yt_dlp/extractor/tiktok.py

   1 import itertools
   2 import json
   3 import random
   4 import string
   5 import time
   6
   7 from .common import InfoExtractor
   8 from ..compat import compat_urllib_parse_unquote, compat_urllib_parse_urlparse
   9 from ..utils import (
  10     ExtractorError,
  11     HEADRequest,
  12     LazyList,
  13     UnsupportedError,
  14     get_element_by_id,
  15     get_first,
  16     int_or_none,
  17     join_nonempty,
  18     qualities,
  19     remove_start,
  20     srt_subtitles_timecode,
  21     str_or_none,
  22     traverse_obj,
  23     try_get,
  24     url_or_none,
  25 )
  26
  27
  28 class TikTokBaseIE(InfoExtractor):
  29     _APP_VERSIONS = [('26.1.3', '260103'), ('26.1.2', '260102'), ('26.1.1', '260101'), ('25.6.2', '250602')]
  30     _WORKING_APP_VERSION = None
  31     _APP_NAME = 'trill'
  32     _AID = 1180
  33     _API_HOSTNAME = 'api16-normal-c-useast1a.tiktokv.com'
  34     _UPLOADER_URL_FORMAT = 'https://www.tiktok.com/@%s'
  35     _WEBPAGE_HOST = 'https://www.tiktok.com/'
  36     QUALITIES = ('360p', '540p', '720p', '1080p')
  37
  38     @staticmethod
  39     def _create_url(user_id, video_id):
  40         return f'https://www.tiktok.com/@{user_id or "_"}/video/{video_id}'
  41
  42     def _get_sigi_state(self, webpage, display_id):
  43         return self._parse_json(get_element_by_id(
  44             'SIGI_STATE|sigi-persisted-data', webpage, escape_value=False), display_id)
  45
  46     def _call_api_impl(self, ep, query, manifest_app_version, video_id, fatal=True,
  47                        note='Downloading API JSON', errnote='Unable to download API page'):
  48         self._set_cookie(self._API_HOSTNAME, 'odin_tt', ''.join(random.choice('0123456789abcdef') for _ in range(160)))
  49         webpage_cookies = self._get_cookies(self._WEBPAGE_HOST)
  50         if webpage_cookies.get('sid_tt'):
  51             self._set_cookie(self._API_HOSTNAME, 'sid_tt', webpage_cookies['sid_tt'].value)
  52         return self._download_json(
  53             'https://%s/aweme/v1/%s/' % (self._API_HOSTNAME, ep), video_id=video_id,
  54             fatal=fatal, note=note, errnote=errnote, headers={
  55                 'User-Agent': f'com.ss.android.ugc.{self._APP_NAME}/{manifest_app_version} (Linux; U; Android 10; en_US; Pixel 4; Build/QQ3A.200805.001; Cronet/58.0.2991.0)',
  56                 'Accept': 'application/json',
  57             }, query=query)
  58
  59     def _build_api_query(self, query, app_version, manifest_app_version):
  60         return {
  61             **query,
  62             'version_name': app_version,
  63             'version_code': manifest_app_version,
  64             'build_number': app_version,
  65             'manifest_version_code': manifest_app_version,
  66             'update_version_code': manifest_app_version,
  67             'openudid': ''.join(random.choice('0123456789abcdef') for _ in range(16)),
  68             'uuid': ''.join([random.choice(string.digits) for _ in range(16)]),
  69             '_rticket': int(time.time() * 1000),
  70             'ts': int(time.time()),
  71             'device_brand': 'Google',
  72             'device_type': 'Pixel 4',
  73             'device_platform': 'android',
  74             'resolution': '1080*1920',
  75             'dpi': 420,
  76             'os_version': '10',
  77             'os_api': '29',
  78             'carrier_region': 'US',
  79             'sys_region': 'US',
  80             'region': 'US',
  81             'app_name': self._APP_NAME,
  82             'app_language': 'en',
  83             'language': 'en',
  84             'timezone_name': 'America/New_York',
  85             'timezone_offset': '-14400',
  86             'channel': 'googleplay',
  87             'ac': 'wifi',
  88             'mcc_mnc': '310260',
  89             'is_my_cn': 0,
  90             'aid': self._AID,
  91             'ssmix': 'a',
  92             'as': 'a1qwert123',
  93             'cp': 'cbfhckdckkde1',
  94         }
  95
  96     def _call_api(self, ep, query, video_id, fatal=True,
  97                   note='Downloading API JSON', errnote='Unable to download API page'):
  98         if not self._WORKING_APP_VERSION:
  99             app_version = self._configuration_arg('app_version', [''], ie_key=TikTokIE.ie_key())[0]
 100             manifest_app_version = self._configuration_arg('manifest_app_version', [''], ie_key=TikTokIE.ie_key())[0]
 101             if app_version and manifest_app_version:
 102                 self._WORKING_APP_VERSION = (app_version, manifest_app_version)
 103                 self.write_debug('Imported app version combo from extractor arguments')
 104             elif app_version or manifest_app_version:
 105                 self.report_warning('Only one of the two required version params are passed as extractor arguments', only_once=True)
 106
 107         if self._WORKING_APP_VERSION:
 108             app_version, manifest_app_version = self._WORKING_APP_VERSION
 109             real_query = self._build_api_query(query, app_version, manifest_app_version)
 110             return self._call_api_impl(ep, real_query, manifest_app_version, video_id, fatal, note, errnote)
 111
 112         for count, (app_version, manifest_app_version) in enumerate(self._APP_VERSIONS, start=1):
 113             real_query = self._build_api_query(query, app_version, manifest_app_version)
 114             try:
 115                 res = self._call_api_impl(ep, real_query, manifest_app_version, video_id, fatal, note, errnote)
 116                 self._WORKING_APP_VERSION = (app_version, manifest_app_version)
 117                 return res
 118             except ExtractorError as e:
 119                 if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0:
 120                     if count == len(self._APP_VERSIONS):
 121                         if fatal:
 122                             raise e
 123                         else:
 124                             self.report_warning(str(e.cause or e.msg))
