X-Git-Url: https://git.kianting.info/?a=blobdiff_plain;f=src%2Findex.js;h=411e166d95894b49998c1f4b59f51f567f5a11d5;hb=aefa3e14791ba05b7959572338ad2377f528ab0b;hp=a1ef267b5eb062cd6765a25f3be82e86485efecd;hpb=a722e739575012ebad8cd0a7d2c12df478a54eb8;p=clo

diff --git a/src/index.js b/src/index.js
index a1ef267..411e166 100644
--- a/src/index.js
+++ b/src/index.js
@@ -1,6 +1,6 @@
 "use strict";
 Object.defineProperty(exports, "__esModule", { value: true });
-exports.tokenize = exports.zeroOrOnceDo = exports.notDo = exports.zeroOrMoreDo = exports.orDo = exports.thenDo = exports.charToCodepoint = exports.matchRange = exports.matchAny = exports.match1Char = void 0;
+exports.tokenize = exports.zeroOrOnceDo = exports.notDo = exports.zeroOrMoreDo = exports.orDo = exports.thenDo = exports.charToCodepoint = exports.matchRange = exports.matchAny = exports.match1Char = exports.TokenType = void 0;
 var fs = require('fs');
 /**
  * wrap a x in a `Some(T)`
@@ -10,6 +10,64 @@ var fs = require('fs');
 function toSome(x) {
     return { _tag: "Some", value: x };
 }
+/**
+ * The types of Token
+ *    NL, // newline
+ *
+ *   SP, // half-width space and tab
+ *
+ * ID, // identifier
+ *
+ * STR, // string
+ *
+ * OP, // operator or something like it
+ *
+ * FLO, // float num
+ *
+ * INT, // integer
+ *
+ * I_* // integer manipulation
+ *
+ * F_* // float manipulation
+ *
+ * SEMI_C// semi-colon
+ */
+var TokenType;
+(function (TokenType) {
+    TokenType[TokenType["NL"] = 0] = "NL";
+    TokenType[TokenType["SP"] = 1] = "SP";
+    TokenType[TokenType["ID"] = 2] = "ID";
+    TokenType[TokenType["STR"] = 3] = "STR";
+    TokenType[TokenType["FLO"] = 4] = "FLO";
+    TokenType[TokenType["INT"] = 5] = "INT";
+    TokenType[TokenType["F_ADD"] = 6] = "F_ADD";
+    TokenType[TokenType["F_SUB"] = 7] = "F_SUB";
+    TokenType[TokenType["F_MUL"] = 8] = "F_MUL";
+    TokenType[TokenType["F_DIV"] = 9] = "F_DIV";
+    TokenType[TokenType["I_ADD"] = 10] = "I_ADD";
+    TokenType[TokenType["I_SUB"] = 11] = "I_SUB";
+    TokenType[TokenType["I_MUL"] = 12] = "I_MUL";
+    TokenType[TokenType["I_DIV"] = 13] = "I_DIV";
+    TokenType[TokenType["L_PAREN"] = 14] = "L_PAREN";
+    TokenType[TokenType["R_PAREN"] = 15] = "R_PAREN";
+    TokenType[TokenType["L_BRACK"] = 16] = "L_BRACK";
+    TokenType[TokenType["R_BRACK"] = 17] = "R_BRACK";
+    TokenType[TokenType["L_BRACE"] = 18] = "L_BRACE";
+    TokenType[TokenType["R_BRACE"] = 19] = "R_BRACE";
+    TokenType[TokenType["COMMA"] = 20] = "COMMA";
+    TokenType[TokenType["DOT"] = 21] = "DOT";
+    TokenType[TokenType["COLON"] = 22] = "COLON";
+    TokenType[TokenType["SEMI_C"] = 23] = "SEMI_C";
+    TokenType[TokenType["AT"] = 24] = "AT";
+    TokenType[TokenType["HASH"] = 25] = "HASH";
+    TokenType[TokenType["EQ"] = 26] = "EQ";
+    TokenType[TokenType["SET"] = 27] = "SET";
+    TokenType[TokenType["GT"] = 28] = "GT";
+    TokenType[TokenType["LT"] = 29] = "LT";
+    TokenType[TokenType["GE"] = 30] = "GE";
+    TokenType[TokenType["LE"] = 31] = "LE";
+    TokenType[TokenType["R_ARROW"] = 32] = "R_ARROW";
+})(TokenType || (exports.TokenType = TokenType = {}));
 /**
  * @description
  * it returns a function which test if the first char of the `remained` part of
@@ -211,19 +269,136 @@ exports.zeroOrOnceDo = zeroOrOnceDo;
 function tokenize(input) {
     var input_matchee_pair = toSome({ matched: "",
         remained: input });
-    // integer = ([+]|[-])\d\d?
-    let integer = (x) => {
-        let wrapped_x = toSome(x);
-        let plusMinus = orDo(match1Char('+'), match1Char('-')); // ([+]|[-])
-        let d = matchRange('0', '9'); // \d
-        return thenDo(thenDo(thenDo(wrapped_x, zeroOrOnceDo(plusMinus)), d), zeroOrMoreDo(d));
+    /**
+     * 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 });
+            }
+            else {
+                col = 0;
+                ln += 1;
+                token_list.push({ text: new_x.value.matched,
+                    type: new_x.value.matched_type,
+                    ln: ln,
+                    col: col });
+            }
+            old_x = toSome({ matched: "",
+                remained: new_x.value.remained });
+            new_x = thenDo(old_x, term_aux);
+        }
+        if (old_x.value.remained.length) {
+            console.log(token_list);
+            throw new Error("the code can't be tokenized is near Ln. " + ln + ", Col." + col
+                + ", starting with " + old_x.value.remained.substring(0, 10));
+        }
+        return token_list;
     };
-    console.log(input + ", result: ");
-    console.log(thenDo(input_matchee_pair, integer));
+    console.log(term([], input_matchee_pair));
+    // TODO: id, string, space, basic operator, 3 marks: @, {, }.
 }
 exports.tokenize = tokenize;
-tokenize("+123");
-tokenize("123");
-tokenize("-123");
-tokenize(" 123");
-tokenize("c123");