comp.lang.ada
 help / color / mirror / Atom feed
From: Tucker Taft <stt@averstar.com>
Subject: Re: access-to-constant dispatching params
Date: 2000/05/09
Date: 2000-05-09T17:07:27+00:00	[thread overview]
Message-ID: <391845CE.256C3D93@averstar.com> (raw)
In-Reply-To: 8f8flg$cj7$1@nnrp1.deja.com

don_harrison@my-deja.com wrote:
> 
>   Tucker Taft <stt@averstar.com> wrote:
> > m_cubed@my-deja.com wrote:
> > >
> > > Ada provides access-to-variable dispatching
> > > parameters.
> 
> [..]
> 
> > > What it doesn't seem to provide is
> > > access-to-constant dispatching parameters.
> >
> > This is one thing being considered for Ada 0X.
> >
> > However, as someone else suggested, plain old "IN" parameters
> > pretty much do the job and are preferable in certain ways.
> 
> In what ways do you consider in-parameters preferable?

IN parameters would be preferable to access-to-constant for most
non-dispatching cases, because they are more flexible, allowing
passing of non-aliased objects, the result of calling a function,
an aggregate or a literal, etc.  For dispatching cases, the two
are quite similar, and choosing between them depends on whether
most callers are dealing with access values or objects.
> 
> Don Harrison

-- 
-Tucker Taft   stt@averstar.com   http://www.averstar.com/~stt/
Technical Director, Commercial Division, AverStar (formerly Intermetrics)
(http://www.averstar.com/services/IT_consulting.html)  Burlington, MA  USA




      reply	other threads:[~2000-05-09  0:00 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2000-05-08  0:00 access-to-constant dispatching params m_cubed
2000-05-08  0:00 ` Andy
2000-05-09  0:00   ` don_harrison
2000-05-08  0:00 ` Tucker Taft
2000-05-09  0:00   ` don_harrison
2000-05-09  0:00     ` Tucker Taft [this message]
replies disabled

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox