Fix GH-17802: \Dom\HTMLDocument querySelector attribute name is case sensitive in HTML

nielsdos · nielsdos · commit 3a4a32038681 · 2025-02-17T19:31:23.000+01:00
According to https://html.spec.whatwg.org/#case-sensitivity-of-selectors, the CSS selector attribute name must be converted to lowercase in HTML elements, and then compared case-sensitive to the attribute name in the element. We implement this not by doing the explicit conversion, but by a manual loop using a function that first converts the rhs characters to lowercase and keeps the lhs characters the same, achieving the same effect.
diff --git a/ext/dom/lexbor/lexbor/selectors-adapted/selectors.c b/ext/dom/lexbor/lexbor/selectors-adapted/selectors.c
@@ -65,7 +65,21 @@ static zend_always_inline bool lxb_selectors_adapted_cmp_local_name_id(const xml
 
 static zend_always_inline const xmlAttr *lxb_selectors_adapted_attr(const xmlNode *node, const lxb_char_t *name)
 {
-	const xmlAttr *attr = xmlHasProp(node, (const xmlChar *) name);
+	const xmlAttr *attr = NULL;
+	ZEND_ASSERT(node->doc != NULL);
+	if (php_dom_ns_is_html_and_document_is_html(node)) {
+		/* No need to handle DTD entities as we're in HTML. */
+		size_t name_bound = strlen((const char *) name) + 1;
+		for (const xmlAttr *cur = node->properties; cur != NULL; cur = cur->next) {
+			if (lexbor_str_data_nlocmp_right(cur->name, name, name_bound)) {
+				attr = cur;
+				break;
+			}
+		}
+	} else {
+		attr = xmlHasProp(node, (const xmlChar *) name);
+	}
+
 	if (attr != NULL && attr->ns != NULL) {
 		return NULL;
 	}
diff --git a/ext/dom/tests/modern/css_selectors/gh17802.phpt b/ext/dom/tests/modern/css_selectors/gh17802.phpt
@@ -0,0 +1,47 @@
+--TEST--
+GH-17802 (\Dom\HTMLDocument querySelector attribute name is case sensitive in HTML)
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$text = <<<TEXT
+<html>
+<head>
+<meta charset="Windows-1252">
+</head>
+<body>
+</body>
+</html>
+TEXT;
+
+$dom = \Dom\HTMLDocument::createFromString($text, options: LIBXML_NOERROR);
+$meta2 = $dom->head->appendChild($dom->createElementNS('urn:x', 'meta'));
+$meta2->setAttribute('charset', 'x');
+echo $dom->saveHtml(), "\n";
+
+echo "--- charseT ---\n";
+
+foreach ($dom->querySelectorAll('meta[charseT]') as $entry) {
+    var_dump($dom->saveHtml($entry));
+}
+
+echo "--- charset ---\n";
+
+foreach ($dom->querySelectorAll('meta[charset]') as $entry) {
+    var_dump($dom->saveHtml($entry));
+}
+
+?>
+--EXPECT--
+<html><head>
+<meta charset="Windows-1252">
+<meta charset="x"></meta></head>
+<body>
+
+</body></html>
+--- charseT ---
+string(29) "<meta charset="Windows-1252">"
+--- charset ---
+string(29) "<meta charset="Windows-1252">"
+string(25) "<meta charset="x"></meta>"
