X-Git-Url: https://jfr.im/git/yt-dlp.git/blobdiff_plain/16aa9ea41dd0a575c6c2d6945d3f8fe3bb42b327..88afe05695464e4242933b9474eaf2b15f8b55bb:/yt_dlp/extractor/youtube.py

diff --git a/yt_dlp/extractor/youtube.py b/yt_dlp/extractor/youtube.py
index a424e0e57..5750e75d7 100644
--- a/yt_dlp/extractor/youtube.py
+++ b/yt_dlp/extractor/youtube.py
@@ -258,7 +258,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
 
     _RESERVED_NAMES = (
         r'channel|c|user|playlist|watch|w|v|embed|e|watch_popup|clip|'
-        r'shorts|movies|results|shared|hashtag|trending|explore|feed|feeds|'
+        r'shorts|movies|results|search|shared|hashtag|trending|explore|feed|feeds|'
         r'browse|oembed|get_video_info|iframe_api|s/player|'
         r'storefront|oops|index|account|reporthistory|t/terms|about|upload|signin|logout')
 
@@ -2422,7 +2422,7 @@ def _decrypt_nsig(self, s, video_id, player_url):
 
     def _extract_n_function_name(self, jscode):
         nfunc, idx = self._search_regex(
-            r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]{3})(\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)',
+            r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]{3})(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)',
             jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
         if not idx:
             return nfunc
@@ -3006,13 +3006,17 @@ def _extract_formats(self, streaming_data, video_id, player_url, is_live):
 
             tbr = float_or_none(
                 fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
+            language_preference = (
+                10 if audio_track.get('audioIsDefault') and 10
+                else -10 if 'descriptive' in (audio_track.get('displayName') or '').lower() and -10
+                else -1)
             dct = {
                 'asr': int_or_none(fmt.get('audioSampleRate')),
                 'filesize': int_or_none(fmt.get('contentLength')),
                 'format_id': itag,
                 'format_note': join_nonempty(
                     '%s%s' % (audio_track.get('displayName') or '',
-                              ' (default)' if audio_track.get('audioIsDefault') else ''),
+                              ' (default)' if language_preference > 0 else ''),
                     fmt.get('qualityLabel') or quality.replace('audio_quality_', ''),
                     throttled and 'THROTTLED', delim=', '),
                 'source_preference': -10 if throttled else -1,
@@ -3022,8 +3026,9 @@ def _extract_formats(self, streaming_data, video_id, player_url, is_live):
                 'tbr': tbr,
                 'url': fmt_url,
                 'width': int_or_none(fmt.get('width')),
-                'language': audio_track.get('id', '').split('.')[0],
-                'language_preference': 1 if audio_track.get('audioIsDefault') else -1,
+                'language': join_nonempty(audio_track.get('id', '').split('.')[0],
+                                          'desc' if language_preference < -1 else ''),
+                'language_preference': language_preference,
             }
             mime_mobj = re.match(
                 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
@@ -3408,6 +3413,9 @@ def process_language(container, base_url, lang_code, sub_name, query):
                         trans_name += format_field(lang_name, template=' from %s')
                     process_language(
                         automatic_captions, base_url, trans_code, trans_name, {'tlang': trans_code})
+                    if lang_code == f'a-{trans_code}':
+                        process_language(
+                            automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {'tlang': trans_code})
             info['automatic_captions'] = automatic_captions
             info['subtitles'] = subtitles
 
@@ -3601,6 +3609,26 @@ def process_language(container, base_url, lang_code, sub_name, query):
 
 class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
 
+    @staticmethod
+    def passthrough_smuggled_data(func):
+        def _smuggle(entries, smuggled_data):
+            for entry in entries:
+                # TODO: Convert URL to music.youtube instead.
+                # Do we need to passthrough any other smuggled_data?
+                entry['url'] = smuggle_url(entry['url'], smuggled_data)
+                yield entry
+
+        @functools.wraps(func)
+        def wrapper(self, url):
+            url, smuggled_data = unsmuggle_url(url, {})
+            if self.is_music_url(url):
+                smuggled_data['is_music_url'] = True
+            info_dict = func(self, url, smuggled_data)
+            if smuggled_data and info_dict.get('entries'):
+                info_dict['entries'] = _smuggle(info_dict['entries'], smuggled_data)
+            return info_dict
+        return wrapper
+
     def _extract_channel_id(self, webpage):
         channel_id = self._html_search_meta(
             'channelId', webpage, 'channel id', default=None)
@@ -4251,16 +4279,9 @@ def _extract_tab_endpoint(self, url, item_id, ytcfg=None, fatal=True, default_cl
             raise ExtractorError(err_note, expected=True)
         self.report_warning(err_note, item_id)
 
