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

import functools
import urllib.parse
import hashlib
import json

from .common import InfoExtractor
from ..utils import (
    ExtractorError,
    OnDemandPagedList,
    int_or_none,
    mimetype2ext,
    qualities,
    traverse_obj,
    unified_timestamp,
)


# https://github.com/yt-dlp/yt-dlp/issues/6671
class IwaraBaseIE(InfoExtractor):
    _USERTOKEN = None
    _MEDIATOKEN = None
    _NETRC_MACHINE = 'iwara'

    def _get_user_token(self, invalidate=False):
        if not invalidate and self._USERTOKEN:
            return self._USERTOKEN

        username, password = self._get_login_info()
        IwaraBaseIE._USERTOKEN = username and self.cache.load(self._NETRC_MACHINE, username)
        if not IwaraBaseIE._USERTOKEN or invalidate:
            IwaraBaseIE._USERTOKEN = self._download_json(
                'https://api.iwara.tv/user/login', None, note='Logging in',
                data=json.dumps({
                    'email': username,
                    'password': password
                }).encode('utf-8'),
                headers={
                    'Content-Type': 'application/json'
                })['token']

            self.cache.store(self._NETRC_MACHINE, username, IwaraBaseIE._USERTOKEN)

        return self._USERTOKEN

    def _get_media_token(self, invalidate=False):
        if not invalidate and self._MEDIATOKEN:
            return self._MEDIATOKEN

        IwaraBaseIE._MEDIATOKEN = self._download_json(
            'https://api.iwara.tv/user/token', None, note='Fetching media token',
            data=b'',  # Need to have some data here, even if it's empty
            headers={
                'Authorization': f'Bearer {self._get_user_token()}',
                'Content-Type': 'application/json'
            })['accessToken']

        return self._MEDIATOKEN


class IwaraIE(IwaraBaseIE):
    IE_NAME = 'iwara'
    _VALID_URL = r'https?://(?:www\.|ecchi\.)?iwara\.tv/videos?/(?P<id>[a-zA-Z0-9]+)'
    _TESTS = [{
        # this video cannot be played because of migration
        'only_matching': True,
        'url': 'https://www.iwara.tv/video/k2ayoueezfkx6gvq',
        'info_dict': {
            'id': 'k2ayoueezfkx6gvq',
            'ext': 'mp4',
            'age_limit': 18,
            'title': 'Defeat of Irybelda - アイリベルダの敗北',
            'description': 'md5:70278abebe706647a8b4cb04cf23e0d3',
            'uploader': 'Inwerwm',
            'uploader_id': 'inwerwm',
            'tags': 'count:1',
            'like_count': 6133,
            'view_count': 1050343,
            'comment_count': 1,
            'timestamp': 1677843869,
            'modified_timestamp': 1679056362,
        },
    }, {
        'url': 'https://iwara.tv/video/1ywe1sbkqwumpdxz5/',
        'md5': '20691ce1473ec2766c0788e14c60ce66',
        'info_dict': {
            'id': '1ywe1sbkqwumpdxz5',
            'ext': 'mp4',
            'age_limit': 18,
            'title': 'Aponia 阿波尼亚SEX  Party Tonight 手动脱衣 大奶 裸腿',
            'description': 'md5:0c4c310f2e0592d68b9f771d348329ca',
            'uploader': '龙也zZZ',
            'uploader_id': 'user792540',
            'tags': [
                'uncategorized'
            ],
            'like_count': 1809,
            'view_count': 25156,
            'comment_count': 1,
            'timestamp': 1678732213,
            'modified_timestamp': 1679110271,
        },
    }, {
        'url': 'https://iwara.tv/video/blggmfno8ghl725bg',
        'info_dict': {
            'id': 'blggmfno8ghl725bg',
            'ext': 'mp4',
            'age_limit': 18,
            'title': 'お外でおしっこしちゃう猫耳ロリメイド',
            'description': 'md5:0342ba9bf6db09edbbb28729657c3611',
            'uploader': 'Fe_Kurosabi',
            'uploader_id': 'fekurosabi',
            'tags': [
                'pee'
            ],
            'like_count': 192,
            'view_count': 12119,
            'comment_count': 0,
            'timestamp': 1598880567,
            'modified_timestamp': 1598908995,
            'availability': 'needs_auth',
        },
    }]

    def _extract_formats(self, video_id, fileurl):
        up = urllib.parse.urlparse(fileurl)
        q = urllib.parse.parse_qs(up.query)
        paths = up.path.rstrip('/').split('/')
        # https://github.com/yt-dlp/yt-dlp/issues/6549#issuecomment-1473771047
        x_version = hashlib.sha1('_'.join((paths[-1], q['expires'][0], '5nFp9kmbNnHdAFhaqMvt')).encode()).hexdigest()

        preference = qualities(['preview', '360', '540', 'Source'])

        files = self._download_json(fileurl, video_id, headers={'X-Version': x_version})
        for fmt in files:
            yield traverse_obj(fmt, {
                'format_id': 'name',
                'url': ('src', ('view', 'download'), {self._proto_relative_url}),
                'ext': ('type', {mimetype2ext}),
                'quality': ('name', {preference}),
                'height': ('name', {int_or_none}),
            }, get_all=False)

    def _real_extract(self, url):
        video_id = self._match_id(url)
        username, password = self._get_login_info()
        headers = {
            'Authorization': f'Bearer {self._get_media_token()}',
        } if username and password else None
        video_data = self._download_json(f'https://api.iwara.tv/video/{video_id}', video_id, expected_status=lambda x: True, headers=headers)
        errmsg = video_data.get('message')
        # at this point we can actually get uploaded user info, but do we need it?
        if errmsg == 'errors.privateVideo':
            self.raise_login_required('Private video. Login if you have permissions to watch')
        elif errmsg == 'errors.notFound' and not username:
            self.raise_login_required('Video may need login to view')
        elif errmsg:  # None if success
            raise ExtractorError(f'Iwara says: {errmsg}')

        if not video_data.get('fileUrl'):
            if video_data.get('embedUrl'):
                return self.url_result(video_data.get('embedUrl'))
            raise ExtractorError('This video is unplayable', expected=True)

        return {
            'id': video_id,
            'age_limit': 18 if video_data.get('rating') == 'ecchi' else 0,  # ecchi is 'sexy' in Japanese
            **traverse_obj(video_data, {
                'title': 'title',
                'description': 'body',
                'uploader': ('user', 'name'),
                'uploader_id': ('user', 'username'),
                'tags': ('tags', ..., 'id'),
                'like_count': 'numLikes',
                'view_count': 'numViews',
                'comment_count': 'numComments',
                'timestamp': ('createdAt', {unified_timestamp}),
                'modified_timestamp': ('updatedAt', {unified_timestamp}),
                'thumbnail': ('file', 'id', {str}, {
                    lambda x: f'https://files.iwara.tv/image/thumbnail/{x}/thumbnail-00.jpg'}),
            }),
            'formats': list(self._extract_formats(video_id, video_data.get('fileUrl'))),
        }

    def _perform_login(self, username, password):
        if self.cache.load(self._NETRC_MACHINE, username) and self._get_media_token():
            self.write_debug('Skipping logging in')
            return

        IwaraBaseIE._USERTOKEN = self._get_user_token(True)
        self._get_media_token(True)
        self.cache.store(self._NETRC_MACHINE, username, IwaraBaseIE._USERTOKEN)


class IwaraUserIE(IwaraBaseIE):
    _VALID_URL = r'https?://(?:www\.)?iwara\.tv/profile/(?P<id>[^/?#&]+)'
    IE_NAME = 'iwara:user'
    _PER_PAGE = 32

    _TESTS = [{
        'url': 'https://iwara.tv/profile/user792540/videos',
        'info_dict': {
            'id': 'user792540',
        },
        'playlist_mincount': 80,
    }, {
        'url': 'https://iwara.tv/profile/theblackbirdcalls/videos',
        'info_dict': {
            'id': 'theblackbirdcalls',
        },
        'playlist_mincount': 723,
    }, {
        'url': 'https://iwara.tv/profile/user792540',
        'only_matching': True,
    }, {
        'url': 'https://iwara.tv/profile/theblackbirdcalls',
        'only_matching': True,
    }]

    def _entries(self, playlist_id, user_id, page):
        videos = self._download_json(
            'https://api.iwara.tv/videos', playlist_id,
            note=f'Downloading page {page}',
            query={
                'page': page,
                'sort': 'date',
                'user': user_id,
                'limit': self._PER_PAGE,
            })
        for x in traverse_obj(videos, ('results', ..., 'id')):
            yield self.url_result(f'https://iwara.tv/video/{x}')

    def _real_extract(self, url):
        playlist_id = self._match_id(url)
        user_info = self._download_json(
            f'https://api.iwara.tv/profile/{playlist_id}', playlist_id,
            note='Requesting user info')
        user_id = traverse_obj(user_info, ('user', 'id'))

        return self.playlist_result(
            OnDemandPagedList(
                functools.partial(self._entries, playlist_id, user_id),
                self._PER_PAGE),
            playlist_id, traverse_obj(user_info, ('user', 'name')))


class IwaraPlaylistIE(IwaraBaseIE):
    # the ID is an UUID but I don't think it's necessary to write concrete regex
    _VALID_URL = r'https?://(?:www\.)?iwara\.tv/playlist/(?P<id>[0-9a-f-]+)'
    IE_NAME = 'iwara:playlist'
    _PER_PAGE = 32

    _TESTS = [{
        'url': 'https://iwara.tv/playlist/458e5486-36a4-4ac0-b233-7e9eef01025f',
        'info_dict': {
            'id': '458e5486-36a4-4ac0-b233-7e9eef01025f',
        },
        'playlist_mincount': 3,
    }]

    def _entries(self, playlist_id, first_page, page):
        videos = self._download_json(
            'https://api.iwara.tv/videos', playlist_id, f'Downloading page {page}',
            query={'page': page, 'limit': self._PER_PAGE}) if page else first_page
        for x in traverse_obj(videos, ('results', ..., 'id')):
            yield self.url_result(f'https://iwara.tv/video/{x}')

    def _real_extract(self, url):
        playlist_id = self._match_id(url)
        page_0 = self._download_json(
            f'https://api.iwara.tv/playlist/{playlist_id}?page=0&limit={self._PER_PAGE}', playlist_id,
            note='Requesting playlist info')

        return self.playlist_result(
            OnDemandPagedList(
                functools.partial(self._entries, playlist_id, page_0),
                self._PER_PAGE),
            playlist_id, traverse_obj(page_0, ('title', 'name')))
Commit	Line	Data
c14af7a7	1	import functools
ac668111	2	import urllib.parse
c14af7a7	3	import hashlib
21b9413c	4	import json
001a5fd3 YCH	5
001a5fd3 YCH	6	from .common import InfoExtractor
caf0f5f8	7	from ..utils import (
95a383be	8	ExtractorError,
c14af7a7	9	OnDemandPagedList,
caf0f5f8 YCH	10	int_or_none,
caf0f5f8 YCH	11	mimetype2ext,
56793f74	12	qualities,
c14af7a7 L	13	traverse_obj,
c14af7a7 L	14	unified_timestamp,
caf0f5f8	15	)
001a5fd3 YCH	16
001a5fd3 YCH	17
21b9413c	18	# https://github.com/yt-dlp/yt-dlp/issues/6671
	19	class IwaraBaseIE(InfoExtractor):
	20	_USERTOKEN = None
	21	_MEDIATOKEN = None
	22	_NETRC_MACHINE = 'iwara'
	23
	24	def _get_user_token(self, invalidate=False):
	25	if not invalidate and self._USERTOKEN:
	26	return self._USERTOKEN
	27
	28	username, password = self._get_login_info()
	29	IwaraBaseIE._USERTOKEN = username and self.cache.load(self._NETRC_MACHINE, username)
	30	if not IwaraBaseIE._USERTOKEN or invalidate:
	31	IwaraBaseIE._USERTOKEN = self._download_json(
	32	'https://api.iwara.tv/user/login', None, note='Logging in',
	33	data=json.dumps({
	34	'email': username,
	35	'password': password
	36	}).encode('utf-8'),
	37	headers={
	38	'Content-Type': 'application/json'
	39	})['token']
	40
	41	self.cache.store(self._NETRC_MACHINE, username, IwaraBaseIE._USERTOKEN)
	42
	43	return self._USERTOKEN
	44
	45	def _get_media_token(self, invalidate=False):
	46	if not invalidate and self._MEDIATOKEN:
	47	return self._MEDIATOKEN
	48
	49	IwaraBaseIE._MEDIATOKEN = self._download_json(
	50	'https://api.iwara.tv/user/token', None, note='Fetching media token',
	51	data=b'', # Need to have some data here, even if it's empty
	52	headers={
	53	'Authorization': f'Bearer {self._get_user_token()}',
	54	'Content-Type': 'application/json'
	55	})['accessToken']
	56
	57	return self._MEDIATOKEN
	58
	59
	60	class IwaraIE(IwaraBaseIE):
c14af7a7	61	IE_NAME = 'iwara'
ab92d865	62	_VALID_URL = r'https?://(?:www\.\|ecchi\.)?iwara\.tv/videos?/(?P<id>[a-zA-Z0-9]+)'
001a5fd3	63	_TESTS = [{
c14af7a7 L	64	# this video cannot be played because of migration
	65	'only_matching': True,
	66	'url': 'https://www.iwara.tv/video/k2ayoueezfkx6gvq',
001a5fd3	67	'info_dict': {
c14af7a7	68	'id': 'k2ayoueezfkx6gvq',
001a5fd3	69	'ext': 'mp4',
001a5fd3	70	'age_limit': 18,
c14af7a7 L	71	'title': 'Defeat of Irybelda - アイリベルダの敗北',
	72	'description': 'md5:70278abebe706647a8b4cb04cf23e0d3',
	73	'uploader': 'Inwerwm',
	74	'uploader_id': 'inwerwm',
	75	'tags': 'count:1',
	76	'like_count': 6133,
	77	'view_count': 1050343,
	78	'comment_count': 1,
	79	'timestamp': 1677843869,
	80	'modified_timestamp': 1679056362,
001a5fd3 YCH	81	},
001a5fd3 YCH	82	}, {
c14af7a7 L	83	'url': 'https://iwara.tv/video/1ywe1sbkqwumpdxz5/',
c14af7a7 L	84	'md5': '20691ce1473ec2766c0788e14c60ce66',
001a5fd3	85	'info_dict': {
c14af7a7	86	'id': '1ywe1sbkqwumpdxz5',
001a5fd3	87	'ext': 'mp4',
001a5fd3	88	'age_limit': 18,
c14af7a7 L	89	'title': 'Aponia 阿波尼亚SEX Party Tonight 手动脱衣大奶裸腿',
	90	'description': 'md5:0c4c310f2e0592d68b9f771d348329ca',
	91	'uploader': '龙也zZZ',
	92	'uploader_id': 'user792540',
	93	'tags': [
	94	'uncategorized'
	95	],
	96	'like_count': 1809,
	97	'view_count': 25156,
	98	'comment_count': 1,
	99	'timestamp': 1678732213,
	100	'modified_timestamp': 1679110271,
001a5fd3	101	},
21b9413c	102	}, {
	103	'url': 'https://iwara.tv/video/blggmfno8ghl725bg',
	104	'info_dict': {
	105	'id': 'blggmfno8ghl725bg',
	106	'ext': 'mp4',
	107	'age_limit': 18,
	108	'title': 'お外でおしっこしちゃう猫耳ロリメイド',
	109	'description': 'md5:0342ba9bf6db09edbbb28729657c3611',
	110	'uploader': 'Fe_Kurosabi',
	111	'uploader_id': 'fekurosabi',
	112	'tags': [
	113	'pee'
	114	],
	115	'like_count': 192,
	116	'view_count': 12119,
	117	'comment_count': 0,
	118	'timestamp': 1598880567,
	119	'modified_timestamp': 1598908995,
	120	'availability': 'needs_auth',
	121	},
001a5fd3 YCH	122	}]
001a5fd3 YCH	123
c14af7a7 L	124	def _extract_formats(self, video_id, fileurl):
	125	up = urllib.parse.urlparse(fileurl)
	126	q = urllib.parse.parse_qs(up.query)
	127	paths = up.path.rstrip('/').split('/')
	128	# https://github.com/yt-dlp/yt-dlp/issues/6549#issuecomment-1473771047
	129	x_version = hashlib.sha1('_'.join((paths[-1], q['expires'][0], '5nFp9kmbNnHdAFhaqMvt')).encode()).hexdigest()
	130
56793f74	131	preference = qualities(['preview', '360', '540', 'Source'])
56793f74	132
c14af7a7 L	133	files = self._download_json(fileurl, video_id, headers={'X-Version': x_version})
	134	for fmt in files:
	135	yield traverse_obj(fmt, {
	136	'format_id': 'name',
	137	'url': ('src', ('view', 'download'), {self._proto_relative_url}),
	138	'ext': ('type', {mimetype2ext}),
56793f74	139	'quality': ('name', {preference}),
c14af7a7 L	140	'height': ('name', {int_or_none}),
	141	}, get_all=False)
	142
001a5fd3 YCH	143	def _real_extract(self, url):
001a5fd3 YCH	144	video_id = self._match_id(url)
21b9413c	145	username, password = self._get_login_info()
	146	headers = {
	147	'Authorization': f'Bearer {self._get_media_token()}',
	148	} if username and password else None
	149	video_data = self._download_json(f'https://api.iwara.tv/video/{video_id}', video_id, expected_status=lambda x: True, headers=headers)
95a383be L	150	errmsg = video_data.get('message')
	151	# at this point we can actually get uploaded user info, but do we need it?
	152	if errmsg == 'errors.privateVideo':
	153	self.raise_login_required('Private video. Login if you have permissions to watch')
21b9413c	154	elif errmsg == 'errors.notFound' and not username:
	155	self.raise_login_required('Video may need login to view')
	156	elif errmsg: # None if success
95a383be	157	raise ExtractorError(f'Iwara says: {errmsg}')
caf0f5f8	158
56793f74	159	if not video_data.get('fileUrl'):
	160	if video_data.get('embedUrl'):
	161	return self.url_result(video_data.get('embedUrl'))
	162	raise ExtractorError('This video is unplayable', expected=True)
	163
caf0f5f8	164	return {
001a5fd3	165	'id': video_id,
c14af7a7 L	166	'age_limit': 18 if video_data.get('rating') == 'ecchi' else 0, # ecchi is 'sexy' in Japanese
	167	**traverse_obj(video_data, {
	168	'title': 'title',
	169	'description': 'body',
	170	'uploader': ('user', 'name'),
	171	'uploader_id': ('user', 'username'),
	172	'tags': ('tags', ..., 'id'),
	173	'like_count': 'numLikes',
	174	'view_count': 'numViews',
	175	'comment_count': 'numComments',
	176	'timestamp': ('createdAt', {unified_timestamp}),
	177	'modified_timestamp': ('updatedAt', {unified_timestamp}),
	178	'thumbnail': ('file', 'id', {str}, {
	179	lambda x: f'https://files.iwara.tv/image/thumbnail/{x}/thumbnail-00.jpg'}),
	180	}),
	181	'formats': list(self._extract_formats(video_id, video_data.get('fileUrl'))),
caf0f5f8	182	}
ff4d7860	183
21b9413c	184	def _perform_login(self, username, password):
	185	if self.cache.load(self._NETRC_MACHINE, username) and self._get_media_token():
	186	self.write_debug('Skipping logging in')
	187	return
	188
	189	IwaraBaseIE._USERTOKEN = self._get_user_token(True)
	190	self._get_media_token(True)
	191	self.cache.store(self._NETRC_MACHINE, username, IwaraBaseIE._USERTOKEN)
	192
ff4d7860	193
21b9413c	194	class IwaraUserIE(IwaraBaseIE):
c14af7a7 L	195	_VALID_URL = r'https?://(?:www\.)?iwara\.tv/profile/(?P<id>[^/?#&]+)'
	196	IE_NAME = 'iwara:user'
	197	_PER_PAGE = 32
ff4d7860	198
ff4d7860	199	_TESTS = [{
c14af7a7	200	'url': 'https://iwara.tv/profile/user792540/videos',
ff4d7860	201	'info_dict': {
c14af7a7	202	'id': 'user792540',
ff4d7860	203	},
c14af7a7	204	'playlist_mincount': 80,
ff4d7860	205	}, {
c14af7a7	206	'url': 'https://iwara.tv/profile/theblackbirdcalls/videos',
ff4d7860	207	'info_dict': {
c14af7a7	208	'id': 'theblackbirdcalls',
ff4d7860	209	},
c14af7a7 L	210	'playlist_mincount': 723,
	211	}, {
	212	'url': 'https://iwara.tv/profile/user792540',
	213	'only_matching': True,
	214	}, {
	215	'url': 'https://iwara.tv/profile/theblackbirdcalls',
	216	'only_matching': True,
ff4d7860	217	}]
ff4d7860	218
c14af7a7 L	219	def _entries(self, playlist_id, user_id, page):
	220	videos = self._download_json(
	221	'https://api.iwara.tv/videos', playlist_id,
	222	note=f'Downloading page {page}',
	223	query={
	224	'page': page,
	225	'sort': 'date',
	226	'user': user_id,
	227	'limit': self._PER_PAGE,
	228	})
	229	for x in traverse_obj(videos, ('results', ..., 'id')):
	230	yield self.url_result(f'https://iwara.tv/video/{x}')
	231
ff4d7860	232	def _real_extract(self, url):
c14af7a7 L	233	playlist_id = self._match_id(url)
	234	user_info = self._download_json(
	235	f'https://api.iwara.tv/profile/{playlist_id}', playlist_id,
	236	note='Requesting user info')
	237	user_id = traverse_obj(user_info, ('user', 'id'))
ff4d7860	238
c14af7a7 L	239	return self.playlist_result(
	240	OnDemandPagedList(
	241	functools.partial(self._entries, playlist_id, user_id),
	242	self._PER_PAGE),
	243	playlist_id, traverse_obj(user_info, ('user', 'name')))
ff4d7860	244
ff4d7860	245
21b9413c	246	class IwaraPlaylistIE(IwaraBaseIE):
c14af7a7 L	247	# the ID is an UUID but I don't think it's necessary to write concrete regex
	248	_VALID_URL = r'https?://(?:www\.)?iwara\.tv/playlist/(?P<id>[0-9a-f-]+)'
	249	IE_NAME = 'iwara:playlist'
	250	_PER_PAGE = 32
ff4d7860	251
ff4d7860	252	_TESTS = [{
c14af7a7	253	'url': 'https://iwara.tv/playlist/458e5486-36a4-4ac0-b233-7e9eef01025f',
661e7253	254	'info_dict': {
c14af7a7	255	'id': '458e5486-36a4-4ac0-b233-7e9eef01025f',
661e7253	256	},
c14af7a7	257	'playlist_mincount': 3,
ff4d7860	258	}]
ff4d7860	259
c14af7a7 L	260	def _entries(self, playlist_id, first_page, page):
	261	videos = self._download_json(
	262	'https://api.iwara.tv/videos', playlist_id, f'Downloading page {page}',
	263	query={'page': page, 'limit': self._PER_PAGE}) if page else first_page
	264	for x in traverse_obj(videos, ('results', ..., 'id')):
	265	yield self.url_result(f'https://iwara.tv/video/{x}')
ff4d7860	266
ff4d7860	267	def _real_extract(self, url):
c14af7a7 L	268	playlist_id = self._match_id(url)
	269	page_0 = self._download_json(
	270	f'https://api.iwara.tv/playlist/{playlist_id}?page=0&limit={self._PER_PAGE}', playlist_id,
	271	note='Requesting playlist info')
ff4d7860	272
ff4d7860	273	return self.playlist_result(
c14af7a7 L	274	OnDemandPagedList(
	275	functools.partial(self._entries, playlist_id, page_0),
	276	self._PER_PAGE),
	277	playlist_id, traverse_obj(page_0, ('title', 'name')))