add UniversalNamespaceCache

CHANGELOG.md

@@ -10,6 +10,7 @@
 * 【poi    】     调整别名策略，clearHeaderAlias和addHeaderAlias同时清除aliasComparator（issue#828@Github）
 * 【core   】     修改StrUtil.equals逻辑，改为contentEquals
 * 【core   】     增加URLUtil.UrlDecoder
+* 【core   】     增加XmlUtil.setNamespaceAware，getByPath支持UniversalNamespaceCache
 
 ### Bug修复
 * 【json   】     修复解析JSON字符串时配置无法传递问题

hutool-core/src/main/java/cn/hutool/core/util/XmlUtil.java

@@ -6,13 +6,17 @@ import cn.hutool.core.exceptions.UtilException;
 import cn.hutool.core.io.FileUtil;
 import cn.hutool.core.io.IoUtil;
 import cn.hutool.core.lang.Assert;
+import cn.hutool.core.map.BiMap;
 import cn.hutool.core.map.MapUtil;
 import org.w3c.dom.Document;
 import org.w3c.dom.Element;
+import org.w3c.dom.NamedNodeMap;
 import org.w3c.dom.Node;
 import org.w3c.dom.NodeList;
 import org.xml.sax.InputSource;
 
+import javax.xml.XMLConstants;
+import javax.xml.namespace.NamespaceContext;
 import javax.xml.namespace.QName;
 import javax.xml.parsers.DocumentBuilder;
 import javax.xml.parsers.DocumentBuilderFactory;
@@ -68,6 +72,11 @@ public class XmlUtil {
 	 */
 	private static String defaultDocumentBuilderFactory = "com.sun.org.apache.xerces.internal.jaxp.DocumentBuilderFactoryImpl";
 
+	/**
+	 * 是否打开命名空间支持
+	 */
+	private static boolean namespaceAware = true;
+
 	/**
 	 * 禁用默认的DocumentBuilderFactory，禁用后如果有第三方的实现（如oracle的xdb包中的xmlparse），将会自动加载实现。
 	 */
@@ -75,6 +84,16 @@ public class XmlUtil {
 		defaultDocumentBuilderFactory = null;
 	}
 
+	/**
+	 * 设置是否打开命名空间支持，默认打开
+	 *
+	 * @param isNamespaceAware 是否命名空间支持
+	 * @since 5.3.1
+	 */
+	synchronized public static void setNamespaceAware(boolean isNamespaceAware) {
+		namespaceAware = isNamespaceAware;
+	}
+
 	// -------------------------------------------------------------------------------------- Read
 
 	/**
@@ -176,7 +195,7 @@ public class XmlUtil {
 			throw new IllegalArgumentException("XML content string is empty !");
 		}
 		xmlStr = cleanInvalid(xmlStr);
-		return readXML(new InputSource(StrUtil.getReader(xmlStr)));
+		return readXML(StrUtil.getReader(xmlStr));
 	}
 
 	/**
@@ -487,7 +506,7 @@ public class XmlUtil {
 			factory = DocumentBuilderFactory.newInstance();
 		}
 		// 默认打开NamespaceAware，getElementsByTagNameNS可以使用命名空间
-		factory.setNamespaceAware(true);
+		factory.setNamespaceAware(namespaceAware);
 		return disableXXE(factory);
 	}
 
