| Gary L. Scott 2005-02-24, 3:58 am |
| James Giles wrote:
> gary.l.scott@lmco.com wrote:
>
>
>
> There's a new data type being proposed right now. It has the
> oxymoronic name TYPELESS, but it implements the data type
> already in the standard described as the BIT model, except that
> it doesn't attribute any special properties to the leading bit of
> a value. Change the name of that type to BINARY and get
> bit-resolution KINDs for it and you've got what you need.
> A BINARY array of KIND-1 elements would be an array of
> BITs. You could index them with ordinary subscripting
> syntax, select sequences (array(5:9)) and so on.
>
> (There are a small number of other problems with the TYPELESS
> proposal as written besides just the name. For one thing, it is
> big-endian throughout, while Fortran is otherwise little-endian.
> So, it needs work. The form it presently has is identical to a
> feature already in a particular vendor's compiler. So, you can
> guess who proposed it to the committee.)
>
I had conversations via email with the author last year. While he was
gracious, appeared to welcomed my enquiry, and was willing to explain
his proposal to me, he did not seem interested in discussing changes,
especially significant ones that my suggestions would entail, indicating
it was "exactly what my customers want". (nothing wrong with that, just
not what I want)
--
Gary Scott
mailto:garyscott@ev1.net
Fortran Library: http://www.fortranlib.com
Support the Original G95 Project: http://www.g95.org
-OR-
Support the GNU GFortran Project: http://gcc.gnu.org/fortran/index.html
Why are there two? God only knows.
Democracy is two wolves and a sheep, voting on what to eat for dinner...
Liberty is a well armed sheep contesting the vote. - Thomas Jefferson
|