support direct writing to top layer

support direct writing to top layer, and hide getTopLayer()

this should remove many of the chrome callers that today call
accessBitmap on the toplayer, so they can read/write those pixels.

The ultimate fix will be to support custom allocation of raster layers
(via GDI/cairo/mac) so we can remove PlatformDevice subclassing in
skia/ext

BUG=skia:
R=bsalomon@google.com, scroggo@google.com

Committed: https://code.google.com/p/skia/source/detail?r=13774

[bsalomon, 2014-03-12 14:43:11]

https://codereview.chromium.org/197433002/diff/20001/include/core/SkCanvas.h
File include/core/SkCanvas.h (right):

https://codereview.chromium.org/197433002/diff/20001/include/core/SkCanvas.h#...
include/core/SkCanvas.h:187: *  local RAM) return the const-address of those
pixels, and if not null,
This comment should be updated to indicate that the pointer is writable.

https://codereview.chromium.org/197433002/diff/20001/src/core/SkCanvas.cpp
File src/core/SkCanvas.cpp (right):

https://codereview.chromium.org/197433002/diff/20001/src/core/SkCanvas.cpp#ne...
src/core/SkCanvas.cpp:1066: return const_cast<void*>(this->onPeekPixels(info,
rowBytes));
onAccessTopLayerPixels()

https://codereview.chromium.org/197433002/diff/20001/src/utils/SkDeferredCanv...
File src/utils/SkDeferredCanvas.cpp (left):

https://codereview.chromium.org/197433002/diff/20001/src/utils/SkDeferredCanv...
src/utils/SkDeferredCanvas.cpp:138: class DeferredDevice : public SkBaseDevice {
Should this be *Sk*DeferredDevice? It's only in this cpp file. I often leave off
Sk/Gr prefixes for such classes.

[reed1, 2014-03-12 15:31:58]

https://codereview.chromium.org/197433002/diff/20001/include/core/SkCanvas.h
File include/core/SkCanvas.h (right):

https://codereview.chromium.org/197433002/diff/20001/include/core/SkCanvas.h#...
include/core/SkCanvas.h:187: *  local RAM) return the const-address of those
pixels, and if not null,
On 2014/03/12 14:43:11, bsalomon wrote:
> This comment should be updated to indicate that the pointer is writable.

Done.

https://codereview.chromium.org/197433002/diff/20001/src/core/SkCanvas.cpp
File src/core/SkCanvas.cpp (right):

https://codereview.chromium.org/197433002/diff/20001/src/core/SkCanvas.cpp#ne...
src/core/SkCanvas.cpp:1066: return const_cast<void*>(this->onPeekPixels(info,
rowBytes));
On 2014/03/12 14:43:11, bsalomon wrote:
> onAccessTopLayerPixels()

Done.

https://codereview.chromium.org/197433002/diff/20001/src/utils/SkDeferredCanv...
File src/utils/SkDeferredCanvas.cpp (left):

https://codereview.chromium.org/197433002/diff/20001/src/utils/SkDeferredCanv...
src/utils/SkDeferredCanvas.cpp:138: class DeferredDevice : public SkBaseDevice {
On 2014/03/12 14:43:11, bsalomon wrote:
> Should this be *Sk*DeferredDevice? It's only in this cpp file. I often leave
off
> Sk/Gr prefixes for such classes.

Turns out it was declared in SkDeferredCanvas.h and (as a friend) in SkDevice.h,
hence my Sk prepend

[bsalomon, 2014-03-12 15:36:33]

lgtm

On 2014/03/12 15:31:58, reed1 wrote:
>
https://codereview.chromium.org/197433002/diff/20001/src/utils/SkDeferredCanv...
> src/utils/SkDeferredCanvas.cpp:138: class DeferredDevice : public SkBaseDevice
{
> On 2014/03/12 14:43:11, bsalomon wrote:
> > Should this be *Sk*DeferredDevice? It's only in this cpp file. I often leave
> off
> > Sk/Gr prefixes for such classes.
> 
> Turns out it was declared in SkDeferredCanvas.h and (as a friend) in
SkDevice.h,
> hence my Sk prepend

gotcha

[scroggo, 2014-03-12 15:36:58]

lgtm

[reed1, 2014-03-12 15:39:20]

The CQ bit was checked by reed@google.com

[commit-bot: I haz the power, 2014-03-12 15:39:26]

CQ is trying da patch. Follow status at
https://skia-tree-status.appspot.com/cq/reed@google.com/197433002/40001

[reed1, 2014-03-12 15:39:32]

The CQ bit was unchecked by reed@google.com

[reed1, 2014-03-12 15:40:25]

The CQ bit was checked by reed@google.com

[commit-bot: I haz the power, 2014-03-12 15:40:26]

CQ is trying da patch. Follow status at
https://skia-tree-status.appspot.com/cq/reed@google.com/197433002/50013

[commit-bot: I haz the power, 2014-03-12 15:48:16]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-03-12 15:48:17]

Retried try job too often on Build-Mac10.8-Clang-x86-Release-Trybot for step(s)
BuildBench, BuildEverything, BuildGm, BuildSkiaLib, BuildTests, BuildTools
http://108.170.219.164:10117/buildstatus?builder=Build-Mac10.8-Clang-x86-Rele...

[reed1, 2014-03-12 17:02:38]

The CQ bit was checked by reed@google.com

[commit-bot: I haz the power, 2014-03-12 17:02:47]

CQ is trying da patch. Follow status at
https://skia-tree-status.appspot.com/cq/reed@google.com/197433002/70001

[commit-bot: I haz the power, 2014-03-12 17:06:33]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-03-12 17:06:37]

Retried try job too often on Build-Mac10.8-Clang-x86-Release-Trybot for step(s)
BuildSkiaLib
http://108.170.219.164:10117/buildstatus?builder=Build-Mac10.8-Clang-x86-Rele...

[reed1, 2014-03-12 17:26:08]

The CQ bit was checked by reed@google.com

[commit-bot: I haz the power, 2014-03-12 17:26:12]

CQ is trying da patch. Follow status at
https://skia-tree-status.appspot.com/cq/reed@google.com/197433002/80001

[reed1, 2014-03-12 18:28:35]

The CQ bit was unchecked by reed@google.com

[reed1, 2014-03-12 18:28:42]

Committed patchset #6 manually as r13774 (presubmit successful).