Vadim Frolov wrote:
> Paul Rohr wrote:
>
> > Hi,
> >
> > Thanks for taking a whack at this. I took a look at the diffs and only had
> > a few comments:
> >
> > 1. Do you have any suggestions about how your approach should scale for use
> > in other non-Latin-1 languages? I got the impression that your changes --
> > especially the mods to NVK processing in ev_UnixKeyboard.cpp -- might only
> > help for the Latin-2 subset you need for cz-CZ, but I couldn't tell for
> > sure. A few more comments here might be useful.
> >
> >
>
> Hello,
>
> I try to make cyrillic support in AW and this is my "IMHOs":
> 1. IMHO, we need working with unicode(16 bit, not UTF8) from getting keyboard
> key to output to screen.
> Why? If the difference between target alphabet and Latin is not big, 2-4 chars
> ( such as europe alphabets), the way of Miksa is good. But russian alphabet,
> ukrainian alphabet has 32 - 35 _NON-latin_ letters and charsets for this
> languages are platform-dependent.
> Russian: CP1251 for Win
> KOI8-R for most Unixes
> ISO-8859-5 for ISO systems
> MacCyrillic for Mac
>
> Ukrainian: CP1251 for Win
> KOI8-U for most Unixes
> ISO-8859-5 for ISO
> MacUkrainian for Mac
>
> I don't know what's about arabic, korean and so on...
>
> WBR,
> Vadim.
P.S.
May be I'm wrong or may be I'm not understand something....
But when I press russian "a" on keyboard, i'v got a 0x06C1 as keyval (FreeBSD,
X11R6)
Unicode for russian "a" - 0x0410
Under console text editor in unix this char will be 0xC1 (Koi8-R), in windows -
0xE0.
And after it i see only one way to working - convert all chars after keyboard to
Unicode, internally working only with Unicode, and covert it to proper platform
charset
when calling "drawchars" function.
WBR, Vadim.
