Fix RenderFrameHostManagerTest.DontSwapProcessWithOnlyTargetBlank for OOPIF

Fix RenderFrameHostManagerTest.DontSwapProcessWithOnlyTargetBlank for OOPIF.

This test checks that following a cross-site link with target=_blank does not create a new SiteInstance.  Per https://crbug.com/24447, we currently keep the new page in the same SiteInstance to avoid breaking script interactions between windows.  However, with --site-per-process we do expect a process swap in this situation, and this CL changes the test's expectation accordingly.

BUG=417518
Committed: https://crrev.com/82e94f8c27cf91a2fb8fdf0a375f1100fe1bd8b4
Cr-Commit-Position: refs/heads/master@{#298610}

[alexmos, 2014-10-07 22:16:59]

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

[alexmos, 2014-10-07 22:23:15]

Hey Charlie, could you please review this CL?

[Charlie Reis, 2014-10-07 22:40:38]

LGTM, but let's rephrase the CL description.

(It's not quite right to say that the desirable / soon-to-be-expected behavior
is a process swap here, since we will be deploying --site-per-process only for
certain sites.  Most will still have no process swap on a target=_blank link. 
Better to just say that we expect a swap with --site-per-process.)

https://codereview.chromium.org/635933002/diff/40001/content/browser/frame_ho...
File content/browser/frame_host/render_frame_host_manager_browsertest.cc
(right):

https://codereview.chromium.org/635933002/diff/40001/content/browser/frame_ho...
content/browser/frame_host/render_frame_host_manager_browsertest.cc:295: if
(base::CommandLine::ForCurrentProcess()->HasSwitch(
nit: Let's invert this so that _EQ comes before _NE, to match how the comment is
phrased.

[alexmos, 2014-10-07 22:55:19]

On 2014/10/07 22:40:38, Charlie Reis wrote:
> LGTM, but let's rephrase the CL description.
> 
> (It's not quite right to say that the desirable / soon-to-be-expected behavior
> is a process swap here, since we will be deploying --site-per-process only for
> certain sites.  Most will still have no process swap on a target=_blank link. 
> Better to just say that we expect a swap with --site-per-process.)

Good point - done.

> 
>
https://codereview.chromium.org/635933002/diff/40001/content/browser/frame_ho...
> File content/browser/frame_host/render_frame_host_manager_browsertest.cc
> (right):
> 
>
https://codereview.chromium.org/635933002/diff/40001/content/browser/frame_ho...
> content/browser/frame_host/render_frame_host_manager_browsertest.cc:295: if
> (base::CommandLine::ForCurrentProcess()->HasSwitch(
> nit: Let's invert this so that _EQ comes before _NE, to match how the comment
is
> phrased.

Done.  Thanks!

[Charlie Reis, 2014-10-07 23:20:57]

Thanks.  Still LGTM.

[alexmos, 2014-10-07 23:32:29]

The CQ bit was checked by alexmos@chromium.org

[commit-bot: I haz the power, 2014-10-07 23:34:05]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/635933002/60001

[commit-bot: I haz the power, 2014-10-07 23:39:20]

Committed patchset #4 (id:60001) as 63fbea03521cd5ebf7d16dbf173055a70f475df3

[commit-bot: I haz the power, 2014-10-07 23:40:16]

Patchset 4 (id:??) landed as
https://crrev.com/82e94f8c27cf91a2fb8fdf0a375f1100fe1bd8b4
Cr-Commit-Position: refs/heads/master@{#298610}