skade_ has joined #jruby
skade_ has quit [Ping timeout: 240 seconds]
camlow325 has quit [Quit: WeeChat 1.5]
skade_ has joined #jruby
skade_ has quit [Ping timeout: 240 seconds]
hoodow has quit [Quit: No Ping reply in 180 seconds.]
hoodow has joined #jruby
skade_ has joined #jruby
skade_ has quit [Ping timeout: 260 seconds]
kares has quit [Quit: ...]
etehtsea has joined #jruby
skade_ has joined #jruby
skade_ has quit [Ping timeout: 240 seconds]
skade_ has joined #jruby
skade_ has quit [Ping timeout: 240 seconds]
skade_ has joined #jruby
donV has joined #jruby
skade_ has quit [Ping timeout: 260 seconds]
donV has quit [Read error: Connection reset by peer]
donV_ has joined #jruby
<donV_> Good morning!
shellac_ has joined #jruby
donV_ has quit [Quit: donV_]
shellac_ has quit [Quit: Computer has gone to sleep.]
donV has joined #jruby
skade_ has joined #jruby
skade_ has quit [Ping timeout: 255 seconds]
donV has quit [Quit: donV]
vtunka has joined #jruby
donV has joined #jruby
<GitHub155> [jruby] donv pushed 1 new commit to ruboto_9k: https://git.io/v9XOB
<GitHub155> jruby/ruboto_9k 2834c3c Uwe Kubosch: Reinstate the use of SeekableByteChannel. Patch in Ruboto instead.
<GitHub137> [jruby] donv pushed 1 new commit to ruboto_9k: https://git.io/v9XOS
<GitHub137> jruby/ruboto_9k 4707f66 Uwe Kubosch: Remove unused import
shellac has quit [Ping timeout: 255 seconds]
shellac has joined #jruby
drbobbeaty has joined #jruby
drbobbeaty has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
shellac_ has joined #jruby
shellac_ has quit [Quit: Computer has gone to sleep.]
etehtsea has quit [Ping timeout: 255 seconds]
shellac has quit [Remote host closed the connection]
shellac has joined #jruby
drbobbeaty has joined #jruby
bbrowning_away is now known as bbrowning
skade_ has joined #jruby
<GitHub163> [jruby] ivoanjo opened issue #4596: Oat gem slower at steady-state with invokedynamic than without it https://git.io/v9XDI
skade_ has quit [Ping timeout: 264 seconds]
Aethenelle has joined #jruby
etehtsea has joined #jruby
etehtsea has quit [Ping timeout: 240 seconds]
enebo has joined #jruby
Aethenelle has quit [Quit: Aethenelle]
donV has quit [Ping timeout: 260 seconds]
etehtsea has joined #jruby
bbrowning is now known as bbrowning_away
donV has joined #jruby
Aethenelle has joined #jruby
camlow325 has joined #jruby
hobodave has joined #jruby
shellac has quit [Quit: Leaving]
subbu is now known as subbu|afk
donV has quit [Quit: donV]
subbu|afk is now known as subbu
<GitHub82> [jruby] madrob opened pull request #4597: Update COPYING for bundled java licenses #4587 (master...master) https://git.io/v918Y
<GitHub23> [jruby] enebo pushed 2 new commits to master: https://git.io/v914V
<GitHub23> jruby/master dc1d2da Thomas E Enebo: Merge pull request #4597 from madrob/master...
<GitHub23> jruby/master 81ff75f Mike Drob: Update COPYING for bundled java licenses #4587
<GitHub9> [jruby] enebo closed pull request #4597: Update COPYING for bundled java licenses #4587 (master...master) https://git.io/v918Y
madrob has joined #jruby
<madrob> Hello JRuby folks! Where do more detailed developer conversations typically happen? Is there a mailing list to use, or do folks prefer github issues to discuss, or something else?
donV has joined #jruby
<enebo> madrob: this is probably not a bad place for conversations but if it appears to be a real issue (and not a "how do I") then open an issue for it
<GitHub9> [jruby] madrob closed issue #4587: COPYING is missing Java libraries https://git.io/v9ufU
<madrob> Ok, I'll start here and open an issue if you think it's appropriate.
<madrob> jruby-complete bundles racc/cparse-jruby.jar which is LGPL licensed. This is a problem for ASF projects and means we won't be able to redistribute you guys.
<enebo> madrob: headius you released that didn't you? Is this just a mistake in licensing?
<enebo> heh addressed that a bit wrong :)
<madrob> What does racc do, how important is it? Is it needed at build time, runtime, or both?
<madrob> Oh, should it be tri-licensed EPL/GPL/LGPL like the rest of JRuby?
<enebo> madrob: it is a yacc in Ruby syntax. headius I think ported the C code to Java a while back
<enebo> which was not simple since it had lots of goto flow control
<madrob> If it's a port from C that was LGPL, we might be stuck with the original license.
<enebo> well I do not think it was LGPL but Ruby artistic
<enebo> but I think it might not be all that much of a port either
<madrob> Ruby artistic is fine for us too!
<enebo> when he is not afk he will clue us in
<madrob> Sure, thanks. Is he US or EU?
<enebo> madrob: he is US and about 8km from my house :)
<enebo> madrob: I was talking to him not too long ago so I am sure he is in transit or just afk a bit
shellac has joined #jruby
donV has quit [Quit: donV]
skade_ has joined #jruby
skade_ has quit [Read error: Connection reset by peer]
shellac has quit [Quit: Computer has gone to sleep.]
donV has joined #jruby
<GitHub2> [jruby] erikogan opened issue #4598: Kernel.system fails to write stdout to a file in a Linux environment. https://git.io/v91iA
donV has quit [Client Quit]
shellac has joined #jruby
<GitHub178> [jruby] etehtsea opened pull request #4599: Fix bootstrapping on AARCH64 (master...fix-aarch64-bootstrap) https://git.io/v911E
donV has joined #jruby
donV has quit [Client Quit]
shellac has quit [Ping timeout: 268 seconds]
<GitHub102> [jruby] etehtsea opened issue #4600: Dir.tmpdir fails with error on AARCH64 https://git.io/v91SY
<GitHub167> [jruby] etehtsea opened pull request #4601: Bump jnr-posix. Fixes #4600 (master...4600-tmpdir-fix) https://git.io/v91HZ
etehtsea has quit [Quit: Textual IRC Client: www.textualapp.com]
donV has joined #jruby
madrob_ has joined #jruby
madrob has quit [Ping timeout: 260 seconds]
subbu is now known as subbu|lunch
bbrowning_away is now known as bbrowning
subbu|lunch is now known as subbu
camlow325 has quit [Quit: WeeChat 1.5]
<headius> madrob_: hey
<madrob_> hello sir
<headius> enebo was mistaken, the racc native ext *is* LGPL
camlow325 has joined #jruby
<headius> so we'll have to do something about that
<headius> madrob_: good news
<headius> and enebo
<madrob_> ?
<headius> ok so here's the deail
<headius> deal
<headius> what MRI ships for racc is just the *runtime* and that's all Ruby license
<headius> that includes the extension I ported
<headius> so the ext is fine
<headius> but we thought they were shipping all of the racc *gem*, which includes other code that is LGPL
<headius> so we do still have some LGPL racc code in JRuby but we don't need to
<headius> and before you say it, yes, it's a mess... racc is two pieces that are licensed differently
<madrob_> all licensing is a mess
<headius> that's for sure
<headius> I was thinking I might get through the month without a licensing headache
<madrob_> so the license info on your racc port needs to be updated, and then the jruby licensing can reflect that
<headius> yes, the C code has Ruby license header, so I'll add the same
<headius> also file a bug with tenderlove/racc to fix their COPYING to point out the Ruby-licensed bits
<headius> also file a bug to get the non-LGPL runtime part released as a separate gem
<headius> and also only commit the Ruby licensed stuff to JRuby for future releases
<headius> *sigh*
<headius> madrob_: can you open a bug for this please?
<headius> mentioning ASF etc
hobodave_ has joined #jruby
hobodave has quit [Ping timeout: 260 seconds]
<GitHub84> [jruby] headius opened pull request #4602: Also increment thresholds in indy sites for method_missing. (master...indy_mm_cache_bust) https://git.io/v91jf
madrob has joined #jruby
<madrob> sure, i'll open an issue. can i copy your response in there?
madrob_ has quit [Ping timeout: 260 seconds]
<headius> just describe the issue and I'll comment like it's new
<headius> my train of thought IRC babble is not what I'd include in a bug :-)
<GitHub128> [jruby] madrob opened issue #4603: ASF projects can't bundle and ship jruby9k https://git.io/v9MvW
<madrob> i tried to clean up your thoughts. lmk if i got it wrong
<headius> thank you
<GitHub38> [jruby] madrob reopened issue #4587: COPYING is missing Java libraries https://git.io/v9ufU
<headius> madrob: I'm making some small edits
<headius> it's unfortunately not just a notational issue since we do have the LGPL parts of racc in our releases (until I commit my local changes)
<headius> enebo: high priority on 9.1.9 now
<madrob> any guesses how long that's been around?
<headius> I'm going to see
<GitHub0> [jruby] headius pushed 1 new commit to master: https://git.io/v9MJS
<GitHub0> jruby/master 6852e57 Charles Oliver Nutter: Unbundle racc gem and commit non-LGPL portions as in MRI....
<headius> that should handle master
<GitHub26> [jruby] headius pushed 1 new commit to master: https://git.io/v9MJx
<GitHub26> jruby/master efb775b Charles Oliver Nutter: Remove LGPL Racc section from COPYING. #4603.
<GitHub30> [jruby] headius pushed 1 new commit to master: https://git.io/v9MUl
<GitHub30> jruby/master 77e3fa7 Charles Oliver Nutter: Update gem counts for now removed racc gem. #4603.
<headius> madrob: this has been in place since 2015, so all JRuby 9k releases have it
<headius> 007e2cefde17ce05b8f044ca9594d6cc3f9b68d9
<headius> enebo: ^^
<madrob> thanks
<GitHub100> [jruby] headius closed pull request #4602: Also increment thresholds in indy sites for method_missing. (master...indy_mm_cache_bust) https://git.io/v91jf
<GitHub173> [jruby] headius closed issue #4596: Oat gem slower at steady-state with invokedynamic than without it https://git.io/v9XDI
<GitHub6> jruby/master ef28367 Charles Oliver Nutter: Merge pull request #4602 from headius/indy_mm_cache_bust...
<GitHub6> jruby/master 2c211bf Charles Oliver Nutter: Also increment thresholds in indy sites for method_missing....
<GitHub6> [jruby] headius pushed 2 new commits to master: https://git.io/v9MTc
<GitHub22> [jruby] madrob opened pull request #4604: Include Ruby Licensed json libs in COPYING #4587 (master...master) https://git.io/v9MT6
<GitHub2> [jruby] headius pushed 2 new commits to master: https://git.io/v9MTx
<GitHub2> jruby/master e143db6 Charles Oliver Nutter: Merge pull request #4604 from madrob/master...
<GitHub2> jruby/master 75ed50c Mike Drob: Include Ruby Licensed json libs in COPYING #4587
<GitHub32> [jruby] headius closed pull request #4604: Include Ruby Licensed json libs in COPYING #4587 (master...master) https://git.io/v9MT6
<madrob> do y'all have a timeline for 9.1.9.0 yet?
<madrob> i'm not in a rush, btw
<headius> well it's "ready"
<headius> it's a matter of verification and release, mostly
<headius> I leave on a trip on Tuesday so I'm hoping we can get it out Monday
<headius> enebo: ^^
kares has joined #jruby
<nirvdrum> headius, enebo: Do either of you happen to know which version of ant should be used for building jffi?
<enebo> nirvdrum: no does newest version not work?
<nirvdrum> I keep getting: "build.xml:398: jar doesn't support the nested "archives" element"
<headius> I have not had to use ant for the last several releases, so it has been a while
<nirvdrum> I've tried three different ant versions now.
<nirvdrum> I'm looking to add support for Linux/SPARC.
<headius> others have been able to build the native bits as recently as the past few months
<enebo> nirvdrum: perhaps you can fix it and we can document it only works with version X.Y and above
<headius> I myself have not tried...I think they just build the .so directly and put it in the right place for the archive/ versions to pick it up
<nirvdrum> Hmm...
<enebo> nirvdrum: if we are using something which got removed we should update it anyways
<nirvdrum> I couldn't really tell if ant was supposed to be invoked directly or via maven somehow.
<enebo> nirvdrum: I assume it probably was deprecated for a decade or something :P
<nirvdrum> So maybe I'm just running it wrong.
<headius> I forget who it was...Aethenelle or Antiarc...that was trying to help us set up a full build on multiple VMs
<Aethenelle> that was me...
<headius> hi there!
<nirvdrum> I'd be happy with a simple README to start :-P
<Aethenelle> i got most of the way through it but also pulled onto other things...
<Aethenelle> nirvdrum: that i should be able to get together in a week or so...
<headius> Aethenelle: any help you can provide would be appreciated :-)
<Aethenelle> however, I think sparc may not fully work in qemu
<headius> I think nirvdrum just wants to know how to properly build things
<headius> and it has been a long time since I tried
<headius> the qemu/vm rig would be amazing but I know that's more work
<headius> I have tried to find public/free or local VMs for a lot of these platforms and they just don't exist
<nirvdrum> Correct. There's a build.xml and a pom.xml and the POM seems to also call into ant, but there's also a bunch commented out.
<enebo> nirvdrum: plain 'ant' WFM with:
<enebo> Apache Ant(TM) version 1.9.6 compiled on February 3 2016
<enebo> on FC25
<nirvdrum> enebo: Okay, thanks.
<enebo> nirvdrum: what are you using?
<nirvdrum> Currently, 1.10.1. But I've tried 1.9.9 and 1.7.1 as well.
<enebo> nirvdrum: I only saw one thing online about this and it was a weird SO question
<enebo> nirvdrum: almost like something with processing the XML file was the issue
<GitHub138> [jruby] headius closed issue #4587: COPYING is missing Java libraries https://git.io/v9ufU
<nirvdrum> The JARs actually build. So maybe that's good enough.
<enebo> AHA
<enebo> [jar] jni/x86_64-Linux/libjffi-1.2.so already added, skipping
<enebo> nirvdrum: you building a new .so right?
drbobbeaty has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
<enebo> hahah well removing them still worked
<nirvdrum> New everything.
madrob has quit [Ping timeout: 260 seconds]
<enebo> nirvdrum: I wonder if this works if the jar exists and fails if it does not exist
<enebo> nirvdrum: we have "update = 'true'"
<nirvdrum> Try deleting the prebuilt one then?
<enebo> nirvdrum: well I am just guessing since I got a message like: [jar] Updating jar: /home/enebo/work/jffi/build/native.jar
<enebo> [jar] Warning: selected jar files include a META-INF/INDEX.LIST which will be replaced by a newly generated one.
<nirvdrum> It's the assemble-native-jar one that's crapping out on me.
<enebo> nirvdrum: so noticing this just made me guess if you are truly new work perhaps we have something weird where this only works if jar already exists
<enebo> nirvdrum: and that is my -native-assemble-jar output
<enebo> err assemble-native-jar
<Aethenelle> oh... that's where you are... let me take a stroll through that code...
<nirvdrum> The jar task indeed makes no mention of <archives>: https://ant.apache.org/manual-1.9.x/Tasks/jar.html
<nirvdrum> Oops. I spoke too soon. It's there, but nested under something else.
<nirvdrum> Although that doesn't work either.
<nirvdrum> Man, it's probably been a decade since I've last written an ant build file.
hobodave_ has quit [Quit: Computer has gone to sleep.]
<enebo> nirvdrum: you have verbose or debug or whatnot enabled?
<Aethenelle> looks like it build fine w/ maven just fails all the tests
<nirvdrum> enebo: It just gives a stacktrace with the same message I've posted.
<nirvdrum> Aethenelle: Will that build the native components? If so, what phase or whatever are you running?
<Aethenelle> maven package
<Aethenelle> I want to say, yes but I don't know for sure...
<nirvdrum> mvn package isn't building the native lib as far as I can tell.
<enebo> nirvdrum: what is the latest hashtag in your repo?
<enebo> mine is 2ef7316476b4e9f055844a020895e772ac86a66a
<Aethenelle> looks like it
<Aethenelle> what repo you in? jnr-ffi or jffi?
<nirvdrum> jffi
<nirvdrum> enebo: Same. This is a fresh clone.
<enebo> nirvdrum: this isn't windows is it?
<nirvdrum> enebo: Linux/SPARC.
<enebo> Gnu/Linux/SPARC
<nirvdrum> Heh.
<nirvdrum> sparcv9-Linux
<enebo> nirvdrum: I do see some conditions in -pre-init so I half wonder if it is something weird with options on v9
<enebo> nirvdrum: I am just throwing random shit out there :)
<Aethenelle> let me do a clean and check the jars in JD-GUI
<nirvdrum> I appreciate it.
<enebo> nirvdrum: you running this with graal? :)
<Aethenelle> [INFO] Building jar: /Users/eroot/repos/jffi/target/jffi-1.2.8-SNAPSHOT-native.jar
<Aethenelle> double checking in JD
<nirvdrum> Heh. Nope.
<nirvdrum> Just Java 8.
<Aethenelle> yes, it's building the native stuff...
<Aethenelle> ugh... this thing needs updating...
<enebo> haha
<Aethenelle> you have a sparc box or a VM?
<enebo> yeah there is no instructions and it uses ant and maven.
<enebo> are
<nirvdrum> Aethenelle: Both? I think it's a T7 with a Linux zone allocated on a Solaris host.
