From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: 
X-Spam-Status: No, score=-1.3 required=5.0 tests=BAYES_00,INVALID_MSGID
	autolearn=no autolearn_force=no version=3.4.4
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 103376,505b52f527b29fd6
X-Google-Attributes: gid103376,public
From: Peter Amey <pna@erlang.praxis.co.uk>
Subject: Re: Are there any "bad" Ada constructs?
Date: 1996/09/05
Message-ID: <Pine.SUN.3.91.960905094006.2507B-100000@erlang.praxis.co.uk>#1/1
X-Deja-AN: 178621017
references: <322E16DC.74B1@swcp.com>
content-type: TEXT/PLAIN; charset=US-ASCII
organization: Praxis plc, U.K.
mime-version: 1.0
newsgroups: comp.lang.ada
Date: 1996-09-05T00:00:00+00:00
List-Id: <comp.lang.ada>



> 
> My question is: Are there any constructs in Ada that are permitted to do
>     two or more different things and still be correct, in a manner that
>     is similar to that listed above?
> 

There are, mostly associated with aliasing, evaluation order and 
elaboration order.  A simple example is:

procedure Init2(X, Y : out integer)
is
begin
  X := 1;
  Y := 2;
end Init2;

The call Init2(A, A); can set A to 1 or 2 because the order of parameter 
association is undefined (I have two validated compilers which actually 
demonstrate this effect).

Ada has a bit of a cop out in that it labels programs whose meaning 
varies in this way "erroneous"; however, applying such a label doesn't 
make the problems any easier to detect or prevent.  Some secure Ada 
subsets such as SPARK are completely free of this form of anomalous 
behaviour.

Peter