Mark, On 4/22/14, 4:18 PM, Mark Thomas wrote: > On 22/04/2014 21:15, Christopher Schultz wrote: >> Mark, > >> On 4/22/14, 3:58 PM, Mark Thomas wrote: >>> On 22/04/2014 20:53, Christopher Schultz wrote: >>>> Mark/Konstantin, >>> >>>> On 4/18/14, 11:29 AM, kkoli...@apache.org wrote: >>>>> Author: kkolinko Revision: 1588102 Modified property: >>>>> svn:log >>>>> >>>>> Modified: svn:log at Fri Apr 18 15:29:06 2014 >>>>> ------------------------------------------------------------------------------ >>>>> >>>>> >>> >>>>> > --- svn:log (original) >>>>> +++ svn:log Fri Apr 18 15:29:06 2014 @@ -1 +1,3 @@ Fix an >>>>> Eclipse nag +Correct signature of SSL.fipsModeGet(), +as >>>>> native code is implemented as throwing an Exception in this >>>>> method. >>> >>>> Why single-out this method only to have its method signature >>>> changed? >>> >>>> Nearly every native method in SSL.java can throw an exception. >>>> I was just following convention. >>> >>> Because you documented the Exception in the Javadoc without >>> declaring it in the code and that triggered an IDE warning. The >>> 8.0.x code is kept warning free so either the Javadoc had to be >>> removed or the declaration corrected. Correcting the declaration >>> seemed to be the better option since it is possible for an >>> exception to be thrown in this case. > >> Fair enough. Shall we change the other methods in SSL.java as >> well? > > That is worth doing but I'd do that as a separate clean up task. > >> Any particular reason you removed the 'final' from fipsModeState? > > Yes. If the reason is not immediately obvious to you I suggest you > restore it and then try and compile the class.
Before posting, I checked just so I wouldn't embarrass myself. It worked just fine. My guess is that you changed the implementation to throw an exception after removing the 'final' which then makes the 'final' legal again. -chris
signature.asc
Description: OpenPGP digital signature
