Re: A little polish of Thai shaper

[Theppitak Karoonboonyanan <thep links nectec or th>]

On Fri, Jun 15, 2001 at 02:58:37PM -0400, Owen Taylor wrote:
> 
> Theppitak Karoonboonyanan <thep links nectec or th> writes:
> 
> > To my understanding: In the Unicode book, most combining characters
> > referred to are put after base character. So, I think it would be
> > natural for choice 1), as the normal advancing method will simply yield
> > the desired result.
> 
> I'm basically fairly skeptical of assuming that you can get things
> right by simply rendering Unicode code points in order as if they were
> roman characters.

It provides a graceful fallback in absence of rendering engine.
Even monospace rendering in the pre-Unicode era gives a tolerable
output with this strategy. So, it becomes somewhat de facto. :-)

However, I think what is yet missing anyway is the special treatment
of X "charcell" fonts to make Thai shaper really complete.

I would like to add that code, but how could I distinguish charcell,
monospace, and proportional fonts to one another in Pango? Or could
I assume that Pango always uses proportional fonts?

> It bothers me a little that people running XFree86 will see the Thai
> parts of the Pango examples rendered wrong until a fixed version is
> widespread.
> 
> Perhaps it might make sense to submit a patch directly to the XFree86
> people to speed up the process of getting a version with correct Thai
> glyphs.

All right. I could try to submit the fix to XFree86 for the time being,
expecting that it would be overridden by the new version from Mark Leisher.

In addition, to make Pango shape elegantly with Thai Windows fonts,
tis620-2.enc is also needed in XFree86. I'll try to submit that, too.

Regards,
-Theppitak.
-- 
Theppitak Karoonboonyanan
Information Research and Development Division, NECTEC
(Was: Software and Language Engineering Laboratory)
http://www.links.nectec.or.th/~thep/  mailto:theppitak nectec or th