Time for action – handling multitouches

There are three methods we need to implement in this game to handle touches. Each method receives, as one of its parameters, a vector of Touch objects:

  1. So add our onTouchesBegan method: 
    void GameLayer::onTouchesBegan(const std::vector<Touch*> &touches, Event* event)
{
   for( auto touch : touches) {
     if(touch != nullptr) {
        auto tap = touch->getLocation();
        for (auto player : _players) {
         if (player->boundingBox().containsPoint(tap)) {
            player->setTouch(touch);
         }
       }
     }
   }
}

    Each GameSprite, if you recall, has a _touch property.

    So we iterate through the touches, grab their location on screen, loop through the players in the vector, and determine if the touch lands on one of the players. If so, we store the touch inside the player's _touch property (from the GameSprite class).

    A similar process is repeated for onTouchesMoved and onTouchesEnded, so you can copy and paste the code and just replace what goes on inside the _players array for loop.

  2. In TouchesMoved, when we loop through the players, we do this: 
    for (auto player : _players) {
  if (player->getTouch() != nullptr && player->getTouch() ==  touch) {
    Point nextPosition = tap;
   if (nextPosition.x < player->radius())
      nextPosition.x = player->radius();
   if (nextPosition.x > _screenSize.width - player->radius())
      nextPosition.x = _screenSize.width - player->radius();
   if (nextPosition.y < player->radius())
      nextPosition.y  = player->radius();
   if (nextPosition.y > _screenSize.height - player->radius())
      nextPosition.y = _screenSize.height - player->radius();
                    
   //keep player inside its court
   if (player->getPositionY() < _screenSize.height* 0.5f) {
      if (nextPosition.y > _screenSize.height* 0.5 -  player->radius()) {
         nextPosition.y = _screenSize.height* 0.5 -  player->radius();
        }
   } else {
      if (nextPosition.y < _screenSize.height* 0.5 +  player->radius()) {
         nextPosition.y = _screenSize.height* 0.5 +  player->radius();
      }
   }              
   player->setNextPosition(nextPosition);
   player->setVector(Vec2(tap.x - player->getPositionX(),  tap.y - player->getPositionY()));
 }   
}

    We check to see if the _touch property stored inside the player is the being moved now. If so, we update the player's position with the touch's current position, but we check to see if the new position is valid: a player cannot move outside the screen and cannot enter its opponent's court. We also update the player's vector of movement; we'll need this when we collide the player with the puck. The vector is based on the player's displacement.

  3. In onTouchesEnded, we add this: 
    for (auto player : _players) {
   if (player->getTouch() != nullptr && player->getTouch() == touch) {
     //if touch ending belongs to this player, clear it
     player->setTouch(nullptr);
     player->setVector(Vec2(0,0));
   }
}

We clear the _touch property stored inside the player if this touch is the one just ending. The player also stops moving, so its vector is set to 0. Notice that we don't need the location of the touch anymore; so in TouchesEnded you can skip that bit of logic.

What just happened?

When you implement logic for multitouch this is pretty much what you will have to do: store the individual touches inside either an array or individual sprites, so you can keep tracking these touches.

Now, for the heart and soul of the game—the main loop.

上一章目录下一章