You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

313 lines
12 KiB

# coding: utf-8
from __future__ import unicode_literals
import re
import hashlib
from .common import InfoExtractor
from ..compat import compat_str
from ..utils import (
class YandexMusicBaseIE(InfoExtractor):
def _handle_error(response):
if isinstance(response, dict):
error = response.get('error')
if error:
raise ExtractorError(error, expected=True)
if response.get('type') == 'captcha' or 'captcha' in response:
def _raise_captcha():
raise ExtractorError(
'YandexMusic has considered youtube-dl requests automated and '
'asks you to solve a CAPTCHA. You can either wait for some '
'time until unblocked and optionally use --sleep-interval '
'in future or alternatively you can go to '
'solve CAPTCHA, then export cookies and pass cookie file to '
'youtube-dl with --cookies',
def _download_webpage_handle(self, *args, **kwargs):
webpage = super(YandexMusicBaseIE, self)._download_webpage_handle(*args, **kwargs)
if 'Нам очень жаль, но запросы, поступившие с вашего IP-адреса, похожи на автоматические.' in webpage:
return webpage
def _download_json(self, *args, **kwargs):
response = super(YandexMusicBaseIE, self)._download_json(*args, **kwargs)
return response
class YandexMusicTrackIE(YandexMusicBaseIE):
IE_NAME = 'yandexmusic:track'
IE_DESC = 'Яндекс.Музыка - Трек'
_VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<album_id>\d+)/track/(?P<id>\d+)'
_TESTS = [{
'url': '',
'md5': 'f496818aa2f60b6c0062980d2e00dc20',
'info_dict': {
'id': '4878838',
'ext': 'mp3',
'title': 'Carlo Ambrosio & Fabio Di Bari - Gypsy Eyes 1',
'filesize': 4628061,
'duration': 193.04,
'track': 'Gypsy Eyes 1',
'album': 'Gypsy Soul',
'album_artist': 'Carlo Ambrosio',
'artist': 'Carlo Ambrosio & Fabio Di Bari',
'release_year': 2009,
'skip': 'Travis CI servers blocked by YandexMusic',
}, {
# multiple disks
'url': '',
'md5': 'ebe7b4e2ac7ac03fe11c19727ca6153e',
'info_dict': {
'id': '705105',
'ext': 'mp3',
'title': 'Hooverphonic - Sometimes',
'filesize': 5743386,
'duration': 239.27,
'track': 'Sometimes',
'album': 'The Best of Hooverphonic',
'album_artist': 'Hooverphonic',
'artist': 'Hooverphonic',
'release_year': 2016,
'genre': 'pop',
'disc_number': 2,
'track_number': 9,
'skip': 'Travis CI servers blocked by YandexMusic',
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
album_id, track_id ='album_id'),'id')
track = self._download_json(
'' % (track_id, album_id),
track_id, 'Downloading track JSON')['track']
track_title = track['title']
download_data = self._download_json(
'' % (track_id, album_id),
track_id, 'Downloading track location url JSON',
headers={'X-Retpath-Y': url})
fd_data = self._download_json(
download_data['src'], track_id,
'Downloading track location JSON',
query={'format': 'json'})
key = hashlib.md5(('XGRlBW9FXlekgbPrRHuSiA' + fd_data['path'][1:] + fd_data['s']).encode('utf-8')).hexdigest()
storage = track['storageDir'].split('.')
f_url = 'http://%s/get-mp3/%s/%s?track-id=%s ' % (fd_data['host'], key, fd_data['ts'] + fd_data['path'], storage[1])
thumbnail = None
cover_uri = track.get('albums', [{}])[0].get('coverUri')
if cover_uri:
thumbnail = cover_uri.replace('%%', 'orig')
if not thumbnail.startswith('http'):
thumbnail = 'http://' + thumbnail
track_info = {
'id': track_id,
'ext': 'mp3',
'url': f_url,
'filesize': int_or_none(track.get('fileSize')),
'duration': float_or_none(track.get('durationMs'), 1000),
'thumbnail': thumbnail,
'track': track_title,
'acodec': download_data.get('codec'),
'abr': int_or_none(download_data.get('bitrate')),
def extract_artist_name(artist):
decomposed = artist.get('decomposed')
if not isinstance(decomposed, list):
return artist['name']
parts = [artist['name']]
for element in decomposed:
if isinstance(element, dict) and element.get('name'):
elif isinstance(element, compat_str):
return ''.join(parts)
def extract_artist(artist_list):
if artist_list and isinstance(artist_list, list):
artists_names = [extract_artist_name(a) for a in artist_list if a.get('name')]
if artists_names:
return ', '.join(artists_names)
albums = track.get('albums')
if albums and isinstance(albums, list):
album = albums[0]
if isinstance(album, dict):
year = album.get('year')
disc_number = int_or_none(try_get(
album, lambda x: x['trackPosition']['volume']))
track_number = int_or_none(try_get(
album, lambda x: x['trackPosition']['index']))
'album': album.get('title'),
'album_artist': extract_artist(album.get('artists')),
'release_year': int_or_none(year),
'genre': album.get('genre'),
'disc_number': disc_number,
'track_number': track_number,
track_artist = extract_artist(track.get('artists'))
if track_artist:
'artist': track_artist,
'title': '%s - %s' % (track_artist, track_title),
track_info['title'] = track_title
return track_info
class YandexMusicPlaylistBaseIE(YandexMusicBaseIE):
def _build_playlist(self, tracks):
return [
'' % (track['albums'][0]['id'], track['id']))
for track in tracks if track.get('albums') and isinstance(track.get('albums'), list)]
class YandexMusicAlbumIE(YandexMusicPlaylistBaseIE):
IE_NAME = 'yandexmusic:album'
IE_DESC = 'Яндекс.Музыка - Альбом'
_VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<id>\d+)/?(\?|$)'
_TESTS = [{
'url': '',
'info_dict': {
'id': '540508',
'title': 'Carlo Ambrosio - Gypsy Soul (2009)',
'playlist_count': 50,
'skip': 'Travis CI servers blocked by YandexMusic',
}, {
'url': '',
'info_dict': {
'id': '3840501',
'title': 'Hooverphonic - The Best of Hooverphonic (2016)',
'playlist_count': 33,
'skip': 'Travis CI servers blocked by YandexMusic',
def _real_extract(self, url):
album_id = self._match_id(url)
album = self._download_json(
'' % album_id,
album_id, 'Downloading album JSON')
entries = self._build_playlist([track for volume in album['volumes'] for track in volume])
title = '%s - %s' % (album['artists'][0]['name'], album['title'])
year = album.get('year')
if year:
title += ' (%s)' % year
return self.playlist_result(entries, compat_str(album['id']), title)
class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE):
IE_NAME = 'yandexmusic:playlist'
IE_DESC = 'Яндекс.Музыка - Плейлист'
_VALID_URL = r'https?://music\.yandex\.(?P<tld>ru|kz|ua|by)/users/(?P<user>[^/]+)/playlists/(?P<id>\d+)'
_TESTS = [{
'url': '',
'info_dict': {
'id': '1245',
'title': 'Что слушают Enter Shikari',
'description': 'md5:3b9f27b0efbe53f2ee1e844d07155cc9',
'playlist_count': 6,
'skip': 'Travis CI servers blocked by YandexMusic',
}, {
# playlist exceeding the limit of 150 tracks shipped with webpage (see
'url': '',
'info_dict': {
'id': '1036',
'title': 'Музыка 90-х',
'playlist_mincount': 300,
'skip': 'Travis CI servers blocked by YandexMusic',
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
tld ='tld')
user ='user')
playlist_id ='id')
playlist = self._download_json(
'' % tld,
playlist_id, 'Downloading missing tracks JSON',
'Referer': url,
'X-Requested-With': 'XMLHttpRequest',
'X-Retpath-Y': url,
'owner': user,
'kinds': playlist_id,
'light': 'true',
'lang': tld,
'external-domain': '' % tld,
'overembed': 'false',
tracks = playlist['tracks']
track_ids = [compat_str(track_id) for track_id in playlist['trackIds']]
# tracks dictionary shipped with playlist.jsx API is limited to 150 tracks,
# missing tracks should be retrieved manually.
if len(tracks) < len(track_ids):
present_track_ids = set([
for track in tracks if track.get('id')])
missing_track_ids = [
track_id for track_id in track_ids
if track_id not in present_track_ids]
missing_tracks = self._download_json(
'' % tld,
playlist_id, 'Downloading missing tracks JSON',
'Referer': url,
'X-Requested-With': 'XMLHttpRequest',
'entries': ','.join(missing_track_ids),
'lang': tld,
'external-domain': '' % tld,
'overembed': 'false',
'strict': 'true',
if missing_tracks:
return self.playlist_result(
playlist.get('title'), playlist.get('description'))