/**
 * @license
 * Copyright The Closure Library Authors.
 * SPDX-License-Identifier: Apache-2.0
 */
/**
 * @fileoverview Abstract cryptographic hash interface.
 *
 * See goog.crypt.Sha1 and goog.crypt.Md5 for sample implementations.
 */
/**
 * Create a cryptographic hash instance.
 *
 * @class
 * @abstract
 */
export class Hash {
    /**
     * The block size for the hasher.
     * @type {number}
     */
    blockSize: number;
    /**
     * Resets the internal accumulator.
     * @abstract
     */
    reset(): void;
    /**
     * Adds a byte array (array with values in [0-255] range) or a string (must
     * only contain 8-bit, i.e., Latin1 characters) to the internal accumulator.
     *
     * Many hash functions operate on blocks of data and implement optimizations
     * when a full chunk of data is readily available. Hence it is often preferable
     * to provide large chunks of data (a kilobyte or more) than to repeatedly
     * call the update method with few tens of bytes. If this is not possible, or
     * not feasible, it might be good to provide data in multiplies of hash block
     * size (often 64 bytes). Please see the implementation and performance tests
     * of your favourite hash.
     *
     * @param {Array<number>|Uint8Array|string} bytes Data used for the update.
     * @param {number=} opt_length Number of bytes to use.
     * @abstract
     */
    update(bytes: Array<number> | Uint8Array | string, opt_length?: number | undefined): void;
    /**
     * @return {!Array<number>} The finalized hash computed
     *     from the internal accumulator.
     * @abstract
     */
    digest(): Array<number>;
}