💾 Archived View for gemini.ctrl-c.club › ~phoebos › logs › freenode-kisslinux-2020-01-10.txt captured on 2024-05-10 at 14:41:26.
⬅️ Previous capture (2021-12-17)
-=-=-=-=-=-=-
2020-01-10T09:03:34 #kisslinux <nixi> my tip is to just get amd :^) 2020-01-10T12:11:53 #kisslinux <icyphox> stupid question: how do i enable the community repo? 2020-01-10T12:12:10 #kisslinux <icyphox> or is it documented somewhere, and i missed it? 2020-01-10T12:16:25 #kisslinux <dylanaraps> It's in the installation guide 2020-01-10T12:16:37 #kisslinux <dylanaraps> https://getkiss.org/pages/install#enable-the-community-repository 2020-01-10T12:16:47 #kisslinux <icyphox> oh jeez 2020-01-10T12:16:52 #kisslinux <icyphox> my bad, lol 2020-01-10T12:17:09 #kisslinux <icyphox> looked everywhere else but there. thanks. 2020-01-10T12:17:24 #kisslinux <icyphox> got xorg built 2020-01-10T12:17:29 #kisslinux <icyphox> that too a long ass while 2020-01-10T12:17:39 #kisslinux <icyphox> took* 2020-01-10T12:18:14 #kisslinux <dylanaraps> Heh 2020-01-10T12:24:29 #kisslinux <dzove855> dylanaraps: since i started following the wiki and so on. I have found that the website isn't nicely structured. I think everything what is in install and so on, should also be in Wiki. This can help poeple to find easily everything in a centralized page. 2020-01-10T12:27:01 #kisslinux <dylanaraps> The reason the guide isn't a part of the Wiki is that I don't want unchecked changes to be made to it. 2020-01-10T12:27:20 #kisslinux <dylanaraps> Anyone can send a PR for the guide right now if they'd like to make changes. 2020-01-10T12:27:32 #kisslinux <dylanaraps> The Wiki is a free for all (as it should be) 2020-01-10T12:27:58 #kisslinux <dylanaraps> The Wiki also has restrictions in place for the markdown to html conversion. 2020-01-10T12:28:08 #kisslinux <dylanaraps> No embedded HTML, no scripts, etc. 2020-01-10T12:29:11 #kisslinux <dzove855> Ah okey, didn't know this. the reason is clear. I think when i have time, i will just use the guide, and create splitted wiki's about it, which could help and while i have seens other things missing which could be nice to add into the wiki. 2020-01-10T12:30:16 #kisslinux <dylanaraps> I'm happy for changes to be made to the guide directly via pull requests if you'd like to do so. 2020-01-10T12:30:33 #kisslinux <dylanaraps> I *could* also move the guide to the Wiki "visually". 2020-01-10T12:30:54 #kisslinux <dylanaraps> So it'd look like it's a part of the Wiki but would still be subject to approval for changes. 2020-01-10T12:31:11 #kisslinux <dylanaraps> I'd like to rewrite the entire website at some stage. 2020-01-10T12:31:15 #kisslinux <dylanaraps> This would be version 4 lol 2020-01-10T12:33:46 #kisslinux <dzove855> Yeah will have a look when everything is working perfectly on my side. :D 2020-01-10T12:33:54 #kisslinux <dylanaraps> Awesome 2020-01-10T12:33:59 #kisslinux <dzove855> haha i sense the motiviation for the rewrite 2020-01-10T12:35:40 #kisslinux <dzove855> also i have seen yesterday while building firefox, that we could add in the manpage of Kiss RUSTFLAGS='-C codegen-units=$digit' which will lead rust to spawn only a limited number of process instead using nproc 2020-01-10T12:35:54 #kisslinux <dzove855> (Note: not sure that it uses the result of nproc) 2020-01-10T12:36:42 #kisslinux <dylanaraps> At the same time we should mention limiting ninja's (samurai really) process count too. 2020-01-10T12:37:15 #kisslinux <dylanaraps> Also CMAKE_GENERATOR needs a fix. 2020-01-10T12:38:17 #kisslinux <dzove855> yeaah, could even be better, just by setting something like KISS_UNITS=2, which will set ninja, rust, make and so on in the runtime. This will not impact anything and would be more user friendly 2020-01-10T12:43:50 #kisslinux <dylanaraps1> Turns out the Ninja developers don't want to add an ENVAR to control job count. 2020-01-10T12:43:58 #kisslinux <dylanaraps1> They recommend '-j #' instead... 2020-01-10T12:44:05 #kisslinux <dylanaraps1> Let me check Samurai (which we use). 2020-01-10T12:47:19 #kisslinux <dzove855> *yey dylanaraps1 is doing incremention* 2020-01-10T12:48:02 #kisslinux <dzove855> Well, if it is the case, then it will be more complicated 2020-01-10T12:49:56 #kisslinux <dylanaraps1> Samurai supports a flag for jobs. 2020-01-10T12:50:10 #kisslinux <dylanaraps1> This is a fun read about Ninja related to it: https://github.com/ninja-build/ninja/issues/1482 2020-01-10T12:51:23 #kisslinux <lieuxnoir> any reason why linux-headers is still not updated to 5.4.10? 2020-01-10T12:52:10 #kisslinux <dylanaraps1> I'm on it 2020-01-10T12:52:32 #kisslinux <dylanaraps1> There's no rush to upgrade it as newer kernels will always work on older headers. 2020-01-10T12:52:59 #kisslinux <dylanaraps1> I check kernel.org a lot less than I check repology.org for outdated packages. 2020-01-10T12:53:00 #kisslinux <lieuxnoir> i also feel like writing a wiki for including microcodes built into the kernel 2020-01-10T12:53:08 #kisslinux <dylanaraps1> Nice 2020-01-10T12:53:11 #kisslinux <dylanaraps1> I'm for that 2020-01-10T12:53:40 #kisslinux <dylanaraps1> Once 5.4 becomes the LTS I'll be pinning our linux-headers version to it. 2020-01-10T12:54:09 #kisslinux <dylanaraps1> As this allows users to use the LTS kernel, current kernel or rc (mainline) kernel without the slight risk of a headers issue. 2020-01-10T12:54:12 #kisslinux <lieuxnoir> so KISS is really switching to lts full time? 2020-01-10T12:54:18 #kisslinux <lieuxnoir> oh nice 2020-01-10T12:54:23 #kisslinux <dylanaraps1> The headers only 2020-01-10T12:54:28 #kisslinux <dylanaraps1> And only once 5.4 becomes LTS. 2020-01-10T12:54:50 #kisslinux <dylanaraps1> And only since newer kernels are always compatible with older headers. 2020-01-10T12:55:03 #kisslinux <dylanaraps1> It makes more sense and saves you downloading the kernel sources every release. :P 2020-01-10T12:56:43 #kisslinux <dylanaraps1> Added SAMUFLAGS to the man page: https://github.com/kisslinux/kiss/commit/feac556f5689526d22ac83494f9ec278f5175b84 2020-01-10T12:57:05 #kisslinux <dylanaraps1> linux-headers update pushed as well. 2020-01-10T13:00:03 #kisslinux <lieuxnoir> kiss uses the original filenames for its sources, right? 2020-01-10T13:00:44 #kisslinux <dylanaraps1> https://website.com/pkg-1.2.3.tar.xz --> pkg-1.2.3.tar.xz 2020-01-10T13:00:53 #kisslinux <dylanaraps1> It runs a basename essentially. 2020-01-10T13:01:15 #kisslinux <dylanaraps1> Whatever is *after* the last / becomes the file name. 2020-01-10T13:02:16 #kisslinux <lieuxnoir> which means `curl "$src" -LO"` would make much more sense, no? 2020-01-10T13:02:35 #kisslinux <dylanaraps1> curl may name the file differently. 2020-01-10T13:03:55 #kisslinux <dylanaraps1> curl strips ?blabla from URLs iirc. 2020-01-10T13:04:21 #kisslinux <dylanaraps1> Whereas I need this information to be a part of the file name. 2020-01-10T13:04:51 #kisslinux <konimex> speaking of curl, one pkg errored when I try to download with curl, some cert error iirc, it was pkgconf(?) 2020-01-10T13:05:16 #kisslinux <dylanaraps1> Let me see if I can reproduce. 2020-01-10T13:05:39 #kisslinux <dylanaraps1> Works fine here. 2020-01-10T13:05:59 #kisslinux <dylanaraps1> Note that busybox wget silenced all SSL errors. 2020-01-10T13:06:03 #kisslinux <konimex> yeah, a package but can't remember which, what about mandoc? 2020-01-10T13:06:17 #kisslinux <dylanaraps1> Both use libressl underneath so I'd assume the error would be in both. 2020-01-10T13:06:20 #kisslinux <dylanaraps1> Let me try it. 2020-01-10T13:06:54 #kisslinux <dylanaraps1> Works fine here too. 2020-01-10T13:07:14 #kisslinux <dylanaraps1> I tested all packages when I made the change. 2020-01-10T13:07:17 #kisslinux <konimex> hmm.. I'll need to take a look later in monday 2020-01-10T13:07:35 #kisslinux <lieuxnoir> correct me if i'm wrong, but isn't -O just does what ${src##*/} is doing? 2020-01-10T13:07:45 #kisslinux <dylanaraps1> lieuxnoir: Nope 2020-01-10T13:08:05 #kisslinux <dylanaraps1> If I include ?no-extract in the URL '-O' will strip it. 2020-01-10T13:08:16 #kisslinux <dylanaraps1> '-o ${src##*/}' won't. 2020-01-10T13:08:42 #kisslinux <lieuxnoir> ahh yes 2020-01-10T13:09:06 #kisslinux <lieuxnoir> but why are they needed though? 2020-01-10T13:09:08 #kisslinux <dylanaraps1> We use ?no-extract as a means of disabling archive extraction when needed. 2020-01-10T13:09:13 #kisslinux <dylanaraps1> See rust's build file. 2020-01-10T13:09:33 #kisslinux <dylanaraps1> https://github.com/kisslinux/repo/blob/master/extra/rust/build 2020-01-10T13:09:42 #kisslinux <dylanaraps1> https://github.com/kisslinux/repo/blob/master/extra/rust/sources 2020-01-10T13:12:12 #kisslinux <lieuxnoir> that explains it 2020-01-10T13:12:58 #kisslinux <dylanaraps1> Basically making use of how URLs work to enable the feature. 2020-01-10T13:13:07 #kisslinux <dylanaraps1> There's a lot of tricks like this in kiss. 2020-01-10T13:17:43 #kisslinux <lieuxnoir> btw, firefox can't detect my webcam 2020-01-10T13:18:36 #kisslinux <lieuxnoir> the tab just crashes when i try to video call someone on messenger 2020-01-10T13:18:54 #kisslinux <dylanaraps1> Are you running with firefox-privacy or without it? 2020-01-10T13:19:01 #kisslinux <lieuxnoir> without it 2020-01-10T13:20:32 #kisslinux <dylanaraps1> Any error in the terminal? 2020-01-10T13:20:48 #kisslinux <dylanaraps1> Can run Firefox with 'firefox >file 2>&1' and then attempt to use the webcam. 2020-01-10T13:21:10 #kisslinux <lieuxnoir> oh, i'll check 2020-01-10T13:21:12 #kisslinux <lieuxnoir> i used dmenu 2020-01-10T13:22:36 #kisslinux <lieuxnoir> brb 2020-01-10T13:28:32 #kisslinux <lieuxnoir> https://termbin.com/bkyn 2020-01-10T13:29:02 #kisslinux <lieuxnoir> the relevant messages are the ones with something like audioipc 2020-01-10T13:30:26 #kisslinux <lieuxnoir> and not on video call, btw, it's when i try to record a video message 2020-01-10T13:31:22 #kisslinux <lieuxnoir> video calls and picture doesn't respond at all, and doesn't log anything as well 2020-01-10T13:33:22 #kisslinux <dylanaraps1> Can you reproduce it on this page?: https://mozilla.github.io/webrtc-landing/gum_test.html 2020-01-10T13:35:23 #kisslinux <lieuxnoir> nope 2020-01-10T13:35:38 #kisslinux <lieuxnoir> everything says success but all i can see are colors 2020-01-10T13:37:08 #kisslinux <dylanaraps1> What about this test?: https://www.onlinemictest.com/webcam-test/ 2020-01-10T13:38:18 #kisslinux <lieuxnoir> colors as well 2020-01-10T13:38:58 #kisslinux <dylanaraps1> I wonder if you can try setting your Facebook profile picture using the webcam. 2020-01-10T13:39:15 #kisslinux <dylanaraps1> I'll see if I can reproduce at all. 2020-01-10T13:40:18 #kisslinux <lieuxnoir> too bad, i don't use facebook 2020-01-10T13:40:30 #kisslinux <lieuxnoir> i just registered on messenger with my number 2020-01-10T13:41:24 #kisslinux <dylanaraps1> Oh yeah. I don't have my webcam drivers enabled. 2020-01-10T13:41:34 #kisslinux <dylanaraps1> It's taped too but the test should've worked anyway. 2020-01-10T13:41:55 #kisslinux <lieuxnoir> `mpv /dev/video0` works fine too, so not a kernel issue, i guess 2020-01-10T13:44:45 #kisslinux <dylanaraps1> It's tricky as web browsers are so complex :( 2020-01-10T13:44:55 #kisslinux <dylanaraps1> Do you have any older Firefox builds in your cache? 2020-01-10T13:45:05 #kisslinux <dylanaraps1> You could try each version to see if all are affected. 2020-01-10T13:45:50 #kisslinux <lieuxnoir> it was also present in 71 2020-01-10T13:46:32 #kisslinux <dylanaraps1> Got anything older by chance? 2020-01-10T13:47:08 #kisslinux <lieuxnoir> i re-installed kiss, so 71 was the first one i built 2020-01-10T13:48:39 #kisslinux <dylanaraps1> Ah 2020-01-10T13:48:45 #kisslinux <dylanaraps1> All good 2020-01-10T13:49:15 #kisslinux <dylanaraps1> I think it's worth opening an issue upstream though they'd most likely ask you to build a debug version... 2020-01-10T13:52:44 #kisslinux <dylanaraps1> This might be a bug?: https://bugzilla.mozilla.org/show_bug.cgi?id=1605118 2020-01-10T13:53:13 #kisslinux <dylanaraps1> Same broken pipe error. 2020-01-10T13:53:35 #kisslinux <dylanaraps1> Actually the only search result for the error in its entirety. 2020-01-10T14:05:55 #kisslinux <lieuxnoir> seems to be fixed on 73 2020-01-10T14:06:36 #kisslinux <dylanaraps1> Nice 2020-01-10T14:06:56 #kisslinux <dylanaraps1> I just pushed the latest beta to the repository (73.0b3) if you'd like to do a build and test. 2020-01-10T14:06:57 #kisslinux <lieuxnoir> is 73 in testing? 2020-01-10T14:07:00 #kisslinux <dylanaraps1> Yes 2020-01-10T14:07:04 #kisslinux <lieuxnoir> i haven't checked 2020-01-10T14:07:04 #kisslinux <dylanaraps1> b1-b3 2020-01-10T14:07:26 #kisslinux <lieuxnoir> oh nice, i might try building that tomorrow 2020-01-10T14:07:31 #kisslinux <dylanaraps1> I'm running b2 and there's a minor UI issue. 2020-01-10T14:07:57 #kisslinux <dylanaraps1> Sometimes the nav buttons (back/forward/reload) disappear. 2020-01-10T14:08:26 #kisslinux <dylanaraps1> b3 should also build fine as only 20~ random commits have been made. 2020-01-10T14:08:29 #kisslinux <dylanaraps1> Minor stuff 2020-01-10T14:10:44 #kisslinux <lieuxnoir> funny, there was also one time before when i click on Customize and firefox just crashes 2020-01-10T14:11:21 #kisslinux <dylanaraps1> Yeah 2020-01-10T14:11:27 #kisslinux <dylanaraps1> I fixed that a while ago. 2020-01-10T14:11:41 #kisslinux <lieuxnoir> btw, sowm doesn't detect keybinds with ShiftMask 2020-01-10T14:12:08 #kisslinux <lieuxnoir> ahh i see 2020-01-10T14:12:20 #kisslinux <lieuxnoir> it's in the issues 2020-01-10T14:13:02 #kisslinux <dylanaraps1> Yup 2020-01-10T14:13:24 #kisslinux <dylanaraps1> It's a tricky issue to solve as I need to make sowm ignore certain modifiers. 2020-01-10T14:13:41 #kisslinux <dylanaraps1> I'm also not very motivated as I run my system with these keys disabled. 2020-01-10T14:13:55 #kisslinux <dylanaraps1> It's why I didn't notice it while writing sowm. :P 2020-01-10T14:14:12 #kisslinux <dylanaraps1> Turns out I'm the 9th most active GitHub user right now. https://commits.top/worldwide.html 2020-01-10T14:14:35 #kisslinux <lieuxnoir> wow congrats 2020-01-10T14:16:05 #kisslinux <dylanaraps1> I'll be 6th once the site updates :P 2020-01-10T14:20:02 #kisslinux <lieuxnoir> big changes? 2020-01-10T14:20:56 #kisslinux <dylanaraps1> It's just 4~ days out of date. 2020-01-10T15:25:21 #kisslinux <icyphox> has anyone experienced their wireless dying because dhcpcd doesn't renew the lease? 2020-01-10T15:25:37 #kisslinux <icyphox> i have to manually do it each time 2020-01-10T15:29:21 #kisslinux <lieuxnoir> how do you bring up your wireless? 2020-01-10T15:33:21 #kisslinux <dylanaraps1> I had no issues when I was using dhcpcd. 2020-01-10T15:33:35 #kisslinux <dylanaraps1> Another option you have is eiwd which can also replace dhcpcd. 2020-01-10T15:43:16 #kisslinux <icyphox> lieuxnoir: wpa_supplicant 2020-01-10T15:43:32 #kisslinux <icyphox> aaah i'm in x 2020-01-10T15:43:47 #kisslinux <icyphox> but uh, looks like my keyboard and touchpad don't work 2020-01-10T15:43:49 #kisslinux <icyphox> hmm 2020-01-10T15:44:10 #kisslinux <lieuxnoir> adduser input <your username> 2020-01-10T15:44:12 #kisslinux <icyphox> dylanaraps: yeah i saw eiwd. i shall try it out. 2020-01-10T15:44:16 #kisslinux <icyphox> oh right 2020-01-10T15:44:20 #kisslinux <icyphox> lol 2020-01-10T15:44:38 #kisslinux <lieuxnoir> >wpa_supplicant 2020-01-10T15:44:50 #kisslinux <icyphox> that's right 2020-01-10T15:44:56 #kisslinux <lieuxnoir> add `update_config=1` in your wpa_supplicant*.conf 2020-01-10T15:45:06 #kisslinux <icyphox> it's there 2020-01-10T15:45:20 #kisslinux <dylanaraps> Do you have the correct input drivers enabled? 2020-01-10T15:45:37 #kisslinux <icyphox> i should think so 2020-01-10T15:45:45 #kisslinux <dylanaraps> Let me send you mine. 2020-01-10T15:45:55 #kisslinux <icyphox> guess we'll find out. i've added my user to input. let me see. 2020-01-10T15:46:25 #kisslinux <icyphox> yeah that didn't work 2020-01-10T15:46:27 #kisslinux <icyphox> bummer 2020-01-10T15:46:34 #kisslinux <icyphox> probably kernel config again 2020-01-10T15:47:09 #kisslinux <lieuxnoir> did you reboot? 2020-01-10T15:47:17 #kisslinux <icyphox> logout / login 2020-01-10T15:47:46 #kisslinux <dylanaraps> Got this?: CONFIG_INPUT_UINPUT=y 2020-01-10T15:48:07 #kisslinux <icyphox> nope 2020-01-10T15:48:11 #kisslinux <dylanaraps> > CONFIG_INPUT_MOUSEDEV=y 2020-01-10T15:48:15 #kisslinux <dylanaraps> > CONFIG_INPUT_MOUSEDEV_PSAUX=y 2020-01-10T15:48:30 #kisslinux <dylanaraps> You also need xf86-input-libinput installed. 2020-01-10T15:48:39 #kisslinux <icyphox> ah yes 2020-01-10T15:49:38 #kisslinux <dylanaraps> You may also need the specific drivers for your hardware. 2020-01-10T15:50:18 #kisslinux <icyphox> i shall check 2020-01-10T15:55:15 #kisslinux <icyphox> heh 2020-01-10T15:55:24 #kisslinux <icyphox> so the mouse works, but the keyboard doesn't 2020-01-10T15:55:39 #kisslinux <icyphox> dylanaraps: got any kernel config lines for the keyboard? 2020-01-10T16:04:14 #kisslinux <dylanaraps> icyphox: let me eat my dinner first. Give me 5 mins :P 2020-01-10T16:04:25 #kisslinux <icyphox> no hurry! haha 2020-01-10T16:04:43 #kisslinux <icyphox> i'm looking at the xorg log, looks like it's not seeing my kb 2020-01-10T16:04:53 #kisslinux <icyphox> or i think that's what it says 2020-01-10T16:05:15 #kisslinux <icyphox> "no input driver specified, ignoring this device" 2020-01-10T16:05:15 #kisslinux <dzove855> i fixed this yesterday with installing xinput 2020-01-10T16:05:19 #kisslinux <icyphox> oh 2020-01-10T16:05:24 #kisslinux <icyphox> let me try that 2020-01-10T16:07:42 #kisslinux <icyphox> hmm don't think that worked 2020-01-10T16:09:17 #kisslinux <dzove855> will be back in few minutes or one hour. Finally finished with working. 2020-01-10T16:10:22 #kisslinux <icyphox> sure 2020-01-10T16:14:44 #kisslinux <dylanaraps> CONFIG_KEYBOARD_ATKBD=y 2020-01-10T16:14:48 #kisslinux <dylanaraps> CONFIG_INPUT_KEYBOARD=y 2020-01-10T16:15:17 #kisslinux <icyphox> thanks let me see 2020-01-10T16:15:54 #kisslinux <icyphox> yeah they're both set 2020-01-10T16:15:56 #kisslinux <icyphox> hm 2020-01-10T16:16:27 #kisslinux <dylanaraps> Ah 2020-01-10T16:16:30 #kisslinux <dylanaraps> Let me dig deeper. 