Posted by gouessej on Jan 08, 2018; 8:35pm
URL: https://forum.jogamp.org/Jogl-Jogamp-on-Java-9-tp4038012p4038493.html

I know that there is a problem with Java 1.9 under OS X, I don't deny it and the modified version of Java3D works around a bug in JOGL caused by changes in Java itself, it doesn't fix the real bug. However, you can still make a pull request against Phil's repository as I'd like to know his opinion. Predrag Bokšić's changes affect Java3D as far as I know, not JOGL. Some questionable workarounds are better than no solution at all but we need a pull request, we need to see exactly why it works even though it doesn't release the context properly and by the way, I still don't understand exactly why you needed to add a DLL for AWT somewhere under Windows to make it work. I cannot go any further and help you both if you don't share your findings in the most suitable way, not by sending a compressed archive without indicating clearly what changes in the source code have been done. If making a pull request is too complicated for you, please post at least a diff file. Ignoring the exceptions isn't a satisfying solution. JOGL and Java3D are under open source licences, we don't hide any part of the source code, either you work in an open and transparent manner like us, or you solve your problems without the help of the JogAmp community, the choice is yours. I'm sorry, maybe I'm a bit touchy but when I read "our new version of jogl" and when I still get no clear explanation of what changes fixed your problems, I'm really upset. I'm a volunteer, I have used JOGL since 2006, I answered your questions even during my holidays. Making a new version of JOGL ready for a release requires a lot more work than just fixing a few things privately. Modifying the source code just to catch some exceptions that shouldn't even be thrown isn't a fix even though it allows you not to use my clean workaround already working without Java Webstart. Of course, I would prefer a solution that works with and without Java Webstart but a dirty solution not fixing the real bugs isn't acceptable on the long term, there is a huge difference between tinkering some code and providing testable strong fixes.