youtube_dl/extractor/vimeo.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import base64
   5 import json
   6 import re
   7 import itertools
   8
   9 from .common import InfoExtractor
  10 from ..compat import (
  11     compat_HTTPError,
  12     compat_str,
  13     compat_urlparse,
  14 )
  15 from ..utils import (
  16     determine_ext,
  17     ExtractorError,
  18     js_to_json,
  19     InAdvancePagedList,
  20     int_or_none,
  21     merge_dicts,
  22     NO_DEFAULT,
  23     parse_filesize,
  24     qualities,
  25     RegexNotFoundError,
  26     sanitized_Request,
  27     smuggle_url,
  28     std_headers,
  29     try_get,
  30     unified_timestamp,
  31     unsmuggle_url,
  32     urlencode_postdata,
  33     unescapeHTML,
  34 )
  35
  36
  37 class VimeoBaseInfoExtractor(InfoExtractor):
  38     _NETRC_MACHINE = 'vimeo'
  39     _LOGIN_REQUIRED = False
  40     _LOGIN_URL = 'https://vimeo.com/log_in'
  41
  42     def _login(self):
  43         username, password = self._get_login_info()
  44         if username is None:
  45             if self._LOGIN_REQUIRED:
  46                 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
  47             return
  48         webpage = self._download_webpage(
  49             self._LOGIN_URL, None, 'Downloading login page')
  50         token, vuid = self._extract_xsrft_and_vuid(webpage)
  51         data = {
  52             'action': 'login',
  53             'email': username,
  54             'password': password,
  55             'service': 'vimeo',
  56             'token': token,
  57         }
  58         self._set_vimeo_cookie('vuid', vuid)
  59         try:
  60             self._download_webpage(
  61                 self._LOGIN_URL, None, 'Logging in',
  62                 data=urlencode_postdata(data), headers={
  63                     'Content-Type': 'application/x-www-form-urlencoded',
  64                     'Referer': self._LOGIN_URL,
  65                 })
  66         except ExtractorError as e:
  67             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 418:
  68                 raise ExtractorError(
  69                     'Unable to log in: bad username or password',
  70                     expected=True)
  71             raise ExtractorError('Unable to log in')
  72
  73     def _verify_video_password(self, url, video_id, webpage):
  74         password = self._downloader.params.get('videopassword')
  75         if password is None:
  76             raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True)
  77         token, vuid = self._extract_xsrft_and_vuid(webpage)
  78         data = urlencode_postdata({
  79             'password': password,
  80             'token': token,
  81         })
  82         if url.startswith('http://'):
  83             # vimeo only supports https now, but the user can give an http url
  84             url = url.replace('http://', 'https://')
  85         password_request = sanitized_Request(url + '/password', data)
  86         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  87         password_request.add_header('Referer', url)
  88         self._set_vimeo_cookie('vuid', vuid)
  89         return self._download_webpage(
  90             password_request, video_id,
  91             'Verifying the password', 'Wrong password')
  92
  93     def _extract_xsrft_and_vuid(self, webpage):
  94         xsrft = self._search_regex(
  95             r'(?:(?P<q1>["\'])xsrft(?P=q1)\s*:|xsrft\s*[=:])\s*(?P<q>["\'])(?P<xsrft>.+?)(?P=q)',
  96             webpage, 'login token', group='xsrft')
  97         vuid = self._search_regex(
  98             r'["\']vuid["\']\s*:\s*(["\'])(?P<vuid>.+?)\1',
  99             webpage, 'vuid', group='vuid')
 100         return xsrft, vuid
 101
 102     def _set_vimeo_cookie(self, name, value):
 103         self._set_cookie('vimeo.com', name, value)
 104
 105     def _vimeo_sort_formats(self, formats):
 106         # Bitrates are completely broken. Single m3u8 may contain entries in kbps and bps
 107         # at the same time without actual units specified. This lead to wrong sorting.
 108         self._sort_formats(formats, field_preference=('preference', 'height', 'width', 'fps', 'tbr', 'format_id'))
 109
 110     def _parse_config(self, config, video_id):
 111         video_data = config['video']
 112         # Extract title
 113         video_title = video_data['title']
 114
 115         # Extract uploader, uploader_url and uploader_id
 116         video_uploader = video_data.get('owner', {}).get('name')
 117         video_uploader_url = video_data.get('owner', {}).get('url')
 118         video_uploader_id = video_uploader_url.split('/')[-1] if video_uploader_url else None
 119
 120         # Extract video thumbnail
 121         video_thumbnail = video_data.get('thumbnail')
 122         if video_thumbnail is None:
 123             video_thumbs = video_data.get('thumbs')
 124             if video_thumbs and isinstance(video_thumbs, dict):
 125                 _, video_thumbnail = sorted((int(width if width.isdigit() else 0), t_url) for (width, t_url) in video_thumbs.items())[-1]
 126
 127         # Extract video duration
 128         video_duration = int_or_none(video_data.get('duration'))
 129
 130         formats = []
 131         config_files = video_data.get('files') or config['request'].get('files', {})
 132         for f in config_files.get('progressive', []):
 133             video_url = f.get('url')
 134             if not video_url:
 135                 continue
 136             formats.append({
 137                 'url': video_url,
 138                 'format_id': 'http-%s' % f.get('quality'),
 139                 'width': int_or_none(f.get('width')),
 140                 'height': int_or_none(f.get('height')),
 141                 'fps': int_or_none(f.get('fps')),
 142                 'tbr': int_or_none(f.get('bitrate')),
 143             })
 144
 145         for files_type in ('hls', 'dash'):
 146             for cdn_name, cdn_data in config_files.get(files_type, {}).get('cdns', {}).items():
 147                 manifest_url = cdn_data.get('url')
 148                 if not manifest_url:
 149                     continue
 150                 format_id = '%s-%s' % (files_type, cdn_name)
 151                 if files_type == 'hls':
 152                     formats.extend(self._extract_m3u8_formats(
 153                         manifest_url, video_id, 'mp4',
 154                         'm3u8_native', m3u8_id=format_id,
 155                         note='Downloading %s m3u8 information' % cdn_name,
 156                         fatal=False))
 157                 elif files_type == 'dash':
 158                     mpd_pattern = r'/%s/(?:sep/)?video/' % video_id
 159                     mpd_manifest_urls = []
 160                     if re.search(mpd_pattern, manifest_url):
 161                         for suffix, repl in (('', 'video'), ('_sep', 'sep/video')):
 162                             mpd_manifest_urls.append((format_id + suffix, re.sub(
 163                                 mpd_pattern, '/%s/%s/' % (video_id, repl), manifest_url)))
 164                     else:
 165                         mpd_manifest_urls = [(format_id, manifest_url)]
 166                     for f_id, m_url in mpd_manifest_urls:
 167                         mpd_formats = self._extract_mpd_formats(
 168                             m_url.replace('/master.json', '/master.mpd'), video_id, f_id,
 169                             'Downloading %s MPD information' % cdn_name,
 170                             fatal=False)
 171                         for f in mpd_formats:
 172                             if f.get('vcodec') == 'none':
 173                                 f['preference'] = -50
 174                             elif f.get('acodec') == 'none':
 175                                 f['preference'] = -40
 176                         formats.extend(mpd_formats)
 177
 178         subtitles = {}
 179         text_tracks = config['request'].get('text_tracks')
 180         if text_tracks:
 181             for tt in text_tracks:
 182                 subtitles[tt['lang']] = [{
 183                     'ext': 'vtt',
 184                     'url': 'https://vimeo.com' + tt['url'],
 185                 }]
 186
 187         return {
 188             'title': video_title,
 189             'uploader': video_uploader,
 190             'uploader_id': video_uploader_id,
 191             'uploader_url': video_uploader_url,
 192             'thumbnail': video_thumbnail,
 193             'duration': video_duration,
 194             'formats': formats,
 195             'subtitles': subtitles,
 196         }
 197
 198
 199 class VimeoIE(VimeoBaseInfoExtractor):
 200     """Information extractor for vimeo.com."""
 201
 202     # _VALID_URL matches Vimeo URLs
 203     _VALID_URL = r'''(?x)
 204                     https?://
 205                         (?:
 206                             (?:
 207                                 www|
 208                                 (?P<player>player)
 209                             )
 210                             \.
 211                         )?
 212                         vimeo(?P<pro>pro)?\.com/
 213                         (?!(?:channels|album)/[^/?#]+/?(?:$|[?#])|[^/]+/review/|ondemand/)
 214                         (?:.*?/)?
 215                         (?:
 216                             (?:
 217                                 play_redirect_hls|
 218                                 moogaloop\.swf)\?clip_id=
 219                             )?
 220                         (?:videos?/)?
 221                         (?P<id>[0-9]+)
 222                         (?:/[\da-f]+)?
 223                         /?(?:[?&].*)?(?:[#].*)?$
 224                     '''
 225     IE_NAME = 'vimeo'
 226     _TESTS = [
 227         {
 228             'url': 'http://vimeo.com/56015672#at=0',
 229             'md5': '8879b6cc097e987f02484baf890129e5',
 230             'info_dict': {
 231                 'id': '56015672',
 232                 'ext': 'mp4',
 233                 'title': "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
 234                 'description': 'md5:509a9ad5c9bf97c60faee9203aca4479',
 235                 'timestamp': 1355990239,
 236                 'upload_date': '20121220',
 237                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user7108434',
 238                 'uploader_id': 'user7108434',
 239                 'uploader': 'Filippo Valsorda',
 240                 'duration': 10,
 241                 'license': 'by-sa',
 242             },
 243         },
 244         {
 245             'url': 'http://vimeopro.com/openstreetmapus/state-of-the-map-us-2013/video/68093876',
 246             'md5': '3b5ca6aa22b60dfeeadf50b72e44ed82',
 247             'note': 'Vimeo Pro video (#1197)',
 248             'info_dict': {
 249                 'id': '68093876',
 250                 'ext': 'mp4',
 251                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/openstreetmapus',
 252                 'uploader_id': 'openstreetmapus',
 253                 'uploader': 'OpenStreetMap US',
 254                 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
 255                 'description': 'md5:fd69a7b8d8c34a4e1d2ec2e4afd6ec30',
 256                 'duration': 1595,
 257             },
 258         },
 259         {
 260             'url': 'http://player.vimeo.com/video/54469442',
 261             'md5': '619b811a4417aa4abe78dc653becf511',
 262             'note': 'Videos that embed the url in the player page',
 263             'info_dict': {
 264                 'id': '54469442',
 265                 'ext': 'mp4',
 266                 'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software 2012',
 267                 'uploader': 'The BLN & Business of Software',
 268                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/theblnbusinessofsoftware',
 269                 'uploader_id': 'theblnbusinessofsoftware',
 270                 'duration': 3610,
 271                 'description': None,
 272             },
 273         },
 274         {
 275             'url': 'http://vimeo.com/68375962',
 276             'md5': 'aaf896bdb7ddd6476df50007a0ac0ae7',
 277             'note': 'Video protected with password',
 278             'info_dict': {
 279                 'id': '68375962',
 280                 'ext': 'mp4',
 281                 'title': 'youtube-dl password protected test video',
 282                 'timestamp': 1371200155,
 283                 'upload_date': '20130614',
 284                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user18948128',
 285                 'uploader_id': 'user18948128',
 286                 'uploader': 'Jaime Marquínez Ferrándiz',
 287                 'duration': 10,
 288                 'description': 'md5:dca3ea23adb29ee387127bc4ddfce63f',
 289             },
 290             'params': {
 291                 'videopassword': 'youtube-dl',
 292             },
 293         },
 294         {
 295             'url': 'http://vimeo.com/channels/keypeele/75629013',
 296             'md5': '2f86a05afe9d7abc0b9126d229bbe15d',
 297             'info_dict': {
 298                 'id': '75629013',
 299                 'ext': 'mp4',
 300                 'title': 'Key & Peele: Terrorist Interrogation',
 301                 'description': 'md5:8678b246399b070816b12313e8b4eb5c',
 302                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/atencio',
 303                 'uploader_id': 'atencio',
 304                 'uploader': 'Peter Atencio',
