Good point. That would obviously restrict the choice for naming the methods in such a protocol, but since we are talking about new protocols that would not be an impediment. It might result in non-optimal method names, of course.
-Thorsten > Am 31.03.2016 um 18:37 schrieb Rob Mayoff via swift-evolution > <[email protected]>: > >> On Thu, Mar 31, 2016 at 10:56 AM, Thorsten Seitz via swift-evolution >> <[email protected]> wrote: > >> >> protocol UIGestureRecognizerDelegate { >> var gestureRecognizerShouldBegin: ((gestureRecognizer: >> UIGestureRecognizer) -> Bool)? { get } >> } > > UIGestureRecognizerDelegate has five methods that are "named" > gestureRecognizer: > > gestureRecognizer(_:shouldRecognizeSimultaneouslyWithGestureRecognizer:) > gestureRecognizer(_:shouldRequireFailureOfGestureRecognizer:) > gestureRecognizer(_:shouldBeRequiredToFailByGestureRecognizer:) > gestureRecognizer(_:shouldReceiveTouch:) > gestureRecognizer(_:shouldReceivePress:) > You can only have a single property named "gestureRecognizer", so you either > have to come up with other names for these, or change the language to allow > closure-typed properties to have multipart names. > > This problem has been noted before, for example here: > http://article.gmane.org/gmane.comp.lang.swift.evolution/8707/ > > _______________________________________________ > swift-evolution mailing list > [email protected] > https://lists.swift.org/mailman/listinfo/swift-evolution
_______________________________________________ swift-evolution mailing list [email protected] https://lists.swift.org/mailman/listinfo/swift-evolution
