ResourceFetcherTest: introduce FetchTestingPlatformSupport

ResourceFetcherTest: introduce FetchTestingPlatformSupport

To move ResourceFetcherTest under platform, we need to use
TestingPlatformSupportWithMockScheduler and to make
WebURLLoaderMockFactory available with the MockScheduler.

FetchTestingPlatformSupport is the TestingPlatformSupport
inheritance that adds WebURLLoaderMockFactory and FetchContext
support. At this point, only ResourceFetcherTest use the
FetchTestingplatformSupport, but eventually, other tests
may want to use it.

Also this patch contains modifications to remove cached resources
and all mock URL entries so that we can cleanup each test safely.

This is needed to pass LinkPreloadResourceMultipleFetchersAnd{Use|Move}
without any retry.

BUG=655920
TEST=webkit_unit_tests --gtest_filter='ResourceFetcherTest.*'

Review-Url: https://codereview.chromium.org/2644083003
Cr-Commit-Position: refs/heads/master@{#445639}
Committed: https://chromium.googlesource.com/chromium/src/+/284fe71823b4516ac5630addd2f44788347ddd07

[Takashi Toyoshima, 2017-01-19 13:55:50]

Description was changed from

==========
ResourceFetcherTest: Use TestingPlatformSupport

To move ResourceFetcherTest under platform, use
TestingPlatformSupportWithMockScheduler and make WebURLLoaderMockFactory
available with MockScheduler and MockFetchContext beforehand.

Also this patch contains modifications to remove cached resources so that
LinkPreloadResourceMultipleFetchersAnd{Use|Move} tests work always.

BUG=655920
TEST=webkit_unit_tests --gtest_filter='ResourceFetcherTest.*'
==========

to

==========
ResourceFetcherTest: Use TestingPlatformSupport

To move ResourceFetcherTest under platform, use
TestingPlatformSupportWithMockScheduler and make WebURLLoaderMockFactory
available with MockScheduler and MockFetchContext beforehand.

Also this patch contains modifications to remove cached resources so that
LinkPreloadResourceMultipleFetchersAnd{Use|Move} tests work always.

BUG=655920
TEST=webkit_unit_tests --gtest_filter='ResourceFetcherTest.*'
==========

[Takashi Toyoshima, 2017-01-19 13:55:50]

toyoshim@chromium.org changed reviewers:
+ yhirano@chromium.org

[Takashi Toyoshima, 2017-01-19 13:56:06]

toyoshim@chromium.org changed reviewers:
+ kinuko@chromium.org

[Takashi Toyoshima, 2017-01-19 14:00:08]

Yep, finally, I made the last patch to run all fetch unit tests under platform.

This still requires minor non-mechanical changes to move under platform, but I
can not have it under core/ due to the build rule restriction.

Once this patch was submitted, I'd rebase the large platform/loader CL. That
should contain the minimum set of non-mechanical changes.

[kinuko, 2017-01-19 18:40:18]

https://codereview.chromium.org/2644083003/diff/1/third_party/WebKit/Source/c...
File third_party/WebKit/Source/core/fetch/MockFetchContext.h (right):

https://codereview.chromium.org/2644083003/diff/1/third_party/WebKit/Source/c...
third_party/WebKit/Source/core/fetch/MockFetchContext.h:59:
RefPtr<WebTaskRunner> loadingTaskRunner() const override { return m_runner; }
Could you help me understand... so the task runner returned by this fetch
context and the task runner that is returned by
Platform::current()->currentThread()->scheduler()->loadingTaskRunner() are
different, is that right?  Wouldn't that become a problem at some point?

