ASU keyboards, again
Carsten Haitzler (The Rasterman)
raster at openmoko.org
Wed Aug 27 16:15:47 CEST 2008
On Wed, 27 Aug 2008 16:05:11 +0200 "Marco Trevisan (Treviño)" <mail at 3v1n0.net>
> Carsten Haitzler (The Rasterman) wrote:
> > On Wed, 27 Aug 2008 12:56:06 +0200 julien cubizolles <j.cubizolles at free.fr>
> > babbled:
> >> Le mercredi 27 août 2008 à 09:55 +1000, Carsten Haitzler a écrit :
> >>> On Tue, 26 Aug 2008 15:31:45 +0200 David Samblas
> >>> <dsamblasomcommunity at gmail.com> babbled:
> >>>> I have followed th Raster indications to install spanish dictionary and
> >>>> it works :)
> >>> fantastic! i never tested. including accented characters? like ñ etc.?
> >> The accents don't work for me with a french dict : the word "abcès" is
> >> in my dictionnary but is never proposed in the list.
> > do you type a "è" on the keyboard? on an "e" (yes - you need to go to
> > numbers to fine an "è") ? is your dictionary utf8?
> Ah, another request: what about considering "e" like "è", "é" and other
> special chars? It would improve a lot writing!
i know. already accounted for it. from illume's code (e_kbd_dict.c right at the
top) ... right now "normalising" is:
// FIXME: ö -> o, ä -> a, Ó -> o etc. - ie normalise to latin-1
return tolower(glyph) & 0x7f;
not great... but hey... better than a kick in the head :)
> I should have somewhere I function I wrote to do that (for different
> purposes, of course) maybe it could help (well, I know it's easy to
> write but so much annoying!).
yes. annoying. i would need a nice big table of all the unicode points and what
to map them to. also the issue was to make it fast - as this is done during
searching to match fuzzily - so every char will have this transform done...
multiple time ( i could limit it to once actually - but not in the code right
> Treviño's World - Life and Linux
> Openmoko community mailing list
> community at lists.openmoko.org
Carsten Haitzler (The Rasterman) <raster at openmoko.org>
More information about the community