Merge a9fc46dc7c into c699bafc50

Closes #11537

Authored by: bashonly

test/test_traversal.py

@@ -481,7 +481,7 @@ class TestTraversalHelpers:
             'id': 'name',
             'data': 'content',
             'url': 'url',
-        }, all, {subs_list_to_dict}]) == {
+        }, all, {subs_list_to_dict(lang=None)}]) == {
             'de': [{'url': 'https://example.com/subs/de.ass'}],
             'en': [{'data': 'content'}],
         }, 'subs with mandatory items missing should be filtered'
@@ -507,6 +507,54 @@ class TestTraversalHelpers:
             {'url': 'https://example.com/subs/en1', 'ext': 'ext'},
             {'url': 'https://example.com/subs/en2', 'ext': 'ext'},
         ]}, '`quality` key should sort subtitle list accordingly'
+        assert traverse_obj([
+            {'name': 'de', 'url': 'https://example.com/subs/de.ass'},
+            {'name': 'de'},
+            {'name': 'en', 'content': 'content'},
+            {'url': 'https://example.com/subs/en'},
+        ], [..., {
+            'id': 'name',
+            'url': 'url',
+            'data': 'content',
+        }, all, {subs_list_to_dict(lang='en')}]) == {
+            'de': [{'url': 'https://example.com/subs/de.ass'}],
+            'en': [
+                {'data': 'content'},
+                {'url': 'https://example.com/subs/en'},
+            ],
+        }, 'optionally provided lang should be used if no id available'
+        assert traverse_obj([
+            {'name': 1, 'url': 'https://example.com/subs/de1'},
+            {'name': {}, 'url': 'https://example.com/subs/de2'},
+            {'name': 'de', 'ext': 1, 'url': 'https://example.com/subs/de3'},
+            {'name': 'de', 'ext': {}, 'url': 'https://example.com/subs/de4'},
+        ], [..., {
+            'id': 'name',
+            'url': 'url',
+            'ext': 'ext',
+        }, all, {subs_list_to_dict(lang=None)}]) == {
+            'de': [
+                {'url': 'https://example.com/subs/de3'},
+                {'url': 'https://example.com/subs/de4'},
+            ],
+        }, 'non str types should be ignored for id and ext'
+        assert traverse_obj([
+            {'name': 1, 'url': 'https://example.com/subs/de1'},
+            {'name': {}, 'url': 'https://example.com/subs/de2'},
+            {'name': 'de', 'ext': 1, 'url': 'https://example.com/subs/de3'},
+            {'name': 'de', 'ext': {}, 'url': 'https://example.com/subs/de4'},
+        ], [..., {
+            'id': 'name',
+            'url': 'url',
+            'ext': 'ext',
+        }, all, {subs_list_to_dict(lang='de')}]) == {
+            'de': [
+                {'url': 'https://example.com/subs/de1'},
+                {'url': 'https://example.com/subs/de2'},
+                {'url': 'https://example.com/subs/de3'},
+                {'url': 'https://example.com/subs/de4'},
+            ],
+        }, 'non str types should be replaced by default id'
 
     def test_trim_str(self):
         with pytest.raises(TypeError):
@@ -525,7 +573,7 @@ class TestTraversalHelpers:
     def test_unpack(self):
         assert unpack(lambda *x: ''.join(map(str, x)))([1, 2, 3]) == '123'
         assert unpack(join_nonempty)([1, 2, 3]) == '1-2-3'
-        assert unpack(join_nonempty(delim=' '))([1, 2, 3]) == '1 2 3'
+        assert unpack(join_nonempty, delim=' ')([1, 2, 3]) == '1 2 3'
         with pytest.raises(TypeError):
             unpack(join_nonempty)()
         with pytest.raises(TypeError):

test/test_utils.py

@@ -72,7 +72,6 @@ from yt_dlp.utils import (
     intlist_to_bytes,
     iri_to_uri,
     is_html,
-    join_nonempty,
     js_to_json,
     limit_length,
     locked_file,
@@ -2158,10 +2157,6 @@ Line 1
         assert int_or_none(v=10) == 10, 'keyword passed positional should call function'
         assert int_or_none(scale=0.1)(10) == 100, 'call after partial application should call the function'
 
