I have a generic class definition similar to this:
public sealed class MyClass<TProperty,TOwner>
{
...
}
Now I'd like any instances of MyClass<TProperty,TOwner> regardless of the types of TProperty or TOwner to share a Hashtable. I thought of creating an internal MyClassBase with a protected internal static field of type Hashtable and inherit from that. I really only want the definition of MyClass to have access to this hashtable.
Is this a sound approach? I can't seal MyClassBase, so this probably could lead to opening a can of worms later on...
Are there other approaches for this?
The protected static Hashtable is a fine approach (make sure you synchronize it). You can't have an internal base class whose derived classes are public - but you can prevent it from being derived outside your assembly by making its default constructor internal:
public abstract class MyBaseClass
{
internal MyBaseClass() { }
private static Hashtable _hashtable;
protected static Hashtable Hashtable
{
get
{
if(_hashtable == null)
{
_hashtable = Hashtable.Synchronized(new Hashtable());
}
return _hashtable;
}
}
}
Another option is to make an internal static class with an exposed member, and only access it from MyClass<T,U>. Sure, it'd be visible within your assembly, but that's easily controlled compared to a protected member. Just make sure to only use it in your MyClass<T,U> and it'll be fine.
Related
I have a scenario that (simplified) is as follows:
public static class Example
{
public const int CONSTANT_VALUE = 1;
public static Example<T> Create<T>(IEnumerable<T> argument)
where T : class
{
return new Example<T>(argument);
}
//More overloads of Create<T>, possibly other static methods, etc..
}
public class Example<T>
where T : class
{
public Example(IEnumerable<T> argument)
{
//Do Stuff
//Nothing like this in the real code, just example
//that constants are used from the other class.
if (something == Example.CONSTANT_VALUE)
{
//Do A Thing
}
}
//Lots more code
}
The basic idea is that I can have static methods, constants, etc. available through the name of the class through the static class, while the actual implementation is in the type-argumented non-static class.
My question is whether or not this is a good way to set this up. Is there a way to put some static methods and constants that don't care what the type argument is on Example<T>? Is there otherwise a more recommended pattern? What I have works fine, but I wanted to know if there are other ways since this is the first time I've ended up doing something like this (not that it's conceptually new to me, just never had need).
This would only make sense if the constants are public. If they are only for internal use inside Example<T> then this is pointless, becuase you can reference them without a fully qualified name.
If the constants are of public use, I wouldn't use this pattern anayways; Example and Example<T> are two different classes, it is potentially confusing to any user, and not immeadiately obvious, that constants defined in the non generic class are aplicable to the generic one.
You are only avoding the user a few keystrokes, I'm not really sure it is worth it.
Update: other options
In this scenario, I'd use the following factory pattern (assuming the users are outside your assembly)
public class Example<T>
{
internal Example() { } //disallow users from instantiating this class
...
}
public static class Example
{
public const int Constant = ...
public static Example<T> Create<T>() { return new ... }
}
And now all users will interact only with Example and avoid using Example<T>. You could even enforce this with users of your own assembly, you'd just need to make Example<T> a private nested class implementing a public interface:
public interface IExample<T>
{
...
}
public static class Example
{
private class Example<T>: IExample<T> { ... }
public static IExample<T> Create<T>() { ... }
....
}
Unless there is a reason this wouldn't work in your case, I would prefer to use a non-static base class Example, and then let Example<T> inherit from this class. That way you get direct access to all the methods in Example, without having to qualify with the name. Of course, this assumes that the Example class is exclusively to be used in connection with the various typed classes Example<T>.
Is there any use of declaring a static class as private.Here is the code below:
static class DerivedClass
{
private static string GetVal()
{
return "Hello";
}
}
The sample code you provided actually illustrates an internal class, not a private class. This is perfectly fine and is done all the time. It means the methods of the class are available from other classes within the same module, but not externally.
If you mean declaring private members of static classes, sure there is.
static class DerivedClass
{
public static string GetVal()
{
return GetValInternal();
}
private static string GetValInternal()
{
return "Hello";
}
}
If you mean declaring a private static nested classes (because only nested classes can be private, according to the documentation), then you can do it, but there's probably no reason to do it.
class SomeClass
{
private static class DerivedClass
{
public static string GetVal()
{
return "Hello";
}
}
}
Is equivalent to
class SomeClass
{
private static string GetVal()
{
return "Hello";
}
}
By default classes with no access modifiers like in your example are internal, not private. See this reference: http://msdn.microsoft.com/en-us/library/ms173121.aspx. This means that you can access this class from anywhere inside the library/project. This makes sense because it allows you to use the class internally without necessarily exposing it to the outside world.
Explicitly declaring it as private however makes sense in some rare cases only in my opinion. I have used it before for nested classes simply to group certain things together and make my code prettier/more readable. However I find that if I am creating nested classes it usually means that I need to redesign my code and pull some of it into separate files and separate classes. Rather try to stick to one class per file.
Here is a piece of code:
private class myClass
{
public static void Main()
{
}
}
'or'
private class myClass
{
public void method()
{
}
}
I know, first one will not work. And second one will.
But why first is not working? Is there any specific reason for it?
Actually looking for a solution in this perspective, thats why made it bold. Sorry
It would be meaningful in this scenario; you have a public class SomeClass, inside which you want to encapsulate some functionality that is only relevant to SomeClass. You could do this by declaring a private class (SomePrivateClass in my example) within SomeClass, as shown below.
public class SomeClass
{
private class SomePrivateClass
{
public void DoSomething()
{
}
}
// Only SomeClass has access to SomePrivateClass,
// and can access its public methods, properties etc
}
This holds true regardless of whether SomePrivateClass is static, or contains public static methods.
I would call this a nested class, and it is explored in another StackOverflow thread.
Richard Ev gave a use case of access inside a nested classes. Another use case for nested classes is private implementation of a public interface:
public class MySpecialCollection<T> : IEnumerable<T>
{
public IEnumerator<T> GetEnumerator()
{
return new MySpecialEnumerator(...);
}
private class MySpecialEnumerator : IEnumerator<T>
{
public bool MoveNext() { ... }
public T Current
{
get { return ...; }
}
// etc...
}
}
This allows one to provide a private (or protected or internal) implementation of a public interface or base class. The consumer need not know nor care about the concrete implementation. This can also be done without nested classes by having the MySpecialEnumerator class be internal, as you cannot have non-nested private classes.
The BCL uses non-public implementations extensively. For example, objects returned by LINQ operators are non-public classes that implement IEnumerable<T>.
This code is syntactically correct. But the big question is: is it useful, or at least usable in the context where you want to use it? Probably not, since the Main method must be in a public class.
Main() method is where application execution begin, so the reason you cannot compile your first class (with public static void Main()) is because you already have Main method somewhere else in your application. The compiler don't know where to begin execute your application.
Your application must have only one Main method to compile with default behavior otherwise you need to add /main option when you compile it.
Given the following:
public abstract class Base
{
// other stuff
public static void StaticMethod()
{
PrivateMethod();
}
// here should be PrivateMethod() declaration somehow
}
public sealed class Derived: Base
{
// other stuff
public void InstanceMethod()
{
// call somehow PrivateMethod
PrivateMethod();
}
}
I need to make use of PrivateMethod() from 2 different contexts (different assemblies). Once calling Base.StaticMethod(), and the second time by using an instance of the Derived class d.InstanceMethod();.
I am looking for a way how to design PrivateMethod() inside the Base class. Of course PrivateMethod() should not be visible outside the Base and Derived classes.
I was thinking something about "protected static PrivateMethod() {}" but I read I should not do that...
What do you recommend guys?
protected static void PrivateMethod() {}
Is OK (apart form the name) and does what you require. You won't need base. when calling it from Derived.
I had never heard this before, so I went looking for something that said what you described. I found this article: New Design Guideline: Avoid Protected Static. However, it only talks about protected static field.
I don't think the article actually makes a good case for what it is trying to say. Rather than just describing how protected statics can lead to complications, it uses a very simple example of the base class designer not setting the right access flags for something that should not be accessed by everyone.
That being said, there is still a point that protected static can lead to complications. Protected static means that any subclass can call a method at any time. This can lead to thread safety concerns if the method is written naively. It seems like the article was written in a way that it conveys "Don't do it" rather than "If you need to do it, be careful."
You could just call the public StaticMethod() from your derived class's InstanceMethod() ... since it indirects back to PrivateMethod() anyway. That way you can leave PrivateMethod() private. The implementation would be something like:
public abstract class Base
{
// other stuff
public static void StaticMethod()
{
PrivateMethod();
}
// here should be PrivateMethod() declaration somehow
private static void PrivateMethod()
{
// do stuff
}
}
public sealed class Derived: Base
{
// other stuff
public void InstanceMethod()
{
// call somehow PrivateMethod
StaticMethod();
}
}
PS: If there is need during StaticMethod to differentiate between a public caller or a derived class caller (from InstanceMethod) it could be either passed as parameter, or determined via reflection.
What are all the difference between an abstract class, and a class with only protected constructor(s)? They seem to be pretty similar to me, in that you can't instantiate either one.
EDIT:
How would you create an instance in a derived class, with a base class with a protected constructor? For instance:
public class ProtectedConstructor
{
protected ProtectedConstructor()
{
}
public static ProtectedConstructor GetInstance()
{
return new ProtectedConstructor(); // this is fine
}
}
public class DerivedClass : ProtectedConstructor
{
public void createInstance()
{
ProtectedConstructor p = new ProtectedConstructor(); // doesn't compile
}
public static ProtectedConstructor getInstance()
{
return new ProtectedConstructor(); // doesn't compile
}
}
You can instantiate a class with protected constructors from within the class itself - in a static constructor or static method. This can be used to implement a singleton, or a factory-type thing.
An abstract class cannot be instantiated at all - the intent is that one or more child classes will complete the implementation, and those classes will get instantiated
Edit:
if you call ProtectedConstructor.GetInstance(); instead of new ProtectedConstructor();, it works. Maybe protected constructors can't be called this way? But protected methods certainly can.
Here is an interesting article on the topic.
Most of the time, there is little practical difference, as both are only able to be generated via a subclass.
However, marking a class abstract has two benefits:
With protected constructors, it's still possible to create an instance of the class in two ways. You can use Activator.CreateInstance with BindingFlags.NonPublic, or you can use a factory method defined in the class (or a subclass) to create an instance of the class. A class marked abstract, however, cannot be created.
You are making your intention more clear by marking the class abstract. Personally, I find this the most compelling reason to do so.
From an outside , black-box perspective, yes they are similar in that you cannot instantiate either one. However, you can never instantiate an abstract class, where you can construct a class with only protected constructors from within the class itself, or from an inheritor.
An abstract class can have abstract methods; methods that consist only of the method signature, but no body, that child classes must implement.
Seriously, not one person mentioned that yet?
Your example is flawed because in the getInstance case because you construct a ProtectedConstructor class and expect to down cast it as a DerivedClass. Instead you need a slightly more complete implementation where the derived class has a constrcutor:
public class ProtectedConstructor
{
protected ProtectedConstructor(string arg)
{
// do something with arg
}
public static ProtectedConstructor GetInstance()
{
return new ProtectedConstructor("test");
}
}
public class DerivedClass : ProtectedConstructor
{
protected DerivedClass(string arg) : base(arg)
{
}
public void createInstance()
{
DerivedClass p = new DerivedClass("test");
}
public static DerivedClass getInstance()
{
return new DerivedClass("test");
}
}
Regardless the major difference usage of abstract classes is to define abstract methods that subclasses must implement but you don't want to provide a default implementation for. For example suppose you have some kind of Thread class that has a Run method. You want to ensure that every call to Run first setups up some logging then does the real work of the thread and then stops logging. You could write an abstract Thread class like this:
public abstract Thread
{
protected Thread()
{
}
public void Run()
{
LogStart();
DoRun();
LogEnd();
}
protected abstract DoRun();
private void LogStart()
{
Console.Write("Starting Thread Run");
}
private void LogEnd()
{
Console.Write("Ending Thread Run");
}
}
public class HelloWorldThread : Thread
{
public HelloWorldThread()
{
}
protected override DoRun()
{
Console.Write("Hello World");
}
}
Another thing to consider, that I didn't see other people mention, is that your code may be maintained in the future. If the maintainer adds a public constructor to a class, then it can be instantiated. This might break your design, so you should prevent it (or design to accommodate it).
To prevent other people from making these kinds of changes, you can comment your code. Or, as other people said, use "abstract" to explicitly document your intent.
Well, the first difference that comes to mind is that an abstract class can not be instantiated, but a class with protected constructors could be instantiated throw another public method.
A common example of this might be something like the Singleton pattern: http://en.wikipedia.org/wiki/Singleton_pattern
if you inherit an abstract class from another abstract class, you do not have to satisfy abstract methods, but you do with a normal class with protected ctors. Examples
public abstract class Parent
{
protected abstract void AMethod();
}
public abstract class Child: Parent
{
// does not implement AMethod, and that's ok
}
public class Child2: Parent
{
// does not implement AMethod, and that will cause a compile error
}
If your intent is to only allow static uses of the class (i.e. not to use it as a pure base class) then you should use the static keyword instead; the CLR will prevent instances of the class being created via any method including Reflection (AFAIK).