[Caml-list] Optional and mandatory labels with the same name in the same function ?

[Caml-list] Optional and mandatory labels with the same name in the same function ?

[Berke Durak]

Has the following any use, and if not, shouldn't Ocaml better give a
warning ?

# let f ?gogo ~gogo () = 33;;
val f : ?gogo:'a -> gogo:'b -> unit -> int = <fun>

(This actually happened in some LablGtk-using code with ?packing ~packing).
-- 
Berke Durak

-------------------
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] Optional and mandatory labels with the same name in the same function ?

[Jacques GARRIGUE]

From: Berke Durak <durak@liafa.jussieu.fr>

> Has the following any use, and if not, shouldn't Ocaml better give a
> warning ?
> 
> # let f ?gogo ~gogo () = 33;;
> val f : ?gogo:'a -> gogo:'b -> unit -> int = <fun>
> 

Then I suppose that
   let f x x = 33
should also cause a warning, because this has just the same
semantics.

If your question was, is it possible to call such a function, then the
answer is yes:
   f ~gogo:1 ~gogo:2 ()

Jacques Garrigue

-------------------
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] Optional and mandatory labels with the same name in the same function ?

[Richard Jones]

On Thu, Jul 15, 2004 at 11:11:53PM +0900, Jacques GARRIGUE wrote:
> Then I suppose that
>    let f x x = 33
> should also cause a warning, because this has just the same
> semantics.

It's a bit more subtle than that.  OCaml 3.07 always picks the last
argument, and ignores the others.  Thus:

# let f x x = x;;
val f : 'a -> 'b -> 'b = <fun>
# let f x x x x = x;;
val f : 'a -> 'b -> 'c -> 'd -> 'd = <fun>

But a future version might pick the first argument, or one in the
middle, or one at random.  I can't find any guarantee in the manual
that the behaviour won't change, so this might cause a bug.

Rich.

-- 
Richard Jones. http://www.annexia.org/ http://www.j-london.com/
Merjis Ltd. http://www.merjis.com/ - improving website return on investment
Perl4Caml lets you use any Perl library in your type-safe Objective
CAML programs. http://www.merjis.com/developers/perl4caml/

-------------------
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] Optional and mandatory labels with the same name in the same function ?

[Jacques GARRIGUE]

From: Richard Jones <rich@annexia.org>

> It's a bit more subtle than that.  OCaml 3.07 always picks the last
> argument, and ignores the others.  Thus:
> 
> # let f x x = x;;
> val f : 'a -> 'b -> 'b = <fun>
> # let f x x x x = x;;
> val f : 'a -> 'b -> 'c -> 'd -> 'd = <fun>
> 
> But a future version might pick the first argument, or one in the
> middle, or one at random.  I can't find any guarantee in the manual
> that the behaviour won't change, so this might cause a bug.

Well, it's not directly related to the question I was answering, but I
can reassure you. This behaviour is well defined, because
  let f x x = x
is no more than a shorthand for
  let f = fun x -> fun x -> x
which is quite natural for anybody versed in lambda-calculus.
I cannot cite you whre the manual specifies that, but this not a case
of undefined behaviour.

More generally, ocaml may leave some runtime behaviour unspecified,
but the static behaviour (type-checking) is expected to be fully
specified.

Jacques Garrigue

-------------------
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