documents4j VS pandoc

Take a look at https://documents4j.com. I'm not sure if it checks all your boxes. I have never set it up my self, but know it to be reliable in a production environment.

On the technically advanced end of the spectrum you'll find John MacFarlane [1], professor of philosophy at Berkeley and creator of pandoc [2]. Some people are just amazing.

[1] https://johnmacfarlane.net/

[2] https://pandoc.org/

Something that is quite suited to little mini projects with python is writing pandoc filters. The basic idea of pandoc is to convert written documents of some file type (the main source type is typically markdown) into another document type, for example latex, pdf, docx, odt... Now of course there will be certain things that pandoc cannot do quite well enough given its versatility. So it gives users the option to write [simple filters](https://pandoc.org/filters.html#summary) and apply them. This way you can also come up with your own syntax/tags and whatever in markdown and convert that to an arbitrary formatting instruction in another document type. Look into something like panflute and maybe you can come up with your own filters.

Pandoc generally works just fine: https://pandoc.org/

Similar to for instance shellcheck to check the syntax of shell scripts, is there an equivalent for the set of roff commands typically used in a (Linux) man page? I'm aware that e.g. pandoc permits the conversion of an other format (e.g., org) to both roff man and roff ms.

pandoc is a format converter, not a format. You still need to write the document. That said, I would rather choose something pandoc supports as an input format than asciidoc, which is a constantly moving target that still doesn't have a specification, is governed by a working group that seemingly never makes decisions, and in practice asciidoctor may as well be the specification.

As an addendum to yesterday's discussion of software estimates, one of the more amusing open source versions is this guy saying in July of 2014 he expected pandoc's asciidoc frontend to be done in a "few months": https://github.com/jgm/pandoc/issues/1456#issuecomment-50805...

You could do it with Pandoc (https://github.com/jgm/pandoc), it supports both Latex and Typst as input and output formats, and can use either as a PDF engine.

> I'm pretty sure the author would say that what you end up with is a KeenWrite document.

Documents written in KeenWrite can be typeset using tools that already exist:

* https://yihui.org/knitr/

* https://pandoc.org/

* https://github.com/jfisteus/html2xhtml

I've kept compatibility with pandoc (CommonMark) and knitr a priority; I wrote KeenWrite to replace a shell script that was calling out to those software packages. My Typesetting Markdown blog posts dive into the script and programs:

https://dave.autonoma.ca/blog/2019/05/22/typesetting-markdow...