[Tugindia] Devnag and unicode

[Duvvuri Venu Gopal]

On Tue, 02 Mar 2004 Anna Choma wrote :
>On 29 Feb 2004 04:25:39 -0000
>"Duvvuri Venu Gopal" <venugopal_duvvuri at rediffmail.com> wrote:
>
> > The method adopted for Hindi - Omega is
> >
> > 1) Translation of Velthuse's scheme to Unicode
> > 2) Translation from Unicode to Velthuse font (Devanagari)
> >
> > If you have a Unicode compatible editor the first step is not required.
> > You can directly go to the second step. So you have to disable the OTP's
> > that convert transliteration to Unicode.
> >
> > In typing unicode it is simple. For full details see the uni2 - OTP's.
> >
> > To get compound letter you have to type Character + halant + character.
> > Then you will get the required character, i.e., to type "kya" you have to
> > type " ka + halant + ya + A". ka means the ka key not k + a of english
> > keyboard and ya means the ya key not y + a of english keyboard.
> >
> > I am not using Yudit. So I am not sure. But it should work. You can
> > disable the Velthu2uni.OTP in odev.sty file.
> >
> > D. Venu Gopal
> >
> > P.A. Can you post me your experience after above suggestions?
> >
> >
>
>Thank you very much.
>
>I tried to disable Velthu2uni.OTP in odev.sty file (comment this line 3
>times), but it changed nothing. But I managed to process file and get
>letters correctly written by putting this sequence after \begin{document}:
>
>\ocp\TexUTF=inutf8
>         \InputTranslation currentfile \TexUTF
>
>Then I saw ligatures and the letters I wrote in Yudit. Excellent!
>On the other hand I couldn't write in polish, russian and arabic properly.
>To do this these lines are nessesary:
>
>\ocp\OCPutf=inutf8
>\ocp\OCParab=uni2cuni % for arab
>\ocp\OCParabout=cuni2oar % for arab
>\ocplist\OCPlistutf=\addbeforeocplist 1000 \OCPutf\nullocplist
>\ocplist\OCPArablistutf=\addbeforeocplist 1000 \OCPutf
>               \addbeforeocplist 1000 \OCParab % for arabic
>               \addbeforeocplist 1000 \OCParabout\nullocplist % for arabic
>\pushocplist\OCPlistutf
>\font\OMLGC=omlgc
>\font\OMARAB=omarab % for arabic
>\OMLGC
>
>\newenvironment{newarab}{% for arabic
>\textdir TRT\pardir TRT\OMARAB
>\pushocplist\OCPArablistutf}{}
>
>Lines nessesery for arabic turned off my hindi and there ware to many
>letters again. Then I carried these lines to {newarab} definition and it
>was all right again. Strange also, that commenting
>\ocplist\OCPlistutf=\addbeforeocplist 1000 \OCPutf\nullocplist
>line for example causes bad processing of devanagari too.
>But, after all, I did it, entirely by accident.  One should redefine the
>\hindi command to
>
>\def\Hindi{
> 	\ocp\TexUTF=inutf8
>         \InputTranslation currentfile \TexUTF
> 	\hindi}
>
>and then the text looks fine.
>The only problem to solve is that Yudit uses old vertical ligatures (kka
>for example), which are not popular in modern hindi. It can be switch on or
>switch off when writing in Velthuis transliteration scheme, but in Yudit?
>I will try to change this.
>
>A, by the way, what is "halant"? In Yudit I write ka (key k, key a) and
>get ka letter. When I want long kaa I write kA (key k, key Shift+a). The
>same for ki, ku... The ligatures generate for themselves. It's very
>convenient (when you want these ligatures offered by Yudit).
>
>Best regards,
>
>Anna Choma
>

Thanks for posting your experience.

In inscript (developed by CDAC, Pune, India) what I do for getting ligatures or compound letters is -

for kya --> k + halant + y (halant = \char94 of Velthuis, it is used in all Indian scripts to get Sanskrit nasal, for more details visit the unicode website and see the Devnagari codes).

If you can kindly write your experiences in an article form and send to the tug-India. The editor may publish in the forthcoming edition.

Venu Gopal