Common Algos

Exercises, solutions, and benchmarks to common algorithms and whiteboarding questions.

Overview

Testing

Each problem has its own folder dedicated to it, along with a test suite to verify expected outcomes. This test file uses a custom describe method which will allow me to pass in an arbitrary number of methods from the index file (imported as import * as ...) and execute the same suite of tests across each exported method without having to duplicate tests.

$ npm test reversestring

 PASS  exercises/reversestring/test.mjs

reversestring
    Naive Reverse
      √ reverses a string (2 ms)
      √ reverses a string with whitespace
    for loop
      √ reverses a string
      √ reverses a string with whitespace (1 ms)
    for..of loop
      √ reverses a string
      √ reverses a string with whitespace
    arr.reduce
      √ reverses a string
      √ reverses a string with whitespace

Test Suites: 1 passed, 1 total
Tests:       8 passed, 8 total
Snapshots:   0 total
Time:        0.558 s, estimated 1 s
Ran all test suites matching /reversestring/i.

Active Filters: filename /reversestring/

Benchmarking

Each exercise is also benchmarkable thanks to BenchmarkJS. Similar to the test suite setup, benchmarking is also set up to run benchmarks on all exported methods. The only restriction being that each exercise file must have an export const data with the argument value to be passed to the methods being benchmarked.

Optionally, each method can have a descriptive testname property which will be used in the benchmarking output. If not present, the function name will be used instead.

$ npm run bench reversestring

  ┌ Running "reversestring" benchmarks...
  ├─ Naive Reverse x 3,072,077 ops/sec ±0.31% (94 runs sampled)
  ├─ for loop x 11,899,016 ops/sec ±0.77% (90 runs sampled)
  ├─ for..of loop x 9,729,595 ops/sec ±1.82% (86 runs sampled)
  ├─ arr.reduce x 6,032,047 ops/sec ±0.76% (91 runs sampled)
  └ Done!

Fastest is "for loop"