<headius>
if it's totally empty and the call succeeded, that seems like the next guess
mcclurmc has joined #jruby
camlow325 has joined #jruby
<imperator>
oops, good to check return values
<imperator>
yeah, the call itself is failing
<headius>
ok
<headius>
probably not mapped right...let me check something
<imperator>
updated gist
<imperator>
note it works ok with the "A" functions
baroquebobcat has joined #jruby
<imperator>
also, i'd rip out all that win9x crap, but that's me
<headius>
I agree
<headius>
we can use W everywhere, can't we?
<headius>
everything is NT-based now
<headius>
imperator: hmm
<headius>
shouldn't you be passing an array that's size * sizeof(wchar_t)?
<headius>
and by you I mean we, if it's our bug
<imperator>
might be my mistake, hang on
calavera has joined #jruby
<headius>
data = "\0".force_encoding('ASCII-8BIT') * unpackdw(size)
<headius>
from registry.rb QueryValue
<imperator>
all that does is append [0].pack('V')
<headius>
some day I'll memorize the pack language
<headius>
yeah
<imperator>
don't think the size is wrong, works with MRI, and doubling it had no effect
<headius>
ok
<headius>
ohh!
<headius>
what is size after the first call?
calavera_ has joined #jruby
<headius>
RegQueryValueEx updates that size, and maybe we're not treating it as an inout param
calavera has quit [Ping timeout: 244 seconds]
<imperator>
aha, with jruby it's 0
<headius>
there ya go
<imperator>
hm, why is it 0 though?
<headius>
yeah I'm looking at msdn now
<imperator>
guess i need to check the error code
<imperator>
error_more_data vs something else
<headius>
there are a few paths in docs that have undefined result
<headius>
which could be 0
<headius>
I love win32: "If the function succeeds, the return value is ERROR_SUCCESS."
fivebats has joined #jruby
<headius>
imperator: I have to bail out for dinner...update an issue with your findings, and feel free to suggest your favorite path forward on improving our versions of this stuff :-)
Hobogrammer has joined #jruby
<headius>
I'd love to see your win32 ffi stuff replace what we hacked together from MRI's logic
<imperator>
ok
yfeldblum has quit [Ping timeout: 272 seconds]
* imperator
updated gist again to show real failure point
sferik has joined #jruby
triple_b has joined #jruby
calavera_ is now known as calavera
<sferik>
headius: I’m working on my first JRuby patch and failing miserably
mitchellhenke has quit [Quit: Computer has gone to sleep.]
mrmargolis has joined #jruby
<chrisseaton>
sferik: what's the problem?
<sferik>
hey chrisseaton…to be honest, I’m not sure
mcclurmc has quit [Remote host closed the connection]
<headius>
that's probably all that was wrong then...by default, even with newer JDKs installed, it seems like OS X tries to load its own Java 6 from 8 years ago
pietr0 has quit [Quit: pietr0]
<sferik>
headius: so, I’m somewhat baffled that this error was even possible in the first place
<headius>
which error do you mean?
<sferik>
headius: the bug that my patch is attempting to fix
<sferik>
headius: it seems like there ought to be a test suite that calls the #methods method for every type of object in CRuby and make sure the equivalent objects in JRuby respond to the same methods
<sferik>
headius: I know JRuby has some extra methods that it adds to Object
<headius>
ahh yes
<sferik>
headius: I guess it’s just very surprising to me that such a fundamental bug could still exist in JRuby
<headius>
well there's lots of those tests, but none of them confirm they're *not* defined too high in the hierarchy
<sferik>
headius: I suspect someone else would have caught it before me ;)
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] chrisseaton pushed 3 new commits to master: http://git.io/Fste
<JRubyGithub>
jruby/master 9549565 Chris Seaton: [Truffle] Remove dead getCurrentMethod (Truffle provides it now).
<JRubyGithub>
jruby/master 17be927 Chris Seaton: [Truffle] -Xtruffle.passalot option to periodically pass between threads.
<JRubyGithub>
jruby/master 0919714 Chris Seaton: [Truffle] Stack server (jstack for Truffle) prototype.
JRubyGithub has left #jruby [#jruby]
<headius>
sferik: that surprises us every bug we get
<headius>
some more than others
mrmargolis has quit [Remote host closed the connection]
<headius>
we have meant to build a nice tool for auditing our method defs versus MRI, but it's a messy script to write and we've just one-offed it periodically
<sferik>
headius: fair enough
<headius>
these have also moved around from version to version
metadave has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
<headius>
a tool would be nice
<sferik>
headius: fair point
<sferik>
headius: (about things moving around)
<sferik>
headius: I was surprised to find Rubinius code in core/src/main/ruby/jruby/truffle/core
<sferik>
headius: how is that code used?
<sferik>
headius: or is it just to make RubySpec happy?
calavera has joined #jruby
<chrisseaton>
sferik: that's Truffle code - it's only used for the new Truffle backend, not conventional JRuby
<chrisseaton>
chrisseaton: we're experimenting with using Rubinius' core library implementation
<chrisseaton>
sferik: it has plusses and minuses we're finding
<sferik>
chrisseaton: what are some of the plusses and minuses?
<headius>
talking to ones' self among them
<chrisseaton>
ha ha
<chrisseaton>
the primitive system is pretty nice, but it seems to define lots of extra non-standard methods and we'd like to hide those
saga65 has quit [Quit: (null)]
<chrisseaton>
also we're finding it's quite complicated in places - I think on Friday I found that Time.now.to_f creates several temporary Rational objects
<chrisseaton>
but all in all, it's massively accelerating our pace of development
<sferik>
chrisseaton: that's good to hear
pietr0 has joined #jruby
<headius>
no coding is almost always faster than any coding :-)
<chrisseaton>
yes that is the advantage which trumps all others
<chrisseaton>
we support 25% of core today
pgokeeffe has joined #jruby
nateberkopec has quit [Quit: Leaving...]
<headius>
imperator: very nice
<imperator>
:)
calavera has quit [Ping timeout: 264 seconds]
mcclurmc has joined #jruby
imperator has quit [Quit: Valete!]
nateberkopec has joined #jruby
nateberkopec has quit [Client Quit]
pgokeeffe has quit [Quit: pgokeeffe]
sferik has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
yfeldblum has quit [Remote host closed the connection]
pietr0 has quit [Quit: pietr0]
pgokeeffe has joined #jruby
ivan\ has quit [Remote host closed the connection]
ivan\ has joined #jruby
ivan\ has joined #jruby
ivan\ has quit [Changing host]
Who has joined #jruby
<headius>
boo overcast, no asteroid-gazing tonight
yfeldblu_ has joined #jruby
Who has quit [Quit: Who]
deobalds has joined #jruby
deobalds has quit [Client Quit]
<yipdw>
that or use Qt and let it handle all the stuff that I don't want to use
<yipdw>
oops
<yipdw>
tricky mice
Neomex has quit [Ping timeout: 252 seconds]
metadave has joined #jruby
colinsurprenant has joined #jruby
mcclurmc has quit [Remote host closed the connection]
colinsurprenant has quit [Client Quit]
colinsurprenant has joined #jruby
colinsurprenant has quit [Client Quit]
metadave has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] headius opened issue #2521: Require order appears to impact JRuby extensions [9k] http://git.io/FsKa
JRubyGithub has left #jruby [#jruby]
diegoviola has quit [Quit: WeeChat 1.1]
pgokeeffe has quit [Quit: pgokeeffe]
colinsurprenant has joined #jruby
kares has joined #jruby
pgokeeffe has joined #jruby
johnmuhl has quit [Quit: Connection closed for inactivity]
triple_b has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
colinsurprenant has quit [Quit: colinsurprenant]
deobalds has joined #jruby
pgokeeffe has quit [Quit: pgokeeffe]
yfeldblu_ has quit [Remote host closed the connection]
yfeldblum has joined #jruby
ivan\ has quit [Remote host closed the connection]
ivan\ has joined #jruby
slyphon has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
subbu has joined #jruby
pchalupa has quit [Quit: Computer has gone to sleep.]
