getOutlines() returns repeated items

[aalvrz]

When obtaining the outline of a PDF file:

with open(file_path, 'rb') as pdf_file:
    pdf_reader = PyPDF2.PdfFileReader(pdf_file)
    outlines = pdf_reader.getOutlines()

The method returns repeated items in the list. For example:

[
{'/Title': 'Working with a virtual environment', '/Page': IndirectObject(327, 0), '/Type': '/XYZ', '/Left': <PyPDF2.generic.NullObject object at 0x7f06aba3f278>, '/Top': <PyPDF2.generic.NullObject object at 0x7f06aba3f2b0>, '/Zoom': <PyPDF2.generic.NullObject object at 0x7f06aba3f2e8>}, 

{'/Title': 'Working with a virtual environment', '/Page': IndirectObject(327, 0), '/Type': '/XYZ', '/Left': <PyPDF2.generic.NullObject object at 0x7f06aba3f278>, '/Top': <PyPDF2.generic.NullObject object at 0x7f06aba3f2b0>, '/Zoom': <PyPDF2.generic.NullObject object at 0x7f06aba3f2e8>}, 

# ...

]

The 1st item should actually be another one with Title of Introduction:

[jayantsolanki]

I am also facing the same issue, one of the bookmark is getting repeated, it is happening for just one pdf. Working perfectly for other pdfs.

[MartinThoma]

Could you share a pdf that causes these issues?

[mtd91429]

Without a PDF to test and compare, I can only guess. However, I believe that this issue is related to #1121 and is fixed by PR #1128

[MartinThoma]

Yes, it looks like that to me as well. Thank you for the hint 🤗

I'm closing this issue now as I assume that PyPDF2==2.6.0 has fixed this issue. Please let me know if you see it again with the latest PyPDF2 version.

[jayantsolanki]

Could you share a pdf that causes these issues?

I am very sorry, I missed this. issue happened when I was interning somewhere. I don't have access to that pdf anymore (was proprietary). #1121 seems to tackle the similar issue. By the way, hugely appreciate the development done by you guys and keeping PyPDF alive. PyPDF2 for me has been a very versatile tool. I referenced your work in one of our paper How SAS® and Python Enhance PDF