Acceptance Tests

Acceptance tests are tests from the perspective of the end-user.

In the Rails world, we sometimes refer to these as Feature or System tests. A tool called Capybara is included to help us simulate a human's actions inside of our tests.

Generally, we are simulating what a user could do from their web browser and ensuring that the app behaves as intended. When a feature is heavily reliant on interaction from a user via the browser, it's a good idea to write automated Acceptance tests to uncover any bugs that might not be apparent from manual testing.

It's important to note that Rails System tests can be fairly slow, so it's better to focus on testing core functionality or pieces of your code that you think might be prone to regressions. Another strategy we use to help us keep acceptance tests fast is :aggregate_failures. :aggregate_failures allows us to make multiple assertions within a single test. The flag signals to Rspec to run each assertion and then report all of the failures back to us rather than just the first failure. This saves time because we only have to setup the conditions for the spec once and then make all of our assertions instead of setting up the conditions repeatedly for each individual assertion.

Acceptance tests can be found in the directory spec/system.

You can run all acceptance tests with:

1
bundle exec rspec spec/system

To run an individual file you can use:

1
bundle exec rspec spec/system/user_views_a_reading_list_spec.rb

To run a specific test case you can use:

1
bundle exec rspec spec/system/user_views_a_reading_list_spec.rb:10

where 10 is the line number of the test case that you want to execute.

You can read the official guide Testing Rails Applications to learn more.