/*
|
MIT License http://www.opensource.org/licenses/mit-license.php
|
Author Tobias Koppers @sokra
|
*/
|
|
"use strict";
|
|
/**
|
* Returns quoted meta.
|
* @param {string} str string
|
* @returns {string} quoted meta
|
*/
|
const quoteMeta = (str) => str.replace(/[-[\]\\/{}()*+?.^$|]/g, "\\$&");
|
|
/**
|
* Quote meta in char class.
|
* @param {string} char character to escape for use in character class
|
* @returns {string} escaped character
|
*/
|
const quoteMetaInCharClass = (char) => {
|
// In character class, only these need escaping: ] \ ^ -
|
if (char === "]" || char === "\\" || char === "^" || char === "-") {
|
return `\\${char}`;
|
}
|
return char;
|
};
|
|
/**
|
* Converts an array of single characters into an optimized character class string
|
* using ranges where possible. E.g., ["1","2","3","4","a"] => "1-4a"
|
* @param {string[]} chars array of single characters (should be sorted)
|
* @returns {string} optimized character class content (without the brackets)
|
*/
|
const charsToCharClassContent = (chars) => {
|
if (chars.length === 0) return "";
|
if (chars.length === 1) return quoteMetaInCharClass(chars[0]);
|
|
// Sort by char code
|
const sorted = [...chars].sort((a, b) => a.charCodeAt(0) - b.charCodeAt(0));
|
|
/** @type {string[]} */
|
const parts = [];
|
let rangeStart = sorted[0];
|
let rangeEnd = sorted[0];
|
|
for (let i = 1; i < sorted.length; i++) {
|
const char = sorted[i];
|
const prevCode = rangeEnd.charCodeAt(0);
|
const currCode = char.charCodeAt(0);
|
|
if (currCode === prevCode + 1) {
|
// Extend the range
|
rangeEnd = char;
|
} else {
|
// Flush the current range
|
parts.push(formatRange(rangeStart, rangeEnd));
|
rangeStart = char;
|
rangeEnd = char;
|
}
|
}
|
// Flush the last range
|
parts.push(formatRange(rangeStart, rangeEnd));
|
|
return parts.join("");
|
};
|
|
/**
|
* Formats a range of characters for use in a character class
|
* @param {string} start start character
|
* @param {string} end end character
|
* @returns {string} formatted range
|
*/
|
const formatRange = (start, end) => {
|
const startCode = start.charCodeAt(0);
|
const endCode = end.charCodeAt(0);
|
const length = endCode - startCode + 1;
|
|
if (length === 1) {
|
return quoteMetaInCharClass(start);
|
}
|
if (length === 2) {
|
// For 2 chars, just list them (e.g., "ab" instead of "a-b")
|
return quoteMetaInCharClass(start) + quoteMetaInCharClass(end);
|
}
|
// For 3+ chars, use range notation
|
return `${quoteMetaInCharClass(start)}-${quoteMetaInCharClass(end)}`;
|
};
|
|
/**
|
* Returns string.
|
* @param {string} str string
|
* @returns {string} string
|
*/
|
const toSimpleString = (str) => {
|
if (`${Number(str)}` === str) {
|
return str;
|
}
|
return JSON.stringify(str);
|
};
|
|
/**
|
* Compile boolean matcher.
|
* @param {Record<string | number, boolean>} map value map
|
* @returns {boolean | ((value: string) => string)} true/false, when unconditionally true/false, or a template function to determine the value at runtime
|
*/
|
const compileBooleanMatcher = (map) => {
|
const positiveItems = Object.keys(map).filter((i) => map[i]);
|
const negativeItems = Object.keys(map).filter((i) => !map[i]);
|
if (positiveItems.length === 0) return false;
|
if (negativeItems.length === 0) return true;
|
return compileBooleanMatcherFromLists(positiveItems, negativeItems);
|
};
|
|
/**
|
* Compile boolean matcher from lists.
|
* @param {string[]} positiveItems positive items
|
* @param {string[]} negativeItems negative items
|
* @returns {(value: string) => string} a template function to determine the value at runtime
|
*/
|
const compileBooleanMatcherFromLists = (positiveItems, negativeItems) => {
|
if (positiveItems.length === 0) return () => "false";
|
if (negativeItems.length === 0) return () => "true";
|
if (positiveItems.length === 1) {
|
return (value) => `${toSimpleString(positiveItems[0])} == ${value}`;
|
}
|
if (negativeItems.length === 1) {
|
return (value) => `${toSimpleString(negativeItems[0])} != ${value}`;
|
}
|
const positiveRegexp = itemsToRegexp(positiveItems);
|
const negativeRegexp = itemsToRegexp(negativeItems);
|
if (positiveRegexp.length <= negativeRegexp.length) {
|
return (value) => `/^${positiveRegexp}$/.test(${value})`;
|
}
|
return (value) => `!/^${negativeRegexp}$/.test(${value})`;
|
};
|
|
/** @typedef {string[][]} ListOfCommonItems */
|
|
/**
|
* Returns list of common items.
|
* @param {Set<string>} itemsSet items set
|
* @param {(str: string) => string | false} getKey get key function
|
* @param {(str: string[]) => boolean} condition condition
|
* @returns {ListOfCommonItems} list of common items
|
*/
|
const popCommonItems = (itemsSet, getKey, condition) => {
|
/** @type {Map<string, string[]>} */
|
const map = new Map();
|
for (const item of itemsSet) {
|
const key = getKey(item);
|
if (key) {
|
let list = map.get(key);
|
if (list === undefined) {
|
/** @type {string[]} */
|
list = [];
|
map.set(key, list);
|
}
|
list.push(item);
|
}
|
}
|
/** @type {ListOfCommonItems} */
|
const result = [];
|
for (const list of map.values()) {
|
if (condition(list)) {
|
for (const item of list) {
|
itemsSet.delete(item);
|
}
|
result.push(list);
|
}
|
}
|
return result;
|
};
|
|
/**
|
* Gets common prefix.
|
* @param {string[]} items items
|
* @returns {string} common prefix
|
*/
|
const getCommonPrefix = (items) => {
|
let prefix = items[0];
|
for (let i = 1; i < items.length; i++) {
|
const item = items[i];
|
for (let p = 0; p < prefix.length; p++) {
|
if (item[p] !== prefix[p]) {
|
prefix = prefix.slice(0, p);
|
break;
|
}
|
}
|
}
|
return prefix;
|
};
|
|
/**
|
* Gets common suffix.
|
* @param {string[]} items items
|
* @returns {string} common suffix
|
*/
|
const getCommonSuffix = (items) => {
|
let suffix = items[0];
|
for (let i = 1; i < items.length; i++) {
|
const item = items[i];
|
for (let p = item.length - 1, s = suffix.length - 1; s >= 0; p--, s--) {
|
if (item[p] !== suffix[s]) {
|
suffix = suffix.slice(s + 1);
|
break;
|
}
|
}
|
}
|
return suffix;
|
};
|
|
/**
|
* Returns regexp.
|
* @param {string[]} itemsArr array of items
|
* @returns {string} regexp
|
*/
|
const itemsToRegexp = (itemsArr) => {
|
if (itemsArr.length === 1) {
|
return quoteMeta(itemsArr[0]);
|
}
|
/** @type {string[]} */
|
const finishedItems = [];
|
|
// merge single char items: (a|b|c|d|ef) => ([abcd]|ef)
|
let countOfSingleCharItems = 0;
|
for (const item of itemsArr) {
|
if (item.length === 1) {
|
countOfSingleCharItems++;
|
}
|
}
|
// special case for only single char items
|
if (countOfSingleCharItems === itemsArr.length) {
|
return `[${charsToCharClassContent(itemsArr)}]`;
|
}
|
/** @type {Set<string>} */
|
const items = new Set(itemsArr.sort());
|
if (countOfSingleCharItems > 2) {
|
/** @type {string[]} */
|
const singleCharItems = [];
|
for (const item of items) {
|
if (item.length === 1) {
|
singleCharItems.push(item);
|
items.delete(item);
|
}
|
}
|
finishedItems.push(`[${charsToCharClassContent(singleCharItems)}]`);
|
}
|
|
// special case for 2 items with common prefix/suffix
|
if (finishedItems.length === 0 && items.size === 2) {
|
const prefix = getCommonPrefix(itemsArr);
|
const suffix = getCommonSuffix(
|
itemsArr.map((item) => item.slice(prefix.length))
|
);
|
if (prefix.length > 0 || suffix.length > 0) {
|
return `${quoteMeta(prefix)}${itemsToRegexp(
|
itemsArr.map((i) => i.slice(prefix.length, -suffix.length || undefined))
|
)}${quoteMeta(suffix)}`;
|
}
|
}
|
|
// special case for 2 items with common suffix
|
if (finishedItems.length === 0 && items.size === 2) {
|
/** @type {SetIterator<string>} */
|
const it = items[Symbol.iterator]();
|
const a = /** @type {string} */ (it.next().value);
|
const b = /** @type {string} */ (it.next().value);
|
if (a.length > 0 && b.length > 0 && a.slice(-1) === b.slice(-1)) {
|
return `${itemsToRegexp([a.slice(0, -1), b.slice(0, -1)])}${quoteMeta(
|
a.slice(-1)
|
)}`;
|
}
|
}
|
|
// find common prefix: (a1|a2|a3|a4|b5) => (a(1|2|3|4)|b5)
|
const prefixed = popCommonItems(
|
items,
|
(item) => (item.length >= 1 ? item[0] : false),
|
(list) => {
|
if (list.length >= 3) return true;
|
if (list.length <= 1) return false;
|
return list[0][1] === list[1][1];
|
}
|
);
|
for (const prefixedItems of prefixed) {
|
const prefix = getCommonPrefix(prefixedItems);
|
finishedItems.push(
|
`${quoteMeta(prefix)}${itemsToRegexp(
|
prefixedItems.map((i) => i.slice(prefix.length))
|
)}`
|
);
|
}
|
|
// find common suffix: (a1|b1|c1|d1|e2) => ((a|b|c|d)1|e2)
|
const suffixed = popCommonItems(
|
items,
|
(item) => (item.length >= 1 ? item.slice(-1) : false),
|
(list) => {
|
if (list.length >= 3) return true;
|
if (list.length <= 1) return false;
|
return list[0].slice(-2) === list[1].slice(-2);
|
}
|
);
|
for (const suffixedItems of suffixed) {
|
const suffix = getCommonSuffix(suffixedItems);
|
finishedItems.push(
|
`${itemsToRegexp(
|
suffixedItems.map((i) => i.slice(0, -suffix.length))
|
)}${quoteMeta(suffix)}`
|
);
|
}
|
|
/** @type {string[]} */
|
const conditional = [...finishedItems, ...Array.from(items, quoteMeta)];
|
if (conditional.length === 1) return conditional[0];
|
return `(${conditional.join("|")})`;
|
};
|
|
compileBooleanMatcher.fromLists = compileBooleanMatcherFromLists;
|
compileBooleanMatcher.itemsToRegexp = itemsToRegexp;
|
|
module.exports = compileBooleanMatcher;
|
