yt_dlp/extractor/twitter.py

   1 import functools
   2 import json
   3 import random
   4 import re
   5 import urllib.parse
   6
   7 from .common import InfoExtractor
   8 from .periscope import PeriscopeBaseIE, PeriscopeIE
   9 from ..networking.exceptions import HTTPError
  10 from ..utils import (
  11     ExtractorError,
  12     dict_get,
  13     filter_dict,
  14     float_or_none,
  15     format_field,
  16     int_or_none,
  17     join_nonempty,
  18     make_archive_id,
  19     remove_end,
  20     str_or_none,
  21     strip_or_none,
  22     traverse_obj,
  23     try_call,
  24     try_get,
  25     unified_timestamp,
  26     update_url_query,
  27     url_or_none,
  28     xpath_text,
  29 )
  30
  31
  32 class TwitterBaseIE(InfoExtractor):
  33     _NETRC_MACHINE = 'twitter'
  34     _API_BASE = 'https://api.x.com/1.1/'
  35     _GRAPHQL_API_BASE = 'https://x.com/i/api/graphql/'
  36     _BASE_REGEX = r'https?://(?:(?:www|m(?:obile)?)\.)?(?:(?:twitter|x)\.com|twitter3e4tixl4xyajtrzo62zg5vztmjuricljdp2c5kshju4avyoid\.onion)/'
  37     _AUTH = 'AAAAAAAAAAAAAAAAAAAAANRILgAAAAAAnNwIzUejRCOuH5E6I8xnZz4puTs%3D1Zv7ttfk8LF81IUq16cHjhLTvJu4FA33AGWWjCpTnA'
  38     _LEGACY_AUTH = 'AAAAAAAAAAAAAAAAAAAAAIK1zgAAAAAA2tUWuhGZ2JceoId5GwYWU5GspY4%3DUq7gzFoCZs1QfwGoVdvSac3IniczZEYXIcDyumCauIXpcAPorE'
  39     _flow_token = None
  40
  41     _LOGIN_INIT_DATA = json.dumps({
  42         'input_flow_data': {
  43             'flow_context': {
  44                 'debug_overrides': {},
  45                 'start_location': {
  46                     'location': 'unknown',
  47                 },
  48             },
  49         },
  50         'subtask_versions': {
  51             'action_list': 2,
  52             'alert_dialog': 1,
  53             'app_download_cta': 1,
  54             'check_logged_in_account': 1,
  55             'choice_selection': 3,
  56             'contacts_live_sync_permission_prompt': 0,
  57             'cta': 7,
  58             'email_verification': 2,
  59             'end_flow': 1,
  60             'enter_date': 1,
  61             'enter_email': 2,
  62             'enter_password': 5,
  63             'enter_phone': 2,
  64             'enter_recaptcha': 1,
  65             'enter_text': 5,
  66             'enter_username': 2,
  67             'generic_urt': 3,
  68             'in_app_notification': 1,
  69             'interest_picker': 3,
  70             'js_instrumentation': 1,
  71             'menu_dialog': 1,
  72             'notifications_permission_prompt': 2,
  73             'open_account': 2,
  74             'open_home_timeline': 1,
  75             'open_link': 1,
  76             'phone_verification': 4,
  77             'privacy_options': 1,
  78             'security_key': 3,
  79             'select_avatar': 4,
  80             'select_banner': 2,
  81             'settings_list': 7,
  82             'show_code': 1,
  83             'sign_up': 2,
  84             'sign_up_review': 4,
  85             'tweet_selection_urt': 1,
  86             'update_users': 1,
  87             'upload_media': 1,
  88             'user_recommendations_list': 4,
  89             'user_recommendations_urt': 1,
  90             'wait_spinner': 3,
  91             'web_modal': 1,
  92         },
  93     }, separators=(',', ':')).encode()
  94
  95     def _extract_variant_formats(self, variant, video_id):
  96         variant_url = variant.get('url')
  97         if not variant_url:
  98             return [], {}
  99         elif '.m3u8' in variant_url:
 100             fmts, subs = self._extract_m3u8_formats_and_subtitles(
 101                 variant_url, video_id, 'mp4', 'm3u8_native',
 102                 m3u8_id='hls', fatal=False)
 103             for f in traverse_obj(fmts, lambda _, v: v['vcodec'] == 'none' and v.get('tbr') is None):
 104                 if mobj := re.match(r'hls-[Aa]udio-(?P<bitrate>\d{4,})', f['format_id']):
 105                     f['tbr'] = int_or_none(mobj.group('bitrate'), 1000)
 106             return fmts, subs
 107         else:
 108             tbr = int_or_none(dict_get(variant, ('bitrate', 'bit_rate')), 1000) or None
 109             f = {
 110                 'url': variant_url,
 111                 'format_id': join_nonempty('http', tbr),
 112                 'tbr': tbr,
 113             }
 114             self._search_dimensions_in_video_url(f, variant_url)
 115             return [f], {}
 116
 117     def _extract_formats_from_vmap_url(self, vmap_url, video_id):
 118         vmap_url = url_or_none(vmap_url)
 119         if not vmap_url:
 120             return [], {}
 121         vmap_data = self._download_xml(vmap_url, video_id)
 122         formats = []
 123         subtitles = {}
 124         urls = []
 125         for video_variant in vmap_data.findall('.//{http://twitter.com/schema/videoVMapV2.xsd}videoVariant'):
 126             video_variant.attrib['url'] = urllib.parse.unquote(
 127                 video_variant.attrib['url'])
 128             urls.append(video_variant.attrib['url'])
 129             fmts, subs = self._extract_variant_formats(
 130                 video_variant.attrib, video_id)
 131             formats.extend(fmts)
 132             subtitles = self._merge_subtitles(subtitles, subs)
 133         video_url = strip_or_none(xpath_text(vmap_data, './/MediaFile'))
 134         if video_url not in urls:
 135             fmts, subs = self._extract_variant_formats({'url': video_url}, video_id)
 136             formats.extend(fmts)
 137             subtitles = self._merge_subtitles(subtitles, subs)
 138         return formats, subtitles
 139
 140     @staticmethod
 141     def _search_dimensions_in_video_url(a_format, video_url):
 142         m = re.search(r'/(?P<width>\d+)x(?P<height>\d+)/', video_url)
 143         if m:
 144             a_format.update({
 145                 'width': int(m.group('width')),
 146                 'height': int(m.group('height')),
 147             })
 148
 149     @property
 150     def is_logged_in(self):
 151         return bool(self._get_cookies(self._API_BASE).get('auth_token'))
 152
 153     @functools.cached_property
 154     def _selected_api(self):
 155         return self._configuration_arg('api', ['graphql'], ie_key='Twitter')[0]
 156
 157     def _fetch_guest_token(self, display_id):
 158         guest_token = traverse_obj(self._download_json(
 159             f'{self._API_BASE}guest/activate.json', display_id, 'Downloading guest token', data=b'',
 160             headers=self._set_base_headers(legacy=display_id and self._selected_api == 'legacy')),
 161             ('guest_token', {str}))
 162         if not guest_token:
 163             raise ExtractorError('Could not retrieve guest token')
 164         return guest_token
 165
 166     def _set_base_headers(self, legacy=False):
 167         bearer_token = self._LEGACY_AUTH if legacy and not self.is_logged_in else self._AUTH
 168         return filter_dict({
 169             'Authorization': f'Bearer {bearer_token}',
 170             'x-csrf-token': try_call(lambda: self._get_cookies(self._API_BASE)['ct0'].value),
 171         })
 172
 173     def _call_login_api(self, note, headers, query={}, data=None):
 174         response = self._download_json(
 175             f'{self._API_BASE}onboarding/task.json', None, note,
 176             headers=headers, query=query, data=data, expected_status=400)
 177         error = traverse_obj(response, ('errors', 0, 'message', {str}))
 178         if error:
 179             raise ExtractorError(f'Login failed, Twitter API says: {error}', expected=True)
 180         elif traverse_obj(response, 'status') != 'success':
 181             raise ExtractorError('Login was unsuccessful')
 182
 183         subtask = traverse_obj(
 184             response, ('subtasks', ..., 'subtask_id', {str}), get_all=False)
 185         if not subtask:
 186             raise ExtractorError('Twitter API did not return next login subtask')
 187
 188         self._flow_token = response['flow_token']
 189
 190         return subtask
 191
 192     def _perform_login(self, username, password):
 193         if self.is_logged_in:
 194             return
 195
 196         guest_token = self._fetch_guest_token(None)
 197         headers = {
 198             **self._set_base_headers(),
 199             'content-type': 'application/json',
 200             'x-guest-token': guest_token,
 201             'x-twitter-client-language': 'en',
 202             'x-twitter-active-user': 'yes',
 203             'Referer': 'https://x.com/',
 204             'Origin': 'https://x.com',
 205         }
 206
 207         def build_login_json(*subtask_inputs):
 208             return json.dumps({
 209                 'flow_token': self._flow_token,
 210                 'subtask_inputs': subtask_inputs,
 211             }, separators=(',', ':')).encode()
 212
 213         def input_dict(subtask_id, text):
 214             return {
 215                 'subtask_id': subtask_id,
 216                 'enter_text': {
 217                     'text': text,
 218                     'link': 'next_link',
 219                 },
 220             }
 221
 222         next_subtask = self._call_login_api(
 223             'Downloading flow token', headers, query={'flow_name': 'login'}, data=self._LOGIN_INIT_DATA)
 224
 225         while not self.is_logged_in:
 226             if next_subtask == 'LoginJsInstrumentationSubtask':
 227                 next_subtask = self._call_login_api(
 228                     'Submitting JS instrumentation response', headers, data=build_login_json({
 229                         'subtask_id': next_subtask,
 230                         'js_instrumentation': {
 231                             'response': '{}',
 232                             'link': 'next_link',
 233                         },
 234                     }))
 235
 236             elif next_subtask == 'LoginEnterUserIdentifierSSO':
 237                 next_subtask = self._call_login_api(
 238                     'Submitting username', headers, data=build_login_json({
 239                         'subtask_id': next_subtask,
 240                         'settings_list': {
 241                             'setting_responses': [{
 242                                 'key': 'user_identifier',
 243                                 'response_data': {
 244                                     'text_data': {
 245                                         'result': username,
 246                                     },
 247                                 },
 248                             }],
 249                             'link': 'next_link',
 250                         },
 251                     }))
 252
 253             elif next_subtask == 'LoginEnterAlternateIdentifierSubtask':
 254                 next_subtask = self._call_login_api(
 255                     'Submitting alternate identifier', headers,
 256                     data=build_login_json(input_dict(next_subtask, self._get_tfa_info(
 257                         'one of username, phone number or email that was not used as --username'))))
 258
 259             elif next_subtask == 'LoginEnterPassword':
 260                 next_subtask = self._call_login_api(
 261                     'Submitting password', headers, data=build_login_json({
 262                         'subtask_id': next_subtask,
 263                         'enter_password': {
 264                             'password': password,
 265                             'link': 'next_link',
 266                         },
 267                     }))
 268
 269             elif next_subtask == 'AccountDuplicationCheck':
 270                 next_subtask = self._call_login_api(
 271                     'Submitting account duplication check', headers, data=build_login_json({
 272                         'subtask_id': next_subtask,
 273                         'check_logged_in_account': {
 274                             'link': 'AccountDuplicationCheck_false',
 275                         },
 276                     }))
 277
 278             elif next_subtask == 'LoginTwoFactorAuthChallenge':
 279                 next_subtask = self._call_login_api(
 280                     'Submitting 2FA token', headers, data=build_login_json(input_dict(
 281                         next_subtask, self._get_tfa_info('two-factor authentication token'))))
 282
 283             elif next_subtask == 'LoginAcid':
 284                 next_subtask = self._call_login_api(
 285                     'Submitting confirmation code', headers, data=build_login_json(input_dict(
 286                         next_subtask, self._get_tfa_info('confirmation code sent to your email or phone'))))
 287
 288             elif next_subtask == 'ArkoseLogin':
 289                 self.raise_login_required('Twitter is requiring captcha for this login attempt', method='cookies')
 290
 291             elif next_subtask == 'DenyLoginSubtask':
 292                 self.raise_login_required('Twitter rejected this login attempt as suspicious', method='cookies')
 293
 294             elif next_subtask == 'LoginSuccessSubtask':
 295                 raise ExtractorError('Twitter API did not grant auth token cookie')
 296
 297             else:
 298                 raise ExtractorError(f'Unrecognized subtask ID "{next_subtask}"')
 299
 300         self.report_login()
 301
 302     def _call_api(self, path, video_id, query={}, graphql=False):
 303         headers = self._set_base_headers(legacy=not graphql and self._selected_api == 'legacy')
 304         headers.update({
 305             'x-twitter-auth-type': 'OAuth2Session',
 306             'x-twitter-client-language': 'en',
 307             'x-twitter-active-user': 'yes',
 308         } if self.is_logged_in else {
 309             'x-guest-token': self._fetch_guest_token(video_id),
 310         })
 311         allowed_status = {400, 401, 403, 404} if graphql else {403}
 312         result = self._download_json(
 313             (self._GRAPHQL_API_BASE if graphql else self._API_BASE) + path,
 314             video_id, headers=headers, query=query, expected_status=allowed_status,
 315             note=f'Downloading {"GraphQL" if graphql else "legacy API"} JSON')
 316
 317         if result.get('errors'):
 318             errors = ', '.join(set(traverse_obj(result, ('errors', ..., 'message', {str}))))
 319             if errors and 'not authorized' in errors:
 320                 self.raise_login_required(remove_end(errors, '.'))
