[webkit-reviews] review denied: [Bug 3714] REGRESSION: HTMLLabelElement.form returns null : [Attachment 2674] Traverse parent nodes looking for a form element

[bugzilla-request-daemon at opendarwin.org]

Darin Adler <darin at apple.com> has denied Anders Carlsson <andersca at mac.com>'s
request for review:
Bug 3714: REGRESSION: HTMLLabelElement.form returns null
http://bugzilla.opendarwin.org/show_bug.cgi?id=3714

Attachment 2674: Traverse parent nodes looking for a form element
http://bugzilla.opendarwin.org/attachment.cgi?id=2674&action=edit

------- Additional Comments from Darin Adler <darin at apple.com>
A few comments:

    1) The old version of HTMLLabelElementImpl::form has a bug -- it's calling
isGenericFormElement() on this rather than on element. That's fine if we are
rewriting it, but worth noting.
    2) The loop looking for a ID_FORM in parents is the same as the existing
getForm() function; we should not write a new copy of that function. (Although
perhaps we should rename it.)
    3) I don't think this is significantly different from the form function in
the base class. Can we just remove the override of the base form class's form
function?

Do we have a test for the case where the <label> element is outside the form
and the form element it points to is inside? What about when the <label> is in
one form, and the form element is in another?