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