 321             raise ExtractorError(f'Error(s) while querying API: {errors or "Unknown error"}')
 322
 323         return result
 324
 325     def _build_graphql_query(self, media_id):
 326         raise NotImplementedError('Method must be implemented to support GraphQL')
 327
 328     def _call_graphql_api(self, endpoint, media_id):
 329         data = self._build_graphql_query(media_id)
 330         query = {key: json.dumps(value, separators=(',', ':')) for key, value in data.items()}
 331         return traverse_obj(self._call_api(endpoint, media_id, query=query, graphql=True), 'data')
 332
 333
 334 class TwitterCardIE(InfoExtractor):
 335     IE_NAME = 'twitter:card'
 336     _VALID_URL = TwitterBaseIE._BASE_REGEX + r'i/(?:cards/tfw/v1|videos(?:/tweet)?)/(?P<id>\d+)'
 337     _TESTS = [
 338         {
 339             'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889',
 340             # MD5 checksums are different in different places
 341             'info_dict': {
 342                 'id': '560070131976392705',
 343                 'ext': 'mp4',
 344                 'title': "Twitter - You can now shoot, edit and share video on Twitter. Capture life's most moving moments from your perspective.",
 345                 'description': 'md5:18d3e24bb4f6e5007487dd546e53bd96',
 346                 'uploader': 'Twitter',
 347                 'uploader_id': 'Twitter',
 348                 'thumbnail': r're:^https?://.*\.jpg',
 349                 'duration': 30.033,
 350                 'timestamp': 1422366112,
 351                 'upload_date': '20150127',
 352                 'age_limit': 0,
 353                 'comment_count': int,
 354                 'tags': [],
 355                 'repost_count': int,
 356                 'like_count': int,
 357                 'display_id': '560070183650213889',
 358                 'uploader_url': 'https://twitter.com/Twitter',
 359             },
 360         },
 361         {
 362             'url': 'https://twitter.com/i/cards/tfw/v1/623160978427936768',
 363             'md5': '7137eca597f72b9abbe61e5ae0161399',
 364             'info_dict': {
 365                 'id': '623160978427936768',
 366                 'ext': 'mp4',
 367                 'title': "NASA - Fly over Pluto's icy Norgay Mountains and Sputnik Plain in this @NASANewHorizons #PlutoFlyby video.",
 368                 'description': "Fly over Pluto's icy Norgay Mountains and Sputnik Plain in this @NASANewHorizons #PlutoFlyby video. https://t.co/BJYgOjSeGA",
 369                 'uploader': 'NASA',
 370                 'uploader_id': 'NASA',
 371                 'timestamp': 1437408129,
 372                 'upload_date': '20150720',
 373                 'uploader_url': 'https://twitter.com/NASA',
 374                 'age_limit': 0,
 375                 'comment_count': int,
 376                 'like_count': int,
 377                 'repost_count': int,
 378                 'tags': ['PlutoFlyby'],
 379             },
 380             'params': {'format': '[protocol=https]'},
 381         },
 382         {
 383             'url': 'https://twitter.com/i/cards/tfw/v1/654001591733886977',
 384             'md5': 'b6d9683dd3f48e340ded81c0e917ad46',
 385             'info_dict': {
 386                 'id': 'dq4Oj5quskI',
 387                 'ext': 'mp4',
 388                 'title': 'Ubuntu 11.10 Overview',
 389                 'description': 'md5:a831e97fa384863d6e26ce48d1c43376',
 390                 'upload_date': '20111013',
 391                 'uploader': 'OMG! UBUNTU!',
 392                 'uploader_id': 'omgubuntu',
 393                 'channel_url': 'https://www.youtube.com/channel/UCIiSwcm9xiFb3Y4wjzR41eQ',
 394                 'channel_id': 'UCIiSwcm9xiFb3Y4wjzR41eQ',
 395                 'channel_follower_count': int,
 396                 'chapters': 'count:8',
 397                 'uploader_url': 'http://www.youtube.com/user/omgubuntu',
 398                 'duration': 138,
 399                 'categories': ['Film & Animation'],
 400                 'age_limit': 0,
 401                 'comment_count': int,
 402                 'availability': 'public',
 403                 'like_count': int,
 404                 'thumbnail': 'https://i.ytimg.com/vi/dq4Oj5quskI/maxresdefault.jpg',
 405                 'view_count': int,
 406                 'tags': 'count:12',
 407                 'channel': 'OMG! UBUNTU!',
 408                 'playable_in_embed': True,
 409             },
 410             'add_ie': ['Youtube'],
 411         },
 412         {
 413             'url': 'https://twitter.com/i/cards/tfw/v1/665289828897005568',
 414             'info_dict': {
 415                 'id': 'iBb2x00UVlv',
 416                 'ext': 'mp4',
 417                 'upload_date': '20151113',
 418                 'uploader_id': '1189339351084113920',
 419                 'uploader': 'ArsenalTerje',
 420                 'title': 'Vine by ArsenalTerje',
 421                 'timestamp': 1447451307,
 422                 'alt_title': 'Vine by ArsenalTerje',
 423                 'comment_count': int,
 424                 'like_count': int,
 425                 'thumbnail': r're:^https?://[^?#]+\.jpg',
 426                 'view_count': int,
 427                 'repost_count': int,
 428             },
 429             'add_ie': ['Vine'],
 430             'params': {'skip_download': 'm3u8'},
 431         },
 432         {
 433             'url': 'https://twitter.com/i/videos/tweet/705235433198714880',
 434             'md5': '884812a2adc8aaf6fe52b15ccbfa3b88',
 435             'info_dict': {
 436                 'id': '705235433198714880',
 437                 'ext': 'mp4',
 438                 'title': "Brent Yarina - Khalil Iverson's missed highlight dunk. And made highlight dunk. In one highlight.",
 439                 'description': "Khalil Iverson's missed highlight dunk. And made highlight dunk. In one highlight. https://t.co/OrxcJ28Bns",
 440                 'uploader': 'Brent Yarina',
 441                 'uploader_id': 'BTNBrentYarina',
 442                 'timestamp': 1456976204,
 443                 'upload_date': '20160303',
 444             },
 445             'skip': 'This content is no longer available.',
 446         },
 447         {
 448             'url': 'https://twitter.com/i/videos/752274308186120192',
 449             'only_matching': True,
 450         },
 451     ]
 452
 453     def _real_extract(self, url):
 454         status_id = self._match_id(url)
 455         return self.url_result(
 456             'https://twitter.com/statuses/' + status_id,
 457             TwitterIE.ie_key(), status_id)
 458
 459
 460 class TwitterIE(TwitterBaseIE):
 461     IE_NAME = 'twitter'
 462     _VALID_URL = TwitterBaseIE._BASE_REGEX + r'(?:(?:i/web|[^/]+)/status|statuses)/(?P<id>\d+)(?:/(?:video|photo)/(?P<index>\d+))?'
