yt_dlp/extractor/iwara.py

   1 import functools
   2 import urllib.parse
   3 import hashlib
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     ExtractorError,
   8     OnDemandPagedList,
   9     int_or_none,
  10     mimetype2ext,
  11     qualities,
  12     traverse_obj,
  13     unified_timestamp,
  14 )
  15
  16
  17 class IwaraIE(InfoExtractor):
  18     IE_NAME = 'iwara'
  19     _VALID_URL = r'https?://(?:www\.|ecchi\.)?iwara\.tv/videos?/(?P<id>[a-zA-Z0-9]+)'
  20     _TESTS = [{
  21         # this video cannot be played because of migration
  22         'only_matching': True,
  23         'url': 'https://www.iwara.tv/video/k2ayoueezfkx6gvq',
  24         'info_dict': {
  25             'id': 'k2ayoueezfkx6gvq',
  26             'ext': 'mp4',
  27             'age_limit': 18,
  28             'title': 'Defeat of Irybelda - アイリベルダの敗北',
  29             'description': 'md5:70278abebe706647a8b4cb04cf23e0d3',
  30             'uploader': 'Inwerwm',
  31             'uploader_id': 'inwerwm',
  32             'tags': 'count:1',
  33             'like_count': 6133,
  34             'view_count': 1050343,
  35             'comment_count': 1,
  36             'timestamp': 1677843869,
  37             'modified_timestamp': 1679056362,
  38         },
  39     }, {
  40         'url': 'https://iwara.tv/video/1ywe1sbkqwumpdxz5/',
  41         'md5': '20691ce1473ec2766c0788e14c60ce66',
  42         'info_dict': {
  43             'id': '1ywe1sbkqwumpdxz5',
  44             'ext': 'mp4',
  45             'age_limit': 18,
  46             'title': 'Aponia 阿波尼亚SEX  Party Tonight 手动脱衣 大奶 裸腿',
  47             'description': 'md5:0c4c310f2e0592d68b9f771d348329ca',
  48             'uploader': '龙也zZZ',
  49             'uploader_id': 'user792540',
  50             'tags': [
  51                 'uncategorized'
  52             ],
  53             'like_count': 1809,
  54             'view_count': 25156,
  55             'comment_count': 1,
  56             'timestamp': 1678732213,
  57             'modified_timestamp': 1679110271,
  58         },
  59     }]
  60
  61     def _extract_formats(self, video_id, fileurl):
  62         up = urllib.parse.urlparse(fileurl)
  63         q = urllib.parse.parse_qs(up.query)
  64         paths = up.path.rstrip('/').split('/')
  65         # https://github.com/yt-dlp/yt-dlp/issues/6549#issuecomment-1473771047
  66         x_version = hashlib.sha1('_'.join((paths[-1], q['expires'][0], '5nFp9kmbNnHdAFhaqMvt')).encode()).hexdigest()
  67
  68         preference = qualities(['preview', '360', '540', 'Source'])
  69
  70         files = self._download_json(fileurl, video_id, headers={'X-Version': x_version})
  71         for fmt in files:
  72             yield traverse_obj(fmt, {
  73                 'format_id': 'name',
  74                 'url': ('src', ('view', 'download'), {self._proto_relative_url}),
  75                 'ext': ('type', {mimetype2ext}),
  76                 'quality': ('name', {preference}),
  77                 'height': ('name', {int_or_none}),
  78             }, get_all=False)
  79
  80     def _real_extract(self, url):
  81         video_id = self._match_id(url)
  82         video_data = self._download_json(f'https://api.iwara.tv/video/{video_id}', video_id, expected_status=lambda x: True)
  83         errmsg = video_data.get('message')
  84         # at this point we can actually get uploaded user info, but do we need it?
