var Component=(()=>{var m=Object.create;var n=Object.defineProperty;var x=Object.getOwnPropertyDescriptor;var y=Object.getOwnPropertyNames;var w=Object.getPrototypeOf,I=Object.prototype.hasOwnProperty;var b=(r,e)=>()=>(e||r((e={exports:{}}).exports,e),e.exports),C=(r,e)=>{for(var a in e)n(r,a,{get:e[a],enumerable:!0})},h=(r,e,a,o)=>{if(e&&typeof e=="object"||typeof e=="function")for(let i of y(e))!I.call(r,i)&&i!==a&&n(r,i,{get:()=>e[i],enumerable:!(o=x(e,i))||o.enumerable});return r};var z=(r,e,a)=>(a=r!=null?m(w(r)):{},h(e||!r||!r.__esModule?n(a,"default",{value:r,enumerable:!0}):a,r)),S=r=>h(n({},"__esModule",{value:!0}),r);var d=b((j,s)=>{s.exports=_jsx_runtime});var T={};C(T,{default:()=>g,frontmatter:()=>k});var t=z(d());var l=MDXCodeBlock;var c=`/** * @param {Array} arr The input integer array to be sorted. * @return {Array} */ export default function heapSort(arr) { // Begin by building a max heap. const size = arr.length; for (let i = Math.floor(size / 2 - 1); i >= 0; i--) { // Start with the index of the last parent node. // heapify: Swaps parent with child as long as child is larger than parent. heapify(arr, size, i); } // Iterate through the heap backwards, swapping the last element of the heap with the max element (the root of a max heap). // Max elements swapped to the end constitute the sorted part of the array (ignored in the next iteration by "i--"). for (let i = size - 1; i >= 0; i--) { [arr[0], arr[i]] = [arr[i], arr[0]]; // Build a max heap again in preparation for the swap in the next iteration. heapify(arr, i, 0); } return arr; } function heapify(arr, size, parentIdx) { let largest = parentIdx; // Initiate largest value's index with parent index. const leftChildIdx = 2 * parentIdx + 1; // Calculate index of left child. const rightChildIdx = 2 * parentIdx + 2; // Calculate index of right child. // Set \`largest\` to index with highest value between parent, left and right child. // See if left child of parent exists and is larger than parent. if (leftChildIdx < size && arr[leftChildIdx] > arr[largest]) { largest = leftChildIdx; } // See if right child of parent exists and is larger than parent. if (rightChildIdx < size && arr[rightChildIdx] > arr[largest]) { largest = rightChildIdx; } // If \`largest\` is not the current parent, swap positions with the current parent. if (largest !== parentIdx) { [arr[parentIdx], arr[largest]] = [arr[largest], arr[parentIdx]]; // Continue to recursively heapify the affected subtree. heapify(arr, size, largest); } } `;var p=`export default function heapSort(arr: Array): Array { // Begin by building a max heap. const size = arr.length; for (let i = Math.floor(size / 2 - 1); i >= 0; i--) { // Start with the index of the last parent node. // heapify: Swaps parent with child as long as child is larger than parent. heapify(arr, size, i); } // Iterate through the heap backwards, swapping the last element of the heap with the max element (the root of a max heap). // Max elements swapped to the end constitute the sorted part of the array (ignored in the next iteration by "i--"). for (let i = size - 1; i >= 0; i--) { [arr[0], arr[i]] = [arr[i], arr[0]]; // Build a max heap again in preparation for the swap in the next iteration. heapify(arr, i, 0); } return arr; } function heapify(arr: Array, size: number, parentIdx: number) { let largest = parentIdx; // Initiate largest value's index with parent index. const leftChildIdx = 2 * parentIdx + 1; // Calculate index of left child. const rightChildIdx = 2 * parentIdx + 2; // Calculate index of right child. // Set \`largest\` to index with highest value between parent, left and right child. // See if left child of parent exists and is larger than parent. if (leftChildIdx < size && arr[leftChildIdx] > arr[largest]) { largest = leftChildIdx; } // See if right child of parent exists and is larger than parent. if (rightChildIdx < size && arr[rightChildIdx] > arr[largest]) { largest = rightChildIdx; } // If \`largest\` is not the current parent, swap positions with the current parent. if (largest !== parentIdx) { [arr[parentIdx], arr[largest]] = [arr[largest], arr[parentIdx]]; // Continue to recursively heapify the affected subtree. heapify(arr, size, largest); } } `;var f=`function heapifyMin(arr, size, parentIdx) { let smallest = parentIdx; // initiate smallest value's index with parent index const leftChildIdx = 2 * parentIdx + 1; // calculate index of left child const rightChildIdx = 2 * parentIdx + 2; // calculate index of right child // set 'smallest' to index with lowest value between parent, left and right child if (leftChildIdx < size && arr[leftChildIdx] < arr[smallest]) { smallest = leftChildIdx; } if (rightChildIdx < size && arr[rightChildIdx] < arr[smallest]) { smallest = rightChildIdx; } // if 'smallest' is not the current parent, swap positions with the current parent if (smallest !== parentIdx) { [arr[parentIdx], arr[smallest]] = [arr[smallest], arr[parentIdx]]; // continue to recursively heapify the affected subtree heapifyMin(arr, size, smallest); } } `;var k=void 0;function u(r){let e={code:"code",h2:"h2",h3:"h3",li:"li",ol:"ol",p:"p",pre:"pre",ul:"ul",...r.components};return(0,t.jsxs)(t.Fragment,{children:[(0,t.jsx)(e.p,{children:"Heap sort is an in-place, comparison-based algorithm that works well for large sized arrays since it has a time complexity of O(nlog(n))."}),` `,(0,t.jsx)(e.p,{children:"It also does not require any additional memory space to sort the data as it is an in-place sorting algorithm. However, it is not a stable sort, which means that the order of elements with equal values may not be preserved."}),` `,(0,t.jsx)(e.h2,{children:"Clarification questions"}),` `,(0,t.jsx)(e.p,{children:"If unspecified:"}),` `,(0,t.jsxs)(e.ul,{children:[` `,(0,t.jsx)(e.li,{children:"Should the data be sorted in an ascending or descending order?"}),` `,(0,t.jsx)(e.li,{children:"Should the data be sorted in-place or is it acceptable to use additional data structures?"}),` `,(0,t.jsxs)(e.li,{children:["What kinds of inputs do we need to handle?",` `,(0,t.jsxs)(e.ul,{children:[` `,(0,t.jsx)(e.li,{children:"Will it just be an array of integers or should we handle other data types?"}),` `,(0,t.jsx)(e.li,{children:"Are there a large number of duplicate elements?"}),` `,(0,t.jsx)(e.li,{children:"How should negative numbers be handled?"}),` `]}),` `]}),` `]}),` `,(0,t.jsx)(e.h2,{children:"Solution"}),` `,(0,t.jsx)(e.p,{children:"Note: This question tackles in-place sorting for an output in ascending order. Refer to the 'Notes' section below on how to handle other cases."}),` `,(0,t.jsx)(l,{languages:{jsx:c,tsx:p}}),` `,(0,t.jsx)(e.h2,{children:"Edge cases"}),` `,(0,t.jsxs)(e.ul,{children:[` `,(0,t.jsx)(e.li,{children:"Non-integer element input: If the input array comprises of elements of different data types, you may need to return an error, settle for sorting on a best-effort basis, for e.g., only sorting a subset of the input data, or even define custom comparison functions for non-integers."}),` `,(0,t.jsx)(e.li,{children:"Single element / Empty input: Either return the original data structure, raise an error, or clarify required handling."}),` `,(0,t.jsx)(e.li,{children:"Already sorted input: You may want to detect sorting and return the input unchanged."}),` `,(0,t.jsx)(e.li,{children:"Invalid input: If the input is not an array, consider throwing an error."}),` `]}),` `,(0,t.jsx)(e.h2,{children:"Notes"}),` `,(0,t.jsxs)(e.ul,{children:[` `,(0,t.jsxs)(e.li,{children:["If asked to sort in descending order: Change comparison operators in ",(0,t.jsx)(e.code,{children:"heapify()"})," function in order to build a min heap instead of a max heap, then use it as per normal in ",(0,t.jsx)(e.code,{children:"heapSort"}),":"]}),` `]}),` `,(0,t.jsx)(l,{children:f}),` `,(0,t.jsxs)(e.ul,{children:[` `,(0,t.jsx)(e.li,{children:"If asked to use additional data structures or to write a pure function rather than sorting the data in-place, create a copy of the array and return a shallow copy of the sorted array:"}),` `]}),` `,(0,t.jsx)(e.pre,{children:(0,t.jsx)(e.code,{className:"language-js",children:`function heapSort(arr) { const result = arr.slice(0); const size = arr.length; for (let i = Math.floor(size / 2 - 1); i >= 0; i--) { heapify(result, size, i); } for (let i = size - 1; i >= 0; i--) { [result[0], result[i]] = [result[i], result[0]]; heapify(result, i, 0); } return result; } `})}),` `,(0,t.jsxs)(e.ul,{children:[` `,(0,t.jsxs)(e.li,{children:["If you're wondering why ",(0,t.jsx)(e.code,{children:"Math.floor(size / 2 - 1)"})," was able to calculate the index of the last parent node, it's because in a complete binary tree, the number of parent nodes is equal to Math.floor(size / 2). The -1 at the end is needed because the indices in the array start at 0, not 1."]}),` `]}),` `,(0,t.jsx)(e.h2,{children:"Big-O analysis"}),` `,(0,t.jsx)(e.p,{children:"Let's analyze the algorithm's time and space complexity."}),` `,(0,t.jsx)(e.h3,{children:"Time complexity"}),` `,(0,t.jsx)(e.p,{children:"The best, average and worst case time complexity of heap sort is O(nlog n). Regardless of whether the elements are already sorted, the algorithm will always take O(nlog(n)) time to complete. This is accurate unless you detect sorted arrays and return them immediately for a O(n) time complexity at the best case with a completely sorted array."}),` `,(0,t.jsxs)(e.p,{children:["To understand the time complexity of heap sort, understanding the time complexity of ",(0,t.jsx)(e.code,{children:"heapify()"})," is critical. In ",(0,t.jsx)(e.code,{children:"heapify()"}),", we walk through the tree from top to bottom, which means the time complexity is proportional to the height of the binary tree, which is O(log(n)) at most for a tree of size n. Therefore, time complexity for the ",(0,t.jsx)(e.code,{children:"heapify()"})," function is O(log(n))."]}),` `,(0,t.jsx)(e.p,{children:"The time complexity of heap sort is O(nlog(n)) as there are 2 steps to the algorithm: Building the heap and then sorting the data."}),` `,(0,t.jsxs)(e.ol,{children:[` `,(0,t.jsxs)(e.li,{children:["Building the heap has a time complexity of O(n) because the ",(0,t.jsx)(e.code,{children:"heapify()"})," method is called for each parent node backward, starting with the last node and ending at the tree root. On average, this can be done optimally at O(n)."]}),` `,(0,t.jsxs)(e.li,{children:["When sorting the data, we call the ",(0,t.jsx)(e.code,{children:"heapify()"})," method n-1 times to maintain the heap property on the unsorted part of the array. Hence, the total time complexity is O(nlog(n))."]}),` `]}),` `,(0,t.jsx)(e.p,{children:"The total time complexity of heap sort is the sum of the time complexity of each step, which is O(n) + O(nlog(n)) = O(nlog(n))."}),` `,(0,t.jsx)(e.h3,{children:"Space complexity"}),` `,(0,t.jsxs)(e.p,{children:["The array is sorted in place and requires no auxiliary storage proportional to its length. However, the shown recursive ",(0,t.jsx)(e.code,{children:"heapify()"})," uses O(log n) call-stack space in the worst case because it descends at most the height of the heap. An iterative ",(0,t.jsx)(e.code,{children:"heapify()"})," can reduce auxiliary space to O(1)."]})]})}function g(r={}){let{wrapper:e}=r.components||{};return e?(0,t.jsx)(e,{...r,children:(0,t.jsx)(u,{...r})}):u(r)}return S(T);})(); ;return Component;

Heap Sort

Examples