From 32e386f1a55f76d827e264d9637870e9bc424dff Mon Sep 17 00:00:00 2001
From: osapon <osa@osa-p.net>
Date: Mon, 25 Nov 2024 11:58:17 +0900
Subject: [PATCH] Improved charset tag recognition accuracy.

---
 src/Document.php | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/src/Document.php b/src/Document.php
index 6db07663..65c299a1 100644
--- a/src/Document.php
+++ b/src/Document.php
@@ -28,11 +28,11 @@ public function __construct(Extractor $extractor)
 
         $encoding = null;
         $contentType = $extractor->getResponse()->getHeaderLine('content-type');
-        preg_match('/charset="?(.*?)(?=$|\s|;|")/i', $contentType, $match);
+        preg_match('/charset=(?:"|\')?(.*?)(?=$|\s|;|"|\'|>)/i', $contentType, $match);
         if (!empty($match[1])) {
             $encoding = trim($match[1], ',');
         } elseif (!empty($html)) {
-            preg_match('/charset="?(.*?)(?=$|\s|;|")/i', $html, $match);
+            preg_match('/charset=(?:"|\')?(.*?)(?=$|\s|;|"|\'|>)/i', $html, $match);
             if (!empty($match[1])) {
                 $encoding = trim($match[1], ',');
             }