@adguard/agtree 3.0.0 → 3.1.0

package/dist/cjs/utils/regexp.js

@@ -1,190 +0,0 @@
-/*
- * AGTree v3.0.0 (build date: Tue, 12 Nov 2024 14:05:06 GMT)
- * (c) 2024 Adguard Software Ltd.
- * Released under the MIT license
- * https://github.com/AdguardTeam/tsurlfilter/tree/master/packages/agtree#readme
- */
-'use strict';
-
-var constants = require('./constants.js');
-
-/**
- * @file Regular expression utilities
- */
-// Special RegExp constants
-const REGEX_START = constants.CARET; // '^'
-const REGEX_END = constants.DOLLAR_SIGN; // '$'
-const REGEX_ANY_CHARACTERS = constants.DOT + constants.ASTERISK; // '.*'
-// Special adblock pattern symbols and their RegExp equivalents
-const ADBLOCK_URL_START = constants.PIPE + constants.PIPE; // '||'
-const ADBLOCK_URL_START_REGEX = '^(http|https|ws|wss)://([a-z0-9-_.]+\\.)?';
-const ADBLOCK_URL_SEPARATOR = constants.CARET; // '^'
-const ADBLOCK_URL_SEPARATOR_REGEX = '([^ a-zA-Z0-9.%_-]|$)';
-const ADBLOCK_WILDCARD = constants.ASTERISK; // '*'
-const ADBLOCK_WILDCARD_REGEX = REGEX_ANY_CHARACTERS;
-// Negation wrapper for RegExp patterns
-const REGEX_NEGATION_PREFIX = '^((?!';
-const REGEX_NEGATION_SUFFIX = ').)*$';
-/**
- * Special RegExp symbols
- *
- * @see https://developer.mozilla.org/en-US/docs/Web/JavaScript/Guide/Regular_Expressions#special-escape
- */
-const SPECIAL_REGEX_SYMBOLS = new Set([
-    constants.ASTERISK,
-    constants.CARET,
-    constants.CLOSE_CURLY_BRACKET,
-    constants.CLOSE_PARENTHESIS,
-    constants.CLOSE_SQUARE_BRACKET,
-    constants.DOLLAR_SIGN,
-    constants.DOT,
-    constants.ESCAPE_CHARACTER,
-    constants.OPEN_CURLY_BRACKET,
-    constants.OPEN_PARENTHESIS,
-    constants.OPEN_SQUARE_BRACKET,
-    constants.PIPE,
-    constants.PLUS,
-    constants.QUESTION_MARK,
-    constants.SLASH,
-]);
-/**
- * Utility functions for working with RegExp patterns
- */
-class RegExpUtils {
-    /**
-     * Checks whether a string possibly is a RegExp pattern.
-     * Flags are not supported.
-     *
-     * Note: it does not perform a full validation of the pattern,
-     * it just checks if the string starts and ends with a slash.
-     *
-     * @param pattern - Pattern to check
-     * @returns `true` if the string is a RegExp pattern, `false` otherwise
-     */
-    static isRegexPattern(pattern) {
-        const trimmedPattern = pattern.trim();
-        // Avoid false positives
-        return trimmedPattern.length > constants.REGEX_MARKER.length * 2
-            && trimmedPattern.startsWith(constants.REGEX_MARKER)
-            && trimmedPattern.endsWith(constants.REGEX_MARKER)
-            && trimmedPattern[constants.REGEX_MARKER.length - 2] !== constants.ESCAPE_CHARACTER;
-    }
-    /**
-     * Negates a RegExp pattern. Technically, this method wraps the pattern in `^((?!` and `).)*$`.
-     *
-     * RegExp modifiers are not supported.
-     *
-     * @param pattern Pattern to negate (can be wrapped in slashes or not)
-     * @returns Negated RegExp pattern
-     */
-    static negateRegexPattern(pattern) {
-        let result = pattern.trim();
-        let slashes = false;
-        // Remove the leading and trailing slashes (/)
-        if (RegExpUtils.isRegexPattern(result)) {
-            result = result.substring(constants.REGEX_MARKER.length, result.length - constants.REGEX_MARKER.length);
-            slashes = true;
-        }
-        // Only negate the pattern if it's not already negated
-        if (!(result.startsWith(REGEX_NEGATION_PREFIX) && result.endsWith(REGEX_NEGATION_SUFFIX))) {
-            // Remove leading caret (^)
-            if (result.startsWith(REGEX_START)) {
-                result = result.substring(REGEX_START.length);
-            }
-            // Remove trailing dollar sign ($)
-            if (result.endsWith(REGEX_END)) {
-                result = result.substring(0, result.length - REGEX_END.length);
-            }
-            // Wrap the pattern in the negation
-            result = `${REGEX_NEGATION_PREFIX}${result}${REGEX_NEGATION_SUFFIX}`;
-        }
-        // Add the leading and trailing slashes back if they were there
-        if (slashes) {
-            result = `${constants.REGEX_MARKER}${result}${constants.REGEX_MARKER}`;
-        }
-        return result;
-    }
-    /**
-     * Converts a basic adblock rule pattern to a RegExp pattern. Based on
-     * https://github.com/AdguardTeam/tsurlfilter/blob/9b26e0b4a0e30b87690bc60f7cf377d112c3085c/packages/tsurlfilter/src/rules/simple-regex.ts#L219
-     *
-     * @param pattern Pattern to convert
-     * @returns RegExp equivalent of the pattern
-     * @see {@link https://kb.adguard.com/en/general/how-to-create-your-own-ad-filters#basic-rules}
-     */
-    static patternToRegexp(pattern) {
-        const trimmed = pattern.trim();
-        // Return regex for any character sequence if the pattern is just |, ||, * or empty
-        if (trimmed === ADBLOCK_URL_START
-            || trimmed === constants.PIPE
-            || trimmed === ADBLOCK_WILDCARD
-            || trimmed === constants.EMPTY) {
-            return REGEX_ANY_CHARACTERS;
-        }
-        // If the pattern is already a RegExp, just return it, but remove the leading and trailing slashes
-        if (RegExpUtils.isRegexPattern(pattern)) {
-            return pattern.substring(constants.REGEX_MARKER.length, pattern.length - constants.REGEX_MARKER.length);
-        }
-        let result = constants.EMPTY;
-        let offset = 0;
-        let len = trimmed.length;
-        // Handle leading pipes
-        if (trimmed[0] === constants.PIPE) {
-            if (trimmed[1] === constants.PIPE) {
-                // Replace adblock url start (||) with its RegExp equivalent
-                result += ADBLOCK_URL_START_REGEX;
-                offset = ADBLOCK_URL_START.length;
-            }
-            else {
-                // Replace single pipe (|) with the RegExp start symbol (^)
-                result += REGEX_START;
-                offset = REGEX_START.length;
-            }
-        }
-        // Handle trailing pipes
-        let trailingPipe = false;
-        if (trimmed.endsWith(constants.PIPE)) {
-            trailingPipe = true;
-            len -= constants.PIPE.length;
-        }
-        // Handle the rest of the pattern, if any
-        for (; offset < len; offset += 1) {
-            if (trimmed[offset] === ADBLOCK_WILDCARD) {
-                // Replace adblock wildcard (*) with its RegExp equivalent
-                result += ADBLOCK_WILDCARD_REGEX;
-            }
-            else if (trimmed[offset] === ADBLOCK_URL_SEPARATOR) {
-                // Replace adblock url separator (^) with its RegExp equivalent
-                result += ADBLOCK_URL_SEPARATOR_REGEX;
-            }
-            else if (SPECIAL_REGEX_SYMBOLS.has(trimmed[offset])) {
-                // Escape special RegExp symbols (we handled pipe (|) and asterisk (*) already)
-                result += constants.ESCAPE_CHARACTER + trimmed[offset];
-            }
-            else {
-                // Just add any other character
-                result += trimmed[offset];
-            }
-        }
-        // Handle trailing pipes
-        if (trailingPipe) {
-            // Replace trailing pipe (|) with the RegExp end symbol ($)
-            result += REGEX_END;
-        }
-        return result;
-    }
-}
-
-exports.ADBLOCK_URL_SEPARATOR = ADBLOCK_URL_SEPARATOR;
-exports.ADBLOCK_URL_SEPARATOR_REGEX = ADBLOCK_URL_SEPARATOR_REGEX;
-exports.ADBLOCK_URL_START = ADBLOCK_URL_START;
-exports.ADBLOCK_URL_START_REGEX = ADBLOCK_URL_START_REGEX;
-exports.ADBLOCK_WILDCARD = ADBLOCK_WILDCARD;
-exports.ADBLOCK_WILDCARD_REGEX = ADBLOCK_WILDCARD_REGEX;
-exports.REGEX_ANY_CHARACTERS = REGEX_ANY_CHARACTERS;
-exports.REGEX_END = REGEX_END;
-exports.REGEX_NEGATION_PREFIX = REGEX_NEGATION_PREFIX;
-exports.REGEX_NEGATION_SUFFIX = REGEX_NEGATION_SUFFIX;
-exports.REGEX_START = REGEX_START;
-exports.RegExpUtils = RegExpUtils;
-exports.SPECIAL_REGEX_SYMBOLS = SPECIAL_REGEX_SYMBOLS;