 305                 'channel_id': 'keypeele',
 306                 'channel_url': r're:https?://(?:www\.)?vimeo\.com/channels/keypeele',
 307                 'timestamp': 1380339469,
 308                 'upload_date': '20130928',
 309                 'duration': 187,
 310             },
 311             'expected_warnings': ['Unable to download JSON metadata'],
 312         },
 313         {
 314             'url': 'http://vimeo.com/76979871',
 315             'note': 'Video with subtitles',
 316             'info_dict': {
 317                 'id': '76979871',
 318                 'ext': 'mp4',
 319                 'title': 'The New Vimeo Player (You Know, For Videos)',
 320                 'description': 'md5:2ec900bf97c3f389378a96aee11260ea',
 321                 'timestamp': 1381846109,
 322                 'upload_date': '20131015',
 323                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/staff',
 324                 'uploader_id': 'staff',
 325                 'uploader': 'Vimeo Staff',
 326                 'duration': 62,
 327             }
 328         },
 329         {
 330             # from https://www.ouya.tv/game/Pier-Solar-and-the-Great-Architects/
 331             'url': 'https://player.vimeo.com/video/98044508',
 332             'note': 'The js code contains assignments to the same variable as the config',
 333             'info_dict': {
 334                 'id': '98044508',
 335                 'ext': 'mp4',
 336                 'title': 'Pier Solar OUYA Official Trailer',
 337                 'uploader': 'Tulio Gonçalves',
 338                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user28849593',
 339                 'uploader_id': 'user28849593',
 340             },
 341         },
 342         {
 343             # contains original format
 344             'url': 'https://vimeo.com/33951933',
 345             'md5': '53c688fa95a55bf4b7293d37a89c5c53',
 346             'info_dict': {
 347                 'id': '33951933',
 348                 'ext': 'mp4',
 349                 'title': 'FOX CLASSICS - Forever Classic ID - A Full Minute',
 350                 'uploader': 'The DMCI',
 351                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/dmci',
 352                 'uploader_id': 'dmci',
 353                 'timestamp': 1324343742,
 354                 'upload_date': '20111220',
 355                 'description': 'md5:ae23671e82d05415868f7ad1aec21147',
 356             },
 357         },
 358         {
 359             # only available via https://vimeo.com/channels/tributes/6213729 and
 360             # not via https://vimeo.com/6213729
 361             'url': 'https://vimeo.com/channels/tributes/6213729',
 362             'info_dict': {
 363                 'id': '6213729',
 364                 'ext': 'mp4',
 365                 'title': 'Vimeo Tribute: The Shining',
 366                 'uploader': 'Casey Donahue',
 367                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/caseydonahue',
 368                 'uploader_id': 'caseydonahue',
 369                 'channel_url': r're:https?://(?:www\.)?vimeo\.com/channels/tributes',
 370                 'channel_id': 'tributes',
 371                 'timestamp': 1250886430,
 372                 'upload_date': '20090821',
 373                 'description': 'md5:bdbf314014e58713e6e5b66eb252f4a6',
 374             },
 375             'params': {
 376                 'skip_download': True,
 377             },
 378             'expected_warnings': ['Unable to download JSON metadata'],
 379         },
 380         {
 381             # redirects to ondemand extractor and should be passed through it
 382             # for successful extraction
 383             'url': 'https://vimeo.com/73445910',
 384             'info_dict': {
 385                 'id': '73445910',
 386                 'ext': 'mp4',
 387                 'title': 'The Reluctant Revolutionary',
 388                 'uploader': '10Ft Films',
 389                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/tenfootfilms',
 390                 'uploader_id': 'tenfootfilms',
 391             },
 392             'params': {
 393                 'skip_download': True,
 394             },
 395         },
 396         {
 397             'url': 'http://player.vimeo.com/video/68375962',
 398             'md5': 'aaf896bdb7ddd6476df50007a0ac0ae7',
 399             'info_dict': {
 400                 'id': '68375962',
 401                 'ext': 'mp4',
 402                 'title': 'youtube-dl password protected test video',
 403                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user18948128',
 404                 'uploader_id': 'user18948128',
 405                 'uploader': 'Jaime Marquínez Ferrándiz',
 406                 'duration': 10,
 407             },
 408             'params': {
 409                 'videopassword': 'youtube-dl',
 410             },
 411         },
 412         {
 413             'url': 'http://vimeo.com/moogaloop.swf?clip_id=2539741',
 414             'only_matching': True,
 415         },
 416         {
 417             'url': 'https://vimeo.com/109815029',
 418             'note': 'Video not completely processed, "failed" seed status',
 419             'only_matching': True,
 420         },
 421         {
 422             'url': 'https://vimeo.com/groups/travelhd/videos/22439234',
 423             'only_matching': True,
 424         },
 425         {
 426             'url': 'https://vimeo.com/album/2632481/video/79010983',
 427             'only_matching': True,
 428         },
 429         {
 430             # source file returns 403: Forbidden
 431             'url': 'https://vimeo.com/7809605',
 432             'only_matching': True,
 433         },
 434         {
 435             'url': 'https://vimeo.com/160743502/abd0e13fb4',
 436             'only_matching': True,
 437         }
 438     ]
 439
 440     @staticmethod
 441     def _smuggle_referrer(url, referrer_url):
 442         return smuggle_url(url, {'http_headers': {'Referer': referrer_url}})
 443
 444     @staticmethod
 445     def _extract_urls(url, webpage):
 446         urls = []
 447         # Look for embedded (iframe) Vimeo player
 448         for mobj in re.finditer(
 449                 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/\d+.*?)\1',
 450                 webpage):
 451             urls.append(VimeoIE._smuggle_referrer(unescapeHTML(mobj.group('url')), url))
 452         PLAIN_EMBED_RE = (
 453             # Look for embedded (swf embed) Vimeo player
 454             r'<embed[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)\1',
 455             # Look more for non-standard embedded Vimeo player
 456             r'<video[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vimeo\.com/[0-9]+)\1',
 457         )
 458         for embed_re in PLAIN_EMBED_RE:
 459             for mobj in re.finditer(embed_re, webpage):
 460                 urls.append(mobj.group('url'))
 461         return urls
 462
 463     @staticmethod
 464     def _extract_url(url, webpage):
 465         urls = VimeoIE._extract_urls(url, webpage)
 466         return urls[0] if urls else None
 467
 468     def _verify_player_video_password(self, url, video_id):
 469         password = self._downloader.params.get('videopassword')
 470         if password is None:
 471             raise ExtractorError('This video is protected by a password, use the --video-password option')
 472         data = urlencode_postdata({
 473             'password': base64.b64encode(password.encode()),
 474         })
 475         pass_url = url + '/check-password'
 476         password_request = sanitized_Request(pass_url, data)
 477         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 478         password_request.add_header('Referer', url)
 479         return self._download_json(
 480             password_request, video_id,
 481             'Verifying the password', 'Wrong password')
 482
 483     def _real_initialize(self):
 484         self._login()
 485
 486     def _real_extract(self, url):
 487         url, data = unsmuggle_url(url, {})
 488         headers = std_headers.copy()
 489         if 'http_headers' in data:
 490             headers.update(data['http_headers'])
 491         if 'Referer' not in headers:
 492             headers['Referer'] = url
 493
 494         channel_id = self._search_regex(
 495             r'vimeo\.com/channels/([^/]+)', url, 'channel id', default=None)
 496
 497         # Extract ID from URL
 498         mobj = re.match(self._VALID_URL, url)
 499         video_id = mobj.group('id')
 500         orig_url = url
 501         if mobj.group('pro') or mobj.group('player'):
 502             url = 'https://player.vimeo.com/video/' + video_id
 503         elif any(p in url for p in ('play_redirect_hls', 'moogaloop.swf')):
 504             url = 'https://vimeo.com/' + video_id
 505
 506         # Retrieve video webpage to extract further information
 507         request = sanitized_Request(url, headers=headers)
 508         try:
 509             webpage, urlh = self._download_webpage_handle(request, video_id)
 510             redirect_url = compat_str(urlh.geturl())
 511             # Some URLs redirect to ondemand can't be extracted with
 512             # this extractor right away thus should be passed through
 513             # ondemand extractor (e.g. https://vimeo.com/73445910)
 514             if VimeoOndemandIE.suitable(redirect_url):
 515                 return self.url_result(redirect_url, VimeoOndemandIE.ie_key())
 516         except ExtractorError as ee:
 517             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 518                 errmsg = ee.cause.read()
 519                 if b'Because of its privacy settings, this video cannot be played here' in errmsg:
 520                     raise ExtractorError(
 521                         'Cannot download embed-only video without embedding '
 522                         'URL. Please call youtube-dl with the URL of the page '
 523                         'that embeds this video.',
 524                         expected=True)
 525             raise
 526
 527         # Now we begin extracting as much information as we can from what we
 528         # retrieved. First we extract the information common to all extractors,
 529         # and latter we extract those that are Vimeo specific.
 530         self.report_extraction(video_id)
 531
 532         vimeo_config = self._search_regex(
 533             r'vimeo\.config\s*=\s*(?:({.+?})|_extend\([^,]+,\s+({.+?})\));', webpage,
 534             'vimeo config', default=None)
 535         if vimeo_config:
 536             seed_status = self._parse_json(vimeo_config, video_id).get('seed_status', {})
 537             if seed_status.get('state') == 'failed':
 538                 raise ExtractorError(
 539                     '%s said: %s' % (self.IE_NAME, seed_status['title']),
 540                     expected=True)
 541
 542         cc_license = None
 543         timestamp = None
 544
 545         # Extract the config JSON
 546         try:
 547             try:
 548                 config_url = self._html_search_regex(
 549                     r' data-config-url="(.+?)"', webpage,
 550                     'config URL', default=None)
 551                 if not config_url:
 552                     # Sometimes new react-based page is served instead of old one that require
 553                     # different config URL extraction approach (see
 554                     # https://github.com/rg3/youtube-dl/pull/7209)
 555                     vimeo_clip_page_config = self._search_regex(
 556                         r'vimeo\.clip_page_config\s*=\s*({.+?});', webpage,
 557                         'vimeo clip page config')
 558                     page_config = self._parse_json(vimeo_clip_page_config, video_id)
 559                     config_url = page_config['player']['config_url']
 560                     cc_license = page_config.get('cc_license')
 561                     timestamp = try_get(
 562                         page_config, lambda x: x['clip']['uploaded_on'],
 563                         compat_str)
 564                 config_json = self._download_webpage(config_url, video_id)
 565                 config = json.loads(config_json)
 566             except RegexNotFoundError:
 567                 # For pro videos or player.vimeo.com urls
 568                 # We try to find out to which variable is assigned the config dic
 569                 m_variable_name = re.search(r'(\w)\.video\.id', webpage)
 570                 if m_variable_name is not None:
 571                     config_re = [r'%s=({[^}].+?});' % re.escape(m_variable_name.group(1))]
 572                 else:
 573                     config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});']
 574                 config_re.append(r'\bvar\s+r\s*=\s*({.+?})\s*;')
 575                 config_re.append(r'\bconfig\s*=\s*({.+?})\s*;')
 576                 config = self._search_regex(config_re, webpage, 'info section',
 577                                             flags=re.DOTALL)
 578                 config = json.loads(config)
 579         except Exception as e:
 580             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
 581                 raise ExtractorError('The author has restricted the access to this video, try with the "--referer" option')
 582
 583             if re.search(r'<form[^>]+?id="pw_form"', webpage) is not None:
 584                 if '_video_password_verified' in data:
 585                     raise ExtractorError('video password verification failed!')
 586                 self._verify_video_password(redirect_url, video_id, webpage)
 587                 return self._real_extract(
 588                     smuggle_url(redirect_url, {'_video_password_verified': 'verified'}))
 589             else:
 590                 raise ExtractorError('Unable to extract info section',
 591                                      cause=e)
 592         else:
 593             if config.get('view') == 4:
 594                 config = self._verify_player_video_password(redirect_url, video_id)
 595
 596         vod = config.get('video', {}).get('vod', {})
 597
 598         def is_rented():
 599             if '>You rented this title.<' in webpage:
 600                 return True
 601             if config.get('user', {}).get('purchased'):
 602                 return True
 603             for purchase_option in vod.get('purchase_options', []):
 604                 if purchase_option.get('purchased'):
 605                     return True
 606                 label = purchase_option.get('label_string')
 607                 if label and (label.startswith('You rented this') or label.endswith(' remaining')):
 608                     return True
 609             return False
 610
 611         if is_rented() and vod.get('is_trailer'):
 612             feature_id = vod.get('feature_id')
 613             if feature_id and not data.get('force_feature_id', False):
 614                 return self.url_result(smuggle_url(
 615                     'https://player.vimeo.com/player/%s' % feature_id,
 616                     {'force_feature_id': True}), 'Vimeo')
 617
 618         # Extract video description
 619
 620         video_description = self._html_search_regex(
 621             r'(?s)<div\s+class="[^"]*description[^"]*"[^>]*>(.*?)</div>',
 622             webpage, 'description', default=None)
 623         if not video_description:
 624             video_description = self._html_search_meta(
 625                 'description', webpage, default=None)
 626         if not video_description and mobj.group('pro'):
 627             orig_webpage = self._download_webpage(
 628                 orig_url, video_id,
 629                 note='Downloading webpage for description',
 630                 fatal=False)
 631             if orig_webpage:
 632                 video_description = self._html_search_meta(
 633                     'description', orig_webpage, default=None)
 634         if not video_description and not mobj.group('player'):
 635             self._downloader.report_warning('Cannot find video description')
 636
 637         # Extract upload date
 638         if not timestamp:
 639             timestamp = self._search_regex(
 640                 r'<time[^>]+datetime="([^"]+)"', webpage,
 641                 'timestamp', default=None)
 642
 643         try:
 644             view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, 'view count'))
 645             like_count = int(self._search_regex(r'UserLikes:(\d+)', webpage, 'like count'))
 646             comment_count = int(self._search_regex(r'UserComments:(\d+)', webpage, 'comment count'))
 647         except RegexNotFoundError:
 648             # This info is only available in vimeo.com/{id} urls
 649             view_count = None
 650             like_count = None
 651             comment_count = None
 652
 653         formats = []
 654         download_request = sanitized_Request('https://vimeo.com/%s?action=load_download_config' % video_id, headers={
 655             'X-Requested-With': 'XMLHttpRequest'})
 656         download_data = self._download_json(download_request, video_id, fatal=False)
 657         if download_data:
 658             source_file = download_data.get('source_file')
 659             if isinstance(source_file, dict):
 660                 download_url = source_file.get('download_url')
 661                 if download_url and not source_file.get('is_cold') and not source_file.get('is_defrosting'):
 662                     source_name = source_file.get('public_name', 'Original')
 663                     if self._is_valid_url(download_url, video_id, '%s video' % source_name):
 664                         ext = (try_get(
 665                             source_file, lambda x: x['extension'],
 666                             compat_str) or determine_ext(
 667                             download_url, None) or 'mp4').lower()
 668                         formats.append({
 669                             'url': download_url,
 670                             'ext': ext,
 671                             'width': int_or_none(source_file.get('width')),
 672                             'height': int_or_none(source_file.get('height')),
 673                             'filesize': parse_filesize(source_file.get('size')),
 674                             'format_id': source_name,
 675                             'preference': 1,
 676                         })
 677
 678         info_dict_config = self._parse_config(config, video_id)
 679         formats.extend(info_dict_config['formats'])
 680         self._vimeo_sort_formats(formats)
 681
 682         json_ld = self._search_json_ld(webpage, video_id, default={})
 683
 684         if not cc_license:
 685             cc_license = self._search_regex(
 686                 r'<link[^>]+rel=["\']license["\'][^>]+href=(["\'])(?P<license>(?:(?!\1).)+)\1',
 687                 webpage, 'license', default=None, group='license')
 688
 689         channel_url = 'https://vimeo.com/channels/%s' % channel_id if channel_id else None
 690
 691         info_dict = {
 692             'id': video_id,
 693             'formats': formats,
 694             'timestamp': unified_timestamp(timestamp),
 695             'description': video_description,
 696             'webpage_url': url,
 697             'view_count': view_count,
 698             'like_count': like_count,
 699             'comment_count': comment_count,
 700             'license': cc_license,
 701             'channel_id': channel_id,
 702             'channel_url': channel_url,
 703         }
 704
 705         info_dict = merge_dicts(info_dict, info_dict_config, json_ld)
 706
 707         return info_dict
 708
 709
 710 class VimeoOndemandIE(VimeoBaseInfoExtractor):
 711     IE_NAME = 'vimeo:ondemand'
 712     _VALID_URL = r'https?://(?:www\.)?vimeo\.com/ondemand/(?P<id>[^/?#&]+)'
 713     _TESTS = [{
 714         # ondemand video not available via https://vimeo.com/id
 715         'url': 'https://vimeo.com/ondemand/20704',
 716         'md5': 'c424deda8c7f73c1dfb3edd7630e2f35',
 717         'info_dict': {
 718             'id': '105442900',
 719             'ext': 'mp4',
 720             'title': 'המעבדה - במאי יותם פלדמן',
 721             'uploader': 'גם סרטים',
 722             'uploader_url': r're:https?://(?:www\.)?vimeo\.com/gumfilms',
 723             'uploader_id': 'gumfilms',
 724         },
 725         'params': {
 726             'format': 'best[protocol=https]',
 727         },
 728     }, {
 729         # requires Referer to be passed along with og:video:url
 730         'url': 'https://vimeo.com/ondemand/36938/126682985',
 731         'info_dict': {
 732             'id': '126682985',
 733             'ext': 'mp4',
 734             'title': 'Rävlock, rätt läte på rätt plats',
 735             'uploader': 'Lindroth & Norin',
 736             'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user14430847',
 737             'uploader_id': 'user14430847',
 738         },
 739         'params': {
 740             'skip_download': True,
 741         },
 742     }, {
 743         'url': 'https://vimeo.com/ondemand/nazmaalik',
 744         'only_matching': True,
 745     }, {
 746         'url': 'https://vimeo.com/ondemand/141692381',
 747         'only_matching': True,
 748     }, {
 749         'url': 'https://vimeo.com/ondemand/thelastcolony/150274832',
 750         'only_matching': True,
 751     }]
 752
 753     def _real_extract(self, url):
 754         video_id = self._match_id(url)
 755         webpage = self._download_webpage(url, video_id)
 756         return self.url_result(
 757             # Some videos require Referer to be passed along with og:video:url
 758             # similarly to generic vimeo embeds (e.g.
