Merge 4cf46e29c1 into 8e15177b41

Closes #9358
Authored by: jakeogh, minamotorin, shoxie007, bbilly1

Co-authored-by: minamotorin <76122224+minamotorin@users.noreply.github.com>
Co-authored-by: shoxie007 <74592022+shoxie007@users.noreply.github.com>
Co-authored-by: Simon <35427372+bbilly1@users.noreply.github.com>

README.md

@@ -666,7 +666,7 @@ ## Filesystem Options:
                                     The name of the browser to load cookies
                                     from. Currently supported browsers are:
                                     brave, chrome, chromium, edge, firefox,
-                                    opera, safari, vivaldi. Optionally, the
+                                    opera, safari, vivaldi, whale. Optionally, the
                                     KEYRING used for decrypting Chromium cookies
                                     on Linux, the name/path of the PROFILE to
                                     load cookies from, and the CONTAINER name

test/test_utils.py

@@ -5,6 +5,7 @@
 import sys
 import unittest
 import warnings
+import datetime as dt
 
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
@@ -27,6 +28,7 @@
     ExtractorError,
     InAdvancePagedList,
     LazyList,
+    NO_DEFAULT,
     OnDemandPagedList,
     Popen,
     age_restricted,
@@ -768,6 +770,11 @@ def test_encode_compat_str(self):
 
     def test_parse_iso8601(self):
         self.assertEqual(parse_iso8601('2014-03-23T23:04:26+0100'), 1395612266)
+        self.assertEqual(parse_iso8601('2014-03-23T23:04:26-07:00'), 1395641066)
+        self.assertEqual(parse_iso8601('2014-03-23T23:04:26', timezone=dt.timedelta(hours=-7)), 1395641066)
+        self.assertEqual(parse_iso8601('2014-03-23T23:04:26', timezone=NO_DEFAULT), None)
+        # default does not override timezone in date_str
+        self.assertEqual(parse_iso8601('2014-03-23T23:04:26-07:00', timezone=dt.timedelta(hours=-10)), 1395641066)
         self.assertEqual(parse_iso8601('2014-03-23T22:04:26+0000'), 1395612266)
         self.assertEqual(parse_iso8601('2014-03-23T22:04:26Z'), 1395612266)
         self.assertEqual(parse_iso8601('2014-03-23T22:04:26.1234Z'), 1395612266)

yt_dlp/cookies.py

@@ -46,7 +46,7 @@
 from .utils._utils import _YDLLogger
 from .utils.networking import normalize_url
 
-CHROMIUM_BASED_BROWSERS = {'brave', 'chrome', 'chromium', 'edge', 'opera', 'vivaldi'}
+CHROMIUM_BASED_BROWSERS = {'brave', 'chrome', 'chromium', 'edge', 'opera', 'vivaldi', 'whale'}
 SUPPORTED_BROWSERS = CHROMIUM_BASED_BROWSERS | {'firefox', 'safari'}
 
 
@@ -219,6 +219,7 @@ def _get_chromium_based_browser_settings(browser_name):
             'edge': os.path.join(appdata_local, R'Microsoft\Edge\User Data'),
             'opera': os.path.join(appdata_roaming, R'Opera Software\Opera Stable'),
             'vivaldi': os.path.join(appdata_local, R'Vivaldi\User Data'),
+            'whale': os.path.join(appdata_local, R'Naver\Naver Whale\User Data'),
         }[browser_name]
 
     elif sys.platform == 'darwin':
@@ -230,6 +231,7 @@ def _get_chromium_based_browser_settings(browser_name):
             'edge': os.path.join(appdata, 'Microsoft Edge'),
             'opera': os.path.join(appdata, 'com.operasoftware.Opera'),
             'vivaldi': os.path.join(appdata, 'Vivaldi'),
+            'whale': os.path.join(appdata, 'Naver/Whale'),
         }[browser_name]
 
     else:
@@ -241,6 +243,7 @@ def _get_chromium_based_browser_settings(browser_name):
             'edge': os.path.join(config, 'microsoft-edge'),
             'opera': os.path.join(config, 'opera'),
             'vivaldi': os.path.join(config, 'vivaldi'),
+            'whale': os.path.join(config, 'naver-whale'),
         }[browser_name]
 
     # Linux keyring names can be determined by snooping on dbus while opening the browser in KDE:
@@ -252,6 +255,7 @@ def _get_chromium_based_browser_settings(browser_name):
         'edge': 'Microsoft Edge' if sys.platform == 'darwin' else 'Chromium',
         'opera': 'Opera' if sys.platform == 'darwin' else 'Chromium',
         'vivaldi': 'Vivaldi' if sys.platform == 'darwin' else 'Chrome',
+        'whale': 'Whale',
     }[browser_name]
 
     browsers_without_profiles = {'opera'}

yt_dlp/extractor/bbc.py

@@ -602,7 +602,7 @@ class BBCIE(BBCCoUkIE):  # XXX: Do not subclass from concrete IE
         'url': 'http://www.bbc.com/news/world-europe-32668511',
         'info_dict': {
             'id': 'world-europe-32668511',
-            'title': 'Russia stages massive WW2 parade',
+            'title': 'Russia stages massive WW2 parade despite Western boycott',
             'description': 'md5:00ff61976f6081841f759a08bf78cc9c',
         },
         'playlist_count': 2,
@@ -623,6 +623,7 @@ class BBCIE(BBCCoUkIE):  # XXX: Do not subclass from concrete IE
         'info_dict': {
             'id': '3662a707-0af9-3149-963f-47bea720b460',
             'title': 'BUGGER',
+            'description': r're:BUGGER  The recent revelations by the whistleblower Edward Snowden were fascinating. .{211}\.{3}$',
         },
         'playlist_count': 18,
     }, {
@@ -631,14 +632,14 @@ class BBCIE(BBCCoUkIE):  # XXX: Do not subclass from concrete IE
         'info_dict': {
             'id': 'p02mprgb',
             'ext': 'mp4',
-            'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
-            'description': 'md5:2868290467291b37feda7863f7a83f54',
+            'title': 'Germanwings crash site aerial video',
+            'description': r're:(?s)Aerial video showed the site where the Germanwings flight 4U 9525, .{156} BFM TV\.$',
             'duration': 47,
             'timestamp': 1427219242,
             'upload_date': '20150324',
+            'thumbnail': 'https://ichef.bbci.co.uk/news/1024/media/images/81879000/jpg/_81879090_81879089.jpg',
         },
         'params': {
-            # rtmp download
             'skip_download': True,
         }
     }, {
@@ -656,21 +657,24 @@ class BBCIE(BBCCoUkIE):  # XXX: Do not subclass from concrete IE
         },
         'params': {
             'skip_download': True,
-        }
+        },
+        'skip': 'now SIMORGH_DATA with no video',
     }, {
         # single video embedded with data-playable containing XML playlists (regional section)
         'url': 'http://www.bbc.com/mundo/video_fotos/2015/06/150619_video_honduras_militares_hospitales_corrupcion_aw',
         'info_dict': {
-            'id': '150619_video_honduras_militares_hospitales_corrupcion_aw',
+            'id': '39275083',
+            'display_id': '150619_video_honduras_militares_hospitales_corrupcion_aw',
             'ext': 'mp4',
             'title': 'Honduras militariza sus hospitales por nuevo escándalo de corrupción',
-            'description': 'md5:1525f17448c4ee262b64b8f0c9ce66c8',
+            'description': 'Honduras militariza sus hospitales por nuevo escándalo de corrupción',
             'timestamp': 1434713142,
             'upload_date': '20150619',
+            'thumbnail': 'https://a.files.bbci.co.uk/worldservice/live/assets/images/2015/06/19/150619132146_honduras_hsopitales_militares_640x360_aptn_nocredit.jpg',
         },
         'params': {
             'skip_download': True,
-        }
+        },
     }, {
         # single video from video playlist embedded with vxp-playlist-data JSON
         'url': 'http://www.bbc.com/news/video_and_audio/must_see/33376376',
@@ -683,22 +687,21 @@ class BBCIE(BBCCoUkIE):  # XXX: Do not subclass from concrete IE
         },
         'params': {
             'skip_download': True,
-        }
+        },
+        'skip': '404 Not Found',
     }, {
-        # single video story with digitalData
+        # single video story with __PWA_PRELOADED_STATE__
         'url': 'http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret',
         'info_dict': {
             'id': 'p02q6gc4',
-            'ext': 'flv',
-            'title': 'Sri Lanka’s spicy secret',
-            'description': 'As a new train line to Jaffna opens up the country’s north, travellers can experience a truly distinct slice of Tamil culture.',
-            'timestamp': 1437674293,
-            'upload_date': '20150723',
+            'ext': 'mp4',
+            'title': 'Tasting the spice of life in Jaffna',
+            'description': r're:(?s)BBC Travel Show’s Henry Golding explores the city of Jaffna .{151} aftertaste\.$',
+            'timestamp': 1646058397,
+            'upload_date': '20220228',
+            'duration': 255,
+            'thumbnail': 'https://ichef.bbci.co.uk/images/ic/1920xn/p02vxvkn.jpg',
         },
-        'params': {
-            # rtmp download
-            'skip_download': True,
-        }
     }, {
         # single video story without digitalData
         'url': 'http://www.bbc.com/autos/story/20130513-hyundais-rock-star',
@@ -710,12 +713,10 @@ class BBCIE(BBCCoUkIE):  # XXX: Do not subclass from concrete IE
             'timestamp': 1415867444,
             'upload_date': '20141113',
         },
-        'params': {
-            # rtmp download
-            'skip_download': True,
-        }
+        'skip': 'redirects to TopGear home page',
     }, {
         # single video embedded with Morph
+        # TODO: replacement test page
         'url': 'http://www.bbc.co.uk/sport/live/olympics/36895975',
         'info_dict': {
             'id': 'p041vhd0',
@@ -726,27 +727,22 @@ class BBCIE(BBCCoUkIE):  # XXX: Do not subclass from concrete IE
             'uploader': 'BBC Sport',
             'uploader_id': 'bbc_sport',
         },
-        'params': {
-            # m3u8 download
-            'skip_download': True,
-        },
-        'skip': 'Georestricted to UK',
+        'skip': 'Video no longer in page',
     }, {
-        # single video with playlist.sxml URL in playlist param
+        # single video in __INITIAL_DATA__
         'url': 'http://www.bbc.com/sport/0/football/33653409',
         'info_dict': {
             'id': 'p02xycnp',
             'ext': 'mp4',
-            'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?',
-            'description': 'BBC Sport\'s David Ornstein has the latest transfer gossip, including rumours of a Manchester United return for Cristiano Ronaldo.',
+            'title': 'Ronaldo to Man Utd, Arsenal to spend?',
+            'description': r're:(?s)BBC Sport\'s David Ornstein rounds up the latest transfer reports, .{359} here\.$',
+            'timestamp': 1437750175,
+            'upload_date': '20150724',
+            'thumbnail': r're:https?://.+/.+media/images/69320000/png/_69320754_mmgossipcolumnextraaugust18.png',
             'duration': 140,
         },
-        'params': {
-            # rtmp download
-            'skip_download': True,
-        }
     }, {
-        # article with multiple videos embedded with playlist.sxml in playlist param
+        # article with multiple videos embedded with Morph.setPayload
         'url': 'http://www.bbc.com/sport/0/football/34475836',
         'info_dict': {
             'id': '34475836',
@@ -754,6 +750,21 @@ class BBCIE(BBCCoUkIE):  # XXX: Do not subclass from concrete IE
             'description': 'Fast-paced football, wit, wisdom and a ready smile - why Liverpool fans should come to love new boss Jurgen Klopp.',
         },
         'playlist_count': 3,
+    }, {
+        # Testing noplaylist
+        'url': 'http://www.bbc.com/sport/0/football/34475836',
+        'info_dict': {
+            'id': 'p034ppnv',
+            'ext': 'mp4',
+            'title': 'All you need to know about Jurgen Klopp',
+            'timestamp': 1444335081,
+            'upload_date': '20151008',
+            'duration': 122.0,
+            'thumbnail': 'https://ichef.bbci.co.uk/onesport/cps/976/cpsprodpb/7542/production/_85981003_klopp.jpg',
+        },
+        'params': {
+            'noplaylist': True,
+        },
     }, {
         # school report article with single video
         'url': 'http://www.bbc.co.uk/schoolreport/35744779',
@@ -762,6 +773,7 @@ class BBCIE(BBCCoUkIE):  # XXX: Do not subclass from concrete IE
             'title': 'School which breaks down barriers in Jerusalem',
         },
         'playlist_count': 1,
+        'skip': 'redirects to Young Reporter home page https://www.bbc.co.uk/news/topics/cg41ylwv43pt',
     }, {
         # single video with playlist URL from weather section
         'url': 'http://www.bbc.com/weather/features/33601775',
@@ -778,18 +790,33 @@ class BBCIE(BBCCoUkIE):  # XXX: Do not subclass from concrete IE
             'thumbnail': r're:https?://.+/.+\.jpg',
             'timestamp': 1437785037,
             'upload_date': '20150725',
+            'duration': 105,
         },
     }, {
         # video with window.__INITIAL_DATA__ and value as JSON string
         'url': 'https://www.bbc.com/news/av/world-europe-59468682',
         'info_dict': {
-            'id': 'p0b71qth',
+            'id': 'p0b779gc',
             'ext': 'mp4',
             'title': 'Why France is making this woman a national hero',
-            'description': 'md5:7affdfab80e9c3a1f976230a1ff4d5e4',
+            'description': r're:(?s)France is honouring the US-born 20th Century singer and activist Josephine .{208} Second World War.',
             'thumbnail': r're:https?://.+/.+\.jpg',
-            'timestamp': 1638230731,
-            'upload_date': '20211130',
+            'timestamp': 1638215626,
+            'upload_date': '20211129',
+            'duration': 125,
+        },
+    }, {
+        # video with script id __NEXT_DATA__ and value as JSON string
+        'url': 'https://www.bbc.com/news/uk-68546268',
+        'info_dict': {
+            'id': 'p0hj0lq7',
+            'ext': 'mp4',
+            'title': 'Nasser Hospital doctor describes his treatment by IDF',
+            'description': r're:(?s)Doctor Abu Sabha said he was detained by Israeli forces after .{276} hostages\."$',
+            'thumbnail': r're:https?://.+/.+\.jpg',
+            'timestamp': 1710188248,
+            'upload_date': '20240311',
+            'duration': 104,
         },
     }, {
         # single video article embedded with data-media-vpid
@@ -817,6 +844,7 @@ class BBCIE(BBCCoUkIE):  # XXX: Do not subclass from concrete IE
             'uploader': 'Radio 3',
             'uploader_id': 'bbc_radio_three',
         },
+        'skip': '404 Not Found',
     }, {
         'url': 'http://www.bbc.co.uk/learningenglish/chinese/features/lingohack/ep-181227',
         'info_dict': {
@@ -824,6 +852,7 @@ class BBCIE(BBCCoUkIE):  # XXX: Do not subclass from concrete IE
             'ext': 'mp4',
             'title': 'md5:2fabf12a726603193a2879a055f72514',
             'description': 'Learn English words and phrases from this story',
+            'thumbnail': 'https://ichef.bbci.co.uk/images/ic/1200x675/p06pq9gk.jpg',
         },
         'add_ie': [BBCCoUkIE.ie_key()],
     }, {
@@ -832,28 +861,30 @@ class BBCIE(BBCCoUkIE):  # XXX: Do not subclass from concrete IE
         'info_dict': {
             'id': 'p07c6sb9',
             'ext': 'mp4',
-            'title': 'How positive thinking is harming your happiness',
-            'alt_title': 'The downsides of positive thinking',
-            'description': 'md5:fad74b31da60d83b8265954ee42d85b4',
+            'title': 'The downsides of positive thinking',
+            'description': 'The downsides of positive thinking',
             'duration': 235,
-            'thumbnail': r're:https?://.+/p07c9dsr.jpg',
-            'upload_date': '20190604',
-            'categories': ['Psychology'],
+            'thumbnail': r're:https?://.+/p07c9dsr\.(?:jpg|webp|png)',
+            'upload_date': '20220223',
+            'timestamp': 1645632746,
         },
     }, {
         # BBC Sounds
-        'url': 'https://www.bbc.co.uk/sounds/play/m001q78b',
+        'url': 'https://www.bbc.co.uk/sounds/play/w3ct5rgx',
         'info_dict': {
-            'id': 'm001q789',
+            'id': 'p0hrw4nr',
             'ext': 'mp4',
-            'title': 'The Night Tracks Mix - Music for the darkling hour',
-            'thumbnail': 'https://ichef.bbci.co.uk/images/ic/raw/p0c00hym.jpg',
-            'chapters': 'count:8',
-            'description': 'md5:815fb51cbdaa270040aab8145b3f1d67',
-            'uploader': 'Radio 3',
-            'duration': 1800,
-            'uploader_id': 'bbc_radio_three',
-        },
+            'title': 'Are our coastlines being washed away?',
+            'description': r're:(?s)Around the world, coastlines are constantly changing .{2000,} Images\)$',
+            'timestamp': 1713556800,
+            'upload_date': '20240419',
+            'duration': 1588,
+            'thumbnail': 'https://ichef.bbci.co.uk/images/ic/raw/p0hrnxbl.jpg',
+            'uploader': 'World Service',
+            'uploader_id': 'bbc_world_service',
+            'series': 'CrowdScience',
+            'chapters': [],
+        }
     }, {  # onion routes
         'url': 'https://www.bbcnewsd73hkzno2ini43t4gblxvycyac5aw4gnv7t2rccijh7745uqd.onion/news/av/world-europe-63208576',
         'only_matching': True,
@@ -1008,8 +1039,7 @@ def _real_extract(self, url):
             webpage, 'group id', default=None)
         if group_id:
             return self.url_result(
-                'https://www.bbc.co.uk/programmes/%s' % group_id,
-                ie=BBCCoUkIE.ie_key())
+                f'https://www.bbc.co.uk/programmes/{group_id}', BBCCoUkIE)
 
