yt_dlp/extractor/mailru.py

   1 import itertools
   2 import json
   3 import re
   4 import urllib.parse
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     int_or_none,
   9     parse_duration,
  10     remove_end,
  11     try_get,
  12     urljoin,
  13 )
  14
  15
  16 class MailRuIE(InfoExtractor):
  17     IE_NAME = 'mailru'
  18     IE_DESC = 'Видео@Mail.Ru'
  19     _VALID_URL = r'''(?x)
  20                     https?://
  21                         (?:(?:www|m|videoapi)\.)?my\.mail\.ru/+
  22                         (?:
  23                             video/.*\#video=/?(?P<idv1>(?:[^/]+/){3}\d+)|
  24                             (?:videos/embed/)?(?:(?P<idv2prefix>(?:[^/]+/+){2})(?:video/(?:embed/)?)?(?P<idv2suffix>[^/]+/\d+))(?:\.html)?|
  25                             (?:video/embed|\+/video/meta)/(?P<metaid>\d+)
  26                         )
  27                     '''
  28     _TESTS = [
  29         {
  30             'url': 'http://my.mail.ru/video/top#video=/mail/sonypicturesrus/75/76',
  31             'md5': 'dea205f03120046894db4ebb6159879a',
  32             'info_dict': {
  33                 'id': '46301138_76',
  34                 'ext': 'mp4',
  35                 'title': 'Новый Человек-Паук. Высокое напряжение. Восстание Электро',
  36                 'timestamp': 1393235077,
  37                 'upload_date': '20140224',
  38                 'uploader': 'sonypicturesrus',
  39                 'uploader_id': 'sonypicturesrus@mail.ru',
  40                 'duration': 184,
  41             },
  42             'skip': 'Not accessible from Travis CI server',
  43         },
  44         {
  45             'url': 'http://my.mail.ru/corp/hitech/video/news_hi-tech_mail_ru/1263.html',
  46             'md5': '00a91a58c3402204dcced523777b475f',
  47             'info_dict': {
  48                 'id': '46843144_1263',
  49                 'ext': 'mp4',
  50                 'title': 'Samsung Galaxy S5 Hammer Smash Fail Battery Explosion',
  51                 'timestamp': 1397039888,
  52                 'upload_date': '20140409',
  53                 'uploader': 'hitech',
  54                 'uploader_id': 'hitech@corp.mail.ru',
  55                 'duration': 245,
  56             },
  57             'skip': 'Not accessible from Travis CI server',
  58         },
  59         {
  60             # only available via metaUrl API
  61             'url': 'http://my.mail.ru/mail/720pizle/video/_myvideo/502.html',
  62             'md5': '3b26d2491c6949d031a32b96bd97c096',
  63             'info_dict': {
  64                 'id': '56664382_502',
  65                 'ext': 'mp4',
  66                 'title': ':8336',
  67                 'timestamp': 1449094163,
  68                 'upload_date': '20151202',
  69                 'uploader': '720pizle@mail.ru',
  70                 'uploader_id': '720pizle@mail.ru',
  71                 'duration': 6001,
  72             },
  73             'skip': 'Not accessible from Travis CI server',
  74         },
  75         {
  76             'url': 'http://m.my.mail.ru/mail/3sktvtr/video/_myvideo/138.html',
  77             'only_matching': True,
  78         },
  79         {
  80             'url': 'https://my.mail.ru/video/embed/7949340477499637815',
  81             'only_matching': True,
  82         },
  83         {
  84             'url': 'http://my.mail.ru/+/video/meta/7949340477499637815',
  85             'only_matching': True,
  86         },
  87         {
  88             'url': 'https://my.mail.ru//list/sinyutin10/video/_myvideo/4.html',
  89             'only_matching': True,
  90         },
  91         {
  92             'url': 'https://my.mail.ru//list//sinyutin10/video/_myvideo/4.html',
  93             'only_matching': True,
  94         },
  95         {
  96             'url': 'https://my.mail.ru/mail/cloud-strife/video/embed/Games/2009',
  97             'only_matching': True,
  98         },
  99         {
 100             'url': 'https://videoapi.my.mail.ru/videos/embed/mail/cloud-strife/Games/2009.html',
 101             'only_matching': True,
 102         },
 103     ]
 104
 105     def _real_extract(self, url):
 106         mobj = self._match_valid_url(url)
 107         meta_id = mobj.group('metaid')
 108
 109         video_id = None
 110         if meta_id:
 111             meta_url = f'https://my.mail.ru/+/video/meta/{meta_id}'
 112         else:
 113             video_id = mobj.group('idv1')
 114             if not video_id:
