yt_dlp/extractor/facebook.py

   1 import json
   2 import re
   3 import urllib.parse
   4
   5 from .common import InfoExtractor
   6 from ..compat import compat_etree_fromstring
   7 from ..networking import Request
   8 from ..networking.exceptions import network_exceptions
   9 from ..utils import (
  10     ExtractorError,
  11     clean_html,
  12     determine_ext,
  13     float_or_none,
  14     format_field,
  15     get_element_by_id,
  16     get_first,
  17     int_or_none,
  18     join_nonempty,
  19     js_to_json,
  20     merge_dicts,
  21     parse_count,
  22     parse_qs,
  23     qualities,
  24     str_or_none,
  25     traverse_obj,
  26     try_get,
  27     url_or_none,
  28     urlencode_postdata,
  29     urljoin,
  30     variadic,
  31 )
  32
  33
  34 class FacebookIE(InfoExtractor):
  35     _VALID_URL = r'''(?x)
  36                 (?:
  37                     https?://
  38                         (?:[\w-]+\.)?(?:facebook\.com|facebookwkhpilnemxj7asaniu7vnjjbiltxjqhye3mhbshg7kx5tfyd\.onion)/
  39                         (?:[^#]*?\#!/)?
  40                         (?:
  41                             (?:
  42                                 permalink\.php|
  43                                 video/video\.php|
  44                                 photo\.php|
  45                                 video\.php|
  46                                 video/embed|
  47                                 story\.php|
  48                                 watch(?:/live)?/?
  49                             )\?(?:.*?)(?:v|video_id|story_fbid)=|
  50                             [^/]+/videos/(?:[^/]+/)?|
  51                             [^/]+/posts/|
  52                             events/(?:[^/]+/)?|
  53                             groups/[^/]+/(?:permalink|posts)/|
  54                             watchparty/
  55                         )|
  56                     facebook:
  57                 )
  58                 (?P<id>pfbid[A-Za-z0-9]+|\d+)
  59                 '''
  60     _EMBED_REGEX = [
  61         r'<iframe[^>]+?src=(["\'])(?P<url>https?://www\.facebook\.com/(?:video/embed|plugins/video\.php).+?)\1',
  62         # Facebook API embed https://developers.facebook.com/docs/plugins/embedded-video-player
  63         r'''(?x)<div[^>]+
  64                 class=(?P<q1>[\'"])[^\'"]*\bfb-(?:video|post)\b[^\'"]*(?P=q1)[^>]+
  65                 data-href=(?P<q2>[\'"])(?P<url>(?:https?:)?//(?:www\.)?facebook.com/.+?)(?P=q2)''',
  66     ]
  67     _LOGIN_URL = 'https://www.facebook.com/login.php?next=http%3A%2F%2Ffacebook.com%2Fhome.php&login_attempt=1'
  68     _CHECKPOINT_URL = 'https://www.facebook.com/checkpoint/?next=http%3A%2F%2Ffacebook.com%2Fhome.php&_fb_noscript=1'
  69     _NETRC_MACHINE = 'facebook'
  70     IE_NAME = 'facebook'
  71
  72     _VIDEO_PAGE_TEMPLATE = 'https://www.facebook.com/video/video.php?v=%s'
  73     _VIDEO_PAGE_TAHOE_TEMPLATE = 'https://www.facebook.com/video/tahoe/async/%s/?chain=true&isvideo=true&payloadtype=primary'
  74
  75     _TESTS = [{
  76         'url': 'https://www.facebook.com/radiokicksfm/videos/3676516585958356/',
  77         'info_dict': {
  78             'id': '3676516585958356',
  79             'ext': 'mp4',
  80             'title': 'dr Adam Przygoda',
  81             'description': 'md5:34675bda53336b1d16400265c2bb9b3b',
  82             'uploader': 'RADIO KICKS FM',
  83             'upload_date': '20230818',
  84             'timestamp': 1692346159,
  85             'thumbnail': r're:^https?://.*',
  86             'uploader_id': '100063551323670',
  87             'duration': 3132.184,
  88             'view_count': int,
  89             'concurrent_view_count': 0,
  90         },
  91     }, {
  92         'url': 'https://www.facebook.com/video.php?v=637842556329505&fref=nf',
  93         'md5': '6a40d33c0eccbb1af76cf0485a052659',
  94         'info_dict': {
  95             'id': '637842556329505',
  96             'ext': 'mp4',
  97             'title': 're:Did you know Kei Nishikori is the first Asian man to ever reach a Grand Slam',
  98             'uploader': 'Tennis on Facebook',
  99             'upload_date': '20140908',
 100             'timestamp': 1410199200,
 101         },
 102         'skip': 'Requires logging in',
 103     }, {
 104         # data.video
 105         'url': 'https://www.facebook.com/video.php?v=274175099429670',
 106         'info_dict': {
 107             'id': '274175099429670',
 108             'ext': 'mp4',
 109             'title': 'Asif',
 110             'description': '',
 111             'uploader': 'Asif Nawab Butt',
 112             'upload_date': '20140506',
 113             'timestamp': 1399398998,
 114             'thumbnail': r're:^https?://.*',
 115             'uploader_id': 'pfbid028wxorhX2ErLFJ578N6P3crHD3PHmXTCqCvfBpsnbSLmbokwSY75p5hWBjHGkG4zxl',
 116             'duration': 131.03,
 117             'concurrent_view_count': int,
 118         },
 119     }, {
 120         'note': 'Video with DASH manifest',
 121         'url': 'https://www.facebook.com/video.php?v=957955867617029',
 122         'md5': 'b2c28d528273b323abe5c6ab59f0f030',
 123         'info_dict': {
 124             'id': '957955867617029',
 125             'ext': 'mp4',
 126             'title': 'When you post epic content on instagram.com/433 8 million followers, this is ...',
 127             'uploader': 'Demy de Zeeuw',
 128             'upload_date': '20160110',
 129             'timestamp': 1452431627,
 130         },
 131         'skip': 'Requires logging in',
 132     }, {
 133         'url': 'https://www.facebook.com/maxlayn/posts/10153807558977570',
 134         'md5': '037b1fa7f3c2d02b7a0d7bc16031ecc6',
 135         'info_dict': {
 136             'id': '544765982287235',
 137             'ext': 'mp4',
 138             'title': '"What are you doing running in the snow?"',
 139             'uploader': 'FailArmy',
 140         },
 141         'skip': 'Video gone',
 142     }, {
 143         'url': 'https://m.facebook.com/story.php?story_fbid=1035862816472149&id=116132035111903',
