User Agent: Mozilla/5.0 (Windows NT 6.2; WOW64; rv:22.0) Gecko/20130330 Firefox/22.0 Build ID: 20130330030828 Steps to reproduce: 1. add an event listener for UI event (for example: "click" event) to an element by using `addEventListener` 2. emit event (for example: clicking the element) 3. remove the element (`evt.currentTarget`) from DOM tree in the event listener 4. check the `evt.view` property in the event listener (`evt` is an argument of event listener) please see: http://jsfiddle.net/nobuoka/qvJfH/ (sample code) Actual results: The value of `evt.view` is `null` (on Firefox 22 nightly 2013-03-30) Expected results: The value of `evt.view` is defaultView (window object) see: http://www.w3.org/TR/DOM-Level-3-Events/#events-UIEvent-view It says "The view attribute identifies the AbstractView from which the event was generated.", so I think it must not changed even when the eventTarget is removed from DOM tree.

Sorry, comment #0 is wrong. Instead, see below. Steps to reproduce: 1. add an event listener for UI event (for example: "click" event) to an element by using `addEventListener` 2. emit event (for example: clicking the element) 3. read the `evt.view` property in the event listener (it should be correct value) 4. read the `evt.view` property again (`evt` is an argument of event listener) please see: http://jsfiddle.net/nobuoka/qvJfH/ (sample code) Actual results: The value of `evt.view` is `null` (on Firefox 22 nightly 2013-03-30) Expected results: The value of `evt.view` is defaultView (window object) see: http://www.w3.org/TR/DOM-Level-3-Events/#events-UIEvent-view

Regression range: http://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=e23e43a2c14e&tochange=8156df33b757 I'll narrow this down.

100 already_AddRefed<nsIDOMWindow> GetView() 101 { 102 nsCOMPtr<nsIDOMWindow> view = mView; 103 return mView.forget(); 104 }

uh, sorry. Patch coming.

Comment on attachment 732378 [details] [diff] [review] patch Review of attachment 732378 [details] [diff] [review]: ----------------------------------------------------------------- I don't know why this needs to return already_AddRefed, but this works.

Comment on attachment 732378 [details] [diff] [review] patch [Approval Request Comment] Bug caused by (feature/regressing bug #): bug 847585 User impact if declined: wrong behavior Testing completed (on m-c, etc.): landed to m-i Risk to taking this patch (and alternatives if risky): super-safe String or IDL/UUID changes made by this patch: NA