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=-1.3 required=5.0 tests=BAYES_00,INVALID_MSGID
	autolearn=no autolearn_force=no version=3.4.4
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 103376,b50bc6538a649497
X-Google-Attributes: gid103376,public
From: Robert A Duff <bobduff@world.std.com>
Subject: Re: constants v variables (was Re: if statements)
Date: 2000/11/09
Message-ID: <wccd7g5f52s.fsf@world.std.com>#1/1
X-Deja-AN: 691680016
Sender: bobduff@world.std.com (Robert A Duff)
References: <3A02CED4.520C2768@brighton.ac.uk> <3A078B6F.D34B024B@erols.com>
 <8ua3m1$bru$1@nnrp1.deja.com> <3A09A39F.2822C01B@cepsz.unizar.es>
 <8udce8$1qi$1@nnrp1.deja.com> <3a0a6f6c$1@pull.gecm.com>
Organization: The World Public Access UNIX, Brookline, MA
Newsgroups: comp.lang.ada
Date: 2000-11-09T00:00:00+00:00
List-Id: <comp.lang.ada>

"Martin Dowie" <martin.dowie@gecm.com> writes:

> Thanks for highlighting this one - I've been bleeting on about this locally
> (and even in c.l.a. once!) for ages to collegues.

Me, too.  IMHO, constant should be the *default*, and you should have to
say "var" or something to declare a variable; then people would be less
likely to make this mistake.  In fact, that *is* the way parameters work
("in" is the default).  It seems inconsistent that object declarations
work differently.

>...I've noticed that apart
> from the analysis-of-the-problem meaning (i.e. it shows that the value has
> been determined to be the same thoughout its scope) there are also
> efficiency benefits, at least on one of our target compilers (in
> non-optimising mode, untried when optimising).

Well, it seems kind of pointless to worry about efficiency if you have
optimization turned off.

> My question is - can these efficiencies be expected from a compiler, by the
> way the ARM defines variables v constants, or is this just because of the
> particular compiler we were using. Will any 'decent' compiler spot when a
> 'variable' should actually have been declared as a constant?

A good compiler will probably be able to spot this if the variable is
local (if you turn on the highest optimization level!).  For globals
(library-package-level), the compiler probably needs the "constant" to
tell it the thing is constant.

- Bob