🏔️ Heaps (Priority Queues)#

Min-heap: parent ≤ children. O(log n) push/pop. O(1) peek. C++ priority_queue defaults to max-heap — use greater<int> for min-heap.

---

Core Patterns#

Pattern	Approach	Example
Top K largest	Min-heap of size K	Keep K largest, peek = Kth
Top K smallest	Max-heap of size K	Keep K smallest
K-way merge	Push (val, list, idx)	Merge K sorted lists
Running median	Max-heap (lower) + min-heap (upper)	Median of data stream

Templates#

// Min-heap
priority_queue<int, vector<int>, greater<int>> minH;

// Top-K largest elements (min-heap of size K)
priority_queue<int, vector<int>, greater<int>> pq;
for (int x : nums) {
    pq.push(x);
    if ((int)pq.size() > k) pq.pop();
}
// pq.top() = Kth largest

// Custom comparator (min-heap by first element of pair)
auto cmp = [](pair<int,int>& a, pair<int,int>& b){ return a.first > b.first; };
priority_queue<pair<int,int>, vector<pair<int,int>>, decltype(cmp)> pq(cmp);

Complexity#

Operation	Time
Push / Pop	O(log n)
Peek (top)	O(1)
Build heap from array	O(n)

---