yt_dlp/extractor/tiktok.py

   1 import itertools
   2 import json
   3 import random
   4 import re
   5 import string
   6 import time
   7
   8 from .common import InfoExtractor
   9 from ..compat import compat_urllib_parse_urlparse
  10 from ..networking import HEADRequest
  11 from ..utils import (
  12     ExtractorError,
  13     LazyList,
  14     UnsupportedError,
  15     UserNotLive,
  16     determine_ext,
  17     format_field,
  18     int_or_none,
  19     join_nonempty,
  20     merge_dicts,
  21     qualities,
  22     remove_start,
  23     srt_subtitles_timecode,
  24     str_or_none,
  25     traverse_obj,
  26     try_call,
  27     try_get,
  28     url_or_none,
  29 )
  30
  31
  32 class TikTokBaseIE(InfoExtractor):
  33     _APP_VERSIONS = [('26.1.3', '260103'), ('26.1.2', '260102'), ('26.1.1', '260101'), ('25.6.2', '250602')]
  34     _WORKING_APP_VERSION = None
  35     _APP_NAME = 'trill'
  36     _AID = 1180
  37     _UPLOADER_URL_FORMAT = 'https://www.tiktok.com/@%s'
  38     _WEBPAGE_HOST = 'https://www.tiktok.com/'
  39     QUALITIES = ('360p', '540p', '720p', '1080p')
  40
  41     @property
  42     def _API_HOSTNAME(self):
  43         return self._configuration_arg(
  44             'api_hostname', ['api16-normal-c-useast1a.tiktokv.com'], ie_key=TikTokIE)[0]
  45
  46     @staticmethod
  47     def _create_url(user_id, video_id):
  48         return f'https://www.tiktok.com/@{user_id or "_"}/video/{video_id}'
  49
  50     def _get_sigi_state(self, webpage, display_id):
  51         return self._search_json(
  52             r'<script[^>]+\bid="(?:SIGI_STATE|sigi-persisted-data)"[^>]*>', webpage,
  53             'sigi state', display_id, end_pattern=r'</script>', default={})
  54
  55     def _get_universal_data(self, webpage, display_id):
  56         return traverse_obj(self._search_json(
  57             r'<script[^>]+\bid="__UNIVERSAL_DATA_FOR_REHYDRATION__"[^>]*>', webpage,
  58             'universal data', display_id, end_pattern=r'</script>', default={}),
  59             ('__DEFAULT_SCOPE__', {dict})) or {}
  60
  61     def _call_api_impl(self, ep, query, manifest_app_version, video_id, fatal=True,
  62                        note='Downloading API JSON', errnote='Unable to download API page'):
  63         self._set_cookie(self._API_HOSTNAME, 'odin_tt', ''.join(random.choices('0123456789abcdef', k=160)))
  64         webpage_cookies = self._get_cookies(self._WEBPAGE_HOST)
  65         if webpage_cookies.get('sid_tt'):
  66             self._set_cookie(self._API_HOSTNAME, 'sid_tt', webpage_cookies['sid_tt'].value)
  67         return self._download_json(
  68             'https://%s/aweme/v1/%s/' % (self._API_HOSTNAME, ep), video_id=video_id,
  69             fatal=fatal, note=note, errnote=errnote, headers={
  70                 'User-Agent': f'com.ss.android.ugc.{self._APP_NAME}/{manifest_app_version} (Linux; U; Android 13; en_US; Pixel 7; Build/TD1A.220804.031; Cronet/58.0.2991.0)',
  71                 'Accept': 'application/json',
  72             }, query=query)
  73
  74     def _build_api_query(self, query, app_version, manifest_app_version):
  75         return {
  76             **query,
  77             'version_name': app_version,
  78             'version_code': manifest_app_version,
  79             'build_number': app_version,
  80             'manifest_version_code': manifest_app_version,
  81             'update_version_code': manifest_app_version,
  82             'openudid': ''.join(random.choices('0123456789abcdef', k=16)),
  83             'uuid': ''.join(random.choices(string.digits, k=16)),
  84             '_rticket': int(time.time() * 1000),
  85             'ts': int(time.time()),
  86             'device_brand': 'Google',
  87             'device_type': 'Pixel 7',
  88             'device_platform': 'android',
  89             'resolution': '1080*2400',
  90             'dpi': 420,
  91             'os_version': '13',
  92             'os_api': '29',
  93             'carrier_region': 'US',
  94             'sys_region': 'US',
  95             'region': 'US',
  96             'app_name': self._APP_NAME,
  97             'app_language': 'en',
  98             'language': 'en',
  99             'timezone_name': 'America/New_York',
 100             'timezone_offset': '-14400',
 101             'channel': 'googleplay',
 102             'ac': 'wifi',
 103             'mcc_mnc': '310260',
 104             'is_my_cn': 0,
 105             'aid': self._AID,
 106             'ssmix': 'a',
 107             'as': 'a1qwert123',
 108             'cp': 'cbfhckdckkde1',
 109         }
 110
 111     def _call_api(self, ep, query, video_id, fatal=True,
 112                   note='Downloading API JSON', errnote='Unable to download API page'):
 113         if not self._WORKING_APP_VERSION:
 114             app_version = self._configuration_arg('app_version', [''], ie_key=TikTokIE.ie_key())[0]
 115             manifest_app_version = self._configuration_arg('manifest_app_version', [''], ie_key=TikTokIE.ie_key())[0]
 116             if app_version and manifest_app_version:
 117                 self._WORKING_APP_VERSION = (app_version, manifest_app_version)
 118                 self.write_debug('Imported app version combo from extractor arguments')
 119             elif app_version or manifest_app_version:
 120                 self.report_warning('Only one of the two required version params are passed as extractor arguments', only_once=True)
 121
 122         if self._WORKING_APP_VERSION:
 123             app_version, manifest_app_version = self._WORKING_APP_VERSION
 124             real_query = self._build_api_query(query, app_version, manifest_app_version)
 125             return self._call_api_impl(ep, real_query, manifest_app_version, video_id, fatal, note, errnote)
 126
 127         for count, (app_version, manifest_app_version) in enumerate(self._APP_VERSIONS, start=1):
 128             real_query = self._build_api_query(query, app_version, manifest_app_version)
 129             try:
 130                 res = self._call_api_impl(ep, real_query, manifest_app_version, video_id, fatal, note, errnote)
 131                 self._WORKING_APP_VERSION = (app_version, manifest_app_version)
 132                 return res
 133             except ExtractorError as e:
 134                 if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0:
 135                     if count == len(self._APP_VERSIONS):
 136                         if fatal:
 137                             raise e
 138                         else:
 139                             self.report_warning(str(e.cause or e.msg))
 140                             return
 141                     self.report_warning('%s. Retrying... (attempt %s of %s)' % (str(e.cause or e.msg), count, len(self._APP_VERSIONS)))
 142                     continue
 143                 raise e
 144
 145     def _extract_aweme_app(self, aweme_id):
 146         feed_list = self._call_api(
 147             'feed', {'aweme_id': aweme_id}, aweme_id, note='Downloading video feed',
 148             errnote='Unable to download video feed').get('aweme_list') or []
 149         aweme_detail = next((aweme for aweme in feed_list if str(aweme.get('aweme_id')) == aweme_id), None)
 150         if not aweme_detail:
 151             raise ExtractorError('Unable to find video in feed', video_id=aweme_id)
 152         return self._parse_aweme_video_app(aweme_detail)
 153
 154     def _get_subtitles(self, aweme_detail, aweme_id):
 155         # TODO: Extract text positioning info
 156         subtitles = {}
 157         # aweme/detail endpoint subs
 158         captions_info = traverse_obj(
 159             aweme_detail, ('interaction_stickers', ..., 'auto_video_caption_info', 'auto_captions', ...), expected_type=dict)
 160         for caption in captions_info:
 161             caption_url = traverse_obj(caption, ('url', 'url_list', ...), expected_type=url_or_none, get_all=False)
 162             if not caption_url:
 163                 continue
 164             caption_json = self._download_json(
 165                 caption_url, aweme_id, note='Downloading captions', errnote='Unable to download captions', fatal=False)
