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,FROM_DOMAIN_NOVOWEL,
	INVALID_DATE,MSGID_SHORT autolearn=no autolearn_force=no version=3.4.4
Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.3 4.3bsd-beta 6/6/85; site sdcrdcf.UUCP
Path: utzoo!linus!decvax!decwrl!pyramid!hplabs!sdcrdcf!steve
From: steve@sdcrdcf.UUCP (Steven Holtsberg)
Newsgroups: net.lang.ada
Subject: Arrays Declarations
Message-ID: <2635@sdcrdcf.UUCP>
Date: Tue, 18-Feb-86 17:27:46 EST
Article-I.D.: sdcrdcf.2635
Posted: Tue Feb 18 17:27:46 1986
Date-Received: Thu, 20-Feb-86 00:10:57 EST
Organization: System Development Corporation, Santa Monica
Keywords: index constraint
List-Id: <comp.lang.ada>

I have read the paragraph on index constraints (page 3-29,
second from bottom) over and over, and I still do not know
what they are talking about.

It says:

   If an index constraint follows a type mark in a subtype
   indication, then the type or subtype denoted by the
   type mark _must not already impose an index constraint_.
   The type mark must denote either an unconstrained array
   type or an access type whose designated type is such an
   array type.

As I understand it, an index constraint (page 3-27) is a list of ranges
or subtype indications.  So, if it follows a type mark, we have
something like:

    INTEGER INTEGER range 0 .. 100

    or 

    INTEGER range 0 .. 100

The first makes no sense, so I assume they mean the second, which
is an index constraint itself (as well as one that follows a type mark).

On the same page, there is an example,

type T is array(POSITIVE range MIN .. MAX) of COMPONENT;

It seems that the type mark in question is POSITIVE and
the index constraint is
range MIN .. MAX.
So, this means that POSITIVE must not already impose an index
constraint.  But this makes no sense: although predefined
(see page 3-32), POSITIVE has the constraint 1 .. INTEGER'LAST.
The only thing I can infer is that the type of subtype denoted
by the type mark cannot impose a user-defined index constraint.
I am not sure of this- is there any agreement on the net?

However, the next sentence really confuses me:

   The type mark must denote either an unconstrained array type
   or an access type whose designated type is such an array type.

POSITIVE certainly does not satisfy this condition.

Does anyone know what they are talking about?