/** * Utilities for hex, bytes, CSPRNG. * @module */ /*! noble-ciphers - MIT License (c) 2023 Paul Miller (paulmillr.com) */ /** * Bytes API type helpers for old + new TypeScript. * * TS 5.6 has `Uint8Array`, while TS 5.9+ made it generic `Uint8Array`. * We can't use specific return type, because TS 5.6 will error. * We can't use generic return type, because most TS 5.9 software will expect specific type. * * Maps typed-array input leaves to broad forms. * These are compatibility adapters, not ownership guarantees. * * - `TArg` keeps byte inputs broad. * - `TRet` marks byte outputs for TS 5.6 and TS 5.9+ compatibility. */ export type TypedArg = T extends BigInt64Array ? BigInt64Array : T extends BigUint64Array ? BigUint64Array : T extends Float32Array ? Float32Array : T extends Float64Array ? Float64Array : T extends Int16Array ? Int16Array : T extends Int32Array ? Int32Array : T extends Int8Array ? Int8Array : T extends Uint16Array ? Uint16Array : T extends Uint32Array ? Uint32Array : T extends Uint8ClampedArray ? Uint8ClampedArray : T extends Uint8Array ? Uint8Array : never; /** Maps typed-array output leaves to narrow TS-compatible forms. */ export type TypedRet = T extends BigInt64Array ? ReturnType : T extends BigUint64Array ? ReturnType : T extends Float32Array ? ReturnType : T extends Float64Array ? ReturnType : T extends Int16Array ? ReturnType : T extends Int32Array ? ReturnType : T extends Int8Array ? ReturnType : T extends Uint16Array ? ReturnType : T extends Uint32Array ? ReturnType : T extends Uint8ClampedArray ? ReturnType : T extends Uint8Array ? ReturnType : never; /** Recursively adapts byte-carrying API input types. See {@link TypedArg}. */ export type TArg = | T | ([TypedArg] extends [never] ? T extends (...args: infer A) => infer R ? ((...args: { [K in keyof A]: TRet }) => TArg) & { [K in keyof T]: T[K] extends (...args: any) => any ? T[K] : TArg; } : T extends [infer A, ...infer R] ? [TArg, ...{ [K in keyof R]: TArg }] : T extends readonly [infer A, ...infer R] ? readonly [TArg, ...{ [K in keyof R]: TArg }] : T extends (infer A)[] ? TArg[] : T extends readonly (infer A)[] ? readonly TArg[] : T extends Promise ? Promise> : T extends object ? { [K in keyof T]: TArg } : T : TypedArg); /** Recursively adapts byte-carrying API output types. See {@link TypedArg}. */ export type TRet = T extends unknown ? T & ([TypedRet] extends [never] ? T extends (...args: infer A) => infer R ? ((...args: { [K in keyof A]: TArg }) => TRet) & { [K in keyof T]: T[K] extends (...args: any) => any ? T[K] : TRet; } : T extends [infer A, ...infer R] ? [TRet, ...{ [K in keyof R]: TRet }] : T extends readonly [infer A, ...infer R] ? readonly [TRet, ...{ [K in keyof R]: TRet }] : T extends (infer A)[] ? TRet[] : T extends readonly (infer A)[] ? readonly TRet[] : T extends Promise ? Promise> : T extends object ? { [K in keyof T]: TRet } : T : TypedRet) : never; export function aarray( item: unknown, title: string, inner: (elm: T, title: string) => void = () => {} ): T[] { if (!Array.isArray(item)) throw new TypeError(`"${title}" expected array, got type=${typeof item}`); for (let i = 0; i < item.length; i++) inner(item[i], `${title}[${i}]`); return item; } /** * Checks if something is Uint8Array. Be careful: nodejs Buffer will return true. * @param a - Value to inspect. * @returns `true` when the value is a Uint8Array view, including Node's `Buffer`. * @example * Guards a value before treating it as raw key material. * * ```ts * isBytes(new Uint8Array()); * ``` */ export function isBytes(a: unknown): a is Uint8Array { // Plain `instanceof Uint8Array` is too strict for some Buffer / proxy / // cross-realm cases. The fallback still requires a real ArrayBuffer view // so plain JSON-deserialized `{ constructor: ... }` // spoofing is rejected, and `BYTES_PER_ELEMENT === 1` keeps the fallback on byte-oriented views. return ( a instanceof Uint8Array || (ArrayBuffer.isView(a) && a.constructor.name === 'Uint8Array' && 'BYTES_PER_ELEMENT' in a && a.BYTES_PER_ELEMENT === 1) ); } // Shared error-message prefix builder. Only called on throw paths, so assert // success paths never pay for the string concatenation. const atitle = (title: string): string => (title ? `"${title}" ` : ''); /** * Asserts something is boolean. * @param value - Value to validate. * @returns The validated boolean. * @throws On wrong argument types. {@link TypeError} * @example * Validates a boolean option before branching on it. * * ```ts * abool(true); * ``` */ export function abool(value: boolean, title: string = ''): boolean { if (typeof value !== 'boolean') throw new TypeError(atitle(title) + 'expected boolean, got type=' + typeof value); return value; } /** * Asserts something is a non-negative safe integer. * @param n - Value to validate. * @returns The validated number. * @throws On wrong argument types. {@link TypeError} * @throws On wrong argument ranges or values. {@link RangeError} * @example * Validates a non-negative length or counter. * * ```ts * anumber(1); * ``` */ export function anumber(n: number, title: string = ''): number { if (typeof n !