Web Development with
MongoDB and NodeJS
Second Edition
Build an interactive and full-featured web application
from scratch using Node.js and MongoDB
Mithun Satheesh
Bruno Joseph D'mello
Jason Krol
Web Development with MongoDB and NodeJS
Second Edition
Copyright © 2015 Packt Publishing
All rights reserved. No part of this book may be reproduced, stored in a retrieval
system, or transmitted in any form or by any means, without the prior written
permission of the publisher, except in the case of brief quotations embedded in
critical articles or reviews.
Every effort has been made in the preparation of this book to ensure the accuracy
of the information presented. However, the information contained in this book is
sold without warranty, either express or implied. Neither the authors nor Packt
Publishing, and its dealers and distributors will be held liable for any damages
caused or alleged to be caused directly or indirectly by this book.
Packt Publishing has endeavored to provide trademark information about all of the
companies and products mentioned in this book by the appropriate use of capitals.
However, Packt Publishing cannot guarantee the accuracy of this information.
First published: October 2015
Production reference: 1261015
Published by Packt Publishing Ltd.
Livery Place
35 Livery Street
Birmingham B3 2PB, UK.
ISBN 978-1-78528-752-7
Technical Editor
Mithun Satheesh
Deepti Tuscano
Bruno Joseph D'mello
Copy Editor
Jason Krol
Merilyn Pereira
Project Coordinator
Huseyin Babal
Suzanne Coutinho
Luke P. Issac
Alexandru-Emil Lupu
Suhas Hoskote Muralidhar
Safis Editing
Sandeep Pandey
Oddur Sigurdsson
Tejal Soni
Commissioning Editor
Nadeem Bagban
Production Coordinator
Manu Joseph
Acquisition Editors
Neha Nagwekar
Cover Work
Manu Joseph
Reshma Raman
Content Development Editor
Zeeyan Pinheiro
About the Authors
Mithun Satheesh is an open source enthusiast and a web developer from India.
He has over five years of web development experience and specializes in JavaScript,
Ruby, PHP, and other frontend engineering technologies. He is the author of a
couple of libraries on Node.js, which are available as opensource via npm. One of
these is called node-rules, which is a forward-chaining rule engine implementation
written initially to handle transaction risks on bookmyshow.com, one of his former
employers. Node rules have a considerable user following both on npm and GitHub.
Apart from this, he has also served as an individual contributor to many open source
projects on GitHub.
He is a regular on programming sites such as Stack Overflow and loves contributing
to the open source world. Apart from programming, he is also interested in
experimenting with various cloud platform solutions. He has a number of applications
listed in the developer spotlight of platform-as-a-service providers such as RedHat's
You can follow him on Twitter at @mithunsatheesh.
I would like to thank my parents for all the support they have given
me. I am thankful to all my teachers for whatever knowledge I have
gained in my life.
Bruno Joseph D'mello is a web application engineer currently working at Built.io.
He is a JavaScript enthusiast and has a keen interest in different web technologies and
the programming paradigms implemented in them.
Thanks to my family for their patience and encouragement.
Jason Krol is a passionate web developer with over 15 years of professional
experience creating highly interactive web applications using the latest in both
client and server technologies.
Over the past few years, Jason has been focusing on developing Single-Page
Applications using JavaScript in the full stack with Node.js, MongoDB, and
Backbone.js. After co-owning and running a successful web development
agency for a number of years, Jason recently joined AWeber Communications,
an e-mail marketing service provider in the suburbs of Philadelphia.
When he isn't writing code for work or side projects, Jason blogs about his
development experiences and opinions at KrolTech.com. Jason loves spending
his free time with his wife and 8-year-old son.
A very special thank you to my wonderful wife for putting up with
me and for always being there to push me whenever I doubt myself.
About the Reviewers
Huseyin Babal is an enthusiastic full stack developer since 2007. He mainly
develops web applications using Java, Node.js, and PHP on the backend, AngularJS,
and Twitter Bootstrap on the frontend, and Elasticsearch and MongoDB for some
research projects. He is the author of NodeJS in Action in Udemy with over
1,500 students. He is also interested in DevOps engineering and applies continuous
delivery principles to his projects. He writes tutorials about full stack development
on Tuts+ and Java Code Geeks and shares his experiences at public conferences.
Besides the computer world, he lives in Istanbul with his wonderful wife and two
cockatiels. He likes to spend his spare time with his wife by walking at least an hour
per day, visiting different places, watching cartoons, and going on summer holidays.
Luke P. Issac is a full stack JavaScript developer with bachelor's of technology
degree (Hons.) in computer science engineering. He has also been a technical writer
for the past two years for www.thegeekstuff.com where he keeps on sharing his
works with the world in a simplified form and mentors an active technical discussion
on the blog.
His experience over the last four years involves his contribution to several successful
e-commerce implementations around the globe. He believes that technology is not
a bar and continues to research different technology stacks to find efficient and
optimized solutions for the continuously evolving Web. Thus, he has hands-on
experience with live e-commerce implementations on the MEAN stack and the
LAMP stack, which makes him very versatile in web development.
He had his thankful contribution on behalf of the Author in reviewing and adding
content for Getting Started with Flurry Analytics, Packt Publishing.
I want to thank my parents, Mr. Issac P. L. and Mrs. Sisily Issac,
who worked hard for my education, and all my friends for their
continuous support and inspiration to explore more and share it
with the world.
Alexandru-Emil Lupu is a CTO and Ruby on Rails developer at 2Performant.com.
Alex has about 10 years of experience in the web development industry, during
which time he learned a lot of skills from e-commerce platforms implementation
and presentation sites, to online games code writing. He is one of the developers
who are constantly learning new programming languages and he has no problem
understanding Ruby, PHP, Python, JavaScript, and Java code.
Alex is very passionate about programming and computer science; when he was a
teenager, he did not have his own computer or an Internet connection (hard to believe,
but true). He would go to an Internet cafe to read about his programming problems
and would then struggle to implement them at home. He fondly remembers those
days and hopes he's the same guy from 10 years ago with much more experience. For
him, passion is the word that describes the challenge he faced while learning. Trust me,
it is not easy to be a youngster, but also willing to learn new stuff. Coming home at 2-3
A.M, determined to install Linux just to learn about it, is not as easy as it sounds. I had
a Pentium I at 133 MHz in the Pentium IV at 1800 MHz era!
He is constantly learning and likes to stay close to well-trained and passionate
people who better motivate him every day! This is the reason he joined the
2Performant (2Parale) team, to face a challenge. He likes teams that work
intelligently and are energetic.
Proof of his perseverance is that he is a Certified Scrum Master and is passionate
about Agile development. His resume also includes four years at eRepublik.com,
an online game, where he was responsible for a long list of tasks including
feature development, performance optimization, but also was the tech lead on an
internal project. He learned the necessarily skills to fulfill his day-to-day tasks at
2Performant.com the hard way.
In the little spare time he has developed small personal projects or reads technical
or project management books or articles. When relaxing, he enjoys watching thriller
movies and playing shooter or strategy games.
He doesn't talk too much, but is willing to teach others programming. If you meet
him at a cafe prepare to be entertained, as he likes to tell a lot of contextual jokes.
His LinkedIn profile is at https://www.linkedin.com/in/alecslupu. Interact with
him on http://github.com/alecslupu.
Suhas Hoskote Muralidhar is a computer enthusiast and is extremely interested
in learning and exploring new technologies. After receiving his undergraduate
degree in computer science, he worked at Intel Corporation as a full stack web
developer. He is currently pursuing his master's degree at the University of Illinois
at Urbana-Champaign. He has kept in sync with his passion for web development by
working as a research assistant at the university by being involved in building full
stack e-learning web applications. He also developed a strong interest in machine
learning and big data analytics and has that as his main research focus. Besides
optimizing his code, he enjoys playing the guitar and going on road trips.
Sandeep Pandey is a full-stack developer for Practo.com, India, and builds products
for doctors and patients. In the past, he worked with CISCO as a consultant on
learning products (https://www.youtube.com/watch?v=dYMAD_L2kkM). He has
been actively involved in design and development of solutions using Node.js as a
tech stack. Some of the solutions he has contributed to include a social media platform
(https://flikstak.com) and an LMS adaptor at CLKS.
Sandeep enjoys working as a full stack developer and providing complete end-to-end
solutions, including UI frameworks such as Ember.js and Node frameworks such as
Experss.js, Restify.js, and so on. Apart from daily development activities, he enjoys
teaching and discussing things related to Node.js via an instructor-led training website
Oddur Sigurdsson is a full-stack JavaScript developer who's passionate about
cutting-edge experiences for the Web and sharing this knowledge with others. After
spending time as a teaching fellow at Fullstack Academy, an immersive JavaScript
program, he now works at Hoefler & Co. in New York, where he contributes to an
array of solutions that provide developers and designers with beautiful typography
for the Web.
Many thanks to Mithun Satheesh, Suzanne, and the excellent team at
Packt Publishing for giving me the opportunity to contribute to this
wonderful book.
Support files, eBooks, discount offers, and more
For support files and downloads related to your book, please visit www.PacktPub.com.
Did you know that Packt offers eBook versions of every book published, with PDF
and ePub files available? You can upgrade to the eBook version at www.PacktPub.
com and as a print book customer, you are entitled to a discount on the eBook copy.
Get in touch with us at service@packtpub.com for more details.
At www.PacktPub.com, you can also read a collection of free technical articles,
sign up for a range of free newsletters and receive exclusive discounts and offers
on Packt books and eBooks.
Do you need instant solutions to your IT questions? PacktLib is Packt's online digital
book library. Here, you can search, access, and read Packt's entire library of books.
Why subscribe?
• Fully searchable across every book published by Packt
• Copy and paste, print, and bookmark content
• On demand and accessible via a web browser
Free access for Packt account holders
If you have an account with Packt at www.PacktPub.com, you can use this to access
PacktLib today and view 9 entirely free books. Simply use your login credentials for
immediate access.
Table of Contents
Chapter 1: Welcome to JavaScript in the Full Stack
A short introduction to Node.js
The advantage that the V8 engine brings in
Node.js is single threaded!
Non-blocking asynchronous execution
npm – the Node Package Manager
Sharing and reusing JavaScript
Not just for building web servers!
Real-time web application with Socket.io
Networking and file IO
The origin of io.js
A simple server with Node.js
When to use Node.js?
The NoSQL movement
A short introduction to MongoDB
Features of MongoDB
Node and MongoDB in the wild
What to expect from this book
Chapter 2: Getting Up and Running
Installing Node.js
Mac OS X
Testing whether Node.js is installed properly
Online documentation
Table of Contents
Installing MongoDB
Mac OS X installation instructions
Windows 7 / Windows 8 installation instructions
Linux installation instructions
Confirming successful MongoDB installation
Bookmark the online documentation
Writing your first app
Creating the sample application
Getting the dependent modules in place
Adding the application code
Understanding the code
Launching the sample app
Checking the actual database
Chapter 3: Node and MongoDB Basics
A JavaScript primer
Syntax basics
Data types
Understanding the scope of variables
Operators and flows
Understanding objects
Understanding arrays
Understanding functions
Anonymous functions and callbacks
The basics of Node.js
Asynchronous execution
The module system
The Node.js core
Installing modules using npm
The basics of MongoDB
The Mongo shell
Inserting data
Updating data
Deleting data
Additional resources
[ ii ]
Table of Contents
Chapter 4: Introducing Express
Chapter 5: Templating with Handlebars
Chapter 6: Controllers and View Models
Web application frameworks
What is Express.js?
Building a complete web application
Designing the web application
Organizing the files
Creating the application's entry point
Booting up the application
Using and understanding middleware
The configure module
Activating the configure module
Routers and controllers
Custom middleware
Handlebars as view engines
Templating engines
Server-side and client-side templating
Client-side templating
Server-side templating
Partial views
The basics of Handlebars
Binding an object to the template
Embedding presentation logic
Handlebars helpers
Global helpers
View-specific helpers
Rendering the views
View models
Updating the home controller
Updating the image controller
Displaying an image
Uploading an image
[ iii ]
Table of Contents
Helpers for reusable code
The sidebar module
The stats module
The images module
The comments module
Testing the sidebar implementation
Iterating on the UI
Chapter 7: Persisting Data with MongoDB
Index – retrieving an image model
Create – inserting an image model
Testing everything out so far
The Like button and updating an image model
Comment – inserting a comment model
Wrapping it up
Using MongoDB with Node.js
Connecting to MongoDB
Inserting a document
Retrieving a document
Introducing Mongoose
Built-in validation
Static methods
Virtual properties
Connecting with Mongoose
Defining the schemas and models
The models index file
Adding CRUD to the controllers
The home controller
The image controller
Introducing the async module
The comments helper
The sidebar helper
The stats helper
The popular images helper
Iterating by adding an image removal capability
Adding a route
Adding a controller handler
Updating the Handlebars image page template
Updating jQuery
[ iv ]
Table of Contents
Refactoring and improvements
Chapter 8: Creating a RESTful API
Chapter 9: Testing Your Code
Chapter 10: Deploying with Cloud-Based Services
What is an API?
What is a RESTful API?
Introducing Postman REST Client
Installation instructions
A quick tour of Postman REST Client
Using the JSONView Chrome extension
Creating a basic API server
Creating sample JSON data
Responding to GET requests
Receiving data – POST and PUT requests
Removing data – DELETE
Consuming external APIs from Node.js
Consuming an API endpoint using request
The tools of the trade
Running tests with the Mocha framework
Asserting tests with Chai.js
Getting started with Chai
Spies and stubs with Sinon.js
Stubbing node modules with Proxyquire
Writing and running your first test
Writing a test helper
Testing the application
Testing the routes
Testing the server
Testing a model
Testing a controller
Spy and stub everything!
Cloud versus traditional hosting
Infrastructure as a Service versus Platform as a Service
An introduction to Git
Deploying your application
Table of Contents
Amazon Web Services
Create a MongoLab account and database
Create and configure the AWS environment
Microsoft Azure
Digital Ocean
Chapter 11: Single-Page Applications with Popular
Frontend Frameworks
Chapter 12: Popular Node.js Web Frameworks
What is a single page application?
Why use a frontend framework?
The TodoMVC project
Frontend development tools
Automated build task managers
Dependency management
HTML template-rendering engines
CSS transpiling
Testing and test-driven development
PhantomJS headless browser
[ vi ]
Node.js and MongoDB are quickly becoming very popular tech stacks for the
Web. Powered by Google's V8 engine, Node.js caters to easily building fast,
scalable network applications while MongoDB is the perfect fit as a scalable,
high-performance, open source NoSQL database solution. Using these two
technologies together, web applications can be built quickly and easily and
deployed to the cloud with very little difficulty.
The book will begin by introducing you to the groundwork needed to set up the
development environment. Here, you will quickly run through the steps necessary
to get the main application server up and running. Then, you will see how to use
Node.js to connect to a MongoDB database and perform data manipulations.
From here on, the book will take you through integration with third-party tools for
interaction with web apps. It then moves on to show you how to use controllers and
view models to generate reusable code that will reduce development time. Toward
the end of the book, we will cover tests to properly execute the code and some
popular frameworks for developing web applications.
By the end of the book, you will have a running web application developed with
MongoDB and Node.js along with their popular frameworks.
What this book covers
Chapter 1, Welcome to JavaScript in Full Stack, introduces you to Node.js and the
advantages of writing Javascript on the backend. In addition to this, it will explain
the overall architecture of the application you are going to build using this book.
Chapter 2, Getting Up and Running, explains how to set up the development
environments for Node.js and MongoDB. You will also be verifying that
everything is set up properly by writing a sample app and running it.
[ vii ]
Chapter 3, Node and MongoDB Basics, teaches you about the fundamental concepts of
JavaScript, Node.js, and MongoDB in this chapter. It will introduce you to NodeJS
and the various concepts around it, as well as MongoDB and its basic shell for
CRUD operations.
Chapter 4, Introducing Express.js, introduces you to the Express framework and its
various components It also walks you through how you will be organizing the
basic application you are building with this framework. It will give you a detailed
overview of the MVC components of Express.js too.
Chapter 5, Templating with Handlebars, introduces you to the concept of using a
templating engine and handlebars. Also, it shows you how to use handlebars in
your application as a templating engine.
Chapter 6, Controllers and View Models, shows you how to organize the code
for the sample application you build into the controllers and views of Express.
It will introduce you to the MVC concepts indirectly via introducing the need of
separating the code into various modules and utilizing the Express framework.
Chapter 7, Persisting Data with MongoDB, shows you how to connect to the MongoDB
server from the Node.js application you are building. It will also introduce you to the
concept of ODM, the most popular one being Mongoose.
Chapter 8, Creating a RESTful API, introduces you to RESTful APIs. Also, it shows you
the importance of RESTful wrapper for the application. Then, it will teach you how
to can change the current application to a REST API based application.
Chapter 9, Testing Your Code, shows you why you need testing incorporated with
your application and also what you should keep in mind for the testability of the
code written.
Chapter 10, Deploying with Cloud-Based Services, discusses the options for hosting the
Node.js MongoDB application you are building. It also compares the various PaaS
solutions available in the market.
Chapter 11, Single-Page Applications with Popular Frontend Frameworks, discusses
Single Page Applications. Also, you will analyze the popular frontend frameworks
available. You will have a look at other frontend aspects such as the automation tools
and transpilers available.
Chapter 12, Popular Node.js Web Frameworks, covers the various web frameworks
available on Node.js, other than Express, which you will be using to build the
application in this book. You will analyze various web frameworks such as Meteor,
Sails, Koa, Hapi, and Flatiron.
[ viii ]
What you need for this book
You will need:
• A computer running OS X, Windows, or Linux
• Samba 4.x Server software
Who this book is for
This book is designed for JavaScript developers of any skill level that want to get up
and running using Node.js and MongoDB to build full-featured web applications. A
basic understanding of JavaScript and HTML is the only requirement for this book.
In this book, you will find a number of text styles that distinguish between different
kinds of information. Here are some examples of these styles and an explanation of
their meaning.
Code words in text, database table names, folder names, filenames, file extensions,
pathnames, dummy URLs, user input, and Twitter handles are shown as follows:
"The function that actually logs Three is known as a callback to the setTimeout
A block of code is set as follows:
setTimeout(function() {
}, 2000);
When we wish to draw your attention to a particular part of a code block, the
relevant lines or items are set in bold:
<div class="panel panel-default">
<div class="panel-heading">
<h3 class="panel-title">
Newest Images
[ ix ]
<div class="panel-body">
{{#each images}}
<div class="col-md-4 text-center" style="padding-bottom:
1em;"><a href="/images/{{ uniqueId }}"><img src="/public/upload/
{{filename}}" alt="{{title}}" style="width: 175px; height: 175px;"
Any command-line input or output is written as follows:
$ command -parameters –etc
New terms and important words are shown in bold. Words that you see on the
screen, for example, in menus or dialog boxes, appear in the text like this: " You
should see The image:index controller testing123 on the screen!"
Warnings or important notes appear in a box like this.
Tips and tricks appear like this.
Reader feedback
Feedback from our readers is always welcome. Let us know what you think about
this book—what you liked or disliked. Reader feedback is important for us as it helps
us develop titles that you will really get the most out of.
To send us general feedback, simply e-mail feedback@packtpub.com, and mention
the book's title in the subject of your message.
If there is a topic that you have expertise in and you are interested in either writing
or contributing to a book, see our author guide at www.packtpub.com/authors.
Customer support
Now that you are the proud owner of a Packt book, we have a number of things to
help you to get the most from your purchase.
Downloading the example code
You can download the example code files from your account at http://www.
packtpub.com for all the Packt Publishing books you have purchased. If you
purchased this book elsewhere, you can visit http://www.packtpub.com/support
and register to have the files e-mailed directly to you.
Although we have taken every care to ensure the accuracy of our content, mistakes
do happen. If you find a mistake in one of our books—maybe a mistake in the text or
the code—we would be grateful if you could report this to us. By doing so, you can
save other readers from frustration and help us improve subsequent versions of this
book. If you find any errata, please report them by visiting http://www.packtpub.
com/submit-errata, selecting your book, clicking on the Errata Submission Form
link, and entering the details of your errata. Once your errata are verified, your
submission will be accepted and the errata will be uploaded to our website or added
to any list of existing errata under the Errata section of that title.
To view the previously submitted errata, go to https://www.packtpub.com/books/
content/support and enter the name of the book in the search field. The required
information will appear under the Errata section.
Piracy of copyrighted material on the Internet is an ongoing problem across all
media. At Packt, we take the protection of our copyright and licenses very seriously.
If you come across any illegal copies of our works in any form on the Internet, please
provide us with the location address or website name immediately so that we can
pursue a remedy.
Please contact us at copyright@packtpub.com with a link to the suspected
pirated material.
We appreciate your help in protecting our authors and our ability to bring you
valuable content.
If you have a problem with any aspect of this book, you can contact us at
questions@packtpub.com, and we will do our best to address the problem.
[ xi ]
Welcome to JavaScript
in the Full Stack
What an exciting time to be a JavaScript developer! What was once only considered
a language to add enhancements and widgets to a web page has since evolved into
its own fully-fledged ecosystem. As of the beginning of 2015, it stands as the second
most popular language in terms of questions tagged on stack overflow, next to only
Java, with around a million questions tagged on it. There are tons of frameworks
and environments to make it possible to run JavaScript almost anywhere. I believe
Atwood's law says it best:
"Anything that can be written in JavaScript will eventually be written
in JavaScript!"
While this quote dates back to 2007, it's never been more true than today. Not only
can you use JavaScript to develop a complete single-page application such as Gmail,
but you will also see how we can achieve the following projects with JavaScript in
the coming chapters of the book:
• Completely power the backend using Node.js and Express.js
• Persist data with a powerful document oriented database such as MongoDB
• Write dynamic HTML pages using Handlebars.js
• Deploy your entire project to the cloud using services such as Heroku and
Amazon Web Services (AWS)
Welcome to JavaScript in the Full Stack
With the introduction of Node.js, JavaScript has officially gone in a direction that
was never even possible before. Now you can use JavaScript on the server and you
can also use it to develop full-scale, enterprise-level applications. When you combine
this with the power of MongoDB and its JSON-powered data, you can work with
JavaScript in every layer of your application.
Let's quickly go through some basic concepts of Node.js and MongoDB, which
would be helpful for you to follow the rest of the chapters in this book.
A short introduction to Node.js
One of the most important things that people get confused about while getting
introduced to Node.js is to understand what exactly it is. Is it a different language
altogether, is it just a framework on top of it, or is it something else? Node.js is
definitely not a new language, and it is not just a framework on JavaScript. It can
be considered as a runtime environment for JavaScript built on top of Google's V8
engine. So, it provides us with a context where we can write JavaScript code on any
platform where Node.js can be installed. Anywhere!
Now a bit about its history! Back in 2009, Ryan Dahl gave a presentation at JSConf
that changed JavaScript forever. During his presentation, he introduced Node.js to
the JavaScript community. After a roughly 45-minute talk, he concluded it, receiving
a standing ovation from the audience in the process. He was inspired to write Node.
js after he saw a simple file upload progress bar on Flickr, the image-sharing site.
Realizing that the site was going about the whole process the wrong way, he
decided that there had to be a better solution.
Now let's go through the features of Node.js, which makes it unique from other
server-side programming languages.
The advantage that the V8 engine brings in
The V8 engine was developed by Google and was open sourced in 2008. As we
all know, JavaScript is an interpreted language and it will not be as efficient as a
compiled language, as each line of code gets interpreted one by one while the code
gets executed. The V8 engine brings in an efficient model here, where the JavaScript
code will be compiled into machine-level code and the executions will happen on
the compiled code instead of interpreting the JavaScript. But even though Node.
js is using the V8 engine, Joyent, which is the company that is maintaining Node.
js development, it does not always update the V8 engine to the latest versions that
Google actively releases. This has led to the new branch named io.js, which we will
discuss later in this chapter.
Chapter 1
Node.js is single threaded!
You might be asking, how does a single threaded model help? Typical PHP, ASP.
NET, Ruby, or Java-based servers follow a model where each client request results
in the instantiation of a new thread or even a process. When it comes to Node.
js, requests are run on the same thread with even shared resources. A common
question that is often asked is what will be the advantage of using such a model? To
understand this, we should understand the problem that Node.js tries to resolve. It
tries to do asynchronous processing on a single thread to provide more performance
and scalability for applications that are supposed to handle too much web traffic.
Imagine web applications that handle millions of concurrent requests; if the server
makes a new thread for handling each request that comes in, it will consume a lot
of resources and we would end up trying to add more and more servers to increase
the scalability of the application. The single threaded asynchronous processing
model has its advantage in the previous context, and you can process much more
concurrent requests with less number of server-side resources. However, there is a
downside to this approach; Node (by default) will not utilize the number of CPU
cores available on the server it is running on without using extra modules like pm2.
The point that Node.js is single threaded doesn't mean that it doesn't use
threads internally. It is just that the developer and the execution context
that the code has exposure to have no control over the threading model
internally used by Node.js.
If you are new to the concept of threads and process, we would suggest
you to go through some preliminary articles about these topics. There
are plenty of YouTube videos as well on the same topic. The following
reference could be used as a starting point:
Non-blocking asynchronous execution
One of the most powerful features of Node is that it is event-driven and
asynchronous. So how does an asynchronous model work? Imagine you have a block
of code and at some nth line you have an operation that is time consuming. So what
happens to the lines that follow the nth line while this code gets executed? In normal
synchronous programming models, the lines that follow the nth line will have to
wait till the operation at that line completes. An asynchronous model handles this
case differently. To handle this scenario in an asynchronous approach, we need
to segment the code that follows the nth line into two sections. The first section is
dependent on the result of the operation at the nth line and the second section is
independent of the result.
Welcome to JavaScript in the Full Stack
We wrap the dependent code in a function with the result of the operation as its
parameter, and register it as a callback to the operation on its success. Once the
operation completes, the callback function will be triggered with its result. Meanwhile,
we can continue executing the result-independent lines without waiting for the result.
In this scenario, the execution is never blocked for a process to complete. It just goes
on with callback functions registered on each ones completion. Simply put, you assign
a callback function to an operation, and when Node determines that the completion
event has been fired, it will execute your callback function at that moment.
We can look at an example to understand the asynchronous nature in detail:
setTimeout(function() {
}, 2000);
In a typical synchronous programming language, executing the preceding code
will yield the following output:
... (2 second delay) ...
However, in an asynchronous approach, the following output is seen:
... (approx. 2 second delay) ...
The function that actually logs Three is known as a callback to the setTimeout
If you are still interested in learning more about asynchronous models
and the callback concept in JavaScript, Mozilla Developer Network
(MDN) has many articles that explain these concepts in detail.
Chapter 1
npm – the Node Package Manager
Writing applications with Node is really enjoyable when you realize the sheer wealth
of information and tools at your disposal! Using Node's built-in package manager
npm, you can find literally tens of thousands of modules that can be installed and
used within your application with just a few keystrokes! One of the reasons for the
biggest success of Node.js is npm, which is one of the best package managers out
there with a very minute learning curve. If this is the first ever package manager
that you are getting exposed to, then you should consider yourself lucky!
On a regular month, npm handles more than a billion downloads and it has around
150,000 packages currently available for you to download. You can view the library
of available modules by visiting www.npmjs.com. Downloading and installing any
module within your application is as simple as executing the npm install package
command. Have you written a module that you want to share with the world? You
can package it up using npm and upload it to the public www.npmjs.org registry just
as easily! If you are not sure how a module you installed works, the source code is
right there in your projects' node_modules/ folder waiting to be explored!!
Package versions of modules in npm follow semantic versioning,
such as major.minor.patch order.
Sharing and reusing JavaScript
While you develop web applications, you will always end up doing the validations
for your UI both at the client and server sides, as the client-side validations are
required for better UI experience and server-side validations are needed for better
security of app. Think about two different languages in action: you will have the
same logic implemented in both the server and client side. With Node.js, you can
think of sharing the common function between server and client, reducing the code
duplication to a large extent.
Ever worked on optimizing the load time for client-side components of your
Single-Page Application (SPA) loaded from template engines such as underscore?
You would end up thinking about a way we could share the rendering of templates
in both server and client at the same time—some call it hybrid templating. Node.js
resolves the context of duplication of client templates better than any other serverside technologies, just because we can use the same JS templating framework and
the templates both at server and client.
Welcome to JavaScript in the Full Stack
If you are taking this point lightly, the problem it resolves is not just the issue
of reusing validations or templates on the server and client. Think about an SPA
being built; you will need to implement the subsets of server-side models in the
client-side MV* framework also. Now, think about the templates, models, and
controller subsets being shared on both client and server. We are solving a higher
scenario of code redundancy.
Not just for building web servers!
Node.js is not just to write JavaScript in the server-side. Yes, we have discussed
this point earlier. Node.js sets up the environment for the JavaScript code to work
anywhere it can be installed. It can be a powerful solution to create command-line
tools as well as fully-featured, locally run applications that have nothing to do with
the Web or a browser. Grunt.js is a great example of a Node-powered command-line
tool that many web developers use daily to automate everyday tasks such as build
processes, compiling CoffeeScript, launching Node servers, running tests, and more.
In addition to command-line tools, Node is increasingly popular among the
hardware crowd with the Node bots movement. Johnny-Five and Cylon.js are two
popular Node libraries that exist to provide a framework to work with robotics. Just
search on YouTube for Node robots and you will see a lot of examples. Also, there is
a chance that you might be using a text editor developed on Node.js. GitHub's open
source editor named Atom, which is hugely popular, is an example.
Real-time web application with Socket.io
One of the important reasons behind the origin of Node.js was to support
real-time web applications. Node.js has a couple of frameworks built for real-time
web applications, which are hugely popular: Socket.io and sock.js. These frameworks
make it very simple to build instant collaboration-based applications such as Google
Drive and Mozilla's together.js. Before the introduction of WebSockets in the modern
browsers, this was achieved via long polling, which was not a great solution for
real-time experience. While WebSockets is a feature that is only supported in
modern browsers, Socket.io acts as a framework, which also features seamless
fallback implementations for legacy browsers.
If you need to understand more on the use of WebSockets in
applications, here's a good resource on MDN that you can explore:
Chapter 1
Networking and file IO
In addition to the powerful non-blocking asynchronous nature of Node, it also has
very robust networking and filesystem tools available via its core modules. With
Node's networking modules, you can create server and client applications that accept
network connections and communicate via streams and pipes.
The origin of io.js
As we mentioned earlier in this chapter, io.js is nothing but a fork of Node.js created
to be updated with the latest development on both V8 and other developments in the
JS community. Joyent was taking care of the releases in Node.js and the process that
was followed in taking care of the release management of Node.js lacked an open
governance model. It leads to scenarios where the newer developments in V8 as well
as the JS community were not incorporated into its releases. For example, if you want
to write JavaScript using the latest EcmaScript6 (ES6) features, you will have to run
it in the harmony mode. Joyent is surely not to be blamed for this, as they were more
concerned about stability of Node.js releases than frequent updates in the stack.
This led to the io.js fork, which is kept up to date with the latest JavaScript and V8
updates. So, it's better to keep your eyes on the releases on both Node and io.js to be
up to date with the Node.js world.
A simple server with Node.js
To see an example of just how lightweight Node can be, let's take a look at some
sample code that starts up an HTTP server and sends Hello World to a browser:
var http = require('http');
http.createServer(function (req, res) {
res.writeHead(200, {'Content-Type': 'text/plain'});
res.end('Hello World\n');
}).listen(8080, 'localhost');
console.log('Server running at http://localhost:8080');
A few basic lines of code are all it takes to write a complete Node application.
Running it with a simple Node app.js command will launch an HTTP server that is
listening on port 8080. Point any browser to http://localhost:8080, and you will
see the simple output Hello World on your screen! While this sample app doesn't
actually do anything useful, it should give you a glimpse of the kind of power you
will have while writing web applications using Node.js. If you don't have the initial
Node.js development environment set up, we will discuss it in the next chapter.
Welcome to JavaScript in the Full Stack
When to use Node.js?
You should have heard of this proverb:
"If all you have is a hammer, everything looks like a nail!"
This makes a lot of sense in this context. Node.js is not a technology to depend on
all the application problems that you intend to solve and if not chosen wisely, the
decision will backfire. Node.js is well suited for applications that are expected to
handle a huge amount of concurrent connections. Also, it should be noted that it is
most suited for applications where each incoming request requires very few CPU
cycles. This means that if you intend to do computation-intensive tasks on requests, it
will end up blocking the event loop—thereby impacting other requests concurrently
processed by the web server. Node.js is well suited for real-time web applications,
such as chat rooms, collaboration tools, online games, and so on. So when deciding
whether to use or not use Node.js, we should analyze the application context
seriously and figure out whether Node.js really suits the context of the application.
It is quite hard to debate over the use cases of Node.js in a detailed
manner. However, the following stackoverflow thread does this so
effectively and we strongly recommend you to go though the answers
on this post if you are more interested in the use cases of Node.js:
As we have briefly gone through the features and concept of Node.js, now let's look
into the NoSQL and MongoDB side.
The NoSQL movement
Let's start by exploring the answers to the question, what exactly is a NoSQL
database? NoSQL is a common term for database technologies that deviate from
the traditional Relational Database Management System (RDBMS) concepts.
The common reason for these database solutions to deviate from RDBMS database
standards is to achieve and set a better standard of availability and partitioning
capabilities than traditional RDBMS solutions.
To introduce you to this concept, we should have a look at the Brewer's theorem,
which is otherwise known as the CAP theorem:
"It is impossible for a distributed computer system to simultaneously provide all three of the
following guarantees: Consistency, Availability, and Partition tolerance."
Chapter 1
Traditional RDBMS solutions are good at consistency and compromise a little once
it comes to providing better availability (data reads) and partitioning capabilities.
Most of the NoSQL solutions have been evolved in this direction to achieve better
heights in data availability and partitioning.
As this is a common term for any database technology that deviates from the
concepts followed by RDBMS solutions such as MySQL, PostgreSQL, and so on,
there are various subsets for NoSQL databases. Most popular subsets of NoSQL are
document stores, key-value stores, and graph-based database solutions. MongoDB,
which is the one we are going to try out, falls in the document store category. There
are many more NoSQL solutions available in the market apart from MongoDB, such
as Cassandra, Redis, Neo4j, Hbase, and so on.
A short introduction to MongoDB
As we discussed in the previous paragraph, MongoDB falls into the document
store category of NoSQL databases. MongoDB is being actively developed by
10gen, which was later renamed to MongoDB I.inc. MongoDB is open source
and its source is available on various platforms such as GitHub.
Features of MongoDB
One of the most important reasons for the popularity of MongoDB is that it is a
JSON-friendly database. It means that documents are stored and retrieved from
MongoDB as JavaScript objects. Internally, this JSON data gets converted to BSON
format while getting persisted to the system. So, this gives an extreme flexibility where
we can use the same data format from client to server and eventually to the database.
A typical document (record) in a MongoDB collection (table) might look like the
following code:
$ mongo
> db.contacts.find({email: 'jason@kroltech.com'}).pretty()
"email" : "jason@kroltech.com",
"phone" : "123-456-7890",
"gravatar" : "751e957d48e31841ff15d8fa0f1b0acf",
"_id" : ObjectId("52fad824392f58ac2452c992"),
"name" : {
"first" : "Jason",
"last" : "Krol"
"__v" : 0
Welcome to JavaScript in the Full Stack
Another important feature of MongoDB is its schemaless nature. With a relational
database, you are required to define (ahead of time) the exact structure of the
data being stored, which is termed as the schema. This means that you must have
defined the exact number of columns, length, and data type for every field in a table,
and that each field must always match that exact set of criteria. Mongo provides a
flexible nature where the documents that you store into the database need not follow
any schema unless the developer enforces it through the application level. This
makes MongoDB a great fit for Agile-based development, as you could carry out
modifications on the application schema on fly.
Other than the JavaScript-friendly nature, one other resemblance of MongoDB
with Node.js is that it is also designed with highly concurrent applications with
heavy read operations in mind.
MongoDB also introduces the concept of sharding, which makes it possible to scale
the database horizontally as well as vertically. If the application owner needs to
increase the database capabilities, they could add up more machines into the stack.
This is a cheaper option compared to investing on RAM of a single machine, which
will be the case in RDBMS solutions.
All the advantages that we discussed come with some impact on the consistency,
as MongoDB does not strictly adhere to the RDBMS standards like ACID
transactions. Also, if you end up creating a data model that might need too many
JOIN operations, then MongoDB won't make a good fit as it is not designed with
too many aggregations even though the aggregations are possible via the MongoDB
aggregation framework. MongoDB may or may not be the right solution for your
application. You should truly weigh the pros and cons of each technology before
making a decision to determine which technology is right for you.
Node and MongoDB in the wild
Both Node and MongoDB are extremely popular and active in the development
community. This is true for enterprises as well. Some of the biggest names in the
Fortune 500 space have fully embraced Node to power their web applications.
This is due in large part to the asynchronous nature of Node, which makes it
a great alternative for high traffic, high I/O applications such as e-commerce
websites and mobile applications.
Here's just a small list of some big companies that are working with Node:
• PayPal
• LinkedIn
• eBay
[ 10 ]
Chapter 1
• Walmart
• Yahoo!
• Microsoft
• Dow Jones
• Uber
• New York Times
MongoDB's use in the enterprise sector is equally as impressive and widespread,
with an increasing number of companies adopting the leading NoSQL database
server. Here's just a small list of some big companies that are working with
• Cisco
• Craigslist Inc.
• Forbes
• FourSquare
• Intuit
• McAfee
• MetLife
• Shutterfly
• Under Armour
What to expect from this book
The remainder of this book is going to be a guided tour that walks you through
creating a complete data-driven website. The website we create will feature almost
every aspect of a typical large-scale web development project. The app will be
developed using a popular Node.js framework called Express, and it will persist data
using MongoDB. In the first few chapters, we will cover the groundwork involved
in getting the core of the server up and serving content. This includes configuring
your environment so you are up and running with Node and MongoDB, and a
basic introduction to the core concepts of both technologies. Then, we will write a
web server from scratch powered by ExpressJS, which will handle serving all of the
necessary files for the website. From there, we will work with the Handlebars template
engine to serve both static and dynamic HTML web pages. Diving deeper, we will
make the application persistent by adding a data layer where the records for the
website will be saved and retrieved via a MongoDB server.
[ 11 ]
Welcome to JavaScript in the Full Stack
We will cover writing a RESTful API so that other people can interact with your
application. Finally, we will go into the details to see how to write and execute
tests for all of your code. A summary is given in the following figure:
Wrapping up, we will take a brief detour as we examine some popular, merging
frontend technologies that are becoming increasingly popular while writing SPAs.
These technologies include Backbone.js, Angular, and Ember.js.
Last but not least, we will go into details about how to deploy your new website
to the Internet using popular cloud-based hosting services such as Heroku and
Amazon Web Services.
Downloading the example code
You can download the example code files from your account at
http://www.packtpub.com for all the Packt Publishing books
you have purchased. If you purchased this book elsewhere, you can
visit http://www.packtpub.com/support and register to have
the files e-mailed directly to you.
In this chapter, we reviewed what is to be expected throughout the rest of this
book. We discussed the amazing current state of JavaScript and how it can be used
to power the full stack of a web application. Not that you needed any convincing
in the first place, but I hope you're excited and ready to get started writing web
applications using Node.js and MongoDB!
Next up, we will set up your development environment and get you up and
running with Node, MongoDB, and npm as well as write and launch a quick
Node app that uses MongoDB!
[ 12 ]
Getting Up and Running
In this chapter, we will cover the necessary steps to set up your development
environment. These will include the following:
1. Installing Node.js on your machine
2. Installing MongoDB on your machine
3. Verifying whether everything is set up properly
Follow these sections carefully, as we need the development environment to be up
and running before we jump into the chapters where we dive into actual coding.
For the remainder of this book, it's going to be assumed that you are using either
a Mac with OS X, Linux, or Windows 7 / Windows 8. You will also need super
user and/or administrator privileges on the computer, as you will be installing the
Node and MongoDB server. The code and examples after this chapter will all be OS
agnostic and should work in any environment, assuming you have taken the steps
I have outlined earlier to be prepared ahead of time.
You will need a proper text editor to write and edit the code. While any text editor you
choose will serve this purpose, choosing a better text editor will hugely improve your
productivity. Sublime Text 3 appears to be the most popular text editor regardless of
the platform at this moment. It is a simple, lightweight editor with unlimited plugins
made available by developers around the world. If you are on a Windows machine,
then Notepad++ is also a good candidate. Also, there are JavaScript-based open source
editors such as Atom and Brackets, which are also worth a try.
[ 13 ]
Getting Up and Running
Finally, you're going to need access to the command line. Linux and Mac have access to
the command line via the Terminal program. A great alternative on the Mac is iTerm2
(http://iterm2.com). For Windows, the default command-line program works, but
isn't the best. A great alternative there is ConEmu (http://conemu.codeplex.com).
For the remainder of this book, any time I reference a command line or command
prompt, it will look like the following:
$ command -parameters –etc
Installing Node.js
The Node.js installer can be easily obtained by visiting the official Node website and
accessing the downloads section. Once there, be sure to download the correct version
depending on your OS and CPU (32 bit or 64 bit). As an alternative, you can also use
OS-specific package managers to install this. Depending on the OS you are using, just
jump into the specific subsection below to get more details on the steps to be followed.
You can jump into the Node.js download sections by following
the link http://nodejs.org/download.
Mac OS X
There is a universal installer available from the Node website specifically for OS X.
We need to follow these steps to install Node.js on a Mac:
1. Visit the download page of the Node.js official website as mentioned earlier
and click on the Mac OS X installer, which is independent of the processor
type (32 or 64 bit).
2. Once the download is complete, double-click on the .pkg file, which will
launch the Node installer.
[ 14 ]
Chapter 2
3. Proceed through each step of the wizard that should be fairly self-explanatory.
Also, note that if you have any of the OS X package managers
installed, then you don't need to manually download the installer.
You may install Node.js via the respective package manager.
You may install Node.js via the respective package manager:
• Installation using Homebrew
brew install node
• Installation using Mac ports
port install nodejs
The installation of Node.js via either an installer or via the
package managers will include npm alongside. So, we don't
need to install it separately.
[ 15 ]
Getting Up and Running
To install Node.js on Windows, we will follow these steps:
1. We need to determine your processor type, 32 or 64 bit. You can do this by
executing the following command at the command prompt:
$ wmic os get osarchitecture
The output is as follows:
2. Download the installer depending on the result of this command.
3. Once the download is complete, double-click on the .msi file, which will
launch the Node installer.
4. Proceed through each step of the wizard.
5. When you get to the custom setup screen, you should notice that the
installation wizard will install not only the Node.js runtime, but also
the npm package manager, and configure a PATH variable.
6. So once the installation is done, Node and npm can be executed from any
folder via the command line.
[ 16 ]
Chapter 2
Also, if you have any of the Windows package managers installed, then you don't
need to manually download the installer. You may install Node.js via the respective
package manager:
• Using chocolatey
cinst nodejs.install
• Using scoop
scoop install nodejs
Because there are so many different flavors and distributions of Linux available,
installing Node isn't quite as straightforward. However, if you're running Linux
to begin with, then you are more than aware of this and probably comfortable
with a few extra steps.
Joyent has an excellent wiki on how to install Node on Linux using the many
different package manager options available. This covers almost all the popular
deb and rpm based package managers. You can read that wiki by visiting:
For Ubuntu 12.04 and above as an example, the steps to install Node would be
as follows:
$ sudo apt-get update
$ sudo apt-get install python-software-properties python g++ make
$ sudo add-apt-repository ppa:chris-lea/node.js
$ sudo apt-get update
$ sudo apt-get install nodejs nodejs-dev npm
Once these steps have been completed, both Node and npm should be installed on
your system.
[ 17 ]
Getting Up and Running
Testing whether Node.js is installed properly
Now that Node has been installed on your system, let's run a quick test to
ensure everything is working properly.
Access a command line via your terminal program and execute the
following command:
$ node --version
$ npm --version
Assuming that your Node installation was successful, you should see the version
number that was installed as an output on the screen right below the command
you executed.
Note that your version numbers will most likely be more recent
than those printed earlier.
You can also launch the Node repl, a command-line shell that lets you execute
JavaScript directly:
$ node
> console.log('Hello world!')
Hello World!
[press Ctrl-C twice to exit]
Online documentation
You'll want to be sure to point your browser to the online documentation for Node
and bookmark it, as it will undoubtedly become a resource that you will want to
access on a regular basis. It is not mandatory that you should go through each and
every section of it, but once you start writing code in Node.js, you will need to refer
to this document frequently to understand more on the apis exposed by Node.js.
The document is available at: http://nodejs.org/api/.
Also, check out the npm registry available at http://npmjs.com where you can find
tens of thousands of modules available for Node developers.
[ 18 ]
Chapter 2
Installing MongoDB
MongoDB can also be easily downloaded by visiting the official MongoDB website
and accessing the downloads section from http://www.mongodb.org/downloads.
Once there, be sure to download the correct version depending on your OS and
CPU (32 or 64 bit). For Windows users, you can opt to download the MSI installer
file, which will make the installation much simpler.
Depending on the version of MongoDB you downloaded, you
will want to replace <version> in the following sections with
the appropriate version number that matches your file.
Mac OS X installation instructions
If you are using the Homebrew package manager, MongoDB can be installed using
the following two commands:
$ brew update
$ brew install mongoDB
The remainder of this chapter assumes you are not using Homebrew and need to
install MongoDB manually. If you are installing MongoDB via Homebrew, you can
proceed directly to the testing MongoDB installation section.
After completing the download, open and extract the contents of the .tgz file. You
will want to move the extracted contents to a destination folder /mongodb. You can
do this either via the Finder or the command line, whichever you prefer, as follows:
$ mkdir -p /mongodb
$ cd ~/Downloads
$ cp -R -n mongodb-osx-x86_64-2.4.9/ mongodb
You will want to ensure that the locations of the MongoDB binaries are configured
in your environment PATH so that you can execute mongod and mongo from any
working directory. To do this, edit the .profile file in your home folder (~/) and
append the location for MongoDB to it. Your .profile file should look something
like the following:
export PATH=~/bin:/some/of/my/stuff:/more/stuff:/mongodb/bin:$PATH
[ 19 ]
Getting Up and Running
If you don't have this line or are missing .bash_profile completely, you can create
one easily by executing the following command:
$ touch .bash_profile
$ [edit] .bash_profile
export PATH=$PATH:/mongodb/bin
You will more than likely have a lot more than what I have in the preceding lines of
code. The important thing is that you append :/mongodb/bin before the $PATH at
the end. The : is a delimiter between different paths (so it's likely that you will be
adding your path to the end of an existing list but before the trailing $PATH).
Here, mongod refers to the MongoDB server instance that you need
to invoke and mongo refers to the Mongo shell, which will be the
console through which you will be interacting with the database.
Next, you need to create a default data folder that MongoDB will use to store all
data documents. From the command line, execute the following:
$ mkdir -p /data/db
$ chown `id -u` /data/db
Once the files have been properly extracted to the /mongodb folder and the data
folders created, you can now start the MongoDB database server by executing the
following command from the command line:
$ mongod
This should dump a bunch of log statements while the server starts up, but
ultimately ends with:
Sun Mar 16 12:26:58.885 [initandlisten] waiting for connections on
port 27017
That's it! Your MongoDB server is up and running. You can type Ctrl-C to cancel
and shut down the server.
Windows 7 / Windows 8 installation
After completing the download, the MongoDB website will automatically redirect
you to a landing page with a link to a Windows Quick Start guide at http://docs.
[ 20 ]
Chapter 2
It is highly recommended that you follow that guide, as it will be the most up to date
and will generally be more detailed than what I can provide here.
Extract the ZIP file that was downloaded to the root c:\ destination. By default, this
should extract a folder named mongodb-osx-x86_64-2.4.9. Depending on the tool
you are using for extraction, you can leave this as is or change the destination folder
to simply mongodb. If you don't change the destination during extraction, you should
rename the folder once complete. Either way, be sure that the files that are extracted
reside in a folder named c:\mongodb.
Next, you need to create a default data folder that MongoDB will use to store all
data documents. Using Windows Explorer or Command Prompt, whichever you are
most comfortable with, create the folder c:\data and then c:\data\db by using the
following command:
$ md data
$ md data\db
Once the files have been properly extracted to the c:\mongodb folder and both the
data folders subsequently created, you can now start the MongoDB database server
by executing the following command from a prompt:
$ c:\mongodb\bin\mongod.exe
This should dump a bunch of log statements while the server starts up, but will
ultimately end with:
Sun Mar 16 16:58:05.182 [initandlisten] waiting for connections on
port 27017
That's it! Your MongoDB server is up and running. You can type Ctrl-C to cancel
and shutdown the server.
Linux installation instructions
Once again, we will face a slightly more challenging installation process with Linux
versus Windows or Mac. The official website http://docs.mongodb.org/manual/
administration/install-on-linux/ has great instructions on how to install
MongoDB on a number of different Linux distributions.
We will continue to use Ubuntu as our flavor of choice and use the APT package
manager for the installation.
$ sudo apt-key adv --keyserver hkp://keyserver.ubuntu.com:80 --recv
[ 21 ]
Getting Up and Running
$ echo 'deb http://downloads-distro.mongodb.org/repo/ubuntu-upstart
dist 10gen' | sudo tee /etc/apt/sources.list.d/mongodb.list
$ sudo apt-get update
$ sudo apt-get install mongodb-10gen
Once these steps are completed, MongoDB should be installed and ready to run on
your system. Execute the following command in the terminal to be sure. This gets the
MongoDB daemon up and running, listening for connections:
$ mongod
Sun Mar 16 12:04:20 [initandlisten] waiting for connections on port
Success! Your MongoDB server is up and running. You can type Ctrl-C to cancel
and shut down the server.
It's important to note that as you are performing local development
on your development machine and not a production server, you don't
need the MongoDB server to always be up and running. This will be
an unnecessary strain on your machine for the majority of the time
you're not developing against the server. Because of this, throughout
the remainder of this book it will always be a requirement that you
manually launch the server every time you launch code that expects
to connect to a MongoDB server. If you want, you can certainly
configure MongoDB to run locally as a service and be always up,
but the instructions to do so are beyond the scope of this chapter.
Confirming successful MongoDB installation
Now that MongoDB has been installed on your system, let's run a quick test to
ensure everything is working properly.
Access a command line via your terminal program and execute the following
$ mongod --version
db version v2.4.8
Sun Mar 16 14:17:18.280 git version: a123b456c789d012e345f678
$ mongo --version
Mongod shell version 2.4.8
[ 22 ]
Chapter 2
Assuming that your MongoDB installation was successful, you should see the
version number that was installed as an output on the screen right below the
command you executed.
Note that your version numbers will most likely be more recent
than those printed earlier.
Bookmark the online documentation
You'll want to be sure to point your browser to the online documentation for
MongoDB available at http://docs.mongodb.org/manual/ and bookmark it as it
will undoubtedly become a resource that you will want to access on a regular basis.
Writing your first app
Now that you have everything installed and confirmed that it's all working, you can
write your first quick app that will use both Node and MongoDB. This will prove
that your environment is good to go, and you're ready to get started. In addition, it
will give you a brief taste of the world of Node and MongoDB development! Don't
worry if a lot of the following is confusing or just doesn't make sense to you—it will
all be made clear throughout the rest of the book!
To begin with, we need to create a folder for our application where this application's
specific code will reside, as follows:
$ mkdir testapp
$ cd testapp
Creating the sample application
The testapp folder that we just created will be the root of our sample Node
application. Even though it's not necessary, it's important and also a best practice
that we should create the package.json file for our Node app, which will hold the
necessary data regarding the application such as its version, name, description and
development, and runtime dependencies. This can be done by issuing the following
command from the testapp folder root:
$ npm init
[ 23 ]
Getting Up and Running
This command will follow up by asking you a few questions such as the name and
version number of your newly created application. It is not necessary that you fill in
all the details in one go, and you can skip the steps by pressing Enter and the default
values will be entered, which you may update later.
Getting the dependent modules in place
Before we start to write any Node.js code, we need to get our dependencies in
place by using npm. Since this is a basic app, we will be using it to test our Node.js
connectivity with the MongoDB server. So, the only dependent module that we need
is the native MongoDB client for Node.js. We can easily install that by executing the
following command:
$ npm install mongodb --save
After npm installs the MongoDB driver, you can list the contents of the directory
and notice that a new folder was created, node_modules. This is where, surprisingly
enough, all Node modules are stored whenever you install them from npm. Inside
the node_modules folder should be a single folder named mongodb. Also, you will
notice that the package.json file of our sample application will be updated by this
new dependency entry.
Adding the application code
Now, let's write the simple application code to test things out. This app is going
to basically connect to our locally running MongoDB server, insert a few records
as seed data, and then provide the output on whether or not the data was inserted
properly into MongoDB.
You can download a Gist of the code via the URL: http://bit.ly/1JpT8QL.
Using your editor of choice, create a new file named app.js and save it to the
application root, which is the testapp folder. Just copy the content of the above
Gist on to the app.js file.
Understanding the code
Now, let's go through and explain what each section of the code is doing.
//require the mongoClient from mongodb module
var MongoClient = require('mongodb').MongoClient;
[ 24 ]
Chapter 2
The preceding line requires the MongoDB Node driver that we installed via
npm. This is the required convention used in Node.js for bringing in external file
dependencies to the current file in context. We will explain more about this in the
coming chapters.
//mongodb configs
var connectionUrl = 'mongodb://localhost:27017/myproject',
sampleCollection = 'chapters';
In the preceding code, we declare the variables for the database server information
and collection we want to work with. Here, myproject is the database we want to
use and chapters is the collection. In MongoDB, if you reference and try to use a
collection that doesn't exist, it will automatically be created.
The next step would be to define some data that we can insert into MongoDB to
verify whether everything is fine. So, we create an array of chapters here, which
can be inserted into the database and collections we set up in the previous steps:
//We need to insert these chapters into mongoDB
var chapters = [{
'Title': 'Snow Crash',
'Author': 'Neal Stephenson'
'Title': 'Snow Crash',
'Author': 'Neal Stephenson'
Now, we can take a look at the rest of the code where we insert this data into the
MongoDB database:
MongoClient.connect(connectionUrl, function(err, db) {
console.log("Connected correctly to server");
// Get some collection
var collection = db.collection(sampleCollection);
//here result will contain an array of records inserted
if(!error) {
console.log("Success :"+result.ops.length+" chapters
} else {
console.log("Some error was encountered!");
[ 25 ]
Getting Up and Running
Here, we initiate a connection with the MongoDB server, and if the connection
was proper, the db variable will have the connection object that we can use for
further operations:
MongoClient.connect(url, function(err, db) {
Look at the preceding code closely, do you remember something that we learned
in Chapter 1, Welcome to JavaScript in the Full Stack. We are using a callback for the
connection call that we are making here. As discussed in the first chapter, this function
will be registered as a callback to trigger once the connection attempt is completed.
Upon connection completion, this will be triggered with either an error or a db object
depending on whether we were able to make proper connectivity or not.
So, if you look at the code in the callback function, we are not checking whether any
error was raised in the connection attempt before logging connected correctly to
the server. Now, that's your task to add and check while we try to run this app!
Take a look at the following code block in this section:
var collection = db.collection(sampleCollection);
This does nothing but use the db object we got in the connection call and get the
collection named chapters. Remember that we set that value to sampleCollection
at the beginning of the code. Once we get the collection, we make an insert call to put
the chapters we have defined in the array chapters. As you can see, this insert call
is also done via an asynchronous call by attaching the callback function. This callback
function will be triggered once the insert operation is completed by the code residing
inside the MongoDB native client, which we required as a dependency.
Next, we will take a look at the code inside the callback function, which we passed to
the insert function call.
if(!error) {
console.log("Success :"+result.ops.length+" chapters
} else {
console.log("Some error was encountered!");
Here, we process the values passed via the callback to find out whether the insert
operation succeeded or not and the data related to the records which have been
inserted. So, we check whether there was an error, and, if not, proceed to print the
number of records that got inserted. Here, the result array will contain the records
that we inserted into MongoDB, if the operation was a success.
[ 26 ]
Chapter 2
Now we can go ahead and try to run this code, as we have understood what it does.
Launching the sample app
Once you have the complete code saved to app.js, it's time to execute it and see what
happens. However, before you can launch an app that clearly relies on a connection to
MongoDB, you need to first boot up the MongoDB daemon instance:
$ mongod
In Windows, if you haven't set a PATH variable for mongod, you
may need to use the full path while executing MongoDB, which is
c:\mongodb\bin\mongod.exe. For your needs, the remainder
of this book will refer to the mongod command, but you may
always need to execute the full path in each instance.
Now, to launch the app itself, execute the following command after moving to the
root folder where app.js is located:
$ node app.js
When the app first executes, you should see the following:
Connected correctly to server
Success :2 chapters inserted!
Checking the actual database
Let's take a quick look at the database itself to see what happened during the execution
of the app. Since the server is currently up and running, we can connect to it using the
Mongo shell—a command line interface to the MongoDB server. Execute the following
commands to connect to the server using Mongo and run a query against the chapters
collection. As you can see in the upcoming code, Mongo shell connects to a default
database named test initially. We need to manually specify the database name to
switch to, if its something other than test:
$ mongo
Mongod shell version: 2.4.8
connecting to: test
> use myproject
> show collections
> db.chapters.find().pretty()
[ 27 ]
Getting Up and Running
Here pretty is used as part of the command to format the result from
the find command. This is used only in a shell context. It does more of
a prettification task for the JSON.
You should see something similar to the following output.
'id' : ObjectId("5547e734cdf16a5ca59531a7"),
'Title': 'Snow Crash',
'Author': 'Neal Stephenson'
'id' : ObjectId("5547e734cdf16a5ca59531a7"),
'Title': 'Snow Crash',
'Author': 'Neal Stephenson'
If you run the Node app again, the records will be inserted into the
Mongo server again. So, if you repeat the command multiple times,
the result will have more records in the output. We are not handling
this case in this chapter as we intend to have only specific code,
which will be simple enough to understand.
In this chapter, we took time to make sure your development environment was
properly configured with both the Node runtime environment as well as the
MongoDB server. After making sure both were properly installed, we wrote a
basic app that utilized both technologies. The app connected to a locally running
MongoDB server, and inserted sample records.
Now that the tedious but necessary tasks of setup and installation are out of the way,
we can move on to some fun and start learning!
In the next chapter, we will review a primer on the JavaScript language and
understand the basics of Node. Then we will review the basic CRUD (create,
read, update, delete) operations with MongoDB using the Mongo shell.
[ 28 ]
Node and MongoDB Basics
Before we dig in and start building a full-blown web application using Node and
MongoDB, it's important that we review some of the basics first. This chapter will
give you a crash course on the syntax and important topics. It is broken down
into two parts, where the first half focuses on JavaScript or Node, and the second
half covers MongoDB. You will gain insight into some of the more common and
powerful tools available to you, and a lot of sample code will be reviewed to get
you up to speed.
In this chapter, we will review the following topics:
• Fundamentals of the JavaScript language
• The basics of Node.js
• Node's package manager, npm
• The basics of MongoDB
By the end of this chapter, you should have a solid understanding of the syntax and
how to use both Node and MongoDB. There's a lot to cover, so let's get started.
A JavaScript primer
As we already know Node.js is not just another language, but JavaScript. The language
syntax and tools you are used to while coding JavaScript on the browser will work
verbatim on the server. Node.js has additional tools that are only available on the
server, but the language and syntax again are the same as JavaScript. I'm assuming
you have a general understanding of the basic JavaScript syntax, but I will introduce
JavaScript to you with a very brief primer on the language just in case.
In general, JavaScript is a fairly simple language when it comes to syntax, and you
only need to know a few important elements.
[ 29 ]
Node and MongoDB Basics
Syntax basics
The most basic thing you can do in pretty much any programming language is
declare a variable. Unlike most other languages, JavaScript is a dynamically typed
language, which means when you declare a variable, its value can be of any type and
can change during the course of its lifetime. However, in contrast, a strongly typed
language dictates that a variable defined as a type string must always be a string
and must always have a value of a string.
To declare a variable in JavaScript, simply use the var keyword before your
variable name:
var myVariable;
// declaring a variable with no value
var myFirstName = "Jason";
var myLastName = "Krol";
var myFullName = myFirstName + ' ' + myLastName;
// => Jason Krol
The preceding code snippet shows how we declare variables and define them
with initial values alongside their declarations. The + operator is used for string
Also, we use camel case for the variable names. It is not mandatory that you use
camel case for variable naming, but it is more common in object-oriented languages
to follow camel case as opposed to the underscore-based approach.
JavaScript won't complain if you forget to put a semicolon at the end
of each statement. Instead it will attempt to put the semicolons for you
if there are proper statement terminations missing. This can lead to
unexpected results. The rules of semicolon insertion are explained in
this article at http://bclary.com/2004/11/07/#a-7.9.1.
Apart from var, ES6 introduced two more keywords for variable declarations, namely
let and const, which we will not be discussing in this chapter to avoid confusion.
In the preceding example, the variables we created held string values. JavaScript
supports a number of data types along with string. Let's take a look at them.
[ 30 ]
Chapter 3
Data types
Data types are the base of any language. The data types available in JavaScript are
as follows:
• Number
• String
• Boolean
• Object
• Null
• Undefined
In our code, every variable we declare will contain values that belong to any of
the types above. Number, string, and Boolean types are quite straightforward to
understand. These come under the primitive data types supported by the language.
Here, one important point is to note that JavaScript has no distinction between
integers or floating points at its type level.
Types such as array, function, regex, and date come under the object data type.
They are considered to be composite data types. So the functions that you define
in your code will also be nothing but objects.
Null and undefined are two special types supported by JavaScript. Null points
to a deliberate non-value, whereas undefined points to an uninitialized value.
So when you just declare a variable and haven't yet initialized it with a value,
the variable will be of undefined type.
Understanding the scope of variables
Understanding the scope of variables is very important in JavaScript to gain a better
hold of the language. Scope may be referred to as a bucket in which your variable
or your function exists. Unlike Java and other popular languages, JavaScript follows
function-level scoping as opposed to block-level scoping. So this means that the
variables you define will be limited to a scope that is bound to its parent function.
Consider the following code snippet:
var outer = 10;
function myFunction() {
var inner = 2;
console.log(inner);// 2
console.log(outer);// 10
[ 31 ]
Node and MongoDB Basics
As you can see, the scope of the inner variable is limited to the parent function
named myFunction. It won't be accessible outside it. Also, the variables available
in outer scopes are available in function scopes, and you don't need to make any
extra effort to access them, as you have seen with the variable named outer in the
preceding example.
An important thing to discuss in this context is the use of the var keyword.
JavaScript won't complain if you missed var while declaring a new variable.
But things can get really bad if that happens. See the following example:
(function (){
(function (){
a = 10;
console.log(a);// 10
Here, as the var keyword was skipped along with the variable declaration inside
the inner function, JavaScript considered that this variable should be searched in its
parent scopes, then attached it to the global scope, and ended up making it available
everywhere. So, to avoid such issues with the code, it is always useful if you pass
your code through code quality tools such as JSHint. The preceding code may have
confused you, as it used self-invoked functions just to induce scopes. So let's take a
look at the various ways to define and use functions, arrays, and objects.
Operators and flows
JavaScript supports similar control structures to other languages in the C family.
Conditional statements are written with if and else, and you can chain together
the statements using else-if ladders.
var a = "some value";
if(a === "other value") {
//do something
} else if (a === "another value") {
//do something
} else {
//do something
[ 32 ]
Chapter 3
Control statements can be written using while, do-while, for, and switch
statements. One important thing to consider while writing conditions in JavaScript
is to understand what equates to true and/or false. Any value greater or less than
zero, not null, and not undefined equates to true. Strings such as 0, null, undefined,
or empty strings equate to false.
Some sample examples using while, do-while, for and switch statements are
given below.
// for loop example
var myVar = 0;
for(var i = 0; i < 100; i += 1) {
myVar = i;
console.log(myVar); // => 0 1 ... 99
// do while example
var x = 0;
do {
x += 1;
console.log(x); // => 1 2 ... 100
} while (x < 100);
// while example
while (x > 90) {
x -= 1;
console.log(x); // => 99 98 ... 90
//switch example
var x = 0;
switch(x) {
case 1 :
case 2 :
} // => "none"
[ 33 ]
Node and MongoDB Basics
Another important thing will be to understand the basic difference between
the comparisons using == and ===. The == comparisons should be used where
the type of variable is not your concern; and if the data type of the variables also
should be compared, then you should opt for a === comparison symbol as given
in the following command:
var a = '5';
var b = 5;
if(a == b) {
//do something
if(a === b) {
//do something
Here in the code snippet, the first condition evaluates to true while the second
doesn't. So while you code, it's always safer to depend on strict (===) equality
checks as a best practice.
It's always advised to run your code through code quality tools such as
JS Hint before you approve them for your application. You can automate
the code quality checks via task runners such as Grunt so that each time
we make a change in your code, the code quality tool runs and presents
if there are any potential issues with the code that has been written.
Understanding objects
In JavaScript objects, the arrays and even functions we create fall into the same data
type: Object. Declaring an object is a fairly straightforward process:
var myObject = {};
// that's it!
You may add properties or attributes to this object, which may belong to any type.
It means you can add arrays, functions, or even other objects as properties of this
object. Adding a new property to this object can be done in any of the two ways
shown here:
var person = {};
person.firstName = 'Jason';
person['lastName'] = 'Krol';
// via dot operator
// via square brackets
[ 34 ]
Chapter 3
Let's look at an example where we add arrays and functions as properties of
this object:
var person = {};
person.firstName = 'Jason';
// properties
person.lastName = 'Krol';
person.fullName = function() { // methods
return this.firstName + ' ' + this.lastName;
person.colors = ['red', 'blue', 'green']; // array property
You can see in the preceding code that we defined a basic object called person and
assigned it some properties and a function. It's important to note the use of the this
keyword in the fullName function. The this keyword refers to the object that the
function is a part of. So via the this keyword, the function will be able to access the
other properties that are part of the object it belongs to.
Apart from the approach of adding properties after the object creation, we can also
attach the initial object properties as part of its creation, as follows:
// define properties during declaration
var book = {
title: 'Web Development with MongoDB and NodeJS',
author: 'Jason Krol',
publisher: 'Packt Publishing'
// => Web Development with MongoDB and NodeJS
book.pageCount = 150;
// add new properties
In the preceding example, we are creating objects without specifying any class from
which they should be created by using {}. So, this results in the creation of this new
object from the Object base class from which other composite types such as Arrays
and Functions are extended. So when you use {}, it is equivalent to a new Object().
Here, the objects we create via use of the object literal {} are instances of
the Object class. To define custom classes for our application, we will
need to use functions and prototypes. Mozilla has a fairly good tutorial
on introducing this whole point, available at https://developer.
[ 35 ]
Node and MongoDB Basics
Understanding arrays
Arrays work the same way in JavaScript as they do in pretty much any other
language. They are zero indexed, and you can declare a variable as an empty
array or a pre-populated array. You can manipulate the items in an array, and
arrays are not fixed in length:
favFoods = ['pizza', 'cheeseburgers', 'french fries'];
stuff = [];
// empty array
moreStuff = new Array();
// empty array
firstFood = favFoods[0];
// => pizza
// array functions:
// add new item
// => ['pizza', 'cheeseburgers', 'french fries', 'salad']
// remove the last item
// => ['pizza', 'cheeseburgers', 'french fries']
var first = favFoods.shift();
// remove the first item
// => first = 'pizza';
// => favFoods = ['cheeseburgers', 'french fries']
To be more precise, you can consider arrays as extended child classes of the base
Object class with extra implementations of Array functions.
Understanding functions
Functions are first-class citizens in JavaScript. What this means is that a function
by itself is an object, so it can be treated as such and extended with properties and
additional functions to the base Object class. We will see a lot of situations where
we pass functions as parameters to other functions and return functions from other
function calls.
Here, we will take a standard function (in this case, myFunction). We will assign
this function a timesRun property, just like we do for any other object during its
execution, and see how to refer to that property later:
var myFunction = function() {
this.timesRun += 1;
this.timesRun = 1;
// do some actual work
[ 36 ]
Chapter 3
// => 1;
// => 2;
// => 3;
As we have seen in the preceding example, by using the var keyword, we
can define functions in the same way as variables:
function sayHello() {
// or
var sayHello = function() {
Both methods are almost identical in the preceding sample code. The first method is
the most common way to define a function, and is called the named function approach.
The second method discussed here is the function expression approach, where you
assign the unnamed function as a reference to a variable and keep it unnamed.
The single most important difference between these two approaches is related to a
concept called JavaScript hoisting. Basically, the difference is that when you adopt
a function expression strategy, the function will not be available in its containing
scope till the point in its definition statement gets executed. In the named function
approach, regardless of the position you define it at, the function will be available
throughout the containing scope as given in the following code:
one();//will display Hello
two();//will trigger error as its definition is yet to happen.
function one() {
var two = function() {
two ();//will display Hello
In the preceding sample code snippet, function one can be invoked from anywhere
in this parent scope. But function two will not be available before the point where
its expression is evaluated.
[ 37 ]
Node and MongoDB Basics
JavaScript hoisting is the process by which the function definitions and
variable declarations are moved to the top of the containing scope by
the JS interpreter before the script is executed. So in the previous case
of named functions, the definition was moved to the top of the scope.
However, for the function expression, only the declaration of the variable
moved to the top of the scope, setting it to undefined till the point in
the script where it was actually executed. You can read more about the
concept of hoisting at http://code.tutsplus.com/tutorials/
Anonymous functions and callbacks
Often, you will need to use a temporary function that you don't necessarily want to
declare ahead of time. In this type of scenario, you can use an anonymous function,
which is simply a function that is declared at the time you need it. This is similar to
the function expression context we explored earlier, with a simple difference that the
function isn't assigned to a variable so it has no way of being referenced to later. The
most common use of anonymous functions is when they are defined as a parameter
to another function (most notably when used as a callback).
One of the most common places to use an anonymous function (which also acts as
a callback even if you didn't realize it) is with setTimeout or setInterval. These
are two standard JavaScript functions that will execute code after a specified delay
(in milliseconds) or repeat the execution of code every specified delay. Here is an
example of one of them, setTimeout, using an anonymous inline function:
setTimeout(function() {
}, 5000);
// => Hello...
// (5000 milliseconds i.e. 5 second delay)
// => World!
You can see that the anonymous function was passed as the first parameter to
setTimeout because setTimeout expects a function. You can, if you so desire,
declare the function ahead of time as a variable and pass that to setTimeout
instead of the inline anonymous function:
var sayWorld = function() {
setTimeout(sayWorld, 5000);
// (5 second delay)
// => World!
[ 38 ]
Chapter 3
The anonymous function just acts as a clean inline disposable function.
Callbacks are important because one of the most powerful (and confusing)
features of JavaScript is that it's asynchronous. This means that every line executes
sequentially, but it doesn't wait around for code that might be taking longer than it
should (even if by design). We have explored this via an example in the first chapter
while looking into the asynchronous nature of Node.js.
Mozilla has a detailed tutorial on JavaScript concepts, which we advise
you to go through once you finish this chapter. The tutorial includes
highly advanced concepts, such as closures, that were not covered
in this chapter due to the depth of the topic. So refer to this Mozilla
Development Network article at https://developer.mozilla.
JavaScript Object Notation (JSON) is the standard syntax used when dealing
with data in JavaScript as well as most other languages and web services. The basic
premise of JSON is that it looks exactly like a standard JavaScript object with a few
strict exceptions:
• JSON is pure text. There are no data types with properties; that is, date
values are stored as strings and so on.
• All names and string values must be in double quotes.
• There can be no functions as properties.
Let's take a quick look at a pretty standard JSON object:
"title": "This is the title",
"description": "Here is where the description would be",
"page-count": 150,
"authors": [
{ "name": "John Smith" },
{ "name": "Jane Doe" },
{ "name": "Andrea Johnson" }
"id": "1234-567-89012345"
[ 39 ]
Node and MongoDB Basics
If you are at all familiar with XML, JSON is somewhat similar, except it is much
easier to read and make sense out of. As described best by the ECMA, JSON is a text
format that facilitates structured data interchange between all programming languages.
The basics of Node.js
With the basics of JavaScript out of the way, let's focus on some of the basics of Node.
At its core, one of the most powerful features of Node is that it is event-driven.
This means that almost all the code you write in Node is going to be written in a
way that is either responding to an event or is itself firing an event (which in turn
will fire other code listening for that event).
Let's take a look at the code that we'll write in a later chapter that handles connecting
to a MongoDB server using Mongoose, a popular Node.js MongoDB Object
Document Mapper (ODM) module:
mongoose.connection.on('open', function() {
console.log("Connected to Mongoose...");
First, we tell our mongoose object to connect to the server provided as a string
parameter to the function. Connecting will take an undetermined amount of time
though, and we have no way of knowing how long. So, what we do is bind a listener to
the open event on the mongoose.connection object. With the use of the on keyword,
we are indicating that when the mongoose.connection object triggers an open event,
it executes the anonymous function that was passed in as the parameter.
Asynchronous execution
Earlier, we reviewed the idea of asynchronous JavaScript code in the browser using
setTimeout—the principles apply more strongly in the world of Node. As you
may be making a number of network-dependent connections to different REST API
services, database servers, and anything else, it is important that your code can execute
smoothly and has proper callback usage in place whenever each service responds.
[ 40 ]
Chapter 3
The module system
In an effort to make the code as modular and reusable as possible, Node uses a
module system that allows you to better organize your code. The basic premise is
that you write a code fulfilling a single concern, and export this code as a module
that serves that single purpose. Then, whenever you need to use that code elsewhere
in your code base, you will require that module:
// ** file: dowork.js
module.exports = {
doWork: function(param1, param2) {
return param1 + param2;
// ** file: testing.js
var worker = require('./dowork'); // note: no .js in the file
var something = 1;
var somethingElse = 2;
var newVal = worker.doWork(something, somethingElse);
// => 3
Using this system, it is simple to reuse the functionality in a module (in this case,
the dowork module) in a number of other files. Furthermore, the individual files of
a module act as a private namespace. Any variables declared and used within the
module file are private to that module and not exposed to any code that uses the
module via require().
This system extends infinitely as well. Within your modules, you can require other
modules and so on and so forth.
The Node.js core
The Node.js core literally has hundreds of modules available for you to use while
writing your applications. These include the following:
• Events
• Filesystems
• Net
[ 41 ]
Node and MongoDB Basics
• Streams
• Timers
Definitely make sure to check out the online docs on
Node (at http://nodejs.org/api) to see the full
list of modules available in Node's core and see plenty
of sample code and explanations.
Installing modules using npm
The module system in Node is so powerful that consuming a third-party module
written by other developers is a piece of cake. Node includes its own package
manager called npm, which is a registry that currently contains over 60,000 unique
modules written in Node. These modules are completely open source and available
to you via a few short commands. In addition, you can release your own personal
modules via npm and allow anyone in the world to use your feature!
Let's say you wanted to include a popular web framework, Express, in your project
(the one we will be using later in this book). There are simply two steps required to
download a module and use it in your code:
$ npm install express
// ** file: usingnpm.js
var express = require('express');
And that's it! Literally, it's that simple! From the command-line of the folder where
your project is located, simply execute npm install package-name, and the
package will be downloaded from npm and stored in a folder called node_modules
within your project. If you browse through the node_modules folder, you will find
a folder for the package you installed, and within that folder, you will find the raw
source code for the package itself. Once the package is downloaded, it's as simple as
using require() from within your code.
There may be times when you want to install a Node package globally, for example,
when using a popular command-line build tool called Grunt.js. To install an npm
package globally, simply include the -g or --global flag, and the module will be
installed as a global executable instead. When installing npm packages globally,
the source files for the package are not stored within the node_modules folder of a
specific project, but instead within a node_modules folder in a system directory of
your machine.
[ 42 ]
Chapter 3
A really powerful feature of npm is that it allows a quick, easy, and consistent way
for other developers to boot up your code in their local environment. Node projects
typically include a special file called package.json that includes information
about the project as well as a list of all npm packages that the project depends on.
A developer with a copy of your local code can simply execute npm install to
have every dependency downloaded and installed locally using this file.
The npm install flag --save or --save-dev is required if you want the dependency
you are installing to be saved to the package.json file. If you are starting a new
project and don't want to create a package.json file by hand, you can simply
execute npm init and answer a few quick questions to get a default package.json
file quickly set up. You can leave every question blank during init and accept the
default values if you want:
$ npm init
$ npm install express --save
$ npm install grunt --save-dev
$ cat package.json
"name": "chapter3",
"version": "0.0.0",
"description": "",
"main": "index.js",
"scripts": {
"test": "echo \"Error: no test specified\" && exit 1"
"author": "",
"license": "ISC",
"dependencies": {
"express": "^3.5.1"
"devDependencies": {
"grunt": "^0.4.4"
[ 43 ]
Node and MongoDB Basics
Note that the dependencies and devDependencies sections have express and
grunt listed. The difference between these two sections is that the dependencies
section is absolutely critical for the app to function properly. The devDependencies
section has only packages that need to be installed for a developer to use during
the development of the project (such as Grunt for various build steps, testing
frameworks, and so on).
If you are confused about the use of the ^ symbol in the package
versions, its used to update the dependency to the most recent
minor version or patch version (the second or third number).
^1.2.3 will match any 1.x.x release including 1.3.0, but will hold
off on 2.0.0. So in our case ^3.5.1 of Express.js will look for the
most recent minor version of Express but will not take 4.0.0, as
it's a major version.
The basics of MongoDB
Since MongoDB is largely powered by JavaScript, the Mongo shell acts as a JavaScript
environment. In addition to being able to execute regular Mongo queries, you can also
execute standard JavaScript statements. Most of the items mentioned earlier in the
JavaScript primer apply directly to the Mongo shell as well.
In this next section, we will focus primarily on the various ways to perform standard
create, read, update, delete (CRUD) operations via the Mongo shell.
The Mongo shell
To access the Mongo shell, simply execute mongo from any terminal. The Mongo
shell requires the mongod server to be currently running and available on the
machine, as the first thing it does is connect to the server. Use the following
command to access the Mongo shell:
$ mongo
MongoDB shell version: 2.4.5
connecting to: test
By default, when you first launch Mongo, you are connected to the local server and
set to use the test database. To display a list of all databases on the server, use the
following command:
> show dbs
[ 44 ]
Chapter 3
To switch databases to any of those listed in the output of show dbs, use the
following command:
> use chapter3
switched to db chapter3
An interesting thing to note is that if you use use on a database that doesn't exist,
one is created automatically. If you are using an existing database and want to view
a list of collections in the database, execute the following command:
> show collections
In the case of my chapter3 database, I had no existing collections since it was
automatically generated as a new database for me. Collections in MongoDB are
similar to the tables in a relational database.
Inserting data
Since we are working with the chapter3 database, which is brand new, there are
currently no collections in it. You can use any collection (table) you want by simply
referring to a new collection name with the db object:
> db.newCollection.find()
Performing a find operation on an empty collection simply returns nothing. Let's
insert some data so we can experiment with some queries:
> db.newCollection.insert({ name: 'Jason Krol', website:
'http://kroltech.com' })
> db.newCollection.find().pretty()
"_id" : ObjectId("5338b749dc8738babbb5a45a"),
"name" : "Jason Krol",
"website" : "http://kroltech.com"
After we perform a simple insertion (basically of a JavaScript JSON object), we perform
another find operation on the collection and get our new record returned, this time
with an additional _id field added. The _id field is Mongo's method for tracking a
unique identifier for every document (record). We also chained the pretty() function
to the end of the find, which outputs the results a little more nicely.
Go ahead and insert a few more records so you have some data to play with in the
next section when we go over querying.
[ 45 ]
Node and MongoDB Basics
Querying and searching for documents in a MongoDB collection is pretty
straightforward. Using the find() function by itself with no parameters will
return every document in the collection. To narrow down the search results, you
can provide a JSON object as the first parameter with as much or as little specific
information to match against as you wish, as shown in the following code:
> db.newCollection.find({ name: 'Jason Krol' })
{ "_id" : ObjectId("533dfb9433519b9339d3d9e1"), "name" : "Jason
Krol", "website" : "http://kroltech.com" }
You can include additional parameters to make the search more precise:
> db.newCollection.find({ name: 'Jason Krol', website:
{ "_id" : ObjectId("533dfb9433519b9339d3d9e1"), "name" : "Jason
Krol", "website" : "http://kroltech.com" }
Note that with each result set, every field is included. If you want to only return a
specific set of fields with the result, you can include a map as the second parameter
to find():
> db.newCollection.find({ name: 'Jason Krol' }, { name: true })
{ "_id" : ObjectId("533dfb9433519b9339d3d9e1"), "name" : "Jason Krol"
> db.newCollection.find({ name: 'Jason Krol' }, { name: true, _id:
false })
{ "name" : "Jason Krol" }
Note that the _id field will always be included by default unless you specifically
state that you don't want it included.
Additionally, you can use query operators to search for things that are within ranges.
These include greater than (or equal to) and less than (or equal to). If you want to
perform a search against a collection of homework, and you want to find every
document with a score within the B range (80–89), you can execute the following
> db.homework_scores.find({ score: { $gte: 80, $lt: 90 } })
Finally, you can use regex while performing a search to return multiple
matching documents:
> db.newCollection.find({ name: { $regex: 'Krol'} })
[ 46 ]
Chapter 3
The preceding query will return every document that contains the word Krol.
You can get as advanced as you want with regex statements.
If you know that you are going to be returning multiple documents on a query and
only want the first result, use findOne() in place of a regular find() operation.
Updating data
To update a record, use the update() function, but include a find query as the
first parameter:
> db.newCollection.update({ name: 'Jason Krol' }, { website:
'http://jasonkrol.com' })
There's a bit of a catch here. If you perform a new find({ name: 'Jason Krol'
}) operation, something strange happens. No data is returned. What happened?
Well, the second parameter in the update() function is actually the new version of
the complete document. Since you only wanted to update the website field, what
actually happened was that the document that was found was replaced with the new
version that consists of only the website field. To reiterate, the reason this happens
at all is because with NoSQL such as MongoDB, the document does not have a set
number of fields (as a relational database does). To fix this problem, you should use
the $set operator instead:
> db.newCollection.update({ name: 'Jason Krol' }, { $set: { website:
'http://jasonkrol.com'} })
There may be a time when you want to update a document, but the document itself
may or may not exist. What happens when the document does not exist, and you'd
like a new one to be created instantly based on the updated values you provide?
Well, there's a handy function just for that. Pass {upsert: true} as the third
parameter to the update() function:
> db.newCollection.update({ name: 'Joe Smith' }, { name: 'Joe Smith',
website: 'http://google.com' }, { upsert: true })
If we have a document with a name field that matches Joe Smith, the website
field will be updated (and the name field preserved). However, if we do not have
a matching document, a new one will be created automatically.
