temp snapshot

src/index.d.ts

@@ -0,0 +1 @@
+export {};

src/index.js

@@ -1,404 +1,61 @@
 "use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
+var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
-exports.tokenize = exports.zeroOrOnceDo = exports.notDo = exports.zeroOrMoreDo = exports.orDo = exports.thenDo = exports.charToCodepoint = exports.matchRange = exports.matchAny = exports.match1Char = exports.TokenType = void 0;
+    if (k2 === undefined) k2 = k;
-var fs = require('fs');
+    var desc = Object.getOwnPropertyDescriptor(m, k);
-/**
+    if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
- * wrap a x in a `Some(T)`
+      desc = { enumerable: true, get: function() { return m[k]; } };
- * @param x : variable to be wrapped.
- * @returns wrapped `x`.
- */
-function toSome(x) {
-    return { _tag: "Some", value: x };
     }
-/**
+    Object.defineProperty(o, k2, desc);
- * The types of Token
+}) : (function(o, m, k, k2) {
- *    NL, // newline
+    if (k2 === undefined) k2 = k;
- *
+    o[k2] = m[k];
- *   SP, // half-width space and tab
+}));
- *
+var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
- * ID, // identifier
+    Object.defineProperty(o, "default", { enumerable: true, value: v });
- *
+}) : function(o, v) {
- * STR, // string
+    o["default"] = v;
- *
+});
- * OP, // operator or something like it
+var __importStar = (this && this.__importStar) || function (mod) {
- *
+    if (mod && mod.__esModule) return mod;
- * FLO, // float num
+    var result = {};
- *
+    if (mod != null) for (var k in mod) if (k !== "default" && Object.prototype.hasOwnProperty.call(mod, k)) __createBinding(result, mod, k);
- * INT, // integer
+    __setModuleDefault(result, mod);
- *
+    return result;
- * I_* // integer manipulation
+};
- *
+Object.defineProperty(exports, "__esModule", { value: true });
- * F_* // float manipulation
+exports.match1token = void 0;
- *
+var fs = require('fs');
- * SEMI_C// semi-colon
+const tk = __importStar(require("./tokenize.js"));
- */
+let b = tk.tokenize("2+2");
-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
  *  the argument of the function is `c`, if it's true, update the `MatchedPair` wrapped
  * in `Some`. Otherwise, it returns `None`.
- *  * @param c : the char to be test.
+ *  * @param t : the char to be test.
  * @returns the updated `MatchedPair` wrapped in `Some(x)` or `None`.
  */
-function match1Char(c) {
+function match1token(t) {
     return (m) => {
         if (m.remained.length == 0) {
             return { _tag: "None" };
         }
-        const charToBeMatched = m.remained[0];
+        const tokenToBeMatched = m.remained[0];
-        if (charToBeMatched === c) {
+        if (tokenToBeMatched === t) {
-            return { _tag: "Some", value: {
+            m.matched.push(tokenToBeMatched);
-                    matched: m.matched + charToBeMatched,
+            return {
-                    remained: m.remained.substring(1)
+                _tag: "Some", value: {
-                } };
+                    matched: m.matched,
+                    remained: m.remained.slice(1)
+                }
+            };
         }
         else {
             return { _tag: "None" };
         }
     };
 }
-exports.match1Char = match1Char;
+exports.match1token = match1token;
 ;
-/**
+let c = tk.toSome(b);
- *
+console.log(thenDo(c, match1token(tk.tokenize("+")[0])));
- * @param m : the `MatcheePair` to be consumed.
- * @returns if the length of `m.remained` >= 1; consumes the matchee by 1 char and wraps it in `Some`,
- * otherwise, returns `None`.
- */
-function matchAny(m) {
-    if (m.remained.length >= 1) {
-        return { _tag: "Some", value: {
-                matched: m.matched + m.remained[0],
-                remained: m.remained.substring(1)
-            } };
-    }
-    else {
-        return { _tag: "None" };
-    }
-}
-exports.matchAny = matchAny;
-/**
- * @description
- * it returns a function which test if the first char of the `remained` part of
- *  the argument of the function is between `l` and `u`, if it's true, update the `MatchedPair` wrapped
- * in `Some`. Otherwise, it returns `None`.
- *  * @param l : lower bound char, 1-char string
- *  * @param u : upper bound char, 1-char string
- * @returns the updated `MatchedPair` wrapped in `Some(x)` or `None`.
- */
-function matchRange(l, u) {
-    let lCodepoint = charToCodepoint(l);
-    let uCodepoint = charToCodepoint(u);
-    if (l > u) {
-        throw new Error("Error: the codepoint of `" + l + "` is not smaller than `" + u + "`)");
-    }
-    return (m) => {
-        if (m.remained.length < 1) {
-            return { _tag: "None" };
-        }
-        const charToBeMatched = m.remained[0];
-        const codePointToBeMatched = charToCodepoint(charToBeMatched);
-        if (codePointToBeMatched >= lCodepoint && codePointToBeMatched <= uCodepoint) {
-            return { _tag: "Some", value: {
-                    matched: m.matched + charToBeMatched,
-                    remained: m.remained.substring(1)
-                } };
-        }
-        else {
-            return { _tag: "None" };
-        }
-    };
-}
-exports.matchRange = matchRange;
-;
-/**
- * convert the one-char string to codepoint.
- * @param s : the string to code point.
- * @returns if `s.length > 1` return error; otherwise, return the codepoint of `s`.
- */
-function charToCodepoint(s) {
-    if (s.length > 1) {
-        throw new Error("Error: the length of input string for " + s + "is " + s.length + `,
-        however, it should be 1.`);
-    }
-    else {
-        return s.charCodeAt(0);
-    }
-}
-exports.charToCodepoint = charToCodepoint;
-/**
- *  @description thendo(input, f, ...) like
- * a ==> f
- * @param input: the wrapped input.
- * @param f: the function to be applied.
- *
- * @returns:the applied wrapped result `MatcheePair`.
- */
-function thenDo(input, f) {
-    if (input._tag == "None") {
-        return input;
-    }
-    else {
-        let inner = input.value;
-        return f(inner);
-    }
-}
-exports.thenDo = thenDo;
-/**
- * @description "or", like the regex `( f1 | f2 )` .
- * It returns a function `f` of which the argument is`x`.
- * if `f1(x)` is None, then `f` returns `f2(x)`. Otherwise,
- * `F` returns `f1(x)`.
- * @param f1 : 1st function to be compared
- * @param f2 : 2nd function to be compared
- * @returns:the combined function
- */
-function orDo(f1, f2) {
-    return (x) => {
-        let f1x = (f1(x));
-        {
-            if (f1x._tag == "None") {
-                return f2(x);
-            }
-            else {
-                return f1x;
-            }
-        }
-    };
-}
-exports.orDo = orDo;
-/**
-* @description repeating matching function `f`
-* zero or more times, like the asterisk `*` in regex `f*` .
-* @param f : the function to be repeated 0+ times.
-* @returns:the combined function
-*/
-function zeroOrMoreDo(f) {
-    return (x) => {
-        var wrapped_old_x = { _tag: "Some", value: x };
-        var wrapped_new_x = wrapped_old_x;
-        while (wrapped_new_x._tag != "None") {
-            wrapped_old_x = wrapped_new_x;
-            wrapped_new_x = thenDo(wrapped_old_x, f);
-        }
-        ;
-        return wrapped_old_x;
-    };
-}
-exports.zeroOrMoreDo = zeroOrMoreDo;
-/**
-* @description Not. like the `^` inside regex of [^f].
-* returns a function `F(x)` such that if  `f(x)` is `None`,
-* returns the x consuming a char; if `f(x)` is not None, F(x)
-* returns `None`.
-* @param f: the function forbidden to be matched.
-* @returns: combined function `F`.
-*/
-function notDo(f) {
-    return (x) => {
-        let wrapped_x = {
-            _tag: "Some",
-            value: x
-        };
-        let f_x = thenDo(wrapped_x, f);
-        if (f_x._tag != "None") {
-            return { _tag: "None" };
-        }
-        else {
-            return thenDo(wrapped_x, matchAny);
-        }
-    };
-}
-exports.notDo = notDo;
-/**
- * if `x` is matched by `f` once, returns `f(x)`. Otherwise,
- * returns x
- * similar to `?` in regex `f?`.
- * @param f : the function to be matched
- * @returns return wrapped f(x)
- */
-function zeroOrOnceDo(f) {
-    return (x) => {
-        var wrapped_old_x = { _tag: "Some", value: x };
-        var wrapped_new_x = thenDo(wrapped_old_x, f);
-        if (wrapped_new_x._tag != "None") {
-            return wrapped_new_x;
-        }
-        else {
-            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 });
-            }
-            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(term([], input_matchee_pair));
-    // TODO: id, string, space, basic operator, 3 marks: @, {, }.
-}
-exports.tokenize = tokenize;

