{-# LANGUAGE NoImplicitPrelude #-} {-# LANGUAGE FlexibleContexts #-} {-# LANGUAGE TypeFamilies #-} module ClassyPrelude ( -- * CorePrelude module CorePrelude , undefined -- * Standard -- ** Monoid , (++) -- ** Semigroup , Semigroup (..) , WrappedMonoid -- ** Monad , module Control.Monad -- ** Mutable references , module Control.Concurrent.MVar.Lifted , module Data.IORef.Lifted -- ** Debugging , trace , traceShow , traceId , traceM , traceShowId , traceShowM , assert -- ** Time (since 0.6.1) , module Data.Time , defaultTimeLocale -- ** Generics (since 0.8.1) , Generic -- * Poly hierarchy , module Data.Foldable , module Data.Traversable -- * Mono hierarchy , module Data.MonoTraversable , module Data.Sequences , module Data.Sequences.Lazy , module Data.Textual.Encoding , module Data.Containers , module Data.Builder , module Data.MinLen -- * I\/O , Handle , stdin , stdout , stderr -- * Non-standard -- ** List-like classes , map , concat , concatMap , length , null , pack , unpack , repack , toList , Traversable.mapM , mapM_ , Traversable.forM , forM_ , any , all , foldl' , foldr , foldM --, split , readMay , intercalate , zip, zip3, zip4, zip5, zip6, zip7 , unzip, unzip3, unzip4, unzip5, unzip6, unzip7 , zipWith, zipWith3, zipWith4, zipWith5, zipWith6, zipWith7 , hashNub , ordNub , ordNubBy , sortWith , compareLength , sum , product , Prelude.repeat -- ** Set-like , (\\) , intersect , unions -- FIXME , mapSet -- ** Text-like , Show (..) , tshow , tlshow -- *** Case conversion , charToLower , charToUpper -- ** IO , IOData (..) , print , hClose -- ** FilePath , fpToString , fpFromString , fpToText , fpFromText , fpToTextWarn , fpToTextEx -- ** Exceptions , module Control.Exception.Enclosed -- ** Force types -- | Helper functions for situations where type inferer gets confused. , asByteString , asLByteString , asHashMap , asHashSet , asText , asLText , asList , asMap , asMaybe , asSet , asVector , asUVector ) where import qualified Prelude import Control.Exception (assert) import Control.Exception.Enclosed import Control.Monad (when, unless, void, liftM, ap, forever, join, sequence, sequence_, replicateM_) import Control.Concurrent.MVar.Lifted import Data.IORef.Lifted import qualified Data.Monoid as Monoid import qualified Data.Traversable as Traversable import Data.Traversable (Traversable) import Data.Foldable (Foldable) import Data.IOData (IOData (..)) import Data.Vector.Instances () import CorePrelude hiding (print, undefined, (<>)) import Data.ChunkedZip import qualified Data.Char as Char import Data.Sequences import Data.MonoTraversable import Data.Containers import Data.Builder import Data.MinLen import qualified Filesystem.Path.CurrentOS as F import System.IO (Handle, stdin, stdout, stderr, hClose) import Debug.Trace (trace, traceShow) import Data.Semigroup (Semigroup (..), WrappedMonoid (..)) import Prelude (Show (..)) import Data.Time ( UTCTime (..) , Day (..) , toGregorian , fromGregorian , formatTime , parseTime , getCurrentTime ) import System.Locale (defaultTimeLocale) import qualified Data.Set as Set import qualified Data.Map as Map import qualified Data.HashSet as HashSet import Data.Textual.Encoding import Data.Sequences.Lazy import GHC.Generics (Generic) tshow :: Show a => a -> Text tshow = fromList . Prelude.show tlshow :: Show a => a -> LText tlshow = fromList . Prelude.show -- | Convert a character to lower case. -- -- Character-based case conversion is lossy in comparison to string-based 'Data.MonoTraversable.toLower'. -- For instance, \'İ\' will be converted to \'i\', instead of \"i̇\". charToLower :: Char -> Char charToLower = Char.toLower -- | Convert a character to upper case. -- -- Character-based case conversion is lossy in comparison to string-based 'Data.MonoTraversable.toUpper'. -- For instance, \'ß\' won't be converted to \"SS\". charToUpper :: Char -> Char charToUpper = Char.toUpper -- Renames from mono-traversable pack :: IsSequence c => [Element c] -> c pack = fromList unpack, toList :: MonoFoldable c => c -> [Element c] unpack = otoList toList = otoList null :: MonoFoldable c => c -> Bool null = onull compareLength :: (Integral i, MonoFoldable c) => c -> i -> Ordering compareLength = ocompareLength sum :: (MonoFoldable c, Num (Element c)) => c -> Element c sum = osum product :: (MonoFoldable c, Num (Element c)) => c -> Element c product = oproduct all :: MonoFoldable c => (Element c -> Bool) -> c -> Bool all = oall any :: MonoFoldable c => (Element c -> Bool) -> c -> Bool any = oany length :: MonoFoldable c => c -> Int length = olength mapM_ :: (Monad m, MonoFoldable c) => (Element c -> m a) -> c -> m () mapM_ = omapM_ forM_ :: (Monad m, MonoFoldable c) => c -> (Element c -> m a) -> m () forM_ = oforM_ concatMap :: (Monoid m, MonoFoldable c) => (Element c -> m) -> c -> m concatMap = ofoldMap foldr :: MonoFoldable c => (Element c -> b -> b) -> b -> c -> b foldr = ofoldr foldl' :: MonoFoldable c => (a -> Element c -> a) -> a -> c -> a foldl' = ofoldl' foldM :: (Monad m, MonoFoldable c) => (a -> Element c -> m a) -> a -> c -> m a foldM = ofoldlM concat :: (MonoFoldable c, Monoid (Element c)) => c -> Element c concat = ofoldMap id readMay :: (Element c ~ Char, MonoFoldable c, Read a) => c -> Maybe a readMay a = -- FIXME replace with safe-failure stuff case [x | (x, t) <- Prelude.reads (otoList a :: String), onull t] of [x] -> Just x _ -> Nothing -- | Repack from one type to another, dropping to a list in the middle. -- -- @repack = pack . unpack@. repack :: (MonoFoldable a, IsSequence b, Element a ~ Element b) => a -> b repack = fromList . toList map :: Functor f => (a -> b) -> f a -> f b map = fmap infixr 5 ++ (++) :: Monoid m => m -> m -> m (++) = mappend {-# INLINE (++) #-} infixl 9 \\{-This comment teaches CPP correct behaviour -} -- | An alias for `difference`. (\\) :: SetContainer a => a -> a -> a (\\) = difference {-# INLINE (\\) #-} -- | An alias for `intersection`. intersect :: SetContainer a => a -> a -> a intersect = intersection {-# INLINE intersect #-} unions :: (MonoFoldable c, SetContainer (Element c)) => c -> Element c unions = ofoldl' union Monoid.mempty intercalate :: (Monoid (Element c), IsSequence c) => Element c -> c -> Element c intercalate xs xss = concat (intersperse xs xss) asByteString :: ByteString -> ByteString asByteString = id asLByteString :: LByteString -> LByteString asLByteString = id asHashMap :: HashMap k v -> HashMap k v asHashMap = id asHashSet :: HashSet a -> HashSet a asHashSet = id asText :: Text -> Text asText = id asLText :: LText -> LText asLText = id asList :: [a] -> [a] asList = id asMap :: Map k v -> Map k v asMap = id asMaybe :: Maybe a -> Maybe a asMaybe = id asSet :: Set a -> Set a asSet = id asVector :: Vector a -> Vector a asVector = id asUVector :: UVector a -> UVector a asUVector = id print :: (Show a, MonadIO m) => a -> m () print = liftIO . Prelude.print -- | Sort elements using the user supplied function to project something out of -- each element. -- Inspired by . sortWith :: (Ord a, IsSequence c) => (Element c -> a) -> c -> c sortWith f = sortBy $ comparing f -- | We define our own @undefined@ which is marked as deprecated. This makes it -- useful to use during development, but let's you more easily getting -- notification if you accidentally ship partial code in production. -- -- The classy prelude recommendation for when you need to really have a partial -- function in production is to use @error@ with a very descriptive message so -- that, in case an exception is thrown, you get more information than -- @Prelude.undefined@. -- -- Since 0.5.5 undefined :: a undefined = error "ClassyPrelude.undefined" {-# DEPRECATED undefined "It is highly recommended that you either avoid partial functions or provide meaningful error messages" #-} -- | -- -- Since 0.5.9 traceId :: String -> String traceId a = trace a a -- | -- -- Since 0.5.9 traceM :: (Monad m) => String -> m () traceM string = trace string $ return () -- | -- -- Since 0.5.9 traceShowId :: (Show a) => a -> a traceShowId a = trace (show a) a -- | -- -- Since 0.5.9 traceShowM :: (Show a, Monad m) => a -> m () traceShowM = traceM . show fpToString :: FilePath -> String fpToString = F.encodeString fpFromString :: String -> FilePath fpFromString = F.decodeString -- | Translates a FilePath to a Text -- Warns if there are non-unicode -- sequences in the file name fpToTextWarn :: MonadIO m => FilePath -> m Text fpToTextWarn fp = case F.toText fp of Right ok -> return ok Left bad -> do putStrLn $ pack $ "non-unicode filepath: " ++ F.encodeString fp return bad -- | Translates a FilePath to a Text -- Throws an exception if there are non-unicode -- sequences in the file name -- -- Use this to assert that you know -- a filename will translate properly into a Text -- If you created the filename, this should be the case. fpToTextEx :: FilePath -> Text fpToTextEx fp = either (const $ error errorMsg) id $ F.toText fp where errorMsg = "fpToTextEx: non-unicode filepath: " ++ F.encodeString fp -- | Translates a FilePath to a Text -- This translation is not correct for a (unix) filename -- which can contain arbitrary (non-unicode) bytes: those bytes will be discarded -- -- This means you cannot translate the Text back to the original file name. -- -- If you control or otherwise understand the filenames -- and believe them to be unicode valid consider using 'fpToTextEx' or 'fpToTextWarn' fpToText :: FilePath -> Text fpToText = either id id . F.toText fpFromText :: Text -> FilePath fpFromText = F.fromText -- Below is a lot of coding for classy-prelude! -- These functions are restricted to lists right now. -- Should eventually exist in mono-foldable and be extended to MonoFoldable -- when doing that should re-run the haskell-ordnub benchmarks -- | same behavior as nub, but requires Hashable & Eq and is O(n log n) -- https://github.com/nh2/haskell-ordnub hashNub :: (Hashable a, Eq a) => [a] -> [a] hashNub = go HashSet.empty where go _ [] = [] go s (x:xs) | x `HashSet.member` s = go s xs | otherwise = x : go (HashSet.insert x s) xs -- | same behavior as nub, but requires Ord and is O(n log n) -- https://github.com/nh2/haskell-ordnub ordNub :: (Ord a) => [a] -> [a] ordNub = go Set.empty where go _ [] = [] go s (x:xs) | x `Set.member` s = go s xs | otherwise = x : go (Set.insert x s) xs -- | same behavior as nubBy, but requires Ord and is O(n log n) -- https://github.com/nh2/haskell-ordnub ordNubBy :: (Ord b) => (a -> b) -> (a -> a -> Bool) -> [a] -> [a] ordNubBy p f = go Map.empty -- When removing duplicates, the first function assigns the input to a bucket, -- the second function checks whether it is already in the bucket (linear search). where go _ [] = [] go m (x:xs) = let b = p x in case b `Map.lookup` m of Nothing -> x : go (Map.insert b [x] m) xs Just bucket | elem_by f x bucket -> go m xs | otherwise -> x : go (Map.insert b (x:bucket) m) xs -- From the Data.List source code. elem_by :: (a -> a -> Bool) -> a -> [a] -> Bool elem_by _ _ [] = False elem_by eq y (x:xs) = y `eq` x || elem_by eq y xs