re2js 2.8.2 → 2.8.4

package/README.md

@@ -264,7 +264,7 @@ This is incredibly powerful for profanity filters, routing engines, or log parse
 ```js
 import { RE2Set } from 're2js'
 
-// Create a new set. You can optionally pass an anchor and public RE2JS flags.
+// Create a new set. You can optionally pass an anchor, public RE2JS flags, and a max memory limit for the DFA.
 // Default anchor: RE2Set.UNANCHORED
 const set = new RE2Set()
 
@@ -286,17 +286,22 @@ console.log(set.match('All systems operational.'))
 // Outputs: []
 ```
 
-#### Anchoring a Set
+#### Anchoring a Set and Memory Limits
 
 You can strictly anchor the entire set by passing an anchor constant to the constructor (`RE2Set.UNANCHORED`, `RE2Set.ANCHOR_START`, or `RE2Set.ANCHOR_BOTH`).
 
-Additionally, you can pass standard public `RE2JS` flags (like `CASE_INSENSITIVE` or `LOOKBEHINDS`) as the second argument to apply them to all patterns in the set.
+You can pass standard public `RE2JS` flags (like `CASE_INSENSITIVE` or `LOOKBEHINDS`) as the second argument to apply them to all patterns in the set.
+
+Additionally, the third argument allows you to specify a `maxMem` limit (in bytes) to prevent the underlying DFA from exploding and consuming too much memory on highly ambiguous patterns.
+The default is 8 * 1024 * 1024 (8MB).
 
 ```js
 import { RE2Set, RE2JS } from 're2js'
 
-// Anchor the set to match the entire string, and make it case-insensitive
-const set = new RE2Set(RE2Set.ANCHOR_BOTH, RE2JS.CASE_INSENSITIVE)
+// Anchor the set to match the entire string, make it case-insensitive,
+// and limit the DFA memory usage to ~4MB (default is 8MB).
+const maxMem = 4 * 1024 * 1024;
+const set = new RE2Set(RE2Set.ANCHOR_BOTH, RE2JS.CASE_INSENSITIVE, maxMem)
 
 set.add('foo') // ID: 0
 set.add('bar') // ID: 1

package/build/index.cjs

@@ -2,7 +2,7 @@
 * re2js
 * RE2JS is the JavaScript port of RE2, a regular expression engine that provides linear time matching
 *
-* @version v2.8.2
+* @version v2.8.4
 * @author Oleksii Vasyliev
 * @homepage https://github.com/le0pard/re2js#readme
 * @repository github:le0pard/re2js
@@ -57,7 +57,7 @@ for (let i = 0; i < ASCII_SIZE; i++) {
 	else ASCII_TO_LOWER[i] = i;
 }
 var Codepoint = class {
-	static CODES = new Map([
+	static CODES = /* @__PURE__ */ new Map([
 		["\x07", 7],
 		["\b", 8],
 		["	", 9],
@@ -168,7 +168,7 @@ var UnicodeRangeTable = class {
 };
 //#endregion
 //#region src/UnicodeTables.js
-const B64_MAP = new Uint8Array(256);
+const B64_MAP = /* @__PURE__ */ new Uint8Array(256);
 for (let i = 0, b = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+-"; i < 64; i++) B64_MAP[b.charCodeAt(i)] = i;
 const decodeVLQ = (str) => {
 	const res = [];
@@ -1251,6 +1251,8 @@ var Matcher = class Matcher {
 	*/
 	resetMatcherInput(input) {
 		if (input === null) throw new Error("input is null");
+		if (!(input instanceof MatcherInputBase)) if (Utils.isByteArray(input)) input = MatcherInput.utf8(input);
+		else input = MatcherInput.utf16(input);
 		this.matcherInput = input;
 		this.reset();
 		return this;
@@ -2208,12 +2210,13 @@ var DFAState = class {
 };
 var DFA = class DFA {
 	static MAX_CACHE_CLEARS = 5;
-	constructor(prog) {
+	static STATE_MEMORY_ESTIMATE = 838;
+	constructor(prog, maxMem = 8388608) {
 		this.prog = prog;
 		this.stateCache = /* @__PURE__ */ new Map();
 		this.stateCount = 0;
 		this.startState = null;
-		this.stateLimit = 1e4;
+		this.stateLimit = Math.max(1, Math.floor(maxMem / DFA.STATE_MEMORY_ESTIMATE));
 		this.cacheClears = 0;
 		this.failed = false;
 		this.clock = 0;
@@ -2425,14 +2428,14 @@ const MAX_BACKTRACK_VECTOR = 256 * 1024;
 var BitState = class {
 	constructor() {
 		this.end = 0;
-		this.cap = new Int32Array(0);
-		this.matchcap = new Int32Array(0);
+		this.cap = /* @__PURE__ */ new Int32Array(0);
+		this.matchcap = /* @__PURE__ */ new Int32Array(0);
 		this.ncap = 0;
 		this.jobPc = new Int32Array(INITIAL_JOB_CAPACITY);
 		this.jobArg = new Uint8Array(INITIAL_JOB_CAPACITY);
 		this.jobPos = new Int32Array(INITIAL_JOB_CAPACITY);
 		this.jobLen = 0;
-		this.visited = new Uint32Array(0);
+		this.visited = /* @__PURE__ */ new Uint32Array(0);
 	}
 	reset(prog, end, ncap) {
 		this.end = end;
@@ -3942,7 +3945,7 @@ const code3 = [
 	97,
 	122
 ];
-const PERL_GROUPS = new Map([
+const PERL_GROUPS = /* @__PURE__ */ new Map([
 	["\\d", new CharGroup(1, code1)],
 	["\\D", new CharGroup(-1, code1)],
 	["\\s", new CharGroup(1, code2)],
@@ -4016,7 +4019,7 @@ const code17 = [
 	97,
 	102
 ];
-const POSIX_GROUPS = new Map([
+const POSIX_GROUPS = /* @__PURE__ */ new Map([
 	["[:alnum:]", new CharGroup(1, code4)],
 	["[:^alnum:]", new CharGroup(-1, code4)],
 	["[:alpha:]", new CharGroup(1, code5)],
@@ -5900,10 +5903,12 @@ var RE2Set = class RE2Set {
 	* Constructs a new RE2Set with the specified anchor mode and flags.
 	* @param {number} [anchor=RE2Set.UNANCHORED] - The anchoring mode (e.g., RE2Set.UNANCHORED).
 	* @param {number} [flags=0] - The public flags to apply to all patterns in the set.
+	* @param {number} [maxMem=8388608] - The maximum memory in bytes to use for the DFA (default 8MB).
 	*/
-	constructor(anchor = RE2Set.UNANCHORED, flags = 0) {
+	constructor(anchor = RE2Set.UNANCHORED, flags = 0, maxMem = 8388608) {
 		this.anchor = anchor;
 		this.jsFlags = flags;
+		this.maxMem = maxMem;
 		let re2Flags = RE2Flags.PERL;
 		if ((flags & PublicFlags.DISABLE_UNICODE_GROUPS) !== 0) re2Flags &= ~RE2Flags.UNICODE_GROUPS;
 		if ((flags & PublicFlags.LOOKBEHINDS) !== 0) re2Flags |= RE2Flags.LOOKBEHIND;
@@ -5938,7 +5943,7 @@ var RE2Set = class RE2Set {
 	compile() {
 		if (this.prog) return;
 		this.prog = Compiler.compileSet(this.regexps);
-		this.dfa = new DFA(this.prog);
+		this.dfa = new DFA(this.prog, this.maxMem);
 		this.dummyRe2 = {
 			prog: this.prog,
 			cond: this.prog.startCond(),

package/build/index.d.cts

@@ -2,7 +2,8 @@
 
 declare class DFA {
 	static MAX_CACHE_CLEARS: number;
-	constructor(prog: any);
+	static STATE_MEMORY_ESTIMATE: number;
+	constructor(prog: any, maxMem?: number);
 	prog: any;
 	stateCache: Map<any, any>;
 	stateCount: number;
@@ -53,10 +54,12 @@ export class RE2Set {
 	 * Constructs a new RE2Set with the specified anchor mode and flags.
 	 * @param {number} [anchor=RE2Set.UNANCHORED] - The anchoring mode (e.g., RE2Set.UNANCHORED).
 	 * @param {number} [flags=0] - The public flags to apply to all patterns in the set.
+	 * @param {number} [maxMem=8388608] - The maximum memory in bytes to use for the DFA (default 8MB).
 	 */
-	constructor(anchor?: number, flags?: number);
+	constructor(anchor?: number, flags?: number, maxMem?: number);
 	anchor: number;
 	jsFlags: number;
+	maxMem: number;
 	re2Flags: number;
 	regexps: any[];
 	prog: Prog;

package/build/index.d.ts

@@ -2,7 +2,8 @@
 
 declare class DFA {
 	static MAX_CACHE_CLEARS: number;
-	constructor(prog: any);
+	static STATE_MEMORY_ESTIMATE: number;
+	constructor(prog: any, maxMem?: number);
 	prog: any;
 	stateCache: Map<any, any>;
 	stateCount: number;
@@ -53,10 +54,12 @@ export class RE2Set {
 	 * Constructs a new RE2Set with the specified anchor mode and flags.
 	 * @param {number} [anchor=RE2Set.UNANCHORED] - The anchoring mode (e.g., RE2Set.UNANCHORED).
 	 * @param {number} [flags=0] - The public flags to apply to all patterns in the set.
+	 * @param {number} [maxMem=8388608] - The maximum memory in bytes to use for the DFA (default 8MB).
 	 */
-	constructor(anchor?: number, flags?: number);
+	constructor(anchor?: number, flags?: number, maxMem?: number);
 	anchor: number;
 	jsFlags: number;
+	maxMem: number;
 	re2Flags: number;
 	regexps: any[];
 	prog: Prog;

package/build/index.js

@@ -2,7 +2,7 @@
 * re2js
 * RE2JS is the JavaScript port of RE2, a regular expression engine that provides linear time matching
 *
-* @version v2.8.2
+* @version v2.8.4
 * @author Oleksii Vasyliev
 * @homepage https://github.com/le0pard/re2js#readme
 * @repository github:le0pard/re2js
@@ -56,7 +56,7 @@ for (let i = 0; i < ASCII_SIZE; i++) {
 	else ASCII_TO_LOWER[i] = i;
 }
 var Codepoint = class {
-	static CODES = new Map([
+	static CODES = /* @__PURE__ */ new Map([
 		["\x07", 7],
 		["\b", 8],
 		["	", 9],
@@ -167,7 +167,7 @@ var UnicodeRangeTable = class {
 };
 //#endregion
 //#region src/UnicodeTables.js
-const B64_MAP = new Uint8Array(256);
+const B64_MAP = /* @__PURE__ */ new Uint8Array(256);
 for (let i = 0, b = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+-"; i < 64; i++) B64_MAP[b.charCodeAt(i)] = i;
 const decodeVLQ = (str) => {
 	const res = [];
@@ -1250,6 +1250,8 @@ var Matcher = class Matcher {
 	*/
 	resetMatcherInput(input) {
 		if (input === null) throw new Error("input is null");
+		if (!(input instanceof MatcherInputBase)) if (Utils.isByteArray(input)) input = MatcherInput.utf8(input);
+		else input = MatcherInput.utf16(input);
 		this.matcherInput = input;
 		this.reset();
 		return this;
@@ -2207,12 +2209,13 @@ var DFAState = class {
 };
 var DFA = class DFA {
 	static MAX_CACHE_CLEARS = 5;
-	constructor(prog) {
+	static STATE_MEMORY_ESTIMATE = 838;
+	constructor(prog, maxMem = 8388608) {
 		this.prog = prog;
 		this.stateCache = /* @__PURE__ */ new Map();
 		this.stateCount = 0;
 		this.startState = null;
-		this.stateLimit = 1e4;
+		this.stateLimit = Math.max(1, Math.floor(maxMem / DFA.STATE_MEMORY_ESTIMATE));
 		this.cacheClears = 0;
 		this.failed = false;
 		this.clock = 0;
@@ -2424,14 +2427,14 @@ const MAX_BACKTRACK_VECTOR = 256 * 1024;
 var BitState = class {
 	constructor() {
 		this.end = 0;
-		this.cap = new Int32Array(0);
-		this.matchcap = new Int32Array(0);
+		this.cap = /* @__PURE__ */ new Int32Array(0);
+		this.matchcap = /* @__PURE__ */ new Int32Array(0);
 		this.ncap = 0;
 		this.jobPc = new Int32Array(INITIAL_JOB_CAPACITY);
 		this.jobArg = new Uint8Array(INITIAL_JOB_CAPACITY);
 		this.jobPos = new Int32Array(INITIAL_JOB_CAPACITY);
 		this.jobLen = 0;
-		this.visited = new Uint32Array(0);
+		this.visited = /* @__PURE__ */ new Uint32Array(0);
 	}
 	reset(prog, end, ncap) {
 		this.end = end;
@@ -3941,7 +3944,7 @@ const code3 = [
 	97,
 	122
 ];
-const PERL_GROUPS = new Map([
+const PERL_GROUPS = /* @__PURE__ */ new Map([
 	["\\d", new CharGroup(1, code1)],
 	["\\D", new CharGroup(-1, code1)],
 	["\\s", new CharGroup(1, code2)],
@@ -4015,7 +4018,7 @@ const code17 = [
 	97,
 	102
 ];
-const POSIX_GROUPS = new Map([
+const POSIX_GROUPS = /* @__PURE__ */ new Map([
 	["[:alnum:]", new CharGroup(1, code4)],
 	["[:^alnum:]", new CharGroup(-1, code4)],
 	["[:alpha:]", new CharGroup(1, code5)],
@@ -5899,10 +5902,12 @@ var RE2Set = class RE2Set {
 	* Constructs a new RE2Set with the specified anchor mode and flags.
 	* @param {number} [anchor=RE2Set.UNANCHORED] - The anchoring mode (e.g., RE2Set.UNANCHORED).
 	* @param {number} [flags=0] - The public flags to apply to all patterns in the set.
+	* @param {number} [maxMem=8388608] - The maximum memory in bytes to use for the DFA (default 8MB).
 	*/
-	constructor(anchor = RE2Set.UNANCHORED, flags = 0) {
+	constructor(anchor = RE2Set.UNANCHORED, flags = 0, maxMem = 8388608) {
 		this.anchor = anchor;
 		this.jsFlags = flags;
+		this.maxMem = maxMem;
 		let re2Flags = RE2Flags.PERL;
 		if ((flags & PublicFlags.DISABLE_UNICODE_GROUPS) !== 0) re2Flags &= ~RE2Flags.UNICODE_GROUPS;
 		if ((flags & PublicFlags.LOOKBEHINDS) !== 0) re2Flags |= RE2Flags.LOOKBEHIND;
@@ -5937,7 +5942,7 @@ var RE2Set = class RE2Set {
 	compile() {
 		if (this.prog) return;
 		this.prog = Compiler.compileSet(this.regexps);
-		this.dfa = new DFA(this.prog);
+		this.dfa = new DFA(this.prog, this.maxMem);
 		this.dummyRe2 = {
 			prog: this.prog,
 			cond: this.prog.startCond(),

package/build/index.umd.js

@@ -2,7 +2,7 @@
 * re2js
 * RE2JS is the JavaScript port of RE2, a regular expression engine that provides linear time matching
 *
-* @version v2.8.2
+* @version v2.8.4
 * @author Oleksii Vasyliev
 * @homepage https://github.com/le0pard/re2js#readme
 * @repository github:le0pard/re2js
@@ -60,7 +60,7 @@
 		else ASCII_TO_LOWER[i] = i;
 	}
 	var Codepoint = class {
-		static CODES = new Map([
+		static CODES = /* @__PURE__ */ new Map([
 			["\x07", 7],
 			["\b", 8],
 			["	", 9],
@@ -171,7 +171,7 @@
 	};
 	//#endregion
 	//#region src/UnicodeTables.js
-	const B64_MAP = new Uint8Array(256);
+	const B64_MAP = /* @__PURE__ */ new Uint8Array(256);
 	for (let i = 0, b = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+-"; i < 64; i++) B64_MAP[b.charCodeAt(i)] = i;
 	const decodeVLQ = (str) => {
 		const res = [];
@@ -1254,6 +1254,8 @@
 		*/
 		resetMatcherInput(input) {
 			if (input === null) throw new Error("input is null");
+			if (!(input instanceof MatcherInputBase)) if (Utils.isByteArray(input)) input = MatcherInput.utf8(input);
+			else input = MatcherInput.utf16(input);
 			this.matcherInput = input;
 			this.reset();
 			return this;
@@ -2211,12 +2213,13 @@
 	};
 	var DFA = class DFA {
 		static MAX_CACHE_CLEARS = 5;
-		constructor(prog) {
+		static STATE_MEMORY_ESTIMATE = 838;
+		constructor(prog, maxMem = 8388608) {
 			this.prog = prog;
 			this.stateCache = /* @__PURE__ */ new Map();
 			this.stateCount = 0;
 			this.startState = null;
-			this.stateLimit = 1e4;
+			this.stateLimit = Math.max(1, Math.floor(maxMem / DFA.STATE_MEMORY_ESTIMATE));
 			this.cacheClears = 0;
 			this.failed = false;
 			this.clock = 0;
@@ -2428,14 +2431,14 @@
 	var BitState = class {
 		constructor() {
 			this.end = 0;
-			this.cap = new Int32Array(0);
-			this.matchcap = new Int32Array(0);
+			this.cap = /* @__PURE__ */ new Int32Array(0);
+			this.matchcap = /* @__PURE__ */ new Int32Array(0);
 			this.ncap = 0;
 			this.jobPc = new Int32Array(INITIAL_JOB_CAPACITY);
 			this.jobArg = new Uint8Array(INITIAL_JOB_CAPACITY);
 			this.jobPos = new Int32Array(INITIAL_JOB_CAPACITY);
 			this.jobLen = 0;
-			this.visited = new Uint32Array(0);
+			this.visited = /* @__PURE__ */ new Uint32Array(0);
 		}
 		reset(prog, end, ncap) {
 			this.end = end;
@@ -3945,7 +3948,7 @@
 		97,
 		122
 	];
-	const PERL_GROUPS = new Map([
+	const PERL_GROUPS = /* @__PURE__ */ new Map([
 		["\\d", new CharGroup(1, code1)],
 		["\\D", new CharGroup(-1, code1)],
 		["\\s", new CharGroup(1, code2)],
@@ -4019,7 +4022,7 @@
 		97,
 		102
 	];
-	const POSIX_GROUPS = new Map([
+	const POSIX_GROUPS = /* @__PURE__ */ new Map([
 		["[:alnum:]", new CharGroup(1, code4)],
 		["[:^alnum:]", new CharGroup(-1, code4)],
 		["[:alpha:]", new CharGroup(1, code5)],
@@ -5903,10 +5906,12 @@
 		* Constructs a new RE2Set with the specified anchor mode and flags.
 		* @param {number} [anchor=RE2Set.UNANCHORED] - The anchoring mode (e.g., RE2Set.UNANCHORED).
 		* @param {number} [flags=0] - The public flags to apply to all patterns in the set.
+		* @param {number} [maxMem=8388608] - The maximum memory in bytes to use for the DFA (default 8MB).
 		*/
-		constructor(anchor = RE2Set.UNANCHORED, flags = 0) {
+		constructor(anchor = RE2Set.UNANCHORED, flags = 0, maxMem = 8388608) {
 			this.anchor = anchor;
 			this.jsFlags = flags;
+			this.maxMem = maxMem;
 			let re2Flags = RE2Flags.PERL;
 			if ((flags & PublicFlags.DISABLE_UNICODE_GROUPS) !== 0) re2Flags &= ~RE2Flags.UNICODE_GROUPS;
 			if ((flags & PublicFlags.LOOKBEHINDS) !== 0) re2Flags |= RE2Flags.LOOKBEHIND;
@@ -5941,7 +5946,7 @@
 		compile() {
 			if (this.prog) return;
 			this.prog = Compiler.compileSet(this.regexps);
-			this.dfa = new DFA(this.prog);
+			this.dfa = new DFA(this.prog, this.maxMem);
 			this.dummyRe2 = {
 				prog: this.prog,
 				cond: this.prog.startCond(),

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "re2js",
-  "version": "2.8.2",
+  "version": "2.8.4",
   "description": "RE2JS is the JavaScript port of RE2, a regular expression engine that provides linear time matching",
   "type": "module",
   "sideEffects": false,
@@ -56,24 +56,24 @@
   },
   "devDependencies": {
     "@eslint/js": "^9.39.4",
-    "@unicode/unicode-17.0.0": "^1.6.16",
-    "@vitest/browser": "^4.1.6",
-    "@vitest/browser-playwright": "^4.1.6",
-    "@vitest/coverage-v8": "^4.1.6",
+    "@unicode/unicode-17.0.0": "^1.6.17",
+    "@vitest/browser": "^4.1.9",
+    "@vitest/browser-playwright": "^4.1.9",
+    "@vitest/coverage-v8": "^4.1.9",
     "@yarnpkg/pnpify": "^4.1.6",
     "dts-bundle-generator": "^9.5.1",
     "eslint": "9.39.4",
     "eslint-config-prettier": "^10.1.8",
-    "eslint-import-resolver-node": "^0.3.10",
+    "eslint-import-resolver-node": "^0.4.0",
     "eslint-plugin-import": "^2.32.0",
-    "globals": "^17.6.0",
+    "globals": "^17.7.0",
     "lodash": "^4.18.1",
-    "playwright": "^1.60.0",
-    "prettier": "^3.8.3",
-    "rolldown": "^1.0.1",
+    "playwright": "^1.61.1",
+    "prettier": "^3.8.4",
+    "rolldown": "^1.1.3",
     "typescript": "^6.0.3",
     "unicode-property-value-aliases": "^3.9.0",
-    "vitest": "^4.1.6"
+    "vitest": "^4.1.9"
   },
-  "packageManager": "yarn@4.14.1"
+  "packageManager": "yarn@4.17.0"
 }