WeakReference - Am I doing it right? - c#

I have a static class, which exposes a event:
public static class MyStaticClass
{
static bool myBool= false;
public static bool MyBool
{
get { return myBool; }
private set
{
myBool= value;
var handler = MyBoolChanged;
if (handler != null)
handler(null, null);
}
}
public static event EventHandler MyBoolChanged;
}
And then I am registering to it using this pattern:
class AnotherClass
{
WeakReference _me;
public MyMethodInAnotherClass()
{
_me = new WeakReference(this);
MyStaticClass.MyBoolChanged+=
(_me.Target as AnotherClass).MyMethodInAnotherClassCallback;
}
private void MyMethodInAnotherClassCallback(some arguments)
{
}
}
What I want to achieve is that MyStaticClass will only execute the handler if the instance of AnotherClass has not been disposed (and has not deregistered).

The best way I can see to use this is to forget about an event, and use some kind of list instead; let's say List<WeakReference>; you could then have:
interface IFoo {
void Bar(some args);
}
with:
static class Whatever {
private static readonly List<WeakReference> items=new List<WeakReference>();
public static void Add(IFoo foo) {
if(foo != null) {
var newRef = new WeakReference(foo);
lock(items) { items.Add(newRef); }
}
}
public static void DoIt(some args) {
lock(items) {
foreach(var item in items) {
IFoo foo = item.IsAlive ? item.Target as IFoo : null;
if(foo != null) foo.Bar(some args);
}
}
}
}
with additional mechanisms to remove a specific IFoo, and to remove all dead foos left todo.
Then you just need AnotherClass : IFoo, with a Bar() implementation that applies your callback.
Additional emphasis: static collections (including events) are fairly dangerous; you must have some kind of sweep occasionally to remove empty items, and try to unsubscribe promptly where possible (in Dispose(), for example). As an illustration:
public static void Remove(IFoo foo) {
lock (items) { // also remove any dead debris
items.RemoveAll(x => !x.IsAlive || x.Target == foo || x.Target == null);
}
}

Related

C#: Share properties, events, and methods across unrelated classes

I need to add the following to several unrelated classes:
private MyClass myclass;
private EventHandler clicked;
public event EventHandler Clicked { ... }
private bool enabled;
public bool Enabled { ... }
private void HandleClicked(object sender, EventArgs e) { ... }
The problem is these classes are third-party and do not necessarily share the same immediate base class though they all eventually inherit from a class called View. Right now, I end up creating my own subclasses for each and copy-pasting the same code which leads to unnecessary duplication.
Any way to meaningfully refactor this?
One of the way is to use composition. Create class which will store all new events\properties\methods:
public class Properties
{
private MyClass myclass;
private EventHandler clicked;
public event EventHandler Clicked { ... }
private bool enabled;
public bool Enabled { ... }
private void HandleClicked(object sender, EventArgs e) { ... }
}
Then use Extension methods to expand required interface (i.e. classA)
public static class NewInterfaces
{
public static Properties Props(this classA)
{ /* lookup required properties, from some associative storage */ }
}
Usage will look like:
var inst = new classA();
inst.Prop.Enabled = !inst.Prop.Enabled;
Second way it still composition, but you will use wrapper for those:
public class Wrapper
{
private object _wrapped;
public Wrapper(classA obj)
{
_wrapped = obj;
}
public Wrapper(classB obj)
{
_wrapped = obj;
}
public int WrappedProperty
{
get
{
var instA = _wrapped as classA;
if (instA != null)
return instA.SomeProperty1;
var instB = _wrapped as classB;
if (instB != null)
return instB.SomeProperty2;
}
}
private MyClass myclass;
private EventHandler clicked;
public event EventHandler Clicked { ... }
private bool enabled;
public bool Enabled { ... }
private void HandleClicked(object sender, EventArgs e) { ... }
}
Second way allow you to create new hierarchy of wrapper which will contain elements without common base class.
Inheritance becomes problematic in time. I recommend using interfaces instead, you will have much more flexibility.
public interface INewInterfaces
{
event EventHandler Clicked;
bool Enabled { get; }
void HandleClicked(object sender, EventArgs e);
}
public class NewClassA : ClassA, INewInterfaces
{
//...
}
public class NewClassB : ClassB, INewInterfaces
{
//...
}
Edit 1:
If you are saying that ClassX's are very similar and you want to use the same HandleClicked implementation in all of these unrelated classes, you may use two other approaches.
1- Still inheritance
Create an interface and add all the common functions across the classes you want to use. This will put the ClassX's in the same family. And then create a class for general use.
public interface IExistingInterfaces
{
void SomeMethod();
}
public class NewClassA : ClassA, IExistingInterfaces
{
//Do nothing
}
public class NewClassB : ClassB, IExistingInterfaces
{
//Do nothing
}
public class MyClassForGeneralUse : IExistingInterfaces
{
private IExistingInterfaces _baseObject;
public MyClassForGeneralUse(IExistingInterfaces baseObject)
{
_baseObject = baseObject;
}
//Write proxy calls for IExistingInterfaces
public void SomeMethod()
{
_baseObject.SomeMethod();
}
//Add new methods here
public void HandleClicked(object sender, EventArgs e)
{
}
//...
//...
}
Not: The first part is Bridge Pattern and the second part is Decorator Pattern
2- Reflection
var propertyInfo = someObject.GetType().GetProperty("property name");
if (propertyInfo == null)
throw new Exception(string.Format("Property does not exist:{0}", condition.Property));
var propertyValue = propertyInfo.GetValue(someObject, null);
long longValue = (long)propertyValue;
//You can get methods in a smilar manner and execute with
result = methodInfo.Invoke(methodInfo, parametersArray);
But reflection may be overkill.

