-# coding: utf-8
-from __future__ import unicode_literals
-
import itertools
+import json
import random
+import re
import string
import time
-import json
from .common import InfoExtractor
-from ..compat import (
- compat_urllib_parse_unquote,
- compat_urllib_parse_urlparse
-)
+from ..compat import compat_urllib_parse_unquote, compat_urllib_parse_urlparse
from ..utils import (
ExtractorError,
HEADRequest,
+ LazyList,
+ UnsupportedError,
+ UserNotLive,
+ format_field,
+ get_element_by_id,
get_first,
int_or_none,
join_nonempty,
- LazyList,
+ merge_dicts,
+ qualities,
+ remove_start,
srt_subtitles_timecode,
str_or_none,
traverse_obj,
+ try_call,
try_get,
url_or_none,
- qualities,
)
class TikTokBaseIE(InfoExtractor):
- _APP_VERSIONS = [('20.9.3', '293'), ('20.4.3', '243'), ('20.2.1', '221'), ('20.1.2', '212'), ('20.0.4', '204')]
+ _APP_VERSIONS = [('26.1.3', '260103'), ('26.1.2', '260102'), ('26.1.1', '260101'), ('25.6.2', '250602')]
_WORKING_APP_VERSION = None
_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', '1080p')
+ @property
+ def _API_HOSTNAME(self):
+ return self._configuration_arg(
+ 'api_hostname', ['api16-normal-c-useast1a.tiktokv.com'], ie_key=TikTokIE)[0]
+
+ @staticmethod
+ def _create_url(user_id, video_id):
+ return f'https://www.tiktok.com/@{user_id or "_"}/video/{video_id}'
+
+ def _get_sigi_state(self, webpage, display_id):
+ return self._parse_json(get_element_by_id(
+ 'SIGI_STATE|sigi-persisted-data', webpage, escape_value=False), display_id)
+
def _call_api_impl(self, ep, query, manifest_app_version, video_id, fatal=True,
note='Downloading API JSON', errnote='Unable to download API page'):
- self._set_cookie(self._API_HOSTNAME, 'odin_tt', ''.join(random.choice('0123456789abcdef') for _ in range(160)))
+ self._set_cookie(self._API_HOSTNAME, 'odin_tt', ''.join(random.choices('0123456789abcdef', k=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://%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/{manifest_app_version} (Linux; U; Android 10; en_US; Pixel 4; Build/QQ3A.200805.001; Cronet/58.0.2991.0)',
+ 'User-Agent': f'com.ss.android.ugc.{self._APP_NAME}/{manifest_app_version} (Linux; U; Android 10; en_US; Pixel 4; Build/QQ3A.200805.001; Cronet/58.0.2991.0)',
'Accept': 'application/json',
}, query=query)
'build_number': app_version,
'manifest_version_code': manifest_app_version,
'update_version_code': manifest_app_version,
- 'openudid': ''.join(random.choice('0123456789abcdef') for _ in range(16)),
- 'uuid': ''.join([random.choice(string.digits) for _ in range(16)]),
+ 'openudid': ''.join(random.choices('0123456789abcdef', k=16)),
+ 'uuid': ''.join(random.choices(string.digits, k=16)),
'_rticket': int(time.time() * 1000),
'ts': int(time.time()),
'device_brand': 'Google',
continue
raise e
+ def _extract_aweme_app(self, aweme_id):
+ 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), None)
+ 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 _get_subtitles(self, aweme_detail, aweme_id):
# TODO: Extract text positioning info
subtitles = {}
+ # aweme/detail endpoint subs
captions_info = traverse_obj(
- aweme_detail, ('interaction_stickers', ..., 'auto_video_caption_info', 'auto_captions', ...), expected_type=dict, default=[])
+ aweme_detail, ('interaction_stickers', ..., 'auto_video_caption_info', 'auto_captions', ...), expected_type=dict)
for caption in captions_info:
caption_url = traverse_obj(caption, ('url', 'url_list', ...), expected_type=url_or_none, get_all=False)
if not caption_url:
f'{i + 1}\n{srt_subtitles_timecode(line["start_time"] / 1000)} --> {srt_subtitles_timecode(line["end_time"] / 1000)}\n{line["text"]}'
for i, line in enumerate(caption_json['utterances']) if line.get('text'))
})
+ # feed endpoint subs
+ if not subtitles:
+ for caption in traverse_obj(aweme_detail, ('video', 'cla_info', 'caption_infos', ...), expected_type=dict):
+ if not caption.get('url'):
+ continue
+ subtitles.setdefault(caption.get('lang') or 'en', []).append({
+ 'ext': remove_start(caption.get('caption_format'), 'web'),
+ 'url': caption['url'],
+ })
+ # webpage subs
+ if not subtitles:
+ for caption in traverse_obj(aweme_detail, ('video', 'subtitleInfos', ...), expected_type=dict):
+ if not caption.get('Url'):
+ continue
+ subtitles.setdefault(caption.get('LanguageCodeName') or 'en', []).append({
+ 'ext': remove_start(caption.get('Format'), 'web'),
+ 'url': caption['Url'],
+ })
return subtitles
def _parse_aweme_video_app(self, aweme_detail):
if auth_cookie:
for f in formats:
self._set_cookie(compat_urllib_parse_urlparse(f['url']).hostname, 'sid_tt', auth_cookie.value)
- self._sort_formats(formats, ('quality', 'codec', 'size', 'br'))
thumbnails = []
for cover_id in ('cover', 'ai_dynamic_cover', 'animated_cover', 'ai_dynamic_cover_bak',
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=[])
+ labels = traverse_obj(aweme_detail, ('hybrid_label', ..., 'text'), expected_type=str)
contained_music_track = traverse_obj(
music_info, ('matched_song', 'title'), ('matched_pgc_sound', 'title'), expected_type=str)
return {
'id': aweme_id,
- 'title': aweme_detail['desc'],
- 'description': aweme_detail['desc'],
+ 'extractor_key': TikTokIE.ie_key(),
+ 'extractor': TikTokIE.IE_NAME,
+ 'webpage_url': self._create_url(author_info.get('uid'), aweme_id),
+ 'title': aweme_detail.get('desc'),
+ 'description': aweme_detail.get('desc'),
'view_count': int_or_none(stats_info.get('play_count')),
'like_count': int_or_none(stats_info.get('digg_count')),
'repost_count': int_or_none(stats_info.get('share_count')),
'uploader_url': user_url,
'track': music_track,
'album': str_or_none(music_info.get('album')) or None,
- 'artist': music_author,
+ 'artist': music_author or None,
'timestamp': int_or_none(aweme_detail.get('create_time')),
'formats': formats,
'subtitles': self.extract_subtitles(aweme_detail, aweme_id),
'availability': self._availability(
is_private='Private' in labels,
needs_subscription='Friends only' in labels,
- is_unlisted='Followers only' in labels)
+ is_unlisted='Followers only' in labels),
+ '_format_sort_fields': ('quality', 'codec', 'size', 'br'),
}
def _parse_aweme_video_web(self, aweme_detail, webpage_url):
'ext': 'mp4',
'width': width,
'height': height,
- } for url in traverse_obj(play_url, (..., 'src'), expected_type=url_or_none, default=[]) if url]
+ } for url in traverse_obj(play_url, (..., 'src'), expected_type=url_or_none) 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:
'height': height,
})
self._remove_duplicate_formats(formats)
- self._sort_formats(formats)
thumbnails = []
for thumbnail_name in ('thumbnail', 'cover', 'dynamicCover', 'originCover'):
'timestamp': int_or_none(aweme_detail.get('createTime')),
'creator': str_or_none(author_info.get('nickname')),
'uploader': str_or_none(author_info.get('uniqueId') or aweme_detail.get('author')),
- 'uploader_id': str_or_none(author_info.get('id') or aweme_detail.get('authorId')),