== 'number') throw new TypeError(atitle(title) + 'expected number, got ' + typeof n); if (!Number.isSafeInteger(n) || n < 0) throw new RangeError(atitle(title) + 'expected integer >= 0, got ' + n); return n; } /** * Asserts something is Uint8Array. * @param value - Value to validate. * @param length - Expected byte length. * @param title - Optional label used in error messages. * @returns The validated byte array. * On Node, `Buffer` is accepted too because it is a Uint8Array view. * @throws On wrong argument types. {@link TypeError} * @throws On wrong argument lengths. {@link RangeError} * @example * Validates a fixed-length nonce or key buffer. * * ```ts * abytes(new Uint8Array([1, 2]), 2); * ``` */ export function abytes( value: TArg, length?: number, title: string = '' ): TRet { // Success path first: this runs at the start of every update() / digestInto(), and the // common `abytes(data)` form must not pay for length handling it does not use. if (isBytes(value) && (length === undefined || value.length === length)) return value as TRet; // Error path: recompute freely to build the exact message. if (length !== undefined) anumber(length, 'length'); const bytes = isBytes(value); const ofLen = length !== undefined ? ` of length ${length}` : ''; const got = bytes ? `length=${value.length}` : `type=${typeof value}`; const message = atitle(title) + 'expected Uint8Array' + ofLen + ', got ' + got; if (!bytes) throw new TypeError(message); throw new RangeError(message); } const aobject = (value: Record, label: string) => { if (value === null || typeof value !== 'object' || Array.isArray(value)) throw new TypeError( label === 'object' ? 'expected valid options object' : `"${label}" expected object, got type=${typeof value}` ); }; /** * Asserts a hash- or MAC-like instance has not been destroyed or finished. * @param instance - Stateful instance to validate. * @param checkFinished - Whether to reject finished instances. * When `false`, only `destroyed` is checked. * @throws If the hash instance has already been destroyed or finalized. {@link Error} * @example * Guards against calling `update()` or `digest()` on a finished hash. * * ```ts * aexists({ destroyed: false, finished: false }); * ``` */ export function aexists(instance: any, checkFinished = true): void { // Runs on every update()/digestInto(); the flags are library-owned booleans, so only their // truthiness is checked - re-validating their type per call was pure hot-path overhead. if (instance.destroyed) throw new Error('hash was destroyed'); if (checkFinished && instance.finished) throw new Error('digest() was already called'); } /** * Asserts output is a sufficiently-sized byte array. * @param out - Output buffer to validate. * @param instance - Hash-like instance providing `outputLen`. * This is the relaxed `digestInto()`-style contract: output must be at least `outputLen`, * unlike one-shot cipher helpers elsewhere in the repo that often require exact lengths. * @throws On wrong argument types. {@link TypeError} * @throws On wrong output buffer lengths. {@link RangeError} * @example * Verifies that a caller-provided output buffer is large enough. * * ```ts * aoutput(new Uint8Array(16), { outputLen: 16 }); * ``` */ export function aoutput(out: any, instance: any): void { abytes(out, undefined, 'output'); // `outputLen` is a library-owned readonly number; the negated comparison keeps failing fast // when it is missing/NaN (comparisons with undefined/NaN are false) without an anumber() call. const min = instance.outputLen; if (!(out.length >= min)) { throw new RangeError('"output" expected length >= ' + min); } } /** * Asserts output is a sufficiently-sized, 4-byte-aligned byte array. * {@link aoutput} plus an {@link isAligned32} check, for `digestInto()` paths * that write through zero-allocation `u32` word views. * @param out - Output buffer to validate. * @param instance - Hash-like instance providing `outputLen`. * @throws On wrong argument types. {@link TypeError} * @throws On wrong output buffer lengths. {@link RangeError} * @throws On wrong output buffer alignment. {@link Error} * @example * Verifies that a caller-provided output buffer is large enough and aligned. * * ```ts * aoutput32(new Uint8Array(16), { outputLen: 16 }); * ``` */ export function aoutput32(out: any, instance: any): void { aoutput(out, instance); if (!isAligned32(out)) throw new Error('invalid output, must be aligned'); } /** One-shot hash helper with `.create()`. */ export type IHash = { (data: string | TArg): TRet; /** Input block size in bytes. */ blockLen: number; /** Digest size in bytes. */ outputLen: number; /** Creates a fresh incremental hash instance of the same algorithm. */ create: any; }; /** One-shot MAC helper with `.create()`. */ export type CMac = { (msg: TArg, key: TArg): TRet; /** Input block size in bytes. */ blockLen: number; /** Digest size in bytes. */ outputLen: number; /** * Creates a fresh incremental MAC instance of the same algorithm. * @param key - MAC key bytes. * @param args - Additional constructor arguments, when the MAC wrapper needs them. * @returns Fresh incremental MAC instance. */ create(key: TArg, ...args: A): H; }; /** Generic type encompassing 8/16/32-bit typed arrays, but not 64-bit. */ // prettier-ignore export type TypedArray = Int8Array | Uint8ClampedArray | Uint8Array | Uint16Array | Int16Array | Uint32Array | Int32Array; /** * Casts a typed-array view to Uint8Array. * @param arr - Typed-array view to reinterpret. * @returns Uint8Array view over the same bytes. * @example * Views 32-bit words as raw bytes without copying. * * ```ts * u8(new Uint32Array([1])); * ``` */ export function u8(arr: TArg): TRet { return new Uint8Array(arr.buffer, arr.byteOffset, arr.byteLength) as TRet; } /** * Casts a typed-array view to Uint32Array. * @param arr - Typed-array view to reinterpret. * @returns Uint32Array view over the same bytes. Callers are expected to provide a * 4-byte-aligned offset; trailing `1..3` bytes are silently dropped. * @example * Views a byte buffer as 32-bit words for block processing. * * ```ts * u32(new Uint8Array(4)); * ``` */ export function u32(arr: TArg): TRet { return new Uint32Array( arr.buffer, arr.byteOffset, Math.floor(arr.byteLength / 4) ) as TRet; } /** * Zeroizes typed arrays in place. * Warning: JS provides no guarantees. * @param arrays - Arrays to wipe. * @example * Wipes a temporary key buffer after use. * * ```ts * const bytes = new Uint8Array([1]); * clean(bytes); * ``` */ export function clean(...arrays: TArg): void { for (let i = 0; i < arrays.length; i++) { arrays[i].fill(0); } } /** * Creates a DataView for byte-level manipulation. * @param arr - Typed-array view to wrap. * @returns DataView over the same bytes. * @example * Creates an endian-aware view for length encoding. * * ```ts * createView(new Uint8Array(4)); * ``` */ export function createView(arr: TArg): DataView { return new DataView(arr.buffer, arr.byteOffset, arr.byteLength); } /** * Whether the current platform is little-endian. * Most are; some IBM systems are not. */ export const isLE: boolean = /* @__PURE__ */ (() => new Uint8Array(new Uint32Array([0x11223344]).buffer)[0] === 0x44)(); /** * Reverses byte order of one 32-bit word. * @param word - Unsigned 32-bit word to swap. * @returns The same word with bytes reversed. * @example * Swaps a big-endian word into little-endian byte order. * * ```ts * byteSwap(0x11223344); * ``` */ export function byteSwap(word: number): number { return ( ((word << 24) & 0xff000000) | ((word << 8) & 0xff0000) | ((word >>> 8) & 0xff00) | ((word >>> 24) & 0xff) ); } /** * Normalizes one 32-bit word to the little-endian representation expected by cipher cores. * @param n - Unsigned 32-bit word to normalize. * @returns Little-endian normalized word on big-endian hosts, else the input word unchanged. * @example * Normalizes a host-endian word before passing it into an ARX/AES core. * * ```ts * swap8IfBE(0x11223344); * ``` */ export const swap8IfBE: (n: number) => number = isLE ? (n: number) => n : (n: number) => byteSwap(n) >>> 0; /** * Byte-swaps every word of a Uint32Array in place. * @param arr - Uint32Array whose words should be swapped. * @returns The same array after in-place byte swapping. * @example * Swaps every 32-bit word in a word-view buffer. * * ```ts * byteSwap32(new Uint32Array([0x11223344])); * ``` */ export function byteSwap32(arr: TArg): TRet { for (let i = 0; i < arr.length; i++) { arr[i] = byteSwap(arr[i]); } return arr as TRet; } /** * Normalizes a Uint32Array view to the little-endian representation expected by cipher cores. * @param u - Word view to normalize in place. * @returns Little-endian normalized word view. * @example * Normalizes a word-view buffer before block processing. * * ```ts * swap32IfBE(new Uint32Array([0x11223344])); * ``` */ export const swap32IfBE: (u: TArg) => TRet = isLE ? (u: TArg) => u as TRet : byteSwap32; // Built-in hex conversion: // {@link https://caniuse.com/mdn-javascript_builtins_uint8array_fromhex | caniuse entry} const hasHexBuiltin: boolean = /* @__PURE__ */ (() => // @ts-ignore typeof Uint8Array.from([]).toHex === 'function' && typeof Uint8Array.fromHex === 'function')(); // Array where index 0xf0 (240) is mapped to string 'f0' const hexes = /* @__PURE__ */ Array.from({ length: 256 }, (_, i) => i.toString(16).padStart(2, '0') ); /** * Convert byte array to hex string. Uses built-in function, when available. * @param bytes - Bytes to encode. * @returns Lowercase hexadecimal string. * @throws On wrong argument types. {@link TypeError} * @example * Formats ciphertext bytes for logs or test vectors. * * ```ts * bytesToHex(Uint8Array.from([0xca, 0xfe, 0x01, 0x23])); // 'cafe0123' * ``` */ export function bytesToHex(bytes: TArg): string { abytes(bytes); // @ts-ignore if (hasHexBuiltin) return bytes.toHex(); // pre-caching improves the speed 6x let hex = ''; for (let i = 0; i < bytes.length; i++) { hex += hexes[bytes[i]]; } return hex; } // Strict ASCII nibble parser: non-ASCII hex lookalikes are rejected as undefined. // ASCII codes: '0'..'9' = 48..57, 'A'..'F' = 65..70, 'a'..'f' = 97..102. // prettier-ignore function asciiToBase16(ch: number): number | undefined { return ch >= 48 && ch <= 57 ? ch - 48 // '2' => 50-48 : ch >= 65 && ch <= 70 ? ch - (65 - 10) // 'B' => 66-(65-10) : ch >= 97 && ch <= 102 ? ch - (97 - 10) // 'b' => 98-(97-10) : undefined; } /** * Convert hex string to byte array. Uses built-in function, when available. * @param hex - hexadecimal string to decode * @returns Decoded bytes. * @throws On wrong argument types. {@link TypeError} * @throws On wrong argument ranges or values. {@link RangeError} * @example * Decode lowercase hexadecimal into bytes. * ```ts * hexToBytes('cafe0123'); // Uint8Array.from([0xca, 0xfe, 0x01, 0x23]) * ``` */ export function hexToBytes(hex: string): TRet { if (typeof hex !