vantezzen / infinite-craft-solver

🕵️‍♂️ Find the quickest way to craft an item in Infinite Craft

Infinite Craft Solver

This project allows searching for the quickest way to craft a given item in the game Infinite Craft.

Visit https://ics.vantezzen.io to use the solver.

Dataset

Internally, this project uses a dataset of over 200k recipes fetched from the worker (see packages/worker). Feel free to download and use the dataset for your own projects.

To use the dataset, download the compressed recipe list from https://icscdn.vantezzen.io/recipes.json. You can use the function in apps/web/lib/compression.ts to decompress the file and get an array of object in the format { first: string, second: string, result: string }. As used by Infinite Craft, first and second are sorted alphabetically to allow for easier searching.

Compression

The dataset uses a simple compression method to reduce repetition of strings. For this, the compression file is a JSON object with the array "items" containing all unique item names and the array "recipes" containing all recipes. Each recipe is an array of three numbers, the first two being the index of the first and second item in the "items" array and the third being the index of the result item in the "items" array.

With this, the dataset can be uncompressed by iterating over the "recipes" array and using the numbers to get the item names from the "items" array:

function decompressRecipes({
  items,
  recipes,
}: {
  items: string[];
  recipes: any[];
}): Recipe[] {
  return recipes.map((recipe) => ({
    first: items[recipe[0]]!,
    second: items[recipe[1]]!,
    result: items[recipe[2]]!,
  }));
}

const recipes = decompressRecipes(compressedData);

Development

The project is built using TurboRepo and pnpm. To start the development server, run the following commands:

pnpm install

# Setup prisma
pnpm dlx turbo db:generate db:push

# Run the worker to fetch recipes
pnpm dlx turbo worker:run

# Start the frontend
pnpm dlx turbo dev

# Build production
pnpm dlx turbo build

# Run worker in background
pm2 start "pnpm dlx turbo worker:run"

# When running on a server, use xvfb to run the worker
sudo apt-get install xvfb
xvfb-run --auto-servernum pnpm dlx turbo worker:run
pm2 start "xvfb-run --auto-servernum pnpm dlx turbo worker:run" --name ics-worker

The project needs a running PostgreSQL database to work. To set this up, create .env in packages/db and insert the configuration for a PostgreSQL database.