Understanding method meaning...

Guest User

Given the following extension\utility\helper method

public static string CommaSeparate(this IEnumerable source, Func func)
{ return string.Join(",", source.Select(s => func(s).ToString()).ToArray()); }

Im having difficulty in understanding it easily\fully...

My concerns are:-

Why are both T and U needed after CommaSeperate and not just T?

How and why is the Select method mapping to the Func delegate func, by simply s => func(s) ?

Regards
Steve

3 Replies

Know the answer? Post it — somebody with the same question will find it here.

Vulpes13y ago
Accepted answer
Hi Steve,

T and U are both needed because the second parameter to the extension method is a Func i.e. a delegate representing a method which takes a parameter of type T and returns a value of type U.

The Select extension method is being passed this lambda expression:

s => func(s).ToString()

So, the parameter 's' is inferred to be of type T because 'source' is of type IEnumerable.
The expression func(s) then passes 's' to func which we know takes a parameter of type T and returns a value of type U.

The ToString() method is then applied to this U value which always works because all classes/structs inherit the virtual ToString() method from type System.Object.

So, less formally, the extension method is taking a sequence of T values, transforming them to U values, converting them to their string equivalent and then converting the whole sequence to a string array.

It is then using the String.Join method to return a single string containing the elements of the string array separated by commas.
Vulpes13y ago
In that example only one type parameter is needed (TSource) because bool is a fixed type.

A type parameter is in effect a variable which can represent (subject to any constraints) any type at all.

However, it's possible to have generic methods where the type parameters aren't used in the parameter list or return type but are used in the body of the method. For example, consider the generic method in this program which returns the default value of the type parameter's actual type:

using System;

class Test
{
static object GetTypeString()
{
return default(T);
}

static void Main()
{
Console.WriteLine(GetTypeString()); // 0
Console.WriteLine(GetTypeString()); // false
Console.ReadKey();
}
}
Guest User13y ago
Brilliant thank you Vulpes.

So when exactly do extension methods with 2 type parameters ie need to be defined?

For example the Where extension method only has 1 type parameter but takes a func input parameter like so

public static IEnumerable Where
(this IEnumerable source, Funcbool> predicate)

Is this because the return value of the func delegate is already defined as bool and therefore a second generic type parameter ie U isnt needed?

Regards

Steve