Re: [Gnumed-devel] xelatex/ xetex

[Karsten Hilbert]

On Tue, Aug 21, 2012 at 08:00:51PM +0200, Hilbert, Sebastian wrote:

> > - undo the renaming
> > 
> > - find client/business/gmForms.py
> > 
> > - replace "pdflatex(.exe)" with "xelatex(.exe"
> > 
> > I don't know xelatex options, though.
> > 
> This is what I was after.

> As soon as you can confirm that the usual suspects in
> GNUmed work as expected with xelatex

I don't expect they will because xelatex does not want

        \usepackage[utf8x]{inputenc}

which pdlatex does and we do, too, (as long as we use
pdflatex, so it's a kind of Catch 22).

> I will do the change
> for you and ship an updated frozen GNUmed.

You can but you'll also have to test with adjusted templates.

> GNUmed will produce the tex file when trying to compile
> it. So you can potentially find the temporary tex file and
> compile it with xetex manually to verify everything works as
> expected.

That's an interesting test, too: remove the {inputenc} line,
insert some Hindi or Gujarati and recompile with xelatex.

The problem will be that one will need to insert specific
fonts in order to be able to output certain scripts. Which
does not help GNUmed because GNUmed will not know which
scripts are going to be needed ...

So maybe xelatex does make it easier to get documents
written in known scripts but it does not make it easier to
produce scripts in any variety of combinations (which is
what we ultimately want) -- unless one has got a font which
covers the entire unicode surface.

Karsten
-- 
GPG key ID E4071346 @ gpg-keyserver.de
E167 67FD A291 2BEA 73BD  4537 78B9 A9F9 E407 1346