 125                             return
 126                     self.report_warning('%s. Retrying... (attempt %s of %s)' % (str(e.cause or e.msg), count, len(self._APP_VERSIONS)))
 127                     continue
 128                 raise e
 129
 130     def _extract_aweme_app(self, aweme_id):
 131         feed_list = self._call_api(
 132             'feed', {'aweme_id': aweme_id}, aweme_id, note='Downloading video feed',
 133             errnote='Unable to download video feed').get('aweme_list') or []
 134         aweme_detail = next((aweme for aweme in feed_list if str(aweme.get('aweme_id')) == aweme_id), None)
 135         if not aweme_detail:
 136             raise ExtractorError('Unable to find video in feed', video_id=aweme_id)
 137         return self._parse_aweme_video_app(aweme_detail)
 138
 139     def _get_subtitles(self, aweme_detail, aweme_id):
 140         # TODO: Extract text positioning info
 141         subtitles = {}
 142         # aweme/detail endpoint subs
 143         captions_info = traverse_obj(
 144             aweme_detail, ('interaction_stickers', ..., 'auto_video_caption_info', 'auto_captions', ...), expected_type=dict)
 145         for caption in captions_info:
 146             caption_url = traverse_obj(caption, ('url', 'url_list', ...), expected_type=url_or_none, get_all=False)
 147             if not caption_url:
 148                 continue
 149             caption_json = self._download_json(
 150                 caption_url, aweme_id, note='Downloading captions', errnote='Unable to download captions', fatal=False)
 151             if not caption_json:
 152                 continue
 153             subtitles.setdefault(caption.get('language', 'en'), []).append({
 154                 'ext': 'srt',
 155                 'data': '\n\n'.join(
 156                     f'{i + 1}\n{srt_subtitles_timecode(line["start_time"] / 1000)} --> {srt_subtitles_timecode(line["end_time"] / 1000)}\n{line["text"]}'
 157                     for i, line in enumerate(caption_json['utterances']) if line.get('text'))
 158             })
 159         # feed endpoint subs
 160         if not subtitles:
 161             for caption in traverse_obj(aweme_detail, ('video', 'cla_info', 'caption_infos', ...), expected_type=dict):
 162                 if not caption.get('url'):
 163                     continue
 164                 subtitles.setdefault(caption.get('lang') or 'en', []).append({
 165                     'ext': remove_start(caption.get('caption_format'), 'web'),
 166                     'url': caption['url'],
 167                 })
 168         # webpage subs
 169         if not subtitles:
 170             for caption in traverse_obj(aweme_detail, ('video', 'subtitleInfos', ...), expected_type=dict):
 171                 if not caption.get('Url'):
 172                     continue
 173                 subtitles.setdefault(caption.get('LanguageCodeName') or 'en', []).append({
 174                     'ext': remove_start(caption.get('Format'), 'web'),
 175                     'url': caption['Url'],
 176                 })
 177         return subtitles
 178
 179     def _parse_aweme_video_app(self, aweme_detail):
 180         aweme_id = aweme_detail['aweme_id']
 181         video_info = aweme_detail['video']
 182
 183         def parse_url_key(url_key):
 184             format_id, codec, res, bitrate = self._search_regex(
 185                 r'v[^_]+_(?P<id>(?P<codec>[^_]+)_(?P<res>\d+p)_(?P<bitrate>\d+))', url_key,
 186                 'url key', default=(None, None, None, None), group=('id', 'codec', 'res', 'bitrate'))
 187             if not format_id:
 188                 return {}, None
 189             return {
 190                 'format_id': format_id,
 191                 'vcodec': 'h265' if codec == 'bytevc1' else codec,
 192                 'tbr': int_or_none(bitrate, scale=1000) or None,
 193                 'quality': qualities(self.QUALITIES)(res),
 194             }, res
 195
 196         known_resolutions = {}
 197
 198         def extract_addr(addr, add_meta={}):
 199             parsed_meta, res = parse_url_key(addr.get('url_key', ''))
 200             if res:
 201                 known_resolutions.setdefault(res, {}).setdefault('height', add_meta.get('height'))
 202                 known_resolutions[res].setdefault('width', add_meta.get('width'))
 203                 parsed_meta.update(known_resolutions.get(res, {}))
 204                 add_meta.setdefault('height', int_or_none(res[:-1]))
 205             return [{
 206                 'url': url,
 207                 'filesize': int_or_none(addr.get('data_size')),
 208                 'ext': 'mp4',
 209                 'acodec': 'aac',
 210                 'source_preference': -2 if 'aweme/v1' in url else -1,  # Downloads from API might get blocked
 211                 **add_meta, **parsed_meta,
 212                 'format_note': join_nonempty(
 213                     add_meta.get('format_note'), '(API)' if 'aweme/v1' in url else None, delim=' ')
 214             } for url in addr.get('url_list') or []]
 215
 216         # Hack: Add direct video links first to prioritize them when removing duplicate formats
 217         formats = []
 218         if video_info.get('play_addr'):
 219             formats.extend(extract_addr(video_info['play_addr'], {
 220                 'format_id': 'play_addr',
 221                 'format_note': 'Direct video',
 222                 'vcodec': 'h265' if traverse_obj(
 223                     video_info, 'is_bytevc1', 'is_h265') else 'h264',  # TODO: Check for "direct iOS" videos, like https://www.tiktok.com/@cookierun_dev/video/7039716639834656002
 224                 'width': video_info.get('width'),
 225                 'height': video_info.get('height'),
 226             }))
 227         if video_info.get('download_addr'):
 228             formats.extend(extract_addr(video_info['download_addr'], {
 229                 'format_id': 'download_addr',
 230                 'format_note': 'Download video%s' % (', watermarked' if video_info.get('has_watermark') else ''),
