cc: Remove ScopedPtrDeque.

cc: Remove ScopedPtrDeque.

This patch removes ScopedPtrDeque and replaces it with a deque of
scoped_ptrs. Note that this also adds a helper container_util file
that contains TakeBack and TakeFront helpers.

R=danakj
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel

Committed: https://crrev.com/36d6abffa95c4ea211d0028c3be07adffef7d80e
Cr-Commit-Position: refs/heads/master@{#359724}

[vmpstr, 2015-11-12 08:56:01]

https://codereview.chromium.org/1441613002/diff/1/cc/base/container_util.h
File cc/base/container_util.h (right):

https://codereview.chromium.org/1441613002/diff/1/cc/base/container_util.h#ne...
cc/base/container_util.h:14: typename Container::value_type element =
std::move(container->front());
Note that this is using std::move, to allow for non scoped ptr types as well. I
can change this to container->front().Pass() instead if that's better for now

[danakj, 2015-11-13 19:54:02]

https://codereview.chromium.org/1441613002/diff/1/cc/base/container_util.h
File cc/base/container_util.h (right):

https://codereview.chromium.org/1441613002/diff/1/cc/base/container_util.h#ne...
cc/base/container_util.h:1: // Copyright 2015 The Chromium Authors. All rights
reserved.
Can you move these to base/stl_util.h?

https://codereview.chromium.org/1441613002/diff/1/cc/base/container_util.h#ne...
cc/base/container_util.h:12: template <typename Container>
please leave comments on these explaining what they do (explain reliance on move
cons/assignment etc

https://codereview.chromium.org/1441613002/diff/1/cc/base/container_util.h#ne...
cc/base/container_util.h:14: typename Container::value_type element =
std::move(container->front());
On 2015/11/12 08:56:01, vmpstr wrote:
> Note that this is using std::move, to allow for non scoped ptr types as well.
I
> can change this to container->front().Pass() instead if that's better for now

Ya, you should use .Pass() and a TODO if you want to land this before we allow
move(). Up to you. The follow-up to switch back to move() would be easy (you can
TBR me on it if you do that).

https://codereview.chromium.org/1441613002/diff/1/cc/resources/resource_provi...
File cc/resources/resource_provider_unittest.cc (right):

https://codereview.chromium.org/1441613002/diff/1/cc/resources/resource_provi...
cc/resources/resource_provider_unittest.cc:360: typedef
std::deque<scoped_ptr<PendingProduceTexture>>
nit: using instead of typedef

but probably kill the typedef and use auto for iterators instead

[vmpstr, 2015-11-13 23:15:38]

ptal

https://codereview.chromium.org/1441613002/diff/1/cc/base/container_util.h
File cc/base/container_util.h (right):

https://codereview.chromium.org/1441613002/diff/1/cc/base/container_util.h#ne...
cc/base/container_util.h:1: // Copyright 2015 The Chromium Authors. All rights
reserved.
On 2015/11/13 19:54:02, danakj (behind on reviews) wrote:
> Can you move these to base/stl_util.h?

I was under the impression that code should live close to where its used. Also,
if we do switch this to std::move, this would introduce a library feature into
base header file which might break crashpad (at least that's my impression). 

I'd prefer to keep it here, but I can put it in base/ if you insist.

https://codereview.chromium.org/1441613002/diff/1/cc/base/container_util.h#ne...
cc/base/container_util.h:12: template <typename Container>
On 2015/11/13 19:54:02, danakj (behind on reviews) wrote:
> please leave comments on these explaining what they do (explain reliance on
move
> cons/assignment etc

Done.

https://codereview.chromium.org/1441613002/diff/1/cc/base/container_util.h#ne...
cc/base/container_util.h:14: typename Container::value_type element =
std::move(container->front());
On 2015/11/13 19:54:02, danakj (behind on reviews) wrote:
> On 2015/11/12 08:56:01, vmpstr wrote:
> > Note that this is using std::move, to allow for non scoped ptr types as
well.
> I
> > can change this to container->front().Pass() instead if that's better for
now
> 
> Ya, you should use .Pass() and a TODO if you want to land this before we allow
> move(). Up to you. The follow-up to switch back to move() would be easy (you
can
> TBR me on it if you do that).

