<    March 2017    >
Su Mo Tu We Th Fr Sa  
          1  2  3  4  
 5  6  7  8  9 10 11  
12 13 14 15 16 17 18  
19 20 21 22 23 24 25  
26 27 28 29 30 31
00:00 kaniini joined
00:19 ostera joined
00:38 rdutra joined
00:45 ostera joined
01:12 ostera joined
01:19 tty` joined
01:43 <gromero> is there a particular reason for leaving javac without a symlink in /usr/bin leaving javac out of $PATH?
01:44 <kaniini> jdk package should install javac skyline
01:45 <kaniini> er symlink
01:51 <gromero> kaniini: https://github.com/alpinelinux/aports/pull/1169 https://github.com/alpinelinux/aports/pull/1170
01:51 <gromero> "jdk package should install javac skyline" ?
01:52 s33se joined
01:52 <gromero> apk add openjdk{7,8} installs javac, but link is not created...
01:52 ppisati joined
01:53 <gromero> I guess travis will spend sometime building openjdk
01:54 <gromero> good night
02:13 <kaniini> hm
02:42 Emperor_Earth joined
04:24 minimalism joined
04:40 dobson joined
04:50 qrwteyrutiyoup joined
05:09 fabled joined
05:14 <TemptorSent> 'evening fabled/
05:15 <TemptorSent> How're you?
05:15 <fabled> TemptorSent, good morning
05:15 <fabled> waking up... :)
05:16 xentec joined
05:17 <TemptorSent> fabled: Bed is still several hours off for me, but it's been a long day.
05:19 <TemptorSent> fabled: I only had a little time to work on the computer today, but hopefully my last push answers a few of the questions that came up.
05:25 tdtrask joined
05:28 <kaniini> hi
05:28 <TemptorSent> Hello.
07:10 t0mmy joined
07:17 scadu joined
07:21 scadu joined
07:34 atmoz joined
07:58 fekepp joined
08:10 royger joined
08:20 <clandmeter> kaniini, regarding wiki, you got a new acc i guess? it takes a few days before you can add them.
08:24 andypost joined
08:47 ostera joined
08:48 ostera joined
08:58 <asie> still getting "Exception in thread "main" java.lang.UnsatisfiedLinkError: /usr/lib/jvm/java-1.8-openjdk/jre/lib/amd64/libfontmanager.so: Error relocating /usr/lib/jvm/java-1.8-openjdk/jre/lib/amd64/libfontmanager.so: AWTFontDefaultChar: symbol not found"
08:59 <asie> reproducible via running any AWT-using piece of Java code
09:23 tty` joined
09:23 fekepp joined
09:27 kaniini joined
09:30 fekepp joined
09:31 vakartel joined
09:35 kaiyou joined
09:41 blueness joined
09:47 <jirutka> sigh, it’s a simple boolean algebra! what sense makes `[ "$CARCH" = ppc64le ] && [ "$CARCH" = s390x ]`?! yeah, none!
09:47 <jirutka> going to clean this mess
09:48 <skarnet> well, you're forgetting the case where ppc64le = s390x
09:48 <jirutka> people should not commit when too tired or drunk
09:49 <jirutka> seriously, rnalrd, tired or drunk?
09:51 <rnalrd> :-/ more like headache
09:51 <jirutka> this must be really massive headache :(
09:58 <rnalrd> jirutka, can I push the fix or you want to do it?
09:59 <jirutka> rnalrd: I’d prefer to not
09:59 <jirutka> rnalrd: or open a PR or send patch here
10:00 <jirutka> currently I’m thinking how to do this clearly
10:04 <rnalrd> thinking for the packages with luajit dependency? or something else?
10:25 <rnalrd> k, i'll surrender the keyboard today :)
10:25 <rnalrd> it seems that missing a couples of nights of sleep kills me
10:25 <rnalrd> sorry for the mess
10:26 BitL0G1c joined
10:27 <clandmeter> rnalrd, i dont think your fix is a fix :)
10:27 <clandmeter> better to use a case and check arch
10:31 <rnalrd> tnx clandmeter, it think I'm able to mess even a simple case statement given the circumstances ;-)
10:33 <jirutka> clandmeter: that’s what I did (pushed it a moment ago)
10:34 <jirutka> I’m thinking how to avoid manually disabling arch when some dependency is not available for some arch
10:36 <clandmeter> rnalrd, also the == is non posix iirc
10:36 blueness joined
10:36 <jirutka> clandmeter: yeah, I’ve also noted this
10:37 <jirutka> I forgot to fix something, IIRC there’s also some pkg with noarch !s390x
10:38 <jirutka> eh, no, it’s all
10:40 czart_ joined
10:40 xsteadfastx joined
10:48 <xsteadfastx> i just sent a patch for the pkg bitlbee-facebook. wasnt there a overview for the patches to be merged?
10:51 <_ikke_> patchworks you mean?
10:54 <xsteadfastx> yes thats it :)
10:57 <_ikke_> https://wiki.alpinelinux.org/wiki/Patch_Workflow#I.27m_an_alpine_developer._How_can_I_start_to_use_patchwork_.3F
10:58 <_ikke_> No clue where it lives
11:00 blueness joined
11:21 vakartel joined
11:30 <^7heo> ncopa: isn't that https://patchwork.alpinelinux.org/patch/2198/ already in?
11:32 <jirutka> ^7heo: it is, but stupid patchwork…
11:32 <^7heo> jirutka: ?
11:32 <jirutka> ^7heo: it didn’t closed it
11:32 <^7heo> ah ok
11:33 lucybun joined
11:58 sean joined
11:58 gromero joined
12:01 <^7heo> jirutka: I think it just happened that people forgot about the patchwork
12:01 <^7heo> jirutka: and worked on imagemagick at the same time.
12:02 <^7heo> I mean, look at f2333fd696b1778d0ef56323a23a178e49e8eadd
12:03 <^7heo> that alone would have prevented the patch in patchwork.
12:03 <^7heo> and it happened 2 days after the patchwork.
12:05 leitao joined
12:06 rdutra joined
12:08 ferseiti joined
12:09 <^7heo> ncopa: my patchwork account is inactive; and I never got a mail.
12:15 farosas joined
12:28 <jirutka> ^7heo: yeah, that’s quite possible :)
12:31 rdutra1 joined
13:16 stwa joined
13:18 ncopa joined
13:18 ncopa joined
13:42 LouisA joined
14:01 tmh1999 joined
14:17 leitao joined
14:22 ferseiti joined
14:25 leo-unglaub joined
14:34 <mitchty> looks like ghc 8.2 is about ready for release
14:34 <mitchty> https://ghc.haskell.org/trac/ghc/wiki/Status/GHC-8.2.1
14:34 <mitchty> once a rc is out should i get that pushed to testing?
14:37 <mitchty> also could upx get migrated to community from testing? ucl its dependency is there, would be nice to keep them together
14:38 leo-unglaub joined
14:55 <jirutka> mitchty: yeah, I’d like to move ghc from testing to community before v3.6
15:06 leo-unglaub joined
15:13 hl[m] joined
15:16 <hl[m]> !join #alpine-offtopic
15:18 <^7heo> matrix...
15:20 hl_ joined
15:20 hl_ joined
15:22 <kaniini> hl: i dont think that works
15:22 <kaniini> ^7heo: in matrix you could be 7heo right now
15:22 <jomatv6> ^7heo: on the official matrix.org server...
15:23 <jomatv6> ... no harm here as this channel is public, but i wouldn't recommend to use any of matrix.org's infrastructure :-D
15:23 <kaniini> i only use their irc relay
15:24 <kaniini> because they have freenode i:lines i do not have
15:25 <hl> kaniini: err, yeah, meant to type it in the bridge window :P
15:25 leo-unglaub joined
15:36 vakartel joined
15:42 stwa joined
15:52 farosas joined
15:58 ostera joined
16:00 <jirutka> anyone using Matrix here? I’d like to install my Matrix node soon
16:02 <hl> I am.
16:03 <jirutka> hl: how good does it work?
16:03 <hl> jirutka: I only just set it up, but it seems to work OK.
16:04 <jirutka> hl: what services did you integrated?
16:04 <hl> jirutka (IRC): hmm?
16:07 <jirutka> hl: I’d like to connect IRC, XMPP, Slack, Twitter PMs and FB Mess-anger via Matrix
16:08 tmh1999 joined
16:09 loopz_ joined
16:09 <loopz_> hey guys, is there any way to make who or w in alpine to work??
16:09 <_ikke_> loopz_: No, musl does not support it
16:11 <loopz_> do you guys have any workarounds to check who's logged into the machine at the moment?
16:12 <jirutka> loopz_: just look into (h)top?
16:21 <rdutra1> ncopa: I sent a patch do linux-vanilla and travis broke. It exceeded the maximum time limit for the job
16:36 leo-unglaub joined
16:43 minimalism joined
16:48 dsabogal joined
17:00 <leitao> aiccu could not be download because curl is getting a 402 error
17:00 <leitao> curl: (22) The requested URL returned error: 402 Payment Required
17:01 <leitao> Any chance someone saw this problem before?
17:12 <mitchty> jirutka: ok cool no worries, one aspect of ghc version updates impacts things that build but since nothing right now depends on it its not a huge deal
17:14 leo-unglaub joined
17:22 gromero joined
17:22 <TemptorSent> ^7heo: Actually, it's not that bad (or it wasn't last time I had to do it)
17:22 gromero joined
17:23 <jomatv6> leitao: sixxs is dead
17:24 <jomatv6> https://www.sixxs.net/sunset/
17:25 <jomatv6> jirutka: i'm running 3 matrix instances. one ldap backed for our space, a public one for friends and one without users, just acting as irc bridge
17:26 <jomatv6> so everyone is able to join #channelname:darkfasel.net directly without the matrix.org-bridge-stuff.
17:26 <jomatv6> the irc bridge is a pile of crap :>
17:27 <jomatv6> and i don't have any other integrations. Oh and to stay on topic: of course it runs on alpine :-D
17:37 <gromero> what's the "meaning" of a log directory empty like: http://build.alpinelinux.org/buildlogs/build-edge-x86_64/main/qemu/ ? Did it failed or did not start yet?
17:41 <gromero> or even it's still building...?
17:42 <Shiz> considering all builders are idle, it's definitely not still building
17:42 <Shiz> my guess would be it's yet to build considering all the other dirs
17:42 <Shiz> there
17:43 <Shiz> looks like only the first package in edge/x86_64/main built...
17:50 <kaniini> jirutka: i am working on packaging for synapse -- but i don't think synapse should be used by alpine infrastructure as there are more modern replacements in development (Go, Rust, C++) with much more emphasis on scalability
17:51 <Shiz> C++, modern?
17:51 <Shiz> :p
17:51 <jirutka> kaniini: Go, modern?!
17:51 <jirutka> kaniini: but I’m glad that there’s also Rust implementation! link?
17:52 <Shiz> (do we package Rust yet?)
17:52 <Shiz> (an up-to-date version)
17:52 <jirutka> kaniini: I’m not considering it for Alpine infra (where?), but for myself
17:52 <kaniini> modern in terms of being built around being production-ready instead of proof of concept like synapse is
17:52 <gromero> Shiz: thx
17:52 <jirutka> Shiz: still waiting for upstream, I should ping them once again…
17:52 <Shiz> jirutka: waiting for musl prebuilds?
17:52 <Shiz> if so, you could use the void ones :p
17:52 <Shiz> temporarily
17:52 <jirutka> Shiz: yes, b/c I don’t wanna cross-compile it every time, I really hate cross-compiling rustc
17:53 <kaniini> i think we can bootstrap rust in the same way we do haskell
17:53 <Shiz> https://github.com/voidlinux/void-packages/blob/master/srcpkgs/rust/template#L25
17:53 <Shiz> :p
17:53 <kaniini> just use docker to build a temporary environment and build it in there
17:53 <jirutka> Shiz: are you sure that Void has newer version? they used my build before
17:53 <kaniini> and then wrap it around an APKBUILD
17:53 <kaniini> :p
17:53 <jirutka> the problem is that they may be many breakages
17:54 <jirutka> I want to be dynamically linked musl finally officially supported and they run CI tests on it
17:54 <Shiz> jirutka: 1.15.1 is far newer than anything alpine has
17:54 <Shiz> and with my limited testing, it seemed to got me through at least part of building rust 1.15
17:54 <Shiz> 1.16*
17:54 shykes__ joined
17:54 mattaitchison_ joined
17:54 <Shiz> of course i ran into other build system issues...
17:54 <jirutka> I’ve tried to build 1.11.0 with our 1.10.0 and it failed to build
17:54 <Shiz> but the bootstrap rust was not the issue
17:54 <Shiz> :p
17:54 <jirutka> I’m happy to hear that, I’ll give it another try
17:55 <jirutka> at least they finally implemented generic support to switch between static and dynamic linking, so it should not be needed to use the hack I used before
17:55 <Shiz> really?
17:55 <Shiz> nice
17:55 <jirutka> yeas
17:56 justincormack__ joined
17:56 <Shiz> happen to have a reference of some kind? curious what switches it needs for that
17:56 <jirutka> to be honest, I was really disgusted from my experience with bootstrapping rustc first time, that’s why I haven’t touch it after that yet
17:56 hanez_ joined
17:57 <jirutka> Shiz: https://github.com/rust-lang/rust/pull/37545
17:57 <kaniini> jirutka: re: alpine infra and matrix, at some point i want to leverage matrix to make it possible to interact with the alpine chat resources via protocol/service of choice
17:58 <kaniini> but synapse could not scale that
17:58 <jirutka> https://github.com/rust-lang/rust/issues/31322
17:58 <Shiz> jirutka: nice!
18:04 <jirutka> Shiz: to be very honest, I actually quite hope that someone will take the best ideas from Rust and create a new, truly lightweight language and implementation; but still Rust is the best we have now
18:04 <Shiz> :p
18:04 <jirutka> but when I see cargo, I wanna cry
18:04 <Shiz> yeah rust is far too complex for me as it is
18:05 <jirutka> cargo is doing tooo many things
18:05 <jirutka> and then there’s that LLVM thing…
18:06 <jirutka> but I hope that someday they will be able to switch to something more sane, they’ve implemented IR and so can do many optimizations on IR level instead of relaying on LLVM
18:07 <jirutka> btw I’ve packaged Emscripten this week… emscripten-fastcomp (their fork of LLVM + clang…) is already in testing, emscripten itself is waiting in PR, therejs some problem that I can’t reproduce locally
18:07 <jirutka> Emscripten is incredible mess, it’s more like a dump of random files
18:11 <Shiz> jirutka: re:issue https://github.com/travis-ci/travis-ci/issues/943
18:11 <jirutka> aha, so it’s not problem in the package!
18:12 <jirutka> aha, it needs /dev/shm?
18:12 <Shiz> yes
18:12 <Shiz> that seems to be the case
18:12 <Shiz> :p
18:29 <pickfire> Hi, I wonder why wine doesn't provide /usr/bin/wine.
18:31 <jirutka> ’cause it’s too drunk? :P
18:31 <jirutka> hm, srsly, there’s something wrong in the current wine pkg https://pkgs.alpinelinux.org/contents?file=wine&path=&name=&branch=&=&repo=&=&arch=&=
18:32 <jirutka> there’s /usr/bin/wine in v3.4 and in v3.5 it’s gone…
18:32 <jirutka> someone messed it
18:39 <asie> wine64
18:39 <asie> ?
18:39 <asie> or wine32?
18:43 t0mmy joined
18:55 <jirutka> asie: right, wine64 is here!
19:05 Emperor_Earth_ joined
19:11 <Shiz> jirutka: http://git.alpinelinux.org/cgit/aports/tree/community/wine/APKBUILD#n60
19:11 <Shiz> this happened
19:11 <Shiz> :p
19:11 <jirutka> but why?
19:13 <Shiz> hmm, actually
19:13 <Shiz> this was the case in 3.4 too...
19:19 tmh1999 joined
19:26 <gromero> jomatv6: so is sixxs going to be removed from alpine 3.6?
19:28 ostera joined
19:51 <^7heo> ncopa: what's the use of `conf.search_device = argv[0];` in mkinitfs?
19:55 <Shiz> ^7heo: if nlplug-findfs is given an argument, it will simply only search for that device
19:55 <Shiz> instead of any block device that matches the filter
19:58 blueness joined
20:11 blueness joined
20:13 <^7heo> Shiz: in my experience, argv[0] isn't really the first argument.
20:14 <^7heo> Shiz: unless it's shifted, lemme check.
20:14 <Shiz> then you're not familiar with ARGBEGIN semantics
20:14 <Shiz> :)
20:14 <^7heo> I was, so that's why I wrote that just now.
20:14 <^7heo> right
20:15 <^7heo> Ok then all good.
20:15 <^7heo> I was wondering why the code was taking the binary's name as a device...
20:15 <^7heo> Thanks Shiz
20:15 <Shiz> :p
20:15 <Shiz> (fwiw, i think ARGBEGIN changing argc/argv is rather bad and this is exactly the reason why)
20:16 <Shiz> macros shouldn't change local variables that aren't passed to them
20:16 <skarnet> we're writing code for Plan9 now?
20:16 <^7heo> skarnet: it's from suckless
20:16 <^7heo> skarnet: so... yeah.
20:16 <Shiz> (so yes)
20:17 <^7heo> Shiz: yeah but the logic behind it AFAICT was that then the ARGUMENTS would start at 0
20:17 <skarnet> wat
20:17 <skarnet> nlplug-findfs is from suckless?
20:17 <^7heo> skarnet: we have a few binaries coming from suckless in our codebase.
20:17 <^7heo> skarnet: initially yes.
20:17 <skarnet> I thought it was from ncopa.
20:17 <^7heo> skarnet: 20h wrote the beginning AFAICT.
20:17 <skarnet> hence the nl.
20:17 <^7heo> I'll have to ask him for the confirmation.
20:18 <Shiz> skarnet: nl just means netlink
20:18 <Shiz> no?
20:18 <^7heo> argv0 = strrchr(argv[0], '/'); if (argv0++ == NULL) ...
20:18 <skarnet> ah, it's possible.
20:18 <Shiz> yeah, 20h wrote the initial version
20:18 <^7heo> is the if block only executed if the argv[0] ends with '/'?
20:19 <^7heo> or is it me?
20:19 <skarnet> it's just that I encountered nl* in ncopa's code, so I figured it stood for n_atanae_l
20:19 <^7heo> skarnet: aaah ;)
20:19 <^7heo> skarnet: you and naming... :D
20:19 <skarnet> but now you're mentioning it, all those programs also dealt with the netlink
20:20 xfix joined
20:20 <clandmeter> its because he likes The Netherlands so much.
20:20 <Shiz> hence nl.alpinelinux.org
20:20 <Shiz> :p
20:20 <^7heo> ahahaha
20:21 <skarnet> clearly it's the netlink server
20:26 <jomatv6> gromero: i don't know how alpine packet maintainers for aiccu (=sixxs client) will react, but it is more or less worthless without sixxsd (the server), which is closed source and I'm not aware of any free alternatives
20:30 <kaniini> nuke it
20:30 <kaniini> that's with all relevant hats on
20:30 ferseiti joined
20:32 faffolter joined
20:32 faffolter joined
20:35 <skarnet> I want a hat that allows me to launch nukes too
20:35 <Shiz> something something maga hat
20:36 <skarnet> no, only one of the hat-wearers can do that
20:36 <skarnet> so the hat is a scam
20:40 <Shiz> "Although our provisioning servers and routing daemon (sixxsd) are very well thought out, they do have some intricate dependencies on how we built SixXS. As such, offering the code base will not be necessarily useful for others. Over time, given effort on Jeroen’s part, this may change. We cannot make any promises at this point."
20:40 <Shiz> :(
20:41 <skarnet> bullshit
20:42 <skarnet> even Google opensourced their shit and God knows it had intricate dependencies
20:42 <skarnet> so yeah, if they won't do it, you can safely question the "very well thought out" part
20:43 <Shiz> sixxs is not google and as such they may weigh different things differently
20:43 <Shiz> no need to jump to conclusions
20:44 <jirutka> agree with skarnet, this msg is total bullshit
20:45 <skarnet> I will happily jump to any conclusion I want if a company gives excuses not to opensource their stuff
20:45 <Shiz> which is your right, doesn't mean it's a justified one
20:45 <Shiz> :)
20:45 <skarnet> how could I not? I don't have the code to verify, it's all speculation at this point!
