> Am 01.05.2016 um 23:34 schrieb Xiaodi Wu <[email protected]>: > > The point is pretty much moot. In Erica's draft proposal, keywords are > required in any Swift code that hasn't been compiled in order to indicate > conformance. Under that scheme, it might be useful to be able to express > something like what Erica asked about in this email thread. I understood it > to be a factual question--is it or is it not possible to exclude one > particular conforming type from an extension of a protocol? The answer seems > to be that the it is not possible in current Swift syntax, which is fine. > > I understand that you are saying that it isn't currently necessary to use > such a feature for retroactive modeling and that it might be possible to > write an alternative proposal for keywords like `required` in which it > continues not to be necessary. I don't think there's any disagreement here.
Ok :-) -Thorsten > On Sun, May 1, 2016 at 16:24 Thorsten Seitz <[email protected] > <mailto:[email protected]>> wrote: > >> Am 01.05.2016 um 03:03 schrieb Erica Sadun <[email protected] >> <mailto:[email protected]>>: >> >> >>> On Apr 29, 2016, at 3:18 PM, Thorsten Seitz <[email protected] >>> <mailto:[email protected]>> wrote: >>> >>> >>>> Am 29.04.2016 um 19:51 schrieb Xiaodi Wu <[email protected] >>>> <mailto:[email protected]>>: >>>> >>>> Yes, certainly this works today. The motivation for Erica's question is >>>> that this would not work without modifying the third-party code if >>>> keywords were required to indicate implementation of protocol requirements. >>> >>> If I understood the motivation correctly, the requirements should protect >>> against typos where I planned to *replace* a method but due to a typo >>> instead created a new method and inherited the default (or superclass) >>> implementation. >> >> extension A where !x:B { >> .... >> } >> >> >> then conforming a type to A and B would exclude the type from inheriting the >> extension. > > I still don’t see the need for that. As my code example demonstrated an > extension will not override an existing method in a conforming type and my > other mail demonstrated that it is not necessary to have „required“ keywords > in existing code which is extended. > > Maybe I still haven’t understood the use case you are trying to solve. In > that case I’d appreciate if you would give a complete example that shows what > should happen why. > > -Thorsten >
_______________________________________________ swift-evolution mailing list [email protected] https://lists.swift.org/mailman/listinfo/swift-evolution
