Merge 9de472f6ea into da252d9d32

Closes #6884
Authored by: bashonly, Grub4K, seproDev

Co-authored-by: Simon Sawicki <contact@grub4k.xyz>
Co-authored-by: sepro <sepro@sepr0.com>

README.md

@@ -342,8 +342,9 @@ If you fork the project on GitHub, you can run your fork's [build workflow](.git
                                     extractor plugins; postprocessor plugins can
                                     only be loaded from the default plugin
                                     directories
-    --flat-playlist                 Do not extract the videos of a playlist,
+    --flat-playlist                 Do not extract a playlist's URL result
-                                    only list them
+                                    entries; some entry metadata may be missing
+                                    and downloading may be bypassed
     --no-flat-playlist              Fully extract the videos of a playlist
                                     (default)
     --live-from-start               Download livestreams from the start.
@@ -612,8 +613,7 @@ If you fork the project on GitHub, you can run your fork's [build workflow](.git
     --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
+    --no-windows-filenames          Sanitize filenames only minimally
-                                    using Windows (default)
     --trim-filenames LENGTH         Limit the filename length (excluding
                                     extension) to the specified number of
                                     characters
@@ -1869,6 +1869,9 @@ The following extractors use this feature:
 #### digitalconcerthall
 * `prefer_combined_hls`: Prefer extracting combined/pre-merged video and audio HLS formats. This will exclude 4K/HEVC video and lossless/FLAC audio formats, which are only available as split video/audio HLS formats
 
+#### sonylivseries
+* `sort_order`: Episode sort order for series extraction - one of `asc` (ascending, oldest first) or `desc` (descending, newest first). Default is `asc`
+
 **Note**: These options may be changed/removed in the future without concern for backward compatibility
 
 <!-- MANPAGE: MOVE "INSTALLATION" SECTION HERE -->

devscripts/changelog_override.json

@@ -234,5 +234,10 @@
         "when": "57212a5f97ce367590aaa5c3e9a135eead8f81f7",
         "short": "[ie/vimeo] Fix API retries (#11351)",
         "authors": ["bashonly"]
+    },
+    {
+        "action": "add",
+        "when": "52c0ffe40ad6e8404d93296f575007b05b04c686",
+        "short": "[priority] **Login with OAuth is no longer supported for YouTube**\nDue to a change made by the site, yt-dlp is longer able to support OAuth login for YouTube. [Read more](https://github.com/yt-dlp/yt-dlp/issues/11462#issuecomment-2471703090)"
     }
 ]

test/test_YoutubeDL.py

@@ -761,6 +761,13 @@ class TestYoutubeDL(unittest.TestCase):
         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

@@ -266,7 +266,9 @@ 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
@@ -1192,8 +1194,7 @@ class YoutubeDL:
 
     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.
+        @param sanitize    Whether to sanitize the output as a filename
-                           For backward compatibility, a function can also be passed
         """
 
         info_dict.setdefault('epoch', int(time.time()))  # keep epoch consistent once set
@@ -1309,14 +1310,23 @@ class YoutubeDL:
 
         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
+        if callable(sanitize):
-        sanitize = bool(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)):
@@ -1399,13 +1409,13 @@ class YoutubeDL:
 
             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 = '{}\0{}'.format(key.replace('%', '%\0'), outer_mobj.group('format'))
             TMPL_DICT[key] = value

yt_dlp/extractor/_extractors.py

@@ -1520,8 +1520,8 @@ from .pgatour import PGATourIE
 from .philharmoniedeparis import PhilharmonieDeParisIE
 from .phoenix import PhoenixIE
 from .photobucket import PhotobucketIE
+from .pialive import PiaLiveIE
 from .piapro import PiaproIE
-from .piaulizaportal import PIAULIZAPortalIE
 from .picarto import (
     PicartoIE,
     PicartoVodIE,
@@ -2250,6 +2250,10 @@ from .ufctv import (
 )
 from .ukcolumn import UkColumnIE
 from .uktvplay import UKTVPlayIE
+from .uliza import (
+    UlizaPlayerIE,
+    UlizaPortalIE,
+)
 from .umg import UMGDeIE
 from .unistra import UnistraIE
 from .unity import UnityIE

yt_dlp/extractor/bandlab.py

@@ -1,4 +1,3 @@
-
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,

yt_dlp/extractor/common.py

@@ -3767,7 +3767,7 @@ class InfoExtractor:
         """ Merge subtitle dictionaries, language by language. """
         if target is None:
             target = {}
-        for d in dicts:
+        for d in filter(None, dicts):
             for lang, subs in d.items():
                 target[lang] = cls._merge_subtitle_items(target.get(lang, []), subs)
         return target

yt_dlp/extractor/ctvnews.py

@@ -1,14 +1,27 @@
+import json
 import re
+import urllib.parse
 
 from .common import InfoExtractor
-from ..utils import orderedSet
+from .ninecninemedia import NineCNineMediaIE
+from ..utils import extract_attributes, orderedSet
+from ..utils.traversal import find_element, traverse_obj
 
 
 class CTVNewsIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:.+?\.)?ctvnews\.ca/(?:video\?(?:clip|playlist|bin)Id=|.*?)(?P<id>[0-9.]+)'
+    _BASE_REGEX = r'https?://(?:[^.]+\.)?ctvnews\.ca/'
+    _VIDEO_ID_RE = r'(?P<id>\d{5,})'
+    _PLAYLIST_ID_RE = r'(?P<id>\d\.\d{5,})'
+    _VALID_URL = [
+        rf'{_BASE_REGEX}video/c{_VIDEO_ID_RE}',
+        rf'{_BASE_REGEX}video(?:-gallery)?/?\?clipId={_VIDEO_ID_RE}',
+        rf'{_BASE_REGEX}video/?\?(?:playlist|bin)Id={_PLAYLIST_ID_RE}',
+        rf'{_BASE_REGEX}(?!video/)[^?#]*?{_PLAYLIST_ID_RE}/?(?:$|[?#])',
+        rf'{_BASE_REGEX}(?!video/)[^?#]+\?binId={_PLAYLIST_ID_RE}',
+    ]
     _TESTS = [{
         'url': 'http://www.ctvnews.ca/video?clipId=901995',
-        'md5': '9b8624ba66351a23e0b6e1391971f9af',
+        'md5': 'b608f466c7fa24b9666c6439d766ab7e',
         'info_dict': {
             'id': '901995',
             'ext': 'flv',
@@ -16,6 +29,33 @@ class CTVNewsIE(InfoExtractor):
             'description': 'md5:958dd3b4f5bbbf0ed4d045c790d89285',
             'timestamp': 1467286284,
             'upload_date': '20160630',
+            'categories': [],
+            'season_number': 0,
+            'season': 'Season 0',
+            'tags': [],
+            'series': 'CTV News National | Archive | Stories 2',
+            'season_id': '57981',
+            'thumbnail': r're:https?://.*\.jpg$',
+            'duration': 764.631,
+        },
+    }, {
+        'url': 'https://barrie.ctvnews.ca/video/c3030933-here_s-what_s-making-news-for-nov--15?binId=1272429',
+        'md5': '8b8c2b33c5c1803e3c26bc74ff8694d5',
+        'info_dict': {
+            'id': '3030933',
+            'ext': 'flv',
+            'title': 'Here’s what’s making news for Nov. 15',
+            'description': 'Here are the top stories we’re working on for CTV News at 11 for Nov. 15',
+            'thumbnail': 'http://images2.9c9media.com/image_asset/2021_2_22_a602e68e-1514-410e-a67a-e1f7cccbacab_png_2000x1125.jpg',
+            'season_id': '58104',
+            'season_number': 0,
+            'tags': [],
+            'season': 'Season 0',
+            'categories': [],
+            'series': 'CTV News Barrie',
+            'upload_date': '20241116',
+            'duration': 42.943,
+            'timestamp': 1731722452,
         },
     }, {
         'url': 'http://www.ctvnews.ca/video?playlistId=1.2966224',
@@ -31,6 +71,72 @@ class CTVNewsIE(InfoExtractor):
             'id': '1.2876780',
         },
         'playlist_mincount': 100,
+    }, {
+        'url': 'https://www.ctvnews.ca/it-s-been-23-years-since-toronto-called-in-the-army-after-a-major-snowstorm-1.5736957',
+        'info_dict':
+        {
+            'id': '1.5736957',
+        },
+        'playlist_mincount': 6,
+    }, {
+        'url': 'https://www.ctvnews.ca/business/respondents-to-bank-of-canada-questionnaire-largely-oppose-creating-a-digital-loonie-1.6665797',
+        'md5': '24bc4b88cdc17d8c3fc01dfc228ab72c',
+        'info_dict': {
+            'id': '2695026',
+            'ext': 'flv',
+            'season_id': '89852',
+            'series': 'From CTV News Channel',
+            'description': 'md5:796a985a23cacc7e1e2fafefd94afd0a',
+            'season': '2023',
+            'title': 'Bank of Canada asks public about digital currency',
+            'categories': [],
+            'tags': [],
+            'upload_date': '20230526',
+            'season_number': 2023,
+            'thumbnail': 'http://images2.9c9media.com/image_asset/2019_3_28_35f5afc3-10f6-4d92-b194-8b9a86f55c6a_png_1920x1080.jpg',
+            'timestamp': 1685105157,
+            'duration': 253.553,
+        },
+    }, {
+        'url': 'https://stox.ctvnews.ca/video-gallery?clipId=582589',
+        'md5': '135cc592df607d29dddc931f1b756ae2',
+        'info_dict': {
+            'id': '582589',
+            'ext': 'flv',
+            'categories': [],
+            'timestamp': 1427906183,
+            'season_number': 0,
+            'duration': 125.559,
+            'thumbnail': 'http://images2.9c9media.com/image_asset/2019_3_28_35f5afc3-10f6-4d92-b194-8b9a86f55c6a_png_1920x1080.jpg',
+            'series': 'CTV News Stox',
+            'description': 'CTV original footage of the rise and fall of the Berlin Wall.',
+            'title': 'Berlin Wall',
+            'season_id': '63817',
+            'season': 'Season 0',
+            'tags': [],
+            'upload_date': '20150401',
+        },
+    }, {
+        'url': 'https://ottawa.ctvnews.ca/features/regional-contact/regional-contact-archive?binId=1.1164587#3023759',
+        'md5': 'a14c0603557decc6531260791c23cc5e',
+        'info_dict': {
+            'id': '3023759',
+            'ext': 'flv',
+            'season_number': 2024,
+            'timestamp': 1731798000,
+            'season': '2024',
+            'episode': 'Episode 125',
+            'description': 'CTV News Ottawa at Six',
+            'duration': 2712.076,
+            'episode_number': 125,
+            'upload_date': '20241116',
+            'title': 'CTV News Ottawa at Six for Saturday, November 16, 2024',
+            'thumbnail': 'http://images2.9c9media.com/image_asset/2019_3_28_35f5afc3-10f6-4d92-b194-8b9a86f55c6a_png_1920x1080.jpg',
+            'categories': [],
+            'tags': [],
+            'series': 'CTV News Ottawa at Six',
+            'season_id': '92667',
+        },
     }, {
         'url': 'http://www.ctvnews.ca/1.810401',
         'only_matching': True,
@@ -42,29 +148,35 @@ class CTVNewsIE(InfoExtractor):
         'only_matching': True,
     }]
 
+    def _ninecninemedia_url_result(self, clip_id):
+        return self.url_result(f'9c9media:ctvnews_web:{clip_id}', NineCNineMediaIE, clip_id)
+
     def _real_extract(self, url):
         page_id = self._match_id(url)
 
-        def ninecninemedia_url_result(clip_id):
+        if mobj := re.fullmatch(self._VIDEO_ID_RE, urllib.parse.urlparse(url).fragment):
-            return {
+            page_id = mobj.group('id')
-                '_type': 'url_transparent',
-                'id': clip_id,
-                'url': f'9c9media:ctvnews_web:{clip_id}',
-                'ie_key': 'NineCNineMedia',
-            }
 
-        if page_id.isdigit():
+        if re.fullmatch(self._VIDEO_ID_RE, page_id):
-            return ninecninemedia_url_result(page_id)
+            return self._ninecninemedia_url_result(page_id)
-        else:
+
-            webpage = self._download_webpage(f'http://www.ctvnews.ca/{page_id}', page_id, query={
+        webpage = self._download_webpage(f'https://www.ctvnews.ca/{page_id}', page_id, query={
             'ot': 'example.AjaxPageLayout.ot',
             'maxItemsPerPage': 1000000,
         })
-            entries = [ninecninemedia_url_result(clip_id) for clip_id in orderedSet(
+        entries = [self._ninecninemedia_url_result(clip_id)
-                re.findall(r'clip\.id\s*=\s*(\d+);', webpage))]
+                   for clip_id in orderedSet(re.findall(r'clip\.id\s*=\s*(\d+);', webpage))]
         if not entries:
             webpage = self._download_webpage(url, page_id)
             if 'getAuthStates("' in webpage:
-                    entries = [ninecninemedia_url_result(clip_id) for clip_id in
+                entries = [self._ninecninemedia_url_result(clip_id) for clip_id in
                            self._search_regex(r'getAuthStates\("([\d+,]+)"', webpage, 'clip ids').split(',')]
+            else:
+                entries = [
+                    self._ninecninemedia_url_result(clip_id) for clip_id in
+                    traverse_obj(webpage, (
+                        {find_element(tag='jasper-player-container', html=True)},
+                        {extract_attributes}, 'axis-ids', {json.loads}, ..., 'axisId', {str}))
+                ]
+
         return self.playlist_result(entries, page_id)

yt_dlp/extractor/facebook.py

@@ -569,7 +569,7 @@ class FacebookIE(InfoExtractor):
             if dash_manifest:
                 formats.extend(self._parse_mpd_formats(
                     compat_etree_fromstring(urllib.parse.unquote_plus(dash_manifest)),
-                    mpd_url=url_or_none(video.get('dash_manifest_url')) or mpd_url))
+                    mpd_url=url_or_none(vid_data.get('dash_manifest_url')) or mpd_url))
 
         def process_formats(info):
             # Downloads with browser's User-Agent are rate limited. Working around

yt_dlp/extractor/litv.py

@@ -1,30 +1,32 @@
 import json
+import uuid
 
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
     int_or_none,
+    join_nonempty,
     smuggle_url,
     traverse_obj,
     try_call,
     unsmuggle_url,
+    urljoin,
 )
 
 
 class LiTVIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?litv\.tv/(?:vod|promo)/[^/]+/(?:content\.do)?\?.*?\b(?:content_)?id=(?P<id>[^&]+)'
+    _VALID_URL = r'https?://(?:www\.)?litv\.tv/(?:[^/?#]+/watch/|vod/[^/?#]+/content\.do\?content_id=)(?P<id>[\w-]+)'
-
+    _URL_TEMPLATE = 'https://www.litv.tv/%s/watch/%s'
-    _URL_TEMPLATE = 'https://www.litv.tv/vod/%s/content.do?content_id=%s'
+    _GEO_COUNTRIES = ['TW']
-
     _TESTS = [{
-        'url': 'https://www.litv.tv/vod/drama/content.do?brc_id=root&id=VOD00041610&isUHEnabled=true&autoPlay=1',
+        'url': 'https://www.litv.tv/drama/watch/VOD00041610',
         'info_dict': {
             'id': 'VOD00041606',
             'title': '花千骨',
         },
         'playlist_count': 51,  # 50 episodes + 1 trailer
     }, {
-        'url': 'https://www.litv.tv/vod/drama/content.do?brc_id=root&id=VOD00041610&isUHEnabled=true&autoPlay=1',
+        'url': 'https://www.litv.tv/drama/watch/VOD00041610',
         'md5': 'b90ff1e9f1d8f5cfcd0a44c3e2b34c7a',
         'info_dict': {
             'id': 'VOD00041610',
@@ -32,16 +34,15 @@ class LiTVIE(InfoExtractor):
             'title': '花千骨第1集',
             'thumbnail': r're:https?://.*\.jpg$',
             'description': '《花千骨》陸劇線上看。十六年前，平靜的村莊內，一名女嬰隨異相出生，途徑此地的蜀山掌門清虛道長算出此女命運非同一般，她體內散發的異香易招惹妖魔。一念慈悲下，他在村莊周邊設下結界阻擋妖魔入侵，讓其年滿十六後去蜀山，並賜名花千骨。',
-            'categories': ['奇幻', '愛情', '中國', '仙俠'],
+            'categories': ['奇幻', '愛情', '仙俠', '古裝'],
             'episode': 'Episode 1',
             'episode_number': 1,
         },
         'params': {
             'noplaylist': True,
         },
-        'skip': 'Georestricted to Taiwan',
     }, {
-        'url': 'https://www.litv.tv/promo/miyuezhuan/?content_id=VOD00044841&',
+        'url': 'https://www.litv.tv/drama/watch/VOD00044841',
         'md5': '88322ea132f848d6e3e18b32a832b918',
         'info_dict': {
             'id': 'VOD00044841',
@@ -55,94 +56,62 @@ class LiTVIE(InfoExtractor):
     def _extract_playlist(self, playlist_data, content_type):
         all_episodes = [
             self.url_result(smuggle_url(
-                self._URL_TEMPLATE % (content_type, episode['contentId']),
+                self._URL_TEMPLATE % (content_type, episode['content_id']),
                 {'force_noplaylist': True}))  # To prevent infinite recursion
-            for episode in traverse_obj(playlist_data, ('seasons', ..., 'episode', lambda _, v: v['contentId']))]
+            for episode in traverse_obj(playlist_data, ('seasons', ..., 'episodes', lambda _, v: v['content_id']))]
 
