| 9 Jun 2024 |
 @irenes:matrix.org | the switch to the Meson build system is the progress in that direction so far | 09:53:00 |
 ma27 | last week I didn't have time for Lix hacking, so I may be a little out of date, but are you aware of error: 'xonsh-unwrapped' was incorporated into xonsh code; call xonsh directly instead. when opening the devshell?
worked around it by doing what the error told me, but don't really know if that even makes sense (and this was purely to get my env back up), so asking here first :) | 13:24:49 |
 julia | that's a nixpkgs update, not lix | 13:35:27 |
| julia | * that's a nixpkgs update, not lix, see pkgs/top-level/aliases.nix | 13:35:38 |
 delroth | ahhh sorry I bumped the nixpkgs version in the flake and clearly didn't test the devshell | 13:39:14 |
| delroth | that's probably why | 13:39:17 |
 puck | hrm, can we have a list of non-trivial things to check on these kinds of bumps? or maybe add devshell to checks | 13:40:18 |
| delroth | building the devshell in CI sounds like a good idea | 13:40:47 |
| delroth | https://git.lix.systems/lix-project/lix/issues/383 | 13:42:36 |
| puck | (we like being able to get rid of these kinds of footguns ^-^) | 13:43:05 |
| delroth | tbf I also have no clue why xonsh is in the devshell by default | 13:43:15 |
| puck | not sure either; jade_ added it and i suspect it might be to deal with an artifact of how devshells work (Qyriad?) | 13:44:31 |
| delroth | To ssh://gerrit.lix.systems:2022/lix
! [remote rejected] HEAD -> refs/for/main (n/a (unpacker error)) | 13:44:57 |
| delroth | that's a new one | 13:44:59 |
| delroth | error: remote unpack failed: error Missing tree e1aff8ff0d78351e5685efc04641fc558b50c2df | 13:45:30 |
 raitobezarius | Transient? | 13:45:43 |
| delroth | it repro'd 3 times | 13:45:54 |
| ma27 | delroth --no-thin helped me. | 13:46:05 |
| raitobezarius | Uh-huh | 13:46:06 |
| delroth | but after rebasing on latest main it worked | 13:46:07 |
| delroth | https://gerrit.lix.systems/c/lix/+/1428 with the caveat that I tested that "nix develop" worked and that "xonsh" starts up in there, but nothing further | 13:46:29 |
| puck | checking error_log, Missing tree e1aff8ff0d78351e5685efc04641fc558b50c2df | 13:46:31 |
| raitobezarius | Maybe git fsck on the FS repo would be good as well or is that unrelated? | 13:47:01 |
| puck | now there's something slightly cursed: that tree exists | 13:47:23 |
| puck | wait did we get rid of sandbox refs | 13:48:13 |
| puck | though, let's see | 13:51:20 |
| puck | git fsck says it's fine, with exception of a few dangling commits | 13:52:36 |
| delroth | it's apparently a commonly reported gerrit bug | 13:54:10 |
| delroth | so I wouldn't worry about it | 13:54:13 |
| puck | (and a larger pile of unreachable commits) | 13:54:14 |
