The boolean member variables of HitTestResult (readonly, active, and mouseMove) are not a part of the RESULT of the hit test. Rather, they are more like conditions for conducting the hit test. So it does not really make sense that they are part of the HitTestResult class.
Created attachment 11289 [details] Creates a struct for the request The thing that I dislike about this patch is that it adds a new boolean parameter to all of the nodeAtPoint()s. The only nodeAtPoint() implementation to use any of the fields of HitTestRequst is RenderFrameSet::nodeAtPoint() which asks about readonly. So instead of sending the entire HitTestRequest into all of the nodeAtPoint()s, I just sent the readonly value. I wish I didn't have to do that since no one else needs it, but it doesn't seem like there is another way. Or is there??
Comment on attachment 11289 [details] Creates a struct for the request I'd suggest passing the whole HitTestRequest to the nodeAtPoint methods even though they only need the one bool, just for consistency r=me either way
Committed with r17494.