@hot_girl_spring@vlad_mihalcea Yep that was me, but quite a while ago and I've never used the acceleration feature. I'd suggest starting with the container docs that the DB2 testcontainer is based on:
https://t.co/iBAKLzrxek
And perhaps try .withEnv("BLU", "true")?
Otherwise, I'd ask on StackOverflow. Hth!
@maxandersen If you preface it with -XX:+IgnoreUnrecognizedVMOptions then it will make the JVM silently ignore --add-opens on JDK 8. That -XX option should work for every major distro of JDK 8.
@nottycode Wow I can't believe you had that horrible experience going for B9. When you got B10 I was like "yea, about time!".
This is why whenever I ask for a reference I always say "would you be willing to give me a _positive_ reference for X?"
@joechacko@OpenLibertyIO@QuarkusIO Happy new year Joe! It was always a pleasure working with you on stuff.
I still remember watching you in awe get a super long sed/awk command right on your first try from memory -- a true legend!
Today was my last day at IBM. I genuinely enjoyed working on @OpenLibertyIO and @QuarkusIO and learned a ton from the brilliant engineers working on those projects!
@DanHeidinga I believe you can just put it together using basedir ($HOME/.m2/repository/ or $M2_HOME) plus expanding the GAV into a dir+file. So https://t.co/xyc3oDUWkY:bar:1.0 goes to $HOME/.m2/repository/org/foo/bar/1.0/bar-1.0.jar
@gunnarmorling@mariofusco@mreinhold@aalmiray I agree with Mario. OSGi gave us reliable configuration and eliminated NCDFE's many years ago. Unfortunately OSGi could not be used to modularize the JDK itself, but it was doing a fine job for the lib/application space.
@javahippie Yea... reactive is difficult to debug right now, but with software maturity and adoption comes serviceability and tooling. Will just take some time.
@vertx_project@rodolfodpk@lukaseder@r2dbc@mauriciojr The DB2 client is in a better state than WIP afaik.
There is a relatively small feature set for now (same for other Vertx drivers) but more can always be added later based on user demand
@lukaseder@rodolfodpk@vertx_project@r2dbc Yep they are fully reactive!
I wrote the DB2 one ~9 months ago from scratch. The IO layer is based on Vertx core which is based on Netty for the transport layer. (Same for all other Vertx drivers too)
@DaliaShea Did you make sure that the charging port is cleaned out? It can build up an impressive amount of pocket lint over time. Otherwise, 5 years on a 6S is a pretty decent run IMO!
@BrianGoetz @rafaelcodes I think it could be treated similarly to how @Override is treaded by javac. If you try to put @Override on something that doesn't match the super/ifc, then compiler blows up. Likewise if you put @NamedParameters on something that doesn't match super/ifc, then compiler blows up
@BrianGoetz @rafaelcodes Perhaps let the impl declare if it's sensitive to parameter names with a source-retention annotation like @NamedParameters on the method/class level? That way it doesn't have to be a global switch for "every parameter name is a potential breaking change"
@hakdoganhoca Yes a new release of MicroShed will come out shortly after Testcontainers 1.15.0 is released. Currently the fix you are seeking isn't in an official release of TC though (just a release candidate). In the the short term, you can override the TC dependency in your project.
All the official upload plugin can do is upload zips. So I'm thinking we'd have to upload HTML test results to a separate static file server for viewing.
Kicking the tires on @github Actions as CI for @OpenLibertyIO and it looks really promising. It's able to handle our thousands of integration tests with relative ease.
But one obvious gap is the lack of easily viewable test results as HTML. Anyone solved this yet?