Examples

Below are some example of how WCC is being used right now.

Table of contents

Server Rendering (SSR)

For the project Greenwood, WCC is used to provide a Next.js like experience by allowing users to author server-side routes using native custom elements! ✨

import '../components/card/card.js';

export default class ArtistsPage extends HTMLElement {
  async connectedCallback() {
    if (!this.shadowRoot) {
      const artists = await fetch('https://www.domain.com/api/artists')
        .then(resp => resp.json());
      const html = artists.map(artist => {
        return `
          <wc-card>
            <h2 slot="title">${artist.name}</h2>
            <img slot="image" src="${artist.imageUrl}" alt="${artist.name}"/>
          </wc-card>
        `;
      }).join('');

      this.attachShadow({ mode: 'open' });
      this.shadowRoot.innerHTML = html;
    }
  }
}

Serverless and Edge Functions

In the talk "Web Components at the Edge" for OpenJS World 2022, WCC was leveraged for all the AWS Lambda serverless function and Netlify Edge function demos. It also shows some clever ways to use WCC in more constrained runtime environments, like an edge runtime where something like fs might not be available. See all the code, slides and demos in GitHub. 🚀

import '../../node_modules/wc-compiler/src/dom-shim.js';

import Greeting from './components/greeting.js';

export default async function (request, context) {
  const countryCode = context.geo.country.code || 'UNKNOWN';
  const countryName = context.geo.country.name || 'UNKNOWN';
  const greeting = new Greeting(countryCode, countryName);

  greeting.connectedCallback();

  const response = new Response(`
    <!DOCTYPE html>
    <html lang="en">
      <body>
        <wc-greeting>
          ${greeting.getInnerHTML({ includeShadowRoots: true })}
          <details slot="details">
            <pre>
              ${JSON.stringify(context.geo)}
            </pre>
          </details>
        </wc-greeting>
      </body>
    </html>
  `);

  response.headers.set('content-type', 'text/html');

  return response;
}

Static Sites (SSG)

Using innerHTML, custom elements can be authored to not use Shadow DOM, which can be useful for a Layout or App component where that top level content specifically should not be rendered in a shadow root, e.g. <template> tag. What's nice about WCC is that by using innerHTML or attachShadow, you can opt-in to either on a per component basis, like is being done for the WCC website. In this case, the content is authored in markdown, but the layout, header, navigation, and footer are all custom elements rendered to static HTML. 🗒️

// layout.js
import './components/footer.js';
import './components/header.js';

class Layout extends HTMLElement {
  connectedCallback() {
    this.innerHTML = `
      <style>
        :root {
          --accent: #367588;
        }

        body {
          display: flex;
          flex-direction: column;
        }

        main {
          max-width: 1200px;
          margin: 20px auto;
        }

        a:visited {
          color: var(--accent);
        }
      </style>

      <wcc-header></wcc-header>

      <main>
        <slot name="content"></slot>
      </main>

      <wcc-footer></wcc-footer>
    `;
  }
}

export default Layout;

HTML (Light DOM) Web Components

As detailed in this excellent blog post, HTML Web Components are a strategy for transcluding content into the Light DOM of a custom element instead of (or in addition to) setting attributes. This can be useful for providing a set of styles to a block of content.

So instead of setting attributes:

<picture-frame img="/path/to/image.png" title="My Image"></picture-frame>

Pass HTML as children:

<picture-frame>
  <h3>My Image<h3>
  <img src="/path/to/image.png" alt="My Image">
</picture-frame>

With a custom element definition like so:

export default class PictureFrame extends HTMLElement {
  connectedCallback() {
    this.innerHTML = `
      <div class="picture-frame">
        ${this.innerHTML}
      </div>
    `;
  }
}

customElements.define('picture-frame', PictureFrame);

Progressive Hydration

Using the metadata information from a custom element with the hydrate=true attribute, you can use use the metadata with an IntersectionObserver to progressively load a custom element. In this case, handler.js builds SliderComponent from HTML and not only uses the hydrate attribute and metadata for lazy hydration, but also passes in the animated color via a CSS custom property set at build time! 🤯

See it in action here by scrolling to the bottom of the page and seeing the animation happen! View the code here in GitHub.

// slider.js
const template = document.createElement('template');

template.innerHTML = `
  <style>
    h6 {
      color: var(--color-secondary);
      font-size: 25px;
    }

    h6.hydrated {
      animation-duration: 3s;
      animation-name: slidein;
    }

    @keyframes slidein {
      from {
        margin-left: 100%;
        width: 300%;
      }

      to {
        font-size: 25px;
      }
    }
  </style>

  <h6>This is a slider component.</h6>
`;

class SliderComponent extends HTMLElement {
  connectedCallback() {
    if (!this.shadowRoot) {
      this.attachShadow({ mode: 'open' });
      this.shadowRoot.appendChild(template.content.cloneNode(true));
    } else {
      const header = this.shadowRoot.querySelector('h6');

      header.style.color = this.getAttribute('color');
      header.classList.add('hydrated');
    }
  }
}

export { SliderComponent };
export default SliderComponent;

customElements.define('wc-slider', SliderComponent);

// handler.js
import { renderFromHTML } from 'wc-compiler';

export async function handler() {
  const { html, metadata } = await renderFromHTML(`
    <wc-slider color="var(--color-accent)">
    </wc-slider>
  `);
  const lazyJs = [];

  for (const asset in metadata) {
    const a = metadata[asset];

    a.tagName = asset;

    if (a.moduleURL.href.endsWith('.js')) {
      if (a.hydrate === 'lazy') {
        lazyJs.push(a);
      }
    }
  }

  return {
    status: 200,
    headers: {
      'cache-control': 'no-cache, no-store, must-revalidate, max-age=0, s-maxage=0',
      'content-type': 'text/html; charset=utf8'
    },
    body: `
      <!DOCTYPE html>
      <html>
        <head>
          <style>
            :root, :host {
              --color-primary: rgb(12, 36, 42);
              --color-secondary: rgb(110, 176, 6);
              --color-tertiary: rgb(110, 176, 6);
              --color-accent: rgb(250, 217, 28);
            }
          </style>
          ${
            lazyJs.map(script => {
              return `
                <script type="module">
                  let initialized = false;

                  window.addEventListener('load', () => {
                    const options = {
                      root: null,
                      rootMargin: '20px',
                      threshold: 1.0
                    }

                    const callback = (entries, observer) => {
                      entries.forEach(entry => {
                        if(!initialized && entry.isIntersecting) {
                          initialized = true;
                          import('${script.moduleURL.pathname.replace(process.cwd(), '')}')
                        }
                      });
                    }

                    const observer = new IntersectionObserver(callback, options);
                    const target = document.querySelector('${script.tagName}');

                    observer.observe(target);
                  })
                </script>
              `;
            }).join('\n')
          }
        </head>
        <body>
          ${html}
        </body>
      </html>
    `
  };
}

JSX

A couple examples of using WCC + JSX are available for reference and reproduction:

Both of these examples can compile JSX for the client or the server using Greenwood, and can even be used with great testing tools like @web/test-runner! 💪