cl-unification-devel May 2005

cl-unification-devel@common-lisp.net

3 participants
4 discussions

[cl-unification-devel] Re: question about CL-UNIFICATION
by Marco Antoniotti 27 May '05

27 May '05

Hi thanks for the comments. On May 27, 2005, at 12:25 PM, Robert P. Goldman wrote: > > I was following up on CL-UNIFICATION based on your email to the cells > list, and read over the web site docs. I was left with the question > of how to unify together two structures or objects. I see how one can > unify a structure (or object) with a template, and how that might be > used for ML-style pattern-matching. > > But what if you really want to know if two structures themselves > unify? Presumably one might want to do something like the following: > > (setf x #S(FOO A 42 S NIL D NIL)) > (setf y #S(FOO A 42 S NIL D NIL)) > > where x and y are EQUALP, but not EQ, and you'd like to do > > (unify x y) > > but this is impossible. Ooops. Not good. > Is the following correct: > > (unify x #T(FOO foo-a (foo-a y) foo-s (foo-s y) foo-d (foo-d y))) Yes. This would work. > > and, if so, is there some less cumbersome way to make this happen? > I.e., to have a unification method definition that would say "whenever > I unify together a FOO and a FOO, I want to apply this template to the > second argument? You make a very good point. Right now, the only defined methods for UNIFY with STRUCTURE-OBJECTs are unify structure-object template unify template structure-object TRT would be to write a method for unify structure-object structure-object This would be a limited method in any case, just testing for equality with EQUALP. E.g. given (setf x #S(FOO A ?x)) (setf y #S(FOO A 42)) (unify x y) would return an empty environment. I.e. the unifier would not recur into the slots. In alternative the unifier could generate two subtests (unify x #T(FOO foo-a (foo-a y) foo-s (foo-s y) foo-d (foo-d y))) (unify #T(FOO foo-a (foo-a x) foo-s (foo-s x) foo-d (foo-d x)) y) This is needed to ensure cross unifications. I know I could do away with all of this by invoking implementation dependent code getting the structure slots. But I am kind of against it. As it is the unification code is completely portable. Yet, given enough pressure, I will concede and add the the extra code. Cheers Marco PS. Do subscribe to the mailing lists. They are very low volume, but most announcements and discussions go there. -- Marco Antoniotti http://bioinformatics.nyu.edu NYU Courant Bioinformatics Group tel. +1 - 212 - 998 3488 715 Broadway 10th FL fax. +1 - 212 - 998 3484 New York, NY, 10003, U.S.A.

2 3

[cl-unification-devel] question about CL-UNIFICATION
by Robert P. Goldman 27 May '05

27 May '05

I was following up on CL-UNIFICATION based on your email to the cells list, and read over the web site docs. I was left with the question of how to unify together two structures or objects. I see how one can unify a structure (or object) with a template, and how that might be used for ML-style pattern-matching. But what if you really want to know if two structures themselves unify? Presumably one might want to do something like the following: (setf x #S(FOO A 42 S NIL D NIL)) (setf y #S(FOO A 42 S NIL D NIL)) where x and y are EQUALP, but not EQ, and you'd like to do (unify x y) but this is impossible. Is the following correct: (unify x #T(FOO foo-a (foo-a y) foo-s (foo-s y) foo-d (foo-d y))) and, if so, is there some less cumbersome way to make this happen? I.e., to have a unification method definition that would say "whenever I unify together a FOO and a FOO, I want to apply this template to the second argument? Thanks, R

1 0

[cl-unification-devel] Re: cl-unification - bugfix + extension
by Marco Antoniotti 20 May '05

