Re: Proposal: Introduction of Neutral Override Character

Dov Grobgeld wrote:
> Hi,
> 
> After reading Behdad's excellent summary of the current state of 
> BiDi in Gtk, I realized that the problem that Behdad describes 
> on Slide 11 could easily be solved by the introduction of a new
> unicode control character "Neutral Override". This new unicode
> character should have similar functionality to:
> 
>    202D;LEFT-TO-RIGHT OVERRIDE;Cf;0;LRO;;;;;N;;;;;
>    202E;RIGHT-TO-LEFT OVERRIDE;Cf;0;RLO;;;;;N;;;;;
> 
> And I suggest that for the time being we give the new character
> the definition:
> 
>    20EB;NEUTRAL OVERRIDE;Cf;0;NO;;;;;N;;;;;
>    
> The effect of the character is terminated by PDF. 

This part seems like a bad idea to me.  A NO..PDF in a text containing
the existing stateful bidi codes would be *badly* misinterpretted
(rather than just ignored) by software unaware of NO because it could
take the PDF as closing a containing explicit code, going out of sync:

   <RLE>foo <NO>quux<PDF> bar<PDF>

would be handled as if it is:

   <RLE>foo quux<PDF> bar<PDF>

It follows that introducing any new stateful bidi code must be
accompanied by a new terminating code for it.  UAX9 got away (did it?)
with the single PDF only because LRE/RLE/LRO/RLO were all introduced at
once (were they?).

(Then agian, the effect of not understanding NO might be bad enough to 
judge this a non-issue in this case; I still feel a new terminating code 
would be cleaner).

-- 
Beni Cherniavsky <[email protected]>
Note: I can only read email on week-ends...