Add chrome.input.ime.activate and chrome.input.ime.deactivate API

The IME extension uses chrome.input.ime.activate() to activate itself so that it can receive the events (e.g. the key events) and call input.ime related functions (e.g. commitText).
And IME extension uses chrome.input.ime.deactivate() to deactivate itself.
These two functions is Non-ChromeOS APIs.

This CL does the following work:
1. Add class input_method::InputMethodEngine which is derived from input_method::InputMethodBase for Linux & Windows functions.
2. Add extensions::InputImeEventRouter related functions.
3. Implement extensions::InputImeAPI related functions.
4. Add class extensions::InputImeActivateFunction and extensions::InputImeDeactivateFunction for implementation of chrome.input.ime.activate/deactivate API.

Some design points that is different from ChromeOS API:
1. There's only one or zero activated InputMethodEngine all the time, so functions CheckProfile() and IsActive() will always return true.
2. “input_components” section is ignored in the extension’s manifest, so the component_id/engineID will always be empty string.

BUG=517773
TEST=No compiling errors.

Committed: https://crrev.com/b8d5846de644695112e79e06d499c143d31a4748
Cr-Commit-Position: refs/heads/master@{#370626}

[Azure Wei, 2015-12-14 15:42:17]

Description was changed from

==========
Add chrome.input.ime.activate and chrome.input.ime.deactivate API

BUG=517773
TEST=No compiling errors.
==========

to

==========
Add chrome.input.ime.activate and chrome.input.ime.deactivate API

BUG=517773
TEST=No compiling errors.
==========

[Azure Wei, 2015-12-14 15:42:18]

azurewei@chromium.org changed reviewers:
+ shuchen@chromium.org

[Azure Wei, 2015-12-16 15:07:00]

Description was changed from

==========
Add chrome.input.ime.activate and chrome.input.ime.deactivate API

BUG=517773
TEST=No compiling errors.
==========

to

==========
The IME extension uses chrome.input.ime.activate() to activate itself so that it
can receive the events (e.g. the key events) and call input.ime related
functions (e.g. commitText).

This CL does the following work:
1. Add class ::input_method::InputMethodInterface which inherits from
ui::IMEEngineHandlerInterface for implementing all-platform functions.
2. Make class chromeos::InputMethodEngine derived from
::input_method::InputMethodInterface for chromeos functions. And add class
::input_method::InputMethodEngine derived from
::input_method::InputMethodInterface for Linux & Windows functions.
3.  Refactor the definition of class extensions::InputImeEventRouter from
input_ime_api.h to input_ime_api_chromeos.h and input_ime_api_nonchromeos.h
respectively, as they have different functions and variables.
4. Change extensions::InputImeAPI related functions on non-chromeos platforms
5. Add class extensions::InputImeActivateFunction and
extensions::InputImeDeactivateFunction for implementation of
chrome.input.ime.activate/deactivate.

BUG=517773
TEST=No compiling errors.
==========

[Azure Wei, 2015-12-16 15:51:05]

azurewei@chromium.org changed reviewers:
+ isherman@chromium.org, thakis@chromium.org

[Azure Wei, 2015-12-16 15:51:06]

Hi Shu, Nico and Ilya Sherman,

can you please help to review this cl? Thanks

[Ilya Sherman, 2015-12-16 16:35:20]

histograms.xml lgtm

[Azure Wei, 2015-12-17 02:05:30]

The CQ bit was checked by azurewei@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2015-12-17 02:36:17]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1528483002/60001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1528483002/60001

[commit-bot: I haz the power, 2015-12-17 03:09:32]

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

[commit-bot: I haz the power, 2015-12-17 03:09:33]

Dry run: Try jobs failed on following builders:
  linux_chromium_compile_dbg_32_ng on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)
  linux_chromium_compile_dbg_ng on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)
  linux_chromium_rel_ng on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)
  win_chromium_rel_ng on tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_rel_...)

[Azure Wei, 2015-12-18 05:28:40]

The CQ bit was checked by azurewei@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2015-12-18 05:28:56]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1528483002/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1528483002/80001

[commit-bot: I haz the power, 2015-12-18 05:47:19]

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

[commit-bot: I haz the power, 2015-12-18 05:47:20]

Dry run: Try jobs failed on following builders:
  chromeos_daisy_chromium_compile_only_ng on tryserver.chromium.linux
(JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromeos_daisy_...)
  chromeos_x86-generic_chromium_compile_only_ng on tryserver.chromium.linux
(JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromeos_x86-ge...)

[Azure Wei, 2015-12-19 04:59:13]

The CQ bit was checked by azurewei@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2015-12-19 04:59:22]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1528483002/100001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1528483002/100001

[commit-bot: I haz the power, 2015-12-19 05:00:47]

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

[commit-bot: I haz the power, 2015-12-19 05:00:47]

Dry run: Try jobs failed on following builders:
  mac_chromium_compile_dbg_ng on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_comp...)
  mac_chromium_gn_rel on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_gn_r...)
  mac_chromium_rel_ng on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_rel_...)

[Azure Wei, 2015-12-19 07:18:12]

The CQ bit was checked by azurewei@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2015-12-19 07:18:27]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1528483002/120001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1528483002/120001

[commit-bot: I haz the power, 2015-12-19 07:29:08]

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

[commit-bot: I haz the power, 2015-12-19 07:29:12]

Dry run: Try jobs failed on following builders:
  mac_chromium_gn_rel on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_gn_r...)

[Azure Wei, 2015-12-19 12:51:31]

The CQ bit was checked by azurewei@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2015-12-19 12:51:36]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1528483002/140001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1528483002/140001

[commit-bot: I haz the power, 2015-12-19 13:50:22]

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

[commit-bot: I haz the power, 2015-12-19 13:50:23]

Dry run: Try jobs failed on following builders:
  win_chromium_compile_dbg_ng on tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_comp...)

[Azure Wei, 2015-12-20 07:24:17]

The CQ bit was checked by azurewei@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2015-12-20 07:24:28]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1528483002/160001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1528483002/160001

[commit-bot: I haz the power, 2015-12-20 08:09:34]

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

[commit-bot: I haz the power, 2015-12-20 08:09:35]

Dry run: Try jobs failed on following builders:
  win_chromium_x64_rel_ng on tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_x64_...)

[Azure Wei, 2015-12-25 11:33:47]

The CQ bit was checked by azurewei@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2015-12-25 11:33:54]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1528483002/180001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1528483002/180001

[commit-bot: I haz the power, 2015-12-25 14:10:00]

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

[commit-bot: I haz the power, 2015-12-25 14:10:02]

Dry run: Try jobs failed on following builders:
  win8_chromium_ng on tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win8_chromium_ng/...)

[Azure Wei, 2015-12-30 06:48:36]

The CQ bit was checked by azurewei@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2015-12-30 06:48:40]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1528483002/200001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1528483002/200001

[commit-bot: I haz the power, 2015-12-30 06:59:24]

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

[commit-bot: I haz the power, 2015-12-30 06:59:25]

Dry run: Try jobs failed on following builders:
  chromeos_amd64-generic_chromium_compile_only_ng on tryserver.chromium.linux
(JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromeos_amd64-...)
  chromeos_daisy_chromium_compile_only_ng on tryserver.chromium.linux
(JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromeos_daisy_...)
  chromeos_x86-generic_chromium_compile_only_ng on tryserver.chromium.linux
(JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromeos_x86-ge...)

[Azure Wei, 2015-12-31 09:10:57]

The CQ bit was checked by azurewei@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2015-12-31 09:11:09]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1528483002/220001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1528483002/220001

[commit-bot: I haz the power, 2015-12-31 09:15:44]

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

[commit-bot: I haz the power, 2015-12-31 09:15:45]

Dry run: Try jobs failed on following builders:
  android_arm64_dbg_recipe on tryserver.chromium.android (JOB_FAILED,
https://build.chromium.org/p/tryserver.chromium.android/builders/android_arm6...)

[Azure Wei, 2016-01-04 13:33:38]

The CQ bit was checked by azurewei@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-01-04 13:33:52]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1528483002/240001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1528483002/240001

[commit-bot: I haz the power, 2016-01-04 13:38:38]

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

[commit-bot: I haz the power, 2016-01-04 13:38:40]

Dry run: Try jobs failed on following builders:
  android_clang_dbg_recipe on tryserver.chromium.android (JOB_FAILED,
https://build.chromium.org/p/tryserver.chromium.android/builders/android_clan...)

[Azure Wei, 2016-01-04 14:07:18]

The CQ bit was checked by azurewei@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-01-04 14:07:35]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1528483002/260001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1528483002/260001

[commit-bot: I haz the power, 2016-01-04 15:17:00]

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

[commit-bot: I haz the power, 2016-01-04 15:17:01]

Dry run: Try jobs failed on following builders:
  win8_chromium_ng on tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win8_chromium_ng/...)

[Azure Wei, 2016-01-05 06:28:34]

The CQ bit was checked by azurewei@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-01-05 06:28:46]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1528483002/280001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1528483002/280001

[commit-bot: I haz the power, 2016-01-05 07:30:44]

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

[commit-bot: I haz the power, 2016-01-05 07:30:46]

Dry run: Try jobs failed on following builders:
  win8_chromium_ng on tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win8_chromium_ng/...)

[Azure Wei, 2016-01-05 11:40:13]

azurewei@chromium.org changed reviewers:
+ rdevlin.cronin@chromium.org, rockot@chromium.org

[Azure Wei, 2016-01-05 13:07:34]

The CQ bit was checked by azurewei@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-01-05 13:07:57]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1528483002/300001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1528483002/300001

[commit-bot: I haz the power, 2016-01-05 14:12:26]

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

[commit-bot: I haz the power, 2016-01-05 14:12:27]

Dry run: This issue passed the CQ dry run.

[Azure Wei, 2016-01-05 14:16:01]

azurewei@chromium.org changed reviewers:
- rockot@chromium.org

[Azure Wei, 2016-01-05 14:17:14]

Hi Shu Chen, Devlin and Nico, can you please help review this cl? Thank you!

[Shu Chen, 2016-01-06 21:24:47]

I think this cl should be rebased on the CLs:
 - https://codereview.chromium.org/1554983002/
 - https://codereview.chromium.org/1562733002/

[Devlin, 2016-01-07 18:52:30]

On 2016/01/06 21:24:47, Shu Chen wrote:
> I think this cl should be rebased on the CLs:
>  - https://codereview.chromium.org/1554983002/
>  - https://codereview.chromium.org/1562733002/

+1.  I'm having a hard time parsing all this.

[Azure Wei, 2016-01-11 08:26:27]

Hi all, this CL has been rebased on these CLs:

https://codereview.chromium.org/1562733002/
https://codereview.chromium.org/1573903002/
https://codereview.chromium.org/1554983002/
https://codereview.chromium.org/1558413002/

can you please help review it? Thank you!

[Azure Wei, 2016-01-11 09:36:40]

Description was changed from

==========
The IME extension uses chrome.input.ime.activate() to activate itself so that it
can receive the events (e.g. the key events) and call input.ime related
functions (e.g. commitText).

This CL does the following work:
1. Add class ::input_method::InputMethodInterface which inherits from
ui::IMEEngineHandlerInterface for implementing all-platform functions.
2. Make class chromeos::InputMethodEngine derived from
::input_method::InputMethodInterface for chromeos functions. And add class
::input_method::InputMethodEngine derived from
::input_method::InputMethodInterface for Linux & Windows functions.
3.  Refactor the definition of class extensions::InputImeEventRouter from
input_ime_api.h to input_ime_api_chromeos.h and input_ime_api_nonchromeos.h
respectively, as they have different functions and variables.
4. Change extensions::InputImeAPI related functions on non-chromeos platforms
5. Add class extensions::InputImeActivateFunction and
extensions::InputImeDeactivateFunction for implementation of
chrome.input.ime.activate/deactivate.

BUG=517773
TEST=No compiling errors.
==========

to

==========
The IME extension uses chrome.input.ime.activate() to activate itself so that it
can receive the events (e.g. the key events) and call input.ime related
functions (e.g. commitText).
And IME extension uses chrome.input.ime.deactivate() to deactivate itself.

This CL does the following work:
1. Add class input_method::InputMethodEngine which is derived from
input_method::InputMethodBase  for Linux & Windows functions.
2. Implement extensions::InputImeAPI related functions on non-chromeos
platforms.
3. Add class extensions::InputImeActivateFunction and
extensions::InputImeDeactivateFunction for implementation of
chrome.input.ime.activate/deactivate API.

BUG=517773
TEST=No compiling errors.
==========

[Azure Wei, 2016-01-11 11:43:56]

Description was changed from

==========
The IME extension uses chrome.input.ime.activate() to activate itself so that it
can receive the events (e.g. the key events) and call input.ime related
functions (e.g. commitText).
And IME extension uses chrome.input.ime.deactivate() to deactivate itself.

This CL does the following work:
1. Add class input_method::InputMethodEngine which is derived from
input_method::InputMethodBase  for Linux & Windows functions.
2. Implement extensions::InputImeAPI related functions on non-chromeos
platforms.
3. Add class extensions::InputImeActivateFunction and
extensions::InputImeDeactivateFunction for implementation of
chrome.input.ime.activate/deactivate API.

BUG=517773
TEST=No compiling errors.
==========

to

==========
The IME extension uses chrome.input.ime.activate() to activate itself so that it
can receive the events (e.g. the key events) and call input.ime related
functions (e.g. commitText).
And IME extension uses chrome.input.ime.deactivate() to deactivate itself.
These two functions is Non-ChromeOS APIs.

This CL does the following work:
1. Add class input_method::InputMethodEngine which is derived from
input_method::InputMethodBase for Linux & Windows functions.
2. Add extensions::InputImeEventRouter related functions. 
3. Implement extensions::InputImeAPI related functions.
4. Add class extensions::InputImeActivateFunction and
extensions::InputImeDeactivateFunction for implementation of
chrome.input.ime.activate/deactivate API.

Some design points that is different from ChromeOS API:
1. “input_components” section will be ignored in the extension’s manifest.
2. There's only one or zero activated InputMethodEngine all the time.
3. The component_id/engineID will always be empty string.

BUG=517773
TEST=No compiling errors.
==========

[Azure Wei, 2016-01-11 11:46:28]

Description was changed from

==========
The IME extension uses chrome.input.ime.activate() to activate itself so that it
can receive the events (e.g. the key events) and call input.ime related
functions (e.g. commitText).
And IME extension uses chrome.input.ime.deactivate() to deactivate itself.
These two functions is Non-ChromeOS APIs.

This CL does the following work:
1. Add class input_method::InputMethodEngine which is derived from
input_method::InputMethodBase for Linux & Windows functions.
2. Add extensions::InputImeEventRouter related functions. 
3. Implement extensions::InputImeAPI related functions.
4. Add class extensions::InputImeActivateFunction and
extensions::InputImeDeactivateFunction for implementation of
chrome.input.ime.activate/deactivate API.

Some design points that is different from ChromeOS API:
1. “input_components” section will be ignored in the extension’s manifest.
2. There's only one or zero activated InputMethodEngine all the time.
3. The component_id/engineID will always be empty string.

BUG=517773
TEST=No compiling errors.
==========

to

==========
The IME extension uses chrome.input.ime.activate() to activate itself so that it
can receive the events (e.g. the key events) and call input.ime related
functions (e.g. commitText).
And IME extension uses chrome.input.ime.deactivate() to deactivate itself.
These two functions is Non-ChromeOS APIs.

This CL does the following work:
1. Add class input_method::InputMethodEngine which is derived from
input_method::InputMethodBase for Linux & Windows functions.
2. Add extensions::InputImeEventRouter related functions. 
3. Implement extensions::InputImeAPI related functions.
4. Add class extensions::InputImeActivateFunction and
extensions::InputImeDeactivateFunction for implementation of
chrome.input.ime.activate/deactivate API.

Some design points that is different from ChromeOS API:
1. There's only one or zero activated InputMethodEngine all the time, so
functions CheckProfile() and IsActive() will always return true.
2. “input_components” section is ignored in the extension’s manifest, so the
component_id/engineID will always be empty string.

BUG=517773
TEST=No compiling errors.
==========

[Devlin, 2016-01-11 21:48:32]

rdevlin.cronin@chromium.org changed reviewers:
+ rockot@chromium.org

[Devlin, 2016-01-11 21:48:34]

Ken speaks much better gyp/gn than I, so adding him for the *.g*

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
File chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc
(right):

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:22:
namespace {
\n

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:71: VLOG(1)
<< "RegisterImeExtension: " << extension_id;
Remove debugging logs.

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:73: //
Already regsitered
typo + add period.

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:90: engine
= nullptr;
What is this supposed to do?  It's just a local raw ptr, so resetting it
(AFAICT) doesn't have any effect...

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:101: return
engine;
ternary.

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:111:
input_method::InputMethodEngine* input_engine =
nit: Wrap in a scoped_ptr until ownership is passed.

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:118: engine
= ui::IMEBridge::Get()->GetCurrentEngineHandler();
Is there any way this is different from |input_engine|?  If so, document how.

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:148:
ui::IMEBridge::Initialize();
If this is guaranteed to be a no-op, put a comment.  Otherwise, ensure we only
call this once.

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
File chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.h
(right):

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.h:45: bool
RunSync() override;
Read comments in extension_function.h above RunSync() (TL;DR:
SyncExtensionFunction is deprecated).

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/input_m...
File chrome/browser/input_method/input_method_engine.h (right):

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/input_m...
chrome/browser/input_method/input_method_engine.h:8: #include <map>
Trim includes

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/input_m...
chrome/browser/input_method/input_method_engine.h:45: bool CheckProfile() const
override;
DISALLOW_COPY_AND_ASSIGN

https://codereview.chromium.org/1528483002/diff/400001/chrome/common/extensio...
File chrome/common/extensions/api/input_ime.json (right):

https://codereview.chromium.org/1528483002/diff/400001/chrome/common/extensio...
chrome/common/extensions/api/input_ime.json:549: "description": "Activates the
self IME extension, so that it can receive events.",
what does "the self IME extension" mean?

[Ken Rockot(use gerrit already), 2016-01-11 21:58:36]

https://codereview.chromium.org/1528483002/diff/400001/chrome/chrome_browser_...
File chrome/chrome_browser_extensions.gypi (right):

https://codereview.chromium.org/1528483002/diff/400001/chrome/chrome_browser_...
chrome/chrome_browser_extensions.gypi:1139:
'../ui/base/ime/ime_engine_handler_interface.cc',
This is almost definitely wrong, in GN as well. You should be depending on the
ui/base/ime target ('../ui/base/ime/ui_base_ime.gyp:ui_base_ime' for gyp and
"//ui/base/ime" for GN).

[Azure Wei, 2016-01-13 02:28:10]

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
File chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc
(right):

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:22:
namespace {
On 2016/01/11 21:48:33, Devlin wrote:
> \n

Done.

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:71: VLOG(1)
<< "RegisterImeExtension: " << extension_id;
On 2016/01/11 21:48:33, Devlin wrote:
> Remove debugging logs.

Done.

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:73: //
Already regsitered
On 2016/01/11 21:48:33, Devlin wrote:
> typo + add period.

Done.

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:90: engine
= nullptr;
On 2016/01/11 21:48:33, Devlin wrote:
> What is this supposed to do?  It's just a local raw ptr, so resetting it
> (AFAICT) doesn't have any effect...

Changed it with ui::IMEBridge()::Get()->SetCurrentEngineHandler(nullptr);

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:101: return
engine;
On 2016/01/11 21:48:33, Devlin wrote:
> ternary.

Done.

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:111:
input_method::InputMethodEngine* input_engine =
On 2016/01/11 21:48:33, Devlin wrote:
> nit: Wrap in a scoped_ptr until ownership is passed.

Done.

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:118: engine
= ui::IMEBridge::Get()->GetCurrentEngineHandler();
On 2016/01/11 21:48:33, Devlin wrote:
> Is there any way this is different from |input_engine|?  If so, document how.

The only difference is that |input_engine| has Initialize() function. I've
changed the sequence of SetCurrentEngineHandler() with engine->Enable() to avoid
using two engine.

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:148:
ui::IMEBridge::Initialize();
On 2016/01/11 21:48:33, Devlin wrote:
> If this is guaranteed to be a no-op, put a comment.  Otherwise, ensure we only
> call this once.

Done.

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
File chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.h
(right):

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.h:45: bool
RunSync() override;
On 2016/01/11 21:48:34, Devlin wrote:
> Read comments in extension_function.h above RunSync() (TL;DR:
> SyncExtensionFunction is deprecated).

Done.

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/input_m...
File chrome/browser/input_method/input_method_engine.h (right):

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/input_m...
chrome/browser/input_method/input_method_engine.h:8: #include <map>
On 2016/01/11 21:48:34, Devlin wrote:
> Trim includes

Done.

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/input_m...
chrome/browser/input_method/input_method_engine.h:45: bool CheckProfile() const
override;
On 2016/01/11 21:48:34, Devlin wrote:
> DISALLOW_COPY_AND_ASSIGN

Done.

https://codereview.chromium.org/1528483002/diff/400001/chrome/chrome_browser_...
File chrome/chrome_browser_extensions.gypi (right):

https://codereview.chromium.org/1528483002/diff/400001/chrome/chrome_browser_...
chrome/chrome_browser_extensions.gypi:1139:
'../ui/base/ime/ime_engine_handler_interface.cc',
On 2016/01/11 21:58:36, Ken Rockot wrote:
> This is almost definitely wrong, in GN as well. You should be depending on the
> ui/base/ime target ('../ui/base/ime/ui_base_ime.gyp:ui_base_ime' for gyp and
> "//ui/base/ime" for GN).

"../ui/base/ime/ui_base_ime.gyp:ui_base_ime" was used for gyp and
"//ui/base/ime" was in the general deps in GN.
If I remove these "sources: ime_engine_handler_interface.cc", it will fail on
Windows building with errors like "unresolved external symbol
IMEEngineHandlerInterface::KeyboardEvent::KeyboardEvent()".
And if I remove the conditions "component=="shared_library", it will fail on
Win8 building with errors like "IMEEngineHandlerInterface already defined".

https://codereview.chromium.org/1528483002/diff/400001/chrome/common/extensio...
File chrome/common/extensions/api/input_ime.json (right):

https://codereview.chromium.org/1528483002/diff/400001/chrome/common/extensio...
chrome/common/extensions/api/input_ime.json:549: "description": "Activates the
self IME extension, so that it can receive events.",
On 2016/01/11 21:48:34, Devlin wrote:
> what does "the self IME extension" mean?

I've changed it with 'Activates the IME extension itself', will it be clearer
than before?

[Devlin, 2016-01-13 18:22:59]

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
File chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc
(right):

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:90: engine
= nullptr;
On 2016/01/13 02:28:10, Azure Wei wrote:
> On 2016/01/11 21:48:33, Devlin wrote:
> > What is this supposed to do?  It's just a local raw ptr, so resetting it
> > (AFAICT) doesn't have any effect...
> 
> Changed it with ui::IMEBridge()::Get()->SetCurrentEngineHandler(nullptr);

Heh that's quite a difference...

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:148:
ui::IMEBridge::Initialize();
On 2016/01/13 02:28:10, Azure Wei wrote:
> On 2016/01/11 21:48:33, Devlin wrote:
> > If this is guaranteed to be a no-op, put a comment.  Otherwise, ensure we
only
> > call this once.
> 
> Done.

I'd prefer just a "// No-op if called multiple times." rather than including
implementation details about the method here.

https://codereview.chromium.org/1528483002/diff/400001/chrome/common/extensio...
File chrome/common/extensions/api/input_ime.json (right):

https://codereview.chromium.org/1528483002/diff/400001/chrome/common/extensio...
chrome/common/extensions/api/input_ime.json:549: "description": "Activates the
self IME extension, so that it can receive events.",
On 2016/01/13 02:28:10, Azure Wei wrote:
> On 2016/01/11 21:48:34, Devlin wrote:
> > what does "the self IME extension" mean?
> 
> I've changed it with 'Activates the IME extension itself', will it be clearer
> than before?

I would just remove the "itself", too.  Just "Activates the IME extension so
that it can receive events" is fine.

https://codereview.chromium.org/1528483002/diff/420001/chrome/browser/extensi...
File chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc
(right):

https://codereview.chromium.org/1528483002/diff/420001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:73: //
Already regsitered.
Typo: registered
Also, let's make this a bit more clear; "Check if the IME extension is already
registered."

https://codereview.chromium.org/1528483002/diff/420001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:91: delete
engine;
This lifetime management is pretty wonky.  Do we have a guarantee that
UnregisterImeExtension will always be called so that the engine doesn't leak?

https://codereview.chromium.org/1528483002/diff/420001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:123:
return;
?

https://codereview.chromium.org/1528483002/diff/420001/chrome/browser/extensi...
File chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.h
(right):

https://codereview.chromium.org/1528483002/diff/420001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.h:22: bool
RegisterImeExtension(const std::string& extension_id);
all these functions should be documented.

https://codereview.chromium.org/1528483002/diff/420001/chrome/browser/input_m...
File chrome/browser/input_method/input_method_engine.h (right):

https://codereview.chromium.org/1528483002/diff/420001/chrome/browser/input_m...
chrome/browser/input_method/input_method_engine.h:24: 
nit: no \n

https://codereview.chromium.org/1528483002/diff/420001/chrome/common/extensio...
File chrome/common/extensions/api/input_ime.json (right):

https://codereview.chromium.org/1528483002/diff/420001/chrome/common/extensio...
chrome/common/extensions/api/input_ime.json:547: "name": "activate",
Could we instead just have one function, setActive?

[Azure Wei, 2016-01-14 14:03:48]

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
File chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc
(right):

https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:148:
ui::IMEBridge::Initialize();
On 2016/01/13 18:22:59, Devlin wrote:
> On 2016/01/13 02:28:10, Azure Wei wrote:
> > On 2016/01/11 21:48:33, Devlin wrote:
> > > If this is guaranteed to be a no-op, put a comment.  Otherwise, ensure we
> only
> > > call this once.
> > 
> > Done.
> 
> I'd prefer just a "// No-op if called multiple times." rather than including
> implementation details about the method here.

Done.

https://codereview.chromium.org/1528483002/diff/400001/chrome/common/extensio...
File chrome/common/extensions/api/input_ime.json (right):

https://codereview.chromium.org/1528483002/diff/400001/chrome/common/extensio...
chrome/common/extensions/api/input_ime.json:549: "description": "Activates the
self IME extension, so that it can receive events.",
On 2016/01/13 18:22:59, Devlin wrote:
> On 2016/01/13 02:28:10, Azure Wei wrote:
> > On 2016/01/11 21:48:34, Devlin wrote:
> > > what does "the self IME extension" mean?
> > 
> > I've changed it with 'Activates the IME extension itself', will it be
clearer
> > than before?
> 
> I would just remove the "itself", too.  Just "Activates the IME extension so
> that it can receive events" is fine.

Done.

https://codereview.chromium.org/1528483002/diff/420001/chrome/browser/extensi...
File chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc
(right):

https://codereview.chromium.org/1528483002/diff/420001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:73: //
Already regsitered.
On 2016/01/13 18:22:59, Devlin wrote:
> Typo: registered
> Also, let's make this a bit more clear; "Check if the IME extension is already
> registered."

Done.

https://codereview.chromium.org/1528483002/diff/420001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:91: delete
engine;
On 2016/01/13 18:22:59, Devlin wrote:
> This lifetime management is pretty wonky.  Do we have a guarantee that
> UnregisterImeExtension will always be called so that the engine doesn't leak?

A new private function DeleteInputMethodEngine() is added. It will be called in
UnregisterImeExtension(), ~InputImeEventRouter() and SetActiveEngine(). I think
these will guarantee the InputMethodEngine will be deleted.

https://codereview.chromium.org/1528483002/diff/420001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:123:
return;
On 2016/01/13 18:22:59, Devlin wrote:
> ?

Deleted.

https://codereview.chromium.org/1528483002/diff/420001/chrome/browser/extensi...
File chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.h
(right):

https://codereview.chromium.org/1528483002/diff/420001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.h:22: bool
RegisterImeExtension(const std::string& extension_id);
On 2016/01/13 18:22:59, Devlin wrote:
> all these functions should be documented.

Done.

https://codereview.chromium.org/1528483002/diff/420001/chrome/browser/input_m...
File chrome/browser/input_method/input_method_engine.h (right):

https://codereview.chromium.org/1528483002/diff/420001/chrome/browser/input_m...
chrome/browser/input_method/input_method_engine.h:24: 
On 2016/01/13 18:22:59, Devlin wrote:
> nit: no \n

Done.

https://codereview.chromium.org/1528483002/diff/420001/chrome/common/extensio...
File chrome/common/extensions/api/input_ime.json (right):

https://codereview.chromium.org/1528483002/diff/420001/chrome/common/extensio...
chrome/common/extensions/api/input_ime.json:547: "name": "activate",
On 2016/01/13 18:22:59, Devlin wrote:
> Could we instead just have one function, setActive?

'setActive' is little wired since we don't have 'getActive' function. Instead,
we have 'onActivate' and 'onDeactivate' events. I think 'activate' and
'deactivate' seem more symmetric and consistent with the two events. Do we have
to change this?

