| 12 Nov 2025 |
φ | is there any way to be able to use nix as a normal user in a non-systemd distro? | 09:22:58 |
toonn | I think Nix still allows a single-user install, and without a daemon there's no need for systemd, IIRC. (Lix's installer does not support a non-daemon install (yet?) IIRC) | 09:25:55 |
φ | so it should work fine if i install cppnix single-user from my user? is that correct? | 09:26:44 |
toonn | If I remember correctly. Single-user installs were disabled for Darwin but I believe Linux still allows them. | 09:27:49 |
φ | they dont work with selinux enabled afaik , but i dont have that so should be fine | 09:28:37 |
φ | do u know of any plans for lix to be able to do this? | 09:28:52 |
toonn | It was asked fairly recently. I think someone said it might come but it didn't seem all that certain. | 09:29:27 |
φ | oki tysm <3 | 09:30:25 |
Zoe Z | in a CPP conditional, is there a way to check lix < 2.93 ? | 16:28:23 |
raitobezarius | let me think | 17:21:01 |
Zoe Z | right now I'm using #if __has_include (<lix/libutil/async.hh>) | 17:23:06 |
raitobezarius | i think we have a version define? | 17:23:15 |
raitobezarius | Zoe Z PACKAGE_VERSION? | 17:25:28 |
just1602 | If there's such thing I have an old CL I could bring up because I didn't find how to check the version and I didn't had time to dig | 17:25:31 |
raitobezarius | autogenerated in config.h | 17:25:39 |
raitobezarius | /*
* Autogenerated by the Meson build system.
* Do not edit, your changes will be lost.
*/
#pragma once
#define CAN_LINK_SYMLINK 1
#define ENABLE_S3 1
#define HAVE_BOEHMGC 1
#define HAVE_CLOSE_RANGE 1
#define HAVE_DTRACE 1
#define HAVE_F_GETPATH 0
#define HAVE_LCHOWN 1
#define HAVE_LIBCPUID 1
#define HAVE_LUTIMES 1
#define HAVE_PIPE2 1
#define HAVE_POSIX_FALLOCATE 1
#define HAVE_SECCOMP 1
#define HAVE_STATVFS 1
#define HAVE_STRSIGNAL 1
#define HAVE_STRUCT_DIRENT_D_TYPE 1
#define HAVE_SYSCONF 1
#define PACKAGE_NAME "lix"
#define PACKAGE_STRING "lix 2.94.0pre20251106-dev_f00d720"
#define PACKAGE_TARNAME "lix"
#define PACKAGE_VERSION "2.94.0pre20251106-dev_f00d720"
#define SYSTEM "x86_64-linux"
| 17:25:45 |
raitobezarius | does that help? | 17:25:49 |
raitobezarius | not exactly sure you can do preprocessor magic enough on such a string… | 17:26:58 |
Zoe Z | yeah, that was what I was struggling with | 17:27:18 |
raitobezarius | yeah i think preprocessor doesn't offer any way | 17:27:44 |
raitobezarius | we need to define this in meson directly i guess | 17:27:51 |
just1602 | https://gerrit.lix.systems/c/lix/+/3025/comment/fdf7cf24_c93473f0/
Yeah, that's what peanne were saying here | 17:28:31 |
raitobezarius | if you are already using async to detect Lix < 2.93, I think you can probably check for one of the usual async helper defined? | 17:28:34 |
raitobezarius | yep | 17:28:45 |
raitobezarius | we should definitely start doing that Zoe Z thanks for the ping | 17:28:57 |
| lavender.pet joined the room. | 19:38:19 |
| 13 Nov 2025 |
hexa | I'm thankful for how reliably lix tells me when it is waiting on a lock | 02:02:47 |
raitobezarius | is it actually reliable on that :D ? | 05:39:41 |
Yureka (she/her) | in staging not staging-next, no? | 07:47:37 |
Yureka (she/her) | at least I can see staging-next is on curl 8.16.0 and doesn't have any extra patches | 07:47:58 |