implementation module lex
+from Text import class Text, instance Text String
+import Control.Monad
import Data.Either
+import Data.Functor
import Data.List
import StdString
-import System.CommandLine
+import StdBool
import StdFile
-import StdMisc
-from StdFunc import id, const
-import Data.Maybe
-import Control.Applicative
-import Control.Monad
-import Control.Monad.State
-import Data.Functor
-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 i = let (out, rest) = runParser (many anyT) i in case rest of
- [] = out
- _ = Left "Unexpected input after last token"
-
-//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 = CharToken <$> (char '\'' *> item <* char '\'')
-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
- _ = zero
-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
-//note, for the anyToken parser the order matters!
-anyT = char2T <|> char1T <|> varT <|> voidT <|> returnT <|> ifT <|> elseT <|> whileT <|>
- trueT <|> falseT <|> numberT <|> identT <|> charLT
+import System.CommandLine
+import qualified Text
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) = fclose (out <<< "\n" <<< toString (lexer toparse)) w
= w
- 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
+readEntireFile :: *File -> *([Char], *File)
+readEntireFile f
+# (b, c, f) = freadc f
+| not b = ([], f)
+# (cs, f) = readEntireFile f
+= ([c:cs], f)
+
+lexer :: [Char] -> LexerOutput
+lexer [] = Right []
+lexer x = case lex x of
+ (Right t, rest) = lexer rest >>= \ts.Right [t:ts]
+ (Left e, _) = Left e
+
+untilNext :: [Char] -> [Char]
+untilNext [] = []
+untilNext ['*':'/':rest] = rest
+untilNext [x:rest] = untilNext rest
+
+lex :: [Char] -> (Either String Token, [Char])
+lex [] = (Right EndOfFileToken, [])
+//Comments
+lex ['/':'/':rest] = lex (tl (dropWhile ((<>) '\n') rest))
+lex ['/':'*':rest] = lex (untilNext rest)
+//Keyword tokens
+lex ['v':'a':'r':rest] = (Right VarToken, rest)
+lex ['V':'o':'i':'d':rest] = (Right VoidToken, rest)
+lex ['r':'e':'t':'u':'r':'n':rest] = (Right ReturnToken, rest)
+lex ['i':'f':rest] = (Right IfToken, rest)
+lex ['e':'l':'s':'e':rest] = (Right ElseToken, rest)
+lex ['w':'h':'i':'l':'e':rest] = (Right WhileToken, rest)
+lex ['T':'r':'u':'e':rest] = (Right TrueToken, rest)
+lex ['F':'a':'l':'s':'e':rest] = (Right FalseToken, rest)
+lex ['I':'n':'t':rest] = (Right IntTypeToken, rest)
+lex ['C':'h':'a':'r':rest] = (Right CharTypeToken, rest)
+lex ['B':'o':'o':'l':rest] = (Right BoolTypeToken, rest)
+//Two character tokens
+lex [':':':':rest] = (Right DoubleColonToken, rest)
+lex ['!':'=':rest] = (Right NotEqualToken, rest)
+lex ['<':'=':rest] = (Right LesserEqToken, rest)
+lex ['>':'=':rest] = (Right GreaterEqToken, rest)
+lex ['=':'=':rest] = (Right EqualsToken, rest)
+lex ['&':'&':rest] = (Right AmpersandsToken, rest)
+lex ['|':'|':rest] = (Right PipesToken, rest)
+lex ['-':'>':rest] = (Right ArrowToken, rest)
+//One character tokens
+lex ['(':rest] = (Right BraceOpenToken, rest)
+lex [')':rest] = (Right BraceCloseToken, rest)
+lex ['{':rest] = (Right CBraceOpenToken, rest)
+lex ['}':rest] = (Right CBraceCloseToken, rest)
+lex ['[':rest] = (Right SquareOpenToken, rest)
+lex [']':rest] = (Right SquareCloseToken, rest)
+lex [',':rest] = (Right CommaToken, rest)
+lex [':':rest] = (Right ColonToken, rest)
+lex [';':rest] = (Right SColonToken, rest)
+lex ['.':rest] = (Right DotToken, rest)
+lex ['+':rest] = (Right PlusToken, rest)
+lex ['-':rest] = (Right DashToken, rest)
+lex ['*':rest] = (Right StarToken, rest)
+lex ['/':rest] = (Right SlashToken, rest)
+lex ['%':rest] = (Right PercentToken, rest)
+lex ['=':rest] = (Right AssignmentToken, rest)
+lex ['<':rest] = (Right LesserToken, rest)
+lex ['>':rest] = (Right BiggerToken, rest)
+lex ['!':rest] = (Right ExclamationToken, rest)
+//Value tokens
+lex ['\'':x:'\'':rest] = (Right (CharToken x), rest)
+lex ['\'':'\\':x:'\'':rest] = case x of
+ 'a' = (Right (CharToken (toChar 7)), rest) //Alarm
+ 'b' = (Right (CharToken '\b'), rest) //Backspace
+ 'f' = (Right (CharToken '\f'), rest) //Formfeed
+ 'n' = (Right (CharToken '\n'), rest) //Newline
+ 'r' = (Right (CharToken '\r'), rest) //Carriage Return
+ 't' = (Right (CharToken '\t'), rest) //Horizontal tab
+ 'v' = (Right (CharToken '\v'), rest) //Vertical tab
+ _ = (Left ("Unknown escape: \\" +++ toString x), [])
+lex [x:xs]
+| isSpace x = lex xs
+| isDigit x = let (is, rest) = span isDigit xs in
+ (Right (NumberToken [x:is]), rest)
+| isAlpha x = let (is, rest) = span (\c.isAlphanum c || c == '_') xs in
+ (Right (IdentToken [x:is]), rest)
+| otherwise = (Left ("Unexpected character: " +++ toString x), [])
+
+instance toString LexerOutput where
+ toString (Left l) = "Error: " +++ l
+ toString (Right x) = 'Text'.concat (print 0 x)
+
+tab :: Int -> [String]
+tab i = replicate i "\t"
+
+print :: Int [Token] -> [String]
+print _ [] = []
+print i [(IdentToken l):rest] = [toString l:print i rest]
+print i [(NumberToken j):rest] = [toString j:print i rest]
+print i [(CharToken c):rest] = ["'":toString c:"'":print i rest]
+print i [VarToken:rest] = ["var ":print i rest]
+print i [ReturnToken:rest] = ["return ":print i rest]
+print i [IfToken:rest] = ["if":print i rest]
+print i [ElseToken:rest] = ["else ":print i rest]
+print i [WhileToken:rest] = ["while":print i rest]
+print i [TrueToken:rest] = ["True":print i rest]
+print i [FalseToken:rest] = ["False":print i rest]
+print i [VoidToken:rest] = ["Void":print i rest]
+print i [IntTypeToken:rest] = ["Int":print i rest]
+print i [CharTypeToken:rest] = ["Char":print i rest]
+print i [BoolTypeToken:rest] = ["Bool":print i rest]
+print i [DoubleColonToken:rest] = [" :: ":print i rest]
+print i [NotEqualToken:rest] = [" != ":print i rest]
+print i [LesserEqToken:rest] = [" <= ":print i rest]
+print i [GreaterEqToken:rest] = [" >= ":print i rest]
+print i [EqualsToken:rest] = [" == ":print i rest]
+print i [AmpersandsToken:rest] = [" && ":print i rest]
+print i [PipesToken:rest] = [" || ":print i rest]
+print i [ArrowToken:rest] = [" -> ":print i rest]
+print i [BraceOpenToken:rest] = ["(":print i rest]
+print i [BraceCloseToken:rest] = [")":print i rest]
+print i [CBraceOpenToken:rest] = ["{\n":tab (i+1)] ++ print (i+1) rest
+print i [CBraceCloseToken:rest] = case rest of
+ [CBraceCloseToken:_] = ["}\n":tab (i-2)] ++ print (i-1) rest
+ _ = ["}\n":tab (i-1)] ++ print (i-1) rest
+print i [SquareOpenToken:rest] = ["[":print i rest]
+print i [SquareCloseToken:rest] = ["]":print i rest]
+print i [CommaToken:rest] = [", ":print i rest]
+print i [ColonToken:rest] = [":":print i rest]
+print i [SColonToken:rest] = case rest of
+ [CBraceCloseToken:_] = [";\n":tab (i-1)] ++ print i rest
+ _ = [";\n":tab i] ++ print i rest
+print i [DotToken:rest] = [".":print i rest]
+print i [PlusToken:rest] = [" + ":print i rest]
+print i [DashToken:rest] = [" - ":print i rest]
+print i [StarToken:rest] = [" * ":print i rest]
+print i [SlashToken:rest] = [" / ":print i rest]
+print i [PercentToken:rest] = [" % ":print i rest]
+print i [AssignmentToken:rest] = [" = ":print i rest]
+print i [LesserToken:rest] = [" < ":print i rest]
+print i [BiggerToken:rest] = [" > ":print i rest]
+print i [ExclamationToken:rest] = ["!":print i rest]
+print i [EndOfFileToken:rest] = ["\n":print i rest]
repositories / cc1516.git / blobdiff