Sort Algorithm Comparator (BETA)

This project provides two main features to help developers evaluate and benchmark sorting algorithms in JavaScript.

---

TL;DR

1. add import('sorting-benchmark') to the file that you wanna analyze
2. check ~/.js-sorting-logs/times.log file to see results

Feature 1: Enhanced Array Sorting

Once the project is integrated into your environment, it modifies the behavior of the native Array.prototype.sort() method. Instead of using the default sorting algorithm, your arrays will be sorted using various algorithms. The time taken for each algorithm will be logged into ~/.js-sorting-logs/times.log, allowing you to track the performance and determine the most efficient algorithm for your use case.

Feature 2: Benchmarking Sorting Algorithms

You can also benchmark multiple algorithms using either predefined or randomly generated datasets.

Data Structure

The data for benchmarking is stored in data.json. Each array object contains two arrays (e.g., firstArray and secondArray). The arrays are compared pairwise, where the first element of firstArray is compared with the first element of secondArray, and so on. Both arrays must have the same length, and each object requires a key.

Example:

{
  "firstArray": [
    [1, 2, 3],
    [4, 5, 6]
  ],
  "secondArray": [
    [1, 3, 2],
    [6, 5, 4]
  ]
}

Generating Test Data

To generate random datasets for testing, use:

npm run generateData

This will populate data.json with ordered, mostly ordered, and random datasets, preparing your project for benchmarking.

Running Benchmarks

Once the data is ready, run:

npm run benchmark

This will compare multiple sorting algorithms, running each algorithm 20 times (by default) on every array. The benchmark calculates the average execution time and displays the results in the terminal.

Known Bugs/Errors

1. Only works with numbers. Every custom sorting algorithm sorting strings differently. Send me an email if you wanna help about that topic.

TODO List

• Add repeat into cli parameters --repeat=20
• Add data into cli parameters --data=path/to/customData.json
• Select algorithms to run --allow=quicksort,timsort
• Select algorithms to disable --disable=quicksort,timsort
• Filter max best case complexity --bc=on2
• Filter max worst case complexity --wc=logn
• Filter max average case complexity --ac=on
• Select output file type --output=json
• Add memory usages into log --with-memory
• Filter minimum array length --minLength=100
• Filter maximum array length --minLength=10000
• Allow adjusting the log file location --log=path/to/logs
• Add visual graphs for benchmarking results