lexer with line numbers

[cc1516.git] / lex.icl

diff --git a/lex.icl b/lex.icl
index e2d679a..82e6139 100644 (file)
--- a/lex.icl
+++ b/lex.icl
@@ -8,7 +8,7 @@ import StdBool
 import StdList
 import StdChar
 
-SingleCharTokens :: Map Char Token
+SingleCharTokens :: Map Char TokenValue
 SingleCharTokens = fromList [
        ('(', BraceOpenToken), (')', BraceCloseToken), ('{', CBraceOpenToken),
        ('}', CBraceCloseToken), ('[', SquareOpenToken), (']', SquareCloseToken),
@@ -22,47 +22,57 @@ 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
+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
 
-lex :: [Char] -> (Either String Token, [Char])
-lex [] = (Right EndOfFileToken, [])
-lex ['/':'/':xs] = lex $ dropWhile ((<>) '\n') xs
-lex ['/':'*':xs] =
-       lex $ tl $ map snd $ dropWhile ((<>) ('*', '/')) $ zip2 [zero:xs] xs
-lex ['v':'a':'r':xs] = (Right VarToken, xs)
-lex ['V':'o':'i':'d':xs] = (Right VoidToken, xs)
-lex ['r':'e':'t':'u':'r':'n':xs] = (Right ReturnToken, xs)
-lex ['i':'f':xs] = (Right IfToken, xs)
-lex ['e':'l':'s':'e':xs] = (Right ElseToken, xs)
-lex ['w':'h':'i':'l':'e':xs] = (Right WhileToken, xs)
-lex ['T':'r':'u':'e':xs] = (Right TrueToken, xs)
-lex ['F':'a':'l':'s':'e':xs] = (Right FalseToken, xs)
-lex ['I':'n':'t':xs] = (Right IntTypeToken, xs)
-lex ['C':'h':'a':'r':xs] = (Right CharTypeToken, xs)
-lex ['B':'o':'o':'l':xs] = (Right BoolTypeToken, xs)
-lex [':':':':xs] = (Right DoubleColonToken, xs)
-lex ['!':'=':xs] = (Right NotEqualToken, xs)
-lex ['<':'=':xs] = (Right LesserEqToken, xs)
-lex ['>':'=':xs] = (Right GreaterEqToken, xs)
-lex ['=':'=':xs] = (Right EqualsToken, xs)
-lex ['&':'&':xs] = (Right AmpersandsToken, xs)
-lex ['|':'|':xs] = (Right PipesToken, xs)
-lex ['-':'>':xs] = (Right ArrowToken, xs)
-lex ['\'':x:'\'':xs] = (Right (CharToken x), xs)
+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 = (Right (CharToken t), xs)
-       _ = (Left ("Unknown escape: \\" +++ toString x), [])
+       Just t = ret (CharToken t) xs
+       _ = err ("Unknown escape: \\" +++ toString x)
 lex ['-':t=:[x:xs]]
-| isDigit x = let (i, r) = span isDigit t in (Right (NumberToken ['-':i]), r)
-| otherwise = (Right DashToken, 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) = (Right tok, xs)
+       (Just tok) = ret tok xs
        Nothing
+       | x == '\n' = \i.lex xs (i+1)
        | isSpace x = lex xs
-       | isDigit x = let (i, r) = span isDigit t in (Right (NumberToken i), r)
-       | isAlpha x = let (i, r) = span isIdent t in (Right (IdentToken i), r)
+       | 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 = (Left ("Unexpected character: " +++ toString x), [])
+       | otherwise = err ("Unexpected character: " +++ toString x)