Done.

https://codereview.chromium.org/1441613002/diff/1/cc/resources/resource_provi...
File cc/resources/resource_provider_unittest.cc (right):

https://codereview.chromium.org/1441613002/diff/1/cc/resources/resource_provi...
cc/resources/resource_provider_unittest.cc:360: typedef
std::deque<scoped_ptr<PendingProduceTexture>>
On 2015/11/13 19:54:02, danakj (behind on reviews) wrote:
> nit: using instead of typedef
> 
> but probably kill the typedef and use auto for iterators instead

Done.

[danakj, 2015-11-13 23:41:06]

LGTM

https://codereview.chromium.org/1441613002/diff/1/cc/base/container_util.h
File cc/base/container_util.h (right):

https://codereview.chromium.org/1441613002/diff/1/cc/base/container_util.h#ne...
cc/base/container_util.h:1: // Copyright 2015 The Chromium Authors. All rights
reserved.
On 2015/11/13 23:15:38, vmpstr wrote:
> On 2015/11/13 19:54:02, danakj (behind on reviews) wrote:
> > Can you move these to base/stl_util.h?
> 
> I was under the impression that code should live close to where its used.
Also,
> if we do switch this to std::move, this would introduce a library feature into
> base header file which might break crashpad (at least that's my impression). 
> 
> I'd prefer to keep it here, but I can put it in base/ if you insist.

Crashpad will define new things in its own compat library. Calling new C++11
methods on existing classes is the problematic thing.

https://codereview.chromium.org/1441613002/diff/20001/cc/resources/resource_p...
File cc/resources/resource_provider_unittest.cc (right):

https://codereview.chromium.org/1441613002/diff/20001/cc/resources/resource_p...
cc/resources/resource_provider_unittest.cc:185: for (auto it =
pending_produce_textures_.begin();
um, actually not sure why not use range-based for loop here.

[vmpstr, 2015-11-13 23:50:49]

https://codereview.chromium.org/1441613002/diff/1/cc/base/container_util.h
File cc/base/container_util.h (right):

https://codereview.chromium.org/1441613002/diff/1/cc/base/container_util.h#ne...
cc/base/container_util.h:1: // Copyright 2015 The Chromium Authors. All rights
reserved.
On 2015/11/13 23:41:06, danakj (behind on reviews) wrote:
> On 2015/11/13 23:15:38, vmpstr wrote:
> > On 2015/11/13 19:54:02, danakj (behind on reviews) wrote:
> > > Can you move these to base/stl_util.h?
> > 
> > I was under the impression that code should live close to where its used.
> Also,
> > if we do switch this to std::move, this would introduce a library feature
into
> > base header file which might break crashpad (at least that's my impression).

> > 
> > I'd prefer to keep it here, but I can put it in base/ if you insist.
> 
> Crashpad will define new things in its own compat library. Calling new C++11
> methods on existing classes is the problematic thing.

Ah OK. That makes sense.

https://codereview.chromium.org/1441613002/diff/20001/cc/resources/resource_p...
File cc/resources/resource_provider_unittest.cc (right):

https://codereview.chromium.org/1441613002/diff/20001/cc/resources/resource_p...
cc/resources/resource_provider_unittest.cc:185: for (auto it =
pending_produce_textures_.begin();
On 2015/11/13 23:41:06, danakj (behind on reviews) wrote:
> um, actually not sure why not use range-based for loop here.

Yep, changed it to that. A lot of these changes were mechanical, so I didn't
necessarily consider better constructs.

[vmpstr, 2015-11-13 23:50:59]

The CQ bit was checked by vmpstr@chromium.org

[vmpstr, 2015-11-13 23:51:00]

The patchset sent to the CQ was uploaded after l-g-t-m from danakj@chromium.org
Link to the patchset: https://codereview.chromium.org/1441613002/#ps40001
(title: " ")

[commit-bot: I haz the power, 2015-11-13 23:51:53]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1441613002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1441613002/40001

[commit-bot: I haz the power, 2015-11-14 01:37:18]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2015-11-14 01:38:15]

Patchset 3 (id:??) landed as
https://crrev.com/36d6abffa95c4ea211d0028c3be07adffef7d80e
Cr-Commit-Position: refs/heads/master@{#359724}