+ 'uploader_id': str_or_none(traverse_obj(author_info, 'id', 'uid', 'authorId')),
'uploader_url': user_url,
'track': str_or_none(music_info.get('title')),
'album': str_or_none(music_info.get('album')) or None,
class TikTokIE(TikTokBaseIE):
- _VALID_URL = r'https?://www\.tiktok\.com/@[\w\.-]+/video/(?P<id>\d+)'
+ _VALID_URL = r'https?://www\.tiktok\.com/(?:embed|@(?P<user_id>[\w\.-]+)?/video)/(?P<id>\d+)'
+ _EMBED_REGEX = [rf'<(?:script|iframe)[^>]+\bsrc=(["\'])(?P<url>{_VALID_URL})']
_TESTS = [{
'url': 'https://www.tiktok.com/@leenabhushan/video/6748451240264420610',
'like_count': int,
'repost_count': int,
'comment_count': int,
+ 'artist': 'Ysrbeats',
+ 'album': 'Lehanga',
+ 'track': 'Lehanga',
}
}, {
'url': 'https://www.tiktok.com/@patroxofficial/video/6742501081818877190?langCountry=en',
'like_count': int,
'repost_count': int,
'comment_count': int,
+ 'artist': 'Evan Todd, Jessica Keenan Wynn, Alice Lee, Barrett Wilbert Weed & Jon Eidson',
+ 'track': 'Big Fun',
}
}, {
# Banned audio, only available on the app
'repost_count': int,
'comment_count': int,
},
- 'expected_warnings': ['Video not available']
+ 'expected_warnings': ['trying with webpage', 'Unable to find video in feed']
+ }, {
+ # Video without title and description
+ 'url': 'https://www.tiktok.com/@pokemonlife22/video/7059698374567611694',
+ 'info_dict': {
+ 'id': '7059698374567611694',
+ 'ext': 'mp4',
+ 'title': 'TikTok video #7059698374567611694',
+ 'description': '',
+ 'uploader': 'pokemonlife22',
+ 'creator': 'Pokemon',
+ 'uploader_id': '6820838815978423302',
+ 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA0tF1nBwQVVMyrGu3CqttkNgM68Do1OXUFuCY0CRQk8fEtSVDj89HqoqvbSTmUP2W',
+ 'track': 'original sound',
+ 'timestamp': 1643714123,
+ 'duration': 6,
+ 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
+ 'upload_date': '20220201',
+ 'artist': 'Pokemon',
+ 'view_count': int,
+ 'like_count': int,
+ 'repost_count': int,
+ 'comment_count': int,
+ },
+ }, {
+ # hydration JSON is sent in a <script> element
+ 'url': 'https://www.tiktok.com/@denidil6/video/7065799023130643713',
+ 'info_dict': {
+ 'id': '7065799023130643713',
+ 'ext': 'mp4',
+ 'title': '#denidil#денидил',
+ 'description': '#denidil#денидил',
+ 'uploader': 'denidil6',
+ 'uploader_id': '7046664115636405250',
+ 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAsvMSzFdQ4ikl3uR2TEJwMBbB2yZh2Zxwhx-WCo3rbDpAharE3GQCrFuJArI3C8QJ',
+ 'artist': 'Holocron Music',
+ 'album': 'Wolf Sounds (1 Hour) Enjoy the Company of the Animal That Is the Majestic King of the Night',
+ 'track': 'Wolf Sounds (1 Hour) Enjoy the Company of the Animal That Is the Majestic King of the Night',
+ 'timestamp': 1645134536,
+ 'duration': 26,
+ 'upload_date': '20220217',
+ 'view_count': int,
+ 'like_count': int,
+ 'repost_count': int,
+ 'comment_count': int,
+ },
+ 'skip': 'This video is unavailable',
}, {
# Auto-captions available
'url': 'https://www.tiktok.com/@hankgreen1/video/7047596209028074758',
'only_matching': True
}]
- def _extract_aweme_app(self, aweme_id):
- 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), None)
- 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)
-
+ video_id, user_id = self._match_valid_url(url).group('id', 'user_id')
try:
return self._extract_aweme_app(video_id)
except ExtractorError as e:
- self.report_warning(f'{e}; Retrying with webpage')
+ self.report_warning(f'{e}; trying with webpage')
- # If we only call once, we get a 403 when downlaoding the video.
- self._download_webpage(url, video_id)
- webpage = self._download_webpage(url, video_id, note='Downloading video webpage')
+ url = self._create_url(user_id, video_id)
+ webpage = self._download_webpage(url, video_id, headers={'User-Agent': 'Mozilla/5.0'})
next_data = self._search_nextjs_data(webpage, video_id, default='{}')
-
if next_data:
status = traverse_obj(next_data, ('props', 'pageProps', 'statusCode'), expected_type=int) or 0
video_data = traverse_obj(next_data, ('props', 'pageProps', 'itemInfo', 'itemStruct'), expected_type=dict)
else:
- sigi_json = self._search_regex(
- r'>\s*window\[[\'"]SIGI_STATE[\'"]\]\s*=\s*(?P<sigi_state>{.+});',
- webpage, 'sigi data', group='sigi_state')
- sigi_data = self._parse_json(sigi_json, video_id)
+ sigi_data = self._get_sigi_state(webpage, video_id)
status = traverse_obj(sigi_data, ('VideoPage', 'statusCode'), expected_type=int) or 0
video_data = traverse_obj(sigi_data, ('ItemModule', video_id), expected_type=dict)
class TikTokUserIE(TikTokBaseIE):
IE_NAME = 'tiktok:user'
_VALID_URL = r'https?://(?:www\.)?tiktok\.com/@(?P<id>[\w\.-]+)/?(?:$|[#?])'
+ _WORKING = False
_TESTS = [{
'url': 'https://tiktok.com/@corgibobaa?lang=en',
'playlist_mincount': 45,
'thumbnail': r're:https://.+_1080x1080\.webp'
},
'expected_warnings': ['Retrying']
+ }, {
+ 'url': 'https://www.tiktok.com/@6820838815978423302',
+ 'playlist_mincount': 5,
+ 'info_dict': {
+ 'id': '6820838815978423302',
+ 'title': '6820838815978423302',
+ 'thumbnail': r're:https://.+_1080x1080\.webp'
+ },
+ 'expected_warnings': ['Retrying']
}, {
'url': 'https://www.tiktok.com/@meme',
'playlist_mincount': 593,
'max_cursor': 0,
'min_cursor': 0,
'retry_type': 'no_retry',
- '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.
+ 'device_id': ''.join(random.choices(string.digits, k=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)
for page in itertools.count(1):
- for retries in itertools.count():
+ for retry in self.RetryManager():
try:
- post_list = self._call_api('aweme/post', query, username,
- note='Downloading user video list page %d%s' % (page, f' (attempt {retries})' if retries != 0 else ''),
- errnote='Unable to download user video list')
+ post_list = self._call_api(
+ 'aweme/post', query, username, note=f'Downloading user video list page {page}',
+ errnote='Unable to download user 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))
+ if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0:
+ retry.error = e
continue
raise
- break
yield from post_list.get('aweme_list', [])
if not post_list.get('has_more'):
break
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')
+ user_id = self._html_search_regex(r'snssdk\d*://user/profile/(\d+)', webpage, 'user ID', default=None) or user_name
videos = LazyList(self._video_entries_api(webpage, user_id, user_name))
thumbnail = traverse_obj(videos, (0, 'author', 'avatar_larger', 'url_list', 0))
return self.playlist_result(self._entries_api(user_id, videos), user_id, user_name, thumbnail=thumbnail)
-class TikTokBaseListIE(TikTokBaseIE):
+class TikTokBaseListIE(TikTokBaseIE): # XXX: Conventionally, base classes should end with BaseIE/InfoExtractor
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))
+ 'device_id': ''.join(random.choices(string.digits, k=19))
}
- max_retries = self.get_param('extractor_retries', 3)
for page in itertools.count(1):
- for retries in itertools.count():
+ for retry in self.RetryManager():
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')
+ post_list = self._call_api(
+ self._API_ENDPOINT, query, display_id, note=f'Downloading video list page {page}',
+ 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))
+ if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0:
+ retry.error = e
continue
raise
- break
for video in post_list.get('aweme_list', []):
yield {
**self._parse_aweme_video_app(video),
class TikTokSoundIE(TikTokBaseListIE):
IE_NAME = 'tiktok:sound'
_VALID_URL = r'https?://(?:www\.)?tiktok\.com/music/[\w\.-]+-(?P<id>[\d]+)[/?#&]?'
+ _WORKING = False
_QUERY_NAME = 'music_id'
_API_ENDPOINT = 'music/aweme'
_TESTS = [{
class TikTokEffectIE(TikTokBaseListIE):
IE_NAME = 'tiktok:effect'
_VALID_URL = r'https?://(?:www\.)?tiktok\.com/sticker/[\w\.-]+-(?P<id>[\d]+)[/?#&]?'
+ _WORKING = False
_QUERY_NAME = 'sticker_id'
_API_ENDPOINT = 'sticker/aweme'
_TESTS = [{
class TikTokTagIE(TikTokBaseListIE):
IE_NAME = 'tiktok:tag'
_VALID_URL = r'https?://(?:www\.)?tiktok\.com/tag/(?P<id>[^/?#&]+)'
+ _WORKING = False
_QUERY_NAME = 'ch_id'
_API_ENDPOINT = 'challenge/aweme'
_TESTS = [{
return self.playlist_result(self._entries(tag_id, display_id), tag_id, display_id)
-class DouyinIE(TikTokIE):
+class DouyinIE(TikTokBaseIE):
_VALID_URL = r'https?://(?:www\.)?douyin\.com/video/(?P<id>[0-9]+)'
_TESTS = [{
'url': 'https://www.douyin.com/video/6961737553342991651',
- 'md5': '10523312c8b8100f353620ac9dc8f067',
+ 'md5': 'a97db7e3e67eb57bf40735c022ffa228',
'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,
+ 'description': '#杨超越 小小水手带你去远航❤️',
'uploader_id': '110403406559',
+ 'uploader_url': 'https://www.douyin.com/user/MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98',
+ 'creator': '杨超越',
+ 'duration': 19782,
+ 'timestamp': 1620905839,
+ 'upload_date': '20210513',
+ 'track': '@杨超越创作的原声',
'view_count': int,
'like_count': int,
'repost_count': int,
'comment_count': int,
- }
+ },
}, {
'url': 'https://www.douyin.com/video/6982497745948921092',
- 'md5': 'd78408c984b9b5102904cf6b6bc2d712',
+ 'md5': '34a87ebff3833357733da3fe17e37c0e',
'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,
+ 'description': '这个夏日和小羊@杨超越 一起遇见白色幻想',
'uploader_id': '408654318141572',
+ 'uploader_url': 'https://www.douyin.com/user/MS4wLjABAAAAZJpnglcjW2f_CMVcnqA_6oVBXKWMpH0F8LIHuUu8-lA',
+ 'creator': '杨超越工作室',
+ 'duration': 42608,
+ 'timestamp': 1625739481,
+ 'upload_date': '20210708',
+ 'track': '@杨超越工作室创作的原声',
'view_count': int,
'like_count': int,
'repost_count': int,
'comment_count': int,
- }
+ },
}, {
'url': 'https://www.douyin.com/video/6953975910773099811',
- 'md5': '72e882e24f75064c218b76c8b713c185',
+ 'md5': 'dde3302460f19db59c47060ff013b902',
'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,
+ 'description': '#一起看海 出现在你的夏日里',
'uploader_id': '110403406559',
+ 'uploader_url': 'https://www.douyin.com/user/MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98',
+ 'creator': '杨超越',
+ 'duration': 17228,
+ 'timestamp': 1619098692,
+ 'upload_date': '20210422',
+ 'track': '@杨超越创作的原声',
'view_count': int,
'like_count': int,
'repost_count': int,
'comment_count': int,
- }
+ },
}, {
'url': 'https://www.douyin.com/video/6950251282489675042',
'md5': 'b4db86aec367ef810ddd38b1737d2fed',
'like_count': int,
'repost_count': int,
'comment_count': int,
- }
+ },
+ 'skip': 'No longer available',
}, {
'url': 'https://www.douyin.com/video/6963263655114722595',
- 'md5': '1abe1c477d05ee62efb40bf2329957cf',
+ 'md5': 'cf9f11f0ec45d131445ec2f06766e122',
'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,
+ 'description': '#哪个爱豆的105度最甜 换个角度看看我哈哈',
'uploader_id': '110403406559',
+ 'uploader_url': 'https://www.douyin.com/user/MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98',
+ 'creator': '杨超越',
+ 'duration': 15115,
+ 'timestamp': 1621261163,
+ 'upload_date': '20210517',
+ 'track': '@杨超越创作的原声',
'view_count': int,
'like_count': int,
'repost_count': int,
'comment_count': int,
- }
