I am defining a class GameState and a class MainMenuGameState. The former is meant to be an abstract class and the latter is inheriting it. But somehow, I am not able to overwrite its properties.
GameState.h
#ifndef _GAME_STATE_H_
#define _GAME_STATE_H_
#include <SDL2/SDL.h>
class GameState {
public:
virtual void loop(Uint32 deltaTime) = 0;
virtual void render() = 0;
virtual void event(SDL_Event * event) = 0;
bool stopRenderPropagation = false;
bool stopLoopPropagation = false;
};
#endif
MainMenuGameState.h
#ifndef _MAIN_MENU_GAME_STATE_H_
#define _MAIN_MENU_GAME_STATE_H_
#include "../Game.h"
class MainMenuGameState : public GameState {
public:
MainMenuGameState(Game * pGame);
void loop(Uint32 deltaTime);
void render();
void event(SDL_Event * event);
bool stopRenderPropagation = true;
bool stopLoopPropagation = true;
private:
Game * game;
int xOffset = 0;
int yOffset = 0;
};
#endif
So after instanciating a MainMenuGameState object, I expected stopRenderPropagation and stopLoopPropagation to be true, but they are false.
I also had no luck overwriting them inside the constructor for some reason.
MainMenuGameState::MainMenuGameState(Game * pGame) {
game = pGame;
xOffset = rand() % 20;
yOffset = rand() % 20;
stopRenderPropagation = true;
stopLoopPropagation = true;
}
After that, they are still true. I don't know weather this is a problem with my constructor or if I misunderstood polymorphism in c++.
The instances of MainMenuGameState get stored in a vector<GameState *>, could this be the problem? I am accessing the properties like this:
if(gameStates.begin() != gameStates.end()) {
std::vector<GameState *>::iterator it = gameStates.end();
do {
--it;
} while(it != gameStates.begin() && (*it)->stopLoopPropagation == false);
while(it != gameStates.end()) {
(*it)->loop(deltaTime);
++it;
}
}
Thank you for your help!
Your derived class is declaring another couple of members with the same name of the members in the base class, thus "hiding" the base ones.
You should accept the initial values for those members in the constructor, or if they are fixed properties of the class that never change you should make them member functions instead like in
class GameState {
public:
...
virtual bool stopRenderPropagation() { return false; }
virtual bool stopLoopPropagation() { return false; }
};
class MainMenuGameState : public GameState {
public:
...
bool stopRenderPropagation() { return true; }
bool stopLoopPropagation() { return true; }
...
};
Inheriting data members (your booleans) does not work in the same way as inheriting and overloading methods. Try putting the bools as protected inherited data members instead (uninitialized), and initialize them in the respective subclass constructors instead.
class MainMenuGameState : public GameState {
public:
MainMenuGameState(Game * pGame);
void loop(Uint32 deltaTime);
void render();
void event(SDL_Event * event);
protected:
bool stopRenderPropagation;
bool stopLoopPropagation;
private:
Game * game;
int xOffset = 0;
int yOffset = 0;
};
You declare new variables in your derived class, which is leading to these problems. Variables that are not private are inherited:
struct A { int x };
struct B : A {}; // has B::x by inheritance
You can just set them in your constructor without redeclaration:
struct A { int x; };
struct B : A { B() : x(1) {} };
Notice that it is usually considered bad practice to declare public variables, it is more common to implement getters and setters instead:
struct A
{
int x() const { return x_; }
int & x() { return x_; }
private:
int x_;
};
Related
I learn C++ OOP-paradigm and want to ask related question:
Assumption
We have a base class:
class Base {
public:
virtual SomeType PowerMethod() { return SomeType{} };
}
We have a variable target and subclass which realizes some calculations with target variable based on the constructor's parameter (simple calculations or complicated calcs):
class Calc : public Base {
public: // using only public access to simplify real code structure
SomeType target;
void Simple() { target = 1; };
void Complex(){ target = 10000; };
explicit Calc(bool isSimple) {
if(isSimple)
Simple();
else
Complex();
}
};
Question
How to optimally realize two classes which based on different methods (Simple or Complex) but provide the same functionality of PowerMethod()?
