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,FREEMAIL_FROM
	autolearn=unavailable autolearn_force=no version=3.4.4
X-Received: by 10.66.243.5 with SMTP id wu5mr4033624pac.29.1429278175452;
        Fri, 17 Apr 2015 06:42:55 -0700 (PDT)
X-Received: by 10.182.22.134 with SMTP id d6mr21296obf.41.1429278175413; Fri,
 17 Apr 2015 06:42:55 -0700 (PDT)
Path: 
 eternal-september.org!reader01.eternal-september.org!reader02.eternal-september.org!news.eternal-september.org!mx02.eternal-september.org!feeder.eternal-september.org!news.glorb.com!l13no6337107iga.0!news-out.google.com!db6ni18153igc.0!nntp.google.com!m20no4084153iga.0!postnews.google.com!glegroupsg2000goo.googlegroups.com!not-for-mail
Newsgroups: comp.lang.ada
Date: Fri, 17 Apr 2015 06:42:55 -0700 (PDT)
Complaints-To: groups-abuse@google.com
Injection-Info: glegroupsg2000goo.googlegroups.com; posting-host=67.0.18.148;
 posting-account=lJ3JNwoAAAAQfH3VV9vttJLkThaxtTfC
NNTP-Posting-Host: 67.0.18.148
User-Agent: G2/1.0
MIME-Version: 1.0
Message-ID: <8def4a86-5835-4368-b414-63e12f1074e8@googlegroups.com>
Subject: Interesting containers problem.
From: Shark8 <onewingedshark@gmail.com>
Injection-Date: Fri, 17 Apr 2015 13:42:55 +0000
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Xref: news.eternal-september.org comp.lang.ada:25535
Date: 2015-04-17T06:42:55-07:00
List-Id: <comp.lang.ada>

Ok, so let's say we have a type for Identifiers:

   SubType Identifier is String
   with Dynamic_Predicate =3D> Is_Valid( Identifier )
     or else raise Parse_Error with "Invalid identifier: """ & Identifier &=
 '"';
--...
   -- ACH is a rename for Ada.Characters.Handling.
   Function Is_Valid( Input: String ) return Boolean is
     ( Input'Length in Positive and then
       ACH.Is_Letter(Input(Input'First)) and then
       ACH.Is_Alphanumeric(Input(Input'Last)) and then
       (for all Ch of Input =3D> Ch =3D '_' or ACH.Is_Alphanumeric(Ch)) and=
 then
       (for all Index in Input'First..Positive'Pred(Input'Last) =3D>
          (if Input(Index) =3D '_' then Input(Index+1) /=3D '_')
       )
     );

And a few types for types:
   Type Data_Type is ( Integer, Float, Fixed );
   Type Variable_Data( Type_Value : Data_Type ) is
    case Type_Value is
     when Integer =3D> Integer_Value : Standard.Integer;
     when Float   =3D> Float_Value   : Standard.Float;
     when Fixed   =3D> Fixed_Value   : Fixed_Type; -- Defined elsewhere.
    end case;
   end record;

Now we can define a Symbol-table using the standard containers simply with =
this:

  Package Symbol_Table is
    new Ada.Containers.Indefinite_Ordered_Maps(
        Key_Type     =3D> Identifier,
        Element_Type =3D> Variable_Data,
        "<"          =3D> Ada.Strings.Less_Case_Insensitive,
        "=3D"          =3D> "=3D"
    );

And that's all well and good; however, there is one limitation here that is=
 revealed when you need nesting scopes -- there is, apparently, no way to d=
efine an Element_Type to feed to Indefinite_Ordered_Maps which is an instan=
ce of Indefinite_Ordered_Maps.Maps or an instance of Variable_Data (by way =
of a variant record; even one which is forward declared).

What would be the proper way to handle this sort of situation?