yfeldblum has quit [Ping timeout: 245 seconds]
yfeldblum has joined #jruby
yfeldblum has quit [Remote host closed the connection]
pgokeeffe has joined #jruby
noop has joined #jruby
skade has joined #jruby
JohnBat26 has joined #jruby
yfeldblum has joined #jruby
skade has quit [Ping timeout: 256 seconds]
skade has joined #jruby
subbu has quit [Ping timeout: 244 seconds]
pchalupa has joined #jruby
temporalfox has joined #jruby
deobalds has quit [Quit: Computer has gone to sleep.]
skade has quit [Quit: Computer has gone to sleep.]
pgokeeffe has quit [Quit: pgokeeffe]
skade has joined #jruby
skade has quit [Ping timeout: 265 seconds]
pgokeeffe has joined #jruby
ivan\ has quit [Read error: Connection reset by peer]
ivan\ has joined #jruby
multibot_ has quit [Remote host closed the connection]
multibot_ has joined #jruby
kares has quit [Ping timeout: 264 seconds]
kares has joined #jruby
shellac has joined #jruby
elia has joined #jruby
pgokeeffe has quit [Quit: pgokeeffe]
shellac has quit [Quit: Computer has gone to sleep.]
dinfuehr has joined #jruby
vtunka has joined #jruby
Hobogrammer has quit [Ping timeout: 255 seconds]
pgokeeffe has joined #jruby
drbobbeaty has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] GennadySpb opened pull request #2522: Implement pack & unpacking ipv6 address in sockaddr (master...master) http://git.io/FGS5
JRubyGithub has left #jruby [#jruby]
yfeldblum has quit [Read error: Connection reset by peer]
yfeldblum has joined #jruby
deobalds has joined #jruby
pgokeeffe has quit [Quit: pgokeeffe]
havenwood has quit [Remote host closed the connection]
drbobbeaty has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
marr has joined #jruby
deobalds has quit [Ping timeout: 245 seconds]
PragTob has joined #jruby
deobalds has joined #jruby
kaawee has joined #jruby
deobalds has quit [Ping timeout: 264 seconds]
deobalds has joined #jruby
ivan\ has quit [Read error: Connection reset by peer]
ivan\ has joined #jruby
kaawee has quit [Quit: Konversation terminated!]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] eregon pushed 1 new commit to master: http://git.io/FZTA
JRubyGithub has left #jruby [#jruby]
<JRubyGithub>
jruby/master 0f550c5 Benoit Daloze: [Truffle] Use a ThreadPassNode directly rather than resolving constant and such....
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] eregon pushed 1 new commit to master: http://git.io/FZIo
<JRubyGithub>
jruby/master a12b6f1 Benoit Daloze: [Truffle] Rename socket to port.
JRubyGithub has left #jruby [#jruby]
dinfuehr_ has joined #jruby
dinfuehr_ has quit [Remote host closed the connection]
dinfuehr has quit [Ping timeout: 255 seconds]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] djberg96 opened pull request #2523: Modified the make_wstr method so that it appends a null character. (master...master) http://git.io/FZmW
<warbot_>
Hello, I've got a small problem building capybara-webkit gem with jruby 1.7.18: mkmf.rb can't find header files for ruby at /jruby-1.7.18/lib/native/include/ruby/ruby.h
<warbot_>
`find /jruby-1.7.18 -name ruby.h -print` gives no result. Could anyone help, please?
tcrawley-away is now known as tcrawley
BurningChrome_ has joined #jruby
deobalds has quit [Quit: Computer has gone to sleep.]
noop has joined #jruby
BurningChrome has quit [Ping timeout: 256 seconds]
shellac has joined #jruby
sferik has joined #jruby
sferik has quit [Client Quit]
<headius>
warbot_: JRuby 1.7 doesn't support C extensions..was it capybara-webkit that failed to install?
sferik has joined #jruby
<warbot_>
headius: yeah
mrmargolis has joined #jruby
colinsurprenant has joined #jruby
<warbot_>
headius: to be precise: `jruby -S gem install capybara-webkit -v '1.4.0'` => Building native extensions. This could take a while... ERROR: Error installing capybara-webkit: ERROR: Failed to build gem native extension.
<warbot_>
so there's no way install capybara-webkit with jruby?
<rtyler>
wat was the error building native extensions? can you gist the whole thing?
<rtyler>
oh, there's the thing up there
<rtyler>
ruby.h
<rtyler>
warbot_: I'm pretty sure you're up shit's creek, unless you feel like patching capybara-webkit to use FFI
<warbot_>
rtyler: :( I see, thanks
<bbrowning>
warbot_: I've had good luck using phantomjs and capybara with the poltergeist gem on jruby
<warbot_>
bbrowning: heh, I was actually looking for webkit because issues with poltergeist's randomness
<chrisseaton>
headius: we're still a bit stuck on making -X+T work in bin/jruby - we're not the ones who made the changes so we're not sure what the best thing to do is to make it work again - any advice?
<headius>
hmm
<headius>
chrisseaton: I thought mkristian was going to take the next step...doesn't a full build put truffle into lib/jruby.jar?
<headius>
I didn't like the way it's structured, exactly, but I thought that was working
<headius>
I see a lot of truffle stuff in lib/jruby.jar
<chrisseaton>
I'll try from scratch...
<headius>
how have you been doing incremental builds?
<chrisseaton>
mvn package
<headius>
my suggestion to mkristian was that we should make the top-level pom.xml responsible for building a working jruby env, which might clean up some of the weirdness
<chrisseaton>
(although we're making our own tool for doing builds and running things - I'll show you next time you want to experiment with Truffle as it makes a lot of things easier)
enebo has joined #jruby
thsig has quit [Remote host closed the connection]
<chrisseaton>
headius: with what I think is a clean build I still get a message saying Truffle's removed - but maybe I'm not cleaning the right thing, or something else, but as I don't understand the big picture I'm not sure what could be wrong
<chrisseaton>
I'll ping mkristian about it, but we're not sure what to do really, and we could do with finding a way to fix it before FOSDEM
colinsurprenant has quit [Quit: colinsurprenant]
mitchellhenke has joined #jruby
<rtyler>
chrisseaton: he and I are both coming fwiw
<headius>
hmmm
<headius>
chrisseaton: ok
<headius>
I think I know the problem
thsig has joined #jruby
subbu has quit [Ping timeout: 245 seconds]
mister_solo has joined #jruby
<headius>
ahh yes
ivan\ has quit [Read error: Connection reset by peer]
<headius>
this explains why it works fine in maven-based test runs
ivan\ has joined #jruby
<chrisseaton>
yeah that was also confusing
<headius>
ruby.class.getClassLoader() will return null when running on boot classpath
<headius>
so if you run with VERIFY_JRUBY=1 it will work
<headius>
or if you run after I fix it, it will work
treehau55 has joined #jruby
subbu has joined #jruby
noop has joined #jruby
<headius>
chrisseaton: I have a fix for you
<chrisseaton>
are you pushing it?
cajone has quit [Remote host closed the connection]
<chrisseaton>
got it
<headius>
chrisseaton: try that and let me know if it's working right
<headius>
worked ok for -e 1
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] eregon pushed 6 new commits to master: http://git.io/FniJ
<JRubyGithub>
jruby/master 27e480c Benoit Daloze: [Truffle] Use the provided HiddenKey so it is not iterated by default.
<JRubyGithub>
jruby/master 737a821 Benoit Daloze: [Truffle] Forgot to create a new array in Array(Object[])#sort.
JRubyGithub has left #jruby [#jruby]
<chrisseaton>
headius: that looks like it fixes it - thanks very much
nirvdrum has joined #jruby
cajone has joined #jruby
mcclurmc has joined #jruby
subbu has quit [Ping timeout: 256 seconds]
noop has quit [Ping timeout: 245 seconds]
mcclurmc has quit [Ping timeout: 272 seconds]
rsim has quit [Quit: Leaving.]
subbu has joined #jruby
mcclurmc has joined #jruby
havenwood has joined #jruby
pchalupa has quit [Quit: Computer has gone to sleep.]
dinfuehr has joined #jruby
vtunka has quit [Ping timeout: 245 seconds]
calavera has joined #jruby
metadave_ has joined #jruby
marr has joined #jruby
vtunka has joined #jruby
metadave has quit [Ping timeout: 245 seconds]
e_dub has joined #jruby
DrShoggoth has joined #jruby
mister_solo has quit [Ping timeout: 252 seconds]
camlow325 has joined #jruby
slyphon has joined #jruby
<eregon>
headius: we have problems that maven also simply never recompiles anything of truffle, so clean+package was needed all the time
kwando has quit [Read error: Connection reset by peer]
kwando has joined #jruby
<eregon>
the compile plugin seems very weird and buggy when there are multiple projects :/ it tells you it did not recompile anything but actually does
calavera has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
<eregon>
how do I get that fix? :)
viking has joined #jruby
viking has joined #jruby
<headius>
eregon: hmm
<headius>
that's not fun
subbu has quit [Ping timeout: 245 seconds]
<headius>
so like, you make a change in truffle stuff and it doesn't pick it up without a clean?
<eregon>
I tried replicating the "eclipse hack" of core's pom.xml and it kind of works except it still does not say anything about when it recompiles
<eregon>
yeah
<headius>
that's weird
<headius>
you might try mvn -pl truffle to force that module to rebuild
<headius>
I think we need to work with mkristian on restructuring the build a bit...it has gotten hard to follow
<headius>
chrisseaton: ok cool
<eregon>
do you have any experience with the forceJavacCompilerUse flag?
<eregon>
[INFO] Nothing to compile - all classes are up to date
<eregon>
but it is a lie very clearly
pgokeeffe_ has joined #jruby
<headius>
eregon: huh, you're right
<headius>
oh what the heck
<headius>
there's still truffle sources in core
<eregon>
where?
balo_ has joined #jruby
<headius>
ahh nevermind, it's just the TruffleBridge interface
<headius>
I think this may be a problem with it seeing the end result is jruby.jar
<headius>
core builds first, so it builds lib/jruby.jar
<headius>
modifications that happened in truffle then look like they're older than the jar, so it doesn't build
<headius>
in theory :-)
<eregon>
ahah
pgokeeffe has quit [Read error: Connection reset by peer]
pgokeeffe_ is now known as pgokeeffe
brocktimus has quit [Ping timeout: 276 seconds]
balo has quit [Remote host closed the connection]
<eregon>
yeah might, although it do seem to work for rtivial changes now, is Maven non-deterministic?
<headius>
see, shit like this just pisses me off...this page is absolutely useless but for the tiny link at the bottom: http://maven.apache.org/ plugins/maven-shade-plugin/usage.html
<headius>
there's basically no information at all
<headius>
and the samples page isn't much better
<eregon>
yeah :(
brocktimus has joined #jruby
<headius>
yeah I think this needs my suggestion...do the final build of lib/jruby.jar in either lib or root poms
<headius>
eregon: I will talk to mkristian about fixing this
<chrisseaton>
there's also still our core Ruby code in core, I'll move that over
<headius>
ok
<eregon>
mmh, now it does look like bad Eclipse/mvn interaction
dfr|work has quit [Ping timeout: 245 seconds]
<enebo>
Will this get rid of the truffle compile error messages if you make a syntax error in main JRuby sources on second build?
<eregon>
headius: sry for bothering you, it seems fixed, I forgot to uncheck some config in Eclipse so it uses a different build dir for all source folders
<enebo>
That annoys the shit out of me :)
<eregon>
enebo: mvn -pl core should do that, no?
<enebo>
I don’t know
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] eregon pushed 1 new commit to master: http://git.io/FcWR
<enebo>
tarcieri: ack. Maybe end of next week. We have spent so much time working towards pre1 that we have done very little for 1.7.19 yet
<enebo>
tarcieri: Need to review what we have marked for 19
<enebo>
tarcieri: but knowing people are hitting common issues in Rails 4.2 is a major motivator
calavera has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
elia has quit [Quit: Computer has gone to sleep.]
shellac has joined #jruby
<headius>
hmmm
<headius>
well we can spin with what we have if we need something out quickly
<headius>
but a triage would be good
<mrmargolis>
enebo: yeah, I am also blocked on 4.2 because of that issue and I think also ar-jdbc.
<enebo>
heh
<enebo>
headius: yeah I just have not seen what has landed on there recently either
<enebo>
multi-branch dev is tough :)
<headius>
indeed
<enebo>
maybe thursday
<enebo>
I guess I will spend some quality 1.7 time and see what we planned on doing and what was landed to make sure we did not introduce something icky
camlow32_ has joined #jruby
<enebo>
tarcieri: If you or someone you work with can test jruby-1_7 branch against 4.2 and give some seal of approval it will work for you that would be really helpful too
camlow32_ has quit [Read error: Connection reset by peer]
<enebo>
I would hate to release for this one API change and have it not be enough
camlow32_ has joined #jruby
<enebo>
mrmargolis: not sure about ar-jdbc. That I think is something kares would be able to answer
calavera has joined #jruby
camlow325 has quit [Ping timeout: 276 seconds]
camlow32_ has quit [Remote host closed the connection]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] headius pushed 1 new commit to master: http://git.io/FCbY
<JRubyGithub>
jruby/master c601151 Charles Oliver Nutter: Uh...
JRubyGithub has left #jruby [#jruby]
<tarcieri>
enebo: someone's gonna give it a try
<enebo>
tarcieri: yeah this is a good opportunity to get something else fixed too :)
camlow325 has joined #jruby
camlow325 has quit [Remote host closed the connection]
<headius>
mkristian: so you talked about a compromise on the classloader thing, making it configurable
yfeldblum has joined #jruby
<headius>
we accept there are cases where it should be the current way, but I think a majority of users are deploying in situations where that's wrong or unnecessary
<headius>
e.g. everyone deploying in a way file potentially setting class collisions
<headius>
way=war
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] michaelbuckner opened issue #2525: undefined method `invoke_with_build_args' for nil:NilClass on Windows 7 http://git.io/FCpF
JRubyGithub has left #jruby [#jruby]
subbu has quit [Ping timeout: 240 seconds]
skade has quit [Quit: Computer has gone to sleep.]
<headius>
sigh...I guess we deserve to get mostly Windows bugs for shipping without a big Windows push
<enebo>
I do wonder how this is happening. I test gem install using windows 7
bbrowning is now known as bbrowning_away
e_dub has quit [Quit: e_dub]
camlow325 has joined #jruby
camlow325 has quit [Remote host closed the connection]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] Who828 opened pull request #2526: Fixed a bug with local_variables ordering and duplicate variables (master...local_variables_bug) http://git.io/FWJA
JRubyGithub has left #jruby [#jruby]
sferik has joined #jruby
baroquebobcat has quit [Quit: baroquebobcat]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] enebo closed pull request #2526: Fixed a bug with local_variables ordering and duplicate variables (master...local_variables_bug) http://git.io/FWJA
JRubyGithub has left #jruby [#jruby]
baroquebobcat has joined #jruby
<mkristian>
headius, I wanted to make the jruby-1_7 way the default and the current one the opt-in, just for those few cases where this might solve problems.
