Hi,
We are doing automated browser compatibility using selenium as an automation tool.The approach we are following are
We first decide a base browser let say IE 10 for internet explorer and then target browser starting from IE8 to IE11 and then we are picking every link /input on the page and find coordinate (x,y) width and height in base browser(IE10 in this case) and then we do similar activity for base browser and comparing the result.
During our comparative study of the result we found that these parameter never matches :) and it has variances of 5 to15 pixel in most case and in worst scenario more than 20.
When we took screenshot manually and did comparison in Photoshop the difference was negligible where selenium was showing variance of 15 or more pixel.
Is this a correct way of doing browser compatability(visual only) and why selenium is not calculating width height left right correctly or we are missing something.
PS We are doing similar excercise for chrome/firefox/safari
Any help/suggestion is highly appreciated.
Thanks!!!
Roshan