Could you let us know how to use stl:map as two dimension array? I wanted to access the individual elements as like mymap[i][j] where I do not know beforehand what the value
You can do
std::map<int, std::map<int, int> > mymap;
For example:
#include <map>
#include <iostream>
int main()
{
std::map<int, std::map<int, int> > mymap;
mymap[9][2] = 7;
std::cout << mymap[9][2] << std::endl;
if (mymap.find(9) != mymap.end() && mymap[9].find(2) != mymap[9].end()) {
std::cout << "My map contains a value for [9][2]" << std::endl;
} else {
std::cout << "My map does not contain a value for [9][2]" << std::endl;
}
return 0;
}
prints 7 on the standard output, followed by "My map contains a value for [9][2]".
An alternative solution to Andrew Stein's which plays nicer with the rest of STL is to simply use
typedef std::map<std::pair<int, int>, int > AMapT;
AMapT mymap;
mymap[std::make_pair(2, 4)] = 10;
...
AMapT::iterator f = mymap.find(std::make_pair(3, 5));
For example, with this way you don't need to chain two calls to map::find
to search for a single value.
Consider using a kd-tree instead. Each level of branching will compare the i an j values in turn. See http://en.wikipedia.org/wiki/Kd-tree.