RuntimeError: cycle in page tree

[hellohr11]

Please provide all mandatory information!

Describe the bug (mandatory)

When open pdf file in multiple functions, sometimes I got runtimeerror: cycle in page tree.
It doesn't happen to every pdf. Sometimes, when the error happens, if I change the pdf file path from absolute path to relative path, the problem will be gone.

To Reproduce (mandatory)

import fitz
FILENAME='blueprint (1).pdf'
def do_some_analysis(filename,pageno):
    doc=fitz.open(filename)
    page=doc[pageno-1]
    
    
doc=fitz.open(FILENAME)

for i in range(doc.page_count):
    page=doc[i]
    print(i)
    do_some_analysis(FILENAME,i+1)

For problems when building or installing PyMuPDF, give the full output of the build/install command so that, for example, all pip/compiler/linker errors/warnings can be seen.

Expected behavior (optional)

Describe what you expected to happen (if not obvious).

Screenshots (optional)

If applicable, add screenshots to help explain your problem.
For the same file, running the code snippet
when I use absolute path

when I use relative path

Your configuration (mandatory)

• Operating system, potentially version and bitness
  ubuntu 22.04
• Python version, bitness
  python 3.10.6
• PyMuPDF version, installation method (wheel or generated from source).
  pip install pymupdf
  version 1.21.0

For example, the output of print(sys.version, "\n", sys.platform, "\n", fitz.__doc__) would be sufficient (for the first two bullets).
3.10.6 (main, May 29 2023, 11:10:38) [GCC 11.3.0]
linux

PyMuPDF 1.21.0: Python bindings for the MuPDF 1.21.0 library.
Version date: 2022-11-08 00:00:01.
Built for Python 3.10 on linux (64-bit).

Additional context (optional)

Add any other context about the problem here.
blueprint (1).pdf

[JorjMcKie]

1. In all generality, this problem is no bug, but a file integrity error.
2. I cannot reproduce the problem with your script and file - neither in Windows nor in Linux.
3. I don't understand, why you open the file multiple times. Why don't do hand the document to do_some_analysis instead of the file name? IAW I would suggest doing this:

import fitz

FILENAME = "blueprint.1.pdf"


def do_some_analysis(doc, pageno):
    page = doc[pageno - 1]


doc = fitz.open(FILENAME)

for i in range(doc.page_count):
    page = doc[i]
    print(i)
    do_some_analysis(doc, i + 1)

[hellohr11]

Thank you @JorjMcKie ,

This script is only an example. We are writing a software analysis pdf files. I passed pdf name and page number to each module. Is there any other way to fix it?

Thanks!

[JorjMcKie]

As long as I don't have a reproducing file, I cannot say much I am afraid.
The file you gave me is ok at least, and there is no difference beween using its relative or absolute name.
Very probably a file error: the message says that a cyclic reference between objects in the page tree occurs.

This script is only an example. We are writing a software analysis pdf files. I passed pdf name and page number to each module. Is there any other way to fix it?

I understand that that code is more symbolic than anything else, that's fine.

[JorjMcKie]

Let's convert this to a Discussion under the hypothesis there is no bug but only file errors.
If you can provide a PDF that reproduces an error, we can revert that decision or open another bug issue as required.