ifreund changed the topic of #river to: river - a dynamic tiling wayland compositor - https://github.com/ifreund/river - channel logs: https://freenode.irclog.whitequark.org/river/
lyhokia has joined #river
lyhokia has joined #river
<lyhokia> Hello everyone! I'm trying river WM these days. Can anyone tell me how to config libinput in river?
<lyhokia> Hello everyone! I'm trying river WM these days. Can anyone tell me how to config libinput in river?
<lyhokia> More specificly, I'm trying to config my touchpad
<lyhokia> More specificly, I'm trying to config my touchpad
<leon-p> lyhokia: we just has the same topic a few hours earlier :P
<leon-p> lyhokia: we just has the same topic a few hours earlier :P
lyhokia has quit [Quit: Connection closed]
lyhokia has quit [Quit: Connection closed]
<leon-p> basically, river does not (yet) have a way to configure libinput
<leon-p> basically, river does not (yet) have a way to configure libinput
<leon-p> however there is a workaround
<leon-p> however there is a workaround
<leon-p> it's a bit hacky, but it should work
<leon-p> it's a bit hacky, but it should work
lyhokia has joined #river
lyhokia has joined #river
<lyhokia> Thanks a lot!
<lyhokia> Thanks a lot!
lyhokia has quit [Ping timeout: 240 seconds]
lyhokia has quit [Ping timeout: 240 seconds]
londoed has quit [Remote host closed the connection]
londoed has quit [Remote host closed the connection]
londoed has joined #river
londoed has joined #river
waleee-cl has quit [Quit: Connection closed for inactivity]
waleee-cl has quit [Quit: Connection closed for inactivity]
londoed has quit [Remote host closed the connection]
londoed has quit [Remote host closed the connection]
londoed has joined #river
londoed has joined #river
yyp has joined #river
yyp has joined #river
dominikh has joined #river
dominikh has joined #river
maringuu has quit [Ping timeout: 240 seconds]
maringuu has quit [Ping timeout: 240 seconds]
maringuu has joined #river
maringuu has joined #river
<daurnimator> ifreund: no I'm not missing wlroots includes.... I think its just not adding -L/usr/include ?
<daurnimator> ifreund: no I'm not missing wlroots includes.... I think its just not adding -L/usr/include ?
<leon-p> just watching the newest showtime, and I think we also need a lizard mascot now :D
<leon-p> just watching the newest showtime, and I think we also need a lizard mascot now :D
<yyp> ^^
<yyp> ^^
<yyp> Every zig project needs one :D
<yyp> Every zig project needs one :D
Bonicgamer[m] has quit [Quit: Idle for 30+ days]
Bonicgamer[m] has quit [Quit: Idle for 30+ days]
londoed has quit [Quit: Leaving]
londoed has quit [Quit: Leaving]
<ifreund> daurnimator: I'm 99% sure linkSystemLibrary("wlroots") is supposed to handle include dirs as well
<ifreund> daurnimator: I'm 99% sure linkSystemLibrary("wlroots") is supposed to handle include dirs as well
<daurnimator> ifreund: I think its not because I pass -Dtarge ?
<daurnimator> ifreund: I think its not because I pass -Dtarge ?
<ifreund> daurnimator: maybe? what build command are you using?
<ifreund> daurnimator: maybe? what build command are you using?
<ifreund> oh nvm it's at the bottom of the paste
<ifreund> oh nvm it's at the bottom of the paste
<daurnimator> zig build -Dtarget=x86_64-linux-gnu.2.32 -Drelease-safe -Dxwayland -Dman-pages
<daurnimator> zig build -Dtarget=x86_64-linux-gnu.2.32 -Drelease-safe -Dxwayland -Dman-pages
<ifreund> daurnimator: yeah even if it did find the headers linking against the system libs would fail if you build the zig code against glibc 2.32
<ifreund> daurnimator: yeah even if it did find the headers linking against the system libs would fail if you build the zig code against glibc 2.32
<ifreund> the only way to build on arch currently is to link the system libc as described in this issue: https://github.com/ifreund/river/issues/232
<ifreund> the only way to build on arch currently is to link the system libc as described in this issue: https://github.com/ifreund/river/issues/232
<ifreund> I'll patch zig to make this the default before 0.8.0
<ifreund> I'll patch zig to make this the default before 0.8.0
<daurnimator> > The problem is that all of the library dependencies of river must be compiled against the same glibc version as river. This is a non-issue with musl as it doesn't use symbol versioning.
<daurnimator> > The problem is that all of the library dependencies of river must be compiled against the same glibc version as river. This is a non-issue with musl as it doesn't use symbol versioning.
<daurnimator> Why would you need the exact same glibc version?
<daurnimator> Why would you need the exact same glibc version?
<daurnimator> Things should be fine as long as the installed/dynamically loaded version is >=max(application, all_dependencies)
<daurnimator> Things should be fine as long as the installed/dynamically loaded version is >=max(application, all_dependencies)
<ifreund> daurnimator: the system libs are linked against 2.33
<ifreund> daurnimator: the system libs are linked against 2.33
<daurnimator> yes. it should be fine for a lib to be against 2.33 and a application against 2.32, as long as the libc loaded is >=2.33
<daurnimator> yes. it should be fine for a lib to be against 2.33 and a application against 2.32, as long as the libc loaded is >=2.33
<ifreund> you're right though that that quote is technically false
<ifreund> you're right though that that quote is technically false
<ifreund> daurnimator: perhaps zig is doing something else wrong here, --verbose-link output might be helpful
<ifreund> daurnimator: perhaps zig is doing something else wrong here, --verbose-link output might be helpful
<ifreund> without the -Dtarget=x86_64-linux-gnu.2.32 ?
<ifreund> without the -Dtarget=x86_64-linux-gnu.2.32 ?
<daurnimator> (which is a known zig issue)
<daurnimator> (which is a known zig issue)
<daurnimator> the workaround for which is specifying 2.32....
<daurnimator> the workaround for which is specifying 2.32....
<daurnimator> ifreund: a build.zig flag to call addIncludeDir for me might help?
<daurnimator> ifreund: a build.zig flag to call addIncludeDir for me might help?
<daurnimator> `--search-prefix` seems to not work?
<daurnimator> `--search-prefix` seems to not work?
<ifreund> the working workaround for that is linking the system libc not specifying 2.32 :P
<ifreund> the working workaround for that is linking the system libc not specifying 2.32 :P
<daurnimator> need to figure out how to get /usr/include in there...
<daurnimator> need to figure out how to get /usr/include in there...
<ifreund> I mean addIncludePath() in the build.zig will work for a quick test
<ifreund> I mean addIncludePath() in the build.zig will work for a quick test
<ifreund> I'm not too optimistic about your linking succeeding though
<ifreund> I'm not too optimistic about your linking succeeding though
_whitelogger_ has quit [Remote host closed the connection]
<daurnimator> oh I think the --search-prefix thing is fixed in zig master :/
_whitelogger_ has joined #river
_whitelogger_ has joined #river
<daurnimator> ifreund: hmm... getting to a different error now
<daurnimator> ifreund: hmm... getting to a different error now
<daurnimator> ifreund: you generate some program at build-time and run it don't you?
<daurnimator> ifreund: you generate some program at build-time and run it don't you?
<daurnimator> ==> I think its trying to install it to /usr
<daurnimator> ==> I think its trying to install it to /usr
<ifreund> no, the code generator is built into the build.zig, no separate executable is generated and run at build time
<ifreund> no, the code generator is built into the build.zig, no separate executable is generated and run at build time
<ifreund> what error are you getting?
<ifreund> what error are you getting?
<daurnimator> note that if I don't pass `--prefix` then I get: http://sprunge.us/xyvtrb
<daurnimator> note that if I don't pass `--prefix` then I get: http://sprunge.us/xyvtrb
<ifreund> daurnimator: hmm, the most recent zig-wayland commit might help, I don't think river's submodule is updated yet
<ifreund> daurnimator: hmm, the most recent zig-wayland commit might help, I don't think river's submodule is updated yet
<ifreund> that crash without --prefix is a bug, I'll fix that
<ifreund> that crash without --prefix is a bug, I'll fix that
<daurnimator> ifreund: new zig-wayland didn't help with that other error
<daurnimator> ifreund: new zig-wayland didn't help with that other error
<ifreund> hrm, the trouble is that the error stack trace is pretty useless for figuring out what path doesn't exist
<ifreund> hrm, the trouble is that the error stack trace is pretty useless for figuring out what path doesn't exist
<ifreund> daurnimator: pushed a fix for the --prefix issue though
<ifreund> daurnimator: pushed a fix for the --prefix issue though
<ifreund> if you feel like digging through strace output to see what the args are for the failing syscall that would probably help a lot
<ifreund> if you feel like digging through strace output to see what the args are for the failing syscall that would probably help a lot
<daurnimator> ifreund: well that's weird... your --prefix fix apparently fixed the other issue too
<daurnimator> ifreund: well that's weird... your --prefix fix apparently fixed the other issue too
<daurnimator> on the other hand; now I can't compile with a config location of /etc/river/ ?
<daurnimator> on the other hand; now I can't compile with a config location of /etc/river/ ?
<ifreund> ugh, did I screw up the fix?
<ifreund> ugh, did I screw up the fix?
<ifreund> if --prefix is "/usr" it should use "/etc/river"
<ifreund> if --prefix is "/usr" it should use "/etc/river"
<daurnimator> well yeah. but I didn't want to pass `--prefix` :)
<daurnimator> well yeah. but I didn't want to pass `--prefix` :)
<ifreund> oh, I see. If you set DESTDIR it defaults to /usr when the prefix is null
<ifreund> oh, I see. If you set DESTDIR it defaults to /usr when the prefix is null
<daurnimator> I might also expect that some people might want to use /usr/local as their installation prefix, but also look in /etc for config
<daurnimator> I might also expect that some people might want to use /usr/local as their installation prefix, but also look in /etc for config
<ifreund> time to duplicate more logic from std.build
<ifreund> time to duplicate more logic from std.build
<ifreund> daurnimator: I don't plan on supporting that use case with logic, maybe with a -Dconfig_path override
<ifreund> daurnimator: I don't plan on supporting that use case with logic, maybe with a -Dconfig_path override
<daurnimator> yeah that's what I was thinking
<daurnimator> yeah that's what I was thinking
<daurnimator> though really I guess I should put in --prefix
<daurnimator> though really I guess I should put in --prefix
<daurnimator> there was no reason not to
<daurnimator> there was no reason not to
<daurnimator> I just didn't expect that unwrap error
<daurnimator> I just didn't expect that unwrap error
<daurnimator> okay putting back in --prefix brings back the AccessDenied
<daurnimator> okay putting back in --prefix brings back the AccessDenied
<daurnimator> [pid 1634] mkdirat(AT_FDCWD, "/etc/river", 0755) = -1 EACCES (Permission denied)
<daurnimator> [pid 1634] mkdirat(AT_FDCWD, "/etc/river", 0755) = -1 EACCES (Permission denied)
<daurnimator> ^ that's what's failing
<daurnimator> ^ that's what's failing
<yyp> sudo !!
<yyp> sudo !!
<ifreund> daurnimator: pushed another tweak, can you try with that?
<ifreund> daurnimator: pushed another tweak, can you try with that?
<daurnimator> ifreund: same problem
<daurnimator> ifreund: same problem
<ifreund> did it fix things when you use DESTDIR without --prefix though?
<ifreund> did it fix things when you use DESTDIR without --prefix though?
<daurnimator> uh, I never did that?
<daurnimator> uh, I never did that?
