backbone-fundamentals

A creative-commons book on Backbone.js for beginners and advanced users alike

6 years after

Developing Backbone.js Applications

A book on Backbone.js targeted at beginners and advanced users alike

This is the home of 'Developing Backbone.js Applications', an (in-progress) book about the Backbone.js framework for structuring JavaScript applications. It is released under a CC-license.

I am quite pleased that this book will be out in physical form in a few months time via O'Reilly Media. Readers will have the option of purchasing the latest version in either print or a number of digital formats then or can grab a copy for free from this repository.

Corrections to existing material are always welcome and I hope that together we can provide the community with an up-to-date resource that is of help.

My extended thanks to these members of the community for their assistance tweaking the project.

Index

Introduction

As JavaScript developers, we are at an interesting point in time where not only do we have mature solutions to help organize the JavaScript powering our applications based on a separation of concerns, but developers looking to build non-trivial projects are almost spoiled for choice for frameworks that can help structure their applications.

Maturity in software (framework) development isn't simply about how long a framework has been around. It's about how solid the framework is and more importantly how well it's evolved to fill its role. Has it become more effective at solving common problems? Does it continue to improve as developers build larger and more complex applications with it?

In this book, I will be covering the popular Backbone.js, which I consider the best of the current family of JavaScript architectural frameworks.

Topics will include MVC theory and how to build applications using Backbone's models, views, collections and routers. I'll also be taking you through advanced topics like modular development with Backbone.js and AMD (via RequireJS), how to build applications using modern software stacks (like Node and Express), how to solve the routing problems with Backbone and jQuery Mobile, tips about scaffolding tools, and a lot more.

If this is your first time looking at Backbone.js and you're still unsure whether or not to give it a try, why not take a look at how a Todo application can be implemented in Backbone and several other popular Javascript frameworks before reading further?

The goal of this book is to create an authoritative and centralized repository of information that can help those developing real-world apps with Backbone. If you come across a section or topic which you think could be improved or expanded on, please feel free to submit a pull-request. It won't take long and you'll be helping other developers avoid problems you've run into before.

Fundamentals

In this section we are going to cover the context into which a framework like Backbone.js fits. Let's begin our journey into understanding Backbone better with a look at code architecture.

MVC, MVP & Backbone.js

Before exploring any JavaScript frameworks that assist in structuring applications, it can be useful to gain a basic understanding of architectural design patterns. Design patterns are proven solutions to common development problems and can suggest structural approaches to help guide developers in adding some organization to their applications.

Patterns are useful because they're a set of practices that build upon the collective experience of skilled developers who have repeatedly solved similar problems. Although developers 10 or 20 years ago may not have been using the same programming languages when implementing patterns in their projects, there are many lessons we can learn from their efforts.

In this section, we're going to review two popular patterns - MVC and MVP. We'll be exploring in greater detail how Backbone.js implements these patterns shortly to better appreciate where it fits in.

MVC

MVC (Model-View-Controller) is an architectural design pattern that encourages improved application organization through a separation of concerns. It enforces the isolation of business data (Models) from user interfaces (Views), with a third component (Controllers) traditionally present to manage logic, user-input and the coordination of models and views. The pattern was originally designed by Trygve Reenskaug while working on Smalltalk-80 (1979), where it was initially called Model-View-Controller-Editor. MVC was described in depth in “Design Patterns: Elements of Reusable Object-Oriented Software” (The "GoF" or “Gang of Four” book) in 1994, which played a role in popularizing its use.

Smalltalk-80 MVC

It's important to understand what the original MVC pattern was aiming to solve as it has changed quite heavily since the days of its origin. Back in the 70's, graphical user-interfaces were far and few between. An approach known as Separated Presentation began to be used as a means to make a clear division between domain objects which modeled concepts in the real world (e.g a photo, a person) and the presentation objects which were rendered to the user's screen.

The Smalltalk-80 implementation of MVC took this concept further and had an objective of separating out the application logic from the user interface. The idea was that decoupling these parts of the application would also allow the reuse of models for other interfaces in the application. There are some interesting points worth noting about Smalltalk-80's MVC architecture:

  • A Domain element was known as a Model and were ignorant of the user-interface (Views and Controllers)
  • Presentation was taken care of by the View and the Controller, but there wasn't just a single view and controller. A View-Controller pair was required for each element being displayed on the screen and so there was no true separation between them
  • The Controller's role in this pair was handling user input (such as key-presses and click events), doing something sensible with them.
  • The Observer pattern was relied upon for updating the View whenever the Model changed

Developers are sometimes surprised when they learn that the Observer pattern (nowadays commonly implemented as a Publish/Subscribe system) was included as a part of MVC's architecture decades ago. In Smalltalk-80's MVC, the View and Controller both observe the Model: anytime the Model changes, the Views react. A simple example of this is an application backed by stock market data - for the application to show real-time information, any change to the data in its Models should result in the View being refreshed instantly.

Martin Fowler has done an excellent job of writing about the origins of MVC over the years and if you are interested in further historical information about Smalltalk-80's MVC, I recommend reading his work.

MVC As We Know It

We've reviewed the 70's, but let us now return to the here and now. The MVC pattern has been applied to a diverse range of programming languages. For example, the popular Ruby on Rails is an implementation of a web application framework based on MVC for the Ruby language. JavaScript now has a number of MVC frameworks, including Ember.js, JavaScriptMVC, and of course Backbone.js. Given the importance of avoiding "spaghetti" code, a term which describes code that is very difficult to read or maintain due to its lack of structure, let's look at what the MVC pattern enables the Javascript developer to do.

MVC is composed of three core components:

Models

Models manage the data for an application. They are concerned with neither the user-interface nor presentation layers, but instead represent structured data that an application may require. When a model changes (e.g when it is updated), it will typically notify its observers (e.g views, a concept we will cover shortly) that a change has occurred so that they may react accordingly.

To understand models better, let us imagine we have a JavaScript photo gallery application. In a photo gallery, a photo would merit its own model, as it represents a unique kind of domain-specific data. The Photo model may represent attributes such as a caption, image source and additional meta-data. A specific photo would be stored in an instance of a model. Here's an example of a simple Photo model implemented with Backbone.js:

var Photo = Backbone.Model.extend({

    // Default attributes for the photo
    defaults: {
      // Ensure that each photo created has an `src`.
      src: "placeholder.jpg",
      caption: "A default image",
      viewed: false
    },

    initialize: function() {
    }

});

The built-in capabilities of models vary across frameworks, however it's common for them to support validation of attributes, where attributes represent the properties of the model, such as a model identifier. When using models in real-world applications we generally also need a way of persisting models. Persistence allows us to edit and update models with the knowledge that their most recent states will be saved somewhere, for example in a web browser's localStorage data-store or synchronized with a database.

A model may also have multiple views observing it. Imagine our Photo model contained meta-data such as the longitude and latitude where the photo was taken, a list of people present in the photo, and a list of tags. A developer could create a single view that displayed all these attributes, or might create three separate views to display each attribute. The important detail is that the Photo model doesn't care how these views are organized, it simply announces updates to its data as necessary. We'll come back to Views in more detail later.

It is not uncommon for modern MVC/MV* frameworks to provide a means to group models together. In Backbone, these groups are called "Collections". Managing models in groups allows us to write application logic based on notifications from the group, should any model it contains change. This avoids the need to manually observe individual model instances.

Here's how we might group Photo models into a simplified Backbone Collection:

var PhotoGallery = Backbone.Collection.extend({

    // Reference to this collection's model.
    model: Photo,

    // Filter down the list of all photos that have been viewed
    viewed: function() {
      return this.filter(function(photo){ return photo.get('viewed'); });
    },

    // Filter down the list to only photos that have not yet been viewed
    unviewed: function() {
      return this.without.apply(this, this.viewed());
    }

});

If you read older texts on MVC, you may come across a description of models as also managing application 'state'. In JavaScript applications "state" has a specific meaning, typically referring to the current "state" of a view or sub-view on a user's screen at a fixed time. State is a topic which is regularly discussed when looking at Single-page applications, where the concept of state needs to be simulated.

Views

Views are a visual representation of models that present a filtered view of their current state. A view typically observes a model and is notified when the model changes, allowing the view to update itself accordingly. Design pattern literature commonly refers to views as 'dumb', given that their knowledge of models and controllers in an application is limited.

Users interact with views, which usually means reading and editing model data. For example, in our photo gallery application example, model viewing might happen in a user interface with a big image, a caption, and a list of tags. Model editing could be done through an "edit" view where a user who has selected a specific photo could edit its caption, tags, or other metadata in a form.

In MVC, the actual task of updating the Model falls to Controllers, which we'll be covering shortly.

Let's explore Views a little further using a simple JavaScript example. Below we can see a function that creates a single Photo view, consuming both a model instance and a controller instance.

We define a render() utility within our view which is responsible for rendering the contents of the photoModel using a JavaScript templating engine (Underscore templating) and updating the contents of our view, referenced by photoEl.

The photoModel then adds our render() callback as one of its subscribers, so that through the Observer pattern it can trigger the view to update when the model changes.

You may wonder where user interaction comes into play here. When users click on any elements within the view, it's not the view's responsibility to know what to do next. A Controller makes this decision. In our sample implementation, this is achieved by adding an event listener to photoEl which will delegate handling the click behavior back to the controller, passing the model information along with it in case it's needed.

The benefit of this architecture is that each component plays its own separate role in making the application function as needed.

var buildPhotoView = function( photoModel, photoController ){

    var base        = document.createElement('div'),
        photoEl     = document.createElement('div');

     base.appendChild(photoEl);

     var render= function(){
        // We use a templating library such as Underscore
        // templating which generates the HTML for our 
        // photo entry
        photoEl.innerHTML = _.template('photoTemplate', {src: photoModel.getSrc()});
     }

     photoModel.addSubscriber( render );

     photoEl.addEventListener('click', function(){
        photoController.handleEvent('click', photoModel );
     });

     var show = function(){
        photoEl.style.display  = '';
     }

     var hide = function(){
        photoEl.style.display  = 'none';
     }

     return{
        showView: show,
        hideView: hide
     }

}

Templating

In the context of JavaScript frameworks that support MVC/MV*, it is worth looking more closely at JavaScript templating and its relationship to Views.

It has long been considered bad practice (and computationally expensive) to manually create large blocks of HTML markup in-memory through string concatenation. Developers using this technique often find themselves iterating through their data, wrapping it in nested divs and using outdated techniques such as document.write to inject the 'template' into the DOM. This approach often means keeping scripted markup inline with standard markup, which can quickly become difficult to read and maintain, especially when building large applications.

JavaScript templating libraries (such as Handlebars.js or Mustache) are often used to define templates for views as HTML markup containing template variables. These template blocks can be either stored externally or within script tags with a custom type (e.g 'text/template'). Variables are deliminated using a variable syntax (e.g {{name}}). Javascript template libraries typically accept data in JSON, and the grunt work of populating templates with data is taken care of by the framework itself. This has a several benefits, particularly when opting to store templates externally as this can let applications load templates dynamically on an as-needed basis.

Let's compare two examples of HTML templates. One is implemented using the popular Handlebars.js library, and the other uses Underscore's 'microtemplates'.

