Replace MessagePumpForUI with MessagePumpDefault in unit tests.

Replace MessagePumpForUI with MessagePumpDefault in unit tests.

Uses the MessagePumpFactory functionality to install a default message pump in
unit tests.  This allows us to stop hard-coding a default message pump in many
separate tests and instead centralizes the logic.  When tests ask for a
MessageLoopForUI now, they will actually receive a message loop that is backed
by a default message pump, which means that calling Run() and Quit() will work
without crashing.

BUG=None
TEST=None

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=148384

[rohitrao (ping after 24h), 2012-07-13 14:55:38]

One more CL for you, Pawel.  This uses the same MessagePumpFactory that android
does, but the implementations are so different that it doesn't seem worth trying
to share any code.

[rohitrao (ping after 24h), 2012-07-25 14:46:57]

Over to Mark instead.

Mark, do you feel ok with reviewing changes in base/test?  Pawel is the only
owner of this directory, and it looks like he's been out for a while.

[Mark Mentovai, 2012-07-25 15:22:00]

http://codereview.chromium.org/10701172/diff/3001/base/test/test_support_ios.h
File base/test/test_support_ios.h (left):

http://codereview.chromium.org/10701172/diff/3001/base/test/test_support_ios....
base/test/test_support_ios.h:5: #include "base/message_loop_proxy.h"
Don’t use svn copy to create a new file. The revision history for this file will
be confusing.

http://codereview.chromium.org/10701172/diff/3001/base/test/test_support_ios.mm
File base/test/test_support_ios.mm (right):

http://codereview.chromium.org/10701172/diff/3001/base/test/test_support_ios....
base/test/test_support_ios.mm:5: #include "base/logging.h"
Unused.

http://codereview.chromium.org/10701172/diff/3001/base/test/test_support_ios....
base/test/test_support_ios.mm:14: };
Semicolon?

[rohitrao (ping after 24h), 2012-07-25 16:36:53]

Gonna try and fix the A+ nonsense.  Will report back.

http://codereview.chromium.org/10701172/diff/3001/base/test/test_support_ios.h
File base/test/test_support_ios.h (left):

http://codereview.chromium.org/10701172/diff/3001/base/test/test_support_ios....
base/test/test_support_ios.h:5: #include "base/message_loop_proxy.h"
On 2012/07/25 15:22:00, Mark Mentovai wrote:
> Don’t use svn copy to create a new file. The revision history for this file
will
> be confusing.

Git is working some sort of magic and auto-detecting copies.  I'll see if I can
find a way to stop it, otherwise I might need to switch to an svn checkout.

http://codereview.chromium.org/10701172/diff/3001/base/test/test_support_ios.mm
File base/test/test_support_ios.mm (right):

http://codereview.chromium.org/10701172/diff/3001/base/test/test_support_ios....
base/test/test_support_ios.mm:5: #include "base/logging.h"
On 2012/07/25 15:22:00, Mark Mentovai wrote:
> Unused.

Done.

http://codereview.chromium.org/10701172/diff/3001/base/test/test_support_ios....
base/test/test_support_ios.mm:14: };
On 2012/07/25 15:22:00, Mark Mentovai wrote:
> Semicolon?

Liberal use of semicolons prevents...I dunno.

[Mark Mentovai, 2012-07-25 16:42:04]

LGTM when you make it not look like a copy.

[rohitrao (ping after 24h), 2012-07-25 16:47:32]

On 2012/07/25 16:42:04, Mark Mentovai wrote:
> LGTM when you make it not look like a copy.
I added this to my .gitconfig and reuploaded.  Seems to have worked.

[diff]
  renamelimit=1

[commit-bot: I haz the power, 2012-07-25 16:48:01]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/rohitrao@chromium.org/10701172/12002

[commit-bot: I haz the power, 2012-07-25 18:23:36]

Try job failure for 10701172-12002 (retry) on mac_rel for step "browser_tests".
It's a second try, previously, step "browser_tests" failed.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=mac_rel&nu...

[commit-bot: I haz the power, 2012-07-25 18:25:16]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/rohitrao@chromium.org/10701172/12002

[commit-bot: I haz the power, 2012-07-25 19:30:00]

Try job failure for 10701172-12002 (retry) on mac_rel for step "browser_tests".
It's a second try, previously, step "browser_tests" failed.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=mac_rel&nu...

[Paweł Hajdan Jr., 2012-08-29 14:51:02]

LGTM

By the way, any base/ OWNER can review base/test changes (e.g. when I'm on
vacation).