Add InputScope support into InputMethodIMM32

Add InputScope support into InputMethodIMM32

RWHVW has directly used InputScope APIs that are available
for IMM32 applications.
In order to migrate from RWHVW to RWHVA as Aura transition,
InputMethodIMM32 should also have InputScope support since
RWHVA delegates all the IME activities into InputMethodIMM32.

BUG=296505, 244688
TEST=manually done with Aura build on Windows 8.1

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=224954

[yoichio, 2013-09-23 12:09:54]

https://codereview.chromium.org/23537064/diff/1/ui/base/ime/input_method_imm3...
File ui/base/ime/input_method_imm32.cc (right):

https://codereview.chromium.org/23537064/diff/1/ui/base/ime/input_method_imm3...
ui/base/ime/input_method_imm32.cc:283: window_handle, GetTextInputType(),
GetTextInputMode());
Can we cache values of GetTextInputType() and *Mode() in this scope?

[yukawa, 2013-09-23 17:54:21]

https://codereview.chromium.org/23537064/diff/1/ui/base/ime/input_method_imm3...
File ui/base/ime/input_method_imm32.cc (right):

https://codereview.chromium.org/23537064/diff/1/ui/base/ime/input_method_imm3...
ui/base/ime/input_method_imm32.cc:283: window_handle, GetTextInputType(),
GetTextInputMode());
On 2013/09/23 12:09:54, yoichio wrote:
> Can we cache values of GetTextInputType() and *Mode() in this scope?

I think we can in this particular case, although this is not a trivial question
in general.

We should make a snapshot of state if and only if
  A. the state is guaranteed to be immutable in the scope
or
  B. the state might be changed in the scope but we should use the initial state
intentionally.

In this case, |imm32_manager_.DisableIME|, |imm32_manager_.EnableIME|, and
|imm32_manager_.SetTextInputMode| might have side side effect.

Apparently InputMethodIMM32::UpdateIMEState is not designed to be reentrant.
Paradoxically it means that we assume the case A is implicitly applied in this
function. This is why I think we can *cache* the state in this particular case.

[yoichio, 2013-09-24 01:38:56]

lgtm

[Seigo Nonaka, 2013-09-24 03:59:45]

lgtm

Please update the description, IIRC the Aura should not go for M31, so this CL
is not necessary for M31 merge.

[yukawa, 2013-09-24 04:21:50]

On 2013/09/24 03:59:45, Seigo Nonaka wrote:
> lgtm
> 
> Please update the description, IIRC the Aura should not go for M31, so this CL
> is not necessary for M31 merge.

done.

[Seigo Nonaka, 2013-09-24 04:22:40]

lgtm again

[commit-bot: I haz the power, 2013-09-24 04:27:37]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/yukawa@chromium.org/23537064/11001

[commit-bot: I haz the power, 2013-09-24 05:09:03]

Retried try job too often on linux_chromeos for step(s) net_unittests,
sandbox_linux_unittests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=linux_chro...

[commit-bot: I haz the power, 2013-09-24 05:40:02]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/yukawa@chromium.org/23537064/11001

[commit-bot: I haz the power, 2013-09-24 08:17:02]

Change committed as 224954