Title: Pointer function references as actual arguments

Submitted by: Aleksandar Donev

Status: For Consideration

References: Interp 000074

Basic Functionality:
Allow references to functions returning an associated pointer to appear as actual 
arguments associated with INTENT(OUT) dummies. This is parallel to the way we treat 
regular pointers.

Example:

FUNCTION storage(key) RESULT(loc)
    INTEGER, INTENT(IN) :: key
    REAL, POINTER :: loc
    loc =>...
END FUNCTION

! Make the following legal
CALL RANDOM_NUMBER(storage(10))

Rationale:
The proposed change makes the language more symmetrical. It arose while looking at an 
interpretation request, number 000074, titled "TARGET dummy arguments and POINTER 
expressions", which decided that in Fortran 2003 function results are not definable 
even when they are pointers associated with a TARGET. However, the functionality of 
allowing code like the one in the example above is needed in the language, and also, 
there is inconsistency with the way pointer variables are treated.

Estimated Impact:
As interp 000074 showed, the standard is already a little murky on whether function 
results are definable, so I think the changes to the standard will be small. The cost 
to implementations will be small.

Detailed Specification:
The text which describes argument association has problems in that it often confuses 
the actual argument with the "thing" the dummy is associated with, and once that text 
is clarified I believe the above example will become legal by miracle.

History:

Comments:
Malcolm Cohen, Nihon NAG, Tokyo:
(1) I have been sent two related Fortran 2003+ 
suggestions, which I would like the U.K. panel to consider.  They are both from 
Aleksandar Donev.

My opinion is that we should definitely put the first one (pointer_args) forward, as 
the current situation is rather peculiar (the dummy argument is associated with the 
variable pointed to by a pointer function reference actual argument, but is not 
definable even though it is associated with a variable).

P.S. If we do decide to put them forward I think the text and examples need a little 
work. I'm willing to help with that.

John Reid, JKR Associates, Oxford:
I will advocate that we go for both of these. It seems to me to be illogical to go for 
one and not the other.

Malcolm Cohen, Nihon NAG, Tokyo:
(2) I am in favour of "pointer args".