Move SkV8Example to SkSurface

Move SkV8Example to SkSurface

BUG=skia:

Committed: http://code.google.com/p/skia/source/detail?r=13609

[robertphillips, 2014-02-26 23:51:08]

Questions, questions, questions

https://codereview.chromium.org/148523019/diff/30001/experimental/SkV8Example...
File experimental/SkV8Example/SkV8Example.cpp (right):

https://codereview.chromium.org/148523019/diff/30001/experimental/SkV8Example...
experimental/SkV8Example/SkV8Example.cpp:28: #include "SkGraphics.h"
Is this #include for the kBGRA_8888_SkColorType definition?

https://codereview.chromium.org/148523019/diff/30001/experimental/SkV8Example...
experimental/SkV8Example/SkV8Example.cpp:98: SkSafeUnref(fCurSurface);
Doesn't this need to get cleaned up in the destructor?

https://codereview.chromium.org/148523019/diff/30001/experimental/SkV8Example...
experimental/SkV8Example/SkV8Example.cpp:104: #if SK_SUPPORT_GPU
So does the base class not support creating a GPU backend?

https://codereview.chromium.org/148523019/diff/30001/experimental/SkV8Example...
experimental/SkV8Example/SkV8Example.cpp:107: SkCanvas* c =
fCurSurface->getCanvas();
How isn't there an unbalanced ref problem here? Is the canvas being leaked?

[jcgregorio, 2014-02-27 14:23:15]

https://codereview.chromium.org/148523019/diff/30001/experimental/SkV8Example...
File experimental/SkV8Example/SkV8Example.cpp (right):

https://codereview.chromium.org/148523019/diff/30001/experimental/SkV8Example...
experimental/SkV8Example/SkV8Example.cpp:28: #include "SkGraphics.h"
On 2014/02/26 23:51:08, robertphillips wrote:
> Is this #include for the kBGRA_8888_SkColorType definition?

I added it for that reason, looks like it isn't needed. Dropped.

https://codereview.chromium.org/148523019/diff/30001/experimental/SkV8Example...
experimental/SkV8Example/SkV8Example.cpp:98: SkSafeUnref(fCurSurface);
Added SkSafeUnref's for all the GPU related objects in the destructor.

On 2014/02/26 23:51:08, robertphillips wrote:
> Doesn't this need to get cleaned up in the destructor?

https://codereview.chromium.org/148523019/diff/30001/experimental/SkV8Example...
experimental/SkV8Example/SkV8Example.cpp:104: #if SK_SUPPORT_GPU
On 2014/02/26 23:51:08, robertphillips wrote:
> So does the base class not support creating a GPU backend?

No, the base class impl is merely new SkCanvas(this-getBitmap()).

https://codereview.chromium.org/148523019/diff/30001/experimental/SkV8Example...
experimental/SkV8Example/SkV8Example.cpp:107: SkCanvas* c =
fCurSurface->getCanvas();
On 2014/02/26 23:51:08, robertphillips wrote:
> How isn't there an unbalanced ref problem here? Is the canvas being leaked?

Yeah, I would expect that fCurSurface->getCanvas() would return a canvas with a
ref count of 2, since it is in theory maintaining a cache of the canvas. Maybe
that's where the change for the refcount should go?

[robertphillips, 2014-02-27 14:39:26]

lgtm

[jcgregorio, 2014-02-27 15:54:10]

The CQ bit was checked by jcgregorio@google.com

[commit-bot: I haz the power, 2014-02-27 15:54:18]

CQ is trying da patch. Follow status at
https://skia-tree-status.appspot.com/cq/jcgregorio@google.com/148523019/50001

[commit-bot: I haz the power, 2014-02-27 17:39:42]

Change committed as 13609