Raphaël G. Git Repositories - youtubedl/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             if response.get('type') == 'captcha' or 'captcha' in response:
  24                 YandexMusicBaseIE._raise_captcha()
  25
  26     @staticmethod
  27     def _raise_captcha():
  28         raise ExtractorError(
  29             'YandexMusic has considered youtube-dl requests automated and '
  30             'asks you to solve a CAPTCHA. You can either wait for some '
  31             'time until unblocked and optionally use --sleep-interval '
  32             'in future or alternatively you can go to https://music.yandex.ru/ '
  33             'solve CAPTCHA, then export cookies and pass cookie file to '
  34             'youtube-dl with --cookies',
  35             expected=True)
  36
  37     def _download_webpage(self, *args, **kwargs):
  38         webpage = super(YandexMusicBaseIE, self)._download_webpage(*args, **kwargs)
  39         if 'Нам очень жаль, но&nbsp;запросы, поступившие с&nbsp;вашего IP-адреса, похожи на&nbsp;автоматические.' in webpage:
  40             self._raise_captcha()
  41         return webpage
  42
  43     def _download_json(self, *args, **kwargs):
  44         response = super(YandexMusicBaseIE, self)._download_json(*args, **kwargs)
  45         self._handle_error(response)
  46         return response
  47
  48
  49 class YandexMusicTrackIE(YandexMusicBaseIE):
  50     IE_NAME = 'yandexmusic:track'
  51     IE_DESC = 'Яндекс.Музыка - Трек'
  52     _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<album_id>\d+)/track/(?P<id>\d+)'
  53
  54     _TEST = {
  55         'url': 'http://music.yandex.ru/album/540508/track/4878838',
  56         'md5': 'f496818aa2f60b6c0062980d2e00dc20',
  57         'info_dict': {
  58             'id': '4878838',
  59             'ext': 'mp3',
  60             'title': 'Carlo Ambrosio & Fabio Di Bari, Carlo Ambrosio - Gypsy Eyes 1',
  61             'filesize': 4628061,
  62             'duration': 193.04,
  63             'track': 'Gypsy Eyes 1',
  64             'album': 'Gypsy Soul',
  65             'album_artist': 'Carlo Ambrosio',
  66             'artist': 'Carlo Ambrosio & Fabio Di Bari, Carlo Ambrosio',
  67             'release_year': '2009',
  68         },
  69         'skip': 'Travis CI servers blocked by YandexMusic',
  70     }
  71
  72     def _get_track_url(self, storage_dir, track_id):
  73         data = self._download_json(
  74             'http://music.yandex.ru/api/v1.5/handlers/api-jsonp.jsx?action=getTrackSrc&p=download-info/%s'
  75             % storage_dir,
  76             track_id, 'Downloading track location JSON')
  77
  78         key = hashlib.md5(('XGRlBW9FXlekgbPrRHuSiA' + data['path'][1:] + data['s']).encode('utf-8')).hexdigest()
  79         storage = storage_dir.split('.')
  80
  81         return ('http://%s/get-mp3/%s/%s?track-id=%s&from=service-10-track&similarities-experiment=default'
  82                 % (data['host'], key, data['ts'] + data['path'], storage[1]))
  83
  84     def _get_track_info(self, track):
  85         thumbnail = None
  86         cover_uri = track.get('albums', [{}])[0].get('coverUri')
  87         if cover_uri:
  88             thumbnail = cover_uri.replace('%%', 'orig')
  89             if not thumbnail.startswith('http'):
  90                 thumbnail = 'http://' + thumbnail
  91
  92         track_title = track['title']
  93         track_info = {
  94             'id': track['id'],
  95             'ext': 'mp3',
  96             'url': self._get_track_url(track['storageDir'], track['id']),
  97             'filesize': int_or_none(track.get('fileSize')),
  98             'duration': float_or_none(track.get('durationMs'), 1000),
  99             'thumbnail': thumbnail,
 100             'track': track_title,
 101         }
 102
 103         def extract_artist(artist_list):
 104             if artist_list and isinstance(artist_list, list):
 105                 artists_names = [a['name'] for a in artist_list if a.get('name')]
 106                 if artists_names:
 107                     return ', '.join(artists_names)
 108
 109         albums = track.get('albums')
 110         if albums and isinstance(albums, list):
 111             album = albums[0]
 112             if isinstance(album, dict):
 113                 year = album.get('year')
 114                 track_info.update({
 115                     'album': album.get('title'),
 116                     'album_artist': extract_artist(album.get('artists')),
 117                     'release_year': compat_str(year) if year else None,
 118                 })
 119
 120         track_artist = extract_artist(track.get('artists'))
 121         if track_artist:
 122             track_info.update({
 123                 'artist': track_artist,
 124                 'title': '%s - %s' % (track_artist, track_title),
 125             })
 126         else:
 127             track_info['title'] = track_title
 128         return track_info
 129
 130     def _real_extract(self, url):
 131         mobj = re.match(self._VALID_URL, url)
 132         album_id, track_id = mobj.group('album_id'), mobj.group('id')
 133
 134         track = self._download_json(
 135             'http://music.yandex.ru/handlers/track.jsx?track=%s:%s' % (track_id, album_id),
