From 011c16868684470f589a6ba53365194955a5e72a Mon Sep 17 00:00:00 2001
From: poire-z <poire-z@users.noreply.github.com>
Date: Fri, 19 May 2023 16:48:36 +0200
Subject: [PATCH] Wikipedia EPUBs: strip out invalid-XHTML <link> (#10462)

---
 frontend/ui/wikipedia.lua | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/frontend/ui/wikipedia.lua b/frontend/ui/wikipedia.lua
index 9a74c21a2..989fee27b 100644
--- a/frontend/ui/wikipedia.lua
+++ b/frontend/ui/wikipedia.lua
@@ -1218,6 +1218,12 @@ table {
     -- OEBPS/content.html
     -- Some small fixes to Wikipedia HTML to make crengine and the user happier
 
+    -- In some articles' HTML, we may get <link rel="mw-deduplicated-inline-style" href="mw-data...">
+    -- (which, by specs, is an empty element) without the proper empty tag ending "/>", which
+    -- would cause crengine's EPUB XHTML parser to wait for a proper </link>, hiding all the
+    -- following content... So, just remove them, as we don't make any use of them.
+    html = html:gsub("<link [^>]*>", "")
+
     -- Most images are in a link to the image info page, which is a useless
     -- external link for us, so let's remove this link.
     html = html:gsub("<a[^>]*>%s*(<%s*img [^>]*>)%s*</a>", "%1")