yt_dlp/extractor/murrtube.py

   1 import functools
   2 import json
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     ExtractorError,
   7     OnDemandPagedList,
   8     determine_ext,
   9     int_or_none,
  10     try_get,
  11 )
  12
  13
  14 class MurrtubeIE(InfoExtractor):
  15     _VALID_URL = r'''(?x)
  16                         (?:
  17                             murrtube:|
  18                             https?://murrtube\.net/videos/(?P<slug>[a-z0-9\-]+)\-
  19                         )
  20                         (?P<id>[a-f0-9]{8}\-[a-f0-9]{4}\-[a-f0-9]{4}\-[a-f0-9]{4}\-[a-f0-9]{12})
  21                     '''
  22     _TEST = {
  23         'url': 'https://murrtube.net/videos/inferno-x-skyler-148b6f2a-fdcc-4902-affe-9c0f41aaaca0',
  24         'md5': '169f494812d9a90914b42978e73aa690',
  25         'info_dict': {
  26             'id': '148b6f2a-fdcc-4902-affe-9c0f41aaaca0',
  27             'ext': 'mp4',
  28             'title': 'Inferno X Skyler',
  29             'description': 'Humping a very good slutty sheppy (roomate)',
  30             'thumbnail': r're:^https?://.*\.jpg$',
  31             'duration': 284,
  32             'uploader': 'Inferno Wolf',
  33             'age_limit': 18,
  34             'comment_count': int,
  35             'view_count': int,
  36             'like_count': int,
  37             'tags': ['hump', 'breed', 'Fursuit', 'murrsuit', 'bareback'],
  38         }
  39     }
  40
  41     def _download_gql(self, video_id, op, note=None, fatal=True):
  42         result = self._download_json(
  43             'https://murrtube.net/graphql',
  44             video_id, note, data=json.dumps(op).encode(), fatal=fatal,
  45             headers={'Content-Type': 'application/json'})
  46         return result['data']
  47
  48     def _real_extract(self, url):
  49         video_id = self._match_id(url)
  50         data = self._download_gql(video_id, {
  51             'operationName': 'Medium',
  52             'variables': {
  53                 'id': video_id,
  54             },
  55             'query': '''\
  56 query Medium($id: ID!) {
  57   medium(id: $id) {
  58     title
  59     description
  60     key
  61     duration
  62     commentsCount
  63     likesCount
  64     viewsCount
  65     thumbnailKey
  66     tagList
  67     user {
  68       name
  69       __typename
  70     }
  71     __typename
  72   }
  73 }'''})
  74         meta = data['medium']
  75
  76         storage_url = 'https://storage.murrtube.net/murrtube/'
  77         format_url = storage_url + meta.get('key', '')
  78         thumbnail = storage_url + meta.get('thumbnailKey', '')
  79
  80         if determine_ext(format_url) == 'm3u8':
  81             formats = self._extract_m3u8_formats(
  82                 format_url, video_id, 'mp4', entry_protocol='m3u8_native', fatal=False)
  83         else:
  84             formats = [{'url': format_url}]
  85
  86         return {
  87             'id': video_id,
  88             'title': meta.get('title'),
  89             'description': meta.get('description'),
  90             'formats': formats,
  91             'thumbnail': thumbnail,
  92             'duration': int_or_none(meta.get('duration')),
  93             'uploader': try_get(meta, lambda x: x['user']['name']),
  94             'view_count': meta.get('viewsCount'),
  95             'like_count': meta.get('likesCount'),
  96             'comment_count': meta.get('commentsCount'),
  97             'tags': meta.get('tagList'),
  98             'age_limit': 18,
  99         }
 100
 101
 102 class MurrtubeUserIE(MurrtubeIE):  # XXX: Do not subclass from concrete IE
 103     IE_DESC = 'Murrtube user profile'
 104     _VALID_URL = r'https?://murrtube\.net/(?P<id>[^/]+)$'
 105     _TEST = {
 106         'url': 'https://murrtube.net/stormy',
 107         'info_dict': {
 108             'id': 'stormy',
 109         },
 110         'playlist_mincount': 27,
 111     }
 112     _PAGE_SIZE = 10
 113
 114     def _fetch_page(self, username, user_id, page):
 115         data = self._download_gql(username, {
 116             'operationName': 'Media',
 117             'variables': {
 118                 'limit': self._PAGE_SIZE,
 119                 'offset': page * self._PAGE_SIZE,
 120                 'sort': 'latest',
 121                 'userId': user_id,
 122             },
 123             'query': '''\
 124 query Media($q: String, $sort: String, $userId: ID, $offset: Int!, $limit: Int!) {
 125   media(q: $q, sort: $sort, userId: $userId, offset: $offset, limit: $limit) {
 126     id
 127     __typename
 128   }
 129 }'''},
 130             'Downloading page {0}'.format(page + 1))
 131         if data is None:
 132             raise ExtractorError(f'Failed to retrieve video list for page {page + 1}')
 133
 134         media = data['media']
 135
 136         for entry in media:
 137             yield self.url_result('murrtube:{0}'.format(entry['id']), MurrtubeIE.ie_key())
 138
 139     def _real_extract(self, url):
 140         username = self._match_id(url)
 141         data = self._download_gql(username, {
 142             'operationName': 'User',
 143             'variables': {
 144                 'id': username,
 145             },
 146             'query': '''\
 147 query User($id: ID!) {
 148   user(id: $id) {
 149     id
 150     __typename
 151   }
 152 }'''},
 153             'Downloading user info')
 154         if data is None:
 155             raise ExtractorError('Failed to fetch user info')
 156
 157         user = data['user']
 158
 159         entries = OnDemandPagedList(functools.partial(
 160             self._fetch_page, username, user.get('id')), self._PAGE_SIZE)
 161
 162         return self.playlist_result(entries, username)