"use strict";
+var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
+ if (k2 === undefined) k2 = k;
+ var desc = Object.getOwnPropertyDescriptor(m, k);
+ if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
+ desc = { enumerable: true, get: function() { return m[k]; } };
+ }
+ Object.defineProperty(o, k2, desc);
+}) : (function(o, m, k, k2) {
+ if (k2 === undefined) k2 = k;
+ o[k2] = m[k];
+}));
+var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
+ Object.defineProperty(o, "default", { enumerable: true, value: v });
+}) : function(o, v) {
+ o["default"] = v;
+});
+var __importStar = (this && this.__importStar) || function (mod) {
+ if (mod && mod.__esModule) return mod;
+ var result = {};
+ if (mod != null) for (var k in mod) if (k !== "default" && Object.prototype.hasOwnProperty.call(mod, k)) __createBinding(result, mod, k);
+ __setModuleDefault(result, mod);
+ return result;
+};
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 = exports.TokenType = void 0;
+exports.m1TType = void 0;
var fs = require('fs');
-/**
- * wrap a x in a `Some(T)`
- * @param x : variable to be wrapped.
- * @returns wrapped `x`.
- */
-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
- */
-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["OP"] = 4] = "OP";
- TokenType[TokenType["FLO"] = 5] = "FLO";
- TokenType[TokenType["INT"] = 6] = "INT";
-})(TokenType || (exports.TokenType = TokenType = {}));
+const node_process_1 = require("node:process");
+const tk = __importStar(require("./tokenize.js"));
+const util = __importStar(require("util"));
/**
* @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
+ * match one token type.
+ *
+ * it returns a function which test if the type of first token of the `remained` part of
+ * the argument of the function is `typ` , if it's true, update the `TokenMatcheePair` wrapped
* in `Some`. Otherwise, it returns `None`.
- * * @param c : the char to be test.
- * @returns the updated `MatchedPair` wrapped in `Some(x)` or `None`.
+ * * @param typ : the type to be test.
+ * @returns the updated `TokenMatcheePair` wrapped in `Some(x)` or `None`.
*/
-function match1Char(c) {
+function m1TType(typ) {
return (m) => {
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)
- } };
+ /**
+ * token to be matched
+ * */
+ const ttbm = m.remained[0];
+ if (ttbm.type == typ) {
+ let new_matched = m.matched.concat(ttbm);
+ let result = {
+ _tag: "Some", value: {
+ matched: new_matched,
+ remained: m.remained.slice(1)
+ }
+ };
+ return result;
}
else {
return { _tag: "None" };
}
};
}
-exports.match1Char = match1Char;
+exports.m1TType = m1TType;
;
+let toSome = tk.toSome;
+let thenDo = tk.thenDo;
+let zeroOrOnceDo = tk.zeroOrOnceDo;
+let orDo = tk.orDo;
+let zeroOrMoreDo = tk.zeroOrMoreDo;
+node_process_1.argv.forEach((val, index) => {
+ console.log(`${index}=${val}`);
+});
+let commandInput = "int a str b"; //argv[2];
+let commandInputTokenized = tk.tokenize(commandInput);
+let commandInputTokenizedFiltered = commandInputTokenized.filter((x) => {
+ return x.type != tk.TokenType.SP &&
+ x.type != tk.TokenType.NL;
+});
+console.log("aaa: " + util.inspect(commandInputTokenizedFiltered, { showHidden: true, depth: null }));
/**
- *
- * @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`.
+ * matchee pair of commandInputTokenized
*/
-function matchAny(m) {
- if (m.remained.length >= 1) {
- return { _tag: "Some", value: {
- matched: m.matched + m.remained[0],
- remained: m.remained.substring(1)
- } };
+let commandTPair = { matched: [],
+ remained: commandInputTokenizedFiltered };
+let tInt = m1TType(tk.TokenType.INT);
+let tFlo = m1TType(tk.TokenType.FLO);
+let tStr = m1TType(tk.TokenType.STR);
+let tId = m1TType(tk.TokenType.ID);
+let tApos = m1TType(tk.TokenType.APOS);
+function tBool(x) {
+ let text = x.remained[0].text;
+ if (text == "true" || text == "false") {
+ return thenDo(toSome(x), m1TType(tk.TokenType.ID));
}
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`.
+ * define the right hand side of a grammar
+ * eg. `LHS ::= a + b`
+ * @param process the right hand side processing : eg. `a + b` in `LHS`
+ * @param arrange define the order (0 starting) of the elements of the result.
+ * ast. : eg. `a + c` is `1 0 2` `(+ a c)`
+ * @returns the processed ast.
*/
-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 + "`)");
- }
+function gramRHS(process, arrange) {
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)
- } };
+ let middle = process(m);
+ console.log("Middle" + util.inspect(middle, { showHidden: true, depth: null }));
+ if (middle._tag == "None") {
+ return middle;
}
else {
- return { _tag: "None" };
+ let matched = middle.value.matched;
+ let arrLength = arrange.length;
+ let returnRrray = Array(arrange.length);
+ arrange.forEach((val, index) => {
+ returnRrray[arrange[index]] = matched[index];
+ });
+ let matchedTmp1Length = matched.length - arrLength;
+ console.log(matchedTmp1Length);
+ var matchedTmp1 = matched
+ .slice(0, matchedTmp1Length);
+ console.log("matchedTmp1" + util.inspect(matchedTmp1, { showHidden: true, depth: null }));
+ console.log("returnRrray" + util.inspect(returnRrray, { showHidden: true, depth: null }));
+ matchedTmp1.push(returnRrray);
+ let result = { _tag: "Some",
+ value: { matched: matchedTmp1,
+ remained: middle.value.remained } };
+ return result;
}
};
}
-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`.
+ * typeABS ::= "'" ID
*/
-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);
+var typeABS = (x) => {
+ var result = thenDo(thenDo(toSome(x), tApos), tId);
+ if (result._tag == "Some" && "text" in result.value.matched[1]) {
+ var realToken = result.value.matched[1];
+ realToken.text = "'" + realToken.text;
+ result.value.matched = [realToken];
}
-}
-exports.charToCodepoint = charToCodepoint;
+ return result;
+};
/**
- * @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`.
+ * TypeId ::= typeABS | ID
*/
-function thenDo(input, f) {
- if (input._tag == "None") {
- return input;
- }
- else {
- let inner = input.value;
- return f(inner);
- }
-}
-exports.thenDo = thenDo;
+var typeName = (x) => {
+ return thenDo(toSome(x), orDo(typeABS, tId));
+};
/**
- * @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
+ * CONST ::= INT | STR | FLO | BOOL
*/
-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;
+ * TODO: 要用 debugger 檢查分析問題
+ */
+var constParser = gramRHS((x) => { return thenDo(toSome(x), orDo(orDo(orDo(tInt, tFlo), tStr), tBool)); }, [0]);
/**
-* @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;
+ * storing the tree
+ */
+var astTree = [];
/**
- * 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)
+ * TYPE_PAIR ::= TYP_ID ID
*/
-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 });
- // integer = ([+]|[-])?\d\d*
- let integer = (x) => {
- let wrapped_x = toSome(x);
- let plusMinus = orDo(match1Char('+'), match1Char('-')); // ([+]|[-])
- let d = matchRange('0', '9'); // \d
- var result = thenDo(thenDo(thenDo(wrapped_x, zeroOrOnceDo(plusMinus)), d), zeroOrMoreDo(d));
- if (result._tag == "Some") {
- result.value.matched_type = TokenType.INT;
- }
- return result;
- };
- let space = (x) => {
- let wrapped_x = toSome(x);
- let s_aux = orDo(match1Char(' '), match1Char('\t')); // (" " | "\t")
- var result = thenDo(thenDo(wrapped_x, s_aux), zeroOrMoreDo(s_aux));
- if (result._tag == "Some") {
- result.value.matched_type = TokenType.SP;
- }
- return result;
- };
- let newline = (x) => {
- let wrapped_x = toSome(x);
- // nl = \r?\n
- let result = thenDo(thenDo(wrapped_x, zeroOrOnceDo(match1Char('\r'))), match1Char('\n'));
- if (result._tag == "Some") {
- result.value.matched_type = TokenType.NL;
- }
- return result;
- };
- let term = (token_list, x) => {
- var ln = 1;
- var col = 0;
- var old_x = x;
- let term_list = [newline, space, integer];
- 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;
+var typePair = (x) => {
+ let a = thenDo(thenDo(x.maybeTokens, typeName), tId);
+ if (a._tag == "Some") {
+ let matched = a.value.matched;
+ let slice = matched.slice(matched.length - 2);
+ console.log("slice" + slice);
+ let b = { maybeTokens: a, ast: slice };
+ return b;
+ }
+ else {
+ let b = { maybeTokens: a, ast: [] };
+ return b;
+ }
+};
+/**
+ * function's arguments
+ * FN_ARGS = TYPE_PAIR ("," TYPE_PAIR)+
+ */
+var fnArgs = (x) => {
+ let wrapper = { maybeTokens: toSome(x), ast: [] };
+ let a = typePair(wrapper);
+ console.log("AAAAA" + util.inspect(a, { showHidden: true, depth: null }));
+ let abanibi = typePair(a);
+ console.log("ABNB" + util.inspect(abanibi, { showHidden: true, depth: null }));
+ return { maybeTokens: abanibi.maybeTokens, ast: [a.ast, abanibi.ast] };
+};
+let tree = fnArgs(commandTPair);
+console.log("CHRANN" + util.inspect(tree, { showHidden: true, depth: null }));