emacs-bidi
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [emacs-bidi] Points composition - status.


From: Yair Friedman (Jerusalem)
Subject: Re: [emacs-bidi] Points composition - status.
Date: Mon, 24 Dec 2001 08:51:11 +0200
User-agent: Gnus/5.090004 (Oort Gnus v0.04) Emacs/21.1

Matan Ninio <address@hidden> writes:

> Eli Zaretskii <address@hidden> writes:
>
>> Yair Friedman (Jerusalem) wrote:
>> > 
>> > Eli Zaretskii <address@hidden> writes:
>> > 
>> > >> From: "Yair Friedman (Jerusalem)" <address@hidden>
>> > >> Date: Thu, 20 Dec 2001 18:25:05 +0200
>> > >>
>> > Yes, but composed characters don't work on character-based terminal in
>> > the first place...
>> 
>> How do you mean ``don't work''?  You should see the first character of the
>> composition, IIRC.  If that doesn't work, perhaps there's some bug.

After more careful testing, I found that you're right, although I'm not
sure this is the correct thing to do. I Don't see anywhere that the user
is signaled about the missing compositions.

>> > It seems that forward/backward commands are working per-base
>> > character, while delete is done per diacritical.
>> 
>> That's possible, but I think having a command that deletes the base
>> character together with its diacriticals is better.  Perhaps Thai and
>> Tibetan users will like it too ;-)  It's even possible that there is such a
>> command already.
>
> What about a toggle-able composite-character behavior (delete/move by
> base characters) and segment based behavior?  we may even want to allow
> decomposition of the buffer for complicated cases.

Yes, this is what I thought would be the best. In case we require
normalization of the diacriticals, then a visible
decompose-edit-normalize-compose sequence make sense.

> btw, have any of you considered that diacriticals may appear on
> non-strong-R2L  base chars, such as " ", "_" , "*" and even English
> letters (Thing of a "fill the letters" quiz, for an easy example.)

Not, although this might be the easier solution. I tend into a more
strict model - diacriticals only above letters, restricting several
types (shin/sin dots, rafe) into letters, no more than one vowel, no
more than two cantillation.

In case that those rules aren't kept, the diacriticals are not
automatically composed - I'm not restricting input of, such
combinations, they just not composed.

In case of a special need, a special code should be written.
  




reply via email to

[Prev in Thread] Current Thread [Next in Thread]