Convert //ppapi to use std::unique_ptr

Convert //ppapi to use std::unique_ptr

BUG=554298
R=bbudge@chromium.org

Committed: https://crrev.com/ced9224fbb7866e6ea7e9da221c0b2570008bd55
Cr-Commit-Position: refs/heads/master@{#385591}

[dcheng, 2016-04-06 19:57:02]

https://codereview.chromium.org/1864293002/diff/1/chrome/browser/renderer_hos...
File chrome/browser/renderer_host/pepper/chrome_browser_pepper_host_factory.h
(right):

https://codereview.chromium.org/1864293002/diff/1/chrome/browser/renderer_hos...
chrome/browser/renderer_host/pepper/chrome_browser_pepper_host_factory.h:21:
std::unique_ptr<ppapi::host::ResourceHost> CreateResourceHost(
I ended up changing this too since it's a HostFactory override.

Guiding rules for #includes that I use:

1. Implementation and test .cc files do not need to duplicate headers included
in the 'primary' header.
2. If a class declared in a header has a parent class, the header does not need
to duplicate headers included by the parent class's header when the only
reference is for overriding methods from that parent class. Thus, in this case,
since the only reference is to override a method from HostFactory, no #include
for <memory> is added.

https://codereview.chromium.org/1864293002/diff/1/content/renderer/pepper/pep...
File content/renderer/pepper/pepper_compositor_host.h (right):

https://codereview.chromium.org/1864293002/diff/1/content/renderer/pepper/pep...
content/renderer/pepper/pepper_compositor_host.h:41: ~PepperCompositorHost()
override;
It doesn't make sense for these to be private or protected: its ancestors have
public destructors anyway.

https://codereview.chromium.org/1864293002/diff/1/content/renderer/pepper/pep...
File content/renderer/pepper/pepper_media_stream_video_track_host.h (right):

https://codereview.chromium.org/1864293002/diff/1/content/renderer/pepper/pep...
content/renderer/pepper/pepper_media_stream_video_track_host.h:43:
~PepperMediaStreamVideoTrackHost() override;
Ditto.

[bbudge, 2016-04-06 21:01:21]

I'm curious, why is base::WrapUnique used when replacing make_scoped_ptr, rather
than the std::unique_ptr ctor? It looks like WrapUnique just calls that
constructor. I didn't see this explained in the bug.

https://codereview.chromium.org/1864293002/diff/1/content/renderer/pepper/pep...
File content/renderer/pepper/pepper_compositor_host.h (right):

https://codereview.chromium.org/1864293002/diff/1/content/renderer/pepper/pep...
content/renderer/pepper/pepper_compositor_host.h:41: ~PepperCompositorHost()
override;
On 2016/04/06 19:57:02, dcheng wrote:
> It doesn't make sense for these to be private or protected: its ancestors have
> public destructors anyway.

OK.

https://codereview.chromium.org/1864293002/diff/1/content/renderer/pepper/pep...
File content/renderer/pepper/pepper_media_stream_video_track_host.h (right):

https://codereview.chromium.org/1864293002/diff/1/content/renderer/pepper/pep...
content/renderer/pepper/pepper_media_stream_video_track_host.h:43:
~PepperMediaStreamVideoTrackHost() override;
On 2016/04/06 19:57:02, dcheng wrote:
> Ditto.

OK.

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/audio_encoder_r...
File ppapi/proxy/audio_encoder_resource.h (right):

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/audio_encoder_r...
ppapi/proxy/audio_encoder_resource.h:10: #include <memory>
Move to .cc file?

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/device_enumerat...
File ppapi/proxy/device_enumeration_resource_helper.h (right):

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/device_enumerat...
ppapi/proxy/device_enumeration_resource_helper.h:9: 
#include <memory> ?

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/file_ref_resour...
File ppapi/proxy/file_ref_resource.h (right):

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/file_ref_resour...
ppapi/proxy/file_ref_resource.h:10: #include <memory>
Move to .cc file?

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/plugin_dispatch...
File ppapi/proxy/plugin_dispatcher_unittest.cc (right):

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/plugin_dispatch...
ppapi/proxy/plugin_dispatcher_unittest.cc:5: #include <memory>
Could this be eliminated?

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/plugin_resource...
File ppapi/proxy/plugin_resource_tracker_unittest.cc (right):

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/plugin_resource...
ppapi/proxy/plugin_resource_tracker_unittest.cc:7: #include <memory>
Needed?

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/ppp_messaging_p...
File ppapi/proxy/ppp_messaging_proxy.cc (right):

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/ppp_messaging_p...
ppapi/proxy/ppp_messaging_proxy.cc:7: #include <algorithm>
#include <memory> ?

It does get pulled in by the includes below, but I don't see it in
ppp_messaging_proxy.h.

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/udp_socket_filt...
File ppapi/proxy/udp_socket_filter.h (right):

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/udp_socket_filt...
ppapi/proxy/udp_socket_filter.h:10: 
#include <memory> ?
I don't see it included in any superclass.

https://codereview.chromium.org/1864293002/diff/1/ppapi/shared_impl/media_str...
File ppapi/shared_impl/media_stream_buffer_manager_unittest.cc (right):

https://codereview.chromium.org/1864293002/diff/1/ppapi/shared_impl/media_str...
ppapi/shared_impl/media_stream_buffer_manager_unittest.cc:9: #include <memory>
Is included by media_stream_buffer_manager.h

https://codereview.chromium.org/1864293002/diff/1/ppapi/thunk/ppb_instance_api.h
File ppapi/thunk/ppb_instance_api.h (right):

https://codereview.chromium.org/1864293002/diff/1/ppapi/thunk/ppb_instance_ap...
ppapi/thunk/ppb_instance_api.h:10: #include <memory>
Is this needed here?

[dcheng, 2016-04-06 21:18:33]

On 2016/04/06 at 21:01:21, bbudge wrote:
> I'm curious, why is base::WrapUnique used when replacing make_scoped_ptr,
rather than the std::unique_ptr ctor? It looks like WrapUnique just calls that
constructor. I didn't see this explained in the bug.
> 

It deduces the template type from the argument. So instead of writing:
  auto x = std::unique_ptr<MyVeryLongTypeName>(new MyVeryLongTypeName);

one can just write:
  auto x = base::WrapUnique(new MyVeryLongTypeName);

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/audio_encoder_r...
File ppapi/proxy/audio_encoder_resource.h (right):

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/audio_encoder_r...
ppapi/proxy/audio_encoder_resource.h:10: #include <memory>
On 2016/04/06 at 21:01:20, bbudge wrote:
> Move to .cc file?

Deleted since it's not used. Note that at the end of the day, the main goal of
these CLs is just about migrating existing code since there are still 60000+
references to convert. If incidental cleanup happens, that is nice, but it is
explicitly not a primary goal to go out of my way to clean up unused headers,
fix IWYU violations that don't break the build, etc.

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/device_enumerat...
File ppapi/proxy/device_enumeration_resource_helper.h (right):

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/device_enumerat...
ppapi/proxy/device_enumeration_resource_helper.h:9: 
On 2016/04/06 at 21:01:20, bbudge wrote:
> #include <memory> ?

Done.

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/file_ref_resour...
File ppapi/proxy/file_ref_resource.h (right):

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/file_ref_resour...
ppapi/proxy/file_ref_resource.h:10: #include <memory>
On 2016/04/06 at 21:01:20, bbudge wrote:
> Move to .cc file?

Deleted since it's not necessary.

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/plugin_dispatch...
File ppapi/proxy/plugin_dispatcher_unittest.cc (right):

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/plugin_dispatch...
ppapi/proxy/plugin_dispatcher_unittest.cc:5: #include <memory>
On 2016/04/06 at 21:01:20, bbudge wrote:
> Could this be eliminated?

Done.

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/plugin_resource...
File ppapi/proxy/plugin_resource_tracker_unittest.cc (right):

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/plugin_resource...
ppapi/proxy/plugin_resource_tracker_unittest.cc:7: #include <memory>
On 2016/04/06 at 21:01:20, bbudge wrote:
> Needed?

Done.

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/ppp_messaging_p...
File ppapi/proxy/ppp_messaging_proxy.cc (right):

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/ppp_messaging_p...
ppapi/proxy/ppp_messaging_proxy.cc:7: #include <algorithm>
On 2016/04/06 at 21:01:20, bbudge wrote:
> #include <memory> ?
> 
> It does get pulled in by the includes below, but I don't see it in
ppp_messaging_proxy.h.

I've changed this to just use base::WrapUnique.

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/udp_socket_filt...
File ppapi/proxy/udp_socket_filter.h (right):

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/udp_socket_filt...
ppapi/proxy/udp_socket_filter.h:10: 
On 2016/04/06 at 21:01:20, bbudge wrote:
> #include <memory> ?
> I don't see it included in any superclass.

Done.

https://codereview.chromium.org/1864293002/diff/1/ppapi/shared_impl/media_str...
File ppapi/shared_impl/media_stream_buffer_manager_unittest.cc (right):

https://codereview.chromium.org/1864293002/diff/1/ppapi/shared_impl/media_str...
ppapi/shared_impl/media_stream_buffer_manager_unittest.cc:9: #include <memory>
On 2016/04/06 at 21:01:20, bbudge wrote:
> Is included by media_stream_buffer_manager.h

Done.

https://codereview.chromium.org/1864293002/diff/1/ppapi/thunk/ppb_instance_api.h
File ppapi/thunk/ppb_instance_api.h (right):

https://codereview.chromium.org/1864293002/diff/1/ppapi/thunk/ppb_instance_ap...
ppapi/thunk/ppb_instance_api.h:10: #include <memory>
On 2016/04/06 at 21:01:21, bbudge wrote:
> Is this needed here?

Done.

[bbudge, 2016-04-06 21:23:44]

LGTM Thanks for doing this!

[piman, 2016-04-06 21:25:02]

piman@chromium.org changed reviewers:
+ piman@chromium.org

[piman, 2016-04-06 21:25:03]

LGTM+nits

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/raw_var_data.cc
File ppapi/proxy/raw_var_data.cc (right):

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/raw_var_data.cc...
ppapi/proxy/raw_var_data.cc:113: return std::unique_ptr<RawVarDataGraph>();
nit: nullptr, here and below?

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/raw_var_data.cc...
ppapi/proxy/raw_var_data.cc:194: return std::unique_ptr<RawVarDataGraph>();
(also here and below)

[dcheng, 2016-04-06 21:29:04]

The CQ bit was checked by dcheng@chromium.org

[dcheng, 2016-04-06 21:29:06]

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/raw_var_data.cc
File ppapi/proxy/raw_var_data.cc (right):

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/raw_var_data.cc...
ppapi/proxy/raw_var_data.cc:113: return std::unique_ptr<RawVarDataGraph>();
On 2016/04/06 at 21:25:03, piman wrote:
> nit: nullptr, here and below?

Done.

https://codereview.chromium.org/1864293002/diff/1/ppapi/proxy/raw_var_data.cc...
ppapi/proxy/raw_var_data.cc:194: return std::unique_ptr<RawVarDataGraph>();
On 2016/04/06 at 21:25:03, piman wrote:
> (also here and below)

FWIW, I'm cautiously optimistic that clang-tidy will help out with this in the
future, so I'm not too worried about fixing all of these.

[dcheng, 2016-04-06 21:29:06]

The patchset sent to the CQ was uploaded after l-g-t-m from bbudge@chromium.org,
piman@chromium.org
Link to the patchset: https://codereview.chromium.org/1864293002/#ps40001
(title: "more nullptr")

[commit-bot: I haz the power, 2016-04-06 21:29:54]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1864293002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1864293002/40001

[commit-bot: I haz the power, 2016-04-07 00:00:21]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2016-04-07 00:01:44]

Description was changed from

==========
Convert //ppapi to use std::unique_ptr

BUG=554298
R=bbudge@chromium.org
==========

to

==========
Convert //ppapi to use std::unique_ptr

BUG=554298
R=bbudge@chromium.org

Committed: https://crrev.com/ced9224fbb7866e6ea7e9da221c0b2570008bd55
Cr-Commit-Position: refs/heads/master@{#385591}
==========

[commit-bot: I haz the power, 2016-04-07 00:01:45]

Patchset 3 (id:??) landed as
https://crrev.com/ced9224fbb7866e6ea7e9da221c0b2570008bd55
Cr-Commit-Position: refs/heads/master@{#385591}