#13 Building a graph

December 02, 2022
See the code for this post on the build-graph branch.

We need a data structure allowing us to move from point A to point B. This data structure is called a graph.

Suppose we have a point on the road at coordinates 12, 40. Let's call this point a node. From this node, we want to be able to move to the adjacent road nodes. So every node in our data structure must include pointers to its neighbors.

When we drew our map in the previous post, we populated the points object, which had a key for every coordinate on the map. If a coordinate was a piece of road, we stored the corresponding rect element to it. If it was an obstacle, we set its value to false.

We can now build our graph using the points object. We start at the node 0,0. At every node, we explore all four adjacent coordinates. If we find a road node at that coordinate (by checking if the coordinate is truthy in points), we initialize that node as an empty object (if it doesn't exist yet), and we call the build function on that node. This way, our function recursively visits all road coordinates and populates the graph.

const build = async ({ x, y }) => {
visited[`${x}:${y}`] = true;
const currentNode = nodes[`${x}:${y}`];
await highlightVisited(`${x}:${y}`);
const neighbours = [
[x, y - 1],
[x + 1, y],
[x, y + 1],
[x - 1, y],
for (const [x, y] of neighbours) {
const coords = `${x}:${y}`;
if (points[coords]) {
nodes[coords] = nodes[coords] || {};
currentNode[coords] = nodes[coords];
if (!visited[coords]) await build({ x, y });
await build({ x: 0, y: 0 });

The video shows how the build algorithm traverses the points in the network. I've added some delay to animate this (hence the async/await in the code). At every node, I log the node object to the console.

The final product is the nodes object. It stores every node in our road network. If we now pick any coordinate on the map, we can explore its neighbors and get to any other point in the network.

See the code for this post on the build-graph branch.