2020-01-10T16:16:52 #kisslinux <icyphox> i can't even find my keyboard in lshw 2020-01-10T16:17:43 #kisslinux <icyphox> ah, they're both handled by the i8042 driver 2020-01-10T16:17:48 #kisslinux <icyphox> that ships in the kernel 2020-01-10T16:17:55 #kisslinux <icyphox> also kb works in the tty 2020-01-10T16:18:06 #kisslinux <icyphox> so it should be an X related issue then 2020-01-10T16:18:11 #kisslinux <dylanaraps> Nice 2020-01-10T16:18:12 #kisslinux <dylanaraps> OK 2020-01-10T16:18:39 #kisslinux <dylanaraps> Try 'libinput list-devices' 2020-01-10T16:19:20 #kisslinux <icyphox> ok let me see 2020-01-10T16:19:33 #kisslinux <icyphox> yep my kb is there 2020-01-10T16:19:47 #kisslinux <icyphox> "AT Translated Set 2 keyboard 2020-01-10T16:19:49 #kisslinux <icyphox> " 2020-01-10T16:20:05 #kisslinux <dylanaraps> Nice 2020-01-10T16:20:08 #kisslinux <dylanaraps> I'll give you my config to try. 2020-01-10T16:20:28 #kisslinux <icyphox> do i need to do some /etc/X11/xorg.conf jazz 2020-01-10T16:21:14 #kisslinux <dylanaraps> Maybe 2020-01-10T16:21:25 #kisslinux <icyphox> hmm 2020-01-10T16:21:33 #kisslinux <icyphox> ok i'll read up on it 2020-01-10T16:22:09 #kisslinux <dylanaraps> https://termbin.com/rczg 2020-01-10T16:22:14 #kisslinux <dylanaraps> In: /etc/X11/xorg.conf.d/40-libinput.conf 2020-01-10T16:22:24 #kisslinux <dylanaraps> You can remove all of the Option lines. 2020-01-10T16:22:39 #kisslinux <dylanaraps> This will force all devices to go through libinput. 2020-01-10T16:22:46 #kisslinux <icyphox> cool 2020-01-10T16:23:12 #kisslinux <icyphox> what's the diff b/w xinput and libinput 2020-01-10T16:23:15 #kisslinux <dylanaraps> Not sure if it will fix it but it's worth a shot. 2020-01-10T16:23:44 #kisslinux <dylanaraps> libinput is the driver for all input devices and xinput is an interface for Xorg devices. 2020-01-10T16:23:53 #kisslinux <dylanaraps> In the end they let you control the same devices. 2020-01-10T16:23:58 #kisslinux <icyphox> ah ok 2020-01-10T16:24:03 #kisslinux <icyphox> do i need both 2020-01-10T16:24:22 #kisslinux <dylanaraps> You need only libinput/xf86-input-libinput 2020-01-10T16:24:34 #kisslinux <icyphox> alright 2020-01-10T16:24:43 #kisslinux <dylanaraps> I personally use xinput as a means of disabling my touchscreen. 2020-01-10T16:25:01 #kisslinux <dylanaraps> I *could* use a Xorg config to do this but I'm lazy :P 2020-01-10T16:25:15 #kisslinux <icyphox> heh 2020-01-10T16:25:47 #kisslinux <lieuxnoir> not related, but i removed the ribbon cable connecting my trackpad to the motherboard haha 2020-01-10T16:26:16 #kisslinux <dylanaraps> lol 2020-01-10T16:27:08 #kisslinux <lieuxnoir> all hail the trackpoint 2020-01-10T16:29:12 #kisslinux <icyphox> aaa nope kb still dead 2020-01-10T16:29:22 #kisslinux <icyphox> i'll see what the logs say 2020-01-10T16:29:38 #kisslinux <icyphox> lieuxnoir: lol 2020-01-10T16:30:58 #kisslinux <lieuxnoir> try re-installing xorg 2020-01-10T16:31:20 #kisslinux <icyphox> hm 2020-01-10T16:31:22 #kisslinux <icyphox> okay 2020-01-10T16:31:25 #kisslinux <icyphox> sure why not 2020-01-10T16:31:59 #kisslinux <lieuxnoir> also remove the configs you've made 2020-01-10T16:32:02 #kisslinux <lieuxnoir> if there are any 2020-01-10T16:32:07 #kisslinux <lieuxnoir> or back em up 2020-01-10T16:32:09 #kisslinux <icyphox> only libinput 2020-01-10T16:32:14 #kisslinux <dylanaraps> Send a full dmesg/xorg log too. 2020-01-10T16:32:26 #kisslinux <icyphox> alright 2020-01-10T16:36:09 #kisslinux <icyphox> Xorg log: https://0x0.st/zRL2.log 2020-01-10T16:36:50 #kisslinux <icyphox> dmesg: https://termbin.com/ibhb 2020-01-10T16:38:13 #kisslinux <dylanaraps> Do you have libinput installed? 2020-01-10T16:38:18 #kisslinux <icyphox> yes 2020-01-10T16:38:19 #kisslinux <dylanaraps> It is absent from your Xorg log. 2020-01-10T16:38:26 #kisslinux <dylanaraps> xf86-input-libinput? 2020-01-10T16:38:39 #kisslinux <icyphox> hm 2020-01-10T16:38:53 #kisslinux <icyphox> yep 2020-01-10T16:39:05 #kisslinux <dylanaraps> Weird. 2020-01-10T16:39:13 #kisslinux <dylanaraps> Looks like Xorg isn't using libinput at all. 2020-01-10T16:39:15 #kisslinux <icyphox> wait, iirc, Xorg.0.log is the most recent? 2020-01-10T16:39:40 #kisslinux <icyphox> oh 2020-01-10T16:39:41 #kisslinux <icyphox> nopeee 2020-01-10T16:39:53 #kisslinux <icyphox> it's incremental apparently 2020-01-10T16:40:07 #kisslinux <dylanaraps> [ 0.848133] input: AT Translated Set 2 keyboard as /devices/platform/i8042/serio0/input/input4 2020-01-10T16:40:08 #kisslinux <icyphox> i sent you an old log then dylanaraps 2020-01-10T16:40:20 #kisslinux <dylanaraps> Your keyboard is detected by udev. 2020-01-10T16:40:26 #kisslinux <icyphox> yes 2020-01-10T16:40:28 #kisslinux <dylanaraps> And by extension the kernel. 2020-01-10T16:40:32 #kisslinux <icyphox> yes 2020-01-10T16:40:54 #kisslinux <lieuxnoir> what is the output of `groups`? 2020-01-10T16:41:10 #kisslinux <dylanaraps> lieuxnoir: Mouse works so it's unrelated me thinks. 2020-01-10T16:41:13 #kisslinux <icyphox> https://0x0.st/zRLp.log -- newest Xorg log 2020-01-10T16:41:20 #kisslinux <lieuxnoir> i see 2020-01-10T16:43:39 #kisslinux <dylanaraps> Hm 2020-01-10T16:43:42 #kisslinux <dylanaraps> Log looks fine to me. 2020-01-10T16:44:05 #kisslinux <lieuxnoir> what's in your .xinitrc? 2020-01-10T16:45:24 #kisslinux <icyphox> let me post 2020-01-10T16:45:26 #kisslinux <icyphox> hold 2020-01-10T16:46:25 #kisslinux <icyphox> https://0x0.st/zRLW.bin -- .xinitrc 2020-01-10T16:46:31 #kisslinux <icyphox> ignore the extension 2020-01-10T16:46:54 #kisslinux <icyphox> nvm let me use termbin 2020-01-10T16:47:12 #kisslinux <dylanaraps> Try removing the setxkbmap line. 2020-01-10T16:47:20 #kisslinux <icyphox> https://termbin.com/v2mue 2020-01-10T16:47:30 #kisslinux <icyphox> but i don't have setxkbmap installed 2020-01-10T16:47:36 #kisslinux <dylanaraps> Oh 2020-01-10T16:47:37 #kisslinux <dylanaraps> lol 2020-01-10T16:47:48 #kisslinux <icyphox> heh so it should have no effect 2020-01-10T16:47:49 #kisslinux <lieuxnoir> st & 2020-01-10T16:47:51 #kisslinux <lieuxnoir> what 2020-01-10T16:48:06 #kisslinux <dylanaraps> sowm and sxhkd might not be a good mix either 2020-01-10T16:48:10 #kisslinux <icyphox> oh yeah i wanted to see if st would launch 2020-01-10T16:48:19 #kisslinux <icyphox> don't have sxhkd either dylanaraps 2020-01-10T16:50:37 #kisslinux <lieuxnoir> try commenting out everything except the sowm line 2020-01-10T16:50:38 #kisslinux <dylanaraps> So st launches and still no input? 2020-01-10T16:51:03 #kisslinux <icyphox> no st doesn't launch 2020-01-10T16:51:13 #kisslinux <lieuxnoir> wait 2020-01-10T16:51:15 #kisslinux <icyphox> i see only my mouse 2020-01-10T16:51:29 #kisslinux <lieuxnoir> is bash installed? 2020-01-10T16:51:32 #kisslinux <icyphox> yes 2020-01-10T16:51:42 #kisslinux <icyphox> that's my shell 2020-01-10T16:51:58 #kisslinux <dylanaraps> Ah 2020-01-10T16:52:13 #kisslinux <dylanaraps> Try making your .xinitrc solely exec bla/bla/sowm 2020-01-10T16:52:22 #kisslinux <icyphox> okay wait 2020-01-10T16:52:30 #kisslinux <dylanaraps> I'm unsure if sowm supports opening applications before it is executed. 2020-01-10T16:52:41 #kisslinux <dylanaraps> It only watches post launch. 2020-01-10T16:52:55 #kisslinux <icyphox> yeah i wanted to test that 2020-01-10T16:53:03 #kisslinux <icyphox> looks like st never maps 2020-01-10T16:53:24 #kisslinux <icyphox> ok only sowm launch 2020-01-10T16:53:28 #kisslinux <icyphox> nothing 2020-01-10T16:53:34 #kisslinux <icyphox> no kb input 2020-01-10T16:53:51 #kisslinux <icyphox> using default sowm config btw 2020-01-10T16:53:57 #kisslinux <icyphox> so mod+return is st 2020-01-10T16:54:14 #kisslinux <icyphox> and mod+d is dmenu, both are installed 2020-01-10T16:55:07 #kisslinux <dylanaraps> You can try exec st as well 2020-01-10T16:55:10 #kisslinux <dylanaraps> That works iirc. 2020-01-10T16:55:55 #kisslinux <icyphox> without wm? 2020-01-10T16:56:21 #kisslinux <icyphox> no that won't work 2020-01-10T16:56:28 #kisslinux <icyphox> ok i'll try brb 2020-01-10T16:57:20 #kisslinux <icyphox> the problem rn is everytime i startx and kb doesn't work, i've to hard shutdown and reboot 2020-01-10T16:57:30 #kisslinux <icyphox> since i can't access the ttys 2020-01-10T16:57:35 #kisslinux <icyphox> (no kb) 2020-01-10T16:58:58 #kisslinux <dylanaraps> Yup. Not ideal. 2020-01-10T17:00:29 #kisslinux <dylanaraps> If you enable the Magic SysRq key you should be able to shutdown properly. 2020-01-10T17:00:30 #kisslinux <dylanaraps> https://en.wikipedia.org/wiki/Magic_SysRq_key 2020-01-10T17:01:08 #kisslinux <dylanaraps> See "REISUB" on the same page. 2020-01-10T17:01:26 #kisslinux <adamantium> What's the issue 2020-01-10T17:01:36 #kisslinux <dylanaraps> "Configuration" shows you how to enable it. 2020-01-10T17:01:41 #kisslinux <icyphox> oh 2020-01-10T17:01:41 #kisslinux <adamantium> oh keyboard 2020-01-10T17:01:42 #kisslinux <icyphox> neat 2020-01-10T17:01:47 #kisslinux <dylanaraps> No keyboard input under Xorg 2020-01-10T17:01:52 #kisslinux <dylanaraps> Mouse works though 2020-01-10T17:02:01 #kisslinux <adamantium> weird 2020-01-10T17:02:16 #kisslinux <dylanaraps> So this rules out the input group and libinput/xf86-input-libinput not being installed. 2020-01-10T17:03:22 #kisslinux <adamantium> It's hard to know btw, which packages we need vs are extra in the Xorg group 2020-01-10T17:03:28 #kisslinux <dylanaraps> The keyboard is also detected by the kernel, udev and libinput. 2020-01-10T17:03:31 #kisslinux <adamantium> I think we could improve this. 2020-01-10T17:03:44 #kisslinux <adamantium> A Xorg-minimal meta package would be nice. 2020-01-10T17:03:55 #kisslinux <adamantium> that included xorg server, libinput, the barebones essential 2020-01-10T17:04:17 #kisslinux <dylanaraps> kiss b xorg-server xf86-input-libinput xinit 2020-01-10T17:04:21 #kisslinux <dylanaraps> That should do it. 2020-01-10T17:04:31 #kisslinux <adamantium> why xinit 2020-01-10T17:04:39 #kisslinux <dylanaraps> To start the server. 2020-01-10T17:04:58 #kisslinux <adamantium> https://termbin.com/ei1l 2020-01-10T17:05:00 #kisslinux <icyphox> exec st doesn't work btw 2020-01-10T17:05:06 #kisslinux <dylanaraps> If I just made these two a dependency of xorg-server we'd have our minimal base. 2020-01-10T17:05:06 #kisslinux <adamantium> I rolled my own xinit 2020-01-10T17:05:39 #kisslinux <adamantium> ;) 2020-01-10T17:05:49 #kisslinux <icyphox> adamantium: lol hahaha 2020-01-10T17:05:52 #kisslinux <dylanaraps> icyphox: https://wiki.archlinux.org/index.php/Xinit#Starting_applications_without_a_window_manager 2020-01-10T17:06:25 #kisslinux <dylanaraps> Oh 2020-01-10T17:06:30 #kisslinux <dylanaraps> xinit is a dependency already. 2020-01-10T17:06:39 #kisslinux <dylanaraps> Basically, kiss b xorg-server xf86-input-libinput 2020-01-10T17:06:41 #kisslinux <icyphox> yeah so here's what happens: X starts, and stops immediately 2020-01-10T17:06:47 #kisslinux <dylanaraps> That gives you all of the required. 2020-01-10T17:06:52 #kisslinux <icyphox> with exec st 2020-01-10T17:06:54 #kisslinux <dylanaraps> Did you do 'exec st' 2020-01-10T17:06:56 #kisslinux <adamantium> cool 2020-01-10T17:06:58 #kisslinux <icyphox> yee 2020-01-10T17:06:59 #kisslinux <dylanaraps> Just those two words? 2020-01-10T17:07:01 #kisslinux <icyphox> yes 2020-01-10T17:07:06 #kisslinux <dylanaraps> Maybe your st is borked lol 2020-01-10T17:07:11 #kisslinux <adamantium> haha 2020-01-10T17:07:14 #kisslinux <icyphox> lol it's from the repos 2020-01-10T17:07:24 #kisslinux <icyphox> i did kiss b st 2020-01-10T17:07:32 #kisslinux <adamantium> ok do we have another application that can accept keyboard input 2020-01-10T17:07:41 #kisslinux <dylanaraps> dmenu 2020-01-10T17:07:46 #kisslinux <dylanaraps> 'exec dmenu' 2020-01-10T17:07:48 #kisslinux <adamantium> lol 2020-01-10T17:07:49 #kisslinux <icyphox> ah let me try that 2020-01-10T17:07:52 #kisslinux <dylanaraps> I doubt anyone has done that befoe lol 2020-01-10T17:07:59 #kisslinux <dylanaraps> before* 2020-01-10T17:08:23 #kisslinux <icyphox> huhhh wtf X starts and stops immediately 2020-01-10T17:08:24 #kisslinux <icyphox> wtf 2020-01-10T17:08:35 #kisslinux <adamantium> thats okay 2020-01-10T17:08:39 #kisslinux <adamantium> dmenu isn't up for the task 2020-01-10T17:08:49 #kisslinux <icyphox> hmm 2020-01-10T17:09:00 #kisslinux <adamantium> kiss b emacs 2020-01-10T17:09:02 #kisslinux <adamantium> exec emacs 2020-01-10T17:09:04 #kisslinux <adamantium> give it a whirl 2020-01-10T17:09:07 #kisslinux <icyphox> never 2020-01-10T17:09:08 #kisslinux <dylanaraps> lol 2020-01-10T17:09:19 #kisslinux <dylanaraps> kiss b gtk+3 2020-01-10T17:09:19 #kisslinux <adamantium> dylanaraps: did you click https://termbin.com/ei1l 2020-01-10T17:09:27 #kisslinux <dylanaraps> Then gtk3-demo-application 2020-01-10T17:09:39 #kisslinux <icyphox> lol ok let me try that 2020-01-10T17:09:44 #kisslinux <dylanaraps> This reminds me, I should purge these from the package. 2020-01-10T17:09:52 #kisslinux <dylanaraps> demos, yuck 2020-01-10T17:10:03 #kisslinux <dylanaraps> I did click on t 2020-01-10T17:10:04 #kisslinux <dylanaraps> it8 2020-01-10T17:10:06 #kisslinux <dylanaraps> it* 2020-01-10T17:10:20 #kisslinux <adamantium> it's pretty rediculous 2020-01-10T17:10:27 #kisslinux <dylanaraps> Oh yeah 2020-01-10T17:10:48 #kisslinux <adamantium> but it's my baby 2020-01-10T17:11:24 #kisslinux <dylanaraps> OK 2020-01-10T17:11:41 #kisslinux <dylanaraps> Can't make xorg-server depend on xf86-input-libinput as the latter already depends on the former. 2020-01-10T17:11:55 #kisslinux <dylanaraps> I think we just need some documentation for the install guide. 2020-01-10T17:12:02 #kisslinux <dylanaraps> A new section for Xorg and libinput. 2020-01-10T17:12:29 #kisslinux <adamantium> tbh tho dylanaraps you could just execute (start-process "Xorg" nil "Xorg" "-nolisten" "tcp" "-nolisten" "local" ":0" "vt1" "v" "-arinterval" "30" "-ardelay" "175") as a shell command , export DISPLAY=:0 and and fork a window manager. 2020-01-10T17:12:34 #kisslinux <adamantium> You dont need xinit 2020-01-10T17:12:46 #kisslinux <dylanaraps> ooo 2020-01-10T17:13:01 #kisslinux <adamantium> ^_^ 2020-01-10T17:13:18 #kisslinux <adamantium> your xinit is bloated 2020-01-10T17:13:23 #kisslinux <adamantium> ^_^ 2020-01-10T17:14:11 #kisslinux <Aarg[m]> icyphox, enable ssh, then kill x via phone 2020-01-10T17:14:21 #kisslinux <Aarg[m]> No need to reboot 2020-01-10T17:14:25 #kisslinux <dylanaraps> Aarg[m]: That's a great idea too. 2020-01-10T17:14:29 #kisslinux <adamantium> Yep 2020-01-10T17:14:34 #kisslinux <adamantium> termux if android 2020-01-10T17:14:43 #kisslinux <icyphox> ah good call 2020-01-10T17:14:48 #kisslinux <icyphox> i'll install ssh 2020-01-10T17:14:52 #kisslinux <dylanaraps> (openssh) 2020-01-10T17:14:55 #kisslinux <icyphox> but first, gtk+3 2020-01-10T17:15:02 #kisslinux <Aarg[m]> ssh always saves me from trouble 2020-01-10T17:15:11 #kisslinux <Aarg[m]> When the machine locks up and shit 2020-01-10T17:15:18 #kisslinux <icyphox> heh 2020-01-10T17:15:20 #kisslinux <dylanaraps> What's good is that you're laying the ground work for Firefox by doing this :P 2020-01-10T17:15:29 #kisslinux <dylanaraps> adamantium: Do you fork Xorg as well? 2020-01-10T17:15:35 #kisslinux <adamantium> yes 2020-01-10T17:15:40 #kisslinux <dylanaraps> ty 2020-01-10T17:15:42 #kisslinux <adamantium> it will stay running by default 2020-01-10T17:15:46 #kisslinux <icyphox> but heck, i've come so far. nearly 3 days in and the fuckin keyboard doesn't work?! 2020-01-10T17:15:57 #kisslinux <adamantium> dylanaraps: unless you specify. it will stay running. Take a glance at Xorg --help 2020-01-10T17:16:08 #kisslinux <dylanaraps> The best part is that you'll understand how your entire system works once done. 2020-01-10T17:16:14 #kisslinux <adamantium> ++ 2020-01-10T17:20:37 #kisslinux <dylanaraps> brb, testing my xinit. 2020-01-10T17:24:49 #kisslinux <Aarg[m]> So the xserver started with only st? 2020-01-10T17:26:12 #kisslinux <Aarg[m]> If yes, have you then tried spawning programs via sxhkd? 2020-01-10T17:26:52 #kisslinux <Aarg[m]> Asking because st won't have focus so you can't type anyway 2020-01-10T17:28:30 #kisslinux <dylanaraps> borked it 2020-01-10T17:28:33 #kisslinux <dylanaraps> lol 2020-01-10T17:30:05 #kisslinux <adamantium> let me see it 2020-01-10T17:31:01 #kisslinux <adamantium> starting x with st should get focus if you put the mouse over it 2020-01-10T17:32:24 #kisslinux <dylanaraps> I nuked it already. 2020-01-10T17:32:45 #kisslinux <adamantium> ah 2020-01-10T17:32:54 #kisslinux <dylanaraps> Was along the lines of 'X arg arg arg &' then '~/.xinitrc' 2020-01-10T17:33:01 #kisslinux <dylanaraps> Executing .xinitrc as a script. 2020-01-10T17:33:07 #kisslinux <dylanaraps> Which included 'sowm &'. 2020-01-10T17:33:10 #kisslinux <Aarg[m]> Are you sure, adamantium? 2020-01-10T17:33:11 #kisslinux <adamantium> nice 2020-01-10T17:33:40 #kisslinux <dylanaraps> Plus a bunch of regular commands which run and exit (xset, wallpaper etc) 2020-01-10T17:34:03 #kisslinux <dylanaraps> X would start but I'd be stuck in vt1. 2020-01-10T17:34:04 #kisslinux <adamantium> Aarg[m]: I'm fairly certain. I startx with exec emacs (vanilla) or even just exec firefox and i can input from my keyboard to it without a problem 2020-01-10T17:34:17 #kisslinux <adamantium> just need to hover with the mouse 2020-01-10T17:34:44 #kisslinux <adamantium> dylanaraps: why would be be stuck in vt1 2020-01-10T17:34:46 #kisslinux <Aarg[m]> Ok, I thought that's the wm doing it 2020-01-10T17:35:18 #kisslinux <dylanaraps> adamantium: I see the regular Xorg startup output but it doesn't actually load anything graphical. 2020-01-10T17:35:24 #kisslinux <dylanaraps> Just stuck in the tty. 2020-01-10T17:35:46 #kisslinux <adamantium> dylanaraps: if I understand correctly, just need to fork some apps then with the DISPLAY 2020-01-10T17:35:58 #kisslinux <adamantium> It's a black screen, the Xorg server, until it has apps 2020-01-10T17:36:33 #kisslinux <dylanaraps> Oh yeah, I also set 'export DISPLAY=:0' before running Xorg. 2020-01-10T17:36:43 #kisslinux <dylanaraps> I fork the WM too fyi. 2020-01-10T17:36:57 #kisslinux <dylanaraps> It leaves me in the same tty in the end though. 2020-01-10T17:37:16 #kisslinux <dylanaraps> Let me see how xinit works. 2020-01-10T17:38:35 #kisslinux <adamantium> dylanaraps: https://termbin.com/v6zb 2020-01-10T17:39:06 #kisslinux <adamantium> change vt1 to vt2 or whatever vt you want it to switch to 2020-01-10T17:40:05 #kisslinux <adamantium> you can make it dynamic with some $(tty) magic 2020-01-10T17:40:09 #kisslinux <dylanaraps> That's pretty much what I had. 2020-01-10T17:40:56 #kisslinux <dylanaraps> Anyway 2020-01-10T17:40:56 #kisslinux <dylanaraps> I'll mess with this later. 2020-01-10T17:40:56 #kisslinux <dylanaraps> I'd rather get icyphox's keyboard working :P 2020-01-10T17:41:01 #kisslinux <adamantium> cool 2020-01-10T17:41:09 #kisslinux <adamantium> I have to go to the store o/ 2020-01-10T17:44:11 #kisslinux <adamantium> When you see how startx/xinit works, you'll be more motivated to do what i'm doing 2020-01-10T17:44:20 #kisslinux <dylanaraps> I just looked 2020-01-10T17:44:22 #kisslinux <dylanaraps> It's big 2020-01-10T17:44:32 #kisslinux <adamantium> ya its crap 2020-01-10T17:44:54 #kisslinux <adamantium> when we can just launch Xorg and start a window manager ... lol 2020-01-10T17:48:23 #kisslinux <adamantium> and before you try your hand at writing a new startx script. you could check out earnestly's "sx" on github 2020-01-10T17:48:35 #kisslinux <adamantium> But my approach is more minimal than his 2020-01-10T17:52:50 #kisslinux <dylanaraps> I've seen sx yeah 2020-01-10T17:57:35 #kisslinux <dylanaraps> brb 2020-01-10T17:57:38 #kisslinux <dylanaraps> testing my new script 2020-01-10T17:57:53 #kisslinux <E5ten> do exports in scripts get used in traps? like if I export a variable, when the trap runs will the running trap's environment also have that variable exported? I've got absolutely no reason to think it wouldn't be but I'm just double checking 2020-01-10T17:58:37 #kisslinux <dylanaraps> woo hoo 2020-01-10T17:58:47 #kisslinux <dylanaraps> https://termbin.com/1bky 2020-01-10T17:59:05 #kisslinux <E5ten> nice lol 2020-01-10T18:00:15 #kisslinux <dylanaraps> E5ten: Try: export var=1; trap 'echo $var' EXIT; exit 2020-01-10T18:00:35 #kisslinux <dylanaraps> yup 2020-01-10T18:00:36 #kisslinux <dylanaraps> Works 2020-01-10T18:00:51 #kisslinux <dylanaraps> trap internally just calls eval on the arguments 2020-01-10T18:00:58 #kisslinux <dylanaraps> lets not talk about this though :^) 2020-01-10T18:01:07 #kisslinux <E5ten> ok lol 2020-01-10T18:01:12 #kisslinux <dylanaraps> shhh 2020-01-10T18:07:30 #kisslinux <E5ten> would you mind looking me at the trap that uses ls in fakeroot and telling me what the fuck you think is going on there? 2020-01-10T18:08:02 #kisslinux <E5ten> cuz my best guess is that uh it's literally using ls -l / because it needs to delay exiting or something and that's something that takes a little bit of time? 2020-01-10T18:08:16 #kisslinux <E5ten> and if that's it I think I'm going to explode 2020-01-10T18:10:11 #kisslinux <dylanaraps> wot 2020-01-10T18:10:36 #kisslinux <dylanaraps> Link? 2020-01-10T18:10:45 #kisslinux <dylanaraps> Found it 2020-01-10T18:10:45 #kisslinux <E5ten> https://salsa.debian.org/clint/fakeroot/blob/master/scripts/fakeroot.in 2020-01-10T18:11:11 #kisslinux <dylanaraps> On exit right? 2020-01-10T18:11:18 #kisslinux <dylanaraps> and int 2020-01-10T18:11:28 #kisslinux <E5ten> yeah 2020-01-10T18:12:08 #kisslinux <dylanaraps> Disgusting 2020-01-10T18:12:25 #kisslinux <dylanaraps> I'm glad I dropped fakeroot early on in KISS 2020-01-10T18:12:52 #kisslinux <E5ten> like I'm making a ton of adjustments to improve this script and then I'll try to make a PR, biggest thing being not using getopt, but like what the fuck do I even do about that 2020-01-10T18:13:15 #kisslinux <E5ten> it's one of the worst things I've ever seen and I just have no clue how to deal with it 2020-01-10T18:13:19 #kisslinux <dylanaraps> Only Debian are "kinda" maintaining it. 2020-01-10T18:13:25 #kisslinux <dylanaraps> There's no upstream iirc. 2020-01-10T18:13:39 #kisslinux <E5ten> I think they are upstream 2020-01-10T18:13:44 #kisslinux <dylanaraps> ls -l / takes 0ms on my machine. 2020-01-10T18:13:58 #kisslinux <dylanaraps> Though it uses LDPRELOAD 2020-01-10T18:14:13 #kisslinux <dylanaraps> Does it depend on the fakeroot overhead? 2020-01-10T18:14:42 #kisslinux <dylanaraps> Or does it depend on the exit status of ls -l /? 2020-01-10T18:15:25 #kisslinux <E5ten> it doesn't depend on the exit status I think, and ls -l through fakeroot for me still takes like no time 2020-01-10T18:15:28 #kisslinux <dylanaraps> Send me a "compiled" fakeroot? 2020-01-10T18:16:24 #kisslinux <E5ten> http://ix.io/272O 2020-01-10T18:16:41 #kisslinux <E5ten> there's like an actual compiled bit too though 2020-01-10T18:16:51 #kisslinux <dylanaraps> Yeah, I know 2020-01-10T18:16:53 #kisslinux <E5ten> and I'm on glibc so I can't send you my binaries 2020-01-10T18:16:55 #kisslinux <dylanaraps> The faked daemon 2020-01-10T18:16:56 #kisslinux <dylanaraps> All good 2020-01-10T18:17:00 #kisslinux <dylanaraps> Not going to run it :P 2020-01-10T18:17:06 #kisslinux <E5ten> ah 2020-01-10T18:17:24 #kisslinux <dylanaraps> sleep 0.1 is already non-portable is sh is the shell. 2020-01-10T18:17:42 #kisslinux <E5ten> yeah 2020-01-10T18:17:47 #kisslinux <dylanaraps> Same with 'local' usage 2020-01-10T18:17:58 #kisslinux <dylanaraps> This must depend on bash as /bin/sh huh 2020-01-10T18:18:18 #kisslinux <E5ten> local is super common and sleep is not a builtin, nor is sleep 0.1 non-portable 2020-01-10T18:18:27 #kisslinux <dylanaraps> getopt can be replaced with getopts no? 2020-01-10T18:18:52 #kisslinux <dylanaraps> POSIX doesn't define local though. 2020-01-10T18:19:04 #kisslinux <E5ten> getopts can't do longopts properly, but I already rewrote it to be manual parsing thing that correctly handles everything 2020-01-10T18:19:13 #kisslinux <E5ten> yeah POSIX isn't portable 2020-01-10T18:19:22 #kisslinux <dylanaraps> heh 2020-01-10T18:19:35 #kisslinux <dylanaraps> You have a point 2020-01-10T18:19:44 #kisslinux <E5ten> whoops 2020-01-10T18:19:51 #kisslinux <E5ten> meant to type local lol 2020-01-10T18:20:17 #kisslinux <E5ten> most common non-bash shells do still support local though which is why nobody notices that it's technically non-POSIX 2020-01-10T18:20:26 #kisslinux <E5ten> but my changes obviously include removing local 2020-01-10T18:20:41 #kisslinux <dylanaraps> yup 2020-01-10T18:20:57 #kisslinux <dylanaraps> Does 'faked' output its PID on execution? 2020-01-10T18:21:02 #kisslinux <dylanaraps> > KEY_PID=`eval $FAKED $FAKEDOPTS $PIPEIN` 2020-01-10T18:21:27 #kisslinux <E5ten> <FAKEROOTKEY>:<PID> I think 2020-01-10T18:21:34 #kisslinux <dylanaraps> Ah 2020-01-10T18:21:36 #kisslinux <dylanaraps> Gross 2020-01-10T18:22:24 #kisslinux <dylanaraps> All I can think of for the ls call is that it "wakes" the daemon in some way... 2020-01-10T18:22:55 #kisslinux <E5ten> the manual option parsing I made for it using just mandatory builtins and [ handles --<longopt>=<arg> --<longopt> <arg> -<shortopt><arg> -<shortopt> <arg> and -<multiple shortopts> 2020-01-10T18:23:17 #kisslinux <dylanaraps> > waits for the user process to 2020-01-10T18:23:17 #kisslinux <dylanaraps> die, and then send a SIGTERM signal to faked, causing 2020-01-10T18:23:17 #kisslinux <dylanaraps> Faked to clear the ipc message queues. 2020-01-10T18:23:18 #kisslinux <E5ten> no getopts because like I said it can't do longopts right 2020-01-10T18:23:23 #kisslinux <dylanaraps> yup 2020-01-10T18:23:37 #kisslinux <dylanaraps> Could also do 'while [ "$1" ]; do' and then have 'shift' below the 'case'. 2020-01-10T18:23:42 #kisslinux <dylanaraps> I do this in neofetch. 2020-01-10T18:23:55 #kisslinux <E5ten> that's what I did 2020-01-10T18:24:12 #kisslinux <dylanaraps> Ah you added support for -- 2020-01-10T18:24:15 #kisslinux <dylanaraps> I see now 2020-01-10T18:24:16 #kisslinux <E5ten> although I did [ "$#" -ne 0 ] because I handle my empty arguments correctly :p 2020-01-10T18:24:34 #kisslinux <E5ten> the one I sent is not my modified one 2020-01-10T18:24:38 #kisslinux <E5ten> that's a normal fakeroot 2020-01-10T18:24:53 #kisslinux <E5ten> it already supported -- 2020-01-10T18:24:54 #kisslinux <dylanaraps> OH 2020-01-10T18:25:02 #kisslinux <dylanaraps> > Will clear ipc message ques 2020-01-10T18:25:02 #kisslinux <dylanaraps> upon receipt of a SIGTERM. Will show debug output upon 2020-01-10T18:25:02 #kisslinux <dylanaraps> receipt of a SIGUSR1 (if started with -d debug option) 2020-01-10T18:25:06 #kisslinux <dylanaraps> Try with debug 2020-01-10T18:25:12 #kisslinux <dylanaraps> Should print what it's doing 2020-01-10T18:25:15 #kisslinux <dylanaraps> """should""" 2020-01-10T18:25:27 #kisslinux <dylanaraps> Great spelling of queues lol 2020-01-10T18:26:18 #kisslinux <dylanaraps> Oh nvm 2020-01-10T18:26:22 #kisslinux <dylanaraps> debug prints nothing of value 2020-01-10T18:26:30 #kisslinux <dylanaraps> "fakeroot: clearing up message queues and semaphores" 2020-01-10T18:26:40 #kisslinux <E5ten> it literally just runs ls -l /, then kill -s TERM <pid>, then sleep 0.1, then kill -s TERM <pid> 2020-01-10T18:26:52 #kisslinux <dylanaraps> OH 2020-01-10T18:26:58 #kisslinux <E5ten> I ran it with -x 2020-01-10T18:27:18 #kisslinux <dylanaraps> It runs ls -l to open up IPC so that the clean up code works even when the user hasn't run any commands. 2020-01-10T18:27:23 #kisslinux <dylanaraps> LOL 2020-01-10T18:27:52 #kisslinux <E5ten> so any external command or thing a subshell or something would work for that? 2020-01-10T18:28:28 #kisslinux <dylanaraps> Have you tried without the ls? 2020-01-10T18:28:29 #kisslinux <E5ten> thing in a* 2020-01-10T18:28:41 #kisslinux <dylanaraps> I could be wrong but I don't know what else it could be there for. 2020-01-10T18:28:51 #kisslinux <dylanaraps> It must be there for a reason right?... 2020-01-10T18:29:22 #kisslinux <dylanaraps> I have an idea 2020-01-10T18:29:51 #kisslinux <E5ten> seems to eb the same? 2020-01-10T18:30:03 #kisslinux <E5ten> to be* 2020-01-10T18:30:08 #kisslinux <dylanaraps> 1https://github.com/mackyle/fakeroot/commit/6a2dceedeb99d9bc1c781b302e933ecb1cda53bc 2020-01-10T18:30:11 #kisslinux <dylanaraps> https://github.com/mackyle/fakeroot/commit/6a2dceedeb99d9bc1c781b302e933ecb1cda53bc 2020-01-10T18:30:13 #kisslinux <dylanaraps> YES 2020-01-10T18:30:15 #kisslinux <dylanaraps> I was right 2020-01-10T18:31:25 #kisslinux <E5ten> so what should I put there instead? 2020-01-10T18:32:14 #kisslinux <dylanaraps> Uh 2020-01-10T18:32:26 #kisslinux <dylanaraps> Honestly, I wouldn't touch this code at all 2020-01-10T18:32:44 #kisslinux <E5ten> I don't get why the ls -l is needed at all 2020-01-10T18:32:49 #kisslinux <E5ten> shouldn't the while loop waiting to kill it be all that's needed? 2020-01-10T18:33:34 #kisslinux <dylanaraps> https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=446351 2020-01-10T18:33:40 #kisslinux <dylanaraps> > When fakeroot terminates, a signal is sent to the faked process which 2020-01-10T18:33:40 #kisslinux <dylanaraps> tells it to exit after processing all pending messages. Unfortunately 2020-01-10T18:33:40 #kisslinux <dylanaraps> some write messages may still be pending in the kernel buffers at this 2020-01-10T18:33:40 #kisslinux <dylanaraps> time, regardless of the communications style (TCP or SysV IPC). 2020-01-10T18:33:53 #kisslinux <dylanaraps> > meaning that the mknod command doesn't make it to the save file, 2020-01-10T18:33:54 #kisslinux <dylanaraps> in time, despite the sleep 1 (the latter being there to discriminate 2020-01-10T18:33:54 #kisslinux <dylanaraps> from another race condition, see bug #381016). 2020-01-10T18:33:59 #kisslinux <dylanaraps> > To fix that one only needs to issue a single synchronous command 2020-01-10T18:33:59 #kisslinux <dylanaraps> that talks to faked, such as "ls -l /", before sending the signal. 2020-01-10T18:34:20 #kisslinux <dylanaraps> Disgusting 2020-01-10T18:34:42 #kisslinux <E5ten> so what's a proper synchronous command that talks to faked that hopefully isn't external but if it is still isn't fucking ls -l / lol? 2020-01-10T18:34:50 #kisslinux <dylanaraps> It must be external. 2020-01-10T18:35:04 #kisslinux <dylanaraps> A similar bug that the same code fixes: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=381016 2020-01-10T18:35:49 #kisslinux <dylanaraps> I'd leave it as-is and write a comment with this info. 2020-01-10T18:35:59 #kisslinux <dylanaraps> Maybe also a message saying "DO NOT TOUCH" 2020-01-10T18:36:00 #kisslinux <E5ten> I refuse 2020-01-10T18:36:04 #kisslinux <E5ten> I'm not leaving ls -l / in it 2020-01-10T18:36:23 #kisslinux <dylanaraps> That comment should probably go at top of the README in bold big letters :^) 2020-01-10T18:36:48 #kisslinux <dylanaraps> It must be an external command and that command must be dynamically linked. 2020-01-10T18:36:57 #kisslinux <dylanaraps> It has to be something that fakeroot can intercept. 2020-01-10T18:37:13 #kisslinux <E5ten> jesus 2020-01-10T18:37:16 #kisslinux <dylanaraps> It also has to have a super short runtime. 2020-01-10T18:37:28 #kisslinux <E5ten> so it's probably gotta be something that uses a permission related function? 2020-01-10T18:37:32 #kisslinux <dylanaraps> Yup 2020-01-10T18:37:36 #kisslinux <dylanaraps> Hence ls -l / 2020-01-10T18:37:44 #kisslinux <dylanaraps> The '/' is important too. 2020-01-10T18:38:02 #kisslinux <dylanaraps> As is the '-l' as it shows ownership and permissions. 2020-01-10T18:38:16 #kisslinux <E5ten> the / is important because it always exists? 2020-01-10T18:38:27 #kisslinux <dylanaraps> Yes 2020-01-10T18:38:40 #kisslinux <dylanaraps> And it shows files owned by root (less important in this case) 2020-01-10T18:38:44 #kisslinux <dylanaraps> files/dirs* 2020-01-10T18:38:54 #kisslinux <dylanaraps> Absolutely foul though 2020-01-10T18:40:03 #kisslinux <E5ten> I don't think it would matter who the owner of the files is would it? wouldn't it just matter that the functions in question are getting wrapped by fakeroot? 2020-01-10T18:41:00 #kisslinux <dylanaraps> True 2020-01-10T18:41:04 #kisslinux <E5ten> wait lol couldn't this be solved just by using --foreground for faked and & 2020-01-10T18:41:09 #kisslinux <E5ten> and then you could use wait? 2020-01-10T18:41:24 #kisslinux <dylanaraps> I don't think so. This is mentioned on one of the commits. 2020-01-10T18:41:44 #kisslinux <E5ten> which one? 2020-01-10T18:41:47 #kisslinux <dylanaraps> hang on 2020-01-10T18:42:17 #kisslinux <E5ten> cuz if it was possible to find a way to do that, that'd definitely be the proper way to be doing this 2020-01-10T18:42:44 #kisslinux <dylanaraps> > Because 'faked' has detached itself from the terminal, it cannot be waited for 2020-01-10T18:42:44 #kisslinux <dylanaraps> with the sh builtin command 'wait'. 2020-01-10T18:42:48 #kisslinux <dylanaraps> https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=381016 2020-01-10T18:43:00 #kisslinux <E5ten> yeah, but it won't detach itself if you use its --foreground flag 2020-01-10T18:43:09 #kisslinux <E5ten> and then you can background it using the shell with & 2020-01-10T18:43:21 #kisslinux <E5ten> and now it'll be backgrounded but controlled by the shell not itself 2020-01-10T18:43:28 #kisslinux <dylanaraps> Try it 2020-01-10T18:44:26 #kisslinux <dylanaraps> iirc there's a remake of fakeroot 2020-01-10T18:44:39 #kisslinux <dylanaraps> One that works with static libs/programs. 2020-01-10T18:45:56 #kisslinux <dylanaraps> http://git.yoctoproject.org/cgit/cgit.cgi/pseudo/log/ 2020-01-10T18:45:57 #kisslinux <dylanaraps> Found it. 2020-01-10T18:46:19 #kisslinux <dylanaraps> README: http://git.yoctoproject.org/cgit/cgit.cgi/pseudo/tree/README 2020-01-10T18:46:27 #kisslinux <E5ten> huh I'll check that out 2020-01-10T18:47:32 #kisslinux <dylanaraps> Also: https://github.com/wrpseudo/pseudo/wiki/WhyNotFakeroot 2020-01-10T18:47:45 #kisslinux <dylanaraps> The GitHub is outdated but is the only source for this page. 2020-01-10T18:48:48 #kisslinux <dylanaraps> There's another fakeroot replacement actually. 2020-01-10T18:49:10 #kisslinux <E5ten> fakeroot-ng had problems I didn't understand when I tried it 2020-01-10T18:49:36 #kisslinux <dylanaraps> Fedora use pseudo fwiw 2020-01-10T18:55:32 #kisslinux <dylanaraps> Yeah, it was fakeroot-ng 2020-01-10T18:58:13 #kisslinux <E5ten> Is there any hackery you know of to save a things output without running it in a subshell, without using a tmpfile and in POSIX sh not bash 2020-01-10T18:58:41 #kisslinux <dylanaraps> oof 2020-01-10T18:59:02 #kisslinux <dylanaraps> Ah 2020-01-10T18:59:05 #kisslinux <dylanaraps> Got an idea 2020-01-10T19:07:45 #kisslinux <E5ten> What is it lol? 2020-01-10T19:11:32 #kisslinux <dylanaraps> Working on it 2020-01-10T19:18:12 #kisslinux <adamantium> dylanaraps: nice X script. Why -keeptty 2020-01-10T19:18:42 #kisslinux <adamantium> dylanaraps: I was I looking in to it a while back, trying to figure out why -keeptty is so common. and couldn't find a good reason to use it. the Xorg --help says it's only useful for debuggiing 2020-01-10T19:18:56 #kisslinux <adamantium> But i'm also not sure what -keeptty does 2020-01-10T19:18:57 #kisslinux <dylanaraps> It keeps Xorg on vt1 2020-01-10T19:19:05 #kisslinux <dylanaraps> I only have vt1 :^) 2020-01-10T19:19:11 #kisslinux <dylanaraps> I don't use gettys 2020-01-10T19:19:21 #kisslinux <adamantium> oh, i explicly specify like vt1 v 2020-01-10T19:20:57 #kisslinux <adamantium> i love that people here are obsessed with choosing the simplest/lightest option to the point of using . vs . read -r and export 2020-01-10T19:21:11 #kisslinux <adamantium> even when it's not a noticeable performance boost 2020-01-10T19:21:12 #kisslinux <adamantium> lol 2020-01-10T19:22:37 #kisslinux <adamantium> dylanaraps: on your configuration, what happens without -keeptty 2020-01-10T19:22:43 #kisslinux <dylanaraps> Uh 2020-01-10T19:22:48 #kisslinux <dylanaraps> Will check in a few mins 2020-01-10T19:22:54 #kisslinux <adamantium> im very curious 2020-01-10T19:22:56 #kisslinux <adamantium> thanks 2020-01-10T19:23:04 #kisslinux <dylanaraps> What do you mean about above? 