<ifreund> well nvm then I guess
<ifreund> well nvm then I guess
<ifreund> I'm very confused why the zig build system is attempting to create /etc/river directly though
<ifreund> I'm very confused why the zig build system is attempting to create /etc/river directly though
<daurnimator> ifreund: this is happening during the build() phase
<daurnimator> ifreund: this is happening during the build() phase
<ifreund> oh
<ifreund> oh
<ifreund> daurnimator: if you don't set --prefix it should work I think
<ifreund> daurnimator: if you don't set --prefix it should work I think
<daurnimator> yeah that's what I said before.... but when I don't set --prefix I can't get the right default config path
<daurnimator> yeah that's what I said before.... but when I don't set --prefix I can't get the right default config path
<ifreund> I'm very confused why the zig build system is trying to create a directory there when you aren't doing zig build install
<ifreund> I'm very confused why the zig build system is trying to create a directory there when you aren't doing zig build install
<ifreund> daurnimator: the lastest commit should have fixed that
<ifreund> daurnimator: the lastest commit should have fixed that
<daurnimator> ifreund: I'm confused by your fix.... why should DESTDIR change the default to /usr
<daurnimator> ifreund: I'm confused by your fix.... why should DESTDIR change the default to /usr
<ifreund> daurnimator: I'm just matching logic from std.build...
<ifreund> daurnimator: I'm just matching logic from std.build...
<daurnimator> if I want to install to /opt/river in a chroot mounted at a /mnt/myfutureroot I would use `DESTDIR=/mnt/myfutureroot zig build --prefix=/opt/river`
<daurnimator> if I want to install to /opt/river in a chroot mounted at a /mnt/myfutureroot I would use `DESTDIR=/mnt/myfutureroot zig build --prefix=/opt/river`
<ifreund> yep that's right and in that case you explicitly set --prefix so the default of /usr if DESTDIR is set isn't used
<ifreund> yep that's right and in that case you explicitly set --prefix so the default of /usr if DESTDIR is set isn't used
<ifreund> DESTDIR changes the default from the cache root to /usr
<ifreund> DESTDIR changes the default from the cache root to /usr
<daurnimator> huh, okay
<daurnimator> huh, okay
<daurnimator> so now I have to set DESTDIR to *anything*? wut lol
<daurnimator> so now I have to set DESTDIR to *anything*? wut lol
<ifreund> ugh, I forgot to duplicate that logic in the other place I check b.install_prefix...
<ifreund> ugh, I forgot to duplicate that logic in the other place I check b.install_prefix...
<ifreund> once stage2 is done I really need to take some time to make the zig build system less ugly
<ifreund> once stage2 is done I really need to take some time to make the zig build system less ugly
<daurnimator> okay now I get ld.lld: error: /lib64/libxkbcommon.so: undefined reference to stat64@GLIBC_2.33 [--no-allow-shlib-undefined]
<daurnimator> okay now I get ld.lld: error: /lib64/libxkbcommon.so: undefined reference to stat64@GLIBC_2.33 [--no-allow-shlib-undefined]
<daurnimator> I guess I need to pass --allow-shlib-undefined
<daurnimator> I guess I need to pass --allow-shlib-undefined
<daurnimator> ifreund: https://github.com/ziglang/zig/pull/8331 escape hatch for future...
<daurnimator> ifreund: https://github.com/ziglang/zig/pull/8331 escape hatch for future...
<daurnimator> well not so much escape hatch as a feature I wish I had now
<daurnimator> well not so much escape hatch as a feature I wish I had now
yyp has quit [Quit: now it's safe to turn off your computer]
yyp has quit [Quit: now it's safe to turn off your computer]
yyp has joined #river
yyp has joined #river
kip_ has joined #river
kip_ has joined #river
kip__ has joined #river
kip__ has joined #river
kip_ has quit [Client Quit]
kip__ has quit [Client Quit]
kip__ has quit [Client Quit]
kip_ has quit [Client Quit]
waleee-cl has joined #river
waleee-cl has joined #river
snakedye has joined #river
snakedye has joined #river
snakedye has quit [Ping timeout: 240 seconds]
snakedye has quit [Ping timeout: 240 seconds]
leon-p has quit [Quit: leaving]
leon-p has quit [Quit: leaving]
yyp has quit [Quit: now it's safe to turn off your computer]
yyp has quit [Quit: now it's safe to turn off your computer]
snakedye has joined #river
snakedye has joined #river
<snakedye> does river have a protocol that allows a client to know the index of the focused view?
<snakedye> does river have a protocol that allows a client to know the index of the focused view?