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
 237         thumbnails = []
 238         for cover_id in ('cover', 'ai_dynamic_cover', 'animated_cover', 'ai_dynamic_cover_bak',
 239                          'origin_cover', 'dynamic_cover'):
 240             cover = video_info.get(cover_id)
 241             if cover:
 242                 for cover_url in cover['url_list']:
 243                     thumbnails.append({
 244                         'id': cover_id,
 245                         'url': cover_url,
 246                     })
 247
 248         stats_info = aweme_detail.get('statistics', {})
 249         author_info = aweme_detail.get('author', {})
 250         music_info = aweme_detail.get('music', {})
 251         user_url = self._UPLOADER_URL_FORMAT % (traverse_obj(author_info,
 252                                                              'sec_uid', 'id', 'uid', 'unique_id',
 253                                                              expected_type=str_or_none, get_all=False))
 254         labels = traverse_obj(aweme_detail, ('hybrid_label', ..., 'text'), expected_type=str, default=[])
 255
 256         contained_music_track = traverse_obj(
 257             music_info, ('matched_song', 'title'), ('matched_pgc_sound', 'title'), expected_type=str)
 258         contained_music_author = traverse_obj(
 259             music_info, ('matched_song', 'author'), ('matched_pgc_sound', 'author'), 'author', expected_type=str)
 260
 261         is_generic_og_trackname = music_info.get('is_original_sound') and music_info.get('title') == 'original sound - %s' % music_info.get('owner_handle')
 262         if is_generic_og_trackname:
 263             music_track, music_author = contained_music_track or 'original sound', contained_music_author
 264         else:
 265             music_track, music_author = music_info.get('title'), music_info.get('author')
 266
 267         return {
 268             'id': aweme_id,
 269             'extractor_key': TikTokIE.ie_key(),
 270             'extractor': TikTokIE.IE_NAME,
 271             'webpage_url': self._create_url(author_info.get('uid'), aweme_id),
 272             'title': aweme_detail.get('desc'),
 273             'description': aweme_detail.get('desc'),
 274             'view_count': int_or_none(stats_info.get('play_count')),
 275             'like_count': int_or_none(stats_info.get('digg_count')),
 276             'repost_count': int_or_none(stats_info.get('share_count')),
 277             'comment_count': int_or_none(stats_info.get('comment_count')),
 278             'uploader': str_or_none(author_info.get('unique_id')),
 279             'creator': str_or_none(author_info.get('nickname')),
 280             'uploader_id': str_or_none(author_info.get('uid')),
 281             'uploader_url': user_url,
 282             'track': music_track,
 283             'album': str_or_none(music_info.get('album')) or None,
 284             'artist': music_author or None,
 285             'timestamp': int_or_none(aweme_detail.get('create_time')),
 286             'formats': formats,
 287             'subtitles': self.extract_subtitles(aweme_detail, aweme_id),
 288             'thumbnails': thumbnails,
 289             'duration': int_or_none(traverse_obj(video_info, 'duration', ('download_addr', 'duration')), scale=1000),
 290             'availability': self._availability(
 291                 is_private='Private' in labels,
 292                 needs_subscription='Friends only' in labels,
 293                 is_unlisted='Followers only' in labels),
 294             '_format_sort_fields': ('quality', 'codec', 'size', 'br'),
 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
 337         thumbnails = []
 338         for thumbnail_name in ('thumbnail', 'cover', 'dynamicCover', 'originCover'):
 339             if aweme_detail.get(thumbnail_name):
 340                 thumbnails = [{
 341                     'url': self._proto_relative_url(aweme_detail[thumbnail_name]),
 342                     'width': width,
 343                     'height': height
 344                 }]
 345
 346         return {
 347             'id': traverse_obj(aweme_detail, 'id', 'awemeId', expected_type=str_or_none),
 348             'title': aweme_detail.get('desc'),
 349             'duration': try_get(aweme_detail, lambda x: x['video']['duration'], int),
 350             'view_count': int_or_none(stats_info.get('playCount')),
 351             'like_count': int_or_none(stats_info.get('diggCount')),
 352             'repost_count': int_or_none(stats_info.get('shareCount')),
 353             'comment_count': int_or_none(stats_info.get('commentCount')),
 354             'timestamp': int_or_none(aweme_detail.get('createTime')),
 355             'creator': str_or_none(author_info.get('nickname')),
 356             'uploader': str_or_none(author_info.get('uniqueId') or aweme_detail.get('author')),
 357             'uploader_id': str_or_none(author_info.get('id') or aweme_detail.get('authorId')),
 358             'uploader_url': user_url,
 359             'track': str_or_none(music_info.get('title')),
 360             'album': str_or_none(music_info.get('album')) or None,
 361             'artist': str_or_none(music_info.get('authorName')),
 362             'formats': formats,
 363             'thumbnails': thumbnails,
 364             'description': str_or_none(aweme_detail.get('desc')),
 365             'http_headers': {
 366                 'Referer': webpage_url
 367             }
 368         }
 369
 370
 371 class TikTokIE(TikTokBaseIE):
 372     _VALID_URL = r'https?://www\.tiktok\.com/(?:embed|@(?P<user_id>[\w\.-]+)/video)/(?P<id>\d+)'
 373     _EMBED_REGEX = [rf'<(?:script|iframe)[^>]+\bsrc=(["\'])(?P<url>{_VALID_URL})']
 374
 375     _TESTS = [{
 376         'url': 'https://www.tiktok.com/@leenabhushan/video/6748451240264420610',
 377         'md5': '736bb7a466c6f0a6afeb597da1e6f5b7',
 378         'info_dict': {
 379             'id': '6748451240264420610',
 380             'ext': 'mp4',
 381             'title': '#jassmanak #lehanga #leenabhushan',
 382             'description': '#jassmanak #lehanga #leenabhushan',
 383             'duration': 13,
 384             'height': 1024,
 385             'width': 576,
 386             'uploader': 'leenabhushan',
 387             'uploader_id': '6691488002098119685',
 388             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA_Eb4t1vodM1IuTy_cvp9CY22RAb59xqrO0Xtz9CYQJvgXaDvZxYnZYRzDWhhgJmy',
 389             'creator': 'facestoriesbyleenabh',
 390             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 391             'upload_date': '20191016',
 392             'timestamp': 1571246252,
 393             'view_count': int,
 394             'like_count': int,
 395             'repost_count': int,
 396             'comment_count': int,
 397             'artist': 'Ysrbeats',
 398             'album': 'Lehanga',
 399             'track': 'Lehanga',
 400         }
 401     }, {
 402         'url': 'https://www.tiktok.com/@patroxofficial/video/6742501081818877190?langCountry=en',
 403         'md5': '6f3cf8cdd9b28cb8363fe0a9a160695b',
 404         'info_dict': {
 405             'id': '6742501081818877190',
 406             'ext': 'mp4',
 407             'title': 'md5:5e2a23877420bb85ce6521dbee39ba94',
 408             'description': 'md5:5e2a23877420bb85ce6521dbee39ba94',
 409             'duration': 27,
 410             'height': 960,
 411             'width': 540,
 412             'uploader': 'patrox',
 413             'uploader_id': '18702747',
 414             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAiFnldaILebi5heDoVU6bn4jBWWycX6-9U3xuNPqZ8Ws',
 415             'creator': 'patroX',
 416             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 417             'upload_date': '20190930',
 418             'timestamp': 1569860870,
 419             'view_count': int,
 420             'like_count': int,
 421             'repost_count': int,
 422             'comment_count': int,
 423             'artist': 'Evan Todd, Jessica Keenan Wynn, Alice Lee, Barrett Wilbert Weed & Jon Eidson',
 424             'track': 'Big Fun',
 425         }
 426     }, {
 427         # Banned audio, only available on the app
 428         'url': 'https://www.tiktok.com/@barudakhb_/video/6984138651336838402',
 429         'info_dict': {
 430             'id': '6984138651336838402',
 431             'ext': 'mp4',
 432             'title': 'Balas @yolaaftwsr hayu yu ? #SquadRandom_ 🔥',
 433             'description': 'Balas @yolaaftwsr hayu yu ? #SquadRandom_ 🔥',
 434             'uploader': 'barudakhb_',
 435             'creator': 'md5:29f238c49bc0c176cb3cef1a9cea9fa6',
 436             'uploader_id': '6974687867511718913',
 437             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAbhBwQC-R1iKoix6jDFsF-vBdfx2ABoDjaZrM9fX6arU3w71q3cOWgWuTXn1soZ7d',
 438             'track': 'Boka Dance',
 439             'artist': 'md5:29f238c49bc0c176cb3cef1a9cea9fa6',
 440             'timestamp': 1626121503,
 441             'duration': 18,
 442             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 443             'upload_date': '20210712',
 444             'view_count': int,
 445             'like_count': int,
 446             'repost_count': int,
 447             'comment_count': int,
 448         }
 449     }, {
 450         # Sponsored video, only available with feed workaround
 451         'url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_/video/7042692929109986561',
 452         'info_dict': {
 453             'id': '7042692929109986561',
 454             'ext': 'mp4',
 455             'title': 'Slap and Run!',
 456             'description': 'Slap and Run!',
 457             'uploader': 'user440922249',
 458             'creator': 'Slap And Run',
 459             'uploader_id': '7036055384943690754',
 460             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_',
 461             'track': 'Promoted Music',
 462             'timestamp': 1639754738,
 463             'duration': 30,
 464             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 465             'upload_date': '20211217',
 466             'view_count': int,
 467             'like_count': int,
 468             'repost_count': int,
 469             'comment_count': int,
 470         },
 471         'expected_warnings': ['trying with webpage', 'Unable to find video in feed']
 472     }, {
 473         # Video without title and description
 474         'url': 'https://www.tiktok.com/@pokemonlife22/video/7059698374567611694',
 475         'info_dict': {
 476             'id': '7059698374567611694',
 477             'ext': 'mp4',
 478             'title': 'TikTok video #7059698374567611694',
 479             'description': '',
 480             'uploader': 'pokemonlife22',
 481             'creator': 'Pokemon',
 482             'uploader_id': '6820838815978423302',
 483             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA0tF1nBwQVVMyrGu3CqttkNgM68Do1OXUFuCY0CRQk8fEtSVDj89HqoqvbSTmUP2W',
 484             'track': 'original sound',
 485             'timestamp': 1643714123,
 486             'duration': 6,
 487             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 488             'upload_date': '20220201',
 489             'artist': 'Pokemon',
 490             'view_count': int,
 491             'like_count': int,
 492             'repost_count': int,
 493             'comment_count': int,
 494         },
 495     }, {
 496         # hydration JSON is sent in a <script> element
 497         'url': 'https://www.tiktok.com/@denidil6/video/7065799023130643713',
 498         'info_dict': {
 499             'id': '7065799023130643713',
 500             'ext': 'mp4',
 501             'title': '#denidil#денидил',
 502             'description': '#denidil#денидил',
 503             'uploader': 'denidil6',
 504             'uploader_id': '7046664115636405250',
 505             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAsvMSzFdQ4ikl3uR2TEJwMBbB2yZh2Zxwhx-WCo3rbDpAharE3GQCrFuJArI3C8QJ',
 506             'artist': 'Holocron Music',
 507             'album': 'Wolf Sounds (1 Hour) Enjoy the Company of the Animal That Is the Majestic King of the Night',
 508             'track': 'Wolf Sounds (1 Hour) Enjoy the Company of the Animal That Is the Majestic King of the Night',
 509             'timestamp': 1645134536,
 510             'duration': 26,
 511             'upload_date': '20220217',
 512             'view_count': int,
 513             'like_count': int,
 514             'repost_count': int,
 515             'comment_count': int,
 516         },
 517         'skip': 'This video is unavailable',
 518     }, {
 519         # Auto-captions available
 520         'url': 'https://www.tiktok.com/@hankgreen1/video/7047596209028074758',
 521         'only_matching': True
 522     }]
 523
 524     def _extract_aweme_app(self, aweme_id):
 525         feed_list = self._call_api('feed', {'aweme_id': aweme_id}, aweme_id,
 526                                    note='Downloading video feed', errnote='Unable to download video feed').get('aweme_list') or []
