18:01 <novakane> lot of new release comming this month, ifreund you started working on zig/wlroots for river already?

18:05 <ifreund> novakane: yep, already got a zig 0.9.0 branch that I'll rebase this weekend

18:05 <ifreund> still working on wlroots features for the release then I'll do the zig-wlroots update

18:22 <novakane> cool, lots of breaking changes in zig?

18:22 <novakane> I just know about the unused parameter

18:39 <ifreund> novakane: std.mem.Allocator got changed :D

18:42 <novakane> ifreund: oh yeah I had an error building zls the other day about that, didn't look at it yet though

19:21 <pkap> Hey novakane, I hope you won't mind a question about agertu.

19:21 <pkap> In this function: https://git.sr.ht/~novakane/agertu/tree/main/item/src/Output.zig#L86 why don't you need to trigger rendering in the else branch?

19:46 <novakane> pkap: beacause the render function is called here https://git.sr.ht/~novakane/agertu/tree/main/item/src/Surface.zig#L122

19:47 <novakane> if I already have a surface I just render on it but in the else branch I need to configure the surface before rendering

19:53 <pkap> Ah I see. Thank you!

21:24 <andrea> Hi, someone in the AUR comments asked for the session file to be added to river-git.

21:26 <andrea> I think it would be better to add it upstream. Sway for example has it on the upstream repo.

21:26 <andrea> https://github.com/swaywm/sway/blob/master/sway.desktop

21:27 <andrea> ifreund: what do you think?

21:30 <ifreund> hmm, I don't really feel like officially supporting login/session managers. I haven't ever used one

21:31 <ifreund> I feel like distros adding a simple 5 line file in a patch to match whatever session manager config they support is a fine solution

21:31 <ifreund> anyone have strong opinions on that?

21:31 <andrea> Ok then, I'll add it to the package

21:32 <andrea> There's always time to change it

21:32 <ifreund> yeah, it's easy to add but if I were to start officially distributing it I can't really stop doing so :D

21:33 <ifreund> without a breaking change that nobody likes :D

21:39 <novakane> a patch sounds better yeah, and once it done it should not be hard to maintain for packager

21:49 <andrea> done :D

21:49 <ifreund> nice, thanks for packaging river!

22:29 <leon-p> ifreund: RE your review, moving the output focusing to the else branch won't work, because seat.setFocusRaw() only focuses the output of a surface if that surface is a view, not a layer surface.

22:29 <leon-p> I think it would make sense to change setFocusRaw() here

22:29 <ifreund> leon-p: oh, I didn't mean the focusOutput() call, just the focus(null) line

22:29 <leon-p> ah, I see

22:31 <leon-p> correction: setFocusRaw() does not change output focus for views, it just asserts it

22:32 <ifreund> I think the current behavior of setFocusRaw() is fine, or at least fine enough that there's not really anything to be gained by changing it