jensnockert has quit [Remote host closed the connection]
vtunka has joined #jruby
<GitHub143>
[jruby] eregon commented on commit 9dd1d4a: @chrisseaton Look in `test/mri/excludes_truffle`, add a line manually with in the file corresponding to the test class and the failing test method. http://git.io/vG4DD
<GitHub139>
[jruby] eregon pushed 1 new commit to master: http://git.io/vG4DH
<GitHub139>
jruby/master 522f220 Benoit Daloze: [Truffle] Update the Eclipse launcher to use full paths for the core library.
havenwood has quit [Ping timeout: 265 seconds]
drbobbeaty has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
dinfuehr has quit [Remote host closed the connection]
jensnockert has quit [Ping timeout: 244 seconds]
havenwood has joined #jruby
skade has joined #jruby
bbrowning_away is now known as bbrowning
pietr0 has joined #jruby
donV has joined #jruby
donV has quit [Client Quit]
erikhatcher has quit [Quit: erikhatcher]
<rtyler>
what's the schedule for 9.0.1.0?
cultureulterior1 has quit [Ping timeout: 240 seconds]
<rtyler>
mkristian told me that a fix we had in 1.7.22 isn't in 9k which makes me le sad
<rtyler>
(two steps forward one step back)
cultureulterior1 has joined #jruby
benlovell has joined #jruby
benlovell has quit [Ping timeout: 240 seconds]
havenwood has quit [Ping timeout: 240 seconds]
cultureulterio-1 has joined #jruby
<headius>
rtyler: what's the fix
<headius>
9.0.1.0 is delayed until Weds so we can get jossl 0.9.10 out
_djbkd has joined #jruby
<headius>
kares has been merging 1.7 to master so I'm not sure why a fix would be missing
cultureulterior1 has quit [Ping timeout: 240 seconds]
<rtyler>
headius: the sub-process invocation of JRuby failing, basically causing extconf.rb's to croak
<rtyler>
it's in master and was in 1.7.22, just not yet release
<rtyler>
*d
<headius>
oh ok
_djbkd has quit [Remote host closed the connection]
<headius>
well, Wednesday
<rtyler>
which I didn't know, so a bit of egg on my face suggesting 9k to people :(
_djbkd has joined #jruby
rsim has quit [Quit: Leaving.]
<headius>
rtyler: ahh, bummer :-(
havenwood has joined #jruby
lance|afk is now known as lanceball
ekinmur has joined #jruby
<nirvdrum>
enebo: I can never keep it straight. Is Struct or StructLayout preferable?
<enebo>
nirvdrum: I think StructLayout but I don’t actually even know the difference
ekinmur has quit [Max SendQ exceeded]
<enebo>
nirvdrum: It is a bit of a sour dough starter for me
<enebo>
nirvdrum: I just look at how FileStat was done and use the same pattern
ekinmur has joined #jruby
<nirvdrum>
Heh. I've been using Timeval as my template.
<enebo>
nirvdrum: great minds…or poor memory ones :)
<nirvdrum>
So, now we've got a sharks & jets situation going on.
<enebo>
nirvdrum: since you use Struct and I use StructLayout?
<nirvdrum>
Yeah. I'd just like to be consistent and have some reasonable justification for it.
<nirvdrum>
headius: Tie breaker?
ekinmur has quit [Max SendQ exceeded]
<headius>
beats me
<headius>
ask wmeissner
* headius
drops the mic
<nirvdrum>
If I could build a time machine, I wouldn't be wasting it on this :-)
ekinmur has joined #jruby
<enebo>
nirvdrum: htf does timeval even work?
bffff_ has quit [Quit: Connection closed for inactivity]
<enebo>
oh I see
<enebo>
same damned thing
<nirvdrum>
I guess Timeval relies on reflection, which may not be preferable.
<enebo>
nirvdrum: but does it?
ekinmur has quit [Max SendQ exceeded]
<enebo>
nirvdrum: the native classes extend that and it has same initializers that StructLayout seems to
<nirvdrum>
Yeah, I'm just looking at: times = Struct.arrayOf(getRuntime(), DefaultNativeTimeval.class, 2);
ekinmur has joined #jruby
<nirvdrum>
But I guess that's an optimization to allocate in native memory.
