10:30 <LarstiQ> Hello! In my quest to migrate from awesomewm so far riverwm seems the most promising. I've cobbled together my own layout generator with inspiration from looking at stacktile/contrib dir/Nick Hasting gist/waybar code. A monocle layout covers a large part of my maximized-window usage, but for minimized-window being skipped over by focus-view it seems I'd need coperation from command/view_operations.zig

10:30 <LarstiQ> getTarget, is that right?

10:40 <leon-p> minimized windows aren't a thing on river and I am not sure it's a concept that will be added

10:40 <leon-p> instead you could move them to a different tag

10:42 <LarstiQ> Ah, I really really miss them for views I don't want to see _most_ of the time (say stdout of firefox). I could move them to a different tag to get them out of the way but then getting them back is a bit harder with many. Thanks, I'll try it!

10:44 <LarstiQ> leon-p: another thing from awesomewm I'm trying to get back is a list of views on the current tag, waybar's wlr/taskbar shows all views on the output which is unusable. Any pointers for how I should go about that?

10:45 <leon-p> not possible right now unfortunately

10:45 <leon-p> will likely be addressed in the future, but there needs to be some infrastructure first which may take a while

10:46 <LarstiQ> right

10:46 <LarstiQ> It makes monocle switching a bit more tedious, maybe I'll try to implement something like the idea from https://github.com/riverwm/river/issues/26 in my generator

10:47 <LarstiQ> leon-p: as a total Zig newbie, I don't suppose that's infrastructure I can help out with soon?

10:47 <leon-p> it's mostly wayland related things, not zig code, that needs to be createdd

10:47 <LarstiQ> an even steeper climb :)

10:48 <leon-p> river needs an implementation of ext-foreign-toplevel-list for unique window IDs, then we need another protocol extension to add river tags to the toplevel handles (toplevel == window in this context)

10:48 <LarstiQ> thanks for your answers! This should already give me a nicer experience than Sway, for my fossilized usage patterns

10:48 <leon-p> have fun!

10:49 <LarstiQ> is that protocol extension a PR to wayland-protocols? I've read some of them and they don't seem to move super fast

10:50 <leon-p> the ext-foreign-toplevel-list protocol already exists and has been shipped upstream

10:50 <leon-p> the river specific protocol is something only river cares about, so no need to upstream it

10:50 <LarstiQ> aha!

10:51 <leon-p> we would have been a bir closer if I had managed to find the time to get the right wlroots PR merged, but oh well 🐦

10:52 <LarstiQ> no worries, I'm chronically low on time, know how that is

15:35 <leon-p> ages ago we talked / fantasized about doing layouts in lisp

15:35 <leon-p> well, I did a thing: https://git.sr.ht/~leon_plickat/riverguile/

15:35 <angry_vincent> wow :)

15:35 <leon-p> it allows you to define layouts in guile scheme

15:36 <leon-p> it's not perfect and my scheme is rusty, but it works

15:37 <leon-p> I chose guile over janet because it's ubiquitous

15:46 <waleee> I didn't know that gcc requires guile, I thought it was (g)make that brought it onto most linuxes

15:47 <waleee> hmm alpine didn't include it in gcc though

15:47 <leon-p> don't stare too long at gcc, it might stare back

15:49 <waleee> must be the flag that enables makefiles to be able to be guile-scripts that brings it in in both gcc and make

15:50 <novakane> leon-p: nice, funny enough I played with guile and Janet this week and thought about making this later too :P

16:52 <tiosgz> ooh, nice. guess i can nuke that old thing of mine that was bad from the start.

16:53 <tiosgz> i thought i would do something for school, but now i need to yet again replicate the thing that you get with rivertile, anyway

16:54 <tiosgz> which is why i absolutely need other layout generators /facepalm

17:18 <novakane> tiosgz: oh lespa right? I cloned it someday but never take the time to try it lol

18:48 <tiosgz> novakane: yeah, that thing. it is useless for most use cases and the periodic idea makes stuff jump around... which didn't really motivate me to continue

18:49 <tiosgz> i still have a redesign draft somewhere, but couldn't make it nice enough that i'd bother with implementation (which was a whole different issue)

18:50 <tiosgz> so it's nice to see a way shorter (in sloc), but also way more flexible thing with a lispy syntax :)

18:51 <tiosgz> (actually, with a lisp syntax, unlike that lisp/y/ syntax of mine)

18:55 <tiosgz> also, cool that i'm not the only one who clones any layout generator they come around but never tries it

18:56 <novakane> yeah I wanted to try yours because of the lispy syntax but an actual lisp and even scheme is better

18:57 <novakane> hehe yeah I try to look around what's going on the layout generators world but I don't always have the motivation :P

19:13 <vimproved> leon-p: oh my god i've wanted this for so long

19:14 <vimproved> waleee: neither gcc nor gmake should depend on guile pretty sure

19:15 <waleee> vimproved: many if not most distributions enable the guile flag for gmake, which introduce a dependency on guile

19:15 <vimproved> oh i see

19:15 <vimproved> yeah i don't have it enabled because gentoo

19:16 <waleee> the distro I use also had guile in a "whatrequires" for gcc, which is probably for more circumspect reasons

19:27 <novakane> leon-p: is landlock required, can't compile with it, but doesn't seem to break if I remove the include

19:27 <leon-p> oh, leftover include from copy-pasting the wayland logic

19:28 <leon-p> might be an interesting idea, but I don't think guile will like it

19:34 <novakane> leon-p: well I don't like it either :P

19:51 <tiosgz> leon-p: fyi riverguile will error with a non-integer number. in addition to inexact_to_exact, round_number is needed in the conversion

19:52 <tiosgz> (inexact_to_exact is still needed; just tried to go without it)

19:52 <leon-p> huh, I was under the impression inexact_to_exact also rounds, at least that's what reading the docs lead me to believe

19:53 <tiosgz> nah, it just converts 1.2 to 5404319552844595/4503599627370496

19:53 <leon-p> it's been ages since I've "seriously" used scheme

19:54 <tiosgz> btw rounding is also weird, but whatevs (2.5 -> 2, 3.5 -> 4)

19:55 <leon-p> scheme rounds to the closest natural number, or to the even one if directly inbetween

19:57 <tiosgz> seems like i'll have to have a closer look at the lisp family... when i have time, lol

20:00 <leon-p> lisp is fun

20:00 <leon-p> btw, feel free to share any layout scripts. I might set up a contrid dir in the repo or something

23:00 <Guest403232323> Hello!