{-# LANGUAGE BangPatterns, CPP #-} #if __GLASGOW_HASKELL__ >= 702 {-# LANGUAGE Trustworthy #-} -- Imports internal modules #endif -- | -- Module : Data.Attoparsec.Combinator -- Copyright : Daan Leijen 1999-2001, Bryan O'Sullivan 2007-2015 -- License : BSD3 -- -- Maintainer : bos@serpentine.com -- Stability : experimental -- Portability : portable -- -- Useful parser combinators, similar to those provided by Parsec. module Data.Attoparsec.Combinator ( -- * Combinators try , (<?>) , choice , count , option , many' , many1 , many1' , manyTill , manyTill' , sepBy , sepBy' , sepBy1 , sepBy1' , skipMany , skipMany1 , eitherP , feed , satisfyElem , endOfInput , atEnd , lookAhead ) where #if !MIN_VERSION_base(4,8,0) import Control.Applicative (Applicative(..), (<$>)) import Data.Monoid (Monoid(mappend)) #endif import Control.Applicative (Alternative(..), empty, liftA2, many, (<|>)) import Control.Monad (MonadPlus(..)) import Data.Attoparsec.Internal.Types (Parser(..), IResult(..)) import Data.Attoparsec.Internal (endOfInput, atEnd, satisfyElem) import Data.ByteString (ByteString) import Data.Text (Text) import qualified Data.Attoparsec.Zepto as Z import Prelude hiding (succ) -- | Attempt a parse, and if it fails, rewind the input so that no -- input appears to have been consumed. -- -- This combinator is provided for compatibility with Parsec. -- attoparsec parsers always backtrack on failure. try :: Parser i a -> Parser i a try p = p {-# INLINE try #-} -- | Name the parser, in case failure occurs. (<?>) :: Parser i a -> String -- ^ the name to use if parsing fails -> Parser i a p <?> msg0 = Parser $ \t pos more lose succ -> let lose' t' pos' more' strs msg = lose t' pos' more' (msg0:strs) msg in runParser p t pos more lose' succ {-# INLINE (<?>) #-} infix 0 <?> -- | @choice ps@ tries to apply the actions in the list @ps@ in order, -- until one of them succeeds. Returns the value of the succeeding -- action. choice :: Alternative f => [f a] -> f a choice = foldr (<|>) empty {-# SPECIALIZE choice :: [Parser ByteString a] -> Parser ByteString a #-} {-# SPECIALIZE choice :: [Parser Text a] -> Parser Text a #-} {-# SPECIALIZE choice :: [Z.Parser a] -> Z.Parser a #-} -- | @option x p@ tries to apply action @p@. If @p@ fails without -- consuming input, it returns the value @x@, otherwise the value -- returned by @p@. -- -- > priority = option 0 (digitToInt <$> digit) option :: Alternative f => a -> f a -> f a option x p = p <|> pure x {-# SPECIALIZE option :: a -> Parser ByteString a -> Parser ByteString a #-} {-# SPECIALIZE option :: a -> Parser Text a -> Parser Text a #-} {-# SPECIALIZE option :: a -> Z.Parser a -> Z.Parser a #-} -- | A version of 'liftM2' that is strict in the result of its first -- action. liftM2' :: (Monad m) => (a -> b -> c) -> m a -> m b -> m c liftM2' f a b = do !x <- a y <- b return (f x y) {-# INLINE liftM2' #-} -- | @many' p@ applies the action @p@ /zero/ or more times. Returns a -- list of the returned values of @p@. The value returned by @p@ is -- forced to WHNF. -- -- > word = many' letter many' :: (MonadPlus m) => m a -> m [a] many' p = many_p where many_p = some_p `mplus` return [] some_p = liftM2' (:) p many_p {-# INLINE many' #-} -- | @many1 p@ applies the action @p@ /one/ or more times. Returns a -- list of the returned values of @p@. -- -- > word = many1 letter many1 :: Alternative f => f a -> f [a] many1 p = liftA2 (:) p (many p) {-# INLINE many1 #-} -- | @many1' p@ applies the action @p@ /one/ or more times. Returns a -- list of the returned values of @p@. The value returned by @p@ is -- forced to WHNF. -- -- > word = many1' letter many1' :: (MonadPlus m) => m a -> m [a] many1' p = liftM2' (:) p (many' p) {-# INLINE many1' #-} -- | @sepBy p sep@ applies /zero/ or more occurrences of @p@, separated -- by @sep@. Returns a list of the values returned by @p@. -- -- > commaSep p = p `sepBy` (char ',') sepBy :: Alternative f => f a -> f s -> f [a] sepBy p s = liftA2 (:) p ((s *> sepBy1 p s) <|> pure []) <|> pure [] {-# SPECIALIZE sepBy :: Parser ByteString a -> Parser ByteString s -> Parser ByteString [a] #-} {-# SPECIALIZE sepBy :: Parser Text a -> Parser Text s -> Parser Text [a] #-} {-# SPECIALIZE sepBy :: Z.Parser a -> Z.Parser s -> Z.Parser [a] #-} -- | @sepBy' p sep@ applies /zero/ or more occurrences of @p@, separated -- by @sep@. Returns a list of the values returned by @p@. The value -- returned by @p@ is forced to WHNF. -- -- > commaSep p = p `sepBy'` (char ',') sepBy' :: (MonadPlus m) => m a -> m s -> m [a] sepBy' p s = scan `mplus` return [] where scan = liftM2' (:) p ((s >> sepBy1' p s) `mplus` return []) {-# SPECIALIZE sepBy' :: Parser ByteString a -> Parser ByteString s -> Parser ByteString [a] #-} {-# SPECIALIZE sepBy' :: Parser Text a -> Parser Text s -> Parser Text [a] #-} {-# SPECIALIZE sepBy' :: Z.Parser a -> Z.Parser s -> Z.Parser [a] #-} -- | @sepBy1 p sep@ applies /one/ or more occurrences of @p@, separated -- by @sep@. Returns a list of the values returned by @p@. -- -- > commaSep p = p `sepBy1` (char ',') sepBy1 :: Alternative f => f a -> f s -> f [a] sepBy1 p s = scan where scan = liftA2 (:) p ((s *> scan) <|> pure []) {-# SPECIALIZE sepBy1 :: Parser ByteString a -> Parser ByteString s -> Parser ByteString [a] #-} {-# SPECIALIZE sepBy1 :: Parser Text a -> Parser Text s -> Parser Text [a] #-} {-# SPECIALIZE sepBy1 :: Z.Parser a -> Z.Parser s -> Z.Parser [a] #-} -- | @sepBy1' p sep@ applies /one/ or more occurrences of @p@, separated -- by @sep@. Returns a list of the values returned by @p@. The value -- returned by @p@ is forced to WHNF. -- -- > commaSep p = p `sepBy1'` (char ',') sepBy1' :: (MonadPlus m) => m a -> m s -> m [a] sepBy1' p s = scan where scan = liftM2' (:) p ((s >> scan) `mplus` return []) {-# SPECIALIZE sepBy1' :: Parser ByteString a -> Parser ByteString s -> Parser ByteString [a] #-} {-# SPECIALIZE sepBy1' :: Parser Text a -> Parser Text s -> Parser Text [a] #-} {-# SPECIALIZE sepBy1' :: Z.Parser a -> Z.Parser s -> Z.Parser [a] #-} -- | @manyTill p end@ applies action @p@ /zero/ or more times until -- action @end@ succeeds, and returns the list of values returned by -- @p@. This can be used to scan comments: -- -- > simpleComment = string "<!--" *> manyTill anyChar (string "-->") -- -- (Note the overlapping parsers @anyChar@ and @string \"-->\"@. -- While this will work, it is not very efficient, as it will cause a -- lot of backtracking.) manyTill :: Alternative f => f a -> f b -> f [a] manyTill p end = scan where scan = (end *> pure []) <|> liftA2 (:) p scan {-# SPECIALIZE manyTill :: Parser ByteString a -> Parser ByteString b -> Parser ByteString [a] #-} {-# SPECIALIZE manyTill :: Parser Text a -> Parser Text b -> Parser Text [a] #-} {-# SPECIALIZE manyTill :: Z.Parser a -> Z.Parser b -> Z.Parser [a] #-} -- | @manyTill' p end@ applies action @p@ /zero/ or more times until -- action @end@ succeeds, and returns the list of values returned by -- @p@. This can be used to scan comments: -- -- > simpleComment = string "<!--" *> manyTill' anyChar (string "-->") -- -- (Note the overlapping parsers @anyChar@ and @string \"-->\"@. -- While this will work, it is not very efficient, as it will cause a -- lot of backtracking.) -- -- The value returned by @p@ is forced to WHNF. manyTill' :: (MonadPlus m) => m a -> m b -> m [a] manyTill' p end = scan where scan = (end >> return []) `mplus` liftM2' (:) p scan {-# SPECIALIZE manyTill' :: Parser ByteString a -> Parser ByteString b -> Parser ByteString [a] #-} {-# SPECIALIZE manyTill' :: Parser Text a -> Parser Text b -> Parser Text [a] #-} {-# SPECIALIZE manyTill' :: Z.Parser a -> Z.Parser b -> Z.Parser [a] #-} -- | Skip zero or more instances of an action. skipMany :: Alternative f => f a -> f () skipMany p = scan where scan = (p *> scan) <|> pure () {-# SPECIALIZE skipMany :: Parser ByteString a -> Parser ByteString () #-} {-# SPECIALIZE skipMany :: Parser Text a -> Parser Text () #-} {-# SPECIALIZE skipMany :: Z.Parser a -> Z.Parser () #-} -- | Skip one or more instances of an action. skipMany1 :: Alternative f => f a -> f () skipMany1 p = p *> skipMany p {-# SPECIALIZE skipMany1 :: Parser ByteString a -> Parser ByteString () #-} {-# SPECIALIZE skipMany1 :: Parser Text a -> Parser Text () #-} {-# SPECIALIZE skipMany1 :: Z.Parser a -> Z.Parser () #-} -- | Apply the given action repeatedly, returning every result. count :: Monad m => Int -> m a -> m [a] count n p = sequence (replicate n p) {-# INLINE count #-} -- | Combine two alternatives. eitherP :: (Alternative f) => f a -> f b -> f (Either a b) eitherP a b = (Left <$> a) <|> (Right <$> b) {-# INLINE eitherP #-} -- | If a parser has returned a 'T.Partial' result, supply it with more -- input. feed :: Monoid i => IResult i r -> i -> IResult i r feed (Fail t ctxs msg) d = Fail (mappend t d) ctxs msg feed (Partial k) d = k d feed (Done t r) d = Done (mappend t d) r {-# INLINE feed #-} -- | Apply a parser without consuming any input. lookAhead :: Parser i a -> Parser i a lookAhead p = Parser $ \t pos more lose succ -> let succ' t' _pos' more' = succ t' pos more' in runParser p t pos more lose succ' {-# INLINE lookAhead #-}