解决方案(在Java中)可能是以下(我相信这里的朋友可以找到更好的): 从字符串的末尾开始交换数字，直到得到一个更高的数字。 也就是说，首先从下位开始移动。然后到达下一个更高的地方，直到你到达下一个更高的地方。 然后对剩下的进行排序。 在你的例子中，你会得到:

38276 --> 38267 (smaller) --> 38627 Found it    
    ^        ^                  ^        

 public static int nextDigit(int number){
    String num = String.valueOf(number);        
    int stop = 0;       
    char [] chars = null;
    outer:
        for(int i = num.length() - 1; i > 0; i--){          
            chars = num.toCharArray();
            for(int j = i; j > 0; j--){
                char temp = chars[j];
                chars[j] = chars[j - 1];
                chars[j - 1] = temp;
                if(Integer.valueOf(new String(chars)) > number){
                    stop = j;                   
                    break outer;                                
                }               
            }               
        }

    Arrays.sort(chars, stop, chars.length); 
    return Integer.valueOf(new String(chars));
}

关于如何做到这一点，请参阅Knuth的“计算机编程艺术:生成所有排列”(.ps.gz)中的“算法L”。

这是另一个Java实现，可以开箱即用，并通过测试完成。 这个解决方案是O(n)个空间和时间，使用老式的动态规划。

如果你想用蛮力，有两种蛮力:

排列所有的东西，然后选择最小值更高的:O(n!) 与此实现类似，但不是DP，而是强制填充的步骤 indexToIndexOfNextSmallerLeft映射将在O(n²)中运行。

import java.util.Arrays;
import java.util.HashMap;
import java.util.Map;

import org.junit.Test;

import static org.junit.Assert.assertEquals;

public class NextHigherSameDigits {

    public long next(final long num) {
        final char[] chars = String.valueOf(num).toCharArray();
        final int[] digits = new int[chars.length];
        for (int i = 0; i < chars.length; i++) {
            digits[i] = Character.getNumericValue(chars[i]);
        }

        final Map<Integer, Integer> indexToIndexOfNextSmallerLeft = new HashMap<>();
        indexToIndexOfNextSmallerLeft.put(1, digits[1] > digits[0] ? 0 : null);
        for (int i = 2; i < digits.length; i++) {
            final int left = digits[i - 1];
            final int current = digits[i];
            Integer indexOfNextSmallerLeft = null;
            if (current > left) {
                indexOfNextSmallerLeft = i - 1;
            } else {
                final Integer indexOfnextSmallerLeftOfLeft = indexToIndexOfNextSmallerLeft.get(i - 1);
                final Integer nextSmallerLeftOfLeft = indexOfnextSmallerLeftOfLeft == null ? null : 
                    digits[indexOfnextSmallerLeftOfLeft];

                if (nextSmallerLeftOfLeft != null && current > nextSmallerLeftOfLeft) {
                    indexOfNextSmallerLeft = indexOfnextSmallerLeftOfLeft;
                } else {
                    indexOfNextSmallerLeft = null;
                }
            }

            indexToIndexOfNextSmallerLeft.put(i, indexOfNextSmallerLeft);
        }

        Integer maxOfindexOfNextSmallerLeft = null;
        Integer indexOfMinToSwapWithNextSmallerLeft = null;
        for (int i = digits.length - 1; i >= 1; i--) {
            final Integer indexOfNextSmallerLeft = indexToIndexOfNextSmallerLeft.get(i);
            if (maxOfindexOfNextSmallerLeft == null ||
                    (indexOfNextSmallerLeft != null && indexOfNextSmallerLeft > maxOfindexOfNextSmallerLeft)) {

                maxOfindexOfNextSmallerLeft = indexOfNextSmallerLeft;
                if (maxOfindexOfNextSmallerLeft != null && (indexOfMinToSwapWithNextSmallerLeft == null || 
                        digits[i] < digits[indexOfMinToSwapWithNextSmallerLeft])) {

                    indexOfMinToSwapWithNextSmallerLeft = i;
                }
            }
        }

        if (maxOfindexOfNextSmallerLeft == null) {
            return -1;
        } else {
            swap(digits, indexOfMinToSwapWithNextSmallerLeft, maxOfindexOfNextSmallerLeft);
            reverseRemainingOfArray(digits, maxOfindexOfNextSmallerLeft + 1);
            return backToLong(digits);
        }
    }

    private void reverseRemainingOfArray(final int[] digits, final int startIndex) {
        final int[] tail = Arrays.copyOfRange(digits, startIndex, digits.length);
        for (int i = tail.length - 1; i >= 0; i--) {
            digits[(digits.length - 1)  - i] = tail[i];                 
        }
    }

    private void swap(final int[] digits, final int currentIndex, final int indexOfNextSmallerLeft) {
        int temp = digits[currentIndex];
        digits[currentIndex] = digits[indexOfNextSmallerLeft];
        digits[indexOfNextSmallerLeft] = temp;
    }

    private long backToLong(int[] digits) {     
        StringBuilder sb = new StringBuilder();
        for (long i : digits) {
            sb.append(String.valueOf(i));
        }

        return Long.parseLong(sb.toString());
    }

    @Test
    public void test() {
        final long input1 =    34722641;
        final long expected1 = 34724126;
        final long output1 = new NextHigherSameDigits().next(input1);
        assertEquals(expected1, output1);

        final long input2 =    38276;
        final long expected2 = 38627;
        final long output2 = new NextHigherSameDigits().next(input2);
        assertEquals(expected2, output2);

        final long input3 =    54321;
        final long expected3 = -1;
        final long output3 = new NextHigherSameDigits().next(input3);
        assertEquals(expected3, output3);

        final long input4 =    123456784987654321L;
        final long expected4 = 123456785123446789L;
        final long output4 = new NextHigherSameDigits().next(input4);
        assertEquals(expected4, output4);

        final long input5 =    9999;
        final long expected5 = -1;
        final long output5 = new NextHigherSameDigits().next(input5);
        assertEquals(expected5, output5);
    }

}

我只测试了两个数字。他们工作。 在去年12月退休之前，我做了8年的IT经理，我关心三件事: 1)准确性:如果它总是有效，那就很好。 2)速度:用户可以接受。 3)明确:我可能没有你聪明，但我付你薪水。确保你用英语解释你在做什么。

奥马尔，祝你好运。

Sub Main()

Dim Base(0 To 9) As Long
Dim Test(0 To 9) As Long

Dim i As Long
Dim j As Long
Dim k As Long
Dim ctr As Long

Const x As Long = 776914648
Dim y As Long
Dim z As Long

Dim flag As Boolean

' Store the digit count for the original number in the Base vector.
    For i = 0 To 9
        ctr = 0
        For j = 1 To Len(CStr(x))
            If Mid$(CStr(x), j, 1) = i Then ctr = ctr + 1
        Next j
        Base(i) = ctr
    Next i

' Start comparing from the next highest number.
    y = x + 1
    Do

' Store the digit count for the each new number in the Test vector.
        flag = False
        For i = 0 To 9
            ctr = 0
            For j = 1 To Len(CStr(y))
                If Mid$(CStr(y), j, 1) = i Then ctr = ctr + 1
            Next j
            Test(i) = ctr
        Next i

' Compare the digit counts.
        For k = 0 To 9
            If Test(k) <> Base(k) Then flag = True
        Next k

' If no match, INC and repeat.
        If flag = True Then
            y = y + 1
            Erase Test()
        Else
            z = y ' Match.
        End If

    Loop Until z > 0

    MsgBox (z), , "Solution"

End Sub

你的想法

我想从找到第一个比个位小的数字的下标开始。然后我将旋转子集中的最后一个数字，这样它是由相同的数字组成的下一个最大的数字，但卡住了。

其实还不错。您不仅要考虑最后一位数字，还要考虑所有比当前考虑的不那么重要的数字。在此之前，我们有一个单调的数字序列，即最右边的数字比它右边的邻居小。把

1234675
    ^

下一个有相同数字的大数是

1234756

将找到的数字交换为最后一位数字(考虑的数字中最小的数字)，其余数字按递增顺序排列。

I didn't know anything about the brute force algorithm when answering this question, so I approached it from another angle. I decided to search the entire range of possible solutions that this number could possibly be rearranged into, starting from the number_given+1 up to the max number available (999 for a 3 digit number, 9999 for 4 digits, etc.). I did this kind of like finding a palindrome with words, by sorting the numbers of each solution and comparing it to the sorted number given as the parameter. I then simply returned the first solution in the array of solutions, as this would be the next possible value.

下面是我的Ruby代码:

def PermutationStep(num)

    a = []
    (num.to_s.length).times { a.push("9") }
    max_num = a.join('').to_i
    verify = num.to_s.split('').sort
    matches = ((num+1)..max_num).select {|n| n.to_s.split('').sort == verify }

    if matches.length < 1
      return -1
    else
      matches[0]
    end
end