"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;
+};
+var __importDefault = (this && this.__importDefault) || function (mod) {
+ return (mod && mod.__esModule) ? mod : { "default": mod };
+};
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.matchAny = exports.tkTreeToSExp = void 0;
var fs = require('fs');
+const js_tokens_1 = __importDefault(require("js-tokens"));
+const util = __importStar(require("util"));
/**
- * wrap a x in a `Some(T)`
- * @param x : variable to be wrapped.
- * @returns wrapped `x`.
+ *
+ * # REPRESENTATION
*/
-function toSome(x) {
- return { _tag: "Some", value: x };
-}
/**
- * @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.value;
}
- };
+ }
+ return str;
}
-exports.match1Char = match1Char;
-;
+exports.tkTreeToSExp = tkTreeToSExp;
+/**inspect the inner of the representation. */
+let repr = (x) => { return util.inspect(x, { 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`.
+ * # PARSER UNITS
*/
-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" };
- }
+function toSome(x) {
+ return { _tag: "Some", value: x };
}
-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`.
+ * like `m ==> f` in ocaml
+ * @param m matchee wrapped
+ * @param f matching function
+ * @returns wrapped result
*/
-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.`);
+function thenDo(m, f) {
+ if (m._tag == "None") {
+ return m;
}
else {
- return s.charCodeAt(0);
+ var a = f(m.value);
+ if (a._tag == "Some") {
+ a.value.ast = m.value.ast.concat(a.value.ast);
+ }
+ return a;
}
}
-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`.
+ * @param m : the `TokenPair` to be consumed.
+ * @returns if the length of `m.remained` >= 1; consumes the matchee by 1 token
+ * and wraps it in `Some`,
+ * otherwise, returns `None`.
*/
-function thenDo(input, f) {
- if (input._tag == "None") {
- return input;
+function matchAny(m) {
+ if (m.remained.length >= 1) {
+ return {
+ _tag: "Some", value: {
+ matched: m.matched.concat(m.remained[0]),
+ remained: m.remained.slice(1),
+ ast: [m.remained[0]],
+ }
+ };
}
else {
- let inner = input.value;
- return f(inner);
+ return { _tag: "None" };
}
}
-exports.thenDo = thenDo;
+exports.matchAny = matchAny;
/**
- * @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
+ * like `f1 | f2` in regex
+ * @param f1 the first tried function
+ * @param f2 the second tried function
+ * @returns wrapped result
*/
function orDo(f1, f2) {
return (x) => {
- let f1x = (f1(x));
- {
- if (f1x._tag == "None") {
- return f2(x);
- }
- else {
- return f1x;
- }
+ let res1 = f1(x);
+ if (res1._tag == "Some") {
+ return res1;
}
- };
-}
-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);
+ else {
+ let res2 = f2(x);
+ return res2;
}
- ;
- 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`.
-*/
+ * like regex [^c]
+ * @param f input token function. one token only.
+ * @returns combined finction
+ */
function notDo(f) {
return (x) => {
- let wrapped_x = {
- _tag: "Some",
- value: x
- };
- let f_x = thenDo(wrapped_x, f);
- if (f_x._tag != "None") {
+ let res1 = f(x);
+ if (res1._tag == "Some") {
return { _tag: "None" };
}
else {
- return thenDo(wrapped_x, matchAny);
+ let res2 = matchAny(x);
+ return res2;
}
};
}
-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;
+function matchToken(typeName, value) {
+ return (t) => {
+ let headToken = t.remained[0];
+ if (headToken.type != typeName) {
+ return { _tag: "None" };
}
else {
- return wrapped_old_x;
+ if (value === undefined || value == headToken.value) {
+ let newTokenPair = {
+ matched: t.matched.concat(headToken),
+ remained: t.remained.slice(1),
+ ast: [headToken]
+ };
+ return { _tag: "Some", value: newTokenPair };
+ }
+ else {
+ return { _tag: "None" };
+ }
}
+ ;
};
}
-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));
- };
- console.log(input + ", result: ");
- console.log(thenDo(input_matchee_pair, integer));
+;
+/**
+ *
+ * # TEST
+ */
+const tokens = Array.from((0, js_tokens_1.default)(`import; foo from\t 'bar';
+import * as util from 'util';
+
+
+花非花，霧\\{非霧 。{{foo();}}下
+一句`));
+console.log("RESULT=" + repr(tokens));
+var mainTokenPair = {
+ matched: [],
+ remained: tokens,
+ ast: []
+};
+let a = thenDo(thenDo(toSome(mainTokenPair), matchToken('IdentifierName')), notDo(matchToken('Punctuator', ';')));
+console.log("RESULT=" + repr(a));
+if (a._tag == "Some") {
+ console.log("SEXP=" + tkTreeToSExp(a.value.ast));
}
-exports.tokenize = tokenize;
-tokenize("+123");
-tokenize("123");
-tokenize("-123");
-tokenize(" 123");
-tokenize("c123");
projects / clo / blobdiff