BinaryHeap

A generic binary heap implementation providing both MinHeap and MaxHeap variants with efficient priority-based operations.

Installation

npm

npm install @msnkr/data-structures

pnpm

pnpm install @msnkr/data-structures

yarn

yarn add @msnkr/data-structures

Deno (JSR)

import { MinHeap, MaxHeap } from 'jsr:@mskr/data-structures';

Browser (CDN)

<script type="module">
  import { MinHeap, MaxHeap } from 'https://esm.sh/jsr/@mskr/data-structures';
</script>

Usage

import { MinHeap, MaxHeap } from '@msnkr/data-structures';

const minHeap = new MinHeap<number>();
const maxHeap = new MaxHeap<number>();

API Reference

Properties

• size: number - Number of elements in the heap
• isEmpty(): boolean - Whether the heap is empty

Methods

Basic Operations

// Insert element - O(log n)
heap.insert(5);

// Remove root element - O(log n)
const root = heap.remove();

// Peek at root element - O(1)
const top = heap.peek();

Performance

Insert and remove operations are O(log n), with O(1) access to the minimum (MinHeap) or maximum (MaxHeap) element.

Searching and Utility

// Check if element exists - O(n)
const exists = heap.contains(42);

// Remove all elements - O(1)
heap.clear();

// Convert to array (level-order) - O(n)
const array = heap.toArray();

Iteration

// Level-order traversal
for (const value of heap) {
  console.log(value);
}

Examples

MinHeap Usage

const minHeap = new MinHeap<number>();

// Insert elements
minHeap.insert(5);
minHeap.insert(3);
minHeap.insert(7);
minHeap.insert(1);

console.log(minHeap.peek()); // 1 (minimum element)
console.log(minHeap.remove()); // 1
console.log(minHeap.peek()); // 3 (new minimum)
console.log(minHeap.size); // 3

MaxHeap Usage

const maxHeap = new MaxHeap<number>();

// Insert elements
maxHeap.insert(5);
maxHeap.insert(3);
maxHeap.insert(7);
maxHeap.insert(10);

console.log(maxHeap.peek()); // 10 (maximum element)
console.log(maxHeap.remove()); // 10
console.log(maxHeap.peek()); // 7 (new maximum)

Efficient Heap Construction

// Create a heap from an array - O(n)
const minHeap = new MinHeap<number>(null, [5, 3, 8, 1, 7, 4]);
console.log(minHeap.peek()); // 1 (minimum element)
console.log(minHeap.size); // 6

// Similarly for MaxHeap
const maxHeap = new MaxHeap<number>(null, [5, 3, 8, 1, 7]);
console.log(maxHeap.peek()); // 8 (maximum element)
console.log(maxHeap.size); // 5

Efficient Initialization

Building a heap from an array uses an O(n) heapify algorithm, which is significantly faster than inserting elements one by one (O(n log n)).

Custom Comparator for Objects

interface Person {
  name: string;
  age: number;
}

// Min-heap by age
const byAge = (a: Person, b: Person) => a.age - b.age;
const minHeap = new MinHeap<Person>(byAge);

minHeap.insert({ name: 'Alice', age: 25 });
minHeap.insert({ name: 'Bob', age: 20 });
minHeap.insert({ name: 'Charlie', age: 30 });

console.log(minHeap.peek()); // { name: "Bob", age: 20 }

Type-Safe Comparable Objects

class ComparablePerson implements Comparable<ComparablePerson> {
  constructor(public name: string, public age: number) {}

  compareTo(other: ComparablePerson): number {
    return this.age - other.age;
  }
}

const heap = new MinHeap<ComparablePerson>();
heap.insert(new ComparablePerson('Alice', 25));
heap.insert(new ComparablePerson('Bob', 20));

console.log(heap.peek()); // ComparablePerson { name: "Bob", age: 20 }

Error Handling

import { EmptyStructureError } from '@msnkr/data-structures';

try {
  const empty = new MinHeap<number>();
  empty.remove(); // Throws EmptyStructureError
} catch (error) {
  if (error instanceof EmptyStructureError) {
    console.log('Heap is empty!');
  }
}

try {
  const empty = new MaxHeap<number>();
  empty.peek(); // Throws EmptyStructureError
} catch (error) {
  if (error instanceof EmptyStructureError) {
    console.log('Cannot peek at empty heap!');
  }
}

Empty Heap

remove() and peek() throw EmptyStructureError when called on an empty heap.

Performance Characteristics

Operation	Time Complexity	Description
insert()	O(log n)	Add element to heap
remove()	O(log n)	Remove root element
peek()	O(1)	View root element
contains()	O(n)	Search for element
toArray()	O(n)	Convert to array
clear()	O(1)	Remove all elements
Construction	O(n)	Build heap from array

Space Complexity: O(n) where n is the number of elements

Implementation Details

Array-Based Storage

The heap is implemented as a complete binary tree stored in an array. For any node at index i:

• Left child: 2i + 1
• Right child: 2i + 2
• Parent: ⌊(i - 1) / 2⌋

       1
      / \
     3   2
    / \
   5   4

Array: [1, 3, 2, 5, 4]
Indices: 0  1  2  3  4

Heap Property

MinHeap: Parent ≤ Children (root is minimum) MaxHeap: Parent ≥ Children (root is maximum)

Heapify Algorithm

When initializing with an array, the heap uses Floyd's heapify algorithm:

• Starts from the last non-leaf node
• Works backwards, "sifting down" each node
• Time complexity: O(n) - more efficient than n insertions (O(n log n))

When to Use BinaryHeap

Perfect for:

• Finding min/max element efficiently
• Priority queue implementations
• K largest/smallest elements problems
• Heap sort algorithm
• Median finding (two-heap approach)
• Streaming data with top-k queries

MinHeap vs MaxHeap

Choose based on your needs:

• MinHeap: Quick access to minimum element
• MaxHeap: Quick access to maximum element
• Can't efficiently access both min and max simultaneously

Comparison with PriorityQueue

Feature	BinaryHeap	PriorityQueue
API	insert(), remove()	enqueue(), dequeue()
Use Case	Low-level heap operations	Queue-like priority handling
Abstraction	Direct heap manipulation	Higher-level queue interface

Which to Use?

• Use PriorityQueue for task scheduling, event queues (queue semantics)
• Use BinaryHeap for algorithms requiring direct heap operations (heap sort, etc.)

See Also

Guides:

• Heap Algorithms - K-largest elements, median finder, and more

Related Data Structures:

• PriorityQueue - Higher-level priority queue built on heap
• Queue - First-In-First-Out (FIFO) queue