Move LazyBackgroundTaskQueueTest into extensions_unittests

Move LazyBackgroundTaskQueueTest into extensions_unittests

Previously it ran as part of Chrome's unit_tests. This requires factoring out
the usage of TestingProfile and swapping in a MockExtensionSystem.

BUG=397164
TEST=extensions_unittests

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=287267

[James Cook, 2014-08-01 16:54:39]

yoz, PTAL

If more tests require these MockExtensionSystems I'll pull it out into a
separate class and set it up in all ExtensionTests.

[Yoyo Zhou, 2014-08-01 22:13:10]

LGTM

https://chromiumcodereview.appspot.com/428953003/diff/20001/extensions/browse...
File extensions/browser/lazy_background_task_queue_unittest.cc (right):

https://chromiumcodereview.appspot.com/428953003/diff/20001/extensions/browse...
extensions/browser/lazy_background_task_queue_unittest.cc:56: class
MockExtensionSystem : public ExtensionSystem {
Thanks for looking at this test - I had gotten a little bit into it earlier this
week but gave up after deciding not to do this effort (and remembering that
browser_context_ is not the only browser context, as you have noticed).

https://chromiumcodereview.appspot.com/428953003/diff/20001/extensions/browse...
extensions/browser/lazy_background_task_queue_unittest.cc:123: friend struct
DefaultSingletonTraits<MockExtensionSystemFactory>;
nit: it's more or less equivalent but I prefer LazyInstance for this.

https://chromiumcodereview.appspot.com/428953003/diff/20001/extensions/browse...
File extensions/browser/test_extensions_browser_client.h (right):

https://chromiumcodereview.appspot.com/428953003/diff/20001/extensions/browse...
extensions/browser/test_extensions_browser_client.h:25: void
set_extension_system_factory(ExtensionSystemProvider* factory) {
It seems slightly odd to set the factory rather than the system. Guess it's a
byproduct of ExtensionSystem::Get.

[James Cook, 2014-08-01 22:48:03]

https://codereview.chromium.org/428953003/diff/20001/extensions/browser/lazy_...
File extensions/browser/lazy_background_task_queue_unittest.cc (right):

https://codereview.chromium.org/428953003/diff/20001/extensions/browser/lazy_...
extensions/browser/lazy_background_task_queue_unittest.cc:123: friend struct
DefaultSingletonTraits<MockExtensionSystemFactory>;
On 2014/08/01 22:13:09, Yoyo Zhou wrote:
> nit: it's more or less equivalent but I prefer LazyInstance for this.

On closer inspection it doesn't need to be a singleton at all, so long as I
tweak ExtensionsTest to clean up the individual services before I destroy the
factory.

https://codereview.chromium.org/428953003/diff/20001/extensions/browser/test_...
File extensions/browser/test_extensions_browser_client.h (right):

https://codereview.chromium.org/428953003/diff/20001/extensions/browser/test_...
extensions/browser/test_extensions_browser_client.h:25: void
set_extension_system_factory(ExtensionSystemProvider* factory) {
On 2014/08/01 22:13:10, Yoyo Zhou wrote:
> It seems slightly odd to set the factory rather than the system. Guess it's a
> byproduct of ExtensionSystem::Get.

Yeah, this is awkward. I don't remember why ExtensionSystemProvider was built
this way. I'm guessing it's because an ExtensionSystemProvider subclass has to
express the ExtensionSystemFactory DependsOn() calls?

I wish ExtensionsBrowserClient just had a GetExtensionSystem() method. Should I
investigate that possibility for another CL?

[Yoyo Zhou, 2014-08-01 23:00:30]

https://codereview.chromium.org/428953003/diff/20001/extensions/browser/test_...
File extensions/browser/test_extensions_browser_client.h (right):

https://codereview.chromium.org/428953003/diff/20001/extensions/browser/test_...
extensions/browser/test_extensions_browser_client.h:25: void
set_extension_system_factory(ExtensionSystemProvider* factory) {
On 2014/08/01 22:48:03, James Cook wrote:
> On 2014/08/01 22:13:10, Yoyo Zhou wrote:
> > It seems slightly odd to set the factory rather than the system. Guess it's
a
> > byproduct of ExtensionSystem::Get.
> 
> Yeah, this is awkward. I don't remember why ExtensionSystemProvider was built
> this way. I'm guessing it's because an ExtensionSystemProvider subclass has to
> express the ExtensionSystemFactory DependsOn() calls?
> 
> I wish ExtensionsBrowserClient just had a GetExtensionSystem() method. Should
I
> investigate that possibility for another CL?

Yes, I think it was for the dependencies on other factories. If you want to look
into it, feel free to, but don't spend too much time on it, I suppose. =)

[James Cook, 2014-08-03 22:12:30]

The CQ bit was checked by jamescook@chromium.org

[commit-bot: I haz the power, 2014-08-03 22:13:22]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/jamescook@chromium.org/428953003/60001

[commit-bot: I haz the power, 2014-08-03 23:49:29]

Change committed as 287267

[tfarina, 2014-08-03 23:59:04]

https://codereview.chromium.org/428953003/diff/60001/extensions/extensions.gyp
File extensions/extensions.gyp (right):

https://codereview.chromium.org/428953003/diff/60001/extensions/extensions.gy...
extensions/extensions.gyp:832: 'browser/lazy_background_task_queue_unittest.cc',
should we say we depend on keyed_service_content? Looks like we inherit it from
extensions_browser though.