yt_dlp/extractor/vk.py

@@ -13,7 +13,6 @@ from ..utils import (
     ExtractorError,
     UserNotLive,
     clean_html,
-    extract_attributes,
     get_element_by_class,
     get_element_html_by_class,
     get_element_html_by_id,
@@ -815,7 +814,7 @@ class VKMusicBaseIE(VKBaseIE):
             # ['Main Artist', 'Feat. Artist']
             'artists': traverse_obj(
                 (*meta[17], *meta[18]) if len_ >= 18 else None,
-                (..., 'name')) or [artist],
+                (..., 'name'), default=[artist]),
 
             'duration': int_or_none(meta[5]) if len_ >= 5 else None,
             'thumbnails': [{'url': thumbnail}] if thumbnail else [],
@@ -1084,7 +1083,6 @@ class VKMusicPlaylistIE(VKMusicBaseIE):
             playlist_id)
         del hash_in_url
 
-        # to remove big scripts and other elements not used by parser
         html = get_element_html_by_class('AudioPlaylistSnippet', webpage)
         del webpage
 
@@ -1107,29 +1105,29 @@ class VKMusicPlaylistIE(VKMusicBaseIE):
             entries.append(self.url_result(
                 audio_url, VKMusicTrackIE, track_id, title, **info))
 
-        header = get_element_html_by_class('AudioPlaylistSnippet__header', html)
+        title = self._html_search_regex(
+            r'class="[^"]*AudioPlaylistSnippet__title--main[^"]*"[^>]*>([^<]+)',
+            html, 'playlist title', fatal=False, group=1)
 
-        title = clean_html(get_element_by_class('AudioPlaylistSnippet__title', header))
+        artist = self._html_search_regex(
-        artist = clean_html(get_element_by_class('AudioPlaylistSnippet__author', header))
+            r'class="[^"]*AudioPlaylistSnippet__author[^"]*"[^>]*>\s*<a(?:\s[^>]*)?>([^<]+)',
+            html, 'playlist author', fatal=False, group=1)
 
-        info_text = clean_html(get_element_by_class('AudioPlaylistSnippet__info', header))
+        description = clean_html(get_element_by_class(
-        info_sep = info_text.find('·')
+            'AudioPlaylistSnippet__description', html))
+        # description = self._html_search_regex(
+        #     r'div\s[^>]*class="[^"]*AudioPlaylistSnippet__description[^"]*">??????',
+        #     html, 'playlist description', fatal=False, group=1)
+
+        genre, year = self._html_search_regex(
+            r'class="[^"]*AudioPlaylistSnippet__info[^"]*"[^>]*>\s*(.+)&nbsp;.*;(\d+)\s*</',
+            html, 'genre and release year', default=(None, None), group=(1, 2))
 
-        year = int_or_none(info_text[info_sep + 1:]) if info_sep != -1 else None
         is_album = year is not None
-        genre = info_text[:info_sep].rstrip() if is_album else None
 
-        del header
+        thumbnail = url_or_none(self._html_search_regex(
-
+            r'class="[^"]*AudioPlaylistSnippet__cover[^"]*"[^>]*style="background-image\s*:\s*url\s*\(\s*\'([^\']+)',
-        description = clean_html(get_element_by_class('AudioPlaylistSnippet__description', html))
+            html, 'playlist thumbnail', fatal=False, group=1))
-
-        thumbnail = url_or_none(self._search_regex(
-            r'background[^:;]*:\s*url\s*\(\s*\'([^\']+)',
-            extract_attributes(
-                get_element_html_by_class(
-                    'AudioPlaylistSnippet__cover',
-                    html)).get('style'),
-            'playlist thumbnail', fatal=False, group=1))
 
         return self.playlist_result(
             entries, playlist_id,
@@ -1140,7 +1138,7 @@ class VKMusicPlaylistIE(VKMusicBaseIE):
             artists=[artist] if is_album else None,
             thumbnails=[{'url': thumbnail}] if thumbnail else [],
             genres=[genre] if genre else None,
-            release_year=year)
+            release_year=int_or_none(year))
 
 
 class VKPlayBaseIE(InfoExtractor):