Reputation: 23138
If I have a function
void Foo(params int[] bar){}
The following runs fine:
int[] a1 = {1, 2, 3};
int[] a2 = {4, 5, 6};
Foo(1, 2, 3);
Foo(a1);
But these give compile errors:
Foo(a1, 1, 2, 3);
Foo(1, 2, a1);
Foo(1, a1, 2);
Foo(a1, a2, 1, 2, 3);
because only the first argument is allowed to be an int[], the rest have to be ints.
The final example is what I would like to do, but the language won't let me without combining the arrays first. I really like the simplicity of the syntax, and I would rather not add to the code more than I have to. Does anyone have a nice way to do this?
Upvotes: 3
Views: 1074
Reputation: 427
Agreed, its very odd Foo(a1, 2, 3)
works.
Heres an extension method for .Concat to make Mehrdad's syntax a little easier on the eyes.
public T[] Concat<T>(this T[] first, params T[] other)
{
T[] output = new T[first.Length+other.Length];
first.CopyTo(output,0);
other.CopyTo(output,first.Length);
return output;
}
Can be used as follows:
Foo(a1.Concat(a2).Concat(1,2,3));
Upvotes: 2
Reputation: 422320
It's weird. Foo(a1, 2, 3)
shouldn't work. You should either pass an array or a bunch of integers. You can't mix them AFAIK. Do you have another overload or something?
There's not really a neat syntax for doing that. The most concise one I can think of is:
Foo(a1.Concat(a2).Concat(new[] {1,2,3}).ToArray());
Upvotes: 11