On Thu, Sep 18, 2008 at 01:31:44AM +0100, Stan Thorovsky wrote:
>
> First hurdle (never had to bother about it on servers) - how to
> properly set Linux Console in /etc/sysconfig/console for UK keyboard
> (LFS 6.3, paragraph 7.6). Googling didn't produce that much but Gentoo
> wiki suggests this configuration:
>
> KEYMAP="uk"
>
> FONT=?
>
> Is keymap correct and what about font?
>
> Thanks in advance - your help will be very much appreciates.
> Stan
I think "Is keymap correct?" is a question for someone with the
exact (unspecified) hardware. Assuming you have a regular desktop
UK keyboard available to you, you can note what the normal mappings
are, then compare that to your laptop to see what, if anything, is
different or missing in the symbols on the keys.
The other thing to do is to run 'showkey' to see what codes your
keyboard produces. On a modern desktop keyboard, the common
variations in keys are between the space bar and the control keys
(Alt,AltGr,Windows,Menu). Also, on UK keyboards the \| key next to
the space bar doesn't always work on usb. Laptops may drop keys, or
may have idiosyncratic keycodes. If you are working in the linux
console, you will need \|~#`'"<> (some of these have been "fun" to
set up on apple ppc laptop keyboards).
You can compare your keys to the maps in
/lib/kbd/keymaps/i386/qwerty/
If you do need to change, it's probably only a few keys which need
to be corrected. I was going to point you to my uk-utf.map in ~/ken
but apache can't access it so I'll attach it to this mail. If you
are using UTF-8, it might work for you (but not all fonts have all
the symbols). If you are using legacy latin1 encoding, you can't use
the U+ forms. The names are in the kbd package, in src/ksyms.c.
Also see keymaps (5)
ĸen
--
das eine Mal als Tragödie, das andere Mal als Farce
# uk.map altered to ease UTF-8, compose, dead accents
#
# The display of non latin1 characters does not work in distributions using
# console-tools, even if you add the unicode.map from the kbd package.
# For some reason, the non-ascii character codes seem to get garbled, at
# least on ppc.
#
# Note that this defaults to latin1, any other characters
# can _only_ be specified by unicode number, and the
# result ('to') of a 'compose' must be in the latin1 charmap.
#
# So, although I really wanted to replicate what happens for me
# in X, I can't do it all - prioritise W. European, plus
# hungarian (polish, czech have too many other letters)
#keymaps 0-2,4-6,8,9,12
# these keymaps are a bit excessive, at the time I couldn't
# find the docs and this does make AltGr and Shift+AltGr work.
keymaps 0-15
alt_is_meta
include "qwerty-layout"
include "linux-with-alt-and-altgr"
# key any mapped unicode letter by ctrl-shift plus XXXX for the hex digits
include "unicode.map"
strings as usual
# NB I use U+ notation for characters not in latin1 - dumpkeys will report the
# name, but loading by name gives messages about assume iso-8859-x (where x !=
1)
# and produces the wrong result.
#
# Strangely, console-tools produces the messages and errors even when I use
# the U+xxxx form. It also translates U+201E to a symbol name which it then
# claims to not recognise.
# Normal Shift AltGr AltGr+Shift
keycode 1 = Escape
keycode 2 = one exclam one U+00A1
keycode 3 = two quotedbl at U+201E
keycode 4 = three sterling
control keycode 4 = Escape
# euro on AltGr 4 - 'currency' doesn't work for me
keycode 5 = four dollar U+20AC
Control_backslash
keycode 6 = five percent
control keycode 6 = Control_bracketright
keycode 7 = six asciicircum
control keycode 7 = Control_asciicircum
keycode 8 = seven ampersand braceleft
Control_underscore
keycode 9 = eight asterisk bracketleft Delete
keycode 10 = nine parenleft bracketright
keycode 11 = zero parenright braceright
# alt keycode 11 = Meta_parenright
keycode 12 = minus underscore backslash U+00BF
#keycode 13 = equal plus dead_cedilla dead_ogonek
keycode 13 = equal plus dead_cedilla
keycode 14 = Delete
control keycode 14 = Control_underscore
keycode 15 = Tab
# łŁ on w
keycode 17 = w W U+0142 U+0141
# u with double acute
keycode 22 = u U U+0171 U+0170
# extra on o ø and Ø can be accessed from compose / o, compose / O
# so put the double-acute here : it isn't latin-1.
keycode 24 = o O U+0151 U+0150
# extra on p
keycode 25 = p P thorn THORN
#keycode 26 = bracketleft braceleft dead_diaeresis dead_ring
keycode 26 = bracketleft braceleft dead_diaeresis
control keycode 26 = Escape
#keycode 27 = bracketright braceright dead_tilde dead_macron
keycode 27 = bracketright braceright dead_tilde
keycode 28 = Return
alt keycode 28 = Meta_Control_m
keycode 29 = Control
# add extras to 'a'
keycode 30 = +a +A ae AE
# extra on s
keycode 31 = s S U+00DF
# extras on 'd'
keycode 32 = d D eth ETH
# add kra to k (iso-8859-4 so specify as U+)
keycode 37 = k K U+0138
# łŁ also on l
keycode 38 = l L U+0142 U+0141
#keycode 39 = semicolon colon dead_acute dead_doubleacute
keycode 39 = semicolon colon dead_acute
#keycode 40 = apostrophe at dead_circumflex dead_caron
keycode 40 = apostrophe at dead_circumflex
control keycode 40 = Control_g
shift control keycode 40 = nul
keycode 41 = grave notsign bar nul
keycode 42 = Shift
keycode 43 = numbersign asciitilde dead_grave dead_breve
control keycode 43 = Control_backslash
# extra on z «
keycode 44 = z Z U+00AB
# extra on x »
keycode 45 = x X U+00BB
keycode 51 = comma less
keycode 52 = period greater
#keycode 53 = slash question slash
dead_abovedot
keycode 53 = slash question slash
control keycode 53 = Delete
keycode 54 = Shift
keycode 56 = Alt
keycode 57 = space
control keycode 57 = nul
keycode 58 = Caps_Lock
keycode 86 = backslash bar bar
Control_backslash
keycode 97 = Control
# right windows key and right menu key both mapped to Compose -
# my normal keyboards have one or other of these
keycode 126 = Compose
keycode 127 = Compose
include "compose.latin1"
# looks as if U+ compose results only work if in the same iso-8859-x set,
# and the default charset is iso-8859-1
# e.g. can put aring in (also on oa, aa)
compose '=' 'a' to U+00E5
--
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page
