metadave_ has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
baroquebobcat has quit [Quit: baroquebobcat]
kl__ has quit [Ping timeout: 245 seconds]
Felystirra has quit []
<mjc_>
headius: autocorrect? "I hate windows" seems more likely :P
skade has quit [Quit: Computer has gone to sleep.]
shellac has quit [Quit: Computer has gone to sleep.]
kl has joined #jruby
erikhatcher_ has joined #jruby
erikhatcher has quit [Read error: Connection reset by peer]
erikhatcher_ is now known as erikhatcher
<Antiarc>
Windows is pretty okay these days if you aren't trying to do POSIX things
zorak8 has quit [Ping timeout: 240 seconds]
multibot_ has quit [Read error: Connection reset by peer]
multibot_ has joined #jruby
e_dub has joined #jruby
zorak8 has joined #jruby
dinfuehr has joined #jruby
tcrawley-away is now known as tcrawley
dinfuehr has quit [Ping timeout: 252 seconds]
e_dub has quit [Quit: e_dub]
marr has quit []
multibot_ has quit [Remote host closed the connection]
multibot_ has joined #jruby
pietr0 has quit [Quit: pietr0]
mrmargolis has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] cheald opened pull request #2504: Fix ascii-only detection of symbol names. (master...utf8_string_symbol_windows_fix) http://git.io/J1Tjdw
JRubyGithub has left #jruby [#jruby]
nirvdrum has quit [Ping timeout: 245 seconds]
e_dub has joined #jruby
camlow325 has quit [Remote host closed the connection]
Hobogrammer has joined #jruby
robbyoconnor has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] kingspider1971 opened issue #2505: IO.popen cannot run 'date' on Windows 7 http://git.io/nwKkpw
JRubyGithub has left #jruby [#jruby]
r0bby_ has joined #jruby
robbyoconnor has quit [Ping timeout: 245 seconds]
mrmargolis has quit [Remote host closed the connection]
r0bby_ is now known as robbyoconnor
lopex has quit [Quit: Connection closed for inactivity]
josh-k has quit [Remote host closed the connection]
deobalds has quit [Ping timeout: 252 seconds]
pgokeeffe has joined #jruby
<headius>
Antiarc: indeed
deobalds has joined #jruby
erikhatcher has quit [Quit: erikhatcher]
Who has joined #jruby
DavidEGrayson has quit [Quit: Leaving.]
Hobogrammer has quit [Ping timeout: 244 seconds]
tcrawley is now known as tcrawley-away
yfeldblu_ has joined #jruby
yfeldblum has quit [Ping timeout: 245 seconds]
kl has quit [Ping timeout: 265 seconds]
Hobogrammer has joined #jruby
zorak8 has quit [Ping timeout: 272 seconds]
anaeem1 has joined #jruby
anaeem1 has quit [Remote host closed the connection]
anaeem1_ has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] digitalextremist opened issue #2506: Mysterious core dump / segfault with 9.0.0.0-pre. http://git.io/f5-5Vw
JRubyGithub has left #jruby [#jruby]
codefinger has quit [Remote host closed the connection]
Who has quit [Ping timeout: 240 seconds]
Who has joined #jruby
colinsurprenant has joined #jruby
codefinger has joined #jruby
DomKM has quit [Quit: Connection closed for inactivity]
colinsurprenant has quit [Quit: colinsurprenant]
havenwood has quit []
deobalds has quit [Ping timeout: 264 seconds]
deobalds has joined #jruby
havenwood has joined #jruby
deobalds has quit [Ping timeout: 252 seconds]
deobalds has joined #jruby
Who has quit [Quit: Who]
yfeldblu_ has quit [Remote host closed the connection]
yfeldblum has joined #jruby
JohnBat26 has joined #jruby
diegoviola has quit [Quit: WeeChat 1.1]
pgokeeffe has quit [Quit: pgokeeffe]
deobalds has quit [Ping timeout: 245 seconds]
deobalds has joined #jruby
djellemah has joined #jruby
donV has quit [Quit: donV]
Who has joined #jruby
deobalds has quit [Ping timeout: 276 seconds]
multibot_ has quit [Remote host closed the connection]
multibot_ has joined #jruby
johnmuhl has quit [Quit: Connection closed for inactivity]
deobalds has joined #jruby
camlow325 has joined #jruby
codefinger has quit [Remote host closed the connection]
havenwood has quit []
noop has joined #jruby
slyphon has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
skade has joined #jruby
donV has joined #jruby
kl has joined #jruby
rsim has joined #jruby
Who has quit [Quit: Who]
robbyoconnor has quit [Quit: Konversation terminated!]
r0bby_ has joined #jruby
codefinger has joined #jruby
codefinger has quit [Ping timeout: 246 seconds]
yfeldblum has quit [Ping timeout: 245 seconds]
skade has quit [Quit: Computer has gone to sleep.]
kares has joined #jruby
deobalds has quit [Quit: Computer has gone to sleep.]
nirvdrum has joined #jruby
dinfuehr has joined #jruby
camlow325 has quit [Remote host closed the connection]
camlow325 has joined #jruby
camlow325 has quit [Ping timeout: 245 seconds]
deobalds has joined #jruby
kl has quit [Ping timeout: 244 seconds]
pchalupa has joined #jruby
deobalds has quit [Ping timeout: 246 seconds]
deobalds has joined #jruby
nirvdrum has quit [Ping timeout: 272 seconds]
skade has joined #jruby
eugenmayer has joined #jruby
<eugenmayer>
good morning. Reaing about warbler and jruby i though to just ask here. I want to keep ruby ( with rails ) while deploying precompiled binaries. The reason is 30% performance, 70% securing the code ( for obvious / easy modifaction ). JRuby suits these needs, right?
shellac has joined #jruby
kl__ has joined #jruby
kl__ has quit [Ping timeout: 276 seconds]
skade has quit [Quit: Computer has gone to sleep.]
shellac has quit [Quit: Computer has gone to sleep.]
Hobogrammer has quit [Ping timeout: 245 seconds]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] raskhadafi opened issue #2507: load error: jopenssl/load in jruby 9.0.0.0.pre1 http://git.io/cvsqew
JRubyGithub has left #jruby [#jruby]
marr has joined #jruby
olleolleolle has joined #jruby
lopex has joined #jruby
vtunka has joined #jruby
vtunka has quit [Client Quit]
vtunka has joined #jruby
josh-k_ has joined #jruby
skade has joined #jruby
yfeldblum has joined #jruby
vtunka has quit [Quit: Leaving]
nirvdrum has joined #jruby
joast has quit [Ping timeout: 252 seconds]
donV has quit [Quit: donV]
donV has joined #jruby
deobalds has quit [Quit: Computer has gone to sleep.]
yfeldblum has quit [Remote host closed the connection]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] mkristian force-pushed test-jossl-0.9.6 from 01d4877 to 0012943: http://git.io/iQPOiA
<eugenmayer>
]good morning. Reaing about warbler and jruby i though to just ask here. I want to keep ruby ( with rails ) while deploying precompiled binaries. The reason is 30% performance, 70% securing the code ( for obvious / easy modifaction ). JRuby suits these needs, right?
josh-k_ has quit [Remote host closed the connection]
portertech has quit []
portertech has joined #jruby
zorak8 has joined #jruby
josh-k has joined #jruby
anaeem1__ has joined #jruby
anaeem1_ has quit [Ping timeout: 264 seconds]
vtunka has joined #jruby
dinfuehr has quit [Ping timeout: 255 seconds]
hosiawak has joined #jruby
yfeldblum has quit [Remote host closed the connection]
metadave has joined #jruby
vtunka has quit [Ping timeout: 265 seconds]
anaeem1__ has quit [Remote host closed the connection]
rcvalle has joined #jruby
vtunka has joined #jruby
bbrowning_away is now known as bbrowning
zorak8 has quit [Ping timeout: 264 seconds]
hosiawak has quit [Remote host closed the connection]
nateberkopec has quit [Quit: Leaving...]
shellac has joined #jruby
amdprophet_ has quit []
amdprophet_ has joined #jruby
vtunka has quit [Quit: Leaving]
<headius>
eugenmayer: it suits the "securing code" aspect, but precompiling in JRuby doesn't really affect performance
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] chrisseaton pushed 1 new commit to master: http://git.io/DufNnQ
<JRubyGithub>
jruby/master 0fb5e4b Chris Seaton: [Truffle] jt run command (work around for broken JRuby launcher).
JRubyGithub has left #jruby [#jruby]
<eugenmayer>
most probably rather its a better perfomance as far as i could read it up?
<eugenmayer>
@headius
<multibot_>
Unknown command, try @list
joast has joined #jruby
dinfuehr has joined #jruby
yfeldblum has joined #jruby
<headius>
eugenmayer: well it avoids some of the warmup caused by lazily compiling Ruby code into JVM bytecode, but it can take longer to start up because the JVM has more bytecode to verify
<headius>
uncompiled apps will eventually reach the same performance in most cases
<eugenmayer>
headius: yea it depends on the amount of requests sure. Just in case, are there any "standalone" daemons for ror?
