Java And OpenDocument Files

Hi Friends,

Now, lets say you have a file, lets call it document.odt

I have noticed that .odt files are not "stand alone" files like .txt or .dat files. Rather they are like zipped files with other defining files in there.

Before I get out of track, let me go to my question.

document.odt contains some text, say "Hello World". I want to edit this text through Java and save the resulting text. How can this be achieved?

Thanks and regards.

OpenOffice has a Java API.

And how can I use the API to open/edit files through a TextArea on my Java application?

Accessing ODF files through the OO Java API and using the data in GUI elements are separate activities. You'll need to dig into the API to figure out how to access the data that you need for the GUI. The AccessingFileFormats wiki page links to a number of articles about the OO Java API and the ODF file format.

Dear Dittmer,

I am simply lost in the array of suggestions in those pages...

The OpenOffice Java API

OpenOffice can read a number of file formats, and makes them accessible through its API. A starting point might be this article and of course the OO developer site
Some introductory information about the OO file format can be found here and here - basic Java code for reading OO files is here

Reading an OpenOffice file is not as simple as reading a plain text file, simply because OpenOffice contains a lot more features than a plain text editor.
[ October 31, 2007: Message edited by: Jesper Young ]

So whats your suggestion on how to open them?

Either use the OO Java API to convert the file to some other format that you do know how to open, or -based on the articles linked above- write code that opens and processes the files.

From my cursory look at the articles OO files appear to be zipped-up XML files. Since Java has APIs for dealing with ZIP and XML files, getting at the actual contents shouldn't be too hard. Making sense of those is a different matter, of course - I'd recommend to start with a simple document, to see if you can manage to extract whatever information you need.

I guess this leads to the next question, my good people.

How do I use the Java API to zip/unzip file packages?

Examples of using the ZIP API -as well as just about all other java.* classes- can be found at the Developer's Almanac: http://www.exampledepot.com/egs/java.util.zip/pkg.html

from the reference you gave me has solved my most immediate problem. Thanks!

This is a sample output:

and the code

retrieves only the first entry whereas I would want the file called content.xml to be extracted, if not the entire odt file, onto a specified folder. How or what do I need to modify in the code?

You need a loop. ZipInputStream.getNextEntry() returns null when there are no more entries, so you can use that as the loop controller.
If you only want to output certain files, then you need to check the name of each ZipEntry object before you write it out. Check the API docs to see if there is a method that gets the name of the ZipEntry object.

My redo of the code is

import java.util.*;
import java.io.*;
import java.util.zip.*;
 
 
public class Unzipper
{
    private String inFilename;
    private ZipInputStream zipInputStream;
    private OutputStream out;
    private ZipEntry entry;
    
    public Unzipper()
    {
        inFilename = "/home/arthur/filename.odt";
        
    }
    
    public void unzip()
    {
        try
        {
            // Open the ZIP file
            
            zipInputStream = new ZipInputStream(new FileInputStream(inFilename));
            
            entry = zipInputStream.getNextEntry();
            
            // Get the first entry
            while(entry != null)
            {
                System.out.println("Opening...");
                // Open the output file
                String outFilename = entry.getName();
                new File("/home/arthur/unzipped/" + outFilename).mkdirs();
                out = new FileOutputStream("/home/arthur/unzipped/" + outFilename);
                
                
                // Transfer bytes from the ZIP file to the output file
                byte[] buf = new byte[1024];
                int len;
                while ((len = zipInputStream.read(buf)) > 0)
                {
                    out.write(buf, 0, len);
                }
                out.flush();
                entry = zipInputStream.getNextEntry();
            }
            // Close the streams
            out.close();
            zipInputStream.close();
        }
        catch (IOException e)
        {
            e.printStackTrace();
        }
    }
 
static public void main(String[] args)
{
new Unzipper().unzip()
}
}

What could be the issue here as it returns lots of null and empty folders?

You create a new FileOutputStream object every time through the loop, but you only close the last one after you exit the loop. Put the out.close() call inside the loop.

There is a piece of software called JODConverter that can open any file that OpenOffice can open and export to any that OpenOffice can export to. I am happy to say that it has really helped me solve the issues.

PS: It is FOSS