(j3.2006) Pointer undefinition

[Van Snyder]

On Sat, 2016-10-01 at 19:25 +0000, Bill Long wrote:
> On Sep 30, 2016, at 4:53 PM, Van Snyder <Van.Snyder at jpl.nasa.gov> wrote:
> 
> > 10.2.2.3p2 says that if the pointer target in a pointer assignment
> > statement is allocatable, it shall be allocated.
> 
> So, if a program tries this, it is not conforming. 

Right.  But some processors might leave the pointer disassociated
instead of junk.  It's better for the standard explicitly to say what
happens, instead of just "don't do this."
 
> > 19.5.2.5 is silent concerning the event that the target is allocatable
> > but not allocated.  Should it be in the list of events that cause the
> > association status of pointers to become undefined?  I assume it ought
> > not to be an event that causes the pointer to become disassociated (it's
> > not in that list either).
> 
> I thought that this subclause was intended to be a collection of cases where the standard has stated semantics - i.e. actually says that the pointer association status becomes undefined.  I didn?t think we intended to include all variants of illegal pointer assignments as cases in this list. 
> 
> Cheers,
> Bill
> 
> > 
> > 
> > _______________________________________________
> > J3 mailing list
> > J3 at mailman.j3-fortran.org
> > http://mailman.j3-fortran.org/mailman/listinfo/j3
> 
> Bill Long                                                                       longb at cray.com
> Fortran Technical Support  &                                  voice:  651-605-9024
> Bioinformatics Software Development                     fax:  651-605-9142
> Cray Inc./ Cray Plaza, Suite 210/ 380 Jackson St./ St. Paul, MN 55101
> 
>