[Ken Rockot(use gerrit already), 2016-01-14 16:30:41]

On 2016/01/13 at 02:28:10, azurewei wrote:
>
https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
> File chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc
(right):
> 
>
https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
> chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:22:
namespace {
> On 2016/01/11 21:48:33, Devlin wrote:
> > \n
> 
> Done.
> 
>
https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
> chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:71:
VLOG(1) << "RegisterImeExtension: " << extension_id;
> On 2016/01/11 21:48:33, Devlin wrote:
> > Remove debugging logs.
> 
> Done.
> 
>
https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
> chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:73: //
Already regsitered
> On 2016/01/11 21:48:33, Devlin wrote:
> > typo + add period.
> 
> Done.
> 
>
https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
> chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:90:
engine = nullptr;
> On 2016/01/11 21:48:33, Devlin wrote:
> > What is this supposed to do?  It's just a local raw ptr, so resetting it
> > (AFAICT) doesn't have any effect...
> 
> Changed it with ui::IMEBridge()::Get()->SetCurrentEngineHandler(nullptr);
> 
>
https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
> chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:101:
return engine;
> On 2016/01/11 21:48:33, Devlin wrote:
> > ternary.
> 
> Done.
> 
>
https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
> chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:111:
input_method::InputMethodEngine* input_engine =
> On 2016/01/11 21:48:33, Devlin wrote:
> > nit: Wrap in a scoped_ptr until ownership is passed.
> 
> Done.
> 
>
https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
> chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:118:
engine = ui::IMEBridge::Get()->GetCurrentEngineHandler();
> On 2016/01/11 21:48:33, Devlin wrote:
> > Is there any way this is different from |input_engine|?  If so, document
how.
> 
> The only difference is that |input_engine| has Initialize() function. I've
changed the sequence of SetCurrentEngineHandler() with engine->Enable() to avoid
using two engine.
> 
>
https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
> chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:148:
ui::IMEBridge::Initialize();
> On 2016/01/11 21:48:33, Devlin wrote:
> > If this is guaranteed to be a no-op, put a comment.  Otherwise, ensure we
only
> > call this once.
> 
> Done.
> 
>
https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
> File chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.h
(right):
> 
>
https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/extensi...
> chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.h:45: bool
RunSync() override;
> On 2016/01/11 21:48:34, Devlin wrote:
> > Read comments in extension_function.h above RunSync() (TL;DR:
> > SyncExtensionFunction is deprecated).
> 
> Done.
> 
>
https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/input_m...
> File chrome/browser/input_method/input_method_engine.h (right):
> 
>
https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/input_m...
> chrome/browser/input_method/input_method_engine.h:8: #include <map>
> On 2016/01/11 21:48:34, Devlin wrote:
> > Trim includes
> 
> Done.
> 
>
https://codereview.chromium.org/1528483002/diff/400001/chrome/browser/input_m...
> chrome/browser/input_method/input_method_engine.h:45: bool CheckProfile()
const override;
> On 2016/01/11 21:48:34, Devlin wrote:
> > DISALLOW_COPY_AND_ASSIGN
> 
> Done.
> 
>
https://codereview.chromium.org/1528483002/diff/400001/chrome/chrome_browser_...
> File chrome/chrome_browser_extensions.gypi (right):
> 
>
https://codereview.chromium.org/1528483002/diff/400001/chrome/chrome_browser_...
> chrome/chrome_browser_extensions.gypi:1139:
'../ui/base/ime/ime_engine_handler_interface.cc',
> On 2016/01/11 21:58:36, Ken Rockot wrote:
> > This is almost definitely wrong, in GN as well. You should be depending on
the
> > ui/base/ime target ('../ui/base/ime/ui_base_ime.gyp:ui_base_ime' for gyp and
> > "//ui/base/ime" for GN).
> 
> "../ui/base/ime/ui_base_ime.gyp:ui_base_ime" was used for gyp and
"//ui/base/ime" was in the general deps in GN.
> If I remove these "sources: ime_engine_handler_interface.cc", it will fail on
Windows building with errors like "unresolved external symbol
IMEEngineHandlerInterface::KeyboardEvent::KeyboardEvent()".
> And if I remove the conditions "component=="shared_library", it will fail on
Win8 building with errors like "IMEEngineHandlerInterface already defined".
> 

It's important to understand WHY a build is failing in these cases. Blindly and
statically linking in the symbols you want is ABSOLUTELY NOT OK and please don't
ever do this. You are essentially adding a duplicate definition of the type and
this can cause all kinds of miserable and subtle problems down the road.

In this case, the problem is that KeyboardEvent is not exported by the component
library. You could fix this by adding a UI_BASE_IME_EXPORT annotation to
KeyboardEvent's definition. But that does change the public API of ui/base/ime.
While this might be OK, it's also possible that the symbol isn't exported for a
reason. Hopefully shuchen@ can weigh in on this.

>
https://codereview.chromium.org/1528483002/diff/400001/chrome/common/extensio...
> File chrome/common/extensions/api/input_ime.json (right):
> 
>
https://codereview.chromium.org/1528483002/diff/400001/chrome/common/extensio...
> chrome/common/extensions/api/input_ime.json:549: "description": "Activates the
self IME extension, so that it can receive events.",
> On 2016/01/11 21:48:34, Devlin wrote:
> > what does "the self IME extension" mean?
> 
> I've changed it with 'Activates the IME extension itself', will it be clearer
than before?

[Azure Wei, 2016-01-15 13:13:24]

On 2016/01/14 16:30:41, Ken Rockot wrote:
> On 2016/01/13 at 02:28:10, azurewei wrote:
> 
> 
> It's important to understand WHY a build is failing in these cases. Blindly
and
> statically linking in the symbols you want is ABSOLUTELY NOT OK and please
don't
> ever do this. You are essentially adding a duplicate definition of the type
and
> this can cause all kinds of miserable and subtle problems down the road.
> 
> In this case, the problem is that KeyboardEvent is not exported by the
component
> library. You could fix this by adding a UI_BASE_IME_EXPORT annotation to
> KeyboardEvent's definition. But that does change the public API of
ui/base/ime.
> While this might be OK, it's also possible that the symbol isn't exported for
a
> reason. Hopefully shuchen@ can weigh in on this.
> 

Thanks a lot for your comments. Another CL:
https://codereview.chromium.org/1590863002/ is created to move the struct
definition from ui::IMEEngineHandlerInterface to
input_method::InputMethodEngineInterface. And this CL has been rebased on that
CL and updated the gyp & gn file. 
Can you please review the newest patch set? Thank you!

[Devlin, 2016-01-15 22:06:02]

extensions lgtm, but please wait for rockot@ for the .g*.

https://codereview.chromium.org/1528483002/diff/420001/chrome/common/extensio...
File chrome/common/extensions/api/input_ime.json (right):

https://codereview.chromium.org/1528483002/diff/420001/chrome/common/extensio...
chrome/common/extensions/api/input_ime.json:547: "name": "activate",
On 2016/01/14 14:03:48, Azure Wei wrote:
> On 2016/01/13 18:22:59, Devlin wrote:
> > Could we instead just have one function, setActive?
> 
> 'setActive' is little wired since we don't have 'getActive' function. Instead,
> we have 'onActivate' and 'onDeactivate' events. I think 'activate' and
> 'deactivate' seem more symmetric and consistent with the two events. Do we
have
> to change this?

I don't really think setActive is weird, and we try to keep these APIs concise
when we can.  I have a slight preference for setActive over activate/deactivate,
but I don't feel strongly, so either way is fine.

https://codereview.chromium.org/1528483002/diff/500001/chrome/browser/extensi...
File chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc
(right):

https://codereview.chromium.org/1528483002/diff/500001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:124:
ui::IMEBridge::Get()->SetCurrentEngineHandler(nullptr);
I'd prefer this to be the first call so that if SetCurrentEngineHandler() ever
tries to do cleanup on the engine, it has a valid pointer.

https://codereview.chromium.org/1528483002/diff/500001/chrome/browser/extensi...
File chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.h
(right):

https://codereview.chromium.org/1528483002/diff/500001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.h:26: //
Registers the extension by saving the |extension_id| in the vector
When making class comments, try to describe the intent of the function, rather
than the implementation details (since those will almost certainly be changed
without updating the comment).  So, for these, I would do something like:
// Registers the extension as an IME extension, allowing it to be the active
engine.
RegisterImeExetnsion(id);

// Unregisters the extension as an IME extension and deactivates the IME engine
for it, if it was active.

https://codereview.chromium.org/1528483002/diff/500001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.h:51:
std::vector<std::string> extension_ids_;
Out of curiosity, why do we even need this list (or RegisterImeExtension)?  It
looks like all we do is check for the active engine in all cases.  Are we going
to use this later?

[Shu Chen, 2016-01-18 09:21:59]

https://codereview.chromium.org/1528483002/diff/500001/chrome/browser/extensi...
File chrome/browser/extensions/BUILD.gn (right):

https://codereview.chromium.org/1528483002/diff/500001/chrome/browser/extensi...
chrome/browser/extensions/BUILD.gn:158: deps += [
"//ui/events:dom_keycode_converter" ]
missing the same deps for "is_win"?

https://codereview.chromium.org/1528483002/diff/500001/chrome/browser/extensi...
File chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc
(right):

https://codereview.chromium.org/1528483002/diff/500001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:135:
event_router->SetActiveEngine(extension_id());
RegisterImeExtension & SetActiveEngine are always called together.
Maybe they can be merged into one method?

https://codereview.chromium.org/1528483002/diff/500001/ui/base/ime/ime_engine...
File ui/base/ime/ime_engine_handler_interface.h (right):

https://codereview.chromium.org/1528483002/diff/500001/ui/base/ime/ime_engine...
ui/base/ime/ime_engine_handler_interface.h:143: // Get the id of the activated
extension.
s/activated extension/IME extension

An engine handler represents an IME extension. GetExtensionId() should return
the extension id despite of whether it is activated.

[Azure Wei, 2016-01-19 03:46:54]

https://codereview.chromium.org/1528483002/diff/500001/chrome/browser/extensi...
File chrome/browser/extensions/BUILD.gn (right):

https://codereview.chromium.org/1528483002/diff/500001/chrome/browser/extensi...
chrome/browser/extensions/BUILD.gn:158: deps += [
"//ui/events:dom_keycode_converter" ]
On 2016/01/18 09:21:59, Shu Chen wrote:
> missing the same deps for "is_win"?

Added. Thanks for pointing it out.

https://codereview.chromium.org/1528483002/diff/500001/chrome/browser/extensi...
File chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc
(right):

https://codereview.chromium.org/1528483002/diff/500001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:124:
ui::IMEBridge::Get()->SetCurrentEngineHandler(nullptr);
On 2016/01/15 22:06:02, Devlin wrote:
> I'd prefer this to be the first call so that if SetCurrentEngineHandler() ever
> tries to do cleanup on the engine, it has a valid pointer.

Done.

https://codereview.chromium.org/1528483002/diff/500001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:135:
event_router->SetActiveEngine(extension_id());
On 2016/01/18 09:21:59, Shu Chen wrote:
> RegisterImeExtension & SetActiveEngine are always called together.
> Maybe they can be merged into one method?

Done.

https://codereview.chromium.org/1528483002/diff/500001/chrome/browser/extensi...
File chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.h
(right):

https://codereview.chromium.org/1528483002/diff/500001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.h:26: //
Registers the extension by saving the |extension_id| in the vector
On 2016/01/15 22:06:02, Devlin wrote:
> When making class comments, try to describe the intent of the function, rather
> than the implementation details (since those will almost certainly be changed
> without updating the comment).  So, for these, I would do something like:
> // Registers the extension as an IME extension, allowing it to be the active
> engine.
> RegisterImeExetnsion(id);
> 
> // Unregisters the extension as an IME extension and deactivates the IME
engine
> for it, if it was active.

Done. Thank you for your suggestion.

https://codereview.chromium.org/1528483002/diff/500001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.h:51:
std::vector<std::string> extension_ids_;
On 2016/01/15 22:06:02, Devlin wrote:
> Out of curiosity, why do we even need this list (or RegisterImeExtension)?  It
> looks like all we do is check for the active engine in all cases.  Are we
going
> to use this later?

Yeah, extension_ids_ doesn't really work now. It will be used for switching IME
extensions. For instance, if we want to change IME extensions with short-cuts,
we will need the IME extensions list that used to be registered/activated.

https://codereview.chromium.org/1528483002/diff/500001/ui/base/ime/ime_engine...
File ui/base/ime/ime_engine_handler_interface.h (right):

https://codereview.chromium.org/1528483002/diff/500001/ui/base/ime/ime_engine...
ui/base/ime/ime_engine_handler_interface.h:143: // Get the id of the activated
extension.
On 2016/01/18 09:21:59, Shu Chen wrote:
> s/activated extension/IME extension
> 
> An engine handler represents an IME extension. GetExtensionId() should return
> the extension id despite of whether it is activated.

Done.

[Shu Chen, 2016-01-19 04:03:06]

lgtm

https://codereview.chromium.org/1528483002/diff/520001/chrome/browser/extensi...
File chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc
(right):

https://codereview.chromium.org/1528483002/diff/520001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:149: void
InputImeAPI::OnExtensionLoaded(content::BrowserContext* browser_context,
Please keep the declare order as:
 - InputImeAPI
 - InputImeEventRouter
 - InputImeXxxFunction

[Ken Rockot(use gerrit already), 2016-01-19 04:06:49]

https://codereview.chromium.org/1528483002/diff/520001/chrome/chrome_browser_...
File chrome/chrome_browser_extensions.gypi (right):

https://codereview.chromium.org/1528483002/diff/520001/chrome/chrome_browser_...
chrome/chrome_browser_extensions.gypi:952:
'browser/input_method/input_method_engine.cc',
This is not the right place for these source files to be listed. They aren't
extensions sources.

As far as I can tell they should be in chrome_browser.gypi under
chrome_browser_chromeos_sources.

[Azure Wei, 2016-01-19 05:24:58]

https://codereview.chromium.org/1528483002/diff/520001/chrome/browser/extensi...
File chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc
(right):

https://codereview.chromium.org/1528483002/diff/520001/chrome/browser/extensi...
chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:149: void
InputImeAPI::OnExtensionLoaded(content::BrowserContext* browser_context,
On 2016/01/19 04:03:06, Shu Chen wrote:
> Please keep the declare order as:
>  - InputImeAPI
>  - InputImeEventRouter
>  - InputImeXxxFunction

Done.

https://codereview.chromium.org/1528483002/diff/520001/chrome/chrome_browser_...
File chrome/chrome_browser_extensions.gypi (right):

https://codereview.chromium.org/1528483002/diff/520001/chrome/chrome_browser_...
chrome/chrome_browser_extensions.gypi:952:
'browser/input_method/input_method_engine.cc',
On 2016/01/19 04:06:49, Ken Rockot wrote:
> This is not the right place for these source files to be listed. They aren't
> extensions sources.
> 
> As far as I can tell they should be in chrome_browser.gypi under
> chrome_browser_chromeos_sources.

I think chrome_browser_chromeos_sources is for Chrome OS platform. These files
should be built on Linux and Widnows platforms. I've moved these to
chrome_browser.gypi under 'chrome_browser_linux_desktop_sources' and
'chrome_browser_win_sources'. Is that OK?

[Azure Wei, 2016-01-19 05:27:26]

azurewei@chromium.org changed reviewers:
- thakis@chromium.org

[Ken Rockot(use gerrit already), 2016-01-19 05:37:30]

On 2016/01/19 at 05:24:58, azurewei wrote:
>
https://codereview.chromium.org/1528483002/diff/520001/chrome/browser/extensi...
> File chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc
(right):
> 
>
https://codereview.chromium.org/1528483002/diff/520001/chrome/browser/extensi...
> chrome/browser/extensions/api/input_ime/input_ime_api_nonchromeos.cc:149: void
InputImeAPI::OnExtensionLoaded(content::BrowserContext* browser_context,
> On 2016/01/19 04:03:06, Shu Chen wrote:
> > Please keep the declare order as:
> >  - InputImeAPI
> >  - InputImeEventRouter
> >  - InputImeXxxFunction
> 
> Done.
> 
>
https://codereview.chromium.org/1528483002/diff/520001/chrome/chrome_browser_...
> File chrome/chrome_browser_extensions.gypi (right):
> 
>
https://codereview.chromium.org/1528483002/diff/520001/chrome/chrome_browser_...
> chrome/chrome_browser_extensions.gypi:952:
'browser/input_method/input_method_engine.cc',
> On 2016/01/19 04:06:49, Ken Rockot wrote:
> > This is not the right place for these source files to be listed. They aren't
> > extensions sources.
> > 
> > As far as I can tell they should be in chrome_browser.gypi under
> > chrome_browser_chromeos_sources.
> 
> I think chrome_browser_chromeos_sources is for Chrome OS platform. These files
should be built on Linux and Widnows platforms. I've moved these to
chrome_browser.gypi under 'chrome_browser_linux_desktop_sources' and
'chrome_browser_win_sources'. Is that OK?

Yes, thanks! LGTM

[Azure Wei, 2016-01-19 05:42:03]

azurewei@chromium.org changed reviewers:
+ sky@chromium.org

[Azure Wei, 2016-01-19 05:43:40]

+sky@ for file: chrome/browser/BUILD.gn

Hi sky@, can you please help review the file: chrome/browser/BUILD.gn, which has
been changed in patch set 28. Thank you!

[sky, 2016-01-19 17:20:41]

Is there a reason you want this in chrome/browser/input_method vs
chrome/browser/ui/input_method? I favor the later as this code seems all ui
related. Or am I missing something?

[Shu Chen, 2016-01-20 00:18:29]

On 2016/01/19 17:20:41, sky wrote:
> Is there a reason you want this in chrome/browser/input_method vs
> chrome/browser/ui/input_method? I favor the later as this code seems all ui
> related. Or am I missing something?

I'm not sure.

The input method stuff for chromeos is under
chrome/browser/chromeos/input_method/...
instead of chrome/browser/chromeos/ui/input_method/...
I guess it is because that those code is for extension APIs & input method
management & utils & views.

For non-chromeos, the input method stuff is only for extension APIs (e.g.
passing events, invoking TextInputClient).
The code under chrome/browser/ui/input_method/... seems more related to UIs
including windows/views.
Is my understand correct?

[sky, 2016-01-20 00:49:04]

The code under c/b/ui is not all platform specific. For example,
c/b/ui/tabs contains ui code that is used by all platforms.
c/b/extensions consumes a far amount of code in c/b/ui already.

On Tue, Jan 19, 2016 at 4:18 PM,  <shuchen@chromium.org> wrote:
> On 2016/01/19 17:20:41, sky wrote:
>>
>> Is there a reason you want this in chrome/browser/input_method vs
>> chrome/browser/ui/input_method? I favor the later as this code seems all
>> ui
>> related. Or am I missing something?
>
>
> I'm not sure.
>
> The input method stuff for chromeos is under
> chrome/browser/chromeos/input_method/...
> instead of chrome/browser/chromeos/ui/input_method/...
> I guess it is because that those code is for extension APIs & input method
> management & utils & views.
>
> For non-chromeos, the input method stuff is only for extension APIs (e.g.
> passing events, invoking TextInputClient).
> The code under chrome/browser/ui/input_method/... seems more related to UIs
> including windows/views.
> Is my understand correct?
>
>
> https://codereview.chromium.org/1528483002/

-- 
You received this message because you are subscribed to the Google Groups "Blink
Reviews" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to blink-reviews+unsubscribe@chromium.org.

[sky, 2016-01-20 00:49:05]

The code under c/b/ui is not all platform specific. For example,
c/b/ui/tabs contains ui code that is used by all platforms.
c/b/extensions consumes a far amount of code in c/b/ui already.

On Tue, Jan 19, 2016 at 4:18 PM,  <shuchen@chromium.org> wrote:
> On 2016/01/19 17:20:41, sky wrote:
>>
>> Is there a reason you want this in chrome/browser/input_method vs
>> chrome/browser/ui/input_method? I favor the later as this code seems all
>> ui
>> related. Or am I missing something?
>
>
> I'm not sure.
>
> The input method stuff for chromeos is under
> chrome/browser/chromeos/input_method/...
> instead of chrome/browser/chromeos/ui/input_method/...
> I guess it is because that those code is for extension APIs & input method
> management & utils & views.
>
> For non-chromeos, the input method stuff is only for extension APIs (e.g.
> passing events, invoking TextInputClient).
> The code under chrome/browser/ui/input_method/... seems more related to UIs
> including windows/views.
> Is my understand correct?
>
>
> https://codereview.chromium.org/1528483002/

-- 
You received this message because you are subscribed to the Google Groups
"Chromium-reviews" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

[Azure Wei, 2016-01-20 07:07:12]

The CQ bit was checked by azurewei@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-01-20 07:07:42]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1528483002/560001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1528483002/560001

[commit-bot: I haz the power, 2016-01-20 07:10:02]

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

[commit-bot: I haz the power, 2016-01-20 07:10:03]

Dry run: Try jobs failed on following builders:
  ios_dbg_simulator_ninja on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/ios_dbg_simulator...)
  ios_rel_device_ninja on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/ios_rel_device_ni...)
  mac_chromium_gn_rel on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_gn_r...)

[Azure Wei, 2016-01-20 07:22:06]

The CQ bit was checked by azurewei@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-01-20 07:22:47]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1528483002/580001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1528483002/580001

[commit-bot: I haz the power, 2016-01-20 08:12:21]

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

[commit-bot: I haz the power, 2016-01-20 08:12:22]

Dry run: Try jobs failed on following builders:
  win_chromium_compile_dbg_ng on tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_comp...)
  win_chromium_x64_rel_ng on tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_x64_...)

[Azure Wei, 2016-01-20 08:32:12]

The CQ bit was checked by azurewei@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-01-20 08:33:35]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1528483002/600001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1528483002/600001

[commit-bot: I haz the power, 2016-01-20 11:35:08]

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

[commit-bot: I haz the power, 2016-01-20 11:35:09]

Dry run: This issue passed the CQ dry run.

[Azure Wei, 2016-01-20 11:52:25]

On 2016/01/20 00:49:05, sky wrote:
> The code under c/b/ui is not all platform specific. For example,
> c/b/ui/tabs contains ui code that is used by all platforms.
> c/b/extensions consumes a far amount of code in c/b/ui already.
> 
> On Tue, Jan 19, 2016 at 4:18 PM,  <mailto:shuchen@chromium.org> wrote:
> > On 2016/01/19 17:20:41, sky wrote:
> >>
> >> Is there a reason you want this in chrome/browser/input_method vs
> >> chrome/browser/ui/input_method? I favor the later as this code seems all
> >> ui
> >> related. Or am I missing something?
> >
> >
> > I'm not sure.
> >
> > The input method stuff for chromeos is under
> > chrome/browser/chromeos/input_method/...
> > instead of chrome/browser/chromeos/ui/input_method/...
> > I guess it is because that those code is for extension APIs & input method
> > management & utils & views.
> >
> > For non-chromeos, the input method stuff is only for extension APIs (e.g.
> > passing events, invoking TextInputClient).
> > The code under chrome/browser/ui/input_method/... seems more related to UIs
> > including windows/views.
> > Is my understand correct?
> >
> >
> > https://codereview.chromium.org/1528483002/
> 
> -- 
> You received this message because you are subscribed to the Google Groups
> "Chromium-reviews" group.
> To unsubscribe from this group and stop receiving emails from it, send an
email
> to mailto:chromium-reviews+unsubscribe@chromium.org.
> 

sky@, I've moved the InputMethodEngine/InputMethodEngineBase files from
chrome/browser/input_method/ to chrome/browser/ui/input_method/, could you take
another look? Thank you!

[sky, 2016-01-20 16:45:27]

Thanks LGTM

[Azure Wei, 2016-01-21 01:33:57]

The CQ bit was checked by azurewei@chromium.org

[Azure Wei, 2016-01-21 01:33:58]

The patchset sent to the CQ was uploaded after l-g-t-m from
isherman@chromium.org, rdevlin.cronin@chromium.org, shuchen@chromium.org,
rockot@chromium.org
Link to the patchset: https://codereview.chromium.org/1528483002/#ps600001
(title: "Fix windows build error.")

[commit-bot: I haz the power, 2016-01-21 01:37:07]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1528483002/600001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1528483002/600001

[commit-bot: I haz the power, 2016-01-21 01:46:03]

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

[commit-bot: I haz the power, 2016-01-21 01:46:04]

Try jobs failed on following builders:
  chromium_presubmit on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)

[Azure Wei, 2016-01-21 02:37:13]

The CQ bit was checked by azurewei@chromium.org

[Azure Wei, 2016-01-21 02:37:14]

The patchset sent to the CQ was uploaded after l-g-t-m from rockot@chromium.org,
rdevlin.cronin@chromium.org, isherman@chromium.org, shuchen@chromium.org,
sky@chromium.org
Link to the patchset: https://codereview.chromium.org/1528483002/#ps620001
(title: "Sync")

[commit-bot: I haz the power, 2016-01-21 02:38:35]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1528483002/620001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1528483002/620001

[commit-bot: I haz the power, 2016-01-21 02:51:24]

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

[commit-bot: I haz the power, 2016-01-21 02:51:26]

Try jobs failed on following builders:
  chromeos_amd64-generic_chromium_compile_only_ng on tryserver.chromium.linux
(JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromeos_amd64-...)

[Azure Wei, 2016-01-21 03:00:04]

The CQ bit was checked by azurewei@chromium.org

[Azure Wei, 2016-01-21 03:00:04]

The patchset sent to the CQ was uploaded after l-g-t-m from rockot@chromium.org,
rdevlin.cronin@chromium.org, isherman@chromium.org, shuchen@chromium.org,
sky@chromium.org
Link to the patchset: https://codereview.chromium.org/1528483002/#ps640001
(title: " ")

[commit-bot: I haz the power, 2016-01-21 03:00:30]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1528483002/640001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1528483002/640001

[commit-bot: I haz the power, 2016-01-21 04:57:00]

Description was changed from

==========
The IME extension uses chrome.input.ime.activate() to activate itself so that it
can receive the events (e.g. the key events) and call input.ime related
functions (e.g. commitText).
And IME extension uses chrome.input.ime.deactivate() to deactivate itself.
These two functions is Non-ChromeOS APIs.

This CL does the following work:
1. Add class input_method::InputMethodEngine which is derived from
input_method::InputMethodBase for Linux & Windows functions.
2. Add extensions::InputImeEventRouter related functions. 
3. Implement extensions::InputImeAPI related functions.
4. Add class extensions::InputImeActivateFunction and
extensions::InputImeDeactivateFunction for implementation of
chrome.input.ime.activate/deactivate API.

Some design points that is different from ChromeOS API:
1. There's only one or zero activated InputMethodEngine all the time, so
functions CheckProfile() and IsActive() will always return true.
2. “input_components” section is ignored in the extension’s manifest, so the
component_id/engineID will always be empty string.

BUG=517773
TEST=No compiling errors.
==========

to

==========
The IME extension uses chrome.input.ime.activate() to activate itself so that it
can receive the events (e.g. the key events) and call input.ime related
functions (e.g. commitText).
And IME extension uses chrome.input.ime.deactivate() to deactivate itself.
These two functions is Non-ChromeOS APIs.

This CL does the following work:
1. Add class input_method::InputMethodEngine which is derived from
input_method::InputMethodBase for Linux & Windows functions.
2. Add extensions::InputImeEventRouter related functions. 
3. Implement extensions::InputImeAPI related functions.
4. Add class extensions::InputImeActivateFunction and
extensions::InputImeDeactivateFunction for implementation of
chrome.input.ime.activate/deactivate API.

Some design points that is different from ChromeOS API:
1. There's only one or zero activated InputMethodEngine all the time, so
functions CheckProfile() and IsActive() will always return true.
2. “input_components” section is ignored in the extension’s manifest, so the
component_id/engineID will always be empty string.

BUG=517773
TEST=No compiling errors.
==========

[commit-bot: I haz the power, 2016-01-21 04:57:02]

Committed patchset #33 (id:640001)

[commit-bot: I haz the power, 2016-01-21 04:58:07]

Description was changed from

==========
The IME extension uses chrome.input.ime.activate() to activate itself so that it
can receive the events (e.g. the key events) and call input.ime related
functions (e.g. commitText).
And IME extension uses chrome.input.ime.deactivate() to deactivate itself.
These two functions is Non-ChromeOS APIs.

This CL does the following work:
1. Add class input_method::InputMethodEngine which is derived from
input_method::InputMethodBase for Linux & Windows functions.
2. Add extensions::InputImeEventRouter related functions. 
3. Implement extensions::InputImeAPI related functions.
4. Add class extensions::InputImeActivateFunction and
extensions::InputImeDeactivateFunction for implementation of
chrome.input.ime.activate/deactivate API.

Some design points that is different from ChromeOS API:
1. There's only one or zero activated InputMethodEngine all the time, so
functions CheckProfile() and IsActive() will always return true.
2. “input_components” section is ignored in the extension’s manifest, so the
component_id/engineID will always be empty string.

BUG=517773
TEST=No compiling errors.
==========

to

==========
The IME extension uses chrome.input.ime.activate() to activate itself so that it
can receive the events (e.g. the key events) and call input.ime related
functions (e.g. commitText).
And IME extension uses chrome.input.ime.deactivate() to deactivate itself.
These two functions is Non-ChromeOS APIs.

This CL does the following work:
1. Add class input_method::InputMethodEngine which is derived from
input_method::InputMethodBase for Linux & Windows functions.
2. Add extensions::InputImeEventRouter related functions.
3. Implement extensions::InputImeAPI related functions.
4. Add class extensions::InputImeActivateFunction and
extensions::InputImeDeactivateFunction for implementation of
chrome.input.ime.activate/deactivate API.

Some design points that is different from ChromeOS API:
1. There's only one or zero activated InputMethodEngine all the time, so
functions CheckProfile() and IsActive() will always return true.
2. “input_components” section is ignored in the extension’s manifest, so the
component_id/engineID will always be empty string.

BUG=517773
TEST=No compiling errors.

Committed: https://crrev.com/b8d5846de644695112e79e06d499c143d31a4748
Cr-Commit-Position: refs/heads/master@{#370626}
==========

[commit-bot: I haz the power, 2016-01-21 04:58:09]

Patchset 33 (id:??) landed as
https://crrev.com/b8d5846de644695112e79e06d499c143d31a4748
Cr-Commit-Position: refs/heads/master@{#370626}