2020-01-10T19:23:12 #kisslinux <dylanaraps> read -r, export, etc 2020-01-10T19:23:33 #kisslinux <adamantium> your comment in the PR about how not sourcing the file with '.' could be faster 2020-01-10T19:23:54 #kisslinux <adamantium> but I mean also that it's cool to be with a group of people who are even more obsessed with correctness and speed than myself 2020-01-10T19:24:59 #kisslinux <dylanaraps> Ah 2020-01-10T19:26:00 #kisslinux <E5ten> ok I don't think the output thing is possible, so what's the least terrible way to store a word in a file for a second so something can read it 2020-01-10T19:31:48 #kisslinux <dylanaraps> E5ten: I have it working lmao 2020-01-10T19:31:53 #kisslinux <dylanaraps> I wouldn't use it though 2020-01-10T19:32:48 #kisslinux <E5ten> how'd you do it lol 2020-01-10T19:33:02 #kisslinux <dylanaraps> It's so gross 2020-01-10T19:33:03 #kisslinux <dylanaraps> Hang on 2020-01-10T19:36:06 #kisslinux <dylanaraps> https://termbin.com/dkdub 2020-01-10T19:36:08 #kisslinux <dylanaraps> BEHOLD 2020-01-10T19:36:29 #kisslinux <dylanaraps> It's beautiful 2020-01-10T19:37:55 #kisslinux <E5ten> oh my god lol 2020-01-10T19:38:11 #kisslinux <dzove855> holy 2020-01-10T19:38:28 #kisslinux <dylanaraps> Totally safe 2020-01-10T19:38:30 #kisslinux <dylanaraps> I swear 2020-01-10T19:38:40 #kisslinux <dylanaraps> You won't see ls' output in one of your random terminals. 2020-01-10T19:39:02 #kisslinux <dzove855> It's a feature! 2020-01-10T19:40:23 #kisslinux <dylanaraps> In reality, as I limit this to fds > 10 I'm probably lying when I say it'll output in a random terminal. 2020-01-10T19:41:11 #kisslinux <dylanaraps> brb testing my xinit without keeptty 2020-01-10T19:41:26 #kisslinux <dylanaraps> LOL 2020-01-10T19:41:34 #kisslinux <dylanaraps> hang on 2020-01-10T19:41:35 #kisslinux <dylanaraps> OK 2020-01-10T19:41:43 #kisslinux <dylanaraps> Vim opens an fd for each open file 2020-01-10T19:41:54 #kisslinux <dylanaraps> My .profile is now the output of ls / 2020-01-10T19:41:56 #kisslinux <dylanaraps> heh 2020-01-10T19:42:06 #kisslinux <adamantium> lol 2020-01-10T19:42:09 #kisslinux <E5ten> that was not the smartest move lmao 2020-01-10T19:42:09 #kisslinux <dzove855> haha 2020-01-10T19:42:10 #kisslinux <adamantium> this is nutty in here 2020-01-10T19:43:34 #kisslinux <dylanaraps> Just run it as sudo as use pid 1's file descriptors. 2020-01-10T19:44:32 #kisslinux <dylanaraps> brb 2020-01-10T19:45:48 #kisslinux <dylanaraps> OK 2020-01-10T19:45:53 #kisslinux <dylanaraps> Works without keeptty. 2020-01-10T19:46:52 #kisslinux <adamantium> dylanaraps: what terminolgy do you prefer log "Running rc.d scripts..."; or log "Running rc.d hooks..."; 2020-01-10T19:47:31 #kisslinux <adamantium> dylanaraps: I suspected it would. 2020-01-10T19:48:08 #kisslinux <adamantium> "-keeptty don't detach controlling tty (for debugging only)" 2020-01-10T19:48:39 #kisslinux <dylanaraps> Look at your PR btw. I left a comment. 2020-01-10T19:48:46 #kisslinux <dylanaraps> Either one is fine. 2020-01-10T19:48:52 #kisslinux <dylanaraps> I'd lean towards hooks though. 2020-01-10T19:49:17 #kisslinux <adamantium> Oh yeah, heh. 2020-01-10T19:49:23 #kisslinux <adamantium> I like hooks too! 2020-01-10T19:49:26 #kisslinux <dylanaraps> I'll be back on in an hour or so. 2020-01-10T19:49:27 #kisslinux <adamantium> I wanted to ask first though. 2020-01-10T19:49:28 #kisslinux <adamantium> okay 2020-01-10T19:49:30 #kisslinux <dylanaraps> Drinky time 2020-01-10T19:49:42 #kisslinux <adamantium> o/ 2020-01-10T19:52:13 #kisslinux <dylanaraps> E5ten: I just figured out how to make that script do 0 damage. 2020-01-10T19:52:30 #kisslinux <dylanaraps> Just read from the hijacked fd first. Save its contents. 2020-01-10T19:52:37 #kisslinux <dylanaraps> Then put it back when you're done. 2020-01-10T19:52:38 #kisslinux <dylanaraps> 10/10 2020-01-10T19:52:40 #kisslinux <dylanaraps> 100% safe 2020-01-10T19:53:31 #kisslinux <dylanaraps> ok 2020-01-10T19:53:34 #kisslinux <dylanaraps> Now I'm going. 2020-01-10T21:32:37 #kisslinux <dylanaraps> Back 2020-01-10T21:34:29 #kisslinux <dylanaraps> https://git.savannah.gnu.org/cgit/grub.git/commit/?id=365e0cc3e7e44151c14dd29514c2f870b49f9755 2020-01-10T21:34:30 #kisslinux <dylanaraps> Nice. 2020-01-10T21:34:47 #kisslinux <dylanaraps> This should make initramfs-less decryption easier. 2020-01-10T21:41:46 #kisslinux <Aarg[m]> dylanaraps, I think it would be better if kiss prompted on stderr 2020-01-10T21:45:17 #kisslinux <dylanaraps> Care to elaborate? 2020-01-10T21:45:27 #kisslinux <dylanaraps> The message should be printed on stderr? or? 2020-01-10T21:46:54 #kisslinux <Aarg[m]> Yeah 2020-01-10T21:47:13 #kisslinux <Aarg[m]> You never see it when redirecting the output 2020-01-10T21:47:18 #kisslinux <dylanaraps> Ah 2020-01-10T21:47:21 #kisslinux <dylanaraps> Noted 2020-01-10T21:47:37 #kisslinux <Aarg[m]> That's why prompts like this are always on stderr ;) 2020-01-10T21:48:18 #kisslinux <dylanaraps> I'll fix it in a sec. 2020-01-10T21:48:55 #kisslinux <Aarg[m]> Nice 2020-01-10T21:50:00 #kisslinux <Aarg[m]> Personally I would go even further and print log() to both, stdout and err when stdout is not a terminal 2020-01-10T21:50:08 #kisslinux <Aarg[m]> But not sure if that's a good idea 2020-01-10T21:51:17 #kisslinux <Aarg[m]> However, you can then redirect to a log file and only see kiss' output on your terminal without all the other noise 2020-01-10T21:53:19 #kisslinux <dylanaraps> Noted 2020-01-10T22:03:02 #kisslinux <E5ten> well my modified fakeroot now uses a tmpfile sadly, but I still think it's less awful than the previous thing cuz it properly manages the faked process and doesn't do fucking ls -l / 2020-01-10T22:04:13 #kisslinux <dylanaraps> What's the temp file used for? 2020-01-10T22:04:31 #kisslinux <E5ten> to save the fakerootkey it outputs 2020-01-10T22:04:38 #kisslinux <E5ten> cuz I can't run it in a subshell cuz I need to background and manage it 2020-01-10T22:05:27 #kisslinux <dylanaraps> Ah 2020-01-10T22:05:33 #kisslinux <dylanaraps> There *might* be a way around this. 2020-01-10T22:05:59 #kisslinux <E5ten> oh do tell if you find one lol I'd very much like to not have to use a temp file 2020-01-10T22:06:15 #kisslinux <dylanaraps> Can I see the code? 2020-01-10T22:06:33 #kisslinux <E5ten> http://ix.io/2749 2020-01-10T22:07:22 #kisslinux <E5ten> I changed the @MACOSX_FALSE@ thing so that it's only on one line, and if it's not macos it gets replaced with nothing and if it is macos that line is deleted 2020-01-10T22:07:39 #kisslinux <dylanaraps> Could you not do: key=$(_faked) ? 2020-01-10T22:08:06 #kisslinux <E5ten> no because then the backgrounded faked is run in the subshell 2020-01-10T22:08:22 #kisslinux <dylanaraps> You can't capture output through '$()'? 2020-01-10T22:08:30 #kisslinux <E5ten> that's a subshell 2020-01-10T22:09:10 #kisslinux <dylanaraps> Oh 2020-01-10T22:09:13 #kisslinux <dylanaraps> Gotcha 2020-01-10T22:09:43 #kisslinux <dylanaraps> You can omit quotes in variable assignments if there's no space fyi 2020-01-10T22:09:52 #kisslinux <dylanaraps> PID=$! is 100% fine. 2020-01-10T22:09:59 #kisslinux <E5ten> I know 2020-01-10T22:10:03 #kisslinux <E5ten> I just like putting quotes on everything 2020-01-10T22:10:10 #kisslinux <dylanaraps> I used to be like that 2020-01-10T22:10:19 #kisslinux <dylanaraps> lol 2020-01-10T22:10:31 #kisslinux <E5ten> you don't need them even if there is a space 2020-01-10T22:10:40 #kisslinux <E5ten> var=$varwithspace is also totally safe 2020-01-10T22:10:46 #kisslinux <dylanaraps> No no 2020-01-10T22:10:54 #kisslinux <dylanaraps> I mean: var='hello world' 2020-01-10T22:11:01 #kisslinux <E5ten> oh 2020-01-10T22:11:04 #kisslinux <dylanaraps> Unless you do var=hello world 2020-01-10T22:12:15 #kisslinux <Aarg[m]> You never have to quote a variable when using case, but I guess you know that 2020-01-10T22:12:21 #kisslinux <E5ten> yeah 2020-01-10T22:12:43 #kisslinux <dylanaraps> yup 2020-01-10T22:13:10 #kisslinux <Aarg[m]> k, just making sure 2020-01-10T22:13:42 #kisslinux <dylanaraps> You could possibly edit a /proc value at runtime to store the value. 2020-01-10T22:14:07 #kisslinux <dylanaraps> I know some database software does this to obfuscate passwords sent on the command-line. 2020-01-10T22:14:57 #kisslinux <E5ten> maybe we can think of a better way for faked to expose that information, cuz if I'm modifying fakeroot.in obviously I can modify faked to support a change I want to make 2020-01-10T22:15:20 #kisslinux <dylanaraps> Yup 2020-01-10T22:15:46 #kisslinux <dylanaraps> It's 2020 so write use dbus and implement a client/server model 2020-01-10T22:15:52 #kisslinux <dylanaraps> s/write// 2020-01-10T22:16:12 #kisslinux <E5ten> of course 2020-01-10T22:16:16 #kisslinux <dylanaraps> I'm a little drunk. I apologize. 2020-01-10T22:16:16 #kisslinux <E5ten> the only solution 2020-01-10T22:17:15 #kisslinux <E5ten> nw lmao 2020-01-10T22:18:05 #kisslinux <dylanaraps> We basically need the simplest IPC possible. 2020-01-10T22:18:14 #kisslinux <dylanaraps> To send a single value. 2020-01-10T22:18:34 #kisslinux <dylanaraps> Can we do better than a file? 2020-01-10T22:19:16 #kisslinux <E5ten> hmmmmmmmmm 2020-01-10T22:20:29 #kisslinux <dylanaraps> ooo 2020-01-10T22:20:45 #kisslinux <dylanaraps> You can write to /proc/PID_OF_FAKED/comm fyi 2020-01-10T22:21:09 #kisslinux <dylanaraps> This is super simple. 