Merge branch 'yt-dlp:master' into networking/curl-impersonate

Authored by: bashonly

.github/workflows/build.yml

@@ -250,7 +250,7 @@ jobs:
           # We need to fuse our own universal2 wheels for curl_cffi
           python3 -m pip install -U --user delocate
           mkdir curl_cffi_whls curl_cffi_universal2
-          python3 devscripts/install_deps.py --print -o --include impersonate > requirements.txt
+          python3 devscripts/install_deps.py --print -o --include curl_cffi > requirements.txt
           for platform in "macosx_11_0_arm64" "macosx_11_0_x86_64"; do
             python3 -m pip download \
               --only-binary=:all: \
@@ -316,7 +316,7 @@ jobs:
         run: |
           brew install coreutils
           python3 devscripts/install_deps.py --user -o --include build
-          python3 devscripts/install_deps.py --user --include pyinstaller --include impersonate
+          python3 devscripts/install_deps.py --user --include pyinstaller --include curl_cffi
 
       - name: Prepare
         run: |
@@ -358,7 +358,7 @@ jobs:
       - name: Install Requirements
         run: | # Custom pyinstaller built with https://github.com/yt-dlp/pyinstaller-builds
           python devscripts/install_deps.py -o --include build
-          python devscripts/install_deps.py --include py2exe --include impersonate
+          python devscripts/install_deps.py --include py2exe --include curl_cffi
           python -m pip install -U "https://yt-dlp.github.io/Pyinstaller-Builds/x86_64/pyinstaller-5.8.0-py3-none-any.whl"
 
       - name: Prepare

Makefile

@@ -38,11 +38,13 @@ MANDIR ?= $(PREFIX)/man
 SHAREDIR ?= $(PREFIX)/share
 PYTHON ?= /usr/bin/env python3
 
-# set SYSCONFDIR to /etc if PREFIX=/usr or PREFIX=/usr/local
-SYSCONFDIR = $(shell if [ $(PREFIX) = /usr -o $(PREFIX) = /usr/local ]; then echo /etc; else echo $(PREFIX)/etc; fi)
+# $(shell) and $(error) are no-ops in BSD Make and the != variable assignment operator is not supported by GNU Make <4.0
+VERSION_CHECK != echo supported
+VERSION_CHECK ?= $(error GNU Make 4+ or BSD Make is required)
+CHECK_VERSION := $(VERSION_CHECK)
 
-# set markdown input format to "markdown-smart" for pandoc version 2 and to "markdown" for pandoc prior to version 2
-MARKDOWN = $(shell if [ `pandoc -v | head -n1 | cut -d" " -f2 | head -c1` = "2" ]; then echo markdown-smart; else echo markdown; fi)
+# set markdown input format to "markdown-smart" for pandoc version 2+ and to "markdown" for pandoc prior to version 2
+MARKDOWN != if [ "`pandoc -v | head -n1 | cut -d' ' -f2 | head -c1`" -ge "2" ]; then echo markdown-smart; else echo markdown; fi
 
 install: lazy-extractors yt-dlp yt-dlp.1 completions
 	mkdir -p $(DESTDIR)$(BINDIR)
@@ -73,17 +75,17 @@ test:
 offlinetest: codetest
 	$(PYTHON) -m pytest -k "not download"
 
-CODE_FOLDERS := $(shell find yt_dlp -type d -not -name '__*' -exec sh -c 'test -e "$$1"/__init__.py' sh {} \; -print)
-CODE_FILES := $(shell for f in $(CODE_FOLDERS); do echo "$$f" | awk '{gsub(/\/[^\/]+/,"/*"); print $$1"/*.py"}'; done | sort -u)
+CODE_FOLDERS != find yt_dlp -type f -name '__init__.py' -exec dirname {} \+ | grep -v '/__' | sort
+CODE_FILES != for f in $(CODE_FOLDERS) ; do echo "$$f" | sed 's,$$,/*.py,' ; done
 yt-dlp: $(CODE_FILES)
 	mkdir -p zip
 	for d in $(CODE_FOLDERS) ; do \
 	  mkdir -p zip/$$d ;\
 	  cp -pPR $$d/*.py zip/$$d/ ;\
 	done
-	cd zip ; touch -t 200001010101 $(CODE_FILES)
+	(cd zip && touch -t 200001010101 $(CODE_FILES))
 	mv zip/yt_dlp/__main__.py zip/
-	cd zip ; zip -q ../yt-dlp $(CODE_FILES) __main__.py
+	(cd zip && zip -q ../yt-dlp $(CODE_FILES) __main__.py)
 	rm -rf zip
 	echo '#!$(PYTHON)' > yt-dlp
 	cat yt-dlp.zip >> yt-dlp
@@ -127,7 +129,7 @@ completions/fish/yt-dlp.fish: $(CODE_FILES) devscripts/fish-completion.in
 	mkdir -p completions/fish
 	$(PYTHON) devscripts/fish-completion.py
 
-_EXTRACTOR_FILES = $(shell find yt_dlp/extractor -name '*.py' -and -not -name 'lazy_extractors.py')
+_EXTRACTOR_FILES != find yt_dlp/extractor -name '*.py' -and -not -name 'lazy_extractors.py'
 yt_dlp/extractor/lazy_extractors.py: devscripts/make_lazy_extractors.py devscripts/lazy_load_template.py $(_EXTRACTOR_FILES)
 	$(PYTHON) devscripts/make_lazy_extractors.py $@
 
@@ -141,6 +143,7 @@ yt-dlp.tar.gz: all
 		--exclude '__pycache__' \
 		--exclude '.pytest_cache' \
 		--exclude '.git' \
+		--exclude '__pyinstaller' \
 		-- \
 		README.md supportedsites.md Changelog.md LICENSE \
 		CONTRIBUTING.md Collaborators.md CONTRIBUTORS AUTHORS \

pyproject.toml

@@ -51,7 +51,7 @@ dependencies = [
 ]
 
 [project.optional-dependencies]
-impersonate = ["curl_cffi>=0.5.10,<0.6.0"]
+curl_cffi = ["curl_cffi>=0.5.10,<0.6.0"]
 secretstorage = [
     "cffi",
     "secretstorage",

yt_dlp/extractor/antenna.py

@@ -78,14 +78,14 @@ class Ant1NewsGrArticleIE(AntennaBaseIE):
 
     _TESTS = [{
         'url': 'https://www.ant1news.gr/afieromata/article/549468/o-tzeims-mpont-sta-meteora-oi-apeiles-kai-o-xesikomos-ton-kalogeron',
-        'md5': '294f18331bb516539d72d85a82887dcc',
+        'md5': '57eb8d12181f0fa2b14b0b138e1de9b6',
         'info_dict': {
             'id': '_xvg/m_cmbatw=',
             'ext': 'mp4',
             'title': 'md5:a93e8ecf2e4073bfdffcb38f59945411',
-            'timestamp': 1603092840,
-            'upload_date': '20201019',
-            'thumbnail': 'https://ant1media.azureedge.net/imgHandler/640/756206d2-d640-40e2-b201-3555abdfc0db.jpg',
+            'timestamp': 1666166520,
+            'upload_date': '20221019',
+            'thumbnail': 'https://ant1media.azureedge.net/imgHandler/1920/756206d2-d640-40e2-b201-3555abdfc0db.jpg',
         },
     }, {
         'url': 'https://ant1news.gr/Society/article/620286/symmoria-anilikon-dikigoros-thymaton-ithelan-na-toys-apoteleiosoyn',
@@ -117,7 +117,7 @@ class Ant1NewsGrEmbedIE(AntennaBaseIE):
     _BASE_PLAYER_URL_RE = r'(?:https?:)?//(?:[a-zA-Z0-9\-]+\.)?(?:antenna|ant1news)\.gr/templates/pages/player'
     _VALID_URL = rf'{_BASE_PLAYER_URL_RE}\?([^#]+&)?cid=(?P<id>[^#&]+)'
     _EMBED_REGEX = [rf'<iframe[^>]+?src=(?P<_q1>["\'])(?P<url>{_BASE_PLAYER_URL_RE}\?(?:(?!(?P=_q1)).)+)(?P=_q1)']
-    _API_PATH = '/news/templates/data/jsonPlayer'
+    _API_PATH = '/templates/data/jsonPlayer'
 
     _TESTS = [{
         'url': 'https://www.antenna.gr/templates/pages/player?cid=3f_li_c_az_jw_y_u=&w=670&h=377',

yt_dlp/extractor/crooksandliars.py

@@ -33,10 +33,7 @@ class CrooksAndLiarsIE(InfoExtractor):
         webpage = self._download_webpage(
             'http://embed.crooksandliars.com/embed/%s' % video_id, video_id)
 
-        manifest = self._parse_json(
-            self._search_regex(
-                r'var\s+manifest\s*=\s*({.+?})\n', webpage, 'manifest JSON'),
-            video_id)
+        manifest = self._search_json(r'var\s+manifest\s*=', webpage, 'manifest JSON', video_id)
 
         quality = qualities(('webm_low', 'mp4_low', 'webm_high', 'mp4_high'))
 

yt_dlp/extractor/facebook.py

@@ -500,6 +500,7 @@ class FacebookIE(InfoExtractor):
                 webpage, 'description', default=None)
             uploader_data = (
                 get_first(media, ('owner', {dict}))
+                or get_first(post, ('video', 'creation_story', 'attachments', ..., 'media', lambda k, v: k == 'owner' and v['name']))
                 or get_first(post, (..., 'video', lambda k, v: k == 'owner' and v['name']))
                 or get_first(post, ('node', 'actors', ..., {dict}))
                 or get_first(post, ('event', 'event_creator', {dict})) or {})
@@ -583,8 +584,8 @@ class FacebookIE(InfoExtractor):
         def extract_relay_prefetched_data(_filter):
             return traverse_obj(extract_relay_data(_filter), (
                 'require', (None, (..., ..., ..., '__bbox', 'require')),
-                lambda _, v: 'RelayPrefetchedStreamCache' in v, ..., ...,
-                '__bbox', 'result', 'data', {dict}), get_all=False) or {}
+                lambda _, v: any(key.startswith('RelayPrefetchedStreamCache') for key in v),
+                ..., ..., '__bbox', 'result', 'data', {dict}), get_all=False) or {}
 
         if not video_data:
             server_js_data = self._parse_json(self._search_regex([

yt_dlp/extractor/funk.py

@@ -1,25 +1,29 @@
 from .common import InfoExtractor
 from .nexx import NexxIE
-from ..utils import (
-    int_or_none,
-    str_or_none,
-)
 
 
 class FunkIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.|origin\.)?funk\.net/(?:channel|playlist)/[^/]+/(?P<display_id>[0-9a-z-]+)-(?P<id>\d+)'
     _TESTS = [{
         'url': 'https://www.funk.net/channel/ba-793/die-lustigsten-instrumente-aus-dem-internet-teil-2-1155821',
-        'md5': '8dd9d9ab59b4aa4173b3197f2ea48e81',
+        'md5': '8610449476156f338761a75391b0017d',
         'info_dict': {
             'id': '1155821',
             'ext': 'mp4',
             'title': 'Die LUSTIGSTEN INSTRUMENTE aus dem Internet - Teil 2',
-            'description': 'md5:a691d0413ef4835588c5b03ded670c1f',
+            'description': 'md5:2a03b67596eda0d1b5125c299f45e953',
             'timestamp': 1514507395,
             'upload_date': '20171229',
+            'duration': 426.0,
+            'cast': ['United Creators PMB GmbH'],
+            'thumbnail': 'https://assets.nexx.cloud/media/75/56/79/3YKUSJN1LACN0CRxL.jpg',
+            'display_id': 'die-lustigsten-instrumente-aus-dem-internet-teil-2',
+            'alt_title': 'Die LUSTIGSTEN INSTRUMENTE aus dem Internet Teil 2',
+            'season_number': 0,
+            'season': 'Season 0',
+            'episode_number': 0,
+            'episode': 'Episode 0',
         },
-
     }, {
         'url': 'https://www.funk.net/playlist/neuesteVideos/kameras-auf-dem-fusion-festival-1618699',
         'only_matching': True,
@@ -27,18 +31,10 @@ class FunkIE(InfoExtractor):
 
     def _real_extract(self, url):
         display_id, nexx_id = self._match_valid_url(url).groups()
-        video = self._download_json(
-            'https://www.funk.net/api/v4.0/videos/' + nexx_id, nexx_id)
         return {
             '_type': 'url_transparent',
-            'url': 'nexx:741:' + nexx_id,
+            'url': f'nexx:741:{nexx_id}',
             'ie_key': NexxIE.ie_key(),
             'id': nexx_id,
-            'title': video.get('title'),
-            'description': video.get('description'),
-            'duration': int_or_none(video.get('duration')),
-            'channel_id': str_or_none(video.get('channelId')),
             'display_id': display_id,
-            'tags': video.get('tags'),
-            'thumbnail': video.get('imageUrlLandscape'),
         }

yt_dlp/extractor/lefigaro.py

@@ -13,7 +13,7 @@ class LeFigaroVideoEmbedIE(InfoExtractor):
 
     _TESTS = [{
         'url': 'https://video.lefigaro.fr/embed/figaro/video/les-francais-ne-veulent-ils-plus-travailler-suivez-en-direct-le-club-le-figaro-idees/',
-        'md5': 'e94de44cd80818084352fcf8de1ce82c',
+        'md5': 'a0c3069b7e4c4526abf0053a7713f56f',
         'info_dict': {
             'id': 'g9j7Eovo',
             'title': 'Les Français ne veulent-ils plus travailler ? Retrouvez Le Club Le Figaro Idées',
@@ -26,7 +26,7 @@ class LeFigaroVideoEmbedIE(InfoExtractor):
         },
     }, {
         'url': 'https://video.lefigaro.fr/embed/figaro/video/intelligence-artificielle-faut-il-sen-mefier/',
-        'md5': '0b3f10332b812034b3a3eda1ef877c5f',
+        'md5': '319c662943dd777bab835cae1e2d73a5',
         'info_dict': {
             'id': 'LeAgybyc',
             'title': 'Intelligence artificielle : faut-il s’en méfier ?',
@@ -41,7 +41,7 @@ class LeFigaroVideoEmbedIE(InfoExtractor):
 
     _WEBPAGE_TESTS = [{
         'url': 'https://video.lefigaro.fr/figaro/video/suivez-en-direct-le-club-le-figaro-international-avec-philippe-gelie-9/',
-        'md5': '3972ddf2d5f8b98699f191687258e2f9',
+        'md5': '6289f9489efb969e38245f31721596fe',
         'info_dict': {
             'id': 'QChnbPYA',
             'title': 'Où en est le couple franco-allemand ? Retrouvez Le Club Le Figaro International',
@@ -55,7 +55,7 @@ class LeFigaroVideoEmbedIE(InfoExtractor):
         },
     }, {
         'url': 'https://video.lefigaro.fr/figaro/video/la-philosophe-nathalie-sarthou-lajus-est-linvitee-du-figaro-live/',
-        'md5': '3ac0a0769546ee6be41ab52caea5d9a9',
+        'md5': 'f6df814cae53e85937621599d2967520',
         'info_dict': {
             'id': 'QJzqoNbf',
             'title': 'La philosophe Nathalie Sarthou-Lajus est l’invitée du Figaro Live',
@@ -73,7 +73,8 @@ class LeFigaroVideoEmbedIE(InfoExtractor):
         display_id = self._match_id(url)
         webpage = self._download_webpage(url, display_id)
 
-        player_data = self._search_nextjs_data(webpage, display_id)['props']['pageProps']['pageData']['playerData']
+        player_data = self._search_nextjs_data(
+            webpage, display_id)['props']['pageProps']['initialProps']['pageData']['playerData']
 
         return self.url_result(
             f'jwplatform:{player_data["videoId"]}', title=player_data.get('title'),

yt_dlp/extractor/linkedin.py

@@ -3,16 +3,15 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
-    clean_html,
-    extract_attributes,
     ExtractorError,
+    extract_attributes,
     float_or_none,
-    get_element_by_class,
     int_or_none,
     srt_subtitles_timecode,
-    strip_or_none,
     mimetype2ext,
+    traverse_obj,
     try_get,
+    url_or_none,
     urlencode_postdata,
     urljoin,
 )
@@ -83,15 +82,29 @@ class LinkedInLearningBaseIE(LinkedInBaseIE):
 
