Problem about font encoding in PDF/A generation

Question

So here is my problem : I'm currently working on an java application that will archive document in a PDF/A-1. I'm using PdfBox for pdf generation and when I can't generate a valid PDF/A-1 pdf, because of the font. The font is embedded inside the pdf file but this website : https://www.pdf-online.com/osa/validate.aspx tell me that this is not a valid PDF/A because of :

The key Encoding has a value Identity-H which is prohibited.

I look on internet on what is this Identity-H encoding and it seem that it's the way that font are encoded, like the ansi encoding.

I've already tried to get different font like Helvetica or arial unicode Ms but nothing work, there is alway this Identity-H encoding.I'm a bit lost with all this mess in encoding so if someone can explain me it'll be great. Also here is the code I write to embedded a font in the pdf :

 // load the font as this needs to be embedded
PDFont font = PDType0Font.load(doc, getClass().getClassLoader().getResourceAsStream(fontfile), true);

if (!font.isEmbedded())
        {
            throw new IllegalStateException("PDF/A compliance requires that all fonts used for"
                    + " text rendering in rendering modes other than rendering mode 3 are embedded.");
        }

Thanks for your help :)

Answer 1

Problem solved :

I used the example of apache : CreatePDFA ( I have no clue why that work and not my code ) : Example in examples/src/main/java/org/apache/pdfbox/examples

I add to fit the PDF/A-3 requirement :

doc.getDocumentCatalog().setLanguage("en-US");

PDMarkInfo mark = new PDMarkInfo(); // new PDMarkInfo(page.getCOSObject());
PDStructureTreeRoot treeRoot = new PDStructureTreeRoot();
doc.getDocumentCatalog().setMarkInfo(mark);
doc.getDocumentCatalog().setStructureTreeRoot(treeRoot);
doc.getDocumentCatalog().getMarkInfo().setMarked(true);

PDDocumentInformation info = doc.getDocumentInformation();
info.setCreationDate(date);
info.setModificationDate(date);
info.setAuthor("KairosPDF");
info.setProducer("KairosPDF");
info.setCreator("KairosPDF");
info.setTitle("Generated PDf");
info.setSubject("PDF/A3-A");

Here is my code to embedded a file to the pdf :

private final PDDocument doc = new PDDocument();
private final PDEmbeddedFilesNameTreeNode efTree = new PDEmbeddedFilesNameTreeNode();
private final PDDocumentNameDictionary names = new PDDocumentNameDictionary(doc.getDocumentCatalog());
private final Map<String, PDComplexFileSpecification> efMap = new HashMap<>();

public void addFile(PDDocument doc, File child) throws IOException {
    File file = new File(child.getPath());

    Calendar date = Calendar.getInstance();

    //first create the file specification, which holds the embedded file
    PDComplexFileSpecification fs = new PDComplexFileSpecification();
    fs.setFileUnicode(child.getName());
    fs.setFile(child.getName());
    InputStream is = new FileInputStream(file);
    PDEmbeddedFile ef = new PDEmbeddedFile(doc, is);

    //Setting
    ef.setSubtype("application/octet-stream");
    ef.setSize((int) file.length() + 1);
    ef.setCreationDate(date);
    ef.setModDate(date);
    COSDictionary dictionary = fs.getCOSObject();
    dictionary.setItem(COSName.getPDFName("AFRelationship"), COSName.getPDFName("Data"));

    fs.setEmbeddedFile(ef);

    efMap.put(child.getName(), fs);
    efTree.setNames(efMap);

    names.setEmbeddedFiles(efTree);
    doc.getDocumentCatalog().setNames(names);
    is.close();
}

The only problem left is this error from the validation :

File specification 'Test.txt' not associated with an object.

Hope it'll help some.