[Caml-list] module types

[Caml-list] module types

[Damien]

Hi,


module type T = sig type 'a t end
module F(M: T): sig
	type a
	type t = a M.t
	(* plus a lot of values... *)
end = struct
	type a = unit
	type t = unit M.t
end

I would like to re-write it like this :

module type T = sig type 'a t end
module type O = 
sig 
	type a 
	type t 
	(* plus a lot of values... *)
end
module F(M: T): O with type t = a M.t =
struct
	type a = unit
	type t = unit M.t
end

but I can't, because "a" is unknown...
(of course, I want to keep it abstract)

any idea ?

thanks,
damien

-------------------
To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr
Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners

Re: [Caml-list] module types

[John Carr]

I also have a problem with module types.  Suppose I have

	module type T = sig type t val f : t -> string end

If I want to declare a module implementing this signature I can write

	module X : T with type t = int

but this similar declaration is not allowed

	module Y : T with type t = A | B | C

The manual implies this restriction is intentional, a result of not
allowing unnamed constructor types.  I also tried

	module Y : sig include T type t = A | B | C end

but the compiler considered this to be an illegal duplicate
definition of type t.

1. Is there a reason not to allow (A | B | C) as an anonymous
type declaration?

2. Is there a way to do what I want without declaring the type name
outside the module?  I prefer to limit the scope of type names.

-------------------
To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr
Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners

Re: [Caml-list] module types

[skaller]

On Fri, 2004-06-04 at 04:33, John Carr wrote:

> 1. Is there a reason not to allow (A | B | C) as an anonymous
> type declaration?

Standrd Sum types are generative: A | B is not a type, a binder is
required.

Use a polymorphic variant instead:

[`A | `B | `C]


-- 
John Skaller, mailto:skaller@users.sf.net
voice: 061-2-9660-0850, 
snail: PO BOX 401 Glebe NSW 2037 Australia
Checkout the Felix programming language http://felix.sf.net



-------------------
To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr
Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners

Re: [Caml-list] module types

[Julien Signoles]

> module type T = sig type 'a t end
> module F(M: T): sig
> 	type a
> 	type t = a M.t
> 	(* plus a lot of values... *)
> end = struct
> 	type a = unit
> 	type t = unit M.t
> end
>
> I would like to re-write it like this :
>
> module type T = sig type 'a t end
> module type O =
> sig
> 	type a
> 	type t
> 	(* plus a lot of values... *)
> end
> module F(M: T): O with type t = a M.t =
> struct
> 	type a = unit
> 	type t = unit M.t
> end
>
> but I can't, because "a" is unknown...
> (of course, I want to keep it abstract)
>
> any idea ?

I already encountered this problem a while ago. I solved it by using the
following technic:

module type T = sig type 'a t end
module type O = sig type a type t end
type b = unit
module F(M:T): O with type t = b M.t and type a = b =
struct
  type a = unit
  type t = unit M.t
end

In the signature of the above structure, [b] should be abstract in order
to keep [a] abstract. Sadly, this technic introduces a new type definition
but I don't see a better way to solve this problem.

Hope this helps,
Julien Signoles
-- 
mailto:Julien.Signoles@lri.fr ; http://www.lri.fr/~signoles
"In theory, practice and theory are the same,
but in practice they are different" (Larry McVoy)

-------------------
To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr
Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners