Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
SproutCore 2.0
JavaScript

This branch is 2 commits ahead, 8812 commits behind emberjs:master

Fetching latest commit…

Cannot retrieve the latest commit at this time

Failed to load latest commit information.
docs
generators
packages
vendor
.gitignore
Gemfile
README.md
Rakefile
VERSION
list_file_sizes.rb
package.json
tests.html

README.md

SproutCore

SproutCore is a JavaScript framework that does all of the heavy lifting that you'd normally have to do by hand. There are tasks that are common to every web app; SproutCore does those things for you, so you can focus on building killer features and UI.

These are the three features that make SproutCore a joy to use:

  1. Bindings
  2. Computed properties
  3. Auto-updating templates

Bindings

Use bindings to keep properties between two different objects in sync. You just declare a binding once, and SproutCore will make sure changes get propagated in either direction.

Here's how you create a binding between two objects:

MyApp.president = SC.Object.create({
  name: "Barack Obama"
});

MyApp.country = SC.Object.create({
  // Ending a property with 'Binding' tells SproutCore to
  // create a binding to the presidentName property.
  presidentNameBinding: 'MyApp.president.name'
});

MyApp.country.get('presidentName');
// "Barack Obama"

Bindings allow you to architect your application using the MVC (Model-View-Controller) pattern, then rest easy knowing that data will always flow correctly from layer to layer.

Computed Properties

Computed properties allow you to treat a function like a property:

MyApp.president = SC.Object.create({
  firstName: "Barack",
  lastName: "Obama",

  fullName: function() {
    return this.get('firstName') + ' ' + this.get('lastName');

    // Call this flag to mark the function as a property
  }.property()
});

MyApp.president.get('fullName');
// "Barack Obama"

Treating a function like a property is useful because they can work with bindings, just like any other property.

Many computed properties have dependencies on other properties. For example, in the above example, the fullName property depends on firstName and lastName to determine its value. You can tell SproutCore about these dependencies like this:

MyApp.president = SC.Object.create({
  firstName: "Barack",
  lastName: "Obama",

  fullName: function() {
    return this.get('firstName') + ' ' + this.get('lastName');

    // Tell SproutCore that this computed property depends on firstName
    // and lastName
  }.property('firstName', 'lastName')
});

Make sure you list these dependencies so SproutCore knows when to update bindings that connect to a computed property.

Auto-updating Templates

SproutCore uses Handlebars, a semantic templating library. To take data from your JavaScript application and put it into the DOM, create a <script> tag and put it into your HTML, wherever you'd like the value to appear:

<script type="text/html">
  The President of the United States is {{MyApp.president.fullName}}.
</script>

Here's the best part: templates are bindings-aware. That means that if you ever change the value of the property that you told us to display, we'll update it for you automatically. And because you've specified dependencies, changes to those properties are reflected as well.

Hopefully you can see how all three of these powerful tools work together: start with some primitive properties, then start building up more sophisticated properties and their dependencies using computed properties. Once you've described the data, you only have to say how it gets displayed once, and SproutCore takes care of the rest. It doesn't matter how the underlying data changes, whether from an XHR request or the user performing an action; your user interface always stays up-to-date. This eliminates entire categories of edge cases that developers struggle with every day.

Getting Started

For new users, we recommend downloading the SproutCore Starter Kit, which includes everything you need to get started.

We also recommend that you check out the annotated Todos example, which shows you the best practices for architecting an MVC-based web application.

To learn more about what we're up to, follow @sproutcore on Twitter, subscribe to the blog, or read the original SproutCore 2.0 announcement.

Building SproutCore 2.0

  1. Run rake to build SproutCore. Two builds will be placed in the dist/ directory.
    • sproutcore.js and sproutcore.min.js - unminified and minified builds of SproutCore 2.0
    • sproutcore-datastore.js and sproutcore-datastore.min.js - unminified and minified of the data store framework

If you are building under Linux, you will need a JavaScript runtime for minification. You can either install nodejs or gem install therubyracer.

How to Run Unit Tests

  1. Install Ruby and Rubygems. There are many resources on the web can help you to do so, one of the best ways may be rvm. (WARNING: There is a bug in RubyGems 1.7.2, please upgrade to 1.8.5+ or downgrade.)

  2. Install Bundler via gem install bundler

  3. Run bundle install to pull in Ruby dependencies.

  4. To start the development web server, run bpm autocompile or bpm preview (depending on your version of BPM).

  5. Then visit: http://localhost:4020/tests.html?package=PACKAGE_NAME. Replace 'PACKAGE_NAME' with the name of the package you want to run. For example:

Adding New Packages

Be sure you include the new package as a dependency in the global package.json and run spaderun update.

Note that unless you are adding new tests or adding a new package you should not need to run spaderun update.

Something went wrong with that request. Please try again.