random-source

Generic basis for random number generators.

Random number generation based on entropy sources able to produce a small but well-defined set of primitive variates. Also includes facilities for "completing" partial implementations, making it easy to define new entropy sources in a way that is naturally forward-compatible. Now DEPRECATED: see README for further details.

This is now deprecated and it is better to use the random package as the source of randomness.

Here's what you might have written before:

import qualified Data.Random as Random
import qualified Data.Random.Distribution.Bernoulli as Bernoulli

main :: IO ()
main = do
  x <- Random.sample $ Bernoulli.Bernoulli (0.5 :: Double) :: IO Double
  print x

And here's what you should write now (but there are many other options):

import qualified System.Random.Stateful as Random.Stateful
import qualified Data.Random as Random
import qualified Data.Random.Distribution.Bernoulli as Bernoulli
import qualified Control.Monad.Reader as Reader

main :: IO ()
main = do
  stdgen <- Random.Stateful.newIOGenM =<< Random.Stateful.newStdGen
  x <- Reader.runReaderT (Random.sample $ Bernoulli.Bernoulli (0.5 :: Double)) stdgen
        :: IO Double
  print x