Allow ApplicationTestBase clients to avoid a default MessageLoop.

Allow ApplicationTestBase clients to avoid a default MessageLoop.

Provides a means for overriding the creation of a default message
loop.  Such a loop is unnecessary for many tests and causes issues
with the bevy of existing chrome tests which expect to be able to
create their own message loop.

BUG=none
TEST=https://codereview.chromium.org/806443002/
R=msw@chromium.org, sky@chromium.org

Committed: https://chromium.googlesource.com/external/mojo/+/dcc2f22a40eb6070ba79b322f8db3a2a7cc1bb22

[DaleCurtis, 2014-12-16 01:09:06]

dalecurtis@chromium.org changed reviewers:
+ msw@chromium.org, sky@chromium.org

[sky, 2014-12-16 03:47:36]

LGTM with a name change

https://codereview.chromium.org/792463007/diff/1/mojo/public/cpp/application/...
File mojo/public/cpp/application/application_test_base.h (right):

https://codereview.chromium.org/792463007/diff/1/mojo/public/cpp/application/...
mojo/public/cpp/application/application_test_base.h:51: virtual bool
IsDefaultLoopEnabled();
This name is a bit confusing. How about ShouldCreateDefaultRunLoop().

[msw, 2014-12-16 15:40:32]

lgtm with sky's proposed name change.

https://codereview.chromium.org/792463007/diff/1/mojo/public/cpp/application/...
File mojo/public/cpp/application/lib/application_test_base.cc (right):

https://codereview.chromium.org/792463007/diff/1/mojo/public/cpp/application/...
mojo/public/cpp/application/lib/application_test_base.cc:62: if
(IsDefaultLoopEnabled())
Hmm, I think all the current apptests would fail without instantiating the
default run loop for application impl init, but I guess
PipelineIntegrationTestBase already creates a base::MessageLoop for your new
PipelineIntegrationTestHost; I guess this is reasonable.

[DaleCurtis, 2014-12-16 19:26:12]

https://codereview.chromium.org/792463007/diff/1/mojo/public/cpp/application/...
File mojo/public/cpp/application/application_test_base.h (right):

https://codereview.chromium.org/792463007/diff/1/mojo/public/cpp/application/...
mojo/public/cpp/application/application_test_base.h:51: virtual bool
IsDefaultLoopEnabled();
On 2014/12/16 03:47:36, sky wrote:
> This name is a bit confusing. How about ShouldCreateDefaultRunLoop().

Done.

[DaleCurtis, 2014-12-16 19:26:18]

The CQ bit was checked by dalecurtis@chromium.org

[commit-bot: I haz the power, 2014-12-16 19:27:56]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-12-16 19:27:57]

Commit queue rejected this change because it did not recognize the base URL.
Please commit your change manually.

[DaleCurtis, 2014-12-16 19:30:26]

Hmm, I have no idea how to land this.  CQ fails and "git cl land" yields
permission denied to my github account:

https://github.com/dalecurtis

What to do?

[msw, 2014-12-16 20:02:05]

msw@chromium.org changed reviewers:
+ jamesr@chromium.org

[msw, 2014-12-16 20:02:05]

On 2014/12/16 19:30:26, DaleCurtis wrote:
> Hmm, I have no idea how to land this.  CQ fails and "git cl land" yields
> permission denied to my github account:
> 
> https://github.com/dalecurtis
> 
> What to do?

Mojo isn't hooked up to the CQ, but git cl land should work.
https://github.com/domokit/mojo/wiki
Maybe James can help by inviting you to the domokit org?

[msw, 2014-12-16 20:03:25]

Maybe click Join here:
https://github.com/orgs/domokit/teams/contributors

[DaleCurtis, 2014-12-16 20:39:33]

On 2014/12/16 20:03:25, msw wrote:
> Maybe click Join here:
> https://github.com/orgs/domokit/teams/contributors

I can't see a Join button, it just gives a 404 for that URL. Backing up a the
path a bit, I get a "This organization has no public members. You must be a
member to see who's a part of this organization."

[jamesr, 2014-12-16 20:43:34]

I can invite you to the organization.  One moment...

[jamesr, 2014-12-16 20:44:16]

OK, I've invited dalecurtis to the domokit contributors team.  Accept that
invitation and then run 'git cl land'.

[DaleCurtis, 2014-12-16 20:52:44]

Committed patchset #2 (id:20001) manually as
dcc2f22a40eb6070ba79b322f8db3a2a7cc1bb22 (presubmit successful).

[DaleCurtis, 2014-12-16 20:52:56]

Thanks!