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