 231                 'vcodec': 'h264',
 232                 'width': video_info.get('width'),
 233                 'height': video_info.get('height'),
 234                 'preference': -2 if video_info.get('has_watermark') else -1,
 235             }))
 236         if video_info.get('play_addr_h264'):
 237             formats.extend(extract_addr(video_info['play_addr_h264'], {
 238                 'format_id': 'play_addr_h264',
 239                 'format_note': 'Direct video',
 240                 'vcodec': 'h264',
 241             }))
 242         if video_info.get('play_addr_bytevc1'):
 243             formats.extend(extract_addr(video_info['play_addr_bytevc1'], {
 244                 'format_id': 'play_addr_bytevc1',
 245                 'format_note': 'Direct video',
 246                 'vcodec': 'h265',
 247             }))
 248
 249         for bitrate in video_info.get('bit_rate', []):
 250             if bitrate.get('play_addr'):
 251                 formats.extend(extract_addr(bitrate['play_addr'], {
 252                     'format_id': bitrate.get('gear_name'),
 253                     'format_note': 'Playback video',
 254                     'tbr': try_get(bitrate, lambda x: x['bit_rate'] / 1000),
 255                     'vcodec': 'h265' if traverse_obj(
 256                         bitrate, 'is_bytevc1', 'is_h265') else 'h264',
 257                     'fps': bitrate.get('FPS'),
 258                 }))
 259
 260         self._remove_duplicate_formats(formats)
 261         auth_cookie = self._get_cookies(self._WEBPAGE_HOST).get('sid_tt')
 262         if auth_cookie:
 263             for f in formats:
 264                 self._set_cookie(compat_urllib_parse_urlparse(f['url']).hostname, 'sid_tt', auth_cookie.value)
 265
 266         thumbnails = []
 267         for cover_id in ('cover', 'ai_dynamic_cover', 'animated_cover', 'ai_dynamic_cover_bak',
 268                          'origin_cover', 'dynamic_cover'):
 269             cover = video_info.get(cover_id)
 270             if cover:
 271                 for cover_url in cover['url_list']:
 272                     thumbnails.append({
 273                         'id': cover_id,
 274                         'url': cover_url,
 275                     })
 276
 277         stats_info = aweme_detail.get('statistics', {})
 278         author_info = aweme_detail.get('author', {})
 279         music_info = aweme_detail.get('music', {})
 280         user_url = self._UPLOADER_URL_FORMAT % (traverse_obj(author_info,
 281                                                              'sec_uid', 'id', 'uid', 'unique_id',
 282                                                              expected_type=str_or_none, get_all=False))
 283         labels = traverse_obj(aweme_detail, ('hybrid_label', ..., 'text'), expected_type=str, default=[])
 284
 285         contained_music_track = traverse_obj(
 286             music_info, ('matched_song', 'title'), ('matched_pgc_sound', 'title'), expected_type=str)
 287         contained_music_author = traverse_obj(
 288             music_info, ('matched_song', 'author'), ('matched_pgc_sound', 'author'), 'author', expected_type=str)
 289
 290         is_generic_og_trackname = music_info.get('is_original_sound') and music_info.get('title') == 'original sound - %s' % music_info.get('owner_handle')
 291         if is_generic_og_trackname:
 292             music_track, music_author = contained_music_track or 'original sound', contained_music_author
 293         else:
 294             music_track, music_author = music_info.get('title'), music_info.get('author')
 295
 296         return {
 297             'id': aweme_id,
 298             'extractor_key': TikTokIE.ie_key(),
 299             'extractor': TikTokIE.IE_NAME,
 300             'webpage_url': self._create_url(author_info.get('uid'), aweme_id),
 301             'title': aweme_detail.get('desc'),
 302             'description': aweme_detail.get('desc'),
 303             'view_count': int_or_none(stats_info.get('play_count')),
 304             'like_count': int_or_none(stats_info.get('digg_count')),
 305             'repost_count': int_or_none(stats_info.get('share_count')),
 306             'comment_count': int_or_none(stats_info.get('comment_count')),
 307             'uploader': str_or_none(author_info.get('unique_id')),
 308             'creator': str_or_none(author_info.get('nickname')),
 309             'uploader_id': str_or_none(author_info.get('uid')),
 310             'uploader_url': user_url,
 311             'track': music_track,
 312             'album': str_or_none(music_info.get('album')) or None,
 313             'artist': music_author or None,
 314             'timestamp': int_or_none(aweme_detail.get('create_time')),
 315             'formats': formats,
 316             'subtitles': self.extract_subtitles(aweme_detail, aweme_id),
 317             'thumbnails': thumbnails,
 318             'duration': int_or_none(traverse_obj(video_info, 'duration', ('download_addr', 'duration')), scale=1000),
 319             'availability': self._availability(
 320                 is_private='Private' in labels,
 321                 needs_subscription='Friends only' in labels,
 322                 is_unlisted='Followers only' in labels),
 323             '_format_sort_fields': ('quality', 'codec', 'size', 'br'),
 324         }
 325
 326     def _parse_aweme_video_web(self, aweme_detail, webpage_url):
 327         video_info = aweme_detail['video']
 328         author_info = traverse_obj(aweme_detail, 'authorInfo', 'author', expected_type=dict, default={})
 329         music_info = aweme_detail.get('music') or {}
 330         stats_info = aweme_detail.get('stats') or {}
 331         user_url = self._UPLOADER_URL_FORMAT % (traverse_obj(author_info,
 332                                                              'secUid', 'id', 'uid', 'uniqueId',
 333                                                              expected_type=str_or_none, get_all=False)
 334                                                 or aweme_detail.get('authorSecId'))
