+lexToken :: Parser Char TokenValue
+lexToken =
+ //Comments
+ (list (fromString "//") >>| until top ((item '\n' >>| return Void) <|> eof) >>| lexToken) <|>
+ (list (fromString "/*") >>| until top (list (fromString "*/")) >>| lexToken) <|>
+ //Keyword tokens
+ (lexKw "var" VarToken) <|>
+ (lexKw "Void" VoidToken) <|>
+ (lexKw "return" ReturnToken) <|>
+ (lexKw "if" IfToken) <|>
+ (lexKw "else" ElseToken) <|>
+ (lexKw "while" WhileToken) <|>
+ (lexKw "True" TrueToken) <|>
+ (lexKw "False" FalseToken) <|>
+ (lexKw "Int" IntTypeToken) <|>
+ (lexKw "Bool" BoolTypeToken) <|>
+ (lexKw "Char" CharTypeToken) <|>
+ //Escape chars tokens
+ (liftM CharToken $ item '\'' *> item '\\' *> lexEscape <* item '\'') <|>
+ //Normal chars tokens
+ (liftM CharToken $ item '\'' *> satisfy ((<>) '\'') <* item '\'') <|>
+ //Two char ops tokens
+ (lexOp "::" DoubleColonToken) <|> (lexOp "!=" NotEqualToken) <|>
+ (lexOp "<=" LesserEqToken) <|> (lexOp ">=" GreaterEqToken) <|>
+ (lexOp "==" EqualsToken) <|> (lexOp "&&" AmpersandsToken) <|>
+ (lexOp "||" PipesToken) <|> (lexOp "[]" EmptyListToken) <|>
+ (lexOp "->" ArrowToken) <|>
+ //One char ops tokens
+ (lexOp "(" BraceOpenToken) <|>
+ (lexOp ")" BraceCloseToken) <|> (lexOp "{" CBraceOpenToken) <|>
+ (lexOp "}" CBraceCloseToken) <|> (lexOp "[" SquareOpenToken) <|>
+ (lexOp "]" SquareCloseToken) <|> (lexOp "," CommaToken) <|>
+ (lexOp ":" ColonToken) <|> (lexOp ";" SColonToken) <|>
+ (lexOp "." DotToken) <|> (lexOp "+" PlusToken) <|>
+ (lexOp "*" StarToken) <|> (lexOp "/" SlashToken) <|>
+ (lexOp "%" PercentToken) <|> (lexOp "=" AssignmentToken) <|>
+ (lexOp "<" LesserToken) <|> (lexOp ">" BiggerToken) <|>
+ (lexOp "!" ExclamationToken) <|> (lexOp "-" DashToken) <|>
+ //Number tokens
+ (liftM (NumberToken o toInt o toString) $ some $ satisfy isDigit) <|>
+ //Ident tokens
+ (liftM IdentToken $ some $ satisfy isIdentChar) <|>
+ (satisfy isSpace >>| lexToken)
+// (eof >>| pure EndOfFileToken)
+ where
+ 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
+ lexEscape = (
+ lexOp "a" (toChar 7) <|> lexOp "b" '\b' <|> lexOp "f" '\f' <|>
+ lexOp "n" '\n' <|> lexOp "r" '\t' <|> lexOp "v" '\v' <|>
+ lexOp "'" '\'') <?> ("Unknown escape", 0)