elia has joined #rubinius
slaught has quit [Quit: slaught]
elia has quit [Quit: Computer has gone to sleep.]
amsi has quit [Quit: Leaving]
sbryant has quit [Ping timeout: 258 seconds]
amclain has joined #rubinius
sbryant has joined #rubinius
goyox86 has joined #rubinius
amclain has quit [Quit: Leaving]
sbryant has quit [Ping timeout: 258 seconds]
goyox86 has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
goyox86 has joined #rubinius
|jemc| has joined #rubinius
Bwild has joined #rubinius
<|jemc|> brixen: not sure what you need from me - you should be an "owner" on the dockerhub rubinius organization already
meh` has quit [Ping timeout: 240 seconds]
<brixen> |jemc|: ah ok
<brixen> |jemc|: docker noobie :)
<brixen> I'll work on getting this stuff set up with Travis so we can automate official PR stuff
<brixen> and stuff
goyox86 has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
<|jemc|> sure; let me know if there's anything you do end up needing from me - I'll do what I can in between family holiday stuff
sbryant has joined #rubinius
<brixen> |jemc|: thanks
nirvdrum has quit [Ping timeout: 240 seconds]
sbryant has quit [Ping timeout: 258 seconds]
sbryant has joined #rubinius
<brixen> cool, basic JIT specs going
<|jemc|> brixen: great :) - it'll be nice to have a good path for preventing regressions there
<brixen> indeed
<brixen> it's quite complicated to do this correctly
<brixen> this is a really quick hack
<brixen> but something to iterate on
<brixen> as a by-product, you can now 'Rubinius::JIT.sync = true; <some ruby code>; Rubinius::JIT.sync = false' :)
<brixen> wish I could finish this keyword crap but probably too tired tonight
<|jemc|> brixen: I was going to take a look at it next week, but saw you were already on it
<|jemc|> (the string kwargs thing)
<brixen> yeah, it was a super mess
<brixen> 2 files changed, 65 insertions(+), 6 deletions(-) more specs
<brixen> maybe I got the edge cases, maybe not :p
<|jemc|> I guess we'll find out when some other app or gem is depending on other oness
<brixen> heh, I guess so
GitHub181 has joined #rubinius
<GitHub181> [rubinius] brixen pushed 6 new commits to master: http://git.io/KaAL6w
<GitHub181> rubinius/master 1736dbd Brian Shirai: Updated CI specs to RubySpec 671ef6d3.
<GitHub181> rubinius/master 02d4e83 Brian Shirai: Added more keyword specs.
<GitHub181> rubinius/master 22cb693 Brian Shirai: Updated block keyword parsing to MRI 2.2.0 behavior.
GitHub181 has left #rubinius [#rubinius]
<brixen> still have to fix the JIT, but I'm guessing khaase's wacky gem isn't going to be running master before tomorrow
<|jemc|> hm... that's unsettling
<|jemc|> ruby exceptions raised from ruby code in my FFI::Functions used as callbacks passed to a C library are failing pretty darn silently
sbryant has quit [Ping timeout: 258 seconds]
sbryant has joined #rubinius
<|jemc|> yeah, I suppose it's to be expected - it's just unfortunate
<brixen> whoa, just refreshed on Travis and got a totally new design
<brixen> and what the hell is going on with os x vms
<jc00ke> wth?
<jc00ke> Weird, I thought maybe there would be more info in the full log file
<brixen> it's just sitting there for 25min trying to download llvm-3.5
<jc00ke> ENOTIMEOUTFORYOU
noop has joined #rubinius
|jemc| has quit [Ping timeout: 264 seconds]
travis-ci has joined #rubinius
<travis-ci> rubinius/rubinius/master (5575076 - Brian Shirai): http://travis-ci.org/rubinius/rubinius/builds/45006265: The build has errored.
travis-ci has left #rubinius [#rubinius]
noop has quit [Ping timeout: 252 seconds]
noop has joined #rubinius
noop has quit [Ping timeout: 256 seconds]
havenwood has quit []
Bwild has quit [Ping timeout: 244 seconds]
noop has joined #rubinius
noop has quit [Read error: Connection reset by peer]
noop has joined #rubinius
saline has quit [Ping timeout: 258 seconds]
saline has joined #rubinius
elia has joined #rubinius
elia has quit [Ping timeout: 244 seconds]
elia has joined #rubinius
elia has quit [Ping timeout: 256 seconds]
sbryant has quit [Ping timeout: 258 seconds]
kagaro has quit [Ping timeout: 244 seconds]
|Blaze| has quit [Ping timeout: 244 seconds]
<yorickpeterse> morning
<Gibheer> evening
sbryant has joined #rubinius
max96at|off is now known as max96at
<yopp> gdday!
<yorickpeterse> weeeeee finally having Oga actually do some XML
<yorickpeterse> Although I need to put a few better tracers in place to see how much time is spent in it
elia has joined #rubinius
elia has quit [Client Quit]
<yorickpeterse> hihi, only 1% spent in Oga
<yorickpeterse> 5ms of parsing xpath, 15ms of evaluation
<yorickpeterse> but for some reason I'm not even seeing the xml parsing timings
kagaro has joined #rubinius
<yopp> yorickpeterse, how much it's comparing to ox?
goyox86 has joined #rubinius
<yorickpeterse> nfi I'm not using that crap
<yorickpeterse> but perf wise it's ox -> nokogiri -> oga -> rexml
<yorickpeterse> (in a standalone benchmark that is)
goyox86 has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
<yorickpeterse> hm, looks like caching xpath might not be a bad idea
<yorickpeterse> but doing that thread-safe is a bit hairy
<yorickpeterse> bbl heading home, yay christmas
<yorickpeterse> not used to going home before 18:30 though :<
elia has joined #rubinius
nirvdrum has joined #rubinius
noop has quit [Ping timeout: 264 seconds]
noop has joined #rubinius
elia has quit [Quit: Computer has gone to sleep.]
elia has joined #rubinius
noop has quit [Quit: Leaving]
noop has joined #rubinius
elia has quit [Quit: Computer has gone to sleep.]
josh-k has joined #rubinius
elia has joined #rubinius
noop has quit [Ping timeout: 264 seconds]
noop has joined #rubinius
cremes has quit [Quit: cremes]
elia has quit [Quit: Computer has gone to sleep.]
<yorickpeterse> "Encoding::ConverterNotFoundError: code converter not found (ASCII-8BIT to ASCII-8BIT)" wtf
<yorickpeterse> Hm, apparently that's our String#scrub going haywire
cremes has joined #rubinius
josh-k has quit []
sshao has joined #rubinius
elia has joined #rubinius
<brixen> oh yay, String#scrub
<brixen> this is really surprising behavior
<brixen> it assigns passed elements in order and then jumps all the way to the last element for the kw
<brixen> I don't even know what to search redmine for
meh` has joined #rubinius
<brixen> hm, the only rationale I could see is something like this https://gist.github.com/brixen/30604b1c58090bc59c6c
<brixen> oh well, if it's a bug, someone will bitch, or not
GitHub78 has joined #rubinius
<GitHub78> [rubinius] brixen pushed 2 new commits to master: http://git.io/SnbPWw
<GitHub78> rubinius/master 0a547fd Brian Shirai: Some more JIT spec setup.
<GitHub78> rubinius/master 07171db Brian Shirai: A few more JIT specs.
GitHub78 has left #rubinius [#rubinius]
<brixen> sshao: pushed, I'll add the JIT predicates and stuff but you should be able to add specs
<brixen> sshao: tag them as usual if you need to
<sshao> awesome, i'll take a shot at that now
Bwild has joined #rubinius
elia has quit [Quit: Computer has gone to sleep.]
<yorickpeterse> brixen: this is Ruby, people will complain anyway :P
<yopp> yorickpeterse, you can safely reduce it to "people will complain anyway :P"
travis-ci has joined #rubinius
<travis-ci> rubinius/rubinius/master (07171db - Brian Shirai): http://travis-ci.org/rubinius/rubinius/builds/45040906: The build failed.
travis-ci has left #rubinius [#rubinius]
elia has joined #rubinius
<brixen> well, that's curious
<yorickpeterse> str_scrub calls rb_str_scrub which calls str_scrub0
<yorickpeterse> ._.
elia has quit [Quit: Computer has gone to sleep.]
<yorickpeterse> I fear for the day there will be a str_scrub1 and str_scrub
<yorickpeterse> * str_strub2
<yorickpeterse> ffff
<yorickpeterse> scrub
* yorickpeterse ruined the joke anyway
* yopp scrubs his back
<yorickpeterse> Hm, looks like String#scrub just returns a copy of self for ASCII input
amsi has joined #rubinius
slaught has joined #rubinius
cremes has quit [Ping timeout: 240 seconds]
GitHub13 has joined #rubinius
<GitHub13> [rubinius] YorickPeterse pushed 2 new commits to master: http://git.io/VI5hAg
<GitHub13> rubinius/master 5714923 Yorick Peterse: Added specs for String#scrub with ASCII/BINARY.
<GitHub13> rubinius/master 07f71e4 Yorick Peterse: Fix usage of String#scrub with BINARY input....
GitHub13 has left #rubinius [#rubinius]
cremes has joined #rubinius
<sshao> brixen: actually, it looks like the JIT issues i wanted to write specs for only happened on the 1.8.7 branch
<brixen> sshao: ah, ok, I'll merge this to 1.8.7 shortly
<sshao> sweet
<brixen> it will have to be version specific, especially around the argument passing
<brixen> yorickpeterse: hm, String#scrub might be a perfect experiment for some new bytecode
<yorickpeterse> Hmm....I seem to've fixed our remaining String#scrub failure (which was tagged) in the most bizarre way
<yorickpeterse> force_encoding all the things
<yorickpeterse> granted we should probably redo the thing without the usage of Encoding::Converter
<brixen> yeah, force_encoding is magic fairy dust for pushing the problem along the way :p
<yorickpeterse> well so in this case the problem is pretty wack:
cremes has quit [Ping timeout: 245 seconds]
<yorickpeterse> the buffer string used during replacement is in BINARY
<yorickpeterse> String#scrub takes a block, which in this spec returns a string in UTF8
<yorickpeterse> That can't be encoded to BINARY without errors
<yorickpeterse> unless you use the wonders of force_encoding
<yorickpeterse> At the very end the buffer is encoded into the correct encoding again
<yorickpeterse> also using force_encoding :P
<brixen> yep, fun fun
<yorickpeterse> brixen: do spec tag removals go in a separate commit as well?
<brixen> they don't have to
<brixen> you don't have to worry about specs in a separate commit either
<brixen> git fp spec/ruby sorts it out
<yorickpeterse> meh I'll do it separately, means I have more commits
<yorickpeterse> which means I'm contributing more :D
* yorickpeterse runs
<brixen> yeah, pump that commit number
<brixen> put each character in a separate commit
<brixen> obviously valid metric
<brixen> hilariously, I've had so many conversations about this in a job context
<brixen> everyone admits that, say, attaching bonus to number of commits would be totally bogus
GitHub87 has joined #rubinius
<GitHub87> [rubinius] YorickPeterse pushed 2 new commits to master: http://git.io/TDNgcg
<GitHub87> rubinius/master 4ff363a Yorick Peterse: Removed spec tags for String#scrub.
<GitHub87> rubinius/master 64f69d3 Yorick Peterse: Fixed String#scrub with custom block encodings....
GitHub87 has left #rubinius [#rubinius]
<yorickpeterse> oh yeah, I had similar talks with co-workers going "damn Yorick commits a lot of stuff"
<brixen> but you get "contributor status" from # of commits
<brixen> lulz
<yorickpeterse> Then I tell them I commit a lot because I just fuck things up
havenwood has joined #rubinius
<yopp> instead of stash use branch and push it server
<yopp> valid way to get a lot of commit points!
<yorickpeterse> In all seriousness, I like splitting things up as it keeps my brain sane when looking back at it
travis-ci has joined #rubinius
<travis-ci> rubinius/rubinius/master (07f71e4 - Yorick Peterse): http://travis-ci.org/rubinius/rubinius/builds/45043944: The build is still failing.
travis-ci has left #rubinius [#rubinius]
<yorickpeterse> :<
<brixen> hold on, pushing a fix
<brixen> if someone wasn't so eager to game commit numbers :p
<yorickpeterse> huh, a segfault
<yorickpeterse> heh
<yorickpeterse> well something tells me you don't really need to worry about that statistic :P
<brixen> exactly
<brixen> so efforts to do so will only result in pain and suffering
<brixen> heh
<brixen> should have released 2.4.2 without fixing keywords
<brixen> oh well
* yopp filled up build queue and now can go and get some grilled stuff
<yopp> I don't know, but installing dependencies on our CI takes 99% of time.
