Can PyMuPDF edit or create new signature fields?

[sk33z3r]

From what I gather in the documentation, it seems like it can't edit or add signature fields, only read info from existing signature fields?

What I'm trying to do:

• Iterate a folder of fillable PDFs
• Iterate each page's widgets looking for field names that match a format
• Do 1 of 2 things with each match:
  1. Change the field_type from 7 to 6 (text to signature)
  2. Delete the old widget, and create a new signature widget using the old widget's rect

Just wondering if what I am trying to accomplish is possible in this tool. Finding my fields and parsing through their metadata is no problem, I'm just running into roadblocks when it comes to the last step (changing/creating the widgets).

[JorjMcKie]

Sorry, this is not supported currently.

[sk33z3r]

No problem, thanks for the quick response!

[sk33z3r]

If anyone else happens to stumble in here looking to do the same process, I found a solution. I use LibreOffice Draw to create the PDF forms initially, which doesn't export proper PDF signature fields. I just create placeholder text fields in LibreOffice Draw with names that start with sig or init, then...

1. Use PyMuPDF to iterate the pages and store the field_name and rect for any matching fields
2. Convert the y coordinates to Cartesian
3. Use pyHanko to add new signature fields based on the stored data

It's worth noting that I ran into a known bug (#3216) but you can easily fix it by removing line 1701 from .venv/fitz/utils.py

A note on the coordinates:
As far as I can tell in PDF land, 72 units = 1 inch. PyMuPDF gives you coordinates where the page's (0,0) coordinate is the top-left, but pyHanko is expecting the page's (0,0) coordinate to be the bottom-left. So I had to subtract the received y values from PyMuPDF to my page's height (in units).

In my case, it is an 8.5" x 11" document, so the total height (in units) is 11 * 72, or 792. To convert my y values, it was as simple as 792 - y and everything lines up perfectly.

Here's my full testing script that gives me the results I want:

import shutil, fitz, os
from pyhanko.pdf_utils.incremental_writer import IncrementalPdfFileWriter
from pyhanko.sign.fields import SigFieldSpec, append_signature_field

og_file = "general_release.pdf"
new_file = "test.pdf"

print("Setting up new file...")
os.remove(new_file)
shutil.copyfile(og_file, new_file)

boxes = {}

print(f"Getting signature field data and removing placeholders...")
pn = 0
fdoc = fitz.open(new_file)
for page in fdoc:
    for field in page.widgets():
        n = field.field_name            
        if n.startswith("sig") or n.startswith("init"):
            print(f"Found field '{n}' on page {pn}...")
            _type = 1 if n.startswith("sig") else 2
            boxes[n] = {
                "page": pn,
                "type": _type,
                "box": (
                    field.rect.x0,
                    792-field.rect.y0,
                    field.rect.x1,
                    792-field.rect.y1
                )
            }
            field.field_name = n + "_orig"
            field.update()
            try:
                page.delete_widget(field)
                print("...removed!")
            except Exception as e:
                print(f"...failed to remove.\nERROR: {e}")
    pn += 1
fdoc.saveIncr()
fdoc.close()

print("Adding new signature fields to document...")
try:
    with open(new_file, 'rb+') as _doc:
        d = IncrementalPdfFileWriter(_doc)
        for _name in boxes.keys():
            _dict = boxes[_name]
            append_signature_field(d, SigFieldSpec(
                                        sig_field_name=_name,
                                        on_page=_dict["page"],
                                        box=_dict["box"]
                                    ))
        d.write_in_place()
    print("...done!")
except Exception as e:
    print(f"...failed to add signature fields.\nERROR: {e}")