Why 1st code gives Memory Limit exceeded and 2nd code works fine???? This is question no 1855. Maximum Distance Between a Pair of Values of Leetcode. I think in both code space complexity should be constant but why first code is giving memory limit exceeded while second code works fine.

1.

class Solution {
public:
    int binarySearch(vector<int> nums, int l, int h, int target) {
        int res=-1;
        
        while(l<=h) {
            int mid = (l h)/2;
            if(nums[mid] >= target)
            {
                res=mid;
                l=mid 1;
            }
            else h=mid-1;
        }     
        return res;
    }
    int maxDistance(vector<int>& nums1, vector<int>& nums2) {
        int maxi = 0;
        
        for(int i=0; i<nums1.size(); i  ) {
            int ind = binarySearch(nums2, i, nums2.size()-1, nums1[i]);
            if(ind != -1) {
                maxi=max(ind-i, maxi);
            }
        }
        
        return maxi;
    }
};

2.

class Solution {
public:
    int maxDistance(vector<int>& nums1, vector<int>& nums2) {
        int maxi = 0;
        
        for(int i=0; i<nums1.size(); i  ) {
            int h=nums2.size()-1, l=i;
            while(l<=h) {
                int mid = (l h)/2;
                if(nums2\[mid\] >= nums1\[i\])
                {
                    maxi=max(mid-i, maxi);
                    l=mid 1;
                }
                else h=mid-1;
            }
        } 
        
        return maxi;
    }
};

CodePudding user response：

One of the possible reasons can be that in the first approach, the vector is passed by value. So whenever the function is called, a copy of the vector nums is created which is used further in the function. I think passing the vector by reference would not lead to MLE.