Re: [emacs-bidi] adding LRM in visual-to-logical transform

[Behdad Esfahbod]

On Fri, 16 Nov 2001, Alex Schroeder wrote:

> Alex Schroeder <address@hidden> writes:
> 
> > "Ehud Karni" <address@hidden> writes:
> >
> >> At least minimal optimization of the defining the R strings should be
> >> done, otherwise you'll have "^ ^" between every Hebrew words. The
> >> minimum is to include spaces between R words. It can be extended to
> >> include all characters that can not be resolved (in any case) into L
> >> type.
> >
> > Hm yes, I'm currently wondering wether we ought to look at R strings
> > before or after resolving weak types.  I'll need to do some more
> > thinking.  :)
> 
> While I'm writing such a thing (adding LRM characters to the string),
> I will need a function to tell me wether two strings containing
> logical representations are "equal" in order to write my test cases --
> for a meaningful definition of "equal".
> 
> I read UAX#9 as saying that two strings should be equal when the
> logical representation has been transformed into the visual
> representation and all LRM RLM RLO LRO and PDF characters have been
> removed.  Correct?

No, in my opinion two strings are equal if the logical representation 
is equal when directional marks removed. The definition of UAX#9 says 
that a visual representation is conformant if and only if when 
removing directional marks, the outputs is the same.
 
> Will such a function be usefull for Emacs as well?
> 
> Alex.
> 

-- 
Behdad
29 Aban 1380, 2001 Nov 20

[Finger for Geek Code]