| 17 Mar 2024 |
 infinisil | Yeah I'm not a fan of that idea | 14:11:10 |
 K900 | And also because it's much bigger and much more expensive to build than the other cases I can see | 14:10:51 |
 Robert Hensing (roberth) | I don't get it. It would work well for git-based channels | 14:11:41 |
| Robert Hensing (roberth) | Currently we don't have any programs.sqlite for those users | 14:11:52 |
| K900 | Also, here's a hot take: the default behavior should be to put that behind an HTTP endpoint | 14:11:46 |
| infinisil | Why does it need to be git-based? | 14:12:10 |
| K900 | And make people opt in if you want it offline | 14:11:59 |
| K900 | In reply to @roberthensing:matrix.org
Currently we don't have any programs.sqlite for those users We have nix-index | 14:12:06 |
| K900 | Though it's also not great because it wastes time scraping Hydra every time | 14:12:27 |
| K900 | But it is correct in being incremental and out of lifecycle | 14:12:39 |
| Robert Hensing (roberth) | Because then Nixpkgs functionality like command-not-found doesn't break when you use a branch | 14:12:47 |
| Robert Hensing (roberth) | nix-index is not a viable replacement for everyone to run by default | 14:13:26 |
| K900 | We can have it not break in other ways | 14:13:27 |
| K900 | That don't involve tracking an sqlite blob in nixpkgs forever | 14:13:38 |
| Robert Hensing (roberth) | We should deliver programs.sqlite to branch-based users, which is the majority by now | 14:13:52 |
| infinisil | Orthogonally to this discussion, programs.sqlite should be replaced with nix-index | 14:14:05 |
| Robert Hensing (roberth) | Most growth has been through flakes apparently | 14:14:07 |
| infinisil | * Orthogonally to this discussion, programs.sqlite should be replaced with a nix-index database | 14:14:12 |
| Robert Hensing (roberth) | Sure, but it needs to be distributed somehow, regardless of the format | 14:14:25 |
| K900 | In reply to @k900:0upti.me
That don't involve tracking an sqlite blob in nixpkgs forever No, we shouldn't | 14:14:24 |
| K900 | We should deliver working command-not-found to those users | 14:14:33 |
| K900 | But that is very much not the same thing | 14:14:40 |
| Robert Hensing (roberth) | And it shouldn't be based on the exact Nixpkgs commit, because then other branches don't work | 14:14:45 |
| infinisil | How about making Flakes support channels | 14:14:56 |
| infinisil | We don't need to invent a new solution for something that already worked in stable Nix | 14:15:06 |
| Robert Hensing (roberth) | No, that doesn't solve the problem that branches don't have the data | 14:15:22 |
| infinisil | Even npins supports channels, it's just a tarball fetch | 14:15:23 |
| K900 | You can actually use a channel as a flake input already | 14:15:22 |
| Robert Hensing (roberth) | It's not a flakes problem | 14:15:35 |
| K900 | It's silly but you can | 14:15:27 |