 166             if not caption_json:
 167                 continue
 168             subtitles.setdefault(caption.get('language', 'en'), []).append({
 169                 'ext': 'srt',
 170                 'data': '\n\n'.join(
 171                     f'{i + 1}\n{srt_subtitles_timecode(line["start_time"] / 1000)} --> {srt_subtitles_timecode(line["end_time"] / 1000)}\n{line["text"]}'
 172                     for i, line in enumerate(caption_json['utterances']) if line.get('text'))
 173             })
 174         # feed endpoint subs
 175         if not subtitles:
 176             for caption in traverse_obj(aweme_detail, ('video', 'cla_info', 'caption_infos', ...), expected_type=dict):
 177                 if not caption.get('url'):
 178                     continue
 179                 subtitles.setdefault(caption.get('lang') or 'en', []).append({
 180                     'ext': remove_start(caption.get('caption_format'), 'web'),
 181                     'url': caption['url'],
 182                 })
 183         # webpage subs
 184         if not subtitles:
 185             for caption in traverse_obj(aweme_detail, ('video', 'subtitleInfos', ...), expected_type=dict):
 186                 if not caption.get('Url'):
 187                     continue
 188                 subtitles.setdefault(caption.get('LanguageCodeName') or 'en', []).append({
 189                     'ext': remove_start(caption.get('Format'), 'web'),
 190                     'url': caption['Url'],
 191                 })
 192         return subtitles
 193
 194     def _parse_aweme_video_app(self, aweme_detail):
 195         aweme_id = aweme_detail['aweme_id']
 196         video_info = aweme_detail['video']
 197
 198         def parse_url_key(url_key):
 199             format_id, codec, res, bitrate = self._search_regex(
 200                 r'v[^_]+_(?P<id>(?P<codec>[^_]+)_(?P<res>\d+p)_(?P<bitrate>\d+))', url_key,
 201                 'url key', default=(None, None, None, None), group=('id', 'codec', 'res', 'bitrate'))
 202             if not format_id:
 203                 return {}, None
 204             return {
 205                 'format_id': format_id,
 206                 'vcodec': 'h265' if codec == 'bytevc1' else codec,
 207                 'tbr': int_or_none(bitrate, scale=1000) or None,
 208                 'quality': qualities(self.QUALITIES)(res),
 209             }, res
 210
 211         known_resolutions = {}
 212
 213         def audio_meta(url):
 214             ext = determine_ext(url, default_ext='m4a')
 215             return {
 216                 'format_note': 'Music track',
 217                 'ext': ext,
 218                 'acodec': 'aac' if ext == 'm4a' else ext,
 219                 'vcodec': 'none',
 220                 'width': None,
 221                 'height': None,
 222             } if ext == 'mp3' or '-music-' in url else {}
 223
 224         def extract_addr(addr, add_meta={}):
 225             parsed_meta, res = parse_url_key(addr.get('url_key', ''))
 226             if res:
 227                 known_resolutions.setdefault(res, {}).setdefault('height', int_or_none(addr.get('height')))
 228                 known_resolutions[res].setdefault('width', int_or_none(addr.get('width')))
 229                 parsed_meta.update(known_resolutions.get(res, {}))
 230                 add_meta.setdefault('height', int_or_none(res[:-1]))
 231             return [{
 232                 'url': url,
 233                 'filesize': int_or_none(addr.get('data_size')),
 234                 'ext': 'mp4',
 235                 'acodec': 'aac',
 236                 'source_preference': -2 if 'aweme/v1' in url else -1,  # Downloads from API might get blocked
 237                 **add_meta, **parsed_meta,
 238                 'format_note': join_nonempty(
 239                     add_meta.get('format_note'), '(API)' if 'aweme/v1' in url else None, delim=' '),
 240                 **audio_meta(url),
 241             } for url in addr.get('url_list') or []]
 242
 243         # Hack: Add direct video links first to prioritize them when removing duplicate formats
 244         formats = []
 245         width = int_or_none(video_info.get('width'))
 246         height = int_or_none(video_info.get('height'))
 247         if video_info.get('play_addr'):
 248             formats.extend(extract_addr(video_info['play_addr'], {
 249                 'format_id': 'play_addr',
 250                 'format_note': 'Direct video',
 251                 'vcodec': 'h265' if traverse_obj(
 252                     video_info, 'is_bytevc1', 'is_h265') else 'h264',  # TODO: Check for "direct iOS" videos, like https://www.tiktok.com/@cookierun_dev/video/7039716639834656002
 253                 'width': width,
 254                 'height': height,
 255             }))
 256         if video_info.get('download_addr'):
 257             download_addr = video_info['download_addr']
 258             dl_width = int_or_none(download_addr.get('width'))
 259             formats.extend(extract_addr(download_addr, {
 260                 'format_id': 'download_addr',
 261                 'format_note': 'Download video%s' % (', watermarked' if video_info.get('has_watermark') else ''),
 262                 'vcodec': 'h264',
 263                 'width': dl_width or width,
 264                 'height': try_call(lambda: int(dl_width / 0.5625)) or height,  # download_addr['height'] is wrong
 265                 'preference': -2 if video_info.get('has_watermark') else -1,
 266             }))
 267         if video_info.get('play_addr_h264'):
 268             formats.extend(extract_addr(video_info['play_addr_h264'], {
 269                 'format_id': 'play_addr_h264',
 270                 'format_note': 'Direct video',
 271                 'vcodec': 'h264',
 272             }))
 273         if video_info.get('play_addr_bytevc1'):
 274             formats.extend(extract_addr(video_info['play_addr_bytevc1'], {
 275                 'format_id': 'play_addr_bytevc1',
 276                 'format_note': 'Direct video',
 277                 'vcodec': 'h265',
 278             }))
 279
 280         for bitrate in video_info.get('bit_rate', []):
 281             if bitrate.get('play_addr'):
 282                 formats.extend(extract_addr(bitrate['play_addr'], {
 283                     'format_id': bitrate.get('gear_name'),
 284                     'format_note': 'Playback video',
 285                     'tbr': try_get(bitrate, lambda x: x['bit_rate'] / 1000),
 286                     'vcodec': 'h265' if traverse_obj(
 287                         bitrate, 'is_bytevc1', 'is_h265') else 'h264',
 288                     'fps': bitrate.get('FPS'),
 289                 }))
 290
 291         self._remove_duplicate_formats(formats)
 292         auth_cookie = self._get_cookies(self._WEBPAGE_HOST).get('sid_tt')
 293         if auth_cookie:
 294             for f in formats:
 295                 self._set_cookie(compat_urllib_parse_urlparse(f['url']).hostname, 'sid_tt', auth_cookie.value)
 296
 297         thumbnails = []
 298         for cover_id in ('cover', 'ai_dynamic_cover', 'animated_cover', 'ai_dynamic_cover_bak',
 299                          'origin_cover', 'dynamic_cover'):
 300             for cover_url in traverse_obj(video_info, (cover_id, 'url_list', ...)):
 301                 thumbnails.append({
 302                     'id': cover_id,
 303                     'url': cover_url,
 304                 })
 305
 306         stats_info = aweme_detail.get('statistics') or {}
 307         author_info = aweme_detail.get('author') or {}
 308         music_info = aweme_detail.get('music') or {}
 309         user_url = self._UPLOADER_URL_FORMAT % (traverse_obj(author_info,
 310                                                              'sec_uid', 'id', 'uid', 'unique_id',
 311                                                              expected_type=str_or_none, get_all=False))
 312         labels = traverse_obj(aweme_detail, ('hybrid_label', ..., 'text'), expected_type=str)
 313
 314         contained_music_track = traverse_obj(
