Turn your manual testers into automation experts Request a Demo Start testRigor Free

Mojolicious Testing

Mojolicious is an open-source full-stack web development framework for the Perl programming language. It was created by Sebastian Riedel in 2009 and is designed to be user-friendly, powerful, and flexible. Mojolicious utilizes an MVC (Model-View-Controller) approach for development. The framework provides many tools and features to aid developers in building web applications quickly and easily. These include a robust routing engine, a built-in web server, template rendering, session management, and support for web sockets, among others.

A key feature of Mojolicious is its emphasis on real-time web development, with built-in support for web sockets and server-sent events. This feature makes it an excellent fit for building interactive web applications, such as chat applications, real-time analytics dashboards, and more. It is also renowned for its modern and expressive syntax, making it easy to write clean and maintainable code. The framework is fully compatible with the latest versions of Perl and has a thriving community of developers and users who contribute to its development and support.

You can test Mojolicious code using the following methods:

Unit testing
Integration testing
End-to-end testing

The goal of unit tests is to inspect various individual units in your code. These units could be part of the model, view, controller, route, or middleware. The main goal is to focus on individual methods and verify their outcomes using techniques such as mocking. Checking scenarios like positive and negative flows, and error-handling cases will help you create high-quality test cases.

Testing using the Test framework

The Test framework offers many helper modules that facilitate the testing of Perl code. Some examples of Test modules in Perl include Test::Exception, Test::MockObject, Test::Deep, and Test::WWW::Mechanize, among many others.

An example of using the Test framework is as follows. The test commences by setting the test plan using the plan() method, specifying the number of tests to be run (14) and a list of tests to be skipped (tests 3 and 4). It then loads the MyModule module and prints a helpful note indicating the module's version being tested.

use strict;
use Test;

# use a BEGIN block so we print our plan before MyModule is loaded
BEGIN { plan tests => 14, todo => [3,4] }

# load your module...
use MyModule;

print "# I'm testing MyModule version $MyModule::VERSION\n";

ok(0); # failure
ok(1); # success

ok(0); # ok, expected failure (see todo list, above)
ok(1); # surprise success!

# and so on...

Next, we'll explore some of the modules that the Test framework offers to assist with testing.

Testing using Test::Mojo

Test::Mojo comes bundled with Mojolicious, offering a means to write tests that simulate HTTP requests and responses to your application. It makes it easy to test controllers, routes, and templates, and is typically used in conjunction with Test::More.

Here is an example of a test case using Test::Mojo:

use Mojo::Base -strict;

use Test::Mojo;
use Test::More;

# Start a Mojolicious app named "Celestial"
my $t = Test::Mojo->new('Celestial');

# Post a JSON document
$t->post_ok('/notifications' => json => {event => 'full moon'})
  ->status_is(201)
  ->json_is('/message' => 'notification created');

# Perform GET requests and look at the responses
$t->get_ok('/sunrise')
  ->status_is(200)
  ->content_like(qr/ am$/);
$t->get_ok('/sunset')
  ->status_is(200)
  ->content_like(qr/ pm$/);

# Post a URL-encoded form
$t->post_ok('/insurance' => form => {name => 'Jimmy', amount => '€3.000.000'})
  ->status_is(200);

# Use Test::More's like() to check the response
like $t->tx->res->dom->at('div#thanks')->text, qr/thank you/, 'thanks';

done_testing();

The above test makes several HTTP requests to the application using Test::Mojo's testing methods, such as post_ok, get_ok, status_is, content_like, and json_is. The responses are then verified using Test::More's like method. The done_testing() method at the end of the script signifies the completion of the unit tests, and the number of tests executed is automatically determined by the testing framework.

Testing using Test::More

Test::More offers a simple yet powerful API for writing and executing tests. It includes an array of methods for making assertions about the behavior of your code, such as ok(), is(), like(), cmp_ok(), and isa_ok(). These methods allow developers to test a wide range of conditions, from simple equality checks to more complex object comparisons. Test::More also includes features for test planning, output formatting, and error reporting.

Testing using Test::Unit

Test::Unit is another module used for testing Perl code. However, for Mojolicious, Test::Mojo is generally the preferred choice. If you're accustomed to Test::Unit, you can certainly write your tests using this module.

Integration Testing

Integration testing focuses on verifying the interaction and communication between different components or modules of a software system. It is performed after unit testing, where individual units or components are tested in isolation. The purpose of integration testing is to expose defects in the interfaces and interactions between these units when they are integrated.

The main objective of integration testing is to ensure that the integrated components work together correctly as a whole and fulfill the intended functionality and requirements. It aims to identify any issues such as incompatible interfaces, data inconsistencies, communication failures, or incorrect assumptions made by the individual components.

These tests are expected to be lighter weight compared to end-to-end tests, and hence, you can use mocking or data fixtures in place of the real deal to simulate your integration tests. However, certain parts of the test, such as the modules being tested for integration, should not be simulated.

