Test cache creation retry via public interface only.

Test cache creation retry via public interface only.

This is a cleanup after:
https://codereview.chromium.org/12794003/

One test for backend force retry is enough, the other test
(CreateBackend_MissingFile) is converted to be backend implementation specific.

The CreateCacheViaPublicInterface() is called only once, but to eliminate it
there is a need to make the cache_thread_ protected. Also, I am sure the name is
not ideal, please suggest another.

The New Eviction allows not to be very strict about version comparison as stored
in the files on disk, this allows creating the BackendImpl with no flags and
then being able to pass the common integrity check.

BUG=173384
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=192558

[rvargas (doing something else), 2013-04-03 17:44:22]

https://codereview.chromium.org/13517004/diff/1/net/disk_cache/backend_unitte...
File net/disk_cache/backend_unittest.cc (right):

https://codereview.chromium.org/13517004/diff/1/net/disk_cache/backend_unitte...
net/disk_cache/backend_unittest.cc:1554: CreateCacheViaPublicInterface();
I think the test would be more clear if it is explicit about what it's doing...
create the cache thread and use it to create the cache.

As a bonus, we don't need the helper anymore.

[pasko-google - do not use, 2013-04-03 17:48:46]

On 2013/04/03 17:44:22, rvargas wrote:
>
https://codereview.chromium.org/13517004/diff/1/net/disk_cache/backend_unitte...
> File net/disk_cache/backend_unittest.cc (right):
> 
>
https://codereview.chromium.org/13517004/diff/1/net/disk_cache/backend_unitte...
> net/disk_cache/backend_unittest.cc:1554: CreateCacheViaPublicInterface();
> I think the test would be more clear if it is explicit about what it's
doing...
> create the cache thread and use it to create the cache.
> 
> As a bonus, we don't need the helper anymore.

OK, I agree, making the cache_thread_ protected ..

[pasko-google - do not use, 2013-04-03 17:50:15]

On 2013/04/03 17:48:46, pasko wrote:
> On 2013/04/03 17:44:22, rvargas wrote:
> >
>
https://codereview.chromium.org/13517004/diff/1/net/disk_cache/backend_unitte...
> > File net/disk_cache/backend_unittest.cc (right):
> > 
> >
>
https://codereview.chromium.org/13517004/diff/1/net/disk_cache/backend_unitte...
> > net/disk_cache/backend_unittest.cc:1554: CreateCacheViaPublicInterface();
> > I think the test would be more clear if it is explicit about what it's
> doing...
> > create the cache thread and use it to create the cache.
> > 
> > As a bonus, we don't need the helper anymore.
> 
> OK, I agree, making the cache_thread_ protected ..

oh no, sorry, creating another cache thread ..

[pasko-google - do not use, 2013-04-03 18:12:30]

https://codereview.chromium.org/13517004/diff/1/net/disk_cache/backend_unitte...
File net/disk_cache/backend_unittest.cc (right):

https://codereview.chromium.org/13517004/diff/1/net/disk_cache/backend_unitte...
net/disk_cache/backend_unittest.cc:1554: CreateCacheViaPublicInterface();
On 2013/04/03 17:44:22, rvargas wrote:
> I think the test would be more clear if it is explicit about what it's
doing...
> create the cache thread and use it to create the cache.
> 
> As a bonus, we don't need the helper anymore.

Done.

[rvargas (doing something else), 2013-04-03 18:25:33]

lgtm

[commit-bot: I haz the power, 2013-04-04 11:24:01]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/pasko@google.com/13517004/18001

[commit-bot: I haz the power, 2013-04-04 13:28:05]

Retried try job too often on win_rel for step(s) net_unittests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win_rel&nu...

[pasko-google - do not use, 2013-04-05 11:15:03]

In Patch Set 3 I am moving the cache integrity check to a point after the
backend is destroyed to avoid flakiness and failures on windows. Gavin, please
take a look.

[gavinp, 2013-04-05 11:19:13]

lgtm.

[commit-bot: I haz the power, 2013-04-05 11:23:09]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/pasko@google.com/13517004/18004

[commit-bot: I haz the power, 2013-04-05 13:27:10]

Change committed as 192558