mojca.miklavec.lists at gmail.com
Mon Jun 16 20:32:23 CEST 2008
On Mon, Jun 16, 2008 at 7:53 PM, Norbert Preining <preining at logic.at> wrote:
> Hi Mojca, hi Karl,
> On Mo, 16 Jun 2008, Mojca Miklavec wrote:
>> language: es
>> name: spanish
>> synonym: espanol
>> lefthyphenmin: 2
>> righthyphenmin: 2
>> file: loadhyph-es.tex
> what about the following:
> - We get rid of the language.XX.dat file
> - replace the
> execute BuildLanguageDat es
> execute language=es name=spanish synonyms=espanol[,...] lefthyphenmin=2 righthyphenmin=2 file=loadhyph-es.tex
> (note that these executes could occur several times in one tlpsrc)
> Important would be that the
> tag is *unique* over *all* files.
> That way we could get rid of *loads* of language.XX.dat files and
> updating would be the usual thing, nothing to be thought twice (change
> here and there).
> If we extend this we could even do something similar with the formats
> and get rid of the fmtutil.XXX.cnf files.
> My job would be to adapt the code generating the
> language.dat/fmtutil.cnf lines, which is quite trivial, and add another
> function to generate language.def lines.
> Could all be done in probably 20min.
> Your (whoever that may be) is to fill in the stuff into the tlpsrc
To me that makes sense (less files, less mess). But note that there
are many comments under English.
Can anyone explain me when/which languages should be grouped together
apart from English?
German should probably include old and new, Greek should include all
variants, Serbian should include both scripts. What about
"Czecho-Slovak" and "Norwegian"? Anything else?
More information about the tex-live