X-Git-Url: https://git.immae.eu/?a=blobdiff_plain;f=Pipes%2FText.hs;h=58b9c26d2158fa78eb04871dba25866e62b7fc72;hb=79917d53aa8a1e2c8332e330337f74440859306d;hp=f85f0ee5ad35bb71fb667caefd107503b826f228;hpb=c8027236c897ebac20a4d9cc8209f216cded20c8;p=github%2Ffretlink%2Ftext-pipes.git diff --git a/Pipes/Text.hs b/Pipes/Text.hs index f85f0ee..58b9c26 100644 --- a/Pipes/Text.hs +++ b/Pipes/Text.hs @@ -1,76 +1,27 @@ -{-# LANGUAGE RankNTypes, TypeFamilies, BangPatterns, CPP #-} -#if __GLASGOW_HASKELL__ >= 702 -{-# LANGUAGE Trustworthy #-} -#endif -{-| This module provides @pipes@ utilities for \"text streams\", which are - streams of 'Text' chunks. The individual chunks are uniformly @strict@, but - a 'Producer' can be converted to and from lazy 'Text's, though this is generally - unwise. Where pipes IO replaces lazy IO, 'Producer Text m r' replaces lazy 'Text'. - An 'IO.Handle' can be associated with a 'Producer' or 'Consumer' according as it is read or written to. - - To stream to or from 'IO.Handle's, one can use 'fromHandle' or 'toHandle'. For - example, the following program copies a document from one file to another: +{-# LANGUAGE RankNTypes, TypeFamilies, BangPatterns, Trustworthy #-} -> import Pipes -> import qualified Data.Text.Pipes as Text -> import System.IO -> -> main = -> withFile "inFile.txt" ReadMode $ \hIn -> -> withFile "outFile.txt" WriteMode $ \hOut -> -> runEffect $ Text.fromHandle hIn >-> Text.toHandle hOut - -To stream from files, the following is perhaps more Prelude-like (note that it uses Pipes.Safe): - -> import Pipes -> import qualified Data.Text.Pipes as Text -> import Pipes.Safe -> -> main = runSafeT $ runEffect $ Text.readFile "inFile.txt" >-> Text.writeFile "outFile.txt" - - You can stream to and from 'stdin' and 'stdout' using the predefined 'stdin' - and 'stdout' pipes, as with the following \"echo\" program: - -> main = runEffect $ Text.stdin >-> Text.stdout - - You can also translate pure lazy 'TL.Text's to and from pipes: - -> main = runEffect $ Text.fromLazy (TL.pack "Hello, world!\n") >-> Text.stdout - - In addition, this module provides many functions equivalent to lazy - 'Text' functions so that you can transform or fold text streams. For - example, to stream only the first three lines of 'stdin' to 'stdout' you - might write: - -> import Pipes -> import qualified Pipes.Text as Text -> import qualified Pipes.Parse as Parse -> -> main = runEffect $ takeLines 3 Text.stdin >-> Text.stdout -> where -> takeLines n = Text.unlines . Parse.takeFree n . Text.lines - - The above program will never bring more than one chunk of text (~ 32 KB) into - memory, no matter how long the lines are. - - Note that functions in this library are designed to operate on streams that - are insensitive to text boundaries. This means that they may freely split - text into smaller texts, /discard empty texts/. However, apart from the - special case of 'concatMap', they will /never concatenate texts/ in order - to provide strict upper bounds on memory usage -- with the single exception of 'concatMap'. --} module Pipes.Text ( - -- * Producers - fromLazy - , stdin - , fromHandle - , readFile + -- * Effectful Text + -- $intro + + -- * Lenses + -- $lenses + + -- ** @view@ \/ @(^.)@ + -- $view - -- * Consumers - , stdout - , toHandle - , writeFile + -- ** @over@ \/ @(%~)@ + -- $over + + -- ** @zoom@ + -- $zoom + + -- * Special types: @Producer Text m (Producer Text m r)@ and @FreeT (Producer Text m) m r@ + -- $special + + -- * Producers + fromLazy -- * Pipes , map @@ -81,7 +32,6 @@ module Pipes.Text ( , dropWhile , filter , scan - , encodeUtf8 , pack , unpack , toCaseFold @@ -106,7 +56,6 @@ module Pipes.Text ( , count -- * Primitive Character Parsers - -- $parse , nextChar , drawChar , unDrawChar @@ -121,30 +70,13 @@ module Pipes.Text ( , group , word , line - - -- * Decoding Lenses - , decodeUtf8 - , codec - - -- * Codecs - , utf8 - , utf16_le - , utf16_be - , utf32_le - , utf32_be - - -- * Other Decoding/Encoding Functions - , decodeIso8859_1 - , decodeAscii - , encodeIso8859_1 - , encodeAscii -- * FreeT Splitters , chunksOf , splitsWith , splits --- , groupsBy --- , groups + , groupsBy + , groups , lines , words @@ -157,63 +89,37 @@ module Pipes.Text ( , unlines , unwords - -- * Re-exports + -- * Re-exports -- $reexports - , Decoding(..) - , streamDecodeUtf8 - , decodeSomeUtf8 - , Codec(..) - , TextException(..) , module Data.ByteString , module Data.Text , module Data.Profunctor - , module Data.Word , module Pipes.Parse , module Pipes.Group ) where -import Control.Exception (throwIO, try) import Control.Applicative ((<*)) -import Control.Monad (liftM, unless, join) +import Control.Monad (liftM, join) import Control.Monad.Trans.State.Strict (StateT(..), modify) -import Data.Monoid ((<>)) import qualified Data.Text as T -import qualified Data.Text.IO as T -import qualified Data.Text.Encoding as TE -import qualified Data.Text.Encoding.Error as TE import Data.Text (Text) import qualified Data.Text.Lazy as TL -import qualified Data.Text.Lazy.IO as TL -import Data.Text.Lazy.Internal (foldrChunks, defaultChunkSize) -import Data.ByteString.Unsafe (unsafeTake, unsafeDrop) import Data.ByteString (ByteString) -import qualified Data.ByteString as B -import qualified Data.ByteString.Char8 as B8 -import Data.Char (ord, isSpace) import Data.Functor.Constant (Constant(Constant, getConstant)) import Data.Functor.Identity (Identity) import Data.Profunctor (Profunctor) import qualified Data.Profunctor -import qualified Data.List as List -import Foreign.C.Error (Errno(Errno), ePIPE) -import qualified GHC.IO.Exception as G import Pipes -import qualified Pipes.ByteString as PB -import qualified Pipes.Text.Internal as PI -import Pipes.Text.Internal -import Pipes.Core (respond, Server') import Pipes.Group (concats, intercalates, FreeT(..), FreeF(..)) import qualified Pipes.Group as PG import qualified Pipes.Parse as PP import Pipes.Parse (Parser) -import qualified Pipes.Safe.Prelude as Safe -import qualified Pipes.Safe as Safe -import Pipes.Safe (MonadSafe(..), Base(..)) +import Pipes.Text.Encoding (Lens'_, Iso'_) import qualified Pipes.Prelude as P -import qualified System.IO as IO import Data.Char (isSpace) import Data.Word (Word8) - +import Foreign.Storable (sizeOf) +import Data.Bits (shiftL) import Prelude hiding ( all, any, @@ -243,87 +149,280 @@ import Prelude hiding ( words, writeFile ) --- | Convert a lazy 'TL.Text' into a 'Producer' of strict 'Text's -fromLazy :: (Monad m) => TL.Text -> Producer' Text m () -fromLazy = foldrChunks (\e a -> yield e >> a) (return ()) -{-# INLINE fromLazy #-} +{- $intro + This package provides @pipes@ utilities for /text streams/ or /character streams/, + realized as streams of 'Text' chunks. The individual chunks are uniformly /strict/, + and thus you will generally want @Data.Text@ in scope. But the type + @Producer Text m r@ ,as we are using it, is a sort of /pipes/ equivalent of the lazy @Text@ type. + + This particular module provides many functions equivalent in one way or another to + the pure functions in + . + They transform, divide, group and fold text streams. Though @Producer Text m r@ + is the type of \'effectful Text\', the functions in this module are \'pure\' + in the sense that they are uniformly monad-independent. + Simple /IO/ operations are defined in @Pipes.Text.IO@ -- as lazy IO @Text@ + operations are in @Data.Text.Lazy.IO@. Inter-operation with @ByteString@ + is provided in @Pipes.Text.Encoding@, which parallels @Data.Text.Lazy.Encoding@. + + The Text type exported by @Data.Text.Lazy@ is basically that of a lazy list of + strict Text: the implementation is arranged so that the individual strict 'Text' + chunks are kept to a reasonable size; the user is not aware of the divisions + between the connected 'Text' chunks. + So also here: the functions in this module are designed to operate on streams that + are insensitive to text boundaries. This means that they may freely split + text into smaller texts and /discard empty texts/. The objective, though, is + that they should /never concatenate texts/ in order to provide strict upper + bounds on memory usage. + + For example, to stream only the first three lines of 'stdin' to 'stdout' you + might write: --- | Stream text from 'stdin' -stdin :: MonadIO m => Producer Text m () -stdin = fromHandle IO.stdin -{-# INLINE stdin #-} +> import Pipes +> import qualified Pipes.Text as Text +> import qualified Pipes.Text.IO as Text +> import Pipes.Group (takes') +> import Lens.Family +> +> main = runEffect $ takeLines 3 Text.stdin >-> Text.stdout +> where +> takeLines n = Text.unlines . takes' n . view Text.lines + + The above program will never bring more than