My solution
class SimpleCalc : public Calc {
bool isSimple = true;
public:
SomeType PowerMethod() override {
Calc CalcInstance(isSimple);
return CalcInstance.target;
};
};
class ComplexCalc : public Calc {
bool isSimple = false;
public:
SomeType PowerMethod() override {
Calc CalcInstance(isSimple);
return CalcInstance.target;
};
};
This solution is pretty "ugly" and I want to ask you how to make it more readable.
Thank you!
I think that in your code, you didn't mean to craete a new Calc object, but instead call it on the superclass. This can be done like so:
Calc::Simple();
You can override the method PowerMethod, but still call the superclass's code:
virtual SomeType PowerMethod() override {
//do something
Base::PowerMethod();
}
If your problem is more complicated, and polymorphism and superclasses can't help you, you can always declare some method protected, so that only subclasses can access it. So, you could for example do this:
class Calc : public Base {
protected:
SomeType target;
void Simple() { target = 1; };
void Complex(){ target = 10000; };
public:
explicit Calc(bool isSimple) {
if(isSimple)
Simple();
else
Complex();
}
};
class SimpleCalc : public Calc {
public:
SomeType PowerMethod() override {
Calc::Simple();
return Calc::target;
};
};
class ComplexCalc : public Calc {
public:
SomeType PowerMethod() override {
Calc::Complex();
return Calc::target;
};
};
If your target is to learn OOP then you can use a factory design pattern to create your final calculator based on isSimple condition:
#include <iostream>
class Base
{
public:
Base()
{
target = 0;
}
int target;
virtual void PowerMethod() = 0;
};
class SimpleCalc : public Base
{
virtual void PowerMethod() { target = 0; }
};
class ComplexCalc : public Base
{
virtual void PowerMethod() { target = 1000; }
};
class CalcFactory
{
public:
virtual Base* createCalc(bool isSimple)
{
if (isSimple)
return new SimpleCalc();
else
return new ComplexCalc();
}
};
int main()
{
CalcFactory factory;
Base * base1 = factory.createCalc(true);
Base * base2 = factory.createCalc(false);
base1->PowerMethod();
base2->PowerMethod();
std::cout << base1->target << std::endl;
std::cout << base2->target << std::endl;
}
I have a code which has many derived class from a single base class. I wrote this code when there is minimum required and currently the specification changes so I need to create some 100+ derived classes.
My earlier implementation was something like
class Base {
public:
Base();
virtual ~Base();
virtual bool isThereError() { return false;}
virtual int configureMe() { return 0; }
virtual int executeMe() { return 0;}
};
class Derived_1 : public Base {
public:
Derived_1() {
errorStatus = false;
//Some initialization code for this class }
virtual ~Derived_1() {}
bool isThereError() { return errorStatus;}
int configureMe() {
// configuration code for this class
return 0;
}
int executeMe() {
//execute the major functionality of this class based on the configuration
return 0;
}
private:
bool errorStatus;
};
class Derived_2 : public Base {
public:
Derived_2() {
errorStatus = false;
//Some initialization code for this class }
virtual ~Derived_2() {}
bool isThereError() { return errorStatus;}
int configureMe() {
// configuration code for this class
return 0;
}
int executeMe() {
//execute the major functionality of this class based on the configuration
return 0;
}
private:
bool errorStatus;
};
Main.cpp:
#include "Base.h"
#include "Derived_1.h"
#include "Derived_2.h"
#include <set>
Derived_1 *dv1Ptr;
Derived_2 *dv2Ptr;
typedef std::set<Base *> ClassPtrList;
int main() {
ClassPtrList cpList;
dv1Ptr = new Derived_1();
dv2Ptr = new Derived_2();
dv1Ptr->configureMe();
if(dv1Ptr->isThereError()){
cpList.insert(dv1Ptr);
}
dv2Ptr->configureMe();
if(dv2Ptr->isThereError()){
cpList.insert(dv2Ptr);
}
while(true) {
for(ClassPtrList::iterator iter = cpList.begin(); iter != cpList.end(); ++iter) {
(*iter)->executeMe();
}
Sleep(1000);
}
}
I found the above implementation would lengthen the number of line and it is also not a good practice to write such a form of code when there are more derived classes. I need to write a code using MACRO or any other type, so that each derive class get instantiated by itself and the ClassPtrList keeps the pointer of all the derived class.
