From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.5-pre1 (2020-06-20) on
	ip-172-31-74-118.ec2.internal
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 required=3.0 tests=BAYES_00 autolearn=ham
	autolearn_force=no version=3.4.5-pre1
Date: 23 Jun 93 02:09:12 GMT
From: pa.dec.com!e2big.mko.dec.com!jac.nuo.dec.com!digits.enet.dec.com!brett@de
cwrl.dec.com  (Bevin R. Brett)
Subject: Re: References in Ada exceptions
Message-ID: <1993Jun23.011133.8132@jac.nuo.dec.com>
List-Id: <comp.lang.ada>

In article <3176.2c26e734@atlas.nafb.trw.com>, mgatlin@atlas.nafb.trw.com write
s...
>[ The following is a request for information by a member of my department.
>  Please reply direct to my email address, and I will forward. mg.]
> 
>I seem to remember that the following is dangerous:
> 
>	for i in 1..10 loop
>	 begin
>	   ...
>	 exception
>		...
>		put(i);
>	 end;
>	end loop;
> 
>The state of the variable i cannot be guaranteed in an exception 
>handler.  We had a problem like this on the DEC Ada compiler, and
>I claim it was not a compiler bug, but a known limitation of Ada 
>validated compilers.  Is there a formal ruling on this?  We require 
>a hard reference.  Please reply by email to mgatlin@atlas.nafb.trw.com.

If a DEC Ada product did this, we would definitely regard it as a bug.
Exceptions within the begin..exception...end should not interrupt operations
such as the loop increment done outside the frame.

I have sent personal mail asking for help reproducing the problem.

/Bevin
DEC Ada team