 144         'md5': '1deb90b6ac27f7efcf6d747c8a27f5e3',
 145         'info_dict': {
 146             'id': '1035862816472149',
 147             'ext': 'mp4',
 148             'title': 'What the Flock Is Going On In New Zealand  Credit: ViralHog',
 149             'uploader': 'S. Saint',
 150         },
 151         'skip': 'Video gone',
 152     }, {
 153         'note': 'swf params escaped',
 154         'url': 'https://www.facebook.com/barackobama/posts/10153664894881749',
 155         'md5': '97ba073838964d12c70566e0085c2b91',
 156         'info_dict': {
 157             'id': '10153664894881749',
 158             'ext': 'mp4',
 159             'title': 'Average time to confirm recent Supreme Court nominees: 67 days Longest it\'s t...',
 160             'thumbnail': r're:^https?://.*',
 161             'timestamp': 1456259628,
 162             'upload_date': '20160223',
 163             'uploader': 'Barack Obama',
 164         },
 165         'skip': 'Gif on giphy.com gone',
 166     }, {
 167         # have 1080P, but only up to 720p in swf params
 168         # data.video.story.attachments[].media
 169         'url': 'https://www.facebook.com/cnn/videos/10155529876156509/',
 170         'md5': 'ca63897a90c9452efee5f8c40d080e25',
 171         'info_dict': {
 172             'id': '10155529876156509',
 173             'ext': 'mp4',
 174             'title': 'Holocaust survivor becomes US citizen',
 175             'description': 'She survived the holocaust — and years later, she’s getting her citizenship so she can vote for Hillary Clinton http://cnn.it/2eERh5f',
 176             'timestamp': 1477818095,
 177             'upload_date': '20161030',
 178             'uploader': 'CNN',
 179             'thumbnail': r're:^https?://.*',
 180             'view_count': int,
 181             'uploader_id': '100059479812265',
 182             'concurrent_view_count': int,
 183             'duration': 44.478,
 184         },
 185     }, {
 186         # bigPipe.onPageletArrive ... onPageletArrive pagelet_group_mall
 187         # data.node.comet_sections.content.story.attachments[].style_type_renderer.attachment.media
 188         'url': 'https://www.facebook.com/yaroslav.korpan/videos/1417995061575415/',
 189         'info_dict': {
 190             'id': '1417995061575415',
 191             'ext': 'mp4',
 192             'title': 'Довгоочікуване відео | By Yaroslav - Facebook',
 193             'description': 'Довгоочікуване відео',
 194             'timestamp': 1486648217,
 195             'upload_date': '20170209',
 196             'uploader': 'Yaroslav Korpan',
 197             'uploader_id': 'pfbid06AScABAWcW91qpiuGrLt99Ef9tvwHoXP6t8KeFYEqkSfreMtfa9nTveh8b2ZEVSWl',
 198             'concurrent_view_count': int,
 199             'thumbnail': r're:^https?://.*',
 200             'view_count': int,
 201             'duration': 11736.446,
 202         },
 203         'params': {
 204             'skip_download': True,
 205         },
 206     }, {
 207         # FIXME: Cannot parse data error
 208         'url': 'https://www.facebook.com/LaGuiaDelVaron/posts/1072691702860471',
 209         'info_dict': {
 210             'id': '1072691702860471',
 211             'ext': 'mp4',
 212             'title': 'md5:ae2d22a93fbb12dad20dc393a869739d',
 213             'timestamp': 1477305000,
 214             'upload_date': '20161024',
 215             'uploader': 'La Guía Del Varón',
 216             'thumbnail': r're:^https?://.*',
 217         },
 218         'skip': 'Requires logging in',
 219     }, {
 220         # data.node.comet_sections.content.story.attachments[].style_type_renderer.attachment.media
 221         'url': 'https://www.facebook.com/groups/1024490957622648/permalink/1396382447100162/',
 222         'info_dict': {
 223             'id': '202882990186699',
 224             'ext': 'mp4',
 225             'title': 'birb (O v O") | Hello? Yes your uber ride is here',
 226             'description': 'Hello? Yes your uber ride is here * Jukin Media Verified * Find this video and others like it by visiting...',
 227             'timestamp': 1486035513,
 228             'upload_date': '20170202',
 229             'uploader': 'Elisabeth Ahtn',
 230             'uploader_id': '100013949973717',
 231         },
 232         'skip': 'Requires logging in',
 233     }, {
 234         # data.node.comet_sections.content.story.attachments[].throwbackStyles.attachment_target_renderer.attachment.target.attachments[].styles.attachment.media
 235         'url': 'https://www.facebook.com/groups/1645456212344334/posts/3737828833107051/',
 236         'info_dict': {
 237             'id': '1569199726448814',
 238             'ext': 'mp4',
 239             'title': 'Pence MUST GO!',
 240             'description': 'Vickie Gentry shared a memory.',
 241             'timestamp': 1511548260,
 242             'upload_date': '20171124',
 243             'uploader': 'Vickie Gentry',
 244             'uploader_id': 'pfbid0FuZhHCeWDAxWxEbr3yKPFaRstXvRxgsp9uCPG6GjD4J2AitB35NUAuJ4Q75KcjiDl',
 245             'thumbnail': r're:^https?://.*',
 246             'duration': 148.435,
 247         },
 248     }, {
 249         # data.node.comet_sections.content.story.attachments[].styles.attachment.media
 250         'url': 'https://www.facebook.com/attn/posts/pfbid0j1Czf2gGDVqeQ8KiMLFm3pWN8GxsQmeRrVhimWDzMuKQoR8r4b1knNsejELmUgyhl',
 251         'info_dict': {
 252             'id': '6968553779868435',
 253             'ext': 'mp4',
 254             'description': 'md5:2f2fcf93e97ac00244fe64521bbdb0cb',
 255             'uploader': 'ATTN:',
 256             'upload_date': '20231207',
 257             'title': 'ATTN:',
 258             'duration': 132.675,
 259             'uploader_id': '100064451419378',
 260             'view_count': int,
 261             'thumbnail': r're:^https?://.*',
 262             'timestamp': 1701975646,
 263         },
 264     }, {
 265         # data.node.comet_sections.content.story.attachments[].styles.attachment.media
