Hello, Since the topic is interesting to me I took a look. Have some nits or points to discuss: The testProvider() function seems to be missnamed, it actually checks the default Implementation (for the right provider). Maybe testDefaultProvider()?
i would also remove the elses, especially as they print different Passed messages. Just one print line after the guards. BTW, i think it is very common for java code to request SHA1Prng unconditionally (with and without provider=sun) i would add that to all OS as a test. I also wonder if this would be the right place to test the strong secure random getter as well? what about testing the sorting logic (based on the egd url property) I think this is still implemented, even with the different seeding modes for Native. Gruss Bernd -- http://bernd.eckenfels.net ----- Ursprüngliche Nachricht ----- Von: "raghu k.nair" <[email protected]> Gesendet: 27.08.2014 11:00 An: "Bradford Wetmore" <[email protected]> Cc: "[email protected]" <[email protected]> Betreff: Please review CR 8048356 Secure Random provider tests Hi Brad, Could you please help in reviewing the following test. webrev: http://cr.openjdk.java.net/~tyan/raghu/8048356/webrev01/ Bug : JDK-8048356 Thanks, Raghu Nair
