Introduce new SplitGen type class
#160
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
lehins
force-pushed
the
lehins/SplitGen
branch
2 times, most recently
from
3411863 to
2cc5a88
Compare
|
+1 Following will cause compile failure for instances which don't define default split :: SplitGen g => g -> (g, g)
split = splitGen |
lehins
force-pushed
the
lehins/SplitGen
branch
from
2cc5a88 to
6ba16ae
Compare
lehins
added a commit
that referenced
this pull request
Jan 28, 2024
lehins
added a commit
that referenced
this pull request
Jan 28, 2024
lehins
added a commit
that referenced
this pull request
Oct 26, 2024
lehins
added a commit
that referenced
this pull request
Oct 27, 2024
netbsd-srcmastr
pushed a commit
to NetBSD/pkgsrc
that referenced
this pull request
Jan 29, 2025
# 1.3.0 * Improve floating point value generation and avoid degenerate cases: [#172](haskell/random#172) * Add `Uniform` instance for `Maybe` and `Either`: [#167](haskell/random#167) * Add `Seed`, `SeedGen`, `seedSize`, `seedSizeProxy`, `mkSeed` and `unSeed`: [#162](haskell/random#162) * Add `mkSeedFromByteString`, `unSeedToByteString`, `withSeed`, `withSeedM`, `withSeedFile`, `seedGenTypeName`, `nonEmptyToSeed`, `nonEmptyFromSeed`, `withSeedM`, `withSeedMutableGen` and `withSeedMutableGen_` * Add `SplitGen` and `splitGen`: [#160](haskell/random#160) * Add `unifromShuffleList` and `unifromShuffleListM`: [#140](haskell/random#140) * Add `uniformWordR`: [#140](haskell/random#140) * Add `mkStdGen64`: [#155](haskell/random#155) * Add `uniformListRM`, `uniformList`, `uniformListR`, `uniforms` and `uniformRs`: [#154](haskell/random#154) * Add compatibility with recently added `ByteArray` to `base`: [#153](haskell/random#153) * Switch to using `ByteArray` for type class implementation instead of `ShortByteString` * Add `unsafeUniformFillMutableByteArray` to `RandomGen` and a helper function `defaultUnsafeUniformFillMutableByteArray` that makes implementation for most instances easier. * Add `uniformByteArray`, `uniformByteString` and `uniformFillMutableByteArray` * Deprecate `genByteString` in favor of `uniformByteString` * Add `uniformByteArrayM` to `StatefulGen` * Add `uniformByteStringM` and `uniformShortByteStringM` * Deprecate `System.Random.Stateful.uniformShortByteString` in favor of `uniformShortByteStringM` for consistent naming and a future plan of removing it from `StatefulGen` type class * Add a pure `System.Random.uniformShortByteString` generating function. * Deprecate `genShortByteString` in favor of `System.Random.uniformShortByteString` * Expose a helper function `fillByteArrayST`, that can be used for defining implementation for `uniformByteArrayM` * Deprecate `genShortByteStringST` and `genShortByteStringIO` in favor of `fillByteArrayST` * Improve `FrozenGen` interface: [#149](haskell/random#149) * Move `thawGen` from `FreezeGen` into the new `ThawGen` type class. Fixes an issue with an unlawful instance of `StateGen` for `FreezeGen`. * Add `modifyGen` and `overwriteGen` to the `FrozenGen` type class * Switch `splitGenM` to use `SplitGen` and `FrozenGen` instead of deprecated `RandomGenM` * Add `splitMutableGenM` * Switch `randomM` and `randomRM` to use `FrozenGen` instead of `RandomGenM` * Deprecate `RandomGenM` in favor of a more powerful `FrozenGen` * Add `isInRangeOrd` and `isInRangeEnum` that can be used for implementing `isInRange`: [#148](haskell/random#148) * Add `isInRange` to `UniformRange`: [#78](haskell/random#78) * Add default implementation for `uniformRM` using `Generics`: [#92](haskell/random#92)
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fixes #97
I've been thinking for a while about this and I think the correct path forward is to introduce a separate class for splittable generators, which I called
SplitGenand deprecate the oldsplitfunction.The benefit of this approach is that current user code does not break and gives a graceful path to upgrade to the new type class.
Other approaches were considered in #94 and discussed at length in idontgetoutmuch#7
I highly doubt that we will provide a generic approach of using some hash function for implementing PRNG splitting. I would rather prefer such functionality be in a separate library that other packages that want to implement
SplitGencan opt-in into using.