Get rid of FileThreadDeserializer.

Get rid of FileThreadDeserializer.

Replace it with modern threading constructs:
 - PostTaskAndReplyWithResult gets rid of most of the logic FileThreadDeserializer was implementing.
 - The remainder logic didn't require any class state so it was moved to anonymous methods.

Also declare JsonPrefStore explicitly NonThreadSafe (the only actions outside the UI thread should happen by posting anonymous tasks to the |sequenced_task_runner_|).

This is a stepping stone in cleaning up JsonPrefStore to eventually get rid of PrefStore's ref-counting scheme.

BUG=393081
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=284754

[gab, 2014-07-22 16:03:24]

Bernhard PTAL :-)!

Cheers,
Gab

[Bernhard Bauer, 2014-07-22 16:22:00]

Awesumz!!!111oneone

https://codereview.chromium.org/400673008/diff/40001/base/prefs/json_pref_sto...
File base/prefs/json_pref_store.cc (right):

https://codereview.chromium.org/400673008/diff/40001/base/prefs/json_pref_sto...
base/prefs/json_pref_store.cc:116: void PostAsyncReadTask(
Could you inline this method?

https://codereview.chromium.org/400673008/diff/40001/base/prefs/json_pref_sto...
File base/prefs/json_pref_store.h (right):

https://codereview.chromium.org/400673008/diff/40001/base/prefs/json_pref_sto...
base/prefs/json_pref_store.h:41: struct ReadResult;
I think the struct can be private?

[gab, 2014-07-22 16:47:43]

https://codereview.chromium.org/400673008/diff/40001/base/prefs/json_pref_sto...
File base/prefs/json_pref_store.cc (right):

https://codereview.chromium.org/400673008/diff/40001/base/prefs/json_pref_sto...
base/prefs/json_pref_store.cc:116: void PostAsyncReadTask(
On 2014/07/22 16:22:00, Bernhard Bauer wrote:
> Could you inline this method?

Indeed :-)!

https://codereview.chromium.org/400673008/diff/40001/base/prefs/json_pref_sto...
File base/prefs/json_pref_store.h (right):

https://codereview.chromium.org/400673008/diff/40001/base/prefs/json_pref_sto...
base/prefs/json_pref_store.h:41: struct ReadResult;
On 2014/07/22 16:22:00, Bernhard Bauer wrote:
> I think the struct can be private?

No because it's used by the anonymous methods in the impl :-(.

I agree that this is weird, but I couldn't find a cleaner way...

[Bernhard Bauer, 2014-07-22 16:56:35]

lgtm

https://codereview.chromium.org/400673008/diff/40001/base/prefs/json_pref_sto...
File base/prefs/json_pref_store.h (right):

https://codereview.chromium.org/400673008/diff/40001/base/prefs/json_pref_sto...
base/prefs/json_pref_store.h:41: struct ReadResult;
On 2014/07/22 16:47:43, gab wrote:
> On 2014/07/22 16:22:00, Bernhard Bauer wrote:
> > I think the struct can be private?
> 
> No because it's used by the anonymous methods in the impl :-(.
> 
> I agree that this is weird, but I couldn't find a cleaner way...

Yeah... one possibility would be to make ReadPrefsFromDisk a private (static)
class method, but I don't think that'd be necessarily cleaner.

[gab, 2014-07-22 17:23:05]

Thanks :-)!

https://codereview.chromium.org/400673008/diff/40001/base/prefs/json_pref_sto...
File base/prefs/json_pref_store.h (right):

https://codereview.chromium.org/400673008/diff/40001/base/prefs/json_pref_sto...
base/prefs/json_pref_store.h:41: struct ReadResult;
On 2014/07/22 16:56:35, Bernhard Bauer wrote:
> On 2014/07/22 16:47:43, gab wrote:
> > On 2014/07/22 16:22:00, Bernhard Bauer wrote:
> > > I think the struct can be private?
> > 
> > No because it's used by the anonymous methods in the impl :-(.
> > 
> > I agree that this is weird, but I couldn't find a cleaner way...
> 
> Yeah... one possibility would be to make ReadPrefsFromDisk a private (static)
> class method, but I don't think that'd be necessarily cleaner.

Yea, I debated both and figured the public fwd-decl was a minor nuisance to keep
the anonymous methods.

[gab, 2014-07-22 17:23:09]

The CQ bit was checked by gab@chromium.org

[commit-bot: I haz the power, 2014-07-22 17:24:58]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/gab@chromium.org/400673008/60001

[commit-bot: I haz the power, 2014-07-22 19:30:25]

Change committed as 284754