From: Martin Pitt <martin@piware.de>
Date: Sat, 3 Jan 2026 16:43:31 +0000 (+0100)
Subject: recipekeeper-split: Fix unescaped URLs
X-Git-Url: https://piware.de/gitweb/?a=commitdiff_plain;h=af0dbf7671f2a183bd5895b024705850efa46bf1;p=bin.git

recipekeeper-split: Fix unescaped URLs
---

diff --git a/recipekeeper-split b/recipekeeper-split
index a56f52e..c510408 100755
--- a/recipekeeper-split
+++ b/recipekeeper-split
@@ -33,6 +33,11 @@ content_fixed = re.sub(r'(<meta\b[^>]*)(?<!/)>', r'\1 />', content)
 # Only match when the value is not already quoted
 content_fixed = re.sub(r'itemprop=([^\s">]+)', r'itemprop="\1"', content_fixed)
 
+# 3. Fix unescaped ampersands in attribute values (& -> &amp;)
+# This handles ampersands in URLs and other attribute values
+# Only escape & that are not already part of an entity
+content_fixed = re.sub(r'&(?!amp;|lt;|gt;|quot;|apos;|#)', r'&amp;', content_fixed)
+
 # Parse as XML
 tree = ET.ElementTree(ET.fromstring(content_fixed))
 root = tree.getroot()