<eugenmayer>
so instead of debootstrapping on every request, keep the CGI / Framework wrapper bootstrapped, handling all requests when they come in
dinfuehr has quit [Ping timeout: 245 seconds]
<headius>
well generally once Rails boots up it stays booted, on all servers
yfeldblum has quit [Ping timeout: 245 seconds]
<headius>
Ruby apps don't really work like CGI...they're always running a standalone server
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] eregon pushed 1 new commit to master: http://git.io/jvJLDw
<JRubyGithub>
[jruby] chrisseaton pushed 1 new commit to master: http://git.io/wfAZAA
<JRubyGithub>
jruby/master ce0df7f Chris Seaton: [Truffle] Add temporary launcher wrapper script while -X+T is broken to interface with CI systems.
JRubyGithub has left #jruby [#jruby]
anaeem1_ has joined #jruby
anaeem1_ has quit [Remote host closed the connection]
anaeem1_ has joined #jruby
donV has quit [Quit: donV]
skade has quit [Quit: Computer has gone to sleep.]
anaeem1_ has quit [Remote host closed the connection]
anaeem1 has joined #jruby
colinsurprenant has joined #jruby
anaeem1 has quit [Remote host closed the connection]
joast has quit [Ping timeout: 256 seconds]
<headius>
chrisseaton: I missed discussions about -X+T...what did mkristian recommend we do?
<headius>
sounds like mostly just modifying launcher to pick the truffle version?
<headius>
strange...truffle stuff is in my jruby.jar but -X+T doesn't work
r0bby_ has quit [Ping timeout: 244 seconds]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] headius closed issue #1389: JRuby can Not load JFFI library under IBM Power 64 with PLinux platform http://git.io/N8MQJQ
JRubyGithub has left #jruby [#jruby]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] headius reopened issue #1389: JRuby can Not load JFFI library under IBM Power 64 with PLinux platform http://git.io/N8MQJQ
JRubyGithub has left #jruby [#jruby]
Who has joined #jruby
yfeldblum has joined #jruby
e_dub has quit [Quit: e_dub]
noop has quit [Ping timeout: 265 seconds]
yfeldblum has quit [Ping timeout: 264 seconds]
eugenmayer has quit [Quit: Leaving.]
DrShoggoth has joined #jruby
mrmargolis has joined #jruby
r0bby_ has joined #jruby
slyphon has joined #jruby
donV has joined #jruby
dinfuehr has joined #jruby
nateberkopec has joined #jruby
kwando has joined #jruby
<bbrowning>
headius: I haven't figured out the exact details yet, but somehow using autoload and regular requires it's possible to load a file more than once, leading to that error when ProxyObject in ActiveSupport tries to undef ==
johnmuhl has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] mkristian created test-build-distribution (+2 new commits): http://git.io/CCdeDw
<JRubyGithub>
jruby/test-build-distribution ec3e503 Christian Meier: pack jni/* native libraries only in lib/jruby.jar
<JRubyGithub>
jruby/test-build-distribution 6ce1792 Christian Meier: cleanup naming of shaded jars
JRubyGithub has left #jruby [#jruby]
<headius>
bbrowning: ahh hmmm
<headius>
bbrowning: so it's trying to undef twice
<headius>
you should be able to track loading with debug.loadService
<bbrowning>
exactly, it tries to undef twice
<headius>
perhaps something is not canonicalizing properly for loaded features
<bbrowning>
I'm actually using a debugger to step through it now since debug.loadService didn't narrow it down enough
<bbrowning>
so in a bit I should have a better idea of how it's getting loaded twice
<headius>
debug.loadService.timing sometimes is better for me, since it shows the hierarchy of loads
<bbrowning>
I can tell you that the first time it tries to load proxy_object.rb via the line "boolean loaded = tryLoadingLibraryOrScript(runtime, state);" that process of loading proxy_object.rb triggers an autoload of ProxyObject, when then loads it again
<bbrowning>
basically two loads of proxy_object.rb get triggered before we mark that feature as loaded
enebo has joined #jruby
<headius>
both are triggered via autoload?
<headius>
does this happen consistently or could it be multithreading issues?