 315             music_info, ('matched_song', 'title'), ('matched_pgc_sound', 'title'), expected_type=str)
 316         contained_music_author = traverse_obj(
 317             music_info, ('matched_song', 'author'), ('matched_pgc_sound', 'author'), 'author', expected_type=str)
 318
 319         is_generic_og_trackname = music_info.get('is_original_sound') and music_info.get('title') == 'original sound - %s' % music_info.get('owner_handle')
 320         if is_generic_og_trackname:
 321             music_track, music_author = contained_music_track or 'original sound', contained_music_author
 322         else:
 323             music_track, music_author = music_info.get('title'), traverse_obj(music_info, ('author', {str}))
 324
 325         return {
 326             'id': aweme_id,
 327             **traverse_obj(aweme_detail, {
 328                 'title': ('desc', {str}),
 329                 'description': ('desc', {str}),
 330                 'timestamp': ('create_time', {int_or_none}),
 331             }),
 332             **traverse_obj(stats_info, {
 333                 'view_count': 'play_count',
 334                 'like_count': 'digg_count',
 335                 'repost_count': 'share_count',
 336                 'comment_count': 'comment_count',
 337             }, expected_type=int_or_none),
 338             **traverse_obj(author_info, {
 339                 'uploader': ('unique_id', {str}),
 340                 'uploader_id': ('uid', {str_or_none}),
 341                 'creators': ('nickname', {str}, {lambda x: [x] if x else None}),  # for compat
 342                 'channel': ('nickname', {str}),
 343                 'channel_id': ('sec_uid', {str}),
 344             }),
 345             'uploader_url': user_url,
 346             'track': music_track,
 347             'album': str_or_none(music_info.get('album')) or None,
 348             'artists': re.split(r'(?:, | & )', music_author) if music_author else None,
 349             'formats': formats,
 350             'subtitles': self.extract_subtitles(aweme_detail, aweme_id),
 351             'thumbnails': thumbnails,
 352             'duration': int_or_none(traverse_obj(video_info, 'duration', ('download_addr', 'duration')), scale=1000),
 353             'availability': self._availability(
 354                 is_private='Private' in labels,
 355                 needs_subscription='Friends only' in labels,
 356                 is_unlisted='Followers only' in labels),
 357             '_format_sort_fields': ('quality', 'codec', 'size', 'br'),
 358         }
 359
 360     def _parse_aweme_video_web(self, aweme_detail, webpage_url, video_id):
 361         video_info = aweme_detail['video']
 362         author_info = traverse_obj(aweme_detail, 'authorInfo', 'author', expected_type=dict, default={})
 363         music_info = aweme_detail.get('music') or {}
 364         stats_info = aweme_detail.get('stats') or {}
 365         channel_id = traverse_obj(author_info or aweme_detail, (('authorSecId', 'secUid'), {str}), get_all=False)
 366         user_url = self._UPLOADER_URL_FORMAT % channel_id if channel_id else None
 367
 368         formats = []
 369         width = int_or_none(video_info.get('width'))
 370         height = int_or_none(video_info.get('height'))
 371
 372         for play_url in traverse_obj(video_info, ('playAddr', ((..., 'src'), None), {url_or_none})):
 373             formats.append({
 374                 'url': self._proto_relative_url(play_url),
 375                 'ext': 'mp4',
 376                 'width': width,
 377                 'height': height,
 378             })
 379
 380         for download_url in traverse_obj(video_info, (('downloadAddr', ('download', 'url')), {url_or_none})):
 381             formats.append({
 382                 'format_id': 'download',
 383                 'url': self._proto_relative_url(download_url),
 384                 'ext': 'mp4',
 385                 'width': width,
 386                 'height': height,
 387             })
 388
 389         self._remove_duplicate_formats(formats)
 390
 391         thumbnails = []
 392         for thumb_url in traverse_obj(aweme_detail, (
 393                 (None, 'video'), ('thumbnail', 'cover', 'dynamicCover', 'originCover'), {url_or_none})):
 394             thumbnails.append({
 395                 'url': self._proto_relative_url(thumb_url),
 396                 'width': width,
 397                 'height': height,
 398             })
 399
 400         return {
 401             'id': video_id,
 402             **traverse_obj(aweme_detail, {
 403                 'title': ('desc', {str}),
 404                 'description': ('desc', {str}),
 405                 'duration': ('video', 'duration', {int_or_none}),
 406                 'timestamp': ('createTime', {int_or_none}),
 407             }),
 408             **traverse_obj(author_info or aweme_detail, {
 409                 'creators': ('nickname', {str}, {lambda x: [x] if x else None}),  # for compat
 410                 'channel': ('nickname', {str}),
 411                 'uploader': (('uniqueId', 'author'), {str}),
 412                 'uploader_id': (('authorId', 'uid', 'id'), {str_or_none}),
 413             }, get_all=False),
 414             **traverse_obj(stats_info, {
 415                 'view_count': 'playCount',
 416                 'like_count': 'diggCount',
 417                 'repost_count': 'shareCount',
 418                 'comment_count': 'commentCount',
 419             }, expected_type=int_or_none),
 420             **traverse_obj(music_info, {
 421                 'track': ('title', {str}),
 422                 'album': ('album', {str}, {lambda x: x or None}),
 423                 'artists': ('authorName', {str}, {lambda x: [x] if x else None}),
 424             }),
 425             'channel_id': channel_id,
 426             'uploader_url': user_url,
 427             'formats': formats,
 428             'thumbnails': thumbnails,
 429             'http_headers': {
 430                 'Referer': webpage_url,
 431             }
 432         }
 433
 434
 435 class TikTokIE(TikTokBaseIE):
 436     _VALID_URL = r'https?://www\.tiktok\.com/(?:embed|@(?P<user_id>[\w\.-]+)?/video)/(?P<id>\d+)'
 437     _EMBED_REGEX = [rf'<(?:script|iframe)[^>]+\bsrc=(["\'])(?P<url>{_VALID_URL})']
 438
 439     _TESTS = [{
 440         'url': 'https://www.tiktok.com/@leenabhushan/video/6748451240264420610',
 441         'md5': '736bb7a466c6f0a6afeb597da1e6f5b7',
 442         'info_dict': {
 443             'id': '6748451240264420610',
 444             'ext': 'mp4',
 445             'title': '#jassmanak #lehanga #leenabhushan',
 446             'description': '#jassmanak #lehanga #leenabhushan',
 447             'duration': 13,
 448             'height': 1024,
 449             'width': 576,
 450             'uploader': 'leenabhushan',
 451             'uploader_id': '6691488002098119685',
 452             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA_Eb4t1vodM1IuTy_cvp9CY22RAb59xqrO0Xtz9CYQJvgXaDvZxYnZYRzDWhhgJmy',
 453             'creator': 'facestoriesbyleenabh',
 454             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 455             'upload_date': '20191016',
 456             'timestamp': 1571246252,
 457             'view_count': int,
 458             'like_count': int,
 459             'repost_count': int,
 460             'comment_count': int,
 461             'artist': 'Ysrbeats',
 462             'album': 'Lehanga',
 463             'track': 'Lehanga',
 464         },
 465         'skip': '404 Not Found',
 466     }, {
 467         'url': 'https://www.tiktok.com/@patroxofficial/video/6742501081818877190?langCountry=en',
 468         'md5': '6f3cf8cdd9b28cb8363fe0a9a160695b',
 469         'info_dict': {
 470             'id': '6742501081818877190',
 471             'ext': 'mp4',
 472             'title': 'md5:5e2a23877420bb85ce6521dbee39ba94',
 473             'description': 'md5:5e2a23877420bb85ce6521dbee39ba94',
 474             'duration': 27,
 475             'height': 960,
 476             'width': 540,
 477             'uploader': 'patrox',
 478             'uploader_id': '18702747',
 479             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAiFnldaILebi5heDoVU6bn4jBWWycX6-9U3xuNPqZ8Ws',
 480             'channel_id': 'MS4wLjABAAAAiFnldaILebi5heDoVU6bn4jBWWycX6-9U3xuNPqZ8Ws',
 481             'channel': 'patroX',
 482             'creators': ['patroX'],
 483             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 484             'upload_date': '20190930',
 485             'timestamp': 1569860870,
 486             'view_count': int,
 487             'like_count': int,
 488             'repost_count': int,
 489             'comment_count': int,
 490             'artists': ['Evan Todd', 'Jessica Keenan Wynn', 'Alice Lee', 'Barrett Wilbert Weed', 'Jon Eidson'],
 491             'track': 'Big Fun',
 492         },
 493     }, {
 494         # Banned audio, only available on the app
 495         'url': 'https://www.tiktok.com/@barudakhb_/video/6984138651336838402',
 496         'info_dict': {
 497             'id': '6984138651336838402',
 498             'ext': 'mp4',
 499             'title': 'Balas @yolaaftwsr hayu yu ? #SquadRandom_ 🔥',
 500             'description': 'Balas @yolaaftwsr hayu yu ? #SquadRandom_ 🔥',
 501             'uploader': 'barudakhb_',
 502             'channel': 'md5:29f238c49bc0c176cb3cef1a9cea9fa6',
 503             'creators': ['md5:29f238c49bc0c176cb3cef1a9cea9fa6'],
 504             'uploader_id': '6974687867511718913',
 505             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAbhBwQC-R1iKoix6jDFsF-vBdfx2ABoDjaZrM9fX6arU3w71q3cOWgWuTXn1soZ7d',
 506             'channel_id': 'MS4wLjABAAAAbhBwQC-R1iKoix6jDFsF-vBdfx2ABoDjaZrM9fX6arU3w71q3cOWgWuTXn1soZ7d',
 507             'track': 'Boka Dance',
 508             'artists': ['md5:29f238c49bc0c176cb3cef1a9cea9fa6'],
 509             'timestamp': 1626121503,
 510             'duration': 18,
 511             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 512             'upload_date': '20210712',
 513             'view_count': int,
 514             'like_count': int,
 515             'repost_count': int,
 516             'comment_count': int,
 517         },
 518     }, {
 519         # Sponsored video, only available with feed workaround
 520         'url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_/video/7042692929109986561',
 521         'info_dict': {
 522             'id': '7042692929109986561',
 523             'ext': 'mp4',
 524             'title': 'Slap and Run!',
 525             'description': 'Slap and Run!',
 526             'uploader': 'user440922249',
 527             'channel': 'Slap And Run',
 528             'creators': ['Slap And Run'],
 529             'uploader_id': '7036055384943690754',
 530             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_',
 531             'channel_id': 'MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_',
 532             'track': 'Promoted Music',
 533             'timestamp': 1639754738,
 534             'duration': 30,
 535             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 536             'upload_date': '20211217',
 537             'view_count': int,
 538             'like_count': int,
 539             'repost_count': int,
 540             'comment_count': int,
 541         },
 542         'params': {'skip_download': True},  # XXX: unable to download video data: HTTP Error 403: Forbidden
 543     }, {
 544         # Video without title and description
 545         'url': 'https://www.tiktok.com/@pokemonlife22/video/7059698374567611694',
 546         'info_dict': {
 547             'id': '7059698374567611694',
 548             'ext': 'mp4',
 549             'title': 'TikTok video #7059698374567611694',
 550             'description': '',
 551             'uploader': 'pokemonlife22',
 552             'channel': 'Pokemon',
 553             'creators': ['Pokemon'],
 554             'uploader_id': '6820838815978423302',
 555             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA0tF1nBwQVVMyrGu3CqttkNgM68Do1OXUFuCY0CRQk8fEtSVDj89HqoqvbSTmUP2W',
 556             'channel_id': 'MS4wLjABAAAA0tF1nBwQVVMyrGu3CqttkNgM68Do1OXUFuCY0CRQk8fEtSVDj89HqoqvbSTmUP2W',
 557             'track': 'original sound',
 558             'timestamp': 1643714123,
 559             'duration': 6,
 560             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 561             'upload_date': '20220201',
 562             'artists': ['Pokemon'],
 563             'view_count': int,
 564             'like_count': int,
 565             'repost_count': int,
 566             'comment_count': int,
 567         },
 568     }, {
 569         # hydration JSON is sent in a <script> element
 570         'url': 'https://www.tiktok.com/@denidil6/video/7065799023130643713',
 571         'info_dict': {
 572             'id': '7065799023130643713',
 573             'ext': 'mp4',
 574             'title': '#denidil#денидил',
 575             'description': '#denidil#денидил',
 576             'uploader': 'denidil6',
 577             'uploader_id': '7046664115636405250',
 578             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAsvMSzFdQ4ikl3uR2TEJwMBbB2yZh2Zxwhx-WCo3rbDpAharE3GQCrFuJArI3C8QJ',
 579             'artist': 'Holocron Music',
 580             'album': 'Wolf Sounds (1 Hour) Enjoy the Company of the Animal That Is the Majestic King of the Night',
 581             'track': 'Wolf Sounds (1 Hour) Enjoy the Company of the Animal That Is the Majestic King of the Night',
 582             'timestamp': 1645134536,
 583             'duration': 26,
 584             'upload_date': '20220217',
 585             'view_count': int,
 586             'like_count': int,
 587             'repost_count': int,
 588             'comment_count': int,
 589         },
 590         'skip': 'This video is unavailable',
 591     }, {
 592         # slideshow audio-only mp3 format
 593         'url': 'https://www.tiktok.com/@_le_cannibale_/video/7139980461132074283',
 594         'info_dict': {
 595             'id': '7139980461132074283',
 596             'ext': 'mp3',
 597             'title': 'TikTok video #7139980461132074283',
 598             'description': '',
 599             'channel': 'Antaura',
 600             'creators': ['Antaura'],
 601             'uploader': '_le_cannibale_',
 602             'uploader_id': '6604511138619654149',
 603             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAoShJqaw_5gvy48y3azFeFcT4jeyKWbB0VVYasOCt2tTLwjNFIaDcHAM4D-QGXFOP',
 604             'channel_id': 'MS4wLjABAAAAoShJqaw_5gvy48y3azFeFcT4jeyKWbB0VVYasOCt2tTLwjNFIaDcHAM4D-QGXFOP',
 605             'artists': ['nathan !'],
 606             'track': 'grahamscott canon',
 607             'upload_date': '20220905',
 608             'timestamp': 1662406249,
 609             'view_count': int,
 610             'like_count': int,
 611             'repost_count': int,
 612             'comment_count': int,
 613             'thumbnail': r're:^https://.+\.(?:webp|jpe?g)',
 614         },
 615     }, {
 616         # only available via web
 617         'url': 'https://www.tiktok.com/@moxypatch/video/7206382937372134662',  # FIXME
 618         'md5': '6aba7fad816e8709ff2c149679ace165',
 619         'info_dict': {
 620             'id': '7206382937372134662',
 621             'ext': 'mp4',
 622             'title': 'md5:1d95c0b96560ca0e8a231af4172b2c0a',
 623             'description': 'md5:1d95c0b96560ca0e8a231af4172b2c0a',
 624             'channel': 'MoxyPatch',
 625             'creators': ['MoxyPatch'],
 626             'uploader': 'moxypatch',
 627             'uploader_id': '7039142049363379205',
 628             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAFhqKnngMHJSsifL0w1vFOP5kn3Ndo1ODp0XuIBkNMBCkALTvwILdpu12g3pTtL4V',
 629             'channel_id': 'MS4wLjABAAAAFhqKnngMHJSsifL0w1vFOP5kn3Ndo1ODp0XuIBkNMBCkALTvwILdpu12g3pTtL4V',
 630             'artists': ['your worst nightmare'],
 631             'track': 'original sound',
 632             'upload_date': '20230303',
 633             'timestamp': 1677866781,
 634             'duration': 10,
 635             'view_count': int,
 636             'like_count': int,
 637             'repost_count': int,
