CC3Layer is a cocos2d CCLayer that supports full 3D rendering in combination with normal cocos2d 2D rendering. More...
#import <CC3Layer.h>
Public Member Functions | |
| (void) | - cc3AddGestureRecognizer: |
| (void) | - cc3RemoveAllGestureRecognizers |
| (void) | - cc3RemoveGestureRecognizer: |
| (void) | - initializeControls |
| (void) | - onCloseCC3Layer |
| (void) | - onOpenCC3Layer |
| (void) | - update: |
| (void) | - updateViewport |
Properties | |
| CCArray * | cc3GestureRecognizers |
| CC3Scene * | cc3Scene |
| CC3Scene *cc3World | DEPRECATED_ATTRIBUTE |
| BOOL | isOpaque |
| BOOL | shouldAlwaysUpdateViewport: 1 |
Detailed Description
CC3Layer is a cocos2d CCLayer that supports full 3D rendering in combination with normal cocos2d 2D rendering.
It forms the bridge between the 2D and 3D drawing environments.
The CC3Layer contains an instance of CC3Scene, and delegates all 3D operations, for both updating and drawing 3D models, to the CC3Scene instance.
In addition, like any cocos2d CCLayer, 2D child CCNodes can be added to this layer and will be rendered either over or under the 3D scene, based on their individual Z-order. In particular, 2D controls such as menus, sprites, labels, health bars, joysticks, etc, can be overlayed on the 3D scene simply by adding them as children of this layer. Similarly, a 2D backdrop could be rendered behind the 3D scene by adding an appropriate CCNode as a child with a negative Z-order.
Like other CCNodes, this layer can be added to another 2D node, and given a contentSize, position, and scale. You can even dynamically move and scale the embedded CC3Layer using CCActions.
Changes to the position and scale of the CC3Layer are propagated to the viewport of the contained CC3Scene, and to any child CC3Layers and CC3Scenes.
However, these properties will only be propagated if the node being moved is a CC3Layer. If the CC3Layer is a child of a regular 2D CCLayer or CCNode, and that node is moved, the resulting changes to the position or scale of the child CC3Layer may not automatically be propagated to the CC3Scene viewport. In this case, you can use the updateViewport method of CC3Layer to ensure that the CC3Scene viewport is aligned with the position and scale of the CC3Layer.
Also, although the 3D scene will be correctly rendered when this, or a parent layer is scaled, be aware that scaling of the 2D nodes affects the interaction between the 2D and 3D environments. Specifically, when the 2D layer is scaled, the following limitation apply:
- a 2D CCNode held by CC3Billboards whose shouldDrawAs2DOverlay property is set to YES, indicating that the 2D CCNode should be drawn as an overlay above the 3D scene, will not be rendered in the correct position, relative to the 3D scene.
- projection and unprojection between the 2D and 3D coordinate systems, including projecting touch events onto 3D nodes, will not work correctly.
CC3Layer descends from CCLayerColor, and will draw a colored background behind both 2D and 3D content if configured with a background color.
To make use of the standard cocos2d model updatating functionality to update and animate the 3D scene, use the scheduleUpdate or schedule:interval: methods of CC3Layer to invoke periodic callbacks to the update: method of the CC3Layer instance. The update: method forwards these callbacks to the CC3Scene instance held by the CC3Layer.
To enable simple single-touch event handling for this layer, set the isTouchEnabled property to YES. Once enabled, single-touch events will automatically be forwarded to the touchEvent:at: method on your customized CC3Scene instance to support user selection of 3D nodes via touches. For more information on handling 3D node selections, see the description of the method nodeSelected:byTouchEvent:at: of CC3Scene.
Since the touch-move events are both voluminous and seldom used, the implementation of ccTouchMoved:withEvent: has been left out of the default CC3Layer implementation. To receive and handle touch-move events for object picking, copy the commented-out ccTouchMoved:withEvent: template method implementation in CC3Layer to your customized CC3Layer subclass.
For more sophisticated touch interfaces, such as multi-touch events or gestures, add event-handing behaviour to your customized CC3Layer, as you would for any cocos2d application and, when required, invoke the touchEvent:at: method on your customized CC3Scene to initiate node selection.
Most 3D games will be displayed in full-screen mode, so typically your custom CC3Layer will be sized to cover the entire screen. However, the CC3Layer can indeed be set to a contentSize less that the full window size, and may be positioned on the window, or within a parent CCLayer like any other CCNode.
You can even dyanamically move your CC3Layer around within the window, by changing the position property (for example, by using a CCMoveTo action).
