[Webkit-unassigned] [Bug 61322] [Qt] Selectstart event tests added by r87096 are failing on Qt

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=61322





--- Comment #9 from Robert Hogan <robert at webkit.org>  2011-05-25 15:09:46 PST ---
Looking at EventSenderQt I can't really see a good reason for matching other ports' behaviour of introducing a delay in the event queue when leapForward() is called. The purpose of leapForward() (if I understand it correctly) is to introduce a delay long enough to simulate a drag event to the port's toolkit.

This isn't necessary for Qt, which lets the application decide when to create a drag event loop. Removing the delay in leapForward() (and replacing it with m_drag=true) allows the remaining two failing tests to pass and causes only one regression in the layout tests.

This can be fixed in the test itself:

[22:33] <rniwa> editing/selection/select-out-of-editable.html?
[22:33] <mwenge> rniwa:yup
[22:33] <rniwa> mwenge: oh you can remove the second mouseMoveTo, mouseDown, and leapForward now
[22:34] <mwenge> ok, why's that?
[22:34] <rniwa> mwenge: there was a bug in EventHandler
[22:34] <rniwa> mwenge: luckily, evmar found a reduction for the bug and I fixed it a couple of weeks ago :)
[22:34] <rniwa> mwenge: http://trac.webkit.org/changeset/81053
[22:34] <rniwa> mwenge: after this revision, the test shouldn't require additional moveMove

Yael, what are your thoughts on this? OK to remove the delay?

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