 463
 464     _TESTS = [{
 465         'url': 'https://twitter.com/freethenipple/status/643211948184596480',
 466         'info_dict': {
 467             'id': '643211870443208704',
 468             'display_id': '643211948184596480',
 469             'ext': 'mp4',
 470             'title': 'FREE THE NIPPLE - FTN supporters on Hollywood Blvd today!',
 471             'thumbnail': r're:^https?://.*\.jpg',
 472             'description': 'FTN supporters on Hollywood Blvd today! http://t.co/c7jHH749xJ',
 473             'channel_id': '549749560',
 474             'uploader': 'FREE THE NIPPLE',
 475             'uploader_id': 'freethenipple',
 476             'duration': 12.922,
 477             'timestamp': 1442188653,
 478             'upload_date': '20150913',
 479             'uploader_url': 'https://twitter.com/freethenipple',
 480             'comment_count': int,
 481             'repost_count': int,
 482             'like_count': int,
 483             'tags': [],
 484             'age_limit': 18,
 485             '_old_archive_ids': ['twitter 643211948184596480'],
 486         },
 487         'skip': 'Requires authentication',
 488     }, {
 489         'url': 'https://twitter.com/giphz/status/657991469417025536/photo/1',
 490         'md5': 'f36dcd5fb92bf7057f155e7d927eeb42',
 491         'info_dict': {
 492             'id': '657991469417025536',
 493             'ext': 'mp4',
 494             'title': 'Gifs - tu vai cai tu vai cai tu nao eh capaz disso tu vai cai',
 495             'description': 'Gifs on Twitter: "tu vai cai tu vai cai tu nao eh capaz disso tu vai cai https://t.co/tM46VHFlO5"',
 496             'thumbnail': r're:^https?://.*\.png',
 497             'uploader': 'Gifs',
 498             'uploader_id': 'giphz',
 499         },
 500         'expected_warnings': ['height', 'width'],
 501         'skip': 'Account suspended',
 502     }, {
 503         'url': 'https://twitter.com/starwars/status/665052190608723968',
 504         'info_dict': {
 505             'id': '665052190608723968',
 506             'display_id': '665052190608723968',
 507             'ext': 'mp4',
 508             'title': r're:Star Wars.*A new beginning is coming December 18.*',
 509             'description': 'A new beginning is coming December 18. Watch the official 60 second #TV spot for #StarWars: #TheForceAwakens. https://t.co/OkSqT2fjWJ',
 510             'channel_id': '20106852',
 511             'uploader_id': 'starwars',
 512             'uploader': r're:Star Wars.*',
 513             'timestamp': 1447395772,
 514             'upload_date': '20151113',
 515             'uploader_url': 'https://twitter.com/starwars',
 516             'comment_count': int,
 517             'repost_count': int,
 518             'like_count': int,
 519             'tags': ['TV', 'StarWars', 'TheForceAwakens'],
 520             'age_limit': 0,
 521             '_old_archive_ids': ['twitter 665052190608723968'],
 522         },
 523     }, {
 524         'url': 'https://twitter.com/BTNBrentYarina/status/705235433198714880',
 525         'info_dict': {
 526             'id': '705235433198714880',
 527             'ext': 'mp4',
 528             'title': "Brent Yarina - Khalil Iverson's missed highlight dunk. And made highlight dunk. In one highlight.",
 529             'description': "Khalil Iverson's missed highlight dunk. And made highlight dunk. In one highlight. https://t.co/OrxcJ28Bns",
 530             'uploader_id': 'BTNBrentYarina',
 531             'uploader': 'Brent Yarina',
 532             'timestamp': 1456976204,
 533             'upload_date': '20160303',
 534             'uploader_url': 'https://twitter.com/BTNBrentYarina',
 535             'comment_count': int,
 536             'repost_count': int,
 537             'like_count': int,
 538             'tags': [],
 539             'age_limit': 0,
 540         },
 541         'params': {
 542             # The same video as https://twitter.com/i/videos/tweet/705235433198714880
 543             # Test case of TwitterCardIE
 544             'skip_download': True,
 545         },
 546         'skip': 'Dead external link',
 547     }, {
 548         'url': 'https://twitter.com/jaydingeer/status/700207533655363584',
 549         'info_dict': {
 550             'id': '700207414000242688',
 551             'display_id': '700207533655363584',
 552             'ext': 'mp4',
 553             'title': 'jaydin donte geer - BEAT PROD: @suhmeduh #Damndaniel',
 554             'description': 'BEAT PROD: @suhmeduh  https://t.co/HBrQ4AfpvZ #Damndaniel https://t.co/byBooq2ejZ',
 555             'thumbnail': r're:^https?://.*\.jpg',
 556             'channel_id': '1383165541',
 557             'uploader': 'jaydin donte geer',
 558             'uploader_id': 'jaydingeer',
 559             'duration': 30.0,
 560             'timestamp': 1455777459,
 561             'upload_date': '20160218',
 562             'uploader_url': 'https://twitter.com/jaydingeer',
 563             'comment_count': int,
 564             'repost_count': int,
 565             'like_count': int,
 566             'tags': ['Damndaniel'],
 567             'age_limit': 0,
 568             '_old_archive_ids': ['twitter 700207533655363584'],
 569         },
 570     }, {
 571         'url': 'https://twitter.com/Filmdrunk/status/713801302971588609',
 572         'md5': '89a15ed345d13b86e9a5a5e051fa308a',
 573         'info_dict': {
 574             'id': 'MIOxnrUteUd',
 575             'ext': 'mp4',
 576             'title': 'Dr.Pepperの飲み方 #japanese #バカ #ドクペ #電動ガン',
 577             'uploader': 'TAKUMA',
 578             'uploader_id': '1004126642786242560',
 579             'timestamp': 1402826626,
 580             'upload_date': '20140615',
 581             'thumbnail': r're:^https?://.*\.jpg',
 582             'alt_title': 'Vine by TAKUMA',
 583             'comment_count': int,
 584             'repost_count': int,
 585             'like_count': int,
 586             'view_count': int,
 587         },
 588         'add_ie': ['Vine'],
 589     }, {
 590         'url': 'https://twitter.com/captainamerica/status/719944021058060289',
 591         'info_dict': {
 592             'id': '717462543795523584',
 593             'display_id': '719944021058060289',
 594             'ext': 'mp4',
 595             'title': 'Captain America - @King0fNerd Are you sure you made the right choice? Find out in theaters.',
 596             'description': '@King0fNerd Are you sure you made the right choice? Find out in theaters. https://t.co/GpgYi9xMJI',
 597             'channel_id': '701615052',
 598             'uploader_id': 'CaptainAmerica',
 599             'uploader': 'Captain America',
 600             'duration': 3.17,
 601             'timestamp': 1460483005,
 602             'upload_date': '20160412',
 603             'uploader_url': 'https://twitter.com/CaptainAmerica',
 604             'thumbnail': r're:^https?://.*\.jpg',
 605             'comment_count': int,
 606             'repost_count': int,
 607             'like_count': int,
 608             'tags': [],
 609             'age_limit': 0,
 610             '_old_archive_ids': ['twitter 719944021058060289'],
 611         },
 612     }, {
 613         'url': 'https://twitter.com/OPP_HSD/status/779210622571536384',
 614         'info_dict': {
 615             'id': '1zqKVVlkqLaKB',
 616             'ext': 'mp4',
 617             'title': 'Sgt Kerry Schmidt - Ontario Provincial Police - Road rage, mischief, assault, rollover and fire in one occurrence',
 618             'upload_date': '20160923',
 619             'uploader_id': '1PmKqpJdOJQoY',
 620             'uploader': 'Sgt Kerry Schmidt - Ontario Provincial Police',
 621             'timestamp': 1474613214,
 622             'thumbnail': r're:^https?://.*\.jpg',
 623         },
 624         'add_ie': ['Periscope'],
 625         'skip': 'Broadcast not found',
 626     }, {
 627         # has mp4 formats via mobile API
 628         'url': 'https://twitter.com/news_al3alm/status/852138619213144067',
 629         'info_dict': {
 630             'id': '852077943283097602',
 631             'ext': 'mp4',
 632             'title': 'عالم الأخبار - كلمة تاريخية بجلسة الجناسي التاريخية.. النائب خالد مؤنس العتيبي للمعارضين : اتقوا الله .. الظلم ظلمات يوم القيامة',
 633             'description': 'كلمة تاريخية بجلسة الجناسي التاريخية.. النائب خالد مؤنس العتيبي للمعارضين : اتقوا الله .. الظلم ظلمات يوم القيامة   https://t.co/xg6OhpyKfN',
 634             'channel_id': '2526757026',
 635             'uploader': 'عالم الأخبار',
 636             'uploader_id': 'news_al3alm',
 637             'duration': 277.4,
 638             'timestamp': 1492000653,
 639             'upload_date': '20170412',
 640             'display_id': '852138619213144067',
 641             'age_limit': 0,
 642             'uploader_url': 'https://twitter.com/news_al3alm',
 643             'thumbnail': r're:^https?://.*\.jpg',
 644             'tags': [],
 645             'repost_count': int,
 646             'like_count': int,
 647             'comment_count': int,
 648             '_old_archive_ids': ['twitter 852138619213144067'],
 649         },
 650     }, {
 651         'url': 'https://twitter.com/i/web/status/910031516746514432',
 652         'info_dict': {
 653             'id': '910030238373089285',
 654             'display_id': '910031516746514432',
 655             'ext': 'mp4',
 656             'title': 'Préfet de Guadeloupe - [Direct] #Maria Le centre se trouve actuellement au sud de Basse-Terre. Restez confinés. Réfugiez-vous dans la pièce la + sûre.',
 657             'thumbnail': r're:^https?://.*\.jpg',
 658             'description': '[Direct] #Maria Le centre se trouve actuellement au sud de Basse-Terre. Restez confinés. Réfugiez-vous dans la pièce la + sûre. https://t.co/mwx01Rs4lo',
 659             'channel_id': '2319432498',
 660             'uploader': 'Préfet de Guadeloupe',
 661             'uploader_id': 'Prefet971',
 662             'duration': 47.48,
 663             'timestamp': 1505803395,
 664             'upload_date': '20170919',
 665             'uploader_url': 'https://twitter.com/Prefet971',
 666             'comment_count': int,
 667             'repost_count': int,
 668             'like_count': int,
 669             'tags': ['Maria'],
 670             'age_limit': 0,
 671             '_old_archive_ids': ['twitter 910031516746514432'],
 672         },
 673         'params': {
 674             'skip_download': True,  # requires ffmpeg
 675         },
 676     }, {
 677         # card via api.twitter.com/1.1/videos/tweet/config
 678         'url': 'https://twitter.com/LisPower1/status/1001551623938805763',
 679         'info_dict': {
 680             'id': '1001551417340022785',
 681             'display_id': '1001551623938805763',
 682             'ext': 'mp4',
 683             'title': 're:.*?Shep is on a roll today.*?',
 684             'thumbnail': r're:^https?://.*\.jpg',
 685             'description': 'md5:37b9f2ff31720cef23b2bd42ee8a0f09',
 686             'channel_id': '255036353',
 687             'uploader': 'Lis Power',
 688             'uploader_id': 'LisPower1',
 689             'duration': 111.278,
 690             'timestamp': 1527623489,
 691             'upload_date': '20180529',
 692             'uploader_url': 'https://twitter.com/LisPower1',
 693             'comment_count': int,
 694             'repost_count': int,
 695             'like_count': int,
 696             'tags': [],
 697             'age_limit': 0,
 698             '_old_archive_ids': ['twitter 1001551623938805763'],
 699         },
 700         'params': {
 701             'skip_download': True,  # requires ffmpeg
 702         },
 703     }, {
 704         'url': 'https://twitter.com/foobar/status/1087791357756956680',
