Since you asked for advice, I advise you to first try to
handle properly written vanilla LaTeX.
That will require deciding what "properly written" LaTeX might
look like.
If you manage to get something useful, then you can think
about whether your next step is to try to handle the quirky
LaTeX people write, or to think about handling LaTeX packages.
Almost surely you will NOT want to go the LaTeX package route.
Many of those concern the appearance of the document. But PreTeXt
source knows nothing about appearances, so it is not clear to me
what you would hope to do with most packages.
My script is not private:
https://github.com/davidfarmer/SL3X
I still offer the service of converting to LaTeX, with a preference
for large documents such as textbooks. Usually people send me a link
to their GitHub repository.
Happy to discuss any ideas here.
More advice: experiment to see if the Pandoc conversion is any good.
If it is, then can't you accomplish what you want by writing python
that handles the file management, calling pandoc, etc?
I'm interested in hearing your evaluation of the Pandoc conversion,
if you decide to go that route.
Regards,
David
> --
> You received this message because you are subscribed to the Google Groups "PreTeXt development"
> group.
> To unsubscribe from this group and stop receiving emails from it, send an email to
>
pretext-dev...@googlegroups.com.
> To view this discussion visithttps://
groups.google.com/d/msgid/pretext-dev/0e55d313-9417-4de8-801b-05166bb3847dn%40googlegroups.
> com.
>
>