I wrote a test program with bindings (facts) between atoms and numbers.
bind(a, 3).
bind(b, 4).
bind(c, 5).
As part of a toy interpreter, I want to be able to perform additions on these atoms using Prolog's native arithmetic operators. For instance, I want to be able to run this query:
% val(X) is the value bound to X
?- X is val(a) + val(b).
X = 7.
However, I'm struggling to find a way to allow this addition. My first approach would have been this one:
% val(X, Y): Y is the value bound to X
val(X, Y) :- bind(X, Y).
% Make val an arithmetic function
:- arithmetic_function(val/1).
However, arithmetic_function/1 is no longer part of Prolog (or at least SWI-Prolog says it's deprecated), so I can't use it. Then I believed the best solution would be to overload the + operator to take this into account:
% val(X, Y): Y is the value bound to X
val(val(X), Y) :- bind(X, Y).
% Overload the + operator
+(val(_X, XVal), val(_Y, YVal)) :- XVal + YVal.
But here I've got my syntax all messed up because I don't really know how to overload a native arithmetic operation. When I type in the sample query from before, SWI-Prolog says ERROR: Arithmetic: ``val(a)' is not a function.
Would you have hints about a possible solution or a better approach or something I missed?
From the docs, I tought you should use function_expansion/3.
But I'm unable to get it to work, instead, goal_expansion could do, but isn't very attractive... for instance, if you save the following definitions in a file bind.pl (just to say)
:- module(bind, [test/0]).
:- dynamic bind/2.
bind(a, 3).
bind(b, 4).
bind(c, 5).
% :- multifile user:goal_expansion/2.
user:goal_expansion(val(X), Y) :- bind(X, Y).
user:goal_expansion(X is Y, X is Z) :- expand_goal(Y, Z).
user:goal_expansion(X + Y, U + V) :- expand_goal(X, U), expand_goal(Y, V).
test :-
X is val(a) + val(b), writeln(X).
and consult it, you can run your test:
?- test.
7
edit
after Paulo suggestion, here is an enhanced solution, that should work for every binary expression.
user:goal_expansion(X is Y, X is Z) :- expr_bind(Y, Z).
expr_bind(val(A), V) :- !, bind(A, V).
expr_bind(X, Y) :-
X =.. [F, L, R], % get operator F and Left,Right expressions
expr_bind(L, S), % bind Left expression
expr_bind(R, T), % bind Right expression
Y =.. [F, S, T]. % pack bound expressions back with same operator
expr_bind(X, X). % oops, I forgot... this clause allows numbers and variables
having defined user as target module for goal_expansion, it works on the CLI:
?- R is val(a)*val(b)-val(c).
R = 7.
edit
now, let's generalize to some other arithmetic operators, using the same skeleton expr_bind uses for binary expressions:
user:goal_expansion(X, Y) :-
X =.. [F,L,R], memberchk(F, [is, =<, <, =:=, >, >=]),
expr_bind(L, S),
expr_bind(R, T),
Y =.. [F, S, T].
and unary operators (I cannot recall no one apart minus, so I show a simpler way than (=..)/2):
...
expr_bind(-X, -Y) :- expr_bind(X, Y).
expr_bind(X, X).
Now we get
?- -val(a)*2 < val(b)-val(c).
true.
One way to do it is using Logtalk parametric objects (Logtalk runs on SWI-Prolog and 11 other Prolog systems; this makes this solution highly portable). The idea is to define each arithmetic operation as a parametric object that understands an eval/1 message. First we define a protocol that will be implemented by the objects representing the arithmetic operations:
:- protocol(eval).
:- public(eval/1).
:- end_protocol.
The basic parametric object understands val/1 and contains the bind/2 table:
:- object(val(_X_), implements(eval)).
eval(X) :-
bind(_X_, X).
bind(a, 3).
bind(b, 4).
bind(c, 5).
:- end_object.
