In reply to @ericadelagnier:multi.coop
When using flakes, I was able to run my flake (yeah !) but I now have a question regarding the dependency environment to run my app with gunicorn since I can't seems to find the right syntax to build it (the README give an default.nix as example and I'm don't know enough about flakes to reproduce it)

I think this will work with nix build .

flake.nix:

{
  description = "Python application packaged using poetry2nix";

  inputs.nixpkgs.url = "github:NixOS/nixpkgs/nixos-unstable";
  inputs.poetry2nix.url = "github:nix-community/poetry2nix";
  inputs.flake-utils.url = "github:numtide/flake-utils";

  outputs =
    {
      self,
      nixpkgs,
      poetry2nix,
      flake-utils,
    }:
    flake-utils.lib.eachDefaultSystem (
      system:
      let
        pkgs = nixpkgs.legacyPackages.${system};
        # create a custom "mkPoetryApplication" API function that under the hood uses
        # the packages and versions (python3, poetry etc.) from our pinned nixpkgs above:
        inherit (poetry2nix.lib.mkPoetry2Nix { inherit pkgs; }) mkPoetryApplication;
      in
      {
        packages = {
          app = mkPoetryApplication {
            packageName = "app";
            projectDir = ./.;
          };

          default = self.packages.${system}.app;
        };
      }
    );
}

In reply to @ericadelagnier:multi.coop
the build works ! thank you :) I can now run the app with ./result/bin/hello, but I'm not sure how to call it with gunicorn ? the README suggest "./result/bin/gunicorn web:app" but I'm not sure how to make gunicorn available in the result

I would suggest to add gunicorn in flake.nix buildInputs, and run it from python something like subprocess.Popen(["gunicorn", "web:app""])

Part of flake.nix with buildInputs:

packages = {
  app = mkPoetryApplication {
    packageName = "app";
    projectDir = ./.;
  };

  buildInputs = with pkgs; [
    python312Packages.gunicorn
  ];

  default = self.packages.${system}.app;
};

In reply to @ericadelagnier:multi.coop
the build works ! thank you :) I can now run the app with ./result/bin/hello, but I'm not sure how to call it with gunicorn ? the README suggest "./result/bin/gunicorn web:app" but I'm not sure how to make gunicorn available in the result

I would suggest to add gunicorn in flake.nix buildInputs, and run it from python something like subprocess.Popen(["gunicorn", "web:app"])

Part of flake.nix with buildInputs:

packages = {
  app = mkPoetryApplication {
    packageName = "app";
    projectDir = ./.;
  };

  buildInputs = with pkgs; [
    python312Packages.gunicorn
  ];

  default = self.packages.${system}.app;
};

The result is produced by using the update operator //. This means nested values of previous overlays are not merged recursively. In other words, previously defined attributes are replaced, ignoring the previous value, unless referenced by the overlay; for example final: prev: { foo = final.foo + 1; }.

The result is produced by using the update operator //. This means nested values of previous overlays are not merged recursively. In other words, previously defined attributes are replaced, ignoring the previous value, unless referenced by the overlay; for example final: prev: { foo = final.foo + 1; }.