josh-k_ has quit [Remote host closed the connection]
josh-k has joined #rubinius
josh-k has quit [Ping timeout: 245 seconds]
byprdct_ has joined #rubinius
byprdct has quit [Ping timeout: 244 seconds]
byprdct has joined #rubinius
byprdct_ has quit [Ping timeout: 250 seconds]
havenwood has quit [Ping timeout: 246 seconds]
meh` has quit [Ping timeout: 260 seconds]
<jnh> Hrm. Is File::FNM_EXTGLOB missing or do I need to require something?
<jnh> rbx-2.2.10 :018 > File.constants.grep /FNM/
<jnh> => [:FNM_NOESCAPE, :FNM_PATHNAME, :FNM_DOTMATCH, :FNM_CASEFOLD, :FNM_SYSCASE]
amclain has quit [Quit: Leaving]
dimday has joined #rubinius
_whitelogger has joined #rubinius
heftig_ has joined #rubinius
heftig has quit [Read error: Connection reset by peer]
diegoviola has quit [Remote host closed the connection]
johnmuhl_ has quit [Quit: Connection closed for inactivity]
havenwood has joined #rubinius
josh-k has joined #rubinius
heftig_ is now known as heftig
josh-k has quit [Remote host closed the connection]
josh-k has joined #rubinius
josh-k has quit [Ping timeout: 244 seconds]
havenwood has quit [Remote host closed the connection]
noop has quit [Ping timeout: 264 seconds]
josh-k has joined #rubinius
|jemc| has quit [Quit: WeeChat 1.0.1]
josh-k_ has joined #rubinius
josh-k has quit [Ping timeout: 264 seconds]
byprdct_ has joined #rubinius
byprdct has quit [Ping timeout: 250 seconds]
noop has joined #rubinius
jnh has quit [Quit: Leaving...]
JohnBat26 has joined #rubinius
dimday has quit [Quit: Leaving.]
byprdct_ has quit []
flavio has joined #rubinius
flavio has joined #rubinius
flavio has quit [Quit: WeeChat 0.4.1]
flavio has joined #rubinius
flavio has joined #rubinius
<yorickpeterse> morning
lbianc has joined #rubinius
jeremyevans has quit [Ping timeout: 245 seconds]
jeremyevans has joined #rubinius
<yorickpeterse> also yay weekly builds enabled again
<yorickpeterse> apparently all I had to install was libedit-devel
lbianc has quit [Quit: ChatZilla 0.9.91 [Firefox 24.8.1/20140923194127]]
_whitelogger has joined #rubinius
<yorickpeterse> weee, first test suite working fine on 2.3
<yorickpeterse> Though I didn't really expect any problems since it's pretty much compatible with 2.2.10
postmodern has quit [Quit: Leaving]
lbianc has joined #rubinius
<kagaro> interesting i installed llvm via brew run rvm install rbx-2.3 it compiles and links and stuff then says not installed :P
<kagaro> check this out yorickpeterse https://gist.github.com/washu/f3e369a6105c4bf5a51b
<kagaro> lol nvm rvm is just being stupid
<mpapis> kagaro, open a ticket for it: https://github.com/wayneeseguin/rvm/issues
<kagaro> i found the problem it and correct .. operator error
lbianc has quit [Quit: ChatZilla 0.9.91 [Firefox 24.8.1/20140923194127]]
<mpapis> kagaro, the bug is in rvm feature "use on install" - I would still appreciate a ticket
<kagaro> ticket added
<mpapis> thank you!
max96at|off is now known as max96at
lbianc has joined #rubinius
johnmuhl_ has joined #rubinius
meh` has joined #rubinius
lbianc has quit [Quit: ChatZilla 0.9.91 [Firefox 24.8.1/20140923194127]]
enebo has joined #rubinius
<kagaro> anyone else get Unrecognized option: --skip-system on centos when trying to install rbx 2.3.0 via rvm
<mpapis> kagaro, might be just removed option from rbx, please open another ticket, I will fix it in few ... maybe an hour
<kagaro> didnt get the error on mavericks just centos
<mpapis> kagaro, that's because RVM does the flags evaluation system dependent and OSX provides LLVM by default
<kagaro> ahh gotcha
<yorickpeterse> kagaro: is this regular CentOS, or Amazon CentOS?
<kagaro> regualr centos
<kagaro> 6.5
<yorickpeterse> hm
<kagaro> i did install llvm 3.4 on the box to be sure
<kagaro> looks like the --skip-system wa already removed for Arch linux a few months back
havenwood has joined #rubinius
_whitelogger has joined #rubinius
<brixen> yorickpeterse: binaries should be up
<yorickpeterse> nice
<kagaro> brixen on a side note the rubysl date failing on ruby startup seems to no longer exist under 2.3.0 i ran a script that looped 1000 times trying to get it to fail and it didnt
<brixen> kagaro: excellent :)
<brixen> does centos use rpm ?
<yorickpeterse> brixen: yes
<yorickpeterse> well the package format that is, the actual package manager is yum
<brixen> who wants to help the packagers make Rubinius packages?
benediktdeicke has joined #rubinius
<yorickpeterse> brixen: what needs to be done for that, besides harassing people?
<brixen> yorickpeterse: we'd need to know who is in charge of that and what help they need
<brixen> and yes, harassing them :)
<yorickpeterse> Well, at least for Arch Linux it seems things are ok version wise
josh-k_ has quit [Remote host closed the connection]
josh-k has joined #rubinius
josh-k has quit [Ping timeout: 258 seconds]
yopp has quit [Ping timeout: 250 seconds]
lbianc has joined #rubinius
Gibheer has quit [Ping timeout: 250 seconds]
Gibheer has joined #rubinius
yopp has joined #rubinius
noop has quit [Ping timeout: 265 seconds]
havenwood has quit [Ping timeout: 246 seconds]
lbianc_ has joined #rubinius
lbianc has quit [Ping timeout: 256 seconds]
lbianc_ is now known as lbianc
Gibheer has quit [Read error: Connection reset by peer]
Gibheer has joined #rubinius
DireFog has quit [Quit: No Ping reply in 180 seconds.]
DireFog has joined #rubinius
|jemc| has joined #rubinius
johnmuhl_ is now known as johnmuhl
Bish_ is now known as Bish
flavio has quit [Quit: WeeChat 0.4.1]
benediktdeicke has quit [Quit: Linkinus - http://linkinus.com]
tenderlove has joined #rubinius
amsi has joined #rubinius
tenderlove has quit [Quit: Leaving...]
tenderlove has joined #rubinius
_whitelogger has joined #rubinius
_whitelogger has joined #rubinius
JohnBat26 has joined #rubinius
<mpapis> kagaro, started working on this right now, give me 5 minutes
<kagaro> k
<kagaro> ok mpapis that gets me past the config issue. error compiling is a differnet story
<kagaro> brixen, yorickpeterse you want a ticket for that?
<|jemc|> kagaro: if there's no ticket for that yet, yes, add one
<|jemc|> if you just want to work around it for now for your own installation you could set your compiler to not treat warnings as errors
<|jemc|> if you file a ticket, please include your compiler version
<kagaro> updated the ticket with compiler version
<brixen> kagaro: you should compile with clang if it's available
<kagaro> just install clang
<kagaro> err just installed it even
<kagaro> trying again
<brixen> ok
<brixen> that has to be one of the all time stupidest gcc warnings ever
<brixen> kagaro: also looks like you might upgrade to gcc 4.6
<brixen> if that's even available
<brixen> since that's what's on Travis and the build passes
<kagaro> switching to clang i think is a better option
<brixen> indeed it is
<brixen> it's also future-proofing :)
<brixen> not going to support gcc much longer
<yorickpeterse> oh?
<yorickpeterse> we'll need to hack mkmf in that case though
<brixen> why?
<brixen> not that I'm opposed to hacking mkmf
<yorickpeterse> mkmf uses RbConfig::CONFIG['CC'] for compiling C-exts. If that's set to clang then users are required to have clang installed when installing C exts
noop has quit [Ping timeout: 255 seconds]
<yorickpeterse> I used to compile rbx binaries with clang until I found that out, hence we reverted back to gcc
GitHub73 has joined #rubinius
<GitHub73> [rubinius] brixen pushed 1 new commit to master: http://git.io/7nwN-g
<GitHub73> rubinius/master f2d9aa6 Brian Shirai: Voidify write() calls. Fixes #3174.
GitHub73 has left #rubinius [#rubinius]
<brixen> kagaro: you can check if master fixes the issue with your gcc
<brixen> kagaro: but you'd probably have to clone + build
<brixen> unless rvm has a --HEAD like option
<brixen> yorickpeterse: yep
<kagaro> yeah will be a bit befor ei can try that i think part of my llvm install is jacked up
<brixen> how does part of one's llvm install get jacked up?