+ },
}]
- _APP_VERSIONS = [('9.6.0', '960')]
+ _APP_VERSIONS = [('23.3.0', '230300')]
_APP_NAME = 'aweme'
_AID = 1128
_API_HOSTNAME = 'aweme.snssdk.com'
try:
return self._extract_aweme_app(video_id)
except ExtractorError as e:
- self.report_warning(f'{e}; Retrying with webpage')
+ e.expected = True
+ self.to_screen(f'{e}; trying with webpage')
webpage = self._download_webpage(url, video_id)
render_data_json = self._search_regex(
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')
+ cookies = self._get_cookies(self._WEBPAGE_HOST)
+ expected = not cookies.get('s_v_web_id') or not cookies.get('ttwid')
+ raise ExtractorError(
+ 'Fresh cookies (not necessarily logged in) are needed', expected=expected)
render_data = self._parse_json(
render_data_json, video_id, transform_source=compat_urllib_parse_unquote)
class TikTokVMIE(InfoExtractor):
- _VALID_URL = r'https?://(?:vm|vt)\.tiktok\.com/(?P<id>\w+)'
+ _VALID_URL = r'https?://(?:(?:vm|vt)\.tiktok\.com|(?:www\.)tiktok\.com/t)/(?P<id>\w+)'
IE_NAME = 'vm.tiktok'
_TESTS = [{
- 'url': 'https://vm.tiktok.com/ZSe4FqkKd',
+ 'url': 'https://www.tiktok.com/t/ZTRC5xgJp',
'info_dict': {
- 'id': '7023491746608712966',
+ 'id': '7170520270497680683',
'ext': 'mp4',
- 'title': 'md5:5607564db90271abbbf8294cca77eddd',
- 'description': 'md5:5607564db90271abbbf8294cca77eddd',
- 'duration': 11,
- 'upload_date': '20211026',
- 'uploader_id': '7007385080558846981',
- 'creator': 'Memes',
- 'artist': 'Memes',
- 'track': 'original sound',
- 'uploader': 'susmandem',
- 'timestamp': 1635284105,
- 'thumbnail': r're:https://.+\.webp.*',
+ 'title': 'md5:c64f6152330c2efe98093ccc8597871c',
+ 'uploader_id': '6687535061741700102',
+ 'upload_date': '20221127',
+ 'view_count': int,
'like_count': int,
+ 'comment_count': int,
+ 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAObqu3WCTXxmw2xwZ3iLEHnEecEIw7ks6rxWqOqOhaPja9BI7gqUQnjw8_5FSoDXX',
+ 'album': 'Wave of Mutilation: Best of Pixies',
+ 'thumbnail': r're:https://.+\.webp.*',
+ 'duration': 5,
+ 'timestamp': 1669516858,
+ 'repost_count': int,
+ 'artist': 'Pixies',
+ 'track': 'Where Is My Mind?',
+ 'description': 'md5:c64f6152330c2efe98093ccc8597871c',
+ 'uploader': 'sigmachaddeus',
+ 'creator': 'SigmaChad',
+ },
+ }, {
+ 'url': 'https://vm.tiktok.com/ZTR45GpSF/',
+ 'info_dict': {
+ 'id': '7106798200794926362',
+ 'ext': 'mp4',
+ 'title': 'md5:edc3e7ea587847f8537468f2fe51d074',
+ 'uploader_id': '6997695878846268418',
+ 'upload_date': '20220608',
'view_count': int,
+ 'like_count': int,
'comment_count': int,
+ 'thumbnail': r're:https://.+\.webp.*',
+ 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAdZ_NcPPgMneaGrW0hN8O_J_bwLshwNNERRF5DxOw2HKIzk0kdlLrR8RkVl1ksrMO',
+ 'duration': 29,
+ 'timestamp': 1654680400,
'repost_count': int,
- 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAXcNoOEOxVyBzuII_E--T0MeCrLP0ay1Sm6x_n3dluiWEoWZD0VlQOytwad4W0i0n',
- }
+ 'artist': 'Akihitoko',
+ 'track': 'original sound',
+ 'description': 'md5:edc3e7ea587847f8537468f2fe51d074',
+ 'uploader': 'akihitoko1',
+ 'creator': 'Akihitoko',
+ },
}, {
'url': 'https://vt.tiktok.com/ZSe4FqkKd',
'only_matching': True,
}]
