Open thedmm opened 2 years ago
I have the same issue, and I think it's actually a duplicate of #4955. In any case, adding seat * keyboard_grouping none
in my config fixed it for me.
(Sway 1.7, Arch Linux, using internal laptop keyboard in fr-oss layout and external keyboard with a custom US-based layout)
In any case, adding
seat * keyboard_grouping none
in my config fixed it for me.
Thanks for the workaround, works for me too !
Running Arch Linux,
sway version 1.7
. I use multiple keyboards (a built-in one, and an external programmable one), and I have a different set of keybindings for each. As such, I decided to specify--input-device
for everybindsym
invocation. Turns out that adding--input-device
subtly changes the behavior of the binding: activations of keyboard-specific bindings are not consumed by sway, and get passed down to applications.Steps
swaymsg -t get_inputs
).swaymsg bindsym --input-device=<ID> Mod4+a nop
).swaymsg bindsym Mod4+b nop
).fish
insidekitty
(cat -v
is likely the best bet -fish
can be configured to handleESC
).[97;9u
is printed), where as the non-specific one does not.Configuration
Here is a sample configuration file which shows this issue (replace
<ID>
appropriately):Simply open Sway with this file, activate
Mod4+t
, (opencat -v
),Mod4+a
,Mod4+b
, note the difference in output, close the shell, and exit withMod4+e
.Log
I have a log, but due to the excess amount of low-level information therein, I would rather not upload it. Here are what appear to be the most relevant snippets: