formal package question

formal package question

[ytomino]

Hello,
Please look this code:

--------
package formalpkg is

   generic
      type T is private;
      with procedure P (X : T) is <>;
   package F is
   end F;

   generic
      with package FA is new F (others => <>);
   package B is
   end B;

   procedure P1 (X : Character) is null;
   package F1 is new F (Character, P1); -- use P => P1

   package B1 is new B (F1); -- Error !!

   procedure P (X : Character) is null;
   package F2 is new F (Character); -- P => P

   package B2 is new B (F2); -- OK

end formalpkg;
--------
% gnatmake formalpkg.ads
gcc -c formalpkg.ads
formalpkg.ads:17:25: actual for "P" in actual instance does not match
formal
gnatmake: "formalpkg.ads" compilation error
--------

B1 was compile error, but B2 is ok. Why?
Where is wrong in my code?
I want to use instances of generic package with some different
subprograms like B1...

(I use gcc-4.5.1)

Re: formal package question

[anon]

Typo:

   package F2 is new F (Character, P); -- P => P



In <a1ce6c12-d389-4685-95b2-68159efefadc@s11g2000prs.googlegroups.com>, ytomino <aghia05@gmail.com> writes:
>Hello,
>Please look this code:
>
>--------
>package formalpkg is
>
>   generic
>      type T is private;
>      with procedure P (X : T) is <>;
>   package F is
>   end F;
>
>   generic
>      with package FA is new F (others => <>);
>   package B is
>   end B;
>
>   procedure P1 (X : Character) is null;
>   package F1 is new F (Character, P1); -- use P => P1
>
>   package B1 is new B (F1); -- Error !!
>
>   procedure P (X : Character) is null;
>   package F2 is new F (Character); -- P => P
>
>   package B2 is new B (F2); -- OK
>
>end formalpkg;
>--------
>% gnatmake formalpkg.ads
>gcc -c formalpkg.ads
>formalpkg.ads:17:25: actual for "P" in actual instance does not match
>formal
>gnatmake: "formalpkg.ads" compilation error
>--------
>
>B1 was compile error, but B2 is ok. Why?
>Where is wrong in my code?
>I want to use instances of generic package with some different
>subprograms like B1...
>
>(I use gcc-4.5.1)

Re: formal package question

[ytomino]

The parameter P of F has default box(<>), it's valid.

On Feb 14, 3:51 pm, a...@att.net wrote:
> Typo:
>
>    package F2 is new F (Character, P); -- P => P

Re: formal package question

[Georg Bauhaus]

On 14.02.11 03:46, ytomino wrote:
> Hello,
> Please look this code:
> 
> --------
> package formalpkg is
> 
>    generic
>       type T is private;
>       with procedure P (X : T) is <>;
>    package F is
>    end F;
> 
>    generic
>       with package FA is new F (others => <>);
>    package B is
>    end B;
> 
>    procedure P1 (X : Character) is null;
>    package F1 is new F (Character, P1); -- use P => P1
> 
>    package B1 is new B (F1); -- Error !!
> 
>    procedure P (X : Character) is null;
>    package F2 is new F (Character); -- P => P
> 
>    package B2 is new B (F2); -- OK
> 
> end formalpkg;
> --------
> % gnatmake formalpkg.ads
> gcc -c formalpkg.ads
> formalpkg.ads:17:25: actual for "P" in actual instance does not match
> formal
> gnatmake: "formalpkg.ads" compilation error
> --------
> 
> B1 was compile error, but B2 is ok. Why?

I get an error for B2, probably using a different GNAT.
Not really an answer, but can you leave out the "others"
in ([others =>] <>) ?

package Formalpkg is

    generic
        type T is private;
        with procedure P (X : T) is <>;
    package F is
    end F;

    generic
        with package FA is new F (<>);
    package B is
    end B;

    procedure P1 (X : Character) is null;
    package F1 is new F (Character, P1);

    package B1 is new B (F1);

    procedure P (X : Character) is null;
    package F2 is new F (Character);

    package B2 is new B (F2);

end Formalpkg;

Re: formal package question

[Adam Beneschan]

