Question

I have a “raffle” C++ program that I use to “draw out of a hat”. I receive an EXC_BAD_ACCESS signal when I try to use it, though. Here is the function:

vector<int> speedRaffle(vector<Player>players,int pNum){
    vector<int> spdtics,order;
    int ticnum,randy;
    vector<int>::iterator iter = spdtics.begin();
    for (int k=0;k<pNum;k++){
        for (int i=0; i<pNum; i++) {
            for (int j=0; j<pow(players[i].speed,2); j++){
                for (int io=0; io<order.size(); io++) {
                    if(order[io]!=i){
                        spdtics.push_back(i);
                        ticnum++;
                    }
                }
            }
        }
        randy=random() % ticnum;
        for(int i=0;i<randy;i++){
            iter++;

        }
        order[k]=*iter; //Thread 1: Program received signal: "EXC_BAD_ACCESS". 
        iter=spdtics.begin();

    }
return order;
}

This function should take all of the players’ speeds and square them. Then, it puts that many (the squares of speeds) “raffle tickets” into spdtics. It then randomly draws one “ticket” from spdtics, and puts the number of the player who owned the ticket into order. Then, it repeats again until all players have been drawn, not drawing the same player twice. It returns the order in which the players won.

The class Player contains an int speed. I call this function like this:

order=speedRaffle(players,pNum);

where players is vector and pNum is int. What am I doing wrong?

Answer 1

1. You are trying to access element at index k in empty vector order

It crashes because vector order is empty when you call order[k] = *iter;, you should use push_back function instead: order.push_back(*iter);.

2. You use loop for “moving” iterator instead of simple advance call

advance(iter, randy - 1); has same effect as this loop: for(int i=0;i<randy;i++){ iter++; }.

3. You call pow in every single iteration

for (int j=0; j<pow(players[i].speed,2); j++)

Note, that this would be much faster:

int maxspeed = pow(players[i].speed,2);
for (int j = 0; j < maxspeed; j++)

4. Elements in vector can be accessed directly by using index
You don’t need any iterator at all in this case.

5. Passing vector by value instead of passing it by reference

vector<int> speedRaffle(vector<Player>players,int pNum)

Note, that copy of vector players is created every time you call this function. You don’t want to do that. You also don’t want to change this vector inside of function, so declaring this argument as const would be much better:

vector<int> speedRaffle(const vector<Player>& players, int pNum)

6. Your code does not do what you need it to do

“It should take all of the players’ speeds and square them. Then, it puts that many (the squares of speeds) “raffle tickets” into spdtics. It then randomly draws one “ticket” from spdtics, and puts the number of the player who owned the ticket into order. Then, it repeats again until all players have been drawn, not drawing the same player twice. It returns the order in which the players won.”

According to this, your function should look like this:

vector<int> speedRaffle(vector<Player>& players)
{
    // create vector of tickets:
    set<int> ticketOwners;
    vector<int> spdtics;
    for (int i = 0; i < players.size(); i++)
    {
        ticketOwners.insert(i);
        int maxspeed = pow(players[i].speed,2);
        for (int j = 0; j < maxspeed; j++)
        {
            spdtics.push_back(i);
        }
    }
    // draw ticket for every player:
    vector<int> order;
    while (!ticketOwners.empty())
    {
        set<int>::const_iterator to;
        int randy;
        do
        {
            randy = random() % spdtics.size();
            to = ticketOwners.find(spdtics[randy]);
        }
        while (to == ticketOwners.end());
        spdtics.erase(spdtics.begin() + randy);
        order.push_back(*to);
        ticketOwners.erase(to);
    }
    return order;
}

Also note that you don’t need pNum argument if it’s equal to players.size().

Hope this will help you.