One possible approach to this problem can be divide & conquer. Step of following describes the approach:
- Say m is the minimum & n is the maximum, within what i wanna get x number of randoms
- Choose a random p between m & n. Save it to an array of answer. decrease x by 1 as we get one answer to our problem.
- Now take a q a random number between m & p-1, another r a random number between p+1 & n. Fill up the answer array with q & r decrease x 1 for q and another 1 for the r.
- Now carry on this process recursively, until the lower bound (m) & higher bound (n) becomes equal or x becomes 0.
Benefit: benefit of this approach is that, in worst case, it's runtime will be O(x), where x is the number of random number required. The best case scenarion is also o(x), as i have to find at least n number of random. These two comprise average case to θ(x) complexity.
import java.util.Random;
class GenerateDistinctRandom{
static int alreadyPut = 0;
static Random rand = new Random();
public static int[] generateDistinctRandom(int howMany, int rangeMin, int rangeMax)
{
int randomNumbers[] = new int[howMany];
GenerateDistinctRandom.recursiveRandomGenerator(rangeMin, rangeMax, randomNumbers, howMany);
return randomNumbers;
}
private static void recursiveRandomGenerator(int rangeMin, int rangeMax, int[] storage ,int storageSize)
{
if(rangeMax - rangeMin <= 0 || GenerateDistinctRandom.alreadyPut == storageSize)
{
return ;
}
int randomNumber = GenerateDistinctRandom.rand.nextInt(rangeMax-rangeMin) + rangeMin;
storage[GenerateDistinctRandom.alreadyPut] = randomNumber;
GenerateDistinctRandom.alreadyPut++;
//calling the left side of the recursion
recursiveRandomGenerator(rangeMin, randomNumber - 1, storage, storageSize);
recursiveRandomGenerator(randomNumber + 1, rangeMax, storage, storageSize);
}
public static void main(String []args){
int howMany = 5;
int distinctNumber[] = GenerateDistinctRandom.generateDistinctRandom(howMany 0, 9);
for(int i = 0;i < howMany;i++)
{
System.out.println(distinctNumber[i]);
}
}
}