Update SkSpecialImage API

Update SkSpecialImage API

This adds some temporary compatibility features:
   internal_getBM & internal_fromBM - to convert back & forth from Bitmaps in the ImageFilters
   internal_getProxy - to allow the special images to interface with the current ImagerFilter system

It also adds a unique ID, opaque flag and size method.
   The unique ID and size method are needed for the image filter's caching system.
   The opaque flag is need for swapping back & forth from Bitmaps

These are all calved off of https://codereview.chromium.org/1695823002/ (Get OffsetImageFilter really working with SkSpecialImages) which actually makes use of them.

GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1705113002

Committed: https://skia.googlesource.com/skia/+/3b087f4010327b304242aaf27ef3689150dfc226

[robertphillips, 2016-02-17 16:32:21]

Description was changed from

==========
Update SkSpecialImage API
==========

to

==========
Update SkSpecialImage API
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

[robertphillips, 2016-02-17 16:47:21]

Description was changed from

==========
Update SkSpecialImage API
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

to

==========
Update SkSpecialImage API

This adds some temporary compatibility features:
   internal_getBM & internal_fromBM - to convert back & forth from Bitmaps in
the ImageFilters
   internal_getProxy - to work allow the special images to interface with the
current ImagerFilter system

It also adds a unique ID, opaque flag and size method.
   The unique ID and size method are needed for the image filter's caching
system.
   The opaque flag is need for swapping back & forth from Bitmaps


GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

[robertphillips, 2016-02-17 16:47:21]

robertphillips@google.com changed reviewers:
+ bsalomon@google.com, senorblanco@google.com

[robertphillips, 2016-02-17 16:47:45]

Description was changed from

==========
Update SkSpecialImage API

This adds some temporary compatibility features:
   internal_getBM & internal_fromBM - to convert back & forth from Bitmaps in
the ImageFilters
   internal_getProxy - to work allow the special images to interface with the
current ImagerFilter system

It also adds a unique ID, opaque flag and size method.
   The unique ID and size method are needed for the image filter's caching
system.
   The opaque flag is need for swapping back & forth from Bitmaps


GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

to

==========
Update SkSpecialImage API

This adds some temporary compatibility features:
   internal_getBM & internal_fromBM - to convert back & forth from Bitmaps in
the ImageFilters
   internal_getProxy - to allow the special images to interface with the current
ImagerFilter system

It also adds a unique ID, opaque flag and size method.
   The unique ID and size method are needed for the image filter's caching
system.
   The opaque flag is need for swapping back & forth from Bitmaps


GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

[robertphillips, 2016-02-17 16:48:51]

Description was changed from

==========
Update SkSpecialImage API

This adds some temporary compatibility features:
   internal_getBM & internal_fromBM - to convert back & forth from Bitmaps in
the ImageFilters
   internal_getProxy - to allow the special images to interface with the current
ImagerFilter system

It also adds a unique ID, opaque flag and size method.
   The unique ID and size method are needed for the image filter's caching
system.
   The opaque flag is need for swapping back & forth from Bitmaps


GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

to

==========
Update SkSpecialImage API

This adds some temporary compatibility features:
   internal_getBM & internal_fromBM - to convert back & forth from Bitmaps in
the ImageFilters
   internal_getProxy - to allow the special images to interface with the current
ImagerFilter system

It also adds a unique ID, opaque flag and size method.
   The unique ID and size method are needed for the image filter's caching
system.
   The opaque flag is need for swapping back & forth from Bitmaps

These are all calved off of https://codereview.chromium.org/1695823002/ (Get
OffsetImageFilter really working with SkSpecialImages) which actually makes use
of them.

GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

[Stephen White, 2016-02-17 17:00:02]

senorblanco@chromium.org changed reviewers:
+ senorblanco@chromium.org

[Stephen White, 2016-02-17 17:00:02]

https://codereview.chromium.org/1705113002/diff/20001/src/core/SkSpecialImage...
File src/core/SkSpecialImage.cpp (right):

https://codereview.chromium.org/1705113002/diff/20001/src/core/SkSpecialImage...
src/core/SkSpecialImage.cpp:94: fDevice.reset(SkGpuDevice::Create(context,
SkSurface::kYes_Budgeted, info, 0, &props,
This doesn't make sense to me. Why do we need to have a per-SkSpecialImage
proxy? It shouldn't need its own storage. Can't we just have a single proxy
that's copied down the chain, or something? I.e., the calling device creates
one, and sets in the primitive's SkSpecialImage, and then the base-class
onFilterImage() (the non-deprecated one) pulls it out and puts it in the surface
it creates? Something like that? So that it can be propagated to the created
SkSpecialImages?

Alternatively, we could make a new proxy that contains the SkSpecialImage, and
implements createDevice() by calling the image's surface creation functions. Not
sure if you can actually extract the device out of that, though.

https://codereview.chromium.org/1705113002/diff/20001/src/core/SkSpecialImage.h
File src/core/SkSpecialImage.h (right):

https://codereview.chromium.org/1705113002/diff/20001/src/core/SkSpecialImage...
src/core/SkSpecialImage.h:57: uint32_t uniqueID,
Out of curiosity, are we going to need this uniqueID when all the conversions
are done, or is it going to go away too?

[robertphillips, 2016-02-17 19:10:28]

https://codereview.chromium.org/1705113002/diff/20001/src/core/SkSpecialImage...
File src/core/SkSpecialImage.cpp (right):

https://codereview.chromium.org/1705113002/diff/20001/src/core/SkSpecialImage...
src/core/SkSpecialImage.cpp:94: fDevice.reset(SkGpuDevice::Create(context,
SkSurface::kYes_Budgeted, info, 0, &props,
On 2016/02/17 17:00:02, Stephen White wrote:
> This doesn't make sense to me. Why do we need to have a per-SkSpecialImage
> proxy? It shouldn't need its own storage. Can't we just have a single proxy
> that's copied down the chain, or something? I.e., the calling device creates
> one, and sets in the primitive's SkSpecialImage, and then the base-class
> onFilterImage() (the non-deprecated one) pulls it out and puts it in the
surface
> it creates? Something like that? So that it can be propagated to the created
> SkSpecialImages?
> 
> Alternatively, we could make a new proxy that contains the SkSpecialImage, and
> implements createDevice() by calling the image's surface creation functions.
Not
> sure if you can actually extract the device out of that, though.

Done.

https://codereview.chromium.org/1705113002/diff/20001/src/core/SkSpecialImage.h
File src/core/SkSpecialImage.h (right):

https://codereview.chromium.org/1705113002/diff/20001/src/core/SkSpecialImage...
src/core/SkSpecialImage.h:57: uint32_t uniqueID,
I believe it will go away when there are only SkImage-based SpecialImages (since
the unique ID will be wrapped in that class).

[Stephen White, 2016-02-17 21:44:03]

https://codereview.chromium.org/1705113002/diff/60001/src/core/SkSpecialImage...
File src/core/SkSpecialImage.cpp (right):

https://codereview.chromium.org/1705113002/diff/60001/src/core/SkSpecialImage...
src/core/SkSpecialImage.cpp:26: // Delete this entry point ASAP
Log a bug, and reference it here? (see below)

https://codereview.chromium.org/1705113002/diff/60001/src/core/SkSpecialImage...
src/core/SkSpecialImage.cpp:63: // Need to test offset case!
You mean an offset in the pixelref? Log a bug maybe?

https://codereview.chromium.org/1705113002/diff/60001/src/core/SkSpecialImage...
src/core/SkSpecialImage.cpp:83: #include "SkGpuDevice.h"
Is this still needed? If not, remove it. If so, could it be in the existing #if
SK_SUPPORT_GPU clause below?

https://codereview.chromium.org/1705113002/diff/60001/src/core/SkSpecialImage.h
File src/core/SkSpecialImage.h (right):

https://codereview.chromium.org/1705113002/diff/60001/src/core/SkSpecialImage...
src/core/SkSpecialImage.h:14: // remove this when internal_getProxy goes away
Nit: log a bug for switching all image filters to the new path, and reference it
here?

https://codereview.chromium.org/1705113002/diff/60001/src/core/SkSpecialImage...
src/core/SkSpecialImage.h:68: // These three internal methods will go away
Nit: reference above bug?

https://codereview.chromium.org/1705113002/diff/60001/src/core/SkSpecialImage...
src/core/SkSpecialImage.h:110: // TODO: remove this ASAP
Same here? (reference a switch-all-filters bug).

https://codereview.chromium.org/1705113002/diff/60001/src/core/SkSpecialSurfa...
File src/core/SkSpecialSurface.h (right):

https://codereview.chromium.org/1705113002/diff/60001/src/core/SkSpecialSurfa...
src/core/SkSpecialSurface.h:92: // TODO: remove this ASAP
Replace w/bug ID?

[robertphillips, 2016-02-18 13:18:39]

https://codereview.chromium.org/1705113002/diff/60001/src/core/SkSpecialImage...
File src/core/SkSpecialImage.cpp (right):

https://codereview.chromium.org/1705113002/diff/60001/src/core/SkSpecialImage...
src/core/SkSpecialImage.cpp:26: // Delete this entry point ASAP
On 2016/02/17 21:44:02, Stephen White wrote:
> Log a bug, and reference it here? (see below)

Done.

https://codereview.chromium.org/1705113002/diff/60001/src/core/SkSpecialImage...
src/core/SkSpecialImage.cpp:63: // Need to test offset case!
On 2016/02/17 21:44:02, Stephen White wrote:
> You mean an offset in the pixelref? Log a bug maybe?

Right - Done.

https://codereview.chromium.org/1705113002/diff/60001/src/core/SkSpecialImage...
src/core/SkSpecialImage.cpp:83: #include "SkGpuDevice.h"
On 2016/02/17 21:44:02, Stephen White wrote:
> Is this still needed? If not, remove it. If so, could it be in the existing
#if
> SK_SUPPORT_GPU clause below?

