Fix GH-21357: XSLTProcessor works with DOMDocument, but fails with Dom\XMLDocument

ndossche · ndossche · commit 284fd7779d92 · 2026-03-11T22:31:03.000+01:00
Registering namespace after the parsing is too late because parsing can fail due to attributes referencing namespaces. So we have to register fake namespaces before the parsing. However, the clone operation reconciles namespaces in the wrong way, so we have to clone via an object. Closes GH-21371.
diff --git a/NEWS b/NEWS
@@ -30,6 +30,10 @@ PHP                                                                        NEWS
 - Sysvshm:
   . Fix memory leak in shm_get_var() when variable is corrupted. (ndossche)
 
+- XSL:
+  . Fix GH-21357 (XSLTProcessor works with DOMDocument, but fails with
+    Dom\XMLDocument). (ndossche)
+
 12 Mar 2026, PHP 8.4.19
 
 - Core:
diff --git a/ext/xsl/tests/auto_registration_namespaces_new_dom.phpt b/ext/xsl/tests/auto_registration_namespaces_new_dom.phpt
@@ -24,7 +24,7 @@ $sheet = Dom\XMLDocument::createFromString(<<<XML
 XML);
 
 // Make sure it will auto-register urn:test
-$sheet->documentElement->append($sheet->createElementNS('urn:test', 'test:dummy'));
+$sheet->documentElement->setAttributeNS('http://www.w3.org/2000/xmlns/', 'xmlns:test', 'urn:test');
 
 $input = Dom\XMLDocument::createFromString(<<<XML
 <root>
diff --git a/ext/xsl/tests/gh21357_1.phpt b/ext/xsl/tests/gh21357_1.phpt
@@ -0,0 +1,20 @@
+--TEST--
+GH-21357 (XSLTProcessor works with \DOMDocument, but fails with \Dom\XMLDocument)
+--EXTENSIONS--
+dom
+xsl
+--CREDITS--
+jacekkow
+--FILE--
+<?php
+$xml = <<<'XML'
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" xmlns:tns="urn:myns" version="1.0">
+	<xsl:template match="tns:referee"/>
+</xsl:stylesheet>
+XML;
+
+$dom = Dom\XMLDocument::createFromString($xml);
+var_dump(new XSLTProcessor()->importStylesheet($dom));
+?>
+--EXPECT--
+bool(true)
diff --git a/ext/xsl/tests/gh21357_2.phpt b/ext/xsl/tests/gh21357_2.phpt
@@ -0,0 +1,35 @@
+--TEST--
+GH-21357 (XSLTProcessor works with \DOMDocument, but fails with \Dom\XMLDocument)
+--EXTENSIONS--
+dom
+xsl
+--CREDITS--
+jacekkow
+--FILE--
+<?php
+$xsl = '<?xml version="1.0" encoding="utf-8"?>
+<xsl:stylesheet
+    version="1.0"
+    xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+    xmlns:old="http://something/old/"
+    xmlns:new="http://something/new/">
+  <xsl:template match="node()|@*">
+    <xsl:copy>
+      <xsl:apply-templates select="node()|@*"/>
+    </xsl:copy>
+  </xsl:template>
+  <xsl:template match="old:*">
+    <xsl:element name="{local-name()}" xmlns="http://something/new/" >
+      <xsl:apply-templates select="node()|@*"/>
+    </xsl:element>
+  </xsl:template>
+</xsl:stylesheet>';
+$dom = Dom\XMLDocument::createFromString($xsl);
+$xsl = new XSLTProcessor();
+$xsl->importStylesheet($dom);
+var_dump($xsl->transformToXml(\Dom\XMLDocument::createFromString('<test xmlns="http://something/old/"/>')));
+?>
+--EXPECT--
+string(138) "<?xml version="1.0"?>
+<test xmlns="http://something/new/" xmlns:ns_1="http://www.w3.org/2000/xmlns/" ns_1:xmlns="http://something/old/"/>
+"
diff --git a/ext/xsl/xsltprocessor.c b/ext/xsl/xsltprocessor.c
@@ -123,65 +123,38 @@ static void xsl_ext_function_trampoline(xmlXPathParserContextPtr ctxt, int nargs
 	}
 }
 
-static void xsl_add_ns_to_map(xmlHashTablePtr table, xsltStylesheetPtr sheet, const xmlNode *cur, const xmlChar *prefix, const xmlChar *uri)
+static void xsl_add_ns_def(xmlNodePtr node)
 {
-	const xmlChar *existing_url = xmlHashLookup(table, prefix);
-	if (existing_url != NULL && !xmlStrEqual(existing_url, uri)) {
-		xsltTransformError(NULL, sheet, (xmlNodePtr) cur, "Namespaces prefix %s used for multiple namespaces\n", prefix);
-		sheet->warnings++;
-	} else if (existing_url == NULL) {
-		xmlHashUpdateEntry(table, prefix, (void *) uri, NULL);
-	}
-}
-
-/* Adds all namespace declaration (not using nsDef) into a hash map that maps prefix to uri. Warns on conflicting declarations. */
-static void xsl_build_ns_map(xmlHashTablePtr table, xsltStylesheetPtr sheet, php_dom_libxml_ns_mapper *ns_mapper, const xmlDoc *doc)
-{
-	const xmlNode *cur = xmlDocGetRootElement(doc);
+	if (node->type == XML_ELEMENT_NODE) {
+		for (xmlAttrPtr attr = node->properties; attr; attr = attr->next) {
+			if (php_dom_ns_is_fast((const xmlNode *) attr, php_dom_ns_is_xmlns_magic_token)) {
+				xmlNsPtr ns = xmlMalloc(sizeof(*ns));
+				if (!ns) {
+					return;
+				}
 
-	while (cur != NULL) {
-		if (cur->type == XML_ELEMENT_NODE) {
-			if (cur->ns != NULL && cur->ns->prefix != NULL) {
-				xsl_add_ns_to_map(table, sheet, cur, cur->ns->prefix, cur->ns->href);
-			}
+				bool should_free;
+				xmlChar *attr_value = php_libxml_attr_value(attr, &should_free);
 
-			for (const xmlAttr *attr = cur->properties; attr != NULL; attr = attr->next) {
-				if (attr->ns != NULL && attr->ns->prefix != NULL && php_dom_ns_is_fast_ex(attr->ns, php_dom_ns_is_xmlns_magic_token)
-					&& attr->children != NULL && attr->children->content != NULL) {
-					/* This attribute declares a namespace, get the relevant instance.
-					 * The declared namespace is not the same as the namespace of this attribute (which is xmlns). */
-					const xmlChar *prefix = attr->name;
-					xmlNsPtr ns = php_dom_libxml_ns_mapper_get_ns_raw_strings_nullsafe(ns_mapper, (const char *) prefix, (const char *) attr->children->content);
-					xsl_add_ns_to_map(table, sheet, cur, prefix, ns->href);
-				}
+				memset(ns, 0, sizeof(*ns));
+				ns->type = XML_LOCAL_NAMESPACE;
+				ns->href = should_free ? attr_value : xmlStrdup(attr_value);
+				ns->prefix = attr->ns->prefix ? xmlStrdup(attr->name) : NULL;
+				ns->next = node->nsDef;
+				node->nsDef = ns;
 			}
 		}
-
-		cur = php_dom_next_in_tree_order(cur, (const xmlNode *) doc);
 	}
 }
 
-/* Apply namespace corrections for new DOM */
-typedef enum {
-	XSL_NS_HASH_CORRECTION_NONE = 0,
-	XSL_NS_HASH_CORRECTION_APPLIED = 1,
-	XSL_NS_HASH_CORRECTION_FAILED = 2
-} xsl_ns_hash_correction_status;
-
-static zend_always_inline xsl_ns_hash_correction_status xsl_apply_ns_hash_corrections(xsltStylesheetPtr sheetp, xmlNodePtr nodep, xmlDocPtr doc)
+static void xsl_add_ns_defs(xmlDocPtr doc)
 {
-	if (sheetp->nsHash == NULL) {
-		dom_object *node_intern = php_dom_object_get_data(nodep);
-		if (node_intern != NULL && php_dom_follow_spec_intern(node_intern)) {
-			sheetp->nsHash = xmlHashCreate(10);
-			if (UNEXPECTED(!sheetp->nsHash)) {
-				return XSL_NS_HASH_CORRECTION_FAILED;
-			}
-			xsl_build_ns_map(sheetp->nsHash, sheetp, php_dom_get_ns_mapper(node_intern), doc);
-			return XSL_NS_HASH_CORRECTION_APPLIED;
-		}
+	xmlNodePtr base = (xmlNodePtr) doc;
+	xmlNodePtr node = base->children;
+	while (node != NULL) {
+		xsl_add_ns_def(node);
+		node = php_dom_next_in_tree_order(node, base);
 	}
-	return XSL_NS_HASH_CORRECTION_NONE;
 }
 
 /* {{{ URL: http://www.w3.org/TR/2003/WD-DOM-Level-3-Core-20030226/DOM3-Core.html#
@@ -190,63 +163,71 @@ static zend_always_inline xsl_ns_hash_correction_status xsl_apply_ns_hash_correc
 PHP_METHOD(XSLTProcessor, importStylesheet)
 {
 	zval *id, *docp = NULL;
-	xmlDoc *doc = NULL, *newdoc = NULL;
+	xmlDoc *newdoc = NULL;
 	xsltStylesheetPtr sheetp;
 	bool clone_docu = false;
 	xmlNode *nodep = NULL;
-	zval *cloneDocu, rv;
+	zval *cloneDocu, rv, clone_zv;
 	zend_string *member;
 
 	id = ZEND_THIS;
 	if (zend_parse_parameters(ZEND_NUM_ARGS(), "o", &docp) == FAILURE) {
 		RETURN_THROWS();
 	}
 
-	nodep = php_libxml_import_node(docp);
+	/* libxslt uses _private, so we must copy the imported
+	 * stylesheet document otherwise the node proxies will be a mess.
+	 * We will clone the object and detach the libxml internals later. */
+	zend_object *clone = Z_OBJ_HANDLER_P(docp, clone_obj)(Z_OBJ_P(docp));
+	if (!clone) {
+		RETURN_THROWS();
+	}
+
+	ZVAL_OBJ(&clone_zv, clone);
+	nodep = php_libxml_import_node(&clone_zv);
 
 	if (nodep) {
-		doc = nodep->doc;
+		newdoc = nodep->doc;
 	}
-	if (doc == NULL) {
+	if (newdoc == NULL) {
+		OBJ_RELEASE(clone);
 		zend_argument_type_error(1, "must be a valid XML node");
 		RETURN_THROWS();
 	}
 
-	/* libxslt uses _private, so we must copy the imported
-	stylesheet document otherwise the node proxies will be a mess */
-	newdoc = xmlCopyDoc(doc, 1);
-	xmlNodeSetBase((xmlNodePtr) newdoc, (xmlChar *)doc->URL);
+	php_libxml_node_object *clone_lxml_obj = Z_LIBXML_NODE_P(&clone_zv);
+
 	PHP_LIBXML_SANITIZE_GLOBALS(parse);
 	ZEND_DIAGNOSTIC_IGNORED_START("-Wdeprecated-declarations")
 	xmlSubstituteEntitiesDefault(1);
 	xmlLoadExtDtdDefaultValue = XML_DETECT_IDS | XML_COMPLETE_ATTRS;
 	ZEND_DIAGNOSTIC_IGNORED_END
 
+	if (clone_lxml_obj->document->class_type == PHP_LIBXML_CLASS_MODERN) {
+		xsl_add_ns_defs(newdoc);
+	}
+
 	sheetp = xsltParseStylesheetDoc(newdoc);
 	PHP_LIBXML_RESTORE_GLOBALS(parse);
 
 	if (!sheetp) {
-		xmlFreeDoc(newdoc);
+		OBJ_RELEASE(clone);
 		RETURN_FALSE;
 	}
 
 	xsl_object *intern = Z_XSL_P(id);
 
-	xsl_ns_hash_correction_status status = xsl_apply_ns_hash_corrections(sheetp, nodep, doc);
-	if (UNEXPECTED(status == XSL_NS_HASH_CORRECTION_FAILED)) {
-		xsltFreeStylesheet(sheetp);
-		xmlFreeDoc(newdoc);
-		RETURN_FALSE;
-	} else if (status == XSL_NS_HASH_CORRECTION_APPLIED) {
-		/* The namespace mappings need to be kept alive.
-		 * This is stored in the ref obj outside of libxml2, but that means that the sheet won't keep it alive
-		 * unlike with namespaces from old DOM. */
-		if (intern->sheet_ref_obj) {
-			php_libxml_decrement_doc_ref_directly(intern->sheet_ref_obj);
-		}
-		intern->sheet_ref_obj = Z_LIBXML_NODE_P(docp)->document;
-		intern->sheet_ref_obj->refcount++;
-	}
+	/* Detach object */
+	clone_lxml_obj->document->ptr = NULL;
+	/* The namespace mappings need to be kept alive.
+	 * This is stored in the ref obj outside of libxml2, but that means that the sheet won't keep it alive
+	 * unlike with namespaces from old DOM. */
+	if (intern->sheet_ref_obj) {
+		php_libxml_decrement_doc_ref_directly(intern->sheet_ref_obj);
+	}
+	intern->sheet_ref_obj = clone_lxml_obj->document;
+	intern->sheet_ref_obj->refcount++;
+	OBJ_RELEASE(clone);
 
 	member = ZSTR_INIT_LITERAL("cloneDocument", 0);
 	cloneDocu = zend_std_read_property(Z_OBJ_P(id), member, BP_VAR_R, NULL, &rv);
