yt_dlp/extractor/kaltura.py

   1 import base64
   2 import json
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import (
   7     compat_urlparse,
   8     compat_parse_qs,
   9 )
  10 from ..utils import (
  11     clean_html,
  12     ExtractorError,
  13     format_field,
  14     int_or_none,
  15     unsmuggle_url,
  16     smuggle_url,
  17     traverse_obj,
  18     remove_start
  19 )
  20
  21
  22 class KalturaIE(InfoExtractor):
  23     _VALID_URL = r'''(?x)
  24                 (?:
  25                     kaltura:(?P<partner_id>\w+):(?P<id>\w+)(?::(?P<player_type>\w+))?|
  26                     https?://
  27                         (:?(?:www|cdnapi(?:sec)?)\.)?kaltura\.com(?::\d+)?/
  28                         (?:
  29                             (?:
  30                                 # flash player
  31                                 index\.php/(?:kwidget|extwidget/preview)|
  32                                 # html5 player
  33                                 html5/html5lib/[^/]+/mwEmbedFrame\.php
  34                             )
  35                         )(?:/(?P<path>[^?]+))?(?:\?(?P<query>.*))?
  36                 )
  37                 '''
  38     _SERVICE_URL = 'http://cdnapi.kaltura.com'
  39     _SERVICE_BASE = '/api_v3/service/multirequest'
  40     # See https://github.com/kaltura/server/blob/master/plugins/content/caption/base/lib/model/enums/CaptionType.php
  41     _CAPTION_TYPES = {
  42         1: 'srt',
  43         2: 'ttml',
  44         3: 'vtt',
  45     }
  46     _TESTS = [
  47         {
  48             'url': 'kaltura:269692:1_1jc2y3e4',
  49             'md5': '3adcbdb3dcc02d647539e53f284ba171',
  50             'info_dict': {
  51                 'id': '1_1jc2y3e4',
  52                 'ext': 'mp4',
  53                 'title': 'Straight from the Heart',
  54                 'upload_date': '20131219',
  55                 'uploader_id': 'mlundberg@wolfgangsvault.com',
  56                 'description': 'The Allman Brothers Band, 12/16/1981',
  57                 'thumbnail': 're:^https?://.*/thumbnail/.*',
  58                 'timestamp': int,
  59             },
  60             'skip': 'The access to this service is forbidden since the specified partner is blocked'
  61         },
  62         {
  63             'url': 'http://www.kaltura.com/index.php/kwidget/cache_st/1300318621/wid/_269692/uiconf_id/3873291/entry_id/1_1jc2y3e4',
  64             'only_matching': True,
  65         },
  66         {
  67             'url': 'https://cdnapisec.kaltura.com/index.php/kwidget/wid/_557781/uiconf_id/22845202/entry_id/1_plr1syf3',
  68             'only_matching': True,
  69         },
  70         {
  71             'url': 'https://cdnapisec.kaltura.com/html5/html5lib/v2.30.2/mwEmbedFrame.php/p/1337/uiconf_id/20540612/entry_id/1_sf5ovm7u?wid=_243342',
  72             'only_matching': True,
  73         },
  74         {
  75             # video with subtitles
  76             'url': 'kaltura:111032:1_cw786r8q',
  77             'only_matching': True,
  78         },
  79         {
  80             # video with ttml subtitles (no fileExt)
  81             'url': 'kaltura:1926081:0_l5ye1133',
  82             'info_dict': {
  83                 'id': '0_l5ye1133',
  84                 'ext': 'mp4',
  85                 'title': 'What Can You Do With Python?',
  86                 'upload_date': '20160221',
  87                 'uploader_id': 'stork',
  88                 'thumbnail': 're:^https?://.*/thumbnail/.*',
  89                 'timestamp': int,
  90                 'subtitles': {
  91                     'en': [{
  92                         'ext': 'ttml',
  93                     }],
  94                 },
  95             },
  96             'skip': 'Gone. Maybe https://www.safaribooksonline.com/library/tutorials/introduction-to-python-anon/3469/',
  97             'params': {
  98                 'skip_download': True,
  99             },
 100         },
 101         {
 102             'url': 'https://www.kaltura.com/index.php/extwidget/preview/partner_id/1770401/uiconf_id/37307382/entry_id/0_58u8kme7/embed/iframe?&flashvars[streamerType]=auto',
 103             'only_matching': True,
 104         },
 105         {
 106             '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',
 107             'only_matching': True,
 108         },
 109         {
 110             # unavailable source format
 111             'url': 'kaltura:513551:1_66x4rg7o',
 112             'only_matching': True,
 113         },
 114         {
 115             # html5lib URL using kwidget player
 116             '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',
 117             'info_dict': {
 118                 'id': '0_c076mna6',
 119                 'ext': 'mp4',
 120                 'title': 'md5:4883e7acbcbf42583a2dddc97dee4855',
 121                 'duration': 3608,
 122                 'uploader_id': 'commons@swinburne.edu.au',
 123                 'timestamp': 1408086874,
 124                 'view_count': int,
 125                 'upload_date': '20140815',
 126                 'thumbnail': 'http://cfvod.kaltura.com/p/691292/sp/69129200/thumbnail/entry_id/0_c076mna6/version/100022',
 127             }
 128         },
 129         {
 130             # html5lib playlist URL using kwidget player
 131             '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',
 132             'info_dict': {
 133                 'id': '1_jovey5nu',
 134                 'title': '00-00 Introduction'
 135             },
 136             'playlist': [
 137                 {
 138                     'info_dict': {
 139                         'id': '1_b1y5hlvx',
 140                         'ext': 'mp4',
 141                         'title': 'CS7646_00-00 Introductio_Introduction',
 142                         'duration': 91,
 143                         'thumbnail': 'http://cfvod.kaltura.com/p/2019031/sp/201903100/thumbnail/entry_id/1_b1y5hlvx/version/100001',
 144                         'view_count': int,
 145                         'timestamp': 1533154447,
 146                         'upload_date': '20180801',
 147                         'uploader_id': 'djoyner3',
 148                     }
 149                 }, {
 150                     'info_dict': {
 151                         'id': '1_jfb7mdpn',
 152                         'ext': 'mp4',
 153                         'title': 'CS7646_00-00 Introductio_Three parts to the course',
 154                         'duration': 63,
 155                         'thumbnail': 'http://cfvod.kaltura.com/p/2019031/sp/201903100/thumbnail/entry_id/1_jfb7mdpn/version/100001',
 156                         'view_count': int,
 157                         'timestamp': 1533154489,
 158                         'upload_date': '20180801',
 159                         'uploader_id': 'djoyner3',
 160                     }
 161                 }, {
 162                     'info_dict': {
 163                         'id': '1_8xflxdp7',
 164                         'ext': 'mp4',
 165                         'title': 'CS7646_00-00 Introductio_Textbooks',
 166                         'duration': 37,
 167                         'thumbnail': 'http://cfvod.kaltura.com/p/2019031/sp/201903100/thumbnail/entry_id/1_8xflxdp7/version/100001',
 168                         'view_count': int,
 169                         'timestamp': 1533154512,
 170                         'upload_date': '20180801',
 171                         'uploader_id': 'djoyner3',
 172                     }
 173                 }, {
 174                     'info_dict': {
 175                         'id': '1_3hqew8kn',
 176                         'ext': 'mp4',
 177                         'title': 'CS7646_00-00 Introductio_Prerequisites',
 178                         'duration': 49,
 179                         'thumbnail': 'http://cfvod.kaltura.com/p/2019031/sp/201903100/thumbnail/entry_id/1_3hqew8kn/version/100001',
 180                         'view_count': int,
 181                         'timestamp': 1533154536,
 182                         'upload_date': '20180801',
 183                         'uploader_id': 'djoyner3',
 184                     }
 185                 }
 186             ]
 187         }
 188     ]
 189
 190     @classmethod
 191     def _extract_embed_urls(cls, url, webpage):
 192         # Embed codes: https://knowledge.kaltura.com/embedding-kaltura-media-players-your-site
 193         finditer = (
 194             list(re.finditer(
 195                 r"""(?xs)
 196                     kWidget\.(?:thumb)?[Ee]mbed\(
 197                     \{.*?
 198                         (?P<q1>['"])wid(?P=q1)\s*:\s*
 199                         (?P<q2>['"])_?(?P<partner_id>(?:(?!(?P=q2)).)+)(?P=q2),.*?
 200                         (?P<q3>['"])entry_?[Ii]d(?P=q3)\s*:\s*
 201                         (?P<q4>['"])(?P<id>(?:(?!(?P=q4)).)+)(?P=q4)(?:,|\s*\})
 202                 """, webpage))
 203             or list(re.finditer(
 204                 r'''(?xs)
 205                     (?P<q1>["'])
 206                         (?:https?:)?//cdnapi(?:sec)?\.kaltura\.com(?::\d+)?/(?:(?!(?P=q1)).)*\b(?:p|partner_id)/(?P<partner_id>\d+)(?:(?!(?P=q1)).)*
 207                     (?P=q1).*?
 208                     (?:
 209                         (?:
 210                             entry_?[Ii]d|
 211                             (?P<q2>["'])entry_?[Ii]d(?P=q2)
 212                         )\s*:\s*|
 213                         \[\s*(?P<q2_1>["'])entry_?[Ii]d(?P=q2_1)\s*\]\s*=\s*
 214                     )
 215                     (?P<q3>["'])(?P<id>(?:(?!(?P=q3)).)+)(?P=q3)
 216                 ''', webpage))
 217             or list(re.finditer(
 218                 r'''(?xs)
 219                     <(?:iframe[^>]+src|meta[^>]+\bcontent)=(?P<q1>["'])\s*
 220                       (?:https?:)?//(?:(?:www|cdnapi(?:sec)?)\.)?kaltura\.com/(?:(?!(?P=q1)).)*\b(?:p|partner_id)/(?P<partner_id>\d+)
 221                       (?:(?!(?P=q1)).)*
 222                       [?&;]entry_id=(?P<id>(?:(?!(?P=q1))[^&])+)
 223                       (?:(?!(?P=q1)).)*
 224                     (?P=q1)
 225                 ''', webpage))
 226         )
 227         urls = []
 228         for mobj in finditer:
 229             embed_info = mobj.groupdict()
 230             for k, v in embed_info.items():
 231                 if v:
 232                     embed_info[k] = v.strip()
 233             embed_url = 'kaltura:%(partner_id)s:%(id)s' % embed_info
 234             escaped_pid = re.escape(embed_info['partner_id'])
 235             service_mobj = re.search(
 236                 r'<script[^>]+src=(["\'])(?P<id>(?:https?:)?//(?:(?!\1).)+)/p/%s/sp/%s00/embedIframeJs' % (escaped_pid, escaped_pid),
 237                 webpage)
 238             if service_mobj:
 239                 embed_url = smuggle_url(embed_url, {'service_url': service_mobj.group('id')})
 240             urls.append(embed_url)
 241         return urls
 242
 243     def _kaltura_api_call(self, video_id, actions, service_url=None, *args, **kwargs):
 244         params = actions[0]
 245         params.update({i: a for i, a in enumerate(actions[1:], start=1)})
 246
 247         data = self._download_json(
 248             (service_url or self._SERVICE_URL) + self._SERVICE_BASE,
 249             video_id, data=json.dumps(params).encode('utf-8'),
 250             headers={
 251                 'Content-Type': 'application/json',
 252                 'Accept-Encoding': 'gzip, deflate, br',
 253             }, *args, **kwargs)
 254
 255         for idx, status in enumerate(data):
 256             if not isinstance(status, dict):
 257                 continue
 258             if status.get('objectType') == 'KalturaAPIException':
 259                 raise ExtractorError(
 260                     '%s said: %s (%d)' % (self.IE_NAME, status['message'], idx))
 261
 262         data[1] = traverse_obj(data, (1, 'objects', 0))
 263
 264         return data
 265
 266     def _get_video_info(self, video_id, partner_id, service_url=None, player_type='html5'):
 267         assert player_type in ('html5', 'kwidget')
 268         if player_type == 'kwidget':
 269             return self._get_video_info_kwidget(video_id, partner_id, service_url)
 270
 271         return self._get_video_info_html5(video_id, partner_id, service_url)
 272
 273     def _get_video_info_html5(self, video_id, partner_id, service_url=None):
 274         actions = [
 275             {
 276                 'apiVersion': '3.3.0',
 277                 'clientTag': 'html5:v3.1.0',
 278                 'format': 1,  # JSON, 2 = XML, 3 = PHP
 279                 'ks': '',
 280                 'partnerId': partner_id,
 281             },
 282             {
 283                 'expiry': 86400,
 284                 'service': 'session',
 285                 'action': 'startWidgetSession',
 286                 'widgetId': self._build_widget_id(partner_id),
 287             },
 288             # info
 289             {
 290                 'action': 'list',
 291                 'filter': {'redirectFromEntryId': video_id},
 292                 'service': 'baseentry',
 293                 'ks': '{1:result:ks}',
 294                 'responseProfile': {
 295                     'type': 1,
 296                     'fields': 'createdAt,dataUrl,duration,name,plays,thumbnailUrl,userId',
 297                 },
 298             },
 299             # flavor_assets
 300             {
 301                 'action': 'getbyentryid',
 302                 'entryId': video_id,
 303                 'service': 'flavorAsset',
 304                 'ks': '{1:result:ks}',
 305             },
 306             # captions
 307             {
 308                 'action': 'list',
 309                 'filter:entryIdEqual': video_id,
 310                 'service': 'caption_captionasset',
 311                 'ks': '{1:result:ks}',
 312             },
 313         ]
 314         return self._kaltura_api_call(
 315             video_id, actions, service_url, note='Downloading video info JSON (Kaltura html5 player)')
 316
 317     def _get_video_info_kwidget(self, video_id, partner_id, service_url=None):
 318         actions = [
 319             {
 320                 'service': 'multirequest',
 321                 'apiVersion': '3.1',
 322                 'expiry': 86400,
 323                 'clientTag': 'kwidget:v2.89',
 324                 'format': 1,  # JSON, 2 = XML, 3 = PHP
 325                 'ignoreNull': 1,
 326                 'action': 'null',
 327             },
 328             # header
 329             {
 330                 'expiry': 86400,
 331                 'service': 'session',
 332                 'action': 'startWidgetSession',
 333                 'widgetId': self._build_widget_id(partner_id),
 334             },
 335             # (empty)
 336             {
 337                 'expiry': 86400,
 338                 'service': 'session',
 339                 'action': 'startwidgetsession',
 340                 'widgetId': self._build_widget_id(partner_id),
 341                 'format': 9,
 342                 'apiVersion': '3.1',
 343                 'clientTag': 'kwidget:v2.89',
 344                 'ignoreNull': 1,
 345                 'ks': '{1:result:ks}'
 346             },
 347             # info
 348             {
 349                 'action': 'list',
 350                 'filter': {'redirectFromEntryId': video_id},
 351                 'service': 'baseentry',
 352                 'ks': '{1:result:ks}',
 353                 'responseProfile': {
 354                     'type': 1,
 355                     'fields': 'createdAt,dataUrl,duration,name,plays,thumbnailUrl,userId',
 356                 },
 357             },
 358             # flavor_assets
 359             {
 360                 'action': 'getbyentryid',
 361                 'entryId': video_id,
 362                 'service': 'flavorAsset',
 363                 'ks': '{1:result:ks}',
 364             },
 365             # captions
 366             {
 367                 'action': 'list',
 368                 'filter:entryIdEqual': video_id,
 369                 'service': 'caption_captionasset',
 370                 'ks': '{1:result:ks}',
 371             },
 372         ]
 373         # second object (representing the second start widget session) is None
 374         header, _, _info, flavor_assets, captions = self._kaltura_api_call(
 375             video_id, actions, service_url, note='Downloading video info JSON (Kaltura kwidget player)')
 376         info = _info['objects'][0]
 377         return header, info, flavor_assets, captions
 378
 379     def _build_widget_id(self, partner_id):
 380         return partner_id if '_' in partner_id else f'_{partner_id}'
 381
 382     IFRAME_PACKAGE_DATA_REGEX = r'window\.kalturaIframePackageData\s*='
 383
 384     def _real_extract(self, url):
 385         url, smuggled_data = unsmuggle_url(url, {})
 386
 387         mobj = self._match_valid_url(url)
 388         partner_id, entry_id, player_type = mobj.group('partner_id', 'id', 'player_type')
 389         ks, captions = None, None
 390         if not player_type:
 391             player_type = 'kwidget' if 'html5lib/v2' in url else 'html5'
 392         if partner_id and entry_id:
 393             _, info, flavor_assets, captions = self._get_video_info(entry_id, partner_id, smuggled_data.get('service_url'), player_type=player_type)
 394         else:
 395             path, query = mobj.group('path', 'query')
 396             if not path and not query:
 397                 raise ExtractorError('Invalid URL', expected=True)
 398             params = {}
 399             if query:
 400                 params = compat_parse_qs(query)
 401             if path:
 402                 splitted_path = path.split('/')
 403                 params.update(dict((zip(splitted_path[::2], [[v] for v in splitted_path[1::2]]))))
 404             if 'wid' in params:
 405                 partner_id = remove_start(params['wid'][0], '_')
 406             elif 'p' in params:
 407                 partner_id = params['p'][0]
 408             elif 'partner_id' in params:
 409                 partner_id = params['partner_id'][0]
 410             else:
 411                 raise ExtractorError('Invalid URL', expected=True)
 412             if 'entry_id' in params:
 413                 entry_id = params['entry_id'][0]
 414                 _, info, flavor_assets, captions = self._get_video_info(entry_id, partner_id, player_type=player_type)
 415             elif 'uiconf_id' in params and 'flashvars[referenceId]' in params:
 416                 reference_id = params['flashvars[referenceId]'][0]
 417                 webpage = self._download_webpage(url, reference_id)
 418                 entry_data = self._search_json(
 419                     self.IFRAME_PACKAGE_DATA_REGEX, webpage,
 420                     'kalturaIframePackageData', reference_id)['entryResult']
 421                 info, flavor_assets = entry_data['meta'], entry_data['contextData']['flavorAssets']
 422                 entry_id = info['id']
 423                 # Unfortunately, data returned in kalturaIframePackageData lacks
 424                 # captions so we will try requesting the complete data using
 425                 # regular approach since we now know the entry_id
 426                 try:
 427                     _, info, flavor_assets, captions = self._get_video_info(
 428                         entry_id, partner_id, player_type=player_type)
 429                 except ExtractorError:
 430                     # Regular scenario failed but we already have everything
 431                     # extracted apart from captions and can process at least
 432                     # with this
 433                     pass
 434             elif 'uiconf_id' in params and 'flashvars[playlistAPI.kpl0Id]' in params:
 435                 playlist_id = params['flashvars[playlistAPI.kpl0Id]'][0]
 436                 webpage = self._download_webpage(url, playlist_id)
 437                 playlist_data = self._search_json(
 438                     self.IFRAME_PACKAGE_DATA_REGEX, webpage,
 439                     'kalturaIframePackageData', playlist_id)['playlistResult']
 440                 return self.playlist_from_matches(
 441                     traverse_obj(playlist_data, (playlist_id, 'items', ..., 'id')),
 442                     playlist_id, traverse_obj(playlist_data, (playlist_id, 'name')),
 443                     ie=KalturaIE, getter=lambda x: f'kaltura:{partner_id}:{x}:{player_type}')
 444             else:
 445                 raise ExtractorError('Invalid URL', expected=True)
 446             ks = params.get('flashvars[ks]', [None])[0]
 447
 448         return self._per_video_extract(smuggled_data, entry_id, info, ks, flavor_assets, captions)
 449
 450     def _per_video_extract(self, smuggled_data, entry_id, info, ks, flavor_assets, captions):
 451         source_url = smuggled_data.get('source_url')
 452         if source_url:
 453             referrer = base64.b64encode(
 454                 '://'.join(compat_urlparse.urlparse(source_url)[:2])
 455                 .encode('utf-8')).decode('utf-8')
 456         else:
 457             referrer = None
 458
 459         def sign_url(unsigned_url):
 460             if ks:
 461                 unsigned_url += '/ks/%s' % ks
 462             if referrer:
 463                 unsigned_url += '?referrer=%s' % referrer
 464             return unsigned_url
 465
 466         data_url = info['dataUrl']
 467         if '/flvclipper/' in data_url:
 468             data_url = re.sub(r'/flvclipper/.*', '/serveFlavor', data_url)
 469
 470         formats = []
 471         subtitles = {}
 472         for f in flavor_assets:
 473             # Continue if asset is not ready
 474             if f.get('status') != 2:
 475                 continue
 476             # Original format that's not available (e.g. kaltura:1926081:0_c03e1b5g)
 477             # skip for now.
 478             if f.get('fileExt') == 'chun':
 479                 continue
 480             # DRM-protected video, cannot be decrypted
 481             if not self.get_param('allow_unplayable_formats') and f.get('fileExt') == 'wvm':
 482                 continue
 483             if not f.get('fileExt'):
 484                 # QT indicates QuickTime; some videos have broken fileExt
 485                 if f.get('containerFormat') == 'qt':
 486                     f['fileExt'] = 'mov'
 487                 else:
 488                     f['fileExt'] = 'mp4'
 489             video_url = sign_url(
 490                 '%s/flavorId/%s' % (data_url, f['id']))
 491             format_id = '%(fileExt)s-%(bitrate)s' % f
 492             # Source format may not be available (e.g. kaltura:513551:1_66x4rg7o)
 493             if f.get('isOriginal') is True and not self._is_valid_url(
 494                     video_url, entry_id, format_id):
 495                 continue
 496             # audio-only has no videoCodecId (e.g. kaltura:1926081:0_c03e1b5g
 497             # -f mp4-56)
 498             vcodec = 'none' if 'videoCodecId' not in f and f.get(
 499                 'frameRate') == 0 else f.get('videoCodecId')
 500             formats.append({
 501                 'format_id': format_id,
 502                 'ext': f.get('fileExt'),
 503                 'tbr': int_or_none(f['bitrate']),
 504                 'fps': int_or_none(f.get('frameRate')),
 505                 'filesize_approx': int_or_none(f.get('size'), invscale=1024),
 506                 'container': f.get('containerFormat'),
 507                 'vcodec': vcodec,
 508                 'height': int_or_none(f.get('height')),
 509                 'width': int_or_none(f.get('width')),
 510                 'url': video_url,
 511             })
 512         if '/playManifest/' in data_url:
 513             m3u8_url = sign_url(data_url.replace(
 514                 'format/url', 'format/applehttp'))
 515             fmts, subs = self._extract_m3u8_formats_and_subtitles(
 516                 m3u8_url, entry_id, 'mp4', 'm3u8_native',
 517                 m3u8_id='hls', fatal=False)
 518             formats.extend(fmts)
 519             self._merge_subtitles(subs, target=subtitles)
 520
 521         if captions:
 522             for caption in captions.get('objects', []):
 523                 # Continue if caption is not ready
 524                 if caption.get('status') != 2:
 525                     continue
 526                 if not caption.get('id'):
 527                     continue
 528                 caption_format = int_or_none(caption.get('format'))
 529                 subtitles.setdefault(caption.get('languageCode') or caption.get('language'), []).append({
 530                     'url': '%s/api_v3/service/caption_captionasset/action/serve/captionAssetId/%s' % (self._SERVICE_URL, caption['id']),
 531                     'ext': caption.get('fileExt') or self._CAPTION_TYPES.get(caption_format) or 'ttml',
 532                 })
 533
 534         return {
 535             'id': entry_id,
 536             'title': info['name'],
 537             'formats': formats,
 538             'subtitles': subtitles,
 539             'description': clean_html(info.get('description')),
 540             'thumbnail': info.get('thumbnailUrl'),
 541             'duration': info.get('duration'),
 542             'timestamp': info.get('createdAt'),
 543             'uploader_id': format_field(info, 'userId', ignore=('None', None)),
 544             'view_count': int_or_none(info.get('plays')),
 545         }