[yandexmusic] Refactor and add support for artist's tracks and albums (closes #11887, closes #22284)

youtube_dl/extractor/extractors.py

@@ -1478,6 +1478,8 @@ from .yandexmusic import (
     YandexMusicTrackIE,
     YandexMusicAlbumIE,
     YandexMusicPlaylistIE,
+    YandexMusicArtistTracksIE,
+    YandexMusicArtistAlbumsIE,
 )
 from .yandexvideo import YandexVideoIE
 from .yapfiles import YapFilesIE

youtube_dl/extractor/yandexmusic.py

@@ -46,57 +46,69 @@ class YandexMusicBaseIE(InfoExtractor):
         self._handle_error(response)
         return response
 
+    def _call_api(self, ep, tld, url, item_id, note, query):
+        return self._download_json(
+            'https://music.yandex.%s/handlers/%s.jsx' % (tld, ep),
+            item_id, note,
+            fatal=False,
+            headers={
+                'Referer': url,
+                'X-Requested-With': 'XMLHttpRequest',
+                'X-Retpath-Y': url,
+            },
+            query=query)
+
 
 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+)'
+    _VALID_URL = r'https?://music\.yandex\.(?P<tld>ru|kz|ua|by)/album/(?P<album_id>\d+)/track/(?P<id>\d+)'
 
     _TESTS = [{
         'url': 'http://music.yandex.ru/album/540508/track/4878838',
-        'md5': 'f496818aa2f60b6c0062980d2e00dc20',
+        'md5': 'dec8b661f12027ceaba33318787fff76',
         'info_dict': {
             'id': '4878838',
             'ext': 'mp3',
-            'title': 'Carlo Ambrosio & Fabio Di Bari - Gypsy Eyes 1',
+            'title': 'md5:c63e19341fdbe84e43425a30bc777856',
-            'filesize': 4628061,
+            'filesize': int,
             'duration': 193.04,
-            'track': 'Gypsy Eyes 1',
+            'track': 'md5:210508c6ffdfd67a493a6c378f22c3ff',
-            'album': 'Gypsy Soul',
+            'album': 'md5:cd04fb13c4efeafdfa0a6a6aca36d01a',
-            'album_artist': 'Carlo Ambrosio',
+            'album_artist': 'md5:5f54c35462c07952df33d97cfb5fc200',
-            'artist': 'Carlo Ambrosio & Fabio Di Bari',
+            'artist': 'md5:e6fd86621825f14dc0b25db3acd68160',
             'release_year': 2009,
         },
-        'skip': 'Travis CI servers blocked by YandexMusic',
+        # 'skip': 'Travis CI servers blocked by YandexMusic',
     }, {
         # multiple disks
         'url': 'http://music.yandex.ru/album/3840501/track/705105',
-        'md5': 'ebe7b4e2ac7ac03fe11c19727ca6153e',
+        'md5': '82a54e9e787301dd45aba093cf6e58c0',
         'info_dict': {
             'id': '705105',
             'ext': 'mp3',
-            'title': 'Hooverphonic - Sometimes',
+            'title': 'md5:f86d4a9188279860a83000277024c1a6',
-            'filesize': 5743386,
+            'filesize': int,
             'duration': 239.27,
-            'track': 'Sometimes',
+            'track': 'md5:40f887f0666ba1aa10b835aca44807d1',
-            'album': 'The Best of Hooverphonic',
+            'album': 'md5:624f5224b14f5c88a8e812fd7fbf1873',
-            'album_artist': 'Hooverphonic',
+            'album_artist': 'md5:dd35f2af4e8927100cbe6f5e62e1fb12',
-            'artist': 'Hooverphonic',
+            'artist': 'md5:dd35f2af4e8927100cbe6f5e62e1fb12',
             'release_year': 2016,
             'genre': 'pop',
             'disc_number': 2,
             'track_number': 9,
         },
-        'skip': 'Travis CI servers blocked by YandexMusic',
+        # 'skip': 'Travis CI servers blocked by YandexMusic',
     }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        album_id, track_id = mobj.group('album_id'), mobj.group('id')
+        tld, album_id, track_id = mobj.group('tld'), mobj.group('album_id'), mobj.group('id')
 
-        track = self._download_json(
+        track = self._call_api(
-            'http://music.yandex.ru/handlers/track.jsx?track=%s:%s' % (track_id, album_id),
+            'track', tld, url, track_id, 'Downloading track JSON',
-            track_id, 'Downloading track JSON')['track']
+            {'track': '%s:%s' % (track_id, album_id)})['track']
         track_title = track['title']
 
         download_data = self._download_json(
@@ -109,8 +121,7 @@ class YandexMusicTrackIE(YandexMusicBaseIE):
             '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'], track['id'])
-        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')
@@ -180,42 +191,85 @@ class YandexMusicTrackIE(YandexMusicBaseIE):
 
 
 class YandexMusicPlaylistBaseIE(YandexMusicBaseIE):
+    def _extract_tracks(self, source, item_id, url, tld):
+        tracks = source['tracks']
+        track_ids = [compat_str(track_id) for track_id in source['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]
+            missing_tracks = self._call_api(
+                'track-entries', tld, url, item_id,
+                'Downloading missing tracks JSON', {
+                    'entries': ','.join(missing_track_ids),
+                    'lang': tld,
+                    'external-domain': 'music.yandex.%s' % tld,
+                    'overembed': 'false',
+                    'strict': 'true',
+                })
+            if missing_tracks:
+                tracks.extend(missing_tracks)
+
+        return tracks
+
     def _build_playlist(self, tracks):
-        return [
+        entries = []
-            self.url_result(
+        for track in tracks:
-                'http://music.yandex.ru/album/%s/track/%s' % (track['albums'][0]['id'], track['id']))
+            track_id = track.get('id') or track.get('realId')
-            for track in tracks if track.get('albums') and isinstance(track.get('albums'), list)]
+            if not track_id:
+                continue
+            albums = track.get('albums')
+            if not albums or not isinstance(albums, list):
+                continue
+            album = albums[0]
+            if not isinstance(album, dict):
+                continue
+            album_id = album.get('id')
+            if not album_id:
+                continue
+            entries.append(self.url_result(
+                'http://music.yandex.ru/album/%s/track/%s' % (album_id, track_id),
+                ie=YandexMusicTrackIE.ie_key(), video_id=track_id))
+        return entries
 
 
 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+)/?(\?|$)'
+    _VALID_URL = r'https?://music\.yandex\.(?P<tld>ru|kz|ua|by)/album/(?P<id>\d+)/?(\?|$)'
 
     _TESTS = [{
         'url': 'http://music.yandex.ru/album/540508',
         'info_dict': {
             'id': '540508',
-            'title': 'Carlo Ambrosio - Gypsy Soul (2009)',
+            'title': 'md5:7ed1c3567f28d14be9f61179116f5571',
         },
         'playlist_count': 50,
-        'skip': 'Travis CI servers blocked by YandexMusic',
+        # 'skip': 'Travis CI servers blocked by YandexMusic',
     }, {
         'url': 'https://music.yandex.ru/album/3840501',
         'info_dict': {
             'id': '3840501',
-            'title': 'Hooverphonic - The Best of Hooverphonic (2016)',
+            'title': 'md5:36733472cdaa7dcb1fd9473f7da8e50f',
         },
         'playlist_count': 33,
-        'skip': 'Travis CI servers blocked by YandexMusic',
+        # 'skip': 'Travis CI servers blocked by YandexMusic',
     }]
 
     def _real_extract(self, url):
-        album_id = self._match_id(url)
+        mobj = re.match(self._VALID_URL, url)
+        tld = mobj.group('tld')
+        album_id = mobj.group('id')
 
-        album = self._download_json(
+        album = self._call_api(
-            'http://music.yandex.ru/handlers/album.jsx?album=%s' % album_id,
+            'album', tld, url, album_id, 'Downloading album JSON',
-            album_id, 'Downloading album JSON')
+            {'album': album_id})
 
         entries = self._build_playlist([track for volume in album['volumes'] for track in volume])
 
@@ -236,21 +290,24 @@ class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE):
         'url': 'http://music.yandex.ru/users/music.partners/playlists/1245',
         'info_dict': {
             'id': '1245',
-            'title': 'Что слушают Enter Shikari',
+            'title': 'md5:841559b3fe2b998eca88d0d2e22a3097',
             'description': 'md5:3b9f27b0efbe53f2ee1e844d07155cc9',
         },
-        'playlist_count': 6,
+        'playlist_count': 5,
-        'skip': 'Travis CI servers blocked by YandexMusic',
+        # 'skip': 'Travis CI servers blocked by YandexMusic',
     }, {
-        # playlist exceeding the limit of 150 tracks shipped with webpage (see
-        # https://github.com/ytdl-org/youtube-dl/issues/6666)
         'url': 'https://music.yandex.ru/users/ya.playlist/playlists/1036',
+        'only_matching': True,
+    }, {
+        # playlist exceeding the limit of 150 tracks (see
+        # https://github.com/ytdl-org/youtube-dl/issues/6666)
+        'url': 'https://music.yandex.ru/users/mesiaz/playlists/1364',
         'info_dict': {
-            'id': '1036',
+            'id': '1364',
-            'title': 'Музыка 90-х',
+            'title': 'md5:b3b400f997d3f878a13ae0699653f7db',
         },
-        'playlist_mincount': 300,
+        'playlist_mincount': 437,
-        'skip': 'Travis CI servers blocked by YandexMusic',
+        # 'skip': 'Travis CI servers blocked by YandexMusic',
     }]
 
     def _real_extract(self, url):
@@ -259,16 +316,8 @@ class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE):
         user = mobj.group('user')
         playlist_id = mobj.group('id')
 
