yt-dlp/youtube_dl/extractor/yandexmusic.py

264 lines
9.8 KiB
Python
Raw Normal View History

# coding: utf-8
2015-03-09 13:06:49 +00:00
from __future__ import unicode_literals
import re
import hashlib
from .common import InfoExtractor
from ..compat import compat_str
from ..utils import (
ExtractorError,
int_or_none,
float_or_none,
)
2015-03-09 13:06:49 +00:00
class YandexMusicBaseIE(InfoExtractor):
@staticmethod
def _handle_error(response):
2016-04-28 15:37:34 +00:00
if isinstance(response, dict):
error = response.get('error')
if error:
raise ExtractorError(error, expected=True)
if response.get('type') == 'captcha' or 'captcha' in response:
YandexMusicBaseIE._raise_captcha()
@staticmethod
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 https://music.yandex.ru/ '
'solve CAPTCHA, then export cookies and pass cookie file to '
'youtube-dl with --cookies',
expected=True)
def _download_webpage(self, *args, **kwargs):
webpage = super(YandexMusicBaseIE, self)._download_webpage(*args, **kwargs)
if 'Нам очень жаль, но запросы, поступившие с вашего IP-адреса, похожи на автоматические.' in webpage:
self._raise_captcha()
return webpage
def _download_json(self, *args, **kwargs):
response = super(YandexMusicBaseIE, self)._download_json(*args, **kwargs)
self._handle_error(response)
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+)'
_TEST = {
'url': 'http://music.yandex.ru/album/540508/track/4878838',
'md5': 'f496818aa2f60b6c0062980d2e00dc20',
'info_dict': {
'id': '4878838',
'ext': 'mp3',
'title': 'Carlo Ambrosio & Fabio Di Bari, Carlo Ambrosio - 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, Carlo Ambrosio',
'release_year': '2009',
},
'skip': 'Travis CI servers blocked by YandexMusic',
}
2015-03-09 13:06:49 +00:00
def _get_track_url(self, storage_dir, track_id):
data = self._download_json(
'http://music.yandex.ru/api/v1.5/handlers/api-jsonp.jsx?action=getTrackSrc&p=download-info/%s'
% storage_dir,
track_id, 'Downloading track location JSON')
2015-03-09 13:06:49 +00:00
key = hashlib.md5(('XGRlBW9FXlekgbPrRHuSiA' + data['path'][1:] + data['s']).encode('utf-8')).hexdigest()
2015-03-09 13:06:49 +00:00
storage = storage_dir.split('.')
return ('http://%s/get-mp3/%s/%s?track-id=%s&from=service-10-track&similarities-experiment=default'
% (data['host'], key, data['ts'] + data['path'], storage[1]))
2015-03-09 13:06:49 +00:00
def _get_track_info(self, track):
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_title = track['title']
track_info = {
'id': track['id'],
'ext': 'mp3',
'url': self._get_track_url(track['storageDir'], track['id']),
'filesize': int_or_none(track.get('fileSize')),
'duration': float_or_none(track.get('durationMs'), 1000),
'thumbnail': thumbnail,
'track': track_title,
}
2015-03-09 13:06:49 +00:00
def extract_artist(artist_list):
if artist_list and isinstance(artist_list, list):
artists_names = [a['name'] 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')
track_info.update({
'album': album.get('title'),
'album_artist': extract_artist(album.get('artists')),
'release_year': compat_str(year) if year else None,
})
track_artist = extract_artist(track.get('artists'))
if track_artist:
track_info.update({
'artist': track_artist,
'title': '%s - %s' % (track_artist, track_title),
})
else:
track_info['title'] = track_title
return track_info
2015-03-09 13:06:49 +00:00
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
album_id, track_id = mobj.group('album_id'), mobj.group('id')
2015-03-09 13:06:49 +00:00
track = self._download_json(
'http://music.yandex.ru/handlers/track.jsx?track=%s:%s' % (track_id, album_id),
track_id, 'Downloading track JSON')['track']
2015-03-09 13:06:49 +00:00
return self._get_track_info(track)
2015-03-09 13:06:49 +00:00
class YandexMusicPlaylistBaseIE(YandexMusicBaseIE):
def _build_playlist(self, tracks):
return [
self.url_result(
'http://music.yandex.ru/album/%s/track/%s' % (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'
2015-03-14 07:56:04 +00:00
_VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<id>\d+)/?(\?|$)'
2015-03-09 13:06:49 +00:00
_TEST = {
'url': 'http://music.yandex.ru/album/540508',
'info_dict': {
'id': '540508',
'title': 'Carlo Ambrosio - Gypsy Soul (2009)',
},
'playlist_count': 50,
'skip': 'Travis CI servers blocked by YandexMusic',
}
2015-03-09 13:06:49 +00:00
def _real_extract(self, url):
album_id = self._match_id(url)
2015-03-09 13:06:49 +00:00
album = self._download_json(
'http://music.yandex.ru/handlers/album.jsx?album=%s' % album_id,
album_id, 'Downloading album JSON')
2015-03-09 13:06:49 +00:00
entries = self._build_playlist(album['volumes'][0])
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)
2015-03-09 13:06:49 +00:00
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': 'http://music.yandex.ru/users/music.partners/playlists/1245',
2015-03-09 13:06:49 +00:00
'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
# https://github.com/rg3/youtube-dl/issues/6666)
'url': 'https://music.yandex.ru/users/ya.playlist/playlists/1036',
'info_dict': {
'id': '1036',
'title': 'Музыка 90-х',
},
2016-05-04 17:18:48 +00:00
'playlist_mincount': 300,
'skip': 'Travis CI servers blocked by YandexMusic',
}]
2015-03-09 13:06:49 +00:00
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
tld = mobj.group('tld')
user = mobj.group('user')
playlist_id = mobj.group('id')
playlist = self._download_json(
'https://music.yandex.%s/handlers/playlist.jsx' % tld,
playlist_id, 'Downloading missing tracks JSON',
fatal=False,
headers={
'Referer': url,
'X-Requested-With': 'XMLHttpRequest',
'X-Retpath-Y': url,
},
query={
'owner': user,
'kinds': playlist_id,
'light': 'true',
'lang': tld,
'external-domain': 'music.yandex.%s' % tld,
'overembed': 'false',
})['playlist']
tracks, track_ids = playlist['tracks'], map(compat_str, 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([
compat_str(track['id'])
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]
2016-05-04 16:33:29 +00:00
missing_tracks = self._download_json(
'https://music.yandex.%s/handlers/track-entries.jsx' % tld,
playlist_id, 'Downloading missing tracks JSON',
fatal=False,
headers={
'Referer': url,
'X-Requested-With': 'XMLHttpRequest',
},
query={
'entries': ','.join(missing_track_ids),
'lang': tld,
'external-domain': 'music.yandex.%s' % tld,
'overembed': 'false',
'strict': 'true',
2016-05-04 16:33:29 +00:00
})
if missing_tracks:
tracks.extend(missing_tracks)
return self.playlist_result(
self._build_playlist(tracks),
compat_str(playlist_id),
playlist.get('title'), playlist.get('description'))