-        assert callable(join_nonempty(delim=', ')), 'varargs positional should apply partially'
-        assert callable(join_nonempty()), 'varargs positional should apply partially'
-        assert join_nonempty(None, delim=', ') == '', 'passed varargs should call the function'
-
 
 if __name__ == '__main__':
     unittest.main()

yt_dlp/YoutubeDL.py

@@ -4381,7 +4381,9 @@ class YoutubeDL:
             return None
 
         for idx, t in list(enumerate(thumbnails))[::-1]:
-            thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
+            thumb_ext = t.get('ext') or determine_ext(t['url'], 'jpg')
+            if multiple:
+                thumb_ext = f'{t["id"]}.{thumb_ext}'
             thumb_display_id = f'{label} thumbnail {t["id"]}'
             thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
             thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))

yt_dlp/extractor/afreecatv.py

@@ -66,6 +66,14 @@ class AfreecaTVBaseIE(InfoExtractor):
             extensions={'legacy_ssl': True}), display_id,
             'Downloading API JSON', 'Unable to download API JSON')
 
+    @staticmethod
+    def _fixup_thumb(thumb_url):
+        if not url_or_none(thumb_url):
+            return None
+        # Core would determine_ext as 'php' from the url, so we need to provide the real ext
+        # See: https://github.com/yt-dlp/yt-dlp/issues/11537
+        return [{'url': thumb_url, 'ext': 'jpg'}]
+
 
 class AfreecaTVIE(AfreecaTVBaseIE):
     IE_NAME = 'soop'
@@ -155,7 +163,7 @@ class AfreecaTVIE(AfreecaTVBaseIE):
             'uploader': ('writer_nick', {str}),
             'uploader_id': ('bj_id', {str}),
             'duration': ('total_file_duration', {int_or_none(scale=1000)}),
-            'thumbnail': ('thumb', {url_or_none}),
+            'thumbnails': ('thumb', {self._fixup_thumb}),
         })
 
         entries = []
@@ -226,8 +234,7 @@ class AfreecaTVCatchStoryIE(AfreecaTVBaseIE):
 
         return self.playlist_result(self._entries(data), video_id)
 
-    @staticmethod
-    def _entries(data):
+    def _entries(self, data):
         # 'files' is always a list with 1 element
         yield from traverse_obj(data, (
             'data', lambda _, v: v['story_type'] == 'catch',
@@ -238,7 +245,7 @@ class AfreecaTVCatchStoryIE(AfreecaTVBaseIE):
                 'title': ('title', {str}),
                 'uploader': ('writer_nick', {str}),
                 'uploader_id': ('writer_id', {str}),
-                'thumbnail': ('thumb', {url_or_none}),
+                'thumbnails': ('thumb', {self._fixup_thumb}),
                 'timestamp': ('write_timestamp', {int_or_none}),
             }))
 

yt_dlp/extractor/common.py

@@ -279,6 +279,7 @@ class InfoExtractor:
     thumbnails:     A list of dictionaries, with the following entries:
                         * "id" (optional, string) - Thumbnail format ID
                         * "url"
+                        * "ext" (optional, string) - actual image extension if not given in URL
                         * "preference" (optional, int) - quality of the image
                         * "width" (optional, int)
                         * "height" (optional, int)

yt_dlp/extractor/rts.py

@@ -1,182 +1,290 @@
-import re
+import functools
 
 from .srgssr import SRGSSRIE
 from ..utils import (
     determine_ext,
     int_or_none,
-    parse_duration,
+    orderedSet,
     parse_iso8601,
-    unescapeHTML,
-    urljoin,
+    parse_resolution,
 )
+from ..utils.traversal import traverse_obj
 
