Missing **-operator in GNAT 3.14?

Missing **-operator in GNAT 3.14?

[Anders Råberg]

Hi all,

I'm trying to compile a program that contains the follwing code: 


    --- mypack.ads ---

    1: with Interfaces;
    2:
    3: package Mypack is
    4:   type Short is new Interfaces.Integer_16;
    5: end Mypack;

    ------------------



    --- mytest.ads ---

    1: with Mypack;
    2:
    3: package Mytest is
    4:    subtype My_Short is Mypack.Short;
    5:
    6:    function "*" (A, B : in My_Short) return My_Short renames
Mypack."*";
    7:
    8:    function "**" (A, B : in My_Short) return My_Short renames
Mypack."**";
    9: end Mytest;

    -----------------

    % gcc -c mytest.ads
    mytest.ads:8:63: operator for type "Short" is not declared in
"Mypack"


When I compile the program with GNAT 3.13, 3.12 etc. it works perfectly
fine, but when I use 3.14 I get the error stated above. Has the 
**-operator been removed or have I missed something?


Regards,
	Anders

-- 
----------------------------------------------------------------------
Anders R�berg, M.Sc. C.S&E                
Software Systems Engineer                  
Ericsson Microwave Systems AB, Sweden
----------------------------------------------------------------------

Re: Missing **-operator in GNAT 3.14?

[David C. Hoos, Sr.]

The renaming declarations are unneccessary because those operators
are implicitly declared for all integer types.

That this is so is evidenced by the successful compilation and
execution of the following program (with the renaming declarations
in Mytest commented out):

with Ada.Text_IO;
with Mypack;
with Mytest;
procedure Main is
   use type Mypack.Short;
   Value : Mypack.Short := 3 ** 4;
   My_Value : Mytest.My_Short := 4 ** 5;
begin
   Ada.Text_IO.Put_Line (Value'Img);
   Ada.Text_IO.Put_Line (My_Value'Img);
end Main;

The fact that "*", and "**" are treated differently by GNAT looks
like a "bug" to me, though, and you should probably report it
to report@gnat.com


----- Original Message -----
From: "Anders R�berg" <anders.raberg@NO.SPAM.PLEASE.emw.ericsson.se>
Newsgroups: comp.lang.ada
To: <comp.lang.ada@ada.eu.org>
Sent: April 02, 2002 2:56 AM
Subject: Missing **-operator in GNAT 3.14?


> Hi all,
>
> I'm trying to compile a program that contains the follwing code:
>
>
>     --- mypack.ads ---
>
>     1: with Interfaces;
>     2:
>     3: package Mypack is
>     4:   type Short is new Interfaces.Integer_16;
>     5: end Mypack;
>
>     ------------------
>
>
>
>     --- mytest.ads ---
>
>     1: with Mypack;
>     2:
>     3: package Mytest is
>     4:    subtype My_Short is Mypack.Short;
>     5:
>     6:    function "*" (A, B : in My_Short) return My_Short renames
> Mypack."*";
>     7:
>     8:    function "**" (A, B : in My_Short) return My_Short renames
> Mypack."**";
>     9: end Mytest;
>
>     -----------------
>
>     % gcc -c mytest.ads
>     mytest.ads:8:63: operator for type "Short" is not declared in
> "Mypack"
>
>
> When I compile the program with GNAT 3.13, 3.12 etc. it works perfectly
> fine, but when I use 3.14 I get the error stated above. Has the
> **-operator been removed or have I missed something?
>
>
> Regards,
> Anders
>
> --
> ----------------------------------------------------------------------
> Anders R�berg, M.Sc. C.S&E
> Software Systems Engineer
> Ericsson Microwave Systems AB, Sweden
> ----------------------------------------------------------------------
> _______________________________________________
> comp.lang.ada mailing list
> comp.lang.ada@ada.eu.org
> http://ada.eu.org/mailman/listinfo/comp.lang.ada
>
>

Re: Missing **-operator in GNAT 3.14?

