yt_dlp/extractor/tiktok.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import itertools
   5 import random
   6 import string
   7 import time
   8 import json
   9
  10 from .common import InfoExtractor
  11 from ..compat import (
  12     compat_urllib_parse_unquote,
  13     compat_urllib_parse_urlparse
  14 )
  15 from ..utils import (
  16     ExtractorError,
  17     HEADRequest,
  18     get_first,
  19     int_or_none,
  20     join_nonempty,
  21     LazyList,
  22     srt_subtitles_timecode,
  23     str_or_none,
  24     traverse_obj,
  25     try_get,
  26     url_or_none,
  27     qualities,
  28 )
  29
  30
  31 class TikTokBaseIE(InfoExtractor):
  32     _APP_VERSIONS = [('20.9.3', '293'), ('20.4.3', '243'), ('20.2.1', '221'), ('20.1.2', '212'), ('20.0.4', '204')]
  33     _WORKING_APP_VERSION = None
  34     _APP_NAME = 'trill'
  35     _AID = 1180
  36     _API_HOSTNAME = 'api-h2.tiktokv.com'
  37     _UPLOADER_URL_FORMAT = 'https://www.tiktok.com/@%s'
  38     _WEBPAGE_HOST = 'https://www.tiktok.com/'
  39     QUALITIES = ('360p', '540p', '720p', '1080p')
  40
  41     def _call_api_impl(self, ep, query, manifest_app_version, video_id, fatal=True,
  42                        note='Downloading API JSON', errnote='Unable to download API page'):
  43         self._set_cookie(self._API_HOSTNAME, 'odin_tt', ''.join(random.choice('0123456789abcdef') for _ in range(160)))
  44         webpage_cookies = self._get_cookies(self._WEBPAGE_HOST)
  45         if webpage_cookies.get('sid_tt'):
  46             self._set_cookie(self._API_HOSTNAME, 'sid_tt', webpage_cookies['sid_tt'].value)
  47         return self._download_json(
  48             'https://%s/aweme/v1/%s/' % (self._API_HOSTNAME, ep), video_id=video_id,
  49             fatal=fatal, note=note, errnote=errnote, headers={
  50                 'User-Agent': f'com.ss.android.ugc.trill/{manifest_app_version} (Linux; U; Android 10; en_US; Pixel 4; Build/QQ3A.200805.001; Cronet/58.0.2991.0)',
  51                 'Accept': 'application/json',
  52             }, query=query)
  53
  54     def _build_api_query(self, query, app_version, manifest_app_version):
  55         return {
  56             **query,
  57             'version_name': app_version,
  58             'version_code': manifest_app_version,
  59             'build_number': app_version,
  60             'manifest_version_code': manifest_app_version,
  61             'update_version_code': manifest_app_version,
  62             'openudid': ''.join(random.choice('0123456789abcdef') for _ in range(16)),
  63             'uuid': ''.join([random.choice(string.digits) for _ in range(16)]),
  64             '_rticket': int(time.time() * 1000),
  65             'ts': int(time.time()),
  66             'device_brand': 'Google',
  67             'device_type': 'Pixel 4',
  68             'device_platform': 'android',
  69             'resolution': '1080*1920',
  70             'dpi': 420,
  71             'os_version': '10',
  72             'os_api': '29',
  73             'carrier_region': 'US',
  74             'sys_region': 'US',
  75             'region': 'US',
  76             'app_name': self._APP_NAME,
  77             'app_language': 'en',
  78             'language': 'en',
  79             'timezone_name': 'America/New_York',
  80             'timezone_offset': '-14400',
  81             'channel': 'googleplay',
  82             'ac': 'wifi',
  83             'mcc_mnc': '310260',
  84             'is_my_cn': 0,
  85             'aid': self._AID,
  86             'ssmix': 'a',
  87             'as': 'a1qwert123',
  88             'cp': 'cbfhckdckkde1',
  89         }
  90
  91     def _call_api(self, ep, query, video_id, fatal=True,
  92                   note='Downloading API JSON', errnote='Unable to download API page'):
  93         if not self._WORKING_APP_VERSION:
  94             app_version = self._configuration_arg('app_version', [''], ie_key=TikTokIE.ie_key())[0]
  95             manifest_app_version = self._configuration_arg('manifest_app_version', [''], ie_key=TikTokIE.ie_key())[0]
  96             if app_version and manifest_app_version:
  97                 self._WORKING_APP_VERSION = (app_version, manifest_app_version)
  98                 self.write_debug('Imported app version combo from extractor arguments')
  99             elif app_version or manifest_app_version:
 100                 self.report_warning('Only one of the two required version params are passed as extractor arguments', only_once=True)
 101
 102         if self._WORKING_APP_VERSION:
 103             app_version, manifest_app_version = self._WORKING_APP_VERSION
 104             real_query = self._build_api_query(query, app_version, manifest_app_version)
 105             return self._call_api_impl(ep, real_query, manifest_app_version, video_id, fatal, note, errnote)
 106
 107         for count, (app_version, manifest_app_version) in enumerate(self._APP_VERSIONS, start=1):
 108             real_query = self._build_api_query(query, app_version, manifest_app_version)
 109             try:
 110                 res = self._call_api_impl(ep, real_query, manifest_app_version, video_id, fatal, note, errnote)
 111                 self._WORKING_APP_VERSION = (app_version, manifest_app_version)
 112                 return res
 113             except ExtractorError as e:
