Teach navigation throttles how to cancel requests in WillProcessResponse.

Teach navigation throttles how to cancel requests in WillProcessResponse.

This is the first step towards moving various response-time checks up
from Blink into the browser. To support things like 'X-Frame-Options',
we'll need to do throttle checks after we can evaluate headers. This
patch starts us down that path by teaching NavigationHandleImpl,
NavigationResourceThrottle, and NavigationThrottle about response-time
checks. This patch shouldn't introduce any behavioral changes, as no
NavigationThrottle actually uses the new functionality yet.

We'll start using the new functionality with the patch at
https://codereview.chromium.org/1617043002.

BUG=555418, 574997
TBR=palmer@chromium.org
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_site_isolation

------------------------------------------------------------------------
This will be a re-land of the patch from
https://crrev.com/409b51d438ea3f99d37025372adaf4d730553857
which was reverted due to timeouts in the Windows builder.
------------------------------------------------------------------------

Committed: https://crrev.com/59145851ada35fbf11d3cc39b4f98d462afdca60
Cr-Commit-Position: refs/heads/master@{#373557}

[Mike West, 2016-02-03 12:46:20]

mkwst@chromium.org changed reviewers:
+ clamy@chromium.org, nasko@chromium.org

[Mike West, 2016-02-03 12:46:21]

Hey clamy@ and nasko@! I'd like to reland this, but it's not at all clear to me
why it's failing on Windows. I split up the relevant test, and it looks like
it's not successfully navigating to a `file:///` URL under Windows. Is there
something special I should be looking at? Halp! :)

[nasko, 2016-02-03 22:44:56]

On 2016/02/03 12:46:21, Mike West (OOO until 2nd) wrote:
> Hey clamy@ and nasko@! I'd like to reland this, but it's not at all clear to
me
> why it's failing on Windows. I split up the relevant test, and it looks like
> it's not successfully navigating to a `file:///` URL under Windows. Is there
> something special I should be looking at? Halp! :)

This looks very similar to crbug.com/574997, where file: URLs to nonexistent
resource never finishes. In the case of the bug, it is in site-isolation mode,
which introduces a thread hop to ensure we commit in the right RenderFrameHost.
It looks like with your changes, we are adding the same thread hop model and
hitting similar failure.

Since it is passing in all other OS versions, I'm ok with disabling it on
Windows and tagging it with crbug.com/574997, so we can make progress on the
rest of the work.

[nasko, 2016-02-03 22:48:27]

LGTM with a note.

https://codereview.chromium.org/1659163006/diff/20001/content/public/browser/...
File content/public/browser/navigation_throttle.h (right):

https://codereview.chromium.org/1659163006/diff/20001/content/public/browser/...
content/public/browser/navigation_throttle.h:65: virtual ThrottleCheckResult
WillProcessResponse();
In general, we don't add public API in a patch of its own, but with the usage of
the API. I missed this in the previous round :(.
Since this is just a small breakout CL (and a reland), can you add the follow up
CL link in the description?

[Mike West, 2016-02-04 08:57:26]

Description was changed from

==========
Teach navigation throttles how to cancel requests in WillProcessResponse.

This is the first step towards moving various response-time checks up
from Blink into the browser. To support things like 'X-Frame-Options',
we'll need to do throttle checks after we can evaluate headers. This
patch starts us down that path by teaching NavigationHandleImpl,
NavigationResourceThrottle, and NavigationThrottle about response-time
checks. This patch shouldn't introduce any behavioral changes, as no
NavigationThrottle actually uses the new functionality yet.

BUG=555418
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_site_isolation

------------------------------------------------------------------------
This will be a re-land of the patch from
https://crrev.com/409b51d438ea3f99d37025372adaf4d730553857
which was reverted due to timeouts in the Windows builder.
------------------------------------------------------------------------
==========

to

==========
Teach navigation throttles how to cancel requests in WillProcessResponse.

This is the first step towards moving various response-time checks up
from Blink into the browser. To support things like 'X-Frame-Options',
we'll need to do throttle checks after we can evaluate headers. This
patch starts us down that path by teaching NavigationHandleImpl,
NavigationResourceThrottle, and NavigationThrottle about response-time
checks. This patch shouldn't introduce any behavioral changes, as no
NavigationThrottle actually uses the new functionality yet.

We'll start using the new functionality with the patch at
https://codereview.chromium.org/1617043002.

BUG=555418
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_site_isolation

------------------------------------------------------------------------
This will be a re-land of the patch from
https://crrev.com/409b51d438ea3f99d37025372adaf4d730553857
which was reverted due to timeouts in the Windows builder.
------------------------------------------------------------------------
==========

[Mike West, 2016-02-04 09:02:39]

Description was changed from

==========
Teach navigation throttles how to cancel requests in WillProcessResponse.

This is the first step towards moving various response-time checks up
from Blink into the browser. To support things like 'X-Frame-Options',
we'll need to do throttle checks after we can evaluate headers. This
patch starts us down that path by teaching NavigationHandleImpl,
NavigationResourceThrottle, and NavigationThrottle about response-time
checks. This patch shouldn't introduce any behavioral changes, as no
NavigationThrottle actually uses the new functionality yet.

We'll start using the new functionality with the patch at
https://codereview.chromium.org/1617043002.

BUG=555418
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_site_isolation

------------------------------------------------------------------------
This will be a re-land of the patch from
https://crrev.com/409b51d438ea3f99d37025372adaf4d730553857
which was reverted due to timeouts in the Windows builder.
------------------------------------------------------------------------
==========

to

==========
Teach navigation throttles how to cancel requests in WillProcessResponse.

This is the first step towards moving various response-time checks up
from Blink into the browser. To support things like 'X-Frame-Options',
we'll need to do throttle checks after we can evaluate headers. This
patch starts us down that path by teaching NavigationHandleImpl,
NavigationResourceThrottle, and NavigationThrottle about response-time
checks. This patch shouldn't introduce any behavioral changes, as no
NavigationThrottle actually uses the new functionality yet.

We'll start using the new functionality with the patch at
https://codereview.chromium.org/1617043002.

BUG=555418, 574997
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_site_isolation

------------------------------------------------------------------------
This will be a re-land of the patch from
https://crrev.com/409b51d438ea3f99d37025372adaf4d730553857
which was reverted due to timeouts in the Windows builder.
------------------------------------------------------------------------
==========

[Mike West, 2016-02-04 09:02:57]

Description was changed from

==========
Teach navigation throttles how to cancel requests in WillProcessResponse.

This is the first step towards moving various response-time checks up
from Blink into the browser. To support things like 'X-Frame-Options',
we'll need to do throttle checks after we can evaluate headers. This
patch starts us down that path by teaching NavigationHandleImpl,
NavigationResourceThrottle, and NavigationThrottle about response-time
checks. This patch shouldn't introduce any behavioral changes, as no
NavigationThrottle actually uses the new functionality yet.

We'll start using the new functionality with the patch at
https://codereview.chromium.org/1617043002.

BUG=555418, 574997
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_site_isolation

------------------------------------------------------------------------
This will be a re-land of the patch from
https://crrev.com/409b51d438ea3f99d37025372adaf4d730553857
which was reverted due to timeouts in the Windows builder.
------------------------------------------------------------------------
==========

to

==========
Teach navigation throttles how to cancel requests in WillProcessResponse.

This is the first step towards moving various response-time checks up
from Blink into the browser. To support things like 'X-Frame-Options',
we'll need to do throttle checks after we can evaluate headers. This
patch starts us down that path by teaching NavigationHandleImpl,
NavigationResourceThrottle, and NavigationThrottle about response-time
checks. This patch shouldn't introduce any behavioral changes, as no
NavigationThrottle actually uses the new functionality yet.

