Retrieve index of element of array stored in vector - c++

Question

I have a 2D array used to store non-repeated values and some entries are randomly picked and push_back-ed into a vector as favorite list.
int num[10][10];
vector<int> fav_list;
int retrieved_i, retrieved_j, retrieve_vector_position;
for(i=0;i<10;i++) for(j=0;j<10;j++)
// ...assign numbers to num...
fav_list.push_back(num[2][3]);
fav_list.push_back(num[4][7]);
fav_list.push_back(num[6][2]);
//...push_back more random selected num[...][...] into fav_list...
The problem is, how can I retrieve the i, j index of particular fav_list[...]?
I've tried to make struct struct Num{int value, index_i, index_j;}num[10][10]; so that I can do in this way
retrieved_i = fav_list[retrieve_vector_position].index_i;
retrieved_j = fav_list[retrieve_vector_position].index_j;
but I wish to know is there any other better/ efficient ways?

Answer 1

Using a plain vector to store your 2D array would solve the problem. You could access elements in the vector by calculating absolute index (i * row_len + j) and store in fav_list absolute indices.
Also, you may want to use std::unordered_map for fav_list. Generally, hash tables is the most efficient data structure for such caches.

Answer 2

There are a few possibilities depending on how often you want to access the i & j indices / the favorite number itself.
One approach is, to save the indices instead of the number (or additional to it). With this approach, more memory is required but the the time to access the indices will be constant, regardless how big your array becomes. This uses std::pair to store 2 elements in your favorite vector.
#include <vector>
#include <iostream>
#include <utility>
using namespace std;
int main(int argc, char* argv[]) {
int num[10][10];
vector<std::pair<int, int>> fav_list;
for (int i = 0; i < 10; i++) {
for (int j = 0; j < 10; j++) {
if (/* your condition for favorites */) {
fav_list.push_back(num[i][j]);
}
}
}
/* example get the indices of the first favorite */
cout << "i: " << fav_list[0].first << "j: " << fav_list[0].second << endl;
/* example get the first favorite */
cout << num[fav_list[0].first][fav_list[0].second] << endl;
return 0;
}
Another approach is to "lookup" the indices, when you require it: it has the condition, that one number is not multiple times contained in your num[][] array (otherwise the first entry is found). There is no additional memory overhead required, but the time to lookup the indices will increase when your array gets bigger.
#include <vector>
#include <iostream>
using namespace std;
int main(int argc, char* argv[]) {
int num[10][10];
vector<int> fav_list;
for (int i = 0; i < 10; i++) {
for (int j = 0; j < 10; j++) {
if (/* your condition for favorites */) {
fav_list.push_back(num[i][j]);
}
}
}
/* example get the indices of the first favorite */
int indexI = -1, indexJ = -1;
for (int i = 0; i < 10; i++) {
for (int j = 0; j < 10; j++) {
if (fav_list[0] == num[i][j]) {
indexI = i;
indexJ = j;
break;
}
}
}
cout << "i: " << indexI << "j: " << indexJ << endl;
/* example get the first favorite */
cout << fav_list[0] << endl;
return 0;
}

Answer 3

Instead of storing 3 variable value, x and y just store a single unsigned int via which you can retrieve x and y.
struct fav_list
{
unsigned int total_rows;
unsigned int total_columns;
fav_list(unsigned int _rows, unsigned int _columns)
{
total_rows = _rows;
total_columns = _columns;
}
unsigned int get_x(unsigned int _index)
{
return v[_index] / total_columns;
}
unsigned int get_y(unsigned int _index)
{
return v[_index] % total_columns;
}
void append_xy_to_list(unsigned int _x, unsigned int _y)
{
v.push_back(_x * total_columns + _y);
}
vector <unsigned int> v;
};
fav_list f(10, 10);
for(x = 0; x < 10; ++x)
{
for(y = 0; y < 10; ++y)
{
//suppose you want to store the indexes of element num[x][y] then:
f.append_xy_to_list(x, y);
}
}
retrieved_i = f.get_x(retrieve_vector_position);
retrieved_j = f.get_y(retrieve_vector_position);