page.apply_redactions gives unwanted black rectangle

[CH-Tam]

Description of the bug

I want to remove all texts and only keep vector graphics (such as straight lines) in PDF, the code and result are shown below.

However, the original PDF does not contain that black rectangle, the output is undesired. What should I do? (the PDF file is non-public, but it can be provided though private channels.)

Both 1.24.3 and 1.24.7 have this error.

How to reproduce the bug

        doc = fitz.open(pdf_path)
                for page in doc:
                    page.add_redact_annot(page.rect + (-600, -600, 600, 600))
                    page.apply_redactions(graphics=0)

The result of 1st page is

PyMuPDF version

1.24.7

Operating system

Linux

Python version

3.10

[JorjMcKie]

Please send me the file via e-mail, thanks.

[JorjMcKie]

@CH-Tam - I received nothing so far ... did you use [email protected]?

[JorjMcKie]

Thanks - file received. This is a problem in the base library for which a fix has been developed. I tested it locally with the improvements and confirmed that it works.

The fix works with MuPDF v1.25.0. (the development version on GitHub).

[JorjMcKie]

This has been fixed by a recent release - probably already 1.24.10.

[CH-Tam]

hi @JorjMcKie, I have tried the new versions (1.24.10 & 1.24.11). In 1.24.10, the issue is completely solved; however in 1.24.11, the above code has no impact at all (cannot clear the text in the PDF).