A correct and based array implementation. 1-based indexing. obviously.
Inspired by Why you're wrong about 0-based indexing by @tjdevries
npm install --save based-array
yarn add based-array
pnpm add based-array
import { BasedArray } from "based-array";
const funkyReduce = (arr: number[]) => {
return arr.reduce((accum, item, index) => {
return accum + item * index;
}, 0);
};
let array = new BasedArray<number>(5).fill(1);
console.log(array[1]);
// 1
console.log(funkyReduce(array));
// 15
array = BasedArray.from([1, 2, 3, 4, 5]);
console.log(array[1]);
// 1
console.log(funkyReduce(array));
// 55
import { BasedArray } from "based-array";
let array = BasedArray.from([1, 2, 3, 4, 5]);
console.log(array[1]);
// 1
console.log(array[0]);
// undefined
// console.warn: you are counting from zero
BasedArray.strict();
array = BasedArray.from([1, 2, 3, 4, 5]);
console.log(array[1]);
// 1
console.log(array[0]);
// Throws error: CountingFromZeroError
Every based array method is tested against its vanilla counterpart in a thrilling batte over in tests/vanilla-vs.spec.
The example above is tested in tests/readme-example.spec.
You can run a simple performance test.
β based-array git:(main) β npm run perf
Starting performance test with 10,000 items.
Vanilla: 0.809ms
Based: 0.657ms
Vanilla: 0.524ms
Based: 0.548ms
Finished performance test.
No but honestly you shouldn't trust these perf test results, it makes no sense for it to be anything but slower than vanilla Array, maybe will get better tests to demonstrate when I'm bored again.