Handlebars.js:

<li class="photo">
  <h2>{{caption}}</h2>
  <img class="source" src="{{src}}"/>
  <div class="meta-data"> 
    {{metadata}}
  </div>
</li>

Underscore.js Microtemplates:

<li class="photo">
  <h2><%= caption %></h2>
  <img class="source" src="<%= src %>"/>
  <div class="meta-data"> 
    <%= metadata %>
  </div>
</li>

You may also use double curly brackets (i.e {{}}) (or any other tag you feel comfortable with) in Microtemplates. In the case of curly brackets, this can be done by setting the Underscore templateSettings attribute as follows:

_.templateSettings = { interpolate : /\{\{(.+?)\}\}/g };

A note on navigation and state

It is also worth noting that in classical web development, navigating between independent views required the use of a page refresh. In single-page JavaScript applications, however, once data is fetched from a server via Ajax, it can be dynamically rendered in a new view within the same page. Since this doesn't automatically update the URL, the role of navigation thus falls to a "router", which assists in managing application state (e.g allowing users to bookmark a particular view they have navigated to). As routers are however neither a part of MVC nor present in every MVC-like framework, I will not be going into them in greater detail in this section.

Controllers

Controllers are an intermediary between models and views which are classically responsible for two tasks: they both update the view when the model changes and update the model when the user manipulates the view.

In our photo gallery application, a controller would be responsible for handling changes the user made to the edit view for a particular photo, updating a specific photo model when a user has finished editing.

It's with controllers that most JavaScript MVC frameworks depart from this interpretation of the MVC pattern. The reasons for this vary, but in my opinion, Javascript framework authors likely initially looked at server-side interpretations of MVC (such as Ruby on Rails), realized that that approach didn't translate 1:1 on the client-side, and so re-interpreted the C in MVC to solve their state management problem. This was a clever approach, but it can make it hard for developers coming to MVC for the first time to understand both the classical MVC pattern and the "proper" role of controllers in other non-Javascript frameworks.

So does Backbone.js have Controllers? Not really. Backbone's Views typically contain "controller" logic, and Routers (discussed below) are used to help manage application state, but neither are true Controllers according to classical MVC.

In this respect, contrary to what might be mentioned in the official documentation or in blog posts, Backbone is neither a truly MVC/MVP nor MVVM framework. It's in fact better to see it a member of the MV family which approaches architecture in its own way. There is of course nothing wrong with this, but it is important to distinguish between classical MVC and MV should you be relying on discussions of MVC to help with your Backbone projects.

Controllers in Spine.js vs Backbone.js

Spine.js

We now know that controllers are traditionally responsible for updating the view when the model changes (and similarly the model when the user updates the view). Since Backbone doesn't have its own explicit controllers, it's useful to review the controller from another MVC framework to appreciate the difference in implementations. Let's take a look at Spine.js:

In this example, we're going to have a controller called PhotosController which will be in charge of individual photos in the application. It will ensure that when the view updates (e.g a user edited the photo meta-data) the corresponding model does too.

(Note: We won't be delving heavily into Spine.js beyond this example, but it's worth looking at it to learn more about Javascript frameworks in general.)

// Controllers in Spine are created by inheriting from Spine.Controller

var PhotosController = Spine.Controller.sub({      
  init: function(){
    this.item.bind("update", this.proxy(this.render));
    this.item.bind("destroy", this.proxy(this.remove));
  },

  render: function(){
    // Handle templating
    this.replace($("#photoTemplate").tmpl(this.item));
    return this;
  },

  remove: function(){
    this.el.remove();
    this.release();
  }
});

In Spine, controllers are considered the glue for an application, adding and responding to DOM events, rendering templates and ensuring that views and models are kept in sync (which makes sense in the context of what we know to be a controller).

What we're doing in the above example is setting up listeners in the update and destroy events using render() and remove(). When a photo entry gets updated, we re-render the view to reflect the changes to the meta-data. Similarly, if the photo gets deleted from the gallery, we remove it from the view. In case you were wondering about the tmpl() function in the code snippet: in the render() function, we're using this to render a JavaScript template called #photoTemplate which simply returns a HTML string used to replace the controller's current element.

What this provides us with is a very lightweight, simple way to manage changes between the model and the view.

Backbone.js

Later on in this section we're going to revisit the differences between Backbone and traditional MVC, but for now let's focus on controllers.

In Backbone, controller logic is shared between Backbone.View and Backbone.Router. Earlier releases of Backbone contained something called Backbone.Controller, but it was renamed to Router to clarify its role.

A Router's main purpose is to translate URL requests into application states. When a user browses to the URL www.example.com/photos/42, a Router could be used to show the photo with that ID, and to define what application behavior should be run in response to that request. Routers can contain traditional controller responsibilities, such as binding the events between models and views, or rendering parts of the page. However, Backbone contributor Tim Branyen has pointed out that it's possible to get away without needing Backbone.Router at all for this, so a way to think about it using the Router paradigm is probably:

var PhotoRouter = Backbone.Router.extend({
  routes: { "photos/:id": "route" },

  route: function(id) {
    var item = photoCollection.get(id);
    var view = new PhotoView({ model: item });

    something.html( view.render().el );
  }
}):

What does MVC give us?

To summarize, the separation of concerns in MVC facilitates modularization of an application's functionality and enables:

  • Easier overall maintenance. When updates need to be made to the application it is clear whether the changes are data-centric, meaning changes to models and possibly controllers, or merely visual, meaning changes to views.
  • Decoupling models and views means that it's straight-forward to write unit tests for business logic
  • Duplication of low-level model and controller code is eliminated across the application
  • Depending on the size of the application and separation of roles, this modularity allows developers responsible for core logic and developers working on the user-interfaces to work simultaneously

Delving deeper

Right now, you likely have a basic understanding of what the MVC pattern provides, but for the curious, we'll explore it a little further.

The GoF (Gang of Four) do not refer to MVC as a design pattern, but rather consider it a "set of classes to build a user interface". In their view, it's actually a variation of three other classical design patterns: the Observer (Pub/Sub), Strategy and Composite patterns. Depending on how MVC has been implemented in a framework, it may also use the Factory and Decorator patterns. I've covered some of these patterns in my other free book, JavaScript Design Patterns For Beginners if you would like to read into them further.

As we've discussed, models represent application data, while views handle what the user is presented on screen. As such, MVC relies on Pub/Sub for some of its core communication (something that surprisingly isn't covered in many articles about the MVC pattern). When a model is changed it "publishes" to the rest of the application that it has been updated. The "subscriber"--generally a Controller--then updates the view accordingly. The observer-viewer nature of this relationship is what facilitates multiple views being attached to the same model.

For developers interested in knowing more about the decoupled nature of MVC (once again, depending on the implementation), one of the goals of the pattern is to help define one-to-many relationships between a topic and its observers. When a topic changes, its observers are updated. Views and controllers have a slightly different relationship. Controllers facilitate views to respond to different user input and are an example of the Strategy pattern.

Summary

Having reviewed the classical MVC pattern, your should now understand how it allows developers to cleanly separate concerns in an application. You should also now appreciate how JavaScript MVC frameworks may differ in their interpretation of MVC, and how they share some of the fundamental concepts of the original pattern.

When reviewing a new JavaScript MVC/MV* framework, remember - it can be useful to step back and consider how it's opted to approach Models, Views, Controllers or other alternatives, as this can better help you grok how the framework expects to be used.

MVP

Model-view-presenter (MVP) is a derivative of the MVC design pattern which focuses on improving presentation logic. It originated at a company named Taligent in the early 1990s while they were working on a model for a C++ CommonPoint environment. Whilst both MVC and MVP target the separation of concerns across multiple components, there are some fundamental differences between them.

For the purposes of this summary we will focus on the version of MVP most suitable for web-based architectures.

Models, Views & Presenters

The P in MVP stands for presenter. It's a component which contains the user-interface business logic for the view. Unlike MVC, invocations from the view are delegated to the presenter, which are decoupled from the view and instead talk to it through an interface. This allows for all kinds of useful things such as being able to mock views in unit tests.

The most common implementation of MVP is one which uses a Passive View (a view which is for all intents and purposes "dumb"), containing little to no logic. MVP models are almost identical to MVC models and handle application data. The presenter acts as a mediator which talks to both the view and model, however both of these are isolated from each other. They effectively bind models to views, a responsibility held by Controllers in MVC. Presenters are at the heart of the MVP pattern and as you can guess, incorporate the presentation logic behind views.

Solicited by a view, presenters perform any work to do with user requests and pass data back to them. In this respect, they retrieve data, manipulate it and determine how the data should be displayed in the view. In some implementations, the presenter also interacts with a service layer to persist data (models). Models may trigger events but it's the presenter's role to subscribe to them so that it can update the view. In this passive architecture, we have no concept of direct data binding. Views expose setters which presenters can use to set data.

The benefit of this change from MVC is that it increases the testability of your application and provides a more clean separation between the view and the model. This isn't however without its costs as the lack of data binding support in the pattern can often mean having to take care of this task separately.

Although a common implementation of a Passive View is for the view to implement an interface, there are variations on it, including the use of events which can decouple the View from the Presenter a little more. As we don't have the interface construct in JavaScript, we're using it more as more a protocol than an explicit interface here. It's technically still an API and it's probably fair for us to refer to it as an interface from that perspective.

There is also a Supervising Controller variation of MVP, which is closer to the MVC and MVVM patterns as it provides data-binding from the Model directly from the View. Key-value observing (KVO) plugins (such as Derick Bailey's Backbone.ModelBinding plugin) introduce this idea of a Supervising Controller to Backbone.

MVP or MVC?

MVP is generally used most often in enterprise-level applications where it's necessary to reuse as much presentation logic as possible. Applications with very complex views and a great deal of user interaction may find that MVC doesn't quite fit the bill here as solving this problem may mean heavily relying on multiple controllers. In MVP, all of this complex logic can be encapsulated in a presenter, which can simplify maintenance greatly.

As MVP views are defined through an interface and the interface is technically the only point of contact between the system and the view (other than a presenter), this pattern also allows developers to write presentation logic without needing to wait for designers to produce layouts and graphics for the application.

Depending on the implementation, MVP may be more easy to automatically unit test than MVC. The reason often cited for this is that the presenter can be used as a complete mock of the user-interface and so it can be unit tested independent of other components. In my experience this really depends on the languages you are implementing MVP in (there's quite a difference between opting for MVP for a JavaScript project over one for say, ASP.net).

At the end of the day, the underlying concerns you may have with MVC will likely hold true for MVP given that the differences between them are mainly semantic. As long as you are cleanly separating concerns into models, views and controllers (or presenters) you should be achieving most of the same benefits regardless of the pattern you opt for.

MVC, MVP and Backbone.js

There are very few, if any architectural JavaScript frameworks that claim to implement the MVC or MVP patterns in their classical form as many JavaScript developers don't view MVC and MVP as being mutually exclusive (we are actually more likely to see MVP strictly implemented when looking at web frameworks such as ASP.net or GWT). This is because it's possible to have additional presenter/view logic in your application and yet still consider it a flavor of MVC.

Backbone contributor Irene Ros subscribes to this way of thinking as when she separates Backbone views out into their own distinct components, she needs something to actually assemble them for her. This could either be a controller route (such as a Backbone.Router, covered later in the book) or a callback in response to data being fetched.

That said, some developers do however feel that Backbone.js better fits the description of MVP than it does MVC . Their view is that:

  • The presenter in MVP better describes the Backbone.View (the layer between View templates and the data bound to it) than a controller does
  • The model fits Backbone.Model (it isn't that different from the classical MVC "Model")
  • The views best represent templates (e.g Handlebars/Mustache markup templates)

A response to this could be that the view can also just be a View (as per MVC) because Backbone is flexible enough to let it be used for multiple purposes. The V in MVC and the P in MVP can both be accomplished by Backbone.View because they're able to achieve two purposes: both rendering atomic components and assembling those components rendered by other views.

We've also seen that in Backbone the responsibility of a controller is shared with both the Backbone.View and Backbone.Router and in the following example we can actually see that aspects of that are certainly true.

Here, our Backbone PhotoView uses the Observer pattern to 'subscribe' to changes to a View's model in the line this.model.bind('change',...). It also handles templating in the render() method, but unlike some other implementations, user interaction is also handled in the View (see events).

var PhotoView = Backbone.View.extend({

    //... is a list tag.
    tagName:  "li",

    // Pass the contents of the photo template through a templating
    // function, cache it for a single photo
    template: _.template($('#photo-template').html()),

    // The DOM events specific to an item.
    events: {
      "click img" : "toggleViewed"
    },

    // The PhotoView listens for changes to its model, re-rendering. Since there's
    // a one-to-one correspondence between a **Photo** and a **PhotoView** in this
    // app, we set a direct reference on the model for convenience.

    initialize: function() {
      _.bindAll(this, 'render');
      this.model.bind('change', this.render);
      this.model.bind('destroy', this.remove);
    },

    // Re-render the photo entry
    render: function() {
      $(this.el).html(this.template(this.model.toJSON()));
      return this;
    },

    // Toggle the `"viewed"` state of the model.
    toggleViewed: function() {
      this.model.viewed();
    }

});

Another (quite different) opinion is that Backbone more closely resembles Smalltalk-80 MVC, which we went through earlier.

As regular Backbone user Derick Bailey has written, it's ultimately best not to force Backbone to fit any specific design patterns. Design patterns should be considered flexible guides to how applications may be structured and in this respect, Backbone doesn't fit either MVC nor MVP perfectly. Instead, it borrows some of the best concepts from multiple architectural patterns and creates a flexible framework that just works well. Call it the Backbone way, MV* or whatever helps reference its flavor of application architecture.

It is however worth understanding where and why these concepts originated, so I hope that my explanations of MVC and MVP have been of help. Most structural JavaScript frameworks will adopt their own take on classical patterns, either intentionally or by accident, but the important thing is that they help us develop applications which are organized, clean and can be easily maintained.

Fast facts

Backbone.js

  • Core components: Model, View, Collection, Router. Enforces its own flavor of MV*
  • Good documentation, with more improvements on the way
  • Used by large companies such as SoundCloud and Foursquare to build non-trivial applications
  • Event-driven communication between views and models. As we'll see, it's relatively straight-forward to add event listeners to any attribute in a model, giving developers fine-grained control over what changes in the view
  • Supports data bindings through manual events or a separate Key-value observing (KVO) library
  • Great support for RESTful interfaces out of the box, so models can be easily tied to a backend
  • Extensive eventing system. It's trivial to add support for pub/sub in Backbone
  • Prototypes are instantiated with the new keyword, which some developers prefer
  • Agnostic about templating frameworks, however Underscore's micro-templating is available by default. Backbone works well with libraries like Handlebars
  • Doesn't support deeply nested models, though there are Backbone plugins such as this which can help
  • Clear and flexible conventions for structuring applications. Backbone doesn't force usage of all of its components and can work with only those needed.

The Basics

What is Backbone?

Backbone.js is one of a number of JavaScript frameworks for creating MVC-like web applications. On the front-end, it's my architectural framework of choice as it's both mature, relatively lightweight and can be easily tested using third-party toolkits such as Jasmine or QUnit. Other MVC frameworks you may be familiar with include Ember.js (SproutCore 2.0), Spine, YUILibrary and JavaScriptMVC.

Backbone is maintained by a number of contributors, most notably: Jeremy Ashkenas, creator of CoffeeScript, Docco and Underscore.js. As Jeremy is a believer in detailed documentation, there's a level of comfort in knowing you're unlikely to run into issues which are either not explained in the official docs or which can't be nailed down with some assistance from the #documentcloud IRC channel. I strongly recommend using the latter if you find yourself getting stuck.

Why should you consider using it?

Backbone's main benefits, regardless of your target platform or device, include helping:

  • Organize the structure to your application
  • Simplify server-side persistence
  • Decouple the DOM from your page's data
  • Model data, views and routers in a succinct manner
  • Provide DOM, model and collection synchronization

The Basics

In this section, you'll learn the essentials of Backbone's models, views, collections and routers, as well as about using namespacing to organize your code. This isn't meant as a replacement for the official documentation, but it will help you understand many of the core concepts behind Backbone before you start building applications with it.

  • Models
  • Collections
  • Routers
  • Views
  • Namespacing

Models

Backbone models contain interactive data for an application as well as the logic around this data. For example, we can use a model to represent the concept of a photo object including its attributes like tags, titles and a location.

Models can be created by extending Backbone.Model as follows:

var Photo = Backbone.Model.extend({
    defaults: {
        src: 'placeholder.jpg',
        title: 'an image placeholder',
        coordinates: [0,0]
    },
    initialize: function(){
        this.bind("change:src", function(){
            var src = this.get("src"); 
            console.log('Image source updated to ' + src);
        });
    },
    changeSrc: function( source ){
        this.set({ src: source });
    }
});

var somePhoto = new Photo({ src: "test.jpg", title:"testing"});
somePhoto.changeSrc("magic.jpg"); // which triggers "change:src" and logs an update message to the console.

Initialization

The initialize() method is called when a new instance of a model is created. Its use is optional, however you'll see why it's good practice to use it below.

var Photo = Backbone.Model.extend({
    initialize: function(){
        console.log('this model has been initialized');
    }
});

/*We can then create our own instance of a photo as follows:*/
var myPhoto = new Photo();

Getters & Setters

Model.get()

Model.get() provides easy access to a model's attributes. Attributes which are passed through to the model on instantiation are instantly available for retrieval.

var myPhoto = new Photo({ title: "My awesome photo", 
                          src:"boston.jpg", 
                          location: "Boston", 
                          tags:['the big game', 'vacation']}),

    title = myPhoto.get("title"), //My awesome photo
    location = myPhoto.get("location"), //Boston
    tags = myPhoto.get("tags"), // ['the big game','vacation']
    photoSrc = myPhoto.get("src"); //boston.jpg

Alternatively, if you wish to directly access all of the attributes in a model's instance directly, you can achieve this as follows:

var myAttributes = myPhoto.attributes;
console.log(myAttributes);

It is best practice to use Model.set() or direct instantiation to set the values of a model's attributes.

