我刚刚搞砸了一次面试，而且我的面试问题几乎毫无进展。
给定一个数，找出下一个与之完全相同的更大的数 作为原始数字的数字集。例如:给定38276返回 38627
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。
面试官还建议我试着一次换一个数字，但我搞不懂算法，只能盯着屏幕看20-30分钟。不用说，我想我还得继续找工作。
当前回答
这里是Java实现
public static int nextHigherNumber(int number) {
Integer[] array = convertToArray(number);
int pivotIndex = pivotMaxIndex(array);
int digitInFirstSequence = pivotIndex -1;
int lowerDigitIndexInSecondSequence = lowerDigitIndex(array[digitInFirstSequence], array, pivotIndex);
swap(array, digitInFirstSequence, lowerDigitIndexInSecondSequence);
doRercursiveQuickSort(array, pivotIndex, array.length - 1);
return arrayToInteger(array);
}
public static Integer[] convertToArray(int number) {
int i = 0;
int length = (int) Math.log10(number);
int divisor = (int) Math.pow(10, length);
Integer temp[] = new Integer[length + 1];
while (number != 0) {
temp[i] = number / divisor;
if (i < length) {
++i;
}
number = number % divisor;
if (i != 0) {
divisor = divisor / 10;
}
}
return temp;
}
private static int pivotMaxIndex(Integer[] array) {
int index = array.length - 1;
while(index > 0) {
if (array[index-1] < array[index]) {
break;
}
index--;
}
return index;
}
private static int lowerDigitIndex(int number, Integer[] array, int fromIndex) {
int lowerMaxIndex = fromIndex;
int lowerMax = array[lowerMaxIndex];
while (fromIndex < array.length - 1) {
if (array[fromIndex]> number && lowerMax > array[fromIndex]) {
lowerMaxIndex = fromIndex;
}
fromIndex ++;
}
return lowerMaxIndex;
}
public static int arrayToInteger(Integer[] array) {
int number = 0;
for (int i = 0; i < array.length; i++) {
number+=array[i] * Math.pow(10, array.length-1-i);
}
return number;
}
这里是单元测试
@Test
public void nextHigherNumberTest() {
assertThat(ArrayUtils.nextHigherNumber(34722641), is(34724126));
assertThat(ArrayUtils.nextHigherNumber(123), is(132));
}
其他回答
至少，这里有几个基于字符串的暴力解决方案的例子，你应该能够马上想到:
38276中的数字排序为23678
38627排序的数字列表是23678
蛮力增量，排序和比较
沿着蛮力解决方案将转换为字符串 然后用这些数字强行找出所有可能的数字。
从它们中创建int，把它们放在一个列表中并排序， 获取目标条目之后的下一个条目。
如果你花了30分钟在这个问题上，却没有想出一个蛮力的方法，我也不会雇用你。
在商业世界中，一个不优雅、缓慢和笨拙但能完成工作的解决方案总是比没有解决方案更有价值，事实上，这几乎描述了所有不优雅、缓慢和笨拙的商业软件。
public static void findNext(long number){
/* convert long to string builder */
StringBuilder s = new StringBuilder();
s.append(number);
int N = s.length();
int index=-1,pivot=-1;
/* from tens position find the number (called pivot) less than the number in right */
for(int i=N-2;i>=0;i--){
int a = s.charAt(i)-'0';
int b = s.charAt(i+1)-'0';
if(a<b){
pivot = a;
index =i;
break;
}
}
/* if no such pivot then no solution */
if(pivot==-1) System.out.println(" No such number ")
else{
/* find the minimum highest number to the right higher than the pivot */
int nextHighest=Integer.MAX_VALUE, swapIndex=-1;
for(int i=index+1;i<N;i++){
int a = s.charAt(i)-'0';
if(a>pivot && a<nextHighest){
nextHighest = a;
swapIndex=i;
}
}
/* swap the pivot and next highest number */
s.replace(index,index+1,""+nextHighest);
s.replace(swapIndex,swapIndex+1,""+pivot);
/* sort everything to right of pivot and replace the sorted answer to right of pivot */
char [] sort = s.substring(index+1).toCharArray();
Arrays.sort(sort);
s.replace(index+1,N,String.copyValueOf(sort));
System.out.println("next highest number is "+s);
}
}
PHP实现
时间复杂度O(n)
$n = "9875";
$n_size = strlen($n);
for($i = $n_size-1; $i > 0; $i-- ) {
if($n[$i] > $n[$i-1]){
$temp = $n[$i];
$n[$i] = $n[$i-1];
$n[$i-1] = $temp;
break;
}
}
if($i == 0){
echo "Next Greater value no possible";
}else{
echo $n;
}
这里是Java实现
public static int nextHigherNumber(int number) {
Integer[] array = convertToArray(number);
int pivotIndex = pivotMaxIndex(array);
int digitInFirstSequence = pivotIndex -1;
int lowerDigitIndexInSecondSequence = lowerDigitIndex(array[digitInFirstSequence], array, pivotIndex);
swap(array, digitInFirstSequence, lowerDigitIndexInSecondSequence);
doRercursiveQuickSort(array, pivotIndex, array.length - 1);
return arrayToInteger(array);
}
public static Integer[] convertToArray(int number) {
int i = 0;
int length = (int) Math.log10(number);
int divisor = (int) Math.pow(10, length);
Integer temp[] = new Integer[length + 1];
while (number != 0) {
temp[i] = number / divisor;
if (i < length) {
++i;
}
number = number % divisor;
if (i != 0) {
divisor = divisor / 10;
}
}
return temp;
}
private static int pivotMaxIndex(Integer[] array) {
int index = array.length - 1;
while(index > 0) {
if (array[index-1] < array[index]) {
break;
}
index--;
}
return index;
}
private static int lowerDigitIndex(int number, Integer[] array, int fromIndex) {
int lowerMaxIndex = fromIndex;
int lowerMax = array[lowerMaxIndex];
while (fromIndex < array.length - 1) {
if (array[fromIndex]> number && lowerMax > array[fromIndex]) {
lowerMaxIndex = fromIndex;
}
fromIndex ++;
}
return lowerMaxIndex;
}
public static int arrayToInteger(Integer[] array) {
int number = 0;
for (int i = 0; i < array.length; i++) {
number+=array[i] * Math.pow(10, array.length-1-i);
}
return number;
}
这里是单元测试
@Test
public void nextHigherNumberTest() {
assertThat(ArrayUtils.nextHigherNumber(34722641), is(34724126));
assertThat(ArrayUtils.nextHigherNumber(123), is(132));
}
@BlueRaja算法的javascript实现。
var Bar = function(num){
num = num.toString();
var max = 0;
for(var i=num.length-2; i>0; i--){
var numArray = num.substr(i).split("");
max = Math.max.apply(Math,numArray);
if(numArray[0]<max){
numArray.sort(function(a,b){return a-b;});
numArray.splice(-1);
numArray = numArray.join("");
return Number(num.substr(0,i)+max+numArray);
}
}
return -1;
};
