00:48 <cruxbot> [contrib.git/3.7]: python3-soupsieve: 2.4.1 -> 2.5

00:48 <cruxbot> [contrib.git/3.7]: firefox: 116.0.3 -> 117.0

00:55 <farkuhar> ukky: https://lists.crux.nu/pipermail/crux/2023-May/007311.html

00:57 <ukky> farkuhar: thanks. Removing python3-libxml2 fixed the issue.

01:00 <ukky> I don't know where to report, but 'linux-pam' has a bug in /etc/security/namespace.init: Line 'mode=$(printf "%o" $((0777 & ~mask)))' should be 'mode=$(printf "%o" $((0777 & ~$mask)))'

01:13 <farkuhar> ukky: namespace.init is not distributed through the CRUX port tree, it comes from upstream. If the file in question is interpreted by bash, then inside the arithmetic expansion $(( )) you're allowed to drop the sigil from shell variables, so "not a bug" is probably how upstream will respond.

01:13 <ukky> I need to RTFM on 'sh' variable substitution: $((0777 & ~$mask)) gives the same result as $((0777 & ~mask))

01:15 <ukky> yes, CRUX does not change that file in any way, it has been changed by upstream sometime after 3.7 ISO release

01:16 <farkuhar> I find it much more readable with the sigil omitted, so I employed this syntax in my revised prtrej: http://ix.io/4FcW

04:33 <cruxbot> [contrib.git/3.7]: open-vm-tools: removed libdnet dependency

04:34 <jaeger> Thanks for mentioning that, I forgot to remove it after they made it off by default in linux

06:52 <cruxbot> [contrib.git/3.7]: openblas: 0.3.23 -> 0.3.24

10:54 <cruxbot> [core.git/3.7]: libnghttp2: 1.55.1 -> 1.56.0

15:01 <cruxbot> [contrib.git/3.7]: libtommath: 1.2.0 -> 1.2.1

15:01 <cruxbot> [contrib.git/3.7]: matrix-synapse: 1.91.0 -> 1.91.1

15:01 <cruxbot> [contrib.git/3.7]: python3-aiohttp-socks: 0.8.0 -> 0.8.1

18:17 <cruxbot> [contrib.git/3.7]: pnpm: 8.7.1 -> 8.7.2

19:45 <r0ni> midnight commander seems awfully broken on any system i've installed 4.8.30 on

19:47 <r0ni> vfs things like "F3" to view an archive contents is broken, can no longer press enter on a archive to browse it's dir structure

19:48 <r0ni> completely shattered my workflow lol

19:49 <SiFuh> r0ni: mc.ext is now mc.ext.ini

19:51 <SiFuh> my mc.ext/mc.ext.ini is more towards what I want rather than their fancy cool scripts. http://ix.io/4FnR

19:51 <r0ni> file exists so i assume it should work

19:53 <SiFuh> To solve my issue I just copied mc.ext to mc.ext.ini

19:53 <SiFuh> GNU Midnight Commander 4.8.29

19:55 <r0ni> .29 worked .30 isn't, and i have /usr/etc/mc.ext.ini

19:55 <r0ni> revdep didn't complain so it's not a lib

19:55 <SiFuh> r0ni: is your mc.ext.ini the exact same file as your mc.ext ?

19:56 <r0ni> there is no mc.ext just mc.ext.ini

19:58 <SiFuh> I hope you are looking in your home directory

19:59 <SiFuh> https://github.com/MidnightCommander/mc/blob/master/misc/mc.ext.ini.in is pretty much the same as the old mc.ext

19:59 <SiFuh> Actually I have no freaking idea why they did that

19:59 <r0ni> sure am not, but that would make hella sense lol

20:00 <SiFuh> Yeah start with ~/.config/mc/

20:00 <SiFuh> If there is nothing you will be using the system default

20:04 <r0ni> welp i just wiped .config/mc dir and things still don't work, i'll have to look at it later, i've used mc for decades and never had to f with a conf file for it lmao

20:05 <SiFuh> I have used MC since 1999 and I do tweak it occasionally :-)

20:06 <SiFuh> But the other day when I upgraded to 4.8.29 it gave me a warning message about how I am an ancient retard still using bullshit old formats. Rather than taking offence I copied the original to the the new and then it stopped insulting me.

20:07 <SiFuh> And then everything started working as I had originally configured again.

20:09 <r0ni> this is a fresh install, no old version to even complain about. just noticed on a vm then another, then my systems as well. they've only had .29 and .30 installed and it's all gone to the shitter

20:10 <SiFuh> You can use my config above to test it

20:10 <SiFuh> You might need to tweak a few things though

20:10 <SiFuh> http://ix.io/4FnR <-- This one

20:11 <SiFuh> At least it will give you an idea of where not to look

20:11 <r0ni> i'll try it out in a few, pushing updates to xfce repo

20:14 <r0ni> removal of gtksourceview3 broke mousepad, so i had to get a port up for that and change the deps, and whiskermenu now needs either elogind or systemd to function so i added elogind and also daemon so xfce can start things properly... now I should test it all lol

20:19 <r0ni> i've got half a distro in this repo at this point lmao

20:31 <r0ni> really need to work out some meta-packages so it's not such a PITA to install it all

20:32 <jaeger> I moved gtksourceview3 into the mate repo because I didn't see anything else using it but I could move it back to contrib if need be

20:32 <r0ni> ahh it's all good, mousepad can use 3 or 4 so i put 4 into my repo

20:33 <jaeger> Same thing with 4, both 3 and 4 are used by mate stuff, heh

20:34 <r0ni> IIRC it has differing features and some ppl prefer one or the other, but it looks the same to me no matter which one is in use

20:39 <ivandi> wrt midnight commander, https://gitlab.gnome.org/GNOME/glib/-/issues/3094, sed -i 's/\\/\\\\/g' /usr/etc/mc/mc.ext.ini

20:44 <ivandi> farkuhar: wrt build logs removal, imho oldfiles is a better place to implement it

20:44 <SiFuh> ivandi: Still wonder why the changed it

20:47 <farkuhar> ivandi: thanks for the input. samsep10l hasn't returned to tell us how the tests went, so I'll assume it's working fine until we hear otherwise.

20:47 <ivandi> SiFuh: no idea, but the glib issue is tagged 'expected behavior' so most likely it wont change

20:49 <farkuhar> My first reaction when samsep10l brought it up again was to advise using an external tool like oldfiles, but he went through the trouble of submitting a patch so I wanted to see if it could be done more cleanly.

20:49 <SiFuh> I thinking I am still on an older version of glib

20:49 <SiFuh> 2.76.4

20:50 <SiFuh> Hmm not that old