Move ExtensionsProcessManager to src/extensions, part 2

Move ExtensionsProcessManager to src/extensions, part 2

* Move observation of NOTIFICATION_BROWSER_WINDOW_READY to Chrome
* Remove remaining references to Profile
* Tidy up ChromeTestSuiteInitializer

BUG=313481
TEST=unit_tests and browser_tests ExtensionProcessManager*
R=sky@chromium.org, yoz@chromium.org

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=233011

[James Cook, 2013-11-01 20:53:12]

yoz, PTAL

https://codereview.chromium.org/56253002/diff/80001/chrome/browser/extensions...
File chrome/browser/extensions/chrome_notification_observer.cc (right):

https://codereview.chromium.org/56253002/diff/80001/chrome/browser/extensions...
chrome/browser/extensions/chrome_notification_observer.cc:42: // non-incognito
window opened.
This code replicates the behavior of the NOTIFICATION_BROWSER_WINDOW_READY
handler in the non-incognito ExtensionProcessManager. I think it will only have
an effect if somehow we get this notification for an incognito window before we
get it for the main window. While that should not happen in practice I wanted to
make sure I was maintaining the behavior of the existing code.

https://codereview.chromium.org/56253002/diff/80001/chrome/test/base/chrome_t...
File chrome/test/base/chrome_test_suite.cc (right):

https://codereview.chromium.org/56253002/diff/80001/chrome/test/base/chrome_t...
chrome/test/base/chrome_test_suite.cc:102: TearDownContentClients();
This teardown has no significant side effects, so I moved it before the
g_browser_process cleanup to mirror the initialization order.

https://codereview.chromium.org/56253002/diff/80001/chrome/test/base/chrome_t...
chrome/test/base/chrome_test_suite.cc:139:
extensions::ExtensionsBrowserClient::Set(extensions_browser_client_.get());
This must be done after content is initialized because it now registers for
notifications.

[Yoyo Zhou, 2013-11-01 22:54:06]

LGTM

https://codereview.chromium.org/56253002/diff/80001/chrome/browser/extensions...
File chrome/browser/extensions/chrome_extensions_browser_client.h (right):

https://codereview.chromium.org/56253002/diff/80001/chrome/browser/extensions...
chrome/browser/extensions/chrome_extensions_browser_client.h:53:
scoped_ptr<ChromeNotificationObserver> notification_observer_;
Can this be a regular member variable?

https://codereview.chromium.org/56253002/diff/80001/extensions/browser/extens...
File extensions/browser/extensions_browser_client.cc (left):

https://codereview.chromium.org/56253002/diff/80001/extensions/browser/extens...
extensions/browser/extensions_browser_client.cc:22: // This can happen in tests.
I assume this was just copied over from ExtensionsClient?

In ExtensionsClient there are some adverse effects when this gets set twice.

[James Cook, 2013-11-04 19:06:18]

https://codereview.chromium.org/56253002/diff/80001/chrome/browser/extensions...
File chrome/browser/extensions/chrome_extensions_browser_client.h (right):

https://codereview.chromium.org/56253002/diff/80001/chrome/browser/extensions...
chrome/browser/extensions/chrome_extensions_browser_client.h:53:
scoped_ptr<ChromeNotificationObserver> notification_observer_;
On 2013/11/01 22:54:06, Yoyo Zhou wrote:
> Can this be a regular member variable?

Done.

https://codereview.chromium.org/56253002/diff/80001/extensions/browser/extens...
File extensions/browser/extensions_browser_client.cc (left):

https://codereview.chromium.org/56253002/diff/80001/extensions/browser/extens...
extensions/browser/extensions_browser_client.cc:22: // This can happen in tests.
On 2013/11/01 22:54:06, Yoyo Zhou wrote:
> I assume this was just copied over from ExtensionsClient?
> 
> In ExtensionsClient there are some adverse effects when this gets set twice.

This should be OK here - the case in ExtensionsClient has side effects each time
it is called.  Things are simpler here - I've got complete control of the
lifetime of this object and just need to set it up and tear it down for each
test invocation.

[James Cook, 2013-11-04 19:09:31]

sky, can I get OWNERS approval for chrome/browser/browser_process_impl.* and
chrome/test/base/chrome_test_suite.cc ?

https://codereview.chromium.org/56253002/diff/160001/chrome/browser/browser_p...
File chrome/browser/browser_process_impl.h (right):

https://codereview.chromium.org/56253002/diff/160001/chrome/browser/browser_p...
chrome/browser/browser_process_impl.h:192:
scoped_ptr<extensions::ExtensionsBrowserClient> extensions_browser_client_;
I converted the ExtensionsBrowserClient from a singleton to an owned object.

https://codereview.chromium.org/56253002/diff/160001/chrome/test/base/chrome_...
File chrome/test/base/chrome_test_suite.cc (right):

https://codereview.chromium.org/56253002/diff/160001/chrome/test/base/chrome_...
chrome/test/base/chrome_test_suite.cc:102: TearDownContentClients();
This teardown has no significant side effects, so I moved it before the
g_browser_process cleanup to mirror the initialization order.

[sky, 2013-11-04 21:15:21]

LGTM

[commit-bot: I haz the power, 2013-11-05 06:23:59]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jamescook@chromium.org/56253002/240001

[commit-bot: I haz the power, 2013-11-05 09:08:37]

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

[James Cook, 2013-11-05 17:09:13]

Committed patchset #5 manually as r233011 (presubmit successful).