Re: [emacs-bidi] Input method with diacritics.

[Yair Friedman (Jerusalem)]

Eli Zaretskii <address@hidden> writes:

> On Thu, 22 Nov 2001, Yair Friedman (Jerusalem) wrote:
>
>> What about CAPS?
>
> Only if we have to: shifted letters still produce Latin caps in the 
> Hebrew input method, and I don't think we should break this unless we 
> have no other choice.

Stupid SI 1452 put the Hebrew diacritics in the CAPS-Shift map - to
produce all the characters I put in the '[' map, you need to press CAPS,
and then shift-letter e.g. CAPS-Shift-8 produce qamats.

Stupid as it is, I'd like to see if I can implement this standard in
addition to hebrew-acc-prefix.

> But using those modifiers was just an idea; it's not the only one.  For 
> example, we could define one of the Fn function keys as a prefix key, so 
> that, say, "F8 i" would produce Hiriq, etc.  (Doesn't MS Word use 
> something like that?)

This is why I defined the prefix method, It uses '[' and ']' as map
change keys.
> The problem with these is that they Shift and Alt are usually taken (Alt 
> produces META, a you know), and AltGr might be unavailable (you get a 
> usual Alt instead).

I have a feeling that Emacs knows inside the real key that was hit - it
know if the left or right (Alt/Ctrl/Shift) keys were hit, it knows the
difference between 1 and KP-1, it just has the default binding of the
keys to the same "logical" key.  What I'd like to do is catch these
differences only when the input method is active, and use them - The
left Alt will stay left while the right one (Sometimes labeled AltGr),
used differently, and also CAPS-Shift would produce different results.

Of course on keyboard that don't have these keys the prefix method is
available as well. and on any case none of the input methods for Hebrew
with diacritics should be the default for Hebrew.

I hope I'm more clear this time.