Why can't I compute the factorial of 100 in scala? - scala

Question

I'm unit testing and I can't compute the factorial of 100. I read that using BigInt would help, but I'm not sure what I'm doing wrong, because it still doesn't work.
Here is my function.
package Factorial{
class factorial
{
def recursive_factorial(number:BigInt) : BigInt =
{
if (number == 0)
return 1
number * recursive_factorial (number - 1)
}
}
}
If I call the function with recursive_factorial(100).. I get an error that the number is too large for an int

Answer 1

The code you posted works fine. The problem is the code you use to test your code (which you posted in a comment to another answer):
assertResult(93326215443944152681699238856266700490715968264‌​38162146859296389521‌​75999932299156089414‌​63976156518286253697‌​92082722375825118521‌​09168640000000000000‌​00000000000) {
factorial.recursive_factorial(100)
}
The problem is that integer literals have type Int and the number 93326... is obviously too large to fit into an int, so you can't use an integer literal to represent it.
One way to create a BigInt of a large number is to write it as a string and convert that to a BigInt. So this will work fine:
assertResult(BigInt("93326215443944152681699238856266700490715968264‌​38162146859296389521‌​75999932299156089414‌​63976156518286253697‌​92082722375825118521‌​09168640000000000000‌​00000000000")) {
factorial.recursive_factorial(100)
}

Answer 2

first of all, your function has the problem of causing a stackoverflow when number is too large. You should write it tail-recursively:
#annotation.tailrec
def recursive_factorial(number: BigInt, result: BigInt = 1): BigInt = {
if (number == 0)
result
else
recursive_factorial(number -1, result * number)
}
otherwise, your code example works perfectly for me and I don't know what your problem is. Probably, you try to write the result of this function (which is a BigInt) into an Int at some point which triggers the error.

Answer 3

The function is just fine. Only limitation is the data type BigInt/BigInteger does not support arbitrary-precision point integer operation which can be done by BigDecimal. It works like a wrapper to BigInt. That's why instead of showing the big number it shows 9.332621544394415268169923885626674E+157. There is a discussion on which data type to use and when in the link below.
Performace of BigDecimal vs. BigInteger and BigDecimal