ur5us has quit [Ping timeout: 260 seconds]
ur5us has joined #jruby
nirvdrum has joined #jruby
nirvdrum has quit [Ping timeout: 240 seconds]
_whitelogger has joined #jruby
ur5us has quit [Ping timeout: 260 seconds]
<kares[m]> yeah jossl tests - on my todo list but the build is so convoluted with ruby-maven it will take time to undo the marriage ...
<kares[m]> there's two ways we're running tests, one of which is the simply ruby way, so that might be okay.
<kares[m]> I think the annoyance at the moment is that mvn (w ruby-maven plugins) fails building on Java 11.
ur5us has joined #jruby
drbobbeaty has quit [Ping timeout: 244 seconds]
drbobbeaty has joined #jruby
<headius[m]> So leave it to just the autoload? Can you show me the patch?
<headius[m]> This is a real world issue yeah? I forget which bug
<headius[m]> People don't often update RG unless there's a security issue
<headius[m]> Hopefullu we can have 9.3 out in another month
<kares[m]> it's very rare (and restart usually helps) but yeah I have it from a real world case ...
<headius[m]> I have a patch for 9.2.12 to skip setting up proxy for non-ezported classes also, missing check in MethodGatherer
<kares[m]> on top of the RGs update, like you've said this does not make sense unless users upgrade RGs
<kares[m]> so I believe it would only make sense with updating RGs to latest and making sure next RGs has the same patch
<kares[m]> * on top of the RGs update, like you've said this does not make sense if users upgrade RGs
<kares[m]> so I believe it would only make sense with updating RGs to latest and making sure next RGs has the same patch
<headius[m]> Hmmm I would expect this to not be fixed by the patch
<headius[m]> Seems like two concurrent autoload issue
<kares[m]> interestingly I did not reproduce the issue with the patch, but yeah its weird
<headius[m]> Actually maybe not, it would just require a concurrent access of the constant
<headius[m]> So there's a brief moment in time where the autoload constant disappears
<kares[m]> yy kind of
<kares[m]> I think the finish auto-load not being cleaned up might play a role here
<headius[m]> It may be possible to fix by making something in the auto load logic more atomic, but I need to look at that code again
<kares[m]> do not bother much, will take a look as/if needed
<kares[m]> was first checking what you guys think of the RGs patch
<kares[m]> as I wasn't sure it's a good solution to move forward with ...
<headius[m]> The require doesn't seem necessary I suppose
<kares[m]> I think this issue has existed for the whole 9.2 line
<headius[m]> Could be sold as a startup improvement 😄
<kares[m]> 😇
<kares[m]> well let's see if we get a reply from RGs today as to where they would land this ...
<headius[m]> Yeah ok, so if not in 3.1 you are thinking we patch and assume it will get into an upcoming release anyway
<headius[m]> That seems fine, we'll see what they say
<headius[m]> enebo want some today so maybe not .12 this week... No Friday releases and he takes most of a day to verify
<headius[m]> Wasn't here today
<headius[m]> Bad mobile
<headius[m]> Hmm that classloader bug in openjdk is not moving along, may want to patch that in .12 also
<headius[m]> We're going to need to patch it anyway because it's unlikely we'll see a fix in jdk 8 for many weeks
<kares[m]> sounds like a plan, I'll fire up a RGs 3.1 update PR to see how CI is looking
<headius[m]> Ok
<headius[m]> Can't really patch this in defaults ☚ī¸
Antiarc_ has joined #jruby
Antiarc has quit [Ping timeout: 265 seconds]
ur5us has quit [Ping timeout: 260 seconds]
nirvdrum has joined #jruby
_Caerus has quit [Ping timeout: 260 seconds]
Caerus has joined #jruby
<headius[m]> Good morning!
<headius[m]> hmm I know someone filed an issue about this exports bug
<headius[m]> ahh right related to javafx ObservableMap
<headius[m]> bummer, search history doesn't have it
<headius[m]> i8her8oat: hey there
<headius[m]> I will have a PR that fixes your ObservableMap issue
<headius[m]> apparently Riot search only searches what's in your buffer
nirvdrum has quit [Ping timeout: 264 seconds]
subbu is now known as subbu|lunch
nirvdrum has joined #jruby
sagax has quit [Ping timeout: 256 seconds]
sagax has joined #jruby
subbu|lunch is now known as subbu
annette[m] has joined #jruby
<kalenp[m]> @head
<kalenp[m]> * headius Pushed some kotlin integration tests. The kotlin integration into the build is just rammed in, but hopefully this is a useful start. https://github.com/jruby/jruby/pull/6295
ur5us has joined #jruby
<headius[m]> excellent!
<headius[m]> doing this might be a model for testing Scala better too, since we're only using simulated classes for that right now
nirvdrum has quit [Ping timeout: 256 seconds]
ur5us has quit [Ping timeout: 260 seconds]
nirvdrum has joined #jruby
nirvdrum has quit [Ping timeout: 265 seconds]
nirvdrum has joined #jruby
ur5us has joined #jruby