Ciao-users June 2003

ciao-users@cliplab.org

4 participants
4 discussions

ITCLS 2003 Call for Papers and Participation
by Jesus Correas Fernandez 17 Jun '03

17 Jun '03

[Apologies if you receive this more than once] Call for Papers and Participation COLOGNET Workshop on Implementation Technology for Computational Logic Systems (ITCLS 2003) Satellite Workshop of FME 2003 Symposium 9 September 2003, Pisa, Italy http://clip.dia.fi.upm.es/Conferences/ITCLS-2003/ CoLogNet, the European Network of Excellence in Computational Logic is organizing the Second Workshop on Implementation Technology for Computational Logic Systems (ITCLS) as a satellite workshop of Formal Methods 2003, to take place in Pisa. The previous edition was held in September 2002 in Madrid, affiliated with Logic-based Program Development and Transformation (LOPSTR'02), and co-located with the International Static Analysis Symposium (SAS'02) and the APPIA-GULP-PRODE Joint Conference on Declarative Programming (AGP'02). The workshop is a forum open to both academia and industry and whose aim is to present and discuss work (possibly in progress) related to the topics mentioned below. The ITCLS workshop builds on the tradition of the series of workshops on Implementations of Logic Programming Systems held in the context of COMPULOG NET, the former Network of Excellence in Computational Logic, but opens up its scope to a broader use of logic and formal methods within computational systems. ITCLS 2003 focuses on implementation techniques from the different areas addressed by CoLogNet. The aim is to bring together researchers from fields such as automated reasoning, formal methods and theorem proving, natural language processing, constraint logic programming, inductive logic programming, non-monotonic logic programming, etc. in order to share common problems and solutions regarding implementation and usability issues which arise in formal and logic-based computational systems. Topics of interest for this workshop include, but are not limited to: * Implementation techniques for all aspects of the implementation of systems related to computational logic and formal systems, including: + Formal Methods and its relationship with Computational Logic, + Compiling Specifications, + Theorem Proving, + Non-monotonic Reasoning, Higher Order, Linear Logics, + Logic Programming, + Constraint Programming, and + Natural Language Processing * Environments for computational logic and formal systems, including: + User-interfaces and User-needs, + Software Development, + Debugging and Visualization Techniques, and + Interfaces to Other (possibly Computational Logic) Systems PROGRAMME The workshop will take place during the afternoon and evening of September 9th. The programme will consist of an invited talk (to be announced later) and the presentation of the accepted papers. SUBMISSIONS AND DEADLINES Authors are invited to contribute to the workshop by submitting extended abstracts (6 pages maximum) or full papers (12 pages maximum) on the workshop themes. Limits are strict, and the llncs format (http://www.springer.de/comp/lncs/authors.html) must be used. Submissions are to be sent in PS or PDF format to the organizing committee by emailing them to: colognet-ws-submission_at_clip.dia.fi.upm.es (substitute _at_ by the @ symbol to send email). Submissions originally prepared with LaTeX are highly encouraged, as this will help in preparing the final proceedings, which will be available at the workshop in printed form, as well as electronically. The table below highlights the important dates: Submission deadline: July 7 Notification to authors: July 28 Camera-ready copies: August 15 Workshop: September 9 ORGANIZING COMMITTEE Jesus Correas, UPM, Madrid, Spain Manuel Hermenegildo, UPM, Madrid, Spain Manuel Carro, UPM, Madrid, Spain Christian Schulte, KTH, Stockholm, Sweden Please address any question directly related to the workshop to colognet-ws_at_clip.dia.fi.upm.es (substitute _at_ by @). ============================================================================== Message: Address: Action: help majordomo(a)clip.dia.fi.upm.es Info. on useful commands subscribe ciao-users-request(a)clip.dia.fi.upm.es Subscribe to this list unsubscribe ciao-users-request(a)clip.dia.fi.upm.es Unsubscribe from this list <whatever> ciao-users(a)clip.dia.fi.upm.es Send message to list ----------------------------------------------------------------------------- Archived messages: http://www.clip.dia.fi.upm.es/Mail/ciao-users/ -----------------------------------------------------------------------------

1 0

Crash during ciao_query_begin_term(), from C
by Jeff Donner 11 Jun '03

11 Jun '03

Hi, What's the right way to query Ciao from C? Attached are two files which show what I'm trying (to get at some particular predicates) but, Ciao crashes: Ciao-Prolog 1.9 #44: Thu May 22 22:21:28 MDT 2003 ?- use_module(library(show_preds)). yes ?- show_preds(x). orig_var=2, ciao_is_variable? yes goal_var=4, ciao_is_variable? yes ciao_is_structure(goal): yes goal is: some_pred(_var_5) About to begin query goal=3 {ERROR: segmentation violation} { Execution aborted } ?- (To re-create the above, put the attached two files into your ciao library directory and follow the session as shown.) - Also, I have no success getting Ciao to accept a 0-arity foreign predicate. The following all fail with a parse error (so says assert anyway): :- true pred show_preds() :: foreign. :- true pred show_preds :: foreign. :- true pred show_preds/0 :: foreign. (where currently I have .. show_preds(in(X)) :: .. is that a real limit or is the syntax of the above wrong? Thanks for any help, Jeff :- module(show_preds, [show_preds/1], [assertions, basicmodes, regtypes, foreign_interface ]). %% This doesn't seem to like /0 predicates; I have tried: %% show_preds :: .. show_preds/0 :: .. show_preds() :: %% So the argument is unused. :- true pred show_preds(in(X)) :: any_term + foreign. :- use_foreign_source(show_preds_c). :- impl_defined([show_preds/1]). #include <stdio.h> #include "ciao_prolog.h" static void print_ciao_term(ciao_term term); void show_preds(ciao_term unused_term) { ciao_term orig_var = ciao_var(); ciao_term goal = ciao_structure("some_pred", 1, orig_var); ciao_term goal_var = ciao_structure_arg(goal, 1); ciao_query* query = 0; printf("orig_var=%d, ciao_is_variable? %s\n", orig_var, ciao_is_variable(orig_var) ? "yes" : "no"); printf("goal_var=%d, ciao_is_variable? %s\n", goal_var, ciao_is_variable(goal_var) ? "yes" : "no"); printf("ciao_is_structure(goal): %s\n", ciao_is_structure(goal) ? "yes" : "no"); printf("goal is:\n"); print_ciao_term(goal); printf("\n"); printf("About to begin query goal=%d\n", goal); query = ciao_query_begin_term(goal); printf("Began ok...\n"); printf("First ciao_query_next query = %p\n", query); while (ciao_query_next(query)) { printf("print_ciao_term(goal)\n"); print_ciao_term(goal); printf("\n... 'nother ciao_query_next\n"); } /* ciao_bool ciao_query_ok(ciao_query *query); ciao_bool ciao_query_next(ciao_query *query); void ciao_query_end(ciao_query *query); */ ciao_query_end(query); } static void print_ciao_term(ciao_term term) { if (ciao_is_atom(term)) { printf("%s", ciao_atom_name(term)); } else if (ciao_is_variable(term)) { /* most likely not the canonical form of the var.. */ printf("_var_%d", term); } else if (ciao_is_structure(term)) { int i; int a; a = ciao_structure_arity(term); printf("%s(", ciao_structure_name(term)); for (i = 1; i <= a; i++) { print_ciao_term(ciao_structure_arg(term, i)); if (i != a) printf(", "); } printf(")"); } else if (ciao_is_list(term)) { printf("["); print_ciao_term(term); printf("]"); } else if (ciao_is_empty_list(term)) { printf("[]"); } else if (ciao_is_integer(term)) { printf("%d", ciao_to_integer(term)); } else if (ciao_is_number(term)) { printf("%f", ciao_to_float(term)); } else { printf("<unknown/>"); } }

1 0

Help with 'sentence' translation
by Jeff Donner 08 Jun '03

08 Jun '03

Hi, I'm doing some 'sentence translation' for a module which parses code related to Prolog. But, some of its syntax is hard to get Ciao to see as being a 'sentence'; I'll be grateful for any suggestions. It's this: compute 1 { not fail }. I can get { not fail }. and compute x. recognized separately by making 'not' and 'compute' be operators, but, no non-infix prolog operator can have two arguments and that's what it seems 'compute' needs. So, how to get such a construct recognized by the sentence / clause / term translation machinery? %% Note that this obvious-seeming version doesn't work. braces_sentence_trans(compute(N){X},[],_M) :- message(['full compute ']). -- If it's too far from prolog syntax for the sentence translation machinery is it possible to wrap it somehow, or to make the sentence translation more flexible? Or is there a more powerful technique? Thanks for any help; Jeff Donner (Attached are simple illustrative files; braces_tr is the translation module, braces is the package, braces_test contains the items which all parse except for the last. To build it put braces and braces_tr in your ciao library directory and make them .po files (emacs/C-c o), do 'braces_tr' before 'braces'). :- load_compilation_module(library(braces_tr)). :- op(900, fy, 'compute'). :- op(900, fy, 'not'). :- add_sentence_trans(braces_sentence_trans/3). :- use_package(library(braces)). { foo(X), bar(Y) } :- sausage(X). { not foo }. compute 9. compute 1 { not fail }. :- module(braces_tr, [braces_sentence_trans/3]). braces_sentence_trans(0,[],Module) :- !, message(['sentence sign-on. Module ',Module,':']). braces_sentence_trans(end_of_file,[],_M) :- !, message(['EOF']), fail. %% handles singles and lists! braces_sentence_trans(({X} :- Body),[],_M) :- message(['braces ']). braces_sentence_trans(not(S),[],_M) :- message(['not ']). braces_sentence_trans(compute(N),[],_M) :- message(['compute ']). %% Note that this doesn't work. braces_sentence_trans((compute(N){X}),[],_M) :- message(['full compute ']).

2 1

Re: Help with 'sentence' translation
by Richard A. O'Keefe 05 Jun '03

05 Jun '03

Jeff Donner <jdonner(a)cs.nmsu.edu> wrote: I'm doing some 'sentence translation' for a module which parses code related to Prolog. ... compute 1 { not fail }. Why not take the public domain Prolog tokeniser and the public domain Prolog parser and modify the parser? Prolog has prefix (unary) operators; postfix (unary) operators; infix (binary) operators. It does not have any 'fxx' operators, although if I recall correctly NU Prolog does support that. It would be a comparatively small matter to modify the public domain parser to support fxx operators. A Prolog extension I thought about several years ago is to allow compound terms to be operators, so you would write :- op(900, fy, [compute/0, compute/1]). ... compute 9 => compute(9) compute(9) {not fail} => compute(9,{}(not(fail))) (compute 9) {not fail} => compute(9,{}(not(fail))) compute 9 {not fail} => ERROR This would be quite a small change to the public domain Prolog parser, rather easier and less troublesome than introducing binary prefix operators. Simplest of all, why not change the syntax to something Prolog _can_ parse? (A) Make compute infix binary instead of prefix binary: 1 compute { not fail }. It has always been legal in Prolog for an atom to be both a prefix and an infix operator. (B) Include a dummy operator in the surface syntax which you strip out before processing: compute 1 times { not fail }. :- load_compilation_module(library(braces_tr)). :- op(900, fy, compute). :- op(900, fy, not). + :- op(850, xfx, times). compute 9 => compute(9). compute 1 times { not fail } => compute(times(1,{}(not(fail)))). ============================================================================== Message: Address: Action: help majordomo(a)clip.dia.fi.upm.es Info. on useful commands subscribe ciao-users-request(a)clip.dia.fi.upm.es Subscribe to this list unsubscribe ciao-users-request(a)clip.dia.fi.upm.es Unsubscribe from this list <whatever> ciao-users(a)clip.dia.fi.upm.es Send message to list ----------------------------------------------------------------------------- Archived messages: http://www.clip.dia.fi.upm.es/Mail/ciao-users/ -----------------------------------------------------------------------------

1 0

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

Ciao-users June 2003