 266         'url': 'https://www.facebook.com/permalink.php?story_fbid=pfbid0fqQuVEQyXRa9Dp4RcaTR14KHU3uULHV1EK7eckNXSH63JMuoALsAvVCJ97zAGitil&id=100068861234290',
 267         'info_dict': {
 268             'id': '270103405756416',
 269             'ext': 'mp4',
 270             'title': 'Lela Evans',
 271             'description': 'Today Makkovik\'s own Pilot Mandy Smith made her inaugural landing on the airstrip in her hometown. What a proud moment as we all cheered and...',
 272             'thumbnail': r're:^https?://.*',
 273             'uploader': 'Lela Evans',
 274             'uploader_id': 'pfbid0shZJipuigyy5mqrUJn9ub5LJFWNHvan5prtyi3LrDuuuJ4NwrURgnQHYR9fywBepl',
 275             'upload_date': '20231228',
 276             'timestamp': 1703804085,
 277             'duration': 394.347,
 278             'view_count': int,
 279         },
 280     }, {
 281         'url': 'https://www.facebook.com/story.php?story_fbid=pfbid0Fnzhm8UuzjBYpPMNFzaSpFE9UmLdU4fJN8qTANi1Dmtj5q7DNrL5NERXfsAzDEV7l&id=100073071055552',
 282         'only_matching': True,
 283     }, {
 284         'url': 'https://www.facebook.com/video.php?v=10204634152394104',
 285         'only_matching': True,
 286     }, {
 287         'url': 'https://www.facebook.com/amogood/videos/1618742068337349/?fref=nf',
 288         'only_matching': True,
 289     }, {
 290         # data.mediaset.currMedia.edges
 291         'url': 'https://www.facebook.com/ChristyClarkForBC/videos/vb.22819070941/10153870694020942/?type=2&theater',
 292         'only_matching': True,
 293     }, {
 294         # data.video.story.attachments[].media
 295         'url': 'facebook:544765982287235',
 296         'only_matching': True,
 297     }, {
 298         # data.node.comet_sections.content.story.attachments[].style_type_renderer.attachment.media
 299         'url': 'https://www.facebook.com/groups/164828000315060/permalink/764967300301124/',
 300         'only_matching': True,
 301     }, {
 302         # data.video.creation_story.attachments[].media
 303         'url': 'https://zh-hk.facebook.com/peoplespower/videos/1135894589806027/',
 304         'only_matching': True,
 305     }, {
 306         # data.video
 307         'url': 'https://www.facebookwkhpilnemxj7asaniu7vnjjbiltxjqhye3mhbshg7kx5tfyd.onion/video.php?v=274175099429670',
 308         'only_matching': True,
 309     }, {
 310         # no title
 311         'url': 'https://www.facebook.com/onlycleverentertainment/videos/1947995502095005/',
 312         'only_matching': True,
 313     }, {
 314         # data.video
 315         'url': 'https://www.facebook.com/WatchESLOne/videos/359649331226507/',
 316         'info_dict': {
 317             'id': '359649331226507',
 318             'ext': 'mp4',
 319             'title': 'Fnatic vs. EG - Group A - Opening Match - ESL One Birmingham Day 1',
 320             'description': '#ESLOne VoD - Birmingham Finals Day#1 Fnatic vs. @Evil Geniuses',
 321             'timestamp': 1527084179,
 322             'upload_date': '20180523',
 323             'uploader': 'ESL One Dota 2',
 324             'uploader_id': '100066514874195',
 325             'duration': 4524.212,
 326             'view_count': int,
 327             'thumbnail': r're:^https?://.*',
 328             'concurrent_view_count': int,
 329         },
 330         'params': {
 331             'skip_download': True,
 332         },
 333     }, {
 334         # data.node.comet_sections.content.story.attachments[].style_type_renderer.attachment.all_subattachments.nodes[].media
 335         'url': 'https://www.facebook.com/100033620354545/videos/106560053808006/',
 336         'info_dict': {
 337             'id': '106560053808006',
 338             'ext': 'mp4',
 339             'title': 'Josef',
 340             'thumbnail': r're:^https?://.*',
 341             'concurrent_view_count': int,
 342             'uploader_id': 'pfbid0cibUN6tV7DYgdbJdsUFN46wc4jKpVSPAvJQhFofGqBGmVn3V3JtAs2tfUwziw2hUl',
 343             'timestamp': 1549275572,
 344             'duration': 3.413,
 345             'uploader': 'Josef Novak',
 346             'description': '',
 347             'upload_date': '20190204',
 348         },
 349     }, {
 350         # data.video.story.attachments[].media
 351         'url': 'https://www.facebook.com/watch/?v=647537299265662',
 352         'only_matching': True,
 353     }, {
 354         # FIXME: https://github.com/yt-dlp/yt-dlp/issues/542
 355         # data.node.comet_sections.content.story.attachments[].style_type_renderer.attachment.all_subattachments.nodes[].media
 356         'url': 'https://www.facebook.com/PankajShahLondon/posts/10157667649866271',
 357         'info_dict': {
 358             'id': '10157667649866271',
 359         },
 360         'playlist_count': 3,
 361         'skip': 'Requires logging in',
 362     }, {
 363         # data.nodes[].comet_sections.content.story.attachments[].style_type_renderer.attachment.media
 364         'url': 'https://m.facebook.com/Alliance.Police.Department/posts/4048563708499330',
 365         'info_dict': {
 366             'id': '117576630041613',
 367             'ext': 'mp4',
 368             # TODO: title can be extracted from video page
 369             'title': 'Facebook video #117576630041613',
 370             'uploader_id': '189393014416438',
 371             'upload_date': '20201123',
 372             'timestamp': 1606162592,
 373         },
 374         'skip': 'Requires logging in',
 375     }, {
 376         # node.comet_sections.content.story.attached_story.attachments.style_type_renderer.attachment.media
 377         'url': 'https://www.facebook.com/groups/ateistiskselskab/permalink/10154930137678856/',
 378         'info_dict': {
 379             'id': '211567722618337',
 380             'ext': 'mp4',
 381             'title': 'Facebook video #211567722618337',
 382             'uploader_id': '127875227654254',
 383             'upload_date': '20161122',
 384             'timestamp': 1479793574,
 385         },
 386         'skip': 'No video',
 387     }, {
 388         # data.video.creation_story.attachments[].media
 389         'url': 'https://www.facebook.com/watch/live/?v=1823658634322275',
 390         'only_matching': True,
 391     }, {
 392         'url': 'https://www.facebook.com/watchparty/211641140192478',
 393         'info_dict': {
 394             'id': '211641140192478',
 395         },
 396         'playlist_count': 1,
 397         'skip': 'Requires logging in',
 398     }, {
 399         # data.event.cover_media_renderer.cover_video
 400         'url': 'https://m.facebook.com/events/1509582499515440',
 401         'info_dict': {
 402             'id': '637246984455045',
 403             'ext': 'mp4',
 404             'title': 'ANALISI IN CAMPO OSCURO " Coaguli nel sangue dei vaccinati"',
 405             'description': 'Other event by Comitato Liberi Pensatori on Tuesday, October 18 2022',
 406             'thumbnail': r're:^https?://.*',
 407             'uploader': 'Comitato Liberi Pensatori',
 408             'uploader_id': '100065709540881',
 409         },
 410     }]
 411     _SUPPORTED_PAGLETS_REGEX = r'(?:pagelet_group_mall|permalink_video_pagelet|hyperfeed_story_id_[0-9a-f]+)'
 412     _api_config = {
 413         'graphURI': '/api/graphql/',
 414     }
 415
 416     def _perform_login(self, username, password):
 417         login_page_req = Request(self._LOGIN_URL)
 418         self._set_cookie('facebook.com', 'locale', 'en_US')
 419         login_page = self._download_webpage(login_page_req, None,
 420                                             note='Downloading login page',
 421                                             errnote='Unable to download login page')
 422         lsd = self._search_regex(
 423             r'<input type="hidden" name="lsd" value="([^"]*)"',
 424             login_page, 'lsd')
 425         lgnrnd = self._search_regex(r'name="lgnrnd" value="([^"]*?)"', login_page, 'lgnrnd')
 426
 427         login_form = {
 428             'email': username,
 429             'pass': password,
 430             'lsd': lsd,
 431             'lgnrnd': lgnrnd,
 432             'next': 'http://facebook.com/home.php',
 433             'default_persistent': '0',
 434             'legacy_return': '1',
 435             'timezone': '-60',
 436             'trynum': '1',
 437         }
 438         request = Request(self._LOGIN_URL, urlencode_postdata(login_form))
 439         request.headers['Content-Type'] = 'application/x-www-form-urlencoded'
 440         try:
 441             login_results = self._download_webpage(request, None,
 442                                                    note='Logging in', errnote='unable to fetch login page')
 443             if re.search(r'<form(.*)name="login"(.*)</form>', login_results) is not None:
 444                 error = self._html_search_regex(
 445                     r'(?s)<div[^>]+class=(["\']).*?login_error_box.*?\1[^>]*><div[^>]*>.*?</div><div[^>]*>(?P<error>.+?)</div>',
 446                     login_results, 'login error', default=None, group='error')
 447                 if error:
 448                     raise ExtractorError(f'Unable to login: {error}', expected=True)
 449                 self.report_warning('unable to log in: bad username/password, or exceeded login rate limit (~3/min). Check credentials or wait.')
 450                 return
 451
 452             fb_dtsg = self._search_regex(
 453                 r'name="fb_dtsg" value="(.+?)"', login_results, 'fb_dtsg', default=None)
 454             h = self._search_regex(
 455                 r'name="h"\s+(?:\w+="[^"]+"\s+)*?value="([^"]+)"', login_results, 'h', default=None)
 456
 457             if not fb_dtsg or not h:
 458                 return
 459
 460             check_form = {
 461                 'fb_dtsg': fb_dtsg,
 462                 'h': h,
 463                 'name_action_selected': 'dont_save',
 464             }
 465             check_req = Request(self._CHECKPOINT_URL, urlencode_postdata(check_form))
 466             check_req.headers['Content-Type'] = 'application/x-www-form-urlencoded'
 467             check_response = self._download_webpage(check_req, None,
 468                                                     note='Confirming login')
 469             if re.search(r'id="checkpointSubmitButton"', check_response) is not None:
 470                 self.report_warning('Unable to confirm login, you have to login in your browser and authorize the login.')
 471         except network_exceptions as err:
 472             self.report_warning(f'unable to log in: {err}')
 473             return
 474
 475     def _extract_from_url(self, url, video_id):
 476         webpage = self._download_webpage(
 477             url.replace('://m.facebook.com/', '://www.facebook.com/'), video_id)
 478
 479         def extract_metadata(webpage):
 480             post_data = [self._parse_json(j, video_id, fatal=False) for j in re.findall(
 481                 r'data-sjs>({.*?ScheduledServerJS.*?})</script>', webpage)]
 482             post = traverse_obj(post_data, (
 483                 ..., 'require', ..., ..., ..., '__bbox', 'require', ..., ..., ..., '__bbox', 'result', 'data'), expected_type=dict) or []
 484             media = traverse_obj(post, (..., 'attachments', ..., lambda k, v: (
 485                 k == 'media' and str(v['id']) == video_id and v['__typename'] == 'Video')), expected_type=dict)
 486             title = get_first(media, ('title', 'text'))
 487             description = get_first(media, ('creation_story', 'comet_sections', 'message', 'story', 'message', 'text'))
 488             page_title = title or self._html_search_regex((
 489                 r'<h2\s+[^>]*class="uiHeaderTitle"[^>]*>(?P<content>[^<]*)</h2>',
 490                 r'(?s)<span class="fbPhotosPhotoCaption".*?id="fbPhotoPageCaption"><span class="hasCaption">(?P<content>.*?)</span>',
 491                 self._meta_regex('og:title'), self._meta_regex('twitter:title'), r'<title>(?P<content>.+?)</title>',
 492             ), webpage, 'title', default=None, group='content')
 493             description = description or self._html_search_meta(
 494                 ['description', 'og:description', 'twitter:description'],
 495                 webpage, 'description', default=None)
 496             uploader_data = (
 497                 get_first(media, ('owner', {dict}))
 498                 or get_first(post, ('video', 'creation_story', 'attachments', ..., 'media', lambda k, v: k == 'owner' and v['name']))
 499                 or get_first(post, (..., 'video', lambda k, v: k == 'owner' and v['name']))
 500                 or get_first(post, ('node', 'actors', ..., {dict}))
 501                 or get_first(post, ('event', 'event_creator', {dict})) or {})
 502             uploader = uploader_data.get('name') or (
 503                 clean_html(get_element_by_id('fbPhotoPageAuthorName', webpage))
 504                 or self._search_regex(
 505                     (r'ownerName\s*:\s*"([^"]+)"', *self._og_regexes('title')), webpage, 'uploader', fatal=False))
 506             timestamp = int_or_none(self._search_regex(
 507                 r'<abbr[^>]+data-utime=["\'](\d+)', webpage,
 508                 'timestamp', default=None))
 509             thumbnail = self._html_search_meta(
 510                 ['og:image', 'twitter:image'], webpage, 'thumbnail', default=None)
 511             # some webpages contain unretrievable thumbnail urls
 512             # like https://lookaside.fbsbx.com/lookaside/crawler/media/?media_id=10155168902769113&get_thumbnail=1
 513             # in https://www.facebook.com/yaroslav.korpan/videos/1417995061575415/
 514             if thumbnail and not re.search(r'\.(?:jpg|png)', thumbnail):
 515                 thumbnail = None
 516             info_dict = {
 517                 'description': description,
 518                 'uploader': uploader,
 519                 'uploader_id': uploader_data.get('id'),
 520                 'timestamp': timestamp,
 521                 'thumbnail': thumbnail,
 522                 'view_count': parse_count(self._search_regex(
 523                     (r'\bviewCount\s*:\s*["\']([\d,.]+)', r'video_view_count["\']\s*:\s*(\d+)'),
 524                     webpage, 'view count', default=None)),
 525                 'concurrent_view_count': get_first(post, (
 526                     ('video', (..., ..., 'attachments', ..., 'media')), 'liveViewerCount', {int_or_none})),
 527             }
 528
 529             info_json_ld = self._search_json_ld(webpage, video_id, default={})
 530             info_json_ld['title'] = (re.sub(r'\s*\|\s*Facebook$', '', title or info_json_ld.get('title') or page_title or '')
 531                                      or (description or '').replace('\n', ' ') or f'Facebook video #{video_id}')
 532             return merge_dicts(info_json_ld, info_dict)
 533
 534         video_data = None
 535
 536         def extract_video_data(instances):
 537             video_data = []
 538             for item in instances:
 539                 if try_get(item, lambda x: x[1][0]) == 'VideoConfig':
 540                     video_item = item[2][0]
 541                     if video_item.get('video_id'):
 542                         video_data.append(video_item['videoData'])
 543             return video_data
 544
 545         server_js_data = self._parse_json(self._search_regex(
 546             [r'handleServerJS\(({.+})(?:\);|,")', r'\bs\.handle\(({.+?})\);'],
 547             webpage, 'server js data', default='{}'), video_id, fatal=False)
 548
 549         if server_js_data:
 550             video_data = extract_video_data(server_js_data.get('instances', []))
 551
 552         def extract_from_jsmods_instances(js_data):
 553             if js_data:
 554                 return extract_video_data(try_get(
 555                     js_data, lambda x: x['jsmods']['instances'], list) or [])
 556
 557         def extract_dash_manifest(video, formats):
 558             dash_manifest = traverse_obj(video, 'dash_manifest', 'playlist', expected_type=str)
 559             if dash_manifest:
 560                 formats.extend(self._parse_mpd_formats(
 561                     compat_etree_fromstring(urllib.parse.unquote_plus(dash_manifest)),
 562                     mpd_url=video.get('dash_manifest_url')))
 563
 564         def process_formats(info):
 565             # Downloads with browser's User-Agent are rate limited. Working around
 566             # with non-browser User-Agent.
 567             for f in info['formats']:
 568                 # Downloads with browser's User-Agent are rate limited. Working around
 569                 # with non-browser User-Agent.
 570                 f.setdefault('http_headers', {})['User-Agent'] = 'facebookexternalhit/1.1'
 571                 # Formats larger than ~500MB will return error 403 unless chunk size is regulated
 572                 f.setdefault('downloader_options', {})['http_chunk_size'] = 250 << 20
 573
 574         def extract_relay_data(_filter):
 575             return self._parse_json(self._search_regex(
 576                 rf'data-sjs>({{.*?{_filter}.*?}})</script>',
 577                 webpage, 'replay data', default='{}'), video_id, fatal=False) or {}
 578
 579         def extract_relay_prefetched_data(_filter):
 580             return traverse_obj(extract_relay_data(_filter), (
 581                 'require', (None, (..., ..., ..., '__bbox', 'require')),
 582                 lambda _, v: any(key.startswith('RelayPrefetchedStreamCache') for key in v),
 583                 ..., ..., '__bbox', 'result', 'data', {dict}), get_all=False) or {}
 584
 585         if not video_data:
 586             server_js_data = self._parse_json(self._search_regex([
 587                 r'bigPipe\.onPageletArrive\(({.+?})\)\s*;\s*}\s*\)\s*,\s*["\']onPageletArrive\s+' + self._SUPPORTED_PAGLETS_REGEX,
 588                 rf'bigPipe\.onPageletArrive\(({{.*?id\s*:\s*"{self._SUPPORTED_PAGLETS_REGEX}".*?}})\);',
 589             ], webpage, 'js data', default='{}'), video_id, js_to_json, False)
 590             video_data = extract_from_jsmods_instances(server_js_data)
 591
 592         if not video_data:
 593             data = extract_relay_prefetched_data(
 594                 r'"(?:dash_manifest|playable_url(?:_quality_hd)?)')
 595             if data:
 596                 entries = []
 597
 598                 def parse_graphql_video(video):
 599                     v_id = video.get('videoId') or video.get('id') or video_id
 600                     reel_info = traverse_obj(
 601                         video, ('creation_story', 'short_form_video_context', 'playback_video', {dict}))
 602                     if reel_info:
 603                         video = video['creation_story']
 604                         video['owner'] = traverse_obj(video, ('short_form_video_context', 'video_owner'))
 605                         video.update(reel_info)
 606                     formats = []
 607                     q = qualities(['sd', 'hd'])
 608                     for key, format_id in (('playable_url', 'sd'), ('playable_url_quality_hd', 'hd'),
 609                                            ('playable_url_dash', ''), ('browser_native_hd_url', 'hd'),
 610                                            ('browser_native_sd_url', 'sd')):
 611                         playable_url = video.get(key)
 612                         if not playable_url:
 613                             continue
 614                         if determine_ext(playable_url) == 'mpd':
 615                             formats.extend(self._extract_mpd_formats(playable_url, video_id))
 616                         else:
 617                             formats.append({
 618                                 'format_id': format_id,
 619                                 # sd, hd formats w/o resolution info should be deprioritized below DASH
 620                                 'quality': q(format_id) - 3,
 621                                 'url': playable_url,
 622                             })
 623                     extract_dash_manifest(video, formats)
 624
 625                     automatic_captions, subtitles = {}, {}
 626                     is_broadcast = traverse_obj(video, ('is_video_broadcast', {bool}))
 627                     for caption in traverse_obj(video, (
 628                         'video_available_captions_locales',
 629                         {lambda x: sorted(x, key=lambda c: c['locale'])},
 630                         lambda _, v: url_or_none(v['captions_url']),
 631                     )):
 632                         lang = caption.get('localized_language') or 'und'
 633                         subs = {
 634                             'url': caption['captions_url'],
 635                             'name': format_field(caption, 'localized_country', f'{lang} (%s)', default=lang),
 636                         }
 637                         if caption.get('localized_creation_method') or is_broadcast:
 638                             automatic_captions.setdefault(caption['locale'], []).append(subs)
 639                         else:
 640                             subtitles.setdefault(caption['locale'], []).append(subs)
 641                     captions_url = traverse_obj(video, ('captions_url', {url_or_none}))
 642                     if captions_url and not automatic_captions and not subtitles:
 643                         locale = self._html_search_meta(
 644                             ['og:locale', 'twitter:locale'], webpage, 'locale', default='en_US')
 645                         (automatic_captions if is_broadcast else subtitles)[locale] = [{'url': captions_url}]
 646
 647                     info = {
 648                         'id': v_id,
 649                         'formats': formats,
 650                         'thumbnail': traverse_obj(
 651                             video, ('thumbnailImage', 'uri'), ('preferred_thumbnail', 'image', 'uri')),
 652                         'uploader_id': traverse_obj(video, ('owner', 'id', {str_or_none})),
 653                         'timestamp': traverse_obj(video, 'publish_time', 'creation_time', expected_type=int_or_none),
 654                         'duration': (float_or_none(video.get('playable_duration_in_ms'), 1000)
 655                                      or float_or_none(video.get('length_in_second'))),
 656                         'automatic_captions': automatic_captions,
 657                         'subtitles': subtitles,
 658                     }
 659                     process_formats(info)
 660                     description = try_get(video, lambda x: x['savable_description']['text'])
 661                     title = video.get('name')
 662                     if title:
 663                         info.update({
 664                             'title': title,
 665                             'description': description,
 666                         })
 667                     else:
 668                         info['title'] = description or f'Facebook video #{v_id}'
 669                     entries.append(info)
 670
 671                 def parse_attachment(attachment, key='media'):
 672                     media = attachment.get(key) or {}
 673                     if media.get('__typename') == 'Video':
 674                         return parse_graphql_video(media)
 675
 676                 nodes = variadic(traverse_obj(data, 'nodes', 'node') or [])
 677                 attachments = traverse_obj(nodes, (
 678                     ..., 'comet_sections', 'content', 'story', (None, 'attached_story'), 'attachments',
 679                     ..., ('styles', 'style_type_renderer', ('throwbackStyles', 'attachment_target_renderer')),
 680                     'attachment', {dict}))
 681                 for attachment in attachments:
 682                     ns = traverse_obj(attachment, ('all_subattachments', 'nodes', ..., {dict}),
 683                                       ('target', 'attachments', ..., 'styles', 'attachment', {dict}))
 684                     for n in ns:
 685                         parse_attachment(n)
 686                     parse_attachment(attachment)
 687
 688                 edges = try_get(data, lambda x: x['mediaset']['currMedia']['edges'], list) or []
 689                 for edge in edges:
 690                     parse_attachment(edge, key='node')
 691
 692                 video = traverse_obj(data, (
 693                     'event', 'cover_media_renderer', 'cover_video'), 'video', expected_type=dict) or {}
 694                 if video:
 695                     attachments = try_get(video, [
 696                         lambda x: x['story']['attachments'],
 697                         lambda x: x['creation_story']['attachments'],
 698                     ], list) or []
 699                     for attachment in attachments:
 700                         parse_attachment(attachment)
 701                     if not entries:
 702                         parse_graphql_video(video)
 703
 704                 if len(entries) > 1:
 705                     return self.playlist_result(entries, video_id)
 706
 707                 video_info = entries[0] if entries else {'id': video_id}
 708                 webpage_info = extract_metadata(webpage)
 709                 # honor precise duration in video info
 710                 if video_info.get('duration'):
 711                     webpage_info['duration'] = video_info['duration']
 712                 # preserve preferred_thumbnail in video info
 713                 if video_info.get('thumbnail'):
 714                     webpage_info['thumbnail'] = video_info['thumbnail']
 715                 return merge_dicts(webpage_info, video_info)
 716
 717         if not video_data:
 718             m_msg = re.search(r'class="[^"]*uiInterstitialContent[^"]*"><div>(.*?)</div>', webpage)
 719             if m_msg is not None:
 720                 raise ExtractorError(