<headius>
like, something regressed wrt thread-safty of autoload
r0bby_ has quit [Ping timeout: 264 seconds]
<bbrowning>
headius: one explicit require and one via autoload
<bbrowning>
and it's not a threading issue, no
<headius>
ok
<bbrowning>
all happening on the same thread
<headius>
that's good
<headius>
are the rails files inside a jar?
slyphon has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
<headius>
so I guess proxy_object.rb is the file autoloaded for ProxyObject in active_support.rb
<headius>
that autoload is in an eager section so eager autoloading should do it on boot
<bbrowning>
that simple reproduction triggers it :)
<bbrowning>
define an autoload and then require a file that defines the constant defined in the autoload
<bbrowning>
end up loading that file twice, which many times would be fine
<headius>
bbrowning: reproduced at command line
slyphon has joined #jruby
<headius>
hahah
<headius>
I just made that same script, but reversed the lines
<bbrowning>
hehe
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] mkristian opened pull request #2509: improving building of distribution archives and artifacts (master...test-build-distribution) http://git.io/VU7Oaw
JRubyGithub has left #jruby [#jruby]
<headius>
high five
<headius>
should be easy to find issue now
<bbrowning>
I'll file a github issue for it so you can keep track
donV has quit [Quit: donV]
<headius>
actually a simple script that's just require 'active_support'; require 'active_support/proxy_object.rb'
<headius>
that's enough to reproduce locally
<bbrowning>
right that works, but requires active_support to be installed
<headius>
indeed
<bbrowning>
and actually activesupport 4.1.x
<headius>
bbrowning: did you see require first and then autoload, or other way around?
<bbrowning>
headius: the require happens first and then when the required file is loading the "class ProxyObject" triggers the autoload
<bbrowning>
basically when a class is defined it's also triggering the autoloading for that class
<headius>
yeah
<headius>
just wanted to confirm that's what you're seeing
<headius>
I bet I know what's wrong
<headius>
the class definition is doing the lookup of the class using autoload
<headius>
it should not
mitchellhenke has joined #jruby
<headius>
yeah that's looking like the problem
metadave_ has joined #jruby
metadave_ has quit [Client Quit]
<headius>
strange...1.7 logic looks the same
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] eregon pushed 1 new commit to master: http://git.io/H7ileQ
<JRubyGithub>
jruby/master 9345e9a Benoit Daloze: Eclipse: move .factorypath to truffle/ and generalize ignore paths.
JRubyGithub has left #jruby [#jruby]
<bbrowning>
I wasn't able to trigger it on 1.7 though
<headius>
yeah
<headius>
so maybe it's not what I thought...maybe it's supposed to see that we're currently doing that autoload and not re-attempt it
metadave has quit [Ping timeout: 255 seconds]
slyphon has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] bbrowning opened issue #2510: Explicit require of autoloaded class triggers double load http://git.io/trBYSw
JRubyGithub has left #jruby [#jruby]
<headius>
oh, or circular require detection is failing
<bbrowning>
huh - I hadn't tried this exact repro and it also fails
<bbrowning>
for me
<headius>
yeah
<bbrowning>
perhaps something's wrong with my repro script then - the real code works fine under 1.7 but not 9k
<headius>
MRI runs it fine
<headius>
so there's a related issue in 1.7 too
colinsurprenant has quit [Quit: colinsurprenant]
<headius>
I think all these cases are problems matching the required name to the circular require lock
<headius>
hmm...though MRI doesn't warn
<bbrowning>
this shouldn't spit out a warning I wouldn't think, since it's not a user-initiated circular require
dinfuehr has quit [Remote host closed the connection]
<headius>
that's certainly possible
<headius>
ahh right I see we turn off the warning for autoload require
slyphon has joined #jruby
<headius>
heh
<headius>
well here's one difference... MRI appears to pull off the extension when searching loaded features and we don't
<bbrowning>
headius: in my debugging, it doesn't appear we've even added this thing to the list of loaded features yet before the 2nd load happens
<bbrowning>
explicit require loads the file and then the file loading itself triggers the autoload which loads the same file again. then, feature gets marked as loaded (only after 2nd load finishes).
<bbrowning>
1st load never finishes because it blows up on the 2nd undef_method
<headius>
true
<bbrowning>
it's almost like something needs to realize the autoload is already satisfied so it doesn't trigger
<bbrowning>
"hey look we're already loading the file that the autoload says to require when this constant is undefined, so ignore actually loading it a second time"
colinsurprenant has joined #jruby
<headius>
right
<headius>
I believe the only place either of us put that info would be in the require locks
<headius>
though I'm walking through MRI's logic now
<bbrowning>
I'm still confused why no one's reported this in 1.7 before
imperator has joined #jruby
<headius>
I also see MRI adding a provided feature before loading
<bbrowning>
that would definitely solve it - if we added to the loaded feature list before doing the actual loading
<headius>
hmmm
<bbrowning>
but right now jruby checks to be sure something actually loaded before adding to that list
<headius>
this might be something else
booly-yam-9117 has joined #jruby
<headius>
ah
<headius>
MRI checks the loading table (require lock table) when checking loaded features too
<headius>
and uses the extension and load path expansion logic to try multiple forms
<headius>
bbrowning, enebo: this may force us to finally duplicate the logic they have for maintaining expanded load path
<headius>
I don't think this is a regression...I think it's a bug that was hidden because we weren't expanding paths when we should
<enebo>
yum
<bbrowning>
ahh
<headius>
now that we expand paths, whatever prevented this double-load in 1.7 fails to work now
<headius>
and the logic in MRI that corresponds uses expanded load path as part of its "currently loading" search
<bbrowning>
I have no idea how widespread the issue will show up - I just hit it using some tests that I recently upgraded to rails 4.1.8 from an earlier version
<bbrowning>
the same tests running rails 4.2 seem to be ok
<bbrowning>
they changed some of the require stuff inside activesupport around for 4.2
<enebo>
I have been testing against 4.2 for pre1
<enebo>
Getting closer to the fruits of victory
yfeldblum has joined #jruby
jsvd has quit []
yfeldblum has quit [Ping timeout: 244 seconds]
baroquebobcat has joined #jruby
DrShoggoth has quit [Quit: Leaving]
skade has joined #jruby
e_dub has joined #jruby
tcrawley is now known as tcrawley-away
skade has quit [Ping timeout: 265 seconds]
dinfuehr has joined #jruby
tcrawley-away is now known as tcrawley
<headius>
bleh
<temporalfox>
hi there
<temporalfox>
I'm almost done with jruby lang for Vert.x 3
<temporalfox>
I have a strange bug
<temporalfox>
in my test suite
<temporalfox>
anyone available to hear the symptoms :-) ?
DrShoggoth has joined #jruby
DrShoggoth has quit [Remote host closed the connection]
DrShoggoth has joined #jruby
<headius>
sure!
<temporalfox>
thanks :-)
<temporalfox>
issue is when I run the full testsuite the bug appears
<temporalfox>
when I run the test alone, it does not
dinfuehr has quit [Remote host closed the connection]
<temporalfox>
and the issue is about an object that is supposed to be true or false literal
<temporalfox>
and when it fails it's because it's not true of false object
<dfr|work>
headius, we're talking about 9k only now, right?
mkristian has joined #jruby
booly-yam-9117 has quit [Ping timeout: 264 seconds]
booly-yam-9117_ has joined #jruby
shellac has quit [Quit: Ex-Chat]
<mkristian>
headius, enebo hi
<mkristian>
what shall I do with those jni/* libraries ? lib/jruby.jar only. or differentiate between the various distribution forms ?
<mkristian>
currently they are both inside jruby.jar as well under lib/jni/*
eugenmayer has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] mkristian pushed 1 new commit to test-build-distribution: http://git.io/D4quRQ
JRubyGithub has left #jruby [#jruby]
<JRubyGithub>
jruby/test-build-distribution 56c9411 Christian Meier: get core and truffle artifact attached
<headius>
mkristian: well I'm pretty sure they don't need to be in both of them
<headius>
both locations
<headius>
I think the lib/jni is the best location to keep, since jruby.jar would have to unpack the libraries to use them anyway
<headius>
dfr|work: yeah so far
<dfr|work>
headius, wootz :D
<headius>
dfr|work: it would be extremely invasive in 1.7 and 1.7 dodges most of the issues by not absolutizing as much
<headius>
that's a bug too but it's a smaller impact :-)
<dfr|work>
headius, I have no idea whether I'll have time, but I've mentally added it to my mental list :)
oblutak has joined #jruby
<dfr|work>
and moving forward with 9k makes perfect sense to me anyway
<dfr|work>
no more speciall 1.8 classes :D
<headius>
ok...Who (who828 on Github) is also interested in looking into it
<headius>
I think the best option is going to be to port their expanded load path and loaded features logic...we can refactor and refine once we have matching logic
calavera has joined #jruby
<mkristian>
headius, then this means I need to pack another jruby.jar without those jni/* for jruby-dist archives only
<headius>
mkristian: how big is the extra size?
<mkristian>
you keep the jruby.jar with jni/* included. let me check
<mkristian>
40k
jwinter_ has quit [Quit: Connection closed for inactivity]