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=-0.3 required=5.0 tests=BAYES_00,
	REPLYTO_WITHOUT_TO_CC autolearn=no autolearn_force=no version=3.4.4
Path: 
 eternal-september.org!reader01.eternal-september.org!reader02.eternal-september.org!news.eternal-september.org!mx02.eternal-september.org!.POSTED!not-for-mail
From: "G.B." <bauhaus@futureapps.invalid>
Newsgroups: comp.lang.ada
Subject: Re: working with diferent instances of the same generic interface
Date: Tue, 8 Sep 2015 19:11:38 +0200
Organization: A noiseless patient Spider
Message-ID: <msn4or$vfe$1@dont-email.me>
References: <625c79d2-f5ed-4cd4-80ea-ea310e381017@googlegroups.com>
Reply-To: nonlegitur@futureapps.de
Mime-Version: 1.0
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Transfer-Encoding: 7bit
Injection-Date: Tue, 8 Sep 2015 17:09:47 +0000 (UTC)
Injection-Info: mx02.eternal-september.org;
 posting-host="b96887e80893c84a90c3007226ca0d1c";
	logging-data="32238"; mail-complaints-to="abuse@eternal-september.org";
	posting-account="U2FsdGVkX1+CznB8d+UDrqkgRw6MsuBiiJXLUqVU1jY="
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.10; rv:38.0)
 Gecko/20100101 Thunderbird/38.2.0
In-Reply-To: <625c79d2-f5ed-4cd4-80ea-ea310e381017@googlegroups.com>
Cancel-Lock: sha1:87veXTZccPZmZdL286RC5ljHgCQ=
Xref: news.eternal-september.org comp.lang.ada:27745
Date: 2015-09-08T19:11:38+02:00
List-Id: <comp.lang.ada>

On 08.09.15 17:10, Aitor Alcrudo Sangros wrote:

> procedure try is
>     man :manager ;
>     elm1 : Element_type_A_Access ;
>     elm2 : Element_type_B_Access ;
> begin
>     man.add(Reporter.Reporting_basic.Reporting_Element_Access(elm1)  );
>     man.add(Reporter.Reporting_basic.Reporting_Element_Access(elm2)  );
> end;
>
> This compiles, but produces a runtime exception with invalid interface conversion.
>
> Is there any way to get this to work?.


Not sure, after skimming. (It looks like a C++ design written
using Ada syntax, although there is a formal tagged type in
the first template's parameters ;-)

I think the problem may be one of trying to convert half-a-bee
into half-a-frog. That's in try's block of statements, where the
other half would be Reporting_Element. But, by the above remark,
I'm not sure. Simplifying, I can get GNAT in trouble using this
program, trying to expose what I think the conversion of elm1
and elm2 amounts to:

procedure Test_Tp is

    package Tp is

       type X is interface;

       type A is tagged null record;
       type B is tagged null record;

       type C is new A and X with null record;
       type D is new B and X with null record;

    end Tp;


    use Tp;

    type X_Ptr is access all X'Class;
    Item : X_Ptr := new D;

    Z : X'Class := X(Item.all);  -- <-- Here
begin
    null;
end Test_Tp;

The running program show memory access failures on some
platforms, referring to the line marked "<-- Here".

Starting gdb, after compiling the library with the program,
I see

Program received signal SIGSEGV, Segmentation fault.
0x000000000040e017 in ada.tags.displace (this=(system.address) 
0x7fffffffd408, t=0x7fffffffdf48 (test_tp.tp.x)) at 
/home/bauhaus/opt/GNAT2014/lib/gcc/x86_64-pc-linux-gnu/4.7.4/adainclude/a-tags.adb:375
375	      Obj_DT_Tag  := To_Tag_Ptr (Obj_Base).all;

So, one might conclude that to much multiple inheritance is
involved...