From: Virgile Prevosto <virgile.prevosto@lip6.fr>
To: Nobuyuki Tomizawa <n-tomizawa@mua.biglobe.ne.jp>
Cc: caml-list@inria.fr
Subject: Re: [Caml-list] Question on typing of class/object and optional argument.
Date: Tue, 17 Dec 2002 12:37:21 +0100 [thread overview]
Message-ID: <20021217123721.43010c7d.virgile.prevosto@lip6.fr> (raw)
In-Reply-To: <20021216.011851.730550325.nobuyuki@mua.biglobe.ne.jp>
Hello,
Nobuyuki Tomizawa a écrit:
> ------------------------------------------------------------
> class foo s = object
> val str : string = s
> method to_string ?(opt = "" ) () = opt ^ s
> end;;
>
> let l = [new foo "a"; new foo "b"; new foo "c" ];;
>
> List.iter (fun e -> print_endline (e#to_string ())) l;;
>
> File "test.ml", line 8, characters 52-53:
> This expression has type foo list but is here used with type
> < to_string : unit -> string > list
> Type foo = < to_string : ?opt:string -> unit -> string >
> is not compatible with type < to_string : unit -> string >
> ------------------------------------------------------------
This is quite normal: since the type of e is not constrained,
ocaml has inferred the most general one from the body of the
function: e must be an object with a method 'to_string' of type
unit -> string . The problem comes from the fact that optional arguments
and type inference do not work very well together
(cf http://pauillac.inria.fr/ocaml/htmlman/manual006.html section 4.1.2).
As suggested in the manual, the best solution might be to add a type annotation
in the function above:
List.iter (fun (e:foo) -> print_endline (e#to_string ())) l;;
or
List.iter (fun (e:#foo) -> print_endline (e#to_string ())) l;;
if you intend to use subclasses of foo
>
> In contrast, I did not get the above message if I rewrote the program
> into module style like:
> module Bar : BAR = struct
> type t = string
> let create t = t
> let to_string ?(opt = "") t = opt ^ t
> end;;
>
> let m = [Bar.create "a"; Bar.create "b"; Bar.create "c" ];;
>
> List.iter (fun e -> print_endline (Bar.to_string e)) m;;
Here, Bar.to_string has a perfectly defined type
(namely ?opt:string -> Bar.t -> string), and optionnal arguments
are treated inside the function. e itself is
not involved in the process: it should only be of type Bar.t, and m is indeed a
Bar.t list, so that everything works fine. In the object version, e is an object
whose method to_string must have a certain type in which optionnal arguments are
not necessarily taken into account.
I'm afraid I'm not very clear here, but I'm not very familiar with labels,
sorry...
--
E tutto per oggi, a la prossima volta
Virgile
-------------------
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
next prev parent reply other threads:[~2002-12-17 11:37 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2002-12-15 16:18 Nobuyuki Tomizawa
2002-12-17 11:37 ` Virgile Prevosto [this message]
2002-12-17 13:22 ` Olivier Andrieu
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=20021217123721.43010c7d.virgile.prevosto@lip6.fr \
--to=virgile.prevosto@lip6.fr \
--cc=caml-list@inria.fr \
--cc=n-tomizawa@mua.biglobe.ne.jp \
/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