Added registerChild; transforms, textures, glKey automatically handled.

Added registerChild; transforms, textures, glKey automatically handled.

BUG=skia:

Committed: https://skia.googlesource.com/skia/+/4b3050b410254d0cb38df9a30ae2e209124fa1a2

[wangyix, 2015-07-31 21:16:14]

wangyix@google.com changed reviewers:
+ bsalomon@google.com, egdaniel@google.com, joshualitt@google.com,
robertphillips@google.com

[joshualitt, 2015-08-03 14:06:02]

https://codereview.chromium.org/1266633003/diff/1/include/gpu/GrFragmentProce...
File include/gpu/GrFragmentProcessor.h (right):

https://codereview.chromium.org/1266633003/diff/1/include/gpu/GrFragmentProce...
include/gpu/GrFragmentProcessor.h:30: virtual void getGLProcessorKey(const
GrGLSLCaps& caps,
Lets rename all of the *IncludeChildProc functions to *, and if any of these
other non-recursive functions are still needed, lets rename them *Priv.

https://codereview.chromium.org/1266633003/diff/1/include/gpu/GrFragmentProce...
include/gpu/GrFragmentProcessor.h:45: for (int i = 0; i <
fChildProcessors.count(); ++i)
We put braces around single line for loops.

https://codereview.chromium.org/1266633003/diff/1/include/gpu/GrFragmentProce...
include/gpu/GrFragmentProcessor.h:53: for (int i = 0; i <
fChildProcessors.count(); ++i)
same

https://codereview.chromium.org/1266633003/diff/1/include/gpu/GrFragmentProce...
include/gpu/GrFragmentProcessor.h:76: for (int i = 0; i <
fChildProcessors.count(); ++i)
same

https://codereview.chromium.org/1266633003/diff/1/include/gpu/GrFragmentProce...
include/gpu/GrFragmentProcessor.h:157: 
delete tab

https://codereview.chromium.org/1266633003/diff/1/src/gpu/gl/builders/GrGLPro...
File src/gpu/gl/builders/GrGLProgramBuilder.cpp (right):

https://codereview.chromium.org/1266633003/diff/1/src/gpu/gl/builders/GrGLPro...
src/gpu/gl/builders/GrGLProgramBuilder.cpp:183: void
GrGLProgramBuilder::emitSamplers(const GrProcessor& processor,
This should be GrFragmentProcessor& so it doesn't instantiate for GrProcessor
base class.  Also, can we make this call the general template, and then just
call itself recursively?  I think we can try upcasting to GrProcessor base
class, or something similar.

https://codereview.chromium.org/1266633003/diff/1/src/gpu/gl/builders/GrGLPro...
src/gpu/gl/builders/GrGLProgramBuilder.cpp:207: void
appendGrFPCoordTransforms(const GrFragmentProcessor* processor,
static void append_gr_fp_coord_transforms

https://codereview.chromium.org/1266633003/diff/1/src/gpu/gl/builders/GrGLPro...
src/gpu/gl/builders/GrGLProgramBuilder.cpp:407: void
GrGLProgramBuilder::emitSamplers(const GrProcessor& processor,
This should be Proc& processor

[wangyix, 2015-08-03 16:01:52]

Patchset #2 (id:20001) has been deleted

[wangyix, 2015-08-03 16:12:47]

https://codereview.chromium.org/1266633003/diff/1/include/gpu/GrFragmentProce...
File include/gpu/GrFragmentProcessor.h (right):

https://codereview.chromium.org/1266633003/diff/1/include/gpu/GrFragmentProce...
include/gpu/GrFragmentProcessor.h:30: virtual void getGLProcessorKey(const
GrGLSLCaps& caps,
On 2015/08/03 14:06:02, joshualitt wrote:
> Lets rename all of the *IncludeChildProc functions to *, and if any of these
> other non-recursive functions are still needed, lets rename them *Priv.

Done.

https://codereview.chromium.org/1266633003/diff/1/include/gpu/GrFragmentProce...
include/gpu/GrFragmentProcessor.h:45: for (int i = 0; i <
fChildProcessors.count(); ++i)
On 2015/08/03 14:06:02, joshualitt wrote:
> We put braces around single line for loops.

Done.

https://codereview.chromium.org/1266633003/diff/1/include/gpu/GrFragmentProce...
include/gpu/GrFragmentProcessor.h:53: for (int i = 0; i <
fChildProcessors.count(); ++i)
On 2015/08/03 14:06:02, joshualitt wrote:
> same

Done.

https://codereview.chromium.org/1266633003/diff/1/include/gpu/GrFragmentProce...
include/gpu/GrFragmentProcessor.h:76: for (int i = 0; i <
fChildProcessors.count(); ++i)
On 2015/08/03 14:06:02, joshualitt wrote:
> same

Done.

https://codereview.chromium.org/1266633003/diff/1/include/gpu/GrFragmentProce...
include/gpu/GrFragmentProcessor.h:157: 
On 2015/08/03 14:06:02, joshualitt wrote:
> delete tab

Done.

https://codereview.chromium.org/1266633003/diff/1/src/gpu/gl/builders/GrGLPro...
File src/gpu/gl/builders/GrGLProgramBuilder.cpp (right):

https://codereview.chromium.org/1266633003/diff/1/src/gpu/gl/builders/GrGLPro...
src/gpu/gl/builders/GrGLProgramBuilder.cpp:183: void
GrGLProgramBuilder::emitSamplers(const GrProcessor& processor,
On 2015/08/03 14:06:02, joshualitt wrote:
> This should be GrFragmentProcessor& so it doesn't instantiate for GrProcessor
> base class.  Also, can we make this call the general template, and then just
> call itself recursively?  I think we can try upcasting to GrProcessor base
> class, or something similar.

Removed this function for the time being; Was discussed.

https://codereview.chromium.org/1266633003/diff/1/src/gpu/gl/builders/GrGLPro...
src/gpu/gl/builders/GrGLProgramBuilder.cpp:207: void
appendGrFPCoordTransforms(const GrFragmentProcessor* processor,
On 2015/08/03 14:06:02, joshualitt wrote:
> static void append_gr_fp_coord_transforms

Done.

https://codereview.chromium.org/1266633003/diff/1/src/gpu/gl/builders/GrGLPro...
src/gpu/gl/builders/GrGLProgramBuilder.cpp:407: void
GrGLProgramBuilder::emitSamplers(const GrProcessor& processor,
On 2015/08/03 14:06:02, joshualitt wrote:
> This should be Proc& processor

Didn't do this; Was discussed.

[bsalomon, 2015-08-03 17:07:08]

I haven't looked in detail at the rest of the files, but had a style note.

https://codereview.chromium.org/1266633003/diff/60001/include/gpu/GrFragmentP...
File include/gpu/GrFragmentProcessor.h (right):

https://codereview.chromium.org/1266633003/diff/60001/include/gpu/GrFragmentP...
include/gpu/GrFragmentProcessor.h:30: virtual void getGLProcessorKeyPriv(const
GrGLSLCaps& caps,
Usually we'd call this "onGetGLProcessorKey" and make it private. Similar for
the other Privs.

[joshualitt, 2015-08-03 17:41:06]

On 2015/08/03 17:07:08, bsalomon wrote:
> I haven't looked in detail at the rest of the files, but had a style note.
> 
>
https://codereview.chromium.org/1266633003/diff/60001/include/gpu/GrFragmentP...
> File include/gpu/GrFragmentProcessor.h (right):
> 
>
https://codereview.chromium.org/1266633003/diff/60001/include/gpu/GrFragmentP...
> include/gpu/GrFragmentProcessor.h:30: virtual void getGLProcessorKeyPriv(const
> GrGLSLCaps& caps,
> Usually we'd call this "onGetGLProcessorKey" and make it private. Similar for
> the other Privs.

Sorry for leading you astray Yixin, but could you rename *Priv to on*

[wangyix, 2015-08-03 17:58:51]

Patchset #3 (id:60001) has been deleted

[wangyix, 2015-08-03 17:59:00]

Patchset #2 (id:40001) has been deleted

[joshualitt, 2015-08-03 19:23:41]

On 2015/08/03 19:14:21, wangyix wrote:

lgtm

[wangyix, 2015-08-03 19:26:25]

The CQ bit was checked by wangyix@google.com

[commit-bot: I haz the power, 2015-08-03 19:26:32]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1266633003/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1266633003/80001

[wangyix, 2015-08-03 19:27:32]

The CQ bit was unchecked by wangyix@google.com

[commit-bot: I haz the power, 2015-08-03 19:27:34]

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

[commit-bot: I haz the power, 2015-08-03 19:27:35]

Try jobs failed on following builders:
  skia_presubmit-Trybot on client.skia.fyi (JOB_FAILED,
http://build.chromium.org/p/client.skia.fyi/builders/skia_presubmit-Trybot/bu...)

[wangyix, 2015-08-03 19:44:07]

The CQ bit was checked by wangyix@google.com

[commit-bot: I haz the power, 2015-08-03 19:44:35]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1266633003/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1266633003/80001

[commit-bot: I haz the power, 2015-08-03 19:45:54]

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

[commit-bot: I haz the power, 2015-08-03 19:45:55]

Try jobs failed on following builders:
  skia_presubmit-Trybot on client.skia.fyi (JOB_FAILED,
http://build.chromium.org/p/client.skia.fyi/builders/skia_presubmit-Trybot/bu...)

[joshualitt, 2015-08-03 19:47:11]

On 2015/08/03 19:45:55, commit-bot: I haz the power wrote:
> Try jobs failed on following builders:
>   skia_presubmit-Trybot on client.skia.fyi (JOB_FAILED,
>
http://build.chromium.org/p/client.skia.fyi/builders/skia_presubmit-Trybot/bu...)

Brian, can you lgtm for presubmit?

[bsalomon, 2015-08-04 14:51:03]

lgtm, but I'd like there to be a follow up to make the names more consistent
("Procs" vs. "Processors" in function names, change from "IncludeChildProcs" to
"IncludingChildren", or ...)

[wangyix, 2015-08-04 14:58:49]

The CQ bit was checked by wangyix@google.com

[commit-bot: I haz the power, 2015-08-04 14:58:59]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1266633003/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1266633003/80001

[commit-bot: I haz the power, 2015-08-04 14:59:40]

Committed patchset #2 (id:80001) as
https://skia.googlesource.com/skia/+/4b3050b410254d0cb38df9a30ae2e209124fa1a2