Re: Singleton variable in branch

On 16/12/2013, at 15:58, FERNANDO SAENZ PEREZ <[email protected]> wrote:

> Hello,
> Given the following system session in SWI-Prolog 6.6.1 (Windows, stable
> version):
> 
> %%%%%%%%%%%%%%%%%%%%%%
> 
> Welcome to SWI-Prolog (Multi-threaded, 32 bits, Version 6.6.1)
> Copyright (c) 1990-2013 University of Amsterdam, VU Amsterdam
> SWI-Prolog comes with ABSOLUTELY NO WARRANTY. This is free software,
> and you are welcome to redistribute it under certain conditions.
> Please visit http://www.swi-prolog.org for details.
> 
> For help, use ?- help(Topic). or ?- apropos(Word).
> 
> 1 ?- [p].
> Warning: c:/fernan/research/bddeduc/des/des3.3.3/p.dl:1:
>        Singleton variable in branch: B
> % p.dl compiled 0.00 sec, 2 clauses
> true.
> 
> 2 ?- listing.
> 
> :- thread_local thread_message_hook/3.
> :- dynamic thread_message_hook/3.
> :- volatile thread_message_hook/3.
> 
> 
> p(A) :-
>        (   functor(A, _, B)
>        ->  (   foo(B)
>            ->  foo(B)
>            ;   foo(B)
>            )
>        ;   foo(B)
>        ).
> true.
> 
> %%%%%%%%%%%%%%%%%%%%%%
> 
> I do not understand the warning. I think that B is used in all branches,
> isn't it?

That's not the issue. If functor/3 fails, B is a singleton variable. If you take into account a failure of functor/3, your code becomes equivalent to:

p(A) :-
       (   fail
       ;   foo(B)
       ).

> In addition, it is said that 2 clauses have been compiled. Is it so? The
> source file includes the very same clause as listed.

My only explanation involves parallel universes :-) But maybe Jan can shed some light here.

Cheers,

Paulo

-----------------------------------------------------------------
Paulo Moura
Logtalk developer

Email: <mailto:[email protected]>
Web:   <http://logtalk.org/>
-----------------------------------------------------------------