From: Nessdoor <tlopez@keio.jp>

This allows for image.nix to be used inside hermetic builds where
builtins.currentSystem is not available as an attribute.
---
Prefer passing `hostPlatform` as a modular option through
`nixpkgs.hostPlatform`, rather than using the now legacy `system`
argument.

 images/nixos/image.nix | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/images/nixos/image.nix b/images/nixos/image.nix
index 227039b..8e23f79 100644
--- a/images/nixos/image.nix
+++ b/images/nixos/image.nix
@@ -1,10 +1,16 @@
-{ pkgs ? import <nixpkgs> {} }:
+{ pkgs ? import <nixpkgs> { }
+, hostPlatform ? { system = builtins.currentSystem; }
+}:
 
 let
   makeDiskImage = import "${pkgs.path}/nixos/lib/make-disk-image.nix";
   evalConfig = import "${pkgs.path}/nixos/lib/eval-config.nix";
   config = (evalConfig {
-    modules = [ (import ./qemu-system-configuration.nix) ];
+    system = null; # Pass system parameters modularly
+    modules = [
+      (import ./qemu-system-configuration.nix)
+      ({ ... }: { nixpkgs.hostPlatform = hostPlatform; })
+    ];
   }).config;
 in
   makeDiskImage {
-- 
2.40.1

builds.sr.ht/patches: FAILED in 4m35s

[images/nixos: pass optional system attribute to evalConfig][0] v2 from [nessdoor][1]

[0]: https://lists.sr.ht/~sircmpwn/sr.ht-dev/patches/44014
[1]: tlopez@keio.jp

✗ #1047938 FAILED  builds.sr.ht/patches/archlinux.yml https://builds.sr.ht/~sircmpwn/job/1047938
✓ #1047937 SUCCESS builds.sr.ht/patches/alpine.yml    https://builds.sr.ht/~sircmpwn/job/1047937
✓ #1047939 SUCCESS builds.sr.ht/patches/debian.yml    https://builds.sr.ht/~sircmpwn/job/1047939

Looks good, thank you

> From: Nessdoor <tlopez@keio.jp>

^ Looks like a header found its way into the commit message, but this
can be fixed at merge time

>> From: Nessdoor <tlopez@keio.jp>
>>
> ^ Looks like a header found its way into the commit message, but this
> can be fixed at merge time

I am very sorry, my eye is not yet trained to spot errors in mailed
patches. I don't know how it ended up in there. If it is a problem and
could result in a delayed merging process, tell me and I will submit a
v3 patch.

On Mon Sep 25, 2023 at 9:57 PM CEST, Tomas Antonio Lopez wrote:
>
> >> From: Nessdoor <tlopez@keio.jp>
> >>
> > ^ Looks like a header found its way into the commit message, but this
> > can be fixed at merge time
>
> I am very sorry, my eye is not yet trained to spot errors in mailed
> patches. I don't know how it ended up in there. If it is a problem and
> could result in a delayed merging process, tell me and I will submit a
> v3 patch.

This isn't an issue, and it's not an error. It was added because your
outgoing email ident does not match the commit author (seems like it's
just the case of the "N"). It will apply normally when I merge it.

Thanks!

To git@git.sr.ht:~sircmpwn/builds.sr.ht
   3cb88b0..0d9ff47  master -> master