Standard library

nox.random

A fast general-purpose pseudo-random number generator (xoshiro256). Suitable for simulations, games, shuffling and tests.

Not for security. Use nox.crypto for tokens, salts, keys and anything an attacker must not be able to guess.

Text

import nox.random

Capability: none. The generator is seeded from the system at start-up; call seed for a reproducible sequence.

Functions#

Function	Description
`seed(s)`	seeds the generator; the same seed gives the same sequence
`randint(lo, hi)`	a uniform integer in `[lo, hi]` — both ends inclusive
`random()`	a uniform `float` in `[0.0, 1.0)`
`normal()`	a standard normal (Gaussian) variate: mean 0, standard deviation 1
`exponential(rate)`	an exponentially distributed variate with the given `rate` (mean `1 / rate`)
`shuffle(xs)`	shuffles a `list[T]` in place (Fisher–Yates)

Nox

import nox.random

nox.random.seed(42)
a: int = nox.random.randint(1, 6)
nox.random.seed(42)
b: int = nox.random.randint(1, 6)
print(a == b, a >= 1 and a <= 6)

x: float = nox.random.random()
print(x >= 0.0 and x < 1.0)

xs: list[int] = [1, 2, 3, 4, 5]
nox.random.shuffle(xs)
print(len(xs), sum(xs))
print(nox.random.exponential(2.0) >= 0.0)

Output

True True
True
5 15
True

Notes#

The generator state is per program (per worker under the multi-core scheduler); seeding in one worker does not affect another.
randint requires lo <= hi.
shuffle works for any element type, including classes and nested lists.

Edit this page on GitHub