make content_shell and content_browsertests works with aura

make content_shell and content_browsertests works with aura

BUG=138525
TEST=content_browsertests


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=148742

[Hongbo Min, 2012-07-26 13:40:54]

The patch set 1 is from http://codereview.chromium.org/10802064 issue.

[Hongbo Min, 2012-07-26 14:34:51]

The latest patch has two more refinements:
1. Remove the packed_resources target dependency by packing ui_string locale and
app_locale_setting resource into content_shell.pak. Now it is on the right way,
I think.

2. Remove MockDBusThreadManager from content_browser_test. The dbus thread
manger instance is already initialized in Shell::PlatformInitialize.

Pls review it again. Thanks very much.

[jam, 2012-07-26 14:58:15]

Thanks!

just to confirm: did you run content_browsertests on linux_chromeos builds? When
it ran on the buildbots for a few runs, it had an assert in dbusmanager code
because it wasn't initialized. I think the current line in PlatformInitialize
might not get hit in content_browsertests

http://codereview.chromium.org/10824043/diff/3001/content/content_shell.gypi
File content/content_shell.gypi (right):

http://codereview.chromium.org/10824043/diff/3001/content/content_shell.gypi#...
content/content_shell.gypi:252:
'<(SHARED_INTERMEDIATE_DIR)/ui/app_locale_settings/app_locale_settings_en-US.pak',
nit: order

[Hongbo Min, 2012-07-27 00:40:55]

I will try it later.

On 2012/07/26 14:58:15, John Abd-El-Malek wrote:
> Thanks!
> 
> just to confirm: did you run content_browsertests on linux_chromeos builds?
When
> it ran on the buildbots for a few runs, it had an assert in dbusmanager code
> because it wasn't initialized. I think the current line in PlatformInitialize
> might not get hit in content_browsertests
> 
> http://codereview.chromium.org/10824043/diff/3001/content/content_shell.gypi
> File content/content_shell.gypi (right):
> 
>
http://codereview.chromium.org/10824043/diff/3001/content/content_shell.gypi#...
> content/content_shell.gypi:252:
>
'<(SHARED_INTERMEDIATE_DIR)/ui/app_locale_settings/app_locale_settings_en-US.pak',
> nit: order

[Hongbo Min, 2012-07-27 01:53:34]

I double-check the issue you concerned,  the PlatformInitialize did get hit in
content_browsertests, the call stacking is as following:

#0  chromeos::DBusThreadManager::Initialize () at
chromeos/dbus/dbus_thread_manager.cc:397
#1  0x0810cd24 in content::Shell::PlatformInitialize () at
content/shell/shell_aura.cc:335
#2  0x081156c0 in content::ShellBrowserMainParts::PreMainMessageLoopRun
(this=0x86050e0) at content/shell/shell_browser_main_parts.cc:70
#3  0xb632ad09 in content::BrowserMainLoop::CreateThreads (this=0x8602900) at
content/browser/browser_main_loop.cc:438
#4  0xb632ce54 in (anonymous namespace)::BrowserMainRunnerImpl::Initialize
(this=0x8605060, parameters=...) at content/browser/browser_main_runner.cc:86
#5  0xb63294ed in BrowserMain (parameters=...) at
content/browser/browser_main.cc:17
#6  0x0811d78d in BrowserTestBase::SetUp (this=0x8605e60) at
content/test/browser_test_base.cc:42
#7  0x081069fe in content::ContentBrowserTest::SetUp (this=0x8605e60) at
content/test/content_browser_test.cc:72

So the global dbus thread manager is actually initialized by Shell.


On 2012/07/27 00:40:55, Hongbo wrote:
> I will try it later.
> 
> On 2012/07/26 14:58:15, John Abd-El-Malek wrote:
> > Thanks!
> > 
> > just to confirm: did you run content_browsertests on linux_chromeos builds?
> When
> > it ran on the buildbots for a few runs, it had an assert in dbusmanager code
> > because it wasn't initialized. I think the current line in
PlatformInitialize
> > might not get hit in content_browsertests
> > 
> > http://codereview.chromium.org/10824043/diff/3001/content/content_shell.gypi
> > File content/content_shell.gypi (right):
> > 
> >
>
http://codereview.chromium.org/10824043/diff/3001/content/content_shell.gypi#...
> > content/content_shell.gypi:252:
> >
>
'<(SHARED_INTERMEDIATE_DIR)/ui/app_locale_settings/app_locale_settings_en-US.pak',
> > nit: order

[jam, 2012-07-27 02:01:08]

On 2012/07/27 01:53:34, Hongbo wrote:
> I double-check the issue you concerned,  the PlatformInitialize did get hit in
> content_browsertests, the call stacking is as following:

Great, thank you.

lgtm

[commit-bot: I haz the power, 2012-07-27 07:05:58]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/hongbo.min@intel.com/10824043/10001

[commit-bot: I haz the power, 2012-07-27 07:55:46]

Try job failure for 10824043-10001 (retry) on linux_rel for step "check_deps".
It's a second try, previously, step "check_deps" failed.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=linux_rel&...

[commit-bot: I haz the power, 2012-07-27 08:23:31]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/hongbo.min@intel.com/10824043/13002

[commit-bot: I haz the power, 2012-07-27 10:55:59]

Change committed as 148742