yt_dlp/extractor/kaltura.py

   1 import base64
   2 import contextlib
   3 import json
   4 import re
   5 import urllib.parse
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     ExtractorError,
  10     clean_html,
  11     format_field,
  12     int_or_none,
  13     remove_start,
  14     smuggle_url,
  15     traverse_obj,
  16     unsmuggle_url,
  17 )
  18
  19
  20 class KalturaIE(InfoExtractor):
  21     _VALID_URL = r'''(?x)
  22                 (?:
  23                     kaltura:(?P<partner_id>\w+):(?P<id>\w+)(?::(?P<player_type>\w+))?|
  24                     https?://
  25                         (?:(?:www|cdnapi(?:sec)?)\.)?kaltura\.com(?::\d+)?/
  26                         (?:
  27                             (?:
  28                                 # flash player
  29                                 index\.php/(?:kwidget|extwidget/preview)|
  30                                 # html5 player
  31                                 html5/html5lib/[^/]+/mwEmbedFrame\.php
  32                             )
  33                         )(?:/(?P<path>[^?]+))?(?:\?(?P<query>.*))?
  34                 )
  35                 '''
  36     _SERVICE_URL = 'http://cdnapi.kaltura.com'
  37     _SERVICE_BASE = '/api_v3/service/multirequest'
  38     # See https://github.com/kaltura/server/blob/master/plugins/content/caption/base/lib/model/enums/CaptionType.php
  39     _CAPTION_TYPES = {
  40         1: 'srt',
  41         2: 'ttml',
  42         3: 'vtt',
  43     }
  44     _TESTS = [
  45         {
  46             'url': 'kaltura:269692:1_1jc2y3e4',
  47             'md5': '3adcbdb3dcc02d647539e53f284ba171',
  48             'info_dict': {
  49                 'id': '1_1jc2y3e4',
  50                 'ext': 'mp4',
  51                 'title': 'Straight from the Heart',
  52                 'upload_date': '20131219',
  53                 'uploader_id': 'mlundberg@wolfgangsvault.com',
  54                 'description': 'The Allman Brothers Band, 12/16/1981',
  55                 'thumbnail': 're:^https?://.*/thumbnail/.*',
  56                 'timestamp': int,
  57             },
  58             'skip': 'The access to this service is forbidden since the specified partner is blocked',
  59         },
  60         {
  61             'url': 'http://www.kaltura.com/index.php/kwidget/cache_st/1300318621/wid/_269692/uiconf_id/3873291/entry_id/1_1jc2y3e4',
  62             'only_matching': True,
  63         },
  64         {
  65             'url': 'https://cdnapisec.kaltura.com/index.php/kwidget/wid/_557781/uiconf_id/22845202/entry_id/1_plr1syf3',
  66             'only_matching': True,
  67         },
  68         {
  69             'url': 'https://cdnapisec.kaltura.com/html5/html5lib/v2.30.2/mwEmbedFrame.php/p/1337/uiconf_id/20540612/entry_id/1_sf5ovm7u?wid=_243342',
  70             'only_matching': True,
  71         },
  72         {
  73             # video with subtitles
  74             'url': 'kaltura:111032:1_cw786r8q',
  75             'only_matching': True,
  76         },
  77         {
  78             # video with ttml subtitles (no fileExt)
  79             'url': 'kaltura:1926081:0_l5ye1133',
  80             'info_dict': {
  81                 'id': '0_l5ye1133',
  82                 'ext': 'mp4',
  83                 'title': 'What Can You Do With Python?',
  84                 'upload_date': '20160221',
  85                 'uploader_id': 'stork',
  86                 'thumbnail': 're:^https?://.*/thumbnail/.*',
  87                 'timestamp': int,
  88                 'subtitles': {
  89                     'en': [{
  90                         'ext': 'ttml',
  91                     }],
  92                 },
  93             },
  94             'skip': 'Gone. Maybe https://www.safaribooksonline.com/library/tutorials/introduction-to-python-anon/3469/',
  95             'params': {
  96                 'skip_download': True,
  97             },
  98         },
  99         {
 100             'url': 'https://www.kaltura.com/index.php/extwidget/preview/partner_id/1770401/uiconf_id/37307382/entry_id/0_58u8kme7/embed/iframe?&flashvars[streamerType]=auto',
 101             'only_matching': True,
 102         },
 103         {
 104             'url': 'https://www.kaltura.com:443/index.php/extwidget/preview/partner_id/1770401/uiconf_id/37307382/entry_id/0_58u8kme7/embed/iframe?&flashvars[streamerType]=auto',
 105             'only_matching': True,
 106         },
 107         {
 108             # unavailable source format
 109             'url': 'kaltura:513551:1_66x4rg7o',
 110             'only_matching': True,
 111         },
 112         {
