I've got the following code which demonstrates a problem :
Code:
struct A {
double x[2];
double i;
};
struct B : virtual public A {
double y[2];
[code]....
I'm wondering if this is a compiler bug. Why doesn't the pointer-to-(derived)-member work for the array if it works for the non-array?
How to put virtual keyword in the function of the base class. Will the function still be over-written properly? What will happen if I
(1) call function from base class pointer
(2) call function from derived class pointer
(3) call function from derived class object
If the virtual keyword is present, the over-written version will be called in all cases, am I mistaken?
I have the following classes and 'dreaded diamond':
A
/
/
B C
/
/
D
|
|
E
Classes B & C both inherit from A using public virtual A.
E is the only concrete class. None of the classes are totally abstract.
Every class has a copy constructor.
All of the copy constructors are chained together through the initialization lists.
E correctly calls D's copy constructor.
D correctly calls B and C's copy constructors.
But neither B nor C call A's copy constructor, although A's default constructor is called. To reiterate B and C have a call to A's copy constructor in their initialization lists.
I guess A's default constructor is being called is because of virtual inheritence, but why isn't its copy constructor called (too)?
A's copy constructor includes some very important code and I could do with calling it. Should I call it from the concrete class' initialization list or is that considered bad form?
I have a simple question about inheritance. Consider the following code:
Code:
Class Base {
int type;
Base(){};
};
Class Derived1 : public Base
[Code] ....
I get the following error: Class "Base" has no member "Function1";
That makes sense - as Base has not declared Function1. But how can I loop through a vector of Bases, and then if the object is of type Derived1, call the function Function1?
I've got the following code with output. I can't figure out myself why it's what printed out there. I believe, it has something to deal with overloading/overriding/virtual functions implementations in C++:
class Base{
public: virtual void f(int);
virtual void f(double);
}
[Code].....
Thus here're my conclusions:
1) in line
d.f(1.0);
for some reason compiler preferred casting double->int of the argument and then call to 'Derived::f(int)'.
2)in line
pb->f(1.0);
for some reason compiler preferred call to 'Base::f(double);'. 'Base' is static type of pb, but the dynamic type is 'Derived'.
I believe the answer has to deal with the fact whether virtual table contains in addition to functions' names also the types of arguments they accept. AFAIK, vTable doesn't include such info.
I defined a virtual class and three other classes based on it. I want to use them like this:
int main() {
Dirichlet_t D;
Neumann_t N;
Cauchy_t C;
PDEBoundary_t * B1=& D;
PDEBoundary_t * B2=& N;
PDEBoundary_t * B3=& C;
[Code] .....
but I got two major errors
1: "object f abstract type is not allowed" error.-----why not?
2: "the derived class must implement the inherited pure virtual method"-----Did't I?
Instead of this:
#include <iostream>
struct Object {
int size; // Want to avoid this because size is (almost always) constant
Object (int s): size(s) {} // for every Object subtype.
[Code] ....
I want this:
#include <iostream>
struct Object {
virtual int getSize() const = 0;
};
struct Block: Object {
int getSize() const {return 5;} // always 5, except once in a blue moon it may change
[Code] ....
The Decorator Pattern works (getSize() can then return 6) but it is a poor choice for my program because it will cause more problems (due to many containers holding its old address, among other things. Any way to achieve this without any change of address, and without creating new storage somewhere, either inside the class or outside the class (too much responsibility to follow that stored value for the rest of the program just for this rare change, and creating a data member uses up too much memory for the many, many Block instances)?
I'm writing a class "Property" for a program that manages different types of properties. This is my .h for y base class. I was trying to write a virtual void function to convert different children classes to strings that can be displayed, but Xcode is freaking out.
I had it as:
virtual void toString()= 0;
and it gave me an error message: "Virtual can only exist in non-static member functions" and "field has incomplete type 'void'"
I changed it to:
virtual string toString() = 0;
and the error message didn't change.
Is this an issue with Xcode or did I do something wrong? Even after changing it to string it told me that it "has incomplete type 'void'"....
Here is an example,
Code:
class A {
public:
virtual void foo(){}
virtual void foo2(){}
virtual void foo3(){}
};
int main() {
A a;
int ret = sizeof(A);
return 0;
}
Basically object a contains a virtual table pointer which is of size 4 bytes. Since class A should have a virtual table which contains three pointers pointing to foo, foo2,foo3 separately. So the virtual table should be of size 12 bytes. I wonder where is virtual table located in memory?
Let's look at this simplified code, it gives compilation error
#include <iostream>
using namespace std;
class A {
public:
void showInfo() { cout << " This is an instance of A" << endl; }
[Code] ....
Without using pointer, how to make this works?
I have found that when I dump a C++ object from memory to a file - it seems that there is a misplacement of the last Virtual-Function-Table pointer - in that appears at the beginning. The result is that the gdump information based on this object dump (using green hills) is incorrect. I copied the contents of the gdump information below. The executable is compiled in linux.
Basically MEIO::CameraStatus contains an item that relates to its parent class (line 188). Then it has 18 items that are all Diagnostics::EventsCounter items. Now for each Diagnostics::EventsCounter item there is a Virtual-Function-Table Info Pointer as its last item. All is fine and good except that the last item of MEIO::CameraStatus which is _selfReset is missing its last item of 4 bytes (which is the Virtual-Function-Table Info Pointer ). On the other hand - right before the first Diagnostics::EventsCounter item ("_vidErrors") - there is an extra 4 bytes which happens to be the Virtual-Function-Table Info Pointer. As I said the gdump information file does not see this.
Why the object memory "moves" the last Virtual-Function-Table Info Pointer to the beginning (right before _vidErrors) and is there a way to "fix" this?
///////////////////////////////////////////////////////////////////////////
"MEIO::CameraStatus" val:0x000002f0 ind208,-1) Struct-Begin Info
188: "" offset 0, Parent-Class Private Info C++ Struct ref = 114
189: "_vidErrors" offset 160, Member Info C++ Struct ref = 128
190: "_vdiErrors" offset 480, Member Info C++ Struct ref = 128
[Code] .....
I have Class A as a base class , and Class B , C derived classes from A and there's class D who have a data member (pointer to Array) of type A (Composition)
class D{
A **a;
int size;
.......
a = new A*[size];
......
};
And i have Print method , in its body i have to specific element (if it from class B or C ) with a given ID(both B and C have a data member ID ) there should be 2 options in print function .. printing elements for class B , or printing elements for class C ? how can i specific the elements ?
Shape base class, line and Point derived classes. What should I declare in .h files and implement in .cpp files that this is array will be work.
My major concern refer to operator [] and assign (=) operator. As far as I understand I should overload ([]) and (=) three times for classes shape , line and point or not... or is it possible made through virtual function? How will be code looks like ?
Code:
// part of main.cpp
Shape* shapes[3]; // Array of pointers to Shape
shapes[0] = new Shape();
shapes[1] = new Line ("line from array ", Point(1,22),Point(33,22));
shapes[2] = new Point(11,44);
cout << "using ToString function" << endl;
for(int i=0; i < 3; i++)
cout << s[i]->ToString();
for(i=0; i < 3; i++)
delete s[i];
Write a program using inheritance allow user to enter grades of his students 5~8 students as a base class and compute the sums for each students in derived class and compute the average of sums in another derived class. I created 3 classes in 1 header file and 1 cpp file how ever i cant seem to get the sum or the average to show up on execution time
header file
#ifndef GRADES_H_INCLUDED
#define GRADES_H_INCLUDED
class Grades {
public:
Grades()
[Code] .....
code:
/part of main function
int *A=NULL;
//load from file
load_from_file(A, &next);
printf("next=%d
",next);
getchar();
printf("A[%d]=%d
",0,A[0]);getchar();//here program crashes
}
[code]....
I initialize array A with NULL in main, and I "load" it with elements from a file. The code without function is working. But when I tried to use a function as above, my array crashes!!!
I am trying to use "remove_if" with a predicate function inside a class. The code intends to remove the grid cells which an agent cannot move into (from among all possible cells).
Code:
void classname::function1()
{
vector<MoorePoint> neighbors;
....
[Code]....
That code would work if it was not in a class and the predicate was not a member function. However, now I receive long error messages which I guess refer to incompatibility of remove_if template with the predicate parameter (one error includes : error C2064: term does not evaluate to a function taking 1 arguments).
I have the following scenario :
Code:
class test {
public :
int num;
};
int main() {
test t1;
test *ptr = &t1;
int test :: *mem_ptr = &test::num;
}
I need to access mem_ptr (pointer to a class member) through :
a. object (t1)
b. pointer to the object (ptr).
How can i do that ?
I have a class with member functions and a pointer like this:
Code:
class MyClass{
private:
typedef void(MyClass::*memFnPtr_t) ();
public:
memFnPtr_t fnptr;
void fn1();
void fn2();
};
Now in a regular function I create an instance of the class and try to assign the pointer:
Code:
MyClass mc;
mc.fnptr = &mc.fn1;
The compiler does not like it and suggests this instead:
Code:
mc.fnptr = &MyClass::fn1;
This seems to work but what if I have two instances of the class:
Code:
MyClass mc1, mc2;
How does the compiler know to distinguish between
Code:
mc1.fn1
and
Code:
mc2.fn1
when the assignment now looks identical:
Code:
mc1.fnptr = &MyClass::fn1;
mc2.fnptr = &MyClass::fn1;
So, I've got this class in SDL Player that has, among other things, an SDL_Texture* to hold an image that represents the player on the screen. I'd assume it's good practice to do get() and set() functions for the class; but because textures are handled via pointers, when I write a get() function I end up returning a pointer to an internal resource; which isn't good practice I hear as it "breaks" encapsulation.
Find my code below:
#ifndef PLAYER_H
#define PLAYER_H
#include "SDL.h"
#include "SDL_image.h"
#include "CTexture.h"
class Player {
[Code] .....
I have the following problem: I am using NLOpt for optimization. The API provides functions to set the objective. This is done as follows:
double objective(const vector<double> &x, vector<double> &grad, void *data)
{
return x[1]*x[0];
}
int main(){
nlopt::opt opti(nlopt::LD_MMA,2);
opti.set_min_objective(objective,NULL);
vector<double> x(2);
[Code]....
Now I want to make the function objective a member of a class:
class Foo {
public:
double objective(...){..}
};
How can I give this method to opti.optimize? If I make objective static I can use
opti.optimize(Foo::objective,NULL);
but I do not want to have a static member. Is it possible to create an object of type Foo and give it to opti.optimize?
I have following:
struct Point {int* a; int b;};
vector<vector<Point> > numbers;
vector<int> example;
The numbers vector has a matrix of a sort and each of the members are pointing to one member in the example vector. A member numbers.at(2).at(3).a is pointing at example.at(3). Now, can I remotely delete a member in the example vector using the pointers? Like so:
delete (*(numbers.at(2).at(3).a));
I know there is a more convenient way to delete members, but this is a very specific case I'm working on.
Here is the assignment: (3pts) Given the following class header file, write the class’ source code for each of the accessor and mutator functions listed. (How the functions have listed their parameters, varying between passing by reference and by value.) Don’t forget to comment your code – it counts!
class Album {
private:
char * artist; // band or singer’s name
char * title; // title of the album
[code]....
The input will be an array. My questions: First, am I on the right track?
When using (char * a) for a function, for example, this is passing the address of a, correct? so then *artist=a; changes what the address of a points to?
also, the functions are bool when I would expect void. Why? for all of the set_" " functions, the parameter is *... but for set_record_label it is *&. That appears to be a mistake to me. Is that right?
what is the difference between *& and * as parameters?
I have the following piece of code.
Code:
#include<iostream>
using namespace std;
class Test {
public:
Test(){cout<<"Test"<<endl;}
void fun() {
int i=5;
[Code] ...
Compiled with g++.
Executing this give output fun5.
It is correct? I have not allocated any object and so this pointer is not created. Then how it is able to run and call the function.
I'm currently programming a server which uses multiple threads- I have a class for one map in the game. Each map has a thread for timed events(tile regeneration, NPC regeneration, etc.), and a thread for handling NPCs(movement, combat, etc.). A basic structure of the class looks like this:
class Region {
public:
/* game values are here, they are public so
they can be accessed from outside of the class
inside of packet-handling functions and such */
int value;
void *Function();
[Code] ....
The program crashes when I use a member of the same class the function is located in- in the context I have shown about it would crash on "value++".
I have little problem which causing memory leaks.
Parent > Multiple Child(Parent parent) > Child destructor deleting parent => next Child destructor crash
Example code: without using:
class Parent {
public:
Parent() {
for(int i = 0; i < x; ++i) {
for(int j = 0; j < y; ++j)
childs[i][j] = new Child(this);
[Code] ....
If you read code, on Parent destructor i = 0 & j = 1 its going crash.
Parent will be deleted aswell, but it give me assert: _block_type_is_valid(phead- nblockuse)
I am trying to create a callback system for input events in my game engine.
Here is a cut down version of the EventManager.h file
#include "Controls.h"
#include "Object.h"
enum MouseEventType{PRESSED, POINTER_AT_POSITION, PRESSED_AT_POSITION };
[Code].....
This works if the function pointer being passed to the event manager is not a member function.
I have two other classes Scene and Object that could potentially use this EventManager to create callback events. Scene and Object are both pure virtual objects. How can I pass a pointer to a member function of the child classes of both Scene and Object? I am fine with just having two separate watchEvent functions for Scene and Object but how do I pass the type of the Object or the type of the Scene? The child classes are unknown as they are being created by someone using this game engine.
For example, if I want to make a player object it would be something like
class PlayerObject : public Object{...};
Now that PlayerObject type has to find its way to PlayerObject::functionToCall(). I think I need templates to do this but, since I never used them before
This is how I intend to use this
class OtherScene : public Scene{
void p_pressed(void){
//pause
}
[Code].....