x1337807x has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
mjelen has quit [Ping timeout: 250 seconds]
mjelen has joined #jruby
x1337807x has joined #jruby
camlow325 has joined #jruby
skade has joined #jruby
x1337807x has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
vikaton has quit [Quit: Connection closed for inactivity]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] deepj opened issue #2957: JRuby 9.0.0.0.pre2 starts handing while reading from a socket http://git.io/vTZsc
JRubyGithub has left #jruby [#jruby]
djbkd has quit [Quit: My people need me...]
baroquebobcat has quit [Quit: baroquebobcat]
baroquebobcat has joined #jruby
skade has quit [Quit: Computer has gone to sleep.]
anaeem1 has joined #jruby
ponga has joined #jruby
mister_solo has joined #jruby
baroquebobcat has quit [Quit: baroquebobcat]
mister_solo has quit [Ping timeout: 244 seconds]
arturaz has joined #jruby
djellemah has joined #jruby
drbobbeaty has joined #jruby
drbobbeaty has quit [Client Quit]
mister_solo has joined #jruby
camlow325 has quit [Remote host closed the connection]
TheBrayn has joined #jruby
elia has joined #jruby
rsim has joined #jruby
rsim has quit [Client Quit]
mister_solo has quit [Ping timeout: 245 seconds]
mister_solo has joined #jruby
TheBrayn has left #jruby ["WeeChat 1.1.1"]
ponga has quit [Remote host closed the connection]
elia has quit [Quit: Computer has gone to sleep.]
mister_solo has quit [Remote host closed the connection]
skade has joined #jruby
mister_solo has joined #jruby
vtunka has joined #jruby
ponga has joined #jruby
drbobbeaty has joined #jruby
mister_solo has quit [Ping timeout: 252 seconds]
ponga has quit [Remote host closed the connection]
shellac has joined #jruby
drbobbeaty has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
drbobbeaty has joined #jruby
elia has joined #jruby
keeguon has joined #jruby
vtunka has quit [Quit: Leaving]
skade has quit [Quit: Computer has gone to sleep.]
<keeguon>
Hey guys, trying desperately to exclude files when generating a war with warbler (1.4.7), I’m using the compiled option as well so my understanding judging from this PR: https://github.com/jruby/warbler/pull/254 is that they also should be excluded when compiling but it’s not the case, for information my “config.excludes” looks like this: “config.excludes = FileList["tmp/restart.txt"] +
drbobbeaty has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
<keeguon>
Nevermind figured it out
mister_solo has quit [Ping timeout: 244 seconds]
cprice has joined #jruby
cprice404 has quit [Ping timeout: 258 seconds]
ponga has quit [Remote host closed the connection]
elia has quit [Ping timeout: 264 seconds]
elia has joined #jruby
ponga has joined #jruby
ponga has quit [Client Quit]
skade has joined #jruby
vtunka has joined #jruby
keeguon has quit [Quit: keeguon]
havenwood has quit [Ping timeout: 264 seconds]
rsim has joined #jruby
Nilbus has joined #jruby
cajone has left #jruby [#jruby]
pjammer has joined #jruby
vikaton has joined #jruby
drbobbeaty has joined #jruby
mister_solo has joined #jruby
<Nilbus>
Good morning! When a Java exception is thrown in my JRuby app, tracebacks can be extremely long and unhelpful for finding the ruby call that called into java. Annotated example: https://gist.github.com/nilbus/62819abc689775f16584 Has this improved since JRuby-1.7.6? Is there a way to get more meaningful (ruby source/line) from these tracebacks?
mister_s_ has joined #jruby
mister_solo has quit [Ping timeout: 245 seconds]
pjammer has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
nirvdrum has joined #jruby
keeguon has joined #jruby
bbrowning_away is now known as bbrowning
ponga has joined #jruby
JRubyGithub has joined #jruby
JRubyGithub has left #jruby [#jruby]
<JRubyGithub>
jruby/master 23eb994 Benoit Daloze: [Truffle] Define even?, odd? and succ on Fixnum.
<JRubyGithub>
[jruby] eregon pushed 1 new commit to master: http://git.io/vTnNj
nirvdrum has quit [Ping timeout: 256 seconds]
nirvdrum has joined #jruby
<kares>
Nilbus: yes
<kares>
... it has improved - but it depends on how/where from it's thrown
ponga has quit [Remote host closed the connection]
pjammer has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
vtunka has quit [Quit: Leaving]
anaeem1 has quit [Remote host closed the connection]
keeguon has joined #jruby
<headius>
good morning!