 705         'info_dict': {
 706             'id': '1087791272830607360',
 707             'display_id': '1087791357756956680',
 708             'ext': 'mp4',
 709             'title': 'X - A new is coming.  Some of you got an opt-in to try it now. Check out the emoji button, quick keyboard shortcuts, upgraded trends, advanced search, and more. Let us know your thoughts!',
 710             'thumbnail': r're:^https?://.*\.jpg',
 711             'description': 'md5:6dfd341a3310fb97d80d2bf7145df976',
 712             'uploader': 'X',
 713             'uploader_id': 'X',
 714             'duration': 61.567,
 715             'timestamp': 1548184644,
 716             'upload_date': '20190122',
 717             'uploader_url': 'https://twitter.com/X',
 718             'comment_count': int,
 719             'repost_count': int,
 720             'like_count': int,
 721             'view_count': int,
 722             'tags': [],
 723             'age_limit': 0,
 724         },
 725         'skip': 'This Tweet is unavailable',
 726     }, {
 727         # not available in Periscope
 728         'url': 'https://twitter.com/ViviEducation/status/1136534865145286656',
 729         'info_dict': {
 730             'id': '1vOGwqejwoWxB',
 731             'ext': 'mp4',
 732             'title': 'Vivi - Vivi founder @lior_rauchy announcing our new student feedback tool live at @EduTECH_AU #EduTECH2019',
 733             'uploader': 'Vivi',
 734             'uploader_id': '1eVjYOLGkGrQL',
 735             'thumbnail': r're:^https?://.*\.jpg',
 736             'tags': ['EduTECH2019'],
 737             'view_count': int,
 738         },
 739         'add_ie': ['TwitterBroadcast'],
 740         'skip': 'Broadcast no longer exists',
 741     }, {
 742         # unified card
 743         'url': 'https://twitter.com/BrooklynNets/status/1349794411333394432?s=20',
 744         'info_dict': {
 745             'id': '1349774757969989634',
 746             'display_id': '1349794411333394432',
 747             'ext': 'mp4',
 748             'title': 'md5:d1c4941658e4caaa6cb579260d85dcba',
 749             'thumbnail': r're:^https?://.*\.jpg',
 750             'description': 'md5:71ead15ec44cee55071547d6447c6a3e',
 751             'channel_id': '18552281',
 752             'uploader': 'Brooklyn Nets',
 753             'uploader_id': 'BrooklynNets',
 754             'duration': 324.484,
 755             'timestamp': 1610651040,
 756             'upload_date': '20210114',
 757             'uploader_url': 'https://twitter.com/BrooklynNets',
 758             'comment_count': int,
 759             'repost_count': int,
 760             'like_count': int,
 761             'tags': [],
 762             'age_limit': 0,
 763             '_old_archive_ids': ['twitter 1349794411333394432'],
 764         },
 765         'params': {
 766             'skip_download': True,
 767         },
 768     }, {
 769         'url': 'https://twitter.com/oshtru/status/1577855540407197696',
 770         'info_dict': {
 771             'id': '1577855447914409984',
 772             'display_id': '1577855540407197696',
 773             'ext': 'mp4',
 774             'title': 'md5:466a3a8b049b5f5a13164ce915484b51',
 775             'description': 'md5:b9c3699335447391d11753ab21c70a74',
 776             'upload_date': '20221006',
 777             'channel_id': '143077138',
 778             'uploader': 'Oshtru',
 779             'uploader_id': 'oshtru',
 780             'uploader_url': 'https://twitter.com/oshtru',
 781             'thumbnail': r're:^https?://.*\.jpg',
 782             'duration': 30.03,
 783             'timestamp': 1665025050,
 784             'comment_count': int,
 785             'repost_count': int,
 786             'like_count': int,
 787             'tags': [],
 788             'age_limit': 0,
 789             '_old_archive_ids': ['twitter 1577855540407197696'],
 790         },
 791         'params': {'skip_download': True},
 792     }, {
 793         'url': 'https://twitter.com/UltimaShadowX/status/1577719286659006464',
 794         'info_dict': {
 795             'id': '1577719286659006464',
 796             'title': 'Ultima Reload - Test',
 797             'description': 'Test https://t.co/Y3KEZD7Dad',
 798             'channel_id': '168922496',
 799             'uploader': 'Ultima Reload',
 800             'uploader_id': 'UltimaShadowX',
 801             'uploader_url': 'https://twitter.com/UltimaShadowX',
 802             'upload_date': '20221005',
 803             'timestamp': 1664992565,
 804             'comment_count': int,
 805             'repost_count': int,
 806             'like_count': int,
 807             'tags': [],
 808             'age_limit': 0,
 809         },
 810         'playlist_count': 4,
 811         'params': {'skip_download': True},
 812     }, {
 813         'url': 'https://twitter.com/MesoMax919/status/1575560063510810624',
 814         'info_dict': {
 815             'id': '1575559336759263233',
 816             'display_id': '1575560063510810624',
 817             'ext': 'mp4',
 818             'title': 'md5:eec26382babd0f7c18f041db8ae1c9c9',
 819             'thumbnail': r're:^https?://.*\.jpg',
 820             'description': 'md5:95aea692fda36a12081b9629b02daa92',
 821             'channel_id': '1094109584',
 822             'uploader': 'Max Olson',
 823             'uploader_id': 'MesoMax919',
 824             'uploader_url': 'https://twitter.com/MesoMax919',
 825             'duration': 21.321,
 826             'timestamp': 1664477766,
 827             'upload_date': '20220929',
 828             'comment_count': int,
 829             'repost_count': int,
 830             'like_count': int,
 831             'tags': ['HurricaneIan'],
 832             'age_limit': 0,
 833             '_old_archive_ids': ['twitter 1575560063510810624'],
 834         },
 835     }, {
 836         # Adult content, fails if not logged in
 837         'url': 'https://twitter.com/Rizdraws/status/1575199173472927762',
 838         'info_dict': {
 839             'id': '1575199163847000068',
 840             'display_id': '1575199173472927762',
 841             'ext': 'mp4',
 842             'title': str,
 843             'description': str,
 844             'channel_id': '1217167793541480450',
 845             'uploader': str,
 846             'uploader_id': 'Rizdraws',
 847             'uploader_url': 'https://twitter.com/Rizdraws',
 848             'upload_date': '20220928',
 849             'timestamp': 1664391723,
 850             'thumbnail': r're:^https?://.+\.jpg',
 851             'like_count': int,
 852             'repost_count': int,
 853             'comment_count': int,
 854             'age_limit': 18,
 855             'tags': [],
 856             '_old_archive_ids': ['twitter 1575199173472927762'],
 857         },
 858         'params': {'skip_download': 'The media could not be played'},
 859         'skip': 'Requires authentication',
 860     }, {
 861         # Playlist result only with graphql API
 862         'url': 'https://twitter.com/Srirachachau/status/1395079556562706435',
 863         'playlist_mincount': 2,
 864         'info_dict': {
 865             'id': '1395079556562706435',
 866             'title': str,
 867             'tags': [],
 868             'channel_id': '21539378',
 869             'uploader': str,
 870             'like_count': int,
 871             'upload_date': '20210519',
 872             'age_limit': 0,
 873             'repost_count': int,
 874             'description': 'Here it is! Finished my gothic western cartoon. Pretty proud of it. It\'s got some goofs and lots of splashy over the top violence, something for everyone, hope you like it https://t.co/fOsG5glUnw',
 875             'uploader_id': 'Srirachachau',
 876             'comment_count': int,
 877             'uploader_url': 'https://twitter.com/Srirachachau',
 878             'timestamp': 1621447860,
 879         },
 880     }, {
 881         'url': 'https://twitter.com/DavidToons_/status/1578353380363501568',
 882         'playlist_mincount': 2,
 883         'info_dict': {
 884             'id': '1578353380363501568',
 885             'title': str,
 886             'channel_id': '2195866214',
 887             'uploader_id': 'DavidToons_',
 888             'repost_count': int,
 889             'like_count': int,
 890             'uploader': str,
 891             'timestamp': 1665143744,
 892             'uploader_url': 'https://twitter.com/DavidToons_',
 893             'description': 'Chris sounds like Linda from Bob\'s Burgers, so as an animator: this had to be done. https://t.co/WgJauwIW1w',
 894             'tags': [],
 895             'comment_count': int,
 896             'upload_date': '20221007',
 897             'age_limit': 0,
 898         },
 899     }, {
 900         'url': 'https://twitter.com/primevideouk/status/1578401165338976258',
 901         'playlist_count': 2,
 902         'info_dict': {
 903             'id': '1578401165338976258',
 904             'title': str,
 905             'description': 'md5:659a6b517a034b4cee5d795381a2dc41',
 906             'channel_id': '19338359',
 907             'uploader': str,
 908             'uploader_id': 'primevideouk',
 909             'timestamp': 1665155137,
 910             'upload_date': '20221007',
 911             'age_limit': 0,
 912             'uploader_url': 'https://twitter.com/primevideouk',
 913             'comment_count': int,
 914             'repost_count': int,
 915             'like_count': int,
 916             'tags': ['TheRingsOfPower'],
 917         },
 918     }, {
 919         # Twitter Spaces
 920         'url': 'https://twitter.com/MoniqueCamarra/status/1550101959377551360',
 921         'info_dict': {
 922             'id': '1lPJqmBeeNAJb',
 923             'ext': 'm4a',
 924             'title': 'EuroFile@6 Ukraine Up-date-Draghi Defenestration-the West',
 925             'uploader': r're:Monique Camarra.+?',
 926             'uploader_id': 'MoniqueCamarra',
 927             'live_status': 'was_live',
 928             'release_timestamp': 1658417414,
 929             'description': r're:Twitter Space participated by Sergej Sumlenny.+',
 930             'timestamp': 1658407771,
 931             'release_date': '20220721',
 932             'upload_date': '20220721',
 933         },
 934         'add_ie': ['TwitterSpaces'],
 935         'params': {'skip_download': 'm3u8'},
 936     }, {
 937         # URL specifies video number but --yes-playlist
 938         'url': 'https://twitter.com/CTVJLaidlaw/status/1600649710662213632/video/1',
 939         'playlist_mincount': 2,
 940         'info_dict': {
 941             'id': '1600649710662213632',
 942             'title': 'md5:be05989b0722e114103ed3851a0ffae2',
 943             'timestamp': 1670459604.0,
 944             'description': 'md5:591c19ce66fadc2359725d5cd0d1052c',
 945             'comment_count': int,
 946             'uploader_id': 'CTVJLaidlaw',
 947             'channel_id': '80082014',
 948             'repost_count': int,
 949             'tags': ['colorectalcancer', 'cancerjourney', 'imnotaquitter'],
 950             'upload_date': '20221208',
 951             'age_limit': 0,
 952             'uploader': 'Jocelyn Laidlaw',
 953             'uploader_url': 'https://twitter.com/CTVJLaidlaw',
 954             'like_count': int,
 955         },
 956     }, {
 957         # URL specifies video number and --no-playlist
 958         'url': 'https://twitter.com/CTVJLaidlaw/status/1600649710662213632/video/2',
 959         'info_dict': {
 960             'id': '1600649511827013632',
 961             'ext': 'mp4',
 962             'title': 'md5:7662a0a27ce6faa3e5b160340f3cfab1',
 963             'thumbnail': r're:^https?://.+\.jpg',
