[Webkit-unassigned] [Bug 30271] [QT] DRT doesn't support double click simulation

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Mon Oct 12 10:34:43 PDT 2009


https://bugs.webkit.org/show_bug.cgi?id=30271





--- Comment #6 from Jakub Wieczorek <faw217 at gmail.com>  2009-10-12 10:34:43 PDT ---
(In reply to comment #5)
> 2. EventServer,  as an Javascript object,  will be instantiated for each test
> page,  two or more subsequent tests will each has its own instance of
> EventSender,  so mouse clicks in 2 test cases won't create a double click event
> in the second test, instead,  2 single-clicks will be created , one for each
> test case .

Actually, all the controller objects, that are exposed to JS, are created only
once in each DRT instance, which is reused as long as possible until one of the
tests crashes or a specific limit of subsequent tests is reached.

Qt DRT uses the QObject-JSC bridge and it needs to readd the relevant objects
to the window object after it is cleared but it does not reinstantiate them.

> 1. class LayoutTestController and EventSender are two totally different classes
> which don't have any connection , it makes no sense to reset EventSender inside
> LayoutTestController.

Yes, I agree. How about DRT::resetToConsistentStateBeforeTesting()? (if that
means making DRT a friend of EventSender, this should be ok I guess).

I believe it may be a good idea to look at the layout tests that Qt is passing
and that use eventSender.leapForward() (which is not implemented in Qt DRT) and
check if there are any regressions. Now that the EventSender would support
double click events, the lack of that feature might introduce some new
failures.
fast/events/frame-click-focus.html might be one example.

I'm not a reviewer so please take it all as suggestions.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.



More information about the webkit-unassigned mailing list