Develop Reference

C# 8 base interface's default method invocation workaround

According to https://learn.microsoft.com/en-us/dotnet/csharp/language-reference/proposals/csharp-8.0/default-interface-methods
It is possible to explicitly invoke an interface base implementation with the following syntax.
base(IInterfaceType).Method();
But this doesn't seem to be implemented yet.
Is there a workaround (e.g reflection) to achieve this?
Example code to illustrate the problem
interface IA
{
void M()
{
Console.WriteLine("IA.M");
}
}
interface IB : IA
{
void IA.M()
{
Console.WriteLine("IB.M");
}
}
interface IC : IA
{
void IA.M()
{
Console.WriteLine("IC.M");
}
}
class D : IA, IB, IC
{
public void M()
{
// base(IB).M(); Is not yet supported apparently
((IB)this).M(); // Throws stack overflow
}
}
class Program
{
static void Main(string[] args)
{
D d = new D();
d.M();
}
}

The link in the question points to a version of the proposal copied from the proposal document in Github
The feature was cut in April 2019
Conclusion
Cut base() syntax for C# 8. We intend to bring this back in the next major release.
The design meeting doc explains that without runtime support (which wouldn't be available in time), the implementation would be workable at best for C# but not VB.NET.
If B.M is not present at run time, A.M() will be called. For base() and interfaces, this is not supported by the runtime, so the call will throw an exception instead. We'd like to add support for this in the runtime, but it is too expensive to make this release.
We have some workarounds, but they do not have the behavior we want, and are not the preferred codegen. Our implementation for C# is somewhat workable, although not exactly what we would like, but the VB implementation would be much more difficult. Moreover, the implementation for VB would require the interface implementation methods to be public API surface.
As for the infinite recursion, this
public void M()
{
((IB)this).M(); // Throws stack overflow
}
That's essentially
public void M()
{
M(); // Throws stack overflow
}
Default interface members are called the same way explicitly implemented interface methods are, through the interface. Besides, you're asking to call the method on this, not base.

There is a workaround.
I got it working, using GetFunctionPointer
Warning do not use this code
static class BaseInterfaceInvocationExtension
{
private static readonly string invalidExpressionMessage = "Invalid expression.";
public static void Base<TInterface>(this TInterface owner, Expression<Action<TInterface>> selector)
{
if (selector.Body is MethodCallExpression methodCallExpression)
{
MethodInfo methodInfo = methodCallExpression.Method;
string name = methodInfo.DeclaringType.FullName + "." + methodInfo.Name;
Type type = owner.GetType();
InterfaceMapping interfaceMapping = type.GetInterfaceMap(typeof(TInterface));
var map = interfaceMapping;
var interfaceMethod = map.InterfaceMethods.First(info =>
info.Name == name);
var functionPointer = interfaceMethod.MethodHandle.GetFunctionPointer();
var x = methodCallExpression.Arguments.Select(expression =>
{
if (expression is ConstantExpression constantExpression)
{
return constantExpression.Value;
}
var lambda = Expression.Lambda(Expression.Convert(expression, expression.Type));
return lambda.Compile().DynamicInvoke();
}).ToArray();
Type actionType = null;
if (x.Length == 0)
{
actionType = typeof(Action);
}else if (x.Length == 1)
{
actionType = typeof(Action<>);
}
else if (x.Length == 2)
{
actionType = typeof(Action<,>);
}
var genericType = actionType.MakeGenericType(methodInfo.GetParameters().Select(t => t.ParameterType).ToArray());
var instance = Activator.CreateInstance(genericType, owner, functionPointer);
instance.GetType().GetMethod("Invoke").Invoke(instance, x);
}
else
{
throw new Exception(invalidExpressionMessage);
}
}
}
class D : IA, IB, IC
{
public void M(int test)
{
this.Base<IB>(d => d.M(test));
}
}
class Program
{
static void Main(string[] args)
{
D d = new D();
d.M(12);
Console.ReadKey();
}
}

This is a workaround. It isn't ideal. Perhaps it will help someone.
class C : IB
{
public void IBM() => (this as IB).M();
}
class D : IA, IB, IC
{
private C _c = new C();
public void M()
{
_c.IBM();
}
}

Alternatively, this could work:
...
interface IB : IA
{
void IA.M()
{
IB_M();
}
void IB_M()
{
Console.WriteLine("IB.M");
}
}
...
class D : IA, IB, IC
{
public void M()
{
(this as IB).IB_M();
}
}
Not ideal either but less verbose.

Generic method with dynamic T

I'm making a small example to check the type of parameter is valid or not.
class A
{
}
class B
{
}
class C
{
}
class D
{
public void SomeMethod<T>(T t) where T : class
{
if (t is A)
{
A a = t as A;
}
else if (t is B)
{
B b = t as B;
}
}
}
Then, I can call:
A a = new A();
SomeMethod<A>(a);
B b = new B();
SomeMethod<B>(b);
Now, I want to prevent to passing class C to SomeMethod. What I want to achieve:
C c = new C();
SomeMethod<C>(c); // error
To do that, I've tried:
public void SomeMethod<T>(T t) where T : A
{
// accept only class A
}
or
public void SomeMethod<T>(T t) where T : B
{
// accept only class B
}
My question is: how to declare SomeMethod with T can be A or B at the same time? Just like:
public void SomeMethod<T>(T t) where T : A, B
{
// accept class A and class B
}

As Lee has mentioned, this defeats the purpose of generics. To ahieve what you're describing just write overloads for each case
class A { }
class B { }
class C { }
class D
{
public void SomeMethod(A a)
{
//Do stuff with a
}
public void SomeMethod(B b)
{
//Do stuff with b
}
}
If you wanted to have a run-time error you could do something like this:
class A { }
class B { }
class C { }
class D
{
public void SomeMethod<T>(T t) where T : class
{
if (t is A)
{
A a = t as A;
}
else if (t is B)
{
B b = t as B;
}
else //if (t is C)
{
throw new ArgumentException();
}
}
}
Though this is a poor solution. The overload solution is still cleaner and will give a compile-time error.

It seems like really bad practice but I think you could do
class D
{
public void SomeMethod<T>(T t) where T : class
{
if (t is A)
A a = t as A;
else if (t is B)
B b = t as B;
else
throw new Exception("Wrong class type.");
}
}
This way you can use the method with only classes A and B, and it will throw an error for class C - and others.

Expose Methods in Properties as Single Class

I needed to break up a WCF service contract that had a massive interface and clientbase class into smaller classes. All of the smaller classes are similar but have different operation contracts. I want to be able to expose the operation contract methods in all the new sub-classes as a single class for backwards compatibility. Ideally it would look something like this:
public class MainClient {
public MainClient() {
Sub1 = new Sub1Client();
Sub2 = new Sub2Client();
}
public static Sub1Client Sub1;
public static Sub2Client Sub2;
}
I would then want to be able to call methods from Sub1 and Sub2 as if those methods were defined in MainClient. So instead of calling (new MainClient()).Sub1.Method1() I would call (new MainClient()).Method1() where Method1 still exists in the Sub1Client class.
Is this possible?

I not sure that clearly understand your question, but check this solution:
public interface IFirst
{
void Method1(string a);
}
public interface ISecond
{
double Method2(int b, bool a);
}
public interface IComplex : IFirst, ISecond
{
}
public class MyException : Exception
{
public MyException(string message) : base(message)
{
}
}
public class Sub1Client : IFirst
{
public void Method1(string a)
{
Console.WriteLine("IFirst.Method1");
Console.WriteLine(a);
}
}
public class Sub2Client : ISecond
{
public double Method2(int b, bool a)
{
Console.WriteLine("ISecond.Method2");
return a ? b : -b;
}
}
public class MainClient : IComplex
{
public MainClient()
{
Sub1 = new Sub1Client();
Sub2 = new Sub2Client();
}
public static Sub1Client Sub1;
public static Sub2Client Sub2;
private T FindAndInvoke<T>(string methodName, params object[] args)
{
foreach(var field in this.GetType().GetFields(BindingFlags.Public | BindingFlags.Static))
{
var method = field.FieldType.GetMethod(methodName);
if(method != null)
return (T)method.Invoke(field.GetValue(this), args);
}
throw new MyException("Method was not found!");
}
public void Method1(string a)
{
FindAndInvoke<object>(MethodBase.GetCurrentMethod().Name, a);
}
public double Method2(int b, bool a)
{
return FindAndInvoke<double>(MethodBase.GetCurrentMethod().Name, b, a);
}
}
public static void Main()
{
var test = new MainClient();
test.Method1("test");
Console.WriteLine(test.Method2(2, true));
}

Generic method declaration

I have hierarchy of classes:
class A{}
class B: A {}
class C:B {}
is it possible to implement method in class A and it would be inherited by derived classes B and C and so on and that method should return value of class type?
A val = A.method(); (val is A)
B val = B.method(); (val is B)
C val = C.method(); (val is C)
And I don't want use of generics in call of this method, ie:
C val = C.method<C>();
Guys, excuse me, one elaboration, this method should be static.
I don't want to use generic in method istelf, because it forces to point type that method should return, whereas method should return type of its class.
class A
{
Method<T>()
{
T result;
return result;
}
}
If I have such method I can change return type:
D result = A.Method<D>();
but I wanted it to return value of type A;

No, that is not possible.
To call the method like that it would have to be static, and static methods are not inherited.
Using B.method() to call a static method in A is the same as using A.method(). The compiler just uses the type to determine where the method is, but it's impossible for the method to know if it was called using the A or B type.

Use an extension method:
class Program
{
static void Main(string[] args)
{
B x = new B();
x.Method();
}
}
public static class Ext
{
public static T Method<T>(this T obj)
where T : A,new()
{
return new T();
}
}
public class A
{
}
public class B : A
{
}
Or a variation thereof. Note that you must have some public member capable of creating an instance of the specified type. To expound, the compiler 'guesses' the value of the type parameter. The method is still generic, but generic syntax is nowhere to be seen when the method is called (usually).

Using some design patterns from C++ makes this easier:
class A
{
protected virtual A method_impl() { return new A(); }
public A method() { return method_impl(); }
}
class B : A
{
protected override A method_impl() { return new B(); }
public new B method() { return (B)method_impl(); }
}
class C : B
{
protected override A method_impl() { return new C(); }
public new C method() { return (C)method_impl(); }
}
Of course, this exact problem never arises in C++, which allows covariant return types for overrides.
Another way, using IoC pattern:
class A
{
protected virtual void method_impl(A a) { a.initialize(); }
public A method() { A result = new A(); method_impl(result); return result; }
}
class B : A
{
public new B method() { B result = new B(); method_impl(result); return result; }
}
class C : B
{
public new C method() { C result = new C(); method_impl(result); return result; }
}

Discriminated union in C#

[Note: This question had the original title "C (ish) style union in C#"
but as Jeff's comment informed me, apparently this structure is called a 'discriminated union']
Excuse the verbosity of this question.
There are a couple of similar sounding questions to mine already in SO but they seem to concentrate on the memory saving benefits of the union or using it for interop.
Here is an example of such a question.
My desire to have a union type thing is somewhat different.
I am writing some code at the moment which generates objects that look a bit like this
public class ValueWrapper
{
public DateTime ValueCreationDate;
// ... other meta data about the value
public object ValueA;
public object ValueB;
}
Pretty complicated stuff I think you will agree. The thing is that ValueA can only be of a few certain types (let's say string, int and Foo (which is a class) and ValueB can be another small set of types. I don't like treating these values as objects (I want the warm snugly feeling of coding with a bit of type safety).
So I thought about writing a trivial little wrapper class to express the fact that ValueA logically is a reference to a particular type. I called the class Union because what I am trying to achieve reminded me of the union concept in C.
public class Union<A, B, C>
{
private readonly Type type;
public readonly A a;
public readonly B b;
public readonly C c;
public A A{get {return a;}}
public B B{get {return b;}}
public C C{get {return c;}}
public Union(A a)
{
type = typeof(A);
this.a = a;
}
public Union(B b)
{
type = typeof(B);
this.b = b;
}
public Union(C c)
{
type = typeof(C);
this.c = c;
}
/// <summary>
/// Returns true if the union contains a value of type T
/// </summary>
/// <remarks>The type of T must exactly match the type</remarks>
public bool Is<T>()
{
return typeof(T) == type;
}
/// <summary>
/// Returns the union value cast to the given type.
/// </summary>
/// <remarks>If the type of T does not exactly match either X or Y, then the value <c>default(T)</c> is returned.</remarks>
public T As<T>()
{
if(Is<A>())
{
return (T)(object)a; // Is this boxing and unboxing unavoidable if I want the union to hold value types and reference types?
//return (T)x; // This will not compile: Error = "Cannot cast expression of type 'X' to 'T'."
}
if(Is<B>())
{
return (T)(object)b;
}
if(Is<C>())
{
return (T)(object)c;
}
return default(T);
}
}
Using this class ValueWrapper now looks like this
public class ValueWrapper2
{
public DateTime ValueCreationDate;
public Union<int, string, Foo> ValueA;
public Union<double, Bar, Foo> ValueB;
}
which is something like what I wanted to achieve but I am missing one fairly crucial element - that is compiler enforced type checking when calling the Is and As functions as the following code demonstrates
public void DoSomething()
{
if(ValueA.Is<string>())
{
var s = ValueA.As<string>();
// .... do somethng
}
if(ValueA.Is<char>()) // I would really like this to be a compile error
{
char c = ValueA.As<char>();
}
}
IMO It is not valid to ask ValueA if it is a char since its definition clearly says it is not - this is a programming error and I would like the compiler to pick up on this. [Also if I could get this correct then (hopefully) I would get intellisense too - which would be a boon.]
In order to achieve this I would want to tell the compiler that the type T can be one of A, B or C
public bool Is<T>() where T : A
or T : B // Yes I know this is not legal!
or T : C
{
return typeof(T) == type;
}
Does anyone have any idea if what I want to achieve is possible? Or am I just plain stupid for writing this class in the first place?

I don't really like the type-checking and type-casting solutions provided above, so here's 100% type-safe union which will throw compilation errors if you attempt to use the wrong datatype:
using System;
namespace Juliet
{
class Program
{
static void Main(string[] args)
{
Union3<int, char, string>[] unions = new Union3<int,char,string>[]
{
new Union3<int, char, string>.Case1(5),
new Union3<int, char, string>.Case2('x'),
new Union3<int, char, string>.Case3("Juliet")
};
foreach (Union3<int, char, string> union in unions)
{
string value = union.Match(
num => num.ToString(),
character => new string(new char[] { character }),
word => word);
Console.WriteLine("Matched union with value '{0}'", value);
}
Console.ReadLine();
}
}
public abstract class Union3<A, B, C>
{
public abstract T Match<T>(Func<A, T> f, Func<B, T> g, Func<C, T> h);
// private ctor ensures no external classes can inherit
private Union3() { }
public sealed class Case1 : Union3<A, B, C>
{
public readonly A Item;
public Case1(A item) : base() { this.Item = item; }
public override T Match<T>(Func<A, T> f, Func<B, T> g, Func<C, T> h)
{
return f(Item);
}
}
public sealed class Case2 : Union3<A, B, C>
{
public readonly B Item;
public Case2(B item) { this.Item = item; }
public override T Match<T>(Func<A, T> f, Func<B, T> g, Func<C, T> h)
{
return g(Item);
}
}
public sealed class Case3 : Union3<A, B, C>
{
public readonly C Item;
public Case3(C item) { this.Item = item; }
public override T Match<T>(Func<A, T> f, Func<B, T> g, Func<C, T> h)
{
return h(Item);
}
}
}
}

