X-Git-Url: https://git.martlubbers.net/?a=blobdiff_plain;f=lex.icl;h=0b6272ced499dfc247406c05823354a78feec4a5;hb=e3f049be4fb2ed9b7f675f06a9a525c81af9946c;hp=cbdcd653ab2f20f4778ffd7f863c721c64a6aadc;hpb=82d6648e3777f5bb74fa1825f18cf41922f3b782;p=cc1516.git

diff --git a/lex.icl b/lex.icl
index cbdcd65..0b6272c 100644
--- a/lex.icl
+++ b/lex.icl
@@ -1,158 +1,76 @@
 implementation module lex
 
+import Control.Monad
 import Data.Either
-import Data.List
+import Data.Maybe
+import Data.Map
 import StdString
-import System.CommandLine
+import StdBool
+import StdList
+import StdChar
 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
-
-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
-	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
+import System.CommandLine
+import qualified Text
+
+SingleCharTokens :: Map Char Token
+SingleCharTokens = fromList [
+	('(', BraceOpenToken), (')', BraceCloseToken), ('{', CBraceOpenToken),
+	('}', CBraceCloseToken), ('[', SquareOpenToken), (']', SquareCloseToken),
+	(',', CommaToken), (':', ColonToken), (';', SColonToken),
+	('.', DotToken), ('+', PlusToken), ('-', DashToken), ('*', StarToken),
+	('/', SlashToken), ('%', PercentToken), ('=', AssignmentToken),
+	('<', LesserToken), ('>', BiggerToken), ('!', ExclamationToken)]
+
+EscapeMap :: Map Char Char
+EscapeMap = fromList [
+	('a', toChar 7), ('b', '\b'), ('f', '\f'), ('n', '\n'), ('r', '\r'),
+	('t', '\t'), ('v', '\v')]
+
+lexer :: [Char] -> LexerOutput
+lexer [] = Right []
+lexer x = case lex x of
+	(Right t, rest) = lexer rest >>= \ts.Right [t:ts]
+	(Left e, _) = Left e
+
+lex :: [Char] -> (Either String Token, [Char])
+lex [] = (Right EndOfFileToken, [])
+//Comments
+lex ['/':'/':x:rest] = if (x == '\n') (lex rest) (lex ['/':'/':rest])
+lex ['/':'*':x1:x2:rest] = if (x1 == '*' && x2 == '/') 
+	(lex rest) (lex ['/':'*':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
+//Value tokens
+lex ['\'':x:'\'':rest] = (Right (CharToken x), rest)
+lex ['\'':'\\':x:'\'':rest] = case get x EscapeMap of
+	Just t = (Right (CharToken t), rest)
+	_ = (Left ("Unknown escape: \\" +++ toString x), [])
+lex [x:rest]
+# t = get x SingleCharTokens
+| isJust t = (Right (fromJust t), rest)
+| isSpace x = lex rest
+| isDigit x = let (is, rest`) = span isDigit rest in
+	(Right (NumberToken [x:is]), rest`)
+| isAlpha x = let (is, rest`) = span (\c.isAlphanum c || c == '_') rest in
+	(Right (IdentToken [x:is]), rest`)
+| otherwise = (Left ("Unexpected character: " +++ toString x), [])