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     srt_subtitles_timecode,
  20     str_or_none,
  21     traverse_obj,
  22     try_get,
  23     url_or_none,
  24 )
  25
  26
  27 class TikTokBaseIE(InfoExtractor):
  28     _APP_VERSIONS = [('26.1.3', '260103'), ('26.1.2', '260102'), ('26.1.1', '260101'), ('25.6.2', '250602')]
  29     _WORKING_APP_VERSION = None
  30     _APP_NAME = 'trill'
  31     _AID = 1180
  32     _API_HOSTNAME = 'api-h2.tiktokv.com'
  33     _UPLOADER_URL_FORMAT = 'https://www.tiktok.com/@%s'
  34     _WEBPAGE_HOST = 'https://www.tiktok.com/'
  35     QUALITIES = ('360p', '540p', '720p', '1080p')
  36
  37     @staticmethod
  38     def _create_url(user_id, video_id):
  39         return f'https://www.tiktok.com/@{user_id or "_"}/video/{video_id}'
  40
  41     def _get_sigi_state(self, webpage, display_id):
  42         return self._parse_json(get_element_by_id(
  43             'SIGI_STATE|sigi-persisted-data', webpage, escape_value=False), display_id)
  44
  45     def _call_api_impl(self, ep, query, manifest_app_version, video_id, fatal=True,
  46                        note='Downloading API JSON', errnote='Unable to download API page'):
  47         self._set_cookie(self._API_HOSTNAME, 'odin_tt', ''.join(random.choice('0123456789abcdef') for _ in range(160)))
  48         webpage_cookies = self._get_cookies(self._WEBPAGE_HOST)
  49         if webpage_cookies.get('sid_tt'):
  50             self._set_cookie(self._API_HOSTNAME, 'sid_tt', webpage_cookies['sid_tt'].value)
  51         return self._download_json(
  52             'https://%s/aweme/v1/%s/' % (self._API_HOSTNAME, ep), video_id=video_id,
  53             fatal=fatal, note=note, errnote=errnote, headers={
  54                 '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)',
  55                 'Accept': 'application/json',
  56             }, query=query)
  57
  58     def _build_api_query(self, query, app_version, manifest_app_version):
  59         return {
  60             **query,
  61             'version_name': app_version,
  62             'version_code': manifest_app_version,
  63             'build_number': app_version,
  64             'manifest_version_code': manifest_app_version,
  65             'update_version_code': manifest_app_version,
  66             'openudid': ''.join(random.choice('0123456789abcdef') for _ in range(16)),
  67             'uuid': ''.join([random.choice(string.digits) for _ in range(16)]),
  68             '_rticket': int(time.time() * 1000),
  69             'ts': int(time.time()),
  70             'device_brand': 'Google',
  71             'device_type': 'Pixel 4',
  72             'device_platform': 'android',
  73             'resolution': '1080*1920',
  74             'dpi': 420,
  75             'os_version': '10',
  76             'os_api': '29',
  77             'carrier_region': 'US',
  78             'sys_region': 'US',
  79             'region': 'US',
  80             'app_name': self._APP_NAME,
  81             'app_language': 'en',
  82             'language': 'en',
  83             'timezone_name': 'America/New_York',
  84             'timezone_offset': '-14400',
  85             'channel': 'googleplay',
  86             'ac': 'wifi',
  87             'mcc_mnc': '310260',
  88             'is_my_cn': 0,
  89             'aid': self._AID,
  90             'ssmix': 'a',
  91             'as': 'a1qwert123',
  92             'cp': 'cbfhckdckkde1',
  93         }
  94
  95     def _call_api(self, ep, query, video_id, fatal=True,
  96                   note='Downloading API JSON', errnote='Unable to download API page'):
  97         if not self._WORKING_APP_VERSION:
  98             app_version = self._configuration_arg('app_version', [''], ie_key=TikTokIE.ie_key())[0]
  99             manifest_app_version = self._configuration_arg('manifest_app_version', [''], ie_key=TikTokIE.ie_key())[0]
 100             if app_version and manifest_app_version:
 101                 self._WORKING_APP_VERSION = (app_version, manifest_app_version)
 102                 self.write_debug('Imported app version combo from extractor arguments')
 103             elif app_version or manifest_app_version:
 104                 self.report_warning('Only one of the two required version params are passed as extractor arguments', only_once=True)
 105
 106         if self._WORKING_APP_VERSION:
 107             app_version, manifest_app_version = self._WORKING_APP_VERSION
 108             real_query = self._build_api_query(query, app_version, manifest_app_version)
 109             return self._call_api_impl(ep, real_query, manifest_app_version, video_id, fatal, note, errnote)
 110
 111         for count, (app_version, manifest_app_version) in enumerate(self._APP_VERSIONS, start=1):
 112             real_query = self._build_api_query(query, app_version, manifest_app_version)
 113             try:
 114                 res = self._call_api_impl(ep, real_query, manifest_app_version, video_id, fatal, note, errnote)
 115                 self._WORKING_APP_VERSION = (app_version, manifest_app_version)
 116                 return res
 117             except ExtractorError as e:
 118                 if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0:
 119                     if count == len(self._APP_VERSIONS):
 120                         if fatal:
 121                             raise e
 122                         else:
 123                             self.report_warning(str(e.cause or e.msg))
 124                             return
 125                     self.report_warning('%s. Retrying... (attempt %s of %s)' % (str(e.cause or e.msg), count, len(self._APP_VERSIONS)))
 126                     continue
 127                 raise e
 128
 129     def _get_subtitles(self, aweme_detail, aweme_id):
 130         # TODO: Extract text positioning info
 131         subtitles = {}
 132         captions_info = traverse_obj(
 133             aweme_detail, ('interaction_stickers', ..., 'auto_video_caption_info', 'auto_captions', ...), expected_type=dict, default=[])
 134         for caption in captions_info:
 135             caption_url = traverse_obj(caption, ('url', 'url_list', ...), expected_type=url_or_none, get_all=False)
 136             if not caption_url:
 137                 continue
 138             caption_json = self._download_json(
 139                 caption_url, aweme_id, note='Downloading captions', errnote='Unable to download captions', fatal=False)
 140             if not caption_json:
 141                 continue
 142             subtitles.setdefault(caption.get('language', 'en'), []).append({
 143                 'ext': 'srt',
 144                 'data': '\n\n'.join(
 145                     f'{i + 1}\n{srt_subtitles_timecode(line["start_time"] / 1000)} --> {srt_subtitles_timecode(line["end_time"] / 1000)}\n{line["text"]}'
 146                     for i, line in enumerate(caption_json['utterances']) if line.get('text'))
 147             })
 148         return subtitles
 149
 150     def _parse_aweme_video_app(self, aweme_detail):
 151         aweme_id = aweme_detail['aweme_id']
 152         video_info = aweme_detail['video']
 153
 154         def parse_url_key(url_key):
 155             format_id, codec, res, bitrate = self._search_regex(
 156                 r'v[^_]+_(?P<id>(?P<codec>[^_]+)_(?P<res>\d+p)_(?P<bitrate>\d+))', url_key,
 157                 'url key', default=(None, None, None, None), group=('id', 'codec', 'res', 'bitrate'))
 158             if not format_id:
 159                 return {}, None
 160             return {
 161                 'format_id': format_id,
 162                 'vcodec': 'h265' if codec == 'bytevc1' else codec,
 163                 'tbr': int_or_none(bitrate, scale=1000) or None,
 164                 'quality': qualities(self.QUALITIES)(res),
 165             }, res
 166
 167         known_resolutions = {}
 168
 169         def extract_addr(addr, add_meta={}):
 170             parsed_meta, res = parse_url_key(addr.get('url_key', ''))
 171             if res:
 172                 known_resolutions.setdefault(res, {}).setdefault('height', add_meta.get('height'))
 173                 known_resolutions[res].setdefault('width', add_meta.get('width'))
 174                 parsed_meta.update(known_resolutions.get(res, {}))
 175                 add_meta.setdefault('height', int_or_none(res[:-1]))
 176             return [{
 177                 'url': url,
 178                 'filesize': int_or_none(addr.get('data_size')),
 179                 'ext': 'mp4',
 180                 'acodec': 'aac',
 181                 'source_preference': -2 if 'aweme/v1' in url else -1,  # Downloads from API might get blocked
 182                 **add_meta, **parsed_meta,
 183                 'format_note': join_nonempty(
 184                     add_meta.get('format_note'), '(API)' if 'aweme/v1' in url else None, delim=' ')
 185             } for url in addr.get('url_list') or []]
 186
 187         # Hack: Add direct video links first to prioritize them when removing duplicate formats
 188         formats = []
 189         if video_info.get('play_addr'):
 190             formats.extend(extract_addr(video_info['play_addr'], {
 191                 'format_id': 'play_addr',
 192                 'format_note': 'Direct video',
 193                 'vcodec': 'h265' if traverse_obj(
 194                     video_info, 'is_bytevc1', 'is_h265') else 'h264',  # TODO: Check for "direct iOS" videos, like https://www.tiktok.com/@cookierun_dev/video/7039716639834656002
 195                 'width': video_info.get('width'),
 196                 'height': video_info.get('height'),
 197             }))
 198         if video_info.get('download_addr'):
 199             formats.extend(extract_addr(video_info['download_addr'], {
 200                 'format_id': 'download_addr',
 201                 'format_note': 'Download video%s' % (', watermarked' if video_info.get('has_watermark') else ''),
 202                 'vcodec': 'h264',
 203                 'width': video_info.get('width'),
 204                 'height': video_info.get('height'),
 205                 'preference': -2 if video_info.get('has_watermark') else -1,
 206             }))
 207         if video_info.get('play_addr_h264'):
 208             formats.extend(extract_addr(video_info['play_addr_h264'], {
 209                 'format_id': 'play_addr_h264',
 210                 'format_note': 'Direct video',
 211                 'vcodec': 'h264',
 212             }))
 213         if video_info.get('play_addr_bytevc1'):
 214             formats.extend(extract_addr(video_info['play_addr_bytevc1'], {
 215                 'format_id': 'play_addr_bytevc1',
 216                 'format_note': 'Direct video',
 217                 'vcodec': 'h265',
 218             }))
 219
 220         for bitrate in video_info.get('bit_rate', []):
 221             if bitrate.get('play_addr'):
 222                 formats.extend(extract_addr(bitrate['play_addr'], {
 223                     'format_id': bitrate.get('gear_name'),
 224                     'format_note': 'Playback video',
 225                     'tbr': try_get(bitrate, lambda x: x['bit_rate'] / 1000),
 226                     'vcodec': 'h265' if traverse_obj(
 227                         bitrate, 'is_bytevc1', 'is_h265') else 'h264',
 228                     'fps': bitrate.get('FPS'),
 229                 }))
 230
 231         self._remove_duplicate_formats(formats)
 232         auth_cookie = self._get_cookies(self._WEBPAGE_HOST).get('sid_tt')
 233         if auth_cookie:
 234             for f in formats:
 235                 self._set_cookie(compat_urllib_parse_urlparse(f['url']).hostname, 'sid_tt', auth_cookie.value)
 236         self._sort_formats(formats, ('quality', 'codec', 'size', 'br'))
 237
 238         thumbnails = []
 239         for cover_id in ('cover', 'ai_dynamic_cover', 'animated_cover', 'ai_dynamic_cover_bak',
 240                          'origin_cover', 'dynamic_cover'):
 241             cover = video_info.get(cover_id)
 242             if cover:
 243                 for cover_url in cover['url_list']:
 244                     thumbnails.append({
 245                         'id': cover_id,
 246                         'url': cover_url,
 247                     })
 248
 249         stats_info = aweme_detail.get('statistics', {})
 250         author_info = aweme_detail.get('author', {})
 251         music_info = aweme_detail.get('music', {})
 252         user_url = self._UPLOADER_URL_FORMAT % (traverse_obj(author_info,
 253                                                              'sec_uid', 'id', 'uid', 'unique_id',
 254                                                              expected_type=str_or_none, get_all=False))
 255         labels = traverse_obj(aweme_detail, ('hybrid_label', ..., 'text'), expected_type=str, default=[])
 256
 257         contained_music_track = traverse_obj(
 258             music_info, ('matched_song', 'title'), ('matched_pgc_sound', 'title'), expected_type=str)
 259         contained_music_author = traverse_obj(
 260             music_info, ('matched_song', 'author'), ('matched_pgc_sound', 'author'), 'author', expected_type=str)
 261
 262         is_generic_og_trackname = music_info.get('is_original_sound') and music_info.get('title') == 'original sound - %s' % music_info.get('owner_handle')
 263         if is_generic_og_trackname:
 264             music_track, music_author = contained_music_track or 'original sound', contained_music_author
 265         else:
 266             music_track, music_author = music_info.get('title'), music_info.get('author')
 267
 268         return {
 269             'id': aweme_id,
 270             'extractor_key': TikTokIE.ie_key(),
 271             'extractor': TikTokIE.IE_NAME,
 272             'webpage_url': self._create_url(author_info.get('uid'), aweme_id),
 273             'title': aweme_detail.get('desc'),
 274             'description': aweme_detail.get('desc'),
 275             'view_count': int_or_none(stats_info.get('play_count')),
 276             'like_count': int_or_none(stats_info.get('digg_count')),
 277             'repost_count': int_or_none(stats_info.get('share_count')),
 278             'comment_count': int_or_none(stats_info.get('comment_count')),
 279             'uploader': str_or_none(author_info.get('unique_id')),
 280             'creator': str_or_none(author_info.get('nickname')),
 281             'uploader_id': str_or_none(author_info.get('uid')),
 282             'uploader_url': user_url,
 283             'track': music_track,
 284             'album': str_or_none(music_info.get('album')) or None,
 285             'artist': music_author or None,
 286             'timestamp': int_or_none(aweme_detail.get('create_time')),
 287             'formats': formats,
 288             'subtitles': self.extract_subtitles(aweme_detail, aweme_id),
 289             'thumbnails': thumbnails,
 290             'duration': int_or_none(traverse_obj(video_info, 'duration', ('download_addr', 'duration')), scale=1000),
 291             'availability': self._availability(
 292                 is_private='Private' in labels,
 293                 needs_subscription='Friends only' in labels,
 294                 is_unlisted='Followers only' in labels)
 295         }
 296
 297     def _parse_aweme_video_web(self, aweme_detail, webpage_url):
 298         video_info = aweme_detail['video']
 299         author_info = traverse_obj(aweme_detail, 'authorInfo', 'author', expected_type=dict, default={})
 300         music_info = aweme_detail.get('music') or {}
 301         stats_info = aweme_detail.get('stats') or {}
 302         user_url = self._UPLOADER_URL_FORMAT % (traverse_obj(author_info,
 303                                                              'secUid', 'id', 'uid', 'uniqueId',
 304                                                              expected_type=str_or_none, get_all=False)
 305                                                 or aweme_detail.get('authorSecId'))
 306
 307         formats = []
 308         play_url = video_info.get('playAddr')
 309         width = video_info.get('width')
 310         height = video_info.get('height')
 311         if isinstance(play_url, str):
 312             formats = [{
 313                 'url': self._proto_relative_url(play_url),
 314                 'ext': 'mp4',
 315                 'width': width,
 316                 'height': height,
 317             }]
 318         elif isinstance(play_url, list):
 319             formats = [{
 320                 'url': self._proto_relative_url(url),
 321                 'ext': 'mp4',
 322                 'width': width,
 323                 'height': height,
 324             } for url in traverse_obj(play_url, (..., 'src'), expected_type=url_or_none, default=[]) if url]
 325
 326         download_url = url_or_none(video_info.get('downloadAddr')) or traverse_obj(video_info, ('download', 'url'), expected_type=url_or_none)
 327         if download_url:
 328             formats.append({
 329                 'format_id': 'download',
 330                 'url': self._proto_relative_url(download_url),
 331                 'ext': 'mp4',
 332                 'width': width,
 333                 'height': height,
 334             })
 335         self._remove_duplicate_formats(formats)
 336         self._sort_formats(formats)
 337
 338         thumbnails = []
 339         for thumbnail_name in ('thumbnail', 'cover', 'dynamicCover', 'originCover'):
 340             if aweme_detail.get(thumbnail_name):
 341                 thumbnails = [{
 342                     'url': self._proto_relative_url(aweme_detail[thumbnail_name]),
 343                     'width': width,
 344                     'height': height
 345                 }]
 346
 347         return {
 348             'id': traverse_obj(aweme_detail, 'id', 'awemeId', expected_type=str_or_none),
 349             'title': aweme_detail.get('desc'),
 350             'duration': try_get(aweme_detail, lambda x: x['video']['duration'], int),
 351             'view_count': int_or_none(stats_info.get('playCount')),
 352             'like_count': int_or_none(stats_info.get('diggCount')),
 353             'repost_count': int_or_none(stats_info.get('shareCount')),
 354             'comment_count': int_or_none(stats_info.get('commentCount')),
 355             'timestamp': int_or_none(aweme_detail.get('createTime')),
 356             'creator': str_or_none(author_info.get('nickname')),
 357             'uploader': str_or_none(author_info.get('uniqueId') or aweme_detail.get('author')),
 358             'uploader_id': str_or_none(author_info.get('id') or aweme_detail.get('authorId')),
 359             'uploader_url': user_url,
 360             'track': str_or_none(music_info.get('title')),
 361             'album': str_or_none(music_info.get('album')) or None,
 362             'artist': str_or_none(music_info.get('authorName')),
 363             'formats': formats,
 364             'thumbnails': thumbnails,
 365             'description': str_or_none(aweme_detail.get('desc')),
 366             'http_headers': {
 367                 'Referer': webpage_url
 368             }
 369         }
 370
 371
 372 class TikTokIE(TikTokBaseIE):
 373     _VALID_URL = r'https?://www\.tiktok\.com/(?:embed|@(?P<user_id>[\w\.-]+)/video)/(?P<id>\d+)'
 374     _EMBED_REGEX = [rf'<(?:script|iframe)[^>]+\bsrc=(["\'])(?P<url>{_VALID_URL})']
 375
 376     _TESTS = [{
 377         'url': 'https://www.tiktok.com/@leenabhushan/video/6748451240264420610',
 378         'md5': '736bb7a466c6f0a6afeb597da1e6f5b7',
 379         'info_dict': {
 380             'id': '6748451240264420610',
 381             'ext': 'mp4',
 382             'title': '#jassmanak #lehanga #leenabhushan',
 383             'description': '#jassmanak #lehanga #leenabhushan',
 384             'duration': 13,
 385             'height': 1024,
 386             'width': 576,
 387             'uploader': 'leenabhushan',
 388             'uploader_id': '6691488002098119685',
 389             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA_Eb4t1vodM1IuTy_cvp9CY22RAb59xqrO0Xtz9CYQJvgXaDvZxYnZYRzDWhhgJmy',
 390             'creator': 'facestoriesbyleenabh',
 391             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 392             'upload_date': '20191016',
 393             'timestamp': 1571246252,
 394             'view_count': int,
 395             'like_count': int,
 396             'repost_count': int,
 397             'comment_count': int,
 398             'artist': 'Ysrbeats',
 399             'album': 'Lehanga',
 400             'track': 'Lehanga',
 401         }
 402     }, {
 403         'url': 'https://www.tiktok.com/@patroxofficial/video/6742501081818877190?langCountry=en',
 404         'md5': '6f3cf8cdd9b28cb8363fe0a9a160695b',
 405         'info_dict': {
 406             'id': '6742501081818877190',
 407             'ext': 'mp4',
 408             'title': 'md5:5e2a23877420bb85ce6521dbee39ba94',
 409             'description': 'md5:5e2a23877420bb85ce6521dbee39ba94',
 410             'duration': 27,
 411             'height': 960,
 412             'width': 540,
 413             'uploader': 'patrox',
 414             'uploader_id': '18702747',
 415             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAiFnldaILebi5heDoVU6bn4jBWWycX6-9U3xuNPqZ8Ws',
 416             'creator': 'patroX',
 417             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 418             'upload_date': '20190930',
 419             'timestamp': 1569860870,
 420             'view_count': int,
 421             'like_count': int,
 422             'repost_count': int,
 423             'comment_count': int,
 424             'artist': 'Evan Todd, Jessica Keenan Wynn, Alice Lee, Barrett Wilbert Weed & Jon Eidson',
 425             'track': 'Big Fun',
 426         }
 427     }, {
 428         # Banned audio, only available on the app
 429         'url': 'https://www.tiktok.com/@barudakhb_/video/6984138651336838402',
 430         'info_dict': {
 431             'id': '6984138651336838402',
 432             'ext': 'mp4',
 433             'title': 'Balas @yolaaftwsr hayu yu ? #SquadRandom_ 🔥',
 434             'description': 'Balas @yolaaftwsr hayu yu ? #SquadRandom_ 🔥',
 435             'uploader': 'barudakhb_',
 436             'creator': 'md5:29f238c49bc0c176cb3cef1a9cea9fa6',
 437             'uploader_id': '6974687867511718913',
 438             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAbhBwQC-R1iKoix6jDFsF-vBdfx2ABoDjaZrM9fX6arU3w71q3cOWgWuTXn1soZ7d',
 439             'track': 'Boka Dance',
 440             'artist': 'md5:29f238c49bc0c176cb3cef1a9cea9fa6',
 441             'timestamp': 1626121503,
 442             'duration': 18,
 443             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 444             'upload_date': '20210712',
 445             'view_count': int,
 446             'like_count': int,
 447             'repost_count': int,
 448             'comment_count': int,
 449         }
 450     }, {
 451         # Sponsored video, only available with feed workaround
 452         'url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_/video/7042692929109986561',
 453         'info_dict': {
 454             'id': '7042692929109986561',
 455             'ext': 'mp4',
 456             'title': 'Slap and Run!',
 457             'description': 'Slap and Run!',
 458             'uploader': 'user440922249',
 459             'creator': 'Slap And Run',
 460             'uploader_id': '7036055384943690754',
 461             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_',
 462             'track': 'Promoted Music',
 463             'timestamp': 1639754738,
 464             'duration': 30,
 465             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 466             'upload_date': '20211217',
 467             'view_count': int,
 468             'like_count': int,
 469             'repost_count': int,
 470             'comment_count': int,
 471         },
 472         'expected_warnings': ['trying with webpage', 'Unable to find video in feed']
 473     }, {
 474         # Video without title and description
 475         'url': 'https://www.tiktok.com/@pokemonlife22/video/7059698374567611694',
 476         'info_dict': {
 477             'id': '7059698374567611694',
 478             'ext': 'mp4',
 479             'title': 'TikTok video #7059698374567611694',
 480             'description': '',
 481             'uploader': 'pokemonlife22',
 482             'creator': 'Pokemon',
 483             'uploader_id': '6820838815978423302',
 484             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA0tF1nBwQVVMyrGu3CqttkNgM68Do1OXUFuCY0CRQk8fEtSVDj89HqoqvbSTmUP2W',
 485             'track': 'original sound',
 486             'timestamp': 1643714123,
 487             'duration': 6,
 488             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 489             'upload_date': '20220201',
 490             'artist': 'Pokemon',
 491             'view_count': int,
 492             'like_count': int,
 493             'repost_count': int,
 494             'comment_count': int,
 495         },
 496     }, {
 497         # hydration JSON is sent in a <script> element
 498         'url': 'https://www.tiktok.com/@denidil6/video/7065799023130643713',
 499         'info_dict': {
 500             'id': '7065799023130643713',
 501             'ext': 'mp4',
 502             'title': '#denidil#денидил',
 503             'description': '#denidil#денидил',
 504             'uploader': 'denidil6',
 505             'uploader_id': '7046664115636405250',
 506             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAsvMSzFdQ4ikl3uR2TEJwMBbB2yZh2Zxwhx-WCo3rbDpAharE3GQCrFuJArI3C8QJ',
 507             'artist': 'Holocron Music',
 508             'album': 'Wolf Sounds (1 Hour) Enjoy the Company of the Animal That Is the Majestic King of the Night',
 509             'track': 'Wolf Sounds (1 Hour) Enjoy the Company of the Animal That Is the Majestic King of the Night',
 510             'timestamp': 1645134536,
 511             'duration': 26,
 512             'upload_date': '20220217',
 513             'view_count': int,
 514             'like_count': int,
 515             'repost_count': int,
 516             'comment_count': int,
 517         },
 518         'skip': 'This video is unavailable',
 519     }, {
 520         # Auto-captions available
 521         'url': 'https://www.tiktok.com/@hankgreen1/video/7047596209028074758',
 522         'only_matching': True
 523     }]
 524
 525     def _extract_aweme_app(self, aweme_id):
 526         feed_list = self._call_api('feed', {'aweme_id': aweme_id}, aweme_id,
 527                                    note='Downloading video feed', errnote='Unable to download video feed').get('aweme_list') or []
 528         aweme_detail = next((aweme for aweme in feed_list if str(aweme.get('aweme_id')) == aweme_id), None)
 529         if not aweme_detail:
 530             raise ExtractorError('Unable to find video in feed', video_id=aweme_id)
 531         return self._parse_aweme_video_app(aweme_detail)
 532
 533     def _real_extract(self, url):
 534         video_id, user_id = self._match_valid_url(url).group('id', 'user_id')
 535         try:
 536             return self._extract_aweme_app(video_id)
 537         except ExtractorError as e:
 538             self.report_warning(f'{e}; trying with webpage')
 539
 540         url = self._create_url(user_id, video_id)
 541         webpage = self._download_webpage(url, video_id, headers={'User-Agent': 'User-Agent:Mozilla/5.0'})
 542         next_data = self._search_nextjs_data(webpage, video_id, default='{}')
 543         if next_data:
 544             status = traverse_obj(next_data, ('props', 'pageProps', 'statusCode'), expected_type=int) or 0
 545             video_data = traverse_obj(next_data, ('props', 'pageProps', 'itemInfo', 'itemStruct'), expected_type=dict)
 546         else:
 547             sigi_data = self._get_sigi_state(webpage, video_id)
 548             status = traverse_obj(sigi_data, ('VideoPage', 'statusCode'), expected_type=int) or 0
 549             video_data = traverse_obj(sigi_data, ('ItemModule', video_id), expected_type=dict)
 550
 551         if status == 0:
 552             return self._parse_aweme_video_web(video_data, url)
 553         elif status == 10216:
 554             raise ExtractorError('This video is private', expected=True)
 555         raise ExtractorError('Video not available', video_id=video_id)
 556
 557
 558 class TikTokUserIE(TikTokBaseIE):
 559     IE_NAME = 'tiktok:user'
 560     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/@(?P<id>[\w\.-]+)/?(?:$|[#?])'
 561     _WORKING = False
 562     _TESTS = [{
 563         'url': 'https://tiktok.com/@corgibobaa?lang=en',
 564         'playlist_mincount': 45,
 565         'info_dict': {
 566             'id': '6935371178089399301',
 567             'title': 'corgibobaa',
 568             'thumbnail': r're:https://.+_1080x1080\.webp'
 569         },
 570         'expected_warnings': ['Retrying']
 571     }, {
 572         'url': 'https://www.tiktok.com/@6820838815978423302',
 573         'playlist_mincount': 5,
 574         'info_dict': {
 575             'id': '6820838815978423302',
 576             'title': '6820838815978423302',
 577             'thumbnail': r're:https://.+_1080x1080\.webp'
 578         },
 579         'expected_warnings': ['Retrying']
 580     }, {
 581         'url': 'https://www.tiktok.com/@meme',
 582         'playlist_mincount': 593,
 583         'info_dict': {
 584             'id': '79005827461758976',
 585             'title': 'meme',
 586             'thumbnail': r're:https://.+_1080x1080\.webp'
 587         },
 588         'expected_warnings': ['Retrying']
 589     }]
 590
 591     r'''  # TODO: Fix by adding _signature to api_url
 592     def _entries(self, webpage, user_id, username):
 593         secuid = self._search_regex(r'\"secUid\":\"(?P<secUid>[^\"]+)', webpage, username)
 594         verifyfp_cookie = self._get_cookies('https://www.tiktok.com').get('s_v_web_id')
 595         if not verifyfp_cookie:
 596             raise ExtractorError('Improper cookies (missing s_v_web_id).', expected=True)
 597         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='
 598         cursor = '0'
 599         for page in itertools.count():
 600             data_json = self._download_json(api_url + cursor, username, note='Downloading Page %d' % page)
 601             for video in data_json.get('itemList', []):
 602                 video_id = video['id']
 603                 video_url = f'https://www.tiktok.com/@{user_id}/video/{video_id}'
 604                 yield self._url_result(video_url, 'TikTok', video_id, str_or_none(video.get('desc')))
 605             if not data_json.get('hasMore'):
 606                 break
 607             cursor = data_json['cursor']
 608     '''
 609
 610     def _video_entries_api(self, webpage, user_id, username):
 611         query = {
 612             'user_id': user_id,
 613             'count': 21,
 614             'max_cursor': 0,
 615             'min_cursor': 0,
 616             'retry_type': 'no_retry',
 617             '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.
 618         }
 619
 620         for page in itertools.count(1):
 621             for retry in self.RetryManager():
 622                 try:
 623                     post_list = self._call_api(
 624                         'aweme/post', query, username, note=f'Downloading user video list page {page}',
 625                         errnote='Unable to download user video list')
 626                 except ExtractorError as e:
 627                     if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0:
 628                         retry.error = e
 629                         continue
 630                     raise
 631             yield from post_list.get('aweme_list', [])
 632             if not post_list.get('has_more'):
 633                 break
 634             query['max_cursor'] = post_list['max_cursor']
 635
 636     def _entries_api(self, user_id, videos):
 637         for video in videos:
 638             yield {
 639                 **self._parse_aweme_video_app(video),
 640                 'extractor_key': TikTokIE.ie_key(),
 641                 'extractor': 'TikTok',
 642                 'webpage_url': f'https://tiktok.com/@{user_id}/video/{video["aweme_id"]}',
 643             }
 644
 645     def _real_extract(self, url):
 646         user_name = self._match_id(url)
 647         webpage = self._download_webpage(url, user_name, headers={
 648             'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)'
 649         })
 650         user_id = self._html_search_regex(r'snssdk\d*://user/profile/(\d+)', webpage, 'user ID', default=None) or user_name
 651
 652         videos = LazyList(self._video_entries_api(webpage, user_id, user_name))
 653         thumbnail = traverse_obj(videos, (0, 'author', 'avatar_larger', 'url_list', 0))
 654
 655         return self.playlist_result(self._entries_api(user_id, videos), user_id, user_name, thumbnail=thumbnail)
 656
 657
 658 class TikTokBaseListIE(TikTokBaseIE):
 659     def _entries(self, list_id, display_id):
 660         query = {
 661             self._QUERY_NAME: list_id,
 662             'cursor': 0,
 663             'count': 20,
 664             'type': 5,
 665             'device_id': ''.join(random.choice(string.digits) for i in range(19))
 666         }
 667
 668         for page in itertools.count(1):
 669             for retry in self.RetryManager():
 670                 try:
 671                     post_list = self._call_api(
 672                         self._API_ENDPOINT, query, display_id, note=f'Downloading video list page {page}',
 673                         errnote='Unable to download video list')
 674                 except ExtractorError as e:
 675                     if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0:
 676                         retry.error = e
 677                         continue
 678                     raise
 679             for video in post_list.get('aweme_list', []):
 680                 yield {
 681                     **self._parse_aweme_video_app(video),
 682                     'extractor_key': TikTokIE.ie_key(),
 683                     'extractor': 'TikTok',
 684                     'webpage_url': f'https://tiktok.com/@_/video/{video["aweme_id"]}',
 685                 }
 686             if not post_list.get('has_more'):
 687                 break
 688             query['cursor'] = post_list['cursor']
 689
 690     def _real_extract(self, url):
 691         list_id = self._match_id(url)
 692         return self.playlist_result(self._entries(list_id, list_id), list_id)
 693
 694
 695 class TikTokSoundIE(TikTokBaseListIE):
 696     IE_NAME = 'tiktok:sound'
 697     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/music/[\w\.-]+-(?P<id>[\d]+)[/?#&]?'
 698     _WORKING = False
 699     _QUERY_NAME = 'music_id'
 700     _API_ENDPOINT = 'music/aweme'
 701     _TESTS = [{
 702         'url': 'https://www.tiktok.com/music/Build-a-Btch-6956990112127585029?lang=en',
 703         'playlist_mincount': 100,
 704         'info_dict': {
 705             'id': '6956990112127585029'
 706         },
 707         'expected_warnings': ['Retrying']
 708     }, {
 709         # Actual entries are less than listed video count
 710         'url': 'https://www.tiktok.com/music/jiefei-soap-remix-7036843036118469381',
 711         'playlist_mincount': 2182,
 712         'info_dict': {
 713             'id': '7036843036118469381'
 714         },
 715         'expected_warnings': ['Retrying']
 716     }]
 717
 718
 719 class TikTokEffectIE(TikTokBaseListIE):
 720     IE_NAME = 'tiktok:effect'
 721     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/sticker/[\w\.-]+-(?P<id>[\d]+)[/?#&]?'
 722     _WORKING = False
 723     _QUERY_NAME = 'sticker_id'
 724     _API_ENDPOINT = 'sticker/aweme'
 725     _TESTS = [{
 726         'url': 'https://www.tiktok.com/sticker/MATERIAL-GWOOORL-1258156',
 727         'playlist_mincount': 100,
 728         'info_dict': {
 729             'id': '1258156',
 730         },
 731         'expected_warnings': ['Retrying']
 732     }, {
 733         # Different entries between mobile and web, depending on region
 734         'url': 'https://www.tiktok.com/sticker/Elf-Friend-479565',
 735         'only_matching': True
 736     }]
 737
 738
 739 class TikTokTagIE(TikTokBaseListIE):
 740     IE_NAME = 'tiktok:tag'
 741     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/tag/(?P<id>[^/?#&]+)'
 742     _WORKING = False
 743     _QUERY_NAME = 'ch_id'
 744     _API_ENDPOINT = 'challenge/aweme'
 745     _TESTS = [{
 746         'url': 'https://tiktok.com/tag/hello2018',
 747         'playlist_mincount': 39,
 748         'info_dict': {
 749             'id': '46294678',
 750             'title': 'hello2018',
 751         },
 752         'expected_warnings': ['Retrying']
 753     }, {
 754         'url': 'https://tiktok.com/tag/fypシ?is_copy_url=0&is_from_webapp=v1',
 755         'only_matching': True
 756     }]
 757
 758     def _real_extract(self, url):
 759         display_id = self._match_id(url)
 760         webpage = self._download_webpage(url, display_id, headers={
 761             'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)'
 762         })
 763         tag_id = self._html_search_regex(r'snssdk\d*://challenge/detail/(\d+)', webpage, 'tag ID')
 764         return self.playlist_result(self._entries(tag_id, display_id), tag_id, display_id)
 765
 766
 767 class DouyinIE(TikTokIE):
 768     _VALID_URL = r'https?://(?:www\.)?douyin\.com/video/(?P<id>[0-9]+)'
 769     _TESTS = [{
 770         'url': 'https://www.douyin.com/video/6961737553342991651',
 771         'md5': '10523312c8b8100f353620ac9dc8f067',
 772         'info_dict': {
 773             'id': '6961737553342991651',
 774             'ext': 'mp4',
 775             'title': '#杨超越  小小水手带你去远航❤️',
 776             'uploader': '杨超越',
 777             'upload_date': '20210513',
 778             'timestamp': 1620905839,
 779             'uploader_id': '110403406559',
 780             'view_count': int,
 781             'like_count': int,
 782             'repost_count': int,
 783             'comment_count': int,
 784         }
 785     }, {
 786         'url': 'https://www.douyin.com/video/6982497745948921092',
 787         'md5': 'd78408c984b9b5102904cf6b6bc2d712',
 788         'info_dict': {
 789             'id': '6982497745948921092',
 790             'ext': 'mp4',
 791             'title': '这个夏日和小羊@杨超越 一起遇见白色幻想',
 792             'uploader': '杨超越工作室',
 793             'upload_date': '20210708',
 794             'timestamp': 1625739481,
 795             'uploader_id': '408654318141572',
 796             'view_count': int,
 797             'like_count': int,
 798             'repost_count': int,
 799             'comment_count': int,
 800         }
 801     }, {
 802         'url': 'https://www.douyin.com/video/6953975910773099811',
 803         'md5': '72e882e24f75064c218b76c8b713c185',
 804         'info_dict': {
 805             'id': '6953975910773099811',
 806             'ext': 'mp4',
 807             'title': '#一起看海  出现在你的夏日里',
 808             'uploader': '杨超越',
 809             'upload_date': '20210422',
 810             'timestamp': 1619098692,
 811             'uploader_id': '110403406559',
 812             'view_count': int,
 813             'like_count': int,
 814             'repost_count': int,
 815             'comment_count': int,
 816         }
 817     }, {
 818         'url': 'https://www.douyin.com/video/6950251282489675042',
 819         'md5': 'b4db86aec367ef810ddd38b1737d2fed',
 820         'info_dict': {
 821             'id': '6950251282489675042',
 822             'ext': 'mp4',
 823             'title': '哈哈哈，成功了哈哈哈哈哈哈',
 824             'uploader': '杨超越',
 825             'upload_date': '20210412',
 826             'timestamp': 1618231483,
 827             'uploader_id': '110403406559',
 828             'view_count': int,
 829             'like_count': int,
 830             'repost_count': int,
 831             'comment_count': int,
 832         }
 833     }, {
 834         'url': 'https://www.douyin.com/video/6963263655114722595',
 835         'md5': '1abe1c477d05ee62efb40bf2329957cf',
 836         'info_dict': {
 837             'id': '6963263655114722595',
 838             'ext': 'mp4',
 839             'title': '#哪个爱豆的105度最甜 换个角度看看我哈哈',
 840             'uploader': '杨超越',
 841             'upload_date': '20210517',
 842             'timestamp': 1621261163,
 843             'uploader_id': '110403406559',
 844             'view_count': int,
 845             'like_count': int,
 846             'repost_count': int,
 847             'comment_count': int,
 848         }
 849     }]
 850     _APP_VERSIONS = [('9.6.0', '960')]
 851     _APP_NAME = 'aweme'
 852     _AID = 1128
 853     _API_HOSTNAME = 'aweme.snssdk.com'
 854     _UPLOADER_URL_FORMAT = 'https://www.douyin.com/user/%s'
 855     _WEBPAGE_HOST = 'https://www.douyin.com/'
 856
 857     def _real_extract(self, url):
 858         video_id = self._match_id(url)
 859
 860         try:
 861             return self._extract_aweme_app(video_id)
 862         except ExtractorError as e:
 863             self.report_warning(f'{e}; trying with webpage')
 864
 865         webpage = self._download_webpage(url, video_id)
 866         render_data_json = self._search_regex(
 867             r'<script [^>]*\bid=[\'"]RENDER_DATA[\'"][^>]*>(%7B.+%7D)</script>',
 868             webpage, 'render data', default=None)
 869         if not render_data_json:
 870             # TODO: Run verification challenge code to generate signature cookies
 871             raise ExtractorError('Fresh cookies (not necessarily logged in) are needed')
 872
 873         render_data = self._parse_json(
 874             render_data_json, video_id, transform_source=compat_urllib_parse_unquote)
 875         return self._parse_aweme_video_web(get_first(render_data, ('aweme', 'detail')), url)
 876
 877
 878 class TikTokVMIE(InfoExtractor):
 879     _VALID_URL = r'https?://(?:vm|vt)\.tiktok\.com/(?P<id>\w+)'
 880     IE_NAME = 'vm.tiktok'
 881
 882     _TESTS = [{
 883         'url': 'https://vm.tiktok.com/ZSe4FqkKd',
 884         'info_dict': {
 885             'id': '7023491746608712966',
 886             'ext': 'mp4',
 887             'title': 'md5:5607564db90271abbbf8294cca77eddd',
 888             'description': 'md5:5607564db90271abbbf8294cca77eddd',
 889             'duration': 11,
 890             'upload_date': '20211026',
 891             'uploader_id': '7007385080558846981',
 892             'creator': 'Memes',
 893             'artist': 'Memes',
 894             'track': 'original sound',
 895             'uploader': 'susmandem',
 896             'timestamp': 1635284105,
 897             'thumbnail': r're:https://.+\.webp.*',
 898             'like_count': int,
 899             'view_count': int,
 900             'comment_count': int,
 901             'repost_count': int,
 902             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAXcNoOEOxVyBzuII_E--T0MeCrLP0ay1Sm6x_n3dluiWEoWZD0VlQOytwad4W0i0n',
 903         }
 904     }, {
 905         'url': 'https://vt.tiktok.com/ZSe4FqkKd',
 906         'only_matching': True,
 907     }]
 908
 909     def _real_extract(self, url):
 910         new_url = self._request_webpage(
 911             HEADRequest(url), self._match_id(url), headers={'User-Agent': 'facebookexternalhit/1.1'}).geturl()
 912         if self.suitable(new_url):  # Prevent infinite loop in case redirect fails
 913             raise UnsupportedError(new_url)
 914         return self.url_result(new_url)