<GitHub55>
[jruby] chrisseaton pushed 1 new commit to truffle-head: http://git.io/vGPbd
<GitHub55>
jruby/truffle-head 6820636 Chris Seaton: Merge branch 'master' into truffle-head
skade has quit [Quit: Computer has gone to sleep.]
erikhatcher has quit [Quit: erikhatcher]
skade has joined #jruby
_whitelogger has joined #jruby
mkristian has joined #jruby
<mkristian>
enebo, hi, are you here ?
<enebo>
mkristian: sure
<enebo>
mkristian: howdy
<enebo>
mkristian: I checked contents of that .jar and it is just core
<mkristian>
so you "-complete" postfix. that is nothing new and just meant to distinguish the jruby-core the maven artifact from lib/jruby.jar
<mkristian>
enebo, so you are saying we do not pack the right jar ?
<enebo>
mkristian: 1.7.22 jruby-jars does not have that
<enebo>
mkristian: it is jruby-core so I think it is the right jar
mje113 has joined #jruby
<mkristian>
enebo, if it is jruby-core on 1.7.22 that is misses all those shaded jars like jnr-*.jar
<enebo>
mkristian: so you are saying jruby-core-version-complete.jar is correct?
dinfuehr has joined #jruby
<mkristian>
so the point here is to split jruby-complete into two jars. the jruby-core artifact is just the classes from ./core/src/main/java without any dependencies shaded into it.
benlovell has quit [Ping timeout: 255 seconds]
<mkristian>
enebo, not sure, in the end it is just naming the thing. does the gem contains the two jars ?
<mkristian>
and they name needs to match the ruby file
<enebo>
mkristian: I believe jruby-core-*.jar is what we think of as just the code files
<enebo>
mkristian: I did not notice for 9.0.0.0 (or currently) that it had -complete.jar on the end
<enebo>
mkristian: 1.7.x has never had this so I was only asking about it
<enebo>
mkristian: if it is so that maven artifact names do not conflict that is ok. It just confused me
<mkristian>
enebo, so jruby-core-9.0.0.0-complete.jar + jruby-stdline.jar == jruby-complete.jar
<mkristian>
jruby-core.jar is the module maven artifact
<enebo>
mkristian: ok so this was to make sure maven did not conflict then
<enebo>
mkristian: we must not have a jruby-stdlib maven artifact then
<enebo>
mkristian: otherwise it would still conflict
<mkristian>
or people do replace the one with the other !
dinfuehr has quit [Ping timeout: 250 seconds]
<enebo>
mkristian: ok well not an issue then. I just thought it looked weird and wanted to know if it was for a reason and that reason is to not overlap names with the same maven artifact
<mkristian>
jruby-complete is not a good maven artifact since it bundles all its dependencies. so org.jruby:jruby has two dependencies jruby-core and jruby-stdlib
<enebo>
mkristian: oh so we have a jruby-stdlib artifact?
<enebo>
mkristian: so current naming of jruby-jars is not quite right since jruby-stdlib has no -complete on the end of its name
<enebo>
mkristian: sure. the word -complete through me off since that is what we call the pre-split version
<enebo>
so seeing core and complete in the same name looked like an error
<mkristian>
other projects call it shaded instead of complete :)
<enebo>
mkristian: fwiw, since jruby-jars is consumed through lib/jruby-jars.rb the name could be anything so I was not worred about it being broken
<enebo>
mkristian: shaded is maven-specific but it does seem like a less overloaded term for us
<enebo>
mkristian: since as you said above stdlib + core = complete
<mkristian>
stdlib + core-shaded = complete
<enebo>
mkristian: yeah seems less confusing to me but I am only one person
Aethenelle has joined #jruby
dumdedum has quit [Read error: Connection reset by peer]
knu has quit [Quit: Reboot...]
SynrG has quit [Read error: Connection reset by peer]
SynrG has joined #jruby
subbu has joined #jruby
knu has joined #jruby
vtunka has quit [Quit: Leaving]
<GitHub118>
[jruby] mkristian pushed 1 new commit to master: http://git.io/vGXEn
<GitHub118>
jruby/master 750becc Christian Meier: [build] added some line to wiki [skip ci]
<mkristian>
enebo, jruby-9.0.1.0 soon ?
<GitHub198>
[jruby] mkristian force-pushed master from 750becc to f2ba397: http://git.io/Xb_O8w
<GitHub198>
jruby/master f2ba397 Christian Meier: [build] added a link to wiki to some build info [skip ci]
<enebo>
mkristian: today!
vyorkin has joined #jruby
<enebo>
mkristian: I made the release bits already but I am running tests and making sure things nistall and package
<mkristian>
:)
<enebo>
mkristian: but it closed on sonatype and the only thing I noticed was what I mentioned above…so looking great so far
<mkristian>
can I make an "extra" maven profile (I know we have already too many) to run more tests like building a war with a rails app and start it, etc. nothing to run on every commit on travis ?
<mkristian>
but before a release or on some daily, weekly base
<enebo>
mkristian: yeah I think that is a good idea and I wish we did it with more libraries/apps
<enebo>
mkristian: it could also be a rake target to
<enebo>
mkristian: unless there is a reason for it to be maven-specific
<enebo>
mkristian: another potential idea is we could make a separate jruby project which ran soem representative app
<enebo>
mkristian: but I am unsure how we could periodically run it via travis (we can do it via cloudbees too)
<nirvdrum>
It'd be kinda nice if there were a way to see results for all projects on Travis running with jruby-head.
<enebo>
someone has an app which keeps track of gems and how we and other impls do but it is not neccesarily head but whatever they have configured
<mkristian>
enebo, the main reason I use maven is that I can easily spawn a "new project" without reference or interference from the actual maven build. I found it really hard in that past to use cucumber or something to setup a rails project from scratch and run it.
<enebo>
thank bjfish2 when he comes back on :)
subbu has quit [Ping timeout: 244 seconds]
<mkristian>
200 pages takes some time to look through ;)
<GitHub9>
[jruby] tdaitx opened issue #3303: strftime_spec.rb SPEC test fails in "returns the timezone with %Z" http://git.io/vGXVh
<enebo>
mkristian: perhaps bjfish2 can share how he is extracting this info and we can make a just one impl failing results page(s)
<mkristian>
and reduce it to jruby
<enebo>
mkristian: yeah
<enebo>
looking at ‘mail’ it looks like we probably only have two issues
<enebo>
knocking some of these down here and there until we get an empty list
<enebo>
It is pretty tiring to have to reduce these things without just getting a reduced report
rcvalle has joined #jruby
bbrowning is now known as bbrowning_away
<GitHub45>
[jruby] chrisseaton pushed 2 new commits to master: http://git.io/vGXru
<GitHub45>
jruby/master cbacdd5 Chris Seaton: [Truffle] More types of cache option.
<GitHub45>
jruby/master 5a4b3f9 Chris Seaton: [Truffle] Have a single option to set all cache size defaults.
samphippen has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]