[sync] Refactor how default sync datatypes are set.

[sync] Refactor how default sync datatypes are set.

Android does not sync all datatypes, so we need a way to differentiate
which datatypes should be enabled by default.

Since parts of this require extensions, we need to register the user preferences for extensions as part of this.

BUG=139057
TEST=Signing in to sync on should enable all the correct default
datatypes.

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=150485

[nyquist, 2012-07-26 19:23:39]

Moving component startup around so it is easier to ifdef out non-Android sync
types.

[nilesh, 2012-07-27 00:59:07]

https://chromiumcodereview.appspot.com/10829019/diff/2001/chrome/browser/sync...
File chrome/browser/sync/profile_sync_components_factory_impl.cc (right):

https://chromiumcodereview.appspot.com/10829019/diff/2001/chrome/browser/sync...
chrome/browser/sync/profile_sync_components_factory_impl.cc:91:
extension_system_(NULL) {
I think we include ExtensionSystemFactory and this is not needed. If this fails
at runtime, we need a stub or something else. 
If you absolutely need to do this, it should be protected by ENABLE_EXTENSIONS
(and not OS_ANDROID)

[akalin, 2012-07-31 22:25:21]

http://codereview.chromium.org/10829019/diff/7002/chrome/browser/prefs/browse...
File chrome/browser/prefs/browser_prefs.cc (right):

http://codereview.chromium.org/10829019/diff/7002/chrome/browser/prefs/browse...
chrome/browser/prefs/browser_prefs.cc:241:
extensions::ExtensionPrefs::RegisterUserPrefs(user_prefs);
mention somewhere why this was moved out?

[nyquist, 2012-08-01 16:36:05]

Added CL description about registering extension user preferences.

Adding yfriedman@ for sanity check.

[nilesh, 2012-08-01 16:43:20]

https://chromiumcodereview.appspot.com/10829019/diff/7002/chrome/browser/sync...
File chrome/browser/sync/profile_sync_components_factory_impl.cc (right):

https://chromiumcodereview.appspot.com/10829019/diff/7002/chrome/browser/sync...
chrome/browser/sync/profile_sync_components_factory_impl.cc:264: return
base::WeakPtr<syncer::SyncableService>();
Now that we are initializing extension system, I think we don't need this ifdef
either.

[nyquist, 2012-08-02 18:28:17]

Addressed all comments. PTAL

http://codereview.chromium.org/10829019/diff/7002/chrome/browser/prefs/browse...
File chrome/browser/prefs/browser_prefs.cc (right):

http://codereview.chromium.org/10829019/diff/7002/chrome/browser/prefs/browse...
chrome/browser/prefs/browser_prefs.cc:241:
extensions::ExtensionPrefs::RegisterUserPrefs(user_prefs);
On 2012/07/31 22:25:22, akalin wrote:
> mention somewhere why this was moved out?

Added reason to CL description. Should help people doing git blame.

http://codereview.chromium.org/10829019/diff/7002/chrome/browser/sync/profile...
File chrome/browser/sync/profile_sync_components_factory_impl.cc (right):

http://codereview.chromium.org/10829019/diff/7002/chrome/browser/sync/profile...
chrome/browser/sync/profile_sync_components_factory_impl.cc:264: return
base::WeakPtr<syncer::SyncableService>();
On 2012/08/01 16:43:20, nileshagrawal1 wrote:
> Now that we are initializing extension system, I think we don't need this
ifdef
> either.

Verified that it still builds and you can sign in to sync without this
downstream.

[nilesh, 2012-08-02 19:15:05]

On 2012/08/02 18:28:17, nyquist wrote:
> Addressed all comments. PTAL
> 
>
http://codereview.chromium.org/10829019/diff/7002/chrome/browser/prefs/browse...
> File chrome/browser/prefs/browser_prefs.cc (right):
> 
>
http://codereview.chromium.org/10829019/diff/7002/chrome/browser/prefs/browse...
> chrome/browser/prefs/browser_prefs.cc:241:
> extensions::ExtensionPrefs::RegisterUserPrefs(user_prefs);
> On 2012/07/31 22:25:22, akalin wrote:
> > mention somewhere why this was moved out?
> 
> Added reason to CL description. Should help people doing git blame.
> 
>
http://codereview.chromium.org/10829019/diff/7002/chrome/browser/sync/profile...
> File chrome/browser/sync/profile_sync_components_factory_impl.cc (right):
> 
>
http://codereview.chromium.org/10829019/diff/7002/chrome/browser/sync/profile...
> chrome/browser/sync/profile_sync_components_factory_impl.cc:264: return
> base::WeakPtr<syncer::SyncableService>();
> On 2012/08/01 16:43:20, nileshagrawal1 wrote:
> > Now that we are initializing extension system, I think we don't need this
> ifdef
> > either.
> 
> Verified that it still builds and you can sign in to sync without this
> downstream.

