yt_dlp/extractor/tumblr.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     ExtractorError,
   4     int_or_none,
   5     traverse_obj,
   6     url_or_none,
   7     urlencode_postdata,
   8 )
   9
  10
  11 class TumblrIE(InfoExtractor):
  12     _VALID_URL = r'https?://(?P<blog_name_1>[^/?#&]+)\.tumblr\.com/(?:post|video|(?P<blog_name_2>[a-zA-Z\d-]+))/(?P<id>[0-9]+)(?:$|[/?#])'
  13     _NETRC_MACHINE = 'tumblr'
  14     _LOGIN_URL = 'https://www.tumblr.com/login'
  15     _OAUTH_URL = 'https://www.tumblr.com/api/v2/oauth2/token'
  16     _TESTS = [{
  17         'url': 'http://tatianamaslanydaily.tumblr.com/post/54196191430/orphan-black-dvd-extra-behind-the-scenes',
  18         'md5': '479bb068e5b16462f5176a6828829767',
  19         'info_dict': {
  20             'id': '54196191430',
  21             'ext': 'mp4',
  22             'title': 'md5:dfac39636969fe6bf1caa2d50405f069',
  23             'description': 'md5:390ab77358960235b6937ab3b8528956',
  24             'uploader_id': 'tatianamaslanydaily',
  25             'uploader_url': 'https://tatianamaslanydaily.tumblr.com/',
  26             'thumbnail': r're:^https?://.*\.jpg',
  27             'duration': 127,
  28             'like_count': int,
  29             'repost_count': int,
  30             'age_limit': 0,
  31             'tags': ['Orphan Black', 'Tatiana Maslany', 'Interview', 'Video', 'OB S1 DVD Extras'],
  32         },
  33     }, {
  34         'note': 'multiple formats',
  35         'url': 'https://maskofthedragon.tumblr.com/post/626907179849564160/mona-talking-in-english',
  36         'md5': 'f43ff8a8861712b6cf0e0c2bd84cfc68',
  37         'info_dict': {
  38             'id': '626907179849564160',
  39             'ext': 'mp4',
  40             'title': 'Mona\xa0“talking” in\xa0“english”',
  41             'description': 'md5:082a3a621530cb786ad2b7592a6d9e2c',
  42             'uploader_id': 'maskofthedragon',
  43             'uploader_url': 'https://maskofthedragon.tumblr.com/',
  44             'thumbnail': r're:^https?://.*\.jpg',
  45             'duration': 7,
  46             'like_count': int,
  47             'repost_count': int,
  48             'age_limit': 0,
  49             'tags': 'count:19',
  50         },
  51         'params': {
  52             'format': 'hd',
  53         },
  54     }, {
  55         'note': 'non-iframe video (with related posts)',
  56         'url': 'https://shieldfoss.tumblr.com/post/675519763813908480',
  57         'md5': '12bdb75661ef443bffe5a4dac1dbf118',
  58         'info_dict': {
  59             'id': '675519763813908480',
  60             'ext': 'mp4',
  61             'title': 'Shieldfoss',
  62             'uploader_id': 'nerviovago',
  63             'uploader_url': 'https://nerviovago.tumblr.com/',
  64             'thumbnail': r're:^https?://.*\.jpg',
  65             'like_count': int,
  66             'repost_count': int,
  67             'age_limit': 0,
  68             'tags': [],
  69         },
  70         'skip': '404',
  71     }, {
  72         'note': 'dashboard only (original post)',
  73         'url': 'https://jujanon.tumblr.com/post/159704441298/my-baby-eating',
  74         'md5': '029f7c91ab386701b211e3d494d2d95e',
  75         'info_dict': {
  76             'id': '159704441298',
  77             'ext': 'mp4',
  78             'title': 'md5:ba79365861101f4911452728d2950561',
  79             'description': 'md5:773738196cea76b6996ec71e285bdabc',
  80             'uploader_id': 'jujanon',
  81             'uploader_url': 'https://jujanon.tumblr.com/',
  82             'thumbnail': r're:^https?://.*\.jpg',
  83             'like_count': int,
  84             'repost_count': int,
  85             'age_limit': 0,
  86             'tags': ['crabs', 'my video', 'my pets'],
  87         },
  88     }, {
  89         'note': 'dashboard only (reblog)',
  90         'url': 'https://bartlebyshop.tumblr.com/post/180294460076/duality-of-bird',
  91         'md5': '04334e7cadb1af680d162912559f51a5',
  92         'info_dict': {
  93             'id': '180294460076',
  94             'ext': 'mp4',
  95             'title': 'duality of bird',
  96             'description': 'duality of bird',
  97             'uploader_id': 'todaysbird',
  98             'uploader_url': 'https://todaysbird.tumblr.com/',
  99             'thumbnail': r're:^https?://.*\.jpg',
 100             'like_count': int,
 101             'repost_count': int,
 102             'age_limit': 0,
 103         },
 104     }, {
 105         'note': 'dashboard only (external)',
 106         'url': 'https://afloweroutofstone.tumblr.com/post/675661759168823296/the-blues-remembers-everything-the-country-forgot',
 107         'info_dict': {
 108             'id': 'q67_fd7b8SU',
 109             'ext': 'mp4',
 110             'title': 'The Blues Remembers Everything the Country Forgot',
 111             'alt_title': 'The Blues Remembers Everything the Country Forgot',
 112             'description': 'md5:1a6b4097e451216835a24c1023707c79',
 113             'creator': 'md5:c2239ba15430e87c3b971ba450773272',
 114             'uploader': 'Moor Mother - Topic',
 115             'upload_date': '20201223',
 116             'uploader_id': 'UCxrMtFBRkFvQJ_vVM4il08w',
 117             'uploader_url': 'http://www.youtube.com/channel/UCxrMtFBRkFvQJ_vVM4il08w',
 118             'thumbnail': r're:^https?://i.ytimg.com/.*',
 119             'channel': 'Moor Mother',
 120             'channel_id': 'UCxrMtFBRkFvQJ_vVM4il08w',
 121             'channel_url': 'https://www.youtube.com/channel/UCxrMtFBRkFvQJ_vVM4il08w',
 122             'channel_follower_count': int,
 123             'duration': 181,
 124             'view_count': int,
 125             'like_count': int,
 126             'age_limit': 0,
 127             'categories': ['Music'],
 128             'tags': 'count:7',
 129             'live_status': 'not_live',
 130             'playable_in_embed': True,
 131             'availability': 'public',
 132             'track': 'The Blues Remembers Everything the Country Forgot',
 133             'artist': 'md5:c2239ba15430e87c3b971ba450773272',
 134             'album': 'Brass',
 135             'release_year': 2020,
 136         },
 137         'add_ie': ['Youtube'],
 138         'skip': 'Video Unavailable',
 139     }, {
 140         'url': 'https://prozdvoices.tumblr.com/post/673201091169681408/what-recording-voice-acting-sounds-like',
 141         'md5': 'cb8328a6723c30556cef59e370202918',
 142         'info_dict': {
 143             'id': 'eomhW5MLGWA',
 144             'ext': 'mp4',
 145             'title': 'what recording voice acting sounds like',
 146             'description': 'md5:1da3faa22d0e0b1d8b50216c284ee798',
 147             'uploader': 'ProZD',
 148             'upload_date': '20220112',
 149             'uploader_id': '@ProZD',
 150             'uploader_url': 'https://www.youtube.com/@ProZD',
 151             'thumbnail': r're:^https?://i.ytimg.com/.*',
 152             'channel': 'ProZD',
 153             'channel_id': 'UC6MFZAOHXlKK1FI7V0XQVeA',
 154             'channel_url': 'https://www.youtube.com/channel/UC6MFZAOHXlKK1FI7V0XQVeA',
 155             'channel_follower_count': int,
 156             'duration': 20,
 157             'view_count': int,
 158             'like_count': int,
 159             'age_limit': 0,
 160             'categories': ['Film & Animation'],
 161             'tags': [],
 162             'live_status': 'not_live',
 163             'playable_in_embed': True,
 164             'availability': 'public',
 165             'heatmap': 'count:100',
 166             'channel_is_verified': True,
 167             'timestamp': 1642014562,
 168             'comment_count': int,
 169         },
 170         'add_ie': ['Youtube'],
 171     }, {
 172         'url': 'https://dominustempori.tumblr.com/post/673572712813297664/youtubes-all-right-for-some-pretty-cool',
 173         'md5': '203e9eb8077e3f45bfaeb4c86c1467b8',
 174         'info_dict': {
 175             'id': '87816359',
 176             'ext': 'mp4',
 177             'title': 'Harold Ramis',
 178             'description': 'md5:c99882405fcca0b1d348ad093f8f1672',
 179             'uploader': 'Resolution Productions Group',
 180             'uploader_id': 'resolutionproductions',
 181             'uploader_url': 'https://vimeo.com/resolutionproductions',
 182             'upload_date': '20140227',
 183             'thumbnail': r're:^https?://i.vimeocdn.com/video/.*',
 184             'timestamp': 1393541719,
 185             'duration': 291,
 186             'comment_count': int,
 187             'like_count': int,
 188             'release_timestamp': 1393541719,
 189             'release_date': '20140227',
 190         },
 191         'add_ie': ['Vimeo'],
 192     }, {
 193         'url': 'http://sutiblr.tumblr.com/post/139638707273',
 194         'md5': '2dd184b3669e049ba40563a7d423f95c',
 195         'info_dict': {
 196             'id': 'ir7qBEIKqvq',
 197             'ext': 'mp4',
 198             'title': 'Vine by sutiblr',
 199             'alt_title': 'Vine by sutiblr',
 200             'uploader': 'sutiblr',
 201             'uploader_id': '1198993975374495744',
 202             'upload_date': '20160220',
 203             'like_count': int,
 204             'comment_count': int,
 205             'repost_count': int,
 206             'thumbnail': r're:^https?://.*\.jpg',
 207             'timestamp': 1455940159,
 208             'view_count': int,
 209         },
 210         'add_ie': ['Vine'],
 211         'skip': 'Vine is unavailable',
 212     }, {
 213         'url': 'https://silami.tumblr.com/post/84250043974/my-bad-river-flows-in-you-impression-on-maschine',
 214         'md5': '3c92d7c3d867f14ccbeefa2119022277',
 215         'info_dict': {
 216             'id': 'nYtvtTPuTl',
 217             'ext': 'mp4',
 218             'title': 'Video by silbulterman',
 219             'description': '#maschine',
 220             'uploader_id': '242859024',
 221             'thumbnail': r're:^https?://.*\.jpg',
 222             'timestamp': 1398801174,
 223             'like_count': int,
 224             'uploader': 'Sil',
 225             'channel': 'silbulterman',
 226             'comment_count': int,
 227             'upload_date': '20140429',
 228         },
 229         'add_ie': ['Instagram'],
 230     }, {
 231         'note': 'new url scheme',
 232         'url': 'https://www.tumblr.com/autumnsister/765162750456578048?source=share',
 233         'info_dict': {
 234             'id': '765162750456578048',
 235             'ext': 'mp4',
 236             'uploader_url': 'https://autumnsister.tumblr.com/',
 237             'tags': ['autumn', 'food', 'curators on tumblr'],
 238             'like_count': int,
 239             'thumbnail': 'https://64.media.tumblr.com/tumblr_sklad89N3x1ygquow_frame1.jpg',
 240             'title': '🪹',
 241             'uploader_id': 'autumnsister',
 242             'repost_count': int,
 243             'age_limit': 0,
 244         },
 245     }, {
 246         'note': 'bandcamp album embed',
 247         'url': 'https://patricia-taxxon.tumblr.com/post/704473755725004800/patricia-taxxon-agnes-hilda-patricia-taxxon',
 248         'info_dict': {
 249             'id': 'agnes-hilda',
 250             'title': 'Agnes & Hilda',
 251             'description': 'The inexplicable joy of an artist. Wash paws after listening.',
 252             'uploader_id': 'patriciataxxon',
 253         },
 254         'playlist_count': 8,
 255     }, {
 256         'note': 'bandcamp track embeds (many)',
 257         'url': 'https://www.tumblr.com/felixcosm/730460905855467520/if-youre-looking-for-new-music-to-write-or',
 258         'info_dict': {
 259             'id': '730460905855467520',
 260             'uploader_id': 'felixcosm',
 261             'repost_count': int,
 262             'tags': 'count:15',
 263             'description': 'md5:2eb3482a3c6987280cbefb6839068f32',
 264             'like_count': int,
 265             'age_limit': 0,
 266             'title': 'If you\'re looking for new music to write or imagine scenerios to: STOP. This is for you.',
 267             'uploader_url': 'https://felixcosm.tumblr.com/',
 268         },
 269         'playlist_count': 10,
 270     }, {
 271         'note': 'soundcloud track embed',
 272         'url': 'https://silverfoxstole.tumblr.com/post/765305403763556352/jamie-robertson-doctor-who-8th-doctor',
 273         'info_dict': {
 274             'id': '1218136399',
 275             'ext': 'opus',
 276             'comment_count': int,
 277             'genres': [],
 278             'repost_count': int,
 279             'uploader': 'Jamie Robertson',
 280             'title': 'Doctor Who - 8th doctor -   Stranded Theme never released and used.',
 281             'duration': 46.106,
 282             'uploader_id': '2731064',
 283             'thumbnail': 'https://i1.sndcdn.com/artworks-MVgcPm5jN42isC5M-6Dz22w-original.jpg',
 284             'timestamp': 1645181261,
 285             'uploader_url': 'https://soundcloud.com/jamierobertson',
 286             'view_count': int,
 287             'upload_date': '20220218',
 288             'description': 'md5:ab924dd9994d0a7d64d6d31bf2af4625',
 289             'license': 'all-rights-reserved',
 290             'like_count': int,
 291         },
 292     }, {
 293         'note': 'soundcloud set embed',
 294         'url': 'https://www.tumblr.com/beyourselfchulanmaria/703505323122638848/chu-lan-maria-the-playlist-%E5%BF%83%E7%9A%84%E5%91%BC%E5%96%9A-call-of-the',
 295         'info_dict': {
 296             'id': '691222680',
 297             'title': '心的呼喚 Call of the heart I',
 298             'description': 'md5:25952a8d178a3aa55e40fcbb646a38c3',
 299         },
 300         'playlist_mincount': 19,
 301     }, {
 302         'note': 'dailymotion video embed',
 303         'url': 'https://www.tumblr.com/funvibecentral/759390024460632064',
 304         'info_dict': {
 305             'id': 'x94cnnk',
 306             'ext': 'mp4',
 307             'description': 'Funny dailymotion shorts.\n#funny #fun#comedy #romantic #exciting',
 308             'uploader': 'FunVibe Central',
 309             'like_count': int,
 310             'view_count': int,
 311             'timestamp': 1724210553,
 312             'title': 'Woman watching other Woman',
 313             'tags': [],
 314             'upload_date': '20240821',
 315             'age_limit': 0,
 316             'uploader_id': 'x32m6ye',
 317             'duration': 20,
 318             'thumbnail': r're:https://(?:s[12]\.)dmcdn\.net/v/Wtqh01cnxKNXLG1N8/x1080',
 319         },
 320     }, {
 321         'note': 'tiktok video embed',
 322         'url': 'https://fansofcolor.tumblr.com/post/660637918605475840/blockquote-class-tiktok-embed',
 323         'info_dict': {
 324             'id': '7000937272010935558',
 325             'ext': 'mp4',
 326             'artists': ['Alicia Dreaming'],
 327             'like_count': int,
 328             'repost_count': int,
 329             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 330             'channel_id': 'MS4wLjABAAAAsJohwz_dU4KfAOc61cbGDAZ46-5hg2ANTXVQlRe1ipDhpX08PywR3PPiple1NTAo',
 331             'uploader': 'aliciadreaming',
 332             'description': 'huge casting news Greyworm will be  #louisdulac #racebending #interviewwiththevampire',
 333             'title': 'huge casting news Greyworm will be  #louisdulac #racebending #interviewwiththevampire',
 334             'channel_url': 'https://www.tiktok.com/@MS4wLjABAAAAsJohwz_dU4KfAOc61cbGDAZ46-5hg2ANTXVQlRe1ipDhpX08PywR3PPiple1NTAo',
 335             'uploader_id': '7000478462196990982',
 336             'uploader_url': 'https://www.tiktok.com/@aliciadreaming',
 337             'timestamp': 1630032733,
 338             'channel': 'Alicia Dreaming',
 339             'track': 'original sound',
 340             'upload_date': '20210827',
 341             'view_count': int,
 342             'comment_count': int,
 343             'duration': 59,
 344         },
 345     }, {
 346         'note': 'tumblr video AND youtube embed',
 347         'url': 'https://www.tumblr.com/anyaboz/765332564457209856/my-music-video-for-selkie-by-nobodys-wolf-child',
 348         'info_dict': {
 349             'id': '765332564457209856',
 350             'uploader_id': 'anyaboz',
 351             'repost_count': int,
 352             'age_limit': 0,
 353             'uploader_url': 'https://anyaboz.tumblr.com/',
 354             'description': 'md5:9a129cf6ce9d87a80ffd3c6dedd4d1e6',
 355             'like_count': int,
 356             'title': 'md5:b18a2ac9387681d20303e485db85c1b5',
 357             'tags': ['music video', 'nobodys wolf child', 'selkie', 'Stop Motion Animation', 'stop Motion', 'room guardians', 'Youtube'],
 358         },
 359         'playlist_count': 2,
 360     }, {
 361         # twitch_live provider - error when linked account is not live
 362         'url': 'https://www.tumblr.com/anarcho-skamunist/722224493650722816/hollow-knight-stream-right-now-going-to-fight',
 363         'only_matching': True,
 364     }]
 365
 366     _providers = {
 367         'instagram': 'Instagram',
 368         'vimeo': 'Vimeo',
 369         'vine': 'Vine',
 370         'youtube': 'Youtube',
 371         'dailymotion': 'Dailymotion',
 372         'tiktok': 'TikTok',
 373         'twitch_live': 'TwitchStream',
 374         'bandcamp': None,
 375         'soundcloud': None,
 376     }
 377     # known not to be supported
 378     _unsupported_providers = {
 379         # seems like podcasts can't be embedded
 380         'spotify',
 381     }
 382
 383     _ACCESS_TOKEN = None
 384
 385     def _initialize_pre_login(self):
 386         login_page = self._download_webpage(
 387             self._LOGIN_URL, None, 'Downloading login page', fatal=False)
 388         if login_page:
 389             self._ACCESS_TOKEN = self._search_regex(
 390                 r'"API_TOKEN":\s*"(\w+)"', login_page, 'API access token', fatal=False)
 391         if not self._ACCESS_TOKEN:
 392             self.report_warning('Failed to get access token; metadata will be missing and some videos may not work')
 393
 394     def _perform_login(self, username, password):
 395         if not self._ACCESS_TOKEN:
 396             return
 397
 398         data = {
 399             'password': password,
 400             'grant_type': 'password',
 401             'username': username,
 402         }
 403         if self.get_param('twofactor'):
 404             data['tfa_token'] = self.get_param('twofactor')
 405
 406         def _call_login():
 407             return self._download_json(
 408                 self._OAUTH_URL, None, 'Logging in',
 409                 data=urlencode_postdata(data),
 410                 headers={
 411                     'Content-Type': 'application/x-www-form-urlencoded',
 412                     'Authorization': f'Bearer {self._ACCESS_TOKEN}',
 413                 },
 414                 errnote='Login failed', fatal=False,
 415                 expected_status=lambda s: 400 <= s < 500)
 416
 417         response = _call_login()
 418         if traverse_obj(response, 'error') == 'tfa_required':
 419             data['tfa_token'] = self._get_tfa_info()
 420             response = _call_login()
 421         if traverse_obj(response, 'error'):
 422             raise ExtractorError(
 423                 f'API returned error {": ".join(traverse_obj(response, (("error", "error_description"), {str})))}')
 424
 425     def _real_extract(self, url):
 426         blog_1, blog_2, video_id = self._match_valid_url(url).groups()
 427         blog = blog_2 or blog_1
 428