CC3Layer directly descends from CC3ControllableLayer, which means that it can optionally be controlled by a CC3UIViewController instance. Doing so enables two features:
- Automatic rotatation the layer (both the 2D and 3D components) when the device orientation changes.
- The CC3Layer can be overlaid on a device camera image stream so that both the 2D and 3D scenes can participate in an augmented reality view perspective.
With the CC3UIViewController attached, either or both of these features can be turned on or off. If neither of these features is required, there is no need to instantiate and attach a CC3UIViewController, and the CC3Layer can be used without it.
For most applications, you will create subclasses of both CC3Layer and CC3Scene. The customized subclass of CC3Scene manages the behaviour of the 3D resources. The customized subclass of CC3Layer manages the 2D artifacts, such as menus, sprites, labels, health bars, joysticks, etc, that you want to overlay on the 3D scene.
Typically, you will create a separate instance of CC3Scene for each 3D scene. You can also create a distinct CC3Layer for each scene as well or, more typically, reuse a single CC3Layer instance across multiple CC3Scene scenes by simply assigning a differnt CC3Scene instance to the layer. Any running actions in the old scene are automatically paused, and any running actions in the new scene are automatically started. For more information on swapping 3D scenes, see the notes on the cc3Scene property.
To create and use your CC3Layer and CC3Scene pair, follow these steps:
- Instantiate your CC3Scene class, including creating or loading 3D file resources in the initializeScene method.
- Instantiate your CC3Layer subclass, adding any 2D controls in the initializeControls method, and managing event handlers and gesture recognizers in the onOpenCC3Layer and onCloseCC3Layer methods.
- Attach your CC3Scene to the cc3Scene property of your CC3Layer.
- Invoke the play method of your CC3Scene to enable dynamic behaviour for the 3D scene.
- Schedule regular updates in your CC3Layer instance by invoking either the scheduleUpdate or schedule:interval: method.
- Optionally create a CC3UIViewController.
- Run your CC3Layer instance either by invoking the runSceneOnNode: method of the CC3UIViewController with your CC3Layer, or by wrapping your CC3Layer in a CCScene and invoking the runWithScene: method of the shared CCDirector instance.
Member Function Documentation
| - (void) cc3AddGestureRecognizer: | (UIGestureRecognizer *) | gesture |
Adds the specified gesture recognizer to the UIView that is displaying this layer, and tracks the gesture recognizer in the cc3GestureRecognizers property.
For applications that use a single CC3Layer to cover the entire UIView, you can override the onOpenCC3Layer method to create gesture recognizers, and you can invoke this method to easily add them to the UIView.
When this layer is removed from the view, the gesture recognizers added using this method are automatically removed from the view, and from the cc3GestureRecognizers property. Whenever this layer is displayed again, new gesture recognizers will be created and attached to the view when the onOpenCC3Layer method runs again.
For applications that diplay several CC3Layers that support gesture recognizers, you may want to create centralized gesture recognizers in some other scope, and bypass adding them using this method.
| - (void) cc3RemoveAllGestureRecognizers |
Removes all gesture recognizers that were previously added using the cc3AddGestureRecognizer: method, and removes them all from the UIView.
This method is invoked automatically when this layer is removed from the view. Usually, the application does not need to invoke this method directly, but if you need to remove all gesture recognizers prior to closing the layer, you can use this method to do so.
| - (void) cc3RemoveGestureRecognizer: | (UIGestureRecognizer *) | gesture |
Removes the specified gesture recognizer from the UIView that is displaying this layer, and removes the gesture recognizer from the cc3GestureRecognizers property.
When this layer is removed from the view, the gesture recognizers added to the cc3GestureRecognizers property using the cc3AddGestureRecognizer: method are automatically removed from the view, and from the cc3GestureRecognizers property. Usually, the application does not need to invoke this method directly.
| - (void) initializeControls |
Template method that is invoked automatically during initialization, regardless of the actual init* method that was invoked.
Subclasses can override to set up their 2D controls and other initial state without having to override all of the possible superclass init methods.
This default implementation does nothing. It is not necessary to invoke this superclass implementation when overriding in a subclass.
| - (void) onCloseCC3Layer |
Template method that is invoked automatically immediately after the CC3Scene has closed, and immediately before this layer is closed.
This default implementation does nothing. You can override this method in your custom subclass to perform tear-down activity prior to the scene disappearing.
