segment fault?

I dont' know what this is... it gives me a run-time error (the program just craps)... I debugged it and the debugger said it was an "access violation (segment fault)" here:

Code:

---

        for(v=0;v<vertical;v++)        //loop through rows
        {
                if(debug)
                          vcount++;
                for(h=0;h<horizontal;h++)        //loop through cols
                {
                        if(debug)
                                hcount++;
                        tank[v][h]=emptychar;        //fill with empty space
                }
        }

---

and that is part of this class:

Code:

---

class lifeClass
{
        public:
                lifeClass(unsigned int,unsigned int,unsigned int);
                          //constructor (tank length,tank width,organism#)
                void step();        //step through program
                bool debg();        //returns bool debug
                bool nextstep();        //returns true for another generation
                ~lifeClass();        //destructor
        private:
                  void randxy(unsigned int&,unsigned int&);        //randomizes X and Y
                  void drawtemptank();        //debug - Draws temporary tank
                void drawtank();        //draws tank to screen
                void copytank();        //copies next tank to current tank
                char live(unsigned int,unsigned int);        //living characters
                char die(unsigned int,unsigned int);        //dying characters
                unsigned int countNeighbors(unsigned int,unsigned int);        //counts neighbors
               
                bool debug;        //debugging statements - see lifeClass
               
                bool cont;        //continue to next generation?
                char tank[][];        //living space
                char emptychar,fillchar,killchar,bornchar;        //characters used
                unsigned int vertical,horizontal;        //length and width of living space
                unsigned int clickcount;        //counts clicks (steps) taken by the program
};

---

I would appreciate some help with what this is, and if there's a better way (which I'm sure there is) to write a 2D (array of arrays) array, please post...

secondly, you define a new char array like this

Code:

---

char *array=new char[int];

---

but how/can you do that for a 2D array? and how do you delete a 2D array?

sorry for all the questions, but, well, I have a lot of questions

Quote:

---

but how/can you do that for a 2D array? and how do you delete a 2D array?

---

I think you just have to allocate it one element at a time or one dimension at a time (?)

Code:

---

  for(int x=0;x<ROWS;x++)
    for(int y=0;y<COLS;y++)
          array[x][y]=new Type;

---

But I could be incorrect...

EDIT:

This seems to work:

Code:

---


int *array[10];
  for(int x=0;x<10;x++)
          array[x]=new int[10];
//or
int *array[10][10]
    for(int x=0;x<10;x++)
        for(int y=0;y<10;y++)
              array[x][y]=new int[10];

---

JaWiB's example is essentially a 3D array (though in essence, correct). To do this, I recommend using stl::vector, or a matrix class of some sort (or other vector class). Anyways, if this is the way you want to do it, here is a small example:

Code:

---

int** array2d = new int*[10];
for(int i = 0; i < 10; ++i)
  array2d[i] = new int[10];

... etc ...

for(int j = 0; j < 10; ++j)
  delete array2d[j];
delete array2d;

---

Cheers

If you allocate something with new[], then you should frre it with delete [] operator. If you free the memory with just delete, it will probably turn wrong, for example memory will be really unallocated, but the destructor will be called only for first element in your array. Here we have integers, so it doesn't matter, but it is still wrong to use delete instead of delete[], because it's unpredictable and compiler dependent.

cool thanks for all the help... I would do vectors, but I don't know how to use them/exactly what they are... =/ I still need to look into that some time...