[Robert Dewar]

"David C. Hoos, Sr." <david.c.hoos.sr@ada95.com> wrote in message news:<mailman.1017746102.31836.comp.lang.ada@ada.eu.org>...

> The fact that "*", and "**" are treated differently by 
> GNAT looks like a "bug" to me, though, and you should 
> probably report it to report@gnat.com

Nothing to report here, the diagnostic is correct! If
some earlier version of GNAT accepted this, it was a bug
in that earlier version, since the code is wrong.

Re: Missing **-operator in GNAT 3.14?

[David C. Hoos]

I do not dispute the correctness of the diagnostic, but what I am
curious about is why the diagnostic complains about "**", but
doesn't complain about "*".

What is the difference between these two operators, in this context?

David Hoos

----- Original Message -----
From: "Robert Dewar" <dewar@gnat.com>
Newsgroups: comp.lang.ada
To: <comp.lang.ada@ada.eu.org>
Sent: Tuesday, April 02, 2002 10:18 AM
Subject: Re: Missing **-operator in GNAT 3.14?


> "David C. Hoos, Sr." <david.c.hoos.sr@ada95.com> wrote in message
news:<mailman.1017746102.31836.comp.lang.ada@ada.eu.org>...
>
> > The fact that "*", and "**" are treated differently by
> > GNAT looks like a "bug" to me, though, and you should
> > probably report it to report@gnat.com
>
> Nothing to report here, the diagnostic is correct! If
> some earlier version of GNAT accepted this, it was a bug
> in that earlier version, since the code is wrong.
> _______________________________________________
> comp.lang.ada mailing list
> comp.lang.ada@ada.eu.org
> http://ada.eu.org/mailman/listinfo/comp.lang.ada
>

Re: Missing **-operator in GNAT 3.14?

[Sergey Koshcheyev]

"David C. Hoos" <david.c.hoos.sr@ada95.com> wrote in message
news:mailman.1017765842.8487.comp.lang.ada@ada.eu.org...
> What is the difference between these two operators, in this context?
>
> David Hoos

The difference apparently lies in this:
7. The highest precedence exponentiation operator ** is predefined for every
specific integer type T with the following specification:
8. function "**"(Left : T; Right : Natural) return T
(http://www.ada-auth.org/arm-html/RM-4-5-6.html)

So the second operand of "**" has to be of type Natural, regardless of the
type of the first operand.

Sergey.

Re: Missing **-operator in GNAT 3.14?

[Robert Dewar]

"Sergey Koshcheyev" <serko84@hotmail.com> wrote in message news:<a8con5$2emm$1@ns.felk.cvut.cz>...
> So the second operand of "**" has to be of type Natural, 
> regardless of the type of the first operand.

Good going Sergey, I was wondering when someone might
actually look up the declaration of "**". After all the
diagnostic said that no matching "**" was declared, you
would think the first reaction would be to go look at the
declared "**" and see why it does not match :-)

Re: Missing **-operator in GNAT 3.14?

[Steve Doiel]

"Robert Dewar" <dewar@gnat.com> wrote in message
news:5ee5b646.0204021702.bb04357@posting.google.com...
> "Sergey Koshcheyev" <serko84@hotmail.com> wrote in message
news:<a8con5$2emm$1@ns.felk.cvut.cz>...
> > So the second operand of "**" has to be of type Natural,
> > regardless of the type of the first operand.
>
> Good going Sergey, I was wondering when someone might
> actually look up the declaration of "**". After all the
> diagnostic said that no matching "**" was declared, you
> would think the first reaction would be to go look at the
> declared "**" and see why it does not match :-)

Gee.  If I didn't know better I would think you were a professor who makes
his students think (and gives positive feedback when someone getss it right)
 :-)

SteveD

Re: Missing **-operator in GNAT 3.14?

[Robert Dewar]

<<I do not dispute the correctness of the diagnostic, but what I am
curious about is why the diagnostic complains about "**", but
doesn't complain about "*".
>>

Because a matching "*" is declared but no matching "**"