Missing validity check in parse_pdf.py

[severin-kacianka]

Dear all,

this paper: https://arxiv.org/abs/2402.13610 seems to trigger a bug in this parse_pdf.py:L281:

File "[redacted]/src/parse_pdf.py", line 137, in <module>
    paper = parse_pdf_only_text(pdf_path)
  File "[redacted]/src/parse_pdf.py", line 88, in parse_pdf_only_text
    paper_dict = scipdf.parse_pdf_to_dict(pdf_path)
  File "[redacted]/.local/lib/python3.10/site-packages/scipdf/pdf/parse_pdf.py", line 412, in parse_pdf_to_dict
    article_dict = convert_article_soup_to_dict(parsed_article, as_list=as_list)
  File "[redacted]/.local/lib/python3.10/site-packages/scipdf/pdf/parse_pdf.py", line 368, in convert_article_soup_to_dict
    article_dict["figures"] = parse_figure_caption(article)
  File "[redacted]/.local/lib/python3.10/site-packages/scipdf/pdf/parse_pdf.py", line 281, in parse_figure_caption
    label = figure.find("label").text
AttributeError: 'NoneType' object has no attribute 'text'

My simple workaround is to just surround the assignment with a try/catch block like this:

try:
    label = figure.find("label").text
except:
    label = ''

I assume this might mask some underlying issues, so I did not submit a pull request; if I have time I might dig deeper in a few days.
Please let me know if I can be of any help or provide further debug output.

Cheers,
Severin