[ie/tiktok:user] Fix extractor (#9661)

Closes #3776, Closes #4996
Authored by: bashonly
This commit is contained in:
bashonly 2024-05-26 16:16:36 -05:00 committed by GitHub
parent 96a134dea6
commit 347f13dd9b
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194

View File

@ -3,6 +3,7 @@
import json import json
import random import random
import re import re
import string
import time import time
import uuid import uuid
@ -11,7 +12,6 @@
from ..networking import HEADRequest from ..networking import HEADRequest
from ..utils import ( from ..utils import (
ExtractorError, ExtractorError,
LazyList,
UnsupportedError, UnsupportedError,
UserNotLive, UserNotLive,
determine_ext, determine_ext,
@ -236,7 +236,7 @@ def _extract_web_data_and_status(self, url, video_id, fatal=True):
return video_data, status return video_data, status
def _get_subtitles(self, aweme_detail, aweme_id, user_url): def _get_subtitles(self, aweme_detail, aweme_id, user_name):
# TODO: Extract text positioning info # TODO: Extract text positioning info
subtitles = {} subtitles = {}
# aweme/detail endpoint subs # aweme/detail endpoint subs
@ -267,9 +267,9 @@ def _get_subtitles(self, aweme_detail, aweme_id, user_url):
}) })
# webpage subs # webpage subs
if not subtitles: if not subtitles:
if user_url: # only _parse_aweme_video_app needs to extract the webpage here if user_name: # only _parse_aweme_video_app needs to extract the webpage here
aweme_detail, _ = self._extract_web_data_and_status( aweme_detail, _ = self._extract_web_data_and_status(
f'{user_url}/video/{aweme_id}', aweme_id, fatal=False) self._create_url(user_name, aweme_id), aweme_id, fatal=False)
for caption in traverse_obj(aweme_detail, ('video', 'subtitleInfos', lambda _, v: v['Url'])): for caption in traverse_obj(aweme_detail, ('video', 'subtitleInfos', lambda _, v: v['Url'])):
subtitles.setdefault(caption.get('LanguageCodeName') or 'en', []).append({ subtitles.setdefault(caption.get('LanguageCodeName') or 'en', []).append({
'ext': remove_start(caption.get('Format'), 'web'), 'ext': remove_start(caption.get('Format'), 'web'),
@ -394,11 +394,7 @@ def extract_addr(addr, add_meta={}):
}) })
stats_info = aweme_detail.get('statistics') or {} stats_info = aweme_detail.get('statistics') or {}
author_info = aweme_detail.get('author') or {}
music_info = aweme_detail.get('music') or {} music_info = aweme_detail.get('music') or {}
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) labels = traverse_obj(aweme_detail, ('hybrid_label', ..., 'text'), expected_type=str)
contained_music_track = traverse_obj( contained_music_track = traverse_obj(
@ -412,6 +408,13 @@ def extract_addr(addr, add_meta={}):
else: else:
music_track, music_author = music_info.get('title'), traverse_obj(music_info, ('author', {str})) music_track, music_author = music_info.get('title'), traverse_obj(music_info, ('author', {str}))
author_info = traverse_obj(aweme_detail, ('author', {
'uploader': ('unique_id', {str}),
'uploader_id': ('uid', {str_or_none}),
'channel': ('nickname', {str}),
'channel_id': ('sec_uid', {str}),
}))
return { return {
'id': aweme_id, 'id': aweme_id,
**traverse_obj(aweme_detail, { **traverse_obj(aweme_detail, {
@ -425,21 +428,20 @@ def extract_addr(addr, add_meta={}):
'repost_count': 'share_count', 'repost_count': 'share_count',
'comment_count': 'comment_count', 'comment_count': 'comment_count',
}, expected_type=int_or_none), }, expected_type=int_or_none),
**traverse_obj(author_info, { **author_info,
'uploader': ('unique_id', {str}), 'channel_url': format_field(author_info, 'channel_id', self._UPLOADER_URL_FORMAT, default=None),
'uploader_id': ('uid', {str_or_none}), 'uploader_url': format_field(
'creators': ('nickname', {str}, {lambda x: [x] if x else None}), # for compat author_info, ['uploader', 'uploader_id'], self._UPLOADER_URL_FORMAT, default=None),
'channel': ('nickname', {str}),
'channel_id': ('sec_uid', {str}),
}),
'uploader_url': user_url,
'track': music_track, 'track': music_track,
'album': str_or_none(music_info.get('album')) or None, 'album': str_or_none(music_info.get('album')) or None,
'artists': re.split(r'(?:, | & )', music_author) if music_author else None, 'artists': re.split(r'(?:, | & )', music_author) if music_author else None,
'formats': formats, 'formats': formats,
'subtitles': self.extract_subtitles(aweme_detail, aweme_id, user_url), 'subtitles': self.extract_subtitles(
aweme_detail, aweme_id, traverse_obj(author_info, 'uploader', 'uploader_id', 'channel_id')),
'thumbnails': thumbnails, 'thumbnails': thumbnails,
'duration': int_or_none(traverse_obj(video_info, 'duration', ('download_addr', 'duration')), scale=1000), 'duration': (traverse_obj(video_info, (
(None, 'download_addr'), 'duration', {functools.partial(int_or_none, scale=1000)}, any))
or traverse_obj(music_info, ('duration', {int_or_none}))),
'availability': self._availability( 'availability': self._availability(
is_private='Private' in labels, is_private='Private' in labels,
needs_subscription='Friends only' in labels, needs_subscription='Friends only' in labels,
@ -447,23 +449,17 @@ def extract_addr(addr, add_meta={}):
'_format_sort_fields': ('quality', 'codec', 'size', 'br'), '_format_sort_fields': ('quality', 'codec', 'size', 'br'),
} }
def _parse_aweme_video_web(self, aweme_detail, webpage_url, video_id): def _extract_web_formats(self, aweme_detail):
video_info = aweme_detail['video']
author_info = traverse_obj(aweme_detail, 'authorInfo', 'author', expected_type=dict, default={})
music_info = aweme_detail.get('music') or {}
stats_info = aweme_detail.get('stats') or {}
channel_id = traverse_obj(author_info or aweme_detail, (('authorSecId', 'secUid'), {str}), get_all=False)
user_url = self._UPLOADER_URL_FORMAT % channel_id if channel_id else None
formats = []
width = int_or_none(video_info.get('width'))
height = int_or_none(video_info.get('height'))
ratio = try_call(lambda: width / height) or 0.5625
COMMON_FORMAT_INFO = { COMMON_FORMAT_INFO = {
'ext': 'mp4', 'ext': 'mp4',
'vcodec': 'h264', 'vcodec': 'h264',
'acodec': 'aac', 'acodec': 'aac',
} }
video_info = traverse_obj(aweme_detail, ('video', {dict})) or {}
play_width = int_or_none(video_info.get('width'))
play_height = int_or_none(video_info.get('height'))
ratio = try_call(lambda: play_width / play_height) or 0.5625
formats = []
for bitrate_info in traverse_obj(video_info, ('bitrateInfo', lambda _, v: v['PlayAddr']['UrlList'])): for bitrate_info in traverse_obj(video_info, ('bitrateInfo', lambda _, v: v['PlayAddr']['UrlList'])):
format_info, res = self._parse_url_key( format_info, res = self._parse_url_key(
@ -488,7 +484,7 @@ def _parse_aweme_video_web(self, aweme_detail, webpage_url, video_id):
else: # landscape: res/dimension is height else: # landscape: res/dimension is height
x = int(dimension * ratio) x = int(dimension * ratio)
format_info.update({ format_info.update({
'width': x - (x % 2), 'width': x + (x % 2),
'height': dimension, 'height': dimension,
}) })
@ -500,15 +496,15 @@ def _parse_aweme_video_web(self, aweme_detail, webpage_url, video_id):
}) })
# We don't have res string for play formats, but need quality for sorting & de-duplication # We don't have res string for play formats, but need quality for sorting & de-duplication
play_quality = traverse_obj(formats, (lambda _, v: v['width'] == width, 'quality', any)) play_quality = traverse_obj(formats, (lambda _, v: v['width'] == play_width, 'quality', any))
for play_url in traverse_obj(video_info, ('playAddr', ((..., 'src'), None), {url_or_none})): for play_url in traverse_obj(video_info, ('playAddr', ((..., 'src'), None), {url_or_none})):
formats.append({ formats.append({
**COMMON_FORMAT_INFO, **COMMON_FORMAT_INFO,
'format_id': 'play', 'format_id': 'play',
'url': self._proto_relative_url(play_url), 'url': self._proto_relative_url(play_url),
'width': width, 'width': play_width,
'height': height, 'height': play_height,
'quality': play_quality, 'quality': play_quality,
}) })
@ -528,8 +524,8 @@ def _parse_aweme_video_web(self, aweme_detail, webpage_url, video_id):
}) })
# Is it a slideshow with only audio for download? # Is it a slideshow with only audio for download?
if not formats and traverse_obj(music_info, ('playUrl', {url_or_none})): if not formats and traverse_obj(aweme_detail, ('music', 'playUrl', {url_or_none})):
audio_url = music_info['playUrl'] audio_url = aweme_detail['music']['playUrl']
ext = traverse_obj(parse_qs(audio_url), ( ext = traverse_obj(parse_qs(audio_url), (
'mime_type', -1, {lambda x: x.replace('_', '/')}, {mimetype2ext})) or 'm4a' 'mime_type', -1, {lambda x: x.replace('_', '/')}, {mimetype2ext})) or 'm4a'
formats.append({ formats.append({
@ -540,23 +536,31 @@ def _parse_aweme_video_web(self, aweme_detail, webpage_url, video_id):
'vcodec': 'none', 'vcodec': 'none',
}) })
thumbnails = [] return formats
for thumb_url in traverse_obj(aweme_detail, (
(None, 'video'), ('thumbnail', 'cover', 'dynamicCover', 'originCover'), {url_or_none})): def _parse_aweme_video_web(self, aweme_detail, webpage_url, video_id, extract_flat=False):
thumbnails.append({ author_info = traverse_obj(aweme_detail, (('authorInfo', 'author', None), {
'url': self._proto_relative_url(thumb_url), 'channel': ('nickname', {str}),
'width': width, 'channel_id': (('authorSecId', 'secUid'), {str}),
'height': height, 'uploader': (('uniqueId', 'author'), {str}),
}) 'uploader_id': (('authorId', 'uid', 'id'), {str_or_none}),
}), get_all=False)
return { return {
'id': video_id, 'id': video_id,
**traverse_obj(music_info, { 'formats': None if extract_flat else self._extract_web_formats(aweme_detail),
'subtitles': None if extract_flat else self.extract_subtitles(aweme_detail, video_id, None),
'http_headers': {'Referer': webpage_url},
**author_info,
'channel_url': format_field(author_info, 'channel_id', self._UPLOADER_URL_FORMAT, default=None),
'uploader_url': format_field(
author_info, ['uploader', 'uploader_id'], self._UPLOADER_URL_FORMAT, default=None),
**traverse_obj(aweme_detail, ('music', {
'track': ('title', {str}), 'track': ('title', {str}),
'album': ('album', {str}, {lambda x: x or None}), 'album': ('album', {str}, {lambda x: x or None}),
'artists': ('authorName', {str}, {lambda x: [x] if x else None}), 'artists': ('authorName', {str}, {lambda x: re.split(r'(?:, | & )', x) if x else None}),
'duration': ('duration', {int_or_none}), 'duration': ('duration', {int_or_none}),
}), })),
**traverse_obj(aweme_detail, { **traverse_obj(aweme_detail, {
'title': ('desc', {str}), 'title': ('desc', {str}),
'description': ('desc', {str}), 'description': ('desc', {str}),
@ -564,26 +568,17 @@ def _parse_aweme_video_web(self, aweme_detail, webpage_url, video_id):
'duration': ('video', 'duration', {int_or_none}, {lambda x: x or None}), 'duration': ('video', 'duration', {int_or_none}, {lambda x: x or None}),
'timestamp': ('createTime', {int_or_none}), 'timestamp': ('createTime', {int_or_none}),
}), }),
**traverse_obj(author_info or aweme_detail, { **traverse_obj(aweme_detail, ('stats', {
'creators': ('nickname', {str}, {lambda x: [x] if x else None}), # for compat
'channel': ('nickname', {str}),
'uploader': (('uniqueId', 'author'), {str}),
'uploader_id': (('authorId', 'uid', 'id'), {str_or_none}),
}, get_all=False),
**traverse_obj(stats_info, {
'view_count': 'playCount', 'view_count': 'playCount',
'like_count': 'diggCount', 'like_count': 'diggCount',
'repost_count': 'shareCount', 'repost_count': 'shareCount',
'comment_count': 'commentCount', 'comment_count': 'commentCount',
}, expected_type=int_or_none), }), expected_type=int_or_none),
'channel_id': channel_id, 'thumbnails': traverse_obj(aweme_detail, (
'uploader_url': user_url, (None, 'video'), ('thumbnail', 'cover', 'dynamicCover', 'originCover'), {
'formats': formats, 'url': ({url_or_none}, {self._proto_relative_url}),
'subtitles': self.extract_subtitles(aweme_detail, video_id, None), },
'thumbnails': thumbnails, )),
'http_headers': {
'Referer': webpage_url,
}
} }
@ -620,21 +615,21 @@ class TikTokIE(TikTokBaseIE):
'skip': '404 Not Found', 'skip': '404 Not Found',
}, { }, {
'url': 'https://www.tiktok.com/@patroxofficial/video/6742501081818877190?langCountry=en', 'url': 'https://www.tiktok.com/@patroxofficial/video/6742501081818877190?langCountry=en',
'md5': '6f3cf8cdd9b28cb8363fe0a9a160695b', 'md5': 'f21112672ee4ce05ca390fb6522e1b6f',
'info_dict': { 'info_dict': {
'id': '6742501081818877190', 'id': '6742501081818877190',
'ext': 'mp4', 'ext': 'mp4',
'title': 'md5:5e2a23877420bb85ce6521dbee39ba94', 'title': 'md5:5e2a23877420bb85ce6521dbee39ba94',
'description': 'md5:5e2a23877420bb85ce6521dbee39ba94', 'description': 'md5:5e2a23877420bb85ce6521dbee39ba94',
'duration': 27, 'duration': 27,
'height': 960, 'height': 1024,
'width': 540, 'width': 576,
'uploader': 'patrox', 'uploader': 'patrox',
'uploader_id': '18702747', 'uploader_id': '18702747',
'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAiFnldaILebi5heDoVU6bn4jBWWycX6-9U3xuNPqZ8Ws', 'uploader_url': 'https://www.tiktok.com/@patrox',
'channel_url': 'https://www.tiktok.com/@MS4wLjABAAAAiFnldaILebi5heDoVU6bn4jBWWycX6-9U3xuNPqZ8Ws',
'channel_id': 'MS4wLjABAAAAiFnldaILebi5heDoVU6bn4jBWWycX6-9U3xuNPqZ8Ws', 'channel_id': 'MS4wLjABAAAAiFnldaILebi5heDoVU6bn4jBWWycX6-9U3xuNPqZ8Ws',
'channel': 'patroX', 'channel': 'patroX',
'creators': ['patroX'],
'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?', 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
'upload_date': '20190930', 'upload_date': '20190930',
'timestamp': 1569860870, 'timestamp': 1569860870,
@ -646,7 +641,7 @@ class TikTokIE(TikTokBaseIE):
'track': 'Big Fun', 'track': 'Big Fun',
}, },
}, { }, {
# Banned audio, only available on the app # Banned audio, was available on the app, now works with web too
'url': 'https://www.tiktok.com/@barudakhb_/video/6984138651336838402', 'url': 'https://www.tiktok.com/@barudakhb_/video/6984138651336838402',
'info_dict': { 'info_dict': {
'id': '6984138651336838402', 'id': '6984138651336838402',
@ -655,9 +650,9 @@ class TikTokIE(TikTokBaseIE):
'description': 'Balas @yolaaftwsr hayu yu ? #SquadRandom_ 🔥', 'description': 'Balas @yolaaftwsr hayu yu ? #SquadRandom_ 🔥',
'uploader': 'barudakhb_', 'uploader': 'barudakhb_',
'channel': 'md5:29f238c49bc0c176cb3cef1a9cea9fa6', 'channel': 'md5:29f238c49bc0c176cb3cef1a9cea9fa6',
'creators': ['md5:29f238c49bc0c176cb3cef1a9cea9fa6'],
'uploader_id': '6974687867511718913', 'uploader_id': '6974687867511718913',
'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAbhBwQC-R1iKoix6jDFsF-vBdfx2ABoDjaZrM9fX6arU3w71q3cOWgWuTXn1soZ7d', 'uploader_url': 'https://www.tiktok.com/@barudakhb_',
'channel_url': 'https://www.tiktok.com/@MS4wLjABAAAAbhBwQC-R1iKoix6jDFsF-vBdfx2ABoDjaZrM9fX6arU3w71q3cOWgWuTXn1soZ7d',
'channel_id': 'MS4wLjABAAAAbhBwQC-R1iKoix6jDFsF-vBdfx2ABoDjaZrM9fX6arU3w71q3cOWgWuTXn1soZ7d', 'channel_id': 'MS4wLjABAAAAbhBwQC-R1iKoix6jDFsF-vBdfx2ABoDjaZrM9fX6arU3w71q3cOWgWuTXn1soZ7d',
'track': 'Boka Dance', 'track': 'Boka Dance',
'artists': ['md5:29f238c49bc0c176cb3cef1a9cea9fa6'], 'artists': ['md5:29f238c49bc0c176cb3cef1a9cea9fa6'],
@ -680,7 +675,6 @@ class TikTokIE(TikTokBaseIE):
'description': 'Slap and Run!', 'description': 'Slap and Run!',
'uploader': 'user440922249', 'uploader': 'user440922249',
'channel': 'Slap And Run', 'channel': 'Slap And Run',
'creators': ['Slap And Run'],
'uploader_id': '7036055384943690754', 'uploader_id': '7036055384943690754',
'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_', 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_',
'channel_id': 'MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_', 'channel_id': 'MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_',
@ -694,7 +688,7 @@ class TikTokIE(TikTokBaseIE):
'repost_count': int, 'repost_count': int,
'comment_count': int, 'comment_count': int,
}, },
'params': {'skip_download': True}, # XXX: unable to download video data: HTTP Error 403: Forbidden 'skip': 'This video is unavailable',
}, { }, {
# Video without title and description # Video without title and description
'url': 'https://www.tiktok.com/@pokemonlife22/video/7059698374567611694', 'url': 'https://www.tiktok.com/@pokemonlife22/video/7059698374567611694',
@ -705,9 +699,9 @@ class TikTokIE(TikTokBaseIE):
'description': '', 'description': '',
'uploader': 'pokemonlife22', 'uploader': 'pokemonlife22',
'channel': 'Pokemon', 'channel': 'Pokemon',
'creators': ['Pokemon'],