 113             # html5lib URL using kwidget player
 114             'url': 'https://cdnapisec.kaltura.com/html5/html5lib/v2.46/mwEmbedFrame.php/p/691292/uiconf_id/20499062/entry_id/0_c076mna6?wid=_691292&iframeembed=true&playerId=kaltura_player_1420508608&entry_id=0_c076mna6&flashvars%5BakamaiHD.loadingPolicy%5D=preInitialize&flashvars%5BakamaiHD.asyncInit%5D=true&flashvars%5BstreamerType%5D=hdnetwork',
 115             'info_dict': {
 116                 'id': '0_c076mna6',
 117                 'ext': 'mp4',
 118                 'title': 'md5:4883e7acbcbf42583a2dddc97dee4855',
 119                 'duration': 3608,
 120                 'uploader_id': 'commons@swinburne.edu.au',
 121                 'timestamp': 1408086874,
 122                 'view_count': int,
 123                 'upload_date': '20140815',
 124                 'thumbnail': 'http://cfvod.kaltura.com/p/691292/sp/69129200/thumbnail/entry_id/0_c076mna6/version/100022',
 125             },
 126         },
 127         {
 128             # html5lib playlist URL using kwidget player
 129             'url': 'https://cdnapisec.kaltura.com/html5/html5lib/v2.89/mwEmbedFrame.php/p/2019031/uiconf_id/40436601?wid=1_4j3m32cv&iframeembed=true&playerId=kaltura_player_&flashvars[playlistAPI.kpl0Id]=1_jovey5nu&flashvars[ks]=&&flashvars[imageDefaultDuration]=30&flashvars[localizationCode]=en&flashvars[leadWithHTML5]=true&flashvars[forceMobileHTML5]=true&flashvars[nextPrevBtn.plugin]=true&flashvars[hotspots.plugin]=true&flashvars[sideBarContainer.plugin]=true&flashvars[sideBarContainer.position]=left&flashvars[sideBarContainer.clickToClose]=true&flashvars[chapters.plugin]=true&flashvars[chapters.layout]=vertical&flashvars[chapters.thumbnailRotator]=false&flashvars[streamSelector.plugin]=true&flashvars[EmbedPlayer.SpinnerTarget]=videoHolder&flashvars[dualScreen.plugin]=true&flashvars[playlistAPI.playlistUrl]=https://canvasgatechtest.kaf.kaltura.com/playlist/details/{playlistAPI.kpl0Id}/categoryid/126428551',
 130             'info_dict': {
 131                 'id': '1_jovey5nu',
 132                 'title': '00-00 Introduction',
 133             },
 134             'playlist': [
 135                 {
 136                     'info_dict': {
 137                         'id': '1_b1y5hlvx',
 138                         'ext': 'mp4',
 139                         'title': 'CS7646_00-00 Introductio_Introduction',
 140                         'duration': 91,
 141                         'thumbnail': 'http://cfvod.kaltura.com/p/2019031/sp/201903100/thumbnail/entry_id/1_b1y5hlvx/version/100001',
 142                         'view_count': int,
 143                         'timestamp': 1533154447,
 144                         'upload_date': '20180801',
 145                         'uploader_id': 'djoyner3',
 146                     },
 147                 }, {
 148                     'info_dict': {
 149                         'id': '1_jfb7mdpn',
 150                         'ext': 'mp4',
 151                         'title': 'CS7646_00-00 Introductio_Three parts to the course',
 152                         'duration': 63,
 153                         'thumbnail': 'http://cfvod.kaltura.com/p/2019031/sp/201903100/thumbnail/entry_id/1_jfb7mdpn/version/100001',
 154                         'view_count': int,
 155                         'timestamp': 1533154489,
 156                         'upload_date': '20180801',
 157                         'uploader_id': 'djoyner3',
 158                     },
 159                 }, {
 160                     'info_dict': {
 161                         'id': '1_8xflxdp7',
 162                         'ext': 'mp4',
 163                         'title': 'CS7646_00-00 Introductio_Textbooks',
 164                         'duration': 37,
 165                         'thumbnail': 'http://cfvod.kaltura.com/p/2019031/sp/201903100/thumbnail/entry_id/1_8xflxdp7/version/100001',
 166                         'view_count': int,
 167                         'timestamp': 1533154512,
 168                         'upload_date': '20180801',
 169                         'uploader_id': 'djoyner3',
 170                     },
 171                 }, {
 172                     'info_dict': {
 173                         'id': '1_3hqew8kn',
 174                         'ext': 'mp4',
 175                         'title': 'CS7646_00-00 Introductio_Prerequisites',
 176                         'duration': 49,
 177                         'thumbnail': 'http://cfvod.kaltura.com/p/2019031/sp/201903100/thumbnail/entry_id/1_3hqew8kn/version/100001',
 178                         'view_count': int,
 179                         'timestamp': 1533154536,
 180                         'upload_date': '20180801',
 181                         'uploader_id': 'djoyner3',
 182                     },
 183                 },
 184             ],
 185         },
 186     ]
 187
 188     @classmethod
 189     def _extract_embed_urls(cls, url, webpage):
 190         # Embed codes: https://knowledge.kaltura.com/embedding-kaltura-media-players-your-site
 191         finditer = (
 192             list(re.finditer(
 193                 r'''(?xs)
 194                     kWidget\.(?:thumb)?[Ee]mbed\(
 195                     \{.*?
 196                         (?P<q1>['"])wid(?P=q1)\s*:\s*
 197                         (?P<q2>['"])_?(?P<partner_id>(?:(?!(?P=q2)).)+)(?P=q2),.*?
 198                         (?P<q3>['"])entry_?[Ii]d(?P=q3)\s*:\s*
 199                         (?P<q4>['"])(?P<id>(?:(?!(?P=q4)).)+)(?P=q4)(?:,|\s*\})
 200                 ''', webpage))
 201             or list(re.finditer(
 202                 r'''(?xs)
 203                     (?P<q1>["'])
 204                         (?:https?:)?//cdnapi(?:sec)?\.kaltura\.com(?::\d+)?/(?:(?!(?P=q1)).)*\b(?:p|partner_id)/(?P<partner_id>\d+)(?:(?!(?P=q1)).)*
 205                     (?P=q1).*?
 206                     (?:
 207                         (?:
 208                             entry_?[Ii]d|
 209                             (?P<q2>["'])entry_?[Ii]d(?P=q2)
 210                         )\s*:\s*|
 211                         \[\s*(?P<q2_1>["'])entry_?[Ii]d(?P=q2_1)\s*\]\s*=\s*
 212                     )
 213                     (?P<q3>["'])(?P<id>(?:(?!(?P=q3)).)+)(?P=q3)
 214                 ''', webpage))
 215             or list(re.finditer(
 216                 r'''(?xs)
 217                     <(?:iframe[^>]+src|meta[^>]+\bcontent)=(?P<q1>["'])\s*
 218                       (?:https?:)?//(?:(?:www|cdnapi(?:sec)?)\.)?kaltura\.com/(?:(?!(?P=q1)).)*\b(?:p|partner_id)/(?P<partner_id>\d+)
 219                       (?:(?!(?P=q1)).)*
 220                       [?&;]entry_id=(?P<id>(?:(?!(?P=q1))[^&])+)
 221                       (?:(?!(?P=q1)).)*
 222                     (?P=q1)
 223                 ''', webpage))
 224         )
 225         urls = []
 226         for mobj in finditer:
 227             embed_info = mobj.groupdict()
 228             for k, v in embed_info.items():
 229                 if v:
 230                     embed_info[k] = v.strip()
 231             embed_url = 'kaltura:{partner_id}:{id}'.format(**embed_info)
 232             escaped_pid = re.escape(embed_info['partner_id'])
 233             service_mobj = re.search(
 234                 rf'<script[^>]+src=(["\'])(?P<id>(?:https?:)?//(?:(?!\1).)+)/p/{escaped_pid}/sp/{escaped_pid}00/embedIframeJs',
 235                 webpage)
 236             if service_mobj:
 237                 embed_url = smuggle_url(embed_url, {'service_url': service_mobj.group('id')})
 238             urls.append(embed_url)
 239         return urls
 240
 241     def _kaltura_api_call(self, video_id, actions, service_url=None, **kwargs):
 242         params = actions[0]
 243         params.update(dict(enumerate(actions[1:], start=1)))
 244
 245         data = self._download_json(
 246             (service_url or self._SERVICE_URL) + self._SERVICE_BASE,
 247             video_id, data=json.dumps(params).encode(),
 248             headers={
 249                 'Content-Type': 'application/json',
 250                 'Accept-Encoding': 'gzip, deflate, br',
 251             }, **kwargs)
 252
 253         for idx, status in enumerate(data):
 254             if not isinstance(status, dict):
 255                 continue
 256             if status.get('objectType') == 'KalturaAPIException':
 257                 raise ExtractorError(
 258                     '{} said: {} ({})'.format(self.IE_NAME, status['message'], idx))
 259
 260         data[1] = traverse_obj(data, (1, 'objects', 0))
 261
 262         return data
 263
 264     def _get_video_info(self, video_id, partner_id, service_url=None, player_type='html5'):
 265         assert player_type in ('html5', 'kwidget')
 266         if player_type == 'kwidget':
 267             return self._get_video_info_kwidget(video_id, partner_id, service_url)
 268
 269         return self._get_video_info_html5(video_id, partner_id, service_url)
 270
 271     def _get_video_info_html5(self, video_id, partner_id, service_url=None):
 272         actions = [
 273             {
 274                 'apiVersion': '3.3.0',
 275                 'clientTag': 'html5:v3.1.0',
 276                 'format': 1,  # JSON, 2 = XML, 3 = PHP
 277                 'ks': '',