 114                 if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0:
 115                     if count == len(self._APP_VERSIONS):
 116                         if fatal:
 117                             raise e
 118                         else:
 119                             self.report_warning(str(e.cause or e.msg))
 120                             return
 121                     self.report_warning('%s. Retrying... (attempt %s of %s)' % (str(e.cause or e.msg), count, len(self._APP_VERSIONS)))
 122                     continue
 123                 raise e
 124
 125     def _get_subtitles(self, aweme_detail, aweme_id):
 126         # TODO: Extract text positioning info
 127         subtitles = {}
 128         captions_info = traverse_obj(
 129             aweme_detail, ('interaction_stickers', ..., 'auto_video_caption_info', 'auto_captions', ...), expected_type=dict, default=[])
 130         for caption in captions_info:
 131             caption_url = traverse_obj(caption, ('url', 'url_list', ...), expected_type=url_or_none, get_all=False)
 132             if not caption_url:
 133                 continue
 134             caption_json = self._download_json(
 135                 caption_url, aweme_id, note='Downloading captions', errnote='Unable to download captions', fatal=False)
 136             if not caption_json:
 137                 continue
 138             subtitles.setdefault(caption.get('language', 'en'), []).append({
 139                 'ext': 'srt',
 140                 'data': '\n\n'.join(
 141                     f'{i + 1}\n{srt_subtitles_timecode(line["start_time"] / 1000)} --> {srt_subtitles_timecode(line["end_time"] / 1000)}\n{line["text"]}'
 142                     for i, line in enumerate(caption_json['utterances']) if line.get('text'))
 143             })
 144         return subtitles
 145
 146     def _parse_aweme_video_app(self, aweme_detail):
 147         aweme_id = aweme_detail['aweme_id']
 148         video_info = aweme_detail['video']
 149
 150         def parse_url_key(url_key):
 151             format_id, codec, res, bitrate = self._search_regex(
 152                 r'v[^_]+_(?P<id>(?P<codec>[^_]+)_(?P<res>\d+p)_(?P<bitrate>\d+))', url_key,
 153                 'url key', default=(None, None, None, None), group=('id', 'codec', 'res', 'bitrate'))
 154             if not format_id:
 155                 return {}, None
 156             return {
 157                 'format_id': format_id,
 158                 'vcodec': 'h265' if codec == 'bytevc1' else codec,
 159                 'tbr': int_or_none(bitrate, scale=1000) or None,
 160                 'quality': qualities(self.QUALITIES)(res),
 161             }, res
 162
 163         known_resolutions = {}
 164
 165         def extract_addr(addr, add_meta={}):
 166             parsed_meta, res = parse_url_key(addr.get('url_key', ''))
 167             if res:
 168                 known_resolutions.setdefault(res, {}).setdefault('height', add_meta.get('height'))
 169                 known_resolutions[res].setdefault('width', add_meta.get('width'))
 170                 parsed_meta.update(known_resolutions.get(res, {}))
 171                 add_meta.setdefault('height', int_or_none(res[:-1]))
 172             return [{
 173                 'url': url,
 174                 'filesize': int_or_none(addr.get('data_size')),
 175                 'ext': 'mp4',
 176                 'acodec': 'aac',
 177                 'source_preference': -2 if 'aweme/v1' in url else -1,  # Downloads from API might get blocked
 178                 **add_meta, **parsed_meta,
 179                 'format_note': join_nonempty(
 180                     add_meta.get('format_note'), '(API)' if 'aweme/v1' in url else None, delim=' ')
 181             } for url in addr.get('url_list') or []]
 182
 183         # Hack: Add direct video links first to prioritize them when removing duplicate formats
 184         formats = []
 185         if video_info.get('play_addr'):
 186             formats.extend(extract_addr(video_info['play_addr'], {
 187                 'format_id': 'play_addr',
 188                 'format_note': 'Direct video',
 189                 'vcodec': 'h265' if traverse_obj(
 190                     video_info, 'is_bytevc1', 'is_h265') else 'h264',  # TODO: Check for "direct iOS" videos, like https://www.tiktok.com/@cookierun_dev/video/7039716639834656002
 191                 'width': video_info.get('width'),
 192                 'height': video_info.get('height'),
 193             }))
 194         if video_info.get('download_addr'):
 195             formats.extend(extract_addr(video_info['download_addr'], {
 196                 'format_id': 'download_addr',
 197                 'format_note': 'Download video%s' % (', watermarked' if video_info.get('has_watermark') else ''),
 198                 'vcodec': 'h264',
 199                 'width': video_info.get('width'),
 200                 'height': video_info.get('height'),
 201                 'preference': -2 if video_info.get('has_watermark') else -1,
 202             }))
 203         if video_info.get('play_addr_h264'):
 204             formats.extend(extract_addr(video_info['play_addr_h264'], {
 205                 'format_id': 'play_addr_h264',
 206                 'format_note': 'Direct video',
 207                 'vcodec': 'h264',
 208             }))
 209         if video_info.get('play_addr_bytevc1'):
 210             formats.extend(extract_addr(video_info['play_addr_bytevc1'], {
 211                 'format_id': 'play_addr_bytevc1',
 212                 'format_note': 'Direct video',
 213                 'vcodec': 'h265',
 214             }))
 215
 216         for bitrate in video_info.get('bit_rate', []):
 217             if bitrate.get('play_addr'):
 218                 formats.extend(extract_addr(bitrate['play_addr'], {
 219                     'format_id': bitrate.get('gear_name'),
 220                     'format_note': 'Playback video',
 221                     'tbr': try_get(bitrate, lambda x: x['bit_rate'] / 1000),
 222                     'vcodec': 'h265' if traverse_obj(
 223                         bitrate, 'is_bytevc1', 'is_h265') else 'h264',
 224                     'fps': bitrate.get('FPS'),
 225                 }))
 226
 227         self._remove_duplicate_formats(formats)
 228         auth_cookie = self._get_cookies(self._WEBPAGE_HOST).get('sid_tt')
 229         if auth_cookie:
 230             for f in formats:
 231                 self._set_cookie(compat_urllib_parse_urlparse(f['url']).hostname, 'sid_tt', auth_cookie.value)
 232         self._sort_formats(formats, ('quality', 'codec', 'size', 'br'))
 233
 234         thumbnails = []
 235         for cover_id in ('cover', 'ai_dynamic_cover', 'animated_cover', 'ai_dynamic_cover_bak',
 236                          'origin_cover', 'dynamic_cover'):
 237             cover = video_info.get(cover_id)
 238             if cover:
 239                 for cover_url in cover['url_list']:
 240                     thumbnails.append({
 241                         'id': cover_id,
 242                         'url': cover_url,
 243                     })
 244
 245         stats_info = aweme_detail.get('statistics', {})
 246         author_info = aweme_detail.get('author', {})
 247         music_info = aweme_detail.get('music', {})
 248         user_url = self._UPLOADER_URL_FORMAT % (traverse_obj(author_info,
 249                                                              'sec_uid', 'id', 'uid', 'unique_id',
 250                                                              expected_type=str_or_none, get_all=False))
 251         labels = traverse_obj(aweme_detail, ('hybrid_label', ..., 'text'), expected_type=str, default=[])
 252
 253         contained_music_track = traverse_obj(
 254             music_info, ('matched_song', 'title'), ('matched_pgc_sound', 'title'), expected_type=str)
 255         contained_music_author = traverse_obj(
 256             music_info, ('matched_song', 'author'), ('matched_pgc_sound', 'author'), 'author', expected_type=str)
 257
 258         is_generic_og_trackname = music_info.get('is_original_sound') and music_info.get('title') == 'original sound - %s' % music_info.get('owner_handle')
 259         if is_generic_og_trackname:
 260             music_track, music_author = contained_music_track or 'original sound', contained_music_author
 261         else:
 262             music_track, music_author = music_info.get('title'), music_info.get('author')
 263
 264         return {
 265             'id': aweme_id,
 266             'title': aweme_detail['desc'],
 267             'description': aweme_detail['desc'],
 268             'view_count': int_or_none(stats_info.get('play_count')),
 269             'like_count': int_or_none(stats_info.get('digg_count')),
 270             'repost_count': int_or_none(stats_info.get('share_count')),
 271             'comment_count': int_or_none(stats_info.get('comment_count')),
 272             'uploader': str_or_none(author_info.get('unique_id')),
 273             'creator': str_or_none(author_info.get('nickname')),
 274             'uploader_id': str_or_none(author_info.get('uid')),
 275             'uploader_url': user_url,
 276             'track': music_track,
 277             'album': str_or_none(music_info.get('album')) or None,
 278             'artist': music_author,
 279             'timestamp': int_or_none(aweme_detail.get('create_time')),
 280             'formats': formats,
 281             'subtitles': self.extract_subtitles(aweme_detail, aweme_id),
 282             'thumbnails': thumbnails,
 283             'duration': int_or_none(traverse_obj(video_info, 'duration', ('download_addr', 'duration')), scale=1000),
 284             'availability': self._availability(
 285                 is_private='Private' in labels,
 286                 needs_subscription='Friends only' in labels,
 287                 is_unlisted='Followers only' in labels)
 288         }
 289
 290     def _parse_aweme_video_web(self, aweme_detail, webpage_url):
 291         video_info = aweme_detail['video']
 292         author_info = traverse_obj(aweme_detail, 'authorInfo', 'author', expected_type=dict, default={})
 293         music_info = aweme_detail.get('music') or {}
 294         stats_info = aweme_detail.get('stats') or {}
 295         user_url = self._UPLOADER_URL_FORMAT % (traverse_obj(author_info,
 296                                                              'secUid', 'id', 'uid', 'uniqueId',
 297                                                              expected_type=str_or_none, get_all=False)
 298                                                 or aweme_detail.get('authorSecId'))
 299
 300         formats = []