 
-class RTSIE(SRGSSRIE):  # XXX: Do not subclass from concrete IE
-    _WORKING = False
+class RTSIE(SRGSSRIE):
+    _GEO_COUNTRIES = ['CH']
     IE_DESC = 'RTS.ch'
-    _VALID_URL = r'rts:(?P<rts_id>\d+)|https?://(?:.+?\.)?rts\.ch/(?:[^/]+/){2,}(?P<id>[0-9]+)-(?P<display_id>.+?)\.html'
+    _VALID_URL = [
+        r'rts:(?P<id>\d+)',
+        r'https?://(?:.+?\.)?rts\.ch/(?:[^/]+/){2,}(?P<id>[0-9]+)-(?P<display_id>.+?)\.html',
+        r'https?://(?:.+?\.)?rts\.ch/(?:[^/]+/){2,}(?P<display_id>.+?)-(?P<id>[0-9]+)\.html',
+    ]
 
     _TESTS = [
         {
+            # article with videos
             'url': 'http://www.rts.ch/archives/tv/divers/3449373-les-enfants-terribles.html',
-            'md5': '753b877968ad8afaeddccc374d4256a5',
             'info_dict': {
                 'id': '3449373',
-                'display_id': 'les-enfants-terribles',
-                'ext': 'mp4',
-                'duration': 1488,
                 'title': 'Les Enfants Terribles',
                 'description': 'France Pommier et sa soeur Luce Feral, les deux filles de ce groupe de 5.',
-                'uploader': 'Divers',
+                'display_id': 'les-enfants-terribles',
+                'tags': ['Divers', 'Archives TV', 'Culture et Arts', 'Les archives', 'Personnalités', 'RTS Archives', 'Années 1960', 'Autres arts', 'Décennies', 'Société'],
+            },
+            'playlist': [{
+                'info_dict': {
+                    'id': '3449373',
+                    'ext': 'mp4',
+                    'title': 'Les Enfants Terribles',
+                    'description': 'France Pommier et sa soeur Luce Feral, les deux filles de ce groupe de 5.',
+                    'thumbnail': r're:^https?://.*\.image',
                     'upload_date': '19680921',
                     'timestamp': -40280400,
-                'thumbnail': r're:^https?://.*\.image',
-                'view_count': int,
+                    'duration': 1488,
+                    'categories': ['Divers'],
                 },
-            'expected_warnings': ['Unable to download f4m manifest', 'Failed to download m3u8 information'],
-        },
-        {
-            'url': 'http://www.rts.ch/emissions/passe-moi-les-jumelles/5624067-entre-ciel-et-mer.html',
-            'info_dict': {
-                'id': '5624065',
-                'title': 'Passe-moi les jumelles',
-            },
-            'playlist_mincount': 4,
+            }],
+            'params': {'skip_download': 'm3u8'},  # 700-byte first fragment
         },
         {
+            # video without text content
             'url': 'http://www.rts.ch/video/sport/hockey/5745975-1-2-kloten-fribourg-5-2-second-but-pour-gotteron-par-kwiatowski.html',
             'info_dict': {
                 'id': '5745975',
                 'display_id': '1-2-kloten-fribourg-5-2-second-but-pour-gotteron-par-kwiatowski',
-                'ext': 'mp4',
-                'duration': 48,
                 'title': '1/2, Kloten - Fribourg (5-2): second but pour Gottéron par Kwiatowski',
                 'description': 'Hockey - Playoff',
-                'uploader': 'Hockey',
+                'tags': ['Hockey', 'Sport', 'RTS Sport'],
+            },
+            'playlist': [{
+                'info_dict': {
+                    'id': '5745975',
+                    'ext': 'mp4',
+                    'title': '1/2, Kloten - Fribourg (5-2): second but pour Gottéron par Kwiatowski',
+                    'description': 'Hockey - Playoff',
+                    'thumbnail': r're:^https?://.*\.image',
                     'upload_date': '20140403',
                     'timestamp': 1396556882,
-                'thumbnail': r're:^https?://.*\.image',
-                'view_count': int,
+                    'duration': 48,
+                    'categories': ['Hockey sur glace'],
                 },
-            'params': {
-                # m3u8 download
-                'skip_download': True,
-            },
-            'expected_warnings': ['Unable to download f4m manifest', 'Failed to download m3u8 information'],
+            }],
+            'params': {'skip_download': 'm3u8'},  # 700-byte first fragment
             'skip': 'Blocked outside Switzerland',
         },
         {
