<clever>
copumpkin: pkgs.makeInitrd is pretty equal to the others, the contents attr lets you set non-store things, like /init being a symlink to the storepath, then it just gets the entire closure automagicaly
<clever>
copumpkin: but make-squashfs.nix is weird, in that it always strips the /nix/store/ path from things, and expects the FS to be mounted to /nix/store/
<clever>
copumpkin: and the way you call make-squashfs.nix is also a bit odd, rather then being 2 lambda's like fetchurl and friends, its a single lamba, and it has to be called in a more specialized way
erasmas has quit [(Quit: leaving)]
<NixOS_GitHub>
[nixpkgs] Mic92 pushed 1 new commit to release-16.09: https://git.io/vDxgh
<NixOS_GitHub>
nixpkgs/release-16.09 cac0a75 Lorenzo Manacorda: Make more visible which fetch approach is best...
<NixOS_GitHub>
[nixpkgs] Mic92 closed pull request #23057: Backport "Make more visible which fetch approach is best" (release-16.09...fetch-cherry-pick) https://git.io/vDxUV
<copumpkin>
and it's not necessarily just stdenv, right?
<copumpkin>
hmm okay but I haven't changed any of that
<clever>
copumpkin: because of these 2 lines, the stdenv will be in the runtime closure of the test vm sandbox, which means it wont have to download the stdenv when rebuilding things under a vm
shlevy has quit [(Ping timeout: 260 seconds)]
<MichaelRaskin>
Does't exportReferencesGraph list the build-dependencies of Nix?
<copumpkin>
yeah, but that isn't necessarily the full build-time closure of a nixos-rebuild
jluttine has quit [(Ping timeout: 240 seconds)]
<MichaelRaskin>
When listing nixClosure in tools.nix
<copumpkin>
MichaelRaskin: I thought it was just runtime
<copumpkin>
MichaelRaskin: either way, I still have that part
<clever>
yeah, i believe exportReferencesGraph is just runtime
<copumpkin>
furthermore I even use exportReferencesGraph on system closure
shlevy has joined #nixos
<copumpkin>
so my question is mostly "what have I changed that would break that?" I guess
<copumpkin>
and it doesn't feel like I've changed anything that would affect what goes into the closure
<clever>
copumpkin: oh, another trick nixos-install uses, NIX_OTHER_STORES and a bind-mount to /mnt/root/nix of the host nix
<copumpkin>
yeah, I copied that, although it shouldn't matter too much
<clever>
copumpkin: this allows the chroot env to access the entire store of the "host" (which will be the sandbox of the nix build in your case)
<copumpkin>
yep
<copumpkin>
I don't keep a /mnt/tmp/root though
<clever>
there is a new option i was thinking of, that might play into things
<copumpkin>
keep in mind that this failure is after the installed machine boots up
ibrahims has joined #nixos
<copumpkin>
so none of my environment variables affect anything
<clever>
i dont remember the syntax, but nix gained support to do limited operations against a storepath not at /
<copumpkin>
my script has long finished executing
eacameron has quit [(Remote host closed the connection)]
<clever>
ah
<copumpkin>
that sounds interesting for simplifying the installer later
<copumpkin>
but I mostly just want this thing to work right now :)
eacameron has joined #nixos
<clever>
from what i remember, you can just tell nix-store to operate on a normal nix store, that is currently mounted at /mnt/nix/store, and it will just prepend /mnt/ to all paths when doing binary-substituion, and accessing db.sqlite
<clever>
it cant do builds via that, but it can do the bulk of the operation
<clever>
though maybe builds do work, via normal sandboxing?, or it can build on the host, and copy over
<clever>
when nix commands are ran as root, all operations go thru an instance of LocalStore, that just directly manipulates db.sqlite and /nix/store/
<copumpkin>
yeah
<copumpkin>
I remember there was some sort of environment variable for the store location
<copumpkin>
but it didn't seem to work for me, but perhaps I was just on an older version of nix
takle_ has joined #nixos
<clever>
LocalFSStore may also be related
<clever>
RemoteStore is for non-root access, proxying everything via a unix socket to nix-daemon
ibrahims has quit [(Ping timeout: 240 seconds)]
<copumpkin>
I see
<ndowens08>
hmm can't substituteinplace change more than one file like substituteInPlace file1 file2 --replace
slack1256 has joined #nixos
<clever>
and then nix-daemon just internaly maps everything to the LocalStore instance inside itself
<clever>
while applying some permission restrictions
eacameron has quit [(Ping timeout: 260 seconds)]
<clever>
both of those require /nix/store to be at /nix/store/
<copumpkin>
yeah
<copumpkin>
any other thoughts on my build-time closure problem?
<copumpkin>
I'll resume trying to debug it in a bit
<clever>
BinaryCacheStore is a recent addition, that treats the binary cache as another store, allowing you to just throw copy-closure logic at it, and get substitutes
takle has quit [(Ping timeout: 260 seconds)]
<clever>
class LocalStore : public LocalFSStore
pmade has joined #nixos
<copumpkin>
I wonder if I'm putting the right closure into the target box...
<MichaelRaskin>
I think I have actually seen machines after nixos-install needing internet connection to nixos-rebuild
<copumpkin>
MichaelRaskin: but the installer test explicitly tests that
<MichaelRaskin>
Maybe it got better recently
<copumpkin>
my PR just broke it and I don't know why :)
<MichaelRaskin>
Hm
<MichaelRaskin>
Maybe if you build the system configuration outside the target, the target never needs to build anything during nixos-install?
<copumpkin>
oh!
<MichaelRaskin>
And in the old version it had to build the system instance and so it had to fetch stdenv?
<copumpkin>
yes, of course that's it
<copumpkin>
gah :)
<clever>
class LocalFSStore : public virtual Store
<copumpkin>
stupid
<MichaelRaskin>
I just looked at the list of missing paths and tried to imagine what the graph would be
<copumpkin>
:) it makes sense
<copumpkin>
now I need to figure out how to fix it, since exportReferencesGraph only gives me runtime closure and I can't talk about .drvs in nix
<MichaelRaskin>
drvPath?
<copumpkin>
oh!
<MichaelRaskin>
Which is what iso-image.nix does
<copumpkin>
I'll see if that fixes it, thanks!
<copumpkin>
so I'd just call exportReferencesGraph on derivation.drvPath?
<copumpkin>
clever: maybe with that stuff I can kill fakechroot!
<copumpkin>
I'd love to do that :)
<clever>
copumpkin: yeah
* copumpkin
gets it working first, then cleans it up
<MichaelRaskin>
Yep, looks like that
<MichaelRaskin>
Note that you will pull much more than the minimum needed
<copumpkin>
how does that differ from building in the target machine?
<copumpkin>
I'm thinking I'll just make it optional
ibrahims has joined #nixos
<MichaelRaskin>
Well, transitive build dependencies include static bootstrap tools tarball
<copumpkin>
so you can ask for a system that can rebuild itself offline
<copumpkin>
MichaelRaskin: but most of the stdenv bootstrapping hoops are to remove all runtime dependencies in stdenv from bootstrap tools
<copumpkin>
and isn't a build-time dependency still just pulling in the runtime dependency of the build-time dependency?
<copumpkin>
or is it build-time dependencies all the way down?
<copumpkin>
is it weird that my runInLinuxVM seems to be able to download stuff?
* copumpkin
reruns with build-use-sandbox true :)
<MichaelRaskin>
For drvs it's transitive build-time dependency closure
<copumpkin>
MichaelRaskin: hmm weird, why wouldn't a build-time dependency pull in the runtime dependency of the build-time dependency?
<copumpkin>
if that question makes sense
<MichaelRaskin>
The dependency list of a drv is the list of the drvs for the build depedndencies
<copumpkin>
e.g., package X has build-time dep on Y, which has runtime dep on Z and build-time dep on Q. It seems like if I ask for package X's build-time dependencies, I should get Y and Z, but not Q
<MichaelRaskin>
Now apply that recursively
<copumpkin>
fair enough
<copumpkin>
but when Nix is actually building stuff itself, it's smarter than that?
<MichaelRaskin>
The dependency list of an output is the list of outputs needed in runtime
<copumpkin>
since it doesn't always pull in bootstrap-tools.tgz
<MichaelRaskin>
Wee-eell, you could just instpect the drv
<Drakonis>
copumpkin, it certainly isn't premature to get rid of channels
<hodapp>
oh, so I didn't even need to install tftpd-hpa
<hodapp>
I kept finding absolutely nothing when looking up "nixos tftp"
<copumpkin>
Drakonis: oh okay, I was concerned but thanks to your unjustified assertion now I'm reassured ;)
<Drakonis>
i'm still writing ty
<copumpkin>
:)
<Drakonis>
see here though
<Drakonis>
the current system of even allowing multiple channels is broken if there's disparities between root and user
<Drakonis>
and if you dare to delete a channel
<Drakonis>
nix-env starts to get wack
<copumpkin>
oh, that's niv-env being shitty though
<copumpkin>
what's wrong with having multiple channels?
<Drakonis>
bugs really
<copumpkin>
are they in the tracker?
<Drakonis>
probably?
<copumpkin>
like I pull in different versions of nixpkgs all the time
<copumpkin>
and use them concurrently
<Drakonis>
hm
<copumpkin>
have never had any issues
<hodapp>
ndowens08: thanks, that's what I needed. Why I couldn't find this prior I don't know
<Drakonis>
the introduction of overlays might phase out the need
<Drakonis>
so you can have the current and assorted packages
<copumpkin>
I keep different projects with locked-down nixpkgs references
<copumpkin>
anyway, I see nothing really making multiple channels a bad idea so far
<copumpkin>
they're roughly equivalent to branches in git
<copumpkin>
I need a mutable pointer into an immutable system
<copumpkin>
and I want to be able to control when it gets updated
<Drakonis>
hm
<Drakonis>
what about using it locally?
<copumpkin>
how do you mean?
<Drakonis>
o boy
<Drakonis>
nix query-options
<Drakonis>
this will be my favorite
Sonarpulse has quit [(Ping timeout: 260 seconds)]
ibrahims has joined #nixos
ibrahims has quit [(Ping timeout: 260 seconds)]
lambdamu_ has joined #nixos
lambdamu has quit [(Read error: Connection reset by peer)]
Drakonis has quit [(Read error: Connection reset by peer)]
Drakonis_ has joined #nixos
ibrahims has joined #nixos
<copumpkin>
shlevy: hi :)
ibrahims has quit [(Ping timeout: 240 seconds)]
Rotaerk has joined #nixos
fhoffmeyer has joined #nixos
seanz has quit [(Quit: Leaving.)]
MP2E has joined #nixos
mbrgm has quit [(Ping timeout: 240 seconds)]
fhoffmeyer has quit [(Ping timeout: 260 seconds)]
mbrgm has joined #nixos
phreedom has quit [(Ping timeout: 260 seconds)]
phreedom has joined #nixos
<ndowens08>
Drakonis_: yup i seen that, which is gonna be nice instead of having to look at the nix file
<lambdamu_>
sounds stupid, but how does one actually log into a vm created with "nixos-rebuild build-vm", that is get past the display manager, nothing works not even creating an extra user with an explict password
derjohn_mob has quit [(Ping timeout: 240 seconds)]
<lambdamu_>
tried slim and lightdm
<ndowens08>
lambdamu_: sounds as it isn't using the configuration file
<ndowens08>
are you editing the current config or create a copy and edit it?
<ndowens08>
mine might look a bit diff, i use more brackets than some shows.
<lambdamu_>
hm ok thanks i'll try with your config
<ndowens08>
i break it down, so i could just add user w/o having to type out extra users.extraUsers
eacameron has joined #nixos
<ndowens08>
lambdamu_: good luck, i got to go to bed
<lambdamu_>
ndowens08: gn8 and ty
<ndowens08>
lambdamu_: welcome :)
<lambdamu_>
actually it works now
<lambdamu_>
not really sure why
hexagoxel has quit [(Ping timeout: 260 seconds)]
Drakonis_ has quit [(Quit: Leaving)]
eacameron has quit [(Ping timeout: 240 seconds)]
sellout- has joined #nixos
hexagoxel has joined #nixos
ndowens08 has quit [(Ping timeout: 260 seconds)]
systemfault has joined #nixos
eacameron has joined #nixos
ibrahims has joined #nixos
eacameron has quit [(Ping timeout: 255 seconds)]
ibrahims has quit [(Ping timeout: 260 seconds)]
fhoffmeyer has joined #nixos
mkoenig has joined #nixos
fhoffmeyer has quit [(Ping timeout: 255 seconds)]
takle_ has quit [(Remote host closed the connection)]
indi_ has quit [(Remote host closed the connection)]
indi_ has joined #nixos
ibrahims has joined #nixos
<NixOS_GitHub>
[nix] copumpkin opened pull request #1246: Explicitly model all settings and fail on unrecognized ones (master...rework-options) https://git.io/vDxHE
alx741 has joined #nixos
ibrahims has quit [(Ping timeout: 260 seconds)]
<copumpkin>
as always, I welcome feedback on the PR I just opened up there :)
<NixOS_GitHub>
[nixpkgs] lukego opened pull request #23070: r-modules: Updated according to procedure in README (BROKEN) (master...R-update-broken) https://git.io/vDxdc
Intensity has quit [(Ping timeout: 256 seconds)]
AllanEspinosa has quit [(Ping timeout: 268 seconds)]
slack1256 has quit [(Ping timeout: 240 seconds)]
systemfault has joined #nixos
takle_ has joined #nixos
takle_ has quit [(Ping timeout: 240 seconds)]
ibrahims has joined #nixos
edvorg has joined #nixos
ibrahims has quit [(Ping timeout: 240 seconds)]
uralbash has quit [(Ping timeout: 240 seconds)]
takle has joined #nixos
fhoffmeyer has joined #nixos
takle has quit [(Ping timeout: 240 seconds)]
fhoffmeyer has quit [(Ping timeout: 240 seconds)]
proteusguy has quit [(Remote host closed the connection)]
<LnL>
oh right, you can't reverence boot.specialFileSystems yet
<srk>
ok..
