Fix preedit handling on IME switching.

Fix preedit handling on IME switching.

This is hack for Korean IME which required to commit preedit when it is disabled or switched to other IME.
The root of the problem is that, in ChromeOS, if the next engine is xkb layout, SetGlobalEngine is not called.
Due to this logic, ibus-daemon can't know the ime switching.
As the result, IBusPreeditFocusMode does not work.

BUG=chrome-os-partner:12798
TEST=Manually done on lumpy

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=154802

[Yusuke Sato, 2012-09-04 16:35:36]

Please write a test for this.

[Seigo Nonaka, 2012-09-04 17:14:40]

On 2012/09/04 16:35:36, Yusuke Sato wrote:
> Please write a test for this.

Sure, done.

[Yusuke Sato, 2012-09-04 17:28:35]

http://codereview.chromium.org/10917061/diff/1003/chrome/browser/chromeos/inp...
File chrome/browser/chromeos/input_method/input_method_manager_impl.cc (right):

http://codereview.chromium.org/10917061/diff/1003/chrome/browser/chromeos/inp...
chrome/browser/chromeos/input_method/input_method_manager_impl.cc:252: //
https://code.google.com/p/chrome-os-partner/issues/detail?id=12798
http://crosbug.com/p/12798

http://codereview.chromium.org/10917061/diff/1003/chrome/browser/chromeos/inp...
chrome/browser/chromeos/input_method/input_method_manager_impl.cc:253: // We
should notify ime switching to ibus-daemon, otherwise
IME

http://codereview.chromium.org/10917061/diff/1003/chrome/browser/chromeos/inp...
chrome/browser/chromeos/input_method/input_method_manager_impl.cc:255: // itself
if the next engine is xkb layout.
XKB

http://codereview.chromium.org/10917061/diff/1003/chrome/browser/chromeos/inp...
chrome/browser/chromeos/input_method/input_method_manager_impl.cc:257: if
(current_input_method_id.empty() ||
Q. when can |current_input_method_id| be empty?

[Seigo Nonaka, 2012-09-04 17:47:34]

http://codereview.chromium.org/10917061/diff/1003/chrome/browser/chromeos/inp...
File chrome/browser/chromeos/input_method/input_method_manager_impl.cc (right):

http://codereview.chromium.org/10917061/diff/1003/chrome/browser/chromeos/inp...
chrome/browser/chromeos/input_method/input_method_manager_impl.cc:252: //
https://code.google.com/p/chrome-os-partner/issues/detail?id=12798
On 2012/09/04 17:28:35, Yusuke Sato wrote:
> http://crosbug.com/p/12798

Done.

http://codereview.chromium.org/10917061/diff/1003/chrome/browser/chromeos/inp...
chrome/browser/chromeos/input_method/input_method_manager_impl.cc:253: // We
should notify ime switching to ibus-daemon, otherwise
On 2012/09/04 17:28:35, Yusuke Sato wrote:
> IME

Done.

http://codereview.chromium.org/10917061/diff/1003/chrome/browser/chromeos/inp...
chrome/browser/chromeos/input_method/input_method_manager_impl.cc:255: // itself
if the next engine is xkb layout.
On 2012/09/04 17:28:35, Yusuke Sato wrote:
> XKB

Done.

http://codereview.chromium.org/10917061/diff/1003/chrome/browser/chromeos/inp...
chrome/browser/chromeos/input_method/input_method_manager_impl.cc:257: if
(current_input_method_id.empty() ||
On 2012/09/04 17:28:35, Yusuke Sato wrote:
> Q. when can |current_input_method_id| be empty?

|current_input_method_id| is empty just after logging in.
And also, some unittest performs with empty.

[Yusuke Sato, 2012-09-04 17:49:53]

lgtm with two nits.

http://codereview.chromium.org/10917061/diff/3002/chrome/browser/chromeos/inp...
File chrome/browser/chromeos/input_method/input_method_manager_impl.cc (right):

http://codereview.chromium.org/10917061/diff/3002/chrome/browser/chromeos/inp...
chrome/browser/chromeos/input_method/input_method_manager_impl.cc:251: // Hack
for fixing http://crbug.com/p/12798
crosbug

please also fix the BUG= line.

[Yusuke Sato, 2012-09-04 18:05:43]

BUG line is still wrong. it should be BUG=chrome-os-partner:12798 I believe.

On 2012/09/04 17:49:53, Yusuke Sato wrote:
> lgtm with two nits.
> 
>
http://codereview.chromium.org/10917061/diff/3002/chrome/browser/chromeos/inp...
> File chrome/browser/chromeos/input_method/input_method_manager_impl.cc
(right):
> 
>
http://codereview.chromium.org/10917061/diff/3002/chrome/browser/chromeos/inp...
> chrome/browser/chromeos/input_method/input_method_manager_impl.cc:251: // Hack
> for fixing http://crbug.com/p/12798
> crosbug
> 
> please also fix the BUG= line.

[Seigo Nonaka, 2012-09-04 18:08:48]

http://codereview.chromium.org/10917061/diff/3002/chrome/browser/chromeos/inp...
File chrome/browser/chromeos/input_method/input_method_manager_impl.cc (right):

http://codereview.chromium.org/10917061/diff/3002/chrome/browser/chromeos/inp...
chrome/browser/chromeos/input_method/input_method_manager_impl.cc:251: // Hack
for fixing http://crbug.com/p/12798
Opps, thanks.
Done.
On 2012/09/04 17:49:53, Yusuke Sato wrote:
> crosbug
> 
> please also fix the BUG= line.

[Yusuke Sato, 2012-09-04 18:10:35]

lgtm, thanks for patching this.

On 2012/09/04 18:08:48, Seigo Nonaka wrote:
>
http://codereview.chromium.org/10917061/diff/3002/chrome/browser/chromeos/inp...
> File chrome/browser/chromeos/input_method/input_method_manager_impl.cc
(right):
> 
>
http://codereview.chromium.org/10917061/diff/3002/chrome/browser/chromeos/inp...
> chrome/browser/chromeos/input_method/input_method_manager_impl.cc:251: // Hack
> for fixing http://crbug.com/p/12798
> Opps, thanks.
> Done.
> On 2012/09/04 17:49:53, Yusuke Sato wrote:
> > crosbug
> > 
> > please also fix the BUG= line.

[commit-bot: I haz the power, 2012-09-04 18:11:23]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/nona@chromium.org/10917061/4

[commit-bot: I haz the power, 2012-09-04 20:33:25]

Change committed as 154802