Re: fyi, GNAT and SPARK GPL 2016 are out

[gautier_niouzes@hotmail.com]

Le samedi 4 juin 2016 18:31:29 UTC+2, Georg Bauhaus a écrit :

> Doesn't your workaround demonstrate just how the behavior
> shown by GNAT contradicts the one to expect from standards
> conformance? Or is -gnatp now overruling the effect which
> 
>      Element (No_Element)
> 
> is supposed to have?

Indeed. Without pragma Suppress(Container_Checks) you get the RM05/12 behavior:
raised CONSTRAINT_ERROR : Test_2016.T_Dic.Element: Position cursor of function Element equals No_Element

(A.18.4 Maps: 33/2
function Element (Position : Cursor) return Element_Type;
34/2
If Position equals No_Element, then Constraint_Error is propagated. Otherwise, Element returns the element component of the node designated by Position.)

with pragma Suppress(Container_Checks) (or with -gnatp) you get
raised PROGRAM_ERROR : EXCEPTION_ACCESS_VIOLATION

No coincidence, since you have in a-cohama.adb (GPL 2016 version):
   function Element (Position : Cursor) return Element_Type is
   begin
      if Checks and then Position.Node = null then
         raise Constraint_Error with
           "Position cursor of function Element equals No_Element";
      end if;
      ...
and (one click away):
   Checks : constant Boolean := Container_Checks'Enabled;

Hence my question about Container_Checks being in a future version of the standard...
_________________________ 
Gautier's Ada programming 
http://sf.net/users/gdemont/