I’m trying to convert a regular flavour PDF to a PDF/A. I already found some articles here that give some insight on how to produce a PDF/A. But every time I save the the PDDocument the resulting file is either empty (0kb) or corrupt. PDFBox 2.0.3 produces a corrupt PDF and PDFBox 3.0 just generates a no size document.

I tried both PDFBox 3 and PDFBox 2 both give me wrong results. What I could see from tests is that as soon as I call the getDocumentCatalog() on the PDDocument and I try to save the PDDocument the problem is there. It actually doesn’t matter what the next actions are.

Here is some of the code

private void initializeDocument() throws Exception {
    try (PDDocument loadedDoc = Loader.loadPDF(new RandomAccessReadBufferedFile(pdfInfo.getInputFile())))

    {
        document = loadedDoc;
        document.setVersion(1.7f);

    }

    catch (IOException ioException) {
        throw new Exception(ioException.getMessage());
    }

}

private void addMetadata() throws Exception {
    catalog = document.getDocumentCatalog();
    XMPMetadata xmp = XMPMetadata.createXMPMetadata();
    Calendar cal = Calendar.getInstance();

    try {
        // set core metadata
        DublinCoreSchema dublinCoreSchema = new DublinCoreSchema(xmp);
        dublinCoreSchema.setTitle(pdfInfo.getOutputFileName());
        dublinCoreSchema.addDate(cal);
        dublinCoreSchema.addCreator(pdfInfo.getCreator());

        XMPBasicSchema xmpBasicSchema = xmp.createAndAddXMPBasicSchema();
        xmpBasicSchema.setAboutAsSimple(pdfInfo.getCreator());
        xmpBasicSchema.setCreateDate(cal);
        xmpBasicSchema.setCreatorTool(pdfInfo.getCreator());

        PDDocumentInformation pdDocumentInformation = new PDDocumentInformation();
        pdDocumentInformation.setTitle(pdfInfo.getOutputFileName());
        pdDocumentInformation.setProducer(pdfInfo.getCreator());
        pdDocumentInformation.setCreator(pdfInfo.getCreator());
        document.setDocumentInformation(pdDocumentInformation);

        PDMarkInfo markinfo = new PDMarkInfo();
        markinfo.setMarked(true);
        this.catalog.setMarkInfo(markinfo);

        PDFAIdentificationSchema pdfaIdentificationSchema = xmp.createAndAddPDFAIdentificationSchema();
        pdfaIdentificationSchema.setConformance("A");
        pdfaIdentificationSchema.setPart(3);
        pdfaIdentificationSchema.setAboutAsSimple("");

        XmpSerializer serializer = new XmpSerializer();
        ByteArrayOutputStream baos = new ByteArrayOutputStream();
        serializer.serialize(xmp, baos, true);

        PDMetadata metadata = new PDMetadata(document);
        metadata.importXMPMetadata(baos.toByteArray());
        this.catalog.setMetadata(metadata);

    }

    catch (BadFieldValueException e) {
        throw new IllegalArgumentException(e.getMessage());
    }

    catch (Exception e) {
        throw new IllegalArgumentException(e.getMessage());
    }
}

private void saveFile() {
    try (OutputStream fileOutputStream = new FileOutputStream(pdfInfo.getOutputFile())) {
        document.save(fileOutputStream);
        document.close();
        fileOutputStream.close();
    } catch (Exception e) {
        // TODO: handle exception
    }

}