Changelog of @haskell/vector 0.13.2.0

Changes in version 0.13.2.0

  • Strict boxed vector Data.Vector.Strict and Data.Vector.Strict.Mutable is added (#488). it ensures that all values in the vector are evaluated to WHNF.
  • DoNotUnboxStrict, DoNotUnboxLazy, and DoNotUnboxNormalForm wrapper are added for defining unbox instances for types that contain not unboxable fields. #503, #508
  • spanR and breakR were added #476. They allow parsing vector from the right.
  • We had some improvements on *.Mutable.{next,prev}Permutation{,By} #498:
    • Add *.Mutable.prevPermutation{,By} and *.Mutable.nextPermutationBy
    • Improve time performance. We may now expect good specialization supported by inlining. The implementation has also been algorithmically updated: in the previous implementation the full enumeration of all the permutations of [1..n] took Omega(n*n!), but it now takes O(n!).
    • Add tests for {next,prev}Permutation
    • Add benchmarks for {next,prev}Permutation
  • Cabal >= 3.0 is now required for building package (#481).
  • vector:benchmarks-O2 public sublibrary containing benchmarks is added (#481).
  • Type family Mutable provides instances for arrays from primitive.
  • Various documentation improvements.

Changes in version 0.13.1.0

  • Specialized variants of findIndexR are reexported for all vector types. #469
  • UnboxViaPrim could be used for deriving Unbox instances (V_UnboxViaPrim constructor is exported) #450
  • Fields of Data.Vector.Fusion.Bundle.Size are now strict #456
  • Compatibility with future GHC 9.10 release #462
  • Test suite no longer fails when built with QuickCheck-2.14 #461
  • Doctests now work with current versions of GHC #465
  • Various documentation improvements

Changes in version 0.13.0.0

  • mkType from Data.Vector.Generic is deprecated in favor of Data.Data.mkNoRepType
  • The role signatures on several Vector types were too permissive, so they have been tightened up:
    • The role signature for Data.Vector.Mutable.MVector is now type role MVector nominal representational (previously, both arguments were phantom). #224

    • The role signature for Data.Vector.Primitive.Vector is now type role Vector nominal (previously, it was phantom). The role signature for Data.Vector.Primitive.Mutable.MVector is now type role MVector nominal nominal (previously, both arguments were phantom). #316

    • The role signature for Data.Vector.Storable.Vector is now type role Vector nominal (previous, it was phantom), and the signature for Data.Vector.Storable.Mutable.MVector is now type role MVector nominal nominal (previous, both arguments were phantom). #235

      We pick nominal for the role of the last argument instead of representational since the internal structure of a Storable vector is determined by the Storable instance of the element type, and it is not guaranteed that the Storable instances between two representationally equal types will preserve this internal structure. One consequence of this choice is that it is no longer possible to coerce between Storable.Vector a and Storable.Vector b if a and b are nominally distinct but representationally equal types. We now provide unsafeCoerce{M}Vector and unsafeCast functions to allow this (the onus is on the user to ensure that no Storable invariants are broken when using these functions).

  • Methods of type classes Data.Vector.Generic.Mutable.MVector and Data.Vector.Generic.Vector use concrete monads (ST, etc) istead of being polymorphic (PrimMonad, etc). #335. This makes it possible to derive Unbox with:
    • GeneralizedNewtypeDeriving
    • via UnboxViaPrim and Prim instance
    • via As and IsoUnbox instance: #378
  • Add MonadFix instance for boxed vectors: #312
  • Re-export PrimMonad and RealWorld from mutable vectors: #320
  • Add maximumOn and minimumOn: #356
  • The functions scanl1, scanl1', scanr1, and scanr1' for immutable vectors are now defined when given empty vectors as arguments, in which case they return empty vectors. This new behavior is consistent with the one of the corresponding functions in Data.List. Prior to this change, applying an empty vector to any of those functions resulted in an error. This change was introduced in: #382
  • Change allocation strategy for unfoldrN: #387
  • Remove CPP driven error reporting in favor of HasCallStack: #397
  • Remove redundant Storable constraints on to/from ForeignPtr conversions: #394
  • Add unsafeCast to Primitive vectors: #401
  • Make (!?) operator strict: #402
  • Add readMaybe: #425
  • Add groupBy and group for Data.Vector.Generic and the specialized version in Data.Vector, Data.Vector.Unboxed, Data.Vector.Storable and Data.Vector.Primitive. #427
  • Add toArraySlice and unsafeFromArraySlice functions for conversion to and from the underlying boxed Array: #434

Changes in version 0.12.3.1

  • Bugfix for ghcjs and Double memset for Storable vector: #410
  • Avoid haddock bug: #383
  • Improve haddock and doctests
  • Disable problematic tests with -boundschecks #407

Changes in version 0.12.3.0

  • Fix performance regression due to introduction of keepAlive# primop in ghc-9.0: #372

  • Add monadic functions for mutable vectors: #338

    • Added folds for monadic functions: mapM_, imapM_, forM_, iforM_, foldl, foldl', foldM, foldM', ifoldl, ifoldl', ifoldM, ifoldM'
    • Added modifyM and unsafeModifyM for mutable vectors
    • Added generate and generateM for mutable vectors

Changes in version 0.12.2.0

  • Add MINIMAL pragma to Vector & MVector type classes: #11
  • Export unstreamM fromfrom Data.Vector.Generic: #70
  • Added unfoldrExactN and unfoldrExactNM: #140
  • Added iforM and iforM_: #262
  • Added MonadFix instance for boxed vectors: #178
  • Added uncons and unsnoc: #212
  • Added foldMap and foldMap': #263
  • Added isSameVector for storable vectors
  • Added toArray, fromArray, toMutableArray and fromMutableArray
  • Added iscanl, iscanl', iscanr, iscanr' to Primitive, Storable and Unboxed
  • Added izipWithM, izipWithM_, imapM and imapM_ to Primitive and Storable
  • Added ifoldM, ifoldM', ifoldM_ and ifoldM'_ to Primitive and Storable
  • Added eqBy and cmpBy
  • Added findIndexR to Generic: #172
  • Added catMaybes: #329
  • Added mapMaybeM and imapMaybeM: #183

Changes in version 0.12.1.2

  • Fix for lost function Data.Vector.Generic.mkType: #287

Changes in version 0.12.1.1 (deprecated)

  • add semigrioups dep to test suite so CI actually runs again on GHC < 8

Changes in version 0.12.1.0 (deprecated)

  • Fix integer overflows in specializations of Bundle/Stream enumFromTo on Integral types
  • Fix possibility of OutOfMemory with take and very large arguments.
  • Fix slice function causing segfault and not checking the bounds properly.
  • updated specialization rule for EnumFromTo on Float and Double to make sure it always matches the version in GHC Base (which changed as of 8.6) Thanks to Aleksey Khudyakov @Shimuuar for this fix.
  • fast rejection short circuiting in eqBy operations
  • the O2 test suite now has reasonable memory usage on every GHC version, special thanks to Alexey Kuleshevich (@lehins).
  • The Mutable type family is now injective on GHC 8.0 or later.
  • Using empty Storable vectors no longer results in division-by-zero errors.
  • The Data instances for Vector types now have well defined implementations for toConstr, gunfold, and dataTypeOf.
  • New function: partitionWith.
  • Add Unbox instances for Identity, Const, Down, Dual, Sum, Product, Min, Max, First, Last, WrappedMonoid, Arg, Any, All, Alt, and Compose.
  • Add NFData1 instances for applicable Vector types.

Changes in version 0.12.0.3

  • Monad Fail support

Changes in version 0.12.0.2

  • Fixes issue #220, compact heap operations crashing on boxed vectors constructed using traverse.
  • backport injective type family support
  • Cleanup the memset code internal to storable vector modules to be compatible with future Primitive releases

Changes in version 0.12.0.1

  • Make sure length can be inlined
  • Include modules that test-suites depend on in other-modules

Changes in version 0.12.0.0

  • Documentation fixes/additions
  • New functions: createT, iscanl/r, iterateNM, unfoldrM, uniq
  • New instances for various vector types: Semigroup, MonadZip
  • Made Storable vectors respect memory alignment
  • Changed some macros to ConstraintKinds
    • Dropped compatibility with old GHCs to support this
  • Add Eq1, Ord1, Show1, and Read1 Vector instances, and related helper functions.
  • Relax context for Unbox (Complex a).

Changes in version 0.11.0.0

  • Define Applicative instances for Data.Vector.Fusion.Util.{Box,Id}
  • Define non-bottom fail for instance Monad Vector
  • New generalized stream fusion framework
  • Various safety fixes
    • Various overflows due to vector size have been eliminated
    • Memory is initialized on creation of unboxed vectors
  • Changes to SPEC usage to allow building under more conditions

Changes in version 0.10.12.3

  • Allow building with primtive-0.6

Changes in version 0.10.12.2

  • Add support for deepseq-1.4.0.0

Changes in version 0.10.12.1

  • Fixed compilation on non-head GHCs

Changes in version 0.10.12.0

  • Export MVector constructor from Data.Vector.Primitive to match Vector's (which was already exported).

  • Fix building on GHC 7.9 by adding Applicative instances for Id and Box

Changes in version 0.10.11.0

  • Support OverloadedLists for boxed Vector in GHC >= 7.8

Changes in version 0.10.10.0

  • Minor version bump to rectify PVP violation occured in 0.10.9.3 release

Changes in version 0.10.9.3 (deprecated)

  • Add support for OverloadedLists in GHC >= 7.8

Changes in version 0.10.9.2

  • Fix compilation with GHC 7.9

Changes in version 0.10.9.1

  • Implement poly-kinded Typeable

Changes in version 0.10.0.1

  • Require primitive to include workaround for a GHC array copying bug

Changes in version 0.10

  • NFData instances
  • More efficient block fills
  • Safe Haskell support removed