 964             'timestamp': 1670459604.0,
 965             'channel_id': '80082014',
 966             'uploader_id': 'CTVJLaidlaw',
 967             'uploader': 'Jocelyn Laidlaw',
 968             'repost_count': int,
 969             'comment_count': int,
 970             'tags': ['colorectalcancer', 'cancerjourney', 'imnotaquitter'],
 971             'duration': 102.226,
 972             'uploader_url': 'https://twitter.com/CTVJLaidlaw',
 973             'display_id': '1600649710662213632',
 974             'like_count': int,
 975             'description': 'md5:591c19ce66fadc2359725d5cd0d1052c',
 976             'upload_date': '20221208',
 977             'age_limit': 0,
 978             '_old_archive_ids': ['twitter 1600649710662213632'],
 979         },
 980         'params': {'noplaylist': True},
 981     }, {
 982         # id pointing to TweetWithVisibilityResults type entity which wraps the actual Tweet over
 983         # note the id different between extraction and url
 984         'url': 'https://twitter.com/s2FAKER/status/1621117700482416640',
 985         'info_dict': {
 986             'id': '1621117577354424321',
 987             'display_id': '1621117700482416640',
 988             'ext': 'mp4',
 989             'title': '뽀 - 아 최우제 이동속도 봐',
 990             'description': '아 최우제 이동속도 봐 https://t.co/dxu2U5vXXB',
 991             'duration': 24.598,
 992             'channel_id': '1281839411068432384',
 993             'uploader': '뽀',
 994             'uploader_id': 's2FAKER',
 995             'uploader_url': 'https://twitter.com/s2FAKER',
 996             'upload_date': '20230202',
 997             'timestamp': 1675339553.0,
 998             'thumbnail': r're:https?://pbs\.twimg\.com/.+',
 999             'age_limit': 18,
1000             'tags': [],
1001             'like_count': int,
1002             'repost_count': int,
1003             'comment_count': int,
1004             '_old_archive_ids': ['twitter 1621117700482416640'],
1005         },
1006         'skip': 'Requires authentication',
1007     }, {
1008         'url': 'https://twitter.com/hlo_again/status/1599108751385972737/video/2',
1009         'info_dict': {
1010             'id': '1599108643743473680',
1011             'display_id': '1599108751385972737',
1012             'ext': 'mp4',
1013             'title': '\u06ea - \U0001F48B',
1014             'channel_id': '1347791436809441283',
1015             'uploader_url': 'https://twitter.com/hlo_again',
1016             'like_count': int,
1017             'uploader_id': 'hlo_again',
1018             'thumbnail': 'https://pbs.twimg.com/ext_tw_video_thumb/1599108643743473680/pu/img/UG3xjov4rgg5sbYM.jpg?name=orig',
1019             'repost_count': int,
1020             'duration': 9.531,
1021             'comment_count': int,
1022             'upload_date': '20221203',
1023             'age_limit': 0,
1024             'timestamp': 1670092210.0,
1025             'tags': [],
1026             'uploader': '\u06ea',
1027             'description': '\U0001F48B https://t.co/bTj9Qz7vQP',
1028             '_old_archive_ids': ['twitter 1599108751385972737'],
1029         },
1030         'params': {'noplaylist': True},
1031     }, {
1032         'url': 'https://twitter.com/MunTheShinobi/status/1600009574919962625',
1033         'info_dict': {
1034             'id': '1600009362759733248',
1035             'display_id': '1600009574919962625',
1036             'ext': 'mp4',
1037             'channel_id': '211814412',
1038             'uploader_url': 'https://twitter.com/MunTheShinobi',
1039             'description': 'This is a genius ad by Apple. \U0001f525\U0001f525\U0001f525\U0001f525\U0001f525 https://t.co/cNsA0MoOml',
1040             'thumbnail': 'https://pbs.twimg.com/ext_tw_video_thumb/1600009362759733248/pu/img/XVhFQivj75H_YxxV.jpg?name=orig',
1041             'age_limit': 0,
1042             'uploader': 'Mün',
1043             'repost_count': int,
1044             'upload_date': '20221206',
1045             'title': 'Mün - This is a genius ad by Apple. \U0001f525\U0001f525\U0001f525\U0001f525\U0001f525',
1046             'comment_count': int,
1047             'like_count': int,
1048             'tags': [],
1049             'uploader_id': 'MunTheShinobi',
1050             'duration': 139.987,
1051             'timestamp': 1670306984.0,
1052             '_old_archive_ids': ['twitter 1600009574919962625'],
1053         },
1054     }, {
1055         # retweeted_status (private)
1056         'url': 'https://twitter.com/liberdalau/status/1623739803874349067',
1057         'info_dict': {
1058             'id': '1623274794488659969',
1059             'display_id': '1623739803874349067',
1060             'ext': 'mp4',
1061             'title': 'Johnny Bullets - Me after going viral to over 30million people:    Whoopsie-daisy',
1062             'description': 'md5:b06864cd3dc2554821cc327f5348485a',
1063             'uploader': 'Johnny Bullets',
1064             'uploader_id': 'Johnnybull3ts',
1065             'uploader_url': 'https://twitter.com/Johnnybull3ts',
1066             'age_limit': 0,
1067             'tags': [],
1068             'duration': 8.033,
1069             'timestamp': 1675853859.0,
1070             'upload_date': '20230208',
1071             'thumbnail': r're:https://pbs\.twimg\.com/ext_tw_video_thumb/.+',
1072             'like_count': int,
1073             'repost_count': int,
1074         },
1075         'skip': 'Protected tweet',
1076     }, {
1077         # retweeted_status
1078         'url': 'https://twitter.com/playstrumpcard/status/1695424220702888009',
1079         'info_dict': {
1080             'id': '1694928337846538240',
1081             'ext': 'mp4',
1082             'display_id': '1695424220702888009',
1083             'title': 'md5:e8daa9527bc2b947121395494f786d9d',
1084             'description': 'md5:004f2d37fd58737724ec75bc7e679938',
1085             'channel_id': '15212187',
1086             'uploader': 'Benny Johnson',
1087             'uploader_id': 'bennyjohnson',
1088             'uploader_url': 'https://twitter.com/bennyjohnson',
1089             'age_limit': 0,
1090             'tags': [],
1091             'duration': 45.001,
1092             'timestamp': 1692962814.0,
1093             'upload_date': '20230825',
1094             'thumbnail': r're:https://pbs\.twimg\.com/amplify_video_thumb/.+',
1095             'like_count': int,
1096             'repost_count': int,
1097             'comment_count': int,
1098             '_old_archive_ids': ['twitter 1695424220702888009'],
1099         },
1100     }, {
1101         # retweeted_status w/ legacy API
1102         'url': 'https://twitter.com/playstrumpcard/status/1695424220702888009',
1103         'info_dict': {
1104             'id': '1694928337846538240',
1105             'ext': 'mp4',
1106             'display_id': '1695424220702888009',
1107             'title': 'md5:e8daa9527bc2b947121395494f786d9d',
1108             'description': 'md5:004f2d37fd58737724ec75bc7e679938',
1109             'channel_id': '15212187',
1110             'uploader': 'Benny Johnson',
1111             'uploader_id': 'bennyjohnson',
1112             'uploader_url': 'https://twitter.com/bennyjohnson',
1113             'age_limit': 0,
1114             'tags': [],
1115             'duration': 45.001,
1116             'timestamp': 1692962814.0,
1117             'upload_date': '20230825',
1118             'thumbnail': r're:https://pbs\.twimg\.com/amplify_video_thumb/.+',
1119             'like_count': int,
1120             'repost_count': int,
1121             '_old_archive_ids': ['twitter 1695424220702888009'],
1122         },
1123         'params': {'extractor_args': {'twitter': {'api': ['legacy']}}},
1124     }, {
1125         # Broadcast embedded in tweet
1126         'url': 'https://twitter.com/JessicaDobsonWX/status/1731121063248175384',
1127         'info_dict': {
1128             'id': '1rmxPMjLzAXKN',
1129             'ext': 'mp4',
1130             'title': 'WAVE Weather Now - Saturday 12/2/23 Update',
1131             'uploader': 'Jessica Dobson',
1132             'uploader_id': 'JessicaDobsonWX',
1133             'uploader_url': 'https://twitter.com/JessicaDobsonWX',
1134             'timestamp': 1701566398,
1135             'upload_date': '20231203',
1136             'live_status': 'was_live',
1137             'thumbnail': r're:https://[^/]+pscp\.tv/.+\.jpg',
1138             'concurrent_view_count': int,
1139             'view_count': int,
1140         },
1141         'add_ie': ['TwitterBroadcast'],
1142     }, {
1143         # Animated gif and quote tweet video
1144         'url': 'https://twitter.com/BAKKOOONN/status/1696256659889565950',
1145         'playlist_mincount': 2,
1146         'info_dict': {
1147             'id': '1696256659889565950',
1148             'title': 'BAKOON - https://t.co/zom968d0a0',
1149             'description': 'https://t.co/zom968d0a0',
1150             'tags': [],
1151             'channel_id': '1263540390',
1152             'uploader': 'BAKOON',
1153             'uploader_id': 'BAKKOOONN',
1154             'uploader_url': 'https://twitter.com/BAKKOOONN',
1155             'age_limit': 18,
1156             'timestamp': 1693254077.0,
1157             'upload_date': '20230828',
1158             'like_count': int,
1159             'comment_count': int,
1160             'repost_count': int,
1161         },
1162         'skip': 'Requires authentication',
1163     }, {
1164         # "stale tweet" with typename "TweetWithVisibilityResults"
1165         'url': 'https://twitter.com/RobertKennedyJr/status/1724884212803834154',
1166         'md5': '511377ff8dfa7545307084dca4dce319',
1167         'info_dict': {
1168             'id': '1724883339285544960',
1169             'ext': 'mp4',
1170             'title': 'md5:cc56716f9ed0b368de2ba54c478e493c',
1171             'description': 'md5:9dc14f5b0f1311fc7caf591ae253a164',
1172             'display_id': '1724884212803834154',
1173             'channel_id': '337808606',
1174             'uploader': 'Robert F. Kennedy Jr',
1175             'uploader_id': 'RobertKennedyJr',
1176             'uploader_url': 'https://twitter.com/RobertKennedyJr',
1177             'upload_date': '20231115',
1178             'timestamp': 1700079417.0,
1179             'duration': 341.048,
1180             'thumbnail': r're:https://pbs\.twimg\.com/amplify_video_thumb/.+',
1181             'tags': ['Kennedy24'],
1182             'repost_count': int,
1183             'like_count': int,
1184             'comment_count': int,
1185             'age_limit': 0,
1186             '_old_archive_ids': ['twitter 1724884212803834154'],
1187         },
1188     }, {
1189         # x.com
1190         'url': 'https://x.com/historyinmemes/status/1790637656616943991',
1191         'md5': 'daca3952ba0defe2cfafb1276d4c1ea5',
1192         'info_dict': {
1193             'id': '1790637589910654976',
1194             'ext': 'mp4',
1195             'title': 'Historic Vids - One of the most intense moments in history',
1196             'description': 'One of the most intense moments in history https://t.co/Zgzhvix8ES',
1197             'display_id': '1790637656616943991',
1198             'uploader': 'Historic Vids',
1199             'uploader_id': 'historyinmemes',
1200             'uploader_url': 'https://twitter.com/historyinmemes',
1201             'channel_id': '855481986290524160',
1202             'upload_date': '20240515',
1203             'timestamp': 1715756260.0,
1204             'duration': 15.488,
1205             'tags': [],
1206             'comment_count': int,
1207             'repost_count': int,
1208             'like_count': int,
1209             'thumbnail': r're:https://pbs\.twimg\.com/amplify_video_thumb/.+',
