data.get('messageType', '')
  97             # Can be ProgramIsGeoBlocked or ChannelIsGeoBlocked*
  98             if 'IsGeoBlocked' in message_type and not self._faked_ip:
  99                 self.report_warning(
 100                     'Video is geo restricted, trying to fake IP')
 101                 self._fake_ip()
 102                 return self._real_extract(url)
 103
 104             MESSAGES = {
 105                 'ProgramRightsAreNotReady': 'Du kan dessverre ikke se eller høre programmet',
 106                 'ProgramRightsHasExpired': 'Programmet har gått ut',
 107                 'ProgramIsGeoBlocked': 'NRK har ikke rettigheter til å vise dette programmet utenfor Norge',
 108             }
 109             raise ExtractorError(
 110                 '%s said: %s' % (self.IE_NAME, MESSAGES.get(
 111                     message_type, message_type)),
 112                 expected=True)
 113
 114         conviva = data.get('convivaStatistics') or {}
 115         series = conviva.get('seriesName') or data.get('seriesTitle')
 116         episode = conviva.get('episodeName') or data.get('episodeNumberOrDate')
 117
 118         thumbnails = None
 119         images = data.get('images')
 120         if images and isinstance(images, dict):
 121             web_images = images.get('webImages')
 122             if isinstance(web_images, list):
 123                 thumbnails = [{
 124                     'url': image['imageUrl'],
 125                     'width': int_or_none(image.get('width')),
 126                     'height': int_or_none(image.get('height')),
 127                 } for image in web_images if image.get('imageUrl')]
 128
 129         description = data.get('description')
 130
 131         common_info = {
 132             'description': description,
 133             'series': series,
 134             'episode': episode,
 135             'age_limit': parse_age_limit(data.get('legalAge')),
 136             'thumbnails': thumbnails,
 137         }
 138
 139         vcodec = 'none' if data.get('mediaType') == 'Audio' else None
 140
 141         # TODO: extract chapters when https://github.com/rg3/youtube-dl/pull/9409 is merged
 142
 143         for entry in entries:
 144             entry.update(common_info)
 145             for f in entry['formats']:
 146                 f['vcodec'] = vcodec
 147
 148         return self.playlist_result(entries, video_id, title, description)
 149
 150
 151 class NRKIE(NRKBaseIE):
 152     _VALID_URL = r'''(?x)
 153                         (?:
 154                             nrk:|
 155                             https?://
 156                                 (?:
 157                                     (?:www\.)?nrk\.no/video/PS\*|
 158                                     v8-psapi\.nrk\.no/mediaelement/
 159                                 )
 160                             )
 161                             (?P<id>[^/?#&]+)
 162                         '''
 163     _API_HOST = 'v8.psapi.nrk.no'
 164     _TESTS = [{
 165         # video
 166         'url': 'http://www.nrk.no/video/PS*150533',
 167         'md5': '2f7f6eeb2aacdd99885f355428715cfa',
 168         'info_dict': {
 169             'id': '150533',
 170             'ext': 'mp4',
 171             'title': 'Dompap og andre fugler i Piip-Show',
 172             'description': 'md5:d9261ba34c43b61c812cb6b0269a5c8f',
 173             'duration': 263,
 174         }
 175     }, {
 176         # audio
 177         'url': 'http://www.nrk.no/video/PS*154915',
 178         # MD5 is unstable
 179         'info_dict': {
 180             'id': '154915',
 181             'ext': 'flv',
 182             'title': 'Slik høres internett ut når du er blind',
 183             'description': 'md5:a621f5cc1bd75c8d5104cb048c6b8568',
 184             'duration': 20,
 185         }
 186     }, {
 187         'url': 'nrk:ecc1b952-96dc-4a98-81b9-5296dc7a98d9',
 188         'only_matching': True,
 189     }, {
 190         'url': 'https://v8-psapi.nrk.no/mediaelement/ecc1b952-96dc-4a98-81b9-5296dc7a98d9',
 191         'only_matching': True,
 192     }]
 193
 194
 195 class NRKTVIE(NRKBaseIE):
 196     IE_DESC = 'NRK TV and NRK Radio'
 197     _VALID_URL = r'https?://(?:tv|radio)\.nrk(?:super)?\.no/(?:serie/[^/]+|program)/(?P<id>[a-zA-Z]{4}\d{8})(?:/\d{2}-\d{2}-\d{4})?(?:#del=(?P<part_id>\d+))?'
 198     _API_HOST = 'psapi-we.nrk.no'
 199
 200     _TESTS = [{
 201         'url': 'https://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
 202         'md5': '4e9ca6629f09e588ed240fb11619922a',
 203         'info_dict': {
 204             'id': 'MUHH48000314AA',
 205             'ext': 'mp4',
 206             'title': '20 spørsmål 23.05.2014',
 207             'description': 'md5:bdea103bc35494c143c6a9acdd84887a',
 208             'duration': 1741,
 209         },
 210     }, {
 211         'url': 'https://tv.nrk.no/program/mdfp15000514',
 212         'md5': '43d0be26663d380603a9cf0c24366531',
 213         'info_dict': {
 214             'id': 'MDFP15000514CA',
 215             'ext': 'mp4',
 216             'title': 'Grunnlovsjubiléet - Stor ståhei for ingenting 24.05.2014',
 217             'description': 'md5:89290c5ccde1b3a24bb8050ab67fe1db',
 218             'duration': 4605,
 219         },
 220     }, {
 221         # single playlist video
 222         'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2',
 223         'md5': 'adbd1dbd813edaf532b0a253780719c2',
 224         'info_dict': {
 225             'id': 'MSPO40010515-part2',
 226             'ext': 'flv',
 227             'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
 228             'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 229         },
 230         'skip': 'Only works from Norway',
 231     }, {
 232         'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015',
 233         'playlist': [{
 234             'md5': '9480285eff92d64f06e02a5367970a7a',
 235             'info_dict': {
 236                 'id': 'MSPO40010515-part1',
 237                 'ext': 'flv',
 238                 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 1:2)',
 239                 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 240             },
 241         }, {
 242             'md5': 'adbd1dbd813edaf532b0a253780719c2',
 243             'info_dict': {
 244                 'id': 'MSPO40010515-part2',
 245                 'ext': 'flv',
 246                 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
 247                 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 248             },
 249         }],
 250         'info_dict': {
 251             'id': 'MSPO40010515',
 252             'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn',
 253             'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
 254             'duration': 6947.52,
 255         },
 256         'skip': 'Only works from Norway',
 257     }, {
 258         'url': 'https://radio.nrk.no/serie/dagsnytt/NPUB21019315/12-07-2015#',
 259         'only_matching': True,
 260     }]
 261
 262
 263 class NRKPlaylistIE(InfoExtractor):
 264     _VALID_URL = r'https?://(?:www\.)?nrk\.no/(?!video|skole)(?:[^/]+/)+(?P<id>[^/]+)'
 265
 266     _TESTS = [{
 267         'url': 'http://www.nrk.no/troms/gjenopplev-den-historiske-solformorkelsen-1.12270763',
 268         'info_dict': {
 269             'id': 'gjenopplev-den-historiske-solformorkelsen-1.12270763',
 270             'title': 'Gjenopplev den historiske solformørkelsen',
 271             'description': 'md5:c2df8ea3bac5654a26fc2834a542feed',
 272         },
 273         'playlist_count': 2,
 274     }, {
 275         'url': 'http://www.nrk.no/kultur/bok/rivertonprisen-til-karin-fossum-1.12266449',
 276         'info_dict': {
 277             'id': 'rivertonprisen-til-karin-fossum-1.12266449',
 278             'title': 'Rivertonprisen til Karin Fossum',
 279             'description': 'Første kvinne på 15 år til å vinne krimlitteraturprisen.',
 280         },
 281         'playlist_count': 5,
 282     }]
 283
 284     def _real_extract(self, url):
 285         playlist_id = self._match_id(url)
 286
 287         webpage = self._download_webpage(url, playlist_id)
 288
 289         entries = [
 290             self.url_result('nrk:%s' % video_id, 'NRK')
 291             for video_id in re.findall(
 292                 r'class="[^"]*\brich\b[^"]*"[^>]+data-video-id="([^"]+)"',
 293                 webpage)
 294         ]
 295
 296         playlist_title = self._og_search_title(webpage)
 297         playlist_description = self._og_search_description(webpage)
 298
 299         return self.playlist_result(
 300             entries, playlist_id, playlist_title, playlist_description)
 301
 302
 303 class NRKSkoleIE(InfoExtractor):
 304     IE_DESC = 'NRK Skole'
 305     _VALID_URL = r'https?://(?:www\.)?nrk\.no/skole/?\?.*\bmediaId=(?P<id>\d+)'
 306
 307     _TESTS = [{
 308         'url': 'https://www.nrk.no/skole/?page=search&q=&mediaId=14099',
 309         'md5': '6bc936b01f9dd8ed45bc58b252b2d9b6',
 310         'info_dict': {
 311             'id': '6021',
 312             'ext': 'mp4',
 313             'title': 'Genetikk og eneggede tvillinger',
 314             'description': 'md5:3aca25dcf38ec30f0363428d2b265f8d',
 315             'duration': 399,
 316         },
 317     }, {
 318         'url': 'https://www.nrk.no/skole/?page=objectives&subject=naturfag&objective=K15114&mediaId=19355',
 319         'only_matching': True,
 320     }]
 321
 322     def _real_extract(self, url):
 323         video_id = self._match_id(url)
 324
 325         webpage = self._download_webpage(
 326             'https://mimir.nrk.no/plugin/1.0/static?mediaId=%s' % video_id,
 327             video_id)
 328
 329         nrk_id = self._parse_json(
 330             self._search_regex(
 331                 r'<script[^>]+type=["\']application/json["\'][^>]*>({.+?})</script>',
 332                 webpage, 'application json'),
 333             video_id)['activeMedia']['psId']
 334
 335         return self.url_result('nrk:%s' % nrk_id)