ChanServ changed the topic of #wayland to: https://wayland.freedesktop.org | Discussion about the Wayland protocol and its implementations, plus libinput
kenny has joined #wayland
Moprius has joined #wayland
Moprius has quit []
nerdopolis has joined #wayland
nerdopolis has quit [Ping timeout: 480 seconds]
fmuellner has quit [Ping timeout: 480 seconds]
Moprius has joined #wayland
Moprius has quit [Remote host closed the connection]
garnacho has quit [Ping timeout: 480 seconds]
<orowith2os[m]>
could we maybe make an ext-background-effect-y protocol that's more like an ext-surface-effect protocol?
<orowith2os[m]>
that works on surfaces, subsurfaces, and layers.
<orowith2os[m]>
I've had some discussion with a cross-platform dev about the DirectComposition Windows interface, and it's really interesting. You attach a base surface with an effect (or more), and you can layer other surfaces on top of it, all with their own effects.
<orowith2os[m]>
I feel like it would make heavy use of subsurfaces, but it would be able to achieve a lot more than the current ext-background-effect.
nerdopolis has joined #wayland
nerdopolis has quit [Ping timeout: 480 seconds]
akimoto has quit [Remote host closed the connection]
kelnos has quit [Remote host closed the connection]
kelnos has joined #wayland
AJ_Z0 has joined #wayland
eluks has quit [Remote host closed the connection]
eluks has joined #wayland
sima has joined #wayland
coldfeet has joined #wayland
coldfeet has quit []
alarumbe has quit []
tzimmermann has joined #wayland
mohan43u has quit [Quit: WeeChat 4.6.3]
mriesch has quit [Remote host closed the connection]
mriesch has joined #wayland
AJ_Z0 has quit [Ping timeout: 480 seconds]
garnacho has joined #wayland
phryk has quit [Remote host closed the connection]
phryk has joined #wayland
rasterman has joined #wayland
Solomoncyj has joined #wayland
<Solomoncyj>
hello
<Solomoncyj>
for PIP i this disable window decorations and enable always on top, right?
leon-anavi has joined #wayland
bindu_ has joined #wayland
bindu has quit [Ping timeout: 480 seconds]
mripard has joined #wayland
<Solomoncyj>
for PIP i this disable window decorations and enable always on top, right?
<kennylevinsen>
solomoncyj: an application cannot request to be "always on top", but yes a client-side decorated window can just not draw decorations to be "frame-less"
daz has quit [Read error: Connection reset by peer]
d42 has joined #wayland
mohit81582263530690 has quit [Quit: mohit81582263530690]
mohit81582263530690 has joined #wayland
fmuellner has joined #wayland
fmuellner_ has joined #wayland
fmuellner has quit [Ping timeout: 480 seconds]
feaneron has joined #wayland
<zamundaaa[m]>
<orowith2os[m]> "that works on surfaces, subsurfa..." <- The protocol works on surfaces of all kinds.
<zamundaaa[m]>
There's no such thing as a layer.
Solomoncyj has quit [Read error: Connection reset by peer]
MrCooper_ has joined #wayland
MrCooper has quit [Ping timeout: 480 seconds]
karenw has joined #wayland
emilio[m] has joined #wayland
<emilio[m]>
A bit sad the rounded corner stuff wasn't done as part of it but it's fair enough, progress is progress :)
LoneStarr has joined #wayland
<LoneStarr>
Thank you, whot, for libinput merge request !1225. Will check out this source very soon. I still have not arrived at an ideal configuration with the problematic laptop. Will send you updates as I make progress.
<Solomoncyj>
so there is no "awlways on top" protocool?
Hypfer is now known as Guest18727
Guest18727 has quit [Read error: Connection reset by peer]
Hypfer has joined #wayland
<any1>
layer-shell?
<kennylevinsen>
sure, but layer-shell is not for normal windows
cyrinux94907 has joined #wayland
akimoto has joined #wayland
nerdopolis has joined #wayland
LoneStarr has quit [Quit: Leaving]
<Solomoncyj>
XDG-TopLevel != "awlways on top" protocool?
<pq>
no
<pq>
top-level is a window type for "a normal application window"
<pq>
like a main window
<Solomoncyj>
so there is currenlt no way for an apliction to declare itself to be always on top to the server/slient/compositor atm?
<Solomoncyj>
*client
<pq>
the name is X11 tradition, I believe
<pq>
Correct.
<pq>
..in general.
<Solomoncyj>
.... so new issue in the wayland repo i assume?
<jadahl>
solomoncyj: ideally, describe the intended use case, and not the low level knob that worked before
<Solomoncyj>
wait... then how does zoom ( the conferencxe software) enter always on top mode without user intervention?
<emersion>
zoom might be a case of xdg-pip
<davidre>
Or xwayland
<Solomoncyj>
nope, its native wayland (the rpm packages)\
<Solomoncyj>
also is adg-pip already released?
<davidre>
no
<davidre>
there are some experimental implementations
<Solomoncyj>
ozone platform?
<davidre>
hidden behind flags/env vars usually though
<kennylevinsen>
xdg-pip is an unreleased, work-in-progress protocol still subject to design discussion
<Solomoncyj>
ah ok
<jadahl>
solomoncyj: you can make one of your toplevels always on top of some other of your toplevels. maybe that's what you're seeing?
<Solomoncyj>
in kde, there is an option when i right click a window in my taskbar to put it on top of all other windoewws, i was thinking if there was way for an aplication to declare that by itself without user intervention
karenw has quit [Quit: Deep into that darkness peering...]
karenw has joined #wayland
<davidre>
No thats a feature in the compositor
feaneron has joined #wayland
<zamundaaa[m]>
[@_oftc_solomoncyj:matrix.org](https://matrix.to/#/@_oftc_solomoncyj:matrix.org) as a user you can use window rules to automate that, if that's what you're after. It would be useful to know exactly what you actually want though, answering questions without context is rarely all that useful
<zamundaaa[m]>
Oro (any/all): the protocol doesn't say anything about subsurfaces or toplevel surfaces. It doesn't have any restrictions on which surface you can apply it to
<zamundaaa[m]>
<orowith2os[m]> "Now my question is, I guess, why..." <- That is a great question. Second question, how did my implementation work with that? Third question, what do we do about it now?
<zamundaaa[m]>
I guess, do a backwards incompatible change to the value, and add a check to wayland-scanner that prevents this nonsense?
<orowith2os[m]>
I have more stuff I'd like to add to background-effect, maybe make a new protocol and mark the old one as "never touch this again"?
<orowith2os[m]>
zamundaaa[m]: You can't pass a subsurface where the protocol asks for a surface, can you?
<zamundaaa[m]>
wl_surface is always the same
<davidre>
A subsurface is a surface role
<zamundaaa[m]>
orowith2os[m]: You don't need a new protocol to add stuff
<orowith2os[m]>
davidre: I. Missed tat entirely.
<orowith2os[m]>
*That
<orowith2os[m]>
zamundaaa[m]: Not need to, but it would also take care of the blur capability stuff?
<zamundaaa[m]>
Sure, but given that the current blur capability can never be set, changing it shouldn't be able to break anything either
<zamundaaa[m]>
And then we don't need a v2 for nothing
<zamundaaa[m]>
Ugh, wayland-scanner can't disallow entries with value 0 in bitfields
<zamundaaa[m]>
The core protocol uses it for dnd_action "none"
<Solomoncyj>
quick q: what prevents keyloggers from declaring all keys as "global keybinds" to listen to all keys?
<zamundaaa[m]>
solomoncyj: apps can't declare keys as global shortcuts
<Solomoncyj>
@zamundaaa[m] so golbal shortcuts need a modifier key?
<zamundaaa[m]>
no. There is no such thing as a global shortcut on the Wayland level
<Solomoncyj>
huh, iirc onlt active apps can lisnen to the keyboard, and apps that need to listen to key binds globaly need to declare them?
<zamundaaa[m]>
Apps can ask for global shortcuts - not keys, just generic actions - through the global shortcuts portal. It asks the user for permission / shows a shortcuts page for that
<Solomoncyj>
ah ok
<zamundaaa[m]>
<zamundaaa[m]> "That is a great question. Second..." <- My implementation worked because there was a code path in the client in which it never checked the capability but just set the blur region anyways
MrCooper_ has joined #wayland
<orowith2os[m]>
<zamundaaa[m]> "Ugh, wayland-scanner can't..." <- A new argument on bit fields to say "yes, 0 is a non-null-effect-y value" would be useful
<zamundaaa[m]>
Yeah, if we only warn when it's not set, that could work
<orowith2os[m]>
Either way, you're breaking somebody. My xdg-decoration proposal uses 0x0 in a valid case, fwiw
<Solomoncyj>
oh, i thout you declared a keybind to a certian action in the app, and the app will report the shortcut to wayland as a global action