== 'string') throw new TypeError('hex string expected, got ' + typeof hex); if (hasHexBuiltin) { try { return (Uint8Array as any).fromHex(hex); } catch (error) { if (error instanceof SyntaxError) throw new RangeError(error.message); throw error; } } const hl = hex.length; const al = hl / 2; if (hl % 2) throw new RangeError('hex string expected, got unpadded hex of length ' + hl); const array = new Uint8Array(al); for (let ai = 0, hi = 0; ai < al; ai++, hi += 2) { const n1 = asciiToBase16(hex.charCodeAt(hi)); // parse first char, multiply it by 16 const n2 = asciiToBase16(hex.charCodeAt(hi + 1)); // parse second char if (n1 === undefined || n2 === undefined) { const char = hex[hi] + hex[hi + 1]; throw new RangeError( 'hex string expected, got non-hex character "' + char + '" at index ' + hi ); } array[ai] = n1 * 16 + n2; // example: 'A9' => 10*16 + 9 } return array; } const _0n = /* @__PURE__ */ BigInt(0); // Used in ff1, via bytesToNumberBE /** * Converts a big-endian hex string into bigint. * @param hex - Hexadecimal string without `0x`. * @returns Parsed bigint value. The empty string is treated as `0n`. * @throws On wrong argument types. {@link TypeError} * @example * Parses a big-endian field element or counter from hex. * * ```ts * hexToNumber('ff'); * ``` */ export function hexToNumber(hex: string): bigint { if (typeof hex !== 'string') throw new TypeError('hex string expected, got ' + typeof hex); // Numeric parser, not byte-hex decoder: odd-length forms like 'f' are valid, // and malformed syntax follows BigInt's native error behavior. return hex === '' ? _0n : BigInt('0x' + hex); // Big Endian } // Used in ff1 // BE: Big Endian, LE: Little Endian /** * Converts big-endian bytes into bigint. * @param bytes - Big-endian bytes. * @returns Parsed bigint value. Empty input is treated as `0n`. * @throws On invalid byte input passed to the internal hex conversion. {@link TypeError} * @example * Reads a big-endian integer from serialized bytes. * * ```ts * bytesToNumberBE(new Uint8Array([1, 0])); * ``` */ export function bytesToNumberBE(bytes: TArg): bigint { return hexToNumber(bytesToHex(bytes)); } /** * Validates that a value is a non-negative bigint or safe integer. * @param n - Value to validate. * @returns The same validated value. * @throws On wrong argument ranges or values. {@link RangeError} */ function abignumber(n: T): T { if (typeof n === 'bigint') { if (!(_0n <= n)) throw new RangeError('positive bigint expected, got ' + n); } else anumber(n); return n; } // Used in ff1 /** * Converts a number into big-endian bytes of fixed length. * @param n - Number to encode. * @param len - Output length in bytes. Must be greater than zero. * @returns Big-endian bytes padded to `len`. * Negative values, `len = 0`, and values that do not fit are rejected before * downstream hex parsing. * @throws On wrong argument ranges or values. {@link RangeError} * @throws If a documented runtime validation or state check fails. {@link Error} * @example * Encodes a counter as fixed-width big-endian bytes. * * ```ts * numberToBytesBE(1, 2); * ``` */ export function numberToBytesBE(n: number | bigint, len: number): TRet { anumber(len); if (len === 0) throw new Error('zero output length is invalid'); n = abignumber(n); const expectedLen = len * 2; const hex = n.toString(16); // Detect overflow before hex parsing so oversized values don't leak the shared odd-hex error. if (hex.length > expectedLen) throw new RangeError('number is too large'); return hexToBytes(hex.padStart(expectedLen, '0')); } // Global symbols, but ts doesn't see them: // {@link https://github.com/microsoft/TypeScript/issues/31535 | TypeScript issue 31535} declare const TextEncoder: any; declare const TextDecoder: any; /** * Converts string to bytes using UTF8 encoding. * @param str - String to encode. * @returns UTF-8 bytes in a detached fresh Uint8Array copy. * @throws On wrong argument types. {@link TypeError} * @example * Encodes application text before encryption or MACing. * * ```ts * utf8ToBytes('abc'); // new Uint8Array([97, 98, 99]) * ``` */ export function utf8ToBytes(str: string): TRet { if (typeof str !== 'string') throw new TypeError('string expected'); return new Uint8Array(new TextEncoder().encode(str)) as TRet; // {@link https://bugzil.la/1681809 | Firefox bug 1681809} } /** * Converts bytes to string using UTF8 encoding. * @param bytes - UTF-8 bytes. * @returns Decoded string. Input validation is delegated to `TextDecoder`, and malformed * UTF-8 is replacement-decoded instead of rejected. * @example * Decodes UTF-8 plaintext back into a string. * * ```ts * bytesToUtf8(new Uint8Array([97, 98, 99])); // 'abc' * ``` */ export function bytesToUtf8(bytes: TArg): string { return new TextDecoder().decode(bytes); } /** * Checks if two U8A use same underlying buffer and overlaps. * This is invalid and can corrupt data. * @param a - First byte view. * @param b - Second byte view. * @returns `true` when the views overlap in memory. * @example * Detects whether two slices alias the same backing buffer. * * ```ts * overlapBytes(new