You are not logged in.
Hey!
Since my new install of arch with KDE, I cannot use ctrl+z anymore.
Kernel Version: 6.6.6-arch1-1
KDE Plasma version: 5.27.10
I do use X11.
Following output is displayed by xev:
KeyPress event, serial 40, synthetic NO, window 0x8800001,
root 0x7a2, subw 0x0, time 1288410, (683,406), root:(1554,877),
state 0x2000, keycode 37 (keysym 0xffe3, Control_L), same_screen YES,
XLookupString gives 0 bytes:
XmbLookupString gives 0 bytes:
XFilterEvent returns: False
FocusOut event, serial 40, synthetic NO, window 0x8800001,
mode NotifyGrab, detail NotifyAncestor
FocusIn event, serial 40, synthetic NO, window 0x8800001,
mode NotifyUngrab, detail NotifyAncestor
KeymapNotify event, serial 40, synthetic NO, window 0x0,
keys: 1 0 0 0 32 0 16 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
KeyRelease event, serial 40, synthetic NO, window 0x8800001,
root 0x7a2, subw 0x0, time 1288716, (683,406), root:(1554,877),
state 0x2004, keycode 52 (keysym 0x7a, z), same_screen YES,
XKeysymToKeycode returns keycode: 34
XLookupString gives 1 bytes: (1a) "▒"
XFilterEvent returns: False
KeyRelease event, serial 40, synthetic NO, window 0x8800001,
root 0x7a2, subw 0x0, time 1288751, (683,406), root:(1554,877),
state 0x2004, keycode 37 (keysym 0xffe3, Control_L), same_screen YES,
XLookupString gives 0 bytes:
XFilterEvent returns: FalseThank you for help.
Last edited by Stormf0x (2023-12-14 09:33:26)
Offline
Please share your shell configuration files. I presume this is bash?
If you're not sure then post
echo $0
echo $SHELLAnd please explain exactly what happens when you try to use <ctrl>+z and how it differs from the expected behaviour. Thanks.
EDIT: and can we also see
echo $TERMWhich terminal emulator is this? Have you tried the console (TTY)?
Last edited by Head_on_a_Stick (2023-12-14 10:02:16)
Jin, Jîyan, Azadî
Offline
Yes, I do use bash.
/usr/bin/bash.bashrc
#
# ~/.bashrc
#
# If not running interactively, don't do anything
[[ $- != *i* ]] && return
alias ls='ls --color=auto'
alias grep='grep --color=auto'
PS1='[\u@\h \W]\$ '.bash_profile only invokes .bashrc
This is the term, but the issue is globally, in every program.
xterm-256colorEDIT:
I did try tty now but seems to be the same issue. The keybind is not available in any program.
Last edited by Stormf0x (2023-12-14 11:59:11)
Offline
Did you set TERM yourself? I don't use Plasma so I'm not sure what their terminal will report. The point is to let the terminal emulator set TERM, do not try to override it.
Jin, Jîyan, Azadî
Offline
I don't fully know what you are trying to accomplish. The issue is not only in the terminal, but everywhere, changing term would only help with the emulator, right?
Offline
please explain exactly what happens when you try to use <ctrl>+z and how it differs from the expected behaviour
Clearly I am misunderstanding the problem...
Jin, Jîyan, Azadî
Offline
Yeah, I guess so. Let me clarify.
The ctrl+z keybind in general, which should revert stuff doesn't work in general. In neither the console nor OpenOffice, nor any other application. It doesn't even matter if I change the keyboard layout, which I tried with a special German Layout (KOY), because that spot is switched with a x in that layout.
When I changed that, ctrl+x now doesn't work, which means it is the spot that makes the issues and not the key in itself.
Using an external keyboard doesn't solve the issue either and I am pretty much out of my depth
Offline
The ctrl+z keybind ... should revert stuff
It should? Says who? Why?
The only general use of Ctrl-Z I'm aware of is the same as what HoaS seems to have been asking about: it stops the foreground process in an interactive shell (in a tty or terminal emulator).
Does this more general function of Ctrl-Z work? Switch to a tty, run `sleep 30`, then while that is sleeping, hit Ctrl-Z. Are you returned to a prompt after a message that the sleep process was stopped?
Some gui text editors or word processors or other gui programs that have text entry fields may also respond to Ctrl-Z as a form of "undo". I gather this is what you are looking for now, right?
I tried with a special German Layout (KOY), because that spot is switched with a x in that layout.
When I changed that, ctrl+x now doesn't work, which means it is the spot that makes the issues and not the key in itself.
That's a useful diagnostic test, though I'm not sure you've interpreted the results correctly. When you switch the layout in that way, hitting Ctrl and the key with an actual 'Y' glyph on it should generate a Ctrl-X signal, right? And this does not do what you expect Ctrl-X to do, right? What do you expect Ctrl-X to do? Does this outcome happen with Ctrl-X in the original keyboard layout?
"UNIX is simple and coherent" - Dennis Ritchie; "GNU's Not Unix" - Richard Stallman
Offline
Some gui text editors or word processors or other gui programs that have text entry fields may also respond to Ctrl-Z as a form of "undo". I gather this is what you are looking for now, right?
Yes, that's what I am looking for.
That's a useful diagnostic test, though I'm not sure you've interpreted the results correctly. When you switch the layout in that way, hitting Ctrl and the key with an actual 'Y' glyph on it should generate a Ctrl-X signal, right? And this does not do what you expect Ctrl-X to do, right? What do you expect Ctrl-X to do? Does this outcome happen with Ctrl-X in the original keyboard layout?
I do expect ctrl+x to cut text in most programs. Which does work on the original keyboard layout (one key to the right). If I replace it with the usual Z glyph key though, it does not work anymore.
Offline
Stormf0x wrote:The ctrl+z keybind ... should revert stuff
It should? Says who? Why?
The only general use of Ctrl-Z I'm aware of is the same as what HoaS seems to have been asking about: it stops the foreground process in an interactive shell (in a tty or terminal emulator).
Does this more general function of Ctrl-Z work? Switch to a tty, run `sleep 30`, then while that is sleeping, hit Ctrl-Z. Are you returned to a prompt after a message that the sleep process was stopped?
I just tested that, and it seems to work.
Offline
So the key is fine - it's just that programs are not doing what you expect in response to it. The only program I have that uses Ctrl-Z as an "undo" is my web browser - so I may not be the best in helping test further, but have you tried it in your browser, and which browser do you use? (i.e., start typing some text in the reply box on this forum then hit Ctrl-Z, my expectation would be that it would remove the last entered text.)
"UNIX is simple and coherent" - Dennis Ritchie; "GNU's Not Unix" - Richard Stallman
Offline
If we ignore the "what keys do in shouldland" stuff (there's no common shortcut theme, be *extremely* carefuly applying any of the shortcuts MS invented for notepad into a posix terminal!) thise here
state 0x2004, keycode 52 (keysym 0x7a, z), same_screen YES,
XKeysymToKeycode returns keycode: 34
XLookupString gives 1 bytes: (1a) "▒"is clealy bogus.
I do use X11.
loginctl session-statusAnd then
setxkbmap -print -query
xmodmap -pkOffline
If we ignore the "what keys do in shouldland" stuff (there's no common shortcut theme, be *extremely* carefuly applying any of the shortcuts MS invented for notepad into a posix terminal!) thise here
state 0x2004, keycode 52 (keysym 0x7a, z), same_screen YES, XKeysymToKeycode returns keycode: 34 XLookupString gives 1 bytes: (1a) "▒"is clealy bogus.
I do use X11.
loginctl session-statusAnd then
setxkbmap -print -query xmodmap -pk
5 - lcqbraendli (1001)
Since: Thu 2023-12-14 10:47:08 CET; 4h 46min ago
State: active
Leader: 5908 (sddm-helper)
Seat: seat0; vc2
TTY: tty2
Remote: no
Service: sddm
Type: x11
Class: user
Desktop: KDE
Idle: no
Unit: session-5.scope
├─5908 /usr/lib/sddm/sddm-helper --socket /tmp/sddm-auth-a1c2e1e5-cf1e-402b-85bf-c2c2714e4424 --id 3 --start /usr/bin/startplasma-x11 --
user lcqbraendli
├─5932 /usr/bin/kwalletd5 --pam-login 12 15
└─5933 /usr/bin/startplasma-x11
Dec 14 10:47:08 archlinux systemd[1]: Started Session 5 of User lcqbraendli.
Dec 14 10:47:09 archlinux sddm-helper[5931]: pam_kwallet5: final socket path: /run/user/1001/kwallet5.socket
Dec 14 10:47:09 archlinux kwalletd5[5932]: kf.wallet.kwalletd: Wallet failed to get opened by PAM, error code is -9
Dec 14 10:47:31 archlinux kwalletd5[5932]: Application ' "Brave" ' using kwallet without parent window!
Dec 14 10:47:37 archlinux kwalletd5[5932]: kf.wallet.kwalletd: Failed to open wallet "kdewallet" "Read error - possibly i
ncorrect password."
Dec 14 10:47:37 archlinux kwalletd5[5932]: Application ' "Brave" ' using kwallet without parent window!
Dec 14 10:47:41 archlinux kwalletd5[5932]: kf.wallet.kwalletd: Failed to open wallet "kdewallet" "Read error - possibly i
ncorrect password."
Dec 14 10:47:41 archlinux kwalletd5[5932]: Application ' "Brave" ' using kwallet without parent window! kb_keymap {
xkb_keycodes { include "evdev+aliases(qwertz)" };
xkb_types { include "complete" };
xkb_compat { include "complete+caps(caps_lock)+misc(assign_shift_left_action)+level5(level5_lock)" };
xkb_symbols { include "pc+de(koy)+de(deadacute):2+us:3+inet(evdev)" };
xkb_geometry { include "pc(pc104)" };
};
rules: evdev
model: pc104
layout: de,de,us
variant: koy,deadacute,bash: xmodmap: command not foundI do use koy as my regular keymap.
Any idea?
Last edited by Stormf0x (2023-12-14 14:39:04)
Offline
So the key is fine - it's just that programs are not doing what you expect in response to it. The only program I have that uses Ctrl-Z as an "undo" is my web browser - so I may not be the best in helping test further, but have you tried it in your browser, and which browser do you use? (i.e., start typing some text in the reply box on this forum then hit Ctrl-Z, my expectation would be that it would remove the last entered text.)
Yes, doesn't work in browser (brave, firefox & chrome either)
Offline
I do use koy as my regular keymap.
Start with stop doing that and test the behavior on the regular qwerty (us) or qwertz (de) layout, configure a single layout to be sure about what is active.
Report back w/ updated setxkbmap output and the results, I'll look into the koy layout later.
Offline
I do use koy as my regular keymap.
Start with stop doing that and test the behavior on the regular qwerty (us) or qwertz (de) layout, configure a single layout to be sure about what is active.
Report back w/ updated setxkbmap output and the results, I'll look into the koy layout later.
I removed KOY and Qwertz now. With this setup everything works as expected. Any idea how to activate KOY as the primary one?
xkb_keymap {
xkb_keycodes { include "evdev+aliases(qwerty)" };
xkb_types { include "complete" };
xkb_compat { include "complete" };
xkb_symbols { include "pc+us+inet(evdev)" };
xkb_geometry { include "pc(pc104)" };
};
rules: evdev
model: pc104
layout: usEDIT:
I think this could be the issue?
xkb_keycodes { include "evdev+aliases(qwertz)" };this should be the koy layout I believe and not qwertz, that could brick the bindings. Could that be the right assumption?
Last edited by Stormf0x (2023-12-14 14:52:41)
Offline
When setting only "-layout de -variant koy" Ctrl+Alt+<+ü (on a regular-ass qwertz keyboard) gets me the expected event:
xkb_keymap {
xkb_keycodes { include "evdev+aliases(qwertz)" };
xkb_types { include "complete" };
xkb_compat { include "complete+caps(caps_lock)+misc(assign_shift_left_action)+level5(level5_lock)" };
xkb_symbols { include "pc+de(koy)+inet(evdev)" };
xkb_geometry { include "pc(pc105)" };
};
rules: evdev
model: pc105
layout: de
variant: koyOffline
When setting only "-layout de -variant koy" Ctrl+Alt+<+ü (on a regular-ass qwertz keyboard) gets me the expected event:
xkb_keymap { xkb_keycodes { include "evdev+aliases(qwertz)" }; xkb_types { include "complete" }; xkb_compat { include "complete+caps(caps_lock)+misc(assign_shift_left_action)+level5(level5_lock)" }; xkb_symbols { include "pc+de(koy)+inet(evdev)" }; xkb_geometry { include "pc(pc105)" }; }; rules: evdev model: pc105 layout: de variant: koy
I can second that. this is my setting if I change to KOY again:
xkb_keymap {
xkb_keycodes { include "evdev+aliases(qwertz)" };
xkb_types { include "complete" };
xkb_compat { include "complete+caps(caps_lock)+misc(assign_shift_left_action)+level5(level5_lock)" };
xkb_symbols { include "pc+de(koy)+inet(evdev)" };
xkb_geometry { include "pc(pc104)" };
};
rules: evdev
model: pc104
layout: de
variant: koyStill the same issue is back. CTRL+X (labeled z) does not cut in any program. It doesn't do the CTRL+Z shortcut either thought, feels like getting ignored, tbh.
Offline
Still the same issue is back. CTRL+X (labeled z) does not cut in any program. It doesn't do the CTRL+Z shortcut either though
Wut?
Please post the xev output, "x" on the 1st level in koy is the tab key, so ctrl+x would be ctrl+tab?
Offline
Still the same issue is back. CTRL+X (labeled z) does not cut in any program. It doesn't do the CTRL+Z shortcut either though
Wut?
Please post the xev output, "x" on the 1st level in koy is the tab key, so ctrl+x would be ctrl+tab?
No, the x key on layer 1 in KOY is at the position labeled z on a us keyboard.
xev (only x pressed)
KeyPress event, serial 40, synthetic NO, window 0x9800001,
root 0x7a2, subw 0x0, time 22478163, (-262,225), root:(609,696),
state 0x0, keycode 52 (keysym 0x78, x), same_screen YES,
XLookupString gives 1 bytes: (78) "x"
XmbLookupString gives 1 bytes: (78) "x"
XFilterEvent returns: False
KeyRelease event, serial 40, synthetic NO, window 0x9800001,
root 0x7a2, subw 0x0, time 22478237, (-262,225), root:(609,696),
state 0x0, keycode 52 (keysym 0x78, x), same_screen YES,
XLookupString gives 1 bytes: (78) "x"
XFilterEvent returns: Falsexev (ctrl_l pressed)
KeyPress event, serial 40, synthetic NO, window 0x9800001,
root 0x7a2, subw 0x0, time 22526701, (456,88), root:(1327,559),
state 0x0, keycode 37 (keysym 0xffe3, Control_L), same_screen YES,
XLookupString gives 0 bytes:
XmbLookupString gives 0 bytes:
XFilterEvent returns: False
KeyRelease event, serial 40, synthetic NO, window 0x9800001,
root 0x7a2, subw 0x0, time 22526808, (456,88), root:(1327,559),
state 0x4, keycode 37 (keysym 0xffe3, Control_L), same_screen YES,
XLookupString gives 0 bytes:
XFilterEvent returns: Falsexev (ctrl_l+x)
KeyPress event, serial 40, synthetic NO, window 0x9800001,
root 0x7a2, subw 0x0, time 22547514, (218,155), root:(1089,626),
state 0x0, keycode 37 (keysym 0xffe3, Control_L), same_screen YES,
XLookupString gives 0 bytes:
XmbLookupString gives 0 bytes:
XFilterEvent returns: False
FocusOut event, serial 40, synthetic NO, window 0x9800001,
mode NotifyGrab, detail NotifyAncestor
FocusIn event, serial 40, synthetic NO, window 0x9800001,
mode NotifyUngrab, detail NotifyAncestor
KeymapNotify event, serial 40, synthetic NO, window 0x0,
keys: 1 0 0 0 32 0 16 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
KeyRelease event, serial 40, synthetic NO, window 0x9800001,
root 0x7a2, subw 0x0, time 22550751, (218,155), root:(1089,626),
state 0x4, keycode 52 (keysym 0x78, x), same_screen YES,
XLookupString gives 1 bytes: (18) "▒"
XFilterEvent returns: False
KeyRelease event, serial 40, synthetic NO, window 0x9800001,
root 0x7a2, subw 0x0, time 22552629, (218,155), root:(1089,626),
state 0x4, keycode 37 (keysym 0xffe3, Control_L), same_screen YES,
XLookupString gives 0 bytes:
XFilterEvent returns: FalseOffline
I get that for LSGT+y on qwertz (alt is actually not relevant) ie. layer 5.
Either way, this here
FocusOut event, serial 40, synthetic NO, window 0x9800001,
mode NotifyGrab, detail NotifyAncestor
FocusIn event, serial 40, synthetic NO, window 0x9800001,
mode NotifyUngrab, detail NotifyAncestor
KeymapNotify event, serial 40, synthetic NO, window 0x0,
keys: 1 0 0 0 32 0 16 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0is a passive grab, some global shortcut eats the key and probably xdotool's the ▒
sleep 10; xdotool key "XF86LogGrabInfo" and make sure to press and hold the key util xdotool fired.
This will print all grabs into your xorg log, the interesting being the active one(s)
Offline
I get that for LSGT+y on qwertz (alt is actually not relevant) ie. layer 5.
Either way, this hereFocusOut event, serial 40, synthetic NO, window 0x9800001, mode NotifyGrab, detail NotifyAncestor FocusIn event, serial 40, synthetic NO, window 0x9800001, mode NotifyUngrab, detail NotifyAncestor KeymapNotify event, serial 40, synthetic NO, window 0x0, keys: 1 0 0 0 32 0 16 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0is a passive grab, some global shortcut eats the key and probably xdotool's the ▒
sleep 10; xdotool key "XF86LogGrabInfo"and make sure to press and hold the key util xdotool fired.
This will print all grabs into your xorg log, the interesting being the active one(s)
These are the active grabs, using your method:
[ 24165.653] (II) Printing all currently active device grabs:
[ 24165.653] Active grab 0x4143501d (core) on device 'Virtual core keyboard' (3):
[ 24165.653] client pid 6077 /usr/bin/kglobalaccel5
[ 24165.653] at 24165601 (from passive grab) (device frozen, state 6)
[ 24165.653] core event mask 0x3
[ 24165.653] passive grab type 2, detail 0x34, activating key 52
[ 24165.653] owner-events true, kb 0 ptr 1, confine 0, cursor 0x0
[ 24165.653] (II) End list of active device grabsThere is alot more output (The registered grabs), but it is a lot. Do you need that stuff as well?
EDIT:
This looks a bit like the kglobalaccel (Thats a shortcut agent, right?) crashes. Or am I misinterpreting that?
Last edited by Stormf0x (2023-12-14 15:58:51)
Offline
Or am I misinterpreting that?
No, you've setup a global shortcut there that matches the key, but you'll have to look into its config to see what that actually is (unless it does something™ notable that gives it away)
Neither ctrl+x nor ctrl+z make any sense as global shortcuts, so that's probably a misconfiguration.
Offline
Or am I misinterpreting that?
No, you've setup a global shortcut there that matches the key, but you'll have to look into its config to see what that actually is (unless it does something™ notable that gives it away)
Neither ctrl+x nor ctrl+z make any sense as global shortcuts, so that's probably a misconfiguration.
The weird thing is, that this is a fresh plasma install. I changed literally nothing.
I installed it like this:
pacman -S xorg plasma plasma-wayland-session kde-applicationsOffline
kglobalaccel5 might grab the "wrong" key (resolving the symbol to the code for the qwert* layout and then misapplying it to the koy one)
1. kill kglobalaccel5
2. run xev again to record the actually significant event (codes produced and everything else), also test the shortcut now to see whether it acts as expected
3. re-start kglobalaccel5 while the koy layout is active and re-try the behavior
Offline