October 2004 - PPL-devel - BUGSENG Mailman

ppl/interfaces/Prolog/Ciao ppl_ciao.cc
by Abramo Bagnara 21 Oct '04

21 Oct '04

CVSROOT: /cvs/ppl Module name: ppl Changes by: abramo(a)cs.unipr.it 2004-10-21 11:59:01 Modified files: interfaces/Prolog/Ciao: ppl_ciao.cc Log message: Generalized string conversion. Patches: http://www.cs.unipr.it/cgi-bin/cvsweb.cgi/ppl/interfaces/Prolog/Ciao/ppl_ci…

1 0

Giant bug in XSB 2.6?
by Roberto Bagnara 21 Oct '04

21 Oct '04

Still on the theme "nigthmares about Prolog arithmetics", can anyone reproduce this (rather incredible) behavior? $ xsb [xsb_configuration loaded] [sysinitrc loaded] [packaging loaded] XSB Version 2.6 (Duff) of June 24, 2003 [i686-pc-linux-gnu; mode: optimal; engine: slg-wam; gc: indirection; scheduling: local] | ?- [user]. [Compiling user] write_intexpr(Term) :- M is Term, write(M). [user compiled, cpu time used: 0.0300 seconds] [user loaded] yes | ?- write_intexpr(1+1). 1 + 1 yes | ?- Term = 1+1, M is Term, write(M). 2 Term = 1 + 1 M = 2 yes | ?- If this is confirmed as a bug, does anyone know if there is a workaround that is general enough? While we are at it, do you know if XSB is still actively maintained? I sent a message to xsb-development back in April and never got a reply. All the best, Roberto -- Prof. Roberto Bagnara Computer Science Group Department of Mathematics, University of Parma, Italy http://www.cs.unipr.it/~bagnara/ mailto:bagnara@cs.unipr.it

1 0

[Fwd: Re: Any Prolog language lawyer out there?]
by Roberto Bagnara 21 Oct '04

21 Oct '04

-------- Original Message -------- Subject: Re: Any Prolog language lawyer out there? Date: Thu, 21 Oct 2004 11:08:04 +0200 From: Bart Demoen <bmd(a)cs.kuleuven.ac.be> Organization: KULeuvenNet Newsgroups: comp.lang.prolog References: <41750912.8040102(a)cs.unipr.it> <1098196164.417498(a)seven.kulnet.kuleuven.ac.be> <slrncnaat8.jkn.jan(a)ct.xs4all.nl> <87acuijutl.fsf(a)gondolin.bb.bawue.de> <1098259920.942884(a)seven.kulnet.kuleuven.ac.be> <1098274154.598785(a)seven.kulnet.kuleuven.ac.be> <77bbf36a.0410210015.1db2726c(a)posting.google.com> Mats Carlsson wrote: >>In the standard "directly" means something like "with no space in between". > > > That must be wishful thinking on your part. There is no such > definition of "directly" in the document. I should indeed have been more careful. In principle, everything in the standard has a well-agreed upon meaning. In an attempt to ensure that, page 2 of the standard says in the 3th paragraph of section 3: # Words and phrases not defined in this glossary are assumed # to have the meaning given in ISO 2382-15; if they do not # appear in ISO 2382-15, then they are assumed to have # their usual meaning. "directly" is not defined in the ISO Prolog glossary (rest of section 3) So we should check ISO 2382-15 ... I couldn't get hold of ISO 2382-15 (it is too expensive to acquire just to set this dispute) and if anybody has access to it and can check whether it defines "directly", please let us know. If it isn't in ISO 2382-15, we must take the usual meaning of "directly". I think (apparently together with GNU Prolog and the makers of http://www.sju.edu/~jhodgson/prolog_proj/framed_spec.html) that in the context as it is used in the explanation of negative numbers, it means "no spaces [or other stuff] in between" You might think differently. Cheers Bart Demoen -- Prof. Roberto Bagnara Computer Science Group Department of Mathematics, University of Parma, Italy http://www.cs.unipr.it/~bagnara/ mailto:bagnara@cs.unipr.it

1 0

[Fwd: Re: Any Prolog language lawyer out there?]
by Roberto Bagnara 21 Oct '04

21 Oct '04

-------- Original Message -------- Subject: Re: Any Prolog language lawyer out there? Date: 21 Oct 2004 01:15:45 -0700 From: matsc(a)sics.se (Mats Carlsson) Organization: http://groups.google.com Newsgroups: comp.lang.prolog References: <41750912.8040102(a)cs.unipr.it> <1098196164.417498(a)seven.kulnet.kuleuven.ac.be> <slrncnaat8.jkn.jan(a)ct.xs4all.nl> <87acuijutl.fsf(a)gondolin.bb.bawue.de> <1098259920.942884(a)seven.kulnet.kuleuven.ac.be> <1098274154.598785(a)seven.kulnet.kuleuven.ac.be> I made a mistake in my previous posting, so here goes again. Bart Demoen <bmd(a)cs.kuleuven.ac.be> wrote in message news:<1098274154.598785(a)seven.kulnet.kuleuven.ac.be>... > I wrote: > > > Goal : integer(- 4) > > Specification result : Must fail > > > > > > > > Goal : integer(-4) > > Specification result : Must succeed > > > > > > So it seems that the space between the - and the 4 is important. > > > > I do not have the final copy of the standard, and although I do remember > > that there has been discussion about it, I do not remember the final > > outcome, > > so can anybody check that's indeed what the standard says ? > > I got hold of a standard ... > > It says at the end of section that Jens was quoting from: > > "A term which is the name - followed directly by a numeric constant, > denotes the corresponding negative constant" > > In the standard "directly" means something like "with no space in between". That must be wishful thinking on your part. There is no such definition of "directly" in the document. I too have the standard document, from which I quote the relevant grammar rules: 6.3.1.2 Negative Numbers term = name, integer ; term = name, float number ; /* where the name must be '-' */ 6.4 Tokens name = [ layout text sequence ], name token ; integer = [ layout text sequence ], integer token ; float number = [ layout text sequence ], float number token ; 6.4.1 Layout text layout text sequence = layout text, { layout text } ; layout text = layout char | comment ; So clearly, there can be space, even comments, between the '-' and the digits. BTW, the same applies to [/*this atom*/] and {/*this atom*/}. > That decides the issue I think: - 4 should not be read as a negative integer; To the contrary, it should. --Mats -- Prof. Roberto Bagnara Computer Science Group Department of Mathematics, University of Parma, Italy http://www.cs.unipr.it/~bagnara/ mailto:bagnara@cs.unipr.it

1 0

[Fwd: Re: Any Prolog language lawyer out there?]
by Roberto Bagnara 21 Oct '04

21 Oct '04

-------- Original Message -------- Subject: Re: Any Prolog language lawyer out there? Date: 21 Oct 2004 01:12:22 -0700 From: matsc(a)sics.se (Mats Carlsson) Organization: http://groups.google.com Newsgroups: comp.lang.prolog References: <41750912.8040102(a)cs.unipr.it> <1098196164.417498(a)seven.kulnet.kuleuven.ac.be> <slrncnaat8.jkn.jan(a)ct.xs4all.nl> <87acuijutl.fsf(a)gondolin.bb.bawue.de> <1098259920.942884(a)seven.kulnet.kuleuven.ac.be> <1098274154.598785(a)seven.kulnet.kuleuven.ac.be> Bart Demoen <bmd(a)cs.kuleuven.ac.be> wrote in message news:<1098274154.598785(a)seven.kulnet.kuleuven.ac.be>... > I wrote: > > > Goal : integer(- 4) > > Specification result : Must fail > > > > > > > > Goal : integer(-4) > > Specification result : Must succeed > > > > > > So it seems that the space between the - and the 4 is important. > > > > I do not have the final copy of the standard, and although I do remember > > that there has been discussion about it, I do not remember the final > > outcome, > > so can anybody check that's indeed what the standard says ? > > I got hold of a standard ... > > It says at the end of section that Jens was quoting from: > > "A term which is the name - followed directly by a numeric constant, > denotes the corresponding negative constant" > > In the standard "directly" means something like "with no space in between". That must be wishful thinking on your part. There is no such definition of "directly" in the document. I too have the standard document, from which I quote the relevant grammar rules: 6.3.1.2 Negative Numbers term = name, integer ; term = name, float number ; /* where the name must be '-' */ 6.4 Tokens name = [ layout text sequence ], name token ; 6.4.1 Layout text layout text sequence = layout text, { layout text } ; layout text = layout char | comment ; 6.4.2 Names name token = letter digit token | graphic token | quoted token | semicolon token | cut token ; letter digit token = small letter char, { alphanumeric char } ; So clearly, there can be space, even comments, between the '-' and the digits. BTW, the same applies to [/*this atom*/] and {/*this atom*/}. > That decides the issue I think: - 4 should not be read as a negative integer; To the contrary, it should. --Mats -- Prof. Roberto Bagnara Computer Science Group Department of Mathematics, University of Parma, Italy http://www.cs.unipr.it/~bagnara/ mailto:bagnara@cs.unipr.it

1 0

[Fwd: Re: Any Prolog language lawyer out there?]
by Roberto Bagnara 21 Oct '04

21 Oct '04

-------- Original Message -------- Subject: Re: Any Prolog language lawyer out there? Date: Wed, 20 Oct 2004 14:09:14 +0200 From: Bart Demoen <bmd(a)cs.kuleuven.ac.be> Organization: KULeuvenNet Newsgroups: comp.lang.prolog References: <41750912.8040102(a)cs.unipr.it> <1098196164.417498(a)seven.kulnet.kuleuven.ac.be> <slrncnaat8.jkn.jan(a)ct.xs4all.nl> <87acuijutl.fsf(a)gondolin.bb.bawue.de> <1098259920.942884(a)seven.kulnet.kuleuven.ac.be> I wrote: > Goal : integer(- 4) > Specification result : Must fail > > > > Goal : integer(-4) > Specification result : Must succeed > > > So it seems that the space between the - and the 4 is important. > > I do not have the final copy of the standard, and although I do remember > that there has been discussion about it, I do not remember the final > outcome, > so can anybody check that's indeed what the standard says ? I got hold of a standard ... It says at the end of section that Jens was quoting from: "A term which is the name - followed directly by a numeric constant, denotes the cirresponding negative constant" In the standard "directly" means something like "with no space in between". That decides the issue I think: - 4 should not be read as a negative integer; and -4 should be read as a negative integer. Cheers Bart Demoen -- Prof. Roberto Bagnara Computer Science Group Department of Mathematics, University of Parma, Italy http://www.cs.unipr.it/~bagnara/ mailto:bagnara@cs.unipr.it

1 0

ppl/src checked_int.inlines.hh
by Abramo Bagnara 20 Oct '04

20 Oct '04

CVSROOT: /cvs/ppl Module name: ppl Changes by: abramo(a)cs.unipr.it 2004-10-20 20:00:13 Modified files: src : checked_int.inlines.hh Log message: Use signed explicitly. Patches: http://www.cs.unipr.it/cgi-bin/cvsweb.cgi/ppl/src/checked_int.inlines.hh.di…

1 0

ppl/demos/ppl_lcdd Makefile.am expected_a
by Roberto Bagnara 20 Oct '04

20 Oct '04

CVSROOT: /cvs/ppl Module name: ppl Changes by: roberto(a)cs.unipr.it 2004-10-20 19:36:02 Modified files: demos/ppl_lcdd : Makefile.am Added files: demos/ppl_lcdd : expected_a Log message: Exclude examples that are too heavy when assertions are enabled. Patches: http://www.cs.unipr.it/cgi-bin/cvsweb.cgi/ppl/demos/ppl_lcdd/expected_a?cvs… http://www.cs.unipr.it/cgi-bin/cvsweb.cgi/ppl/demos/ppl_lcdd/Makefile.am.di…

1 0

ppl configure.ac
by Roberto Bagnara 20 Oct '04

20 Oct '04

CVSROOT: /cvs/ppl Module name: ppl Changes by: roberto(a)cs.unipr.it 2004-10-20 18:46:08 Modified files: . : configure.ac Log message: Introduced new Automake conditional ASSERTIONS_ENABLED. Patches: http://www.cs.unipr.it/cgi-bin/cvsweb.cgi/ppl/configure.ac.diff?cvsroot=ppl…

1 0

ppl TODO
by Roberto Bagnara 20 Oct '04

20 Oct '04

CVSROOT: /cvs/ppl Module name: ppl Changes by: roberto(a)cs.unipr.it 2004-10-20 16:17:21 Modified files: . : TODO Log message: We now allow the specification of alternative coefficient implementations. Patches: http://www.cs.unipr.it/cgi-bin/cvsweb.cgi/ppl/TODO.diff?cvsroot=ppl&r1=1.10…

1 0