I've been creating a test suite using Keyword based tests and a variety of region checkpoints. Most are created via a window region, some are created with the finder tool - both are giving me headaches.
Neither type of region seems to pass consistently, the biggest problem I have is that the region will pass just fine one day (over the course of many runs during that day), and the next day using the exact same build and exact same test script with the exact same regions it will fail. The most common text in the 'Remarks' field is "The regions are not equal. The size of the images is different".
Can someone help me? This is driving me absolutely crazy!!
First of all, please check whether the problem can be reproduced with TestComplete 8.10 which has been released recently - a lot of problems were fixed in this version. If this suggestion does not help, please zip your entire project suite folder along with the log of the failed test execution and send us the archive via our Contact Support form (http://www.automatedqa.com/support/message). Make sure that messages in your log correspond to the latest version of your tests.
I recently upgraded to TestComplete 8.10 and have narrowed down the root of the problem. The issue was that I was not familiar with the comparison mechanics of TestComplete; I expected TestComplete to search the screen and find the image it's looking for - but instead it searches ONLY in the area in which the image was originally captured. This is leading to some inconsistencies with me because the image may be a few pixels to the left or right of its original location depending on the date (12/10/2010 pushes the image farther to the right than 12/9/2010).
I understand there is a FindRegion function in TestComplete that may solve my problem - but I stay away from scripted tests as much as possible due to my extremely limited coding knowledge. Is there a way for me to use the FindRegion function with a Keyword test?
btw BINGONOW - I am using the same computer every day.