Change rtc::VideoSinkWants to have target and a max pixel count

webrtc/video/vie_encoder.h

Index: webrtc/video/vie_encoder.h
diff --git a/webrtc/video/vie_encoder.h b/webrtc/video/vie_encoder.h
index 0c564724d4b96e49d9bc9d3e9718a3d0d93efde1..6d22f07693e190ab1274afc95c32bf535f6145ef 100644
--- a/webrtc/video/vie_encoder.h
+++ b/webrtc/video/vie_encoder.h
@@ -218,10 +218,16 @@ class ViEEncoder : public rtc::VideoSinkInterface<VideoFrame>,
   VideoSendStream::DegradationPreference degradation_preference_
       ACCESS_ON(&encoder_queue_);
 
-  // Pixel count last time the resolution was requested to be changed down.
-  rtc::Optional<int> max_pixel_count_ ACCESS_ON(&encoder_queue_);
-  // Pixel count last time the resolution was requested to be changed up.
-  rtc::Optional<int> max_pixel_count_step_up_ ACCESS_ON(&encoder_queue_);
+  struct AdaptationRequest {
+    // The pixel count produced by the source at the time of the adaptation.
+    int input_pixel_count_;
+    // True if request was to adapt down, false otherwise.
+    bool adapt_down_;

[kthelgason, 2017/02/06 12:13:21]

Can we make this an enum instead? IMO having explicit direction = kDown or kUp
is clearer than a boolean.

[sprang_webrtc, 2017/02/06 13:18:07]

Done.

+  };
+  // Stores a snapshot of the last adaptation request triggered by an AdaptUp
+  // or AdaptDown signal.
+  rtc::Optional<AdaptationRequest> last_adaptation_request_
+      ACCESS_ON(&encoder_queue_);
 
   rtc::RaceChecker incoming_frame_race_checker_
       GUARDED_BY(incoming_frame_race_checker_);