[Takashi Toyoshima, 2017-01-20 04:13:16]

https://codereview.chromium.org/2644083003/diff/1/third_party/WebKit/Source/c...
File third_party/WebKit/Source/core/fetch/MockFetchContext.h (right):

https://codereview.chromium.org/2644083003/diff/1/third_party/WebKit/Source/c...
third_party/WebKit/Source/core/fetch/MockFetchContext.h:59:
RefPtr<WebTaskRunner> loadingTaskRunner() const override { return m_runner; }
The right approach to obtain the task runner in fetch is to access it via
FetchContext since it's per-frame runner. But unfortunately as you noticed, we
are using currentThread()->scheduler()->loadingTaskRunner() in some modules,
e.g. MemoryCache, Resource, and RawResource. Probably, we could change to use
per-frame task runner for *Resource, but I have no idea about the MemoryCache.
Probably, we will continue to use per-frame runner.

Once we move this test to platform, currentThread()->scheduler() is mocked by
TestingPlatformSupportWithMockScheduler. That does not have a MessageLoop, and
need to call another runUntilIdle method that the testing Platform instance has.
So anyway, current WebURLLoaderMockFactory's RunLoop does not work.

If all MockFetchContext users are moved to platform, we could change this to
return a task runner mocked by TestingPlatform..., but since this is used by
core/loader tests, using FakeWebTaskRunner here is a point of compromise.

[Takashi Toyoshima, 2017-01-20 04:14:51]

> Probably, we will continue to use per-frame runner.

Sorry, this means, *per-thread* runner in the MemoryCache.

[kinuko, 2017-01-20 05:00:33]

https://codereview.chromium.org/2644083003/diff/1/third_party/WebKit/Source/c...
File third_party/WebKit/Source/core/fetch/MockFetchContext.h (right):

https://codereview.chromium.org/2644083003/diff/1/third_party/WebKit/Source/c...
third_party/WebKit/Source/core/fetch/MockFetchContext.h:59:
RefPtr<WebTaskRunner> loadingTaskRunner() const override { return m_runner; }
On 2017/01/20 04:13:16, toyoshim wrote:
> The right approach to obtain the task runner in fetch is to access it via
> FetchContext since it's per-frame runner. But unfortunately as you noticed, we
> are using currentThread()->scheduler()->loadingTaskRunner() in some modules,
> e.g. MemoryCache, Resource, and RawResource. Probably, we could change to use
> per-frame task runner for *Resource, but I have no idea about the MemoryCache.
> Probably, we will continue to use per-frame runner.

When you say MemoryCache do you mean the task observers for pruning?  And I see
it comes from the browser... I feel we might be able to deprecate the Web Cache
manager and this prune code path as we now have memory coordinator, but that's
probably another story.

> Once we move this test to platform, currentThread()->scheduler() is mocked by
> TestingPlatformSupportWithMockScheduler. That does not have a MessageLoop, and
> need to call another runUntilIdle method that the testing Platform instance
has.
> So anyway, current WebURLLoaderMockFactory's RunLoop does not work.

Yep I noticed it.
 
> If all MockFetchContext users are moved to platform, we could change this to
> return a task runner mocked by TestingPlatform..., but since this is used by
> core/loader tests, using FakeWebTaskRunner here is a point of compromise.

Do you mean you're assuming we can't use TestingPlatform if the test's not in
platform?

[Takashi Toyoshima, 2017-01-20 05:45:47]

Ah, good point. TestingPlatformSupport is designed to be used in both core and
platform. So we can make other MockFetchContext users to use it.
Probably, I'd create FetchTestingPlatformSupport that inherits
TestingPlatformSupportWithMockScheduler and provides MockFetchContext, and use
the platform from related tests. Thus, we can use a same mocked task runner for
both.

[kinuko, 2017-01-20 05:51:52]

On 2017/01/20 05:45:47, toyoshim wrote:
> Ah, good point. TestingPlatformSupport is designed to be used in both core and
> platform. So we can make other MockFetchContext users to use it.
> Probably, I'd create FetchTestingPlatformSupport that inherits
> TestingPlatformSupportWithMockScheduler and provides MockFetchContext, and use
> the platform from related tests. Thus, we can use a same mocked task runner
for
> both.

Yeah I think that's probably nicer and we can have more coherent testing
environment.

[Takashi Toyoshima, 2017-01-20 07:08:35]

Description was changed from

==========
ResourceFetcherTest: Use TestingPlatformSupport

To move ResourceFetcherTest under platform, use
TestingPlatformSupportWithMockScheduler and make WebURLLoaderMockFactory
available with MockScheduler and MockFetchContext beforehand.

Also this patch contains modifications to remove cached resources so that
LinkPreloadResourceMultipleFetchersAnd{Use|Move} tests work always.

BUG=655920
TEST=webkit_unit_tests --gtest_filter='ResourceFetcherTest.*'
==========

to

==========
ResourceFetcherTest: introduce FetchTestingPlatformSupport

To move ResourceFetcherTest under platform, we need to use
TestingPlatformSupportWithMockScheduler and to make
WebURLLoaderMockFactory available with the MockScheduler.

FetchTestingPlatformSupport is the TestingPlatformSupport
inheritance that adds WebURLLoaderMockFactory and FetchContext
support. At this point, ResourceFetcherTest use the platform,
but eventually, other tests may want to use it.

Also this patch contains modifications to remove cached resources so that
LinkPreloadResourceMultipleFetchersAnd{Use|Move} tests work always.

BUG=655920
TEST=webkit_unit_tests --gtest_filter='ResourceFetcherTest.*'
==========

[Takashi Toyoshima, 2017-01-20 07:11:44]

PTAL

https://codereview.chromium.org/2644083003/diff/20001/third_party/WebKit/Sour...
File third_party/WebKit/Source/core/fetch/MockFetchContext.h (right):

https://codereview.chromium.org/2644083003/diff/20001/third_party/WebKit/Sour...
third_party/WebKit/Source/core/fetch/MockFetchContext.h:32: // TODO(toyoshim):
Disallow to pass nullptr for |taskRunner|, and force to use
Let me leave this TODO now because it's a little stressful to check my changes
work for both core and platform.

I'd adopt another CL to use FetchTestingPlatformSupport for other
MockFetchContext users, and remove old interfaces that use the
FakeWebTaskRunner.

[Takashi Toyoshima, 2017-01-20 09:17:08]

The CQ bit was checked by toyoshim@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2017-01-20 09:17:16]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-01-20 10:29:50]

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

[commit-bot: I haz the power, 2017-01-20 10:29:51]

Dry run: Try jobs failed on following builders:
  android_arm64_dbg_recipe on master.tryserver.chromium.android (JOB_FAILED,
https://build.chromium.org/p/tryserver.chromium.android/builders/android_arm6...)

[Takashi Toyoshima, 2017-01-20 10:40:21]

The CQ bit was checked by toyoshim@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2017-01-20 10:40:33]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[kinuko, 2017-01-20 12:30:34]

Thanks for keeping updating this one, I have a few more comments...! (I've
wanted to cleanup TestingPlatformSupport & loading tests for sometime)

https://codereview.chromium.org/2644083003/diff/20001/third_party/WebKit/Sour...
File third_party/WebKit/Source/core/fetch/FetchTestingPlatformSupport.cpp
(right):

https://codereview.chromium.org/2644083003/diff/20001/third_party/WebKit/Sour...
third_party/WebKit/Source/core/fetch/FetchTestingPlatformSupport.cpp:24: }
I still feel this is a bit unfortunate. I think it'd be nice if
WebURLLoaderMockFactory can just use the right runUntilIdle depending on what
the underlying platform supports.

What do you think if we make WebURLLoaderMockFactory optionally take
TestingPlatformSupport and call its runUntilIdle?  Like -->
https://codereview.chromium.org/2647903002 (diff from this patch)

[commit-bot: I haz the power, 2017-01-20 14:07:45]

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

[commit-bot: I haz the power, 2017-01-20 14:07:45]

Dry run: This issue passed the CQ dry run.

[Takashi Toyoshima, 2017-01-23 06:39:16]

> What do you think if we make WebURLLoaderMockFactory optionally take
> TestingPlatformSupport and call its runUntilIdle?

Or we may want to pass unique_ptr<Function<void()>> for runUntilIdle that could
internally use weak pointer to be safer?
But I feel current approach is enough for testing. (well, we have not manage
life-cycles for Platform related things carefully though we have a scheme to
switch it for testing...)

Patch Set 5 is just importing kinuko-san's proposal, and fix some minor style
nits (I'm not sure why cl format can not catch them).

PTAL.

[yhirano, 2017-01-23 07:42:15]

If we can assume that ScopedTestingPlatformSupport is used, can we use
Platform::current instead of passing round the object?

https://codereview.chromium.org/2644083003/diff/20001/third_party/WebKit/Sour...
File third_party/WebKit/Source/core/fetch/ResourceFetcherTest.cpp (right):

https://codereview.chromium.org/2644083003/diff/20001/third_party/WebKit/Sour...
third_party/WebKit/Source/core/fetch/ResourceFetcherTest.cpp:557:
memoryCache()->remove(resource);
How about calling MemoryCache::evictResources in the destructor?

https://codereview.chromium.org/2644083003/diff/80001/third_party/WebKit/Sour...
File third_party/WebKit/Source/core/fetch/ResourceFetcherTest.cpp (right):

https://codereview.chromium.org/2644083003/diff/80001/third_party/WebKit/Sour...
third_party/WebKit/Source/core/fetch/ResourceFetcherTest.cpp:399:
ScopedMockRedirectRequester(MockFetchContext* context) : m_context(context) {}
explicit

https://codereview.chromium.org/2644083003/diff/80001/third_party/WebKit/Sour...
File
third_party/WebKit/Source/platform/testing/weburl_loader_mock_factory_impl.h
(right):

https://codereview.chromium.org/2644083003/diff/80001/third_party/WebKit/Sour...
third_party/WebKit/Source/platform/testing/weburl_loader_mock_factory_impl.h:35:
WebURLLoaderMockFactoryImpl(TestingPlatformSupport*);
+explicit

[yhirano, 2017-01-23 07:46:20]

On 2017/01/23 07:42:15, yhirano wrote:
> If we can assume that ScopedTestingPlatformSupport is used, can we use
> Platform::current instead of passing round the object?
> 
Ah, sorry, we would need a down cast then. Never mind.

[yhirano, 2017-01-23 07:50:17]

On 2017/01/23 07:46:20, yhirano wrote:
> On 2017/01/23 07:42:15, yhirano wrote:
> > If we can assume that ScopedTestingPlatformSupport is used, can we use
> > Platform::current instead of passing round the object?
> > 
> Ah, sorry, we would need a down cast then. Never mind.

... or we could have TestingPlatformSupport::current().

[kinuko, 2017-01-23 08:15:15]

On 2017/01/23 07:50:17, yhirano wrote:
> On 2017/01/23 07:46:20, yhirano wrote:
> > On 2017/01/23 07:42:15, yhirano wrote:
> > > If we can assume that ScopedTestingPlatformSupport is used, can we use
> > > Platform::current instead of passing round the object?
> > > 
> > Ah, sorry, we would need a down cast then. Never mind.
> 
> ... or we could have TestingPlatformSupport::current().

Hmm, yeah that could be an option too.  Might look cleaner?

[kinuko, 2017-01-23 08:20:11]

Aside that, the current code change itself lgtm.

[Takashi Toyoshima, 2017-01-23 09:18:13]

There are several options for better handling of runUntilIdle, but I have no
idea which is the best at this point. IMO, Platform could have virtual
runUntilIdle() directly because we already have some virtual methods for
testing, you know getURLLoaderMockFactory in the discussion is exactly such a
thing. Or we could move testing only interfaces into TestingPlatform abstract
class, and could have Platform::testing() that may return null for production.

Actually, we have some options to call runUntilIdle-ish methods in testing, e.g.
testing::runPendingTasks is another option for different situations, and some
tests have own method. If we could have such interface on Platform directly, we
might solve this confused situation eventually?

https://codereview.chromium.org/2644083003/diff/20001/third_party/WebKit/Sour...
File third_party/WebKit/Source/core/fetch/ResourceFetcherTest.cpp (right):

https://codereview.chromium.org/2644083003/diff/20001/third_party/WebKit/Sour...
third_party/WebKit/Source/core/fetch/ResourceFetcherTest.cpp:557:
memoryCache()->remove(resource);
Aha. Sounds nice.
Since now we have an actual implementation of ResourceFetcherTest, I'd move
ScopedMockRedirectRequester's cleanUp() to ResourceFetcherTest::TearDown() so
that it can clean up for all test cases safely.

https://codereview.chromium.org/2644083003/diff/80001/third_party/WebKit/Sour...
File third_party/WebKit/Source/core/fetch/ResourceFetcherTest.cpp (right):

https://codereview.chromium.org/2644083003/diff/80001/third_party/WebKit/Sour...
third_party/WebKit/Source/core/fetch/ResourceFetcherTest.cpp:399:
ScopedMockRedirectRequester(MockFetchContext* context) : m_context(context) {}
On 2017/01/23 07:42:14, yhirano wrote:
> explicit

Done.

[Takashi Toyoshima, 2017-01-23 09:24:07]

Description was changed from

==========
ResourceFetcherTest: introduce FetchTestingPlatformSupport

To move ResourceFetcherTest under platform, we need to use
TestingPlatformSupportWithMockScheduler and to make
WebURLLoaderMockFactory available with the MockScheduler.

FetchTestingPlatformSupport is the TestingPlatformSupport
inheritance that adds WebURLLoaderMockFactory and FetchContext
support. At this point, ResourceFetcherTest use the platform,
but eventually, other tests may want to use it.

Also this patch contains modifications to remove cached resources so that
LinkPreloadResourceMultipleFetchersAnd{Use|Move} tests work always.

BUG=655920
TEST=webkit_unit_tests --gtest_filter='ResourceFetcherTest.*'
==========

to

==========
ResourceFetcherTest: introduce FetchTestingPlatformSupport

To move ResourceFetcherTest under platform, we need to use
TestingPlatformSupportWithMockScheduler and to make
WebURLLoaderMockFactory available with the MockScheduler.

FetchTestingPlatformSupport is the TestingPlatformSupport
inheritance that adds WebURLLoaderMockFactory and FetchContext
support. At this point, only ResourceFetcherTest use the
FetchTestingplatformSupport, but eventually, other tests
may want to use it.

Also this patch contains modifications to remove cached resources
and all mock URL entries so that we can cleanup each test safely.

This is needed to pass LinkPreloadResourceMultipleFetchersAnd{Use|Move}
without any retry.

BUG=655920
TEST=webkit_unit_tests --gtest_filter='ResourceFetcherTest.*'
==========

[Takashi Toyoshima, 2017-01-23 09:24:49]

description was also modified to be up to date.
PTAL.

[Takashi Toyoshima, 2017-01-23 09:26:06]

The CQ bit was checked by toyoshim@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2017-01-23 09:26:23]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[kinuko, 2017-01-23 10:55:13]

On 2017/01/23 09:18:13, toyoshim wrote:
> There are several options for better handling of runUntilIdle, but I have no
> idea which is the best at this point. IMO, Platform could have virtual
> runUntilIdle() directly because we already have some virtual methods for
> testing, you know getURLLoaderMockFactory in the discussion is exactly such a
> thing. Or we could move testing only interfaces into TestingPlatform abstract
> class, and could have Platform::testing() that may return null for production.

Fyi, we used to have something like this before, which I removed away sometime
ago.
https://codereview.chromium.org/1845323002

Platform.h is a public layer that is also exposed outside blink, I'm not too
motivated to add a test-only method on it unless we need it outside blink.

