Oilpan: Fix a raw pointer in FrameLoaderClientImplTest.

Oilpan: Fix a raw pointer in FrameLoaderClientImplTest.

A simple follow-up of https://codereview.chromium.org/1306793003/.

BUG=509911
R=haraken@chromium.org, oilpan-reviews@chromium.org, sigbjornf@opera.com

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=201125

[haraken, 2015-08-25 09:42:18]

LGTM

https://codereview.chromium.org/1314703003/diff/1/Source/web/tests/FrameLoade...
File Source/web/tests/FrameLoaderClientImplTest.cpp (right):

https://codereview.chromium.org/1314703003/diff/1/Source/web/tests/FrameLoade...
Source/web/tests/FrameLoaderClientImplTest.cpp:101:
RawPtrWillBePersistent<FrameLoaderClientImpl> m_frameLoaderClientImpl;

I'd prefer using RefPtrWillBePersistent (to clarify that keeping a strong
reference here won't cause a leak).

[sof, 2015-08-25 09:47:10]

https://codereview.chromium.org/1314703003/diff/1/Source/web/tests/FrameLoade...
File Source/web/tests/FrameLoaderClientImplTest.cpp (right):

https://codereview.chromium.org/1314703003/diff/1/Source/web/tests/FrameLoade...
Source/web/tests/FrameLoaderClientImplTest.cpp:101:
RawPtrWillBePersistent<FrameLoaderClientImpl> m_frameLoaderClientImpl;
On 2015/08/25 09:42:17, haraken wrote:
> 
> I'd prefer using RefPtrWillBePersistent (to clarify that keeping a strong
> reference here won't cause a leak).

Just avoid the issue and not keep the cached field to start with? This is test
code after all.

[Yuta Kitamura, 2015-08-25 10:00:55]

https://codereview.chromium.org/1314703003/diff/1/Source/web/tests/FrameLoade...
File Source/web/tests/FrameLoaderClientImplTest.cpp (right):

https://codereview.chromium.org/1314703003/diff/1/Source/web/tests/FrameLoade...
Source/web/tests/FrameLoaderClientImplTest.cpp:101:
RawPtrWillBePersistent<FrameLoaderClientImpl> m_frameLoaderClientImpl;
On 2015/08/25 09:47:10, sof wrote:
> On 2015/08/25 09:42:17, haraken wrote:
> > 
> > I'd prefer using RefPtrWillBePersistent (to clarify that keeping a strong
> > reference here won't cause a leak).
> 
> Just avoid the issue and not keep the cached field to start with? This is test
> code after all.

Gave another shot according to sof's comment. PTAL.

[haraken, 2015-08-25 10:21:46]

On 2015/08/25 10:00:55, Yuta Kitamura wrote:
>
https://codereview.chromium.org/1314703003/diff/1/Source/web/tests/FrameLoade...
> File Source/web/tests/FrameLoaderClientImplTest.cpp (right):
> 
>
https://codereview.chromium.org/1314703003/diff/1/Source/web/tests/FrameLoade...
> Source/web/tests/FrameLoaderClientImplTest.cpp:101:
> RawPtrWillBePersistent<FrameLoaderClientImpl> m_frameLoaderClientImpl;
> On 2015/08/25 09:47:10, sof wrote:
> > On 2015/08/25 09:42:17, haraken wrote:
> > > 
> > > I'd prefer using RefPtrWillBePersistent (to clarify that keeping a strong
> > > reference here won't cause a leak).
> > 
> > Just avoid the issue and not keep the cached field to start with? This is
test
> > code after all.
> 
> Gave another shot according to sof's comment. PTAL.

Looks nicer. LGTM

[haraken, 2015-08-25 10:21:50]

The CQ bit was checked by haraken@chromium.org

[commit-bot: I haz the power, 2015-08-25 10:22:11]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1314703003/2
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1314703003/2

[sof, 2015-08-25 10:37:14]

lgtm

https://codereview.chromium.org/1314703003/diff/2/Source/web/tests/FrameLoade...
File Source/web/tests/FrameLoaderClientImplTest.cpp (right):

https://codereview.chromium.org/1314703003/diff/2/Source/web/tests/FrameLoade...
Source/web/tests/FrameLoaderClientImplTest.cpp:89: WTF::CString userAgent =
frameLoaderClientImpl().userAgent(dummyURL).utf8();
You can use frameLoaderClient() for this also (and avoid the need for
frameLoaderClientImpl()).

[commit-bot: I haz the power, 2015-08-25 11:11:07]

Committed patchset #2 (id:2) as
https://src.chromium.org/viewvc/blink?view=rev&revision=201125

[Yuta Kitamura, 2015-08-26 05:32:30]

https://codereview.chromium.org/1314703003/diff/2/Source/web/tests/FrameLoade...
File Source/web/tests/FrameLoaderClientImplTest.cpp (right):

https://codereview.chromium.org/1314703003/diff/2/Source/web/tests/FrameLoade...
Source/web/tests/FrameLoaderClientImplTest.cpp:89: WTF::CString userAgent =
frameLoaderClientImpl().userAgent(dummyURL).utf8();
On 2015/08/25 10:37:14, sof wrote:
> You can use frameLoaderClient() for this also (and avoid the need for
> frameLoaderClientImpl()).

OK, I'll follow up on this. Follow-up of a follow-up ;)