ChanServ changed the topic of #wayland to: https://wayland.freedesktop.org | Discussion about the Wayland protocol and its implementations, plus libinput | register your nick to speak
eroux has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
flacks has quit [Quit: Quitter]
flacks has joined #wayland
riteo has joined #wayland
<riteo>
hiiii!
<riteo>
I'm that xdg_popup crash person
<riteo>
i came here like a month ago and only now I could finally find enough continuous time to sit down and report properly the whole popup issue with tests and whatnot
<riteo>
I meant f1fe6ec776f77115cadf308530da50994aab3018, "xdg-shell: Mark xdg_popup surfaces as mapped"
<riteo>
so now I don't know what to do, was this already fixed specifically or was this "accidentally" fixed? I'll look into this further, but before going forward I wanted to ask if y'all rightfully got tired of waiting a whole month lol
columbarius has joined #wayland
co1umbarius has quit [Ping timeout: 480 seconds]
riteo_ has joined #wayland
riteo_ has quit []
riteo is now known as Guest3100
riteo has joined #wayland
<riteo>
dang it I think I timed out
Guest3100 has quit [Ping timeout: 480 seconds]
<riteo>
small update: my weston version is the same
<riteo>
now I'm even more confused
riteo has quit [Quit: zzz]
riteo has joined #wayland
<riteo>
I have no idea wft's going on, both my sway and my weston session closed at random, although I have no idea whether this is connected to my godot test session
<riteo>
if someone knows whether there was some intention behind this bugfix, please tell me. I'll also look at the logs once in a while so even if I'm not online I'll still know.
<ManMower>
broken popups were fixed intentionally
riteo has quit [Ping timeout: 480 seconds]
riteo has joined #wayland
<riteo>
sorry for my timeouts, should have fixed my wifi now
<riteo>
ManMower: What was the fix? Because I can't seem to replicate this issue on my old weston version
<riteo>
or perhaps I didn't test enough
<riteo>
also, since it was intentional, I'm really sorry for not having fixed it myself
<ManMower>
well, you didn't break it, so no worries :)
<ManMower>
it was a series of patches recently. and the bug was introduced in the 10.0 cycle somewhere
<ManMower>
never in a release
<riteo>
weird, I have a release
<riteo>
are you talking about weston?
<riteo>
or like wayland the library? I don't know all wayland's project version schemes
<ManMower>
yes
<riteo>
ok I see
<ManMower>
are we talking about the recent weston bug where popups would randomly cause it to crash?
<riteo>
I think not
<ManMower>
oh, ok, I have no idea which bug we're discussing then
<riteo>
I talked about this issue a month ago or so
<riteo>
No worries
<riteo>
I replicated this bug on 10.0.0, stable release
<riteo>
lemme refresh my memory quickly
<riteo>
here, it crashed on committing of a dead popup
<riteo>
a new buffer attach I think
<riteo>
I still didn't update weston, so the bug's probably still there on my machine, although it looks like they fixed it on sway
<ManMower>
I wouldn't expect sway and weston to share a lot of bugs
<riteo>
actually they did sort of in this case
<ManMower>
neat
<riteo>
it failed a bit later, but still did
<riteo>
well, perhaps I got a bit too excited, I didn't touch this laptop in a whole month and I really want to finish my port
<riteo>
I'll, try to replicate seriously, update and look if it's there.
<riteo>
sorry for the whole mess. I think I'll quit for now to avoid other unnecessary quit messages but I'll still read the logs. Cya!
<wlb>
weston Merge request !841 closed (Revert "clients: Drop corresponding touch points when destroying windows")
wvanhauwaert has joined #wayland
shankaru has quit [Quit: Leaving.]
Company has joined #wayland
cool110 has quit [Remote host closed the connection]
cool110 has joined #wayland
fmuellner has joined #wayland
<MrCooper>
if a client were to attach a buffer to a surface, then destroy the buffer before wl_surface_commit, what is supposed to happen? In particular, should the compositor apply the contents of the attached buffer to the surface (if possible)?
<emersion>
there was a recent issue about this on the wayland repo
<emersion>
but it was about destroy right after commit iirc
<ManMower>
wasn't that about destroying an attached (post commit) buffer?
<ManMower>
seems like you shouldn't commit if the attached buffer is destroyed. attach something else? I think this is a client bug.
