]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/yandexmusic.py
Merge tag 'upstream/2015.11.10'
[youtubedl] / 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 thumbnail = None
50 cover_uri = track.get('albums', [{}])[0].get('coverUri')
51 if cover_uri:
52 thumbnail = cover_uri.replace('%%', 'orig')
53 if not thumbnail.startswith('http'):
54 thumbnail = 'http://' + thumbnail
55 return {
56 'id': track['id'],
57 'ext': 'mp3',
58 'url': self._get_track_url(track['storageDir'], track['id']),
59 'title': '%s - %s' % (track['artists'][0]['name'], track['title']),
60 'filesize': int_or_none(track.get('fileSize')),
61 'duration': float_or_none(track.get('durationMs'), 1000),
62 'thumbnail': thumbnail,
63 }
64
65 def _real_extract(self, url):
66 mobj = re.match(self._VALID_URL, url)
67 album_id, track_id = mobj.group('album_id'), mobj.group('id')
68
69 track = self._download_json(
70 'http://music.yandex.ru/handlers/track.jsx?track=%s:%s' % (track_id, album_id),
71 track_id, 'Downloading track JSON')['track']
72
73 return self._get_track_info(track)
74
75
76 class YandexMusicPlaylistBaseIE(InfoExtractor):
77 def _build_playlist(self, tracks):
78 return [
79 self.url_result(
80 'http://music.yandex.ru/album/%s/track/%s' % (track['albums'][0]['id'], track['id']))
81 for track in tracks if track.get('albums') and isinstance(track.get('albums'), list)]
82
83
84 class YandexMusicAlbumIE(YandexMusicPlaylistBaseIE):
85 IE_NAME = 'yandexmusic:album'
86 IE_DESC = 'Яндекс.Музыка - Альбом'
87 _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/album/(?P<id>\d+)/?(\?|$)'
88
89 _TEST = {
90 'url': 'http://music.yandex.ru/album/540508',
91 'info_dict': {
92 'id': '540508',
93 'title': 'Carlo Ambrosio - Gypsy Soul (2009)',
94 },
95 'playlist_count': 50,
96 }
97
98 def _real_extract(self, url):
99 album_id = self._match_id(url)
100
101 album = self._download_json(
102 'http://music.yandex.ru/handlers/album.jsx?album=%s' % album_id,
103 album_id, 'Downloading album JSON')
104
105 entries = self._build_playlist(album['volumes'][0])
106
107 title = '%s - %s' % (album['artists'][0]['name'], album['title'])
108 year = album.get('year')
109 if year:
110 title += ' (%s)' % year
111
112 return self.playlist_result(entries, compat_str(album['id']), title)
113
114
115 class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE):
116 IE_NAME = 'yandexmusic:playlist'
117 IE_DESC = 'Яндекс.Музыка - Плейлист'
118 _VALID_URL = r'https?://music\.yandex\.(?:ru|kz|ua|by)/users/[^/]+/playlists/(?P<id>\d+)'
119
120 _TESTS = [{
121 'url': 'http://music.yandex.ru/users/music.partners/playlists/1245',
122 'info_dict': {
123 'id': '1245',
124 'title': 'Что слушают Enter Shikari',
125 'description': 'md5:3b9f27b0efbe53f2ee1e844d07155cc9',
126 },
127 'playlist_count': 6,
128 }, {
129 # playlist exceeding the limit of 150 tracks shipped with webpage (see
130 # https://github.com/rg3/youtube-dl/issues/6666)
131 'url': 'https://music.yandex.ru/users/ya.playlist/playlists/1036',
132 'info_dict': {
133 'id': '1036',
134 'title': 'Музыка 90-х',
135 },
136 'playlist_count': 310,
137 }]
138
139 def _real_extract(self, url):
140 playlist_id = self._match_id(url)
141
142 webpage = self._download_webpage(url, playlist_id)
143
144 mu = self._parse_json(
145 self._search_regex(
146 r'var\s+Mu\s*=\s*({.+?});\s*</script>', webpage, 'player'),
147 playlist_id)
148
149 playlist = mu['pageData']['playlist']
150 tracks, track_ids = playlist['tracks'], playlist['trackIds']
151
152 # tracks dictionary shipped with webpage is limited to 150 tracks,
153 # missing tracks should be retrieved manually.
154 if len(tracks) < len(track_ids):
155 present_track_ids = set([compat_str(track['id']) for track in tracks if track.get('id')])
156 missing_track_ids = set(map(compat_str, track_ids)) - set(present_track_ids)
157 request = compat_urllib_request.Request(
158 'https://music.yandex.ru/handlers/track-entries.jsx',
159 compat_urllib_parse.urlencode({
160 'entries': ','.join(missing_track_ids),
161 'lang': mu.get('settings', {}).get('lang', 'en'),
162 'external-domain': 'music.yandex.ru',
163 'overembed': 'false',
164 'sign': mu.get('authData', {}).get('user', {}).get('sign'),
165 'strict': 'true',
166 }).encode('utf-8'))
167 request.add_header('Referer', url)
168 request.add_header('X-Requested-With', 'XMLHttpRequest')
169
170 missing_tracks = self._download_json(
171 request, playlist_id, 'Downloading missing tracks JSON', fatal=False)
172 if missing_tracks:
173 tracks.extend(missing_tracks)
174
175 return self.playlist_result(
176 self._build_playlist(tracks),
177 compat_str(playlist_id),
178 playlist['title'], playlist.get('description'))