 527         aweme_detail = next((aweme for aweme in feed_list if str(aweme.get('aweme_id')) == aweme_id), None)
 528         if not aweme_detail:
 529             raise ExtractorError('Unable to find video in feed', video_id=aweme_id)
 530         return self._parse_aweme_video_app(aweme_detail)
 531
 532     def _real_extract(self, url):
 533         video_id, user_id = self._match_valid_url(url).group('id', 'user_id')
 534         try:
 535             return self._extract_aweme_app(video_id)
 536         except ExtractorError as e:
 537             self.report_warning(f'{e}; trying with webpage')
 538
 539         url = self._create_url(user_id, video_id)
 540         webpage = self._download_webpage(url, video_id, headers={'User-Agent': 'User-Agent:Mozilla/5.0'})
 541         next_data = self._search_nextjs_data(webpage, video_id, default='{}')
 542         if next_data:
 543             status = traverse_obj(next_data, ('props', 'pageProps', 'statusCode'), expected_type=int) or 0
 544             video_data = traverse_obj(next_data, ('props', 'pageProps', 'itemInfo', 'itemStruct'), expected_type=dict)
 545         else:
 546             sigi_data = self._get_sigi_state(webpage, video_id)
 547             status = traverse_obj(sigi_data, ('VideoPage', 'statusCode'), expected_type=int) or 0
 548             video_data = traverse_obj(sigi_data, ('ItemModule', video_id), expected_type=dict)
 549
 550         if status == 0:
 551             return self._parse_aweme_video_web(video_data, url)
 552         elif status == 10216:
 553             raise ExtractorError('This video is private', expected=True)
 554         raise ExtractorError('Video not available', video_id=video_id)
 555
 556
 557 class TikTokUserIE(TikTokBaseIE):
 558     IE_NAME = 'tiktok:user'
 559     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/@(?P<id>[\w\.-]+)/?(?:$|[#?])'
 560     _WORKING = False
 561     _TESTS = [{
 562         'url': 'https://tiktok.com/@corgibobaa?lang=en',
 563         'playlist_mincount': 45,
 564         'info_dict': {
 565             'id': '6935371178089399301',
 566             'title': 'corgibobaa',
 567             'thumbnail': r're:https://.+_1080x1080\.webp'
 568         },
 569         'expected_warnings': ['Retrying']
 570     }, {
 571         'url': 'https://www.tiktok.com/@6820838815978423302',
 572         'playlist_mincount': 5,
 573         'info_dict': {
 574             'id': '6820838815978423302',
 575             'title': '6820838815978423302',
 576             'thumbnail': r're:https://.+_1080x1080\.webp'
 577         },
 578         'expected_warnings': ['Retrying']
 579     }, {
 580         'url': 'https://www.tiktok.com/@meme',
 581         'playlist_mincount': 593,
 582         'info_dict': {
 583             'id': '79005827461758976',
 584             'title': 'meme',
 585             'thumbnail': r're:https://.+_1080x1080\.webp'
 586         },
 587         'expected_warnings': ['Retrying']
 588     }]
 589
 590     r'''  # TODO: Fix by adding _signature to api_url
 591     def _entries(self, webpage, user_id, username):
 592         secuid = self._search_regex(r'\"secUid\":\"(?P<secUid>[^\"]+)', webpage, username)
 593         verifyfp_cookie = self._get_cookies('https://www.tiktok.com').get('s_v_web_id')
 594         if not verifyfp_cookie:
 595             raise ExtractorError('Improper cookies (missing s_v_web_id).', expected=True)
 596         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='
 597         cursor = '0'
 598         for page in itertools.count():
 599             data_json = self._download_json(api_url + cursor, username, note='Downloading Page %d' % page)
 600             for video in data_json.get('itemList', []):
 601                 video_id = video['id']
 602                 video_url = f'https://www.tiktok.com/@{user_id}/video/{video_id}'
 603                 yield self._url_result(video_url, 'TikTok', video_id, str_or_none(video.get('desc')))
 604             if not data_json.get('hasMore'):
 605                 break
 606             cursor = data_json['cursor']
 607     '''
 608
 609     def _video_entries_api(self, webpage, user_id, username):
 610         query = {
 611             'user_id': user_id,
 612             'count': 21,
 613             'max_cursor': 0,
 614             'min_cursor': 0,
 615             'retry_type': 'no_retry',
 616             '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.
 617         }
 618
 619         for page in itertools.count(1):
 620             for retry in self.RetryManager():
 621                 try:
 622                     post_list = self._call_api(
 623                         'aweme/post', query, username, note=f'Downloading user video list page {page}',
 624                         errnote='Unable to download user video list')
 625                 except ExtractorError as e:
 626                     if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0:
 627                         retry.error = e
 628                         continue
 629                     raise
 630             yield from post_list.get('aweme_list', [])
 631             if not post_list.get('has_more'):
 632                 break
 633             query['max_cursor'] = post_list['max_cursor']
 634
 635     def _entries_api(self, user_id, videos):
 636         for video in videos:
 637             yield {
 638                 **self._parse_aweme_video_app(video),
 639                 'extractor_key': TikTokIE.ie_key(),
 640                 'extractor': 'TikTok',
 641                 'webpage_url': f'https://tiktok.com/@{user_id}/video/{video["aweme_id"]}',
 642             }
 643
 644     def _real_extract(self, url):
 645         user_name = self._match_id(url)
 646         webpage = self._download_webpage(url, user_name, headers={
 647             'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)'
 648         })
 649         user_id = self._html_search_regex(r'snssdk\d*://user/profile/(\d+)', webpage, 'user ID', default=None) or user_name
 650
 651         videos = LazyList(self._video_entries_api(webpage, user_id, user_name))
 652         thumbnail = traverse_obj(videos, (0, 'author', 'avatar_larger', 'url_list', 0))
 653
 654         return self.playlist_result(self._entries_api(user_id, videos), user_id, user_name, thumbnail=thumbnail)
 655
 656
 657 class TikTokBaseListIE(TikTokBaseIE):  # XXX: Conventionally, base classes should end with BaseIE/InfoExtractor
 658     def _entries(self, list_id, display_id):
 659         query = {
 660             self._QUERY_NAME: list_id,
 661             'cursor': 0,
 662             'count': 20,
 663             'type': 5,
 664             'device_id': ''.join(random.choice(string.digits) for i in range(19))
 665         }
 666
 667         for page in itertools.count(1):
 668             for retry in self.RetryManager():
 669                 try:
 670                     post_list = self._call_api(
 671                         self._API_ENDPOINT, query, display_id, note=f'Downloading video list page {page}',
 672                         errnote='Unable to download video list')
 673                 except ExtractorError as e:
 674                     if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0:
 675                         retry.error = e
 676                         continue
 677                     raise
 678             for video in post_list.get('aweme_list', []):
 679                 yield {
 680                     **self._parse_aweme_video_app(video),
 681                     'extractor_key': TikTokIE.ie_key(),
 682                     'extractor': 'TikTok',
 683                     'webpage_url': f'https://tiktok.com/@_/video/{video["aweme_id"]}',
 684                 }
 685             if not post_list.get('has_more'):
 686                 break
 687             query['cursor'] = post_list['cursor']
 688
 689     def _real_extract(self, url):
 690         list_id = self._match_id(url)
 691         return self.playlist_result(self._entries(list_id, list_id), list_id)
 692
 693
 694 class TikTokSoundIE(TikTokBaseListIE):
 695     IE_NAME = 'tiktok:sound'
 696     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/music/[\w\.-]+-(?P<id>[\d]+)[/?#&]?'
 697     _WORKING = False
 698     _QUERY_NAME = 'music_id'
 699     _API_ENDPOINT = 'music/aweme'
 700     _TESTS = [{
 701         'url': 'https://www.tiktok.com/music/Build-a-Btch-6956990112127585029?lang=en',
 702         'playlist_mincount': 100,
 703         'info_dict': {
 704             'id': '6956990112127585029'
 705         },
 706         'expected_warnings': ['Retrying']
 707     }, {
 708         # Actual entries are less than listed video count
 709         'url': 'https://www.tiktok.com/music/jiefei-soap-remix-7036843036118469381',
 710         'playlist_mincount': 2182,
 711         'info_dict': {
 712             'id': '7036843036118469381'
 713         },
 714         'expected_warnings': ['Retrying']
 715     }]
 716
 717
 718 class TikTokEffectIE(TikTokBaseListIE):
 719     IE_NAME = 'tiktok:effect'
 720     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/sticker/[\w\.-]+-(?P<id>[\d]+)[/?#&]?'
