26 Aug
2005
26 Aug
'05
6 p.m.
On Sat, 2005-08-27 at 01:17 +0400, en.karpachov@ospaz.ru wrote:
If the adapter is a function, the function may not implement the interface. The actual expected behavior is that the result of *calling* the adapter *provides* the interface. An adapter class may well implement the interface in order for this to happen, but it's not a necessary condition.
That's true, and I don't see how does it differ from my pseudocode example.
Your pseudo-code does indeed match what I said, and would be useful to have. I have a bug open for "mode of running that is slower but does a lot more runtime checking" and this should probably be one of the features.