Jeannette Smith - 17 January 2018
Wow! It’s been quite a year in QA at Testplant. We’ve implemented so many new, big features, providing even more ways to expand our quiver of testing solutions—and yours.
We added API and Selenium testing capabilities to our already powerful, intelligent image and text recognition and search. It’s been really fun and interesting to use them all, see their different strengths, and identify where there are advantages to one over the other. For example, while there are benefits to being able to directly specify web elements with SenseTalk and our Selenium WebDriver, when you’re new to Selenium, learning how to find and specify the element you want can be a challenge. Or, for web pages with frames, our image-based searches are much easier. We continue to learn more about all of these. We also added a roaming license capability for our RLM License Servers. This is a really handy feature for us when we’re traveling or sitting on a plane.
In addition to enhancing our product in cool, new ways, we’ve been automating more of our internal tests. As testers of an automation software solutions, we write tests that automate our own automation software, which definitely helps us better understand what it’s like to be one of our customers. As part of this, we’ve been expanding our regression suite: Our own Eggplant Manager and Eggplant Automation Cloud solutions are, and have been, critical to effectively managing our regression suite execution.
When I started to write this blog, someone asked me how we use Eggplant Manager and Eggplant Automation Cloud in our regression tests. So, I thought I’d pass along that information. We actually use these solutions to automate the whole execution process:
- We develop our scripts in Eggplant Functional against the SUTs in our Eggplant Automation Cloud.
- We check our tests into source control, and then check them out into the tests we define in Eggplant Manager.
- We create Eggplant Manager tests that typically run multiple scripts, specifying the order in which they run to accommodate dependencies.
- In the test definition, we can specify the SUT at the Eggplant Manager test level or at the individual script level when the test requires a SUT. Eggplant Automation Cloud reserves the SUTs that our automated tests use when they run. Also, in the test definition, we specify which Eggplant Manager agent should run the test. The agents are installed on VMs of different operating systems. This enables us to achieve our cross-platform coverage on all the supported Eggplant Functional platforms.
- The tests run nightly as cron jobs, but can also be run on demand at any time.
- We receive email notifications of the status of the test runs, and the results are uploaded to our test case management system.
- Then, we investigate any failures by navigating to the test results in Eggplant Manager. There, we can see where the failure occurred, down to which command in which script. If the test runs on a SUT, we even get a screenshot of the SUT when the failure occurred.
Finally, we’ve been improving and defining our internal test processes, especially around test results reporting and metrics gathering. We implemented a new test case management system to which we now upload our test results from Eggplant Manager. This is giving us a clearer picture of where we are overall, where we need to improve, and what we’re doing well. Armed with that information, we can continue to implement new capabilities and make enhancements to provide you with a high-quality testing suite.
So, with lots of work still to be done, we should get cracking (as our UK colleagues like to say). On behalf of all of us on the Eggplant Functional test team, we wish you a happy and productive new testing year.P.S. If you haven’t already, check out our Community page for an Eggplant Functional interactive guide, training and certification courses, our user forum, and more.