Apex problem w/aliased variables

Apex problem w/aliased variables

[paul williams]

We are developing a large scientific application using Ada95 
with the Rational Apex tool. Our design depends heavily on being 
able to take advantage of the new access mechanism for 
non-allocated variables. However, the Apex analyzer reports an 
error when attempting to assign the access attribute of an 
aliased variable to a pointer that is appropriately declared for 
the entire class. See code sample followed by error report 
below. Is anyone else using Apex and having this problem? If so, 
have you determined a suitable workaround? I'm told that 
Rational knows about the problem but has no patches. Any 
suggestions?

package Test is
    type Object is tagged record
        Component : Integer;
    end record;
    type Object_Ptr is access all Object'Class;
    type My_Object is new Object with null record;

    The_Object : aliased My_Object;
    The_Ptr : Object_Ptr := The_Object'Access;
end Test;

16:35:46 >>> THE_PTR : OBJECT_PTR := THE_OBJECT'ACCESS;
16:35:46 *** THE_OBJECT is not a value of OBJECT'CLASS

Thanks in advance for any assistance.
Regards,
Paul

problemes mathematiques

[koniez tayeb]

Salut,

J'aimerai savoir comment on obtient
sous ada95 les op�rateurs suivants :

- racine carr�e.
- racine n i�me.
- sin, cos, arc-cos et autres fonctions
  trigonom�triques.

Si vous avez une documentation �lectronique
sur ce sujet cela m'int�resserai au plus haut
point.

Mille merci d'avance .

A+

Re: problemes mathematiques

[Laurent Guerby]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 864 bytes --]


[Note: there's currently a CFV for fr.comp.lang.ada]

 > J'aimerai savoir comment on obtient sous ada95 les op�rateurs
 > suivants :

with Ada.Numeric.Elementary_Functions; use Ada.Numeric.Elementary_Functions;

 > - racine carr�e.

   Sqrt (X)

 > - racine n i�me.  
 
   ~= Exp (Log (X) / N)   

 > sin, cos, arc-cos et autres fonctions trigonom�triques.

   Sin, Cos, Arccos, ...

 > Si vous avez une documentation �lectronique sur ce sujet cela
 > m'int�resserai au plus haut point.

URLs: http://www.adahome.com/rm95/rm9x-A-05.html
      http://www.adahome.com/rm95/rm9x-G.html
      http://www.adahome.com/LRM/95/Rationale/rat95html/rat95-p3-a.html#3
      http://www.adahome.com/LRM/95/Rationale/rat95html/rat95-p3-g.html

-- 
Laurent Guerby <guerby@gnat.com>, Team Ada.
   "Use the Source, Luke. The Source will be with you, always (GPL)."

Re: problemes mathematiques

[Keith Thompson]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 747 bytes --]


In <01bbd5f1$b994c530$262a6282@cln49ae> "Pascal Obry" <pascal.obry@der.edfgdf.fr> writes:
[...]
> Ce ne sont pas des op�rateurs mais des fonctions que tu trouveras dans le
> package Ada.Numerics.Elementary_Functions.
> 
> racine carr�e x : Sqrt (x)
> racine n i�me   : Sqrt (x) / Sqrt (n)

I presume "racine n i�me" is "nth root" (cube root, fourth root, etc.).
The formula given above, Sqrt (x) / Sqrt (n), is incorrect.  The correct
formula is x ** 1.0/n (you'll have convert n to the appropriate
floating-point type first).

-- 
Keith Thompson (The_Other_Keith) kst@thomsoft.com <*>
TeleSoft^H^H^H^H^H^H^H^H Alsys^H^H^H^H^H Thomson Software Products
10251 Vista Sorrento Parkway, Suite 300, San Diego, CA, USA, 92121-2706
FIJAGDWOL

Re: problemes mathematiques

[Samuel Tardieu]

>>>>> "Keith" == Keith Thompson <kst@thomsoft.com> writes:

Keith> I presume "racine n i�me" is "nth root" (cube root, fourth
Keith> root, etc.). The formula given above, Sqrt (x) / Sqrt (n), is
Keith> incorrect.  The correct formula is x ** 1.0/n (you'll have
Keith> convert n to the appropriate floating-point type first).

I presume you mean x ** (1.0/n) ? (I thought first that it was a
mathematical formula rather than an Ada expression, but your comment
about n being converted to the appropriate floating-point type makes me
think differently :-)

  Sam
-- 
  Samuel Tardieu -- sam@ada.eu.org

Re: problemes mathematiques

[Keith Thompson]

In <qw67mnfd6ls.fsf@esmeralda.enst.fr> Samuel Tardieu <sam@ada.eu.org> writes:
> >>>>> "Keith" == Keith Thompson <kst@thomsoft.com> writes:
> Keith> The correct formula is x ** 1.0/n (you'll have to
> Keith> convert n to the appropriate floating-point type first).
> 
> I presume you mean x ** (1.0/n) ?

Actually, I was using spacing to indicate the association of the operands.
Oh, you mean your compiler doesn't implement this extension?

(Ok, ok, I forgot the parentheses.)

-- 
Keith Thompson (The_Other_Keith) kst@aonix.com <http://www.aonix.com> <*>
TeleSo^H^H^H^H^H^H Alsy^H^H^H^H Thomson Softw^H^H^H^H^H^H^H^H^H^H^H^H^H Aonix
10251 Vista Sorrento Parkway, Suite 300, San Diego, CA, USA, 92121-2706
"SPOON!" -- The Tick

Re: problemes mathematiques

[lecomteg]

I'm not sure the parentheses are needed. I think I remember the priority
of
** was a little silly (or did I dream it?).

A propos, the LRM is defenitely *not* the best book on Ada (just a bunch
of
legalese). Did THEY really intend it to be read (and understood) ?

Well, I recognise that having one handy would have helped me
being sure about that ** vs / priority... ;-)

G Lecomte

<<<Strong typing is for people with weak memories>>>