diff --git a/Data/List/Unique.hs b/Data/List/Unique.hs --- a/Data/List/Unique.hs +++ b/Data/List/Unique.hs @@ -26,7 +26,7 @@ -- | 'sortUniq' sorts the list and removes the duplicates of elements. Example: -- --- sortUniq "foo bar" == " abfor" +-- > sortUniq "foo bar" == " abfor" sortUniq :: Ord a => [a] -> [a] sortUniq = sort . nub @@ -36,14 +36,14 @@ -- | 'repeated' finds only the elements that are present more than once in the list. Example: -- --- repeated "foo bar" == "o" +-- > repeated "foo bar" == "o" repeated :: Ord a => [a] -> [a] repeated = repeatedBy (>1) -- | The repeatedBy function behaves just like repeated, except it uses a user-supplied equality predicate. -- --- repeatedBy (>2) "This is the test line" == " eist" +-- > repeatedBy (>2) "This is the test line" == " eist" repeatedBy :: Ord a => (Int -> Bool) -> [a] -> [a] repeatedBy p = map head . filterByLength p @@ -51,28 +51,28 @@ -- | 'unique' gets only unique elements, that do not have duplicates. -- It sorts them. Example: -- --- unique "foo bar" == " abfr" +-- > unique "foo bar" == " abfr" unique :: Ord a => [a] -> [a] unique = concat . filterByLength (==1) -- | 'count' of each element in the list, it sorts by keys (elements). Example: -- --- count "foo bar" == [(' ',1),('a',1),('b',1),('f',1),('o',2),('r',1)] +-- > count "foo bar" == [(' ',1),('a',1),('b',1),('f',1),('o',2),('r',1)] count :: Ord a => [a] -> [(a, Int)] count = map (\x -> (head x,length x)) . group . sort -- | 'count_' of each elements in the list, it sorts by their number. Example: -- --- count_ "foo bar" == [(' ',1),('a',1),('b',1),('f',1),('r',1),('o',2)] +-- > count_ "foo bar" == [(' ',1),('a',1),('b',1),('f',1),('r',1),('o',2)] count_ :: Ord a => [a] -> [(a, Int)] count_ = map (\(x,y) -> (y,x)) . sort . map (\x -> (length x,head x)) . group . sort -- | 'countElem' gets the number of occurrences of the specified element. Example: -- --- countElem 'o' "foo bar" == 2 +-- > countElem 'o' "foo bar" == 2 countElem :: Eq a => a -> [a] -> Int countElem x [] = 0 diff --git a/Data/List/UniqueStrict.hs b/Data/List/UniqueStrict.hs --- a/Data/List/UniqueStrict.hs +++ b/Data/List/UniqueStrict.hs @@ -28,14 +28,14 @@ -- | The 'repeatedBy' function behaves just like repeated, except it uses a user-supplied equality predicate. -- --- repeatedBy (>2) "This is the test line" == " eist" +-- > repeatedBy (>2) "This is the test line" == " eist" repeatedBy :: Ord a => (Int -> Bool) -> [a] -> [a] repeatedBy p = MS.keys . MS.filter p . countMap -- | 'repeated' finds only the elements that are present more than once in the list. Example: -- --- repeated "foo bar" == "o" +-- > repeated "foo bar" == "o" repeated :: Ord a => [a] -> [a] repeated = repeatedBy (>1) @@ -43,21 +43,21 @@ -- | 'unique' gets only unique elements, that do not have duplicates. -- It sorts them. Example: -- --- unique "foo bar" == " abfr" +-- > unique "foo bar" == " abfr" unique :: Ord a => [a] -> [a] unique = repeatedBy (==1) -- | 'count' of each element in the list, it sorts by keys (elements). Example: -- --- count "foo bar" == [(' ',1),('a',1),('b',1),('f',1),('o',2),('r',1)] +-- > count "foo bar" == [(' ',1),('a',1),('b',1),('f',1),('o',2),('r',1)] count :: Ord a => [a] -> [(a, Int)] count = MS.toList . countMap -- | 'count_' of each elements in the list, it sorts by their number. Example: -- --- count_ "foo bar" == [(' ',1),('a',1),('b',1),('f',1),('r',1),('o',2)] +-- > count_ "foo bar" == [(' ',1),('a',1),('b',1),('f',1),('r',1),('o',2)] count_ :: Ord a => [a] -> [(a, Int)] count_ = fromIntMap . toIntMap . MS.toList . countMap diff --git a/Data/List/UniqueUnsorted.hs b/Data/List/UniqueUnsorted.hs new file mode 100644 --- /dev/null +++ b/Data/List/UniqueUnsorted.hs @@ -0,0 +1,66 @@ +----------------------------------------------------------------------------- +-- | +-- Module : Data.List.UniqueUnsorted +-- Copyright : (c) Volodymyr Yaschenko +-- License : BSD3 +-- +-- Maintainer : ualinuxcn@gmail.com +-- Stability : Unstable +-- Portability : portable +-- +-- Library provides functions to find unique and duplicate elements in the list. +-- Unlike Unique or UniqueStrict modules this one uses Data.HashMap.Strict for calculation. +-- +-- The elements in the list can be unsorted (do not have an instance of Ord class, but Hashable is needed). +-- This implementation is good for ByteStrings. + +module Data.List.UniqueUnsorted + ( repeated + , repeatedBy + , unique + , count + , count_ ) where + +import Data.Hashable +import qualified Data.HashMap.Strict as HS (HashMap (..),fromListWith,toList,filter,keys) +import qualified Data.IntMap.Strict as IM (IntMap (..), fromListWith,toList) + +countMap :: (Hashable a, Eq a) => [a] -> HS.HashMap a Int +countMap = HS.fromListWith (+) . flip zip (repeat 1) + +-- | The 'repeatedBy' function behaves just like 'repeated', except it uses a user-supplied equality predicate. +-- +-- > repeatedBy (>2) "This is the test line" == " stei" + +repeatedBy :: (Hashable a, Eq a) => (Int -> Bool) -> [a] -> [a] +repeatedBy p = HS.keys . HS.filter p . countMap + +-- | 'repeated' finds only the elements that are present more than once in the list. Example: +-- +-- > repeated "foo bar" == "o" + +repeated :: (Hashable a, Eq a) => [a] -> [a] +repeated = repeatedBy (>1) + +-- | 'unique' gets only unique elements, that do not have duplicates. +-- +-- > unique "foo bar" == " abrf" + +unique :: (Hashable a, Eq a) => [a] -> [a] +unique = repeatedBy (==1) + +-- | 'count' of each element in the list. Example: +-- +-- > count "This is the test line" == [(' ',4),('s',3),('T',1),('t',3),('e',3),('h',2),('i',3),('l',1),('n',1)] + +count :: (Hashable a, Eq a) => [a] -> [(a, Int)] +count = HS.toList . countMap + +-- | 'count_' of each elements in the list, it sorts by their number. Example: +-- +-- > count_ "This is the test line" == [('n',1),('l',1),('T',1),('h',2),('i',3),('e',3),('t',3),('s',3),(' ',4)] + +count_ :: (Hashable a, Eq a) => [a] -> [(a, Int)] +count_ = fromIntMap . toIntMap . HS.toList . countMap + where toIntMap = IM.fromListWith (++) . map (\(x,y) -> (y,[x])) + fromIntMap = concatMap (\(x,y) -> zip y $ repeat x) . IM.toList diff --git a/Unique.cabal b/Unique.cabal --- a/Unique.cabal +++ b/Unique.cabal @@ -10,8 +10,7 @@ -- PVP summary: +-+------- breaking API changes -- | | +----- non-breaking API additions -- | | | +--- code changes with no API change ---version: 0.2.0.5 -version: 0.3 +version: 0.4 -- A short (one-line) description of the package. synopsis: It provides the functionality like unix "uniq" utility @@ -50,7 +49,8 @@ library -- Modules exported by the library. exposed-modules: Data.List.Unique, - Data.List.UniqueStrict + Data.List.UniqueStrict, + Data.List.UniqueUnsorted -- Modules included in this library but not exported. -- other-modules: @@ -60,7 +60,9 @@ -- Other library packages from which modules are imported. build-depends: base >=4.0 && <= 4.8, - containers >= 0.4 && <= 0.9 + containers , + hashable , + unordered-containers -- Directories containing source files. -- hs-source-dirs: