yt_dlp/extractor/redbee.py

   1 import json
   2 import re
   3 import time
   4 import urllib.parse
   5 import uuid
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     ExtractorError,
  10     float_or_none,
  11     int_or_none,
  12     strip_or_none,
  13     traverse_obj,
  14     try_call,
  15     unified_timestamp,
  16 )
  17
  18
  19 class RedBeeBaseIE(InfoExtractor):
  20     _DEVICE_ID = str(uuid.uuid4())
  21
  22     @property
  23     def _API_URL(self):
  24         """
  25         Ref: https://apidocs.emp.ebsd.ericsson.net
  26         Subclasses must set _REDBEE_CUSTOMER, _REDBEE_BUSINESS_UNIT
  27         """
  28         return f'https://exposure.api.redbee.live/v2/customer/{self._REDBEE_CUSTOMER}/businessunit/{self._REDBEE_BUSINESS_UNIT}'
  29
  30     def _get_bearer_token(self, asset_id, jwt=None):
  31         request = {
  32             'deviceId': self._DEVICE_ID,
  33             'device': {
  34                 'deviceId': self._DEVICE_ID,
  35                 'name': 'Mozilla Firefox 102',
  36                 'type': 'WEB',
  37             },
  38         }
  39         if jwt:
  40             request['jwt'] = jwt
  41
  42         return self._download_json(
  43             f'{self._API_URL}/auth/{"gigyaLogin" if jwt else "anonymous"}',
  44             asset_id, data=json.dumps(request).encode('utf-8'), headers={
  45                 'Content-Type': 'application/json;charset=utf-8'
  46             })['sessionToken']
  47
  48     def _get_formats_and_subtitles(self, asset_id, **kwargs):
  49         bearer_token = self._get_bearer_token(asset_id, **kwargs)
  50         api_response = self._download_json(
  51             f'{self._API_URL}/entitlement/{asset_id}/play',
  52             asset_id, headers={
  53                 'Authorization': f'Bearer {bearer_token}',
  54                 'Accept': 'application/json, text/plain, */*'
  55             })
  56
  57         formats, subtitles = [], {}
  58         for format in api_response['formats']:
  59             if not format.get('mediaLocator'):
  60                 continue
  61
  62             fmts, subs = [], {}
  63             if format.get('format') == 'DASH':
  64                 fmts, subs = self._extract_mpd_formats_and_subtitles(
  65                     format['mediaLocator'], asset_id, fatal=False)
  66             elif format.get('format') == 'SMOOTHSTREAMING':
  67                 fmts, subs = self._extract_ism_formats_and_subtitles(
  68                     format['mediaLocator'], asset_id, fatal=False)
  69             elif format.get('format') == 'HLS':
  70                 fmts, subs = self._extract_m3u8_formats_and_subtitles(
  71                     format['mediaLocator'], asset_id, fatal=False)
  72
  73             if format.get('drm'):
  74                 for f in fmts:
  75                     f['has_drm'] = True
  76
  77             formats.extend(fmts)
  78             self._merge_subtitles(subs, target=subtitles)
  79
  80         return formats, subtitles
  81
  82
  83 class ParliamentLiveUKIE(RedBeeBaseIE):
  84     IE_NAME = 'parliamentlive.tv'
  85     IE_DESC = 'UK parliament videos'
  86     _VALID_URL = r'(?i)https?://(?:www\.)?parliamentlive\.tv/Event/Index/(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'
  87
  88     _REDBEE_CUSTOMER = 'UKParliament'
  89     _REDBEE_BUSINESS_UNIT = 'ParliamentLive'
  90
  91     _TESTS = [{
  92         'url': 'http://parliamentlive.tv/Event/Index/c1e9d44d-fd6c-4263-b50f-97ed26cc998b',
  93         'info_dict': {
  94             'id': 'c1e9d44d-fd6c-4263-b50f-97ed26cc998b',
  95             'ext': 'mp4',
  96             'title': 'Home Affairs Committee',
  97             'timestamp': 1395153872,
  98             'upload_date': '20140318',
  99             'thumbnail': r're:https?://[^?#]+c1e9d44d-fd6c-4263-b50f-97ed26cc998b[^/]*/thumbnail',
 100         },
 101     }, {
 102         'url': 'http://parliamentlive.tv/event/index/3f24936f-130f-40bf-9a5d-b3d6479da6a4',
 103         'only_matching': True,
 104     }, {
 105         'url': 'https://parliamentlive.tv/Event/Index/27cf25e4-e77b-42a3-93c5-c815cd6d7377',
 106         'info_dict': {
 107             'id': '27cf25e4-e77b-42a3-93c5-c815cd6d7377',
 108             'ext': 'mp4',
 109             'title': 'House of Commons',
 110             'timestamp': 1658392447,
 111             'upload_date': '20220721',
 112             'thumbnail': r're:https?://[^?#]+27cf25e4-e77b-42a3-93c5-c815cd6d7377[^/]*/thumbnail',
 113         },
 114     }]
 115
 116     def _real_extract(self, url):
 117         video_id = self._match_id(url)
 118
 119         formats, subtitles = self._get_formats_and_subtitles(video_id)
 120         self._sort_formats(formats)
 121
 122         video_info = self._download_json(
 123             f'https://www.parliamentlive.tv/Event/GetShareVideo/{video_id}', video_id, fatal=False)
 124
 125         self._sort_formats(formats, ['res', 'proto'])
 126
 127         return {
 128             'id': video_id,
 129             'formats': formats,
 130             'subtitles': subtitles,
 131             'title': traverse_obj(video_info, ('event', 'title')),
 132             'thumbnail': traverse_obj(video_info, 'thumbnailUrl'),
 133             'timestamp': traverse_obj(
 134                 video_info, ('event', 'publishedStartTime'), expected_type=unified_timestamp),
 135         }
 136
 137
 138 class RTBFIE(RedBeeBaseIE):
 139     _VALID_URL = r'''(?x)
 140         https?://(?:www\.)?rtbf\.be/
 141         (?:
 142             video/[^?]+\?.*\bid=|
 143             ouftivi/(?:[^/]+/)*[^?]+\?.*\bvideoId=|
 144             auvio/[^/]+\?.*\b(?P<live>l)?id=
 145         )(?P<id>\d+)'''
 146     _NETRC_MACHINE = 'rtbf'
 147
 148     _REDBEE_CUSTOMER = 'RTBF'
 149     _REDBEE_BUSINESS_UNIT = 'Auvio'
 150
 151     _TESTS = [{
 152         'url': 'https://www.rtbf.be/video/detail_les-diables-au-coeur-episode-2?id=1921274',
 153         'md5': '8c876a1cceeb6cf31b476461ade72384',
 154         'info_dict': {
 155             'id': '1921274',
 156             'ext': 'mp4',
 157             'title': 'Les Diables au coeur (épisode 2)',
 158             'description': '(du 25/04/2014)',
 159             'duration': 3099.54,
 160             'upload_date': '20140425',
 161             'timestamp': 1398456300,
 162         },
 163         'skip': 'No longer available',
 164     }, {
 165         # geo restricted
 166         'url': 'http://www.rtbf.be/ouftivi/heros/detail_scooby-doo-mysteres-associes?id=1097&videoId=2057442',
 167         'only_matching': True,
 168     }, {
 169         'url': 'http://www.rtbf.be/ouftivi/niouzz?videoId=2055858',
 170         'only_matching': True,
 171     }, {
 172         'url': 'http://www.rtbf.be/auvio/detail_jeudi-en-prime-siegfried-bracke?id=2102996',
 173         'only_matching': True,
 174     }, {
 175         # Live
 176         'url': 'https://www.rtbf.be/auvio/direct_pure-fm?lid=134775',
 177         'only_matching': True,
 178     }, {
 179         # Audio
 180         'url': 'https://www.rtbf.be/auvio/detail_cinq-heures-cinema?id=2360811',
 181         'only_matching': True,
 182     }, {
 183         # With Subtitle
 184         'url': 'https://www.rtbf.be/auvio/detail_les-carnets-du-bourlingueur?id=2361588',
 185         'only_matching': True,
 186     }, {
 187         'url': 'https://www.rtbf.be/auvio/detail_investigation?id=2921926',
 188         'md5': 'd5d11bb62169fef38d7ce7ac531e034f',
 189         'info_dict': {
 190             'id': '2921926',
 191             'ext': 'mp4',
 192             'title': 'Le handicap un confinement perpétuel - Maladie de Lyme',
 193             'description': 'md5:dcbd5dcf6015488c9069b057c15ccc52',
 194             'duration': 5258.8,
 195             'upload_date': '20220727',
 196             'timestamp': 1658934000,
 197             'series': '#Investigation',
 198             'thumbnail': r're:^https?://[^?&]+\.jpg$',
 199         },
 200     }, {
 201         'url': 'https://www.rtbf.be/auvio/detail_la-belgique-criminelle?id=2920492',
 202         'md5': '054f9f143bc79c89647c35e5a7d35fa8',
 203         'info_dict': {
 204             'id': '2920492',
 205             'ext': 'mp4',
 206             'title': '04 - Le crime de la rue Royale',
 207             'description': 'md5:0c3da1efab286df83f2ab3f8f96bd7a6',
 208             'duration': 1574.6,
 209             'upload_date': '20220723',
 210             'timestamp': 1658596887,
 211             'series': 'La Belgique criminelle - TV',
 212             'thumbnail': r're:^https?://[^?&]+\.jpg$',
 213         },
 214     }]
 215
 216     _IMAGE_HOST = 'http://ds1.ds.static.rtbf.be'
 217     _PROVIDERS = {
 218         'YOUTUBE': 'Youtube',
 219         'DAILYMOTION': 'Dailymotion',
 220         'VIMEO': 'Vimeo',
 221     }
 222     _QUALITIES = [
 223         ('mobile', 'SD'),
 224         ('web', 'MD'),
 225         ('high', 'HD'),
 226     ]
 227     _LOGIN_URL = 'https://login.rtbf.be/accounts.login'
 228     _GIGYA_API_KEY = '3_kWKuPgcdAybqnqxq_MvHVk0-6PN8Zk8pIIkJM_yXOu-qLPDDsGOtIDFfpGivtbeO'
 229     _LOGIN_COOKIE_ID = f'glt_{_GIGYA_API_KEY}'
 230
 231     def _perform_login(self, username, password):
 232         if self._get_cookies(self._LOGIN_URL).get(self._LOGIN_COOKIE_ID):
 233             return
 234
 235         self._set_cookie('.rtbf.be', 'gmid', 'gmid.ver4', secure=True, expire_time=time.time() + 3600)
 236
 237         login_response = self._download_json(
 238             self._LOGIN_URL, None, data=urllib.parse.urlencode({
 239                 'loginID': username,
 240                 'password': password,
 241                 'APIKey': self._GIGYA_API_KEY,
 242                 'targetEnv': 'jssdk',
 243                 'sessionExpiration': '-2',
 244             }).encode('utf-8'), headers={
 245                 'Content-Type': 'application/x-www-form-urlencoded',
 246             })
 247
 248         if login_response['statusCode'] != 200:
 249             raise ExtractorError('Login failed. Server message: %s' % login_response['errorMessage'], expected=True)
 250
 251         self._set_cookie('.rtbf.be', self._LOGIN_COOKIE_ID, login_response['sessionInfo']['login_token'],
 252                          secure=True, expire_time=time.time() + 3600)
 253
 254     def _get_formats_and_subtitles(self, url, media_id):
 255         login_token = self._get_cookies(url).get(self._LOGIN_COOKIE_ID)
 256         if not login_token:
 257             self.raise_login_required()
 258
 259         session_jwt = try_call(lambda: self._get_cookies(url)['rtbf_jwt'].value) or self._download_json(
 260             'https://login.rtbf.be/accounts.getJWT', media_id, query={
 261                 'login_token': login_token.value,
 262                 'APIKey': self._GIGYA_API_KEY,
 263                 'sdk': 'js_latest',
 264                 'authMode': 'cookie',
 265                 'pageURL': url,
 266                 'sdkBuild': '13273',
 267                 'format': 'json',
 268             })['id_token']
 269
 270         return super()._get_formats_and_subtitles(media_id, jwt=session_jwt)
 271
 272     def _real_extract(self, url):
 273         live, media_id = self._match_valid_url(url).groups()
 274         embed_page = self._download_webpage(
 275             'https://www.rtbf.be/auvio/embed/' + ('direct' if live else 'media'),
 276             media_id, query={'id': media_id})
 277
 278         media_data = self._html_search_regex(r'data-media="([^"]+)"', embed_page, 'media data', fatal=False)
 279         if not media_data:
 280             if re.search(r'<div[^>]+id="js-error-expired"[^>]+class="(?![^"]*hidden)', embed_page):
 281                 raise ExtractorError('Livestream has ended.', expected=True)
 282             if re.search(r'<div[^>]+id="js-sso-connect"[^>]+class="(?![^"]*hidden)', embed_page):
 283                 self.raise_login_required()
 284
 285             raise ExtractorError('Could not find media data')
 286
 287         data = self._parse_json(media_data, media_id)
 288
 289         error = data.get('error')
 290         if error:
 291             raise ExtractorError('%s said: %s' % (self.IE_NAME, error), expected=True)
 292
 293         provider = data.get('provider')
 294         if provider in self._PROVIDERS:
 295             return self.url_result(data['url'], self._PROVIDERS[provider])
 296
 297         title = traverse_obj(data, 'subtitle', 'title')
 298         is_live = data.get('isLive')
 299         height_re = r'-(\d+)p\.'
 300         formats, subtitles = [], {}
 301
 302         # The old api still returns m3u8 and mpd manifest for livestreams, but these are 'fake'
 303         # since all they contain is a 20s video that is completely unrelated.
 304         # https://github.com/yt-dlp/yt-dlp/issues/4656#issuecomment-1214461092
 305         m3u8_url = None if data.get('isLive') else traverse_obj(data, 'urlHlsAes128', 'urlHls')
 306         if m3u8_url:
 307             fmts, subs = self._extract_m3u8_formats_and_subtitles(
 308                 m3u8_url, media_id, 'mp4', m3u8_id='hls', fatal=False)
 309             formats.extend(fmts)
 310             self._merge_subtitles(subs, target=subtitles)
 311
 312         fix_url = lambda x: x.replace('//rtbf-vod.', '//rtbf.') if '/geo/drm/' in x else x
 313         http_url = data.get('url')
 314         if formats and http_url and re.search(height_re, http_url):
 315             http_url = fix_url(http_url)
 316             for m3u8_f in formats[:]:
 317                 height = m3u8_f.get('height')
 318                 if not height:
 319                     continue
 320                 f = m3u8_f.copy()
 321                 del f['protocol']
 322                 f.update({
 323                     'format_id': m3u8_f['format_id'].replace('hls-', 'http-'),
 324                     'url': re.sub(height_re, '-%dp.' % height, http_url),
 325                 })
 326                 formats.append(f)
 327         else:
 328             sources = data.get('sources') or {}
 329             for key, format_id in self._QUALITIES:
 330                 format_url = sources.get(key)
 331                 if not format_url:
 332                     continue
 333                 height = int_or_none(self._search_regex(
 334                     height_re, format_url, 'height', default=None))
 335                 formats.append({
 336                     'format_id': format_id,
 337                     'url': fix_url(format_url),
 338                     'height': height,
 339                 })
 340
 341         mpd_url = None if data.get('isLive') else data.get('urlDash')
 342         if mpd_url and (self.get_param('allow_unplayable_formats') or not data.get('drm')):
 343             fmts, subs = self._extract_mpd_formats_and_subtitles(
 344                 mpd_url, media_id, mpd_id='dash', fatal=False)
 345             formats.extend(fmts)
 346             self._merge_subtitles(subs, target=subtitles)
 347
 348         audio_url = data.get('urlAudio')
 349         if audio_url:
 350             formats.append({
 351                 'format_id': 'audio',
 352                 'url': audio_url,
 353                 'vcodec': 'none',
 354             })
 355
 356         for track in (data.get('tracks') or {}).values():
 357             sub_url = track.get('url')
 358             if not sub_url:
 359                 continue
 360             subtitles.setdefault(track.get('lang') or 'fr', []).append({
 361                 'url': sub_url,
 362             })
 363
 364         if not formats:
 365             fmts, subs = self._get_formats_and_subtitles(url, f'live_{media_id}' if is_live else media_id)
 366             formats.extend(fmts)
 367             self._merge_subtitles(subs, target=subtitles)
 368
 369         self._sort_formats(formats, ['res', 'proto'])
 370         return {
 371             'id': media_id,
 372             'formats': formats,
 373             'title': title,
 374             'description': strip_or_none(data.get('description')),
 375             'thumbnail': data.get('thumbnail'),
 376             'duration': float_or_none(data.get('realDuration')),
 377             'timestamp': int_or_none(data.get('liveFrom')),
 378             'series': data.get('programLabel'),
 379             'subtitles': subtitles,
 380             'is_live': is_live,
 381         }