<fuzzypixelz>
trying to make a PR that fixes the Plasma themes not installing correctly, it changes the activation script of plasma in the nixos/modules/services/x11/desktop-managers/plasma5.nix
<fuzzypixelz>
what should I name the commit
<hazel[m]>
`jdk12` won't build for me -- how do I pin down the source of what package is even pulling it?
<Kyndig>
Does anyone have any ideas about what would make my personal user's "nix-env -iA ..." command suddenly start trying to read /etc/nixos/configuration.nix?
dycan has joined #nixos
liuyb has quit [Ping timeout: 252 seconds]
<Kyndig>
my .nix-defexpr points to my per-user profile...
<Kyndig>
i don't see how the system config get's involved with nix-env
<Church->
Okay asking in the right channel this time. :p Can I give a relative path to builtins.readFile somehow? Trying to read a file kept in a subdir of the dir my nix expression lives in.
<Church->
And `builtins.readFile ./secrets/foo.zone` errors out
fuzzypixelz has quit [Ping timeout: 268 seconds]
<samueldr>
what does "errors out" mean?
turlando has quit [Quit: Leaving]
fuzzypixelz has joined #nixos
fuzzypixelz has left #nixos [#nixos]
<dotdotok>
samueldr: as in it breaks when you try to use it
<dotdotok>
/build it
<samueldr>
dotdotok: are you Church-?
<samueldr>
what I wanted to know is the exact error message you get
<samueldr>
because otherwise it's a vague issue
<dotdotok>
ah read the message sequence wrong, sorry samueldr
<dotdotok>
read it as church -> you -> church and not church -> church -> you
<samueldr>
no worries, mistakes can happen, and it happens I know that some users have multiple handles here
<samueldr>
(that's why I was asking if you were them, not as a combative response)
h0m1 has quit [Ping timeout: 258 seconds]
bitmapper has quit [Quit: Connection closed for inactivity]
<matthewcroughan>
samueldr: thank God you're here :D
<samueldr>
oh, the age old issue of .profile and friend sourcing being an absolute mess to understand in all unix-likes
<samueldr>
probably not a home-manager issue per se
<samueldr>
I personally, a long time ago, took time to read all the litterature about that, and made all things source each others in ways I believe were easier to get
<samueldr>
but uh... that would probably break some use cases? I never had any issues though
<samueldr>
and with NixOS I even disabled most of it, and my system setup relies only on the equivalent files in /etc/
<samueldr>
I guess in your case, figure out what gdm and sddm *do* read for "profile"
<samueldr>
it looks like that upstream sddm would use ~/.xprofile
<samueldr>
sure, I only realized after the fact that you wanted a login dialog, and not the smallest way to autologin
<samueldr>
from my own bias, where I want autologin to happen
<matthewcroughan>
samueldr: I want to have to enter a password, of course.
<samueldr>
from cold boot, you need a passphrase to get to the graphical session... and anyway I have to trust the session locker I use, so first thing that happens when my session start is to launch the locker
<samueldr>
in the off chance I walk away early during boot
<matthewcroughan>
? You need a passphrase?
spudly- has joined #nixos
<samueldr>
to unlock the full disk encryption
<matthewcroughan>
I'm happy as long as a passphrase is required.
<matthewcroughan>
Yeah I do have FDE..
<matthewcroughan>
I guess
<matthewcroughan>
But what happens when you exit sway?
<matthewcroughan>
what happens if sway crashes?
<samueldr>
I don't know because I'm using X11 still
<matthewcroughan>
LOL
<samueldr>
but that depends on the DM I guess
<matthewcroughan>
Okay, assume your thing crashse
<matthewcroughan>
it will give someone full access to your user, if you're just using a tty
<matthewcroughan>
that's why I don't want to use a tty
<samueldr>
here if my session crashes, SDDM restarts it, and if it crashes too often in a short while it knows to not autologin
<samueldr>
but since it's an autologin through a DM, that's not giving them a console
<samueldr>
and anyway, sway shouldn't crash ;)
<matthewcroughan>
Not an argument ;D
<matthewcroughan>
If the program you're running crashes, you return to a user shell.
<matthewcroughan>
then, anyone can run rm -rf *
spudly- is now known as spudly
<samueldr>
sure, that's why I'm not launching it from a user session on the tty
<matthewcroughan>
I quite like the journalctl diagnostics that gdm gives me anyway
<matthewcroughan>
samueldr: Do you have any idea why `xdg.portal.gtkUsePortal = true;` would be delaying the startup of `services.xserver.gdm.enable = true;` by like 30 seconds?
<matthewcroughan>
because it totally does
<matthewcroughan>
and it drives me mental :)
<samueldr>
nope
liuyb has joined #nixos
ahmedelgabri has joined #nixos
kaction has joined #nixos
ahmedelgabri has quit [Ping timeout: 258 seconds]
Dr8128 has quit [Ping timeout: 252 seconds]
mmmattyx has quit [Quit: Connection closed for inactivity]
<matthewcroughan>
loginctl shows me that the session is still active, and things spawned by sway are still active.
<matthewcroughan>
Even if I set `services.logind.killUserProcesses = true;` I still get this behavior
ahmedelgabri has quit [Ping timeout: 258 seconds]
Heirlung has joined #nixos
zeenk has joined #nixos
sveitser2 has quit [Quit: WeeChat 2.2]
codygman____ has joined #nixos
<codygman____>
I just updated my flake inputs and the new nix broke my system :/ I say this to warn others and to signal-boost the issue I filed https://github.com/NixOS/nix/issues/4705
<{^_^}>
nix#4705 (by codygman, 2 minutes ago, open): error creating statement 'insert or replace into realisations" broke nix installation on update
<matthewcroughan>
codygman____: how do you know this was the thing that broke it?
<matthewcroughan>
And what input were you trying to update?
<codygman____>
I installed programs earlier today with my older flake.lock and things were fine. Then I did 'nix flake update', got this flake.lock file https://github.com/NixOS/nix/issues/4705#issuecomment-817262927, then I did `sudo nix shell ".#nixosConfigurations.razer.config.system.build.toplevel" -c switch-to-configuration switch` and got the error above. Plus now none of my nix commands work.
<afontain_>
<matthewcroughan "How can I put stuff in /usr/shar"> With or without a FHS?
<afontain_>
davidak: for the record, the "regular" desktop file works, but it seems the invisible unv:// one doesn't
<afontain_>
if you want to try to install it and see if it works for you, please do — it really should work, so maybe it's just my setup being broken
<afontain_>
ah, I've got an idea of protocol handler: mailto:xyz@xyz.xyz
<afontain_>
(when running the command with an existing unvanquished window)
<davidak[m]>
thanks. will test later. have to sleep now
<afontain_>
sure
<afontain_>
for the record, I installed with `nix-env -f . -Ai unvanquished`
philr has quit [Ping timeout: 265 seconds]
<afontain_>
this $ xdg-open unv://88.198.122.215:27960 returns error 4: gio: unv://88.198.122.215:27960: The specified location is not supported
<afontain_>
and error 4 means "The action failed."
<afontain_>
so this suggests it's not "A required tool could not be found.", "One of the files passed on the command line did not exist." or "Error in command line syntax."
<davidak[m]>
it would be interesting to see if it works on other distros
<oleks>
Hi, often when working with stack on NixOS, I run into stack telling me something like error: attribute 'ghc844' missing, at (string):1:61 Things start working again once I upgrade to some more recent LTS, but why does this happen in the first place?
<afontain_>
this looks like a nix error
astylian has quit [Ping timeout: 252 seconds]
<oleks>
Indeed, it is something to do with nix and stack not playing well together
<afontain_>
"upgrade to some more recent LTS" of what?
<oleks>
It's an option in my stack.yaml
<oleks>
So LTS of Stack
<afontain_>
I assume ghc8.4.4 didn't exist yet in that lts
<oleks>
That's what I understand as well. But now I'm trying to start a new project
<oleks>
And stack init fails
<oleks>
With that sort of error
<oleks>
So.. no stack.yaml
<afontain_>
och
<afontain_>
you can probably open a bug report on nixpkgs, or on stack
VideoGameEnjoyer has quit [Quit: Konversation terminated!]
ahmedelgabri has joined #nixos
BenjiProd has joined #nixos
<sterni>
oleks: We don't have GHC 8.8.4 in nixpkgs anymore, but the stack nix integration expects it to be present
<virkony>
I'm trying to migrate system configuration to flakes and I stumbled on this git submodules not being properly supported (NixOS/nix#4423). And attempt to use flake input in module for system conf results in error for lastModified...
<virkony>
it is standard environment to build Nix packages
<sterni>
well stdenv doesn't really contain build tools it contains stuff you need to bootstrap nixpkgs that is a different focus
<sterni>
you don't need a debugger in there for example
<virkony>
there is also nixpkgs.metaBuildEnv that have gcc included, though without gdb
quinn has joined #nixos
<tad-lispy[m]>
Hello! With Home Manager, how can I switch to previous generation?
<tad-lispy[m]>
* Hello! With Home Manager, how can I switch to a previous generation?
<virkony>
tad-lispy[m]: most likely you want to go to #home-manager with such question. In short, it have sub-command generations and you can find link to activation script
<afontain_>
tad-lispy: IRC folks can't see you 👍️ reaction
<tad-lispy[m]>
So I switched back, but it didn't help.
<tad-lispy[m]>
The problem is that the `super` key stopped working. I'm using Gnome on Wayland.
ahmedelgabri has quit [Ping timeout: 260 seconds]
<{^_^}>
[nixpkgs] @sternenseemann merged pull request #113696 → ocamlPackages.janeStreet{,_0_9_0}: join the ocamlPackages fix point, allowing overriding to work as expected → https://github.com/NixOS/nixpkgs/pull/113696
<tad-lispy[m]>
I upgraded both NixOS (`sudo nixos-rebuild switch`) and home manager. Then I noticed the key doesn't work. I've rebooted to previous NixOS generation. Still didn't work. I activated previous Home Manager generation and rebooted. Still doesn't work. Any ideas?
Siyo has quit [Quit: Bye]
Siyo has joined #nixos
zebrag has joined #nixos
ahmedelgabri has joined #nixos
Dr8128 has joined #nixos
philr has joined #nixos
ahmedelgabri has quit [Client Quit]
lb5tr has quit [Quit: WeeChat 1.4]
<tad-lispy[m]>
I've found the problem and the solution and it has nothing to do with NixOS or Home Manager or Wayland or Gnome :D
<kolaente_>
gchristensen: I have a habit of cleaning up garbage regularily after updates, but it does not seem to clean up much (~1GB last time)
<gchristensen>
is this nixos?
<kolaente_>
nix-store --gc --print-roots returns 652 paths which, while a lot, does not seem that excessive to me
<kolaente_>
yes, nixos
<__monty__>
kolaente_: And you delete generations regularly?
<gchristensen>
b/c you'll need to periodically need to delete old system generations: ls /nix/var/nix/profiles/system | wc -l
<kolaente_>
Ah well looks like I don't __monty__
<kolaente_>
I guess I just assumed nix-collect-garbage would do that for me
<gchristensen>
that is what --delete-older-than does
<gchristensen>
and you'll need to run it as root
<gchristensen>
each system generation is considered not garbage until you ask it to prune old generations :$
<__monty__>
You have explicit control over which generations to keep because that's how you do rollbacks.
<kolaente_>
ohhhhh and if I don't pass it an --delete-older-than it removes none of the old generations
<kolaente_>
I think that was my mistake
<__monty__>
I usually use nix-env to manage generations, both user and system.
<__monty__>
Keep at least one known-working generation around.
ahmedelgabri has joined #nixos
<__monty__>
nix-env --delete-generations +2 would keep the last three generations.
<kolaente_>
TIL
<__monty__>
For system generations you have to pass --profile /nix/var/nix/profiles/system
<kolaente_>
I just assumed if you run nix-collect-garbage without any options it would delete everyting currently not in use but it makes a lot of sense to need to specify removing old generations explicitly
<kolaente_>
Running nix-collect-garbage --delete-older-than 3d just removed 196GB of old stuff, thanks everyone for clarifying!
<__monty__>
Yes, old generations are in use, for rollbacks : )
<{^_^}>
#76615 (by JonathanReeve, 1 year ago, open): Can't install a new KDE theme
<fuzzypixelz100>
I'm getting file:///nix/store/rabw98zbqi7hvnayixmy57qw2fnmd1ph-kdeclarative-5.79.0-bin/lib/qt-5.15.2/qml/org/kde/kcm/GridDelegate.qml:0: ReferenceError: theme is not defined
<fuzzypixelz100>
I'm on plasma 5.21.3
<kolaente_>
__monty__ using the --profile option to remove system profiles gives me "error: unrecognised flag '--profile'"
<__monty__>
kolaente_: Try passing it before --delete-generations.
<gchristensen>
you don't have to pass a specific profile actually
<fuzzypixelz100>
then `kf.newstuff.core: Install job finished with error state 113 and description 113`
<__monty__>
With nix-env you do, afaik.
<gchristensen>
oh yeah
<gchristensen>
sorry
<__monty__>
But indeed with nix-collect-garbage running as root would do the trick iirc.
<kolaente_>
__monty__ ah nix-env ofc (I've been trying nix-collect-garbage ._.
<__monty__>
Reason I don't use --delete-older-than is sometimes my previous generation is quite old, so there's not really a fixed age I can use. And you can inspect your generations with nix-env --list-generations.
<evanjs>
tried adding cargoBuildHook and cargoInstallHook and that doesn't do anything, though I'm not sure if it would have an affect on a buildPython{Package,Application} closure to begin with
<urkk>
evanjs: with nix log I can specify a custom cache store to retrieve the log, but I would like to distribute all the closure logs in a file
emp has joined #nixos
<urkk>
so that all the output paths and logs are self contained
ahmedelgabri has quit [Ping timeout: 250 seconds]
<urkk>
maybe I can just copy the corresponding entries in /nix/var/log ?
MatrixBot25 has joined #nixos
domogled has quit [Quit: domogled]
chiefgoat has quit [Ping timeout: 240 seconds]
cfricke has joined #nixos
aaaaaaaaaaaareis has joined #nixos
<aaaaaaaaaaaareis>
I'm trying to add a vim plugin to nixpkgs, but when i run the update.py script (WITH the regquired args, as said in the readme), it prints plenty of too many requests errors, until it crashes about 4m later:
<aaaaaaaaaaaareis>
```
<aaaaaaaaaaaareis>
I'm trying to add a vim plugin to nixpkgs, but when i run the update.py script (WITH the regquired args, as said in the readme), it prints plenty of too many requests errors, until it crashes about 4m later:
aaaaaaaaaaaareis has quit [Client Quit]
aaaaaaaaaaaareis has joined #nixos
vidbina has joined #nixos
<aaaaaaaaaaaareis>
sorry for this ;-; I'm super new to IRC
<evanjs>
aaaaaaaaaaaareis: you need to set up GitHub auth IIRC. Does it point you to a page that tells you what to do?
<aveltras>
anyone using nixos-unstable with nixpkgs-wayland overlay ? it seems i can't rebuild successfully because of hash mismatch in rust packages, anyone has the same issue ?
<pie_>
nschoe: are you comparing it to the kicad .nix file or to how its being called from all-packages.nix?
<pie_>
they may be different
<nschoe>
pie_: I'mnot sure I understand.
Henson has joined #nixos
<sterni>
nschoe: srcs is not an argument, but an derivation attribute, so you need to use overrideAttrs
<sterni>
arguments are always function inputs so usually dependencies
<nschoe>
pie_: all-packages.nix provie a "kicad" (normal), "kicad-small" which is defined as "kicad.override {with3d = false;}", and a "kicad-unstable".
<pie_>
ok so this is whats in all-packages, so nothing misleading there; kicad-unstable = kicad.override { pname = "kicad-unstable"; stable = false; };
<nschoe>
sterni: ohhh right. Thanks. Then the comments in the kicad file are wrong?
<Kinnison>
Good afternoon all, I'm having some trouble with a nixos system I've been putting together using nix flakes.
<Kinnison>
It has been working well for a while, but for a few days now, if I update the nixpkgs flake in the lock, then I cannot build the system because home-manager can't build
simukis_ has joined #nixos
<evanjs>
Do you have some output that describes the error?
<Kinnison>
the error is:
<Kinnison>
error: cannot import '/nix/store...-nmd', since path '/nix/store/....-nmd.drv' is not valid (with different ...s)
<Kinnison>
the line in question failing is line 16 of home-manager's doc/default.nix which is nmd = import nmdSrc { inherit lib pkgs; };
<Kinnison>
Sorry I can't easily copy/paste since the machine in question is a VM
<Kinnison>
I should try and ssh into it so I can, one tic
<workerdrone>
How does one try and build a package from their checked out copy of nixpkgs? I am trying to do my first package, and when trying to call nix-build on it, it of course complains that it can't auto-call a function that has an argument without a default value (fair enough, it's written in nixpkg style, have read up that those have to be wrapped in callPackage one way or another), so I decided that I should try building an already existent
<workerdrone>
package this way, but I don't really see how one would do that.
aveltras has quit [Quit: Connection closed for inactivity]
<virkony>
workerdrone: I think you need to use something like pkgs.callPackage, where pkgs is something like (import <nixpkgs> {}) for example
jophish__ has quit [Ping timeout: 240 seconds]
sangoma has quit [Ping timeout: 240 seconds]
cfricke has joined #nixos
<Kinnison>
evanjs: I think I've tracked down to a single merge where before the merge to nixpkgs I don't get the error and afterwards I do
<Kinnison>
evanjs: 67fe7b4b81486e197a0d4e2ec48d9e5d1fd20cce works, 26e33dc1e92ae496cbc121be1dc4415dedf957a8 does not
<virkony>
if you are checking out nixpkgs and fine loading all packages from that repo, you probably can use nix-build . -A attribute-to-install
<Kinnison>
evanjs: afaict that's just adding some patches to curl
<Kinnison>
evanjs: It looks like nmd consumes curl as a nativeBuildInputs - I wonder if this is something going iffy with a security update vs. how flakes work?
lordcirth has quit [Ping timeout: 258 seconds]
<evanjs>
Kinnison: yeah, I haven't dealt with any buildInputs vs nativeBuildInputs issues in a while, but I have seen some things become more strict lately
mschwaig has joined #nixos
<evanjs>
like some rust packages that worked before where buildInputs and nativeBuildInputs were conflated now require the inputs are defined in the right places, etc.
<Kinnison>
Who knows how many rookie mistakes I've made so far :D
<magnetophon>
My laptop stopped changing power state: no upower events, wrong value in /sys/class/power_supply/AC/online Odly, booting into an older NixOS generation does not fix it! It always says it's charging, even when unplugged, yet it always discharges, even when plugged. It might have started when I did tlp recalibrate, but I'm not sure.
<Guest16900>
Anyone online who can help me out configuring Nixos-shell (the QEMU VM helper)? I'm having trouble with setting up permissions for my user.
ahmedelgabri has joined #nixos
<evanjs>
Kinnison: seeing what happens when I try and build locally
<Kinnison>
evanjs: thank you, I'm trying to do that on a spare box I have here (debian, but with nix installed) since that might be easier for me to strace etc.
<Kinnison>
if I end up needing to
<workerdrone>
Is there any tradition as to how packages in nixpkgs/pkgs/top-level/all-packages.nix should be ordered?
jonringer has joined #nixos
ahmed_elgabri has quit [Ping timeout: 260 seconds]
supersandro2000 has quit [Remote host closed the connection]
supersandro2000 has joined #nixos
jonringer has quit [Remote host closed the connection]
jonringer has joined #nixos
<Kinnison>
evanjs: When I try to do the build on my desktop, it doesn't seem to fail in the same way, which suggests I've messed something up on my test VM perhaps?
<Kinnison>
evanjs: On my desktop, `nix build git+file:///path/to/nixos-hosts\#nixosConfigurations.\"test\".config.system.build.toplevel --dry-run` works
<Kinnison>
evanjs: the equivalent on the test VM gives the error I started the conversation with
* Kinnison
ponders clearing his entire store on the desktop
<Kinnison>
evanjs: Interestingly if I take --dry-run away I think it's able to build it, so I wonder if there's something trying to import from the store rather than being pure?
<Kinnison>
(I say "I think" because it's still building on the vm)
fendor has quit [Ping timeout: 252 seconds]
<evanjs>
I mean I did see a few things going on in the Makefile outside of `nix build`
<evanjs>
sops and gpg updates, etc.
<Kinnison>
Oh goodness, the makefile in that repo is a mess
<Kinnison>
Right, so without --dry-run I can build it with an updated flake lock
<Kinnison>
That's, erm, interesting
<Kinnison>
So it couldn't dry-run and gave the odd error about not being able to import something directly from the store, but without dry-run it built
miscalculated has joined #nixos
<evanjs>
Oh well sometimes dry-run might not be able to run for various reasons
<evanjs>
I have made a few issues for myself with e.g. IFD (import from derivations), etc.
<Kinnison>
so I guess if I hit this kind of thing in the future I should do a non-dry build
* Kinnison
learns something new
<Kinnison>
evanjs: thank you for helping me thinking through this
<evanjs>
Rubber ducking is great--I mean I never do that :P
philr has quit [Ping timeout: 240 seconds]
<evanjs>
`nixos-rebuild test` might also help if you want to see what happens before actually "switching"
<Kritnich>
Does anyone have some pointers to booting a Pi 4 with NixOS via TFTP? I have everything set up and it's able to pull the boot files but gets stuck in U-Boot.
<Kinnison>
evanjs: is that more sensible than dry-build ?
VideoGameEnjoyer has quit [Remote host closed the connection]
Mirrexagon has joined #nixos
<Kinnison>
evanjs: oooh dry-activate looks good
* Kinnison
wonders if bringing in the nixos-* tools from nixpkgs-unstable would stop the `warning: 'nix flake info' is a deprecated alias for 'nix flake metadata'`
<Kinnison>
and if so, what the nix equivalent of dpkg -S bin/nixos-rebuild would be
<Kinnison>
without the switch it gives me the help, and with --rollback switch I get the error as above
VideoGameEnjoyer has joined #nixos
<clever>
nixos-rebuild --rollback switch
<clever>
Kinnison: you can also rollback a little more manually, ls -lh /nix/var/nix/profiles/system
<clever>
what does this output?
<Kinnison>
er /nix/var/nix/profiles/system -> system-4-link
<hodapp>
part of the reason I became quite adept at un-mangling disks and partitions is because of how many times as a kid I'd have to fix something before my parents found out
<hodapp>
but NixOS has worked well enough for me that I have no similar experience
<clever>
Kinnison: we are relying on the .override here, to make nixos-rebuild use a flake capable nix
<clever>
and the overlay, then changes what pkgs.nixos-rebuild refers to
<Kyndig>
Does anyone know what config or env var settings might make nix-env try to read /etc/nixos/configuration.nix? This started happening after an upgrade from nixos 20.03 to 20.09:
<Kyndig>
ahh; wow, I set that up in the configuration.nix like a year ago when trying to get global overlays to work (never did; documentation was conflicting and had issues)
<Kyndig>
can't believe i didn't notice until now
<Kyndig>
something in my shell must've been working around it
<Kyndig>
I'll remove that now from config; thanks clever
<wunderbrick>
What's the best way to sort a set alphabetically by keys?
<pie_>
wunderbrick: attrsets are already sorted?
<pie_>
well, depends on what you mean by alphabetically i guess
zupo has joined #nixos
<wunderbrick>
Oh yeah, duh. Haha. I'm just looking at a huge attr set in a file and needing to alphabetize. I can just import the file and spit it back out.
<wunderbrick>
Well sets aren't really sorted like I need them to be if I just import the set and write it back to a file. It seems the order that way is capital letters, then special characters like _, then lowercase letters when I guess I need special characters first, then everything else, case-insensitive. What I'm actually trying to do is sort maintainer-list.nix per a reviewer's request and I'm not sure the
<wunderbrick>
best way to go about it, or really why it should be done for my particular pull request. lol.
est31 has quit [Remote host closed the connection]
<rmcgibbo[m]>
Does anyone have any examples of a configuration.nix that imports NixOS modules from out-of-tree (e.g. from github)?
<rmcgibbo[m]>
I tried to put a fetchFromGithub into my `imports = [ ... ]` section, and I appear to be crashing and burning with infinite recursion in ways that don't really make sense to me.
quinn has joined #nixos
<simpson>
I suspect that you can make it work by putting a let/in at the top of your module, right after you've closed over `pkgs`.
<{^_^}>
[nixpkgs] @mweinelt pushed to revert-118719-home-assistant-tests « Revert "nixos/home-assistant: use override before overridePythonAttrs" »: https://git.io/JOLfv
<samueldr>
matthewcroughan: why have you pinged me?
<pinpox>
hexa-: Ok, let me try to add that check
<matthewcroughan>
because you were in chat recently and have knowledge and I'm asking you a question :D
<matthewcroughan>
If you want me to not ping you, I will not, sorry.
<Andoriyu>
okay, theme imported. now how would one request a variant of said theme in home-manager
codygman__ has quit [Remote host closed the connection]
codygman__ has joined #nixos
MinceR has joined #nixos
BlessJah has quit [Ping timeout: 276 seconds]
codygman__ has quit [Remote host closed the connection]
<elvishjerricco>
I'm trying to do something like `buildEnv { name="foo"; paths=[bash coreutils]; pathsToLink=["/bin"]; }` and I'm getting an output with no symlinks in $out/bin
<elvishjerricco>
I feel like I'm going nuts. Isn't that exactly what buildEnv is for?
Qwerky has quit [Remote host closed the connection]
philr has joined #nixos
zeta_0 has quit [Quit: rcirc on GNU Emacs 27.1]
<SumnerEvans[m]>
When I run `nix-shell -p nixpkgs-review --run "nixpkgs-review wip"`, it tells me that "No diff detected, stopping review...". What is necessary for that to work?
<{^_^}>
[rfc39-record] @grahamc pushed commit from rfc39 to main « Automated team sync results. »: https://git.io/JOLIR
<supersandro2000>
Raito_Bezarius: or the PATH does not contain it
ahmed_elgabri has joined #nixos
<rmcgibbo[m]>
Sumner Evans: I think that's because you've already made a commit. The `wip` is looking for uncommitted stuff in your checkout. If you've made a commit you can use `nixpkgs-review rev {sha}`.
Henson has joined #nixos
ahmed_elgabri has quit [Ping timeout: 260 seconds]
<Henson>
hi everyone. I'm trying to make the DAViCal software work with NixOS. One part of the installation is initializing the database with a script. Is there a proper way to do this initialization only once? I've seen some software that runs a service do this as part of the pre-start stage, which it checks for the existence of a file, if it's not present it does the initialization and then writes...
<Henson>
the file, but Davical doesn't run as a service, it's a PHP script that runs with Apache. Is there a way I could do this initialization without a service?
<Henson>
or maybe just make the user do the initialization themselves as part of the installation?
<Henson>
the few other web service modules I've looked at don't do the initialization themselves (only looked at wordpress, mediawiki, and dokuwiki).