yt_dlp/extractor/iwara.py

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