LeetCode 3487: Maximum Unique Subarray Sum After Deletion ·

Table of Contents

Meta Data
#

Difficulty: easy First Attempt: 2025-03-16

Submissions: Wrong Answer ×2 → Accepted (2025-07-25)

Approach
#

Any elements may be deleted (just not all), so any set of elements with distinct values can be left behind as a subarray. The problem becomes: pick distinct values with the largest sum.

Take one of every non-negative value (0 does not change the sum).
If everything is negative, something must remain, so keep the largest.

This is the “sort then scan” form from Greedy, without even needing the sort.

My Attempts
#

Version	Problem
Attempt 1	Collected only positives; with all negatives the set was empty and it returned 0. For `[-3]` the answer is -3
Attempt 2	Special-cased length 1, but `[-5, -1]` still returned 0 instead of -1
AC	When the set is empty, return the largest negative

Solution
#

class Solution:
    def maxSum(self, nums: List[int]) -> int:
        keep = set(x for x in nums if x >= 0)
        return sum(keep) if keep else max(nums)

Complexity
#

Time O(n), space O(n).

Mistakes to Avoid
#

The result cannot be empty: with all negatives, return the maximum, not 0.
Special-casing length 1 is not enough; the real check is whether any non-negative value exists.

Interview Explanation
#

Since deletion is free, any set of distinct values can become the subarray, so I add each non-negative value once. If there are none, one element must remain, so I return the largest negative. Time O(n).