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 = [('20.9.3', '293'), ('20.4.3', '243'), ('20.2.1', '221'), ('20.1.2', '212'), ('20.0.4', '204')]
  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     _session_initialized = False
  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 _real_initialize(self):
  47         if self._session_initialized:
  48             return
  49         self._request_webpage(HEADRequest('https://www.tiktok.com'), None, note='Setting up session', fatal=False)
  50         TikTokBaseIE._session_initialized = True
  51
  52     def _call_api_impl(self, ep, query, manifest_app_version, video_id, fatal=True,
  53                        note='Downloading API JSON', errnote='Unable to download API page'):
  54         self._set_cookie(self._API_HOSTNAME, 'odin_tt', ''.join(random.choice('0123456789abcdef') for _ in range(160)))
  55         webpage_cookies = self._get_cookies(self._WEBPAGE_HOST)
  56         if webpage_cookies.get('sid_tt'):
  57             self._set_cookie(self._API_HOSTNAME, 'sid_tt', webpage_cookies['sid_tt'].value)
  58         return self._download_json(
  59             'https://%s/aweme/v1/%s/' % (self._API_HOSTNAME, ep), video_id=video_id,
  60             fatal=fatal, note=note, errnote=errnote, headers={
  61                 '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)',
  62                 'Accept': 'application/json',
  63             }, query=query)
  64
  65     def _build_api_query(self, query, app_version, manifest_app_version):
  66         return {
  67             **query,
  68             'version_name': app_version,
  69             'version_code': manifest_app_version,
  70             'build_number': app_version,
  71             'manifest_version_code': manifest_app_version,
  72             'update_version_code': manifest_app_version,
  73             'openudid': ''.join(random.choice('0123456789abcdef') for _ in range(16)),
  74             'uuid': ''.join([random.choice(string.digits) for _ in range(16)]),
  75             '_rticket': int(time.time() * 1000),
  76             'ts': int(time.time()),
  77             'device_brand': 'Google',
  78             'device_type': 'Pixel 4',
  79             'device_platform': 'android',
  80             'resolution': '1080*1920',
  81             'dpi': 420,
  82             'os_version': '10',
  83             'os_api': '29',
  84             'carrier_region': 'US',
  85             'sys_region': 'US',
  86             'region': 'US',
  87             'app_name': self._APP_NAME,
  88             'app_language': 'en',
  89             'language': 'en',
  90             'timezone_name': 'America/New_York',
  91             'timezone_offset': '-14400',
  92             'channel': 'googleplay',
  93             'ac': 'wifi',
  94             'mcc_mnc': '310260',
  95             'is_my_cn': 0,
  96             'aid': self._AID,
  97             'ssmix': 'a',
  98             'as': 'a1qwert123',
  99             'cp': 'cbfhckdckkde1',
 100         }
 101
 102     def _call_api(self, ep, query, video_id, fatal=True,
 103                   note='Downloading API JSON', errnote='Unable to download API page'):
 104         if not self._WORKING_APP_VERSION:
 105             app_version = self._configuration_arg('app_version', [''], ie_key=TikTokIE.ie_key())[0]
 106             manifest_app_version = self._configuration_arg('manifest_app_version', [''], ie_key=TikTokIE.ie_key())[0]
 107             if app_version and manifest_app_version:
 108                 self._WORKING_APP_VERSION = (app_version, manifest_app_version)
 109                 self.write_debug('Imported app version combo from extractor arguments')
 110             elif app_version or manifest_app_version:
 111                 self.report_warning('Only one of the two required version params are passed as extractor arguments', only_once=True)
 112
 113         if self._WORKING_APP_VERSION:
 114             app_version, manifest_app_version = self._WORKING_APP_VERSION
 115             real_query = self._build_api_query(query, app_version, manifest_app_version)
 116             return self._call_api_impl(ep, real_query, manifest_app_version, video_id, fatal, note, errnote)
 117
 118         for count, (app_version, manifest_app_version) in enumerate(self._APP_VERSIONS, start=1):
 119             real_query = self._build_api_query(query, app_version, manifest_app_version)
 120             try:
 121                 res = self._call_api_impl(ep, real_query, manifest_app_version, video_id, fatal, note, errnote)
 122                 self._WORKING_APP_VERSION = (app_version, manifest_app_version)
 123                 return res
 124             except ExtractorError as e:
 125                 if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0:
 126                     if count == len(self._APP_VERSIONS):
 127                         if fatal:
 128                             raise e
 129                         else:
 130                             self.report_warning(str(e.cause or e.msg))
 131                             return
 132                     self.report_warning('%s. Retrying... (attempt %s of %s)' % (str(e.cause or e.msg), count, len(self._APP_VERSIONS)))
 133                     continue
 134                 raise e
 135
 136     def _get_subtitles(self, aweme_detail, aweme_id):
 137         # TODO: Extract text positioning info
 138         subtitles = {}
 139         captions_info = traverse_obj(
 140             aweme_detail, ('interaction_stickers', ..., 'auto_video_caption_info', 'auto_captions', ...), expected_type=dict, default=[])
 141         for caption in captions_info:
 142             caption_url = traverse_obj(caption, ('url', 'url_list', ...), expected_type=url_or_none, get_all=False)
 143             if not caption_url:
 144                 continue
 145             caption_json = self._download_json(
 146                 caption_url, aweme_id, note='Downloading captions', errnote='Unable to download captions', fatal=False)
 147             if not caption_json:
 148                 continue
 149             subtitles.setdefault(caption.get('language', 'en'), []).append({
 150                 'ext': 'srt',
 151                 'data': '\n\n'.join(
 152                     f'{i + 1}\n{srt_subtitles_timecode(line["start_time"] / 1000)} --> {srt_subtitles_timecode(line["end_time"] / 1000)}\n{line["text"]}'
 153                     for i, line in enumerate(caption_json['utterances']) if line.get('text'))
 154             })
 155         return subtitles
 156
 157     def _parse_aweme_video_app(self, aweme_detail):
 158         aweme_id = aweme_detail['aweme_id']
 159         video_info = aweme_detail['video']
 160
 161         def parse_url_key(url_key):
 162             format_id, codec, res, bitrate = self._search_regex(
 163                 r'v[^_]+_(?P<id>(?P<codec>[^_]+)_(?P<res>\d+p)_(?P<bitrate>\d+))', url_key,
 164                 'url key', default=(None, None, None, None), group=('id', 'codec', 'res', 'bitrate'))
 165             if not format_id:
 166                 return {}, None
 167             return {
 168                 'format_id': format_id,
 169                 'vcodec': 'h265' if codec == 'bytevc1' else codec,
 170                 'tbr': int_or_none(bitrate, scale=1000) or None,
 171                 'quality': qualities(self.QUALITIES)(res),
 172             }, res
 173
 174         known_resolutions = {}
 175
 176         def extract_addr(addr, add_meta={}):
 177             parsed_meta, res = parse_url_key(addr.get('url_key', ''))
 178             if res:
 179                 known_resolutions.setdefault(res, {}).setdefault('height', add_meta.get('height'))
 180                 known_resolutions[res].setdefault('width', add_meta.get('width'))
 181                 parsed_meta.update(known_resolutions.get(res, {}))
 182                 add_meta.setdefault('height', int_or_none(res[:-1]))
 183             return [{
 184                 'url': url,
 185                 'filesize': int_or_none(addr.get('data_size')),
 186                 'ext': 'mp4',
 187                 'acodec': 'aac',
 188                 'source_preference': -2 if 'aweme/v1' in url else -1,  # Downloads from API might get blocked
 189                 **add_meta, **parsed_meta,
 190                 'format_note': join_nonempty(
 191                     add_meta.get('format_note'), '(API)' if 'aweme/v1' in url else None, delim=' ')
 192             } for url in addr.get('url_list') or []]
 193
 194         # Hack: Add direct video links first to prioritize them when removing duplicate formats
 195         formats = []
 196         if video_info.get('play_addr'):
 197             formats.extend(extract_addr(video_info['play_addr'], {
 198                 'format_id': 'play_addr',
 199                 'format_note': 'Direct video',
 200                 'vcodec': 'h265' if traverse_obj(
 201                     video_info, 'is_bytevc1', 'is_h265') else 'h264',  # TODO: Check for "direct iOS" videos, like https://www.tiktok.com/@cookierun_dev/video/7039716639834656002
 202                 'width': video_info.get('width'),
 203                 'height': video_info.get('height'),
 204             }))
 205         if video_info.get('download_addr'):
 206             formats.extend(extract_addr(video_info['download_addr'], {
 207                 'format_id': 'download_addr',
 208                 'format_note': 'Download video%s' % (', watermarked' if video_info.get('has_watermark') else ''),
 209                 'vcodec': 'h264',
 210                 'width': video_info.get('width'),
 211                 'height': video_info.get('height'),
 212                 'preference': -2 if video_info.get('has_watermark') else -1,
 213             }))
 214         if video_info.get('play_addr_h264'):
 215             formats.extend(extract_addr(video_info['play_addr_h264'], {
 216                 'format_id': 'play_addr_h264',
 217                 'format_note': 'Direct video',
 218                 'vcodec': 'h264',
 219             }))
 220         if video_info.get('play_addr_bytevc1'):
 221             formats.extend(extract_addr(video_info['play_addr_bytevc1'], {
 222                 'format_id': 'play_addr_bytevc1',
 223                 'format_note': 'Direct video',
 224                 'vcodec': 'h265',
 225             }))
 226
 227         for bitrate in video_info.get('bit_rate', []):
 228             if bitrate.get('play_addr'):
 229                 formats.extend(extract_addr(bitrate['play_addr'], {
 230                     'format_id': bitrate.get('gear_name'),
 231                     'format_note': 'Playback video',
 232                     'tbr': try_get(bitrate, lambda x: x['bit_rate'] / 1000),
 233                     'vcodec': 'h265' if traverse_obj(
 234                         bitrate, 'is_bytevc1', 'is_h265') else 'h264',
 235                     'fps': bitrate.get('FPS'),
 236                 }))
 237
 238         self._remove_duplicate_formats(formats)
 239         auth_cookie = self._get_cookies(self._WEBPAGE_HOST).get('sid_tt')
 240         if auth_cookie:
 241             for f in formats:
 242                 self._set_cookie(compat_urllib_parse_urlparse(f['url']).hostname, 'sid_tt', auth_cookie.value)
 243         self._sort_formats(formats, ('quality', 'codec', 'size', 'br'))
 244
 245         thumbnails = []
 246         for cover_id in ('cover', 'ai_dynamic_cover', 'animated_cover', 'ai_dynamic_cover_bak',
 247                          'origin_cover', 'dynamic_cover'):
 248             cover = video_info.get(cover_id)
 249             if cover:
 250                 for cover_url in cover['url_list']:
 251                     thumbnails.append({
 252                         'id': cover_id,
 253                         'url': cover_url,
 254                     })
 255
 256         stats_info = aweme_detail.get('statistics', {})
 257         author_info = aweme_detail.get('author', {})
 258         music_info = aweme_detail.get('music', {})
 259         user_url = self._UPLOADER_URL_FORMAT % (traverse_obj(author_info,
 260                                                              'sec_uid', 'id', 'uid', 'unique_id',
 261                                                              expected_type=str_or_none, get_all=False))
 262         labels = traverse_obj(aweme_detail, ('hybrid_label', ..., 'text'), expected_type=str, default=[])
 263
 264         contained_music_track = traverse_obj(
 265             music_info, ('matched_song', 'title'), ('matched_pgc_sound', 'title'), expected_type=str)
 266         contained_music_author = traverse_obj(
 267             music_info, ('matched_song', 'author'), ('matched_pgc_sound', 'author'), 'author', expected_type=str)
 268
 269         is_generic_og_trackname = music_info.get('is_original_sound') and music_info.get('title') == 'original sound - %s' % music_info.get('owner_handle')
 270         if is_generic_og_trackname:
 271             music_track, music_author = contained_music_track or 'original sound', contained_music_author
 272         else:
 273             music_track, music_author = music_info.get('title'), music_info.get('author')
 274
 275         return {
 276             'id': aweme_id,
 277             'extractor_key': TikTokIE.ie_key(),
 278             'extractor': TikTokIE.IE_NAME,
 279             'webpage_url': self._create_url(author_info.get('uid'), aweme_id),
 280             'title': aweme_detail.get('desc'),
 281             'description': aweme_detail.get('desc'),
 282             'view_count': int_or_none(stats_info.get('play_count')),
 283             'like_count': int_or_none(stats_info.get('digg_count')),
 284             'repost_count': int_or_none(stats_info.get('share_count')),
 285             'comment_count': int_or_none(stats_info.get('comment_count')),
 286             'uploader': str_or_none(author_info.get('unique_id')),
 287             'creator': str_or_none(author_info.get('nickname')),
 288             'uploader_id': str_or_none(author_info.get('uid')),
 289             'uploader_url': user_url,
 290             'track': music_track,
 291             'album': str_or_none(music_info.get('album')) or None,
 292             'artist': music_author,
 293             'timestamp': int_or_none(aweme_detail.get('create_time')),
 294             'formats': formats,
 295             'subtitles': self.extract_subtitles(aweme_detail, aweme_id),
 296             'thumbnails': thumbnails,
 297             'duration': int_or_none(traverse_obj(video_info, 'duration', ('download_addr', 'duration')), scale=1000),
 298             'availability': self._availability(
 299                 is_private='Private' in labels,
 300                 needs_subscription='Friends only' in labels,
 301                 is_unlisted='Followers only' in labels)
 302         }
 303
 304     def _parse_aweme_video_web(self, aweme_detail, webpage_url):
 305         video_info = aweme_detail['video']
 306         author_info = traverse_obj(aweme_detail, 'authorInfo', 'author', expected_type=dict, default={})
 307         music_info = aweme_detail.get('music') or {}
 308         stats_info = aweme_detail.get('stats') or {}
 309         user_url = self._UPLOADER_URL_FORMAT % (traverse_obj(author_info,
 310                                                              'secUid', 'id', 'uid', 'uniqueId',
 311                                                              expected_type=str_or_none, get_all=False)
 312                                                 or aweme_detail.get('authorSecId'))
 313
 314         formats = []
 315         play_url = video_info.get('playAddr')
 316         width = video_info.get('width')
 317         height = video_info.get('height')
 318         if isinstance(play_url, str):
 319             formats = [{
 320                 'url': self._proto_relative_url(play_url),
 321                 'ext': 'mp4',
 322                 'width': width,
 323                 'height': height,
 324             }]
 325         elif isinstance(play_url, list):
 326             formats = [{
 327                 'url': self._proto_relative_url(url),
 328                 'ext': 'mp4',
 329                 'width': width,
 330                 'height': height,
 331             } for url in traverse_obj(play_url, (..., 'src'), expected_type=url_or_none, default=[]) if url]
 332
 333         download_url = url_or_none(video_info.get('downloadAddr')) or traverse_obj(video_info, ('download', 'url'), expected_type=url_or_none)
 334         if download_url:
 335             formats.append({
 336                 'format_id': 'download',
 337                 'url': self._proto_relative_url(download_url),
 338                 'ext': 'mp4',
 339                 'width': width,
 340                 'height': height,
 341             })
 342         self._remove_duplicate_formats(formats)
 343         self._sort_formats(formats)
 344
 345         thumbnails = []
 346         for thumbnail_name in ('thumbnail', 'cover', 'dynamicCover', 'originCover'):
 347             if aweme_detail.get(thumbnail_name):
 348                 thumbnails = [{
 349                     'url': self._proto_relative_url(aweme_detail[thumbnail_name]),
 350                     'width': width,
 351                     'height': height
 352                 }]
 353
 354         return {
 355             'id': traverse_obj(aweme_detail, 'id', 'awemeId', expected_type=str_or_none),
 356             'title': aweme_detail.get('desc'),
 357             'duration': try_get(aweme_detail, lambda x: x['video']['duration'], int),
 358             'view_count': int_or_none(stats_info.get('playCount')),
 359             'like_count': int_or_none(stats_info.get('diggCount')),
 360             'repost_count': int_or_none(stats_info.get('shareCount')),
 361             'comment_count': int_or_none(stats_info.get('commentCount')),
 362             'timestamp': int_or_none(aweme_detail.get('createTime')),
 363             'creator': str_or_none(author_info.get('nickname')),
 364             'uploader': str_or_none(author_info.get('uniqueId') or aweme_detail.get('author')),
 365             'uploader_id': str_or_none(author_info.get('id') or aweme_detail.get('authorId')),
 366             'uploader_url': user_url,
 367             'track': str_or_none(music_info.get('title')),
 368             'album': str_or_none(music_info.get('album')) or None,
 369             'artist': str_or_none(music_info.get('authorName')),
 370             'formats': formats,
 371             'thumbnails': thumbnails,
 372             'description': str_or_none(aweme_detail.get('desc')),
 373             'http_headers': {
 374                 'Referer': webpage_url
 375             }
 376         }
 377
 378
 379 class TikTokIE(TikTokBaseIE):
 380     _VALID_URL = r'https?://www\.tiktok\.com/(?:embed|@(?P<user_id>[\w\.-]+)/video)/(?P<id>\d+)'
 381     _EMBED_REGEX = [rf'<(?:script|iframe)[^>]+\bsrc=(["\'])(?P<url>{_VALID_URL})']
 382
 383     _TESTS = [{
 384         'url': 'https://www.tiktok.com/@leenabhushan/video/6748451240264420610',
 385         'md5': '736bb7a466c6f0a6afeb597da1e6f5b7',
 386         'info_dict': {
 387             'id': '6748451240264420610',
 388             'ext': 'mp4',
 389             'title': '#jassmanak #lehanga #leenabhushan',
 390             'description': '#jassmanak #lehanga #leenabhushan',
 391             'duration': 13,
 392             'height': 1024,
 393             'width': 576,
 394             'uploader': 'leenabhushan',
 395             'uploader_id': '6691488002098119685',
 396             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA_Eb4t1vodM1IuTy_cvp9CY22RAb59xqrO0Xtz9CYQJvgXaDvZxYnZYRzDWhhgJmy',
 397             'creator': 'facestoriesbyleenabh',
 398             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 399             'upload_date': '20191016',
 400             'timestamp': 1571246252,
 401             'view_count': int,
 402             'like_count': int,
 403             'repost_count': int,
 404             'comment_count': int,
 405             'artist': 'Ysrbeats',
 406             'album': 'Lehanga',
 407             'track': 'Lehanga',
 408         }
 409     }, {
 410         'url': 'https://www.tiktok.com/@patroxofficial/video/6742501081818877190?langCountry=en',
 411         'md5': '6f3cf8cdd9b28cb8363fe0a9a160695b',
 412         'info_dict': {
 413             'id': '6742501081818877190',
 414             'ext': 'mp4',
 415             'title': 'md5:5e2a23877420bb85ce6521dbee39ba94',
 416             'description': 'md5:5e2a23877420bb85ce6521dbee39ba94',
 417             'duration': 27,
 418             'height': 960,
 419             'width': 540,
 420             'uploader': 'patrox',
 421             'uploader_id': '18702747',
 422             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAiFnldaILebi5heDoVU6bn4jBWWycX6-9U3xuNPqZ8Ws',
 423             'creator': 'patroX',
 424             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 425             'upload_date': '20190930',
 426             'timestamp': 1569860870,
 427             'view_count': int,
 428             'like_count': int,
 429             'repost_count': int,
 430             'comment_count': int,
 431             'artist': 'Evan Todd, Jessica Keenan Wynn, Alice Lee, Barrett Wilbert Weed & Jon Eidson',
 432             'track': 'Big Fun',
 433         }
 434     }, {
 435         # Banned audio, only available on the app
 436         'url': 'https://www.tiktok.com/@barudakhb_/video/6984138651336838402',
 437         'info_dict': {
 438             'id': '6984138651336838402',
 439             'ext': 'mp4',
 440             'title': 'Balas @yolaaftwsr hayu yu ? #SquadRandom_ 🔥',
 441             'description': 'Balas @yolaaftwsr hayu yu ? #SquadRandom_ 🔥',
 442             'uploader': 'barudakhb_',
 443             'creator': 'md5:29f238c49bc0c176cb3cef1a9cea9fa6',
 444             'uploader_id': '6974687867511718913',
 445             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAbhBwQC-R1iKoix6jDFsF-vBdfx2ABoDjaZrM9fX6arU3w71q3cOWgWuTXn1soZ7d',
 446             'track': 'Boka Dance',
 447             'artist': 'md5:29f238c49bc0c176cb3cef1a9cea9fa6',
 448             'timestamp': 1626121503,
 449             'duration': 18,
 450             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 451             'upload_date': '20210712',
 452             'view_count': int,
 453             'like_count': int,
 454             'repost_count': int,
 455             'comment_count': int,
 456         }
 457     }, {
 458         # Sponsored video, only available with feed workaround
 459         'url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_/video/7042692929109986561',
 460         'info_dict': {
 461             'id': '7042692929109986561',
 462             'ext': 'mp4',
 463             'title': 'Slap and Run!',
 464             'description': 'Slap and Run!',
 465             'uploader': 'user440922249',
 466             'creator': 'Slap And Run',