 638             'comment_count': int,
 639             'thumbnail': r're:^https://.+',
 640             'thumbnails': 'count:3',
 641         },
 642         'expected_warnings': ['Unable to find video in feed'],
 643     }, {
 644         # 1080p format
 645         'url': 'https://www.tiktok.com/@tatemcrae/video/7107337212743830830',  # FIXME
 646         'md5': '982512017a8a917124d5a08c8ae79621',
 647         'info_dict': {
 648             'id': '7107337212743830830',
 649             'ext': 'mp4',
 650             'title': 'new music video 4 don’t come backkkk🧸🖤 i hope u enjoy !! @musicontiktok',
 651             'description': 'new music video 4 don’t come backkkk🧸🖤 i hope u enjoy !! @musicontiktok',
 652             'uploader': 'tatemcrae',
 653             'uploader_id': '86328792343818240',
 654             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA-0bQT0CqebTRr6I4IkYvMDMKSRSJHLNPBo5HrSklJwyA2psXLSZG5FP-LMNpHnJd',
 655             'channel_id': 'MS4wLjABAAAA-0bQT0CqebTRr6I4IkYvMDMKSRSJHLNPBo5HrSklJwyA2psXLSZG5FP-LMNpHnJd',
 656             'channel': 'tate mcrae',
 657             'creators': ['tate mcrae'],
 658             'artists': ['tate mcrae'],
 659             'track': 'original sound',
 660             'upload_date': '20220609',
 661             'timestamp': 1654805899,
 662             'duration': 150,
 663             'view_count': int,
 664             'like_count': int,
 665             'repost_count': int,
 666             'comment_count': int,
 667             'thumbnail': r're:^https://.+\.webp',
 668         },
 669         'skip': 'Unavailable via feed API, no formats available via web',
 670     }, {
 671         # Slideshow, audio-only m4a format
 672         'url': 'https://www.tiktok.com/@hara_yoimiya/video/7253412088251534594',
 673         'md5': '2ff8fe0174db2dbf49c597a7bef4e47d',
 674         'info_dict': {
 675             'id': '7253412088251534594',
 676             'ext': 'm4a',
 677             'title': 'я ред флаг простите #переписка #щитпост #тревожныйтиппривязанности #рекомендации ',
 678             'description': 'я ред флаг простите #переписка #щитпост #тревожныйтиппривязанности #рекомендации ',
 679             'uploader': 'hara_yoimiya',
 680             'uploader_id': '6582536342634676230',
 681             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAIAlDxriiPWLE-p8p1R_0Bx8qWKfi-7zwmGhzU8Mv25W8sNxjfIKrol31qTczzuLB',
 682             'channel_id': 'MS4wLjABAAAAIAlDxriiPWLE-p8p1R_0Bx8qWKfi-7zwmGhzU8Mv25W8sNxjfIKrol31qTczzuLB',
 683             'channel': 'лампочка',
 684             'creators': ['лампочка'],
 685             'artists': ['Øneheart'],
 686             'album': 'watching the stars',
 687             'track': 'watching the stars',
 688             'upload_date': '20230708',
 689             'timestamp': 1688816612,
 690             'view_count': int,
 691             'like_count': int,
 692             'comment_count': int,
 693             'repost_count': int,
 694             'thumbnail': r're:^https://.+\.(?:webp|jpe?g)',
 695         },
 696     }, {
 697         # Auto-captions available
 698         'url': 'https://www.tiktok.com/@hankgreen1/video/7047596209028074758',
 699         'only_matching': True
 700     }]
 701
 702     def _real_extract(self, url):
 703         video_id, user_id = self._match_valid_url(url).group('id', 'user_id')
 704         try:
 705             return self._extract_aweme_app(video_id)
 706         except ExtractorError as e:
 707             e.expected = True
 708             self.report_warning(f'{e}; trying with webpage')
 709
 710         url = self._create_url(user_id, video_id)
 711         webpage = self._download_webpage(url, video_id, headers={'User-Agent': 'Mozilla/5.0'})
 712
 713         if universal_data := self._get_universal_data(webpage, video_id):
 714             self.write_debug('Found universal data for rehydration')
 715             status = traverse_obj(universal_data, ('webapp.video-detail', 'statusCode', {int})) or 0
 716             video_data = traverse_obj(universal_data, ('webapp.video-detail', 'itemInfo', 'itemStruct', {dict}))
 717
 718         elif sigi_data := self._get_sigi_state(webpage, video_id):
 719             self.write_debug('Found sigi state data')
 720             status = traverse_obj(sigi_data, ('VideoPage', 'statusCode', {int})) or 0
 721             video_data = traverse_obj(sigi_data, ('ItemModule', video_id, {dict}))
 722
 723         elif next_data := self._search_nextjs_data(webpage, video_id, default='{}'):
 724             self.write_debug('Found next.js data')
 725             status = traverse_obj(next_data, ('props', 'pageProps', 'statusCode', {int})) or 0
 726             video_data = traverse_obj(next_data, ('props', 'pageProps', 'itemInfo', 'itemStruct', {dict}))
 727
 728         else:
 729             raise ExtractorError('Unable to extract webpage video data')
 730
 731         if video_data and status == 0:
 732             return self._parse_aweme_video_web(video_data, url, video_id)
 733         elif status == 10216:
 734             raise ExtractorError('This video is private', expected=True)
 735         raise ExtractorError(f'Video not available, status code {status}', video_id=video_id)
 736
 737
 738 class TikTokUserIE(TikTokBaseIE):
 739     IE_NAME = 'tiktok:user'
 740     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/@(?P<id>[\w\.-]+)/?(?:$|[#?])'
 741     _WORKING = False
 742     _TESTS = [{
 743         'url': 'https://tiktok.com/@corgibobaa?lang=en',
 744         'playlist_mincount': 45,
 745         'info_dict': {
 746             'id': '6935371178089399301',
 747             'title': 'corgibobaa',
 748             'thumbnail': r're:https://.+_1080x1080\.webp'
 749         },
 750         'expected_warnings': ['Retrying']
 751     }, {
 752         'url': 'https://www.tiktok.com/@6820838815978423302',
 753         'playlist_mincount': 5,
 754         'info_dict': {
 755             'id': '6820838815978423302',
 756             'title': '6820838815978423302',
 757             'thumbnail': r're:https://.+_1080x1080\.webp'
 758         },
 759         'expected_warnings': ['Retrying']
 760     }, {
 761         'url': 'https://www.tiktok.com/@meme',
 762         'playlist_mincount': 593,
 763         'info_dict': {
 764             'id': '79005827461758976',
 765             'title': 'meme',
 766             'thumbnail': r're:https://.+_1080x1080\.webp'
 767         },
 768         'expected_warnings': ['Retrying']
 769     }]
 770
 771     r'''  # TODO: Fix by adding _signature to api_url
 772     def _entries(self, webpage, user_id, username):
 773         secuid = self._search_regex(r'\"secUid\":\"(?P<secUid>[^\"]+)', webpage, username)
 774         verifyfp_cookie = self._get_cookies('https://www.tiktok.com').get('s_v_web_id')
 775         if not verifyfp_cookie:
 776             raise ExtractorError('Improper cookies (missing s_v_web_id).', expected=True)
 777         api_url = f'https://m.tiktok.com/api/post/item_list/?aid=1988&cookie_enabled=true&count=30&verifyFp={verifyfp_cookie.value}&secUid={secuid}&cursor='
 778         cursor = '0'
 779         for page in itertools.count():
 780             data_json = self._download_json(api_url + cursor, username, note='Downloading Page %d' % page)
 781             for video in data_json.get('itemList', []):
 782                 video_id = video['id']
 783                 video_url = f'https://www.tiktok.com/@{user_id}/video/{video_id}'
 784                 yield self._url_result(video_url, 'TikTok', video_id, str_or_none(video.get('desc')))
 785             if not data_json.get('hasMore'):
 786                 break
 787             cursor = data_json['cursor']
 788     '''
 789
 790     def _video_entries_api(self, webpage, user_id, username):
 791         query = {
 792             'user_id': user_id,
 793             'count': 21,
 794             'max_cursor': 0,
 795             'min_cursor': 0,
 796             'retry_type': 'no_retry',
 797             'device_id': ''.join(random.choices(string.digits, k=19)),  # Some endpoints don't like randomized device_id, so it isn't directly set in _call_api.
