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     _WORKING = False
  16     _VALID_URL = r'''(?x)
  17                         (?:
  18                             murrtube:|
  19                             https?://murrtube\.net/videos/(?P<slug>[a-z0-9\-]+)\-
  20                         )
  21                         (?P<id>[a-f0-9]{8}\-[a-f0-9]{4}\-[a-f0-9]{4}\-[a-f0-9]{4}\-[a-f0-9]{12})
  22                     '''
  23     _TEST = {
  24         'url': 'https://murrtube.net/videos/inferno-x-skyler-148b6f2a-fdcc-4902-affe-9c0f41aaaca0',
  25         'md5': '169f494812d9a90914b42978e73aa690',
  26         'info_dict': {
  27             'id': '148b6f2a-fdcc-4902-affe-9c0f41aaaca0',
  28             'ext': 'mp4',
  29             'title': 'Inferno X Skyler',
  30             'description': 'Humping a very good slutty sheppy (roomate)',
  31             'thumbnail': r're:^https?://.*\.jpg$',
  32             'duration': 284,
  33             'uploader': 'Inferno Wolf',
  34             'age_limit': 18,
  35             'comment_count': int,
  36             'view_count': int,
  37             'like_count': int,
  38             'tags': ['hump', 'breed', 'Fursuit', 'murrsuit', 'bareback'],
  39         },
  40     }
  41
  42     def _download_gql(self, video_id, op, note=None, fatal=True):
  43         result = self._download_json(
  44             'https://murrtube.net/graphql',
  45             video_id, note, data=json.dumps(op).encode(), fatal=fatal,
  46             headers={'Content-Type': 'application/json'})
  47         return result['data']
  48
  49     def _real_extract(self, url):
  50         video_id = self._match_id(url)
  51         data = self._download_gql(video_id, {
  52             'operationName': 'Medium',
  53             'variables': {
  54                 'id': video_id,
  55             },
  56             'query': '''\
  57 query Medium($id: ID!) {
  58   medium(id: $id) {
  59     title
  60     description
  61     key
  62     duration
  63     commentsCount
  64     likesCount
  65     viewsCount
  66     thumbnailKey
  67     tagList
  68     user {
  69       name
  70       __typename
  71     }
  72     __typename
  73   }
  74 }'''})
  75         meta = data['medium']
  76
  77         storage_url = 'https://storage.murrtube.net/murrtube/'
  78         format_url = storage_url + meta.get('key', '')
  79         thumbnail = storage_url + meta.get('thumbnailKey', '')
  80
  81         if determine_ext(format_url) == 'm3u8':
  82             formats = self._extract_m3u8_formats(
  83                 format_url, video_id, 'mp4', entry_protocol='m3u8_native', fatal=False)
  84         else:
  85             formats = [{'url': format_url}]
  86
  87         return {
  88             'id': video_id,
  89             'title': meta.get('title'),
  90             'description': meta.get('description'),
  91             'formats': formats,
  92             'thumbnail': thumbnail,
  93             'duration': int_or_none(meta.get('duration')),
  94             'uploader': try_get(meta, lambda x: x['user']['name']),
  95             'view_count': meta.get('viewsCount'),
  96             'like_count': meta.get('likesCount'),
  97             'comment_count': meta.get('commentsCount'),
  98             'tags': meta.get('tagList'),
  99             'age_limit': 18,
 100         }
 101
 102
 103 class MurrtubeUserIE(MurrtubeIE):  # XXX: Do not subclass from concrete IE
 104     _WORKING = False
 105     IE_DESC = 'Murrtube user profile'
 106     _VALID_URL = r'https?://murrtube\.net/(?P<id>[^/]+)$'
 107     _TEST = {
 108         'url': 'https://murrtube.net/stormy',
 109         'info_dict': {
 110             'id': 'stormy',
 111         },
 112         'playlist_mincount': 27,
 113     }
 114     _PAGE_SIZE = 10
 115
 116     def _fetch_page(self, username, user_id, page):
 117         data = self._download_gql(username, {
 118             'operationName': 'Media',
 119             'variables': {
 120                 'limit': self._PAGE_SIZE,
 121                 'offset': page * self._PAGE_SIZE,
 122                 'sort': 'latest',
 123                 'userId': user_id,
 124             },
 125             'query': '''\
 126 query Media($q: String, $sort: String, $userId: ID, $offset: Int!, $limit: Int!) {
 127   media(q: $q, sort: $sort, userId: $userId, offset: $offset, limit: $limit) {
 128     id
 129     __typename
 130   }
 131 }'''},
 132             f'Downloading page {page + 1}')
 133         if data is None:
 134             raise ExtractorError(f'Failed to retrieve video list for page {page + 1}')
 135
 136         media = data['media']
 137
 138         for entry in media:
 139             yield self.url_result('murrtube:{}'.format(entry['id']), MurrtubeIE.ie_key())
 140
 141     def _real_extract(self, url):
 142         username = self._match_id(url)
 143         data = self._download_gql(username, {
 144             'operationName': 'User',
 145             'variables': {
 146                 'id': username,
 147             },
 148             'query': '''\
 149 query User($id: ID!) {
 150   user(id: $id) {
 151     id
 152     __typename
 153   }
 154 }'''},
 155             'Downloading user info')
 156         if data is None:
 157             raise ExtractorError('Failed to fetch user info')
 158
 159         user = data['user']
 160
 161         entries = OnDemandPagedList(functools.partial(
 162             self._fetch_page, username, user.get('id')), self._PAGE_SIZE)
 163
 164         return self.playlist_result(entries, username)