Cleanup: Mark some overridden methods with 'SK_OVERRIDE'.

Cleanup: Mark some overridden methods with 'SK_OVERRIDE'.

This fixes errors like this:

../../include/gpu/effects/GrPorterDuffXferProcessor.h:27:25: error:
'name' overrides a member function but is not marked 'override'
[-Werror,-Winconsistent-missing-override]

BUG=skia:3075
TEST=ninja -C out/Debug skia_lib
TBR=bsalomon@google.com

Committed: https://skia.googlesource.com/skia/+/912ed6ebb8e2813e72ed7a3dec3b6710ba7e7405

[tfarina, 2014-12-14 15:03:22]

TBRing... (rs?)

[tfarina, 2014-12-14 15:03:32]

The CQ bit was checked by tfarina@chromium.org

[commit-bot: I haz the power, 2014-12-14 15:04:19]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/804813002/1

[mtklein, 2014-12-14 23:19:27]

mtklein@google.com changed reviewers:
+ mtklein@google.com

[mtklein, 2014-12-14 23:19:28]

lgtm

CLs that add SK_OVERRIDE where legal are always very welcome!

Isn't there a Clang tool to add override to every place it could legally go? 
Why don't we follow up with that (and s/override/SK_OVERRIDE) and get all the
rest in one go?

[commit-bot: I haz the power, 2014-12-14 23:20:15]

Committed patchset #1 (id:1) as
https://skia.googlesource.com/skia/+/912ed6ebb8e2813e72ed7a3dec3b6710ba7e7405

[tfarina, 2014-12-15 01:19:36]

I'm not aware of such tool. Ccing Daniel, he might know.

Is Skia C++11 ready?

[dcheng, 2014-12-15 01:30:52]

On 2014/12/15 at 01:19:36, tfarina wrote:
> I'm not aware of such tool. Ccing Daniel, he might know.
> 
> Is Skia C++11 ready?

It is theoretically possible to do this with the Clang plugin. I guess the
bigger question is if it's worth the effort. I have a simple patch which can
automatically apply fixits generated by clang to source
(https://codereview.chromium.org/598073004). The workflow is basically something
like:

1. ninja -C out/Debug -k 1000 > fixits
2. tools/clang/scripts/apply-fixits.sh < fixits
3. repeat steps 1-2 until everything compiles

However, the Chrome clang plugin enforces (or will soon enforce) things that
probably aren't desirable if you aren't using C++11 (it prefers you only use one
of {virtual,override,final} and will soon generate fixits to remove redundant
specifiers) and it also emits fixits to insert 'override' rather than
SK_OVERRIDE. These aren't insurmountable obstacles, but they might make it more
effort than it's worth, if there aren't that many things to fix up.

[mtklein, 2014-12-15 14:58:01]

On 2014/12/15 01:30:52, dcheng wrote:
> On 2014/12/15 at 01:19:36, tfarina wrote:
> > I'm not aware of such tool. Ccing Daniel, he might know.
> > 
> > Is Skia C++11 ready?

Skia's supports back to C++98, but sure, it's ready for C++11.  We build and
test it with -std=c++11 on a few of our bots.

> It is theoretically possible to do this with the Clang plugin. I guess the
> bigger question is if it's worth the effort. I have a simple patch which can
> automatically apply fixits generated by clang to source
> (https://codereview.chromium.org/598073004). The workflow is basically
something
> like:
> 
> 1. ninja -C out/Debug -k 1000 > fixits
> 2. tools/clang/scripts/apply-fixits.sh < fixits
> 3. repeat steps 1-2 until everything compiles
> 
> However, the Chrome clang plugin enforces (or will soon enforce) things that
> probably aren't desirable if you aren't using C++11 (it prefers you only use
one
> of {virtual,override,final} and will soon generate fixits to remove redundant
> specifiers) and it also emits fixits to insert 'override' rather than
> SK_OVERRIDE. These aren't insurmountable obstacles, but they might make it
more
> effort than it's worth, if there aren't that many things to fix up.

I think I was thinking of clang-modernize (-add-override -override-macros). 
I'll play around with that.