Question

I’m trying to solve a problem using java, iText, and the Java advanced imaging library. My software system uses ghostscript to create jpg thumbnail images etc… from PDF files. However on CentOS 5.x the highest version of ghostscript is 8.7 which has a known issue of not being able to handle PDF files containing JPEG 2000 images in them. My plan is to scan the file first and see if it contains jpeg2000 images (I’ve already got this part figured out); if so, then use iText and the Java Advanced Imaging library (contains the jpeg2000 read & write codecs) to convert the contained jpeg2000 files into regular jpeg files & then pass the new PDF file to ghostscript. The code below attempts this, but results in another file containing jpeg2000 files. Any help with this would be much appreciated.

public class ImageReplacer{
    public static void main(String [] args){
        try{
            String RESULT = "";
            PdfReader reader = new PdfReader("pdf_containing_jpeg2000_images.pdf");
            PdfReaderContentParser parser = new PdfReaderContentParser(reader);
            MyImageRenderListener listener = new MyImageRenderListener(RESULT);
            MyImageConverterListener clistener = new MyImageConverterListener(RESULT);
            clistener.setReader(reader);
            for (int i = 1; i <= reader.getNumberOfPages(); i++) {
                parser.processContent(i, clistener);
            }   
            PdfStamper stamper = new PdfStamper(reader, new FileOutputStream("out.pdf"));
            stamper.close();            
        }catch(Exception e){
            e.printStackTrace();    
        }
    }
}
class MyImageConverterListener implements RenderListener {
    protected String path = "";
    protected PdfReader reader;
    public MyImageConverterListener(String path) {
        this.path = path;
    }
    public void beginTextBlock() { }
    public void endTextBlock() { }
    public void renderImage(ImageRenderInfo renderInfo) {
        try {
            PdfImageObject image = renderInfo.getImage();
            PdfName filter = (PdfName)image.get(PdfName.FILTER);
            if (PdfName.JPXDECODE.equals(filter)) {
                if(image.getDictionary().isStream()){
                    BufferedImage bi = image.getBufferedImage();
                    if (bi == null) return; 
                    int width = (int)bi.getWidth();
                    int height = (int)bi.getHeight();
                    ByteArrayOutputStream imgBytes = new ByteArrayOutputStream(); 
                    ImageIO.write(bi, "JPG", imgBytes);
                    PRStream stream = new PRStream(reader,imgBytes.toByteArray());
                    stream.clear();
                    stream.setData(imgBytes.toByteArray(), false, PRStream.NO_COMPRESSION);
                    stream.put(PdfName.TYPE, PdfName.XOBJECT);
                    stream.put(PdfName.SUBTYPE, PdfName.IMAGE);
                    stream.put(new PdfName("foo"+Math.random()), new PdfName("bar"+Math.random()));
                    stream.put(PdfName.FILTER, PdfName.DCTDECODE);
                    stream.put(PdfName.WIDTH, new PdfNumber(width));
                    stream.put(PdfName.HEIGHT, new PdfNumber(height));
                    stream.put(PdfName.BITSPERCOMPONENT, new PdfNumber(8));
                    stream.put(PdfName.COLORSPACE, PdfName.DEVICERGB);
                }
            }
        } catch (Exception e) {
            e.printStackTrace();
        }
    }
    public void renderText(TextRenderInfo renderInfo) { }
    public void setReader(PdfReader r){
        reader = r;
    }
}

Answer 1

So I managed to solve this one on my own (with a little help from iText in action by Bruno Lowagie – great book). Just to re-iterate, my intention is to scan a PDF using iText to see if it contains any JPEG2000 images and if it does output the same PDF but with the inner JPEG2000 images replaced with regular JPEG images. This solves the fatal ghostscript 8.7 ‘Unable to process JPXDecode data’ error, but would also provide useful for making PDF’s iOS compatible.

So without further a do kids; here goes…

Step 1) Download iText 5.x .jar file, and download jai_imageio-1.1.jar (the Java advanced imaging library that allows you to convert JPEG2000 files)

Step 2) Create a file called PDFConverter.java and put this code in it:

import com.itextpdf.text.pdf.PdfReader;
import com.itextpdf.text.pdf.PdfName;
import com.itextpdf.text.pdf.PdfObject;
import com.itextpdf.text.pdf.PRStream;
import com.itextpdf.text.pdf.parser.PdfImageObject;
import com.itextpdf.text.pdf.PdfNumber;
import javax.imageio.ImageIO;
import java.awt.image.BufferedImage;
import com.itextpdf.text.pdf.PdfStamper;
import java.io.*;

public class PDFConverter{
    public static void main(String [] args){
        if(args.length==1){
            if(hasJpeg2000(args[0])){
                System.out.println("Contains JPEG2000 images: Converting them to JPEG..."); 
                convertPDF(args[0]);
                System.out.println("Done...");
            }else{
                System.out.println("Doesn't contain any JPEG2000 images: Nothing to be done...");   
            }
        }else{
            System.out.println("Please specify a PDF filename as a command line argument!");    
        }
    }
    public static boolean hasJpeg2000(String s){
        try{
            PdfReader reader = new PdfReader(s); 
            int n = reader.getXrefSize();
            PdfObject object; 
            PRStream stream; 
            for (int i = 0; i < n; i++) {
                object = reader.getPdfObject(i); 
                if (object == null || !object.isStream())continue; 
                stream = (PRStream)object; 
                PdfImageObject image = new PdfImageObject(stream);
                PdfName filter = (PdfName)image.get(PdfName.FILTER);
                if (PdfName.JPXDECODE.equals(filter)) {
                    return true;
                }
            }
        }catch(Exception e){
            e.printStackTrace();
        }
        return false;
    }
    public static void convertPDF(String s){
        try{
            PdfReader reader = new PdfReader(s); 
            int n = reader.getXrefSize();
            PdfObject object; 
            PRStream stream; 
            for (int i = 0; i < n; i++) {
                object = reader.getPdfObject(i); 
                if (object == null || !object.isStream())continue; 
                stream = (PRStream)object; 
                PdfImageObject image = new PdfImageObject(stream);
                PdfName filter = (PdfName)image.get(PdfName.FILTER);
                if (PdfName.JPXDECODE.equals(filter)) {
                    BufferedImage bi = image.getBufferedImage(); 
                    if (bi == null) continue; 
                    int width = (int)(bi.getWidth());
                    int height = (int)(bi.getHeight());
                    ByteArrayOutputStream imgBytes = new ByteArrayOutputStream(); 
                    ImageIO.write(bi, "JPG", imgBytes); 
                    stream.clear(); 
                    stream.setData(imgBytes.toByteArray(),false, PRStream.NO_COMPRESSION); 
                    stream.put(PdfName.TYPE, PdfName.XOBJECT); 
                    stream.put(PdfName.SUBTYPE, PdfName.IMAGE); 
                    stream.put(new PdfName("foo"+Math.random()), new PdfName("bar"+Math.random())); 
                    stream.put(PdfName.FILTER, PdfName.DCTDECODE); 
                    stream.put(PdfName.WIDTH, new PdfNumber(width)); 
                    stream.put(PdfName.HEIGHT, new PdfNumber(height)); 
                    stream.put(PdfName.BITSPERCOMPONENT,new PdfNumber(8)); 
                    stream.put(PdfName.COLORSPACE, PdfName.DEVICERGB);
                }
            }
            PdfStamper stamper = new PdfStamper(reader, new FileOutputStream("out.pdf")); stamper.close();
        }catch(Exception e){
            e.printStackTrace();
        }
    }
} 

Step 3)Compile the above file in the following manner:

javac -cp .:iText-5.0.4.jar:jai_imageio-1.1.jar PDFConverter.java

Step 4)Run the program with a PDF…

java -cp .:iText-5.0.4.jar:jai_imageio-1.1.jar PDFConverter PDFFileName.pdf

Booyah…