 721                     f'The video is not available, Facebook said: "{m_msg.group(1)}"',
 722                     expected=True)
 723             elif any(p in webpage for p in (
 724                     '>You must log in to continue',
 725                     'id="login_form"',
 726                     'id="loginbutton"')):
 727                 self.raise_login_required()
 728
 729         if not video_data and '/watchparty/' in url:
 730             post_data = {
 731                 'doc_id': 3731964053542869,
 732                 'variables': json.dumps({
 733                     'livingRoomID': video_id,
 734                 }),
 735             }
 736
 737             prefetched_data = extract_relay_prefetched_data(r'"login_data"\s*:\s*{')
 738             if prefetched_data:
 739                 lsd = try_get(prefetched_data, lambda x: x['login_data']['lsd'], dict)
 740                 if lsd:
 741                     post_data[lsd['name']] = lsd['value']
 742
 743             relay_data = extract_relay_data(r'\[\s*"RelayAPIConfigDefaults"\s*,')
 744             for define in (relay_data.get('define') or []):
 745                 if define[0] == 'RelayAPIConfigDefaults':
 746                     self._api_config = define[2]
 747
 748             living_room = self._download_json(
 749                 urljoin(url, self._api_config['graphURI']), video_id,
 750                 data=urlencode_postdata(post_data))['data']['living_room']
 751
 752             entries = []
 753             for edge in (try_get(living_room, lambda x: x['recap']['watched_content']['edges']) or []):
 754                 video = try_get(edge, lambda x: x['node']['video']) or {}
 755                 v_id = video.get('id')
 756                 if not v_id:
 757                     continue
 758                 v_id = str(v_id)
 759                 entries.append(self.url_result(
 760                     self._VIDEO_PAGE_TEMPLATE % v_id,
 761                     self.ie_key(), v_id, video.get('name')))
 762
 763             return self.playlist_result(entries, video_id)
 764
 765         if not video_data:
 766             # Video info not in first request, do a secondary request using
 767             # tahoe player specific URL
 768             tahoe_data = self._download_webpage(
 769                 self._VIDEO_PAGE_TAHOE_TEMPLATE % video_id, video_id,
 770                 data=urlencode_postdata({
 771                     '__a': 1,
 772                     '__pc': self._search_regex(
 773                         r'pkg_cohort["\']\s*:\s*["\'](.+?)["\']', webpage,
 774                         'pkg cohort', default='PHASED:DEFAULT'),
 775                     '__rev': self._search_regex(
 776                         r'client_revision["\']\s*:\s*(\d+),', webpage,
 777                         'client revision', default='3944515'),
 778                     'fb_dtsg': self._search_regex(
 779                         r'"DTSGInitialData"\s*,\s*\[\]\s*,\s*{\s*"token"\s*:\s*"([^"]+)"',
 780                         webpage, 'dtsg token', default=''),
 781                 }),
 782                 headers={
 783                     'Content-Type': 'application/x-www-form-urlencoded',
 784                 })
 785             tahoe_js_data = self._parse_json(
 786                 self._search_regex(
 787                     r'for\s+\(\s*;\s*;\s*\)\s*;(.+)', tahoe_data,
 788                     'tahoe js data', default='{}'),
 789                 video_id, fatal=False)
 790             video_data = extract_from_jsmods_instances(tahoe_js_data)
 791
 792         if not video_data:
 793             raise ExtractorError('Cannot parse data')
 794
 795         if len(video_data) > 1:
 796             entries = []
 797             for v in video_data:
 798                 video_url = v[0].get('video_url')
 799                 if not video_url:
 800                     continue
 801                 entries.append(self.url_result(urljoin(
 802                     url, video_url), self.ie_key(), v[0].get('video_id')))
 803             return self.playlist_result(entries, video_id)
 804         video_data = video_data[0]
 805
 806         formats = []
 807         subtitles = {}
 808         for f in video_data:
 809             format_id = f['stream_type']
 810             if f and isinstance(f, dict):
 811                 f = [f]
 812             if not f or not isinstance(f, list):
 813                 continue
 814             for quality in ('sd', 'hd'):
 815                 for src_type in ('src', 'src_no_ratelimit'):
 816                     src = f[0].get(f'{quality}_{src_type}')
 817                     if src:
 818                         # sd, hd formats w/o resolution info should be deprioritized below DASH
 819                         # TODO: investigate if progressive or src formats still exist
 820                         preference = -10 if format_id == 'progressive' else -3
 821                         if quality == 'hd':
 822                             preference += 1
 823                         formats.append({
 824                             'format_id': f'{format_id}_{quality}_{src_type}',
 825                             'url': src,
 826                             'quality': preference,
 827                             'height': 720 if quality == 'hd' else None,
 828                         })
 829             extract_dash_manifest(f[0], formats)
 830             subtitles_src = f[0].get('subtitles_src')
 831             if subtitles_src:
 832                 subtitles.setdefault('en', []).append({'url': subtitles_src})
 833
 834         info_dict = {
 835             'id': video_id,
 836             'formats': formats,
 837             'subtitles': subtitles,
 838         }
 839         process_formats(info_dict)
 840         info_dict.update(extract_metadata(webpage))
 841
 842         return info_dict
 843
 844     def _real_extract(self, url):
 845         video_id = self._match_id(url)
 846
 847         real_url = self._VIDEO_PAGE_TEMPLATE % video_id if url.startswith('facebook:') else url
 848         return self._extract_from_url(real_url, video_id)
 849
 850
 851 class FacebookPluginsVideoIE(InfoExtractor):
 852     _VALID_URL = r'https?://(?:[\w-]+\.)?facebook\.com/plugins/video\.php\?.*?\bhref=(?P<id>https.+)'
 853
 854     _TESTS = [{
 855         'url': 'https://www.facebook.com/plugins/video.php?href=https%3A%2F%2Fwww.facebook.com%2Fgov.sg%2Fvideos%2F10154383743583686%2F&show_text=0&width=560',
 856         'md5': '5954e92cdfe51fe5782ae9bda7058a07',
 857         'info_dict': {
 858             'id': '10154383743583686',
 859             'ext': 'mp4',
 860             # TODO: Fix title, uploader
 861             'title': 'What to do during the haze?',
 862             'uploader': 'Gov.sg',
 863             'upload_date': '20160826',
 864             'timestamp': 1472184808,
 865         },
 866         'add_ie': [FacebookIE.ie_key()],
 867     }, {
 868         'url': 'https://www.facebook.com/plugins/video.php?href=https%3A%2F%2Fwww.facebook.com%2Fvideo.php%3Fv%3D10204634152394104',
 869         'only_matching': True,
 870     }, {
 871         'url': 'https://www.facebook.com/plugins/video.php?href=https://www.facebook.com/gov.sg/videos/10154383743583686/&show_text=0&width=560',
 872         'only_matching': True,
 873     }]
 874
 875     def _real_extract(self, url):
 876         return self.url_result(
 877             urllib.parse.unquote(self._match_id(url)),
 878             FacebookIE.ie_key())
 879
 880
 881 class FacebookRedirectURLIE(InfoExtractor):
 882     IE_DESC = False  # Do not list
 883     _VALID_URL = r'https?://(?:[\w-]+\.)?facebook\.com/flx/warn[/?]'
