fix: No longer serialize any namespaces with an empty URI (#244)

xmldom · Jun 9, 2021 · be3b3e6 · be3b3e6
1 parent 23570ce
commit be3b3e6
Show file tree

Hide file tree

Showing 3 changed files with 39 additions and 26 deletions.
diff --git a/lib/dom.js b/lib/dom.js
@@ -962,10 +962,17 @@ function nodeSerializeToString(isHtml,nodeFilter){
 	return buf.join('');
 }
 
-function needNamespaceDefine(node,isHTML, visibleNamespaces) {
-	var prefix = node.prefix||'';
+function needNamespaceDefine(node, isHTML, visibleNamespaces) {
+	var prefix = node.prefix || '';
 	var uri = node.namespaceURI;
-	if (!prefix && !uri){
+	// According to [Namespaces in XML 1.0](https://www.w3.org/TR/REC-xml-names/#ns-using) ,
+	// and more specifically https://www.w3.org/TR/REC-xml-names/#nsc-NoPrefixUndecl :
+	// > In a namespace declaration for a prefix [...], the attribute value MUST NOT be empty.
+	// in a similar manner [Namespaces in XML 1.1](https://www.w3.org/TR/xml-names11/#ns-using)
+	// and more specifically https://www.w3.org/TR/xml-names11/#nsc-NSDeclared :
+	// > [...] Furthermore, the attribute value [...] must not be an empty string.
+	// so serializing empty namespace value like xmlns:ds="" would produce an invalid XML document.
+	if (!uri) {
 		return false;
 	}
 	if (prefix === "xml" && uri === NAMESPACE.XML || uri === NAMESPACE.XMLNS) {
@@ -976,8 +983,8 @@ function needNamespaceDefine(node,isHTML, visibleNamespaces) {
 	while (i--) {
 		var ns = visibleNamespaces[i];
 		// get namespace prefix
-		if (ns.prefix == prefix){
-			return ns.namespace != uri;
+		if (ns.prefix === prefix) {
+			return ns.namespace !== uri;
 		}
 	}
 	return true;
@@ -1035,13 +1042,9 @@ function serializeToString(node,buf,isHTML,nodeFilter,visibleNamespaces){
 		if (needNamespaceDefine(node,isHTML, visibleNamespaces)) {
 			var prefix = node.prefix||'';
 			var uri = node.namespaceURI;
-			if (uri) {
-				// Avoid empty namespace value like xmlns:ds=""
-				// Empty namespace URL will we produce an invalid XML document
-				var ns = prefix ? ' xmlns:' + prefix : " xmlns";
-				buf.push(ns, '="' , uri , '"');
-				visibleNamespaces.push({ prefix: prefix, namespace:uri });
-			}
+			var ns = prefix ? ' xmlns:' + prefix : " xmlns";
+			buf.push(ns, '="' , uri , '"');
+			visibleNamespaces.push({ prefix: prefix, namespace:uri });
 		}
 
 		if(child || isHTML && !/^(?:meta|link|img|br|hr|input)$/i.test(nodeName)){

diff --git a/test/dom/serializer.test.js b/test/dom/serializer.test.js
@@ -17,4 +17,28 @@ describe('XML Serializer', () => {
 			'<script xmlns="http://www.w3.org/1999/xhtml"></script>'
 		)
 	})
+
+	describe('does not serialize namespaces with an empty URI', () => {
+		// for more details see the comments in lib/dom.js:needNamespaceDefine
+		it('that are used in a node', () => {
+			const source = '<w:p><w:r>test1</w:r><w:r>test2</w:r></w:p>'
+			const { documentElement } = new DOMParser().parseFromString(source)
+
+			expect(documentElement.firstChild.firstChild).toMatchObject({
+				nodeValue: 'test1',
+			})
+			expect(documentElement.lastChild.firstChild).toMatchObject({
+				nodeValue: 'test2',
+			})
+
+			expect(documentElement.toString()).toStrictEqual(source)
+		})
+
+		it('that are used in an attribute', () => {
+			const source = '<w:p w:attr="val"/>'
+			const { documentElement } = new DOMParser().parseFromString(source)
+
+			expect(documentElement.toString()).toStrictEqual(source)
+		})
+	})
 })
diff --git a/test/parse/node.test.js b/test/parse/node.test.js
@@ -87,20 +87,6 @@ describe('XML Node Parse', () => {
 		})
 	})
 
-	it('prefixed without empty namespace', () => {
-		const source = '<w:p><w:r>test1</w:r><w:r>test2</w:r></w:p>'
-		const { documentElement } = new DOMParser().parseFromString(source)
-
-		expect(documentElement.firstChild.firstChild).toMatchObject({
-			nodeValue: 'test1',
-		})
-		expect(documentElement.lastChild.firstChild).toMatchObject({
-			nodeValue: 'test2',
-		})
-
-		expect(documentElement.toString()).toStrictEqual(source)
-	})
-
 	it('cdata comment', () => {
 		const { documentElement } = new DOMParser().parseFromString(
 			'<xml>start <![CDATA[<encoded>]]> <!-- comment -->end</xml>'