<kagaro> deps didnt install correctly via yum
<brixen> like, you were racing your llvm install down the city streets and high-centered on a fire hydrant?
<kagaro> or i didnt install something that isnt marked as a dep
<kagaro> think thats it
<kagaro> i didnt install llvm-static
<mpapis> brixen, kagaro: rvm install rbx-head
<kagaro> swiching to clang worked
<brixen> yorickpeterse: btw, we're the one who sets RbConfig::CONFIG['CC'] and we can set it to what's available
<brixen> if we actually hit ABI issues between clang and gcc, we can insist people use clang
<brixen> but just because rbx is built with clang doesn't require C-exts to be built with it
<yorickpeterse> brixen: True, one idea would be to just always set CC to gcc, even though we use clang ourselves
<brixen> nah, if clang is available and the correct version, we should prefer it
josh-k has joined #rubinius
<yxhuvud> speaking of gcc, did you see that they are working at adding a jit to it?
<yorickpeterse> wat
<yorickpeterse> JITting precompiled code? We must go deeper
<yorickpeterse> (heh, JITting)
<yorickpeterse> probably should be JIT'ing instead
<brixen> I usually go with ' for awkward words like that
<yxhuvud> maybe it is more interesting to the japanese guy playing around with making mri jit though :D
<brixen> I'm sure a gcc jit would be phenomenal
<brixen> a phenomenal pain in that ass, that is
<yorickpeterse> the two often go hand in hand
<yorickpeterse> yxhuvud: Ah, so this looks more like something LLVM-alike
<yorickpeterse> I was thinking of a JIT for C, which seemed a bit odd
<yorickpeterse> "We're running a JIT on precompiled C code for optimal performance. But wait, there's more!"
<yorickpeterse> "Call now and get a copy of "Debugging gcc output" for free!"
<headius> yeah looks very llvm-like
<brixen> it's one of the thing LLVM suggested from the very beginning
<brixen> things*
<brixen> yorickpeterse: adaptive compilation for C isn't an odd or far-fetched idea
<yorickpeterse> I don't see much use of that when everything is already precompiled though
<yorickpeterse> Ideally that would be just as good as whatever a JIT produces
<chrisseaton> yorickpeterse: Truffle uses a JIT to run C extensions - it can be faster as you can inline through C code and Ruby code
<yorickpeterse> (note the "ideally" here :P)
<yorickpeterse> also note the precompiled part
<chrisseaton> yorickpeterse: a static compiler can't see the data running through your program - only the code - which is only half the story
<brixen> unless your types are good enough
<yorickpeterse> I was about to say "...if you have static typing"
<yorickpeterse> then all your problems would go away
<yxhuvud> that is quite quoteworthy.
* yorickpeterse crawls back into his cave of CSS selectors
<yorickpeterse> honestly these bastards are such a drain of motivation, ugh
<yorickpeterse> can't wait until I'm done with it
GitHub175 has joined #rubinius
<GitHub175> [rubinius] dubek opened pull request #3175: Accept a nil argument for Array#flatten. (master...array-flatten-nil) http://git.io/CJJ4rg
GitHub175 has left #rubinius [#rubinius]
* yorickpeterse pulls out the rubber glove
<brixen> yorickpeterse: think of the expertise you are acquiring
<yorickpeterse> well yeah, that's the only reason I'm doing it :P
<yorickpeterse> besides myself actually needing CSS support for the time being
<brixen> ah man, was so confused
<brixen> why is building rbx trying to install EM
<brixen> but of course it's not
<brixen> it's that shit from rubygems auto-compiling C-exts
<yorickpeterse> That's still happening?
<yorickpeterse> Haven't had that problem in a while
<brixen> system ruby still had rubygems 2.2.2 afaics
<kagaro> took a bit but now i can install puppet it over a bunch of web servers now
<kagaro> had to pass the -C -cc=clang
<yorickpeterse> make sure you also set cxx in that case
<yorickpeterse> otherwise it will still use g++ for C++ code
<yorickpeterse> which doesn't work when combined with using clang for CC
<yorickpeterse> brixen: on a sidenote, you mentioned something about moving Enova over to Oga in some shape or form. Would that actually be for serious internet business, or just small tooling things?
max96at is now known as max96at|off
jnh has joined #rubinius
<brixen> very serious internet business
<brixen> is there any other kind? :)
jnh has quit [Remote host closed the connection]
<yorickpeterse> well, there's one that involves me getting a better insurance policy, and one that doesn't :P
<brixen> heh
<yorickpeterse> But no, I'm curious what people will be using it for
<|jemc|> I've taught the chickens on my farm to output XML reports about their expected egg rates for the week based on their diet and habits and a variety of other environmental factors
<|jemc|> so I'll be using oga to count my chickens before they hatch
<brixen> seems like MRI 2.1.4 likes to segv running melbourne
<yorickpeterse> |jemc|: ok sanity check here
<yorickpeterse> |jemc|: that's a joke?
<yorickpeterse> Because if so, fuck you got me :P
<yorickpeterse> if not: wat
<|jemc|> :D
<yorickpeterse> honestly with programmers that would be totally plausible
<brixen> seems like a good use of XML
<brixen> basically, comes out of the chickens with the rest of their poo
<yorickpeterse> haha
<yorickpeterse> Then it's only a short time before SCAP
<yorickpeterse> Simple Chicken Access Protocol
<yorickpeterse> Complete with a Chicken Service Description Language (CSDL)
<|jemc|> yorickpeterse: you should know that my social subroutines adjust my rate of jokes and sarcasm directed at a given individual to be proportional to the jokes and sarcasm I see/hear from them
<brixen> or SCRAP, Simple Chicken cRap Protocol
<|jemc|> heh
<yorickpeterse> |jemc|: heh
<brixen> er missed Access
<brixen> Access Protocol
<yorickpeterse> it's not often that I meet somebody with a similar acquired taste of humor
<yorickpeterse> (or lack thereof depending on your point of view)
<yorickpeterse> also fuck, that joke doesn't fit in a tweet
<yorickpeterse> W3 logic: provide basic examples for stupidly complex selectors such as :nth-child(n), provide much better examples for a simple one such as :first-child
<yorickpeterse> :areyoukidding"
<yorickpeterse> fek
<yorickpeterse> щ(゚д゚щ)
<brixen> yorickpeterse: do you have macros for those or do you type them from memory every time?
<yorickpeterse> I have a plugin for Weechat for that
<yorickpeterse> but I type them by hand
<yorickpeterse> Sometimes I cat the emoji db to see what they were called again :P
<brixen> heh
<yorickpeterse> ooooooooh, there's actually an autocomplete
<yorickpeterse> I did not know that
<|jemc|> yorickpeterse: what is the weechat plugin called?
<yorickpeterse> I removed some of the more questionable emoji from the DB, there's some weird shit in there
<brixen> I am not surprised, somehow
<yorickpeterse> I also have...
<yorickpeterse> 4 |*£|_|6i|\| 7h47 £375 |\/|3 |)0 7hi5
<yorickpeterse> and I have one that colours text, but that's sadly (or luckily) disabled in all channels I sit in
<yorickpeterse> and of course my most favourite one: rslap
* yorickpeterse slaps yorickpeterse around with a copy of "The Pragmatic Trout"
<yorickpeterse> comes with customizable "slaps"
<brixen> hah
<brixen> for i in channel; do ... done
<yorickpeterse> I also have a parse.y reference in the list, but it's almost never picked :/
GitHub10 has joined #rubinius
<GitHub10> [rubinius] YorickPeterse closed pull request #3175: Accept a nil argument for Array#flatten. (master...array-flatten-nil) http://git.io/CJJ4rg
GitHub10 has left #rubinius [#rubinius]
<yorickpeterse> 20 more selectors to go, by this rate that should be ~2 weeks
<yorickpeterse> which means I'm "only" 2,5 weeks overdue
<headius> pragmatic trout, nice
JohnBat26 has quit [Ping timeout: 245 seconds]
|jemc| has quit [Quit: WeeChat 1.0.1]
|jemc| has joined #rubinius
|jemc-bot| has quit [Remote host closed the connection]
|jemc-bot| has joined #rubinius
enebo has quit [Quit: enebo]
<yorickpeterse> well fuck
<yorickpeterse> I think I just wrapped up the parsing side of CSS selectors
<yorickpeterse> A whole bunch of selectors are specific to web browsers, so I can just drop those
<yorickpeterse> yup, boom done
<yorickpeterse> now I need to make sure that the generated XPath is actually correct
<yorickpeterse> so this basically reduced release time by ~1 week
|jemc| has quit [Ping timeout: 256 seconds]