Fix sanitizer function for libxml2 >= 2.14 and HTML5

renanrodrigo · renanrodrigo · commit 66416e08a231 · 2025-08-07T00:01:52.000-03:00
This is adapting the sanitize function to remove commented XML tags from the resulting markdown. Fixes: #263
diff --git a/src/HtmlConverter.php b/src/HtmlConverter.php
@@ -233,7 +233,7 @@ protected function sanitize(string $markdown): string
          * Removing unwanted tags. Tags should be added to the array in the order they are expected.
          * XML, html and body opening tags should be in that order. Same case with closing tags
          */
-        $unwanted = ['<?xml encoding="UTF-8">', '<html>', '</html>', '<body>', '</body>', '<head>', '</head>', '&#xD;'];
+        $unwanted = ['<?xml encoding="UTF-8">', '<!--?xml encoding="UTF-8"-->', '<html>', '</html>', '<body>', '<!--?xml encoding="UTF-8"-->', '</body>', '<head>', '</head>', '&#xD;'];
 
         foreach ($unwanted as $tag) {
             if (\strpos($tag, '/') === false) {
