Undetected character styles

[JorjMcKie]

Description of the bug

Underline and strikeout are not detected on the page of the attached PDF.

test.pdf

Reproducer:

mutool convert -o test.stext -O collect-styles test.pdf

For example, this line should show underlined and strikeout characters.

The relevant vector graphics are stroked with a rectangle height of 0 and a line width of 0.5.

How to reproduce the bug

Standard install python -m pip install -U pymupdf.
Using MuPDF v1.26.0.

PyMuPDF version

1.25.5

Operating system

Windows

Python version

3.13

[JorjMcKie]

Bugzilla link https://bugs.ghostscript.com/show_bug.cgi?id=708545

[sebras]

This bug appears to have been fixed upstream. Please test it! :)

[JorjMcKie]

This bug appears to have been fixed upstream. Please test it! :)

Confirming that the bug is mostly fixed in MuPDF version post 1.26.1 (aka 1.27.0).
The current v1.26.1 does not contain commit 7a437b3899fcafe7fc6a4fbaf8940e36c765d541.

In the quoted line above the marked word "and," is not recorded as strikeout:

I will also add a comment in bugzilla. Here is the produced .stext file, please look at line 1428.

test.zip

[JorjMcKie]

MuPDF version 1.26.3 has now completely resolved the problem.

[julian-smith-artifex-com]

Do we have a test for this?

[JorjMcKie]

All I currently have is non-PyMuPDF because the error was detected using mutool convert -o test.stext -O collect-styles,accurate-bboxes test.pdf.

We certainly could make a PyMuPDF script that confirms the highlighted "and," above is now indeed strikeout.