Sliding window patterns

Sliding window is the two-pointer family member for contiguous subarrays/substrings. Two variants, one decision.

Fixed window

The window size k is given. Slide by adding the entering element and removing the leaving one — never recompute the whole window.

let sum = 0;
for (let i = 0; i < nums.length; i++) {
  sum += nums[i];
  if (i >= k) sum -= nums[i - k];
  if (i >= k - 1) best = Math.max(best, sum);
}

Dynamic window

The window grows on the right and shrinks from the left whenever an invariant breaks ("at most K distinct", "sum < target", "no repeats").

let left = 0;
for (let right = 0; right < s.length; right++) {
  add(s[right]);
  while (invariantBroken()) remove(s[left++]);
  best = Math.max(best, right - left + 1);
}

The while is the whole trick: each element enters once and leaves once, so it's O(n) even with the nested loop.

The three tell-tale signs

1. The answer is about a contiguous run — subarray, substring, never a subsequence
2. There's a constraint to maintain — at most K distinct, sum below X, no duplicates
3. The asked quantity is a max/min length or count over such runs

If "contiguous" is missing, it's probably prefix sums, DP, or sorting — not a window.