From 7592749cbe377675688dfcad5b7c1d46bbb684e1 Mon Sep 17 00:00:00 2001
From: nixxo <nixxo@protonmail.com>
Date: Sat, 1 Jan 2022 20:50:27 +0100
Subject: [PATCH] [extractor] Extract thumbnails from JSON-LD (#2195)
Authored by: nixxo
---
test/test_InfoExtractor.py | 26 ++++++++++++++++++++++++++
yt_dlp/extractor/common.py | 3 ++-
2 files changed, 28 insertions(+), 1 deletion(-)
diff --git a/test/test_InfoExtractor.py b/test/test_InfoExtractor.py
index cf06dbde46..25bc00e0b9 100644
--- a/test/test_InfoExtractor.py
+++ b/test/test_InfoExtractor.py
@@ -208,6 +208,32 @@ class TestInfoExtractor(unittest.TestCase):
},
{'expected_type': 'NewsArticle'},
),
+ (
+ # test multiple thumbnails in a list
+ r'''
+<script type="application/ld+json">
+{"@context":"https://schema.org",
+"@type":"VideoObject",
+"thumbnailUrl":["https://www.rainews.it/cropgd/640x360/dl/img/2021/12/30/1640886376927_GettyImages.jpg"]}
+</script>''',
+ {
+ 'thumbnails': [{'url': 'https://www.rainews.it/cropgd/640x360/dl/img/2021/12/30/1640886376927_GettyImages.jpg'}],
+ },
+ {},
+ ),
+ (
+ # test single thumbnail
+ r'''
+<script type="application/ld+json">
+{"@context":"https://schema.org",
+"@type":"VideoObject",
+"thumbnailUrl":"https://www.rainews.it/cropgd/640x360/dl/img/2021/12/30/1640886376927_GettyImages.jpg"}
+</script>''',
+ {
+ 'thumbnails': [{'url': 'https://www.rainews.it/cropgd/640x360/dl/img/2021/12/30/1640886376927_GettyImages.jpg'}],
+ },
+ {},
+ )
]
for html, expected_dict, search_json_ld_kwargs in _TESTS:
expect_dict(
diff --git a/yt_dlp/extractor/common.py b/yt_dlp/extractor/common.py
index 3821b71835..1d694293e0 100644
--- a/yt_dlp/extractor/common.py
+++ b/yt_dlp/extractor/common.py
@@ -1436,7 +1436,8 @@ class InfoExtractor(object):
'url': url_or_none(e.get('contentUrl')),
'title': unescapeHTML(e.get('name')),
'description': unescapeHTML(e.get('description')),
- 'thumbnail': url_or_none(e.get('thumbnailUrl') or e.get('thumbnailURL')),
+ 'thumbnails': [{'url': url_or_none(url)}
+ for url in variadic(traverse_obj(e, 'thumbnailUrl', 'thumbnailURL'))],
'duration': parse_duration(e.get('duration')),
'timestamp': unified_timestamp(e.get('uploadDate')),
# author can be an instance of 'Organization' or 'Person' types.