Thread: Structures? *brain melts*

Originally posted by Bubba

Code:

: Gosub Module1
:
: Module1:
:    PRINT "This is a module"
:    RETURN

Also worth mentioning that in QB you never have to use GOSUB except in very special circumstances.

And that is not a module: You have defined module1: to be a label, not a module. The entire file is the module.

sorry this isn't helping.

But for what it's worth, I know how to program modules in QB, for god's sake, it's in the damn textbook... MODULES... and it's using the same structure as I am...

ANYWAY... BACK TO C++ PLEASE??!

No one let me bring up QB again!

But for this discussion structs and classes have little or nothing to do with the concept of a module. So to get back to the point a module is just one source code file in a project whether it be in PASCAL, assembly, BASIC, or C/C++.

Actually, I think the DoomMaster was more concerned with the struct code that he had.

In order to create an instance of a class you must provide a constructor. The constructor is called at creation time. The way in which you instantiated the class is incorrect and will generate an error from the compiler.

I have often omitted a constructor. I have NEVER had an error.

Classes should be instantiated with new - they are like structs but they cannot be thought of exactly in the same terms.

Every tutorial I used never did it with 'new'. Nothing ever went wrong. Could you explain a bit further, as I do not have a C++ book.

Originally posted by Bubba
Classes should be instantiated with new...

There's no more reason to dynamically allocate instances of a structure than there is a primitive type.

Originally posted by bennyandthejets
Classes/structs allow you to group data which has something in common. For example, here is the traditional Cat class:

Code:

class Cat
{
public:
      int Age;
      int Weight;
      void Meow();
      void Sleep();
};

You group all the things related to Cats in one object. But, what if you have two cats?

Code:

Cat Tintin;
Cat Chelsea;

(Those are the names of my cats)
Those statements create two objects based on the type 'Cat'. Ie, both objects have Age and Weight, and they can both Meow() and Sleep(). You cannot do this:

Code:

Cat.Age=5;

Because Cat is not an object, it is a type. Therefore, there is actually no variable in memory which can be assigned the value '5'.

Hmm... so, Cat isn't a user-defined name? same with Example?

hmmmmmmmm.... I get mad when I can't figure something out. Is it just me being stupid, or is this really hard to figure out?

Hmm... so, Cat isn't a user-defined name? same with Example?

I'm not sure what you mean by user-defined name.

Additionally, with constructors:

If you provide none, the compiler will generate for you a default constructor. Unless you explicitly declare them away, it will also generate a shallow copy constructor and assignment operator. When creating an object from its default constructor, the parens can be omitted. That is, the following statements are equivalent:

MyClass a;
MyClass a();

Same with:

MyClass* a = new MyClass;
MyClass* a = new MyClass();

Originally posted by MrDoomMaster
Hmm... so, Cat isn't a user-defined name? same with Example?

hmmmmmmmm.... I get mad when I can't figure something out. Is it just me being stupid, or is this really hard to figure out?

It is a user defined type. A type just like int.

Tintin is an instance of Cat. That is, the following are analogous:

Cat Tintin; // Create a new Cat named Tintin
int x; // Create a new int named x

You can instantiate classes w/o new but again I don't recommend doing this as it will only serve to confuse you

Believe me, the only thing confusing me is YOU. I've been using classes for a couple years now, and I use them both ways, with new, and without. I've never been confused.

Originally posted by bennyandthejets
I'm not sure what you mean by user-defined name.

well, I mean, for example... you can't initialize a variable called INT, because that is a reserved word... right?

So, INT is not user-defined... yet, a variable name like int123 would be user defined, because the programmer made up the name himself... got it?

Now, I know we're not talking about variables in the question of structures, but I was just using variables to explain to you what I mean't by user-defined. I'm sorry if my terminology is off, but I hope you catch my drift!

I'm a work in progress, I understand how hard it is to be patient with people that just take FOREVER to understand something that you see so simple, but I'm really trying as hard as I can. Remember, we all had to learn this at one point. I thank everyone for your help, including you Bubba, even though I have no idea what you're saying most of the time lol. Thanks though It's the thought that counts!


How about I just tell you how I understand structures so far, and you can fill in the gaps?

Okay, explaining this as simple as I can... A structure, to me, is a scope enclosed with curly braces that contains variable initializations, for example:

Code:

struct myvariables
{
    int variable1
    int variable2
}

and there has to be some way to call on one of these variables that are contained inside of the above "scope" right? So, I figure you would put the name of the structure, then a period, and the name of the variable inside of the structure that you want to change, like so:

myvariables.variable1=3

To me, this would say "Look inside of structure 'myvariables' and find 'variable1' then set it to 3!"

See how I see things now? That's why the "example an_example" throws me off! It doesn't fit in with what I understand! Anyhow, I hope this helps!

Thanks for everyone's patience!

Originally posted by Bubba
Well if you do not provide a constructor then when your object is created the compiler will provide one. However, this is not a good practice. Let's say for instance you have a pointer in your class. If you do not provide a constructor and init the pointer to 0 you have a NULL pointer. Accessing it later w/o first initializing it will crash the system. IMO all classes should have a constructor and I'm not sure what the ISO C++ standard says about it.

New is basically malloc(sizeof(object)). New is not C it is C++ which is why your books don't use it. But classes are C++ so I can't understand why a C book would include classes in it. You can instantiate classes w/o new but again I don't recommend doing this as it will only serve to confuse you. For that matter you can put constructors in structs, but again we are getting way off topic here.


Where are you guys getting these books?

Yes, there are reasons to provide constructors a lot of the time. There is not always a reason, however. Additionally, if you want to get rid of the copy semantics, and default constructor, you must declare them away:

Code:

class X
{
public:
  // Whatever constructors you want.
private:
  // Declare away!
  X();
  X(const X&);
  X& operator=(const X&);
};

Using dynamic allocation (new) will probably be more confusing at this state. Not the 'new' operator itself, but the fact that you are responsible for cleaning up (where appropriate, and not before) any memory you allocate.

Just go with the example, and create it on the stack (i.e. without new).