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=-1.9 required=5.0 tests=BAYES_00 autolearn=ham
	autolearn_force=no version=3.4.4
X-Google-Thread: 103376,42795ada22dedd1b
X-Google-Attributes: gid103376,public,usenet
X-Google-Language: ENGLISH,ASCII-7-bit
Path: 
 g2news2.google.com!postnews.google.com!i38g2000prf.googlegroups.com!not-for-mail
From: Adam Beneschan <adam@irvine.com>
Newsgroups: comp.lang.ada
Subject: Re: Legal usage of downward closures
Date: Thu, 16 Aug 2007 10:08:30 -0700
Organization: http://groups.google.com
Message-ID: <1187284110.812863.280360@i38g2000prf.googlegroups.com>
References: <1187280678.505830.294720@a39g2000hsc.googlegroups.com>
NNTP-Posting-Host: 66.126.103.122
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
X-Trace: posting.google.com 1187284111 24693 127.0.0.1 (16 Aug 2007 17:08:31
 GMT)
X-Complaints-To: groups-abuse@google.com
NNTP-Posting-Date: Thu, 16 Aug 2007 17:08:31 +0000 (UTC)
In-Reply-To: <1187280678.505830.294720@a39g2000hsc.googlegroups.com>
User-Agent: G2/1.0
X-HTTP-UserAgent: Mozilla/5.0 (X11; U; Linux x86_64; en-US;
 rv:1.7.12) Gecko/20050922 Fedora/1.7.12-1.3.1,gzip(gfe),gzip(gfe)
Complaints-To: groups-abuse@google.com
Injection-Info: i38g2000prf.googlegroups.com; posting-host=66.126.103.122;
   posting-account=ps2QrAMAAAA6_jCuRt2JEIpn5Otqf_w0
Xref: g2news2.google.com comp.lang.ada:1462
Date: 2007-08-16T10:08:30-07:00
List-Id: <comp.lang.ada>

On Aug 16, 9:11 am, mala...@magic.fr wrote:
> Hi,
>
> I have problems understanding the RAT05 section on downward closures.
>
> I thought that anonymous access to a sub-program, as parameter of a
> sub-program, would prevent local copies and further usage out of
> scope.
> Surprisingly, the following example compiles an runs (on GNAT GPL
> 2007).

The assignment statements are illegal.  Every "anonymous access type"
definition defines its own type; therefore Acc1 and Arg don't have the
same type, and neither of those types is the same as Acc2_Type.  GNAT
has a bug if it accepts them.

Furthermore, this would be illegal due to accessibility level rules:

    Acc2 := Acc_Type(Arg);

                               -- Adam




> package Pack is
>   procedure Store (Arg : access procedure);
>   procedure Call;
> end Pack;
>
> package body Pack is
>   Acc1 : access procedure;
>   type Acc_Type is access procedure;
>   Acc2 : Acc_Type;
>
>   procedure Store (Arg : access procedure) is
>   begin
>     Acc1 := Arg;  -- ILLEGAL
>     Acc2 := Arg;  -- ILLEGAL
>   end Store;
>   procedure Call is
>   begin
>     Acc1.all;
>     Acc2.all;
>   end Call;
>
> end Pack;