yt_dlp/extractor/linkedin.py

@@ -3,8 +3,8 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
-    ExtractorError,
     extract_attributes,
+    ExtractorError,
     float_or_none,
     int_or_none,
     srt_subtitles_timecode,
@@ -82,27 +82,29 @@ class LinkedInLearningBaseIE(LinkedInBaseIE):
 
 
 class LinkedInIE(LinkedInBaseIE):
-    _VALID_URL = r'https?://(?:www\.)?linkedin\.com/posts/[^/?#]+-(?P<id>\d+)-\w{4}/?(?:[?#]|$)'
+    _VALID_URL = r'https?://(?:www\.)?linkedin\.com/posts/(?P<id>[^/?#]+)'
     _TESTS = [{
         'url': 'https://www.linkedin.com/posts/mishalkhawaja_sendinblueviews-toronto-digitalmarketing-ugcPost-6850898786781339649-mM20',
         'info_dict': {
-            'id': '6850898786781339649',
+            'id': 'C4E05AQG7hCp7zIeciw',
+            'display_id': 'mishalkhawaja_sendinblueviews-toronto-digitalmarketing-ugcPost-6850898786781339649-mM20',
             'ext': 'mp4',
             'title': 'Mishal K. on LinkedIn: #sendinblueviews #toronto #digitalmarketing #nowhiring #sendinblue…',
             'description': 'md5:2998a31f6f479376dd62831f53a80f71',
             'uploader': 'Mishal K.',
-            'thumbnail': 're:^https?://media.licdn.com/dms/image/.*$',
+            'thumbnail': 'https://media.licdn.com/dms/image/C4E05AQG7hCp7zIeciw/feedshare-thumbnail_720_1280/0/1633381554858?e=2147483647&v=beta&t=teGgAD-Z8A2K4UrwAtEgHz9xGMrRLmUK6cE3vm3csK0',
             'like_count': int
         },
     }, {
         'url': 'https://www.linkedin.com/posts/the-mathworks_2_what-is-mathworks-cloud-center-activity-7151241570371948544-4Gu7',
         'info_dict': {
-            'id': '7151241570371948544',
+            'id': 'D5610AQFKo9M0zqY2_g',
+            'display_id': 'the-mathworks_2_what-is-mathworks-cloud-center-activity-7151241570371948544-4Gu7',
             'ext': 'mp4',
             'title': 'MathWorks on LinkedIn: What Is MathWorks Cloud Center?',
             'description': 'md5:95f9d4eeb6337882fb47eefe13d7a40c',
             'uploader': 'MathWorks',
-            'thumbnail': 're:^https?://media.licdn.com/dms/image/.*$',
+            'thumbnail': 'https://media.licdn.com/dms/image/D5610AQFKo9M0zqY2_g/ads-video-thumbnail_720_1280/0/1704988806715?e=2147483647&v=beta&t=0vg-ksOY2KrL_QFlNacCv5Tmuk0tum9FJ_4dlJ56Gyw',
             'like_count': int,
             'subtitles': 'mincount:1'
         },
@@ -112,8 +114,17 @@ class LinkedInIE(LinkedInBaseIE):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
+        title = self._og_search_title(webpage, default=None) or self._html_extract_title(webpage)
+        description = self._og_search_description(webpage, default=None)
+        json_ld_list = list(self._yield_json_ld(webpage, video_id))
+        uploader = traverse_obj(
+            json_ld_list, (lambda _, v: v['@type'] == 'SocialMediaPosting', 'author'), get_all=False).get('name')
+        like_count = int_or_none(self._search_regex(r'<a href=[^>]+\sdata-num-reactions="(\d+)"', webpage, 'reactions', default=None))
+
         video_attrs = extract_attributes(self._search_regex(r'(<video[^>]+>)', webpage, 'video'))
+        print('video_attrs:', video_attrs)
         sources = self._parse_json(video_attrs['data-sources'], video_id)
+        id = str(video_attrs.get('data-digitalmedia-asset-urn') or video_id).replace('urn:li:digitalmediaAsset:', '')
         formats = [{
             'url': source['src'],
             'ext': mimetype2ext(source.get('type')),
@@ -125,16 +136,14 @@ class LinkedInIE(LinkedInBaseIE):
         }]} if url_or_none(video_attrs.get('data-captions-url')) else {}
 
         return {
-            'id': video_id,
+            'id': id,
+            'display_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)),
+            'uploader': uploader,
-            '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,
         }