  85         if errmsg == 'errors.privateVideo':
  86             self.raise_login_required('Private video. Login if you have permissions to watch')
  87         elif errmsg:
  88             raise ExtractorError(f'Iwara says: {errmsg}')
  89
  90         if not video_data.get('fileUrl'):
  91             if video_data.get('embedUrl'):
  92                 return self.url_result(video_data.get('embedUrl'))
  93             raise ExtractorError('This video is unplayable', expected=True)
  94
  95         return {
  96             'id': video_id,
  97             'age_limit': 18 if video_data.get('rating') == 'ecchi' else 0,  # ecchi is 'sexy' in Japanese
  98             **traverse_obj(video_data, {
  99                 'title': 'title',
 100                 'description': 'body',
 101                 'uploader': ('user', 'name'),
 102                 'uploader_id': ('user', 'username'),
 103                 'tags': ('tags', ..., 'id'),
 104                 'like_count': 'numLikes',
 105                 'view_count': 'numViews',
 106                 'comment_count': 'numComments',
 107                 'timestamp': ('createdAt', {unified_timestamp}),
 108                 'modified_timestamp': ('updatedAt', {unified_timestamp}),
 109                 'thumbnail': ('file', 'id', {str}, {
 110                     lambda x: f'https://files.iwara.tv/image/thumbnail/{x}/thumbnail-00.jpg'}),
 111             }),
 112             'formats': list(self._extract_formats(video_id, video_data.get('fileUrl'))),
 113         }
 114
 115
 116 class IwaraUserIE(InfoExtractor):
 117     _VALID_URL = r'https?://(?:www\.)?iwara\.tv/profile/(?P<id>[^/?#&]+)'
 118     IE_NAME = 'iwara:user'
 119     _PER_PAGE = 32
 120
 121     _TESTS = [{
 122         'url': 'https://iwara.tv/profile/user792540/videos',
 123         'info_dict': {
 124             'id': 'user792540',
 125         },
 126         'playlist_mincount': 80,
 127     }, {
 128         'url': 'https://iwara.tv/profile/theblackbirdcalls/videos',
 129         'info_dict': {
 130             'id': 'theblackbirdcalls',
 131         },
 132         'playlist_mincount': 723,
 133     }, {
 134         'url': 'https://iwara.tv/profile/user792540',
 135         'only_matching': True,
 136     }, {
 137         'url': 'https://iwara.tv/profile/theblackbirdcalls',
 138         'only_matching': True,
 139     }]
 140
 141     def _entries(self, playlist_id, user_id, page):
 142         videos = self._download_json(
 143             'https://api.iwara.tv/videos', playlist_id,
 144             note=f'Downloading page {page}',
 145             query={
 146                 'page': page,
 147                 'sort': 'date',
 148                 'user': user_id,
 149                 'limit': self._PER_PAGE,
 150             })
 151         for x in traverse_obj(videos, ('results', ..., 'id')):
 152             yield self.url_result(f'https://iwara.tv/video/{x}')
 153
 154     def _real_extract(self, url):
 155         playlist_id = self._match_id(url)
 156         user_info = self._download_json(
 157             f'https://api.iwara.tv/profile/{playlist_id}', playlist_id,
 158             note='Requesting user info')
 159         user_id = traverse_obj(user_info, ('user', 'id'))
 160
 161         return self.playlist_result(
 162             OnDemandPagedList(
 163                 functools.partial(self._entries, playlist_id, user_id),
 164                 self._PER_PAGE),
 165             playlist_id, traverse_obj(user_info, ('user', 'name')))
 166
 167
 168 class IwaraPlaylistIE(InfoExtractor):
 169     # the ID is an UUID but I don't think it's necessary to write concrete regex
 170     _VALID_URL = r'https?://(?:www\.)?iwara\.tv/playlist/(?P<id>[0-9a-f-]+)'
 171     IE_NAME = 'iwara:playlist'
 172     _PER_PAGE = 32
 173
 174     _TESTS = [{
 175         'url': 'https://iwara.tv/playlist/458e5486-36a4-4ac0-b233-7e9eef01025f',
 176         'info_dict': {
 177             'id': '458e5486-36a4-4ac0-b233-7e9eef01025f',
 178         },
 179         'playlist_mincount': 3,
 180     }]
 181
 182     def _entries(self, playlist_id, first_page, page):
 183         videos = self._download_json(
 184             'https://api.iwara.tv/videos', playlist_id, f'Downloading page {page}',
 185             query={'page': page, 'limit': self._PER_PAGE}) if page else first_page
 186         for x in traverse_obj(videos, ('results', ..., 'id')):
 187             yield self.url_result(f'https://iwara.tv/video/{x}')
 188
 189     def _real_extract(self, url):
 190         playlist_id = self._match_id(url)
 191         page_0 = self._download_json(
 192             f'https://api.iwara.tv/playlist/{playlist_id}?page=0&limit={self._PER_PAGE}', playlist_id,
 193             note='Requesting playlist info')
 194
 195         return self.playlist_result(
 196             OnDemandPagedList(
 197                 functools.partial(self._entries, playlist_id, page_0),
 198                 self._PER_PAGE),
 199             playlist_id, traverse_obj(page_0, ('title', 'name')))