Hi all!

Ok, this had me confused for a while. I have figured it out but I don't know why it works this way, since the manual states otherwise.

Here is the thing: I have defined the following struct.

Code:

---

typedef struct
{
	PANEL* p;
	var row;
	var col;
} Card;

---

When I know try to access the panel pointer, I have to do it in a special way. In the next example "card" is an array of card pointers:

Code:

---

(card[i])->p = pan_create("bmap = card_base", 1); // works fine
((card[i])->p).pos_x = 25; // does NOT work, instead it overwrites (card[i])->p
((card[i])->p)->pos_x = 25; // works as intended

---

So it seems in this case I have to use the -> and not the . to access the structure. However, the manual states:

Quote:

---

In standard C / C++, members of structs are accessed by a dot '.' and members of struct pointers are accessed by an arrow '->'. In lite-C or C-Script the compiler automatically detects whether it's a struct or a struct pointer, and the dot '.' can be used for both.

---

So, is this a bug? Or was I doing something wrong? Anyway, I am glad this works now, it had me confused for a while. By the way, if I declare a global panel pointer "test" and use this one with test.pos_x etc. everything works.

Gnometech

well, yes, it's a pointer to a panel, isn't it? so you obviously have to use -> instead of .

joey.

Sure, it makes sense, but I thought in Lite-C the . and the -> could both be used. Obviously not in all cases, that was just the point I was trying to make.

As I said I am not sure if this is intended behaviour - the quote from the manual states that both . and -> can be used and this is not true in every case, as the example shows.

Gnometech

My guess is that the compiler simply can't always do it right. It has a reason that there exists . as well as -> and depending on what you want, the compiler might decide to do it differently than what you expect.

Maybe it doesn't work in structs - for my. / my-> it works as intended.

The replacement of -> by . works for simple objects, but not for objects that are part of a struct. I've changed the manual description accordingly.

I see, thank you for the clarification. Should've posted it in "Blame the manual" then, I guess.

Thanks and regards,
Gnometech