我喜欢此页面上的 Nomura 解决方案。我对谓词和要解决的事实稍作修改。
enum Digit { N0,N1,N2,N3,N4,N5,N6,N7,N8,N9 }
one sig Code {a,b,c: Digit}
pred hint(code: Code, d1,d2,d3: Digit, correct, wellPlaced:Int) {
correct = #((code.a + code.b + code.c)&(d1 + d2 + d3))
wellPlaced = #((0->code.a + 1->code.b + 2->code.c)&(0->d1 + 1->d2 + 2->d3))
}
fact {
some code: Code |
hint[code, N6,N8,N2, 1,1] and
hint[code, N6,N1,N4, 1,0] and
hint[code, N2,N0,N6, 2,0] and
hint[code, N7,N3,N8, 0,0] and
hint[code, N7,N8,N0, 1,0]
}
run {}
更新(2020-12-29):
Nomura (https://*.com/a/61022419/5005552) 提出的新难题证明了原始解决方案的一个弱点:它没有考虑代码中数字的多次使用。对“正确”表达式的修改解决了这个问题。将每个猜测的数字与传递代码中的数字的并集相交,并将它们相加以获得真正的基数。我将匹配封装在一个函数中,它会为每个数字返回 0 或 1。
enum Digit {N0,N1,N2,N3,N4,N5,N6,N7,N8,N9}
let sequence[a,b,c] = 0->a + 1->b + 2->c
one sig Code {c1, c2, c3: Digit}
fun match[code: Code, d: Digit]: Int { #((code.c1 + code.c2 + code.c3) & d) }
pred hint(code: Code, d1,d2,d3: Digit, correct, wellPlaced:Int) {
// The intersection of each guessed digit with the code (unordered) tells us
// whether any of the digits match each other and how many
correct = match[code,d1].plus[match[code,d2]].plus[match[code,d3]]
// The intersection of the sequences of digits (ordered) tells us whether
// any of the digits are correct AND in the right place in the sequence
wellPlaced = #(sequence[code.c1,code.c2,code.c3] & sequence[d1, d2, d3])
}
pred originalLock {
some code: Code |
hint[code, N6,N8,N2, 1,1] and
hint[code, N6,N1,N4, 1,0] and
hint[code, N2,N0,N6, 2,0] and
hint[code, N7,N3,N8, 0,0] and
hint[code, N7,N8,N0, 1,0]
}
pred newLock {
some code: Code |
hint[code, N1,N2,N3, 0,0] and
hint[code, N4,N5,N6, 0,0] and
hint[code, N7,N8,N9, 1,0] and
hint[code, N9,N0,N0, 3,1]
}
run originalLock
run newLock
run test {some code: Code | hint[code, N9,N0,N0, 3,1]}