<chrisseaton>
Antiarc: those comments all talk about self being a method - I thought it was a keyword isn't it? That's how I've implemented it in Truffle.
<Antiarc>
chrisseaton: It is a keyword, the comments talk about using it as a method
triple_b has joined #jruby
<Antiarc>
The resolution was to use #itself as a method which returns the receiver
SynrG has quit [Quit: No Ping reply in 180 seconds.]
baroquebobcat has quit [Quit: baroquebobcat]
SynrG has joined #jruby
johnsonch_afk is now known as johnsonch
<Antiarc>
headius: The open3 errors are because AbstractSelectableChannel is being passed SelectionKey.OP_WRITE | SelectionKey.OP_CONNECT, when it only accepts SelectionKey.OP_WRITE. Do you have an immediate intuition of what the problem should be, or should I dig in?
SynrG has quit [Quit: No Ping reply in 180 seconds.]
<Antiarc>
Hm, this is being called from the capture stuff, which should be read-only
johnsonch is now known as johnsonch_afk
SynrG has joined #jruby
iamjarvo has joined #jruby
josh-k has quit [Remote host closed the connection]
phrinx_ has quit [Remote host closed the connection]
josh-k has joined #jruby
<Antiarc>
The channel is a sun.nio.ch.SinkChannelImpl, which defines validOps() as SelectionKey.OP_WRITE, but OpenFile attempts to register it with OP_WRITE | OP_CONNECT and it's barfing
josh-k has quit [Ping timeout: 256 seconds]
x1337807x has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
<diegoviola>
will jruby run fine on jre8-openjdk?
<rtyler>
yep
* rtyler
is all JDK8 for dev
robbyoconnor has quit [Ping timeout: 245 seconds]
iamjarvo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
josh-k has joined #jruby
Haifeng has quit [Ping timeout: 245 seconds]
Haifeng has joined #jruby
yfeldblum has quit [Remote host closed the connection]
fivebats has joined #jruby
yfeldblum has joined #jruby
diegovio1 has joined #jruby
diegoviola is now known as Guest26343
diegovio1 is now known as diegoviola
Guest26343 has quit [Ping timeout: 255 seconds]
<jc00ke>
Hi there
<jc00ke>
Is there a findAll-type predicate that's usuable?
<jc00ke>
I'm thinking of predicates like what I remember from C#, years ago
josh-k has quit [Remote host closed the connection]
<Antiarc>
Maybe it's just noobishness on my part, but I'm not sure what you mean
<jc00ke>
But I instead just did it the long way, nbd.
<jc00ke>
I'm not familiar with Java, but I thought later versions had generics and predicates like C# had bitd
subbu has joined #jruby
<Antiarc>
jc00ke: Looks like Guava has it, but jruby-core doesn't depend on Guava. Java 8 has 'em, too, looks like
<jc00ke>
Antiarc: yeah, that's what I figured. No biggie. Thanks for looking!
<Antiarc>
The lack of that stuff in Java 7 drives me absolutely insane
<Antiarc>
Every time I end up wanting to do something like that I go looking for it and then get sad when I find out it's not available on my target platforms!
<diegoviola>
how do I install stuff with maven, anyone please?
<Antiarc>
Usually just "mvn"
<diegoviola>
mvn install foo?
<Antiarc>
mvn will usually run out and download declared dependencies
<Antiarc>
you just add that snippet to your POM and maven should take care of it + dependencies
<Antiarc>
headius: down to 89 failures locally!
havenwood has quit [Remote host closed the connection]
<diegoviola>
okay
<diegoviola>
ty
<Antiarc>
20 of those failures are TracePoint stuff which I'm not gonna touch because I'm pretty sure it would require me to sacrifice a goat or something
havenwood has joined #jruby
triple_b has quit [Remote host closed the connection]
triple_b has joined #jruby
triple_b has quit [Remote host closed the connection]
triple_b has joined #jruby
tlarevo has joined #jruby
tlarevo has quit [Remote host closed the connection]
tlarevo has joined #jruby
robbyoconnor has joined #jruby
robbyoconnor has quit [Client Quit]
triple_b has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
x1337807x has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] subbuss pushed 1 new commit to master: http://git.io/tfRIXw
<JRubyGithub>
jruby/master 0845943 Subramanya Sastry: Fix #2098: Lambdas are yielded to differently in ruby 2.2 vs 2.1...
JRubyGithub has left #jruby [#jruby]
<Antiarc>
\o/
<subbu>
:)
<subbu>
Antiarc till you report some other failure :)
<Antiarc>
subbu: I'll test it against update_stdlib shortly
<subbu>
k
<Antiarc>
There are a few 2.2 stdlib tests that were what exposed that
tenderlove has quit [Remote host closed the connection]
mister_solo has joined #jruby
subbu has quit [Ping timeout: 250 seconds]
rsim has joined #jruby
mister_solo has quit [Ping timeout: 272 seconds]
x1337807x has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
baroquebobcat has joined #jruby
JohnBat26 has joined #jruby
noop has joined #jruby
tenderlove has joined #jruby
robbyoconnor has joined #jruby
kares has quit [Ping timeout: 256 seconds]
<tarcieri>
I am still weirded out by djb wearing my t-shirt
<tarcieri>
lol
frobs has joined #jruby
tenderlove has quit [Ping timeout: 244 seconds]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] headius pushed 2 new commits to update_stdlib: http://git.io/9EDGaA
<JRubyGithub>
jruby/update_stdlib f488a44 Chris Heald: Implement Method#curry, which simply delegates to Proc#curry....
<JRubyGithub>
jruby/update_stdlib c723013 Charles Oliver Nutter: Merge pull request #2133 from cheald/method_curry...
JRubyGithub has left #jruby [#jruby]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] headius pushed 2 new commits to update_stdlib: http://git.io/Hfuwdw
<JRubyGithub>
jruby/update_stdlib 3c56776 Chris Heald: Exclude a couple of tests which depend on MRI-specific behavior
JRubyGithub has left #jruby [#jruby]
<JRubyGithub>
jruby/update_stdlib 17a3c17 Charles Oliver Nutter: Merge pull request #2134 from cheald/hash_exclusions...
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] headius pushed 2 new commits to update_stdlib: http://git.io/VFenMg
<JRubyGithub>
jruby/update_stdlib c40909c Chris Heald: Always taint StringIO by it input. Fixes ruby-bug 9769.
<JRubyGithub>
jruby/update_stdlib 80ec619 Charles Oliver Nutter: Merge pull request #2135 from cheald/fix_stringio_infect...
JRubyGithub has left #jruby [#jruby]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] headius pushed 2 new commits to update_stdlib: http://git.io/X7ozcQ
<JRubyGithub>
jruby/update_stdlib 7cbd2e5 Chris Heald: Implement Object#itself, per ruby-feature 6373
<JRubyGithub>
jruby/update_stdlib 08df7ae Charles Oliver Nutter: Merge pull request #2136 from cheald/object_identity...
JRubyGithub has left #jruby [#jruby]
kares has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] headius pushed 2 new commits to update_stdlib: http://git.io/bnbDAg
<JRubyGithub>
jruby/update_stdlib 5145b47 Chris Heald: Fix ruby-bug 8625 and ruby-bug 9847...
<JRubyGithub>
jruby/update_stdlib 50c5434 Charles Oliver Nutter: Merge pull request #2138 from cheald/fix_io_buffers...
JRubyGithub has left #jruby [#jruby]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] headius pushed 2 new commits to update_stdlib: http://git.io/qUfbhg
<JRubyGithub>
jruby/update_stdlib 650b075 Chris Heald: Don't swallow missing user/group errors for Etc.getgrnam/getpwname
<JRubyGithub>
jruby/update_stdlib 723686d Charles Oliver Nutter: Merge pull request #2140 from cheald/fix_chmod_chgrp...
JRubyGithub has left #jruby [#jruby]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] headius pushed 1 new commit to update_stdlib: http://git.io/ZVZztw
<JRubyGithub>
jruby/update_stdlib 8335011 Charles Oliver Nutter: Merge remote-tracking branch 'origin/master' into update_stdlib
JRubyGithub has left #jruby [#jruby]
<headius>
well, let's see where we are now
mister_solo has joined #jruby
baroquebobcat has quit [Quit: baroquebobcat]
justinmcp_ has quit []
<Antiarc>
I just added another PR to correct the behavior of RubyIO#ensureYieldClose, but the test that it's associated with is still failing. I think there are multiple failures at play there.
<Antiarc>
But it un-swallows a bunch of exceptions, which exposed another broken test, which I then fixed.
<JRubyGithub>
[jruby] mkristian force-pushed uniform-jruby-home from 86aab7c to af5f0a7: http://git.io/ScsKNw
<JRubyGithub>
jruby/uniform-jruby-home 476acae Christian Meier: run tests/jruby.index with jruby-complete
<JRubyGithub>
jruby/uniform-jruby-home d8592b3 Christian Meier: uniform setting path to jruby home - no detection just use the one from the classloader
<JRubyGithub>
jruby/uniform-jruby-home af5f0a7 Christian Meier: warn when jruby home is not set to one of the tested cases
JRubyGithub has left #jruby [#jruby]
toshe_tm has quit [Read error: Connection reset by peer]
toshetm has quit [Read error: Connection reset by peer]
bbrowning has joined #jruby
toshe_tm has joined #jruby
toshetm has joined #jruby
toshetm has quit [Read error: Connection reset by peer]
toshe_tm has quit [Read error: Connection reset by peer]
toshetm has joined #jruby
toshe_tm has joined #jruby
toshetm has quit [Read error: Connection reset by peer]
toshe_tm has quit [Read error: Connection reset by peer]
toshetm has joined #jruby
toshe_tm has joined #jruby
anaeem1 has quit [Remote host closed the connection]
mister_solo has quit [Ping timeout: 255 seconds]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] mkristian pushed 1 new commit to master: http://git.io/JJzFug
<JRubyGithub>
jruby/master a8eb2bf Christian Meier: run tests/jruby.index with jruby-complete...
JRubyGithub has left #jruby [#jruby]
JRubyGithub has joined #jruby
<JRubyGithub>
jruby/uniform-jruby-home 2a159b6 Christian Meier: warn when jruby home is not set to one of the tested cases
<JRubyGithub>
jruby/uniform-jruby-home adcbdc7 Christian Meier: uniform setting path to jruby home - no detection just use the one from the classloader
<JRubyGithub>
[jruby] mkristian force-pushed uniform-jruby-home from af5f0a7 to 2a159b6: http://git.io/ScsKNw
JRubyGithub has left #jruby [#jruby]
etehtsea has joined #jruby
mister_solo has joined #jruby
toshe_tmtm has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] eregon pushed 1 new commit to master: http://git.io/diqkBw
<JRubyGithub>
jruby/master 5d93cdc Benoit Daloze: [Truffle] Improve BooleanCastNode so to better reflect its cousin RubyContext.isTruthy()....
JRubyGithub has left #jruby [#jruby]
yfeldblum has quit [Remote host closed the connection]
toshetm has quit [Ping timeout: 260 seconds]
toshe_tm has quit [Ping timeout: 260 seconds]
toshetm has joined #jruby
triple_b has joined #jruby
toshe_tmtm has quit [Ping timeout: 240 seconds]
nirvdrum has quit [Quit: Leaving]
lance|afk is now known as lanceball
tcrawley-away is now known as tcrawley
yfeldblum has joined #jruby
tenderlove has joined #jruby
yfeldbl__ has joined #jruby
yfeldblum has quit [Ping timeout: 264 seconds]
tbuehlmann has joined #jruby
<headius>
usebrn_: leak?
rsim has quit [Quit: Leaving.]
<headius>
Antiarc: still no word from Johann unfortunately :-\
tenderlove has quit [Ping timeout: 272 seconds]
yfeldbl__ has quit [Ping timeout: 258 seconds]
<headius>
usebrn_: your suggested fix seems like a good one
<headius>
JRuby master no longer uses a static map of files so the problem should not exist there
deobalds has quit [Quit: Computer has gone to sleep.]
kares has quit [Quit: Ex-Chat]
rsim has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] headius pushed 2 new commits to update_stdlib: http://git.io/Pr_lCw
<JRubyGithub>
jruby/update_stdlib 5a2d482 Chris Heald: Update RubyIO#ensureYieldClose to match ruby-bug 10153 behavior...
<JRubyGithub>
jruby/update_stdlib d183f8d Charles Oliver Nutter: Merge pull request #2141 from cheald/update_io_ensure_close...
JRubyGithub has left #jruby [#jruby]
pglombar_ has joined #jruby
pglombar_ has quit [Client Quit]
tlarevo_ has quit [Remote host closed the connection]
tlarevo has joined #jruby
tlarevo has quit [Ping timeout: 265 seconds]
baroquebobcat has joined #jruby
joast has quit [Ping timeout: 244 seconds]
vyorkin has joined #jruby
johnsonch_afk is now known as johnsonch
yfeldblum has joined #jruby
e_dub has quit [Quit: ZZZzzz…]
mkristian has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] headius pushed 3 new commits to update_stdlib: http://git.io/2xCA3g
<JRubyGithub>
jruby/update_stdlib b4ff008 Chris Heald: Add Method#super_method
<JRubyGithub>
jruby/update_stdlib 7359d48 Charles Oliver Nutter: Merge branch 'super_method' of github.com:cheald/jruby into cheald-super_method...
<JRubyGithub>
jruby/update_stdlib 53a43ca Charles Oliver Nutter: Merge branch 'cheald-super_method' into update_stdlib
<mkristian>
headius, no this was not my doing - I just added a test for jruby-complete which fails and is allowed to fail. there was one commit before which probably fixed master
<headius>
mkristian: can you make that fix for both branches?
<headius>
subbu: yes...we could have just excluded the lot but there's a bunch of 2.2 features we were able to knock down
<mkristian>
headius, on jruby-1_7 we do not overwrite loadClass so only master ;)
<headius>
or rather, Antiarc was able to knock down...he's done most of it
<headius>
mkristian: oh, so this would not exhibit on 1.7 most likely
<headius>
ok
<mkristian>
headius, and on jdk6 it might not even work
<headius>
ahh right, that method was added for 7
<headius>
I read through the Java issue that prompted their change
mister_solo has quit [Ping timeout: 244 seconds]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] chrisseaton pushed 1 new commit to truffle-head: http://git.io/qaYvoQ
<JRubyGithub>
jruby/truffle-head 7192c00 Chris Seaton: Merge branch 'master' into truffle-head...
JRubyGithub has left #jruby [#jruby]
triple_b has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
vyorkin has joined #jruby
vtunka has quit [Quit: Leaving]
mkristian has quit [Ping timeout: 255 seconds]
drbobbeaty has quit [Read error: Connection reset by peer]
drbobbeaty_ has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] mkristian pushed 1 new commit to master: http://git.io/SB7zjA
<JRubyGithub>
jruby/master ac9f2eb Christian Meier: synchronized class loading properly - fixes #2108...
JRubyGithub has left #jruby [#jruby]
<headius>
oh, and then he times out :-(
<headius>
I wonder if that will fix the j2ee suite
<headius>
same kind of exception being raised
triple_b has joined #jruby
vyorkin has quit [Ping timeout: 255 seconds]
triple_b has quit [Client Quit]
Aethenelle has joined #jruby
yfeldblum has joined #jruby
ReactiveIO has joined #jruby
yfeldblum has quit [Ping timeout: 240 seconds]
benlovell has quit [Ping timeout: 250 seconds]
triple_b has joined #jruby
ReactiveIO has quit [Remote host closed the connection]
ReactiveIO has joined #jruby
mkristian has joined #jruby
vyorkin has joined #jruby
Aethenelle_ has joined #jruby
Aethenelle has quit [Ping timeout: 255 seconds]
Aethenelle_ is now known as Aethenelle
<Aethenelle>
hrm... the truffle stuff breaking compilation for anyone else on master?
tlarevo has joined #jruby
<headius>
hmm, I'm not on master at the moment
vyorkin has quit [Ping timeout: 245 seconds]
<Aethenelle>
... according to travis... that's a no...
Xzyx987X has quit [Ping timeout: 265 seconds]
<Aethenelle>
i forget... is there a maven task to clear out everything?
<subbu>
mvn clean?
Xzyx987X has joined #jruby
<headius>
and git clean has several levels of destructiveness
<Aethenelle>
i though mvn clean only nuked the class and jar files in target
pglombardo has joined #jruby
<Aethenelle>
huh... it was a stray duplicated mthod in RubyModule... weird
ReactiveIO has quit [Remote host closed the connection]
triple_b has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
triple_b has joined #jruby
ReactiveIO has joined #jruby
anaeem1 has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] headius pushed 3 new commits to update_stdlib: http://git.io/mOTJrw
<JRubyGithub>
jruby/update_stdlib bea6e70 Charles Oliver Nutter: Fix excludes logic for included modules....
<JRubyGithub>
jruby/update_stdlib a65e66f Charles Oliver Nutter: Update tags for #2142.
<JRubyGithub>
jruby/update_stdlib 8f628ed Charles Oliver Nutter: Add frozen string optz and improve string handling in JIT....
JRubyGithub has left #jruby [#jruby]
<headius>
Aethenelle: bad merge?
iamjarvo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<headius>
lots of fingers in the pie the past week
<Aethenelle>
headius: only on my end...
<headius>
no problem
<headius>
ok, frozen string sidebar over, back to greening up 2.2 stdlib
<Aethenelle>
rebased the other prpend impl onto the current master so i can fix it up while i work on a patch for MRI
<chrisseaton>
Aethenelle: need any help?
anaeem1 has quit [Remote host closed the connection]
<headius>
chrisseaton: when are you going to be in SD?
<headius>
enebo and I are there from Sunday to Friday I believe
<Aethenelle>
chrisseaton: no, the current(correct) impl was merged yesterday... just needs some cleanup... it's the new one that got a bit borked by the recent JIT stuff...
<Aethenelle>
thanks though
<chrisseaton>
I get in on Saturday, so have Sunday free at the moment, and leave early Thursday (SF on Friday)
<Aethenelle>
headius: you live in SD?
<headius>
stopping in at the mothership Friday?
<Aethenelle>
i was just there...
<headius>
Aethenelle: no, going there for RubyConf week after next
<Aethenelle>
ahh.... that makes more sense...
<headius>
enebo and I are both firmly based in the Minneapolis area
<Aethenelle>
it was a little weird having 70-80 degree weather in october the whole time i was there.
<headius>
Antiarc: I'll have to talk to enebo about how to handle this in licensing, but with their only requirement being attribution (2-clause BSD I think we're ok
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] headius pushed 2 new commits to update_stdlib: http://git.io/13SYdw
<JRubyGithub>
jruby/update_stdlib 2f39afa Chris Heald: Add direct BubbleBabble power from OpenSSH
<JRubyGithub>
jruby/update_stdlib 5a42b1b Charles Oliver Nutter: Merge pull request #2144 from cheald/clean_bubblebabble...
JRubyGithub has left #jruby [#jruby]
<Antiarc>
Yup!
<headius>
finishing up a half-assed Etc.uname
<headius>
I don't feel like binding with FFI right now
<Antiarc>
I kept meaning to get to that one
<Antiarc>
Haha, yup
<headius>
uname struct is probably different across revisions of frickin libc
<headius>
the windows version of it might be easier
<headius>
hahah that's awesome
<subbu>
oh rubyconf is next week, eh?
<Aethenelle>
from the man page, it looks like it's just a few pointers and part of POSIX.1 ... shouldn't be too bad...
<headius>
subbu: week after
<headius>
next week we write presentations, finish up preview release, and have team meeting in NC :-D
<subbu>
ok, so 1 week to preview then?
<headius>
well, Red Hat team meeting I mean
<headius>
yeah about that
<headius>
enebo probably isn't excited about releasing on the road, but hey
<headius>
Aethenelle: whew
<subbu>
:)
<headius>
I'll add a bug and maybe mark it as beginner (for a beginner familiar with C functions and Java, anyway)
<Aethenelle>
you should be able to edit the rakefile generating that stuff to use auto layout instead which might make it way easier than it currently is...
marr has quit [Ping timeout: 250 seconds]
<headius>
Aethenelle: if we do it in Ruby that would work
<headius>
really all of etc could (and probably should) be in ruby with FFI
<headius>
rbx impl might be usable...a new version wouldn't look much different
Aethenelle_ has joined #jruby
<headius>
oh, but unfortunately rubysl-etc is one of the rubinius-specific rubysl gems
<headius>
I guess we do our own then
Aethenelle has quit [Ping timeout: 245 seconds]
Aethenelle_ is now known as Aethenelle
<Antiarc>
`"foo-bar": 2` as a valid hash key is a parser change, yeah?
<headius>
yes
<headius>
enebo was working on it
<Antiarc>
k, not touching that one
noop has quit [Ping timeout: 258 seconds]
GregMefford has joined #jruby
<headius>
might not be able to push until he's back from killing forest creaturs
<headius>
I'm guessing the deer shack doesn't have internet
<subbu>
headius, aha .. that is where he is , eh? :)
<headius>
indeed...yearly trip with dad, I keep meaning to go but it never lines up right
<headius>
November sucks
<subbu>
k
pglombar_ has joined #jruby
<headius>
subbu: I'm going to land stdlib today so you'll be able to pull latest of everything before you leave
pglombardo has quit [Quit: #rubylang]
<headius>
we're poking at the last few dozen failures now
<dfr|work>
seems to me that the only way to get it if self is null, which seems that things are really out of whack for whatever reason =/
auxbuss has quit [Quit: I'm gone.]
<headius>
dfr|work: that's exactly what it means, and that would happen if some other code sent a null into the call chain without expecting too
<Antiarc>
Should be easy enough to track down via debugger, no?
<Antiarc>
have a repro case?
<headius>
usually that's a bad core method that got a null when it didn't expect it...but it's a serious pain to find these sometimes
<Antiarc>
I'm already set up for debugging
<headius>
you'll kinda have to trace back through the call stack and see what this value SHOULD have been and figure out why it wasn't
<Antiarc>
yup
<dfr|work>
Antiarc, not an easy repro case, hehe. It's blowign up deep in my ruby code... I'm likely at fault, I'm just trying to figure out how to debug it
<Antiarc>
Have you seen intellij 14's new inline debug values?
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] headius pushed 2 new commits to update_stdlib: http://git.io/9JidwA
<JRubyGithub>
jruby/update_stdlib 10f7229 Charles Oliver Nutter: Add minor deprecation warning in Comparable.
<JRubyGithub>
jruby/update_stdlib 9576d09 Charles Oliver Nutter: Implement a half-assed Etc.uname....
JRubyGithub has left #jruby [#jruby]
<dfr|work>
Antiarc, also... I don't think I cannot easily attach a debugger :(
<dfr|work>
headius, okay, thanks =/
<Antiarc>
dfr|work: intellij has remote debugging, super easy :)
<dfr|work>
it's a bit hard to trace what's going on 'cause the rest of stack is a bunch of ruby parsing methods, naturally :(
<dfr|work>
Antiarc, I'm not going down that rabbit hole ;)
<headius>
lopex: how do we update the unicode tables in jcodings?
<Antiarc>
you just launc your process with -J-agentlib:jdwp=transport=dt_socket,server=y,suspend=n,address=5005
<Antiarc>
and then intellij can connect to it and debug it
<headius>
2.2 did an update to unicode 7 (I think?) and I suspect some of our failures are because of new code points
<Antiarc>
it was surprisingly painless
<headius>
Antiarc: I hate that flag...it's like it's designed to be unfriendly
<Antiarc>
headius: Same, except for every Java flag ever
bbrowning_away is now known as bbrowning
<dfr|work>
Antiarc, that's after I spend multiple hours setting up intelliJ and waiting for it to index my huge java codebase ;)
<headius>
have wanted to add a --remote-debug=port to jruby launcher for a long time
<headius>
or similar
<lopex>
headius: via scripts, both encoding and folding tables ?
<headius>
lopex: ideally everything
<lopex>
headius: those two scripts
<lopex>
headius: and a mri master
GitHub105 has joined #jruby
<GitHub105>
[jcodings] headius merged central_loop into master: http://git.io/rs3sTg
GitHub105 has left #jruby [#jruby]
<headius>
I didn't realize I hadn't merged that
<headius>
lopex: ok, easy
<lopex>
headius: reading the tables from unicode.org was too risky given they change everything everywhere
zeroecco has joined #jruby
iamjarvo has joined #jruby
<headius>
that's fine...it looks like the import scripts still work ok
pglombardo1 has joined #jruby
<lopex>
headius: there' quite a few assersions there
<headius>
ah-ha, and I see some EUC-KR changes that might just be what I need
iamjarvo has quit [Client Quit]
<headius>
lopex: I'm going to spin jcodings release 1.1, since it has working transcoder loop now
<headius>
or should it be 2.0?
pglombardo has quit [Ping timeout: 264 seconds]
<lopex>
headius: I guess it could be for 1.1x ?
<lopex>
oh jcodings version
<headius>
eah
<lopex>
I dont know ?
<lopex>
there's been a lot of changes right ?
robbyoconnor has quit [Ping timeout: 245 seconds]
<headius>
jcodings updates in 1.7 would be fine too, I didn't do anything to code 1.7 uses except fixes
<headius>
like the entire transcoder loop state machine thingy
<lopex>
headius: all the "special" transcoders implemented ?
<headius>
everything
<lopex>
you're a machine
<headius>
majority of the MRI transcoder tests pass now
<headius>
well, I'm a porting machine like you, but I'm proud of my central loop
<headius>
that took a lot of fucking work
<headius>
SO great that 9k won't have to pass all transcoding through char[] anymore
<lopex>
nuts
<lopex>
jcodings is 1.0.12-SNAPSHOT now
postmodern has joined #jruby
<headius>
hell, I'm going 2.0
<headius>
it's a major rev
<lopex>
headius: what about keeping utf-16 for strings leaking from java integration ?
<headius>
1.0 was just encoding tables, really, and 2.0 is transcoding as well
<lopex>
ok
<headius>
lopex: transcode to UTF-16 when Java string enters Ruby, you mean?
<lopex>
yeah
<Antiarc>
headius: re: string#scrub, I've been using https://github.com/knu/ruby-unf in my jruby projects to provide it. Is java.text.Normalizer something that we could use to bootstrap that?
<lopex>
headius: wrt that torqbox discussion
<Antiarc>
(I'm looking through all the 9k milestone tickets)
<lopex>
it has also about hashing right ?
<lopex>
Antiarc: is that a lot of code in mri ?
<subbu>
headius, i guess you saw .. but the LJE issue is an interesting bug .. but probably a real edge case.
<lopex>
wrt scrub ?
<Antiarc>
lopex: I haven't looked, tbh
<Antiarc>
I've been avoiding String stuff because all the encoding stuff gives me the heebie jeebies
<lopex>
Antiarc: might be easier just to port that
toshe_tm has quit [Ping timeout: 250 seconds]
<headius>
Antiarc: String#scrub should be the same as MRI now
<Antiarc>
Oh, cool
<subbu>
fixing that properly requires an explicit LJE test prior to cloned ensure bodies as far as i can tell.
<headius>
subbu: yeah super edge-casey, but I didn't want it to get away
<headius>
Antiarc: oh, yup
<headius>
go for it
<lopex>
headius: what the ratio for transcoder stuff passing ?
<Antiarc>
I can't close the ones that aren't mine :P
<headius>
lopex: yeah it would be interesting to slip the switch and make Java strings encode as UTF-16, just to see if anything out there breaks
<headius>
if it works, we could unsafe copy the chars directly into a byte[]
<lopex>
yeah
<headius>
cheapest transcoding evar
<lopex>
and cow of course
<headius>
ratio for transcoder stuff passing... 99%
<lopex>
ah, copy
<headius>
I think there's a couple oddities I didn't track down
<lopex>
I meant directly stealing it :)
<headius>
lopex: that would be pretty wild if possible
<lopex>
headius: and no there's no begin/index in 1.7 since update 05
<headius>
it sucks that all the methods to access char[] from String make a copy too
<lopex>
*in java 7
<headius>
lopex: oh right, they eliminated shared backing store for strings didn't they
<lopex>
headius: now java copies all strings even on substrig etc
<headius>
a little bird told me they're trying to do a JEP to use byte[] inside String in OpenJDK
<lopex>
why ?
tlarevo has quit [Remote host closed the connection]
<headius>
probably for the same reasons we want it...reduced overhead reading utf-8 off the wire
<headius>
we're just way ahead of the JVM
tlarevo has joined #jruby
<lopex>
headius: so given the ration it looks the tables are ok ?
<lopex>
*ratio
<lopex>
I was worried about endianness there
<headius>
yeah I didn't run into any problems
<headius>
and the transcoding tests are pretty extensive
x1337807x has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
<headius>
I didn't have to touch table generation at all
<lopex>
headius: one could do a test comparing them to c object file
<lopex>
given the offset
<headius>
mmm yeah true
x1337807x has joined #jruby
GitHub46 has joined #jruby
<GitHub46>
[jcodings] headius pushed 1 new commit to master: http://git.io/ouCOzw
<GitHub46>
jcodings/master 3c9bf52 Charles Oliver Nutter: Update transcoding tables from MRI r48316.
GitHub46 has left #jruby [#jruby]
<headius>
Antiarc: closed that bug
<lopex>
headius: ah, there was also a though about separating the tables
<lopex>
*thought
<lopex>
as a maven dep or something ?
<headius>
to load less into memory, eh?
<headius>
oh right
<lopex>
headius: they're huge
<headius>
yeah
<lopex>
headius: oh, and now ruboto folks will complain again
<headius>
I could see separating them, but that almost feels more like you'd want to separate transcod ing as a whole
<headius>
transcoding won't work at all without the tables
<lopex>
since it affects their startup
<lopex>
headius: but the larges ones are the least used
<headius>
hah...well, ruboto folks won't be complaining about much wrt 9k until I make it work without indy
tlarevo has quit [Ping timeout: 245 seconds]
<lopex>
headius: like the big5 etc
<headius>
yeah
<lopex>
headius: it's worth it
<headius>
I agree it's worth it to split them off
<headius>
I also want to put a nicer Java API in front of jcodings
<headius>
the transcoding API is way too C
<headius>
lopex: you should file an issue with details on moving the tables out
<headius>
maybe we'll get to it soon
<headius>
or you will :-D
<lopex>
headius: deadline ?
toshetm has joined #jruby
x1337807x has quit [Client Quit]
<headius>
9k final is end of year
<headius>
I'll wait to release jcodings until after 2.2 is done, so we have fully up-to-date tables
<lopex>
yeah, I may have some spare time, but I have given enough promises :)
baroquebobcat has joined #jruby
<lopex>
now I'm woking on some crazy idea
<headius>
lopex: there's probably a lot of room for optimization in the transcoder too...it's a pretty close port of the C code but I had to use a lot of carrier objects for pointer references, etc
<lopex>
and it uses jruby anyways :)
<headius>
yay, back on the team!
<lopex>
headius: yeah, I though about that too
<lopex>
headius: a framework to ultrafast implement legacy systems
<lopex>
a reboot
fivebats has quit [Remote host closed the connection]
fivebats has joined #jruby
zeroecco has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
<headius>
legacy systems?
zeroecco has joined #jruby
<lopex>
headius: like, very quick and flexible scaffolding
<lopex>
for legacy dbs
Aethenelle has joined #jruby
rsim has joined #jruby
<lopex>
headius: I havent seen anything like that
etehtsea has quit [Ping timeout: 240 seconds]
<headius>
oh right, I see
<headius>
yeah that sounds interesting
<lopex>
headius: the application is all reflective and tree based
<headius>
sounds like a good candidate then
x1337807x has joined #jruby
x1337807x has quit [Max SendQ exceeded]
x1337807x has joined #jruby
Aethenelle has quit [Quit: Aethenelle]
Aethenelle has joined #jruby
Aethenelle has quit [Client Quit]
<headius>
bleh, merged jcodings fails tests now
<headius>
wtf
<headius>
ugh, transcoding tables not loading now?
<dfr|work>
headius, Antiarc: By the way, seems like the issue probably stems from activerecord-jdbc-adapter although not sure where form exactly
<dfr|work>
but I think I've figured high-level where I messed up, so I think I'll get by. Thanks :)
<headius>
ok cool
baroquebobcat has quit [Quit: baroquebobcat]
marr has joined #jruby
yfeldblum has joined #jruby
e_dub has joined #jruby
e_dub has quit [Client Quit]
pglombardo has joined #jruby
pglombardo1 has quit [Ping timeout: 255 seconds]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] eregon pushed 11 new commits to master: http://git.io/zglbIw
<JRubyGithub>
jruby/master 5164fc0 Benoit Daloze: [Truffle] Bignums have no singleton class.
<headius>
I'm trying to fix another block-yielding exception swallow path now
<headius>
in IO
<headius>
Antiarc: I just want to merge it back so it and other fixes and changes I've made can settle
<headius>
we can continue to fix things on master
<Antiarc>
The static RubyIO.close
<Antiarc>
>?
