In the context of the CTS. I could see how the systems wouldn't align well
at all.
This paper now makes sense to me in the context of a prototyping language.
The difference being that types are not modifiable at runtime, only compile
time.

I haven't looked much at CTS, but to translate this into the C++ type
system. I would leave (Bruno et al) classes out of the resulting object code
entirely. They would be used only as a compile time construct to enforce
interfaces.  Objects, then would correspond to structs and classes fairly
naturally.

Sealed classes would construct their own SealedClass object which would
reference their object and method instances. This would probably require
some runtime overhead for reflection and GADT enforcement.
_______________________________________________
bitc-dev mailing list
[email protected]
http://www.coyotos.org/mailman/listinfo/bitc-dev

Reply via email to