@@ -533,6 +552,7 @@ public class XmlUtil {
 
 	/**
 	 * 获取节点所在的Document
+	 *
 	 * @param node 节点
 	 * @return {@link Document}
 	 * @since 5.3.0
@@ -728,7 +748,31 @@ public class XmlUtil {
 	 * @since 3.2.0
 	 */
 	public static Object getByXPath(String expression, Object source, QName returnType) {
+		NamespaceContext nsContext = null;
+		if (source instanceof Node) {
+			nsContext = new UniversalNamespaceCache((Node) source, false);
+		}
+		return getByXPath(expression, source, returnType, nsContext);
+	}
+
+	/**
+	 * 通过XPath方式读取XML节点等信息<br>
+	 * Xpath相关文章：<br>
+	 * https://www.ibm.com/developerworks/cn/xml/x-javaxpathapi.html<br>
+	 * https://www.ibm.com/developerworks/cn/xml/x-nmspccontext/
+	 *
+	 * @param expression XPath表达式
+	 * @param source     资源，可以是Docunent、Node节点等
+	 * @param returnType 返回类型，{@link javax.xml.xpath.XPathConstants}
+	 * @param nsContext  {@link NamespaceContext}
+	 * @return 匹配返回类型的值
+	 * @since 5.3.1
+	 */
+	public static Object getByXPath(String expression, Object source, QName returnType, NamespaceContext nsContext) {
 		final XPath xPath = createXPath();
+		if (null != nsContext) {
+			xPath.setNamespaceContext(nsContext);
+		}
 		try {
 			if (source instanceof InputSource) {
 				return xPath.evaluate(expression, (InputSource) source, returnType);
@@ -1182,6 +1226,103 @@ public class XmlUtil {
 		}
 		return dbf;
 	}
+
+	/**
+	 * 全局命名空间上下文<br>
+	 * 见：https://www.ibm.com/developerworks/cn/xml/x-nmspccontext/
+	 */
+	public static class UniversalNamespaceCache implements NamespaceContext {
+		private static final String DEFAULT_NS = "DEFAULT";
+		private final BiMap<String, String> prefixUri = new BiMap<>(new HashMap<>());
+
+		/**
+		 * This constructor parses the document and stores all namespaces it can
+		 * find. If toplevelOnly is true, only namespaces in the root are used.
+		 *
+		 * @param node         source Node
+		 * @param toplevelOnly restriction of the search to enhance performance
+		 */
+		public UniversalNamespaceCache(Node node, boolean toplevelOnly) {
+			examineNode(node.getFirstChild(), toplevelOnly);
+		}
+
+		/**
+		 * A single node is read, the namespace attributes are extracted and stored.
+		 *
+		 * @param node            to examine
+		 * @param attributesOnly, if true no recursion happens
+		 */
+		private void examineNode(Node node, boolean attributesOnly) {
+			NamedNodeMap attributes = node.getAttributes();
+			for (int i = 0; i < attributes.getLength(); i++) {
+				Node attribute = attributes.item(i);
+				storeAttribute(attribute);
+			}
+
+			if (false == attributesOnly) {
+				NodeList childNodes = node.getChildNodes();
+				for (int i = 0; i < childNodes.getLength(); i++) {
+					Node item = childNodes.item(i);
+					if (item.getNodeType() == Node.ELEMENT_NODE)
+						examineNode(item, false);
+				}
+			}
+		}
+
+		/**
+		 * This method looks at an attribute and stores it, if it is a namespace
+		 * attribute.
+		 *
+		 * @param attribute to examine
+		 */
+		private void storeAttribute(Node attribute) {
+			// examine the attributes in namespace xmlns
+			if (attribute.getNamespaceURI() != null
+					&& attribute.getNamespaceURI().equals(
+					XMLConstants.XMLNS_ATTRIBUTE_NS_URI)) {
+				// Default namespace xmlns="uri goes here"
+				if (attribute.getNodeName().equals(XMLConstants.XMLNS_ATTRIBUTE)) {
+					prefixUri.put(DEFAULT_NS, attribute.getNodeValue());
+				} else {
+					// The defined prefixes are stored here
+					prefixUri.put(attribute.getLocalName(), attribute.getNodeValue());
+				}
+			}
+
+		}
+
+		/**
+		 * This method is called by XPath. It returns the default namespace, if the
+		 * prefix is null or "".
+		 *
+		 * @param prefix to search for
+		 * @return uri
+		 */
+		@Override
+		public String getNamespaceURI(String prefix) {
+			if (prefix == null || prefix.equals(XMLConstants.DEFAULT_NS_PREFIX)) {
+				return prefixUri.get(DEFAULT_NS);
+			} else {
+				return prefixUri.get(prefix);
+			}
+		}
+
+		/**
+		 * This method is not needed in this context, but can be implemented in a
+		 * similar way.
+		 */
+		@Override
+		public String getPrefix(String namespaceURI) {
+			return prefixUri.getInverse().get(namespaceURI);
+		}
+
+		@Override
+		public Iterator<?> getPrefixes(String namespaceURI) {
+			// Not implemented
+			return null;
+		}
+
+	}
 	// ---------------------------------------------------------------------------------------- Private method end
 
 }

hutool-core/src/test/java/cn/hutool/core/util/XmlUtilTest.java

@@ -147,4 +147,22 @@ public class XmlUtilTest {
 		String xml = XmlUtil.mapToXmlStr(map, true);
 		Assert.assertEquals("<xml><name>ddatsh</name></xml>", xml);
 	}
+
+	@Test
+	public void getByPathTest(){
+		String xmlStr = "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n" +
+				"<soap:Envelope xmlns:soap=\"http://schemas.xmlsoap.org/soap/envelope/\">\n" +
+				"  <soap:Body>\n" +
+				"    <ns2:testResponse xmlns:ns2=\"http://ws.xxx.com/\">\n" +
+				"      <return>2020/04/15 21:01:21</return>\n" +
+				"    </ns2:testResponse>\n" +
+				"  </soap:Body>\n" +
+				"</soap:Envelope>\n";
+
+		Document document = XmlUtil.readXML(xmlStr);
+		Object value = XmlUtil.getByXPath(
+				"//soap:Envelope/soap:Body/ns2:testResponse/return",
+				document,XPathConstants.STRING);//
+		Assert.assertEquals("2020/04/15 21:01:21", value);
+	}
 }