<cwre> Infinisil: what do you do?
<Infinisil> Soon I may even ditch macOS from my main SSD and put it on an external one
<gchristensen> imalsogreg: nice! no idea, I haven't used nix without the daemon for more than a few minutes :P
<Infinisil> cwre: Do you mean job?
<cwre> Infinisil: well, why do you need so much efficiency?
<Infinisil> Ahh
<cwre> That's why I don't run windows in a VM.
<cwre> I need it for gaming :(
<cwre> Wish I didn't.
<Infinisil> cwre: Because my mac's battery life is very very bad
<simpson> cwre: I just finished my last Windows-only game, and I'm now finally retiring that machine. Quite happy.
<cwre> Ahh, fair enough.
<cwre> simpson: I can't finish my windows-only game.
<imalsogreg> gchristensen: One thing - I see the thorough uninstall instructions now that I reread the logs. But when I was having the thought "time to delete nix", I didn't remember that the uninstall instructions existed.
<gchristensen> imalsogreg: thank you _very_ much for your help on this project of mine.
<cwre> Infinisil: what do you do for a job?
<imalsogreg> Happy to help :) Thanks for the patience.
<cwre> How old is your mac btw
<Infinisil> cwre: Studying computer science
<cwre> Infinisil: nice nice.
<Infinisil> cwre: It's from 2012, a MacBook Air 13"
<Infinisil> Yours?
<cwre> Ahhh, nice and old.
<gchristensen> imalsogreg: I understand, it is a different use case, too, when doing the install again for the second time "just to see"
<cwre> 2015 MBP with Discrete, haha....
<Infinisil> Oh nice
<cwre> Yeah, my mother split cost with me.
<cwre> But honestly I still regret getting it.
k0001_ has quit [(Remote host closed the connection)]
<cwre> Because the performance for gaming is still abysmal.
<imalsogreg> gchristensen: Yes, maybe best to optimize the normal path. Really excellent experience, running that script!
<Infinisil> Yeah mac is pretty bad for gaming..
<cwre> Infinisil: yeah, even with bootcamp.
<gchristensen> imalsogreg: <3 I'm so happy to hear that. I'll add you as a happy customer :PP
<Infinisil> I have a super powerful windows machine for gaming, but I'll put NixOS on it eventually
<cwre> Infinisil: I probably never will...
<cwre> Unless Linux becomes viable for AAA gaming.
<cwre> Witcher 3, Senua's Sacrifice, Path of Exile....
<cwre> I need them, man.
<cwre> And I just end up spending all my time in Windows because it's easier than rebooting to Linux.
<Infinisil> cwre: Well I meant I'll keep windows on it, maybe via virtualbox or somethnig
<cwre> Infinisil: yeah, but my current rig doesn't support GPU passthrough...
<cwre> :((((
<Infinisil> Ohh
<cwre> The only god damn proc in Ivy Bridge that doesn't support it
<cwre> >:((((((
<cwre> Literally i5's support it but my K skew can't do it.
<Infinisil> Whew, what a bummer
<cwre> K skew i7.
<cwre> The 4th gen can support it.
<cwre> UGH I'm still mad about that, haha.
<cwre> And I found out about this like last year.
<Infinisil> Well sometimes things don't work out
<cwre> Yeah, I'll be building a new rig after I pay off student debt.
<cwre> So in like 2 or 3 years probably.
<cwre> Infinisil: what's the config option for importing?
<cwre> imports = []?
<cwre> imports = [ mozilla.nix ]?
<Infinisil> cwre: You should already have an imports statement at the beginning of your configuration.nix
<cwre> If it's in the same dir as configurations.nix
<cwre> Oh, awesome
<Infinisil> ./mozilla.nix
<Infinisil> It needs a "/" to know it's a path
<cwre> Yeah
<Infinisil> so probably it looks like imports = [ ./hardware-configuration.nix ./mozilla.nix ];
<cwre> Yeah
<cwre> Now it's telling me I don't have unFree = True
<cwre> But....
<cwre> I do.
<Infinisil> cwre: Maybe try putting { allowUnfree = true; } in /root/.config/nixpkgs/config.nix
<cwre> It's in my configuration.nix
Wizek_ has quit [(Ping timeout: 260 seconds)]
<Infinisil> Apparently you need it, don't know why, it doesn't work without it for me
<cwre> Do I need the brackets?
<cwre> *braces
<Infinisil> cwre: Yup, this file is a nix expression just like everthing else, and to declare stuff it needs to be an attribute set, which is declared with braces
<cwre> I don't have a .config in ~root
<cwre> Do you mean ~user?
<Infinisil> cwre: Yeah you need to create it. It needs to be in root because you're calling rebuild with sudo
<cwre> Infinisil: okay, thanks for bearing with me.
<Infinisil> cwre: No problem :). Does it work like that?
<Infinisil> Ohh actually I might just know how that file isn't required
<cwre> What do you mean?
<Infinisil> Ah no it doesn't work, never mind
mudri has quit [(Ping timeout: 260 seconds)]
<Infinisil> clever: Do you know why this requires allowUnfree in the root's config? https://gist.github.com/Infinisil/66c70bddeb88fe362dde8bdd3f832ee5
ennui has quit [(Ping timeout: 248 seconds)]
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] dywedir opened pull request #27246: ion: init at 1.0.3 (master...ion) https://git.io/vQ659
NixOS_GitHub has left #nixos []
<cwre> Infinisil: My firefox is still only 54!
<cwre> D:
<Infinisil> cwre: What does `firefox --version` say?
<Infinisil> You also need to remove the normal firefox from systemPackages because this would give a collision
<Infinisil> And you need to quit every old firefox version you're running, otherwise it's just gonna create new windows from that
<cwre> It's still saying 54. I did all of those things D:
<cwre> Should I try a reboot?
spinus has joined #nixos
<Infinisil> cwre: Reboot won't do anything
<Infinisil> cwre: Did you see it being installed when rebuilding?
<cwre> I saw it being cloned...
<Infinisil> And a big download should've been visible
<Infinisil> And you should've seen the install path of it
tmaekawa has quit [(Quit: tmaekawa)]
<Infinisil> Which was /nix/store/d2mv00gza0cf72jc45dawapihv7p4n0y-firefox-nightly-bin-56.0a1-20170705170343 for me
<cwre> Infinisil: still have it installed in my env too
<cwre> I got it
<cwre> Thanks!
<Infinisil> Ahh
<cwre> nix-env -e firefox got it to work
<Infinisil> yeah that overrides it
<cwre> Anyone know why I'm getting a UTF box on my terminal? I have powerline-fonts in my system...
<cwre> It's supposed to be a git-fork
<Infinisil> cwre: Then it's probably not using the right font, you need to set this to the right one in the terminals config
<cwre> Infinisil: what's a patched font?
<Infinisil> cwre: A font created by someone, but patched because you needed little modifications
<cwre> Nooo I mean what is a patched font installed with that package? Like I said it's already installed on my system, but it doesn't really say which ones are patched.
<Infinisil> cwre: I don't really know where you're trying to get with this, why does it matter if they're patched?
<cwre> The ZSH theme I'm using recommends it to ensure font compatibility
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] peterhoeg pushed 2 new commits to master: https://git.io/vQ6db
<NixOS_GitHub> nixpkgs/master 73c21aa Peter Hoeg: kronometer: fix build
<NixOS_GitHub> nixpkgs/master 4a8d67e Peter Hoeg: krusader: fix compilation
NixOS_GitHub has left #nixos []
<Infinisil> cwre: Don't really know
<cwre> Mmm, okay...
<cwre> I'm gonna call this project a wrap for today, haha.
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] rnhmjoj opened pull request #27247: wine: add missing dependency (master...wine) https://git.io/vQ6Fq
NixOS_GitHub has left #nixos []
Lisanna_ has joined #nixos
hellrazo1 has joined #nixos
<Lisanna_> Hey guys, I'm trying to do something relatively simple but I need help with the syntax ): I'm installing packages with the declarative format in configuration.nix, and for one of them I would like to apply a patch on top of it (i.e., I need to set patches = mypatch), but I'm not sure how to set the variable for just the package I want to. Any advice?
hellrazor has quit [(Ping timeout: 248 seconds)]
spinus has quit [(Ping timeout: 240 seconds)]
tmaekawa has joined #nixos
tmaekawa has quit [(Client Quit)]
<Infinisil> Lisanna_: Can you show what you got for now?
<gchristensen> Lisanna_: (pkgs.myPackage.overrideAttrs (x: { patches = [ ... your patches ... ]; })) in your systempackages mightdo it
<Lisanna_> I'd like to set the "patches" variable for JUST the st package
<Lisanna_> because the recipe for st interprets that variable if it's set
<Lisanna_> gchristensen you mean put that in the () after the package in question?
nix-gsc-io`bot has joined #nixos
<nix-gsc-io`bot> Channel nixpkgs-unstable advanced to https://github.com/NixOS/nixpkgs/commit/3a227e673a (from 12 hours ago, history: https://channels.nix.gsc.io/nixpkgs-unstable)
nix-gsc-io`bot has quit [(Client Quit)]
<gchristensen> oh that one is a bit special Lisanna_
justelex_ has joined #nixos
<gchristensen> outside of systemPackages, put: nixpkgs.config.st.patches = [ your patches ];
<gchristensen> in pkgs/top-level/all-packages.nix you'll find `st = ...` with: conf = config.st.conf or null;patches = config.st.patches or null; is how I found that
<Lisanna_> oh okay
<Lisanna_> well that's easy
<Lisanna_> is that a way to set variables like that for all nix package recipes?
<Lisanna_> or just for package definitions that allow that
<gchristensen> Lisanna_: just for packages set up specifically like this:
<Lisanna_> OK
<Infinisil> You basically have to read the source to know this :(
<gchristensen> yeah :(
justelex has quit [(Ping timeout: 240 seconds)]
<Lisanna_> Infinisil it also doesn't help that I still struggle with the syntax ^^
<gchristensen> Lisanna_: have you read the nix pills?
<Lisanna_> gchristensen I see, so certain packages can expose certain environment things for easy assignment, that makes sense
<gchristensen> they can be really helpful in understanding lots of nix stuff
<gchristensen> Lisanna_: yeah, but usually only packages that cannot be compiled at run time do this
<gchristensen> Lisanna_: https://nixcloud.io/tour/?id=1 this is helpful too
<Infinisil> gchristensen: Do you know why st includes such a patch config?
<Infinisil> It goes against the standard
<gchristensen> b/c of how st is designed :(
<Infinisil> s/standard/convention
<Infinisil> gchristensen: Wouldn't just overrideAttrs work too?
<Lisanna_> gchristensen nix... pills?
<Lisanna_> Oh okay
<Infinisil> I actually haven't read it myself, but it's being referenced a lot
<Lisanna_> gchristensen wow, https://nixcloud.io/tour/?id=1 looks awesome
<Lisanna_> (even though I told myself I would be spending the weekend learning emacs; I got sidetracked as I frequently do with configuring my Nix environment ^^)
ertes has joined #nixos
MP2E has quit [(Quit: leaving)]
lambdamu_ has joined #nixos
mkoenig has quit [(Remote host closed the connection)]
<Infinisil> Lisanna_: I constantly get sidetracked by every possible thing you could get sidetracked by
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] peterhoeg pushed 1 new commit to master: https://git.io/vQ6bm
<NixOS_GitHub> nixpkgs/master 5906a27 Peter Hoeg: krusader: add missing dependencies
NixOS_GitHub has left #nixos []
lambdamu has quit [(Ping timeout: 240 seconds)]
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] peterhoeg pushed 1 new commit to master: https://git.io/vQ6bY
<NixOS_GitHub> nixpkgs/master a8c1419 Peter Hoeg: peruse: add missing dependencies
NixOS_GitHub has left #nixos []
<gchristensen> Lisanna_: as an emacs and nix user, nix pays off better IMO :)
phinxy has quit [(Quit: Leaving)]
<gchristensen> Lisanna_: but emacs has a better tutorial :D
<Lisanna_> gchristensen haha, yeah, it probably does
<gchristensen> emacs also has a 30 year head start
<Infinisil> I too have to get more into emacs, I'm probably not better at using emacs than my mom is at using a computer
duzzy2 has joined #nixos
bennofs has quit [(Ping timeout: 258 seconds)]
<clever> Infinisil: not sure, can you gist the entire error?
s33se has joined #nixos
Supersonic112 has quit [(Disconnected by services)]
Supersonic112_ has joined #nixos
Supersonic112_ is now known as Supersonic112
nix-gsc-io`bot has joined #nixos
<nix-gsc-io`bot> Channel nixpkgs-unstable advanced to https://github.com/NixOS/nixpkgs/commit/50831d543d (from 8 hours ago, history: https://channels.nix.gsc.io/nixpkgs-unstable)
nix-gsc-io`bot has quit [(Client Quit)]
s33se_ has quit [(Ping timeout: 240 seconds)]
<Lisanna_> Ugh... there's some arcane problem with applying the patch and I don't know how to debug it...
mbrgm has quit [(Ping timeout: 268 seconds)]
<gchristensen> oh?
mbrgm has joined #nixos
<Lisanna_> I downloaded the same st source that Nix does
<Lisanna_> made a patch file with git diff
<Lisanna_> checked it against the patch file that nix applies by default... same syntax exactly, looks identical (https://github.com/NixOS/nixpkgs/blob/master/pkgs/applications/misc/st/st-fix-deletekey.patch)
<Lisanna_> but when I include the file, it says Hunk #1 failed
<Lisanna_> so the diff isn't applying but I have no idea why...
<gchristensen> can you share your file?
<Lisanna_> OH, it's applying it to st-0.6?!
<Lisanna_> well there's my problem
<gchristensen> ahhh run nixos-version
<Lisanna_> I probably need to update :)
* Lisanna_ tries to remember the update command
<gchristensen> ahhh run nixos-version?
<Lisanna_> Flounder
<gchristensen> 16.09?
<Lisanna_> yes
<gchristensen> yikes, yeah time to upgrade
<Lisanna_> (sorry, don't have good copy/paste setup)
<gchristensen> this will upgrade you to 17.03: nix-channel --add https://nixos.org/channels/nixos-17.03 nixos
<gchristensen> then nixos-rebuild boot --upgrade
<Lisanna_> gchristensen thanks for holding my hand ^^ Nix is a very... different way of doing things, still getting used to stuff
<gchristensen> you're welcome! I'm always glad to help people get started. paying forward all the help I got when I started.
iyzsong has joined #nixos
<Lisanna_> gchristensen Cool! Everything is working perfectly now - thank you so much!
<gchristensen> awesome :D
<gchristensen> note that nixos-rebuild boot --upgrade means you need to reboot to finish the upgrade
<gchristensen> (which is good, for such a big updateE)
<Lisanna_> hence the name I imagine ^^
<gchristensen> yep!
<gchristensen> ok I'm going to bed, good luck Lisanna_ -- I hope to see you around!
<Lisanna_> gchristensen okay, thanks!
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] peterhoeg pushed 1 new commit to master: https://git.io/vQ6N8
<NixOS_GitHub> nixpkgs/master e9c803f Peter Hoeg: peruse: one last missing dependency
NixOS_GitHub has left #nixos []
<Infinisil> clever: It's just the usual error: Package ‘firefox-nightly-bin-unwrapped-56.0a1-20170705170343’ in ‘/nix/store/19rm9wpkqkfqzv6i6agsn5s1aq7kx3hs-7q45a7b44s3bmwal70lzr7nq4x3pbfp9-nixpkgs-channels-ed070354a9e307fdf20a94cb2af749738562385d-src/pkgs/applications/networking/browsers/firefox-bin/default.nix:175’ has an unfree license (‘unknown’), refusing to evaluate.
<Infinisil> clever: When importing https://gist.github.com/Infinisil/66c70bddeb88fe362dde8bdd3f832ee5 without allowUnfree in /root/.config/nixpkgs/config.nix
Fare has joined #nixos
<Infinisil> clever: I feel like nixpkgs-mozilla needs to be modified to not require that
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] peterhoeg pushed 1 new commit to master: https://git.io/vQ6NQ
<NixOS_GitHub> nixpkgs/master 6e2579c Peter Hoeg: partition-manager: add missing dependencies
NixOS_GitHub has left #nixos []
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] therealpxc opened pull request #27248: kbd: fix unicode_start and unicode_stop scripts (master...kbd) https://git.io/vQ6N7
NixOS_GitHub has left #nixos []
<clever> Infinisil: firefox-bin is a pre-compiled thing, while firefox is built by hydra
ebzzry has joined #nixos
spear2 has quit [(Quit: ChatZilla 0.9.93 [Firefox 46.0.1/20160521140538])]
<Infinisil> clever: Well whatever, it works for now, I was talking about the nixpkgs-mozilla stuff, which doesn't use the config set in configuration.nix's nixpkgs.config option
<clever> Infinisil: ah, any time you do import <nixpkgs>{};, it reads config.nix from $HOME
<clever> you have to do import <nixpkgs> { config = config.nixpkgs.config; };
vaibhavsagar has joined #nixos
erlandsona has joined #nixos
<erlandsona> I'm getting pretty close to figuring out how to fix my broken surround sound speakers in my laptop! That said, I need to figure out how to get the contents of hda-jack-retask.fw into the "nix" version of /lib/firmware/hda-jack-retask.fw and then point to that file with a kernel entry `options snd-hda-intel patch=hda-jack-retask.fw,hda-jack-retask.fw...`
<erlandsona> Thanks in advance to the person who can point me in the right direction!
<clever> erlandsona: and also look at the description of the other firmware related options in that file
<clever> it also looks like this already tells the kernel where to look
copumpkin1 is now known as copumpkin
copumpkin is now known as Guest68517
<erlandsona> Or even better clever: can you help me figure out why I'm getting http://nixpaste.lbr.uno/txZujDel
contrapumpkin has joined #nixos
<erlandsona> Basically hdajackretask is trying to run a script from a tmp directory with a #!/bin/bash and that's not getting picked up so I'm running the commands inside the script in my own shell environment, usually just calls to mv some generated files into fhs specific locations.
<clever> erlandsona: usualy, you would run patchShebangs on such a script to fix it
Guest68517 is now known as copumpkin
copumpkin has quit [(Changing host)]
copumpkin has joined #nixos
<erlandsona> Could I just change the shebangs in the script to something like #!/run/current-system/sw/bin/bash ?
<clever> "#!/usr/bin/env bash" is usualy more portable, but $PATH wont be set when the kernel runs something directly
<clever> so what you typed may be better
<clever> if bash isnt required, you can also #!/bin/sh
<erlandsona> Sorry the script is already referencing #!/bin/sh not #!/bin/bash...
magnetophon has joined #nixos
<erlandsona> Alright well I'll dig more into those scripts... looks like writeTextFile might be relevant?
<clever> and also writeScriptBin
<erlandsona> thanks clever:
<erlandsona> writeScriptBin spits out an executable I'm assuming? Fortunately I have access to config files, the script is just to put them in their proper places.
<erlandsona> clever: any idea why I'm getting device or resource busy from `echo 1 | sudo tee /sys/class/sound/hwC0D0/reconfig`?
<clever> check dmesg
<clever> and also, is pulseaudio in use?
<erlandsona> pulseaudio is probably in use... what am I looking for in dmesg?
<erlandsona> [ 5869.673473] snd_hda_codec_realtek hdaudioC0D0: hda-codec: reconfiguring [ 5869.673494] snd_hda_codec_realtek hdaudioC0D0: The codec is being used, can't reconfigure.
cheshircat has quit [(Quit: ZNC 1.6.5 - http://znc.in)]
<clever> you may need to close all programs using audio, including pavucontrol
<clever> and pavucontrol will also tell you what those programs are
<erlandsona> Ran pulseaudio --kill and was able to get it to go!
<erlandsona> clever: Thanks a ton!
<clever> ive also found that if the PA daemon crashes, several programs mess up in fun ways
<clever> skype will silently go offline, while claiming to be online
<clever> teamspeak will loose all audio, and take a minute to quit
<clever> chromium will loose the ability to capture from the mic
<erlandsona> God what have I gotten myself into.
<Infinisil> Sounds like fun
<erlandsona> Infinisil: So I'm literally running a quadrant in tmux one running hdajackretask, two: running `pkttyagent -p 14038` attached to the hdajackretask process..., three running alsamixer to make sure all my finagaling isn't screwing with the volumes... and four to check the scripts hdajackretask is generating to be able to run those commands manually to check the jack's I'm retasking are actually correct.
<erlandsona> Meanwhile using a youtube video to check sound. I just go the front two speakers working again but now no sub, and who knows if I've totally borked my mic / headphone jacks! Hahaha
<erlandsona> I feel like a mad scientist right now.
<clever> :D
<clever> a few months ago, i was patching pulseaudio to support all the output channels in my card
<clever> i had to hook up 2 sets of headphones and a sound bar
<erlandsona> kek the things we do for the benfits of Nix. Like who knew rollbacks would be worth all this trouble?
ndrei has joined #nixos
<clever> even outside of nix, pulse didnt support this
<clever> i had to patch the source of pulse to make it work right
hyphon81 has joined #nixos
<erlandsona> damn clever:
vaibhavsagar has quit [(Ping timeout: 240 seconds)]
eacameron has joined #nixos
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] therealpxc closed pull request #27248: kbd: fix unicode_start and unicode_stop scripts (master...kbd) https://git.io/vQ6N7
NixOS_GitHub has left #nixos []
eacameron has quit [(Ping timeout: 248 seconds)]
<Infinisil> Someone stop me, I'm constantly seeing Nix as the solution to the problems they're having in #linux
Fare has quit [(Ping timeout: 260 seconds)]
jmeredith has quit [(Quit: Connection closed for inactivity)]
marsel has joined #nixos
Rajsun has quit [(Quit: Leaving)]
cheshircat has joined #nixos
eacameron has joined #nixos
<chominist[m]> Has anyone tried haiku? How does that compare with nixos?
eacameron has quit [(Ping timeout: 260 seconds)]
filterfish has joined #nixos
mkoenig has joined #nixos
eacameron has joined #nixos
eacameron has quit [(Ping timeout: 248 seconds)]
eacameron has joined #nixos
eacameron has quit [(Ping timeout: 248 seconds)]
eacameron has joined #nixos
eacameron has quit [(Ping timeout: 240 seconds)]
eacameron has joined #nixos
erlandsona has left #nixos []
hellrazo1 has quit [(Ping timeout: 240 seconds)]
eacamero_ has joined #nixos
eacameron has quit [(Ping timeout: 248 seconds)]
eacamero_ has quit [(Ping timeout: 260 seconds)]
marsel has quit [(Ping timeout: 248 seconds)]
erlandsona has joined #nixos
ebzzry has quit [(Ping timeout: 240 seconds)]
<erlandsona> clever: or whoever? I'm trying to preserve a working config... basically I need a nix package representing / wrapping the file hda-jack-retask.fw either as the file itself or just the string inside it.? I'm planning to include it in the hardware.firmware list but it needs to be a package.
<erlandsona> ?
<clever> erlandsona: for hardware.firmware, it would need to be a package containing a /lib/firmware/hda-jack-retask.fw file, i believe
<erlandsona> yeh I don't know how to make a package? Is there a way to do it without the hashing thing? Like just a generic "I just want this file / string to be a part of the build" ?
<clever> hashing thing?
<erlandsona> sha256, last time I tried to write a derivation was to build tmux1.9, but then I was using fetchUrl and I generated a sha and overall had to do a lot, but that made sense for building an entire program like tmux. In this case the file is only 10 lines, I could use a string instead? But the extraModprobeConfig points to hda-jack-retask.fw...
<clever> do you already have hda-jack-retask.fw in a local file, or do you want to fetch it from a url?
<erlandsona> local file
<clever> then you can just do something like this
<clever> runCommand "name" {} ''
<clever> mkdir -pv $out/lib/firmware
<clever> cp -vi ${./hda-jack-retask.fw} $out/lib/firmware/hda-jack-retask.fw
<clever> ''
<erlandsona> Yup that's what I was looking for!
* Christian[m]1 wants to try getting to know apparmor on NixOs, smashing his head against not working at all apparmor-utils
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] jensbin opened pull request #27249: corebird: 1.5 -> 1.5.1 (master...corebird) https://git.io/vQief
NixOS_GitHub has left #nixos []
<Christian[m]1> starts with utils not finding their settings files, continues with settings containing paths to binries that are not present and now i'm stuck in python27 vs. python3 stuf
erlandsona has left #nixos []
Coyin has joined #nixos
<Christian[m]1> tried upgrading apparmor to 2.10.1 as this issue https://bugs.launchpad.net/apparmor/+bug/1513880 tells me that particular problem is fixed but it is not :/
ebzzry has joined #nixos
erlandsona has joined #nixos
<erlandsona> clever: Where does that config go? I put it inside hardware.firmware [(pkgs.runCommand "hda-jack-retask" {} 'shell things you mentioned...')] but I'm not seeing the file under /run/current-system/sw/lib/firmware/ ?
<Coyin> Hello everyone , is exist anyway to search package to install , faster than "nix-env -qaP mypackage", a way like in nox with a cache"
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] ck3d opened pull request #27250: emsciptenStdenv: allow reuse without ./autogen.sh (master...fix-emscriptenStdenv-autogen) https://git.io/vQiec
NixOS_GitHub has left #nixos []
<clever> erlandsona: did you do a nixos-rebuild switch?
<clever> Coyin: i just tab-complete in nix-repl '<nixpkgs>'
<erlandsona> clever: scratch that, its under /run/current-system/sw/firmware not run/current-system/sw/lib/firmware. MyB! Thanks again!
<clever> yep
erlandsona has quit [(Quit: Leaving.)]
erlandsona has joined #nixos
<erlandsona> clever: it's pathetic how happy I am to have the speakers on my laptop finally working. I've had this machine for almost 4 months now and now it finally makes sound out of all the speakers! And I didn't break the headphone port!
<erlandsona> Thank you for all your time, help and tips today!
<clever> yep
<clever> most of my laptops only have 1 speaker, cheap buggers
<erlandsona> Hehehe Well this laptop certainly saved me from having to buy a mac, and from what I've done with Bitwig-Studio on it so far and anything else, it just runs like a champ!
erlandsona has left #nixos []
<Coyin> clever , thanks but after that you have to "nix-env -iA nixos.mypackage", so no real alternative to nox
<clever> Coyin: yeah, and nox does have caching, but i think it rebuilds too often
<Coyin> clever thx
duzzy1 has joined #nixos
duzzy2 has quit [(Read error: Connection reset by peer)]
ebzzry has quit [(Ping timeout: 248 seconds)]
Infinisil has quit [(Quit: ZZzzz.z.zzz)]
<unlmtd> is 'xserver' going to remain as a historical option to setup a wayland display?
Ivanych has joined #nixos
ebzzry has joined #nixos
eacameron has joined #nixos
ebzzry has quit [(Ping timeout: 240 seconds)]
eacameron has quit [(Ping timeout: 255 seconds)]
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] jfrankenau opened pull request #27251: feh: 2.18.2 -> 2.19 (master...update-feh) https://git.io/vQivN
NixOS_GitHub has left #nixos []
ebzzry has joined #nixos
vaibhavsagar has joined #nixos
<vaibhavsagar> would anyone be able to help me fix broken tests on a haskell package?
<vaibhavsagar> I think it's a GHC Paths issue
hellrazor has joined #nixos
iyzsong has quit [(Read error: Connection reset by peer)]
ebzzry has quit [(Ping timeout: 240 seconds)]
iyzsong has joined #nixos
justelex_ has quit [(Ping timeout: 240 seconds)]
iyzsong has quit [(Quit: ZNC 1.6.5 - http://znc.in)]
Ivanych has quit [(Ping timeout: 260 seconds)]
wowi42 has joined #nixos
iyzsong has joined #nixos
<srhb> vaibhavsagar: Which one?
<vaibhavsagar> I'm trying to get IHaskell set up
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] vbgl pushed 1 new commit to master: https://git.io/vQiJq
<NixOS_GitHub> nixpkgs/master ee5a175 Karn Kallio: re2c : Fix build by adjusting sourceRoot.
NixOS_GitHub has left #nixos []
<srhb> vaibhavsagar: And that's the one with the broken tests?
<vaibhavsagar> it kind of works when I build it with release.nix
<vaibhavsagar> but I have a `dontCheck` in there
<vaibhavsagar> and when I remove it the tests break
<vaibhavsagar> I got the tests running successfully under `nix-shell` after futzing around manually
<vaibhavsagar> after I added the snippet here to my ~/.bashrc: https://github.com/NixOS/nixpkgs/issues/213#issuecomment-39615862
<srhb> Yikes, looks complicated.
<vaibhavsagar> indeed
Ivanych has joined #nixos
<srhb> vaibhavsagar: I wonder if you can just override the checkPhase for ihaskell to bring the right PATH in
<vaibhavsagar> thanks, let me look into that
mudri has joined #nixos
<srhb> vaibhavsagar: I took a look, and I'm afraid it's a bit beyond my skill level to give any meaningful advice. Sorry.
<vaibhavsagar> no worries :)
elninja44 has joined #nixos
mkoenig has quit [(Ping timeout: 240 seconds)]
simukis has joined #nixos
mkoenig has joined #nixos
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] aristidb pushed 1 new commit to master: https://git.io/vQiJw
<NixOS_GitHub> nixpkgs/master 19dfb63 Aristid Breitkreuz: sshuttle: 0.78.0 -> 0.78.3, use py3 for tests to succeed
NixOS_GitHub has left #nixos []
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] vcunat pushed 3 new commits to staging: https://git.io/vQiJ6
<NixOS_GitHub> nixpkgs/staging e7521e2 Volth: gitRepoToName: make it compatible with old fetchFromGitHub's "${repo}-${rev}-scr" to avoid mass rebuild
<NixOS_GitHub> nixpkgs/staging ab8dd33 Volth: support old naming, just add escaping of unsafe chars
<NixOS_GitHub> nixpkgs/staging d10c3cc Vladimír Čunát: Merge: more compatibility for git* fetchers...
NixOS_GitHub has left #nixos []
lambdael has quit [(Quit: WeeChat 1.7.1)]
<Christian[m]1> ok, my bad -- apparmor-utils is indeed fixed in 2.10.1, stupid me forgot to change the hash in src so that the new install changed the version in the package name but did not fetch different sources.
<aristid> why is my /etc/hosts a file and not a symlink? oO
justan0theruser has joined #nixos
justan0theruser has quit [(Client Quit)]
justan0theruser has joined #nixos
justanotheruser has quit [(Ping timeout: 248 seconds)]
sophiag has joined #nixos
<sophiag> hi. so...my box just crashed and gnome3 was freezing after login so to fix that i had to rollback all the way to 16.09. i tend to reboot very rarely relative to rebuilds, but would think i'd have between now and then. not sure what could have changed otherwise. any ideas?
<sophiag> i know there's not a huge nixos/gnome overlap
<sophiag> probably for good reason...
Filystyn has joined #nixos
Filystyn has quit [(Changing host)]
Filystyn has joined #nixos
Wizek_ has joined #nixos
<sphalerite> sophiag: not really a good reason, just not enough people using it to maintain it and debug issues like this. gnome-shell also crashes on a device with a touchscreen when it's touched
<sphalerite> sophiag: Could you get any useful information from the journal?
thc202 has joined #nixos
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] joachifm pushed 2 new commits to master: https://git.io/vQiUW
<NixOS_GitHub> nixpkgs/master 1df820c Johannes Frankenau: feh: 2.18.2 -> 2.19
<NixOS_GitHub> nixpkgs/master 079b495 Joachim F: Merge pull request #27251 from jfrankenau/update-feh...
NixOS_GitHub has left #nixos []
<sophiag> i couldn't launch a tty when all that was happening. i'm sure i could have edited the grub script to launch into a shell, but forgot how to :(
<sophiag> my channels still say 17.03...is that right?
<sophiag> i'm pretty shaky on how rollbacks work
pie_ has joined #nixos
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] joachifm pushed 1 new commit to release-17.03: https://git.io/vQiUr
<NixOS_GitHub> nixpkgs/release-17.03 a27b43c Johannes Frankenau: feh: 2.18.2 -> 2.19...
NixOS_GitHub has left #nixos []
<sphalerite> sophiag: if you have a persistent journal you can look at the logs from the previous boots. Does journalctl --list-boots list more than one?
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] joachifm pushed 2 new commits to master: https://git.io/vQiUo
<NixOS_GitHub> nixpkgs/master cf515b0 Jens Binkert: corebird: 1.5 -> 1.5.1
<NixOS_GitHub> nixpkgs/master 12216a6 Joachim F: Merge pull request #27249 from jensbin/corebird...
NixOS_GitHub has left #nixos []
<sphalerite> but if you couldn't even switch to a tty it might be a graphics driver issue. Are you on nvidia by any chance?
<sphalerite> aristid: that's a good question. On NixOS, right?
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] joachifm closed pull request #27193: qutebrowser: 0.10.1 -> 0.11.0 (master...upd_qutebrowser-0.11.0) https://git.io/vQr93
NixOS_GitHub has left #nixos []
<aristid> sphalerite: yup
mudri has quit [(Ping timeout: 260 seconds)]
<aristid> sphalerite: i think it may be because i tried sshuttle
<sophiag> yup, hand me down brick-like gaming laptop
nox_ has joined #nixos
<sophiag> is this a moment to link to that linus video? :)
<sphalerite> sophiag: are you using nvidia proprietary drivers?
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] vcunat pushed 1 new commit to master: https://git.io/vQiU9
<NixOS_GitHub> nixpkgs/master 986c177 Vladimír Čunát: Merge: more compatibility for git* fetchers...
NixOS_GitHub has left #nixos []
<sophiag> i haven't installed any
<sophiag> so presumably no
<sphalerite> aristid: I'm trying to work out where in the modules /etc/hosts normally gets generated but I can't even tell o.O
Turion has joined #nixos
<sophiag> i've noticed recently since fb added webgl support it's been crashing the tab in chrome, but i doubt that's related
<sophiag> oh, i pulled up the journal
<sphalerite> sophiag: so nouveau? Does lsmod | grep 'n.*v.*' say anything about nouveau or nvidia?
<sphalerite> great
<sophiag> what am i looking for in the journal?
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] joachifm closed pull request #27055: mpd service: Start when needed and harden (master...mpd-startWhenNeeded) https://git.io/vQ0xC
NixOS_GitHub has left #nixos []
<sophiag> it shows my 9 failed login attempts
<sophiag> it does seem like i haven't rebooted since before upgrading to 17.03...
<sphalerite> impressive
<aristid> sphalerite: search for ::1 in nixos/modules/config/networking.nix
<sophiag> ha
<sphalerite> You might want to get yourself on a newer kernel though :p
<aristid> sphalerite: but that just generates a regular symlink
<sphalerite> aristid: weird, nixos-option environment.etc.hosts.text didn't find that for me
<Turion> In the nixpkgs repository, there is a qgo package under packages/games, but I can't find out the correct attribute to install it
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] joachifm pushed 2 new commits to master: https://git.io/vQiUA
<NixOS_GitHub> nixpkgs/master 52ceb98 Pascal Bach: gitlab-runner_1_11: 1.11.4 -> 1.11.5
<NixOS_GitHub> nixpkgs/master c5fdfff Joachim F: Merge pull request #27155 from bachp/gitlab-runner-1.11.5...
NixOS_GitHub has left #nixos []
<sophiag> um, so your regex didn't just pull up nvidia drivers
<Turion> nix-env -qaP 'qgo.*' doesn't give anything
<sphalerite> sophiag: yeah, likely to include nvram and stuff as well. But do either of nvidia and nouveau show up?
<aristid> sphalerite: ok, it must've been some other application because nixos-rebuild switch restored it to a symlink :)
<sphalerite> Turion: are you on 17.03? It was only added recetnly and won't be in 17.03 yet
<Turion> sphalerite: Oh, yes!
<sphalerite> The attribute is simply qgo
<sophiag> sphalerite: one sec. also btw, kernel is 4.9 so not too old
<Turion> Ok :)
<sphalerite> sophiag: the major version doesn't matter as much as the minor versions
<sophiag> ok, i see 4 nouveau drivers
<sphalerite> 4? Huh
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] vcunat pushed 1 new commit to master: https://git.io/vQiTJ
<NixOS_GitHub> nixpkgs/master 3a891c6 Vladimír Čunát: re2c: fix sourceRoot in a more robust way...
NixOS_GitHub has left #nixos []
<sophiag> so, you're saying i should bump to 4.12? i'm afraid if i do i'll just have to rollback again
<sphalerite> no, but from 4.9.xy to the latest 4.9 minor
<sphalerite> nixos-rebuild switch can upgrade most stuff, but you can't upgrade the kernel without rebooting
<sophiag> sphalerite: button, i2c_algo_bit, drm, ttm, mxm_wmi
<sophiag> so 5
<sphalerite> that's just in general, not sure it will help with your gnome issue but who knows
<sphalerite> I'd definitely reboot after an OS version upgrade
<sophiag> oh you meant patch version...
<sphalerite> oops yes
filterfish has quit [(Ping timeout: 240 seconds)]
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] 0xABAB opened pull request #27253: wine: add missing dependency libcap on Linux (master...patch-4) https://git.io/vQiTL
NixOS_GitHub has left #nixos []
<sophiag> if memory serves me i may have not rebooted since upgrading to 4.9 and that may have been a major version upgrade...
<sophiag> is that possible? how old was 3?
<sphalerite> Quite
mudri has joined #nixos
<sophiag> i think we're leaning towards driver issue seeing as it was just a gnome thing, right? i just have no idea what those nouveau drivers do. seem mostly irrelevant
<sphalerite> but AFAIU the major version doesn't mean much in the kernel anyway, it's more of a "Linus feels like it" thing
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] FRidh pushed 1 new commit to master: https://git.io/vQiTO
<NixOS_GitHub> nixpkgs/master bd129c2 Karn Kallio: unzipNLS : Fix build by removing patch fuzz.
NixOS_GitHub has left #nixos []
<sophiag> well, i'll bump to 4.12 after fixing any potential nvidia stuff
<sophiag> speaking of linus feels like it :)
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] LnL7 pushed 2 new commits to master: https://git.io/vQiTs
<NixOS_GitHub> nixpkgs/master 1f70f38 Yacine Hmito: Make zshrc more predictable...
<NixOS_GitHub> nixpkgs/master 8189811 Daiderd Jordan: Merge pull request #25648 from yacinehmito/custom...
NixOS_GitHub has left #nixos []
<sphalerite> sophiag: well the main reason I asked is because nvidia's proprietary drivers tend to horribly break stuff, but if you're on nouveau I think it's less likely to be that
<sophiag> ah i see
<sphalerite> no big reason to upgrade to 4.12 really, but I would uppgrade to the latest patch version
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] FRidh closed pull request #27214: unzip: Use less lines of context in patch so NLS patch commutes (staging...unzip-nls) https://git.io/vQoNa
NixOS_GitHub has left #nixos []
<sophiag> i'm scanning visually at a list of ~50, but don't see any nvidia. besides wouldn't i had to have installed a proprietary driver?
<sophiag> i woudl certainly remember that
<sphalerite> yes, you would
<sophiag> ha
<sphalerite> and if you have nouveau you can't have the proprietary driver
<sophiag> well, would you recommend ugrading the kernel and then rebooting on the new build?
justelex_ has joined #nixos
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] FRidh closed pull request #27253: wine: add missing dependency libcap on Linux (master...patch-4) https://git.io/vQiTL
NixOS_GitHub has left #nixos []
<sphalerite> I would absolutely recommend rebooting in general :)
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] FRidh closed pull request #27247: wine: add missing dependency (master...wine) https://git.io/vQ6Fq
NixOS_GitHub has left #nixos []
<sophiag> oh, and for the future...i'm about to optimize a library and unsure the linear algebra library i'll be using calls CUDA. would i need proprietary drivers for that vs. opencl?
Turion has quit [(Ping timeout: 248 seconds)]
ennui has joined #nixos
<sophiag> it may be irrelevant, but not sure. i'll likely only use the features that call BLAS level one. but i'm unsure how this library works as far as all that. it's a clojure thing
<sphalerite> I think you might
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] LnL7 closed pull request #26716: Generalize building of Elixir interpreter (master...generalize-elixir) https://git.io/vQvLV
NixOS_GitHub has left #nixos []
<sophiag> sphalerite: but i'm here now (instead of sleeping...) beacuse a reboot caused me to have to rollback to 16.09...
<sphalerite> but I'm almost sure that gnome won't work on nvidia proprietary
<sphalerite> oh right
<sophiag> good to know. i should transition to just xmonad with no desktop manager anyway
<pie_> oh god the version of tor in nixpkgs is REALLY OLD
<aristid> sphalerite: why wouldn't gnome work on nvidia proprietary? oO
<aristid> pie_: pull request it!
<sophiag> i sort of doubt the kernel is what's screwing with gnome, but who knows? waiting a long time between reboots adds too many variables. not smart of me :/
<pie_> either that or i did something wrong xD
<pie_> aristid, yeh :D
<pie_> aristid, maybe i should have grabbed from unstable, thats a bit better
<sphalerite> aristid: something something wayland something nvidia sucks
<sophiag> should i just link to the video???
<aristid> pie_: tor was upgraded on july 1, in unstable :)
<aristid> sphalerite: you can use gnome with X11, though?
<srhb> sophiag: Can't hurt.
filterfish has joined #nixos
filterfish has quit [(Remote host closed the connection)]
filterfish has joined #nixos
<sophiag> oh wait, should i not specify a kernel version with linuxPackages?
<sophiag> does that override the one nix should be giving me with 17.03?
<sophiag> ugh and now i'm reminded why i did this upgrade...dropbox is broken
<nox_> Hello everyone I'm trying to write a nixpkgs overlay to add a kernel module, but I have quite a limited understanding of nix and nixos :) I would like to "extend" the linuxPackagesFor function from all-packages.nix in the overlay but it may not be possible. And even if I don't I would still need access to the "kernel" variable inside the overlay but I don't know how.
Ivanych has quit [(Ping timeout: 260 seconds)]
<nox_> The configuration (I am building a cd iso but it shouldn’t change anything) : http://nixpaste.lbr.uno/m4GdiZaQ?nix
<nox_> And the kernel module code : http://nixpaste.lbr.uno/zFiwkN87?nix
<srhb> sophiag: It does override it, yes.
<srhb> sophiag: Which you may or may not want
<sophiag> hmm
<sophiag> does anyone know what 17.03 defaults to?
<sophiag> also linuxPackages_4_12 can't be found
<srhb> sophiag: 4.9.36 it looks like
<sophiag> oh, so i was likely overriding down a patch version
<sophiag> that's damn stupid
<sophiag> ok, lemme give that a shot
<srhb> sophiag: Probably not catastrophically so :)
<sophiag> if not i think i'm back to 16.09 and going to sleep
<srhb> sophiag: yay for rollbacks. :)
<sophiag> srhb, i'm only awake in here because i had to roll back months
<srhb> sophiag: Ow :/
<sophiag> so we assume this is the culprit
<sophiag> gnome was frozen after login
<sophiag> i have no prop nvidia drivers
<sophiag> not sure what else it could be
<srhb> It doesn't sound convincing to me, but worth a shot I guess.
<sophiag> yeah me neither...
<srhb> sophiag: I do see _4_12 in nixos-17.03 though, so that's a bit odd.
<cmcdragonkai> Why does nix-env --query --installed -P give numbers on the left column instead of the attribute paths of installed packages?
<sophiag> but unless you have any ideas
<sophiag> uh, rebuild is lagging on "failed to stop -.slice"
<pie_> tor-browser on unstable seems pretty broken :/ tabs keep crashing
<sophiag> oy
<sophiag> well i deleted that line and rebuilt
<sophiag> i rolled back to 16.09 but my channels still said 17.03
<srhb> sophiag: I don't think channels are affected by rollbacks? Not sure though.
<srhb> sophiag: I don't use them.
<sophiag> appears not
<sophiag> anyway, i suppose i'll try to restart :(
<sophiag> otherwise have to knock out
<srhb> sophiag: Good luck :)
<pie_> disabling multiproces firefox stuff seems to fix it (browser.tabs.remote.autostart)
spinus has joined #nixos
sophiag has quit [(Ping timeout: 260 seconds)]
nox_ has quit [(Ping timeout: 260 seconds)]
mudri has quit [(Ping timeout: 240 seconds)]
sophiag has joined #nixos
<sophiag> welp, no luck
nox_ has joined #nixos
<sophiag> but i'm unsure because when i pulled up the builds i didn't see one from today when i got rid of the kernel override. does it omit the latest build from the list?
slyfox has quit [(Remote host closed the connection)]
<symphorien> sophiag: there is nix-channel --rollback i think
<srhb> sophiag: The boot list? No, it shouldn't.
<srhb> sophiag: Assuming you did nixos-rebuild boot
<sophiag> ah, no i did switch
<sophiag> i always do switch
<srhb> It also adds a boot entry.
<sophiag> right, that's why my others were there
Turion has joined #nixos
<sophiag> not sure about the one from right now
<srhb> No, that sounds odd.
<srhb> Did you accidentally forget to sudo ?
<srhb> That should give you an error, but I've missed it on occasion.
slyfox has joined #nixos
<sophiag> nope
<srhb> Hm. No idea then.
Ivanych has joined #nixos
<sophiag> not sure if it could be because i was rollbacked when i rebuilt?
<srhb> I still think the new generation with the new kernel package would get a new generation number.
<sophiag> oh, i just tried rebuilding again and it's lagging on that slice error again
<srhb> As in, failing to alter some systemd slice?
<srhb> Probably because the process is hanging.
<srhb> Might be able to coerce it through pkill
<sophiag> "Failed to stop -.slice: Operation refused,"
Coyin has quit [(Ping timeout: 248 seconds)]
<srhb> Yeech.
<sophiag> and then i cmd-c
<sophiag> and then rebuild again
<sophiag> seems fine
<srhb> OK.
<sophiag> but doesn't actually rebuild
<sophiag> that's the problem
<srhb> Probably because it already did
<srhb> It just failed some final systemd service changes
<sophiag> nope, not in the log
<srhb> As in, restart, stop, etc.
