Make cricket::VideoCapturer implement VideoSourceInterface

webrtc/media/base/videosourceinterface.h

+/*
+ *  Copyright (c) 2016 The WebRTC project authors. All Rights Reserved.
+ *
+ *  Use of this source code is governed by a BSD-style license
+ *  that can be found in the LICENSE file in the root of the source
+ *  tree. An additional intellectual property rights grant can be found
+ *  in the file PATENTS.  All contributing project authors may
+ *  be found in the AUTHORS file in the root of the source tree.
+ */
+
+#ifndef WEBRTC_MEDIA_BASE_VIDEOSOURCEINTERFACE_H_
+#define WEBRTC_MEDIA_BASE_VIDEOSOURCEINTERFACE_H_
+
+#include "webrtc/media/base/videosinkinterface.h"
+#include "webrtc/base/callback.h"
+
+namespace rtc {
+
+// VideoWants is a used for notifying the source of properties a video frame
+// should have when it is delivered to a certain sink.
+struct VideoWants {
+  bool operator==(const VideoWants& rh) const {
+    return rotation_applied == rh.rotation_applied;
+  }
+  bool operator!=(const VideoWants& rh) const { return !operator==(rh); }
+
+  // Tells the source whether the sink wants frames with rotation applied.
+  // By default, the rotation is applied by the source.
+  bool rotation_applied = true;
+};
+
+template <typename VideoFrameT>
+class VideoSourceInterface {
+ public:
+  virtual void AddOrUpdateSink(VideoSinkInterface<VideoFrameT>* sink,

[nisse-webrtc, 2016/02/09 14:10:54]

I don't insist on this, but I think it will be prettier if this method is
renamed to just AddSink, together with a comment saying that the same method is
used to change the wants of an already registered sink.

[pthatcher1, 2016/02/09 17:13:41]

It would be prettier.  The tradeoff is when you see code later that is:

source->AddSink(sink, wants)

for a sink that's already been added.  Reading that code, you'd be confused
unless you want to read the comment.  

So, which is more important?  Pretty or readable?  

I could go either way.

[perkj_webrtc, 2016/02/10 10:54:36]

I vote on AddOrUpdate... 

Guess what - we are more or less back on patchset 3(4) with the difference that
we don't have to call RemoveSink to update VideoWants + naming.

[pthatcher1, 2016/02/10 19:29:19]

Yeah, I realized that, too.  I guess you convinced me of the wisdom of your
ideas.  Sorry it took me so long.

+                               const VideoWants& wants) = 0;
+  // RemoveSink must guarantee that at the time the method returns,
+  // there is no current and no future calls to VideoSinkInterface::OnFrame.
+  virtual void RemoveSink(VideoSinkInterface<VideoFrameT>* sink) = 0;
+
+ protected:
+  virtual ~VideoSourceInterface() {}
+};
+
+}  // namespace rtc
+#endif  // WEBRTC_MEDIA_BASE_VIDEOSOURCEINTERFACE_H_