LGTM

[nyquist, 2012-08-03 00:33:05]

bauerb@ for chrome/browser/prefs

[Bernhard Bauer, 2012-08-04 09:12:49]

LGTM

[akalin, 2012-08-07 00:45:59]

On 2012/08/04 09:12:49, Bernhard Bauer wrote:
> LGTM

lgtm

[akalin, 2012-08-07 00:49:05]

http://codereview.chromium.org/10829019/diff/7005/chrome/browser/prefs/browse...
File chrome/browser/prefs/browser_prefs.cc (right):

http://codereview.chromium.org/10829019/diff/7005/chrome/browser/prefs/browse...
chrome/browser/prefs/browser_prefs.cc:242:
extensions::ExtensionPrefs::RegisterUserPrefs(user_prefs);
out of curiosity, where exactly is this needed?  i.e. what code that is run on
android is looking at the extension prefs?

[nilesh, 2012-08-07 16:42:16]

http://codereview.chromium.org/10829019/diff/7005/chrome/browser/prefs/browse...
File chrome/browser/prefs/browser_prefs.cc (right):

http://codereview.chromium.org/10829019/diff/7005/chrome/browser/prefs/browse...
chrome/browser/prefs/browser_prefs.cc:242:
extensions::ExtensionPrefs::RegisterUserPrefs(user_prefs);
On 2012/08/07 00:49:05, akalin wrote:
> out of curiosity, where exactly is this needed?  i.e. what code that is run on
> android is looking at the extension prefs?

extensions::ExtensionSystem::Get() causes ExtensionSystem to be created.
ExtensionSystemImpl reads extension prefs during initialization.

One such call is in
ProfileSyncComponentsFactoryImpl::ProfileSyncComponentsFactoryImpl()

[commit-bot: I haz the power, 2012-08-07 16:48:21]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/nyquist@chromium.org/10829019/7005

[commit-bot: I haz the power, 2012-08-07 16:48:22]

Failed to apply patch for chrome/browser/prefs/browser_prefs.cc:
While running patch -p1 --forward --force;
patching file chrome/browser/prefs/browser_prefs.cc
Hunk #1 FAILED at 239.
1 out of 1 hunk FAILED -- saving rejects to file
chrome/browser/prefs/browser_prefs.cc.rej

[nilesh, 2012-08-07 17:10:43]

http://codereview.chromium.org/10829019/diff/7005/chrome/browser/prefs/browse...
File chrome/browser/prefs/browser_prefs.cc (right):

http://codereview.chromium.org/10829019/diff/7005/chrome/browser/prefs/browse...
chrome/browser/prefs/browser_prefs.cc:242:
extensions::ExtensionPrefs::RegisterUserPrefs(user_prefs);
Tommy, please move this up with the default ones.

[commit-bot: I haz the power, 2012-08-08 00:12:32]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/nyquist@chromium.org/10829019/17002

[commit-bot: I haz the power, 2012-08-08 00:13:22]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/nyquist@chromium.org/10829019/17002

[commit-bot: I haz the power, 2012-08-08 02:10:03]

Change committed as 150485

[nilesh, 2012-08-08 18:04:08]

Looks like this broke AutofillManagerTest.UpdatePasswordSyncState
which depends on password sync.

CRITICAL:root:AutofillManagerTest.UpdatePasswordSyncState
CRITICAL:root:08-08 02:58:40.421 30768 30768 I chromium:
[INFO:profile_sync_service.cc(1768)] ConfigureDataTypeManager not invoked
because backend is not initialized


08-08 02:58:40.421 30768 30768 I chromium: [INFO:profile_sync_service.cc(1768)]
ConfigureDataTypeManager not invoked because backend is not initialized


08-08 02:58:40.421 30768 30768 V chromium: *** Failure in
chrome/browser/autofill/autofill_manager_unittest.cc:2945


08-08 02:58:40.421 30768 30768 V chromium: Value of:
autofill_manager_->GetSentStates().size()


08-08 02:58:40.421 30768 30768 V chromium:   Actual: 0


08-08 02:58:40.421 30768 30768 V chromium: Expected: 1u


08-08 02:58:40.421 30768 30768 V chromium: Which is: 1