_ Git - youtube-dl/blob - youtube_dl/extractor/yandexmusic.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import hashlib
   6
   7 from .common import InfoExtractor
   8 from ..compat import (
   9     compat_str,
  10     compat_urllib_parse,
  11 )
  12 from ..utils import (
  13     ExtractorError,
  14     int_or_none,
  15     float_or_none,
  16     sanitized_Request,
  17 )
  18
  19
  20 class YandexMusicBaseIE(InfoExtractor):
  21     @staticmethod
  22     def _handle_error(response):
  23         error = response.get('error')
  24         if error:
  25             raise ExtractorError(error, expected=True)
  26
  27     def _download_json(self, *args, **kwargs):
  28         response = super(YandexMusicBaseIE, self)._download_json(*args, **kwargs)
  29         self._handle_error(response)
  30         return response
  31
  32
  33 class YandexMusicTrackIE(YandexMusicBaseIE):
  34     IE_NAME = 'yandexmusic:track'
  35     IE_DESC = 'Яндекс.Музыка - Трек'
  36     _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<album_id>\d+)/track/(?P<id>\d+)'
  37
  38     _TEST = {
  39         'url': 'http://music.yandex.ru/album/540508/track/4878838',
  40         'md5': 'f496818aa2f60b6c0062980d2e00dc20',
  41         'info_dict': {
  42             'id': '4878838',
  43             'ext': 'mp3',
  44             'title': 'Carlo Ambrosio - Gypsy Eyes 1',
  45             'filesize': 4628061,
  46             'duration': 193.04,
  47         }
  48     }
  49
  50     def _get_track_url(self, storage_dir, track_id):
  51         data = self._download_json(
  52             'http://music.yandex.ru/api/v1.5/handlers/api-jsonp.jsx?action=getTrackSrc&p=download-info/%s'
  53             % storage_dir,
  54             track_id, 'Downloading track location JSON')
  55
  56         key = hashlib.md5(('XGRlBW9FXlekgbPrRHuSiA' + data['path'][1:] + data['s']).encode('utf-8')).hexdigest()
  57         storage = storage_dir.split('.')
  58
  59         return ('http://%s/get-mp3/%s/%s?track-id=%s&from=service-10-track&similarities-experiment=default'
  60                 % (data['host'], key, data['ts'] + data['path'], storage[1]))
  61
  62     def _get_track_info(self, track):
  63         thumbnail = None
  64         cover_uri = track.get('albums', [{}])[0].get('coverUri')
  65         if cover_uri:
  66             thumbnail = cover_uri.replace('%%', 'orig')
  67             if not thumbnail.startswith('http'):
  68                 thumbnail = 'http://' + thumbnail
  69         return {
  70             'id': track['id'],
  71             'ext': 'mp3',
  72             'url': self._get_track_url(track['storageDir'], track['id']),
  73             'title': '%s - %s' % (track['artists'][0]['name'], track['title']),
  74             'filesize': int_or_none(track.get('fileSize')),
  75             'duration': float_or_none(track.get('durationMs'), 1000),
  76             'thumbnail': thumbnail,
  77         }
  78
  79     def _real_extract(self, url):
  80         mobj = re.match(self._VALID_URL, url)
  81         album_id, track_id = mobj.group('album_id'), mobj.group('id')
  82
  83         track = self._download_json(
  84             'http://music.yandex.ru/handlers/track.jsx?track=%s:%s' % (track_id, album_id),
  85             track_id, 'Downloading track JSON')['track']
  86
  87         return self._get_track_info(track)
  88
  89
  90 class YandexMusicPlaylistBaseIE(YandexMusicBaseIE):
  91     def _build_playlist(self, tracks):
  92         return [
  93             self.url_result(
  94                 'http://music.yandex.ru/album/%s/track/%s' % (track['albums'][0]['id'], track['id']))
  95             for track in tracks if track.get('albums') and isinstance(track.get('albums'), list)]
  96
  97
  98 class YandexMusicAlbumIE(YandexMusicPlaylistBaseIE):
  99     IE_NAME = 'yandexmusic:album'
 100     IE_DESC = 'Яндекс.Музыка - Альбом'
