v1.24 breaks XPS-table example

[martin-dreyer-teck]

Description of the bug

With v1.23.26 running page.find_tables() with the provided example file XPS-table.pdf returns one table.
With v1.24, doing the same returns ZERO tables.

How to reproduce the bug

pip install PyMuPDF==1.24

import fitz

doc = fitz.open("XPS-table.pdf")
page = doc[0]
tabs = page.find_tables()
print(f"len={len(tabs.tables)}")

output: len=0

PyMuPDF version

1.24.0

Operating system

Linux

Python version

3.10

[julian-smith-artifex-com]

Could you provide the file XPS-table.pdf?

[JorjMcKie]

This is not a table-related problem. When taking a closer look it turns out that no text is extracted from the page.
Based on this, of course no table can ever be identified.

[JorjMcKie]

Closed b/o extended time without response.

[martin-dreyer-teck]

@JorjMcKie apologies, I missed the messages on this hence the late response. FYI the file XPS-table.pdf is in the pymupdf repo and presented in examples of how to do table extraction.

https://github.com/pymupdf/PyMuPDF-Utilities/tree/master/table-analysis