Done.

https://codereview.chromium.org/1705113002/diff/60001/src/core/SkSpecialImage.h
File src/core/SkSpecialImage.h (right):

https://codereview.chromium.org/1705113002/diff/60001/src/core/SkSpecialImage...
src/core/SkSpecialImage.h:14: // remove this when internal_getProxy goes away
On 2016/02/17 21:44:03, Stephen White wrote:
> Nit: log a bug for switching all image filters to the new path, and reference
it
> here?

Done.

https://codereview.chromium.org/1705113002/diff/60001/src/core/SkSpecialImage...
src/core/SkSpecialImage.h:68: // These three internal methods will go away
On 2016/02/17 21:44:02, Stephen White wrote:
> Nit: reference above bug?

Done.

https://codereview.chromium.org/1705113002/diff/60001/src/core/SkSpecialImage...
src/core/SkSpecialImage.h:110: // TODO: remove this ASAP
On 2016/02/17 21:44:03, Stephen White wrote:
> Same here? (reference a switch-all-filters bug).

Done.

https://codereview.chromium.org/1705113002/diff/60001/src/core/SkSpecialSurfa...
File src/core/SkSpecialSurface.h (right):

https://codereview.chromium.org/1705113002/diff/60001/src/core/SkSpecialSurfa...
src/core/SkSpecialSurface.h:92: // TODO: remove this ASAP
On 2016/02/17 21:44:03, Stephen White wrote:
> Replace w/bug ID?

Done.

[Stephen White, 2016-02-18 14:58:47]

LGTM

[robertphillips, 2016-02-18 16:00:33]

The CQ bit was checked by robertphillips@google.com to run a CQ dry run

[commit-bot: I haz the power, 2016-02-18 16:00:39]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1705113002/100001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1705113002/100001

[commit-bot: I haz the power, 2016-02-18 16:02:50]

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

[commit-bot: I haz the power, 2016-02-18 16:02:51]

Dry run: Try jobs failed on following builders:
  Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Debug-Trybot on client.skia (JOB_FAILED,
http://build.chromium.org/p/client.skia/builders/Test-Ubuntu-GCC-GCE-CPU-AVX2...)
  Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-Shared-Trybot on client.skia
(JOB_FAILED,
http://build.chromium.org/p/client.skia/builders/Test-Ubuntu-GCC-GCE-CPU-AVX2...)

[robertphillips, 2016-02-18 16:21:57]

The CQ bit was checked by robertphillips@google.com to run a CQ dry run

[commit-bot: I haz the power, 2016-02-18 16:22:06]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1705113002/120001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1705113002/120001

[commit-bot: I haz the power, 2016-02-18 16:43:25]

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

[commit-bot: I haz the power, 2016-02-18 16:43:26]

Dry run: This issue passed the CQ dry run.

[robertphillips, 2016-02-18 16:47:14]

The CQ bit was checked by robertphillips@google.com

[robertphillips, 2016-02-18 16:47:15]

The patchset sent to the CQ was uploaded after l-g-t-m from
senorblanco@chromium.org
Link to the patchset: https://codereview.chromium.org/1705113002/#ps120001
(title: "Fix no-gpu build")

[commit-bot: I haz the power, 2016-02-18 16:47:23]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1705113002/120001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1705113002/120001

[commit-bot: I haz the power, 2016-02-18 16:48:06]

Description was changed from

==========
Update SkSpecialImage API

This adds some temporary compatibility features:
   internal_getBM & internal_fromBM - to convert back & forth from Bitmaps in
the ImageFilters
   internal_getProxy - to allow the special images to interface with the current
ImagerFilter system

It also adds a unique ID, opaque flag and size method.
   The unique ID and size method are needed for the image filter's caching
system.
   The opaque flag is need for swapping back & forth from Bitmaps

These are all calved off of https://codereview.chromium.org/1695823002/ (Get
OffsetImageFilter really working with SkSpecialImages) which actually makes use
of them.

GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

to

==========
Update SkSpecialImage API

This adds some temporary compatibility features:
   internal_getBM & internal_fromBM - to convert back & forth from Bitmaps in
the ImageFilters
   internal_getProxy - to allow the special images to interface with the current
ImagerFilter system

It also adds a unique ID, opaque flag and size method.
   The unique ID and size method are needed for the image filter's caching
system.
   The opaque flag is need for swapping back & forth from Bitmaps

These are all calved off of https://codereview.chromium.org/1695823002/ (Get
OffsetImageFilter really working with SkSpecialImages) which actually makes use
of them.

GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...

Committed:
https://skia.googlesource.com/skia/+/3b087f4010327b304242aaf27ef3689150dfc226
==========

[commit-bot: I haz the power, 2016-02-18 16:48:07]

Committed patchset #7 (id:120001) as
https://skia.googlesource.com/skia/+/3b087f4010327b304242aaf27ef3689150dfc226