yt_dlp/extractor/iwara.py

   1 import functools
   2 import hashlib
   3 import json
   4 import time
   5 import urllib.error
   6 import urllib.parse
   7
   8 from .common import InfoExtractor
   9 from ..utils import (
  10     ExtractorError,
  11     OnDemandPagedList,
  12     int_or_none,
  13     jwt_decode_hs256,
  14     mimetype2ext,
  15     qualities,
  16     traverse_obj,
  17     try_call,
  18     unified_timestamp,
  19 )
  20
  21
  22 class IwaraBaseIE(InfoExtractor):
  23     _NETRC_MACHINE = 'iwara'
  24     _USERTOKEN = None
  25     _MEDIATOKEN = None
  26
  27     def _is_token_expired(self, token, token_type):
  28         # User token TTL == ~3 weeks, Media token TTL == ~1 hour
  29         if (try_call(lambda: jwt_decode_hs256(token)['exp']) or 0) <= int(time.time() - 120):
  30             self.to_screen(f'{token_type} token has expired')
  31             return True
  32
  33     def _get_user_token(self):
  34         username, password = self._get_login_info()
  35         if not username or not password:
  36             return
  37
  38         user_token = IwaraBaseIE._USERTOKEN or self.cache.load(self._NETRC_MACHINE, username)
  39         if not user_token or self._is_token_expired(user_token, 'User'):
  40             response = self._download_json(
  41                 'https://api.iwara.tv/user/login', None, note='Logging in',
  42                 headers={'Content-Type': 'application/json'}, data=json.dumps({
  43                     'email': username,
  44                     'password': password
  45                 }).encode(), expected_status=lambda x: True)
  46             user_token = traverse_obj(response, ('token', {str}))
  47             if not user_token:
  48                 error = traverse_obj(response, ('message', {str}))
  49                 if 'invalidLogin' in error:
  50                     raise ExtractorError('Invalid login credentials', expected=True)
  51                 else:
  52                     raise ExtractorError(f'Iwara API said: {error or "nothing"}')
  53
  54             self.cache.store(self._NETRC_MACHINE, username, user_token)
  55
  56         IwaraBaseIE._USERTOKEN = user_token
  57
  58     def _get_media_token(self):
  59         self._get_user_token()
  60         if not IwaraBaseIE._USERTOKEN:
  61             return  # user has not passed credentials
  62
  63         if not IwaraBaseIE._MEDIATOKEN or self._is_token_expired(IwaraBaseIE._MEDIATOKEN, 'Media'):
  64             IwaraBaseIE._MEDIATOKEN = self._download_json(
  65                 'https://api.iwara.tv/user/token', None, note='Fetching media token',
  66                 data=b'', headers={
  67                     'Authorization': f'Bearer {IwaraBaseIE._USERTOKEN}',
  68                     'Content-Type': 'application/json'
  69                 })['accessToken']
  70
  71         return {'Authorization': f'Bearer {IwaraBaseIE._MEDIATOKEN}'}
  72
  73     def _perform_login(self, username, password):
  74         self._get_media_token()
  75
  76
  77 class IwaraIE(IwaraBaseIE):
  78     IE_NAME = 'iwara'
  79     _VALID_URL = r'https?://(?:www\.|ecchi\.)?iwara\.tv/videos?/(?P<id>[a-zA-Z0-9]+)'
  80     _TESTS = [{
  81         'url': 'https://www.iwara.tv/video/k2ayoueezfkx6gvq',
  82         'info_dict': {
  83             'id': 'k2ayoueezfkx6gvq',
  84             'ext': 'mp4',
  85             'age_limit': 18,
  86             'title': 'Defeat of Irybelda - アイリベルダの敗北',
  87             'description': 'md5:70278abebe706647a8b4cb04cf23e0d3',
  88             'uploader': 'Inwerwm',
  89             'uploader_id': 'inwerwm',
  90             'tags': 'count:1',
  91             'like_count': 6133,
