Oilpan: have DocumentInit use transition types throughout.

Oilpan: have DocumentInit use transition types throughout.

For DocumentInit, the context document is no longer mediated
as a WeakPtr, but as a Member. It is sufficient that the
Document holds onto this reference weakly.

R=haraken@chromium.org, tkent@chromium.org
BUG=357163
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=175260

[sof, 2014-06-02 06:39:44]

Please take a look.

https://codereview.chromium.org/304323010/diff/30001/Source/core/dom/Document...
File Source/core/dom/Document.cpp (right):

https://codereview.chromium.org/304323010/diff/30001/Source/core/dom/Document...
Source/core/dom/Document.cpp:5824: visitor->trace(m_contextDocument);
The alternative is to do the clearing of this WeakMember manually in
clearWeakMembers(). I'm not sure which one we prefer if a weak callback is
otherwise required.

[haraken, 2014-06-02 07:17:03]

LGTM

https://codereview.chromium.org/304323010/diff/30001/Source/core/dom/Document...
File Source/core/dom/Document.cpp (right):

https://codereview.chromium.org/304323010/diff/30001/Source/core/dom/Document...
Source/core/dom/Document.cpp:5824: visitor->trace(m_contextDocument);
On 2014/06/02 06:39:44, sof wrote:
> The alternative is to do the clearing of this WeakMember manually in
> clearWeakMembers(). I'm not sure which one we prefer if a weak callback is
> otherwise required.

Either is fine with me, but I'd prefer the approach in this CL because the
overhead of weak processing won't matter in this case.

https://codereview.chromium.org/304323010/diff/30001/Source/core/dom/Document...
File Source/core/dom/DocumentInit.h (right):

https://codereview.chromium.org/304323010/diff/30001/Source/core/dom/Document...
Source/core/dom/DocumentInit.h:87: WeakPtr<Document> m_contextDocument;

Can we introduce WeakPtrWillBeMember?

[sof, 2014-06-02 07:20:05]

https://codereview.chromium.org/304323010/diff/30001/Source/core/dom/Document...
File Source/core/dom/DocumentInit.h (right):

https://codereview.chromium.org/304323010/diff/30001/Source/core/dom/Document...
Source/core/dom/DocumentInit.h:87: WeakPtr<Document> m_contextDocument;
On 2014/06/02 07:17:03, haraken wrote:
> 
> Can we introduce WeakPtrWillBeMember?

I considered it too specialized & a one-off occurrence, but perhaps it is worth
doing so?

[haraken, 2014-06-02 07:26:46]

https://codereview.chromium.org/304323010/diff/30001/Source/core/dom/Document...
File Source/core/dom/DocumentInit.h (right):

https://codereview.chromium.org/304323010/diff/30001/Source/core/dom/Document...
Source/core/dom/DocumentInit.h:87: WeakPtr<Document> m_contextDocument;
On 2014/06/02 07:20:06, sof wrote:
> On 2014/06/02 07:17:03, haraken wrote:
> > 
> > Can we introduce WeakPtrWillBeMember?
> 
> I considered it too specialized & a one-off occurrence, but perhaps it is
worth
> doing so?

Either is fine, but I might to use it in modules/websocket :)

[sof, 2014-06-02 07:31:04]

https://codereview.chromium.org/304323010/diff/30001/Source/core/dom/Document...
File Source/core/dom/DocumentInit.h (right):

https://codereview.chromium.org/304323010/diff/30001/Source/core/dom/Document...
Source/core/dom/DocumentInit.h:87: WeakPtr<Document> m_contextDocument;
On 2014/06/02 07:26:47, haraken wrote:
> On 2014/06/02 07:20:06, sof wrote:
> > On 2014/06/02 07:17:03, haraken wrote:
> > > 
> > > Can we introduce WeakPtrWillBeMember?
> > 
> > I considered it too specialized & a one-off occurrence, but perhaps it is
> worth
> > doing so?
> 
> Either is fine, but I might to use it in modules/websocket :)

Thanks, let's add it here then :)

[sof, 2014-06-02 07:35:40]

wtf OWNERS: added a WeakPtr(std::nullptr_t) constructor here -- acceptable?

[tkent, 2014-06-02 07:56:06]

lgtm

https://codereview.chromium.org/304323010/diff/50001/Source/core/dom/Document...
File Source/core/dom/DocumentInit.h (right):

https://codereview.chromium.org/304323010/diff/50001/Source/core/dom/Document...
Source/core/dom/DocumentInit.h:50: explicit DocumentInit(const KURL& = KURL(),
LocalFrame* = 0, WeakPtrWillBeRawPtr<Document> = nullptr, HTMLImportsController*
= 0);
WeakPtr(std::nullptr_t) is necessary because RawPtr needs nullptr initialization
here, right?

[sof, 2014-06-02 08:01:34]

https://codereview.chromium.org/304323010/diff/50001/Source/core/dom/Document...
File Source/core/dom/DocumentInit.h (right):

https://codereview.chromium.org/304323010/diff/50001/Source/core/dom/Document...
Source/core/dom/DocumentInit.h:50: explicit DocumentInit(const KURL& = KURL(),
LocalFrame* = 0, WeakPtrWillBeRawPtr<Document> = nullptr, HTMLImportsController*
= 0);
On 2014/06/02 07:56:07, tkent wrote:
> WeakPtr(std::nullptr_t) is necessary because RawPtr needs nullptr
initialization
> here, right?

Yes, exactly so. As RawPtr() (for the Oilpan case) represent an uninitialized
pointer, and we do want it that way, reconciled this invocation of a transition
type constructor by extending WeakPtr's range of constructors.

[sof, 2014-06-02 08:02:57]

The CQ bit was checked by sigbjornf@opera.com

[commit-bot: I haz the power, 2014-06-02 08:03:26]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/sigbjornf@opera.com/304323010/50001

[Mads Ager (chromium), 2014-06-02 08:17:16]

LGTM2

[commit-bot: I haz the power, 2014-06-02 09:04:50]

Change committed as 175260