Uint8Array(4), new Uint8Array(4)); * ``` */ export function overlapBytes(a: TArg, b: TArg): boolean { // Zero-length views cannot overwrite anything, even if their offset sits inside another range. if (!a.byteLength || !b.byteLength) return false; return ( a.buffer === b.buffer && // best we can do, may fail with an obscure Proxy a.byteOffset < b.byteOffset + b.byteLength && // a starts before b end b.byteOffset < a.byteOffset + a.byteLength // b starts before a end ); } /** * If input and output overlap and input starts before output, we will overwrite end of input before * we start processing it, so this is not supported by forward-processing ciphers. * @param input - Input bytes. * @param output - Output bytes. * @throws If the output view would overwrite unread input bytes. {@link Error} * @example * Rejects an in-place layout that would overwrite unread input bytes. * * ```ts * const buffer = new Uint8Array(8); * complexOverlapBytes(buffer.subarray(0, 4), buffer.subarray(2, 6)); * ``` */ export function complexOverlapBytes(input: TArg, output: TArg): void { // This is very cursed. It works somehow, but I'm completely unsure, // reasoning about overlapping aligned windows is very hard. if (overlapBytes(input, output) && input.byteOffset < output.byteOffset) throw new Error('complex overlap of input and output is not supported'); } /** * Copies several Uint8Arrays into one. * @param arrays - Byte arrays to concatenate. * @returns Combined byte array. * @throws On wrong argument types inside the byte-array list. {@link TypeError} * @example * Builds a `nonce || ciphertext` style buffer. * * ```ts * concatBytes(new Uint8Array([1]), new Uint8Array([2])); * ``` */ export function concatBytes(...arrays: TArg): TRet { let sum = 0; for (let i = 0; i < arrays.length; i++) { const a = arrays[i]; abytes(a); sum += a.length; } const res = new Uint8Array(sum); for (let i = 0, pad = 0; i < arrays.length; i++) { const a = arrays[i]; res.set(a, pad); pad += a.length; } return res as TRet; } // Used in ARX only type EmptyObj = {}; /** * Merges user options into defaults. * @param defaults - Default option values. * @param opts - User-provided overrides. * @returns Combined options object. * `defaults` is a library-owned mutable object; user-provided `opts` only need to be * object-shaped, since "plain object" checks reject valid proxy/cross-realm containers. * The merge mutates `defaults` in place and returns the same object, so direct callers * should pass a fresh defaults object unless they intentionally want shared state updated. * @throws If options are missing or not an object. {@link Error} * @example * Applies user overrides to the default cipher options. * * ```ts * checkOpts({ rounds: 20 }, { rounds: 8 }); * ``` */ export function checkOpts( defaults: T1, opts: T2 ): T1 & T2 { aobject(defaults as Record, 'defaults'); aobject(opts as Record, 'opts'); // Mutates defaults by design. __proto__ follows Object.assign semantics here and can only // affect this local option object; callers already control this low-level options surface. const merged = Object.assign(defaults, opts); return merged as T1 & T2; } /** * Compares two byte arrays in kinda constant time once lengths already match. * @param a - First byte array. * @param b - Second byte array. * @returns `true` when the arrays contain the same bytes. Different lengths still return early. * @example * Compares an expected authentication tag with the received one. * * ```ts * equalBytes(new Uint8Array([1]), new Uint8Array([1])); * ``` */ export function equalBytes(a: TArg, b: TArg): boolean { a = abytes(a); b = abytes(b); if (a.length !== b.length) return false; let diff = 0; for (let i = 0; i < a.length; i++) diff |= a[i] ^ b[i]; return diff === 0; } // TODO: remove /** Incremental hash interface used internally. */ export interface IHash2 { /** Bytes processed per compression block. */ blockLen: number; /** Bytes produced by the final digest. */ outputLen: number; /** * Absorbs one more chunk into the hash state. * @param buf - Data chunk to hash. * @returns The same hash instance for chaining. */ update(buf: string | TArg): this; /** * Writes the final digest into a caller-provided buffer. * @param buf - Destination buffer for the digest bytes. * @returns Nothing. Implementations write into `buf` in place. */ digestInto(buf: TArg): void; /** * Finalizes the hash and returns a fresh digest buffer. * @returns Digest bytes. */ digest(): TRet; /** * Resets internal state. Makes Hash instance unusable. * Reset is impossible for keyed hashes if key is consumed into state. If digest is not consumed * by user, they will need to manually call `destroy()` when zeroing is necessary. */ destroy(): void; } /** * Wraps a keyed MAC constructor into a one-shot helper with `.create()`. * @param keyLen - Valid probe-key length used to read static metadata once. * The probe key is only used for `outputLen` / `blockLen`, so callers with several valid key sizes * can pass any representative size as long as those values stay fixed. * @param macCons - Keyed MAC constructor or factory. * @param fromMsg - Optional adapter that derives extra constructor args from the one-shot message. * @returns Callable MAC helper with `.create()`. */ export