How to save XFA properly to my PDF

[daniel-manaog]

Hi PyPDF community,

I am currently creating a project where I can automatically fill up the PDF forms using Python. I stumbled upon this library which is a perfect tool for me to edit PDF.

I am currently experimenting a bit and I have two pdf I have imported, one is not filled up but the other one is. I am only targeting the datasets in XFA to copy since, that is the data that I want to fill/edit. Here's is the code that I have used.

Upon execution, and trying to open the file, I'm seeing warning from Adobe Reader:

Also, if I try to fill up a form, it shows this, and I cannot save the file.

I believe the PDF file is encrypted and signed, and when using my code to edit it, it loses it's encryption (SECURED).
I looked into the documentations and tried incremental=True but I cannot save any XFA changes using this method.

Additionally, I can't send the PDF since it contains confidential information.

[stefan6419846]

I cannot speak for XFA as I have never used it. But

I believe the PDF file is encrypted and signed, and when using my code to edit it, it loses it's encryption (SECURED).

does not sound right. I suspect that the signature might become invalid, but the encryption should not be affected (if in place).

[daniel-manaog]

Here's the encryption info

Original PDF:

Filled-up PDF:

Saved PDF using pypdf:

[daniel-manaog]

More context:

The encryption works if I use the incremental=True. But it doesn't apply the changes I made on XFA.
I wonder if I could retain the encryption while editing XFA.

[daniel-manaog]

For anyone looking a solution for this, I decided to use iTextPDF5 in C# (alternative is itext7 in Java).
Unfortunately, even other python libraries are not supporting this kind of XFA modify.

The Stamping Properties and append in iTextPDF5 in C# works properly.