<Aethenelle> darn... not a cross platform vm...
<nirvdrum> No, sorry.
bbrowning is now known as bbrowning_away
<Aethenelle> just fexed the pom a little
<Aethenelle> looks like qemu fixed the sparc emulator mostly
<nirvdrum> Aethenelle: If i replace that <archives><zips> bit with <zipgroupfileset>, it seems to do the right thing.
<nirvdrum> But I don't know ant well enough to say if that's an adequate replacement.
<Aethenelle> it seems to work fine for me as is...maybe... what's the replacement w/ zipgroupfileset look like?
<Aethenelle> mvn clean package package
<nirvdrum> Since I can't get it to build the other way, I can't really compare the build/native.jar before & after.
<Aethenelle> I will here...
<nirvdrum> Thanks.
<nirvdrum> I'm still trying to work out why maven doesn't actually build any of the native stuff.
<nirvdrum> <ant antfile="version.xml" dir="." target="-generate-version-source" />
<nirvdrum> I think that's why.
<nirvdrum> The commented out section in the POM ran "-assemble-native-jar". This one just generates a version file.
<Aethenelle> it's building it here... just not in the native jar for some reason...
<nirvdrum> But reports itself as "build-native-library"
donV has quit [Quit: donV]
<nirvdrum> enebo: Any chance you recall doing this? https://github.com/jnr/jffi/commit/147d4bbe26a9ef4b7728a6595931871e1444521f
<nirvdrum> Aethenelle: As far as I can tell, the POM doesn't use the build.xml file. But the build.xml seems to be the only way to actually build the native parts.
<nirvdrum> I guess you're supposed to use ant to bootstrap manually and then use maven from there on out?
<enebo> nirvdrum: not totally but I think it was something which did not work in either my env or ci so I backed it out
<enebo> nirvdrum: I think it was somewhere where I messaged kristian and said it did not work and I was going to revert anticipating he would tweak or something
<enebo> nirvdrum: which I guess did not happen
<Aethenelle> if I rm the jffi-Darwin.jar I still get the library in the non-native jar...
<nirvdrum> Do you see output for make?
<nirvdrum> I don't see how the maven config can build anything native.
<nirvdrum> Unless that fix you alluded to does it. I don't see that pushed yet.
<Aethenelle> and that's because maven doesn't kill the build dir...
<Aethenelle> nah... was a minor fix...
<Aethenelle> uncomment the commented section in pom.xml
<Aethenelle> the native.jar filename needs to be fixed.
<Aethenelle> hrm... not yet enough...
<nirvdrum> I think that ant filename needs to be changed, too.
<Aethenelle> <unjar src="${project.build.directory}/${project.artifactId}-${project.version}-native.jar" dest="${project.build.outputDirectory}" />
<Aethenelle> yup... maven just runs ant...
<Aethenelle> ant build-native gets the build...
<Aethenelle> or would if it wasn't borked...
<nirvdrum> Save for that <archives> issue I'm hitting.
<Aethenelle> jar then build-native succeeds
<nirvdrum> ant jar does build-native though, doesn't it?
<nirvdrum> It's the default ant task.
<nirvdrum> Well, if you're able to confirm the build.xml change is equivalent, I can at least cobble this together manually and submit a PR.
<nirvdrum> Otherwise, I still don't know why ant is complaining.
<nirvdrum> Aethenelle, enebo: Do you see any activity when you from the -assemble-native-jar task?
<Aethenelle> I'm seeing compiling w/ ant build-native
<Aethenelle> ant jar build-native && mv build/native.jar && mvn package
<Aethenelle> w/ custom-build.xml name fixed but not native.jar
<Aethenelle> however, jffi-1.2.8-SNAPSHOT-native.jar doesn't include the darwin library but jffi-1.2.8-SNAPSHOT.jar does
<nirvdrum> Aethenelle: Are you going to push the changes to the POM or do you want me to?
<Aethenelle> I'll push... needs more than just the pom... just firgured you might want to follow along
<nirvdrum> I appreciate you taking the time.
<nirvdrum> I'm just getting ready to leave the office in the next 30 min. or so, so I wanted to make sure I'm not blocking anything.
<Aethenelle> nah.. I should be able to get this sorted mostly by the time i get home
<nirvdrum> Cool. I'm heading home then eating dinner. AFK for a couple hours.
<Aethenelle> maybe not... this is pretty borked... I can at least get it less busted...
Aethenelle has quit [Quit: Aethenelle]
enebo has quit [Quit: Leaving.]