discuss.visual-prolog.com

Hello Thomas,

just a small suggestion regarding interface synonyms, maybe this could be made working in some future version:

You can just write:

True, that makes the example's code even shorter. In large use cases however things can be the other way around. I.e. when declarations made in an interface obj{@TypeA, @TypeB, ...} are refered to many times in the code and in each place type parameters are substituted by same lengthy type expressions.

Merry Christmas to you and all

An interface defines an object type, and as such it defines the predicates and properties that such objects have.

For convenience an interface is however also a scope that can declare constants and domains, but these constants and domains belongs to the scope not to the interface type (and they are not accessible through the objects).

So an interface definition is a single textural construction that defines two things, an object type and a static/global scope.

We have no intensions about extending object types to carry more than the predicates and properties of the corresponding objects.

Likewise we have no intensions about extending an synonym type to do more than defining a synonym of a type.

So the definition
Defines a synonym of the type, but it is not the type that contains the myType domain, so you cannot write:
But you can define this synonym type (i.e. because is it a type):
You may also be able to handle things by opening obj{unsigned}.

Thank you for the detailed explanation. Now I see the concept behind scenes.

I suppose the issue is connected to Supports Qualification for Domains. Similarly a supports qualification extends the type of an interface but not its scope.

Martin Meyer wrote: ↑19 Dec 2017 15:09Similarly a supports qualification extends the type of an interface but not its scope.

Exactly.