.\" keymaps.5 - Copyright (C) Andries Brouwer 1998

.\" May be freely distributed.

.\" @(#)keymaps.5 1.10 940130 aeb

.TH KEYMAPS 5 "24 April 1998"

.SH NAME

keymaps \- keyboard table descriptions for loadkeys and dumpkeys

.SH DESCRIPTION

.IX "keymaps" "" "\fLkeymaps\fR \(em keyboard table descriptions for loadkeys and dumpkeys" ""

.IX "loadkeys" "keyboard table descriptions" "\fLloadkeys\fR" "keyboard table descriptions"

.IX "dumpkeys" "keyboard table descriptions" "\fLdumpkeys\fR" "keyboard table descriptions"

.IX keyboard "table descriptions for loadkeys and dumpkeys" keyboard "table descriptions for \fLloadkeys\fR and \fLdumpkeys\fR"

.IX "translation tables"

.LP

These files are used by

.BR loadkeys (1)

to modify the translation tables used by the kernel keyboard driver

and generated by

.BR dumpkeys (1)

from those translation tables.

.LP

The format of these files is vaguely similar to the one accepted by

.BR xmodmap (1).

The file consists of charset or key or string definition lines

interspersed with comments.

.LP

Comments are introduced with

.B !

or

.B #

characters and continue to the end of the line. Anything following one

of these characters on that line is ignored. Note that comments need

not begin from column one as with

.BR xmodmap (1).

.LP

The syntax of keymap files is line oriented; a complete definition

must fit on a single logical line. Logical lines can, however, be split

into multiple physical lines by ending each subline with the backslash

character (\\).

.SH "INCLUDE FILES"

A keymap can include other keymaps using the syntax

.LP

.RS

include "pathname"

.RE

.LP

.SH "CHARSET DEFINITIONS"

A character set definition line is of the form:

.LP

.RS

.EX

charset "iso-8859-x"

.EE

.RE

.LP

It defines how following keysyms are to be interpreted.

For example, in iso-8859-1 the symbol mu (or micro) has code 0265,

while in iso-8859-7 the letter mu has code 0354.

.SH "COMPLETE KEYCODE DEFINITIONS"

Each complete key definition line is of the form:

.LP

.RS

.nf

.BI keycode " keynumber " = " keysym keysym keysym" \fR...

.fi

.RE

.LP

.I keynumber

is the internal identification number of the key, roughly equivalent to

the scan code of it.

.I keynumber

can be given in decimal, octal or hexadecimal notation.

Octal is denoted by a leading zero and hexadecimal by the prefix

.B 0x.

.LP

Each of the

.I keysyms

represent keyboard actions, of which up to 256 can be bound to a single

key. The actions available include outputting character codes or

character sequences, switching consoles or keymaps, booting the machine

etc. (The complete list can be obtained from dumpkeys(1) by saying

.BI " dumpkeys -l"

\&.)

.LP

Each

.I keysym

may be prefixed by a '+' (plus sign), in wich case this keysym is treated

as a "letter" and therefore affected by the "CapsLock" the same way as by

"Shift" (to be correct, the CapsLock inverts the Shift state).

The ASCII letters ('a'-'z' and 'A'-'Z') are made CapsLock'able by default.

If Shift+CapsLock should not produce a lower case symbol, put lines like

.LP

.RS

.nf

.BI "keycode 30 = +a  A"

.fi

.RE

.LP

in the map file.

.LP

Which of the actions bound to a given key is taken when it is pressed

depends on what modifiers are in effect at that moment.

The keyboard driver supports 9 modifiers. These modifiers are labeled

(completely arbitrarily) Shift, AltGr, Control, Alt, ShiftL, ShiftR,

CtrlL, CtrlR and CapsShift.

Each of these modifiers has an associated weight of power of two

according to the following table:

.ev table

.LP

.RS

.TP 20

.I modifier

.I weight

.P

.ta T 24R

Shift	1

.br

AltGr	2

.br

Control	4

.br

Alt	8

.br

ShiftL	16

.br

ShiftR	32

.br

CtrlL	64

.br

CtrlR	128

.br

CapsShift	256

.RE

.LP

.ev

The effective action of a key is found out by adding up the weights of

all the modifiers in effect. By default, no modifiers are in effect, so

action number zero, i.e. the one in the first column in a key definition

line, is taken when the key is pressed or released. When e.g. Shift and

Alt modifiers are in effect, action number nine (from the 10th column)

is the effective one.

.LP

Changing the state of what modifiers are in effect can be achieved by

binding appropriate key actions to desired keys. For example, binding

the symbol Shift to a key sets the Shift modifier in effect when that

key is pressed and cancels the effect of that modifier when the key is

released. Binding AltGr_Lock to a key sets AltGr in effect when the key

is pressed and cancels the effect when the key is pressed again.

(By default Shift, AltGr, Control and Alt are bound to the keys that bear

a similar label; AltGr may denote the right Alt key.)

.LP

Note that you should be very careful when binding the modifier keys,

otherwise you can end up with an unusable keyboard mapping. If you for

example define a key to have Control in its first column and leave the

rest of the columns to be VoidSymbols, you're in trouble. This is

because pressing the key puts Control modifier in effect and the

following actions are looked up from the fifth column (see the table

above). So, when you release the key, the action from the fifth column

is taken. It has VoidSymbol in it, so nothing happens. This means that

the Control modifier is still in effect, although you have released the key.

Re-pressing and releasing the key has no effect. To avoid this,

you should always define all the columns to have the same modifier

symbol. There is a handy short-hand notation for this, see below.

.LP

.I keysyms

can be given in decimal, octal, hexadecimal, unicode or symbolic notation.

The numeric notations use the same format as with

.IR keynumber .

Unicode notation is "U+" followed by four hexadecimal digits.

The symbolic notation resembles that used by

.BR xmodmap (1).

Notable differences are the number symbols. The numeric

symbols '0', ..., '9' of

.BR xmodmap (1)

are replaced with the corresponding words 'zero', 'one', ... 'nine' to

avoid confusion with the numeric notation.

.LP

It should be noted that using numeric notation for the

.I keysyms

is highly unportable as the key action numbers may vary from one kernel

version to another and the use of numeric notations is thus strongly

discouraged. They are intended to be used only when you know there is a

supported keyboard action in your kernel for which your current version

of

.BR loadkeys (1)

has no symbolic name.

.LP

There is a number of short-hand notations to add readability and reduce

typing work and the probability of typing-errors.

.LP

First of all, you can give a map specification line, of the form

.LP

.RS

.EX

keymaps 0-2,4-5,8,12

.EE

.RE

.LP

to indicate that the lines of the keymap will not specify all 256 columns,

but only the indicated ones. (In the example: only the plain, Shift,

AltGr, Control, Control+Shift, Alt and Control+Alt maps, that is, 7 columns

instead of 256.)

When no such line is given, the keymaps 0-M will be defined, where

M+1 is the maximum number of entries found in any definition line.

.LP

Next, you can leave off any trailing VoidSymbol entries from a key

definition line. VoidSymbol denotes a keyboard action which produces no

output and has no other effects either. For example, to define key

number 30 to output 'a' unshifted, 'A' when pressed with Shift and do

nothing when pressed with AltGr or other modifiers, you can write

.LP

.RS

.nf

keycode  30 = a	A

.fi

.RE

.LP

instead of the more verbose

.LP

.RS

.nf

keycode  30 = a	A	VoidSymbol	VoidSymbol \\

		VoidSymbol VoidSymbol VoidSymbol ...

.fi

.RE

.LP

For added convenience, you can usually get off with still more terse

definitions. If you enter a key definition line with only and exactly

one action code after the equals sign, it has a special meaning. If the

code (numeric or symbolic) is not an ASCII letter, it means the code

is implicitly replicated through all columns being defined.

If, on the other hand, the action code is an ASCII character in the

range 'a', ..., 'z' or 'A', ..., 'Z' in the ASCII collating sequence,

the following definitions are made for the different modifier combinations,

provided these are actually being defined.

(The table lists the two possible cases:

either the single action code is a lower case letter,

denoted by 'x' or an upper case letter, denoted by 'Y'.)

.LP

.RS 4

.TP 24

.I modifier

.I symbol

.TP 24

none

x			Y

.PD 0

.TP 24

Shift

X			y

.TP 24

AltGr

x			Y

.TP 24

Shift+AltGr

X			y

.TP 24

Control

Control_x		Control_y

.TP 24

Shift+Control

Control_x		Control_y

.TP 24

AltGr+Control

Control_x		Control_y

.TP 24

Shift+AltGr+Control

Control_x		Control_y

.TP 24

Alt

Meta_x		Meta_Y

.TP 24

Shift+Alt

Meta_X		Meta_y

.TP 24

AltGr+Alt

Meta_x		Meta_Y

.TP 24

Shift+AltGr+Alt

Meta_X		Meta_y

.TP 24

Control+Alt

Meta_Control_x	Meta_Control_y

.TP 24

Shift+Control+Alt

Meta_Control_x	Meta_Control_y

.TP 24

AltGr+Control+Alt

Meta_Control_x	Meta_Control_y

.TP 24

Shift+AltGr+Control+Alt

Meta_Control_x	Meta_Control_y

.PD

.RE

.LP

.SH "SINGLE MODIFIER DEFINITIONS"

All the previous forms of key definition lines always define all the M+1

possible modifier combinations being defined, whether the line actually

contains that many action codes or not.

There is, however, a variation of the definition

syntax for defining only single actions to a particular modifier

combination of a key. This is especially useful, if you load a keymap

which doesn't match your needs in only some modifier combinations, like

AltGr+function keys. You can then make a small local file redefining

only those modifier combinations and loading it after the main file.

The syntax of this form is:

.LP

.BR "" { " plain " "| <modifier sequence> } " keycode

.I keynumber

.B =

.I keysym

.LP

, e.g.,

.RS

.EX

.nf

plain keycode 14 = BackSpace

control alt keycode 83 = Boot

alt keycode 105 = Decr_Console

alt keycode 106 = Incr_Console

.fi

.EE

.RE

Using "plain" will define only the base entry of a

key (i.e. the one with no modifiers in effect) without affecting the

bindings of other modifier combinations of that key.

.SH "STRING DEFINITIONS"

In addition to comments and key definition lines, a keymap can

contain string definitions. These are used to define what each function

key action code sends. The syntax of string definitions is:

.LP

.RS

.B string

.I keysym

.B =

.BI

"text"

.RE

.LP

.I text

can contain literal characters, octal character codes in the format of

backslash followed by up to three octal digits, and the three escape

sequences \fB\\n\fP, \fB\\\\\fP, and \fB\\"\fP,

for newline, backslash and quote, respectively.

.SH "COMPOSE DEFINITIONS"

Then there may also be compose definitions. They have syntax

.LP

.RS

.BI "compose '" char "' '" char "' to '" char "'"

.RE

and describe how two bytes are combined to form a third one

(when a dead accent or compose key is used).

This is used to get accented letters and the like on a standard

keyboard.

.SH ABBREVIATIONS

Various abbreviations can be used with kbd-0.96 and later.

.TP

.B "strings as usual"

Defines the usual values of the strings (but not the keys

they are bound to).

.TP

\fBcompose as usual for "iso-8859-1"\fP

Defines the usual compose combinations.

.LP

To find out what

.I keysyms

there are available for use in keymaps, use the command

.LP

.RS

.nf

.B dumpkeys --long-info

.fi

.RE

.LP

Unfortunately, there is currently no description of what each symbol

does. It has to be guessed from the name or figured out from the kernel

sources.

.LP

.SH EXAMPLES

(Be careful to use a keymaps line, like the first line of `dumpkeys`,

or "keymaps 0-15" or so.)

.LP

The following entry exchanges the left Control key and the Caps Lock

key on the keyboard:

.LP

.RS

.nf

keycode  58 = Control

keycode  29 = Caps_Lock

.fi

.RE

.LP

Key number 58 is normally the Caps Lock key, and key number 29 is

normally the Control key.

.LP

The following entry sets the Shift and Caps Lock keys to behave more

nicely, like in older typewriters. That is, pressing Caps Lock key once

or more sets the keyboard in CapsLock state and pressing either of the

Shift keys releases it.

.LP

.RS

.nf

keycode  42 = Uncaps_Shift

keycode  54 = Uncaps_Shift

keycode  58 = Caps_On

.fi

.RE

.LP

The following entry sets the layout of the edit pad in the enhanced

keyboard to be more like that in the VT200 series terminals:

.LP

.RS

.nf

keycode 102 = Insert

keycode 104 = Remove

keycode 107 = Prior

shift keycode 107 = Scroll_Backward

keycode 110 = Find

keycode 111 = Select

control alt   keycode 111 = Boot

control altgr keycode 111 = Boot

.fi

.RE

.LP

Here's an example to bind the string "du\\ndf\\n" to the key AltGr-D. We use

the "spare" action code F100 not normally bound to any key.

.LP

.RS

.nf

altgr keycode 32 = F100

string F100 = "du\\ndf\\n"

.LP

.SH "SEE ALSO"

.BR loadkeys (1),

.BR dumpkeys (1),

.BR showkey (1),

.BR xmodmap (1)