| 20 Jul 2021 |
 David Arnold (blaggacao) | Do tests pass on https://github.com/gytis-ivaskevicius/flake-utils-plus/pull/84? | 21:15:33 |
| David Arnold (blaggacao) | (Let's go!) | 21:15:39 |
| David Arnold (blaggacao) | Gytis Ivaskevicius: as a shared module and overlay api between fup and digga, could we settle on composition at the user level via fup.nixosModules.<name> & fup.overlays.<name>? | 21:19:04 |
| David Arnold (blaggacao) | fufexan: would you mind also looking over digga's develop branch with the critical eye that inspired https://github.com/gytis-ivaskevicius/flake-utils-plus/pull/84 ? | 21:37:18 |
| 21 Jul 2021 |
| David Arnold (blaggacao) | After refactoring digga to comply with the recent upstream fup refactoring done by fufexan, users might start to see things like:
trace: warning: You are accessing a deprecated item of the digga lib.
Please update timely, it will be remove soon.
trace: warning: Deprecated Attribute Set: lib.importers.
Please use import* functions instead:
lib.importers.overlays -> lib.importOverlays
lib.importers.modules -> lib.importModules
lib.importers.hosts -> lib.importHosts
lib.importers.rakeLeaves -> lib.rakeLeaves
lib.importers.flattenTree -> lib.flattenTree
| 00:04:31 |
| David Arnold (blaggacao) | * After refactoring digga to comply with the recent upstream fup refactoring done by fufexan, users might start to see things like:
trace: warning: You are accessing a deprecated item of the digga lib.
Please update timely, it will be remove soon.
trace: warning: Deprecated Attribute Set: lib.importers.
Please use import* functions instead:
lib.importers.overlays -> lib.importOverlays
lib.importers.modules -> lib.importModules
lib.importers.hosts -> lib.importHosts
lib.importers.rakeLeaves -> lib.rakeLeaves
lib.importers.flattenTree -> lib.flattenTree
This stuff will probably be removed once develop is merged back into master.
The full deprecation list is here: https://github.com/divnix/digga/blob/develop/deprecated.nix
| 00:06:31 |
| David Arnold (blaggacao) | this devos commit exemplifies the changes. Thx @fufexan for isnpiring the downstream cleanup with his refactoring. | 00:14:15 |
| David Arnold (blaggacao) | * this devos commit exemplifies the changes. Thx @fufexan for inspiring the downstream cleanup with his refactoring. | 00:14:25 |
| David Arnold (blaggacao) | * This devos commit exemplifies the changes. Thx @fufexan for inspiring the downstream cleanup with his refactoring. | 00:14:30 |
 @timdeh:matrix.org | I got pinged somewhere here today, but I can't find it 😅 | 01:26:14 |
| @timdeh:matrix.org | nvm, I think it was my username in the nix.conf file segment 😆 | 01:29:16 |
| David Arnold (blaggacao) | develop went away — now we are on main (still no release, though) | 03:17:42 |
| David Arnold (blaggacao) | * develop went away on digga — now we are on main (still no release, though) | 03:22:04 |
| David Arnold (blaggacao) | * The develop branch went away on digga — now we are on main (still no release, though) | 03:22:15 |
| David Arnold (blaggacao) | * The develop branch went away on digga — now we are on main (still no new release, though) | 03:22:48 |
 @gytis-ivaskevicius:matrix.org | In reply to @blaggacao:matrix.org
Gytis Ivaskevicius: as a shared module and overlay api between fup and digga, could we settle on composition at the user level via fup.nixosModules.<name> & fup.overlays.<name>? Im thinking maybe we could push something like what is required for Nix options to nixpkgs | 08:21:39 |
| David Arnold (blaggacao) | Oh I forgot: https://digga.divnix.com/ | 16:15:17 |
| David Arnold (blaggacao) | * Oh I forgot: https://digga.divnix.com/ (initial digga docs are live) | 16:15:29 |
 fufexan | In reply to @blaggacao:matrix.org
fufexan: would you mind also looking over digga's develop branch with the critical eye that inspired https://github.com/gytis-ivaskevicius/flake-utils-plus/pull/84 ? yes, sure thing | 20:15:40 |
| fufexan | how does the lib.exportOverlays work in practice? I'm trying to understand what would get output as "nixpkgs/winestreamproxy" coming from this overlay
{ inputs }:
final: prev: {
osu-stable = prev.callPackage ./osu-stable {
wine = final.wine-tkg;
winetricks = prev.winetricks.override { wine = final.wine-tkg; };
inherit (final) winestreamproxy;
};
winestreamproxy = prev.callPackage ./winestreamproxy { wine = final.wine-tkg; };
wine-tkg = prev.callPackage ./wine-tkg {
wine =
if prev.system == "x86_64-linux"
then final.wineWowPackages.unstable
else final.wineUnstable;
inherit (inputs) tkgPatches oglfPatches;
};
}
| 20:16:55 |
| fufexan | I get that it outputs something like "nixpkgs/winestreamproxy" = final: prev: { }; as written in the doc, but I don't really get what's in the set | 20:17:36 |
| David Arnold (blaggacao) | So it tries to give clearly identifiable names "channel/winestreamproxy" to every overlay. Each overlay was crafted for a specific channel, which still is a local reference. But having a clue what the author regarded as "nixpkgs" at the time of crafting an overlay might get the consumer an idea with wich channel of his own this overlay probably might have a better chance of working. | 20:36:35 |
| David Arnold (blaggacao) | * So it tries to give clearly identifiable names "channel/winestreamproxy" to every overlay. Each overlay was crafted for a specific channel, which still is a local reference. But having a clue what the author regarded as "nixpkgs" at the time of crafting an overlay might get the consumer an idea with which channel of his own this overlay probably might have a better chance of working. | 20:36:56 |
| David Arnold (blaggacao) | in the end, the cuonsumer has to try to know for sure, but if there is something like nixos-19.03/some, then the user might probably be able to guess that it's not going to work with master | 20:37:39 |
| David Arnold (blaggacao) | * in the end, the consumer has to try to know for sure, but if there is something like nixos-19.03/some, then the user might probably be able to guess that it's not going to work with master | 20:37:44 |
| David Arnold (blaggacao) | In reply to @fufexan:matrix.org
I get that it outputs something like "nixpkgs/winestreamproxy" = final: prev: { }; as written in the doc, but I don't really get what's in the set So this would be completed as: final: prev: { winestreamproxy = ...} | 20:38:26 |
| David Arnold (blaggacao) | In reply to @fufexan:matrix.org
I get that it outputs something like "nixpkgs/winestreamproxy" = final: prev: { }; as written in the doc, but I don't really get what's in the set * So this would be completed as: "nixpkgs/winestreamproxy" = final: prev: { winestreamproxy = ...} | 20:38:43 |
| David Arnold (blaggacao) | As an author, you probably group some overlays together. But that doesn't mean as a consumer you want to be forced to consume the entire group. That's the reason why each overlay attribute is put in a spearate overlay. | 20:39:38 |
| David Arnold (blaggacao) | * As an author, you probably group some overlays together. But that doesn't mean as a consumer you want to be forced to consume the entire group. That's the reason why each overlay attribute is put in a separate overlay. | 20:39:44 |
| David Arnold (blaggacao) | * As an author, you probably group some overlays together. But that doesn't mean as a consumer you want to be forced to consume the entire group. That's the reason why each overlay attribute is put in a separate overlay. A good example would be if you manage a single overlays.nix file with all your overlays defined according to your personal needs. A consumer might want to pick one out of these. | 20:40:46 |
