Extending callbacks

A tscript code trick i recently discovered may be useful to some here. Maybe most of you will be saying 'no s**t sherlock', but i wanted to share something that wasn't obvious to me at first as a tscript newbie.

One thing that was 'difficult' to work with for me was the single callback function for multiple objects. For example, i could have 30 sceneobjects that would all use the same fxSceneObject2D::onCollision callback block. Handling all of this collision logic in that single block quickly led to spaghetti code. I recently saw a post by member 'Luke D' that gave me insight on handling this situation much more elegantly. It takes nice advantage of the tscript's 'call' method, which effectivly calls the function which is passed to it as a string.

In the following example, i define my player and enemy sprite objects, and i define the custom callback function each will use. I then add a short block of code in the fxSceneObject2D::onCollision method which subsequently calls my custom callback function for the appropriate objects:

function createPlayer() {
                $player = new fxStaticSprite2D() { scenegraph = t2dSceneGraph; };
                // other init code deleted for brevity


                // define special callback method
                $player.collisionCB = "handlePlayerCollision";   
}

function createBugEnemy() {
                $benemy= new fxStaticSprite2D() { scenegraph = t2dSceneGraph; };
                // other init code deleted for brevity
                 
                // define special callback method
                $benemy.collisionCB = "handleBugEnemyCollision"; 
}


function fxSceneObject2D::onCollision( %srcObj, %dstObj.......... ){ 



       // if either collision object has a 
       //collision callback special method defined,
       // call the defined function:



        if (!(%srcObj.collisionCB $= "") ) {   
                call(%srcObj.collisionCB, %dstObj);        
        }
        
        if (!(%dstObj.collisionCB $= "")) { 
                call(%dstObj.collisionCB, %srcObj);
        }       
}

function handleBugEnemyCollision(%collisionObject) {  
        
       //my custom collision code for this specific object can go here
        
       %collisionObject.safeDelete();  
}

function handlePlayerCollision(%collisionObject) {  
       
       //my custom collision code for this specific object can go here
        
       %collisionObject.safeDelete();  
}

Here you see how i was able to shrink the logic in the actual fxSceneObject2D::onCollision block. I was then able to move object specific logic into each appropriate collision handling method. For me, this makes code much more readable and manageable.

One thing to note is that in the 'call' method, you can pass each required target function argument as an additional argument within the 'call' parameters. In my example, i only needed to pass the other collision object. However, if more of the original fxSceneObject2D::onCollision arguments were required by your child function logic, you could easily extend these.

This approach can also be used to seperate logic used in other callbacks.

As I stated before, i am a tscript newbie, so i may be missing a better way to accomplish what was done above. If that is the case, i'd be grateful to know the cleaner methodology.