Ensure WebUI bindings are properly set

Ensure WebUI bindings are properly set

When a RenderViewHost is created in swapped out state, it doesn't have
any WebUI bindings. If later on a navigation is performed that uses
the same RenderViewHost, we need to ensure it has WebUI bindings.
Currently the code does that conditionally inside RFHM::CreateRenderFrame
and it should be done unconditionally.

BUG=357747
TEST=Existing browser tests pass, including WebUIGetsBindings

Committed: https://crrev.com/25c15f366899cea3ec6d72904cb97d61d78e79f3
Cr-Commit-Position: refs/heads/master@{#330956}

[nasko, 2015-05-20 23:28:28]

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

[nasko, 2015-05-20 23:28:28]

Hey Charlie,
Can you review this CL for me? It moves the WebUI bindings code to be executed
unconditionally in RFHM::CreateRenderFrame.
I tried poking at the other usages of AllowBindings, but didn't seem
straightforward to alter. As such, I'm leaving them as is and will be good to
revisit this whole area as we need to move bindings from RenderViewHost to
RenderFrameHost.
Thanks in advance!
Nasko

[Charlie Reis, 2015-05-21 18:35:03]

Looks reasonable.  If I understand correctly, we don't actually need it in the
other branch today, but it makes the bindings work in your next CL(s) when we
stop using swappedout:// in the main frame, right?  LGTM.

[nasko, 2015-05-21 18:50:25]

On 2015/05/21 18:35:03, Charlie Reis wrote:
> Looks reasonable.  If I understand correctly, we don't actually need it in the
> other branch today, but it makes the bindings work in your next CL(s) when we
> stop using swappedout:// in the main frame, right?  LGTM.

That is correct. The reason is that with my next patch, we no longer use
CreateRenderFrame to create swapped out RFHs, but directly create proxies, which
don't assign any bindings to the associated RenderViewHost. With this being
unconditional, we will always set the proper bindings on the RVH at the moment
we create the first RFH associated with it.

[nasko, 2015-05-21 18:50:33]

The CQ bit was checked by nasko@chromium.org

[commit-bot: I haz the power, 2015-05-21 19:40:53]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1145353002/1

[commit-bot: I haz the power, 2015-05-21 19:47:33]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2015-05-21 19:48:16]

Patchset 1 (id:??) landed as
https://crrev.com/25c15f366899cea3ec6d72904cb97d61d78e79f3
Cr-Commit-Position: refs/heads/master@{#330956}