 301         play_url = video_info.get('playAddr')
 302         width = video_info.get('width')
 303         height = video_info.get('height')
 304         if isinstance(play_url, str):
 305             formats = [{
 306                 'url': self._proto_relative_url(play_url),
 307                 'ext': 'mp4',
 308                 'width': width,
 309                 'height': height,
 310             }]
 311         elif isinstance(play_url, list):
 312             formats = [{
 313                 'url': self._proto_relative_url(url),
 314                 'ext': 'mp4',
 315                 'width': width,
 316                 'height': height,
 317             } for url in traverse_obj(play_url, (..., 'src'), expected_type=url_or_none, default=[]) if url]
 318
 319         download_url = url_or_none(video_info.get('downloadAddr')) or traverse_obj(video_info, ('download', 'url'), expected_type=url_or_none)
 320         if download_url:
 321             formats.append({
 322                 'format_id': 'download',
 323                 'url': self._proto_relative_url(download_url),
 324                 'ext': 'mp4',
 325                 'width': width,
 326                 'height': height,
 327             })
 328         self._remove_duplicate_formats(formats)
 329         self._sort_formats(formats)
 330
 331         thumbnails = []
 332         for thumbnail_name in ('thumbnail', 'cover', 'dynamicCover', 'originCover'):
 333             if aweme_detail.get(thumbnail_name):
 334                 thumbnails = [{
 335                     'url': self._proto_relative_url(aweme_detail[thumbnail_name]),
 336                     'width': width,
 337                     'height': height
 338                 }]
 339
 340         return {
 341             'id': traverse_obj(aweme_detail, 'id', 'awemeId', expected_type=str_or_none),
 342             'title': aweme_detail.get('desc'),
 343             'duration': try_get(aweme_detail, lambda x: x['video']['duration'], int),
 344             'view_count': int_or_none(stats_info.get('playCount')),
 345             'like_count': int_or_none(stats_info.get('diggCount')),
 346             'repost_count': int_or_none(stats_info.get('shareCount')),
 347             'comment_count': int_or_none(stats_info.get('commentCount')),
 348             'timestamp': int_or_none(aweme_detail.get('createTime')),
 349             'creator': str_or_none(author_info.get('nickname')),
 350             'uploader': str_or_none(author_info.get('uniqueId') or aweme_detail.get('author')),
 351             'uploader_id': str_or_none(author_info.get('id') or aweme_detail.get('authorId')),
 352             'uploader_url': user_url,
 353             'track': str_or_none(music_info.get('title')),
 354             'album': str_or_none(music_info.get('album')) or None,
 355             'artist': str_or_none(music_info.get('authorName')),
 356             'formats': formats,
 357             'thumbnails': thumbnails,
 358             'description': str_or_none(aweme_detail.get('desc')),
 359             'http_headers': {
 360                 'Referer': webpage_url
 361             }
 362         }
 363
 364
 365 class TikTokIE(TikTokBaseIE):
 366     _VALID_URL = r'https?://www\.tiktok\.com/@[\w\.-]+/video/(?P<id>\d+)'
 367
 368     _TESTS = [{
 369         'url': 'https://www.tiktok.com/@leenabhushan/video/6748451240264420610',
 370         'md5': '736bb7a466c6f0a6afeb597da1e6f5b7',
 371         'info_dict': {
 372             'id': '6748451240264420610',
 373             'ext': 'mp4',
 374             'title': '#jassmanak #lehanga #leenabhushan',
 375             'description': '#jassmanak #lehanga #leenabhushan',
 376             'duration': 13,
 377             'height': 1024,
 378             'width': 576,
 379             'uploader': 'leenabhushan',
 380             'uploader_id': '6691488002098119685',
 381             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA_Eb4t1vodM1IuTy_cvp9CY22RAb59xqrO0Xtz9CYQJvgXaDvZxYnZYRzDWhhgJmy',
 382             'creator': 'facestoriesbyleenabh',
 383             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 384             'upload_date': '20191016',
 385             'timestamp': 1571246252,
 386             'view_count': int,
 387             'like_count': int,
 388             'repost_count': int,
 389             'comment_count': int,
 390         }
 391     }, {
 392         'url': 'https://www.tiktok.com/@patroxofficial/video/6742501081818877190?langCountry=en',
 393         'md5': '6f3cf8cdd9b28cb8363fe0a9a160695b',
 394         'info_dict': {
 395             'id': '6742501081818877190',
 396             'ext': 'mp4',
 397             'title': 'md5:5e2a23877420bb85ce6521dbee39ba94',
 398             'description': 'md5:5e2a23877420bb85ce6521dbee39ba94',
 399             'duration': 27,
 400             'height': 960,
 401             'width': 540,
 402             'uploader': 'patrox',
 403             'uploader_id': '18702747',
 404             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAiFnldaILebi5heDoVU6bn4jBWWycX6-9U3xuNPqZ8Ws',
 405             'creator': 'patroX',
 406             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 407             'upload_date': '20190930',
 408             'timestamp': 1569860870,
 409             'view_count': int,
 410             'like_count': int,
 411             'repost_count': int,
 412             'comment_count': int,
 413         }
 414     }, {
 415         # Banned audio, only available on the app
 416         'url': 'https://www.tiktok.com/@barudakhb_/video/6984138651336838402',
 417         'info_dict': {
 418             'id': '6984138651336838402',
 419             'ext': 'mp4',
 420             'title': 'Balas @yolaaftwsr hayu yu ? #SquadRandom_ 🔥',
 421             'description': 'Balas @yolaaftwsr hayu yu ? #SquadRandom_ 🔥',
 422             'uploader': 'barudakhb_',
 423             'creator': 'md5:29f238c49bc0c176cb3cef1a9cea9fa6',
 424             'uploader_id': '6974687867511718913',
 425             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAbhBwQC-R1iKoix6jDFsF-vBdfx2ABoDjaZrM9fX6arU3w71q3cOWgWuTXn1soZ7d',
 426             'track': 'Boka Dance',
 427             'artist': 'md5:29f238c49bc0c176cb3cef1a9cea9fa6',
 428             'timestamp': 1626121503,
 429             'duration': 18,
 430             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 431             'upload_date': '20210712',
 432             'view_count': int,
 433             'like_count': int,
 434             'repost_count': int,
 435             'comment_count': int,
 436         }
 437     }, {
 438         # Sponsored video, only available with feed workaround
 439         'url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_/video/7042692929109986561',
 440         'info_dict': {
 441             'id': '7042692929109986561',
 442             'ext': 'mp4',
 443             'title': 'Slap and Run!',
 444             'description': 'Slap and Run!',
 445             'uploader': 'user440922249',
 446             'creator': 'Slap And Run',
 447             'uploader_id': '7036055384943690754',
 448             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_',
 449             'track': 'Promoted Music',
 450             'timestamp': 1639754738,
 451             'duration': 30,
 452             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 453             'upload_date': '20211217',
 454             'view_count': int,
 455             'like_count': int,
 456             'repost_count': int,
 457             'comment_count': int,
 458         },
 459         'expected_warnings': ['Video not available']
 460     }, {
 461         # Auto-captions available
 462         'url': 'https://www.tiktok.com/@hankgreen1/video/7047596209028074758',
 463         'only_matching': True
 464     }]
 465
 466     def _extract_aweme_app(self, aweme_id):
 467         try:
 468             aweme_detail = self._call_api('aweme/detail', {'aweme_id': aweme_id}, aweme_id,
 469                                           note='Downloading video details', errnote='Unable to download video details').get('aweme_detail')
 470             if not aweme_detail:
 471                 raise ExtractorError('Video not available', video_id=aweme_id)
 472         except ExtractorError as e:
 473             self.report_warning(f'{e}; Retrying with feed workaround')
 474             feed_list = self._call_api('feed', {'aweme_id': aweme_id}, aweme_id,
 475                                        note='Downloading video feed', errnote='Unable to download video feed').get('aweme_list') or []
 476             aweme_detail = next((aweme for aweme in feed_list if str(aweme.get('aweme_id')) == aweme_id), None)
 477             if not aweme_detail:
 478                 raise ExtractorError('Unable to find video in feed', video_id=aweme_id)
 479         return self._parse_aweme_video_app(aweme_detail)
 480
 481     def _real_extract(self, url):
 482         video_id = self._match_id(url)
 483
 484         try:
 485             return self._extract_aweme_app(video_id)
 486         except ExtractorError as e:
 487             self.report_warning(f'{e}; Retrying with webpage')
 488
 489         # If we only call once, we get a 403 when downlaoding the video.
 490         self._download_webpage(url, video_id)
 491         webpage = self._download_webpage(url, video_id, note='Downloading video webpage')
 492         next_data = self._search_nextjs_data(webpage, video_id, default='{}')
 493
 494         if next_data:
 495             status = traverse_obj(next_data, ('props', 'pageProps', 'statusCode'), expected_type=int) or 0
 496             video_data = traverse_obj(next_data, ('props', 'pageProps', 'itemInfo', 'itemStruct'), expected_type=dict)
 497         else:
 498             sigi_json = self._search_regex(
 499                 r'>\s*window\[[\'"]SIGI_STATE[\'"]\]\s*=\s*(?P<sigi_state>{.+});',
 500                 webpage, 'sigi data', group='sigi_state')
 501             sigi_data = self._parse_json(sigi_json, video_id)
 502             status = traverse_obj(sigi_data, ('VideoPage', 'statusCode'), expected_type=int) or 0
 503             video_data = traverse_obj(sigi_data, ('ItemModule', video_id), expected_type=dict)
 504
 505         if status == 0:
 506             return self._parse_aweme_video_web(video_data, url)
 507         elif status == 10216:
 508             raise ExtractorError('This video is private', expected=True)
 509         raise ExtractorError('Video not available', video_id=video_id)
 510
 511
 512 class TikTokUserIE(TikTokBaseIE):
 513     IE_NAME = 'tiktok:user'
 514     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/@(?P<id>[\w\.-]+)/?(?:$|[#?])'
 515     _TESTS = [{
 516         'url': 'https://tiktok.com/@corgibobaa?lang=en',
 517         'playlist_mincount': 45,
 518         'info_dict': {
 519             'id': '6935371178089399301',
 520             'title': 'corgibobaa',
 521             'thumbnail': r're:https://.+_1080x1080\.webp'
 522         },
 523         'expected_warnings': ['Retrying']
 524     }, {
 525         'url': 'https://www.tiktok.com/@meme',
 526         'playlist_mincount': 593,
 527         'info_dict': {
 528             'id': '79005827461758976',
 529             'title': 'meme',
 530             'thumbnail': r're:https://.+_1080x1080\.webp'
 531         },
 532         'expected_warnings': ['Retrying']
 533     }]
 534
 535     r'''  # TODO: Fix by adding _signature to api_url
 536     def _entries(self, webpage, user_id, username):
 537         secuid = self._search_regex(r'\"secUid\":\"(?P<secUid>[^\"]+)', webpage, username)
 538         verifyfp_cookie = self._get_cookies('https://www.tiktok.com').get('s_v_web_id')
 539         if not verifyfp_cookie:
 540             raise ExtractorError('Improper cookies (missing s_v_web_id).', expected=True)
 541         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='
 542         cursor = '0'
 543         for page in itertools.count():
 544             data_json = self._download_json(api_url + cursor, username, note='Downloading Page %d' % page)
 545             for video in data_json.get('itemList', []):
 546                 video_id = video['id']
 547                 video_url = f'https://www.tiktok.com/@{user_id}/video/{video_id}'
 548                 yield self._url_result(video_url, 'TikTok', video_id, str_or_none(video.get('desc')))
 549             if not data_json.get('hasMore'):
 550                 break
 551             cursor = data_json['cursor']
 552     '''
 553
 554     def _video_entries_api(self, webpage, user_id, username):
 555         query = {
 556             'user_id': user_id,
 557             'count': 21,
 558             'max_cursor': 0,
 559             'min_cursor': 0,
 560             'retry_type': 'no_retry',
 561             '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.
 562         }
 563
 564         max_retries = self.get_param('extractor_retries', 3)
 565         for page in itertools.count(1):
 566             for retries in itertools.count():
 567                 try:
 568                     post_list = self._call_api('aweme/post', query, username,
 569                                                note='Downloading user video list page %d%s' % (page, f' (attempt {retries})' if retries != 0 else ''),
 570                                                errnote='Unable to download user video list')
 571                 except ExtractorError as e:
 572                     if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0 and retries != max_retries:
 573                         self.report_warning('%s. Retrying...' % str(e.cause or e.msg))
 574                         continue
 575                     raise
 576                 break
 577             yield from post_list.get('aweme_list', [])
 578             if not post_list.get('has_more'):
 579                 break
 580             query['max_cursor'] = post_list['max_cursor']
 581
 582     def _entries_api(self, user_id, videos):
 583         for video in videos:
 584             yield {
 585                 **self._parse_aweme_video_app(video),
 586                 'extractor_key': TikTokIE.ie_key(),
 587                 'extractor': 'TikTok',
 588                 'webpage_url': f'https://tiktok.com/@{user_id}/video/{video["aweme_id"]}',
 589             }
 590
 591     def _real_extract(self, url):
 592         user_name = self._match_id(url)
 593         webpage = self._download_webpage(url, user_name, headers={
 594             'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)'
 595         })
 596         user_id = self._html_search_regex(r'snssdk\d*://user/profile/(\d+)', webpage, 'user ID')
 597
 598         videos = LazyList(self._video_entries_api(webpage, user_id, user_name))
 599         thumbnail = traverse_obj(videos, (0, 'author', 'avatar_larger', 'url_list', 0))
 600
 601         return self.playlist_result(self._entries_api(user_id, videos), user_id, user_name, thumbnail=thumbnail)
 602
 603
 604 class TikTokBaseListIE(TikTokBaseIE):
 605     def _entries(self, list_id, display_id):
 606         query = {
 607             self._QUERY_NAME: list_id,
 608             'cursor': 0,
 609             'count': 20,
 610             'type': 5,
 611             'device_id': ''.join(random.choice(string.digits) for i in range(19))
 612         }
 613
 614         max_retries = self.get_param('extractor_retries', 3)
 615         for page in itertools.count(1):
 616             for retries in itertools.count():
 617                 try:
 618                     post_list = self._call_api(self._API_ENDPOINT, query, display_id,
 619                                                note='Downloading video list page %d%s' % (page, f' (attempt {retries})' if retries != 0 else ''),
 620                                                errnote='Unable to download video list')
 621                 except ExtractorError as e:
 622                     if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0 and retries != max_retries:
 623                         self.report_warning('%s. Retrying...' % str(e.cause or e.msg))
 624                         continue
 625                     raise
 626                 break
 627             for video in post_list.get('aweme_list', []):
 628                 yield {
 629                     **self._parse_aweme_video_app(video),
 630                     'extractor_key': TikTokIE.ie_key(),
 631                     'extractor': 'TikTok',
 632                     'webpage_url': f'https://tiktok.com/@_/video/{video["aweme_id"]}',
 633                 }
 634             if not post_list.get('has_more'):
 635                 break
 636             query['cursor'] = post_list['cursor']
 637
 638     def _real_extract(self, url):
 639         list_id = self._match_id(url)
 640         return self.playlist_result(self._entries(list_id, list_id), list_id)
 641
 642
 643 class TikTokSoundIE(TikTokBaseListIE):
 644     IE_NAME = 'tiktok:sound'
 645     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/music/[\w\.-]+-(?P<id>[\d]+)[/?#&]?'
 646     _QUERY_NAME = 'music_id'
 647     _API_ENDPOINT = 'music/aweme'
 648     _TESTS = [{
 649         'url': 'https://www.tiktok.com/music/Build-a-Btch-6956990112127585029?lang=en',
 650         'playlist_mincount': 100,
 651         'info_dict': {
 652             'id': '6956990112127585029'
 653         },
 654         'expected_warnings': ['Retrying']
 655     }, {
 656         # Actual entries are less than listed video count
 657         'url': 'https://www.tiktok.com/music/jiefei-soap-remix-7036843036118469381',
 658         'playlist_mincount': 2182,
 659         'info_dict': {
 660             'id': '7036843036118469381'
 661         },
 662         'expected_warnings': ['Retrying']
 663     }]
 664
 665
 666 class TikTokEffectIE(TikTokBaseListIE):
 667     IE_NAME = 'tiktok:effect'
 668     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/sticker/[\w\.-]+-(?P<id>[\d]+)[/?#&]?'
 669     _QUERY_NAME = 'sticker_id'
 670     _API_ENDPOINT = 'sticker/aweme'
 671     _TESTS = [{
 672         'url': 'https://www.tiktok.com/sticker/MATERIAL-GWOOORL-1258156',
 673         'playlist_mincount': 100,
 674         'info_dict': {
 675             'id': '1258156',
 676         },
 677         'expected_warnings': ['Retrying']
 678     }, {
 679         # Different entries between mobile and web, depending on region
 680         'url': 'https://www.tiktok.com/sticker/Elf-Friend-479565',
 681         'only_matching': True
 682     }]
 683
 684
 685 class TikTokTagIE(TikTokBaseListIE):
 686     IE_NAME = 'tiktok:tag'
 687     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/tag/(?P<id>[^/?#&]+)'
 688     _QUERY_NAME = 'ch_id'
 689     _API_ENDPOINT = 'challenge/aweme'
 690     _TESTS = [{
 691         'url': 'https://tiktok.com/tag/hello2018',
 692         'playlist_mincount': 39,
 693         'info_dict': {
 694             'id': '46294678',
 695             'title': 'hello2018',
 696         },
 697         'expected_warnings': ['Retrying']
 698     }, {
 699         'url': 'https://tiktok.com/tag/fypシ?is_copy_url=0&is_from_webapp=v1',
 700         'only_matching': True
 701     }]
 702
 703     def _real_extract(self, url):
 704         display_id = self._match_id(url)
 705         webpage = self._download_webpage(url, display_id, headers={
 706             'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)'
 707         })
 708         tag_id = self._html_search_regex(r'snssdk\d*://challenge/detail/(\d+)', webpage, 'tag ID')
 709         return self.playlist_result(self._entries(tag_id, display_id), tag_id, display_id)
 710
 711
 712 class DouyinIE(TikTokIE):
 713     _VALID_URL = r'https?://(?:www\.)?douyin\.com/video/(?P<id>[0-9]+)'
 714     _TESTS = [{
 715         'url': 'https://www.douyin.com/video/6961737553342991651',
 716         'md5': '10523312c8b8100f353620ac9dc8f067',
 717         'info_dict': {
 718             'id': '6961737553342991651',
 719             'ext': 'mp4',
 720             'title': '#杨超越  小小水手带你去远航❤️',
 721             'uploader': '杨超越',
 722             'upload_date': '20210513',
 723             'timestamp': 1620905839,
 724             'uploader_id': '110403406559',
 725             'view_count': int,
 726             'like_count': int,
 727             'repost_count': int,
 728             'comment_count': int,
 729         }
 730     }, {
 731         'url': 'https://www.douyin.com/video/6982497745948921092',
 732         'md5': 'd78408c984b9b5102904cf6b6bc2d712',
 733         'info_dict': {
 734             'id': '6982497745948921092',
 735             'ext': 'mp4',
 736             'title': '这个夏日和小羊@杨超越 一起遇见白色幻想',
 737             'uploader': '杨超越工作室',
 738             'upload_date': '20210708',
 739             'timestamp': 1625739481,
 740             'uploader_id': '408654318141572',
 741             'view_count': int,
 742             'like_count': int,
 743             'repost_count': int,
 744             'comment_count': int,
 745         }
 746     }, {
 747         'url': 'https://www.douyin.com/video/6953975910773099811',
 748         'md5': '72e882e24f75064c218b76c8b713c185',
 749         'info_dict': {
 750             'id': '6953975910773099811',
 751             'ext': 'mp4',
 752             'title': '#一起看海  出现在你的夏日里',
 753             'uploader': '杨超越',
 754             'upload_date': '20210422',
 755             'timestamp': 1619098692,
 756             'uploader_id': '110403406559',
 757             'view_count': int,
 758             'like_count': int,
 759             'repost_count': int,
 760             'comment_count': int,
 761         }
 762     }, {
 763         'url': 'https://www.douyin.com/video/6950251282489675042',
 764         'md5': 'b4db86aec367ef810ddd38b1737d2fed',
 765         'info_dict': {
 766             'id': '6950251282489675042',
 767             'ext': 'mp4',
 768             'title': '哈哈哈，成功了哈哈哈哈哈哈',
 769             'uploader': '杨超越',
 770             'upload_date': '20210412',
 771             'timestamp': 1618231483,
 772             'uploader_id': '110403406559',
 773             'view_count': int,
 774             'like_count': int,
 775             'repost_count': int,
 776             'comment_count': int,
 777         }
 778     }, {
 779         'url': 'https://www.douyin.com/video/6963263655114722595',
 780         'md5': '1abe1c477d05ee62efb40bf2329957cf',
 781         'info_dict': {
 782             'id': '6963263655114722595',
 783             'ext': 'mp4',
 784             'title': '#哪个爱豆的105度最甜 换个角度看看我哈哈',
 785             'uploader': '杨超越',
 786             'upload_date': '20210517',
 787             'timestamp': 1621261163,
 788             'uploader_id': '110403406559',
 789             'view_count': int,
 790             'like_count': int,
 791             'repost_count': int,
 792             'comment_count': int,
 793         }
 794     }]
 795     _APP_VERSIONS = [('9.6.0', '960')]
 796     _APP_NAME = 'aweme'
 797     _AID = 1128
 798     _API_HOSTNAME = 'aweme.snssdk.com'
 799     _UPLOADER_URL_FORMAT = 'https://www.douyin.com/user/%s'
 800     _WEBPAGE_HOST = 'https://www.douyin.com/'
 801
 802     def _real_extract(self, url):
 803         video_id = self._match_id(url)
 804
 805         try:
 806             return self._extract_aweme_app(video_id)
 807         except ExtractorError as e:
 808             self.report_warning(f'{e}; Retrying with webpage')
 809
 810         webpage = self._download_webpage(url, video_id)
 811         render_data_json = self._search_regex(
 812             r'<script [^>]*\bid=[\'"]RENDER_DATA[\'"][^>]*>(%7B.+%7D)</script>',
 813             webpage, 'render data', default=None)
 814         if not render_data_json:
 815             # TODO: Run verification challenge code to generate signature cookies
 816             raise ExtractorError('Fresh cookies (not necessarily logged in) are needed')
 817
 818         render_data = self._parse_json(
 819             render_data_json, video_id, transform_source=compat_urllib_parse_unquote)
 820         return self._parse_aweme_video_web(get_first(render_data, ('aweme', 'detail')), url)
 821
 822
 823 class TikTokVMIE(InfoExtractor):
 824     _VALID_URL = r'https?://(?:vm|vt)\.tiktok\.com/(?P<id>\w+)'
 825     IE_NAME = 'vm.tiktok'
 826
 827     _TESTS = [{
 828         'url': 'https://vm.tiktok.com/ZSe4FqkKd',
 829         'info_dict': {
 830             'id': '7023491746608712966',
 831             'ext': 'mp4',
 832             'title': 'md5:5607564db90271abbbf8294cca77eddd',
 833             'description': 'md5:5607564db90271abbbf8294cca77eddd',
 834             'duration': 11,
 835             'upload_date': '20211026',
 836             'uploader_id': '7007385080558846981',
 837             'creator': 'Memes',
 838             'artist': 'Memes',
 839             'track': 'original sound',
 840             'uploader': 'susmandem',
 841             'timestamp': 1635284105,
 842             'thumbnail': r're:https://.+\.webp.*',
 843             'like_count': int,
 844             'view_count': int,
 845             'comment_count': int,
 846             'repost_count': int,
 847             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAXcNoOEOxVyBzuII_E--T0MeCrLP0ay1Sm6x_n3dluiWEoWZD0VlQOytwad4W0i0n',
 848         }
 849     }, {
 850         'url': 'https://vt.tiktok.com/ZSe4FqkKd',
 851         'only_matching': True,
 852     }]
 853
 854     def _real_extract(self, url):
 855         return self.url_result(self._request_webpage(
 856             HEADRequest(url), self._match_id(url), headers={'User-Agent': 'facebookexternalhit/1.1'}).geturl(), TikTokIE)