 136             track_id, 'Downloading track JSON')['track']
 137
 138         return self._get_track_info(track)
 139
 140
 141 class YandexMusicPlaylistBaseIE(YandexMusicBaseIE):
 142     def _build_playlist(self, tracks):
 143         return [
 144             self.url_result(
 145                 'http://music.yandex.ru/album/%s/track/%s' % (track['albums'][0]['id'], track['id']))
 146             for track in tracks if track.get('albums') and isinstance(track.get('albums'), list)]
 147
 148
 149 class YandexMusicAlbumIE(YandexMusicPlaylistBaseIE):
 150     IE_NAME = 'yandexmusic:album'
 151     IE_DESC = 'Яндекс.Музыка - Альбом'
 152     _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<id>\d+)/?(\?|$)'
 153
 154     _TEST = {
 155         'url': 'http://music.yandex.ru/album/540508',
 156         'info_dict': {
 157             'id': '540508',
 158             'title': 'Carlo Ambrosio - Gypsy Soul (2009)',
 159         },
 160         'playlist_count': 50,
 161         'skip': 'Travis CI servers blocked by YandexMusic',
 162     }
 163
 164     def _real_extract(self, url):
 165         album_id = self._match_id(url)
 166
 167         album = self._download_json(
 168             'http://music.yandex.ru/handlers/album.jsx?album=%s' % album_id,
 169             album_id, 'Downloading album JSON')
 170
 171         entries = self._build_playlist(album['volumes'][0])
 172
 173         title = '%s - %s' % (album['artists'][0]['name'], album['title'])
 174         year = album.get('year')
 175         if year:
 176             title += ' (%s)' % year
 177
 178         return self.playlist_result(entries, compat_str(album['id']), title)
 179
 180
 181 class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE):
 182     IE_NAME = 'yandexmusic:playlist'
 183     IE_DESC = 'Яндекс.Музыка - Плейлист'
 184     _VALID_URL = r'https?://music\.yandex\.(?P<tld>ru|kz|ua|by)/users/(?P<user>[^/]+)/playlists/(?P<id>\d+)'
 185
 186     _TESTS = [{
 187         'url': 'http://music.yandex.ru/users/music.partners/playlists/1245',
 188         'info_dict': {
 189             'id': '1245',
 190             'title': 'Что слушают Enter Shikari',
 191             'description': 'md5:3b9f27b0efbe53f2ee1e844d07155cc9',
 192         },
 193         'playlist_count': 6,
 194         'skip': 'Travis CI servers blocked by YandexMusic',
 195     }, {
 196         # playlist exceeding the limit of 150 tracks shipped with webpage (see
 197         # https://github.com/rg3/youtube-dl/issues/6666)
 198         'url': 'https://music.yandex.ru/users/ya.playlist/playlists/1036',
 199         'info_dict': {
 200             'id': '1036',
 201             'title': 'Музыка 90-х',
 202         },
 203         'playlist_mincount': 300,
 204         'skip': 'Travis CI servers blocked by YandexMusic',
 205     }]
 206
 207     def _real_extract(self, url):
 208         mobj = re.match(self._VALID_URL, url)
 209         tld = mobj.group('tld')
 210         user = mobj.group('user')
 211         playlist_id = mobj.group('id')
 212
 213         playlist = self._download_json(
 214             'https://music.yandex.%s/handlers/playlist.jsx' % tld,
 215             playlist_id, 'Downloading missing tracks JSON',
 216             fatal=False,
 217             headers={
 218                 'Referer': url,
 219                 'X-Requested-With': 'XMLHttpRequest',
 220                 'X-Retpath-Y': url,
 221             },
 222             query={
 223                 'owner': user,
 224                 'kinds': playlist_id,
 225                 'light': 'true',
 226                 'lang': tld,
 227                 'external-domain': 'music.yandex.%s' % tld,
 228                 'overembed': 'false',
 229             })['playlist']
 230
 231         tracks, track_ids = playlist['tracks'], map(compat_str, playlist['trackIds'])
 232
 233         # tracks dictionary shipped with playlist.jsx API is limited to 150 tracks,
 234         # missing tracks should be retrieved manually.
 235         if len(tracks) < len(track_ids):
 236             present_track_ids = set([
 237                 compat_str(track['id'])
 238                 for track in tracks if track.get('id')])
 239             missing_track_ids = [
 240                 track_id for track_id in track_ids
 241                 if track_id not in present_track_ids]
 242             missing_tracks = self._download_json(
 243                 'https://music.yandex.%s/handlers/track-entries.jsx' % tld,
 244                 playlist_id, 'Downloading missing tracks JSON',
 245                 fatal=False,
 246                 headers={
 247                     'Referer': url,
 248                     'X-Requested-With': 'XMLHttpRequest',
 249                 },
 250                 query={
 251                     'entries': ','.join(missing_track_ids),
 252                     'lang': tld,
 253                     'external-domain': 'music.yandex.%s' % tld,
 254                     'overembed': 'false',
 255                     'strict': 'true',
 256                 })
 257             if missing_tracks:
 258                 tracks.extend(missing_tracks)
 259
 260         return self.playlist_result(
 261             self._build_playlist(tracks),
 262             compat_str(playlist_id),
 263             playlist.get('title'), playlist.get('description'))