ifreund changed the topic of #river to: river - a dynamic tiling wayland compositor || https://github.com/riverwm/river || channel logs: https://libera.irclog.whitequark.org/river/
<Anderson-D> leon-p: I've submitted a patch to allow cycling through multiple sublayouts in stacktile. I was not sure if this functionality is welcome, but it's very useful for me coming from awesomewm. I'll be glad to hear any feedback or comments.
<Anderson-D> Example usage: `riverctl map normal $mod Space send-layout-cmd stacktile 'primary_sublayout full,columns'`
<leon-p> Anderson-D: I'll have a look tomorrow
<Anderson-D> Sure. Thanks!
<Anderson-D> I'm using it with `--per-tag-config` and it works great.
<leon-p> although I just quickly glanced at it and I can already tell you that I don't like `#define MAX_SUBLAYOUTS 16`. It should be set to the exact amount of available sublayouts. Specifying the same sublayout twice therefore should be an error.
Guest58 has joined #river
Guest58 has quit [Client Quit]
<Anderson-D> Thanks. I agree, I'll change it. Quick question: shall I reset the latest commit and then re-send the entire patch with -v2 or do I just add one more commit and send just this commit with -v2?
_whitelogger has joined #river
snakedye has quit [Ping timeout: 260 seconds]
waleee has joined #river
hspak has quit [Remote host closed the connection]
hspak has joined #river
<leon-p> Anderson-D: I'll prefer a v2. Also instead of having MAX_SUBLAYOUTS be a macro, just put it as last entry into the sublayout enum, that is more idiomatic
<leon-p> ah, I see you already did exactly that :)
<novakane> ifreund: it looks like so but for zig 0.9 extern enum you went with `enum(c_int)` for all?
<Anderson-D> leon-p: eventually I had to bump to v3 since I messed up some indentation :)
<leon-p> don't worry too much about style, I can fix that myself for such a small patch
<leon-p> Anderson-D: send you a review
<leon-p> btw, you don't need a cover-letter for single patches, you can just put that into the commit description
<Anderson-D> Gotcha. I just wanted to makes sure I explain the proposed functionality without bloating the commit message too much
snakedye has joined #river
<Anderson-D> Still I'm fairly new to mailing list patch contribution flow, so please excluse me in advance :)
<leon-p> Anderson-D: I don't mind large commit messages. In fact I prefer them. When I look back at a patch in a few years, I want to know everything about it.
<leon-p> Anderson-D: no worries :)
<Anderson-D> Sounds reasonable, I agree
<Anderson-D> For next patch, shall I re-submit all changes as a new v4 patch or is it possible to somehow follow-up on the existing mailing list thread with a new commit?
<Anderson-D> I'm not sure how sourcehut handles those mails
<leon-p> v4 would be the idiomatic thing to do
<Anderson-D> So a new thread is OK, right?
<leon-p> totally fine
<novakane> Anderson-D: did you check https://git-send-email.io/ ?
<leon-p> as long as the patch lands in my INBOX, I will not complain :D
<Anderson-D> novakane: Yep, however it was unclear how v2+ patches are prepared
<Anderson-D> novakane: actually, I just saw it's documented in there. My bad. :)
<novakane> yeah it's not super in depth, it takes a bit of time to be used to it
notzmv has quit [Ping timeout: 240 seconds]
<novakane> Anderson-D: there also some good info here https://git-scm.com/docs/MyFirstContribution#howto-git-send-email
<Anderson-D> Actually the git-send-email.io was very helpful, just took me few times to read through to fully comprehend it
<Anderson-D> leon-p: I've added the requested comment, I hope this change is what you were expecting
<novakane> yeah it's good to look it back after you used a bit git send-email
<leon-p> Anderson-D: I'll have a look in a few hours, am a bit busy right now. But it's probably fine, aside some minor details your patch is sound.
<Anderson-D> leon-p: thank you for your time, and thanks for bearing with me :)
<Anderson-D> Please feel free to judge other minor details, I'll be happy to fix that
<Anderson-D> leon-p: I'll need to submit v5 since sourcehut did not show all your comments, only the first one
notzmv has joined #river
dbuckley has quit [Read error: Connection reset by peer]
dbuckley has joined #river
waleee has quit [Ping timeout: 250 seconds]
elshize has quit [Ping timeout: 252 seconds]
elshize has joined #river
notzmv has quit [Remote host closed the connection]
notzmv has joined #river
waleee has joined #river
chipps_ is now known as chipps
ahti has joined #river
ahti has quit [Quit: Lost terminal]
Guest97 has joined #river
<Guest97> Sorry if this isn't the right place to ask for support, but I couldn't get riverctl working. river starts  fine, except all riverctl and rivertile commands return "Unable to connect to Wayland server" or error: ConnectFailed with only ??? as characters in the error traceback.
<leon-p> Guest97: where are you running riverctl? In a TTY?
<Guest97> from the init script, but running riverctl in terminal while running river wm results in the same (obviously)
<leon-p> weird, what does `echo $WAYLAND_DISPLAY` say?
<Guest97> wayland-2; (I'm currenlty running it inside sway)
<Guest97> as in, running river inside sway
<leon-p> if you are running inside sway, then you must run riverctl with the WAYLAND_DISPLAY of river, not sway
<leon-p> riverctl uses a protocol specific to river that sway does not implement
<leon-p> if you do `WAYLAND_DISPLAY=wayland-2 <your terminal>`, where does the terminal spawn, in river or sway?
<Guest97> inside river, but without the variable also in river
<leon-p> and in the terminal inside river, riverctl does not work?
<Guest97> correct
<leon-p> It would be very helpful if you build river from source, so that we can have debug symbols and a useful backtrace
<Guest97> Alright I'll be back
<leon-p> do we even have the "Unable to connect to Wayland server" string in current rivertile? I can't find it. Maybe it's a mismatch between river and rivertile versions?
<leon-p> s/rivertile/riverctl/
<Guest97> How do compile with debug options (I'm not zig developer, only c/c++)
<leon-p> ah, ok. I misread it, you have issues with both rivertile and riverctl, I only looked at the latter
<leon-p> Guest97: debug mode is on by default if you do not turn on any of the release modes
<leon-p> make sure to also get the git submodules: `git submodule update --init`
<leon-p> Anderson-D: Applied your patch, thanks for the contribution!
<Guest97> weird, it works now, but the ebuild from gentoo doesn't
<leon-p> it must be doing something funky then
<Guest97> version numers report the same
<Guest97> a well, thanx for the help, I figure out how to update the ebuild
Guest97 has quit [Quit: Client closed]
Guest97 has joined #river
Guest97 has quit [Client Quit]