DCG prolog testing several sentences

问题:

If I have the below piece of code, how would I make it produce Answer= 5 and Answer2= 10?. I run the goal ?- test(Data),lpsolve(Data, [Answer1,Answer2]).

    :-use_module(library(clpfd)).
    test([the, variable, X, is, five,fullstop,
          the,variable, Y, is, ten, fullstop]).
    lpsolve(Data, [Answer,Answer2]):- sentence(Answer, Data,[]).

    sentence(X) --> nounphrase, verbphrase(X).
    nounphrase --> [the], [variable].
    verbphrase(X) --> [X], [is], [five],[fullstop], {X = 5}.

    sentence(Y) --> nounphrase, verbphrase(Y).
    nounphrase --> [the], [variable].
    verbphrase(Y) --> [Y], [is], [ten],[fullstop], {Y = 10}.

Example of a program that actually runs and is closely related is the following:

:-use_module(library(clpfd)).
test([the, variable, X, is, five,fullstop]).
lpsolve(Data, Answer):- sentence(Answer, Data,[]).

sentence(X) --> nounphrase, verbphrase(X).
nounphrase --> [the], [variable].
verbphrase(X) --> [X], [is], [five],[fullstop], {X = 5}.

I have just one sentence to test and the goal succeeds as shown below.

 ?- test(Data),lpsolve(Data, Answer).
    Data = [the, variable, 5, is, five],
    Answer = 5.

EDIT

I try the following as per the first comment:

:-use_module(library(clpfd)).

test([the, variable, x, is, five,fullstop,
      the,variable, y, is, ten, fullstop]).
lpsolve(Data, Answer):- sentence(Answer, Data,[]).

sentence(X) --> nounphrase, verbphrase(X).
nounphrase --> [the], [variable].
verbphrase(X) --> [x], [is], [five],[fullstop], {X = 5}. 
verbphrase(Y) --> [y], [is], [ten],[fullstop], {Y = 10}.

I get the following:

-? test(Data),lpsolve(Data, Answer).
false.

回答1:

I'm not really sure what you're trying to do here, but I feel your DCG is broken down in completely strange ways and you may benefit from seeing another way to arrange it.

You have a list of variable bindings, so you should already be thinking in terms of obtaining a list of results rather than a single result:

sentences([S|Ss]) --> sentence(S), sentences(Ss).
sentences([]) --> [].

What is a sentence? It is a noun phrase and a verb phrase, in your simple system. But you have broken the noun and verb phrases apart incorrectly for English, where a sentence like "The variable X is 5" should be broken into a noun phrase subject "The variable X" and a verb phrase "is 5". Ideally, the verb phrase should be decomposed further into a verb another noun phrase, but we're ignoring that detail for now. I am looking for the verb "is" to relate it to the Prolog predicate =/2, and handling fullstop here:

sentence(N1=N2) --> nounphrase(N1), verbphrase(is, N2), [fullstop].

OK, so now we need your noun phrase:

nounphrase(N) --> [the, variable, N].

And your verb phrase:

verbphrase(is, V) --> [is], value(V).

I'm only handling the two decodings and I'm doing it implicitly in the DCG definition:

value(5) --> [five].
value(10) --> [ten].

You'll find this works for the use case you've defined above:

?- phrase(sentences(S), [the,variable,X,is,five,fullstop,the,variable,Y,is,ten,fullstop]).
  S = [X=5, Y=10] ;
  false.