Delete method webrtc::VideoFrame::Reset.

Delete method webrtc::VideoFrame::Reset.

Mainly affects VideoCaptureInput.

BUG=webrtc:5682
Committed: https://crrev.com/c7fe3c27ebdf4f33934c17a938841a3aa98ef5c6
Cr-Commit-Position: refs/heads/master@{#12439}

[nisse-webrtc, 2016-04-13 12:15:27]

nisse@webrtc.org changed reviewers:
+ pbos@webrtc.org, perkj@webrtc.org

[nisse-webrtc, 2016-04-13 12:15:27]

Deletes the Reset method. A step towards ensuring that the video_frame_buffer
method never can return NULL.

Adds a little allocation overhead (of a small object) in VideoCaptureInput.

[pbos-webrtc, 2016-04-13 12:31:52]

lgtm, maybe getting rid of this heap allocation is a later problem

https://codereview.webrtc.org/1889443002/diff/1/webrtc/video/video_capture_in...
File webrtc/video/video_capture_input.cc (right):

https://codereview.webrtc.org/1889443002/diff/1/webrtc/video/video_capture_in...
webrtc/video/video_capture_input.cc:87: captured_frame_.reset(new VideoFrame);
Consider getting rid of this heap allocation.

[nisse-webrtc, 2016-04-13 12:45:26]

https://codereview.webrtc.org/1889443002/diff/1/webrtc/video/video_capture_in...
File webrtc/video/video_capture_input.cc (right):

https://codereview.webrtc.org/1889443002/diff/1/webrtc/video/video_capture_in...
webrtc/video/video_capture_input.cc:87: captured_frame_.reset(new VideoFrame);
On 2016/04/13 12:31:52, pbos-webrtc wrote:
> Consider getting rid of this heap allocation.

It's kind-of tricky. 

1. I aim to make sure that a VideoFrame always has a valid pointer to an
underlying VideoFrameBuffer. 

2. The GetVideoFrame method here wants to drop that reference.

The only straightforward way to get both is to destroy the VideoFrame object.

A better way might be to change GetVideoFrame return a unique_ptr<VideoFrame>?
Then we can also eliminate the (shallow) copy.

[pbos-webrtc, 2016-04-13 13:44:35]

On 2016/04/13 12:45:26, nisse-webrtc wrote:
>
https://codereview.webrtc.org/1889443002/diff/1/webrtc/video/video_capture_in...
> File webrtc/video/video_capture_input.cc (right):
> 
>
https://codereview.webrtc.org/1889443002/diff/1/webrtc/video/video_capture_in...
> webrtc/video/video_capture_input.cc:87: captured_frame_.reset(new VideoFrame);
> On 2016/04/13 12:31:52, pbos-webrtc wrote:
> > Consider getting rid of this heap allocation.
> 
> It's kind-of tricky. 
> 
> 1. I aim to make sure that a VideoFrame always has a valid pointer to an
> underlying VideoFrameBuffer. 
> 
> 2. The GetVideoFrame method here wants to drop that reference.
> 
> The only straightforward way to get both is to destroy the VideoFrame object.
> 
> A better way might be to change GetVideoFrame return a unique_ptr<VideoFrame>?
> Then we can also eliminate the (shallow) copy.

Sounds good to me. :)

[nisse-webrtc, 2016-04-20 09:11:32]

The CQ bit was checked by nisse@webrtc.org

[commit-bot: I haz the power, 2016-04-20 09:11:37]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1889443002/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1889443002/1

[commit-bot: I haz the power, 2016-04-20 10:25:40]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2016-04-20 10:25:45]

Description was changed from

==========
Delete method webrtc::VideoFrame::Reset.

Mainly affects VideoCaptureInput.

BUG=webrtc:5682
==========

to

==========
Delete method webrtc::VideoFrame::Reset.

Mainly affects VideoCaptureInput.

BUG=webrtc:5682

Committed: https://crrev.com/c7fe3c27ebdf4f33934c17a938841a3aa98ef5c6
Cr-Commit-Position: refs/heads/master@{#12439}
==========

[commit-bot: I haz the power, 2016-04-20 10:25:45]

Patchset 1 (id:??) landed as
https://crrev.com/c7fe3c27ebdf4f33934c17a938841a3aa98ef5c6
Cr-Commit-Position: refs/heads/master@{#12439}