Re: Problem with LOC intrinsic for assumed size arrays

[Mark Hesselink <mhesseli at caltech dot edu>]

On Jan 24, 2006, at 2:37 PM, Steve Kargl wrote:

> On Tue, Jan 24, 2006 at 02:24:10PM -0800, Mark Hesselink wrote:
> > While trying to compile FMS (http://www.gfdl.noaa.gov/fms/) I came
> > across a problem with the LOC intrinsic for assumed size arrays. The
> > following code produces the error below:
> >
> > PROGRAM loc_1
> > CONTAINS
> >    SUBROUTINE f (x)
> >       INTEGER, DIMENSION(*)   :: x
> >       INTEGER                 :: address
> >       address=LOC(*)
> >    END SUBROUTINE f
> > END PROGRAM loc_1
> >
> > [aardschokker:fms/trunk/obj] mark% /usr/local/bin/gfortran -c 
> > loc_1.f90
> >  In file loc_1.f90:6
> >
> >       address=LOC(*)
> >                  1
> > Error: Syntax error in argument list at (1)
>
> I don't use LOC, but this won't stop me from asking a question.
> Do you really mean "address=LOC(*)"?  LOC() returns the address
> of a variable and * isn't a variable.  Is this suppose to be
> "address=LOC(x)"?

I am so sorry. Yes, of course it is supposed to be LOC(x).  Changing 
the line ``address=LOC(*)'' to ``address=LOC(x)'' produces the 
following result:

[aardschokker:fms/trunk/obj] mark% /usr/local/bin/gfortran -c loc_1.f90
 In file loc_1.f90:6

      address=LOC(x)
                 1
Error: The upper bound in the last dimension must appear in the 
reference to the assumed size array 'x' at (1).

Thanks,

Mark

P.S. I apologize for the noise. I should have paid more attention 
before sending the email.