"Nicholls, Mark" <[EMAIL PROTECTED]> wrote: > At best I think the error message to be misleading and there is a > significant hole in the type inference mechanism,
I disagree that there is a hole. I'm pretty sure subtyping constraints like in this example should never be used in inference, because it would imply that adding a new ancestor interface to a class might be a breaking change, which should be pretty counter-intuitive when you think about it. In the functional world, subtyping and polymorphism are known not to play particularly well with type inference. > at worst it wouldn't > surprise me if there is a bug in there...the error would seem to imply > it thinks it's proved no solution exists, when at least one does. I think the error message is poor. I guess that the type inference is using early sanity-checking to reduce the contenders: when the number of generic type arguments is greater than the uses of the generic type arguments in the method declaration, it's excluding that method from the list of inferable candidates early on, but when it gets to the inference resolver it finds it has no potential methods to work with, so it issues this error message. -- Barry -- http://barrkel.blogspot.com/ =================================== This list is hosted by DevelopMentor® http://www.develop.com View archives and manage your subscription(s) at http://discuss.develop.com
