Move the bits of Prefs where production code has only trivially easy

Move the bits of Prefs where production code has only trivially easy
to break dependencies back to Chrome.

This leaves the following in chrome/ for now, which will move later:

- PrefService, to move once sync dependencies and dependencies on
  Chrome-specific pref stores are externalized.

- PrefNotifierImpl and PrefObserverMock, to move once
  NotificationSource dependency is broken.

- PrefValueStore, to move once PrefModelAssociatior dependency is
  externalized.

- ScopedUserPrefUpdate, to move once PrefService moves.

TBR=ben@chromium.org
BUG=155525
Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=163354

[Jói, 2012-10-22 14:38:12]

Hi Mattias,

As before, the interesting files are:
DEPS
base/prefs/DEPS
base/prefs/README
chrome/DEPS
chrome/*.gypi

Everything else is the result of using tools/refactor/move_file.py for moving
each moved file, followed by [ tools/git/for-all-touched-files.py -c
"tools/sort-headers.py [[FILENAME]]" ] to sort headers in all touched files.

Each A+ file should have a matching D, and is modified only to update include
paths and change header guards.  Each M file should just be an update to include
paths.

The plan for building these is also as before, until the move is done and all
dependencies are broken (so that base/prefs/ contains all of the
non-Chrome-specific Prefs implementation, and only depends on base/), these
sources continue to be built from chrome/*.gypi, but will move to a separate
'base_prefs' target in base/base.gyp at the end.

Cheers,
Jói

[Jói, 2012-10-22 14:42:38]

+ben@ for .gypi changes, but I plan to TBR once I have Mattias's LGTM since this
is as we had discussed and this change needs to land quickly, before merge
problems creep in (it happens quickly with ~150 files with changed includes).

Cheers,
Jói

[Mattias Nissler (ping if slow), 2012-10-22 15:03:11]

LGTM with with nits.

https://codereview.chromium.org/11243002/diff/5001/base/prefs/README
File base/prefs/README (right):

https://codereview.chromium.org/11243002/diff/5001/base/prefs/README#newcode1
base/prefs/README:1: At the moment, src/base/prefs has a bunch of dependencies
back to
Maybe add a first line saying what this is, something along the lines of
"General-purpose key-value store for application preferences" or something?

https://codereview.chromium.org/11243002/diff/5001/base/prefs/public/pref_ser...
File base/prefs/public/pref_service_base.h (right):

https://codereview.chromium.org/11243002/diff/5001/base/prefs/public/pref_ser...
base/prefs/public/pref_service_base.h:34: class PrefServiceBase {
Just to verify: I guess the plan is to reconcile PrefService and PrefServiceBase
after the PrefService refactor?

https://codereview.chromium.org/11243002/diff/5001/chrome/browser/chromeos/se...
File chrome/browser/chromeos/settings/device_settings_provider.h (right):

https://codereview.chromium.org/11243002/diff/5001/chrome/browser/chromeos/se...
chrome/browser/chromeos/settings/device_settings_provider.h:18: #include
"base/prefs/pref_value_map.h"
I guess I should have realized that one include is enough when I put this in...
Care to fix? ;)

[Jói, 2012-10-22 15:31:57]

Thanks, addressed your comments, will CQ now.

Cheers,
Jói

https://codereview.chromium.org/11243002/diff/5001/base/prefs/README
File base/prefs/README (right):

https://codereview.chromium.org/11243002/diff/5001/base/prefs/README#newcode1
base/prefs/README:1: At the moment, src/base/prefs has a bunch of dependencies
back to
On 2012/10/22 15:03:12, Mattias Nissler wrote:
> Maybe add a first line saying what this is, something along the lines of
> "General-purpose key-value store for application preferences" or something?

Done.

https://codereview.chromium.org/11243002/diff/5001/base/prefs/public/pref_ser...
File base/prefs/public/pref_service_base.h (right):

https://codereview.chromium.org/11243002/diff/5001/base/prefs/public/pref_ser...
base/prefs/public/pref_service_base.h:34: class PrefServiceBase {
On 2012/10/22 15:03:12, Mattias Nissler wrote:
> Just to verify: I guess the plan is to reconcile PrefService and
PrefServiceBase
> after the PrefService refactor?

Yes, once dependencies back to chrome/ have been broken.

https://codereview.chromium.org/11243002/diff/5001/chrome/browser/chromeos/se...
File chrome/browser/chromeos/settings/device_settings_provider.h (right):

https://codereview.chromium.org/11243002/diff/5001/chrome/browser/chromeos/se...
chrome/browser/chromeos/settings/device_settings_provider.h:18: #include
"base/prefs/pref_value_map.h"
On 2012/10/22 15:03:12, Mattias Nissler wrote:
> I guess I should have realized that one include is enough when I put this
in...
> Care to fix? ;)

Done.

[commit-bot: I haz the power, 2012-10-22 15:32:50]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/joi@chromium.org/11243002/9001

[Ben Goodger (Google), 2012-10-22 16:07:34]

Very cool!


On Mon, Oct 22, 2012 at 8:32 AM, <commit-bot@chromium.org> wrote:

> CQ is trying da patch. Follow status at
>
https://chromium-status.**appspot.com/cq/joi@chromium.**org/11243002/9001<htt...
>
>
>
https://chromiumcodereview.**appspot.com/11243002/<https://chromiumcodereview...
>

[commit-bot: I haz the power, 2012-10-22 20:02:11]

Change committed as 163354