Merge 513d4c358e into f2a4983df7

Closes #11526
Authored by: jshumphrey

yt_dlp/extractor/archiveorg.py

@@ -205,6 +205,26 @@ class ArchiveOrgIE(InfoExtractor):
                 },
             },
         ],
+    }, {
+        # The reviewbody is None for one of the reviews; just need to extract data without crashing
+        'url': 'https://archive.org/details/gd95-04-02.sbd.11622.sbeok.shnf/gd95-04-02d1t04.shn',
+        'info_dict': {
+            'id': 'gd95-04-02.sbd.11622.sbeok.shnf/gd95-04-02d1t04.shn',
+            'ext': 'mp3',
+            'title': 'Stuck Inside of Mobile with the Memphis Blues Again',
+            'creators': ['Grateful Dead'],
+            'duration': 338.31,
+            'track': 'Stuck Inside of Mobile with the Memphis Blues Again',
+            'description': 'md5:764348a470b986f1217ffd38d6ac7b72',
+            'display_id': 'gd95-04-02d1t04.shn',
+            'location': 'Pyramid Arena',
+            'uploader': 'jon@archive.org',
+            'album': '1995-04-02 - Pyramid Arena',
+            'upload_date': '20040519',
+            'track_number': 4,
+            'release_date': '19950402',
+            'timestamp': 1084927901,
+        },
     }]
 
     @staticmethod
@@ -335,7 +355,7 @@ class ArchiveOrgIE(InfoExtractor):
                 info['comments'].append({
                     'id': review.get('review_id'),
                     'author': review.get('reviewer'),
-                    'text': str_or_none(review.get('reviewtitle'), '') + '\n\n' + review.get('reviewbody'),
+                    'text': join_nonempty('reviewtitle', 'reviewbody', from_dict=review, delim='\n\n'),
                     'timestamp': unified_timestamp(review.get('createdate')),
                     'parent': 'root'})
 

yt_dlp/extractor/chaturbate.py

@@ -9,7 +9,7 @@ from ..utils import (
 
 
 class ChaturbateIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:[^/]+\.)?chaturbate\.com/(?:fullvideo/?\?.*?\bb=)?(?P<id>[^/?&#]+)'
+    _VALID_URL = r'https?://(?:[^/]+\.)?chaturbate\.(?P<tld>com|eu|global)/(?:fullvideo/?\?.*?\bb=)?(?P<id>[^/?&#]+)'
     _TESTS = [{
         'url': 'https://www.chaturbate.com/siswet19/',
         'info_dict': {
@@ -29,15 +29,24 @@ class ChaturbateIE(InfoExtractor):
     }, {
         'url': 'https://en.chaturbate.com/siswet19/',
         'only_matching': True,
+    }, {
+        'url': 'https://chaturbate.eu/siswet19/',
+        'only_matching': True,
+    }, {
+        'url': 'https://chaturbate.eu/fullvideo/?b=caylin',
+        'only_matching': True,
+    }, {
+        'url': 'https://chaturbate.global/siswet19/',
+        'only_matching': True,
     }]
 
     _ROOM_OFFLINE = 'Room is currently offline'
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        video_id, tld = self._match_valid_url(url).group('id', 'tld')
 
         webpage = self._download_webpage(
-            f'https://chaturbate.com/{video_id}/', video_id,
+            f'https://chaturbate.{tld}/{video_id}/', video_id,
             headers=self.geo_verification_headers())
 
         found_m3u8_urls = []

yt_dlp/extractor/facebook.py

@@ -563,13 +563,13 @@ class FacebookIE(InfoExtractor):
                 return extract_video_data(try_get(
                     js_data, lambda x: x['jsmods']['instances'], list) or [])
 
-        def extract_dash_manifest(video, formats):
+        def extract_dash_manifest(vid_data, formats, mpd_url=None):
             dash_manifest = traverse_obj(
-                video, 'dash_manifest', 'playlist', 'dash_manifest_xml_string', expected_type=str)
+                vid_data, 'dash_manifest', 'playlist', 'dash_manifest_xml_string', 'manifest_xml', expected_type=str)
             if dash_manifest:
                 formats.extend(self._parse_mpd_formats(
                     compat_etree_fromstring(urllib.parse.unquote_plus(dash_manifest)),
-                    mpd_url=url_or_none(video.get('dash_manifest_url'))))
+                    mpd_url=url_or_none(video.get('dash_manifest_url')) or mpd_url))
 
         def process_formats(info):
             # Downloads with browser's User-Agent are rate limited. Working around
@@ -619,9 +619,12 @@ class FacebookIE(InfoExtractor):
                         video = video['creation_story']
                         video['owner'] = traverse_obj(video, ('short_form_video_context', 'video_owner'))
                         video.update(reel_info)
-                    fmt_data = traverse_obj(video, ('videoDeliveryLegacyFields', {dict})) or video
+
                     formats = []
                     q = qualities(['sd', 'hd'])
+
+                    # Legacy formats extraction
+                    fmt_data = traverse_obj(video, ('videoDeliveryLegacyFields', {dict})) or video
                     for key, format_id in (('playable_url', 'sd'), ('playable_url_quality_hd', 'hd'),
                                            ('playable_url_dash', ''), ('browser_native_hd_url', 'hd'),
                                            ('browser_native_sd_url', 'sd')):
@@ -629,7 +632,7 @@ class FacebookIE(InfoExtractor):
                         if not playable_url:
                             continue
                         if determine_ext(playable_url) == 'mpd':
-                            formats.extend(self._extract_mpd_formats(playable_url, video_id))
+                            formats.extend(self._extract_mpd_formats(playable_url, video_id, fatal=False))
                         else:
                             formats.append({
                                 'format_id': format_id,
@@ -638,6 +641,28 @@ class FacebookIE(InfoExtractor):
                                 'url': playable_url,
                             })
                     extract_dash_manifest(fmt_data, formats)
+
+                    # New videoDeliveryResponse formats extraction
+                    fmt_data = traverse_obj(video, ('videoDeliveryResponseFragment', 'videoDeliveryResponseResult'))
+                    mpd_urls = traverse_obj(fmt_data, ('dash_manifest_urls', ..., 'manifest_url', {url_or_none}))
+                    dash_manifests = traverse_obj(fmt_data, ('dash_manifests', lambda _, v: v['manifest_xml']))
+                    for idx, dash_manifest in enumerate(dash_manifests):
+                        extract_dash_manifest(dash_manifest, formats, mpd_url=traverse_obj(mpd_urls, idx))
+                    if not dash_manifests:
+                        # Only extract from MPD URLs if the manifests are not already provided
+                        for mpd_url in mpd_urls:
+                            formats.extend(self._extract_mpd_formats(mpd_url, video_id, fatal=False))
+                    for prog_fmt in traverse_obj(fmt_data, ('progressive_urls', lambda _, v: v['progressive_url'])):
+                        format_id = traverse_obj(prog_fmt, ('metadata', 'quality', {str.lower}))
+                        formats.append({
+                            'format_id': format_id,
+                            # sd, hd formats w/o resolution info should be deprioritized below DASH
+                            'quality': q(format_id) - 3,
+                            'url': prog_fmt['progressive_url'],
+                        })
+                    for m3u8_url in traverse_obj(fmt_data, ('hls_playlist_urls', ..., 'hls_playlist_url', {url_or_none})):
+                        formats.extend(self._extract_m3u8_formats(m3u8_url, video_id, 'mp4', fatal=False, m3u8_id='hls'))
+
                     if not formats:
                         # Do not append false positive entry w/o any formats
                         return

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,
+    parse_resolution,
-    urljoin,
 )
+from ..utils.traversal import traverse_obj
 
 
-class RTSIE(SRGSSRIE):  # XXX: Do not subclass from concrete IE
+class RTSIE(SRGSSRIE):
-    _WORKING = False
+    _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',
+                    'duration': 1488,
-                'view_count': int,
+                    'categories': ['Divers'],
                 },
-            'expected_warnings': ['Unable to download f4m manifest', 'Failed to download m3u8 information'],
+            }],
-        },
+            'params': {'skip_download': 'm3u8'},  # 700-byte first fragment
-        {
-            '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,
         },
         {
+            # 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',
+                    'duration': 48,
-                'view_count': int,
+                    'categories': ['Hockey sur glace'],
                 },
-            'params': {
+            }],
-                # m3u8 download
+            'params': {'skip_download': 'm3u8'},  # 700-byte first fragment
-                'skip_download': True,
-            },
-            'expected_warnings': ['Unable to download f4m manifest', 'Failed to download m3u8 information'],
             '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',
+                'display_id': 'urban-hippie-de-damien-krisl',
-                'timestamp': 1396551600,
+                '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)
+        webpage, urlh = self._download_webpage_handle(url, self._match_id(url))
-        media_id = m.group('rts_id') or m.group('id')
+        if urlh.url != url:
-        display_id = m.group('display_id') or media_id
+            return self.url_result(urlh.url)
 
-        def download_json(internal_id):
+        mobj = self._match_valid_url(url)
-            return self._download_json(
+        display_id = traverse_obj(mobj, 'display_id', default=mobj.group('id')) or mobj.group('id')
-                f'http://www.rts.ch/a/{internal_id}.html?f=json/article',
-                display_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', {lambda x: media_list.append(x) if x else None}))
+        traverse_obj(article_details, ('innerMediaElements', {lambda x: media_list.extend(x)}))
+        traverse_obj(article_details, ('mediaElements', {lambda x: media_list.extend(x)}))
+        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', []):
+            if fmts := self._extract_formats(media_info, media_id):
-                item_url = item.get('url')
+                entries.append({
-                if not item_url:
+                    'id': media_info['id'],
-                    continue
+                    'title': media_info['title'],
-                entries.append(self.url_result(item_url, 'RTS'))
+                    '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:
+        return self.playlist_result(
-                page, urlh = self._download_webpage_handle(url, display_id)
+            entries, article_details.get('oid'), article_details.get('title'),
-                if re.match(self._VALID_URL, urlh.url).group('id') != media_id:
+            article_details.get('lead'), display_id=display_id,
-                    return self.url_result(urlh.url, 'RTS')
+            tags=traverse_obj(article_details, ('tags', ..., 'name')))
-
-                # 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']
 
+    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 idx, stream in enumerate(traverse_obj(
-        for format_id, format_url in streams.items():
+                media_info, ('resourceList', lambda _, v: v['url']))):
-            if format_id == 'hds_sd' and 'hds' in streams:
+            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)
-
+        return formats
-        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')),
-        }

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': {