chrome/browser/predictors componentization part-1

chrome/browser/predictors componentization part-1

Refactor PredictorDatabase into to two separate logic entities:
  1) PredictorDatabaseService: responsible for  profile/ProfileKeyed[Service,
     factory] management.
  2) PredictorDatabase: responsible for core database and
     [AutocompleteActionPredictor,ResourcePrefetchPredictor]Tables management.


BUG=175026

[Gajen, 2013-02-08 02:42:36]

PTAL

[Shishir, 2013-02-08 02:52:11]

Can you give us some context why this is being done?

[Gajen, 2013-02-08 06:48:44]

On 2013/02/08 02:52:11, Shishir wrote:
> Can you give us some context why this is being done?

I have mentioned a preview of the plan in bug
http://code.google.com/p/chromium/issues/detail?id=175026.
Please let me know if that makes sense.

[Jói, 2013-02-08 13:02:29]

I haven't yet reviewed in detail (so no nits for now) but in general
the approach looks right.  It is similar to how we've broken up other
classes that were ProfileKeyedServices to enable componentizing their
"business logic" parts.

Gajen, in the future, it might work better to get buy-in from OWNERS
for components you are thinking about working on before sending out
the first changelist for the work.  I will wait to hear from owners of
this component that they are willing to work with you on this before
doing my review.

Shishir and Dominich, if you are willing to work with Gajen, I would
be happy to do the first round of reviews for changes in this area, to
make sure they follow conventions that have been established for
componentization so far.  One thing to note right away is that when
stuff moves to src/components, the OWNERS list is copied into the
OWNERS file for the new component.

Cheers,
Jói



On Fri, Feb 8, 2013 at 6:48 AM,  <wxjg68@motorola.com> wrote:
> On 2013/02/08 02:52:11, Shishir wrote:
>>
>> Can you give us some context why this is being done?
>
>
> I have mentioned a preview of the plan in bug
> http://code.google.com/p/chromium/issues/detail?id=175026.
> Please let me know if that makes sense.
>
> https://codereview.chromium.org/12225093/

[Gajen, 2013-02-11 19:07:01]

Shishir, My apologies for not discussing prior of landing this patch, But would
be happy to know if this still needs to be continued..???

[dominich, 2013-05-16 22:33:59]

On 2013/02/11 19:07:01, Gajen wrote:
> Shishir, My apologies for not discussing prior of landing this patch, But
would
> be happy to know if this still needs to be continued..???

Sorry for the delay - i'm not getting emails on reviews any more for some
reason.

I'm happy to see this happen.