Re: Overloading

From: Brian Rogoff (bpr@best.com)
Date: Tue Mar 10 1998 - 22:35:29 MET


Date: Tue, 10 Mar 1998 13:35:29 -0800 (PST)
From: Brian Rogoff <bpr@best.com>
To: caml-list@inria.fr
Subject: Re: Overloading
In-Reply-To: <199803101942.OAA09854@wagga.cs.umass.edu>

On Tue, 10 Mar 1998, Adam P. Jenkins wrote:
> Brian Rogoff writes:
> > Hi Caml'ers,
> > One of the things I miss the most when I'm working in Caml is
> > overloading. There are numerous situations, such as arithmetic, linear
> > algebra (where I may want multiplication between scalars, vectors,
> > matrices, and higher order tensors), I/O (read/write/open), etc. where
> > it is IMO the "right thing". Are there any plans to add some form of
> > overloading to Caml in the future? I know that the Haskell folks plan to
> > remove the single parameter restriction in type classes and gain more
> > expressiveness in these areas, but none of the ML family of languages
> > I know of support any overloading.
> >
> > My experience with overloading in Ada is almost entirely positive.
> >
>
> I too have missed overloading in ML/Caml. However, it seems to me
> that function overloading is opposed to type inference; you can't have
> both at the same time without too many explicit casts, which destroys
> the point of function overloading.

Yes, I know the marriage of the type inference and overloading is a
difficult one.

> Are there any languages which have
> both automatic type inference and function overloading?
>
> Adam

Haskell has some limited ad-hoc polymorphism via its type classes, and
there was some work at adding a form of ad-hoc polymorphism to Caml. It is
on the Caml page, look for "extensional" polymorphism. I haven't read it
yet. Perhaps some of the Caml implementors can comment on what it
describes. Also under Francois Rouaix's page is a description of "Alcool",
which has another approach to overloading in an ML like language.

C++ (which has overloading) supports a limited form of type inference for
templated functions. There was a proposal to add something similar to Ada
in an old Tri-Ada proceedings. It is usually called "automatic instantiation"
or something similar in those communities.

In response to Frank Christoph, I am using O'Caml. What I really want in
this case is overloading, not OO. I think the reasons for this desire and
the arguments for overloading are known, so I won't repeat them unless
someone wants me to. I have seen O'Labl, but since I use Windows NT at
home I haven't used it yet (when I get Linux that will change ;-). It
looks very promising, and I also miss named parameters from Ada, but
still, it isn't quite what I want.

-- Brian



This archive was generated by hypermail 2b29 : Sun Jan 02 2000 - 11:58:14 MET