-        return self.playlist_result(all_episodes, playlist_data['contentId'], playlist_data.get('title'))
+        return self.playlist_result(all_episodes, playlist_data['content_id'], playlist_data.get('title'))
 
     def _real_extract(self, url):
         url, smuggled_data = unsmuggle_url(url, {})
-
         video_id = self._match_id(url)
-
         webpage = self._download_webpage(url, video_id)
+        vod_data = self._search_nextjs_data(webpage, video_id)['props']['pageProps']
 
-        if self._search_regex(
+        program_info = traverse_obj(vod_data, ('programInformation', {dict})) or {}
-                r'(?i)<meta\s[^>]*http-equiv="refresh"\s[^>]*content="[0-9]+;\s*url=https://www\.litv\.tv/"',
+        playlist_data = traverse_obj(vod_data, ('seriesTree'))
-                webpage, 'meta refresh redirect', default=False, group=0):
+        if playlist_data and self._yes_playlist(program_info.get('series_id'), video_id, smuggled_data):
-            raise ExtractorError('No such content found', expected=True)
+            return self._extract_playlist(playlist_data, program_info.get('content_type'))
 
-        program_info = self._parse_json(self._search_regex(
+        asset_id = traverse_obj(program_info, ('assets', 0, 'asset_id', {str}))
-            r'var\s+programInfo\s*=\s*([^;]+)', webpage, 'VOD data', default='{}'),
+        if asset_id:  # This is a VOD
-            video_id)
+            media_type = 'vod'
-
+        else:  # This is a live stream
-        # In browsers `getProgramInfo` request is always issued. Usually this
+            asset_id = program_info['content_id']
-        # endpoint gives the same result as the data embedded in the webpage.
+            media_type = program_info['content_type']
-        # If, for some reason, there are no embedded data, we do an extra request.
-        if 'assetId' not in program_info:
-            program_info = self._download_json(
-                'https://www.litv.tv/vod/ajax/getProgramInfo', video_id,
-                query={'contentId': video_id},
-                headers={'Accept': 'application/json'})
-
-        series_id = program_info['seriesId']
-        if self._yes_playlist(series_id, video_id, smuggled_data):
-            playlist_data = self._download_json(
-                'https://www.litv.tv/vod/ajax/getSeriesTree', video_id,
-                query={'seriesId': series_id}, headers={'Accept': 'application/json'})
-            return self._extract_playlist(playlist_data, program_info['contentType'])
-
-        video_data = self._parse_json(self._search_regex(
-            r'uiHlsUrl\s*=\s*testBackendData\(([^;]+)\);',
-            webpage, 'video data', default='{}'), video_id)
-        if not video_data:
-            payload = {'assetId': program_info['assetId']}
         puid = try_call(lambda: self._get_cookies('https://www.litv.tv/')['PUID'].value)
         if puid:
-                payload.update({
+            endpoint = 'get-urls'
-                    'type': 'auth',
-                    'puid': puid,
-                })
-                endpoint = 'getUrl'
         else:
-                payload.update({
+            puid = str(uuid.uuid4())
-                    'watchDevices': program_info['watchDevices'],
+            endpoint = 'get-urls-no-auth'
-                    'contentType': program_info['contentType'],
-                })
-                endpoint = 'getMainUrlNoAuth'
         video_data = self._download_json(
-                f'https://www.litv.tv/vod/ajax/{endpoint}', video_id,
+            f'https://www.litv.tv/api/{endpoint}', video_id,
-                data=json.dumps(payload).encode(),
+            data=json.dumps({'AssetId': asset_id, 'MediaType': media_type, 'puid': puid}).encode(),
             headers={'Content-Type': 'application/json'})
 
-        if not video_data.get('fullpath'):
+        if error := traverse_obj(video_data, ('error', {dict})):
-            error_msg = video_data.get('errorMessage')
+            error_msg = traverse_obj(error, ('message', {str}))
-            if error_msg == 'vod.error.outsideregionerror':
+            if error_msg and 'OutsideRegionError' in error_msg:
                 self.raise_geo_restricted('This video is available in Taiwan only')
-            if error_msg:
+            elif error_msg:
                 raise ExtractorError(f'{self.IE_NAME} said: {error_msg}', expected=True)
-            raise ExtractorError(f'Unexpected result from {self.IE_NAME}')
+            raise ExtractorError(f'Unexpected error from {self.IE_NAME}')
 
         formats = self._extract_m3u8_formats(
-            video_data['fullpath'], video_id, ext='mp4',
+            video_data['result']['AssetURLs'][0], video_id, ext='mp4', m3u8_id='hls')
-            entry_protocol='m3u8_native', m3u8_id='hls')
         for a_format in formats:
             # LiTV HLS segments doesn't like compressions
             a_format.setdefault('http_headers', {})['Accept-Encoding'] = 'identity'
 
-        title = program_info['title'] + program_info.get('secondaryMark', '')
-        description = program_info.get('description')
-        thumbnail = program_info.get('imageFile')
-        categories = [item['name'] for item in program_info.get('category', [])]
-        episode = int_or_none(program_info.get('episode'))
-
         return {
             'id': video_id,
             'formats': formats,
-            'title': title,
+            'title': join_nonempty('title', 'secondary_mark', delim='', from_dict=program_info),
-            'description': description,
+            **traverse_obj(program_info, {
-            'thumbnail': thumbnail,
+                'description': ('description', {str}),
-            'categories': categories,
+                'thumbnail': ('picture', {urljoin('https://p-cdnstatic.svc.litv.tv/')}),
-            'episode_number': episode,
+                'categories': ('genres', ..., 'name', {str}),
+                'episode_number': ('episode', {int_or_none}),
+            }),
         }

yt_dlp/extractor/pialive.py

@@ -0,0 +1,122 @@
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    clean_html,
+    extract_attributes,
+    get_element_by_class,
+    get_element_html_by_class,
+    multipart_encode,
+    str_or_none,
+    unified_timestamp,
+    url_or_none,
+)
+from ..utils.traversal import traverse_obj
+
+
+class PiaLiveIE(InfoExtractor):
+    _VALID_URL = r'https?://player\.pia-live\.jp/stream/(?P<id>[\w-]+)'
+    _PLAYER_ROOT_URL = 'https://player.pia-live.jp/'
+    _PIA_LIVE_API_URL = 'https://api.pia-live.jp'
+    _API_KEY = 'kfds)FKFps-dms9e'
+    _TESTS = [{
+        'url': 'https://player.pia-live.jp/stream/4JagFBEIM14s_hK9aXHKf3k3F3bY5eoHFQxu68TC6krUDqGOwN4d61dCWQYOd6CTxl4hjya9dsfEZGsM4uGOUdax60lEI4twsXGXf7crmz8Gk__GhupTrWxA7RFRVt76',
+        'info_dict': {
+            'id': '88f3109a-f503-4d0f-a9f7-9f39ac745d84',
+            'display_id': '2431867_001',
+            'title': 'こながめでたい日２０２４の視聴ページ | PIA LIVE STREAM(ぴあライブストリーム)',
+            'live_status': 'was_live',
+            'comment_count': int,
+        },
+        'params': {
+            'getcomments': True,
+            'skip_download': True,
+            'ignore_no_formats_error': True,
+        },
+        'skip': 'The video is no longer available',
+    }, {
+        'url': 'https://player.pia-live.jp/stream/4JagFBEIM14s_hK9aXHKf3k3F3bY5eoHFQxu68TC6krJdu0GVBVbVy01IwpJ6J3qBEm3d9TCTt1d0eWpsZGj7DrOjVOmS7GAWGwyscMgiThopJvzgWC4H5b-7XQjAfRZ',
+        'info_dict': {
+            'id': '9ce8b8ba-f6d1-4d1f-83a0-18c3148ded93',
+            'display_id': '2431867_002',
+            'title': 'こながめでたい日２０２４の視聴ページ | PIA LIVE STREAM(ぴあライブストリーム)',
+            'live_status': 'was_live',
+            'comment_count': int,
+        },
+        'params': {
+            'getcomments': True,
+            'skip_download': True,
+            'ignore_no_formats_error': True,
+        },
+        'skip': 'The video is no longer available',
+    }]
+
+    def _extract_var(self, variable, html):
+        return self._search_regex(
+            rf'(?:var|const|let)\s+{variable}\s*=\s*(["\'])(?P<value>(?:(?!\1).)+)\1',
+            html, f'variable {variable}', group='value')
+
+    def _real_extract(self, url):
+        video_key = self._match_id(url)
+        webpage = self._download_webpage(url, video_key)
+
+        program_code = self._extract_var('programCode', webpage)
+        article_code = self._extract_var('articleCode', webpage)
+        title = self._html_extract_title(webpage)
+
+        if get_element_html_by_class('play-end', webpage):
+            raise ExtractorError('The video is no longer available', expected=True, video_id=program_code)
+
+        if start_info := clean_html(get_element_by_class('play-waiting__date', webpage)):
+            date, time = self._search_regex(
+                r'(?P<date>\d{4}/\d{1,2}/\d{1,2})\([月火水木金土日]\)(?P<time>\d{2}:\d{2})',
+                start_info, 'start_info', fatal=False, group=('date', 'time'))
+            if date and time:
+                release_timestamp_str = f'{date} {time} +09:00'
+                release_timestamp = unified_timestamp(release_timestamp_str)
+                self.raise_no_formats(f'The video will be available after {release_timestamp_str}', expected=True)
+                return {
+                    'id': program_code,
+                    'title': title,
+                    'live_status': 'is_upcoming',
+                    'release_timestamp': release_timestamp,
+                }
+
+        payload, content_type = multipart_encode({
+            'play_url': video_key,
+            'api_key': self._API_KEY,
+        })
+        api_data_and_headers = {
+            'data': payload,
+            'headers': {'Content-Type': content_type, 'Referer': self._PLAYER_ROOT_URL},
+        }
+
+        player_tag_list = self._download_json(
+            f'{self._PIA_LIVE_API_URL}/perf/player-tag-list/{program_code}', program_code,
+            'Fetching player tag list', 'Unable to fetch player tag list', **api_data_and_headers)
+
+        return self.url_result(
+            extract_attributes(player_tag_list['data']['movie_one_tag'])['src'],
+            url_transparent=True, title=title, display_id=program_code,
+            __post_extractor=self.extract_comments(program_code, article_code, api_data_and_headers))
+
+    def _get_comments(self, program_code, article_code, api_data_and_headers):
+        chat_room_url = traverse_obj(self._download_json(
+            f'{self._PIA_LIVE_API_URL}/perf/chat-tag-list/{program_code}/{article_code}', program_code,
+            'Fetching chat info', 'Unable to fetch chat info', fatal=False, **api_data_and_headers),
+            ('data', 'chat_one_tag', {extract_attributes}, 'src', {url_or_none}))
+        if not chat_room_url:
+            return
+        comment_page = self._download_webpage(
+            chat_room_url, program_code, 'Fetching comment page', 'Unable to fetch comment page',
+            fatal=False, headers={'Referer': self._PLAYER_ROOT_URL})
+        if not comment_page:
+            return
+        yield from traverse_obj(self._search_json(
+            r'var\s+_history\s*=', comment_page, 'comment list',
+            program_code, contains_pattern=r'\[(?s:.+)\]', fatal=False), (..., {
+                'timestamp': (0, {int}),
+                'author_is_uploader': (1, {lambda x: x == 2}),
+                'author': (2, {str}),
+                'text': (3, {str}),
+                'id': (4, {str_or_none}),
+            }))

yt_dlp/extractor/piaulizaportal.py

@@ -1,70 +0,0 @@
-from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
-    int_or_none,
-    parse_qs,
-    time_seconds,
-    traverse_obj,
-)
-
-
-class PIAULIZAPortalIE(InfoExtractor):
-    IE_DESC = 'ulizaportal.jp - PIA LIVE STREAM'
-    _VALID_URL = r'https?://(?:www\.)?ulizaportal\.jp/pages/(?P<id>[\da-f]{8}-(?:[\da-f]{4}-){3}[\da-f]{12})'
-    _TESTS = [{
-        'url': 'https://ulizaportal.jp/pages/005f18b7-e810-5618-cb82-0987c5755d44',
-        'info_dict': {
-            'id': '005f18b7-e810-5618-cb82-0987c5755d44',
-            'title': 'プレゼンテーションプレイヤーのサンプル',
-            'live_status': 'not_live',
-        },
-        'params': {
-            'skip_download': True,
-            'ignore_no_formats_error': True,
-        },
-    }, {
-        'url': 'https://ulizaportal.jp/pages/005e1b23-fe93-5780-19a0-98e917cc4b7d?expires=4102412400&signature=f422a993b683e1068f946caf406d211c17d1ef17da8bef3df4a519502155aa91&version=1',
-        'info_dict': {
-            'id': '005e1b23-fe93-5780-19a0-98e917cc4b7d',
-            'title': '【確認用】視聴サンプルページ（ULIZA）',
-            'live_status': 'not_live',
-        },
-        'params': {
-            'skip_download': True,
-            'ignore_no_formats_error': True,
-        },
-    }]
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-
-        expires = int_or_none(traverse_obj(parse_qs(url), ('expires', 0)))
-        if expires and expires <= time_seconds():
-            raise ExtractorError('The link is expired.', video_id=video_id, expected=True)
-
-        webpage = self._download_webpage(url, video_id)
-
-        player_data = self._download_webpage(
-            self._search_regex(
-                r'<script [^>]*\bsrc="(https://player-api\.p\.uliza\.jp/v1/players/[^"]+)"',
-                webpage, 'player data url'),
-            video_id, headers={'Referer': 'https://ulizaportal.jp/'},
-            note='Fetching player data', errnote='Unable to fetch player data')
-
-        formats = self._extract_m3u8_formats(
-            self._search_regex(
-                r'["\'](https://vms-api\.p\.uliza\.jp/v1/prog-index\.m3u8[^"\']+)', player_data,
-                'm3u8 url', default=None),
-            video_id, fatal=False)
-        m3u8_type = self._search_regex(
-            r'/hls/(dvr|video)/', traverse_obj(formats, (0, 'url')), 'm3u8 type', default=None)
-
-        return {
-            'id': video_id,
-            'title': self._html_extract_title(webpage),
-            'formats': formats,
-            'live_status': {
-                'video': 'is_live',
-                'dvr': 'was_live',  # short-term archives
-            }.get(m3u8_type, 'not_live'),  # VOD or long-term archives
-        }

yt_dlp/extractor/rutube.py

@@ -13,7 +13,10 @@ from ..utils import (
     unified_timestamp,
     url_or_none,
 )
-from ..utils.traversal import traverse_obj
+from ..utils.traversal import (
+    subs_list_to_dict,
+    traverse_obj,
+)
 
 
 class RutubeBaseIE(InfoExtractor):
@@ -92,11 +95,11 @@ class RutubeBaseIE(InfoExtractor):
                 hls_url, video_id, 'mp4', fatal=False, m3u8_id='hls')
             formats.extend(fmts)
             self._merge_subtitles(subs, target=subtitles)
-        for caption in traverse_obj(options, ('captions', lambda _, v: url_or_none(v['file']))):
+        self._merge_subtitles(traverse_obj(options, ('captions', ..., {
-            subtitles.setdefault(caption.get('code') or 'ru', []).append({
+            'id': 'code',
-                'url': caption['file'],
+            'url': 'file',
-                'name': caption.get('langTitle'),
+            'name': ('langTitle', {str}),
-            })
+        }, all, {subs_list_to_dict(lang='ru')})), target=subtitles)
         return formats, subtitles
 
     def _download_and_extract_formats_and_subtitles(self, video_id, query=None):

yt_dlp/extractor/sonyliv.py

@@ -199,8 +199,9 @@ class SonyLIVSeriesIE(InfoExtractor):
         },
     }]
     _API_BASE = 'https://apiv2.sonyliv.com/AGL'
+    _SORT_ORDERS = ('asc', 'desc')
 
-    def _entries(self, show_id):
+    def _entries(self, show_id, sort_order):
         headers = {
             'Accept': 'application/json, text/plain, */*',
             'Referer': 'https://www.sonyliv.com',
@@ -215,6 +216,9 @@ class SonyLIVSeriesIE(InfoExtractor):
                 'from': '0',
                 'to': '49',
             }), ('resultObj', 'containers', 0, 'containers', lambda _, v: int_or_none(v['id'])))
+
+        if sort_order == 'desc':
+            seasons = reversed(seasons)
         for season in seasons:
             season_id = str(season['id'])
             note = traverse_obj(season, ('metadata', 'title', {str})) or 'season'
@@ -226,7 +230,7 @@ class SonyLIVSeriesIE(InfoExtractor):
                         'from': str(cursor),
                         'to': str(cursor + 99),
                         'orderBy': 'episodeNumber',
-                        'sortOrder': 'asc',
+                        'sortOrder': sort_order,
                     }), ('resultObj', 'containers', 0, 'containers', lambda _, v: int_or_none(v['id'])))
                 if not episodes:
                     break
@@ -237,4 +241,10 @@ class SonyLIVSeriesIE(InfoExtractor):
 
     def _real_extract(self, url):
         show_id = self._match_id(url)
-        return self.playlist_result(self._entries(show_id), playlist_id=show_id)
+
+        sort_order = self._configuration_arg('sort_order', [self._SORT_ORDERS[0]])[0]
+        if sort_order not in self._SORT_ORDERS:
+            raise ValueError(
+                f'Invalid sort order "{sort_order}". Allowed values are: {", ".join(self._SORT_ORDERS)}')
+
+        return self.playlist_result(self._entries(show_id, sort_order), playlist_id=show_id)

yt_dlp/extractor/soundcloud.py

@@ -241,7 +241,7 @@ class SoundcloudBaseIE(InfoExtractor):
                     format_urls.add(format_url)
                     formats.append({
                         'format_id': 'download',
-                        'ext': urlhandle_detect_ext(urlh) or 'mp3',
+                        'ext': urlhandle_detect_ext(urlh, default='mp3'),
                         'filesize': int_or_none(urlh.headers.get('Content-Length')),
                         'url': format_url,
                         'quality': 10,

yt_dlp/extractor/uliza.py

@@ -0,0 +1,113 @@
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    int_or_none,
+    make_archive_id,
+    parse_qs,
+    time_seconds,
+)
+from ..utils.traversal import traverse_obj
+
+
+class UlizaPlayerIE(InfoExtractor):
+    _VALID_URL = r'https://player-api\.p\.uliza\.jp/v1/players/[^?#]+\?(?:[^#]*&)?name=(?P<id>[^#&]+)'
+    _TESTS = [{
+        'url': 'https://player-api.p.uliza.jp/v1/players/timeshift-disabled/pia/admin?type=normal&playerobjectname=ulizaPlayer&name=livestream01_dvr&repeatable=true',
+        'info_dict': {
+            'id': '88f3109a-f503-4d0f-a9f7-9f39ac745d84',
+            'ext': 'mp4',
+            'title': '88f3109a-f503-4d0f-a9f7-9f39ac745d84',
+            'live_status': 'was_live',
+            '_old_archive_ids': ['piaulizaportal 88f3109a-f503-4d0f-a9f7-9f39ac745d84'],
+        },
+    }, {
+        'url': 'https://player-api.p.uliza.jp/v1/players/uliza_jp_gallery_normal/promotion/admin?type=presentation&name=cookings&targetid=player1',
+        'info_dict': {
+            'id': 'ae350126-5e22-4a7f-a8ac-8d0fd448b800',
+            'ext': 'mp4',
+            'title': 'ae350126-5e22-4a7f-a8ac-8d0fd448b800',
+            'live_status': 'not_live',
+            '_old_archive_ids': ['piaulizaportal ae350126-5e22-4a7f-a8ac-8d0fd448b800'],
+        },
+    }, {
+        'url': 'https://player-api.p.uliza.jp/v1/players/default-player/pia/admin?type=normal&name=pia_movie_uliza_fix&targetid=ulizahtml5&repeatable=true',
+        'info_dict': {
+            'id': '0644ecc8-e354-41b4-b957-3b08a2d63df1',
+            'ext': 'mp4',
+            'title': '0644ecc8-e354-41b4-b957-3b08a2d63df1',
+            'live_status': 'not_live',
+            '_old_archive_ids': ['piaulizaportal 0644ecc8-e354-41b4-b957-3b08a2d63df1'],
+        },
+    }]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        player_data = self._download_webpage(
+            url, display_id, headers={'Referer': 'https://player-api.p.uliza.jp/'},
+            note='Fetching player data', errnote='Unable to fetch player data')
+
+        m3u8_url = self._search_regex(
+            r'["\'](https://vms-api\.p\.uliza\.jp/v1/prog-index\.m3u8[^"\']+)', player_data, 'm3u8 url')
+        video_id = parse_qs(m3u8_url).get('ss', [display_id])[0]
+
+        formats = self._extract_m3u8_formats(m3u8_url, video_id)
+        m3u8_type = self._search_regex(
+            r'/hls/(dvr|video)/', traverse_obj(formats, (0, 'url')), 'm3u8 type', default=None)
+        return {
+            'id': video_id,
+            'title': video_id,
+            'formats': formats,
+            'live_status': {
+                'video': 'is_live',
+                'dvr': 'was_live',  # short-term archives
+            }.get(m3u8_type, 'not_live'),  # VOD or long-term archives
+            '_old_archive_ids': [make_archive_id('PIAULIZAPortal', video_id)],
+        }
+
+
+class UlizaPortalIE(InfoExtractor):
+    IE_DESC = 'ulizaportal.jp'
+    _VALID_URL = r'https?://(?:www\.)?ulizaportal\.jp/pages/(?P<id>[\da-f]{8}-(?:[\da-f]{4}-){3}[\da-f]{12})'
+    _TESTS = [{
+        'url': 'https://ulizaportal.jp/pages/005f18b7-e810-5618-cb82-0987c5755d44',
+        'info_dict': {
+            'id': 'ae350126-5e22-4a7f-a8ac-8d0fd448b800',
+            'display_id': '005f18b7-e810-5618-cb82-0987c5755d44',
+            'title': 'プレゼンテーションプレイヤーのサンプル',
+            'live_status': 'not_live',
+            '_old_archive_ids': ['piaulizaportal ae350126-5e22-4a7f-a8ac-8d0fd448b800'],
+        },
+        'params': {
+            'skip_download': True,
+            'ignore_no_formats_error': True,
+        },
+    }, {
+        'url': 'https://ulizaportal.jp/pages/005e1b23-fe93-5780-19a0-98e917cc4b7d?expires=4102412400&signature=f422a993b683e1068f946caf406d211c17d1ef17da8bef3df4a519502155aa91&version=1',
+        'info_dict': {
+            'id': '0644ecc8-e354-41b4-b957-3b08a2d63df1',
+            'display_id': '005e1b23-fe93-5780-19a0-98e917cc4b7d',
+            'title': '【確認用】視聴サンプルページ（ULIZA）',
+            'live_status': 'not_live',
+            '_old_archive_ids': ['piaulizaportal 0644ecc8-e354-41b4-b957-3b08a2d63df1'],
+        },
+        'params': {
+            'skip_download': True,
+            'ignore_no_formats_error': True,
+        },
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        expires = int_or_none(traverse_obj(parse_qs(url), ('expires', 0)))
+        if expires and expires <= time_seconds():
+            raise ExtractorError('The link is expired', video_id=video_id, expected=True)
+
+        webpage = self._download_webpage(url, video_id)
+
+        player_data_url = self._search_regex(
+            r'<script [^>]*\bsrc="(https://player-api\.p\.uliza\.jp/v1/players/[^"]+)"',
+            webpage, 'player data url')
+        return self.url_result(
+            player_data_url, UlizaPlayerIE, url_transparent=True,
+            display_id=video_id, video_title=self._html_extract_title(webpage))

