The matching author function author_match does a double test instead of using the match result.

The current author matching is

pattern in name

This leads to many false positive and should be replaced.

Affects:

• LatexDocument

• and common tools

Sometimes macros are not in math mode. They are not parsed in the markdown extraction

To specify which figures you want to appear alongside your paper, one can leave a comment in any .tex file before uploading to arXiv, as in this example:

%@arxiver{fig1.pdf,fig4.png,fig15.eps}

It seems that for some papers (e.g., 2303.02816 the parsing of bib does not work.

---------------------------------------------------------------------------
IndexError                                Traceback (most recent call last)
Cell In[13], line 6
      4 except Exception as e:
      5     print(e)
----> 6     raise e

Cell In[13], line 2
      1 try:
----> 2     bibdata = latex_bib.LatexBib.from_doc(doc)
      3     full_md = latex_bib.replace_citations(full_md, bibdata)
      4 except Exception as e:

File ~/Work/arxiv_display/venv/lib/python3.11/site-packages/arxiv_on_deck_2/latex_bib.py:144, in LatexBib.from_doc(cls, doc)
    141         bib_database = parse_file(bibtex_file)
    142     bib_data.append(bib_database)
--> 144 b0 = bib_data[0]
    145 for bdata in bib_data[1:]:
    146     b0.add_entries(bdata.entries)

IndexError: list index out of range

Not all accentuated letters are correctly cleaned up from the tex file

e.g. Jes'us Vega-Ferrero

date = soup.find_all('div', {'class': 'list-dateline'})[0].text.replace('\n', '').split(',')[-1].strip()

This div has been removed for an h3 text.

Need to make this more robust to errors.

• Title, see 2302.08736
  \title{CRIRES$^+$ detection of CO emissions lines and temperature inversions on the dayside of WASP-18b and ...}

• this paper would not be selected anymore as a paper candidate after correcting for #8. No MPIA author (code would not even check the affiliations).
• Patch needed on the LatexDocument.get_title() .
  • the content.find_all() breaks the title into smaller bits because of the math mode.

title = '"'.join([str(k) for k in doc.content.find_all('title')[0].contents])

• subtitle, see 2302.07916

Example paper: 2302.10943

\title[Brightest cluster galaxies in IllustrisTNG]{The growth of brightest cluster galaxies in the TNG300 simulation: dissecting the contributions from mergers and \textit{in situ} star formation} 

renders as

Brightest cluster galaxies in IllustrisTNGThe growth of brightest cluster galaxies in the TNG300 simulation: dissecting the contributions from mergers and \textit{in situ} star formation