>
> First, the phraseology "passing allocatable arrays" is vague enough
> to be confusing. It is perfectly fine to "pass" an allocatable
> array as an actual argument; has been since allocatables were
> introduced in f90. Nothing unusual about it at all. What gets
> passed is just the allocated array. The part that needs f2003 is
> just having an allocatable dummy argument, which you need if you
> want to do allocation or deallocation inside of the subroutine.
Yes sorry what I meant was having it allocatable as a dummy otherwise
wrt to inside the subroutine its just like a pre-f90 array (including
lack of bounds info ... which this extension solves). I've checked
and the compilers I'm currently using (g95, Intel V9 and Pathscale)
all implement the allocatable extensions so if some kind soul could
fill me on wrt SGI that would be great because this eliminates most
of the need I have for pointers and makes the code simpler re
deallocation and bounds.
Cheers
Garry Willgoose
====================================================================
NOTE: New Contact Details
====================================================================
Prof Garry Willgoose,
Australian Professorial Fellow in Environmental Engineering,
School of Engineering, The University of Newcastle,
Callaghan, 2308
Australia.
Phone: (International) +61 2 4921 6050
FAX: (International) +61 2 4921 6991
Secretary: (International) +61 2 4921 6042
Telluric Research, 100 Barton Street,
Scone, 2337
Australia.
ABN 13 132 312 352
Phone/FAX: (International) +61 2 6545 9574
WWW : www.telluricresearch.com
email: [log in to unmask] (all other addresses forward
to here)
====================================================================
"Do not go where the path may lead, go instead where there is no path
and leave a trail"
Ralph Waldo Emerson
====================================================================
|