X-Git-Url: https://git.martlubbers.net/?a=blobdiff_plain;f=lex.icl;h=82e61393b9b9b1e7ac266e02c1e68b7723bb4e46;hb=345b80b49413bb1b52735fb155faded41456a6b7;hp=34d913c0ee42ed21cdbf0354a4b4337f4657ff1e;hpb=d5190ca02578d0c678bf3f724feeaeb37ac766f8;p=cc1516.git

diff --git a/lex.icl b/lex.icl
index 34d913c..82e6139 100644
--- a/lex.icl
+++ b/lex.icl
@@ -1,40 +1,78 @@
-module lex
+implementation module lex
 
-:: Token
-	//Values
-	= IdentToken String // Identifier
-	| NumberToken Int   // Integer
-	//Keywords
-	| VarToken          // var keyword
-	| VoidToken         // void keyword
-	| ReturnToken       // return keyword
-	| IfToken           // if keyword
-	| ElseToken         // else keyword
-	| WhileToken        // while keyword
-	| TrueToken         // True keyword
-	| FalseToken        // False keyword
-	//Single Characters
-	| BraceOpenToken    // (
-	| BraceCloseToken   // )
-	| CBraceOpenToken   // {
-	| CBraceCloseToken  // {
-	| SquareOpenToken   // [
-	| SquareCloseToken  // ]
-	| CommaToken        // ,
-	| ColonToken        // :
-	| SColonToken       // ;
-	| DotToken          // .
-	| PlusToken         // +
-	| DashToken         // -
-	| StarToken         // *
-	| SlashToken        // /
-	| PercentToken      // %
-	| EqualsToken       // =
-	| LesserToken       // <
-	| BiggerToken       // >
-	| ExclamationToken  // !
-	| AmpersandToken    // &
-	| PipeToken         // |
-	| SingleQuoteToken  // '
+import Control.Monad
+import Data.Either, Data.Maybe, Data.Map, Data.Func
+import StdString
+import StdTuple
+import StdBool
+import StdList
+import StdChar
 
-Start = "Hello World!\n"
+SingleCharTokens :: Map Char TokenValue
+SingleCharTokens = fromList [
+	('(', BraceOpenToken), (')', BraceCloseToken), ('{', CBraceOpenToken),
+	('}', CBraceCloseToken), ('[', SquareOpenToken), (']', SquareCloseToken),
+	(',', CommaToken), (':', ColonToken), (';', SColonToken),
+	('.', DotToken), ('+', PlusToken), ('*', 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 x = lexer` x 1
+	where
+		lexer` :: [Char] Int -> LexerOutput
+		lexer` [] i = Right []
+		lexer` x i = case lex x i of
+			(Right t, rest, i) = lexer` rest i >>= \ts.Right [t:ts]
+			(Left e, _, _) = Left e
+
+ret:: TokenValue [Char] Int -> (Either String Token, [Char], Int)
+ret tv rest i = (Right (i, tv), rest, i)
+
+err :: String Int -> (Either String Token, [Char], Int)
+err s i = (Left ("Line " +++ toString i +++ ": " +++ s), [], i)
+
+lex :: [Char] -> (Int -> (Either String Token, [Char], Int))
+lex [] = ret EndOfFileToken []
+lex ['/':'/':xs] = lex (dropWhile ((<>) '\n') xs)
+lex ['/':'*':xs] =  \i.let (c, r) = span ((<>) ('*','/')) $ zip2 [' ':xs] xs in
+	lex (tl $ map snd $ tl $ r) (i + length [x\\(x, _)<-c | x == '\n'])
+lex ['v':'a':'r':xs] = ret VarToken xs
+lex ['V':'o':'i':'d':xs] = ret VoidToken xs
+lex ['r':'e':'t':'u':'r':'n':xs] = ret ReturnToken xs
+lex ['i':'f':xs] = ret IfToken xs
+lex ['e':'l':'s':'e':xs] = ret ElseToken xs
+lex ['w':'h':'i':'l':'e':xs] = ret WhileToken xs
+lex ['T':'r':'u':'e':xs] = ret TrueToken xs
+lex ['F':'a':'l':'s':'e':xs] = ret FalseToken xs
+lex ['I':'n':'t':xs] = ret IntTypeToken xs
+lex ['C':'h':'a':'r':xs] = ret CharTypeToken xs
+lex ['B':'o':'o':'l':xs] = ret BoolTypeToken xs
+lex [':':':':xs] = ret DoubleColonToken xs
+lex ['!':'=':xs] = ret NotEqualToken xs
+lex ['<':'=':xs] = ret LesserEqToken xs
+lex ['>':'=':xs] = ret GreaterEqToken xs
+lex ['=':'=':xs] = ret EqualsToken xs
+lex ['&':'&':xs] = ret AmpersandsToken xs
+lex ['|':'|':xs] = ret PipesToken xs
+lex ['-':'>':xs] = ret ArrowToken xs
+lex ['\'':x:'\'':xs] = ret (CharToken x) xs
+lex ['\'':'\\':x:'\'':xs] = case get x EscapeMap of
+	Just t = ret (CharToken t) xs
+	_ = err ("Unknown escape: \\" +++ toString x)
+lex ['-':t=:[x:xs]]
+| isDigit x = let (n, r) = span isDigit t in ret (NumberToken ['-':n]) r
+| otherwise = ret DashToken  xs
+lex t=:[x:xs] = case get x SingleCharTokens of
+	(Just tok) = ret tok xs
+	Nothing
+	| x == '\n' = \i.lex xs (i+1)
+	| isSpace x = lex xs
+	| isDigit x = let (v, r) = span isDigit t in ret (NumberToken v) r
+	| isAlpha x = let (v, r) = span isIdent t in ret (IdentToken v) r
+		with isIdent c = isAlphanum c || c == '_'
+	| otherwise = err ("Unexpected character: " +++ toString x)