+            # video player; redirection: https://www.rts.ch/play/tv/lactu-en-video/video/londres-cachee-par-un-epais-smog?urn=urn:rts:video:5745356
             'url': 'http://www.rts.ch/video/info/journal-continu/5745356-londres-cachee-par-un-epais-smog.html',
-            'md5': '9bb06503773c07ce83d3cbd793cebb91',
             'info_dict': {
                 'id': '5745356',
-                'display_id': 'londres-cachee-par-un-epais-smog',
                 'ext': 'mp4',
-                'duration': 33,
+                'duration': 33.76,
                 'title': 'Londres cachée par un épais smog',
                 'description': 'Un important voile de smog recouvre Londres depuis mercredi, provoqué par la pollution et du sable du Sahara.',
-                'uploader': 'L\'actu en vidéo',
                 'upload_date': '20140403',
                 'timestamp': 1396537322,
                 'thumbnail': r're:^https?://.*\.image',
-                'view_count': int,
+                'webpage_url': 'srgssr:rts:video:5745356',
             },
-            'expected_warnings': ['Unable to download f4m manifest', 'Failed to download m3u8 information'],
+            'params': {'skip_download': 'm3u8'},  # 700-byte first fragment
         },
         {
+            # audio & podcast
             'url': 'http://www.rts.ch/audio/couleur3/programmes/la-belle-video-de-stephane-laurenceau/5706148-urban-hippie-de-damien-krisl-03-04-2014.html',
-            'md5': 'dd8ef6a22dff163d063e2a52bc8adcae',
             'info_dict': {
                 'id': '5706148',
-                'display_id': 'urban-hippie-de-damien-krisl-03-04-2014',
-                'ext': 'mp3',
-                'duration': 123,
                 'title': '"Urban Hippie", de Damien Krisl',
                 'description': 'Des Hippies super glam.',
-                'upload_date': '20140403',
-                'timestamp': 1396551600,
+                'display_id': 'urban-hippie-de-damien-krisl',
+                'tags': ['Media Radio', 'Couleur3'],
             },
+            'playlist': [{
+                'info_dict': {
+                    'id': '5706148',
+                    'ext': 'mp3',
+                    'title': '"Urban Hippie", de Damien Krisl',
+                    'description': 'Des Hippies super glam.',
+                    'thumbnail': r're:^https?://.*\.image',
+                    'upload_date': '20140403',
+                    'timestamp': 1396546481,
+                    'duration': 123,
+                    'categories': ['La belle vidéo de Stéphane Laurenceau'],
+                },
+            }, {
+                'info_dict': {
+                    'id': '5747185',
+                    'ext': 'mp3',
+                    'title': 'Le musée du psychédélisme',
+                    'description': 'md5:72f8662f48c32050ae817e3bde7e0acc',
+                    'thumbnail': r're:^https?://.*\.image',
+                    'upload_date': '20140402',
+                    'timestamp': 1396476000,
+                    'duration': 274,
+                    'categories': ['Happy Culture'],
+                },
+            }, {
+                'info_dict': {
+                    'id': '5706149',
+                    'ext': 'mp3',
+                    'title': 'Silk Art Hippie Culture',
+                    'description': 'md5:8e3b9d8d84d85ca8a1905cf50b39bba4',
+                    'thumbnail': r're:^https?://.*\.image',
+                    'upload_date': '20140403',
+                    'timestamp': 1396545649,
+                    'duration': 161,
+                    'categories': ['Happy Pics'],
+                },
+            }, {
+                'info_dict': {
+                    'id': '5706148',
+                    'ext': 'mp3',
+                    'title': '"Urban Hippie", de Damien Krisl',
+                    'description': 'Des Hippies super glam.',
+                    'thumbnail': r're:^https?://.*\.image',
+                    'upload_date': '20140403',
+                    'timestamp': 1396546481,
+                    'duration': 123,
+                    'categories': ['La belle vidéo de Stéphane Laurenceau'],
+                },
+            }],
         },
         {
             # article with videos on rhs
