[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-core-devel
Subject:    Re: The future of Sonnet
From:       Simon Hausmann <hausmann () kde ! org>
Date:       2007-01-02 18:11:09
Message-ID: 200701021911.13366.hausmann () kde ! org
[Download RAW message or body]


On Tuesday 02 January 2007 18:49, Jacob R Rideout wrote:
> > Qt already includes this, albeit privately. We're also working on adding
> > all character names to kdeui (because of the KCharMap class).
> >
> > What property do you need from a character that you can't find in QChar?
>
> As far as I know, Qt uses data from unicode 4. I'd like to
> incorporated some of the changes that occured in Unicode 5, but if Qt
> included the data I needed, then 4 would be ok.

Qt 4.3 uses the latest Unicode 5.0 tables.

> QChar doesn't include an ability to query which block (usually this
> means script type) a character belongs to. It also doesn't allow you
> to query some other the other data included in the technical reports
> or annexes, including tr29 which is used for the text breaks class.
> This data includes a bunch of special categorization that only is
> relevant, if used in the context of text breaking code. This is
> especially important for handling the more exotic orthographies of the
> world. I'd also like to be able to allow users or language specific
> distros to change the mappings, although this may be unnecessary
> and/or foolish.

We actually consider exposing a bit more public unicode API in Qt 4.3. Some of 
it is already visible, like title casing and basic surrogate helpers.

Simon

[Attachment #3 (application/pgp-signature)]

[prev in list] [next in list] [prev in thread] [next in thread]