Performance fix for Vista+ where non-IME input is reported as IME

On Windows Vista and its follow-ons, ImmIsIME() returns always true, thus
non-IME input languages were classified as IME.
This caused extra traffic of IPCs.

This CL uses ImmGetFileName() to check if a given
keyboard layout has any associated IME or not.

BUG=241342
TEST=manually check if IME works properly on Vista+

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=201035

[kochi, 2013-05-16 07:08:11]

Hi Yukawa-san,

Could you review this?

Thanks,

[Yohei Yukawa, 2013-05-16 08:55:02]

https://codereview.chromium.org/14988010/diff/8001/ui/base/win/ime_input.cc
File ui/base/win/ime_input.cc (right):

https://codereview.chromium.org/14988010/diff/8001/ui/base/win/ime_input.cc#n...
ui/base/win/ime_input.cc:140: CComPtr<ITfInputProcessorProfileMgr> prof_mgr;
We prefer to use our own scoped pointer to maingain COM object.
Use base::win::ScopedComPtr<ITfInputProcessorProfileMgr> instead.

[kochi, 2013-05-16 10:29:31]

Thanks for the review!

Nona-san, once I get LGTM from Yukawa-san, could you OWNERS review?

https://codereview.chromium.org/14988010/diff/8001/ui/base/win/ime_input.cc
File ui/base/win/ime_input.cc (right):

https://codereview.chromium.org/14988010/diff/8001/ui/base/win/ime_input.cc#n...
ui/base/win/ime_input.cc:140: CComPtr<ITfInputProcessorProfileMgr> prof_mgr;
On 2013/05/16 08:55:02, Yohei Yukawa wrote:
> We prefer to use our own scoped pointer to maingain COM object.
> Use base::win::ScopedComPtr<ITfInputProcessorProfileMgr> instead.

Done.

[Yohei Yukawa, 2013-05-16 10:35:15]

lgtm

[Seigo Nonaka, 2013-05-17 03:58:22]

lgtm but I'm not an owner of this file.
Please send to ui/ owners.

[kochi, 2013-05-17 04:01:03]

Hi Scott,

Could you take an OWNERS look?

Thanks,

[sky, 2013-05-17 15:43:08]

LGTM

[commit-bot: I haz the power, 2013-05-19 15:30:55]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/kochi@chromium.org/14988010/14001

[commit-bot: I haz the power, 2013-05-19 16:57:16]

Retried try job too often on win_rel for step(s) unit_tests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win_rel&nu...

[commit-bot: I haz the power, 2013-05-20 02:26:17]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/kochi@chromium.org/14988010/34001

[commit-bot: I haz the power, 2013-05-20 05:49:50]

Change committed as 201035