Accessing Model.attributes directly is generally discouraged. Instead, should you need to read or clone data, Model.toJSON() is recommended for this purpose. If you would like to access or copy a model's attributes for purposes such as JSON stringification (e.g. for serialization prior to being passed to a view), this can be achieved using Model.toJSON():

var myAttributes = myPhoto.toJSON();
console.log(myAttributes);
/* this returns { title: "My awesome photo", 
             src:"boston.jpg", 
             location: "Boston", 
             tags:['the big game', 'vacation']}*/

Model.set()

Model.set() allows us to pass attributes into an instance of our model. Attributes can either be set during initialization or at any time afterwards. It's important to avoid trying to set a Model's attributes directly (for example Model.caption = 'A new caption'). Backbone uses Model.set() to know when to broadcast that a model's data has changed.

var Photo = Backbone.Model.extend({
    initialize: function(){
        console.log('this model has been initialized');
    }
});

/*Setting the value of attributes via instantiation*/
var myPhoto = new Photo({ title: 'My awesome photo', location: 'Boston' });

var myPhoto2 = new Photo();

/*Setting the value of attributes through Model.set()*/
myPhoto2.set({ title:'Vacation in Florida', location: 'Florida' });

Default values

There are times when you want your model to have a set of default values (e.g. in a scenario where a complete set of data isn't provided by the user). This can be set using a property called defaults in your model.

var Photo = Backbone.Model.extend({
    defaults:{
        title: 'Another photo!',
        tags:  ['untagged'],
        location: 'home',
        src: 'placeholder.jpg'
    },
    initialize: function(){
    }
});

var myPhoto = new Photo({ location: "Boston", 
                          tags:['the big game', 'vacation']}),
    title   = myPhoto.get("title"), //Another photo!
    location = myPhoto.get("location"), //Boston
    tags = myPhoto.get("tags"), // ['the big game','vacation']
    photoSrc = myPhoto.get("src"); //placeholder.jpg

Listening for changes to your model

Any and all of the attributes in a Backbone model can have listeners bound to them which detect when their values change. Listeners can be added to the initialize() function:

this.bind('change', function(){
    console.log('values for this model have changed');
});

In the following example, we log a message whenever a specific attribute (the title of our Photo model) is altered.

var Photo = Backbone.Model.extend({
    defaults:{
        title: 'Another photo!',
        tags:  ['untagged'],
        location: 'home',
        src: 'placeholder.jpg'
    },
    initialize: function(){
        console.log('this model has been initialized');
        this.bind("change:title", function(){
            var title = this.get("title");
            console.log("My title has been changed to.. " + title);
        });
    },

    setTitle: function(newTitle){
        this.set({ title: newTitle });
    }
});

var myPhoto = new Photo({ title:"Fishing at the lake", src:"fishing.jpg"});
myPhoto.setTitle('Fishing at sea'); 
//logs My title has been changed to.. Fishing at sea

Validation

Backbone supports model validation through Model.validate(), which allows checking the attribute values for a model prior to them being set.

Validation functions can be as simple or complex as necessary. If the attributes provided are valid, nothing should be returned from .validate(). If they are invalid, a custom error can be returned instead.

A basic example for validation can be seen below:

var Photo = Backbone.Model.extend({
    validate: function(attribs){
        if(attribs.src === undefined){
            return "Remember to set a source for your image!";
        }
    },

    initialize: function(){
        console.log('this model has been initialized');
        this.bind("error", function(model, error){
            console.log(error);
        });
    }
});

var myPhoto = new Photo();
myPhoto.set({ title: "On the beach" });
//logs Remember to set a source for your image!

Views

Views in Backbone don't contain the markup for your application, but rather they are there to support models by defining the logic for how they should be represented to the user. This is usually achieved using JavaScript templating (e.g. Mustache, jQuery-tmpl, etc.). A view's render() function can be bound to a model's change() event, allowing the view to always be up to date without requiring a full page refresh.

Creating new views

Similar to the previous sections, creating a new view is relatively straight-forward. To create a new View, simply extend Backbone.View. I'll explain this code in detail below:

var PhotoSearch = Backbone.View.extend({
    el: $('#results'),
    render: function( event ){
        var compiled_template = _.template( $("#results-template").html() );
        this.el.html( compiled_template(this.model.toJSON()) );
        return this; //recommended as this enables calls to be chained.
    },
    events: {
        "submit #searchForm":  "search",
        "click .reset": "reset",
        "click .advanced": "switchContext"
    },
    search: function( event ){
        //executed when a form '#searchForm' has been submitted
    },
    reset: function( event ){
        //executed when an element with class "reset" has been clicked.
    },
    switchContext: function( event ){
        //executed when an element with class "advanced" has been clicked.
    }
});

What is el?

el is basically a reference to a DOM element and all views must have one. It allows for all of the contents of a view to be inserted into the DOM at once, which makes for faster rendering as browser performs the minimum required reflows and repaints.

There are two ways to attach a DOM element to a view: the element already exists in the page or a new element is created for the view and added manually by the developer. If the element already exists in the page, you can set el as either a CSS selector that matches the element or a simple reference to the DOM element.

el: '#footer', 
// OR
el: document.getElementById( 'footer' )

If you want to create a new element for you view, set any combination of the following view's properties: tagName, id and className. A new element will be created for you by the framework and a reference to it will be available at the el property.

tagName: 'p', // required, but defaults to 'div' if not set
className: 'container', // optional, you can assign multiple classes to this property like so 'container homepage'
id: 'header', // optional

The above code creates the DOMElement below but doesn't append it to the DOM.

<p id="header" class="container"></p>

Understanding render()

render() is an optional function that defines the logic for rendering a template. We'll use Underscore's micro-templating in these examples, but remember you can use other templating frameworks if you prefer.

The _.template method in Underscore compiles JavaScript templates into functions which can be evaluated for rendering. In the above view, I'm passing the markup from a template with id results-template to _.template() to be compiled. Next, I set the html of the el DOM element to the output of processing a JSON version of the model associated with the view through the compiled template.

Presto! This populates the template, giving you a data-complete set of markup in just a few short lines of code.

The events attribute

The Backbone events attribute allows us to attach event listeners to either custom selectors, or directly to el if no selector is provided. An event takes the form {"eventName selector": "callbackFunction"} and a number of event-types are supported, including click, submit, mouseover, dblclick and more.

What isn't instantly obvious is that under the bonnet, Backbone uses jQuery's .delegate() to provide instant support for event delegation but goes a little further, extending it so that this always refers to the current view object. The only thing to really keep in mind is that any string callback supplied to the events attribute must have a corresponding function with the same name within the scope of your view.

Collections

Collections are sets of Models and are created by extending Backbone.Collection.

Normally, when creating a collection you'll also want to pass through a property specifying the model that your collection will contain, as well as any instance properties required.

In the following example, we create a PhotoCollection that will contain our Photo models:

var PhotoCollection = Backbone.Collection.extend({
    model: Photo
});

Getters and Setters

There are a few different ways to retrieve a model from a collection. The most straight-forward is to use Collection.get() which accepts a single id as follows:

var skiingEpicness = PhotoCollection.get(2);

Sometimes you may also want to get a model based on its client id. The client id is a property that Backbone automatically assigns models that have not yet been saved. You can get a model's client id from its .cid property.

var mySkiingCrash = PhotoCollection.getByCid(456);

Backbone Collections don't have setters as such, but do support adding new models via .add() and removing models via .remove().

var a = new Backbone.Model({ title: 'my vacation'}),
    b = new Backbone.Model({ title: 'my holiday'});

var photoCollection = new PhotoCollection([a,b]);
photoCollection.remove([a,b]);

Listening for events

As collections represent a group of items, we're also able to listen for add and remove events for when new models are added or removed from the collection. Here's an example:

var PhotoCollection = new Backbone.Collection();
PhotoCollection.bind("add", function(photo) {
  console.log("I liked " + photo.get("title") + ' its this one, right? '  + photo.get("src"));
});

PhotoCollection.add([
  {title: "My trip to Bali", src: "bali-trip.jpg"},
  {title: "The flight home", src: "long-flight-oofta.jpg"},
  {title: "Uploading pix", src: "too-many-pics.jpg"}
]);

In addition, we're able to bind a change event to listen for changes to models in the collection.

PhotoCollection.bind("change:title", function(){
    console.log('there have been updates made to this collections titles');    
});

Fetching models from the server

Collections.fetch() retrieves a default set of models from the server in the form of a JSON array. When this data returns, the current collection's contents will be replaced with the contents of the array.

var PhotoCollection = new Backbone.Collection;
PhotoCollection.url = '/photos';
PhotoCollection.fetch();

Under the covers, Backbone.sync is the function called every time Backbone tries to read or save models to the server. It uses jQuery or Zepto's ajax implementations to make these RESTful requests, however this can be overridden as per your needs.

In the above example if we wanted to log an event when .sync() was called, we could do this:

Backbone.sync = function(method, model) {
  console.log("I've been passed " + method + " with " + JSON.stringify(model));
};

Resetting/Refreshing Collections

Rather than adding or removing models individually, you might occasionally wish to update an entire collection at once. Collection.reset() allows us to replace an entire collection with new models as follows:

PhotoCollection.reset([
  {title: "My trip to Scotland", src: "scotland-trip.jpg"},
  {title: "The flight from Scotland", src: "long-flight.jpg"},
  {title: "Latest snap of lock-ness", src: "lockness.jpg"}]);

Underscore utility functions

As Backbone requires Underscore as a hard dependency, we're able to use many of the utilities it has to offer to aid with our application development. Here's an example of how Underscore's sortBy() method can be used to sort a collection of photos based on a particular attribute.

var sortedByAlphabet = PhotoCollection.sortBy(function (photo) {
    return photo.get("title").toLowerCase();
});

The complete list of what Underscore can do is beyond the scope of this guide, but can be found in its official docs.

Routers

In Backbone, routers are used to help manage application state and for connecting URLs to application events. This is achieved using hash-tags with URL fragments, or using the browser's pushState and History API. Some examples of routes may be seen below:

http://unicorns.com/#whatsup
http://unicorns.com/#search/seasonal-horns/page2

Note: An application will usually have at least one route mapping a URL route to a function that determines what happens when a user reaches that particular route. This relationship is defined as follows:

"route" : "mappedFunction"

Let us now define our first controller by extending Backbone.Router. For the purposes of this guide, we're going to continue pretending we're creating a photo gallery application that requires a GalleryRouter.

Note the inline comments in the code example below as they continue the rest of the lesson on routers.

var GalleryRouter = Backbone.Router.extend({
    /* define the route and function maps for this router */
    routes: {
        "about" : "showAbout",
        /*Sample usage: http://unicorns.com/#about*/

        "photos/:id" : "getPhoto",
        /*This is an example of using a ":param" variable which allows us to match 
        any of the components between two URL slashes*/
        /*Sample usage: http://unicorns.com/#photos/5*/

        "search/:query" : "searchPhotos"
        /*We can also define multiple routes that are bound to the same map function,
        in this case searchPhotos(). Note below how we're optionally passing in a 
        reference to a page number if one is supplied*/
        /*Sample usage: http://unicorns.com/#search/lolcats*/

        "search/:query/p:page" : "searchPhotos",
        /*As we can see, URLs may contain as many ":param"s as we wish*/
        /*Sample usage: http://unicorns.com/#search/lolcats/p1*/

        "photos/:id/download/*imagePath" : "downloadPhoto",
        /*This is an example of using a *splat. splats are able to match any number of 
        URL components and can be combined with ":param"s*/
        /*Sample usage: http://unicorns.com/#photos/5/download/files/lolcat-car.jpg*/

        /*If you wish to use splats for anything beyond default routing, it's probably a good 
        idea to leave them at the end of a URL otherwise you may need to apply regular
        expression parsing on your fragment*/

        "*other"    : "defaultRoute"
        /*This is a default route that also uses a *splat. Consider the
        default route a wildcard for URLs that are either not matched or where
        the user has incorrectly typed in a route path manually*/
        /*Sample usage: http://unicorns.com/#anything*/

    },

    showAbout: function(){
    },

    getPhoto: function(id){
        /* 
        Note that the id matched in the above route will be passed to this function
        */
        console.log("You are trying to reach photo " + id);
    },

    searchPhotos: function(query, page){
        console.log("Page number: " + page + " of the results for " + query);
    },

    downloadPhoto: function(id, path){
    },

    defaultRoute: function(other){
        console.log("Invalid. You attempted to reach:" + other);
    }
});

/* Now that we have a router setup, remember to instantiate it*/

var myGalleryRouter = new GalleryRouter();

As of Backbone 0.5+, it's possible to opt-in for HTML5 pushState support via window.history.pushState. This permits you to define routes such as http://www.scriptjunkie.com/just/an/example. This will be supported with automatic degradation when a user's browser doesn't support pushState. For the purposes of this tutorial, we'll use the hashtag method.

Backbone.history

Next, we need to initialize Backbone.history as it handles hashchange events in our application. This will automatically handle routes that have been defined and trigger callbacks when they've been accessed.

The Backbone.history.start() method will simply tell Backbone that it's OK to begin monitoring all hashchange events as follows:

Backbone.history.start();
Router.navigate();

As an aside, if you would like to save application state to the URL at a particular point you can use the .navigate() method to achieve this. It simply updates your URL fragment without the need to trigger the hashchange event:

/*Lets imagine we would like a specific fragment for when a user zooms into a photo*/
zoomPhoto: function(factor){
    this.zoom(factor); //imagine this zooms into the image
    this.navigate("zoom/" + factor); //updates the fragment for us, but doesn't trigger the route
}

It is also possible for Router.navigate() to trigger the route as well as updating the URL fragment.

zoomPhoto: function(factor){
    this.zoom(factor); //imagine this zooms into the image
    this.navigate("zoom/" + factor, true); //updates the fragment for us and triggers the route
}

Namespacing

When learning how to use Backbone, an important and commonly overlooked area by tutorials is namespacing. If you already have experience with namespacing in JavaScript, the following section will provide some advice on how to specifically apply concepts you know to Backbone, however I will also be covering explanations for beginners to ensure everyone is on the same page.

What is namespacing?

The basic idea around namespacing is to avoid collisions with other objects or variables in the global namespace. They're important as it's best to safeguard your code from breaking in the event of another script on the page using the same variable names as you are. As a good 'citizen' of the global namespace, it's also imperative that you do your best to similarly not prevent other developer's scripts executing due to the same issues.

JavaScript doesn't really have built-in support for namespaces like other languages, however it does have closures which can be used to achieve a similar effect.

In this section we'll be taking a look shortly at some examples of how you can namespace your models, views, routers and other components specifically. The patterns we'll be examining are:

  • Single global variables
  • Object Literals
  • Nested namespacing

Single global variables

One popular pattern for namespacing in JavaScript is opting for a single global variable as your primary object of reference. A skeleton implementation of this where we return an object with functions and properties can be found below:

var myApplication = (function(){
    function(){
      // ...
    },
    return {
      // ...
    }
})();

You've probably seen this technique before. A Backbone-specific example might look like this:

var myViews = (function(){
    return {
        PhotoView: Backbone.View.extend({ .. }),
        GalleryView: Backbone.View.extend({ .. }),
        AboutView: Backbone.View.extend({ .. });
        //etc.
    };
})();

Here we can return a set of views, but the same technique could return an entire collection of models, views and routers depending on how you decide to structure your application. Although this works for certain situations, the biggest challenge with the single global variable pattern is ensuring that no one else has used the same global variable name as you have in the page.

One solution to this problem, as mentioned by Peter Michaux, is to use prefix namespacing. It's a simple concept at heart, but the idea is you select a common prefix name (in this example, myApplication_) and then define any methods, variables or other objects after the prefix.

var myApplication_photoView = Backbone.View.extend({}),
myApplication_galleryView = Backbone.View.extend({});

This is effective from the perspective of trying to lower the chances of a particular variable existing in the global scope, but remember that a uniquely named object can have the same effect. This aside, the biggest issue with the pattern is that it can result in a large number of global objects once your application starts to grow.

For more on Peter's views about the single global variable pattern, read his excellent post on them.

Note: There are several other variations on the single global variable pattern out in the wild, however having reviewed quite a few, I felt the prefixing approach applied best to Backbone.

Object Literals

Object Literals have the advantage of not polluting the global namespace but assist in organizing code and parameters logically. They're beneficial if you wish to create easily readable structures that can be expanded to support deep nesting. Unlike simple global variables, Object Literals often also take into account tests for the existence of a variable by the same name, which helps reduce the chances of collision.

This example demonstrates two ways you can check to see if a namespace already exists before defining it. I commonly use Option 2.

/*Doesn't check for existence of myApplication*/
var myApplication = {};

/*
Does check for existence. If already defined, we use that instance.
Option 1:   if(!myApplication) myApplication = {};
Option 2:   var myApplication = myApplication || {};
We can then populate our object literal to support models, views and collections (or any data, really):
*/

var myApplication = {
    models : {},
    views : {
        pages : {}
    },
    collections : {}
};

One can also opt for adding properties directly to the namespace (such as your views, in the following example):

var myGalleryViews = myGalleryViews || {};
myGalleryViews.photoView = Backbone.View.extend({});
myGalleryViews.galleryView = Backbone.View.extend({});

The benefit of this pattern is that you're able to easily encapsulate all of your models, views, routers etc. in a way that clearly separates them and provides a solid foundation for