(j3.2006) (SC22WG5.3820) Response on the TR29113 draft N1761

Van Snyder Van.Snyder
Thu Dec 11 13:41:19 EST 2008


On Thu, 2008-12-11 at 03:20 -0800, Robert Corbett wrote:
> I agree that adding a function RANK makes sense, even if it can
> be implemented another way.  Most people would not immediately
> recognized what SIZE(SHAPE(X)) does.  Most compilers would if it
> became a common idiom, but Fortran code should be understandable
> by humans as well as machines.

given
  dimension X(:,:,:), Y(..)
size(shape(x)) is not an initialization expression, but rank(x) would
be.  size(lbound(x)) ought to be an initialization expression, but some
vendors have gotten that wrong.  It's hard to imagine not recognizing
that rank(x) is an initialization expression, even though rank(y) would
not be an initialization expression.

-- 
Van Snyder                    |  What fraction of Americans believe 
Van.Snyder at jpl.nasa.gov       |  Wrestling is real and NASA is fake?
Any alleged opinions are my own and have not been approved or
disapproved by JPL, CalTech, NASA, the President, or anybody else.




More information about the J3 mailing list