1210             'age_limit': 0,
1211             '_old_archive_ids': ['twitter 1790637656616943991'],
1212         },
1213     }, {
1214         # onion route
1215         'url': 'https://twitter3e4tixl4xyajtrzo62zg5vztmjuricljdp2c5kshju4avyoid.onion/TwitterBlue/status/1484226494708662273',
1216         'only_matching': True,
1217     }, {
1218         # Twitch Clip Embed
1219         'url': 'https://twitter.com/GunB1g/status/1163218564784017422',
1220         'only_matching': True,
1221     }, {
1222         # promo_video_website card
1223         'url': 'https://twitter.com/GunB1g/status/1163218564784017422',
1224         'only_matching': True,
1225     }, {
1226         # promo_video_convo card
1227         'url': 'https://twitter.com/poco_dandy/status/1047395834013384704',
1228         'only_matching': True,
1229     }, {
1230         # appplayer card
1231         'url': 'https://twitter.com/poco_dandy/status/1150646424461176832',
1232         'only_matching': True,
1233     }, {
1234         # video_direct_message card
1235         'url': 'https://twitter.com/qarev001/status/1348948114569269251',
1236         'only_matching': True,
1237     }, {
1238         # poll2choice_video card
1239         'url': 'https://twitter.com/CAF_Online/status/1349365911120195585',
1240         'only_matching': True,
1241     }, {
1242         # poll3choice_video card
1243         'url': 'https://twitter.com/SamsungMobileSA/status/1348609186725289984',
1244         'only_matching': True,
1245     }, {
1246         # poll4choice_video card
1247         'url': 'https://twitter.com/SouthamptonFC/status/1347577658079641604',
1248         'only_matching': True,
1249     }]
1250
1251     _MEDIA_ID_RE = re.compile(r'_video/(\d+)/')
1252
1253     @property
1254     def _GRAPHQL_ENDPOINT(self):
1255         if self.is_logged_in:
1256             return 'zZXycP0V6H7m-2r0mOnFcA/TweetDetail'
1257         return '2ICDjqPd81tulZcYrtpTuQ/TweetResultByRestId'
1258
1259     def _graphql_to_legacy(self, data, twid):
1260         result = traverse_obj(data, (
1261             'threaded_conversation_with_injections_v2', 'instructions', 0, 'entries',
1262             lambda _, v: v['entryId'] == f'tweet-{twid}', 'content', 'itemContent',
1263             'tweet_results', 'result', ('tweet', None), {dict},
1264         ), default={}, get_all=False) if self.is_logged_in else traverse_obj(
1265             data, ('tweetResult', 'result', {dict}), default={})
1266
1267         typename = result.get('__typename')
1268         if typename not in ('Tweet', 'TweetWithVisibilityResults', 'TweetTombstone', 'TweetUnavailable', None):
1269             self.report_warning(f'Unknown typename: {typename}', twid, only_once=True)
1270
1271         if 'tombstone' in result:
1272             cause = remove_end(traverse_obj(result, ('tombstone', 'text', 'text', {str})), '. Learn more')
1273             raise ExtractorError(f'Twitter API says: {cause or "Unknown error"}', expected=True)
1274         elif typename == 'TweetUnavailable':
1275             reason = result.get('reason')
1276             if reason == 'NsfwLoggedOut':
1277                 self.raise_login_required('NSFW tweet requires authentication')
1278             elif reason == 'Protected':
1279                 self.raise_login_required('You are not authorized to view this protected tweet')
1280             raise ExtractorError(reason or 'Requested tweet is unavailable', expected=True)
1281         # Result for "stale tweet" needs additional transformation
1282         elif typename == 'TweetWithVisibilityResults':
1283             result = traverse_obj(result, ('tweet', {dict})) or {}
1284
1285         status = result.get('legacy', {})
1286         status.update(traverse_obj(result, {
1287             'user': ('core', 'user_results', 'result', 'legacy'),
1288             'card': ('card', 'legacy'),
1289             'quoted_status': ('quoted_status_result', 'result', 'legacy'),
1290             'retweeted_status': ('legacy', 'retweeted_status_result', 'result', 'legacy'),
1291         }, expected_type=dict, default={}))
1292
1293         # extra transformations needed since result does not match legacy format
1294         if status.get('retweeted_status'):
1295             status['retweeted_status']['user'] = traverse_obj(status, (
1296                 'retweeted_status_result', 'result', 'core', 'user_results', 'result', 'legacy', {dict})) or {}
1297
1298         binding_values = {
1299             binding_value.get('key'): binding_value.get('value')
1300             for binding_value in traverse_obj(status, ('card', 'binding_values', ..., {dict}))
1301         }
1302         if binding_values:
1303             status['card']['binding_values'] = binding_values
1304
1305         return status
1306
1307     def _build_graphql_query(self, media_id):
1308         return {
1309             'variables': {
1310                 'focalTweetId': media_id,
1311                 'includePromotedContent': True,
1312                 'with_rux_injections': False,
1313                 'withBirdwatchNotes': True,
1314                 'withCommunity': True,
1315                 'withDownvotePerspective': False,
1316                 'withQuickPromoteEligibilityTweetFields': True,
1317                 'withReactionsMetadata': False,
1318                 'withReactionsPerspective': False,
1319                 'withSuperFollowsTweetFields': True,
1320                 'withSuperFollowsUserFields': True,
1321                 'withV2Timeline': True,
1322                 'withVoice': True,
1323             },
1324             'features': {
1325                 'graphql_is_translatable_rweb_tweet_is_translatable_enabled': False,
1326                 'interactive_text_enabled': True,
1327                 'responsive_web_edit_tweet_api_enabled': True,
1328                 'responsive_web_enhance_cards_enabled': True,
1329                 'responsive_web_graphql_timeline_navigation_enabled': False,
1330                 'responsive_web_text_conversations_enabled': False,
1331                 'responsive_web_uc_gql_enabled': True,
1332                 'standardized_nudges_misinfo': True,
1333                 'tweet_with_visibility_results_prefer_gql_limited_actions_policy_enabled': False,
1334                 'tweetypie_unmention_optimization_enabled': True,
1335                 'unified_cards_ad_metadata_container_dynamic_card_content_query_enabled': True,
1336                 'verified_phone_label_enabled': False,
1337                 'vibe_api_enabled': True,
1338             },
1339         } if self.is_logged_in else {
1340             'variables': {
1341                 'tweetId': media_id,
1342                 'withCommunity': False,
1343                 'includePromotedContent': False,
1344                 'withVoice': False,
1345             },
1346             'features': {
1347                 'creator_subscriptions_tweet_preview_api_enabled': True,
1348                 'tweetypie_unmention_optimization_enabled': True,
1349                 'responsive_web_edit_tweet_api_enabled': True,
1350                 'graphql_is_translatable_rweb_tweet_is_translatable_enabled': True,
1351                 'view_counts_everywhere_api_enabled': True,
1352                 'longform_notetweets_consumption_enabled': True,
1353                 'responsive_web_twitter_article_tweet_consumption_enabled': False,
1354                 'tweet_awards_web_tipping_enabled': False,
1355                 'freedom_of_speech_not_reach_fetch_enabled': True,
1356                 'standardized_nudges_misinfo': True,
1357                 'tweet_with_visibility_results_prefer_gql_limited_actions_policy_enabled': True,
1358                 'longform_notetweets_rich_text_read_enabled': True,
1359                 'longform_notetweets_inline_media_enabled': True,
1360                 'responsive_web_graphql_exclude_directive_enabled': True,
1361                 'verified_phone_label_enabled': False,
1362                 'responsive_web_media_download_video_enabled': False,
1363                 'responsive_web_graphql_skip_user_profile_image_extensions_enabled': False,
1364                 'responsive_web_graphql_timeline_navigation_enabled': True,
1365                 'responsive_web_enhance_cards_enabled': False,
1366             },
1367             'fieldToggles': {
1368                 'withArticleRichContentState': False,
1369             },
1370         }
1371
1372     def _call_syndication_api(self, twid):
1373         self.report_warning(
1374             'Not all metadata or media is available via syndication endpoint', twid, only_once=True)
1375         status = self._download_json(
1376             'https://cdn.syndication.twimg.com/tweet-result', twid, 'Downloading syndication JSON',
1377             headers={'User-Agent': 'Googlebot'}, query={
1378                 'id': twid,
1379                 # TODO: token = ((Number(twid) / 1e15) * Math.PI).toString(36).replace(/(0+|\.)/g, '')
1380                 'token': ''.join(random.choices('123456789abcdefghijklmnopqrstuvwxyz', k=10)),
1381             })
1382         if not status:
1383             raise ExtractorError('Syndication endpoint returned empty JSON response')
1384         # Transform the result so its structure matches that of legacy/graphql
1385         media = []
1386         for detail in traverse_obj(status, ((None, 'quoted_tweet'), 'mediaDetails', ..., {dict})):
1387             detail['id_str'] = traverse_obj(detail, (
1388                 'video_info', 'variants', ..., 'url', {self._MEDIA_ID_RE.search}, 1), get_all=False) or twid
1389             media.append(detail)
1390         status['extended_entities'] = {'media': media}
1391
1392         return status
1393
1394     def _extract_status(self, twid):
1395         if self._selected_api not in ('graphql', 'legacy', 'syndication'):
1396             raise ExtractorError(f'{self._selected_api!r} is not a valid API selection', expected=True)
1397
1398         try:
1399             if self.is_logged_in or self._selected_api == 'graphql':
1400                 status = self._graphql_to_legacy(self._call_graphql_api(self._GRAPHQL_ENDPOINT, twid), twid)
1401             elif self._selected_api == 'legacy':
1402                 status = self._call_api(f'statuses/show/{twid}.json', twid, {
1403                     'cards_platform': 'Web-12',
1404                     'include_cards': 1,
1405                     'include_reply_count': 1,
1406                     'include_user_entities': 0,
1407                     'tweet_mode': 'extended',
1408                 })
1409         except ExtractorError as e:
1410             if not isinstance(e.cause, HTTPError) or not e.cause.status == 429:
1411                 raise
1412             self.report_warning('Rate-limit exceeded; falling back to syndication endpoint')
1413             status = self._call_syndication_api(twid)
1414
1415         if self._selected_api == 'syndication':
1416             status = self._call_syndication_api(twid)
1417
1418         return traverse_obj(status, 'retweeted_status', None, expected_type=dict) or {}
1419
1420     def _real_extract(self, url):
1421         twid, selected_index = self._match_valid_url(url).group('id', 'index')
1422         status = self._extract_status(twid)
1423
1424         title = description = traverse_obj(
1425             status, (('full_text', 'text'), {lambda x: x.replace('\n', ' ')}), get_all=False) or ''
1426         # strip  'https -_t.co_BJYgOjSeGA' junk from filenames
1427         title = re.sub(r'\s+(https?://[^ ]+)', '', title)
1428         user = status.get('user') or {}
1429         uploader = user.get('name')
1430         if uploader:
1431             title = f'{uploader} - {title}'
