From 93f9a0dd18b767b5c1200cf7da29b0b23ff2cd09 Mon Sep 17 00:00:00 2001
From: Marko Jovanovic <marko.jovanovic@in.tum.de>
Date: Wed, 29 Oct 2025 12:22:05 +0100
Subject: [PATCH] Sanitize to UTF-8

---
 src/Command/CleanMobileCommand.php | 22 +++++++++++++++++++++-
 1 file changed, 21 insertions(+), 1 deletion(-)

diff --git a/src/Command/CleanMobileCommand.php b/src/Command/CleanMobileCommand.php
index 26dae5d..e907ebd 100644
--- a/src/Command/CleanMobileCommand.php
+++ b/src/Command/CleanMobileCommand.php
@@ -92,6 +92,26 @@ final class CleanMobileCommand extends Command
             '17',
         ];
 
+        $sanitiseUtf8 = static function(?string $raw): ?string {
+            $utf8 = null;
+            if (mb_check_encoding($raw, 'UTF-8')) {
+                $utf8 = $raw;
+            } else {
+                $encodings = ['Windows-1252', 'ISO-8859-1', 'CP1252', 'ASCII'];
+                foreach ($encodings as $src) {
+                    $utf8 = @iconv($src, 'UTF-8//TRANSLIT//IGNORE', $raw);
+                    if ($utf8 !== false) {
+                        break;
+                    }
+                }
+                if ($utf8 === false) {
+                    // Could not be converted – treat it as “invalid”.
+                    return null;
+                }
+            }
+            return $utf8;
+        };
+
         // -------------------------------------------------------------
         // 4️⃣  Helper closures
         // -------------------------------------------------------------
@@ -206,7 +226,7 @@ final class CleanMobileCommand extends Command
                 $contact->setParsedAt(new \DateTimeImmutable());
                 $contact->setStudyId($study_id);
                 $contact->setParsedFileLinenum($rowCount + 1);
-                $contact->setParsedFileLine(implode(';', $row));
+                $contact->setParsedFileLine($sanitiseUtf8(implode(';', $row)));
                 $contact->setMsgContentType($rowCount % 2 ? 1 : 2);
 
                 try {