| 3 Mar 2026 |
|  lineargraph changed their display name from nea to lineargraph. | 13:43:13 |
|  tekila joined the room. | 15:13:39 |
|  Nicole/Drezil 🏳️⚧️ (she/her) changed their profile picture. | 16:16:33 |
 FireFly | nix-shell -p as implemented is a mess, but I think the UX is pretty comfy... what I've thought before is that something like it (and tbh a lot of the CLI) should be implementable "downstream" from the nix impl itself (so for this case, a script provided as its own thing within nixpkgs since it's tightly coupled to it) | 18:03:20 |
| FireFly | there might be arguments against that, it's been a while since I thought about the nix CLI.. but like, the way -p is implemented with synthesizing an inline script seems like it could be done with a script wrapping -E expr, plus I guess spawning a shell with modified $PATH to include the output in it | 18:05:24 |
 goldstein | you totally could build a downstream CLI with whatever plumbing we have today (and people did)
e.g. nix-direnv uses manual mkShell + nix print-dev-env from nix3 | 18:21:09 |
| goldstein | (no idea why they went with that instead of nix-store --print-env | 18:21:25 |
| goldstein | * (no idea why they went with that instead of nix-store --print-env) | 18:21:28 |
 Icewind | Is there a way to check if a derivation is in the caching without having to download it?
Context is a CI system that runs nix build .... as the sole step. It works fine, but I would like to skip the download for the case where the build is already cached. | 19:41:47 |
| Icewind | Is there a way to check if a derivation is in the cache without having to download it?
Context is a CI system that runs nix build .... as the sole step. It works fine, but I would like to skip the download for the case where the build is already cached. | 19:41:57 |
|  Cyrill set a profile picture. | 19:42:09 |
| goldstein | as in “in local store” or “available from a substituter”? | 19:45:33 |
| Icewind | substituter | 19:45:44 |
 raitobezarius | you can do curl $cache/$hashPart.narinfo | 19:47:50 |
| raitobezarius | if it returns a non-404, then it has a strong chance to exist | 19:47:58 |
| raitobezarius | if you want to ensure it, you can follow the redirect in the narinfo and verify it's also a non-404 | 19:48:06 |
| Icewind | That would work yes. | 19:50:09 |
| Icewind | Not as nice as having it in the nix-cli, but it will do | 19:50:59 |
| Icewind | Thanks | 19:52:00 |
| raitobezarius | an issue to help co design the right Nix CLI verb would be welcome | 21:24:53 |
 neobrain | oh hey raitobezarius , thanks for fixing that keep-env-derivations issue I reported :) | 22:35:32 |
| raitobezarius | hopefully it makes it into the 2.95 release :) | 22:35:50 |
| 4 Mar 2026 |
 iFreilicht | Maybe this was discussed already, but is it already known at what time the talks will start on LixCon's first day? I'd like to book my trains sooner rather than later :) | 17:04:30 |
| raitobezarius | 10am ish is my plan atm | 17:07:07 |
| 5 Mar 2026 |
| iFreilicht | Okay that makes sense, thank you! | 00:06:38 |
| iFreilicht | Venue isn't yet finalized, right? | 00:07:36 |
|  bngs joined the room. | 01:18:21 |
 delroth | it's organized at ENS, so I suspect similar arrangement to nixcon paris a few years ago? | 02:38:00 |
| delroth | and the address is on https://pretix.dgnum.eu/dgnum/lixcon2026/ if that's the info you need | 02:38:15 |
| iFreilicht | In reply to @delroth:delroth.net
and the address is on https://pretix.dgnum.eu/dgnum/lixcon2026/ if that's the info you need That's indeed what I needed, no idea how I missed this 😅 | 07:26:21 |