<srhb> Perhaps try sticking with nixos-rebuild boot while that service is acting up, it will be less confusing. But if it doesn't rebuild on nixos-rebuild switch, then it's already done.
<sophiag> ok, this is not the issue
<srhb> (Or we have a serious bug)
<sophiag> the issue is using 17.03 with gnome
* srhb nods
<sophiag> i actually can't use 17.03 with a whole ton of packages
<sophiag> because virtualbox has been broken for over a month so they're all outdated
<sophiag> but on 16.09 i can't use dropbox period
<sophiag> i'm uploading with the web interface right now
<srhb> Perhaps you'll be better of cherry picking dropbox from 17.03
<srhb> I didn't notice virtualbox being broken though
<srhb> I think my work laptop is on 17.03 and has virtualbox.
<sophiag> i mean the virtualbox build of nix...
<srhb> Oh!
<sophiag> this package from may 23rd is not even in the testing branch: https://github.com/NixOS/nixpkgs/issues/26141
<sophiag> there hasn't been an update to even the testing branch in 1.5 months...
<sophiag> it's getting to a point where i need to choose to be a full time nix developer or switch oses :(
<srhb> sophiag: What testing branch? nixos-unstable?
<sophiag> no, testing
<sophiag> i think?
<srhb> I'm not sure what testing is.
<srhb> But 9f4a10220 is in nixos-unstable for sure.
<sophiag> ok, maybe i have it the other way around
<sophiag> oh, maybe i haven't checked in a week or two
<sophiag> good to know
<sophiag> except i can't even use 17.03 now. awesome
mudri has joined #nixos
<srhb> Sorry, no idea how to help with Gnome.
<sophiag> it's hard for me to diagnose...there are 99 open issues related to gnome
<sophiag> some are four years old...
<sophiag> there's a two year old one marked severe:security
<sophiag> yikes
<sophiag> "Many of our NixOS services run as root and shouldn't"
<sophiag> glad to know it can get much much worse, i guess?
<sophiag> though to be fair, all the gnome issues that sound like mine are nvidia related and i definitely never installed any prop drivers
<srhb> I don't know. Is there nothing useful in your logs?
<sophiag> yup, went through this right before you came on
<srhb> Ah, sorry.
<sophiag> no, i appreciate your help
<sophiag> i'm too cranky frankly
<sophiag> need to deal with this tomorrow
<srhb> Having a broken system tends to do that to people.
<sophiag> i should probably try to start a local nix meetup
<srhb> Where are you located?
<sophiag> get the folks who convinced me to use it to go
<sophiag> nyc
<srhb> Ah, sounds doable. :)
<sophiag> well, folks at a consultancy that uses it for work convinced me to give it a try this past winter so hopefuly they'd be down
<sophiag> that's what it'd take for me to have a system that's not constantly half broken tho
<srhb> I wonder why you experience that. I've never had any issues, really.
<srhb> Though I've heard a lot of people complain about Gnome.
<sophiag> i have to ditch gnome
<sophiag> plus this is a stupid gaming laptop
<srhb> Might be a viable workaround.
<sophiag> they really only play nicely with windows
<srhb> I'll happilly champion xmonad sans gnome :-)
<sophiag> yeah, i've tried xmonad a little
<srhb> And that's a good tip, I was considering a gaming laptop next time.
<srhb> Might have to rethink that
<sophiag> i only use 2 apps normally
<sophiag> emacs (right now) and chrome
<sophiag> sometimes spotify
<sophiag> but that's it for the gui
<srhb> That's about what I do too, plus terminals.
<srhb> (And, well, steam and games)
<sophiag> gaming laptops are just a bad idea in general
<sophiag> worst of both worlds
<sophiag> why not get a really light notebook and build a desktop?
<srhb> I want to be able to bring it to bed. I'm weird like that. :-P
<sophiag> ha
<srhb> Am considering an eGPU instead though
<sophiag> hmm?
<srhb> An external gpu driving the internal lcd
<srhb> For when I need the power
<sophiag> ah right. that's popular now
<sophiag> like that razor addon
<sophiag> that's the thing...i wouldn't want things like an optical drive, dvd, etc. on a laptop
<sophiag> you can still have a nice graphics card and not have it be heavy and no battery life tho
<srhb> I just want a really powerful CPU, lots of memory and the option of gaming grade graphics
<sophiag> i suppose there's a difference between like a razor and a macbook pro
<srhb> Oh, and a good screen.
* srhb nods
<sophiag> yeah the razor screen is amazing
<sophiag> but the battery life...
<srhb> My battery has literally zero seconds capacity by now.
<srhb> It's fine...Fiiiine.
<srhb> :-P
<sophiag> haha
<sophiag> i feel literally like fire dog when i have this thing on my lap
<srhb> Ow :-P
<sophiag> my friend says i could use it as a hot plate
<srhb> To be fair the T530 isn't exactly lap-friendly either
<sophiag> honestly, i miss my five year old mac air
<srhb> I'd buy an X series as replacement if the screen and touchpad weren't so shitty...
<sophiag> when integrated graphics came out i was all butthurt and then i appreciated them for coding webgl and knowing how it'd render on a phone
<srhb> Ah. :)
<sophiag> anyway, idk what i'd get now. i'm not ever buying an apple product again
<srhb> I'm waiting for next year. But you sorta ruined my razor plan now.
<srhb> Oh well.
<srhb> :-)
<sophiag> but i'd like a nice screen. not razor nice, but retina nice. adn probably a graphics card
<sophiag> but still super lite and long battery life
<srhb> Razer*
mpcsh has quit [(Quit: THE NUMERICONS! THEY'RE ATTACKING!)]
<sophiag> ha
<sophiag> anyway, i've been up for over 24 hours :((
<srhb> Go sleep!
<srhb> Perhaps it will be easier to find a solution tomorrow
mpcsh has joined #nixos
<sophiag> i suppose i'll whine to someone on the dev team if i can
<sophiag> they're usually quite nice and helpful
<srhb> Sounds like a plan
<nox_> (I resend my message to get visibility, sorry if you’ve already seen it) Hello everyone I'm trying to write a nixpkgs overlay to add a kernel module, but I have quite a limited understanding of nix and nixos :) I would like to « extend" the linuxPackagesFor function from all-packages.nix in the overlay but it may not be possible. And even if I don't I would still have to access to the "kernel" variable inside the overlay but I don't k
<nox_> building a cd iso but it shouldn’t change anything) : http://nixpaste.lbr.uno/m4GdiZaQ?nix The module code : http://nixpaste.lbr.uno/zFiwkN87?nix
mudri has quit [(Ping timeout: 248 seconds)]
filterfish has quit [(Ping timeout: 255 seconds)]
pie_ has quit [(Changing host)]
pie_ has joined #nixos
hotfuzz has joined #nixos
<unlmtd> the tree is getting pretty big, and it doesnt cover all the systems even yes. has there been talk of breaking it up to better allow it to expand?
<unlmtd> it might make it easier to manage issues too
hotfuzz_ has quit [(Ping timeout: 240 seconds)]
<srhb> unlmtd: Sounds harder, you'd still need to collect all the subtrees to build basically anything.
<srhb> And then you're back to what we have.
<srhb> Besides, it's not really a big deal compared to, say linux.
<unlmtd> I mean there could be different trees for different {kernel,libc,init}
<srhb> unlmtd: I don't see how that would be a good idea.
<unlmtd> ok
<unlmtd> so theyd basically be independant forks that borrow/exchange
<srhb> Yes, I get what you're saying, but I don't understand the benefit.
<srhb> Perhaps you could explain that.
<unlmtd> I want to use nix to basically build like alpine does
<srhb> unlmtd: How does alpine build?
<unlmtd> different clib, different init
<unlmtd> usually more minimal build options
<unlmtd> more hardened
<unlmtd> so that tree would be pretty different a lil bit everywhere
<nox_> If it can be useful for someone I made my code work by passing kernel = self.linuxPackages_latest.kernel to the package in the overlay but it's still far from ideal because the compatible kernel version is hardcoded
<srhb> Different init is basically an entirely new NixOS. Not that that would be a bad thing, but I doubt there's anyone willing to take up that challenge.
<unlmtd> but maybe that can be abstracted down to a single tree for every clib
<srhb> clib is easier, I think.
<unlmtd> lol, well i guess your chatting with him now
<unlmtd> :P
<srhb> Oh, cool. :)
<unlmtd> linux for me is stuck, but nixos is cool, so I want to use it on other kernels, so best start with abstracting the init
filterfish has joined #nixos
<srhb> Well, if you can show some work that details how it would ease your job of abstracting out systemd, that might motivate people. :)
<unlmtd> id probably start by offering infrastucture
<unlmtd> so hydra and binary caches and a website i guess
<unlmtd> I could start with clever's repo
<unlmtd> how does nixos host their infrastructure? amazon?
<srhb> I think Hydra is sponsored by University of Delft.
<unlmtd> nice
<srhb> Channels are hosted on S3, I think.
<srhb> But I don't know what kind of bribe would be sufficient to get the ball rolling. I would think you'd have to show how it would benefit NixOS(systemd) first.
<unlmtd> it must be relatively cheap/easy to maintain different channels with nixos
<unlmtd> bribe who?
<unlmtd> I dont wanna bribe anyone
<srhb> That was tongue in cheek
<unlmtd> :P
<srhb> I meant convince people that abstracting out systemd is a good idea
<unlmtd> i dont need to
<unlmtd> theres lots already
<srhb> Not people who actually work on it, I think :)
<unlmtd> sure there are
<srhb> OK. :)
<aristid> looks like the wireguard nixos module stores the private keys in the nix store /o\
<unlmtd> theyre just disperssed we need to bring the efforts together
<srhb> unlmtd: Good plan.
<unlmtd> thx
<nox_> Is it possible to extend a function in nix in some way ? r are there other place/time in the day where or when I can get help :) ?
<srhb> nox_: Sunday is often quiet. This time yesterday was very active.
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] aristidb pushed 1 new commit to master: https://git.io/vQiIH
<NixOS_GitHub> nixpkgs/master 01557ee Aristid Breitkreuz: sshuttle: clean up expression a little bit
NixOS_GitHub has left #nixos []
magnetophon has quit [(Ping timeout: 240 seconds)]
<nox_> shrb: thanks, I will try another day :)
<srhb> nox_: Sorry I can't help. :)
<LnL> nox_: it's in the linuxPackages attrset
<LnL> you can use something like callPackage ./foo.nix { inherit (linuxPackages) kernel; } to add it to the callPackage scope
marsel has joined #nixos
<nox_> LnL: That means that if I use linuxPackages.kernel instead of linuxPackages_latest.kernel (with boot.kernelPackages = pkgs.linuxPackages_latest in my configuration) it will automatically use linuxPackages_latest instead of linuxPackages_4_9 ?
<LnL> a nixpkgs overlay doesn't know anything about your nixos configuration
Turion has quit [(Ping timeout: 258 seconds)]
filterfish_ has joined #nixos
filterfish has quit [(Read error: Connection reset by peer)]
filterfish_ has quit [(Remote host closed the connection)]
filterfish_ has joined #nixos
<LnL> but you can use config.boot.kernelPackages if it's in a nixos module
filterfish_ has quit [(Remote host closed the connection)]
filterfish_ has joined #nixos
sibi has joined #nixos
<nox_> LnL: That's why I wanted to directly do something like extend linuxPackagesFor function in all-packages.nix to make my package work for every kernel version and to only specify the kernel version in boot.kernelPackages and boot.extraKernelModules options. But at least I have a working solution and use inherit now, which is cleaner :))
<LnL> oh you mean like adding it to the linuxPackages_* attributes?
<nox_> LnL: yes, exactly
nwspk has quit [(Quit: Quit: *.banana *.split)]
<nox_> nox: But it's only if I have to redistribute my code to other people who want to conveniently use other kernels
nwspk has joined #nixos
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] bachp opened pull request #27256: WIP: squid service: intial service based on default config (master...squid-service) https://git.io/vQiLu
NixOS_GitHub has left #nixos []
Turion has joined #nixos
ebzzry has joined #nixos
Thra11 has joined #nixos
ebzzry has quit [(Ping timeout: 248 seconds)]
<LnL> nox_: and I think this works if you want to extend all the versions https://gist.github.com/LnL7/33232b7310af2de12c793c4d27369a4c
magnetophon has joined #nixos
Legenda has joined #nixos
Turion has quit [(Ping timeout: 248 seconds)]
tokudan has joined #nixos
<nox_> LnL: Thank you very much, I think it is working (i'm testing it in a vm so it is quite slow) :)
<Gravious> oh dear
<Gravious> how do i wrap a program to include its own path without triggering a recursive error
* Gravious has an idea
Legenda has quit [(K-Lined)]
<vaibhavsagar> can anyone help me debug a weird test/packaging failure with a haskell package?
<spinus> vaibhavsagar: what's up
<vaibhavsagar> Hey spinus, I'm trying to get the IHaskell tests to pass
<Gravious> oh
<Gravious> warpping a binary does not do what i expected
<vaibhavsagar> and I've got it mostly working except for one test
<spinus> vaibhavsagar: trying
<spinus> ah
<Gravious> i expected the creation of a shell script that set some env vars before calling the binary
<spinus> have an error?
<Gravious> instead, i have another binary?
<vaibhavsagar> so I have a release.nix here: https://github.com/vaibhavsagar/IHaskell/tree/nix
<vaibhavsagar> and if I do `nix-build release.nix` the first test in the Eval module fails by somehow getting output from the Parser module tests
<vaibhavsagar> very bizarre failure IMHO
<Gravious> i could manually create such a wrapper i suppose, but it seems like i shouldn't need to do that
<spinus> but where do you have release.nix? (I don't see in the repo)
<spinus> ah
orivej has quit [(Ping timeout: 240 seconds)]
pie__ has joined #nixos
<Gravious> oh, wrapProgram is working fine, the program calling it is calling the wrapped program instead of the wrapper, cool
<spinus> still compiles
pie__ has quit [(Remote host closed the connection)]
pie___ has joined #nixos
pie___ has quit [(Changing host)]
pie___ has joined #nixos
<vaibhavsagar> I had a `dontCheck` in there earlier
<vaibhavsagar> I just really want these tests to pass
<spinus> nah, tests are overrated, we don't need them :)
<spinus> vaibhavsagar: src/tests/IHaskell/Test/Eval.hs:54:
<spinus> 1) Code Evaluation evaluates expressions
<spinus> expected: "3"
<spinus> that failed for me, same for you?
pie_ has quit [(Ping timeout: 240 seconds)]
<vaibhavsagar> yes, and look what it got instead
<spinus> chunk of text
<vaibhavsagar> the contents of the IHaskell.Test.Parser module
<vaibhavsagar> which makes no sense to me
<spinus> but it's not exactly content of the module, it looks like it's hspec descriptions
<vaibhavsagar> yup, without the 'describe' and 'it'
<vaibhavsagar> but why is this happening?
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] Profpatsch closed pull request #14598: unixODBCDrivers: fix broken module invocation (master...odbc-drivers-overhaul) https://git.io/vV7iq
NixOS_GitHub has left #nixos []
<spinus> probably broken test, maybe variable shadowing
<spinus> checking
<vaibhavsagar> thanks
<vaibhavsagar> let me try giving each test its own hspec
<spinus> vaibhavsagar: you could run hspec with pattern to select only 1 test
<vaibhavsagar> I don't know how to do that inside Nix
<vaibhavsagar> you mean pop open a nix-shell and debug from there?
<spinus> nah, just cabal :)
<spinus> or whatever ghci
<vaibhavsagar> I'm not seeing this error when I use stack or a nix-shell
<vaibhavsagar> and it only started when I did my path stuff in release.nix
<vaibhavsagar> before that the tests were mostly failing because IHaskell.Display wasn't visible on the package path
<spinus> I see
<spinus> vaibhavsagar: I'll try to play with it and rerun it few times, grab me in an hour or so
<vaibhavsagar> sure, no rush
<vaibhavsagar> thanks for your help :)
<vaibhavsagar> If you want to try nix-shell there's a shell.nix at this branch https://github.com/vaibhavsagar/IHaskell/tree/nix-test-failure
<vaibhavsagar> but it only works if you have GHC Path rewriting stuff in your .bashrc as per https://github.com/NixOS/nixpkgs/issues/213#issuecomment-39615862
<spinus> thx
<spinus> ok, why actually you need that rewriting?
nix-gsc-io`bot has joined #nixos
<nix-gsc-io`bot> Channel nixos-17.03-small advanced to https://github.com/NixOS/nixpkgs/commit/a27b43c2af (from 4 hours ago, history: https://channels.nix.gsc.io/nixos-17.03-small)
nix-gsc-io`bot has quit [(Client Quit)]
oaeirjg has joined #nixos
Turion has joined #nixos
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] Mic92 closed pull request #27235: btrfs-progs : fix bash-completion (master...fix-25793) https://git.io/vQ6t4
NixOS_GitHub has left #nixos []
ilyaigpetrov has joined #nixos
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] Mic92 pushed 2 new commits to master: https://git.io/vQiYf
<NixOS_GitHub> nixpkgs/master c4ecdfd Yann Hodique: gx: 20160601-f84ddf7 -> 0.12.0
<NixOS_GitHub> nixpkgs/master aacb816 Jörg Thalheim: Merge pull request #27213 from sigma/pr/gx...
NixOS_GitHub has left #nixos []
nox_ has quit [(Ping timeout: 260 seconds)]
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] hectorj opened pull request #27257: zipkin: init at 1.28.1 (master...init-zipkin-server-1.28.1) https://git.io/vQiYl
NixOS_GitHub has left #nixos []
Turion has quit [(Quit: Leaving.)]
Fare has joined #nixos
phinxy has joined #nixos
proteusguy has quit [(Ping timeout: 240 seconds)]
proteusguy has joined #nixos
bennofs has joined #nixos
oaeirjg has quit [(Ping timeout: 248 seconds)]
ericsagnes has quit [(Ping timeout: 255 seconds)]
<vaibhavsagar> oops, sorry spinus
<vaibhavsagar> I think because the interpreter uses ghc-paths to access libraries
MercurialAlchemi has joined #nixos
<vaibhavsagar> you can confirm this by loading IHaskell.Eval.Evaluate in `cabal repl` and running `testEvaluate "7"`
<vaibhavsagar> it complains about a missing IHaskell.Display because it can't find it in the libdir
<vaibhavsagar> although I'm getting the same error even after I have the rewriting... interesting.
<spinus> vaibhavsagar: I was playing with it for a moment, it's strange to me that you needed to add those paths, I tried to add IHaskell.Display as dependency to the main package but no luck
<vaibhavsagar> IHaskell.Display is included in ihaskell
<spinus> ah
<vaibhavsagar> I think it expects to already be loaded when it's running those tests
<spinus> I thought I had to add ihsakell-blaze as dependency to ihaskell
94KAA1N6F has joined #nixos
<94KAA1N6F> [nixpkgs] Mic92 closed pull request #27246: ion: init at 1.0.3 (master...ion) https://git.io/vQ659
94KAA1N6F has left #nixos []
18WAALHV2 has joined #nixos
<18WAALHV2> [nixpkgs] Mic92 pushed 2 new commits to master: https://git.io/vQiOo
<18WAALHV2> nixpkgs/master 3483c09 dywedir: ion: init at 1.0.3
<18WAALHV2> nixpkgs/master 154efd8 Jörg Thalheim: Merge pull request #27246 from dywedir/ion...
18WAALHV2 has left #nixos []
<vaibhavsagar> ihaskell-blaze is an IHaskell.Display instance for pretty-printing Blaze HTML in an IHaskell notebook
<vaibhavsagar> none of those things are strictly required
<spinus> ok
<vaibhavsagar> minimal build should have ghc-parser, ipython-kernel, and ihaskell
<joepie91> error: infinite recursion encountered, at /nix/var/nix/profiles/per-user/root/channels/nixos/nixpkgs/lib/modules.nix:151:28
<joepie91> woohoo...
<vaibhavsagar> I've been given commit access to IHaskell but I don't know to create a new release or if I should
<LnL> urgh, I still didn't configure this properly
ericsagnes has joined #nixos
<LnL> what's the correct way to add a secondary ip?
phinxy has quit [(Quit: Leaving)]
<joepie91> well, this is a fantastically useless stacktrace
magnetophon has quit [(Read error: Connection reset by peer)]
<joepie91> unless I'm overlooking something, it doesn't even include any frames in my own code...
<mkoenig> anyone using nixops with EC2? deployment doesn't work. "wating for SSH" all the time...
<gchristensen> incredible, joepie91
<gchristensen> mkoenig: sounds like your security groups don't permit ssh
<joepie91> gchristensen: is there something I'm missing, or would this be a valid "useless stacktrace" bug?
<joepie91> it feels incomplete
<mkoenig> gchristensen: hey, thanks, you saved me an hour!
<gchristensen> while evaluating the module argument `pkgs' in ":anon-2:anon-1:anon-1": pretty sure this is wherer the rubber meets the road, and you have a bunch of anon funcs
<gchristensen> maybe package overrides?
<joepie91> and... how do I go about finding those?
<joepie91> lol
<joepie91> nope, no package overrides
<gchristensen> I dunno :)
<joepie91> and this is in a nixops deployment spec
<gchristensen> mkoenig: cool!
<joepie91> I feel like anonymous functions should probably still include a line number...
<gchristensen> yeah
<gchristensen> by all means, do report :D
<gchristensen> nix needs better errors
<joepie91> very yes. :P
<joepie91> problem is, nothing to report until I figure out where the error is...
<joepie91> are there any tools to help me with this at all?
<gchristensen> I dunno, I'd go report it as a UX problem
<gchristensen> as is
<joepie91> wouldn't the first question be "testcase please"?
<gchristensen> no
<joepie91> hm, okay
<gchristensen> I mean, I wouldn't need one to realize the problem and have some ideas at solutions
<gchristensen> if you come up with a test case, please add it too your issue though
<gchristensen> test cases are critical for execution bugs, IMO less critical for experience bugs
JagaJaga has joined #nixos
<gchristensen> cool
<gchristensen> now I just need to go learn C++ to figure out how to help
<joepie91> gchristensen: commented out an imports entry, now I get a different error...
ndrei has quit [(Quit: WeeChat 1.7.1)]
<joepie91> or well
<joepie91> same error, different stack
<joepie91> looks like this is going to be a wild goose hunt :)
<gchristensen> that is a problem with fixed points I think, they can make the stack traces difficult to grok
<joepie91> gchristensen: it would be a lot easier if it gave me the line number! :P
<gchristensen> :)
magnetophon has joined #nixos
ixxie has joined #nixos
GRBurst has joined #nixos
Ivanych has quit [(Ping timeout: 258 seconds)]
magnetophon has quit [(Read error: Connection reset by peer)]
<catern> hey #nixos, if I did import "${fetchurl {...}}/foo.nix"
<catern> would that force the fetchurl to happen every time I used any of the attributes of the resulting value?
<catern> (hmm is it even possible to import from a fetchurl in that way?)
<gchristensen> you can on your system but not on hydra
<catern> what I want to do is, I want to write a expression which downloads a tarball for which I have a hash, evaluates the default.nix in there, and gives me the resulting store path
<catern> and I want it to be cached
<gchristensen> and no because fetchurl will cache by the sha256
<LnL> yes that works and a fixed output drv like fetchurl will only be downloaded once
jonte has joined #nixos
<catern> ah okay right
<catern> but what I'm really concerned about is caching the evaluation of foo.nix or default.nix
<catern> (because that's some lengthy build)
<LnL> builtins.fetchTarball doesn't have a hash so that will redownload every time
<chominist[m]> This is a weird question
<catern> yeah, assume I'm using some fetchurl that does have a hash :)
GRBurst has quit [(Ping timeout: 240 seconds)]
<chominist[m]> Where can I find a mapping of key to symbol/encoding for xkb?
<chominist[m]> I am trying to swap semicolon/colon with single/double quotation marks
<catern> chominist[m]: do the .xml files in /usr/share/X11/xkb (on a conventional distro) help?
<catern> chominist[m]: also try the "xev" command
<chominist[m]> catern (IRC): I can't find /usr/share
<chominist[m]> Oh
<chominist[m]> Right
<chominist[m]> Will give that a shot
GRBurst has joined #nixos
<catern> LnL: gchristensen: so yeah my question is, if I load a nix expression from a fixed output derivation A, and then that nix expression evaluates to a derivation B (which is a build of something, or whatever), will derivation B be cached?
<gchristensen> it will, but will nix reevaluate it? I don't know
<catern> hmmm
<gchristensen> if something is cached can only be known after it has been evaluated
<gchristensen> will nix, within a single execution, remember if it is cached? I guess depends on how you do it, I don't know these nitty gritty details
<catern> what I'm ultimately thinking about is, say Nix is super successful and everyone puts a default.nix into their source repos. and then I want to download some repo and build it using the default.nix in that repo
<catern> but if the resulting build is always re-evaluated even though it ultimately stems from a fixed-output derivation, then that's quite inefficient
<gchristensen> I don't know if that happens now
GRBurst has quit [(Ping timeout: 240 seconds)]
<gchristensen> but bugs get fixed as bugs annoy people
spinus has quit [(Read error: Connection reset by peer)]
arianvp2 has joined #nixos
spinus has joined #nixos
<catern> gchristensen: well, I mean, is it even *possible* to not re-evaluate it each time? like what if I loaded a nix file that was an attrset? and used some attributes? would the attrset be cached in the store somehow???
<gchristensen> I don't understand what we're talking about anymore
<catern> ¯\_(ツ)_/¯
<LnL> for each drv a corresponding file is created in the store, but you have to evaluate the expression in order to know if it already exists
arianvp2 has quit [(Read error: Connection reset by peer)]
arianvp2 has joined #nixos
<catern> which means i need to realize the fixed output derivation that the expression is coming from, just to evaluate the expression, even though it has to be the same
<catern> which, if the fixed output derivation is a source tarball, is going to be frustrating
<catern> one would rather just skip right to downloading a binary substitute
magnetophon has joined #nixos
GRBurst has joined #nixos
<LnL> nix optimise that case and only look at name/outputHash for fixed output drvs I don't know
mrcheeks has quit [(Quit: ERC (IRC client for Emacs 26.0.50))]
<LnL> nix might*
<chominist[m]> I can't for the life of me figure out how to set semicolon to apostrophe, colon to quotedbl on xkbOptions
<chominist[m]> I tried services.xserver.xkbOptions = "apostrophe:semicolon,caps:escape"
<chominist[m]> "caps:escape" works
magnetophon has quit [(Ping timeout: 276 seconds)]
<chominist[m]> And localectl does not work
<chominist[m]> "apostrophe:semicolon" does not work
<viric> Is there any command that will give me the "man configuration.nix" of the not-yet-build-nixos?
<viric> something like "nixos-rebuild --man" would be nice
<Thra11> I have a problem: nixos-rebuild complains of infinite recursion. configuration.nix and nixos-rebuild output with --show-trace here: https://gist.github.com/anonymous/0880c8f4823c8e70b7cf9df4bbe1b4a1 . Nix is version 1.10, nixpkgs is a local checkout of 17.03 channel. Running on an Odroid XU4 (armv7h).
sophiag has quit [(Ping timeout: 248 seconds)]
<viric> GC Warning: Failed to expand heap by 402653184 bytes
<viric> pf
Filystyn has quit [(Quit: Konversation terminated!)]
GRBurst has quit [(Ping timeout: 276 seconds)]
Coyin has joined #nixos
Coyin has quit [(Client Quit)]
<imalsogreg> gchristensen: After more testing last night, I found I had trouble connecting to nix binary caches. I added my username to `trusted-users` and the binary cache server to `trusted-binary-caches` in `/etc/nix/nix.conf`, but when I tried to enter a nix shell indirectly (via a script called try-reflex, if you're familiar with that project), download-from-binary-cache.pl gave me a warning that the nix cache wasn't trusted. I'm not sure if
<imalsogreg> this issue has more to do with OSX, nix-daemon or try-reflex.
<gchristensen> you shouldn't need to be a trusted user or trusted binary cache
xadi has joined #nixos
magnetophon has joined #nixos
<joepie91> gchristensen: made a thing: https://i.imgur.com/QrXz8o8.png
<imalsogreg> (the cache in question: https://nixcache.reflex-frp.org)
tosmi` has quit [(Remote host closed the connection)]
<gchristensen> what was the error you got?
<gchristensen> can you paste your exact long?
<LnL> joepie91: you can't user pkgs in nixpkgs.config
<LnL> since that's used to evaluate nixpkgs and set pkgs
<joepie91> but... I'm not?
jonte has quit [(Ping timeout: 240 seconds)]
<magnetophon> joepie91: nice! where do I find that tool?
<imalsogreg> gchristensen: Sure thing. It'll take me a bit. I reverted to single-user last night to sanity check the binary caches, so I'll re-revert to multi-user and reprocude the issue. shlevy informed me that maybe the problem was that I hadn't restarted nix-daemon after modifying trusted-binary-caches config
<joepie91> magnetophon: nowhere yet :P I just cobbled it together from the stacktrace parser I had laying around and some quickly written code
<joepie91> magnetophon: have to polish it a bit and publish it probably
<gchristensen> oh was shlevy around?
<imalsogreg> I have a shlevy bat phone :)
<gchristensen> oh nice :)
<viric> Pf
<joepie91> magnetophon: I hereby consider you subscribed to the "when is the error reporting tool published" notification list :p
<viric> again, nixos-rebuild --show-trace and I get 200 lines with no clue of what option made it fail
<LnL> joepie91: oh wrong person
<joepie91> oh
<joepie91> heh
<joepie91> I got really confused there for a bit :P
<LnL> Thra11: you can't use pkgs in nixpkgs.config
arianvp2 has quit [(Read error: Connection reset by peer)]
<joepie91> magnetophon: I've been slowly working on some out-of-band debugging tools for Nix
<viric> How to use plasma5.extraPackages in nixos? Nothing works for me there
arianvp2 has joined #nixos
<joepie91> magnetophon: another such tool is https://git.cryto.net/joepie91/find-missing-files, which takes in strace output, and analyzes the output for missing files and libraries and such, and prints a report of what was found and what was missing
<joepie91> magnetophon: useful for packaging tricky stuff that does a lot of runtime loading and where the deps aren't clear upfront
<joepie91> (also not formally published yet, but you can git clone and use it after an `npm install`)
<joepie91> (also desperately needs more polishing...)
<imalsogreg> gchristensen: Found the log: http://lpaste.net/4758308213273657344
<viric> Whatever I put to plasma5.extraPackages, I get "cannot find attribute "unwrapped""
<joepie91> (I have no idea what file the issue even occurs in)
<LnL> yeah that's because it's in a module defined with { imports = [ { ... } ]; } instead of with a path
<Gravious> gchristensen, you still looking for things to merge? :p
<joepie91> LnL: hmm. I thought that was supposed to work?
