Firstly, do you really have to have a mutable struct? They're almost always a bad idea. Likewise public fields. There are some very occasional contexts in which they're reasonable (usually both parts together, as with ValueTuple
) but they're pretty rare in my experience.
Other than that, I'd just create a constructor taking the two bits of data:
class SomeClass
{
struct MyStruct
{
private readonly string label;
private readonly int id;
public MyStruct (string label, int id)
{
this.label = label;
this.id = id;
}
public string Label { get { return label; } }
public string Id { get { return id; } }
}
static readonly IList<MyStruct> MyArray = new ReadOnlyCollection<MyStruct>
(new[] {
new MyStruct ("a", 1),
new MyStruct ("b", 5),
new MyStruct ("q", 29)
});
}
Note the use of ReadOnlyCollection instead of exposing the array itself - this will make it immutable, avoiding the problem exposing arrays directly. (The code show does initialize an array of structs - it then just passes the reference to the constructor of ReadOnlyCollection<>
.)
与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…