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_handle(self, *args, **kwargs):
  38         webpage = super(YandexMusicBaseIE, self)._download_webpage_handle(*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, Carlo Ambrosio & Fabio Di Bari - 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, Carlo Ambrosio & Fabio Di Bari',
  67             'release_year': 2009,
  68         },
  69         'skip': 'Travis CI servers blocked by YandexMusic',
  70     }
  71
  72     def _real_extract(self, url):
  73         mobj = re.match(self._VALID_URL, url)
  74         album_id, track_id = mobj.group('album_id'), mobj.group('id')
  75
  76         track = self._download_json(
  77             'http://music.yandex.ru/handlers/track.jsx?track=%s:%s' % (track_id, album_id),
  78             track_id, 'Downloading track JSON')['track']
  79         track_title = track['title']
  80
  81         download_data = self._download_json(
  82             'https://music.yandex.ru/api/v2.1/handlers/track/%s:%s/web-album_track-track-track-main/download/m' % (track_id, album_id),
  83             track_id, 'Downloading track location url JSON',
  84             headers={'X-Retpath-Y': url})
  85
  86         fd_data = self._download_json(
  87             download_data['src'], track_id,
  88             'Downloading track location JSON',
  89             query={'format': 'json'})
  90         key = hashlib.md5(('XGRlBW9FXlekgbPrRHuSiA' + fd_data['path'][1:] + fd_data['s']).encode('utf-8')).hexdigest()
  91         storage = track['storageDir'].split('.')
  92         f_url = 'http://%s/get-mp3/%s/%s?track-id=%s ' % (fd_data['host'], key, fd_data['ts'] + fd_data['path'], storage[1])
  93
  94         thumbnail = None
  95         cover_uri = track.get('albums', [{}])[0].get('coverUri')
  96         if cover_uri:
  97             thumbnail = cover_uri.replace('%%', 'orig')
  98             if not thumbnail.startswith('http'):
  99                 thumbnail = 'http://' + thumbnail
 100
 101         track_info = {
 102             'id': track_id,
 103             'ext': 'mp3',
 104             'url': f_url,
 105             'filesize': int_or_none(track.get('fileSize')),
 106             'duration': float_or_none(track.get('durationMs'), 1000),
 107             'thumbnail': thumbnail,
 108             'track': track_title,
 109             'acodec': download_data.get('codec'),
 110             'abr': int_or_none(download_data.get('bitrate')),
 111         }
 112
 113         def extract_artist(artist_list):
 114             if artist_list and isinstance(artist_list, list):
 115                 artists_names = [a['name'] for a in artist_list if a.get('name')]
 116                 if artists_names:
 117                     return ', '.join(artists_names)
 118
 119         albums = track.get('albums')
 120         if albums and isinstance(albums, list):
 121             album = albums[0]
 122             if isinstance(album, dict):
 123                 year = album.get('year')
 124                 track_info.update({
 125                     'album': album.get('title'),
 126                     'album_artist': extract_artist(album.get('artists')),
 127                     'release_year': int_or_none(year),
 128                 })
 129
 130         track_artist = extract_artist(track.get('artists'))
 131         if track_artist:
 132             track_info.update({
 133                 'artist': track_artist,
 134                 'title': '%s - %s' % (track_artist, track_title),
 135             })
 136         else:
 137             track_info['title'] = track_title
 138
 139         return track_info
 140
 141
 142 class YandexMusicPlaylistBaseIE(YandexMusicBaseIE):
 143     def _build_playlist(self, tracks):
 144         return [
 145             self.url_result(
 146                 'http://music.yandex.ru/album/%s/track/%s' % (track['albums'][0]['id'], track['id']))
 147             for track in tracks if track.get('albums') and isinstance(track.get('albums'), list)]
 148
 149
 150 class YandexMusicAlbumIE(YandexMusicPlaylistBaseIE):
 151     IE_NAME = 'yandexmusic:album'
 152     IE_DESC = 'Яндекс.Музыка - Альбом'
 153     _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<id>\d+)/?(\?|$)'
 154
 155     _TEST = {
 156         'url': 'http://music.yandex.ru/album/540508',
 157         'info_dict': {
 158             'id': '540508',
 159             'title': 'Carlo Ambrosio - Gypsy Soul (2009)',
 160         },
 161         'playlist_count': 50,
 162         'skip': 'Travis CI servers blocked by YandexMusic',
 163     }
 164
 165     def _real_extract(self, url):
 166         album_id = self._match_id(url)
 167
 168         album = self._download_json(
 169             'http://music.yandex.ru/handlers/album.jsx?album=%s' % album_id,
 170             album_id, 'Downloading album JSON')
 171
 172         entries = self._build_playlist(album['volumes'][0])
 173
 174         title = '%s - %s' % (album['artists'][0]['name'], album['title'])
 175         year = album.get('year')
 176         if year:
 177             title += ' (%s)' % year
 178
 179         return self.playlist_result(entries, compat_str(album['id']), title)
 180
 181
 182 class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE):
 183     IE_NAME = 'yandexmusic:playlist'
 184     IE_DESC = 'Яндекс.Музыка - Плейлист'
 185     _VALID_URL = r'https?://music\.yandex\.(?P<tld>ru|kz|ua|by)/users/(?P<user>[^/]+)/playlists/(?P<id>\d+)'
 186
 187     _TESTS = [{
 188         'url': 'http://music.yandex.ru/users/music.partners/playlists/1245',
 189         'info_dict': {
 190             'id': '1245',
 191             'title': 'Что слушают Enter Shikari',
 192             'description': 'md5:3b9f27b0efbe53f2ee1e844d07155cc9',
 193         },
 194         'playlist_count': 6,
 195         'skip': 'Travis CI servers blocked by YandexMusic',
 196     }, {
 197         # playlist exceeding the limit of 150 tracks shipped with webpage (see
 198         # https://github.com/ytdl-org/youtube-dl/issues/6666)
 199         'url': 'https://music.yandex.ru/users/ya.playlist/playlists/1036',
 200         'info_dict': {
 201             'id': '1036',
 202             'title': 'Музыка 90-х',
 203         },
 204         'playlist_mincount': 300,
 205         'skip': 'Travis CI servers blocked by YandexMusic',
 206     }]
 207
 208     def _real_extract(self, url):
 209         mobj = re.match(self._VALID_URL, url)
 210         tld = mobj.group('tld')
 211         user = mobj.group('user')
 212         playlist_id = mobj.group('id')
 213
 214         playlist = self._download_json(
 215             'https://music.yandex.%s/handlers/playlist.jsx' % tld,
 216             playlist_id, 'Downloading missing tracks JSON',
 217             fatal=False,
 218             headers={
 219                 'Referer': url,
 220                 'X-Requested-With': 'XMLHttpRequest',
 221                 'X-Retpath-Y': url,
 222             },
 223             query={
 224                 'owner': user,
 225                 'kinds': playlist_id,
 226                 'light': 'true',
 227                 'lang': tld,
 228                 'external-domain': 'music.yandex.%s' % tld,
 229                 'overembed': 'false',
 230             })['playlist']
 231
 232         tracks = playlist['tracks']
 233         track_ids = [compat_str(track_id) for track_id in playlist['trackIds']]
 234
 235         # tracks dictionary shipped with playlist.jsx API is limited to 150 tracks,
 236         # missing tracks should be retrieved manually.
 237         if len(tracks) < len(track_ids):
 238             present_track_ids = set([
 239                 compat_str(track['id'])
 240                 for track in tracks if track.get('id')])
 241             missing_track_ids = [
 242                 track_id for track_id in track_ids
 243                 if track_id not in present_track_ids]
 244             missing_tracks = self._download_json(
 245                 'https://music.yandex.%s/handlers/track-entries.jsx' % tld,
 246                 playlist_id, 'Downloading missing tracks JSON',
 247                 fatal=False,
 248                 headers={
 249                     'Referer': url,
 250                     'X-Requested-With': 'XMLHttpRequest',
 251                 },
 252                 query={
 253                     'entries': ','.join(missing_track_ids),
 254                     'lang': tld,
 255                     'external-domain': 'music.yandex.%s' % tld,
 256                     'overembed': 'false',
 257                     'strict': 'true',
 258                 })
 259             if missing_tracks:
 260                 tracks.extend(missing_tracks)
 261
 262         return self.playlist_result(
 263             self._build_playlist(tracks),
 264             compat_str(playlist_id),
 265             playlist.get('title'), playlist.get('description'))