set_page_label fails due to page_labels missing /Nums key

[evilpythonist]

While trying to relabel pages in a pdf so that they agree with its contents I encountered a KeyError (see code and traceback below). I found a workaround (also see below).

I didn't file a bug report because I can't provide the pdf file. Still I wanted to share my workaround in case someone else encounters this issue.

Environment

$ python -m platform
Windows-10-10.0.19045-SP0

$ python --version
Python 3.13.5                   

$ python -c "import pypdf;print(pypdf.__version__)"    
5.8.0

Code

from pypdf import PdfReader, PdfWriter

filename = 'filename'

reader = PdfReader(f'{filename}.pdf')
writer = PdfWriter()

writer.clone_document_from_reader(reader)

beg = 0
mid = 8
end = writer.get_num_pages()

writer.set_page_label(page_index_from=beg, page_index_to=mid - 1, style='/r', prefix='', start=2)
writer.set_page_label(page_index_from=mid, page_index_to=end - 1, style='/D', prefix='', start=1)

writer.write(f'{filename}-new.pdf')

writer.close()
reader.close()

Traceback

This is the complete traceback I see:

Traceback (most recent call last):
  File "$\Documents\renum.py", line 14, in <module>
    writer.set_page_label(page_index_from=beg, page_index_to=mid - 1, style='/r', prefix='', start=2)
    ~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "$\AppData\Local\Programs\Python\Python313\Lib\site-packages\pypdf\_writer.py", line 3379, in set_page_label
    self._set_page_label(page_index_from, page_index_to, style, prefix, start)
    ~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "$\AppData\Local\Programs\Python\Python313\Lib\site-packages\pypdf\_writer.py", line 3435, in _set_page_label
    nums = cast(ArrayObject, page_labels[NameObject("/Nums")])
                             ~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^
  File "$\AppData\Local\Programs\Python\Python313\Lib\site-packages\pypdf\generic\_data_structures.py", line 477, in __getitem__
    return dict.__getitem__(self, key).get_object()
           ~~~~~~~~~~~~~~~~^^^^^^^^^^^
KeyError: '/Nums'

Workaround

I got it to work by replacing

writer.clone_document_from_reader(reader)

with

for page in reader.pages:
    writer.add_page(page)