I'm not in love with error prone. if it's blocking us to move to a newer JDK i say it goes away (at least for now).. if at a later point it's again compatible we put it back on.
On Fri, May 29, 2020 at 12:21 PM Emmanuel Hugonnet <ehugo...@redhat.com> wrote: > > Hello, > > Alas yes as otherwise you need to configure the compilation with error-prone > and jboss logging annotation processor., and future annotation > processors. > > It was the less intrusive way from my point of view. > > Emmanuel > > Le 29/05/2020 à 18:10, Robbie Gemmell a écrit : > > Is it really necessary to add error-prone definition to almost every > > module? Presumably theres some module it isnt defined in. No other > > workarounds for that? > > > > I also wonder about this bit: > > https://github.com/apache/activemq-artemis/compare/master...clebertsuconic:java_11#diff-600376dffeb79835ede4a0b285078036R1437 > > > > It refers to a javac.version property that doesnt seem to exist? Other > > exmaples of error-prone config using that seem to set it explicitly. > > > > Robbie > > > > On Fri, 29 May 2020 at 16:56, Clebert Suconic <clebert.suco...@gmail.com> > > wrote: > >> @Emmanuel: I rebased your branch here: > >> https://github.com/clebertsuconic/activemq-artemis/tree/java_11 > >> > >> Perhaps you may want to take a look. > >> > >> On Fri, May 29, 2020 at 11:54 AM Clebert Suconic > >> <clebert.suco...@gmail.com> wrote: > >>> so, if we switched to JDK 11 on the CI, those would still work with > >>> the trick you used? > >>> > >>> On Fri, May 29, 2020 at 11:39 AM Emmanuel Hugonnet <ehugo...@redhat.com> > >>> wrote: > >>>> I changed the way the classloaders where working with JDK 11 using the > >>>> plateform classloader as the parent instead of null so that I could > >>>> access the required modules. > >>>> > >>>> Emmanuel > >>>> > >>>> Le 29/05/2020 à 17:37, Clebert Suconic a écrit : > >>>>> We should come back into this... > >>>>> > >>>>> > >>>>> How did you fix the compatibility tests with JDK 11? do you need JDK 8 > >>>>> to run the compatibility tests? > >>>>> > >>>>> On Fri, May 8, 2020 at 3:19 PM Emmanuel Hugonnet <ehugo...@redhat.com> > >>>>> wrote: > >>>>>> No relation to WildFly, just for the 'fun' of it and because switching > >>>>>> my JAVA_HOME and PATH each time I wxanted to build was itching me too > >>>>>> much ;) > >>>>>> > >>>>>> It can wait and get some baking :) > >>>>>> > >>>>>> Emmanuel > >>>>>> > >>>>>> > >>>>>> Le 08/05/2020 à 19:47, Clebert Suconic a écrit : > >>>>>>> Oh wow.. that's awesome. > >>>>>>> > >>>>>>> > >>>>>>> Can we do the migration after the 2.13.0 release? I have pretty much > >>>>>>> everything ready to go (besides a few changes we have to make next > >>>>>>> week). doing this migration now would probably delay the release. > >>>>>>> > >>>>>>> > >>>>>>> Or you have some requirement for Wildfly that you must do this now? if > >>>>>>> you do we can eventually delay it.. but I would prefer doing it for > >>>>>>> later. > >>>>>>> > >>>>>>> On Fri, May 8, 2020 at 6:24 AM Emmanuel Hugonnet > >>>>>>> <ehugo...@redhat.com> wrote: > >>>>>>>> Hello, > >>>>>>>> > >>>>>>>> When i need to work on Apache ActiveMQ Artemis I need to switch my > >>>>>>>> local environement to use OpenJDK 8 instead of the default OpenJDK 11 > >>>>>>>> which I use to develop. > >>>>>>>> > >>>>>>>> I have started a branch > >>>>>>>> https://github.com/ehsavoie/activemq-artemis/tree/java_11 which > >>>>>>>> builds on OpenJDK 8 and 11 with the fast-tests profile. > >>>>>>>> > >>>>>>>> mvn clean install -Pfast-tests is passing for me locally :) > >>>>>>>> > >>>>>>>> I don't think that the CI or the release should use Java 11 but I > >>>>>>>> hope this will smoothen the time when the migration occurs. > >>>>>>>> > >>>>>>>> Also I couldn't use the --release 8 flag for OpenJDK 11 because of > >>>>>>>> the use of Unsafe, if someone has a better alternative I'm all hears. > >>>>>>>> > >>>>>>>> Do you want me to sumbit a PR for this ? Does this make sense to the > >>>>>>>> community ? > >>>>>>>> > >>>>>>>> Emmanuel > >>>>>>>> > >>>>>>>> > >>>>>>>> > >>> > >>> -- > >>> Clebert Suconic > >> > >> > >> -- > >> Clebert Suconic > -- Clebert Suconic