<pstn> The tt-rss service is hard dependent onto a locallly started db server at runtime via the systemd service file. I want to change that behaviour. Should I add another option like tt-rss.remoteDatabase or just check whether the DB Host is set?
<joepie91> LnL: passing a set into imports, that is
<joepie91> iirc I've done something with that before
<gchristensen> imalsogreg: doese https://nixcache.reflex-frp.org provide a signature?
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] peterhoeg pushed 1 new commit to master: https://git.io/vQinZ
<NixOS_GitHub> nixpkgs/master f2be840 Peter Hoeg: re2c: minor cleanup
NixOS_GitHub has left #nixos []
<viric> Anyone using plasma5?
<LnL> joepie91: yes it works, but then the errors aren't nice as you can see with the anon-1 paths
<pstn> I can't check whether the db host is localhost because you could have a DB in another container that you connect to via localhost.
<joepie91> LnL: so if I comment out that item, I get a different error
<magnetophon> joepie91: looks great already. Will you write the mailing list when you publish it?
<joepie91> oh hold on, sec
arianvp2 has quit [(Read error: Connection reset by peer)]
<Thra11> LnL: I see. Any idea what I should replace that line with? (I'm new to nixos. I didn't write this configuration.nix myself, I believe qknight did a couple of years ago...)
arianvp2 has joined #nixos
<joepie91> magnetophon: unlikely, I don't like mailing lists :P
<Gravious> https://github.com/NixOS/nixpkgs/pull/27243 is trivial and fixes a broken package :)
<Gravious> the CI screwed up for some reason and ran for 49 minutes, doubt it's anything to do with the change
<LnL> Thra11: platforms = (import <nixpkgs> {}).platforms.odroid_xu4; would work
<joepie91> magnetophon: naturally, anybody else is free to post it to the mailing list once released :P
<obadz> zimbatm: shall we schedule a meetup to squash some 17.09 issues for late august?
<gchristensen> oh my, 17.09 is coming up :o
<obadz> when does it go into beta? late aug/early sep no?
<Thra11> LnL: Ok, thanks. I'll try that.
<imalsogreg> gchristensen: I'm not sure how to know whether that server provides a signature.
<joepie91> LnL: any idea on my error? perhaps I'm overlooking something, but the code seems valid to me...
marsel has quit [(Ping timeout: 268 seconds)]
<magnetophon> joepie91: OK, I'll follow you on GH then.
<joepie91> magnetophon: my new projects don't appear on github anymore (and my old projects will be moved away from it over time), only on my own git server
<LnL> not sure, maybe it's in another file?
Filystyn has joined #nixos
Filystyn has quit [(Changing host)]
Filystyn has joined #nixos
<Gravious> gchristensen, apologies, you said this yesterday
<Gravious> 20:44 < gchristensen> if y'all have a list of PRs you think are ready to merge, send 'em my way :
<Gravious> so i thought it was ok to ask
<Olgierd> hi
<gchristensen> Gravious: sorry, I just missed your message. sure
<Gravious> gchristensen, ah, sorry :(
<gchristensen> sorry for what? :P
<gchristensen> imalsogreg: until someone gets back to me in their channel, add it to your trusted caches and restart the nix daemon, then try again
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] barrucadu opened pull request #27259: ezstream: init at 0.6.0 (master...master) https://git.io/vQinS
NixOS_GitHub has left #nixos []
<magnetophon> joepie91: OK, thanks for the heads up.
<joepie91> magnetophon: anyhow, I'll just make sure to ping you on IRC :)
<Olgierd> what's good for getting a deeper understanding of nix/nixos internals? if I would like to write a package or derivation, or implement a config option without copypasting
<gchristensen> imalsogreg: oh actually, remove signed-binary-caches = *
<Gravious> Olgierd, i found the nix-pills series good
<magnetophon> joepie91: thanks
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] grahamc pushed 2 new commits to master: https://git.io/vQin7
<NixOS_GitHub> nixpkgs/master 32dec43 Richard Ipsum: lrexlib: 2.7.2 (broken) -> 2.8.0...
<NixOS_GitHub> nixpkgs/master 26a69b1 Graham Christensen: Merge pull request #27243 from richardipsum/richardipsum/fix-lrexlib...
NixOS_GitHub has left #nixos []
<imalsogreg> gchristensen: Progress. I have a public key for that server (I'm not using it in the nix.conf yet). try-reflex now does try to use the nix cache there, giving this error: download-from-binary-cache.pl: NAR info file ‘https://nixcache.reflex-frp.org/yn7djr2fj4m067ivvsi27nrhdrpn9cgd.narinfo’ is signed by unknown key ‘ryantrinkle.com-1’; ignoring
<Gravious> gchristensen, wow nice thanks! :D
<gchristensen> ryantrinkle: can you share your public key?
<joepie91> LnL: I just have no idea how to figure it out :P anyway, I'm seriously sleep-deprived by now so I should probably just go to sleep, heh
<gchristensen> see also nixcache.reflex-frp
<joepie91> and look at it with fresh eyes tomorrow
<gchristensen> Gravious: are you 0xabab?
<Gravious> gchristensen, no
<imalsogreg> gchristensen: After adding ryantrinkle's public key to nix.conf binary-cache-public-keys and restarting nix-daemon, downloads are working :)
<gchristensen> oh ok
<gchristensen> nice!
<magnetophon> joepie91: had a look at you site: goed bezig ! :)
<magnetophon> s/you/your/
arianvp2 has quit [(Quit: arianvp2)]
arianvp2 has joined #nixos
Supersonic112 has quit [(Quit: Bye)]
CrazedProgrammer has quit [(Read error: Connection reset by peer)]
CrazedProgrammer has joined #nixos
<Filystyn> does tor work for you guys?
<Filystyn> (unstable)
<chominist[m]> How to restart x windowing system?
<gchristensen> chominist[m]: systemctl restart display-managere
<spinus> chominist[m]: you could try `systemctl restart display-manager`
[0x4A6F] has joined #nixos
<chominist[m]> Thanks
<chominist[m]> Also how do I change terminal bg and stuff
<spinus> I assume it depends on what X env do you use
<chominist[m]> (And also how do you deal with xkb, nothing besides caps:escape is working)
<chominist[m]> xmonad
<chominist[m]> So look there?
<bennofs> How do I find the latest nixpkgs-unstable channel without commit 8a62a9b064ed52da5ab2fd9520cdd218ce4445cb ?
<spinus> I don't know xmonad so can't help here
<gchristensen> bennofs: you can look through https://channels.nix.gsc.io/nixpkgs-unstable/history
ryantrinkle has quit [(Ping timeout: 248 seconds)]
<spinus> mhm, this is nice
<magnetophon> Filystyn: nope: "Gah. Your tab just crashed."
endformationage has joined #nixos
<spinus> vaibhavsagar: what I found is, the first call fails as I think this is that interpreter is showing on first invocation
<spinus> I added another test before
<Filystyn> spinus got the same
<Filystyn> thx
<spinus> and the same thing happens, but this time, the test which was failing is now working
<vaibhavsagar> wow, thanks for looking into it!
<vaibhavsagar> so what happens between the first and second invocation?
<spinus> I think interpreter is printing something on the screen when it starts
<spinus> no clue :)
<spinus> so I think, as easy and dirty, you could add dummy test as first which discards output
<vaibhavsagar> super weird that this isn't happening with nix-shell or stack
Supersonic112 has joined #nixos
<spinus> maybe one of the env variables is triggering the behaviour
Filystyn has quit [(Remote host closed the connection)]
<vaibhavsagar> it's worth a try
<spinus> I mean, if you have power to debug starting section of ihaskell, than yeah :)
<spinus> another thing could be to fix `become` to discard anything which is not correct output
<vaibhavsagar> that sounds like it would break the tests
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] Mic92 closed pull request #27229: minio service: add additional config options (master...minio-more-config) https://git.io/vQKH3
NixOS_GitHub has left #nixos []
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] calbrecht opened pull request #27261: nixos/auditd: init at 2.7.6 (master...upstream-auditd-11864) https://git.io/vQiWn
NixOS_GitHub has left #nixos []
seif[m] has joined #nixos
ng0 has joined #nixos
iyzsong has quit [(Quit: bye.)]
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] globin pushed 1 new commit to openssl-1.1: https://git.io/vQiW1
<NixOS_GitHub> nixpkgs/openssl-1.1 322a4e1 Robin Gloster: mariadb: lib -> connector-c
NixOS_GitHub has left #nixos []
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] Mic92 pushed 1 new commit to master: https://git.io/vQiWA
<NixOS_GitHub> nixpkgs/master 4aad6d8 Michael Walker: ezstream: init at 0.6.0 (#27259)...
NixOS_GitHub has left #nixos []
johnw_ has joined #nixos
Infinisil has joined #nixos
johnw has quit [(Ping timeout: 276 seconds)]
<vaibhavsagar> spinus: it worked!
bennofs has quit [(Ping timeout: 260 seconds)]
radvendii has joined #nixos
<spinus> :)
<spinus> nice
<radvendii> So I'm writing a nix package, and I want to set the CFLAGS variable before the `make` command is run. I've gathered that the right way to do this is by setting `makeFlagsArray` in `preBuild`. But when I do that, I run into errors because it seems to override some flags that are anyways being passed into gcc. What am I doing wrong?
<radvendii> like, some flags that are being passed into gcc are no longer being passed in when I set makeFlagsArray
<spinus> radvendii: you could do something like "export CFLAGS="$CFLAGS --more --flags""
<radvendii> spinus: also in preBuild?
<spinus> sure
<radvendii> okay thanks, I'll try that. There isn't a cleaner nix-y way of doing all of this?
<spinus> radvendii: you could do CFLAGS=["--flag" "--flag2"]; in derivation
<spinus> but I don't know if that overrides something or not
* Gravious wonders if you can pass a glob to wrapProgram
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] zimbatm pushed 1 new commit to master: https://git.io/vQi8J
<NixOS_GitHub> nixpkgs/master 466e7e2 zimbatm: caddy: 0.10.3 -> 0.10.4...
NixOS_GitHub has left #nixos []
<Gravious> it would seem not :<
<spinus> Gravious: what you are trying to do? wrap multiple programs?
<Gravious> i'd like to be able to say "wrap all the binaries" and there are about 6
<spinus> ah
<Gravious> spinus, yeah
<radvendii> Gravious: can you use xargs?
bennofs has joined #nixos
<radvendii> Gravious: ls | xargs wrapProgram
aristid is now known as aristid2
<Gravious> radvendii, nice idea :)
<radvendii> Gravious: yeah, that's one of my favorite utilities
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] vcunat pushed 1 new commit to staging: https://git.io/vQi8s
<NixOS_GitHub> nixpkgs/staging bfb7ef8 Vladimír Čunát: Merge branch 'master' into staging...
NixOS_GitHub has left #nixos []
aristid has joined #nixos
<Gravious> oh wait is wrapProgram a nix function rather than an actual command?
<radvendii> Gravious: It seems to be a command
<Gravious> xargs: wrapProgram: No such file or directory
<Gravious> i see people doing,
<Gravious> 53 for f in $out/bin/*; do 54 wrapProgram $f \ 55 --prefix PYTHONPATH : $PYTHONPATH
<Gravious> erm, sorry..
<gchristensen> Gravious: it is a shell function, which means it can't be used in xargs
<radvendii> gchristensen: O.o I didn't realize there were limitations like that on xargs
<radvendii> :(
<Gravious> gchristensen, ah, probably will just use the loop then
<gchristensen> well xargs calls things in $PATH via exec
<aristid> so how _do_ you invoke a shell function with xargs-alike? :D
<radvendii> why doesn't it just unwrap into a loop?
<gchristensen> because xargs is an executable
<gchristensen> not a shell builtin
<aristid> radvendii: because that's something completely different
<radvendii> ah
<aristid> radvendii: xargs rm will pass all the arguments to _one_ call of rm
<aristid> which is faster too, so that's good
<radvendii> interesting
<radvendii> spinus: when I try setting CFLAGS in preBuild, it doesn't seem to have *any* effect
<radvendii> oh whoops
<radvendii> i forgot to export
ryantrinkle has joined #nixos
jluttine has quit [(Quit: WeeChat 1.7.1)]
phinxy has joined #nixos
jluttine has joined #nixos
arianvp2 has quit [(Quit: arianvp2)]
arianvp2 has joined #nixos
vaibhavsagar has quit [(Ping timeout: 260 seconds)]
MercurialAlchemi has quit [(Ping timeout: 240 seconds)]
bennofs has quit [(Ping timeout: 240 seconds)]
alx741_ has joined #nixos
alx741 has quit [(Ping timeout: 240 seconds)]
Sonarpulse has joined #nixos
magnetophon has quit [(Ping timeout: 255 seconds)]
bennofs has joined #nixos
ng0 has quit [(Quit: Alexa, when is the end of world?)]
<wmertens[m]> grahamc: ping
<gchristensen> wmertens[m]! hi!
<spinus> radvendii: hmm, I'm not sure if exporting would help. Basically when you put something like that in mkDerivation, variable is set in the build environment.
<wmertens[m]> So, what can I break^H^H^H^H^Htest?
<gchristensen> :D
arianvp2 has quit [(Quit: arianvp2)]
<radvendii> spinus: it's working now. i was running CFLAGS="blah" in prebuild instead of export CFLAGS="blah"
<spinus> cool
<radvendii> in fact, everything is working. thank you so much
<gchristensen> wmertens[m]: I have two options for you to try, the first is where you download the nix.tar.bz2 and inside you run ./install (I'd rather you test this one, as it is closer to the final result of this project), or the second where you run my script, and it downloads the nix.tar.bz2 from nixos.org
<radvendii> :)
<radvendii> there should be a nix package for ring daemon very soon
<gchristensen> wmertens[m]: which would you prefer?
<spinus> radvendii: what's ring daemon?
<spinus> IM?
<radvendii> spinus: it's a chat client
<radvendii> spinus: but it does calling
<radvendii> and it's a daemon
<radvendii> so you can call people with scripts
<radvendii> :D
<wmertens[m]> gchristensen: I'll take the option you prefer
<radvendii> or keyboard sohrtcuts!
<gchristensen> wmertens[m]: okay, give me ~5 minutes :D
<radvendii> (i'm so excited)
<spinus> :)
<spinus> I'm happy to test
ixxie has quit [(Quit: Lost terminal)]
<radvendii> Sweet. I'll let you know when it's up.
<gchristensen> wmertens[m]: actually I'd like to run one more test before you do. how long do you have?
<wmertens[m]> plenty, I'm on vacation, but I may need to do social stuff in between :)
<gchristensen> ah okay
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] Mic92 closed pull request #27261: nixos/auditd: init at 2.7.6 (master...upstream-auditd-11864) https://git.io/vQiWn
NixOS_GitHub has left #nixos []
<gchristensen> wmertens[m]: ok here we go: curl -o install.tar.bz2 http://gsc.io/nix.tar.bz2 && echo "1587e440bac9974733ec7182b22aeb7f883aa262db02486d05949979a8dec177 install.tar.bz2" | shasum -c && tar -xf ./install.tar.bz2 && cd nix-* && ./install
lambdael has joined #nixos
magnetophon has joined #nixos
bennofs has quit [(Ping timeout: 276 seconds)]
peacememories has joined #nixos
peacememories has quit [(Client Quit)]
peacememories has joined #nixos
endformationage has quit [(Read error: Connection reset by peer)]
endformationage has joined #nixos
nckx has quit [(Quit: restarting my GuixSD server)]
nckx has joined #nixos
<wmertens[m]> gchristensen: ok I'll report back later, family time :)
<gchristensen> cool :) thank you! ping me before you do
<radvendii> What happens if you bump the version of something in nixpkgs, but that breaks some other package?
