Standard window objects may be present in an application under test, but UFT may not recognise them. In certain cases, objects of the type button, link, or other can be declared as virtual objects (VO) so that user actions can be mimicked on the virtual objects during execution.
For example : Let's pretend we're automating a Microsoft Word scenario. We open the MS Word application and click on any of the ribbon icons. For example, on the Ribbon, the user selects the Insert tab, then selects the "Picture" button. Here, a button is recognised as a WinObject, emphasising the relevance of virtual objects.
Following are the limitations of virtual objects in UFT :
* For analog or low-level recording, QTP does not allow virtual objects.
* Virtual Objects cannot have checkpoints added to them.
* Object Repository has no control over Virtual Objects.
* Although we map an object to a specific class (button or List), Virtual objects do not provide all of the original object's methods.
* Virtual Objects are not supported by Object Spy.
* If the screen resolution changes as the coordinates change, the test will fail.
* Virtual objects should be recorded correctly if the application window is the same size as the screen.
