Mailing list for all users of the OCaml language and system.
 help / color / mirror / Atom feed
From: Markus Mottl <mottl@miss.wu-wien.ac.at>
To: Jerome.Vouillon@inria.fr (Jerome Vouillon)
Cc: caml-list@inria.fr (OCAML)
Subject: Re: Problem binding type parameters in modules + subtyping and inheritance
Date: Mon, 18 Jan 1999 21:37:20 +0100 (MET)	[thread overview]
Message-ID: <199901182037.VAA04696@miss.wu-wien.ac.at> (raw)
In-Reply-To: <19990118201616.45587@pauillac.inria.fr> from "Jerome Vouillon" at Jan 18, 99 08:16:16 pm

Hello,

> Another solution is to turn "empty" into a function :
>    module type FOO1 = sig
>      type 'a foo
>      val empty : unit -> 'a foo
>    end;;
>    module Bar : FOO1 = struct
>      type 'a foo = 'a list Lazy.t
>      let empty () = lazy []
>    end;;

Would certainly also work. But I guess it is better to stay with the
functor - modules parameterized with a new value are not so common as
function invocations. Thus, I won't have to change this much code porting
Okasaki's sources.

> -----------------------------------------------------------

> Actually, it is still possible to make the coercion when the type of
> self only occur in covariant position (in particular, only on the
> right sides of arrows) :
>    class c = object (self)
>      method m = self
>    end;;
>    class d = object (self)
>      inherit c
>      method n = 1
>    end;;
>    let x = (new d :> c);;
> 
> There is a difficulty with "binary methods", where self type occurs in
> contravariant position :
>    class c = object (self : 'a)
>      method x = 1
>      method compare (o : 'a) = self#x = o#x
>    end;;
>    class d = object
>      inherit c
>      method y = 1
>    end;;
>    let x = (new d :> c);;     (* Fails *)

Unfortunately, it's the binary methods I need...

> However, you can coerce to a common supertype of c and d which has no
> method "compare" :
>    class type c' = object
>      method x : int
>    end;;
>    let l = [(new c :> c'); (new d :> c')];; (* Succeeds *)

That's the problem:
Actually the only method I would need is "compare" ;-)

Regards,
Markus

-- 
Markus Mottl, mottl@miss.wu-wien.ac.at, http://miss.wu-wien.ac.at/~mottl




  reply	other threads:[~1999-01-19 16:55 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
1999-01-16 12:21 Problem binding type parameters in modules Markus Mottl
1999-01-18 14:59 ` Sylvain BOULM'E
1999-01-18 17:07   ` Problem binding type parameters in modules + subtyping and inheritance Markus Mottl
1999-01-18 19:16     ` Jerome Vouillon
1999-01-18 20:37       ` Markus Mottl [this message]
1999-01-19  8:50 ` Problem binding type parameters in modules Hendrik Tews

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=199901182037.VAA04696@miss.wu-wien.ac.at \
    --to=mottl@miss.wu-wien.ac.at \
    --cc=Jerome.Vouillon@inria.fr \
    --cc=caml-list@inria.fr \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox