Add Support for New Naming Strategy Based on First Available Declared Interface about byte-buddy

[raphw]

Thanks for your suggestion. I thought about this before, there is however one caveat with this strategy. Interface methods are never package-private, super class methods might however be (often implicitly, as non-public classes implement their public methods as package-private ones). At the same time, the interface type itself might be package-private where your suggested naming strategy makes more sense.

You are also right about the fact that package-private would not be an issue when subclassing Object. I thing it makes sense to apply this strategy when a user specifies "subclassing" an interface. This should however not consider the first interface but the specified one as one might have defined the implementation of an interface on the ByteBuddy configuration.

I'll give it some thought and implement the changes for version 0.5. Thanks for the input.

[raphw]

This is now implemented in the most recent 0.5 snapshot and is included in the next iteration.

[raphw]

Version 0.5 is released.