Multiple --csl arguments give bad error message

[mgajda]

Explain the problem.

If using pandoc within a build pipeline, one may mistakenly give two --csl arguments:

pandoc -s input.md --csl style1.csl --csl style2.csl -o output.pdf 

This will give a spurious error message when pandoc tries to find style1.csl style2.csl which is not a file.

The issue stems from the fact that Text.Pandoc.App.CommandLineOptions.addMeta tries to append data to the previous meta.
Unfortunately, --csl option does not treat the input as a list of styles, instead merges it.

Would it be more convenient to use separate replaceMeta (possibly with a warning when previous metadata was non-empty), or change parsing of meta only for CSL?

Currently Text.Pandoc.Citeproc.getStyle simply uses metaValueToText before trying to read it as a full path.

Pandoc version?

$ pandoc --version
pandoc 2.14.0.3
Compiled with pandoc-types 1.22, texmath 0.12.3.1, skylighting 0.10.5.2,
citeproc 0.4.0.1, ipynb 0.1.0.1
User data directory: /home/m/.local/share/pandoc
Copyright (C) 2006-2021 John MacFarlane. Web:  https://pandoc.org
This is free software; see the source for copying conditions. There is no
warranty, not even for merchantability or fitness for a particular purpose.

This is Linux version.

[mgajda]

I described the source of the error is appending to meta, and then using meta as a single file path.

[jgm]

Would it be more convenient to use separate replaceMeta (possibly with a warning when previous metadata was non-empty), or change parsing of meta only for CSL?

Yes, I think that's a good idea. If you specify two --csl options, one should be ignored, with a warning.
Is this something you want to do a PR for?

[pt2121]

Can I take a stab at this?

[jgm]

Go for it!

[pt2121]

I am looking into this and wondering if we should

• let the user run pandoc with a warning,
• or not run it and show an error message.

Any opinions on this?