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