| 26 Apr 2022 |
|  Jane Jasperous set a profile picture. | 10:12:54 |
|  phaer changed their display name from Paul Haerle to phaer. | 10:29:13 |
| 30 Apr 2022 |
|  duponin joined the room. | 05:32:14 |
 David Arnold (blaggacao) | Zhaofeng Li: congrats to 0.3.0! Shall I rebase https://github.com/zhaofengli/colmena/pull/73
Just in te.yo continue my experients to make a 'native' colmenaConfiguration clade for divnix/std 🙂 | 16:20:20 |
 Zhaofeng Li | In reply to @blaggacao:matrix.org
Zhaofeng Li: congrats to 0.3.0! Shall I rebase https://github.com/zhaofengli/colmena/pull/73
Just in te.yo continue my experients to make a 'native' colmenaConfiguration clade for divnix/std 🙂 Yes, please do ♥️ | 16:55:45 |
| 1 May 2022 |
|  Chinchilla Washington changed their display name from ContinuousWave to Chinchilla Washington. | 06:51:33 |
| David Arnold (blaggacao) | Zhaofeng Li: host dns or static ips or ssh config alias might only be known after an initial (metal) deployment task has returned. I think it might make sense to get a map of these values from an auxiliary json file. Have you thought about this workflow / use case, already? | 17:47:21 |
| David Arnold (blaggacao) | I mean it's easy with nix. Just want to make sure I'm not actually re-inventing something. | 17:48:09 |
| David Arnold (blaggacao) | Rebased. | 18:39:34 |
| David Arnold (blaggacao) | I'm also thinking maybe it's in order for keeping colmena maintainable to refactor it to use two different libraries, one for classical nix and one for flakes. Switch case at the leaves is the source of all evil, imo. | 18:42:37 |
| David Arnold (blaggacao) | * I'm also thinking maybe it's in order for keeping `colmena` maintainable to refactor it to use two different libraries, one for classical nix and one for flakes. Switch case at the leaves is the source of all evil, imo.
I've decent trouble unwinding the double lane reasoning when going through the code.
Took me maybe already 2-3 days. | 18:43:42 |
| David Arnold (blaggacao) | In any way, ideally the compatibility layer wouldn't bleed too much into the code, if that makes sense. | 18:44:47 |
| David Arnold (blaggacao) | Yet another low level suggestion: are you interested in clap_derive? I find that so much more readable in most circumstances. And I think it also uses less LoCs | 19:10:43 |
| Zhaofeng Li | Yes, migrating to the new Derive API has been in the back burner for a while | 20:02:44 |
| Zhaofeng Li | That's already possible to do yourself with fromJSON. I'm a bit wary of blessing any specific format in Colmena itself, to keep it lean and "provisioner-agnostic" | 20:07:36 |
| David Arnold (blaggacao) | K, so I implemented a hidden flag --eval that can swap eval.nix for a different implementation. This way, I can query .#__std metadata to autodetect colmenaConfigurations and make them available to the colmena CLI. 🎉 Will be testing this in a bit. | 20:17:44 |
| David Arnold (blaggacao) | Totally agree! | 20:18:15 |
| Zhaofeng Li | I'm not really sure about this, as I don't want to guarantee any compatibility for the internal eval API. It limits refactorability and can lead to errors indecipherable to the average user. | 20:29:48 |
| David Arnold (blaggacao) | Hm, I see. Alternatively I could just patch colmena, but that would make me responsible for building and caching, which I generally try to avoid. | 20:42:02 |
