Adds MessageLoopUIApplication for use on iOS.

Adds MessageLoopUIApplication for use on iOS.

BUG=None
TEST=None


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=146553

[Mark Mentovai, 2012-07-12 01:08:19]

http://codereview.chromium.org/10689161/diff/1/base/base.gyp
File base/base.gyp (right):

http://codereview.chromium.org/10689161/diff/1/base/base.gyp#newcode571
base/base.gyp:571: ['exclude', '^message_loop_unittest.cc$'],
Give me a \\ on each dot in here.

http://codereview.chromium.org/10689161/diff/1/base/message_pump_mac.mm
File base/message_pump_mac.mm (right):

http://codereview.chromium.org/10689161/diff/1/base/message_pump_mac.mm#newco...
base/message_pump_mac.mm:539: void MessagePumpUIApplication::Attach(Delegate*
delegate) {
DCHECK(!run_loop_); ? That’d require you to initialize the run_loop_ field to
NULL in the constructor, and if you did that, you should take the constructor
body out of line from the .h and put it here in this .mm. But checking run_loop_
for consistency might be a good idea to prevent multiple Attach calls. Unless
you have some reason to want to Attach multiple times. Oy.

http://codereview.chromium.org/10689161/diff/1/base/message_pump_mac.mm#newco...
base/message_pump_mac.mm:545:
MessagePumpUIApplication::~MessagePumpUIApplication() {}
Try to keep things ordered sort of consistently, put the destructor before the
member functions. And if you move the constructor to this file, put that before
Gozer the destructor.

[rohitrao (ping after 24h), 2012-07-12 14:20:03]

http://codereview.chromium.org/10689161/diff/1/base/base.gyp
File base/base.gyp (right):

http://codereview.chromium.org/10689161/diff/1/base/base.gyp#newcode81
base/base.gyp:81: ['exclude', '_nss\.cc$'],
Should this also be a double-backslash?

http://codereview.chromium.org/10689161/diff/1/base/base.gyp#newcode571
base/base.gyp:571: ['exclude', '^message_loop_unittest.cc$'],
On 2012/07/12 01:08:19, Mark Mentovai wrote:
> Give me a \\ on each dot in here.

Fixed everywhere in this file, with one question on a line above.

http://codereview.chromium.org/10689161/diff/1/base/message_pump_mac.mm
File base/message_pump_mac.mm (right):

http://codereview.chromium.org/10689161/diff/1/base/message_pump_mac.mm#newco...
base/message_pump_mac.mm:539: void MessagePumpUIApplication::Attach(Delegate*
delegate) {
On 2012/07/12 01:08:19, Mark Mentovai wrote:
> DCHECK(!run_loop_); ?

Done.  Attach() should never be called multiple times, so a DCHECK is good.

http://codereview.chromium.org/10689161/diff/1/base/message_pump_mac.mm#newco...
base/message_pump_mac.mm:545:
MessagePumpUIApplication::~MessagePumpUIApplication() {}
On 2012/07/12 01:08:19, Mark Mentovai wrote:
> Try to keep things ordered sort of consistently, put the destructor before the
> member functions. And if you move the constructor to this file, put that
before
> Gozer the destructor.

Ooh, I know what I'm watching tonight.

[Mark Mentovai, 2012-07-12 15:59:03]

LGTM otherwise

http://codereview.chromium.org/10689161/diff/1/base/base.gyp
File base/base.gyp (right):

http://codereview.chromium.org/10689161/diff/1/base/base.gyp#newcode81
base/base.gyp:81: ['exclude', '_nss\.cc$'],
rohitrao wrote:
> Should this also be a double-backslash?

Yes.

http://codereview.chromium.org/10689161/diff/4001/base/base.gyp
File base/base.gyp (right):

http://codereview.chromium.org/10689161/diff/4001/base/base.gyp#newcode562
base/base.gyp:562: ['exclude', '^metrics/stats_table_unittest\\.cc'],
Now put dollar signs at the ends of the ones in here that don’t already have
them.

http://codereview.chromium.org/10689161/diff/4001/base/base.gyp#newcode642
base/base.gyp:642: ['include', 'mac/objc_property_releaser_unittest\\.mm'],
Dollar signs on these two also. And these ones are also missing their carets.

[rohitrao (ping after 24h), 2012-07-12 16:30:56]

http://codereview.chromium.org/10689161/diff/4001/base/base.gyp
File base/base.gyp (right):

http://codereview.chromium.org/10689161/diff/4001/base/base.gyp#newcode562
base/base.gyp:562: ['exclude', '^metrics/stats_table_unittest\\.cc'],
On 2012/07/12 15:59:03, Mark Mentovai wrote:
> Now put dollar signs at the ends of the ones in here that don’t already have
> them.

Done (on the ones that are full paths).

http://codereview.chromium.org/10689161/diff/4001/base/base.gyp#newcode642
base/base.gyp:642: ['include', 'mac/objc_property_releaser_unittest\\.mm'],
On 2012/07/12 15:59:03, Mark Mentovai wrote:
> Dollar signs on these two also. And these ones are also missing their carets.

Done.

[commit-bot: I haz the power, 2012-07-12 16:31:12]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/rohitrao@chromium.org/10689161/2010

[commit-bot: I haz the power, 2012-07-12 17:39:44]

Try job failure for 10689161-2010 (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, 2012-07-12 18:01:09]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/rohitrao@chromium.org/10689161/2010

[commit-bot: I haz the power, 2012-07-12 19:10:37]

Try job failure for 10689161-2010 (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...

[rohitrao (ping after 24h), 2012-07-12 19:13:34]

Sigh.  I see this RLZ test failing on many other try jobs, so I don't think it's
related to my CL?  If I had broken the message pump, I would expect to see every
browser_test fail.

I'll try again later tonight and see what happens.

[commit-bot: I haz the power, 2012-07-13 09:45:15]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/rohitrao@chromium.org/10689161/2010

[commit-bot: I haz the power, 2012-07-13 11:02:58]

Change committed as 146553