20 May '05

Hi I fixed the bugs you reported. Thanks a lot for spotting them. I also clarified the disclaimer and noted that the COPYING file contains a Berkeley-style license. I have not yet added your INTERFACE template. I am quibbling on the name. I would like to call it an ACCESS template, as INTERFACE is a bit overloaded. How do you feel about it? Cheers -- Marco On May 17, 2005, at 2:52 PM, norman werner wrote: > Oops - New email-webinterface. > > > I have to apologize. > > > Norman > file in included and as attachment > > ;; Copyright (c) 2005 Norman Werner > ;; All rights reserved. > > ;; Permission is hereby granted, without written agreement and without > ;; license or royalty fees, to use, copy, modify, and distribute this > ;; software and its documentation for any purpose, provided that the > ;; above copyright notice and the following two paragraphs appear in > all > ;; copies of this software. > > ;; IN NO EVENT SHALL THE AUTHOR(S) BE LIABLE TO ANY PARTY FOR DIRECT, > ;; INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES ARISING OUT > OF > ;; THE USE OF THIS SOFTWARE AND ITS DOCUMENTATION, EVEN IF THE > AUTHOR(S), > ;; HAVE BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. > > ;; THE AUTHOR SPECIFICALLY DISCLAIMS ANY WARRANTIES, > ;; INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF > ;; MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE SOFTWARE > ;; PROVIDED HEREUNDER IS ON AN "AS IS" BASIS, AND THE AUTHOR(S) HAVE NO > ;; OBLIGATION TO PROVIDE MAINTENANCE, SUPPORT, UPDATES, ENHANCEMENTS, > OR > ;; MODIFICATIONS. > > > > ;;MOTIVATION: > ;;When something is specified by its interface we dont want to restrict > ;;the unifier to a certain implementation. > > ;;eg. A tree may be specified by the functions (get-node-value ..) and > (get-leafs ...). > ;;But a tree maybe implemented as a list, a structure or a clos-bject. > > ;;Therfore a interface-template is provided which is usefull on its > own and > ;;which may be used to implement other templates. > > > ;;IMPLEMENTATION: > ;;the following implementation tries to imitate the style of > ;;the original library. Only little testing was done. > > > ;;this is not rocket-science! > (in-package "CL.EXT.DACF.UNIFICATION") > > (defclass interface-template (type-template) ()) > > (defgeneric interface-template-p (x) > (:method ((x interface-template)) t) > (:method ((x t)) nil)) > > ;;SYNTAX: > ;;#T(interface read1 var1 read2 var2 ...) > (defmethod make-template ((kind (eql 'interface)) (spec cons)) > (declare (ignore kind)) > (make-instance 'interface-template :spec (rest spec))) > > (defun pairify (list) > "groups list into a list of pairs" > ;;I don't know how to use loop - so you may want to fix this > (cond > ((null list) ()) > ((= (length list) 1) (error "invalid template-spec")) > ( t (cons (list (first list) (second list)) (pairify (cddr > list)))))) > > (defmethod collect-template-vars ((template interface-template)) > (mapcan #'(lambda (l) (destructuring-bind (reader var) l > (assert (or (symbolp reader) > (functionp reader))) > (collect-template-vars var))) > (pairify (template-spec template)))) > > ;;;I basically copied this from the unifier-method of structure-objects > (defmethod unify ((a t) (b interface-template) &optional (env > (make-empty-environment))) > (if (template-spec b) > (loop for (reader value-template) on (template-spec b) by #'cddr > ;;FIXME/TODO: do we want to catch errors from (funcall reader a) > ?? > for mgu = (unify (funcall reader a) value-template env) > then (unify (funcall reader a) value-template mgu) > finally (return mgu)) > env)) > > (defmethod unify ((b interface-template) (a t) &optional (env > (make-empty-environment))) > (unify a b)) > > > ;;PROBLEMS: > ;;(setf e (unify '(1 2 3) #T(interface first ?x))) > ;; won't work because > ;; other method are more specific: > > ;;(defmethod unify ((a list) (b