yt_dlp/extractor/tiktok.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import itertools
   5 import random
   6 import string
   7 import time
   8 import json
   9
  10 from .common import InfoExtractor
  11 from ..compat import (
  12     compat_urllib_parse_unquote,
  13     compat_urllib_parse_urlparse
  14 )
  15 from ..utils import (
  16     ExtractorError,
  17     int_or_none,
  18     join_nonempty,
  19     LazyList,
  20     str_or_none,
  21     traverse_obj,
  22     try_get,
  23     url_or_none,
  24     qualities,
  25 )
  26
  27
  28 class TikTokBaseIE(InfoExtractor):
  29     _APP_VERSION = '20.1.0'
  30     _MANIFEST_APP_VERSION = '210'
  31     _APP_NAME = 'trill'
  32     _AID = 1180
  33     _API_HOSTNAME = 'api-h2.tiktokv.com'
  34     _UPLOADER_URL_FORMAT = 'https://www.tiktok.com/@%s'
  35     _WEBPAGE_HOST = 'https://www.tiktok.com/'
  36     QUALITIES = ('360p', '540p', '720p', '1080p')
  37
  38     def _call_api(self, ep, query, video_id, fatal=True,
  39                   note='Downloading API JSON', errnote='Unable to download API page'):
  40         real_query = {
  41             **query,
  42             'version_name': self._APP_VERSION,
  43             'version_code': self._MANIFEST_APP_VERSION,
  44             'build_number': self._APP_VERSION,
  45             'manifest_version_code': self._MANIFEST_APP_VERSION,
  46             'update_version_code': self._MANIFEST_APP_VERSION,
  47             'openudid': ''.join(random.choice('0123456789abcdef') for _ in range(16)),
  48             'uuid': ''.join([random.choice(string.digits) for _ in range(16)]),
  49             '_rticket': int(time.time() * 1000),
  50             'ts': int(time.time()),
  51             'device_brand': 'Google',
  52             'device_type': 'Pixel 4',
  53             'device_platform': 'android',
  54             'resolution': '1080*1920',
  55             'dpi': 420,
  56             'os_version': '10',
  57             'os_api': '29',
  58             'carrier_region': 'US',
  59             'sys_region': 'US',
  60             'region': 'US',
  61             'app_name': self._APP_NAME,
  62             'app_language': 'en',
  63             'language': 'en',
  64             'timezone_name': 'America/New_York',
  65             'timezone_offset': '-14400',
  66             'channel': 'googleplay',
  67             'ac': 'wifi',
  68             'mcc_mnc': '310260',
  69             'is_my_cn': 0,
  70             'aid': self._AID,
  71             'ssmix': 'a',
  72             'as': 'a1qwert123',
  73             'cp': 'cbfhckdckkde1',
  74         }
  75         self._set_cookie(self._API_HOSTNAME, 'odin_tt', ''.join(random.choice('0123456789abcdef') for _ in range(160)))
  76         webpage_cookies = self._get_cookies(self._WEBPAGE_HOST)
  77         if webpage_cookies.get('sid_tt'):
  78             self._set_cookie(self._API_HOSTNAME, 'sid_tt', webpage_cookies['sid_tt'].value)
  79         return self._download_json(
  80             'https://%s/aweme/v1/%s/' % (self._API_HOSTNAME, ep), video_id=video_id,
  81             fatal=fatal, note=note, errnote=errnote, headers={
  82                 'User-Agent': f'com.ss.android.ugc.trill/{self._MANIFEST_APP_VERSION} (Linux; U; Android 10; en_US; Pixel 4; Build/QQ3A.200805.001; Cronet/58.0.2991.0)',
  83                 'Accept': 'application/json',
  84             }, query=real_query)
  85
  86     def _parse_aweme_video_app(self, aweme_detail):
  87         aweme_id = aweme_detail['aweme_id']
  88         video_info = aweme_detail['video']
  89
  90         def parse_url_key(url_key):
  91             format_id, codec, res, bitrate = self._search_regex(
  92                 r'v[^_]+_(?P<id>(?P<codec>[^_]+)_(?P<res>\d+p)_(?P<bitrate>\d+))', url_key,
  93                 'url key', default=(None, None, None, None), group=('id', 'codec', 'res', 'bitrate'))
  94             if not format_id:
  95                 return {}, None
  96             return {
  97                 'format_id': format_id,
  98                 'vcodec': 'h265' if codec == 'bytevc1' else codec,
  99                 'tbr': int_or_none(bitrate, scale=1000) or None,
 100                 'quality': qualities(self.QUALITIES)(res),
 101             }, res
 102
 103         known_resolutions = {}
 104
 105         def extract_addr(addr, add_meta={}):
 106             parsed_meta, res = parse_url_key(addr.get('url_key', ''))
 107             if res:
 108                 known_resolutions.setdefault(res, {}).setdefault('height', add_meta.get('height'))
 109                 known_resolutions[res].setdefault('width', add_meta.get('width'))
 110                 parsed_meta.update(known_resolutions.get(res, {}))
 111                 add_meta.setdefault('height', int_or_none(res[:-1]))
 112             return [{
 113                 'url': url,
 114                 'filesize': int_or_none(addr.get('data_size')),
 115                 'ext': 'mp4',
 116                 'acodec': 'aac',
 117                 'source_preference': -2 if 'aweme/v1' in url else -1,  # Downloads from API might get blocked
 118                 **add_meta, **parsed_meta,
 119                 'format_note': join_nonempty(
 120                     add_meta.get('format_note'), '(API)' if 'aweme/v1' in url else None, delim=' ')
 121             } for url in addr.get('url_list') or []]
 122
 123         # Hack: Add direct video links first to prioritize them when removing duplicate formats
 124         formats = []
 125         if video_info.get('play_addr'):
 126             formats.extend(extract_addr(video_info['play_addr'], {
 127                 'format_id': 'play_addr',
 128                 'format_note': 'Direct video',
 129                 'vcodec': 'h265' if traverse_obj(
 130                     video_info, 'is_bytevc1', 'is_h265') else 'h264',  # TODO: Check for "direct iOS" videos, like https://www.tiktok.com/@cookierun_dev/video/7039716639834656002
 131                 'width': video_info.get('width'),
 132                 'height': video_info.get('height'),
 133             }))
 134         if video_info.get('download_addr'):
 135             formats.extend(extract_addr(video_info['download_addr'], {
 136                 'format_id': 'download_addr',
 137                 'format_note': 'Download video%s' % (', watermarked' if video_info.get('has_watermark') else ''),
 138                 'vcodec': 'h264',
 139                 'width': video_info.get('width'),
 140                 'height': video_info.get('height'),
 141                 'preference': -2 if video_info.get('has_watermark') else -1,
 142             }))
 143         if video_info.get('play_addr_h264'):
 144             formats.extend(extract_addr(video_info['play_addr_h264'], {
 145                 'format_id': 'play_addr_h264',
 146                 'format_note': 'Direct video',
 147                 'vcodec': 'h264',
 148             }))
 149         if video_info.get('play_addr_bytevc1'):
 150             formats.extend(extract_addr(video_info['play_addr_bytevc1'], {
 151                 'format_id': 'play_addr_bytevc1',
 152                 'format_note': 'Direct video',
 153                 'vcodec': 'h265',
 154             }))
 155
 156         for bitrate in video_info.get('bit_rate', []):
 157             if bitrate.get('play_addr'):
 158                 formats.extend(extract_addr(bitrate['play_addr'], {
 159                     'format_id': bitrate.get('gear_name'),
 160                     'format_note': 'Playback video',
 161                     'tbr': try_get(bitrate, lambda x: x['bit_rate'] / 1000),
 162                     'vcodec': 'h265' if traverse_obj(
 163                         bitrate, 'is_bytevc1', 'is_h265') else 'h264',
 164                     'fps': bitrate.get('FPS'),
 165                 }))
 166
 167         self._remove_duplicate_formats(formats)
 168         auth_cookie = self._get_cookies(self._WEBPAGE_HOST).get('sid_tt')
 169         if auth_cookie:
 170             for f in formats:
 171                 self._set_cookie(compat_urllib_parse_urlparse(f['url']).hostname, 'sid_tt', auth_cookie.value)
 172         self._sort_formats(formats, ('quality', 'codec', 'size', 'br'))
 173
 174         thumbnails = []
 175         for cover_id in ('cover', 'ai_dynamic_cover', 'animated_cover', 'ai_dynamic_cover_bak',
 176                          'origin_cover', 'dynamic_cover'):
 177             cover = video_info.get(cover_id)
 178             if cover:
 179                 for cover_url in cover['url_list']:
 180                     thumbnails.append({
 181                         'id': cover_id,
 182                         'url': cover_url,
 183                     })
 184
 185         stats_info = aweme_detail.get('statistics', {})
 186         author_info = aweme_detail.get('author', {})
 187         music_info = aweme_detail.get('music', {})
 188         user_url = self._UPLOADER_URL_FORMAT % (traverse_obj(author_info,
 189                                                              'sec_uid', 'id', 'uid', 'unique_id',
 190                                                              expected_type=str_or_none, get_all=False))
 191         labels = traverse_obj(aweme_detail, ('hybrid_label', ..., 'text'), expected_type=str, default=[])
 192
 193         contained_music_track = traverse_obj(
 194             music_info, ('matched_song', 'title'), ('matched_pgc_sound', 'title'), expected_type=str)
 195         contained_music_author = traverse_obj(
 196             music_info, ('matched_song', 'author'), ('matched_pgc_sound', 'author'), 'author', expected_type=str)
 197
 198         is_generic_og_trackname = music_info.get('is_original_sound') and music_info.get('title') == 'original sound - %s' % music_info.get('owner_handle')
 199         if is_generic_og_trackname:
 200             music_track, music_author = contained_music_track or 'original sound', contained_music_author
 201         else:
 202             music_track, music_author = music_info.get('title'), music_info.get('author')
 203
 204         return {
 205             'id': aweme_id,
 206             'title': aweme_detail['desc'],
 207             'description': aweme_detail['desc'],
 208             'view_count': int_or_none(stats_info.get('play_count')),
 209             'like_count': int_or_none(stats_info.get('digg_count')),
 210             'repost_count': int_or_none(stats_info.get('share_count')),
 211             'comment_count': int_or_none(stats_info.get('comment_count')),
 212             'uploader': str_or_none(author_info.get('unique_id')),
 213             'creator': str_or_none(author_info.get('nickname')),
 214             'uploader_id': str_or_none(author_info.get('uid')),
 215             'uploader_url': user_url,
 216             'track': music_track,
 217             'album': str_or_none(music_info.get('album')) or None,
 218             'artist': music_author,
 219             'timestamp': int_or_none(aweme_detail.get('create_time')),
 220             'formats': formats,
 221             'thumbnails': thumbnails,
 222             'duration': int_or_none(traverse_obj(video_info, 'duration', ('download_addr', 'duration')), scale=1000),
 223             'availability': self._availability(
 224                 is_private='Private' in labels,
 225                 needs_subscription='Friends only' in labels,
 226                 is_unlisted='Followers only' in labels)
 227         }
 228
 229     def _parse_aweme_video_web(self, aweme_detail, webpage_url):
 230         video_info = aweme_detail['video']
 231         author_info = traverse_obj(aweme_detail, 'authorInfo', 'author', expected_type=dict, default={})
 232         music_info = aweme_detail.get('music') or {}
 233         stats_info = aweme_detail.get('stats') or {}
 234         user_url = self._UPLOADER_URL_FORMAT % (traverse_obj(author_info,
 235                                                              'secUid', 'id', 'uid', 'uniqueId',
 236                                                              expected_type=str_or_none, get_all=False)
 237                                                 or aweme_detail.get('authorSecId'))
 238
 239         formats = []
 240         play_url = video_info.get('playAddr')
 241         width = video_info.get('width')
 242         height = video_info.get('height')
 243         if isinstance(play_url, str):
 244             formats = [{
 245                 'url': self._proto_relative_url(play_url),
 246                 'ext': 'mp4',
 247                 'width': width,
 248                 'height': height,
 249             }]
 250         elif isinstance(play_url, list):
 251             formats = [{
 252                 'url': self._proto_relative_url(url),
 253                 'ext': 'mp4',
 254                 'width': width,
 255                 'height': height,
 256             } for url in traverse_obj(play_url, (..., 'src'), expected_type=url_or_none, default=[]) if url]
 257
 258         download_url = url_or_none(video_info.get('downloadAddr')) or traverse_obj(video_info, ('download', 'url'), expected_type=url_or_none)
 259         if download_url:
 260             formats.append({
 261                 'format_id': 'download',
 262                 'url': self._proto_relative_url(download_url),
 263                 'ext': 'mp4',
 264                 'width': width,
 265                 'height': height,
 266             })
 267         self._remove_duplicate_formats(formats)
 268         self._sort_formats(formats)
 269
 270         thumbnails = []
 271         for thumbnail_name in ('thumbnail', 'cover', 'dynamicCover', 'originCover'):
 272             if aweme_detail.get(thumbnail_name):
 273                 thumbnails = [{
 274                     'url': self._proto_relative_url(aweme_detail[thumbnail_name]),
 275                     'width': width,
 276                     'height': height
 277                 }]
 278
 279         return {
 280             'id': traverse_obj(aweme_detail, 'id', 'awemeId', expected_type=str_or_none),
 281             'title': aweme_detail.get('desc'),
 282             'duration': try_get(aweme_detail, lambda x: x['video']['duration'], int),
 283             'view_count': int_or_none(stats_info.get('playCount')),
 284             'like_count': int_or_none(stats_info.get('diggCount')),
 285             'repost_count': int_or_none(stats_info.get('shareCount')),
 286             'comment_count': int_or_none(stats_info.get('commentCount')),
 287             'timestamp': int_or_none(aweme_detail.get('createTime')),
 288             'creator': str_or_none(author_info.get('nickname')),
 289             'uploader': str_or_none(author_info.get('uniqueId') or aweme_detail.get('author')),
 290             'uploader_id': str_or_none(author_info.get('id') or aweme_detail.get('authorId')),
 291             'uploader_url': user_url,
 292             'track': str_or_none(music_info.get('title')),
 293             'album': str_or_none(music_info.get('album')) or None,
 294             'artist': str_or_none(music_info.get('authorName')),
 295             'formats': formats,
 296             'thumbnails': thumbnails,
 297             'description': str_or_none(aweme_detail.get('desc')),
 298             'http_headers': {
 299                 'Referer': webpage_url
 300             }
 301         }
 302
 303
 304 class TikTokIE(TikTokBaseIE):
 305     _VALID_URL = r'https?://www\.tiktok\.com/@[\w\.-]+/video/(?P<id>\d+)'
 306
 307     _TESTS = [{
 308         'url': 'https://www.tiktok.com/@leenabhushan/video/6748451240264420610',
 309         'md5': '736bb7a466c6f0a6afeb597da1e6f5b7',
 310         'info_dict': {
 311             'id': '6748451240264420610',
 312             'ext': 'mp4',
 313             'title': '#jassmanak #lehanga #leenabhushan',
 314             'description': '#jassmanak #lehanga #leenabhushan',
 315             'duration': 13,
 316             'height': 1024,
 317             'width': 576,
 318             'uploader': 'leenabhushan',
 319             'uploader_id': '6691488002098119685',
 320             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA_Eb4t1vodM1IuTy_cvp9CY22RAb59xqrO0Xtz9CYQJvgXaDvZxYnZYRzDWhhgJmy',
 321             'creator': 'facestoriesbyleenabh',
 322             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 323             'upload_date': '20191016',
 324             'timestamp': 1571246252,
 325             'view_count': int,
 326             'like_count': int,
 327             'repost_count': int,
 328             'comment_count': int,
 329         }
 330     }, {
 331         'url': 'https://www.tiktok.com/@patroxofficial/video/6742501081818877190?langCountry=en',
 332         'md5': '6f3cf8cdd9b28cb8363fe0a9a160695b',
 333         'info_dict': {
 334             'id': '6742501081818877190',
 335             'ext': 'mp4',
 336             'title': 'md5:5e2a23877420bb85ce6521dbee39ba94',
 337             'description': 'md5:5e2a23877420bb85ce6521dbee39ba94',
 338             'duration': 27,
 339             'height': 960,
 340             'width': 540,
 341             'uploader': 'patrox',
 342             'uploader_id': '18702747',
 343             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAiFnldaILebi5heDoVU6bn4jBWWycX6-9U3xuNPqZ8Ws',
 344             'creator': 'patroX',
 345             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 346             'upload_date': '20190930',
 347             'timestamp': 1569860870,
 348             'view_count': int,
 349             'like_count': int,
 350             'repost_count': int,
 351             'comment_count': int,
 352         }
 353     }, {
 354         # Banned audio, only available on the app
 355         'url': 'https://www.tiktok.com/@barudakhb_/video/6984138651336838402',
 356         'info_dict': {
 357             'id': '6984138651336838402',
 358             'ext': 'mp4',
 359             'title': 'Balas @yolaaftwsr hayu yu ? #SquadRandom_ 🔥',
 360             'description': 'Balas @yolaaftwsr hayu yu ? #SquadRandom_ 🔥',
 361             'uploader': 'barudakhb_',
 362             'creator': 'md5:29f238c49bc0c176cb3cef1a9cea9fa6',
 363             'uploader_id': '6974687867511718913',
 364             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAbhBwQC-R1iKoix6jDFsF-vBdfx2ABoDjaZrM9fX6arU3w71q3cOWgWuTXn1soZ7d',
 365             'track': 'Boka Dance',
 366             'artist': 'md5:29f238c49bc0c176cb3cef1a9cea9fa6',
 367             'timestamp': 1626121503,
 368             'duration': 18,
 369             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 370             'upload_date': '20210712',
 371             'view_count': int,
 372             'like_count': int,
 373             'repost_count': int,
 374             'comment_count': int,
 375         }
 376     }, {
 377         # Sponsored video, only available with feed workaround
 378         'url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_/video/7042692929109986561',
 379         'info_dict': {
 380             'id': '7042692929109986561',
 381             'ext': 'mp4',
 382             'title': 'Slap and Run!',
 383             'description': 'Slap and Run!',
 384             'uploader': 'user440922249',
 385             'creator': 'Slap And Run',
 386             'uploader_id': '7036055384943690754',
 387             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_',
 388             'track': 'Promoted Music',
 389             'timestamp': 1639754738,
 390             'duration': 30,
 391             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
 392             'upload_date': '20211217',
 393             'view_count': int,
 394             'like_count': int,
 395             'repost_count': int,
 396             'comment_count': int,
 397         },
 398         'expected_warnings': ['Video not available']
 399     }]
 400
 401     def _extract_aweme_app(self, aweme_id):
 402         try:
 403             aweme_detail = self._call_api('aweme/detail', {'aweme_id': aweme_id}, aweme_id,
 404                                           note='Downloading video details', errnote='Unable to download video details').get('aweme_detail')
 405             if not aweme_detail:
 406                 raise ExtractorError('Video not available', video_id=aweme_id)
 407         except ExtractorError as e:
 408             self.report_warning(f'{e}; Retrying with feed workaround')
 409             feed_list = self._call_api('feed', {'aweme_id': aweme_id}, aweme_id,
 410                                        note='Downloading video feed', errnote='Unable to download video feed').get('aweme_list') or []
 411             aweme_detail = next((aweme for aweme in feed_list if str(aweme.get('aweme_id')) == aweme_id), None)
 412             if not aweme_detail:
 413                 raise ExtractorError('Unable to find video in feed', video_id=aweme_id)
 414         return self._parse_aweme_video_app(aweme_detail)
 415
 416     def _real_extract(self, url):
 417         video_id = self._match_id(url)
 418
 419         try:
 420             return self._extract_aweme_app(video_id)
 421         except ExtractorError as e:
 422             self.report_warning(f'{e}; Retrying with webpage')
 423
 424         # If we only call once, we get a 403 when downlaoding the video.
 425         self._download_webpage(url, video_id)
 426         webpage = self._download_webpage(url, video_id, note='Downloading video webpage')
 427         next_json = self._search_regex(
 428             r'id=\"__NEXT_DATA__\"\s+type=\"application\/json\"\s*[^>]+>\s*(?P<next_data>[^<]+)',
 429             webpage, 'next data', group='next_data', default=None)
 430
 431         if next_json:
 432             next_data = self._parse_json(next_json, video_id)
 433             status = traverse_obj(next_data, ('props', 'pageProps', 'statusCode'), expected_type=int) or 0
 434             video_data = traverse_obj(next_data, ('props', 'pageProps', 'itemInfo', 'itemStruct'), expected_type=dict)
 435         else:
 436             sigi_json = self._search_regex(
 437                 r'>\s*window\[[\'"]SIGI_STATE[\'"]\]\s*=\s*(?P<sigi_state>{.+});',
 438                 webpage, 'sigi data', group='sigi_state')
 439             sigi_data = self._parse_json(sigi_json, video_id)
 440             status = traverse_obj(sigi_data, ('VideoPage', 'statusCode'), expected_type=int) or 0
 441             video_data = traverse_obj(sigi_data, ('ItemModule', video_id), expected_type=dict)
 442
 443         if status == 0:
 444             return self._parse_aweme_video_web(video_data, url)
 445         elif status == 10216:
 446             raise ExtractorError('This video is private', expected=True)
 447         raise ExtractorError('Video not available', video_id=video_id)
 448
 449
 450 class TikTokUserIE(TikTokBaseIE):
 451     IE_NAME = 'tiktok:user'
 452     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/@(?P<id>[\w\.-]+)/?(?:$|[#?])'
 453     _TESTS = [{
 454         'url': 'https://tiktok.com/@corgibobaa?lang=en',
 455         'playlist_mincount': 45,
 456         'info_dict': {
 457             'id': '6935371178089399301',
 458             'title': 'corgibobaa',
 459             'thumbnail': r're:https://.+_1080x1080\.webp'
 460         },
 461         'expected_warnings': ['Retrying']
 462     }, {
 463         'url': 'https://www.tiktok.com/@meme',
 464         'playlist_mincount': 593,
 465         'info_dict': {
 466             'id': '79005827461758976',
 467             'title': 'meme',
 468             'thumbnail': r're:https://.+_1080x1080\.webp'
 469         },
 470         'expected_warnings': ['Retrying']
 471     }]
 472
 473     r'''  # TODO: Fix by adding _signature to api_url
 474     def _entries(self, webpage, user_id, username):
 475         secuid = self._search_regex(r'\"secUid\":\"(?P<secUid>[^\"]+)', webpage, username)
 476         verifyfp_cookie = self._get_cookies('https://www.tiktok.com').get('s_v_web_id')
 477         if not verifyfp_cookie:
 478             raise ExtractorError('Improper cookies (missing s_v_web_id).', expected=True)
 479         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='
 480         cursor = '0'
 481         for page in itertools.count():
 482             data_json = self._download_json(api_url + cursor, username, note='Downloading Page %d' % page)
 483             for video in data_json.get('itemList', []):
 484                 video_id = video['id']
 485                 video_url = f'https://www.tiktok.com/@{user_id}/video/{video_id}'
 486                 yield self._url_result(video_url, 'TikTok', video_id, str_or_none(video.get('desc')))
 487             if not data_json.get('hasMore'):
 488                 break
 489             cursor = data_json['cursor']
 490     '''
 491
 492     def _video_entries_api(self, webpage, user_id, username):
 493         query = {
 494             'user_id': user_id,
 495             'count': 21,
 496             'max_cursor': 0,
 497             'min_cursor': 0,
 498             'retry_type': 'no_retry',
 499             '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.
 500         }
 501
 502         max_retries = self.get_param('extractor_retries', 3)
 503         for page in itertools.count(1):
 504             for retries in itertools.count():
 505                 try:
 506                     post_list = self._call_api('aweme/post', query, username,
 507                                                note='Downloading user video list page %d%s' % (page, f' (attempt {retries})' if retries != 0 else ''),
 508                                                errnote='Unable to download user video list')
 509                 except ExtractorError as e:
 510                     if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0 and retries != max_retries:
 511                         self.report_warning('%s. Retrying...' % str(e.cause or e.msg))
 512                         continue
 513                     raise
 514                 break
 515             yield from post_list.get('aweme_list', [])
 516             if not post_list.get('has_more'):
 517                 break
 518             query['max_cursor'] = post_list['max_cursor']
 519
 520     def _entries_api(self, user_id, videos):
 521         for video in videos:
 522             yield {
 523                 **self._parse_aweme_video_app(video),
 524                 'extractor_key': TikTokIE.ie_key(),
 525                 'extractor': 'TikTok',
 526                 'webpage_url': f'https://tiktok.com/@{user_id}/video/{video["aweme_id"]}',
 527             }
 528
 529     def _real_extract(self, url):
 530         user_name = self._match_id(url)
 531         webpage = self._download_webpage(url, user_name, headers={
 532             'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)'
 533         })
 534         user_id = self._html_search_regex(r'snssdk\d*://user/profile/(\d+)', webpage, 'user ID')
 535
 536         videos = LazyList(self._video_entries_api(webpage, user_id, user_name))
 537         thumbnail = traverse_obj(videos, (0, 'author', 'avatar_larger', 'url_list', 0))
 538
 539         return self.playlist_result(self._entries_api(user_id, videos), user_id, user_name, thumbnail=thumbnail)
 540
 541
 542 class TikTokBaseListIE(TikTokBaseIE):
 543     def _entries(self, list_id, display_id):
 544         query = {
 545             self._QUERY_NAME: list_id,
 546             'cursor': 0,
 547             'count': 20,
 548             'type': 5,
 549             'device_id': ''.join(random.choice(string.digits) for i in range(19))
 550         }
 551
 552         max_retries = self.get_param('extractor_retries', 3)
 553         for page in itertools.count(1):
 554             for retries in itertools.count():
 555                 try:
 556                     post_list = self._call_api(self._API_ENDPOINT, query, display_id,
 557                                                note='Downloading video list page %d%s' % (page, f' (attempt {retries})' if retries != 0 else ''),
 558                                                errnote='Unable to download video list')
 559                 except ExtractorError as e:
 560                     if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0 and retries != max_retries:
 561                         self.report_warning('%s. Retrying...' % str(e.cause or e.msg))
 562                         continue
 563                     raise
 564                 break
 565             for video in post_list.get('aweme_list', []):
 566                 yield {
 567                     **self._parse_aweme_video_app(video),
 568                     'extractor_key': TikTokIE.ie_key(),
 569                     'extractor': 'TikTok',
 570                     'webpage_url': f'https://tiktok.com/@_/video/{video["aweme_id"]}',
 571                 }
 572             if not post_list.get('has_more'):
 573                 break
 574             query['cursor'] = post_list['cursor']
 575
 576     def _real_extract(self, url):
 577         list_id = self._match_id(url)
 578         return self.playlist_result(self._entries(list_id, list_id), list_id)
 579
 580
 581 class TikTokSoundIE(TikTokBaseListIE):
 582     IE_NAME = 'tiktok:sound'
 583     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/music/[\w\.-]+-(?P<id>[\d]+)[/?#&]?'
 584     _QUERY_NAME = 'music_id'
 585     _API_ENDPOINT = 'music/aweme'
 586     _TESTS = [{
 587         'url': 'https://www.tiktok.com/music/Build-a-Btch-6956990112127585029?lang=en',
 588         'playlist_mincount': 100,
 589         'info_dict': {
 590             'id': '6956990112127585029'
 591         },
 592         'expected_warnings': ['Retrying']
 593     }, {
 594         # Actual entries are less than listed video count
 595         'url': 'https://www.tiktok.com/music/jiefei-soap-remix-7036843036118469381',
 596         'playlist_mincount': 2182,
 597         'info_dict': {
 598             'id': '7036843036118469381'
 599         },
 600         'expected_warnings': ['Retrying']
 601     }]
 602
 603
 604 class TikTokEffectIE(TikTokBaseListIE):
 605     IE_NAME = 'tiktok:effect'
 606     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/sticker/[\w\.-]+-(?P<id>[\d]+)[/?#&]?'
 607     _QUERY_NAME = 'sticker_id'
 608     _API_ENDPOINT = 'sticker/aweme'
 609     _TESTS = [{
 610         'url': 'https://www.tiktok.com/sticker/MATERIAL-GWOOORL-1258156',
 611         'playlist_mincount': 100,
 612         'info_dict': {
 613             'id': '1258156',
 614         },
 615         'expected_warnings': ['Retrying']
 616     }, {
 617         # Different entries between mobile and web, depending on region
 618         'url': 'https://www.tiktok.com/sticker/Elf-Friend-479565',
 619         'only_matching': True
 620     }]
 621
 622
 623 class TikTokTagIE(TikTokBaseListIE):
 624     IE_NAME = 'tiktok:tag'
 625     _VALID_URL = r'https?://(?:www\.)?tiktok\.com/tag/(?P<id>[^/?#&]+)'
 626     _QUERY_NAME = 'ch_id'
 627     _API_ENDPOINT = 'challenge/aweme'
 628     _TESTS = [{
 629         'url': 'https://tiktok.com/tag/hello2018',
 630         'playlist_mincount': 39,
 631         'info_dict': {
 632             'id': '46294678',
 633             'title': 'hello2018',
 634         },
 635         'expected_warnings': ['Retrying']
 636     }, {
 637         'url': 'https://tiktok.com/tag/fypシ?is_copy_url=0&is_from_webapp=v1',
 638         'only_matching': True
 639     }]
 640
 641     def _real_extract(self, url):
 642         display_id = self._match_id(url)
 643         webpage = self._download_webpage(url, display_id, headers={
 644             'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)'
 645         })
 646         tag_id = self._html_search_regex(r'snssdk\d*://challenge/detail/(\d+)', webpage, 'tag ID')
 647         return self.playlist_result(self._entries(tag_id, display_id), tag_id, display_id)
 648
 649
 650 class DouyinIE(TikTokIE):
 651     _VALID_URL = r'https?://(?:www\.)?douyin\.com/video/(?P<id>[0-9]+)'
 652     _TESTS = [{
 653         'url': 'https://www.douyin.com/video/6961737553342991651',
 654         'md5': '10523312c8b8100f353620ac9dc8f067',
 655         'info_dict': {
 656             'id': '6961737553342991651',
 657             'ext': 'mp4',
 658             'title': '#杨超越  小小水手带你去远航❤️',
 659             'uploader': '杨超越',
 660             'upload_date': '20210513',
 661             'timestamp': 1620905839,
 662             'uploader_id': '110403406559',
 663             'view_count': int,
 664             'like_count': int,
 665             'repost_count': int,
 666             'comment_count': int,
 667         }
 668     }, {
 669         'url': 'https://www.douyin.com/video/6982497745948921092',
 670         'md5': 'd78408c984b9b5102904cf6b6bc2d712',
 671         'info_dict': {
 672             'id': '6982497745948921092',
 673             'ext': 'mp4',
 674             'title': '这个夏日和小羊@杨超越 一起遇见白色幻想',
 675             'uploader': '杨超越工作室',
 676             'upload_date': '20210708',
 677             'timestamp': 1625739481,
 678             'uploader_id': '408654318141572',
 679             'view_count': int,
 680             'like_count': int,
 681             'repost_count': int,
 682             'comment_count': int,
 683         }
 684     }, {
 685         'url': 'https://www.douyin.com/video/6953975910773099811',
 686         'md5': '72e882e24f75064c218b76c8b713c185',
 687         'info_dict': {
 688             'id': '6953975910773099811',
 689             'ext': 'mp4',
 690             'title': '#一起看海  出现在你的夏日里',
 691             'uploader': '杨超越',
 692             'upload_date': '20210422',
 693             'timestamp': 1619098692,
 694             'uploader_id': '110403406559',
 695             'view_count': int,
 696             'like_count': int,
 697             'repost_count': int,
 698             'comment_count': int,
 699         }
 700     }, {
 701         'url': 'https://www.douyin.com/video/6950251282489675042',
 702         'md5': 'b4db86aec367ef810ddd38b1737d2fed',
 703         'info_dict': {
 704             'id': '6950251282489675042',
 705             'ext': 'mp4',
 706             'title': '哈哈哈，成功了哈哈哈哈哈哈',
 707             'uploader': '杨超越',
 708             'upload_date': '20210412',
 709             'timestamp': 1618231483,
 710             'uploader_id': '110403406559',
 711             'view_count': int,
 712             'like_count': int,
 713             'repost_count': int,
 714             'comment_count': int,
 715         }
 716     }, {
 717         'url': 'https://www.douyin.com/video/6963263655114722595',
 718         'md5': '1abe1c477d05ee62efb40bf2329957cf',
 719         'info_dict': {
 720             'id': '6963263655114722595',
 721             'ext': 'mp4',
 722             'title': '#哪个爱豆的105度最甜 换个角度看看我哈哈',
 723             'uploader': '杨超越',
 724             'upload_date': '20210517',
 725             'timestamp': 1621261163,
 726             'uploader_id': '110403406559',
 727             'view_count': int,
 728             'like_count': int,
 729             'repost_count': int,
 730             'comment_count': int,
 731         }
 732     }]
 733     _APP_VERSION = '9.6.0'
 734     _MANIFEST_APP_VERSION = '960'
 735     _APP_NAME = 'aweme'
 736     _AID = 1128
 737     _API_HOSTNAME = 'aweme.snssdk.com'
 738     _UPLOADER_URL_FORMAT = 'https://www.douyin.com/user/%s'
 739     _WEBPAGE_HOST = 'https://www.douyin.com/'
 740
 741     def _real_extract(self, url):
 742         video_id = self._match_id(url)
 743
 744         try:
 745             return self._extract_aweme_app(video_id)
 746         except ExtractorError as e:
 747             self.report_warning(f'{e}; Retrying with webpage')
 748
 749         webpage = self._download_webpage(url, video_id)
 750         render_data_json = self._search_regex(
 751             r'<script [^>]*\bid=[\'"]RENDER_DATA[\'"][^>]*>(%7B.+%7D)</script>',
 752             webpage, 'render data', default=None)
 753         if not render_data_json:
 754             # TODO: Run verification challenge code to generate signature cookies
 755             raise ExtractorError('Fresh cookies (not necessarily logged in) are needed')
 756
 757         render_data = self._parse_json(
 758             render_data_json, video_id, transform_source=compat_urllib_parse_unquote)
 759         return self._parse_aweme_video_web(
 760             traverse_obj(render_data, (..., 'aweme', 'detail'), get_all=False), url)