| 30 Jul 2025 |
 fzakaria | makes sense; i would probably calculate my own stable key | 15:23:42 |
| fzakaria | * makes sense; i would probably calculate my own stable key off the requisite fields | 15:23:53 |
| fzakaria | i have rarely ever used the serialization format itself as inputs to things | 15:24:37 |
 John Ericson | is this preserves thing widely implemented? | 15:24:44 |
| fzakaria | but maybe its because i never have canonicalization :P | 15:24:50 |
| John Ericson | I would like to make it easier to implement "nixlets" | 15:25:00 |
 emily | no, as I said ecosystem is the main drawback. it's being used in the OCapN effort to standardize CapTP between Agoric/Spritely/I think Cap'n Proto too, and it's used in https://syndicate-lang.org, but it's not the kind of thing you'd expect to find a library in any random language for. I'm not sure what the "second-best" thing is. maybe "whichever of the several CBOR canonical forms is what CTAP uses", but IIRC that's specifically not the RFC one, so that's a nightmare.
I don't know about BCS but at a glance it seems like it does a reasonable job, assuming everyone agrees on what the format actually is. it's not self-describing though, so extensibility will indeed suffer
| 15:28:27 |
| John Ericson | yeah I don't think we can use BCS | 15:29:18 |
| John Ericson | it would not be good to use a map for an extensible struct | 15:30:44 |
| John Ericson | because one then gets unnecessary enum tags on the values, which hinders extensibility | 15:31:00 |
| John Ericson | (you can freely choose your own not-yet-used key, but you need to also choose a not-yet-used enum tag, which is harder) | 15:31:23 |
 magic_rb | NDBF - Nix Derivation Binary Format | 17:00:42 |
 flokli | https://xkcd.com/927/ | 22:35:11 |
| 31 Jul 2025 |
|  AshKaashh joined the room. | 22:51:16 |
| fzakaria | I worked with Yannik on a small test -- https://github.com/NixOS/nix/pull/13673
Small but maybe interesting. | 23:28:29 |
| fzakaria | interesting change is making "ref" available to the flake attributes. I only include if impure. | 23:28:58 |
| 1 Aug 2025 |
|  @mel05saq:inphima.de left the room. | 12:13:53 |
| 2 Aug 2025 |
|  @sorimaxuki:matrix.org joined the room. | 01:18:20 |
| @sorimaxuki:matrix.org left the room. | 01:23:14 |
|  sinan changed their profile picture. | 06:58:39 |
| sinan | 06:58:46 |
| 3 Aug 2025 |
| sinan changed their profile picture. | 15:07:47 |
| fzakaria | Sergei Zimmerman (xokdvium): i'll work on that meson thing tonight; its been bugging me when i work on functional tests and edit code | 23:07:02 |
| fzakaria | feels like a "simple" thing that should be possible:
Recompile the code when i change it before I run the tests. | 23:07:18 |
| fzakaria | works locally.. dunno about that PR | 23:07:25 |
| fzakaria | * works locally.. dunno about that PR failure. | 23:07:36 |
 Sergei Zimmerman (xokdvium) | No clue. Maybe worth asking upstream Meson about how this could be made to work or if the current behavior is how it's supposed to work with override_find_program | 23:08:50 |
| Sergei Zimmerman (xokdvium) | Seems severely undocumented and possibly broken | 23:09:15 |
| 4 Aug 2025 |
| fzakaria | the meson developers | 01:00:28 |
| fzakaria | don't seem to like how it's all setup lol | 01:00:32 |