         # single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
         programme_id = self._search_regex(
@@ -1069,83 +1099,133 @@ def _real_extract(self, url):
                 }
 
         # Morph based embed (e.g. http://www.bbc.co.uk/sport/live/olympics/36895975)
-        # There are several setPayload calls may be present but the video
-        # seems to be always related to the first one
-        morph_payload = self._parse_json(
-            self._search_regex(
-                r'Morph\.setPayload\([^,]+,\s*({.+?})\);',
-                webpage, 'morph payload', default='{}'),
-            playlist_id, fatal=False)
+        # Several setPayload calls may be present but the video(s)
+        # should be in one that mentions leadMedia or videoData
+        morph_payload = self._search_json(
+            r'\bMorph\s*\.\s*setPayload\s*\([^,]+,', webpage, 'morph payload', playlist_id,
+            contains_pattern=r'{(?s:(?:(?!</script>).)+(?:"leadMedia"|\\"videoData\\")\s*:.+)}',
+            default={})
         if morph_payload:
-            components = try_get(morph_payload, lambda x: x['body']['components'], list) or []
-            for component in components:
-                if not isinstance(component, dict):
-                    continue
-                lead_media = try_get(component, lambda x: x['props']['leadMedia'], dict)
-                if not lead_media:
-                    continue
-                identifiers = lead_media.get('identifiers')
-                if not identifiers or not isinstance(identifiers, dict):
-                    continue
-                programme_id = identifiers.get('vpid') or identifiers.get('playablePid')
+            for lead_media in traverse_obj(morph_payload, (
+                    'body', 'components', ..., 'props', 'leadMedia', {dict})):
+                programme_id = traverse_obj(lead_media, ('identifiers', ('vpid', 'playablePid'), {str}, any))
                 if not programme_id:
                     continue
-                title = lead_media.get('title') or self._og_search_title(webpage)
                 formats, subtitles = self._download_media_selector(programme_id)
-                description = lead_media.get('summary')
-                uploader = lead_media.get('masterBrand')
-                uploader_id = lead_media.get('mid')
-                duration = None
-                duration_d = lead_media.get('duration')
-                if isinstance(duration_d, dict):
-                    duration = parse_duration(dict_get(
-                        duration_d, ('rawDuration', 'formattedDuration', 'spokenDuration')))
                 return {
                     'id': programme_id,
-                    'title': title,
-                    'description': description,
-                    'duration': duration,
-                    'uploader': uploader,
-                    'uploader_id': uploader_id,
+                    'title': lead_media.get('title') or self._og_search_title(webpage),
+                    **traverse_obj(lead_media, {
+                        'description': ('summary', {str}),
+                        'duration': ('duration', ('rawDuration', 'formattedDuration', 'spokenDuration'), {parse_duration}),
+                        'uploader': ('masterBrand', {str}),
+                        'uploader_id': ('mid', {str}),
+                    }),
                     'formats': formats,
                     'subtitles': subtitles,
                 }
+            body = self._parse_json(traverse_obj(morph_payload, (
+                'body', 'content', 'article', 'body')), playlist_id, fatal=False)
+            for video_data in traverse_obj(body, (lambda _, v: v['videoData']['pid'], 'videoData')):
+                if video_data.get('vpid'):
+                    video_id = video_data['vpid']
+                    formats, subtitles = self._download_media_selector(video_id)
+                    entry = {
+                        'id': video_id,
+                        'formats': formats,
+                        'subtitles': subtitles,
+                    }
+                else:
+                    video_id = video_data['pid']
+                    entry = self.url_result(
+                        f'https://www.bbc.co.uk/programmes/{video_id}', BBCCoUkIE,
+                        video_id, url_transparent=True)
+                entry.update({
+                    'timestamp': traverse_obj(morph_payload, (
+                        'body', 'content', 'article', 'dateTimeInfo', 'dateTime', {parse_iso8601})
+                    ),
+                    **traverse_obj(video_data, {
+                        'thumbnail': (('iChefImage', 'image'), {url_or_none}, any),
+                        'title': (('title', 'caption'), {str}, any),
+                        'duration': ('duration', {parse_duration}),
+                    }),
+                })
+                if video_data.get('isLead') and not self._yes_playlist(playlist_id, video_id):
+                    return entry
+                entries.append(entry)
+            if entries:
+                playlist_title = traverse_obj(morph_payload, (
+                    'body', 'content', 'article', 'headline', {str})) or playlist_title
+                return self.playlist_result(
+                    entries, playlist_id, playlist_title, playlist_description)
 
-        preload_state = self._parse_json(self._search_regex(
-            r'window\.__PRELOADED_STATE__\s*=\s*({.+?});', webpage,
-            'preload state', default='{}'), playlist_id, fatal=False)
-        if preload_state:
-            current_programme = preload_state.get('programmes', {}).get('current') or {}
-            programme_id = current_programme.get('id')
-            if current_programme and programme_id and current_programme.get('type') == 'playable_item':
-                title = current_programme.get('titles', {}).get('tertiary') or playlist_title
-                formats, subtitles = self._download_media_selector(programme_id)
-                synopses = current_programme.get('synopses') or {}
-                network = current_programme.get('network') or {}
-                duration = int_or_none(
-                    current_programme.get('duration', {}).get('value'))
-                thumbnail = None
-                image_url = current_programme.get('image_url')
-                if image_url:
-                    thumbnail = image_url.replace('{recipe}', 'raw')
+        # various PRELOADED_STATE JSON
+        preload_state = self._search_json(
+            r'window\.__(?:PWA_)?PRELOADED_STATE__\s*=', webpage,
+            'preload state', playlist_id, transform_source=js_to_json, default={})
+        # PRELOADED_STATE with current programmme
+        current_programme = traverse_obj(preload_state, ('programmes', 'current', {dict}))
+        programme_id = traverse_obj(current_programme, ('id', {str}))
+        if programme_id and current_programme.get('type') == 'playable_item':
+            title = traverse_obj(current_programme, ('titles', ('tertiary', 'secondary'), {str}, any)) or playlist_title
+            formats, subtitles = self._download_media_selector(programme_id)
+            return {
+                'id': programme_id,
+                'title': title,
+                'formats': formats,
+                **traverse_obj(current_programme, {
+                    'description': ('synopses', ('long', 'medium', 'short'), {str}, any),
+                    'thumbnail': ('image_url', {lambda u: url_or_none(u.replace('{recipe}', 'raw'))}),
+                    'duration': ('duration', 'value', {int_or_none}),
+                    'uploader': ('network', 'short_title', {str}),
+                    'uploader_id': ('network', 'id', {str}),
+                    'timestamp': ((('availability', 'from'), ('release', 'date')), {parse_iso8601}, any),
+                    'series': ('titles', 'primary', {str}),
+                }),
+                'subtitles': subtitles,
+                'chapters': traverse_obj(preload_state, (
+                    'tracklist', 'tracks', lambda _, v: float(v['offset']['start']), {
+                        'title': ('titles', {lambda x: join_nonempty(
+                            'primary', 'secondary', 'tertiary', delim=' - ', from_dict=x)}),
+                        'start_time': ('offset', 'start', {float_or_none}),
+                        'end_time': ('offset', 'end', {float_or_none}),
+                    })
+                ),
+            }
+
+        # PWA_PRELOADED_STATE with article video asset
+        asset_id = traverse_obj(preload_state, (
+            'entities', 'articles', lambda k, _: k.rsplit('/', 1)[-1] == playlist_id,
+            'assetVideo', 0, {str}, any))
+        if asset_id:
+            video_id = traverse_obj(preload_state, ('entities', 'videos', asset_id, 'vpid', {str}))
+            if video_id:
+                article = traverse_obj(preload_state, (
+                    'entities', 'articles', lambda _, v: v['assetVideo'][0] == asset_id, any))
+
+                def image_url(image_id):
+                    return traverse_obj(preload_state, (
+                        'entities', 'images', image_id, 'url',
+                        {lambda u: url_or_none(u.replace('$recipe', 'raw'))}))
+
+                formats, subtitles = self._download_media_selector(video_id)
                 return {
-                    'id': programme_id,
-                    'title': title,
-                    'description': dict_get(synopses, ('long', 'medium', 'short')),
-                    'thumbnail': thumbnail,
-                    'duration': duration,
-                    'uploader': network.get('short_title'),
-                    'uploader_id': network.get('id'),
+                    'id': video_id,
+                    **traverse_obj(preload_state, ('entities', 'videos', asset_id, {
+                        'title': ('title', {str}),
+                        'description': (('synopsisLong', 'synopsisMedium', 'synopsisShort'), {str}, any),
+                        'thumbnail': (0, {image_url}),
+                        'duration': ('duration', {int_or_none}),
+                    })),
                     'formats': formats,
                     'subtitles': subtitles,
-                    'chapters': traverse_obj(preload_state, (
-                        'tracklist', 'tracks', lambda _, v: float_or_none(v['offset']['start']), {
-                            'title': ('titles', {lambda x: join_nonempty(
-                                'primary', 'secondary', 'tertiary', delim=' - ', from_dict=x)}),
-                            'start_time': ('offset', 'start', {float_or_none}),
-                            'end_time': ('offset', 'end', {float_or_none}),
-                        })) or None,
+                    'timestamp': traverse_obj(article, ('displayDate', {parse_iso8601})),
                 }
+            else:
+                return self.url_result(
+                    f'https://www.bbc.co.uk/programmes/{asset_id}', BBCCoUkIE,
+                    asset_id, playlist_title, display_id=playlist_id,
+                    description=playlist_description)
 
         bbc3_config = self._parse_json(
             self._search_regex(
@@ -1191,6 +1271,28 @@ def _real_extract(self, url):
                 return self.playlist_result(
                     entries, playlist_id, playlist_title, playlist_description)
 
+        def parse_model(model):
+            """Extract single video from model structure"""
+            item_id = traverse_obj(model, ('versions', 0, 'versionId', {str}))
+            if not item_id:
+                return
+            formats, subtitles = self._download_media_selector(item_id)
+            return {
+                'id': item_id,
+                'formats': formats,
+                'subtitles': subtitles,
+                **traverse_obj(model, {
+                    'title': ('title', {str}),
+                    'thumbnail': ('imageUrl', {lambda u: urljoin(url, u.replace('$recipe', 'raw'))}),
+                    'description': ('synopses', ('long', 'medium', 'short'), {str}, {lambda x: x or None}, any),
+                    'duration': ('versions', 0, 'duration', {int}),
+                    'timestamp': ('versions', 0, 'availableFrom', {functools.partial(int_or_none, scale=1000)}),
+                })
+            }
+
+        def is_type(*types):
+            return lambda _, v: v['type'] in types
+
         initial_data = self._search_regex(
             r'window\.__INITIAL_DATA__\s*=\s*("{.+?}")\s*;', webpage,
             'quoted preload state', default=None)
@@ -1202,6 +1304,19 @@ def _real_extract(self, url):
             initial_data = self._parse_json(initial_data or '"{}"', playlist_id, fatal=False)
         initial_data = self._parse_json(initial_data, playlist_id, fatal=False)
         if initial_data:
+            for video_data in traverse_obj(initial_data, (
+                    'stores', 'article', 'articleBodyContent', is_type('video'))):
+                model = traverse_obj(video_data, (
+                    'model', 'blocks', is_type('aresMedia'),
+                    'model', 'blocks', is_type('aresMediaMetadata'),
+                    'model', {dict}, any))
+                entry = parse_model(model)
+                if entry:
+                    entries.append(entry)
+            if entries:
+                return self.playlist_result(
+                    entries, playlist_id, playlist_title, playlist_description)
+
             def parse_media(media):
                 if not media:
                     return
@@ -1234,27 +1349,90 @@ def parse_media(media):
                         'subtitles': subtitles,
                         'timestamp': item_time,
                         'description': strip_or_none(item_desc),
+                        'duration': int_or_none(item.get('duration')),
                     })
-            for resp in (initial_data.get('data') or {}).values():
-                name = resp.get('name')
+
+            for resp in traverse_obj(initial_data, ('data', lambda _, v: v['name'])):
+                name = resp['name']
                 if name == 'media-experience':
                     parse_media(try_get(resp, lambda x: x['data']['initialItem']['mediaItem'], dict))
                 elif name == 'article':
-                    for block in (try_get(resp,
-                                          (lambda x: x['data']['blocks'],
-                                           lambda x: x['data']['content']['model']['blocks'],),
-                                          list) or []):
-                        if block.get('type') not in ['media', 'video']:
-                            continue
-                        parse_media(block.get('model'))
+                    for block in traverse_obj(resp, (
+                            'data', (None, ('content', 'model')), 'blocks',
+                            is_type('media', 'video'), 'model', {dict})):
+                        parse_media(block)
             return self.playlist_result(
                 entries, playlist_id, playlist_title, playlist_description)
 
+        # extract from SIMORGH_DATA hydration JSON
+        simorgh_data = self._search_json(
+            r'window\s*\.\s*SIMORGH_DATA\s*=', webpage,
+            'simorgh data', playlist_id, default={})
+        if simorgh_data:
+            done = False
+            for video_data in traverse_obj(simorgh_data, (
+                    'pageData', 'content', 'model', 'blocks', is_type('video', 'legacyMedia'))):
+                model = traverse_obj(video_data, (
+                    'model', 'blocks', is_type('aresMedia'),
+                    'model', 'blocks', is_type('aresMediaMetadata'),
+                    'model', {dict}, any))
+                if video_data['type'] == 'video':
+                    entry = parse_model(model)
+                else:  # legacyMedia: no duration, subtitles
+                    block_id, entry = traverse_obj(model, ('blockId', {str})), None
+                    media_data = traverse_obj(simorgh_data, (
+                        'pageData', 'promo', 'media',
+                        {lambda x: x if x['id'] == block_id else None}))
+                    formats = traverse_obj(media_data, ('playlist', lambda _, v: url_or_none(v['url']), {
+                        'url': ('url', {url_or_none}),
+                        'ext': ('format', {str}),
+                        'tbr': ('bitrate', {functools.partial(int_or_none, scale=1000)}),
+                    }))
+                    if formats:
+                        entry = {
+                            'id': block_id,
+                            'display_id': playlist_id,
+                            'formats': formats,
+                            'description': traverse_obj(simorgh_data, ('pageData', 'promo', 'summary', {str})),
+                            **traverse_obj(model, {
+                                'title': ('title', {str}),
+                                'thumbnail': ('imageUrl', {lambda u: urljoin(url, u.replace('$recipe', 'raw'))}),
+                                'description': ('synopses', ('long', 'medium', 'short'), {str}, any),
+                                'timestamp': ('firstPublished', {functools.partial(int_or_none, scale=1000)}),
+                            }),
+                        }
+                        done = True
+                if entry:
+                    entries.append(entry)
+                if done:
+                    break
+            if entries:
+                return self.playlist_result(
+                    entries, playlist_id, playlist_title, playlist_description)
+
         def extract_all(pattern):
             return list(filter(None, map(
                 lambda s: self._parse_json(s, playlist_id, fatal=False),
                 re.findall(pattern, webpage))))
 
+        # US accessed article with single embedded video (e.g.
+        # https://www.bbc.com/news/uk-68546268)
+        next_data = traverse_obj(self._search_nextjs_data(webpage, playlist_id, default={}),
+                                 ('props', 'pageProps', 'page'))
+        model = traverse_obj(next_data, (
+            ..., 'contents', is_type('video'),
+            'model', 'blocks', is_type('media'),
+            'model', 'blocks', is_type('mediaMetadata'),
+            'model', {dict}, any))
+        if model and (entry := parse_model(model)):
+            if not entry.get('timestamp'):
+                entry['timestamp'] = traverse_obj(next_data, (
+                    ..., 'contents', is_type('timestamp'), 'model',
+                    'timestamp', {functools.partial(int_or_none, scale=1000)}, any))
+            entries.append(entry)
+            return self.playlist_result(
+                entries, playlist_id, playlist_title, playlist_description)
+
         # Multiple video article (e.g.
         # http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460)
         EMBED_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:[^/]+/)+%s(?:\b[^"]+)?' % self._ID_REGEX

yt_dlp/extractor/common.py

@@ -957,7 +957,8 @@ def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=
         if urlh is False:
             assert not fatal
             return False
-        content = self._webpage_read_content(urlh, url_or_request, video_id, note, errnote, fatal, encoding=encoding)
+        content = self._webpage_read_content(urlh, url_or_request, video_id, note, errnote, fatal,
+                                             encoding=encoding, data=data)
         return (content, urlh)
 
     @staticmethod
@@ -1005,8 +1006,10 @@ def __check_blocked(self, content):
                 'Visit http://blocklist.rkn.gov.ru/ for a block reason.',
                 expected=True)
 
-    def _request_dump_filename(self, url, video_id):
-        basen = f'{video_id}_{url}'
+    def _request_dump_filename(self, url, video_id, data=None):
+        if data is not None:
+            data = hashlib.md5(data).hexdigest()
+        basen = join_nonempty(video_id, data, url, delim='_')
         trim_length = self.get_param('trim_file_name') or 240
         if len(basen) > trim_length:
             h = '___' + hashlib.md5(basen.encode('utf-8')).hexdigest()
@@ -1028,16 +1031,18 @@ def __decode_webpage(self, webpage_bytes, encoding, headers):
         except LookupError:
             return webpage_bytes.decode('utf-8', 'replace')
 
