Initially reported to sway: swaywm/sway#5104
Those repro steps might be a bit too sway-specific, but in short, run sleep 5s && printf "\e]2;%s\a" "new title", hide the kitty window and observe the title - it doesn't change until kitty window is made visible again.
As suggested in that thread, the issue happens because kitty is blocking when window is hidden, and is not listening on frame events - and for example alacritty does not have this issue.
Is there anything we can do on kitty side to support this?
On the one hand, there is GLFW issue: glfw/glfw#1350
On the other hand, it seems something like this has caused a copy-paste bug in the past (#1051 and swaywm/sway#2770) and it was somehow worked around in kitty itself?
As far as I understand, those frame events that kitty is not listening on, is not something sway-specific, but something common in Wayland...
Initially reported to sway: swaywm/sway#5104
Those repro steps might be a bit too sway-specific, but in short, run
sleep 5s && printf "\e]2;%s\a" "new title", hide the kitty window and observe the title - it doesn't change until kitty window is made visible again.As suggested in that thread, the issue happens because kitty is blocking when window is hidden, and is not listening on frame events - and for example alacritty does not have this issue.
Is there anything we can do on kitty side to support this?
On the one hand, there is GLFW issue: glfw/glfw#1350
On the other hand, it seems something like this has caused a copy-paste bug in the past (#1051 and swaywm/sway#2770) and it was somehow worked around in kitty itself?
As far as I understand, those frame events that kitty is not listening on, is not something sway-specific, but something common in Wayland...