Mojo C++ Bindings: Additional support for associated binding sets

Mojo C++ Bindings: Additional support for associated binding sets

Implements three orthogonal changes to the public bindings API:

 1. Adds the ability to remove bindings from a binding set. Uses
    a new mojo::BindingId integer alias to identify added bindings.

 2. Adds the ability to set a pre-dispatch hook on binding sets.
    In some cases simply making the dispatch_context() available
    is insufficient, for example when a consumer owns multiple
    binding sets and wants a common path for establishing context
    among any of them.

 3. Adds mojo::GetDummyProxy, a helper that can bind an
    AssociatedInterfacePtr<T> so that it has some valid internal
    state and silently drops all outgoing messages. Unlike
    regular message pipe interfaces, without this feature
    associated interfaces are difficult to stub out in test
    environments.

Part a series of CLs to enable and demonstrate WebContents associated
interfaces:

  1. https://codereview.chromium.org/2301123004
  2. This CL
  3. https://codereview.chromium.org/2310563002
  4. https://codereview.chromium.org/2310583002

BUG=612500
R=yzshen@chromium.org

Committed: https://crrev.com/4b8bca190cdc55bae81974aa09cc10f422f022fb
Cr-Commit-Position: refs/heads/master@{#417126}

[Ken Rockot(use gerrit already), 2016-09-02 20:01:37]

Description was changed from

==========
Mojo C++ Bindings: Additional support for associated binding sets

Implements three orthogonal changes to the public bindings API:

 1. Adds the ability to remove bindings from a binding set. Uses
    a new mojo::BindingId integer alias to identify added bindings.

 2. Adds the ability to set a pre-dispatch hook on binding sets.
    In some cases simply making the dispatch_context() available
    is insufficient, for example when a consumer owns multiple
    binding sets and wants a common path for establishing context
    among any of them.

 3. Adds mojo::GetDummyProxy, a helper that can bind an
    AssociatedInterfacePtr<T> so that it has some valid internal
    state and silently drops all outgoing messages. Unlike
    regular message pipe interfaces, without this feature
    associated interfaces are difficult to stub out in test
    environments.

These changes are prerequisite for follow-up patches to add more
support for route-associated interfaces between RenderFrame/Host.

BUG=612500
R=yzshen@chromium.org
==========

to

==========
Mojo C++ Bindings: Additional support for associated binding sets

Implements three orthogonal changes to the public bindings API:

 1. Adds the ability to remove bindings from a binding set. Uses
    a new mojo::BindingId integer alias to identify added bindings.

 2. Adds the ability to set a pre-dispatch hook on binding sets.
    In some cases simply making the dispatch_context() available
    is insufficient, for example when a consumer owns multiple
    binding sets and wants a common path for establishing context
    among any of them.

 3. Adds mojo::GetDummyProxy, a helper that can bind an
    AssociatedInterfacePtr<T> so that it has some valid internal
    state and silently drops all outgoing messages. Unlike
    regular message pipe interfaces, without this feature
    associated interfaces are difficult to stub out in test
    environments.

Part a series of CLs to enable and demonstrate WebContents associated
interfaces:

  1. https://codereview.chromium.org/2301123004
  2. This CL
  3. https://codereview.chromium.org/2310563002
  4. https://codereview.chromium.org/2310583002

BUG=612500
R=yzshen@chromium.org
==========

[Ken Rockot(use gerrit already), 2016-09-02 20:09:06]

Please take a look. Just some miscellaneous support that made sense to me. WDYT?

[yzshen1, 2016-09-02 22:28:57]

On 2016/09/02 20:09:06, Ken Rockot wrote:
> Please take a look. Just some miscellaneous support that made sense to me.
WDYT?


It seems the GetDummyProxy() helper can be done without any change to associated
group/controller, right?
We could create a MultiplexRouter, and then use the associated group from it to
create a WILL_PASS_REQUEST pair of handle, using the AssociatedInterfacePtrInfo
to construct an associated pointer.

And it may make sense to move it to some testing utility file.

WDYT? Thanks!