def _real_extract(self, url):
- return self.url_result(self._request_webpage(
- HEADRequest(url), self._match_id(url), headers={'User-Agent': 'facebookexternalhit/1.1'}).geturl(), TikTokIE)
+ new_url = self._request_webpage(
+ HEADRequest(url), self._match_id(url), headers={'User-Agent': 'facebookexternalhit/1.1'}).geturl()
+ if self.suitable(new_url): # Prevent infinite loop in case redirect fails
+ raise UnsupportedError(new_url)
+ return self.url_result(new_url)
+
+
+class TikTokLiveIE(TikTokBaseIE):
+ _VALID_URL = r'''(?x)https?://(?:
+ (?:www\.)?tiktok\.com/@(?P<uploader>[\w.-]+)/live|
+ m\.tiktok\.com/share/live/(?P<id>\d+)
+ )'''
+ IE_NAME = 'tiktok:live'
+
+ _TESTS = [{
+ 'url': 'https://www.tiktok.com/@weathernewslive/live',
+ 'info_dict': {
+ 'id': '7210809319192726273',
+ 'ext': 'mp4',
+ 'title': r're:ウェザーニュースLiVE[\d\s:-]*',
+ 'creator': 'ウェザーニュースLiVE',
+ 'uploader': 'weathernewslive',
+ 'uploader_id': '6621496731283095554',
+ 'uploader_url': 'https://www.tiktok.com/@weathernewslive',
+ 'live_status': 'is_live',
+ 'concurrent_view_count': int,
+ },
+ 'params': {'skip_download': 'm3u8'},
+ }, {
+ 'url': 'https://www.tiktok.com/@pilarmagenta/live',
+ 'info_dict': {
+ 'id': '7209423610325322522',
+ 'ext': 'mp4',
+ 'title': str,
+ 'creator': 'Pilarmagenta',
+ 'uploader': 'pilarmagenta',
+ 'uploader_id': '6624846890674683909',
+ 'uploader_url': 'https://www.tiktok.com/@pilarmagenta',
+ 'live_status': 'is_live',
+ 'concurrent_view_count': int,
+ },
+ 'skip': 'Livestream',
+ }, {
+ 'url': 'https://m.tiktok.com/share/live/7209423610325322522/?language=en',
+ 'only_matching': True,
+ }, {
+ 'url': 'https://www.tiktok.com/@iris04201/live',
+ 'only_matching': True,
+ }]
+
+ def _call_api(self, url, param, room_id, uploader, key=None):
+ response = traverse_obj(self._download_json(
+ url, room_id, fatal=False, query={
+ 'aid': '1988',
+ param: room_id,
+ }), (key, {dict}), default={})
+
+ # status == 2 if live else 4
+ if int_or_none(response.get('status')) == 2:
+ return response
+ # If room_id is obtained via mobile share URL and cannot be refreshed, do not wait for live
+ elif not uploader:
+ raise ExtractorError('This livestream has ended', expected=True)
+ raise UserNotLive(video_id=uploader)
+
+ def _real_extract(self, url):
+ uploader, room_id = self._match_valid_url(url).group('uploader', 'id')
+ webpage = self._download_webpage(
+ url, uploader or room_id, headers={'User-Agent': 'Mozilla/5.0'}, fatal=not room_id)
+
+ if webpage:
+ data = try_call(lambda: self._get_sigi_state(webpage, uploader or room_id))
+ room_id = (traverse_obj(data, ('UserModule', 'users', ..., 'roomId', {str_or_none}), get_all=False)
+ or self._search_regex(r'snssdk\d*://live\?room_id=(\d+)', webpage, 'room ID', default=None)
+ or room_id)
+ uploader = uploader or traverse_obj(
+ data, ('LiveRoom', 'liveRoomUserInfo', 'user', 'uniqueId'),
+ ('UserModule', 'users', ..., 'uniqueId'), get_all=False, expected_type=str)
+
+ if not room_id:
+ raise UserNotLive(video_id=uploader)
+
+ formats = []
+ live_info = self._call_api(
+ 'https://webcast.tiktok.com/webcast/room/info', 'room_id', room_id, uploader, key='data')
+
+ get_quality = qualities(('SD1', 'ld', 'SD2', 'sd', 'HD1', 'hd', 'FULL_HD1', 'uhd', 'ORIGION', 'origin'))
+ parse_inner = lambda x: self._parse_json(x, None)
+
+ for quality, stream in traverse_obj(live_info, (
+ 'stream_url', 'live_core_sdk_data', 'pull_data', 'stream_data',
+ {parse_inner}, 'data', {dict}), default={}).items():
+
+ sdk_params = traverse_obj(stream, ('main', 'sdk_params', {parse_inner}, {
+ 'vcodec': ('VCodec', {str}),
+ 'tbr': ('vbitrate', {lambda x: int_or_none(x, 1000)}),
+ 'resolution': ('resolution', {lambda x: re.match(r'(?i)\d+x\d+|\d+p', x).group().lower()}),
+ }))
+
+ flv_url = traverse_obj(stream, ('main', 'flv', {url_or_none}))
+ if flv_url:
+ formats.append({
+ 'url': flv_url,
+ 'ext': 'flv',
+ 'format_id': f'flv-{quality}',
+ 'quality': get_quality(quality),
+ **sdk_params,
+ })
+
+ hls_url = traverse_obj(stream, ('main', 'hls', {url_or_none}))
+ if hls_url:
+ formats.append({
+ 'url': hls_url,
+ 'ext': 'mp4',
+ 'protocol': 'm3u8_native',
+ 'format_id': f'hls-{quality}',
+ 'quality': get_quality(quality),
+ **sdk_params,
+ })
+
+ def get_vcodec(*keys):
+ return traverse_obj(live_info, (
+ 'stream_url', *keys, {parse_inner}, 'VCodec', {str}))
+
+ for stream in ('hls', 'rtmp'):
+ stream_url = traverse_obj(live_info, ('stream_url', f'{stream}_pull_url', {url_or_none}))
+ if stream_url:
+ formats.append({
+ 'url': stream_url,
+ 'ext': 'mp4' if stream == 'hls' else 'flv',
+ 'protocol': 'm3u8_native' if stream == 'hls' else 'https',
+ 'format_id': f'{stream}-pull',
+ 'vcodec': get_vcodec(f'{stream}_pull_url_params'),
+ 'quality': get_quality('ORIGION'),
+ })
+
+ for f_id, f_url in traverse_obj(live_info, ('stream_url', 'flv_pull_url', {dict}), default={}).items():
+ if not url_or_none(f_url):
+ continue
+ formats.append({
+ 'url': f_url,
+ 'ext': 'flv',
+ 'format_id': f'flv-{f_id}'.lower(),
+ 'vcodec': get_vcodec('flv_pull_url_params', f_id),
+ 'quality': get_quality(f_id),
+ })
+
+ # If uploader is a guest on another's livestream, primary endpoint will not have m3u8 URLs
+ if not traverse_obj(formats, lambda _, v: v['ext'] == 'mp4'):
+ live_info = merge_dicts(live_info, self._call_api(
+ 'https://www.tiktok.com/api/live/detail/', 'roomID', room_id, uploader, key='LiveRoomInfo'))
+ if url_or_none(live_info.get('liveUrl')):
+ formats.append({
+ 'url': live_info['liveUrl'],
+ 'ext': 'mp4',
+ 'protocol': 'm3u8_native',
+ 'format_id': 'hls-fallback',
+ 'vcodec': 'h264',
+ 'quality': get_quality('origin'),
+ })
+
+ uploader = uploader or traverse_obj(live_info, ('ownerInfo', 'uniqueId'), ('owner', 'display_id'))
+
+ return {
+ 'id': room_id,
+ 'uploader': uploader,
+ 'uploader_url': format_field(uploader, None, self._UPLOADER_URL_FORMAT) or None,
+ 'is_live': True,
+ 'formats': formats,
+ '_format_sort_fields': ('quality', 'ext'),
+ **traverse_obj(live_info, {
+ 'title': 'title',
+ 'uploader_id': (('ownerInfo', 'owner'), 'id', {str_or_none}),
+ 'creator': (('ownerInfo', 'owner'), 'nickname'),
+ 'concurrent_view_count': (('user_count', ('liveRoomStats', 'userCount')), {int_or_none}),
+ }, get_all=False),
+ }
projects / yt-dlp.git / blobdiff