Add unittest for updating mozc candidates.

Add unittest for updating mozc candidates.

BUG=chromium-os:23640, 106862, 106850
TEST=run unit_tests


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=113601

Reverted: http://src.chromium.org/viewvc/chrome?view=rev&revision=113624

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=114603

[Seigo Nonaka, 2011-12-07 12:00:42]

Dear reviewers,

I found some try bot failed but I think it is not problematic, because this
change mostly affects compile and unit_tests phase. And they are
passed(linux_chromeos_clang is broken now?).

Please review this CL.

Thank you.

Seigo

[Seigo Nonaka, 2011-12-07 12:00:42]

Dear reviewers,

I found some try bot failed but I think it is not problematic, because this
change mostly affects compile and unit_tests phase. And they are
passed(linux_chromeos_clang is broken now?).

Please review this CL.

Thank you.

Seigo

[Hiro Komatsu, 2011-12-08 02:14:56]

http://codereview.chromium.org/8833008/diff/1/chrome/browser/chromeos/input_m...
File chrome/browser/chromeos/input_method/candidate_window_view_unittest.cc
(right):

http://codereview.chromium.org/8833008/diff/1/chrome/browser/chromeos/input_m...
chrome/browser/chromeos/input_method/candidate_window_view_unittest.cc:67:
MessageLoopForUI message_loop_;
I guess this is not explicitly used, but needed to be created for
initialization.  If so, please add a comment so.

[Seigo Nonaka, 2011-12-08 02:31:21]

http://codereview.chromium.org/8833008/diff/1/chrome/browser/chromeos/input_m...
File chrome/browser/chromeos/input_method/candidate_window_view_unittest.cc
(right):

http://codereview.chromium.org/8833008/diff/1/chrome/browser/chromeos/input_m...
chrome/browser/chromeos/input_method/candidate_window_view_unittest.cc:67:
MessageLoopForUI message_loop_;
On 2011/12/08 02:14:57, komatsu wrote:
> I guess this is not explicitly used, but needed to be created for
> initialization.  If so, please add a comment so.

Done.

[Hiro Komatsu, 2011-12-08 04:50:59]

lgtm

LGTM.

Please make sure it does not break the build.

[Seigo Nonaka, 2011-12-08 05:09:32]

Thank you for your review!

Of course I will pay attention for build break.

Thank you.
On 2011/12/08 04:50:59, komatsu wrote:
> lgtm
> 
> LGTM.
> 
> Please make sure it does not break the build.

[Seigo Nonaka, 2011-12-08 05:10:16]

+yusukes

Could you review my CL?

Thank you

Seigo.

[horo, 2011-12-08 08:03:35]

On 2011/12/08 08:03:14, horo wrote:

lgtm

[Yusuke Sato, 2011-12-08 09:00:18]

lgtm, thanks for adding this.

[commit-bot: I haz the power, 2011-12-08 12:42:49]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/nona@chromium.org/8833008/4001

[commit-bot: I haz the power, 2011-12-08 14:27:44]

Change committed as 113601

[Seigo Nonaka, 2011-12-08 19:48:57]

Sorry reviewers, I have no idea to fix memory leak now.
Please wait until I will announce.

Thank you.

[Seigo Nonaka, 2011-12-14 03:01:15]

Dear reviewers,

I fixed memory leakage and also fixed aura test failure.
So I think it's time to re-start reviewing.

Please review again?

Thank you.
Nona

[Yusuke Sato, 2011-12-14 04:41:09]

Which part is the fix for aura?

On 2011/12/14 03:01:15, nona1 wrote:
> Dear reviewers,
> 
> I fixed memory leakage and also fixed aura test failure.
> So I think it's time to re-start reviewing.
> 
> Please review again?
> 
> Thank you.
> Nona

[Seigo Nonaka, 2011-12-14 05:12:07]

Sorry I forget to explain,

The reason of build break or memory leak is for my mis-usage of widget (mainly
forget to call "CloseNow")

If we do not call "CloseNow",...

In the case of GTK, there are no crush but memory leakage.
In the case of Aura, crash on destructor.

I confirmed both case are fixed.

On 2011/12/14 04:41:09, Yusuke Sato wrote:
> Which part is the fix for aura?
> 
> On 2011/12/14 03:01:15, nona1 wrote:
> > Dear reviewers,
> > 
> > I fixed memory leakage and also fixed aura test failure.
> > So I think it's time to re-start reviewing.
> > 
> > Please review again?
> > 
> > Thank you.
> > Nona

[Yusuke Sato, 2011-12-14 05:18:02]

LGTM with one comment.

linux_chromeos_valgrind trybot is still red. Please confirm that the errors
reported by the bot are not yours.

On 2011/12/14 05:12:07, nona1 wrote:
> Sorry I forget to explain,
> 
> The reason of build break or memory leak is for my mis-usage of widget (mainly
> forget to call "CloseNow")
> 
> If we do not call "CloseNow",...
> 
> In the case of GTK, there are no crush but memory leakage.
> In the case of Aura, crash on destructor.
> 
> I confirmed both case are fixed.
> 
> On 2011/12/14 04:41:09, Yusuke Sato wrote:
> > Which part is the fix for aura?
> > 
> > On 2011/12/14 03:01:15, nona1 wrote:
> > > Dear reviewers,
> > > 
> > > I fixed memory leakage and also fixed aura test failure.
> > > So I think it's time to re-start reviewing.
> > > 
> > > Please review again?
> > > 
> > > Thank you.
> > > Nona

[horo, 2011-12-14 05:36:47]

lgtm

[Seigo Nonaka, 2011-12-14 06:23:35]

Sorry for reviewer,

Thanks to yusuke-san, I found another memory leack both gtk and aura.
Such error (might) be shown on linux_chromeos_valgrind log text. But my local
valgrind report memory leak. So I investigate again.

Thank you.
Seigo

On 2011/12/14 05:36:47, horo wrote:
> lgtm

[Seigo Nonaka, 2011-12-14 12:42:35]

Dear reviewers,

After investigation, there are no memory leak.
(My local environment is wrong...)
So, it is safe to remove suppression entry from suppressions.txt.

Thank you

On 2011/12/14 06:23:35, nona1 wrote:
> Sorry for reviewer,
> 
> Thanks to yusuke-san, I found another memory leack both gtk and aura.
> Such error (might) be shown on linux_chromeos_valgrind log text. But my local
> valgrind report memory leak. So I investigate again.
> 
> Thank you.
> Seigo
> 
> On 2011/12/14 05:36:47, horo wrote:
> > lgtm

[commit-bot: I haz the power, 2011-12-15 04:56:25]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/nona@chromium.org/8833008/25001

[commit-bot: I haz the power, 2011-12-15 06:11:33]

Change committed as 114603