docx parsing was broken in 1.24.7

[ichenhe]

Description of the bug

Can not read the .docx file. It worked perfectly on v1.24.6.

Logs:

Traceback (most recent call last):
  File "~/d.py", line 15, in <module>
    print(extract_text(local_file_path))
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "~/d.py", line 8, in extract_text
    content += page.get_text() + '\n\n'
               ^^^^^^^^^^^^^^^
  File "~/python3.11/site-packages/pymupdf/utils.py", line 798, in get_text
    cb = page.cropbox
         ^^^^^^^^^^^^
  File "~/python3.11/site-packages/pymupdf/__init__.py", line 8535, in cropbox
    page = self._pdf_page()
           ^^^^^^^^^^^^^^^^
  File "~/python3.11/site-packages/pymupdf/__init__.py", line 8051, in _pdf_page
    return _as_pdf_page(self.this)
           ^^^^^^^^^^^^^^^^^^^^^^^
  File "~/python3.11/site-packages/pymupdf/__init__.py", line 337, in _as_pdf_page
    assert ret.m_internal
AssertionError

How to reproduce the bug

TEST.docx

import fitz


def extract_text(file: str) -> str:
    content = ""
    with fitz.open(file) as document:
        for page in document:
            content += page.get_text() + '\n\n'
    content = content.strip()
    return content


if __name__ == '__main__':
    local_file_path = '/path/to/TEST.docx'
    print(extract_text(local_file_path))

PyMuPDF version

1.24.7

Operating system

MacOS

Python version

3.11

[JorjMcKie]

Please always provide a reproducing file when reporting a bug.

[ichenhe]

Please always provide a reproducing file when reporting a bug.

@JorjMcKie Sorry, I have added the test docx file and a reproducible code snippet under reproduce section.

[JorjMcKie]

Thank you for letting us have a reproducing file.

[julian-smith-artifex-com]

Fixed in 1.24.8.