Re: [emacs-bidi] status? news?

[Roozbeh Pournader]

On Thu, 18 Jan 2001, Eli Zaretskii wrote:

> I don't think it is reasonable to request that the user inserts an LRM in 
> such simple cases.

By "you", I meant the editor. I believe that the user need not to even
know that things named LRM and RLM exist.

> > The definition is short-minded, I think. I define the logical order as the
> > order of characters in the file (I know that's not exact ;)).
> 
> I believe ``logical order'' is defined (not by me) as the order in which 
> characters are typed or read.

I agree with you. But I want to tell that there are two logical orders.
The one you want it to be (order of typing and reading), and the one
Unicode specifies (order of storing in files, or interchanging over the
Internet). Unicode version is based on Microsoft's (or others') ideas,
which is short-minded in languages other than Arabic and Hebrew (it has
problems even with those two, I agree). 

You should /not/ take these two the same.

> It is possible that this will be needed in more complicated cases.  But 
> I'd expect the simple case of mixing L2R and R2L text to not require such 
> additional information, because implicit embedding can be easily deduced 
> from the characters you find in the buffer.

Yes, but here the exclamation mark problem you mentioned, arises.

> This goes without saying.  However, I'm currently concerned with the 
> default behavior being smart enough to be useful, and not by how to make 
> that customizable, because I think the default behavior should be good 
> enough for the majority of users and uses.  (The default can be language- 
> and locale-dependent, of course.)

By customizablity, I meant locale-dependecy. I agree that true
customizablity may be hard for a start.

> If the text is kept in logical order in the buffer, the display needs to 
> know about this as well.

Yes, but I'm always with the visual buffer plus additional data. If the
editor is supposed to support both logical and visual buffers, Unicode
bidi awareness in the display code is also needed, as you said.

> > But I don't know if that's possible with Emacs.
> 
> What is the aspect that you think might not be possible with Emacs?

I meant storing the additional information, like the embedding level.

--roozbeh