I just saw that pandoc.write has been implemented in Lua, so sorry for the noise on that score. An extension for simple tables in JSON would still be appreciated, provided it is easy to implement.

Den tors 13 jan. 2022 18:18BPJ <melroch@gmail.com> skrev:


Den tors 13 jan. 2022 10:27Bastien DUMONT <bastien.dumont@posteo.net> skrev:
I guess that it is can now be achieved using pandoc.write:

  * Create a new Pandoc object containing your Div;
  * Pass it to pandoc.write;

AFAIK there is no pandoc.write available in Lua filters. Might that change? I for one could find use for it, e.g. to have content other than just tags/commands be different between HTML and LaTeX one might produce a Markdown version with some content converted to each format in two raw elements and do some manual editing of each. 

In Perl JSON filters it used to be pretty easy to shell out to pandoc to do such things but sadly Pandoc::Elements hasn't been updated for the new table model. I'm not whining since I could and should have contributed to do something about it. Still a pandoc extension which would write/read simple tables in JSON would be welcome, assuming that the code used in the Lua module to convert between table models could be reused. I really have problems conceptualizing the new table module, *especially* as represented in JSON, sadly.

Would it be worth the while to file two issues for this, i.e. is there realistic hope that either of them will be implemented?


  * Modify the returned string;
  * Return it as the content of a RawBlock.

By the way, this could also solve the problem discuss here, but I would like to test it before closing the issue: https://github.com/jgm/pandoc/issues/7262

Le Thursday 13 January 2022 à 05:37:04AM, denis.maier@unibe.ch a écrit :
> Well, the invalid file will only be temporary.
> Let's say I a div with content-type="parallel" for parallel texts. In some cases the one column or both will have righttoleft text, so I need for multiple attributes:
>
> content-type="parallel-left-rtl"
> content-type="parallel-right-rtl"
> content-type="parallel-both-rtl"
>
> But this will cause redundant rules in the xslt for post-processing:
>
> I could do this:
>
> content-type="parallel right-rtl"
> content-type="parallel left-rtl"
> content-type="parallel both-rtl"
>
> Also not ideal (because processing such lists is ugly in XSLT). I thought about this:
>
>
> content-type="parallel" right-dir="rtl"
> content-type="parallel" left-dir="rtl"
> content-type="parallel" left-dir="rtl" right-dir="rtl"
>
> I would then pick this up via xslt and construct valid colgroups based on the values of right-dir and left-dir. But maybe there's an easier way?
>
> Denis
>
> > -----Ursprüngliche Nachricht-----
> > Von: John MacFarlane <jgm-TVLZxgkOlNX2fBVCVOL8/A@public.gmane.org>
> > Gesendet: Donnerstag, 13. Januar 2022 01:50
> > An: Maier, Denis Christian (UB) <denis.maier@unibe.ch>; pandoc-
> > discuss-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
> > Betreff: Re: Force use of attributes in jats writer
> >
> >
> > I don't understand.  You want to create an invalid JATS document, with an
> > attribute that is not supported for that element?
> >
> > <denis.maier-NSENcxR/0n0@public.gmane.org> writes:
> >
> > > Hi
> > >
> > > Is there a way to < convince > pandoc to use an attribute on a div that is not
> > in the list for boxed-text in the list of allowed attributes in the jats writer?
> > > Background: I want to use some non-standard attributes and pick them up
> > later during post-processing. (The intermediary jats file will not be valid, the
> > final result will be.) I can add that stuff to one of the allowed attributes, but
> > that will make post-processing with xslt more complex.
> > >
> > > Best,
> > > Denis
> > >
> > > --
> > > You received this message because you are subscribed to the Google
> > Groups "pandoc-discuss" group.
> > > To unsubscribe from this group and stop receiving emails from it, send an
> > email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org.
> > > To view this discussion on the web visit
> > https://groups.google.com/d/msgid/pandoc-
> > discuss/b4b8786382dd497bb7bb231edcc5f93a%40unibe.ch.
>
> --
> You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org.
> To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/b68835526c9e4d45afb6d3d122b7da91%40unibe.ch.

--
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/Yd/vtdosfda4eefI%40localhost.

--
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org.
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/CADAJKhAKbdJa57TC8_PNfb7cK4BOPBwSr81XydHrazE5WhjOCg%40mail.gmail.com.