There are N children standing in a line. Each child is assigned a rating value.
You are giving candies to these children subjected to the following requirements:
- Each child must have at least one candy.
- Children with a higher rating get more candies than their neighbors.
What is the minimum candies you must give?
由上述分析可以看出,我们需要记录递减子序列的长度descLen,以及子序列开头第一个人的candy数量descBegCan,在判断当前属于递减子序列的人的candy数时,我们并不能仅仅给当前candy总数加上这个长度,因为可能递减子序列的开头那个人(仅仅是开头这个人)本身已经有很多candy了,那么它不需要加1(就像上面判断到第五六个人3,2时,我们并不需要给第四个人4的candy数加1,因为它已经有了3个了,3>2>1 成立),而要看当前的长度descLen与descBegCan的大小关系,如果长度小于开头那个人的数量,那么仅加上长度减1 即可;如果二者相等,那么需要加上长度。
代码如下:
class Solution {
public:
int candy(vector<int> &ratings) {
int res=0;
int lastRat = -1, lastCan = 0, can = 0;
int descLen = 0, descBegCan=0; //连续递减子序列的长度和递减序列头部的值
for(int i=0; i<ratings.size(); ++i){
if(ratings[i]>lastRat){
can = lastCan+1;
descLen = 0;
}else if(ratings[i]==lastRat){ //相邻且rating相等的人candy可以不一样
can = 1;
descLen = 0;
}else{
can = 1;
if(descLen==0)
descBegCan = lastCan;
descLen++;
if(can==lastCan){
if(descLen<descBegCan)
res += descLen-1;
else if(descLen==descBegCan){
res += descLen;
descBegCan++;
}
}
}
res += can;
lastRat = ratings[i];
lastCan = can;
}
return res;
}
};
本方法的空间复杂度为O(1),时间复杂度为O(n),应该属于最好的方法之一了。