------------------------------------------------------------------------ -- | Parser combinators with support for left recursion, following -- Johnson\'s \"Memoization in Top-Down Parsing\". -- -- This implementation is based on an implementation due to Atkey -- (attached to an edlambda-members mailing list message from -- 2011-02-15 titled \'Slides for \"Introduction to Parser -- Combinators\"\'). -- -- Note that non-memoised left recursion is not guaranteed to work. -- -- The code contains an important deviation from Johnson\'s paper: the -- check for subsumed results is not included. This means that one can -- get the same result multiple times when parsing using ambiguous -- grammars. As an example, parsing the empty string using @S ∷= ε | -- ε@ succeeds twice. This change also means that parsing fails to -- terminate for some cyclic grammars that would otherwise be handled -- successfully, such as @S ∷= S | ε@. However, the library is not -- intended to handle infinitely ambiguous grammars. (It is unclear to -- the author of this module whether the change leads to more -- non-termination for grammars that are not cyclic.) {-# LANGUAGE CPP #-} module Agda.Utils.Parser.MemoisedCPS ( ParserClass(..) , sat, token, tok, doc , DocP, bindP, choiceP, seqP, starP, atomP , Parser , ParserWithGrammar ) where import Control.Applicative ( Alternative((<|>), empty, many, some) ) import Control.Monad (ap, liftM2) import Control.Monad.State.Strict (State, evalState, runState, get, put, modify') import Data.Array import Data.Hashable import qualified Data.HashMap.Strict as Map import Data.HashMap.Strict (HashMap) import qualified Data.HashSet as Set import Data.HashSet (HashSet) import qualified Data.IntMap.Strict as IntMap import Data.IntMap.Strict (IntMap) import qualified Data.List as List import Data.Maybe import Text.PrettyPrint.HughesPJ hiding (empty) import qualified Text.PrettyPrint.HughesPJ as PP import Agda.Utils.Pretty ( mparens ) #include "undefined.h" import Agda.Utils.Impossible -- | Positions. type Pos = Int -- | State monad used by the parser. type M k r tok b = State (IntMap (HashMap k (Value k r tok b))) -- | Continuations. type Cont k r tok b a = Pos -> a -> M k r tok b [b] -- | Memoised values. data Value k r tok b = Value { results :: !(IntMap [r]) , continuations :: [Cont k r tok b r] } -- | The parser type. -- -- The parameters of the type @Parser k r tok a@ have the following -- meanings: -- -- [@k@] Type used for memoisation keys. -- -- [@r@] The type of memoised values. (Yes, all memoised values have -- to have the same type.) -- -- [@tok@] The token type. -- -- [@a@] The result type. newtype Parser k r tok a = P { unP :: forall b. Array Pos tok -> Pos -> Cont k r tok b a -> M k r tok b [b] } instance Monad (Parser k r tok) where return = pure P p >>= f = P $ \input i k -> p input i $ \j x -> unP (f x) input j k instance Functor (Parser k r tok) where fmap f (P p) = P $ \input i k -> p input i $ \i -> k i . f instance Applicative (Parser k r tok) where pure x = P $ \_ i k -> k i x P p1 <*> P p2 = P $ \input i k -> p1 input i $ \i f -> p2 input i $ \i x -> k i (f x) instance Alternative (Parser k r tok) where empty = P $ \_ _ _ -> return [] P p1 <|> P p2 = P $ \input i k -> liftM2 (++) (p1 input i k) (p2 input i k) class (Functor p, Applicative p, Alternative p, Monad p) => ParserClass p k r tok | p -> k, p -> r, p -> tok where -- | Runs the parser. parse :: p a -> [tok] -> [a] -- | Tries to print the parser, or returns 'PP.empty', depending on -- the implementation. This function might not terminate. grammar :: Show k => p a -> Doc -- | Parses a token satisfying the given predicate. The computed -- value is returned. sat' :: (tok -> Maybe a) -> p a -- | Uses the given function to modify the printed representation -- (if any) of the given parser. annotate :: (DocP -> DocP) -> p a -> p a -- | Memoises the given parser. -- -- Every memoised parser must be annotated with a /unique/ key. -- (Parametrised parsers must use distinct keys for distinct -- inputs.) memoise :: (Eq k, Hashable k, Show k) => k -> p r -> p r -- | Memoises the given parser, but only if printing, not if -- parsing. -- -- Every memoised parser must be annotated with a /unique/ key. -- (Parametrised parsers must use distinct keys for distinct -- inputs.) memoiseIfPrinting :: (Eq k, Hashable k, Show k) => k -> p r -> p r -- | Uses the given document as the printed representation of the -- given parser. The document's precedence is taken to be 'atomP'. doc :: ParserClass p k r tok => Doc -> p a -> p a doc d = annotate (\_ -> (d, atomP)) -- | Parses a token satisfying the given predicate. sat :: ParserClass p k r tok => (tok -> Bool) -> p tok sat p = sat' (\t -> if p t then Just t else Nothing) -- | Parses a single token. token :: ParserClass p k r tok => p tok token = doc (text "·") (sat' Just) -- | Parses a given token. tok :: (ParserClass p k r tok, Eq tok, Show tok) => tok -> p tok tok t = doc (text (show t)) (sat (t ==)) instance ParserClass (Parser k r tok) k r tok where parse p toks = flip evalState IntMap.empty $ unP p (listArray (0, n - 1) toks) 0 $ \j x -> if j == n then return [x] else return [] where n = List.genericLength toks grammar _ = PP.empty sat' p = P $ \input i k -> if inRange (bounds input) i then case p (input ! i) of Nothing -> return [] Just x -> (k $! (i + 1)) $! x else return [] annotate _ p = p memoiseIfPrinting _ p = p memoise key p = P $ \input i k -> do let alter j zero f m = IntMap.alter (Just . f . fromMaybe zero) j m lookupTable = fmap (\m -> Map.lookup key =<< IntMap.lookup i m) get insertTable v = modify' $ alter i Map.empty (Map.insert key v) v <- lookupTable case v of Nothing -> do insertTable (Value IntMap.empty [k]) unP p input i $ \j r -> do Just (Value rs ks) <- lookupTable insertTable (Value (alter j [] (r :) rs) ks) concat <$> mapM (\k -> k j r) ks -- See note [Reverse ks?]. Just (Value rs ks) -> do insertTable (Value rs (k : ks)) concat . concat <$> mapM (\(i, rs) -> mapM (k i) rs) (IntMap.toList rs) -- [Reverse ks?] -- -- If ks were reversed, then the code would be productive for some -- infinitely ambiguous grammars, including S ∷= S | ε. However, in -- some cases the results would not be fair (some valid results would -- never be returned). -- | An extended parser type, with some support for printing parsers. data ParserWithGrammar k r tok a = PG (Bool -> Either (Parser k r tok a) (Docs k)) -- ^ Invariant: If the boolean is 'True', then the result must be -- @'Left' something@, and if the boolean is 'False', then the -- result must be @'Right' something@. -- | Documents paired with precedence levels. type DocP = (Doc, Int) -- | Precedence of @>>=@. bindP :: Int bindP = 10 -- | Precedence of @<|>@. choiceP :: Int choiceP = 20 -- | Precedence of @<*>@. seqP :: Int seqP = 30 -- | Precedence of @⋆@ and @+@. starP :: Int starP = 40 -- | Precedence of atoms. atomP :: Int atomP = 50 -- | The extended parser type computes one top-level document, plus -- one document per encountered memoisation key. -- -- 'Nothing' is used to mark that a given memoisation key has been -- seen, but that no corresponding document has yet been stored. type Docs k = State (HashMap k (Maybe DocP)) DocP -- | A smart constructor. pg :: Parser k r tok a -> Docs k -> ParserWithGrammar k r tok a pg p d = PG $ \b -> if b then Left p else Right d -- | Extracts the parser. parser :: ParserWithGrammar k r tok a -> Parser k r tok a parser (PG p) = either id __IMPOSSIBLE__ (p True) -- | Extracts the documents. docs :: ParserWithGrammar k r tok a -> Docs k docs (PG p) = either __IMPOSSIBLE__ id (p False) instance Monad (ParserWithGrammar k r tok) where return = pure p >>= f = pg (parser p >>= parser . f) ((\(d, p) -> (mparens (p < bindP) d <+> text ">>= ?", bindP)) <$> docs p) instance Functor (ParserWithGrammar k r tok) where fmap f p = pg (fmap f (parser p)) (docs p) instance Applicative (ParserWithGrammar k r tok) where pure x = pg (pure x) (return (text "ε", atomP)) p1 <*> p2 = pg (parser p1 <*> parser p2) (liftM2 (\(d1, p1) (d2, p2) -> (sep [ mparens (p1 < seqP) d1 , mparens (p2 < seqP) d2 ], seqP)) (docs p1) (docs p2)) -- | A helper function. starDocs :: String -> ParserWithGrammar k r tok a -> Docs k starDocs s p = (\(d, p) -> (mparens (p < starP) d <+> text s, starP)) <$> docs p instance Alternative (ParserWithGrammar k r tok) where empty = pg empty (return (text "∅", atomP)) p1 <|> p2 = pg (parser p1 <|> parser p2) (liftM2 (\(d1, p1) (d2, p2) -> (sep [ mparens (p1 < choiceP) d1 , text "|" , mparens (p2 < choiceP) d2 ], choiceP)) (docs p1) (docs p2)) many p = pg (many (parser p)) (starDocs "⋆" p) some p = pg (some (parser p)) (starDocs "+" p) -- | Pretty-prints a memoisation key. prettyKey :: Show k => k -> DocP prettyKey key = (text ("<" ++ show key ++ ">"), atomP) -- | A helper function. memoiseDocs :: (Eq k, Hashable k, Show k) => k -> ParserWithGrammar k r tok r -> Docs k memoiseDocs key p = do r <- Map.lookup key <$> get case r of Just _ -> return () Nothing -> do modify' (Map.insert key Nothing) d <- docs p modify' (Map.insert key (Just d)) return (prettyKey key) instance ParserClass (ParserWithGrammar k r tok) k r tok where parse p = parse (parser p) sat' p = pg (sat' p) (return (text "", atomP)) annotate f p = pg (parser p) (f <$> docs p) memoise key p = pg (memoise key (parser p)) (memoiseDocs key p) memoiseIfPrinting key p = pg (parser p) (memoiseDocs key p) grammar p = d $+$ nest 2 (foldr1 ($+$) $ text "where" : map (\(k, d) -> fst (prettyKey k) <+> text "∷=" <+> maybe __IMPOSSIBLE__ fst d) (Map.toList ds)) where ((d, _), ds) = runState (docs p) Map.empty