one chunk of text (~ 32 KB) into + memory, no matter how long the lines are. -{-| Convert a 'IO.Handle' into a text stream using a text size - determined by the good sense of the text library; note that this - is distinctly slower than @decideUtf8 (Pipes.ByteString.fromHandle h)@ - but uses the system encoding and has other `Data.Text.IO` features -} +{- $lenses + As this example shows, one superficial difference from @Data.Text.Lazy@ + is that many of the operations, like 'lines', are \'lensified\'; this has a + number of advantages (where it is possible); in particular it facilitates their + use with 'Parser's of Text (in the general + sense.) The disadvantage, famously, is that the messages you get for type errors can be + a little alarming. The remarks that follow in this section are for non-lens adepts. + + Each lens exported here, e.g. 'lines', 'chunksOf' or 'splitAt', reduces to the + intuitively corresponding function when used with @view@ or @(^.)@. Instead of + writing: + + > splitAt 17 producer + + as we would with the Prelude or Text functions, we write + + > view (splitAt 17) producer + + or equivalently + + > producer ^. splitAt 17 -fromHandle :: MonadIO m => IO.Handle -> Producer Text m () -fromHandle h = go where - go = do txt <- liftIO (T.hGetChunk h) - unless (T.null txt) ( do yield txt - go ) -{-# INLINABLE fromHandle#-} + This may seem a little indirect, but note that many equivalents of + @Text -> Text@ functions are exported here as 'Pipe's. Here too we recover the intuitively + corresponding functions by prefixing them with @(>->)@. Thus something like +> stripLines = Text.unlines . Group.maps (>-> Text.stripStart) . view Text.lines -{-| Stream text from a file in the simple fashion of @Data.Text.IO@ + would drop the leading white space from each line. ->>> runSafeT $ runEffect $ Text.readFile "hello.hs" >-> Text.map toUpper >-> hoist lift Text.stdout -MAIN = PUTSTRLN "HELLO WORLD" --} + The lenses in this library are marked as /improper/; this just means that + they don't admit all the operations of an ideal lens, but only /getting/ and /focusing/. + Just for this reason, though, the magnificent complexities of the lens libraries + are a distraction. The lens combinators to keep in mind, the ones that make sense for + our lenses, are @view@ \/ @(^.)@), @over@ \/ @(%~)@ , and @zoom@. -readFile :: MonadSafe m => FilePath -> Producer Text m () -readFile file = Safe.withFile file IO.ReadMode fromHandle -{-# INLINE readFile #-} + One need only keep in mind that if @l@ is a @Lens'_ a b@, then: +-} +{- $view + @view l@ is a function @a -> b@ . Thus @view l a@ (also written @a ^. l@ ) + is the corresponding @b@; as was said above, this function will be exactly the + function you think it is, given its name. Thus to uppercase the first n characters + of a Producer, leaving the rest the same, we could write: -{-| Stream text to 'stdout' - Unlike 'toHandle', 'stdout' gracefully terminates on a broken output pipe. + > upper n p = do p' <- p ^. Text.splitAt n >-> Text.toUpper + > p' +-} +{- $over + @over l@ is a function @(b -> b) -> a -> a@. Thus, given a function that modifies + @b@s, the lens lets us modify an @a@ by applying @f :: b -> b@ to + the @b@ that we can \"see\" through the lens. So @over l f :: a -> a@ + (it can also be written @l %~ f@). + For any particular @a@, then, @over l f a@ or @(l %~ f) a@ is a revised @a@. + So above we might have written things like these: + + > stripLines = Text.lines %~ maps (>-> Text.stripStart) + > stripLines = over Text.lines (maps (>-> Text.stripStart)) + > upper n = Text.splitAt n %~ (>-> Text.toUpper) - Note: For best performance, it might be best just to use @(for source (liftIO . putStr))@ - instead of @(source >-> stdout)@ . -} -stdout :: MonadIO m => Consumer' Text m () -stdout = go - where - go = do - txt <- await - x <- liftIO $ try (T.putStr txt) - case x of - Left (G.IOError { G.ioe_type = G.ResourceVanished - , G.ioe_errno = Just ioe }) - | Errno ioe == ePIPE - -> return () - Left e -> liftIO (throwIO e) - Right () -> go -{-# INLINABLE stdout #-} +{- $zoom + @zoom l@, finally, is a function from a @Parser b m r@ + to a @Parser a m r@ (or more generally a @StateT (Producer b m x) m r@). + Its use is easiest to see with an decoding lens like 'utf8', which + \"sees\" a Text producer hidden inside a ByteString producer: + @drawChar@ is a Text parser, returning a @Maybe Char@, @zoom utf8 drawChar@ is + a /ByteString/ parser, returning a @Maybe Char@. @drawAll@ is a Parser that returns + a list of everything produced from a Producer, leaving only the return value; it would + usually be unreasonable to use it. But @zoom (splitAt 17) drawAll@ + returns a list of Text chunks containing the first seventeen Chars, and returns the rest of + the Text Producer for further parsing. Suppose that we want, inexplicably, to + modify the casing of a Text Producer according to any instruction it might + contain at the start. Then we might write something like this: + +> obey :: Monad m => Producer Text m b -> Producer Text m b +> obey p = do (ts, p') <- lift $ runStateT (zoom (Text.splitAt 7) drawAll) p +> let seven = T.concat ts +> case T.toUpper seven of +> "TOUPPER" -> p' >-> Text.toUpper +> "TOLOWER" -> p' >-> Text.toLower +> _ -> do yield seven +> p' + + +> >>> let doc = each ["toU","pperTh","is document.\n"] +> >>> runEffect $ obey doc >-> Text.stdout +> THIS DOCUMENT. + + The purpose of exporting lenses is the mental economy achieved with this three-way + applicability. That one expression, e.g. @lines@ or @splitAt 17@ can have these + three uses is no more surprising than that a pipe can act as a function modifying + the output of a producer, namely by using @>->@ to its left: @producer >-> pipe@ + -- but can /also/ modify the inputs to a consumer by using @>->@ to its right: + @pipe >-> consumer@ + + The three functions, @view@ \/ @(^.)@, @over@ \/ @(%~)@ and @zoom@ are supplied by + both and + The use of 'zoom' is explained + in + and to some extent in the @Pipes.Text.Encoding@ module here. +-} +{- $special + These simple 'lines' examples reveal a more important difference from @Data.Text.Lazy@ . + This is in the types that are most closely associated with our central text type, + @Producer Text m r@. In @Data.Text@ and @Data.Text.Lazy@ we find functions like -{-| Convert a text stream into a 'Handle' +> splitAt :: Int -> Text -> (Text, Text) +> lines :: Text -> [Text] +> chunksOf :: Int -> Text -> [Text] - Note: again, for best performance, where possible use - @(for source (liftIO . hPutStr handle))@ instead of @(source >-> toHandle handle)@. --} -toHandle :: MonadIO m => IO.Handle -> Consumer' Text m r -toHandle h = for cat (liftIO . T.hPutStr h) -{-# INLINABLE toHandle #-} + which relate a Text with a pair of Texts or a list of Texts. + The corresponding functions here (taking account of \'lensification\') are -{-# RULES "p >-> toHandle h" forall p h . - p >-> toHandle h = for p (\txt -> liftIO (T.hPutStr h txt)) - #-} +> view . splitAt :: (Monad m, Integral n) => n -> Producer Text m r -> Producer Text m (Producer Text m r) +> view lines :: Monad m => Producer Text m r -> FreeT (Producer Text m) m r +> view . chunksOf :: (Monad m, Integral n) => n -> Producer Text m r -> FreeT (Producer Text m) m r + + Some of the types may be more readable if you imagine that we have introduced + our own type synonyms + +> type Text m r = Producer T.Text m r +> type Texts m r = FreeT (Producer T.Text m) m r + + Then we would think of the types above as + +> view . splitAt :: (Monad m, Integral n) => n -> Text m r -> Text m (Text m r) +> view lines :: (Monad m) => Text m r -> Texts m r +> view . chunksOf :: (Monad m, Integral n) => n -> Text m r -> Texts m r + + which brings one closer to the types of the similar functions in @Data.Text.Lazy@ + + In the type @Producer Text m (Producer Text m r)@ the second + element of the \'pair\' of effectful Texts cannot simply be retrieved + with something like 'snd'. This is an \'effectful\' pair, and one must work + through the effects of the first element to arrive at the second Text stream, even + if you are proposing to throw the Text in the first element away. + Note that we use Control.Monad.join to fuse the pair back together, since it specializes to + +> join :: Monad m => Producer Text m (Producer m r) -> Producer m r + + The return type of 'lines', 'words', 'chunksOf' and the other /splitter/ functions, + @FreeT (Producer m Text) m r@ -- our @Texts m r@ -- is the type of (effectful) + lists of (effectful) texts. The type @([Text],r)@ might be seen to gather + together things of the forms: + +> r +> (Text,r) +> (Text, (Text, r)) +> (Text, (Text, (Text, r))) +> (Text, (Text, (Text, (Text, r)))) +> ... + + (We might also have identified the sum of those types with @Free ((,) Text) r@ + -- or, more absurdly, @FreeT ((,) Text) Identity r@.) + + Similarly, our type @Texts m r@, or @FreeT (Text m) m r@ -- in fact called + @FreeT (Producer Text m) m r@ here -- encompasses all the members of the sequence: + +> m r +> Text m r +> Text m (Text m r) +> Text m (Text m (Text m r)) +> Text m (Text m (Text m (Text m r))) +> ... + + We might have used a more specialized type in place of @FreeT (Producer a m) m r@, + or indeed of @FreeT (Producer Text m) m r@, but it is clear that the correct + result type of 'lines' will be isomorphic to @FreeT (Producer Text m) m r@ . + + One might think that + +> lines :: Monad m => Lens'_ (Producer Text m r) (FreeT (Producer Text m) m r) +> view . lines :: Monad m => Producer Text m r -> FreeT (Producer Text m) m r + + should really have the type + +> lines :: Monad m => Pipe Text Text m r + + as e.g. 'toUpper' does. But this would spoil the control we are + attempting to maintain over the size of chunks. It is in fact just + as unreasonable to want such a pipe as to want + +> Data.Text.Lazy.lines :: Text -> Text + to 'rechunk' the strict Text chunks inside the lazy Text to respect + line boundaries. In fact we have --- | Stream text into a file. Uses @pipes-safe@. -writeFile :: (MonadSafe m) => FilePath -> Consumer' Text m () -writeFile file = Safe.withFile file IO.WriteMode toHandle -{-# INLINE writeFile #-} +> Data.Text.Lazy.lines :: Text -> [Text] +> Prelude.lines :: String -> [String] + where the elements of the list are themselves lazy Texts or Strings; the use + of @FreeT (Producer Text m) m r@ is simply the 'effectful' version of this. + + The @Pipes.Group@ module, which can generally be imported without qualification, + provides many functions for working with things of type @FreeT (Producer a m) m r@. + In particular it conveniently exports the constructors for @FreeT@ and the associated + @FreeF@ type -- a fancy form of @Either@, namely + +> data FreeF f a b = Pure a | Free (f b) + + for pattern-matching. Consider the implementation of the 'words' function, or + of the part of the lens that takes us to the words; it is compact but exhibits many + of the points under discussion, including explicit handling of the @FreeT@ and @FreeF@ + constuctors. Keep in mind that + +> newtype FreeT f m a = FreeT (m (FreeF f a (FreeT f m a))) +> next :: Monad m => Producer a m r -> m (Either r (a, Producer a m r)) + + Thus the @do@ block after the @FreeT@ constructor is in the base monad, e.g. 'IO' or 'Identity'; + the later subordinate block, opened by the @Free@ constructor, is in the @Producer@ monad: + +> words :: Monad m => Producer Text m r -> FreeT (Producer Text m) m r +> words p = FreeT $ do -- With 'next' we will inspect p's first chunk, excluding spaces; +> x <- next (p >-> dropWhile isSpace) -- note that 'dropWhile isSpace' is a pipe, and is thus *applied* with '>->'. +> return $ case x of -- We use 'return' and so need something of type 'FreeF (Text m) r (Texts m r)' +> Left r -> Pure r -- 'Left' means we got no Text chunk, but only the return value; so we are done. +> Right (txt, p') -> Free $ do -- If we get a chunk and the rest of the producer, p', we enter the 'Producer' monad +> p'' <- view (break isSpace) -- When we apply 'break isSpace', we get a Producer that returns a Producer; +> (yield txt >> p') -- so here we yield everything up to the next space, and get the rest back. +> return (words p'') -- We then carry on with the rest, which is likely to begin with space. + +-} -type Lens' a b = forall f . Functor f => (b -> f b) -> (a -> f a) +-- | Convert a lazy 'TL.Text' into a 'Producer' of strict 'Text's +fromLazy :: (Monad m) => TL.Text -> Producer' Text m () +fromLazy = TL.foldrChunks (\e a -> yield e >> a) (return ()) +{-# INLINE fromLazy #-} -type Iso' a b = forall f p . (Functor f, Profunctor p) => p b (f b) -> p a (f a) (^.) :: a -> ((b -> Constant b b) -> (a -> Constant b a)) -> b a ^. lens = getConstant (lens Constant a) @@ -348,16 +447,6 @@ concatMap f = P.map (T.concatMap f) p >-> concatMap f = for p (\txt -> yield (T.concatMap f txt)) #-} --- | Transform a Pipe of 'Text' into a Pipe of 'ByteString's using UTF-8 --- encoding; @encodeUtf8 = Pipes.Prelude.map