17:01 MrCooper: if the client attaches & commits a wl_buffer and then destroys it before processing the release event, can libwayland-client call the release event callback for that buffer?
17:09 ManMower: I want to say no
17:09 ManMower: but I also want to paste https://ppaalanen.blogspot.com/2014/07/wayland-protocol-design-object-lifespan.html as the authoritative answer (the section "The boogeyman")
17:10 ManMower: however, I appear to be answering the question "*should* libwayland-client call the relase event"...
17:21 emersion: libwayland is just generated code in this regard
17:21 emersion: i'd rather not add special cases
17:22 emersion: (on top of that, would be a breaking change, wouldn't allow distinguishing between the two cases)
17:22 emersion: etc etc
17:23 ManMower: I didn't see a request for change, is that conversation taking place somewhere else?
17:23 ManMower: I thought we were just talking about current behaviour
17:24 emersion: my understanding is that MrCooper is suggesting a change
17:24 ManMower: ah
17:24 ManMower: well, I'm on team emersion for this one if we need to pick sides. definitely not a fan of trying to special case a thing in a way that's a change from how it's always been
17:24 MrCooper: nope, I'm just asking
17:25 ManMower: phew :D
17:25 emersion: if you're asking what the current behavior is, the answer is that it doesn't
17:25 MrCooper: thanks for confirming
17:57 wlb: wayland Merge request !447 opened by () protocol: Clarify wl_buffer.release description https://gitlab.freedesktop.org/wayland/wayland/-/merge_requests/447
17:57 MrCooper: this is what I'm getting at
21:55 wlb: wayland-protocols Issue #237 opened by () Proposal: Gamma control Protocol https://gitlab.freedesktop.org/wayland/wayland-protocols/-/issues/237