Writing web applications in Clojure
Functional Web
Clojure looks like Lisp and runs wherever Java is installed. Practical tools and sophisticated libraries provide the underpinnings for rapid development of modern web applications.
Clojure [1] made its first appearance in the software landscape in 2007. The functional programming language combines Lisp-like syntax with an implementation on the Java Virtual Machine (JVM). Within a few years, a Clojure community with user groups [2] and conferences (e.g., Clojure West [3] and Clojure/conj [4]) has been established.
An extensive ecosystem of libraries has emerged for the language, which is released under the Eclipse Public License. Clojure benefits from its Java origins, which lets it share the VM with Java and JRuby, and impresses with abstraction over concurrency [5] on multiprocessor systems. The name evokes the closure programming concept, with the "j" coming from Java.
Clojure advocates are enthusiastic about the improved productivity this expression-rich language permits. I strongly encourage you to take a look at an essay by Lisp advocate and entrepreneur Paul Graham [6] that describes the implications of the advantages Lisp programming confers, leading some Internet startups to focus on the language they use in large web applications. In this article, I show you how to use Clojure to implement a simple chat for the web browser (Figure 1).
Building with Leiningen
Most Clojure users rely on the Leiningen build system [7], which also takes care of installing the required libraries. If your package manager does not include it, you can follow the README file that comes with the source code for the install. The lein -v
command outputs the Leiningen version; when first called, the build system takes a few moments to download a few libraries. I used version 2.3.4 for this article, along with the most recent stable release of Clojure (1.5.1), with OpenJDK 7 as the Java implementation.
The project directory for the webchat is set by the
lein new webapp
command, which also sets up a skeleton for the application. Figure 2 shows the contents of webapp
: The project.clj
file contains information about the project and the components it needs – at present, only the current Clojure release.
Additionally, the src/
directory contains the first namespace of the webapp.core
application. Namespaces are the Clojure programmer's primary means of separating modules with different tasks. Each namespace is allowed to reference all objects or individual objects in other namespaces. However, it is possible to make certain objects private.
Interactive: REPL
The src/webapp/core.clj
file already contains a function. It can be called at the interactive command prompt, with which Clojure developers typically work, as well as in the editor. As usual in the Lisp family, the prompt is named REPL (Read Eval Print Loop). In the terminal, the developer types CD
to change to the project directory and then calls lein repl
.
The following REPL session uses the require
function and the in-ns
function to change into webapp.core
. Finally, it calls the foo
function with the argument "It works!"
:
user=> (require 'webapp.core) nil user=> (in-ns 'webapp.core) #<Namespace webapp.core> webapp.core=> (foo "It works!") It works! Hello, World! nil
A simple change to the source file demonstrates how REPL responds to changed code. In Listing 1 [8], the order of arguments in the println
function is reversed. After saving in the editor, reload the namespace in the same REPL to see the changed output:
Listing 1
src/webapp/core.clj
webapp.core=> (require 'webapp.core :reload) nil webapp.core=> (foo "It works!") Hello, World! It works! nil
This ability of REPL allows the developer to test code changes continuously. Editors like vim or emacs can integrate a REPL.
Bundled Libraries
The Clojure world is scarcely aware of large, monolithic frameworks. Most applications use a collection of small libraries, each of which serves a clearly limited purpose. Typical Clojure web applications rely on the Ring library, which represents an abstraction of the HTTP protocol. The sample application does not directly use Ring, however, but Compojure [9], which is based on Ring. On top of this, the project.clj
file defines some other libraries as dependencies (Listing 2).
Listing 2
project.clj Dependency List
Adding a new library is one of the few occasions the developer needs to restart the REPL. Although special tools can do this, entering Ctrl+D then lein repl
does the trick, and Leiningen automatically downloads the required files. This is followed by more changes to the core.clj
file in the editor. The following lines import functions from the Compojure library into the current namespace:
(ns webapp.core (:require [compojure.core :refer [defroutes GET]]))
This imports defroutes
and GET
from compojure.core
and makes them available for a first simple application:
(defroutes app (GET "/" [] "<h1>Hello World</h1>"))
Here, app
defines a Ring handler. It welcomes web surfers who access the /
address using HTTP-GET with a friendly "Hello World".
For this to work, the application requires an HTTP server, which is supplied by Httpkit [10] and is added to the required namespace in core.clj
:
(ns webapp.core (:require [compojure.core :refer [defroutes GET]][org.httpkit.server :as http]))
Now all functions can access org.httpkit.server
under the http/
prefix. This means you can try out the web server in the REPL:
user=> (require 'webapp.core :reload) nil user=> (ns webapp.core) nil webapp.core=> (def srv (http/run-server app {:port 8080})) #'webapp.core/srv
The http/run-server
function starts the application; def
binds to the return value to srv
. The browser application is now accessible under http://localhost:8080
. The function for launching the web server in turn returns a function. This can be checked in the REPL with the fn?
predicate. If you call the function itself, the server shuts down:
webapp.core=> (fn? srv) true webapp.core=> (srv) nil
The server can be restarted using (http/run-server app {:port 8080}))
without restarting the REPL, and you can even run multiple instances on different ports.
Buy this article as PDF
(incl. VAT)
Buy Linux Magazine
Subscribe to our Linux Newsletters
Find Linux and Open Source Jobs
Subscribe to our ADMIN Newsletters
Support Our Work
Linux Magazine content is made possible with support from readers like you. Please consider contributing when you’ve found an article to be beneficial.
News
-
Systemd Fixes Bug While Facing New Challenger in GNU Shepherd
The systemd developers have fixed a really nasty bug amid the release of the new GNU Shepherd init system.
-
AlmaLinux 10.0 Beta Released
The AlmaLinux OS Foundation has announced the availability of AlmaLinux 10.0 Beta ("Purple Lion") for all supported devices with significant changes.
-
Gnome 47.2 Now Available
Gnome 47.2 is now available for general use but don't expect much in the way of newness, as this is all about improvements and bug fixes.
-
Latest Cinnamon Desktop Releases with a Bold New Look
Just in time for the holidays, the developer of the Cinnamon desktop has shipped a new release to help spice up your eggnog with new features and a new look.
-
Armbian 24.11 Released with Expanded Hardware Support
If you've been waiting for Armbian to support OrangePi 5 Max and Radxa ROCK 5B+, the wait is over.
-
SUSE Renames Several Products for Better Name Recognition
SUSE has been a very powerful player in the European market, but it knows it must branch out to gain serious traction. Will a name change do the trick?
-
ESET Discovers New Linux Malware
WolfsBane is an all-in-one malware that has hit the Linux operating system and includes a dropper, a launcher, and a backdoor.
-
New Linux Kernel Patch Allows Forcing a CPU Mitigation
Even when CPU mitigations can consume precious CPU cycles, it might not be a bad idea to allow users to enable them, even if your machine isn't vulnerable.
-
Red Hat Enterprise Linux 9.5 Released
Notify your friends, loved ones, and colleagues that the latest version of RHEL is available with plenty of enhancements.
-
Linux Sees Massive Performance Increase from a Single Line of Code
With one line of code, Intel was able to increase the performance of the Linux kernel by 4,000 percent.