(j3.2006) finalizer infelicity

Bill Long longb
Wed Nov 21 15:59:36 EST 2007

Aleksandar Donev wrote:
> On Wednesday 21 November 2007 11:11, Michael Ingrassia wrote:
>> do our semantics really require us to call a
>> finalizer ([59:5] "it is called") when that
>> finalizer could not have been called explicitly ?  Do they forbid us
>> to bind such finalizers?
> I agree with Michael that this seems ill-defined. Maybe this again points out 
> to the conclusion that Jim's objections also raised: The dummies should be 
> required to be assumed-shape arrays. Or better, the whole non-elemental 

Assumed-size would be a better choice in this case.  Seems inefficient 
to create a dope vector just to call a final routine.

But this seems just a special case of the more general problem of 
calling a subroutine with actual arguments that have a size that varies 
from one call to the next.  If that's what you anticipate happening, it 
is reasonable to declare the dummy argument as assumed-size.  If you 
declare the dummy as explicit-shape, expect a "not-so-good-programmer"  
hat and a chair in the corner facing the wall.


> finalizer business should be deleted. I am not even kidding :-\
> Aleks
> _______________________________________________
> J3 mailing list
> J3 at j3-fortran.org
> http://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., 1340 Mendota Heights Rd., Mendota Heights, MN, 55120


More information about the J3 mailing list