_ 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     compat_urllib_request,
  12 )
  13 from ..utils import (
  14     int_or_none,
  15     float_or_none,
  16 )
  17
  18
  19 class YandexMusicTrackIE(InfoExtractor):
  20     IE_NAME = 'yandexmusic:track'
  21     IE_DESC = 'Яндекс.Музыка - Трек'
  22     _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<album_id>\d+)/track/(?P<id>\d+)'
  23
  24     _TEST = {
  25         'url': 'http://music.yandex.ru/album/540508/track/4878838',
  26         'md5': 'f496818aa2f60b6c0062980d2e00dc20',
  27         'info_dict': {
  28             'id': '4878838',
  29             'ext': 'mp3',
  30             'title': 'Carlo Ambrosio - Gypsy Eyes 1',
  31             'filesize': 4628061,
  32             'duration': 193.04,
  33         }
  34     }
  35
  36     def _get_track_url(self, storage_dir, track_id):
  37         data = self._download_json(
  38             'http://music.yandex.ru/api/v1.5/handlers/api-jsonp.jsx?action=getTrackSrc&p=download-info/%s'
  39             % storage_dir,
  40             track_id, 'Downloading track location JSON')
  41
  42         key = hashlib.md5(('XGRlBW9FXlekgbPrRHuSiA' + data['path'][1:] + data['s']).encode('utf-8')).hexdigest()
  43         storage = storage_dir.split('.')
  44
  45         return ('http://%s/get-mp3/%s/%s?track-id=%s&from=service-10-track&similarities-experiment=default'
  46                 % (data['host'], key, data['ts'] + data['path'], storage[1]))
  47
  48     def _get_track_info(self, track):
  49         album = track['albums'][0]
  50         a_thumb = None
  51
  52         if 'coverUri' in album:
  53             a_thumb = album['coverUri']
  54             if a_thumb:
  55                 a_thumb = 'http://' + a_thumb.replace('%%', '1000x1000')
  56
  57         return {
  58             'id': track['id'],
  59             'ext': 'mp3',
  60             'url': self._get_track_url(track['storageDir'], track['id']),
  61             'title': '%s - %s' % (track['artists'][0]['name'], track['title']),
  62             'filesize': int_or_none(track.get('fileSize')),
  63             'duration': float_or_none(track.get('durationMs'), 1000),
  64             'thumbnail': a_thumb,
  65         }
  66
  67     def _real_extract(self, url):
  68         mobj = re.match(self._VALID_URL, url)
  69         album_id, track_id = mobj.group('album_id'), mobj.group('id')
  70
  71         track = self._download_json(
  72             'http://music.yandex.ru/handlers/track.jsx?track=%s:%s' % (track_id, album_id),
  73             track_id, 'Downloading track JSON')['track']
  74
  75         return self._get_track_info(track)
  76
  77
  78 class YandexMusicPlaylistBaseIE(InfoExtractor):
  79     def _build_playlist(self, tracks):
  80         return [
  81             self.url_result(
  82                 'http://music.yandex.ru/album/%s/track/%s' % (track['albums'][0]['id'], track['id']))
  83             for track in tracks if track.get('albums') and isinstance(track.get('albums'), list)]
  84
  85
  86 class YandexMusicAlbumIE(YandexMusicPlaylistBaseIE):
  87     IE_NAME = 'yandexmusic:album'
  88     IE_DESC = 'Яндекс.Музыка - Альбом'
  89     _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<id>\d+)/?(\?|$)'
  90
  91     _TEST = {
  92         'url': 'http://music.yandex.ru/album/540508',
  93         'info_dict': {
  94             'id': '540508',
  95             'title': 'Carlo Ambrosio - Gypsy Soul (2009)',
  96         },
  97         'playlist_count': 50,
  98     }
  99
 100     def _real_extract(self, url):
 101         album_id = self._match_id(url)
 102
 103         album = self._download_json(
 104             'http://music.yandex.ru/handlers/album.jsx?album=%s' % album_id,
 105             album_id, 'Downloading album JSON')
 106
 107         entries = self._build_playlist(album['volumes'][0])
 108
 109         title = '%s - %s' % (album['artists'][0]['name'], album['title'])
 110         year = album.get('year')
 111         if year:
 112             title += ' (%s)' % year
 113
 114         return self.playlist_result(entries, compat_str(album['id']), title)
 115
 116
 117 class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE):
 118     IE_NAME = 'yandexmusic:playlist'
 119     IE_DESC = 'Яндекс.Музыка - Плейлист'
 120     _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/users/[^/]+/playlists/(?P<id>\d+)'
 121
 122     _TESTS = [{
 123         'url': 'http://music.yandex.ru/users/music.partners/playlists/1245',
 124         'info_dict': {
 125             'id': '1245',
 126             'title': 'Что слушают Enter Shikari',
 127             'description': 'md5:3b9f27b0efbe53f2ee1e844d07155cc9',
 128         },
 129         'playlist_count': 6,
 130     }, {
 131         # playlist exceeding the limit of 150 tracks shipped with webpage (see
 132         # https://github.com/rg3/youtube-dl/issues/6666)
 133         'url': 'https://music.yandex.ru/users/ya.playlist/playlists/1036',
 134         'info_dict': {
 135             'id': '1036',
 136             'title': 'Музыка 90-х',
 137         },
 138         'playlist_count': 310,
 139     }]
 140
 141     def _real_extract(self, url):
 142         playlist_id = self._match_id(url)
 143
 144         webpage = self._download_webpage(url, playlist_id)
 145
 146         mu = self._parse_json(
 147             self._search_regex(
 148                 r'var\s+Mu\s*=\s*({.+?});\s*</script>', webpage, 'player'),
 149             playlist_id)
 150
 151         playlist = mu['pageData']['playlist']
 152         tracks, track_ids = playlist['tracks'], playlist['trackIds']
 153
 154         # tracks dictionary shipped with webpage is limited to 150 tracks,
 155         # missing tracks should be retrieved manually.
 156         if len(tracks) < len(track_ids):
 157             present_track_ids = set([compat_str(track['id']) for track in tracks if track.get('id')])
 158             missing_track_ids = set(map(compat_str, track_ids)) - set(present_track_ids)
 159             request = compat_urllib_request.Request(
 160                 'https://music.yandex.ru/handlers/track-entries.jsx',
 161                 compat_urllib_parse.urlencode({
 162                     'entries': ','.join(missing_track_ids),
 163                     'lang': mu.get('settings', {}).get('lang', 'en'),
 164                     'external-domain': 'music.yandex.ru',
 165                     'overembed': 'false',
 166                     'sign': mu.get('authData', {}).get('user', {}).get('sign'),
 167                     'strict': 'true',
 168                 }).encode('utf-8'))
 169             request.add_header('Referer', url)
 170             request.add_header('X-Requested-With', 'XMLHttpRequest')
 171
 172             missing_tracks = self._download_json(
 173                 request, playlist_id, 'Downloading missing tracks JSON', fatal=False)
 174             if missing_tracks:
 175                 tracks.extend(missing_tracks)
 176
 177         return self.playlist_result(
 178             self._build_playlist(tracks),
 179             compat_str(playlist_id),
 180             playlist['title'], playlist.get('description'))