AndroidStudio MotionEvent: detect only ACTION_UP on a second view, starting ACTION_DOWN on first view

Question

Solved it by using emandt's suggestion. My personal Solution added below.

I'm using Android Studio for this.

I searched for solutions but couldn't find anything resembling this.

I want to know on which ImageView an UP action occurs while starting the DOWN action on a different ImageView (to eventually be able to drag one image over the other and make it snap to the same position by getting the position of the image I dragged over).

My example has two ImageViews with the id imageView (left) and imageView2(right). In my example I'm not dragging anything yet, I just want to touch the left image, see "Action was down" in the log and lift the finger over the right image showing "Action was up2".

I don't know if this is easily possible.

As far as I can tell from testing, the MotionEvent.ACTION_UP only fires for an ImageView when you also pressed down on it beforehand. So when I release on top of imageView2 it only shows "Action was up" from the left image.

I wondered if it was possible by playing with return false, since the return value tells if an ActionEvent is consumed so I thought if the UP event of imageView returns false, maybe it does trigger the UP event of imageView2 but no. (Either complete misunderstanding on my part or it doesn't recognise UP on the second because it didn't start with a DOWN and MotionEvents probably always have to start with a DOWN).

public class MainActivity extends Activity {
ImageView imageView;
ImageView imageView2;
String DEBUG_TAG = "action";

@Override
protected void onCreate(Bundle savedInstanceState) {
    super.onCreate(savedInstanceState);
    setContentView(R.layout.activity_main);
    imageView = findViewById(R.id.imageView);
    imageView2 = findViewById(R.id.imageView2);

    imageView.setOnTouchListener(new View.OnTouchListener() {
        public boolean onTouch(View v, MotionEvent event) {
            //int action = MotionEventCompat.getActionMasked(event);
            int action = event.getActionMasked();

            switch(action) {
                case (MotionEvent.ACTION_DOWN) :
                    Log.d(DEBUG_TAG,"Action was DOWN"+v.toString());
                    return true;
                case (MotionEvent.ACTION_MOVE) :
                    //Log.d(DEBUG_TAG,"Action was MOVE");
                    return true;
                case (MotionEvent.ACTION_UP) :
                    Log.d(DEBUG_TAG,"Action was UP"+v.toString());
                    return false;

                default :
                    //return true;
            }
            return true;
        }
    });

    imageView2.setOnTouchListener(new View.OnTouchListener() {
        public boolean onTouch(View v, MotionEvent event) {
            //int action = MotionEventCompat.getActionMasked(event);
            int action = event.getActionMasked();

            switch(action) {
                case (MotionEvent.ACTION_DOWN) :
                    Log.d(DEBUG_TAG,"Action was DOWN2"+v.toString());
                    return true;
                case (MotionEvent.ACTION_MOVE) :
                    //Log.d(DEBUG_TAG,"Action was MOVE");
                    return true;
                case (MotionEvent.ACTION_UP) :
                    Log.d(DEBUG_TAG,"Action was UP2"+v.toString());
                    return true;

                default :
                    //return true;
            }
            return true;
        }
    });
}

}

If there is no simple way to do this, I'm thinking about solving this mathematically, but maybe some of you can help.

So my question is, is there a way to recognise an UP action on a second ImageView while currently being in a MotionEvent of another ImageView?

SOLUTION (see emandt's answer)

I ditched the second OnClickListener because I realised that the 2nd image doesn't need any, I just need its position.

Added this method:

 @Nullable
private View getDroppedView(View droppedView, int x, int y, List<View> arrayOfPossibilities) {
    Rect cVisibleBoundsRect = new Rect();
    for (View cView : arrayOfPossibilities) {
        //if currently iterated view doesn't have values for getGlobalVisibleRect, skip the .contains part
        //ignore the item which is your current active item (which would potentially be dropped)
        //getGlobalVisibleRect sets cVisibleBoundsRect immediately to the Rect given as parameter
        if (!cView.getGlobalVisibleRect(cVisibleBoundsRect)||(cView.equals(droppedView))) continue;
        if (cVisibleBoundsRect.contains(x, y)) {
            Log.d(DEBUG_TAG,"Found something");
            //THIS "cView" IS THE VIEW WHERE YOU RELEASED THE FINGER
            return cView;
        }
    }
    Log.d(DEBUG_TAG,"Found nothing");
    return null;
}

And added this in onUP:

case (MotionEvent.ACTION_UP) :
                    View dropTarget;
                    Log.d(DEBUG_TAG,"Action was UP"+v.toString());
                    dropTarget = getDroppedView(v, (int)event.getRawX(), (int)event.getRawY(), listOfViews);

                   if (dropTarget != null){
                        v.setX(dropTarget.getX());
                        v.setY(dropTarget.getY());
                    }

Answer 1

I think you want to know which is the View where you release the finger from the screen, am I right?

To do this you can use the same "View.OnTouchListener()" for all of your Views and in the ACTION_UP you have to call a new method similar to this (pseudo-code):

....
case (MotionEvent.ACTION_UP) :
    View[] cArrayOfPossibileViews = new View[]{ findViewById(IMAGE_1), findViewById(IMAGE2) }
    getDroppedView(v, event.getRawX(), event.getRawY(), cArrayOfPossibileViews);
    break;
}
....

@Nullable
private View getDroppedView(View view, int x, int y, View[] arrayOfPossibilities) {
    Rect cVisibleBoundsRect = new Rect();
    for (View cView : arrayOfPossibilities) {
        if (!cView.getGlobalVisibleRect(cVisibleBoundsRect)) continue;
        if (cVisibleBoundsRect.contains(x, y)) {
            //THIS "cView" IS THE VIEW WHERE YOU RELEASED THE FINGER
            return cView;
        }
    }
    return null;
}

This method get View bounds and compare them avains X and Y of your Touch Event. If X and Y are contained inside a View bounds it means that View is the one you need.