How to properly dispose objects: injected vs. owned

I have a question about disposing objects.
Consider this IDisposable class
public class MyClass : DisposableParentClass
{
private MyProp _prop;
public MyClass(MyProp prop)
{
_prop = prop;
}
public MyClass()
{
_prop = new MyProp();
}
protected override void Dispose(bool disposing)
{
if (disposing)
{
_prop.Dispose();
}
base.Dispose(disposing);
}
}
On the first constructor, MyProp is injected. So MyClass is not the owner of the object. But on the second constructor, MyProp is created locally.
Should I always dispose MyProp, or should I check first if it is injected or not.
public class MyClass : DisposableParentClass
{
private MyProp _prop;
private bool _myPropInjected = false;
public MyClass(MyProp prop)
{
_prop = prop;
_myPropInjected = true;
}
public MyClass()
{
_prop = new MyProp();
}
protected override void Dispose(bool disposing)
{
if (disposing)
{
if (!_myPropInjected) { _prop.Dispose(); }
}
base.Dispose(disposing);
}
}
If your class should handle these two situations:
It is not the owner of the provided object, it should not dispose of it
It is the owner of the created object, it should dispose of it
Then yes, you need to have a mechanism that tells these two situations apart.
A common method (common to me anyway) is to use naming convention like this:
private MyProp _prop;
private bool _ownsProp = false;
ie. reverse the meaning of your flags, but this is details, your solution is just fine, and yes, you need to have a solution like this.
If you have a ton of these fields, where each must have its own bool field to handle this, it might be worth creating a helper class, such as this LINQPad program demonstrates:
void Main()
{
Injectable i1 = new Injectable();
Injectable i2 = new Injectable(new Injected("A"));
Injectable i3 = new Injectable(new Injected("A"), new Injected("B"));
Debug.WriteLine("dispose a and b");
i1.Dispose();
Debug.WriteLine("dispose b");
i2.Dispose();
Debug.WriteLine("no dispose");
i3.Dispose();
}
public class Injected : IDisposable
{
public Injected(string name) { Name = name; }
public string Name { get; set; }
public void Dispose() { Debug.WriteLine(Name + " disposed"); }
}
public class Injectable : IDisposable
{
private Ownable<Injected> _A;
private Ownable<Injected> _B;
public Injectable(Injected a, Injected b)
{
_A = Ownable.NotOwned(a);
_B = Ownable.NotOwned(b);
}
public Injectable(Injected a)
{
_A = Ownable.NotOwned(a);
_B = Ownable.Owned(new Injected("B"));
}
public Injectable()
{
_A = Ownable.Owned(new Injected("A"));
_B = Ownable.Owned(new Injected("B"));
}
public void Dispose()
{
_A.Dispose();
_B.Dispose();
}
}
public class Ownable<T> : IDisposable
where T : class
{
private readonly T _Instance;
private readonly Action _CleanupAction;
public Ownable(T instance, bool isOwned)
{
_Instance = instance;
if (isOwned)
{
IDisposable disposable = instance as IDisposable;
if (disposable == null)
throw new NotSupportedException("Unable to clean up owned object, does not implement IDisposable");
_CleanupAction = () => disposable.Dispose();
}
}
public Ownable(T instance, Action cleanupAction)
{
_Instance = instance;
_CleanupAction = cleanupAction;
}
public T Instance { get { return _Instance; } }
public void Dispose()
{
if (_CleanupAction != null)
_CleanupAction();
}
}
public static class Ownable
{
public static Ownable<T> Owned<T>(T instance)
where T : class
{
return new Ownable<T>(instance, true);
}
public static Ownable<T> Owned<T>(T instance, Action cleanupAction)
where T : class
{
return new Ownable<T>(instance, cleanupAction);
}
public static Ownable<T> NotOwned<T>(T instance)
where T : class
{
return new Ownable<T>(instance, false);
}
}
A different note can be made here either.
It depends on what is your MyClass is doing actually.
For example, if we are talking about a class that reads video stream from device, after applies some filters to it and writes data to a user specified file, where file writing is made by stream passed from the outside, say like this:
public class VideoProcessor : IDisposable {
private FileStream _videoFile = null;
private VideoProcessor() {}
//user specified FileStream
public VideoProcessor(FileStream fs) {_videoFile = fs;}
public void Dispose() {
_videoFile.Dispose(); //Dispose user passed FileStream
}
}
disposing passed stream object during dispose call, makes actually sence.
In other cases, yes, it's better to not destroy object, if you are not an owner of it. Leave it to the caller to decide when it is appropriate time to do that.

Passing dependency into Factory

We are using factory to create an instance of Subscribers. Each subscriber can have its own dependency.
Each subscriber will use constructor injection.
Should I pass dependency into subscribers through Subscriber Factory? Every time adding new dependency in any subscriber will change Subscriber factory?
public interface IMessageSubscriber
{
bool Process(string message)
}
public class MessageSubscriber1 : IMessageSubscriber
{
public bool Process(string message)
{
//Some custom logic
}
}
public class MessageSubscriber2 : IMessageSubscriber
{
public bool Process(string message)
{
//Some custom logic
}
}
public class MessageSubscriberFactory
{
//SubscriberType is enum
public IMessageSubscriber Get(SubscriberType type)
{
if(type == 1)
{
return new MessageSubscriber1();
}
else if(type == 2)
{
return new MessageSubscriber2();
}
}
}
//Main class
public class Process
{
public static void Main(string[] args)
{
MessageSubscriberFactory fac = new MessageSubscriberFactory();
foreach SubscriberType
{
string = "Message";
IMessageSubscriber subscriber = fac.Get(type);
subscriber.Process(message)
}
}
}
One approach would be to use named registrations with a DI/IOC container. This would involve using the container in a service locator fashion (which some people oppose), but I think it could make sense in this case. The example below is pretty crude, but it does give you an approach to handle subscribers with different dependencies without passing them into the factory. I used Unity here and you'd want to wrap the container reference rather than referencing directly, but this gets the point across.
public interface ILowerCaseWriter
{
void Write(string message);
}
public class LowerCaseWriter : ILowerCaseWriter
{
public void Write(string message)
{
Console.WriteLine(message.ToLower());
}
}
public interface IUpperCaseWriter
{
void Write(string message, int number);
}
public class UpperCaseWriter : IUpperCaseWriter
{
public void Write(string message, int number)
{
Console.WriteLine("{0}:{1}", number, message.ToUpper());
}
}
public interface ISubscriber
{
void Write();
}
public class Subscriber1 : ISubscriber
{
private ILowerCaseWriter _writer;
public Subscriber1(ILowerCaseWriter writer)
{
_writer = writer;
}
public void Write()
{
_writer.Write("Using subscriber 1");
}
}
public class Subscriber2 : ISubscriber
{
private IUpperCaseWriter _writer;
public Subscriber2(IUpperCaseWriter writer)
{
_writer = writer;
}
public void Write()
{
_writer.Write("Using subscriber 2", 2);
}
}
public class SubscriberFactory
{
private UnityContainer _container;
public SubscriberFactory()
{
_container = new UnityContainer();
_container.RegisterType<ILowerCaseWriter, LowerCaseWriter>();
_container.RegisterType<IUpperCaseWriter, UpperCaseWriter>();
_container.RegisterType<ISubscriber, Subscriber1>("Subscriber1");
_container.RegisterType<ISubscriber, Subscriber2>("Subscriber2");
}
public ISubscriber GetSubscriber(int type)
{
switch (type)
{
case 1:
return _container.Resolve<ISubscriber>("Subscriber1");
case 2:
return _container.Resolve<ISubscriber>("Subscriber2");
default:
throw new Exception();
}
}
}
class Program
{
private static void Main(string[] args)
{
var factory = new SubscriberFactory();
var subscriber = factory.GetSubscriber(1);
subscriber.Write();
Console.ReadLine();
}
}

RegisterCallback<T>(Action<T> func) , how do I store this function pointer in a class?

I'm trying to expose an API such that, I do the following
RegisterCallback<T>(Action<T> func)
{
someObj.FuncPointer = func;
}
Later on, I call func(obj) .. and the obj is of type T that the user said.
More concrete example:
var callbackRegistrar = new CBRegistrar();
callbackRegistrar.RegisterCallback<ISomeClass>(SomeFunc);
public static void SomeFunc(ISomeClass data)
{
//
}
EDIT: So I may not have been clear, so I'll add more code:
I want to make only "one" object of CBRegistrar, and connect it with many Callbacks, as such:
var callbackRegistrar = new CBRegistrar();
callbackRegistrar.RegisterCallback<ISomeClass>(SomeFunc);
callbackRegistrar.RegisterCallback<ISomeOtherClass>(SomeFunc2);
...
In fact the above code is called by reflecting over a directory of plugins.
The user puts this in their code -->
public static void SomeFunc(ISomeClass data)
{
//
}
public static void SumFunc2(ISomeOtherClass data)
{
//
}
It looks to me as if this is not possible using Generics, etc. What it looks like I might have to do is make an interface called IPlugin or something, and ask the user to do this ..
[PluginIdentifier(typeof(ISomeClass))]
public static void SomeFunc(IPluginData data)
{
var castedStuff = data as ISomeClass; // ISomeClass inherits from IPluginData
}
Seems like asking the user to do stuff that we should take care of, but anyway ...
You need a Action<T> func to store it in. There is a semantic check to make here: if someone calls RegisterCallback twice (with different values), do you want to replace the callback, or keep both ? Assuming the latter, someObj probably wants an event (indeed, this entire API could be exposed as an event), so - in the someObj class:
public event Action<T> FuncPointer;
private void InvokeCallback(T data) {
var handler = FuncPointer;
if(handler != null) handler(data);
}
Noting that RegisterCallback could be replaced entirely, still keeping the data on obj:
public event Action<T> Completed {
add { obj.FuncPointer += value; }
remove { obj.FuncPointer -= value; }
}
Then usage would be:
var callbackRegistrar = new CBRegistrar();
callbackRegistrar.Completed += SomeFunc;
Callback functions are not much used in C#. They've been replaced by events which are more elegant and easier to work with.
class CBRegistrar
{
public delegate void ActionRequiredEventHandler(object sender, ISomeClass e);
public event ActionRequiredEventHandler ActionRequired;
void RaiseActionRequiredEvent(ISomeClass parm)
{
if ( ActionRequired != null)
{
ActionRequired(this, parm);
}
}
}
class APIConsumer
{
var callbackRegistrar = new CBRegistrar();
public APIConsumer()
{
callbackRegistrar.ActionRequired += SomeFunc;
}
public void SomeFunc(object sender, ISomeClass data)
{
}
}
If you still want to use Callbacks, you can use Delegates which are more or less function pointer.
The CBRegistrar will need to be generic (if it's OK to keep a single callback type) or it can do some internal casting (if several callback types need to be registered).
public class CBRegistrar<T>
{
private Action<T> callback;
private Dictionary<Type, object> callbackMap;
public CBRegistrar()
{
this.callbackMap = new Dictionary<Type, object>();
}
public void RegisterCallback(Action<T> func)
{
this.callback = func;
}
public void RegisterGenericCallback<U>(Action<U> func)
{
this.callbackMap[typeof(U)] = func;
}
public Action<U> GetCallback<U>()
{
return this.callbackMap[typeof(U)] as Action<U>;
}
}
public interface ISomeClass
{
string GetName();
}
public class SomeClass : ISomeClass
{
public string GetName()
{
return this.GetType().Name;
}
}
namespace ConsoleApplication
{
class Program
{
static void Main(string[] args)
{
var callbackRegistrar = new CBRegistrar<ISomeClass>();
callbackRegistrar.RegisterCallback(SomeFunc);
callbackRegistrar.RegisterGenericCallback<ISomeClass>(SomeFunc);
var someone = new SomeClass();
callbackRegistrar.GetCallback<ISomeClass>()(someone);
}
public static void SomeFunc(ISomeClass data)
{
// Do something
Console.WriteLine(data.GetName());
}
}
}

Implementing Observers and Subjects using IObserver/IObservable

I want to create a class that can be used to represent a dynamically computed value, and another class that represents a value can be the source (subject) for these dynamically computed values. The goal is that when the subject changes, the computed value is updated automatically.
It seems to me that using IObservable/IObserver is the way to go. Unfortunately I can't use the Reactive Extensions library, so I am forced to implement the subject/observer pattern from scratch.
Enough blabla, here are my classes:
public class Notifier<T> : IObservable<T>
{
public Notifier();
public IDisposable Subscribe(IObserver<T> observer);
public void Subscribe(Action<T> action);
public void Notify(T subject);
public void EndTransmission();
}
public class Observer<T> : IObserver<T>, IDisposable
{
public Observer(Action<T> action);
public void Subscribe(Notifier<T> tracker);
public void Unsubscribe();
public void OnCompleted();
public void OnError(Exception error);
public void OnNext(T value);
public void Dispose();
}
public class ObservableValue<T> : Notifier<T>
{
public T Get();
public void Set(T x);
}
public class ComputedValue<T>
{
public T Get();
public void Set(T x);
}
My implementation is lifted mostly from: http://msdn.microsoft.com/en-us/library/dd990377.aspx.
So what would the "right" way to do this be? Note: I don't care about LINQ or multi-threading or even performance. I just want it to be simple and easy to understand.
If I were you I would try to implement your classes as closely as possible to the way Rx has been implemented.
One of the key underlying principles is the use of relatively few concrete classes that are combined using a large number of operations. So you should create a few basic building blocks and use composition to bring them all together.
There are two classes I would take an initial look at under Reflector.NET: AnonymousObservable<T> & AnonymousObserver<T>. In particular AnonymousObservable<T> is used through-out Rx as the basis for instantiating observables. In fact, if you look at the objects that derive from IObservable<T> there are a few specialized implementations, but only AnonymousObservable<T> is for general purpose use.
The static method Observable.Create<T>() is essentially a wrapper to AnonymousObservable<T>.
The other Rx class that is clearly a fit for your requirements is BehaviorSubject<T>. Subjects are both observables and observers and BehaviorSubject fits your situation because it remembers the last value that is received.
Given these basic classes then you almost have all of the bits you need to create your specific objects. Your objects shouldn't inherit from the above code, but instead use composition to bring together the behaviour that you need.
Now, I would suggest some changes to your class designs to make them more compatible with Rx and thus more composible and robust.
I would drop your Notifier<T> class in favour of using BehaviourSubject<T>.
I would drop your Observer<T> class in favour of using AnonymousObserver<T>.
Then I would modify ObservableValue<T> to look like this:
public class ObservableValue<T> : IObservable<T>, IDisposable
{
public ObservableValue(T initial) { ... }
public T Value { get; set; }
public IDisposable Subscribe(IObserver<T> observer);
public void Dispose();
}
The implementation of ObservableValue<T> would wrap BehaviourSubject<T> rather than inherit from it as exposing the IObserver<T> members would allow access to OnCompleted & OnError which wouldn't make too much sense since this class represents a value and not a computation. Subscriptions would use AnonymousObservable<T> and Dispose would clean up the wrapped BehaviourSubject<T>.
Then I would modify ComputedValue<T> to look like this:
public class ComputedValue<T> : IObservable<T>, IDisposable
{
public ComputedValue(IObservable<T> source) { ... }
public T Value { get; }
public IDisposable Subscribe(IObserver<T> observer);
public void Dispose();
}
The ComputedValue<T> class would wrap AnonymousObservable<T> for all subscribers and and use source to grab a local copy of the values for the Value property. The Dispose method would be used to unsubscribe from the source observable.
These last two classes are the only real specific implementation your design appears to need - and that's only because of the Value property.
Next you need a static ObservableValues class for your extension methods:
public static class ObservableValues
{
public static ObservableValue<T> Create<T>(T initial)
{ ... }
public static ComputedValue<V> Compute<T, U, V>(
this IObservable<T> left,
IObservable<U> right,
Func<T, U, V> computation)
{ ... }
}
The Compute method would use an AnonymousObservable<V> to perform the computation and produce an IObservable<V> to pass to the constructor of ComputedValue<V> that is returned by the method.
With all this in place you can now write this code:
var ov1 = ObservableValues.Create(1);
var ov2 = ObservableValues.Create(2);
var ov3 = ObservableValues.Create(3);
var cv1 = ov1.Compute(ov2, (x, y) => x + y);
var cv2 = ov3.Compute(cv1, (x, y) => x * y);
//cv2.Value == 9
ov1.Value = 2;
ov2.Value = 3;
ov3.Value = 4;
//cv2.Value == 20
Please let me know if this is helpful and/or if there is anything that I can elaborate on.
EDIT: Also need some disposables.
You'll also need to implement AnonymousDisposable & CompositeDisposable to manage your subscriptions particularly in the Compute extension method. Take a look at the Rx implementations using Reflector.NET or use my versions below.
public sealed class AnonymousDisposable : IDisposable
{
private readonly Action _action;
private int _disposed;
public AnonymousDisposable(Action action)
{
_action = action;
}
public void Dispose()
{
if (Interlocked.Exchange(ref _disposed, 1) == 0)
{
_action();
}
}
}
public sealed class CompositeDisposable : IEnumerable<IDisposable>, IDisposable
{
private readonly List<IDisposable> _disposables;
private bool _disposed;
public CompositeDisposable()
: this(new IDisposable[] { })
{ }
public CompositeDisposable(IEnumerable<IDisposable> disposables)
{
if (disposables == null) { throw new ArgumentNullException("disposables"); }
this._disposables = new List<IDisposable>(disposables);
}
public CompositeDisposable(params IDisposable[] disposables)
{
if (disposables == null) { throw new ArgumentNullException("disposables"); }
this._disposables = new List<IDisposable>(disposables);
}
public void Add(IDisposable disposable)
{
if (disposable == null) { throw new ArgumentNullException("disposable"); }
lock (_disposables)
{
if (_disposed)
{
disposable.Dispose();
}
else
{
_disposables.Add(disposable);
}
}
}
public IDisposable Add(Action action)
{
if (action == null) { throw new ArgumentNullException("action"); }
var disposable = new AnonymousDisposable(action);
this.Add(disposable);
return disposable;
}
public IDisposable Add<TDelegate>(Action<TDelegate> add, Action<TDelegate> remove, TDelegate handler)
{
if (add == null) { throw new ArgumentNullException("add"); }
if (remove == null) { throw new ArgumentNullException("remove"); }
if (handler == null) { throw new ArgumentNullException("handler"); }
add(handler);
return this.Add(() => remove(handler));
}
public void Clear()
{
lock (_disposables)
{
var disposables = _disposables.ToArray();
_disposables.Clear();
Array.ForEach(disposables, d => d.Dispose());
}
}
public void Dispose()
{
lock (_disposables)
{
if (!_disposed)
{
this.Clear();
}
_disposed = true;
}
}
public IEnumerator<IDisposable> GetEnumerator()
{
lock (_disposables)
{
return _disposables.ToArray().AsEnumerable().GetEnumerator();
}
}
IEnumerator IEnumerable.GetEnumerator()
{
return this.GetEnumerator();
}
public bool IsDisposed
{
get
{
return _disposed;
}
}
}

Categories

Resources