> > Notice that with my patch (p3_conv_8bit_to_uni.patch > > at http://chris.heathens.co.nz/linux/utf8.html ), > > the "dumpkeys | loadkeys --unicode" is not needed at all. :-) > > If p3_conv_8bit_to_uni.patch uses _8bit_ keymaps in the kernel, > converting it to Unicode at key pressing time, > then it is not so good solution, sorry :-/ > It seems to me that 8bit keymaps in the kernel is generally a bad practice,
I agree, the console fonts are already stored/handled as Unicode in the Kernel, I guess the keymap should behave the same. ... and then translating the key-press trough the same map (the „Unicode console map“ or the „Application charset map“) that the console out uses. > because the keyboard charset trouble will remain despite p3_conv_8bit_to_uni.patch > if I would change one 8bit charset to another _8bit_, > e.g. KOI8-R to Windows-1251 . > > IMHO: kernel keyboard conversion Unicode->8bit will be a good way to 8bit when > needed, > but kernel keyboard conversion 8bit->something is a bad way to something > because 8bit keymaps are not labelled with codepage when loaded to keyboard.c, > so keymap's codepage is undefined. -- Damjan Georgievski jabberID: [EMAIL PROTECTED] -- Linux-UTF8: i18n of Linux on all levels Archive: http://mail.nl.linux.org/linux-utf8/
