This repo contains the utility library to create and run Advent of Code solutions.
To create the AoC solutions project run (requires Node 16 LTS or higher: v16.13.0+
):
npx aocrunner init
It will guide you through the configuration with simple CLI menu.
git init
) (optional)AOC_SESSION_KEY
to the .env
file (you can find it in cookie file when you sign in at adventofcode.com) (optional)src/template
(optional)start <day_number>
command with your package manager, for example:npm start 1
// or
yarn start 1
// or
pnpm start 1
You can join my private leaderboard for JS/TS programmers:
Code:
107172-b51ab08f
AoC Runner respects the concerns of the AoC creator, and does not send unnecessary requests. In fact, it reduces the number of requests sent to the AoC server when compared to doing things manually:
Starting from version 1.7.0 AoC Runner sets the correct request header as requested by AoC creator. If you use an older version, please upgrade.
This library creates modern, ESM compatible project - that means that you have to specify the extensions in your imports (that are not imported from node_modules
).
Always use .js
extension for custom imports, even when importing .ts
files (TypeScript ignores them, but the compiled code relies on them).
Examples:
// correct:
import _ from "lodash"
import myLib from "../utils/myLib.js"
import { myUtil } from "../utils/index.js"
// incorrect:
import _ from "lodash"
import myLib from "../utils/myLib.ts"
import { myUtil } from "../utils/index.ts"
// also incorrect:
import _ from "lodash"
import myLib from "../utils/myLib"
import { myUtil } from "../utils"
run
functionThe run
function takes care of reading the output and supplying it to the solution functions, executes tests and measures your code performance.
It takes an object that describes your solutions, and optionally, as a second argument - custom path to the input file.
The solution description have four keys (each of them is optional):
{
part1: {/* tests and the solution function for part one of the puzzle */},
part2: {/* tests and the solution function for part two of the puzzle */},
trimTestInputs: true, // boolean switch for preparing test inputs
onlyTests: false, // boolean switch to run tests only - useful for debugging
}
part1
and part2
Both part1
and part2
keys accept an object in format:
{
tests: [{ input: `some test input`, expected: "expected value"}], // optional
solution: part1, // required
}
tests
key accepts an array of tests (you can add as many as you want), each test takes input
and expected
result, you can also provide an optional name
for each test that will be displayed when the tests are executed.
solution
key accepts a function that takes the raw input as an argument and returns the solution as string
/ number
/ bigint
(return value will be converted to string before sending a solution)
trimTestInputs
Usually, the test inputs will be some multiline strings, so we can provide them using template literals. Be careful - any white spaces / indents within the template literal are a part of the string, so without trimTestInputs: true
you have to format your test input accordingly.
Let's say that we have this test input:
foo
bar
baz
With trimTestInputs: false
we cannot have any indents (unless they are a part of the input):
/* Correct */
run({
part1: {
tests: [
{
input: `foo
bar
baz`,
expected: "my-result",
},
],
solution: part1,
},
trimTestInputs: false,
})
/* Incorrect */
run({
part1: {
tests: [
{
input: `foo
bar
baz`,
expected: "my-result",
},
],
solution: part1,
},
trimTestInputs: false,
})
/* Also incorrect */
run({
part1: {
tests: [
{
input: `
foo
bar
baz
`,
expected: "my-result",
},
],
solution: part1,
},
trimTestInputs: false,
})
That looks ugly, that's where trimTestInputs: true
come in handy - it will remove empty lines at the beginning and the end of the test input, and will remove first level of indentation.
With trimTestInputs: true
:
/* Still correct */
run({
part1: {
tests: [
{
input: `foo
bar
baz`,
expected: "my-result",
},
],
solution: part1,
},
trimTestInputs: true,
})
/* Incorrect - first line has no indentation */
run({
part1: {
tests: [
{
input: `foo
bar
baz`,
expected: "my-result",
},
],
solution: part1,
},
trimTestInputs: true,
})
/* Now it works! All lines have the same level of indentation,
* so aocrunner will transform it into a correct input.
*/
run({
part1: {
tests: [
{
input: `
foo
bar
baz
`,
expected: "my-result",
},
],
solution: part1,
},
trimTestInputs: true,
})
onlyTests
With onlyTests
switch set to true
your solutions won't be run on the real input, only the test inputs. It can come in handy when you are using console.log
to print intermediate values when solving the puzzle (real input can be really big).
Default value is set to false
, so you can set onlyTests: true
in your code (or even your template) and switch between modes just by commenting and uncommenting this line.
Example:
run({
part1: {
tests: [
{
input: `some test input`,
expected: "some result",
},
],
solution: part1,
},
part2: {
tests: [
{
input: `some test input`,
expected: "some result",
},
],
solution: part2,
},
trimTestInputs: true,
onlyTests: true, // <- Yay! Comment this line to quickly switch mode.
})
Project is under open, non-restrictive ISC license.