HTMLEditorKit adds spaces and tags. Can I make it stop?

Question

I've developed an application that scores boat races and outputs the results in very compact HTML. One way to keep the HTML compact is to make tactical decisions about what tags to leave out, such as "</td>".
Another is to put in virtually no white space unless it is part of the end user's display.
This is all well and good when I save directly to the disk. But if I first display the string as HTML in a JEditorPane, it seems that the HTMLEditorKit wants to reformat the document, adding all the missing tags and a bunch of spaces and returns as well. THe prettier document is now twice the size (28k vs 14k). This is without me having made any editing changes.
I can, of course save the original string directly, but I would like to give users the ability to edit before saving. Is there some sort of "don't make changes I did not ask for" fix to this item?

Answer 1

I don't know of any easy way to do what you want with JEditorPane
There are some packages that do HTML formatting for you. For example, you might be able to use JTidy (I've never tried it) to format the text before saving it. It has different options which you may be able to use to control the level of formatting.
Or if you want full control over formatting, you could try using the ParserCallback class. It parses a text string and notifies you every time it finds a tag or attributes. You could then simply format the tag and attributes in the format you want. This example might get you started:
import java.io.*;
import java.net.*;
import javax.swing.text.*;
import javax.swing.text.html.*;
import javax.swing.text.html.parser.*;
public class ParserCallbackTest extends HTMLEditorKit.ParserCallback
     private int tabLevel = 1;
     public void handleComment(char[] data, int pos)
          displayData(new String(data));
     public void handleEndOfLineString(String eol)
     public void handleEndTag(HTML.Tag tag, int pos)
          tabLevel--;
          displayData("/" + tag);
     public void handleError(String errorMsg, int pos)
          displayData(pos + ":" + errorMsg);
     public void handleMutableTag(HTML.Tag tag, MutableAttributeSet a, int pos)
          displayData("mutable:" + tag + ": " + pos + ": " + a);
     public void handleSimpleTag(HTML.Tag tag, MutableAttributeSet a, int pos)
          displayData( tag + "::" + a );
//          tabLevel++;
     public void handleStartTag(HTML.Tag tag, MutableAttributeSet a, int pos)
          displayData( tag + ":" + a );
          tabLevel++;
     public void handleText(char[] data, int pos)
          displayData( new String(data) );
     private void displayData(String text)
          for (int i = 0; i < tabLevel; i++)
               System.out.print("\t");
          System.out.println(text);
     public static void main(String[] args)
     throws IOException
          ParserCallbackTest parser = new ParserCallbackTest();
          // args[0] is the file to parse
          Reader reader = new FileReader(args[0]);
//          URLConnection conn = new URL(args[0]).openConnection();
//          Reader reader = new InputStreamReader(conn.getInputStream());
          try
               new ParserDelegator().parse(reader, parser, true);
          catch (IOException e)
               System.out.println(e);
}You can use a StringReader class as the input to the parser.