[Chromium] REGRESSION(r121909): m_currentInputEvent never set
Created attachment 151539 [details]
Created attachment 151541 [details]
Can we manufacture a test verifying the correct behavior of window.open?
(In reply to comment #3)
> Can we manufacture a test verifying the correct behavior of window.open?
That would be great, but I'm not sure this can be done as a layout test. Does DRT expose anything regarding "background" vs "foreground" "tabs"?
Landing for now, as the code it's fixing is brand new and clearly wrong. Will poke around to see about adding a test in the meantime.
Thx for catching & fixing. Chat w/ Darin indicates that this is a painful bit of code to test in webkit, as DRT likely doesn't have concepts such as tabs, or logic to have it be foreground or background. Easiest test coverage may be via a browser test. I've created a chrome issue for that, since we don't know where the test should be implemented yet. https://code.google.com/p/chromium/issues/detail?id=136614
Committed r122284: <http://trac.webkit.org/changeset/122284>