We'll start using the new functionality with the patch at
https://codereview.chromium.org/1617043002.

BUG=555418, 574997
TBR=palmer@chromium.org
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_site_isolation

------------------------------------------------------------------------
This will be a re-land of the patch from
https://crrev.com/409b51d438ea3f99d37025372adaf4d730553857
which was reverted due to timeouts in the Windows builder.
------------------------------------------------------------------------
==========

[Mike West, 2016-02-04 09:03:45]

mkwst@chromium.org changed reviewers:
+ palmer@chromium.org

[Mike West, 2016-02-04 09:03:46]

Thanks, nasko@. I've addressed your comments.

TBRing palmer@ for the test change, as I assume he added it. Hi Chris!

[Mike West, 2016-02-04 09:03:49]

The CQ bit was checked by mkwst@chromium.org

[Mike West, 2016-02-04 09:03:49]

The patchset sent to the CQ was uploaded after l-g-t-m from nasko@chromium.org
Link to the patchset: https://codereview.chromium.org/1659163006/#ps40001
(title: "Disable the test on WIN.")

[commit-bot: I haz the power, 2016-02-04 09:04:16]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1659163006/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1659163006/40001

[commit-bot: I haz the power, 2016-02-04 12:58:57]

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

[commit-bot: I haz the power, 2016-02-04 12:58:59]

Try jobs failed on following builders:
  linux_site_isolation on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_site_isol...)

[nasko, 2016-02-04 17:20:52]

The CQ bit was checked by nasko@chromium.org

[commit-bot: I haz the power, 2016-02-04 17:21:23]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1659163006/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1659163006/40001

[commit-bot: I haz the power, 2016-02-04 17:39:36]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2016-02-04 17:41:00]

Description was changed from

==========
Teach navigation throttles how to cancel requests in WillProcessResponse.

This is the first step towards moving various response-time checks up
from Blink into the browser. To support things like 'X-Frame-Options',
we'll need to do throttle checks after we can evaluate headers. This
patch starts us down that path by teaching NavigationHandleImpl,
NavigationResourceThrottle, and NavigationThrottle about response-time
checks. This patch shouldn't introduce any behavioral changes, as no
NavigationThrottle actually uses the new functionality yet.

We'll start using the new functionality with the patch at
https://codereview.chromium.org/1617043002.

BUG=555418, 574997
TBR=palmer@chromium.org
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_site_isolation

------------------------------------------------------------------------
This will be a re-land of the patch from
https://crrev.com/409b51d438ea3f99d37025372adaf4d730553857
which was reverted due to timeouts in the Windows builder.
------------------------------------------------------------------------
==========

to

==========
Teach navigation throttles how to cancel requests in WillProcessResponse.

This is the first step towards moving various response-time checks up
from Blink into the browser. To support things like 'X-Frame-Options',
we'll need to do throttle checks after we can evaluate headers. This
patch starts us down that path by teaching NavigationHandleImpl,
NavigationResourceThrottle, and NavigationThrottle about response-time
checks. This patch shouldn't introduce any behavioral changes, as no
NavigationThrottle actually uses the new functionality yet.

We'll start using the new functionality with the patch at
https://codereview.chromium.org/1617043002.

BUG=555418, 574997
TBR=palmer@chromium.org
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_site_isolation

------------------------------------------------------------------------
This will be a re-land of the patch from
https://crrev.com/409b51d438ea3f99d37025372adaf4d730553857
which was reverted due to timeouts in the Windows builder.
------------------------------------------------------------------------

Committed: https://crrev.com/59145851ada35fbf11d3cc39b4f98d462afdca60
Cr-Commit-Position: refs/heads/master@{#373557}
==========

[commit-bot: I haz the power, 2016-02-04 17:41:01]

Patchset 3 (id:??) landed as
https://crrev.com/59145851ada35fbf11d3cc39b4f98d462afdca60
Cr-Commit-Position: refs/heads/master@{#373557}