From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.5-pre1 (2020-06-20) on
	ip-172-31-74-118.ec2.internal
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 required=3.0 tests=BAYES_00,MSGID_SHORT
	autolearn=no autolearn_force=no version=3.4.5-pre1
Date: 12 Dec 91 15:36:34 GMT
From: elroy.jpl.nasa.gov!sdd.hp.com!zaphod.mps.ohio-state.edu!unix.cis.pitt.edu
!dsinc!gvlf3.gvl.unisys.com!email!parkhill@ames.arc.nasa.gov  (parkhill)
Subject: Re: 'SIZE attribute of a type
Message-ID: <5901@email.sp.unisys.com>
List-Id: <comp.lang.ada>

Mr Mark Biggar writes:

>I don't thing anyone would implement this as an automatic feature of an
>Ada compiler.  But, it is perfectly reasonable for the rep-clause

> for b'size use 1;

> to cause a compiler to do just what you want, while not having the
> rep-clause would make b'size something like 8, 16 or 32.

> b'size is the actual storage used for the type not the minimum possible.


If I am correctly understanding the interpretation of Type_Name'Size
the attribute should return the minimum number of bits objects of this
type might require.  So when a compiler usually uses 16 bits to represent
objects of type boolean, but in the case of pragma pack on a boolean array
only uses 1 bit, Boolean'Size must return 1.  In the case of of compiler
that can at some time understand and interpret baised types and produce
a very small range, than the compiler should generate and return that 
small number of bits for type_name'Size.  Remember, even if you do not use
pragma Pack in your program Boolean'Size still returns 1.  It doesn't
matter if you use the feature at all, it only depends upon whether the
compiler is capable of packing boolean arrays. 

I believe Object_Name'Size returns the actual storage used for an object.
There is little correlation between Object_Name'Size and Type_Name'Size.

If I am wrong anyone feel free to post a correction. I am clearly NOT an
Ada expert.

Robert Parkhill