_ 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 (
   9     compat_str,
  10     compat_urllib_parse,
  11     compat_urllib_request,
  12 )
  13 from ..utils import (
  14     int_or_none,
  15     float_or_none,
  16 )
  17
  18
  19 class YandexMusicTrackIE(InfoExtractor):
  20     IE_NAME = 'yandexmusic:track'
  21     IE_DESC = 'Яндекс.Музыка - Трек'
  22     _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<album_id>\d+)/track/(?P<id>\d+)'
  23
  24     _TEST = {
  25         'url': 'http://music.yandex.ru/album/540508/track/4878838',
  26         'md5': 'f496818aa2f60b6c0062980d2e00dc20',
  27         'info_dict': {
  28             'id': '4878838',
  29             'ext': 'mp3',
  30             'title': 'Carlo Ambrosio - Gypsy Eyes 1',
  31             'filesize': 4628061,
  32             'duration': 193.04,
  33         }
  34     }
  35
  36     def _get_track_url(self, storage_dir, track_id):
  37         data = self._download_json(
  38             'http://music.yandex.ru/api/v1.5/handlers/api-jsonp.jsx?action=getTrackSrc&p=download-info/%s'
  39             % storage_dir,
  40             track_id, 'Downloading track location JSON')
  41
  42         key = hashlib.md5(('XGRlBW9FXlekgbPrRHuSiA' + data['path'][1:] + data['s']).encode('utf-8')).hexdigest()
  43         storage = storage_dir.split('.')
  44
  45         return ('http://%s/get-mp3/%s/%s?track-id=%s&from=service-10-track&similarities-experiment=default'
  46                 % (data['host'], key, data['ts'] + data['path'], storage[1]))
  47
  48     def _get_track_info(self, track):
  49         return {
  50             'id': track['id'],
  51             'ext': 'mp3',
  52             'url': self._get_track_url(track['storageDir'], track['id']),
  53             'title': '%s - %s' % (track['artists'][0]['name'], track['title']),
  54             'filesize': int_or_none(track.get('fileSize')),
  55             'duration': float_or_none(track.get('durationMs'), 1000),
  56         }
  57
  58     def _real_extract(self, url):
  59         mobj = re.match(self._VALID_URL, url)
  60         album_id, track_id = mobj.group('album_id'), mobj.group('id')
  61
  62         track = self._download_json(
  63             'http://music.yandex.ru/handlers/track.jsx?track=%s:%s' % (track_id, album_id),
  64             track_id, 'Downloading track JSON')['track']
  65
  66         return self._get_track_info(track)
  67
  68
  69 class YandexMusicPlaylistBaseIE(InfoExtractor):
  70     def _build_playlist(self, tracks):
  71         return [
  72             self.url_result(
  73                 'http://music.yandex.ru/album/%s/track/%s' % (track['albums'][0]['id'], track['id']))
  74             for track in tracks if track.get('albums') and isinstance(track.get('albums'), list)]
  75
  76
  77 class YandexMusicAlbumIE(YandexMusicPlaylistBaseIE):
  78     IE_NAME = 'yandexmusic:album'
  79     IE_DESC = 'Яндекс.Музыка - Альбом'
  80     _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<id>\d+)/?(\?|$)'
  81
  82     _TEST = {
  83         'url': 'http://music.yandex.ru/album/540508',
  84         'info_dict': {
  85             'id': '540508',
  86             'title': 'Carlo Ambrosio - Gypsy Soul (2009)',
  87         },
  88         'playlist_count': 50,
  89     }
  90
  91     def _real_extract(self, url):
  92         album_id = self._match_id(url)
  93
  94         album = self._download_json(
  95             'http://music.yandex.ru/handlers/album.jsx?album=%s' % album_id,
  96             album_id, 'Downloading album JSON')
  97
  98         entries = self._build_playlist(album['volumes'][0])
  99
 100         title = '%s - %s' % (album['artists'][0]['name'], album['title'])
 101         year = album.get('year')
 102         if year:
 103             title += ' (%s)' % year
 104
 105         return self.playlist_result(entries, compat_str(album['id']), title)
 106
 107
 108 class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE):
 109     IE_NAME = 'yandexmusic:playlist'
 110     IE_DESC = 'Яндекс.Музыка - Плейлист'
 111     _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/users/[^/]+/playlists/(?P<id>\d+)'
 112
 113     _TESTS = [{
 114         'url': 'http://music.yandex.ru/users/music.partners/playlists/1245',
 115         'info_dict': {
 116             'id': '1245',
 117             'title': 'Что слушают Enter Shikari',
 118             'description': 'md5:3b9f27b0efbe53f2ee1e844d07155cc9',
 119         },
 120         'playlist_count': 6,
 121     }, {
 122         # playlist exceeding the limit of 150 tracks shipped with webpage (see
 123         # https://github.com/rg3/youtube-dl/issues/6666)
 124         'url': 'https://music.yandex.ru/users/ya.playlist/playlists/1036',
 125         'info_dict': {
 126             'id': '1036',
 127             'title': 'Музыка 90-х',
 128         },
 129         'playlist_count': 310,
 130     }]
 131
 132     def _real_extract(self, url):
 133         playlist_id = self._match_id(url)
 134
 135         webpage = self._download_webpage(url, playlist_id)
 136
 137         mu = self._parse_json(
 138             self._search_regex(
 139                 r'var\s+Mu\s*=\s*({.+?});\s*</script>', webpage, 'player'),
 140             playlist_id)
 141
 142         playlist = mu['pageData']['playlist']
 143         tracks, track_ids = playlist['tracks'], playlist['trackIds']
 144
 145         # tracks dictionary shipped with webpage is limited to 150 tracks,
 146         # missing tracks should be retrieved manually.
 147         if len(tracks) < len(track_ids):
 148             present_track_ids = set([compat_str(track['id']) for track in tracks if track.get('id')])
 149             missing_track_ids = set(map(compat_str, track_ids)) - set(present_track_ids)
 150             request = compat_urllib_request.Request(
 151                 'https://music.yandex.ru/handlers/track-entries.jsx',
 152                 compat_urllib_parse.urlencode({
 153                     'entries': ','.join(missing_track_ids),
 154                     'lang': mu.get('settings', {}).get('lang', 'en'),
 155                     'external-domain': 'music.yandex.ru',
 156                     'overembed': 'false',
 157                     'sign': mu.get('authData', {}).get('user', {}).get('sign'),
 158                     'strict': 'true',
 159                 }).encode('utf-8'))
 160             request.add_header('Referer', url)
 161             request.add_header('X-Requested-With', 'XMLHttpRequest')
 162
 163             missing_tracks = self._download_json(
 164                 request, playlist_id, 'Downloading missing tracks JSON', fatal=False)
 165             if missing_tracks:
 166                 tracks.extend(missing_tracks)
 167
 168         return self.playlist_result(
 169             self._build_playlist(tracks),
 170             compat_str(playlist_id),
 171             playlist['title'], playlist.get('description'))