src/index.ts

@@ -1,115 +1,14 @@
-import { match } from "assert";
-
 var fs = require('fs');
 
-export type Some<T> = { _tag: "Some"; value: T };
+import * as tk from './tokenize.js';
-export type None = {_tag: "None"};
 
 
-/**
- * wrap a x in a `Some(T)`
- * @param x : variable to be wrapped.
- * @returns wrapped `x`.
- */
-function toSome<T>(x: T): Some<T>{
-    return  { _tag: "Some", value: x};  
-}
-/**
- * @description Like the `Some(a)` and `None` in Rust.
- *
- * @example
- * ```ts
- * let exam1 : Maybe<Number> = { _tag: "Some", value: 12 };
- * let exam2 : Maybe<Number> = None;
- * ```
- */
-export type Maybe<T> = Some<T> | None;
 
+let b : Array<tk.Token> = tk.tokenize("2+2");
 
-/**
+export interface TokenMatcheePair {
- * @description
+    matched: tk.Token[]
- * the pair of the string to be matched later and the string that have been matched
+    remained: tk.Token[]
- * @var matched : have been matched
- * @var remained : will be tested whether it'll be matched.
- * @var matched_type (optional): the type of the matched string
-*/
-export interface MatcheePair {
-    matched : string
-    remained : string
-    matched_type?: TokenType 
-}
-
-/**
- * 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
- */
-export enum TokenType{
-    NL, // newlinw
-    SP, // half-width space and tab
-    ID, // identifier
-    STR, // string
-    FLO, // float num
-    INT, // integer
-    F_ADD,
-    F_SUB,
-    F_MUL,
-    F_DIV,
-    I_ADD, 
-    I_SUB, 
-    I_MUL, 
-    I_DIV,
-    L_PAREN, // (
-    R_PAREN, // )
-    L_BRACK, // [
-    R_BRACK, // ]
-    L_BRACE, // {
-    R_BRACE, // }
-    COMMA, // ,
-    DOT, // .
-    COLON, // :
-    SEMI_C, // ;
-    AT, // @
-    HASH, // #
-    EQ, // ==
-    SET, // =
-    GT, // > greater than
-    LT, // <less than
-    GE, // >=
-    LE, // <=
-    R_ARROW, // ->
-
-}
-
-/**
- * tokenized token.
- * @var text : the content text
- * @var type (optional): the type of the token
- * @var col : the column number
- * @var ln : the line number
- */
-export interface Token{
-    text: string,
-    type?: TokenType,
-    col: number,
-    ln: number,
 }
 
