Intersection Observer for Lazy-Loaded Images

IntersectionObserver is a browser API that allows you to detect when an element is visible in the window or within a scrollable element. Browser support is pretty decent with Safari being the main holdout at the time of this article, but there is a polyfill to support Apple users.

Packages using IntersectionObserver:

  • QuickLink - Adds a preload tag to visible links on the page.
  • Vanilla Lazy Load - Lazy loaded images.
  • Fireship.io - We use it to lazily load the comments at the bottom of the page.

Do not try to perform long-running or cpu intensive task in the observer’s event handler. It runs on the main thread and may block other important tasks in the event loop.

Lazy Loaded Images

Let’s imagine we have some images in an HTML document. Rather than define a src attribute (this will cause the browser to load the image immediately), let’s set a custom data attribute. We will read the data-lazy attribute when it becomes visible in the viewport, then use JavaScript to set the src.

file_type_html index.html

<!-- regular image -->
<img src="img/pig.jpeg">

<!-- our lazy image -->
<img data-lazy="img/cow.jpeg">

Scroll Version

The code below works, but it does not scale well on a page with many images. Scrolling fires off many events and the browser will need to recalculate every element in the DOM each time. This is very inefficient and will cause jank, especially on mobile and/or pages with iframes.

file_type_js bad.js
// 💩 Scroll Listener 

const targets = document.querySelectorAll('img');

window.addEventListener('scroll', (event) => {
    targets.forEach(img => {
        console.log('💩')
        const rect = img.getBoundingClientRect().top;
        if (rect <= window.innerHeight) {
            const src = img.getAttribute('data-lazy');
            img.setAttribute('src', src);
            img.classList.add('fade');
        }
    })
})

IntersectionObserver Version

The code below does not need to send an event on every scroll change. In addition, we do not need to perform any calculations because the isIntersecting value tells us whether or not the image is visible. Once the image is visible we can disconnect the observer to further optimize efficiency.

file_type_js good.js
const targets = document.querySelectorAll('img');

const lazyLoad = target => {
  const io = new IntersectionObserver((entries, observer) => {
    console.log(entries)
    entries.forEach(entry => {
      console.log('😍');

      if (entry.isIntersecting) {
        const img = entry.target;
        const src = img.getAttribute('data-lazy');

        img.setAttribute('src', src);
        img.classList.add('fade');

        observer.disconnect();
      }
    });
  });

  io.observe(target)
};

targets.forEach(lazyLoad);

Questions? Let's chat

Open Discord