This is just to let you know that I have deployed our tests on JDK 1.9. Some of them had to be silenced (ignored) waiting for fixes inside RestEasy, but overall the build is passing after forced upgrades to Groovy and some JVM switch tweaks.
Thanks everyone for help. While I still think some of the changes in jigsaw will be painful in practice, I also clearly see the benefits of the new module system. The aesthetics of splitting rt.jar into more manageable pieces is an additional pleasant aspect as well. :) Dawid On Fri, Apr 8, 2016 at 6:25 PM, Mandy Chung <mandy.ch...@oracle.com> wrote: > >> On Apr 8, 2016, at 4:58 AM, Dawid Weiss <dawid.we...@gmail.com> wrote: >> >> Thank you for the magic switch, Mandy! Here's the full output, >> including stack traces: >> >> http://pastebin.com/uUb3aLwq >> >> The one referencing the problematic class is: >> >> PROXY: jdk.proxy2/com.sun.proxy.jdk.proxy2.$Proxy65 defined by >> jdk.internal.loader.ClassLoaders$PlatformClassLoader@523884b2 >> java.annotations.common/javax.ws.rs.ext.Providers module-private >> loader jdk.internal.loader.ClassLoaders$PlatformClassLoader@523884b2 >> > > I see Alan has explained this and the issue is now understood. > > One thing to add is that a proxy class is only generated in dynamic module if > one of the proxy interface is in a named module and not exported. If you see > a proxy class generated in a dynamic module and you run your applications as > it is today, it would likely invoke creating a proxy class with some > interface in JDK modules, like java.annotations.common module in this case. > > If all proxy interfaces are either in unnamed module, or exported from named > modules, the proxy class will be generated in unnamed module and existing > code should work as it is today. > >> RestEasy creates this proxy class in >> InjectorFactoryImpl.createContextProxy, but its only reason to exist >> is (presumably) to keep a cache of proxy classes for the same set of >> interfaces (it later calls ContextParameterInjector with a cached >> proxy class). This doesn't look correct to me as newProxyInstance >> reuses proxy classes internally (although the javadoc here could be >> improved). >> > > A related note to caching of proxy classes: The performance of > Proxy::newProxyInstance has been improved in JDK 8 and Peter has a patch that > brings jdk9 proxy performance even better than JDK 8 [1]. > > Mandy > [1] > http://mail.openjdk.java.net/pipermail/core-libs-dev/2016-April/040015.html >
