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