             'url': 'http://www.rts.ch/sport/hockey/6693917-hockey-davos-decroche-son-31e-titre-de-champion-de-suisse.html',
             'info_dict': {
                 'id': '6693917',
-                'title': 'Hockey: Davos décroche son 31e titre de champion de Suisse',
+                'title': 'Davos décroche le 31e titre de son histoire',
+                'description': 'md5:3c9a767b2a332413eda33c526024578c',
+                'display_id': 'hockey-davos-decroche-son-31e-titre-de-champion-de-suisse',
+                'tags': ['Hockey', 'Tout le sport', 'RTS Info', 'LNA', "Toute l'info", 'RTS Sport'],
             },
             'playlist_mincount': 5,
+            'skip': 'Blocked outside Switzerland',
+        },
+        {
+            # articles containing recordings of TV shows
+            'url': 'https://www.rts.ch/info/regions/valais/12865814-un-bouquetin-emporte-par-un-aigle-royal-sur-les-hauts-de-fully-vs.html',
+            'info_dict': {
+                'id': '12865814',
+                'title': 'Un bouquetin emporté par un aigle royal sur les hauts de Fully (VS)',
+                'description': 'md5:9b511f89075e2730bd2dd59915c25574',
+                'display_id': 'un-bouquetin-emporte-par-un-aigle-royal-sur-les-hauts-de-fully-vs',
+                'tags': ['Régions', 'RTS Info', 'Valais', "Toute l'info"],
+            },
+            'playlist': [{
+                'info_dict': {
+                    'id': '12861415',
+                    'ext': 'mp4',
+                    'title': 'En Valais, un bouquetin emporté dans les airs par un aigle royal. Décryptage d’une image rare.',
+                    'thumbnail': r're:^https?://.*\.image',
+                    'timestamp': 1644690600,
+                    'upload_date': '20220212',
+                    'duration': 107,
+                    'categories': ['19h30'],
+                },
+            }],
+            'params': {'skip_download': 'm3u8'},  # 700-byte first fragment
+            'expected_warnings': ['Ignoring subtitle tracks found in the HLS manifest'],
+        },
+        {
+            # new URL format; article with videos
+            'url': 'https://www.rts.ch/info/suisse/2024/article/doris-leuthard-il-y-a-des-alternatives-au-nucleaire-qui-sont-moins-risquees-28631869.html',
+            'info_dict': {
+                'id': '28631869',
+                'title': 'Doris Leuthard: "Il y a des alternatives au nucléaire qui sont moins risquées"',
+                'description': 'md5:ba9930e218dcd177801a34b89a16b86e',
+                'display_id': 'doris-leuthard-il-y-a-des-alternatives-au-nucleaire-qui-sont-moins-risquees',
+                'tags': 'count:13',
+            },
+            'playlist': [{
+                'info_dict': {
+                    'id': '15162786',
+                    'ext': 'mp4',
+                    'title': 'L\'invitée de La Matinale (vidéo) - Doris Leuthard, co-présidente du projet d\'exposition nationale Svizra27',
+                    'thumbnail': r're:^https?://.*\.image',
+                    'upload_date': '20240916',
+                    'timestamp': 1726462800,
+                    'duration': 860,
+                    'categories': ['La Matinale'],
+                },
+            }, {
+                'info_dict': {
+                    'id': '15164848',
+                    'ext': 'mp4',
+                    'title': 'Le Centre pourrait faire pencher la balance en faveur de la construction de nouvelles centrales nucléaires',
+                    'thumbnail': r're:^https?://.*\.image',
+                    'upload_date': '20240916',
+                    'timestamp': 1726502400,
+                    'duration': 227,
+                    'categories': ['Forum'],
+                },
+            }],
+            'params': {'skip_download': 'm3u8'},  # 700-byte first fragment
+            'expected_warnings': ['Ignoring subtitle tracks found in the HLS manifest'],
         },
         {
             'url': 'http://pages.rts.ch/emissions/passe-moi-les-jumelles/5624065-entre-ciel-et-mer.html',
             'only_matching': True,
         },
+        {
+            'url': 'http://www.rts.ch/emissions/passe-moi-les-jumelles/5624067-entre-ciel-et-mer.html',
+            'only_matching': True,
+        },
     ]
 
     def _real_extract(self, url):
-        m = self._match_valid_url(url)
-        media_id = m.group('rts_id') or m.group('id')
-        display_id = m.group('display_id') or media_id
+        webpage, urlh = self._download_webpage_handle(url, self._match_id(url))
+        if urlh.url != url:
+            return self.url_result(urlh.url)
 
-        def download_json(internal_id):
-            return self._download_json(
-                f'http://www.rts.ch/a/{internal_id}.html?f=json/article',
-                display_id)
+        mobj = self._match_valid_url(url)
+        display_id = traverse_obj(mobj, 'display_id', default=mobj.group('id')) or mobj.group('id')
 
-        all_info = download_json(media_id)
+        media_list = []
+        article_details = self._search_json(r'articleDetails\s*=\s*', webpage, 'article details', display_id)
+        traverse_obj(article_details, ('mainMedia', filter, {media_list.append}))
+        traverse_obj(article_details, ('innerMediaElements', filter, {media_list.extend}))
+        traverse_obj(article_details, ('mediaElements', filter, {media_list.extend}))
+        media_list = orderedSet(media_list)
 
-        # media_id extracted out of URL is not always a real id
-        if 'video' not in all_info and 'audio' not in all_info:
         entries = []
+        for media in media_list:
+            media_id = media['oid']
+            media_info = self._get_media_data('rts', media['type'], media_id)
 
-            for item in all_info.get('items', []):
-                item_url = item.get('url')
-                if not item_url:
-                    continue
-                entries.append(self.url_result(item_url, 'RTS'))
+            if fmts := self._extract_formats(media_info, media_id):
+                entries.append({
+                    'id': media_info['id'],
+                    'title': media_info['title'],
+                    'formats': fmts,
+                    'description': media_info.get('description'),
+                    'thumbnails': [traverse_obj(media_info, ('imageUrl', {lambda x: {
+                        'url': x,
+                        **parse_resolution(x),
+                    }}))],
+                    'timestamp': parse_iso8601(media_info.get('date')),
+                    'duration': traverse_obj(media_info, ('duration', {functools.partial(int_or_none, scale=1000)})),
+                    'categories': [media.get('category')],
+                })
 
-            if not entries:
-                page, urlh = self._download_webpage_handle(url, display_id)
-                if re.match(self._VALID_URL, urlh.url).group('id') != media_id:
-                    return self.url_result(urlh.url, 'RTS')
-
-                # article with videos on rhs
-                videos = re.findall(
-                    r'<article[^>]+class="content-item"[^>]*>\s*<a[^>]+data-video-urn="urn:([^"]+)"',
-                    page)
-                if not videos:
-                    videos = re.findall(
-                        r'(?s)<iframe[^>]+class="srg-player"[^>]+src="[^"]+urn:([^"]+)"',
-                        page)
-                if videos:
-                    entries = [self.url_result(f'srgssr:{video_urn}', 'SRGSSR') for video_urn in videos]
-
-            if entries:
-                return self.playlist_result(entries, media_id, all_info.get('title'))
-
-            internal_id = self._html_search_regex(
-                r'<(?:video|audio) data-id="([0-9]+)"', page,
-                'internal video id')
-            all_info = download_json(internal_id)
-
-        media_type = 'video' if 'video' in all_info else 'audio'
-
-        # check for errors
-        self._get_media_data('rts', media_type, media_id)
-
-        info = all_info['video']['JSONinfo'] if 'video' in all_info else all_info['audio']
-
-        title = info['title']
+        return self.playlist_result(
+            entries, article_details.get('oid'), article_details.get('title'),
+            article_details.get('lead'), display_id=display_id,
+            tags=traverse_obj(article_details, ('tags', ..., 'name')))
 
+    def _extract_formats(self, media_info, media_id):
         def extract_bitrate(url):
             return int_or_none(self._search_regex(
                 r'-([0-9]+)k\.', url, 'bitrate', default=None))
 
         formats = []
-        streams = info.get('streams', {})
-        for format_id, format_url in streams.items():
-            if format_id == 'hds_sd' and 'hds' in streams:
+        for idx, stream in enumerate(traverse_obj(
+                media_info, ('resourceList', lambda _, v: v['url']))):
+            format_id = stream.get('protocol') or str(idx)
+            format_url = stream['url']
+            if format_id == 'hds_sd' and 'hds' in stream:
                 continue
-            if format_id == 'hls_sd' and 'hls' in streams:
+            if format_id == 'hls_sd' and 'hls' in stream:
                 continue
             ext = determine_ext(format_url)
             if ext in ('m3u8', 'f4m'):
@@ -195,37 +303,5 @@ class RTSIE(SRGSSRIE):  # XXX: Do not subclass from concrete IE
                     'tbr': extract_bitrate(format_url),
                 })
 
-        download_base = 'http://rtsww{}-d.rts.ch/'.format('-a' if media_type == 'audio' else '')
-        for media in info.get('media', []):
-            media_url = media.get('url')
-            if not media_url or re.match(r'https?://', media_url):
-                continue
-            rate = media.get('rate')
-            ext = media.get('ext') or determine_ext(media_url, 'mp4')
-            format_id = ext
-            if rate:
-                format_id += '-%dk' % rate
-            formats.append({
-                'format_id': format_id,
-                'url': urljoin(download_base, media_url),
-                'tbr': rate or extract_bitrate(media_url),
-            })
-
         self._check_formats(formats, media_id)
-
-        duration = info.get('duration') or info.get('cutout') or info.get('cutduration')
-        if isinstance(duration, str):
-            duration = parse_duration(duration)
-
-        return {
-            'id': media_id,
-            'display_id': display_id,
-            'formats': formats,
-            'title': title,
-            'description': info.get('intro'),
-            'duration': duration,
-            'view_count': int_or_none(info.get('plays')),
-            'uploader': info.get('programName'),
-            'timestamp': parse_iso8601(info.get('broadcast_date')),
-            'thumbnail': unescapeHTML(info.get('preview_image_url')),
-        }
+        return formats

yt_dlp/extractor/srgssr.py

@@ -57,7 +57,7 @@ class SRGSSRIE(InfoExtractor):
     def _get_media_data(self, bu, media_type, media_id):
         query = {'onlyChapters': True} if media_type == 'video' else {}
         full_media_data = self._download_json(
-            f'https://il.srgssr.ch/integrationlayer/2.0/{bu}/mediaComposition/{media_type}/{media_id}.json',
+            f'https://il.srgssr.ch/integrationlayer/2.0/mediaComposition/byUrn/urn:{bu}:{media_type}:{media_id}.json',
             media_id, query=query)['chapterList']
         try:
             media_data = next(
@@ -165,7 +165,7 @@ class SRGSSRPlayIE(InfoExtractor):
 
     _TESTS = [{
         'url': 'http://www.srf.ch/play/tv/10vor10/video/snowden-beantragt-asyl-in-russland?id=28e1a57d-5b76-4399-8ab3-9097f071e6c5',
-        'md5': '6db2226ba97f62ad42ce09783680046c',
+        'md5': '81c6ad90d774c46e3c54ea2f01a94db3',
         'info_dict': {
             'id': '28e1a57d-5b76-4399-8ab3-9097f071e6c5',
             'ext': 'mp4',
@@ -173,7 +173,7 @@ class SRGSSRPlayIE(InfoExtractor):
             'title': 'Snowden beantragt Asyl in Russland',
             'timestamp': 1372708215,
             'duration': 113.827,
-            'thumbnail': r're:^https?://.*1383719781\.png$',
+            'thumbnail': r're:^https?://download-media\.srf\.ch/.*\.(?:png|jpg)$',
         },
         'expected_warnings': ['Unable to download f4m manifest'],
     }, {
@@ -185,6 +185,7 @@ class SRGSSRPlayIE(InfoExtractor):
             'title': 'Saira: Tujetsch - tuttina cuntinuar cun Sedrun Mustér Turissem',
             'timestamp': 1444709160,
             'duration': 336.816,
+            'thumbnail': r're:^https?://download-media\.srf\.ch/.*\.(?:png|jpg)$',
         },
         'params': {
             # rtmp download
@@ -217,7 +218,7 @@ class SRGSSRPlayIE(InfoExtractor):
             'duration': 94.0,
             'upload_date': '20170215',
             'timestamp': 1487173560,
-            'thumbnail': r're:https?://www\.swissinfo\.ch/srgscalableimage/42961964',
+            'thumbnail': r're:https?://cdn\.prod\.swi-services\.ch/.+',
             'subtitles': 'count:9',
         },
         'params': {

