jaeger changed the topic of #crux to: CRUX 3.6 | Homepage: https://crux.nu/ | Ports: https://crux.nu/portdb/ https://crux.ninja/portdb/ | Logs: https://libera.irclog.whitequark.org/crux/
farkuhar has quit [*.net *.split]
jaeger has quit [*.net *.split]
rmull_ has quit [*.net *.split]
plow has quit [*.net *.split]
jaeger has joined #crux
plow has joined #crux
rmull has joined #crux
ocb has quit [Ping timeout: 268 seconds]
ocb has joined #crux
_moth_ has quit [Quit: _moth_]
plow has quit [Changing host]
plow has joined #crux
ppetrov^ has joined #crux
ty3r0x has joined #crux
SiFuh_ has quit [Quit: leaving]
guido_rokepo has joined #crux
SiFuh has joined #crux
Romster has quit [Quit: Geeks shall inherit properties and methods of object earth.]
Romster has joined #crux
ppetrov^ has quit [Quit: Leaving]
ty3r0x_ has joined #crux
ty3r0x has quit [Ping timeout: 240 seconds]
todrik has joined #crux
_moth_ has joined #crux
farkuhar has joined #crux
ocb has quit [Quit: leaving]
ocb has joined #crux
ocb has quit [Quit: leaving]
guido_rokepo has quit [Quit: guido_rokepo]
<farkuhar> whoever is able to build graphviz successfully with ghostscript installed, can you report whether ~/.config/paperspecs exists on your system? Its absence seems to trigger the segmentation fault that ppetrov^ observed, according to the stack trace: https://dpaste.com/FGF8WP3GH
ocb has joined #crux
ardo has quit [Read error: Connection reset by peer]
ardo has joined #crux
ardo has quit [Quit: ZNC 1.8.2 - https://znc.in]
ardo has joined #crux
ardo has quit [Ping timeout: 268 seconds]
ardo has joined #crux
ardo has quit [Read error: Connection reset by peer]
_moth_ has quit [Remote host closed the connection]
_moth_ has joined #crux
ardo has joined #crux
ardo has quit [Read error: Connection reset by peer]
ardo has joined #crux
ppetrov^ has joined #crux
todrik has quit [Ping timeout: 252 seconds]
<SiFuh> farkuhar: Mine segfaults during compilation. Heh. I will sysup first
todrik has joined #crux
z812 has quit [Quit: Bye!]
z812 has joined #crux
_moth_ has quit [Remote host closed the connection]
_moth_ has joined #crux
ppetrov^ has quit [Quit: Leaving]
ty3r0x_ has quit [Ping timeout: 240 seconds]
ty3r0x_ has joined #crux
<SiFuh> farkuhar: rm -f cdt.3.pdf; pdffile=cdt.3.pdf; psfile=${pdffile%pdf}ps; \ groff -Tps -man cdt.3 > $psfile || { rm -f $psfile; exit 1; }; \ ps2pdf $psfile && rm -f $psfile || { rm -f $psfile; exit 1; } Segmentation fault
<SiFuh> That's what I got when building :-P
<farkuhar> SiFuh: thanks for the report. I tested again on two different sysupped machines. The first one gave the segfault error shown in my previous stack trace. The second machine gave a "free(): invalid size" error, as shown here: http://sprunge.us/dibP4w
<SiFuh> I am rebuilding ghostscript and then test again
<SiFuh> Then I will rebuild groff and test again
<farkuhar> The problematic files, /etc/paperspecs and /etc/papersize, are actually owned by opt/paper, not ghostscript or groff. In a container build they wouldn't be present (since opt/paper is not a hard dependency), and so the compilation errors would go undetected.
<SiFuh> Yeah I saw the port was libpaper or paper or something
<SiFuh> rebuilding ghostscript and groff didn't work
ty3r0x has joined #crux
ty3r0x_ has quit [Ping timeout: 272 seconds]
<SiFuh> polkit the disease won't build says mozjs-91 is missing. Why is polkit even installed? I don't use anything special. It is just a flight simulator
<farkuhar> SiFuh: I just got a successful build of graphviz. It was necessary to uninstall opt/paper and then rebuild ghostscript. If opt/paper is installed when ghostscript is built, then the ps2pdf program will look for the problematic files (paperspecs and papersize), which causes the segfault.
