On Fri, Apr 06, 2001 at 12:12:14PM +0100, [EMAIL PROTECTED] wrote: > All the above are bodges, not a fixes. > > Bodges will come back to fester later. Always fix problems properly > so it never re-appears. Don't introduce new problems that can be > easily worked around.
Just my thoughts - however, radically changing everything for 2.5 is not a fix, instead it introduces *yet*another* complication! We're stuck with this mapping, changing it is no use: In practice, you'd end up maintaining duplicate maps to support both 2.4 and 2.5, not my idea of an improvement. > Andries has the best solution by far to date. Hm, how about a version of loadkeys which if told to load keycodes from a file named "foo.XXX.kmap", looks for a lowlevel-to-lowlevel map in "/usr/share/keymaps/codemap.XXX"? For example, when you tell loadkeys to load "uk.pc.kmap", it will read the data, then map the PC scancodes to RiscPC scancodes using /usr/share/keymaps/codemap.pc, and use the resulting keybindings. Cheers, Richard -- __ _ |_) /| Richard Atterer | CS student at the Technische | GnuPG key: | \/¯| http://atterer.net | Universität München, Germany | 0x888354F7 ¯ ´` ¯