  92             'view_count': 1050343,
  93             'comment_count': 1,
  94             'timestamp': 1677843869,
  95             'modified_timestamp': 1679056362,
  96         },
  97         'skip': 'this video cannot be played because of migration',
  98     }, {
  99         'url': 'https://iwara.tv/video/1ywe1sbkqwumpdxz5/',
 100         'md5': '7645f966f069b8ec9210efd9130c9aad',
 101         'info_dict': {
 102             'id': '1ywe1sbkqwumpdxz5',
 103             'ext': 'mp4',
 104             'age_limit': 18,
 105             'title': 'Aponia アポニア SEX  Party Tonight 手の脱衣 巨乳 ',
 106             'description': 'md5:3f60016fff22060eef1ef26d430b1f67',
 107             'uploader': 'Lyu ya',
 108             'uploader_id': 'user792540',
 109             'tags': [
 110                 'uncategorized'
 111             ],
 112             'like_count': int,
 113             'view_count': int,
 114             'comment_count': int,
 115             'timestamp': 1678732213,
 116             'modified_timestamp': int,
 117             'thumbnail': 'https://files.iwara.tv/image/thumbnail/581d12b5-46f4-4f15-beb2-cfe2cde5d13d/thumbnail-00.jpg',
 118             'modified_date': '20230614',
 119             'upload_date': '20230313',
 120         },
 121     }, {
 122         'url': 'https://iwara.tv/video/blggmfno8ghl725bg',
 123         'info_dict': {
 124             'id': 'blggmfno8ghl725bg',
 125             'ext': 'mp4',
 126             'age_limit': 18,
 127             'title': 'お外でおしっこしちゃう猫耳ロリメイド',
 128             'description': 'md5:0342ba9bf6db09edbbb28729657c3611',
 129             'uploader': 'Fe_Kurosabi',
 130             'uploader_id': 'fekurosabi',
 131             'tags': [
 132                 'pee'
 133             ],
 134             'like_count': int,
 135             'view_count': int,
 136             'comment_count': int,
 137             'timestamp': 1598880567,
 138             'modified_timestamp': int,
 139             'upload_date': '20200831',
 140             'modified_date': '20230605',
 141             'thumbnail': 'https://files.iwara.tv/image/thumbnail/7693e881-d302-42a4-a780-f16d66b5dadd/thumbnail-00.jpg',
 142             # 'availability': 'needs_auth',
 143         },
 144     }]
 145
 146     def _extract_formats(self, video_id, fileurl):
 147         up = urllib.parse.urlparse(fileurl)
 148         q = urllib.parse.parse_qs(up.query)
 149         paths = up.path.rstrip('/').split('/')
 150         # https://github.com/yt-dlp/yt-dlp/issues/6549#issuecomment-1473771047
 151         x_version = hashlib.sha1('_'.join((paths[-1], q['expires'][0], '5nFp9kmbNnHdAFhaqMvt')).encode()).hexdigest()
 152
 153         preference = qualities(['preview', '360', '540', 'Source'])
 154
 155         files = self._download_json(fileurl, video_id, headers={'X-Version': x_version})
 156         for fmt in files:
 157             yield traverse_obj(fmt, {
 158                 'format_id': 'name',
 159                 'url': ('src', ('view', 'download'), {self._proto_relative_url}),
 160                 'ext': ('type', {mimetype2ext}),
 161                 'quality': ('name', {preference}),
 162                 'height': ('name', {int_or_none}),
 163             }, get_all=False)
 164
 165     def _real_extract(self, url):
 166         video_id = self._match_id(url)
 167         username, _ = self._get_login_info()
 168         video_data = self._download_json(
 169             f'https://api.iwara.tv/video/{video_id}', video_id,
 170             expected_status=lambda x: True, headers=self._get_media_token())
 171         errmsg = video_data.get('message')
 172         # at this point we can actually get uploaded user info, but do we need it?