goibhniu has joined #nixos
MarcelineVQ has quit [(Ping timeout: 260 seconds)]
MarcelineVQ has joined #nixos
MarcelineVQ has quit [(Remote host closed the connection)]
phreedom has quit [(Ping timeout: 268 seconds)]
Shou has joined #nixos
<srk>
its fine during boot but causes warning during rebuild switch - mount: unknown filesystem type 'ramfs'
mkoenig has quit [(Ping timeout: 260 seconds)]
markus1219 has quit [(Ping timeout: 260 seconds)]
markus1189 has quit [(Ping timeout: 240 seconds)]
<srk>
think I should just make a few pull request(s) as well
ckauhaus has joined #nixos
fendor has quit [(Quit: Page closed)]
grantwu has joined #nixos
<grantwu>
Has anyone experienced stack failing with <command line>: can't load .so/.DLL for: libgmp.so (libgmp.so: cannot open shared object file: No such file or directory) for stack exec -- ghci, and stack setup failing with http://lpaste.net/8694464595516981248
Shou has quit [(Ping timeout: 260 seconds)]
samvher has joined #nixos
<samvher>
Does anyone here have experience deploying nixos on amazon EC2? I was looking here: https://nixos.org/wiki/NixOS_on_Amazon_EC2 but the public AMI's link listed there is broken and it says EC2 support is a work in process, so I'm wondering if it's my best option. I've been developing a website in a virtualbox nixos machine, and I'm hoping to deploy it easily
eacameron has joined #nixos
samvher has quit [(Ping timeout: 260 seconds)]
eacameron has quit [(Ping timeout: 260 seconds)]
samvher has joined #nixos
ibrahims has joined #nixos
eacameron has joined #nixos
MarcelineVQ has joined #nixos
pi3r has joined #nixos
markus1189 has joined #nixos
mojjo has quit [(Remote host closed the connection)]
markus1199 has joined #nixos
<timor>
Should pull requests always be made against master?
<dmj`>
samvher: did you try to deploy it to aws?
eacameron has quit [(Ping timeout: 260 seconds)]
<dmj`>
timor: "For pull-requests, please rebase onto nixpkgs master."
<samvher>
dmj`: I'm still figuring out how to do all of this, I thought to deploy I would set up an EC2 instance with nixos
<dmj`>
grantwu: that is an error from ghc that has been installed w/ stack. Trying installing ghc with nix, “nix-shell -p haskell.compiler.ghc802"
<timor>
dmj`: thanks
<samvher>
but if I have to build my own instance I'm not sure that's going to make my life easier, so I was hoping I could find something ready-made
<dmj`>
timor: np
<dmj`>
samvher: you won’t have to, nixops will use the AWS api to do all of that for you
<dmj`>
samvher: you could try GCE instead too
<grantwu>
What does nix-shell -p do? is that different than nix-env -i o.O
<samvher>
ah I can use nixops for that? actually I'm not at home in nixops yet, not even sure what it does exactly
<dmj`>
grantwu: if you read the man page (type /-p after you enter), it will show you that -p stands for “packages”. So this is going to put you into a shell where ghc802 is present
<samvher>
I just have a nixos installation in a virtualbox environment, so I have a configuration.nix that should be relatively straightforward to move to somewhere else
<dmj`>
samvher: be brave
<dmj`>
nix-env -iA nixops -f ‘<nixpkgs>'
<dmj`>
samvher: if you’re deploying to virtualbox, aren’t you using nixops?
<dmj`>
oh I’m sorry, I read developing as deploying
<samvher>
dmj`: no I just installed nixos from a disk image :)
<grantwu>
Wait, so I have to install ghc with nix?
<dmj`>
grantwu: why use stack if you you’re using nix
<dmj`>
grantwu: nixpkgs follows LTS
<grantwu>
Because I may have existing stack workflows?
Ainieco has joined #nixos
<Ainieco>
hello
<dmj`>
grantwu: nix workflows > stack workflows
<samvher>
I'm also using stack actually
markus1189 has quit [(Ping timeout: 260 seconds)]
<Ainieco>
how much time should pass from merged PR in https://github.com/NixOS/nixpkgs for the package to be available in stable nixos?
markus1199 has quit [(Ping timeout: 255 seconds)]
<dmj`>
stack and cabal-install are front ends for resolving package versions. If nix handles the resolving, it pretty much obviates those tools. Nix builds haskell packages with a GHC setup script
<samvher_>
I didn't manage to find any alternatives
<ikwildrpepper>
samvher_: nixpkgs/nixos/modules/virtualisation/ec2-amis.nix has the latest ami ids
<samvher_>
ikwildrpepper: ah thanks, that looks like it will be helpful
xadi has joined #nixos
<ikwildrpepper>
samvher_: make sure to use an ec2 keypair that you have the ssh private key from
<ikwildrpepper>
(when creating the instance)
<ikwildrpepper>
otherwise you cannot log in to the machine ;)
<samvher_>
right :)
mudri has joined #nixos
Itkovian_ has joined #nixos
arjen-jonathan has joined #nixos
Itkovian has quit [(Ping timeout: 268 seconds)]
takle has joined #nixos
mudri has quit [(Ping timeout: 240 seconds)]
mudri has joined #nixos
Itkovian_ has quit [(Quit: My MacBook has gone to sleep. ZZZzzz…)]
fre has quit [(Ping timeout: 240 seconds)]
Itkovian has joined #nixos
proteusguy has quit [(Remote host closed the connection)]
bennofs has joined #nixos
samvher has quit [(Ping timeout: 255 seconds)]
<samvher_>
dmj`: I just started looking a bit more into using nix rather than stack, and it seems a lot easier to switch than I expected + rather pleasant, thanks
<dmj`>
Nice
<samvher_>
I generated a shell.nix with cabal2nix, most of that seems to work fine but I'm getting different versions than stack uses - you mentioned nixpkgs follows lts, is there some way I can set my lts version?
fre has joined #nixos
Itkovian has quit [(Quit: My MacBook has gone to sleep. ZZZzzz…)]
Shou has quit [(Ping timeout: 260 seconds)]
markus1189 has joined #nixos
Shou has joined #nixos
fre2 has joined #nixos
fre2 has quit [(Client Quit)]
fre has quit [(Read error: Connection reset by peer)]
ibrahims has quit [(Ping timeout: 260 seconds)]
Itkovian has joined #nixos
ibrahims has joined #nixos
fhoffmeyer has joined #nixos
Itkovian has quit [(Client Quit)]
bara1 has quit [(Remote host closed the connection)]
lsix has joined #nixos
Itkovian has joined #nixos
bara has joined #nixos
goibhniu has quit [(Ping timeout: 260 seconds)]
Itkovian has quit [(Client Quit)]
arjen-jonathan has quit [(Quit: WeeChat 1.7)]
fhoffmeyer has quit [(Ping timeout: 255 seconds)]
deepfire has joined #nixos
<bennofs>
samvher_: any given nixpkgs revision only supports a single stackage snapshot (lts release) i believe
<samvher_>
bennofs: ah ok, thanks
deepfire has quit [(Ping timeout: 240 seconds)]
watersoul has joined #nixos
Itkovian has joined #nixos
tommynono111 has joined #nixos
Itkovian has quit [(Client Quit)]
<tommynono111>
Hello. I just upgraded my unstable NixOS installation and I'm getting strange error, when trying to run command with sudo. Error is: sudo: /run/current-system/sw/bin/sudo must be owned by uid 0 and have the setuid bit set
<tommynono111>
How should I fix this?
Itkovian has joined #nixos
goibhniu has joined #nixos
<ekleog>
you can relogin ;)
<Ainieco>
how much time should pass from merged PR in https://github.com/NixOS/nixpkgs for the package to be available in stable nixos?
<ekleog>
it's due to the setuid update iirc, bug reported last sunday
proteusguy has joined #nixos
ibrahims has quit [(Ping timeout: 260 seconds)]
tommynono111 has quit [(Client Quit)]
<bennofs>
Ainieco: stable does not generally get package updates, only security updates
<bennofs>
Ainieco: so it would generally take a new stable nixos release
<NixOS_GitHub>
[nixpkgs] grahamc pushed 1 new commit to master: https://git.io/vDpoO
<NixOS_GitHub>
nixpkgs/master cc4919d Graham Christensen: xen: patch for XSAs: 197, 199, 207, 208, 209...
<NixOS_GitHub>
[nixpkgs] grahamc pushed 1 new commit to release-16.09: https://git.io/vDpoH
<NixOS_GitHub>
nixpkgs/release-16.09 142fd06 Graham Christensen: xen: patch for XSAs: 197, 199, 207, 208, 209...
<gchristensen>
^ xen has a great XSA process, <3 them.
Wizek__ has joined #nixos
eddyb_ has joined #nixos
eddyb has quit [(Ping timeout: 252 seconds)]
tnias has quit [(Ping timeout: 252 seconds)]
eddyb_ is now known as eddyb
tnias has joined #nixos
<fpletz>
gchristensen: haven't you heard? they won't publish XSA for every CVE in the future, like information leaks :/
<gchristensen>
oh.
<gchristensen>
well, that is annoying
<NixOS_GitHub>
[nixpkgs] 7c6f434c pushed 2 new commits to master: https://git.io/vDpK2
<NixOS_GitHub>
nixpkgs/master a8bf876 Michael Raskin: kde5.applications.kig: init at 16.12.2
<NixOS_GitHub>
nixpkgs/master 194d137 Michael Raskin: wireshark: patch for CVE-2017-6041
griff_ has quit [(Quit: griff_)]
<gchristensen>
but their page that lists all the patches with descriptions and impact and links to patches with instructions on which ones to use: very, very well done
<NixOS_GitHub>
[nixpkgs] 7c6f434c pushed 1 new commit to release-16.09: https://git.io/vDpKX
<NixOS_GitHub>
nixpkgs/release-16.09 6e17350 Michael Raskin: wireshark: patch for CVE-2017-6041...
ookhoi has quit [(Quit: for another day, then :-))]
<joepie91>
(had to target the .cloudfront.net for the latter since apparently there's no IPv6 records for cache.nixos.org or so? it complained at me anyway)
<gchristensen>
cloudfront's availability is 99.9%
<gchristensen>
and there is nothing for us to do here, unfortunately
<joepie91>
right, just trying to narrow down the problem :)
<joepie91>
it really does seem that the perf issue is on cloudfront's side
xeviox has quit [(Ping timeout: 260 seconds)]
<gchristensen>
it probably is, 100% of it is out of our hands, cloudfront is backed by s3
<LnL>
I wonder if it's related to the size of our bucket
thoughtpolice has quit [(Ping timeout: 240 seconds)]
<the-kenny>
Ugh. I'm too out of the loop for c++ now. What causes our gcc to complain that 'not' isn't defined (as in if(not some_fun()) instead of if(!some_fun())) ?
<copumpkin>
nah
<copumpkin>
buckets get arbitrarily large
<copumpkin>
niksnut: what is ours, like a TB or two?
<LnL>
yeah in theory
<copumpkin>
I dunno, I have some pretty big ones for work
<NixOS_GitHub>
[nixpkgs] league opened pull request #23075: logisim: use better jar wrapper, closes #23068 (master...logisim-args) https://git.io/vDp6u
<LnL>
copumpkin: I think it's x10
takle has quit [(Remote host closed the connection)]
<joepie91>
I'd be surprised if bucket size affects this
<joepie91>
like, in a naive implementation of storage it could
<joepie91>
but I somehow doubt that AWS uses a naive implementation :P
<copumpkin>
S3 backs most of the internet nowadays
<copumpkin>
:P
<copumpkin>
including much of AWS's own service
thoughtpolice has joined #nixos
chakerbenhamed has left #nixos ["WeeChat 0.4.2"]
xeviox has joined #nixos
<copumpkin>
EC2, dynamodb, and S3 are the dogfood building blocks
<gchristensen>
it does not
<niksnut>
copumpkin: 40 TB
<joepie91>
(not that I consider AWS to be a fantastically performant and high-quality service, because I *know* it isn't, but come on, they have competent techs working there... and given that scalability and integration is literally their selling point, the first thing to fix would be "bucket size affects perf")
<copumpkin>
niksnut: oh wow, bigger than I thought, but it should still be able to handle that fine
<gchristensen>
and our file naming scheme also avoids hot key issues that S3 can sometimes have.
mudri has joined #nixos
<gchristensen>
40TB is nothing for S3.
<copumpkin>
niksnut: is logicblox getting sick of paying for that? :P
* copumpkin
sends another donation to the foundation
<niksnut>
the-kenny: is "not" a trigraph? I seem to remember they were removed from C++ recently
<niksnut>
well, I should probably revive garbage collection for the bucket
<joepie91>
... removed? C++ removes things?
<joepie91>
this is news to me :)
<the-kenny>
niksnut: Yeah, it's a trigraph. I didn't know they were removed. Boost still uses them in some places, causing digikam5 to fail.
<rfish>
Hello, I wanna use a program that *has to* edit the /etc/hosts. Someone told me to use buildFHSUserEnv, but the /etc/hosts is read-only inside the sandbox. Do you know why?
<copumpkin>
buildFHSUserEnv I think makes things read-only
<copumpkin>
there's the other one which might not be read-only but is uglier
ibrahims has joined #nixos
<rfish>
copumpkin: Oh! Do you think about buildFHSChrootEnv?
<copumpkin>
I think that might be it
<copumpkin>
but does it really _have_ to write to /etc/hosts?
<copumpkin>
patch that out :P
<copumpkin>
it seems like unfriendly behavior
<rfish>
copumpkin: buildFHSChrootEnv has been delete in PR#160301
<copumpkin>
oh
<copumpkin>
it was kinda awkward, so not necessarily a bad thing
<rfish>
copumpkin: Yhea editing /etc/hosts is weird, but I don't have many choice here. The program which try to do this is a plugin for vagrant (vagrant-hostmanager)
<copumpkin>
it should follow the nss-mymachines lead :P
ibrahims has quit [(Ping timeout: 240 seconds)]
<copumpkin>
not that that works out of the box on NixOS unless you have nscd turned on...
avn has quit [(Ping timeout: 260 seconds)]
<rfish>
copumpkin: I can make an issue on the vagrant project. But in the meanwhile, any idea to let this soft edit my /etc/hosts?
xeviox has quit [(Ping timeout: 260 seconds)]
<NixOS_GitHub>
[nixpkgs] vcunat pushed 1 new commit to master: https://git.io/vDpy9
<NixOS_GitHub>
nixpkgs/master a048495 Vladimír Čunát: fstrm: init at 0.3.1
Wizek__ has quit [(Ping timeout: 268 seconds)]
<copumpkin>
rfish: you could make it think it's editing /etc/hosts with a separate namespace, but for that to really be useful you want it showing up in the host...
Wizek has quit [(Read error: Connection reset by peer)]
<copumpkin>
which won't work because /etc/hosts is in the store in NixOS
<copumpkin>
so I don't think it can work
<gchristensen>
you could probably, in the fhs, delete /etc/hosts and recreate it
<copumpkin>
rfish: well, you could write an nscd plugin that accepts additional /etc/hosts files and merges them into the overall view of hosts
<rfish>
copumpkin: I first try to use overlayfs to do that but /etc/hosts still appear to be read-only
<copumpkin>
and then write a wrapper of your program that unshares filesystem mount, bind mounts your custom file, and then informs your nss library that the other file is a /etc/hosts-alike
<rfish>
gchristensen: thx, I will give it a try
<copumpkin>
you can definitely move it out of the way, but any run of the activation script will put it back in place
<rfish>
gchristensen: copumpkin: I try both. Thx for your help.
<copumpkin>
mine sounds like quite a bit of work, tbc
<copumpkin>
would probably be easier to just add an nss plugin to the upstream project :)
<rfish>
copumpkin: :D
<copumpkin>
domenkozar: omg travis builds for nix
<copumpkin>
oh I guess it isn't a build
<copumpkin>
still cool though
cpennington has joined #nixos
flemming has quit [(Ping timeout: 240 seconds)]
<NixOS_GitHub>
[nixpkgs] wizeman opened pull request #23078: nixos.timezone: set TZ to :/etc/localtime (master...u/fix-tz-stat) https://git.io/vDpHK
<sheenobu>
winehq is pretty dead.. hrm. anyone using a gamepad with wine in NixOS?
ibrahims has joined #nixos
ckauhaus has quit [(Remote host closed the connection)]
fhoffmeyer has joined #nixos
sellout- has quit [(Ping timeout: 260 seconds)]
sellout-1 has joined #nixos
lsix_ has quit [(Quit: Konversation terminated!)]
stepcut has quit [()]
<NixOS_GitHub>
[nix] edolstra pushed 1 new commit to master: https://git.io/vDp5b
MercurialAlchemi has quit [(Ping timeout: 240 seconds)]
ixxie has joined #nixos
Guest85325 has quit [(Ping timeout: 255 seconds)]
erasmas has joined #nixos
Shou has joined #nixos
<gchristensen>
anyone know of a way to pass initial commands to to the test runner driver? I want a way to automatically run the driver, passing `startAll; $machine->forwardPort(2223, 22);`
<savanni>
A hopefully short question: do any of you know how to add the canon printer drivers to the printing configuration? I can see the drivers sitting in the package tree, but don't know the address to them.
<NixOS_GitHub>
[nixpkgs] nixy opened pull request #23080: linode-cli: init at 1.4.7 (master...init/linode-cli) https://git.io/vDhUI
<gchristensen>
savanni: I never bothered to figure out how :(, since I could lean on my wife's computer :(
<savanni>
Hmmm.
<savanni>
The office printer is a Canon and is the only printer I've access to.
<savanni>
Although I can always try the generic drivers.
<NixOS_GitHub>
[nixpkgs] Kendos-Kenlen opened pull request #23081: Rubymine (master...rubymine) https://git.io/vDhLG
reinzelmann has quit [(Quit: Leaving)]
xadi has quit [(Ping timeout: 240 seconds)]
griff_ has joined #nixos
xadi has joined #nixos
roconnor has joined #nixos
fhoffmeyer has joined #nixos
<NixOS_GitHub>
[nixpkgs] gebner pushed 1 new commit to master: https://git.io/vDhO6
<NixOS_GitHub>
nixpkgs/master b66ec60 Gabriel Ebner: idris: jailbreak...
<NixOS_GitHub>
[nixpkgs] vcunat pushed 1 new commit to staging: https://git.io/vDhOi
<NixOS_GitHub>
nixpkgs/staging 2f1945d Vladimír Čunát: python-3.6: fix random numbers with glibc-2.25...
Shou has joined #nixos
<wrl>
hey can nixos do multilib systems? i.e. 64bit but has the requisite 32bit libs for running 32 bit binaries
fhoffmeyer has quit [(Ping timeout: 240 seconds)]
<simpson>
wrl: Yes.
<wrl>
simpson: are there any docs on that?
<simpson>
Well, only amd64 + i686 AFAIK.
<wrl>
that's my usecase so
<simpson>
I can't wait for the days when we want to have multilib ARM. That'll be "fun".
<savanni>
Well, on the print escapade, I got past the point of error messages. But then nothing actually got to the printer. :(
sellout- has quit [(Quit: Leaving.)]
<simpson>
wrl: Generally packages will depend on the correct multilibs as needed. There's a few options in configuration.nix for setting up 32-bit GL libraries.
<gchristensen>
(the \s+ at the beginning was changed to \s* to account for name lines with no leading space, and the ^ anchors it to the beginning of a line) also you may need to add the multi-line flag to the python code?
<FRidh>
that was using text[slice(*re.compile('\s+name\s+=\s+"(?P<name>.*)";').search(text).span())]
<FRidh>
gchristensen: using your new regex with re.MULTILINE removes the \n, that's good. There's some whitespace left though
<FRidh>
actually, it doesn't work anymore
<gchristensen>
well you could remove the ^\s* and see where that gets you
<FRidh>
is there a "ignore everything until you see" ?
<NixOS_GitHub>
[nixpkgs] vcunat pushed 2 new commits to staging: https://git.io/vDhWm
<NixOS_GitHub>
nixpkgs/staging b1c6a9b Tim Steinbach: curl: 7.52.1 -> 7.53.0
<FRidh>
can also just cheat and use search with the regex per line
<NixOS_GitHub>
[nixpkgs] vcunat pushed 1 new commit to release-16.09: https://git.io/vDhW2
<NixOS_GitHub>
nixpkgs/release-16.09 6fcd2d2 Tim Steinbach: curl: 7.52.1 -> 7.53.0 (#23063)...
<gchristensen>
thaht'll be what happens if you remove the ^\s,which is saying "Start at the beginning (^)" and then "probably some whitespace (\s* means expect a whitespace character (\s) zero or more times (*))
wak-work has joined #nixos
[0x4A6F] has joined #nixos
cpennington has joined #nixos
danharaj has joined #nixos
<NixOS_GitHub>
[nixpkgs] vcunat pushed 1 new commit to master: https://git.io/vDhl6
<NixOS_GitHub>
nixpkgs/master 838e29d Vladimír Čunát: Merge branch 'staging'...
Sonarpulse has joined #nixos
ibrahims has quit [(Ping timeout: 240 seconds)]
rfish has quit [(Remote host closed the connection)]
ibrahims has joined #nixos
flemming has joined #nixos
Shou has quit [(Ping timeout: 240 seconds)]
fhoffmeyer has joined #nixos
teknico has left #nixos []
arjen-jonathan has quit [(Ping timeout: 260 seconds)]
mkoenig_ has quit [(Read error: Connection reset by peer)]
<copumpkin>
domenkozar: btw, hercules is exciting :)
fhoffmeyer has quit [(Ping timeout: 240 seconds)]
<LnL>
what's the state of hercules?
<copumpkin>
I dunno, it seems to be progressing
avn_ has quit [(Ping timeout: 260 seconds)]
nschoe has quit [(Quit: Program. Terminated.)]
<domenkozar>
really really early stage :)
davidak has joined #nixos
mkoenig has joined #nixos
griff_ has quit [(Ping timeout: 240 seconds)]
<jophish_>
copumpkin: It's not a thing yet!
flemming has quit [(Ping timeout: 240 seconds)]
<domenkozar>
jophish_: is currently rocking that stage, but I expect to be able to contribute more in a month
<eacameron>
LnL: I'm trying to use your nix-docker repo but I can't figure out how to SSH in for the first time to swap out the keys.
<eacameron>
LnL: readme says "chmod 600 ssh/insecure_rsa" but I'm not sure where that file is...
<garbas>
oh shit i didn't ran the spell checker, huh
<FRidh>
gchristensen: this is funny...if I copy the nix expression in pytex, it matches. But if I read the expression with open(...,'r').read() it won't match...
hotfuzz_ has quit [(Ping timeout: 268 seconds)]
<copumpkin>
garbas: yay, thanks. Is there a twitter account I can retweet that tweets these?
<bennofs>
I'm getting 500 internal server error frm cloudfront
<bennofs>
Is this some cache invalidation issue perhaps?
<garbas>
copumpkin: it should get retweeted via @nixos_org once rss gets picked up
<copumpkin>
jophish_: so I shouldn't talk about it :)
layus_ has joined #nixos
<bennofs>
wow, now I get a 404 for the same request
<eacameron>
garbas: Hah...had no idea people actually saw my twitter post about the Louisville Haskell meetup talk on Nix.
<domenkozar>
bennofs: did you see my nix-channelstatus project? :)
<garbas>
eacameron: i see everything :)
<eacameron>
garbas: Apparently!
<bennofs>
domenkozar: yeah, took a look at it. I'm working on nixpkgs-locate more as a project to learn a bit about rust, so your project will probably be usable much sooner ;)
<FRidh>
gchristensen: nevermind. Had to use .fromall instead of .search
<gchristensen>
ah :)
layus has quit [(Ping timeout: 240 seconds)]
<domenkozar>
bennofs: oh it's usable already, just need to find time to create nixos module for it
<bennofs>
domenkozar: also, I want to experiment with compressing the index a bit
<domenkozar>
oh yeah it's 1GB stored in postgresql
<domenkozar>
need to look into that as well
<bennofs>
domenkozar: my plain-text path list is 200MB
<domenkozar>
yeah this is with indexes, etc
<domenkozar>
and not normalized
griff_ has joined #nixos
<bennofs>
domenkozar: locate-db style (front-compressed, i.e. shared prefix are encoded efficiently) is ~50MB. XZ compressed is 10MB
<domenkozar>
bennofs: I want to be able to search libs
<domenkozar>
and then magically try out autopatchelf implementation
<domenkozar>
that would parse ldd output and figure out packages on it's own
<bennofs>
domenkozar: I want to try some custom compression, like encoding common directory names shorter (there's probably a lot of "share" in there :)
<bennofs>
domenkozar: it would be really nice if store-paths.xz included the hydra job name :/
<bennofs>
domenkozar: I'm using nix-env -qaP --no-name --out-path --xml right now to get store path + nixpkgs attribute
<domenkozar>
bennofs: yeah I have to add that, basically you have to evaluate everything
<domenkozar>
indeed
<domenkozar>
that's still missing for my implementation
<bennofs>
but that gives ~12k paths, while store paths is about 36k
<jophish_>
copumpkin: I'll want you to shout about it when it hits 1.0 :)
<copumpkin>
will do :)
<domenkozar>
bennofs: hydra uses special evaluator
<domenkozar>
copumpkin: you can shout out to those willing to help :)
<copumpkin>
I think there are a few things that are currently in hydra but should be in Nix
<NixOS_GitHub>
[nixpkgs] vcunat pushed 1 new commit to master: https://git.io/vDhu7
<NixOS_GitHub>
nixpkgs/master f5eea8b Vladimír Čunát: libevent: apply security patches from Debian...
<bennofs>
domenkozar: which makes about sense, since we build for 3 architectures
<copumpkin>
domenkozar, jophish_: one thing I think this would benefit greatly from in the readme would be a "why we're doing this". Since everyone's first question is going to be "isn't Hydra already that? Why make another one?"
<copumpkin>
tbc, I want it myself and have plenty of motivations for wanting this over Hydra, but it might not always be clear
<domenkozar>
copumpkin: indeed
<domenkozar>
copumpkin: will add :)
<copumpkin>
thanks :)
<jophish_>
thanks for the idea copumpkin :)
<NixOS_GitHub>
[nixpkgs] vcunat pushed 1 new commit to release-16.09: https://git.io/vDhzg
<NixOS_GitHub>
nixpkgs/release-16.09 3de44ca Vladimír Čunát: libevent: apply security patches from Debian...
goibhniu has quit [(Ping timeout: 240 seconds)]
<jophish_>
I'm doing it *because* I maintain a hydra server, hehe
<copumpkin>
:P
<domenkozar>
I maintain three..
Guest94867 has joined #nixos
Guest94867 is now known as jgertm
<LnL>
eacameron: did you get it working?
<eacameron>
LnL: I'm trying now. My goal is to write a script that sets up a build slave with a single command. I have users/clients that don't know the first thing about nix but they use macOS and need to build stuff.
<eacameron>
LnL: And unfortunately nixops refuses to build things on virtualbox even if the machine is fully functional...
<NixOS_GitHub>
[nixpkgs] fadenb opened pull request #23082: graylog: update + module plugin support (master...graylog_update) https://git.io/vDhg1
derjohn_mob has quit [(Ping timeout: 260 seconds)]
<LnL>
making it easier to build linux packages on darwin was one of the reasons that I made the image
<eacameron>
LnL: Your work is very much appreciated. :D
jensens has quit [(Ping timeout: 260 seconds)]
<eacameron>
LnL: Why do you need to set up a signing keypair even though the docker image is running builds as root?
<LnL>
don't remember exactly
<LnL>
but I think it's needed because osx has a single user install by default
<LnL>
so the store there is owned by the current user
sheenobu has quit [(Quit: Leaving)]
<eacameron>
LnL: I'll try it without and see how it goes.
[0x4A6F] has quit [(Ping timeout: 260 seconds)]
<eacameron>
LnL: I've been able to avoid it with my vbox-build-slave setup.
<gchristensen>
I've been eyeing that, no idea how we stand
<ekleog>
dccp is =m, so vulnerable
<ekleog>
(cf. gunzip /proc/config.gz)
griff__ has joined #nixos
<LnL>
mass rebuild day, feels like it's been a while
<gchristensen>
LnL: what'd you find?
<LnL>
dunno, my hydra is rebuilding the world
griff_ has quit [(Ping timeout: 255 seconds)]
griff__ is now known as griff_
<ekleog>
I guess removing DCCP wouldn't break anything, though
<ekleog>
(it's an alternative to TCP/UDP/ICMP that I've never heard being used, fwiw)
<gchristensen>
hmm
<gchristensen>
sounds good for unstable, not sure I'm +1 on turning it off on stable though
<ekleog>
(that said, cherrypicking the commit works too, but it's still a huge pile of mostly unused code)
DutchWolfie has quit [(Quit: Konversation terminated!)]
<LnL>
gchristensen: might be libevent
<NixOS_GitHub>
[nixpkgs] vcunat pushed 1 new commit to staging: https://git.io/vDh6d
<NixOS_GitHub>
nixpkgs/staging 753c18e Vladimír Čunát: Merge branch 'master' into staging...
griff_ has quit [(Quit: griff_)]
<gchristensen>
nice
bennofs has joined #nixos
reinzelmann has joined #nixos
<manveru>
deploying via nixops on osx still needs docker, right? :|
<LnL>
nixops will bootstrap the builds when deploying to virtualbox
<manveru>
well, i want to deploy to hetzner
proteus-guy has joined #nixos
<manveru>
when will they finally allow installing nixos on their machines without jumping through hoops :P
<LnL>
then you need to setup a remote builder
<NixOS_GitHub>
[nix] veprbl opened pull request #1247: use --cacert instead of --capath (1.11-maintenance...ssl_fix2) https://git.io/vDhPm
<manveru>
i was afraid you'd say that... is remote building any simpler now than last year
proteusguy has quit [(Ping timeout: 240 seconds)]
sheenobu has quit [(Quit: Leaving)]
<manveru>
well, actually i think it's easier if i just install nixos on hetzner and manage via configuration.nix since it's only one machine
<manveru>
what was the name of the new nixos infect script?
<gchristensen>
manveru: if you have a virtualbox vm in your network, it'll use the vbox vm automatically
<gchristensen>
not great / ideal, but easy
<LnL>
that reminds me, I should redeploy to update my redis instance
<manveru>
nixos-assimilate i think
<philipp[m]>
I try to get an owncloud running (just to get out the contacts and calendars via DAV, so it doesn't have to be pretty) I've set this up as my nginx config http://pastebin.ca/3770877 oc keeps telling me that it can't write in the config directory but that is owned by nginx.
<LnL>
manveru: you can do a manual install on hetzner
<NixOS_GitHub>
[nix] edolstra pushed 2 new commits to 1.11-maintenance: https://git.io/vDhXZ
<NixOS_GitHub>
nix/1.11-maintenance 06b46f6 Dmitry Kalinkin: use --cacert instead of --capath...
<NixOS_GitHub>
[nix] edolstra closed pull request #1247: use --cacert instead of --capath (1.11-maintenance...ssl_fix2) https://git.io/vDhPm
<manveru>
LnL: via installimage ?
<LnL>
that bootloader warning is starting to annoy me
<LnL>
with the recovery machine, or whatever they call it
fhoffmeyer has joined #nixos
<pmade>
What's the best way do "install" a modular configuration.nix setup? Right now I have a makefile that installs all my *.nix files into /etc/nixos but that feels dirty. Is there a good way to put all my files into the store? What are others doing?
<philipp[m]>
I have a git repo with configuration.nix in .gitignore. Then I add a folder for each of my machines with their config in it. Most of that config then gets pulled again from a few shared folders.
<LnL>
you could make a profile for it :p
bfrog has joined #nixos
<eacameron>
LnL: Sweet, my script seems to be working.
ThatDocsLady has quit [(Quit: Arma-geddin-outta-here!)]
arjen-jonathan has quit [(Ping timeout: 240 seconds)]
<pmade>
LnL: Thanks. I guess I have some code to read now ;)
mudri has joined #nixos
<manveru>
gchristensen: how's the vbox supposed to work?
<gchristensen>
manveru: just create it in your network.nix just like you were using nixops to deploy to it
<gchristensen>
it should start vbox, do all the building in that vbox, even for the hetzner
nh2 has joined #nixos
<manveru>
do you have an example? i don't understand :(
<gchristensen>
which part?
<manveru>
i can't even `nix-env -i nixops`
<gchristensen>
wait, what?
<LnL>
gchristensen: oh like that, I was wondering how nixops would magically find the virtualbox machine
<nh2>
when I use the remote build feature (NIX_BUILD_HOOK + NIX_REMOTE_SYSTEMS), the remote system doesn't build in parallel (but the local one does if I don't set the build hook) -- any idea why or what I could do about it?
<gchristensen>
LnL: :)
<manveru>
gchristensen: it fails on pyjwt
<manveru>
`E AttributeError: 'module' object has no attribute '_RSAobj'`
<gchristensen>
well that is broken
derjohn_mob has joined #nixos
<copumpkin>
garbas: it made it to @nixos_org, but the format doesn't make it clear what it is. The tweet just says #03 - Title
arjen-jonathan has joined #nixos
<garbas>
ikwildrpepper: ^^^ we have some problems with weekly -> twitter automation
<ndowens08>
yeah most sites call sway a wayland window manager
<ndowens08>
instead of a compositor
MP2E has joined #nixos
ibrahims has joined #nixos
<nwspk>
Mic92: so I had to build it by myself because it was canceled?
Itkovian has quit [(Quit: My MacBook has gone to sleep. ZZZzzz…)]
hamishmack has quit [(Quit: hamishmack)]
<Mic92>
nwspk: to be honest, I don't know, when channels get updated, when they were canceled they shouldn't be updated.
bennofs1 has quit [(Ping timeout: 255 seconds)]
<Mic92>
gchristensen: do you have more insights how one can check what the build state of a package in a published channel is?
ibrahims has quit [(Ping timeout: 260 seconds)]
<nwspk>
Mic92: okay‥ I'm going to continue reading and now I've got my packet anyway. Thank you (:
bennofs1 has joined #nixos
<dtz>
hey, thought I'd try out unstable in prep for 17.03..... now I'm seeing that many of the manpages on my system are double-compressed: http://ix.io/nHV . Is this my fault / is this a known issue / should I file an issue?
Itkovian has joined #nixos
<Mic92>
nwspk: this is indeed not the average case. usually everything is pre-build
<Mic92>
dtz: I have a open pull request to fix this
<dtz>
and to be clear, double-compressed makes them not work, man says "warning: /path/to/double-compressed-manpage.gz.gz: ignoring bogus filename"
<Mic92>
ndowens08: I would put the opening double quotes on the same line as the equal sign and some other small stuff, but in a gist this is hard to comment.
<ndowens08>
Mic92: just wanted opinion; seems like having indention issues for some reason :)
ilja_kuklic has joined #nixos
simendsjo has quit [(Remote host closed the connection)]
<Mic92>
ndowens08: indention in general looks ok.
nh2 has quit [(Remote host closed the connection)]
ibrahims has joined #nixos
arjen-jonathan has quit [(Ping timeout: 240 seconds)]
HappyEnte has left #nixos []
<LnL>
ndowens08: are you using the builtin nix support or something else?
<sphalerite>
can I do that without modiying the package?
<sphalerite>
ongoing work isn't very useful to me if it's not working right now, my goal is to debug st crashing :p
thc202 has quit [(Ping timeout: 240 seconds)]
<Mic92>
at the moment you can only rebuild the package with debug symbols, for example in an packageOverride
ndowens08 has quit [(Quit: WeeChat 1.7)]
griff_ has joined #nixos
<sphalerite>
I want both st with debugging symbols and libXft with debugging symbols. How can I do that? I know I could use overrideAttrs st {dontStrip = true;}, or st.override {libXft = overrideAttrs libXft {dontStrip = true;};} but how can I combine them?
ibrahims has joined #nixos
ndowens08 has joined #nixos
<ndowens08>
LnL: you're right, forgot there is multiple vims :), maybe vimNox would be good
jgertm has quit [(Ping timeout: 255 seconds)]
ibrahims has quit [(Ping timeout: 240 seconds)]
<LnL>
yeah, there's also vim_configurable that allows you to override a bunch of stuff like your vimrc and plugins
<ndowens08>
LnL: problem is, that nix-env -i vimnox doesnt come up, so I wonder if this module will install it w/o error
<LnL>
the behaviour of nix-env is a little different then other things
<ndowens08>
LnL: good :)
<LnL>
nix-env -i foo, will search through all the packages for one with a name that matches
<ndowens08>
yea tried that
Wizek has quit [(Quit: Leaving)]
Wizek_ is now known as Wizek
<sphalerite>
Mic92: does that libXft change actually work correctly though? Because with .override it's an argument to the expression function, and gets passwd into buildInputs. I don't see how your expression would result in the modified Xft going into buildInputs
<LnL>
if you want similar behaviour to other things, that use attributes you can do something like this nix-env -f '<nixpkgs>' -iA foo
<ndowens08>
but it is listed as nixos.vimNox so maybe it will work anyway
wizeman has quit [(Quit: Connection closed for inactivity)]
<LnL>
using attribute paths is not only faster because it doesn't have to search through all the names, it also works for nested attributes and packages with special names (like vimNox)
<LnL>
ndowens08: nix-instantiate --eval '<nixpkgs>' -A vimNox.name # "vim_configurable-..."
<LnL>
that's the reason nix-env -i can't find vimNox by name
<Mic92>
sphalerite: every package is a function (if you take a look at st/default.nix) and libxft is an argument of that function
<Mic92>
sphalerite: wait a sec
avn is now known as avn_nb
<sphalerite>
Mic92: yes I realise that, but with overrideDerivation you're not overriding the arguments to the function, you're overriding attributes of the set passed to mkDerivation
ibrahims has joined #nixos
<Mic92>
sphalerite: yes, maybe you have to override buildInputs instead or chain an override call after that
<sphalerite>
ooooh yeah I could chain an override call, perfect
<sphalerite>
thanks
indi_ has quit [(Remote host closed the connection)]
nckx has quit [(Ping timeout: 240 seconds)]
nckx has joined #nixos
indi_ has joined #nixos
indi_ has quit [(Remote host closed the connection)]
bfrog has quit [(Ping timeout: 240 seconds)]
indi_ has joined #nixos
ibrahims has quit [(Ping timeout: 260 seconds)]
ryanartecona has quit [(Quit: ryanartecona)]
JagaJaga has quit [(Ping timeout: 260 seconds)]
JagaJaga has joined #nixos
<ndowens08>
LnL: hmm looks there is already a VIM module that isnt listed in the nixos options
<ndowens08>
so mine isn't needed; darn thought i found something else to add to NIX
<LnL>
are you on 16.09, might have been added afterwards
<ndowens08>
nah i am in unstable
<ndowens08>
s/in/on
[0x4A6F] has quit [(Quit: [0x4A6F])]
<ndowens08>
just looked the repo trying to decide which dir it would belong in and then i found it
schoppenhauer has quit [(Quit: Adé)]
<LnL>
looks like the current module doesn't have a package option either, you could add it if you want
<ndowens08>
and overwrite it?
ibrahims has joined #nixos
schoppenhauer has joined #nixos
<LnL>
just update it, add the package option and use cfg.package instead of pkgs.vim
bfrog has joined #nixos
<NixOS_GitHub>
[nixpkgs] peti pushed 1 new commit to master: https://git.io/vDjlR
<NixOS_GitHub>
nixpkgs/master 7803a69 Peter Simons: smartmontools: update drivedb.h to latest SVN version 4391