 115                 video_id = mobj.group('idv2prefix') + mobj.group('idv2suffix')
 116             webpage = self._download_webpage(url, video_id)
 117             page_config = self._parse_json(self._search_regex([
 118                 r'(?s)<script[^>]+class="sp-video__page-config"[^>]*>(.+?)</script>',
 119                 r'(?s)"video":\s*({.+?}),'],
 120                 webpage, 'page config', default='{}'), video_id, fatal=False)
 121             if page_config:
 122                 meta_url = page_config.get('metaUrl') or page_config.get('video', {}).get('metaUrl') or page_config.get('metadataUrl')
 123             else:
 124                 meta_url = None
 125
 126         video_data = None
 127
 128         # fix meta_url if missing the host address
 129         if re.match(r'\/\+\/', meta_url):
 130             meta_url = urljoin('https://my.mail.ru', meta_url)
 131
 132         if meta_url:
 133             video_data = self._download_json(
 134                 meta_url, video_id or meta_id, 'Downloading video meta JSON',
 135                 fatal=not video_id)
 136
 137         # Fallback old approach
 138         if not video_data:
 139             video_data = self._download_json(
 140                 f'http://api.video.mail.ru/videos/{video_id}.json?new=1',
 141                 video_id, 'Downloading video JSON')
 142
 143         video_key = self._get_cookies('https://my.mail.ru').get('video_key')
 144
 145         formats = []
 146         for f in video_data['videos']:
 147             video_url = f.get('url')
 148             if not video_url:
 149                 continue
 150             if video_key:
 151                 self._set_cookie(urllib.parse.urlparse(video_url).hostname, 'video_key', video_key.value)
 152             format_id = f.get('key')
 153             height = int_or_none(self._search_regex(
 154                 r'^(\d+)[pP]$', format_id, 'height', default=None)) if format_id else None
 155             formats.append({
 156                 'url': video_url,
 157                 'format_id': format_id,
 158                 'height': height,
 159             })
 160
 161         meta_data = video_data['meta']
 162         title = remove_end(meta_data['title'], '.mp4')
 163
 164         author = video_data.get('author')
 165         uploader = author.get('name')
 166         uploader_id = author.get('id') or author.get('email')
 167         view_count = int_or_none(video_data.get('viewsCount') or video_data.get('views_count'))
 168
 169         acc_id = meta_data.get('accId')
 170         item_id = meta_data.get('itemId')
 171         content_id = f'{acc_id}_{item_id}' if acc_id and item_id else video_id
 172
 173         thumbnail = meta_data.get('poster')
 174         duration = int_or_none(meta_data.get('duration'))
 175         timestamp = int_or_none(meta_data.get('timestamp'))
 176
 177         return {
 178             'id': content_id,
 179             'title': title,
 180             'thumbnail': thumbnail,
 181             'timestamp': timestamp,
 182             'uploader': uploader,
 183             'uploader_id': uploader_id,
 184             'duration': duration,
 185             'view_count': view_count,
 186             'formats': formats,
 187         }
 188
 189
 190 class MailRuMusicSearchBaseIE(InfoExtractor):
 191     def _search(self, query, url, audio_id, limit=100, offset=0):
 192         search = self._download_json(
 193             'https://my.mail.ru/cgi-bin/my/ajax', audio_id,
 194             f'Downloading songs JSON page {offset // limit + 1}',
 195             headers={
 196                 'Referer': url,
 197                 'X-Requested-With': 'XMLHttpRequest',
 198             }, query={
 199                 'xemail': '',
 200                 'ajax_call': '1',
 201                 'func_name': 'music.search',
 202                 'mna': '',
 203                 'mnb': '',
 204                 'arg_query': query,
 205                 'arg_extended': '1',
 206                 'arg_search_params': json.dumps({
 207                     'music': {
 208                         'limit': limit,
 209                         'offset': offset,
 210                     },
 211                 }),
 212                 'arg_limit': limit,
 213                 'arg_offset': offset,
 214             })
 215         return next(e for e in search if isinstance(e, dict))
 216
 217     @staticmethod
 218     def _extract_track(t, fatal=True):