 335
 336         formats = []
 337         play_url = video_info.get('playAddr')
 338         width = video_info.get('width')
 339         height = video_info.get('height')
 340         if isinstance(play_url, str):
 341             formats = [{
 342                 'url': self._proto_relative_url(play_url),
 343                 'ext': 'mp4',
 344                 'width': width,
 345                 'height': height,
 346             }]
 347         elif isinstance(play_url, list):
 348             formats = [{
 349                 'url': self._proto_relative_url(url),
 350                 'ext': 'mp4',
 351                 'width': width,
 352                 'height': height,
 353             } for url in traverse_obj(play_url, (..., 'src'), expected_type=url_or_none, default=[]) if url]
 354
 355         download_url = url_or_none(video_info.get('downloadAddr')) or traverse_obj(video_info, ('download', 'url'), expected_type=url_or_none)
 356         if download_url:
 357             formats.append({
 358                 'format_id': 'download',
 359                 'url': self._proto_relative_url(download_url),
 360                 'ext': 'mp4',
 361                 'width': width,
 362                 'height': height,
 363             })
 364         self._remove_duplicate_formats(formats)
 365
 366         thumbnails = []
 367         for thumbnail_name in ('thumbnail', 'cover', 'dynamicCover', 'originCover'):
 368             if aweme_detail.get(thumbnail_name):
 369                 thumbnails = [{
 370                     'url': self._proto_relative_url(aweme_detail[thumbnail_name]),
 371                     'width': width,
 372                     'height': height
 373                 }]
 374
 375         return {
 376             'id': traverse_obj(aweme_detail, 'id', 'awemeId', expected_type=str_or_none),
 377             'title': aweme_detail.get('desc'),
 378             'duration': try_get(aweme_detail, lambda x: x['video']['duration'], int),
 379             'view_count': int_or_none(stats_info.get('playCount')),
 380             'like_count': int_or_none(stats_info.get('diggCount')),
 381             'repost_count': int_or_none(stats_info.get('shareCount')),
 382             'comment_count': int_or_none(stats_info.get('commentCount')),
 383             'timestamp': int_or_none(aweme_detail.get('createTime')),
 384             'creator': str_or_none(author_info.get('nickname')),
 385             'uploader': str_or_none(author_info.get('uniqueId') or aweme_detail.get('author')),
 386             'uploader_id': str_or_none(traverse_obj(author_info, 'id', 'uid', 'authorId')),
 387             'uploader_url': user_url,
 388             'track': str_or_none(music_info.get('title')),
 389             'album': str_or_none(music_info.get('album')) or None,
 390             'artist': str_or_none(music_info.get('authorName')),
 391             'formats': formats,
 392             'thumbnails': thumbnails,
 393             'description': str_or_none(aweme_detail.get('desc')),
 394             'http_headers': {
 395                 'Referer': webpage_url
 396             }
 397         }
 398
 399
 400 class TikTokIE(TikTokBaseIE):
 401     _VALID_URL = r'https?://www\.tiktok\.com/(?:embed|@(?P<user_id>[\w\.-]+)/video)/(?P<id>\d+)'
 402     _EMBED_REGEX = [rf'<(?:script|iframe)[^>]+\bsrc=(["\'])(?P<url>{_VALID_URL})']
 403
 404     _TESTS = [{
 405         'url': 'https://www.tiktok.com/@leenabhushan/video/6748451240264420610',
 406         'md5': '736bb7a466c6f0a6afeb597da1e6f5b7',
 407         'info_dict': {
 408             'id': '6748451240264420610',
 409             'ext': 'mp4',
 410             'title': '#jassmanak #lehanga #leenabhushan',
 411             'description': '#jassmanak #lehanga #leenabhushan',
 412             'duration': 13,
 413             'height': 1024,
 414             'width': 576,
 415             'uploader': 'leenabhushan',
 416             'uploader_id': '6691488002098119685',
 417             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA_Eb4t1vodM1IuTy_cvp9CY22RAb59xqrO0Xtz9CYQJvgXaDvZxYnZYRzDWhhgJmy',
 418             'creator': 'facestoriesbyleenabh',
 419             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 420             'upload_date': '20191016',
 421             'timestamp': 1571246252,
 422             'view_count': int,
 423             'like_count': int,
 424             'repost_count': int,
 425             'comment_count': int,
 426             'artist': 'Ysrbeats',
 427             'album': 'Lehanga',
 428             'track': 'Lehanga',
 429         }
 430     }, {
 431         'url': 'https://www.tiktok.com/@patroxofficial/video/6742501081818877190?langCountry=en',
 432         'md5': '6f3cf8cdd9b28cb8363fe0a9a160695b',
 433         'info_dict': {
 434             'id': '6742501081818877190',
 435             'ext': 'mp4',
 436             'title': 'md5:5e2a23877420bb85ce6521dbee39ba94',
 437             'description': 'md5:5e2a23877420bb85ce6521dbee39ba94',
 438             'duration': 27,
 439             'height': 960,
 440             'width': 540,
 441             'uploader': 'patrox',
 442             'uploader_id': '18702747',
 443             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAiFnldaILebi5heDoVU6bn4jBWWycX6-9U3xuNPqZ8Ws',
 444             'creator': 'patroX',
 445             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 446             'upload_date': '20190930',
 447             'timestamp': 1569860870,
 448             'view_count': int,
 449             'like_count': int,
 450             'repost_count': int,
 451             'comment_count': int,
 452             'artist': 'Evan Todd, Jessica Keenan Wynn, Alice Lee, Barrett Wilbert Weed & Jon Eidson',
 453             'track': 'Big Fun',
 454         }
 455     }, {
 456         # Banned audio, only available on the app
 457         'url': 'https://www.tiktok.com/@barudakhb_/video/6984138651336838402',
 458         'info_dict': {
 459             'id': '6984138651336838402',
 460             'ext': 'mp4',
 461             'title': 'Balas @yolaaftwsr hayu yu ? #SquadRandom_ 🔥',
 462             'description': 'Balas @yolaaftwsr hayu yu ? #SquadRandom_ 🔥',
 463             'uploader': 'barudakhb_',
 464             'creator': 'md5:29f238c49bc0c176cb3cef1a9cea9fa6',
 465             'uploader_id': '6974687867511718913',
 466             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAbhBwQC-R1iKoix6jDFsF-vBdfx2ABoDjaZrM9fX6arU3w71q3cOWgWuTXn1soZ7d',
 467             'track': 'Boka Dance',
 468             'artist': 'md5:29f238c49bc0c176cb3cef1a9cea9fa6',
 469             'timestamp': 1626121503,
 470             'duration': 18,
 471             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 472             'upload_date': '20210712',
 473             'view_count': int,
 474             'like_count': int,
 475             'repost_count': int,
 476             'comment_count': int,
 477         }
 478     }, {
 479         # Sponsored video, only available with feed workaround
 480         'url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_/video/7042692929109986561',
 481         'info_dict': {
 482             'id': '7042692929109986561',
 483             'ext': 'mp4',
 484             'title': 'Slap and Run!',
 485             'description': 'Slap and Run!',
 486             'uploader': 'user440922249',
 487             'creator': 'Slap And Run',
 488             'uploader_id': '7036055384943690754',
 489             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_',
 490             'track': 'Promoted Music',
 491             'timestamp': 1639754738,
 492             'duration': 30,
 493             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 494             'upload_date': '20211217',
 495             'view_count': int,
 496             'like_count': int,
 497             'repost_count': int,
 498             'comment_count': int,
 499         },
 500         'expected_warnings': ['trying with webpage', 'Unable to find video in feed']
 501     }, {
 502         # Video without title and description
 503         'url': 'https://www.tiktok.com/@pokemonlife22/video/7059698374567611694',
 504         'info_dict': {
 505             'id': '7059698374567611694',
 506             'ext': 'mp4',
 507             'title': 'TikTok video #7059698374567611694',
 508             'description': '',
 509             'uploader': 'pokemonlife22',
 510             'creator': 'Pokemon',
 511             'uploader_id': '6820838815978423302',
 512             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA0tF1nBwQVVMyrGu3CqttkNgM68Do1OXUFuCY0CRQk8fEtSVDj89HqoqvbSTmUP2W',
 513             'track': 'original sound',
 514             'timestamp': 1643714123,
 515             'duration': 6,
 516             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 517             'upload_date': '20220201',
 518             'artist': 'Pokemon',
 519             'view_count': int,
 520             'like_count': int,
 521             'repost_count': int,
 522             'comment_count': int,
 523         },
 524     }, {
 525         # hydration JSON is sent in a <script> element
 526         'url': 'https://www.tiktok.com/@denidil6/video/7065799023130643713',
 527         'info_dict': {
 528             'id': '7065799023130643713',
 529             'ext': 'mp4',
 530             'title': '#denidil#денидил',
 531             'description': '#denidil#денидил',
 532             'uploader': 'denidil6',
 533             'uploader_id': '7046664115636405250',
 534             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAsvMSzFdQ4ikl3uR2TEJwMBbB2yZh2Zxwhx-WCo3rbDpAharE3GQCrFuJArI3C8QJ',
 535             'artist': 'Holocron Music',
 536             'album': 'Wolf Sounds (1 Hour) Enjoy the Company of the Animal That Is the Majestic King of the Night',
 537             'track': 'Wolf Sounds (1 Hour) Enjoy the Company of the Animal That Is the Majestic King of the Night',
 538             'timestamp': 1645134536,
 539             'duration': 26,
 540             'upload_date': '20220217',
 541             'view_count': int,
 542             'like_count': int,
 543             'repost_count': int,
 544             'comment_count': int,
 545         },
 546         'skip': 'This video is unavailable',
 547     }, {
 548         # Auto-captions available
 549         'url': 'https://www.tiktok.com/@hankgreen1/video/7047596209028074758',
 550         'only_matching': True
 551     }]
 552
 553     def _real_extract(self, url):
 554         video_id, user_id = self._match_valid_url(url).group('id', 'user_id')
 555         try:
 556             return self._extract_aweme_app(video_id)
 557         except ExtractorError as e:
 558             self.report_warning(f'{e}; trying with webpage')
 559
 560         url = self._create_url(user_id, video_id)
 561         webpage = self._download_webpage(url, video_id, headers={'User-Agent': 'User-Agent:Mozilla/5.0'})
 562         next_data = self._search_nextjs_data(webpage, video_id, default='{}')
 563         if next_data:
 564             status = traverse_obj(next_data, ('props', 'pageProps', 'statusCode'), expected_type=int) or 0
 565             video_data = traverse_obj(next_data, ('props', 'pageProps', 'itemInfo', 'itemStruct'), expected_type=dict)
 566         else:
 567             sigi_data = self._get_sigi_state(webpage, video_id)
 568             status = traverse_obj(sigi_data, ('VideoPage', 'statusCode'), expected_type=int) or 0
 569             video_data = traverse_obj(sigi_data, ('ItemModule', video_id), expected_type=dict)
 570
 571         if status == 0:
 572             return self._parse_aweme_video_web(video_data, url)
 573         elif status == 10216:
 574             raise ExtractorError('This video is private', expected=True)
 575         raise ExtractorError('Video not available', video_id=video_id)
 576
 577
 578 class TikTokUserIE(TikTokBaseIE):
 579     IE_NAME = 'tiktok:user'
 580     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/@(?P<id>[\w\.-]+)/?(?:$|[#?])'