2020-01-10T22:21:31 #kisslinux <dylanaraps> Just write to faked's PID's comm file. 2020-01-10T22:21:37 #kisslinux <dylanaraps> Then read it in the script. 2020-01-10T22:22:20 #kisslinux <E5ten> '/proc/' isn't POSIX is it? 2020-01-10T22:22:30 #kisslinux <dylanaraps> Does fakeroot run outside of linux? 2020-01-10T22:22:39 #kisslinux <adamantium> nice job dylanaraps !!!!!! 2020-01-10T22:22:45 #kisslinux <E5ten> I hadn't really thought of it 2020-01-10T22:22:47 #kisslinux <dylanaraps> Ah, it does 2020-01-10T22:22:50 #kisslinux <adamantium> posix killall5 is a nice touch 2020-01-10T22:22:54 #kisslinux <dylanaraps> https://www.freshports.org/security/fakeroot 2020-01-10T22:23:14 #kisslinux <dylanaraps> adamantium: Thanks 2020-01-10T22:23:32 #kisslinux <dylanaraps> Only FreeBSD has fakeroot in ports. 2020-01-10T22:23:38 #kisslinux <dylanaraps> No other BSD provides it. 2020-01-10T22:24:18 #kisslinux <E5ten> damn so we're back to a file then I guess? 2020-01-10T22:25:11 #kisslinux <dylanaraps> You could use /proc if available and fallback to a file. 2020-01-10T22:25:16 #kisslinux <dylanaraps> Though this equates to more code. 2020-01-10T22:25:49 #kisslinux <dylanaraps> You can change a processes command-line from C portably iirc. 2020-01-10T22:25:53 #kisslinux <E5ten> if I'm falling back to a file anyway I'd rather just keep the crap I have in the script rn than modify the C 2020-01-10T22:26:22 #kisslinux <dylanaraps> Yup 2020-01-10T22:26:33 #kisslinux <dylanaraps> Where are you writing the file? /tmp? 2020-01-10T22:26:35 #kisslinux <E5ten> I didn't even actually think about what it'd be like if you change the process name, that's definitely not a solution lol 2020-01-10T22:26:41 #kisslinux <E5ten> ${TMPDIR:-/tmp} 2020-01-10T22:26:57 #kisslinux <dylanaraps> Does fakeroot use some kind of tmp directory from C? 2020-01-10T22:27:03 #kisslinux <dylanaraps> It keeps a database no? 2020-01-10T22:27:28 #kisslinux <dylanaraps> E5ten: Not the process name. Just appending to the arguments list. 2020-01-10T22:27:51 #kisslinux <E5ten> ohhh 2020-01-10T22:28:08 #kisslinux <E5ten> then you're getting into grepping ps and at that point I'm out 2020-01-10T22:28:36 #kisslinux <dylanaraps> no no 2020-01-10T22:28:46 #kisslinux <dylanaraps> Access through /proc using a simple read 2020-01-10T22:29:06 #kisslinux <E5ten> but /proc isn't POSIX so we're back to having a fallback and it's still not worth it 2020-01-10T22:29:08 #kisslinux <dylanaraps> On FreeBSD this would be moot 2020-01-10T22:29:10 #kisslinux <dylanaraps> Yup 2020-01-10T22:29:17 #kisslinux <dylanaraps> Files are the simplest IPC 2020-01-10T22:29:29 #kisslinux <dylanaraps> Look into where faked stores its database in this case. 2020-01-10T22:29:36 #kisslinux <dylanaraps> Use the same dir if possible. 2020-01-10T22:29:39 #kisslinux <E5ten> why can't it just be possible to catch output without subshells ;( 2020-01-10T22:29:51 #kisslinux <dylanaraps> Why don't you ditch the key and use pid in its place? 2020-01-10T22:30:15 #kisslinux <E5ten> it's the library and faked that use key though 2020-01-10T22:30:24 #kisslinux <E5ten> the script just gets it for them 2020-01-10T22:30:41 #kisslinux <E5ten> oh you mean modify them to use PID instead? 2020-01-10T22:31:05 #kisslinux <dylanaraps> Yup 2020-01-10T22:31:07 #kisslinux <E5ten> I bet that'd be a problem with like PID namespaces or something 2020-01-10T22:31:18 #kisslinux <dylanaraps> The PID should be unique enough. 2020-01-10T22:32:15 #kisslinux <dylanaraps> Could also make it more unique from the shell by mixing in the time of invocation or some other data. 2020-01-10T22:36:24 #kisslinux <dylanaraps> brb 2020-01-10T22:42:08 #kisslinux <dylanaraps> The new init changes work beautifully. 2020-01-10T22:44:07 #kisslinux <adamantium> dylanaraps: put up a screenshot, gave kiss linux a shoutout. Hit 90 stars in like a day 2020-01-10T22:44:10 #kisslinux <adamantium> https://github.com/a-schaefers/systemE 2020-01-10T22:44:26 #kisslinux <adamantium> anyone new come in here asking about kiss stuff? 2020-01-10T22:45:04 #kisslinux <E5ten> I think that the db can be stored as either an inode or a file from what I can see in the configure script, and the default seems to not be file 2020-01-10T22:45:29 #kisslinux <dylanaraps> adamantium: Nice 2020-01-10T22:45:33 #kisslinux <dylanaraps> I read the comments 2020-01-10T22:46:09 #kisslinux <dylanaraps> E5ten: Where does it store it? 2020-01-10T22:47:20 #kisslinux <E5ten> not sure yet 2020-01-10T22:50:04 #kisslinux <E5ten> oh shit 2020-01-10T22:50:33 #kisslinux <E5ten> you can specify a key to it but only if it's been compiled to use sysv IPC, which means I can't make use of that in fakeroot cuz obviously fakeroot needs to work with all configurations of faked :( 2020-01-10T22:51:01 #kisslinux <E5ten> so were --key available to faked compiled to use sysv IPC or TCP I could use that to use PID as key 2020-01-10T22:51:27 #kisslinux <E5ten> or just generate a key the same way faked does in shell and then use that generated key lol 2020-01-10T22:56:34 #kisslinux <dylanaraps> Hm 2020-01-10T23:01:02 #kisslinux <E5ten> ok so tcp faked's keys are ports, and can be passed using --port 2020-01-10T23:01:29 #kisslinux <E5ten> sysvipc faked's keys are keys, not sure what they actually are yet, and they can be passed using --key 2020-01-10T23:02:30 #kisslinux <E5ten> at this point I feel like I'm already in too deep for this to be worth the effort instead of just keeping my temp file crap lol 2020-01-10T23:02:48 #kisslinux <adamantium> dylanaraps: i'm going to test this for job control issues, brb 2020-01-10T23:03:02 #kisslinux <adamantium> it's a weird issue i've run into when forking gettys 2020-01-10T23:03:12 #kisslinux <adamantium> it'll screw up a system badly if we get it wrong 2020-01-10T23:03:24 #kisslinux <dylanaraps> ok 2020-01-10T23:04:55 #kisslinux <adamantium> hmm 2020-01-10T23:05:02 #kisslinux <adamantium> dylanaraps: can you send me a tarball of the new baseinit 2020-01-10T23:06:04 #kisslinux <dylanaraps> https://github.com/kisslinux/init/archive/31810d07444637cf3a3158793dd9ed649fd80d0a.tar.gz 2020-01-10T23:07:19 #kisslinux <adamantium> grrr 2020-01-10T23:07:20 #kisslinux <adamantium> no build file 2020-01-10T23:08:05 #kisslinux <adamantium> I shall move it in to place 2020-01-10T23:08:05 #kisslinux <adamantium> lol 2020-01-10T23:08:51 #kisslinux <dylanaraps> Just make your own package 2020-01-10T23:08:59 #kisslinux <dylanaraps> Takes no time 2020-01-10T23:09:21 #kisslinux <dylanaraps> cp -r /var/db/kiss/repo/core/baseinit /path/to/myrepo 2020-01-10T23:09:35 #kisslinux <dylanaraps> Then update sources/version and kiss c baseinit 2020-01-10T23:10:31 #kisslinux <adamantium> I was doing that 2020-01-10T23:10:35 #kisslinux <adamantium> I thought our work was done 2020-01-10T23:10:39 #kisslinux <adamantium> So I'm not setup for it anymore 2020-01-10T23:10:41 #kisslinux <adamantium> It's my fault 2020-01-10T23:10:46 #kisslinux <adamantium> I got it set now 2020-01-10T23:10:47 #kisslinux <adamantium> BRB 2020-01-10T23:10:50 #kisslinux <E5ten> so aside from the temp file thing anything else I should change still in the fakeroot script? 2020-01-10T23:11:05 #kisslinux <dylanaraps> I'll take a look in a bit. 2020-01-10T23:12:54 #kisslinux <adamanti`> dylanaraps: line 55 rc.boot export something something : bad variable name 2020-01-10T23:13:00 #kisslinux <adamanti`> my boot was hung 2020-01-10T23:14:24 #kisslinux <dylanaraps> Contents of your rc.conf? 2020-01-10T23:14:52 #kisslinux <adamanti`> looks like pars_conf did it 2020-01-10T23:15:10 #kisslinux <adamanti`> oh 2020-01-10T23:15:12 #kisslinux <adamanti`> LOL 2020-01-10T23:15:19 #kisslinux <dylanaraps> Yup 2020-01-10T23:15:23 #kisslinux <adamanti`> https://termbin.com/5dxm8 2020-01-10T23:15:35 #kisslinux <adamanti`> this is why we test things 2020-01-10T23:15:42 #kisslinux <dylanaraps> Ah 2020-01-10T23:15:43 #kisslinux <adamanti`> I still think it's good to leave them commented 2020-01-10T23:15:49 #kisslinux <adamanti`> This is a safety feature. 2020-01-10T23:15:55 #kisslinux <dylanaraps> I'll fix it. 2020-01-10T23:15:58 #kisslinux <adamanti`> no 2020-01-10T23:16:02 #kisslinux <adamanti`> They must stay commented 2020-01-10T23:16:08 #kisslinux <adamanti`> by default 2020-01-10T23:16:31 #kisslinux <adamanti`> The user needs to switch grub to use /usr/bin/sinit and uncomment /etc/rc.conf to enable this stuff 2020-01-10T23:17:10 #kisslinux <adamanti`> Otherwise, they'll end up with a frankenstein system with like 12 gettys fighting for 6 ttys and two service supervisors and so forth 2020-01-10T23:18:57 #kisslinux <adamantium> dylanaraps: You're right though. It needs to still boot if rc.conf is empty 2020-01-10T23:19:22 #kisslinux <adamantium> I was thinking about the conf file, but the pars function needs to not make init bail if rc.conf is empty 2020-01-10T23:51:10 #kisslinux <E5ten> I wonder if faked could be rewritten and the sysvipc and tcp methods both removed and replaced with POSIX IPC