1432         uploader_id = user.get('screen_name')
1433
1434         info = {
1435             'id': twid,
1436             'title': title,
1437             'description': description,
1438             'uploader': uploader,
1439             'timestamp': unified_timestamp(status.get('created_at')),
1440             'channel_id': str_or_none(status.get('user_id_str')) or str_or_none(user.get('id_str')),
1441             'uploader_id': uploader_id,
1442             'uploader_url': format_field(uploader_id, None, 'https://twitter.com/%s'),
1443             'like_count': int_or_none(status.get('favorite_count')),
1444             'repost_count': int_or_none(status.get('retweet_count')),
1445             'comment_count': int_or_none(status.get('reply_count')),
1446             'age_limit': 18 if status.get('possibly_sensitive') else 0,
1447             'tags': traverse_obj(status, ('entities', 'hashtags', ..., 'text')),
1448         }
1449
1450         def extract_from_video_info(media):
1451             media_id = traverse_obj(media, 'id_str', 'id', expected_type=str_or_none)
1452             self.write_debug(f'Extracting from video info: {media_id}')
1453
1454             formats = []
1455             subtitles = {}
1456             for variant in traverse_obj(media, ('video_info', 'variants', ...)):
1457                 fmts, subs = self._extract_variant_formats(variant, twid)
1458                 subtitles = self._merge_subtitles(subtitles, subs)
1459                 formats.extend(fmts)
1460
1461             thumbnails = []
1462             media_url = media.get('media_url_https') or media.get('media_url')
1463             if media_url:
1464                 def add_thumbnail(name, size):
1465                     thumbnails.append({
1466                         'id': name,
1467                         'url': update_url_query(media_url, {'name': name}),
1468                         'width': int_or_none(size.get('w') or size.get('width')),
1469                         'height': int_or_none(size.get('h') or size.get('height')),
1470                     })
1471                 for name, size in media.get('sizes', {}).items():
1472                     add_thumbnail(name, size)
1473                 add_thumbnail('orig', media.get('original_info') or {})
1474
1475             return {
1476                 'id': media_id,
1477                 'formats': formats,
1478                 'subtitles': subtitles,
1479                 'thumbnails': thumbnails,
1480                 'view_count': traverse_obj(media, ('mediaStats', 'viewCount', {int_or_none})),  # No longer available
1481                 'duration': float_or_none(traverse_obj(media, ('video_info', 'duration_millis')), 1000),
1482                 # Prioritize m3u8 formats for compat, see https://github.com/yt-dlp/yt-dlp/issues/8117
1483                 '_format_sort_fields': ('res', 'proto:m3u8', 'br', 'size'),  # http format codec is unknown
1484             }
1485
1486         def extract_from_card_info(card):
1487             if not card:
1488                 return
1489
1490             self.write_debug(f'Extracting from card info: {card.get("url")}')
1491             binding_values = card['binding_values']
1492
1493             def get_binding_value(k):
1494                 o = binding_values.get(k) or {}
1495                 return try_get(o, lambda x: x[x['type'].lower() + '_value'])
1496
1497             card_name = card['name'].split(':')[-1]
1498             if card_name == 'player':
1499                 yield {
1500                     '_type': 'url',
1501                     'url': get_binding_value('player_url'),
1502                 }
1503             elif card_name == 'periscope_broadcast':
1504                 yield {
1505                     '_type': 'url',
1506                     'url': get_binding_value('url') or get_binding_value('player_url'),
1507                     'ie_key': PeriscopeIE.ie_key(),
1508                 }
1509             elif card_name == 'broadcast':
1510                 yield {
1511                     '_type': 'url',
1512                     'url': get_binding_value('broadcast_url'),
1513                     'ie_key': TwitterBroadcastIE.ie_key(),
1514                 }
1515             elif card_name == 'audiospace':
1516                 yield {
1517                     '_type': 'url',
1518                     'url': f'https://twitter.com/i/spaces/{get_binding_value("id")}',
1519                     'ie_key': TwitterSpacesIE.ie_key(),
1520                 }
1521             elif card_name == 'summary':
1522                 yield {
1523                     '_type': 'url',
1524                     'url': get_binding_value('card_url'),
1525                 }
1526             elif card_name == 'unified_card':
1527                 unified_card = self._parse_json(get_binding_value('unified_card'), twid)
1528                 yield from map(extract_from_video_info, traverse_obj(
1529                     unified_card, ('media_entities', ...), expected_type=dict))
1530             # amplify, promo_video_website, promo_video_convo, appplayer,
1531             # video_direct_message, poll2choice_video, poll3choice_video,
1532             # poll4choice_video, ...
1533             else:
1534                 is_amplify = card_name == 'amplify'
1535                 vmap_url = get_binding_value('amplify_url_vmap') if is_amplify else get_binding_value('player_stream_url')
1536                 content_id = get_binding_value('%s_content_id' % (card_name if is_amplify else 'player'))
1537                 formats, subtitles = self._extract_formats_from_vmap_url(vmap_url, content_id or twid)
1538
1539                 thumbnails = []
1540                 for suffix in ('_small', '', '_large', '_x_large', '_original'):
1541                     image = get_binding_value('player_image' + suffix) or {}
1542                     image_url = image.get('url')
1543                     if not image_url or '/player-placeholder' in image_url:
1544                         continue
1545                     thumbnails.append({
1546                         'id': suffix[1:] if suffix else 'medium',
1547                         'url': image_url,
1548                         'width': int_or_none(image.get('width')),
1549                         'height': int_or_none(image.get('height')),
1550                     })
1551
1552                 yield {
1553                     'formats': formats,
1554                     'subtitles': subtitles,
1555                     'thumbnails': thumbnails,
1556                     'duration': int_or_none(get_binding_value(
1557                         'content_duration_seconds')),
1558                 }
1559
1560         videos = traverse_obj(status, (
1561             (None, 'quoted_status'), 'extended_entities', 'media', lambda _, m: m['type'] != 'photo', {dict}))
1562
1563         if self._yes_playlist(twid, selected_index, video_label='URL-specified video number'):
1564             selected_entries = (*map(extract_from_video_info, videos), *extract_from_card_info(status.get('card')))
1565         else:
1566             desired_obj = traverse_obj(status, (
1567                 (None, 'quoted_status'), 'extended_entities', 'media', int(selected_index) - 1, {dict}), get_all=False)
1568             if not desired_obj:
1569                 raise ExtractorError(f'Video #{selected_index} is unavailable', expected=True)
1570             elif desired_obj.get('type') != 'video':
1571                 raise ExtractorError(f'Media #{selected_index} is not a video', expected=True)
1572
1573             # Restore original archive id and video index in title
1574             for index, entry in enumerate(videos, 1):
1575                 if entry.get('id') != desired_obj.get('id'):
1576                     continue
1577                 if index == 1:
1578                     info['_old_archive_ids'] = [make_archive_id(self, twid)]
1579                 if len(videos) != 1:
1580                     info['title'] += f' #{index}'
1581                 break
1582
1583             return {**info, **extract_from_video_info(desired_obj), 'display_id': twid}
1584
1585         entries = [{**info, **data, 'display_id': twid} for data in selected_entries]
1586         if not entries:
1587             expanded_url = traverse_obj(status, ('entities', 'urls', 0, 'expanded_url'), expected_type=url_or_none)
1588             if not expanded_url or expanded_url == url:
1589                 self.raise_no_formats('No video could be found in this tweet', expected=True)
1590                 return info
1591
1592             return self.url_result(expanded_url, display_id=twid, **info)
1593
1594         entries[0]['_old_archive_ids'] = [make_archive_id(self, twid)]
1595
1596         if len(entries) == 1:
1597             return entries[0]
1598
1599         for index, entry in enumerate(entries, 1):
1600             entry['title'] += f' #{index}'
1601
1602         return self.playlist_result(entries, **info)
1603
1604
1605 class TwitterAmplifyIE(TwitterBaseIE):
1606     IE_NAME = 'twitter:amplify'
1607     _VALID_URL = r'https?://amp\.twimg\.com/v/(?P<id>[0-9a-f\-]{36})'
1608
1609     _TEST = {
1610         'url': 'https://amp.twimg.com/v/0ba0c3c7-0af3-4c0a-bed5-7efd1ffa2951',
1611         'md5': 'fec25801d18a4557c5c9f33d2c379ffa',
1612         'info_dict': {
1613             'id': '0ba0c3c7-0af3-4c0a-bed5-7efd1ffa2951',
1614             'ext': 'mp4',
1615             'title': 'Twitter Video',
1616             'thumbnail': 're:^https?://.*',
1617         },
1618         'params': {'format': '[protocol=https]'},
1619     }
1620
1621     def _real_extract(self, url):
1622         video_id = self._match_id(url)
1623         webpage = self._download_webpage(url, video_id)
1624
1625         vmap_url = self._html_search_meta(
1626             'twitter:amplify:vmap', webpage, 'vmap url')
1627         formats, _ = self._extract_formats_from_vmap_url(vmap_url, video_id)
1628
1629         thumbnails = []
1630         thumbnail = self._html_search_meta(
1631             'twitter:image:src', webpage, 'thumbnail', fatal=False)
1632
1633         def _find_dimension(target):
1634             w = int_or_none(self._html_search_meta(
1635                 f'twitter:{target}:width', webpage, fatal=False))
1636             h = int_or_none(self._html_search_meta(
1637                 f'twitter:{target}:height', webpage, fatal=False))
1638             return w, h
1639
1640         if thumbnail:
1641             thumbnail_w, thumbnail_h = _find_dimension('image')
1642             thumbnails.append({
1643                 'url': thumbnail,
1644                 'width': thumbnail_w,
1645                 'height': thumbnail_h,
1646             })
1647
1648         video_w, video_h = _find_dimension('player')
1649         formats[0].update({
1650             'width': video_w,
1651             'height': video_h,
1652         })
1653
1654         return {
1655             'id': video_id,
1656             'title': 'Twitter Video',
1657             'formats': formats,
1658             'thumbnails': thumbnails,
1659         }
1660
1661
1662 class TwitterBroadcastIE(TwitterBaseIE, PeriscopeBaseIE):
1663     IE_NAME = 'twitter:broadcast'
1664     _VALID_URL = TwitterBaseIE._BASE_REGEX + r'i/broadcasts/(?P<id>[0-9a-zA-Z]{13})'
1665
1666     _TESTS = [{
1667         # untitled Periscope video
1668         'url': 'https://twitter.com/i/broadcasts/1yNGaQLWpejGj',
1669         'info_dict': {
1670             'id': '1yNGaQLWpejGj',
1671             'ext': 'mp4',
1672             'title': 'Andrea May Sahouri - Periscope Broadcast',
1673             'uploader': 'Andrea May Sahouri',
1674             'uploader_id': 'andreamsahouri',