20:45 <Shiz> on a happier note, https://dlbjbjzgnk95t.cloudfront.net/0906000/906395/https-ecf-flsd-uscourts-gov-doc1-051117728570.pdf
20:45 <jirutka> *OR* they have more responsibility and awareness than Google, so they know that it’s a shit and so rather take it closed
20:46 <skarnet> closed shit is *more* harmful than open shit: the latter can be fixed over time
20:46 <jirutka> well…
20:46 <jirutka> it depends
20:48 <jirutka> if they are the only one who runs it and have people who now how to keep it running and safe, then it’s a better situation than giving it away, so dozens of irresponsible and incompetent ppl would run it and just creating a security hazard
20:49 <jirutka> and we know how totally broken things are fixed during time… they are not…
20:49 <jirutka> sometimes the only real fix is to burn it with fire and reimplement
20:50 <skarnet> that's also true for the world
20:50 <* skarnet> laughs maniacally.
20:50 <^7heo> I can provide the fire.
20:50 <jirutka> when some solution, even very bad, exists, then it often does not motivate ppl enough to implement AND start using an alternative
20:51 <jirutka> skarnet: yes, very true, this concept is generic
20:53 <jirutka> I don’t know company behind Sixxs, but if it’s true that they stopped their service to increase pressure to ISPs to finally implement IPv6, then it may be really about responsibility
20:53 <jirutka> this reminds me, I should write to my ISP…
20:54 <^7heo> wasn't it he.net doing sixxs?
20:54 <Shiz> no
20:54 <Shiz> he.net have their own tunnels
20:54 <Shiz> i'm using them right now
20:54 <^7heo> ok then I'm mixing things.
20:54 <Shiz> jirutka: sixxs have always been pretty adamant about pressurizing ISPs
20:54 <Shiz> they stopped taking new applications a couple years back
20:54 <Shiz> with the message 'write to your ISPs instead'
20:56 <^7heo> I can imagine the dinner during which sixxs guys were with $ISP and $ISP was like "heh thanks guys for doing our work for free. I guess you'll also pay us that dinner, right?"
21:01 ostera joined
21:04 <jirutka> why the hell all Google Groups requires to log-in? it used to be open (I mean groups accessible without login for read-only)
21:05 <jirutka> I know that the answer is probably "’cause f*cking Google, ’cause they can”, but does anyone know something more about it?
21:06 <jirutka> I usually use google cache to read the content, but that’s not always applicable
21:07 <^7heo> jirutka: link?
21:08 <jirutka> ^7heo: for example https://groups.google.com/forum/#!topic/ruby-security-ann/Dy7YiKb_pMM
21:09 <^7heo> I can totally read it.
21:09 <^7heo> [ANN] nokogiri security update - 1.6.7.2
21:09 <^7heo> "Nokogiri version 1.6.7.2 has been released, pulling in several upstream patches to the vendored libxml2 to address the following CVE..."
21:10 <jirutka> are you logged into some google account?
21:11 <^7heo> I do not even have a google account.
21:11 <jirutka> maybe some of my anti-social blockers or adblock is doing it :/
21:11 <^7heo> I use no-script, ghostery and ABP.
21:11 <^7heo> no exceptions on any; but no-script, which is allowing google.com and gstatic.com
21:12 <^7heo> However
21:12 <^7heo> I passed a very advanced captcha earlier, while connecting to youtube.
21:13 <Shiz> i can read it with uBlock + uMatrix with exempts set for {api,groups,www}.google.com and *.gstatic.com
21:13 <jirutka> aha!
21:13 <jirutka> some cookies
21:14 <jirutka> maybe I’ve accidentally logged into some google account in non-private mode :/
21:22 leo-unglaub joined
21:26 czart joined
22:06 TemptorSent joined
22:16 farosas joined
22:50 <Klowner> hrm, seems like abuild -F to build as root works to build, but then it fails at the end telling me not to run as root :|
23:05 blueness joined
23:05 ostera joined
23:19 ostera joined
23:23 <TemptorSent> Just pushed latest mkimage to my branch including fully rewritten mkinitfs and compatability wrapper.
23:24 <TemptorSent> I have not made all the backwards compat features themselves yet, but everything else, including all flags, should work.
23:28 <Klowner> 'win 5
23:28 <Klowner> :|
23:32 ostera joined
23:46 ostera joined