Add support for MojoChannel field trial

Add support for MojoChannel field trial

This adds support for the upcoming MojoChannel field trial.

It also simplifies the logic around enabling Mojo channel
in general, getting rid of compile-time overrides and instead
controlling the feature only at runtime via a single flag or
(in the absence of said flag) the field trial.

BUG=595082
R=jam@chromium.org

Committed: https://crrev.com/b0abbd20ecfc1b84f7570ee7f5df100f043010e8
Cr-Commit-Position: refs/heads/master@{#381490}

[Ken Rockot(use gerrit already), 2016-03-15 22:17:01]

The actual experiment is still awaiting launch approval but here's the
client-side change.

[jam, 2016-03-15 22:50:48]

lgtm

https://codereview.chromium.org/1805653003/diff/1/content/public/common/mojo_...
File content/public/common/mojo_channel_switches.cc (right):

https://codereview.chromium.org/1805653003/diff/1/content/public/common/mojo_...
content/public/common/mojo_channel_switches.cc:32: // the Mojo channel.
nit: no need for a comment that just describes simple code per
https://google.github.io/styleguide/cppguide.html#Implementation_Comments

[Ken Rockot(use gerrit already), 2016-03-15 22:54:07]

https://codereview.chromium.org/1805653003/diff/1/content/public/common/mojo_...
File content/public/common/mojo_channel_switches.cc (right):

https://codereview.chromium.org/1805653003/diff/1/content/public/common/mojo_...
content/public/common/mojo_channel_switches.cc:32: // the Mojo channel.
On 2016/03/15 at 22:50:47, jam wrote:
> nit: no need for a comment that just describes simple code per
https://google.github.io/styleguide/cppguide.html#Implementation_Comments

Done

[Ken Rockot(use gerrit already), 2016-03-16 15:32:31]

I'm going to land this since the switch is useful and the field trial is a no-op
until approved.

[Ken Rockot(use gerrit already), 2016-03-16 15:32:34]

The CQ bit was checked by rockot@chromium.org

[Ken Rockot(use gerrit already), 2016-03-16 15:32:34]

The patchset sent to the CQ was uploaded after l-g-t-m from jam@chromium.org
Link to the patchset: https://codereview.chromium.org/1805653003/#ps20001
(title: "nit")

[commit-bot: I haz the power, 2016-03-16 15:32:46]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1805653003/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1805653003/20001

[commit-bot: I haz the power, 2016-03-16 17:50:47]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-03-16 17:53:30]

Description was changed from

==========
Add support for MojoChannel field trial

This adds support for the upcoming MojoChannel field trial.

It also simplifies the logic around enabling Mojo channel
in general, getting rid of compile-time overrides and instead
controlling the feature only at runtime via a single flag or
(in the absence of said flag) the field trial.

BUG=595082
R=jam@chromium.org
==========

to

==========
Add support for MojoChannel field trial

This adds support for the upcoming MojoChannel field trial.

It also simplifies the logic around enabling Mojo channel
in general, getting rid of compile-time overrides and instead
controlling the feature only at runtime via a single flag or
(in the absence of said flag) the field trial.

BUG=595082
R=jam@chromium.org

Committed: https://crrev.com/b0abbd20ecfc1b84f7570ee7f5df100f043010e8
Cr-Commit-Position: refs/heads/master@{#381490}
==========

[commit-bot: I haz the power, 2016-03-16 17:53:31]

Patchset 2 (id:??) landed as
https://crrev.com/b0abbd20ecfc1b84f7570ee7f5df100f043010e8
Cr-Commit-Position: refs/heads/master@{#381490}

[Will Harris, 2016-04-15 16:59:54]

wfh@chromium.org changed reviewers:
+ wfh@chromium.org

[Will Harris, 2016-04-15 16:59:54]

https://codereview.chromium.org/1805653003/diff/20001/content/public/common/m...
File content/public/common/mojo_channel_switches.cc (right):

https://codereview.chromium.org/1805653003/diff/20001/content/public/common/m...
content/public/common/mojo_channel_switches.cc:36:
base::FieldTrialList::FindFullName(kMojoChannelExperimentName);
the query to FindFullName should always take place, otherwise some clients will
not be placed in any experiment group, and your data will be wrong.