nixos/drupal: config extend

[OulipianSummer]

This MR extends the available config options for Drupal.

There is a new configSyncDir setting that allows users to define the location of the config sync dir.

There is also a new extraConfig setting that accepts raw PHP code that will be injected into the Drupal settings.php file. The settings that users write to both attributes will automatically be available to Drupal.

These changes help bring the Drupal service one step closer to being managed by the nix language and NixOS.

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.

[OulipianSummer]

@NixOS/nix-formatting I'm working on this PR and I can't seem to get the formatting and linting tests to pass.

I've run the following commands on my local, all within the root of nixpkgs, with the latest code from the master branch.

• nix fmt
• nix-shell --run treefmt
• nix-shell -p nixfmt-tree && treefmt

None of them have yielded any changes to the files I'm working on, which seems to indicate the formatting is correct. I've even rebased my commit a few times to make sure that wasn't the culprit.

When using nix-shell, it looks like the version of treefmt that comes with nixfmt-tree is v2.3.1
I am running this using nix version 2.28.4.

The failures (at least according to the job logs) don't appear to be related to this package.

Any clue why this job keeps failing?

[jfly]

@OulipianSummer this is not your fault. I believe these checks run not on the code in your branch, but instead on your branch if it were merged.

With the nixfmt 1.0 upgrade, we recently changed our formatting rules, and now we have a bunch of PRs that are green from checks run before the upgrade. For example, your most recent CI failure complained about the formatting of pkgs/by-name/gi/git-spice/package.nix, which was just fixed in #428350.

I believe setting up github's merge queue is our plan to get PRs merged to main without breaking main. Until then, I guess we suffer. Unless there's some way to "expire" all these known stale checks?

[OulipianSummer]

@jfly Thanks for that information. OK, I thought what I was looking at seemed unrelated, but it's good to know for sure. I've noticed that the logs have changed each time I push to this branch, so I assume this is sort of slowly getting resolved by other maintainers as we speak.

I guess my strategy right now will be to just wait for the formatting errors to get resolved and then re-open and rebase my branch when that happens. A merge train/queue does sound like the fix for the future, but as for the old stuff I don't know of a way to expire jobs. Unless you add a new merge job that runs nxmt on everything before it gets merged and then figure out what fails from that, but that might just make a bigger problem.

[drupol]

Thanks, amazing stuff !

[OulipianSummer]

Thanks again!