TE.encodeUtf8@ so more complex --- encoding pipes can easily be constructed with the functions in @Data.Text.Encoding@ -encodeUtf8 :: Monad m => Pipe Text ByteString m r -encodeUtf8 = P.map TE.encodeUtf8 -{-# INLINEABLE encodeUtf8 #-} - -{-# RULES "p >-> encodeUtf8" forall p . - p >-> encodeUtf8 = for p (\txt -> yield (TE.encodeUtf8 txt)) - #-} -- | Transform a Pipe of 'String's into one of 'Text' chunks pack :: Monad m => Pipe String Text m r @@ -379,7 +468,7 @@ unpack = for cat (\t -> yield (T.unpack t)) -- | @toCaseFold@, @toLower@, @toUpper@ and @stripStart@ are standard 'Text' utilities, -- here acting as 'Text' pipes, rather as they would on a lazy text -toCaseFold :: Monad m => Pipe Text Text m () +toCaseFold :: Monad m => Pipe Text Text m r toCaseFold = P.map T.toCaseFold {-# INLINEABLE toCaseFold #-} @@ -389,7 +478,7 @@ toCaseFold = P.map T.toCaseFold -- | lowercase incoming 'Text' -toLower :: Monad m => Pipe Text Text m () +toLower :: Monad m => Pipe Text Text m r toLower = P.map T.toLower {-# INLINEABLE toLower #-} @@ -398,7 +487,7 @@ toLower = P.map T.toLower #-} -- | uppercase incoming 'Text' -toUpper :: Monad m => Pipe Text Text m () +toUpper :: Monad m => Pipe Text Text m r toUpper = P.map T.toUpper {-# INLINEABLE toUpper #-} @@ -487,13 +576,15 @@ filter predicate = P.map (T.filter predicate) scan :: (Monad m) => (Char -> Char -> Char) -> Char -> Pipe Text Text m r -scan step begin = go begin +scan step begin = do + yield (T.singleton begin) + go begin where go c = do txt <- await let txt' = T.scanl step c txt c' = T.last txt' - yield txt' + yield (T.tail txt') go c' {-# INLINABLE scan #-} @@ -591,7 +682,6 @@ minimum = P.fold step Nothing id Just c -> Just (min c (T.minimum txt)) {-# INLINABLE minimum #-} - -- | Find the first element in the stream that matches the predicate find :: (Monad m) @@ -613,12 +703,12 @@ count c p = P.fold (+) 0 id (p >-> P.map (fromIntegral . T.count c)) {-# INLINABLE count #-} -{-| Consume the first character from a stream of 'Text' +-- | Consume the first character from a stream of 'Text' +-- +-- 'next' either fails with a 'Left' if the 'Producer' has no more characters or +-- succeeds with a 'Right' providing the next character and the remainder of the +-- 'Producer'. - 'next' either fails with a 'Left' if the 'Producer' has no more characters or - succeeds with a 'Right' providing the next character and the remainder of the - 'Producer'. --} nextChar :: (Monad m) => Producer Text m r @@ -634,9 +724,8 @@ nextChar = go Just (c, txt') -> return (Right (c, yield txt' >> p')) {-# INLINABLE nextChar #-} -{-| Draw one 'Char' from a stream of 'Text', returning 'Left' if the - 'Producer' is empty --} +-- | Draw one 'Char' from a stream of 'Text', returning 'Left' if the 'Producer' is empty + drawChar :: (Monad m) => Parser Text m (Maybe Char) drawChar = do x <- PP.draw @@ -663,7 +752,9 @@ unDrawChar c = modify (yield (T.singleton c) >>) > Left _ -> return () > Right c -> unDrawChar c > return x + -} + peekChar :: (Monad m) => Parser Text m (Maybe Char) peekChar = do x <- drawChar @@ -690,35 +781,11 @@ isEndOfChars = do {-# INLINABLE isEndOfChars #-} -{- | An improper lens into a stream of 'ByteString' expected to be UTF-8 encoded; the associated - stream of Text ends by returning a stream of ByteStrings beginning at the point of failure. - -} - -decodeUtf8 :: Monad m => Lens' (Producer ByteString m r) - (Producer Text m (Producer ByteString m r)) -decodeUtf8 k p0 = fmap (\p -> join (for p (yield . TE.encodeUtf8))) - (k (go B.empty PI.streamDecodeUtf8 p0)) where - go !carry dec0 p = do - x <- lift (next p) - case x of Left r -> return (if B.null carry - then return r -- all bytestring input was consumed - else (do yield carry -- a potentially valid fragment remains - return r)) - - Right (chunk, p') -> case dec0 chunk of - PI.Some text carry2 dec -> do yield text - go carry2 dec p' - PI.Other text bs -> do yield text - return (do yield bs -- an invalid blob remains - p') -{-# INLINABLE decodeUtf8 #-} - - -- | Splits a 'Producer' after the given number of characters splitAt :: (Monad m, Integral n) => n - -> Lens' (Producer Text m r) + -> Lens'_ (Producer Text m r) (Producer Text m (Producer Text m r)) splitAt n0 k p0 = fmap join (k (go n0 p0)) where @@ -740,13 +807,14 @@ splitAt n0 k p0 = fmap join (k (go n0 p0)) {-# INLINABLE splitAt #-} -{-| Split a text stream in two, where the first text stream is the longest - consecutive group of text that satisfy the predicate --} +-- | Split a text stream in two, producing the longest +-- consecutive group of characters that satisfies the predicate +-- and returning the rest + span :: (Monad m) => (Char -> Bool) - -> Lens' (Producer Text m r) + -> Lens'_ (Producer Text m r) (Producer Text m (Producer Text m r)) span predicate k p0 = fmap join (k (go p0)) where @@ -765,13 +833,13 @@ span predicate k p0 = fmap join (k (go p0)) return (yield suffix >> p') {-# INLINABLE span #-} -{-| Split a text stream in two, where the first text stream is the longest +{-| Split a text stream in two, producing the longest consecutive group of characters that don't satisfy the predicate -} break :: (Monad m) => (Char -> Bool) - -> Lens' (Producer Text m r) + -> Lens'_ (Producer Text m r) (Producer Text m (Producer Text m r)) break predicate = span (not . predicate) {-# INLINABLE break #-} @@ -782,7 +850,7 @@ break predicate = span (not . predicate) groupBy :: (Monad m) => (Char -> Char -> Bool) - -> Lens' (Producer Text m r) + -> Lens'_ (Producer Text m r) (Producer Text m (Producer Text m r)) groupBy equals k p0 = fmap join (k ((go p0))) where go p = do @@ -796,7 +864,7 @@ groupBy equals k p0 = fmap join (k ((go p0))) where -- | Improper lens that splits after the first succession of identical 'Char' s group :: Monad m - => Lens' (Producer Text m r) + => Lens'_ (Producer Text m r) (Producer Text m (Producer Text m r)) group = groupBy (==) {-# INLINABLE group #-} @@ -806,7 +874,7 @@ group = groupBy (==) Unlike 'words', this does not drop leading whitespace -} word :: (Monad m) - => Lens' (Producer Text m r) + => Lens'_ (Producer Text m r) (Producer Text m (Producer Text m r)) word k p0 = fmap join (k (to p0)) where @@ -817,7 +885,7 @@ word k p0 = fmap join (k (to p0)) line :: (Monad m) - => Lens' (Producer Text m r) + => Lens'_ (Producer Text m r) (Producer Text m (Producer Text m r)) line = break (== '\n') @@ -849,7 +917,7 @@ intersperse c = go0 -- | Improper isomorphism between a 'Producer' of 'ByteString's and 'Word8's -packChars :: Monad m => Iso' (Producer Char m x) (Producer Text m x) +packChars :: Monad m => Iso'_ (Producer Char m x) (Producer Text m x) packChars = Data.Profunctor.dimap to (fmap from) where -- to :: Monad m => Producer Char m x -> Producer Text m x @@ -861,13 +929,16 @@ packChars = Data.Profunctor.dimap to (fmap from) -- from :: Monad m => Producer Text m x -> Producer Char m x from p = for p (each . T.unpack) + {-# INLINABLE packChars #-} +defaultChunkSize :: Int +defaultChunkSize = 16384 - (sizeOf (undefined :: Int) `shiftL` 1) -- | Split a text stream into 'FreeT'-delimited text streams of fixed size chunksOf :: (Monad m, Integral n) - => n -> Lens' (Producer Text m r) + => n -> Lens'_ (Producer Text m r) (FreeT (Producer Text m) m r) chunksOf n k p0 = fmap concats (k (FreeT (go p0))) where @@ -913,7 +984,7 @@ splitsWith predicate p0 = FreeT (go0 p0) -- | Split a text stream using the given 'Char' as the delimiter splits :: (Monad m) => Char - -> Lens' (Producer Text m r) + -> Lens'_ (Producer Text m r) (FreeT (Producer Text m) m r) splits c k p = fmap (PG.intercalates (yield (T.singleton c))) (k (splitsWith (c ==) p)) @@ -925,7 +996,7 @@ splits c k p = groupsBy :: Monad m => (Char -> Char -> Bool) - -> Lens' (Producer Text m x) (FreeT (Producer Text m) m x) + -> Lens'_ (Producer Text m x) (FreeT (Producer Text m) m x) groupsBy equals k p0 = fmap concats (k (FreeT (go p0))) where go p = do x <- next p case x of Left r -> return (Pure r) @@ -940,7 +1011,7 @@ groupsBy equals k p0 = fmap concats (k (FreeT (go p0))) where -- | Like 'groupsBy', where the equality predicate is ('==') groups :: Monad m - => Lens' (Producer Text m x) (FreeT (Producer Text m) m x) + => Lens'_ (Producer Text m x) (FreeT (Producer Text m) m x) groups = groupsBy (==) {-# INLINABLE groups #-} @@ -949,7 +1020,7 @@ groups = groupsBy (==) {-| Split a text stream into 'FreeT'-delimited lines -} lines - :: (Monad m) => Iso' (Producer Text m r) (FreeT (Producer Text m) m r) + :: (Monad m) => Iso'_ (Producer Text m r) (FreeT (Producer Text m) m r) lines = Data.Profunctor.dimap _lines (fmap _unlines) where _lines p0 = FreeT (go0 p0) @@ -980,7 +1051,7 @@ lines = Data.Profunctor.dimap _lines (fmap _unlines) -- | Split a text stream into 'FreeT'-delimited words words - :: (Monad m) => Iso' (Producer Text m r) (FreeT (Producer Text m) m r) + :: (Monad m) => Iso'_ (Producer Text m r) (FreeT (Producer Text m) m r) words = Data.Profunctor.dimap go (fmap _unwords) where go p = FreeT $ do @@ -1045,10 +1116,6 @@ unwords unwords = intercalate (yield $ T.singleton ' ') {-# INLINABLE unwords #-} -{- $parse - The following parsing utilities are single-character analogs of the ones found - @pipes-parse@. --} {- $reexports @@ -1057,106 +1124,4 @@ unwords = intercalate (yield $ T.singleton ' ') @Pipes.Parse@ re-exports 'input', 'concat', 'FreeT' (the type) and the 'Parse' synonym. -} -{- | Use a 'Codec' as a pipes-style 'Lens' into a byte stream; the available 'Codec' s are - 'utf8', 'utf16_le', 'utf16_be', 'utf32_le', 'utf32_be' . The 'Codec' concept and the - individual 'Codec' definitions follow the enumerator and conduit libraries. - - Utf8 is handled differently in this library -- without the use of 'unsafePerformIO' &co - to catch 'Text' exceptions; but the same 'mypipe ^. codec utf8' interface can be used. - 'mypipe ^. decodeUtf8' should be the same, but has a somewhat more direct and thus perhaps - better implementation. - - -} -codec :: Monad m => Codec -> Lens' (Producer ByteString m r) (Producer Text m (Producer ByteString m r)) -codec (Codec _ enc dec) k p0 = fmap (\p -> join (for p (yield . fst . enc))) - (k (decoder (dec B.empty) p0) ) where - decoder :: Monad m => PI.Decoding -> Producer ByteString m r -> Producer Text m (Producer ByteString m r) - decoder !d p0 = case d of - PI.Other txt bad -> do yield txt - return (do yield bad - p0) - PI.Some txt extra dec -> do yield txt - x <- lift (next p0) - case x of Left r -> return (do yield extra - return r) - Right (chunk,p1) -> decoder (dec chunk) p1 - -{- | ascii and latin encodings only represent a small fragment of 'Text'; thus we cannot - use the pipes 'Lens' style to work with them. Rather we simply define functions - each way. - - 'encodeAscii' : Reduce as much of your stream of 'Text' actually is ascii to a byte stream, - returning the rest of the 'Text' at the first non-ascii 'Char' --} -encodeAscii :: Monad m => Producer Text m r -> Producer ByteString m (Producer Text m r) -encodeAscii = go where - go p = do echunk <- lift (next p) - case echunk of - Left r -> return (return r) - Right (chunk, p') -> - if T.null chunk - then go p' - else let (safe, unsafe) = T.span (\c -> ord c <= 0x7F) chunk - in do yield (B8.pack (T.unpack safe)) - if T.null unsafe - then go p' - else return $ do yield unsafe - p' -{- | Reduce as much of your stream of 'Text' actually is iso8859 or latin1 to a byte stream, - returning the rest of the 'Text' upon hitting any non-latin 'Char' - -} -encodeIso8859_1 :: Monad m => Producer Text m r -> Producer ByteString m (Producer Text m r) -encodeIso8859_1 = go where - go p = do etxt <- lift (next p) - case etxt of - Left r -> return (return r) - Right (txt, p') -> - if T.null txt - then go p' - else let (safe, unsafe) = T.span (\c -> ord c <= 0xFF) txt - in do yield (B8.pack (T.unpack safe)) - if T.null unsafe - then go p' - else return $ do yield unsafe - p' - -{- | Reduce a byte stream to a corresponding stream of ascii chars, returning the - unused 'ByteString' upon hitting an un-ascii byte. - -} -decodeAscii :: Monad m => Producer ByteString m r -> Producer Text m (Producer ByteString m r) -decodeAscii = go where - go p = do echunk <- lift (next p) - case echunk of - Left r -> return (return r) - Right (chunk, p') -> - if B.null chunk - then go p' - else let (safe, unsafe) = B.span (<= 0x7F) chunk - in do yield (T.pack (B8.unpack safe)) - if B.null unsafe - then go p' - else return $ do yield unsafe - p' - -{- | Reduce a byte stream to a corresponding stream of ascii chars, returning the - unused 'ByteString' upon hitting the rare un-latinizable byte. - -} -decodeIso8859_1 :: Monad m => Producer ByteString m r -> Producer Text m (Producer ByteString m r) -decodeIso8859_1 = go where - go p = do echunk <- lift (next p) - case echunk of - Left r -> return (return r) - Right (chunk, p') -> - if B.null chunk - then go p' - else let (safe, unsafe) = B.span (<= 0xFF) chunk - in do yield (T.pack (B8.unpack safe)) - if B.null unsafe - then go p' - else return $ do yield unsafe - p' - - - - - +