[kenneth2k1]

I'm new to C, and I'm on Apple's development site going through their tutorial thing. I'm reading about "object models" and I can understand some of these terms, but I am not wrapping my head around the idea of a class. And it's probably because of these stupid types of definitions. I don't get it, so perhaps someone could explain it in dumber terms.

class In the Objective-C language, a prototype for a particular kind of object. A class definition declares instance variables and defines methods for all members of the class. Objects that have the same types of instance variables and have access to the same methods belong to the same class.


Thanks for any help, and any other info you want to give a beginning C developer.

[AGCurry]

Classes are really not an integral part of C programming, but they are a part of C++ and Java and any other object-oriented language.

Do you know what a "struct" is in C? A struct is a structure, which in even older languages we call a record. It may contain scalar variables, arrays, other structs, and pointers.

In C we can use pointers to functions. It's kind of an advanced concept and may be a little hard to wrap your head around, but if you think about how a program looks when it's in memory, a function resides at a certain address, just like a variable. So, we can create a variable which is a pointer to that address, and we call it a pointer to a function.

So, it wasn't long before programmers began to include these pointers to functions in structs. In that way, you can "wrap" functions related to the data in the struct inside the struct itself.

This works perfectly well, but some programmers wanted to implement this capability in a more programmer-friendly way, so the concept of the class was developed as a wrapper, to kind of hide the ugly nuts and bolts of implementation.

As a dyed-in-the-wool plain old C programmer, it helps me to understand classes in this way - as an implementation of pointers to functions. In the object-oriented world, these functions are called methods.

The final piece of the puzzle involves the difference between "definition" and "instantiation."
In a C program, when I code:

struct something
{
int c ;
char d ;
} ;

This defines what struct something looks like, but it does not result in an INSTANCE of a struct something. This is analogous to coding a class, or to putting a new word in the dictionary before it has actually ever been used.

To make an INSTANCE of the struct that, I must code:

struct something something_instance ;

This is a little bit of an oversimplification, but not by much. Hope it helps.

[kenneth2k1]

Well that is a better explanation. I am starting to see how it fits. I think once I start programming I will see a lot better, which should be today sometime. Thanks.

[kenneth2k1]

Ok, so I'm writing my first program, and the compiler keeps returning an error. Since I just kinda jumped into this thing, perhaps you can help me. I wrote a simple program the way it was described to me here's the lines of code:

#include <stdio.h>
void main()
{
printf("Hello World");
}

When I compile it, I get an error that says:

'main' must return 'int'
so I delete void and put int in the parenthesis. No errors are returned, but then the program doesn't execute anything. My guess is that if it's supposed to return an integer, it's not going to do anything because I'm not outing any integers, right?

[blaqDeaph]

your best bet is to make it "int main()", and add a "return 1;" as the second line of the function.

And BTW, the only real difference between a struct and class in C & C++ is that with a class you can control access to variables and functions by using "public", "private", and "protected".

[kenneth2k1]

Thanks for the info on class. I am going through the tutorial that came with the compiler program I got, it's a little easier to follow. I changed the code to read like this:

#include<stdio.h>
main()
{
printf("Hello World\n");
}

That's a hell of a lot easier to understand to me.

I'm on to loops now, and things are going nicely.

[blaqDeaph]

This int and return 1 thing is another left over from the old DOS days, where programs returned whether or not they were completed successfully. The number used to (not sure if it's still) be stored under ERRORLEVEL 255 which the OS (or other subsequent programs) would then access.

[AGCurry]

Originally Posted by blaqDeaph

This int and return 1 thing is another left over from the old DOS days, where programs returned whether or not they were completed successfully. The number used to (not sure if it's still) be stored under ERRORLEVEL 255 which the OS (or other subsequent programs) would then access.

This is still the preferred approach. All programs should return an exit status. However, the convention is to return 0 for success and nonzero for errors. This is a UNIX convention, which allows shell coding like:

if myprogram
then
do_something
else
echo "An error occurred in myprogram"
fi

[blaqDeaph]

Yea, DOS would be a little more complicated, if my memory serves me right, you have to call the program, then test for the errorlevel, instead of directly testing the program.