In Part 1, Getting Started With Expressjs, we did learn a lot about how Expressjs works as a Node.js framework. We built a simple and generic web server in the previous article that had two routes to function. We then discussed other paradigms such as devDependencies, REST services and the general anatomy of an Express application. In this article, we are going to continue to build upon the previous article, and we will start by generating a boilerplate project that can be used in any scenario, whether creating an application using Express entirely or just building a REST API using it.
Introducing Express Generator
Building a web server from scratch, time after time can be a tedious process. As a human being, it will be time-consuming, and in 2 out of 10 cases, there will be chances that you might forget about one configuration or the other, and the whole server is not working. To overcome these problems, Express comes with a module that can be installed globally on a local machine, called express-generator. It is a generator tool through which as a developer you can create and populate different routes using database calls and views or template (i.e., if using a template engine).
We will build rest of the series using this tool so it will be better if you install it. From your terminal, run this command:
npm install express-generator -g
To create our project, we will continue to build in the same directory. We will also be using a default template engine which will be explained later in this article.
express --view=pug
If you want to build the app from scratch, the only option you are missing from the above command is the name of the project:
express express-server --view=pug
Installing express-generator gives access to express command which can be used to scaffold a project structure and a basic web server. There are various options you can choose when starting out with a new project. Such as:
$ express --help
Usage: express [options] [dir]
Options:
-h, --help output usage information
--version output the version number
-e, --ejs add ejs engine support
--pug add pug engine support
--hbs add handlebars engine support
-H, --hogan add hogan.js engine support
-v, --view <engine> add view <engine> support (ejs|hbs|hjs|jade|pug|twig|vash) (defaults to jade)
-c, --css <engine> add stylesheet <engine> support (less|stylus|compass|sass) (defaults to plain css)
--git add .gitignore
-f, --force force on non-empty directory
You can always choose a view/template engine using –view along with a CSS engine using –css since there is massive support for various commonly used css generators such as SASS and LESS.
Running the Express-Generator Project
Notice the directory structure of the newly created application. It is close to MVC (model-view-controller) design pattern. We will be building our controllers part later, but for now, we can use routes/ for it. The Model part will be our database that is not defined in this structure, and we will have to add it manually. The views are available to us in views/ directory.
The package.json defines the application dependencies and other information. Let us take a look at it:
{
"name": "node-js",
"version": "0.0.0",
"private": true,
"scripts": {
"start": "node ./bin/www"
},
"dependencies": {
"cookie-parser": "~1.4.3",
"debug": "~2.6.9",
"express": "~4.16.0",
"http-errors": "~1.6.2",
"morgan": "~1.9.0",
"pug": "2.0.0-beta11"
}
}
By default, there are no devDependencies installed in this project, and we will be adding one sooner than later. On comparing to the previous project, we defined bootstrapping configuration in a file called index.js. This time, using this tool, our the configuration to run the Express server is available to us in /bin/www. This file further loads another file called app.js which we will be using to define our custom configuration and other essential parts such as defining a template engine, defining routes and other middleware functions.
The dependencies that come with this project are bare minimum but enough trigger a basic web server up and running. First of all the most important dependency of all express. Then there is debug module which you will see it in action below and is a node debugging utility, pug our template engine, http-errors to catch HTTP methods errors gracefully, and cookie-parser which is used to parse the information such as cookie header and populate req.cookies between a client and server. Lastly, morgan which is an HTTP request logger middleware for Node.js.
For now, we are going to install another dependency called body-parser which is another essential dependency we need and is used to parse the body of an incoming HTTP request. It makes our job as a developer to be easier as it helps in extracting different part of an incoming HTTP request as information useful to our application. Mostly, we need this when we are working with forms or any other POST HTTP requests from our client.
npm install --save body-parser
To run this project, we will have to install dependencies available to us by default.
npm install
Then we can run this project by running:
DEBUG=node-js:* npm start
To verify if our project is running, open the browser window and visit http://localhost:3000. You know have a basic working web server plus Expressjs application. At this step, we will also enable auto server restart on file changes as this functionality is not handled by Express framework by default. We will be using nodemon for this. To know how to install and use nodemon refer to article one in this series. Also, the DEBUG is provided to us by debug dependency.
To close this section, we will be changing our scripts object in a package.json file to make use of nodemon.
"scripts": {
"start": "node ./bin/www",
"devstart": "nodemon ./bin/www"
}
We can start our development server now by calling npm run devstart and nodemon will watch for file changes.
What is app.js?
the app.js file is where define our most configuration and other essentials such as routes to build the web server. This file is created by express-generator and name of the file is a convention with the community. Along with setting up the application, this file also defines various middleware functions and lastly exports app as a module to be used in other parts of the application. On practical example of using this app module is in /bin/www:
#!/usr/bin/env node
/**
* Module dependencies.
*/
var app = require('../app');
This is how our app.js file looks like now:
var createError = require('http-errors');
var express = require('express');
var path = require('path');
var cookieParser = require('cookie-parser');
var logger = require('morgan');
var indexRouter = require('./routes/index');
var usersRouter = require('./routes/users');
var app = express();
// view engine setup
app.set('views', path.join(__dirname, 'views'));
app.set('view engine', 'pug');
app.use(logger('dev'));
app.use(express.json());
app.use(express.urlencoded({ extended: false }));
app.use(cookieParser());
app.use(express.static(path.join(__dirname, 'public')));
app.use('/', indexRouter);
app.use('/users', usersRouter);
// catch 404 and forward to error handler
app.use(function(req, res, next) {
next(createError(404));
});
// error handler
app.use(function(err, req, res, next) {
// set locals, only providing error in development
res.locals.message = err.message;
res.locals.error = req.app.get('env') === 'development' ? err : {};
// render the error page
res.status(err.status || 500);
res.render('error');
});
module.exports = app;
We can see that the file starts by importing various dependencies. We use require() function. After that, we set views to specify the folder where our templates will be stored and then we set view engine value to pug to determine which template engine we will be using such that Express framework can identify by the file extension of any views file .pug.
Middleware Functions
app.use() is how we can use any middleware function. Middleware functions are those functions that have access to the request object (req), the response object (res), and the next function in an application’s request-response cycle. The motive of these functions is to modify incoming requests and outgoing response objects by performing tasks like parsing request bodies, adding response headers, or by making any other change to request-response cycle. Using a middleware function, you can even end the request-response cycle such as done using a custom error handling middleware function.
The next function is the most important of a middleware function, and this is what separates it from other regular functions. If a middleware function does include next(), that means the request-response cycle is ended there. The name of the function next() here is arbitrary, and you can name it whatever you like. However, the name is now stuck within the community and is more of a convention just like app.js.
In our app.js you can see the use of both a third party middleware and a custom one. For third-party middleware, we require morgan, express.json(), express.urlencoded etc. For a custom middleware function, see the one that is labeled as to catch 404 errors.
app.use(function(req, res, next) {
next(createError(404));
});
Error Handling Middleware Function
Another custom middleware function you can notice in our app.js file is the Error middleware function. What separates an error handling middleware function and a normal and how an express application differentiates between the two is the number of arguments passed. Notice in the below error handling middleware function:
// error handler
app.use(function(err, req, res, next) {
// set locals, only providing error in development
res.locals.message = err.message;
res.locals.error = req.app.get('env') === 'development' ? err : {};
// render the error page
res.status(err.status || 500);
res.render('error');
});
There are four arguments passed if compared to a normal middleware function as we learned in the previous section. The extra argument, err is what separates the two lets the express application differentiate between the two. Also, any error middleware must be defined last in the order. The order of every middleware function is important, and you can just replace one with another. The application or the web server might not work. So it is important to keep them in order. This is important because of the request-response cycle.
Serving Static Files
By default, we do not have to install any dependency use Express as a static server. express.static() will serve any file defined using a path. We use global variable __dirname together with path module to define such paths.
app.use(express.static(path.join(__dirname, 'public')));
As a convention, this middleware helps us to serve all static files in the /public directory of project’s root.
Parsing The HTTP Body
Remember, earlier we installed a module from npm called body-parser? Well, I am going to tell you now that you will have to remove it from your project. The reason is that the latest Express version now includes a second built-in middleware function called express.json and express.urlencoded that behave exactly in the same manner as body-parser.
app.use(express.json());
app.use(express.urlencoded({ extended: true }));
// can be written if using `body-parser`
var bodyParser = require('body-parser');
app.use(bodyParser.json());
app.use(bodyParser.urlencoded({ extended: false }));
body-parser have both of these functions. The reason I made you installed this package is that as a Node.js developer you are going to come across older Express versioned projects and you should know the difference, or in this case, the similarity between the two/ You can uninstall this package by using below command:
npm uninstall --save body-parser
Template Engine
Express support various template engines such as handlebars, pug, its previous version jade, ejs, twigs, etc. A Template engine is a library that allows us to use any template language. A template language is a special set of instructions that instruct this library to process data and display to on the client. Using a template engine is easy with Express. However, Express comes with a default template engine, Jade, which is the first released version of Pug. To demonstrate how to use a Template Engine, we will be using Pug.
Notice the file index.pug in the directory views/:
extends layout
block content
h1= title
p Welcome to #{title}
Do you remember at the start when we ran our server we were displayed with a title called Express? The corresponding route for this template is rendering that title for us. You can find the corresponding route in the routes/index.js file.
var express = require('express');
var router = express.Router();
/* GET home page. */
router.get('/', function(req, res, next) {
res.render('index', { title: 'Express' });
});
module.exports = router;
We will now add one more route to this file called /about.
router.get('/about', function(req, res, next) {
res.render('about', { title: 'About' });
});
Notice the res.render() which takes the first argument as the name of the template file and then we can send some options that will be used in our views. All this happens quickly behind the scenes. We will add the corresponding template for this route views/about.pug.
extends layout
block content
h1= title
p Welcome to #{title}
If you visit http://localhost:3000/about URL in your browser, you will notice those changes.
Logging
One more thing to notice here is that the use logger or morgan module we installed as middleware function early in our app.js. If you visit the terminal window, you will see the following output:
GET / 304 331.932 ms - -
GET /stylesheets/style.css 304 10.878 ms - -
GET /about 304 23.367 ms - -
GET /stylesheets/style.css 304 0.888 ms - -
GET /stylesheets/style.css 304 0.596 ms - -
The output is descriptive in itself. The GET / is the route part followed by an HTTP status like 304. Logging is one of the most useful components when building an application. In production mode, it can be a tremendous asset if used because it will be working as monitoring utility. Managing a logging component in a Node.js application can help the developer discover and omit problems, errors, etc. quickly and save time. With the help of a proper logging system only, you as a developer will be able to debug your Node.js application if encountered any error. The aim of this article is to help you setup and use logging in a Node.js application. First, there are some requirements for logging a typical Node.js application. Let us see what they are.
There are three requirements for a module logging work. You can think of them as a convention, but all these requirements will be helpful to you in development. They are:
- Timestamps: to convey the time of an event that happened in an application.
- Format: there are various logging formats used in the industry. To quickly understand, choose a format that maintains a good balance between being brief and descriptive enough
- Logging Levels: Sometimes having different levels to describe different events can be helpful. These levels could be warning, error, success, etc.
Using console methods are a huge part of JavaScript community, but they should not be used or relied upon to log an Expressjs application. To get the logging right in your Node.js application, the first thing that comes to mind and practice is using the console module. The console module is provided by Node.js core API. It is the basic kind of logging and is built in. It can be used for small and spontaneous purposes.
The problem with any console method is that once a problem is encountered, the logs cannot be switched off. It will always print. This is a huge disadvantage at the application level of this module. Another thing that lacks from console API is that you as a developer cannot assign levels of logging.
This where a logging module like morgan comes into the picture. We are using the dev format option and passing it as an argument to our module. Dev is the logging format we are using. There are various formats available such as tiny. Most formats will have a predefined structure just like the dev format:
:method :url :status :response-time ms - :res[content-length]
Using Environment Variables with Express Generator
Almost every web server needs some custom configuration. You can define them in a separate file using a convention called config.js but managing these configurations in applications can be tricky and as the project scales, hard to maintain. Node.js provides us a way to manage these configurations by defining separate environment variables in one application source code which makes the app itself easy to manage. To implement this process, you can use third party npm modules like dotenv(), but with Express generator tool we are going to use a different approach.
Create a directory in the root of our project config inside we will define a file called config.json where we will store our environmental variables.
{
"server": {
"port": 3001
}
}
We will edit our bin/www script to use this config.
#!/usr/bin/env node
/**
* Module dependencies.
*/
var app = require('../app');
var debug = require('debug')('node-js:server');
var http = require('http');
// add this line
const config = require('../config/config.json');
/**
* Get port from environment and store in Express.
*/
// change this line
var port = normalizePort(process.env.PORT || config.server.port);
app.set('port', port);
Now if you visit http://localhost:3001/ you will get the desired result.
This completes our second article in this series, and I hope it does add value to our education of creating a web server using Expressjs and Node.js. You download the source code of this article from here.