• Liked Shashi Gowda
    keyboard_arrow_down

    What can you do with Virtual DOM on the server?

    Shashi Gowda
    Shashi Gowda
    schedule 1 year ago
    Sold Out!
    45 mins
    Demonstration
    Beginner
    Facebook's React pioneered the idea of the Virtual DOM - a clever trick which resulted in a framework that supports a declarative programming style, allows pleasant modularity, and surprising efficiency. This talk will explore the ability to represent a web application as the Virtual DOM on the server side. The Escher.jl Julia package is attempt to implement these ideas. Escher introduces two twists in the usual Virtual DOM story:
     
    1. Extend the Virtual DOM idea to HTML5 Custom Elements. In Escher, things like event listeners, entities that send messages over web sockets are custom elements. You can attach these to other elements to make them behave in interesting ways. Escher also comes out-of-the box with a rich library of pure functions that result in DOM nodes. These DOM nodes address various needs: Markdown, Vector Graphics (via Compose), Plots (via Gadfly), LaTeX, Layouts, Typography, Styles, Input Widgets, Clickable and Keyboard behaviors, even pages, tabs, menus, slideshows are all supported out-of-the box. This is done using Escher's custom elements (mostly the bits that support FRP) and the Polymer library (everything else, pretty much).  This library is entirely functional, and deals only with immutable values.
     
    2. Represent DOM on the server side: The Patchwork.jl package provides the ability to represent DOM on the server. It is essentially a mirror of a VDom node in virtual-dom - a pure JavaScript Virtual DOM library by Matt Esch. Escher sends the browser a JSON formatted Virtual DOM, and subsequently, sequences of patches sent as the UI needs to change.
     
    Escher works seamlessly with Reactive.jl - an FRP library derived from the Elm language's Signal library written for Julia. The result is a strangely beautiful pure Julia web programming experience which lets you do beautiful and bold visualizations of data, write interactive / explorable explanations, and teach better.
Sorry, no proposals found under this section.