[ 47 ]
Node and MongoDB Basics
Deleting data
Deleting documents works almost exactly like find(), except instead of finding and
returning results, it deletes those documents that match the search criteria:
> db.newCollection.remove({ name: 'Jason Krol' })
If you want the nuclear option, you can use the drop() function, which will remove
every document in a collection:
> db.newCollection.drop()
Additional resources
For additional learning with JavaScript, I suggest you check out some of the
following resources:
• Mozilla Developer Network at https://developer.mozilla.org/en-US/
• Secrets of the JavaScript Ninja, John Resig, Bear Bibeault, Manning
• Learning JavaScript Design Patterns, Addy Osmani, O'Reilly
• JavaScript: The Good Parts, Douglas Crockford, O'Reilly
The Node API online documentation is going to be your best bet for fully
understanding everything that's available within the Node core set of modules.
The Node API docs can be found at http://nodejs.org/api.
Additionally, there is a great website that teaches Node using actual programming
problems that you must solve. The emphasis with these exercises is to understand
the nuts and bolts of how Node works and get down into the fundamentals of
working with streams, asynchronous I/O, promises, and more. Node school can
be found at http://nodeschool.io.
Finally, the creators of MongoDB offer an amazing 7–8 week online training and
certification program completely free of charge, where you will learn everything
you need to be a true MongoDB master. This can be found at MongoDB University
at https://university.mongodb.com.
Now it's time to dive in and start writing some real code!
[ 48 ]
Chapter 3
In this chapter, you took a crash course on the basics of JavaScript, Node.js,
and MongoDB. In addition, you learned about Node's package manager, npm.
For further learning, additional resources were provided for JavaScript, Node.js,
and MongoDB.
In the next chapter, you will write your first Node web server using Express.js and
get started with creating a complete web application.
[ 49 ]
Introducing Express
When we need to build a full-fledged web application, writing the whole application
from scratch is not the best approach to take. We can think of using a well-maintained
and well-written web application framework to build our application to reduce the
development effort and increase the maintainability.
In this chapter:
• We will explore the Express.js web application framework
• We will explore the various elements of Express.js
• We will develop the necessary code to bootstrap a web application
using Express
Web application frameworks
Simply put, a web framework makes it easier to develop a web application. Consider
the aspect of breaking down commonly used functionality into reusable modules.
That is exactly what frameworks do. They come with a lot of reusable modules
and enforce a standard structure for the code so that it will be easier for developers
across the world to go through and understand the application. Apart from all
these advantages, web frameworks mostly get maintained by lots of developers
across the world. So, the effort of developers in incorporating the newer bug fixes
and features of underlying languages is minimized to a step where we just need to
upgrade the framework version that is being used by the application. So, the use
of web frameworks for building web applications brings a lot of advantages to the
development and maintenance phases of an application.
[ 51 ]
Introducing Express
The Express.js framework, which we are going to use throughout this book, is a
Model View Controller (MVC) based web application framework. MVC is just an
architectural design pattern:
• Models are used to represent the data or entities of the web application.
They lie more close to the instances, which store the data of the
application—typically a database or a web service.
• Views are responsible for how the application gets presented to the end user.
So, a view can be considered as the presentation layer of the application.
• Now, you may be wondering about the role of controllers in the web
application. Well, the role of controllers is nothing but to just glue together
the models with the respective views and to take care of the request from the
user for a particular web page in our application end to end.
This may be a bit hard for you to grasp if you are hearing this concept for the first
time. But after going through this chapter, you will get used to these concepts while
we present the various examples to you.
What is Express.js?
As described perfectly on its home page, Express is a minimal and flexible Node.js
web application framework, providing a robust set of features for building single,
multi-page, and hybrid web applications. In other words, it provides all the tools
and basic building blocks you need to get a web server up and running by writing
very little code. It puts the power to focus on writing your app and not worry about
the nuts and bolts that go into making the basic stuff work in your hands.
The Express framework is one of the most popular Node-based web frameworks and
one of the single most popular packages available in npm. It is built based on the
Sinatra web framework, which is quite popular in the Ruby world. There are a lot
of frameworks across languages that take inspiration from Sinatra's simplicity such
as PHP's Laravel framework. So Express is the Sinatra-based web framework in the
Node.js world.
[ 52 ]
Chapter 4
If you look at a sample piece of code, one of the most basic implementations of
Express, you can see how easy it is to get a web server up and running, for example:
var express = require('express');
var app = express();
app.get('/', function(req, res){
res.send('Hello World');
The beauty of Express is that it makes building and maintaining the server code for a
website simple.
Building a complete web application
Beginning with this chapter, we are going to build a complete web application.
The web application that we build will be a clone of a popular social image-sharing
site, imgur.com. We'll call our site imgPloadr.io.
Designing the web application
The requirements of the site are as follows:
• The home page will allow visitors to upload an image as well as browse the
existing uploads, which will be ordered based on newest to oldest.
• Each uploaded image will be presented via its own page that shows its title,
description, and a large image display. Visitors will be able to like the image
and post comments.
• A consistent shared sidebar will be visible on both pages and will showcase
some general statistics about the site, the most popular images, and the most
recent comments.
The site will use Bootstrap, so that it has a nice professional design and is responsive
on any device.
[ 53 ]
Introducing Express
The following screenshot is from the home page of the completed site:
The next screenshot is an image's details page from the site:
[ 54 ]
Chapter 4
Organizing the files
Before you start writing any code, we want to make sure that you have a project
folder set up correctly with the proper folder structure to house all of the various
files that you will be creating. Get started by creating a new folder for your project,
and name it anything you like. Then, inside that folder, create additional folders to
match the following structure:
/(project root)
Each of these folders will contain important modules that we will write throughout
the remainder of this chapter and book.
If you use an Express-based generator through Yeoman,
you will get the necessary folder structure and the
dependencies imported with boilerplate code. However,
as our intention is to understand this framework, we
will skip this. Visit http://yeoman.io/ for more
information on the features of Yeoman.
You are going to need a package.json file for this project, and the easiest way to
create one of these is by simply executing the following command from the root of
the project folder:
$ npm init
Respond to each of the questions as you are prompted or simply press Enter
repeatedly to accept the default values. Now, let's install Express and its necessary
dependencies via npm:
$ npm install express morgan body-parser cookie-parser methodoverride errorhandler express-handlebars --save
[ 55 ]
Introducing Express
This will install the Express framework in the node_modules folder and also add
Express to the package.json file in the dependencies section. Note that at the time of
writing this book, Express.js is in its 4.x.x versions. Here, as you can see, Express is a
completely decoupled framework that doesn't come with a lot of packaged modules
within itself. Instead, you can figure out the dependencies of your application and
plug in and out of the application anytime. If you have been following Express
development from the beginning, you must have noted that these changes where
introduced as part of the Express 4.x.x versions. Prior to this version, Express used
to come with a lot of built-in modules bundled within it. Here, the modules we
installed alongside Express are the various dependencies our application has when
we start building the complete web application. We will discuss the use of each
module in a later section of this chapter!
Creating the application's entry point
The next step in developing the application after installing Express and the necessary
dependencies will be to create a file, which will serve as the default entry point
of this application. We will be executing this file to start our web application and
it will contain the necessary code to require dependent modules and boot up the
application to listen to a specified port in the development server.
We are going to name the entry point file server.js for now and will keep it pretty
lean so that its contents are quite self-explanatory. Any major logic that is going to be
executed within this file will actually be deferred to external modules hosted within
other files.
Before we can do anything within server.js, we require a few modules that we're
going to work with, specifically Express:
var express = require('express'),
// config = require('./server/configure'),
app = express();
In the preceding code, we are assigning the express module to the express variable.
The config module is actually going to be our own module that we will write shortly,
but for now, since it doesn't exist, we will leave that line commented out. Finally, we
will declare a variable called app that is actually what the Express framework returns
when it is executed. This app object powers our entire app application, which is how it
was so cleverly named.
[ 56 ]
Chapter 4
Throughout this chapter and the remainder of the book, I may
include commented out code in the samples (code that starts
with //). This is so that following along will be easier when we
use the commented lines as reference points, or we can enable
those features by simply uncommenting the code.
Next up, we will set a few simple settings via the app object using the app.set()
function. These settings are really just a way for us to define some app-level
constants that we can use throughout the rest of our code as handy shortcuts:
app.set('port', process.env.PORT || 3300);
app.set('views', __dirname + '/views');
// app = config(app);
The code is explained as follows:
• The first two lines of the preceding code use built-in constants in Node.
The process.env.PORT constant is an environment setting that is set on the
actual machine for the default port value to the server. If no port value is set
on the machine, we will hardcode a default value of 3300 to use in its place.
• After that, we set the location of our views (HTML templates) to
__dirname + '/views or, using another Node constant, the /views
folder from within the current working directory.
• The third line of code is referencing the config module, which we haven't
written yet, so that the line is commented out.
• Last but not least, we will create a HTTP server using our app object and tell
it to listen for connections:
app.get('/', function(req, res){
res.send('Hello World');
app.listen(app.get('port'), function() {
console.log('Server up: http://localhost:' + app.get('port'));
Here, we set a route in our application to respond with a Hello world message. If
any user requests for the root of our application, it will respond with a Hello World
message. The last section of the code is to call the listen() function on our app that
tells it which port to listen to and to pass in a simple anonymous callback function that
will execute once the server is up and listening by executing a simple console.log()
message. That's it! Again, make sure to save this file with the name server.js within
the root of the project. You're ready to run your server and see if it works.
[ 57 ]
Introducing Express
Booting up the application
Let's take your server for a spin and see how you're doing so far:
$ node server.js
Server up: http://localhost:3300
Perfect! At this point, your server doesn't actually do anything great. Try pointing
your browser to http://localhost:3300. You should receive a pretty basic
message that says Hello World! If you request any other routes on the port, such
as http://localhost:3300/, it will respond with a cannot get response. This is
because you haven't configured any routes or any actual logic in your server to
say how to handle certain requests, but only a GET request to the default route of /.
Before you set up your routes, we should understand the concept of middleware in
Express, which will be essential to understanding how the modules that are custom
dependencies of our application, get integrated with our normal application flow.
You can set any number of environment variables right from
the command line before you run your server by executing
something like the following command:
$ PORT=5500 node server.js
Server up: http://localhost:5500
You can also set environment variables in your environment
settings permanently. This can be done typically by editing
your .profile file or equivalent.
Using and understanding middleware
One of the most powerful features available with Express is the concept of
middleware. The idea behind middleware is that it acts like a stack of filters that
every request to your server passes through. Since every request passes through
each filter, each filter can perform a specific task against the request before it passes
through to the next filter. Typically, these filters are used for tasks such as cookie
parsing, form field handling, session handling, authentication, and error handling
and logging. The list goes on and on and you can use hundreds of third-party
modules as well as simply write your own.
The order that the middleware is called in is very important. Again, using the
concept of filters, as a request passes through each filter, you want to be sure that
they are performing their responsibilities in the correct order. A great example
of this is implementing a cookie parser before a session handler—since sessions
typically rely on cookies to maintain state with a user between requests.
[ 58 ]
Chapter 4
Another great example of how the order of middleware is important involves error
handling. If any of your middleware encounters an error, they will simply pass that
error along to the next middleware in the stack. If the last middleware, regardless
of what it is, doesn't gracefully handle that error, it will basically show up in your
application as a stack trace (and that's bad). Having an error handler configured as
one of the last middleware is like saying if everything else fails, and at any point in the
previous middleware a failure occurs, deal with it gracefully.
The various dependencies we have installed to use in this application are going to
be integrated into our code as middlewares. We are going to carry out this task of
integrating the various middlewares through the config module as it will help us
to make our server.js file leaner to add more readability to the code base.
The configure module
Since we are leaving our server.js file very lean, there is still a fair amount of logic
that is required in configuring our server. For this, we will defer to a custom module
that we'll create called configure. To get started, create a configure.js file in the
server folder. We have already installed the custom dependencies when we were
installing Express in the first place.
Now that the module is installed and ready to be used, let's start writing the
configure.js file. First, like any of our modules, we will declare our dependencies:
var path = require('path'),
routes = require('./routes'),
exphbs = require('express-handlebars'),),
express = require('express'),
bodyParser = require('body-parser'),
cookieParser = require('cookie-parser'),
morgan = require('morgan'),
methodOverride = require('method-override'),
errorHandler = require('errorhandler');
module.exports = function(app) {
routes(app);//moving the routes to routes folder.
[ 59 ]
Introducing Express
app.use('/public/', express.static(path.join(__dirname,
if ('development' === app.get('env')) {
return app;
In the preceding code, we declared variables for each of the modules that we will
be using in our custom configure module. Then, we defined the actual module
that will be exported by this code file, more specifically a function that accepts our
app object as a parameter as well as returns that same object (after we make some
configuration modifications to it).
You should see that we require Connect, which is actually
installed by default with Express.js as it's one of its core
dependencies. Connect is a popular third-party middleware
framework that we will learn more about later in this chapter.
Let's take a look at each of the Connect middleware we have used in the
preceding code:
• morgan: This is the module responsible for logging. This is very helpful for
debugging your Node server.
• bodyParser: This helps facilitate the packing of any form fields that are
submitted via a HTML form submission from a browser. Form fields
that are submitted via a POST request will be made available via the
req.body property.
• methodOverride: For older browsers that don't properly support REST
HTTP verbs, such as UPDATE and PUT, the methodOverride middleware
allows this to be faked using a special hidden input field.
• cookieParser: This allows cookies to be sent and received.
• errorHandler: This handles any errors that occur throughout the entire
middleware process. Typically, you would write your own custom
errorHandler that might render a default 404 HTML page, log the
error to a data store, and so on.
• Handlebars: This is the templating engine we are going to use with the
views. We will explain more about integrating it in the upcoming sections.
[ 60 ]
Chapter 4
The routes(app) line is a special component of Express that says you are actually
using a router with your server, and you can respond to requests such as GET,
POST, PUT, and UPDATE. Since you are using the Express router as one of the last
middleware, we will also define the actual routes in the next section.
Finally, the express.static() middleware is used to render static content files to
the browser from a predefined static resource directory. This is important so that
the server can serve up static files, such as .js, .css, images, and regular .html,
as well as any other files you might need to serve up. The static middleware will
serve up any static files from the public directory like the following code:
It's important that your static middleware is defined after the app.router() so that
static assets aren't inadvertently taking priority over a matching route that you may
have defined.
Activating the configure module
Now that your configure.js file is complete, you're ready to call it from your main
server.js file. If you recall, we included two lines of code that were commented out
for our configure module. It's time to uncomment these two lines so that when you
run your server, your configure module will do its part. The two lines should now
look like this:
config = require('./server/configure'),
app = config(app);
Boot up your server again by executing the server.js node and everything should
still be running smoothly. Now, it's time to incorporate more routes into our
application in addition to the Hello World route we added earlier.
Routers and controllers
So far, you have your server.js file and a configure module that is used to wire
up all of the necessary middleware for the application. The next step is to implement
proper routers and the necessary controllers.
[ 61 ]
Introducing Express
The router is going to be a map of each of the available URL paths for the app. Every
route on the server will correspond to a function in a controller. Here is what our
routes table will look like for the particular application we are writing:
GET /(index) - home.index (render the homepage of the site)
GET /images/image_id - image.index (render the page for a specific
POST /images - image.create (when a user submits and uploads a new
POST /images/image_id/like - image.like (when a user clicks the Like
POST /images/image_id/comment - image.comment (when a user posts a
You can see that we are handling two different GET requests and three different POST
requests. In addition, we have two main controllers: home and image. Controllers
are really just modules with different functions defined that match up with the
corresponding routes. As pointed out earlier, they are called controllers in the Model
View Controller (MVC) design pattern. Typically, every route will correspond to
a controller. This controller will more than likely render a view, and that view will
more than likely have its own model (any data that is displayed in the view).
Let's write our router as its own module, matching the table I outlined. First, create a
routes.js file within the server folder. The routes file is going to be pretty simple,
and the only dependencies it requires will be the controllers we define:
var express = require('express'),
router = express.Router(),
home = require('../controllers/home'),
image = require('../controllers/image');
module.exports = function(app) {
router.get('/', home.index);
router.get('/images/:image_id', image.index);
router.post('/images', image.create);
router.post('/images/:image_id/like', image.like);
router.post('/images/:image_id/comment', image.comment);
Right off the bat, we declare a router variable and require each from the
controllers folder to assign each application route (we haven't yet created
these files but that's coming up next). Here, we are assigning each route to
its corresponding function in the controllers. Then, we export a module that,
when invoked by itself, will attach all these routes to the app instance.
[ 62 ]
Chapter 4
The first parameter for a route is the string value of the route itself, which can contain
variable values as subpaths. You can see with second router.get, we assign a route
value of /images/:image_id that basically equates to /image/ANYVALUE in the
browser address bar. When we write the image.index controller, you will see how
to retrieve the value for :image_id and use it within the controller function itself.
The second parameter for a route is a callback function. You can completely omit
the idea of using controllers and just define your callbacks as inline anonymous
functions; however, as your routes grow, this file will get larger and larger, and the
code will start to become a mess. It's always good practice to break your code up into
as many small and manageable modules as possible to keep yourself sane!
The first two router.get routes are typical routes that would be called whenever a
visitor points their browser to yourdomain.com/routepath—the browser sends a
GET request to the server by default. The other three router.post routes are defined
to handle when the browser posts a request to the server, typically done via a HTML
form submission.
With all of our routes defined, let's now create the matching controllers. Within the
controller's folder, create both the home.js and image.js files. The home.js file is
very basic:
module.exports = {
index: function(req, res) {
res.send('The home:index controller');
With this module, we are actually exporting an object that has a single function called
index. The function signature for index is the signature that is required for every
route using Express. The first parameter is a request object and the second parameter
is a response object. Every detail specific to the request that the browser sent to the
server will be available via the request object. In addition, the request object will
be modified using all of the middleware that was declared earlier. You will use the
response object to send a response back to the client—this may be a rendered HTML
page, static asset, JSON data, an error, or whatever you determine. For now, our
controllers just respond with simple text so you can see that they are all working.
[ 63 ]
Introducing Express
Let's create an image controller that has a few more functions. Edit the
/controllers/image.js file and insert the following code:
module.exports = {
index: function(req, res) {
res.send('The image:index controller ' +
create: function(req, res) {
res.send('The image:create POST controller');
like: function(req, res) {
res.send('The image:like POST controller');
comment: function(req, res) {
res.send('The image:comment POST controller');
Here, we defined the index function, just like we did in the home controller, except
that we will also display image_id, which is set in the route when this controller
function is executed. The params property was added to the request object via the
urlencoded feature, which is part of the body parser module!
Note that neither controller currently requires any dependencies (there were no
require declarations defined at the top of the file). This will change as we actually
flesh out the controller functions and start to do things such as inserting records into
our MongoDB database and using other third-party npm modules.
Now that your controllers are created and ready to be used, you just need to activate
your routes. To do this, we will insert one last line of code into our configure.js
file, right above the return app; line:
Don't forget to uncomment the routes = require('./routes') line at the top of
the file as well. What we're doing here is using the routes module we defined and
executing the initialize function, which will actually wire up our routes via our app
object. We will need to comment out the redundant code that we just moved to routes,
which is still present in server.js, as you can see in the following code snippet:
[ 64 ]
Chapter 4
As a recap of each of the files you have created so far, here are the uninterrupted files
listed so you can view the full code:
First, we need to boot up with server.js:
var express = require('express'),
config = require('./server/configure'),
app = express();
app.set('port', process.env.PORT || 3300);
app.set('views', __dirname + '/views');
app = config(app);
// app.get('/', function(req, res){
res.send('Hello World');
// });
var server = app.listen(app.get('port'), function() {
console.log('Server up: http://localhost:' + app.get('port'));
Next, we will configure the server with server/configure.js:
var path = require('path'),
routes = require('./routes'),
exphbs = require('express-handlebars'),
express = require('express'),
bodyParser = require('body-parser'),
cookieParser = require('cookie-parser'),
morgan = require('morgan'),
methodOverride = require('method-override'),
errorHandler = require('errorhandler');
module.exports = function(app) {
uploadDir:path.join(__dirname, 'public/upload/temp')
app.use('/public/', express.static(path.join(__dirname,
[ 65 ]
Introducing Express
if ('development' === app.get('env')) {
return app;
Then, we have our routes defined in server/routes.js:
var express = require('express'),
router = express.Router(),
home = require('../controllers/home'),
image = require('../controllers/image');
module.exports = function(app) {
router.get('/', home.index);
router.get('/images/:image_id', image.index);
router.post('/images', image.create);
router.post('/images/:image_id/like', image.like);
router.post('/images/:image_id/comment', image.comment);
Finally, we will define our controllers with controllers/home.js:
module.exports = {
index: function(req, res) {
res.send('The home:index controller');
Furthermore, we will also define our controllers with controllers/image.js:
module.exports = {
index: function(req, res) {
res.send('The image:index controller ' +
create: function(req, res) {
res.send('The image:create POST controller');
like: function(req, res) {
res.send('The image:like POST controller');
comment: function(req, res) {
res.send('The image:comment POST controller');
[ 66 ]
Chapter 4
Let's fire up the server one last time and check if it's all working.
Execute the server.js node, and this time point your browser to
http://localhost:3300. Now, you should be seeing some responses in the
browser. Go to http://localhost:3300/images/testing123. You should
see The image:index controller testing123 on the screen!
Custom middleware
There will undoubtedly come a time when you want to write your own custom
middleware in addition to the existing middleware provided by Connect or any
other third party. Before you write your own custom middleware in Node, make
it a habit to search through https://www.npmjs.org/ first, as there's a fairly big
chance someone else has already done the work.
Writing your own custom middleware is pretty simple. You simply need to write a
function that accepts four parameters: err, req, res, and next.
• The first parameter is an error object, and if there were any stack errors prior
to your middleware running, that error will be passed to your middleware
so that you can handle it accordingly.
• You are already familiar with the req and res parameters, having written
your routes.
• The fourth parameter is actually a reference to a callback. This next
parameter is how the middleware stack is able to behave like a stack—each
executing and ensuring that the next middleware in the pipeline is returned
and called via next. Here is a super basic example of a custom middleware:
app.use(function(err, req, res, next) {
// do whatever you want here, alter req, alter res, throw err,
return next();
The only important thing to keep in mind when writing your own custom
middleware is that you have the correct parameters and that you return next().
The rest is completely up to you!
[ 67 ]
Introducing Express
Handlebars as view engines
By default, Express can and will happily render static HTML documents and serve
them back to the client. However, unless you're building a purely static, contentdriven site, which is doubtful, you're more than likely going to want to render your
HTML dynamically. That is, you want to generate portions of the HTML on the fly
as pages are requested, perhaps using loops, conditional statements, data-driven
content, and so on. In order to render dynamic HTML pages, you need to use a
rendering engine.
This is where Handlebars comes in. The rendering engine is given its name because
of the syntax it uses to display data, namely double pairs of braces, {{ and }}. Using
Handlebars, you can have sections of your HTML pages that are determined at
runtime based on data passed to it. For example:
<p>Hello there {{ name }}!
Todays date is {{ timestamp }}</p>
The actual HTML that would wind up on a visitor's browser would be:
<p>Hello there Jason!
Todays date is Sun Apr 13</p>
The first thing we want to take care of in our configure module is to register
Handlebars as the default view rendering engine. In the configure.js file, above
the return(app); line, you should insert the following code:
app.engine('handlebars', exphbs.create({
defaultLayout: 'main',
layoutsDir: app.get('views') + '/layouts',
partialsDir: [app.get('views') + '/partials']
app.set('view engine', 'handlebars');
First, using the Express app object that was passed into the configure function,
we define our rendering engine of choice by calling the engine function of app. The
first parameter to the engine function is the file extension that the rendering engine
should look for, namely handlebars.
[ 68 ]
Chapter 4
The second parameter builds the engine by calling the express-hbs module's
create function. This create function takes an options object as a parameter, and
this options object defines a number of constants for our server. Most importantly,
we will define which layout is our default layout and also where our layouts will
be stored. If you recall, in server.js, we used app.set to set a views property of
our app that pointed to the current working directory + /views. This setting is used
when we configure the options for our rendering engine as well. You'll notice that
the partialsDir property uses an array (with a single item) and a single string
value for layoutsDir. Both of these methods are interchangeable, and I just wanted
to demonstrate that you could have more than one partials directory, and it could
just be an array of string values.
With that set, our server now knows that any time we try to render a HTML page
that has a file extension of handlebars, it will use the Handlebars engine to perform
the rendering. This means that we need to be sure to use Handlebars-specific syntax
in our dynamic HTML pages.
We will be learning more about Handlebars and how to write dynamic HTML pages
in the next chapter.
Using .handlebars as a file extension was purely a personal
choice. Some people prefer .hbs, and if you want, you can use
anything you like. Just make sure that the first parameter to
the app.engine() function and the second parameter in the
app.set('view engine') function are identical.
To learn more about the many template engines available for
use with Node, check out this list on the official Joyent GitHub
wiki at https://github.com/joyent/node/wiki/
In this chapter, we learned about the Express web framework for Node and wrote
a basic web server using Express that will act as the foundation for the image
uploading website that we will build throughout the remainder of the book.
The web server you wrote handles requests to specific routes, uses controllers to
handle the logic for those routes, and supports all of the standard requirements a
typical web server should.
In the next chapter, we will cover using the Handlebars template engine to write
each of the dynamic HTML pages that the website needs. In addition, we will update
the image and home controllers to include the necessary logic to properly render
those HTML pages.
[ 69 ]
Templating with Handlebars
In this chapter, we will be exploring the following topics:
• The Handlebars templating framework
• The steps to develop the templates necessary to build the presentation layer
of our application
Before we start exploring the capabilities of Handlebars, we need to have an idea of
what a templating framework generally does.
Templating engines
As we already know, a Model View Controller (MVC) application framework
divides the application-specific code into models, views, and controllers. Controllers
are supposed to be handling the task of binding the appropriate data to its relevant
views to generate the output for an incoming web application request. So, views
are supposed to be independent of the data and only contain code relevant to the
presentation of the data, which will be mostly HTML. Apart from HTML, views will
need to contain presentation logic, which will be conditions written on the data passed
to them via controllers. Then, the main task that templating frameworks do in this case
is that they make this process of embedding presentational logic simpler and readable.
Also, they attempt to segregate the views into more understandable subcomponents.
[ 71 ]
Templating with Handlebars
Server-side and client-side templating
Templating solutions can be generally classified into two as client-side and
server-side templating solutions. The web applications we build usually follow
a server-side or client-side templating approach or even a hybrid of both.
Client-side templating
Imagine a case where the web application, after loading the page, makes an API call
via AJAX and gets a JSON response in return. How will it render the data it received
into its corresponding HTML? Client-side templates are required in this case to
keep our JavaScript code neat and clean, or else we will end up putting too much
unreadable HTML code as strings inside the JavaScript code. Client-side templating
frameworks allow us to dump the templates corresponding to the components
of a page in the markup inside specific tags and render them via JavaScript code
whenever necessary. The common disadvantage of following a client-side approach
is the impact it has on the initial render time of this page.
Another important advantage of using client-side templating is that it helps to
offload the templating effort from the server to the client. This helps to reduce
the computational load on your server to a great extent, as the templating logic
is executed only in the browser via JavaScript.
Server-side templating
Server-side templating is where we will render the markup via calling the
corresponding views before sending the HTML response back to the web browser.
This is what we are going to explore in this chapter via Handlebars.
Many different rendering engines are available, which can be used with Node and
Express. The most popular among them are Jade, EJS, and Handlebars. The particular
engine we will explore in this book is Handlebars.js.
A view is what we refer to as a HTML page. They are called views because of the
MVC design pattern. As we already discussed, a model is the data that is going to
be displayed on a page, the view is the page itself, and the controller is the brain
that communicates between the model and the view.
Our particular application is going to need two views. The first view is the home
page, and the second view is the image page.
[ 72 ]
Chapter 5
The HTML in the following section relies heavily on Bootstrap,
a popular HTML framework created by Twitter, which provides
a standard set of user interface elements. These include buttons,
fonts, layout grids, color schemes, and a whole lot more. Using
Bootstrap allows us to not only present our application with a
nice clean UI, but also build it so that it is responsive and will
look correct on any device that is viewing it. You can learn more
about Bootstrap by visiting http://getbootstrap.com.
Let's start by creating the home page view. Create a new file within the views folder,
name it index.handlebars, and insert the following HTML code:
<div class="panel panel-primary">
<div class="panel-heading">
<h3 class="panel-title">
Upload an Image
The reason we named our file index.handlebars is a
purely personal choice, but it is also based on common naming
conventions on the Web. Typically, a HTML page that acts
as the root file for any website is named index.whatever
(.php, .aspx, .html, and so on). Again, this is just common
convention and not something you need to specifically adhere to.
Create a basic HTML form and set method to post and action to /images. Be sure
to set the enctype attribute of the form, since we will be uploading files as well as
submitting data via form fields:
<form method="post" action="/images" enctype="multipart/formdata">
<div class="panel-body form-horizontal">
<div class="form-group col-md-12">
<label class="col-sm-2 control-label"
Here, we have included a standard HTML input field for the file to be uploaded:
<div class="col-md-10">
<input class="form-control" type="file"
name="file" id="file">
<div class="form-group col-md-12">
[ 73 ]
Templating with Handlebars
<label class="col-md-2 control-label"
<div class="col-md-10">
Another standard HTML input field for the title of the file can be whatever the user
wants, as shown in the following code:
<input class="form-control" type="text"
<div class="form-group col-md-12">
<label class="col-md-2 control-label"
<div class="col-md-10">
And a standard HTML textarea input field to allow for a description is as follows:
<textarea class="form-control"
name="description" rows="2"></textarea>
<div class="form-group col-md-12">
<div class="col-md-12 text-right">
A standard HTML button is provided that will submit the form to the server.
Using Bootstrap classes, we provide btn and btn-success to make this look like a
Bootstrap-style button with the default color for success (green):
<button type="submit" id="login-btn"
class="btn btn-success" type="button">
<i class="fa fa-cloud-upload ">
</i> Upload Image</button>
After the upload form section, we will display a list of the newest images uploaded
to the website. Please refer to the each block in the code snippet that follows. It is a
keyword supported by Handlebars to loop over the data provided to the template in
case of reusable HTML blocks. We will discuss this in detail in the following code:
<div class="panel panel-default">
<div class="panel-heading">
<h3 class="panel-title">
[ 74 ]
Chapter 5
Newest Images
<div class="panel-body">
{{#each images}}
<div class="col-md-4 text-center" style="padding-bottom:
1em;"><a href="/images/{{ uniqueId }}"><img
src="/public/upload/{{filename}}" alt="{{title}}"
style="width: 175px; height: 175px;" class="imgthumbnail"></a></div>
There are two important sections in the main home page HTML code. The first is
the form we define, which will be the main way users will upload images to the
website. As we will be accepting image files as well as the details of the image
(title, description, and so on), we need to ensure that the form is set up to accept
multipart data. We also set the form action to point to the /images route we
defined earlier in our routes and image controller modules. When a user completes
the form and clicks on the Submit button, the form will send a POST request to
http://localhost:3300/images, and our router will catch that and forward it to
our image controller. From there, the image controller will handle the processing of
the data and will save it to the database, saving the image file to the filesystem, and
redirecting to the image details page. We will actually be writing this logic in the
next chapter. For now, nothing will actually happen if you submit the form.
Below the main image uploading form on the home page, we also have a section that
performs a Handlebars loop using each and iterates through an image collection,
displaying each image as a thumbnail and the link to the image page. The images
collection will be populated from our home controller. It's important to take note here
that when you are inside an {{#each}} loop in a Handlebars template, your context
changes. That is, the path you use to access data inside each is now based on each
item in the collection. Here, our object, which will be bound to the view, will have
an image collection, and each item in the image collection will have a property for
uniqueid, filename, and title. With the home page view out of the way, let's set
up the view for the image page.
Create another file in the views folder and name it image.handlebars. This file is
going to have a bit more functionality, so I'm going to break it down into chunks so
that you can review each section. First, insert the following block of code:
<div class="panel panel-primary">
<div class="panel-heading">
<h2 class="panel-title">{{ image.title }}</h2>
[ 75 ]
Templating with Handlebars
<div class="panel-body">
<p>{{ image.description }}</p>
<div class="col-md-12 text-center">
<img src="/public/upload/{{image.filename}}"
<div class="panel-footer">
<div class="row">
<div class="col-md-8">
<button class="btn btn-success" id="btn-like"
data-id="{{ image.uniqueId }}"><i class="fa faheart">
<strong class="likes-count">{{ image.likes }}
</strong> &nbsp; - &nbsp; <i class="fa fa-eye">
</i> <strong>{{ image.views }}</strong>
&nbsp; - &nbsp; Posted: <em class="text-muted">{{
timeago image.timestamp }}</em>
This block of code defines the bulk of the content that will be displayed on the page
for a specific image. The viewModel for this page is going to consist of an image
object that has various properties defined that you see being used throughout the
code; properties such as title, description, filename, likes, views, and the timestamp
of when the image upload was performed.
You may have noticed a slightly different piece of syntax in the code specific to the
{{ timeago image.timestamp }} timestamp. That is actually a Handlebars helper.
It is a custom function we will write shortly, which will perform some special string
formatting—specifically, converting a timestamp string to how long it was sometime
ago (that is, 2 days ago, 12 hours ago, 15 minutes ago, and so on).
We want to allow users to post comments to images, so let's include a simple form
for that:
<div class="panel panel-default">
<div class="panel-heading">
<div class="row">
<div class="col-md-8">
<strong class="panel-title">Comments</strong>
[ 76 ]
Chapter 5
<div class="col-md-4 text-right">
<button class="btn btn-default btn-sm" id="btncomment" data-id="{{ image.uniqueId }}">
<i class="fa fa-comments-o"> Post
<div class="panel-body">
<blockquote id="post-comment">
<div class="row">
The following is another standard HTML form with the method and action set. This
form allows a user to enter, via standard HTML input fields, his/her name, e-mail
address, and comments. Another Submit button is provided to save the comment:
<form method="post" action="/images/{{
image.uniqueId }}/comment">
<div class="form-group col-sm-12">
<label class="col-sm-2 control-label"
<div class="col-sm-10">
<input class="form-control"
type="text" name="name">
<div class="form-group col-sm-12">
<label class="col-sm-2 control-label"
<div class="col-sm-10">
<input class="form-control"
type="text" name="email">
<div class="form-group col-sm-12">
<label class="col-sm-2 control-label"
<div class="col-sm-10">
<textarea class="form-control"
name="comment" rows="2"></textarea>
<div class="form-group col-sm-12">
<div class="col-sm-12 text-right">
[ 77 ]
Templating with Handlebars
<button type="submit" id="comment-btn"
class="btn btn-success" type="button">
<i class="fa fa-comment"></i>
The form action for comments is set to /images/{{ image.uniqueid }}/comment.
Again, if you recall from the routes we set up, we specifically defined a route to
handle this.
Finally, we want to display any comments that have been submitted for this image.
Our viewModel includes a collection of comments in addition to the image details, so
we can simply iterate over that collection by using the Handlebars #each block helper:
<ul class="media-list">
{{#each comments}}
<li class="media">
<a class="pull-left" href="#">
<img class="media-object img-circle"
<div class="media-body">
{{ comment }}
<br/><strong class="media-heading">{{ name
}}</strong> <small class="text-muted">{{
timeago timestamp }}</small>
Much like the loop we performed on the home page to display a collection of images,
here we simply iterate through every comment in the comments collection and
display the comment and string-formatted timestamp (again using our timeago
global helper). We also use Gravatar to display universal avatar images for users
who have commented (assuming they have provided their e-mail addresses).
[ 78 ]
Chapter 5
Gravatar is a service provided by https://wordpress.com/
that allows a user's personal profile image to be provided via
his/her e-mail address. Many popular web services rely on
Gravatar as a quick and easy way to display a user's personal
profile image, without requiring the additional functionality to
support such a feature. You can learn more about Gravatar at
So far we've created two specific views for our website, one for the home page and
one for the details of an image. However, there's no consistent UI wrapping both of
these pages together. We have no consistent navigation or logo. There's no common
footer with standard copyright or additional information.
Usually, with any website that you create, you're going to want to have some form of
a standard layout or master template that every page will use. This layout typically
includes the website logo and title, main navigation, sidebar (if any), and the footer.
It would be bad practice to include the HTML code for the layout in every single
page on the website because if you wanted to make even the smallest change to
the main layout, you would have to edit every single page as a result. Fortunately,
Handlebars helps lessen the work of utilizing a layout file.
Let's create a layout file for our app by creating a new file named main.handlebars
within the views/layouts folder and inserting the following HTML code:
<html lang="en">
bootstrap.min.css" rel="stylesheet">
<link href="//netdna.bootstrapcdn.com/font-awesome/
4.0.3/css/font-awesome.min.css" rel="stylesheet">
<link rel="stylesheet" type="text/css"
<div class="page-header">
<div class="container">
<div class="col-md-6">
<h1><a href="/">imgPloadr.io</a></h1>
[ 79 ]
Templating with Handlebars
<div class="container">
<div class="row">
<div class="col-sm-8">
{{{ body }}}
<div class="col-sm-4">
{{> stats this }}
{{> popular this }}
{{> comments this }}
<div style="border-top: solid 1px #eee; padding-top:
<div class="container">
<div class="row">
<div class="col-sm-12 text-center">
<p class="text-muted">imgPloadr.io |
&copy; Copyright 2014, All Rights
<p class="text-center">
<i class="fa fa-twitter-square fa-2x
<i class="fa fa-facebook-square fa-2x
<script src="//ajax.googleapis.com/ajax/libs/jquery/
<script type="text/javascript"
Most of the preceding code is just HTML, and a lot of it uses Bootstrap for the actual
physical layout of the page as well as a few other UI-related elements. The most
important part is the highlighted section in the middle with {{{ body }}} and the
few lines below that, as they pertain to the use of Handlebars.
[ 80 ]
Chapter 5
{{{ body }}} is a reserved word in Handlebars that is used specifically with
layouts. What we are basically saying is that any page we render that uses our
default layout file will have its content inserted into the area where {{{ body }}} is
defined. If you recall from the configure module we created earlier, we defined our
default layout file when we set up Handlebars as our rendering engine. The slightly
odd use of {{{ and }}} around the body is due to the fact that Handlebars escapes
HTML by default when using {{ and }}. Since our views contain mostly HTML, we
want this to stay intact so that we use {{{ and }}} instead.
The other three lines that use {{ > ... }} render Handlebars partials, which are
like the shared HTML code blocks that we will cover next.
Partial views
So far we've created a view, which acts as the bulk of the HTML for a specific page,
and a layout, which acts as the wrapper for the consistent parts of the website on
every page. Next, let's take a look at creating partials, which are really just small
views that we can reuse and inject inside our layouts or views.
Partials are a terrific way to create reusable components in a website and reduce code
duplication. Consider the comments in our application. We have an HTML form
defined that a user uses to submit a comment, but what if we wanted to allow users to
post comments from a number of different areas throughout the website? This type of
scenario is a great candidate for moving our comment form out to its own partial and
then just including that partial anywhere we want to display the comment form.
For this app, we're using partials specifically for the sidebar in the main layout.
With every view's viewModel, we will include a JavaScript object called sidebar
that will contain the data specifically for the stats, popular images, and recent
comments found within the sidebar partial.
Let's create the HTML for each of the partials. First, create a file named
stats.handlebars within the views/partials/ path and include the
following HTML code:
<div class="panel panel-default">
<div class="panel-heading">
<h3 class="panel-title">
<div class="panel-body">
<div class="row">
<div class="col-md-2 text-left">Images:</div>
[ 81 ]
Templating with Handlebars
<div class="col-md-10 text-right">{{
sidebar.stats.images }}</div>
<div class="row">
<div class="col-md-2 text-left">Comments:</div>
<div class="col-md-10 text-right">{{
sidebar.stats.comments }}</div>
<div class="row">
<div class="col-md-2 text-left">Views:</div>
<div class="col-md-10 text-right">{{
sidebar.stats.views }}</div>
<div class="row">
<div class="col-md-2 text-left">Likes:</div>
<div class="col-md-10 text-right">{{
sidebar.stats.likes }}</div>
Next, create views/partials/popular.handlebars and insert the following HTML
code into it:
<div class="panel panel-default">
<div class="panel-heading">
<h3 class="panel-title">
Most Popular
<div class="panel-body">
{{#each sidebar.popular}}
<div class="col-md-4 text-center" style="paddingbottom: .5em;">
<a href="/images/{{uniqueId}}"><img
style="width: 75px; height: 75px;"
[ 82 ]
Chapter 5
Finally, create views/partials/comments.handlebars and insert the following
HTML code into it:
<div class="panel panel-default">
<div class="panel-heading">
<h3 class="panel-title">
Latest Comments
<div class="panel-body">
<ul class="media-list">
{{#each sidebar.comments}}
<li class="media">
<a class="pull-left" href="/images/{{
image.uniqueId }}">
<img class="media-object" width="45"
height="45" src="/public/upload/
{{ image.filename }}">
<div class="media-body">
<strong class="media-heading">
{{name}}</strong> <small class="text-muted">
{{timeago timestamp }}</small>
The basics of Handlebars
Handlebars is a really simple and easy-to-use templating framework. Let's go
through and explore the basic syntax of writing a Handlebars template.
Binding an object to the template
Let's assume the following JavaScript object is passed to a Handlebars template:
var model = {
name: 'World'
[ 83 ]
Templating with Handlebars
The template file itself will contain the following markup:
Hello {{ name }}!
This file will render to a browser in the following way:
Hello World!
Embedding presentation logic
Of course, there's a lot more that you can do than just this! Handlebars also supports
conditional statements:
var model = {
name: 'World',
description: 'This will only appear because its set.'
Hello {{ name }}!<br/><br/>
{{#if description}}
Using an if block helper, as shown in the preceding code, you can check for
true conditionals and only display HTML and/or data if the condition is true.