 
 class LinkedInIE(LinkedInBaseIE):
-    _VALID_URL = r'https?://(?:www\.)?linkedin\.com/posts/.+?(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?linkedin\.com/posts/[^/?#]+-(?P<id>\d+)-\w{4}/?(?:[?#]|$)'
     _TESTS = [{
         'url': 'https://www.linkedin.com/posts/mishalkhawaja_sendinblueviews-toronto-digitalmarketing-ugcPost-6850898786781339649-mM20',
         'info_dict': {
             'id': '6850898786781339649',
             'ext': 'mp4',
-            'title': 'Mishal K. on LinkedIn: #sendinblueviews #toronto #digitalmarketing',
-            'description': 'md5:be125430bab1c574f16aeb186a4d5b19',
-            'creator': 'Mishal K.'
+            'title': 'Mishal K. on LinkedIn: #sendinblueviews #toronto #digitalmarketing #nowhiring #sendinblue…',
+            'description': 'md5:2998a31f6f479376dd62831f53a80f71',
+            'uploader': 'Mishal K.',
+            'thumbnail': 're:^https?://media.licdn.com/dms/image/.*$',
+            'like_count': int
+        },
+    }, {
+        'url': 'https://www.linkedin.com/posts/the-mathworks_2_what-is-mathworks-cloud-center-activity-7151241570371948544-4Gu7',
+        'info_dict': {
+            'id': '7151241570371948544',
+            'ext': 'mp4',
+            'title': 'MathWorks on LinkedIn: What Is MathWorks Cloud Center?',
+            'description': 'md5:95f9d4eeb6337882fb47eefe13d7a40c',
+            'uploader': 'MathWorks',
+            'thumbnail': 're:^https?://media.licdn.com/dms/image/.*$',
+            'like_count': int,
+            'subtitles': 'mincount:1'
         },
     }]
 
@@ -99,26 +112,30 @@ class LinkedInIE(LinkedInBaseIE):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
-        title = self._html_extract_title(webpage)
-        description = clean_html(get_element_by_class('share-update-card__update-text', webpage))
-        like_count = int_or_none(get_element_by_class('social-counts-reactions__social-counts-numRections', webpage))
-        creator = strip_or_none(clean_html(get_element_by_class('comment__actor-name', webpage)))
-
-        sources = self._parse_json(extract_attributes(self._search_regex(r'(<video[^>]+>)', webpage, 'video'))['data-sources'], video_id)
+        video_attrs = extract_attributes(self._search_regex(r'(<video[^>]+>)', webpage, 'video'))
+        sources = self._parse_json(video_attrs['data-sources'], video_id)
         formats = [{
             'url': source['src'],
             'ext': mimetype2ext(source.get('type')),
             'tbr': float_or_none(source.get('data-bitrate'), scale=1000),
         } for source in sources]
+        subtitles = {'en': [{
+            'url': video_attrs['data-captions-url'],
+            'ext': 'vtt',
+        }]} if url_or_none(video_attrs.get('data-captions-url')) else {}
 
         return {
             'id': video_id,
             'formats': formats,
-            'title': title,
-            'like_count': like_count,
-            'creator': creator,
+            'title': self._og_search_title(webpage, default=None) or self._html_extract_title(webpage),
+            'like_count': int_or_none(self._search_regex(
+                r'\bdata-num-reactions="(\d+)"', webpage, 'reactions', default=None)),
+            'uploader': traverse_obj(
+                self._yield_json_ld(webpage, video_id),
+                (lambda _, v: v['@type'] == 'SocialMediaPosting', 'author', 'name', {str}), get_all=False),
             'thumbnail': self._og_search_thumbnail(webpage),
-            'description': description,
+            'description': self._og_search_description(webpage, default=None),
+            'subtitles': subtitles,
         }
 
 

yt_dlp/extractor/magellantv.py

@@ -28,12 +28,24 @@ class MagellanTVIE(InfoExtractor):
             'tags': ['Investigation', 'True Crime', 'Justice', 'Europe'],
         },
         'params': {'skip_download': 'm3u8'},
+    }, {
+        'url': 'https://www.magellantv.com/watch/celebration-nation',
+        'info_dict': {
+            'id': 'celebration-nation',
+            'ext': 'mp4',
+            'tags': ['Art & Culture', 'Human Interest', 'Anthropology', 'China', 'History'],
+            'duration': 2640.0,
+            'title': 'Ancestors',
+        },
+        'params': {'skip_download': 'm3u8'},
     }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
-        data = self._search_nextjs_data(webpage, video_id)['props']['pageProps']['reactContext']['video']['detail']
+        data = traverse_obj(self._search_nextjs_data(webpage, video_id), (
+            'props', 'pageProps', 'reactContext',
+            (('video', 'detail'), ('series', 'currentEpisode')), {dict}), get_all=False)
         formats, subtitles = self._extract_m3u8_formats_and_subtitles(data['jwpVideoUrl'], video_id)
 
         return {

yt_dlp/extractor/rai.py

@@ -1,6 +1,7 @@
 import re
 
 from .common import InfoExtractor
+from ..networking import HEADRequest
 from ..utils import (
     clean_html,
     determine_ext,
@@ -91,7 +92,7 @@ class RaiBaseIE(InfoExtractor):
             self.raise_geo_restricted(countries=self._GEO_COUNTRIES, metadata_available=True)
 
         if not audio_only and not is_live:
-            formats.extend(self._create_http_urls(media_url, relinker_url, formats))
+            formats.extend(self._create_http_urls(media_url, relinker_url, formats, video_id))
 
         return filter_dict({
             'is_live': is_live,
@@ -99,7 +100,7 @@ class RaiBaseIE(InfoExtractor):
             'formats': formats,
         })
 
-    def _create_http_urls(self, manifest_url, relinker_url, fmts):
+    def _create_http_urls(self, manifest_url, relinker_url, fmts, video_id):
         _MANIFEST_REG = r'/(?P<id>\w+)(?:_(?P<quality>[\d\,]+))?(?:\.mp4)?(?:\.csmil)?/playlist\.m3u8'
         _MP4_TMPL = '%s&overrideUserAgentRule=mp4-%s'
         _QUALITY = {
@@ -166,6 +167,14 @@ class RaiBaseIE(InfoExtractor):
                 'fps': 25,
             }
 
+        # Check if MP4 download is available
+        try:
+            self._request_webpage(
+                HEADRequest(_MP4_TMPL % (relinker_url, '*')), video_id, 'Checking MP4 availability')
+        except ExtractorError as e:
+            self.to_screen(f'{video_id}: MP4 direct download is not available: {e.cause}')
+            return []
+
         # filter out single-stream formats
         fmts = [f for f in fmts
                 if not f.get('vcodec') == 'none' and not f.get('acodec') == 'none']

yt_dlp/extractor/svt.py

@@ -7,8 +7,6 @@ from ..utils import (
     determine_ext,
     dict_get,
     int_or_none,
-    str_or_none,
-    strip_or_none,
     traverse_obj,
     try_get,
     unified_timestamp,
@@ -388,15 +386,55 @@ class SVTSeriesIE(SVTPlayBaseIE):
             dict_get(series, ('longDescription', 'shortDescription')))
 