<yorickpeterse> yopp: heh, we cache gems for that reason
<yorickpeterse> well "cache", we just don't uninstall them :P
<yopp> We had some issues with that back then, so we decided to rebuild stuff from scratch each time. But seems like our custom-made gem server is now a bottleneck
<yopp> Hell knows why. :|
GitHub186 has joined #rubinius
<GitHub186> [rubinius] brixen pushed 2 new commits to master: http://git.io/fzPpnA
<GitHub186> rubinius/master 9bc6050 Brian Shirai: Added temporary tags for JIT specs.
<GitHub186> rubinius/master dc5b82f Brian Shirai: Re-removed obsoleted specs....
GitHub186 has left #rubinius [#rubinius]
travis-ci has joined #rubinius
<travis-ci> rubinius/rubinius/master (64f69d3 - Yorick Peterse): http://travis-ci.org/rubinius/rubinius/builds/45044711: The build is still failing.
travis-ci has left #rubinius [#rubinius]
<brixen> hm, we could use new Travis container builds if not for LLVM
<brixen> it's annoying that sudo is needed when sudo is actually not needed
<brixen> yorickpeterse: sorry, should have pushed those commits right away, would have saved your good name from the tarnish of failed builds :)
travis-ci has joined #rubinius
<travis-ci> rubinius/rubinius/master (dc5b82f - Brian Shirai): http://travis-ci.org/rubinius/rubinius/builds/45045436: The build was fixed.
travis-ci has left #rubinius [#rubinius]
<yorickpeterse> brixen: heh
<brixen> possibly tasty
<brixen> may have to experiment with using this for a Rubinius distribution channel
<yopp> It is based on node.js and Chromium
<yopp> oh my
<brixen> indeed
<brixen> your OS is based on C
<brixen> everything can't be great
<brixen> ;)
<yorickpeterse> https://gist.github.com/YorickPeterse/8324bbaf8f165967bcb4 this grammar is starting to look like something
<yorickpeterse> I think I now managed to get all first/first conflicts out
<yorickpeterse> doesn't parse everything it should parse though
<yorickpeterse> (e.g. still need to add a bunch of operators)
<yorickpeterse> and I'm not sure about the epsilon syntax ("...")
<yorickpeterse> Toying with the idea of "_" but that looks equally stupid
<yorickpeterse> using the +, *, ? operators you'd be able to write this instead: https://gist.github.com/YorickPeterse/e18fe311ec3a82a5e4cf
<yorickpeterse> soon ™
noop has quit [Ping timeout: 244 seconds]
|jemc| has joined #rubinius
havenwood has quit [Remote host closed the connection]
havenwood has joined #rubinius
nirvdrum has quit [Ping timeout: 265 seconds]
<brixen> yorickpeterse: neat
lopex has quit [Quit: Connection closed for inactivity]
kagaro has quit [Ping timeout: 250 seconds]
jfredett-w1 has quit [Ping timeout: 245 seconds]
diegoviola has joined #rubinius
havenwood has quit [Remote host closed the connection]
kagaro has joined #rubinius
GitHub197 has joined #rubinius
<GitHub197> [rubinius] brixen pushed 2 new commits to master: http://git.io/hFX_Tg
<GitHub197> rubinius/master 2efb85d Brian Shirai: Small keyword tweak.
<GitHub197> rubinius/master eec1ad3 Brian Shirai: Updated Gemfile.lock and gems_list.txt.
GitHub197 has left #rubinius [#rubinius]
<brixen> happy holidays folks
<brixen> if I get the JIT fixed for keywords this weekend, I'll drop 2.4.2
josh-k has joined #rubinius
<jc00ke> happy holidays :)
travis-ci has joined #rubinius
<travis-ci> rubinius/rubinius/master (eec1ad3 - Brian Shirai): http://travis-ci.org/rubinius/rubinius/builds/45055708: The build passed.
travis-ci has left #rubinius [#rubinius]
slaught has quit [Quit: slaught]
diegoviola has quit [Ping timeout: 252 seconds]
havenwood has joined #rubinius
johnmuhl has quit [Quit: Connection closed for inactivity]
max96at is now known as max96at|off
Bwild has quit [Ping timeout: 258 seconds]
sshao has quit []
diegoviola has joined #rubinius
diegoviola has quit [Quit: WeeChat 1.0.1]
diegoviola has joined #rubinius
Bwild has joined #rubinius
diegoviola has quit [Ping timeout: 258 seconds]
josh-k_ has joined #rubinius