 278                 'partnerId': partner_id,
 279             },
 280             {
 281                 'expiry': 86400,
 282                 'service': 'session',
 283                 'action': 'startWidgetSession',
 284                 'widgetId': self._build_widget_id(partner_id),
 285             },
 286             # info
 287             {
 288                 'action': 'list',
 289                 'filter': {'redirectFromEntryId': video_id},
 290                 'service': 'baseentry',
 291                 'ks': '{1:result:ks}',
 292                 'responseProfile': {
 293                     'type': 1,
 294                     'fields': 'createdAt,dataUrl,duration,name,plays,thumbnailUrl,userId',
 295                 },
 296             },
 297             # flavor_assets
 298             {
 299                 'action': 'getbyentryid',
 300                 'entryId': video_id,
 301                 'service': 'flavorAsset',
 302                 'ks': '{1:result:ks}',
 303             },
 304             # captions
 305             {
 306                 'action': 'list',
 307                 'filter:entryIdEqual': video_id,
 308                 'service': 'caption_captionasset',
 309                 'ks': '{1:result:ks}',
 310             },
 311         ]
 312         return self._kaltura_api_call(
 313             video_id, actions, service_url, note='Downloading video info JSON (Kaltura html5 player)')
 314
 315     def _get_video_info_kwidget(self, video_id, partner_id, service_url=None):
 316         actions = [
 317             {
 318                 'service': 'multirequest',
 319                 'apiVersion': '3.1',
 320                 'expiry': 86400,
 321                 'clientTag': 'kwidget:v2.89',
 322                 'format': 1,  # JSON, 2 = XML, 3 = PHP
 323                 'ignoreNull': 1,
 324                 'action': 'null',
 325             },
 326             # header
 327             {
 328                 'expiry': 86400,
 329                 'service': 'session',
 330                 'action': 'startWidgetSession',
 331                 'widgetId': self._build_widget_id(partner_id),
 332             },
 333             # (empty)
 334             {
 335                 'expiry': 86400,
 336                 'service': 'session',
 337                 'action': 'startwidgetsession',
 338                 'widgetId': self._build_widget_id(partner_id),
 339                 'format': 9,
 340                 'apiVersion': '3.1',
 341                 'clientTag': 'kwidget:v2.89',
 342                 'ignoreNull': 1,
 343                 'ks': '{1:result:ks}',
 344             },
 345             # info
 346             {
 347                 'action': 'list',
 348                 'filter': {'redirectFromEntryId': video_id},
 349                 'service': 'baseentry',
 350                 'ks': '{1:result:ks}',
 351                 'responseProfile': {
 352                     'type': 1,
 353                     'fields': 'createdAt,dataUrl,duration,name,plays,thumbnailUrl,userId',
 354                 },
 355             },
 356             # flavor_assets
 357             {
 358                 'action': 'getbyentryid',
 359                 'entryId': video_id,
 360                 'service': 'flavorAsset',
 361                 'ks': '{1:result:ks}',
 362             },
 363             # captions
 364             {
 365                 'action': 'list',
 366                 'filter:entryIdEqual': video_id,
 367                 'service': 'caption_captionasset',
 368                 'ks': '{1:result:ks}',
 369             },
 370         ]
 371         # second object (representing the second start widget session) is None
 372         header, _, _info, flavor_assets, captions = self._kaltura_api_call(
 373             video_id, actions, service_url, note='Downloading video info JSON (Kaltura kwidget player)')
 374         info = _info['objects'][0]
 375         return header, info, flavor_assets, captions
 376
 377     def _build_widget_id(self, partner_id):
 378         return partner_id if '_' in partner_id else f'_{partner_id}'
 379
 380     IFRAME_PACKAGE_DATA_REGEX = r'window\.kalturaIframePackageData\s*='
 381
 382     def _real_extract(self, url):
 383         url, smuggled_data = unsmuggle_url(url, {})
 384
 385         mobj = self._match_valid_url(url)
 386         partner_id, entry_id, player_type = mobj.group('partner_id', 'id', 'player_type')
 387         ks, captions = None, None
 388         if not player_type:
 389             player_type = 'kwidget' if 'html5lib/v2' in url else 'html5'
 390         if partner_id and entry_id:
 391             _, info, flavor_assets, captions = self._get_video_info(entry_id, partner_id, smuggled_data.get('service_url'), player_type=player_type)
 392         else:
 393             path, query = mobj.group('path', 'query')
 394             if not path and not query:
 395                 raise ExtractorError('Invalid URL', expected=True)
 396             params = {}
 397             if query:
 398                 params = urllib.parse.parse_qs(query)
 399             if path:
 400                 splitted_path = path.split('/')
 401                 params.update(dict(zip(splitted_path[::2], [[v] for v in splitted_path[1::2]])))
 402             if 'wid' in params:
 403                 partner_id = remove_start(params['wid'][0], '_')
 404             elif 'p' in params:
 405                 partner_id = params['p'][0]
 406             elif 'partner_id' in params:
 407                 partner_id = params['partner_id'][0]
 408             else:
 409                 raise ExtractorError('Invalid URL', expected=True)
 410             if 'entry_id' in params:
 411                 entry_id = params['entry_id'][0]
 412                 _, info, flavor_assets, captions = self._get_video_info(entry_id, partner_id, player_type=player_type)
 413             elif 'uiconf_id' in params and 'flashvars[referenceId]' in params:
 414                 reference_id = params['flashvars[referenceId]'][0]
 415                 webpage = self._download_webpage(url, reference_id)
 416                 entry_data = self._search_json(
 417                     self.IFRAME_PACKAGE_DATA_REGEX, webpage,
 418                     'kalturaIframePackageData', reference_id)['entryResult']
 419                 info, flavor_assets = entry_data['meta'], entry_data['contextData']['flavorAssets']
 420                 entry_id = info['id']
 421                 # Unfortunately, data returned in kalturaIframePackageData lacks
 422                 # captions so we will try requesting the complete data using
 423                 # regular approach since we now know the entry_id
 424                 # Even if this fails we already have everything extracted
 425                 # apart from captions and can process at least with this
 426                 with contextlib.suppress(ExtractorError):
 427                     _, info, flavor_assets, captions = self._get_video_info(
 428                         entry_id, partner_id, player_type=player_type)
 429             elif 'uiconf_id' in params and 'flashvars[playlistAPI.kpl0Id]' in params:
 430                 playlist_id = params['flashvars[playlistAPI.kpl0Id]'][0]
 431                 webpage = self._download_webpage(url, playlist_id)
 432                 playlist_data = self._search_json(
 433                     self.IFRAME_PACKAGE_DATA_REGEX, webpage,
 434                     'kalturaIframePackageData', playlist_id)['playlistResult']
 435                 return self.playlist_from_matches(
 436                     traverse_obj(playlist_data, (playlist_id, 'items', ..., 'id')),
 437                     playlist_id, traverse_obj(playlist_data, (playlist_id, 'name')),
 438                     ie=KalturaIE, getter=lambda x: f'kaltura:{partner_id}:{x}:{player_type}')
 439             else:
 440                 raise ExtractorError('Invalid URL', expected=True)
 441             ks = params.get('flashvars[ks]', [None])[0]
 442
 443         return self._per_video_extract(smuggled_data, entry_id, info, ks, flavor_assets, captions)
 444
 445     def _per_video_extract(self, smuggled_data, entry_id, info, ks, flavor_assets, captions):
 446         source_url = smuggled_data.get('source_url')
 447         if source_url:
 448             referrer = base64.b64encode(
 449                 '://'.join(urllib.parse.urlparse(source_url)[:2])
 450                 .encode()).decode('utf-8')
 451         else:
 452             referrer = None
 453
 454         def sign_url(unsigned_url):
 455             if ks:
 456                 unsigned_url += f'/ks/{ks}'
 457             if referrer:
 458                 unsigned_url += f'?referrer={referrer}'
 459             return unsigned_url
 460
 461         data_url = info['dataUrl']
 462         if '/flvclipper/' in data_url:
 463             data_url = re.sub(r'/flvclipper/.*', '/serveFlavor', data_url)
 464
 465         formats = []
 466         subtitles = {}
 467         for f in flavor_assets:
 468             # Continue if asset is not ready
 469             if f.get('status') != 2:
 470                 continue
 471             # Original format that's not available (e.g. kaltura:1926081:0_c03e1b5g)
 472             # skip for now.
 473             if f.get('fileExt') == 'chun':
 474                 continue
 475             # DRM-protected video, cannot be decrypted
 476             if not self.get_param('allow_unplayable_formats') and f.get('fileExt') == 'wvm':
 477                 continue
 478             if not f.get('fileExt'):
 479                 # QT indicates QuickTime; some videos have broken fileExt
 480                 if f.get('containerFormat') == 'qt':
 481                     f['fileExt'] = 'mov'
 482                 else:
 483                     f['fileExt'] = 'mp4'
 484             video_url = sign_url(
 485                 '{}/flavorId/{}'.format(data_url, f['id']))
 486             format_id = '{fileExt}-{bitrate}'.format(**f)
 487             # Source format may not be available (e.g. kaltura:513551:1_66x4rg7o)
 488             if f.get('isOriginal') is True and not self._is_valid_url(
 489                     video_url, entry_id, format_id):
 490                 continue
 491             # audio-only has no videoCodecId (e.g. kaltura:1926081:0_c03e1b5g
 492             # -f mp4-56)
 493             vcodec = 'none' if 'videoCodecId' not in f and f.get(
 494                 'frameRate') == 0 else f.get('videoCodecId')
 495             formats.append({
 496                 'format_id': format_id,
 497                 'ext': f.get('fileExt'),
 498                 'tbr': int_or_none(f['bitrate']),
 499                 'fps': int_or_none(f.get('frameRate')),
 500                 'filesize_approx': int_or_none(f.get('size'), invscale=1024),
 501                 'container': f.get('containerFormat'),
 502                 'vcodec': vcodec,
 503                 'height': int_or_none(f.get('height')),
 504                 'width': int_or_none(f.get('width')),
 505                 'url': video_url,
 506             })
 507         if '/playManifest/' in data_url:
 508             m3u8_url = sign_url(data_url.replace(
 509                 'format/url', 'format/applehttp'))
 510             fmts, subs = self._extract_m3u8_formats_and_subtitles(
 511                 m3u8_url, entry_id, 'mp4', 'm3u8_native',
 512                 m3u8_id='hls', fatal=False)
 513             formats.extend(fmts)
 514             self._merge_subtitles(subs, target=subtitles)
 515
 516         if captions:
 517             for caption in captions.get('objects', []):
 518                 # Continue if caption is not ready
 519                 if caption.get('status') != 2:
 520                     continue
 521                 if not caption.get('id'):
 522                     continue
 523                 caption_format = int_or_none(caption.get('format'))
 524                 subtitles.setdefault(caption.get('languageCode') or caption.get('language'), []).append({
 525                     'url': '{}/api_v3/service/caption_captionasset/action/serve/captionAssetId/{}'.format(self._SERVICE_URL, caption['id']),
 526                     'ext': caption.get('fileExt') or self._CAPTION_TYPES.get(caption_format) or 'ttml',
 527                 })
 528
 529         return {
 530             'id': entry_id,
 531             'title': info['name'],
 532             'formats': formats,
 533             'subtitles': subtitles,
 534             'description': clean_html(info.get('description')),
 535             'thumbnail': info.get('thumbnailUrl'),
 536             'duration': info.get('duration'),
 537             'timestamp': info.get('createdAt'),
 538             'uploader_id': format_field(info, 'userId', ignore=('None', None)),
 539             'view_count': int_or_none(info.get('plays')),
 540         }