KBCT is yet another tool that will help to remap keys across the desktop environment.
When is KBCT useful?
-
If you want to have a custom ergonomic keyboard layout (when your fingers almost never need to leave the home row keys).
-
If you're an ex MacOS user and want something similar to Karabiner Elements.
-
If you want to achieve something similar to QMK layers on your laptop keyboard.
-
If you want to have system-wide vim-like navigation mapping
some_modifier + hjkl
to arrow keys. -
If you find
xbindkeys
,xmodmap
andsetxkbmap
impossiblehard to configure. -
If you want your mapping configuration to work on both Wayland and X11.
-
If you want the configuration to be simple and intuitive.
*However, KBCT is not a tool that can be used to configure macros or arbitrary command execution on a key press. Also note that KBCT requires sudo access.
**KBCT is in active development so expect to see some bugs, however it should be stable enough for simple use cases. In any case create an issue if you encounter something unexpected.
There are several ways of installing KBCT
-
Download the pre-built x86_64 AppImage binary from releases.
cd ~/Downloads wget https://github.com/samvel1024/kbct/releases/latest/download/kbct-x86_64.AppImage chmod +x kbct-x86_64.AppImage #Check that it works sudo ./kbct-x86_64.AppImage list-devices
-
Compile from the sources by first installing
libudev1
package (available for all known distributions).sudo apt install libudev1 # for ubuntu/debian
Then assuming that you have a Rust toolchain installed run the following.
cd /tmp && git clone https://github.com/samvel1024/kbct && \ cd kbct && \ cargo build --release && \ ./target/release/kbct --help
-
Install from the AUR
If you are an Arch Linux user, you can install it from AUR:
$ yay -S kbct-git
Note: The configuration file is expected to be in
/etc/kbct/config.yml
.After the installation, run the systemd service:
$ systemctl start kbct
KBCT uses yaml files as configuration. It allows to apply different mapping rules for different keyboards. There are two main types of key mappings
-
keymap
: maps keys 1-1 regardless of any layer modifiers. (e.gcapslock -> leftctrl
) -
layers
: maps keys based on the active layer. Layer is a key map that will activate and override the existing mapping if a given set of keys are pressed. Much likefn
key is combined withF1-F12
keys. (e.grightalt+i=up
orrightalt+leftctrl+comma=volumeup
)
The following is an exhaustive example configuration of KBCT
# Apply this configuratoin to two keyboards (if connected)
- keyboards: [ "Lenovo TrackPoint Keyboard II", "AT Translated Set 2 keyboard"]
keymap:
leftalt: leftctrl
capslock: leftalt
sysrq: rightmeta
# Specify layered configurations (much similar to fn+F keys)
layers:
# Specify the modifiers of the layer
- modifiers: ['rightalt']
keymap:
i: up
j: left
k: down
l: right
u: pageup
o: pagedown
p: home
semicolon: end
As a result the above configuration will have the following effect
# ↓/↑ stand for press/release events
# One to one example
leftalt↓ ⟶ leftctrl↓
leftalt↑ ⟶ leftctrl↑
# Layer example
rightalt↓ ⟶ rightalt↓
i↓ ⟶ rightalt↑ up↓
i↑ ⟶ up↑
rightalt↑ ⟶ ∅
To start kbct based on yaml configuration file run
sudo kbct remap --config ~/.config/kbct.yaml
Here you can find all the available key names to use in the configuration. Essentially those are taken from Linux API headers. In case you want to disable a key map it to reserved
. For example disabling capslock will look like this capslock: reserved
.
In order to list all the available keyboard devices and their respective names run the following.
sudo kbct list-devices
Most often a keyboard laptop will be named AT Translated Set 2 keyboard
. If you're not sure what the name of your keyboard is, use the evtest
utility to find out.
Important note: KBCT is treating leftshift
/rightshift
, leftalt
/rightalt
, etc. as different keys, so if you want to map both you need to define the mapping twice. This is done on purpose to give fine grained control over configuration.
KBCT is operating on a low enough level to be independent from the window manager or the desktop environment. It is achieved by the following steps
Since KBCT should be run as root it has enough privileges to read and grab the output of the keyboard (e.g the output of /dev/input/event2
). Which means that it becomes readable only for KBCT and the display manager is no longer able to read from the keyboard device.
Then KBCT creates another virtual uinput
device (e.g. /dev/input/event6
), and sends customized key events to that device. The new mapped keyboard is successfully read by the window manager, which as a result reads customized key events.
See the examples.md for user-submitted kbct remap examples.