-    def _webpage_read_content(self, urlh, url_or_request, video_id, note=None, errnote=None, fatal=True, prefix=None, encoding=None):
+    def _webpage_read_content(self, urlh, url_or_request, video_id, note=None, errnote=None, fatal=True,
+                              prefix=None, encoding=None, data=None):
         webpage_bytes = urlh.read()
         if prefix is not None:
             webpage_bytes = prefix + webpage_bytes
+        url_or_request = self._create_request(url_or_request, data)
         if self.get_param('dump_intermediate_pages', False):
             self.to_screen('Dumping request to ' + urlh.url)
             dump = base64.b64encode(webpage_bytes).decode('ascii')
             self._downloader.to_screen(dump)
         if self.get_param('write_pages'):
-            filename = self._request_dump_filename(urlh.url, video_id)
+            filename = self._request_dump_filename(urlh.url, video_id, url_or_request.data)
             self.to_screen(f'Saving request to {filename}')
             with open(filename, 'wb') as outf:
                 outf.write(webpage_bytes)
@@ -1098,7 +1103,7 @@ def download_content(self, url_or_request, video_id, note=note, errnote=errnote,
                              impersonate=None, require_impersonation=False):
             if self.get_param('load_pages'):
                 url_or_request = self._create_request(url_or_request, data, headers, query)
-                filename = self._request_dump_filename(url_or_request.url, video_id)
+                filename = self._request_dump_filename(url_or_request.url, video_id, url_or_request.data)
                 self.to_screen(f'Loading request from {filename}')
                 try:
                     with open(filename, 'rb') as dumpf:

yt_dlp/extractor/youtube.py

@@ -1325,6 +1325,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
                 'uploader_id': '@PhilippHagemeister',
                 'heatmap': 'count:100',
+                'timestamp': 1349198244,
             }
         },
         {
@@ -1368,6 +1369,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
                 'uploader_id': '@PhilippHagemeister',
                 'heatmap': 'count:100',
+                'timestamp': 1349198244,
             },
             'params': {
                 'skip_download': True,
@@ -1454,6 +1456,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'comment_count': int,
                 'channel_is_verified': True,
                 'heatmap': 'count:100',
+                'timestamp': 1401991663,
             },
         },
         {
@@ -1513,6 +1516,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader': 'Projekt Melody',
                 'uploader_url': 'https://www.youtube.com/@ProjektMelody',
                 'uploader_id': '@ProjektMelody',
+                'timestamp': 1577508724,
             },
         },
         {
@@ -1618,6 +1622,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader_url': 'https://www.youtube.com/@Olympics',
                 'uploader_id': '@Olympics',
                 'channel_is_verified': True,
+                'timestamp': 1440707674,
             },
             'params': {
                 'skip_download': 'requires avconv',
@@ -1651,6 +1656,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader': '孫ᄋᄅ',
                 'uploader_url': 'https://www.youtube.com/@AllenMeow',
                 'uploader_id': '@AllenMeow',
+                'timestamp': 1299776999,
             },
         },
         # url_encoded_fmt_stream_map is empty string
@@ -1794,6 +1800,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 },
             }],
             'params': {'skip_download': True},
+            'skip': 'Not multifeed anymore',
         },
         {
             # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
@@ -1902,6 +1909,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader': 'The Berkman Klein Center for Internet & Society',
                 'uploader_id': '@BKCHarvard',
                 'uploader_url': 'https://www.youtube.com/@BKCHarvard',
+                'timestamp': 1422422076,
             },
             'params': {
                 'skip_download': True,
@@ -1937,6 +1945,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader_id': '@BernieSanders',
                 'channel_is_verified': True,
                 'heatmap': 'count:100',
+                'timestamp': 1447987198,
             },
             'params': {
                 'skip_download': True,
@@ -2000,6 +2009,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader_id': '@Vsauce',
                 'comment_count': int,
                 'channel_is_verified': True,
+                'timestamp': 1484761047,
             },
             'params': {
                 'skip_download': True,
@@ -2155,6 +2165,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader': 'l\'Or Vert asbl',
                 'uploader_url': 'https://www.youtube.com/@ElevageOrVert',
                 'uploader_id': '@ElevageOrVert',
+                'timestamp': 1497343210,
             },
             'params': {
                 'skip_download': True,
@@ -2193,6 +2204,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader_id': '@Csharp-video-tutorialsBlogspot',
                 'channel_is_verified': True,
                 'heatmap': 'count:100',
+                'timestamp': 1377976349,
             },
             'params': {
                 'skip_download': True,
@@ -2275,6 +2287,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader_id': '@CBSMornings',
                 'comment_count': int,
                 'channel_is_verified': True,
+                'timestamp': 1405513526,
             }
         },
         {
@@ -2292,7 +2305,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'view_count': int,
                 'channel': 'Walk around Japan',
                 'tags': ['Ueno Tokyo', 'Okachimachi Tokyo', 'Ameyoko Street', 'Tokyo attraction', 'Travel in Tokyo'],
-                'thumbnail': 'https://i.ytimg.com/vi_webp/cBvYw8_A0vQ/hqdefault.webp',
+                'thumbnail': 'https://i.ytimg.com/vi/cBvYw8_A0vQ/hqdefault.jpg',
                 'age_limit': 0,
                 'availability': 'public',
                 'channel_url': 'https://www.youtube.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
@@ -2302,6 +2315,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader': 'Walk around Japan',
                 'uploader_url': 'https://www.youtube.com/@walkaroundjapan7124',
                 'uploader_id': '@walkaroundjapan7124',
+                'timestamp': 1605884416,
             },
             'params': {
                 'skip_download': True,
@@ -2396,6 +2410,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'comment_count': int,
                 'channel_is_verified': True,
                 'heatmap': 'count:100',
+                'timestamp': 1395685455,
             }, 'params': {'format': 'mhtml', 'skip_download': True}
         }, {
             # Ensure video upload_date is in UTC timezone (video was uploaded 1641170939)
@@ -2425,9 +2440,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader_url': 'https://www.youtube.com/@LeonNguyen',
                 'uploader_id': '@LeonNguyen',
                 'heatmap': 'count:100',
+                'timestamp': 1641170939,
             }
         }, {
             # Same video as above, but with --compat-opt no-youtube-prefer-utc-upload-date
+            # todo: remove compat opt? no longer works
             'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
             'info_dict': {
                 'id': '2NUZ8W2llS4',
@@ -2487,38 +2504,41 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'comment_count': int,
                 'channel_is_verified': True,
                 'heatmap': 'count:100',
+                'timestamp': 1641172509,
             }
         },
-        {   # continuous livestream. Microformat upload date should be preferred.
-            # Upload date was 2021-06-19 (not UTC), while stream start is 2021-11-27
-            'url': 'https://www.youtube.com/watch?v=kgx4WGK0oNU',
+        {   # continuous livestream.
+            # Upload date was 2022-07-12T05:12:29-07:00, while stream start is 2022-07-12T15:59:30+00:00
+            'url': 'https://www.youtube.com/watch?v=jfKfPfyJRdk',
             'info_dict': {
-                'id': 'kgx4WGK0oNU',
-                'title': r're:jazz\/lofi hip hop radio🌱chill beats to relax\/study to \[LIVE 24\/7\] \d{4}-\d{2}-\d{2} \d{2}:\d{2}',
+                'id': 'jfKfPfyJRdk',
                 'ext': 'mp4',
-                'channel_id': 'UC84whx2xxsiA1gXHXXqKGOA',
-                'availability': 'public',
-                'age_limit': 0,
-                'release_timestamp': 1637975704,
-                'upload_date': '20210619',
-                'channel_url': 'https://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
-                'live_status': 'is_live',
-                'thumbnail': 'https://i.ytimg.com/vi/kgx4WGK0oNU/maxresdefault.jpg',
-                'channel': 'Abao in Tokyo',
-                'channel_follower_count': int,
-                'release_date': '20211127',
-                'tags': 'count:39',
-                'categories': ['People & Blogs'],
+                'channel_id': 'UCSJ4gkVC6NrvII8umztf0Ow',
                 'like_count': int,
-                'view_count': int,
-                'playable_in_embed': True,
-                'description': 'md5:2ef1d002cad520f65825346e2084e49d',
+                'uploader': 'Lofi Girl',
+                'categories': ['Music'],
                 'concurrent_view_count': int,
-                'uploader': 'Abao in Tokyo',
-                'uploader_url': 'https://www.youtube.com/@abaointokyo',
-                'uploader_id': '@abaointokyo',
+                'playable_in_embed': True,
+                'timestamp': 1657627949,
+                'release_date': '20220712',
+                'channel_url': 'https://www.youtube.com/channel/UCSJ4gkVC6NrvII8umztf0Ow',
+                'description': 'md5:13a6f76df898f5674f9127139f3df6f7',
+                'age_limit': 0,
+                'thumbnail': 'https://i.ytimg.com/vi/jfKfPfyJRdk/maxresdefault.jpg',
+                'release_timestamp': 1657641570,
+                'uploader_url': 'https://www.youtube.com/@LofiGirl',
+                'channel_follower_count': int,
+                'channel_is_verified': True,
+                'title': r're:^lofi hip hop radio 📚 - beats to relax/study to',
+                'view_count': int,
+                'live_status': 'is_live',
+                'tags': 'count:32',
+                'channel': 'Lofi Girl',
+                'availability': 'public',
+                'upload_date': '20220712',
+                'uploader_id': '@LofiGirl',
             },
-            'params': {'skip_download': True}
+            'params': {'skip_download': True},
         }, {
             'url': 'https://www.youtube.com/watch?v=tjjjtzRLHvA',
             'info_dict': {
@@ -2544,6 +2564,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader_id': '@lesmiscore',
                 'uploader': 'Lesmiscore',
                 'uploader_url': 'https://www.youtube.com/@lesmiscore',
+                'timestamp': 1648005313,
             }
         }, {
             # Prefer primary title+description language metadata by default
@@ -2571,6 +2592,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader_url': 'https://www.youtube.com/@coletdjnz',
                 'uploader_id': '@coletdjnz',
                 'uploader': 'cole-dlp-test-acc',
+                'timestamp': 1662677394,
             },
             'params': {'skip_download': True}
         }, {
@@ -2584,7 +2606,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'duration': 5,
                 'live_status': 'not_live',
                 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
-                'upload_date': '20220728',
+                'upload_date': '20220729',
                 'view_count': int,
                 'categories': ['People & Blogs'],
                 'thumbnail': r're:^https?://.*\.jpg',
@@ -2597,6 +2619,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader_url': 'https://www.youtube.com/@coletdjnz',
                 'uploader_id': '@coletdjnz',
                 'uploader': 'cole-dlp-test-acc',
+                'timestamp': 1659073275,
+                'like_count': int,
             },
             'params': {'skip_download': True, 'extractor_args': {'youtube': {'lang': ['fr']}}},
             'expected_warnings': [r'Preferring "fr" translated fields'],
@@ -2662,6 +2686,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader': 'Projekt Melody',
                 'uploader_id': '@ProjektMelody',
                 'uploader_url': 'https://www.youtube.com/@ProjektMelody',
+                'timestamp': 1577508724,
             },
             'params': {'extractor_args': {'youtube': {'player_client': ['tv_embedded']}}, 'format': '251-drc'},
         },
@@ -2696,6 +2721,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader_id': '@sana_natori',
                 'channel_is_verified': True,
                 'heatmap': 'count:100',
+                'timestamp': 1671798112,
             },
         },
         {
@@ -2765,6 +2791,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader_url': 'https://www.youtube.com/@ChristopherSykesDocumentaries',
                 'uploader_id': '@ChristopherSykesDocumentaries',
                 'heatmap': 'count:100',
+                'timestamp': 1211825920,
             },
             'params': {
                 'skip_download': True,
@@ -3317,7 +3344,36 @@ def _extract_heatmap(self, data):
                 'value': ('intensityScoreNormalized', {float_or_none}),
             })) or None
 
-    def _extract_comment(self, comment_renderer, parent=None):
+    def _extract_comment(self, entities, parent=None):
+        comment_entity_payload = get_first(entities, ('payload', 'commentEntityPayload', {dict}))
+        if not (comment_id := traverse_obj(comment_entity_payload, ('properties', 'commentId', {str}))):
+            return
+
+        toolbar_entity_payload = get_first(entities, ('payload', 'engagementToolbarStateEntityPayload', {dict}))
+        time_text = traverse_obj(comment_entity_payload, ('properties', 'publishedTime', {str})) or ''
+
+        return {
+            'id': comment_id,
+            'parent': parent or 'root',
+            **traverse_obj(comment_entity_payload, {
+                'text': ('properties', 'content', 'content', {str}),
+                'like_count': ('toolbar', 'likeCountA11y', {parse_count}),
+                'author_id': ('author', 'channelId', {self.ucid_or_none}),
+                'author': ('author', 'displayName', {str}),
+                'author_thumbnail': ('author', 'avatarThumbnailUrl', {url_or_none}),
+                'author_is_uploader': ('author', 'isCreator', {bool}),
+                'author_is_verified': ('author', 'isVerified', {bool}),
+                'author_url': ('author', 'channelCommand', 'innertubeCommand', (
+                    ('browseEndpoint', 'canonicalBaseUrl'), ('commandMetadata', 'webCommandMetadata', 'url')
+                ), {lambda x: urljoin('https://www.youtube.com', x)}),
+            }, get_all=False),
+            'is_favorited': (None if toolbar_entity_payload is None else
+                             toolbar_entity_payload.get('heartState') == 'TOOLBAR_HEART_STATE_HEARTED'),
+            '_time_text': time_text,  # FIXME: non-standard, but we need a way of showing that it is an estimate.
+            'timestamp': self._parse_time_text(time_text),
+        }
+
+    def _extract_comment_old(self, comment_renderer, parent=None):
         comment_id = comment_renderer.get('commentId')
         if not comment_id:
             return
@@ -3398,21 +3454,39 @@ def extract_header(contents):
                 break
             return _continuation
 
