implementation module lex
import Control.Monad
-import Data.Either
-import Data.Maybe
-import Data.Map
+import Data.Either, Data.Maybe, Data.Map, Data.Func
import StdString
+import StdTuple
import StdBool
import StdList
import StdChar
-import StdFile
-import System.CommandLine
-import qualified Text
-SingleCharTokens :: Map Char Token
+SingleCharTokens :: Map Char TokenValue
SingleCharTokens = fromList [
('(', BraceOpenToken), (')', BraceCloseToken), ('{', CBraceOpenToken),
('}', CBraceCloseToken), ('[', SquareOpenToken), (']', SquareCloseToken),
(',', CommaToken), (':', ColonToken), (';', SColonToken),
- ('.', DotToken), ('+', PlusToken), ('-', DashToken), ('*', StarToken),
- ('/', SlashToken), ('%', PercentToken), ('=', AssignmentToken),
- ('<', LesserToken), ('>', BiggerToken), ('!', ExclamationToken)]
+ ('.', 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')]
+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, [])
-//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), [])
+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 (toInt $ toString ['-':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 (toInt $ toString 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)