function wrapMacConstructor( keyLen: number, macCons: TArg<(key: Uint8Array, ...args: A) => H>, fromMsg?: TArg<(msg: Uint8Array) => A> ): TRet> { const mac = macCons as (key: TArg, ...args: A) => H; const getArgs = (fromMsg || (() => [] as unknown as A)) as (msg: TArg) => A; const macC: any = (msg: TArg, key: TArg): TRet => mac(key, ...getArgs(msg)) .update(msg) .digest(); const tmp = mac(new Uint8Array(keyLen), ...getArgs(new Uint8Array(0))); macC.outputLen = tmp.outputLen; macC.blockLen = tmp.blockLen; macC.create = (key: TArg, ...args: A) => mac(key, ...args); return macC as TRet>; } // This will allow to re-use with composable things like packed & base encoders // Also, we probably can make tags composable /** Sync cipher: takes byte array and returns byte array. */ export type Cipher = { /** * Encrypts plaintext bytes. * @param plaintext - Data to encrypt. * @returns Ciphertext bytes. */ encrypt(plaintext: TArg): TRet; /** * Decrypts ciphertext bytes. * @param ciphertext - Data to decrypt. * @returns Plaintext bytes. */ decrypt(ciphertext: TArg): TRet; }; /** Async cipher e.g. from built-in WebCrypto. */ export type AsyncCipher = { /** * Encrypts plaintext bytes. * @param plaintext - Data to encrypt. * @returns Promise resolving to ciphertext bytes. */ encrypt(plaintext: TArg): Promise>; /** * Decrypts ciphertext bytes. * @param ciphertext - Data to decrypt. * @returns Promise resolving to plaintext bytes. */ decrypt(ciphertext: TArg): Promise>; }; /** Cipher with `output` argument which can optimize by doing 1 less allocation. */ export type CipherWithOutput = Cipher & { /** * Encrypts plaintext bytes into an optional caller-provided buffer. * @param plaintext - Data to encrypt. * @param output - Optional destination buffer. * @returns Ciphertext bytes. */ encrypt(plaintext: TArg, output?: TArg): TRet; /** * Decrypts ciphertext bytes into an optional caller-provided buffer. * @param ciphertext - Data to decrypt. * @param output - Optional destination buffer. * @returns Plaintext bytes. */ decrypt(ciphertext: TArg, output?: TArg): TRet; }; /** * Params are outside of return type, so it is accessible before calling constructor. * If function support multiple nonceLength's, we return the best one. */ export type CipherParams = { /** Cipher block size in bytes. */ blockSize: number; /** Nonce length in bytes when the cipher uses a fixed nonce size. */ nonceLength?: number; /** Authentication-tag length in bytes for AEAD modes. */ tagLength?: number; /** Whether constructor accepts associated data after nonce/options args. */ withAAD?: true; /** Whether nonce length is variable at runtime. */ varSizeNonce?: boolean; }; /** * ARX AEAD cipher, like salsa or chacha. * @param key - Secret key bytes. * @param nonce - Nonce bytes. * @param AAD - Optional associated data. * @returns Cipher instance with caller-managed output buffers. */ export type ARXCipher = (( key: TArg, nonce: TArg, AAD?: TArg ) => CipherWithOutput) & { blockSize: number; nonceLength: number; tagLength: number; }; /** * Cipher constructor signature. * @param key - Secret key bytes. * @param args - Additional constructor arguments, such as nonce or IV. * @returns Cipher instance. */ export type CipherCons = (key: TArg, ...args: T) => Cipher; /** * Wraps a cipher: validates args, ensures encrypt() can only be called once. * Used internally by the exported cipher constructors. * Output-buffer support is inferred from the wrapped `encrypt` / `decrypt` * arity (`fn.length === 2`), so wrapped output-capable methods must use a normal * second parameter, not a default/rest parameter. AAD support is explicit in * `params.withAAD`; optional AAD starts after the nonce slot when one is present. * @__NO_SIDE_EFFECTS__ * @param params - Static cipher metadata. See {@link CipherParams}. * @param constructor - Cipher constructor. * @returns Wrapped constructor with validation. */ export const wrapCipher = , P extends CipherParams>( params: P, constructor: C ): C & P => { function wrappedCipher(key: TArg, ...args: any[]): TRet { // Validate key abytes(key, undefined, 'key'); // Validate nonce if nonceLength is present if (params.nonceLength !== undefined) { const nonce = args[0]; abytes(nonce, params.varSizeNonce ? undefined : params.nonceLength, 'nonce'); } // Keep tag length available for decrypt-size checks after constructor validation. const tagl = params.tagLength; const aadStart = params.nonceLength !== undefined ? 1 : 0; // No-AAD constructors otherwise silently ignore byte args meant as AAD. if (!params.withAAD) { for (let i = aadStart; i < args.length; i++) if (isBytes(args[i])) throw new Error('AAD not supported'); } // Validate the first AAD slot early; rest-arg AAD constructors validate the tail themselves. if (params.withAAD && args[aadStart] !== undefined) abytes(args[aadStart], undefined, 'AAD'); const cipher = constructor(key, ...args); const checkOutput = (fnLength: number, output?: TArg) => { if (output !== undefined) { if (fnLength !== 2) throw new Error('cipher output not supported'); abytes(output, undefined, 'output'); } }; // Create wrapped cipher with validation and single-use encryption let called = false; const wrCipher = { encrypt(data: TArg, output?: TArg) { if (called) throw new Error('cannot encrypt() twice with same key + nonce'); // Any encrypt attempt consumes the instance, even if validation rejects below. called = true; abytes(data, undefined, 'data'); checkOutput(cipher.encrypt.length, output); return (cipher as CipherWithOutput).encrypt(data, output); }, decrypt(data: TArg, output?: TArg) { abytes(data, undefined, 'data'); if (tagl && data.length < tagl) throw new Error('"ciphertext" expected length >= tagLength=' + tagl); checkOutput(cipher.decrypt.length, output); return (cipher as CipherWithOutput).decrypt(data, output); }, }; return wrCipher as TRet; } Object.assign(wrappedCipher, params); return wrappedCipher as C & P; }; /** * Represents a Salsa or ChaCha xor stream. * @param key - Secret key bytes. * @param nonce - Nonce bytes. * @param data - Input bytes to xor with the keystream. * @param output - Optional destination buffer. * @param counter - Optional starting block counter. * @returns Output bytes. */ export type XorStream = ( key: TArg, nonce: TArg, data: TArg, output?: TArg, counter?: number ) => TRet; /** * By default, returns u8a of length. * When out is available, it checks it for validity and uses it. * @param expectedLength - Required output length. * @param out - Optional destination buffer. * @param onlyAligned - Whether `out` must be 4-byte aligned. * @returns Output buffer ready for writing. * @throws On wrong argument types. {@link TypeError} * @throws If the provided output buffer has the wrong size. {@link RangeError} * @throws If the provided output buffer has the wrong alignment. {@link Error} * @example * Reuses a caller-provided output buffer when lengths match. * * ```ts * getOutput(16, new Uint8Array(16)); * ``` */ export function getOutput( expectedLength: number, out?: TArg, onlyAligned = true ): TRet { if (out === undefined) return new Uint8Array(expectedLength) as TRet; // Keep Buffer/cross-realm Uint8Array support here instead of trusting a shape-compatible object. abytes(out, expectedLength, 'output'); if (onlyAligned && !isAligned32(out)) throw new Error('invalid output, must be aligned'); return out as TRet; } /** * Encodes data and AAD lengths into a 16-byte buffer. * @param dataLength - Data length. Units are caller-defined: GCM passes bit * lengths, ChaCha20-Poly1305 passes byte lengths — the helper writes the raw values. * @param aadLength - AAD length, same unit convention as `dataLength`. * The serialized block is still `aadLength || dataLength`, matching GCM/Poly1305 * conventions even though the helper parameter order is `(dataLength, aadLength)`. * @param isLE - Whether to encode lengths as little-endian. * @returns 16-byte length block. * @throws On wrong argument types passed to the endian validator. {@link TypeError} * @throws On wrong argument ranges or values. {@link RangeError} * @example * Builds the length block appended by GCM and Poly1305. * * ```ts * u64Lengths(16, 8, true); * ``` */ export function u64Lengths(dataLength: number, aadLength: number, isLE: boolean): TRet { // Reject coercible non-number lengths like '10' and true before BigInt(...) accepts them. anumber(dataLength); anumber(aadLength); abool(isLE); const num = new Uint8Array(16); const view = createView(num); view.setBigUint64(0, BigInt(aadLength), isLE); view.setBigUint64(8, BigInt(dataLength), isLE); return num as TRet; } /** * Checks whether a byte array is aligned to a 4-byte offset. * @param bytes - Byte array to inspect. * @returns `true` when the view is 4-byte aligned. * @example * Checks whether a buffer can be safely viewed as Uint32Array. * * ```ts * isAligned32(new Uint8Array(4)); * ``` */ export function isAligned32(bytes: TArg): boolean { return bytes.byteOffset % 4 === 0; } /** * Copies bytes into a new Uint8Array. * @param bytes - Bytes to copy. * @returns Copied byte array. * @throws On wrong argument types. {@link TypeError} * @example * Copies input into an aligned Uint8Array before block processing. * * ```ts * copyBytes(new Uint8Array([1, 2])); * ``` */ export function copyBytes(bytes: TArg): TRet { // `Uint8Array.from(...)` would also accept arrays / other typed arrays. Keep this helper strict // because callers use it at byte-validation boundaries before mutating the detached copy. return Uint8Array.from(abytes(bytes)) as TRet; } /** * Cryptographically secure PRNG backed by `crypto.getRandomValues`. * @param bytesLength - number of random bytes to generate * @returns Random bytes. * The platform `getRandomValues()` implementation still defines any * single-call length cap, and this helper rejects oversize requests * with a stable library `RangeError` instead of host-specific errors. * @throws On wrong argument types. {@link TypeError} * @throws On wrong argument ranges or values. {@link RangeError} * @throws If the current runtime does not provide `crypto.getRandomValues`. {@link Error} * @example * Generate a fresh random key or nonce. * ```ts * const key = randomBytes(16); * ``` */ export function randomBytes(bytesLength = 32): TRet { // Match the repo's other length-taking helpers instead of relying on Uint8Array coercion. anumber(bytesLength, 'bytesLength'); const cr = typeof globalThis === 'object' ? (globalThis as any).crypto : null; if (typeof cr?.getRandomValues !