* ManMower
is in favor of just disconnecting the client in the hopes that such bugs are fixed sooner rather than later, but is aware that this is probably an unpopular opinion. :)
eroux has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<daniels>
(I got annoyed earlier when I was writing some new tests and had to keep remembering export ASAN_OPTIONS=detect_leaks=0 and wtf tbh)
<ManMower>
thanks - I'll have some more mem leaks for you to fix shortly. ;)
<daniels>
you're so selfless
CuteMath has joined #wayland
<CuteMath>
Hello. I am trying to implement copy and paste. I've managed to implement paste, but not the copy. I have found this https://github.com/glfw/glfw/blob/master/src/wl_window.c and I am staring at line 2689. In my code (which is not exactly that) I am destroy the old data source, I create a new data source, I add the listener, I advert the mime type and I set the selection. The listener has functions for target, send and
<CuteMath>
cancelled. Those functions do nothing other than a simple printf. I receive no events and no callback is invoked. I call wl-paste in another terminal but it gets old data.
<CuteMath>
Device manager is correctly retrieved and the data device is correctly created since I can get the clipboard content. I made sure I am not mistakenly destroying the data source right after setting the selection. The only moment I destroy it is when I need to create a new one.
devilhorns has quit []
wvanhauwaert has quit [Read error: Connection reset by peer]
<that_guy>
is that your blog? that was really helpful when I was trying to figure out the data device flow, thanks
<CuteMath>
thank you emersion
<emersion>
yup :)
sozuba has quit [Quit: sozuba]
<CuteMath>
how was I supposed to know that the serial that I need to pass to set_selection is the keyboard enter event serial >:(
<CuteMath>
emersion thank you again, found tha problem
<ifreund>
CuteMath: that's not the only serial you can pass, any recent input event will likely work for most compositors
<ifreund>
what were you passing before that didn't work?
st3r4g has joined #wayland
<CuteMath>
I was thinking about passing an simple int I would increased everytime, whatever.
<ifreund>
CuteMath: you can't do that, it must be a recent input serial. Ideally the "serial number of the event that triggered this request" as the protocol states
MajorBiscuit has quit [Ping timeout: 480 seconds]
<CuteMath>
got it
<wlb>
weston Merge request !929 closed (clients/terminal: Mem leak fixes)
<riteo>
the popup segfault thing is still there :(
<riteo>
I mean, it's not *that* bad of a news, since my setup was the same and it would've been weirder/scarirer if it fixed itself somehow
<riteo>
so now I'm writing my own popup grab compositor-side destruction test thing
<riteo>
how much should I simulate the real repro setup? (make two windows, open a grabbing popup on one, select the other and let the compositor close the popup while it's doing stuff)
<riteo>
I'm not sure how much I can exactly do with the test suite (expecially the client one)
<riteo>
for example, what does the weston_test::activate_surface protocol method do?
floof58 has quit [Remote host closed the connection]
floof58 has joined #wayland
tzimmermann has quit [Quit: Leaving]
hergertme has joined #wayland
___nick___ has joined #wayland
st3r4g has quit [Ping timeout: 480 seconds]
___nick___ has quit []
___nick___ has joined #wayland
___nick___ has quit []
___nick___ has joined #wayland
MajorBiscuit has joined #wayland
MajorBiscuit has quit [Ping timeout: 480 seconds]
___nick___ has quit [Ping timeout: 480 seconds]
cool110 has quit [Remote host closed the connection]
cool110 has joined #wayland
<riteo>
welp, gotta go. I'll still read the logs if someone wants to tell me something.
<riteo>
Byeeeee
riteo has quit [Quit: epic test moment]
rasterman has quit [Quit: Gettin' stinky!]
mvlad has quit [Remote host closed the connection]
txtsd is now known as Guest3170
txtsd has joined #wayland
Guest3170 has quit [Ping timeout: 480 seconds]
mbalmer_ has quit []
rasterman has joined #wayland
eroux has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
hardening_ has quit [Ping timeout: 480 seconds]
MajorBiscuit has joined #wayland
CuteMath has quit [Ping timeout: 480 seconds]
MajorBiscuit has quit [Ping timeout: 480 seconds]
dcz_ has quit [Ping timeout: 480 seconds]
st3r4g has joined #wayland
ybogdano has quit [Read error: Connection reset by peer]