> Neil Carlson asked:
> >Two questions about the IBITS intrinsic (F95):
> >
> >1) Is it required that the POS and LEN arguments be of the same
> > integer kind as the main integer argument I? The books I have
>
> No. The standard does not require this, it allows them to be of any kind.
>
> >2) Is a compiler required to catch an incorrect call to an intrinsic
> > when the actual argument types/kinds don't match what is allowed?
>
> No. These requirements are not constraints.
>
But, bullet (7) in section 1.5 on conformance says the processor must
be able to
"detect and report the use ... of intrinsic procedures whose names
are not defined in Section 13"
This obviously requires detection of extensions like a "system"
or "ranf" intrinsic routine.
But, doesn't it also require detection of overloaded intrinsics?
If a processor chooses to overload the MAX routine to accept
complex arguments it has to detect that, doesn't it? The phrase
"whose names" can't really apply to overloaded names and allow
for willy-nilly undetected extensions. Otherwise
a standard conforming compiler could allow
call max ('echo "hello world"')
and claim that the name "max" was defined in chapter 13.
Dick Hendrickson
> Cheers,
> --
> ...........................Malcolm Cohen, NAG Ltd., Oxford, U.K.
> ([log in to unmask])
>
> ________________________________________________________________________
> This e-mail has been scanned for all viruses by Star. The
> service is powered by MessageLabs. For more information on a proactive
> anti-virus service working around the clock, around the globe, visit:
> http://www.star.net.uk
> ________________________________________________________________________
|