liveBook · Manning

StdGen

As explained in Learn Haskell by Example

StdGen is a Haskell generator-state type representing the state of a pseudo-random number generator.

StdGen is Haskell’s typical pseudo-random generator state. Create one from a seed, pass it to a generation function, and keep the returned state for the next value. section 9.1.2 section 9.1.3

Use the generator with the main value-generation functions:

  • random generates a value through Random and returns the updated generator. section 9.1.3
  • randomR generates a Random value within a supplied range. section 9.1.3
  • uniform generates uniformly distributed values through Uniform. section 9.1.3
  • uniformR generates uniformly distributed values within a range. section 9.1.3

Generating a random list

Define randomListN :: Random a => StdGen -> Int -> ([a], StdGen), recursively call random with the updated generator, and return the final state. A separate randomList function can first choose a bounded length with uniformR. section 9.1.3

Listing 9.3 Functions to produce random lists of random sizes
randomListN :: (Random a) => StdGen -> Int -> ([a], StdGen)
randomListN gen n
  | n <= 0 = ([], gen)
  | otherwise =
    let (v, gen') = random gen  #1
        (xs, gen'') = randomListN gen' (n - 1)  #2
     in (v : xs, gen'')

randomList :: (Random a) => StdGen -> Int -> ([a], StdGen)
randomList gen maxVal = randomListN gen' n  #3
  where
    (n, gen') = uniformR (0, maxVal) gen  #4

randomList' :: (Random a) => StdGen -> ([a], StdGen)
randomList' = flip randomList 100  #5

Browse the complete random-list definitions, including bounded list lengths and a convenient randomList' specialization. section 9.1.3

StdGen can also be accessed globally inside IO. getStdGen returns a copy, so code using getStdGen and setStdGen must write the modified generator back. For concurrent use, prefer applyAtomicGen with globalStdGen over these deprecated functions. section 9.1.4

Listing 9.5 An IO action that applies a function to the global pseudo-random generator
applyGlobalStdGen :: (StdGen -> (a, StdGen)) -> IO a
applyGlobalStdGen f = applyAtomicGen f globalStdGen  #1

Browse the compact helper that applies a StdGen transformation to the safer global generator interface. section 9.1.4

How Learn Haskell by Example covers it

How deep it goes
Explains seeding, state threading, polymorphic APIs, random lists, and global access, while deferring generator algorithm internals.
Examples
Demonstrates seeded GHCi generation, recursive random lists, bounded list sizes, and IO actions returning random values.
What it assumes
Assumes basic Haskell functions and IO while introducing generator state, random type classes, and explicit state threading.
What you can do after
Apply it via System.Random to build polymorphic random-value functions and thread generator state through pure code.
How it connects
Connects StdGen to System.Random, RandomGen, Random, Uniform, IO, and concurrency-safe global generator operations.
Cover of Learn Haskell by Example
liveBookmanning.com
Read Learn Haskell by Example on liveBook Get the book at manning.com

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Learn Haskell by Example. The section links point to the parts of the book it draws on.