Extended 'Succ attribute?

Extended 'Succ attribute?

[Jérémie Lumbroso]

Hello,

I just started Ada and was wondering if there is a way to make the Succ
attribute of discrete types behave in a cyclic manner (akin to the
modular type):

    type T_DAY is (Mon, Tue, Wed, Thu, Fri, Sat, Sun);

    ...

    day := T_DAY'(Sun);
    day := T_DAY'Succ(day);  -- that this returns 'Mon' instead of
                             -- a type constraint error

I could of course define my own function to do this, but I wish to know
if there is a way to get Ada to do it itself.

Regards,

Jérémie

Re: Extended 'Succ attribute?

[Adam Beneschan]

Jérémie Lumbroso wrote:
> Hello,
>
> I just started Ada and was wondering if there is a way to make the Succ
> attribute of discrete types behave in a cyclic manner (akin to the
> modular type):
>
>     type T_DAY is (Mon, Tue, Wed, Thu, Fri, Sat, Sun);
>
>     ...
>
>     day := T_DAY'(Sun);
>     day := T_DAY'Succ(day);  -- that this returns 'Mon' instead of
>                              -- a type constraint error
>
> I could of course define my own function to do this, but I wish to know
> if there is a way to get Ada to do it itself.

No.

The best you could do would be to declare a modular type, and then
declare Mon, Tue, etc. as constants of that type:

    type T_DAY is mod 7;
    Mon : constant T_DAY := 0;
    Tue : constant T_DAY := 1;
etc.

Then T_DAY'Succ(X) will wrap around.  But I don't see any advantage to
doing things this way.  I'd
just make it an enumeration literal and define your own Succ function.

                        -- Adam

Re: Extended 'Succ attribute?

[Jeffrey R. Carter]

J�r�mie Lumbroso wrote:
> 
> I could of course define my own function to do this, but I wish to know
> if there is a way to get Ada to do it itself.

No, but rather than write it yourself, you could simply instantiate 
PragmARC.Wrapping.

http://pragmada.home.mchsi.com/

-- 
Jeff Carter
"We burst our pimples at you."
Monty Python & the Holy Grail
16

Re: Extended 'Succ attribute?

[Peter Hermann]

Adam Beneschan <adam@irvine.com> wrote:
> J?r?mie Lumbroso wrote:
[snip all]

I agree 100% to Adam Benescha.
Moreover, my experience says, that array accesses are fastest.
However, that still depends on tests.
I played around with the following:


-- package Ada.Calendar.Formatting is
-- -- Day of the week:
-- type Day_Name is (Monday, Tuesday, Wednesday, Thursday,
-- Friday, Saturday, Sunday);
-- function Day_of_Week (Date : Time) return Day_Name;

with Ada.Calendar.Formatting;
 use Ada.Calendar.Formatting;
with ada.text_io;
procedure ada2005calendartest3ph is
  type next_day_type is array(day_name) of day_name;
  next_day : next_day_type :=
  (Tuesday, Wednesday, Thursday, Friday, Saturday, Sunday, Monday);
  specimen : day_name := Friday;
begin
  ada.text_io.put_line("starting program ada2005calendartest3ph");
  for i in 1..9 loop
    ada.text_io.put_line(day_name'image(specimen));
    specimen:=next_day(specimen);
  end loop;
--result:
--starting program ada2005calendartest3ph
--FRIDAY
--SATURDAY
--SUNDAY
--MONDAY
--TUESDAY
--WEDNESDAY
--THURSDAY
--FRIDAY
--SATURDAY
end ada2005calendartest3ph; --20061124ph
-- 
--Peter.Hermann@ihr.uni-stuttgart.de        (+49)0711-685-872-44(Fax79)
--Nobelstr.19 Raum 0.030, D-70569 Stuttgart IHR Hoechstleistungsrechnen
--http://www.ihr.uni-stuttgart.de/