Action<T> vs virtual methods - c#

Consider I have a game-creation framework, with a basic class called Game.
// Action version
public class Game
{
public Action<float> OnUpdate;
public void Update(float mFrameTime) { OnUpdate.Invoke(mFrameTime); }
}
// Virtual version
public class Game
{
public virtual void Update(float mFrameTime) {}
}
Which one is the best approach?
(design-wise and performance-wise)
Subscribing something to the OnUpdate action (and not inheriting the Game class), or inheriting the Game class and overriding the virtual method?

This depends on who the target audience is for the Update notification. If it's for only derived classes then a virtual method is a fine approach. If it's for derived classes and or arbitrary consumer then the event style pattern is the correct approach.

I would rather raise events than have the Action<T>. This is a common pattern in .NET:
public class Game
{
public event EventHandler<SomeEventArgs> Update;
protected virtual void OnUpdate(float mFrameTime) { // invoke event here }
}

It depends what the rest of your framework looks like and what you plan to do with this class. If this class simply alerts other classes that its time to update, then you'll want an event based system. If this is a base class for all games to derive from, then you'll probably want an override. No answer is 100% correct - it all depends on what exactly you are trying to accomplish with your design.

Unfortunately I lack the knowledge to answer which of those options would be more efficient from a performance perspective, but from a design perspective I would advise overriding the Update method for the following reasons:
It keeps all your update logic in one spot.
It allows you to easily monitor the ordering of update actions.
As others have said, it depends on the purpose of the class, and I am making assumptions based on the fact that it's a 'game' class. There's no reason you couldn't implement both, with a virtual method that triggers an event. This would allow for the bulk of update code to be located centrally, with the option for elements to subscribe to the update event as needed.
When I've approached the same problem myself I've typically adopted a virtual-method approach, though I've broken down update into multiple sections for convenience and control purposes.

Related

Mimicking multiple inheritance in C# without interfaces

I know that multiple inheritance in C# is only allowed by using Interfaces and that there are very valid reasons why multiple inheritance can quickly become a real headache. (Working in .NET Framework if that makes any difference to the answers)
However.
In working on various projects accross many classes I find myself returning to the same patterns to handle behaviour.
For example I have an Interface IXMLSavable which requires the functions GetXML() and SetFromXML(XElement e) to be implemented. The way I implement this in every class is, that I write different functions for different versions of the XML (If I changed something in the GetXML() I want to maintain backwards compatibility...). And according to a version-attribute on the root Element I switch case to the right ReadVersionX(XElement e) function so all my data stays consitent.
Another example would be centered around eventing. If for example I want to implement a "stop firing events for the time being"-Lock I would go about thusly:
private bool suppressEvents;
public bool SuppressEvents
{
get { return suppressEvents; }
set
{
bool prevValue=SuppressEvents;
suppressEvents=value;
if(prevValue!=SuppressEvents && !SuppressEvents) TheChangeEvent?.Invoke();
}
}
So I can run multiple operations on the object in question without it giving of a right old firework display of events. Again: This code will be almost unchanged for a lot of classes.
For the XML one I could refactor this to a class that has a Dictionary<int,delegate> ReadFunctions which I could then set in every implementation (I concede that there needs to be a bit of customisation in the "implementing" class) and reduce the amount of bolierplate for every class (the explicit switching on the version attribute) to just filling this dictionary.
The eventing one could go into a class on its own quite readily, I would probably only need to hook up the right event to the invokation, but that could easily be remedied by an abstract function I will have to implement (again: customisation still necessary but much less boilerplate).
Each "now-class-was-interface" on its own would make a splendid base class for any object. I could use functionality down an inheritance tree and customise it by overwriting functionality with new if I would need it.
The problem starts when I want to combine the two now-classes together. Due to the limitation in C# (which, again, is there for a reason) I cannot inherit from both above described classes at the same time. This would only be possible if I have one of these classes inherit from the other. Which would be possible, but would lead to a whole lot of a different headache when I want one functionality but not the other. Or the other functionality and not the one. The way around that would be to create a plethora of permutation classes (so one class for each combination of the functionaities). And while that would solve the problem it would probably be a nightmare to maintain.
So the real question is: Is there a way to correctly plug in different already implemented functionality into a class in an inheritance like manner that allows the addition of multiple distinct functionality packages as opposed to interfaces that cannot by the very nature of themselves provide any concrete implementation.
In many cases you can avoid inheritance with the use of interfaces/default interface methods/extension methods, decorators, or some other pattern.
In your case with xml you could simply change your interface to have one ReadMethod per version, and use a extension method to select the correct one
public interface IXMLReadable{
void ReadVersion1(XElement e);
void ReadVersion2(XElement e);
}
public static class IXMLReadableExtensions {
public static void Read(this IXMLReadable self, XElement e){
// Read version from xml, call ReadVersion1 or ReadVersion2
}
}
default interface methods would do more or less the same thing, with the added advantage of allowing the class to override the Read-method if it wants some other behavior.
However, my preferred solution would be to instead convert your object to a Data Transfer Object (DTO), add any required serialization attributes to this object, and use a library to serialize this. Added fields etc can usually be accommodated by just marking it as optional. Larger changes can usually be done by creating a new DTO class.
One way to solve your event problem could be to move this logic to a separate class
public class SuppressibleEvent
{
private bool suppressEvents;
private bool pendingEvent;
public void Raise()
{
if (!suppressEvents)
{
TheChangeEvent?.Invoke(this, EventArgs.Empty);
}
else
{
pendingEvent = true;
}
}
public event EventHandler TheChangeEvent;
public bool SuppressEvents
{
get => suppressEvents;
set
{
suppressEvents = value;
if (!suppressEvents && pendingEvent)
{
TheChangeEvent?.Invoke(this, EventArgs.Empty);
pendingEvent = false;
}
}
}
}
Optionally you may add a interface, so that only the owner can raise the event, but others can listen and register. You could also add methods/events to your class that just forwards to the actual implementation.
The overall point is that there is usually a better pattern to use than implementation inheritance. Some might require a bit more code, but usually gain a bit of flexibility as a result.

Testable code: Attach event handler in constructor

as for my understanding, part of writing (unit-)testable code, a constructor should not do real work in constructor and only assigning fields. This worked pretty well so far. But I came across with a problem and I'm not sure what is the best way to solve it. See the code sample below.
class SomeClass
{
private IClassWithEvent classWithEvent;
public SomeClass(IClassWithEvent classWithEvent)
{
this.classWithEvent = classWithEvent;
// (1) attach event handler in ctor.
this.classWithEvent.Event += OnEvent;
}
public void ActivateEventHandling()
{
// (2) attach event handler in method
this.classWithEvent.Event += OnEvent;
}
private void OnEvent(object sender, EventArgs args)
{
}
}
For me option (1) sounds fine, but it the constructor should only assign fields. Option (2) feels a bit "too much".
Any help is appreciated.
A unit test would test SomeClass at most. Therefore you would typically mock classWithEvent. Using some kind of injection for classWithEvent in ctor is fine.
Just as Thomas Weller said wiring is field assignment.
Option 2 is actually bad IMHO. As if you omit a call to ActivateEventHandling you end up with a improperly initialized class and need to transport knowledge of the requirement to call ActivateEventHandling in comments or somehow else, which make the class harder to use and probably results in a class-usage that was not even tested by you, as you have called ActivateEventHandling and tested it but an uninformed user omitting the activation didn't, and you have certainly not tested your class when ActivateEventHandling was not called, right? :)
Edit: There may be alternative approaches here which are worth mentioning it
Depending on the paradigm it may be wise to avoid wiring events in the class at all. I need to relativize my comment on Stephen Byrne's answer.
Wiring can be regarded as context knowledge. The single responsibility principle says a class should do only one task. Furthermore a class can be used more versatile if it does not have a dependency to something else. A very loosely coupled system would provide many classes witch have events and handlers and do not know other classes.
The environment is then responsible for wiring all the classes together to connect events properly with handlers.
The environment would create the context in which the classes interact with each-other in a meaningful way.
A class in this case does therefore not know to whom it will be bound and it actually does not care. If it requires a value, it asks for it, whom it asks should be unknown to it. In that case there wouldn't even be an interface injected into the ctor to avoid a dependency. This concept is similar to neurons in a brain as they also emit messages to the environment and expect answer not knowing neighbouring neurons.
However I regard a dependency to an interface, if it is injected by some means of a dependency injection container just another paradigm and not less wrong.
The non trivial task of the environment to wire up all classes on start may lead to runtime errors (which are mitigated by a very good test coverage of functional and integration tests, which may be a hard task for large projects) and it gets very annoying if you need to wire dozens of classes and probably hundreds of events on startup manually.
While I agree that wiring in an environment and not in the class itself can be nice, it is not practical for large scale code.
Ralf Westphal (one of the founders of the clean code developer initiative (sorry german only)) has written a software that performs the wiring automatically in a concept called "event based components" (not necessarily coined by himself). It uses naming conventions and signature matching with reflection to bind events and handlers together.
Wiring events is field assignment (because delegates are nothing but simple reference variables that point to methods).
So option(1) is fine.
The point of constructor is not to "assign fields". It is to establish invariants of your object, i. e. something that never changes during its lifetime.
So if in other methods of class you depend on being always subscribed to some object, you'd better do it in the constructor.
On the other hand, if subscriptions come and go (probably not the case here), you can move this code to another method.
The single responsibility principle dictates that that wiring should be avoided. Your class should not care how, or where from it receives data. It would make sense to rename OnEvent method to something more meaningful, and make it public.
Then some other class (bootstrapper, configurator, whatever) should be responsible for the wiring. Your class should only be responsible for what happens when a new data come's in.
Pseudo code:
public interface IEventProvider //your IClassWithEvent
{
Event MyEvent...
}
public class EventResponder : IEventResponder
{
public void OnEvent(object sender, EventArgs args){...}
}
public class Boostrapper
{
public void WireEvent(IEventProvider eventProvider, IEventResponder eventResponder)
{
eventProvider>event += eventResponder.OnEvent;
}
}
Note, the above is pseudo code, and it's only for the purpose to describe the idea.
How your bootstrapper actually is implemented depends on many things. It can be your "main" method, or your global.asax, or whatever you have in place to actually configure and prepare your application.
The idea is, that whatever is responsible to prepare the application to run, should compose it, not the classes themselves, as they should be as single purpose as possible, and should not care too much about how and where they are used.

How can I layout a decoupled class structure for a simple game?

Right now I have six classes:
Listener - manages socket connections
World - a collection of entities and tasks
Ticker - coordinates updating the world
MessageProcessor - receives commands from players
Intelligence - governs the behavior of non-player characters
Tasks - tracking and execution of tasks
But they are like spaghetti with reference to each other all over the place... The World is a data model which the MessageProcessor, Intelligence, and Tasks classes modify. The Ticker coordiates those three classes updating the World. The Listener is used by the MessageProcessor for incomming messages, and by the other classes to push updates.
How can I improve this situation?
I gave a related answer not long ago. The subject was on improving the testability of code, for which the general solution is to loosen coupling. The main focus on that previous answer was on decoupling networking related code from the world and it's logic, because networking code is not unit testable and is a pain to mock too.
The solution given there was to use an interface for incoming messages, such that you decouple the MessageProcessor (named Handler in other post) from the network code, and similarly, decouple the UpdateNotifier from the World.
The dashed line is just an indirect reference handled either by an interface or delegate. There exists no direct relation between the World and networking component now, making it testable. This is really just an application of the Model View Adapter pattern.
This doesn't seem dissimilar to the design you've described having, except perhaps you are missing a few interfaces. With this pattern of interface based UpdateNotifiers used to push updates, I essentially reuse the same architecture for handling NPCs, tasks, or anything else which is processed elsewhere. You cherry pick the event notifiers you need for a particular area, and implement a concrete Notifier class for them, such that you have multiple adapters on the same model.
And that really only looks more complicated than it is. The World object has no direct dependencies on anything else, and each other class has at most one direct dependency. You can also isolate the timer from the World, as it probably isn't needed there - but perhaps one of the biggest hurdles is handling synchronization between the different adapters.
Well, I'm not sure I have a full picture of what the issues you are having are, but I have a few possibilities from what you have laid out so far. (I may be actually suggesting some things that are already done since I'm not sure I have enough from the one-liner descriptions to understand fully.
The Model
I would say from what you've described, the main thing that strikes me is that you'll want to start implementing common functionality in a class model; you will want either interfaces or base classes that you can use to derive your high-level objects from.
This way you can handle things consistently with little extra effort. I think the idea of "architectural layers" can be useful as a first cut of how to think about it, (e.g. low-level hardware stuff, socket handling etc., then middle-layers stuff like what kind of things happen in your game, and the details behind how game mechanics work, etc., and high-level stuff like what can the PC or NPCs do, what's the environment doing, etc.. and also the idea that you never want to "jump" layers). However, when it comes down to it the important thing is to just find the right abstractions for your game, and keep everything organized in such a way as you never feel like the bit of code you're working on is doing two completely different kinds of things.
So, first, let's take the fact that it sounds like (naturally) there are a lot of things interacting with world state. For something like this, it's probably advantageous to factor a lot of the 'stuff' out into a single class, and then mostly only have the one class doing that job. Ideally you implement, say, event communication/message passing, in it's own little group, so that there's no need to pollute your higher-level objects with the nitty-gritty of handling stuff.
e.g., you want to focus on what things are doing at a high level in the higher-level objects: in an AI perhaps "begin movement toward a location", "set my haste", "stop movement"; and in an environment subsystem do "start raining", "increase windspeed", "dim lights"; in a user class "fire weapon", "sleep", "cast spell". But I wouldn't want any of my high-level classes to even know about things like "send message to world", or "reset thirst timer", or "receive socket data", or "health cycle tick". (These are all just elucidations, not suggestions. ;D)
Events
For instance, I think it may be valuable to keep one object in charge of dispatching events to the World, that way you no longer have everyone talking to everyone. I would likely just create a set of stuff to handle events in general. So, maybe EventMain and an enumEvents that you use so that each type of event has a special ID. And then use Event as the base class for particular events that need extra functionality. (I have both the ID as well as a derivation model in mind, so that things thing like the Dispatcher which likely only need to know very basic things about the event don't have to also know about the derived classes. For instance, the dispatcher could take an event in and send it out without ever having to know the internals of a derived event. This may or may not turn out to be useful, but it's good to have the options.) You could also have an EventDispatcher that has a queue of events to be send to other subsystems.
You will want something common for recieving and sending events. You could do EventSourcer and EventSinker standalone classes that can be set up in any class that is generating or receiving events. Or, you could instead do IEventSource and IEventSink so that you could implement a common interface on multiple classes, or perhaps a common class EventSourceAndSink that implements both, and which is part of your base class model, so that anything that might need to handle events can just derive from it.
I would probably make ProtocolEncoder and ProtocolDecoder classes. You can always combine them into a single object, but it may be valuable, and shouldn't cause any issues if done adequately, to have them be two separate pieces of code. You could also have a ProtocolHelper class that factors out anything in common between the two. The encoders sole job is to receive messages from the network and turn them into events for your game, which it will then pass on to the EventDispatcher. The decoder class will take events from the dispatcher that need to go out to the network, and it will take the data from them and send it out.
How to Get Where You're Going
Since you do have working code, I would recommend that you just start doing it wherever seems natural. That could be things that are bogging you down, or things you've noticed to be very similar in different places that you could make regular with a class or some other type of abstraction, then pull out the old and put in the new. Once you have figured out a workable first cut of a class model, that should give you ideas based on what you already have and as you go be constantly reconsidering your model, fixing the things that are a problem, lather, rinse, repeat.
It doesn't have to be a lot of work, in fact, some of the most gratifying moments I've had working on code was when I was able to do a neat refactor that left a formerly hideous mess in much better shape, removing a lot of hard-to-understand code, and replacing it with something that's easier to understand, in fewer lines of code, and that opened up a path toward my next addition being a pleasure instead of another "zomg I don't have to touch that code again do I?" moment.
Best of luck, follows is a nominal guide to the things I was talking about; the first bit is more detailed because the main event class is one of the more important concepts, and then I try to just give a nominal overview of the classes and how they interact. I'm sure I could spend even more hours on this, but at this point I'll just say: ask me if you have questions and I'll do what I can to give you a good answer :)
Ideas in Code
Oh, one more thing of note is I didn't deal at all with the complexities added if you have multiple threads; there are things ranging from simple to intricate to manage it all if you do, but that's another exercise. :)
using System;
using System.Collections;
using System.Collections.Generic;
using System.ComponentModel;
using System.Data;
using System.Drawing;
using System.Linq;
using System.Text;
using System.Windows.Forms;
// this is internal to the project namespace, say, TimsWorld_o_Hurt
// I'm now resisting calling everything Xxxx_o_Hurt :)
// examples o' hurt
using EventHandlingLibrary;
namespace EventHandlingLibrary
{
// this will provide the base class for all the events, and can
// also have static methods like factory methods, destination
// lookups etc.
// I have the enums set to protected with the intent being that
// specific factory functions should be called by other classes.
// You should change this if it turns out to be too cumbersome.
public class EventOfHurt
{
#region Event Definitions
protected enum EEventType
{
// System Events
SystemInitializing,
SubsystemInitComplete,
FatalErrorNotification,
SubsystemPingReponse,
SubsystemPingRequest,
// Network Events
FrameRateError,
ThroughputData,
ServerTimeout,
ServerPingRequest,
ServerPingResponse,
// User Events
WeaponsFire,
MovementNotification,
FatigueUpdate
// and so forth
}
protected enum ESubsystem
{
System,
Dispatcher,
TickerTimer,
WorldEntity,
WorldTaskManager,
UserMessageProcessor,
NetworkListener,
NetworkTransmitter,
ProtocolEncoder,
ProtocolDecoder,
PlayerCharacter,
NonPlayerCharacter,
EventSink,
EventSource
// and such
}
#endregion
#region Event Information
public Guid EventId { get; protected set; }
public EEventType EventType { get; protected set; }
public ESubsystem SourceSubsystem { get; protected set; }
public ESubsystem DestSubsystem { get; protected set; }
private List<Tuple<EventOfHurt, DateTime>>
myEventReferences;
// the event(s) that triggered it, if any, and when rec'd
public Tuple<EventOfHurt, DateTime>[]
EventReferences
{
get { return myEventReferences.ToArray(); }
}
public DateTime Timestamp { get; private set; }
#endregion
// we'll be using factor methods to create events
// so keep constructors private; no default constructor
private EventOfHurt(
EEventType evt,
ESubsystem src,
ESubsystem dest = ESubsystem.Dispatcher
)
{
EventType = evt;
SourceSubsystem = src;
DestSubsystem = dest;
EventId = Guid.NewGuid();
Timestamp = DateTime.UtcNow;
}
// called to create a non-derived event for simple things;
// but keep other classes limited to calling specific factory
// methods
protected static EventOfHurt CreateGeneric(
EEventType evt, ESubsystem src,
ESubsystem dest = ESubsystem.Dispatcher,
Tuple<EventOfHurt, DateTime>[] reasons = null
)
{
EventOfHurt RetVal;
if (dest == null)
dest = ESubsystem.Dispatcher;
List<Tuple<EventOfHurt, DateTime>> ReasonList =
new List<Tuple<EventOfHurt,DateTime>>();
if (reasons != null)
ReasonList.AddRange(reasons);
// the initializer after the constructor allows for a
// lot more flexibility than e.g., optional params
RetVal = new EventOfHurt(evt, src) {
myEventReferences = ReasonList
};
return RetVal;
}
// some of the specific methods can just return a generic
// non-derived event
public static EventOfHurt CreateTickerTimerEvent(
EEventType evt, ESubsystem dest
)
{
ESubsystem src = ESubsystem.TickerTimer;
return CreateGeneric(evt, src, dest, null);
}
// some may return actual derived classes
public static EventOfHurt CreatePlayerActionEvent(
EEventType evt, ESubsystem dest,
Tuple<EventOfHurt, DateTime>[] reasons
)
{
PlayerEvent PE = new PlayerActionEvent(42);
return PE;
}
}
// could have some specific info relevant to player
// events in this class, world location, etc.
public class PlayerEvent :
EventOfHurt
{
};
// and even further speciailzation here, weapon used
// speed, etc.
public class PlayerActionEvent :
PlayerEvent
{
public PlayerActionEvent(int ExtraInfo)
{
}
};
}
namespace EntitiesOfHurt
{
public class LatchedBool
{
private bool myValue = false;
public bool Value
{
get { return myValue; }
set {
if (!myValue)
myValue = value;
}
}
}
public class EventOfHurtArgs :
EventArgs
{
public EventOfHurtArgs(EventOfHurt evt)
{
myDispatchedEvent = evt;
}
private EventOfHurt myDispatchedEvent;
public EventOfHurt DispatchedEvent
{
get { return myDispatchedEvent; }
}
}
public class MultiDispatchEventArgs :
EventOfHurtArgs
{
public MultiDispatchEventArgs(EventOfHurt evt) :
base(evt)
{
}
public LatchedBool isHandled;
}
public interface IEventSink
{
// could do this via methods like this, or by attching to the
// events in a source
void MultiDispatchRecieve(object sender, MultiDispatchEventArgs e);
void EventOfHurt(object sender, EventOfHurtArgs e);
// to allow attaching an event source and notifying that
// the events need to be hooked
void AttachEventSource(IEventSource evtSource);
void DetachEventSource(IEventSource evtSource);
}
// you could hook things up in your app so that most requests
// go through the Dispatcher
public interface IEventSource
{
// for IEventSinks to map
event EventHandler<MultiDispatchEventArgs> onMultiDispatchEvent;
event EventHandler<EventOfHurtArgs> onEventOfHurt;
void FireEventOfHurt(EventOfHurt newEvent);
void FireMultiDispatchEvent(EventOfHurt newEvent);
// to allow attaching an event source and notifying that
// the events need to be hooked
void AttachEventSink(IEventSink evtSink);
void DetachEventSink(IEventSink evtSink);
}
// to the extent that it works with your model, I think it likely
// that you'll want to keep the event flow being mainly just
// Dispatcher <---> Others and to minimize except where absolutely
// necessary (e.g., performance) Others <---> Others.
// DON'T FORGET THREAD SAFETY! :)
public class Dispatcher :
IEventSource, IEventSink
{
}
public class ProtocolDecoder :
IEventSource
{
}
public class ProtocolEncoder :
IEventSink
{
}
public class NetworkListener
{
// just have these as members, then you can have the
// functionality of both on the listener, but the
// listener will not send or receive events, it will
// focus on the sockets.
private ProtocolEncoder myEncoder;
private ProtocolDecoder myDecoder;
}
public class TheWorld :
IEventSink, IEventSource
{
}
public class Character
{
}
public class NonPlayerCharacter :
Character,
IEventSource, IEventSink
{
}
public class PlayerCharacter :
Character,
IEventSource, IEventSink
{
}
}

Is there such a thing as too many events?

public class Human
{
public void Run(){}
public void Jump(){}
public void Eat(){}
//Generalized approach
public EventHandler<HumanActivityProgressChanged> ActivityProgressChanged;
public EventHandler<HumanActivityCompleted> ActivityCompleted;
//Per-method approach
public EventHandler<HumanActivityProgressChanged> Running;
public EventHandler<HumanActivityCompleted> Ran;
public EventHandler<HumanActivityProgressChanged> Jumping;
public EventHandler<HumanActivityCompleted> Jumped;
public EventHandler<HumanActivityProgressChanged> Eating;
public EventHandler<HumanActivityCompleted> Ate;
}
I have different methods that implements Event-based Async pattern. These methods trigger a ProgressChanged eventargs and a Completed eventargs. They all trigger the same eventargs (As shown in the code above).
Does it make sense to provide an event per each async method? Or just provide a generalized event for all async methods? Is there such a thing as too much events?
Both are valid. It really depends on what your intention is.
Will you expect a listener to want to listen to all events and repond in similar ways? Go fot the generalized event.
If you are expecting a load of disparate listeners each interested in different aspects, performing wildly different tasks on these events, go for the second.
The idea behind api design is not to impose a certain way of usage on the clients (Rails users may disagree), but you will give strong hints in the way you design it..
The fact that all events have the same EventArgs is an indication that you might replace these events with a single event and pass the activity as a property of the HumanActivityProgressChanged and HumanActivityCompleted types.
There is no law that tells you to do so. It all depends on what you wish to expose and what clients would expect/need.
It's a question of taste, but I personally prefer the first approach.
For one thing, maintenance is going to be much easier.
And it is simply more efficient to code that way.

Adding behaviour to a set of classes

I have defined an Event class:
Event
and all the following classes inherit from Event:
SportEventType1 SportEventType2 SportEventType3 SportEventType4
Right now I will only have SportEvents but I don't know if in the future I'll want some other kind of events that doesn't even have anything to do with Sports.
Later, I will want to draw some graphics with info taken from Events, and the drawing logic can be a bit complex. But, for the moment, I think I shouldn't think of how the drawing will be done and I believe that maybe it'd be better if that drawing part was not put as an integral part of the Event/SportEventX class chain.
I am looking for solutions for this problem. I know I could just make Event have an instance variable(attribute, for the java crowds) pointing to something as an IDrawInterface, but that would make the Event class "assume" it will be later used for drawing. I would like to make the Event class oblivious to this if possible.
Thanks!
Your intent to keep knowledge of the drawing process outside the Event class hierarchy is good.
A common way to handle this sort of thing in an OO language is the Visitor Pattern.
If you can't actually change the Event class to add the accept(Visitor v) needed for Visitor, you might consider using a Decorator or an Adapter. Getting the accept method to vary by subclass might be painful with these though. I'll think about this a bit more and maybe add further notes tonight. For now, I've got to get to work.
Here's a more complex but pretty flexible approach. We'll define an interface
for a type that can draw some events:
interface IEventRenderer
{
// Draw the given event, if it can. Return true if the event was drawn,
// false otherwise.
bool Draw(Event event);
}
It does two things: it checks to see if it can draw a given event, and, if so,
draws it. Otherwise it bails and returns false.
For example, a class that can render Sport1Events looks like:
class Sport1EventRenderer : IEventRenderer
{
public bool Draw(Event event)
{
var sportEvent = event as Sport1Event;
// can only draw this type
if (sportEvent == null) return false;
// draw the event...
return true;
}
}
Then we'll define a registry class. It's job is to maintain a collection of
these renderers and hand off the work of drawing an event to the appropriate
one:
class EventRendererRegistry
{
public void Add(IEventRenderer renderer)
{
mRenderers.Add(renderer);
}
public void Draw(Event event)
{
foreach (var renderer in mRenderers)
{
if (renderer.Draw(event)) break;
}
}
private readonly List<IEventRenderer> mRenderers = new List<IEventRenderer>();
}
All it does is find the first renderer that can successfully draw the event.
You would then use this like:
var registry = new EventRendererRegistry();
registry.Add(new Sport1EventRenderer());
registry.Draw(someEvent);
Pros:
Event types are not coupled to any rendering code.
Renderers are not coupled to each other.
Renderers are only coupled to the events they care about. (For example a
Sport2EventRenderer would not need to be coupled to Sport1Event.)
Renders can do arbitrary logic to determine if they're appropriate. We're just
doing a type test here, but we could see if the event implements a certain
interface, has a certain property, is in a certain state, etc.
Relatively fast. No reflection beyond simple casting.
Cons:
Fairly complex.
Can fail at runtime to find a matching renderer.
Have to iterate through renderer collection each time to find a match.
One other solution might be to have an DrawbleEvent abstract class where different sportingevents can inherit from.
public abstract DrawableEvent
{
Event event;
IDrawingStrategy drawingstrategy;
public Draw()
{
drawingStrategy.Draw();
}
}
public SportingEvent1 : DrawableEvent
{
SprortingEvent1(Event event, IdrawingStrategy strategy)
{
this.event=event;
this.drawingstrategy = strategy;
}
}
The Event reference can go to the strategy or to the sprorting event depending on where it is needed.

Categories

Resources