Implicit "this"?

[lerno]

Possibly I would introduce an implicit this parameter for method functions, and I'd like to hear what you think about it. Below you see three possible versions: A (current), B (implicit), C (explicit but without the type). I'm still collecting opinions here, so nothing is decided yet.

fn void test(Foo* f) {
  printf("%d", f.bar(123));
}
// A explicit "this" parameter:
fn int Foo.bar(Foo* foo, int x) {
  return foo.z * x;
}
// B implicit "this" parameter (but still explicit use)
fn int Foo.bar(int x) {
  return this.z * x;
}
// C explicit "this" parameter but without type:
fn int Foo.bar(foo, int x) {
  return foo.z * x;
}

[zweimach]

(D) Explicit but with Go-style method syntax

// D explicit but with Go-style method syntax:
fn int (Foo* foo) bar(int x) {
    return foo.z * x;
}

[lerno]

(D) Explicit but with Go-style method syntax

I'm afraid that I find the Go style syntax extremely hard to visually parse and also feels very different from the rest of the syntax.

[data-man]

(D) (E)
this -> $self like FreePascal:
Programming Using Objects
Programming Using Objects Page 2

fn void Point.inc(int sx, int sy) {
  with $self {
      x += sx;
      y += sy;
  }
}

[lerno]

The suggestion is not for using member access without this. The $ prefix is reserved for compile time variables.

[00shiv]

What about option A, but allow dot calls on any value of the first argument type?

fn void test(Foo* f) {
  printf("%d", f.bar(123));
}
// A explicit "this" parameter; but can call it without first argument via dot notation
fn int bar(Foo* foo, int x) {
  return foo.z * x;
}

[lerno]

That would be function overloading, which C3 doesn't have.

[lerno]

This is long since resolved. So closing it.