<Antiarc>
Cool
<Antiarc>
I'll probably poke at it more this weekend
<Antiarc>
I've had a lot of fun working on it :)
<headius>
Antiarc: I moved your fix for "closed stream" into io_close like MRI, and I'm looking at why this case still fails
<headius>
it uses IO.pipe { } so the path is a little different
<Antiarc>
Yeah, I know which one you're talking about
<headius>
I think it's a bug in how IO.open acquires a ChannelFD for the given fileno
anaeem1_ has quit [Remote host closed the connection]
<Antiarc>
getting the wrong type, or what?
x1337807x has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
e_dub has quit [Quit: It's a hard knock life]
<headius>
ahh I see what it is
<headius>
when we have a real native fileno in hand it always creates a new ChannelFD and NativeDeviceChannel
<headius>
so the double close isn't triggering like it should
<headius>
side effect of close logic in NIO
<headius>
there we go
<headius>
now hopefully that change didn't break a ton of other stuff :-)
<headius>
53F/E before this change...hopefully 52F/E or fewer after :-)
drbobbeaty has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
<headius>
Antiarc: thanks for all your help this week...we wouldn't be anywhere near this close otherwise
<headius>
starting to be all one-off failures at this point
<Antiarc>
Yeah, I think we're out of the low-hanging fruit unfortunately
<headius>
yeesh, we need a test:mri:fast
<headius>
these subprocess tests are murder
<Antiarc>
Haha, yup
<headius>
I think the transcoding update wiped out those EUC_KR failures
<headius>
there were KR transcoding updates when I ran the script
<Antiarc>
The Numerics stuff is the last big chunk, but that looked like the consequence of an MRI change that I didn't fully grok
<headius>
which stuff is that? I only see one Numeric failure
<headius>
eregon: if you want a break from truffle, there's pathname failures on update_stdlib (soon to merge to master with excludes)
<Antiarc>
Uh, let me look
<Antiarc>
Oh, most of them went away. The coercion change is what I was thinking about though
<Antiarc>
How many tests are excluded? Is it worth taking a pass through the "needs investigation" cases to see if there are things to be implemented?
<headius>
harumph...53F/E still, but that one went away :-(
<headius>
Antiarc: oh for sure...that has been a to-do forever, go through tagged specs and excluded MRI tests and fix them
<Antiarc>
Did you solve those String#slice tests?
<headius>
no, haven't touched that
<headius>
I'll push everything I have in a moment
<Antiarc>
Basically where under certain failures a string would remain ascii_only? => true even after non-ascii characters were added to it. I poked around in there a bit, but I don't trust my knowledge of encodings and codepoints and whatnot enough to fix it properly
<Antiarc>
I've been doing this stuff for 15 years and it's still mostly incomprehensible to me
<headius>
yeah, that stuff goes pretty deep
<headius>
a lot of it has not been updated (by comparing with MRI) in some years
<headius>
most of it should be 100% portable now, though, since the transcoding and encoding stuff is like 99% matching MRI now
<headius>
they're not line-by-line ports but it's usually possible to follow along and figure out where they changed logic
zeroecco has quit [Ping timeout: 255 seconds]
<headius>
Antiarc: I see your ensureYieldClose change grabs $! and resets it if we ignore the error...did you see that in MRI somewhere?
<headius>
the logic I'm looking at lives in io_close itself and doesn't do that
calavera has quit [Read error: Connection reset by peer]
<headius>
MRI trunk as of today
calavera has joined #jruby
<headius>
zing
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] headius pushed 4 new commits to update_stdlib: http://git.io/B95HtQ
<JRubyGithub>
jruby/update_stdlib a7878d0 Charles Oliver Nutter: Additional cleanup for IO#close exception handling after blocks.
<JRubyGithub>
jruby/update_stdlib 76242f4 Charles Oliver Nutter: Try to use an existing ChannelFD for native FDs, for close logic....
<JRubyGithub>
jruby/update_stdlib de133c9 Charles Oliver Nutter: Make RubyIO.toString use inspect output.
JRubyGithub has left #jruby [#jruby]
Hobogrammer has joined #jruby
<Antiarc>
headius: No, that was just what was there before
<Antiarc>
I just kept that there and ported the new MRI changes
<headius>
ahh ok
calavera has quit [Read error: Connection reset by peer]
x1337807x has joined #jruby
<headius>
perhaps I need to restore those lines then
<Antiarc>
As long as it's not swallowing not-explicitly-ignored exceptions the mechanism probably doesn't matter
calavera has joined #jruby
<Antiarc>
That was the "only swallow some exceptions" mechanism, as best as I can tell
<headius>
well I can see why we might have wanted it, since we don't want close to necessarily clobber $!...but MRI doesn't do it, and I'm not sure if it's a behavioral difference or if they just don't need to restore it
<Antiarc>
My naive guess would be that RubyException sets $! when it's created, whereas that wouldn't happen in C
<headius>
yeah I believe that's part of it
x1337807x has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
nirvdrum has quit [Ping timeout: 258 seconds]
vyorkin has quit [Ping timeout: 258 seconds]
rsim has quit [Quit: Leaving.]
<Antiarc>
headius: re: BigDecimal.log, I saw that too
<Antiarc>
I can invoke it manually from irb
<Antiarc>
but the test fails. Hell if I know why.
<headius>
they do run this test in a subprocess, but I don't know why it wouldn't see log there either
<Antiarc>
not required or something?
<headius>
heh, and the test calls log a few lines earlier too
<headius>
yeah, spossible... I do see the -r in the test, but maybe it's not passing through