 721     _WORKING = False
 722     _QUERY_NAME = 'sticker_id'
 723     _API_ENDPOINT = 'sticker/aweme'
 724     _TESTS = [{
 725         'url': 'https://www.tiktok.com/sticker/MATERIAL-GWOOORL-1258156',
 726         'playlist_mincount': 100,
 727         'info_dict': {
 728             'id': '1258156',
 729         },
 730         'expected_warnings': ['Retrying']
 731     }, {
 732         # Different entries between mobile and web, depending on region
 733         'url': 'https://www.tiktok.com/sticker/Elf-Friend-479565',
 734         'only_matching': True
 735     }]
 736
 737
 738 class TikTokTagIE(TikTokBaseListIE):
 739     IE_NAME = 'tiktok:tag'
 740     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/tag/(?P<id>[^/?#&]+)'
 741     _WORKING = False
 742     _QUERY_NAME = 'ch_id'
 743     _API_ENDPOINT = 'challenge/aweme'
 744     _TESTS = [{
 745         'url': 'https://tiktok.com/tag/hello2018',
 746         'playlist_mincount': 39,
 747         'info_dict': {
 748             'id': '46294678',
 749             'title': 'hello2018',
 750         },
 751         'expected_warnings': ['Retrying']
 752     }, {
 753         'url': 'https://tiktok.com/tag/fypシ?is_copy_url=0&is_from_webapp=v1',
 754         'only_matching': True
 755     }]
 756
 757     def _real_extract(self, url):
 758         display_id = self._match_id(url)
 759         webpage = self._download_webpage(url, display_id, headers={
 760             'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)'
 761         })
 762         tag_id = self._html_search_regex(r'snssdk\d*://challenge/detail/(\d+)', webpage, 'tag ID')
 763         return self.playlist_result(self._entries(tag_id, display_id), tag_id, display_id)
 764
 765
 766 class DouyinIE(TikTokIE):  # XXX: Do not subclass from concrete IE
 767     _VALID_URL = r'https?://(?:www\.)?douyin\.com/video/(?P<id>[0-9]+)'
 768     _TESTS = [{
 769         'url': 'https://www.douyin.com/video/6961737553342991651',
 770         'md5': '10523312c8b8100f353620ac9dc8f067',
 771         'info_dict': {
 772             'id': '6961737553342991651',
 773             'ext': 'mp4',
 774             'title': '#杨超越  小小水手带你去远航❤️',
 775             'uploader': '杨超越',
 776             'upload_date': '20210513',
 777             'timestamp': 1620905839,
 778             'uploader_id': '110403406559',
 779             'view_count': int,
 780             'like_count': int,
 781             'repost_count': int,
 782             'comment_count': int,
 783         }
 784     }, {
 785         'url': 'https://www.douyin.com/video/6982497745948921092',
 786         'md5': 'd78408c984b9b5102904cf6b6bc2d712',
 787         'info_dict': {
 788             'id': '6982497745948921092',
 789             'ext': 'mp4',
 790             'title': '这个夏日和小羊@杨超越 一起遇见白色幻想',
 791             'uploader': '杨超越工作室',
 792             'upload_date': '20210708',
 793             'timestamp': 1625739481,
 794             'uploader_id': '408654318141572',
 795             'view_count': int,
 796             'like_count': int,
 797             'repost_count': int,
 798             'comment_count': int,
 799         }
 800     }, {
 801         'url': 'https://www.douyin.com/video/6953975910773099811',
 802         'md5': '72e882e24f75064c218b76c8b713c185',
 803         'info_dict': {
 804             'id': '6953975910773099811',
 805             'ext': 'mp4',
 806             'title': '#一起看海  出现在你的夏日里',
 807             'uploader': '杨超越',
 808             'upload_date': '20210422',
 809             'timestamp': 1619098692,
 810             'uploader_id': '110403406559',
 811             'view_count': int,
 812             'like_count': int,
 813             'repost_count': int,
 814             'comment_count': int,
 815         }
 816     }, {
 817         'url': 'https://www.douyin.com/video/6950251282489675042',
 818         'md5': 'b4db86aec367ef810ddd38b1737d2fed',
 819         'info_dict': {
 820             'id': '6950251282489675042',