 429         url = f'http://{blog}.tumblr.com/post/{video_id}'
 430         webpage, urlh = self._download_webpage_handle(
 431             url, video_id, headers={'User-Agent': 'WhatsApp/2.0'})  # whatsapp ua bypasses problems
 432
 433         redirect_url = urlh.url
 434
 435         api_only = bool(self._search_regex(
 436             r'(tumblr.com|^)/(safe-mode|login_required|blog/view)',
 437             redirect_url, 'redirect', default=None))
 438
 439         if api_only and not self._ACCESS_TOKEN:
 440             raise ExtractorError('Cannot get data for dashboard-only post without access token')
 441
 442         post_json = {}
 443         if self._ACCESS_TOKEN:
 444             post_json = traverse_obj(
 445                 self._download_json(
 446                     f'https://www.tumblr.com/api/v2/blog/{blog}/posts/{video_id}/permalink',
 447                     video_id, headers={'Authorization': f'Bearer {self._ACCESS_TOKEN}'}, fatal=False),
 448                 ('response', 'timeline', 'elements', 0, {dict})) or {}
 449         content_json = traverse_obj(post_json, ((('trail', 0), None), 'content', ..., {dict}))
 450
 451         # the url we're extracting from might be an original post or it might be a reblog.
 452         # if it's a reblog, og:description will be the reblogger's comment, not the uploader's.
 453         # content_json is always the op, so if it exists but has no text, there's no description
 454         if content_json:
 455             description = '\n\n'.join(
 456                 item.get('text') for item in content_json if item.get('type') == 'text') or None
 457         else:
 458             description = self._og_search_description(webpage, default=None)
 459         uploader_id = traverse_obj(post_json, 'reblogged_root_name', 'blog_name')
 460
 461         info_dict = {
 462             'id': video_id,
 463             'title': post_json.get('summary') or (blog if api_only else self._html_search_regex(
 464                 r'(?s)<title>(?P<title>.*?)(?: \| Tumblr)?</title>', webpage, 'title', default=blog)),
 465             'description': description,
 466             'uploader_id': uploader_id,
 467             'uploader_url': f'https://{uploader_id}.tumblr.com/' if uploader_id else None,
 468             **traverse_obj(post_json, {
 469                 'like_count': ('like_count', {int_or_none}),
 470                 'repost_count': ('reblog_count', {int_or_none}),
 471                 'tags': ('tags', ..., {str}),
 472             }),
 473             'age_limit': {True: 18, False: 0}.get(post_json.get('is_nsfw')),
 474         }
 475
 476         # for tumblr's own video hosting
 477         fallback_format = None
 478         formats = []
 479         video_url = self._og_search_video_url(webpage, default=None)
 480         # for external video hosts
 481         entries = []
 482         ignored_providers = set()
 483         unknown_providers = set()
 484
 485         for video_json in traverse_obj(content_json, lambda _, v: v['type'] in ('video', 'audio')):
 486             media_json = video_json.get('media') or {}
 487             if api_only and not media_json.get('url') and not video_json.get('url'):
 488                 raise ExtractorError('Failed to find video data for dashboard-only post')
 489             provider = video_json.get('provider')
 490
 491             if provider in ('tumblr', None):
 492                 fallback_format = {
 493                     'url': media_json.get('url') or video_url,
 494                     'width': int_or_none(
 495                         media_json.get('width') or self._og_search_property('video:width', webpage, default=None)),
 496                     'height': int_or_none(
 497                         media_json.get('height') or self._og_search_property('video:height', webpage, default=None)),
 498                 }
 499                 continue
 500             elif provider in self._unsupported_providers:
 501                 ignored_providers.add(provider)
 502                 continue
 503             elif provider and provider not in self._providers:
 504                 unknown_providers.add(provider)
 505             if video_json.get('url'):
 506                 # external video host
 507                 entries.append(self.url_result(
 508                     video_json['url'], self._providers.get(provider)))
 509
 510         duration = None
 511
 512         # iframes can supply duration and sometimes additional formats, so check for one
 513         iframe_url = self._search_regex(
 514             fr'src=\'(https?://www\.tumblr\.com/video/{blog}/{video_id}/[^\']+)\'',
 515             webpage, 'iframe url', default=None)
 516         if iframe_url:
 517             iframe = self._download_webpage(
 518                 iframe_url, video_id, 'Downloading iframe page',
 519                 headers={'Referer': redirect_url})
 520
 521             options = self._parse_json(
 522                 self._search_regex(
 523                     r'data-crt-options=(["\'])(?P<options>.+?)\1', iframe,
 524                     'hd video url', default='', group='options'),
 525                 video_id, fatal=False)
 526             if options:
 527                 duration = int_or_none(options.get('duration'))
 528
 529                 hd_url = options.get('hdUrl')
 530                 if hd_url:
 531                     # there are multiple formats; extract them
 532                     # ignore other sources of width/height data as they may be wrong
 533                     sources = []
 534                     sd_url = self._search_regex(
 535                         r'<source[^>]+src=(["\'])(?P<url>.+?)\1', iframe,
 536                         'sd video url', default=None, group='url')
 537                     if sd_url:
 538                         sources.append((sd_url, 'sd'))
 539                     sources.append((hd_url, 'hd'))
 540
 541                     formats = [{
 542                         'url': video_url,
 543                         'format_id': format_id,
 544                         'height': int_or_none(self._search_regex(
 545                             r'_(\d+)\.\w+$', video_url, 'height', default=None)),
 546                         'quality': quality,
 547                     } for quality, (video_url, format_id) in enumerate(sources)]
 548
 549         if not formats and fallback_format:
 550             formats.append(fallback_format)
 551
 552         if formats:
 553             # tumblr's own video is always above embeds
 554             entries.insert(0, {
 555                 **info_dict,
 556                 'formats': formats,
 557                 'duration': duration,
 558                 'thumbnail': (traverse_obj(video_json, ('poster', 0, 'url', {url_or_none}))
 559                               or self._og_search_thumbnail(webpage, default=None)),
 560             })
 561
 562         if ignored_providers:
 563             if not entries:
 564                 raise ExtractorError(f'None of embed providers are supported: {", ".join(ignored_providers)!s}', video_id=video_id, expected=True)
 565             else:
 566                 self.report_warning(f'Skipped embeds from unsupported providers: {", ".join(ignored_providers)!s}', video_id)
 567         if unknown_providers:
 568             self.report_warning(f'Unrecognized providers, please report: {", ".join(unknown_providers)!s}', video_id)
 569
 570         if not entries:
 571             self.raise_no_formats('No video could be found in this post', expected=True, video_id=video_id)
 572         if len(entries) == 1:
 573             return {
 574                 **info_dict,
 575                 **entries[0],
 576             }
 577         return {
 578             **info_dict,
 579             '_type': 'playlist',
 580             'entries': entries,
 581         }