template) &optional (env > (make-empty-environment))) > ;; (declare (ignore env)) > ;; (error 'unification-failure > ;; :format-control "Cannot unify a list with a non-list > template: ~S ~S." > ;; :format-arguments (list a b))) > > ;; this applies of yourse also for other usages. > > ;; code was tested with the following shortcircuit: > ;;(defmethod unify ((a list) (b interface-template) &optional (env > (make-empty-environment))) > ;; (if (template-spec b) > ;; (loop for (reader value-template) on (template-spec b) by > #'cddr > ;; for mgu = (unify (funcall reader a) value-template env) > ;; then (unify (funcall reader a) value-template mgu) > ;; finally (return mgu)) > ;; env)) > ;; > > > > > ;; Questions / Remarks / BUGS: > > ;; Examples in the Documentation for matching of structures and > objects doesn't work ?! > ;; -> Documentation-Bug? > > ;; Description of Number-template in the doc is incorrect: > ;; <quote> > ;; The NUMBER-TEMPLATE class denotes those object that are used to > unify against a VECTOR. > ;; </quote> ?! > > ;; Is a Facility / Documentation for extending the syntax nessecary > (hash-tables, streams, > ;; closures there is a plethora of built-ins in CL) ?! > > ;; You mention in the doc that the license isn't settled yet. > ;; I consider this a critical bug. > > ;; maybe 'match' should be renamed to with-match or something else, to > make clear that > ;; new bindings are introduced?! > > ;; The following form: > ;; (unify '(foo ?x 42) '(foo baz ?x)) > ;; will unify despite your documentation and common > ;; practice. > > ;;This should fix it: > (defmethod unify ((a symbol) (b number) &optional (env > (make-empty-environment))) > (cond ((variable-any-p a) env) > ((variablep a) (var-unify a b env)) > (t (error 'unification-failure > :format-control "Cannot unify a number ~S with a > symbol ~S." > :format-arguments (list b a))))) > > (defmethod unify ((b number) (a symbol) &optional (env > (make-empty-environment))) > (unify a b env)) > > > ;;(unify '(?x ?x) '("hase" "dodo")) > ;; unifies despite common practice. > > ;; I changed the defmethods > ;; (defmethod unify ((a symbol) (b string) &optional (env > (make-empty-environment))) > ;; (cond ((variable-any-p a) env) > ;; ((variablep a) (extend-environment a b env)) > ;; (t (error 'unification-failure > ;; :format-control "Cannot unify a symbol with a string: > ~S ~S." > ;; :format-arguments (list a b))))) > > > ;; (defmethod unify ((b string) (a symbol) &optional (env > (make-empty-environment))) > ;; (cond ((variable-any-p a) env) > ;; ((variablep a) (extend-environment a b env)) > ;; (t (error 'unification-failure > ;; :format-control "Cannot unify a string with a > symbol: ~S ~S." > ;; :format-arguments (list b a))))) > > ;;to > > (defmethod unify ((a symbol) (b string) &optional (env > (make-empty-environment))) > (cond ((variable-any-p a) env) > ((variablep a) (var-unify a b env)) > (t (error 'unification-failure > :format-control "Cannot unify symbol with a string: ~S > ~S." > :format-arguments (list a b))))) > > > (defmethod unify ((b string) (a symbol) &optional (env > (make-empty-environment))) > (unify a b env)) > > __________________________________________________________ > Mit WEB.DE FreePhone mit hoechster Qualitaet ab 0 Ct./Min. > weltweit telefonieren! > http://freephone.web.de/?mc=021201<norman-contributes.lisp> -- Marco Antoniotti http://bioinformatics.nyu.edu NYU Courant Bioinformatics Group tel. +1 - 212 - 998 3488 715 Broadway 10th FL fax. +1 - 212 - 998 3484 New York, NY, 10003, U.S.A.

1 0

[cl-unification-devel] Test 3
by Marco Antoniotti 17 May '05

17 May '05

disregard -- Marco Antoniotti http://bioinformatics.nyu.edu NYU Courant Bioinformatics Group tel. +1 - 212 - 998 3488 715 Broadway 10th FL fax. +1 - 212 - 998 3484 New York, NY, 10003, U.S.A.

1 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

cl-unification-devel May 2005