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
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45

/// Gen raw gaussian value with distribution mean at 0.
pub fn gen_raw<R>(mut rng: R) -> f64
where
    R: rand::Rng,
{
    rng.sample(rand_distr::StandardNormal)
}

/// Generates a raw gaussian value between [0.0, 1.0) whose distribution's mean is at `value` with
/// the given amount of `randomness` between [0.0, 1.0).
///
/// **Panic**s if the `value` is less than 0.0 or greater than or equal to 1.0.
///
/// **Panic**s if the `randomness` is less than 0.0 or greater than or equal to 1.0.
pub fn gen<R>(mut rng: R, value: f64, randomness: f64) -> f64
where
    R: rand::Rng,
{
    assert!(value >= 0.0);
    assert!(value < 1.0);
    assert!(randomness >= 0.0);
    assert!(randomness <= 1.0);

    // If there is no randomness, return the value as it was given.
    if randomness == 0.0 {
        return value;
    }

    // If there is complete randomness, generate a uniform distribution value.
    if randomness == 1.0 {
        return rand::Rng::gen_range(&mut rng, 0.0, 1.0);
    }

    // Offset the value over the normal distributions
    let offset_value = value * 2.0 - 1.0;

    // Keep attempting values until we get one that falls within the range.
    loop {
        let normal = gen_raw(&mut rng);
        let attempt = normal * randomness + offset_value;
        if -1.0 <= attempt && attempt < 1.0 {
            return (attempt + 1.0) * 0.5;
        }
    }
}