[Mac] Fix handling of download shelf auto close.

[Mac] Fix handling of download shelf auto close.

If the mouse is not on the download shelf when an auto-close is
scheduled, the shelf would not close automatically. Fix this by
closing immediately if the mouse isn't on the shelf.

Also adds unit-tests and update the shelf to use CrTrackingArea
instead of NSTrackingArea.

BUG=none
TEST=(1) Start a download that opens automatically.
     (2) Move the mouse off the shelf and wait for completion.
     (3) The shelf should close automatically.

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=195080

[asanka, 2013-03-29 15:58:22]

Depends on and is a follow up to https://codereview.chromium.org/12995025/

I don't know if the testing strategy is idiomatic. I chose to introduce a
protocol since creating a DownloadItemController seemed too expensive for a test
that didn't rely on its functionality.

[Nico, 2013-03-30 04:33:41]

On 2013/03/29 15:58:22, asanka wrote:
> Depends on and is a follow up to https://codereview.chromium.org/12995025/
> 
> I don't know if the testing strategy is idiomatic. I chose to introduce a
> protocol since creating a DownloadItemController seemed too expensive for a
test
> that didn't rely on its functionality.

Re testing: Since Objective-C is dynamic, it's possible to create mocks at
runtime if a dependency on the real object is too heavy (is it?): See
http://ocmock.org/ for documentation on the mocking framework we use in chrome.
You can use cs.chromium.org (query "#include.*ocmock") to find examples.

[asanka, 2013-04-04 00:54:00]

On 2013/03/30 04:33:41, Nico wrote:
> On 2013/03/29 15:58:22, asanka wrote:
> > Depends on and is a follow up to https://codereview.chromium.org/12995025/
> > 
> > I don't know if the testing strategy is idiomatic. I chose to introduce a
> > protocol since creating a DownloadItemController seemed too expensive for a
> test
> > that didn't rely on its functionality.
> 
> Re testing: Since Objective-C is dynamic, it's possible to create mocks at
> runtime if a dependency on the real object is too heavy (is it?): See
> http://ocmock.org/ for documentation on the mocking framework we use in
chrome.
> You can use http://cs.chromium.org (query "#include.*ocmock") to find
examples.

PTAL? Using OCMock now instead of introducing an extra protocol.

The concrete DownloadItemController itself isn't too slow for the test. But
rather it requires scaffolding for icon loading and verbose expectations for the
MockDownloadItem it would use.

[asanka, 2013-04-16 15:27:25]

On 2013/04/04 00:54:00, asanka wrote:
> On 2013/03/30 04:33:41, Nico wrote:
> > On 2013/03/29 15:58:22, asanka wrote:
> > > Depends on and is a follow up to https://codereview.chromium.org/12995025/
> > > 
> > > I don't know if the testing strategy is idiomatic. I chose to introduce a
> > > protocol since creating a DownloadItemController seemed too expensive for
a
> > test
> > > that didn't rely on its functionality.
> > 
> > Re testing: Since Objective-C is dynamic, it's possible to create mocks at
> > runtime if a dependency on the real object is too heavy (is it?): See
> > http://ocmock.org/ for documentation on the mocking framework we use in
> chrome.
> > You can use http://cs.chromium.org (query "#include.*ocmock") to find
> examples.
> 
> PTAL? Using OCMock now instead of introducing an extra protocol.
> 
> The concrete DownloadItemController itself isn't too slow for the test. But
> rather it requires scaffolding for icon loading and verbose expectations for
the
> MockDownloadItem it would use.

Ping?

[Nico, 2013-04-16 21:58:44]

lgtm

Sorry about the delay.

[asanka, 2013-04-16 22:15:40]

On 2013/04/16 21:58:44, Nico wrote:
> lgtm
> 
> Sorry about the delay.

Thanks!

[commit-bot: I haz the power, 2013-04-18 17:31:45]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/asanka@chromium.org/13318002/8002

[commit-bot: I haz the power, 2013-04-18 18:50:01]

Sorry for I got bad news for ya.
Compile failed with a clobber build on win7_aura.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win7_aura&...
Your code is likely broken or HEAD is junk. Please ensure your
code is not broken then alert the build sheriffs.
Look at the try server FAQ for more details.

[commit-bot: I haz the power, 2013-04-18 22:23:50]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/asanka@chromium.org/13318002/8002

[commit-bot: I haz the power, 2013-04-19 03:49:55]

Change committed as 195080