[Webkit-unassigned] [Bug 34382] When a live iframe element is moved between pages, it still depends on the old page.

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=34382





--- Comment #28 from Dmitry Titov <dimich at chromium.org>  2010-02-12 01:54:49 PST ---
(In reply to comment #27)

> I'm surprised that you call transferChildFrameToNewDocument on each
> child frame of the frame being reparented.  I'm guessing this is b/c
> you want to have the WebKit layer update the WebFrameClients for each
> of the child frames.  That might be better left to the FrameLoaderClient
> implementation since it will know if that is necessary.  As is, the
> notification isn't really correct:  the child frame didn't get moved to
> a new document, and it didn't get reparented.

Child Frames should get a new Page pointer. This is the main reason to iterate
child frames in Frame::transferChildFrameTonewDocument. So this loop exists for
both WebCore and WebKit reasons and as such it's reasonable to have it in Frame
rather then in FrameLoaderClient.

Speaking about WebFrameClient notification, I see it's a bit awkward . Do we
really need it (in chromium WebKit layer)? What would we do in it? The
signature you proposed:

void didReparent(WebFrame*, WebFrame* oldParent);

doesn't also look ideal because for child frames it doesn't make a lot of sense
- the child frames are not really reparented, and their parent frame does not
change.

How about we just don't have it until we have a need for it?

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