X-Git-Url: https://jfr.im/git/yt-dlp.git/blobdiff_plain/0fd6661edb7e671eb7b131de12fa89bb85a6cbaf..4b9353239ee921440442b74f467fae298a284907:/yt_dlp/extractor/tiktok.py diff --git a/yt_dlp/extractor/tiktok.py b/yt_dlp/extractor/tiktok.py index 4b0efd4a3..c34235e96 100644 --- a/yt_dlp/extractor/tiktok.py +++ b/yt_dlp/extractor/tiktok.py @@ -8,12 +8,15 @@ import json from .common import InfoExtractor +from ..compat import compat_urllib_parse_unquote from ..utils import ( ExtractorError, int_or_none, + join_nonempty, str_or_none, traverse_obj, try_get, + url_or_none, qualities, ) @@ -21,6 +24,11 @@ class TikTokBaseIE(InfoExtractor): _APP_VERSION = '20.9.3' _MANIFEST_APP_VERSION = '291' + _APP_NAME = 'trill' + _AID = 1180 + _API_HOSTNAME = 'api-h2.tiktokv.com' + _UPLOADER_URL_FORMAT = 'https://www.tiktok.com/@%s' + _WEBPAGE_HOST = 'https://www.tiktok.com/' QUALITIES = ('360p', '540p', '720p') def _call_api(self, ep, query, video_id, fatal=True, @@ -32,8 +40,8 @@ def _call_api(self, ep, query, video_id, fatal=True, 'build_number': self._APP_VERSION, 'manifest_version_code': self._MANIFEST_APP_VERSION, 'update_version_code': self._MANIFEST_APP_VERSION, - 'openudid': ''.join(random.choice('0123456789abcdef') for i in range(16)), - 'uuid': ''.join([random.choice(string.digits) for num in range(16)]), + 'openudid': ''.join(random.choice('0123456789abcdef') for _ in range(16)), + 'uuid': ''.join([random.choice(string.digits) for _ in range(16)]), '_rticket': int(time.time() * 1000), 'ts': int(time.time()), 'device_brand': 'Google', @@ -46,7 +54,7 @@ def _call_api(self, ep, query, video_id, fatal=True, 'carrier_region': 'US', 'sys_region': 'US', 'region': 'US', - 'app_name': 'trill', + 'app_name': self._APP_NAME, 'app_language': 'en', 'language': 'en', 'timezone_name': 'America/New_York', @@ -55,20 +63,23 @@ def _call_api(self, ep, query, video_id, fatal=True, 'ac': 'wifi', 'mcc_mnc': '310260', 'is_my_cn': 0, - 'aid': 1180, + 'aid': self._AID, 'ssmix': 'a', 'as': 'a1qwert123', 'cp': 'cbfhckdckkde1', } - self._set_cookie('.tiktokv.com', 'odin_tt', ''.join(random.choice('0123456789abcdef') for i in range(160))) + self._set_cookie(self._API_HOSTNAME, 'odin_tt', ''.join(random.choice('0123456789abcdef') for _ in range(160))) + webpage_cookies = self._get_cookies(self._WEBPAGE_HOST) + if webpage_cookies.get('sid_tt'): + self._set_cookie(self._API_HOSTNAME, 'sid_tt', webpage_cookies['sid_tt'].value) return self._download_json( - 'https://api-t2.tiktokv.com/aweme/v1/%s/' % ep, video_id=video_id, + 'https://%s/aweme/v1/%s/' % (self._API_HOSTNAME, ep), video_id=video_id, fatal=fatal, note=note, errnote=errnote, headers={ '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)', 'Accept': 'application/json', }, query=real_query) - def _parse_aweme_video(self, aweme_detail): + def _parse_aweme_video_app(self, aweme_detail): aweme_id = aweme_detail['aweme_id'] video_info = aweme_detail['video'] @@ -101,8 +112,8 @@ def extract_addr(addr, add_meta={}): 'acodec': 'aac', 'source_preference': -2 if 'aweme/v1' in url else -1, # Downloads from API might get blocked **add_meta, **parsed_meta, - 'format_note': ' '.join(filter(None, ( - add_meta.get('format_note'), '(API)' if 'aweme/v1' in url else ''))) + 'format_note': join_nonempty( + add_meta.get('format_note'), '(API)' if 'aweme/v1' in url else None, delim=' ') } for url in addr.get('url_list') or []] # Hack: Add direct video links first to prioritize them when removing duplicate formats @@ -146,6 +157,7 @@ def extract_addr(addr, add_meta={}): 'tbr': try_get(bitrate, lambda x: x['bit_rate'] / 1000), 'vcodec': 'h265' if traverse_obj( bitrate, 'is_bytevc1', 'is_h265') else 'h264', + 'fps': bitrate.get('FPS'), })) self._remove_duplicate_formats(formats) @@ -165,7 +177,10 @@ def extract_addr(addr, add_meta={}): stats_info = aweme_detail.get('statistics', {}) author_info = aweme_detail.get('author', {}) music_info = aweme_detail.get('music', {}) - user_id = str_or_none(author_info.get('nickname')) + user_url = self._UPLOADER_URL_FORMAT % (traverse_obj(author_info, + 'sec_uid', 'id', 'uid', 'unique_id', + expected_type=str_or_none, get_all=False)) + labels = traverse_obj(aweme_detail, ('hybrid_label', ..., 'text'), expected_type=str, default=[]) contained_music_track = traverse_obj( music_info, ('matched_song', 'title'), ('matched_pgc_sound', 'title'), expected_type=str) @@ -187,16 +202,93 @@ def extract_addr(addr, add_meta={}): 'repost_count': int_or_none(stats_info.get('share_count')), 'comment_count': int_or_none(stats_info.get('comment_count')), 'uploader': str_or_none(author_info.get('unique_id')), - 'creator': user_id, + 'creator': str_or_none(author_info.get('nickname')), 'uploader_id': str_or_none(author_info.get('uid')), - 'uploader_url': f'https://www.tiktok.com/@{user_id}' if user_id else None, + 'uploader_url': user_url, 'track': music_track, 'album': str_or_none(music_info.get('album')) or None, 'artist': music_author, 'timestamp': int_or_none(aweme_detail.get('create_time')), 'formats': formats, 'thumbnails': thumbnails, - 'duration': int_or_none(traverse_obj(video_info, 'duration', ('download_addr', 'duration')), scale=1000) + 'duration': int_or_none(traverse_obj(video_info, 'duration', ('download_addr', 'duration')), scale=1000), + 'availability': self._availability( + is_private='Private' in labels, + needs_subscription='Friends only' in labels, + is_unlisted='Followers only' in labels) + } + + def _parse_aweme_video_web(self, aweme_detail, webpage_url): + video_info = aweme_detail['video'] + author_info = traverse_obj(aweme_detail, 'author', 'authorInfo', default={}) + music_info = aweme_detail.get('music') or {} + stats_info = aweme_detail.get('stats') or {} + user_url = self._UPLOADER_URL_FORMAT % (traverse_obj(author_info, + 'secUid', 'id', 'uid', 'uniqueId', + expected_type=str_or_none, get_all=False)) + + formats = [] + play_url = video_info.get('playAddr') + width = video_info.get('width') + height = video_info.get('height') + if isinstance(play_url, str): + formats = [{ + 'url': self._proto_relative_url(play_url), + 'ext': 'mp4', + 'width': width, + 'height': height, + }] + elif isinstance(play_url, list): + formats = [{ + 'url': self._proto_relative_url(url), + 'ext': 'mp4', + 'width': width, + 'height': height, + } for url in traverse_obj(play_url, (..., 'src'), expected_type=url_or_none, default=[]) if url] + + download_url = url_or_none(video_info.get('downloadAddr')) or traverse_obj(video_info, ('download', 'url'), expected_type=url_or_none) + if download_url: + formats.append({ + 'format_id': 'download', + 'url': self._proto_relative_url(download_url), + 'ext': 'mp4', + 'width': width, + 'height': height, + }) + self._remove_duplicate_formats(formats) + self._sort_formats(formats) + + thumbnails = [] + for thumbnail_name in ('thumbnail', 'cover', 'dynamicCover', 'originCover'): + if aweme_detail.get(thumbnail_name): + thumbnails = [{ + 'url': self._proto_relative_url(aweme_detail[thumbnail_name]), + 'width': width, + 'height': height + }] + + return { + 'id': traverse_obj(aweme_detail, 'id', 'awemeId', expected_type=str_or_none), + 'title': aweme_detail.get('desc'), + 'duration': try_get(aweme_detail, lambda x: x['video']['duration'], int), + 'view_count': int_or_none(stats_info.get('playCount')), + 'like_count': int_or_none(stats_info.get('diggCount')), + 'repost_count': int_or_none(stats_info.get('shareCount')), + 'comment_count': int_or_none(stats_info.get('commentCount')), + 'timestamp': int_or_none(aweme_detail.get('createTime')), + 'creator': str_or_none(author_info.get('nickname')), + 'uploader': str_or_none(author_info.get('uniqueId')), + 'uploader_id': str_or_none(author_info.get('id')), + 'uploader_url': user_url, + 'track': str_or_none(music_info.get('title')), + 'album': str_or_none(music_info.get('album')) or None, + 'artist': str_or_none(music_info.get('authorName')), + 'formats': formats, + 'thumbnails': thumbnails, + 'description': str_or_none(aweme_detail.get('desc')), + 'http_headers': { + 'Referer': webpage_url + } } @@ -205,18 +297,18 @@ class TikTokIE(TikTokBaseIE): _TESTS = [{ 'url': 'https://www.tiktok.com/@leenabhushan/video/6748451240264420610', - 'md5': '34a7543afd5a151b0840ba6736fb633b', + 'md5': '736bb7a466c6f0a6afeb597da1e6f5b7', 'info_dict': { 'id': '6748451240264420610', 'ext': 'mp4', 'title': '#jassmanak #lehanga #leenabhushan', 'description': '#jassmanak #lehanga #leenabhushan', 'duration': 13, - 'height': 1280, - 'width': 720, + 'height': 1024, + 'width': 576, 'uploader': 'leenabhushan', 'uploader_id': '6691488002098119685', - 'uploader_url': 'https://www.tiktok.com/@leenabhushan', + 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA_Eb4t1vodM1IuTy_cvp9CY22RAb59xqrO0Xtz9CYQJvgXaDvZxYnZYRzDWhhgJmy', 'creator': 'facestoriesbyleenabh', 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?', 'upload_date': '20191016', @@ -228,7 +320,7 @@ class TikTokIE(TikTokBaseIE): } }, { 'url': 'https://www.tiktok.com/@patroxofficial/video/6742501081818877190?langCountry=en', - 'md5': '06b9800d47d5fe51a19e322dd86e61c9', + 'md5': '6f3cf8cdd9b28cb8363fe0a9a160695b', 'info_dict': { 'id': '6742501081818877190', 'ext': 'mp4', @@ -239,7 +331,7 @@ class TikTokIE(TikTokBaseIE): 'width': 540, 'uploader': 'patrox', 'uploader_id': '18702747', - 'uploader_url': 'https://www.tiktok.com/@patrox', + 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAiFnldaILebi5heDoVU6bn4jBWWycX6-9U3xuNPqZ8Ws', 'creator': 'patroX', 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?', 'upload_date': '20190930', @@ -255,60 +347,20 @@ class TikTokIE(TikTokBaseIE): 'only_matching': True, }] - def _extract_aweme(self, props_data, webpage, url): - video_info = try_get( - props_data, lambda x: x['pageProps']['itemInfo']['itemStruct'], dict) - author_info = try_get( - props_data, lambda x: x['pageProps']['itemInfo']['itemStruct']['author'], dict) or {} - music_info = try_get( - props_data, lambda x: x['pageProps']['itemInfo']['itemStruct']['music'], dict) or {} - stats_info = try_get(props_data, lambda x: x['pageProps']['itemInfo']['itemStruct']['stats'], dict) or {} - - user_id = str_or_none(author_info.get('uniqueId')) - download_url = try_get(video_info, (lambda x: x['video']['playAddr'], - lambda x: x['video']['downloadAddr'])) - height = try_get(video_info, lambda x: x['video']['height'], int) - width = try_get(video_info, lambda x: x['video']['width'], int) - thumbnails = [{ - 'url': video_info.get('thumbnail') or self._og_search_thumbnail(webpage), - 'width': width, - 'height': height - }] - tracker = try_get(props_data, lambda x: x['initialProps']['$wid']) - - return { - 'id': str_or_none(video_info.get('id')), - 'url': download_url, - 'ext': 'mp4', - 'height': height, - 'width': width, - 'title': video_info.get('desc') or self._og_search_title(webpage), - 'duration': try_get(video_info, lambda x: x['video']['duration'], int), - 'view_count': int_or_none(stats_info.get('playCount')), - 'like_count': int_or_none(stats_info.get('diggCount')), - 'repost_count': int_or_none(stats_info.get('shareCount')), - 'comment_count': int_or_none(stats_info.get('commentCount')), - 'timestamp': try_get(video_info, lambda x: int(x['createTime']), int), - 'creator': str_or_none(author_info.get('nickname')), - 'uploader': user_id, - 'uploader_id': str_or_none(author_info.get('id')), - 'uploader_url': f'https://www.tiktok.com/@{user_id}', - 'track': str_or_none(music_info.get('title')), - 'album': str_or_none(music_info.get('album')) or None, - 'artist': str_or_none(music_info.get('authorName')), - 'thumbnails': thumbnails, - 'description': str_or_none(video_info.get('desc')), - 'webpage_url': self._og_search_url(webpage), - 'http_headers': { - 'Referer': url, - 'Cookie': 'tt_webid=%s; tt_webid_v2=%s' % (tracker, tracker), - } - } - def _extract_aweme_app(self, aweme_id): - aweme_detail = self._call_api('aweme/detail', {'aweme_id': aweme_id}, aweme_id, - note='Downloading video details', errnote='Unable to download video details')['aweme_detail'] - return self._parse_aweme_video(aweme_detail) + try: + aweme_detail = self._call_api('aweme/detail', {'aweme_id': aweme_id}, aweme_id, + note='Downloading video details', errnote='Unable to download video details').get('aweme_detail') + if not aweme_detail: + raise ExtractorError('Video not available', video_id=aweme_id) + except ExtractorError as e: + self.report_warning(f'{e}; Retrying with feed workaround') + feed_list = self._call_api('feed', {'aweme_id': aweme_id}, aweme_id, + note='Downloading video feed', errnote='Unable to download video feed').get('aweme_list') or [] + aweme_detail = next(aweme for aweme in feed_list if str(aweme.get('aweme_id')) == aweme_id) + if not aweme_detail: + raise ExtractorError('Unable to find video in feed', video_id=aweme_id) + return self._parse_aweme_video_app(aweme_detail) def _real_extract(self, url): video_id = self._match_id(url) @@ -330,7 +382,7 @@ def _real_extract(self, url): # Chech statusCode for success status = props_data.get('pageProps').get('statusCode') if status == 0: - return self._extract_aweme(props_data, webpage, url) + return self._parse_aweme_video_web(props_data['pageProps']['itemInfo']['itemStruct'], url) elif status == 10216: raise ExtractorError('This video is private', expected=True) @@ -345,13 +397,17 @@ class TikTokUserIE(TikTokBaseIE): 'playlist_mincount': 45, 'info_dict': { 'id': '6935371178089399301', + 'title': 'corgibobaa', }, + 'expected_warnings': ['Retrying'] }, { 'url': 'https://www.tiktok.com/@meme', 'playlist_mincount': 593, 'info_dict': { 'id': '79005827461758976', + 'title': 'meme', }, + 'expected_warnings': ['Retrying'] }] r''' # TODO: Fix by adding _signature to api_url @@ -380,7 +436,7 @@ def _entries_api(self, webpage, user_id, username): 'max_cursor': 0, 'min_cursor': 0, 'retry_type': 'no_retry', - 'device_id': ''.join(random.choice(string.digits) for i in range(19)), # Some endpoints don't like randomized device_id, so it isn't directly set in _call_api. + '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. } max_retries = self.get_param('extractor_retries', 3) @@ -398,18 +454,240 @@ def _entries_api(self, webpage, user_id, username): break for video in post_list.get('aweme_list', []): yield { - **self._parse_aweme_video(video), - 'ie_key': TikTokIE.ie_key(), + **self._parse_aweme_video_app(video), + 'extractor_key': TikTokIE.ie_key(), 'extractor': 'TikTok', + 'webpage_url': f'https://tiktok.com/@{user_id}/video/{video["aweme_id"]}', } if not post_list.get('has_more'): break query['max_cursor'] = post_list['max_cursor'] def _real_extract(self, url): - user_id = self._match_id(url) - webpage = self._download_webpage(url, user_id, headers={ + user_name = self._match_id(url) + webpage = self._download_webpage(url, user_name, headers={ + 'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)' + }) + user_id = self._html_search_regex(r'snssdk\d*://user/profile/(\d+)', webpage, 'user ID') + return self.playlist_result(self._entries_api(webpage, user_id, user_name), user_id, user_name) + + +class TikTokBaseListIE(TikTokBaseIE): + def _entries(self, list_id, display_id): + query = { + self._QUERY_NAME: list_id, + 'cursor': 0, + 'count': 20, + 'type': 5, + 'device_id': ''.join(random.choice(string.digits) for i in range(19)) + } + + max_retries = self.get_param('extractor_retries', 3) + for page in itertools.count(1): + for retries in itertools.count(): + try: + post_list = self._call_api(self._API_ENDPOINT, query, display_id, + note='Downloading video list page %d%s' % (page, f' (attempt {retries})' if retries != 0 else ''), + errnote='Unable to download video list') + except ExtractorError as e: + if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0 and retries != max_retries: + self.report_warning('%s. Retrying...' % str(e.cause or e.msg)) + continue + raise + break + for video in post_list.get('aweme_list', []): + yield { + **self._parse_aweme_video_app(video), + 'extractor_key': TikTokIE.ie_key(), + 'extractor': 'TikTok', + 'webpage_url': f'https://tiktok.com/@_/video/{video["aweme_id"]}', + } + if not post_list.get('has_more'): + break + query['cursor'] = post_list['cursor'] + + def _real_extract(self, url): + list_id = self._match_id(url) + return self.playlist_result(self._entries(list_id, list_id), list_id) + + +class TikTokSoundIE(TikTokBaseListIE): + IE_NAME = 'tiktok:sound' + _VALID_URL = r'https?://(?:www\.)?tiktok\.com/music/[\w\.-]+-(?P[\d]+)[/?#&]?' + _QUERY_NAME = 'music_id' + _API_ENDPOINT = 'music/aweme' + _TESTS = [{ + 'url': 'https://www.tiktok.com/music/Build-a-Btch-6956990112127585029?lang=en', + 'playlist_mincount': 100, + 'info_dict': { + 'id': '6956990112127585029' + }, + 'expected_warnings': ['Retrying'] + }, { + # Actual entries are less than listed video count + 'url': 'https://www.tiktok.com/music/jiefei-soap-remix-7036843036118469381', + 'playlist_mincount': 2182, + 'info_dict': { + 'id': '7036843036118469381' + }, + 'expected_warnings': ['Retrying'] + }] + + +class TikTokEffectIE(TikTokBaseListIE): + IE_NAME = 'tiktok:effect' + _VALID_URL = r'https?://(?:www\.)?tiktok\.com/sticker/[\w\.-]+-(?P[\d]+)[/?#&]?' + _QUERY_NAME = 'sticker_id' + _API_ENDPOINT = 'sticker/aweme' + _TESTS = [{ + 'url': 'https://www.tiktok.com/sticker/MATERIAL-GWOOORL-1258156', + 'playlist_mincount': 100, + 'info_dict': { + 'id': '1258156', + }, + 'expected_warnings': ['Retrying'] + }, { + # Different entries between mobile and web, depending on region + 'url': 'https://www.tiktok.com/sticker/Elf-Friend-479565', + 'only_matching': True + }] + + +class TikTokTagIE(TikTokBaseListIE): + IE_NAME = 'tiktok:tag' + _VALID_URL = r'https?://(?:www\.)?tiktok\.com/tag/(?P[^/?#&]+)' + _QUERY_NAME = 'ch_id' + _API_ENDPOINT = 'challenge/aweme' + _TESTS = [{ + 'url': 'https://tiktok.com/tag/hello2018', + 'playlist_mincount': 39, + 'info_dict': { + 'id': '46294678', + 'title': 'hello2018', + }, + 'expected_warnings': ['Retrying'] + }, { + 'url': 'https://tiktok.com/tag/fypシ?is_copy_url=0&is_from_webapp=v1', + 'only_matching': True + }] + + def _real_extract(self, url): + display_id = self._match_id(url) + webpage = self._download_webpage(url, display_id, headers={ 'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)' }) - own_id = self._html_search_regex(r'snssdk\d*://user/profile/(\d+)', webpage, 'user ID') - return self.playlist_result(self._entries_api(webpage, own_id, user_id), user_id) + tag_id = self._html_search_regex(r'snssdk\d*://challenge/detail/(\d+)', webpage, 'tag ID') + return self.playlist_result(self._entries(tag_id, display_id), tag_id, display_id) + + +class DouyinIE(TikTokIE): + _VALID_URL = r'https?://(?:www\.)?douyin\.com/video/(?P[0-9]+)' + _TESTS = [{ + 'url': 'https://www.douyin.com/video/6961737553342991651', + 'md5': '10523312c8b8100f353620ac9dc8f067', + 'info_dict': { + 'id': '6961737553342991651', + 'ext': 'mp4', + 'title': '#杨超越 小小水手带你去远航❤️', + 'uploader': '杨超越', + 'upload_date': '20210513', + 'timestamp': 1620905839, + 'uploader_id': '110403406559', + 'view_count': int, + 'like_count': int, + 'repost_count': int, + 'comment_count': int, + } + }, { + 'url': 'https://www.douyin.com/video/6982497745948921092', + 'md5': 'd78408c984b9b5102904cf6b6bc2d712', + 'info_dict': { + 'id': '6982497745948921092', + 'ext': 'mp4', + 'title': '这个夏日和小羊@杨超越 一起遇见白色幻想', + 'uploader': '杨超越工作室', + 'upload_date': '20210708', + 'timestamp': 1625739481, + 'uploader_id': '408654318141572', + 'view_count': int, + 'like_count': int, + 'repost_count': int, + 'comment_count': int, + } + }, { + 'url': 'https://www.douyin.com/video/6953975910773099811', + 'md5': '72e882e24f75064c218b76c8b713c185', + 'info_dict': { + 'id': '6953975910773099811', + 'ext': 'mp4', + 'title': '#一起看海 出现在你的夏日里', + 'uploader': '杨超越', + 'upload_date': '20210422', + 'timestamp': 1619098692, + 'uploader_id': '110403406559', + 'view_count': int, + 'like_count': int, + 'repost_count': int, + 'comment_count': int, + } + }, { + 'url': 'https://www.douyin.com/video/6950251282489675042', + 'md5': 'b4db86aec367ef810ddd38b1737d2fed', + 'info_dict': { + 'id': '6950251282489675042', + 'ext': 'mp4', + 'title': '哈哈哈，成功了哈哈哈哈哈哈', + 'uploader': '杨超越', + 'upload_date': '20210412', + 'timestamp': 1618231483, + 'uploader_id': '110403406559', + 'view_count': int, + 'like_count': int, + 'repost_count': int, + 'comment_count': int, + } + }, { + 'url': 'https://www.douyin.com/video/6963263655114722595', + 'md5': '1abe1c477d05ee62efb40bf2329957cf', + 'info_dict': { + 'id': '6963263655114722595', + 'ext': 'mp4', + 'title': '#哪个爱豆的105度最甜 换个角度看看我哈哈', + 'uploader': '杨超越', + 'upload_date': '20210517', + 'timestamp': 1621261163, + 'uploader_id': '110403406559', + 'view_count': int, + 'like_count': int, + 'repost_count': int, + 'comment_count': int, + } + }] + _APP_VERSION = '9.6.0' + _MANIFEST_APP_VERSION = '960' + _APP_NAME = 'aweme' + _AID = 1128 + _API_HOSTNAME = 'aweme.snssdk.com' + _UPLOADER_URL_FORMAT = 'https://www.douyin.com/user/%s' + _WEBPAGE_HOST = 'https://www.douyin.com/' + + def _real_extract(self, url): + video_id = self._match_id(url) + + try: + return self._extract_aweme_app(video_id) + except ExtractorError as e: + self.report_warning(f'{e}; Retrying with webpage') + + webpage = self._download_webpage(url, video_id) + render_data_json = self._search_regex( + r'', + webpage, 'render data', default=None) + if not render_data_json: + # TODO: Run verification challenge code to generate signature cookies + raise ExtractorError('Fresh cookies (not necessarily logged in) are needed') + + render_data = self._parse_json( + render_data_json, video_id, transform_source=compat_urllib_parse_unquote) + return self._parse_aweme_video_web( + traverse_obj(render_data, (..., 'aweme', 'detail'), get_all=False), url)