It's always a good idea to show the actual compiler errors. The whole point of them is that they're supposed to tell you (us) what's wrong!

It would also be a good idea to simmer down the code to a minimal compilable example. For example, I don't know if you've omitted header guards/"#pragma once" for brevity or if they really don't exist.

So I'm going to work on these assumptions:

1. You have appropriate header guards, or are using "#pragma once"
   
2. I'll assume the compiler error is something along the lines of "C++ forbids declaration of 'PlayerClass' with no type".
   

Ok, note how you have circular inclusion: playerclass.h includes weaponclass.h includes playerclass.h ...
Recall also that #include is somewhat like copy-and-pasting a header at the point of inclusion.

So after the preprocessor has done its work, we end up with something like this:


#ifndef WEAPONCLASS_H_
#define WEAPONCLASS_H_

// Filename: weaponclass.h

// --- begin #included playerclass.h ---
#ifndef PLAYERCLASS_H_
#define PLAYERCLASS_H_
// Filename: playerclass.h.

// --- begin #included weaponclass.h ---
#ifndef WEAPONCLASS_H_
// WEAPONCLASS_H_ already defined at this point due to header guards.
// Circular inclusion has no effect.
#endif // WEAPONCLASS_H_
// --- end #included weaponclass.h ---

class PlayerClass
{
// blah blah blah
private:
WeaponClass *weapon;
};

#endif // PLAYERCLASS_H_
// --- end #included playerclass.h ---

class WeaponClass
{
// blah blah blah
private:
PlayerClass *weapon;
};

#endif // WEAPONCLASS_H_


Note how in the resulting translation unit, we've attempted to use the symbol "WeaponClass" as the type of a member of PlayerClass, even though "WeaponClass" hasn't been declared anywhere, thanks to the header guards and circular inclusion. Because there's no declaration, the compiler doesn't know what "WeaponClass" is, or how to treat it.

The way around this is to use forward declarations rather than inclusion. For example:


#ifndef PLAYERCLASS_H_
#define PLAYERCLASS_H_
// Filename: playerclass.h

class WeaponClass;

class PlayerClass
{
//blah blah blah
private:
WeaponClass *weapon;
};

#endif // PLAYERCLASS_H_


Then do the same thing in weaponclass.h. In your .cpp files you can #include both headers. I'll leave you to figure out why that works (just follow the copy-and-paste the preprocesor does and remember that it's valid to have multiple declarations that aren't definitions).

EDIT: and perhaps somewhat off topic, but having a weapon that knows about the player sounds a bit fishy to me.

Weapon has to be able to talk BACK to player class to check for ordnance collisions and stuff (FYI haven't got a clue how to do that yet, but paving the way!).
[/quote]
This is a dubious requirement. There are better ways to design this.

On a side note, there is probably no need to call your classes FooClass. Most coding conventions use an uppercase type names, with lowercase variable names to disambiguate:

class Weapon;

class Player
{
// ...
private:
Weapon *weapon;
};

It is up to you though, it is a stylistic choice.

Why does the player class check for collisions?

I can understand that the player might expose a bounding box (or cylinder, or whatever) and its velocity to allow external collision routines to work out any collisions, and what the collision response should be. Roughly:


bounds, velocity = world.calc_collision_response(player.bounds, player.velocity, dt)
player.bounds = bounds
player.velocity = velocity


The collision detection/response code is then entirely divorced from the player. This means it can be used for other things too (NPCs, vehicles, ...)

Similarly for weapons. If you've got stuff like sniper rifles, shotguns, etc, all that's really needed is:


rays = player.weapon.get_spread(player.position, player.orientation); // might have many rays for a weapon with 'spread', like a shotgun

for ray in rays:
target = world.check_ray_hit(ray.origin, ray.direction)
if target
target.register_hit(ray.force, ray.penetration)


Here, the player's details (position and orientation) are indeed used to calculate what the weapon would hit. But that doesn't mean the weapon has to know about who's holding it. In fact, the code would look exactly the same for the player, or for an NPC.

The code is probably overly simplistic, but you really want to avoid the situation where every class knows about every other class. In other words, you want to reduce coupling. This means that making changes reduces the rippling effect that's often seen when refactoring.

There are other benefits to this kind of separation too. You can group stuff like collision tests together to improve locality and allow vectorization.