Make RestoreSubframeFileAccessForHistoryNavigation pass in OOPIF modes.

Make RestoreSubframeFileAccessForHistoryNavigation pass in OOPIF modes.

Fixing the test properly will require tracking FrameTreeNodes that are
pending deletion, so that we can hear UpdateState messages from
subframes after doing a cross-process navigation in their ancestor.

Since we're only losing the last second of PageState changes at the
moment, we can work around this for now to let us proceed with moving
to the new navigation path.  Adding an in-page navigation ensures that
the PageState arrives in the browser process, letting the test pass.

BUG=609963
TEST=RestoreSubframeFileAccessForHistoryNavigation passes in
     --isolate-sites-for-testing=*.is or --site-per-process.
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_site_isolation

Committed: https://crrev.com/75504cc6c988e7765a7c61e7c0251b9886cd84dc
Cr-Commit-Position: refs/heads/master@{#401924}

[Charlie Reis, 2016-06-24 17:36:11]

Description was changed from

==========
Make RestoreSubframeFileAccessForHistoryNavigation pass in OOPIF modes.

Fixing the test properly will require tracking FrameTreeNodes that are
pending deletion, so that we can hear UpdateState messages from
subframes after doing a cross-process navigation in their ancestor.

Since we're only losing the last second of PageState changes at the
moment, we can work around this for now to let us proceed with moving
to the new navigation path.  Adding an in-page navigation ensures that
the PageState arrives in the browser process, letting the test pass.

BUG=609963
TEST=RestoreSubframeFileAccessForHistoryNavigation passes in
     --isolate-sites-for-testing=*.is or --site-per-process.
==========

to

==========
Make RestoreSubframeFileAccessForHistoryNavigation pass in OOPIF modes.

Fixing the test properly will require tracking FrameTreeNodes that are
pending deletion, so that we can hear UpdateState messages from
subframes after doing a cross-process navigation in their ancestor.

Since we're only losing the last second of PageState changes at the
moment, we can work around this for now to let us proceed with moving
to the new navigation path.  Adding an in-page navigation ensures that
the PageState arrives in the browser process, letting the test pass.

BUG=609963
TEST=RestoreSubframeFileAccessForHistoryNavigation passes in
     --isolate-sites-for-testing=*.is or --site-per-process.
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_site_isolation
==========

[Charlie Reis, 2016-06-24 18:03:14]

creis@chromium.org changed reviewers:
+ nick@chromium.org

[Charlie Reis, 2016-06-24 18:03:15]

Nick, here's the workaround that will let us pass the test (simulating receiving
the PageState 1 second after the dialog is chosen).  We can remove the
workaround once we have the pending delete FTN architecture.

[ncarter (slow), 2016-06-24 18:19:06]

lgtm

[Charlie Reis, 2016-06-24 18:23:32]

The CQ bit was checked by creis@chromium.org

[commit-bot: I haz the power, 2016-06-24 18:24:16]

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, 2016-06-24 19:20:25]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-06-24 19:24:28]

Description was changed from

==========
Make RestoreSubframeFileAccessForHistoryNavigation pass in OOPIF modes.

Fixing the test properly will require tracking FrameTreeNodes that are
pending deletion, so that we can hear UpdateState messages from
subframes after doing a cross-process navigation in their ancestor.

Since we're only losing the last second of PageState changes at the
moment, we can work around this for now to let us proceed with moving
to the new navigation path.  Adding an in-page navigation ensures that
the PageState arrives in the browser process, letting the test pass.

BUG=609963
TEST=RestoreSubframeFileAccessForHistoryNavigation passes in
     --isolate-sites-for-testing=*.is or --site-per-process.
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_site_isolation
==========

to

==========
Make RestoreSubframeFileAccessForHistoryNavigation pass in OOPIF modes.

Fixing the test properly will require tracking FrameTreeNodes that are
pending deletion, so that we can hear UpdateState messages from
subframes after doing a cross-process navigation in their ancestor.

Since we're only losing the last second of PageState changes at the
moment, we can work around this for now to let us proceed with moving
to the new navigation path.  Adding an in-page navigation ensures that
the PageState arrives in the browser process, letting the test pass.

BUG=609963
TEST=RestoreSubframeFileAccessForHistoryNavigation passes in
     --isolate-sites-for-testing=*.is or --site-per-process.
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_site_isolation

Committed: https://crrev.com/75504cc6c988e7765a7c61e7c0251b9886cd84dc
Cr-Commit-Position: refs/heads/master@{#401924}
==========

[commit-bot: I haz the power, 2016-06-24 19:24:30]

Patchset 2 (id:??) landed as
https://crrev.com/75504cc6c988e7765a7c61e7c0251b9886cd84dc
Cr-Commit-Position: refs/heads/master@{#401924}