Fix NavigationHandleImplBrowserTest flakiness.

Fix NavigationHandleImplBrowserTest flakiness.

NavigationHandleImplBrowserTest.VerifyRequestContextTypeForFrameTree has been
flaky because it relied on pointer comparisons for checking that new throttles
were being properly created for each new navigation. This is an unsafe solution
and has already caused problems in the past as it's not uncommon that new
instances of the same type of object are created at precisely the same memory
address. This is especially the case on Windows, where this test has been flaky.

This change uses a throttle-install-counter instead that works just as well for
(indirectly) checking that new throttles were installed without the pointer
comparison drawbacks. It also re-enables that test on Windows.

Moreover it moves class member default initializations to be inline with their
declarations to simplify the code and cut back on line count.

BUG=696915
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation

Review-Url: https://codereview.chromium.org/2728783003
Cr-Commit-Position: refs/heads/master@{#454318}
Committed: https://chromium.googlesource.com/chromium/src/+/87331d860f547f0a5b5df862b77e7a353ea2724e

[carlosk, 2017-03-02 00:10:27]

Description was changed from

==========
Fix NavigationHandleImplBrowserTest flakiness.

NavigationHandleImplBrowserTest.VerifyRequestContextTypeForFrameTree has been
flaky because it relied on pointer comparisons for checking that new throttles
were being properly created for each new navigation. This is an unsafe solution
and has already caused problems in the past as it's not uncommon that new
instances of the same type of object are created at precisely the same memory
address. This is especially the case on Windows, where this test has been flaky.

This change uses a throttle-install-counter instead that works just as well for
(indirectly) checking that new throttles were installed without the pointer
comparison drawbacks. It also reenables that test on Windows.

Moreover it moves class member default initializations to be inline with their
declarations to simplify the code and cut back on line count.

BUG=696915
==========

to

==========
Fix NavigationHandleImplBrowserTest flakiness.

NavigationHandleImplBrowserTest.VerifyRequestContextTypeForFrameTree has been
flaky because it relied on pointer comparisons for checking that new throttles
were being properly created for each new navigation. This is an unsafe solution
and has already caused problems in the past as it's not uncommon that new
instances of the same type of object are created at precisely the same memory
address. This is especially the case on Windows, where this test has been flaky.

This change uses a throttle-install-counter instead that works just as well for
(indirectly) checking that new throttles were installed without the pointer
comparison drawbacks. It also reenables that test on Windows.

Moreover it moves class member default initializations to be inline with their
declarations to simplify the code and cut back on line count.

BUG=696915
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
==========

[carlosk, 2017-03-02 00:10:29]

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

[carlosk, 2017-03-02 00:10:47]

Description was changed from

==========
Fix NavigationHandleImplBrowserTest flakiness.

NavigationHandleImplBrowserTest.VerifyRequestContextTypeForFrameTree has been
flaky because it relied on pointer comparisons for checking that new throttles
were being properly created for each new navigation. This is an unsafe solution
and has already caused problems in the past as it's not uncommon that new
instances of the same type of object are created at precisely the same memory
address. This is especially the case on Windows, where this test has been flaky.

This change uses a throttle-install-counter instead that works just as well for
(indirectly) checking that new throttles were installed without the pointer
comparison drawbacks. It also reenables that test on Windows.

Moreover it moves class member default initializations to be inline with their
declarations to simplify the code and cut back on line count.

BUG=696915
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
==========

to

==========
Fix NavigationHandleImplBrowserTest flakiness.

NavigationHandleImplBrowserTest.VerifyRequestContextTypeForFrameTree has been
flaky because it relied on pointer comparisons for checking that new throttles
were being properly created for each new navigation. This is an unsafe solution
and has already caused problems in the past as it's not uncommon that new
instances of the same type of object are created at precisely the same memory
address. This is especially the case on Windows, where this test has been flaky.

This change uses a throttle-install-counter instead that works just as well for
(indirectly) checking that new throttles were installed without the pointer
comparison drawbacks. It also re-enables that test on Windows.

Moreover it moves class member default initializations to be inline with their
declarations to simplify the code and cut back on line count.

BUG=696915
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
==========

[commit-bot: I haz the power, 2017-03-02 00:11:28]

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

[carlosk, 2017-03-02 00:11:35]

carlosk@chromium.org changed reviewers:
+ clamy@chromium.org

[carlosk, 2017-03-02 00:11:36]

clamy@: PTAL.

[commit-bot: I haz the power, 2017-03-02 01:10:09]

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

[commit-bot: I haz the power, 2017-03-02 01:10:10]

Dry run: This issue passed the CQ dry run.

[clamy, 2017-03-02 13:36:54]

Thanks for the cleanup! Lgtm.

[carlosk, 2017-03-02 18:36:15]

The CQ bit was checked by carlosk@chromium.org

[commit-bot: I haz the power, 2017-03-02 18:36:37]

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, 2017-03-02 18:43:45]

CQ is committing da patch.

Bot data: {"patchset_id": 1, "attempt_start_ts": 1488479775721710, "parent_rev":
"866596040a23db3ab4152dba1d41f0a574fc2b83", "commit_rev":
"87331d860f547f0a5b5df862b77e7a353ea2724e"}

[commit-bot: I haz the power, 2017-03-02 18:45:18]

Description was changed from

==========
Fix NavigationHandleImplBrowserTest flakiness.

NavigationHandleImplBrowserTest.VerifyRequestContextTypeForFrameTree has been
flaky because it relied on pointer comparisons for checking that new throttles
were being properly created for each new navigation. This is an unsafe solution
and has already caused problems in the past as it's not uncommon that new
instances of the same type of object are created at precisely the same memory
address. This is especially the case on Windows, where this test has been flaky.

This change uses a throttle-install-counter instead that works just as well for
(indirectly) checking that new throttles were installed without the pointer
comparison drawbacks. It also re-enables that test on Windows.

Moreover it moves class member default initializations to be inline with their
declarations to simplify the code and cut back on line count.

BUG=696915
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
==========

to

==========
Fix NavigationHandleImplBrowserTest flakiness.

NavigationHandleImplBrowserTest.VerifyRequestContextTypeForFrameTree has been
flaky because it relied on pointer comparisons for checking that new throttles
were being properly created for each new navigation. This is an unsafe solution
and has already caused problems in the past as it's not uncommon that new
instances of the same type of object are created at precisely the same memory
address. This is especially the case on Windows, where this test has been flaky.

This change uses a throttle-install-counter instead that works just as well for
(indirectly) checking that new throttles were installed without the pointer
comparison drawbacks. It also re-enables that test on Windows.

Moreover it moves class member default initializations to be inline with their
declarations to simplify the code and cut back on line count.

BUG=696915
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation

Review-Url: https://codereview.chromium.org/2728783003
Cr-Commit-Position: refs/heads/master@{#454318}
Committed:
https://chromium.googlesource.com/chromium/src/+/87331d860f547f0a5b5df862b77e...
==========

[commit-bot: I haz the power, 2017-03-02 18:45:19]

Committed patchset #1 (id:1) as
https://chromium.googlesource.com/chromium/src/+/87331d860f547f0a5b5df862b77e...