<LnL> nox-review is supposed to catch that in most cases
<radvendii> oh okay, so I don't have to worry about it myself *whew*
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nix] grahamc opened pull request #1453: Multi user darwin (1.11-maintenance...multi-user-darwin) https://git.io/vQiRS
NixOS_GitHub has left #nixos []
<gchristensen> wmertens[m]: I also documented it here: https://github.com/NixOS/nix/pull/1453#issue-241539316
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] Radvendii opened pull request #27262: pjsip: 2.5.5 -> 2.6 (master...master) https://git.io/vQi09
NixOS_GitHub has left #nixos []
<radvendii> I think I did it right? (I'm always terrified of forgetting something when I make a pull request)
<radvendii> If I want to update multiple packages in nixpkgs at the same time, and submit pull requests, what's the best way to do that? If I change them on my forked copy, the changes will accumulate which is not what I want since I want them to be in separate pull requests, no?
<radvendii> by at the same time I mean in parallel, not in the same pull request
<manveru> branches
<simpson> radvendii: You can always have multiple branches on your fork.
<radvendii> simpson: ohhh good call
bennofs has joined #nixos
ison111 has quit [(Ping timeout: 240 seconds)]
niddhoggur has joined #nixos
ison111_ has joined #nixos
Ivanych has joined #nixos
endformationage has quit [(Quit: WeeChat 1.7)]
<catern> does copumpkin use irc?
<LnL> yes
<radvendii> he's on this channel currently
<catern> oh, under the same name too :)
<niddhoggur> hey all :)
niddhoggur is now known as NickSeagull
<NickSeagull> woops wrong nickname :P
<catern> copumpkin: did you ever find out a good solution for https://www.mail-archive.com/nix-dev@lists.science.uu.nl/msg34912.html ? none of the responses seemed very useful...
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] kirelagin closed pull request #10539: fetchzip: respect the `name` argument (master...fetchzip-basename) https://git.io/vWckj
NixOS_GitHub has left #nixos []
<NickSeagull> Is a copumpkin a watermelon?
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] Mic92 pushed 2 new commits to master: https://git.io/vQize
<NixOS_GitHub> nixpkgs/master 6152f4a Maximilian Bosch: Jenkinsfile-vim-syntax: init
<NixOS_GitHub> nixpkgs/master f939630 Jörg Thalheim: Merge pull request #27232 from Ma27/init/jenkinsfile-vim-support...
NixOS_GitHub has left #nixos []
<LnL> catern: I think overlays are a nice solution for that
<catern> LnL: I haven't used overlays, so maybe I'm missing it, but how exactly?
peacememories has quit [(Quit: My MacBook has gone to sleep. ZZZzzz…)]
<catern> yes, I just read that
<catern> I don't see how that solves the problem
<LnL> ofcorse that's more like the second approach
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] Mic92 pushed 2 new commits to master: https://git.io/vQizs
<NixOS_GitHub> nixpkgs/master 56923a4 Taeer Bar-Yam: pjsip: 2.5.5 -> 2.6
<NixOS_GitHub> nixpkgs/master 37d9a9b Jörg Thalheim: Merge pull request #27262 from Radvendii/master...
NixOS_GitHub has left #nixos []
thaega has quit [(Ping timeout: 255 seconds)]
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] Radvendii opened pull request #27263: asio: 1.10.6 -> 1.10.8 (master...asio) https://git.io/vQizy
NixOS_GitHub has left #nixos []
<radvendii> is there a way to mark things as having two possible licenses? I'm making a package which is licensed under "either CC 1.0 or Apache 2.0"
<radvendii> should I just pick one?
tmaekawa has joined #nixos
tmaekawa has quit [(Client Quit)]
<Sonarpulse> can one nix-shell without source stdenv.setup?
<radvendii> gchristensen: sweet thanks. didn't realize that could be a list
<gchristensen> :)
<radvendii> that's some pretty weak typing though :,(
<simpson> radvendii: What would you prefer instead?
<radvendii> It could always be a list
<simpson> Oh, you don't like how it could be a list or a single object?
sigmundv_ has joined #nixos
<radvendii> yeah
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nix] grahamc opened pull request #1454: Add nix profile suitable for using the daemon (master...add-nix-profile-daemon) https://git.io/vQi2q
NixOS_GitHub has left #nixos []
Ivanych has quit [(Ping timeout: 240 seconds)]
taeradan has quit [(Remote host closed the connection)]
<gchristensen> it makes me chuckle a little at how many attempts of multi-user nix installers are out there, some for darwin, some for other linuxes, some that upgrade from single-user to multi-user, and it makes me a little sad that none of them are official >.>
<gchristensen> I don't care if mine is merged or not, I care that there is an official way to do it
thaega has joined #nixos
<Infinisil> gchristensen: I agree, what is yours?
Supersonic112 has quit [(Quit: Bye)]
sigmundv_ has quit [(Remote host closed the connection)]
Supersonic112 has joined #nixos
sigmundv_ has joined #nixos
<Infinisil> gchristensen: Looks nice! I'll have to reboot into osx now, may be back on the other side though
<gchristensen> woohoo!
magnetophon has quit [(Ping timeout: 240 seconds)]
<gchristensen> Infinisil: this is how I'd recommend testing it: curl -o install.tar.bz2 http://gsc.io/nix.tar.bz2 && echo "1587e440bac9974733ec7182b22aeb7f883aa262db02486d05949979a8dec177 install.tar.bz2" | shasum -c && tar -xf ./install.tar.bz2 && cd nix-* && ./install
<NickSeagull> What is the correct way to install npm packages that result in a binary with NixOS?
<Infinisil> gchristensen: Ohh, I'm so sorry! I actually need to reboot into mac for another reason, but I could test that too if you'd like! I'm not using osx anymore really, only rebooting into it to transfer some stuff
<NickSeagull> I mean, is there a helper like buildPythonApplication but for npm modules?
<clever> gchristensen: did you figure out how sandboxing works on OSX?
<gchristensen> Infinisil: oh, well, if you would:P but you don't need to ;)
<gchristensen> clever: hmm what do you mean?
<Infinisil> I'll bbl
<clever> gchristensen: the nix sandboxing
Infinisil has quit [(Quit: leaving)]
<NickSeagull> Great thanks Infinisil
<catern> gchristensen: fyi I believe there is autodetection for NIX_REMOTE in recent master
<catern> gchristensen: if the store isn't writable by the current user Nix will default to trying to use the daemon
dejanr has joined #nixos
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] joachifm pushed 2 new commits to master: https://git.io/vQiVe
<NixOS_GitHub> nixpkgs/master b5c9438 gnidorah: gtk-engine-bluecurve: init at 1.0
<NixOS_GitHub> nixpkgs/master 8c1c2a5 Joachim F: Merge pull request #27136 from gnidorah/gtk-bluecurve...
NixOS_GitHub has left #nixos []
<gchristensen> catern: neat
<gchristensen> catern: I'm targeting 1.11 though
<catern> ah indeed
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] edolstra pushed 1 new commit to master: https://git.io/vQiVZ
<NixOS_GitHub> nixpkgs/master 7b52b0c Eelco Dolstra: youtube-dl: 2017.06.18 -> 2017.07.09
NixOS_GitHub has left #nixos []
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] Radvendii opened pull request #27264: libargon2: init at 20161029 (master...argon) https://git.io/vQiVc
NixOS_GitHub has left #nixos []
<clever> gchristensen: after reading the source i linked above, i think its enough to setup build-sandbox-paths just like on nixos
<clever> /bin/sh=<pathtosh> <pathtoglibc> ... <the full closure of sh>
<gchristensen> I think there are more impurities than that, like various Kits... copumpkin would know for sure
oida has joined #nixos
<radvendii> What's the right way to package a github project that has been updated recently but hasn't had any releases recently?
<radvendii> do you just package the latest commit? and if so, what do you use as the version number?
<gchristensen> the date, see the contributing docs when you open a PR
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] 239 opened pull request #27265: opera: 45.0.2552.898 -> 46.0.2597.39 (master...patch-1) https://git.io/vQiVA
NixOS_GitHub has left #nixos []
magnetophon has joined #nixos
oida_ has quit [(Ping timeout: 240 seconds)]
yegortimoshenko has joined #nixos
<yegortimoshenko> is it possible to use options w/ nix (not NixOS)?
sigmundv_ has quit [(Ping timeout: 240 seconds)]
<gchristensen> no
civodul has joined #nixos
<radvendii> gchristensen: I don't see anything in the contributing docs about what to do if there's no version number, but thank you. I will use the date. Is there a specific way that should be formatted? people seem to be using YYYY-MM-DD. Is that right?
<grahamc> Yes
Infinisil has joined #nixos
<LnL> it should be in there somewhere
<Infinisil> Reporting from macOS, can not recommend, very laggy for some reason
<yegortimoshenko> LnL: why nix-darwin uses darwin-rebuild instead of nixos-rebuild?
<yegortimoshenko> what i mean is, for consistency's sake it could have been named nixos-rebuild...
<Infinisil> gchristensen: You may know this, I want to install osx on an external disk, do I need to keep anything on the main disk for it to boot? I'd like to not have a single bit of osx left on it
<Infinisil> I feel like it should work, but I'm not sure
<gchristensen> I have no idea :/
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] Mic92 closed pull request #27263: asio: 1.10.6 -> 1.10.8 (master...asio) https://git.io/vQizy
NixOS_GitHub has left #nixos []
mounty has quit [(Quit: Konversation terminated!)]
<Infinisil> gchristensen: I'll just try it then :p
<Infinisil> Coming back to osx I have no idea how I ever managed to be productive without a tiling window manager
<gchristensen> yegortimoshenko: well, simply, because darwin-rebuild isn't nixos
yegortimoshenko has quit [(Remote host closed the connection)]
yegortimoshenko has joined #nixos
ryanartecona has joined #nixos
<copumpkin> Impurities are /bin/sh, env, dyld, and libSystem and sub-libs
<dejanr> Infinisil i keep switching back and forth, once i have proper setup of virtualization with gpu passtrought i might not need to do it :D
<copumpkin> Then more as you move into framework land but lots of stuff avoids that
<LnL> yegortimoshenko: dunno, it's not nixos :)
<LnL> but I often type nixos on darwin and the other way around :p
yegortimoshenko has quit [(Ping timeout: 248 seconds)]
<radvendii> what am I supposed to do if I need an old version of a package for something?
<radvendii> like as a dependency
<simpson> radvendii: They're often still in-tree under versioned names, like pythonPackages.django_1_6.
yegortimoshenko has joined #nixos
<radvendii> wait, so is it bad form to bump a version number up? Should I really be making a different package for the newer version?
arianvp2 has joined #nixos
<simpson> It really depends on how radically different the newer version is. What package is this?
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] bugworm opened pull request #27266: antigen: init at 2.2.1 (master...antigen) https://git.io/vQirV
NixOS_GitHub has left #nixos []
<copumpkin> No, in general we try to avoid keeping multiple versions of stuff around unless other packages need older versions
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nix] shlevy closed pull request #1454: Add nix profile suitable for using the daemon (1.11-maintenance...add-nix-profile-daemon) https://git.io/vQi2q
NixOS_GitHub has left #nixos []
yegortimoshenko has quit [(Ping timeout: 248 seconds)]
<gchristensen> shlevy: !!! :o
<radvendii> sorry, the actual situation is I'm trying to package dring, which expects restbed version 4.0. Restbed is not packaged at all by nixpkgs, so I'm writing a package for it. The most recent version of restbed is 4.5
<simpson> Ah, then yeah, just package 4.0.
<simpson> Maybe leave a comment saying that upgrading it breaks dring.
<radvendii> where should I leave that comment? In the default.nix file?
<simpson> Yeah.
marsel has joined #nixos
yegortimoshenko has joined #nixos
arianvp2 has quit [(Read error: Connection reset by peer)]
arianvp2 has joined #nixos
obadz has quit [(Ping timeout: 240 seconds)]
erictapen has joined #nixos
obadz has joined #nixos
yegortimoshenko has quit [(Ping timeout: 240 seconds)]
<Gravious> oh weird, you can't swap -iA for -eA and expect that to work the same?
<Gravious> oh well you can just give it the path, fine
<radvendii> simpson: what if what I really need is some random commit (not an actual version number)? Is it reasonable to package that? I feel like it's really messy, but I don't know what a better way to do it would be
erlandsona has joined #nixos
<simpson> radvendii: You could always shame upstream until they fix their shit, but historically that hasn't worked for most upstreams.
<erlandsona> So I have a list of packages installed at the system-level, and I know I'm supposed to move them to the user-level eg: currently installed with `nixos-rebuild switch` vs `nix-env -irA users-env`... how do you know which packages need to be at the system level verses the user-level?
<radvendii> yeah, dring is a super janky project. They expect everything to be vendored by them basically
<erlandsona> Also if I install like mysql for example as a user-level package, will it still work like normal?
orivej has joined #nixos
<radvendii> simpson: Oh. If what I need is a super specific version (with certian patches applied no less), does it make sense to just include that in the dring package, and not even bind it at the top level?
<radvendii> since nobody else is likely to need it
<simpson> radvendii: Probably.
<simpson> erlandsona: Do you want a package available for everybody, or just for one user?
<erlandsona> I'm the only user so it doesn't really matter I just know some things usually need to be installed as root or sudo right? Like Databases or services right?
<simpson> Most services don't require anything in environment.systemPackages, or they'll put it there themselves.
<erlandsona> Oh well sweet then! In which case I should just be able to move all my specified environment.systemPackages into my local users .config/nixpkgs/config.nix file eh?
<himmAllRight> How can I get virtmanager and vagrant to detect my default hypervisor?
zraexy has quit [(Ping timeout: 255 seconds)]
<radvendii> erlandsona: why would you want to?
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] bjornfor pushed 2 new commits to master: https://git.io/vQiKr
<NixOS_GitHub> nixpkgs/master 35e9127 Niklas Hambüchen: maintainers: Add nh2
<NixOS_GitHub> nixpkgs/master a37f3a3 Niklas Hambüchen: hotspot: init at 1.0.0
NixOS_GitHub has left #nixos []
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] bjornfor closed pull request #27181: hotspot: init at 1.0.0 (master...add-hotspot) https://git.io/vQw96
NixOS_GitHub has left #nixos []
endformationage has joined #nixos
arianvp2 has quit [(Quit: arianvp2)]
<erlandsona> radvendii: Things aren't supposed to be installed by root right? systemPackages makes me feel like I'll get burned by something trying to rm -rf a file with whitespace or something?
<radvendii> Oh interesting. I don't think there are any problems with that, I think systemPackages is just for things that you want to be installed for all users
<radvendii> I don't think they run with special permissions or anything
guillaum1 has quit [(Quit: Lost terminal)]
<erlandsona> So maybe I'
<radvendii> But I'm not that knowledgeable
<radvendii> I might be wrong
<radvendii> can someone confirm/deny?
<LnL> do we have a general policy on logfiles vs journald for services?
<erlandsona> I'm just confused then... if they don't have special permissions then why does there seem to be so much interest in user-level configuration?
krav1 has joined #nixos
magnetophon has quit [(Ping timeout: 255 seconds)]
<erlandsona> LnL: Do you have any thoughts on motivations / differences between using environment.systemPackages with `nixos-rebuild switch` vs user-level configuration with `nix-env -irA user-env`?
<radvendii> erlandsona: I think the interest is that then we can have different configuration for different users
<LnL> not really, depends on what you want
<simpson> erlandsona: Because the typical NixOS user of the future, for whom we want to make things easy, does *not* have superuser powers.