 219         audio_url = t['URL'] if fatal else t.get('URL')
 220         if not audio_url:
 221             return
 222
 223         audio_id = t['File'] if fatal else t.get('File')
 224         if not audio_id:
 225             return
 226
 227         thumbnail = t.get('AlbumCoverURL') or t.get('FiledAlbumCover')
 228         uploader = t.get('OwnerName') or t.get('OwnerName_Text_HTML')
 229         uploader_id = t.get('UploaderID')
 230         duration = int_or_none(t.get('DurationInSeconds')) or parse_duration(
 231             t.get('Duration') or t.get('DurationStr'))
 232         view_count = int_or_none(t.get('PlayCount') or t.get('PlayCount_hr'))
 233
 234         track = t.get('Name') or t.get('Name_Text_HTML')
 235         artist = t.get('Author') or t.get('Author_Text_HTML')
 236
 237         if track:
 238             title = f'{artist} - {track}' if artist else track
 239         else:
 240             title = audio_id
 241
 242         return {
 243             'extractor_key': MailRuMusicIE.ie_key(),
 244             'id': audio_id,
 245             'title': title,
 246             'thumbnail': thumbnail,
 247             'uploader': uploader,
 248             'uploader_id': uploader_id,
 249             'duration': duration,
 250             'view_count': view_count,
 251             'vcodec': 'none',
 252             'abr': int_or_none(t.get('BitRate')),
 253             'track': track,
 254             'artist': artist,
 255             'album': t.get('Album'),
 256             'url': audio_url,
 257         }
 258
 259
 260 class MailRuMusicIE(MailRuMusicSearchBaseIE):
 261     IE_NAME = 'mailru:music'
 262     IE_DESC = 'Музыка@Mail.Ru'
 263     _VALID_URL = r'https?://my\.mail\.ru/+music/+songs/+[^/?#&]+-(?P<id>[\da-f]+)'
 264     _TESTS = [{
 265         'url': 'https://my.mail.ru/music/songs/%D0%BC8%D0%BB8%D1%82%D1%85-l-a-h-luciferian-aesthetics-of-herrschaft-single-2017-4e31f7125d0dfaef505d947642366893',
 266         'md5': '0f8c22ef8c5d665b13ac709e63025610',
 267         'info_dict': {
 268             'id': '4e31f7125d0dfaef505d947642366893',
 269             'ext': 'mp3',
 270             'title': 'L.A.H. (Luciferian Aesthetics of Herrschaft) single, 2017 - М8Л8ТХ',
 271             'uploader': 'Игорь Мудрый',
 272             'uploader_id': '1459196328',
 273             'duration': 280,
 274             'view_count': int,
 275             'vcodec': 'none',
 276             'abr': 320,
 277             'track': 'L.A.H. (Luciferian Aesthetics of Herrschaft) single, 2017',
 278             'artist': 'М8Л8ТХ',
 279         },
 280     }]
 281
 282     def _real_extract(self, url):
 283         audio_id = self._match_id(url)
 284
 285         webpage = self._download_webpage(url, audio_id)
 286
 287         title = self._og_search_title(webpage)
 288         music_data = self._search(title, url, audio_id)['MusicData']
 289         t = next(t for t in music_data if t.get('File') == audio_id)
 290
 291         info = self._extract_track(t)
 292         info['title'] = title
 293         return info
 294
 295
 296 class MailRuMusicSearchIE(MailRuMusicSearchBaseIE):
 297     IE_NAME = 'mailru:music:search'
 298     IE_DESC = 'Музыка@Mail.Ru'
 299     _VALID_URL = r'https?://my\.mail\.ru/+music/+search/+(?P<id>[^/?#&]+)'
 300     _TESTS = [{
 301         'url': 'https://my.mail.ru/music/search/black%20shadow',
 302         'info_dict': {
 303             'id': 'black shadow',
 304         },
 305         'playlist_mincount': 532,
 306     }]
 307
 308     def _real_extract(self, url):
 309         query = urllib.parse.unquote(self._match_id(url))
 310
 311         entries = []
 312
 313         LIMIT = 100
 314         offset = 0
 315
 316         for _ in itertools.count(1):
 317             search = self._search(query, url, query, LIMIT, offset)
 318
 319             music_data = search.get('MusicData')
 320             if not music_data or not isinstance(music_data, list):
 321                 break
 322
 323             for t in music_data:
 324                 track = self._extract_track(t, fatal=False)
 325                 if track:
 326                     entries.append(track)
 327
 328             total = try_get(
 329                 search, lambda x: x['Results']['music']['Total'], int)
 330
 331             if total is not None:
 332                 if offset > total:
 333                     break
 334
 335             offset += LIMIT
 336
 337         return self.playlist_result(entries, query)