19 Oct 2024 |
V. 🏳️⚧️ | I get that but if e.g. we use clangStdenv, and someone goes out of their way to use gcc, surely they're responsible for the nasal demons | 12:41:40 |
V. 🏳️⚧️ | * I get that but if e.g. we use clangStdenv, and someone goes out of their way to override and use gcc, surely they're responsible for the nasal demons | 12:42:04 |
Linux Hackerman | https://gerrit.lix.systems/c/lix/+/2057 it looks like the buildbot failure is unrelated to my change, is there anything I can do to retrigger it? | 13:41:19 |
Linux Hackerman | also, would anyone care to review https://gerrit.lix.systems/c/lix/+/2069 ? :) | 13:41:41 |
piegames | So std::c32rtomb only does UTF-8 if I correctly set the locale | 14:09:46 |
piegames | where can I get off this unhinged ride? | 14:09:52 |
aloisw | just write the thing yourself | 14:10:41 |
piegames | yep :/ | 14:21:32 |
piegames | thanks, I hate it | 14:21:37 |
raitobezarius | In reply to @linus:schreibt.jetzt https://gerrit.lix.systems/c/lix/+/2057 it looks like the buildbot failure is unrelated to my change, is there anything I can do to retrigger it? i ran rebuild | 16:35:08 |
raitobezarius | i can bestow you the right later on | 16:35:11 |
Linux Hackerman | In reply to @raitobezarius:matrix.org i ran rebuild thanks! | 16:35:23 |
raitobezarius | jade_ do you think we could enroll lix-module on gerrit to obtain CI as well? | 16:35:34 |
jade_ | sure we could do that, but this problem would not have even slightly been fixed with CI | 17:50:27 |
jade_ | i ran the tests | 17:50:32 |
jade_ | the tests were simply inadequate | 17:50:36 |
jade_ | wait i did not even fuck up a backport ???? is it just broken on main??! | 17:52:09 |
jade_ | agh okay no it was a fucked up backport combined with my main being old | 17:54:13 |
9999years | https://github.com/NixOS/nix/pull/8817#issuecomment-2423714552 | 20:28:06 |
9999years | struggling to not say "have you tried lix" | 20:28:16 |
9999years |
I'm not sure that simple solution [pinning Nix 2.18] works for me. I've encountered NAR hash mismatches and behaviour divergences which mean that I'm now stuck on a ratchet of increasing my nix version to keep things working
| 20:28:55 |
K900 | Determinate Flakes | 20:29:08 |
9999years | https://github.com/NixOS/nix/pull/10200#issuecomment-2422953563 | 20:34:54 |
9999years | i couldn't resist for this one | 20:35:02 |
just1602 | It's a good thing. With a couple of friends we did a nix workshop in the unconference section of a conf, and we end up with si much questions about DetSys that in the end we point everyone toward lix and we explain why the patches and the roadmap is awesome and why the community is so much welcoming. | 20:42:24 |
just1602 | I still refrain myself from posting link to lix under YouTube video about nix, but I don't know how long I'll be able to. | 20:43:01 |
20 Oct 2024 |
KFears (tragedy arc) | In reply to @k900:0upti.me Determinate Flakes I actually noticed something very cursed between Flakes and DetSys nix-installer a few days ago | 08:38:59 |
KFears (tragedy arc) | Lockfile in flakes is pretty much C++ internals serialized, so you can never change anything that touches flakes even slightly in C++ code, but also the versioning is utterly broken and useless | 08:40:49 |
KFears (tragedy arc) | Receipt in nix-installer is... Literally serializing Rust internals into JSON | 08:41:17 |
KFears (tragedy arc) | They also have "versioning", but it's pretty much just checking if installer version matches the version in JSON, and otherwise you just die | 08:41:43 |