In this tutorial, we will try to add Coin and Obstacles into our parkour game.
After this tutorial, our player should be able to collect the coin when he is running and he will die when he collides with the obstacle.
We will also cover how to design a game level with tiled map editor. Since the game logic is a little bit complex than before, so we will refactor the code before we adding new game components.
Before we start, let's finish the preparation stuff.
Since we will add two more game elements into our parkour game. So we need add some more global integer tags to identify each game items.
Let's add the following code snippets at the end of globals.js:
// collision type for chipmunk
if(typeof SpriteTag == "undefined") {
var SpriteTag = {};
SpriteTag.runner = 0;
SpriteTag.coin = 1;
SpriteTag.rock = 2;
};
Here we use 0,1,2 to represent runner,coin and rock.
We also introduce another spritesheet named background.png and background.plist. We have packed the coins and rocks sprites into the spritesheet named background.png.
The details of how to pack these sprites are leave out for the next subsection.
Next, let's copy the resource files into our res directory and add two more variables for further referring.
var s_background = "background.png";
var s_background_plist = "background.plist";
var g_resources = [
//image
{src:s_HelloBG},
{src:s_start_n},
{src:s_start_s},
{src:s_runner},
{src:s_runnerplist},
{src:s_background},
{src:s_background_plist},
{src:s_map},
{src:s_map00},
{src:s_map01}
];
In the previous chapter, we have learned how to pack a bunch of small sprites into a big large compact spritesheet. Let's pack another spritesheet.
At first, you should launch TexturePacker and drag all the assets under res/TexturePacker/coins and rocks director.(Note: You can get the whole game resource from the download as before.)
After dragging the resource, you should specify the Data file and Texture format with some path like xxx/chapter8/res/background.png or xxx/chapter8/res/background.plist.
If you don't want to any optimization of the spritesheet, just leave them out and press Publish to generate the final spritesheet.
We have used TiledMap for our level map, but it lacks game items. So in this section, we will cover how to design level items with TiledMap object layer. ###Add Coin Object Layer At first, we'll add Coin object layer.
-
Launch Tiled and open map00.tmx and map01.tmx.
-
Create an Object layer named coin in map00.tmx and map01.tmx.
-
Design object layer by dragging and dropping rectangle object into the map. You can change the rectangle size and it's position. You can also duplicate or delete the objects.
- Some tips on designing object layer: You can change opacity of the layers in the tiled map so that you can easily place the object.
###Add Rock Object Layer The process to create the Rock object layer is more or less the same as creating coin object layer.
So we will leave it out for your own implementation.
Sometimes, when you are coding, you may find that it is extremely hard to add new functionality into the existing structure.
It is a bad code smell and we should stop and do refactor work right now.
Since we will add Chipmunk physic body into our background, so we need a method to obtain the space object created in PlayScene.
Let's change the name of ctor function in Background Layer and pass a parameter named space into it. We should also add a new member variable into the BackgroundLayer class. Here is the code snippets:
ctor:function (space) {
this._super();
// clean old array here
this.objects = [];
this.space = space;
this.init();
},
Here we have added additional init code. We added a array named objects and initialize it to an empty array.
(Note: You should call this.init() method right after the assignment of this.space = space. Because we will create physic objects in the init method)
- Add more member variables into BackgroundLayer class:
space:null,
spriteSheet:null,
objects:[],
- Initialize spritesheet in the init method:
// create sprite sheet
cc.SpriteFrameCache.getInstance().addSpriteFrames(s_background_plist);
this.spriteSheet = cc.SpriteBatchNode.create(s_background);
this.addChild(this.spriteSheet);
- Add a method named loadObject to initialize rock and coins.
loadObjects:function (map, mapIndex) {
// add coins
var coinGroup = map.getObjectGroup("coin");
var coinArray = coinGroup.getObjects();
for (var i = 0; i < coinArray.length; i++) {
var coin = new Coin(this.spriteSheet,
this.space,
cc.p(coinArray[i]["x"] + this.mapWidth * mapIndex,coinArray[i]["y"]));
coin.mapIndex = mapIndex;
this.objects.push(coin);
}
// add rock
var rockGroup = map.getObjectGroup("rock");
var rockArray = rockGroup.getObjects();
for (var i = 0; i < rockArray.length; i++) {
var rock = new Rock(this.spriteSheet,
this.space,
rockArray[i]["x"] + this.mapWidth * mapIndex);
rock.mapIndex = mapIndex;
this.objects.push(rock);
}
},
Here we iterate all the objects info in the tiled map and create responding Chipmunk rigid bodies. Finally we store these object into the objects array.
All these code are self-explanation. You should only pay attention to the mapIndex parameter. We use the parameter to calculate where we should place the rigid body.
We need call loadObject method at the end of init method to create the physic objects in the first two screen maps.
this.loadObjects(this.map00, 0);
this.loadObjects(this.map01, 1);
- Add another two helper methods for removing unused chipmunk rigid bodies.
The first method is called removeObjects. It removes a object by mapIndex. Here is the implementation:
removeObjects:function (mapIndex) {
while((function (obj, index) {
for (var i = 0; i < obj.length; i++) {
if (obj[i].mapIndex == index) {
obj[i].removeFromParent();
obj.splice(i, 1);
return true;
}
}
return false;
})(this.objects, mapIndex));
},
The other method is called removeObjectByShape:
removeObjectByShape:function (shape) {
for (var i = 0; i < this.objects.length; i++) {
if (this.objects[i].getShape() == shape) {
this.objects[i].removeFromParent();
this.objects.splice(i, 1);
break;
}
}
},
This method will remove a chipmunk object by its shape.
###Wrap it up: Add Creation and Disposable Logic in checkAndReload Method When the map is moved, we should also call loadObject method to recreate the "Coins & Rocks".
And also we sould remove all unused objects by calling removeObjects method.
Here is the code snippets:
checkAndReload:function (eyeX) {
var newMapIndex = parseInt(eyeX / this.mapWidth);
if (this.mapIndex == newMapIndex) {
return false;
}
if (0 == newMapIndex % 2) {
// change mapSecond
this.map01.setPositionX(this.mapWidth * (newMapIndex + 1));
this.loadObjects(this.map01, newMapIndex + 1);
} else {
// change mapFirst
this.map00.setPositionX(this.mapWidth * (newMapIndex + 1));
this.loadObjects(this.map00, newMapIndex + 1);
}
this.removeObjects(newMapIndex - 1);
this.mapIndex = newMapIndex;
return true;
},
Now it's time to add the coin and rock implementation. Despite the implementation details, you should also pay attention to the design idea behind these two classes. Here we prefer to inherit from cc.Class instead of cc.Sprite. We let each object to own a instance of cc.Sprite.
-
Create a new file named Coin.js. We will define our Coin class in this file. Make sure you have this filed located in your src directory.
-
Derived a class named Coin from cc.Class, let's take a look at the whole implementation:
var Coin = cc.Class.extend({
space:null,
sprite:null,
shape:null,
_mapIndex:0,// which map belongs to
get mapIndex() {
return this._mapIndex;
},
set mapIndex(index) {
this._mapIndex = index;
},
/** Constructor
* @param {cc.SpriteBatchNode *}
* @param {cp.Space *}
* @param {cc.p}
*/
ctor:function (spriteSheet, space, pos) {
this.space = space;
// init coin animation
var animFrames = [];
for (var i = 0; i < 8; i++) {
var str = "coin" + i + ".png";
var frame = cc.SpriteFrameCache.getInstance().getSpriteFrame(str);
animFrames.push(frame);
}
var animation = cc.Animation.create(animFrames, 0.2);
var action = cc.RepeatForever.create(cc.Animate.create(animation));
this.sprite = cc.PhysicsSprite.createWithSpriteFrameName("coin0.png");
// init physics
var radius = 0.95 * this.sprite.getContentSize().width / 2;
var body = new cp.StaticBody();
body.setPos(pos);
this.sprite.setBody(body);
this.shape = new cp.CircleShape(body, radius, cp.vzero);
this.shape.setCollisionType(SpriteTag.coin);
//Sensors only call collision callbacks, and never generate real collisions
this.shape.setSensor(true);
this.space.addStaticShape(this.shape);
// add sprite to sprite sheet
this.sprite.runAction(action);
spriteSheet.addChild(this.sprite, 1);
},
removeFromParent:function () {
this.space.removeStaticShape(this.shape);
this.shape = null;
this.sprite.removeFromParent();
this.sprite = null;
},
getShape:function () {
return this.shape;
}
});
Let's explain the code piece by piece.
At first, we add three member variables named: space, sprite and shape. We will use these variables to create the coin object's physic body and its display attribution.
Then, we added another member variable _mapIndex. We used the get/set syntax sugar to define accessor of the variable.
