On Fri, 30 Apr 2021 09:46:08 GMT, Hannes Wallnöfer <hann...@openjdk.org> wrote:
> Thanks for the review; I agree we should think about normalizing modifiers. > > FWIW, the JLS describes methods in final classes as merely "behaving as if > they were final" (8.4.3.3.), so at least technically it is still the modifier > that decides. Other constructs are actually declared as implicitly final, > with some of them allowing the modifier (such as constants in interfaces) and > some not (enums). Here I think we usually show modifiers if they are allowed, > although I haven't done a systematic review. I would be concerned that adding (or removing) a redundant `final` keyword could be perceived as a specification change if it's reflected in the API documentation. But I am happy for the normalization issue to be revisited later. ------------- PR: https://git.openjdk.java.net/jdk/pull/3716