I like the direction of the accepted solution but it doesn't scale well for unions of more than three items (e.g. a union of 9 items would require 9 class definitions).
Here is another approach that is also 100% type-safe at compile-time, but that is easy to grow to large unions.
public class UnionBase<A>
{
dynamic value;
public UnionBase(A a) { value = a; }
protected UnionBase(object x) { value = x; }
protected T InternalMatch<T>(params Delegate[] ds)
{
var vt = value.GetType();
foreach (var d in ds)
{
var mi = d.Method;
// These are always true if InternalMatch is used correctly.
Debug.Assert(mi.GetParameters().Length == 1);
Debug.Assert(typeof(T).IsAssignableFrom(mi.ReturnType));
var pt = mi.GetParameters()[0].ParameterType;
if (pt.IsAssignableFrom(vt))
return (T)mi.Invoke(null, new object[] { value });
}
throw new Exception("No appropriate matching function was provided");
}
public T Match<T>(Func<A, T> fa) { return InternalMatch<T>(fa); }
}
public class Union<A, B> : UnionBase<A>
{
public Union(A a) : base(a) { }
public Union(B b) : base(b) { }
protected Union(object x) : base(x) { }
public T Match<T>(Func<A, T> fa, Func<B, T> fb) { return InternalMatch<T>(fa, fb); }
}
public class Union<A, B, C> : Union<A, B>
{
public Union(A a) : base(a) { }
public Union(B b) : base(b) { }
public Union(C c) : base(c) { }
protected Union(object x) : base(x) { }
public T Match<T>(Func<A, T> fa, Func<B, T> fb, Func<C, T> fc) { return InternalMatch<T>(fa, fb, fc); }
}
public class Union<A, B, C, D> : Union<A, B, C>
{
public Union(A a) : base(a) { }
public Union(B b) : base(b) { }
public Union(C c) : base(c) { }
public Union(D d) : base(d) { }
protected Union(object x) : base(x) { }
public T Match<T>(Func<A, T> fa, Func<B, T> fb, Func<C, T> fc, Func<D, T> fd) { return InternalMatch<T>(fa, fb, fc, fd); }
}
public class Union<A, B, C, D, E> : Union<A, B, C, D>
{
public Union(A a) : base(a) { }
public Union(B b) : base(b) { }
public Union(C c) : base(c) { }
public Union(D d) : base(d) { }
public Union(E e) : base(e) { }
protected Union(object x) : base(x) { }
public T Match<T>(Func<A, T> fa, Func<B, T> fb, Func<C, T> fc, Func<D, T> fd, Func<E, T> fe) { return InternalMatch<T>(fa, fb, fc, fd, fe); }
}
public class DiscriminatedUnionTest : IExample
{
public Union<int, bool, string, int[]> MakeUnion(int n)
{
return new Union<int, bool, string, int[]>(n);
}
public Union<int, bool, string, int[]> MakeUnion(bool b)
{
return new Union<int, bool, string, int[]>(b);
}
public Union<int, bool, string, int[]> MakeUnion(string s)
{
return new Union<int, bool, string, int[]>(s);
}
public Union<int, bool, string, int[]> MakeUnion(params int[] xs)
{
return new Union<int, bool, string, int[]>(xs);
}
public void Print(Union<int, bool, string, int[]> union)
{
var text = union.Match(
n => "This is an int " + n.ToString(),
b => "This is a boolean " + b.ToString(),
s => "This is a string" + s,
xs => "This is an array of ints " + String.Join(", ", xs));
Console.WriteLine(text);
}
public void Run()
{
Print(MakeUnion(1));
Print(MakeUnion(true));
Print(MakeUnion("forty-two"));
Print(MakeUnion(0, 1, 1, 2, 3, 5, 8));
}
}

I wrote some blog posts on this subject that might be useful:
Union Types in C#
Implementing Tic-Tac-Toe Using State Classes
Let's say you have a shopping cart scenario with three states: "Empty", "Active" and "Paid", each with different behavior.
You create have a ICartState interface that all states have in common (and it could just be an empty marker interface)
You create three classes that implement that interface. (The classes do not have to be in an inheritance relationship)
The interface contains a "fold" method, whereby you pass a lambda in for each state or case that you need to handle.
You could use the F# runtime from C# but as a lighter weight alternative, I have written a little T4 template for generating code like this.
Here's the interface:
partial interface ICartState
{
ICartState Transition(
Func<CartStateEmpty, ICartState> cartStateEmpty,
Func<CartStateActive, ICartState> cartStateActive,
Func<CartStatePaid, ICartState> cartStatePaid
);
}
And here's the implementation:
class CartStateEmpty : ICartState
{
ICartState ICartState.Transition(
Func<CartStateEmpty, ICartState> cartStateEmpty,
Func<CartStateActive, ICartState> cartStateActive,
Func<CartStatePaid, ICartState> cartStatePaid
)
{
// I'm the empty state, so invoke cartStateEmpty
return cartStateEmpty(this);
}
}
class CartStateActive : ICartState
{
ICartState ICartState.Transition(
Func<CartStateEmpty, ICartState> cartStateEmpty,
Func<CartStateActive, ICartState> cartStateActive,
Func<CartStatePaid, ICartState> cartStatePaid
)
{
// I'm the active state, so invoke cartStateActive
return cartStateActive(this);
}
}
class CartStatePaid : ICartState
{
ICartState ICartState.Transition(
Func<CartStateEmpty, ICartState> cartStateEmpty,
Func<CartStateActive, ICartState> cartStateActive,
Func<CartStatePaid, ICartState> cartStatePaid
)
{
// I'm the paid state, so invoke cartStatePaid
return cartStatePaid(this);
}
}
Now let's say you extend the CartStateEmpty and CartStateActive with an AddItem method which is not implemented by CartStatePaid.
And also let's say that CartStateActive has a Pay method that the other states dont have.
Then here's some code that shows it in use -- adding two items and then paying for the cart:
public ICartState AddProduct(ICartState currentState, Product product)
{
return currentState.Transition(
cartStateEmpty => cartStateEmpty.AddItem(product),
cartStateActive => cartStateActive.AddItem(product),
cartStatePaid => cartStatePaid // not allowed in this case
);
}
public void Example()
{
var currentState = new CartStateEmpty() as ICartState;
//add some products
currentState = AddProduct(currentState, Product.ProductX);
currentState = AddProduct(currentState, Product.ProductY);
//pay
const decimal paidAmount = 12.34m;
currentState = currentState.Transition(
cartStateEmpty => cartStateEmpty, // not allowed in this case
cartStateActive => cartStateActive.Pay(paidAmount),
cartStatePaid => cartStatePaid // not allowed in this case
);
}
Note that this code is completely typesafe -- no casting or conditionals anywhere, and compiler errors if you try to pay for an empty cart, say.

I have written a library for doing this at https://github.com/mcintyre321/OneOf
Install-Package OneOf
It has the generic types in it for doing DUs e.g. OneOf<T0, T1> all the way to
OneOf<T0, ..., T9>. Each of those has a .Match, and a .Switch statement which you can use for compiler safe typed behaviour, e.g.:
```
OneOf<string, ColorName, Color> backgroundColor = getBackground();
Color c = backgroundColor.Match(
str => CssHelper.GetColorFromString(str),
name => new Color(name),
col => col
);
```

I am not sure I fully understand your goal. In C, a union is a structure that uses the same memory locations for more than one field. For example:
typedef union
{
float real;
int scalar;
} floatOrScalar;
The floatOrScalar union could be used as a float, or an int, but they both consume the same memory space. Changing one changes the other. You can achieve the same thing with a struct in C#:
[StructLayout(LayoutKind.Explicit)]
struct FloatOrScalar
{
[FieldOffset(0)]
public float Real;
[FieldOffset(0)]
public int Scalar;
}
The above structure uses 32bits total, rather than 64bits. This is only possible with a struct. Your example above is a class, and given the nature of the CLR, makes no guarantee about memory efficiency. If you change a Union<A, B, C> from one type to another, you are not necessarily reusing memory...most likely, you are allocating a new type on the heap and dropping a different pointer in the backing object field. Contrary to a real union, your approach may actually cause more heap thrashing than you would otherwise get if you did not use your Union type.

char foo = 'B';
bool bar = foo is int;
This results in a warning, not an error. If you're looking for your Is and As functions to be analogs for the C# operators, then you shouldn't be restricting them in that way anyhow.

If you allow multiple types, you cannot achieve type safety (unless the types are related).
You can't and won't achieve any kind of type safety, you could only achieve byte-value-safety using FieldOffset.
It would make much more sense to have a generic ValueWrapper<T1, T2> with T1 ValueA and T2 ValueB, ...
P.S.: when talking about type-safety I mean compile-time type-safety.
If you need a code wrapper (performing bussiness logic on modifications you can use something along the lines of:
public class Wrapper
{
public ValueHolder<int> v1 = 5;
public ValueHolder<byte> v2 = 8;
}
public struct ValueHolder<T>
where T : struct
{
private T value;
public ValueHolder(T value) { this.value = value; }
public static implicit operator T(ValueHolder<T> valueHolder) { return valueHolder.value; }
public static implicit operator ValueHolder<T>(T value) { return new ValueHolder<T>(value); }
}
For an easy way out you could use (it has performance issues, but it is very simple):
public class Wrapper
{
private object v1;
private object v2;
public T GetValue1<T>() { if (v1.GetType() != typeof(T)) throw new InvalidCastException(); return (T)v1; }
public void SetValue1<T>(T value) { v1 = value; }
public T GetValue2<T>() { if (v2.GetType() != typeof(T)) throw new InvalidCastException(); return (T)v2; }
public void SetValue2<T>(T value) { v2 = value; }
}
//usage:
Wrapper wrapper = new Wrapper();
wrapper.SetValue1("aaaa");
wrapper.SetValue2(456);
string s = wrapper.GetValue1<string>();
DateTime dt = wrapper.GetValue1<DateTime>();//InvalidCastException

Here is my attempt. It does compile time checking of types, using generic type constraints.
class Union {
public interface AllowedType<T> { };
internal object val;
internal System.Type type;
}
static class UnionEx {
public static T As<U,T>(this U x) where U : Union, Union.AllowedType<T> {
return x.type == typeof(T) ?(T)x.val : default(T);
}
public static void Set<U,T>(this U x, T newval) where U : Union, Union.AllowedType<T> {
x.val = newval;
x.type = typeof(T);
}
public static bool Is<U,T>(this U x) where U : Union, Union.AllowedType<T> {
return x.type == typeof(T);
}
}
class MyType : Union, Union.AllowedType<int>, Union.AllowedType<string> {}
class TestIt
{
static void Main()
{
MyType bla = new MyType();
bla.Set(234);
System.Console.WriteLine(bla.As<MyType,int>());
System.Console.WriteLine(bla.Is<MyType,string>());
System.Console.WriteLine(bla.Is<MyType,int>());
bla.Set("test");
System.Console.WriteLine(bla.As<MyType,string>());
System.Console.WriteLine(bla.Is<MyType,string>());
System.Console.WriteLine(bla.Is<MyType,int>());
// compile time errors!
// bla.Set('a');
// bla.Is<MyType,char>()
}
}
It could use some prettying-up. Especially, I couldn't figure out how to get rid of the type parameters to As/Is/Set (isn't there a way to specify one type parameter and let C# figure the other one?)

So I've hit this same problem many times, and I just came up with a solution that gets the syntax I want (at the expense of some ugliness in the implementation of the Union type.)
To recap: we want this sort of usage at the call site.
Union<int, string> u;
u = 1492;
int yearColumbusDiscoveredAmerica = u;
u = "hello world";
string traditionalGreeting = u;
var answers = new SortedList<string, Union<int, string, DateTime>>();
answers["life, the universe, and everything"] = 42;
answers["D-Day"] = new DateTime(1944, 6, 6);
answers["C#"] = "is awesome";
We want the following examples to fail to compile, however, so that we get a modicum of type safety.
DateTime dateTimeColumbusDiscoveredAmerica = u;
Foo fooInstance = u;
For extra credit, let's also not take up more space than absolutely needed.
With all that said, here's my implementation for two generic type parameters. The implementation for three, four, and so on type parameters is straight-forward.
public abstract class Union<T1, T2>
{
public abstract int TypeSlot
{
get;
}
public virtual T1 AsT1()
{
throw new TypeAccessException(string.Format(
"Cannot treat this instance as a {0} instance.", typeof(T1).Name));
}
public virtual T2 AsT2()
{
throw new TypeAccessException(string.Format(
"Cannot treat this instance as a {0} instance.", typeof(T2).Name));
}
public static implicit operator Union<T1, T2>(T1 data)
{
return new FromT1(data);
}
public static implicit operator Union<T1, T2>(T2 data)
{
return new FromT2(data);
}
public static implicit operator Union<T1, T2>(Tuple<T1, T2> data)
{
return new FromTuple(data);
}
public static implicit operator T1(Union<T1, T2> source)
{
return source.AsT1();
}
public static implicit operator T2(Union<T1, T2> source)
{
return source.AsT2();
}
private class FromT1 : Union<T1, T2>
{
private readonly T1 data;
public FromT1(T1 data)
{
this.data = data;
}
public override int TypeSlot
{
get { return 1; }
}
public override T1 AsT1()
{
return this.data;
}
public override string ToString()
{
return this.data.ToString();
}
public override int GetHashCode()
{
return this.data.GetHashCode();
}
}
private class FromT2 : Union<T1, T2>
{
private readonly T2 data;
public FromT2(T2 data)
{
this.data = data;
}
public override int TypeSlot
{
get { return 2; }
}
public override T2 AsT2()
{
return this.data;
}
public override string ToString()
{
return this.data.ToString();
}
public override int GetHashCode()
{
return this.data.GetHashCode();
}
}
private class FromTuple : Union<T1, T2>
{
private readonly Tuple<T1, T2> data;
public FromTuple(Tuple<T1, T2> data)
{
this.data = data;
}
public override int TypeSlot
{
get { return 0; }
}
public override T1 AsT1()
{
return this.data.Item1;
}
public override T2 AsT2()
{
return this.data.Item2;
}
public override string ToString()
{
return this.data.ToString();
}
public override int GetHashCode()
{
return this.data.GetHashCode();
}
}
}

And my attempt on minimal yet extensible solution using nesting of Union/Either type.
Also usage of default parameters in Match method naturally enables "Either X Or Default" scenario.
using System;
using System.Reflection;
using NUnit.Framework;
namespace Playground
{
[TestFixture]
public class EitherTests
{
[Test]
public void Test_Either_of_Property_or_FieldInfo()
{
var some = new Some(false);
var field = some.GetType().GetField("X");
var property = some.GetType().GetProperty("Y");
Assert.NotNull(field);
Assert.NotNull(property);
var info = Either<PropertyInfo, FieldInfo>.Of(field);
var infoType = info.Match(p => p.PropertyType, f => f.FieldType);
Assert.That(infoType, Is.EqualTo(typeof(bool)));
}
[Test]
public void Either_of_three_cases_using_nesting()
{
var some = new Some(false);
var field = some.GetType().GetField("X");
var parameter = some.GetType().GetConstructors()[0].GetParameters()[0];
Assert.NotNull(field);
Assert.NotNull(parameter);
var info = Either<ParameterInfo, Either<PropertyInfo, FieldInfo>>.Of(parameter);
var name = info.Match(_ => _.Name, _ => _.Name, _ => _.Name);
Assert.That(name, Is.EqualTo("a"));
}
public class Some
{
public bool X;
public string Y { get; set; }
public Some(bool a)
{
X = a;
}
}
}
public static class Either
{
public static T Match<A, B, C, T>(
this Either<A, Either<B, C>> source,
Func<A, T> a = null, Func<B, T> b = null, Func<C, T> c = null)
{
return source.Match(a, bc => bc.Match(b, c));
}
}
public abstract class Either<A, B>
{
public static Either<A, B> Of(A a)
{
return new CaseA(a);
}
public static Either<A, B> Of(B b)
{
return new CaseB(b);
}
public abstract T Match<T>(Func<A, T> a = null, Func<B, T> b = null);
private sealed class CaseA : Either<A, B>
{
private readonly A _item;
public CaseA(A item) { _item = item; }
public override T Match<T>(Func<A, T> a = null, Func<B, T> b = null)
{
return a == null ? default(T) : a(_item);
}
}
private sealed class CaseB : Either<A, B>
{
private readonly B _item;
public CaseB(B item) { _item = item; }
public override T Match<T>(Func<A, T> a = null, Func<B, T> b = null)
{
return b == null ? default(T) : b(_item);
}
}
}
}

You could throw exceptions once there's an attempt to access variables that haven't been initialized, ie if it's created with an A parameter and later on there's an attempt to access B or C, it could throw, say, UnsupportedOperationException. You'd need a getter to make it work though.

The C# Language Design Team discussed discriminated unions in January 2017 https://github.com/dotnet/csharplang/blob/master/meetings/2017/LDM-2017-01-10.md#discriminated-unions-via-closed-types
You can vote for the feature request at https://github.com/dotnet/csharplang/issues/113

You can export a pseudo-pattern matching function, like I use for the Either type in my Sasa library. There's currently runtime overhead, but I eventually plan to add a CIL analysis to inline all the delegates into a true case statement.

It's not possible to do with exactly the syntax you've used but with a bit more verbosity and copy/paste it's easy to make overload resolution do the job for you:
// this code is ok
var u = new Union("");
if (u.Value(Is.OfType()))
{
u.Value(Get.ForType());
}
// and this one will not compile
if (u.Value(Is.OfType()))
{
u.Value(Get.ForType());
}
By now it should be pretty obvious how to implement it:
public class Union
{
private readonly Type type;
public readonly A a;
public readonly B b;
public readonly C c;
public Union(A a)
{
type = typeof(A);
this.a = a;
}
public Union(B b)
{
type = typeof(B);
this.b = b;
}
public Union(C c)
{
type = typeof(C);
this.c = c;
}
public bool Value(TypeTestSelector _)
{
return typeof(A) == type;
}
public bool Value(TypeTestSelector _)
{
return typeof(B) == type;
}
public bool Value(TypeTestSelector _)
{
return typeof(C) == type;
}
public A Value(GetValueTypeSelector _)
{
return a;
}
public B Value(GetValueTypeSelector _)
{
return b;
}
public C Value(GetValueTypeSelector _)
{
return c;
}
}
public static class Is
{
public static TypeTestSelector OfType()
{
return null;
}
}
public class TypeTestSelector
{
}
public static class Get
{
public static GetValueTypeSelector ForType()
{
return null;
}
}
public class GetValueTypeSelector
{
}
There are no checks for extracting the value of the wrong type, e.g.:
var u = Union(10);
string s = u.Value(Get.ForType());
So you might consider adding necessary checks and throw exceptions in such cases.

I am currently trying to create a Julia Runtime in .NET. Julia has types like Union{Int, String}... Etc. I am currently trying to simulate this .NET (without doing weird IL that would not be able to be called from c#).
Here is a compile time implementation of a union of structures. I will be creating more unions for object unions, and cross object and struct unions (this will be the most complex case).
public struct Union<T1,T2> where T1 : struct where T2 : struct{
private byte type;
[FieldOffset(1)] private T1 a1;
[FieldOffset(1)] private T2 a2;
public T1 A1 {
get => a1;
set {
a1 = value;
type = 1;
}
}
public T2 A2 {
get => a2;
set {
a2 = value;
type = 2;
}
}
public Union(int _ = 0) {
type = 0;
a1 = default;
a2 = default;
}
public Union(T1 a) : this() => A1 = a;
public Union(T2 a) : this() => A2 = a;
public bool HasValue => type < 1 || type > 2;
public bool IsNull => !HasValue;
public bool IsT1 => type == 1;
public bool IsT2 => type == 2;
public Type GetType() {
switch (type) {
case 1: return typeof(T1);
case 2: return typeof(T2);
default: return null;
}
}
}
You can use the above like the following:
Union<int, long> myUnion(5); \\Set int inside
myUnion.a2 = 5;
Type theTypeInside = myUnion.GetType(); //long
myUnion.a1 = 5;
theTypeInside = myUnion.GetType(); //int
I will also be creating dynamic union generators or aligned unions for the cross object and struct union.
Take a look at:Generated Struct Union Output to see the current compile time unions I am using.
If you want to create a union of any size take a look at Generator for Struct Unions
If anyone has any improvements for the above let me know! Implementing julia into .NET is an extraordinarily hard task!

I use own of Union Type.
Consider an example to make it clearer.
Imagine we have Contact class:
public class Contact
{
public string Name { get; set; }
public string EmailAddress { get; set; }
public string PostalAdrress { get; set; }
}
These are all defined as simple strings, but really are they just strings?
Of course not. The Name can consist of First Name and Last Name. Or is an Email just a set of symbols? I know that at least it should contain # and it is necessarily.
Let's improve us domain model
public class PersonalName
{
public PersonalName(string firstName, string lastName) { ... }
public string Name() { return _fistName + " " _lastName; }
}
public class EmailAddress
{
public EmailAddress(string email) { ... }
}
public class PostalAdrress
{
public PostalAdrress(string address, string city, int zip) { ... }
}
In this classes will be validations during creating and we will eventually have valid models. Consturctor in PersonaName class require FirstName and LastName at the same time. This means that after the creation, it can not have invalid state.
And contact class respectively
public class Contact
{
public PersonalName Name { get; set; }
public EmailAdress EmailAddress { get; set; }
public PostalAddress PostalAddress { get; set; }
}
In this case we have same problem, object of Contact class may be in invalid state. I mean it may have EmailAddress but haven't Name
var contact = new Contact { EmailAddress = new EmailAddress("foo#bar.com") };
Let's fix it and create Contact class with constructor which requires PersonalName, EmailAddress and PostalAddress:
public class Contact
{
public Contact(
PersonalName personalName,
EmailAddress emailAddress,
PostalAddress postalAddress
)
{
...
}
}
But here we have another problem. What if Person have only EmailAdress and haven't PostalAddress?
If we think about it there we realize that there are three possibilities of valid state of Contact class object:
A contact only has an email address
A contact only has a postal address
A contact has both an email address and a postal address
Let's write out domain models. For the beginning we will create Contact Info class which state will be corresponding with above cases.
public class ContactInfo
{
public ContactInfo(EmailAddress emailAddress) { ... }
public ContactInfo(PostalAddress postalAddress) { ... }
public ContactInfo(Tuple<EmailAddress,PostalAddress> emailAndPostalAddress) { ... }
}
And Contact class:
public class Contact
{
public Contact(
PersonalName personalName,
ContactInfo contactInfo
)
{
...
}
}
Let's try use it:
var contact = new Contact(
new PersonalName("James", "Bond"),
new ContactInfo(
new EmailAddress("agent#007.com")
)
);
Console.WriteLine(contact.PersonalName()); // James Bond
Console.WriteLine(contact.ContactInfo().???) // here we have problem, because ContactInfo have three possible state and if we want print it we would write `if` cases
Let's add Match method in ContactInfo class
public class ContactInfo
{
// constructor
public TResult Match<TResult>(
Func<EmailAddress,TResult> f1,
Func<PostalAddress,TResult> f2,
Func<Tuple<EmailAddress,PostalAddress>> f3
)
{
if (_emailAddress != null)
{
return f1(_emailAddress);
}
else if(_postalAddress != null)
{
...
}
...
}
}
In the match method, we can write this code, because the state of the contact class is controlled with constructors and it may have only one of the possible states.
Let's create an auxiliary class, so that each time do not write as many code.
public abstract class Union<T1,T2,T3>
where T1 : class
where T2 : class
where T3 : class
{
private readonly T1 _t1;
private readonly T2 _t2;
private readonly T3 _t3;
public Union(T1 t1) { _t1 = t1; }
public Union(T2 t2) { _t2 = t2; }
public Union(T3 t3) { _t3 = t3; }
public TResult Match<TResult>(
Func<T1, TResult> f1,
Func<T2, TResult> f2,
Func<T3, TResult> f3
)
{
if (_t1 != null)
{
return f1(_t1);
}
else if (_t2 != null)
{
return f2(_t2);
}
else if (_t3 != null)
{
return f3(_t3);
}
throw new Exception("can't match");
}
}
We can have such a class in advance for several types, as is done with delegates Func, Action. 4-6 generic type parameters will be in full for Union class.
Let's rewrite ContactInfo class:
public sealed class ContactInfo : Union<
EmailAddress,
PostalAddress,
Tuple<EmaiAddress,PostalAddress>
>
{
public Contact(EmailAddress emailAddress) : base(emailAddress) { }
public Contact(PostalAddress postalAddress) : base(postalAddress) { }
public Contact(Tuple<EmaiAddress, PostalAddress> emailAndPostalAddress) : base(emailAndPostalAddress) { }
}
Here the compiler will ask override for at least one constructor. If we forget to override the rest of the constructors we can't create object of ContactInfo class with another state. This will protect us from runtime exceptions during Matching.
var contact = new Contact(
new PersonalName("James", "Bond"),
new ContactInfo(
new EmailAddress("agent#007.com")
)
);
Console.WriteLine(contact.PersonalName()); // James Bond
Console
.WriteLine(
contact
.ContactInfo()
.Match(
(emailAddress) => emailAddress.Address,
(postalAddress) => postalAddress.City + " " postalAddress.Zip.ToString(),
(emailAndPostalAddress) => emailAndPostalAddress.Item1.Name + emailAndPostalAddress.Item2.City + " " emailAndPostalAddress.Item2.Zip.ToString()
)
);
That's all.
I hope you enjoyed.
Example taken from the site F# for fun and profit