I started with something like,
#define CTOR_DERIVED(drvClass) return new drvClass()
but I'm not sure how to avoid creating pointer to update the list. I need to create 287 such derived classes.
Thanks in advance.
I have two classes:
class CEnemy : CObject
{
protected:
int hitPoints;
};
class COgro : public CEnemy
{
COgro::COgro() {hitPoints = 100}
};
and in other file I have class 'CRocket', which can collide with COgro, there is it function:
void CRocket::OnCollision(CObject *collisionObject)
{
if (typeid(*collisionObject) == typeid(COgro))
{
//collisionObject->hitPoints -= 10; ?? or what?
}
}
I want to shoot 10 times to ogro before it dies. How to do this?
I've already tried:
collisionObject->hitPoints -= 10;
(CEnemy)collisionObject->hitPoints -= 10;
but I can't compile it...how to edit this hitPoints value, but without changing '(CObject *collisionObject)'?
Thx
EDIT:
//===============================================================
//------------------------------------CLASS CRocket-----------------------
class CRocket : public CObject
{
protected:
void OnAnimate(scalar_t deltaTime);
void OnCollision(CObject *collisionObject);
void OnDraw(CCamera *camera);
public:
float pitch;
float distanceTravel;
CVector forward;
bool isExplosion;
CTexture *explosionTex;
CExplosion *explosion;
CRocket();
~CRocket();
void Load();
void Unload();
};
void CRocket::OnCollision(CObject *collisionObject)
{
if (typeid(*collisionObject) == typeid(COgroEnemy))
{
isExplosion = true;
velocity = CVector(0.0, 0.0, 0.0);
explosion = new CExplosion(500, position, 8.0, explosionTex->texID);
PlaySound();
}
}
//-----------------------------------------class CObject
class CObject : public CNode
{
protected:
virtual void OnAnimate(scalar_t deltaTime)
{
position += velocity * deltaTime;
velocity += acceleration * deltaTime;
}
virtual void OnDraw(CCamera *camera) {}
virtual void OnCollision(CObject *collisionObject) {}
virtual void OnPrepare()
{
ProcessCollisions(FindRoot());
}
public:
CVector position;
CVector velocity;
CVector acceleration;
scalar_t size;
bool isDead;
CObject() {isDead = false;}
~CObject() {}
...
...
...
}
//---------------------------------------class CEnemy
class CEnemy : public CObject
{
public:
int hitPoints;
protected:
float distFromPlayer;
float runSpeed;
AIState_t aiState;
virtual void OnProcessAI() {}
void OnCollision(CObject *collisionObject)
{
// if this enemy collides with another enemy
if (typeid(*collisionObject) == typeid(CEnemy))
{
modelState = MODEL_IDLE;
velocity = CVector(0.0, 0.0, 0.0);
}
// if this enemy collides with the terrain (always)
else if (typeid(*collisionObject) == typeid(CTerrain))
{
position.y = ((CTerrain*)collisionObject)->GetHeight(position.x, position.z) + size;
}
else
{
}
}
public:
CPlayer *player;
...
...
//----------------------------------class COgro-------------------------
class COgroEnemy : public CEnemy
{
protected:
void OnProcessAI();
void OnCollision(CObject *collisionObject);
void OnPrepare();
public:
COgroEnemy() { Load(); }
COgroEnemy(float x, float z) { position.x = x; position.z = z; Load(); }
~COgroEnemy() {}
void Load();
};
You'll need to cast the pointer to a pointer type CEnemy* (or a subclass), or the dereferenced pointer to a reference type CEnemy&. For maximum safety, I'd suggest dynamic_cast, rather than an evil C-style cast; although that's slightly paranoid since you're checking the type before casting.
// no checks, undefined behaviour if type is wrong
((CEnemy*)collisionObject)->hitPoints -= 10;
static_cast<CEnemy*>(collisionObject)->hitPoints -= 10;
// throws if type is wrong
dynamic_cast<CEnemy&>(*collisionObject).hitPoints -= 10;
// does nothing if type is wrong
if (CEnemy* enemy = dynamic_cast<CEnemy*>(collisionObject)) {
enemy->hitPoints -= 10;
}
You might combine that with the type check, rather than using typeid:
if (COgro * ogro = dynamic_cast<COgro*>(collisionObject)) {
ogro->hitPoints -= 10;
}
Note that this isn't exactly the same as your test: it will pass if the object is a subtype of COgro, while your test checks for an exact match.
You code is not compiling because you are trying to access a class's protected data member from an external source.
The collisionObject parameter is an instance of CObject, which does not have a hitPoints data member.
Also, when you pass around pointers to base classes to functions, the functions should assume that they can only access the interface or features of the base class.
You should write another overloaded method:
void CRocket::OnCollision(CEnemy& enemy);
Or move the hitPoints data member to the CObject class.
I get this:
This is a sample code of the two classes:
main.h
class CControl
{
protected:
int m_X;
int m_Y;
public:
void SetX( int X ) { m_X = X; }
void SetY( int Y ) { m_Y = Y; }
int GetX() { return m_X; }
int GetY() { return m_Y; }
CControl *m_ChildControls;
CControl *m_NextSibling;
CControl *m_PreviousSibling;
CControl *m_Parent;
CControl *m_FocusControl;
};
class CButton : public CControl
{
protected:
bool m_Type;
bool m_Selected;
bool m_Focused;
public:
CButton( bool Type );
~CButton();
};
CButton::CButton( bool Type )
{
}
This is just the declarations of the two classes (they're not complete, but the problem comes in also in the full coded version).
main.cpp
#include <windows.h>
#include "main.h"
int main()
{
CButton *g_Button;
g_Button = new CButton( 1 );
return 0;
}
This is just the application main func where I declare g_Button as a new CButton object for making a debugging analysis.
Are you referring to the CControl * members? Since you didn't initialize them in the constructor it's normal that they are at some "random" value; in particular, the value you see is the pattern used in debug builds in VC++ to mark uninitialized memory.
The same holds also for the other fields of your class (-842150451 is the 32-bit signed integer interpretation of 0xcdcdcdcd).
The pointers could be anything, because they're not initialized.
The compiler generated default constructor for CControl doesn't initialize POD members. You'd need to write your own:
CControl() : m_ChildControls(NULL), m_NextSibling(NULL), m_PreviousSibling(NULL)
m_Parent(NULL), m_FocusControl(NULL)
{
}
You need to initialize the data members of the class in the constructor.
CButton::CButton( bool Type )
{
m_Type = Type;
m_X = m_Y = 0;
m_ChildControls = NULL;
// ...
}
For the below code snippet, how do I initialize instances of class Enemy with variables (such as x, y, type)? I have it working correctly, it triggers the instances no matter how many of them I insert... I just need to know the best way of creating an enemy with certain variables that will differ for each of my instances... particularly when some of those variables are in the base class and others are not.
class BaseObject
{
public:
virtual void Render() = 0;
int x;
int y;
};
class Enemy : public BaseObject
{
public:
Enemy() { }
virtual void Render()
{
cout << "Render! Enemy" << endl;
}
typedef std::set<BaseObject *> GAMEOBJECTS;
GAMEOBJECTS g_gameObjects;
int main()
{
g_gameObjects.insert(new Enemy());
g_lootObjects.insert(new Loot());
for(GAMEOBJECTS::iterator it = g_gameObjects.begin();
it != g_gameObjects.end();
it++)
{
(*it)->Render();
}
for(GAMEOBJECTS::iterator it = g_lootObjects.begin();
it != g_lootObjects.end();
it++)
{
(*it)->Render();
}
return 0;
}
Include the arguments in the enemy constructor and Base constructors. You can then use those to initialize the member variables.
class BaseObject
{
public:
BaseObject(int x, int y) : x(x), y(y){ }
virtual void Render() = 0;
int x;
int y;
};
and
class Enemy : public BaseObject
{
public:
Enemy(int x, int y, int foo) : BaseObject(x,y), foo(foo) { }
int foo;
...
};