read_term and anonymous variables

[Uwe Lesta]

Hello all,

if i ask
:- assert(a(1,2)).
:- assert(a(3,4)).
:-a(_,_).

The top level returns true.


if i ask

:- read_term(T, [variables(V), variable_names(VN)]), call(T).

and enter a(_,_). I get false which seems to be logical if i see

6 ?- read_term(T, [variables(V), variable_names(VN)]).
|: a(_,_).
T = a(_G588, _G588),
V = [_G588],
VN = ['_'=_G588].


What is the right behavior?

How can I read a term to get the top-level behavior and the variablenames?


Thanks in advance

--

Kind regards

Uwe Lesta




_______________________________________________
SWI-Prolog mailing list
SWI-Prolog@...
https://mailbox.iai.uni-bonn.de/mailman/listinfo.cgi/swi-prolog

[Paulo Moura]

On 2009/10/26, at 08:32, Uwe Lesta wrote:

> Hello all,
>
> if i ask
> :- assert(

> :- assert(a(3,4)).
> :-a(_,_).
>
> The top level returns true.
>
>
> if i ask
>
> :- read_term(T, [variables(V), variable_names(VN)]), call(T).
>
> and enter a(_,_). I get false which seems to be logical if i see
>
> 6 ?- read_term(T, [variables(V), variable_names(VN)]).
> |: a(_,_).
> T = a(_G588, _G588),
> V = [_G588],
> VN = ['_'=_G588].
>
>
> What is the right behavior?

No. You found a bug. Every occurrence of an anonymous variable works  
as a distinct variable. Follows the above queries using GNU Prolog:

| ?- read_term(T, [variables(V), variable_names(VN)]), call(T).
a(_, _).

T = a(3,4)
V = [3,4]
VN = [] ? ;

T = a(1,2)
V = [1,2]
VN = []

(1 ms) yes

| ?- read_term(T, [variables(V), variable_names(VN)]).
a(_, _).

T = a(A,B)
V = [A,B]
VN = []

yes

Cheers,

Paulo


-----------------------------------------------------------------
Paulo Jorge Lopes de Moura, PhD
Assistant Professor
Dep. of Computer Science, University of Beira Interior
6201-001 Covilhã, Portugal

Office 3.18  Ext. 3276
Phone: [phone] Fax: [phone]
Email: <mailto:pmoura@...>

Home page: <http://www.di.ubi.pt/~pmoura>
Research:  <http://logtalk.org/> Blog: <http://blog.logtalk.org/>
-----------------------------------------------------------------







_______________________________________________
SWI-Prolog mailing list
SWI-Prolog@...
https://mailbox.iai.uni-bonn.de/mailman/listinfo.cgi/swi-prolog

[Jan Wielemaker-3]

On Monday 26 October 2009 09:54:29 am Paulo Moura wrote:

> On 2009/10/26, at 08:32, Uwe Lesta wrote:
> > Hello all,
> >
> > if i ask
> >
> > :- assert(
> > :
> > :- assert(a(3,4)).
> > :-a(_,_).
> >
> > The top level returns true.
> >
> >
> > if i ask
> >
> > :- read_term(T, [variables(V), variable_names(VN)]), call(T).
> >
> > and enter a(_,_). I get false which seems to be logical if i see
> >
> > 6 ?- read_term(T, [variables(V), variable_names(VN)]).
> >
> > |: a(_,_).
> >
> > T = a(_G588, _G588),
> > V = [_G588],
> > VN = ['_'=_G588].
> >
> >
> > What is the right behavior?
>
> No. You found a bug. Every occurrence of an anonymous variable works
>
> as a distinct variable. Follows the above queries using GNU Prolog:
> | ?- read_term(T, [variables(V), variable_names(VN)]), call(T).
>
> a(_, _).
>
> T = a(3,4)
> V = [3,4]
> VN = [] ? ;
>
> T = a(1,2)
> V = [1,2]
> VN = []
>
> (1 ms) yes

Yip.  Fixed.  Using variables and variable_names in combination is
appearently not often use.  Patch is in development GIT.  Should come
in 5.9.0.

Of course, you can also just use the variable_names option and use
term_variables on the term read to find all variables ...  I wonder
who believed the option variables to read_term makes sense ...

     --- Jan

_______________________________________________
SWI-Prolog mailing list
SWI-Prolog@...
https://mailbox.iai.uni-bonn.de/mailman/listinfo.cgi/swi-prolog

[Uwe Lesta]

Jan Wielemaker schrieb:

<snip>

> Yip.  Fixed.  Using variables and variable_names in combination is
> appearently not often use.  Patch is in development GIT.  Should come
> in 5.9.0.
>
> Of course, you can also just use the variable_names option and use
> term_variables on the term read to find all variables ...  I wonder
> who believed the option variables to read_term makes sense ...

Hello Jan,

Thank you and Paulo for clarification.

I'll have changed the call of read_term/2 to

        read_term(T, [variable_names(VN)])

and build the list of variables in CSharp
( leaving a comment about term_variables/2 ;-)


IMO the option variables/1 of read_tern/2 is superfluous.
A link to term_variables/2 in the comment about the option
variable_names/1 might be useful.

Thanks a lot.

--

Kind regards

Uwe Lesta



_______________________________________________
SWI-Prolog mailing list
SWI-Prolog@...
https://mailbox.iai.uni-bonn.de/mailman/listinfo.cgi/swi-prolog