<nirvdrum>
Code-wise, the Timeval example has much fewer implementations. But I'm not sure if that's just a poor example.
<enebo>
heh…so Struct is in jffi and StructLayout is in jnr-ffi
havenwood has quit [Ping timeout: 252 seconds]
<nirvdrum>
Speaking of which, what is the difference between those two?
<enebo>
StructLayout extends Type and Struct extends Aggregate which extends Type
<enebo>
It is like evolution where some animals put their photoreceptors on opposite of the eye
_djbkd has quit [Remote host closed the connection]
<enebo>
nirvdrum: wayne was constantly improving/changing APIs so you will see like three different Platforms in different packages
<enebo>
nirvdrum: but to answer your question I have no idea what difference there is. I would guess jnr-ffi is written on top of jffi which is very little help
<nirvdrum>
Fair enough.
<enebo>
I would be more embarrassed but I never wrote any code beneath jnr-posix
<enebo>
which I guess might be embarrassing enough
_djbkd has joined #jruby
<GitHub170>
[jruby] bjfish commented on commit 9dd1d4a: That's right, Thx @eregon . Tests within modules go inside folders. The test class names are the names of the exclude files. To exclude a whole test, you can comment it out in mri_truffle.index http://git.io/vGEnr
<enebo>
nirvdrum: I would say we should consider a version X or some big new major number for these and clean house but there is also a JEP so it might be a waste of effort to not just limp along with what we got
<enebo>
kares: I did a release in Maven but does this need a gem push too?
<nirvdrum>
enebo: We can certainly limp along . . . it's what we've been doing and things work. But cleaning house might still make sense if JRuby is going to continue to support older JDK releases.
ekinmur has quit [Quit: left the channel]
<enebo>
nirvdrum: yeah I guess in that case if there is a blessed API for this it might be better to then work on a back compat project
<enebo>
nirvdrum: in any case projects like jnr-posix will still need to exist and be built on something
_djbkd has quit [Remote host closed the connection]
_djbkd has joined #jruby
<xardion>
I take it the spoon gem isn't necessary at all on jruby 9k?
<headius>
xardion: yeah not really
<xardion>
forgot I had that in my code :P
<headius>
spawn works now using posix_spawn and supports most of the behavior of MRI's spawn
_djbkd has quit [Remote host closed the connection]
_djbkd has joined #jruby
tcrawley is now known as tcrawley-away
subbu is now known as subbu|lunch
tcrawley-away is now known as tcrawley
donV has joined #jruby
<nirvdrum>
enebo: Another one for you. Is there a way to read a Pointer into a Struct? Or do I need to walk the Struct fields manually?
<headius>
kares: I'm going to build and push jossl 0.9.10
<headius>
I mean 0.9.11
<xardion>
weird, I'm having an odd posix_spawn issue
<headius>
xardion: do tell
<enebo>
nirvdrum: That’s why we make those struct extending classes right?
<xardion>
thought it was spoon doing it, but its happening with pure ruby
<xardion>
so I'm trying to spawn a java process and background it (with Process.detach), but I keep getting this error:
<nirvdrum>
enebo: I suppose you're right. So I guess I need to dig into jffi/jnr-ffi to see how the translation happens.
<xardion>
error: No such file or directory - /usr/bin/java
<xardion>
should probably clarify this is on OS X, so there may be a gotcha there. works fine on Linux
<headius>
xardion: this is using spoon or spawn?
<xardion>
both
<xardion>
or either
<headius>
hmm
<headius>
I see it's a link, wonder if that has something to do with it
<xardion>
yeah, it's definitely a symlink
<xardion>
but, on the linux box I run it on, it's also a symlink (gentoo)
<headius>
hmm
DomKM has joined #jruby
<headius>
only happening with java or what?
<xardion>
haven't tried with other things
<headius>
ah ok
<xardion>
This particular script I hadn't tried running locally before
<xardion>
so this may have been an issue with spoon and 1.7
<headius>
hmm
<headius>
I don't get it with a simple spawn '/usr/bin/java'
<headius>
but I do get it if I put any arguments into that string...could there be a rogue space at the end or something?
<xardion>
fails with the same error (albeit a different path) if I don't put in the full path to java
<xardion>
well, I should say without the path
<xardion>
i.e. - error: No such file or directory - java
<headius>
weird
<headius>
I removed -jar and everything after that and it works ok here
<headius>
spawn *[...]
<xardion>
maybe I should try it as a single command so that it spawns a shell
<xardion>
right
<headius>
do you get a backtrace?
<xardion>
No, just that error
<xardion>
I think it's the call to posix_spawn that returns that
<xardion>
yeah, fails the same way as a single command line, just spits out the entire command instead
benlovell has joined #jruby
benlovell has quit [Ping timeout: 240 seconds]
<xardion>
just to sanity check, I made sure that command line works from my shell :P
<xardion>
I suppose I could also see if just plain backticks work
<headius>
tried CRuby?
<headius>
checked executable flag?
<headius>
I'm grasping here
<xardion>
oh wait, I think I might've found it...
<donV>
Hi all! How are you all doing?
<xardion>
yup, figured it out
<xardion>
there were line feeds in there
<xardion>
I have no idea how linux was handling that but os x wasn't
<headius>
xardion: ahh so it was whitespace after all
<xardion>
yuuuuup :P
<headius>
that was a shot in the dark
<xardion>
sneaky whitespace
subbu|lunch is now known as subbu
donV has quit [Ping timeout: 244 seconds]
donV has joined #jruby
bffff_ has joined #jruby
subbu has quit [Ping timeout: 256 seconds]
xardion has quit [Ping timeout: 255 seconds]
xardion has joined #jruby
<headius>
xardion: it would be good to know if we're supposed to be chomping that off...can you check MRI?
<headius>
spoon definitely doesn't do any chomping
subbu has joined #jruby
badfish129 has joined #jruby
<badfish129>
I have two jars I need to include in my jruby project, jar A is packaged with dependencies and jar B is not. However, all of jar B’s deps are in jar A. Is there a way to make this work without packaging jar B with all the dependencies also, right now I can only get java_import to work when B is fat too
<headius>
if you require them in order A, B, it doesn't work?
<headius>
it should
yfeldblum has quit [Ping timeout: 246 seconds]
<badfish129>
ah that might be the issue let me check
<badfish129>
still no, I get a NameError when I try to do a java_import on one of the classes in B. If I reinclude all the deps in B then it works fine
skade has quit [Quit: Computer has gone to sleep.]
badfish129 has quit [Quit: badfish129]
cremes has quit [Read error: Connection reset by peer]
cremes_ has joined #jruby
havenwood has joined #jruby
skade has joined #jruby
rsim has joined #jruby
<nirvdrum>
enebo: It looks like you create the struct, then call useMemory().
<enebo>
nirvdrum: ok
badfish129 has joined #jruby
kwando has joined #jruby
skade has quit [Quit: Computer has gone to sleep.]
bbrowning is now known as bbrowning_away
<kares>
enebo: back ... master and 1.7 both seem fine with the staging 0.9.11 release
<enebo>
kares: ah ok great. we just need you to push any local commits (tag) and ‘gem push'
<enebo>
kares: I did a release on maven already
<kares>
cool ... I'll do that now than
benlovell has joined #jruby
<enebo>
crazy nearly all day and no jnr-constants showing up on maven
<enebo>
sonatype shows is in releases repo but search.maven.org nada
<GitHub95>
[jruby-openssl] kares tagged v0.9.11 at master: http://git.io/vGuKu
<GitHub77>
[jruby-openssl] kares pushed 1 new commit to master: http://git.io/vGuKE
<GitHub77>
jruby-openssl/master 3b4bc77 kares: preparing for 0.9.11
<kares>
enebo: seen it take 24h+
<GitHub179>
[jruby] kares pushed 1 new commit to jruby-1_7: http://git.io/vGuK7
<GitHub179>
jruby/jruby-1_7 30f1f4c kares: remove staging repo for jruby-openssl 0.9.11 (use released version)
<kares>
with jruby-rack
badfish129 has quit [Quit: badfish129]
badfish129 has joined #jruby
<kares>
enebo: jruby-openssl gem is pushed and jruby-1_7 is updated to use 0.9.11
<havenwood>
headius: The only blockers for an RVM stable release are mvnw wrapper for systems with mvn 3.3.1 already installed and precompiled CRuby binaries for Linux. (Though we could ship without these.) I've reviewed all the issues and have written release notes.
