00:44 <remiliascarlet> SiFuh: Regarding your MPV problem, you might want to set `hwdex=no` in your `mpv.conf` file. This solved the problem for me on my Core 2 Duo laptop.

03:06 <ukky> Compiling pkgutils with '-static' removed allows installing packages into any writable sysroot directory by non-root user, via fakeroot. This is much safer for testing or benchmarking pkgutils modifications, and does not affects host root FS.

03:26 <remiliascarlet> Finally managed to upgrade my CRUX installation's SSD from 500 GiB to 1 TiB. Not that I need it, but feels better to not have wasted money for nothing.

05:42 <SiFuh> Thanks remiliascarlet I will try it

05:43 <SiFuh> remiliascarlet: There is a hwdec=no in my config

06:50 <ppetrov^> hmm... apache here breaks and it seems to be libxml2 related

10:35 <sajcho> ukky: I've been running core/pkgutils and core/prt-get with disable static for almost a year now and no problem yet.

10:35 <SiFuh> jaeger: whitequark is free right?

10:51 <cruxbot> [opt.git/3.7]: apapche: fix build with libxml2 2.12

11:24 <cruxbot> [opt.git/3.7]: squid: fix build with libxml2 2.12

13:05 <ukky> sajcho: good to know, I will use non-static build for host updates too then

15:16 <jaeger> I assume so

15:30 <SiFuh> Wonder if a static build of a system is actually smaller than a dynamic.

15:31 <SiFuh> A fully complete system where compilation deps are no removed

15:31 <SiFuh> are now removed*

15:38 <dlcusa> No comments yet: https://www.theregister.com/2023/11/29/rhel_10_dropping_x11/

15:41 <SiFuh> Seems a bit extreme to drop the main xserver for wayland

15:44 <remiliascarlet> So as expected, RHEL dropped the display server for Gayland.

15:44 <remiliascarlet> s/Gayland/Wayland

15:46 <dlcusa> If X12 isn't supportable, this seems inevitable. CRUX certainly doesn't have the resources to make that happen.

17:31 <farkuhar> Is it lack of resources that prevents us from fulfilling zorz's request (a tool like cpan2crux, but in the python ecosystem instead of perl)? Maybe if python devs could settle on one canonical way to package software, and not reinvent the wheel every year with a flashy new build system.

17:40 <ppetrov^> cpan2crux is a life saver

17:40 <ppetrov^> a similar tool for python would be great

19:01 <joe9> is the maintainer of this repo ( https://crux.nu/portdb/index.php?a=repo&q=baguette ) here?

19:03 <farkuhar> joe9: that would be groovy3shoes, so yes.

19:04 <joe9> thanks.

19:04 <joe9> groovy3shoes: would you be able to update plan9port to a more recent version, please?

19:04 <joe9> thanks.

19:06 <farkuhar> ppetrov^: I wonder if zorz tried asking ChatGPT "write me a CRUX Pkgfile for python3-selenium". Would be interesting to see what the LLM comes up with.

19:10 <ppetrov^> farkuhar, I have asked ChatGTP to write me SlackBuild scripts... it uses old format and does not resolve dependencies

19:10 <ppetrov^> *GPT

20:00 <dlcusa> Over 16 comments now re: RH dropping X11 at https://forums.theregister.com/forum/all/2023/11/29/rhel_10_dropping_x11/ all worth reading.

20:19 <joe9> First systemd and then this. Why are they screwing up a working system?

20:26 <joe9> does crux put the core files from a segmentation fault somewhere/

20:26 <joe9> ?

20:26 <joe9> I do not see a core file in the current directory.

20:27 <ppetrov^> some comments are hilarious

20:27 <jaeger> crux doesn't do anything with them by default, you might need to increase the core size with ulimit -c

20:42 <joe9> I notice that crux does not install gdb by default? Is that correct?

21:46 <jaeger> Indeed but it's available in opt

22:57 <cruxbot> [contrib.git/3.7]: inkscape: updated to 1.3.2

23:06 <sajcho> Maybe someone will be interested. Firefox can be built using python 3.12. x86_64, arm64 => OK - please don't stonewall me

23:08 <sajcho> ukky: https://github.com/sajcho/core-x86_64/blob/main/pkgutils/pkgutils-5.40.10-by-saux.patch

23:21 <groovy3shoes> joe9, done! let me know if you have any issues with it

23:58 <farkuhar> sajcho: thanks for the report on firefox in a CRUX 3.8 environment. There aren't many of us testing that toolchain yet, but it's good to know that the errors reported previously by beerman have been fixed by now.