-        def extract_thread(contents):
+        def extract_thread(contents, entity_payloads):
             if not parent:
                 tracker['current_page_thread'] = 0
             for content in contents:
                 if not parent and tracker['total_parent_comments'] >= max_parents:
                     yield
                 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
-                comment_renderer = get_first(
-                    (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]],
-                    expected_type=dict, default={})
 
-                comment = self._extract_comment(comment_renderer, parent)
+                # old comment format
+                if not entity_payloads:
+                    comment_renderer = get_first(
+                        (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]],
+                        expected_type=dict, default={})
+
+                    comment = self._extract_comment_old(comment_renderer, parent)
+
+                # new comment format
+                else:
+                    view_model = (
+                        traverse_obj(comment_thread_renderer, ('commentViewModel', 'commentViewModel', {dict}))
+                        or traverse_obj(content, ('commentViewModel', {dict})))
+                    comment_keys = traverse_obj(view_model, (('commentKey', 'toolbarStateKey'), {str}))
+                    if not comment_keys:
+                        continue
+                    entities = traverse_obj(entity_payloads, lambda _, v: v['entityKey'] in comment_keys)
+                    comment = self._extract_comment(entities, parent)
+                    if comment:
+                        comment['is_pinned'] = traverse_obj(view_model, ('pinnedText', {str})) is not None
+
                 if not comment:
                     continue
                 comment_id = comment['id']
+
                 if comment.get('is_pinned'):
                     tracker['pinned_comment_ids'].add(comment_id)
                 # Sometimes YouTube may break and give us infinite looping comments.
@@ -3505,7 +3579,7 @@ def extract_thread(contents):
             check_get_keys = None
             if not is_forced_continuation and not (tracker['est_total'] == 0 and tracker['running_total'] == 0):
                 check_get_keys = [[*continuation_items_path, ..., (
-                    'commentsHeaderRenderer' if is_first_continuation else ('commentThreadRenderer', 'commentRenderer'))]]
+                    'commentsHeaderRenderer' if is_first_continuation else ('commentThreadRenderer', 'commentViewModel', 'commentRenderer'))]]
             try:
                 response = self._extract_response(
                     item_id=None, query=continuation,
@@ -3529,6 +3603,7 @@ def extract_thread(contents):
                 raise
             is_forced_continuation = False
             continuation = None
+            mutations = traverse_obj(response, ('frameworkUpdates', 'entityBatchUpdate', 'mutations', ..., {dict}))
             for continuation_items in traverse_obj(response, continuation_items_path, expected_type=list, default=[]):
                 if is_first_continuation:
                     continuation = extract_header(continuation_items)
@@ -3537,7 +3612,7 @@ def extract_thread(contents):
                         break
                     continue
 
-                for entry in extract_thread(continuation_items):
+                for entry in extract_thread(continuation_items, mutations):
                     if not entry:
                         return
                     yield entry
@@ -4562,19 +4637,35 @@ def process_language(container, base_url, lang_code, sub_name, query):
             'uploader_id': channel_handle,
             'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
         })
+
+        # We only want timestamp IF it has time precision AND a timezone
+        # Currently the uploadDate in microformats appears to be in US/Pacific timezone.
+        timestamp = (
+            parse_iso8601(get_first(microformats, 'uploadDate'), timezone=NO_DEFAULT)
+            or parse_iso8601(search_meta('uploadDate'), timezone=NO_DEFAULT)
+        )
+        upload_date = (
+            dt.datetime.fromtimestamp(timestamp, dt.timezone.utc).strftime('%Y%m%d') if timestamp else
+            (
+                unified_strdate(get_first(microformats, 'uploadDate'))
+                or unified_strdate(search_meta('uploadDate'))
+            ))
+
+        # In the case we cannot get the timestamp:
         # The upload date for scheduled, live and past live streams / premieres in microformats
         # may be different from the stream date. Although not in UTC, we will prefer it in this case.
         # See: https://github.com/yt-dlp/yt-dlp/pull/2223#issuecomment-1008485139
-        upload_date = (
-            unified_strdate(get_first(microformats, 'uploadDate'))
-            or unified_strdate(search_meta('uploadDate')))
         if not upload_date or (
-            live_status in ('not_live', None)
+            not timestamp
+            and live_status in ('not_live', None)
             and 'no-youtube-prefer-utc-upload-date' not in self.get_param('compat_opts', [])
         ):
+            # this should be in UTC, as configured in the cookie/client context
             upload_date = strftime_or_none(
                 self._parse_time_text(self._get_text(vpir, 'dateText'))) or upload_date
+
         info['upload_date'] = upload_date
+        info['timestamp'] = timestamp
 
         if upload_date and live_status not in ('is_live', 'post_live', 'is_upcoming'):
             # Newly uploaded videos' HLS formats are potentially problematic and need to be checked

yt_dlp/utils/_utils.py

@@ -1134,7 +1134,7 @@ def is_path_like(f):
     return isinstance(f, (str, bytes, os.PathLike))
 
 
-def extract_timezone(date_str):
+def extract_timezone(date_str, default=None):
     m = re.search(
         r'''(?x)
             ^.{8,}?                                              # >=8 char non-TZ prefix, if present
@@ -1146,21 +1146,25 @@ def extract_timezone(date_str):
                 (?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2})       # hh[:]mm
             $)
         ''', date_str)
+    timezone = None
+
     if not m:
         m = re.search(r'\d{1,2}:\d{1,2}(?:\.\d+)?(?P<tz>\s*[A-Z]+)$', date_str)
         timezone = TIMEZONE_NAMES.get(m and m.group('tz').strip())
         if timezone is not None:
             date_str = date_str[:-len(m.group('tz'))]
-        timezone = dt.timedelta(hours=timezone or 0)
+            timezone = dt.timedelta(hours=timezone)
     else:
         date_str = date_str[:-len(m.group('tz'))]
-        if not m.group('sign'):
-            timezone = dt.timedelta()
-        else:
+        if m.group('sign'):
             sign = 1 if m.group('sign') == '+' else -1
             timezone = dt.timedelta(
                 hours=sign * int(m.group('hours')),
                 minutes=sign * int(m.group('minutes')))
+
+    if timezone is None and default is not NO_DEFAULT:
+        timezone = default or dt.timedelta()
+
     return timezone, date_str
 
 
@@ -1172,10 +1176,9 @@ def parse_iso8601(date_str, delimiter='T', timezone=None):
 
     date_str = re.sub(r'\.[0-9]+', '', date_str)
 
-    if timezone is None:
-        timezone, date_str = extract_timezone(date_str)
+    timezone, date_str = extract_timezone(date_str, timezone)
 
-    with contextlib.suppress(ValueError):
+    with contextlib.suppress(ValueError, TypeError):
         date_format = f'%Y-%m-%d{delimiter}%H:%M:%S'
         dt_ = dt.datetime.strptime(date_str, date_format) - timezone
         return calendar.timegm(dt_.timetuple())