_ 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 )
  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     _TESTS = [{
  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 - 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',
  67             'release_year': 2009,
  68         },
  69         'skip': 'Travis CI servers blocked by YandexMusic',
  70     }, {
  71         # multiple disks
  72         'url': 'http://music.yandex.ru/album/3840501/track/705105',
  73         'md5': 'ebe7b4e2ac7ac03fe11c19727ca6153e',
  74         'info_dict': {
  75             'id': '705105',
  76             'ext': 'mp3',
  77             'title': 'Hooverphonic - Sometimes',
  78             'filesize': 5743386,
  79             'duration': 239.27,
  80             'track': 'Sometimes',
  81             'album': 'The Best of Hooverphonic',
  82             'album_artist': 'Hooverphonic',
  83             'artist': 'Hooverphonic',
  84             'release_year': 2016,
  85             'genre': 'pop',
  86             'disc_number': 2,
  87             'track_number': 9,
  88         },
  89         'skip': 'Travis CI servers blocked by YandexMusic',
  90     }]
  91
  92     def _real_extract(self, url):
  93         mobj = re.match(self._VALID_URL, url)
  94         album_id, track_id = mobj.group('album_id'), mobj.group('id')
  95
  96         track = self._download_json(
  97             'http://music.yandex.ru/handlers/track.jsx?track=%s:%s' % (track_id, album_id),
  98             track_id, 'Downloading track JSON')['track']
  99         track_title = track['title']
 100
 101         download_data = self._download_json(
 102             'https://music.yandex.ru/api/v2.1/handlers/track/%s:%s/web-album_track-track-track-main/download/m' % (track_id, album_id),
 103             track_id, 'Downloading track location url JSON',
 104             headers={'X-Retpath-Y': url})
 105
 106         fd_data = self._download_json(
 107             download_data['src'], track_id,
 108             'Downloading track location JSON',
 109             query={'format': 'json'})
 110         key = hashlib.md5(('XGRlBW9FXlekgbPrRHuSiA' + fd_data['path'][1:] + fd_data['s']).encode('utf-8')).hexdigest()
 111         storage = track['storageDir'].split('.')
 112         f_url = 'http://%s/get-mp3/%s/%s?track-id=%s ' % (fd_data['host'], key, fd_data['ts'] + fd_data['path'], storage[1])
 113
 114         thumbnail = None
 115         cover_uri = track.get('albums', [{}])[0].get('coverUri')
 116         if cover_uri:
 117             thumbnail = cover_uri.replace('%%', 'orig')
 118             if not thumbnail.startswith('http'):
 119                 thumbnail = 'http://' + thumbnail
 120
 121         track_info = {
 122             'id': track_id,
 123             'ext': 'mp3',
 124             'url': f_url,
 125             'filesize': int_or_none(track.get('fileSize')),
 126             'duration': float_or_none(track.get('durationMs'), 1000),
 127             'thumbnail': thumbnail,
 128             'track': track_title,
 129             'acodec': download_data.get('codec'),
 130             'abr': int_or_none(download_data.get('bitrate')),
 131         }
 132
 133         def extract_artist_name(artist):
 134             decomposed = artist.get('decomposed')
 135             if not isinstance(decomposed, list):
 136                 return artist['name']
 137             parts = [artist['name']]
 138             for element in decomposed:
 139                 if isinstance(element, dict) and element.get('name'):
 140                     parts.append(element['name'])
 141                 elif isinstance(element, compat_str):
 142                     parts.append(element)
 143             return ''.join(parts)
 144
 145         def extract_artist(artist_list):
 146             if artist_list and isinstance(artist_list, list):
 147                 artists_names = [extract_artist_name(a) for a in artist_list if a.get('name')]
 148                 if artists_names:
 149                     return ', '.join(artists_names)
 150
 151         albums = track.get('albums')
 152         if albums and isinstance(albums, list):
 153             album = albums[0]
 154             if isinstance(album, dict):
 155                 year = album.get('year')
 156                 disc_number = int_or_none(try_get(
 157                     album, lambda x: x['trackPosition']['volume']))
 158                 track_number = int_or_none(try_get(
 159                     album, lambda x: x['trackPosition']['index']))
 160                 track_info.update({
 161                     'album': album.get('title'),
 162                     'album_artist': extract_artist(album.get('artists')),
 163                     'release_year': int_or_none(year),
 164                     'genre': album.get('genre'),
 165                     'disc_number': disc_number,
 166                     'track_number': track_number,
 167                 })
 168
 169         track_artist = extract_artist(track.get('artists'))
 170         if track_artist:
 171             track_info.update({
 172                 'artist': track_artist,
 173                 'title': '%s - %s' % (track_artist, track_title),
 174             })
 175         else:
 176             track_info['title'] = track_title
 177
 178         return track_info
 179
 180
 181 class YandexMusicPlaylistBaseIE(YandexMusicBaseIE):
 182     def _build_playlist(self, tracks):
 183         return [
 184             self.url_result(
 185                 'http://music.yandex.ru/album/%s/track/%s' % (track['albums'][0]['id'], track['id']))
 186             for track in tracks if track.get('albums') and isinstance(track.get('albums'), list)]
 187
 188
 189 class YandexMusicAlbumIE(YandexMusicPlaylistBaseIE):
 190     IE_NAME = 'yandexmusic:album'
 191     IE_DESC = 'Яндекс.Музыка - Альбом'
 192     _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<id>\d+)/?(\?|$)'
 193
 194     _TESTS = [{
 195         'url': 'http://music.yandex.ru/album/540508',
 196         'info_dict': {
 197             'id': '540508',
 198             'title': 'Carlo Ambrosio - Gypsy Soul (2009)',
 199         },
 200         'playlist_count': 50,
 201         'skip': 'Travis CI servers blocked by YandexMusic',
 202     }, {
 203         'url': 'https://music.yandex.ru/album/3840501',
 204         'info_dict': {
 205             'id': '3840501',
 206             'title': 'Hooverphonic - The Best of Hooverphonic (2016)',
 207         },
 208         'playlist_count': 33,
 209         'skip': 'Travis CI servers blocked by YandexMusic',
 210     }]
 211
 212     def _real_extract(self, url):
 213         album_id = self._match_id(url)
 214
 215         album = self._download_json(
 216             'http://music.yandex.ru/handlers/album.jsx?album=%s' % album_id,
 217             album_id, 'Downloading album JSON')
 218
 219         entries = self._build_playlist([track for volume in album['volumes'] for track in volume])
 220
 221         title = '%s - %s' % (album['artists'][0]['name'], album['title'])
 222         year = album.get('year')
 223         if year:
 224             title += ' (%s)' % year
 225
 226         return self.playlist_result(entries, compat_str(album['id']), title)
 227
 228
 229 class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE):
 230     IE_NAME = 'yandexmusic:playlist'
 231     IE_DESC = 'Яндекс.Музыка - Плейлист'
 232     _VALID_URL = r'https?://music\.yandex\.(?P<tld>ru|kz|ua|by)/users/(?P<user>[^/]+)/playlists/(?P<id>\d+)'
 233
 234     _TESTS = [{
 235         'url': 'http://music.yandex.ru/users/music.partners/playlists/1245',
 236         'info_dict': {
 237             'id': '1245',
 238             'title': 'Что слушают Enter Shikari',
 239             'description': 'md5:3b9f27b0efbe53f2ee1e844d07155cc9',
 240         },
 241         'playlist_count': 6,
 242         'skip': 'Travis CI servers blocked by YandexMusic',
 243     }, {
 244         # playlist exceeding the limit of 150 tracks shipped with webpage (see
 245         # https://github.com/ytdl-org/youtube-dl/issues/6666)
 246         'url': 'https://music.yandex.ru/users/ya.playlist/playlists/1036',
 247         'info_dict': {
 248             'id': '1036',
 249             'title': 'Музыка 90-х',
 250         },
 251         'playlist_mincount': 300,
 252         'skip': 'Travis CI servers blocked by YandexMusic',
 253     }]
 254
 255     def _real_extract(self, url):
 256         mobj = re.match(self._VALID_URL, url)
 257         tld = mobj.group('tld')
 258         user = mobj.group('user')
 259         playlist_id = mobj.group('id')
 260
 261         playlist = self._download_json(
 262             'https://music.yandex.%s/handlers/playlist.jsx' % tld,
 263             playlist_id, 'Downloading missing tracks JSON',
 264             fatal=False,
 265             headers={
 266                 'Referer': url,
 267                 'X-Requested-With': 'XMLHttpRequest',
 268                 'X-Retpath-Y': url,
 269             },
 270             query={
 271                 'owner': user,
 272                 'kinds': playlist_id,
 273                 'light': 'true',
 274                 'lang': tld,
 275                 'external-domain': 'music.yandex.%s' % tld,
 276                 'overembed': 'false',
 277             })['playlist']
 278
 279         tracks = playlist['tracks']
 280         track_ids = [compat_str(track_id) for track_id in playlist['trackIds']]
 281
 282         # tracks dictionary shipped with playlist.jsx API is limited to 150 tracks,
 283         # missing tracks should be retrieved manually.
 284         if len(tracks) < len(track_ids):
 285             present_track_ids = set([
 286                 compat_str(track['id'])
 287                 for track in tracks if track.get('id')])
 288             missing_track_ids = [
 289                 track_id for track_id in track_ids
 290                 if track_id not in present_track_ids]
 291             missing_tracks = self._download_json(
 292                 'https://music.yandex.%s/handlers/track-entries.jsx' % tld,
 293                 playlist_id, 'Downloading missing tracks JSON',
 294                 fatal=False,
 295                 headers={
 296                     'Referer': url,
 297                     'X-Requested-With': 'XMLHttpRequest',
 298                 },
 299                 query={
 300                     'entries': ','.join(missing_track_ids),
 301                     'lang': tld,
 302                     'external-domain': 'music.yandex.%s' % tld,
 303                     'overembed': 'false',
 304                     'strict': 'true',
 305                 })
 306             if missing_tracks:
 307                 tracks.extend(missing_tracks)
 308
 309         return self.playlist_result(
 310             self._build_playlist(tracks),
 311             compat_str(playlist_id),
 312             playlist.get('title'), playlist.get('description'))