PPF-1298: Created fields have no entries in the Interactive Form (AcroForm) Dictionary

[MosheLaks]

Version

PyPDFForm=3.8.2

Issue Description

When I use the create_field or bulk_create_fields methods, the fields are created, but they have no entries in the Interactive Form Dictionary which is supposed to list all the fields in a PDF Form. This causes the PyPDF library to not see the fields (I use PyPDF to get the coordinates of fields on an existing PDF, since this library does not expose that data)

The PDF Spec describes the Interactive Form dictionary in section 12.7.2 (here).
See also section 7.7.2 for a description of the Document Catalog, which is supposed to link the Interactive Form in the AcroForm entry
In the example below, there is no AcroForm entry at all; other times I have seen it with an empty field list. I used this site to inspect my PDF, but maybe you have better tools.

Code Snippet

from PyPDFForm import PdfWrapper, BlankPage, Fields

def demo(output:str):
    wrapper = PdfWrapper(BlankPage())
    fields = [
        Fields.TextField(name="Test1", page_number=1, x=100, y=600),
        Fields.TextField(name="Test2", page_number=1, x=100, y=500),
    ]
    wrapper.bulk_create_fields(fields)
    wrapper.create_field(
        Fields.TextField(
            name="Test3", page_number=1, x=100, y=400
        )
    )
    wrapper.write(output)

PDF Form Template

test.pdf

[chinapandaman]

Hey thanks for posting.

I think it might be true that the standard asks for (root) fields to be presented in the Fields list of the AcroForm entry. However from a personal experience this Fields list serve little to no purpose, as for many PDF forms the form fields are rarely the root fields and this list does not affect how PDF softwares render the final result.

What I found to be a more reliable way of checking fields is through each page's /Annots object. So in your case you would like to know the coordinates, so something like this worked for me:

from PyPDFForm import PdfWrapper, BlankPage, Fields
from pypdf import PdfReader
from io import BytesIO

wrapper = PdfWrapper(BlankPage())
fields = [
    Fields.TextField(name="Test1", page_number=1, x=100, y=600),
    Fields.TextField(name="Test2", page_number=1, x=100, y=500),
]
wrapper.bulk_create_fields(fields)
wrapper.create_field(Fields.TextField(name="Test3", page_number=1, x=100, y=400))

buff = BytesIO(wrapper.read())
buff.seek(0)
reader = PdfReader(buff)

for page in reader.pages:
    for annot in page["/Annots"]:
        print(annot["/T"], annot["/Rect"])

[MosheLaks]

Hi, thanks for responding! Also, while we're at it, thanks for the library.
I was actually in the middle of creating another Issue with a feature request for more inspection data (especially coordinates) on existing form fields when I saw your reply. I'd rather not have to fall back to PyPDF in my code. In the meantime, I'll probably borrow the excerpt you provided.
If I ever get around to trying my hand at fixing this issue, would you mind? Anything I should know before I try?