<mkristian>
will make some time for it tomorrow
ivan\ has quit [Read error: Connection reset by peer]
ivan\ has joined #jruby
<headius>
mkristian: that will help us sleep a little easier :-)
<headius>
mkristian: I have a local change to make lib build lib/jruby/jar
<headius>
spent a little time fiddling with other parts of build to try to speed it up but I can't get below 21s for mvn with no changes
sferik_ has joined #jruby
<mkristian>
headius, I was thinking about this. would it be possible to have lib/jruby-core.jar + lib/jruby-truffle.jar ?
<mkristian>
bin/jruby.bash can handle this more or less good
<headius>
it would, but does it gain us anything over shading?
Who has joined #jruby
<headius>
the main challenge there is that jruby-launcher needs to work on 1.7 and 9k and would have to know different layout for 9k
<headius>
a legacy issue
<headius>
anything we do to jruby.bash has to be done to the launcher
sferik__ has joined #jruby
sferik_ has quit [Read error: Connection reset by peer]
<headius>
enebo: what was your issue about building twice after a syntax error?
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] chuckremes opened issue #2527: Can't set a range of cells in Excel using win32ole http://git.io/FWsT
JRubyGithub has left #jruby [#jruby]
<mkristian>
the gain would be: mvn -pl core would build the core and mvn -pl truffle the truffle. so whatever you work on you just need to shade one part
sferik has quit [Ping timeout: 264 seconds]
<enebo>
headius: if you have a syntax error in somethng you are working on it sometimes will end up complaining about truffle and not the jruby syntax error
<enebo>
headius: tbh I do not always get it but when I see it I know I can mvn clean to see my error again
<headius>
oh that one...has been around for quite a while?
<headius>
so not related to the truffle split
<headius>
mkristian: that's a very good point
<headius>
honestly lib/jruby.jar could also just look for jruby-truffle.jar in same dir, since we're already doing classloader stuff to get it
<headius>
that doesn't even require a launcher change
<mkristian>
headius, classloader stuff with truffle - surprised !?
<headius>
what I have now is 1:42 for a mvn clean package and :21 for mvn immediately after that
<headius>
mkristian: I had to fix your getClass().getClasLoader() call to load truffle...it uses JRubyClassLoader now
<headius>
Ruby.class.getClassLoader() is null when running from boot classpath
bbrowning_away is now known as bbrowning
<mkristian>
so bin/jruby.bash just adds all the jars from lib/* to the classpath. jruby.jar goes into the bootclassloader
<mkristian>
ok the null classloader again - yes.
<headius>
correct
<headius>
heh, and you just said a magic word if true... the launcher already adds lib/* to classpath?
<mkristian>
I played around with lib/jruby-core.jar and lib/jruby-truffle.jar which just worked.
thsig has quit [Remote host closed the connection]
<headius>
mkristian: yeah this can work for me
<mkristian>
only lib/jruby.jar goes into the bootclassloader
<headius>
with my fix it should be fine
<headius>
then truffle guys get even better build with -pl truffle if they only want to build truffle
<headius>
right
<mkristian>
yes
<mkristian>
and even core is a bit faster
<headius>
yeah, less sources
<headius>
fewer
<headius>
less source
<headius>
whatever
<headius>
stupid English
<headius>
mkristian: ok, let's do that
<headius>
I will back out the shade moves I have locally and just commit one or two other small things
<headius>
mkristian: so just so I understand, truffle and core will each have an additional shade step that dumps into lib
<mkristian>
yes, unless you want the oracle-truffle jar as well in lib/
<headius>
jruby-truffle shading in oracle truffle and core shading what it did before truffle
<headius>
I don't think we need to do that
<mkristian>
yes
<headius>
just shade them to be standalone for lib
<mkristian>
yes, core and truffle both shade their dependencies
<headius>
ok
<headius>
lib should still depend on truffle, so "mvn" builds both
<headius>
I will push that
<mkristian>
yes, it should do
sferik has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] headius pushed 2 new commits to master: http://git.io/FWBW
<JRubyGithub>
jruby/master 09721d6 Charles Oliver Nutter: Hack to make package-info.java emit a dummy class....
<JRubyGithub>
jruby/master dbdfc4e Charles Oliver Nutter: Make lib depend on truffle for jruby-truffle jar.
JRubyGithub has left #jruby [#jruby]
<headius>
mkristian: your turn!
sferik__ has quit [Ping timeout: 264 seconds]
<mkristian>
headius, but I guess I do the classloader thingy first - for the peace of mind ;)
<headius>
ok, I can make the shading changes
skade has joined #jruby
skade has quit [Client Quit]
<mkristian>
it should be already there since I need the exact same split for jruby-jars.gem - probably just profile or so
<headius>
yeah I'll figure it out
subbu has quit [Ping timeout: 246 seconds]
e_dub has joined #jruby
camlow325 has joined #jruby
subbu has joined #jruby
Who has quit [Quit: Who]
elia has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] nbroyles opened issue #2528: RegexpError: invalid multibyte character http://git.io/FW2w
JRubyGithub has left #jruby [#jruby]
enebo has quit [Quit: enebo]
subbu has quit [Ping timeout: 245 seconds]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] headius pushed 1 new commit to master: http://git.io/FWw0
<JRubyGithub>
jruby/master 576ecb3 Charles Oliver Nutter: Have jruby-core and jruby-truffle shade into lib.
JRubyGithub has left #jruby [#jruby]
<headius>
mkristian: ok, it's in
<mkristian>
:)
<headius>
chrisseaton, eregon: let me know how that feels... mvn at root should build everything correctly, but mvn -pl truffle is only 6 seconds with no changes
<headius>
mvn -pl core is 12 seconds, mvn with no changes is 23s
<headius>
I think that's decent
calavera has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
_whitelogger__ has joined #jruby
kares has quit [Ping timeout: 272 seconds]
Freaky has quit [Remote host closed the connection]
<mkristian>
I just pulled master and test/pom.xml and test/pom.rb match
<headius>
hmmm
baroquebobcat has quit [Quit: baroquebobcat]
<mkristian>
headius, had a chat with jason v.zyl yesterday - after groovy has an uncertain future he is keen in getting the ruby DSL into proper maven. so there is small chance it will happen ;)
baroquebobcat has joined #jruby
<headius>
mkristian: that would be spectacular
<headius>
rake test:jruby
<headius>
oops
baroquebobcat has quit [Client Quit]
rcvalle has joined #jruby
triple_b has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
<enebo>
nirvdrum: so it is funny to see OSS projects on windows because they will be using like a 10 year old version of ICC
sferik has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
skade has quit [Read error: Connection reset by peer]
skade has joined #jruby
<nirvdrum>
Ahh. The only ones I've seen have used VC++. But then it's always fun to see those using the express editions complaining that they can't compile if the project uses ATL.
subbu has joined #jruby
<nirvdrum>
I don't know if that mess is cleared up with the newest community version.
<enebo>
nirvdrum: ifyou want to use win32ole you need to have a non-free version because they do not distribute the header files
<headius>
ugh
<headius>
I wasn't using the launcher either
sferik has joined #jruby
<enebo>
YAY
<enebo>
EPIC RAGE QUIT
fridim_ has quit [Ping timeout: 264 seconds]
<headius>
actually maybe I was
<headius>
it does work ok
<headius>
yeah it works ok...I think the windows logic in jruby-laucnher is just missing the lib stuff