Re: Tibetan stacks

From: Martin J. Duerst (mduerst@ifi.unizh.ch)
Date: Fri Jan 24 1997 - 06:52:37 EST


On Thu, 23 Jan 1997 Ken Whistler wrote:

> > <A><spacing up><two dots><spacing down>
>
> This sounds like an application for the general concept
> of kerning pairs (which should have complete freedom of
> relative placement in two dimensions, if the font models
> are properly implemented), rather than something which
> calls for some arbitrary control codes/glyphs.
>
> "<spacing up>" begs the question: up by how much? Why not
> use the generic coordinate mechanisms already available
> in font technology?

If you have kerning pairs available, of course you can try to
use them. But not all font technologies have them available.
BDF fonts on X11 don't. And I am not sure that all others
allow kerning in two dimensions.

Also, pair kerning inserts displacements between pairs
of letters, and thus
        <A>[pair kern]<two dots>
is equivalent to
        <A><spacing up><two dots>
but what we need is
        <A><spacing up><two dots><spacing down>
Otherwise, the rest of the text may just be positioned
too high.

The question of "how much" is of course, in the example
I made, "as much as necessary to position a diacritic
atop an upper-case letter". For other purposes, there
might be other <spacing up> glyphs.

Regards, Martin.



This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:33 EDT