 759             # https://vimeo.com/ondemand/36938/126682985).
 760             VimeoIE._smuggle_referrer(self._og_search_video_url(webpage), url),
 761             VimeoIE.ie_key())
 762
 763
 764 class VimeoChannelIE(VimeoBaseInfoExtractor):
 765     IE_NAME = 'vimeo:channel'
 766     _VALID_URL = r'https://vimeo\.com/channels/(?P<id>[^/?#]+)/?(?:$|[?#])'
 767     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
 768     _TITLE = None
 769     _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"'
 770     _TESTS = [{
 771         'url': 'https://vimeo.com/channels/tributes',
 772         'info_dict': {
 773             'id': 'tributes',
 774             'title': 'Vimeo Tributes',
 775         },
 776         'playlist_mincount': 25,
 777     }]
 778
 779     def _page_url(self, base_url, pagenum):
 780         return '%s/videos/page:%d/' % (base_url, pagenum)
 781
 782     def _extract_list_title(self, webpage):
 783         return self._TITLE or self._html_search_regex(self._TITLE_RE, webpage, 'list title')
 784
 785     def _login_list_password(self, page_url, list_id, webpage):
 786         login_form = self._search_regex(
 787             r'(?s)<form[^>]+?id="pw_form"(.*?)</form>',
 788             webpage, 'login form', default=None)
 789         if not login_form:
 790             return webpage
 791
 792         password = self._downloader.params.get('videopassword')
 793         if password is None:
 794             raise ExtractorError('This album is protected by a password, use the --video-password option', expected=True)
 795         fields = self._hidden_inputs(login_form)
 796         token, vuid = self._extract_xsrft_and_vuid(webpage)
 797         fields['token'] = token
 798         fields['password'] = password
 799         post = urlencode_postdata(fields)
 800         password_path = self._search_regex(
 801             r'action="([^"]+)"', login_form, 'password URL')
 802         password_url = compat_urlparse.urljoin(page_url, password_path)
 803         password_request = sanitized_Request(password_url, post)
 804         password_request.add_header('Content-type', 'application/x-www-form-urlencoded')
 805         self._set_vimeo_cookie('vuid', vuid)
 806         self._set_vimeo_cookie('xsrft', token)
 807
 808         return self._download_webpage(
 809             password_request, list_id,
 810             'Verifying the password', 'Wrong password')
 811
 812     def _title_and_entries(self, list_id, base_url):
 813         for pagenum in itertools.count(1):
 814             page_url = self._page_url(base_url, pagenum)
 815             webpage = self._download_webpage(
 816                 page_url, list_id,
 817                 'Downloading page %s' % pagenum)
 818
 819             if pagenum == 1:
 820                 webpage = self._login_list_password(page_url, list_id, webpage)
 821                 yield self._extract_list_title(webpage)
 822
 823             # Try extracting href first since not all videos are available via
 824             # short https://vimeo.com/id URL (e.g. https://vimeo.com/channels/tributes/6213729)
 825             clips = re.findall(
 826                 r'id="clip_(\d+)"[^>]*>\s*<a[^>]+href="(/(?:[^/]+/)*\1)(?:[^>]+\btitle="([^"]+)")?', webpage)
 827             if clips:
 828                 for video_id, video_url, video_title in clips:
 829                     yield self.url_result(
 830                         compat_urlparse.urljoin(base_url, video_url),
 831                         VimeoIE.ie_key(), video_id=video_id, video_title=video_title)
 832             # More relaxed fallback
 833             else:
 834                 for video_id in re.findall(r'id=["\']clip_(\d+)', webpage):
 835                     yield self.url_result(
 836                         'https://vimeo.com/%s' % video_id,
 837                         VimeoIE.ie_key(), video_id=video_id)
 838
 839             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 840                 break
 841
 842     def _extract_videos(self, list_id, base_url):
 843         title_and_entries = self._title_and_entries(list_id, base_url)
 844         list_title = next(title_and_entries)
 845         return self.playlist_result(title_and_entries, list_id, list_title)
 846
 847     def _real_extract(self, url):
 848         mobj = re.match(self._VALID_URL, url)
 849         channel_id = mobj.group('id')
 850         return self._extract_videos(channel_id, 'https://vimeo.com/channels/%s' % channel_id)
 851
 852
 853 class VimeoUserIE(VimeoChannelIE):
 854     IE_NAME = 'vimeo:user'
 855     _VALID_URL = r'https://vimeo\.com/(?!(?:[0-9]+|watchlater)(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)'
 856     _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
 857     _TESTS = [{
 858         'url': 'https://vimeo.com/nkistudio/videos',
 859         'info_dict': {
 860             'title': 'Nki',
 861             'id': 'nkistudio',
 862         },
 863         'playlist_mincount': 66,
 864     }]
 865
 866     def _real_extract(self, url):
 867         mobj = re.match(self._VALID_URL, url)
 868         name = mobj.group('name')
 869         return self._extract_videos(name, 'https://vimeo.com/%s' % name)
 870
 871
 872 class VimeoAlbumIE(VimeoChannelIE):
 873     IE_NAME = 'vimeo:album'
 874     _VALID_URL = r'https://vimeo\.com/album/(?P<id>\d+)(?:$|[?#]|/(?!video))'
 875     _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
 876     _TESTS = [{
 877         'url': 'https://vimeo.com/album/2632481',
 878         'info_dict': {
 879             'id': '2632481',
 880             'title': 'Staff Favorites: November 2013',
 881         },
 882         'playlist_mincount': 13,
