_ 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 )
  14
  15
  16 class YandexMusicBaseIE(InfoExtractor):
  17     @staticmethod
  18     def _handle_error(response):
  19         if isinstance(response, dict):
  20             error = response.get('error')
  21             if error:
  22                 raise ExtractorError(error, expected=True)
  23
  24     def _download_webpage(self, *args, **kwargs):
  25         webpage = super(YandexMusicBaseIE, self)._download_webpage(*args, **kwargs)
  26         if 'Нам очень жаль, но&nbsp;запросы, поступившие с&nbsp;вашего IP-адреса, похожи на&nbsp;автоматические.' in webpage:
  27             raise ExtractorError(
  28                 'YandexMusic has considered youtube-dl requests automated and '
  29                 'asks you to solve a CAPTCHA. You can either wait for some '
  30                 'time until unblocked and optionally use --sleep-interval '
  31                 'in future or alternatively you can go to https://music.yandex.ru/ '
  32                 'solve CAPTCHA, then export cookies and pass cookie file to '
  33                 'youtube-dl with --cookies',
  34                 expected=True)
  35         return webpage
  36
  37     def _download_json(self, *args, **kwargs):
  38         response = super(YandexMusicBaseIE, self)._download_json(*args, **kwargs)
  39         self._handle_error(response)
  40         return response
  41
  42
  43 class YandexMusicTrackIE(YandexMusicBaseIE):
  44     IE_NAME = 'yandexmusic:track'
  45     IE_DESC = 'Яндекс.Музыка - Трек'
  46     _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<album_id>\d+)/track/(?P<id>\d+)'
  47
  48     _TEST = {
  49         'url': 'http://music.yandex.ru/album/540508/track/4878838',
  50         'md5': 'f496818aa2f60b6c0062980d2e00dc20',
  51         'info_dict': {
  52             'id': '4878838',
  53             'ext': 'mp3',
  54             'title': 'Carlo Ambrosio & Fabio Di Bari, Carlo Ambrosio - Gypsy Eyes 1',
  55             'filesize': 4628061,
  56             'duration': 193.04,
  57             'track': 'Gypsy Eyes 1',
  58             'album': 'Gypsy Soul',
  59             'album_artist': 'Carlo Ambrosio',
  60             'artist': 'Carlo Ambrosio & Fabio Di Bari, Carlo Ambrosio',
  61             'release_year': '2009',
  62         },
  63         'skip': 'Travis CI servers blocked by YandexMusic',
  64     }
  65
  66     def _get_track_url(self, storage_dir, track_id):
  67         data = self._download_json(
  68             'http://music.yandex.ru/api/v1.5/handlers/api-jsonp.jsx?action=getTrackSrc&p=download-info/%s'
  69             % storage_dir,
  70             track_id, 'Downloading track location JSON')
  71
  72         key = hashlib.md5(('XGRlBW9FXlekgbPrRHuSiA' + data['path'][1:] + data['s']).encode('utf-8')).hexdigest()
  73         storage = storage_dir.split('.')
  74
  75         return ('http://%s/get-mp3/%s/%s?track-id=%s&from=service-10-track&similarities-experiment=default'
  76                 % (data['host'], key, data['ts'] + data['path'], storage[1]))
  77
  78     def _get_track_info(self, track):
  79         thumbnail = None
  80         cover_uri = track.get('albums', [{}])[0].get('coverUri')
  81         if cover_uri:
  82             thumbnail = cover_uri.replace('%%', 'orig')
  83             if not thumbnail.startswith('http'):
  84                 thumbnail = 'http://' + thumbnail
  85
  86         track_title = track['title']
  87         track_info = {
  88             'id': track['id'],
  89             'ext': 'mp3',
  90             'url': self._get_track_url(track['storageDir'], track['id']),
  91             'filesize': int_or_none(track.get('fileSize')),
  92             'duration': float_or_none(track.get('durationMs'), 1000),
  93             'thumbnail': thumbnail,
  94             'track': track_title,
  95         }
  96
  97         def extract_artist(artist_list):
  98             if artist_list and isinstance(artist_list, list):
  99                 artists_names = [a['name'] for a in artist_list if a.get('name')]
 100                 if artists_names:
 101                     return ', '.join(artists_names)
 102
 103         albums = track.get('albums')
 104         if albums and isinstance(albums, list):
 105             album = albums[0]
 106             if isinstance(album, dict):
 107                 year = album.get('year')
 108                 track_info.update({
 109                     'album': album.get('title'),
 110                     'album_artist': extract_artist(album.get('artists')),
 111                     'release_year': compat_str(year) if year else None,
 112                 })
 113
 114         track_artist = extract_artist(track.get('artists'))
 115         if track_artist:
 116             track_info.update({
 117                 'artist': track_artist,
 118                 'title': '%s - %s' % (track_artist, track_title),
 119             })
 120         else:
 121             track_info['title'] = track_title
 122         return track_info
 123
 124     def _real_extract(self, url):
 125         mobj = re.match(self._VALID_URL, url)
 126         album_id, track_id = mobj.group('album_id'), mobj.group('id')
 127
 128         track = self._download_json(
 129             'http://music.yandex.ru/handlers/track.jsx?track=%s:%s' % (track_id, album_id),