I exemplify here only the implementation for arithmetic addition:
:- object(_X_ + _Y_, implements(eval)).
eval(Result) :-
_X_::eval(X), _Y_::eval(Y),
Result is X + Y.
:- end_object.
Sample call (assuming the entities above are saved in an eval.lgt file):
% swilgt
...
?- {eval}.
% [ /Users/pmoura/Desktop/eval.lgt loaded ]
% (0 warnings)
true.
?- (val(a) + val(b))::eval(R).
R = 7.
This can be an interesting solution if you plan to implement more functionality other than expression evaluation. E.g. a similar solution but for symbolic differentiation of arithmetic expressions can be found at:
https://github.com/LogtalkDotOrg/logtalk3/tree/master/examples/symdiff
This solution will also work in the case of runtime generated expressions (term-expansion based solutions usually only work at source file compile time and at the top-level).
If you're only interested in expression evaluation, Capelli's solution is more compact and retains is/2 for evaluation. It can also be made more portable if necessary using Logtalk's portable term-expansion mechanism (but note the caveat in the previous paragraph).
This is perhaps not exactly what I was looking for, but I had an idea:
compute(val(X) + val(Y), Out) :-
bind(X, XVal),
bind(Y, YVal),
Out is XVal + YVal.
Now I can run the following query:
?- compute(val(a) + val(c), Out).
Out = 8.
Now I need to define compute for every arithmetic operation I'm interested in, then get my interpreter to run expressions through it.
Related
I'm trying to trigger backtracking on a goal but in a dynamic way, if it's possible. To better exemplify my issue let's say we have the following PROLOG code:
num(1).
num(2).
num(3).
num(4).
num(5).
Then I head to SWI-Prolog and call: num(X). This triggers backtracking looking for all solutions, by typing ; .
What I would like is to remove those facts (num(1),num(2), etc) and replace that code with something thata generates those facts dynamically. Is there any way in which I can achieve this? Someting of the sorts,maybe?
num(X):- for X in 1..5
that yields the same solutions as the code above?
As far as I know, the findall predicate returns a list, which is not what I'm looking for. I would like to backtrack through all answers and look through them using ; in the console.
Yes there is, and you were already very close!
:- use_module(library(clpfd)).
num(X) :-
X in 1..5.
?- num(X).
X in 1..5.
?- num(X), X #>3.
X in 4..5.
?- num(X), labeling([], [X]).
X = 1
; X = 2
; X = 3
; X = 4
; X = 5.
SWI-Prolog has the (non-ISO) predicate between/3 for that:
num(X) :- between(1, 5, X).
You can implement the predicate (for other Prologs and for further tweaking) like this:
between2(A, A, A) :- !. % green cut
between2(A, B, A) :- A < B.
between2(A, B, C) :-
A < B,
A1 is A + 1,
between2(A1, B, C).
The signature for both between/3 and between2/3 is (+From,+To,?X). It means that the From and To must be bound and X can be either bound or not. Also note that From and To must be integers such that From <= To. (Oh, and these integers must be written using Arabic numerals with an optional plus or minus sign before. And using ASCII. Is something non-obvious still missed? And the integers must not be too large or too small, although SWI-Prolog is usually compiled with unbounded integer support, so both between(1, 100000000000000000000000000000000000000000000, X) and between2(1, 100000000000000000000000000000000000000000000, X) usually work.)
I'm new to Prolog and would like to define a simple predicate which calculates the result depending on which function I choose to use in the arithmetic expression.
So, this was my idea:
operation(X,Y, Op, Result):-
Result is X Op Y.
Now, I was expecting this from Prolog:
operation(3,4,'+', X).
X = 7.
But as you can probably guess, Prolog cannot identify Op as an arithmetic operation.
Does anyone have an idea how this is possible?
I could not find anything on the internet yet, even though it is rather basic, I think.
Thanks in advance!
Although the answers by Tudor and gokhans deliver the wanted result, I think there is a more elegant solution.
Portable solution
The following will work in most Prolog implementations:
operation(X, Y, Operator, Result):-
Goal =.. [Operator, X, Y],
Result is Goal.
Extended but SWI-Prolog specific solution
SWI-Prolog allows the definition of custom arithmetic functions. The following code extends the above for use with such user-defined functions coming from other modules:
:- meta_predicate(operation(+,+,2,-)).
operation(X, Y, Module:Operator, Result):-
Goal =.. [Operator, X, Y],
Module:(Result is Goal).
Notice that support for user-defined functions is deprecated in SWI-Prolog and does not work in other Prologs that do not have this feature.
Usage examples
Some examples of using these implementations of operation/4:
?- operation(1, 2, mod, X).
X = 1.
?- operation(1, 2, //, X).
X = 0.
?- operation(1, 2, /, X).
X = 0.5.
?- operation(1, 2, -, X).
X = -1.
?- operation(1, 2, +, X).
X = 3.
You need to tell Prolog if Op is '+' then sum X and Y. You can do that as following
operation(X,Y, Op, Result) :-
Op = '+',
Result is X + Y
;
Op = '-',
Result is X - Y.
You can increase number of operations.
I have created a program, list(X,Y) to check whether all the elements in list Y are smaller than X.
The codes are as follows.
list(X,[]).
list(X,[Y|Z]):-X>Y,list(X,Z).
It works fine when I type list(3,[1,2]). However, if I type list(3,Y) in order to find lists which only contain elements smaller than 3, there is an error.
?- list(3,[1,2]).
true .
?- list(3,Y).
Y = [] ;
ERROR: >/2: Arguments are not sufficiently instantiated
I have read some posts which got the same error, but I still don't understand which part of my codes goes wrong.
Here comes a similar example found from internet.
greater(X,Y,Z) returns the part Z of Y that is greater than X.
greater(X,[],[]).
greater(X,[H|Y],[H|Z]) :- H>X, greater(X,Y,Z).
greater(X,[H|Y],Z) :- H=<X, greater(X,Y,Z).
?- greater(2,[1,2,3], Y).
Y = [3].
The question is, what is the difference between the codes of greater(X,Y,Z) and list(X,Y) so that there is no error when calling greater(2,[1,2,3], Y)..
Thanks for any help provided.
Since - judging from your comment - you seem to be reasoning over integers: That's a textbook example for using finite domain constraints, which are available in almost all modern Prolog implementations and generalize integer arithmetic so that you can use it in all directions.
Your code works exactly as expected with, among others, B-Prolog, GNU Prolog, SICStus, SWI and YAP if you just use the finite domain constraint (#>)/2 instead of the lower-level arithmetic primitive (>)/2:
:- use_module(library(clpfd)).
list(X, []).
list(X, [Y|Z]):- X#>Y, list(X,Z).
Constraints allow you to use this predicate, which you can also express with maplist/2 as in the queries below, in all directions:
?- maplist(#>(3), [1,2]).
true.
?- maplist(#>(X), [1,2]).
X in 3..sup.
?- maplist(#>(3), [A,B]).
A in inf..2,
B in inf..2.
?- maplist(#>(X), [Y,Z]).
Z#=<X+ -1,
Y#=<X+ -1.
Even the most general query, where none of the arguments is instantiated, gives useful answers:
?- maplist(#>(X), Ls).
Ls = [] ;
Ls = [_G1022],
_G1022#=<X+ -1 ;
Ls = [_G1187, _G1190],
_G1190#=<X+ -1,
_G1187#=<X+ -1 ;
etc.
EDIT: Also the example you now added can be made much more general with finite domain constraints:
:- use_module(library(clpfd)).
greater(_, [], []).
greater(X, [H|Y], [H|Z]) :- H #> X, greater(X, Y, Z).
greater(X, [H|Y], Z) :- H #=< X, greater(X, Y, Z).
You can now use it in all directions, for example:
?- greater(X, [E], Ls).
Ls = [E],
X#=<E+ -1 ;
Ls = [],
X#>=E.
This is not possible with the original version, whose author may not have been aware of constraints.
I have come across an unfamiliar bit of Prolog syntax in Lee Naish's paper Higher-order logic programming in Prolog. Here is the first code sample from the paper:
% insertion sort (simple version)
isort([], []).
isort(A.As, Bs) :-
isort(As, Bs1),
isort(A, Bs1, Bs).
% insert number into sorted list
insert(N, [], [N]).
insert(N, H.L, N.H.L) :-
N =< H.
insert(N, H.LO, H.L) :-
N > H,
insert(N, LO, L).
My confusion is with A.As in isort(A.As, Bs) :-. From the context, it appears to be an alternate cons syntax for lists, the equivalent of isort([A|As], Bs) :-.
As well N.H.L appears to be a more convenient way to say [N|[H|L]].
But SWI Prolog won't accept this unusual syntax (unless I'm doing something wrong).
Does anyone recognize it? is my hypothesis correct? Which Prolog interpreter accepts that as valid syntax?
The dot operator was used for lists in the very first Prolog system of 1972, written in Algol-W, sometimes called Prolog 0. It is inspired by similar notation in LISP systems. The following exemple is from the paper The birth of Prolog by Alain Colmerauer and Philippe Roussel – the very creators of Prolog.
+ELEMENT(*X, *X.*Y).
+ELEMENT(*X, *Y.*Z) -ELEMENT(*X, *Z).
At that time, [] used to be NIL.
The next Prolog version, written in Fortran by Battani & Meloni, used cases to distinguish atoms and variables. Then DECsystem 10 Prolog introduced the square bracket notation replacing nil and X.Xs with [] and [X,..Xs] which in later versions of DECsystem 10 received [X|Xs] as an alternative. In ISO Prolog, there is only [X|Xs], .(X,Xs), and as canonical syntax '.'(X,Xs).
Please note that the dot has many different rôles in ISO Prolog. It serves already as
end token when followed by a % or a layout character like SPACE, NEWLINE, TAB.
decimal point in a floating point number, like 3.14159
graphic token char forming graphic tokens as =..
So if you are now declaring . as an infix operator, you have to be very careful. Both with what you write and what Prolog systems will read. A single additional space can change the meaning of a term. Consider two lists of numbers in both notations:
[1,2.3,4]. [5].
1 .2.3.4.[]. 5.[].
Please note that you have to add a space after 1. In this context, an additional white space in front of a number may change the meaning of your terms. Like so:
[1|2.3]. [4]. 5. [].
1 .2.3. 4.[]. 5. [].
Here is another example which might be even more convincing:
[1,-2].
1.(-2).[].
Negative numbers require round brackets within dot-lists.
Today, there is only YAP and XSB left that still offer infix . by default – and they do it differently. And XSB does not even recognize above dot syntax: you need round brackets around some of the nonnegative numbers.
You wrote that N.H.L appears to be a more convenient way to say [N|[H|L]]. There is a simple rule-of-thumb to simplify such expressions in ISO Prolog: Whenever you see within a list the tokens | and [ immediately after each other, you can replace them by , (and remove the corresponding ] on the right side). So you can now write: [N,H|L] which does not look that bad.
You can use that rule also in the other direction. If we have a list [1,2,3,4,5] we can use | as a "razor blade" like so: [1,2,3|[4,5]].
Another remark, since you are reading Naish's paper: In the meantime, it is well understood that only call/N is needed! And ISO Prolog supports call/1, call/2 up to call/8.
Yes, you are right, the dot it's the list cons infix operator. It's actually required by ISO Prolog standard, but usually hidden. I found (and used) that syntax some time ago:
:- module(eog, []).
:- op(103, xfy, (.)).
% where $ARGS appears as argument, replace the call ($ARGS) with a VAR
% the calle goes before caller, binding the VAR (added as last ARG)
funcs(X, (V, Y)) :-
nonvar(X),
X =.. W.As,
% identify meta arguments
( predicate_property(X, meta_predicate M)
% explicitly exclude to handle test(dcg)
% I'd like to handle this case in general way...
, M \= phrase(2, ?, ?)
-> M =.. W.Ms
; true
),
seek_call(As, Ms, Bs, V),
Y =.. W.Bs.
% look for first $ usage
seek_call([], [], _Bs, _V) :-
!, fail.
seek_call(A.As, M.Ms, A.Bs, V) :-
M #>= 0, M #=< 9, % skip meta arguments
!, seek_call(As, Ms, Bs, V).
seek_call(A.As, _, B.As, V) :-
nonvar(A),
A = $(F),
F =.. Fp.FAs,
( current_arithmetic_function(F) % inline arith
-> V = (PH is F)
; append(FAs, [PH], FBs),
V =.. Fp.FBs
),
!, B = PH.
seek_call(A.As, _.Ms, B.As, V) :-
nonvar(A),
A =.. F.FAs,
seek_call(FAs, Ms, FBs, V),
!, B =.. F.FBs.
seek_call(A.As, _.Ms, A.Bs, V) :-
!, seek_call(As, Ms, Bs, V).
:- multifile user:goal_expansion/2.
user:goal_expansion(X, Y) :-
( X = (_ , _) ; X = (_ ; _) ; X = (_ -> _) )
-> !, fail % leave control flow unchanged (useless after the meta... handling?)
; funcs(X, Y).
/* end eog.pl */
I was advised against it. Effectively, the [A|B] syntax it's an evolution of the . operator, introduced for readability.
OT: what's that code?
the code above it's my attempt to sweeten Prolog with functions. Namely, introduces on request, by means of $, the temporary variables required (for instance) by arithmetic expressions
fact(N, F) :-
N > 1 -> F is N * $fact($(N - 1)) ; F is 1.
each $ introduce a variable. After expansion, we have a more traditional fact/2
?- listing(fact).
plunit_eog:fact(A, C) :-
( A>1
-> B is A+ -1,
fact(B, D),
C is A*D
; C is 1
).
Where we have many expressions, that could be useful...
This syntax comes from NU-Prolog. See here. It's probably just the normal list functor '.'/2 redefined as an infix operator, without the need for a trailing empty list:
?- L= .(a,.(b,[])).
L = [a,b]
Yes (0.00s cpu)
?- op(500, xfy, '.').
Yes (0.00s cpu)
?- L = a.b.[].
L = [a,b]
Yes (0.00s cpu)
The following code isn't working
:- arithmetic_function(i/2).
i(X,Y,Z) :-
Z is X+Y.
calcola :-
write('Give me an expression'),nl,
read(ESP),
Z is ESP,nl,nl,
write(Z).
but the following is
:- arithmetic_function(i/2).
i(X,Y,Z) :-
Z is X+Y.
calcola :-
write('Give me an expression'),nl,
Z is 4 i 2,nl,nl,
write(Z).
Why is that? Seems like the "read" function isn't working properly
from SWI-Prolog mailing list ([SWIPL] Ann: SWI-Prolog 5.11.23, 23 Jun):
MODIFIED: User-defined arithmetic functions have been removed from
the kernel. There is a new library(arithmetic) that emulates the
old behaviour PARTIALLY. Notably:
This library must be loaded before arithmetic_function/1 is
used.
It only covers arithmetic functions that are visible as an argument
to is/2, >/2, etc. at compile-time.
A new predicate arithmetic_expression_value/2 can be used to
evaluate expressions with embedded user arithmetic that become
instantiated at runtime.
Well as a lead, when I test it with is/2 it fails but when I use arithmetic_expression_value/2 it succeeds :
:- arithmetic_function(i/2).
:- op(20, xfx, i).
i(X, Y, Z) :-
Z is X + Y.
calcola :-
writeln('Give me an expression'),
read(ESP),
arithmetic_expression_value(ESP, Z), nl,
write(Z).
For #gusbro, it works out of the box. I'm using windows swi-pl here, for the record !
Others may have clues about why it fails for us !