Hydra | 386 Members | |
| 110 Servers |
| Sender | Message | Time |
|---|---|---|
| 19 Aug 2022 | ||
| I must have misunderstood your previous message then | 11:02:46 | |
In reply to @ulrikstrid:matrix.orghttps://github.com/nixOS/nix/pulls?q=is%3Apr+is%3Amerged+rewrite+c%2B%2B | 11:03:03 | |
| Also yes, it partially was perl | 11:03:23 | |
| oh actually https://github.com/NixOS/nix/milestone/6?closed=1 is a better overview. | 11:04:06 | |
| Perl has very little code that justifies being written in C++, I'm pretty sure it was picked for the website because it both had a popular web framework and the Nix contributors had experience | 13:48:52 | |
| * Hydra has very little code that justifies being written in C++, I'm pretty sure it was picked for the website because it both had a popular web framework and the Nix contributors had experience | 13:48:58 | |
| * Hydra has very little code that justifies being written in C++, I'm pretty sure Perl was picked for the website because it both had a popular web framework and the Nix contributors had experience | 13:49:02 | |
| 21 Aug 2022 | ||
| am I alone in thinking it is a surprising behaviour when executing hydra with foreman that evals being affected by the ambiant NIX_PATH is surprising? | 04:47:45 | |
the eval used import <nixpkgs> {}, and nixpkgs was an input of the build (a git checkout) | 04:48:44 | |
it ended-up trying to import <nixos-config> for its overlay, and failing on a further path import disallowed by restricted eval | 04:49:13 | |
btw, I'm using a foreman-started hydra as a sort of better UX around nix-build for a complex project... it's hard to get a good bird's eye view of what is going on | 04:51:44 | |
In reply to @k900:0upti.meNot to be too annoying but bump | 07:12:01 | |
| 09:25:58 | ||
| ignore the fact that it's not a valid way to evaluate Nixpkgs | 21:18:26 | |
| * ignoring the fact that it's not a valid way to evaluate Nixpkgs | 21:18:32 | |
| am I missing something obvious with hydra jobset inputs? | 21:19:04 | |
| 21:19:13 | |
*
This is "working" as expected | 21:19:25 | |
Download image.png | 21:19:50 | |
assuming the inputs work like NIX_PATH does, I would expect this to eval at least up to failing within the eval of Nixpkgs stuff | 21:20:36 | |
| 21:20:42 | |
relatedly, I'm trying to simulate another more involved issue where similarly import <some-url-to-tarball> within hydra will try to import the raw tarball, but when provided as NIX_PATH to a norma Nix eval it unpacks it before making it available | 21:24:33 | |
* relatedly, I'm trying to simulate another more involved issue where similarly import <some-url-to-tarball> within hydra will try to import the raw tarball, but when provided as NIX_PATH to a normal Nix eval it unpacks it before making it available | 21:44:00 | |
*
This is "working" as expected in a Nix eval | 22:05:19 | |
| 22 Aug 2022 | ||
| 07:48:01 | ||
| I'd like to be able to use declarative jobsets in hydra to spawn jobsets for all unmerged branches(essentially the parsed result of Alternatively, is there a way to get branch merged information when doing | 21:37:06 | |
| * I'd like to be able to use declarative jobsets in hydra to spawn jobsets for all unmerged branches(essentially the parsed result of Alternatively, is there a way to get branch merged information when doing | 21:41:30 | |
| * I'd like to be able to use declarative jobsets in hydra to spawn jobsets for all unmerged branches(essentially the parsed result of Alternatively, is there a way to get branch merged information when doing Edit: seems like I might be looking for | 22:01:53 | |
| * I'd like to be able to use declarative jobsets in hydra to spawn jobsets for all unmerged branches(essentially the parsed result of Alternatively, is there a way to get branch merged information when doing Edit: not complete but, seems like I might be looking for | 22:16:55 | |
| 23 Aug 2022 | ||
| 00:48:41 | ||