"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.tkTreeToSExp = void 0;
var fs = require('fs');
+const util = __importStar(require("util"));
+const p = __importStar(require("typescript-parsec"));
/**
- * 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
+ * # REPRESENTATION
*/
-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 = {}));
/**
- * @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`.
+ * convert a `tkTree` AST to S-expr string
+ * @param t the `tkTree`
+ * @returns S-expr String
*/
-function match1Char(c) {
- 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)
- } };
+function tkTreeToSExp(t) {
+ var str = "";
+ if (Array.isArray(t)) {
+ let strArray = t.map((x) => tkTreeToSExp(x));
+ str = "(" + strArray.join(" ") + ")";
+ }
+ else {
+ if (t === undefined) {
+ str = "%undefined";
}
else {
- return { _tag: "None" };
+ str = t;
}
- };
+ }
+ return str;
}
-exports.match1Char = match1Char;
-;
+exports.tkTreeToSExp = tkTreeToSExp;
+/**inspect the inner of the representation. */
+let repr = (x) => { return util.inspect(x, { depth: null }); };
+var TokenKind;
+(function (TokenKind) {
+ TokenKind[TokenKind["Seperator"] = 0] = "Seperator";
+ TokenKind[TokenKind["Semicolon"] = 1] = "Semicolon";
+ TokenKind[TokenKind["Number"] = 2] = "Number";
+ TokenKind[TokenKind["Op"] = 3] = "Op";
+ TokenKind[TokenKind["ExprMark"] = 4] = "ExprMark";
+ TokenKind[TokenKind["ExcapeAt"] = 5] = "ExcapeAt";
+ TokenKind[TokenKind["Paren"] = 6] = "Paren";
+ TokenKind[TokenKind["SpaceNL"] = 7] = "SpaceNL";
+ TokenKind[TokenKind["Id"] = 8] = "Id";
+ TokenKind[TokenKind["Str"] = 9] = "Str";
+ TokenKind[TokenKind["Comment"] = 10] = "Comment";
+})(TokenKind || (TokenKind = {}));
+/**
+ * Parsing
+ */
+const lexer = p.buildLexer([
+ [true, /^\d+(\.\d+)?/g, TokenKind.Number],
+ [true, /^\\\@/g, TokenKind.ExcapeAt],
+ [true, /^\/\*([^/]|\/[^*])*\*\//g, TokenKind.Comment],
+ [true, /^\;/g, TokenKind.Semicolon],
+ [true, /^[-][-][-]/g, TokenKind.Seperator],
+ [true, /^[\+\-\*\/\&\|\!\^\<\>\~\=\?]+/g, TokenKind.Op],
+ [true, /^\@/g, TokenKind.ExprMark],
+ [true, /^[()\[\]{}]/g, TokenKind.Paren],
+ [true, /^["]([\"]|[\\].)*["]/g, TokenKind.Str],
+ [true, /^[']([\']|[\\].)*[']/g, TokenKind.Str],
+ [true, /^[()\[\]{}]/g, TokenKind.Paren],
+ [true, /^[^\/\\\@\s\n\t\r;]+/g, TokenKind.Id],
+ [true, /^(\s|\n|\r|\t)+/g, TokenKind.SpaceNL],
+]);
/**
*
- * @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`.
