简体   繁体   English

从 Prolog 的列表中删除谓词

[英]Delete predicates from the list in Prolog

I want to delete all predicates named a from the list.我想从列表中删除所有名为a的谓词。 The result must be as shown below:结果必须如下图所示:

?- delete_all(a(_), [a(1),a(2),a(3),b(1)], R).
R = [b(1)]

Please, do not offer me built in solutions of SWI or others, because the code must be in Amzi-Prolog.请不要向我提供 SWI 或其他内置解决方案,因为代码必须在 Amzi-Prolog 中。

Thanks.谢谢。

Edit: I have tried the following code but it is working properly only for atoms:编辑:我尝试了以下代码,但它仅适用于原子:

remove_all(X,[],[]).
remove_all(X,[X|L],R):-remove_all(X,L,R).
remove_all(X,[Y|L],R):-not(X=Y), remove_all(X,L,M), R=[Y|M].

?-remove_all(a(_), [a(1),a(2),a(3),b(1)], R).
R=[a(2),a(3),b(1)]

which is not true:(这是不正确的:(

use findall使用 findall

findall(X, (member(X,[a(1),a(2),a(3),b(1)]),\+(X=a(_))) ,V).

I know you said no swi-prolog.我知道你说没有 swi-prolog。 However it's an easy task (compared to the main one), to implement a recursion that behaves the same as the exclude/3 used, the rest should be ISO prolog or present in amzi too.然而,实现与使用的exclude/3行为相同的递归是一项简单的任务(与主要任务相比),rest 应该是 ISO prolog 或也存在于 amzi 中。 It useslambda.pl , a library that allows easier higher order programming:它使用lambda.pl ,一个允许更简单的高阶编程的库:

:- [lambda].
filter(Term, List, Result) :-
    Term =.. [Pred|Args],
    length(Args, Arity),
    exclude(\X^(X =.. [Pred2|Args2],
                length(Args2, Arity2),
                Pred == Pred2,
                Arity == Arity2), List, Result).

This solution has the advantage of staying away from the unpure findall/3 .该解决方案的优点是远离不纯的findall/3

Hope this helps.希望这可以帮助。

Little fix, and it should work:小修复,它应该工作:

% remove_same_indicator(+Callable,+List,-List)
remove_same_indicator(_, [], []).
remove_same_indicator(X, [Y|L], R) :- 
       functor(X, F, N), 
       functor(Y, F, N), 
       !, 
       remove_same_indicator(X, L, R).
remove_same_indicator(X, [Y|L], [Y|R]) :- 
       remove_same_indicator(X, L, R).

Let's give it a try:试一试吧:

?- remove_same_indicator(a(_), [a(1),a(2),a(3),b(1)], R).
R = [b(1)]

Advantage over findall solution, one does not loose variables.优于 findall 解决方案,一个不会丢失变量。 For example one can do:例如可以这样做:

?- remove_same_indicator(a(_), [a(A),a(B),a(C),b(A)], R).
R = [b(A)]

But with the findall solution we get:但是使用 findall 解决方案我们得到:

?- L=[a(A),a(B),a(C),b(A)], findall(X, (member(X,L),\+ (X = a(_))), R).
L = [a(A), a(B), a(C), b(A)],
R = [b(_I)]

The argument of b is not anymore bound to A, since findall creates copies and thus new variables. b 的参数不再绑定到 A,因为 findall 创建副本并因此创建新变量。

Bye再见

functor/3 is ISO and also in Amzi! functor/3 是 ISO,也在 Amzi 中!
http://www.amzi.com/manuals/amzi/pro/ref_manipulating_terms.htm#functorTermFunctorN http://www.amzi.com/manuals/amzi/pro/ref_manipulating_terms.htm#functorTermFunctorN

remove_all(_, [], []).
remove_all(X, [Y|R], L):- \+ X \= Y, remove_all(X, R, L).
remove_all(X, [Y|R], [Y|R2]):- X \= Y, remove_all(X, R2, L).

声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.

 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM