In the thesis, we analyzed and compared the old and new automated testing frameworks for web ads at Celtra d.o.o. The comparison focused on key aspects such as time efficiency, test stability, and unreliable tests. In the new framework, we observed faster test execution and better support for parallelization due to the transition to the cloud service BrowserStack, which allows testing on multiple devices simultaneously. Additionally, the new framework uses CSS locator checks instead of visual comparisons, which contributed to greater stability. Despite initial challenges, the new framework achieved comparable, and in some cases, even better stability than the old one after its implementation.
|