 884     _TESTS = [{
 885         'url': 'https://www.facebook.com/flx/warn/?h=TAQHsoToz&u=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DpO8h3EaFRdo&s=1',
 886         'info_dict': {
 887             'id': 'pO8h3EaFRdo',
 888             'ext': 'mp4',
 889             'title': 'Tripeo Boiler Room x Dekmantel Festival DJ Set',
 890             'description': 'md5:2d713ccbb45b686a1888397b2c77ca6b',
 891             'channel_id': 'UCGBpxWJr9FNOcFYA5GkKrMg',
 892             'playable_in_embed': True,
 893             'categories': ['Music'],
 894             'channel': 'Boiler Room',
 895             'uploader_id': 'brtvofficial',
 896             'uploader': 'Boiler Room',
 897             'tags': 'count:11',
 898             'duration': 3332,
 899             'live_status': 'not_live',
 900             'thumbnail': 'https://i.ytimg.com/vi/pO8h3EaFRdo/maxresdefault.jpg',
 901             'channel_url': 'https://www.youtube.com/channel/UCGBpxWJr9FNOcFYA5GkKrMg',
 902             'availability': 'public',
 903             'uploader_url': 'http://www.youtube.com/user/brtvofficial',
 904             'upload_date': '20150917',
 905             'age_limit': 0,
 906             'view_count': int,
 907             'like_count': int,
 908         },
 909         'add_ie': ['Youtube'],
 910         'params': {'skip_download': 'Youtube'},
 911     }]
 912
 913     def _real_extract(self, url):
 914         redirect_url = url_or_none(parse_qs(url).get('u', [None])[-1])
 915         if not redirect_url:
 916             raise ExtractorError('Invalid facebook redirect URL', expected=True)
 917         return self.url_result(redirect_url)
 918
 919
 920 class FacebookReelIE(InfoExtractor):
 921     _VALID_URL = r'https?://(?:[\w-]+\.)?facebook\.com/reel/(?P<id>\d+)'
 922     IE_NAME = 'facebook:reel'
 923
 924     _TESTS = [{
 925         'url': 'https://www.facebook.com/reel/1195289147628387',
 926         'md5': 'f13dd37f2633595982db5ed8765474d3',
 927         'info_dict': {
 928             'id': '1195289147628387',
 929             'ext': 'mp4',
 930             'title': 'md5:b05800b5b1ad56c0ca78bd3807b6a61e',
 931             'description': 'md5:22f03309b216ac84720183961441d8db',
 932             'uploader': 'md5:723e6cb3091241160f20b3c5dc282af1',
 933             'uploader_id': '100040874179269',
 934             'duration': 9.579,
 935             'timestamp': 1637502609,
 936             'upload_date': '20211121',
 937             'thumbnail': r're:^https?://.*',
 938         },
 939     }]
 940
 941     def _real_extract(self, url):
 942         video_id = self._match_id(url)
 943         return self.url_result(
 944             f'https://m.facebook.com/watch/?v={video_id}&_rdr', FacebookIE, video_id)
 945
 946
 947 class FacebookAdsIE(InfoExtractor):
 948     _VALID_URL = r'https?://(?:[\w-]+\.)?facebook\.com/ads/library/?\?(?:[^#]+&)?id=(?P<id>\d+)'
 949     IE_NAME = 'facebook:ads'
 950
 951     _TESTS = [{
 952         'url': 'https://www.facebook.com/ads/library/?id=899206155126718',
 953         'info_dict': {
 954             'id': '899206155126718',
 955             'ext': 'mp4',
 956             'title': 'video by Kandao',
 957             'uploader': 'Kandao',
 958             'uploader_id': '774114102743284',
 959             'uploader_url': r're:^https?://.*',
 960             'timestamp': 1702548330,
 961             'thumbnail': r're:^https?://.*',
 962             'upload_date': '20231214',
 963             'like_count': int,
 964         },
 965     }, {
 966         'url': 'https://www.facebook.com/ads/library/?id=893637265423481',
 967         'info_dict': {
 968             'id': '893637265423481',
 969             'title': 'Jusqu\u2019\u00e0 -25% sur une s\u00e9lection de vins p\u00e9tillants italiens ',
 970             'uploader': 'Eataly Paris Marais',
 971             'uploader_id': '2086668958314152',
 972             'uploader_url': r're:^https?://.*',
 973             'timestamp': 1703571529,
 974             'upload_date': '20231226',
 975             'like_count': int,
 976         },
 977         'playlist_count': 3,
 978     }, {
 979         'url': 'https://es-la.facebook.com/ads/library/?id=901230958115569',
 980         'only_matching': True,
 981     }, {
 982         'url': 'https://m.facebook.com/ads/library/?id=901230958115569',
 983         'only_matching': True,
 984     }]
 985
 986     _FORMATS_MAP = {
 987         'watermarked_video_sd_url': ('sd-wmk', 'SD, watermarked'),
 988         'video_sd_url': ('sd', None),
 989         'watermarked_video_hd_url': ('hd-wmk', 'HD, watermarked'),
 990         'video_hd_url': ('hd', None),
 991     }
 992
 993     def _extract_formats(self, video_dict):
 994         formats = []
 995         for format_key, format_url in traverse_obj(video_dict, (
 996             {dict.items}, lambda _, v: v[0] in self._FORMATS_MAP and url_or_none(v[1]),
 997         )):
 998             formats.append({
 999                 'format_id': self._FORMATS_MAP[format_key][0],
1000                 'format_note': self._FORMATS_MAP[format_key][1],
1001                 'url': format_url,
1002                 'ext': 'mp4',
1003                 'quality': qualities(tuple(self._FORMATS_MAP))(format_key),
1004             })
1005         return formats
1006
1007     def _real_extract(self, url):
1008         video_id = self._match_id(url)
1009         webpage = self._download_webpage(url, video_id)
1010
1011         post_data = [self._parse_json(j, video_id, fatal=False)
1012                      for j in re.findall(r's\.handle\(({.*})\);requireLazy\(', webpage)]
1013         data = traverse_obj(post_data, (
1014             ..., 'require', ..., ..., ..., 'props', 'deeplinkAdCard', 'snapshot', {dict}), get_all=False)
1015         if not data:
1016             raise ExtractorError('Unable to extract ad data')
1017
1018         title = data.get('title')
1019         if not title or title == '{{product.name}}':
1020             title = join_nonempty('display_format', 'page_name', delim=' by ', from_dict=data)
1021
1022         info_dict = traverse_obj(data, {
1023             'description': ('link_description', {str}, {lambda x: x if x != '{{product.description}}' else None}),
1024             'uploader': ('page_name', {str}),
1025             'uploader_id': ('page_id', {str_or_none}),
1026             'uploader_url': ('page_profile_uri', {url_or_none}),
1027             'timestamp': ('creation_time', {int_or_none}),
1028             'like_count': ('page_like_count', {int_or_none}),
1029         })
1030
1031         entries = []
1032         for idx, entry in enumerate(traverse_obj(
1033             data, (('videos', 'cards'), lambda _, v: any(url_or_none(v[f]) for f in self._FORMATS_MAP))), 1,
1034         ):
1035             entries.append({
1036                 'id': f'{video_id}_{idx}',
1037                 'title': entry.get('title') or title,
1038                 'description': entry.get('link_description') or info_dict.get('description'),
1039                 'thumbnail': url_or_none(entry.get('video_preview_image_url')),
1040                 'formats': self._extract_formats(entry),
1041             })
1042
1043         if len(entries) == 1:
1044             info_dict.update(entries[0])
1045
1046         elif len(entries) > 1:
1047             info_dict.update({
1048                 'title': entries[0]['title'],
1049                 'entries': entries,
1050                 '_type': 'playlist',
1051             })
1052
1053         info_dict['id'] = video_id
1054
1055         return info_dict