- return wrapped_old_x;
- }
- };
-}
-exports.zeroOrOnceDo = zeroOrOnceDo;
-function tokenize(input) {
- var input_matchee_pair = toSome({ matched: "",
- remained: input });
- /**
- * generate a parser of a basic term (b_term)
- * @param pattern : the pattern parser
- * @param token_type : the returning token type
- * @returns a wrapped parser.
- */
- function bTerm(pattern, token_type) {
- return (x) => {
- let wrapped_x = toSome(x);
- let result = pattern(wrapped_x);
- if (result._tag == "Some") {
- result.value.matched_type = token_type;
- }
- return result;
- };
- }
- let d = matchRange('0', '9'); // \d
- // [+-]
- let plusMinus = orDo(match1Char('+'), match1Char('-'));
- let s_aux = orDo(match1Char(' '), match1Char('\t')); // (" " | "\t")
- // integer = ([+]|[-])?\d\d*
- let integer = bTerm((x) => thenDo(thenDo(thenDo(x, zeroOrOnceDo(plusMinus)), d), zeroOrMoreDo(d)), TokenType.INT);
- // space = [ \t]+
- let space = bTerm((x) => thenDo(thenDo(x, s_aux), zeroOrMoreDo(s_aux)), TokenType.INT);
- // newline = \r?\n
- let newline = bTerm((x) => thenDo(thenDo(x, zeroOrOnceDo(match1Char('\r'))), match1Char('\n')), TokenType.NL);
- // [_A-Za-z]
- let idHead = orDo(orDo(matchRange('a', 'z'), matchRange('A', 'Z')), match1Char('_'));
- let idRemained = orDo(idHead, matchRange('0', '9')); // [_A-Za-z0-9]
- // id = [_A-Za-z][_A-Za-z0-9]*
- let id = bTerm((x) => thenDo(thenDo(x, idHead), zeroOrMoreDo(idRemained)), TokenType.ID);
- let doublequote = match1Char("\"");
- // [\\][\"]
- let escapeReverseSlash = (x) => thenDo(thenDo(toSome(x), match1Char("\\")), doublequote);
- // ([\\]["]|[^\"])*
- let stringInnerPattern = zeroOrMoreDo(orDo(escapeReverseSlash, notDo(match1Char("\""))));
- // str = ["]([\\]["]|[^"])*["]
- let str = bTerm((x) => thenDo(thenDo(thenDo(x, doublequote), stringInnerPattern), doublequote), TokenType.STR);
- // float = [+-]?\d+[.]\d+
- function floatPattern(x) {
- return thenDo(thenDo(thenDo(thenDo(thenDo(thenDo(x, zeroOrOnceDo(plusMinus)), d), zeroOrMoreDo(d)), match1Char(".")), d), zeroOrMoreDo(d));
- }
- ;
- let float = bTerm(floatPattern, TokenType.FLO);
- // operators
- // +.
- let floatAdd = bTerm((x) => thenDo(thenDo(x, match1Char("+")), match1Char(".")), TokenType.F_ADD);
- // +.
- let floatSub = bTerm((x) => thenDo(thenDo(x, match1Char("-")), match1Char(".")), TokenType.F_SUB);
- // *.
- let floatMul = bTerm((x) => thenDo(thenDo(x, match1Char("*")), match1Char(".")), TokenType.F_MUL);
- // /.
- let floatDiv = bTerm((x) => thenDo(thenDo(x, match1Char("/")), match1Char(".")), TokenType.F_DIV);
- // ==
- let eq = bTerm((x) => thenDo(thenDo(x, match1Char("=")), match1Char("=")), TokenType.EQ);
- // >=
- let ge = bTerm((x) => thenDo(thenDo(x, match1Char(">")), match1Char("=")), TokenType.GE);
- // <=
- let le = bTerm((x) => thenDo(thenDo(x, match1Char("<")), match1Char("=")), TokenType.LE);
- // ->
- let rightArrow = bTerm((x) => thenDo(thenDo(x, match1Char("-")), match1Char(">")), TokenType.R_ARROW);
- /**
- * unary operator : generating the pattern of basic unary operator
- * @param char : uniry char for the operator
- * @param token_type : the corresponding token_type
- */
- function unaryOp(char, token_type) {
- return bTerm((x) => thenDo(x, match1Char(char)), token_type);
- }
- ;
- let intAdd = unaryOp('+', TokenType.I_ADD);
- let intSub = unaryOp('-', TokenType.I_SUB);
- let intMul = unaryOp('*', TokenType.I_MUL);
- let intDiv = unaryOp('/', TokenType.I_DIV);
- let lParen = unaryOp('(', TokenType.L_PAREN);
- let rParen = unaryOp(')', TokenType.R_PAREN);
- let lBracket = unaryOp('[', TokenType.L_BRACK);
- let rBracket = unaryOp(']', TokenType.R_BRACK);
- let lBrace = unaryOp('{', TokenType.L_BRACE);
- let rBrace = unaryOp('}', TokenType.R_BRACE);
- let comma = unaryOp(',', TokenType.COMMA);
- let dot = unaryOp('.', TokenType.DOT);
- let colon = unaryOp(':', TokenType.COLON);
- let semicolon = unaryOp(';', TokenType.SEMI_C);
- let at = unaryOp('@', TokenType.AT);
- let hash = unaryOp('#', TokenType.HASH);
- let set = unaryOp('=', TokenType.SET);
- let greaterthan = unaryOp('>', TokenType.GT);
- let lessthan = unaryOp('<', TokenType.LE);
- let term = (token_list, x) => {
- var ln = 1;
- var col = 0;
- var old_x = x;
- let term_list = [float, newline, space, integer, str, id,
- floatAdd, floatSub, floatMul, floatDiv,
- intAdd, intSub, intMul, intDiv,
- eq, ge, le, rightArrow,
- lParen, rParen, lBracket, rBracket, lBrace, rBrace,
- comma, dot, colon, semicolon, at, hash,
- set, greaterthan, lessthan];
- let term_aux = term_list.reduce((x, y) => orDo(x, y));
- var new_x = thenDo(old_x, term_aux);
- while (new_x._tag != "None") {
- if (new_x.value.matched_type != TokenType.NL) {
- col += new_x.value.matched.length;
- token_list.push({ text: new_x.value.matched,
- type: new_x.value.matched_type,
- ln: ln,
- col: col });