Re: Unicode support

[Eli Zaretskii]

On Wed, 25 Jul 2001, Roozbeh Pournader wrote:

> Yes, we certainly need an *incremental* bidi algorithm, equivalent
> to the original one. But this is exactly what I call academic
> research.

I gave up waiting for this to materialize, and decided to roll my own,
highly non-academic, extremely amateurish incremental algorithm.  I 
simply could no longer explain to myself why Emacs still doesn't support 
bidi, more than 2 years after we started seriously discussing that.

I have a prototype code that is almost finished, and plan to begin
testing it in a few weeks.  It's a stand-alone implementation for now,
but it's written with Emacs as its primary client, and should plug in 
neatly into the code that generates glyph matrices from buffer text.

>       http://www.cs.fit.edu/~satkin/docs/bidi.pdf

Thanks for the pointer, I will look that up.