fixed correct encoding when xml pi is in content, fixes #46

Rct567 · Jan 18, 2024 · b3357e7 · b3357e7
1 parent 20e2e25
commit b3357e7
Show file tree

Hide file tree

Showing 2 changed files with 33 additions and 5 deletions.
diff --git a/src/Rct567/DomQuery/DomQueryNodes.php b/src/Rct567/DomQuery/DomQueryNodes.php
@@ -353,16 +353,19 @@ public function addDomNode(\DOMNode $dom_node, $prepend=false)
      */
     public function loadContent(string $content, $encoding='UTF-8')
     {
+
         $this->preserve_no_newlines = (strpos($content, '<') !== false && strpos($content, "\n") === false);
 
+        $content_has_leading_pi = stripos($content, '<?xml') === 0;
+
         if (!\is_bool($this->xml_mode)) {
-            $this->xml_mode = (stripos($content, '<?xml') === 0);
+            $this->xml_mode = $content_has_leading_pi;
         }
 
-        $this->xml_print_pi = (stripos($content, '<?xml') === 0);
+        $this->xml_print_pi = $content_has_leading_pi;
 
         $xml_pi_node_added = false;
-        if (!$this->xml_mode && $encoding && stripos($content, '<?xml') === false) {
+        if (!$this->xml_mode && $encoding && !$content_has_leading_pi) {
             $content = '<?xml encoding="'.$encoding.'">'.$content; // add pi node to make libxml use the correct encoding
             $xml_pi_node_added = true;
         }

diff --git a/tests/Rct567/DomQuery/Tests/DomQueryTest.php b/tests/Rct567/DomQuery/Tests/DomQueryTest.php
@@ -103,6 +103,33 @@ public function testLoadingUf8AndGettingSameContent()
         $this->assertEquals('Iñtërnâtiônàlizætiøn', $dom->text());
     }
 
+    /*
+     * Test loading utf8 html with irrelevant pi in content
+     */
+    public function testLoadingUf8AndGettingSameContentWithPiInContent()
+    {
+        $html = '<div><h1>Iñtërnâtiônàlizætiøn</h1></div><a>k</a><?xml version="1.0" encoding="iso-8859-1"?>';
+        $dom = new DomQuery($html);
+
+        $this->assertEquals($html, (string) $dom); // same result
+        $this->assertEquals('<h1>Iñtërnâtiônàlizætiøn</h1>', (string) $dom->find('h1')); // same header
+    }
+
+    /*
+     * Test loading utf8 content with leading pi
+     */
+    public function testLoadingUf8AndGettingSameContentWithLeadingPi()
+    {
+        $html = '<?xml version="1.0" encoding="UTF-8"?><div><h1>Iñtërnâtiônàlizætiøn</h1></div>';
+        $dom = new DomQuery($html);
+
+        $this->assertTrue($dom->xml_mode);
+        $this->assertEquals($html, (string) $dom); // same result
+        $this->assertEquals('<h1>Iñtërnâtiônàlizætiøn</h1>', (string) $dom->find('h1')); // same header
+        $dom->xml_mode = false;
+        $this->assertEquals('<div><h1>Iñtërnâtiônàlizætiøn</h1></div>', (string) $dom); // without pi
+    }
+
     /*
      * Test loading html with new lines
      */
@@ -121,8 +148,6 @@ public function testPreserverAttributeWithoutValue()
         $this->assertEquals('<div selected>a</div>', (string) $dom);
     }
 
-
-
     /*
      * Test change attribute without value in xml write mode
      */