On Feb 14, 1:59 am, Georg Bauhaus <rm.dash-bauh...@futureapps.de>
wrote:
> On 14.02.11 03:46, ytomino wrote:
>
>
>
>
>
> > Hello,
> > Please look this code:
>
> > --------
> > package formalpkg is
>
> >    generic
> >       type T is private;
> >       with procedure P (X : T) is <>;
> >    package F is
> >    end F;
>
> >    generic
> >       with package FA is new F (others => <>);
> >    package B is
> >    end B;
>
> >    procedure P1 (X : Character) is null;
> >    package F1 is new F (Character, P1); -- use P => P1
>
> >    package B1 is new B (F1); -- Error !!
>
> >    procedure P (X : Character) is null;
> >    package F2 is new F (Character); -- P => P
>
> >    package B2 is new B (F2); -- OK
>
> > end formalpkg;
> > --------
> > % gnatmake formalpkg.ads
> > gcc -c formalpkg.ads
> > formalpkg.ads:17:25: actual for "P" in actual instance does not match
> > formal
> > gnatmake: "formalpkg.ads" compilation error
> > --------
>
> > B1 was compile error, but B2 is ok. Why?
>
> I get an error for B2, probably using a different GNAT.
> Not really an answer, but can you leave out the "others"
> in ([others =>] <>) ?

Yes, it compiles for me if I say F(<>) but gets an error with F(others
=> <>).  There should be no difference between the two syntaxes, so
this is an error in the compiler.

                               -- Adam

Re: formal package question

[ytomino]

I tried to remove "others =>", and got same result that it compiled
correctly.
umm...
This behavior of compiler limits using formal parameter...
I want to write partial parameters and use defaults for remaining
parameters.

But gcc(GNAT) can compile the example in
http://www.adaic.org/resources/add_content/standards/05rm/html/RM-12-7.html
I deduce, this behavior is applied only to generic package having
formal package having formal subprograms ???

Thank you.

On Feb 14, 6:59 pm, Georg Bauhaus <rm.dash-bauh...@futureapps.de>
wrote:
>
> I get an error for B2, probably using a different GNAT.
> Not really an answer, but can you leave out the "others"
> in ([others =>] <>) ?
>
> package Formalpkg is
>
>     generic
>         type T is private;
>         with procedure P (X : T) is <>;
>     package F is
>     end F;
>
>     generic
>         with package FA is new F (<>);
>     package B is
>     end B;
>
>     procedure P1 (X : Character) is null;
>     package F1 is new F (Character, P1);
>
>     package B1 is new B (F1);
>
>     procedure P (X : Character) is null;
>     package F2 is new F (Character);
>
>     package B2 is new B (F2);
>
> end Formalpkg;

Re: formal package question

[Ludovic Brenta]

ytomino wrote on comp.lang.ada:
> I tried to remove "others =>", and got same result that it compiled
> correctly.
> umm...
> This behavior of compiler limits using formal parameter...
> I want to write partial parameters and use defaults for remaining
> parameters.
>
> But gcc(GNAT) can compile the example inhttp://www.adaic.org/resources/add_content/standards/05rm/html/RM-12-...
> I deduce, this behavior is applied only to generic package having
> formal package having formal subprograms ???
>
> Thank you.

I reported this bug as http://gcc.gnu.org/PR47748 for you.

--
Ludovic Brenta.

Re: formal package question

[ytomino]

> I reported this bug ashttp://gcc.gnu.org/PR47748for you.
Thank you very much!

>      with package FA is new F (others => <>); -- line 17
An added comment "-- line 17" is misplaced. The real line 17 is
>   package B1 is new B (F1); -- Error !!

On Feb 15, 5:17 pm, Ludovic Brenta <ludo...@ludovic-brenta.org> wrote:
>
> I reported this bug ashttp://gcc.gnu.org/PR47748for you.
>
> --
> Ludovic Brenta.

Re: formal package question

[Ludovic Brenta]

ytomino writes on comp.lang.ada:
>> I reported this bug as http://gcc.gnu.org/PR47748 for you.
> Thank you very much!
>
>>      with package FA is new F (others => <>); -- line 17
> An added comment "-- line 17" is misplaced. The real line 17 is
>>   package B1 is new B (F1); -- Error !!

I realized that after the fact ut it doesn't matter because, in the
comments, I say: "On line 17, replacing "others => <>" with just "<>"
makes the package compile cleanly.".  I might as well have said "on line
A".

-- 
Ludovic Brenta.