<radvendii> simpson: oh that too
sibi has quit [(Quit: Connection closed for inactivity)]
<erlandsona> simpson: Why do you say that? You mean like on servers? Or on personal machines? The only way I could imagine someone who doesn't have sudo permissions is some one who's not a dev, but if they aren't a dev they wouldn't know about nix anyways right?
<erlandsona> Granted I don't work for any monolith corp's either so maybe it's different?
<LnL> erlandsona: I personally put very few things in my system/user profiles and mainly use nix-shell for all my project/language specific things
<erlandsona> Okay I get that people do it that way, I'm primarily asking why? the nix-shell for project stuff makes sense because it keeps you from litering your local system with dependencies for a particular project.
<erlandsona> I guess does the same logic apply for root? Rather, how do you know what needs to be installed at the system level vs the user level?
<erlandsona> Or is it more obvious and you'll just know when it happens?
<simpson> erlandsona: I know it sounds bizarre, but there do exist Linux systems with many unprivileged users.
<simpson> There also exist systems that do almost all of their work as a single unprivileged user without wheel.
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] domenkozar force-pushed osx_private_sdk/cleanup from d660ce2 to e211504: https://git.io/vQ6y5
<NixOS_GitHub> nixpkgs/osx_private_sdk/cleanup e211504 Domen Kožar: Get rid of some usages of osx_private_sdk...
NixOS_GitHub has left #nixos []
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] domenkozar closed pull request #27245: Get rid of some usages of osx_private_sdk (master...osx_private_sdk/cleanup) https://git.io/vQ6yj
NixOS_GitHub has left #nixos []
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nix] grahamc opened pull request #1455: Create a profile suitable for multi-user installs (master...add-nix-profile-daemon-master) https://git.io/vQi6H
NixOS_GitHub has left #nixos []
<erlandsona> simpson: I mean when I first learned about it, it was just some arbitrary thing that programs told me to run... or something I needed to work with directories outside of /home/user. I mean I guess having priviledges makes some sense. I deal with authentication / authorization in web apps all the time, but for a PC the concept is a bit more abstract.
tmaekawa has joined #nixos
<erlandsona> simpson: Could you explain some use cases?
<erlandsona> Or post an article I could read more about it?
krav_ has joined #nixos
<simpson> erlandsona: Remember the golden rule of security: "'Do what thou wilt' shall be the whole of the law!"
<simpson> erlandsona: Use case: I have a big machine. I give people accounts on the machine. I do not give people wheel.
mudri has joined #nixos
<simpson> This is basically how UNIX was deployed, once upon a time. The last time I logged into a Linux machine with this configuration was this morning, and I've been logged in ever since.
magnetophon has joined #nixos
<erlandsona> Yup, uhhh... over my head. Like anyone can do anything so be prepared? Oh maybe like super computers?
<simpson> "If you can do it, you may do it."
<simpson> It's very simple, I promise.
<simpson> Or another phrasing: "The ability to do it is the permission to do it."
LinArcX has joined #nixos
<simpson> erlandsona: Starting to make sense yet?
krav1 has quit [(Quit: WeeChat 1.7.1)]
<erlandsona> simpson: Oh the second phrasing makes a lot more sense. Sorry got lost in an article on satanism after looking up the first phrase. Mildly interesting read. kek
<erlandsona> More like, if you can do it no one is stopping you.
<erlandsona> Not more like, I'm rephrasing myself for understanding sorry.
<simpson> erlandsona: Indeed, Satanists and Discordians might recognize the first version from their holy texts.
<krav_> fnord
magnetophon has quit [(Ping timeout: 268 seconds)]
<simpson> You might recognize a security principle derived from this rule, called "need-to-know" informally, or formally the *Principle of Least Authority*.
<simpson> Nix is an example of a system which follows POLA.
<erlandsona> That makes sense now! So the next question though that radventii: brought up is, when you install a package with environment.systemPackages does that give that package root privaledges? Or does it just make that package available to all users with normal user privaledges?
<erlandsona> Becuase if it's the latter than for a single user system it makes no difference whether a package is installed via systemPackages or with a user-level ~/.config/nixpkgs/config.nix & nix-env right?
bennofs has quit [(Ping timeout: 246 seconds)]
<simpson> From a user's perspective, the store is enumerable, so packages don't have privileges based on which environments they're in.
<simpson> And stuff like setuid wrappers are configured at the NixOS level, but not as part of environment.systemPackages.
<simpson> So it shouldn't matter.
LinArcX has quit [(Remote host closed the connection)]
justbeingglad has joined #nixos
justbeingglad has left #nixos []
LinArcX has joined #nixos
NickSeagull has quit [(Quit: leaving)]
<erlandsona> Interesting, thanks for the help simpson.
erlandsona has left #nixos []
<LinArcX> Why don't put all of these golden tips on wiki pages, for ever?
<radvendii> LinArcX: they killed the wiki, didn't they?
<LinArcX> As arch users do that. And they have biggest wiki among other distributions
<Infinisil> LinArcX: I wouldn't really know where to put this info
<Infinisil> We really need some wiki-similar thing, but better than the old one
<LinArcX> I suggest to create a wiki-base system to save theses greater effort
<simpson> LinArcX, Infinisil: https://github.com/nixos-users/wiki/wiki
<simpson> LinArcX: If you want to write something down forever in the official docs, then send in a PR. It's not hard.
<Infinisil> simpson: Yeah, I find this a bit meh though
simukis has quit [(Ping timeout: 248 seconds)]
* Infinisil feels the urge to start the xml docs discussion again
Mateon1 has quit [(Ping timeout: 240 seconds)]
Mateon3 has joined #nixos
<simpson> Infinisil: So what? It's not supposed to be exciting. It's documentation.
<Infinisil> simpson: Well I think most of this stuff here should go into a QnA section, it's not really documentation
<LinArcX> @simpson, thanks. But many people have a lot of questions about nixos. that don't exists in the wiki. They are in irc. I suggest all people create wiki page instead response people in this place.
<simpson> Infinisil: FAQs are documentation.
LinArcX has quit [(Remote host closed the connection)]
Mateon3 is now known as Mateon1
<simpson> Ah, they don't like IRC.
tmaekawa has quit [(Quit: tmaekawa)]
<Infinisil> simpson: Where is the FAQ?
LinArcX has joined #nixos
<Infinisil> In the current docs?
<simpson> Infinisil: I mean to say that "it's not really documentation" is silly. Words describing stuff count as documentation. What features of a "wiki-similar thing" are missing from the new user wiki?
bennofs has joined #nixos
<Infinisil> simpson: Yeah
<Infinisil> simpson: I'll have a closer look
magnetophon1 has joined #nixos
<Infinisil> simpson: Well the github wiki is as the name says really just a wiki so I guess none
<Infinisil> I also just noticed that it's really easy to add a page
yegortimoshenko has joined #nixos
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] Mic92 closed pull request #27265: opera: 45.0.2552.898 -> 46.0.2597.39 (master...patch-1) https://git.io/vQiVA
NixOS_GitHub has left #nixos []
<Infinisil> The unofficial wiki is a bit on the low end on content, so it's nice to be able to add pages easily
<Taneb> shlevy, do you happen to be about? I'm trying to understand hail's documentation but I think I'm too much of a nix newb
<Infinisil> I guess I'll keep this in mind, for when I feel like something needs to be aded
Supersonic112 has quit [(Ping timeout: 240 seconds)]
Sonarpulse has quit [(Ping timeout: 258 seconds)]
Supersonic112 has joined #nixos
yegortimoshenko has quit [(Remote host closed the connection)]
LinArcX has quit [(Remote host closed the connection)]
ng0 has joined #nixos
magnetophon1 has quit [(Ping timeout: 248 seconds)]
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] globin pushed 1 new commit to openssl-1.1: https://git.io/vQiXr
<NixOS_GitHub> nixpkgs/openssl-1.1 d880360 Robin Gloster: mariadb: more lib -> connector-c
NixOS_GitHub has left #nixos []
hiratara_ has joined #nixos
thblt has joined #nixos
hiratara has quit [(Ping timeout: 255 seconds)]
hiratara_ is now known as hiratara
<catern> hey #nixos
<catern> are there any examples of packages where the sources are themselves an input?
<catern> hmm, yes
<catern> release.nix
<spinus> catern: :)
<catern> hmm
<catern> what is the fundamental difference between, say, the nixpkgs default.nix for patchelf, and the release.nix found in the patchelf repo?
tokudan has quit [(Quit: Leaving)]
<catern> it seems like they are two different specifications of the same thing: how to build the package
ilyaigpetrov has quit [(Quit: Connection closed for inactivity)]
proteusguy has quit [(Ping timeout: 248 seconds)]
<catern> release.nix builds a bunch of additional things as well of course
magnetophon has joined #nixos
<catern> more importantly I suppose:
<catern> is it possible, even theoretically, for fetchgit to ever not require a sha256 hash to be supplied, while still being a fixed-output derivation?
<catern> if you computed the store path from the git commit hash instead...
<clever> catern: the problem is that git is a bit of a blockchain design, and to confirm the files you cloned are from that commit, you have to clone the entire history
<catern> clever: ah right
<catern> but what if you trusted the repo you were cloning from?
<clever> catern: the commit is a hash over (commitmsg, parent commit, treeid)
<clever> and the treeid is a hash over a list of (filename, objectid) (the contents of the root folder)
<clever> one of the key things with fixed-output derivations, is that you dont have to trust the repo or the network
<catern> hmm... could you supply the treeid instead, then?
<clever> nix will enforce it always being the exact same source
<clever> you could, but git doesnt allow you to clone a random tree of objects
<catern> heh
<clever> you must start a clone from a commit that is at the tip of a branch
<catern> okay, i see, fair enough
<clever> and then you have things like fetchFromGithub, that entirely ignore the git protocol
<clever> it just builds a url to the .tar.gz archive
<clever> so the treeid doesnt even exist
<catern> hmmmmmmmm
<catern> well, if you have a conventional git repo
<catern> it should be fine, right?
<catern> to use (commit hash,tree id) instead of (commit hash, sha256)
<clever> and everything in git is sha1, which is close to becoming useless
<clever> every object in git is stored as sha1(value)=value, all objects have a type, and sometimes a size prepended to the value
<clever> and a directory is just a list of names and objectid's
civodul has quit [(Quit: ERC (IRC client for Emacs 25.2.1))]
proteusguy has joined #nixos
magnetophon has quit [(Ping timeout: 240 seconds)]
[0x4A6F] has quit [(Ping timeout: 248 seconds)]
erlandsona has joined #nixos
proteusguy has quit [(Client Quit)]
<erlandsona> Random other question... I'm trying to get packageOverrides working in my .config/nixpkgs/config.nix so I don't have to specify the path to the file in `nix-env -irA erlandsona` but I keep getting error: attribute ‘erlandsona’ in selection path ‘erlandsona’ not found
<erlandsona> Thoughts?
<clever> erlandsona: you need nixos.erlandsona
<swflint> Has anyone else had problems with getting documents that use algorithm2e to compile?
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] NeQuissimus pushed 2 new commits to master: https://git.io/vQiML
<NixOS_GitHub> nixpkgs/master fca0b36 Tim Steinbach: linux-copperhead: 4.12.b -> 4.12.c
<NixOS_GitHub> nixpkgs/master d38656b Tim Steinbach: linux-copperhead: 4.12.c -> 4.12.d
NixOS_GitHub has left #nixos []
<erlandsona> oh. thanks clever: !!!
<clever> the nixos. tells it which channel to load
<clever> that channel then loads config.nix, which mutates the package set
hamishmack has quit [(Quit: hamishmack)]
hiratara has quit [(Quit: ZNC - http://znc.in)]
hiratara has joined #nixos
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] fpletz pushed 4 new commits to master: https://git.io/vQiMr
<NixOS_GitHub> nixpkgs/master 1f926aa Franz Pletz: irssi: 1.0.3 -> 1.0.4 for multiple CVEs...
<NixOS_GitHub> nixpkgs/master 586f780 Franz Pletz: php56: 5.6.30 -> 5.6.31 for multiple CVEs...
<NixOS_GitHub> nixpkgs/master f15f580 Franz Pletz: php70: 7.0.19 -> 7.0.21 for multiple CVEs...
NixOS_GitHub has left #nixos []
ryantrinkle has quit [(Ping timeout: 268 seconds)]
bugworm has joined #nixos
<swflint> So, I have this error only on my Nix install: if I \usepackage{algorithm2e} and compile with lualatex, I get a unicode error.
bugworm has quit [(Client Quit)]
bugworm has joined #nixos
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] fpletz pushed 4 new commits to release-17.03: https://git.io/vQiM1
<NixOS_GitHub> nixpkgs/release-17.03 1c6b2e9 Franz Pletz: irssi: 1.0.3 -> 1.0.4 for multiple CVEs...
<NixOS_GitHub> nixpkgs/release-17.03 3d539ab Franz Pletz: php56: 5.6.30 -> 5.6.31 for multiple CVEs...
<NixOS_GitHub> nixpkgs/release-17.03 3ac3db9 Franz Pletz: php70: 7.0.19 -> 7.0.21 for multiple CVEs...
NixOS_GitHub has left #nixos []
thblt has quit [(Remote host closed the connection)]
tg has quit [(Ping timeout: 255 seconds)]
tg has joined #nixos
erlandsona has quit [(Quit: Leaving.)]
thblt has joined #nixos
<swflint> The specific error is String contains an invalid utf-8 sequence
magnetophon1 has joined #nixos
bugworm has quit [(Remote host closed the connection)]
erictapen has quit [(Ping timeout: 240 seconds)]
roconnor has joined #nixos
roconnor has quit [(Client Quit)]
roconnor has joined #nixos
roconnor has quit [(Client Quit)]
roconnor has joined #nixos
roconnor has quit [(Client Quit)]
pie___ has quit [(Remote host closed the connection)]
pie_ has joined #nixos
pie_ has quit [(Changing host)]
pie_ has joined #nixos
indi_ has joined #nixos
ryanartecona has quit [(Quit: ryanartecona)]
hamishmack has joined #nixos
magnetophon1 has quit [(Ping timeout: 248 seconds)]
thc202 has quit [(Ping timeout: 260 seconds)]
justelex_ has quit [(Ping timeout: 240 seconds)]
markus1199 has joined #nixos
bennofs has quit [(Ping timeout: 240 seconds)]
markus1189 has quit [(Ping timeout: 240 seconds)]
roconnor has joined #nixos
marsel has quit [(Ping timeout: 260 seconds)]
magnetophon1 has joined #nixos
JagaJaga has quit [(Ping timeout: 260 seconds)]
nix-gsc-io`bot has joined #nixos
<nix-gsc-io`bot> Channel nixos-17.03-small advanced to https://github.com/NixOS/nixpkgs/commit/b752a01eda (from 82 minutes ago, history: https://channels.nix.gsc.io/nixos-17.03-small)
nix-gsc-io`bot has quit [(Client Quit)]
ng0 has quit [(Quit: Alexa, when is the end of world?)]
NixOS_GitHub has joined #nixos
<NixOS_GitHub> [nixpkgs] globin pushed 2 new commits to openssl-1.1: https://git.io/vQi9c
<NixOS_GitHub> nixpkgs/openssl-1.1 63a54ee Robin Gloster: qt5.qtbase: disable bindnow
<NixOS_GitHub> nixpkgs/openssl-1.1 5066963 Robin Gloster: qt5.qtbase: patch for openssl 1.1
NixOS_GitHub has left #nixos []