Merge pull request #64 from goetas/i63

Case insensitive tags
Masterminds · Dec 17, 2014 · 398ebb6 · 398ebb6
2 parents de3a3af + 4df8acd
commit 398ebb6
Show file tree

Hide file tree

Showing 2 changed files with 28 additions and 4 deletions.
diff --git a/src/HTML5/Parser/Tokenizer.php b/src/HTML5/Parser/Tokenizer.php
@@ -203,7 +203,9 @@ protected function rcdata()
         $sequence = '</' . $this->untilTag . '>';
         $txt = '';
         $tok = $this->scanner->current();
-        while ($tok !== false && ! ($tok == '<' && ($this->sequenceMatches($sequence) || $this->sequenceMatches(strtoupper($sequence))))) {
+
+        $caseSensitive = !Elements::isHtml5Element($this->untilTag);
+        while ($tok !== false && ! ($tok == '<' && ($this->sequenceMatches($sequence, $caseSensitive)))) {
             if ($tok == '&') {
                 $txt .= $this->decodeCharacterReference();
                 $tok = $this->scanner->current();
@@ -891,7 +893,7 @@ protected function readUntilSequence($sequence)
             $buffer .= $this->scanner->charsUntil($first);
 
             // Stop as soon as we hit the stopping condition.
-            if ($this->sequenceMatches($sequence) || $this->sequenceMatches(strtoupper($sequence))) {
+            if ($this->sequenceMatches($sequence, false)) {
                 return $buffer;
             }
             $buffer .= $this->scanner->current();
@@ -916,7 +918,7 @@ protected function readUntilSequence($sequence)
      * see if the input stream is at the start of a
      * '</script>' string.
      */
-    protected function sequenceMatches($sequence)
+    protected function sequenceMatches($sequence, $caseSensitive = true)
     {
         $len = strlen($sequence);
         $buffer = '';
@@ -932,7 +934,7 @@ protected function sequenceMatches($sequence)
         }
 
         $this->scanner->unconsume($len);
-        return $buffer == $sequence;
+        return $caseSensitive ? $buffer == $sequence : strcasecmp($buffer, $sequence) === 0;
     }
 
     /**

diff --git a/test/HTML5/Parser/DOMTreeBuilderTest.php b/test/HTML5/Parser/DOMTreeBuilderTest.php
@@ -56,6 +56,28 @@ public function testDocument()
         $this->assertInstanceOf('\DOMDocument', $doc);
         $this->assertEquals('html', $doc->documentElement->tagName);
     }
+
+    public function testStrangeCapitalization()
+    {
+        $html = "<!doctype html>
+        <html>
+            <head>
+                <Title>Hello, world!</TitlE>
+            </head>
+            <body>TheBody<script>foo</script></body>
+        </html>";
+        $doc = $this->parse($html);
+
+        $this->assertInstanceOf('\DOMDocument', $doc);
+        $this->assertEquals('html', $doc->documentElement->tagName);
+
+        $xpath = new \DOMXPath( $doc );
+        $xpath->registerNamespace( "x", "http://www.w3.org/1999/xhtml" );
+
+        $this->assertEquals("Hello, world!", $xpath->query( "//x:title" )->item( 0 )->nodeValue);
+        $this->assertEquals("foo", $xpath->query( "//x:script" )->item( 0 )->nodeValue);
+    }
+
     public function testDocumentFakeAttrAbsence()
     {
         $html = "<!DOCTYPE html><html xmlns=\"http://www.w3.org/1999/xhtml\"><body>foo</body></html>";