Alternatively, you can use the unless helper to do the opposite, and display
HTML only if a condition is false:
var model = {
name: 'World'
Hello {{ name }}!<br/><br/>
{{#unless description}}
<p>NOTE: no description found.</p>
You can use both if and else as well as unless the same way you would use
conditional if/else in other programming languages.
[ 84 ]
Chapter 5
Handlebars helpers
Handlebars supports the idea of helpers, which are special custom functions you
can write to perform some special logic from within the template during runtime.
This will encourage developers to migrate the common presentation logic present
in views into helpers and reuse them, thereby adding much more readability to the
views. A great example of a helper would be the date string formatter we've been
using. Helpers can be registered globally and made available to every template file,
or they can be defined per view and passed to the template on an as-needed basis
as part of the viewModel.
Global helpers
First, let's create a global helper that will be available to every Handlebars template
we render. The global helper that we will create will be used to format a timestamp
so that it is worded according to how long ago the event occurred. We will use this
throughout our application for things such as comments and image timestamps.
The first thing we need to do is update our server/configure.js module, where
we originally configured Handlebars as our rendering engine. We are going to add
a section to define our helpers:
app.engine('handlebars', exphbs.create({
defaultLayout: 'main',
layoutsDir: app.get('views') + '/layouts',
partialsDir: [app.get('views') + '/partials'],
helpers: {
timeago: function(timestamp) {
return moment(timestamp).startOf('minute').fromNow();
[ 85 ]
Templating with Handlebars
As you can see from the additional code we added (highlighted in the preceding
code), we defined the helpers property of the configuration options within
create(). Inside the helpers property, we can define any number of functions we
want. In this case, we defined a simple timeago function that actually uses another
npm module called moment. The moment module is a great library for performing
numerous different types of date string formatting. As we are using a new module,
we need to be sure to perform require()at the top of our configure module:
var path = require('path'),
routes = require('./routes'),
exphbs = require('express3-handlebars'),
bodyParser = require('body-parser'),
cookieParser = require('cookie-parser'),
morgan = require('morgan'),
methodOverride = require('method-override'),
errorHandler = require('errorhandler'),
moment = require('moment');
As well as actually install it via npm:
$ npm install moment --save
View-specific helpers
While defining helpers globally is nice because they are available to every view that's
rendered, sometimes you might only need to define a helper for use within a single
view. In this case, you can include the helper with the model itself when calling
res.render(), as shown in the following code:
var viewModel = {
name: 'Jason',
helpers: {
timeago: function(timestamp) {
return 'a long time ago!';
res.render('index', viewModel);
Not only are we defining a custom helper that can be used specifically from this view
in its model object, but in this particular instance, we are overriding the existing
timeago global helper with a slightly different version that is perfectly valid.
[ 86 ]
Chapter 5
Rendering the views
Let's take a minute to do a quick recap and see what we've done up to this point. So
far, we have:
• Created index.handlebars and image.handlebars—the views for the two
main pages of the application
• Created layouts/main.handelbars—the main layout file for every page in
the application
• Created partials/comments.handlebars, popular.handlebars, and
• Created a global timeago Handlebars helper
So far, so good; however, none of these views actually do anything, receive any
viewModels, or even appear when you run the application! Let's make a few quick
minor modifications to our controllers to get our views to render properly.
Open /controllers/home.js so that you can edit the home controller module.
Update the contents of that file so that it looks identical to the following block of code:
module.exports = {
index: function(req, res) {
Instead of performing res.send, which just sends a simple response, we are calling
res.render and passing in the name of the template file we want to render as the
only parameter (for now). Using the defaults that were defined in our configure
module, the index file will be loaded from our views folder. Again, also using the
defaults, we configured the default layout of main that will be applied to this view
in our configure module.
Let's update the image controller to do the same thing as well. Edit /controllers/
image.js and change the index function so that it looks identical to the following
block of code:
index: function(req, res) {
[ 87 ]
Templating with Handlebars
And that's it! Let's fire up the server, open the app in our browser, and see how
it looks:
$ npm start
$ open http://localhost:3300
$ open http://localhost:3300/images/1
Success! Hopefully, you see something that closely resembles the following
screenshot of the home page:
[ 88 ]
Chapter 5
Additionally, if you provide a random URL to a specific image, for example
http://localhost:3300/images/1, you should see the following screenshot:
The preceding command uses npm start to start the
application. Please note that this command will work only
if you have the application entry point file configured in the
package.json file. If this is not working, then you have to
set the main attribute in the package.json and set it to the
server.js file. Also, as an alternative, you can manually
invoke the server.js file by using node server.js.
[ 89 ]
Templating with Handlebars
In this chapter, the Handlebars template-rendering engine was introduced,
following which we reviewed the syntax used when creating dynamic HTML
pages. We created a main layout for our app as well as the home page and image
page views. We included partial views for the sidebar in the layout and created a
global Handlebars helper to display custom formatted dates.
Even though none of the views are currently displaying any data (because we aren't
passing models to the views yet), you can see that things are starting to come along
nicely! In the next chapter, we will wire up the actual logic in the controllers for each
page and build up the Model objects so that we start seeing some actual content on
our screens.
[ 90 ]
Controllers and View Models
Up until this point, the controllers we wrote for our application have been extremely
basic. They were started with a simple task of sending text responses to the client. In
the previous chapter, we updated the controllers so that they render an HTML view
and send the HTML code to the client, instead of a simple text response. The primary
job of a controller is to act as the entity, which holds the logic that makes all of the
necessary decisions to properly render a response to the client. In our case, this means
retrieving and/or generating the data necessary for a page to appear complete.
In this chapter, we will:
• Modify the controllers so that they generate a data model and pass it to
a view
• Include logic to support uploading and saving image files
• Update the controllers to actually render dynamic HTML
• Include helpers for the partials that generate the website statistics
• Iterate on the UI to include improved usability via jQuery
A controller can be defined as the entity that will be responsible for manipulating
models and initiating the view render process with the data received from the
corresponding models. In the code we have developed so far, we can see that the
express router instance is used to tie functions to a corresponding route. These
functions are nothing but controllers. For every route that we create in our router,
two parameters are necessary:
• The first parameter is the string for the route itself, which is
[ 91 ]
Controllers and View Models
• The second parameter is a controller function that will be executed when that
route is accessed
For any route that has to do with images, we rely on the images controller. Likewise,
any route that has to do with the home page relies on the home controller, and so on
and so forth.
The steps we will take to define our controllers in our app are purely organizational
and based on personal preference. We created our controllers as modules so that our
router wasn't a big, long convoluted mess of spaghetti code. We could have just as
easily kept all of the logic contained in our controllers as functions directly within the
routes themselves, but this would have been an organizational mess and made for
very hard-to-read code to maintain later.
As our sample app is fairly small, we only have two controllers currently: home and
image. It is the responsibility of these controllers to build the appropriate view models
for our HTML pages and render the actual pages as well. Any logic that is required to
execute per page and build the view model will be done so via our controllers.
View models
Given a single HTML view in our app, we need to be able to attach data to that
page so that the template that is being rendered can be included in such a way
that the dynamic areas of the page are replaced with real content. To do this, we
need to generate a view model. During the render, the template engine will parse
the template itself and look for special syntax that indicates that specific sections
should be replaced at runtime with values from the view model itself. We have
seen examples of this while we explored the Handlebars template framework in
the previous chapter. Think of this as a fancy runtime find and replace of your
HTML templates—finding variables and replacing them with values stored in the
view model sent to the template. It is important to note that this process happens at
the server and the result is only sent as the response to the HTTP request that our
application receives.
A view model is typically just a single JavaScript object that can be passed to the
template. The template contains all of the necessary logic we need to properly render
the page. It is the task of the templating engine to produce the corresponding HTML
by processing the templates with the associated models. The view model for a page
will typically contain all of the data necessary to render the content-specific portions of
that page. Using our application as an example, the view model for a specific image's
page might contain the title of the image, its description, and the information necessary
to display the image and various stats such as the number of likes, views, and a
collection of comments. A view model can be as simple or as complex as you like.
[ 92 ]
Chapter 6
The term view model is used here to refer to the data form of model,
which will be tied to the template while rendering the HTML via any
templating framework.
Updating the home controller
If you take a look at our current home controller (controllers/home.js), you can
see that the index function barely has any code in it whatsoever:
The first thing we want to do is build a basic view model using sample data so that
we can see our view model at work. Replace that single res.render call with the
following updated code:
var viewModel = {
images: [
}, {
}, {
}, {
'Sample Image 1',
'Sample Image 2',
'Sample Image 3',
[ 93 ]
Controllers and View Models
'Sample Image 4',
res.render('index', viewModel);
In this code, we built a basic JavaScript collection of objects:
• The variable we declare is called viewModel, but the name of this variable
doesn't actually matter and can be whatever you want. The viewModel
variable is an object that contains a single property called images, which
is itself an array.
• The images array contains four sample images, each with a few basic
properties— we came up with the most obvious properties while deciding
what kind of information we want per image. Each image in the collection
has a uniqueId, title, description, filename, views, and likes count,
and a timestamp property.
Once we have set up our viewModel, we simply pass it as the second parameter
to the res.render call. Doing this while rendering a view makes the data in it
available to the view itself. Now, if you recall from some of the template code we
wrote for the home index.handlebars view, we had a {{#each images}} loop that
iterated through each image in the images collection of the view model passed to the
template. Taking another look at our view model we created, we see that it only has
a single property named images. The HTML code inside the Handlebars loop will
then specifically reference the uniqueId, filename, and title properties for each
image in the images array.
[ 94 ]
Chapter 6
Save the changes to the home controller, launch your app again, and point your
browser to http://localhost:3300. You should see the four images that now
appear on the homepage in the Newest Images section (although, as you can see
in the following screenshot, the images are still broken, as we didn't actually create
any image files):
The homepage has a fairly simple controller and view model, and you might have
noticed that the sidebar is still completely empty. We'll cover the sidebar a little later
in this chapter.
[ 95 ]
Controllers and View Models
Updating the image controller
Let's create the controller and view model for the image page. The controller for the
image will be a little more complex, as we'll write the logic to handle uploading and
saving of the image files via the form on the homepage.
Displaying an image
The index function in the image controller will look almost identical to the index
function from the home controller. The only difference is that instead of generating
an array of images, we will build a view model for a single image. However, the
view model for this image will have a little more information than the one from
the homepage, since we are building a page that renders a more detailed view of
an image (versus the thumbnail collection on the homepage). The most noteworthy
inclusion is that of a comments array for the image.
Taking another look at the original index function in our controllers/image.js
file, we can see the simple existing res.render line of code:
We want to replace this line with a view model and an updated res.render
statement using the following code:
var viewModel = {
image: {
comments: [
'Sample Image 1',
'This is a sample.',
'Test Tester',
[ 96 ]
Chapter 6
'This is a test comment...',
'Test Tester',
'Another followup comment!',
res.render('image', viewModel);
Here we declare a new viewModel variable again, this time with an image property
that contains the properties for the single image. In addition to the image property,
there is also a comments property, which is an array of comment objects. You can
see that each comment has various properties specific to a comment for each image.
This JavaScript object is actually a pretty good preview of what our real data will
wind up looking like once we include logic to connect our app to MongoDB!
After we build our sample image object and its collection of comments, we pass
that along to our res.render call, thus sending this new viewModel directly to
our image's Handlebars template. Again, if you review the HTML code in the
image.handlebars file, you can see where each property of the viewModel is
being displayed.
Again, let's run the application and make sure our image page is appearing properly:
$ node server.js
[ 97 ]
Controllers and View Models
Once the app is running and you've launched it in your browser, click on any of the
images that are listed in the Newest Images section of the homepage. This should
take you to an individual image page where you will see something like the page
shown in the following screenshot:
Notice that title, description, likes and views, and timestamp are all now
appearing on the page. In addition, you can see a few comments listed below the
image as well!
Uploading an image
The next feature we need to implement in our image controller is the logic to handle
when a user submits an image via the Image Upload form on the homepage. Even
though the form is on the homepage of our app, we decided to house the logic to
handle uploading within our image controller because logically, this makes the most
sense (since this feature has primarily to do with images and not the homepage per se).
This was purely a personal decision, and you can house the logic wherever you please.
[ 98 ]
Chapter 6
You should note that the HTML for the form on the homepage has its action set
to /images and its method is post. This perfectly matches the route we set up
previously, where we listen for a post to the /images route and call the image
controller's create function.
The create function in our image controller will have a few key responsibilities:
• It should generate a unique filename for the image, which will also act as
an identifier
• It should save the uploaded file to the filesystem and ensure that it is an
image file
• It should redirect to the image/image_id route once its task is complete to
display the actual image
As we are going to be working with the filesystem in this function, we are going to
need to include a few modules from the Node.js core set of modules, specifically the
filesystem (fs) and the path (path) modules.
Before we begin with adding the necessary code for the upload image section, we
need a small fix to be applied on the configuration of the application. If you look at
the configure.js file, which we created in the beginning, you should see the lines
specific to configuring the upload directory using the bodyParser module:
uploadDir:path.join(__dirname, 'public/upload/temp')
Here, the issue is that bodyParser supports only JSON and URL encoded form
submissions and not multipart form submissions that we need in the case of our
file upload. So we need to add an extra module in the configure file to support the
file uploads, namely multer. Add it as a dependency to our application using the
following command:
npm install multer -–save
Now, go to the configure file and require it via
multer = require('multer');
You can place this below the initial require calls in the file. Then, find the following
bodyParser configuration line:
uploadDir:path.join(__dirname, 'public/upload/temp')
[ 99 ]
Controllers and View Models
Replace the preceding line with the following line of code:
app.use(multer({ dest: path.join(__dirname,
Now, our upload actions will work fine, as expected.
Let's begin by first editing the controllers/image.js file and inserting the two new
require statements at the very top of the file:
var fs = require('fs'),
path = require('path');
Next, take the create function's original code:
res.send('The image:create POST controller');
Replace this original code with the following code:
var saveImage = function() {
// to do...
Here, we created a function called saveImage, and we executed it immediately after
we declared it. This might look a little odd, but the reason for this will become clear
when we implement database calls in the following chapter. The main reason is that
we are going to call saveImage repeatedly to ensure that the unique identifier we
generated is in fact unique and doesn't already exist in the database (as a previously
saved image's identifier).
Let's review a breakdown of the code that will be inserted in the saveImage function
(replacing the // to do... comment). I will cover each line of code for this function
and then give you the entire block of code at the end:
var possible = 'abcdefghijklmnopqrstuvwxyz0123456789',
imgUrl = '';
We need to generate a random six-digit alphanumeric string to represent a unique
identifier for an image. This identifier will work similar to other websites that
provide tiny URLs for unique links (that is, bit.ly). To do this, we first provide a
string of possible characters that can be used while generating the random string:
for(var i=0; i < 6; i+=1) {
imgUrl += possible.charAt(Math.floor(Math.random() *
[ 100 ]
Chapter 6
Then, loop six times and randomly pull out a single character from our string
of possible characters, appending it in each cycle. By the end of this for loop,
we should have a string that consists of six random letters and/or numbers,
for example a8bd73:
var tempPath = req.files.file.path,
ext = path.extname(req.files.file.name).toLowerCase(),
targetPath = path.resolve('./public/upload/' + imgUrl + ext);
Here, we declare three variables; where our uploaded files will be stored temporarily,
the file extension of the file that was uploaded (that is .png, .jpg, and so on), and a
destination where the uploaded image should ultimately reside. For both the latter
variables, we use the Path node module, which works great while dealing with
file names and paths and getting information from a file (such as a file extension).
Next, we move the image from its temporary upload path to its final destination:
if (ext === '.png' || ext === '.jpg' || ext === '.jpeg' || ext ===
'.gif') {
fs.rename(tempPath, targetPath, function(err) {
if (err) throw err;
res.redirect('/images/'+ imgUrl);
} else {
fs.unlink(tempPath, function () {
if (err) throw err;
res.json(500, {error: 'Only image files are allowed.'});
This code performs some validation. Specifically, it conducts checks to make sure that
the uploaded file extension matches a list of allowable extensions—namely, known
image file types. If a valid image file was uploaded, it is moved from the temp folder
via the filesystem's rename function. Notice how the filesystem (fs) rename function
takes three parameters: the original file, the new file, and a callback function.
The callback function is executed once the rename is complete. If the node didn't
work this way (always relying on callback functions), it's quite likely your code will
execute immediately following the execution of the rename function and try to work
against a file that doesn't exist yet (that is, the rename function didn't even finish
doing its work). By using a callback function, we are effectively telling the node that
once the rename of the file is finished and the file is ready and where it should be,
then execute the following code.
[ 101 ]
Controllers and View Models
The else condition that follows handles the situation when the uploaded file was
invalid (that is, not an image), so we call the unlink function of the fs module,
which will delete the original file (from the temp directory it was uploaded to)
and then send a simple JSON 500 with an error message.
Here is the complete saveImage function (again, the following code will replace
// to do... from earlier):
var possible = 'abcdefghijklmnopqrstuvwxyz0123456789',
imgUrl = '';
for(var i=0; i < 6; i+=1) {
imgUrl += possible.charAt(Math.floor(Math.random() *
var tempPath = req.files.file.path,
ext = path.extname(req.files.file.name).toLowerCase(),
targetPath = path.resolve('./public/upload/' + imgUrl + ext);
if (ext === '.png' || ext === '.jpg' || ext === '.jpeg' || ext ===
'.gif') {
fs.rename(tempPath, targetPath, function(err) {
if (err) throw err;
res.redirect('/images/' + imgUrl);
} else {
fs.unlink(tempPath, function () {
if (err) throw err;
res.json(500, {error: 'Only image files are allowed.'});
With this code in place, we can now successfully upload an image file via the form
on the homepage. Give it a try by launching the app and opening it in a browser.
Once there, click on the Browse button in the main form, and select an image file
from your computer. If successful, the image file should exist within the public/
upload folder of your project with a new random filename.
[ 102 ]
Chapter 6
Ensure that you have the public/upload folders created in your
project, or you will get runtime errors when you attempt to write
files to a location that doesn't exist. Write permissions might need
to be set on the folder depending on your OS and security access.
After the upload form completes and the create controller function does its work,
it will redirect to the individual image page for the image that was uploaded.
Helpers for reusable code
So far, each of the pages that we have rendered display their viewModel data
perfectly, but that pesky sidebar still remains blank. We're going to fix this by
creating a few modules for the sidebar content by implementing them as helper
modules. These helper modules will be used repeatedly by various parts of our
application and don't necessarily belong to the controller folder or the server
folder. So, we'll just create a new home called helpers and store these modules
As we are just loading temporary fixture data into our
view models, the data we set up in the helpers as well as
the controllers will all be replaced with actual live data
in the next chapter once we implement MongoDB.
The sidebar module
First, we will create a module for the entire sidebar. This module will be responsible
for calling multiple other modules to populate viewModel for each section of the
sidebar. As we are going to be populating each page's own viewModel with data
specifically for the sidebar, the sidebar module's function will accept that original
viewModel as a parameter. This is so that we can append data to the existing
viewModel for each page.
Here, we will be appending a sidebar property (which is a JavaScript object) that
contains properties for each of the sections of the sidebar.
To get started, first create a file named helpers/sidebar.js and insert the
following code:
var Stats = require('./stats'),
Images = require('./images'),
Comments = require('./comments');
[ 103 ]
Controllers and View Models
module.exports = function(viewModel, callback){
viewModel.sidebar = {
stats: Stats(),
popular: Images.popular(),
comments: Comments.newest()
In the preceding code, you can see that we first required a module for each section
of the sidebar. The existing viewModel for any given page that displays the sidebar
is the first parameter of the function. We add a sidebar property to viewModel and
set values for each property by calling the module for each section of the sidebar.
Finally, we execute a callback that was passed in as the second parameter to the
sidebar module. This callback is an anonymous function that we will use to execute
the rendering of the HTML page.
Let's update the home and image controllers to include a call to the sidebar module
as well as defer rendering the HTML template for each page to the callback for the
sidebar module.
Edit controllers/home.js and take the following line of code:
res.render('index', viewModel);
Replace it with this new block of code:
sidebar(viewModel, function(viewModel) {
res.render('index', viewModel);
Make the exact same changes to the controllers/image.js file replacing index
with image:
res.render('image', viewModel);
Again, notice how we execute the sidebar module and pass the existing viewModel
as the first parameter and a basic anonymous function as a callback for the second
parameter. What this is doing is waiting to render the HTML for the view until after
the sidebar has completed populating viewModel. This is because of the asynchronous
nature of Node.js. Suppose we wrote the code in the following way instead:
res.render('index', viewModel);
[ 104 ]
Chapter 6
Here, it's quite likely that the res.render statement will execute before sidebar
has even finished doing any work. This is going to become very important once
we introduce MongoDB in the next chapter.
Additionally, as we are now using the sidebar module in each controller, be sure
to require it at the top of both controllers by including the following code:
var sidebar = require('../helpers/sidebar');
Now that our sidebar module is complete, and it's being called from both controllers,
let's complete the sidebar by creating each of the submodules that are required.
The stats module
The stats module is going to display a few random pieces of statistics about our
app. Specifically, it will show the count for the total number of images, comments,
views, and likes for the entire website.
Create the helpers/stats.js file and insert the following code:
module.exports = function() {
var stats = {
return stats;
This module is pretty basic and all it does is create a standard JavaScript object with
a few properties for the various stats, each set initially to 0.
The images module
The images module is responsible for returning various collections of images.
Initially, we will create a popular function that will be used to return a collection
of the most popular images on the website. Initially, this collection will simply be
an array of image objects with the sample fixture data present.
[ 105 ]
Controllers and View Models
Create the helpers/images.js file and insert the following code:
module.exports = {
popular: function() {
var images = [
}, {
}, {
}, {
return images;
'Sample Image 1',
'Sample Image 2',
'Sample Image 3',
'Sample Image 4',
[ 106 ]
Chapter 6
The comments module
Similar to the image's helper module, the comments module will return a collection
of the newest comments posted to the site. The idea of particular interest is that each
comment also has an image attached to it so that the actual image for each comment
can be displayed as a thumbnail while displaying the list of comments (otherwise,
we lose context when we see a random list of comments with no related image).
Create the helpers/comments.js file and insert the following code:
module.exports = {
newest: function() {
var comments = [
'Test Tester',
gravatar: 'http://lorempixel.com/75/75/animals/1',
'This is a test comment...',
timestamp: Date.now(),
image: {
'Sample Image 1',
}, {
Test Tester',
gravatar: 'http://lorempixel.com/75/75/animals/2',
'Another followup comment!',
timestamp: Date.now(),
image: {
'Sample Image 1',
[ 107 ]
Controllers and View Models
return comments;
Again, this is just a basic JavaScript array of objects with a few properties for
each comment, one of which is an actual image and its properties (the image
property should look familiar since it's the same as one of the items in the
images helper module).
Testing the sidebar implementation
Now that our sidebar module is complete along with its dependent submodules
for the various stats, images, and comments, it's time to give our application another
test run. Launch the node server and open the application in your browser.
You should now see the sidebar complete with content on both the homepage as well
as the image landing page.
Iterating on the UI
Now that we have our application working fairly well and we can actually interact
with it, it's time to step back and take a look at some of the areas we might be able
to improve.
One area is the Post Comment form on the image page. I don't think it's necessary
that this form is always visible, but instead, it should be made available only when
someone actually wants to post a comment.
Additionally, I'd love the Like button to not have to post a full form submission to
the server and cause the entire page to reload (like the form on the homepage does
when it uploads an image). We will use jQuery to submit an AJAX call to the server
to handle the likes, and send and retrieve data in real time without ever reloading
the page.
To make these tweaks, we're going to need to introduce a small amount of JavaScript
on the page to add a little interactivity. To make things even easier, we'll use the
popular jQuery JavaScript library to make creating interactive features a breeze.
[ 108 ]
Chapter 6
jQuery has been around for a number of years and has been
explosively popular in frontend development. Simply put, jQuery
in JavaScript with training wheels. It allows you to manipulate the
Document Object Model (DOM)—the HTML structure of any
page—extremely easily; you will see this in the next section.
You can learn more about jQuery at http://jquery.com.
You might not have noticed, but, in the HTML code that was provided for the main.
handlebars layout file, jQuery has already been included as an external script tag
(referencing jQuery hosted on a CDN). Additionally, a local scripts.js tag is also
included, where we will put our custom jQuery JavaScript code for the changes we're
going to make to the UI. When you look at the very bottom of main.handlebars,
you can see the following code:
<script type="text/javascript"
The first script tag points to Google's code Content Delivery Network (CDN), which
means, we don't have to worry about hosting that file with our code. The second file,
however, is our own file, so we will need to make sure that exists.
CDN is a means of delivering a file from a globally distributed network
of caching servers. What this means, generally speaking, is that, files
that are very commonly downloaded by web visitors (such as jQuery,
for example) can be loaded much quicker via a regionally closer
download source as well as improved caching. If multiple websites use
the same CDN URL to host jQuery, for example, it stands to reason that
a visitor to your website might have already downloaded jQuery from
visiting a previous unrelated website. Therefore, your website will load
that much quicker!
Create the public/js/scripts.js file and insert the following code:
// to do...
[ 109 ]
Controllers and View Models
This is a standard code block that you'll see almost every time anyone uses jQuery.
What this code does is execute an anonymous function within the $() jQuery
wrapper, which is shorthand for writing the following code:
// to do...
This code basically just means that the callback function will wait until the page is
fully loaded and ready before executing. This is important because we don't want
to apply UI event handlers and/or effects to DOM elements that don't actually exist
yet because the page is still loading. This is also another reason why the script tags
in the main.handlebars layout file are the last lines of the page; so, they are the last
to load ensuring that the document has already been fully downloaded and is ready
to be manipulated.
First, let's address the Post Comment functionality. We want to hide the comment
form by default, and then display it only when a user clicks on the Post Comment
button below an image (to the right of the Like button). Insert the following code
inside the callback function where the // to do... comment exists:
$('#btn-comment').on('click', function(event) {
The first line of code executes the hide function on the HTML div that has a postcomment ID. We then immediately apply an event handler to the HTML button with
a btn-comment ID. The event handler we apply is for onClick because we want it to
execute the anonymous function we provided whenever a user clicks on that button.
That function simply prevents the default behavior (the default behavior for that
particular element; in this case, a button) and then calls the show jQuery function,
which reveals the post-comment div that was previously hidden. The event.
preventDefault() part is important because if we didn't include that, the action of
clicking on the button would do what a browser expects it to do and try to execute
our custom JavaScript function at the same time. If we didn't include this, it's likely
that our UI will behave in ways that are less than ideal. A good example is if you
want to override the default behavior of a standard HTML link, you can assign an
onClick event handler and do whatever you want. However, if you don't perform
event.preventDefault(), the browser is going to send the user to the HREF for
that link, regardless of what your code is trying to do.
[ 110 ]
Chapter 6
Now let's add some code to handle the Like button functionality. We are going to
add an event handler for the button, the same way we did for the Post Comment
button, using jQuery's .on function. After the code that you added previously,
insert this additional block of code inside the ready statement:
$('#btn-like').on('click', function(event) {
var imgId = $(this).data('id');
$.post('/images/' + imgId + '/like').done(function(data) {
The preceding code attaches an onClick event handler to the btn-like button.
The event handler first retrieves the data-id attribute from the Like button itself
(assigned via the image.handlebars HTML template code and viewModel) and
then performs a jQuery AJAX post to the /images/:image_id/like route. Recall
the following line from our Node server/routes.js file:
app.post('/images/:image_id/like', image.like);
Once that AJAX call is done, another anonymous callback function will be executed
that will change the text of the HTML element with a likes-count class and replace
it with the data that was returned from the AJAX call—in this case, the updated total
count of likes (typically, it would be whatever it was previously plus one).
In order to test this functionality, we are going to need to implement some fixture
data in our like function inside the image controller. Edit controllers/image.js
and within the like function, replace the existing res.send function call with the
following code:
like: function(req, res) {
res.json({likes: 1});
All this code does is return JSON to the client with a simple object that contains a
single likes property with a value of 1. In the next chapter, when we introduce
MongoDB to the app, we'll update this code to actually increase the count of likes
and return the true value for the liked image.
[ 111 ]
Controllers and View Models
With all of those changes in place, you should be able to relaunch the node server
and open the website in your browser. Click on any image on the homepage to view
the image page and then click on the Like button to see it change from 0 to 1. Don't
forget to check out the fancy new Post Comment button too—clicking on this should
reveal the comment form!
At the beginning of this chapter, we had some basic HTML pages that appear in a
browser via our application, but they contained no content and no logic whatsoever.
We implemented the logic for each of our controllers and learned about the view
model and how to populate pages with content.
In addition to displaying content on our pages via a view model, we also implemented
the code to handle uploading and saving image files to the local filesystem.
We tweaked the UI slightly to include some subtle enhancements using jQuery by
revealing the comment form and used AJAX to track likes instead of relying on a
full page postback.
Now that the groundwork has been laid for our view models and controllers, let's tie
it all together using MongoDB and start working with real data. In the next chapter,
we will update the controllers once again, this time implementing the logic to read
from and save data to our MongoDB server.
[ 112 ]
Persisting Data with
With almost any application written for the Web nowadays, a highly interactive
application is of limited value if the interaction between its users isn't permanently
saved. You have to integrate your application with a proper database to solve this
issue. Imagine a case where all of the data for your application (registered users, order
transactions, and social interactions) was stored within the temporary memory of the
server the application is running on. The moment that server is turned off or rebooted,
all of your application data will be lost. Relying on a database to store this data
permanently is crucial to any dynamic application.
In this chapter, the following topics will be covered:
• Connecting to MongoDB
• An introduction to Mongoose
• Schemas and models
• Adding CRUD to our controllers
In the previous chapter, we wrote and accounted for the actual logic of our application.
The next step in building our application is to connect it to a database so that our user's
interactions and data can be permanently saved and retrieved. Technically, we can get
around this by storing data in the memory, but the moment our web server restarts or
crashes, all of that data will be lost. Without connecting our application to a database
server to persist data, every input interacted by a visitor will be obsolete. Without
some kind of database server to store our data, most of the websites we interact with
on a daily basis wouldn't even exist.
[ 113 ]
Persisting Data with MongoDB
Here is a general breakdown of how our data is going to be persisted for every
visitor interaction in our app:
Consider the preceding diagram, which reflects the typical lifecycle of a web
application request:
1. A visitor submits a request to view a page on our application via their
web browser.
2. Node.js server receives this request and queries a MongoDB server for
any data.
3. MongoDB server returns the queried data back to our Node.js server.
4. The Node.js server takes that data, builds it into the view model, and then
sends the rendered HTML page back to the browser.
5. The web browser receives the response from our Node.js server and
renders the HTML.
6. This cycle repeats typically for every interaction by every visitor.
For the purposes of this book, we are using MongoDB as our primary data store—
but the reality is that we can use anything to store data: MySQL, PostgreSQL, MS
SQL, the filesystem, and so on.
Using MongoDB with Node.js
Before we officially implement MongoDB into our actual application, let's first take a
look at some basic examples of connecting to a MongoDB server from within Node.js.
Create a new project folder to store some sample code to experiment with. I'll call
my folder mongotest. Inside this folder, create a new file called test.js. In this file,
we will play around with some code to test how to connect to MongoDB and how to
insert and retrieve some data. The first thing we need to do in order to connect to a
MongoDB server from node.js is to require a mongodb module.
[ 114 ]
Chapter 7
To get started, change directories into the new mongotest folder and install the
mongodb module using npm:
$ cd mongotest
$ npm install mongodb
Don't be confused by the module's name. The mongodb npm
module isn't MongoDB itself, but rather a third-party npm module
that facilitates communicating to a MongoDB server from within
Node.js. Also, because this is just a sample project to experiment
with, we don't require the --save flag with npm install since
we aren't maintaining a package.json file.
Connecting to MongoDB
Now that the mongodb module is installed, we can use it in our experimentation file.
Boot up your editor and create a file named test.js. Insert the following block of
code into it:
var MongoClient = require('mongodb').MongoClient;
function(err, db) {
console.log('Connected to MongoDB!');
Executing the preceding code should log Connected to MongoDB! to your screen.
The first thing you'll notice is that we require the mongodb module, but we
specifically use the MongoClient component of the module. This component is the
actual interface we use to actively open a connection to a MongoDB server. Using
MongoClient, we pass the mongodb://localhost:27017/mongotest string URL
to our local server as the first parameter. Notice that the path in the URL points to
the server and then the name of the database itself.
Remember to make sure you have your local MongoDB server instance running in
another terminal for the duration of this chapter. To do so, open a command-line
terminal window and execute $ mongod. Your server should launch and log
information to the screen ending with [initandlisten] waiting for
connections on port 27017.
[ 115 ]
Persisting Data with MongoDB
You might find that when you run your application, you receive
a stack trace error with something like the following code:
thrower; // Unhandled 'error' event
Error: failed to connect to [localhost:27017]
If this happens, you should realize that it failed to connect to
localhost on port 27017—which is the port that our local
mongod server runs under.
Once we have an active connection to our database server, it's as if we are running
the Mongo shell command. The MongoClient callback function returns a database
connection object (that we named db in our code, but could have been named
anything), which is almost exactly the same object we work with in the Mongo shell
when we execute use <databasename>. Knowing this, at this point, we can use
the db object to do anything we can do via the mongo shell. The syntax is slightly
different, but the idea is generally the same.
Inserting a document
Let's test out our new db object by inserting a record into a collection:
var MongoClient = require('mongodb').MongoClient;
function(err, db) {
console.log('Connected to MongoDB!');
// using the db connection object, save the collection 'testing'
to a separate variable:
var collection = db.collection('testing');
// insert a new item using the collection's insert function:
collection.insert({'title': 'Snowcrash'}, function(err, docs) {
// on successful insertion, log to the screen the new
// collection's details:
console.log(docs.length + ' record inserted.');
console.log(docs[0].title + ' – ' + docs[0]._id);
// finally close the connection:
[ 116 ]
Chapter 7
In the preceding code, we establish a connection to the database and execute a
callback once the connection is complete. That callback receives two parameters,
the second of which is the db object itself. Using the db object, we can get a collection
we want to work with. In this case, we save that collection as a variable so that we
can more easily work with it throughout the rest of our code. Using the collection
variable, we execute a simple insert command and pass in the JSON object we want
to insert into the database as the first parameter.
The callback function that executes after insert accepts two parameters, the second
of which is an array of documents affected by the command; in this case, an array of
documents that we inserted. Once insert is complete and we are inside the callback
function, we log some data. You can see that the length of the docs.ops array is
1 as we only inserted a single document. Furthermore, you can see that the single
document in the array is the document we inserted, although now it has an extra
_id field since MongoDB handles that automatically.
Retrieving a document
Let's prove our code a little bit more by adding the findOne function call to
look up the document we just inserted. Change the code in test.js to match
the following example:
var MongoClient = require('mongodb').MongoClient;
function(err, db) {
console.log('Connected to MongoDB!');
var collection = db.collection('testing');
'title': 'Snowcrash'
}, function(err, docs) {
console.log(docs.ops.length + ' record inserted.');
console.log(docs[0]._id + ' - ' + docs[0].title);
title: 'Snowcrash'
}, function(err, doc) {
console.log(doc._id + ' - ' + doc.title);
[ 117 ]
Persisting Data with MongoDB
In this code, we insert a record in exactly the same way as before; only this time,
we perform findOne on title. The findOne function accepts a JSON object to
match against (this can be as precise or loose as you want) as its first parameter.
The callback function that executes after findOne will contain the single document
that was found as its second parameter. If we executed a find operation, we would
have received an array of matching documents based on the search criteria.
The output of the last mentioned code should be:
$ node test.js
Connected to MongoDB!
1 record inserted.
538bc3c1a39448868f7013b4 - Snowcrash
538bc3c1a39448868f7013b4 – Snowcrash
In your output, you might notice that the _id parameter being reported on insert
doesn't match the one from findOne. This is likely the result of running the code
multiple times, which results in multiple records with the same title being inserted.
The findOne function will return the first document found in no particular order,
so chances are the document returned might not be the last one inserted.
Now that you have a basic understanding of how to easily connect and communicate
with a MongoDB server from Node.js. Let's take a look at how we can work with
MongoDB in a way that's a little less raw.
Introducing Mongoose
While working directly with the mongodb module is great, it's also a bit raw and
lacks any sense of the developer friendliness that we've come to expect working with
frameworks such as Express in Node.js. Mongoose is a great third-party framework
that makes working with MongoDB a breeze. Mongoose is an elegant mongodb object
modeling for Node.js.
What that basically means is that Mongoose gives us the power to organize our
database by using schemas (also known as model definitions) and provide powerful
features to our models such as validation, virtual properties, and more. Mongoose
is a great tool as it makes working with collections and documents in MongoDB feel
much more elegant. The original mongodb module is a dependency of Mongoose, so
you can think of Mongoose as being a wrapper on top of mongodb much like Express
is a wrapper on top of Node.js—both abstract away a lot of the raw feeling and give
you easier tools to work with directly.
[ 118 ]
Chapter 7
It's important to note that Mongoose is still MongoDB, so everything you're familiar
with and used to will work pretty much the same way; only the syntax will change
slightly. This means that the queries, inserts, and updates that we know and love
from MongoDB work perfectly fine with Mongoose.
The first thing we need to do is install Mongoose so that it's available to use within
our mongotest project:
$ npm install mongoose
After installation, let's take a look at some of the features that Mongoose has to offer
and what we'll take advantage of to make our lives easier when developing apps that
rely heavily on a MongoDB database.
In Mongoose, schemas are what we use to define our models. Think of them as a
blueprint that all models you create throughout the app will derive from. Using
schemas, you can define much more than the simple blueprint of a MongoDB model.
You can also take advantage of the built-in validation that Mongoose provides by
default, add static methods, virtual properties, and more.
The first thing we do while defining a schema for a model is build a list of every field
we think we will need for a particular document. The fields are defined by type, and
the standard data types you would expect are available as well as a few others:
• String: This type stores a string value.
• Number: This type stores a number value, with restrictions.
• Date: This type holds a date and time object.
• Buffer:
• Boolean: This type is used to store a Boolean (true/ false) value.
• Mixed: This is basically an unstructured object that can contain anything.
Consider this when storing JSON type data or data that is arbitrary and
can literally be any JSON representation. It doesn't need to be predefined.
• ObjectID: This is typically used when you want to store the ObjectID of
another document in a field, for example, when defining a relationship.
• Array: This is a collection of other schemas (models).
[ 119 ]
Persisting Data with MongoDB
Here is an example of a basic Mongoose schema definition:
var mongoose = require('mongoose'),
Schema = mongoose.Schema;
var Account = new Schema({
username: { type: String },
date_created: { type: Date, default: Date.now },
visits: { type: Number, default: 0 },
active: { type: Boolean, default: false }
Here, we define our schema for an Accounts collection. The first thing we do is
require Mongoose and then define a schema object using mongoose.Schema in our
module. We define a schema by creating a new Schema instance with a constructor
object that defines the schema. Each field in the definition is a basic JavaScript object
with type and then an optional default value.
A model in Mongoose is a class that can be instantiated (defined by a schema).
Using schemas, we define models and then use them as a regular JavaScript object.
The benefit is that the model object has the added bonus of being backed by Mongoose,
so it also includes features such as saving, finding, creating, and removing. Let's take a
look at defining a model using a schema and then instantiating a model and working
with it. Add another file called model.js to your experimentation folder, mongotest/
model.js, and include the following block in it:
var mongoose = require('mongoose'),
Schema = mongoose.Schema;
mongoose.connection.on('open', function() {
console.log('Mongoose connected.');
var Account = new Schema({
username: { type: String },
date_created: { type: Date, default: Date.now },
visits: { type: Number, default: 0 },
active: { type: Boolean, default: false }
[ 120 ]
Chapter 7
var AccountModel = mongoose.model('Account', Account);
var newUser = new AccountModel({ username: 'randomUser'
Running the preceding code should result in something similar to the following:
$ node model.js
Mon Jun 02 2014 13:23:28 GMT-0400 (EDT)
Creating a new model is great when you're working with new documents and you
want a way to create a new instance, to populate its values, and then to save it to
the database:
var AccountModel = mongoose.model('Account', Account);
var newUser = new AccountModel({ username: 'randomUser' });
Calling .save on a mongoose model will trigger a command to MongoDB that
will perform the necessary insert or update statements to update the server.
When you switch over to your mongo shell, you can see the new user was indeed
saved to the database:
>use mongotest
switched to db mongotest
{ "username" : "randomUser", "_id" :
ObjectId("538cb4cafa7c430000070f66"), "active" : false, "visits" : 0,
"date_created" : ISODate("2014-06-02T17:30:50.330Z"), "__v" : 0 }
Note that without calling .save() on the model, the changes to the model won't
actually be persisted to the database. Working with Mongoose models in your Node
code is just that—code. You have to execute MongoDB functions on a model for any
actual communication to occur with the database server.
[ 121 ]
Persisting Data with MongoDB
You can use AccountModel to perform a find operation and return an array of
AccountModel objects based on some search criteria that retrieve results from the
MongoDB database:
assuming our collection has the
{ username: 'randomUser1', age:
{ username: 'randomUser2', age:
{ username: 'randomUser3', age:
{ username: 'randomUser4', age:
following 4 records:
21 }
25 }
18 }
32 }
AccountModel.find({ age: { $gt : 18, $lt : 30} }, function(err,
// => 2
// => randomUser1
Here, we use the standard MongoDB commands $gt and $lt for the value of age
when passing in our query parameter to find documents (that is, find any document
where the age is above 18 and below 30). The callback function that executes after find
references an accounts array, which is a collection of AccountModel objects returned
from the query to MongoDB. As a general means of good housekeeping, we close the
connection to the MongoDB server after we are finished.
Built-in validation
One of the core concepts of Mongoose is that it enforces a schema on top of a
schema-less design such as MongoDB. In doing so, we gain a number of new
features, including built-in validation. By default, every schema type has a
built-in required validator available. Furthermore, numbers have both min and
max validators and strings have enumeration and matching validators. Custom
validators can also be defined via your schemas. Let's take a brief look at some
validation added to our example schema from earlier:
var Account = new Schema({
username: { type: String, required: true },
date_created: { type: Date, default: Date.now },
visits: { type: Number, default: 0 },
active: { type: Boolean, default: false },
age: { type: Number, required: true, min: 13, max: 120 }
[ 122 ]
Chapter 7
The validation we added to our schema is that the username parameter is now
required, and we included a new field called age, which is a number that must be
between 13 and 120 (years). If either value doesn't match the validation requirements
(that is username is blank or age is less than 13 or greater than 120), an error will
be thrown.
Validation will fire automatically whenever a model's .save() function is called;
however, you can also manually validate by calling a model's .validate() function
with a callback to handle the response. Building on the example, add the following
code, which will create a new mongoose model from the schema defined:
var AccountModel = mongoose.model('Account', Account);
var newUser = new AccountModel({ username: 'randomUser', age: 11 });
newUser.validate(function(err) {
// the same error would occur if we executed:
// newUser.save();
Running the preceding code should log the following error to the screen:
{ message: 'Validation failed',
name: 'ValidationError',
{ age:
{ message: 'Path 'age' (11) is less than minimum allowed value
name: 'ValidatorError',
path: 'age',
type: 'min',
value: 11 } } }
You can see that the error object that is returned from validate is pretty useful
and provides a lot of information that can help when validating your model and
returning helpful error messages back to the user.
Validation is a very good example of why it's so important to always accept an error
object as the first parameter to any callback function in Node. It's equally important
that you check the error object and handle it appropriately.
[ 123 ]
Persisting Data with MongoDB
Static methods
Schemas are flexible enough so that you can easily add your own custom static
methods to them, which then become available to all of your models that are defined
by that schema. Static methods are great for adding helper utilities and functions
that you know you're going to want to use with most of your models. Let's take our
simple age query from earlier and refactor it so that it's a static method and a little
more flexible:
var Account = new Schema({
username: { type: String },
date_created: { type: Date, default: Date.now },
visits: { type: Number, default: 0 },
active: { type: Boolean, default: false },
age: { type: Number, required: true, min: 13, max: 120 }
Account.statics.findByAgeRange = function(min, max, callback) {
this.find({ age: { $gt : min, $lte : max} }, callback);
var AccountModel = mongoose.model('Account', Account);
AccountModel.findByAgeRange(18, 30, function(err, accounts){
// => 2
Static methods are pretty easy to implement and will make your models much more
powerful once you start taking full advantage of them!
Virtual properties
Virtual properties are exactly what they sound like—fake properties that don't
actually exist in your MongoDB documents, but you can fake them by combining
other real properties. The most obvious example of a virtual property would be
a field for fullname, when only firstname and lastname are actual fields in the
MongoDB collection. For fullname, you simply want to say, return the model's
first and last name combined as a single string and label it fullname:
// assuming the Account schema has firstname and lastname defined:
.get(function() {
return this.firstname + ' ' + this.lastname;
[ 124 ]
Chapter 7
.set(function(fullname) {
var parts = fullname.split(' ');
this.firstname = parts[0];
this.lastname = parts[1];
We call the .get() and .set() functions. It's not required to provide both, although
it's fairly common.
In this example, our get() function simply performs basic string concatenation and
returns a new value. Our .set() function performs the reverse—splitting a string
on a space and assigning the models firstname and lastname field values with
each result. You can see that the .set() implementation is a little flakey if someone
attempts to set a model's fullname with a value of say, Dr. Kenneth Noisewater.
It's important to note that virtual properties are not persisted to MongoDB since they
are not real fields in the document or collection.
There's a lot more you can do with Mongoose, and we only just barely scratched
the surface. Fortunately, it has a fairly in-depth guide you can refer to at the
following link:
Definitely spend some time reviewing the Mongoose documentation so that you are
familiar with all of the powerful tools and options available.
That concludes our introduction to Mongoose's models, schemas, and validation.
Next up, let's dive back into our main application and write the schemas and
models that we will be using to replace our existing sample viewModels as well
as connecting with Mongoose.
Connecting with Mongoose
The act of connecting to a MongoDB server with Mongoose is almost identical to the
method we used earlier when we used the mongodb module.
First, we need to ensure that Mongoose is installed. At this point, we are going
to be using Mongoose in our main app, so we want to install it in the main project
directory and also update the package.json file. Using your command-line terminal
program, change locations to your projects folder, and install Mongoose via npm,
making sure to use the --save flag so that the package.json file is updated:
$ cd ~/projects/imgPloadr
$ npm install mongoose --save
[ 125 ]
Persisting Data with MongoDB
With Mongoose installed and the package.json file updated for the project,
we're ready to open a connection to our MongoDB server. For our app, we are
going to open a connection to the MongoDB server once the app itself boots up
and maintains an open connection to the database server for the duration of the
app's lifetime. Let's edit the server.js file to include the connection code we need.
First, include Mongoose in the app by requiring it at the very top of the file:
var express = require('express'),
config = require('./server/configure'),
app = express(),
mongoose = require('mongoose');
Then, insert the following code right after the app = config(app); line:
mongoose.connection.on('open', function() {
console.log('Mongoose connected.');
That's it! Those few simple lines of code are all it takes to open a connection to a
MongoDB server, and our app is ready to start communicating with the database.
The only parameter we pass to the connect function of Mongoose is a URL string
to our locally running MongoDB server and a path to the collection we want to use.
Then, we add an event listener to the open event of the mongoose.connection
object and, when that fires, we simply log an output message that the database
server has connected.
Defining the schemas and models
For the purposes of the application we are building, we're only going to have two
different unique schemas and associated models: an Image model and a Comment
model. If we were to take this application to production and really build it out with
all of the necessary features, we could expect to have many more models as well.
First, create a new directory labeled models in your project and we will store the
Node.js modules for each of our models here. Create three files named image.js,
comment.js, and index.js in this directory. Let's take a look at the Image model
first. Copy the following block of code into the models/image.js file:
var mongoose = require('mongoose'),
Schema = mongoose.Schema,
path = require('path');
var ImageSchema = new Schema({
[ 126 ]
Chapter 7
String },
String },
String },
Number, 'default': 0 },
Number, 'default': 0 },
Date, 'default': Date.now }
.get(function() {
return this.filename.replace(path.extname(this.filename), '');
module.exports = mongoose.model('Image', ImageSchema);
First, we define our ImageSchema with the various fields that we are going to want
to store in MongoDB for each of the images. We created a virtual property of
uniqueId, which is just the filename with the file extension removed. As we want
our Image model to be available throughout the rest of our app, we export it using
module.exports. Note that we are exporting the model not the schema (as the
schema itself is fairly useless to us). Let's set up a similar model for comments.
Copy the following block of code into the models/comment.js file:
var mongoose = require('mongoose'),
Schema = mongoose.Schema,
ObjectId = Schema.ObjectId;
var CommentSchema =
{ type:
{ type:
{ type:
{ type:
{ type:
timestamp: { type:
new Schema({
ObjectId },
String },
String },
String },
String },
Date, 'default': Date.now }
this._image = image;
}).get(function() {
return this._image;
module.exports = mongoose.model('Comment', CommentSchema);
[ 127 ]
Persisting Data with MongoDB
There are a few important things to take note of with this model. First, we have
a field labeled image_id, which has an ObjectId type. We're going to use this
field to store the relationship between comment and image that it was posted to.
The ObjectId that gets stored in this field is the _id of the related image document
from MongoDB.
We also define virtual on the comment schema labeled image, which we provide a
getter and setter for. The image virtual property will be how we attach the related
image when we retrieve comments later in our controllers. For every comment, we
are going to iterate through and look up its associated image and attach that image
object as a property of the comment.
Handling the name of collections
You name your models using singular terms, and Mongoose
will recognize this and create your collections using a
pluralized model name. So, a model defined as Image will
have a collection in MongoDB named images. Mongoose tries
to be smart about this; however, a model defined as Person
will have a corresponding collection named people and so on.
(And yes, octopus will result in octopi!)
The models index file
There's one last file in the models folder that we haven't yet touched on in our
project. The index.js file within any folder in Node.js acts as an index file for
the modules within it. This is by convention, so you don't have to adhere to this
if you don't want to.
Since our models folder will contain a number of different files, each a unique
module for one of our models, it would be nice if we could just include all of our
models in a single require statement. Using the index.js file, we can do so pretty
easily too. Copy the following block of code into the models/index.js file:
module.exports = {
'Image': require('./image'),
'Comment': require('./comment')
[ 128 ]
Chapter 7
The index.js file inside the models directory simply defines a JavaScript object
that consists of a name-value pair for each module in our directory. We manually
maintain this object, but this is the simplest implementation of the concept. Now,
thanks to this basic file, we can perform require('./models') anywhere in our
application and know that we have a dictionary of each of our models via that
module. To reference a specific model in that module, we simply refer to the specific
model as a property of the module. If we only wanted to require a specific model
somewhere in our app instead, we can perform require('./models/image') just
as easily! You will see more of this a little later and it will become much more clear.
Because our two models are so closely related, we are typically always going to require
the models dictionary using require('./models') throughout our application.
Adding CRUD to the controllers
CRUD stands for Create, Read, Update, and Delete. Now that our schemas are
defined and our models are ready, we need to start using them throughout our
application by updating our controllers with various CRUD methods, where
necessary. Up until this point, our controllers have consisted of only fixture, or fake,
data so we can prove that our controllers are working and our view models were
wired up to our templates. The next logical step in our development is to populate
our view models with data directly from MongoDB. It would be even better if we
could just pass our Mongoose models right to our templates as the viewModel itself.
The home controller
If you recall from the Updating the home controller section of Chapter 6, Controllers
and View Models, we originally created viewModel in our home controller, which
consisted of an array of JavaScript objects that were just placeholder fixture data:
var viewModel = {
images: [
'Sample Image 1',
}, {
[ 129 ]
Persisting Data with MongoDB
}, {
}, {
'Sample Image 2',
'Sample Image 3',
'Sample Image 4',
We are going to replace that viewModel with a very stripped down version that we
will then populate with real data from our Mongoose models:
var viewModel = {
images: []
Before we can populate that viewModel with real data, we need to first make sure
our home controller can use our models. To do so, we must require the models
module. Include this at the very top of the controllers/home.js file:
var sidebar = require('../helpers/sidebar'),
ImageModel = require('../models').Image;
[ 130 ]
Chapter 7
We could have required the full models module and have had access to both the
Comment model and the Image model; however, for the homepage, we really only
need to use the Image model. Now our mongoose model for Image is available to
our home controller, we can perform a find operation to retrieve a list of the newest
images to display on the homepage. Replace the existing sidebar() call in your
home controller with this updated version of the code:
ImageModel.find({}, {}, { sort: { timestamp: -1 }},
function(err, images) {
if (err) { throw err; }
viewModel.images = images;
sidebar(viewModel, function(viewModel) {
res.render('index', viewModel);
Using ImageModel, we execute a MongoDB find query, but we provide no specifics
for the actual query (a blank JavaScript object), which means it will return every
document. The second parameter is also a blank JavaScript object, which means we
aren't specifying how to map the results, so the full schema will be returned. The
third parameter is an options object where we can specify things such as the sort
field and order. In this particular query, we are retrieving every single image in the
images collection sorted by timestamp in descending order (ascending order would
have had a value of 1 instead of -1).
The callback function that executes after a successful find query to the MongoDB
database server will return both an error object and an images array of matching
models; in our case, every image in the database. Using the array that's returned
from the query, we simply attach it to our viewModel via its images property.
Then, we call our sidebar function exactly as we did previously.
At this point, we are no longer populating viewModel with fixture data but instead
populating it with exactly what is returned from the database when we perform a basic
find query using our Mongoose Image model. The homepage for the application is
officially data driven. Here is a recap of the entire controllers/home.js file:
var sidebar = require('../helpers/sidebar'),
ImageModel = require('../models').Image;
module.exports = {
index: function(req, res) {
var viewModel = {
[ 131 ]
Persisting Data with MongoDB
images: []
ImageModel.find({}, {}, { sort: { timestamp: -1 }},
function(err, images) {
if (err) { throw err; }
viewModel.images = images;
sidebar(viewModel, function(viewModel) {
res.render('index', viewModel);
If you were to run the app and open it in a browser, you wouldn't actually see
anything on the homepage. That's because we haven't actually inserted any data yet.
That's coming up next. However, note that the page itself still works and you didn't
get any errors. This is because MongoDB is simply returning an empty array from
the find on ImageModel, which the Handlebars homepage template is handling fine
because it's performing an each operation against an empty array so it's displaying
zero images on the homepage.
The image controller
The image controller is by far the biggest component of our application. It contains
most, if not all, of the logic that's powering our app. This includes displaying all of
the details for an image, handling the uploading of images, and handling likes and
comments. There's a lot to cover in this controller, so let's break it down by each section.
Index – retrieving an image model
The primary responsibility of the index function in our image controller is to retrieve
the details for a single specific image and display that via its viewModel. In addition
to the details for the actual image, the comments for an image are also displayed on
the page in the form of a list. Whenever an image is viewed, we also need to update
the views count for the image and increment it by one.
Begin by editing the controllers/image.js file and updating the list of required
modules at the top to include our models module:
var fs = require('fs'),
path = require('path'),
sidebar = require('../helpers/sidebar'),
Models = require('../models');
[ 132 ]
Chapter 7
We also want to strip viewModel down to its most basic form, exactly as we did in
the home controller. Replace the existing viewModel object variable with this new,
lighter version:
var viewModel = {
image: {},
comments: []
After defining blank viewModel, let's include a find call on the Image model so that
we can look up an image specifically by its filename:
Models.Image.findOne({ filename: { $regex: req.params.image_id }
function(err, image) {
if (err) { throw err; }
if (image) {
// to do...
} else {
In the preceding code, we are using the Models module's Image model and
performing findOne, which is identical to find, except it will only ever return
a single document (matching or not) instead of an array as find returns. By
convention, we use a singular variable name in our callback's second parameter
versus a plural, just so we as developers can easily tell we are working with a
single object or an array/collection of objects.
The query object we provide as the first parameter matches the filename field of an
image document using MongoDB's regex filter and compares this to req.params.
image_id, which is the value of the parameter in the URL as defined in our routes
file. The URL for an image page will always be http://localhost:3300/images/
abcdefg, where abcdefg will be the value of req.params.image_id. If you recall, we
are randomly generating this value in the create function when an image is uploaded.
After checking to make sure our err object isn't null, we then check to make sure our
image object is also not null. If it's not null, that means a model was returned from
MongoDB; so, we found our image and we're good to go. If an image model wasn't
returned because we tried searching for an image by a filename that doesn't exist,
we simply redirect the user back to the homepage.
[ 133 ]
Persisting Data with MongoDB
Let's now populate our viewModel by inserting the following lines into the area
where we have the // to do... placeholder comment:
image.views = image.views + 1;
viewModel.image = image;
We attach the image model that was returned from findOne to our viewModel.
image property, but not before incrementing the views property of that model by
1 (so that we represent our actual plus one view as we load the page). Since we
modified the model (by incrementing its views count), we need to ensure that it's
saved back to MongoDB so we call the model's save function.
Now that viewModel has been updated with the image model and the views
count has been incremented and saved, we next need to retrieve a list of comments
associated with the image. Let's include a little bit more code to query the Comment
model and find any comments that belong to the image. Insert the following block
of code immediately after image.save(); from earlier:
Models.Comment.find({ image_id: image._id}, {}, { sort: {
'timestamp': 1 }},
function(err, comments){
if (err) { throw err; }
viewModel.comments = comments;
sidebar(viewModel, function(viewModel) {
res.render('image', viewModel);
Using find on our Comment model, we can pass in an object that contains our query
as the first parameter; in this case, we are specifying that we want all comments
where the image_id field is equal to the _id property of the main image model
we attached to our viewModel earlier.
That code might look a little odd, so let's elaborate. Remember that the image
object that is returned from the original Models.Image.findOne() call is available
throughout the entire scope of that callback function. No matter how deep we get
nesting callback functions, we will always have access to that original image model.
Therefore, we can access it and its properties inside the callback function that fires
when our Model.Comment.find() has executed.
[ 134 ]
Chapter 7
Once inside the find callback of Comment, we attach the comments array that was
returned to our viewModel and then call our sidebar function exactly as we did
before when we first opened the controller and started editing this index function.
As a review, here is the entire index function inside the controllers/image.js file
after it's been completely updated:
index: function(req, res) {
// declare our empty viewModel variable object:
var viewModel = {
image: {},
comments: []
// find the image by searching the filename matching the url
Models.Image.findOne({ filename: { $regex: req.params.image_id }
function(err, image) {
if (err) { throw err; }
if (image) {
// if the image was found, increment its views counter
image.views = image.views + 1;
// save the image object to the viewModel:
viewModel.image = image;
// save the model (since it has been updated):
// find any comments with the same image_id as the image:
Models.Comment.find({ image_id: image._id},{},{ sort: {
'timestamp': 1 }},
function(err, comments){
// save the comments collection to the viewModel:
viewModel.comments = comments;
// build the sidebar sending along the viewModel:
sidebar(viewModel, function(viewModel) {
// render the page view with its viewModel:
res.render('image', viewModel);
} else {
// if no image was found, simply go back to the homepage:
[ 135 ]
Persisting Data with MongoDB
Let's quickly recall all of the index controller's responsibilities and tasks:
• Create a new empty viewModel object
• Create the findOne imagefindOneimage model where the filename
is a regex match to the URL image_id parameter
• Increment the found views of image by one
• Attach the found image model to viewModel
• Save the image model since its views have been updated
• Find all comments with the image_id property equal to the _id of
the original image model
• Attach the array of found comments to viewModel
• Render the page using sidebar, passing in the viewModel and
callback function
Create – inserting an image model
We already have the functionality in place in our create function to handle
randomly naming and uploading an image file. Now we need to save that
information to MongoDB for the uploaded image.
Let's update the original saveImage function inside controllers/images.
js:create and include the functionality to tie it into the database.
Our goal with the saveImage function is two-fold. First, we want to make sure
that we never save an image to the database with the same randomly generated
filename as an already existing image. Second, we want to ensure that we only
insert the image into the database after it has been successfully uploaded, renamed,
and saved to the filesystem. We are going to make two modifications to the existing
code to achieve this.
The first modification is to wrap the bulk of the logic with find against the randomly
generated filename, and if any documents are returned from MongoDB as a match,
we need to start the process over to repeat this as many times as necessary until we
achieve a truly unique filename. The code to perform the search is as follows:
Models.Image.find({ filename: imgUrl }, function(err, images) {
if (images.length> 0) {
} else {
// do all the existing work...
[ 136 ]
Chapter 7
If an images array that is returned from find has a length greater than zero, it means
at least one image was found to have the same filename as was generated with our
random for loop. If that's the case, we want to call saveImage again, which will
repeat the whole process (randomly generate a new name and perform a find on
the database for that new name). We do this by previously defining the saveImage
function as a variable so that within the saveImage function itself, we can execute it
again by calling the original variable as the function.
A function that calls itself is called a recursive function.
Assuming no images were returned from find, it means we have generated a truly
unique filename for our image and we are safe to rename the file and upload it to the
server as well as save a record to the database.
Originally, the last step of the create function was to redirect the visitor to the
image's page within the callback that fired when the filesystem rename was finished.
This is where we're going to want to create a new Mongoose image model. We
should redirect only when the database server has finished saving the image (again
relying on a callback function). Consider the following line in the original function:
res.redirect('/images/' + imgUrl);
Replace this with this new block of code:
var newImg = new Models.Image({
title: req.body.title,
description: req.body.description,
filename: imgUrl + ext
newImg.save(function(err, image) {
console.log('Successfully inserted image: ' + image.filename);
res.redirect('/images/' + image.uniqueId);
Here, we create a brand new Image model and pass in the default values via its
constructor. The title and description fields get set right from the values passed
in via the HTML form using req.body and the form field names (.title, title,
and .description). The filename parameter is what we build the same way we
did originally when we set its destination for renaming it, except we don't include
the path and directory names, just the randomly generated filename and the image's
original extension.
[ 137 ]
Persisting Data with MongoDB
We call the model's .save() function (just as we did earlier when we updated the
image's views property in the index controller function). The save function accepts
a second parameter in its callback, which will be the updated version of itself.
Once the save is completed, and the image has been inserted into the MongoDB
database, we then redirect to the image's page. The reason the callback returns the
updated version of itself is because MongoDB will automatically include additional
information such as _id.
As a review and sanity check, here is the complete code for the saveImage function
in controllers/image.js:create, with the new lines of code clearly highlighted:
var saveImage = function() {
var possible = 'abcdefghijklmnopqrstuvwxyz0123456789',
imgUrl = '';
for(var i=0; i< 6; i+=1) {
imgUrl += possible.charAt(Math.floor(Math.random() *
/* Start new code: */
// search for an image with the same filename by performing a find:
Models.Image.find({ filename: imgUrl }, function(err, images) {
if (images.length> 0) {
// if a matching image was found, try again (start over):
} else {
/* end new code:*/
var tempPath = req.files.file.path,
ext = path.extname(req.files.file.name).toLowerCase(),
targetPath = path.resolve('./public/upload/' + imgUrl + ext);
if (ext === '.png' || ext === '.jpg' || ext === '.jpeg' || ext ===
'.gif') {
fs.rename(tempPath, targetPath, function(err) {
if (err) { throw err; }
/* Start new code: */
// create a new Image model, populate its details:
var newImg = new Models.Image({
title: req.body.title,
filename: imgUrl + ext,
description: req.body.description
[ 138 ]
Chapter 7
// and save the new Image
newImg.save(function(err, image) {
res.redirect('/images/' + image.uniqueId);
/* End new code: */
} else {
fs.unlink(tempPath, function () {
if (err) { throw err; }
res.json(500, {error: 'Only image files are allowed.'});
/* Start new code: */
/* End new code: */
Don't forget to initially execute saveImage() right after the function is defined;
otherwise, nothing will happen!
Testing everything out so far
At this point, we have most of the key functionalities wrapped with MongoDB
integration, and our app should really feel like it's coming together. Let's give it
a test run and make sure all of our endpoints are working so far. Launch the app
and open it in a browser:
$ node server.js
Server up: http://localhost:3300
Mongoose connected.
[ 139 ]
Persisting Data with MongoDB
Open up a browser, point it to http://localhost:3300, and you should see your
application up and running, as shown in the following screenshot:
[ 140 ]
Chapter 7
Go ahead and use the form on the homepage to search for an image file on your
computer and select it. Provide an input for Title and Description and click on the
Upload button. You should be taken directly to the image page with the details for
your uploaded image displayed:
[ 141 ]
Persisting Data with MongoDB
Go back to the homepage, and you should now see your new image displayed under
the Newest Images section:
The Like button and updating an image model
Next, let's add support for the Like button. Remember that our Like button works
a little differently. It uses AJAX with jQuery so that data can be sent and received in
real time without reloading the entire page. The experience for the user is seamless
and enjoyable, as they don't lose their scroll place on the page or experience any
other jarring UI-related issues.
The endpoint that the Like button hits is /images/:image_id/like, so we are going
to use the value in the URL for image_id to find and retrieve the image in MongoDB,
increment its likes value by 1, and then return the new total number of likes for
the image (so that the UI can update with the new value).
[ 142 ]
Chapter 7
Currently, the like function in controllers/image.js only does a simple JSON
response with a hardcoded value of 1:
res.json({likes: 1});
Let's replace that original code with new code that will use the Mongoose Image model
to find an image with a filename that matches the image_id passed in via the URL:
Models.Image.findOne({ filename: { $regex: req.params.image_id }
function(err, image) {
if (!err && image) {
image.likes = image.likes + 1;
image.save(function(err) {
if (err) {
} else {
res.json({ likes: image.likes });
Assuming the callback function receives a valid image model response from the
query, we'll then increment its likes property, and since the model is then modified,
we need to execute its save function. Inside the save function's callback, we send a
JSON response back to the browser with the real current value of the image's likes.
Sometimes, we will use shorthand in JavaScript and write
something similar to the following:
if (!err && image)
In the if statement in the preceding example, we are saying,
"if the err object is false (that is null) and the image object
is true (that is not null), then we're good to go!"
With this code in place, you can run the app again and test out the Like button
by viewing the image that you uploaded earlier and simply clicking on Like. If it
worked, the counter next to the button should increase by one. Refresh the page,
and the likes count should remain as the new value.
[ 143 ]
Persisting Data with MongoDB
Comment – inserting a comment model
Inserting comments will work almost exactly the same way as the likes for an image.
The only difference is that we are creating a new comment model instead of updating
an image model. The original code we had in our comment function was:
res.send('The image:comment POST controller');
Let's replace this with some code that will find the image by image_id in the URL
again, but this time instead of updating its likes, we are going to create a new
comment and assign the comment's image_id value with the _id of the image
we are currently viewing (this is to attach a relationship to the comment so that it
actually belongs to an image). Replace the entire comment function in controllers/
image.js with the following block of code:
Models.Image.findOne({ filename: { $regex: req.params.image_id }
function(err, image) {
if (!err && image) {
var newComment = new Models.Comment(req.body);
newComment.gravatar = md5(newComment.email);
newComment.image_id = image._id;
newComment.save(function(err, comment) {
if (err) { throw err; }
res.redirect('/images/' + image.uniqueId + '#' + comment._id);
} else {
Here, you can see that we are using the same code from the like function to query
MongoDB and find the image with the matching filename from the URL.
Assuming a valid image is returned as a match, we create a new comment object
called newComment and actually pass the entire HTML form body into the constructor.
This is a bit of a cheat as it's a coincidence (not accidental) that our HTML form uses
form fields that have the same name and structure as that of a comment model. If you
were to perform the console.log operation on the req.body object, you would see
something like the following:
name: 'Jason Krol',
email: 'jason@kroltech.com',
comment: 'This is what a comment looks like?!'
[ 144 ]
Chapter 7
That's identical to what we would have built manually anyway, so we just take
a shortcut and pass the whole thing in as it is! After that, we update a few more
properties on the newComment model. First, we manually set a gravatar property,
which is where we will store the MD5 hash value of the commenter's e-mail address
so that we can retrieve their Gravatar profile picture. Gravatar is a universal avatar
service that stores profile pictures based on a user's e-mail address. However, the
unique ID they use for each profile is an MD5 hash value, which is why we have
to store that value.
As we are relying on the third-party MD5 module, we need to ensure that it's
installed in our project and saved to our package.json file as a dependency.
From your project's root folder, execute the following command:
$ npm install MD5 --save
In addition, we need to require the module in the controllers/image.js file at
the very top, along with the other modules we are requiring:
var fs = require('fs'),
path = require('path'),
sidebar = require('../helpers/sidebar'),
Models = require('../models'),
md5 = require('MD5');
Finally, we set the image_id property of newComment to the _id property of the
image we found at the beginning of the function. Then, we call the comment model's
.save() function and redirect the user back to the image page. For convenience,
we append a bookmark to the new comment's _id to the URL so that when the
page loads it will automatically scroll down to the users' comments that have just
been posted.
With that functionality in place, go ahead and fire up the app and open it in your
browser. Visit the image page for any images you've uploaded and post a comment.
Once the comment posts and the page reloads, you should see something like the
following screenshot under an image:
[ 145 ]
Persisting Data with MongoDB
We could have chosen to handle comments using jQuery and AJAX the
same way we handled the Like button. However, this introduces a bit
more complexity because if we were to do that, we would have needed
a slightly more sophisticated way to display that inserted comment to
the screen. This would have involved relying heavily on jQuery to do
some advanced DOM manipulation to display the comment after it
was posted using AJAX.
In a later chapter, when we review Single-Page Applications, we will
take a brief look at some JavaScript frameworks that perform this kind
of functionality and a lot of other advanced features!
That concludes the code and functionality for the image controller.
Wrapping it up
Let's do a quick recap of all of the changes we've made to this controller:
• We updated the index function to retrieve an image from MongoDB and
populate viewModel with the details of the image model. We also found
all the comments related to that image and attached an array of those to
viewModel as well.
• We tweaked the create function to insert a new image model into the
database once it has been successfully renamed and saved to the filesystem.
• The like function was updated to actually increment the value of an image's
likes property and save that value to the database as well as return the new
value via a JSON response.
• Comments are now inserted for a particular image via the comment function.
Not only is a comment model inserted into the database, but its corresponding
image is also found and the image model's _id value is attached to the
comment to solidify a relationship.
The last piece of the puzzle and last area we need to tie into MongoDB is the sidebar.
To do this, we are going to need to update each of the helpers we previously created.
Most of the helpers that we write code for will be using the concepts and functionality
that we've already covered in this chapter. However, there is the addition of one new
concept that I want to focus on before we take a look at the code.
[ 146 ]
Chapter 7
Introducing the async module
As JavaScript by its very nature is asynchronous, there undoubtedly comes a time
when we need a way to handle executing a number of different asynchronous
functions at the same time. The big issue here is that if we tried to perform three
different queries on a MongoDB server for example, how will we know when all three
are finished before we move on and do work with the results? Up until this point,
we've simply been relying on a single callback function, which works great for a single
call. How can we assign a single callback function to multiple asynchronous calls?
The answer is we can't—not directly anyway. You can use a lot of nested callbacks to
achieve this, but that is generally considered bad practice and will significantly reduce
the readability of your code. We can use a third-party module, however, that was
designed very specifically for this exact need.
async is a powerful node module that can be downloaded and installed via npm,
which provides a number of extremely useful utility functions all designed to help
when working with a series of asynchronous functions. Two functions that we are
going to work with in this chapter are series and parallel. The series function
allows us to execute asynchronous functions sequentially, each waiting until the
previous function finishes before executing a single callback function at the end. The
parallel function allows us to do the opposite—execute a number of asynchronous
functions simultaneously, waiting until they all complete executing a single callback
function when the last function is finished. How does a single callback function
handle the responses of a number of different asynchronous functions, you ask?
By accepting an array of the responses of each function as a parameter!
Since we are going to be using async for our project, let's install it via npm and make
sure our package.json file is updated as well. Within the root of your project
folder, execute the following from the command line:
$ npm install --save async
The comments helper
Let's take a look at the first use of async in one of our helpers—the comments
helper. Originally, helpers/comments.js was a module that had a newest
function that returned an array of fixture data with some sample comments.
We are going to completely remove this code and instead query MongoDB for
the newest comments and return those as an array. Start by clearing the comment
helper module and start from scratch (note that we included a new callback
parameter to the newest function):
var models = require('../models'),
async = require('async');
[ 147 ]
Persisting Data with MongoDB
module.exports = {
newest: function(callback) {
// to do...
Notice that we added the additional require statements at the top of the file for
our models and async. Within the newest function, let's replace the // to do...
comment with code to query MongoDB and find the five most recent comments:
models.Comment.find({}, {}, { limit: 5, sort: { 'timestamp': -1 }
function(err, comments){
// to do – attach an image to each comment...
Notice that the first parameter in the find query is an empty JavaScript object,
meaning we will retrieve every comment in the database. For the third parameter,
however, we're using limit and sort so that we limit the number of records
returned to five, and we sort the query by timestamp in descending order.
Now that we have an array of comments, ideally, we'd like the image that each
comment belongs to to be returned as well. Typically, this would be accomplished
by using an aggregate query in MongoDB to join different collections together
(such as a JOIN in SQL). We will see aggregate in more detail in the next chapter.
For the purposes of our code, we're going to instead query MongoDB separately for
each comment and retrieve the image associated with the comment's image_id value.
First, let's define a function that will query MongoDB and retrieve and attach an
image model to a comment model:
var attachImage = function(comment, next) {
models.Image.findOne({ _id : comment.image_id},
function(err, image) {
if (err) throw err;
comment.image = image;
This function will accept a comment model as the first parameter and a callback
function as the second parameter (named next). The next callback as the second
parameter is important because it's the key to how async is able to function. Imagine
that the next callback acts as a chain link. Since the same function is going to be called
for every item in a collection, there needs to be a way to daisy-chain the calls together.
This is performed via the callback.
[ 148 ]
Chapter 7
Basically, every time the callback is called for an item in the array, it performs its
work and then executes the same callback with the next item in the array, and so on
and so forth, which is why we named the callback function parameter next.
Another important element to point out with this function is that when we attach the
image model to the comment's image property, we are using the virtual property
we set up earlier in the main comment's schema. If you recall, when we set the image
property, we are actually setting the private _image property. Likewise, when we get
the image property, we are actually retrieving the private _image property.
After we have defined the attachImage function, we need to use the each function
of async to apply that function to every item in the comments collection:
async.each(comments, attachImage,
function(err) {
if (err) throw err;
callback(err, comments);
The each function of async will loop through every item in the collection in the first
parameter, and send each item as a parameter to a callback function in the second
parameter. The third parameter is the final callback function that is executed once
the entire series is finished with the collection. In this case, every comment in the
comment's array will be passed individually to the attachImage function. When
the entire collection has been iterated through, the final callback will execute, which
basically fires the very first callback function that was passed into the newest
function as its only parameter. Boy, that was a mouthful! Let's try to break this
down a little further so it makes a bit more sense:
• The newest function of the comment helper module accepts a single
parameter named callback—this is the function that will be called
once all of the work is finished in this entire function.
• The first thing the newest function does is find the latest five comments
and returns them as an array to an anonymously defined inline function.
• First, we define a function and store it in a variable named attachImage.
• The attachImage function accepts two parameters, an individual comment
model, and a callback function that we named next.
• The attachImage function will query MongoDB to find an image with
an _id value that is the same as the image_id property of the comment
that was passed into it as the first parameter.
• Once that image is found, it is attached to the comment via its image
property and then the next callback function is executed.
[ 149 ]
Persisting Data with MongoDB
• We use async.each to loop through every comment in the comments array
that was passed as the first parameter to each.
• Pass the attachImage function as the second parameter, which is the
function that will be called for every comment in the comment's array.
• Finally, define an inline anonymous function that will be executed once the
last item in the comments collection has been iterated on. This inline function
itself only accepts an error object as its parameter. Assuming every iteration
of the comments collection was successful, this function will be executed
with no error. Inside this function, we execute the original function named
callback that was the only parameter to the newest function, and callback
is called with the newly updated comment's array as its second parameter.
OK, the hardest part is over! You survived a crash course on the async module and
came out, hopefully, unscathed! Just to be safe, here is the code for the helpers/
comments.js module file in its entirety:
var models = require('../models'),
async = require('async');
module.exports = {
newest: function(callback) {
models.Comment.find({}, {}, { limit: 5, sort: { 'timestamp': -1 } },
function(err, comments){
var attachImage = function(comment, next) {
models.Image.findOne({ _id : comment.image_id},
function(err, image) {
if (err) throw err;
comment.image = image;
async.each(comments, attachImage,
function(err) {
if (err) throw err;
callback(err, comments);
[ 150 ]
Chapter 7
Callback, callback, callbacks everywhere!
At this point, it's probably getting a little confusing with the number
of callbacks we've been dealing with. A part of the problem is
the terminology we've been using. Any function that is passed as
a parameter and only executed after certain conditions are met,
typically as the end result of the original function, is referred to as a
callback. The popular convention with JavaScript is to label a callback
function in a parameter literally with the variable name callback,
so that it's obvious. This works great when you are reading code,
but not so much when you are explaining code and referring to a
function named callback that's also known as the callback!
The sidebar helper
OK! So, of course, there's a catch right!? Well, kind of. Since we introduced async
in our comments helper module, we now need to introduce it in our sidebar
helper. This is because of the simple fact that our comments helper is now really
asynchronous, so anything that uses our comments module needs to deal with that.
As our sidebar module currently stands, it's just expecting the comments helper
module to return an array and do it instantly; so, it's not expecting to have to wait
around for the actual data. Because of this, if we ran our code as is, our comments
sidebar would remain blank (because the sidebar would have rendered the page
before the MongoDB calls were even finished thinking within the comments module).
Let's fix this by updating our sidebar helper module to use async as well.
First, let's edit the helpers/sidebar.js file and replace its entire contents with this
slightly modified version that uses async.parallel:
var Stats = require('./stats'),
Images = require('./images'),
Comments = require('./comments'),
async = require('async');
module.exports = function(viewModel, callback){
function(next) {
next(null, Stats());
function(next) {
next(null, Images.popular());
function(next) {
[ 151 ]
Persisting Data with MongoDB
], function(err, results){
viewModel.sidebar = {
stats: results[0],
popular: results[1],
comments: results[2]
The first thing we did was make sure async was included as a required module at
the top of the file. Inside the primary exports function, we basically wrapped our
existing code and integrated it into async.parallel so that we can easily tweak
it a little later as we update each section of the sidebar helpers. Since we've so far
only completed the comments helper module, that's the only one that's actually been
changed. The other Stats and Images.popular calls are being forcibly used with
async.parallel even though it doesn't quite make sense to do that right now. It
will once those two sections become more asynchronous in the next sections.
The parallel function of async works in a similar way to its each function that we
used earlier. The main difference is that parallel isn't performing the same function
in a loop through a collection, but is instead performing a series of unique functions
all at the same time. If you look closely, you can see that the first parameter to
parallel is actually an array, and each item in the array is a unique function. Every
function in the array accepts a next callback parameter function, which is executed
at the conclusion of each of the functions. The second parameter in the next callback
is the result of the work that was performed within the function itself. In the case of
Stats and Images.popular, those two functions simply return values instantly with
no asynchronous calls to anything else, so we just expect the results to be returned by
executing them directly.
However, as you can see with the Comments.newest section, we are passing in the
next callback function as a parameter because we want its execution to be deferred
until the last second (until Comments.newest has completed all of its work). Once
that next callback function is called, it is passed the results of all of its work.
[ 152 ]
Chapter 7
The last parameter to the parallel function is an inline function that accepts a
results array as its second parameter. This array is a collection of each of the results
that were returned from each of the functions in the array in the first parameter.
You can see that when we build viewModel now, we are referring to indexes in the
results array. The index order is the order that the functions were defined in the
original array. We know that the first function was to retrieve Stats, the second
function to retrieve Images.popular, and the third function to retrieve Comments.
newest. So, we can reliably assign results[0] to viewModel.Stats and so on.
As a reference, here is what the viewModel definition originally looked like in the
sidebar module:
viewModel.sidebar = {
stats: Stats(),
popular: Images.popular(),
comments: Comments.newest()
You can compare this with the updated version that uses async:
viewModel.sidebar = {
stats: results[0],
popular: results[1],
comments: results[2]
Now that the sidebar is set up to properly handle the helper modules that are (and
eventually will be) asynchronous, we can run the application and test it to ensure
that our sidebar is properly displaying the top five most recent comments on the
website. Run the application and launch it in a browser. If you haven't already
posted any comments to an image, do so now so that you can see those comments
appearing in the sidebar along with a thumbnail of the image they belong to.
At this point, we've covered and implemented a large number of changes to our
application. It's understandable that something might be broken for you, so let's
run through a quick checklist to make sure we haven't missed any trouble spots
that might be preventing your app from running properly:
• Make sure you've installed all of the required modules for this chapter and
that they are saved to your package.json file. This includes mongoose,
async, and MD5.
• Make sure the appropriate dependency modules are being required at the
top of each of the module files they are used in.
[ 153 ]
Persisting Data with MongoDB
• Make sure you remember to launch mongod in another terminal instance
whenever you run the application.
• When in doubt, pay attention to the stack trace output that node is giving you
in your terminal when it fails, as it is typically pretty obvious what's wrong.
It will also give you the filename and line number of the offending module.
• When all else fails, execute console.log everywhere!
Next up, let's update the stats helper module to use parallel as well so we can get
some real stats for the application.
The stats helper
The primary responsibility of the stats helper module is to gather some totals
for our application. These stats are for things such as the total number of images
uploaded, total number of comments, total views for all the images combined, and
total likes for all the images combined. Your first inclination might be to assume that
we are going to query MongoDB for all the images and loop through every image
to track all of the views and totals. That's one way to do it, but it's pretty inefficient.
Fortunately, MongoDB has some built-in functionalities, which makes generating
these kinds of values a snap.
As we are going to be making a number of calls to MongoDB, we are going to rely
on the async.parallel function again much like we did in the sidebar module.
The original helpers/stats.js file was very bare bones, so let's completely replace
that file with this new version, which uses parallel:
var models = require('../models'),
async = require('async');
module.exports = function(callback) {
function(next) {
next(null, 0);
function(next) {
next(null, 0);
function(next) {
next(null, 0);
function(next) {
next(null, 0);
[ 154 ]
Chapter 7
], function(err, results){
callback(null, {
comments: results[1],
This code does exactly what the module originally did; only it's a little more
verbose! I'm pretty sure we don't want to just return 0 for all of our stats forever
though, as that'd be pretty useless and unimpressive to say the least! Let's update
the each function to properly query MongoDB and get some stats. Looking at
the object returned in the callback in the last function, we can see that we already
defined the order of the functions that are being executed in parallel. Let's start
with images. Replace the next(null, 0); line in the first function with the
following code snippet:
models.Image.count({}, next);
Easy! Just use MongoDB's count method to find the total number of documents in
the images collection matching any criteria (the first parameter). Then, we just pass
the next function as the callback because, coincidentally enough, the parameter
signatures match. If we didn't want to use shorthand here, we could have written
this the long way, as follows:
models.Image.count({}, function(err, total){
next(err, total);
However, who feels like typing all that when you don't have to! Let's do the same
thing for the second function in the parallel array for total comments. Replace the
next(null, 0); line in the second function with the following line of code:
models.Comment.count({}, next);
Again, this was a piece of cake!
Now, the next two functions are going to be a little different, but they are almost
identical to each other. What we want to do with next is get the total views and
likes for every image. We can't use MongoDB's count method because that only
counts individual documents in a collection. We need to use MongoDB's aggregate
functionality instead.
[ 155 ]
Persisting Data with MongoDB
Using aggregate, we can perform a mathematical operation, such as $sum,
to tally results for us. Replace the next(null, 0); line in the third function
with the following code snippet:
models.Image.aggregate({ $group : {
_id : '1',
viewsTotal : { $sum : '$views' }
}}, function(err, result) {
var viewsTotal = 0;
if (result.length> 0) {
viewsTotal += result[0].viewsTotal;
next(null, viewsTotal);
Using MongoDB's aggregate function, we are telling MongoDB to group every
document together and sum up all of their views into a single new field called
viewsTotal. The resulting collection that is returned to the callback function is
an array of documents with the _id and viewsTotal fields. In this case, the results
array will only contain a single document with the grand total because we weren't
that tricky with our aggregate functionality. If there aren't any images in the
collection at all, we need to handle that and check accordingly. Finally, the next
callback function is called with the actual value for viewsTotal.
Let's use the same exact functionality to total up the likes for all images. Replace
the next(null, 0); line of code in the fourth and final function in parallel with the
following code snippet:
models.Image.aggregate({ $group : {
_id : '1',
likesTotal : { $sum : '$likes' }
}}, function (err, result) {
var likesTotal = 0;
if (result.length> 0) {
likesTotal += result[0].likesTotal;
next(null, likesTotal);
Now that the sidebar helper module has been updated and is complete with the
async.parallel functionality, let's make a minor tweak to our sidebar module to
ensure we are calling the Stats module correctly so that it's properly asynchronous.
The original line in helpers/sidebar.js was:
next(null, Stats());
[ 156 ]
Chapter 7
Replace that line of code with this slightly different version:
Last but not least, let's take care of the most popular helper module of the
images sidebar.
The popular images helper
Again, the original helpers/images.js file was mostly filled with fixture data
and placeholder code that's fairly useless. Let's replace the entire file with this new
version that's actually pretty tame in comparison to all of the other helper modules:
var models = require('../models');
module.exports = {
popular: function(callback) {
models.Image.find({}, {}, { limit: 9, sort: { likes: -1 }},
function(err, images) {
if (err) throw err;
callback(null, images);
At this point, this code should be pretty familiar to you. We just query MongoDB
and find the top nine most liked images by sorting the images by total such as
count in descending order and limiting the results to nine documents.
Let's edit the helpers/sidebar.js file again to include the updated call to the
Images.popular function. Consider the original code:
next(null, Images.popular());
Replace this with the following slightly newer version:
And now the sidebar is completely finished and completely dynamic. No more
fixture data or placeholder variables anywhere. Running the application should
yield a fully functional website with all of the features we set out to implement
working perfectly! Give it a spin and make sure it's working correctly.
[ 157 ]
Persisting Data with MongoDB
Iterating by adding an image removal
At this point, I think our application is pretty awesome, but there's something missing
that's nagging me. During testing, I've been creating all kinds of new images and
uploading them to the application but it's starting to get a bit cluttered and messy.
It dawned on me that the most obvious thing that's missing is the ability to delete
an image!
In reality, I left out this feature on purpose so that we could use this opportunity
to incorporate a completely new functionality that touches almost every area of
the application. This seemingly simple addition is actually going to require the
following changes:
• Update routes.js to include a new route to handle Delete requests
• Update controllers/image.js to include a new function for the route
This should not only remove the image from the database, but also delete the
file and all related comments
• Update the image.handlebars HTML template to include a Remove button
• Update the public/js/scripts.js file with an AJAX handler for the
Remove button
Adding a route
The first thing we need to update in order to add this new functionality is the main
routes list. Here, we will add a new endpoint that handles the delete function and
points to a function within the image controller. Edit the server/routes.js file and
insert the following new line of code:
router.delete('/images/:image_id', image.remove);
Adding a controller handler
Now that we have added a new route, we need to create the controller function that
it's using as its callback (image.remove). Edit controllers/image.js and add the
following new function code after the existing comment: function(req, res){}
operation (don't forget to add a trailing comma after the comment function since you
are adding a new function):
remove: function(req, res) {
Models.Image.findOne({ filename: { $regex: req.params.image_id }
[ 158 ]
Chapter 7
function(err, image) {
if (err) { throw err; }
fs.unlink(path.resolve('./public/upload/' + image.filename),
function(err) {
if (err) { throw err; }
Models.Comment.remove({ image_id: image._id},
function(err) {
image.remove(function(err) {
if (!err) {
} else {
This function performs four primary functions (and as such nests four layers deep
with callbacks—we could have used the async's series method here to prevent the
crazy amount of nesting). The first task is to find the image that we are attempting
to remove. Once that image is found, the file associated with the image should be
deleted. Next, find the comments associated with the image and remove them. Once
they have been removed, the last step is to remove the image itself. Assuming all of
that was a success, simply send a true Boolean JSON response back to the browser.
Updating the Handlebars image page template
Now that we have a route and controller function to support deleting an image,
we need a way for the UI to send the request. The most obvious solution is to add a
Delete button somewhere on the page. Edit the views/image.handlebars file and,
after the existing HTML, where we had the Like button, we are going to add new
HTML for a Delete button:
<div class="col-md-8">
<button class="btnbtn-success" id="btn-like" ...
// existing HTML for Like button and misc details
<div class="col-md-4 text-right">
<button class="btnbtn-danger" id="btn-delete" data-id="{{
image.uniqueId }}">
[ 159 ]
Persisting Data with MongoDB
<i class="fafa-times"></i>
Here, we just include new div that's set to four columns using Bootstrap and are
right aligned. The UI here is that the Like button and the stats are the leftmost
portion of the row, and the Delete button (an X icon from Font Awesome) is all
the way to the right of the same row (and red since we use Bootstrap's danger
color class).
Updating jQuery
Finally, we are going to tie it all together by implementing code similar to the Like
button, where we send an AJAX delete method to the server with the URL and the
image ID when the button is clicked on. To be safe, we display a standard JavaScript
confirmation dialog to ensure the button wasn't clicked on by accident.
Assuming the server responds with a true value, we will turn the button green and
change the icon to a checkmark with the word Deleted! in place. Edit public/js/
scripts.js and insert the following block of code after the existing code (be sure
to insert the new code inside the $(function(){ ... }) jQuery function):
$('#btn-delete').on('click', function(event) {
var $this = $(this);
var remove = confirm('Are you sure you want to delete this image?');
if (remove) {
var imgId = $(this).data('id');
url: '/images/' + imgId,
type: 'DELETE'
}).done(function(result) {
if (result) {
$this.append('<span> Deleted!</span>');
[ 160 ]
Chapter 7
Let's test this brand new functionality by launching the application, loading it up
in a browser, finding an image we no longer want, and viewing its image page.
The Delete button should now show up in place.
Refactoring and improvements
At this point, the application that we've been building is pretty much complete!
Before we iterate anymore on the project and continue to build it out and make it
ready for production, we should probably consider some refactoring and/or general
improvements. Some areas that I would personally take a look at to refactor and/or
rewrite to improve the application's performance and overall sanity are as follows:
• I might rethink working directly with models so much within the controllers
and instead create a utility that I can wrap a lot of that noise and rely on
more basic CRUD calls to my models and only provide a callback to each.
This is most visible in the image controller with like, comment, and remove.
• There is literally no validation in the project that we wrote and that's mostly
for brevity. In reality, we should have included validation on any input fields
a user interfaces with. Validation should be provided both on the frontend
via jQuery or plain old vanilla JavaScript as well as on the backend with
Node. The validation should protect users from submitting invalid and/or
malicious code (that is XSS or Cross-Site Scripting).
• Right now, our application is open to the general public, which means
any visitor that comes along can upload images as well as delete them!
It would be fairly simple to include a user authentication process within
our application. Passport.js is a great third-party module to integrate user
authentication into Node.js applications.
• Instead of attaching images to comments for the purposes of the sidebar (the
newest comments), we should consider creating a more robust aggregate
query using MongoDB to retrieve a hybrid collection of comments that
includes the image provided directly from MongoDB.
This chapter was a monster but was also the last piece of the puzzle to complete
our app and have a fully dynamic database-driven Node.js app that uses MongoDB.
Congratulations on making it this far and sticking with it! You're well on your way
to being a true full-stack JavaScript developer.
In the next chapter, we'll step away from our application for a bit as we take a look at
working with REST APIs using Node.js.
[ 161 ]
Creating a RESTful API
Now that your application is complete and ready for the world to see, you could
start thinking of ways to make it more popular. What if you wanted to allow external
systems with access rights to your data in a way that they could mass produce inserts
to your website without the need for users to visit the actual website? One example
that comes to mind almost immediately is how users of another website, say www.
facebook.com, can upload an image to Facebook and have it automatically uploaded
to your website as well.
The only way to make a scenario like this possible is by providing an API to your data
and the code that gives external developers access to a suite of tools that will allow
them to perform actions without the need to interact with the actual web pages.
In this chapter, we will review the following topics:
• Introducing RESTful APIs
• Installing a few basic tools
• Creating a basic API server and sample JSON data
• Responding to GET requests
• Updating data with POST and PUT
• Removing data with DELETE
• Consuming external APIs from Node.js
[ 163 ]
Creating a RESTful API
What is an API?
An Application Programming Interface (API) is a set of tools that a system makes
available so that unrelated systems or software have the ability to interact with each
other. Typically, a developer uses an API when writing software that will interact
with a closed, external, software system. The external software system provides
an API as a standard set of tools that all developers can use. Many popular social
networking sites provide developers access to APIs to build tools to support those
sites. The most obvious examples are Facebook and Twitter. Both have a robust
API that provides developers with the ability to build plugins and work with data
directly, without them being granted full access as a general security precaution.
As you will see with this chapter, providing your own API is not only fairly simple,
but it also empowers you to provide your users with access to your data. You also
have the added peace of mind knowing that you are in complete control over what
level of access you can grant, what sets of data you can make read-only, as well as
what data can be inserted and updated.
What is a RESTful API?
Representational State Transfer (REST) is a fancy way of saying CRUD over HTTP.
What this means is, when you use a REST API, you have a uniform means to create,
read, and update data using simple HTTP URLs with a standard set of HTTP verbs.
The most basic form of a REST API will accept one of the HTTP verbs at a URL and
return some kind of data as a response.
Typically, a REST API GET request will always return some kind of data, such as
JSON, XML, HTML, or plain text. A POST or PUT request to a RESTful API URL will
accept data to create or update. The URL for a RESTful API is known as an endpoint,
and while working with these endpoints, it is typically said that you are consuming
them. The standard HTTP verbs used while interfacing with REST APIs include:
• GET: This retrieves data
• POST: This submits data for a new record
• PUT: This submits data to update an existing record
• PATCH: This submits a date to update only specific parts of an existing record
• DELETE: This deletes a specific record
[ 164 ]
Chapter 8
Typically, RESTful API endpoints are defined in a way that they mimic the data
models and have semantic URLs that are somewhat representative of the data
models. What this means is that to request a list of models, for example, you
would access an API endpoint of /models. Likewise, to retrieve a specific model
by its ID, you would include that in the endpoint URL via /models/:Id.
Some sample RESTful API endpoint URLs are as follows:
• GET http://myapi.com/v1/accounts: This returns a list of accounts
• GET http://myapi.com/v1/accounts/1: This returns a single account
by Id: 1
• POST http://myapi.com/v1/accounts: This creates a new account
(data submitted as a part of the request)
• PUT http://myapi.com/v1/accounts/1: This updates an existing
account by Id: 1 (data submitted as part of the request)
• GET http://myapi.com/v1/accounts/1/orders: This returns a list of
orders for account Id: 1
• GET http://myapi.com/v1/accounts/1/orders/21345: This returns the
details for a single order by Order Id: 21345 for account Id: 1
It's not a requirement that the URL endpoints match this pattern; it's just
common convention.
Introducing Postman REST Client
Before we get started, there are a few tools that will make life much easier when
you're working directly with APIs. The first of these tools is called Postman REST
Client, and it's a Google Chrome application that can run right in your browser or
as a standalone packaged application. Using this tool, you can easily make any kind
of request to any endpoint you want. The tool provides many useful and powerful
features that are very easy to use and, best of all, free!
Installation instructions
Postman REST Client can be installed in two different ways, but both require Google
Chrome to be installed and running on your system. The easiest way to install the
application is by visiting the Chrome Web Store at https://chrome.google.com/
[ 165 ]
Creating a RESTful API
Perform a search for Postman REST Client and multiple results will be returned.
There is the regular Postman REST Client that runs as an application built into your
browser, and then separate Postman REST Client (packaged app) that runs as a
standalone application on your system in its own dedicated window. Go ahead and
install your preference. If you install the application as the standalone packaged
app, an icon to launch it will be added to your dock or taskbar. If you installed it as
a regular browser app, you can launch it by opening a new tab in Google Chrome,
going to Apps, and finding the Postman REST Client icon.
After you've installed and launched the app, you should be presented with an output
similar to the following screenshot:
[ 166 ]
Chapter 8
A quick tour of Postman REST Client
Using Postman REST Client, we're able to submit REST API calls to any endpoint we
want, as well as modify the type of request. Then, we can have complete access to
the data that's returned from the API, as well as any errors that might have occurred.
To test an API call, enter the URL to your favorite website in the Enter request URL
here field and leave the dropdown next to it as GET. This will mimic a standard GET
request that your browser performs anytime you visit a website. Click on the blue
Send button. The request is made and the response is displayed at the bottom half
of the screen.
In the following screenshot, I sent a simple GET request to http://kroltech.com
and the HTML is returned as follows:
[ 167 ]
Creating a RESTful API
If we change this URL to that of the RSS feed URL for my website, you can see the
XML returned:
The XML view has a few more features as it exposes the sidebar to the right that
gives you a glimpse at the tree structure of the XML data. Not only that, you can
now see a history of the requests we've made so far along the left sidebar. This is
great when we're performing more advanced POST or PUT requests and don't
want to repeat the data setup for each request while testing an endpoint.
[ 168 ]
Chapter 8
Here is a sample API endpoint I submitted a GET request to that returns the JSON
data in its response:
A really nice thing about making API calls to endpoints that return JSON using
Postman Client is that it parses and displays the JSON in a very nicely formatted
way, and each node in the data is expandable and collapsible.
The app is very intuitive, so make sure you spend some time playing around and
experimenting with different types of calls to different URLs.
[ 169 ]
Creating a RESTful API
Using the JSONView Chrome extension
There is one other tool I want to talk to you about (while extremely minor) that is
actually a really big deal. The JSONView Chrome extension is a very small plugin
that will instantly convert any JSON you view directly via the browser into a more
usable JSON tree (exactly like Postman Client). Here is an example of pointing to a
URL that returns JSON from Chrome before JSONView is installed:
Here is that same URL after JSONView has been installed:
You should install the JSONView Google Chrome extension the same way you
installed Postman REST Client—access the Chrome Web Store and perform a
search for JSONView.
[ 170 ]
Chapter 8
Now that you have the tools to be able to easily work with and test API endpoints,
let's take a look at writing your own and handling the different request types.
Creating a basic API server
Let's create a super basic Node.js server using Express that we'll use to create our
own API. Then, we can send tests to the API using Postman REST Client to see how
it all works. In a new project workspace, first install the npm modules that we're
going to need in order to get our server up and running:
$ npm init
$ npm install --save express body-parser underscore
Now that the package.json file for this project has been initialized and the modules
installed, let's create a basic server file to bootstrap an Express server. Create a file
named server.js and insert the following block of code:
var express = require('express'),
bodyParser = require('body-parser'),
_ = require('underscore'),
json = require('./movies.json'),
app = express();
app.set('port', process.env.PORT || 3500);
var router = new express.Router();
// TO DO: Setup endpoints ...
app.use('/', router);
var server = app.listen(app.get('port'), function() {
console.log('Server up: http://localhost:' + app.get('port'));
Most of this should look familiar to you. In the server.js file, we are requiring the
express, body-parser, and underscore modules. We're also requiring a file named
movies.json, which we'll create next.
After our modules are required, we set up the standard configuration for an Express
server with the minimum amount of configuration needed to support an API server.
Notice that we didn't set up Handlebars as a view-rendering engine because we aren't
going to be rendering any HTML with this server, just pure JSON responses.
[ 171 ]
Creating a RESTful API
Creating sample JSON data
Let's create the sample movies.json file that will act as our temporary data store
(even though the API we build for the purpose of demonstration won't actually
persist data beyond the app's life cycle):
"Id": "1",
"Title": "Aliens",
"Director": "James Cameron",
"Year": "1986",
"Rating": "8.5"
"Id": "2",
"Title": "Big Trouble in Little China",
"Director": "John Carpenter",
"Year": "1986",
"Rating": "7.3"
"Id": "3",
"Title": "Killer Klowns from Outer Space",
"Director": "Stephen Chiodo",
"Year": "1988",
"Rating": "6.0"
"Id": "4",
"Title": "Heat",
"Director": "Michael Mann",
"Year": "1995",
"Rating": "8.3"
"Id": "5",
"Title": "The Raid: Redemption",
"Director": "Gareth Evans",
"Year": "2011",
"Rating": "7.6"
[ 172 ]
Chapter 8
This is just a really simple JSON list of a few of my favorite movies. Feel free to
populate it with whatever you like. Boot up the server to make sure you aren't
getting any errors (note that we haven't set up any routes yet, so it won't actually
do anything if you tried to load it via a browser):
$ node server.js
Server up: http://localhost:3500
Responding to GET requests
Adding a simple GET request support is fairly simple, and you've seen this before
already in the app we built. Here is some sample code that responds to a GET request
and returns a simple JavaScript object as JSON. Insert the following code in the routes
section where we have the // TO DO: Setup endpoints ... waiting comment:
router.get('/test', function(req, res) {
var data = {
name: 'Jason Krol',
website: 'http://kroltech.com'
Just like we set up viewModel in Chapter 5, Templating with Handlebars, we create a
basic JavaScript object that we can then send directly as a JSON response using res.
json instead of res.render. Let's tweak the function a little bit and change it so that it
responds to a GET request against the root URL (that is /) route and returns the JSON
data from our movies file. Add this new route after the /test route added previously:
router.get('/', function(req, res) {
The res (response) object in Express has a few different methods
to send data back to the browser. Each of these ultimately fall back
on the base send method, which includes header information,
statusCodes, and so on. res.json and res.jsonp will
automatically format JavaScript objects into JSON and then send
using res.send. res.render will render a template view as a
string and then send it using res.send as well.
[ 173 ]
Creating a RESTful API
With that code in place, if we launch the server.js file, the server will be listening
for a GET request to the / URL route and will respond with the JSON data of our
movies collection. Let's first test it out using the Postman REST Client tool:
[ 174 ]
Chapter 8
GET requests are nice because we could have just as easily pulled that same URL via
our browser and received the same result:
However, we're going to use Postman for the remainder of our endpoint testing as
it's a little more difficult to send POST and PUT requests using a browser.
Receiving data – POST and PUT requests
When we want to allow those using our API to insert or update data, we need to
accept a request from a different HTTP verb. When inserting new data, the POST
verb is the preferred method to accept data and know it's for an insert. Let's take
a look at code that accepts a POST request and data along with the request, inserts
a record into our collection, and returns the updated JSON.
[ 175 ]
Creating a RESTful API
Insert the following block of code after the route you added previously for GET:
router.post('/', function(req, res) {
// insert the new item into the collection (validate first)
if(req.body.Id && req.body.Title && req.body.Director &&
req.body.Year && req.body.Rating) {
} else {
res.json(500, { error: 'There was an error!' });
The first thing we do in the POST function is check to make sure the required fields
were submitted along with the actual request. Assuming our data checks out and all
the required fields are accounted for (in our case, every field), we insert the entire req.
body object into the array as is using the array's push function. If any of the required
fields aren't submitted with the request, we return a 500 error message instead. Let's
submit a POST request this time to the same endpoint using the Postman REST Client.
(Don't forget to make sure your API server is running with node server.js.):
[ 176 ]
Chapter 8
First, we submitted a POST request with no data, so you can clearly see the 500 error
response that was returned:
[ 177 ]
Creating a RESTful API
Next, we provided the actual data using the x-www-form-urlencoded option in
Postman and provided each of the name/value pairs with some new custom data.
You can see from the results that STATUS was 200, which is a success and the
updated JSON data was returned as a result. Reloading the main GET endpoint
in a browser yields our original movies collection with the new one added.
PUT requests will work in almost exactly the same way except, traditionally, the
Id property of the data is handled a little differently. In our example, we are going
to require the Id attribute as a part of the URL and not accept it as a parameter in
the data that's submitted (since it's usually not common for an update function to
change the actual Id of the object it's updating). Insert the following code for the
PUT route after the existing POST route you added earlier:
router.put('/:id', function(req, res) {
// update the item in the collection
if(req.params.id && req.body.Title && req.body.Director &&
req.body.Year && req.body.Rating) {
_.each(json, function(elem, index) {
// find and update:
[ 178 ]
Chapter 8
if (elem.Id === req.params.id) {
elem.Title = req.body.Title;
elem.Director = req.body.Director;
elem.Year = req.body.Year;
elem.Rating = req.body.Rating;
} else {
res.json(500, { error: 'There was an error!' });
This code again validates that the required fields are included with the data that
was submitted along with the request. Then, it performs an _.each loop (using the
underscore module) to look through the collection of movies and find the one whose
Id parameter matches that of the Id included in the URL parameter. Assuming there's
a match, the individual fields for that corresponding object are updated with the new
values that were sent with the request. Once the loop is complete, the updated JSON
data is sent back as the response. Similarly, in the POST request, if any of the required
fields are missing, a simple 500 error message is returned. The following screenshot
demonstrates a successful PUT request updating an existing record:
[ 179 ]
Creating a RESTful API
The response from Postman after including the value 1 in the URL as the Id parameter,
which provides the individual fields to update as x-www-form-urlencoded values,
and finally sending as PUT shows that the original item in our movies collection is
now the original Alien (not Aliens its sequel, as we originally had).
Removing data – DELETE
The final stop on our whirlwind tour of the different REST API HTTP verbs is
DELETE. It should be no surprise that sending a DELETE request should do exactly
what it sounds like. Let's add another route that accepts DELETE requests and deletes
an item from our movies collection. Here is the code that takes care of the DELETE
requests that should be placed after the existing block of code from the previous PUT:
router.delete('/:id', function(req, res) {
var indexToDel = -1;
_.each(json, function(elem, index) {
if (elem.Id === req.params.id) {
indexToDel = index;
if (~indexToDel) {
json.splice(indexToDel, 1);
This code will loop through the collection of movies and find a matching item by
comparing the values of Id. If a match is found, the array index for the matched
item is held until the loop is finished. Using the array.splice function, we can
remove an array item at a specific index. Once the data has been updated by
removing the requested item, the JSON data is returned. Notice in the following
screenshot that the updated JSON that's returned is in fact no longer displaying
the original second item we deleted.
Note that ~ use in JavaScript! That's a little bit of JavaScript black magic!
The tilde (~) in JavaScript will bit flip a value. In other words, take a
value and return the negative of that value incremented by one, that is ~n
=== -(n+1). Typically, the tilde is used with functions that return -1 as
a false response. By using ~ on -1, you are converting it to 0. If you were
to perform a Boolean check on -1 in JavaScript, it would return true.
You will see that ~ is used primarily with the indexOf function and
jQuery's $.inArray(); both return -1 as a false response.
[ 180 ]
Chapter 8
All of the endpoints defined in this chapter are extremely rudimentary, and most
of these should never see the light of day in a production environment! Whenever
you have an API that accepts anything other than GET requests, you need to be sure
to enforce extremely strict validation and authentication rules. After all, you are
basically giving your users direct access to your data.
Consuming external APIs from Node.js
There will undoubtedly be a time when you want to consume an API directly from
within your Node.js code. Perhaps your own API endpoint needs to first fetch data
from some other unrelated third-party API before sending a response. Whatever
the reason, the act of sending a request to an external API endpoint and receiving a
response can be done fairly easily using a popular and well-known npm module called
request. The request module was written by Mikeal Rogers and is currently the third
most popular (and most relied upon) npm module after async and underscore.
[ 181 ]
Creating a RESTful API
Request is basically a super simple HTTP client, so everything you've been doing
with Postman REST Client so far is basically what Request can do, only the resulting
data is available to you in your Node code as well as the response status codes and/
or errors, if any.
Consuming an API endpoint using request
Let's do a neat trick and actually consume our own endpoint as if it was some
third-party external API. First, we need to ensure we have request installed
and can include it in our app:
$ npm install --save request
Next, edit server.js and make sure you include request as a required module
at the start of the file:
var express = require('express'),
bodyParser = require('body-parser'),
_ = require('underscore'),
json = require('./movies.json'),
app = express(),
request = require('request');
Now, let's add a new endpoint after our existing routes, which will be an endpoint
accessible in our server via a GET request to /external-api. This endpoint, however,
will actually consume another endpoint on another server, but for the purposes of
this example, the other server is actually the same server we're currently running!
The request module accepts an options object with a number of different
parameters and settings, but for this particular example, we only care about a few.
We're going to pass an object that has a setting for the method (GET, POST, PUT, and
so on) and the URL of the endpoint we want to consume. After the request is made
and a response is received, we want an inline callback function to execute.
Place the following block of code after your existing list of routes in server.js:
router.get('/external-api', function(req, res) {
method: 'GET',
uri: 'http://localhost:' + (process.env.PORT || 3500),
}, function(error, response, body) {
if (error) { throw error; }
var movies = [];
_.each(JSON.parse(body), function(elem, index) {
[ 182 ]
Chapter 8
Title: elem.Title,
Rating: elem.Rating
res.json(_.sortBy(movies, 'Rating').reverse());
The callback function accepts three parameters: error, response, and body.
The response object is like any other response that Express handles and has all
of the various parameters as such. The third parameter, body, is what we're really
interested in. That will contain the actual result of the request to the endpoint that
we called. In this case, it is the JSON data from our main GET route we defined earlier
that returns our own list of movies. It's important to note that the data returned from
the request is returned as a string. We need to use JSON.parse to convert that string
to actual usable JSON data.
We manipulate the data that came back from request to suit our needs. In this example,
we took the master list of movies and just returned a new collection that consists of
only the Title and Rating of each movie and then sorts the results by the top scores.
Load this new endpoint by pointing your browser to http://localhost:3500/
external-api, and you can see the new transformed JSON output to the screen.
Let's take a look at another example that's a little more real world. Let's say that we
want to display a list of similar movies for each one in our collection, but we want
to look up that data somewhere, such as on www.imdb.com. Here is the sample code
that will send a GET request to IMDB's JSON API, specifically for the word aliens,
and returns a list of related movies by the Title and Year. Go ahead and place this
block of code after the previous route for external-api:
router.get('/imdb', function(req, res) {
method: 'GET',
uri: 'http://sg.mediaimdb.com/suggests/a/aliens.json',
}, function(err, response, body) {
var data = body.substring(body.indexOf('(')+1);
data = JSON.parse(data.substring(0,data.length-1));
var related = [];
_.each(data.d, function(movie, index) {
Title: movie.l,
Year: movie.y,
[ 183 ]
Creating a RESTful API
Poster: movie.i ? movie.i[0] : ''
If we take a look at this new endpoint in a browser, we can see the JSON data that's
returned from our /imdb endpoint is actually itself retrieving and returning data
from some other API endpoint:
Note that the JSON endpoint I am using for IMDB isn't actually from
their API, but rather what they use on their homepage when you type in
the main search box. This would not really be the most appropriate way
to use their data, but it's more of a hack to show this example. In reality,
to use their API (like most other APIs), you would need to register and
get an API key that you would use so that they can properly track how
much data you are requesting on a daily or hourly basis. Most APIs will
require you to use a private key with them for this reason.
[ 184 ]
Chapter 8
In this chapter, we took a brief look at how APIs work in general, the RESTful API
approach to semantic URL paths and arguments, and created a bare bones API.
We used Postman REST Client to interact with the API by consuming endpoints and
testing the different types of request methods (GET, POST, PUT, and so on). You also
learned how to consume an external API endpoint by using the third-party node
module request.
In the next chapter, we will revisit our original application as we implement
best practices by introducing testing in Node.js. We'll take a look at popular
testing frameworks and write tests for the application to prove that our code
works as expected.
[ 185 ]
Testing Your Code
Up until this point, we've been pretty much flying by the seat of our pants when
it comes to the code we've been writing! We've literally had no way of knowing
whether the code worked until we tested it out in an actual browser.
In this chapter, we will cover the following topics:
• Running tests with the Mocha test framework
• Writing tests with the Chai assertion library
• Spies and stubs with Sinon and Proxyquire
• Writing your first test
• Testing our application
Tests are great for making sure your code functions properly, but they're also
awesome for preventing new unexpected bugs from suddenly popping up
because of an innocent little change you made to some unsuspecting code.
The tools of the trade
Let's start by taking a look at the various tools and libraries we're going to be using
to run and write our tests. There are three main concepts we need to cover before
we can actually start writing real tests.
The first is a test runner or the framework we use to run our suite of tests.
Most of the frameworks follow Test Driven Development (TDD) and its process
relies on the following steps:
1. It defines a unit test
2. Implements the unit
3. Executes the test and verifies that the test passes
[ 187 ]
Testing Your Code
The second concept is the assertion library itself—the language we use to write our
tests. A special version of using assertion language, to design and build pieces of
functionality incrementally guided by an expected behavior is Behavior Driven
Development (BDD).
For both TDD and BDD, we can use the Mocha testing framework; however, we will
be writing assertions using a special assertion library called Chai.js
Finally, we'll take a look at the idea of spies and stubs, which are fake representatives
of certain parts of our code that are relied on when we need to track function calls to
ensure an expected behavior.
Running tests with the Mocha framework
When writing tests for an application, you typically write them in batches that are
module specific. These batches are referred to as suites or specs. Each suite typically
contains a batch of tests organized in a way that almost mirrors the application itself.
With Node, the idea is no different in that each suite of tests we write will be specific
to an individual module. You'll require the module you want to test against, and
write a collection of tests for each part of the module's functionality.
Since you'll have many different test files testing each and every component of your
application, you'll want a way to quickly execute all of the tests. This is where the
test runner comes in. The test runner that we've decided to use is called Mocha.
You can install Mocha globally like any other npm package:
$ npm install –g mocha
You might require security privileges while installing on Linux or OS X, which can
done simply using sudo before npm.
Once installed, the Mocha command line tool is now available. Simply executing
mocha from a command line will execute the test run with a few default options.
The test runner will look for a folder named test and any .js file within. In
our case, we haven't actually set up any tests yet so executing mocha alone won't
accomplish anything; instead, it will throw an error of cannot resolve path.
When the Mocha test runner does find .js files, it executes them like any other Node
file except it looks for a few specific keywords within the file.
[ 188 ]
Chapter 9
Here is some sample code for a typical test block:
var expect = require('chai').expect;
describe('The code', function() {
// optional preparation for each test
// optional cleanup after each test
it('should test something', function(){
var something = 1;
// here we "expect" some condition to declare our test
// in this case, we expect the variable to exist
// more on the assertion syntax a little later
it('should test something_else', function(){
var something_else = false;
// now we test a different variable against its value
// and expect that value to equal false
The first thing Mocha will scan the file for is a describe block. A describe block
is a way to define a specific group of test cases in single line. You can have many
describe blocks in a test file, and each describe block can have many specific
tests. In addition, describe blocks can be nested as deep as you like to better
organize your tests.
Once a describe block is found, a few other items are executed within it.
A beforeEach and afterEach block is checked for to see whether there is any
pretest work that needs to be executed before each test is executed. Likewise, any
cleanup that needs to occur between tests can be taken care of within the afterEach
block. Both of these blocks are optional and therefore not required. A good example
of when you would want to use a beforeEach block is if you need to instantiate an
object that you will be testing, you would want to create a new instance before every
single test. This way, whatever changes a test might push to the object will be reset
and will not inadvertently affect any other tests. Likewise, any changes you've made
during a test to any other related objects can be reset during an afterEach block.
[ 189 ]
Testing Your Code
Within the describe block, defining individual tests is done with it statements.
Within each it statement, it's generally considered good practice to include a single
expect to assert the actual test (although you can include as many expect function
calls as you like, its still only considered a single test because of the single it).
We're using the BDD style syntax when writing our suites, which allows our tests
to read like user stories. With the preceding example, you can read the tests as The
code should test something and The code should test something_else.
In fact, if we ran the previous tests, we would see the following output:
The code
should test something
should test something_else
2 passing (5ms)
Asserting tests with Chai.js
As you saw in the previous example, we used special blocks to define our test groups
with Mocha but we used a separate language when defining our actual individual
tests. These tests are called assertions and we chose to use the Chai.js library. This is
purely a personal preference, as there are a number of different assertion libraries that
exist. Each library does basically the same thing with slight variations on the syntax
and style of actually writing the tests. Since Chai.js is project specific and based on
a personal preference, we're going to install it as a project dependency (instead of
globally). In addition, as our tests are not actually required for our application to run,
we'll include Chai.js under the devDependencies list in our package.json file).
Performing npm install on the developer 's machine will yield all packages under
normal dependencies as well as devDependencies from package.json. When the
environment changes to production, we need to execute npm install --prod to
specify the environment.
This will help npm to install packages under dependencies and not devDependencies
from package.json.
In order to include Chai.js in our project as a devDependency, we will use the
--save-dev flag instead of --save when performing the npm install:
$ npm install --save-dev chai
[ 190 ]
Chapter 9
Getting started with Chai
Chai itself has a few different flavors of API styles that can be used when writing
tests. The BDD API, which is what we will use for the tests we write, uses expect
and should. There's also the assert API, which is more of a TDD style. The benefit
of using the BDD style with expect/should is that you can chain the assertion
methods to improve readability of the tests.
You can learn more about BDD as well as TDD by accessing the following
Wikipedia page:
Using the BDD assertion API with Chai.js provides a number of methods at
our disposal such as to, be, is, and many more. They have no testing capability,
but instead they improve the assertion readability. All the getters are listed at
All of these getters will follow an expect() statement and can be coupled with not,
in case we want to flip the assertion negatively.
The preceding getters are combined with Chai assertion methods such as ok, equal,
within, and so on, to determine a test's outcome. All those methods are listed at
Let's start constructing simple assertions. Chai provides three different assertion
styles expect, should, and assert. Consider some simple illustrations as follows:
{ pets: [ 'dog', 'cat', 'mouse' ] };
assert.equal(foo, 'bar', 'Foo equal bar');
As you can see, the expect/should function is based on self-descriptive language
chains. Both differ in a way of declaration, the expect function provides a starting
point for a chain whereas the should interface extends a Object.prototype.
[ 191 ]
Testing Your Code
The assert interface provides simple but powerful TDD style assertions. Apart from
the preceding example that yields a deep equality assertion, exception testing, and
instance of are also available.
For a more in-depth learning, refer to the Chai documentation at
Spies and stubs with Sinon.js
Testing your code will prove to be extremely difficult if there isn't an easy way to spy
on functions and know whenever they are called. Additionally, when one of your
functions is called, it will be nice to know what arguments were passed to it and
what was returned. A spy in testing is a special placeholder function that replaces an
existing function when you want to check specifically if/when it was called. Spies
track a number of properties for a function when it's called, and they can also pass
through to the expected functionality of the original function. The Sinon.js library
provides both spy and stub functionalities and is quite extensive. For a complete list
of the different options available with this powerful framework, I strongly recommend
you spend some time reading the documentation at http://sinonjs.org/docs.
Since we will be using Sinon.js with our tests, we should install it as another
devDependency exactly the same way we did with Chai.js. In addition, we should
also install the sinon-chai helper, which provides additional Chai assertion verbs
specifically for use with Sinon:
$ npm install --save-dev sinon sinon-chai
The inclusion of sinon-chai allows us to write special assertions such as to.be.
calledWith, which would otherwise not work with Chai alone.
Imagine you have a function that simply adds two numbers together and returns
the sum:
var sum = function(a, b) {
return a + b;
var doWork = function() {
var x = 1,
y = 2;
[ 192 ]
Chapter 9
When writing a test for the doWork function, we want to assert that the sum function
was called. We don't necessarily care what the function does or whether it even
works; we just want to make sure since doWork relies on sum that it's actually calling
the function() function. In this scenario, the only way we could be sure is if we
had a way to spy on the sum function and know if it was ever called. Using spy, we
can do just that:
var sinon = require("sinon");
var sinonChai = require("sinon-chai");
describe('doWork', function() {
var sum;
it('should call sum', function() {
sum = sinon.spy();
In the preceding scenario, the sum function is replaced with a spy function. So
its actual functionalities will no longer exist. If we want to ensure that the sum
function is not only spied on but still functions the way we expect, we need to
attach .andCallThrough() after sinon.spy():
describe('doWork', function() {
var sum;
console.log = sinon.spy();
it('should call sum', function() {
sum = sinon.spy().andCallThrough();
Notice that by including andCallThrough on our sum spy, we're able to not only
spy on it and assert that it was called but also spy on the console.log function
and assert that it was called with the correct value returned by sum.
[ 193 ]
Testing Your Code
Where a spy is typically just a watcher to a function and only reports whether
the function was called, a stub allows you to provide custom functionalities for a
function on the fly during test execution. Test stubs are said to be preprogrammed
behavioral functions that are used to test wrapped boilerplate code required as a
module dependency in an application.
Think of stub as a super spy, where it reports the same things that a spy does, but
also performs whatever specific tasks you want as well. Using the same example,
let's stub the sum function to always return the same value:
it('should console.log sum response', function(){
// replace the existing sum function with a new stub,
// a generic function that does exactly what we specify
// in this case always just return the number 2
sum = sinon.stub(function(){
return 2;
// lets replace the standard console.log function
// with a spy
console.log = sinon.spy();
// call our doWork function (which itself uses console.log)
// and if doWork executed the way its supposed to, console.log
// should have been called and the parameter 2 passed to it
Stubbing a function is great when a function performs work that might yield
unexpected results, and you just want to force the response for the purposes of your
test. Stubbing is also handy when you're doing TDD and you're testing against
a function that you haven't even written yet.
Stubbing node modules with Proxyquire
Spies and stubs are great when writing tests against code within the same module,
but when you need to spy on or stub a module required within another node
module, things get a little trickier. Fortunately, there's a tool called Proxyquire that
will allow you to stub modules that are required from your code.
[ 194 ]
Chapter 9
Examine the following code sample:
// google.js
var request = require('request');
module.exports = function() {
request('http://www.google.com', function (err, res, body) {
You can see that we require the request module. The request module accepts
two parameters, the second of which is a callback function. This is where things
start to get tricky. How are we going to implement spies and/or stubs in this type
of scenario? Furthermore, how can we prevent our tests from explicitly making
a network call to fetch google.com? What if google.com is down (ha!) when we
run our tests?
In order to be able to spy on the request module, we need a way to intercept actual
require and attach our own stubbed version of request instead. The request
module is actually a great example of a module that you would want to stub because
request is used to make a network call, and that's something that you want to make
sure your tests never actually do. You don't want your tests relying on an external
resource like a network connection or being dependent on the data returned from a
live request.
Using Proxyquire, we can actually set up our tests in a way that they'll intercept
the require module and replace what gets executed with our own stub. Here's an
example of a test file written against the module we created earlier:
var log = sinon.spy(),
requestStub = sinon.stub().callsArgWith(1, null, null,
google = proxyquire('../google', { 'request': requestStub });
describe('google module', function(){
beforeEach(function() {
it('should request google.com', function() {
it('should log google body', function(){
[ 195 ]
Testing Your Code
expect(callback).to.be.calledWith(null, null,
The first thing the test suite does is set up a spy and generic stub function that
will be used as the request module. Then, we include our google module but
we include it using proxyquire instead of a typical require module. Using
proxyquire, we pass the path to the module the same way we would with require,
except the second parameter is the module that would be required within that
module and the stub function to use in its place.
Before each test, we execute the original google module and assert against our stub
that it was in fact called. Additionally, we assert that the log spy was called with
whatever data was returned from the request module. Since we are in control of
that module, we can test, quite literally, that the string google.com was returned
when a request was made to http://google.com (which we know for a fact is not
true—not only that, but we know that a network call was never sent to www.google.
com either).
We're using a special power of a stub that allows us to execute a particular parameter
to the stubbed function assuming it was a callback function. Here, we're using
callsArgWith and including the argument index (zero based) as the first parameter;
in this case, 1. Of the two parameters that were passed to the request, the first (index
0) was the URL itself and the second (index 1) was the callback function. By using
callsArgWith, we can execute the callback function and specifically provide its
parameters; in this case, null, null, and a string.
Like Sinon.js and Chai.js, Proxyquire will also need to be included in our project as
$ npm install --save-dev proxyquire
Writing and running your first test
Up to this point, all of the test code we've seen has just been demos and examples
and we haven't actually run any tests. Let's set up the basic structure of our
application so that we can start writing real tests.
[ 196 ]
Chapter 9
The first thing to do is set up the folder structure that will house all of our tests.
Consider the following steps:
1. Within the root of the application project folder, create a folder named tests.
2. Within the tests folder, create three more folders for controllers, models,
and server:
/(existing app root)
Writing a test helper
Before we start writing the tests for our application, there's a small amount of
overhead we need to take care of to prepare for our tests. To take care of this
overhead, we're going to write a testhelper file that will be included and run with
every test file we execute via Mocha.
Create a file named testhelper.js within the tests folder and insert the following
block of code:
var chai = require('chai'),
sinon = require('sinon'),
sinonChai = require('sinon-chai');
global.expect = chai.expect;
global.sinon = sinon;
This is code that we would typically need to include at the top of every one of our
test files, but by including it in a single file we can instruct Mocha to automatically
require this file for every test file that is run. The file itself just includes the chai and
sinon modules and defines a few global variables as shortcuts for our test writing.
Additionally, it instructs chai to use the sinonChai module so that our syntax is
extended and we can write Sinon-specific Chai assertions.
The command to actually run our suite of tests is:
$ mocha -r tests/testhelper.js -R spec tests/**/*.test.js
[ 197 ]
Testing Your Code
Remember that we installed Mocha globally earlier so that we can execute
the mocha command from anywhere.
Based on the path to our tests in the preceding command, it's assumed that the
command will be executed from the root of the application project folder. The –r flag
instructs Mocha to require the testhelper.js module. The –R flag is an option to
define the style of the test reporting output. We chose to use the spec style, which
lists our report in a nested indentation style with each describe and it statement
along with a green checkmark for the passed tests. Finally, the last argument is the
path to our test files; in this case, we provided wildcards so that all of our tests will
be run.
Mocha has a few different reporting styles that you can choose from.
These include dot (repeating dots for each test), list, progress (a
percentage bar), JSON and spec. One of the more interesting, albeit
somewhat useless, is the –R nyan reporting style.
Let's write a quick sample test to make sure our project is properly set up.
Within the tests folder, create a new file named mocha.test.js and include
the following code:
describe('Mocha', function() {
describe('First Test', function() {
it('should assert 1 equals 1', function() {
The preceding test is pretty straightforward and simply asserts that 1 is equal to
1. Save this file and run the Mocha test command again, and you should get the
following output:
$ mocha -r tests/testhelper.js -R spec tests/mocha.test.js
First Test
should assert 1 equals 1
1 passing (5ms)
[ 198 ]
Chapter 9
You might find it tiresome and frustrating to remember and execute that long
convoluted command for Mocha. Fortunately, there's a pretty easy solution. Edit the
package.json file in the application and add the following section:
"scripts": {
"start": "node server.js",
"test": "mocha -r tests/testhelper.js -R spec
By making this tweak in the package.json file, you can now simply execute
npm test from a command line as a quick and easy shortcut. This is a standard
convention with the package.json file so any developer will know to simply
execute npm test as follows:
$ npm test
> chapter9@0.0.0 test /Users/jasonk/repos/nodebook/chapter9
> mocha -r tests/testhelper.js -R spec tests/**/*.test.js
First Test
should assert 1 equals 1
1 passing (5ms)
Now that your project is set up to properly run and execute tests, let's start writing
some real tests for the application.
Testing the application
With all of that background information out of the way, let's focus on writing some
real tests for the application we've built. In the following sections, we will write tests
for the routes, servers, models, and controllers in our application.
Testing the routes
Let's start things a little slow by taking a look at one of the most basic files in our
application, the routes.js file. This file simply defines the number of routes that
the application should respond to. This is going to be one of the easiest files to write
tests for.
[ 199 ]
Testing Your Code
Since the routes.js file is in the server folder within our main application, let's
put its corresponding test file in a similar location. Within the tests/server folder,
create a file named routes.test.js. Since the routes.test.js file is going to
be testing the functionalities of our routes.js file, we need it to require the same
modules. Include the following code in test/server/routes.test.js:
var home = require('../../controllers/home'),
image = require('../../controllers/image'),
routes = require('../../server/routes');
Notice that the paths are different since we require modules from within our
test/server folder, but we also require app-specific modules. Also, note that in
addition to the modules that our original routes.js file requires, we also require
the routes module itself. How else are we going to be able to test the functionalities
of the module if it isn't included? Next, let's set up the structure of the test suite and
create a few spies. Include this new block of code following the previous code within
describe('Routes', function(){
var app = {
get: sinon.spy(),
post: sinon.spy(),
delete: sinon.spy()
// to do: write tests...
If you recall, the routes module's initialize function accepted a single parameter,
an app object. In our tests, we defined app as a simple anonymous object with three
functions such as get, post, and delete, each of which is a spy. We include a
beforeEach block to execute the initialize function before every one of our test
Now let's include some tests. First, we'll test that the GET endpoints are configured
correctly. Immediately after the // to do: write tests... comment, place the
following block of code:
describe('GETs', function() {
it('should handle /', function(){
expect(app.get).to.be.calledWith('/', home.index);
[ 200 ]
Chapter 9
it('should handle /images/:image_id', function(){
Then, test the POST endpoints:
describe('POSTs', function() {
it('should handle /images', function(){
it('should handle /images/:image_id/like', function(){
it('should handle /images/:image_id/comment', function(){
Finally, test the DELETE endpoint:
describe('DELETEs', function() {
it('should handle /images/:image_id', function(){
.to.be.calledWith('/images/:image_id', image.remove);
Each of these tests assert the same thing, that the app object's corresponding
the get, post, or delete function was executed with the correct parameters for
each route. We were able to test against the parameters because the app object
we used was a spy.
If you run the mocha command to execute the suite of tests, you should see the
following output:
$ npm test
should handle /
[ 201 ]
Testing Your Code
should handle /images/:image_id
should handle /images
should handle /images/:image_id/like
should handle /images/:image_id/comment
should handle /images/:image_id
6 passing (14ms)
Testing the server
Testing the server.js file will be slightly different than any of our other files. The
file runs as the root of our application, so it doesn't export a module or any object
that we can directly test. Since we launch our server using server.js, we need to
emulate launching our server from our code. We'll create a function called server,
which will require the server.js file using Proxyquire, and stub each of the
modules that it requires itself. Executing the server() function will be exactly the
same as executing node server.js from a command line. All of the code within the
file will execute via that function, and then we can test against each of the calls that
are made using stubs from within Proxyquire.
Create a file named server.test.js within the tests/server/ folder and insert
the following block of code:
var proxyquire, expressStub, configStub, mongooseStub, app,
server = function() {
proxyquire('../../server', {
'express': expressStub,
'./server/configure': configStub,
'mongoose': mongooseStub
describe('Server', function() {
proxyquire = require('proxyquire'),
app = {
set: sinon.spy(),
get: sinon.stub().returns(3300),
listen: sinon.spy()
[ 202 ]
Chapter 9
expressStub = sinon.stub().returns(app),
configStub = sinon.stub().returns(app),
mongooseStub = {
connect: sinon.spy(),
connection: {
on: sinon.spy()
delete process.env.PORT;
// to do: write tests...
Before each test is run for our server, we reset the stubs for all of the major
components of the server. These stubs include the app object itself, express, config,
and mongoose. We're stubbing each of these modules since we want to spy on them
(and we use a stub because some of them need to return objects that we'll work with
in our file). Now that we have all of our spies in place and our app object scaffold set
up, we can start testing the main functionalities of our code.
We need to check whether the following conditions pass:
• An application is created
• The views directory is set
• The port is set and can be configured and/or set to default
• The app itself is configured (config is called with it)
• Mongoose connects to a database URI string
• Finally, the app itself is launched
Replace the // to do: write tests... comment in the earlier code with the
following block of code:
describe('Bootstrapping', function(){
it('should create the app', function(){
it('should set the views', function(){
[ 203 ]
Testing Your Code
it('should configure the app', function(){
it('should connect with mongoose', function(){
it('should launch the app', function(){
In the preceding group of tests, we are testing the bootstrapping of our server, which
is all of the functionalities that initially run within server.js. The names of the tests
are pretty self-explanatory. We're checking against the various methods of the app
object, ensuring that they're called and/or the correct parameters were passed in.
For the tests, we want to test that a specific type of parameter was called, not literally
what the parameter value was; we use Sinon's match element, which allows our tests
to be a little more generic. We wouldn't want to hard code the MongoDB URI string
in our tests because that's just another place we would have to maintain—although
you could very well do this if you wanted your test to be that strict (that is to assert
that quite literally the exact URI string was passed).
In the second set of tests, we want to ensure that the port is set, that it defaults to
3300, and that it can be changed via the use of a node environment variable:
describe('Port', function(){
it('should be set', function() {
it('should default to 3300', function() {
it('should be configurable', function() {
process.env.PORT = '5500';
[ 204 ]
Chapter 9
With these tests in place, run the npm test command again and you should get the
following output:
$ npm test
should create the app (364ms)
should set the views
should configure the app
should connect with mongoose
should launch the app
should be set
should default to 3300
should be configurable
Testing a model
When testing our models, we want to include the model module itself and then write
tests against it. The easiest solution here is that we create a test model object and then
assert that the model has all of the fields that we expect, as well as any virtuals we
might have created.
Create the tests/models/image.test.js file and insert the following code:
var ImageModel = require('../../models/image');
describe('Image Model', function() {
var image;
it('should have a mongoose schema', function(){
image = new ImageModel({
title: 'Test',
description: 'Testing',
filename: 'testfile.jpg'
[ 205 ]
Testing Your Code
// to do: write tests...
First, we include ImageModel using require (note the path for the require
statement). The very first test we run is to make sure that ImageModel has a
mongoose schema property. After this test, we define the beforeEach block that
we'll rely on for the remainder of our tests. Before every test, we want to instantiate
a new ImageModel object that we can test against. We do this in a beforeEach
block so that we're sure we're dealing with a fresh object in every test and that it
hasn't been tainted by any tests that were previously run. It's also important to note
that the order of the first test and the beforeEach block doesn't actually matter, as
the beforeEach block will run before every test in its parent describe function
regardless of the order it was defined in.
Include the following suite of tests replacing the placeholder // to do: write
tests... comment:
describe('Schema', function() {
it('should have a title string', function(){
it('should have a description string', function(){
it('should have a filename string', function(){
it('should have a views number default to 0', function(){
it('should have a likes number default to 0', function(){
it('should have a timestamp date', function(){
[ 206 ]
Chapter 9
Here, we check to ensure that each property we expect an ImageModel instance
to have is defined. For the properties that have default values set, we also check
to ensure the default values are set as well.
Next, we test against the virtuals we expect ImageModel to have, and verify that
they function the way they're supposed to:
describe('Virtuals', function(){
describe('uniqueId', function(){
it('should be defined', function(){
it('should get filename without extension', function(){
When testing the uniqueId virtual, it should return the image model's filename
without the extension. As the beforeEach defined our image model with a filename
of testfile.jpg, we can assert with our test that the uniqueId returned is equal
to testfile (the filename without the extension).
Running the tests for our model should output the following results:
$ npm test
Image Model
should have a mongoose schema
should have a title string
should have a description string
should have a filename string
should have a views number default to 0
should have a likes number default to 0
should have a timestamp date
should be defined
should get filename without extension
[ 207 ]
Testing Your Code
Testing a controller
Last but not least, let's take a look at the image controller, and specifically test for the
main index function. Because the index function does a lot of work and performs a
number of different tasks, the test file will make extensive use of stubs and spies. The
first thing we need to do before any tests is declare a number of global variables for
our tests as well as set up all of our stubs, spies, and placeholder objects for use with
Proxyquire. Then, we require the actual image controller using Proxyquire. Create a
file named tests/controllers/image.test.js and insert the following code:
var proxyquire = require('proxyquire'),
callback = sinon.spy(),
sidebarStub = sinon.stub(),
fsStub = {},
pathStub = {},
md5Stub = {},
ModelsStub = {
Image: {
findOne: sinon.spy()
Comment: {
find: sinon.spy()
image = proxyquire('../../controllers/image', {
'../helpers/sidebar': sidebarStub,
'../models': ModelsStub,
'fs': fsStub,
'path': pathStub,
'md5': md5Stub
res = {},
req = {},
testImage = {};
With this code, we define a number of global variables as spies, stubs, or empty
placeholder JavaScript objects. Once our stubs are prepared, we call proxyquire to
include our image controller (ensuring that the required modules within the image
controller are actually replaced with our various stubs and spies). Now that all of
our globals, stubs, and spies are prepared, let's include some tests.
[ 208 ]
Chapter 9
Include the following code after the previous block of code:
describe('Image Controller', function(){
beforeEach(function() {
res = {
render: sinon.spy(),
json: sinon.spy(),
redirect: sinon.spy()
req.params = {
image_id: 'testing'
testImage = {
_id: 1,
title: 'Test Image',
views: 0,
likes: 0,
save: sinon.spy()
// to do: write tests...
Once again, we build up some setup using a beforeEach block for our tests. This
sets spies on each of the res object's functions including render, JSON and redirect
(each of these are used throughout the image controller). We fake the query string
parameter by setting the req.params object with an image_id property. Finally, we
create a test image object that will be used by our fake mongoose image model stub
to emulate a database object being returned from MongoDB:
describe('Index', function(){
it('should be defined', function(){
it('should call Models.Image.findOne', function(){
ModelsStub.Image.findOne = sinon.spy();
image.index(req, res);
it('should find Image by parameter id', function(){
ModelsStub.Image.findOne = sinon.spy();
image.index(req, res);
{ filename: { $regex: 'testing' } },
[ 209 ]
Testing Your Code
// to do: write more tests...
The first test we run is to ensure that the index function actually exists. Within the
index function, the very first action that occurs is that the image model is found via
the Models.Image.findOne function. In order to test that function, we need to first
set it as spy. The reason we do this here and not in beforeEach is because we might
want the findOne method to behave slightly differently in each test, so we don't
want to set a strict rule to be applied for all the tests.
In order to emulate that a GET call was posted to our server and our image index
controller function was hit, we can just fire the function manually. We do this using
image.index(req, res) and pass in our fake request and response objects (defined
earlier as globals and stubbed in the beforeEach function).
Since ModelsStub.Image.findOne is a spy, we can test that it was called, and then
separately test that it was called specifically with the parameters we expect it to be
called with. In the case of findOne, where the second parameter is a callback function,
we don't care or want to test the very specific function that was included, only ensure
that an actual function was included. To do this, we can use Sinon's matcher API and
specify that a func, or function, was included as the second parameter.
This last set of tests tests the code that executes when an image is found and
returned from the findOne function:
describe('with found image model', function() {
ModelsStub.Image.findOne =
it('should incremement views by 1 and save', function(){
image.index(req, res);
it('should find related comments', function(){
image.index(req, res);
{image_id: 1},
{ sort: { 'timestamp': 1 }},
[ 210 ]
Chapter 9
it('should execute sidebar', function(){
ModelsStub.Comment.find =
sinon.stub().callsArgWith(3, null, [1,2,3]);
image.index(req, res);
{image: testImage, comments: [1,2,3]},
it('should render image template with image and comments',
ModelsStub.Comment.find =
sinon.stub().callsArgWith(3, null, [1,2,3]);
sidebarStub.callsArgWith(1, {image: testImage, comments:
image.index(req, res);
expect(res.render).to.be.calledWith('image', {image:
testImage, comments: [1,2,3]});
The first thing to notice here is that findOne is no longer a spy in these tests, but
a stub that will manually fire the callback function that's provided as its second
parameter. The callback function that's fired will include our test image model.
With this stub, we are emulating that the database call was in fact made via findOne
and that a valid image model was returned. Then, we can test the remainder of the
code that executes within that main callback. We perform a similar setup with the
Comment.find call as well.
When the sidebarStub gets executed, we use the callsArgWith Sinon function—
which fires the callback function that was originally included as a parameter. Within
that callback function, we include fake viewModel as a parameter.
Once sidebarStub does its job, we expect res.render to have been called, and we
specify the exact parameters we expect it to have been called with.
Running the tests for the image controller should yield the following output:
$ npm test
Image Controller
should be defined
should call Models.Image.findOne
should find Image by parameter id
with found image model
[ 211 ]
Testing Your Code
should incremement views by 1 and save
should find related comments
should execute sidebar
should render image template with image and comments
Spy and stub everything!
When in doubt, the safest thing you can do when writing your tests is spy on
everything and stub everything else. There are always going to be times that you'll
want a function to execute naturally; in that case, leave it alone. Ultimately, you
never want your tests to be dependent on any other system—that includes database
servers, other network servers, other APIs, and so on. You only want to test that your
own code works, nothing more. If your code is expected to make a call to an API, spy
on the actual call and just assert that your code attempted to make the call. Likewise,
fake the response from the server via a stub and ensure that your code handles the
response properly.
The easiest way to check for dependencies in your code is to stop any other services
from running (your local node app, and so on), as well as, possibly even, disabling
your network connection. If your tests timeout or fail somewhere unexpectedly, it's
likely because you missed a function you needed to spy on or stub along the way.
Don't get stuck going down a rabbit hole when writing your tests. It's easy to get
carried away and start testing functionalities that should safely be assumed to
be working. An example of this is writing tests to ensure a third-party module is
performing correctly. If it's not a module you wrote, don't test it. Don't worry
about writing tests to prove the module does what it says it should.
To learn more about TDD specifically when writing JavaScript, I would highly
recommend Christian Johansen's beast of a book Test-Driven JavaScript Development.
This book is huge and speaks of the sheer volume of information related to TDD. In
some circles, TDD truly is a way of life and it will define the style with which you
write your code.
[ 212 ]
Chapter 9
This has definitely been a crash course on testing, but the groundwork has been laid
and I hope that you have a solid understanding of the tool chain that you can use
to write your own tests. Trust this powerhouse suite of tools and you'll be writing
bulletproof code in no time!
The ultimate goal with writing tests is to have 100 percent complete code coverage
and have unit tests exist for every line of code you write. From here, the true test is
to switch to TDD, which dictates that you write tests before any code exists at all.
Obviously, tests against nonexistent code will fail, so then you'll need to write the
least amount of code to get it to pass, and repeat!
In the next chapter, we'll take a look at a number of cloud-based hosting options
available to get your application up and running online.
[ 213 ]
Deploying with
Cloud-Based Services
Inevitably, you'll want the application you've been building to be online and
available to the world—whether you want to host your application online during its
development process or when it's complete and ready for production. There are a
number of different hosting options currently available for Node.js and MongoDBbased apps, and in this chapter, we'll take a look at deploying to a few different
popular services.
In this chapter, we will cover:
• Cloud versus traditional web hosting
• Introduction to Git source control
• Deploying an application with Nodejitsu
• Deploying an application with Heroku
• Deploying an application with Amazon Web Services
• Deploying an application with Microsoft Azure
• A brief look at Digital Ocean
[ 215 ]
Deploying with Cloud-Based Services
Cloud versus traditional hosting
If you had any previous experience with website hosting in the past, which I'll
refer to as traditional hosting, you're probably pretty familiar with the process of
using FTP to upload your web files to your hosting provider. With traditional web
hosting, service providers typically offer shared space to every user, each configured
with their own public folder that houses the web files. In a scenario like this, every
customer hosts the same kind of website, and their files are all stored and served
from a single web server.
Traditional web hosting is fairly inexpensive because a single web server can host
literally hundreds, if not thousands, of individual websites. Scaling is typically a
problem with traditional hosting because if your website demanded more power,
it would need to be moved to another server (with more hardware) and could
experience potential downtime during this move. As a side effect, if a website on the
same server as your own is being particularly demanding of the hardware, every site
on that server could suffer.
With cloud-based hosting, every instance of a website or service is hosted on its own
Virtual Private Server (VPS). When a customer uploads a copy of their website,
that website is running in its own isolated environment, and the environment is
specifically designed to run only that website. Virtual private servers are instances of
a server, typically all running simultaneously on the same hardware. Because of their
isolated nature, VPS scales very well because settings simply need to be changed
for hardware allocation and the server restarts. If your VPS is hosted on the same
hardware as others, and they are experiencing high-volume spikes, your website
will not suffer because of the isolated nature of the VPS.
Infrastructure as a Service versus Platform
as a Service
The beauty of the cloud is that the level and amount of service one can obtain
varies greatly. From something as simple as a basic hosting plan to run your web
application, you can use any number of services that are considered a Platform
as a Service (PaaS). This is a service that provides a platform for you to host and
run your web application. Increasing in scale and complexity, you can turn to an
Infrastructure as a Service (IaaS) provider, which offers an entire cloud-based
data center at your disposal.
[ 216 ]
Chapter 10
You can learn more about the differences between IaaS, PaaS, and
Software as a Service (SaaS) by reading this detailed article at http://
Cloud-based hosting costs can vary greatly because of the simple fact that they
are so scalable. Your costs could fluctuate throughout a single month dramatically
depending directly on your need for power (that is more demanding times of the
month and/or big social media hits such as HackerNews or Reddit). On the flip side,
if you require very little power for a server, often you can get cloud hosting for free!
Traditional web hosting service providers include GoDaddy, Dreamhost,
1&1, HostGator, and Network Solutions. Popular cloud-based hosting
options include Nodejitsu (PaaS), Heroku (PaaS), Amazon Web Services
(IaaS), Microsoft Azure (IaaS), and Digital Ocean.
An introduction to Git
With traditional hosting providers, the standard method for connecting to your
server and uploading your files was to use File Transfer Protocol (FTP). You would
connect using any standard FTP software and push a copy of your files to the server
and those changes would be reflected instantly online when accessing your website
URL. With cloud-based hosting providers, the standard typically is to use the Git
source control. Git is a source control technology that allows you to track changes
and history with your project source code as well as provide an easy-use means of
collaboration with multiple developers. The most popular Git online code repository
provider currently is www.github.com.
We are going to use Git in order to track our application project source code as well as
the method of pushing our code up to the various cloud-hosting providers. When you
push code using Git, you are effectively transferring all or only the changed version
of your code to an online repository (for example, Git as well as www.github.com are
relatively easy to get into but can seem intimidating and complex). If you're unfamiliar
with Git and/or https://GitHub.com, I would strongly suggest taking a moment to
get acquainted by checking out the following guides:
• https://help.github.com/articles/set-up-git
• https://gist.github.com/andrewpmiller/9668225
[ 217 ]
Deploying with Cloud-Based Services
The guides will take you through the following concepts:
• Downloading and installing Git
• Registering an account at https://github.com
• Authenticating your machine with https://github.com and creating your
first repository
• Committing your project source code to the repository
Once you have your project source code configured as a local Git repository and all
of the code committed to the master branch, proceed to the following sections.
Deploying your application
Now that you have your project set up as a local GitHub repository, its time to take
that code and get it online! The following sections will each cover the process of
deploying your application to a few different popular cloud-based hosting providers.
Feel free to explore and experiment with each as most have free or relatively
inexpensive plans. Each provider has its strengths and weaknesses, so I'll leave it up
to you to decide which to stick with for your particular needs. The services that we
cover aren't presented in any particular order.
Note that for the purposes of this chapter, I will consistently name my
app imgploadr; however, your app name needs to be different and
unique. Wherever I include imgploadr in this chapter, you should
replace it with your own app's unique name.
To get started with Nodejitsu, first visit www.nodejitsu.com and register for a free
account. After providing your e-mail address, username, and password, you will
be presented with a pricing plan page where you can configure your service. If you
just want to create the free account and experiment, simply click on the No Thanks
button and the registration process is complete. Then, simply click on the Login
button in the upper right corner to log in and proceed to your Apps dashboard:
[ 218 ]
Chapter 10
Deploying your app to Nodejitsu is going to require a new command-line interface
tool, specifically, the jitsu CLI. Clicking on the big blue Deploy an app with jitsu
button will take you to the www.github.com repository for this tool. You can skip
that step and just install the CLI manually using the following npm command:
$ sudo npm install –g–g install jitsu
Note that the sudo part of the command to install an npm package
globally (using the –g flag) is sometimes required. Depending on
your access level of the machine you are using, you may or may
not need to include sudo.
[ 219 ]
Deploying with Cloud-Based Services
Now that the jitsu CLI is installed, you can use this handy tool to log in to your
Nodejitsu account, create an app, and deploy your project code. First, let's log in:
$ jitsu login
Welcome to Nodejitsu
jitsu v0.13.18, node v0.10.26
It worked if it ends with Nodejitsu ok
Executing command login
An activated nodejitsu account is required to login
To create a new account use the jitsu signup command
prompt: username: jkat98
prompt: password:
Authenticated as jkat98
Nodejitsu ok
You can see that after successfully providing your username and password, you are
now authenticated with Nodejitsu and ready to go.
Before we can deploy the actual application, we need to first configure the MongoDB
database in our Nodejitsu dashboard. Switch back to your browser, and on the
Nodejitsu Apps dashboard, switch sections by clicking on the Databases tab:
[ 220 ]
Chapter 10
Let's choose MongoHQ for our needs by clicking on the large MongoHQ button.
You will be prompted for a name for the new database and then it will be listed at
the bottom of the screen in the Your databases section. The important part we need
is the connection string, and there's a convenient copy link right next to it to copy it
to your clipboard.
Edit the server.js file and update the mongoose.connect line to use this new
connection string you copied for your Nodejitsu database:
mongoose.connection.on('open', function() {
console.log('Mongoose connected.');
The only thing left remaining is to open a terminal, change directories to your project
home, and execute the following command to package up your app and push it to
$ jitsu deploy
Welcome to Nodejitsu jkat98
jitsu v0.13.18, node v0.10.26
It worked if it ends with Nodejitsu ok
Executing command deploy
The package.json file is missing required fields:
Subdomain name
Prompting user for required fields.
Press ^C at any time to quit.
prompt: Subdomain name: (jkat98-imgploadr) imgploadr
About to write /Users/jasonk/repos/nodebook/imgploadr/
... (a lot of npm install output) ...
Done creating snapshot 0.0.1
Updating app myapp
Activating snapshot 0.0.1 for myapp
Starting app myapp
App myapp is now started
http://imgploadr.nodejitsu.com on Port 80
Nodejitsu ok
[ 221 ]
Deploying with Cloud-Based Services
After executing jitsu deploy, the CLI first prompts to confirm what the subdomain
will be under the www.nodejitsu.com domain. Feel free to change this to whatever
you like (it will check to confirm availability). It will then make a few minor
modifications to your package.json file, specifically including the subdomain
option with whatever value you provided. Finally, it uploads your source code and
performs a remote npm install operation. Assuming all went well, the app should
be deployed and a confirmation of the URL outputs to the screen. Feel free to open
that URL in your browser to view the app online!
You can also now see that the app is listed in your Apps dashboard:
Now that the application has been successfully uploaded, launch it via its URL and
give it a test run by attempting to upload a new image. The first thing you should
notice is that attempting to upload an image fails with a fairly useless error (you can
see the following error by accessing the Logs tab from your app's dashboard):
400 Error: ENOENT, open
[ 222 ]
Chapter 10
This error is far from helpful! Basically, what's happening here is that the application
is attempting to upload and save the image to the temp folder that doesn't actually
exist! We need to add a snippet of code to our application to check for this condition
and create the folders if necessary.
Edit the server/configure.js file and insert the following code snippet between
routes (app); and return app;:
// Ensure the temporary upload folders exist
fs.mkdir(path.join(__dirname, '../public/upload'),
Here we're using the filesystem fs module to create both the parent upload folder
as well as the temp subfolder. Don't forget to require the fs module at the top of
the file too:
var connect = require('connect'),
path = require('path'),
routes = require('./routes'),
exphbs = require('express3-handlebars'),
moment = require('moment'),
fs = require('fs');
There is an npm module called node-mkdirp that will perform a
recursive mkdir, which basically would accomplish the double mkdir
we called in the preceding example. The only reason I didn't include it
was for brevity and to not unnecessarily include additional instructions
to install the module, require it, and use it. More info can be found at
[ 223 ]
Deploying with Cloud-Based Services
With the mentioned changes made to your code, you need to deploy your
application again. Simply execute another jitsu deploy and a fresh copy
of your code will be uploaded to your instance:
$ jitsu deploy
Open your app URL again, and this time you should be able to interact with the
application and successfully upload a new image! Congratulations, you have
successfully deployed your application and it is now online using the Nodejitsu
hosting service!
Another popular cloud-based hosting provider for Node.js apps is www.Heroku.com.
The one thing that sets Heroku apart from other providers is the number of powerful
add-ons that are available. Any kind of service you can imagine your application
will require is available as an add-on, including data stores, search, logging and
analytics, e-mail and SMS, workers and queuing, monitoring, and media! Each of
these add-ons can be quickly and easily added to your service and integrated into
your application with ease.
Like Nodejitsu, Heroku allows you to register a free account and work within the
confines of their "sandbox" pricing plans. The plans are free, but limited in scope
with regard to bandwidth, processing power, and so on. Most, if not all, of the
add-ons typically also offer some sort of free sandbox or trial-based plan. Just like
Nodejitsu, one of the add-ons we will be using with our Heroku app is MongoHQ, a
cloud-based MongoDB service provider.
To get started, first go to http://heroku.com and sign up for your free account.
While registration doesn't require a credit card, in order to include any add-ons with
your application, you will have to have a credit card on file (even though it won't be
charged unless you choose to scale up the services). After registering, click on the
link in the confirmation e-mail and provide a password; you will be presented with
your Apps dashboard:
[ 224 ]
Chapter 10
You'll notice that the first thing you need to do is download the Heroku Toolbelt
(again, much like the jitsu CLI for Nodejitsu). Click on the Download button to
download and install the Toolbelt. The Toolbelt is a CLI, specifically to create and
deploy apps to Heroku, and gives you the heroku command.
Once you have the Toolbelt installed, open a command-line terminal and change
directories to your project's root. From there, execute the following command
to log in to Heroku:
$ heroku login
Enter your Heroku credentials.
Email: jkat98@gmail.com
Password (typing will be hidden):
Authentication successful.
Now that you're logged in, you can issue commands directly to your Heroku
account and use those commands to create an application, install add-ons, and
deploy your project.
[ 225 ]
Deploying with Cloud-Based Services
The first thing you'll want to do is create a new application. Do so by executing
heroku create from the command line:
$ heroku create
Creating secret-shore-2839... done, stack is cedar
http://secret-shore-2839.herokuapp.com/ | git@heroku.com:secretshore-2839.git
After creating the app, Heroku randomly assigns it a unique name; in my case,
secret-shore-2839 (don't worry though as this can easily be changed):
$ heroku apps:rename imgploadr --app secret-shore-2839
Renaming secret-shore-2839 to imgploadr... done
http://imgploadr.herokuapp.com/ | git@heroku.com:imgploadr.git
Don't forget to update your Git remotes on any local checkouts.
Let's address that last part next. Heroku relies on the Git source control on your
machine in order to push your project source code up to your server, unlike
Nodejitsu, which uses its own file transfer mechanism. Assuming you followed the
directions earlier with regard to Git and www.github.com your project source code
should be all set and committed to the master branch and ready to go. What we
need to do next is add a new remote for Git on your machine to point to Heroku
and specifically your new app.
Let's start with git init to initialize git in the current working directory and then
execute the following command to create a new remote for Heroku:
$ git remote add heroku git@heroku.com:imgploadr.git
Before you can push your source code up to your Heroku account, we need to take
care of a few things first.
A special file is required before your application will be able to run on your Heroku
server. This file is called Procfile, and it specifically contains the command
necessary to launch your application. Create this new file named Procfile (no
extension) in the root of your project and include the following line:
web: node server.js
That's it! With that file, Heroku will use that command to launch your application.
Now that you have Procfile set up and your project source code ready, there's only
one thing left to do—install the MongoHQ add-on and configure your app to use it:
$ heroku addons:create mongohq --app imgploadr
Adding mongohq on imgploadr... done, v3 (free)
Use 'heroku addons:docs mongohq' to view documentation.
[ 226 ]
Chapter 10
With the MongoHQ add-on added, you can now configure the database itself and
retrieve the connection string (much like you did earlier with Nodejitsu). Access
your http://heroku.com Apps dashboard, and it should look something like the
following screenshot:
The app's dashboard screen is a great place to get a snapshot of your application and
a quick glance at its current cost. Since I'm using the sandbox and/or free plans for
my application and add-ons, my current estimated monthly cost is $0.00. However,
you can quickly and easily scale your apps should you demand more power. Pay
attention, as you can also just as quickly and easily escalate your monthly cost
through the roof! (Scaling everything to maximum, I was able to get my estimated
cost to roughly $60,000 per month!)
[ 227 ]
Deploying with Cloud-Based Services
To configure your MongoHQ database, simply click on the MongoHQ link under the
Add-ons section of your app's dashboard:
Click on the Admin tab with the gear icon below the Collections tab. Click on the
Users tab and provide a username and password that your application will use to
connect with your MongoHQ database. This will create the imgploadrdb username
with a secure password. With the new user added, switch back to the Overview tab
and copy the Mongo URI string.
Again, just like with Nodejitsu, edit the server.js file in your project and replace
the mongoose.connect string with the new URI you just copied. Edit the string and
replace <username> and <password> with the appropriate values based on the new
user account you just created. The server.js mongoose.connect code should look
like the following:
mongoose.connection.on('open', function() {
console.log('Mongoose connected.');
[ 228 ]
Chapter 10
Since you just made changes to your project's source code, you need to remember
to commit those changes to your Git repository master branch so that they can get
uploaded to Heroku. Execute the following command to permanently commit these
changes to your source code and upload your code to your Heroku server:
$ git commit -am "Update mongoose connection string"
$ git push heroku master
Initializing repository, done.
Counting objects: 50, done.
Delta compression using up to 8 threads.
Compressing objects: 100% (43/43), done.
Writing objects: 100% (50/50), 182.80 KiB | 0 bytes/s, done.
Total 50 (delta 3), reused 0 (delta 0)
... npm install output ...
To git@heroku.com:imgploadr.git
* [new branch]
master -> master
The final step to get your application up and running is to create an instance of your
server (basically the equivalent of turning it on). To do this, execute the following
$ heroku ps:scale web=1 --app imgploadr
Scaling dynos... done, now running web at 1:1X.
$ heroku open
Opening imgploadr... done
Success! Hopefully, your browser launched and your website is up and running. Go
ahead, give it a try and upload an image! Thanks to the bug we caught during the
Nodejitsu deployment, this updated version of the application should work just fine.
While deploying with Heroku seems more complicated than Nodejitsu, this is
probably because it uses Git source control to facilitate the transfer of your project
files. Also, because Heroku is so flexible with the power of its scaling and add-ons,
the toolbelt CLI is a little more robust.
[ 229 ]
Deploying with Cloud-Based Services
Amazon Web Services
While Nodejitsu and Heroku can be considered developer-level service providers
because they are Platforms as a Service, Amazon Web Services (and Microsoft Azure)
would be considered enterprise-level services because they are more Infrastructure
as a Service. The sheer volume of options and services available with AWS and
Azure is staggering. These are definitely top-tier services and hosting an application
like ours is kind of like using a bazooka to kill a fly!
AWS does provide its own NoSQL database called DynamoDB, but for our
purposes, we want to continue to work with MongoDB and use Mongoose in our
app. To do this, we can use a third-party MongoDB provider. If you recall, when we
originally set up Nodejitsu, one of the MongoDB providers listed was MongoLab.
MongoLab provides "MongoDB-as-a-Service", which means we can use their service
to host our MongoDB database, but use all of the power of AWS to host our Node.js
application (this is not unlike what's already happening with Nodejitsu and Heroku
already—they just streamline the process a little better). Remember that AWS is
an Infrastructure as a Service provider, so you could also just create another server
instance and install MongoDB on it yourself and use that as your data source. That's
slightly beyond the scope of this chapter however.
Create a MongoLab account and database
In order to use MongoLab with our app in AWS, we first need to register a new
account on http://mongolab.com and create an AWS database subscription. After
you register a new account and activate it using the link they send you via e-mail,
you can create your first database subscription.
From your main dashboard, click on the Create new button (with the lightning
bolt icon):
[ 230 ]
Chapter 10
From the Create new subscription page, configure the following settings:
• Cloud provider: amazon web services
• Location: (whichever region you prefer)
• Plan: Select Single-node (development)
Select Sandbox (shared/free)
• MongoDB version: 2.4.x
• Database name: anything_you_want (I chose imgploadr)
• Confirm that the price is $0 per month
• Click on Create new MongoDB deployment
Going back to your main dashboard, you should now see that your new database
has been created and is ready to go. The next thing we need to do is create a user
account that our app will use to connect to the server. Click on the database listed
on the main dashboard and then select the Users tab. Provide a new username and
password. After the new user account has been added, copy the URI located at
the top of the screen (it only appears after a user has been added) that starts with
[ 231 ]
Deploying with Cloud-Based Services
Now that you have the new URI connection string, we need to update server.js
to include this new connection string in our mongoose.connect. Edit the file and
update the following code:
mongoose.connection.on('open', function() {
console.log('Mongoose connected.');
Make sure you replace <username> and <password> with the appropriate
information from the user account you created on the MongoLab dashboard.
With our application code updated to point to the new MongoLab database
connection string, we need to zip up the project files so that they can be uploaded via
the AWS dashboard. From your computer's file browser, locate the project root that
contains all of your applications' source code files, select all of them and right-click
on them to add to an archive or ZIP file. The name of the ZIP file can be whatever
you choose. One thing to note is that you shouldn't include the node_modules folder
with this ZIP file (the easiest solution might be to simply delete the folder altogether).
The AWS online docs have a great write-up on creating ZIP files if you need more
information (https://docs.aws.amazon.com/elasticbeanstalk/latest/dg/
Once your source code has been updated to use the new MongoLab connection
string and you've created a ZIP file of the entire project (excluding the node_modules
folder), you're ready to create the new AWS application and deploy your app.
Create and configure the AWS environment
If you don't already have an account with Amazon, you're going to need one to
use their AWS services. Point your browser to http://aws.amazon.com and click
on Sign Up (even if you already have an Amazon account). From the screen that
follows, you can log in using your existing Amazon account or register for a new
account. Once you've registered and logged in, you should be presented with the
entire suite of cloud services AWS has to offer.
[ 232 ]
Chapter 10
The primary service we're interested in is Elastic Beanstalk (located under
Deployment and Managed with a green icon):
From this screen, click on the Create New Application link in the upper right corner.
The screens that follow will walk you through a multistep wizard process where you
will configure the environment in which the application will reside. Configure the
following settings where appropriate:
• Application Information:
Application name: anything_you_want
• Environment Type:
Environment tier: Web Server
Predefined configuration: Node.js
Environment type: Load balancing, autoscaling
• Application Version:
Upload your own (choose the ZIP file that you created earlier)
[ 233 ]
Deploying with Cloud-Based Services
• Environment Information:
Environment name: anything_you_want
Environment URL: anything_you_want (this is the subdomain for
your app)
• Configuration Details:
Instance type: t1.micro
The remaining fields can be left blank or at their default values
• Environment Tags: Skip this step— it is unnecessary for this app
• The final step is to review the configuration settings and then launch the
environment (by clicking on the blue Launch button). It might take a few
minutes for Elastic Beanstalk to configure and launch your environment
and application so you might need to sit tight:
With the environment officially launched and the application online, go ahead and
open your app (by clicking on the link at the top of the page) and give it a test run.
Assuming everything went according to plan, your application should be up and
running and working just fine!
[ 234 ]
Chapter 10
Microsoft Azure
Microsoft's Azure service is very similar to Amazon's AWS. Both can be considered
enterprise-level services and both offer a tremendous level of flexibility and power
with a really slick UI. Surprisingly, even though it's a Microsoft product, you can
spin up instances of Linux environments using Azure as well as hosting your
Node.js and MongoDB apps!
The first thing you're going to need, like any other service, is a registered account
at http://azure.microsoft.com. You can use an existing Microsoft Live Login if
you have one; otherwise, you can register a new account fairly easily. Once you're
logged into the Azure service, the first thing you'll be presented with is your primary
dashboard. The icons to the left are all of the various services and options available
with Azure:
[ 235 ]
Deploying with Cloud-Based Services
Clicking on the +NEW icon at the bottom-left corner will present you with the
main dialog you can use to add any new service. For our purposes, we want to
add a website:
1. Select Compute, Web Site, and From Gallery.
2. Select Node JS Empty Site from the long list of gallery options. This will
create the necessary environment so that you have somewhere you can put
your application.
3. On the screen that follows, provide the URL for your app.
4. Leave the remaining fields as their default values.
5. Click on the checkmark icon to complete the setup process, and your website
will be created.
6. The next step is to set up the database server. Again, similar to AWS or
Nodejitsu, we are going to once again select MongoLab as our database
service provider:
7. Click on the +NEW icon again and select Store and browse the list until you
find and select MongoLab.
8. Click on the next arrow and browse through the various plans. For our
needs, we will leave Sandbox selected (since it's free).
9. Provide a name for your database; in my case, I entered imgploadrdb.
10. Click Next again to review and confirm the plan and monthly price (which
should be 0.00 $ per month).
11. Finally, click on the checkmark icon to Purchase this new subscription plan.
[ 236 ]
Chapter 10
After a few seconds, you should be taken back to your dashboard where you will see
entries for both the website and database app service listed:
Now that the database has been created and is ready, we need to include its
connection string in our application before we can upload our code:
1. Click on the database row to select it and go to its overview.
2. The bottom of this screen will contain a few icons, one of which is labeled
Connection Info (and has an icon that looks like >i). Click on that icon to
pop up a modal window that contains the connection string URI for your
new MongoLab database server.
3. Copy that URI to your clipboard.
4. Edit server.js in your local app and replace the mongoose.connect
connection string with the new string you just copied. There is no need to
update username and password as Azure has already taken care of this for
you by using the following code:
mongoose.connection.on('open', function() {
console.log('Mongoose connected.');
[ 237 ]
Deploying with Cloud-Based Services
Once that change has been made, save the file and don't forget to update your local
Git repository with the change, as we'll be using Git in the next section to push your
code to Azure (just like we did earlier with Heroku):
$ git commit -am "Azure connection string"
Back at the Azure dashboard, click on Web Site in the All Items list (or filter by
websites using the icons on the left toolbar). From this overview screen, locate
the Integrate source control section towards the bottom and click on the Set up
deployment from source control link. The following screenshot shows what you
should see at this point:
Select Local Git repository and then continue by clicking the next arrow icon.