 130             track_id, 'Downloading track JSON')['track']
 131
 132         return self._get_track_info(track)
 133
 134
 135 class YandexMusicPlaylistBaseIE(YandexMusicBaseIE):
 136     def _build_playlist(self, tracks):
 137         return [
 138             self.url_result(
 139                 'http://music.yandex.ru/album/%s/track/%s' % (track['albums'][0]['id'], track['id']))
 140             for track in tracks if track.get('albums') and isinstance(track.get('albums'), list)]
 141
 142
 143 class YandexMusicAlbumIE(YandexMusicPlaylistBaseIE):
 144     IE_NAME = 'yandexmusic:album'
 145     IE_DESC = 'Яндекс.Музыка - Альбом'
 146     _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<id>\d+)/?(\?|$)'
 147
 148     _TEST = {
 149         'url': 'http://music.yandex.ru/album/540508',
 150         'info_dict': {
 151             'id': '540508',
 152             'title': 'Carlo Ambrosio - Gypsy Soul (2009)',
 153         },
 154         'playlist_count': 50,
 155         'skip': 'Travis CI servers blocked by YandexMusic',
 156     }
 157
 158     def _real_extract(self, url):
 159         album_id = self._match_id(url)
 160
 161         album = self._download_json(
 162             'http://music.yandex.ru/handlers/album.jsx?album=%s' % album_id,
 163             album_id, 'Downloading album JSON')
 164
 165         entries = self._build_playlist(album['volumes'][0])
 166
 167         title = '%s - %s' % (album['artists'][0]['name'], album['title'])
 168         year = album.get('year')
 169         if year:
 170             title += ' (%s)' % year
 171
 172         return self.playlist_result(entries, compat_str(album['id']), title)
 173
 174
 175 class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE):
 176     IE_NAME = 'yandexmusic:playlist'
 177     IE_DESC = 'Яндекс.Музыка - Плейлист'
 178     _VALID_URL = r'https?://music\.yandex\.(?P<tld>ru|kz|ua|by)/users/(?P<user>[^/]+)/playlists/(?P<id>\d+)'
 179
 180     _TESTS = [{
 181         'url': 'http://music.yandex.ru/users/music.partners/playlists/1245',
 182         'info_dict': {
 183             'id': '1245',
 184             'title': 'Что слушают Enter Shikari',
 185             'description': 'md5:3b9f27b0efbe53f2ee1e844d07155cc9',
 186         },
 187         'playlist_count': 6,
 188         'skip': 'Travis CI servers blocked by YandexMusic',
 189     }, {
 190         # playlist exceeding the limit of 150 tracks shipped with webpage (see
 191         # https://github.com/rg3/youtube-dl/issues/6666)
 192         'url': 'https://music.yandex.ru/users/ya.playlist/playlists/1036',
 193         'info_dict': {
 194             'id': '1036',
 195             'title': 'Музыка 90-х',
 196         },
 197         'playlist_mincount': 300,
 198         'skip': 'Travis CI servers blocked by YandexMusic',
 199     }]
 200
 201     def _real_extract(self, url):
 202         mobj = re.match(self._VALID_URL, url)
 203         tld = mobj.group('tld')
 204         user = mobj.group('user')
 205         playlist_id = mobj.group('id')
 206
 207         playlist = self._download_json(
 208             'https://music.yandex.%s/handlers/playlist.jsx' % tld,
 209             playlist_id, 'Downloading missing tracks JSON',
 210             fatal=False,
 211             headers={
 212                 'Referer': url,
 213                 'X-Requested-With': 'XMLHttpRequest',
 214                 'X-Retpath-Y': url,
 215             },
 216             query={
 217                 'owner': user,
 218                 'kinds': playlist_id,
 219                 'light': 'true',
 220                 'lang': tld,
 221                 'external-domain': 'music.yandex.%s' % tld,
 222                 'overembed': 'false',
 223             })['playlist']
 224
 225         tracks, track_ids = playlist['tracks'], map(compat_str, playlist['trackIds'])
 226
 227         # tracks dictionary shipped with playlist.jsx API is limited to 150 tracks,
 228         # missing tracks should be retrieved manually.
 229         if len(tracks) < len(track_ids):
 230             present_track_ids = set([
 231                 compat_str(track['id'])
 232                 for track in tracks if track.get('id')])
 233             missing_track_ids = [
 234                 track_id for track_id in track_ids
 235                 if track_id not in present_track_ids]
 236             missing_tracks = self._download_json(
 237                 'https://music.yandex.%s/handlers/track-entries.jsx' % tld,
 238                 playlist_id, 'Downloading missing tracks JSON',
 239                 fatal=False,
 240                 headers={
 241                     'Referer': url,
 242                     'X-Requested-With': 'XMLHttpRequest',
 243                 },
 244                 query={
 245                     'entries': ','.join(missing_track_ids),
 246                     'lang': tld,
 247                     'external-domain': 'music.yandex.%s' % tld,
 248                     'overembed': 'false',
 249                     'strict': 'true',
 250                 })
 251             if missing_tracks:
 252                 tracks.extend(missing_tracks)
 253
 254         return self.playlist_result(
 255             self._build_playlist(tracks),
 256             compat_str(playlist_id),
 257             playlist.get('title'), playlist.get('description'))