[reland] Routing IME Result Calls to the Correct RenderWidgetHost (Aura Only)

[reland] Routing IME Result Calls to the Correct RenderWidgetHost (Aura Only)

This is a reland of the original CL: https://codereview.chromium.org/2045363002/ which was
reverted by this CL https://codereview.chromium.org/2095813002. The reason for the revert
was the regression of several SiteInstanceTests on Dr Memory bots.

The cause of the issue was that the newly introduced content unit tests in the original CL
were not deleting a RenderWidgetHostImpl which was keeping a RenderProcessHost alive. This
was interfering with the SiteInstanceTests which have explicit asserts on the number of
RenderProceessHosts alive.

BUG=578168, 622793, 602723
Committed: https://crrev.com/8cba7886c3e170ca2dc324dbbd31b28dae623887
Cr-Commit-Position: refs/heads/master@{#401939}

[EhsanK, 2016-06-24 14:05:25]

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

[commit-bot: I haz the power, 2016-06-24 14:05:33]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/2092103002/40001

[EhsanK, 2016-06-24 14:09:39]

Hello Charlie,

Sorry for the trouble, but could you please take a look.

I verified this locally by

1- Renaming my tests to something like SiteInstanceTestEhsan.*
2- Running Dr Memory light locally with --gtest_filter=*SiteInstanceTest* to
make sure that my tests and SiteInstance ones run together.

Without the fix I got the failures. After adding this all tests pass. I think
this should do.

https://codereview.chromium.org/2092103002/diff/1/content/browser/renderer_ho...
File content/browser/renderer_host/render_widget_host_view_aura_unittest.cc
(right):

https://codereview.chromium.org/2092103002/diff/1/content/browser/renderer_ho...
content/browser/renderer_host/render_widget_host_view_aura_unittest.cc:4080: //
This one is in-process with tab. The base class will clean the widget.
This comment is wrong and so was not deleting the RWH.

[commit-bot: I haz the power, 2016-06-24 14:52:03]

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

[commit-bot: I haz the power, 2016-06-24 14:52:04]

Dry run: This issue passed the CQ dry run.

[EhsanK, 2016-06-24 17:23:00]

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

[EhsanK, 2016-06-24 17:23:00]

Adding the creis@ whom I forgot to add in the earlier message.

[Charlie Reis, 2016-06-24 18:22:30]

Thanks for the quick fix!  LGTM with nit.

https://codereview.chromium.org/2092103002/diff/40001/content/browser/rendere...
File content/browser/renderer_host/render_widget_host_view_aura_unittest.cc
(right):

https://codereview.chromium.org/2092103002/diff/40001/content/browser/rendere...
content/browser/renderer_host/render_widget_host_view_aura_unittest.cc:4080:
RenderWidgetHost* host_for_first_process =
Ah, the confusing thing here is "host_for_first_process" is a RWH and not a RPH.
 Let's rename all of these to be widget_for_{Nth}_process to be clearer.

[EhsanK, 2016-06-24 18:37:55]

Thanks for the reviews!

https://codereview.chromium.org/2092103002/diff/40001/content/browser/rendere...
File content/browser/renderer_host/render_widget_host_view_aura_unittest.cc
(right):

https://codereview.chromium.org/2092103002/diff/40001/content/browser/rendere...
content/browser/renderer_host/render_widget_host_view_aura_unittest.cc:4080:
RenderWidgetHost* host_for_first_process =
On 2016/06/24 18:22:30, Charlie Reis wrote:
> Ah, the confusing thing here is "host_for_first_process" is a RWH and not a
RPH.
>  Let's rename all of these to be widget_for_{Nth}_process to be clearer.

Acknowledged. Also, moving this block to the beginning of the method to make
things in increasing order (first->second->third rather than
second->third->first).

