freebangfont-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Freebangfont-devel] RFC: design of bengali opentype font


From: Deepayan Sarkar
Subject: Re: [Freebangfont-devel] RFC: design of bengali opentype font
Date: Mon, 31 Mar 2003 12:12:15 -0600
User-agent: KMail/1.5.1

On Monday 31 March 2003 11:40 am, Kaushik Ghose wrote:
> Hi,
> If these cases raise the size of the font a lot, perhaps use the half
> forms for stuff not found in grammar books. (Do I understand this right,
> half forms will be used by the font when no actual glyph is found to
> "compose" the cahracter..)

Right. The rule (in VOLT terminology) is supposed to be like this (somewhat 
simplified) -- 

Let k1, k2, etc denote consonants. Then

k1 + hasanta + k2 + hasanta + k3

becomes 

half-k1 + half-k2 + k3

This is done by the layout engine (not the font) (this of course needs the 
half-forms to be in the font. Otherwise the full form + hasanta is used). The 
ligature substitution rules that actually specify the conjuncts are supposed 
to be defined in terms of the half-forms. e.g.,

half-Na + Da = N_Da

Anything that's not susbstituted is obviously left as it is.

For devanagari, in particular, this is very useful, since it has very few real 
conjunct glyphs.


> Also is the explicit representation of jawphola necessary - can you fit a
> rule that will add on the jawphola symbol barring the exception of
> ontestojaw-y-reph ?

Yes, that's why I used a different indentation for those. These were in the 
juktolist.txt file, but don't need separate glyphs. (There are some 
exceptions, like n+yaphala, h+yaphala, which often have separate glyphs, so 
I'll include them).

I don't know if the r+ya thing is fixable, I will have to play around in VOLT 
to find out. There's still a lot to do with the font before I start with 
that.

Deepayan




reply via email to

[Prev in Thread] Current Thread [Next in Thread]