This is story is brought to you by Hacker Noon’s weekly sponsor, mabl. ML-driven test automation. mabl, uses machine intelligence to make testing easy
For some, ultimatums set a flame under their rump and motivate them in the moment, but is it a sustainable, healthy way to drive people to change?
“Marry me or I’m leaving you.”
Or is it really just a last-resort tactic for desperate times?
“Do or die.”
We often hear this ultimatum in the software testing world when it comes to automating manual tests:
“Start coding or get left behind.”
Though we understand the intent behind it, forcing an entire industry to evolve on the fly isn’t the only thing we can do to improve automated testing — at least for now, while we wait for the skills to organically catch up with demand.
It’s 2018, and we can now leverage machine learning to create practical solutions to everyday problems, such as overcoming the high learning curve of automating test scripting. mabl uses machine intelligence to allow everyone, developers and testers alike, to create automated, reliable, repeatable tests in seconds.
In an earlier post, we showed you how you can create automated smoke tests without a line of code with mabl. Now we’ll show you the next step — creating custom test flows for critical user journeys.
In the mabl dashboard, we have the options to create a new Journey, create a new application, or invite teammates to our Workspace. We’ll create a new Journey:
In this case we’ll use the mabl marketing website as our application:
Immediately, mabl takes us to our application URL. mabl gives us some tips on how to get the cleanest recordings possible, such as logging out of your app and clearing your cookies. On the right, you can see the mabl trainer, which is a Chrome extension that installs in seconds. Currently, we support Chrome, with more browser support to come.
Let’s create our first custom Journey that tests the Early Access email form on the site:
mabl will begin to record every interaction with your app. Clicks are recorded by default, and hovers and JavaScript animations can be too if enabled. The mabl trainer also allows us to add waits if you have long running processes, add assertions, create randomly generated variables to insert into forms, and find elements by CSS or XPath selectors.
I want to make sure that the email address form validation is working, so I’ll create a randomly generated variable called Email and insert it into the email address form. In the last image below, you can see all the steps leading up to the email form, and our newly created email variable being inserted into the form.
Our Email variable is now in the form, and with a randomly generated string as my email address, I expect to receive a notification label that lets me know my email address is invalid upon form submission:
Let’s add a check for this notification label using the Add Assertion feature in the mabl trainer. The trainer supports many types of asserts. In this case, I’ll apply an equals assertion to check that the notification label equals “Please change your email address to continue.”:
We’ll save this Journey and mabl will automatically create a Plan for the Journey, which dictates the schedule on which Journey will be executed. The default Plan runs the Journey every 30 minutes:
And just like that, we have our first codeless, automated test script under our belts!
Let’s create another simple Journey that checks to make sure the mabl blog is loading properly. This is a good chance for us to use the Find Elements feature in the mabl trainer, since new blogs are published all the time; we don’t want flakey tests, so searching by CSS property instead of by a dynamic blog title would be much more effective:
This time, we’ll navigate to the blog and find a CSS element called .post-header h2. Once that CSS element is found, I’ll apply an assertion that it is present:
I’ll save this Journey and again, a default Plan is created to run the Journey every 30 minutes. Let’s go into the Test Execution History dashboard for our Email Address Form Validation test. We see that it already completed twice by the time I finished writing this blog:
It only took me a couple minutes to create these tests, and now they’ll run on a regular interval in the background. I can configure mabl to notify me of issues via Slack, so I don’t have to hang out in the mabl app if I don’t want to.
So that’s how mabl automatically writes automated tests for you, runs them, and infers actionable insights for you when something goes wrong, such as regressions in performance, newly broken links or broken JavaScript, and more. Now anyone has the ability and time to maintain automated tests for their apps. You can try it too — just sign up for a free mabl account.
This is story is brought to you by Hacker Noon’s weekly sponsor, mabl. ML-driven test automation. mabl, uses machine intelligence to make testing easy
Originally published at www.mabl.com.