Using TAB in a binding does not work as desired anymore...

[Eclextic]

Hyprland version

Hyprland 0.56.0 built from branch main at commit 4bb6844 clean (render: detect software rendering from the GL renderer ( 16343)). Date: Sun Sep 27 23:02:32 2026 Tag: v0.56.0-209-g4bb6844b0, commits: 7834 Libraries: Hyprgraphics: built against 0.5.1, system has 0.5.1 Hyprutils: built against 0.14.2, system has 0.14.2 Hyprcursor: built against 0.1.13, system has 0.1.13 Hyprlang: built against 0.6.8, system has 0.6.8 Aquamarine: built against 0.15.1, system has 0.15.1 Version ABI string: 4bb6844b0351e4fbf2e3d4e46ae71b551a0e0a42_aq_0.15_hu_0.14_hg_0.5_hc_0.1_hlg_0.6 no flags were set

Is this a regression?

It works normally on the latest (hyprland-0.56.2-3.1) Arch Package. (NOT AUR)

What's the bug?

This used to work:
hl.bind("TAB + T", hl.dsp.exec_cmd("kitty"))

Now while I can launch kitty using the key combination, the letter "T" now does not get sent to other applications, i.e. typing into a browser "T" just does not register or in a terminal or anywhere for that matter.

Reproduction steps

Type the letter after the "TAB" part in the keybinding in any interface you want.
"TAB + T" fucks up the "T"
"TAB + A" fucks up the "A" and so on

Attachments

No response

I acknowledge that:

• I have searched all (open and closed) discussions and issues and my problem is not there.

• I have read the Code of Conduct and agree to follow it.

• I wrote this request with my own hands, my own keyboard, and in my own words.

[vaxerski]

on the latest hyprland it's definitely wrong, because both TAB and T are forwarded IIRC.

I think it should consume tab regardless here, and only consume T if TAB is held.

Would this work for you or are you expecting something else?

[Eclextic]

I think it should consume tab regardless here, and only consume T if TAB is held.

Sure that is IIRC how it used to work. However what happens now is that the bind does get run and the window does get launched, but the button T just does not work at all even when TAB is not held so that is obviously not intended behavior.

Also I noticed another thing regarding keybinds. I have this config here to take a screenshot:

hl.bind("PRINT", hl.dsp.submap("screenshot"))
hl.define_submap("screenshot", "reset", function()
    hl.bind("PRINT", hl.dsp.exec_cmd(hyprshot .. "-m output -m active" .. output_dir .. "-f .output.tmp.png"))
    -- 133 stands for the Super/Win key, but Hyprland does not detect it as a key but rather as a modifier without code:133
    hl.bind("code:133", hl.dsp.exec_cmd(hyprshot .. "-m window -m active" .. output_dir .. "-f .window.tmp.png"))
    hl.bind("SPACE", hl.dsp.exec_cmd(hyprshot .. "-m region" .. output_dir .. "-f .screenshot.tmp.png"))

    hl.bind("Escape", hl.dsp.submap("reset"))
end)

code:133 was the Super Key. That previously worked as a way to use the Super key as a button. However it does not anymore. I guess that, it being allowed as a button is not intended as "SUPER" is considered a modifier?

[Eclextic]

on the latest hyprland it's definitely wrong, because both TAB and T are forwarded IIRC.

I think it should consume tab regardless here, and only consume T if TAB is held.

Would this work for you or are you expecting something else?

But yes to answer your question: That behavior is what I am expecting as well.

[vaxerski]

@Eclextic check #16470