hmath 0.1.17

Big Integers and Rational Numbers
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29

use crate::{Ratio, exp_iter, ln_iter};

/// It returns `a^b`. It gets more accurate as `iter` gets bigger. If `b` is an integer, try `Ratio::pow_i32` instead. It panics when `a` is less than 0. `0^0` is 0.
pub fn pow_iter(a: &Ratio, b: &Ratio, iter: usize) -> Ratio {

    if a.is_zero() {
        return Ratio::zero();
    }

    // a^b = e^(b*ln(a))
    exp_iter(&b.mul_rat(&ln_iter(a, iter)), iter)
}

#[cfg(test)]
mod tests {
    use crate::{Ratio, pow_iter};

    #[test]
    fn pow_iter_test() {
        assert_eq!(Ratio::zero(), pow_iter(&Ratio::zero(), &Ratio::zero(), 1));
        assert_eq!(Ratio::one(), pow_iter(&Ratio::one(), &Ratio::zero(), 1));
        assert_eq!(Ratio::zero(), pow_iter(&Ratio::zero(), &Ratio::one(), 1));
        assert_eq!(Ratio::one(), pow_iter(&Ratio::one(), &Ratio::one(), 1));
        assert_eq!("4617933561212708776.4", pow_iter(&Ratio::from_i32(2), &Ratio::from_denom_and_numer_i32(512, 62 * 512 + 1), 12).to_approx_string(21));
        assert_eq!("3.162277660168", pow_iter(&Ratio::from_i32(10), &Ratio::from_ieee754_f32(0.5).unwrap(), 12).to_approx_string(14));
        assert_eq!("16777215.99999", pow_iter(&Ratio::from_i32(8), &Ratio::from_i32(8), 12).to_approx_string(14));
    }

}