Using Chrome Puppeteer on a collaborative drawing board

To add some more fun at work, we have made it a tradition at Agoda to host a collaborative drawing board in the last week of the year.

So this time when I saw this message in our team chat, I quickly tried to find the piece of code I’d written the last time using Chrome Puppeteer.

I was annoyed by not finding it (again!), so this time I decided come what may, I will now

  1. Write a blog post about it
  2. Post the code to GitHub, so that I never lose it. (Though the piece of code is super trivial and takes only a few minutes to write, not being able to find when you need it is even more painful.)

So here we go!

Getting started

  1. I use VS Code which has integrated terminal support, using which I run yarn add puppeteer to install puppeteer. (At the time of writing this post the node version on my machine is 10.13.0)
  2. Once installed, you will see a package.json file will be created with puppeteer as a dependency.
  3. Next is to create an index.js file (which for now will have a very basic code using puppeteer) to load a page and then take a screenshot.
const puppeteer = require("puppeteer");
async function run() {
    const browser = await puppeteer.launch();
    const page = await browser.newPage();
    await page.goto("");
    await page.screenshot({ path: "google.png" });

4. Then we run this file by using node index.js

5. Once the run finishes, we can see VS Code has already identified a new file in the explorer google.png

Screenshot of the page taken using Chrome Puppeteer.

So now we have managed to get a Hello World kind of program up and running using chrome puppeteer.

Visual debugging

Before we continue, I am going to add a setting to the launch method of puppeteer called headless:false to enable visual debugging.

const browser = await puppeteer.launch({ headless: false });

Let’s begin by drawing a simple box

To draw a box, we need to select 4 (x,y) coordinates. For this example, here are the 4 coordinates I have selected — (50, 50) → (500, 50) → (500, 500) → (50, 500).

I will be using the mouse class to move and click at different coordinates. The mouse class operates in main-frame CSS pixels, relative to the top-left corner of the viewport.

I have also added a 500 ms delay between each click to visually see the box being drawn.

const puppeteer = require("puppeteer");
const delay = time => new Promise(res => setTimeout(res, 500));
async function run() {
const browser = await puppeteer.launch({ headless: false });
const page = await browser.newPage();
   await page.goto("xxx");
   await mouseClick(page, 50, 50);
   await mouseClick(page, 500, 50);
   await mouseClick(page, 500, 500);
   await mouseClick(page, 50, 500);
   await mouseClick(page, 50, 50);
async function mouseClick(page, x, y) {
   await delay();
   await, y);

Below is the output on running the above piece of code —

Drawing a basic box using chrome puppeteer

Now that we have got this working, let’s go ahead and add a little bit of complexity. We’ll modify the logic and keep drawing smaller inner boxes until we run out of space.

let x = 50;
let y = 500;
while (y > 0) {
   await mouseClick(page, x, x);
   await mouseClick(page, y, x);
   await mouseClick(page, y, y);
   await mouseClick(page, x, y);
   await mouseClick(page, x, x);
   x = x + 2;
   y = y - 2;
Drawing infinite boxes

And that’s just the beginning of the things one could do using Chrome puppeteer. As promised, posting a link to my code (and hope to add more examples) on Github down below —