_ 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 compat_str
   9 from ..utils import (
  10     ExtractorError,
  11     int_or_none,
  12     float_or_none,
  13     sanitized_Request,
  14     urlencode_postdata,
  15 )
  16
  17
  18 class YandexMusicBaseIE(InfoExtractor):
  19     @staticmethod
  20     def _handle_error(response):
  21         error = response.get('error')
  22         if error:
  23             raise ExtractorError(error, expected=True)
  24
  25     def _download_webpage(self, *args, **kwargs):
  26         webpage = super(YandexMusicBaseIE, self)._download_webpage(*args, **kwargs)
  27         if 'Нам очень жаль, но&nbsp;запросы, поступившие с&nbsp;вашего IP-адреса, похожи на&nbsp;автоматические.' in webpage:
  28             raise ExtractorError('Blocked by YandexMusic', expected=True)
  29         return webpage
  30
  31     def _download_json(self, *args, **kwargs):
  32         response = super(YandexMusicBaseIE, self)._download_json(*args, **kwargs)
  33         self._handle_error(response)
  34         return response
  35
  36
  37 class YandexMusicTrackIE(YandexMusicBaseIE):
  38     IE_NAME = 'yandexmusic:track'
  39     IE_DESC = 'Яндекс.Музыка - Трек'
  40     _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<album_id>\d+)/track/(?P<id>\d+)'
  41
  42     _TEST = {
  43         'url': 'http://music.yandex.ru/album/540508/track/4878838',
  44         'md5': 'f496818aa2f60b6c0062980d2e00dc20',
  45         'info_dict': {
  46             'id': '4878838',
  47             'ext': 'mp3',
  48             'title': 'Carlo Ambrosio & Fabio Di Bari, Carlo Ambrosio - Gypsy Eyes 1',
  49             'filesize': 4628061,
  50             'duration': 193.04,
  51             'track': 'Gypsy Eyes 1',
  52             'album': 'Gypsy Soul',
  53             'album_artist': 'Carlo Ambrosio',
  54             'artist': 'Carlo Ambrosio & Fabio Di Bari, Carlo Ambrosio',
  55             'release_year': '2009',
  56         },
  57         'skip': 'Travis CI servers blocked by YandexMusic',
  58     }
  59
  60     def _get_track_url(self, storage_dir, track_id):
  61         data = self._download_json(
  62             'http://music.yandex.ru/api/v1.5/handlers/api-jsonp.jsx?action=getTrackSrc&p=download-info/%s'
  63             % storage_dir,
  64             track_id, 'Downloading track location JSON')
  65
  66         key = hashlib.md5(('XGRlBW9FXlekgbPrRHuSiA' + data['path'][1:] + data['s']).encode('utf-8')).hexdigest()
  67         storage = storage_dir.split('.')
  68
  69         return ('http://%s/get-mp3/%s/%s?track-id=%s&from=service-10-track&similarities-experiment=default'
  70                 % (data['host'], key, data['ts'] + data['path'], storage[1]))
  71
  72     def _get_track_info(self, track):
  73         thumbnail = None
  74         cover_uri = track.get('albums', [{}])[0].get('coverUri')
  75         if cover_uri:
  76             thumbnail = cover_uri.replace('%%', 'orig')
  77             if not thumbnail.startswith('http'):
  78                 thumbnail = 'http://' + thumbnail
  79
  80         track_title = track['title']
  81         track_info = {
  82             'id': track['id'],
  83             'ext': 'mp3',
  84             'url': self._get_track_url(track['storageDir'], track['id']),
  85             'filesize': int_or_none(track.get('fileSize')),
  86             'duration': float_or_none(track.get('durationMs'), 1000),
  87             'thumbnail': thumbnail,
  88             'track': track_title,
  89         }
  90
  91         def extract_artist(artist_list):
  92             if artist_list and isinstance(artist_list, list):
  93                 artists_names = [a['name'] for a in artist_list if a.get('name')]
  94                 if artists_names:
  95                     return ', '.join(artists_names)
  96
  97         albums = track.get('albums')
  98         if albums and isinstance(albums, list):
  99             album = albums[0]
 100             if isinstance(album, dict):
 101                 year = album.get('year')
 102                 track_info.update({
 103                     'album': album.get('title'),
 104                     'album_artist': extract_artist(album.get('artists')),
 105                     'release_year': compat_str(year) if year else None,
 106                 })
 107
 108         track_artist = extract_artist(track.get('artists'))
 109         if track_artist:
 110             track_info.update({
 111                 'artist': track_artist,
 112                 'title': '%s - %s' % (track_artist, track_title),
 113             })
 114         else:
 115             track_info['title'] = track_title
 116         return track_info
 117
 118     def _real_extract(self, url):
 119         mobj = re.match(self._VALID_URL, url)
 120         album_id, track_id = mobj.group('album_id'), mobj.group('id')
 121
 122         track = self._download_json(
 123             'http://music.yandex.ru/handlers/track.jsx?track=%s:%s' % (track_id, album_id),
 124             track_id, 'Downloading track JSON')['track']
 125
 126         return self._get_track_info(track)
 127
 128
 129 class YandexMusicPlaylistBaseIE(YandexMusicBaseIE):
 130     def _build_playlist(self, tracks):
 131         return [
 132             self.url_result(
 133                 'http://music.yandex.ru/album/%s/track/%s' % (track['albums'][0]['id'], track['id']))
 134             for track in tracks if track.get('albums') and isinstance(track.get('albums'), list)]
 135
 136
 137 class YandexMusicAlbumIE(YandexMusicPlaylistBaseIE):
 138     IE_NAME = 'yandexmusic:album'
 139     IE_DESC = 'Яндекс.Музыка - Альбом'
 140     _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<id>\d+)/?(\?|$)'
 141
 142     _TEST = {
 143         'url': 'http://music.yandex.ru/album/540508',
 144         'info_dict': {
 145             'id': '540508',
 146             'title': 'Carlo Ambrosio - Gypsy Soul (2009)',
 147         },
 148         'playlist_count': 50,
 149         'skip': 'Travis CI servers blocked by YandexMusic',
 150     }
 151
 152     def _real_extract(self, url):
 153         album_id = self._match_id(url)
 154
 155         album = self._download_json(
 156             'http://music.yandex.ru/handlers/album.jsx?album=%s' % album_id,
 157             album_id, 'Downloading album JSON')
 158
 159         entries = self._build_playlist(album['volumes'][0])
 160
 161         title = '%s - %s' % (album['artists'][0]['name'], album['title'])
 162         year = album.get('year')
 163         if year:
 164             title += ' (%s)' % year
 165
 166         return self.playlist_result(entries, compat_str(album['id']), title)
 167
 168
 169 class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE):
 170     IE_NAME = 'yandexmusic:playlist'
 171     IE_DESC = 'Яндекс.Музыка - Плейлист'
 172     _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/users/[^/]+/playlists/(?P<id>\d+)'
 173
 174     _TESTS = [{
 175         'url': 'http://music.yandex.ru/users/music.partners/playlists/1245',
 176         'info_dict': {
 177             'id': '1245',
 178             'title': 'Что слушают Enter Shikari',
 179             'description': 'md5:3b9f27b0efbe53f2ee1e844d07155cc9',
 180         },
 181         'playlist_count': 6,
 182         'skip': 'Travis CI servers blocked by YandexMusic',
 183     }, {
 184         # playlist exceeding the limit of 150 tracks shipped with webpage (see
 185         # https://github.com/rg3/youtube-dl/issues/6666)
 186         'url': 'https://music.yandex.ru/users/ya.playlist/playlists/1036',
 187         'info_dict': {
 188             'id': '1036',
 189             'title': 'Музыка 90-х',
 190         },
 191         'playlist_count': 310,
 192         'skip': 'Travis CI servers blocked by YandexMusic',
 193     }]
 194
 195     def _real_extract(self, url):
 196         playlist_id = self._match_id(url)
 197
 198         webpage = self._download_webpage(url, playlist_id)
 199
 200         mu = self._parse_json(
 201             self._search_regex(
 202                 r'var\s+Mu\s*=\s*({.+?});\s*</script>', webpage, 'player'),
 203             playlist_id)
 204
 205         playlist = mu['pageData']['playlist']
 206         tracks, track_ids = playlist['tracks'], playlist['trackIds']
 207
 208         # tracks dictionary shipped with webpage is limited to 150 tracks,
 209         # missing tracks should be retrieved manually.
 210         if len(tracks) < len(track_ids):
 211             present_track_ids = set([compat_str(track['id']) for track in tracks if track.get('id')])
 212             missing_track_ids = set(map(compat_str, track_ids)) - set(present_track_ids)
 213             request = sanitized_Request(
 214                 'https://music.yandex.ru/handlers/track-entries.jsx',
 215                 urlencode_postdata({
 216                     'entries': ','.join(missing_track_ids),
 217                     'lang': mu.get('settings', {}).get('lang', 'en'),
 218                     'external-domain': 'music.yandex.ru',
 219                     'overembed': 'false',
 220                     'sign': mu.get('authData', {}).get('user', {}).get('sign'),
 221                     'strict': 'true',
 222                 }))
 223             request.add_header('Referer', url)
 224             request.add_header('X-Requested-With', 'XMLHttpRequest')
 225
 226             missing_tracks = self._download_json(
 227                 request, playlist_id, 'Downloading missing tracks JSON', fatal=False)
 228             if missing_tracks:
 229                 tracks.extend(missing_tracks)
 230
 231         return self.playlist_result(
 232             self._build_playlist(tracks),
 233             compat_str(playlist_id),
 234             playlist['title'], playlist.get('description'))