What is the advantage of specifying two types when creating a typedef'd struct?

Question

example 1:

struct T{
    int a;
};

creates the type struct T

---

example 2:

typedef struct {
    int a;
} T;

creates the type T

---

example 3:

typedef struct T{
    int a;
} T;

creates both the types struct T and T

---

I tend to see example 3 a lot, and I'm not sure why someone would choose it over example 1 or 2.

• Are there any advantages you gain from doing it this way?
• Are there reasons people do this for compatibility?
• Is it advantageous for some kind of scoping reason?

I would like to avoid doing it the example 3 way, because it is less maintenance on the type, and it restricts multiple ways of declaring the same thing. However, I would reconsider it, if there are benefits to this "double naming" technique.

Answer 1

I tend to see example 3 a lot, and I'm not sure why someone would choose it over example 1 or 2.

• Are there any advantages you gain from doing it this way?

I hold this truth to be self-evident, namely that cumbersome code is cumbersome. Everyone would prefer to write

T object;

instead of

struct T object;

However, a hard-core C coder might think hey, T is a struct, better call it like that and also, it mitigates the chance for confusion you'd get when doing

struct {int a; } T; typedef int T; 

• Are there reasons people do this for compatibility?

Yes. This way, structs in C can be used like they would be used in C++.

• Is it advantageous for some kind of scoping reason?

no, not that I'd be aware of.