Implements N:M forwarding of OmniboxEditModel events on iOS.

Implements N:M forwarding of OmniboxEditModel events on iOS.

RLZTrackerDelegate implementation on iOS need to listen to events from
any OmniboxEditModel (all other clients of the same events on iOS also
do not care about which OmniboxEditModel sends them).

Introduce OmniboxEditModelObserverRelay to allow N:M forwarding. This
class will be informed of events sent by all OmniboxEditModel and will
forward them to all interested observers.

BUG=511570, 371536
Committed: https://crrev.com/62fe105c137ec2d3752c306b03cda5a2d3f0cda2
Cr-Commit-Position: refs/heads/master@{#339640}

[sdefresne, 2015-07-20 16:30:55]

sdefresne@chromium.org changed reviewers:
+ blundell@chromium.org

[sdefresne, 2015-07-20 16:30:55]

PTAL

[sdefresne, 2015-07-20 17:11:04]

Patchset #1 (id:1) has been deleted

[sdefresne, 2015-07-21 08:26:22]

https://codereview.chromium.org/1245683002/diff/20001/ios/chrome/browser/omni...
File ios/chrome/browser/omnibox/omnibox_edit_model_observer_relay.h (right):

https://codereview.chromium.org/1245683002/diff/20001/ios/chrome/browser/omni...
ios/chrome/browser/omnibox/omnibox_edit_model_observer_relay.h:19: class
OmniboxEditModelObserverRelay {
I don't like this name, what do you think of "OmniboxEditModelEventDispatcher"
instead?

[blundell, 2015-07-21 09:11:57]

lgtm

https://codereview.chromium.org/1245683002/diff/20001/ios/chrome/browser/omni...
File ios/chrome/browser/omnibox/omnibox_edit_model_observer_relay.h (right):

https://codereview.chromium.org/1245683002/diff/20001/ios/chrome/browser/omni...
ios/chrome/browser/omnibox/omnibox_edit_model_observer_relay.h:17: //
OmniboxEditModelObserverRelay does N:M matching between all OmniboxEditModel
How about something like:

"Omnibox code tracks events on a per-BrowserState basis, but there are several
clients who need to observe these events for all BrowserStates. This class
serves as an intermediary to bridge the gap: omnibox code calls the
OmniboxEventGlobalTracker singleton on an event of interest, and it then
forwards the event to its registered observers."

https://codereview.chromium.org/1245683002/diff/20001/ios/chrome/browser/omni...
ios/chrome/browser/omnibox/omnibox_edit_model_observer_relay.h:19: class
OmniboxEditModelObserverRelay {
On 2015/07/21 08:26:22, sdefresne wrote:
> I don't like this name, what do you think of "OmniboxEditModelEventDispatcher"
> instead?

How about "OmniboxEventGlobalTracker"?

https://codereview.chromium.org/1245683002/diff/20001/ios/chrome/browser/omni...
ios/chrome/browser/omnibox/omnibox_edit_model_observer_relay.h:26: // Registers
|cb| to be invoked when user open an URL from the omnibox.
nit: s/an URL/a URL

[sdefresne, 2015-07-21 12:05:07]

The CQ bit was checked by sdefresne@chromium.org

[sdefresne, 2015-07-21 12:05:07]

The patchset sent to the CQ was uploaded after l-g-t-m from
blundell@chromium.org
Link to the patchset: https://codereview.chromium.org/1245683002/#ps40001
(title: "Address comments by blundell")

[commit-bot: I haz the power, 2015-07-21 12:05:22]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1245683002/40001

[sdefresne, 2015-07-21 12:05:32]

https://codereview.chromium.org/1245683002/diff/20001/ios/chrome/browser/omni...
File ios/chrome/browser/omnibox/omnibox_edit_model_observer_relay.h (right):

https://codereview.chromium.org/1245683002/diff/20001/ios/chrome/browser/omni...
ios/chrome/browser/omnibox/omnibox_edit_model_observer_relay.h:17: //
OmniboxEditModelObserverRelay does N:M matching between all OmniboxEditModel
On 2015/07/21 at 09:11:57, blundell wrote:
> How about something like:
> 
> "Omnibox code tracks events on a per-BrowserState basis, but there are several
clients who need to observe these events for all BrowserStates. This class
serves as an intermediary to bridge the gap: omnibox code calls the
OmniboxEventGlobalTracker singleton on an event of interest, and it then
forwards the event to its registered observers."

Sounds better, done.

https://codereview.chromium.org/1245683002/diff/20001/ios/chrome/browser/omni...
ios/chrome/browser/omnibox/omnibox_edit_model_observer_relay.h:19: class
OmniboxEditModelObserverRelay {
On 2015/07/21 at 09:11:57, blundell wrote:
> On 2015/07/21 08:26:22, sdefresne wrote:
> > I don't like this name, what do you think of
"OmniboxEditModelEventDispatcher"
> > instead?
> 
> How about "OmniboxEventGlobalTracker"?

SGTM.

https://codereview.chromium.org/1245683002/diff/20001/ios/chrome/browser/omni...
ios/chrome/browser/omnibox/omnibox_edit_model_observer_relay.h:26: // Registers
|cb| to be invoked when user open an URL from the omnibox.
On 2015/07/21 at 09:11:57, blundell wrote:
> nit: s/an URL/a URL

As discussed offline, both are correct and depends on how you pronounce "URL".

[commit-bot: I haz the power, 2015-07-21 12:14:40]

Committed patchset #2 (id:40001)

[commit-bot: I haz the power, 2015-07-21 12:15:38]

Patchset 2 (id:??) landed as
https://crrev.com/62fe105c137ec2d3752c306b03cda5a2d3f0cda2
Cr-Commit-Position: refs/heads/master@{#339640}