nixos/modules: remove with lib

[kuflierl]

This PR tries to remove generic with expressions that make it hard to discern the source of a function (and for LSP's to notice mistakes).

see also: #208242

All changes were made by hand with assistance from regex and lsp.

Some removed with expressions include:

• with lib
• with types

Things done

• Built on platform:
  • x86_64-linux
  • aarch64-linux
  • x86_64-darwin
  • aarch64-darwin
• Tested, as applicable:
  • NixOS tests in nixos/tests.
  • Package tests at passthru.tests.
  • Tests in lib/tests or pkgs/test for functions and "core" functionality.
• Ran nixpkgs-review on this PR. See nixpkgs-review usage.
• Tested basic functionality of all binary files, usually in ./result/bin/.
• Nixpkgs Release Notes
  • Package update: when the change is major or breaking.
• NixOS Release Notes
  • Module addition: when adding a new NixOS module.
  • Module update: when the change is significant.
• Fits CONTRIBUTING.md, pkgs/README.md, maintainers/README.md and other READMEs.

---

Add a 👍 reaction to pull requests you find important.

[nixos-discourse]

This pull request has been mentioned on NixOS Discourse. There might be relevant details there:

https://discourse.nixos.org/t/prs-ready-for-review/3032/5933

[osbm]

I couldnt review maubot and synapse modules because of the huge diff but rest seems extremely well done.

I just have 1 thing to point out but this seems very much ready.

Thank you so much for this this makes the codes very readable in my opinion

[nixos-discourse]

This pull request has been mentioned on NixOS Discourse. There might be relevant details there:

https://discourse.nixos.org/t/prs-ready-for-review/3032/6008

[mmlb]

Did you consider inhert (lib) ... instead? I gave it a quick try on uni-sync.nix and the diff is much nicer:

[11:04:59]-[~/c/g/n/nixpkgs]-[manny@zennix]
jj diff
diff --git a/nixos/modules/hardware/uni-sync.nix b/nixos/modules/hardware/uni-sync.nix
index 2fc9377256..0512b4d6bc 100644
--- a/nixos/modules/hardware/uni-sync.nix
+++ b/nixos/modules/hardware/uni-sync.nix
@@ -4,9 +4,17 @@
   pkgs,
   ...
 }:
-with lib;
 let
   cfg = config.hardware.uni-sync;
+  inherit (lib)
+    literalExpression
+    maintainers
+    mkEnableOption
+    mkIf
+    mkOption
+    mkPackageOption
+    types
+    ;
 in
 {
   meta.maintainers = with maintainers; [ yunfachi ];
[11:05:04]-[~/c/g/n/nixpkgs]-[manny@zennix]
nil diagnostics nixos/modules/hardware/uni-sync.nix
[--:--:--]-[~/c/g/n/nixpkgs]-[manny@zennix]

Its not super obvious but in the page @AndersonTorres links to in #208242 this is shown as the alternative in the tip.

[kuflierl]

Did you consider inhert (lib) ... instead? I gave it a quick try on uni-sync.nix and the diff is much nicer:

[11:04:59]-[~/c/g/n/nixpkgs]-[manny@zennix]
jj diff
diff --git a/nixos/modules/hardware/uni-sync.nix b/nixos/modules/hardware/uni-sync.nix
index 2fc9377256..0512b4d6bc 100644
--- a/nixos/modules/hardware/uni-sync.nix
+++ b/nixos/modules/hardware/uni-sync.nix
@@ -4,9 +4,17 @@
   pkgs,
   ...
 }:
-with lib;
 let
   cfg = config.hardware.uni-sync;
+  inherit (lib)
+    literalExpression
+    maintainers
+    mkEnableOption
+    mkIf
+    mkOption
+    mkPackageOption
+    types
+    ;
 in
 {
   meta.maintainers = with maintainers; [ yunfachi ];
[11:05:04]-[~/c/g/n/nixpkgs]-[manny@zennix]
nil diagnostics nixos/modules/hardware/uni-sync.nix
[--:--:--]-[~/c/g/n/nixpkgs]-[manny@zennix]

Its not super obvious but in the page @AndersonTorres links to in #208242 this is shown as the alternative in the tip.

Yes, that is an option, but i really don't like it due to the fact that i can not determine if it's a builtin or if its coming from lib. But i did do it for "lib.types" in some places

[mmlb]

Did you consider inhert (lib) ... instead? I gave it a quick try on uni-sync.nix and the diff is much nicer:

[11:04:59]-[~/c/g/n/nixpkgs]-[manny@zennix]
jj diff
diff --git a/nixos/modules/hardware/uni-sync.nix b/nixos/modules/hardware/uni-sync.nix
index 2fc9377256..0512b4d6bc 100644
--- a/nixos/modules/hardware/uni-sync.nix
+++ b/nixos/modules/hardware/uni-sync.nix
@@ -4,9 +4,17 @@
   pkgs,
   ...
 }:
-with lib;
 let
   cfg = config.hardware.uni-sync;
+  inherit (lib)
+    literalExpression
+    maintainers
+    mkEnableOption
+    mkIf
+    mkOption
+    mkPackageOption
+    types
+    ;
 in
 {
   meta.maintainers = with maintainers; [ yunfachi ];
[11:05:04]-[~/c/g/n/nixpkgs]-[manny@zennix]
nil diagnostics nixos/modules/hardware/uni-sync.nix
[--:--:--]-[~/c/g/n/nixpkgs]-[manny@zennix]

Its not super obvious but in the page @AndersonTorres links to in #208242 this is shown as the alternative in the tip.

Yes, that is an option, but i really don't like it due to the fact that i can not determine if it's a builtin or if its coming from lib. But i did do it for "lib.types" in some places

I think lib is preferred over builtins in nixpkgs, which should probably be enforced if true, but that seems like a separate issue. I don't really like the full path on each use myself, but imo minimizing the diff would be the nicest option.

[AndersonTorres]

In theory lib is more future- and past-safe. Sometimes a functionality is promoted to builtin or demoted to nix code. The lib in nixpkgs includes guards for cases when the interpreter does not have this or that builtin.

Further, treating everything as lib avoids the bikeshedding :)

[wolfgangwalther]

Merge conflicted after #443046. Not closing, because the other PR was only about meta = with lib, so there is a lot more here.

[kuflierl]

I currently don't have much time and can not really do any major changes to my change-set.
I would honestly love it if this could be merged or dealt with in another way.