Oz and Mozart Hackers Mailing List

Re: [Fwd: building constraint systems]

From: Daniel Mahler (mahler@cyc.com)
Date: Fri Nov 14 2003 - 21:55:22 CET

Next message: Andreas Rossberg: "Re: 1.3.0 release schedule"
Previous message: Gustavo Adolfo Gutierrez Sabogal: "[Fwd: building constraint systems]"
In reply to: Gustavo Adolfo Gutierrez Sabogal: "[Fwd: building constraint systems]"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Mail actions: [ respond to this message ] [ mail a new topic ]

Why is this message is being sent out over and over?

Daniel

Gustavo Adolfo Gutierrez Sabogal writes:
>
> Return-path: <owner-oz-users@ps.uni-sb.de>
> Received: from [134.96.7.3] (helo=indyio.rz.uni-saarland.de) by
> escher.puj.edu.co with esmtp (Exim 3.34 #1 (Debian)) id 1AK1Vm-0002yo-00;
> Wed, 12 Nov 2003 15:22:47 -0500
> Received: from cs.uni-sb.de (cs.uni-sb.de [134.96.254.254]) by
> indyio.rz.uni-saarland.de (8.12.10/8.12.9) with ESMTP id hACKUTFX43514999;
> Wed, 12 Nov 2003 21:30:29 +0100 (CET)
> Received: from mail.cs.uni-sb.de (mail.cs.uni-sb.de [134.96.254.200]) by
> cs.uni-sb.de (8.12.10/2003091100) with ESMTP id hACKT7hV021983; Wed, 12 Nov
> 2003 21:30:11 +0100 (CET)
> Received: from ps.uni-sb.de (grizzly.ps.uni-sb.de [134.96.186.68]) by
> mail.cs.uni-sb.de (8.12.10/2003073000) with ESMTP id hACKRtkF016276; Wed,
> 12 Nov 2003 21:27:56 +0100 (CET)
> X-Authentication-Warning: email: Host grizzly.ps.uni-sb.de [134.96.186.68]
> claimed to be ps.uni-sb.de
> Received: (from majordomo@localhost) by ps.uni-sb.de (8.11.6/8.11.0) id
> hACKRC910911 for oz-users-internal; Wed, 12 Nov 2003 21:27:12 +0100
> X-Authentication-Warning: grizzly.ps.uni-sb.de: majordomo set sender to
> owner-oz-users@ps.uni-sb.de using -f
> X-Authentication-Warning: uni-sb.de: Host [200.24.111.12] claimed to be
> puj.edu.co
> Subject: building constraint systems
> From: Gustavo Adolfo Gutierrez Sabogal <ggutierrez@escher.puj.edu.co>
> To: users@mozart-oz.org
> Content-Type: text/plain
> Message-Id: <1068667875.31218.18.camel@lucia.puj.edu.co>
> Mime-Version: 1.0
> X-Mailer: Ximian Evolution 1.4.5
> Date: Wed, 12 Nov 2003 15:11:16 -0500
> X-DCC-NIET-Metrics: uni-sb.de 1080; Body=1 Fuz1=1 Fuz2=1
> Sender: owner-oz-users@ps.uni-sb.de
> Precedence: bulk
> Reply-To: users@mozart-oz.org
> Status:
> Content-Transfer-Encoding: 7bit
>
> Hello All,
>
> I'm trying to make some modifications to the RI module in order to
> create a constraint system a bit different. The RI module assumes that
> floating point numbers are the elements of the constraint system, in
> the constraint system i am trying to build elements are real numbers;
> since real numbers can't be represented then i can say that the
> elements of the constraint system are intervals.
>
> I am a bit familiar with the classes needed to implement a new
> constraint system, according to the documentation those are:
> - A class extending OZ_CtDefinition
> - A class extending OZ_CtWakeUp
> - A class extending OZ_CtProfile
> - A class extending OZ_Ct
> - A class extending OZ_CtVar
>
> In the documentation you can find as example the Real Interval
> constraint system and an explanation of the classes above. Also there
> are some methods whose are not implemented in the module, for example
> documentation describes the methods:
> virtual OZ_Ct * unify(OZ_Ct * r)
> virtual OZ_Boolean unify(OZ_Term rvt)
> but those methods are not implemented in the RI module.
>
> Looking at mozart sources, the class OZ_Ct, from RI class inherits,
> doesn't have any virtual methods called
> virtual OZ_Ct * unify(OZ_Ct * r)
> virtual OZ_Boolean unify(OZ_Term rvt)
> but there is a *commented* class OZ_Ct declaring those methods.
>
> What do that means?, may be those methods are not needed or deprecated?
> if that how Mozart unify two variables of the new constraint system?
>
> The RI module has float numbers as elements of the CS and OZ has the
> basic type float for which it provides operations such OZ_isFloat,
> OZ_floatToC, etc.
>
> What happen with a constraint system that do not have elements that can
> be represented by a OZ basic type?, for example in my case, OZ doesn't
> have the basic type interval, so what i need to return in the method
> OZ_Term RI::toValue(void) ?
> I try to look at the FS implementation, but it seems that the class
> FSetValue doesn't inherit from OZ_Ct. The class OZ_FSetValue has a
> method called
> int oz_isFsetValue(TaggedRef t) which is used by the method OZ_Boolean
> FSetValue::unify(OZ_Term t). Notice that FSetValue (in my opinion) is
> analog to the RI class and the first one provides a unification method
> but the second one don't. Why??
>
> Is there any document presenting the Mozart design, may be a class
> diagram or something like that? I am very interested in how FD and FS
> are implemented.
>
> Thanks in advance for your answer and excuse me for the message length.
>
> Regards,
> Gustavo Gutierrez
>
> -
> Please send submissions to users@mozart-oz.org
> and administriva mail to users-request@mozart-oz.org.
> The Mozart Oz web site is at http://www.mozart-oz.org/.
> Please send bug reports to bugs@mozart-oz.org.

-
Please send submissions to hackers@mozart-oz.org
and administriva mail to hackers-request@mozart-oz.org.
The Mozart Oz web site is at http://www.mozart-oz.org/.

Next message: Andreas Rossberg: "Re: 1.3.0 release schedule"
Previous message: Gustavo Adolfo Gutierrez Sabogal: "[Fwd: building constraint systems]"
In reply to: Gustavo Adolfo Gutierrez Sabogal: "[Fwd: building constraint systems]"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Mail actions: [ respond to this message ] [ mail a new topic ]

This archive was generated by hypermail 2b29.