 467             'uploader_id': '7036055384943690754',
 468             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_',
 469             'track': 'Promoted Music',
 470             'timestamp': 1639754738,
 471             'duration': 30,
 472             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 473             'upload_date': '20211217',
 474             'view_count': int,
 475             'like_count': int,
 476             'repost_count': int,
 477             'comment_count': int,
 478         },
 479         'expected_warnings': ['trying with webpage', 'Unable to find video in feed']
 480     }, {
 481         # Video without title and description
 482         'url': 'https://www.tiktok.com/@pokemonlife22/video/7059698374567611694',
 483         'info_dict': {
 484             'id': '7059698374567611694',
 485             'ext': 'mp4',
 486             'title': 'TikTok video #7059698374567611694',
 487             'description': '',
 488             'uploader': 'pokemonlife22',
 489             'creator': 'Pokemon',
 490             'uploader_id': '6820838815978423302',
 491             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA0tF1nBwQVVMyrGu3CqttkNgM68Do1OXUFuCY0CRQk8fEtSVDj89HqoqvbSTmUP2W',
 492             'track': 'original sound',
 493             'timestamp': 1643714123,
 494             'duration': 6,
 495             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 496             'upload_date': '20220201',
 497             'artist': 'Pokemon',
 498             'view_count': int,
 499             'like_count': int,
 500             'repost_count': int,
 501             'comment_count': int,
 502         },
 503     }, {
 504         # hydration JSON is sent in a <script> element
 505         'url': 'https://www.tiktok.com/@denidil6/video/7065799023130643713',
 506         'info_dict': {
 507             'id': '7065799023130643713',
 508             'ext': 'mp4',
 509             'title': '#denidil#денидил',
 510             'description': '#denidil#денидил',
 511             'uploader': 'denidil6',
 512             'uploader_id': '7046664115636405250',
 513             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAsvMSzFdQ4ikl3uR2TEJwMBbB2yZh2Zxwhx-WCo3rbDpAharE3GQCrFuJArI3C8QJ',
 514             'artist': 'Holocron Music',
 515             'album': 'Wolf Sounds (1 Hour) Enjoy the Company of the Animal That Is the Majestic King of the Night',
 516             'track': 'Wolf Sounds (1 Hour) Enjoy the Company of the Animal That Is the Majestic King of the Night',
 517             'timestamp': 1645134536,
 518             'duration': 26,
 519             'upload_date': '20220217',
 520             'view_count': int,
 521             'like_count': int,
 522             'repost_count': int,
 523             'comment_count': int,
 524         },
 525         'expected_warnings': ['trying feed workaround', 'Unable to find video in feed']
 526     }, {
 527         # Auto-captions available
 528         'url': 'https://www.tiktok.com/@hankgreen1/video/7047596209028074758',
 529         'only_matching': True
 530     }]
 531
 532     def _extract_aweme_app(self, aweme_id):
 533         try:
 534             aweme_detail = self._call_api('aweme/detail', {'aweme_id': aweme_id}, aweme_id,
 535                                           note='Downloading video details', errnote='Unable to download video details').get('aweme_detail')
 536             if not aweme_detail:
 537                 raise ExtractorError('Video not available', video_id=aweme_id)
 538         except ExtractorError as e:
 539             self.report_warning(f'{e.orig_msg}; trying feed workaround')
 540             feed_list = self._call_api('feed', {'aweme_id': aweme_id}, aweme_id,
 541                                        note='Downloading video feed', errnote='Unable to download video feed').get('aweme_list') or []
 542             aweme_detail = next((aweme for aweme in feed_list if str(aweme.get('aweme_id')) == aweme_id), None)
 543             if not aweme_detail:
 544                 raise ExtractorError('Unable to find video in feed', video_id=aweme_id)
 545         return self._parse_aweme_video_app(aweme_detail)
 546
 547     def _real_extract(self, url):
 548         video_id, user_id = self._match_valid_url(url).group('id', 'user_id')
 549         try:
 550             return self._extract_aweme_app(video_id)
 551         except ExtractorError as e:
 552             self.report_warning(f'{e}; trying with webpage')
 553
 554         url = self._create_url(user_id, video_id)
 555         webpage = self._download_webpage(url, video_id, headers={'User-Agent': 'User-Agent:Mozilla/5.0'})
 556         next_data = self._search_nextjs_data(webpage, video_id, default='{}')
 557         if next_data:
 558             status = traverse_obj(next_data, ('props', 'pageProps', 'statusCode'), expected_type=int) or 0
 559             video_data = traverse_obj(next_data, ('props', 'pageProps', 'itemInfo', 'itemStruct'), expected_type=dict)
 560         else:
 561             sigi_data = self._get_sigi_state(webpage, video_id)
 562             status = traverse_obj(sigi_data, ('VideoPage', 'statusCode'), expected_type=int) or 0
 563             video_data = traverse_obj(sigi_data, ('ItemModule', video_id), expected_type=dict)
 564
 565         if status == 0:
 566             return self._parse_aweme_video_web(video_data, url)
 567         elif status == 10216:
 568             raise ExtractorError('This video is private', expected=True)
 569         raise ExtractorError('Video not available', video_id=video_id)
 570
 571
 572 class TikTokUserIE(TikTokBaseIE):
 573     IE_NAME = 'tiktok:user'
 574     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/@(?P<id>[\w\.-]+)/?(?:$|[#?])'
 575     _TESTS = [{
 576         'url': 'https://tiktok.com/@corgibobaa?lang=en',
 577         'playlist_mincount': 45,
 578         'info_dict': {
 579             'id': '6935371178089399301',
 580             'title': 'corgibobaa',
 581             'thumbnail': r're:https://.+_1080x1080\.webp'
 582         },
 583         'expected_warnings': ['Retrying']
 584     }, {
 585         'url': 'https://www.tiktok.com/@6820838815978423302',
 586         'playlist_mincount': 5,
 587         'info_dict': {
 588             'id': '6820838815978423302',
 589             'title': '6820838815978423302',
 590             'thumbnail': r're:https://.+_1080x1080\.webp'
 591         },
 592         'expected_warnings': ['Retrying']
 593     }, {
 594         'url': 'https://www.tiktok.com/@meme',
 595         'playlist_mincount': 593,
 596         'info_dict': {
 597             'id': '79005827461758976',
 598             'title': 'meme',
 599             'thumbnail': r're:https://.+_1080x1080\.webp'
 600         },
 601         'expected_warnings': ['Retrying']
 602     }]
 603
 604     r'''  # TODO: Fix by adding _signature to api_url
 605     def _entries(self, webpage, user_id, username):
 606         secuid = self._search_regex(r'\"secUid\":\"(?P<secUid>[^\"]+)', webpage, username)
 607         verifyfp_cookie = self._get_cookies('https://www.tiktok.com').get('s_v_web_id')
 608         if not verifyfp_cookie:
 609             raise ExtractorError('Improper cookies (missing s_v_web_id).', expected=True)
 610         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='
 611         cursor = '0'
 612         for page in itertools.count():
 613             data_json = self._download_json(api_url + cursor, username, note='Downloading Page %d' % page)
 614             for video in data_json.get('itemList', []):
 615                 video_id = video['id']
 616                 video_url = f'https://www.tiktok.com/@{user_id}/video/{video_id}'
 617                 yield self._url_result(video_url, 'TikTok', video_id, str_or_none(video.get('desc')))
 618             if not data_json.get('hasMore'):
 619                 break
 620             cursor = data_json['cursor']
 621     '''
 622
 623     def _video_entries_api(self, webpage, user_id, username):
 624         query = {
 625             'user_id': user_id,
 626             'count': 21,
 627             'max_cursor': 0,
 628             'min_cursor': 0,
 629             'retry_type': 'no_retry',
 630             '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.
 631         }
 632
 633         for page in itertools.count(1):
 634             for retry in self.RetryManager():
 635                 try:
 636                     post_list = self._call_api(
 637                         'aweme/post', query, username, note=f'Downloading user video list page {page}',
 638                         errnote='Unable to download user video list')
 639                 except ExtractorError as e:
 640                     if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0:
 641                         retry.error = e
 642                         continue
 643                     raise
 644             yield from post_list.get('aweme_list', [])
 645             if not post_list.get('has_more'):
 646                 break
 647             query['max_cursor'] = post_list['max_cursor']
 648
 649     def _entries_api(self, user_id, videos):
 650         for video in videos:
 651             yield {
 652                 **self._parse_aweme_video_app(video),
 653                 'extractor_key': TikTokIE.ie_key(),
 654                 'extractor': 'TikTok',
 655                 'webpage_url': f'https://tiktok.com/@{user_id}/video/{video["aweme_id"]}',
 656             }
 657
 658     def _real_extract(self, url):
 659         user_name = self._match_id(url)
 660         webpage = self._download_webpage(url, user_name, headers={
 661             'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)'
 662         })
 663         user_id = self._html_search_regex(r'snssdk\d*://user/profile/(\d+)', webpage, 'user ID', default=None) or user_name
 664
 665         videos = LazyList(self._video_entries_api(webpage, user_id, user_name))
 666         thumbnail = traverse_obj(videos, (0, 'author', 'avatar_larger', 'url_list', 0))
 667
 668         return self.playlist_result(self._entries_api(user_id, videos), user_id, user_name, thumbnail=thumbnail)
 669
 670
 671 class TikTokBaseListIE(TikTokBaseIE):
 672     def _entries(self, list_id, display_id):
 673         query = {
 674             self._QUERY_NAME: list_id,
 675             'cursor': 0,
 676             'count': 20,
 677             'type': 5,
 678             'device_id': ''.join(random.choice(string.digits) for i in range(19))
 679         }
 680
 681         for page in itertools.count(1):
 682             for retry in self.RetryManager():
 683                 try:
 684                     post_list = self._call_api(
 685                         self._API_ENDPOINT, query, display_id, note=f'Downloading video list page {page}',
 686                         errnote='Unable to download video list')
 687                 except ExtractorError as e:
 688                     if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0:
 689                         retry.error = e
 690                         continue
 691                     raise
 692             for video in post_list.get('aweme_list', []):
 693                 yield {
 694                     **self._parse_aweme_video_app(video),
 695                     'extractor_key': TikTokIE.ie_key(),
 696                     'extractor': 'TikTok',
 697                     'webpage_url': f'https://tiktok.com/@_/video/{video["aweme_id"]}',
 698                 }
 699             if not post_list.get('has_more'):
 700                 break
 701             query['cursor'] = post_list['cursor']
 702
 703     def _real_extract(self, url):
 704         list_id = self._match_id(url)
 705         return self.playlist_result(self._entries(list_id, list_id), list_id)
 706
 707
 708 class TikTokSoundIE(TikTokBaseListIE):
 709     IE_NAME = 'tiktok:sound'
 710     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/music/[\w\.-]+-(?P<id>[\d]+)[/?#&]?'
 711     _QUERY_NAME = 'music_id'
 712     _API_ENDPOINT = 'music/aweme'
 713     _TESTS = [{
 714         'url': 'https://www.tiktok.com/music/Build-a-Btch-6956990112127585029?lang=en',
 715         'playlist_mincount': 100,
 716         'info_dict': {
 717             'id': '6956990112127585029'
 718         },
 719         'expected_warnings': ['Retrying']
 720     }, {
 721         # Actual entries are less than listed video count
 722         'url': 'https://www.tiktok.com/music/jiefei-soap-remix-7036843036118469381',
 723         'playlist_mincount': 2182,
 724         'info_dict': {
 725             'id': '7036843036118469381'
 726         },
 727         'expected_warnings': ['Retrying']
 728     }]
 729
 730
 731 class TikTokEffectIE(TikTokBaseListIE):
 732     IE_NAME = 'tiktok:effect'
 733     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/sticker/[\w\.-]+-(?P<id>[\d]+)[/?#&]?'
 734     _QUERY_NAME = 'sticker_id'
 735     _API_ENDPOINT = 'sticker/aweme'
 736     _TESTS = [{
 737         'url': 'https://www.tiktok.com/sticker/MATERIAL-GWOOORL-1258156',
 738         'playlist_mincount': 100,
 739         'info_dict': {
 740             'id': '1258156',
 741         },
 742         'expected_warnings': ['Retrying']
 743     }, {
 744         # Different entries between mobile and web, depending on region
 745         'url': 'https://www.tiktok.com/sticker/Elf-Friend-479565',
 746         'only_matching': True
 747     }]
 748
 749
 750 class TikTokTagIE(TikTokBaseListIE):
 751     IE_NAME = 'tiktok:tag'
 752     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/tag/(?P<id>[^/?#&]+)'
 753     _QUERY_NAME = 'ch_id'
 754     _API_ENDPOINT = 'challenge/aweme'
 755     _TESTS = [{
 756         'url': 'https://tiktok.com/tag/hello2018',
 757         'playlist_mincount': 39,
 758         'info_dict': {
 759             'id': '46294678',
 760             'title': 'hello2018',
 761         },
 762         'expected_warnings': ['Retrying']
 763     }, {
 764         'url': 'https://tiktok.com/tag/fypシ?is_copy_url=0&is_from_webapp=v1',
 765         'only_matching': True
 766     }]
 767
 768     def _real_extract(self, url):
 769         display_id = self._match_id(url)
 770         webpage = self._download_webpage(url, display_id, headers={
 771             'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)'
 772         })
 773         tag_id = self._html_search_regex(r'snssdk\d*://challenge/detail/(\d+)', webpage, 'tag ID')
 774         return self.playlist_result(self._entries(tag_id, display_id), tag_id, display_id)
 775
 776
 777 class DouyinIE(TikTokIE):
 778     _VALID_URL = r'https?://(?:www\.)?douyin\.com/video/(?P<id>[0-9]+)'
 779     _TESTS = [{
 780         'url': 'https://www.douyin.com/video/6961737553342991651',
 781         'md5': '10523312c8b8100f353620ac9dc8f067',
 782         'info_dict': {
 783             'id': '6961737553342991651',
 784             'ext': 'mp4',
 785             'title': '#杨超越  小小水手带你去远航❤️',
 786             'uploader': '杨超越',
 787             'upload_date': '20210513',
 788             'timestamp': 1620905839,
 789             'uploader_id': '110403406559',
 790             'view_count': int,
 791             'like_count': int,
 792             'repost_count': int,
 793             'comment_count': int,
 794         }
 795     }, {
 796         'url': 'https://www.douyin.com/video/6982497745948921092',
 797         'md5': 'd78408c984b9b5102904cf6b6bc2d712',
 798         'info_dict': {
 799             'id': '6982497745948921092',
 800             'ext': 'mp4',
 801             'title': '这个夏日和小羊@杨超越 一起遇见白色幻想',
 802             'uploader': '杨超越工作室',
 803             'upload_date': '20210708',
 804             'timestamp': 1625739481,
 805             'uploader_id': '408654318141572',
 806             'view_count': int,
 807             'like_count': int,
 808             'repost_count': int,
 809             'comment_count': int,
 810         }
 811     }, {
 812         'url': 'https://www.douyin.com/video/6953975910773099811',
 813         'md5': '72e882e24f75064c218b76c8b713c185',
 814         'info_dict': {
 815             'id': '6953975910773099811',
 816             'ext': 'mp4',
 817             'title': '#一起看海  出现在你的夏日里',
 818             'uploader': '杨超越',
 819             'upload_date': '20210422',
 820             'timestamp': 1619098692,
 821             'uploader_id': '110403406559',
 822             'view_count': int,
 823             'like_count': int,
 824             'repost_count': int,
 825             'comment_count': int,
 826         }
 827     }, {
 828         'url': 'https://www.douyin.com/video/6950251282489675042',
 829         'md5': 'b4db86aec367ef810ddd38b1737d2fed',
 830         'info_dict': {
 831             'id': '6950251282489675042',
 832             'ext': 'mp4',
 833             'title': '哈哈哈，成功了哈哈哈哈哈哈',
 834             'uploader': '杨超越',
 835             'upload_date': '20210412',
 836             'timestamp': 1618231483,
 837             'uploader_id': '110403406559',
 838             'view_count': int,
 839             'like_count': int,
 840             'repost_count': int,
 841             'comment_count': int,
 842         }
 843     }, {
 844         'url': 'https://www.douyin.com/video/6963263655114722595',
 845         'md5': '1abe1c477d05ee62efb40bf2329957cf',
 846         'info_dict': {
 847             'id': '6963263655114722595',
 848             'ext': 'mp4',
 849             'title': '#哪个爱豆的105度最甜 换个角度看看我哈哈',
 850             'uploader': '杨超越',
 851             'upload_date': '20210517',
 852             'timestamp': 1621261163,
 853             'uploader_id': '110403406559',
 854             'view_count': int,
 855             'like_count': int,
 856             'repost_count': int,
 857             'comment_count': int,
 858         }
 859     }]
 860     _APP_VERSIONS = [('9.6.0', '960')]
 861     _APP_NAME = 'aweme'
 862     _AID = 1128
 863     _API_HOSTNAME = 'aweme.snssdk.com'
 864     _UPLOADER_URL_FORMAT = 'https://www.douyin.com/user/%s'
 865     _WEBPAGE_HOST = 'https://www.douyin.com/'
 866
 867     def _real_extract(self, url):
 868         video_id = self._match_id(url)
 869
 870         try:
 871             return self._extract_aweme_app(video_id)
 872         except ExtractorError as e:
 873             self.report_warning(f'{e}; trying with webpage')
 874
 875         webpage = self._download_webpage(url, video_id)
 876         render_data_json = self._search_regex(
 877             r'<script [^>]*\bid=[\'"]RENDER_DATA[\'"][^>]*>(%7B.+%7D)</script>',
 878             webpage, 'render data', default=None)
 879         if not render_data_json:
 880             # TODO: Run verification challenge code to generate signature cookies
 881             raise ExtractorError('Fresh cookies (not necessarily logged in) are needed')
 882
 883         render_data = self._parse_json(
 884             render_data_json, video_id, transform_source=compat_urllib_parse_unquote)
 885         return self._parse_aweme_video_web(get_first(render_data, ('aweme', 'detail')), url)
 886
 887
 888 class TikTokVMIE(InfoExtractor):
 889     _VALID_URL = r'https?://(?:vm|vt)\.tiktok\.com/(?P<id>\w+)'
 890     IE_NAME = 'vm.tiktok'
 891
 892     _TESTS = [{
 893         'url': 'https://vm.tiktok.com/ZSe4FqkKd',
 894         'info_dict': {
 895             'id': '7023491746608712966',
 896             'ext': 'mp4',
 897             'title': 'md5:5607564db90271abbbf8294cca77eddd',
 898             'description': 'md5:5607564db90271abbbf8294cca77eddd',
 899             'duration': 11,
 900             'upload_date': '20211026',
 901             'uploader_id': '7007385080558846981',
 902             'creator': 'Memes',
 903             'artist': 'Memes',
 904             'track': 'original sound',
 905             'uploader': 'susmandem',
 906             'timestamp': 1635284105,
 907             'thumbnail': r're:https://.+\.webp.*',
 908             'like_count': int,
 909             'view_count': int,
 910             'comment_count': int,
 911             'repost_count': int,
 912             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAXcNoOEOxVyBzuII_E--T0MeCrLP0ay1Sm6x_n3dluiWEoWZD0VlQOytwad4W0i0n',
 913         }
 914     }, {
 915         'url': 'https://vt.tiktok.com/ZSe4FqkKd',
 916         'only_matching': True,
 917     }]
 918
 919     def _real_extract(self, url):
 920         new_url = self._request_webpage(
 921             HEADRequest(url), self._match_id(url), headers={'User-Agent': 'facebookexternalhit/1.1'}).geturl()
 922         if self.suitable(new_url):  # Prevent infinite loop in case redirect fails
 923             raise UnsupportedError(new_url)
 924         return self.url_result(new_url)