[ 238 ]
Chapter 10
The screen that follows will present instructions on how to push your local code to
the remote Git repository that has just been created for your Azure website. The gist
is to add a new Git remote (much like we did earlier with Heroku) that points to
your Azure repository and then push your code:
$ git remote add azure SPECIFIC_URL_FOR_YOUR_SERVER
$ git push azure master
You should notice the Git information screen in your Azure dashboard update in
real-time as your code starts to push up after the git push command. From the
command line, you will see a lot of remote npm install output as well. Once
completed, the deployment history in your Azure dashboard will update showing
the information for the last active deployment:
[ 239 ]
Deploying with Cloud-Based Services
Now that your code has been deployed to your Azure website and your website
connection string is pointing to your MongoLab Azure app service, you're
ready to give the website a test run! Launch it by pointing your browser to
Azure does a lot of things right (UI/UX) and has some really powerful options and
scaling features available! Taking a quick glance at the dashboard for a website
(the preceding screenshot) you can see that there is a lot going on! There are many
different configuration options as well as health monitoring and general information
(FTP settings, website URL, usage metrics, and so on), so feel free to poke around
and explore.
Digital Ocean
The last service I wanted to mention and briefly take a look at is Digital Ocean
http://digitalocean.com. Digital Ocean is a true Virtual Private Server (VPS)
service provider and is a good example of a service that gets you just about as "close
to the metal" as possible. What this means is that Digital Ocean doesn't really have
all the bells and whistles that the other services we've seen offer. What Digital Ocean
does offer, however, is direct unfiltered access to the Linux server instance you spin
up; in this case, referred to as Droplets:
[ 240 ]
Chapter 10
Digital Ocean allows you to boot up new Linux virtual server instances very quickly.
They offer very competitive prices and they're a great service if you need to get a
Linux server super fast because you only need one for a short period of time, or you
want to boot up your own Linux server that you plan to use to host a production
environment. The only "downside" (if I had to refer to it as such) is that you have
to be pretty familiar with Linux, specifically administering a server and all the
responsibilities that come with that.
You can very easily clone your project using Git on a new Droplet, but an example
of the actual raw nature of a new Droplet is that Git is not installed on the server
by default. You need to manually install Git before you can clone your repository.
Depending on which image you decided to clone when creating a new Droplet, you
might need to install and configure Node.js as well as MongoDB. Fortunately, Digital
Ocean offers a number of predefined servers you can choose from when creating a
new server—one of which includes the MEAN (MongoDB, Express, Angular, and
Node.js) stack. Beyond that, actually launching your app will only run as a process
during your currently logged in session—once you log out, your application will go
down. You would need to further administer the server to configure your app to run
as a service.
[ 241 ]
Deploying with Cloud-Based Services
Digital Ocean allows you to connect directly to your server using the console
access tool within the website itself, or using SSH directly from a terminal on
your own machine:
I mention Digital Ocean only because a lot of people will find this kind of raw power
quite refreshing and will want to do their own hands-on kind of configuration and
maintenance of their server. Digital Ocean is an awesome service but it's not for
everyone. I wanted to talk about it specifically because I feel that it rounds out and
completes the list of services we've covered so far.
We've covered the full spectrum of cloud-based hosting service providers and
walked through configuring your service and deploying your project code. Nodejitsu
and Heroku are great services that cater more to developers and give them a lot of
power through very accessible and slick user interfaces. Amazon and Microsoft, both
industry juggernauts, are representative of the kind of power and sophistication
you'd expect with enterprise-level service providers. Digital Ocean is a no-frills,
"close to the metal" cloud-based VPS provider that sacrifices bells and whistles for
raw and direct access to the server.
All of the hosting options we covered are great and not the only choices out there.
They're just a sample but speak to the power of the Cloud! Within a few minutes
and very little to no cost at all, you can have an environment configured, and your
website up and running online!
In the next chapter, we will take a look at the concept of Single-Page Applications
and popular client-side development frameworks and tools.
[ 242 ]
Single-Page Applications
with Popular Frontend
In this chapter, we will take a look at web application development from the
frontend perspective, specifically with a Single-Page Application (SPA), also
referred to as thick client apps. With SPA, a large chunk of the presentation layer
is off-loaded to the browser, and the browser is responsible for rendering pages,
handling navigation, and making data calls to an API.
In this chapter, we will cover:
• What exactly a Single-Page Application is
• Why we use a frontend framework such as Backbone.js, Ember.js,
or Angular.js
• Popular frontend development tools, such as Grunt, Gulp, Browserify,
SAAS, and Handlebars
• Test-driven development on the frontend
What is a Single-Page Application?
The current trend with sophisticated web applications is to emulate desktop
applications and veer away from the "feel" of a traditional website. With traditional
websites, every interaction with the server would require a full-page postback that
makes a complete round trip. As our web applications become more sophisticated,
the need to send and retrieve data to and from the server increases.
[ 243 ]
Single-Page Applications with Popular Frontend Frameworks
If we rely on full-page postbacks every time we need to facilitate one of these
requests, our app will feel sluggish and unresponsive as the user will have to wait for
a full round trip with every request. Users demand more from their apps these days,
and if you think about the application we've written, the Like button is a perfect
example. Having to send a full-page postback to the server just because we wanted
to increment a counter by one seems like a lot of unnecessary overhead. Fortunately,
we were able to easily rectify this using jQuery and AJAX. This is a perfect example
of how a Single-Page Application works (only on a much larger scale).
A great example of one of the first, standout Single-Page Applications is Google's
Gmail. Gmail gives you an interface that is similar to Microsoft Outlook or any
traditional desktop-based e-mail client. User interaction with the application feels
just as responsive as a desktop application—the "page" never reloads, you can
switch panes and tabs within the application with ease, and data is constantly being
refreshed and updated in real time.
Creating a Single-Page Application typically involves having a single HTML page
as the source of the application that loads all of the necessary JavaScript in order to
trigger a series of events that include:
• Bootstrapping the app: This means connecting to the server via AJAX to
download the necessary startup data.
• Rendering the screens based on user actions: This means monitoring events
triggered by the user and manipulating the DOM so that sections of the
app are hidden, revealed, or redrawn, which emulates the feel of a desktop
• Communicating with the server: This means using AJAX to constantly send
and receive data from the server, which maintains the illusion of a stateful
connection via the browser.
Why use a frontend framework?
We use frameworks to increase our productivity, keep us sane, and generally make our
development process more enjoyable. In most of the chapters throughout this book,
we worked with the Express.js MVC framework for Node.js. This framework allows
us to organize our code and it extrapolates out a lot of boilerplate code, freeing up our
time to focus on our custom business logic. The same should be said for the front of
an application as well. Any amount of complex code is eventually going to need to be
properly organized and use a standard set of reusable tools to achieve common tasks.
Express.js makes our life easy while writing our backend code with Node.js. There are
a number of popular frontend frameworks that you can rely on as well.
[ 244 ]
Chapter 11
The TodoMVC project
When deciding which frontend framework to choose for your next large scale
frontend project, the decision-making process can be crippling! Keeping track of all
of the different frameworks and the pros and cons of each can seem like an exercise
in futility. Luckily, people have answered the call and a handy website exists to not
only demonstrate the same application written in nearly every framework, but also
to offer the complete annotated source code for each as well!
The TodoMVC project, http://todomvc.com, is a website that focuses on creating
a simple, single page, to-do application, which is written using each of the proven
JavaScript MVC frameworks—there's even one written in vanilla JavaScript! So the
easiest way to jump into a framework would be to have a look at its TodoMVC code
Definitely spend some time checking out the website and digging into each of the
featured frameworks. You can get a really good feel of the different frameworks by
seeing the same code written in completely different ways; no two are identical, and
ultimately, it's up to you to evaluate and figure out which you prefer and why.
[ 245 ]
Single-Page Applications with Popular Frontend Frameworks
For the sake of brevity, I'm going to focus on the three that I personally like and
believe are at the top of the current list of front runners.
Backbone.js is an extremely lightweight (6.5 KB in production) MV* Framework that
has been around for a few years. It has an extremely large established user base, and
many very large-scale web applications have been written using this framework.
Some companies that have embraced Backbone.js for development of their flagship
products include:
• USA Today
• Hulu
• LinkedIn
• Trello
• Disqus
• Khan Academy
• Walmart Mobile
Backbone.js is a great framework to start with if you're comfortable with jQuery and
have been using it for a while and want to start improving your code organization
and modularity. Additionally, Backbone.js requires jQuery and integrates it pretty
tightly, so that's one less thing to worry about learning as you ease into this new
world of frontend development.
Backbone.js works on the basic idea of models, collections, views, and routers.
• Models are the basic elements that store and manage all of the data in your
• Collections store models
• Views render HTML to the screen, retrieving dynamic data from models and
• Routers power the URL of your application, allowing each individual section
of your application its own unique URL (without actually loading live URLs)
and ultimately tying the whole thing together
As Backbone.js is so lightweight, an extremely small and simple set of sample code
can be put together very quickly:
var Person = Backbone.Model.extend();
var PersonView = Backbone.View.extend({
[ 246 ]
Chapter 11
tag: 'div',
render: function() {
var html = [
return this;
var person = new Person({
name: 'Jason Krol',
website: 'http://kroltech.com'
view = new PersonView({ model: person });
The one thing to notice is that Backbone.js by its very nature is so lightweight that
it doesn't include most of the functionalities that you'd expect to work right out of
the box. As you can see in the preceding code, in the View object that we created, we
had to provide a render function that manually renders the HTML for us. For this
reason, many people shy away from Backbone.js, but others embrace it for the raw
power and flexibility it gives to developers.
Traditionally, you wouldn't put all of your code into a single file like the earlier
example. You would organize your models, collections, and views into individual
folders in a structure, just like how we organized the code in our Node.js application.
Bundling all of the code together would be the job of a build tool (which will be
discussed later in this chapter).
You can learn more about Backbone.js by visiting its official website at http://
backbonejs.org. Also, don't forget to check out the Backbone.js implementation of
the to-do application on the TodoMVC website!
I maintain a repository on GitHub that has a boilerplate web
application with complete code that uses the full stack we've
covered in this book, as well as Backbone.js with Marionette for
the frontend. Feel free to check it out at http://github.com/
jkat98/benm (Backbone, Express, Node, and MongoDB).
[ 247 ]
Single-Page Applications with Popular Frontend Frameworks
Ember.js bills itself as the framework for creating ambitious web applications. Ember's
goal is to target fairly large-scale SPAs, so the idea of using it to build something
very simple might seem like overkill but it is certainly doable. A fair assessment
is to take a look at the production file size of the Ember library, which comes in at
around 90 KB (versus 6.5 KB for Backbone.js). That being said, if you are building
something very robust with a very large codebase, the added 90 KB might not be
a big deal for you.
Here is a very small sample application using Ember.js:
var App = Ember.Application.create(),
movies = [{
title: "Big Trouble in Little China",
year: "1986"
}, {
title: "Aliens",
year: "1986"
App.IndexRoute = Ember.Route.extend({
model: function() {
return movies;
<script type="text/x-handlebars" data-template-name="index">
{{title}} - {{year}}<br/>
Ember.js's code looks somewhat similar to that of Backbone.js, and it's no surprise
that a lot of seasoned Backbone.js developers find themselves migrating to Ember.js
as their need for more robust solutions increase. Ember.js uses familiar items including
views, models, collections, and routes, as well as an Application object.
Additionally, Ember.js features components, which is one of its more powerful and
beloved features. Giving a sneak preview of the future of the Web, components allow
you to create small, modular, reusable HTML components that you can plug into
your application as needed. With components, you can basically create your own
custom HTML tags that look and behave exactly how you define them, and they
can easily be reused throughout an application.
[ 248 ]
Chapter 11
Developing with Ember.js is all about convention. Unlike Backbone.js, Ember.js tries
to get a lot of the boilerplate out of the way and makes certain assumptions for you.
Because of this, you need to do things a certain way, and controllers, views, and
routes need to follow a somewhat strict pattern with regards to naming conventions.
The Ember.js website features incredible online documentation and gettingstarted guides. If you're interested in learning more about Ember.js, check it out at
http://emberjs.com/guides/. Also, don't forget to take a look at the TodoMVC
AngularJS exploded onto the scene because of the simple fact that it's built by
Google (it is open source). AngularJS is basically like putting HTML on steroids.
The applications and pages that you create use regular HTML that we're all used
to, but they include a number of new and custom directives that extend the core
functionality of HTML, giving it awesome new power.
Another great feature of AngularJS that has seasoned non-web developers flocking
to it is that it is built from the group to be heavily tested and supports dependency
injection. It's a framework that doesn't makes creating sophisticated web applications
feel like traditional web development. This is an extremely robust framework,
clocking in at the largest file size of the three we're looking at, with 111 KB of
compressed production code:
<!doctype html>
<html ng-app>
<script src=
<input type="text" ng-model="yourName" placeholder="Enter a
name here">
<h1>Hello {{yourName}}!</h1>
[ 249 ]
Single-Page Applications with Popular Frontend Frameworks
You can see by the sample code provided that no custom JavaScript was written at
all, yet the page features real-time data binding between an input field and an h1 tag.
This is stock power and functionality right out of the box, and a demonstration of the
extended nature of regular HTML that AngularJS provides.
Make no mistake, however, that JavaScript still plays a huge role in the development
of AngularJS. AngularJS features controllers, directives, providers, and many more
Learn more about AngularJS by visiting its website at http://angularjs.org, and
take a look at the TodoMVC implementation as well!
Frontend frameworks have recently taken on somewhat religious undertones. Post
a negative comment or criticism about a particular framework and it's likely you'll
get blasted by its supporters. Likewise, talk positively about a particular framework
and, again, it's likely you'll get attacked about how much better a different framework
handles the same topic. The bottom line when deciding which framework is right for
you and/or your project is typically going to be about personal preference. Each of the
frameworks featured on the TodoMVC website can clearly accomplish the same goals,
each in its own unique way. Take some time to evaluate a few and decide for yourself!
Frontend development tools
Due to the sophisticated nature of Single-Page Applications, there exists a growing
suite of tools a frontend developer needs to be familiar with to manage many
day-to-day, and sometimes, minute-to-minute tasks.
Automated build task managers
A build tool is just what it sounds like- a tool used to build your application. When
a frontend developer creates and maintains an application, there could be a number
of tasks that need to be repeated literally every time a file is changed and saved.
Using a build tool, a developer can free up time and mental resources by offloading
the responsibility to an automated task manager that can watch files for changes and
execute any number of tasks needed. These tasks might include any number of the
Uglification and obfuscation
Dependency installation and preparation
[ 250 ]
Chapter 11
Custom script firing
Concurrent watchers
Server launching
Test automation
Some of the more popular build tools today include Grunt, Gulp, and Broccoli.
Grunt.js has been around for a number of years and is very well-established in the
development community. Gulp and Broccoli are fairly new but quickly gaining
traction and work a little differently than Grunt. With Grunt, you define and manage
your tasks using a configuration file whereas with Gulp and Broccoli, you write
Node.js code and use the raw power of streams. Many developers find working with
Grunt's configuration file to be fairly convoluted and frustrating and find working
with Gulp to be a refreshing change. However, it's hard to dispute Grunt's history
and popularity.
All three are feature-extensive ecosystems of plugins that help automate literally
everything and anything you can think of with your build process.
Here is a sample output from a typical Grunt build command:
[ 251 ]
Single-Page Applications with Popular Frontend Frameworks
In a typical Single-Page Application, the build manager can be responsible for
downloading and installing dependencies, concatenating multiple JavaScript files
into a single file, compiling and shimming Browserify modules, linting JavaScript
files for syntax errors, transpiling LESS files into production-ready CSS files, copying
files to a runtime destination, watching files for changes to repeat any of the tasks
again, and finally, running appropriate tests any time the code is changed—all from
a single command!
Grunt can be installed using npm and should be installed globally. Execute the
following command to install the Grunt CLI on your machine:
$ npm install -g grunt-cli
Refer to the getting-started guide on the official Grunt.js website for more
information at http://gruntjs.com/getting-started
Additionally, feel free to check out Gulp and Broccoli as well for more information:
• http://gulpjs.com/
• https://github.com/broccolijs/broccoli
Dependency management
There are literally millions of JavaScript libraries that exist to help you with
everything from DOM manipulation (jquery) to timestamp formatting (moment.js).
Managing these libraries and dependencies can sometimes be a bit of a headache. For
the frontend, the dependency manager of choice is Bower.io.
Bower works almost in exactly the same way as npm; it manages the packages in
the bower.json file. While working on the frontend (you need a known JavaScript
library or plugin such as underscore, for example), simply execute bower install
underscore and the JavaScript files will be downloaded to a local bower_
components folder in your project. From there, you can automate the inclusion of
those scripts by updating your build process, or simply copying the file, including a
script tag in your HTML, and then you're all set.
Bower can be installed using npm and should be installed globally. Execute the
following command to install Bower on your machine:
$ npm install –g bower
$ bower install jquery
bower cached
bower validate
2.1.0 against git://github.com/jquery/jquery.git#*
bower new
version for git://github.com/jquery/jquery.git#*
[ 252 ]
Chapter 11
bower resolve
bower download
bower extract
jquery#* archive.tar.gz
bower resolved
bower install
jquery#2.1.1 bower_components/jquery
Visit the Bower.io website (http://bower.io) for more information, as well as the
full directory of scripts available to be installed via bower install.
When writing large JavaScript applications, the key is to keep your source code
well-organized and structurally sane. Unfortunately, JavaScript doesn't inherently
support the idea of modular code right out of-the-box very well. To solve this
problem, two popular libraries exist to allow you to write modular code and rely
on only the modules you need within each individual piece of code.
An absolute must-read and incredible resource for frontend design
patterns is Addy Osmandi's Learning JavaScript Design Patterns,
which you can read for free by visiting the following URL:
Require.js and Browserify are two of the most popular module loaders today. Each
has a very unique syntax and its own set of benefits. Require.js follows asynchronous
module definition, which means each piece of code needs to define its own
dependencies. Personally, I've worked with Require.js in the past, and recently I've
found that I really like working with Browserify. One of Browserify's strengths is
that it uses the same modular pattern as Node.js; so, writing frontend code using
Browserify feels identical to that of Node. You use module.exports and require
on the frontend, and you don't have to worry about syntax context switching if you
go back and forth between Node and the frontend within the same application.
Using a module loader in conjunction with one of the popular MVC frameworks
mentioned earlier is almost a requirement because the two go together like peanut
butter and jelly!
For more information, visit the following links:
• http://browserify.org/
• http://requirejs.org/
[ 253 ]
Single-Page Applications with Popular Frontend Frameworks
HTML template-rendering engines
Fortunately, we've covered the idea of HTML template-rendering engines already
throughout the course of this book. The topics and concepts transfer directly to
frontend applications as well. There are many different HTML template engines to
choose from for use in the browser.
Many template engines will be mustache-based, meaning that they use {{ and }}
for merge variables. Handlebars is currently my personal favorite, mainly because
it works so well in the backend and frontend of an application, and I really like
working with its helpers. Underscore.js has a built-in "lite" template-rendering
engine for use with Backbone.js but its syntax uses <% and %> (much like classic
ASP or ASP.net MVC Razor syntax). Typically, most frontend MVC frameworks
allow you to customize the template-rendering engine and use any engine you
want. For example, Backbone.js can be very easily set up to use Handlebars.js,
instead of Underscore.js by default.
Here's just a small sample list of some of the currently available frontend templaterendering engines:
• Underscore.js: http://underscorejs.org
• Handlebars: http://handlebarsjs.com
• Mustache: http://mustache.github.io
• Dust.js: http://akdubya.github.io/dustjs
• EJS: http://embeddedjs.com
Some of these will work at the backend as well as on the frontend.
CSS transpiling
The idea of using variables and logic within a CSS file sounds like a dream come
true, right? We aren't quite there yet (in the browser, anyway), however there are a
few tools that will let us use variables and logic in our CSS files and compile them
during our build step. LESS and SASS are two of the most popular CSS transpilers
currently available. They behave almost identically, with only slight differences in
syntax and features. The big difference is that LESS was written using JavaScript
and Node, whereas SASS uses Ruby; therefore, each has different requirements to
get running on your machine.
Here is a sample SASS style sheet file:
@mixin radius($radius) {
[ 254 ]
Chapter 11
-moz-border-radius: $radius;
-webkit-border-radius: $radius;
-ms-border-radius: $radius;
border-radius: $radius;
.upload-button {
border-bottom: solid 2px #005A8B;
background: transparent $sprite-bg no-repeat;
@include radius(4px);
cursor: pointer;
#step-status {
color:#dbdbdb; font-size:14px;
span.active {
&.basic-adjust, &.message-editor {
width: 525px;
.icon {
width: 50px;
Looking at the sample code, you can see that we have a few new elements that
wouldn't typically work in a regular CSS file. Some of these include:
• Defining custom variables for use throughout the style sheet
• Defining mixins, which act as pseudo functions for reusable styles (with
dynamic parameters)
• Including mixins and variables within our style definitions
• Nesting styles with parent/child relationships
[ 255 ]
Single-Page Applications with Popular Frontend Frameworks
When the previous code is transpiled using LESS (or in the case of the sample
code SASS), the output is a standard .css style sheet that adheres to all the normal
browser rules and syntax.
For more information on LESS and SASS, check out the following links:
• http://lesscss.org
• http://sass-lang.com
Testing and test-driven development
The development of a sophisticated frontend application is no different than any
other software application. The code is going to be complicated and robust, and
there's no reason not to write tests as well as a practice test-driven development. The
availability of testing frameworks and languages for the frontend is just as robust as
any other language. All of the tools and concepts we've used for testing the Node.js
code that we've written in this book can be used directly on the frontend as well.
Some other tools to consider for testing your frontend JavaScript are:
• Karma for running tests: http://karma-runner.github.io
• Jasmine for writing tests: http://jasmine.github.io
PhantomJS headless browser
One thing I'd like to point out with testing frontend code is that, typically, the test
runners want to run in a browser window. This is great and makes perfect sense,
but in the real world, automating your tests or quickly executing them with TDD
can be a bit painful when a browser window wants to open every time your test
suite runs. PhantomJS is a headless browser available that works perfectly in this
kind of scenario. A headless browser simply means it's a browser that runs from
the command line, in memory, with no actual interface (like a typical browser).
You can easily configure Karma to launch the test suite using PhantomJS instead of
your browser of choice. When using PhantomJS as your browser, your tests execute
behind the scenes and only errors are reported. Here is a sample output of a test suite
running with Karma using PhantomJS:
[ 256 ]
Chapter 11
This was a whirlwind tour of some of the most common frontend tools and
frameworks used when doing typical web development. We took a look at the
TodoMVC project and reviewed three popular JavaScript frameworks to build
robust and sophisticated frontend applications.
Popular build tools such as Grunt.js, Gulp, and Broccoli help developers streamline
their workflow process by automating a lot of the repetitive tasks that need to
occur every time a file is modified. From concatenating scripts into a single file,
to minifying and compressing, and to executing automated test suites, the task
runners can be configured to handle pretty much everything under the sun!
We took a look at two popular CSS transpilers with LESS and SASS and saw how
they can make creating and managing CSS style sheets dynamic with the use of
mixins, variables, and nesting.
Finally, you learned about PhantomJS, the headless browser, and using it when
running frontend tests so that the tests can be executed quickly and easily from the
command line using a test runner like Karma.
In the next and final chapter, we'll review some alternative frameworks to develop
web applications using Node.js and MongoDB.
[ 257 ]
Popular Node.js
Web Frameworks
Throughout this book, we've focused exclusively on using Express.js as our web
framework of choice, primarily because it's one of the most popular web development
frameworks for Node.js. It has been around for quite a while, and is very widely
used. However, there are a number of alternate frameworks available that I want to
introduce you to. Some of these frameworks are much more powerful and robust than
Express.js, while others are right in line or slightly less feature packed.
Most of the frameworks that exist today and discussed in this chapter are still in their
early stages of development. Some have not even reached a 1.0 status. The use of
these in a production environment should be considered carefully and under a fair
amount of scrutiny.
In this chapter, we will take a brief look at the following frameworks:
• Koa
• Meteor
• Sails
• Hapi
• Flatiron
[ 259 ]
Popular Node.js Web Frameworks
Koa is a new web framework designed by the same team that created Express.js.
The goal of Koa is to be smaller, more expressive, and a more robust foundation for
web applications. T J Holowaychuk, who is the creator of the Express framework,
is the same person who created Koa as well and you can see it focuses much of its
capabilities on generators, which is a feature found in other popular programming
languages such as Python, C#, and Ruby. Generators were introduced to JavaScript
with ECMAScript 6. Generators prevent the standard callback hell that is so popular
with the development process in Node.js. Koa has a lightweight architecture, so
it doesn't come with any middleware, rather, it leaves the choice of implementing
certain features to the developers.
The Koa website, http://koajs.com, is shown in the following screenshot:
More information about Koa and sample implementations can be found on its
website as well as at https://github.com/koajs/koa.
Meteor is a simple and complete web framework with the goal of giving developers
of any skillset the ability to build robust web applications in a fraction of the time.
It features a handy CLI tool that you can use to scaffold new projects very quickly.
[ 260 ]
Chapter 12
Meteor provides some core projects/libraries such as blaze, DDP, livequery, and so
on with a unified build system. This simplifies the overall development process and
yields an integrated developer experience. The overall architecture that the Meteor
platform handles is shown as follows:
Meteor aims to build real-time applications by providing its two core features, the
distributed data protocol on the server side and transparent reactive rendering on
the client side. For more details, please visit: http://meteor.com/features.
Another notable feature of this framework is its extensive package system, named
atmosphere, which has modules developed for most of the common application
use cases.
It is quickly gaining traction and becoming increasingly popular every day.
Currently, its GitHub repo has over 27,000 stars!
More information about Meteor can be found on its website as well as on its official
GitHub repo at https://github.com/meteor/meteor.
[ 261 ]
Popular Node.js Web Frameworks
Sails is another great MVC framework for building web applications using Node.js
that sometimes compares itself to Ruby on Rails. Unlike Meteor, Sails is database
agnostic, so it doesn't matter which data store you choose. Sails includes some
handy scaffolding tools such as an automatic RESTful API generation. Socket.io,
a real-time communication framework for Node.js, is built into Sails; so including
real-time functionalities in your application should be a breeze. Sails features some
nice production-level automation that would typically need to be handled by a
tool such as Grunt.js or Gulp (this includes the minification and bundling of CSS
and JavaScript for the frontend). Sails also includes basic security and role-based
authentication for your app, should you require that level of functionality. Sails
could be considered a more full-fledged enterprise level framework than Express,
as it has almost every feature that a popular framework such as rails has.
The Sails website is at http://sailsjs.com.
More information about Sails can be found on its website as well as its official
GitHub repo at https://github.com/balderdashy/sails.
[ 262 ]
Chapter 12
Hapi is the result of the team behind Walmart's online mobile website. The team
that built that website developed a rich set of Node.js utilities and libraries that
can be found under the Spumko umbrella. Considering the tremendous amount of
traffic Walmart's website receives on any given day, it's no wonder that the team at
WalmartLabs is on the top of their game when it comes to Node.js development and
best practices. Hapi is the web framework that was born from the ashes of real-world
trial and error.
The Hapi website is at http://hapijs.com:
More information about Hapi can be found on its website as well as its official
GitHub repo at https://github.com/spumko/hapi.
[ 263 ]
Popular Node.js Web Frameworks
Flatiron is yet another Node.js MVC web application framework. What sets Flatiron
apart from other frameworks is its package-based approach. Since it gives the power
and freedom to decide how much or how little the framework should include,
developers can pick and choose the packages they want to work with and include in
their project. It handles a lot of the basic data management responsibilities and CRUD
for you by supplying a powerful ODM that takes care of a lot of the heavy lifting.
The Flatiron website (http://flatironjs.com) is shown in the following screenshot:
More information about Flatiron can be found on its website as well as its official
GitHub repo at https://github.com/flatiron/flatiron.
[ 264 ]
Chapter 12
Even though we used Express.js exclusively throughout this book, there are many
other options available when creating web applications using Node.js. We examined
a few of those options in this chapter, including Meteor, Sails, Hapi, Koa, and Flatiron.
Each framework features its own strengths and weaknesses and its own unique
approach to the standard functionality that a web application requires.
That's all folks. I hope all the different facets of building a web app using Node.js
and MongoDB brought the readers to a progressive way of learning and developing
an awesome idea. Well, this is just the start. I would recommend you follow the
developer communities of all the technologies or libraries that you are going to use
in your own app.
The beauty of web development with Node.js is that there's no shortage of opinions
on how to accomplish a single task. MVC frameworks are no exception, and you can
see from this chapter there are a lot of really robust and feature-packed frameworks
to choose from.
[ 265 ]
Amazon Web Services (AWS)
about 1, 230
application, deploying with 230
about 249
URL 250
anonymous functions, JavaScript 38
API endpoint
consuming, Request used 182-184
deploying 218
deploying, with Amazon Web
Services (AWS) 230
deploying, with Heroku 224-229
deploying, with Microsoft Azure 235-240
deploying, with Nodejitsu 218-224
testing 199
writing 23
application code
about 24-26
actual database, checking 27
writing 24
Application Programming Interface
(API) 164
arrays, JavaScript 36
asynchronous execution, Node.js 40
async module 147
Atom 13
automated build task managers 250-252
AWS environment
configuring 232-234
creating 232-234
about 246
URL 247
basic API server
creating 171
basic Mongoose schema definition
example 120
Behavior Driven Development (BDD)
about 188
reference link 191
URL 73
URL 253
Brackets 13
URL 252
about 253
URL 253
built-in validation 122, 123
callback 151
callback, JavaScript 38
Chai 191
references 191, 192
tests, asserting with 190
Chrome Web Store
URL 165
client-side templating 72
[ 267 ]
versus traditional hosting 216
comments helper 147-149
comments module 107, 108
URL 14
configure module
about 59, 60
activating 61
Connect middleware
about 60
bodyParser 60
cookieParser 60
errorHandler 60
Handlebars 60
methodOverride 60
morgan 60
Content Delivery Network (CDN) 109
controller handler
adding 158, 159
about 52, 91, 92
CRUD, adding to 129
defining 61-67
testing 208-211
create, read, update, delete (CRUD)
about 44, 129
adding, to controllers 129
adding, to home controller 130-132
CSS transpiling 254-256
custom middleware 67
receiving 175-180
removing 180, 181
data types
array 119
Boolean 119
buffer 119
date 119
mixed 119
number 119
ObjectID 119
string 119
data types, JavaScript 31
DELETE request 180
dependency management 252, 253
dependent modules
obtaining 24
Digital Ocean
about 240-242
URL 240
Document Object Model (DOM) 109
URL 254
EcmaScript6 (ES6) 7
URL 254
about 248, 249
URL 249
endpoint URLs, RESTful API 165
event-driven, Node.js 40
Express.js 52, 244
external APIs
consuming, from Node.js 181, 182
file IO 7
File Transfer Protocol (FTP) 217
first test
running 196
writing 196
about 264
references 264
flow, JavaScript 32-34
frontend design patterns
reference link 253
frontend development tools
about 250
automated build task managers 250-252
CSS transpiling 254-256
dependency management 252, 253
HTML template-rendering engines 254
modularity 253
[ 268 ]
frontend framework
AngularJS 249, 250
Backbone.js 246, 247
Ember.js 248, 249
TodoMVC project 245, 246
using, reasons 244
frontend template-rendering engines
Dust.js 254
EJS 254
Handlebars 254
Mustache 254
Underscore.js 254
functions, JavaScript 36, 37
GET requests
responding to 173-175
about 217
references 217
global helpers, Handlebars 85, 86
Gmail 244
about 79
URL 79
Grunt 252
URL 252
URL 252
basics 83
object, binding to template 83
presentation logic, embedding 84
URL 254
using, as view engine 68, 69
Handlebars helpers
about 85
global helpers 85, 86
view-specific helpers 86
Handlebars image page template
updating 159, 160
about 263
references 263
about 146
async module 147
comments helper 147-149
helper sidebar 151-153
popular images helper 157
stats helper 154-156
helpers, for reusable code
about 103
comments module 107, 108
images module 105
sidebar module 103-105
stats module 105
helper sidebar 151-153
about 224
application, deploying with 224-229
URL 224
hoisting concept
reference link 38
home controller
about 129
CRUD, adding to 130-132
updating 93-95
home page view
creating 73-78
HTML template-rendering engines 254
HTTP verbs
GET 164
POST 164
PUT 164
displaying 96-98
uploading 98-103
image controller
about 132
comment model, inserting 144-146
image model, inserting 136-139
[ 269 ]
image model, retrieving 132-136
image model, updating 142, 143
support, adding for Like button 142, 143
test run, performing 139-142
image page view
creating 73-78
image removal capability
adding 158
images module 105
URL 183
imgPloadr.io 53
Infrastructure as a Service (IaaS)
about 216
versus Platform as a Service 216
installation instructions, Postman
REST Client 165-169
installation, modules
npm used 42-44
installation, MongoDB
about 19
on Linux 21, 22
on Mac OS X 19, 20
on Windows 7 / Windows 8 20, 21
verifying 22
installation, Node.js
about 14
on Linux 17
on Mac OS X 14, 15
on Windows 16, 17
testing 18
origin 7
URL 14
about 29
anonymous functions 38
arrays 36
callbacks 38
data types 31
flow 32-34
functions 36, 37
objects 34, 35
operators 32-34
references 48
reusing 5
sharing 5
syntax 30
variable, declaring in 30
variable scope 31, 32
updating 160, 161
URL 109
JSONView Chrome extension
using 170
Karma 256
about 260
references 260
layout file
creating, for app 79, 80
layouts 79
URL 256
MongoDB, installing on 21, 22
Node.js, installing on 17
Mac OS X
MongoDB, installing on 19, 20
Node.js, installing on 14, 15
MEAN (MongoDB, Express, Angular,
and Node.js) 241
about 260
references 261
Microsoft Azure
about 235
application, deploying with 235-240
URL 235
using 58, 59
[ 270 ]
Mocha framework
tests, running with 188-190
about 52, 120-122
defining 126-128
testing 205-207
models index file 128
Model View Controller (MVC) design
pattern 62
modularity 253
installing, npm used 42-44
module system, Node.js 41
about 9, 44
connecting to 115, 116
data, deleting 48
data, inserting 45
data, querying 46
data, updating 47
document, inserting 116, 117
document, retrieving 117, 118
features 9, 10
installation, verifying 22
online documentation 23
URL, for downloads section 19
URL, for online documentation 23
users 10, 11
using, with Node 114, 115
MongoDB installation, for Linux systems
reference link 21
MongoDB, installing
about 19
on Linux 21, 22
on Mac OS X 19, 20
on Windows 7 / Windows 8 20, 21
URL 230
MongoLab account
creating 230, 231
MongoLab database
creating 230, 231
about 118
built-in validation 122, 123
connecting with 125
models 120-122
schemas 119
static methods 124
URL, for guide 125
virtual properties 124
Mongo shell
accessing 44, 45
Mozilla Developer Network (MDN) 4
URL 254
networking 7
MongoDB, using with 114, 115
users 10
Node API
URL, for online documentation 48
about 218
application, deploying with 218-224
URL 218
about 2, 40
asynchronous execution 40
code, for starting HTTP server 7
environment setup, for JavaScript code 6
event-driven 40
external APIs, consuming from 181, 182
installation, testing 18
module system 41
non-blocking asynchronous execution 3, 4
online documentation 18
reference link 8
single threaded model 3
URL, for download sections 14
URL, for online documentation 18
using 8
Node.js core 41
Node.js, installing
about 14
on Linux 17
on Mac OS X 14, 15
on Windows 16, 17
Node.js Web frameworks
about 259
[ 271 ]
Flatiron 264
Hapi 263
Koa 260
Meteor 260, 261
Sails 262
about 223
reference link 223
node modules
stubbing, with Proxyquire 194-196
Notepad++ 13
about 5
URL 67
used, for installing modules 42-44
npm registry
reference link 18
Object Document Mapper (ODM) 40
Object-Oriented JavaScript
reference link 35
objects, JavaScript 34, 35
operators, JavaScript 32-34
partial views 81-83
PhantomJS headless browser 256
Platform as a Service (PaaS)
about 216
versus Infrastructure as a Service 216
popular images helper 157
Postman REST Client
about 165
installation instructions 165-169
POST request
using 175-180
node modules, stubbing with 194-196
PUT request
using 175-180
Relational Database Management
System (RDBMS) 8
Representational State Transfer (REST) 164
used, for consuming API
endpoint 182-184
about 253
URL 253
about 164
endpoint URLs 165
adding 158
testing 199-201
defining 61-67
references 262
sample application
creating 23
launching 27
sample JSON data
creating 172, 173
URL 256
about 119
defining 126-128
semicolon insertion
reference link, for article 30
testing 202-204
server-side templating 72
sidebar module
about 103-105
implementation, testing 108
Single-Page Application (SPA)
about 243
app, bootstrapping 244
communication, with server 244
[ 272 ]
creating 244
screens, rendering based on user
actions 244
URL 192
real-time web application 6
Software as a Service (SaaS) 217
spies 192-194
static methods 124
stats helper 154-156
stats module 105
stubs 192-194
Sublime Text 3 13
syntax basics, JavaScript 30
template engines, Node
reference link 69
templating engines 71
test-driven development 256
Test Driven Development (TDD) 187
test helper
writing 197-199
testing 256
asserting, with Chai.js 190
running, with Mocha framework 188-190
reference link 3
TodoMVC project
about 245, 246
URL 245
tools, for running frontend JavaScript
Jasmine 256
Karma 256
traditional hosting
versus Cloud 216
iterating on 108-112
URL 254
V8 engine
advantages 2
declaring, in JavaScript 30
variable scope, JavaScript 31, 32
view engine
Handlebars, using as 68, 69
view models 92
about 52, 72
rendering 87-89
view-specific helpers, Handlebars 86
Virtual Private Server (VPS) 216, 240
virtual properties 124
web application
booting up 58
building 53
designing 53, 54
entry point, creating 56, 57
files, organizing 55
web application frameworks 51
web application request
lifecycle 114
reference link 6
Node.js, installing on 16, 17
Windows 7
MongoDB, installing on 21
Windows 8
MongoDB, installing on 21
Windows Quick Start guide, MongoDB
reference link 20
URL 79
reference link 55
[ 273 ]
Thank you for buying
Web Development with MongoDB and NodeJS
Second Edition
About Packt Publishing
Packt, pronounced 'packed', published its first book, Mastering phpMyAdmin for Effective
MySQL Management, in April 2004, and subsequently continued to specialize in publishing
highly focused books on specific technologies and solutions.
Our books and publications share the experiences of your fellow IT professionals in adapting
and customizing today's systems, applications, and frameworks. Our solution-based books
give you the knowledge and power to customize the software and technologies you're using
to get the job done. Packt books are more specific and less general than the IT books you have
seen in the past. Our unique business model allows us to bring you more focused information,
giving you more of what you need to know, and less of what you don't.
Packt is a modern yet unique publishing company that focuses on producing quality,
cutting-edge books for communities of developers, administrators, and newbies alike.
For more information, please visit our website at www.packtpub.com.
About Packt Open Source
In 2010, Packt launched two new brands, Packt Open Source and Packt Enterprise, in order
to continue its focus on specialization. This book is part of the Packt Open Source brand,
home to books published on software built around open source licenses, and offering
information to anybody from advanced developers to budding web designers. The Open
Source brand also runs Packt's Open Source Royalty Scheme, by which Packt gives a royalty
to each open source project about whose software a book is sold.
Writing for Packt
We welcome all inquiries from people who are interested in authoring. Book proposals should
be sent to author@packtpub.com. If your book idea is still at an early stage and you would
like to discuss it first before writing a formal book proposal, then please contact us; one of our
commissioning editors will get in touch with you.
We're not just looking for published authors; if you have strong technical skills but no writing
experience, our experienced editors can help you develop a writing career, or simply get some
additional reward for your expertise.
Web Development with MongoDB
and Node.js
ISBN: 978-1-78398-730-6
Paperback: 294 pages
Build an interactive and full-featured web application
from scratch using Node.js and MongoDB
Configure your development environment to
use Node.js and MongoDB.
Explore the power of development using
JavaScript in the full stack of a web application.
A practical guide with clear instructions to
design and develop a complete web application
from start to finish.
Node.js Blueprints
ISBN: 978-1-78328-733-8
Paperback: 268 pages
Develop stunning web and desktop applications with
the definitive Node.js
Utilize libraries and frameworks to develop
real-world applications using Node.js.
Explore Node.js compatibility with AngularJS,
Socket.io, BackboneJS, EmberJS, and GruntJS.
Step-by-step tutorials that will help you to
utilize the enormous capabilities of Node.js.
Please check www.PacktPub.com for information on our titles
MongoDB Cookbook
ISBN: 978-1-78216-194-3
Paperback: 388 pages
Over 80 practical recipes to design, deploy, and
administer MongoDB
Gain a thorough understanding of some of the
key features of MongoDB.
Learn the techniques necessary to solve
frequent MongoDB problems.
Packed full of step-by-step recipes to help you
with installation, design, and deployment.
Learning MongoDB [Video]
ISBN: 978-1-78398-392-6
Duration: 03:26 hours
A comprehensive guide to using MongoDB for
ultra-fast, fault tolerant management of big data,
including advanced data analysis
Master MapReduce and the MongoDB
aggregation framework for sophisticated
manipulation of large sets of data.
Manage databases and collections, including
backup, recovery, and security.
Discover how to secure your data using SSL,
both from the client and via programming
Please check www.PacktPub.com for information on our titles