Basic sorting-based problem

When a problem is described only as sorting-based, clarify what relation sorting should expose: equal values becoming adjacent, intervals sharing a timeline, or a target sum becoming monotonic. Preserve original indices by sorting pairs if the answer needs them, and sort a copy if mutation is disallowed. After sorting, explain the second phase rather than stopping after the sort. For example, duplicate removal scans runs, interval merging maintains a result tail whose end covers all processed overlaps, and a sorted two-sum scan moves left when the sum is too small and right when it is too large. Test duplicate keys, negative values, already-sorted input, and an answer using endpoints. Comparison sorting costs O(n log n) time, after which the scan is O(n); additional space is O(n) if copying or pairing indices. If values have a small integer range, discuss counting sort only when its O(n plus range) memory tradeoff is acceptable.