missing function to repair non-reference fields in corrupted PDF

[MisterStump]

Edit 2/14: Title changed. The issue is the PDF sample included does not have its fields data structured correctly. This issue item is now for the possible addition of handling of this form of corruption

---

On some PDFs, get_fields and get_form_text_fields both return incomplete data. This only occurs on 1 or 2 PDFs I've encountered out of a couple dozen. In the included example PDF, there are 15 fields but only 8 are returned. There are no errors or trackback I can find.

Environment

Which environment were you using when you encountered the problem?

$ python -m platform
Windows-10-10.0.19045-SP0

$ python -c "import pypdf;print(pypdf._debug_versions)"
pypdf==4.0.1, crypt_provider=('local_crypt_fallback', '0.0.0'), PIL=none

Code + PDF

This is a minimal, complete example that shows the issue:

from pypdf import PdfReader, PdfWriter
import os

filePath = "ExampleForm.pdf"
fileObj = open(filePath, 'rb')
reader = PdfReader(fileObj)

getFields = reader.get_fields()
getTextFields = reader.get_form_text_fields()

print("getFields: ", len(getFields))
print("getTextFields: ", len(getTextFields))
print(getTextFields)
#Print results:
"""
getFields:  8
getTextFields:  8
{'txtAgentExecName': None, 'txtTitleEx': None, 'Entity Name': None, 'Notice Date': '  ', 'Entity Name (Contact Info)': None, 'Zip': None, 'Registered Agent': '  ', 'Registered Agent (Signature)': ' '}
"""

PDF from Example:
ExampleForm.pdf

You may add this to your tests.

Traceback

This is the complete traceback I see:

No traceback. I don't see any erroring/warning.

[pubpub-zz]

Your pdf is not in accordance with standards: It contains some fields that are not child of another field (no parent child) and that are not referenced in pdf Root/Acroform/Fields

[MisterStump]

Thank you, I don't know enough about the format to recognize that, but I understand if it is a PDF-specific. Should I close this out?

[pubpub-zz]

I propose you to leave it open but to replace the title with "missing function to repair non refrence fields in corrupter PDF" I have in mind a way to propose a new fixing feature

[MisterStump]

Updated the title and description, and I will leave it open. Thank you!