(j3.2006) (SC22WG5.3774) Response on the TR29113 draft N1761

[Reinhold Bader]

Aleksandar Donev schrieb:
> Hi,
>
> I am still processing Reinhold's proposals, but this I have thought 
> about already:
>
>   
>> Issue 2 - polymorphism of assumed-type entity:
>> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>> Since no changes to the definition of C_LOC() have been introduced,
>> and this function is used to cast an object of TYPE(*) to a usable
>> type, the text beginning in line 92 of N1761 should be replaced by
>>     
> I am not sure I understand. If the dummy is TYPE(*), and the actual is 
> polymorphic, then the assumed type of the dummy becomes the dynamic 
> type of the actual. It is no longer polymorphic once you are inside the 
> procedure. 
That is how the TR in its present form wants it to be, and I intuitively 
followed the otherwise valid
rule that a CLASS(x) actual cannot ever be matched do a TYPE(x) dummy in 
formulating my
issue.
And the implications of allowing polymorphism here still make me squirm 
because of
unintended side effects like having a TYPE(*), VOLATILE dummy and 
coaxing the external
mechanism into changing the type  of this object from outside :-)


> So what is the problem with doing C_LOC on it?
>
>   
>> "In the association of actual and dummy arguments, an assumed-type
>>  dummy argument is type and kind compatible with a non-polymorphic
>>  actual data argument of any type."
>>     
> I think there be some restriction that the type should be interoperable 
> if the procedure is BIND(C). But I am not sure it can be made to work. 
> It definitely requires more work.
>
> Best,
> Aleks
>
> _______________________________________________
> J3 mailing list
> J3 at j3-fortran.org
> http://j3-fortran.org/mailman/listinfo/j3
>