MuPDF error: syntax error: cannot find ExtGState resource 'BlendMode0'

[AndresEV22]

Description of the bug

Hi, first, sorry for my bad English, but I will try to explain my bug or error. Yesterday I updated my pyproject.toml dependencies, and at this moment I am getting an error when the convert_to_pdf() is executed. This error or bug happens when I try to convert a PDF with more than one page; if the PDF has just one page, this doesn't happen.  After searching on the internet, I found something about this error. It says that the error is because the PDFs are corrupt, but I tried to repair the PDF with the "gs -o corrected.pdf -sDEVICE=pdfwrite -dPDFSETTINGS=/prepress damage.pdf" Linux command or in another way with ILovePDF, but the error persists.

I applied the migration between 1.22.5 and 1.24.4 with these changes:

• import fitz to import pymupdf

How to reproduce the bug

The extact error is:

MuPDF error: syntax error: cannot find ExtGState resource 'BlendMode0'

MuPDF error: syntax error: cannot find ExtGState resource 'BlendMode0'

MuPDF error: syntax error: cannot find ExtGState resource 'BlendMode0'

MuPDF error: syntax error: cannot find ExtGState resource 'BlendMode0'

syntax error: cannot find ExtGState resource 'BlendMode0'
encountered syntax errors; page may not be correct
syntax error: cannot find ExtGState resource 'BlendMode0'
encountered syntax errors; page may not be correct
syntax error: cannot find ExtGState resource 'BlendMode0'
encountered syntax errors; page may not be correct
syntax error: cannot find ExtGState resource 'BlendMode0'
encountered syntax errors; page may not be correct

and this is my function:

def fill_pdf_form(
    pdf_form_stream: bytes,
    widget_values: dict[str, Any],
    image_boxes: list[tuple[pymupdf.Rect, bytes, int]],
) -> bytes:
    with pymupdf.Document(stream=pdf_form_stream) as pdf_document:
        for image_box, image_stream, page_number in image_boxes:
            pdf_page_to_image = cast(
                pymupdf.Page, pdf_document.load_page(page_number - 1)
            )
            pdf_page_to_image.insert_image(image_box, stream=image_stream)

        for pdf_page in pdf_document:
            for widget in pdf_page.widgets():
                widget.field_value = str(widget_values[cast(str, widget.field_name)])
                widget.update()

        return pdf_document.convert_to_pdf()

and my proyect environment is based on python hacth. To clarify this "error," permits work well but are not ideal.

PyMuPDF version

1.24.4

Operating system

Linux

Python version

3.12

[JorjMcKie]

As stated in the documentation: There can no guarantee for PDF-to-PDF conversion to work. Neither is "repairing" a PDF leading to an error-free PDF in all cases.

So this is no bug - just a PDF messed-up beyond a way to repair it. Why do you want to do that conversion anyway?

[AndresEV22]

Because I have a pdf with form fields, when these fields are fully complete, I need to remove this one, but the text can't disappear.

[JorjMcKie]

Because I have a pdf with form fields, when these fields are fully complete, I need to remove this one, but the text can't disappear.

Aha! Thought it must be this. Please look at Document method bake. It will remove annotations and / or form fields and make their current appearance a permanent part of the page.

[AndresEV22]

Oh, I see. My bad. Thanks for the help. I am grateful.