 /**
@@ -117,19 +16,23 @@ export interface Token{
  * it returns a function which test if the first char of the `remained` part of
  *  the argument of the function is `c`, if it's true, update the `MatchedPair` wrapped
  * in `Some`. Otherwise, it returns `None`.
- *  * @param c : the char to be test.
+ *  * @param t : the char to be test.
  * @returns the updated `MatchedPair` wrapped in `Some(x)` or `None`.
  */
-export function match1Char(c : string) : (m: MatcheePair) => Maybe<MatcheePair> {
+export function match1token(t: tk.Token): (m: TokenMatcheePair) => tk.Maybe<TokenMatcheePair> {
-    return (m : MatcheePair)=>{
+    return (m: TokenMatcheePair) => {
         if (m.remained.length == 0) {
             return { _tag: "None" };
         }
-        const charToBeMatched = m.remained[0];
+        const tokenToBeMatched = m.remained[0];
-        if (charToBeMatched === c){
+        if (tokenToBeMatched === t) {
-            return {_tag: "Some", value :{
+            m.matched.push(tokenToBeMatched);
-                    matched : m.matched + charToBeMatched,
+            return {
-                    remained : m.remained.substring(1)}};
+                _tag: "Some", value: {
+                    matched: m.matched,
+                    remained: m.remained.slice(1)
+                }
+            };
         }
         else {
             return { _tag: "None" };
@@ -137,376 +40,7 @@ export function match1Char(c : string) : (m: MatcheePair) => Maybe<MatcheePair>
     }
 };
 
-/**
- * 
- * @param m : the `MatcheePair` to be consumed.
- * @returns if the length of `m.remained` >= 1; consumes the matchee by 1 char and wraps it in `Some`,
- * otherwise, returns `None`.
- */
-export function matchAny(m : MatcheePair) : Maybe<MatcheePair>{
-    if (m.remained.length >= 1){
-        return {_tag: "Some", value :{
-            matched : m.matched + m.remained[0],
-            remained : m.remained.substring(1)}};
-    }else{
-        return {_tag: "None"};
-    }
-}
-
-/**
- * @description
- * it returns a function which test if the first char of the `remained` part of
- *  the argument of the function is between `l` and `u`, if it's true, update the `MatchedPair` wrapped
- * in `Some`. Otherwise, it returns `None`.
- *  * @param l : lower bound char, 1-char string
- *  * @param u : upper bound char, 1-char string
- * @returns the updated `MatchedPair` wrapped in `Some(x)` or `None`.
- */
-export function matchRange(l : string, u : string) : (m: MatcheePair) => Maybe<MatcheePair> {
-    let lCodepoint = charToCodepoint(l);
-    let uCodepoint = charToCodepoint(u);
-    if (l > u){
-        throw new Error("Error: the codepoint of `"+l+"` is not smaller than `"+u+"`)");
-    }
-    return (m : MatcheePair)=>{
-        if (m.remained.length < 1){
-            return {_tag : "None"};
-        }
-        const charToBeMatched = m.remained[0];
-        const codePointToBeMatched = charToCodepoint(charToBeMatched);
-        if (codePointToBeMatched >= lCodepoint && codePointToBeMatched <= uCodepoint){
-            return {_tag: "Some", value :{
-                    matched : m.matched + charToBeMatched,
-                    remained : m.remained.substring(1)}};
-        }
-        else{
-            return {_tag: "None"};
-        }
-    }
-};
-
-/**
- * convert the one-char string to codepoint.
- * @param s : the string to code point.
- * @returns if `s.length > 1` return error; otherwise, return the codepoint of `s`.
- */
-export function charToCodepoint(s : string): number{
-    if (s.length > 1){
-        throw new Error("Error: the length of input string for "+s+ "is "+s.length+`,
-        however, it should be 1.`);
-    }else{
-        return s.charCodeAt(0);
-    }
-}
-
-/**
- *  @description thendo(input, f, ...) like
- * a ==> f
- * @param input: the wrapped input. 
- * @param f: the function to be applied.
- * 
- * @returns:the applied wrapped result `MatcheePair`.
- */
-export function thenDo<T>(input : Maybe<T>, f : Function) : Maybe<T>{
-    if (input._tag == "None"){
-        return input;
-    }
-    else{
-        let inner = input.value;
-        return f(inner);
-    }
-}
-
-/**
- * @description "or", like the regex `( f1 | f2 )` . 
- * It returns a function `f` of which the argument is`x`.
- * if `f1(x)` is None, then `f` returns `f2(x)`. Otherwise, 
- * `F` returns `f1(x)`.
- * @param f1 : 1st function to be compared
- * @param f2 : 2nd function to be compared
- * @returns:the combined function
- */
-export function orDo<T>(f1 : Function, f2: Function) : (x : T ) => Maybe<T>{
-    return (x) => {
-        let f1x : Maybe<T> = (f1(x));
-        {
-            if (f1x._tag == "None"){
-                return f2(x);
-            }
-            else{
-                return f1x;
-            }
-        }
-    };
-}
-
-
-/**
-* @description repeating matching function `f` 
-* zero or more times, like the asterisk `*` in regex `f*` . 
-* @param f : the function to be repeated 0+ times.
-* @returns:the combined function
-*/
-export function zeroOrMoreDo<T>(f : Function): (x : T) => Maybe<T>{
-    return (x)=>{
-        var wrapped_old_x : Maybe<T> = {_tag: "Some", value : x};
-        var wrapped_new_x : Maybe<T> = wrapped_old_x;
-
-        while (wrapped_new_x._tag != "None"){
-            wrapped_old_x = wrapped_new_x;
-            wrapped_new_x = thenDo(wrapped_old_x, f);
-        };
-
-        return wrapped_old_x;
-    };
-}
-
-/**
-* @description Not. like the `^` inside regex of [^f].
-* returns a function `F(x)` such that if  `f(x)` is `None`,
-* returns the x consuming a char; if `f(x)` is not None, F(x)
-* returns `None`.
-* @param f: the function forbidden to be matched.
-* @returns: combined function `F`.
-*/
-export function notDo<T>(f : Function): (x : T) => Maybe<T>{
-    return (x)=>{
-        let wrapped_x : Maybe<T> = {
-            _tag : "Some",
-            value : x
-        };
-        let f_x = thenDo(wrapped_x, f);
-        
-        if (f_x._tag != "None"){
-            return {_tag:"None"};
-        }else{
-            return thenDo(wrapped_x, matchAny);
-        }
-    };
-}
-
-/**
- * if `x` is matched by `f` once, returns `f(x)`. Otherwise, 
- * returns x
- * similar to `?` in regex `f?`.
- * @param f : the function to be matched
- * @returns return wrapped f(x)
- */
-export function zeroOrOnceDo<T>(f : Function): (x : T) => Maybe<T>{
-    return (x)=>{
-        var wrapped_old_x : Maybe<T> = {_tag: "Some", value : x};
-        var wrapped_new_x = thenDo(wrapped_old_x, f);
-
-        if (wrapped_new_x._tag != "None"){
-            return wrapped_new_x;
-        }else{
-        return wrapped_old_x;
-        }
-    };
-}
-
-
-export function tokenize(input : string){
-    var input_matchee_pair : Maybe<MatcheePair> = 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 : Function, token_type : TokenType){
-        return (x : MatcheePair) =>{
-            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 : Maybe<MatcheePair>)=>
-                            thenDo(thenDo(thenDo(x, 
-                            zeroOrOnceDo(plusMinus)),d),
-                            zeroOrMoreDo(d)),
-                        TokenType.INT);
-    // space = [ \t]+
-    let space = bTerm((x : Maybe<MatcheePair>)=>
-        thenDo(thenDo(x, s_aux), zeroOrMoreDo(s_aux)),
-        TokenType.INT);
-
-    // newline = \r?\n
-    let newline = bTerm((x : Maybe<MatcheePair>)=>
-    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 : Maybe<MatcheePair>)=>
-        thenDo(thenDo(x, 
-            idHead),
-            zeroOrMoreDo(idRemained)),
-            TokenType.ID);
-    let doublequote = match1Char("\"");
-    // [\\][\"]
-    let escapeReverseSlash = (x:MatcheePair)=>
-        thenDo(thenDo(toSome(x), match1Char("\\")), doublequote);
-    // ([\\]["]|[^\"])*
-    let stringInnerPattern = zeroOrMoreDo(
-        orDo(escapeReverseSlash, notDo(match1Char("\""))));
-
-    // str = ["]([\\]["]|[^"])*["]
-    let str = bTerm((x : Maybe<MatcheePair>)=>
-        thenDo(thenDo(thenDo(x,doublequote),
-        stringInnerPattern),doublequote),
-        TokenType.STR);
-
-    // float = [+-]?\d+[.]\d+
-    function floatPattern(x : Maybe<MatcheePair>){
-        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 : Maybe<MatcheePair>)=>
-        thenDo(thenDo(x, match1Char("+")),match1Char(".")),
-        TokenType.F_ADD);
-    // +.
-    let floatSub = bTerm((x : Maybe<MatcheePair>)=>
-        thenDo(thenDo(x, match1Char("-")),match1Char(".")),
-        TokenType.F_SUB);  
-
-    // *.
-    let floatMul = bTerm((x : Maybe<MatcheePair>)=>
-        thenDo(thenDo(x, match1Char("*")),match1Char(".")),
-        TokenType.F_MUL);
-
-    // /.
-    let floatDiv = bTerm((x : Maybe<MatcheePair>)=>
-        thenDo(thenDo(x, match1Char("/")),match1Char(".")),
-        TokenType.F_DIV);
-
-    // ==
-    let eq = bTerm((x : Maybe<MatcheePair>)=>
-    thenDo(thenDo(x, match1Char("=")),match1Char("=")),
-    TokenType.EQ);
-
-    // >=
-    let ge = bTerm((x : Maybe<MatcheePair>)=>
-    thenDo(thenDo(x, match1Char(">")),match1Char("=")),
-    TokenType.GE);
-
-    // <=
-    let le = bTerm((x : Maybe<MatcheePair>)=>
-    thenDo(thenDo(x, match1Char("<")),match1Char("=")),
-    TokenType.LE);
-
-    // ->
-    let rightArrow = bTerm((x : Maybe<MatcheePair>)=>
-    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: string, token_type: TokenType) {
-        return bTerm((x : Maybe<MatcheePair>)=>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 : Array<Token>, x :  Some<MatcheePair>)=>{
-        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 : Maybe<MatcheePair> = 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(term([], input_matchee_pair));
-
-    // TODO: id, string, space, basic operator, 3 marks: @, {, }.
-
-}
 
 
+let c = tk.toSome(b);
+console.log(thenDo(c,match1token(tk.tokenize("+")[0])));

src/tokenize.ts

@@ -0,0 +1,542 @@
+
+var fs = require('fs');
+
+export type Some<T> = { _tag: "Some"; value: T };
+export type None = { _tag: "None" };
+/**
+ * part for tokenize the input string
+ */
+
+/**
+ * wrap a x in a `Some(T)`
+ * @param x : variable to be wrapped.
+ * @returns wrapped `x`.
+ */
+export function toSome<T>(x: T): Some<T> {
+    return { _tag: "Some", value: x };
+}
+/**
+ * @description Like the `Some(a)` and `None` in Rust.
+ *
+ * @example
+ * ```ts
+ * let exam1 : Maybe<Number> = { _tag: "Some", value: 12 };
+ * let exam2 : Maybe<Number> = None;
+ * ```
+ */
+export type Maybe<T> = Some<T> | None;
+
+
+/**
+ * @description
+ * the pair of the string to be matched later and the string that have been matched
+ * @var matched : have been matched
+ * @var remained : will be tested whether it'll be matched.
+ * @var matched_type (optional): the type of the matched string
+*/
+export interface MatcheePair {
+    matched: string
+    remained: string
+    matched_type?: TokenType
+}
+
+/**
+ * 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
+ */
+export enum TokenType {
+    NL, // newlinw
+    SP, // half-width space and tab
+    ID, // identifier
+    STR, // string
+    FLO, // float num
+    INT, // integer
+    F_ADD,
+    F_SUB,
+    F_MUL,
+    F_DIV,
+    I_ADD,
+    I_SUB,
+    I_MUL,
+    I_DIV,
+    L_PAREN, // (
+    R_PAREN, // )
+    L_BRACK, // [
+    R_BRACK, // ]
+    L_BRACE, // {
+    R_BRACE, // }
+    COMMA, // ,
+    DOT, // .
+    COLON, // :
+    SEMI_C, // ;
+    AT, // @
+    HASH, // #
+    EQ, // ==
+    SET, // =
+    GT, // > greater than
+    LT, // <less than
+    GE, // >=
+    LE, // <=
+    NE, // <>
+    APOS, // '
+    R_ARROW, // ->
+
+}
+
+/**
+ * tokenized token.
+ * @var text : the content text
+ * @var type (optional): the type of the token
+ * @var col : the column number
+ * @var ln : the line number
+ */
+export interface Token {
+    text: string,
+    type?: TokenType,
+    col: number,
+    ln: number,
+}
+
+/**
+ * @description
+ * it returns a function which test if the first char of the `remained` part of
+ *  the argument of the function is `c`, if it's true, update the `MatchedPair` wrapped
+ * in `Some`. Otherwise, it returns `None`.
+ *  * @param c : the char to be test.
+ * @returns the updated `MatchedPair` wrapped in `Some(x)` or `None`.
+ */
+export function match1Char(c: string): (m: MatcheePair) => Maybe<MatcheePair> {
+    return (m: MatcheePair) => {
+        if (m.remained.length == 0) {
+            return { _tag: "None" };
+        }
+        const charToBeMatched = m.remained[0];
+        if (charToBeMatched === c) {
+            return {
+                _tag: "Some", value: {
+                    matched: m.matched + charToBeMatched,
+                    remained: m.remained.substring(1)
+                }
+            };
+        }
+        else {
+            return { _tag: "None" };
+        }
+    }
+};
+
+/**
+ * 
+ * @param m : the `MatcheePair` to be consumed.
+ * @returns if the length of `m.remained` >= 1; consumes the matchee by 1 char and wraps it in `Some`,
+ * otherwise, returns `None`.
+ */
+export function matchAny(m: MatcheePair): Maybe<MatcheePair> {
+    if (m.remained.length >= 1) {
+        return {
+            _tag: "Some", value: {
+                matched: m.matched + m.remained[0],
+                remained: m.remained.substring(1)
+            }
+        };
+    } else {
+        return { _tag: "None" };
+    }
+}
+
+/**
+ * @description
+ * it returns a function which test if the first char of the `remained` part of
+ *  the argument of the function is between `l` and `u`, if it's true, update the `MatchedPair` wrapped
+ * in `Some`. Otherwise, it returns `None`.
+ *  * @param l : lower bound char, 1-char string
+ *  * @param u : upper bound char, 1-char string
+ * @returns the updated `MatchedPair` wrapped in `Some(x)` or `None`.
+ */
+export function matchRange(l: string, u: string): (m: MatcheePair) => Maybe<MatcheePair> {
+    let lCodepoint = charToCodepoint(l);
+    let uCodepoint = charToCodepoint(u);
+    if (l > u) {
+        throw new Error("Error: the codepoint of `" + l + "` is not smaller than `" + u + "`)");
+    }
+    return (m: MatcheePair) => {
+        if (m.remained.length < 1) {
+            return { _tag: "None" };
+        }
+        const charToBeMatched = m.remained[0];
+        const codePointToBeMatched = charToCodepoint(charToBeMatched);
+        if (codePointToBeMatched >= lCodepoint && codePointToBeMatched <= uCodepoint) {
+            return {
+                _tag: "Some", value: {
+                    matched: m.matched + charToBeMatched,
+                    remained: m.remained.substring(1)
+                }
+            };
+        }
+        else {
+            return { _tag: "None" };
+        }
+    }
+};
+
+/**
+ * convert the one-char string to codepoint.
+ * @param s : the string to code point.
+ * @returns if `s.length > 1` return error; otherwise, return the codepoint of `s`.
+ */
+export function charToCodepoint(s: string): number {
+    if (s.length > 1) {
+        throw new Error("Error: the length of input string for " + s + "is " + s.length + `,
+        however, it should be 1.`);
+    } else {
+        return s.charCodeAt(0);
+    }
+}
+
+/**
+ *  @description thendo(input, f, ...) like
+ * a ==> f
+ * @param input: the wrapped input. 
+ * @param f: the function to be applied.
+ * 
+ * @returns:the applied wrapped result `MatcheePair`.
+ */
+export function thenDo<T>(input: Maybe<T>, f: Function): Maybe<T> {
+    if (input._tag == "None") {
+        return input;
+    }
+    else {
+        let inner = input.value;
+        return f(inner);
+    }
+}
+
+/**
+ * @description "or", like the regex `( f1 | f2 )` . 
+ * It returns a function `f` of which the argument is`x`.
+ * if `f1(x)` is None, then `f` returns `f2(x)`. Otherwise, 
+ * `F` returns `f1(x)`.
+ * @param f1 : 1st function to be compared
+ * @param f2 : 2nd function to be compared
+ * @returns:the combined function
+ */
+export function orDo<T>(f1: Function, f2: Function): (x: T) => Maybe<T> {
+    return (x) => {
+        let f1x: Maybe<T> = (f1(x));
+        {
+            if (f1x._tag == "None") {
+                return f2(x);
+            }
+            else {
+                return f1x;
+            }
+        }
+    };
+}
+
+
+/**
+* @description repeating matching function `f` 
+* zero or more times, like the asterisk `*` in regex `f*` . 
+* @param f : the function to be repeated 0+ times.
+* @returns:the combined function
+*/
+export function zeroOrMoreDo<T>(f: Function): (x: T) => Maybe<T> {
+    return (x) => {
+        var wrapped_old_x: Maybe<T> = { _tag: "Some", value: x };
+        var wrapped_new_x: Maybe<T> = wrapped_old_x;
+
+        while (wrapped_new_x._tag != "None") {
+            wrapped_old_x = wrapped_new_x;
+            wrapped_new_x = thenDo(wrapped_old_x, f);
+        };
+
+        return wrapped_old_x;
+    };
+}
+
+/**
+* @description Not. like the `^` inside regex of [^f].
+* returns a function `F(x)` such that if  `f(x)` is `None`,
+* returns the x consuming a char; if `f(x)` is not None, F(x)
+* returns `None`.
+* @param f: the function forbidden to be matched.
+* @returns: combined function `F`.
+*/
+export function notDo<T>(f: Function): (x: T) => Maybe<T> {
+    return (x) => {
+        let wrapped_x: Maybe<T> = {
+            _tag: "Some",
+            value: x
+        };
+        let f_x = thenDo(wrapped_x, f);
+
+        if (f_x._tag != "None") {
+            return { _tag: "None" };
+        } else {
+            return thenDo(wrapped_x, matchAny);
+        }
+    };
+}
+
+/**
+ * if `x` is matched by `f` once, returns `f(x)`. Otherwise, 
+ * returns x
+ * similar to `?` in regex `f?`.
+ * @param f : the function to be matched
+ * @returns return wrapped f(x)
+ */
+export function zeroOrOnceDo<T>(f: Function): (x: T) => Maybe<T> {
+    return (x) => {
+        var wrapped_old_x: Maybe<T> = { _tag: "Some", value: x };
+        var wrapped_new_x = thenDo(wrapped_old_x, f);
+
+        if (wrapped_new_x._tag != "None") {
+            return wrapped_new_x;
+        } else {
+            return wrapped_old_x;
+        }
+    };
+}
+
+
+export function tokenize(input: string): Array<Token> {
+    var input_matchee_pair: Maybe<MatcheePair> = 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: Function, token_type: TokenType) {
+        return (x: MatcheePair) => {
+            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: Maybe<MatcheePair>) =>
+        thenDo(thenDo(thenDo(x,
+            zeroOrOnceDo(plusMinus)), d),
+            zeroOrMoreDo(d)),
+        TokenType.INT);
+    // space = [ \t]+
+    let space = bTerm((x: Maybe<MatcheePair>) =>
+        thenDo(thenDo(x, s_aux), zeroOrMoreDo(s_aux)),
+        TokenType.INT);
+
+    // newline = \r?\n
+    let newline = bTerm((x: Maybe<MatcheePair>) =>
+        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: Maybe<MatcheePair>) =>
+        thenDo(thenDo(x,
+            idHead),
+            zeroOrMoreDo(idRemained)),
+        TokenType.ID);
+    let doublequote = match1Char("\"");
+    // [\\][\"]
+    let escapeReverseSlash = (x: MatcheePair) =>
+        thenDo(thenDo(toSome(x), match1Char("\\")), doublequote);
+    // ([\\]["]|[^\"])*
+    let stringInnerPattern = zeroOrMoreDo(
+        orDo(escapeReverseSlash, notDo(match1Char("\""))));
+
+    // str = ["]([\\]["]|[^"])*["]
+    let str = bTerm((x: Maybe<MatcheePair>) =>
+        thenDo(thenDo(thenDo(x, doublequote),
+            stringInnerPattern), doublequote),
+        TokenType.STR);
+
+    // float = [+-]?\d+[.]\d+
+    function floatPattern(x: Maybe<MatcheePair>) {
+        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: Maybe<MatcheePair>) =>
+        thenDo(thenDo(x, match1Char("+")), match1Char(".")),
+        TokenType.F_ADD);
+    // +.
+    let floatSub = bTerm((x: Maybe<MatcheePair>) =>
+        thenDo(thenDo(x, match1Char("-")), match1Char(".")),
+        TokenType.F_SUB);
+
+    // *.
+    let floatMul = bTerm((x: Maybe<MatcheePair>) =>
+        thenDo(thenDo(x, match1Char("*")), match1Char(".")),
+        TokenType.F_MUL);
+
+    // /.
+    let floatDiv = bTerm((x: Maybe<MatcheePair>) =>
+        thenDo(thenDo(x, match1Char("/")), match1Char(".")),
+        TokenType.F_DIV);
+
+    // ==
+    let eq = bTerm((x: Maybe<MatcheePair>) =>
+        thenDo(thenDo(x, match1Char("=")), match1Char("=")),
+        TokenType.EQ);
+
+    // >=
+    let ge = bTerm((x: Maybe<MatcheePair>) =>
+        thenDo(thenDo(x, match1Char(">")), match1Char("=")),
+        TokenType.GE);
+
+    // <=
+    let le = bTerm((x: Maybe<MatcheePair>) =>
+        thenDo(thenDo(x, match1Char("<")), match1Char("=")),
+        TokenType.LE);
+
+    // !=
+    let ne = bTerm((x: Maybe<MatcheePair>) =>
+        thenDo(thenDo(x, match1Char("!")), match1Char("=")),
+        TokenType.NE);
+
+    // ->
+    let rightArrow = bTerm((x: Maybe<MatcheePair>) =>
+        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: string, token_type: TokenType) {
+        return bTerm((x: Maybe<MatcheePair>) => 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 apos = unaryOp('\'', TokenType.APOS);
+
+
+
+    let term = (token_list: Array<Token>, x: Some<MatcheePair>) => {
+        var ln = 1;
+        var col = 0;
+        var old_x = x;
+        let term_list = [
+            floatAdd, floatSub, floatMul, floatDiv,
+            intAdd, intSub, intMul, intDiv,
+            eq, ge, le, ne, rightArrow,
+            lParen, rParen, lBracket, rBracket, lBrace, rBrace,
+            comma, dot, colon, semicolon, at, hash,
+            set, greaterthan, lessthan, apos,
+            float, newline, space, integer, str, id];
+        let term_aux = term_list.reduce((x, y) => orDo(x, y));
+
+        var new_x: Maybe<MatcheePair> = 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;
+    }
+
+    return term([], input_matchee_pair);
+
+    // TODO: id, string, space, basic operator, 3 marks: @, {, }.
+
+}
+
+

tests/index.d.ts

@@ -0,0 +1,86 @@
+declare let assert: any;
+declare let tokenize: any;
+declare let a: any;
+declare let example1: any;
+declare let example2: any;
+declare let thenDo: any;
+declare let orDo: any;
+declare let zeroOrMoreDo: any;
+declare let notDo: any;
+declare let matchAny: any;
+declare let compPart1: any;
+declare let compPart2: any;
+declare let doThenTestee1: {
+    _tag: string;
+    value: {
+        matched: string;
+        remained: string;
+    };
+};
+declare let doTestRes1: any;
+declare let doThenTestee2: {
+    _tag: string;
+    value: {
+        matched: string;
+        remained: string;
+    };
+};
+declare let doTestRes2: any;
+declare let doThenTestee3: {
+    _tag: string;
+    value: {
+        matched: string;
+        remained: string;
+    };
+};
+declare let doTestRes3: any;
+declare let doThenTestee4: {
+    _tag: string;
+    value: {
+        matched: string;
+        remained: string;
+    };
+};
+declare let doTestRes4: any;
+declare let doThenTestee5: {
+    _tag: string;
+    value: {
+        matched: string;
+        remained: string;
+    };
+};
+declare let doTestRes5: any;
+declare let doThenTestee6: {
+    _tag: string;
+    value: {
+        matched: string;
+        remained: string;
+    };
+};
+declare let doTestRes6: any;
+declare let doThenTestee7: {
+    _tag: string;
+    value: {
+        matched: string;
+        remained: string;
+    };
+};
+declare let doTestRes7: any;
+declare let doThenTestee8: {
+    _tag: string;
+    value: {
+        matched: string;
+        remained: string;
+    };
+};
+declare let doTestRes8: any;
+declare let doThenTestee9: {
+    _tag: string;
+    value: {
+        matched: string;
+        remained: string;
+    };
+};
+declare let doTestRes9: any;
+declare let harfbuzz: any;
+declare let pdfManipulate: any;

tests/index.js

@@ -1,22 +1,21 @@
 "use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-const src_1 = require("../src");
 let assert = require("assert");
-let cloMain = require("../src");
+let tokenize = require("../src/tokenize");
-let a = cloMain.match1Char("我");
+let a = tokenize.match1Char("我");
 let example1 = a({ matched: "", remained: "我的" });
 assert(example1._tag == "Some");
 assert(example1.value.matched == "我");
 assert(example1.value.remained == "的");
 let example2 = a({ matched: "", remained: "妳的" });
 assert(example2._tag == "None");
-let thenDo = cloMain.thenDo;
+let thenDo = tokenize.thenDo;
-let orDo = cloMain.orDo;
+let orDo = tokenize.orDo;
-let zeroOrMoreDo = cloMain.zeroOrMoreDo;
+let zeroOrMoreDo = tokenize.zeroOrMoreDo;
-let notDo = cloMain.notDo;
+let notDo = tokenize.notDo;
+let matchAny = tokenize.matchAny;
 // composed part x
-let compPart1 = cloMain.match1Char("我");
+let compPart1 = tokenize.match1Char("我");
-let compPart2 = cloMain.match1Char("的");
+let compPart2 = tokenize.match1Char("的");
 let doThenTestee1 = { _tag: "Some", value: { matched: "", remained: "我的貓" } };
 let doTestRes1 = thenDo(thenDo(doThenTestee1, compPart1), compPart2);
 assert(doTestRes1._tag == "Some");
@@ -53,39 +52,40 @@ let doTestRes8 = thenDo(thenDo(doThenTestee8, notDo(compPart1)), compPart2);
 assert(doTestRes8._tag == "Some");
 assert(doTestRes8.value.matched == "妳的");
 let doThenTestee9 = { _tag: "Some", value: { matched: "", remained: "妳的" } };
-let doTestRes9 = thenDo(doThenTestee9, src_1.matchAny);
+let doTestRes9 = thenDo(doThenTestee9, matchAny);
 assert(doTestRes9._tag == "Some");
 assert(doTestRes9.value.matched == "妳");
 assert(doTestRes9.value.remained == "的");
-(0, src_1.tokenize)("+123");
+tokenize.tokenize("+123");
-(0, src_1.tokenize)("123");
+tokenize.tokenize("123");
-(0, src_1.tokenize)("-123");
+tokenize.tokenize("-123");
-(0, src_1.tokenize)(" 123");
+tokenize.tokenize(" 123");
 try {
-    (0, src_1.tokenize)("c123");
+    tokenize.tokenize("c123");
 }
 catch (error) {
     console.log(error);
 }
-(0, src_1.tokenize)("  ");
+tokenize.tokenize("  ");
-(0, src_1.tokenize)(" ");
+tokenize.tokenize(" ");
-(0, src_1.tokenize)(" \t");
+tokenize.tokenize(" \t");
-(0, src_1.tokenize)(" \t123");
+tokenize.tokenize(" \t123");
 try {
-    (0, src_1.tokenize)(" \t123aaa456");
+    tokenize.tokenize(" \t123aaa456");
 }
 catch (error) {
     console.log(error);
 }
-(0, src_1.tokenize)(" \t123\n456");
+tokenize.tokenize(" \t123\n456");
-(0, src_1.tokenize)("\"\"");
+tokenize.tokenize("\"\"");
-(0, src_1.tokenize)("\"123\"");
+tokenize.tokenize("\"123\"");
-(0, src_1.tokenize)("\"1\\\"23\"");
+tokenize.tokenize("\"1\\\"23\"");
-(0, src_1.tokenize)("\"1\\\"23\"  abc123");
+tokenize.tokenize("\"1\\\"23\"  abc123");
-(0, src_1.tokenize)("+0.012");
+tokenize.tokenize("+0.012");
-(0, src_1.tokenize)("0.0");
+tokenize.tokenize("0.0");
-(0, src_1.tokenize)("-222.0");
+tokenize.tokenize("-222.0");
-(0, src_1.tokenize)("1+1 ==2; 3+8 foo(12)");
+tokenize.tokenize("1+1 ==2; 3+8 foo(12)");
+console.log(tokenize.tokenize("2+2"));
 // harfbuzz test
 let harfbuzz = require("../src/harfbuzz.js");
 harfbuzz.harfbuzzTest("123.abc");

tests/index.ts

@@ -1,9 +1,8 @@
-import { matchAny, tokenize } from "../src";
 
 let assert = require("assert");
-let cloMain = require("../src");
+let tokenize = require("../src/tokenize");
 
-let a = cloMain.match1Char("我");
+let a = tokenize.match1Char("我");
 
 let example1 = a({matched: "", remained: "我的"});
 assert(example1._tag == "Some");
@@ -15,14 +14,15 @@ assert(example2._tag == "None");
 
 
 
-let thenDo = cloMain.thenDo;
+let thenDo = tokenize.thenDo;
-let orDo = cloMain.orDo;
+let orDo = tokenize.orDo;
-let zeroOrMoreDo = cloMain.zeroOrMoreDo;
+let zeroOrMoreDo = tokenize.zeroOrMoreDo;
-let notDo = cloMain.notDo;
+let notDo = tokenize.notDo;
+let matchAny = tokenize.matchAny;
 
 // composed part x
-let compPart1 = cloMain.match1Char("我");
+let compPart1 = tokenize.match1Char("我");
-let compPart2 = cloMain.match1Char("的");
+let compPart2 = tokenize.match1Char("的");
 
 let doThenTestee1 = {_tag : "Some",value : {matched: "", remained: "我的貓"}};
 let doTestRes1 = thenDo(thenDo(doThenTestee1, compPart1), compPart2);
@@ -74,40 +74,42 @@ assert(doTestRes9._tag == "Some");
 assert(doTestRes9.value.matched == "妳");
 assert(doTestRes9.value.remained == "的");
 
-tokenize("+123");
+tokenize.tokenize("+123");
-tokenize("123");
+tokenize.tokenize("123");
-tokenize("-123");
+tokenize.tokenize("-123");
-tokenize(" 123");
+tokenize.tokenize(" 123");
 try {
-    tokenize("c123");
+    tokenize.tokenize("c123");
 
 } catch (error) {
     console.log(error);
 }
 
-tokenize("  ");
+tokenize.tokenize("  ");
-tokenize(" ");
+tokenize.tokenize(" ");
-tokenize(" \t");
+tokenize.tokenize(" \t");
-tokenize(" \t123");
+tokenize.tokenize(" \t123");
 
 try {
-    tokenize(" \t123aaa456");
+    tokenize.tokenize(" \t123aaa456");
 
 
 } catch (error) {
     console.log(error);
 }
-tokenize(" \t123\n456");
+tokenize.tokenize(" \t123\n456");
 
-tokenize("\"\"");
+tokenize.tokenize("\"\"");
-tokenize("\"123\"");
+tokenize.tokenize("\"123\"");
-tokenize("\"1\\\"23\"");
+tokenize.tokenize("\"1\\\"23\"");
-tokenize("\"1\\\"23\"  abc123");
+tokenize.tokenize("\"1\\\"23\"  abc123");
 
-tokenize("+0.012");
+tokenize.tokenize("+0.012");
-tokenize("0.0");
+tokenize.tokenize("0.0");
-tokenize("-222.0");
+tokenize.tokenize("-222.0");
-tokenize("1+1 ==2; 3+8 foo(12)");
+tokenize.tokenize("1+1 ==2; 3+8 foo(12)");
+
+console.log(tokenize.tokenize("2+2"));
 
 // harfbuzz test
 let harfbuzz = require("../src/harfbuzz.js");