[yt-dlp.git] / yt_dlp / extractor / triller.py

import itertools
import json
import re

from .common import InfoExtractor
from ..networking import HEADRequest
from ..utils import (
    ExtractorError,
    UnsupportedError,
    determine_ext,
    int_or_none,
    parse_resolution,
    str_or_none,
    traverse_obj,
    unified_timestamp,
    url_basename,
    url_or_none,
    urljoin,
)


class TrillerBaseIE(InfoExtractor):
    _NETRC_MACHINE = 'triller'
    _API_BASE_URL = 'https://social.triller.co/v1.5'
    _API_HEADERS = {'Origin': 'https://triller.co'}

    def _perform_login(self, username, password):
        if self._API_HEADERS.get('Authorization'):
            return

        headers = {**self._API_HEADERS, 'Content-Type': 'application/json'}
        user_check = traverse_obj(self._download_json(
            f'{self._API_BASE_URL}/api/user/is-valid-username', None, note='Checking username',
            fatal=False, expected_status=400, headers=headers,
            data=json.dumps({'username': username}, separators=(',', ':')).encode()), 'status')

        if user_check:  # endpoint returns `"status":false` if username exists
            raise ExtractorError('Unable to login: Invalid username', expected=True)

        login = self._download_json(
            f'{self._API_BASE_URL}/user/auth', None, note='Logging in', fatal=False,
            expected_status=400, headers=headers, data=json.dumps({
                'username': username,
                'password': password,
            }, separators=(',', ':')).encode()) or {}

        if not login.get('auth_token'):
            if login.get('error') == 1008:
                raise ExtractorError('Unable to login: Incorrect password', expected=True)
            raise ExtractorError('Unable to login')

        self._API_HEADERS['Authorization'] = f'Bearer {login["auth_token"]}'

    def _get_comments(self, video_id, limit=15):
        comment_info = self._download_json(
            f'{self._API_BASE_URL}/api/videos/{video_id}/comments_v2',
            video_id, fatal=False, note='Downloading comments API JSON',
            headers=self._API_HEADERS, query={'limit': limit}) or {}
        if not comment_info.get('comments'):
            return
        yield from traverse_obj(comment_info, ('comments', ..., {
            'id': ('id', {str_or_none}),
            'text': 'body',
            'author': ('author', 'username'),
            'author_id': ('author', 'user_id'),
            'timestamp': ('timestamp', {unified_timestamp}),
        }))

    def _parse_video_info(self, video_info, username, user_id, display_id=None):
        video_id = str(video_info['id'])
        display_id = display_id or video_info.get('video_uuid')

        if traverse_obj(video_info, (
                None, ('transcoded_url', 'video_url', 'stream_url', 'audio_url'),
                {lambda x: re.search(r'/copyright/', x)}), get_all=False):
            self.raise_no_formats('This video has been removed due to licensing restrictions', expected=True)

        def format_info(url):
            return {
                'url': url,
                'ext': determine_ext(url),
                'format_id': url_basename(url).split('.')[0],
            }

        formats = []

        if determine_ext(video_info.get('transcoded_url')) == 'm3u8':
            formats.extend(self._extract_m3u8_formats(
                video_info['transcoded_url'], video_id, 'mp4', m3u8_id='hls', fatal=False))

        for video in traverse_obj(video_info, ('video_set', lambda _, v: url_or_none(v['url']))):
            formats.append({
                **format_info(video['url']),
                **parse_resolution(video.get('resolution')),
                'vcodec': video.get('codec'),
                'vbr': int_or_none(video.get('bitrate'), 1000),
            })

        video_url = traverse_obj(video_info, 'video_url', 'stream_url', expected_type=url_or_none)
        if video_url:
            formats.append({
                **format_info(video_url),
                'vcodec': 'h264',
                **traverse_obj(video_info, {
                    'width': 'width',
                    'height': 'height',
                    'filesize': 'filesize',
                }, expected_type=int_or_none),
            })

        audio_url = url_or_none(video_info.get('audio_url'))
        if audio_url:
            formats.append(format_info(audio_url))

        comment_count = traverse_obj(video_info, ('comment_count', {int_or_none}))

        return {
            'id': video_id,
            'display_id': display_id,
            'uploader': username,
            'uploader_id': user_id or traverse_obj(video_info, ('user', 'user_id', {str_or_none})),
            'webpage_url': urljoin(f'https://triller.co/@{username}/video/', display_id),
            'uploader_url': f'https://triller.co/@{username}',
            'extractor_key': TrillerIE.ie_key(),
            'extractor': TrillerIE.IE_NAME,
            'formats': formats,
            'comment_count': comment_count,
            '__post_extractor': self.extract_comments(video_id, comment_count),
            **traverse_obj(video_info, {
                'title': ('description', {lambda x: x.replace('\r\n', ' ')}),
                'description': 'description',
                'creator': ((('user'), ('users', lambda _, v: str(v['user_id']) == user_id)), 'name'),
                'thumbnail': ('thumbnail_url', {url_or_none}),
                'timestamp': ('timestamp', {unified_timestamp}),
                'duration': ('duration', {int_or_none}),
                'view_count': ('play_count', {int_or_none}),
                'like_count': ('likes_count', {int_or_none}),
                'artist': 'song_artist',
                'track': 'song_title',
            }, get_all=False),
        }


class TrillerIE(TrillerBaseIE):
    _VALID_URL = r'''(?x)
            https?://(?:www\.)?triller\.co/
            @(?P<username>[\w.]+)/video/(?P<id>[\da-f]{8}-(?:[\da-f]{4}-){3}[\da-f]{12})
        '''
    _TESTS = [{
        'url': 'https://triller.co/@theestallion/video/2358fcd7-3df2-4c77-84c8-1d091610a6cf',
        'md5': '228662d783923b60d78395fedddc0a20',
        'info_dict': {
            'id': '71595734',
            'ext': 'mp4',
            'title': 'md5:9a2bf9435c5c4292678996a464669416',
            'thumbnail': r're:^https://uploads\.cdn\.triller\.co/.+\.jpg$',
            'description': 'md5:9a2bf9435c5c4292678996a464669416',
            'uploader': 'theestallion',
            'uploader_id': '18992236',
            'creator': 'Megan Thee Stallion',
            'timestamp': 1660598222,
            'upload_date': '20220815',
            'duration': 47,
            'view_count': int,
            'like_count': int,
            'artist': 'Megan Thee Stallion',
            'track': 'Her',
            'uploader_url': 'https://triller.co/@theestallion',
            'comment_count': int,
        },
        'skip': 'This video has been removed due to licensing restrictions',
    }, {
        'url': 'https://triller.co/@charlidamelio/video/46c6fcfa-aa9e-4503-a50c-68444f44cddc',
        'md5': '874055f462af5b0699b9dbb527a505a0',
        'info_dict': {
            'id': '71621339',
            'ext': 'mp4',
            'title': 'md5:4c91ea82760fe0fffb71b8c3aa7295fc',
            'display_id': '46c6fcfa-aa9e-4503-a50c-68444f44cddc',
            'thumbnail': r're:^https://uploads\.cdn\.triller\.co/.+\.jpg$',
            'description': 'md5:4c91ea82760fe0fffb71b8c3aa7295fc',
            'uploader': 'charlidamelio',
            'uploader_id': '1875551',
            'creator': 'charli damelio',
            'timestamp': 1660773354,
            'upload_date': '20220817',
            'duration': 16,
            'view_count': int,
            'like_count': int,
            'artist': 'Dixie',
            'track': 'Someone to Blame',
            'uploader_url': 'https://triller.co/@charlidamelio',
            'comment_count': int,
        },
    }, {
        'url': 'https://triller.co/@theestallion/video/07f35f38-1f51-48e2-8c5f-f7a8e829988f',
        'md5': 'af7b3553e4b8bfca507636471ee2eb41',
        'info_dict': {
            'id': '71837829',
            'ext': 'mp4',
            'title': 'UNGRATEFUL VIDEO OUT NOW 👏🏾👏🏾👏🏾 💙💙 link my bio  #womeninhiphop',
            'display_id': '07f35f38-1f51-48e2-8c5f-f7a8e829988f',
            'thumbnail': r're:^https://uploads\.cdn\.triller\.co/.+\.jpg$',
            'description': 'UNGRATEFUL VIDEO OUT NOW 👏🏾👏🏾👏🏾 💙💙 link my bio\r\n #womeninhiphop',
            'uploader': 'theestallion',
            'uploader_id': '18992236',
            'creator': 'Megan Thee Stallion',
            'timestamp': 1662486178,
            'upload_date': '20220906',
            'duration': 30,
            'view_count': int,
            'like_count': int,
            'artist': 'Unknown',
            'track': 'Unknown',
            'uploader_url': 'https://triller.co/@theestallion',
            'comment_count': int,
        },
    }]

    def _real_extract(self, url):
        username, display_id = self._match_valid_url(url).group('username', 'id')

        video_info = self._download_json(
            f'{self._API_BASE_URL}/api/videos/{display_id}', display_id,
            headers=self._API_HEADERS)['videos'][0]

        return self._parse_video_info(video_info, username, None, display_id)


class TrillerUserIE(TrillerBaseIE):
    _VALID_URL = r'https?://(?:www\.)?triller\.co/@(?P<id>[\w.]+)/?(?:$|[#?])'
    _TESTS = [{
        'url': 'https://triller.co/@theestallion',
        'playlist_mincount': 12,
        'info_dict': {
            'id': '18992236',
            'title': 'theestallion',
            'thumbnail': r're:^https://uploads\.cdn\.triller\.co/.+\.jpg$',
        },
    }, {
        'url': 'https://triller.co/@charlidamelio',
        'playlist_mincount': 150,
        'info_dict': {
            'id': '1875551',
            'title': 'charlidamelio',
            'thumbnail': r're:^https://uploads\.cdn\.triller\.co/.+\.jpg$',
        },
    }]

    def _real_initialize(self):
        if not self._API_HEADERS.get('Authorization'):
            guest = self._download_json(
                f'{self._API_BASE_URL}/user/create_guest', None,
                note='Creating guest session', data=b'', headers=self._API_HEADERS, query={
                    'platform': 'Web',
                    'app_version': '',
                })
            if not guest.get('auth_token'):
                raise ExtractorError('Unable to fetch required auth token for user extraction')

            self._API_HEADERS['Authorization'] = f'Bearer {guest["auth_token"]}'

    def _entries(self, username, user_id, limit=6):
        query = {'limit': limit}
        for page in itertools.count(1):
            videos = self._download_json(
                f'{self._API_BASE_URL}/api/users/{user_id}/videos',
                username, note=f'Downloading user video list page {page}',
                headers=self._API_HEADERS, query=query)

            for video in traverse_obj(videos, ('videos', ...)):
                yield self._parse_video_info(video, username, user_id)

            query['before_time'] = traverse_obj(videos, ('videos', -1, 'timestamp'))
            if not query['before_time']:
                break

    def _real_extract(self, url):
        username = self._match_id(url)

        user_info = traverse_obj(self._download_json(
            f'{self._API_BASE_URL}/api/users/by_username/{username}',
            username, note='Downloading user info', headers=self._API_HEADERS), ('user', {dict})) or {}

        if user_info.get('private') and user_info.get('followed_by_me') not in (True, 'true'):
            raise ExtractorError('This user profile is private', expected=True)
        elif traverse_obj(user_info, (('blocked_by_user', 'blocking_user'), {bool}), get_all=False):
            raise ExtractorError('The author of the video is blocked', expected=True)

        user_id = str_or_none(user_info.get('user_id'))
        if not user_id:
            raise ExtractorError('Unable to extract user ID')

        return self.playlist_result(
            self._entries(username, user_id), user_id, username, thumbnail=user_info.get('avatar_url'))


class TrillerShortIE(InfoExtractor):
    _VALID_URL = r'https?://v\.triller\.co/(?P<id>\w+)'
    _TESTS = [{
        'url': 'https://v.triller.co/WWZNWk',
        'md5': '5eb8dc2c971bd8cd794ec9e8d5e9d101',
        'info_dict': {
            'id': '66210052',
            'ext': 'mp4',
            'title': 'md5:2dfc89d154cd91a4a18cd9582ba03e16',
            'display_id': 'f4480e1f-fb4e-45b9-a44c-9e6c679ce7eb',
            'thumbnail': r're:^https://uploads\.cdn\.triller\.co/.+\.jpg$',
            'description': 'md5:2dfc89d154cd91a4a18cd9582ba03e16',
            'uploader': 'statefairent',
            'uploader_id': '487545193',
            'creator': 'Official Summer Fair of LA',
            'timestamp': 1629655457,
            'upload_date': '20210822',
            'duration': 19,
            'view_count': int,
            'like_count': int,
            'artist': 'Unknown',
            'track': 'Unknown',
            'uploader_url': 'https://triller.co/@statefairent',
            'comment_count': int,
        },
    }]

    def _real_extract(self, url):
        real_url = self._request_webpage(HEADRequest(url), self._match_id(url)).url
        if self.suitable(real_url):  # Prevent infinite loop in case redirect fails
            raise UnsupportedError(real_url)
        return self.url_result(real_url)
Commit	Line	Data
92aa6d68	1	import itertools
92aa6d68	2	import json
33b737be	3	import re
92aa6d68	4
92aa6d68	5	from .common import InfoExtractor
3d2623a8	6	from ..networking import HEADRequest
92aa6d68	7	from ..utils import (
d2c8aadf	8	ExtractorError,
33b737be	9	UnsupportedError,
33b737be	10	determine_ext,
92aa6d68	11	int_or_none,
33b737be	12	parse_resolution,
92aa6d68	13	str_or_none,
92aa6d68	14	traverse_obj,
92aa6d68	15	unified_timestamp,
92aa6d68	16	url_basename,
33b737be	17	url_or_none,
e897bd82	18	urljoin,
92aa6d68	19	)
	20
	21
	22	class TrillerBaseIE(InfoExtractor):
	23	_NETRC_MACHINE = 'triller'
92aa6d68	24	_API_BASE_URL = 'https://social.triller.co/v1.5'
d6f88719	25	_API_HEADERS = {'Origin': 'https://triller.co'}
92aa6d68	26
92aa6d68	27	def _perform_login(self, username, password):
d6f88719	28	if self._API_HEADERS.get('Authorization'):
92aa6d68	29	return
92aa6d68	30
33b737be	31	headers = {**self._API_HEADERS, 'Content-Type': 'application/json'}
33b737be	32	user_check = traverse_obj(self._download_json(
92aa6d68	33	f'{self._API_BASE_URL}/api/user/is-valid-username', None, note='Checking username',
33b737be	34	fatal=False, expected_status=400, headers=headers,
	35	data=json.dumps({'username': username}, separators=(',', ':')).encode()), 'status')
	36
	37	if user_check: # endpoint returns `"status":false` if username exists
92aa6d68	38	raise ExtractorError('Unable to login: Invalid username', expected=True)
92aa6d68	39
92aa6d68	40	login = self._download_json(
33b737be	41	f'{self._API_BASE_URL}/user/auth', None, note='Logging in', fatal=False,
	42	expected_status=400, headers=headers, data=json.dumps({
	43	'username': username,
	44	'password': password,
	45	}, separators=(',', ':')).encode()) or {}
	46
92aa6d68	47	if not login.get('auth_token'):
	48	if login.get('error') == 1008:
	49	raise ExtractorError('Unable to login: Incorrect password', expected=True)
	50	raise ExtractorError('Unable to login')
	51
d6f88719	52	self._API_HEADERS['Authorization'] = f'Bearer {login["auth_token"]}'
92aa6d68	53
	54	def _get_comments(self, video_id, limit=15):
	55	comment_info = self._download_json(
	56	f'{self._API_BASE_URL}/api/videos/{video_id}/comments_v2',
	57	video_id, fatal=False, note='Downloading comments API JSON',
d6f88719	58	headers=self._API_HEADERS, query={'limit': limit}) or {}
92aa6d68	59	if not comment_info.get('comments'):
92aa6d68	60	return
33b737be	61	yield from traverse_obj(comment_info, ('comments', ..., {
	62	'id': ('id', {str_or_none}),
	63	'text': 'body',
	64	'author': ('author', 'username'),
	65	'author_id': ('author', 'user_id'),
	66	'timestamp': ('timestamp', {unified_timestamp}),
	67	}))
92aa6d68	68
33b737be	69	def _parse_video_info(self, video_info, username, user_id, display_id=None):
	70	video_id = str(video_info['id'])
	71	display_id = display_id or video_info.get('video_uuid')
	72
	73	if traverse_obj(video_info, (
	74	None, ('transcoded_url', 'video_url', 'stream_url', 'audio_url'),
	75	{lambda x: re.search(r'/copyright/', x)}), get_all=False):
	76	self.raise_no_formats('This video has been removed due to licensing restrictions', expected=True)
	77
	78	def format_info(url):
	79	return {
	80	'url': url,
	81	'ext': determine_ext(url),
	82	'format_id': url_basename(url).split('.')[0],
	83	}
92aa6d68	84
92aa6d68	85	formats = []
33b737be	86
	87	if determine_ext(video_info.get('transcoded_url')) == 'm3u8':
	88	formats.extend(self._extract_m3u8_formats(
	89	video_info['transcoded_url'], video_id, 'mp4', m3u8_id='hls', fatal=False))
	90
	91	for video in traverse_obj(video_info, ('video_set', lambda _, v: url_or_none(v['url']))):
92aa6d68	92	formats.append({
33b737be	93	**format_info(video['url']),
33b737be	94	**parse_resolution(video.get('resolution')),
92aa6d68	95	'vcodec': video.get('codec'),
92aa6d68	96	'vbr': int_or_none(video.get('bitrate'), 1000),
92aa6d68	97	})
33b737be	98
	99	video_url = traverse_obj(video_info, 'video_url', 'stream_url', expected_type=url_or_none)
	100	if video_url:
92aa6d68	101	formats.append({
33b737be	102	**format_info(video_url),
	103	'vcodec': 'h264',
	104	**traverse_obj(video_info, {
	105	'width': 'width',
	106	'height': 'height',
	107	'filesize': 'filesize',
	108	}, expected_type=int_or_none),
92aa6d68	109	})
92aa6d68	110
33b737be	111	audio_url = url_or_none(video_info.get('audio_url'))
	112	if audio_url:
	113	formats.append(format_info(audio_url))
92aa6d68	114
33b737be	115	comment_count = traverse_obj(video_info, ('comment_count', {int_or_none}))
92aa6d68	116
92aa6d68	117	return {
33b737be	118	'id': video_id,
	119	'display_id': display_id,
	120	'uploader': username,
	121	'uploader_id': user_id or traverse_obj(video_info, ('user', 'user_id', {str_or_none})),
	122	'webpage_url': urljoin(f'https://triller.co/@{username}/video/', display_id),
92aa6d68	123	'uploader_url': f'https://triller.co/@{username}',
	124	'extractor_key': TrillerIE.ie_key(),
	125	'extractor': TrillerIE.IE_NAME,
	126	'formats': formats,
	127	'comment_count': comment_count,
	128	'__post_extractor': self.extract_comments(video_id, comment_count),
33b737be	129	**traverse_obj(video_info, {
	130	'title': ('description', {lambda x: x.replace('\r\n', ' ')}),
	131	'description': 'description',
	132	'creator': ((('user'), ('users', lambda _, v: str(v['user_id']) == user_id)), 'name'),
	133	'thumbnail': ('thumbnail_url', {url_or_none}),
	134	'timestamp': ('timestamp', {unified_timestamp}),
	135	'duration': ('duration', {int_or_none}),
	136	'view_count': ('play_count', {int_or_none}),
	137	'like_count': ('likes_count', {int_or_none}),
	138	'artist': 'song_artist',
	139	'track': 'song_title',
	140	}, get_all=False),
92aa6d68	141	}
	142
	143
	144	class TrillerIE(TrillerBaseIE):
	145	_VALID_URL = r'''(?x)
	146	https?://(?:www\.)?triller\.co/
33b737be	147	@(?P<username>[\w.]+)/video/(?P<id>[\da-f]{8}-(?:[\da-f]{4}-){3}[\da-f]{12})
92aa6d68	148	'''
	149	_TESTS = [{
	150	'url': 'https://triller.co/@theestallion/video/2358fcd7-3df2-4c77-84c8-1d091610a6cf',
	151	'md5': '228662d783923b60d78395fedddc0a20',
	152	'info_dict': {
	153	'id': '71595734',
	154	'ext': 'mp4',
	155	'title': 'md5:9a2bf9435c5c4292678996a464669416',
	156	'thumbnail': r're:^https://uploads\.cdn\.triller\.co/.+\.jpg$',
	157	'description': 'md5:9a2bf9435c5c4292678996a464669416',
	158	'uploader': 'theestallion',
	159	'uploader_id': '18992236',
	160	'creator': 'Megan Thee Stallion',
	161	'timestamp': 1660598222,
	162	'upload_date': '20220815',
	163	'duration': 47,
92aa6d68	164	'view_count': int,
	165	'like_count': int,
	166	'artist': 'Megan Thee Stallion',
	167	'track': 'Her',
92aa6d68	168	'uploader_url': 'https://triller.co/@theestallion',
92aa6d68	169	'comment_count': int,
33b737be	170	},
33b737be	171	'skip': 'This video has been removed due to licensing restrictions',
92aa6d68	172	}, {
	173	'url': 'https://triller.co/@charlidamelio/video/46c6fcfa-aa9e-4503-a50c-68444f44cddc',
	174	'md5': '874055f462af5b0699b9dbb527a505a0',
	175	'info_dict': {
	176	'id': '71621339',
	177	'ext': 'mp4',
	178	'title': 'md5:4c91ea82760fe0fffb71b8c3aa7295fc',
33b737be	179	'display_id': '46c6fcfa-aa9e-4503-a50c-68444f44cddc',
92aa6d68	180	'thumbnail': r're:^https://uploads\.cdn\.triller\.co/.+\.jpg$',
	181	'description': 'md5:4c91ea82760fe0fffb71b8c3aa7295fc',
	182	'uploader': 'charlidamelio',
	183	'uploader_id': '1875551',
	184	'creator': 'charli damelio',
	185	'timestamp': 1660773354,
	186	'upload_date': '20220817',
	187	'duration': 16,
92aa6d68	188	'view_count': int,
	189	'like_count': int,
	190	'artist': 'Dixie',
	191	'track': 'Someone to Blame',
92aa6d68	192	'uploader_url': 'https://triller.co/@charlidamelio',
92aa6d68	193	'comment_count': int,
33b737be	194	},
	195	}, {
	196	'url': 'https://triller.co/@theestallion/video/07f35f38-1f51-48e2-8c5f-f7a8e829988f',
	197	'md5': 'af7b3553e4b8bfca507636471ee2eb41',
	198	'info_dict': {
	199	'id': '71837829',
	200	'ext': 'mp4',
	201	'title': 'UNGRATEFUL VIDEO OUT NOW 👏🏾👏🏾👏🏾 💙💙 link my bio #womeninhiphop',
	202	'display_id': '07f35f38-1f51-48e2-8c5f-f7a8e829988f',
	203	'thumbnail': r're:^https://uploads\.cdn\.triller\.co/.+\.jpg$',
	204	'description': 'UNGRATEFUL VIDEO OUT NOW 👏🏾👏🏾👏🏾 💙💙 link my bio\r\n #womeninhiphop',
	205	'uploader': 'theestallion',
	206	'uploader_id': '18992236',
	207	'creator': 'Megan Thee Stallion',
	208	'timestamp': 1662486178,
	209	'upload_date': '20220906',
	210	'duration': 30,
	211	'view_count': int,
	212	'like_count': int,
	213	'artist': 'Unknown',
	214	'track': 'Unknown',
	215	'uploader_url': 'https://triller.co/@theestallion',
	216	'comment_count': int,
	217	},
92aa6d68	218	}]
	219
	220	def _real_extract(self, url):
33b737be	221	username, display_id = self._match_valid_url(url).group('username', 'id')
	222
	223	video_info = self._download_json(
	224	f'{self._API_BASE_URL}/api/videos/{display_id}', display_id,
	225	headers=self._API_HEADERS)['videos'][0]
92aa6d68	226
33b737be	227	return self._parse_video_info(video_info, username, None, display_id)
92aa6d68	228
	229
	230	class TrillerUserIE(TrillerBaseIE):
33b737be	231	_VALID_URL = r'https?://(?:www\.)?triller\.co/@(?P<id>[\w.]+)/?(?:$\|[#?])'
92aa6d68	232	_TESTS = [{
92aa6d68	233	'url': 'https://triller.co/@theestallion',
33b737be	234	'playlist_mincount': 12,
92aa6d68	235	'info_dict': {
	236	'id': '18992236',
	237	'title': 'theestallion',
	238	'thumbnail': r're:^https://uploads\.cdn\.triller\.co/.+\.jpg$',
33b737be	239	},
92aa6d68	240	}, {
92aa6d68	241	'url': 'https://triller.co/@charlidamelio',
33b737be	242	'playlist_mincount': 150,
92aa6d68	243	'info_dict': {
	244	'id': '1875551',
	245	'title': 'charlidamelio',
	246	'thumbnail': r're:^https://uploads\.cdn\.triller\.co/.+\.jpg$',
33b737be	247	},
92aa6d68	248	}]
	249
	250	def _real_initialize(self):
d6f88719	251	if not self._API_HEADERS.get('Authorization'):
92aa6d68	252	guest = self._download_json(
33b737be	253	f'{self._API_BASE_URL}/user/create_guest', None,
33b737be	254	note='Creating guest session', data=b'', headers=self._API_HEADERS, query={
92aa6d68	255	'platform': 'Web',
	256	'app_version': '',
	257	})
	258	if not guest.get('auth_token'):
	259	raise ExtractorError('Unable to fetch required auth token for user extraction')
	260
d6f88719	261	self._API_HEADERS['Authorization'] = f'Bearer {guest["auth_token"]}'
92aa6d68	262
33b737be	263	def _entries(self, username, user_id, limit=6):
33b737be	264	query = {'limit': limit}
92aa6d68	265	for page in itertools.count(1):
33b737be	266	videos = self._download_json(
	267	f'{self._API_BASE_URL}/api/users/{user_id}/videos',
	268	username, note=f'Downloading user video list page {page}',
	269	headers=self._API_HEADERS, query=query)
	270
	271	for video in traverse_obj(videos, ('videos', ...)):
	272	yield self._parse_video_info(video, username, user_id)
	273
	274	query['before_time'] = traverse_obj(videos, ('videos', -1, 'timestamp'))
92aa6d68	275	if not query['before_time']:
	276	break
	277
92aa6d68	278	def _real_extract(self, url):
92aa6d68	279	username = self._match_id(url)
33b737be	280
39837ae3	281	user_info = traverse_obj(self._download_json(
92aa6d68	282	f'{self._API_BASE_URL}/api/users/by_username/{username}',
39837ae3	283	username, note='Downloading user info', headers=self._API_HEADERS), ('user', {dict})) or {}
	284
	285	if user_info.get('private') and user_info.get('followed_by_me') not in (True, 'true'):
	286	raise ExtractorError('This user profile is private', expected=True)
	287	elif traverse_obj(user_info, (('blocked_by_user', 'blocking_user'), {bool}), get_all=False):
	288	raise ExtractorError('The author of the video is blocked', expected=True)
92aa6d68	289
92aa6d68	290	user_id = str_or_none(user_info.get('user_id'))
33b737be	291	if not user_id:
33b737be	292	raise ExtractorError('Unable to extract user ID')
92aa6d68	293
92aa6d68	294	return self.playlist_result(
33b737be	295	self._entries(username, user_id), user_id, username, thumbnail=user_info.get('avatar_url'))
	296
	297
	298	class TrillerShortIE(InfoExtractor):
	299	_VALID_URL = r'https?://v\.triller\.co/(?P<id>\w+)'
	300	_TESTS = [{
	301	'url': 'https://v.triller.co/WWZNWk',
	302	'md5': '5eb8dc2c971bd8cd794ec9e8d5e9d101',
	303	'info_dict': {
	304	'id': '66210052',
	305	'ext': 'mp4',
	306	'title': 'md5:2dfc89d154cd91a4a18cd9582ba03e16',
	307	'display_id': 'f4480e1f-fb4e-45b9-a44c-9e6c679ce7eb',
	308	'thumbnail': r're:^https://uploads\.cdn\.triller\.co/.+\.jpg$',
	309	'description': 'md5:2dfc89d154cd91a4a18cd9582ba03e16',
	310	'uploader': 'statefairent',
	311	'uploader_id': '487545193',
	312	'creator': 'Official Summer Fair of LA',
	313	'timestamp': 1629655457,
	314	'upload_date': '20210822',
	315	'duration': 19,
	316	'view_count': int,
	317	'like_count': int,
	318	'artist': 'Unknown',
	319	'track': 'Unknown',
	320	'uploader_url': 'https://triller.co/@statefairent',
	321	'comment_count': int,
	322	},
	323	}]
	324
	325	def _real_extract(self, url):
3d2623a8	326	real_url = self._request_webpage(HEADRequest(url), self._match_id(url)).url
33b737be	327	if self.suitable(real_url): # Prevent infinite loop in case redirect fails
	328	raise UnsupportedError(real_url)
	329	return self.url_result(real_url)