<headius>
subbu, chrisseaton: I know PHP has a reference-counting memory model that has proven tricky for alternate impls to support
djellemah_ has joined #jruby
<headius>
and you can't implement PHP without supporting or porting PHP exts, since there's really no PHP without them
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] chrisseaton pushed 1 new commit to master: http://git.io/vTcGG
<JRubyGithub>
jruby/master d78025c Chris Seaton: [Truffle] Use #StringValue instead of manually checking for types in digest.
JRubyGithub has left #jruby [#jruby]
djellemah has quit [Ping timeout: 256 seconds]
ponga has quit [Remote host closed the connection]
<Nilbus>
No luck using a newer version.
codefinger has joined #jruby
<headius>
Nilbus: try running with -Xjit.threshold=0
<headius>
the Java traceback issue is known and hard to fix because we want to just rethrow the original Java exception directly
skade has quit [Quit: Computer has gone to sleep.]
<Nilbus>
We run our app in 1.9 mode, so it's not surprising to me that it's having trouble with JRuby 9000 in 2.x mode. That's just not something I can test for now.
skade has joined #jruby
<headius>
Nilbus: what problem are you having related to 2.x? I have not heard of many cases of issues making that move
mitchellhenke has joined #jruby
<Nilbus>
The immediate issue I have trying to run this spec is this: LoadError: load error: jopenssl/load -- java.lang.SecurityException: class "org.bouncycastle.asn1.ASN1Primitive"'s signer information does not match signer information of other classes in the same package
<Nilbus>
bouncycastle being a library we use
<Nilbus>
I have no reason to think that's related to ruby 2x really
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] chrisseaton pushed 1 new commit to master: http://git.io/vTc8u
<JRubyGithub>
jruby/master d265a61 Chris Seaton: New specs for zsuper with optional arguments.
JRubyGithub has left #jruby [#jruby]
<Nilbus>
kares, are you familiar with the specific ways exceptions are thrown that make the difference you're thinking of?
skade has quit [Quit: Computer has gone to sleep.]
ponga has joined #jruby
tcrawley-away is now known as tcrawley
<chrisseaton>
headius: the other day we were discussing why we do our own core library for Truffle - the other day I thought of a really simple example of the benefit...
deobalds has joined #jruby
<chrisseaton>
headius: when you do something like Array#pack() you'd like to do one allocation of the output ByteList, of exactly the right length. You can try to do that by caching the previous resulting length...
<chrisseaton>
headius: but to be useful that needs to be cached per source location, and really you'd like to keep moving that cache up the call graph so if you have a method between you and Array#pack() you get a separate cache for each of the callers of those methods instead
<headius>
Nilbus: ah definitely not...I know what that is
<chrisseaton>
headius: and Java doesn't give you anywhere to store that kind of cache, but Truffle nodes do
<headius>
Nilbus: this happens when bouncycastle (the library we use to implement 'openssl') gets re-packaged into a different jar... there's signature files that come with it that no longer match the jar
mitchellhenke has quit [Quit: Computer has gone to sleep.]
<headius>
I'm pretty sure that's what it is
iamjarvo has joined #jruby
<headius>
so I'm wondering how you've built your app
iamjarvo has quit [Max SendQ exceeded]
iamjarvo has joined #jruby
<headius>
chrisseaton: mmm sure
<headius>
one thing I'd like to see is how the specialization of all those core classes affects generated code size
<chrisseaton>
headius: maybe that it something you could put into the 292 API though
<headius>
there's a balance to be maintained in how much code gets generated versus how much is used generically
<chrisseaton>
headius: yeah I might do a review of code size and compilation time, warmup etc for my thesis in the next month
<headius>
we never specialize core classes and you do for many, many cases
<headius>
so you'll end up with many copies of basically the same code
yfeldblum has joined #jruby
<headius>
but yeah, that's an interesting case
<headius>
I could see doing that with 292, yes, since we could bind custom logic to wrap any method at the call site
<chrisseaton>
i wonder if we could reduplicate generated code - it goes into the compiler from different places, but ends up the same so drop what you generated and use the original instead
<chrisseaton>
deduplicate - not reduplicate
<headius>
yeah I don't see why you couldn't
<headius>
if they actually come out the same
<headius>
the cost is compiling and then throwing it away once you realize you've already got it
<headius>
hotspot doesn't do that either
<chrisseaton>
ah there might be a reason then :)
<headius>
probably could do the comparison at graal IR level
<headius>
we have had to recommend bumping the hotspot code cache size up just for plain jruby
<headius>
so we know we already generate a ton of native code, even without specialization of anything
<headius>
by bump up I'm talking hundreds of MB for medium-sized apps
<headius>
computers are hard :-\
<chrisseaton>
while you were away we got a basic web server working
<headius>
that's great!
skade has joined #jruby
<headius>
time to post a hello world benchmark
<chrisseaton>
turned into a massive rabbit hole of FFI stuff though, and we're not going to be running on Windows very soon
<headius>
chrisseaton: ahh that's an aspect of using rbx code I didn't think of
<headius>
they've never had it running on Windows in any form
<chrisseaton>
The actual calls are done via JFFI, but is really proving its worth again
<headius>
just to clarify...you're not using jffi directly, right? jnr-ffi you mean?