You can write your integration tests using the Test::Mojo modules. Some interesting modules available within this framework are:

Mojo::Client: This module allows you to make requests to URLs, handling the low-level details of setting up the connection, sending the request, and managing the response. It offers various methods for making different types of requests, including GET, POST, PUT, DELETE, and PATCH. Additionally, it also supports WebSocket connections using the websocket() method.
Mojo::DOM: Built on top of the Mojo::DOM::HTML and Mojo::DOM::XML parsers (which are in turn based on the Mojo::DOM::Tree module), Mojo::DOM can manage different markup languages and encodings, automatically detecting and repairing errors in the input document. With Mojo::DOM, you can select elements in an HTML or XML document using CSS-like selectors and then manipulate those elements by adding or modifying attributes, changing their text content, or even adding or removing entire nodes.
Test::WWW::Mechanize::Mojo: This module enables you to test Mojolicious applications using the WWW::Mechanize interface, incorporating features of web application testing. It supports various HTTP methods and offers methods for checking the status code, headers, and content of the response. Importantly, the module doesn't require you to launch a server or make real HTTP requests.
Test::Mojo::Role::*: With roles, you can extend the Test::Mojo module with additional functionalities. Some examples of available roles include Test::Mojo::Role::Session, Test::Mojo::Role::Debug, and Test::Mojo::Role::SubmitForm.

Here's an example of testing client-server interaction using web sockets. This test uses the Test::Mojo client's websocket() method to simulate a WebSocket connection with the server and to test the messages exchanged between the client and the server. The code checks the contents of the messages received from the server using the like() method, which is an assertion typically used in integration testing to verify that a response matches a particular pattern. The code also keeps track of the number of messages received using the $req_number variable, which is used to perform different checks on different messages.

my $req_number = 0;
$client->websocket(
  # First request
  '/' => sub {
    my $self = shift;
    $self->receive_message(
      sub {
        my ($self, $message) = @_;
        if ($req_number == 0) { # first request
          like($message, qr/{"text":"Guest\d+ has joined chat"}/);
        } elsif {
        ...
        # end of last message 
        $self->finish;
      }
    $req_number++;
  });

This example uses the websocket() method to simulate a WebSocket connection and perform integration testing on client-server interaction.

End-to-End Testing

When discussing end-to-end testing, the focus becomes ensuring the operability of business scenarios for the user. This involves interactions with the UI and server launching. Applications built with Mojolicious can be assessed from a user's perspective using the following testing tools:

Test::Mojo's modules
testRigor

Test::Mojo's modules

Test::Mojo provides modules for writing and testing end-to-end scenarios. Some of these modules include Test::Mojo::Role::Phantom, Test::Mojo::Role::Selenium, Test::Mojo::Role::JSON, Test::Mojo::Role::XML, Test::Mojo::Role::Content, Test::Mojo::Role::Header, Test::Mojo::Role::UA, Test::Mojo::Role::WebSocket, and Test::Mojo::Role::Moai, which are used for different testing purposes.

For instance, here is a Selenium test case example. This test script uses the Test::Mojo::WithRoles module for end-to-end testing of a Mojolicious application. The test verifies the current URL and checks for the presence of the "GUIDES" link on the page. The test manipulates the form on the page and checks the URL's evolution and the value of the search input field.

use Mojo::Base -strict;
use Test::Mojo::WithRoles "Selenium";
use Test::More;

$ENV{MOJO_SELENIUM_DRIVER} ||= 'Selenium::Chrome';

my $t = Test::Mojo::WithRoles->new->setup_or_skip_all;

$t->set_window_size([1024, 768]);

$t->navigate_ok('/perldoc');
$t->current_url_is("http://mojolicious.org/perldoc");
$t->live_text_is('a[href="#GUIDES"]' => 'GUIDES');

$t->driver->execute_script(qq[document.querySelector("form").removeAttribute("target")]);
$t->element_is_displayed("input[name=q]")
  ->send_keys_ok("input[name=q]", ["render", \"return"]);

$t->wait_until(sub { $_->get_current_url =~ qr{q=render} });
$t->live_value_is("input[name=search]", "render");

done_testing;

testRigor

While Mojolicious offers built-in support for web testing, implementing complex business scenarios can be challenging. For those unfamiliar with Perl or coding, such as manual testers and business analysts, a more accessible tool like testRigor may be necessary.

testRigor is a no-code cloud-platform that supports web, mobile and desktop testing. The scripts are written in plain English, thus eliminating the hassle of writing code. Moreover, test runs are so stable that you can even use them to monitor your application's health. Known for its ease of test maintenance, it is a great tool for end-to-end automation testing. It comes loaded with features like visual testing, API testing, referencing UI elements using relative locations (instead of XPaths or other locators), generating test data, accessing databases, audio testing, and many more. It comes with support for third-party integrations with various CI/CD tools, test case management, infrastructure and databases.

Here is the testRigor version of the end-to-end test written using Test::Mojo's Selenium module. You can see how easily one can write the test steps, without worrying about coding. However, if you do wish to write code at certain points like JavaScript for certain UI interactions or properties of UI locators, then you still have the flexibility to do so.

check that “GUIDES” is clickable 
execute JavaScript in the browser text starting from next line and ending with [END]
document.querySelector("form").removeAttribute("target");
[END]
check if page contains “Search…” 
enter “render” in “Search…” 
enter enter
grab url and save it as "URL"
check that stored value "URL" itself contains "q=render"
check that input “Search” has value “render”

Conclusion

No application is entirely immune to bugs, even one as powerful as Mojolicious. Therefore, it's important to have a robust testing strategy in place. The tools discussed above are among the best available in the market and will likely deliver the desired results.

Request a Demo

Start testRigor Free