+ * # TEST
*/
-function matchAny(m) {
- if (m.remained.length >= 1) {
- return { _tag: "Some", value: {
- matched: m.matched + m.remained[0],
- remained: m.remained.substring(1)
- } };
+const inputTxt = `import a as b; /*bacourt*/
+/* ba choir
+ipsum lorem*/
+
+import you as john;
+---
+
+臺中市\\\@
+
+公園
+@1+2==3;
+
+console.log("122");@
+
+山頂
+`;
+const PROG = p.rule();
+const SEGMENT = p.rule();
+const IMPORT = p.rule();
+const IMPORTS = p.rule();
+const SEMICOLON = p.rule();
+const EXCAPE_AT = p.rule();
+const NOT_AT_TEXT = p.rule();
+const CONTENT = p.rule();
+let doubleMinus = { type: 'Punctuator', value: '--' };
+let doubleMinus2 = p.str('--');
+const TERM = p.rule();
+function applySegment(input) {
+ let unpackedInnerExprs = input[1].map((x) => { return x.text; });
+ return ["%exprs", unpackedInnerExprs];
+}
+function applySemiColon(value) {
+ return value.text;
+}
+function applyParts(first, second) {
+ return ["%clo", first, second[1]];
+}
+function applyComment(value) {
+ return [value.text];
+}
+function applyImport(input) {
+ let importTail = input[1].map(x => x.text);
+ return ["import"].concat(importTail);
+}
+;
+/*
+function applyImportComment(input: [Token<TokenKind>,Token<TokenKind>[],
+ tkTree, Token<TokenKind.Comment>]) : tkTree{
+ let importTail = input[1].map(x=>x.text);
+ let comment = [input[3].text];
+ return ["import"].concat(importTail).concat(comment);
+};*/
+function applyImports(input) {
+ let resultBody = [input[0]].concat(input[1]);
+ let resultWrapper = ["%import", resultBody];
+ return resultWrapper;
+}
+;
+function applyNotAtText(value) {
+ if (value.text == "\\\@") {
+ return '@';
}
else {
- return { _tag: "None" };
+ return value.text;
}
}
-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" };
- }
- };
+;
+function applyText(input) {
+ return ["%text", input];
+}
+;
+function applyContent(input) {
+ return ["%content", input];
}
-exports.matchRange = matchRange;
;
+function applySpaceNL(value) {
+ return value.text;
+}
/**
- * 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`.
+ * IMPORTEE: Number, Op, Paren, Id, Str, Comment,
*/
-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;
+let IMPORTEE = p.alt(p.tok(TokenKind.Number), p.tok(TokenKind.Op), p.tok(TokenKind.Paren), p.tok(TokenKind.Id), p.tok(TokenKind.Str), p.tok(TokenKind.SpaceNL), p.tok(TokenKind.Comment));
+let NOT_AT = p.alt(p.tok(TokenKind.Seperator), p.tok(TokenKind.Semicolon), p.tok(TokenKind.Number), p.tok(TokenKind.ExcapeAt), p.tok(TokenKind.Op), p.tok(TokenKind.Paren), p.tok(TokenKind.SpaceNL), p.tok(TokenKind.Id), p.tok(TokenKind.Str), p.tok(TokenKind.Comment));
/**
- * @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`.
+ * PROG : IMPORTS '---' CONTENT;
*/
-function thenDo(input, f) {
- if (input._tag == "None") {
- return input;
- }
- else {
- let inner = input.value;
- return f(inner);
- }
-}
-exports.thenDo = thenDo;
+PROG.setPattern(p.lrec_sc(IMPORTS, p.seq(p.str('---'), CONTENT), applyParts));
/**
- * @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
+ * NOT_AT_TEXT : NOT_AT
*/
-function orDo(f1, f2) {
- return (x) => {
- let f1x = (f1(x));
- {
- if (f1x._tag == "None") {
- return f2(x);
- }
- else {
- return f1x;
- }
- }
- };
-}
-exports.orDo = orDo;
+NOT_AT_TEXT.setPattern(p.apply(NOT_AT, applyNotAtText));
+IMPORTS.setPattern(p.apply(p.seq(IMPORT, p.rep(IMPORT)), applyImports));
/**
-* @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;
+ * IMPORT :
+ * 'import' IMPORTEE* SEMICOLON |
+ * COMMENT |
+ */
+IMPORT.setPattern(p.alt(p.apply(p.seq(p.str('import'), p.rep_sc(IMPORTEE), SEMICOLON), applyImport), p.apply(p.tok(TokenKind.Comment), applyComment), p.apply(p.tok(TokenKind.SpaceNL), applySpaceNL)));
/**
-* @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;
+ * SEMICOLON : ';';
+ */
+SEMICOLON.setPattern(p.apply(p.tok(TokenKind.Semicolon), applySemiColon));
/**
- * 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)
+ * SEGMENT : '@' NOT_AT* '@' |
+ * (NOT_AT_TEXT | EXCAPE_AT)*
*/
-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;
+SEGMENT.setPattern(p.alt(p.apply(p.rep_sc(NOT_AT_TEXT), applyText), p.apply(p.seq(p.str('@'), p.rep(NOT_AT), p.str('@')), applySegment)));
+/**
+ * CONTENT : SEGMENT*
+ */
+CONTENT.setPattern(p.apply(p.rep(SEGMENT), applyContent));
+let tree = p.expectSingleResult(p.expectEOF(PROG.parse(lexer.parse(inputTxt))));
+console.log("RESULT=" + tkTreeToSExp(tree));