== 'function') throw new Error('crypto.getRandomValues must be defined'); // Web Cryptography API Level 2 §10.1.1: // if `byteLength > 65536`, throw `QuotaExceededError`. // Keep the guard explicit so callers can see the quota in code // instead of discovering it by reading the spec or host errors. // This wrapper surfaces the same quota as a stable library RangeError. if (bytesLength > 65536) throw new RangeError(`"bytesLength" expected <= 65536, got ${bytesLength}`); return cr.getRandomValues(new Uint8Array(bytesLength)); } /** * The pseudorandom number generator doesn't wipe current state: * instead, it generates new one based on previous state + entropy. * Not reseed/rekey, since AES CTR DRBG does rekey on each randomBytes, * which is in fact `reseed`, since it changes counter too. */ export interface PRG { /** * Mixes fresh entropy into the current generator state. * @param seed - Entropy bytes to absorb. */ addEntropy(seed: TArg): void; /** * Produces a requested number of pseudorandom bytes. * @param bytesLength - Number of bytes to generate. * @returns Random byte array. */ randomBytes(bytesLength: number): TRet; /** Destroys the generator state. */ clean(): void; } /** Removes the nonce argument from a cipher constructor type. */ export type RemoveNonce any> = T extends ( arg0: any, arg1: any, ...rest: infer R ) => infer Ret ? (key: TArg, ...args: R) => Ret : never; /** * Cipher constructor that requires a nonce argument. * @param key - Secret key bytes. * @param nonce - Nonce bytes. * @param args - Additional cipher-specific arguments. * @returns Cipher instance. */ export type CipherWithNonce = (( key: TArg, nonce: TArg, ...args: any[] ) => Cipher | AsyncCipher) & { nonceLength: number; }; /** * Uses CSPRNG for nonce, nonce injected in ciphertext. * For `encrypt`, a `nonceBytes`-length buffer is fetched from CSPRNG and * prepended to encrypted ciphertext. For `decrypt`, first `nonceBytes` of ciphertext * are treated as nonce. The wrapper always allocates a fresh `nonce || ciphertext` * buffer on encrypt and intentionally does not support caller-provided destination buffers. * Too-short decrypt inputs are split into short/empty nonce views and then delegated * to the wrapped cipher instead of being rejected here first. * * NOTE: Under the same key, using random nonces (e.g. `managedNonce`) with AES-GCM and ChaCha * should be limited to `2**23` (8M) messages to get a collision chance of * `2**-50`. Stretching to `2**32` (4B) messages would raise that chance to * `2**-33`, still negligible but creeping up. * @param fn - Cipher constructor that expects a nonce. * @param randomBytes_ - Random-byte source used for nonce generation. * @returns Cipher constructor that prepends the nonce to ciphertext. * @throws On wrong argument types. {@link TypeError} * @throws On invalid nonce lengths observed at wrapper construction or use. {@link RangeError} * @example * Prepends a fresh random nonce to every ciphertext. * * ```ts * import { gcm } from '@noble/ciphers/aes.js'; * import { managedNonce, randomBytes } from '@noble/ciphers/utils.js'; * const wrapped = managedNonce(gcm); * const key = randomBytes(16); * const ciphertext = wrapped(key).encrypt(new Uint8Array([1, 2, 3])); * wrapped(key).decrypt(ciphertext); * ``` */ export function managedNonce( fn: T, randomBytes_: typeof randomBytes = randomBytes ): TRet> { if (typeof fn !== 'function') throw new TypeError('"fn" expected cipher constructor, got type=' + typeof fn); if (typeof randomBytes_ !== 'function') throw new TypeError('"randomBytes_" expected function, got type=' + typeof randomBytes_); const { nonceLength } = fn; anumber(nonceLength, 'fn.nonceLength'); const addNonce = ( nonce: TArg, ciphertext: TArg, plaintext: TArg ) => { const out = concatBytes(nonce, ciphertext); // Wrapped ciphers may alias caller plaintext on encrypt(); never zero // caller-owned buffers here. if (!overlapBytes(plaintext, ciphertext)) ciphertext.fill(0); return out; }; // NOTE: we cannot support DST here, it would be mistake: // - we don't know how much dst length cipher requires // - nonce may unalign dst and break everything // - we create new u8a anyway (concatBytes) // - previously we passed all args to cipher, but that was mistake! const res = ((key: TArg, ...args: any[]): any => ({ encrypt(plaintext: TArg) { abytes(plaintext, undefined, 'data'); const nonce = randomBytes_(nonceLength); const encrypted = fn(key, nonce, ...args).encrypt(plaintext); // @ts-ignore if (encrypted instanceof Promise) return encrypted.then((ct) => addNonce(nonce, ct, plaintext)); return addNonce(nonce, encrypted, plaintext); }, decrypt(ciphertext: TArg) { abytes(ciphertext, undefined, 'data'); const nonce = ciphertext.subarray(0, nonceLength); const decrypted = ciphertext.subarray(nonceLength); return fn(key, nonce, ...args).decrypt(decrypted); }, })) as RemoveNonce & { blockSize?: number; tagLength?: number; withAAD?: true }; // Auto-nonce wrappers still preserve the wrapped payload geometry. if ('blockSize' in fn) res.blockSize = (fn as any).blockSize; if ('tagLength' in fn) res.tagLength = (fn as any).tagLength; if ('withAAD' in fn) res.withAAD = (fn as any).withAAD; return res as TRet>; } /** `Uint8Array.of()` return type helper for TS 5.9. */ export type Uint8ArrayBuffer = TRet;