diff --git a/CHANGELOG.md b/CHANGELOG.md deleted file mode 100644 --- a/CHANGELOG.md +++ /dev/null @@ -1,12 +0,0 @@ -## Bitset-word8 0.1.1.0 - -* Changed backend from ByteString to unboxed Word64 based on benchmark result. -* Added benchmarks with some backend implementations. - -## Bitset-word8 0.1.0.1 - -* Minor modifications in documents. - -## Bitset-word8 0.1.0.0 - -* Initial release. diff --git a/ChangeLog.md b/ChangeLog.md new file mode 100644 --- /dev/null +++ b/ChangeLog.md @@ -0,0 +1,16 @@ +## Bitset-word8 0.1.1.1 + +* Move to hpack from hand written cabal project file. + +## Bitset-word8 0.1.1.0 + +* Changed backend from ByteString to unboxed Word64 based on benchmark result. +* Added benchmarks with some backend implementations. + +## Bitset-word8 0.1.0.1 + +* Minor modifications in documents. + +## Bitset-word8 0.1.0.0 + +* Initial release. diff --git a/bench/Bench.hs b/bench/Bench.hs --- a/bench/Bench.hs +++ b/bench/Bench.hs @@ -1,13 +1,11 @@ {-# LANGUAGE DeriveAnyClass #-} {-# LANGUAGE DeriveGeneric #-} -{-# LANGUAGE MagicHash #-} -{-# LANGUAGE UnboxedTuples #-} import Control.DeepSeq (NFData, force) import Criterion.Main import Data.Bits (testBit) import Data.ByteString (ByteString, index, pack) -import Data.Vector.Unboxed (Vector (..), (!), fromList) +import Data.Vector.Unboxed (Vector, fromList, (!)) import Data.Word (Word64, Word8) import GHC.Generics (Generic) @@ -63,7 +61,7 @@ newtype BitSetWord8Vector64 = BitSetWord8Vector64 (Vector Word64) deriving (Eq, Generic, NFData, Show) memberVector64 :: BitSetWord8Vector64 -> Word8 -> Bool -memberVector64 (BitSetWord8Vector64 v) w = testBit (v ! (fromIntegral (w `div` 64))) (fromIntegral (w `mod` 64)) +memberVector64 (BitSetWord8Vector64 v) w = testBit (v ! fromIntegral (w `div` 64)) (fromIntegral (w `mod` 64)) allZeroVector64 :: BitSetWord8Vector64 allZeroVector64 = force $ BitSetWord8Vector64 $ fromList [ 0,0,0,0 ] @@ -77,7 +75,7 @@ newtype BitSetWord8Vector8 = BitSetWord8Vector8 (Vector Word8) deriving (Eq, Generic, NFData, Show) memberVector8 :: BitSetWord8Vector8 -> Word8 -> Bool -memberVector8 (BitSetWord8Vector8 v) w = testBit (v ! (fromIntegral (w `div` 8))) (fromIntegral (w `mod` 8)) +memberVector8 (BitSetWord8Vector8 v) w = testBit (v ! fromIntegral (w `div` 8)) (fromIntegral (w `mod` 8)) allZeroVector8 :: BitSetWord8Vector8 allZeroVector8 = force $ BitSetWord8Vector8 $ fromList [ 0,0,0,0,0,0,0,0 diff --git a/bitset-word8.cabal b/bitset-word8.cabal --- a/bitset-word8.cabal +++ b/bitset-word8.cabal @@ -1,55 +1,84 @@ -name: bitset-word8 -version: 0.1.1.0 -synopsis: Space efficient set of Word8 and some pre-canned sets useful for parsing HTTP -description: This package is intended to provide O(1) membership test on any subset of ASCII - and Latin-1 character set in order to write efficient HTTP related parser. -homepage: https://github.com/nshimaza/bitset-word8#readme -license: MIT -license-file: LICENSE -author: Naoto Shimazaki -maintainer: Naoto.Shimazaki@gmail.com -copyright: Copyright: (c) 2017 Naoto Shimazaki -category: Data -build-type: Simple -extra-source-files: CHANGELOG.md - , README.md -cabal-version: >=1.10 +cabal-version: 1.12 +-- This file has been generated from package.yaml by hpack version 0.31.0. +-- +-- see: https://github.com/sol/hpack +-- +-- hash: 6ddab4e773037869bcf89a8b168d28296949766d6abfd03c9bfa0886b18700dc + +name: bitset-word8 +version: 0.1.1.1 +synopsis: Space efficient set of Word8 and some pre-canned sets useful for parsing HTTP +description: Please see the README on GitHub at +category: Concurrency +homepage: https://github.com/nshimaza/bitset-word8#readme +bug-reports: https://github.com/nshimaza/bitset-word8/issues +author: Naoto Shimazaki +maintainer: Naoto.Shimazaki@gmail.com +copyright: 2017,2018 Naoto Shimazaki +license: MIT +license-file: LICENSE +build-type: Simple +extra-source-files: + README.md + ChangeLog.md + +source-repository head + type: git + location: https://github.com/nshimaza/bitset-word8 + library - hs-source-dirs: src - exposed-modules: Data.BitSetWord8 - , Data.BitSetWord8.CharSets - , Data.BitSetWord8.Internal - build-depends: base >= 4.7 && < 5 - , containers - , template-haskell - , th-lift-instances - default-language: Haskell2010 + exposed-modules: + Data.BitSetWord8 + Data.BitSetWord8.CharSets + Data.BitSetWord8.Internal + other-modules: + Paths_bitset_word8 + hs-source-dirs: + src + ghc-options: -Wall -Wcompat -Wincomplete-record-updates -Wincomplete-uni-patterns -Wredundant-constraints + build-depends: + base >=4.7 && <5 + , containers + , template-haskell + , th-lift-instances + default-language: Haskell2010 test-suite bitset-word8-test - type: exitcode-stdio-1.0 - hs-source-dirs: test - main-is: Spec.hs - other-modules: Data.BitSetWord8Spec - build-depends: base - , hspec - , QuickCheck - , bitset-word8 - ghc-options: -threaded -rtsopts -with-rtsopts=-N - default-language: Haskell2010 + type: exitcode-stdio-1.0 + main-is: Spec.hs + other-modules: + Data.BitSetWord8Spec + Paths_bitset_word8 + hs-source-dirs: + test + ghc-options: -Wall -Wcompat -Wincomplete-record-updates -Wincomplete-uni-patterns -Wredundant-constraints -threaded -rtsopts -with-rtsopts=-N + build-depends: + QuickCheck + , base >=4.7 && <5 + , bitset-word8 + , containers + , hspec + , template-haskell + , th-lift-instances + default-language: Haskell2010 benchmark bitset-word8-bench - type: exitcode-stdio-1.0 - hs-source-dirs: bench - main-is: Bench.hs - build-depends: base - , bytestring - , criterion - , deepseq - , vector - ghc-options: -threaded -rtsopts -with-rtsopts=-N - default-language: Haskell2010 - -source-repository head - type: git - location: https://github.com/nshimaza/bitset-word8 + type: exitcode-stdio-1.0 + main-is: Bench.hs + other-modules: + Paths_bitset_word8 + hs-source-dirs: + bench + ghc-options: -Wall -Wcompat -Wincomplete-record-updates -Wincomplete-uni-patterns -Wredundant-constraints -threaded -rtsopts -with-rtsopts=-N + build-depends: + base >=4.7 && <5 + , bitset-word8 + , bytestring + , containers + , criterion + , deepseq + , template-haskell + , th-lift-instances + , vector + default-language: Haskell2010 diff --git a/src/Data/BitSetWord8.hs b/src/Data/BitSetWord8.hs --- a/src/Data/BitSetWord8.hs +++ b/src/Data/BitSetWord8.hs @@ -40,8 +40,6 @@ @ -} -{-# LANGUAGE DeriveLift #-} - module Data.BitSetWord8 ( -- * Types @@ -75,5 +73,5 @@ , member ) where -import Data.BitSetWord8.Internal -import Data.BitSetWord8.CharSets +import Data.BitSetWord8.CharSets +import Data.BitSetWord8.Internal diff --git a/src/Data/BitSetWord8/CharSets.hs b/src/Data/BitSetWord8/CharSets.hs --- a/src/Data/BitSetWord8/CharSets.hs +++ b/src/Data/BitSetWord8/CharSets.hs @@ -20,7 +20,7 @@ , rfc7230QuotedPair ) where -import Language.Haskell.TH.Syntax (Lift, lift) +import Language.Haskell.TH.Syntax (lift) import Data.BitSetWord8.Internal @@ -39,6 +39,3 @@ -- | quoted-string of RFC7230 in BitSetWord8. Evaluated at compile time. rfc7230QuotedPair :: BitSetWord8 rfc7230QuotedPair = $(lift $ fromList rfc7230QuotedPair') - - - diff --git a/src/Data/BitSetWord8/Internal.hs b/src/Data/BitSetWord8/Internal.hs --- a/src/Data/BitSetWord8/Internal.hs +++ b/src/Data/BitSetWord8/Internal.hs @@ -10,9 +10,7 @@ This file contains additional useful character sets but they aren't evaluated at compile time. -} -{-# LANGUAGE DeriveLift #-} -{-# LANGUAGE MagicHash #-} -{-# LANGUAGE UnboxedTuples #-} +{-# LANGUAGE DeriveLift #-} module Data.BitSetWord8.Internal where @@ -21,8 +19,7 @@ import Data.List (foldl', splitAt) import Data.Semigroup ((<>)) import qualified Data.Set as Set (Set, fromList, member) -import Data.Word (Word8, Word64) -import Instances.TH.Lift +import Data.Word (Word64, Word8) import Language.Haskell.TH.Syntax (Lift) @@ -115,6 +112,7 @@ doMember (BitSetWord8 _ w _ _) 1 ind = testBit w (fromIntegral ind) doMember (BitSetWord8 _ _ w _) 2 ind = testBit w (fromIntegral ind) doMember (BitSetWord8 _ _ _ w) 3 ind = testBit w (fromIntegral ind) + doMember _ _ _ = error "Impossible happen. Word8 `dev` 64 cannot be greater than 3." -- | Convert given list of 'Char' into 'Set' of 'Word8'. Any 'Char' having code point greater than 0xff is ignored. toWord8Set :: [Char] -> Set.Set Word8 @@ -138,6 +136,14 @@ Any 'Char' having code point greater than 0xff is ignored. -} fromList :: [Char] -> BitSetWord8 -fromList cs = BitSetWord8 w0 w1 w2 w3 +fromList = fromWord64List . toWord64List . toBoolList . toWord8Set where - (w0:w1:w2:w3:_) = toWord64List . toBoolList . toWord8Set $ cs + -- Convert 4 packed 'Word64' list into single 'BitSetWord8' + fromWord64List :: [Word64] -> BitSetWord8 + fromWord64List (w0:w1:w2:w3:_) = BitSetWord8 w0 w1 w2 w3 + fromWord64List [_, _, _] = raiseError + fromWord64List [_, _] = raiseError + fromWord64List [_] = raiseError + fromWord64List [] = raiseError + + raiseError = error "Impossible happen. Arg of fromList must be converted to 4 elements list." diff --git a/test/Data/BitSetWord8Spec.hs b/test/Data/BitSetWord8Spec.hs --- a/test/Data/BitSetWord8Spec.hs +++ b/test/Data/BitSetWord8Spec.hs @@ -12,35 +12,34 @@ spec :: Spec spec = do describe "fromList" $ do - prop "creates empty BitSetWord8 from empty list" $ forAll (choose (0, 0xff :: Word8)) $ \n -> do + prop "creates empty BitSetWord8 from empty list" $ forAll (choose (0, 0xff :: Word8)) $ \n -> member (fromList []) n `shouldBe` False - prop "creates fullfilled BitSetWord8 from fullfilled list" $ forAll (choose (0, 0xff :: Word8)) $ \n -> do + prop "creates fullfilled BitSetWord8 from fullfilled list" $ forAll (choose (0, 0xff :: Word8)) $ \n -> member (fromList [chr 0 .. chr 0xff]) n `shouldBe` True describe "rfc5234Digit'" $ do - prop "recognizes any digit" $ forAll (choose (0x30, 0x39 :: Word8)) $ \n -> do + prop "recognizes any digit" $ forAll (choose (0x30, 0x39 :: Word8)) $ \n -> member (fromList rfc5234Digit') n `shouldBe` True - prop "unrecognizes any non-digit characters (lower)" $ forAll (choose (0, 0x2f :: Word8)) $ \n -> do + prop "unrecognizes any non-digit characters (lower)" $ forAll (choose (0, 0x2f :: Word8)) $ \n -> member (fromList rfc5234Digit') n `shouldBe` False - prop "unrecognizes any non-digit characters (higher)" $ forAll (choose (0x3a, 0xff :: Word8)) $ \n -> do + prop "unrecognizes any non-digit characters (higher)" $ forAll (choose (0x3a, 0xff :: Word8)) $ \n -> member (fromList rfc5234Digit') n `shouldBe` False describe "rfc5234Alpha'" $ do - prop "recognizes any upper case alphabet" $ forAll (choose (0x41, 0x5a :: Word8)) $ \n -> do + prop "recognizes any upper case alphabet" $ forAll (choose (0x41, 0x5a :: Word8)) $ \n -> member (fromList rfc5234Alpha') n `shouldBe` True - prop "recognizes any lower case alphabet" $ forAll (choose (0x61, 0x7a :: Word8)) $ \n -> do + prop "recognizes any lower case alphabet" $ forAll (choose (0x61, 0x7a :: Word8)) $ \n -> member (fromList rfc5234Alpha') n `shouldBe` True - prop "unrecognizes any non-alphabet (lower)" $ forAll (choose (0, 0x40 :: Word8)) $ \n -> do + prop "unrecognizes any non-alphabet (lower)" $ forAll (choose (0, 0x40 :: Word8)) $ \n -> member (fromList rfc5234Alpha') n `shouldBe` False - prop "unrecognizes any non-alphabet (middle)" $ forAll (choose (0x5b, 0x60 :: Word8)) $ \n -> do + prop "unrecognizes any non-alphabet (middle)" $ forAll (choose (0x5b, 0x60 :: Word8)) $ \n -> member (fromList rfc5234Alpha') n `shouldBe` False - prop "unrecognizes any non-alphabet (higher)" $ forAll (choose (0x7b, 0xff :: Word8)) $ \n -> do + prop "unrecognizes any non-alphabet (higher)" $ forAll (choose (0x7b, 0xff :: Word8)) $ \n -> member (fromList rfc5234Alpha') n `shouldBe` False -