yt_dlp/extractor/youtube.py

@@ -5087,7 +5087,7 @@ class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
     def _rich_entries(self, rich_grid_renderer):
         renderer = traverse_obj(
             rich_grid_renderer,
-            ('content', ('videoRenderer', 'reelItemRenderer', 'playlistRenderer', 'shortsLockupViewModel'), any)) or {}
+            ('content', ('videoRenderer', 'reelItemRenderer', 'playlistRenderer', 'shortsLockupViewModel', 'lockupViewModel'), any)) or {}
         video_id = renderer.get('videoId')
         if video_id:
             yield self._extract_video(renderer)
@@ -5114,6 +5114,18 @@ class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
                 })),
                 thumbnails=self._extract_thumbnails(renderer, 'thumbnail', final_key='sources'))
             return
+        # lockupViewModel extraction
+        content_id = renderer.get('contentId')
+        if content_id and renderer.get('contentType') == 'LOCKUP_CONTENT_TYPE_PODCAST':
+            yield self.url_result(
+                f'https://www.youtube.com/playlist?list={content_id}',
+                ie=YoutubeTabIE, video_id=content_id,
+                **traverse_obj(renderer, {
+                    'title': ('metadata', 'lockupMetadataViewModel', 'title', 'content', {str}),
+                }),
+                thumbnails=self._extract_thumbnails(renderer, (
+                    'contentImage', 'collectionThumbnailViewModel', 'primaryThumbnail', 'thumbnailViewModel', 'image'), final_key='sources'))
+            return
 
     def _video_entry(self, video_renderer):
         video_id = video_renderer.get('videoId')
@@ -6706,22 +6718,22 @@ class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
         },
         'playlist_count': 0,
     }, {
-        # Podcasts tab, with rich entry playlistRenderers
+        # Podcasts tab, with rich entry lockupViewModel
         'url': 'https://www.youtube.com/@99percentinvisiblepodcast/podcasts',
         'info_dict': {
             'id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
             'channel_id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
             'uploader_url': 'https://www.youtube.com/@99percentinvisiblepodcast',
             'description': 'md5:3a0ed38f1ad42a68ef0428c04a15695c',
-            'title': '99 Percent Invisible - Podcasts',
+            'title': '99% Invisible - Podcasts',
-            'uploader': '99 Percent Invisible',
+            'uploader': '99% Invisible',
             'channel_follower_count': int,
             'channel_url': 'https://www.youtube.com/channel/UCVMF2HD4ZgC0QHpU9Yq5Xrw',
             'tags': [],
-            'channel': '99 Percent Invisible',
+            'channel': '99% Invisible',
             'uploader_id': '@99percentinvisiblepodcast',
         },
-        'playlist_count': 0,
+        'playlist_count': 5,
     }, {
         # Releases tab, with rich entry playlistRenderers (same as Podcasts tab)
         'url': 'https://www.youtube.com/@AHimitsu/releases',

yt_dlp/options.py

@@ -419,7 +419,9 @@ def create_parser():
     general.add_option(
         '--flat-playlist',
         action='store_const', dest='extract_flat', const='in_playlist', default=False,
-        help='Do not extract the videos of a playlist, only list them')
+        help=(
+            'Do not extract a playlist\'s URL result entries; '
+            'some entry metadata may be missing and downloading may be bypassed'))
     general.add_option(
         '--no-flat-playlist',
         action='store_false', dest='extract_flat',
@@ -1368,12 +1370,12 @@ def create_parser():
         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,