I am trying to write a Java program to calculate factorial of a large number. It seems BigInteger
is not able to hold such a large number.
The below is
The issue isn't BigInteger, it is your use of a recursive method call (getFactorial()
).
Try this instead, an iterative algorithm:
public static BigInteger getFactorial(int num) {
BigInteger fact = BigInteger.valueOf(1);
for (int i = 1; i <= num; i++)
fact = fact.multiply(BigInteger.valueOf(i));
return fact;
}
The Guava libraries from Google have a highly optimized implementation of factorial that outputs BigIntegers. Check it out. (It does more balanced multiplies and optimizes away simple shifts.)
The problem here looks like its a stack overflow from too much recursion (5000 recursive calls looks like about the right number of calls to blow out a Java call stack) and not a limitation of BigInteger
. Rewriting the factorial function iteratively should fix this. For example:
public static BigInteger factorial(BigInteger n) {
BigInteger result = BigInteger.ONE;
while (!n.equals(BigInteger.ZERO)) {
result = result.multiply(n);
n = n.subtract(BigInteger.ONE);
}
return result;
}
Hope this helps!
Naive implementations of factorial don't work out in real situations.
If you have a serious need, the best thing to do is to write a gamma function (or ln(gamma)
function) that will work not only for integers but is also correct for decimal numbers. Memoize results so you don't have to keep repeating calculations using a WeakHashMap
and you're in business.