yt_dlp/utils/_utils.py

@@ -216,7 +216,7 @@ def partial_application(func):
     sig = inspect.signature(func)
     required_args = [
         param.name for param in sig.parameters.values()
-        if param.kind in (inspect.Parameter.POSITIONAL_ONLY, inspect.Parameter.POSITIONAL_OR_KEYWORD, inspect.Parameter.VAR_POSITIONAL)
+        if param.kind in (inspect.Parameter.POSITIONAL_ONLY, inspect.Parameter.POSITIONAL_OR_KEYWORD)
         if param.default is inspect.Parameter.empty
     ]
 
@@ -4837,7 +4837,6 @@ def number_of_digits(number):
     return len('%d' % number)
 
 
-@partial_application
 def join_nonempty(*values, delim='-', from_dict=None):
     if from_dict is not None:
         values = (traversal.traverse_obj(from_dict, variadic(v)) for v in values)

yt_dlp/utils/traversal.py

@@ -332,14 +332,14 @@ class _RequiredError(ExtractorError):
 
 
 @typing.overload
-def subs_list_to_dict(*, ext: str | None = None) -> collections.abc.Callable[[list[dict]], dict[str, list[dict]]]: ...
+def subs_list_to_dict(*, lang: str | None = 'und', ext: str | None = None) -> collections.abc.Callable[[list[dict]], dict[str, list[dict]]]: ...
 
 
 @typing.overload
-def subs_list_to_dict(subs: list[dict] | None, /, *, ext: str | None = None) -> dict[str, list[dict]]: ...
+def subs_list_to_dict(subs: list[dict] | None, /, *, lang: str | None = 'und', ext: str | None = None) -> dict[str, list[dict]]: ...
 
 
-def subs_list_to_dict(subs: list[dict] | None = None, /, *, ext=None):
+def subs_list_to_dict(subs: list[dict] | None = None, /, *, lang='und', ext=None):
     """
     Convert subtitles from a traversal into a subtitle dict.
     The path should have an `all` immediately before this function.
@@ -352,7 +352,7 @@ def subs_list_to_dict(subs: list[dict] | None = None, /, *, ext=None):
     `quality`  The sort order for each subtitle
     """
     if subs is None:
-        return functools.partial(subs_list_to_dict, ext=ext)
+        return functools.partial(subs_list_to_dict, lang=lang, ext=ext)
 
     result = collections.defaultdict(list)
 
@@ -360,9 +360,15 @@ def subs_list_to_dict(subs: list[dict] | None = None, /, *, ext=None):
         if not url_or_none(sub.get('url')) and not sub.get('data'):
             continue
         sub_id = sub.pop('id', None)
-        if sub_id is None:
+        if not isinstance(sub_id, str):
+            if not lang:
                 continue
-        if ext is not None and not sub.get('ext'):
+            sub_id = lang
+        sub_ext = sub.get('ext')
+        if not isinstance(sub_ext, str):
+            if not ext:
+                sub.pop('ext', None)
+            else:
                 sub['ext'] = ext
         result[sub_id].append(sub)
     result = dict(result)
@@ -452,9 +458,9 @@ def trim_str(*, start=None, end=None):
     return trim
 
 
-def unpack(func):
+def unpack(func, **kwargs):
     @functools.wraps(func)
-    def inner(items, **kwargs):
+    def inner(items):
         return func(*items, **kwargs)
 
     return inner