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-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 103376,1116ece181be1aea
X-Google-Attributes: gid103376,public
X-Google-ArrivalTime: 2003-10-09 01:09:46 PST
Path: 
 archiver1.google.com!news1.google.com!newsfeed.stanford.edu!news.uchicago.edu!yellow.newsread.com!netaxs.com!newsread.com!newsfeed.stueberl.de!fu-berlin.de!uni-berlin.de!tar-alcarin.cbb-automation.DE!not-for-mail
From: Dmitry A. Kazakov <mailbox@dmitry-kazakov.de>
Newsgroups: comp.lang.ada
Subject: Re: Counter-proposal for variable arrays
Date: Thu, 09 Oct 2003 10:19:14 +0200
Message-ID: <725aovsqh10dv5p5nh1e7k2qvjjcqustsa@4ax.com>
References: <3F7B1076.8060106@comcast.net>
 <5mknnv4u96qqudrt4bd8n4t1cljp2fjlp8@4ax.com>
 <CZOdnWbTwIxOruGiU-KYgw@gbronline.com>
 <blklum$dala6$1@ID-77047.news.uni-berlin.de>
 <1065215180.95094@master.nyc.kbcfp.com>
 <blm0av$dkoad$3@ID-77047.news.uni-berlin.de>
 <akOfb.23883$kD3.11661@nwrdny03.gnilink.net>
 <19p2ovk1jh4krn2h5cql44p37ovf6va99i@4ax.com>
 <FyOgb.36403$541.32150@nwrdny02.gnilink.net>
 <4pe7ov0gsepk8p8cq88ropvq4sookdhatk@4ax.com>
 <1065635640.270857@master.nyc.kbcfp.com>
NNTP-Posting-Host: tar-alcarin.cbb-automation.de (212.79.194.111)
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Trace: news.uni-berlin.de 1065686985 18848241 212.79.194.111 (16 [77047])
X-Newsreader: Forte Agent 1.8/32.548
Xref: archiver1.google.com comp.lang.ada:520
Date: 2003-10-09T10:19:14+02:00
List-Id: <comp.lang.ada>

On Wed, 08 Oct 2003 13:54:00 -0400, Hyman Rosen <hyrosen@mail.com>
wrote:

>Dmitry A. Kazakov wrote:
>> Only when you create a concrete instance of the type, an object,
> > you have to constrain it, i.e. to say which types are allowed there
> > or how many characters it will hold. This is the whole idea.
>
>So now you will have arrays whose stride as well as bounds
>is a runtime property.

Exactly.

>We should equally allow records with
>unconstrained field types, and let the constraints get set
>upon creation.

It is already possible. Though there should be a work done for the
tagged types. My another proposal is to treat the type tag of a
class-wide as a discrimant. This would allow a lot of very interesting
things to do. To start with, one will be able to assign class-wide
objects, intrigued?

>Then the field offsets themselves could be
>runtime properties too.

It is already so. Consider:

type Tricky (I, J : Positive) is record
   X : String (1..I);
   Y : String (1..J);
end record;

The offset to Y is variable and depends on the value of I.

>Then we could have classwide objects
>in records.

Right. See above, the tags has to be exposed as discriminants.

>It's all doable. I guess the question is whether you get
>enough bang for the buck to make it worthwhile to specify
>and implement. My inclination is no, but anyone who really
>wants it could have a go at the GNAT sources and make a
>proof of concept.

For good or bad, neither Ada is C++ nor I have Stroustrup's influence
on the language design. There should be some consensus reached. And of
course, any change undertaken have to be carefully reviewed. So far I
observe little interest in further evolving Ada's ADT concept. It is a
pity, because in my view Ada has the best potential to become the
first of the next generation languages. (:-()

---
Regards,
Dmitry Kazakov
www.dmitry-kazakov.de