[racket] [racket-users]Is Racklog suitable to learn Prolog? A simple example just fails.

[WarGrey Gyoudmon Ju]

This example is rewritten from the datalog version.

#lang racket

(require racklog)

(define %edge

  (%rel ()

    [('a 'b)]

    [('b 'c)]

    [('c 'd)]

    [('d 'a)]))

(define %path

  (%rel (X Y)

    [(X Y) (%edge X Y)]

    [(X Y) (%edge X 'Z)

             (%path 'Z Y)]))

(%find-all (X Y) (%path X Y))

It only gives four results.

#lang racklog

edge(a, b). 

edge(b, c).

edge(c, d).

edge(d, a).

path(X, Y) :- edge(X, Y).

path(X, Y) :- edge(X, Z), path(Z, Y).

path(X, Y)?

and this version will not terminate.

Thanks in advance.

[Matthias Felleisen]

If you want to learn Prolog, why not use a free Prolog implementation?

If you want to learn about relational programming, use miniKanren, which is available of cKanren. Then you can also get to constraints. Maintainer cc-ed.

-- Matthias

> ____________________
> Racket Users list:
> http://lists.racket-lang.org/users


____________________
Racket Users list:
http://lists.racket-lang.org/users

[WarGrey Gyoudmon Ju]

Sorry, my example is not correct.

But this one still does not work:

#lang at-exp racket

(require racklog)

(define %edge

  (%rel ()

    [('a 'b)]

    [('b 'c)]

    [('c 'd)]

    [('d 'a)]))

(define %path

  (%rel (X Y Z)

    [(X Y) (%edge X Y)]

    [(X Y) (%edge X Z) (%path Z Y)]))

(let answer ([r (%which (X Y) (%path X Y))])

  (printf "~a --> ~a~n" (cdar r) (cdadr r))

  (answer (%more)))

This version also will not terminate.

On Thu, Jul 3, 2014 at 9:04 PM, Matthias Felleisen <matt...@ccs.neu.edu> wrote:

If you want to learn Prolog, why not use a free Prolog implementation?

Simply speaking, Modern Prolog seems to cut the illumos-based SunOS off, I cannot build it without headache and they does not answer me about this issue.

So I wander if Racklog is a legacy package?

[Matthias Felleisen]

Yes, I corrected your program and ran into this bug. 

Try to use Kanren instead of Racklog. It isn't completely a legacy package but I doubt we can get a quick fix here. 

-- Matthias

[WarGrey Gyoudmon Ju]

Okay, Thank you.

[Scott Moore]

I'm not sure what behavior you were expecting for this program. Under standard prolog semantics (goal-directed) this program will yield infinitely many solutions, so this may not be a bug in racklog but rather in the program.

That said, Matthias is probably right that using a currently maintained package is probably a good idea.

Cheers,
Scott

[Matthias Felleisen]

Racklog should have generated a stream of solutions here. That's my problem with it. -- Matthias

[WarGrey Gyoudmon Ju]

Under standard prolog semantics Racklog is right.

The semantics between prolog and datalog is different.

Okay, thank you for pointing me.

[Jay McCarthy]

Matthias,

Racklog did generate a stream of solutions, but WarGrey's program
iterates through them all by calling (answer (%more)).

I also recommend kanren, though.

Jay

--
Jay McCarthy
http://jeapostrophe.github.io

"Wherefore, be not weary in well-doing,
for ye are laying the foundation of a great work.
And out of small things proceedeth that which is great."
- D&C 64:33

[Matthias Felleisen]

Mea cupla.