'uploader_id': '6820838815978423302', 'uploader_id': '6820838815978423302',
'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA0tF1nBwQVVMyrGu3CqttkNgM68Do1OXUFuCY0CRQk8fEtSVDj89HqoqvbSTmUP2W', 'uploader_url': 'https://www.tiktok.com/@pokemonlife22',
'channel_url': 'https://www.tiktok.com/@MS4wLjABAAAA0tF1nBwQVVMyrGu3CqttkNgM68Do1OXUFuCY0CRQk8fEtSVDj89HqoqvbSTmUP2W',
'channel_id': 'MS4wLjABAAAA0tF1nBwQVVMyrGu3CqttkNgM68Do1OXUFuCY0CRQk8fEtSVDj89HqoqvbSTmUP2W', 'channel_id': 'MS4wLjABAAAA0tF1nBwQVVMyrGu3CqttkNgM68Do1OXUFuCY0CRQk8fEtSVDj89HqoqvbSTmUP2W',
'track': 'original sound', 'track': 'original sound',
'timestamp': 1643714123, 'timestamp': 1643714123,
@ -752,13 +746,14 @@ class TikTokIE(TikTokBaseIE):
'title': 'TikTok video #7139980461132074283', 'title': 'TikTok video #7139980461132074283',
'description': '', 'description': '',
'channel': 'Antaura', 'channel': 'Antaura',
'creators': ['Antaura'],
'uploader': '_le_cannibale_', 'uploader': '_le_cannibale_',
'uploader_id': '6604511138619654149', 'uploader_id': '6604511138619654149',
'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAoShJqaw_5gvy48y3azFeFcT4jeyKWbB0VVYasOCt2tTLwjNFIaDcHAM4D-QGXFOP', 'uploader_url': 'https://www.tiktok.com/@_le_cannibale_',
'channel_url': 'https://www.tiktok.com/@MS4wLjABAAAAoShJqaw_5gvy48y3azFeFcT4jeyKWbB0VVYasOCt2tTLwjNFIaDcHAM4D-QGXFOP',
'channel_id': 'MS4wLjABAAAAoShJqaw_5gvy48y3azFeFcT4jeyKWbB0VVYasOCt2tTLwjNFIaDcHAM4D-QGXFOP', 'channel_id': 'MS4wLjABAAAAoShJqaw_5gvy48y3azFeFcT4jeyKWbB0VVYasOCt2tTLwjNFIaDcHAM4D-QGXFOP',
'artists': ['nathan !'], 'artists': ['nathan !'],
'track': 'grahamscott canon', 'track': 'grahamscott canon',
'duration': 10,
'upload_date': '20220905', 'upload_date': '20220905',
'timestamp': 1662406249, 'timestamp': 1662406249,
'view_count': int, 'view_count': int,
@ -769,18 +764,18 @@ class TikTokIE(TikTokBaseIE):
}, },
}, { }, {
# only available via web # only available via web
'url': 'https://www.tiktok.com/@moxypatch/video/7206382937372134662', # FIXME 'url': 'https://www.tiktok.com/@moxypatch/video/7206382937372134662',
'md5': '6aba7fad816e8709ff2c149679ace165', 'md5': '4cdefa501ac8ac20bf04986e10916fea',
'info_dict': { 'info_dict': {
'id': '7206382937372134662', 'id': '7206382937372134662',
'ext': 'mp4', 'ext': 'mp4',
'title': 'md5:1d95c0b96560ca0e8a231af4172b2c0a', 'title': 'md5:1d95c0b96560ca0e8a231af4172b2c0a',
'description': 'md5:1d95c0b96560ca0e8a231af4172b2c0a', 'description': 'md5:1d95c0b96560ca0e8a231af4172b2c0a',
'channel': 'MoxyPatch', 'channel': 'MoxyPatch',
'creators': ['MoxyPatch'],
'uploader': 'moxypatch', 'uploader': 'moxypatch',
'uploader_id': '7039142049363379205', 'uploader_id': '7039142049363379205',
'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAFhqKnngMHJSsifL0w1vFOP5kn3Ndo1ODp0XuIBkNMBCkALTvwILdpu12g3pTtL4V', 'uploader_url': 'https://www.tiktok.com/@moxypatch',
'channel_url': 'https://www.tiktok.com/@MS4wLjABAAAAFhqKnngMHJSsifL0w1vFOP5kn3Ndo1ODp0XuIBkNMBCkALTvwILdpu12g3pTtL4V',
'channel_id': 'MS4wLjABAAAAFhqKnngMHJSsifL0w1vFOP5kn3Ndo1ODp0XuIBkNMBCkALTvwILdpu12g3pTtL4V', 'channel_id': 'MS4wLjABAAAAFhqKnngMHJSsifL0w1vFOP5kn3Ndo1ODp0XuIBkNMBCkALTvwILdpu12g3pTtL4V',
'artists': ['your worst nightmare'], 'artists': ['your worst nightmare'],
'track': 'original sound', 'track': 'original sound',
@ -809,7 +804,6 @@ class TikTokIE(TikTokBaseIE):
'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA-0bQT0CqebTRr6I4IkYvMDMKSRSJHLNPBo5HrSklJwyA2psXLSZG5FP-LMNpHnJd', 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA-0bQT0CqebTRr6I4IkYvMDMKSRSJHLNPBo5HrSklJwyA2psXLSZG5FP-LMNpHnJd',
'channel_id': 'MS4wLjABAAAA-0bQT0CqebTRr6I4IkYvMDMKSRSJHLNPBo5HrSklJwyA2psXLSZG5FP-LMNpHnJd', 'channel_id': 'MS4wLjABAAAA-0bQT0CqebTRr6I4IkYvMDMKSRSJHLNPBo5HrSklJwyA2psXLSZG5FP-LMNpHnJd',
'channel': 'tate mcrae', 'channel': 'tate mcrae',
'creators': ['tate mcrae'],
'artists': ['tate mcrae'], 'artists': ['tate mcrae'],
'track': 'original sound', 'track': 'original sound',
'upload_date': '20220609', 'upload_date': '20220609',
@ -821,7 +815,7 @@ class TikTokIE(TikTokBaseIE):
'comment_count': int, 'comment_count': int,
'thumbnail': r're:^https://.+\.webp', 'thumbnail': r're:^https://.+\.webp',
}, },
'skip': 'Unavailable via feed API, no formats available via web', 'skip': 'Unavailable via feed API, only audio available via web',
}, { }, {
# Slideshow, audio-only m4a format # Slideshow, audio-only m4a format
'url': 'https://www.tiktok.com/@hara_yoimiya/video/7253412088251534594', 'url': 'https://www.tiktok.com/@hara_yoimiya/video/7253412088251534594',
@ -833,13 +827,14 @@ class TikTokIE(TikTokBaseIE):
'description': 'я ред флаг простите #переписка #щитпост #тревожныйтиппривязанности #рекомендации ', 'description': 'я ред флаг простите #переписка #щитпост #тревожныйтиппривязанности #рекомендации ',
'uploader': 'hara_yoimiya', 'uploader': 'hara_yoimiya',
'uploader_id': '6582536342634676230', 'uploader_id': '6582536342634676230',
'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAIAlDxriiPWLE-p8p1R_0Bx8qWKfi-7zwmGhzU8Mv25W8sNxjfIKrol31qTczzuLB', 'uploader_url': 'https://www.tiktok.com/@hara_yoimiya',
'channel_url': 'https://www.tiktok.com/@MS4wLjABAAAAIAlDxriiPWLE-p8p1R_0Bx8qWKfi-7zwmGhzU8Mv25W8sNxjfIKrol31qTczzuLB',
'channel_id': 'MS4wLjABAAAAIAlDxriiPWLE-p8p1R_0Bx8qWKfi-7zwmGhzU8Mv25W8sNxjfIKrol31qTczzuLB', 'channel_id': 'MS4wLjABAAAAIAlDxriiPWLE-p8p1R_0Bx8qWKfi-7zwmGhzU8Mv25W8sNxjfIKrol31qTczzuLB',
'channel': 'лампочка', 'channel': 'лампочка(!)',
'creators': ['лампочка'],
'artists': ['Øneheart'], 'artists': ['Øneheart'],
'album': 'watching the stars', 'album': 'watching the stars',
'track': 'watching the stars', 'track': 'watching the stars',
'duration': 60,
'upload_date': '20230708', 'upload_date': '20230708',
'timestamp': 1688816612, 'timestamp': 1688816612,
'view_count': int, 'view_count': int,
@ -876,102 +871,141 @@ def _real_extract(self, url):
class TikTokUserIE(TikTokBaseIE): class TikTokUserIE(TikTokBaseIE):
IE_NAME = 'tiktok:user' IE_NAME = 'tiktok:user'
_VALID_URL = r'https?://(?:www\.)?tiktok\.com/@(?P<id>[\w\.-]+)/?(?:$|[#?])' _VALID_URL = r'(?:tiktokuser:|https?://(?:www\.)?tiktok\.com/@)(?P<id>[\w.-]+)/?(?:$|[#?])'
_WORKING = False
_TESTS = [{ _TESTS = [{
'url': 'https://tiktok.com/@corgibobaa?lang=en', 'url': 'https://tiktok.com/@corgibobaa?lang=en',
'playlist_mincount': 45, 'playlist_mincount': 45,
'info_dict': { 'info_dict': {
'id': '6935371178089399301', 'id': 'MS4wLjABAAAAepiJKgwWhulvCpSuUVsp7sgVVsFJbbNaLeQ6OQ0oAJERGDUIXhb2yxxHZedsItgT',
'title': 'corgibobaa', 'title': 'corgibobaa',
'thumbnail': r're:https://.+_1080x1080\.webp'
}, },
'expected_warnings': ['Retrying']
}, { }, {
'url': 'https://www.tiktok.com/@6820838815978423302', 'url': 'https://www.tiktok.com/@6820838815978423302',
'playlist_mincount': 5, 'playlist_mincount': 5,
'info_dict': { 'info_dict': {
'id': '6820838815978423302', 'id': 'MS4wLjABAAAA0tF1nBwQVVMyrGu3CqttkNgM68Do1OXUFuCY0CRQk8fEtSVDj89HqoqvbSTmUP2W',
'title': '6820838815978423302', 'title': '6820838815978423302',
'thumbnail': r're:https://.+_1080x1080\.webp'
}, },
'expected_warnings': ['Retrying']
}, { }, {
'url': 'https://www.tiktok.com/@meme', 'url': 'https://www.tiktok.com/@meme',
'playlist_mincount': 593, 'playlist_mincount': 593,
'info_dict': { 'info_dict': {
'id': '79005827461758976', 'id': 'MS4wLjABAAAAiKfaDWeCsT3IHwY77zqWGtVRIy9v4ws1HbVi7auP1Vx7dJysU_hc5yRiGywojRD6',
'title': 'meme', 'title': 'meme',
'thumbnail': r're:https://.+_1080x1080\.webp'
}, },
'expected_warnings': ['Retrying'] }, {
'url': 'tiktokuser:MS4wLjABAAAAM3R2BtjzVT-uAtstkl2iugMzC6AtnpkojJbjiOdDDrdsTiTR75-8lyWJCY5VvDrZ',
'playlist_mincount': 31,
'info_dict': {
'id': 'MS4wLjABAAAAM3R2BtjzVT-uAtstkl2iugMzC6AtnpkojJbjiOdDDrdsTiTR75-8lyWJCY5VvDrZ',
},
}] }]
_USER_AGENT = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:115.0) Gecko/20100101 Firefox/115.0'
_API_BASE_URL = 'https://www.tiktok.com/api/creator/item_list/'
r''' # TODO: Fix by adding _signature to api_url def _build_web_query(self, sec_uid, cursor):
def _entries(self, webpage, user_id, username): return {
secuid = self._search_regex(r'\"secUid\":\"(?P<secUid>[^\"]+)', webpage, username) 'aid': '1988',
verifyfp_cookie = self._get_cookies('https://www.tiktok.com').get('s_v_web_id') 'app_language': 'en',
if not verifyfp_cookie: 'app_name': 'tiktok_web',
raise ExtractorError('Improper cookies (missing s_v_web_id).', expected=True) 'browser_language': 'en-US',
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=' 'browser_name': 'Mozilla',
cursor = '0' 'browser_online': 'true',
for page in itertools.count(): 'browser_platform': 'Win32',
data_json = self._download_json(api_url + cursor, username, note='Downloading Page %d' % page) 'browser_version': '5.0 (Windows)',
for video in data_json.get('itemList', []): 'channel': 'tiktok_web',
video_id = video['id'] 'cookie_enabled': 'true',
video_url = f'https://www.tiktok.com/@{user_id}/video/{video_id}' 'count': '15',
yield self._url_result(video_url, 'TikTok', video_id, str_or_none(video.get('desc'))) 'cursor': cursor,
if not data_json.get('hasMore'): 'device_id': self._DEVICE_ID,
break 'device_platform': 'web_pc',
cursor = data_json['cursor'] 'focus_state': 'true',
''' 'from_page': 'user',
'history_len': '2',
def _video_entries_api(self, webpage, user_id, username): 'is_fullscreen': 'false',
query = { 'is_page_visible': 'true',
'user_id': user_id, 'language': 'en',
'count': 21, 'os': 'windows',
'max_cursor': 0, 'priority_region': '',
'min_cursor': 0, 'referer': '',
'retry_type': 'no_retry', 'region': 'US',
'device_id': self._DEVICE_ID, # Some endpoints don't like randomized device_id, so it isn't directly set in _call_api. 'screen_height': '1080',
'screen_width': '1920',
'secUid': sec_uid,
'type': '1', # pagination type: 0 == oldest-to-newest, 1 == newest-to-oldest
'tz_name': 'UTC',
'verifyFp': f'verify_{"".join(random.choices(string.hexdigits, k=7))}',
'webcast_language': 'en',
} }
for page in itertools.count(1): def _entries(self, sec_uid, user_name):
for retry in self.RetryManager(): display_id = user_name or sec_uid
try:
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:
retry.error = e
continue
raise
yield from post_list.get('aweme_list', [])
if not post_list.get('has_more'):
break
query['max_cursor'] = post_list['max_cursor']
def _entries_api(self, user_id, videos): cursor = int(time.time() * 1E3)
for video in videos: for page in itertools.count(1):
yield { response = self._download_json(
**self._parse_aweme_video_app(video), self._API_BASE_URL, display_id, f'Downloading page {page}',
'extractor_key': TikTokIE.ie_key(), query=self._build_web_query(sec_uid, cursor), headers={'User-Agent': self._USER_AGENT})
'extractor': 'TikTok',
'webpage_url': f'https://tiktok.com/@{user_id}/video/{video["aweme_id"]}', for video in traverse_obj(response, ('itemList', lambda _, v: v['id'])):
} video_id = video['id']
webpage_url = self._create_url(display_id, video_id)
yield self.url_result(
webpage_url, TikTokIE,
**self._parse_aweme_video_web(video, webpage_url, video_id, extract_flat=True))
old_cursor = cursor
cursor = traverse_obj(
response, ('itemList', -1, 'createTime', {functools.partial(int_or_none, invscale=1E3)}))
if not cursor:
# User may not have posted within this ~1 week lookback, so manually adjust cursor
cursor = old_cursor - 7 * 86_400_000
# In case 'hasMorePrevious' is wrong, break if we have gone back before TikTok existed
if cursor < 1472706000000 or not traverse_obj(response, 'hasMorePrevious'):
break
def _get_sec_uid(self, user_url, user_name, msg):
webpage = self._download_webpage(
user_url, user_name, fatal=False, headers={'User-Agent': 'Mozilla/5.0'},
note=f'Downloading {msg} webpage', errnote=f'Unable to download {msg} webpage') or ''
return (traverse_obj(self._get_universal_data(webpage, user_name),
('webapp.user-detail', 'userInfo', 'user', 'secUid', {str}))
or traverse_obj(self._get_sigi_state(webpage, user_name),
('LiveRoom', 'liveRoomUserInfo', 'user', 'secUid', {str}),
('UserModule', 'users', ..., 'secUid', {str}, any)))
def _real_extract(self, url): def _real_extract(self, url):
user_name = self._match_id(url) user_name, sec_uid = self._match_id(url), None
webpage = self._download_webpage(url, user_name, headers={ if mobj := re.fullmatch(r'MS4wLjABAAAA[\w-]{64}', user_name):
'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)' user_name, sec_uid = None, mobj.group(0)
}) else:
user_id = self._html_search_regex(r'snssdk\d*://user/profile/(\d+)', webpage, 'user ID', default=None) or user_name sec_uid = (self._get_sec_uid(self._UPLOADER_URL_FORMAT % user_name, user_name, 'user')
or self._get_sec_uid(self._UPLOADER_URL_FORMAT % f'{user_name}/live', user_name, 'live'))
videos = LazyList(self._video_entries_api(webpage, user_id, user_name)) if not sec_uid:
thumbnail = traverse_obj(videos, (0, 'author', 'avatar_larger', 'url_list', 0)) webpage = self._download_webpage(
f'https://www.tiktok.com/embed/@{user_name}', user_name,
note='Downloading user embed page', fatal=False) or ''
data = traverse_obj(self._search_json(
r'<script[^>]+\bid=[\'"]__FRONTITY_CONNECT_STATE__[\'"][^>]*>',
webpage, 'data', user_name, default={}),
('source', 'data', f'/embed/@{user_name}', {dict}))
return self.playlist_result(self._entries_api(user_id, videos), user_id, user_name, thumbnail=thumbnail) for aweme_id in traverse_obj(data, ('videoList', ..., 'id', {str})):
webpage_url = self._create_url(user_name, aweme_id)
video_data, _ = self._extract_web_data_and_status(webpage_url, aweme_id, fatal=False)
sec_uid = self._parse_aweme_video_web(
video_data, webpage_url, aweme_id, extract_flat=True).get('channel_id')
if sec_uid:
break
if not sec_uid:
raise ExtractorError(
'Unable to extract secondary user ID. If you are able to get the channel_id '
'from a video posted by this user, try using "tiktokuser:channel_id" as the '
'input URL (replacing `channel_id` with its actual value)', expected=True)
return self.playlist_result(self._entries(sec_uid, user_name), sec_uid, user_name)
class TikTokBaseListIE(TikTokBaseIE): # XXX: Conventionally, base classes should end with BaseIE/InfoExtractor class TikTokBaseListIE(TikTokBaseIE): # XXX: Conventionally, base classes should end with BaseIE/InfoExtractor
@ -1098,7 +1132,6 @@ class DouyinIE(TikTokBaseIE):
'uploader_url': 'https://www.douyin.com/user/MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98', 'uploader_url': 'https://www.douyin.com/user/MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98',
'channel_id': 'MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98', 'channel_id': 'MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98',
'channel': '杨超越', 'channel': '杨超越',
'creators': ['杨超越'],
'duration': 19, 'duration': 19,
'timestamp': 1620905839, 'timestamp': 1620905839,
'upload_date': '20210513', 'upload_date': '20210513',
@ -1123,7 +1156,6 @@ class DouyinIE(TikTokBaseIE):
'uploader_url': 'https://www.douyin.com/user/MS4wLjABAAAAZJpnglcjW2f_CMVcnqA_6oVBXKWMpH0F8LIHuUu8-lA', 'uploader_url': 'https://www.douyin.com/user/MS4wLjABAAAAZJpnglcjW2f_CMVcnqA_6oVBXKWMpH0F8LIHuUu8-lA',
'channel_id': 'MS4wLjABAAAAZJpnglcjW2f_CMVcnqA_6oVBXKWMpH0F8LIHuUu8-lA', 'channel_id': 'MS4wLjABAAAAZJpnglcjW2f_CMVcnqA_6oVBXKWMpH0F8LIHuUu8-lA',
'channel': '杨超越工作室', 'channel': '杨超越工作室',
'creators': ['杨超越工作室'],
'duration': 42, 'duration': 42,
'timestamp': 1625739481, 'timestamp': 1625739481,
'upload_date': '20210708', 'upload_date': '20210708',
@ -1148,7 +1180,6 @@ class DouyinIE(TikTokBaseIE):
'uploader_url': 'https://www.douyin.com/user/MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98', 'uploader_url': 'https://www.douyin.com/user/MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98',
'channel_id': 'MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98', 'channel_id': 'MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98',
'channel': '杨超越', 'channel': '杨超越',
'creators': ['杨超越'],
'duration': 17, 'duration': 17,
'timestamp': 1619098692, 'timestamp': 1619098692,
'upload_date': '20210422', 'upload_date': '20210422',
@ -1190,7 +1221,6 @@ class DouyinIE(TikTokBaseIE):
'uploader_url': 'https://www.douyin.com/user/MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98', 'uploader_url': 'https://www.douyin.com/user/MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98',
'channel_id': 'MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98', 'channel_id': 'MS4wLjABAAAAEKnfa654JAJ_N5lgZDQluwsxmY0lhfmEYNQBBkwGG98',
'channel': '杨超越', 'channel': '杨超越',
'creators': ['杨超越'],
'duration': 15, 'duration': 15,
'timestamp': 1621261163, 'timestamp': 1621261163,
'upload_date': '20210517', 'upload_date': '20210517',