<reepca>I find myself a bit distracted improving OriansJ's forth, but I don't plan on spending much time on it after tonight... after I finally get displaying a number to work. Rather difficult to debug without that.
<rekado>instead of using “guix build” just use “guix package -i” to install the package to your profile
<balduin>okay, thanks now I have that in my profile.
<balduin>However, my question is how can I add the binaries in the /gnu/store.... location to my path. I installed guix in Fedora and if I try to execute the binary from the package I build I get the message program not found.
<rekado>for the gnu-build-system we run “make check” (or any other target)
<rekado>for other build systems it’s a different thing we run.
<balduin>ah, okay. Still the test depends on the author of the software and not on the author of the package (the scheme file).
<balduin>how to you handle packages which are based on electron? Electron is the chromium based framework/building block for applications developed by GitHub. The most well known application based on Electron is the Atom editor.
<rekado>I don’t think we have any Electron applications in Guix yet.
<sneek>ng0, efraim says: you might have to pass --sysconfdir=/etc to configure; if you find more E21 oddities I'm happy to,push fixes
<ng0>sneek: later tell efraim: I have more E21 odities (see open Enlightenment related bugs), I just had no time to push fixes for them, and for one I received no alternative solutions so I'll just prepare a patch for it.
<rekado>sneek: later tell ng0 Comments like “For possible past reviews search the guix-devel <at> gnu.org archive” make it less likely that a review will happen in time, because it takes extra time to find those past reviews.
<sneek>efraim, ng0 says: I have more E21 odities (see open Enlightenment related bugs), I just had no time to push fixes for them, and for one I received no alternative solutions so I'll just prepare a patch for it.
<sneek>efraim, ng0 says: I meant 'to create fixes for them'
<efraim>They really should have torrents for those
<civodul>efraim: but we don't need them anymore, do we? :-)
<janneke>what's with guix gc --help' saying PATHS instead of FILES/DIRECTORIES?
<janneke>civodul: i'm currently at: "probably not, but do we want to?"
<civodul>also, what kind of "handling" are we talking about?
<janneke>"it seems" that when installing two packages with colliding circular symlinks, the collision checker starts to follow waay too deep into the circular symlink tree, finding more and more collisions until memory is exhausted
<janneke>but i would to create a proper test and possibly rather find a way to fix my package
<janneke>as much as i like the young exwm and exwm-x, i cannot seem to get in touch with the developers or other users -- emacs at 29G/14G now brb
<civodul>janneke: oh, memory exhaustion is not good
<TopSekret_>Hey, I have a problem. Installing GuixSD lasts really long, and at some point it requires to use --fallback. I have been advised that this is because of outdated mirror. Is it true? If not, how can I avoid it?
<reepca>janneke: I don't know if "joining" is the right word, but if I have spare time it'd be neat to put my forth hobbyist status to work. But for now, I have several thousands of lines of C++ to read through (or attempt to reverse-engineer by mostly guessing, whichever is easier).
<reepca>OriansJ: could EXECUTE be made so that instead of taking the address of a pointer to the assembly, it takes the address of the assembly itself? That way the stuff we compile with , and the stuff we EXECUTE directly are one and the same
<OriansJ>reepca: it does take the address of the assembly itself
<OriansJ>whatever is on top of the stack is the address that is directly jumped to
<reepca>Huh. That's weird. When I give it the address of a colon definition, it crashes.
<janneke>hmm, how do i set/override "--substitute-urls" in the repl?
<OriansJ>reepca: for example if you run EXECUTE with 0 on top of the stack it will jump to address 0
<janneke>i'm getting lost trying to read guix/scripts/build.scm
<reepca>Oh, I think I see now - I had it backwards. What I've been compiling with , is the address of a pointer to the assembly.