Oilpan: Prepare to move UA shadow elemnts for date/time input types to Oilpan heap.

Oilpan: Prepare to move UA shadow elemnts for date/time input types to Oilpan heap.

This CL is a preparation to move the following classes:

- DateTimeEditElement

- DateTimeEditElement::EditControlOwner as GarbageCollectedMixin
  BaseMultipleFieldsDateAndTimeInputType implements it.

- DateTimeFieldElement and its subclasses

- DateTimeFieldElement::FieldOwner as GarbageCollectedMixin
  DateTimeEditElement implements it.

BUG=357163
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=173533

[tkent, 2014-05-07 09:55:50]

Please review this.

[haraken, 2014-05-07 10:19:46]

LGTM

[haraken, 2014-05-07 10:19:58]

https://codereview.chromium.org/267413002/diff/1/Source/core/html/shadow/Date...
File Source/core/html/shadow/DateTimeEditElement.cpp (right):

https://codereview.chromium.org/267413002/diff/1/Source/core/html/shadow/Date...
Source/core/html/shadow/DateTimeEditElement.cpp:459: // We can't trace
Vector<RawPtr<T>>.

Nit: We normally don't have this kind of comment.

[zerny-chromium, 2014-05-07 10:44:13]

lgtm

https://codereview.chromium.org/267413002/diff/1/Source/core/html/shadow/Date...
File Source/core/html/shadow/DateTimeFieldElement.h (right):

https://codereview.chromium.org/267413002/diff/1/Source/core/html/shadow/Date...
Source/core/html/shadow/DateTimeFieldElement.h:80: virtual void trace(Visitor*);
Nit: OVERRIDE

[tkent, 2014-05-07 14:12:54]

The CQ bit was checked by tkent@chromium.org

[commit-bot: I haz the power, 2014-05-07 14:13:59]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/tkent@chromium.org/267413002/40001

[commit-bot: I haz the power, 2014-05-07 14:43:00]

Change committed as 173533

[Mads Ager (chromium), 2014-05-07 14:55:08]

This looks great. Can we use strong owner pointers in this hierarchy?

(I think we already are because these elements are in the same shadow subtree
and therefore they are all traced together.)

https://codereview.chromium.org/267413002/diff/40001/Source/core/html/shadow/...
File Source/core/html/shadow/DateTimeEditElement.h (right):

https://codereview.chromium.org/267413002/diff/40001/Source/core/html/shadow/...
Source/core/html/shadow/DateTimeEditElement.h:154:
RawPtrWillBeWeakMember<EditControlOwner> m_editControlOwner;
Can this be a Member instead of a WeakMember? I looks like this entire structure
should always live and die together? I think we can just have Member references
from owner to child and from child to owner. Then this entire structure will die
in the same GC round and there is no need for weak processing.

https://codereview.chromium.org/267413002/diff/40001/Source/core/html/shadow/...
File Source/core/html/shadow/DateTimeFieldElement.h (right):

https://codereview.chromium.org/267413002/diff/40001/Source/core/html/shadow/...
Source/core/html/shadow/DateTimeFieldElement.h:103:
RawPtrWillBeWeakMember<FieldOwner> m_fieldOwner;
Ditto for this one? It would be great if this could just be a Member (and I
think it can?)

[tkent, 2014-05-08 01:11:36]

https://codereview.chromium.org/267413002/diff/40001/Source/core/html/shadow/...
File Source/core/html/shadow/DateTimeEditElement.h (right):

https://codereview.chromium.org/267413002/diff/40001/Source/core/html/shadow/...
Source/core/html/shadow/DateTimeEditElement.h:154:
RawPtrWillBeWeakMember<EditControlOwner> m_editControlOwner;
On 2014/05/07 14:55:08, Mads Ager (chromium) wrote:
> Can this be a Member instead of a WeakMember? I looks like this entire
structure
> should always live and die together? I think we can just have Member
references
> from owner to child and from child to owner. Then this entire structure will
die
> in the same GC round and there is no need for weak processing.

Right.  It can be a Member.  I'll make a CL.
It's a weak pointer only if !ENABLE(OILPAN).