Conformance problem in VAX Ada?

Conformance problem in VAX Ada?

[Dave Papay M/S 7G32 x2791]

While writing some code, I accidentally put different default expressions for
a parameter in the declaration and body of a procedure.  Here's what it looked
like, except that procedure Error did some real work ;-) :

   with Text_IO;
   procedure Test is

      procedure Error (I  : in Integer := Integer'First);

      procedure Error (I  : in Integer := Integer'Last) is
      begin
         Text_IO.Put_Line (Integer'Image (I));
      end Error;

   begin
      Error;
   end Test;

I compiled it using VAX Ada v2.3-3 under VAX/VMS v5.5-2.  The compiler didn't
complain, and I didn't discover the error until I started seeing the effects
of using Integer'Last, rather than Integer'First (which is what I really
wanted).  I read over the conformance rules for subprograms, and (as far as I
can determine) I don't think these two specifications conform.  The compiler
thinks they do.  Is this a compiler problem, or am I missing something in the
conformace rules?

                             |         |
David Papay                  |    o    |
papay@acm.org  (preferred)   |  ^/---  |    insert your favorite
dpapay@aol.com (alternate)   |  />     |      disclaimer here
                             |         |
                             | fleche! |

Re: Conformance problem in VAX Ada?

[Norman H. Cohen]

In article <00988B51.1523CFA0.69@gtewd.mtv.gtegsc.com>, Dave Papay
M/S 7G32 x2791 <papayd@GTEWD.MTV.GTEGSC.COM> writes: 

|> While writing some code, I accidentally put different default expressions for
|> a parameter in the declaration and body of a procedure.  Here's what it looked
|> like, except that procedure Error did some real work ;-) : 
|>
|>    with Text_IO;
|>    procedure Test is
|>
|>       procedure Error (I  : in Integer := Integer'First);
|>
|>       procedure Error (I  : in Integer := Integer'Last) is
|>       begin
|>          Text_IO.Put_Line (Integer'Image (I));
|>       end Error;
|>
|>    begin
|>       Error;
|>    end Test;
|>
|> I compiled it using VAX Ada v2.3-3 under VAX/VMS v5.5-2.  The compiler didn't
|> complain
...
|>                  Is this a compiler problem, or am I missing something in the
|> conformace rules?

Definitely a compiler problem.  Funny thing is, I tried it on a compiler
with a quite different lineage and got the same result.  When I changed
the default value in the Error declaration from Integer'First to 0,
however, the compiler caught the mismatch.  When I changed it to
Natural'Size, the mismatch was not caught.  The compiler's strategy seems
to be, "If you've seen one attribute, you've seen them all."

--
Norman H. Cohen    ncohen@watson.ibm.com

Re: Conformance problem in VAX Ada?

[Andreas Schulz RIT14]

In article <00988B51.1523CFA0.69@gtewd.mtv.gtegsc.com>, Dave Papay M/S 7G32 x2791 <papayd@GTEWD.MTV.GTEGSC.COM> writes:
|> I compiled it using VAX Ada v2.3-3 under VAX/VMS v5.5-2.  The compiler didn't
|> complain, and I didn't discover the error until I started seeing the effects
|> of using Integer'Last, rather than Integer'First (which is what I really
|> wanted).  I read over the conformance rules for subprograms, and (as far as I
|> can determine) I don't think these two specifications conform.  The compiler
|> thinks they do.  Is this a compiler problem, or am I missing something in the
|> conformace rules?

Well, SunAda thinks they don't conform :

Sun Ada Compiler, Copyright 1984, 1992
Sun-4 SunOS Release 4.0 and 4.1, Version 1.1
Mon Jan 27 13:31:12 PST 1992, Sun Ada 1.1(f)

File: /.../conform.a
        compiled Tue Dec 13 09:55:55 1994

unit:   subprogram test
        2 SEMANTIC ERRORS               UNIT UNCHANGED

3 statements    14 lines

*********************************  conform.a  *********************************

   6:      procedure Error (I  : in Integer := Integer'Last) is
A -------------------^
A:error: RM 6.3.1: subprogram does not conform with specification from:
A:error:        error, line 4

Re: Conformance problem in VAX Ada?

[Robert Dewar]

I certainly agree that this is a compiler problem (failure to diagnose
different default expressions as conformance problem). Here is the GNAT
output:

     1. with Text_IO;
     2. procedure Test is
     3.
     4.    procedure Error (I  : in Integer := Integer'First);
     5.
     6.    procedure Error (I  : in Integer := Integer'Last) is
                     |
        >>> not fully conformant with declaration at line 4
        >>> default expression for "I" does not match

     7.    begin
     8.       Text_IO.Put_Line (Integer'Image (I));
     9.    end Error;
    10.
    11. begin
    12.    Error;
    13. end Test;

THe checking of default expressions is tricky, because they don't have
to be identical (for instance X.A can match A, and things get worse in
9X where "+(a,b) can match a+b. On the other hand 3 does not match 1+2,
so if your compiler folds 1+2 to 3 (which you can expect given that 1+2
is static), it had better make sure this does not happen at a point that
will result in false conformance.

Still this particular case seems pretty straightforward, and would suggest
that the compiler(s) involved are taking some shortcuts on conformance
analysis.