Re: Editing and formatting characters

[Pablo Saratxaga <pablo mandrakesoft com>]

Kaixo!

On Tue, Nov 14, 2000 at 07:56:55PM -0500, Owen Taylor wrote:
 
> So, I do like the idea that insertions happen before the
> paired-formatting character.

IMHO it should be:
Before the *closing* paired one, and before the *opening* closed one.
That is, adding text should get the poperty of previous text (doesn't
that already work like that for things like bold, font face, etc attributes?
I think the case of paired formatters fits perfectly and could be handled
with the same code)

> Does this mean that they also should
> happen before non-paired formatting characters? I'm not sure.

Imho: only for LRM and RLM, if adding neutral chars. Eg:

HEBREW TEXT [HYPHEN] y [RLM] x
(x and y being postions; the cursor being at x (and could not be at y))

adding a neutral char (eg, another hyphen) should be done at y position;

> The argument for is that in the case of:
>  
>   HEBREW TEXT [HYPHEN] [RLM] 1234
> 
> The user probably thinks of this as a region of strong-right-to-left
> extending from the hebrew text past the HYPHEN, so another neutral
> character inserted after the HYPHEN would naturally extend this region.

Yes, I fully agree.

> Tenatively, I think you've convinced me that invisible formatting
> characters should be associated with the character after.

In most cases, yes. but not always.
consider:

<arabic letter> <zwj> <non arabic char>

About your example; I think the RLM must be associated to the hyphen,
not the first digit, because it modifis the behaviour of the hyphen, not of
the digit (the hyphen has its behaviour depending on the following (and
preceding too?) char, the role of RLM is to isolate from the hard LTR 
char. I wonder if the use of ZWNBS or other ZWsomething wouldn't have
a similar effect btw.

In your example it must be associated with the hyphen, otherwise if the
user goes between the hyphen and the digit and inserts another digit the
display will change and the hyphen behaviour too; that is counter-intuitive.

The rule must be: attach to the char that is modofied; if both sides are
and in a same "intensity", choose one side (attaching to next one is my
preferred choice in this case)

-- 
Ki ça vos våye bén,
Pablo Saratxaga

http://www.srtxg.easynet.be/		PGP Key available, key ID: 0x8F0E4975