<headius>
I know the names are unfortunately similar
<chrisseaton>
I'm doing the thing where I have an interface and then I can get an instance of that backed with native code with one call
<headius>
ok, jnr-ffi
<headius>
jffi is literally just invoke(function pointer, other crap)
<chrisseaton>
I wasn't sure what to do about int/long for my function signatures? The interface is x-platform, right? So how does it resolve that
ponga has quit [Remote host closed the connection]
<headius>
the types in Java have some standard translation to platform types, and it adjusts based on the platform
<headius>
for exotic types Java doesn't have you need to specify (unsigned etc)
<headius>
"exotic"
<headius>
I never thought I'd refer to unsigned integers as exotic
<chrisseaton>
It all just seemed to work fine for things like getaddrinfo (I'll push that down into jnr-unix-sockets at some point)
<headius>
wmeissner was a wizard
iamjarvo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] eregon pushed 2 new commits to master: http://git.io/vTcKs
<JRubyGithub>
[jruby] florf opened issue #2959: JRuby 9k-pre2 throws IOError when reading from $stdin http://git.io/vTcHZ
JRubyGithub has left #jruby [#jruby]
<Nilbus>
I'll see if updating to 1.5.0147 helps
colinsurprenant has joined #jruby
mitchellhenke has quit [Read error: Connection reset by peer]
kfpratt has joined #jruby
iamjarvo has joined #jruby
subbu has joined #jruby
mitchellhenke has joined #jruby
<kares>
Nilbus: JRuby-OpenSSL only supports >= 1.47 < 1.51
<kares>
I mean the latest ones jruby-openssl 0.9.6/0.9.7
<headius>
they have weird numbering
<kares>
there's also some work by mkristian to be able to isolate the BC required by OpenSSL (so you can still use the one you want in the app) but it depends on how you deploy
<kares>
... and I'm not sure how it works since I did not try :)
anaeem1 has joined #jruby
iamjarvo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
anaeem1 has quit [Ping timeout: 246 seconds]
enebo has joined #jruby
Aethenelle has joined #jruby
iamjarvo has joined #jruby
vtunka has quit [Quit: Leaving]
<Nilbus>
kares, Thanks! I'll try that update when I get a minute.
<kares>
Nilbus: also no need to go 9K if it does not work but try updating to latest 1.7.x as well (if you keep hitting issues)
baroquebobcat has joined #jruby
<Nilbus>
Yes, we're planning on migrating there first. Hopefully soon.
ponga has quit [Remote host closed the connection]
baroquebobcat has quit [Ping timeout: 252 seconds]
anaeem1_ has joined #jruby
anaeem1_ has quit [Read error: Connection reset by peer]
<mberg>
Running DateTime.parse on ~ 800k strings? 340 seconds
<mberg>
Using SimpleDateFormat#parse on the same ~ 800k strings? 2.8 seconds
skade has quit [Quit: Computer has gone to sleep.]
benlovell has joined #jruby
mitchellhenke has quit [Quit: Computer has gone to sleep.]
mcclurmc has quit [Remote host closed the connection]
skade has joined #jruby
anaeem___ has quit [Remote host closed the connection]
anaeem1 has joined #jruby
djbkd has quit [Remote host closed the connection]
anaeem1 has quit [Ping timeout: 272 seconds]
<mjc_>
mberg: sweet
baroquebobcat has quit [Quit: baroquebobcat]
enebo has quit [Quit: enebo]
codefinger has joined #jruby
codefing_ has joined #jruby
lance|afk is now known as lanceball
<mberg>
That's probably an undersold feature of JRuby. The like to like comparisons with MRI and Rubinius running pure Ruby code don't reflect the potential performance gains of the platform.
codefinger has quit [Ping timeout: 246 seconds]
<mberg>
Almost every time I've found a chokepoint I've been able to make it absurdly faster by refactoring to use Java.
djbkd has joined #jruby
camlow325 has quit [Remote host closed the connection]