Any gesture recognizers added in the onOpenCC3Layer method by invoking cc3AddGestureRecognizer: will be removed automatically after this method runs. You do not need to use this method to remove any gesture recognizers that you added using the cc3AddGestureRecognizer method. However, if you have bypassed the cc3AddGestureRecognizer method to create and add gesture recognizers, you can use this method to remove them.
| - (void) onOpenCC3Layer |
Template method that is invoked automatically immediately after this layer has opened on the underlying view, and before the CC3Scene is opened.
This default implementation does nothing. You can override this method in your custom subclass to perform set-up activity prior to the scene becoming visible, such as adding gesture recognizers or event handlers.
You can invoke the cc3AddGestureRecognizer method from this method to add gesture recognizers. When creating gesture recognizers, you should use your custom CC3Layer as the target of the action messages from the recognizers. You can then use the cc3Convert... family of methods on this instance to convert locations and movements from the gesture recognizers into the coordinate system of this layer.
If your application contains several CC3Layers on-screen at once, you may want to register gesture recongizers within the onEnter method of a parent grouping CCNode, instead of from within each CC3Layer.
| - (void) update: | (ccTime) | dt |
This method is invoked periodically when the components in the CC3Scene are to be updated.
The dt argument gives the interval, in seconds, since the previous update.
This implementation forwards this update to the updateScene: method of the contained CC3Scene instance. Subclasses can override to perform updates to 2D nodes added to this layer, but should be sure to invoke this superclass implementation, or to invoke updateScene: on the CC3Scene directly.
Typcially this method is scheduled to be invoked automatically at a periodic interval by using the scheduleUpdate or schedule:interval: methods of this instance, but may also be invoked by some other periodic operation, or even directly by the application.
This method is invoked asynchronously to the frame rendering animation loop, to keep the processing of model updates separate from OpenGL ES drawing.
| - (void) updateViewport |
Updates the viewport of the contained CC3Scene instance with the dimensions of this layer and the device orientation.
This method is invoked automatically when the position, size, scale, or orientation of this layer changes. You do not need to invoke this method when changing the position or scale of the layer. These changes are forwarded to the CC3Scene viewport automatically.
Usually, the application should never need to invoke this method directly. However, if your application changes the orientation of this layer in a manner that is not automatically detected, you can use this method to align the CC3Scene viewport with the updated layer.
Property Documentation
- (CCArray *) cc3GestureRecognizers [read, assign] |
Returns a collection of UIGestureRecognizers that were added using the cc3AddGestureRecognizer: method.
- (CC3Scene *) cc3Scene [read, write, retain] |
The CC3Scene instance that maintains the 3D models and draws the 3D content.
If your application contains multiple 3D scenes, you can swap between these scenes by simply setting the value of this property to the new scene. The old CC3Scene instance is released. So if you want to swap that old scene back into this layer at some point in the future, you should cache it somewhere, or recreated it.
When the old scene is released, it will clean up after itself, including all the nodes and meshes it contains.
If this layer already has a CC3Scene assigned, the wasRemoved method of the existing CC3Scene to stop and remove any CCActions running on it and the nodes it contains.
You can set the shouldStopActionsWhenRemoved of the CC3Scene to NO if you want the CCActions attached to the scene and its nodes to be paused, but not stopped and removed. Be aware that CCActions that are paused, but not stopped, will retain the CC3Scene, and could be cause for memory leaks if not managed correctly. Please see the notes of the CC3Node shouldStopActionsWhenRemoved property and the CC3Node wasRemoved method for more information.
Setting this property while this layer is being displayed automatically invokes the open method on the new scene to ensure that the transforms are up to date before the next frame is rendered.
- (CC3Scene *cc3World) DEPRECATED_ATTRIBUTE [read, write, retain] |
- Deprecated:
- Renamed to cc3Scene.
- (BOOL) isOpaque [read, assign] |
Returns whether this layer is opaque.
Return YES if the isColored property returns YES and the opacity property returns 255, otherwise returns NO.
- (BOOL) shouldAlwaysUpdateViewport [read, write, assign] |
Indicates whether this layer should update the 3D viewport on each rendering frame.
If the value of this property is YES, the 3D viewport will be updated before each frame is drawn. This is sometimes useful if the layer is changing in a way that is not automatically tracked by the 3D scene.
You do not need to set this property when changing the position or scale of the layer. These changes are forwarded to the 3D scene automatically.
The initial value of this property is NO. Unless you encounter issues when modifying the layer, leave this property set to NO, to avoid the overhead of calculating an unnecessary transformation matrix on each frame render.
As an alternate to updating the viewport on every frame render, consider invoking the updateViewport method whenever your application changes the orientation of this layer in a manner that is not automatically propagated to the CC3Scene viewport.
The documentation for this class was generated from the following file:
