implementation module lex
-import Data.Either
-import Data.List
+import Control.Monad, Control.Applicative
+import Data.Either, Data.Func, Data.Void
+import Data.Map
+from StdFunc import o
+import StdBool
+import StdList
+import StdChar
import StdString
-import System.CommandLine
-import StdFile
-import StdMisc
-from StdFunc import id, const
-import Data.Maybe
-import Control.Applicative
-import Control.Monad
-import Control.Monad.State
-from Data.Func import $
-
-// Misschien moeten we hier continuation style van maken
-instance toString lexerOutput where
- toString l = "dit is een lexer output, danwel error\n"
-
-lexer :: [Char] -> LexerOutput [Token]
-lexer _ = Left "Not Implemented"
-
-//lexer functions
-identT = alpha >>= \a -> many (char '_' <|> alphaNum) >>= \as -> return $ IdentToken [a:as]
-numberT = optional (char '-') >>= \sign -> (some digit) >>= \n -> case sign of
- Nothing -> return $ NumberToken $ 5 //fromString n
- _ -> return $ NumberToken $ -5 //(fromString n) * -1
-charLT = liftM CharToken item
-char2T = item >>= \c1 -> case c1 of
- ':' = char ':' >>| return DoubleColonToken
- '<' = char '=' >>| return LesserEqToken
- '>' = char '=' >>| return GreaterEqToken
- '=' = char '=' >>| return EqualsToken
- '&' = char '&' >>| return AmpersandsToken
- '|' = char '|' >>| return PipesToken
- '-' = char '>' >>| return ArrowToken
-char1T = item >>= \c1 -> findT c1 charTokenMap
-varT = string (fromString "var") >>| return VarToken
-voidT = string (fromString "Void") >>| return VoidToken
-returnT = string (fromString "return") >>| return ReturnToken
-ifT = string (fromString "if") >>| return IfToken
-elseT = string (fromString "else") >>| return ElseToken
-whileT = string (fromString "while") >>| return WhileToken
-trueT = string (fromString "True") >>| return TrueToken
-falseT = string (fromString "False") >>| return FalseToken
-
-Start :: *World -> *World
-Start w
-# (args, w) = getCommandLine w // We lezen nu nog standaard van stdin
-# (out, w) = stdio w
-# (toparse, out) = readEntireFile out
-# out = out <<< toString (lexer toparse)
-# (b, w) = fclose out w
-| not b = setReturnCode 1 w
-= w
+import StdTuple
+
+import yard
+import AST
+
+:: LexItem
+ = LexToken Int TokenValue
+ | LexSpace Int Int
+ | LexNL
+ | LexEOF
+ | LexItemError String
+
+escapes :: Map Char Char
+escapes = fromList [('a', toChar 7), ('b', '\b'), ('f', '\f'), ('n', '\n'),
+ ('r', '\r'), ('t', '\t'), ('v', '\v'), ('\'', '\''), ('"', '"')]
+
+lexer :: [Char] -> LexerOutput
+lexer r = fst $ runParser (lexProgram 1 1) r
+
+lexProgram :: Int Int -> Parser Char [Token]
+lexProgram line column = lexToken >>= \t->case t of
+ LexEOF = pure []
+ LexNL = lexProgram (line+1) 1
+ (LexSpace l c) = lexProgram (line+l) (column+c)
+ (LexItemError e) = fail <?>
+ PositionalError line column ("LexerError: " +++ e)
+ (LexToken c t) = lexProgram line (column+c)
+ >>= \rest->pure [({line=line,col=column}, t):rest]
+
+lexToken :: Parser Char LexItem
+lexToken =
+ //Comments
+ lexBlockComment <|> lexComment <|>
+ //Keyword tokens
+ lexKw "var" VarToken <|> lexKw "Void" VoidToken <|>
+ lexKw "return" ReturnToken <|> lexKw "if" IfToken <|>
+ lexKw "else" ElseToken <|> lexKw "while" WhileToken <|>
+ lexKw "True" TrueToken <|> lexKw "False" FalseToken <|>
+ lexKw "Int" IntTypeToken <|> lexKw "Bool" BoolTypeToken <|>
+ lexKw "Char" CharTypeToken <|>
+ lexKw "let" LetToken <|>
+ //Character tokens
+ lexEscape <|> lexCharacter <|>
+ //Two char ops tokens
+ lexWord "::" DoubleColonToken <|> lexWord "!=" NotEqualToken <|>
+ lexWord "<=" LesserEqToken <|> lexWord ">=" GreaterEqToken <|>
+ lexWord "==" EqualsToken <|> lexWord "&&" AmpersandsToken <|>
+ lexWord "||" PipesToken <|> lexWord "[]" EmptyListToken <|>
+ lexWord "->" ArrowToken <|>
+ //One char ops tokens
+ lexWord "(" BraceOpenToken <|> lexWord ")" BraceCloseToken <|>
+ lexWord "{" CBraceOpenToken <|> lexWord "}" CBraceCloseToken <|>
+ lexWord "[" SquareOpenToken <|> lexWord "]" SquareCloseToken <|>
+ lexWord "," CommaToken <|> lexWord ":" ColonToken <|>
+ lexWord ";" SColonToken <|> lexWord "." DotToken <|>
+ lexWord "+" PlusToken <|> lexWord "*" StarToken <|>
+ lexWord "/" SlashToken <|> lexWord "%" PercentToken <|>
+ lexWord "=" AssignmentToken <|> lexWord "<" LesserToken <|>
+ lexWord ">" BiggerToken <|> lexWord "!" ExclamationToken <|>
+ lexWord "-" DashToken <|> lexWord "\\" BackslashToken <|>
+ //Number and identifier tokens
+ lexString <|> lexNumber <|> lexIdentifier <|>
+ (item '\n' >>| pure LexNL) <|>
+ //Whitespace
+ (satisfy isSpace >>| (pure $ LexSpace 0 1)) <|>
+ (eof >>| pure LexEOF)
where
- readEntireFile :: *File -> *([Char], *File)
- readEntireFile f
- # (b, c, f) = freadc f
- | not b = ([], f)
- # (cs, f) = readEntireFile f
- = ([c:cs], f)
-
-
-
-charTokenMap = [('(', BraceOpenToken)
- ,(')', BraceCloseToken)
- ,('{', CBraceOpenToken)
- ,('}', CBraceCloseToken)
- ,('[', SquareOpenToken)
- ,(']', SquareCloseToken)
- ,(',', CommaToken)
- ,(':', ColonToken)
- ,(';', SColonToken)
- ,('.', DotToken)
- ,('+', PlusToken)
- ,('-', DashToken)
- ,('*', StarToken)
- ,('/', SlashToken)
- ,('%', PercentToken)
- ,('=', AssignmentToken)
- ,('<', LesserToken)
- ,('>', BiggerToken)
- ,('!', ExclamationToken)]
-findT c [] = fail "Unrecognized character"
-findT c [(k,v):xs] = if (c==k) (return v) (findT c xs)
-
-
-// Clean adaption of Yard, a parsec like parser combinator
-:: Parser a = Parser ([Char] -> (LexerOutput a, [Char]))
-
-runParser :: (Parser a) [Char] -> (LexerOutput a, [Char])
-runParser (Parser p) s = p s
-
-instance Functor Parser where
- fmap f s = liftM f s
-
-instance Applicative Parser where
- pure a = Parser $ \s -> (Right a, s)
- (<*>) sf s = ap sf s
-
-instance Monad Parser where
- bind p f = Parser $ \s -> let (out, rest) = runParser p s in case out of
- Left e = (Left e, rest)
- Right t = runParser (f t) rest
-
-//gives us some, many and optional
-instance Alternative Parser where
- empty = zero
- (<|>) p1 p2 = Parser $ \s -> let (out, rest) = runParser p1 s in case out of
- Left e = runParser p2 s
- Right t = (Right t, rest)
-
-//parser that fails with error
-fail :: String -> Parser a
-fail e = Parser $ \s -> (Left e, s)
-
-//parser that always fails
-zero :: Parser a
-zero = fail "Zero parser"
-
-//matches exactly one Char
-item :: Parser Char
-item = Parser $ \s -> case s of
- [] = (Left "Unexpected empty input", s)
- [x:xs] = (Right x, xs)
-
-//matches any char which satisfies f
-satisfy :: (Char -> Bool) -> Parser Char
-satisfy f = item >>= (\r -> if (f r) (return r) zero)
-
-//tries a parser, if it fails returns a default value
-optionalDef :: a (Parser a) -> Parser a
-optionalDef def p = p <|> return def
-
-//matched given char
-char :: Char -> Parser Char
-char c = satisfy (\i -> c==i) //I hate that we can't do: satisfy (==c)
-
-alpha :: Parser Char
-alpha = satisfy isAlpha
-
-digit :: Parser Char
-digit = satisfy isDigit
-
-alphaNum :: Parser Char
-alphaNum = alpha <|> digit
-
-//matches a given String
-string :: [Char] -> Parser [Char]
-string s = mapM_ char s >>| return s
\ No newline at end of file
+ lexWord :: String TokenValue -> Parser Char LexItem
+ lexWord s tv = list ls >>| pure (LexToken (length ls) tv)
+ where ls = fromString s
+
+ lexKw :: String TokenValue -> Parser Char LexItem
+ lexKw kw tv = lexWord kw tv <* check (not o isIdentChar)
+
+ lexComment :: Parser Char LexItem
+ lexComment = list (fromString "//")
+ >>| top until (eof <|> (item '\n' >>| pure Void)) >>| pure LexNL
+
+ lexBlockComment :: Parser Char LexItem
+ lexBlockComment = list (fromString "/*")
+ >>| (top until (list (fromString "*/")))
+ >>= \chars->pure $ widthHeight chars 0 0
+ where
+ widthHeight :: [Char] Int Int -> LexItem
+ widthHeight [] l c = LexSpace l c
+ widthHeight ['\n':xs] l _ = widthHeight xs (l+1) 0
+ widthHeight [x:xs] l c = widthHeight xs l (c+1)
+
+ lexNumber :: Parser Char LexItem
+ lexNumber = some (satisfy isDigit) >>= \si->pure $
+ LexToken (length si) (NumberToken $ toInt $ toString si)
+
+ lexIdentifier :: Parser Char LexItem
+ lexIdentifier = some (satisfy isIdentChar)
+ >>= \si->pure $ LexToken (length si) (IdentToken $ toString si)
+
+ isIdentChar c = isAlphanum c || c == '_'
+
+ lexCharacter :: Parser Char LexItem
+ lexCharacter = item '\'' *> satisfy ((<>) '\'') <* item '\''
+ >>= \char->pure $ LexToken 3 (CharToken char)
+
+ lexEscape :: Parser Char LexItem
+ lexEscape = item '\'' *> item '\\' *> top <* item '\''
+ >>= \char->pure case get char escapes of
+ Just e = LexToken 4 (CharToken e)
+ Nothing = LexItemError $ "Unknown escape: " +++ toString char
+
+ lexString :: Parser Char LexItem
+ lexString = item '"' *> (
+ many (
+ (satisfy (\c->c <> '"' && c <> '\\')) <|>
+ (item '\\' *> top >>= \char->case get char escapes of
+ Just e = pure e
+ Nothing = empty)
+ ))<* item '"' >>= \cs-> pure $ LexToken (length cs) (StringToken cs)
repositories / cc1516.git / blobdiff