Merge 9de472f6ea into 12d8ea8246

Closes #9554
Authored by: coletdjnz, bashonly

Co-authored-by: bashonly <88596187+bashonly@users.noreply.github.com>

README.md

@@ -613,8 +613,7 @@ ## Filesystem Options:
     --no-restrict-filenames         Allow Unicode characters, "&" and spaces in
                                     filenames (default)
     --windows-filenames             Force filenames to be Windows-compatible
-    --no-windows-filenames          Make filenames Windows-compatible only if
-                                    using Windows (default)
+    --no-windows-filenames          Sanitize filenames only minimally
     --trim-filenames LENGTH         Limit the filename length (excluding
                                     extension) to the specified number of
                                     characters
@@ -666,7 +665,7 @@ ## 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
@@ -1760,7 +1759,7 @@ # EXTRACTOR ARGUMENTS
 #### youtube
 * `lang`: Prefer translated metadata (`title`, `description` etc) of this language code (case-sensitive). By default, the video primary language metadata is preferred, with a fallback to `en` translated. See [youtube.py](https://github.com/yt-dlp/yt-dlp/blob/c26f9b991a0681fd3ea548d535919cec1fbbd430/yt_dlp/extractor/youtube.py#L381-L390) for list of supported content language codes
 * `skip`: One or more of `hls`, `dash` or `translated_subs` to skip extraction of the m3u8 manifests, dash manifests and [auto-translated subtitles](https://github.com/yt-dlp/yt-dlp/issues/4090#issuecomment-1158102032) respectively
-* `player_client`: Clients to extract video data from. The main clients are `web`, `android` and `ios` with variants `_music`, `_embedded`, `_embedscreen`, `_creator` (e.g. `web_embedded`); and `mweb`, `mweb_embedscreen`, `mediaconnect` and `tv_embedded` (agegate bypass) with no variants. By default, `ios,android,web` is used, but `tv_embedded` and `creator` variants are added as required for age-gated videos. Similarly, the music variants are added for `music.youtube.com` urls. You can use `all` to use all the clients, and `default` for the default clients.
+* `player_client`: Clients to extract video data from. The main clients are `web`, `ios` and `android`, with variants `_music`, `_embedded`, `_embedscreen`, `_creator` (e.g. `web_embedded`); and `mweb`, `mweb_embedscreen` and `tv_embedded` (agegate bypass) with no variants. By default, `ios,web` is used, but `tv_embedded` and `creator` variants are added as required for age-gated videos. Similarly, the music variants are added for `music.youtube.com` urls. The `android` clients will always be given lowest priority since their formats are broken. You can use `all` to use all the clients, and `default` for the default clients.
 * `player_skip`: Skip some network requests that are generally needed for robust extraction. One or more of `configs` (skip client configs), `webpage` (skip initial webpage), `js` (skip js player). While these options can help reduce the number of requests needed or avoid some rate-limiting, they could cause some issues. See [#860](https://github.com/yt-dlp/yt-dlp/pull/860) for more details
 * `player_params`: YouTube player parameters to use for player requests. Will overwrite any default ones set by yt-dlp.
 * `comment_sort`: `top` or `new` (default) - choose comment sorting mode (on YouTube's side)
@@ -1813,8 +1812,8 @@ #### tiktok
 * `app_name`: Default app name to use with mobile API calls, e.g. `trill`
 * `app_version`: Default app version to use with mobile API calls - should be set along with `manifest_app_version`, e.g. `34.1.2`
 * `manifest_app_version`: Default numeric app version to use with mobile API calls, e.g. `2023401020`
-* `aid`: Default app ID to use with API calls, e.g. `1180`
-* `app_info`: One or more app info strings in the format of `<iid>/[app_name]/[app_version]/[manifest_app_version]/[aid]`, where `iid` is the unique app install ID. `iid` is the only required value; all other values and their `/` separators can be omitted, e.g. `tiktok:app_info=1234567890123456789` or `tiktok:app_info=123,456/trill///1180,789//34.0.1/340001`
+* `aid`: Default app ID to use with mobile API calls, e.g. `1180`
+* `app_info`: Enable mobile API extraction with one or more app info strings in the format of `<iid>/[app_name]/[app_version]/[manifest_app_version]/[aid]`, where `iid` is the unique app install ID. `iid` is the only required value; all other values and their `/` separators can be omitted, e.g. `tiktok:app_info=1234567890123456789` or `tiktok:app_info=123,456/trill///1180,789//34.0.1/340001`
 
 #### rokfinchannel
 * `tab`: Which tab to download - one of `new`, `top`, `videos`, `podcasts`, `streams`, `stacks`

test/test_YoutubeDL.py

@@ -707,6 +707,13 @@ def test(tmpl, expected, *, info=None, **params):
         test('%(width)06d.%%(ext)s', 'NA.%(ext)s')
         test('%%(width)06d.%(ext)s', '%(width)06d.mp4')
 
+        # Sanitization options
+        test('%(title3)s', (None, 'foo⧸bar⧹test'))
+        test('%(title5)s', (None, 'aei_A'), restrictfilenames=True)
+        test('%(title3)s', (None, 'foo_bar_test'), windowsfilenames=False, restrictfilenames=True)
+        if sys.platform != 'win32':
+            test('%(title3)s', (None, 'foo⧸bar\\test'), windowsfilenames=False)
+
         # ID sanitization
         test('%(id)s', '_abcd', info={'id': '_abcd'})
         test('%(some_id)s', '_abcd', info={'some_id': '_abcd'})

yt_dlp/YoutubeDL.py

@@ -254,7 +254,9 @@ class YoutubeDL:
     outtmpl_na_placeholder: Placeholder for unavailable meta fields.
     restrictfilenames: Do not allow "&" and spaces in file names
     trim_file_name:    Limit length of filename (extension excluded)
-    windowsfilenames:  Force the filenames to be windows compatible
+    windowsfilenames:  True: Force filenames to be Windows compatible
+                       False: Sanitize filenames only minimally
+                       This option has no effect when running on Windows
     ignoreerrors:      Do not stop on download/postprocessing errors.
                        Can be 'only_download' to ignore only download errors.
                        Default is 'only_download' for CLI, but False for API
@@ -1175,8 +1177,7 @@ def _copy_infodict(info_dict):
 
     def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
         """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
-        @param sanitize    Whether to sanitize the output as a filename.
-                           For backward compatibility, a function can also be passed
+        @param sanitize    Whether to sanitize the output as a filename
         """
 
         info_dict.setdefault('epoch', int(time.time()))  # keep epoch consistent once set
@@ -1292,14 +1293,23 @@ def get_value(mdict):
 
         na = self.params.get('outtmpl_na_placeholder', 'NA')
 
-        def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
+        def filename_sanitizer(key, value, restricted):
             return sanitize_filename(str(value), restricted=restricted, is_id=(
                 bool(re.search(r'(^|[_.])id(\.|$)', key))
                 if 'filename-sanitization' in self.params['compat_opts']
                 else NO_DEFAULT))
 
-        sanitizer = sanitize if callable(sanitize) else filename_sanitizer
-        sanitize = bool(sanitize)
+        if callable(sanitize):
+            self.deprecation_warning('Passing a callable "sanitize" to YoutubeDL.prepare_outtmpl is deprecated')
+        elif not sanitize:
+            pass
+        elif (sys.platform != 'win32' and not self.params.get('restrictfilenames')
+                and self.params.get('windowsfilenames') is False):
+            def sanitize(key, value):
+                return value.replace('/', '\u29F8').replace('\0', '')
+        else:
+            def sanitize(key, value):
+                return filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames'))
 
         def _dumpjson_default(obj):
             if isinstance(obj, (set, LazyList)):
@@ -1382,13 +1392,13 @@ def create_key(outer_mobj):
 
             if sanitize:
                 # If value is an object, sanitize might convert it to a string
-                # So we convert it to repr first
+                # So we manually convert it before sanitizing
                 if fmt[-1] == 'r':
                     value, fmt = repr(value), str_fmt
                 elif fmt[-1] == 'a':
                     value, fmt = ascii(value), str_fmt
                 if fmt[-1] in 'csra':
-                    value = sanitizer(last_field, value)
+                    value = sanitize(last_field, value)
 
             key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
             TMPL_DICT[key] = value

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/cda.py

@@ -16,7 +16,6 @@
     merge_dicts,
     multipart_encode,
     parse_duration,
-    random_birthday,
     traverse_obj,
     try_call,
     try_get,
@@ -63,38 +62,57 @@ class CDAIE(InfoExtractor):
             'description': 'md5:60d76b71186dcce4e0ba6d4bbdb13e1a',
             'thumbnail': r're:^https?://.*\.jpg$',
             'uploader': 'crash404',
-            'view_count': int,
             'average_rating': float,
             'duration': 137,
             'age_limit': 0,
+            'upload_date': '20160220',
+            'timestamp': 1455968218,
         }
     }, {
-        # Age-restricted
-        'url': 'http://www.cda.pl/video/1273454c4',
+        # Age-restricted with vfilm redirection
+        'url': 'https://www.cda.pl/video/8753244c4',
+        'md5': 'd8eeb83d63611289507010d3df3bb8b3',
         'info_dict': {
-            'id': '1273454c4',
+            'id': '8753244c4',
             'ext': 'mp4',
-            'title': 'Bronson (2008) napisy HD 1080p',
-            'description': 'md5:1b6cb18508daf2dc4e0fa4db77fec24c',
+            'title': '[18+] Bez Filtra: Rezerwowe Psy czyli...  najwulgarniejsza polska gra?',
+            'description': 'md5:ae80bac31bd6a9f077a6cce03c7c077e',
             'height': 1080,
-            'uploader': 'boniek61',
+            'uploader': 'arhn eu',
             'thumbnail': r're:^https?://.*\.jpg$',
-            'duration': 5554,
+            'duration': 991,
             'age_limit': 18,
-            'view_count': int,
             'average_rating': float,
-        },
+            'timestamp': 1633888264,
+            'upload_date': '20211010',
+        }
+    }, {
+        # Age-restricted without vfilm redirection
+        'url': 'https://www.cda.pl/video/17028157b8',
+        'md5': 'c1fe5ff4582bace95d4f0ce0fbd0f992',
+        'info_dict': {
+            'id': '17028157b8',
+            'ext': 'mp4',
+            'title': 'STENDUPY MICHAŁ OGIŃSKI',
+            'description': 'md5:5851f3272bfc31f762d616040a1d609a',
+            'height': 480,
+            'uploader': 'oginski',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 18855,
+            'age_limit': 18,
+            'average_rating': float,
+            'timestamp': 1699705901,
+            'upload_date': '20231111',
+        }
     }, {
         'url': 'http://ebd.cda.pl/0x0/5749950c',
         'only_matching': True,
     }]
 
     def _download_age_confirm_page(self, url, video_id, *args, **kwargs):
-        form_data = random_birthday('rok', 'miesiac', 'dzien')
-        form_data.update({'return': url, 'module': 'video', 'module_id': video_id})
-        data, content_type = multipart_encode(form_data)
+        data, content_type = multipart_encode({'age_confirm': ''})
         return self._download_webpage(
-            urljoin(url, '/a/validatebirth'), video_id, *args,
+            url, video_id, *args,
             data=data, headers={
                 'Referer': url,
                 'Content-Type': content_type,
@@ -164,7 +182,7 @@ def _real_extract(self, url):
         if 'Authorization' in self._API_HEADERS:
             return self._api_extract(video_id)
         else:
-            return self._web_extract(video_id, url)
+            return self._web_extract(video_id)
 
     def _api_extract(self, video_id):
         meta = self._download_json(
@@ -197,9 +215,9 @@ def _api_extract(self, video_id):
             'view_count': meta.get('views'),
         }
 
-    def _web_extract(self, video_id, url):
+    def _web_extract(self, video_id):
         self._set_cookie('cda.pl', 'cda.player', 'html5')
-        webpage = self._download_webpage(
+        webpage, urlh = self._download_webpage_handle(
             f'{self._BASE_URL}/video/{video_id}/vfilm', video_id)
 
         if 'Ten film jest dostępny dla użytkowników premium' in webpage:
@@ -209,10 +227,10 @@ def _web_extract(self, video_id, url):
             self.raise_geo_restricted()
 
         need_confirm_age = False
-        if self._html_search_regex(r'(<form[^>]+action="[^"]*/a/validatebirth[^"]*")',
+        if self._html_search_regex(r'(<button[^>]+name="[^"]*age_confirm[^"]*")',
                                    webpage, 'birthday validate form', default=None):
             webpage = self._download_age_confirm_page(
-                url, video_id, note='Confirming age')
+                urlh.url, video_id, note='Confirming age')
             need_confirm_age = True
 
         formats = []
@@ -222,9 +240,6 @@ def _web_extract(self, video_id, url):
             (?:<\1[^>]*>[^<]*</\1>|(?!</\1>)(?:.|\n))*?
             <(span|meta)[^>]+itemprop=(["\'])name\4[^>]*>(?P<uploader>[^<]+)</\3>
         ''', webpage, 'uploader', default=None, group='uploader')
-        view_count = self._search_regex(
-            r'Odsłony:(?:\s|&nbsp;)*([0-9]+)', webpage,
-            'view_count', default=None)
         average_rating = self._search_regex(
             (r'<(?:span|meta)[^>]+itemprop=(["\'])ratingValue\1[^>]*>(?P<rating_value>[0-9.]+)',
              r'<span[^>]+\bclass=["\']rating["\'][^>]*>(?P<rating_value>[0-9.]+)'), webpage, 'rating', fatal=False,
@@ -235,7 +250,6 @@ def _web_extract(self, video_id, url):
             'title': self._og_search_title(webpage),
             'description': self._og_search_description(webpage),
             'uploader': uploader,
-            'view_count': int_or_none(view_count),
             'average_rating': float_or_none(average_rating),
             'thumbnail': self._og_search_thumbnail(webpage),
             'formats': formats,

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/tiktok.py

@@ -45,19 +45,18 @@ class TikTokBaseIE(InfoExtractor):
         # "app id": aweme = 1128, trill = 1180, musical_ly = 1233, universal = 0
         'aid': '0',
     }
-    _KNOWN_APP_INFO = [
-        '7351144126450059040',
-        '7351149742343391009',
-        '7351153174894626592',
-    ]
     _APP_INFO_POOL = None
     _APP_INFO = None
     _APP_USER_AGENT = None
 
+    @property
+    def _KNOWN_APP_INFO(self):
+        return self._configuration_arg('app_info', ie_key=TikTokIE)
+
     @property
     def _API_HOSTNAME(self):
         return self._configuration_arg(
-            'api_hostname', ['api22-normal-c-useast2a.tiktokv.com'], ie_key=TikTokIE)[0]
+            'api_hostname', ['api16-normal-c-useast1a.tiktokv.com'], ie_key=TikTokIE)[0]
 
     def _get_next_app_info(self):
         if self._APP_INFO_POOL is None:
@@ -66,13 +65,10 @@ def _get_next_app_info(self):
                 for key, default in self._APP_INFO_DEFAULTS.items()
                 if key != 'iid'
             }
-            app_info_list = (
-                self._configuration_arg('app_info', ie_key=TikTokIE)
-                or random.sample(self._KNOWN_APP_INFO, len(self._KNOWN_APP_INFO)))
             self._APP_INFO_POOL = [
                 {**defaults, **dict(
                     (k, v) for k, v in zip(self._APP_INFO_DEFAULTS, app_info.split('/')) if v
-                )} for app_info in app_info_list
+                )} for app_info in self._KNOWN_APP_INFO
             ]
 
         if not self._APP_INFO_POOL:
@@ -757,11 +753,13 @@ class TikTokIE(TikTokBaseIE):
 
     def _real_extract(self, url):
         video_id, user_id = self._match_valid_url(url).group('id', 'user_id')
-        try:
-            return self._extract_aweme_app(video_id)
-        except ExtractorError as e:
-            e.expected = True
-            self.report_warning(f'{e}; trying with webpage')
+
+        if self._KNOWN_APP_INFO:
+            try:
+                return self._extract_aweme_app(video_id)
+            except ExtractorError as e:
+                e.expected = True
+                self.report_warning(f'{e}; trying with webpage')
 
         url = self._create_url(user_id, video_id)
         webpage = self._download_webpage(url, video_id, headers={'User-Agent': 'Mozilla/5.0'})

yt_dlp/extractor/twitter.py

@@ -36,7 +36,7 @@ class TwitterBaseIE(InfoExtractor):
     _NETRC_MACHINE = 'twitter'
     _API_BASE = 'https://api.twitter.com/1.1/'
     _GRAPHQL_API_BASE = 'https://twitter.com/i/api/graphql/'
-    _BASE_REGEX = r'https?://(?:(?:www|m(?:obile)?)\.)?(?:twitter\.com|twitter3e4tixl4xyajtrzo62zg5vztmjuricljdp2c5kshju4avyoid\.onion)/'
+    _BASE_REGEX = r'https?://(?:(?:www|m(?:obile)?)\.)?(?:(?:twitter|x)\.com|twitter3e4tixl4xyajtrzo62zg5vztmjuricljdp2c5kshju4avyoid\.onion)/'
     _AUTH = 'AAAAAAAAAAAAAAAAAAAAANRILgAAAAAAnNwIzUejRCOuH5E6I8xnZz4puTs%3D1Zv7ttfk8LF81IUq16cHjhLTvJu4FA33AGWWjCpTnA'
     _LEGACY_AUTH = 'AAAAAAAAAAAAAAAAAAAAAIK1zgAAAAAA2tUWuhGZ2JceoId5GwYWU5GspY4%3DUq7gzFoCZs1QfwGoVdvSac3IniczZEYXIcDyumCauIXpcAPorE'
     _flow_token = None
@@ -1191,6 +1191,31 @@ class TwitterIE(TwitterBaseIE):
             'age_limit': 0,
             '_old_archive_ids': ['twitter 1724884212803834154'],
         },
+    }, {
+        # x.com
+        'url': 'https://x.com/historyinmemes/status/1790637656616943991',
+        'md5': 'daca3952ba0defe2cfafb1276d4c1ea5',
+        'info_dict': {
+            'id': '1790637589910654976',
+            'ext': 'mp4',
+            'title': 'Historic Vids - One of the most intense moments in history',
+            'description': 'One of the most intense moments in history https://t.co/Zgzhvix8ES',
+            'display_id': '1790637656616943991',
+            'uploader': 'Historic Vids',
+            'uploader_id': 'historyinmemes',
+            'uploader_url': 'https://twitter.com/historyinmemes',
+            'channel_id': '855481986290524160',
+            'upload_date': '20240515',
+            'timestamp': 1715756260.0,
+            'duration': 15.488,
+            'tags': [],
+            'comment_count': int,
+            'repost_count': int,
+            'like_count': int,
+            'thumbnail': r're:https://pbs\.twimg\.com/amplify_video_thumb/.+',
+            'age_limit': 0,
+            '_old_archive_ids': ['twitter 1790637656616943991'],
+        }
     }, {
         # onion route
         'url': 'https://twitter3e4tixl4xyajtrzo62zg5vztmjuricljdp2c5kshju4avyoid.onion/TwitterBlue/status/1484226494708662273',

yt_dlp/extractor/youtube.py

@@ -2353,6 +2353,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'format': '17',  # 3gp format available on android
                 'extractor_args': {'youtube': {'player_client': ['android']}},
             },
+            'skip': 'android client broken',
         },
         {
             # Skip download of additional client configs (remix client config in this case)
@@ -2730,7 +2731,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'heatmap': 'count:100',
             },
             'params': {
-                'extractor_args': {'youtube': {'player_client': ['android'], 'player_skip': ['webpage']}},
+                'extractor_args': {'youtube': {'player_client': ['ios'], 'player_skip': ['webpage']}},
             },
         },
     ]
@@ -3317,7 +3318,36 @@ 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 +3428,39 @@ 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 +3553,7 @@ 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 +3577,7 @@ 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 +3586,7 @@ 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
@@ -3614,8 +3663,6 @@ def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg,
         yt_query = {
             'videoId': video_id,
         }
-        if _split_innertube_client(client)[0] in ('android', 'android_embedscreen'):
-            yt_query['params'] = 'CgIIAQ=='
 
         pp_arg = self._configuration_arg('player_params', [None], casesense=True)[0]
         if pp_arg:
@@ -3631,19 +3678,24 @@ def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg,
 
     def _get_requested_clients(self, url, smuggled_data):
         requested_clients = []
-        default = ['ios', 'android', 'web']
+        android_clients = []
+        default = ['ios', 'web']
         allowed_clients = sorted(
             (client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'),
             key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
         for client in self._configuration_arg('player_client'):
-            if client in allowed_clients:
-                requested_clients.append(client)
-            elif client == 'default':
+            if client == 'default':
                 requested_clients.extend(default)
             elif client == 'all':
                 requested_clients.extend(allowed_clients)
-            else:
+            elif client not in allowed_clients:
                 self.report_warning(f'Skipping unsupported client {client}')
+            elif client.startswith('android'):
+                android_clients.append(client)
+            else:
+                requested_clients.append(client)
+        # Force deprioritization of broken Android clients for format de-duplication
+        requested_clients.extend(android_clients)
         if not requested_clients:
             requested_clients = default
 
@@ -3862,6 +3914,14 @@ def build_fragments(f):
                     f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True)
 
             client_name = fmt.get(STREAMING_DATA_CLIENT_NAME)
+            # Android client formats are broken due to integrity check enforcement
+            # Ref: https://github.com/yt-dlp/yt-dlp/issues/9554
+            is_broken = client_name and client_name.startswith(short_client_name('android'))
+            if is_broken:
+                self.report_warning(
+                    f'{video_id}: Android client formats are broken and may yield HTTP Error 403. '
+                    'They will be deprioritized', only_once=True)
+
             name = fmt.get('qualityLabel') or quality.replace('audio_quality_', '') or ''
             fps = int_or_none(fmt.get('fps')) or 0
             dct = {
@@ -3874,7 +3934,7 @@ def build_fragments(f):
                     name, fmt.get('isDrc') and 'DRC',
                     try_get(fmt, lambda x: x['projectionType'].replace('RECTANGULAR', '').lower()),
                     try_get(fmt, lambda x: x['spatialAudioType'].replace('SPATIAL_AUDIO_TYPE_', '').lower()),
-                    throttled and 'THROTTLED', is_damaged and 'DAMAGED',
+                    throttled and 'THROTTLED', is_damaged and 'DAMAGED', is_broken and 'BROKEN',
                     (self.get_param('verbose') or all_formats) and client_name,
                     delim=', '),
                 # Format 22 is likely to be damaged. See https://github.com/yt-dlp/yt-dlp/issues/3372
@@ -3892,8 +3952,8 @@ def build_fragments(f):
                 'language': join_nonempty(audio_track.get('id', '').split('.')[0],
                                           'desc' if language_preference < -1 else '') or None,
                 'language_preference': language_preference,
-                # Strictly de-prioritize damaged and 3gp formats
-                'preference': -10 if is_damaged else -2 if itag == '17' else None,
+                # Strictly de-prioritize broken, damaged and 3gp formats
+                'preference': -20 if is_broken else -10 if is_damaged else -2 if itag == '17' else None,
             }
             mime_mobj = re.match(
                 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')

yt_dlp/options.py

@@ -1356,12 +1356,12 @@ def _alias_callback(option, opt_str, value, parser, opts, nargs):
         help='Allow Unicode characters, "&" and spaces in filenames (default)')
     filesystem.add_option(
         '--windows-filenames',
-        action='store_true', dest='windowsfilenames', default=False,
+        action='store_true', dest='windowsfilenames', default=None,
         help='Force filenames to be Windows-compatible')
     filesystem.add_option(
         '--no-windows-filenames',
         action='store_false', dest='windowsfilenames',
-        help='Make filenames Windows-compatible only if using Windows (default)')
+        help='Sanitize filenames only minimally')
     filesystem.add_option(
         '--trim-filenames', '--trim-file-names', metavar='LENGTH',
         dest='trim_file_name', default=0, type=int,