From 0e1f22fa9af690c6e1f469d32752e6bf2fc4f70d Mon Sep 17 00:00:00 2001
From: vincent-peugnet <vincent-peugnet@riseup.net>
Date: Wed, 14 Dec 2022 18:34:33 +0100
Subject: [PATCH] fix utf8 decoding bug

---
 app/class/Servicerender.php | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/app/class/Servicerender.php b/app/class/Servicerender.php
index 270d83ce..57fb5083 100644
--- a/app/class/Servicerender.php
+++ b/app/class/Servicerender.php
@@ -446,9 +446,12 @@ private function autourl($text): string
      */
     private function htmlink(string $text): string
     {
-        $dom = new DOMDocument();
+        $dom = new DOMDocument('1.0', 'UTF-8');
+        /** Force UTF-8 encoding for loadHTML by defining it in the content itself with an XML tag that need to be removed later */
+        $text = '<?xml encoding="utf-8" ?>' . $text;
         /** @phpstan-ignore-next-line Error supposed to be thrown here but is'nt */
-        $dom->loadHTML($text, LIBXML_NOERROR + LIBXML_HTML_NODEFDTD + LIBXML_HTML_NOIMPLIED);
+        $dom->loadHTML($text, LIBXML_NOERROR | LIBXML_HTML_NODEFDTD | LIBXML_HTML_NOIMPLIED);
+        $dom->removeChild($dom->firstChild);
         $links = $dom->getElementsByTagName('a');
         foreach ($links as $link) {
             assert($link instanceof DOMElement);