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=-0.6 required=5.0 tests=BAYES_00,DATE_IN_PAST_24_48,
	FREEMAIL_FROM autolearn=no autolearn_force=no version=3.4.4
X-Google-Thread: a07f3367d7,69b3e9746c62ec93,start
X-Google-Attributes: gida07f3367d7,public,usenet
X-Google-NewGroupId: yes
X-Google-Language: ENGLISH,UTF8
Received: by 10.180.94.98 with SMTP id db2mr1675862wib.2.1351478331933;
        Sun, 28 Oct 2012 19:38:51 -0700 (PDT)
Path: 
 q13ni92136wii.0!nntp.google.com!feeder1.cambriumusenet.nl!feeder3.cambriumusenet.nl!feed.tweaknews.nl!81.171.88.16.MISMATCH!eweka.nl!hq-usenetpeers.eweka.nl!news.astraweb.com!border5.a.newsrouter.astraweb.com!border2.nntp.ams.giganews.com!nntp.giganews.com!newsreader4.netcologne.de!news.netcologne.de!newsfeed.straub-nv.de!news2.arglkargh.de!news.mixmin.net!aioe.org!.POSTED!not-for-mail
From: =?utf-8?Q?Yannick_Duch=C3=AAne_=28Hibou57?=
 =?utf-8?Q?=29?=
 <yannick_duchene@yahoo.fr>
Newsgroups: comp.lang.ada
Subject: Conditional holder for a type with unknown discriminant
Date: Sat, 27 Oct 2012 04:50:52 +0200
Organization: Ada @ Home
Message-ID: <op.wmtgu2dfule2fv@cardamome>
NNTP-Posting-Host: q6HlfgUjnjacTi52J0+pXw.user.speranza.aioe.org
Mime-Version: 1.0
X-Complaints-To: abuse@aioe.org
User-Agent: Opera Mail/12.02 (Linux)
X-Notice: Filtered by postfilter v. 0.8.2
Content-Type: text/plain; charset=utf-8; format=flowed; delsp=yes
Content-Transfer-Encoding: Quoted-Printable
Date: 2012-10-27T04:50:52+02:00
List-Id: <comp.lang.ada>

Hi Ada hackers (lol, does this word can comes Ada?),

Not the first time I face this question, just that it came back to me.

The matter: a conditional holder =C3=A0=E2=80=91la SML is an idiom easil=
y expressed in  =

Ada, except when the object to be wrapped also has to have an unknown  =

discriminant.

Example:

     type E_Type is private;

     type W_Type
       (Is_Null : Boolean)
        is record
           case Is_Null is
              when True =3D> null;
              when False =3D> E : E_Type;
           end case;
        end record;

     function F return W_Type;

`W_Type` is the result of a request returning an element of type `E_Type=
`,  =

which may return nothing.

Pros: avoid testing of condition dissociated from the result.
Cons: troubles when `E_Type` has unknown discriminant.

     type E_Type (<>) is private;

     type W_Type
       (Is_Null : Boolean)
        is record
           case Is_Null is
              when True =3D> null;
              when False =3D> E : E_Type;
           end case;
        end record;

     function F return W_Type;

Oops, unconstrained subtype in record definition.

You may give default values to discriminant, but no there may be no well=
  =

good enough defaults, and you may also want the client side to not creat=
e  =

an instance of `E_Type`, which would not be meaningful.

I tried to have this:

     type E_Type (<>) is private;

     type W_Type
       (Is_Null : Boolean)
        is tagged record
           case Is_Null is
              when True =3D> null;
              when False =3D> E : not null access E_Type;
           end case;
        end record;

     function F return W_Type'Class;

Then, a type derived from `W_Type` extends it with an added `E_Type`  =

aliased member=E2=80=A6 I bet you guess I would like to be able to assig=
n `E` a  =

reference to that member, but that's a no way: compiling the specificati=
on  =

is OK, but compiling any implementation fails.

Should I give up with this and use a precondition instead?

What a pity we can't have a structure referencing it's own component. Th=
at  =

would be perfectly valid from an access point of view, but we can't use =
a  =

declaration before it's terminated, so access discriminant does not help=
  =

here, as what is required, is an access to an inner element:

     X : X_Type
       (...
        E_Object =3D> (...),
        E =3D> X.E_Object'Access); -- Fails


It happens I miss we could do that (providing `X_Type` would be a limite=
d  =

type, of course).


-- =

=E2=80=9CSyntactic sugar causes cancer of the semi-colons.=E2=80=9D [1]
=E2=80=9CStructured Programming supports the law of the excluded muddle.=
=E2=80=9D [1]
[1]: Epigrams on Programming =E2=80=94 Alan J. =E2=80=94 P. Yale Univers=
ity