iamjarvo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
iamjarvo has joined #jruby
mrmargolis has quit []
noop has quit [Ping timeout: 264 seconds]
johnsonch_afk is now known as johnsonch
josh-k has joined #jruby
johnsonch is now known as johnsonch_afk
josh-k_ has joined #jruby
yfeldblum has joined #jruby
BobFunk has joined #jruby
josh-k has quit [Ping timeout: 264 seconds]
yfeldblum has quit [Ping timeout: 260 seconds]
iamjarvo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
tlarevo has joined #jruby
iamjarvo has joined #jruby
x1337807x has joined #jruby
nirvdrum has quit [Quit: Leaving]
fivebats has joined #jruby
fivebats has quit [Client Quit]
johnsonch_afk is now known as johnsonch
johnsonch is now known as johnsonch_afk
brettporter has quit []
<subbu>
headius, it seems like a webrick bug to me ...
<subbu>
the LJE looks legit ...
<subbu>
headius, if you look at ~/jruby/lib/ruby/stdlib/webrick/server.rb ... the start method can return immediately if the server type is thread (which it is in the test).
<subbu>
which means when the block executes later, the return is left hanging ... with nowhere to return to.
BobFunk has quit [Ping timeout: 244 seconds]
<subbu>
so, that code seems to be relying on the fact that the thread yields immediately to the block.
<subbu>
which it maybe in MRI but not necessarily in JRuby (there is a comment to this effect in RubyThread.java lines 562-563)
<subbu>
so, the test is "racy".
pietr0 has joined #jruby
pietr0 has quit [Client Quit]
robbyoconnor has quit [Ping timeout: 244 seconds]
robbyoconnor has joined #jruby
<subbu>
headius, that return on line 176 in webrick/server.rb should actually be a break
<subbu>
maybe .. anyway, bed time.
yfeldblum has joined #jruby
subbu has quit [Ping timeout: 265 seconds]
johnsonch_afk is now known as johnsonch
purplefox has joined #jruby
johnsonch is now known as johnsonch_afk
x1337807x has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
noop has joined #jruby
x1337807x has joined #jruby
BobFunk has joined #jruby
x1337807x has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
tlarevo has quit [Ping timeout: 256 seconds]
dumdedum has joined #jruby
x1337807x has joined #jruby
x1337807x has quit [Client Quit]
tlarevo has joined #jruby
x1337807x has joined #jruby
pchalupa has joined #jruby
x1337807x has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
johnsonch_afk is now known as johnsonch
johnsonch is now known as johnsonch_afk
cajone has left #jruby [#jruby]
cajone has joined #jruby
iamjarvo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
BobFunk has quit [Ping timeout: 256 seconds]
rsim has joined #jruby
thsig has joined #jruby
JohnBat26 has joined #jruby
tlarevo_ has joined #jruby
tlarevo has quit [Ping timeout: 250 seconds]
thsig has quit [Remote host closed the connection]
Usuario has joined #jruby
johnsonch_afk is now known as johnsonch
Usuario is now known as oba
oba is now known as obs
johnsonch is now known as johnsonch_afk
deobalds has joined #jruby
deobalds has quit [Remote host closed the connection]
mister_solo has joined #jruby
thsig has joined #jruby
ephemerian has joined #jruby
deobalds has joined #jruby
vtunka has joined #jruby
johnsonch_afk is now known as johnsonch
johnsonch is now known as johnsonch_afk
mister_solo has quit [Ping timeout: 264 seconds]
tlarevo has joined #jruby
tlarevo_ has quit [Read error: Connection reset by peer]
shellac has joined #jruby
jeremyevans has quit [Ping timeout: 245 seconds]
jeremyevans has joined #jruby
mister_solo has joined #jruby
johnsonch_afk is now known as johnsonch
johnsonch is now known as johnsonch_afk
drbobbeaty has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
drbobbeaty has joined #jruby
skade has joined #jruby
drbobbeaty has quit [Client Quit]
erikhatcher has joined #jruby
_whitelogger has joined #jruby
snovity has joined #jruby
zorak8 has joined #jruby
marr has joined #jruby
postmodern has quit [Quit: Leaving]
tlarevo_ has joined #jruby
thsig has quit [Remote host closed the connection]
tlarevo has quit [Ping timeout: 245 seconds]
yfeldblum has quit [Remote host closed the connection]
erikhatcher has quit [Quit: erikhatcher]
kares has joined #jruby
toshetm has joined #jruby
deobalds has joined #jruby
iamjarvo has joined #jruby
drbobbeaty has joined #jruby
yfeldblum has joined #jruby
yfeldblu_ has joined #jruby
yfeldblum has quit [Ping timeout: 244 seconds]
yfeldblu_ has quit [Ping timeout: 264 seconds]
mister_solo has quit [Ping timeout: 264 seconds]
Areessell_ has joined #jruby
erikhatcher has joined #jruby
zorak8 has quit [Ping timeout: 250 seconds]
mister_solo has joined #jruby
johnmuhl_ has joined #jruby
thsig has joined #jruby
yfeldblum has joined #jruby
lance|afk is now known as lanceball
yfeldblum has quit [Ping timeout: 265 seconds]
kares has quit [Quit: Ex-Chat]
bbrowning_away is now known as bbrowning
momomomomo has joined #jruby
tlarevo_ has quit [Remote host closed the connection]
triple_b has joined #jruby
tcrawley-away is now known as tcrawley
vtunka has quit [Quit: Leaving]
iamjarvo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<Antiarc>
headius: Process#groups is implemented and working, I just need to extract it into a separate branch so that I can PR it by itself for authorization by the Sugar author
<subbu>
nirvdrum, i was telling enebo y'day that on activesupport 3.0.10 ... with 9k, the test run fails with "unable to load active_support/tzinfo" .. but not with 1.7 ... if i set JRUBY_OPTS to "-X-C" and run jruby -X-C -S rake .. it runs to completion without failing. that indicates something is broken in our JIT -> interp handoff sometimes when jitting fails.
<nirvdrum>
subbu: Okay. In this particular case, I get the same failure with JIT disabled.
<subbu>
the JRUBY_OPTS setting is to ensure that all jruby invocations run with -X-C.
<subbu>
ok.
towski has quit [Ping timeout: 264 seconds]
<subbu>
ya, enebo and i think yours is a different problem .. so, an update about what i found trying to get activesupport running.
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] eregon pushed 1 new commit to master: http://git.io/C1WLEA
<enebo>
yeah I will say for some of these I had to restart jobs 2-3 times each befor the full build greened
<enebo>
but 1.7 seems to be less flakuy like that than master
<headius>
asarih: I *really* hate that the list of builds unexpands when I go into a build and then go back
<headius>
need to do a URL rewrite with a pages=X query or something
nirvdrum has joined #jruby
havenwood has joined #jruby
<enebo>
oh looks like mkristian’s expanded tests might be running for -Pmain
<enebo>
I am not sure -Pmain was the destined location for those
<headius>
hmm
<chrisseaton>
are we keeping a list of the flaky tests? when I get a failure how can I tell if it's flaky or a new failure?
<enebo>
looks like complete might also be running those
<headius>
chrisseaton: we have been excluding them as we find them usually, marking as intermittent in tag or exclude
<headius>
perhaps not with consistent language
<enebo>
headius: for 1.7 I think these are all new tests which got enabled on these three jobs
<headius>
oddly enough the three failing suites appear to start failing when I updated 2.0.0 stdlib
<enebo>
headius: I thought we had disabled them largely because they take a long time to run but these might be different perhaps (maybe a faster subset). In any case they are not passing
<enebo>
headius: heh I hope we are not running —2.0 tests on 1.7.x
joelwallis has joined #jruby
<enebo>
last week I had master down to one non-green until I untagged those predefined_tags
<headius>
yeah that's why I'm confused
<enebo>
err perhaps week before. I don’t remember when I (actually chrisseaton) untagged those extra specs
<headius>
yeah, we're nearly green...it's not like it has gone for months
<enebo>
headius: ^ you moved those clow spawn specs out or tagged them?
<Antiarc>
headius: I submitted a few minor PRs for the 2.2 stdlib. Will poke at it more as I have time!
<headius>
enebo: I haven't done anything recently... I added spec:ruby:fast on master only
<headius>
Antiarc: awesome!
<enebo>
headius: oh and that replaced travis_ci_spec or whatever it was
<headius>
enebo: oh, that commit *did* also update 1.8.7 and 1.9.3 stdlib
<Antiarc>
The really big outstanding one is the each/each_with_index params count mismatch issue, but that code's a little too dense for me to wrap my head around so far
<headius>
and the travis jobs seem to show a missing lib or something
<enebo>
ah
<headius>
I'll figure it out anyway
<enebo>
so probably those other tests were running each time
<Antiarc>
but my intuition is that it's yielding "1" and thus receiving 1 parameter when each_with_index is expecting 2 params (val, index), and the index isn't being yielded for some reason
<headius>
Antiarc: that affects a lot of tests?
<Antiarc>
headius: It was like 40-50 failures
<headius>
that could be an IR bug
<headius>
ahh nice
<headius>
ok
noop has quit [Ping timeout: 255 seconds]
<headius>
if you can just reduce it to an issue that would be helpful
<Antiarc>
kk, will do.
<headius>
there's a bunch of LocalJumpError that we need to reduce too
<Antiarc>
yeah, that's in webrick/server
<Antiarc>
That seems like a jruby bug, since there's no yield/return at the cited line, just an implicit return
<headius>
right...subbu thinks the webrick code is wrong, but it works in MRI...so we have to figure out what's up
<headius>
yeah
<headius>
I didn't dig into it yet
<headius>
Antiarc: btw, you can run those tests individually using the test/mri/runner.rb script
<enebo>
headius: it is definitely wrong but there is another problem
<subbu>
Antiarc, there is a non-local return in server.rb @ line 176
<headius>
jruby test/mri/runner.rb <path under test/mri, like "psych">
<enebo>
headius: read my email
<subbu>
Thread. start { return }.join throws a LJE in mri as well.
<Antiarc>
headius: Perfect, running it directly didn't seem to run all the tests in all cases
<Antiarc>
@subbu line 216 was the one I had in my stack traces, which is the last line of the ensure block in #start
<multibot_>
Unknown command, try @list
<enebo>
yeah MRI does not pass the if confitional but master does with is either determinism bug between client/setrver or a bug with IO.select
<Antiarc>
are you thinking that there's the LJE which resolves to the end of the ensure because the ensure runs despite the return?
<Antiarc>
that'd make sense
<headius>
enebo: subbu: all, so there's a thread in between the non-local return and its intended target?
<headius>
Antiarc: IRB does do things differently so I guess it's not *too* surprising
<Antiarc>
oh?
ludyte has joined #jruby
Areessell_ has joined #jruby
x1337807x has joined #jruby
e_dub has joined #jruby
subbu|lunch is now known as subbu
subbu has quit [Ping timeout: 256 seconds]
Aethenelle has quit [Quit: Aethenelle]
<Antiarc>
Oh, bin/irb is using my system jruby. Whoops.
<Antiarc>
headius: looks like those tests pass under the 1.7.16 release, but break on master. So that should be easy to bisect.
lanceball is now known as lance|afk
<Antiarc>
or rather, break on head of update_stdlib
<enebo>
Antiarc: Enumeriable Block parameter binding is completely different in IR from 1.7
<Antiarc>
That's likely the cause then :)
<enebo>
Antiarc: yeah if you can simplify them even more than the tests it can help (unless they are simple already)
<Antiarc>
They're very simple, fortunately
<enebo>
Antiarc: cool
mister_solo has quit [Ping timeout: 245 seconds]
<Antiarc>
obj = Object.new; class << obj; include Enumerable; def each; yield 1; end; end; a = []; cond = ->(v, i) { a << v }; obj.each_with_index.cycle(2, &cond)
<Antiarc>
That breaks, but this works: obj.each_with_index.cycle(2) {|v, i| a << v }
<Antiarc>
I'm winding my way through the code, but it's not immediately obvious to me what the issue is. Trudging through it though!
<enebo>
Antiarc: explicit lambda perhaps
<enebo>
Antiarc: change -> to proc and see if it passes
<Antiarc>
yeah, that works
<enebo>
Antiarc: also possible it is something weird with singletons but I would be more inclined to pick lambda first (-> is a syntactical lambda so we prepare that as IRBuild time)
<Antiarc>
Proc.new {|v, i| a << x} passes the test just fine
<Antiarc>
err, a<< v
<Antiarc>
you get the idea
<enebo>
Antiarc: ok IRBuilder.buildLmabda is doing something wrong of the interpret of BuildLmabdaInstr is wrong
<enebo>
Antiarc: Also make sure it is not a JIT bug and run with -X-C
<Antiarc>
nope, fails with -X-C
<enebo>
Antiarc: good. We want that
<enebo>
Antiarc: in most cases now fixing interp bugs automatically will fix the JIT
<Antiarc>
Nice
<enebo>
Unless it is more primal to the runtime
pietr0_ has joined #jruby
pietr0 has quit [Ping timeout: 264 seconds]
pietr0_ is now known as pietr0
pietr0 has quit [Remote host closed the connection]
pietr0 has joined #jruby
yfeldblum has joined #jruby
<chrisseaton>
how on earth do they plan for FOSDEM if there's no registration?
subbu has joined #jruby
Aethenelle has joined #jruby
joelwallis has quit [Quit: Saindo]
BobFunk has quit [Ping timeout: 244 seconds]
lance|afk is now known as lanceball
auxbuss has joined #jruby
<headius>
unclear
<headius>
it's free but I assume they want to know numbers
<headius>
enebo: I did land rational, complex, and range support for JIT so it should be on par except for the remaining known gaps (END, BEGIN, big code)
<enebo>
headius: yay
havenwood has quit [Remote host closed the connection]
BobFunk has joined #jruby
phrinx_ has joined #jruby
phrinx has quit [Read error: Connection reset by peer]
ludyte has quit [Quit: ludyte]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] headius pushed 1 new commit to jruby-1_7: http://git.io/xBwu-Q
<JRubyGithub>
jruby/jruby-1_7 78999e2 Charles Oliver Nutter: Revert rdoc to 4.0.1 and remove from sync script....
JRubyGithub has left #jruby [#jruby]
bbrowning_away is now known as bbrowning
robbyoconnor has joined #jruby
JohnBat26 has quit [Ping timeout: 245 seconds]
auxbuss has quit [Quit: I'm gone.]
phrinx_ has quit [Read error: Connection reset by peer]
tsunamie has quit [Ping timeout: 264 seconds]
phrinx has joined #jruby
tsunamie has joined #jruby
tsunamie has quit [Max SendQ exceeded]
tsunamie has joined #jruby
tsunamie has quit [Max SendQ exceeded]
<Antiarc>
This test is failing in JRuby, but I'm not sure it should be fixed.
<chrisseaton>
eregon had looked into how tagging floats works in MRI
<Antiarc>
Anyhow, I'm not sure what the right behavior is - MRI is mathematically wrong, so it's a question of whether jruby should implement proper behavior, or if it should emulate MRI
<Antiarc>
(personally I'd prefer a domain exception in that kind of case, but I realize that's not the given behavior!)
<headius>
Antiarc: we should exclude it and link to an MRI issue asking for clarification
tsunamie has joined #jruby
tsunamie has quit [Max SendQ exceeded]
robbyoconnor has quit [Remote host closed the connection]
robbyoconnor has joined #jruby
tsunamie has joined #jruby
subbu_ has joined #jruby
multibot_ has quit [Read error: Connection reset by peer]
subbu has quit [Ping timeout: 260 seconds]
multibot_ has joined #jruby
zorak8 has joined #jruby
pietr0 has quit [Ping timeout: 264 seconds]
<headius>
huh, there's like 5 talks related to JRuby at RubyConf
<headius>
that's pretty good...almost enough for a whole track for a day
pietr0 has joined #jruby
<headius>
two rubymotion
<headius>
should be a fun conf
<headius>
tenderlove: you have any connection with rdoc these days?
subbu_ is now known as subbu
<headius>
trying to figure out if it's ok to use 4.1.2 with 1.8 mode
<tenderlove>
headius: no.... only via drbrain
<headius>
ok
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] enebo pushed 1 new commit to jruby-1_7: http://git.io/ghj5Mw
<JRubyGithub>
jruby/jruby-1_7 8801d55 Thomas E. Enebo: Fixes #2088. Allow partial but not actual esoteric interpolation sequences to be regexp values
JRubyGithub has left #jruby [#jruby]
mkristian has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] enebo pushed 2 new commits to jruby-1_7: http://git.io/CeVvSw
<JRubyGithub>
jruby/jruby-1_7 2b7fecb kiichi: improve handling of non ascii character after backslash
<JRubyGithub>
jruby/jruby-1_7 2c58354 Thomas E Enebo: Merge pull request #2064 from k77ch7/improve_handling_of_nonascii_character_after_backslash...
JRubyGithub has left #jruby [#jruby]
bbrowning is now known as bbrowning_away
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] enebo pushed 1 new commit to jruby-1_7: http://git.io/8gGJNg
<JRubyGithub>
jruby/jruby-1_7 cccc1d7 Thomas E. Enebo: WTF is all this logic for? Make readlink pass some more rubyspecs
JRubyGithub has left #jruby [#jruby]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] headius pushed 1 new commit to jruby-1_7: http://git.io/g36DiQ
<JRubyGithub>
jruby/jruby-1_7 bb8992e Charles Oliver Nutter: Update to rdoc 4.1.2 and set it up as a full default gem.
JRubyGithub has left #jruby [#jruby]
lanceball is now known as lance|afk
tcrawley is now known as tcrawley-away
e_dub has quit [Quit: ZZZzzz…]
<Antiarc>
This one's definitely a jruby bug - ((2 ** 53) + 1).to_f == ((2 ** 53) + 0).to_f - but I'm not immediately sure where the issue is. I suspect a range check somewhere - any thoughts?
<headius>
that looks like a fixnum range different
<multibot_>
<hint>:1:22: parse error on input ‘=>’
<headius>
hmm
<Antiarc>
It's just + 1 and + 0 that are the same, so I suspect someone used <= when they should have used <, or similar :)
<headius>
yeah weird
drbobbeaty has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
<headius>
Antiarc: sometimes new failures are associated with fixed MRI bugs...if we can find those there's often a one-line bug we ported over
<Antiarc>
Got it. I'll dig around and see if I can locate it
<headius>
enebo: 1.7 looks like it will go green...the triplets of failures are at least working ok with rdoc 4.0.1 fix
<headius>
then we'll let rdoc 4.2.1 commit run
<enebo>
headius: cool
<Aethenelle>
just out of curiosity... has anyone worked on a yard plugin to do Java/JRuby doc?
<enebo>
headius: The issues I have marked on 1.7.17 should all get resolved before we release. Pretty much all are regressions for various 1.7 releases
<headius>
Aethenelle: like, to document jruby core?
zenspider has left #jruby ["ERC Version 5.3 (IRC client for Emacs)"]
<headius>
enebo: ok
<Aethenelle>
headius: that's one thing it could be used for... I was thinking more of documenting Java based gems...
<headius>
I have a few things I could do...work on 1.7 issues, work on update_stdlib, get master green
<headius>
Aethenelle: ahh right
<headius>
enebo worked on a tool to do that at one point
<headius>
mixed doco across ruby/java
<headius>
dunno if that documented extension-bound methods though
<enebo>
Aethenelle: noridoc which was based on javadoc + was meant to hook rdoc
<enebo>
Aethenelle: but after talking to yokolet she thought a plugin to rdoc would be better
<enebo>
Aethenelle: but Yard would also be fine and someone suggested that
<enebo>
Aethenelle: I think it is an important tool so if you have interest :)
<headius>
enebo: I'm going to get master green
<Aethenelle>
i think I have enough on my stack for now...
<headius>
I'll need to do that for update_stdlib anyway