Backbone.Model
's are great at storing the state of your objects and persisting them back to your server. But as your Backbone.View
's become more complex, it's useful to have a model to store all view related attributes. These view attributes should be stored on a separate model than persistence attributes for 2 reasons:
- Separation of concerns
- If you store view attributes on the same model as your persistence model, when you call
.save()
on that model, the view attributes will be sent to the server too, eek!
Let's say you have a fairly standard Model and View:
// defining your classes
var Tweet = Backbone.Model.extend({});
var TweetView = Backbone.View.extend({});
// creating an instance of your model
var myTweet = new Tweet({text: "I love backbone!"});
You define a Backbone.ViewModel
class with a computed_attributes
object:
var TweetViewModel = Backbone.ViewModel.extend({
computed_attributes: {
"truncated_text" : function(){
return this.get("source_model").get("text").substring(0,10) + "…";
},
"escaped_text" : function(){
return encodeURIComponent(this.get("source_model").get("text"));
}
}
});
Initialize your ViewModel
and pass your persistence model as source_model
:
var myTweetViewModel = new TweetViewModel({
source_model: myTweet
});
Now, pass your ViewModel
to your View
:
var myTweetView = new TweetView({
model: myTweetViewModel
});
When your ViewModel
is initialized or whenever any attribute on your source_model
changes, all of the computed_attributes
will be processed and set
on your ViewModel
:
// The view attributes are set on the ViewModel
myTweetViewModel.get("truncated_text") // => "I love bac…"
// They can be used easily in your View Template
{{ truncated_text }} <a href="#">View more</a>
If your computed_attributes
depend on multiple source models, you can initialize your ViewModel
with a source_models
attribute that contains a mapping of attribute to model pairs:
var myOldestTweet = new Tweet({text: "Just joined twitter!"});
var myNewestTweet = new Tweet({text: "I love backbone!"});
var TweetSummaryViewModel = Backbone.ViewModel.extend({
computed_attributes: {
alpha_omega: function(){
return this.get("source_models").oldestTweet.get("text") + "…" + this.get("source_models").newestTweet.get("text");
}
}
});
var myTweetSummary = new TweetSummaryViewModel({
source_models: {
oldestTweet: myOldestTweet,
newestTweet: myNewestTweet
}
});
As with the single source_model
approach the computed_attributes
will be processed when the ViewModel
is created, and when any of the source_models
change.
console.log(myTweetSummary.get("alpha_omega")); // Prints 'Just joined twitter!…I love backbone!'.
myNewestTweet.set({text: "Taking a lunch break"});
console.log(myTweetSummary.get("alpha_omega")); // Prints 'Just joined twitter!…Taking a lunch break'.
To install, include the src/view-model.js
file in your HTML page, after Backbone and it's dependencies.
This project uses QUnit for it's automated tests.
You can run the automated tests in one of two ways:
-
Open the following files in your browser:
backbone-view-model/test/basic.html
andbackbone-view-model/test/require-js.html
. -
Karma: Right now this is broken. TODO - can someone help fix this?
- Make sure the tests are green.
- Add tests for your new features/fixes, so I don't break them in the future.
- Add documentation to the README, so people know about your new feature.