 101     _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<id>\d+)/?(\?|$)'
 102
 103     _TEST = {
 104         'url': 'http://music.yandex.ru/album/540508',
 105         'info_dict': {
 106             'id': '540508',
 107             'title': 'Carlo Ambrosio - Gypsy Soul (2009)',
 108         },
 109         'playlist_count': 50,
 110     }
 111
 112     def _real_extract(self, url):
 113         album_id = self._match_id(url)
 114
 115         album = self._download_json(
 116             'http://music.yandex.ru/handlers/album.jsx?album=%s' % album_id,
 117             album_id, 'Downloading album JSON')
 118
 119         entries = self._build_playlist(album['volumes'][0])
 120
 121         title = '%s - %s' % (album['artists'][0]['name'], album['title'])
 122         year = album.get('year')
 123         if year:
 124             title += ' (%s)' % year
 125
 126         return self.playlist_result(entries, compat_str(album['id']), title)
 127
 128
 129 class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE):
 130     IE_NAME = 'yandexmusic:playlist'
 131     IE_DESC = 'Яндекс.Музыка - Плейлист'
 132     _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/users/[^/]+/playlists/(?P<id>\d+)'
 133
 134     _TESTS = [{
 135         'url': 'http://music.yandex.ru/users/music.partners/playlists/1245',
 136         'info_dict': {
 137             'id': '1245',
 138             'title': 'Что слушают Enter Shikari',
 139             'description': 'md5:3b9f27b0efbe53f2ee1e844d07155cc9',
 140         },
 141         'playlist_count': 6,
 142     }, {
 143         # playlist exceeding the limit of 150 tracks shipped with webpage (see
 144         # https://github.com/rg3/youtube-dl/issues/6666)
 145         'url': 'https://music.yandex.ru/users/ya.playlist/playlists/1036',
 146         'info_dict': {
 147             'id': '1036',
 148             'title': 'Музыка 90-х',
 149         },
 150         'playlist_count': 310,
 151     }]
 152
 153     def _real_extract(self, url):
 154         playlist_id = self._match_id(url)
 155
 156         webpage = self._download_webpage(url, playlist_id)
 157
 158         mu = self._parse_json(
 159             self._search_regex(
 160                 r'var\s+Mu\s*=\s*({.+?});\s*</script>', webpage, 'player'),
 161             playlist_id)
 162
 163         playlist = mu['pageData']['playlist']
 164         tracks, track_ids = playlist['tracks'], playlist['trackIds']
 165
 166         # tracks dictionary shipped with webpage is limited to 150 tracks,
 167         # missing tracks should be retrieved manually.
 168         if len(tracks) < len(track_ids):
 169             present_track_ids = set([compat_str(track['id']) for track in tracks if track.get('id')])
 170             missing_track_ids = set(map(compat_str, track_ids)) - set(present_track_ids)
 171             request = sanitized_Request(
 172                 'https://music.yandex.ru/handlers/track-entries.jsx',
 173                 compat_urllib_parse.urlencode({
 174                     'entries': ','.join(missing_track_ids),
 175                     'lang': mu.get('settings', {}).get('lang', 'en'),
 176                     'external-domain': 'music.yandex.ru',
 177                     'overembed': 'false',
 178                     'sign': mu.get('authData', {}).get('user', {}).get('sign'),
 179                     'strict': 'true',
 180                 }).encode('utf-8'))
 181             request.add_header('Referer', url)
 182             request.add_header('X-Requested-With', 'XMLHttpRequest')
 183
 184             missing_tracks = self._download_json(
 185                 request, playlist_id, 'Downloading missing tracks JSON', fatal=False)
 186             if missing_tracks:
 187                 tracks.extend(missing_tracks)
 188
 189         return self.playlist_result(
 190             self._build_playlist(tracks),
 191             compat_str(playlist_id),
 192             playlist['title'], playlist.get('description'))