(Left e, _) = Left $ toString e
lexProgram :: Parser Char [Token]
-lexProgram = some lexToken <* eof >>= \ts->pure (map (\t->(0, 0, t)) ts)
+lexProgram = some lexToken <* many (satisfy isSpace) <* eof
+ >>= \ts->pure $ (map (\t->(0, 0, t)) ts) ++ [(0, 0, EndOfFileToken)]
lexToken :: Parser Char TokenValue
lexToken =
+ //Comments
+ (list (fromString "//") >>| lexUntilNL >>| lexToken) <|>
+ (list (fromString "/*") >>| lexUntilCommentClose >>| lexToken) <|>
//Keyword tokens
(lexKw "var" VarToken) <|>
(lexKw "Void" VoidToken) <|>
(lexOp "<=" LesserEqToken) <|> (lexOp ">=" GreaterEqToken) <|>
(lexOp "==" EqualsToken) <|> (lexOp "&&" AmpersandsToken) <|>
(lexOp "||" PipesToken) <|> (lexOp "[]" EmptyListToken) <|>
- (lexOp "->" ArrowToken) <|> (lexOp "(" BraceOpenToken) <|>
+ (lexOp "->" ArrowToken) <|>
//One char ops tokens
+ (lexOp "(" BraceOpenToken) <|>
(lexOp ")" BraceCloseToken) <|> (lexOp "{" CBraceOpenToken) <|>
(lexOp "}" CBraceCloseToken) <|> (lexOp "[" SquareOpenToken) <|>
(lexOp "]" SquareCloseToken) <|> (lexOp "," CommaToken) <|>
//Number tokens
(liftM (NumberToken o toInt o toString) $ some $ satisfy isDigit) <|>
//Ident tokens
- (liftM IdentToken $ some $ satisfy isIdentChar) <|>
- (satisfy isSpace >>| lexToken) //<|>
-// (eof >>| pure EndOfFileToken)
+ (liftM (IdentToken o toString) $ some $ satisfy isIdentChar) <|>
+ (satisfy isSpace >>| lexToken)
where
+ lexUntilNL = top until (eof <|> (item '\n' >>| pure Void))
+ lexUntilCommentClose = top until list (fromString "*/")
isIdentChar c = isAlphanum c || c == '_'
lexOp s tv = list (fromString s) >>| pure tv
lexKw kw tv = lexOp kw tv <* check (not o isIdentChar) >>| pure tv