| 11 Jul 2023 |
 nbp | adisbladis: I had not thought of using the final as the source for the input, but S.O.S. is kind of layed out this way. With the same disadvantage that the update operator is becoming overwhelmingly present. I think this is part of the solution, except that as mkDerivation is still present. Thus you still need overrideAttrs to modify its argument.
To which the solution is simply to let people override whatever they want before applying mkDerivation at the very last moment.
| 09:38:13 |
| nbp | The idea of S.O.S. is that derivations are available only through final / self, and prev / super provides the recipes. | 09:38:54 |
| nbp | You want to override anything, use prev. You want to use a program, then use final.
I warned people that I would be happy to break their code if they do not follow the rule. | 09:40:15 |
 @piegames:matrix.org | Speaking of mkPackage, a lot of the problems that package modules attempt to solve are actually mkDerivation problems, and I'd like to see them solved first separately | 09:56:35 |
| @piegames:matrix.org | Passing random shit as attribute set down through a chain of function calls should be an anti pattern. builtins.removeAttrs should not exist | 09:57:32 |
 K900 (deprecated) | removeAttrs is useful in other contexts | 10:00:19 |
| K900 (deprecated) | But I agree with the core of the argument | 10:00:29 |
| @piegames:matrix.org | In reply to @k900:conduit.0upti.me
removeAttrs is useful in other contexts Agreed | 10:05:34 |
 profpatsch | roberthensing: I don’t think it’s any question | 10:32:28 |
| profpatsch | nixos as-is is way to slow already | 10:32:35 |
| profpatsch | and that’s only a couple hundred modules | 10:32:45 |
| @piegames:matrix.org | Wait, would be the goal to have one shared global module system for packages? o.O | 10:38:44 |
| @piegames:matrix.org | I thought the idea was to have local module system instances for individual packages or groups thereof | 10:39:01 |
| profpatsch | that sounds … even worse? | 11:12:29 |
| @piegames:matrix.org | I don't know | 11:16:56 |
| @piegames:matrix.org | IMO we should rewrite all builders so that there is exactly only one override function (or maybe two for now). Because I'm pretty sure this should be possible | 11:18:56 |
 toonn | Has any profiling been done? I'd be interested to see which parts of Nixpkgs a typical system build spends most time in. | 11:20:30 |
 raitobezarius | In reply to @profpatsch:augsburg.one
roberthensing: I don’t think it’s any question I think you are forgetting you are discussing this with the person who works actively on NixOS modules and has ideas to make it fast… | 11:31:33 |
| profpatsch | raitobezarius: It’s fundamentally impossible to make it fast as far as I understand, but I’d love if it were otherwise | 12:27:34 |
| profpatsch | If you can get it to work without having to strictly evaluate all options first, then you might be able to | 12:28:15 |
| profpatsch | but then you have something that’s not the nixos module system anymore, because the semantics is different | 12:28:36 |
| nbp | It would be possible to make NixOS module use less memory and potentially be faster … by implementing it in C++.
But one must understand that the module system, while not evolving frequently is still evolving. And from my point of view, of someone who never managed to get anything large accepted in Nix source code, this is a not an option. | 12:30:54 |
 infinisil | It could be implemented as a Nix plugin | 12:31:50 |
| nbp | Not mentioning the fact that evolution would be restricted to evolution of the Nix binary. | 12:31:54 |
| nbp | Can Nix plugin extend the grammar, to experiment new syntax / primitive? | 12:34:01 |
| @piegames:matrix.org | In reply to @infinisil:matrix.org
It could be implemented as a Nix plugin Are these a thing already? What can they do? | 12:42:09 |
| @piegames:matrix.org | In reply to @nbp:mozilla.org
Can Nix plugin extend the grammar, to experiment new syntax / primitive? Simply being able to efficiently implement external functions exposed in Nix would probably be sufficient | 12:42:47 |
| * @piegames:matrix.org dreams of a WASM interface to Nix, to be able to implement complex functions efficiently in other languages, without being tied to the Nix code base | 12:43:36 |
| nbp | the reason I am asking is for S.O.S. implementation. | 12:43:38 |
| nbp | * the reason I am asking is for S.O.S. implementation draft. | 12:43:59 |
