Question about removing a node.. getting 'not found in this context' error

Question

hi, i have the following code:
     public boolean deleteRunType(String runTypeName){
          //in this function we
          NodeList nl = Doc.getElementsByTagName("Run-Type");
          int numberOfNodes = nl.getLength();
          String attribute;
          NamedNodeMap attributes;
          for(int i=0;i<numberOfNodes;i++){
               attributes = nl.item(i).getAttributes();
               if(attributes!=null){
                    Node attributeNode = attributes.getNamedItem("name");
                    if(attributeNode!=null){
                         attribute = attributeNode.getNodeValue();
                         System.out.println("Comparing:  attribute = " + attribute + "   to runTypeName = " + runTypeName);
                         if(attribute.compareTo(runTypeName) == 0){
                              System.out.println("Match!");
                              Doc.removeChild(nl.item(i));
                    }//end if
               }//end if
          }//end for
          return true;
//definition of Doc:
                    DocumentBuilder DB  = DocumentBuilderFactory.newInstance().newDocumentBuilder();
                    Doc           = DB.parse(xmlFile);anyways, i am getting an error:
org.apache.crimson.tree.DomEx: NOT_FOUND_ERR: That node does not exist in this context.
     at org.apache.crimson.tree.ParentNode.removeChild(Unknown Source)
     at RunType.RunTypeXMLWriter.deleteRunType(RunTypeXMLWriter.java:58)
     at RunType.RunTypeXMLWriter.main(RunTypeXMLWriter.java:81)when i run it. it crashes on the Doc.removeChild(nl.item(0)); line.
why is this? I grabbed the child directly from the list, so it should exist in this
context. Maybe i don't understand the context.
Any ideas or suggestions? Thank you SO much for all the help. i am finally
starting to get the hang of XML, but i still have some hangups!

Answer 1

This is not an uncommon problem.
When you do getElementsByTagName() on a document, you get all elements by that name that anywhere in the tree of the document.
When you do a removeChild, you need to be pointing to the parent of the element you are removing.
If you change the removeChild line to :
Node parent =  nl.item(i).getParentNode();
parent.removeChild( nl.item(i) );Also, it is generally not done to use a capital letter to start the name of an instance, such as "Doc". Initial capital letters are usually reserved for class names.
Dave Patterson