<SiFuh> Oh okay, I will try that
<farkuhar> Besides parsing the output of `ldd /usr/bin/gs`, there isn't a clean way to tell whether opt/paper was present when ghostscript was built. That's why my first fix (http://sprunge.us/GjK6Zw ) took the blunt approach of disabling PDF manpages regardless of how ghostscript was built.
<SiFuh> This is an issue. Building against a pure system may work, but once other packages are installed it can fail
<SiFuh> The only two solutions are 1) provide binaries or 2) compile the entire ports tree of core/opt/contrib/compat completely for each port update.
<SiFuh> I can't think of any other solutions except change distros :-P
<farkuhar> At some point it should no longer be a CRUX problem, but something that upstream needs to look into. But I don't see any report of ghostscript problems on the graphviz gitlab site.
<SiFuh> I see it as a CRUX problem to be honest
<SiFuh> But not in a bad way
ty3r0x has quit [Ping timeout: 268 seconds]
<SiFuh> But you are correct.
_moth_ has quit [Ping timeout: 264 seconds]
<farkuhar> If upstream graphviz is providing the file cdt.3 and a Makefile that tries to process it with groff and ps2pdf, then they ought to test those commands across a variety of possible configurations of groff and ps2pdf. We as distro maintainers shouldn't have to patch the Makefile because their testing was incomplete.
<SiFuh> Yes, and you are right. But it is also a CRUX problem since pull non-maintainer deps is a possibility but not really looked into much
<SiFuh> It is mostly left to the user to figure it out
<SiFuh> farkuhar: it worked here
<farkuhar> SiFuh: polkit also suffered from insufficient testing. beerman neglected to check whether it would build against the version of mozjs in our ports tree, not realizing that polkit requires mozjs91. In the meantime you can adapt beerman's fork of opt/mozjs78 and create your own port for mozjs91.
<farkuhar> Otherwise, if nothing else on your system depends on mozjs78, just remove mozjs78 and install duktape instead. That should allow polkit to build cleanly.
todrik has quit [Remote host closed the connection]
todrik has joined #crux
ocb has quit [Ping timeout: 268 seconds]
ocb has joined #crux
ty3r0x has joined #crux
mnkydeth has joined #crux
<SiFuh> I will remove it as I don't know how it even got installed
<SiFuh> Anyways, I will put everything on hold for now. Since to many things are erroring out do to certain port updates breaking everything
<SiFuh> Might wait until 3.7 is released
<SiFuh> What I should do is make my own CRUX SDR version
<SiFuh> Keep the ports a little bit out-of-date to maintain stability and build everything against itself
ty3r0x has quit [Ping timeout: 268 seconds]
ppetrov^ has joined #crux
ivandi has quit [Quit: WeeChat 3.6]
ivandi has joined #crux
ivandi has quit [Client Quit]
ivandi has joined #crux
mnkydeth has quit [Quit: mnkydeth]
ty3r0x has joined #crux
<cruxbot> [opt.git/3.6]: mupdf: add optional dependencies
<cruxbot> [opt.git/3.6]: polkit: remove old patches, fix mozjs detection, whitespaces
ppetrov^ has quit [Quit: Leaving]
<cruxbot> [opt.git/3.6]: spaces are way better than tabs
ppetrov^ has joined #crux
<SiFuh> HAHAHAHA
<SiFuh> 100% agree, I don't like to use tabs. So annoying.
<ppetrov^> please, remove qt6-location as a dependency for qt6-webengine
<ppetrov^> or add a port :P
<SiFuh> ppetrov^: #crux-devel
<ppetrov^> i am too shy to go there...
<SiFuh> Put on some make up
<ppetrov^> i'm a guy, man...
<SiFuh> So?
<SiFuh> HAHAHA
<ppetrov^> although it does not matter
<ppetrov^> right, right...
<SiFuh> I was kidding but I laughed so loud I the dogs a block away are barking
<ppetrov^> heh..
<SiFuh> Just join, dump your issue, and leave. Someone will look at it, look into it and maybe something happens
<ppetrov^> they may find me here...
<SiFuh> You know #crux and #crux-devel are basically the same right? Just the off topic gossip is spoken here in #crux
<SiFuh> Although some members have moved to #crux-devel only because they are hard working and #crux gossip can be distracting.
<SiFuh> Apart from that, it is the same. Just less distracting
<ppetrov^> i am kidding
<ppetrov^> i am a bit drunk
<SiFuh> I drink 24/7
<SiFuh> Even at work
<ppetrov^> are you complaining or bragging about it?
<SiFuh> Enjoying it
<ppetrov^> that's the spirit
<SiFuh> Today I tested the new Fuji C-50 burner. Cut my distillation time from 1 hour and 40 minutes to 30 minutes.
<ppetrov^> and by spirit i don't mean hard drink
<SiFuh> The heat up time, not the complete distillation (Sorry)
<ppetrov^> so, you brew your own?
<SiFuh> Yeah, I make 90 percent I guess of all my alcohol.
<SiFuh> I sell maybe 99 percent of the vodka.
<ppetrov^> this legal there?
<SiFuh> I don't seel my beer or Asian alcohol. That is for me and the family/friends
<SiFuh> Hell no
<SiFuh> Muslim country
<ppetrov^> :D
<SiFuh> Australia and NZ it is legal.
<SiFuh> But in Australia if the distiller is over 5 litres you should apply for a licence from the tax department.
<ppetrov^> interesting
<SiFuh> Beer and wine is legal in Australia
<SiFuh> NZ all alcohol manufacturing is legal regardless of size
<ppetrov^> sounds great
<SiFuh> The world is changing
<SiFuh> SO for now, I am just a moonshiner and my friend is a bootlegger :-)
<ppetrov^> heh
<SiFuh> I have 320 litres to do. So will take about 14 days
<ppetrov^> well, i don't know much about distillation but sounds impressive
<SiFuh> I tested the new distiller a couple of days ago and it failed. Blew up the tempered glass stove, then the next day popped the distiller. So now I am back on the basic pot still to get the job done before the mash becomes vinegar
<SiFuh> Heat the shit, catch the shit, cool the shit, collect the shit.
<ppetrov^> ye, this i know :p
<SiFuh> That is it. But when you start going fancy, you can make a home made bomb.
<SiFuh> The best moonshine (in my opinion) is made in a pot still using recycled mash. I don't care copper or stainless steel.
<SiFuh> A thumpkeg is a bonus for flavour. But it can create back pressure
<SiFuh> Then there is a lot of other stuff but the topic will become huge explaining it all
<ppetrov^> maybe make another channel dedicated to it? like #crux-distil
<SiFuh> Actually I would if there was other moonshiners here
<SiFuh> were*
<ppetrov^> maybe they are hiding?
<SiFuh> But living in Malaysia, knowing this, from when I was young. My neighbour was a brewer. I use to mess around helping him and not really caring, just doing what he said, would end up being my lively-hood. That was in the late 80's early 90's. Who would have thought?
<SiFuh> Probably better to create or even check if a #distiller channel exists
<ppetrov^> i'll join
<SiFuh> #moonshiners and distillers doesn't exist
<ppetrov^> is it complicated to create a channel?
<SiFuh> No
<SiFuh> But I need to check plurals too
<SiFuh> I haven't searched for channels in decades. Wondering how
<SiFuh> LIST This command could not be completed because it has been used recently, and is rate-limited.2
<ppetrov^> i have no idea what this means
<SiFuh> #homebrewtalk.com and ##homebrew exist, both empty
<ppetrov^> go get 'em
<ppetrov^> does it cost anything?
<SiFuh> No
<SiFuh> I will make a distillers channel
<SiFuh> I will need to think about the name though
<SiFuh> Maybe moonshine/r/s
<ppetrov^> in Fallout 3, there's moonshine drink
<SiFuh> There was in Matric 1 too
<SiFuh> Tnak made it, good for two things
<SiFuh> Tank*
<ppetrov^> :D
<SiFuh> I don't know how these fuckers can make it illegal
<ppetrov^> brewing your own?
<SiFuh> I open a tin of pineapple slices and throw into a sealed container and put in the refrigerator and 1 month later it is turning into alcohol
<SiFuh> BOOM excise tax, you broke the freaking law
<ppetrov^> everything turns into alcohol
<ppetrov^> the mircale of fermentation
DaViruz has joined #crux
<SiFuh> I make rice tea and close the lid. The amylase in my salive breaks the rice down into fermentable sugars and the yeast in the air consumes it.
<ppetrov^> well, not everything, but sugers, at least...
<SiFuh> BOOM excise tax, you broke the freaking law
<SiFuh> Lactobacillus is a bacterium that can create alcohol, I use a small amount of it in my beer
<SiFuh> Yeast is a (like a fungus) that is used mostly for creating alcohol. It is everywhere. :-P EVERYWHERE
<ppetrov^> amylase breaks sugers into smaller sugars...
<SiFuh> I can wash and not cook vegetables and soak in water and it ferments because of the natural yeast
<SiFuh> BOOM excise tax, you broke the freaking law
<SiFuh> Yeah yeast can't digest long or large amounts of sugar
<SiFuh> So you need some enzymes to break it down smaller
<ppetrov^> you need enzymes to convert it
<SiFuh> It is why the Tibetans spit into the warm cooked rice first before turning it into Chang (Like Sake)
<ppetrov^> you know, it's bee qwuite some times i studied these...
<SiFuh> For barley we usually let the barley germinate. The germiantion process creates amylase. Then they dry it to stop the process and then use it for making beer
<SiFuh> I never knew it myself until I read a French and a Korean paper about it
<SiFuh> That was in 2019
<ppetrov^> heh
<SiFuh> For me it was Malt, sugar, yeast
<SiFuh> Didn't know Malt was malted barley. Malted barley was semi-germinated then dried barley
<ppetrov^> there you go
<ppetrov^> the miracle
<SiFuh> In Australia we just bought a tin of malt syrup and made beer from that
<SiFuh> Yeah that is it
<SiFuh> Acetlydhyde is interesting though. It depends on temperature the amount you get
<SiFuh> And even though that image is correct, they should swap the Ethanol and Acetyldehyde around
<SiFuh> Majority comes after
<SiFuh> Alcohol fermentation occurs inside your intestinal tract as well
<SiFuh> I have a processed sugar free diet. But those how eat processed sugar create alcohol everytime they eat. Some even get Candida or yeast infections :-P
<SiFuh> 3How about the channel #alcohol ?
jue has joined #crux
<SiFuh> Hmm aparently Marty or some other clown from back to the future registered it
<SiFuh> I will register moonshiners
<ppetrov^> man, you don't have to sell me on alcohol fermentation. i'm already sold
<SiFuh> reading how to set a topic
<SiFuh> I ended up making #moonshiners because everthing else was empty but registered
<SiFuh> We will see if anyone finds and joins for now it will be stagnant and boring hahaha
<cruxbot> [core.git/3.6]: ca-certificates: update to 20220719
<farkuhar> jue: we figured out why ps2pdf was segfaulting during the graphviz build. It seems to happen only if ghostscript was built on a system with opt/paper present. ghostscript built on a clean system will not segfault when processing the requested file(s).
ty3r0x has quit [Quit: Konversation terminated!]
<SiFuh> farkuhar: No, it was your good work
todrik has quit [Ping timeout: 252 seconds]
<cruxbot> [opt.git/3.6]: paper: update to 1.2.2
<jue> farkuhar: nice find, thanks
<jue> farkuhar: paper version 1.2.2 seems to fix the issue
<ppetrov^> farkuhar, thanks for solving this mystery
todrik has joined #crux
<farkuhar> My hunch about it being an upstream problem turned out to be correct, but I was looking at the wrong upstream. Here's the original bug report that prompted the version bump just committed by jue: https://github.com/rrthomas/libpaper/issues/24
ppetrov^ has quit [Quit: Leaving]
jue has quit [Ping timeout: 272 seconds]
_moth_ has joined #crux
todrik has quit [Remote host closed the connection]
tilman has quit [Ping timeout: 272 seconds]
tilman has joined #crux
groovy2shoes has quit [Ping timeout: 244 seconds]
darfo has quit [Quit: Leaving]
groovy2shoes has joined #crux