Is it possible to append or merge one or more XFA form-based PDF files together with iText?

Question

I have one PDF file that has an embedded form that is based on XFA (XML) forms. The first PDF has a table which holds a list of people. If that table overflows, the subsequent list of people are handled by an addendum page which is also a PDF (XFA based form). Is it possible to merge all XFA-based PDFs into one PDF using iText?

Answer 1

@BrunoLowagie Thanks for your response. Actually, I managed to get iText to concatenate PDF interactive forms to create a custom PDF packet. Let me explain how I did this.

From using Adobe Acrobat XI Pro, I learned that when the XFA PDF is loaded, I cannot edit the form if I go to Tools->Edit (it gives me the usual warning that this PDF was created by LiveCycle Designer), but when I go to Pages->Extract, then select all pages to be extracted, the whole XFA-based PDF is extracted and converted over to an AcroForms based PDF. So if I had 25 fields in the XFA-based PDF, it had successfully converted all 25 fields into AcroForm fields. Somehow Adobe Acrobat had to determine the variable names based on the XML structure. (ie xpath) //form1/Page1/variable1 was converted to acro field name: form1[0].Page1[0].variable1[0]. All visible (editable form) fields were present and aligned (pixel-perfect) as usual.

If I had flattened the XFA PDF, I would again need to place the form fields back on each page, which would be tedious. By using Pages->Extract->All Pages, it converts everything for me (no flattening needed - flattening which strips all fields also; also no XFA worker lib needed).

However, my PDF packet is static and I would like to repeat the addendum page in case data overflows from the 2nd page. I know I could have modified the initial XFA to handle this overflow, but the client wants to use the exact page look for the addendum, with headers/footers intact.

I found that I can achieve this by extracting the addendum page separately via Adobe Acrobat Pro->Pages->Extract->(Select Addendum Page), then it was converted to a PDF form w/AcroForm intact.

I took the original PDF packet and attempted to concatenated the addendum PDF page. So for the moment, the main packet has AcroForm fields and the Addendum PDF page also has AcroForm fields.

When I used PdfCopy or PdfConcatenate to do the concatenation, I noticed that I lost all form fields when I called form.getFields()

When I used the (deprecated) PdfCopyFields to do the concatenation, all AcroForm fields were intact. (Exactly what I needed!) I also tested the PdfCopyFields where some fields were filled/saved and the PdfCopyFields still worked & carried over the prefilled values over. I looked at the reason why PdfCopyFields was marked for deprecation, saying that we can either merge accessible files and lose the forms or merge the forms or lose the accessibility (tagged PDFs). What if I don't care about tagged PDFs or accessibility?....and still need PDFCopyFields to carry over the form fields intact. So far I'm forced to keep using the PDFCopyFields since it does exactly what I need for PDF concatenation with Interactive Form Fields. Is it possible to update PDFCopy to have an option to copy over the fields if PDFCopyFields is going away?