From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from mail4-relais-sop.national.inria.fr (mail4-relais-sop.national.inria.fr [192.134.164.105])
	by walapai.inria.fr (8.13.6/8.13.6) with ESMTP id p83AcJ1w025119
	for <caml-list@sympa-roc.inria.fr>; Sat, 3 Sep 2011 12:38:19 +0200
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: AioCAGADYk5V2gB5hWdsb2JhbABCqQABAQEKCwsbJoFGAQEFJyAkDhALRhQoIYgIpwiNfQ6GXASkPQ
X-IronPort-AV: E=Sophos;i="4.68,324,1312149600"; 
   d="scan'208";a="107571289"
Received: from emailfrontal2.citycable.ch ([85.218.0.121])
  by mail4-smtp-sop.national.inria.fr with SMTP; 03 Sep 2011 12:37:53 +0200
X-Alinto-smtpauth-localdomain: Yes
Received: from seldon (unknown [85.218.93.239])
	(Authenticated sender: guillaume.yziquel@citycable.ch)
	by emailfrontal2.citycable.ch (Postfix) with ESMTPA id AA71520C158;
	Sat,  3 Sep 2011 12:37:46 +0200 (CEST)
Received: from yziquel by seldon with local (Exim 4.72)
	(envelope-from <guillaume.yziquel@citycable.ch>)
	id 1Qznan-0002OC-QX; Sat, 03 Sep 2011 12:36:53 +0200
Date: Sat, 3 Sep 2011 12:36:53 +0200
From: Guillaume Yziquel <guillaume.yziquel@citycable.ch>
To: Goswin von Brederlow <goswin-v-b@web.de>
Cc: caml-list@inria.fr
Message-ID: <20110903103653.GX15100@localhost>
References: <87ty8uc5ph.fsf@frosties.localnet>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
In-Reply-To: <87ty8uc5ph.fsf@frosties.localnet>
User-Agent: Mutt/1.5.20 (2009-06-14)
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by walapai.inria.fr id p83AcJ1w025119
Subject: Re: [Caml-list] Odd failure to infer types

Le Saturday 03 Sep 2011 à 11:53:30 (+0200), Goswin von Brederlow a écrit :
> Hi,
> 
> I'm implementing a solver for the game Atomix. If you don't know it then
> don't worry. It isn't relevant.
> 
> I split things up into submodules and now one of the submodules does not
> infere the right types:
> 
> File "Atomix.ml", line 168, characters 11-876:
> Error: The type of this module,
>        sig
>          type dir = NORTH | SOUTH | WEST | EAST
>          val max_moves : int
>          val cache : (string, unit) Hashtbl.t
>          val states :
>            ('_a list * (char * int * int) array * string) list array
>          val string_of_dir : dir -> string
>          val print :
>            (int * int * dir) list * (char * int * int) array * string -> unit
>          val num_states : int
>        end, contains type variables that cannot be generalized
> 
> I believe this is wrong. In S.num_states the call to "print state"
> fixates the type for state and the "states.(d) <- state::states.(d);"
> should then fixate the missing '_a in the type for states.
> 
> Anyone know why?

It also seems quite wrong to me. You should perhaps file a bug into
Mantis if no typing expert answers.

Did you try adding type annotations one at a time near the call to print
and the states.(d) assignment in your anonymous List.fold-ing function?
To check precisely what the type inferencer gets right and what it gets
wrong?

I'd be curious to know whether annotating state in "states.(d) <-
state::states.(d)" solves your problem. Since it's here that the '_a in
the type of states should be fixated.

-- 
     Guillaume Yziquel