 883     }, {
 884         'note': 'Password-protected album',
 885         'url': 'https://vimeo.com/album/3253534',
 886         'info_dict': {
 887             'title': 'test',
 888             'id': '3253534',
 889         },
 890         'playlist_count': 1,
 891         'params': {
 892             'videopassword': 'youtube-dl',
 893         }
 894     }, {
 895         'url': 'https://vimeo.com/album/2632481/sort:plays/format:thumbnail',
 896         'only_matching': True,
 897     }, {
 898         # TODO: respect page number
 899         'url': 'https://vimeo.com/album/2632481/page:2/sort:plays/format:thumbnail',
 900         'only_matching': True,
 901     }]
 902
 903     def _page_url(self, base_url, pagenum):
 904         return '%s/page:%d/' % (base_url, pagenum)
 905
 906     def _real_extract(self, url):
 907         album_id = self._match_id(url)
 908         return self._extract_videos(album_id, 'https://vimeo.com/album/%s' % album_id)
 909
 910
 911 class VimeoGroupsIE(VimeoAlbumIE):
 912     IE_NAME = 'vimeo:group'
 913     _VALID_URL = r'https://vimeo\.com/groups/(?P<name>[^/]+)(?:/(?!videos?/\d+)|$)'
 914     _TESTS = [{
 915         'url': 'https://vimeo.com/groups/rolexawards',
 916         'info_dict': {
 917             'id': 'rolexawards',
 918             'title': 'Rolex Awards for Enterprise',
 919         },
 920         'playlist_mincount': 73,
 921     }]
 922
 923     def _extract_list_title(self, webpage):
 924         return self._og_search_title(webpage)
 925
 926     def _real_extract(self, url):
 927         mobj = re.match(self._VALID_URL, url)
 928         name = mobj.group('name')
 929         return self._extract_videos(name, 'https://vimeo.com/groups/%s' % name)
 930
 931
 932 class VimeoReviewIE(VimeoBaseInfoExtractor):
 933     IE_NAME = 'vimeo:review'
 934     IE_DESC = 'Review pages on vimeo'
 935     _VALID_URL = r'https://vimeo\.com/[^/]+/review/(?P<id>[^/]+)'
 936     _TESTS = [{
 937         'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d',
 938         'md5': 'c507a72f780cacc12b2248bb4006d253',
 939         'info_dict': {
 940             'id': '75524534',
 941             'ext': 'mp4',
 942             'title': "DICK HARDWICK 'Comedian'",
 943             'uploader': 'Richard Hardwick',
 944             'uploader_id': 'user21297594',
 945         }
 946     }, {
 947         'note': 'video player needs Referer',
 948         'url': 'https://vimeo.com/user22258446/review/91613211/13f927e053',
 949         'md5': '6295fdab8f4bf6a002d058b2c6dce276',
 950         'info_dict': {
 951             'id': '91613211',
 952             'ext': 'mp4',
 953             'title': 're:(?i)^Death by dogma versus assembling agile . Sander Hoogendoorn',
 954             'uploader': 'DevWeek Events',
 955             'duration': 2773,
 956             'thumbnail': r're:^https?://.*\.jpg$',
 957             'uploader_id': 'user22258446',
 958         }
 959     }, {
 960         'note': 'Password protected',
 961         'url': 'https://vimeo.com/user37284429/review/138823582/c4d865efde',
 962         'info_dict': {
 963             'id': '138823582',
 964             'ext': 'mp4',
 965             'title': 'EFFICIENT PICKUP MASTERCLASS MODULE 1',
 966             'uploader': 'TMB',
 967             'uploader_id': 'user37284429',
 968         },
 969         'params': {
 970             'videopassword': 'holygrail',
 971         },
 972         'skip': 'video gone',
 973     }]
 974
 975     def _real_initialize(self):
 976         self._login()
 977
 978     def _get_config_url(self, webpage_url, video_id, video_password_verified=False):
 979         webpage = self._download_webpage(webpage_url, video_id)
 980         config_url = self._html_search_regex(
 981             r'data-config-url=(["\'])(?P<url>(?:(?!\1).)+)\1', webpage,
 982             'config URL', default=None, group='url')
 983         if not config_url:
 984             data = self._parse_json(self._search_regex(
 985                 r'window\s*=\s*_extend\(window,\s*({.+?})\);', webpage, 'data',
 986                 default=NO_DEFAULT if video_password_verified else '{}'), video_id)
 987             config_url = data.get('vimeo_esi', {}).get('config', {}).get('configUrl')
 988         if config_url is None:
 989             self._verify_video_password(webpage_url, video_id, webpage)
 990             config_url = self._get_config_url(
 991                 webpage_url, video_id, video_password_verified=True)
 992         return config_url
 993
 994     def _real_extract(self, url):
 995         video_id = self._match_id(url)
 996         config_url = self._get_config_url(url, video_id)
 997         config = self._download_json(config_url, video_id)
 998         info_dict = self._parse_config(config, video_id)
 999         self._vimeo_sort_formats(info_dict['formats'])
1000         info_dict['id'] = video_id
1001         return info_dict
1002
1003
1004 class VimeoWatchLaterIE(VimeoChannelIE):
1005     IE_NAME = 'vimeo:watchlater'
1006     IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)'
1007     _VALID_URL = r'https://vimeo\.com/(?:home/)?watchlater|:vimeowatchlater'
1008     _TITLE = 'Watch Later'
1009     _LOGIN_REQUIRED = True
1010     _TESTS = [{
1011         'url': 'https://vimeo.com/watchlater',
1012         'only_matching': True,
1013     }]
1014
1015     def _real_initialize(self):
1016         self._login()
1017
1018     def _page_url(self, base_url, pagenum):
1019         url = '%s/page:%d/' % (base_url, pagenum)
1020         request = sanitized_Request(url)
1021         # Set the header to get a partial html page with the ids,
1022         # the normal page doesn't contain them.