[EhsanK, 2016-06-24 18:38:09]

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

[commit-bot: I haz the power, 2016-06-24 18:38:31]

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

[Charlie Reis, 2016-06-24 18:40:58]

Thanks, LGTM

[commit-bot: I haz the power, 2016-06-24 19:35:57]

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

[commit-bot: I haz the power, 2016-06-24 19:35:58]

Dry run: This issue passed the CQ dry run.

[EhsanK, 2016-06-24 19:36:22]

The CQ bit was checked by ekaramad@chromium.org

[commit-bot: I haz the power, 2016-06-24 19:37:16]

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 19:42:44]

Committed patchset #4 (id:60001)

[commit-bot: I haz the power, 2016-06-24 19:46:01]

Description was changed from

==========
[reland] Routing IME Result Calls to the Correct RenderWidgetHost (Aura Only)

This is a reland of the original CL: https://codereview.chromium.org/2045363002/
which was
reverted by this CL https://codereview.chromium.org/2095813002. The reason for
the revert
was the regression of several SiteInstanceTests on Dr Memory bots.

The cause of the issue was that the newly introduced content unit tests in the
original CL
were not deleting a RenderWidgetHostImpl which was keeping a RenderProcessHost
alive. This
was interfering with the SiteInstanceTests which have explicit asserts on the
number of
RenderProceessHosts alive.

BUG=578168,622793
==========

to

==========
[reland] Routing IME Result Calls to the Correct RenderWidgetHost (Aura Only)

This is a reland of the original CL: https://codereview.chromium.org/2045363002/
which was
reverted by this CL https://codereview.chromium.org/2095813002. The reason for
the revert
was the regression of several SiteInstanceTests on Dr Memory bots.

The cause of the issue was that the newly introduced content unit tests in the
original CL
were not deleting a RenderWidgetHostImpl which was keeping a RenderProcessHost
alive. This
was interfering with the SiteInstanceTests which have explicit asserts on the
number of
RenderProceessHosts alive.

BUG=578168,622793

Committed: https://crrev.com/8cba7886c3e170ca2dc324dbbd31b28dae623887
Cr-Commit-Position: refs/heads/master@{#401939}
==========

[commit-bot: I haz the power, 2016-06-24 19:46:02]

Patchset 4 (id:??) landed as
https://crrev.com/8cba7886c3e170ca2dc324dbbd31b28dae623887
Cr-Commit-Position: refs/heads/master@{#401939}

[EhsanK, 2016-07-05 15:22:47]

Description was changed from

==========
[reland] Routing IME Result Calls to the Correct RenderWidgetHost (Aura Only)

This is a reland of the original CL: https://codereview.chromium.org/2045363002/
which was
reverted by this CL https://codereview.chromium.org/2095813002. The reason for
the revert
was the regression of several SiteInstanceTests on Dr Memory bots.

The cause of the issue was that the newly introduced content unit tests in the
original CL
were not deleting a RenderWidgetHostImpl which was keeping a RenderProcessHost
alive. This
was interfering with the SiteInstanceTests which have explicit asserts on the
number of
RenderProceessHosts alive.

BUG=578168,622793

Committed: https://crrev.com/8cba7886c3e170ca2dc324dbbd31b28dae623887
Cr-Commit-Position: refs/heads/master@{#401939}
==========

to

==========
[reland] Routing IME Result Calls to the Correct RenderWidgetHost (Aura Only)

This is a reland of the original CL: https://codereview.chromium.org/2045363002/
which was
reverted by this CL https://codereview.chromium.org/2095813002. The reason for
the revert
was the regression of several SiteInstanceTests on Dr Memory bots.

The cause of the issue was that the newly introduced content unit tests in the
original CL
were not deleting a RenderWidgetHostImpl which was keeping a RenderProcessHost
alive. This
was interfering with the SiteInstanceTests which have explicit asserts on the
number of
RenderProceessHosts alive.

BUG=578168,622793,602723

Committed: https://crrev.com/8cba7886c3e170ca2dc324dbbd31b28dae623887
Cr-Commit-Position: refs/heads/master@{#401939}
==========