 
-class SVTPageIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?svt\.se/(?P<path>(?:[^/]+/)*(?P<id>[^/?&#]+))'
+class SVTPageIE(SVTBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?svt\.se/(?:[^/?#]+/)*(?P<id>[^/?&#]+)'
     _TESTS = [{
+        'url': 'https://www.svt.se/nyheter/lokalt/skane/viktor-18-forlorade-armar-och-ben-i-sepsis-vill-ateruppta-karaten-och-bli-svetsare',
+        'info_dict': {
+            'title': 'Viktor, 18, förlorade armar och ben i sepsis – vill återuppta karaten och bli svetsare',
+            'id': 'viktor-18-forlorade-armar-och-ben-i-sepsis-vill-ateruppta-karaten-och-bli-svetsare',
+        },
+        'playlist_count': 2,
+    }, {
+        'url': 'https://www.svt.se/nyheter/lokalt/skane/forsvarsmakten-om-trafikkaoset-pa-e22-kunde-inte-varit-dar-snabbare',
+        'info_dict': {
+            'id': 'jXvk42E',
+            'title': 'Försvarsmakten om trafikkaoset på E22: Kunde inte varit där snabbare',
+            'ext': 'mp4',
+            "duration": 80,
+            'age_limit': 0,
+            'timestamp': 1704370009,
+            'episode': 'Försvarsmakten om trafikkaoset på E22: Kunde inte varit där snabbare',
+            'series': 'Lokala Nyheter Skåne',
+            'upload_date': '20240104'
+        },
+        'params': {
+            'skip_download': True,
+        }
+    }, {
+        'url': 'https://www.svt.se/nyheter/svtforum/2023-tungt-ar-for-svensk-media',
+        'info_dict': {
+            'title': '2023 tungt år för svensk media',
+            'id': 'ewqAZv4',
+            'ext': 'mp4',
+            "duration": 3074,
+            'age_limit': 0,
+            'series': '',
+            'timestamp': 1702980479,
+            'upload_date': '20231219',
+            'episode': 'Mediestudier'
+        },
+        'params': {
+            'skip_download': True,
+        }
+    }, {
         'url': 'https://www.svt.se/sport/ishockey/bakom-masken-lehners-kamp-mot-mental-ohalsa',
         'info_dict': {
             'id': '25298267',
             'title': 'Bakom masken – Lehners kamp mot mental ohälsa',
         },
         'playlist_count': 4,
+        'skip': 'Video is gone'
     }, {
         'url': 'https://www.svt.se/nyheter/utrikes/svenska-andrea-ar-en-mil-fran-branderna-i-kalifornien',
         'info_dict': {
@@ -404,6 +442,7 @@ class SVTPageIE(InfoExtractor):
             'title': 'Svenska Andrea redo att fly sitt hem i Kalifornien',
         },
         'playlist_count': 2,
+        'skip': 'Video is gone'
     }, {
         # only programTitle
         'url': 'http://www.svt.se/sport/ishockey/jagr-tacklar-giroux-under-intervjun',
@@ -414,6 +453,7 @@ class SVTPageIE(InfoExtractor):
             'duration': 27,
             'age_limit': 0,
         },
+        'skip': 'Video is gone'
     }, {
         'url': 'https://www.svt.se/nyheter/lokalt/vast/svt-testar-tar-nagon-upp-skrapet-1',
         'only_matching': True,
@@ -427,26 +467,23 @@ class SVTPageIE(InfoExtractor):
         return False if SVTIE.suitable(url) or SVTPlayIE.suitable(url) else super(SVTPageIE, cls).suitable(url)
 
     def _real_extract(self, url):
-        path, display_id = self._match_valid_url(url).groups()
+        display_id = self._match_id(url)
 
-        article = self._download_json(
-            'https://api.svt.se/nss-api/page/' + path, display_id,
-            query={'q': 'articles'})['articles']['content'][0]
+        webpage = self._download_webpage(url, display_id)
+        title = self._og_search_title(webpage)
 
-        entries = []
+        urql_state = self._search_json(
+            r'window\.svt\.nyh\.urqlState\s*=', webpage, 'json data', display_id)
 
-        def _process_content(content):
-            if content.get('_type') in ('VIDEOCLIP', 'VIDEOEPISODE'):
-                video_id = compat_str(content['image']['svtId'])
-                entries.append(self.url_result(
-                    'svt:' + video_id, SVTPlayIE.ie_key(), video_id))
+        data = traverse_obj(urql_state, (..., 'data', {str}, {json.loads}), get_all=False) or {}
 
-        for media in article.get('media', []):
-            _process_content(media)
+        def entries():
+            for video_id in set(traverse_obj(data, (
+                'page', (('topMedia', 'svtId'), ('body', ..., 'video', 'svtId')), {str}
+            ))):
+                info = self._extract_video(
+                    self._download_json(f'https://api.svt.se/video/{video_id}', video_id), video_id)
+                info['title'] = title
+                yield info
 
-        for obj in article.get('structuredBody', []):
-            _process_content(obj.get('content') or {})
-
-        return self.playlist_result(
-            entries, str_or_none(article.get('id')),
-            strip_or_none(article.get('title')))
+        return self.playlist_result(entries(), display_id, title)