I made a test program:
#include <iostream>
using namespace std;
class Bug {
private:
char a[25];
int& view (int i) {
return *reinterpret_cast<int*>(&a[i]);}
public:
Bug () {}
void overwrite () {
view(a[0]) = 2;
cout << "value of a[0] is: " << view(a[0]) << endl;
}
};
int main () {
Bug b;
b.overwrite();
return 0;
}
... where I'm testing the view function. Everything works as I'd expect; the value beginning at a[0] is 2, as the print statement in the void method shows.
However, I modify the program slightly:
#include <iostream>
using namespace std;
class Bug {
private:
int x;
char a[25];
int& view (int i) {
return *reinterpret_cast<int*>(&a[i]);}
public:
Bug () {}
void overwrite () {
view(a[0]) = 2;
cout << "value of a[0] is: " << view(a[0]) << endl;
}
};
int main () {
Bug b;
b.overwrite();
return 0;
}
Notice that I added a private member variable int x. This was the only change. The moment I do this, the print statement in the overwrite method no longer prints 2, as in the previous program -- it returns a garbage value (and sometimes the correct value, 2).
More importantly, sometimes the correct value will show, but the program will segfault.
Anybody know what's happening under the hood?
Bonus question. for values lower than 25, my private variable char a[value] will show compiler warnings that mention stack smashing (using g++-4.7).
Related
My code is below. This works, It allows me to have exactly one range in my lambda.
So I guess what my question is, is how do I achieve the same results without using
"if(LOOP > 2 && LOOP < 5){int THERANGEVALUE = 2; FUNC[THERANGEVALUE]();}"?
And instead initialize an item in my captureless lambda as being ranged instead. aka, item_2 being item_range(2,4). And then also being able to continue my lambda normally, whereas Item_3 will equate to item_5.
Thank you for any help in advance, I will gladly add more input if requested.
#include <iostream>
using namespace std;
void (*FUNC[3])(void) = { //captureless lambda.
/*ITEM_0*/[](){ cout << "x" << endl;},
/*ITEM_1*/[](){cout << "y" << endl;},
/*ITEM_2->ITEM_4*/[](){cout<<"z";}
};
/*Here the [](){code;} lambda is acting as a stand-in for void FUNC() so it shouldn't touch anything outside of its scope*/
int LOOP = 4;
int main()
{
if(LOOP > 2 && LOOP < 5){int THERANGEVALUE = 2; FUNC[THERANGEVALUE]();}
FUNC[LOOP]();
return 0;
}
Adding on to this, below is the solution I came up with after asking a friend. To my surprise it was actually a lot simpler than I expected. While I couldn't initialize each item in the lambda in a range easily, I could pass it into an array and set a range inside of the array instead. So while it's not quite what I was looking for, it's...good enough for my purposes. Thanks Jaime if you see this. Otherwise I'd use PilouPili's answer below.
#include <iostream>
using namespace std;
void (*FUNC[4])(void) = { //captureless lambda.
/*ITEM_0*/ [](){ cout << "x" << endl;},
/*ITEM_1*/ [](){cout << "y" << endl;},
/*ITEM_2->ITEM_4*/[](){cout<<"z";},
/*ITEM_5*/ [](){cout<<"z";}
};
int LOOP = 4;
int main()
{
int ARR[5]={};
for(int I = 0; I < 6;I=I+1){//handling of ranged values.
if(I>2 && I<5){ARR[I]=2;} else {ARR[I]=I;}
}
FUNC[ARR[LOOP]]();
return 0;
}
I only see to way :
either extend your function array -> That's FUNC1 in the next example
change the value given in operator [] -> That's FUNC2 in the next example
#include <iostream>
#include <vector>
using namespace std;
std::vector<void (*)(void)> init_FUNC()
{
std::vector<void (*)(void)> func(5, [](){cout<<"z";});
func[0]=[](){ cout << "x" << endl;};
func[1]=[](){ cout << "y" << endl;};
return func;
}
std::vector<void (*)(void)> FUNC1= init_FUNC();
class FUNC_MAP
{
void (*_FUNC[3])(void) = { //captureless lambda.
/*ITEM_0*/[](){ cout << "x" << endl;},
/*ITEM_1*/[](){cout << "y" << endl;},
/*ITEM_2->ITEM_4*/[](){cout<<"z";}
};
typedef void (*FUNC_MAP_OUT)(void);
public:
FUNC_MAP_OUT operator[](int i)
{
if(i>2 && i<5)
{return _FUNC[2];}
else
{return _FUNC[i];}
}
};
FUNC_MAP FUNC2;
/*Here the [](){code;} lambda is acting as a stand-in for void FUNC() so it shouldn't touch anything outside of its scope*/
int LOOP = 1;
int main()
{
FUNC1[LOOP]();
FUNC2[LOOP]();
return 0;
}
Im trying to write a class that stores an id and a value in an container class.
Im using an nested class as my data structure.
When im compiling the code sometimes it prints perfectly, sometimes it prints nothing and sometimes it prints half of the data then stops.
When i debug the code the same weird behavior occours, when it fails during debug it throws an error "Map.exe has triggered a breakpoint.", the Error occours in the print method when im using cout.
cmap.h
#pragma once
class CMap
{
public:
CMap();
~CMap();
CMap& Add(int id, int value);
void print() const;
private:
class container
{
public:
~container();
int container_id = 0;
int container_value = 0;
};
container* p_komp_;
int dim_ = -1;
void resize();
};
cmap.cpp
#include "cmap.h"
#include <iostream>
using namespace std;
CMap::CMap()
{
p_komp_ = new container[0];
}
CMap::~CMap()
{
p_komp_ = nullptr;
cout << "destroy cmap";
}
CMap& CMap::Add(int id, int value)
{
resize();
p_komp_[dim_].container_id = id;
p_komp_[dim_].container_value = value;
return *this;
}
void CMap::resize()
{
container* temp_array = new container[++dim_];
if (dim_ == 0)
{
temp_array[0].container_id = p_komp_[0].container_id;
temp_array[0].container_value = p_komp_[0].container_value;
}
for (unsigned i = 0; i < dim_; i++)
{
temp_array[i].container_id = p_komp_[i].container_id;
temp_array[i].container_value = p_komp_[i].container_value;
}
p_komp_ = temp_array;
}
void CMap::print() const
{
for (unsigned i = 0; i <= dim_; i++)
{
cout << p_komp_[i].container_id;
cout << p_komp_[i].container_value;
}
}
CMap::container::~container()
{
cout << "destruct container";
}
Map.cpp
#include "cmap.h"
#include <iostream>
using namespace std;
void main(void)
{
CMap m2;
m2.Add(1, 7);
m2.Add(3, 5);
m2.print();
}
These two things are a possible reason for your problem:
int dim_ = -1;
and
container* temp_array = new container[++dim_];
When you allocate, you increase dim_ from -1 to 0. That is you create a zero-sized "array", where every indexing into it will be out of bounds and lead to undefined behavior.
You also have memory leaks since you never delete[] what you new[]. I didn't look for more problems, but there probably a more.
And an "array" (created at compile-time or through new[]) will have indexes from 0 to size - 1 (inclusive). You seem to think that the "size" you provide is the top index. It's not, it's the number of elements.
It seems to me that you might need to take a few steps back, get a couple of good books to read, and almost start over.
#include <iostream>
using namespace std;
class simpleFunction
{
protected:
int score;
public:
simpleFunction()
{
score = 5;
}
int retScore()
{
return score;
}
};
class changeFunction : public simpleFunction
{
public:
void change()
{
score = 6;
}
};
int main()
{
simpleFunction SimpleFunction;
changeFunction ChangeFunction;
cout << SimpleFunction.retScore() << endl;
ChangeFunction.change(); // Changes Score To 6
cout << SimpleFunction.retScore() << endl; // Should Return 6 But Returns 5 Instead
return 0;
}
ive set the score to 5 and when i use the change function it should change it to 6 but instead it returns the value 5.
the only way ive managed to make it work as intended to is by changing the int score variable to a global varible but all the classes can access it which makes it flawed can anyone help or try to explain this problem to me.
The Program Works but im having a problem with returning the correct values just to clear up any misunderstadnings
whenever I run the program, there is no output, the program just ends. Am i doing something wrong? I'm sure there's something i missed but i can't seem to figure it out.
#include <iostream>
#include <string>
using namespace std;
class Addr
{
public:
Addr(int i = 0){
total = i;
}
void addNum(int num){
total += num;
}
int getNum(){
return total; }
friend int print(Addr& var);
private:
int total;
};
int print(Addr& var){
return var.total;
}
int main()
{
Addr object1;
object1.addNum(3);
print(object1);
return 0;
}
Your program behaves correctly. There is no output because you are not printing anything to the console in your program.
The print function merely returns the total.
If you wish to print the value to the console then you could for example change the definition as follows:
int print(Addr& var){
cout << var.total << endl; // this prints to the console output
return var.total;
}
There is no issue with your code. The fact is that no print function is used. I have modified your main function.
int main()
{
Addr object1;
object1.addNum(3);
cout<<print(object1);
return 0;
}
I was trying to save the binary equivalent of a 32 bit number in an array A. For testing my showbits() function , I choosed 8,9 when I came across this thing:
I am facing an unreasonable thing in my code when I am placing memset in the function showbits(),I am geting absurd integers while I expect an output something as
00000000000000000000000000001000
that is the binary equivalent of 8 . While when I place memset in the main() method, it works properly and gives me the right output.Am I going out of bounds(I cannot see it !) .
My code :
SHOWBITS:
void showbits(int A[32],int num)
{
int k=0;
memset(A,0,sizeof(A));
while(num>0)
{
A[k] = num&1;
k++;
num>>=1;
}
return ;
}
Note: I have placed memset in showbits ,and I am getting incorrect answers!
MAIN:
int main()
{
int A[32],i;
showbits(A,8);
for(i=31;i>=0;i--)
printf("%d",A[i]);
return 0;
}
Whole program for testing:
#include<stdio.h>
#include<algorithm>
#include<cstring>
using namespace std;
void showbits(int A[32],int num)
{
int k=0;
memset(A,0,sizeof(A));
while(num>0)
{
A[k] = num&1;
k++;
num>>=1;
}
return ;
}
int main()
{
int A[32],i;
showbits(A,8);
for(i=31;i>=0;i--)
printf("%d",A[i]);
return 0;
}
When I place that memset statement in Main method before showbits() , I am getting correct output!
EDIT
If someone is interested in what I am getting
398420075242008462686872420075219611920941961187434-2205336646196127610926869242
68672826866724200752000202903316219611874341961187478819611565142686716196182637
61961141748268665201000
The A[32] in the method is actually just a pointer to A. Therefore, sizeof is the size of *int. Take the following test code:
void szof(int A[32])
{
std::cout << "From method: " << sizeof(A) << "\n";
}
int main(int argc, char *argv[])
{
int B[32];
std::cout << "From main: " << sizeof(B) << "\n";
szof(B);
return 0;
}
which give the following output:
From main: 128
From method: 8
Thus, the memset sets fewer bits than you think.
You must pass A by reference
void showbits(int (&A)[32],int num)
See here for more details: C++ pass an array by reference
Avi explained the problem in your code already. Another possible solution is to use C++-style arrays, instead of C-style arrays and memset. Then there is no possibility of a memset length error. Also there is no loss of performance.
#include <array>
void showbits(std::array<int, 32> &A, int num)
{
A = {}; // set all entries to 0
// ...
}
int main()
{
std::array<int, 32> A;
// ...
}