-        playlist = self._download_json(
+        playlist = self._call_api(
-            'https://music.yandex.%s/handlers/playlist.jsx' % tld,
+            'playlist', tld, url, playlist_id, 'Downloading playlist JSON', {
-            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',
@@ -277,37 +326,103 @@ class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE):
                 'overembed': 'false',
             })['playlist']
 
-        tracks = playlist['tracks']
+        tracks = self._extract_tracks(playlist, playlist_id, url, tld)
-        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([
-                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]
-            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',
-                })
-            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'))
+
+
+class YandexMusicArtistBaseIE(YandexMusicPlaylistBaseIE):
+    def _call_artist(self, tld, url, artist_id):
+        return self._call_api(
+            'artist', tld, url, artist_id,
+            'Downloading artist %s JSON' % self._ARTIST_WHAT, {
+                'artist': artist_id,
+                'what': self._ARTIST_WHAT,
+                'sort': self._ARTIST_SORT or '',
+                'dir': '',
+                'period': '',
+                'lang': tld,
+                'external-domain': 'music.yandex.%s' % tld,
+                'overembed': 'false',
+            })
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        tld = mobj.group('tld')
+        artist_id = mobj.group('id')
+        data = self._call_artist(tld, url, artist_id)
+        tracks = self._extract_tracks(data, artist_id, url, tld)
+        title = try_get(data, lambda x: x['artist']['name'], compat_str)
+        return self.playlist_result(
+            self._build_playlist(tracks), artist_id, title)
+
+
+class YandexMusicArtistTracksIE(YandexMusicArtistBaseIE):
+    IE_NAME = 'yandexmusic:artist:tracks'
+    IE_DESC = 'Яндекс.Музыка - Артист - Треки'
+    _VALID_URL = r'https?://music\.yandex\.(?P<tld>ru|kz|ua|by)/artist/(?P<id>\d+)/tracks'
+
+    _TESTS = [{
+        'url': 'https://music.yandex.ru/artist/617526/tracks',
+        'info_dict': {
+            'id': '617526',
+            'title': 'md5:131aef29d45fd5a965ca613e708c040b',
+        },
+        'playlist_count': 507,
+        # 'skip': 'Travis CI servers blocked by YandexMusic',
+    }]
+
+    _ARTIST_SORT = ''
+    _ARTIST_WHAT = 'tracks'
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        tld = mobj.group('tld')
+        artist_id = mobj.group('id')
+        data = self._call_artist(tld, url, artist_id)
+        tracks = self._extract_tracks(data, artist_id, url, tld)
+        artist = try_get(data, lambda x: x['artist']['name'], compat_str)
+        title = '%s - %s' % (artist or artist_id, 'Треки')
+        return self.playlist_result(
+            self._build_playlist(tracks), artist_id, title)
+
+
+class YandexMusicArtistAlbumsIE(YandexMusicArtistBaseIE):
+    IE_NAME = 'yandexmusic:artist:albums'
+    IE_DESC = 'Яндекс.Музыка - Артист - Альбомы'
+    _VALID_URL = r'https?://music\.yandex\.(?P<tld>ru|kz|ua|by)/artist/(?P<id>\d+)/albums'
+
+    _TESTS = [{
+        'url': 'https://music.yandex.ru/artist/617526/albums',
+        'info_dict': {
+            'id': '617526',
+            'title': 'md5:55dc58d5c85699b7fb41ee926700236c',
+        },
+        'playlist_count': 8,
+        # 'skip': 'Travis CI servers blocked by YandexMusic',
+    }]
+
+    _ARTIST_SORT = 'year'
+    _ARTIST_WHAT = 'albums'
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        tld = mobj.group('tld')
+        artist_id = mobj.group('id')
+        data = self._call_artist(tld, url, artist_id)
+        entries = []
+        for album in data['albums']:
+            if not isinstance(album, dict):
+                continue
+            album_id = album.get('id')
+            if not album_id:
+                continue
+            entries.append(self.url_result(
+                'http://music.yandex.ru/album/%s' % album_id,
+                ie=YandexMusicAlbumIE.ie_key(), video_id=album_id))
+        artist = try_get(data, lambda x: x['artist']['name'], compat_str)
+        title = '%s - %s' % (artist or artist_id, 'Альбомы')
+        return self.playlist_result(entries, artist_id, title)