1675             'uploader_url': 'https://twitter.com/andreamsahouri',
1676             'timestamp': 1590973638,
1677             'upload_date': '20200601',
1678             'thumbnail': r're:^https?://[^?#]+\.jpg\?token=',
1679             'view_count': int,
1680         },
1681     }, {
1682         'url': 'https://twitter.com/i/broadcasts/1ZkKzeyrPbaxv',
1683         'info_dict': {
1684             'id': '1ZkKzeyrPbaxv',
1685             'ext': 'mp4',
1686             'title': 'Starship | SN10 | High-Altitude Flight Test',
1687             'uploader': 'SpaceX',
1688             'uploader_id': 'SpaceX',
1689             'uploader_url': 'https://twitter.com/SpaceX',
1690             'timestamp': 1614812942,
1691             'upload_date': '20210303',
1692             'thumbnail': r're:^https?://[^?#]+\.jpg\?token=',
1693             'view_count': int,
1694         },
1695     }, {
1696         'url': 'https://twitter.com/i/broadcasts/1OyKAVQrgzwGb',
1697         'info_dict': {
1698             'id': '1OyKAVQrgzwGb',
1699             'ext': 'mp4',
1700             'title': 'Starship Flight Test',
1701             'uploader': 'SpaceX',
1702             'uploader_id': 'SpaceX',
1703             'uploader_url': 'https://twitter.com/SpaceX',
1704             'timestamp': 1681993964,
1705             'upload_date': '20230420',
1706             'thumbnail': r're:^https?://[^?#]+\.jpg\?token=',
1707             'view_count': int,
1708         },
1709     }]
1710
1711     def _real_extract(self, url):
1712         broadcast_id = self._match_id(url)
1713         broadcast = self._call_api(
1714             'broadcasts/show.json', broadcast_id,
1715             {'ids': broadcast_id})['broadcasts'][broadcast_id]
1716         if not broadcast:
1717             raise ExtractorError('Broadcast no longer exists', expected=True)
1718         info = self._parse_broadcast_data(broadcast, broadcast_id)
1719         info['title'] = broadcast.get('status') or info.get('title')
1720         info['uploader_id'] = broadcast.get('twitter_username') or info.get('uploader_id')
1721         info['uploader_url'] = format_field(broadcast, 'twitter_username', 'https://twitter.com/%s', default=None)
1722         if info['live_status'] == 'is_upcoming':
1723             return info
1724
1725         media_key = broadcast['media_key']
1726         source = self._call_api(
1727             f'live_video_stream/status/{media_key}', media_key)['source']
1728         m3u8_url = source.get('noRedirectPlaybackUrl') or source['location']
1729         if '/live_video_stream/geoblocked/' in m3u8_url:
1730             self.raise_geo_restricted()
1731         m3u8_id = urllib.parse.parse_qs(urllib.parse.urlparse(
1732             m3u8_url).query).get('type', [None])[0]
1733         state, width, height = self._extract_common_format_info(broadcast)
1734         info['formats'] = self._extract_pscp_m3u8_formats(
1735             m3u8_url, broadcast_id, m3u8_id, state, width, height)
1736         return info
1737
1738
1739 class TwitterSpacesIE(TwitterBaseIE):
1740     IE_NAME = 'twitter:spaces'
1741     _VALID_URL = TwitterBaseIE._BASE_REGEX + r'i/spaces/(?P<id>[0-9a-zA-Z]{13})'
1742
1743     _TESTS = [{
1744         'url': 'https://twitter.com/i/spaces/1RDxlgyvNXzJL',
1745         'info_dict': {
1746             'id': '1RDxlgyvNXzJL',
1747             'ext': 'm4a',
1748             'title': 'King Carlo e la mossa Kansas City per fare il Grande Centro',
1749             'description': 'Twitter Space participated by annarita digiorgio, Signor Ernesto, Raffaello Colosimo, Simone M. Sepe',
1750             'uploader': r're:Lucio Di Gaetano.*?',
1751             'uploader_id': 'luciodigaetano',
1752             'live_status': 'was_live',
1753             'timestamp': 1659877956,
1754             'upload_date': '20220807',
1755             'release_timestamp': 1659904215,
1756             'release_date': '20220807',
1757         },
1758         'skip': 'No longer available',
1759     }, {
1760         # post_live/TimedOut but downloadable
1761         'url': 'https://twitter.com/i/spaces/1vAxRAVQWONJl',
1762         'info_dict': {
1763             'id': '1vAxRAVQWONJl',
1764             'ext': 'm4a',
1765             'title': 'Framing Up FinOps: Billing Tools',
1766             'description': 'Twitter Space participated by rupa, Alfonso Hernandez',
1767             'uploader': 'Google Cloud',
1768             'uploader_id': 'googlecloud',
1769             'live_status': 'post_live',
1770             'timestamp': 1681409554,
1771             'upload_date': '20230413',
1772             'release_timestamp': 1681839000,
1773             'release_date': '20230418',
1774             'protocol': 'm3u8',  # ffmpeg is forced
1775             'container': 'm4a_dash',  # audio-only format fixup is applied
1776         },
1777         'params': {'skip_download': 'm3u8'},
1778     }, {
1779         # Needs ffmpeg as downloader, see: https://github.com/yt-dlp/yt-dlp/issues/7536
1780         'url': 'https://twitter.com/i/spaces/1eaKbrQbjoRKX',
1781         'info_dict': {
1782             'id': '1eaKbrQbjoRKX',
1783             'ext': 'm4a',
1784             'title': 'あ',
1785             'description': 'Twitter Space participated by nobody yet',
1786             'uploader': '息根とめる',
1787             'uploader_id': 'tomeru_ikinone',
1788             'live_status': 'was_live',
1789             'timestamp': 1685617198,
1790             'upload_date': '20230601',
1791             'protocol': 'm3u8',  # ffmpeg is forced
1792             'container': 'm4a_dash',  # audio-only format fixup is applied
1793         },
1794         'params': {'skip_download': 'm3u8'},
1795     }, {
1796         # Video Space
1797         'url': 'https://x.com/i/spaces/1DXGydznBYWKM',
1798         'info_dict': {
1799             'id': '1DXGydznBYWKM',
1800             'ext': 'mp4',
1801             'title': 'America and Israel’s “special relationship”',
1802             'description': 'Twitter Space participated by nobody yet',
1803             'uploader': 'Candace Owens',
1804             'uploader_id': 'RealCandaceO',
1805             'live_status': 'was_live',
1806             'timestamp': 1723931351,
1807             'upload_date': '20240817',
1808             'release_timestamp': 1723932000,
1809             'release_date': '20240817',
1810             'protocol': 'm3u8_native',  # not ffmpeg, detected as video space
1811         },
1812         'params': {'skip_download': 'm3u8'},
1813     }]
1814
1815     SPACE_STATUS = {
1816         'notstarted': 'is_upcoming',
1817         'ended': 'was_live',
1818         'running': 'is_live',
1819         'timedout': 'post_live',
1820     }
1821
1822     def _build_graphql_query(self, space_id):
1823         return {
1824             'variables': {
1825                 'id': space_id,
1826                 'isMetatagsQuery': True,
1827                 'withDownvotePerspective': False,
1828                 'withReactionsMetadata': False,
1829                 'withReactionsPerspective': False,
1830                 'withReplays': True,
1831                 'withSuperFollowsUserFields': True,
1832                 'withSuperFollowsTweetFields': True,
1833             },
1834             'features': {
1835                 'dont_mention_me_view_api_enabled': True,
1836                 'interactive_text_enabled': True,
1837                 'responsive_web_edit_tweet_api_enabled': True,
1838                 'responsive_web_enhance_cards_enabled': True,
1839                 'responsive_web_uc_gql_enabled': True,
1840                 'spaces_2022_h2_clipping': True,
1841                 'spaces_2022_h2_spaces_communities': False,
1842                 'standardized_nudges_misinfo': True,
1843                 'tweet_with_visibility_results_prefer_gql_limited_actions_policy_enabled': False,
1844                 'vibe_api_enabled': True,
1845             },
1846         }
1847
1848     def _real_extract(self, url):
1849         space_id = self._match_id(url)
1850         space_data = self._call_graphql_api('HPEisOmj1epUNLCWTYhUWw/AudioSpaceById', space_id)['audioSpace']
1851         if not space_data:
1852             raise ExtractorError('Twitter Space not found', expected=True)
1853
1854         metadata = space_data['metadata']
1855         live_status = try_call(lambda: self.SPACE_STATUS[metadata['state'].lower()])
1856         is_live = live_status == 'is_live'
1857
1858         formats = []
1859         headers = {'Referer': 'https://twitter.com/'}
1860         if live_status == 'is_upcoming':
1861             self.raise_no_formats('Twitter Space not started yet', expected=True)
1862         elif not is_live and not metadata.get('is_space_available_for_replay'):
1863             self.raise_no_formats('Twitter Space ended and replay is disabled', expected=True)
1864         elif metadata.get('media_key'):
1865             source = traverse_obj(
1866                 self._call_api(f'live_video_stream/status/{metadata["media_key"]}', metadata['media_key']),
1867                 ('source', ('noRedirectPlaybackUrl', 'location'), {url_or_none}), get_all=False)
1868             is_audio_space = source and 'audio-space' in source
1869             formats = self._extract_m3u8_formats(
1870                 source, metadata['media_key'], 'm4a' if is_audio_space else 'mp4',
1871                 # XXX: Some audio-only Spaces need ffmpeg as downloader
1872                 entry_protocol='m3u8' if is_audio_space else 'm3u8_native',
1873                 live=is_live, headers=headers, fatal=False) if source else []
1874             if is_audio_space:
1875                 for fmt in formats:
1876                     fmt.update({'vcodec': 'none', 'acodec': 'aac'})
1877                     if not is_live:
1878                         fmt['container'] = 'm4a_dash'
1879
1880         participants = ', '.join(traverse_obj(
1881             space_data, ('participants', 'speakers', ..., 'display_name'))) or 'nobody yet'
1882
1883         if not formats and live_status == 'post_live':
1884             self.raise_no_formats('Twitter Space ended but not downloadable yet', expected=True)
1885
1886         return {
1887             'id': space_id,
1888             'title': metadata.get('title'),
1889             'description': f'Twitter Space participated by {participants}',
1890             'uploader': traverse_obj(
1891                 metadata, ('creator_results', 'result', 'legacy', 'name')),
1892             'uploader_id': traverse_obj(
1893                 metadata, ('creator_results', 'result', 'legacy', 'screen_name')),
1894             'live_status': live_status,
1895             'release_timestamp': try_call(
1896                 lambda: int_or_none(metadata['scheduled_start'], scale=1000)),
1897             'timestamp': int_or_none(metadata.get('created_at'), scale=1000),
1898             'formats': formats,
1899             'http_headers': headers,
1900         }
1901
1902
1903 class TwitterShortenerIE(TwitterBaseIE):
1904     IE_NAME = 'twitter:shortener'
1905     _VALID_URL = r'https?://t\.co/(?P<id>[^?#]+)|tco:(?P<eid>[^?#]+)'
1906     _BASE_URL = 'https://t.co/'
1907
1908     def _real_extract(self, url):
1909         mobj = self._match_valid_url(url)
1910         eid, shortcode = mobj.group('eid', 'id')
1911         if eid:
1912             shortcode = eid
1913             url = self._BASE_URL + shortcode
1914         new_url = self._request_webpage(url, shortcode, headers={'User-Agent': 'curl'}).url
1915         __UNSAFE_LINK = 'https://twitter.com/safety/unsafe_link_warning?unsafe_link='
1916         if new_url.startswith(__UNSAFE_LINK):
1917             new_url = new_url.replace(__UNSAFE_LINK, '')
1918         return self.url_result(new_url)