Bugzilla

The Layout team has proposed that this method become a method on InspectorUtils, instead of on Element. The reasoning is that the necessary analysis can be done after the frame tree is constructed and doesn't need to be done during reflow itself. The implementation in LayoutUtils will be easier to structure as a frame tree traversal.

Additionally, the Layout team has made a compelling argument that the method should return *all* child elements that overflow the visible area.

Manas, the easiest thing to do is to mark the existing patch abandoned on Phabricator, and build a new patch based around changes to InspectorUtils.webidl, .h, and .cpp. The webidl signature will be `sequence<Element> getOverflowingChildren();` and won't need the `[ChromeOnly, Pure]` qualifier ahead of the declaration since all of InspectorUtils is chrome-only.

The Layout team has proposed that this method become a method on InspectorUtils, instead of on Element. The reasoning is that the necessary analysis can be done after the frame tree is constructed and doesn't need to be done during reflow itself. The implementation in LayoutUtils will be easier to structure as a frame tree traversal.

Additionally, the Layout team has made a compelling argument that the method should return *all* child elements that overflow the visible area.

Manas, the easiest thing to do is to mark the existing patch abandoned on Phabricator, and build a new patch based around changes to InspectorUtils.webidl, .h, and .cpp. The webidl signature will be `NodeList getOverflowingChildren();` and won't need the `[ChromeOnly, Pure]` qualifier ahead of the declaration since all of InspectorUtils is chrome-only.

The Layout team has proposed that this method become a method on InspectorUtils, instead of on Element. The reasoning is that the necessary analysis can be done after the frame tree is constructed and doesn't need to be done during reflow itself. The implementation in LayoutUtils will be easier to structure as a frame tree traversal.

Additionally, the Layout team has made a compelling argument that the method should return *all* child elements that overflow the visible area.

Manas, the easiest thing to do is to mark the existing patch abandoned on Phabricator, and build a new patch based around changes to InspectorUtils.webidl, .h, and .cpp. The webidl signature will be `[NewObject] NodeList getOverflowingChildren();` and won't need the `[ChromeOnly, Pure]` qualifier ahead of the declaration since all of InspectorUtils is chrome-only.

The Layout team has proposed that this method become a method on InspectorUtils, instead of on Element. The reasoning is that the necessary analysis can be done after the frame tree is constructed and doesn't need to be done during reflow itself. The implementation in LayoutUtils will be easier to structure as a frame tree traversal.

Additionally, the Layout team has made a compelling argument that the method should return *all* child elements that overflow the visible area.

Manas, the easiest thing to do is to mark the existing patch abandoned on Phabricator, and build a new patch based around changes to InspectorUtils.webidl, .h, and .cpp. The webidl signature will be `[NewObject] NodeList getOverflowingChildrenOfElement(Element element); ` and won't need the `[ChromeOnly, Pure]` qualifier ahead of the declaration since all of InspectorUtils is chrome-only.