Re: ANNOUNCE: SPARK toolset 6.1 now available

[Simon Wright <simon@pushface.org>]

"Randy Brukardt" <randy@rrsoftware.com> writes:

> Rod Chapman wrote in message ...
> >dennison@telepath.com (Ted Dennison) wrote in message news:
> >> By "dispatching" do you mean runtime polymorphisim as opposed to
> >> compile-time polymorphisim?
> >
> >Correct. Run-time polymorhpism is not permitted.  Calling an
> >inherited (or overridden) basic operation with a specific tagged
> >subtype is OK.
> 
> 
> Do you do that by prohibiting T'Class, or do you allow it for uses
> like hetrogeneous lists? If the latter, it seems as if there
> wouldn't be any problem supporting run-time polymorphism. And if the
> former is true, its hard to imagine what uses type extension can be
> put to...

Well, I suppose you can use implementation inheritance and save
yourself repeating the code for inherited operations.

I would be _very_ surprised if SPARKada allowed T'Class
(uncertainty being something you don't want in safety-related
software).