RE: [sv-ac] Mantis #1647: Updated proposal

> Regarding the use of type(expression) in place of the type-query
> functions: As far as I could see in P1800-2008 LRM, the type
comparison
> is strict. In the case of the type-query functions, $isintegral
> encompasses all specific types of integral kind. Similarly $isproperty
> accepts properties, sequences and integral types. The definition of
> type() does not seem to allow this.
[DR>] 
The definition of type() is defined in 6.9.1 and can updated as needed.
For example 'byte' is equivalent to 'bit signed [0:7]'. Maybe my
proposal won't work, but I would rather see a general mechanism for
checkin IS-A relationships instead of individual system tasks for each
type.

We have the same issue with inheritance of class extensions. For example

class Animal;
endclass
class Mammal tends Animal;
endclass
class Dog extenda Mammal;

Animal A;
Mammal M;
Dog D;

I would like to be able check that the type of D is-a Mammal

Brad also has interests with checking for is-a 'signed' or 'packed'
type.

Dave


-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.