| 6 Dec 2025 |
 Robert Hensing (roberth) | Because nix is starting to be affected by this new git feature (https://github.com/NixOS/nix/issues/14716) | 13:05:01 |
 Sergei Zimmerman (xokdvium) | In reply to @roberthensing:matrix.org
We could use some help with https://github.com/libgit2/libgit2/pull/7117 Would be nice to figure out why upstream is dead silent. John did mention that on libgit2’s slack people were saying that the maintainer is MIA. There want much activity since their discussion about sha256 iiuc | 13:07:03 |
| Sergei Zimmerman (xokdvium) | In reply to @roberthensing:matrix.org
We could use some help with https://github.com/libgit2/libgit2/pull/7117 * Would be nice to figure out why upstream is dead silent. John did mention that on libgit2’s slack people were saying that the maintainer is MIA. There wasn’t much activity since their discussion about sha256 iiuc | 13:07:17 |
|  strangeoceans joined the room. | 23:03:08 |
| 7 Dec 2025 |
| Robert Hensing (roberth) | Here's an opportunity to solve some community concerns about flakes https://discourse.nixos.org/t/outlining-the-differences-between-flakes-and-nix-configs/72996?u=roberth | 15:38:03 |
| Robert Hensing (roberth) | Not quite the solve-bugs-first approach, but valid nonetheless | 15:38:18 |
| Robert Hensing (roberth) | adopt this rust code https://fosstodon.org/@ole/115679363422551728 | 17:12:28 |
| Robert Hensing (roberth) | I've opened an RFC to help the flakes project incorporate old community feedback. Would love to have your feedback on https://github.com/NixOS/rfcs/pull/193 | 23:07:01 |
| Robert Hensing (roberth) | "TOML flakes" | 23:07:16 |
| 8 Dec 2025 |
| Robert Hensing (roberth) | Now also https://github.com/NixOS/rfcs/pull/194 for the inputs.entrypoint part | 10:13:43 |
| Robert Hensing (roberth) | I've split it into two | 10:14:10 |
 dramforever | cc vcunat i can't reproduce the aarch64-darwin build failure on nix-community darwin01, /nix/store/1jgxwz43j7rjbfywcldlqxss0iwqgsby-nix-functional-tests-2.31.2+1.drv succeeds there, reproducibly | 11:52:06 |
| dramforever | * re https://github.com/NixOS/nixpkgs/pull/468208#issuecomment-3626314109 vcunat i can't reproduce the aarch64-darwin build failure on nix-community darwin01, /nix/store/1jgxwz43j7rjbfywcldlqxss0iwqgsby-nix-functional-tests-2.31.2+1.drv succeeds there, reproducibly | 11:52:23 |
| dramforever | my only idea is can you possibly try making it not have a plus sign in the name? | 11:52:43 |
 vcunat | I wonder. What I tested is unsandboxed nix config. | 11:53:13 |
| vcunat | (I think Hydra's builders are also unsandboxed?) | 11:53:25 |
| dramforever | oh | 11:54:07 |
| dramforever | like as in --no-sandbox? | 11:54:36 |
| Sergei Zimmerman (xokdvium) | In reply to @dramforever:matrix.org
my only idea is can you possibly try making it not have a plus sign in the name? Mhm that got fixed on maintenance branch | 11:55:25 |
| vcunat | As in no sandbox mentioned in nix.conf. | 11:55:29 |
| Sergei Zimmerman (xokdvium) | In reply to @xokdvium:matrix.org
Mhm that got fixed on maintenance branch It’s the borked soname with appendPatches | 11:56:21 |
| dramforever | oh my god | 11:56:49 |
| dramforever | but why does it work on darwin01 | 11:57:40 |
| Sergei Zimmerman (xokdvium) | Darwin linker doesn’t like plus in soname | 11:57:45 |
| dramforever | is it like, macos version dependent | 11:58:18 |
| dramforever | do you have a link or remember any keywords in the plus sign fix or something | 12:02:13 |
| dramforever | i can't find it | 12:02:26 |
| vcunat | I'd think it's best to just avoid + in SONAMEs... | 12:02:57 |
| vcunat | (seems like pointless risks/trouble) | 12:03:26 |
| dramforever | found it https://github.com/NixOS/nix/pull/14432 | 12:07:18 |