[kinuko, 2017-01-23 10:57:25]

On 2017/01/23 10:55:13, kinuko wrote:
> On 2017/01/23 09:18:13, toyoshim wrote:
> > There are several options for better handling of runUntilIdle, but I have no
> > idea which is the best at this point. IMO, Platform could have virtual
> > runUntilIdle() directly because we already have some virtual methods for
> > testing, you know getURLLoaderMockFactory in the discussion is exactly such
a
> > thing. Or we could move testing only interfaces into TestingPlatform
abstract
> > class, and could have Platform::testing() that may return null for
production.
> 
> Fyi, we used to have something like this before, which I removed away sometime
> ago.
> https://codereview.chromium.org/1845323002
> 
> Platform.h is a public layer that is also exposed outside blink, I'm not too
> motivated to add a test-only method on it unless we need it outside blink.

I mean, I prefer converging these around TestingPlatformSupport as much as
possible, but if we have several cases that don't work well in that layer we
could weigh other options too.

[commit-bot: I haz the power, 2017-01-23 11:27:08]

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

[commit-bot: I haz the power, 2017-01-23 11:27:09]

Dry run: This issue passed the CQ dry run.

[yhirano, 2017-01-23 11:38:56]

https://codereview.chromium.org/2644083003/diff/100001/third_party/WebKit/Sou...
File
third_party/WebKit/Source/platform/testing/weburl_loader_mock_factory_impl.cc
(right):

https://codereview.chromium.org/2644083003/diff/100001/third_party/WebKit/Sou...
third_party/WebKit/Source/platform/testing/weburl_loader_mock_factory_impl.cc:24:
std::unique_ptr<WebURLLoaderMockFactory> WebURLLoaderMockFactory::create()
How about adding a (non-null) TestingPlatformSupport parameter to this function?
IIUC the only call site is [1] and we can provide a TestingPlatformSupport
object.

With that we can delete the if statement in RunUntilIdle.

1:
https://cs.chromium.org/chromium/src/content/test/test_blink_web_unit_test_su...

[kinuko, 2017-01-24 00:39:05]

https://codereview.chromium.org/2644083003/diff/100001/third_party/WebKit/Sou...
File
third_party/WebKit/Source/platform/testing/weburl_loader_mock_factory_impl.cc
(right):

https://codereview.chromium.org/2644083003/diff/100001/third_party/WebKit/Sou...
third_party/WebKit/Source/platform/testing/weburl_loader_mock_factory_impl.cc:24:
std::unique_ptr<WebURLLoaderMockFactory> WebURLLoaderMockFactory::create()
On 2017/01/23 11:38:56, yhirano wrote:
> How about adding a (non-null) TestingPlatformSupport parameter to this
function?
> IIUC the only call site is [1] and we can provide a TestingPlatformSupport
> object.
> 
> With that we can delete the if statement in RunUntilIdle.
> 
> 1:
>
https://cs.chromium.org/chromium/src/content/test/test_blink_web_unit_test_su...

[1] is in content, how can we provide TestingPlatformSupport?

[yhirano, 2017-01-24 03:33:28]

On 2017/01/24 00:39:05, kinuko wrote:
>
https://codereview.chromium.org/2644083003/diff/100001/third_party/WebKit/Sou...
> File
> third_party/WebKit/Source/platform/testing/weburl_loader_mock_factory_impl.cc
> (right):
> 
>
https://codereview.chromium.org/2644083003/diff/100001/third_party/WebKit/Sou...
>
third_party/WebKit/Source/platform/testing/weburl_loader_mock_factory_impl.cc:24:
> std::unique_ptr<WebURLLoaderMockFactory> WebURLLoaderMockFactory::create()
> On 2017/01/23 11:38:56, yhirano wrote:
> > How about adding a (non-null) TestingPlatformSupport parameter to this
> function?
> > IIUC the only call site is [1] and we can provide a TestingPlatformSupport
> > object.
> > 
> > With that we can delete the if statement in RunUntilIdle.
> > 
> > 1:
> >
>
https://cs.chromium.org/chromium/src/content/test/test_blink_web_unit_test_su...
> 
> [1] is in content, how can we provide TestingPlatformSupport?

Sorry I was confused by similar names.

I'm uncomfortable with the virtual runUntilIdle + the if statement in
RunUntilIdle. We need one of them, not both, right? If we need the latter, Can
we store a TestingPlatformSupportWithMockScheduler* and make runUntilIdle
non-virtual?

[Takashi Toyoshima, 2017-01-24 03:43:33]

> Platform.h is a public layer that is also exposed outside blink, I'm not too
> motivated to add a test-only method on it unless we need it outside blink.

I see. It sounds reasonable. It looks a similar policy with content public API.
So, TestingPlatformSupport could be a _impl like thing in content.

As kinuko-san said, TestingPlatformSupport is Blink platform internal thing and
not exposed to public interface.

By the way, it looks Platform::getURLLoaderMockFactory() is called only inside
Blink.
So, probably we can remove this from the public Platform interface, and move it
to TestingPlatformSupport?

I didn't check details about the test_blink_web_unit_test_support, but it looks
only used in content_unittest? If so, we can just remove the caller of
test_blink_web_unit_test_support.
I can try removing it from the public interface, but I want to make it a
separate CL.

[yhirano, 2017-01-24 04:14:34]

lgtm

[Takashi Toyoshima, 2017-01-24 04:16:38]

The CQ bit was checked by toyoshim@chromium.org

[Takashi Toyoshima, 2017-01-24 04:16:39]

The patchset sent to the CQ was uploaded after l-g-t-m from kinuko@chromium.org
Link to the patchset: https://codereview.chromium.org/2644083003/#ps100001
(title: "review #23")

[commit-bot: I haz the power, 2017-01-24 04:16:56]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-01-24 04:20:54]

CQ is committing da patch.

Bot data: {"patchset_id": 100001, "attempt_start_ts": 1485231398888810,
"parent_rev": "44fadca5234816d4defc7380bf7ae8c2376494f6", "commit_rev":
"284fe71823b4516ac5630addd2f44788347ddd07"}

[commit-bot: I haz the power, 2017-01-24 04:21:50]

Description was changed from

==========
ResourceFetcherTest: introduce FetchTestingPlatformSupport

To move ResourceFetcherTest under platform, we need to use
TestingPlatformSupportWithMockScheduler and to make
WebURLLoaderMockFactory available with the MockScheduler.

FetchTestingPlatformSupport is the TestingPlatformSupport
inheritance that adds WebURLLoaderMockFactory and FetchContext
support. At this point, only ResourceFetcherTest use the
FetchTestingplatformSupport, but eventually, other tests
may want to use it.

Also this patch contains modifications to remove cached resources
and all mock URL entries so that we can cleanup each test safely.

This is needed to pass LinkPreloadResourceMultipleFetchersAnd{Use|Move}
without any retry.

BUG=655920
TEST=webkit_unit_tests --gtest_filter='ResourceFetcherTest.*'
==========

to

==========
ResourceFetcherTest: introduce FetchTestingPlatformSupport

To move ResourceFetcherTest under platform, we need to use
TestingPlatformSupportWithMockScheduler and to make
WebURLLoaderMockFactory available with the MockScheduler.

FetchTestingPlatformSupport is the TestingPlatformSupport
inheritance that adds WebURLLoaderMockFactory and FetchContext
support. At this point, only ResourceFetcherTest use the
FetchTestingplatformSupport, but eventually, other tests
may want to use it.

Also this patch contains modifications to remove cached resources
and all mock URL entries so that we can cleanup each test safely.

This is needed to pass LinkPreloadResourceMultipleFetchersAnd{Use|Move}
without any retry.

BUG=655920
TEST=webkit_unit_tests --gtest_filter='ResourceFetcherTest.*'

Review-Url: https://codereview.chromium.org/2644083003
Cr-Commit-Position: refs/heads/master@{#445639}
Committed:
https://chromium.googlesource.com/chromium/src/+/284fe71823b4516ac5630addd2f4...
==========

[commit-bot: I haz the power, 2017-01-24 04:21:53]

Committed patchset #6 (id:100001) as
https://chromium.googlesource.com/chromium/src/+/284fe71823b4516ac5630addd2f4...