(no commit message)

[utils] / support / src / org / wamblee / xml / DomUtils.java

diff --git a/support/src/org/wamblee/xml/DomUtils.java b/support/src/org/wamblee/xml/DomUtils.java
index 4fdb52e752b6517440ed68538c36c1a86b2a9be9..75ba7f6ddefeb1d4a20a35e9f30651d05608d5b7 100644 (file)
--- a/support/src/org/wamblee/xml/DomUtils.java
+++ b/support/src/org/wamblee/xml/DomUtils.java
@@ -20,6 +20,10 @@ import java.io.ByteArrayOutputStream;
 import java.io.IOException;
 import java.io.InputStream;
 import java.io.OutputStream;
+import java.util.ArrayList;
+import java.util.List;
+import java.util.Map;
+import java.util.TreeMap;
 
 import javax.xml.parsers.DocumentBuilder;
 import javax.xml.parsers.DocumentBuilderFactory;
@@ -30,7 +34,12 @@ import org.apache.xml.serialize.XMLSerializer;
 import org.dom4j.DocumentException;
 import org.dom4j.io.DOMReader;
 import org.dom4j.io.DOMWriter;
+import org.w3c.dom.Attr;
 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.SAXException;
 
 /**
@@ -96,4 +105,53 @@ public final class DomUtils {
     public static org.dom4j.Document convert(Document aDocument) { 
         return new DOMReader().read(aDocument); 
     }
+    
+    /**
+     * Removes duplicate attributes from a DOM tree.This is useful for postprocessing the
+     * output of JTidy as a workaround for a bug in JTidy. 
+     * 
+     * @param aNode
+     *            Node to remove duplicate attributes from (recursively).
+     *            Attributes of the node itself are not dealt with. Only the
+     *            child nodes are dealt with.
+     */
+    public static void removeDuplicateAttributes(Node aNode) {
+        NodeList list = aNode.getChildNodes();
+        for (int i = 0; i < list.getLength(); i++) {
+            Node node = list.item(i);
+            if (node instanceof Element) {
+                removeDuplicateAttributes((Element) node);
+                removeDuplicateAttributes(node);
+            }
+        }
+    }
+
+    /**
+     * Removes duplicate attributes from an element.
+     * 
+     * @param aElement
+     *            Element.
+     */
+    private static void removeDuplicateAttributes(Element aElement) {
+        NamedNodeMap attributes = aElement.getAttributes();
+        Map<String, Attr> uniqueAttributes = new TreeMap<String, Attr>();
+        List<Attr> attlist = new ArrayList<Attr>();
+        for (int i = 0; i < attributes.getLength(); i++) {
+            Attr attribute = (Attr) attributes.item(i);
+            if (uniqueAttributes.containsKey(attribute.getNodeName())) {
+                System.out.println("Detected duplicate attribute '"
+                        + attribute.getNodeName() + "'");
+            }
+            uniqueAttributes.put(attribute.getNodeName(), attribute);
+            attlist.add(attribute);
+        }
+        // Remove all attributes from the element.
+        for (Attr att : attlist) {
+            aElement.removeAttributeNode(att);
+        }
+        // Add the unique attributes back to the element.
+        for (Attr att : uniqueAttributes.values()) {
+            aElement.setAttributeNode(att);
+        }
+    }
 }