 581     _WORKING = False
 582     _TESTS = [{
 583         'url': 'https://tiktok.com/@corgibobaa?lang=en',
 584         'playlist_mincount': 45,
 585         'info_dict': {
 586             'id': '6935371178089399301',
 587             'title': 'corgibobaa',
 588             'thumbnail': r're:https://.+_1080x1080\.webp'
 589         },
 590         'expected_warnings': ['Retrying']
 591     }, {
 592         'url': 'https://www.tiktok.com/@6820838815978423302',
 593         'playlist_mincount': 5,
 594         'info_dict': {
 595             'id': '6820838815978423302',
 596             'title': '6820838815978423302',
 597             'thumbnail': r're:https://.+_1080x1080\.webp'
 598         },
 599         'expected_warnings': ['Retrying']
 600     }, {
 601         'url': 'https://www.tiktok.com/@meme',
 602         'playlist_mincount': 593,
 603         'info_dict': {
 604             'id': '79005827461758976',
 605             'title': 'meme',
 606             'thumbnail': r're:https://.+_1080x1080\.webp'
 607         },
 608         'expected_warnings': ['Retrying']
 609     }]
 610
 611     r'''  # TODO: Fix by adding _signature to api_url
 612     def _entries(self, webpage, user_id, username):
 613         secuid = self._search_regex(r'\"secUid\":\"(?P<secUid>[^\"]+)', webpage, username)
 614         verifyfp_cookie = self._get_cookies('https://www.tiktok.com').get('s_v_web_id')
 615         if not verifyfp_cookie:
 616             raise ExtractorError('Improper cookies (missing s_v_web_id).', expected=True)
 617         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='
 618         cursor = '0'
 619         for page in itertools.count():
 620             data_json = self._download_json(api_url + cursor, username, note='Downloading Page %d' % page)
 621             for video in data_json.get('itemList', []):
 622                 video_id = video['id']
 623                 video_url = f'https://www.tiktok.com/@{user_id}/video/{video_id}'
 624                 yield self._url_result(video_url, 'TikTok', video_id, str_or_none(video.get('desc')))
 625             if not data_json.get('hasMore'):
 626                 break
 627             cursor = data_json['cursor']
 628     '''
 629
 630     def _video_entries_api(self, webpage, user_id, username):
 631         query = {
 632             'user_id': user_id,
 633             'count': 21,
 634             'max_cursor': 0,
 635             'min_cursor': 0,
 636             'retry_type': 'no_retry',
 637             'device_id': ''.join(random.choice(string.digits) for _ in range(19)),  # Some endpoints don't like randomized device_id, so it isn't directly set in _call_api.
 638         }
 639
 640         for page in itertools.count(1):
 641             for retry in self.RetryManager():
 642                 try:
 643                     post_list = self._call_api(
 644                         'aweme/post', query, username, note=f'Downloading user video list page {page}',
 645                         errnote='Unable to download user video list')
 646                 except ExtractorError as e:
 647                     if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0:
 648                         retry.error = e
 649                         continue
 650                     raise
 651             yield from post_list.get('aweme_list', [])
 652             if not post_list.get('has_more'):
 653                 break
 654             query['max_cursor'] = post_list['max_cursor']
 655
 656     def _entries_api(self, user_id, videos):
 657         for video in videos:
 658             yield {
 659                 **self._parse_aweme_video_app(video),
 660                 'extractor_key': TikTokIE.ie_key(),
 661                 'extractor': 'TikTok',
 662                 'webpage_url': f'https://tiktok.com/@{user_id}/video/{video["aweme_id"]}',
 663             }
 664
 665     def _real_extract(self, url):
 666         user_name = self._match_id(url)
 667         webpage = self._download_webpage(url, user_name, headers={
 668             'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)'
 669         })
 670         user_id = self._html_search_regex(r'snssdk\d*://user/profile/(\d+)', webpage, 'user ID', default=None) or user_name
 671
 672         videos = LazyList(self._video_entries_api(webpage, user_id, user_name))
 673         thumbnail = traverse_obj(videos, (0, 'author', 'avatar_larger', 'url_list', 0))
 674
 675         return self.playlist_result(self._entries_api(user_id, videos), user_id, user_name, thumbnail=thumbnail)
 676
 677
 678 class TikTokBaseListIE(TikTokBaseIE):  # XXX: Conventionally, base classes should end with BaseIE/InfoExtractor
 679     def _entries(self, list_id, display_id):
 680         query = {
 681             self._QUERY_NAME: list_id,
 682             'cursor': 0,
 683             'count': 20,
 684             'type': 5,
 685             'device_id': ''.join(random.choice(string.digits) for i in range(19))
 686         }
 687
 688         for page in itertools.count(1):
 689             for retry in self.RetryManager():
 690                 try:
 691                     post_list = self._call_api(
 692                         self._API_ENDPOINT, query, display_id, note=f'Downloading video list page {page}',
 693                         errnote='Unable to download video list')
 694                 except ExtractorError as e:
 695                     if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0:
 696                         retry.error = e
 697                         continue
 698                     raise
 699             for video in post_list.get('aweme_list', []):
 700                 yield {
 701                     **self._parse_aweme_video_app(video),
 702                     'extractor_key': TikTokIE.ie_key(),
 703                     'extractor': 'TikTok',
 704                     'webpage_url': f'https://tiktok.com/@_/video/{video["aweme_id"]}',
 705                 }
 706             if not post_list.get('has_more'):
 707                 break
 708             query['cursor'] = post_list['cursor']
 709
 710     def _real_extract(self, url):
 711         list_id = self._match_id(url)
 712         return self.playlist_result(self._entries(list_id, list_id), list_id)
 713
 714
 715 class TikTokSoundIE(TikTokBaseListIE):
 716     IE_NAME = 'tiktok:sound'
 717     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/music/[\w\.-]+-(?P<id>[\d]+)[/?#&]?'
 718     _WORKING = False
 719     _QUERY_NAME = 'music_id'
 720     _API_ENDPOINT = 'music/aweme'
 721     _TESTS = [{
 722         'url': 'https://www.tiktok.com/music/Build-a-Btch-6956990112127585029?lang=en',
 723         'playlist_mincount': 100,
 724         'info_dict': {
 725             'id': '6956990112127585029'
 726         },
 727         'expected_warnings': ['Retrying']
 728     }, {
 729         # Actual entries are less than listed video count
 730         'url': 'https://www.tiktok.com/music/jiefei-soap-remix-7036843036118469381',
 731         'playlist_mincount': 2182,
 732         'info_dict': {
 733             'id': '7036843036118469381'
 734         },
 735         'expected_warnings': ['Retrying']
 736     }]
 737
 738
 739 class TikTokEffectIE(TikTokBaseListIE):
 740     IE_NAME = 'tiktok:effect'
 741     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/sticker/[\w\.-]+-(?P<id>[\d]+)[/?#&]?'
 742     _WORKING = False
 743     _QUERY_NAME = 'sticker_id'
 744     _API_ENDPOINT = 'sticker/aweme'
 745     _TESTS = [{
 746         'url': 'https://www.tiktok.com/sticker/MATERIAL-GWOOORL-1258156',
 747         'playlist_mincount': 100,
 748         'info_dict': {
 749             'id': '1258156',
 750         },
 751         'expected_warnings': ['Retrying']
 752     }, {
 753         # Different entries between mobile and web, depending on region
 754         'url': 'https://www.tiktok.com/sticker/Elf-Friend-479565',
 755         'only_matching': True
 756     }]
 757
 758
 759 class TikTokTagIE(TikTokBaseListIE):
 760     IE_NAME = 'tiktok:tag'
 761     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/tag/(?P<id>[^/?#&]+)'
 762     _WORKING = False
 763     _QUERY_NAME = 'ch_id'
 764     _API_ENDPOINT = 'challenge/aweme'
 765     _TESTS = [{
 766         'url': 'https://tiktok.com/tag/hello2018',
 767         'playlist_mincount': 39,
 768         'info_dict': {
 769             'id': '46294678',
 770             'title': 'hello2018',
 771         },
 772         'expected_warnings': ['Retrying']
 773     }, {
 774         'url': 'https://tiktok.com/tag/fypシ?is_copy_url=0&is_from_webapp=v1',
 775         'only_matching': True
 776     }]
 777
 778     def _real_extract(self, url):
 779         display_id = self._match_id(url)
 780         webpage = self._download_webpage(url, display_id, headers={
 781             'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)'
 782         })
 783         tag_id = self._html_search_regex(r'snssdk\d*://challenge/detail/(\d+)', webpage, 'tag ID')
 784         return self.playlist_result(self._entries(tag_id, display_id), tag_id, display_id)
 785
 786
 787 class DouyinIE(TikTokBaseIE):
 788     _VALID_URL = r'https?://(?:www\.)?douyin\.com/video/(?P<id>[0-9]+)'
 789     _TESTS = [{
 790         'url': 'https://www.douyin.com/video/6961737553342991651',
 791         'md5': 'a97db7e3e67eb57bf40735c022ffa228',
 792         'info_dict': {
 793             'id': '6961737553342991651',
 794             'ext': 'mp4',
 795             'title': '#杨超越  小小水手带你去远航❤️',
 796             'description': '#杨超越  小小水手带你去远航❤️',
 797             'uploader_id': '110403406559',
 798             'uploader_url': 'https://www.douyin.com/user/MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98',
 799             'creator': '杨超越',
 800             'duration': 19782,
 801             'timestamp': 1620905839,
 802             'upload_date': '20210513',
 803             'track': '@杨超越创作的原声',
 804             'view_count': int,
 805             'like_count': int,
 806             'repost_count': int,
 807             'comment_count': int,
 808         },
 809     }, {
 810         'url': 'https://www.douyin.com/video/6982497745948921092',
 811         'md5': '34a87ebff3833357733da3fe17e37c0e',
 812         'info_dict': {
 813             'id': '6982497745948921092',
 814             'ext': 'mp4',
 815             'title': '这个夏日和小羊@杨超越 一起遇见白色幻想',
 816             'description': '这个夏日和小羊@杨超越 一起遇见白色幻想',
 817             'uploader_id': '408654318141572',
 818             'uploader_url': 'https://www.douyin.com/user/MS4wLjABAAAAZJpnglcjW2f_CMVcnqA_6oVBXKWMpH0F8LIHuUu8-lA',
 819             'creator': '杨超越工作室',
 820             'duration': 42608,
 821             'timestamp': 1625739481,
 822             'upload_date': '20210708',
 823             'track': '@杨超越工作室创作的原声',
 824             'view_count': int,
 825             'like_count': int,
 826             'repost_count': int,
 827             'comment_count': int,
 828         },
 829     }, {
 830         'url': 'https://www.douyin.com/video/6953975910773099811',
 831         'md5': 'dde3302460f19db59c47060ff013b902',
 832         'info_dict': {
 833             'id': '6953975910773099811',
 834             'ext': 'mp4',
 835             'title': '#一起看海  出现在你的夏日里',
 836             'description': '#一起看海  出现在你的夏日里',
 837             'uploader_id': '110403406559',
 838             'uploader_url': 'https://www.douyin.com/user/MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98',
 839             'creator': '杨超越',
 840             'duration': 17228,
 841             'timestamp': 1619098692,
 842             'upload_date': '20210422',
 843             'track': '@杨超越创作的原声',
 844             'view_count': int,
 845             'like_count': int,
 846             'repost_count': int,
 847             'comment_count': int,
 848         },
 849     }, {
 850         'url': 'https://www.douyin.com/video/6950251282489675042',
 851         'md5': 'b4db86aec367ef810ddd38b1737d2fed',
 852         'info_dict': {
 853             'id': '6950251282489675042',
 854             'ext': 'mp4',
 855             'title': '哈哈哈，成功了哈哈哈哈哈哈',
 856             'uploader': '杨超越',
 857             'upload_date': '20210412',
 858             'timestamp': 1618231483,
 859             'uploader_id': '110403406559',
 860             'view_count': int,
 861             'like_count': int,
 862             'repost_count': int,
 863             'comment_count': int,
 864         },
 865         'skip': 'No longer available',
 866     }, {
 867         'url': 'https://www.douyin.com/video/6963263655114722595',
 868         'md5': 'cf9f11f0ec45d131445ec2f06766e122',
 869         'info_dict': {
 870             'id': '6963263655114722595',
 871             'ext': 'mp4',
 872             'title': '#哪个爱豆的105度最甜 换个角度看看我哈哈',
 873             'description': '#哪个爱豆的105度最甜 换个角度看看我哈哈',
 874             'uploader_id': '110403406559',
 875             'uploader_url': 'https://www.douyin.com/user/MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98',
 876             'creator': '杨超越',
 877             'duration': 15115,
 878             'timestamp': 1621261163,
 879             'upload_date': '20210517',
 880             'track': '@杨超越创作的原声',
 881             'view_count': int,
 882             'like_count': int,
 883             'repost_count': int,
 884             'comment_count': int,
 885         },
 886     }]
 887     _APP_VERSIONS = [('23.3.0', '230300')]
 888     _APP_NAME = 'aweme'
 889     _AID = 1128
 890     _API_HOSTNAME = 'aweme.snssdk.com'
 891     _UPLOADER_URL_FORMAT = 'https://www.douyin.com/user/%s'
 892     _WEBPAGE_HOST = 'https://www.douyin.com/'
 893
 894     def _real_extract(self, url):
 895         video_id = self._match_id(url)
 896
 897         try:
 898             return self._extract_aweme_app(video_id)
 899         except ExtractorError as e:
 900             e.expected = True
 901             self.to_screen(f'{e}; trying with webpage')
 902
 903         webpage = self._download_webpage(url, video_id)
 904         render_data_json = self._search_regex(
 905             r'<script [^>]*\bid=[\'"]RENDER_DATA[\'"][^>]*>(%7B.+%7D)</script>',
 906             webpage, 'render data', default=None)
 907         if not render_data_json:
 908             # TODO: Run verification challenge code to generate signature cookies
 909             cookies = self._get_cookies(self._WEBPAGE_HOST)
 910             expected = not cookies.get('s_v_web_id') or not cookies.get('ttwid')
 911             raise ExtractorError(
 912                 'Fresh cookies (not necessarily logged in) are needed', expected=expected)
 913
 914         render_data = self._parse_json(
 915             render_data_json, video_id, transform_source=compat_urllib_parse_unquote)
 916         return self._parse_aweme_video_web(get_first(render_data, ('aweme', 'detail')), url)
 917
 918
 919 class TikTokVMIE(InfoExtractor):
 920     _VALID_URL = r'https?://(?:(?:vm|vt)\.tiktok\.com|(?:www\.)tiktok\.com/t)/(?P<id>\w+)'
 921     IE_NAME = 'vm.tiktok'
 922
 923     _TESTS = [{
 924         'url': 'https://www.tiktok.com/t/ZTRC5xgJp',
 925         'info_dict': {
 926             'id': '7170520270497680683',
 927             'ext': 'mp4',
 928             'title': 'md5:c64f6152330c2efe98093ccc8597871c',
 929             'uploader_id': '6687535061741700102',
 930             'upload_date': '20221127',
 931             'view_count': int,
 932             'like_count': int,
 933             'comment_count': int,
 934             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAObqu3WCTXxmw2xwZ3iLEHnEecEIw7ks6rxWqOqOhaPja9BI7gqUQnjw8_5FSoDXX',
 935             'album': 'Wave of Mutilation: Best of Pixies',
 936             'thumbnail': r're:https://.+\.webp.*',
 937             'duration': 5,
 938             'timestamp': 1669516858,
 939             'repost_count': int,
 940             'artist': 'Pixies',
 941             'track': 'Where Is My Mind?',
 942             'description': 'md5:c64f6152330c2efe98093ccc8597871c',
 943             'uploader': 'sigmachaddeus',
 944             'creator': 'SigmaChad',
 945         },
 946     }, {
 947         'url': 'https://vm.tiktok.com/ZSe4FqkKd',
 948         'only_matching': True,
 949     }, {
 950         'url': 'https://vt.tiktok.com/ZSe4FqkKd',
 951         'only_matching': True,
 952     }]
 953
 954     def _real_extract(self, url):
 955         new_url = self._request_webpage(
 956             HEADRequest(url), self._match_id(url), headers={'User-Agent': 'facebookexternalhit/1.1'}).geturl()
 957         if self.suitable(new_url):  # Prevent infinite loop in case redirect fails
 958             raise UnsupportedError(new_url)
 959         return self.url_result(new_url)