How do I remove elements and their children from Xml using Java and Stax

Question

I have a large file containing data like this

<releases>
   <release>
   ......
      <companies>
         <company>
         </company>
      </companies>
   </release>
   <release>
   ......
</releases>

I want to produce and output file the same as the input file but with all the companies elements and their children removed.I'm trying to use Stax, I thought that it iterated down through the elements and then if I got a match I could just not write that element and the section would be removed. ie if I throw away element that causes elements within it and to be removed but it seems that it just removes the element itself and nothing else, is that right ?

ie

<releases>
   <release>
   ......
   </release>
   <release>
   ......
</releases>

This is the code I currently have:

        XMLInputFactory  inputFactory = XMLInputFactory.newInstance();
        XMLOutputFactory outputFactory = XMLOutputFactory.newInstance();
        InputStream in = new FileInputStream(source);
        XMLEventReader reader = inputFactory.createXMLEventReader(in);

        OutputStream out = new FileOutputStream(target);
        XMLEventWriter writer =  outputFactory.createXMLEventWriter(out);
        XMLEvent event;
        while(reader.hasNext()){
            event = reader.nextEvent();
            writer.add(event);
            if(event.getEventType() == XMLStreamConstants.START_ELEMENT)
            {
                if(event.asStartElement().getName().toString().equalsIgnoreCase("companies"))
                {
                    System.out.println("Deleting:"+event);
                    continue;
                }
                else
                {
                    writer.add(event);
                }
            }
            else
            {
                writer.add(event);
            } 
        }

Answer 1

I think I have it now, Stax does process element by element so I needed to mark when the starting element is found, set a variable to indicate all subsequent data should be ignored until get to the ending element.

Code Example:

import javax.xml.stream.*;
import javax.xml.stream.events.XMLEvent;
import java.io.*;

public class FixDb
{

    public static void main(String[] args) throws Exception
    {
        File source = new File(args[0]);
        if(!source.exists())
        {
            System.out.println("File:"+source+ " does not exist");
        }

        File target = new File(source+".new");

        XMLInputFactory  inputFactory = XMLInputFactory.newInstance();
        XMLOutputFactory outputFactory = XMLOutputFactory.newInstance();
        InputStream in = new FileInputStream(source);
        XMLEventReader reader = inputFactory.createXMLEventReader(in);

        OutputStream out = new FileOutputStream(target);
        XMLEventWriter writer =  outputFactory.createXMLEventWriter(out);
        XMLEvent event;

        boolean deleteSection = false;
        while(reader.hasNext())
        {
            event = reader.nextEvent();
            if(event.getEventType() == XMLStreamConstants.START_ELEMENT && event.asStartElement().getName().toString().equalsIgnoreCase("companies"))
            {
                deleteSection=true;
                continue;
            }
            else if(event.getEventType() == XMLStreamConstants.END_ELEMENT && (event.asEndElement().getName().toString().equalsIgnoreCase("companies")))
            {
                deleteSection=false;
                continue;
            }
            else if(deleteSection)
            {
                continue;
            }
            else
            {
                writer.add(event);
            }
        }
    }
}