18 Oct 2025 |
raboof | things like https://cache.nixos.org/5yzw0vhkyszf2d179m0qfkgxmp5wjjx4.narinfo have no deriver but are in the cache anyway... do those ever get fetched? wouldn't they always be present locally because they're created as part of instantiation? | 15:26:14 |
dramforever | you don't have to eval to download stuff from cache.nixos.org, there's nix copy | 15:27:54 |
dramforever | uh, with --substitute-on-destination | 15:28:16 |
dramforever | or you can nix copy --from https://cache.nixos.org a path | 15:28:31 |
dramforever | but the point is no, it won't always be present locally | 15:28:55 |
raboof | right, you can, but I wondered if/when that happens in practice | 15:30:58 |
dramforever | --substitute-on-destination definitely happens in practice | 15:32:11 |
fzakaria | anyone want to pair/hack tonight? | 20:11:25 |
fzakaria | Question I was confused by Jeremy Fleischman (jfly) :
When calculating store-paths, Nix does "Hashing modulo fixed-output derivations" -- this makes sense since you want the store-path to be stable for multiple drvs for the FOD.
His question: Can we replace all the inputDrvs with their store-path and also achieve the same thing? (replace via the same massaging that we do for FODs)
| 20:54:21 |
fzakaria |  Download image.png | 21:04:07 |
fzakaria | The part we are confused about:
B.drv may have many different A.drv ;
That's OK because eventually A.drv is replaced with A.outPath... BUT
C depends on B.drv ... wouldn't that path have changed since the contents of the drv can have N A.drv paths | 21:04:58 |
Jeremy Fleischman (jfly) | My question is stronger than "can we?". I'm saying "mustn't we?" | 21:42:26 |
19 Oct 2025 |
lovesegfault | I'd appreciate some feedback on https://github.com/NixOS/nix/pull/14298, in particular regarding this:
One open question is that, with this fix, we accept copies of paths with any signature, even if that signature is not trusted. Perhaps we ought to only accept trusted signatures?
| 01:59:05 |
| cedricgc joined the room. | 03:38:04 |