Improve default subtitle language selection (#6240)

Authored by: sdht0

yt_dlp/YoutubeDL.py

@@ -2810,10 +2810,14 @@ class YoutubeDL:
                     self.params.get('subtitleslangs'), {'all': all_sub_langs}, use_regex=True)
             except re.error as e:
                 raise ValueError(f'Wrong regex for subtitlelangs: {e.pattern}')
-        elif normal_sub_langs:
-            requested_langs = ['en'] if 'en' in normal_sub_langs else normal_sub_langs[:1]
         else:
-            requested_langs = ['en'] if 'en' in all_sub_langs else all_sub_langs[:1]
+            requested_langs = LazyList(itertools.chain(
+                ['en'] if 'en' in normal_sub_langs else [],
+                filter(lambda f: f.startswith('en'), normal_sub_langs),
+                ['en'] if 'en' in all_sub_langs else [],
+                filter(lambda f: f.startswith('en'), all_sub_langs),
+                normal_sub_langs, all_sub_langs,
+            ))[:1]
         if requested_langs:
             self.to_screen(f'[info] {video_id}: Downloading subtitles: {", ".join(requested_langs)}')
 

yt_dlp/extractor/crunchyroll.py

@@ -160,7 +160,7 @@ class CrunchyrollBetaIE(CrunchyrollBaseIE):
         episode_response = self._download_json(
             f'{api_domain}/cms/v2{bucket}/episodes/{internal_id}', display_id,
             note='Retrieving episode metadata', query=params)
-        if episode_response.get('is_premium_only') and not episode_response.get('playback'):
+        if episode_response.get('is_premium_only') and not bucket.endswith('crunchyroll'):
             if self.is_logged_in:
                 raise ExtractorError('This video is for premium members only', expected=True)
             else:

yt_dlp/extractor/youtube.py

@@ -4120,7 +4120,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
             'description': video_description,
             'uploader': get_first(video_details, 'author'),
-            'uploader_id': self._search_regex(r'/(?:channel|user)/([^/?&#]+)', owner_profile_url, 'uploader id') if owner_profile_url else None,
+            'uploader_id': self._search_regex(r'/(?:channel/|user/|@)([^/?&#]+)', owner_profile_url, 'uploader id', default=None),
             'uploader_url': owner_profile_url,
             'channel_id': channel_id,
             'channel_url': format_field(channel_id, None, 'https://www.youtube.com/channel/%s'),