list - 如何删除所有出现的情况？

Question

我试图理解为什么代码 1 工作正常但代码 2 输出错误:

remove(X,[],[]).
remove(X,[X|Y],Z):-remove(X,Y,Z). % [X|Y] is the input list
remove(X,[F|Y],[F|Z]):-remove(X,Y,Z). % code 1


remove(X,[],[]).
remove(X,[X|Y],Z):-remove(X,Y,Z). % [X|Y] is the input list
remove(X,[F|Y],Z):-remove(X,Y,[F|Z]). % code 2

Answer 1

您提出的两个谓词 - 代码 1 和 代码 2 - 均已损坏。

您没有注意到这一点的原因是:查询。

---

#1)考虑 this previous "answer" 中显示的查询:

?- remove(x, [x,a,b,c,x,d], R).R = [a,b,c,d]                      % okay

Okay? Yes, but there may be more answers. Let's press ;↵!

;  R = [a,b,c,x,d]                 % BAD!;  R = [x,a,b,c,d]                 % BAD!;  R = [x,a,b,c,x,d]               % BAD!;  false.                          % (no more answers)

These three answers are invalid, as each R contains some x.

The bottom like: Don't just look at some query answers. Look at all query answers!

---

#2)Prolog programs comprise different kinds of clauses: facts, rules, and queries.

Queries are a very—if not the most—important part of your program.

Queries are both documentation and specification. Most importantly, they enable you to delegate the "mechanical" part of program development to the Prolog processor.

So which queries to write?

• Start with the most general query:

  ?- remove(A,B,C).

• Queries you expect to succeed:

  ?- remove(x,[x,a,x,p],[a,p]).   % x is first element
  ?- remove(x,[a,x,p,x],[a,p]).   % x is last element
  ?- remove(x,[a,x,p],[a,p]).
  

• 您预计会失败的查询

  ?- \+ (remove(x,[a,x,p],Ls), dif(Ls,[a,p])).
  

• 地面查询:

  ?- remove(x,[],[]).
  

• 非基础查询(带变量的查询):

  ?- remove(X,[a,b,a,c,d,c,a,b],Xs).
  

底线:如果没有查询，您就不是在编写代码，您只是在编写文本。

---

#3)现在让我们开始修复代码 1:看看两个递归子句!

remove(X,[X|Y],Z) :- remove(X,Y,Z).remove(X,[F|Y],[F|Z]) :- remove(X,Y,Z).

第一条规则将 X 与第一个列表 [X|Y] 相关。好的!

第二条规则不将 X 与 [F|Y] 或 [F|Z] 结合使用。糟糕!

通过添加prolog-dif目标 dif/2 到第二个子句，我们可以建立该连接。

---

#4)完成!以下是谓词 remove/3 的完整代码:

remove(_X, [], []).remove(X, [X|Y], Z) :-   remove(X, Y, Z).remove(X, [F|Y], [F|Z]) :-   dif(X, F),   remove(X, Y, Z).