你正在寻找一个双射NxN - >n映射。这些是用于例如燕尾。请看这个PDF文件，它介绍了所谓的配对函数。维基百科介绍了一个特定的配对函数，即康托配对函数:

备注:三个

As others have made clear, if you plan to implement a pairing function, you may soon find you need arbitrarily large integers (bignums). If you don't want to make a distinction between the pairs (a, b) and (b, a), then sort a and b before applying the pairing function. Actually I lied. You are looking for a bijective ZxZ -> N mapping. Cantor's function only works on non-negative numbers. This is not a problem however, because it's easy to define a bijection f : Z -> N, like so: f(n) = n * 2 if n >= 0 f(n) = -n * 2 - 1 if n < 0

再简单一点:给定两个数字，A和B让str为串联:'A' + ';' + 'B'。然后让输出为hash(str)。我知道这不是一个数学答案，但一个简单的python(有一个内置的哈希函数)脚本应该做这项工作。

下面是基于@nawfal给出的方法将@DoctorJ的代码扩展到无界整数。它可以编码和解码。它适用于普通数组和numpy数组。

#!/usr/bin/env python
from numbers import Integral    

def tuple_to_int(tup):
    """:Return: the unique non-negative integer encoding of a tuple of non-negative integers."""
    if len(tup) == 0:  # normally do if not tup, but doesn't work with np
        raise ValueError('Cannot encode empty tuple')
    if len(tup) == 1:
        x = tup[0]
        if not isinstance(x, Integral):
            raise ValueError('Can only encode integers')
        return x
    elif len(tup) == 2:
        # print("len=2")
        x, y = tuple_to_int(tup[0:1]), tuple_to_int(tup[1:2])  # Just to validate x and y

        X = 2 * x if x >= 0 else -2 * x - 1  # map x to positive integers
        Y = 2 * y if y >= 0 else -2 * y - 1  # map y to positive integers
        Z = (X * X + X + Y) if X >= Y else (X + Y * Y)  # encode

        # Map evens onto positives
        if (x >= 0 and y >= 0):
            return Z // 2
        elif (x < 0 and y >= 0 and X >= Y):
            return Z // 2
        elif (x < 0 and y < 0 and X < Y):
            return Z // 2
        # Map odds onto negative
        else:
            return (-Z - 1) // 2
    else:
        return tuple_to_int((tuple_to_int(tup[:2]),) + tuple(tup[2:]))  # ***speed up tuple(tup[2:])?***


def int_to_tuple(num, size=2):
    """:Return: the unique tuple of length `size` that encodes to `num`."""
    if not isinstance(num, Integral):
        raise ValueError('Can only encode integers (got {})'.format(num))
    if not isinstance(size, Integral) or size < 1:
        raise ValueError('Tuple is the wrong size ({})'.format(size))
    if size == 1:
        return (num,)
    elif size == 2:

        # Mapping onto positive integers
        Z = -2 * num - 1 if num < 0 else 2 * num

        # Reversing Pairing
        s = isqrt(Z)
        if Z - s * s < s:
            X, Y = Z - s * s, s
        else:
            X, Y = s, Z - s * s - s

        # Undoing mappint to positive integers
        x = (X + 1) // -2 if X % 2 else X // 2  # True if X not divisible by 2
        y = (Y + 1) // -2 if Y % 2 else Y // 2  # True if Y not divisible by 2

        return x, y

    else:
        x, y = int_to_tuple(num, 2)
        return int_to_tuple(x, size - 1) + (y,)


def isqrt(n):
    """":Return: the largest integer x for which x * x does not exceed n."""
    # Newton's method, via http://stackoverflow.com/a/15391420
    x = n
    y = (x + 1) // 2
    while y < x:
        x = y
        y = (x + n // x) // 2
    return x

对于作为参数的正整数和参数顺序无关的情况:

下面是一个无序配对函数: < x, y > = x * y + trunc ((x - y | | - 1) ^ 2 / 4) = < y、x > 对于x≠y，这里有一个唯一的无序配对函数: <x, y> = if x < y: X * (y - 1) + trunc((y - X - 2)²/ 4) 如果x > y: (x - 1) * y + trunc((x - y - 2)^2 / 4) = <y, x>

如果A和B可以用2个字节表示，那么可以用4个字节组合它们。把A放在最有效的一半，B放在最不有效的一半。

在C语言中，这给出了(假设sizeof(short)=2和sizeof(int)=4):

unsigned int combine(unsigned short A, unsigned short B)
{
    return ((unsigned)A<<16) | (unsigned)B;
}

unsigned short getA(unsigned int C)
{
    return C>>16;
}

unsigned short getB(unsigned int C)
{
    return C & 0xFFFF;    // or  return (unsigned short)C;
}

使输入unsigned short或uint16_t确保他们在你|或+他们一起之前零扩展。否则- B会将上面的位设置为全1或，或者如果你添加，则从上半部分减去1。

强制转换(unsigned)A可以避免将窄类型默认提升为带符号int后左移的带符号溢出UB。对于更广泛的类型，也必须避免转移出位你保持，如((uint64_t)A << 32 | B，因为默认提升停止在int。

(unsigned)B强制转换是不必要的;重要的是它一开始是无符号空头B。左边的|是无符号的意味着它也将转换为无符号的。

你可以将它用于有符号类型，至少是getA和getB，你可以从combine返回有符号int，但是输入需要0 -extend，所以在C中你需要它们在扩大之前是无符号的short。比如((unsigned)(unsigned空头)A << 16) | (unsigned空头)B

你可能想要使用uint16_t和uint32_t，来定义类型宽度，以匹配你正在使用的移位计数。

看看这个:http://en.wikipedia.org/wiki/Pigeonhole_principle。如果A, B, C是同一类型，就不能做。如果A和B是16位整数，而C是32位整数，那么您可以简单地使用移位。

哈希算法的本质是它们不能为每个不同的输入提供唯一的哈希。