1. Volto Client - Server duality – Effective Volto – Architecture

1. Volto Client - Server duality#

Volto is based on the following three main components of the overall architecture:

  • Plone (and Zope) backend server

  • Volto Nodejs SSR server

  • the Single Page Application JavaScript code that runs in the browser

1.1. The Plone backend#

To be able to work as a Volto backend, a Plone website needs to activate its plone.restapi addon, which exposes several JSON-api based endpoints.

1.2. Volto Nodejs server#

Volto includes a Server-Side Rendering Nodejs-based HTTP server which runs the same code that's served to the Client (the browsers) and renders it to a single HTML string that can be served as the initial loaded page. This type of application is called an isomorphic application and, in Volto, is built on top of the Razzle framework.

So Volto websites work (in basic mode) even with JavaScript disabled. All the HTML received by browser should contain the markup needed to render all the content on the page.

This server process is based on the popular Express framework.

1.3. Volto - the SPA client bundle#

Once the HTML is loaded by the browser, the Volto JavaScript client bundle is interpreted and executed. Volto, the React app is then created in the browser memory. From now on, the browser will no longer fetch HTML from the server, but instead communicate via JSON with Plone, the backend server. The JSON-based communication protocol, all the available endpoints, etc. are provided by the plone.restapi package.

Sometimes view components have code that fetches content via async network calls. In case that content should be rendered together with the rest of the generated HTML, you need to follow the Integrate with Volto’s asyncConnect for SSR chapter.