<enebo>
kares: can you update it on master too
<kares>
enebo: ok
<enebo>
kares: thanks
<havenwood>
headius: When I'm done I need to ping mpapis for final release.
<kares>
I'll just merge jruby-1_7 ... there's only the jruby-openssl update commits
benlovell has quit [Ping timeout: 272 seconds]
<havenwood>
I'll try to do the mrvm wrapper today and ping jrgifford for binaries of they're going to make this release.
<headius>
if there's anything we can do to help that doesn't involve writing bash, let us know :-)
<havenwood>
headius: thanks!
badfish129 has quit [Quit: badfish129]
<havenwood>
A stable release is long overdue. I've just had time to triage issues. I think the star have actually aligned for a release this week though. Fingers crossed!
<nirvdrum>
badfish129: You don't require a class. You require a file. And you should be able to get that from $LOADED_FEATURES. If you really want all the classes known, then I guess you'd want to look at ObjectSpace.
<badfish129>
sorry I meant is there a way to list the classes from the files that are required, loaded features just prints the jars themselves, and objectspace is disabled in jruby
<nirvdrum>
It's disabled by default, but you can enable it.
<nirvdrum>
With -Xobjectspace.enabled=true
<badfish129>
is there a way to enable it just for jirb
drbobbeaty has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
<chrisseaton>
ObjectSpace works for classes by default though
<nirvdrum>
It's just a command-line option. I don't use jirb, but you may need to set JRUBY_OPTS="-Xobjectspace.enabled=true" as part of starting jirb.
<nirvdrum>
chrisseaton: Ahh, right. Yeah, this is just for enabling each_object, I think.
<chrisseaton>
Can you tell us what your end-goal is though? There might be a different way entirely to achieve it.
<badfish129>
okay, I have a couple jars Im including, jar a and jar b. Jar a is a fat jar with all deps, while jar b isnt. Im able to import a class from jar b if I package all the deps, but I get a link error if I don’t, so I’m trying to figure out which deps are missing and resolve them independently
enebo has quit [Quit: enebo]
<chrisseaton>
Could you simply used defined? ClassYoureInterestedIn
<chrisseaton>
Otherwise ObjectSpace.each_object(Class) (which is enabled by default) may be best
<badfish129>
well how would I figure out whats missing? I don’t know which classes are not being included
tcrawley is now known as tcrawley-away
<chrisseaton>
Is there an example class that you know is one of the dependencies included in the fat jar? look for that an it implies you need to load the rest
skade has quit [Quit: Computer has gone to sleep.]
<GitHub31>
[jruby] chrisseaton created truffle-runner-executable (+2 new commits): http://git.io/vGz49
<GitHub31>
jruby/truffle-runner-executable 99fd7b6 Petr Chalupa: [Truffle] add app/gem runner
<GitHub31>
jruby/truffle-runner-executable 8538a43 Chris Seaton: [Truffle] Make the runner executable.
samphippen has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
badfish129 has quit [Quit: badfish129]
<GitHub75>
[jruby] chrisseaton closed pull request #3297: [Truffle] Shim to make it easier to use bundler. (master...truffle-bundler-shim) http://git.io/vGlBF
flori has quit [Remote host closed the connection]
badfish129 has joined #jruby
temporalfox has quit [Read error: Connection reset by peer]