| 25 Oct 2023 |
 cpcloud | It's close. Waiting on the status of the osx build. All linux tests are passing | 11:01:07 |
 K900 | mypy takes longer to build than a kernel | 11:01:32 |
| K900 | What a world | 11:01:33 |
| cpcloud | Oh yeah. Do we really need to test that? | 11:01:51 |
| K900 | I think so | 11:01:57 |
| K900 | But that's not even about CI | 11:02:00 |
| cpcloud | I am very tempted to set it prefer wheels | 11:02:01 |
| K900 | I'm just building a mypy locally | 11:02:04 |
| cpcloud | * I am very tempted to set it to prefer wheels | 11:02:08 |
| K900 | For nixops | 11:02:11 |
| cpcloud | I think it's generating some 600k line monstrosity | 11:02:53 |
| K900 | @adisbladis btw would appreciate your review on https://github.com/NixOS/nixops/pull/1569 too | 11:03:07 |
| cpcloud | * I think it's generating some 600k line C monstrosity | 11:03:07 |
 adisbladis | Osx builds are taking way longer than usual | 11:03:38 |
| cpcloud | I enabled --keep-going | 11:05:18 |
| cpcloud | Otherwise there's some netcdf failure early on | 11:05:37 |
| cpcloud | I suspect it's failing because I'm using a patched nixos-unstable-small | 11:06:12 |
| K900 | Maybe run it on Winter's box? | 11:06:18 |
| K900 | Might be faster | 11:06:20 |
| cpcloud | I don't know what that means | 11:06:27 |
| K900 | https://github.com/winterqt/darwin-build-box/ | 11:06:41 |
| K900 | I have access if you want | 11:06:45 |
| cpcloud | Eh, not too sure about shared nature of it :) | 11:10:54 |
| K900 | It's about 99% idle | 11:13:27 |
| K900 | Right now | 11:13:29 |
| K900 | Also I don't mean running it there in CI | 11:13:38 |
| K900 | I mean running it there for now | 11:13:42 |
| K900 | Just to iterate faster | 11:13:49 |
| cpcloud | Re sphinx5, I don't think this can be done in an override | 11:20:40 |
| cpcloud | The recursion is happening before overrides are evaluated, while converting the lockfile to nix expressions | 11:21:14 |
