negative numbers are lexed

[cc1516.git] / lex.icl

diff --git a/lex.icl b/lex.icl
index 34d913c..0adb20c 100644 (file)
--- a/lex.icl
+++ b/lex.icl
@@ -1,40 +1,72 @@
-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
+import Data.Maybe
+import Data.Map
+import Data.Func
+import StdString
+import StdTuple
+import StdBool
+import StdList
+import StdChar
+import qualified Text
 
-Start = "Hello World!\n"
+SingleCharTokens :: Map Char Token
+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 [] = 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, [])
+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)
+lex ['\'':'\\':x:'\'':xs] = case get x EscapeMap of
+       Just t = (Right (CharToken t), xs)
+       _ = (Left ("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)
+lex t=:[x:xs] = case get x SingleCharTokens of
+       (Just tok) = (Right tok, xs)
+       Nothing
+       | 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)
+               with isIdent c = isAlphanum c || c == '_'
+       | otherwise = (Left ("Unexpected character: " +++ toString x), [])