Makefile

@@ -38,13 +38,11 @@ MANDIR ?= $(PREFIX)/man
 SHAREDIR ?= $(PREFIX)/share
 PYTHON ?= /usr/bin/env python3
 
-# $(shell) and $(error) are no-ops in BSD Make and the != variable assignment operator is not supported by GNU Make <4.0
+# set SYSCONFDIR to /etc if PREFIX=/usr or PREFIX=/usr/local
-VERSION_CHECK != echo supported
+SYSCONFDIR = $(shell if [ $(PREFIX) = /usr -o $(PREFIX) = /usr/local ]; then echo /etc; else echo $(PREFIX)/etc; fi)
-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
+# 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
+MARKDOWN = $(shell 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)
@@ -75,17 +73,17 @@ test:
 offlinetest: codetest
 	$(PYTHON) -m pytest -k "not download"
 
-CODE_FOLDERS != find yt_dlp -type f -name '__init__.py' -exec dirname {} \+ | grep -v '/__' | sort
+CODE_FOLDERS := $(shell find yt_dlp -type d -not -name '__*' -exec sh -c 'test -e "$$1"/__init__.py' sh {} \; -print)
-CODE_FILES != for f in $(CODE_FOLDERS) ; do echo "$$f" | sed 's,$$,/*.py,' ; done
+CODE_FILES := $(shell for f in $(CODE_FOLDERS); do echo "$$f" | awk '{gsub(/\/[^\/]+/,"/*"); print $$1"/*.py"}'; done | sort -u)
 yt-dlp: $(CODE_FILES)
 	mkdir -p zip
 	for d in $(CODE_FOLDERS) ; do \
 	  mkdir -p zip/$$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
@@ -129,7 +127,7 @@ completions/fish/yt-dlp.fish: $(CODE_FILES) devscripts/fish-completion.in
 	mkdir -p completions/fish
 	$(PYTHON) devscripts/fish-completion.py
 
-_EXTRACTOR_FILES != find yt_dlp/extractor -name '*.py' -and -not -name 'lazy_extractors.py'
+_EXTRACTOR_FILES = $(shell 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 $@
 
@@ -143,7 +141,6 @@ 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 \

yt_dlp/extractor/facebook.py

@@ -500,7 +500,6 @@ 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 {})
@@ -584,8 +583,8 @@ class FacebookIE(InfoExtractor):
         def extract_relay_prefetched_data(_filter):
             return traverse_obj(extract_relay_data(_filter), (
                 'require', (None, (..., ..., ..., '__bbox', 'require')),
-                lambda _, v: any(key.startswith('RelayPrefetchedStreamCache') for key in v),
+                lambda _, v: 'RelayPrefetchedStreamCache' in v, ..., ...,
-                ..., ..., '__bbox', 'result', 'data', {dict}), get_all=False) or {}
+                '__bbox', 'result', 'data', {dict}), get_all=False) or {}
 
         if not video_data:
             server_js_data = self._parse_json(self._search_regex([

yt_dlp/extractor/linkedin.py

@@ -3,15 +3,16 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
-    ExtractorError,
+    clean_html,
     extract_attributes,
+    ExtractorError,
     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,
 )
@@ -82,29 +83,15 @@ class LinkedInLearningBaseIE(LinkedInBaseIE):
 
 
 class LinkedInIE(LinkedInBaseIE):
-    _VALID_URL = r'https?://(?:www\.)?linkedin\.com/posts/[^/?#]+-(?P<id>\d+)-\w{4}/?(?:[?#]|$)'
+    _VALID_URL = r'https?://(?:www\.)?linkedin\.com/posts/.+?(?P<id>\d+)'
     _TESTS = [{
         'url': 'https://www.linkedin.com/posts/mishalkhawaja_sendinblueviews-toronto-digitalmarketing-ugcPost-6850898786781339649-mM20',
         'info_dict': {
             'id': '6850898786781339649',
             'ext': 'mp4',
-            'title': 'Mishal K. on LinkedIn: #sendinblueviews #toronto #digitalmarketing #nowhiring #sendinblue…',
+            'title': 'Mishal K. on LinkedIn: #sendinblueviews #toronto #digitalmarketing',
-            'description': 'md5:2998a31f6f479376dd62831f53a80f71',
+            'description': 'md5:be125430bab1c574f16aeb186a4d5b19',
-            'uploader': 'Mishal K.',
+            'creator': 'Mishal K.'
-            'thumbnail': 're:^https?://media.licdn.com/dms/image/.*$',
-            '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'
         },
     }]
 
@@ -112,30 +99,26 @@ class LinkedInIE(LinkedInBaseIE):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
-        video_attrs = extract_attributes(self._search_regex(r'(<video[^>]+>)', webpage, 'video'))
+        title = self._html_extract_title(webpage)
-        sources = self._parse_json(video_attrs['data-sources'], video_id)
+        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)
         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': self._og_search_title(webpage, default=None) or self._html_extract_title(webpage),
+            'title': title,
-            'like_count': int_or_none(self._search_regex(
+            'like_count': like_count,
-                r'\bdata-num-reactions="(\d+)"', webpage, 'reactions', default=None)),
+            'creator': creator,
-            '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': self._og_search_description(webpage, default=None),
+            'description': description,
-            'subtitles': subtitles,
         }
 
 

yt_dlp/extractor/rai.py

@@ -1,7 +1,6 @@
 import re
 
 from .common import InfoExtractor
-from ..networking import HEADRequest
 from ..utils import (
     clean_html,
     determine_ext,
@@ -92,7 +91,7 @@ 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, video_id))
+            formats.extend(self._create_http_urls(media_url, relinker_url, formats))
 
         return filter_dict({
             'is_live': is_live,
@@ -100,7 +99,7 @@ class RaiBaseIE(InfoExtractor):
             'formats': formats,
         })
 
-    def _create_http_urls(self, manifest_url, relinker_url, fmts, video_id):
+    def _create_http_urls(self, manifest_url, relinker_url, fmts):
         _MANIFEST_REG = r'/(?P<id>\w+)(?:_(?P<quality>[\d\,]+))?(?:\.mp4)?(?:\.csmil)?/playlist\.m3u8'
         _MP4_TMPL = '%s&overrideUserAgentRule=mp4-%s'
         _QUALITY = {
@@ -167,14 +166,6 @@ 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']