1023         request.add_header('X-Requested-With', 'XMLHttpRequest')
1024         return request
1025
1026     def _real_extract(self, url):
1027         return self._extract_videos('watchlater', 'https://vimeo.com/watchlater')
1028
1029
1030 class VimeoLikesIE(InfoExtractor):
1031     _VALID_URL = r'https://(?:www\.)?vimeo\.com/(?P<id>[^/]+)/likes/?(?:$|[?#]|sort:)'
1032     IE_NAME = 'vimeo:likes'
1033     IE_DESC = 'Vimeo user likes'
1034     _TESTS = [{
1035         'url': 'https://vimeo.com/user755559/likes/',
1036         'playlist_mincount': 293,
1037         'info_dict': {
1038             'id': 'user755559_likes',
1039             'description': 'See all the videos urza likes',
1040             'title': 'Videos urza likes',
1041         },
1042     }, {
1043         'url': 'https://vimeo.com/stormlapse/likes',
1044         'only_matching': True,
1045     }]
1046
1047     def _real_extract(self, url):
1048         user_id = self._match_id(url)
1049         webpage = self._download_webpage(url, user_id)
1050         page_count = self._int(
1051             self._search_regex(
1052                 r'''(?x)<li><a\s+href="[^"]+"\s+data-page="([0-9]+)">
1053                     .*?</a></li>\s*<li\s+class="pagination_next">
1054                 ''', webpage, 'page count', default=1),
1055             'page count', fatal=True)
1056         PAGE_SIZE = 12
1057         title = self._html_search_regex(
1058             r'(?s)<h1>(.+?)</h1>', webpage, 'title', fatal=False)
1059         description = self._html_search_meta('description', webpage)
1060
1061         def _get_page(idx):
1062             page_url = 'https://vimeo.com/%s/likes/page:%d/sort:date' % (
1063                 user_id, idx + 1)
1064             webpage = self._download_webpage(
1065                 page_url, user_id,
1066                 note='Downloading page %d/%d' % (idx + 1, page_count))
1067             video_list = self._search_regex(
1068                 r'(?s)<ol class="js-browse_list[^"]+"[^>]*>(.*?)</ol>',
1069                 webpage, 'video content')
1070             paths = re.findall(
1071                 r'<li[^>]*>\s*<a\s+href="([^"]+)"', video_list)
1072             for path in paths:
1073                 yield {
1074                     '_type': 'url',
1075                     'url': compat_urlparse.urljoin(page_url, path),
1076                 }
1077
1078         pl = InAdvancePagedList(_get_page, page_count, PAGE_SIZE)
1079
1080         return {
1081             '_type': 'playlist',
1082             'id': '%s_likes' % user_id,
1083             'title': title,
1084             'description': description,
1085             'entries': pl,
1086         }
1087
1088
1089 class VHXEmbedIE(InfoExtractor):
1090     IE_NAME = 'vhx:embed'
1091     _VALID_URL = r'https?://embed\.vhx\.tv/videos/(?P<id>\d+)'
1092
1093     def _call_api(self, video_id, access_token, path='', query=None):
1094         return self._download_json(
1095             'https://api.vhx.tv/videos/' + video_id + path, video_id, headers={
1096                 'Authorization': 'Bearer ' + access_token,
1097             }, query=query)
1098
1099     def _real_extract(self, url):
1100         video_id = self._match_id(url)
1101         webpage = self._download_webpage(url, video_id)
1102         credentials = self._parse_json(self._search_regex(
1103             r'(?s)credentials\s*:\s*({.+?}),', webpage,
1104             'config'), video_id, js_to_json)
1105         access_token = credentials['access_token']
1106
1107         query = {}
1108         for k, v in credentials.items():
1109             if k in ('authorization', 'authUserToken', 'ticket') and v and v != 'undefined':
1110                 if k == 'authUserToken':
1111                     query['auth_user_token'] = v
1112                 else:
1113                     query[k] = v
1114         files = self._call_api(video_id, access_token, '/files', query)
1115
1116         formats = []
1117         for f in files:
1118             href = try_get(f, lambda x: x['_links']['source']['href'])
1119             if not href:
1120                 continue
1121             method = f.get('method')
1122             if method == 'hls':
1123                 formats.extend(self._extract_m3u8_formats(
1124                     href, video_id, 'mp4', 'm3u8_native',
1125                     m3u8_id='hls', fatal=False))
1126             elif method == 'dash':
1127                 formats.extend(self._extract_mpd_formats(
1128                     href, video_id, mpd_id='dash', fatal=False))
1129             else:
1130                 fmt = {
1131                     'filesize': int_or_none(try_get(f, lambda x: x['size']['bytes'])),
1132                     'format_id': 'http',
1133                     'preference': 1,
1134                     'url': href,
1135                     'vcodec': f.get('codec'),
1136                 }
1137                 quality = f.get('quality')
1138                 if quality:
1139                     fmt.update({
1140                         'format_id': 'http-' + quality,
1141                         'height': int_or_none(self._search_regex(r'(\d+)p', quality, 'height', default=None)),
1142                     })
1143                 formats.append(fmt)
1144         self._sort_formats(formats)
1145
1146         video_data = self._call_api(video_id, access_token)
1147         title = video_data.get('title') or video_data['name']
1148
1149         subtitles = {}
1150         for subtitle in try_get(video_data, lambda x: x['tracks']['subtitles'], list) or []:
1151             lang = subtitle.get('srclang') or subtitle.get('label')
1152             for _link in subtitle.get('_links', {}).values():
1153                 href = _link.get('href')
1154                 if not href:
1155                     continue
1156                 subtitles.setdefault(lang, []).append({
1157                     'url': href,
1158                 })
1159
1160         q = qualities(['small', 'medium', 'large', 'source'])
1161         thumbnails = []
1162         for thumbnail_id, thumbnail_url in video_data.get('thumbnail', {}).items():
1163             thumbnails.append({
1164                 'id': thumbnail_id,
1165                 'url': thumbnail_url,
1166                 'preference': q(thumbnail_id),
1167             })
1168
1169         return {
1170             'id': video_id,
1171             'title': title,
1172             'description': video_data.get('description'),
1173             'duration': int_or_none(try_get(video_data, lambda x: x['duration']['seconds'])),
1174             'formats': formats,
1175             'subtitles': subtitles,
1176             'thumbnails': thumbnails,
1177             'timestamp': unified_timestamp(video_data.get('created_at')),
1178             'view_count': int_or_none(video_data.get('plays_count')),
1179         }