Feedback from Hacking Java 9 at the Chicago Java Users Group

Martijn Verburg martijnverburg at
Tue Jun 28 17:32:46 UTC 2016

Hi Freddy,

OK, I'll make an attempt to pick out which report should go where.  Are you
happy to then contact the original reporter and then have them mail the
correct list directly?  It will help increase their knowledge/participation
of OpenJDK and allow the OpenJDK folks to talk directly to the bug
submitted to get extra info as required.

I'm CCing in quality-discuss as well so Rory and his team are aware.

Right, my answers inline below.


On 27 June 2016 at 17:44, Freddy Guime <freddy.guime at> wrote:

> not yet, we'd like some help identifying the right lists : )
> Thanks!
> On 27 June 2016 at 14:31, Freddy Guime <freddy.guime at> wrote:
>>> Hello everyone! We had a Hacking JDK 9 session a little while ago, but
>>> finally collected and put all the comments from our members on feedback. It
>>> was great having Heather and Charlie move things along and work with the
>>> Chicago Java Community to try and play with JDK9.
>>> Let me know if you want me to go back and get more details on any of
>>> these.
>>> Thanks!
>>> ————————
>>> Jan Fetyko:
>>> Trying out jdk9 build 9-ea+112: I can compile and run. But jvisualvm
>>> crashes with exception : Exception in thread "main"
>>> java.lang.SecurityException: setContextClassLoader at
>>> jdk.internal.misc.InnocuousThread.setContextClassLoader(java.base at 9-ea
>>> /
Project page for more info - and
the mailing list: users(at)

> Jonathan Ross:
>>> Ran into a couple of issues with playing with Java 9 on one of my
>>> projects today:
>>> * A regression in type inference (easily fixed by simplifying the
>>> expressions/spelling out the type parameters)
Will need more details on what the regression was and the workaround is but
please report to: jdk9-dev(at)

> * A peculiar change involving @Nullable @Assisted injects in guava (fixed
>>> by using Optional<...> instead)
Will need more details on what the bug was and the exact workaround is but
please report to: jdk9-dev(at) - it may need reporting to Guava
as well but let's see.

> * Using IntelliJ 2016.1, I could compile and run everything (to 1.8
>>> target) fine using the 111 build, but when I switched to 112 things broke:
>>> Error:IntelliLang Pattern Validator: sun/misc/Resource
I *think* this has been fixed in the latest build, but JIC please report to
jigsaw-dev(at) and jdk9dev(at)

> * When playing with jshell, I found that java.util.ServiceLoader was not
>>> working as expected.
How so?  Please report the details to kulla-dev(at)

> Mary G:
>>> Question about the Build 112 for MacOS/X, there's no zip or tar.gz
>>> distribution, and as a result, when i install the dmg, the java9-ea version
>>> has overwritten my previous java8 installation. Would it be possible to
>>> have the zip'ped distribution for Mac? Thanks!
I think that's actually one for Rory's team to feedback internally?

> John Krupka:
>>> Tried to start groovyConsole and it crashed. We also tried installing
>>> Leiningen for Clojure and that also exploded with missing sun.misc.Launcher.
Please report to jigsaw-dev(at) and jdk9dev(at) - I
suspect however that Leiningen and perhaps Groovy need to fix their launch
scripts to access the private sun.misc.Launcher

> Matt Sicker:
>>> Got a weird issue with locales
>>> assertEquals("12,00 €", new MessageFormat("{0,number,currency}",
>>> Locale.FRANCE).format(new Object[] {12}));
>>> Same problem with Locale.GERMANY, Locale.ITALY, etc.
Probably needs a bit more detail but please report to i18n-devat)
and and jdk9dev(at)

Thanks again for running this hackday and helping get Java 9 ready for
prime time!


More information about the adoption-discuss mailing list