linkert has quit [Remote host closed the connection]
stefur has joined #river
stefur has quit [Ping timeout: 260 seconds]
<plumeus>
Does River have a runtime flag to disable XWayland? I felt like it might be useful for debugging.
<ifreund>
No it doesnt, perhaps one could just kill the Xwayland process though
<plumeus>
hmm, never thought about doing that
<plumeus>
Was trying to debug a QT6 package on NixOS that doesn't seem to have qtwayland support
upsala has quit [Remote host closed the connection]
<leon-p>
am really looking forward to that new foreign-toplevel-list protocol to replace foreign-toplevel-management. It simplifies lswt soo much. And since compositors will likely have custom extensions for it, it also allows lswt to print actually useful data finally.
<leon-p>
also since outputs are no longer part of it, there is an additional roundtrip involved should there be an extension for getting the output, meaning that hack of making sure to bind the interface after all outputs are bound is likely no longer necessary :)
<tleydxdy[m]>
seeing a bug where if I have brave floating as a small window, close it, then launch brave again (it'll be in tiling) the border show up as the wrong size, that of the small window
<plumeus>
Is the foreign-toplevel-list supposed to be useful for display bars and similar?
<leon-p>
plumeus: it's intended as a basis for every protocol that needs to identify toplevels of other clients. As such, it on its own only lists all toplevels, their app-ids and titles and does not provide any further information about them or any control mechanisms
<leon-p>
for controlling toplevels, like a bar would, you'd create an additional protocol extension that takes this one as a basis
<leon-p>
this also allows compositors to extend it in specific ways. For example, river could add an extension to provide the tags of toplevels.
<leon-p>
and we can avoid implementing extensions we wouldn't support. The current zwlr-foreign-toplevel-management protocol advertises the "minimized" and "maximized" state of toplevels for example, which are concepts river does not have.
<plumeus>
That makes sense.
<leon-p>
also this highlights why the base design of Wayland makes a lot of sense: I just refactored lswt to support the new protocol, but I also kept support for the old one (will deprecate that one in a year or so I think), which was trivial, and supporting extensions of the toplevel-list protocol will also be really simple
<waleee>
leon-p: is the zig rewrite of stacktile in a working state atm?
<leon-p>
yes. it's also document :)
<leon-p>
*documented
<leon-p>
have been using it for a while now
angry_vincent_ has joined #river
angry_vincent_ has quit [Client Quit]
angry_vincent has quit [Ping timeout: 276 seconds]
linkert has joined #river
jao has joined #river
linkert has quit [Remote host closed the connection]
angry_vincent has joined #river
angry_vincent has quit [Changing host]
angry_vincent has joined #river
taupiqueur has quit [Quit: WeeChat 3.8]
taupiqueur has joined #river
n8r has joined #river
<n8r>
Does anyone use wofi? When I spawn wofi on my monitor, it often appears as if it is rendering *off* of my screen to the left, and I am only able to see the right end of it entering my monitor.
<n8r>
This is in river, of course
linkert has joined #river
angry_vincent has quit [Remote host closed the connection]
linkert has quit [Quit: linkert]
youpilai has joined #river
upsala has joined #river
<youpilai>
leon-p : i've been using last stacktile commit with the rule mecanism for some days now but i can't use it properly. I've added an alternative layout in config.ini, with the new parameter tags=20 in order to activate this new layout when tag 20 is populated but it doesn't trigger. I've also added the new layout to the list in the output section
Szadek has joined #river
<leon-p>
youpilai: ah, 20 doesn't mean the tag 20
<leon-p>
river tags are represented as a bitfield
<leon-p>
so when you put 20 there, it really is looking for the tags that represent the 1 bits in the binary representation of 20
<leon-p>
I wonder if there is a way we could better explain what river tags are and how they behave.
<youpilai>
yes i suspected that youwould use the same as river, so i've tried 1048576 (ie 1 << 20) as well but with no success
<leon-p>
try 1 << 20
<leon-p>
eh, I meant 1 << 19
n8r has quit [Ping timeout: 250 seconds]
<leon-p>
that would be 524288
youpilai has quit [Quit: Client closed]
youpilai has joined #river
<youpilai>
i'm trying to activate it with "riverctl map normal $Super M toggle-view-tags $((1 << 19))" with no success
<leon-p>
that's the wrong river command
<youpilai>
oh
<leon-p>
stacktile watches for the currently focused tags, and it has to match them exactly
<leon-p>
so you need set-focused-tags
youpilai has quit [Client Quit]
youpilai has joined #river
<youpilai>
i see, so i can't use it as i though, to change layout with bindings
<leon-p>
no, not yet. However you can change layout parameters with binds
<leon-p>
since you can chain stacktile usercommands, you can change almost everything about a layout with a single bind
<youpilai>
yes but the sublayout parameter if i read the man page correctly
<youpilai>
*not
<leon-p>
yeah, that one you can't change, that's true
<leon-p>
there will be a way to change active layouts eventually, just haven't figured out a sane design yet
<youpilai>
i'll wait for that so, not in a rush (i've read it in you code comments ^^), Thanks for your work anyway
<leon-p>
glad you like it :)
<youpilai>
yes, the only thing i miss from the previous version is the "all_primary" parameter which could simulate a monocle mode, but the new zig version works perfectly (when using the right commands) ;)
<plumeus>
man, imv is so screwed. The unclutter thing doesn't work well on it. Once the cursor is hidden, moving the cursor inside an imv window won't make it appear again
<plumeus>
I really need to find a different image viewer
<leon-p>
that's because imv never sets a cursor ever, so river is a bit confused by that
<plumeus>
imv also has many other issues
<plumeus>
so yeah
<leon-p>
I have switched to the default gnome image viewer for now. Works well enough.
<plumeus>
hmm, will try it. Should be the program called eog.
<leon-p>
yep, that one
<leon-p>
you can change a few settings to hide some of the UI elements
<plumeus>
I'm kind of annoyed of the weird decoration issues with gtk programs
<plumeus>
which I think can be toggled with gsettings
<plumeus>
but NixOS is a bit screwed and idk how to mess with gsettings on it
<youpilai>
i'm using pqiv for that, no perfect but work well enough for my needs
<plumeus>
My favourite when I was using X11 was sxiv
<leon-p>
oh, I remember that one as well
<leon-p>
maybe it can be ported
<youpilai>
i found it when search for a replacement for sxiv indeed
<plumeus>
oh, so pqiv is similar to sxiv?
<plumeus>
I'll give it a try
<plumeus>
welp, pqiv shows the file name on the top but doesn't render CJK fonts properly
<plumeus>
I presume it ships with its own fonts
<youpilai>
i didn't dig that far, maybe there is a way to add font via the pqivrc file
<plumeus>
hmm, I could just use zathura...
youpilai has quit [Ping timeout: 260 seconds]
talismanick has quit [Ping timeout: 260 seconds]
<Nosrep>
i used feh on xwayland for a thing where imv broke
<waleee>
qimgv is ok if you set it to use the mimial ui mode
<waleee>
should with pure wayland now I think since qt supports it?
upsala has quit [Remote host closed the connection]
n8r has joined #river
<n8r>
exit
n8r has quit [Client Quit]
<tleydxdy[m]>
lol what is this debug print std.debug.print("made it here ig\n", .{});
<plumeus>
lol
<plumeus>
One of those messages that were sensible when debugging but wasn't removed?
duncaen has quit [Read error: Connection reset by peer]
duncaen has joined #river
talismanick has joined #river
kotto has quit [Ping timeout: 252 seconds]
kotto has joined #river
<Nosrep>
now feh's broken on fullscreen (inconsistently)
<Nosrep>
time to go image viewer shopping again
<Nosrep>
oh and imv fixed itself
<Nosrep>
that's nice
<Nosrep>
or i guess a river commit fixed it more accurately