Replace the remaining scoped_ptr with unique_ptr in webrtc/modules/

Replace the remaining scoped_ptr with unique_ptr in webrtc/modules/

(This is a re-land of https://codereview.webrtc.org/1921233002, which
got reverted for breaking Chromium.)

BUG=webrtc:5520
Committed: https://crrev.com/84be511ac0c047b5cf7b2797fe2de38ba714c656
Cr-Commit-Position: refs/heads/master@{#12522}

[kwiberg-webrtc, 2016-04-27 00:04:27]

kwiberg@webrtc.org changed reviewers:
+ tommi@webrtc.org

[kwiberg-webrtc, 2016-04-27 00:04:28]

I'm thinking it was a mistake to remove #include "scoped_ptr.h" from the header
files---if users rely on an interface that used to make use of scoped_ptr to
include scoped_ptr.h for them (which it had to do), then removing that include
before all users have replaced scoped_ptr with unique_ptr will cause breakage.
(This was in fact why the original attempt to land this CL failed.)

Instead, I'll leave the #includes in all .h files, and remove them at the same
time as I remove the scoped_ptr.h file itself.

[tommi, 2016-04-27 07:32:56]

On 2016/04/27 00:04:28, kwiberg-webrtc wrote:
> I'm thinking it was a mistake to remove #include "scoped_ptr.h" from the
header
> files---if users rely on an interface that used to make use of scoped_ptr to
> include scoped_ptr.h for them (which it had to do), then removing that include
> before all users have replaced scoped_ptr with unique_ptr will cause breakage.
> (This was in fact why the original attempt to land this CL failed.)
> 
> Instead, I'll leave the #includes in all .h files, and remove them at the same
> time as I remove the scoped_ptr.h file itself.

Sounds reasonable. lgtm.
btw, is the VoiceActivityIndicatorReturns1WithSpeechOn failure flake or
something that started happening recently?

[kwiberg-webrtc, 2016-04-27 07:49:44]

On 2016/04/27 07:32:56, tommi-webrtc wrote:
> On 2016/04/27 00:04:28, kwiberg-webrtc wrote:
> > I'm thinking it was a mistake to remove #include "scoped_ptr.h" from the
> header
> > files---if users rely on an interface that used to make use of scoped_ptr to
> > include scoped_ptr.h for them (which it had to do), then removing that
include
> > before all users have replaced scoped_ptr with unique_ptr will cause
breakage.
> > (This was in fact why the original attempt to land this CL failed.)
> > 
> > Instead, I'll leave the #includes in all .h files, and remove them at the
same
> > time as I remove the scoped_ptr.h file itself.
> 
> Sounds reasonable. lgtm.
> btw, is the VoiceActivityIndicatorReturns1WithSpeechOn failure flake or
> something that started happening recently?

Pretty sure it's a flake. A change like this one doesn't seem likely to cause
that sort of trouble (especially since it didn't when we landed it the first
time).

[kwiberg-webrtc, 2016-04-27 07:49:54]

The CQ bit was checked by kwiberg@webrtc.org

[commit-bot: I haz the power, 2016-04-27 07:50:00]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1923133002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1923133002/20001

[commit-bot: I haz the power, 2016-04-27 08:20:03]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-04-27 08:20:16]

Description was changed from

==========
Replace the remaining scoped_ptr with unique_ptr in webrtc/modules/

(This is a re-land of https://codereview.webrtc.org/1921233002, which
got reverted for breaking Chromium.)

BUG=webrtc:5520
==========

to

==========
Replace the remaining scoped_ptr with unique_ptr in webrtc/modules/

(This is a re-land of https://codereview.webrtc.org/1921233002, which
got reverted for breaking Chromium.)

BUG=webrtc:5520

Committed: https://crrev.com/84be511ac0c047b5cf7b2797fe2de38ba714c656
Cr-Commit-Position: refs/heads/master@{#12522}
==========

[commit-bot: I haz the power, 2016-04-27 08:20:18]

Patchset 2 (id:??) landed as
https://crrev.com/84be511ac0c047b5cf7b2797fe2de38ba714c656
Cr-Commit-Position: refs/heads/master@{#12522}