 798         }
 799
 800         for page in itertools.count(1):
 801             for retry in self.RetryManager():
 802                 try:
 803                     post_list = self._call_api(
 804                         'aweme/post', query, username, note=f'Downloading user video list page {page}',
 805                         errnote='Unable to download user video list')
 806                 except ExtractorError as e:
 807                     if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0:
 808                         retry.error = e
 809                         continue
 810                     raise
 811             yield from post_list.get('aweme_list', [])
 812             if not post_list.get('has_more'):
 813                 break
 814             query['max_cursor'] = post_list['max_cursor']
 815
 816     def _entries_api(self, user_id, videos):
 817         for video in videos:
 818             yield {
 819                 **self._parse_aweme_video_app(video),
 820                 'extractor_key': TikTokIE.ie_key(),
 821                 'extractor': 'TikTok',
 822                 'webpage_url': f'https://tiktok.com/@{user_id}/video/{video["aweme_id"]}',
 823             }
 824
 825     def _real_extract(self, url):
 826         user_name = self._match_id(url)
 827         webpage = self._download_webpage(url, user_name, headers={
 828             'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)'
 829         })
 830         user_id = self._html_search_regex(r'snssdk\d*://user/profile/(\d+)', webpage, 'user ID', default=None) or user_name
 831
 832         videos = LazyList(self._video_entries_api(webpage, user_id, user_name))
 833         thumbnail = traverse_obj(videos, (0, 'author', 'avatar_larger', 'url_list', 0))
 834
 835         return self.playlist_result(self._entries_api(user_id, videos), user_id, user_name, thumbnail=thumbnail)
 836
 837
 838 class TikTokBaseListIE(TikTokBaseIE):  # XXX: Conventionally, base classes should end with BaseIE/InfoExtractor
 839     def _entries(self, list_id, display_id):
 840         query = {
 841             self._QUERY_NAME: list_id,
 842             'cursor': 0,
 843             'count': 20,
 844             'type': 5,
 845             'device_id': ''.join(random.choices(string.digits, k=19))
 846         }
 847
 848         for page in itertools.count(1):
 849             for retry in self.RetryManager():
 850                 try:
 851                     post_list = self._call_api(
 852                         self._API_ENDPOINT, query, display_id, note=f'Downloading video list page {page}',
 853                         errnote='Unable to download video list')
 854                 except ExtractorError as e:
 855                     if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0:
 856                         retry.error = e
 857                         continue
 858                     raise
 859             for video in post_list.get('aweme_list', []):
 860                 yield {
 861                     **self._parse_aweme_video_app(video),
 862                     'extractor_key': TikTokIE.ie_key(),
 863                     'extractor': 'TikTok',
 864                     'webpage_url': f'https://tiktok.com/@_/video/{video["aweme_id"]}',
 865                 }
 866             if not post_list.get('has_more'):
 867                 break
 868             query['cursor'] = post_list['cursor']
 869
 870     def _real_extract(self, url):
 871         list_id = self._match_id(url)
 872         return self.playlist_result(self._entries(list_id, list_id), list_id)
 873
 874
 875 class TikTokSoundIE(TikTokBaseListIE):
 876     IE_NAME = 'tiktok:sound'
 877     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/music/[\w\.-]+-(?P<id>[\d]+)[/?#&]?'
 878     _WORKING = False
 879     _QUERY_NAME = 'music_id'
 880     _API_ENDPOINT = 'music/aweme'
 881     _TESTS = [{
 882         'url': 'https://www.tiktok.com/music/Build-a-Btch-6956990112127585029?lang=en',
 883         'playlist_mincount': 100,
 884         'info_dict': {
 885             'id': '6956990112127585029'
 886         },
 887         'expected_warnings': ['Retrying']
 888     }, {
 889         # Actual entries are less than listed video count
 890         'url': 'https://www.tiktok.com/music/jiefei-soap-remix-7036843036118469381',
 891         'playlist_mincount': 2182,
 892         'info_dict': {
 893             'id': '7036843036118469381'
 894         },
 895         'expected_warnings': ['Retrying']
 896     }]
 897
 898
 899 class TikTokEffectIE(TikTokBaseListIE):
 900     IE_NAME = 'tiktok:effect'
 901     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/sticker/[\w\.-]+-(?P<id>[\d]+)[/?#&]?'
 902     _WORKING = False
 903     _QUERY_NAME = 'sticker_id'
 904     _API_ENDPOINT = 'sticker/aweme'
 905     _TESTS = [{
 906         'url': 'https://www.tiktok.com/sticker/MATERIAL-GWOOORL-1258156',
 907         'playlist_mincount': 100,
 908         'info_dict': {
 909             'id': '1258156',
 910         },
 911         'expected_warnings': ['Retrying']
 912     }, {
 913         # Different entries between mobile and web, depending on region
 914         'url': 'https://www.tiktok.com/sticker/Elf-Friend-479565',
 915         'only_matching': True
 916     }]
 917
 918
 919 class TikTokTagIE(TikTokBaseListIE):
 920     IE_NAME = 'tiktok:tag'
 921     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/tag/(?P<id>[^/?#&]+)'
 922     _WORKING = False
 923     _QUERY_NAME = 'ch_id'
 924     _API_ENDPOINT = 'challenge/aweme'
 925     _TESTS = [{
 926         'url': 'https://tiktok.com/tag/hello2018',
 927         'playlist_mincount': 39,
 928         'info_dict': {
 929             'id': '46294678',
 930             'title': 'hello2018',
 931         },
 932         'expected_warnings': ['Retrying']
 933     }, {
 934         'url': 'https://tiktok.com/tag/fypシ?is_copy_url=0&is_from_webapp=v1',
 935         'only_matching': True
 936     }]
 937
 938     def _real_extract(self, url):
 939         display_id = self._match_id(url)
 940         webpage = self._download_webpage(url, display_id, headers={
 941             'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)'
 942         })
 943         tag_id = self._html_search_regex(r'snssdk\d*://challenge/detail/(\d+)', webpage, 'tag ID')
 944         return self.playlist_result(self._entries(tag_id, display_id), tag_id, display_id)
 945
 946
 947 class DouyinIE(TikTokBaseIE):
 948     _VALID_URL = r'https?://(?:www\.)?douyin\.com/video/(?P<id>[0-9]+)'
 949     _TESTS = [{
 950         'url': 'https://www.douyin.com/video/6961737553342991651',
 951         'md5': '9ecce7bc5b302601018ecb2871c63a75',
 952         'info_dict': {
 953             'id': '6961737553342991651',
 954             'ext': 'mp4',
 955             'title': '#杨超越  小小水手带你去远航❤️',
 956             'description': '#杨超越  小小水手带你去远航❤️',
 957             'uploader': '6897520xka',
 958             'uploader_id': '110403406559',
 959             'uploader_url': 'https://www.douyin.com/user/MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98',
 960             'channel_id': 'MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98',
 961             'channel': '杨超越',
 962             'creators': ['杨超越'],
 963             'duration': 19,
 964             'timestamp': 1620905839,
 965             'upload_date': '20210513',
 966             'track': '@杨超越创作的原声',
 967             'artists': ['杨超越'],
 968             'view_count': int,
 969             'like_count': int,
 970             'repost_count': int,
 971             'comment_count': int,
 972             'thumbnail': r're:https?://.+\.jpe?g',
 973         },
 974     }, {
 975         'url': 'https://www.douyin.com/video/6982497745948921092',
 976         'md5': '15c5e660b7048af3707304e3cc02bbb5',
 977         'info_dict': {
 978             'id': '6982497745948921092',
 979             'ext': 'mp4',
 980             'title': '这个夏日和小羊@杨超越 一起遇见白色幻想',
 981             'description': '这个夏日和小羊@杨超越 一起遇见白色幻想',
 982             'uploader': '0731chaoyue',
 983             'uploader_id': '408654318141572',
 984             'uploader_url': 'https://www.douyin.com/user/MS4wLjABAAAAZJpnglcjW2f_CMVcnqA_6oVBXKWMpH0F8LIHuUu8-lA',
 985             'channel_id': 'MS4wLjABAAAAZJpnglcjW2f_CMVcnqA_6oVBXKWMpH0F8LIHuUu8-lA',
 986             'channel': '杨超越工作室',
 987             'creators': ['杨超越工作室'],
 988             'duration': 42,
 989             'timestamp': 1625739481,
 990             'upload_date': '20210708',
 991             'track': '@杨超越工作室创作的原声',
 992             'artists': ['杨超越工作室'],
 993             'view_count': int,
 994             'like_count': int,
 995             'repost_count': int,
 996             'comment_count': int,
 997             'thumbnail': r're:https?://.+\.jpe?g',
 998         },
 999     }, {
1000         'url': 'https://www.douyin.com/video/6953975910773099811',
1001         'md5': '0e6443758b8355db9a3c34864a4276be',
1002         'info_dict': {
1003             'id': '6953975910773099811',
1004             'ext': 'mp4',
1005             'title': '#一起看海  出现在你的夏日里',
1006             'description': '#一起看海  出现在你的夏日里',
1007             'uploader': '6897520xka',
1008             'uploader_id': '110403406559',
1009             'uploader_url': 'https://www.douyin.com/user/MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98',
1010             'channel_id': 'MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98',
1011             'channel': '杨超越',
1012             'creators': ['杨超越'],
1013             'duration': 17,
1014             'timestamp': 1619098692,
1015             'upload_date': '20210422',
1016             'track': '@杨超越创作的原声',
1017             'artists': ['杨超越'],
1018             'view_count': int,
1019             'like_count': int,
1020             'repost_count': int,
1021             'comment_count': int,
1022             'thumbnail': r're:https?://.+\.jpe?g',
1023         },
1024     }, {
1025         'url': 'https://www.douyin.com/video/6950251282489675042',
1026         'md5': 'b4db86aec367ef810ddd38b1737d2fed',
1027         'info_dict': {
1028             'id': '6950251282489675042',
1029             'ext': 'mp4',
1030             'title': '哈哈哈，成功了哈哈哈哈哈哈',
1031             'uploader': '杨超越',
1032             'upload_date': '20210412',
1033             'timestamp': 1618231483,
1034             'uploader_id': '110403406559',
1035             'view_count': int,
1036             'like_count': int,
1037             'repost_count': int,
1038             'comment_count': int,
1039         },
1040         'skip': 'No longer available',
1041     }, {
1042         'url': 'https://www.douyin.com/video/6963263655114722595',
1043         'md5': '1440bcf59d8700f8e014da073a4dfea8',
1044         'info_dict': {
1045             'id': '6963263655114722595',
1046             'ext': 'mp4',
1047             'title': '#哪个爱豆的105度最甜 换个角度看看我哈哈',
1048             'description': '#哪个爱豆的105度最甜 换个角度看看我哈哈',
1049             'uploader': '6897520xka',
1050             'uploader_id': '110403406559',
1051             'uploader_url': 'https://www.douyin.com/user/MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98',
1052             'channel_id': 'MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98',
1053             'channel': '杨超越',
1054             'creators': ['杨超越'],
1055             'duration': 15,
1056             'timestamp': 1621261163,
1057             'upload_date': '20210517',
1058             'track': '@杨超越创作的原声',
1059             'artists': ['杨超越'],
1060             'view_count': int,
1061             'like_count': int,
1062             'repost_count': int,
1063             'comment_count': int,
1064             'thumbnail': r're:https?://.+\.jpe?g',
1065         },
1066     }]
1067     _UPLOADER_URL_FORMAT = 'https://www.douyin.com/user/%s'
1068     _WEBPAGE_HOST = 'https://www.douyin.com/'
1069
1070     def _real_extract(self, url):
1071         video_id = self._match_id(url)
1072
1073         detail = traverse_obj(self._download_json(
1074             'https://www.douyin.com/aweme/v1/web/aweme/detail/', video_id,
1075             'Downloading web detail JSON', 'Failed to download web detail JSON',
1076             query={'aweme_id': video_id}, fatal=False), ('aweme_detail', {dict}))
1077         if not detail:
1078             # TODO: Run verification challenge code to generate signature cookies
1079             raise ExtractorError(
1080                 'Fresh cookies (not necessarily logged in) are needed',
1081                 expected=not self._get_cookies(self._WEBPAGE_HOST).get('s_v_web_id'))
1082
1083         return self._parse_aweme_video_app(detail)
1084
1085
1086 class TikTokVMIE(InfoExtractor):
1087     _VALID_URL = r'https?://(?:(?:vm|vt)\.tiktok\.com|(?:www\.)tiktok\.com/t)/(?P<id>\w+)'
1088     IE_NAME = 'vm.tiktok'
1089
1090     _TESTS = [{
1091         'url': 'https://www.tiktok.com/t/ZTRC5xgJp',
1092         'info_dict': {
1093             'id': '7170520270497680683',
1094             'ext': 'mp4',
1095             'title': 'md5:c64f6152330c2efe98093ccc8597871c',
1096             'uploader_id': '6687535061741700102',
1097             'upload_date': '20221127',
1098             'view_count': int,
1099             'like_count': int,
1100             'comment_count': int,
1101             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAObqu3WCTXxmw2xwZ3iLEHnEecEIw7ks6rxWqOqOhaPja9BI7gqUQnjw8_5FSoDXX',
1102             'album': 'Wave of Mutilation: Best of Pixies',
1103             'thumbnail': r're:https://.+\.webp.*',
1104             'duration': 5,
1105             'timestamp': 1669516858,
1106             'repost_count': int,
1107             'artist': 'Pixies',
1108             'track': 'Where Is My Mind?',
1109             'description': 'md5:c64f6152330c2efe98093ccc8597871c',
1110             'uploader': 'sigmachaddeus',
1111             'creator': 'SigmaChad',
1112         },
1113     }, {
1114         'url': 'https://vm.tiktok.com/ZTR45GpSF/',
1115         'info_dict': {
1116             'id': '7106798200794926362',
1117             'ext': 'mp4',
1118             'title': 'md5:edc3e7ea587847f8537468f2fe51d074',
1119             'uploader_id': '6997695878846268418',
1120             'upload_date': '20220608',
1121             'view_count': int,
1122             'like_count': int,
1123             'comment_count': int,
1124             'thumbnail': r're:https://.+\.webp.*',
1125             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAdZ_NcPPgMneaGrW0hN8O_J_bwLshwNNERRF5DxOw2HKIzk0kdlLrR8RkVl1ksrMO',
1126             'duration': 29,
1127             'timestamp': 1654680400,
1128             'repost_count': int,
1129             'artist': 'Akihitoko',
1130             'track': 'original sound',
1131             'description': 'md5:edc3e7ea587847f8537468f2fe51d074',
1132             'uploader': 'akihitoko1',
1133             'creator': 'Akihitoko',
1134         },
1135     }, {
1136         'url': 'https://vt.tiktok.com/ZSe4FqkKd',
1137         'only_matching': True,
1138     }]
1139
1140     def _real_extract(self, url):
1141         new_url = self._request_webpage(
1142             HEADRequest(url), self._match_id(url), headers={'User-Agent': 'facebookexternalhit/1.1'}).url
1143         if self.suitable(new_url):  # Prevent infinite loop in case redirect fails
1144             raise UnsupportedError(new_url)
1145         return self.url_result(new_url)
1146
1147
1148 class TikTokLiveIE(TikTokBaseIE):
1149     _VALID_URL = r'''(?x)https?://(?:
1150         (?:www\.)?tiktok\.com/@(?P<uploader>[\w.-]+)/live|
1151         m\.tiktok\.com/share/live/(?P<id>\d+)
1152     )'''
1153     IE_NAME = 'tiktok:live'
1154