The ctor method is the constructor of Coin class. We will create a Coin class with a spritesheet, a space and a position object later.
Since the coins are circular shape, so we have created CircleShape attached to the rigid body. The remaining part of the ctor function is self-explanation.
At last, we need to define a method to do the cleanup work. It's the removeFromParent method. It firstly remove the rigid body from the space and then remove the sprite from its parent. The getShape method is just a getter method used for accessing the shape object stored in the coin object.
The principle of designing the Rock class is more or less as the Coin class except for the rigid shape type part.
Because our Rock class is a rectangle box. So we use cp.BoxShape to replace the cc.CircleShape in Coin class.
Here is the full source code of Rock.js:
var Rock = cc.Class.extend({
space:null,
sprite:null,
shape:null,
_map:0,// which map belong to
get map() {
return this._map;
},
set map(newMap) {
this._map = newMap;
},
/** Constructor
* @param {cc.SpriteBatchNode *}
* @param {cp.Space *}
* @param {cc.p}
*/
ctor:function (spriteSheet, space, posX) {
this.space = space;
this.sprite = cc.PhysicsSprite.createWithSpriteFrameName("rock.png");
var body = new cp.StaticBody();
body.setPos(cc.p(posX, this.sprite.getContentSize().height / 2 + g_groundHight));
this.sprite.setBody(body);
this.shape = new cp.BoxShape(body,
this.sprite.getContentSize().width,
this.sprite.getContentSize().height);
this.shape.setCollisionType(SpriteTag.rock);
this.space.addStaticShape(this.shape);
spriteSheet.addChild(this.sprite);
},
removeFromParent:function () {
this.space.removeStaticShape(this.shape);
this.shape = null;
this.sprite.removeFromParent();
this.sprite = null;
},
getShape:function () {
return this.shape;
}
});
- At first, let's add a extra array named shapesToRemove and initialize it at the beginning of onEnter function.
//the following line goes in init member variable define area
shapesToRemove :[],
//the following line goes at the beginning of the *onEnter* function.
this.shapesToRemove = [];
- Secondly, modify the creation of BackgroundLayer. Here we simply pass the space object into BackgroundLayer's constructor.
this.addChild(new BackgroundLayer(this.space), 0, TagOfLayer.background);
- At first, we should call these two functions at the end of initPhyiscs method:
// setup chipmunk CollisionHandler
this.space.addCollisionHandler(SpriteTag.runner, SpriteTag.coin,
this.collisionCoinBegin.bind(this), null, null, null);
this.space.addCollisionHandler(SpriteTag.runner, SpriteTag.rock,
this.collisionRockBegin.bind(this), null, null, null);
The addCollisionHandler method needs a callback when collision occurs.
- Then, let's define these two callbacks to handle player collide with coins and rocks.
collisionCoinBegin:function (arbiter, space) {
var shapes = arbiter.getShapes();
// shapes[0] is runner
this.shapesToRemove.push(shapes[1]);
},
collisionRockBegin:function (arbiter, space) {
cc.log("==game over");
},
- Delete unused rigid bodies in background layer. You should add the following code at the end of update method:
// Simulation cpSpaceAddPostStepCallback
for(var i = 0; i < this.shapesToRemove.length; i++) {
var shape = this.shapesToRemove[i];
this.getChildByTag(TagOfLayer.background).removeObjectByShape(shape);
}
this.shapesToRemove = [];
We can't delete physic bodies during the physic simulation process. so we use an extra array named shapesToRemove to hold the temporal data needed to be deleted.
Congratulations! You almost reach the end. Before we hit the debug button to see the results. Let's add some extra glue code to connect everything together.
Open cocos2d.js and append two more array items at the end of appFiles array.
appFiles:[
'src/resource.js',
'src/myApp.js',
'src/AnimationLayer.js',
'src/BackgroundLayer.js',
'src/PlayScene.js',
'src/StatusLayer.js',
'src/globals.js',
'src/Coin.js',
'src/Rock.js'
]
Build and run! Cheers, we did it!:)
Let's see our final fruits:
In this tutorial, we have enjoyed a very long journey. But worth it, isn't it?
We have learned how to use TiledMap's object layer to design complex game levels and how to customize your own class to extend your code structure.
You can download the full source code from here.
In the next tutorial, we'll cover how to update game HUD constantly and we'll also add game over logic and simple gesture recognizer into our game to make the player to jump over the obstacles. Keep tuning!