(j3.2006) WG23 Fortran Annex Draft

[Reinhold Bader]

Hello,

Dan Nagle schrieb:
> Hi,
> 
> I have put a new version of Fortran_Annex.txt in Tutorials
> addressing Reinhold's comments.

The wording is not quite right, though. C_F_POINTER does not take
a function argument, but converts a C_PTR object to a pointer entity;
in a sense it is the inverse of C_LOC.

There are analogous procedures C_FUNLOC and C_F_PROCPOINTER
for handling interoperable functions. It might be more appropriate to
treat these in 3.41.

> 
> On Jun 9, 2009, at 3:36 AM, Reinhold Bader wrote:
> 
>> Hello Dan,
>>
>>   In section 3.12 Bit Representations [STR] you mention TRANSFER and  
>> equivalence,
>>   making a recommendation against using TRANSFER.
>>
>>   However, the C_LOC / C_F_POINTER combination as well as unlimited  
>> polymorphic
>>   entities having targets of non-extensible derived type appear to  
>> me to have the same
>>   potential problems as TRANSFER. Their use should therefore also be  
>> recommended against
>>   (unless type inconsistencies are detectable).
>>
>> Regards
>> Reinhold
>>
>>
>>
>> _______________________________________________
>> J3 mailing list
>> J3 at j3-fortran.org
>> http://j3-fortran.org/mailman/listinfo/j3
>