1155     _TESTS = [{
1156         'url': 'https://www.tiktok.com/@weathernewslive/live',
1157         'info_dict': {
1158             'id': '7210809319192726273',
1159             'ext': 'mp4',
1160             'title': r're:ウェザーニュースLiVE[\d\s:-]*',
1161             'creator': 'ウェザーニュースLiVE',
1162             'uploader': 'weathernewslive',
1163             'uploader_id': '6621496731283095554',
1164             'uploader_url': 'https://www.tiktok.com/@weathernewslive',
1165             'live_status': 'is_live',
1166             'concurrent_view_count': int,
1167         },
1168         'params': {'skip_download': 'm3u8'},
1169     }, {
1170         'url': 'https://www.tiktok.com/@pilarmagenta/live',
1171         'info_dict': {
1172             'id': '7209423610325322522',
1173             'ext': 'mp4',
1174             'title': str,
1175             'creator': 'Pilarmagenta',
1176             'uploader': 'pilarmagenta',
1177             'uploader_id': '6624846890674683909',
1178             'uploader_url': 'https://www.tiktok.com/@pilarmagenta',
1179             'live_status': 'is_live',
1180             'concurrent_view_count': int,
1181         },
1182         'skip': 'Livestream',
1183     }, {
1184         'url': 'https://m.tiktok.com/share/live/7209423610325322522/?language=en',
1185         'only_matching': True,
1186     }, {
1187         'url': 'https://www.tiktok.com/@iris04201/live',
1188         'only_matching': True,
1189     }]
1190
1191     def _call_api(self, url, param, room_id, uploader, key=None):
1192         response = traverse_obj(self._download_json(
1193             url, room_id, fatal=False, query={
1194                 'aid': '1988',
1195                 param: room_id,
1196             }), (key, {dict}), default={})
1197
1198         # status == 2 if live else 4
1199         if int_or_none(response.get('status')) == 2:
1200             return response
1201         # If room_id is obtained via mobile share URL and cannot be refreshed, do not wait for live
1202         elif not uploader:
1203             raise ExtractorError('This livestream has ended', expected=True)
1204         raise UserNotLive(video_id=uploader)
1205
1206     def _real_extract(self, url):
1207         uploader, room_id = self._match_valid_url(url).group('uploader', 'id')
1208         webpage = self._download_webpage(
1209             url, uploader or room_id, headers={'User-Agent': 'Mozilla/5.0'}, fatal=not room_id)
1210
1211         if webpage:
1212             data = self._get_sigi_state(webpage, uploader or room_id)
1213             room_id = (traverse_obj(data, ('UserModule', 'users', ..., 'roomId', {str_or_none}), get_all=False)
1214                        or self._search_regex(r'snssdk\d*://live\?room_id=(\d+)', webpage, 'room ID', default=None)
1215                        or room_id)
1216             uploader = uploader or traverse_obj(
1217                 data, ('LiveRoom', 'liveRoomUserInfo', 'user', 'uniqueId'),
1218                 ('UserModule', 'users', ..., 'uniqueId'), get_all=False, expected_type=str)
1219
1220         if not room_id:
1221             raise UserNotLive(video_id=uploader)
1222
1223         formats = []
1224         live_info = self._call_api(
1225             'https://webcast.tiktok.com/webcast/room/info', 'room_id', room_id, uploader, key='data')
1226
1227         get_quality = qualities(('SD1', 'ld', 'SD2', 'sd', 'HD1', 'hd', 'FULL_HD1', 'uhd', 'ORIGION', 'origin'))
1228         parse_inner = lambda x: self._parse_json(x, None)
1229
1230         for quality, stream in traverse_obj(live_info, (
1231                 'stream_url', 'live_core_sdk_data', 'pull_data', 'stream_data',
1232                 {parse_inner}, 'data', {dict}), default={}).items():
1233
1234             sdk_params = traverse_obj(stream, ('main', 'sdk_params', {parse_inner}, {
1235                 'vcodec': ('VCodec', {str}),
1236                 'tbr': ('vbitrate', {lambda x: int_or_none(x, 1000)}),
1237                 'resolution': ('resolution', {lambda x: re.match(r'(?i)\d+x\d+|\d+p', x).group().lower()}),
1238             }))
1239
1240             flv_url = traverse_obj(stream, ('main', 'flv', {url_or_none}))
1241             if flv_url:
1242                 formats.append({
1243                     'url': flv_url,
1244                     'ext': 'flv',
1245                     'format_id': f'flv-{quality}',
1246                     'quality': get_quality(quality),
1247                     **sdk_params,
1248                 })
1249
1250             hls_url = traverse_obj(stream, ('main', 'hls', {url_or_none}))
1251             if hls_url:
1252                 formats.append({
1253                     'url': hls_url,
1254                     'ext': 'mp4',
1255                     'protocol': 'm3u8_native',
1256                     'format_id': f'hls-{quality}',
1257                     'quality': get_quality(quality),
1258                     **sdk_params,
1259                 })
1260
1261         def get_vcodec(*keys):
1262             return traverse_obj(live_info, (
1263                 'stream_url', *keys, {parse_inner}, 'VCodec', {str}))
1264
1265         for stream in ('hls', 'rtmp'):
1266             stream_url = traverse_obj(live_info, ('stream_url', f'{stream}_pull_url', {url_or_none}))
1267             if stream_url:
1268                 formats.append({
1269                     'url': stream_url,
1270                     'ext': 'mp4' if stream == 'hls' else 'flv',
1271                     'protocol': 'm3u8_native' if stream == 'hls' else 'https',
1272                     'format_id': f'{stream}-pull',
1273                     'vcodec': get_vcodec(f'{stream}_pull_url_params'),
1274                     'quality': get_quality('ORIGION'),
1275                 })
1276
1277         for f_id, f_url in traverse_obj(live_info, ('stream_url', 'flv_pull_url', {dict}), default={}).items():
1278             if not url_or_none(f_url):
1279                 continue
1280             formats.append({
1281                 'url': f_url,
1282                 'ext': 'flv',
1283                 'format_id': f'flv-{f_id}'.lower(),
1284                 'vcodec': get_vcodec('flv_pull_url_params', f_id),
1285                 'quality': get_quality(f_id),
1286             })
1287
1288         # If uploader is a guest on another's livestream, primary endpoint will not have m3u8 URLs
1289         if not traverse_obj(formats, lambda _, v: v['ext'] == 'mp4'):
1290             live_info = merge_dicts(live_info, self._call_api(
1291                 'https://www.tiktok.com/api/live/detail/', 'roomID', room_id, uploader, key='LiveRoomInfo'))
1292             if url_or_none(live_info.get('liveUrl')):
1293                 formats.append({
1294                     'url': live_info['liveUrl'],
1295                     'ext': 'mp4',
1296                     'protocol': 'm3u8_native',
1297                     'format_id': 'hls-fallback',
1298                     'vcodec': 'h264',
1299                     'quality': get_quality('origin'),
1300                 })
1301
1302         uploader = uploader or traverse_obj(live_info, ('ownerInfo', 'uniqueId'), ('owner', 'display_id'))
1303
1304         return {
1305             'id': room_id,
1306             'uploader': uploader,
1307             'uploader_url': format_field(uploader, None, self._UPLOADER_URL_FORMAT) or None,
1308             'is_live': True,
1309             'formats': formats,
1310             '_format_sort_fields': ('quality', 'ext'),
1311             **traverse_obj(live_info, {
1312                 'title': 'title',
1313                 'uploader_id': (('ownerInfo', 'owner'), 'id', {str_or_none}),
1314                 'creator': (('ownerInfo', 'owner'), 'nickname'),
1315                 'concurrent_view_count': (('user_count', ('liveRoomStats', 'userCount')), {int_or_none}),
1316             }, get_all=False),
1317         }