[Ken Rockot(use gerrit already), 2016-09-06 17:37:25]

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

[commit-bot: I haz the power, 2016-09-06 17:37:54]

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

[Ken Rockot(use gerrit already), 2016-09-06 17:39:06]

On 2016/09/02 at 22:28:57, yzshen wrote:
> On 2016/09/02 20:09:06, Ken Rockot wrote:
> > Please take a look. Just some miscellaneous support that made sense to me.
WDYT?
> 
> 
> It seems the GetDummyProxy() helper can be done without any change to
associated group/controller, right?
> We could create a MultiplexRouter, and then use the associated group from it
to create a WILL_PASS_REQUEST pair of handle, using the
AssociatedInterfacePtrInfo to construct an associated pointer.

SGTM, done

> 
> And it may make sense to move it to some testing utility file.
> 
> WDYT? Thanks!

I've renamed it to GetDummyProxyForTesting. This will still have to be
referenced in code which isn't strictly test code, since it will conditionally
be used by a public template helper that is called from both testing and
non-testing environments.

[commit-bot: I haz the power, 2016-09-06 20:14:35]

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

[commit-bot: I haz the power, 2016-09-06 20:14:36]

Dry run: This issue passed the CQ dry run.

[Ken Rockot(use gerrit already), 2016-09-07 22:27:00]

Gentle ping!

[yzshen1, 2016-09-07 22:32:59]

On 2016/09/07 22:27:00, Ken Rockot wrote:
> Gentle ping!

Sorry, I thought I had replied LGTM!

[Ken Rockot(use gerrit already), 2016-09-07 22:42:14]

The CQ bit was checked by rockot@chromium.org

[commit-bot: I haz the power, 2016-09-07 22:42:59]

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, 2016-09-08 00:14:53]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-09-08 00:16:58]

Description was changed from

==========
Mojo C++ Bindings: Additional support for associated binding sets

Implements three orthogonal changes to the public bindings API:

 1. Adds the ability to remove bindings from a binding set. Uses
    a new mojo::BindingId integer alias to identify added bindings.

 2. Adds the ability to set a pre-dispatch hook on binding sets.
    In some cases simply making the dispatch_context() available
    is insufficient, for example when a consumer owns multiple
    binding sets and wants a common path for establishing context
    among any of them.

 3. Adds mojo::GetDummyProxy, a helper that can bind an
    AssociatedInterfacePtr<T> so that it has some valid internal
    state and silently drops all outgoing messages. Unlike
    regular message pipe interfaces, without this feature
    associated interfaces are difficult to stub out in test
    environments.

Part a series of CLs to enable and demonstrate WebContents associated
interfaces:

  1. https://codereview.chromium.org/2301123004
  2. This CL
  3. https://codereview.chromium.org/2310563002
  4. https://codereview.chromium.org/2310583002

BUG=612500
R=yzshen@chromium.org
==========

to

==========
Mojo C++ Bindings: Additional support for associated binding sets

Implements three orthogonal changes to the public bindings API:

 1. Adds the ability to remove bindings from a binding set. Uses
    a new mojo::BindingId integer alias to identify added bindings.

 2. Adds the ability to set a pre-dispatch hook on binding sets.
    In some cases simply making the dispatch_context() available
    is insufficient, for example when a consumer owns multiple
    binding sets and wants a common path for establishing context
    among any of them.

 3. Adds mojo::GetDummyProxy, a helper that can bind an
    AssociatedInterfacePtr<T> so that it has some valid internal
    state and silently drops all outgoing messages. Unlike
    regular message pipe interfaces, without this feature
    associated interfaces are difficult to stub out in test
    environments.

Part a series of CLs to enable and demonstrate WebContents associated
interfaces:

  1. https://codereview.chromium.org/2301123004
  2. This CL
  3. https://codereview.chromium.org/2310563002
  4. https://codereview.chromium.org/2310583002

BUG=612500
R=yzshen@chromium.org

Committed: https://crrev.com/4b8bca190cdc55bae81974aa09cc10f422f022fb
Cr-Commit-Position: refs/heads/master@{#417126}
==========

[commit-bot: I haz the power, 2016-09-08 00:16:59]

Patchset 2 (id:??) landed as
https://crrev.com/4b8bca190cdc55bae81974aa09cc10f422f022fb
Cr-Commit-Position: refs/heads/master@{#417126}

[lushnikov, 2016-09-08 01:01:36]

A revert of this CL (patchset #2 id:20001) has been created in
https://codereview.chromium.org/2324623003/ by lushnikov@chromium.org.

The reason for reverting is: This broke compilation on Win x64 builder:

https://build.chromium.org/p/chromium.webkit/builders/WebKit%20Win%20x64%20Bu...

.