pdf / A合并itext 5 java

Question

I'm trying merge two PDFs with itext 5. When the process ends, one PDF is generated, but the foxit reader doesn't read the PDF. Invalid format This is my code

    public static byte[] mergePdf(byte[][] streamOfPDFFiles) throws IOException, DocumentException {

      ByteArrayOutputStream baos = new ByteArrayOutputStream();
      float MARGIN_OF_ONE_CM = 28.8f;
      Document document = new Document(PageSize.A4, MARGIN_OF_ONE_CM, MARGIN_OF_ONE_CM, MARGIN_OF_ONE_CM,MARGIN_OF_ONE_CM);
      document.addLanguage("pt-BR");
      List<PdfReader> readers = new ArrayList<PdfReader>();       
      int totalPages = 0;
      byte[][] pdfs = streamOfPDFFiles;
      try {

        // Create Readers for the pdfs.
        for (byte[] bs : pdfs) {
            PdfReader pdfReader = new PdfReader(bs);
            readers.add(pdfReader);
            totalPages += pdfReader.getNumberOfPages();
        }

    } catch (Exception e) {
        e.printStackTrace();
    }
      // step 2
      PdfACopy copy = new PdfACopy(document, baos, PdfAConformanceLevel.PDF_A_1A);
      // step 3
      document.open();
      // step 4

      int n;
      // loop over the documents you want to concatenate
      for (int i = 0; i < readers.size(); i++) {              
          // loop over the pages in that document
          n = readers.get(i).getNumberOfPages();
          for (int page = 1; page <= n;) {
              copy.addDocument(readers.get(i));
              ++page;
          }
          copy.freeReader(readers.get(i));
          readers.get(i).close();
      }
      byte[] pdfRetorno = baos.toByteArray();
      return pdfRetorno;
}

Can anyone helps me? thanks in advance

Answer 1

Whenever you open something, you should eventually close it again, in particular as closing may involve more than freeing resources, it may also include adding the final piece to written data.

This also is the case here, you open your Document document :

document.open();

but don't close it thereafter. Thus, the PDF written to your ByteArrayOutputStream baos is incomplete, at least the final PDF objects cross references table and the trailer are missing.

Thus, close the document before working with the result in baos :

for (int i = 0; i < readers.size(); i++) {
    [...]
}
document.close();
byte[] pdfRetorno = baos.toByteArray();
return pdfRetorno;

Answer 2

Here, the correct code:

public static byte[] mergePdf(byte[][] streamOfPDFFiles) throws IOException, DocumentException {

    final float MARGIN_OF_ONE_CM = 28.8f;
    ByteArrayOutputStream baos = new ByteArrayOutputStream();
    Document document = new Document(PageSize.A4, MARGIN_OF_ONE_CM, MARGIN_OF_ONE_CM, MARGIN_OF_ONE_CM,
            MARGIN_OF_ONE_CM);
    List<PdfReader> readers = new ArrayList<PdfReader>();

    byte[][] pdfs = streamOfPDFFiles;
    try {

        // Create Readers for the pdfs.
        for (byte[] bs : pdfs) {
            PdfReader pdfReader = new PdfReader(bs);
            readers.add(pdfReader);             
        }
    } catch (Exception e) {
        e.printStackTrace();
    }

    PdfACopy copy = new PdfACopy(document, baos, com.itextpdf.text.pdf.PdfAConformanceLevel.PDF_A_1A);

    copy.setTagged();
    document.open();
    copy.createXmpMetadata();
    copy.setLanguage("pt-br");                  

    int n;
    for (int i = 0; i < readers.size(); i++) {          
        n = readers.get(i).getNumberOfPages();
        for (int page = 1; page <= n;) {
            copy.addDocument(readers.get(i));
            ++page;
        }
        copy.freeReader(readers.get(i));            
    }

    document.close();
    byte[] pdfRetorno = baos.toByteArray();
    return pdfRetorno;
}