 173         if errmsg == 'errors.privateVideo':
 174             self.raise_login_required('Private video. Login if you have permissions to watch', method='password')
 175         elif errmsg == 'errors.notFound' and not username:
 176             self.raise_login_required('Video may need login to view', method='password')
 177         elif errmsg:  # None if success
 178             raise ExtractorError(f'Iwara says: {errmsg}')
 179
 180         if not video_data.get('fileUrl'):
 181             if video_data.get('embedUrl'):
 182                 return self.url_result(video_data.get('embedUrl'))
 183             raise ExtractorError('This video is unplayable', expected=True)
 184
 185         return {
 186             'id': video_id,
 187             'age_limit': 18 if video_data.get('rating') == 'ecchi' else 0,  # ecchi is 'sexy' in Japanese
 188             **traverse_obj(video_data, {
 189                 'title': 'title',
 190                 'description': 'body',
 191                 'uploader': ('user', 'name'),
 192                 'uploader_id': ('user', 'username'),
 193                 'tags': ('tags', ..., 'id'),
 194                 'like_count': 'numLikes',
 195                 'view_count': 'numViews',
 196                 'comment_count': 'numComments',
 197                 'timestamp': ('createdAt', {unified_timestamp}),
 198                 'modified_timestamp': ('updatedAt', {unified_timestamp}),
 199                 'thumbnail': ('file', 'id', {str}, {
 200                     lambda x: f'https://files.iwara.tv/image/thumbnail/{x}/thumbnail-00.jpg'}),
 201             }),
 202             'formats': list(self._extract_formats(video_id, video_data.get('fileUrl'))),
 203         }
 204
 205
 206 class IwaraUserIE(IwaraBaseIE):
 207     _VALID_URL = r'https?://(?:www\.)?iwara\.tv/profile/(?P<id>[^/?#&]+)'
 208     IE_NAME = 'iwara:user'
 209     _PER_PAGE = 32
 210
 211     _TESTS = [{
 212         'url': 'https://iwara.tv/profile/user792540/videos',
 213         'info_dict': {
 214             'id': 'user792540',
 215             'title': 'Lyu ya',
 216         },
 217         'playlist_mincount': 70,
 218     }, {
 219         'url': 'https://iwara.tv/profile/theblackbirdcalls/videos',
 220         'info_dict': {
 221             'id': 'theblackbirdcalls',
 222             'title': 'TheBlackbirdCalls',
 223         },
 224         'playlist_mincount': 723,
 225     }, {
 226         'url': 'https://iwara.tv/profile/user792540',
 227         'only_matching': True,
 228     }, {
 229         'url': 'https://iwara.tv/profile/theblackbirdcalls',
 230         'only_matching': True,
 231     }, {
 232         'url': 'https://www.iwara.tv/profile/lumymmd',
 233         'info_dict': {
 234             'id': 'lumymmd',
 235             'title': 'Lumy MMD',
 236         },
 237         'playlist_mincount': 1,
 238     }]
 239
 240     def _entries(self, playlist_id, user_id, page):
 241         videos = self._download_json(
 242             'https://api.iwara.tv/videos', playlist_id,
 243             note=f'Downloading page {page}',
 244             query={
 245                 'page': page,
 246                 'sort': 'date',
 247                 'user': user_id,
 248                 'limit': self._PER_PAGE,
 249             }, headers=self._get_media_token())
 250         for x in traverse_obj(videos, ('results', ..., 'id')):
 251             yield self.url_result(f'https://iwara.tv/video/{x}')
 252
 253     def _real_extract(self, url):
 254         playlist_id = self._match_id(url)
 255         user_info = self._download_json(
 256             f'https://api.iwara.tv/profile/{playlist_id}', playlist_id,
 257             note='Requesting user info')
 258         user_id = traverse_obj(user_info, ('user', 'id'))
 259
 260         return self.playlist_result(
 261             OnDemandPagedList(
 262                 functools.partial(self._entries, playlist_id, user_id),
 263                 self._PER_PAGE),
 264             playlist_id, traverse_obj(user_info, ('user', 'name')))
 265
 266
 267 class IwaraPlaylistIE(IwaraBaseIE):
 268     _VALID_URL = r'https?://(?:www\.)?iwara\.tv/playlist/(?P<id>[0-9a-f-]+)'
 269     IE_NAME = 'iwara:playlist'
 270     _PER_PAGE = 32
 271
 272     _TESTS = [{
 273         'url': 'https://iwara.tv/playlist/458e5486-36a4-4ac0-b233-7e9eef01025f',
 274         'info_dict': {
 275             'id': '458e5486-36a4-4ac0-b233-7e9eef01025f',
 276         },
 277         'playlist_mincount': 3,
 278     }]
 279
 280     def _entries(self, playlist_id, first_page, page):
 281         videos = self._download_json(
 282             'https://api.iwara.tv/videos', playlist_id, f'Downloading page {page}',
 283             query={'page': page, 'limit': self._PER_PAGE},
 284             headers=self._get_media_token()) if page else first_page
 285         for x in traverse_obj(videos, ('results', ..., 'id')):
 286             yield self.url_result(f'https://iwara.tv/video/{x}')
 287
 288     def _real_extract(self, url):
 289         playlist_id = self._match_id(url)
 290         page_0 = self._download_json(
 291             f'https://api.iwara.tv/playlist/{playlist_id}?page=0&limit={self._PER_PAGE}', playlist_id,
 292             note='Requesting playlist info', headers=self._get_media_token())
 293
 294         return self.playlist_result(
 295             OnDemandPagedList(
 296                 functools.partial(self._entries, playlist_id, page_0),
 297                 self._PER_PAGE),
 298             playlist_id, traverse_obj(page_0, ('title', 'name')))