Description (leetcode)

You are given a sorted integer array nums and an integer k.

Return an array such that each distinct element appears at most k times, while preserving the relative order of the elements in nums.

Note: If a distinct element appears at least k times, then it must appear exactly k times in the resulting array.

Example 1:

Input: nums = [1,1,1,2,2,3], k = 2

Output: [1,1,2,2,3]

Explanation:

Each element can appear at most 2 times.

  • The element 1 appears 3 times, so only 2 occurrences are kept.
  • The element 2 appears 2 times, so both occurrences are kept.
  • The element 3 appears 1 time, so it is kept.

Thus, the resulting array is [1, 1, 2, 2, 3].

Example 2:

Input: nums = [1,2,3], k = 1

Output: [1,2,3]

Explanation:

All elements are distinct and already appear at most once, so the array remains unchanged.

Constraints:

  • 1 <= nums.length <= 100
  • 1 <= nums[i] <= 100
  • nums is sorted in non-decreasing order.
  • 1 <= k <= nums.length

Follow-up:

  • Can you solve this in-place using O(1) extra space?
  • Note that the space used for returning or resizing the result does not count toward the space complexity mentioned above, as some languages do not support in-place resizing.

submission

click to reveal
impl Solution {
    // this should be fairly idiomatic, in theory we could avoid realloc
    pub fn limit_occurrences(nums: Vec<i32>, k: i32) -> Vec<i32> {
        nums.chunk_by(|a, b| a == b)
            .flat_map(|chunk| chunk.iter().copied().take(k as usize))
            .collect()
    }
}