Issue with Accented Characters in PDF Form Fields (checkbox and radiobutton)

[JorjMcKie]

Discussed in #4971

^{Originally posted by dantard April 20, 2026}
Hello,

I'm working with a PDF file that includes form fields such as radio buttons and checkboxes. After a few hours of debugging, I identified a specific issue: when the on_state value of a checkbox or radio button contains accented characters, calling widget.update() fails. That is:

widget.field_value = True
widget.update()

throws:

File "/home/danilo/python-venv/lib/python3.12/site-packages/pymupdf/mupdf.py", line 53520, in pdf_set_field_value
    return _mupdf.pdf_set_field_value(doc, field, text, ignore_trigger_events)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
TypeError: in method 'pdf_set_field_value', argument 3 of type 'char const *'

In my particular case, the on_state is set to S#ED, which is intended to represent “Sí”. This encoding seems to be the root of the problem.

From what I can tell, this may be related to how the PDF library or underlying encoding handles non-ASCII characters in form field values.

Has anyone encountered a similar issue? If so, how did you handle or work around it?

Any insights or suggestions would be appreciated.

Thanks is advance.

[JorjMcKie]

The widget on_state() method must be changed to replace any occurrence of "#LR" in the "ON" string by chr(0xLR) - IAW "L" and "R" must be regarded as hexadecimal digits.