Join GitHub today
GitHub is home to over 28 million developers working together to host and review code, manage projects, and build software together.Sign up
Support the JDK's ThreadLocalRandom #1052
referenced this pull request
Feb 1, 2017
This looks great. The only thing I was thinking about was trying to avoid the branch per call by making the JdkDelegate an instance class, creating a new delegate class for our backfill ThreadLocalRandom, and an interface that defines just the
That's fantastic! Thank you! You've tested this in JDK 6? I'm curious because just by looking at it, I'm not sure whether NoClassDefFound error would be thrown during the instantiation of JdkProvider, or if it would be thrown while calling
Yes, I've run the suite in JDK6. If I add a
I suppose the exception is thrown when the
Feb 7, 2017
added a commit
this pull request
Feb 15, 2017
Ryan, as you wondered : "I'm not sure whether NoClassDefFound error would be thrown during the instantiation of JdkProvider, or if it would be thrown while calling JdkProvider.current", in fact, testing 3.1.4, on Weblogic 10.3 and Oracle JDK 6, the exception is thrown only when JdkProvider.current is called, not when JdkProvider constructor is called, so 3.1.4 is KO. Curiously, on Tomcat 7 and the same Oracle JDK 6, the exception is thrown when the JdkProvider constructor is called.
I made a pull request for my tested workaround (#1129).