 821             'ext': 'mp4',
 822             'title': '哈哈哈，成功了哈哈哈哈哈哈',
 823             'uploader': '杨超越',
 824             'upload_date': '20210412',
 825             'timestamp': 1618231483,
 826             'uploader_id': '110403406559',
 827             'view_count': int,
 828             'like_count': int,
 829             'repost_count': int,
 830             'comment_count': int,
 831         }
 832     }, {
 833         'url': 'https://www.douyin.com/video/6963263655114722595',
 834         'md5': '1abe1c477d05ee62efb40bf2329957cf',
 835         'info_dict': {
 836             'id': '6963263655114722595',
 837             'ext': 'mp4',
 838             'title': '#哪个爱豆的105度最甜 换个角度看看我哈哈',
 839             'uploader': '杨超越',
 840             'upload_date': '20210517',
 841             'timestamp': 1621261163,
 842             'uploader_id': '110403406559',
 843             'view_count': int,
 844             'like_count': int,
 845             'repost_count': int,
 846             'comment_count': int,
 847         }
 848     }]
 849     _APP_VERSIONS = [('9.6.0', '960')]
 850     _APP_NAME = 'aweme'
 851     _AID = 1128
 852     _API_HOSTNAME = 'aweme.snssdk.com'
 853     _UPLOADER_URL_FORMAT = 'https://www.douyin.com/user/%s'
 854     _WEBPAGE_HOST = 'https://www.douyin.com/'
 855
 856     def _real_extract(self, url):
 857         video_id = self._match_id(url)
 858
 859         try:
 860             return self._extract_aweme_app(video_id)
 861         except ExtractorError as e:
 862             self.report_warning(f'{e}; trying with webpage')
 863
 864         webpage = self._download_webpage(url, video_id)
 865         render_data_json = self._search_regex(
 866             r'<script [^>]*\bid=[\'"]RENDER_DATA[\'"][^>]*>(%7B.+%7D)</script>',
 867             webpage, 'render data', default=None)
 868         if not render_data_json:
 869             # TODO: Run verification challenge code to generate signature cookies
 870             raise ExtractorError('Fresh cookies (not necessarily logged in) are needed')
 871
 872         render_data = self._parse_json(
 873             render_data_json, video_id, transform_source=compat_urllib_parse_unquote)
 874         return self._parse_aweme_video_web(get_first(render_data, ('aweme', 'detail')), url)
 875
 876
 877 class TikTokVMIE(InfoExtractor):
 878     _VALID_URL = r'https?://(?:vm|vt)\.tiktok\.com/(?P<id>\w+)'
 879     IE_NAME = 'vm.tiktok'
 880
 881     _TESTS = [{
 882         'url': 'https://vm.tiktok.com/ZSe4FqkKd',
 883         'info_dict': {
 884             'id': '7023491746608712966',
 885             'ext': 'mp4',
 886             'title': 'md5:5607564db90271abbbf8294cca77eddd',
 887             'description': 'md5:5607564db90271abbbf8294cca77eddd',
 888             'duration': 11,
 889             'upload_date': '20211026',
 890             'uploader_id': '7007385080558846981',
 891             'creator': 'Memes',
 892             'artist': 'Memes',
 893             'track': 'original sound',
 894             'uploader': 'susmandem',
 895             'timestamp': 1635284105,
 896             'thumbnail': r're:https://.+\.webp.*',
 897             'like_count': int,
 898             'view_count': int,
 899             'comment_count': int,
 900             'repost_count': int,
 901             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAXcNoOEOxVyBzuII_E--T0MeCrLP0ay1Sm6x_n3dluiWEoWZD0VlQOytwad4W0i0n',
 902         }
 903     }, {
 904         'url': 'https://vt.tiktok.com/ZSe4FqkKd',
 905         'only_matching': True,
 906     }]
 907
 908     def _real_extract(self, url):
 909         new_url = self._request_webpage(
 910             HEADRequest(url), self._match_id(url), headers={'User-Agent': 'facebookexternalhit/1.1'}).geturl()
 911         if self.suitable(new_url):  # Prevent infinite loop in case redirect fails
 912             raise UnsupportedError(new_url)
 913         return self.url_result(new_url)