-    @staticmethod
-    def _smuggle_data(entries, data):
-        for entry in entries:
-            if data:
-                entry['url'] = smuggle_url(entry['url'], data)
-            yield entry
-
     _SEARCH_PARAMS = None
 
-    def _search_results(self, query, params=NO_DEFAULT, client=None):
+    def _search_results(self, query, params=NO_DEFAULT, default_client='web'):
         data = {'query': query}
         if params is NO_DEFAULT:
             params = self._SEARCH_PARAMS
@@ -4281,7 +4302,7 @@ def _search_results(self, query, params=NO_DEFAULT, client=None):
             data.update(continuation_list[0] or {})
             search = self._extract_response(
                 item_id='query "%s" page %s' % (query, page_num), ep='search', query=data,
-                default_client=client, check_get_keys=check_get_keys)
+                default_client=default_client, check_get_keys=check_get_keys)
             slr_contents = traverse_obj(search, *content_keys)
             yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list)
             if not continuation_list[0]:
@@ -4960,18 +4981,10 @@ def suitable(cls, url):
         return False if YoutubeIE.suitable(url) else super(
             YoutubeTabIE, cls).suitable(url)
 
-    def _real_extract(self, url):
-        url, smuggled_data = unsmuggle_url(url, {})
-        if self.is_music_url(url):
-            smuggled_data['is_music_url'] = True
-        info_dict = self.__real_extract(url, smuggled_data)
-        if info_dict.get('entries'):
-            info_dict['entries'] = self._smuggle_data(info_dict['entries'], smuggled_data)
-        return info_dict
-
     _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(not_channel)|(?P<tab>/\w+))?(?P<post>.*)$')
 
-    def __real_extract(self, url, smuggled_data):
+    @YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
+    def _real_extract(self, url, smuggled_data):
         item_id = self._match_id(url)
         url = compat_urlparse.urlunparse(
             compat_urlparse.urlparse(url)._replace(netloc='www.youtube.com'))
@@ -5306,7 +5319,14 @@ class YoutubeSearchIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
     IE_NAME = 'youtube:search'
     _SEARCH_KEY = 'ytsearch'
     _SEARCH_PARAMS = 'EgIQAQ%3D%3D'  # Videos only
-    _TESTS = []
+    _TESTS = [{
+        'url': 'ytsearch5:youtube-dl test video',
+        'playlist_count': 5,
+        'info_dict': {
+            'id': 'youtube-dl test video',
+            'title': 'youtube-dl test video',
+        }
+    }]
 
 
 class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
@@ -5314,12 +5334,20 @@ class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
     _SEARCH_KEY = 'ytsearchdate'
     IE_DESC = 'YouTube search, newest videos first'
     _SEARCH_PARAMS = 'CAISAhAB'  # Videos only, sorted by date
+    _TESTS = [{
+        'url': 'ytsearchdate5:youtube-dl test video',
+        'playlist_count': 5,
+        'info_dict': {
+            'id': 'youtube-dl test video',
+            'title': 'youtube-dl test video',
+        }
+    }]
 
 
 class YoutubeSearchURLIE(YoutubeTabBaseInfoExtractor):
     IE_DESC = 'YouTube search URLs with sorting and filter support'
     IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
-    _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?:[^&]+)(?:[&]|$)'
+    _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:results|search)\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
     _TESTS = [{
         'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
         'playlist_mincount': 5,
@@ -5347,7 +5375,7 @@ def _real_extract(self, url):
 
 
 class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor):
-    IE_DESC = 'YouTube music search URLs with sorting and filter support'
+    IE_DESC = 'YouTube music search URLs with selectable sections (Eg: #songs)'
     IE_NAME = 'youtube:music:search_url'
     _VALID_URL = r'https?://music\.youtube\.com/search\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
     _TESTS = [{
@@ -5396,10 +5424,10 @@ def _real_extract(self, url):
             if not params:
                 section = None
         title = join_nonempty(query, section, delim=' - ')
-        return self.playlist_result(self._search_results(query, params, client='web_music'), title, title)
+        return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title)
 
 
-class YoutubeFeedsInfoExtractor(YoutubeTabIE):
+class YoutubeFeedsInfoExtractor(InfoExtractor):
     """
     Base class for feed extractors
     Subclasses must define the _FEED_NAME property.
@@ -5413,8 +5441,7 @@ def IE_NAME(self):
 
     def _real_extract(self, url):
         return self.url_result(
-            'https://www.youtube.com/feed/%s' % self._FEED_NAME,
-            ie=YoutubeTabIE.ie_key())
+            f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key())
 
 
 class YoutubeWatchLaterIE(InfoExtractor):