Stacia Hartleben wrote:
> I was looking at a previous mail that said you could edit a file to
> put in Tipa codes...this got me thinking that perhaps I could do the
> same for the Devanagari package.
>
> However IPA and Devanagari are reallly different. A big problem
> especially is the "i" vowel which appears *before* the characters but
> in unicode is actually encoded afterwards. Using the devanagari
> package, in shorthand "ki" would be {\dn Ek}. Uh oh. Also, one of the
> vowels ("u") is encoded like this for "ku" - {\dn \7{k}} Not sure how
> this would work, or if any of it would work because the vowel and the
> consonant have to go together to create a proper symbol - for example,
> {\dn E} {\dn k} creates a space between the two characters that
> shouldn't be there!
>
> I'd be willing to update the unicode file if someone gave me tips on
> how to make the Devanagari package work - website is here:
> http://sarovar.org/projects/devnag/
The unicode file can only handle simple replacements: There needs to be a
1:1 relationship between the UCS4 code point and the LaTeX command. It does
also work for characters that are marked as "combining" in unicode (e.g.
0x0303 COMBINING TILDE), they can be combined with any base character.
Everything that is more complicated does not work.
> If we could get this working, it would be really neat, since it would
> effectively get around the need for the preprocessor. My Sanskrit
> teacher told me that I make a lot of mistakes in my assignments when I
> use LaTeX no doubt due to the preprocessor making it difficult to
> catch my mistakes.
>From what you describe it looks like the unicode file will not work for
Devanagari. More sophisticated support in the LyX sources is probably
needed.
Georg
