Python中的简短示例:

def comb(sofar, rest, n):
    if n == 0:
        print sofar
    else:
        for i in range(len(rest)):
            comb(sofar + rest[i], rest[i+1:], n-1)

>>> comb("", "abcde", 3)
abc
abd
abe
acd
ace
ade
bcd
bce
bde
cde

为了解释，递归方法用下面的例子描述:

示例:A B C D E 3的所有组合是:

A与其余2的所有组合(B C D E) B与其余2的所有组合(C D E) C与其余2的所有组合(D E)

在Python中，利用递归的优势和所有事情都是通过引用完成的事实。对于非常大的集合，这将占用大量内存，但其优点是初始集合可以是一个复杂的对象。它只会找到唯一的组合。

import copy

def find_combinations( length, set, combinations = None, candidate = None ):
    # recursive function to calculate all unique combinations of unique values
    # from [set], given combinations of [length].  The result is populated
    # into the 'combinations' list.
    #
    if combinations == None:
        combinations = []
    if candidate == None:
        candidate = []

    for item in set:
        if item in candidate:
            # this item already appears in the current combination somewhere.
            # skip it
            continue

        attempt = copy.deepcopy(candidate)
        attempt.append(item)
        # sorting the subset is what gives us completely unique combinations,
        # so that [1, 2, 3] and [1, 3, 2] will be treated as equals
        attempt.sort()

        if len(attempt) < length:
            # the current attempt at finding a new combination is still too
            # short, so add another item to the end of the set
            # yay recursion!
            find_combinations( length, set, combinations, attempt )
        else:
            # the current combination attempt is the right length.  If it
            # already appears in the list of found combinations then we'll
            # skip it.
            if attempt in combinations:
                continue
            else:
                # otherwise, we append it to the list of found combinations
                # and move on.
                combinations.append(attempt)
                continue
    return len(combinations)

你可以这样使用它。传递'result'是可选的，所以你可以用它来获取可能组合的数量…尽管这样做效率很低(最好通过计算来完成)。

size = 3
set = [1, 2, 3, 4, 5]
result = []

num = find_combinations( size, set, result ) 
print "size %d results in %d sets" % (size, num)
print "result: %s" % (result,)

您应该从测试数据中得到以下输出:

size 3 results in 10 sets
result: [[1, 2, 3], [1, 2, 4], [1, 2, 5], [1, 3, 4], [1, 3, 5], [1, 4, 5], [2, 3, 4], [2, 3, 5], [2, 4, 5], [3, 4, 5]]

如果你的集合是这样的，它也会工作得很好:

set = [
    [ 'vanilla', 'cupcake' ],
    [ 'chocolate', 'pudding' ],
    [ 'vanilla', 'pudding' ],
    [ 'chocolate', 'cookie' ],
    [ 'mint', 'cookie' ]
]

为此，我在SQL Server 2005中创建了一个解决方案，并将其发布在我的网站上:http://www.jessemclain.com/downloads/code/sql/fn_GetMChooseNCombos.sql.htm

下面是一个例子来说明用法:

SELECT * FROM dbo.fn_GetMChooseNCombos('ABCD', 2, '')

结果:

Word
----
AB
AC
AD
BC
BD
CD

(6 row(s) affected)

假设你的字母数组是这样的:"ABCDEFGH"。你有三个下标(i, j, k)来表示你要用哪个字母来表示当前单词。

A B C D E F G H
^ ^ ^
i j k

首先你改变k，所以下一步看起来像这样:

A B C D E F G H
^ ^   ^
i j   k

如果你到达终点，你继续改变j和k。

A B C D E F G H
^   ^ ^
i   j k

A B C D E F G H
^   ^   ^
i   j   k

一旦j达到G, i也开始变化。

A B C D E F G H
  ^ ^ ^
  i j k

A B C D E F G H
  ^ ^   ^
  i j   k
...

用代码写出来是这样的

void print_combinations(const char *string)
{
    int i, j, k;
    int len = strlen(string);

    for (i = 0; i < len - 2; i++)
    {
        for (j = i + 1; j < len - 1; j++)
        {
            for (k = j + 1; k < len; k++)
                printf("%c%c%c\n", string[i], string[j], string[k]);
        }
    }
}

现在又出现了祖辈COBOL，一种饱受诟病的语言。

让我们假设一个包含34个元素的数组，每个元素8个字节(完全是任意选择)。其思想是枚举所有可能的4元素组合，并将它们加载到一个数组中。

我们使用4个指标，每个指标代表4个组中的每个位置

数组是这样处理的:

    idx1 = 1
    idx2 = 2
    idx3 = 3
    idx4 = 4

我们把idx4从4变到最后。对于每个idx4，我们得到一个唯一的组合 四人一组。当idx4到达数组的末尾时，我们将idx3增加1，并将idx4设置为idx3+1。然后再次运行idx4到最后。我们以这种方式继续，分别增加idx3、idx2和idx1，直到idx1的位置距离数组末端小于4。算法就完成了。

1          --- pos.1
2          --- pos 2
3          --- pos 3
4          --- pos 4
5
6
7
etc.

第一次迭代:

1234
1235
1236
1237
1245
1246
1247
1256
1257
1267
etc.

一个COBOL的例子:

01  DATA_ARAY.
    05  FILLER     PIC X(8)    VALUE  "VALUE_01".
    05  FILLER     PIC X(8)    VALUE  "VALUE_02".
  etc.
01  ARAY_DATA    OCCURS 34.
    05  ARAY_ITEM       PIC X(8).

01  OUTPUT_ARAY   OCCURS  50000   PIC X(32).

01   MAX_NUM   PIC 99 COMP VALUE 34.

01  INDEXXES  COMP.
    05  IDX1            PIC 99.
    05  IDX2            PIC 99.
    05  IDX3            PIC 99.
    05  IDX4            PIC 99.
    05  OUT_IDX   PIC 9(9).

01  WHERE_TO_STOP_SEARCH          PIC 99  COMP.

* Stop the search when IDX1 is on the third last array element:

COMPUTE WHERE_TO_STOP_SEARCH = MAX_VALUE - 3     

MOVE 1 TO IDX1

PERFORM UNTIL IDX1 > WHERE_TO_STOP_SEARCH
   COMPUTE IDX2 = IDX1 + 1
   PERFORM UNTIL IDX2 > MAX_NUM
      COMPUTE IDX3 = IDX2 + 1
      PERFORM UNTIL IDX3 > MAX_NUM
         COMPUTE IDX4 = IDX3 + 1
         PERFORM UNTIL IDX4 > MAX_NUM
            ADD 1 TO OUT_IDX
            STRING  ARAY_ITEM(IDX1)
                    ARAY_ITEM(IDX2)
                    ARAY_ITEM(IDX3)
                    ARAY_ITEM(IDX4)
                    INTO OUTPUT_ARAY(OUT_IDX)
            ADD 1 TO IDX4
         END-PERFORM
         ADD 1 TO IDX3
      END-PERFORM
      ADD 1 TO IDX2
   END_PERFORM
   ADD 1 TO IDX1
END-PERFORM.

下面是Clojure版本，它使用了我在OCaml实现答案中描述的相同算法:

(defn select
  ([items]
     (select items 0 (inc (count items))))
  ([items n1 n2]
     (reduce concat
             (map #(select % items)
                  (range n1 (inc n2)))))
  ([n items]
     (let [
           lmul (fn [a list-of-lists-of-bs]
                     (map #(cons a %) list-of-lists-of-bs))
           ]
       (if (= n (count items))
         (list items)
         (if (empty? items)
           items
           (concat
            (select n (rest items))
            (lmul (first items) (select (dec n) (rest items))))))))) 

它提供了三种调用方法:

(a)按问题要求，选出n项:

  user=> (count (select 3 "abcdefgh"))
  56

(b) n1至n2个选定项目:

user=> (select '(1 2 3 4) 2 3)
((3 4) (2 4) (2 3) (1 4) (1 3) (1 2) (2 3 4) (1 3 4) (1 2 4) (1 2 3))

(c)在0至所选项目的集合大小之间:

user=> (select '(1 2 3))
(() (3) (2) (1) (2 3) (1 3) (1 2) (1 2 3))