@nberlette/utf8/esm/text_encoder.js

Blazing fast universal ponyfills for TextEncoder and TextDecoder.

/**
 * This module provides a high performance dependency-free ponyfill for the
 * native `TextEncoder` Web API, allowing you to encode strings into UTF-8
 * encoded `Uint8Array` buffers in any ES2015+ environment.
 *
 * @module text-encoder
 */
import { getCodePoint, Uint8Array, Uint8ArrayPrototypeSubarray, utf8BytesNeeded, } from "./_internal.js";
/**
 * Encode strings into binary data (in the form of a `Uint8Array`) using the
 * UTF-8 encoding standard. This is a high performance ponyfill for the native
 * `TextEncoder` class.
 *
 * @category Encoding
 * @tags utf-8, encoder
 */
export class TextEncoder {
    /**
     * The encoding standard to use. This is always `"utf-8"`.
     * @returns "utf-8"
     */
    get encoding() {
        return "utf-8";
    }
    /**
     * Encodes a string into a `Uint8Array` with UTF-8 encoding.
     *
     * @param input The string to encode. Defaults to an empty string.
     * @returns A `Uint8Array` containing the UTF-8 encoded bytes.
     */
    encode(input = "") {
        // speculatively allocate 4 B per character and trim the result later
        const buffer = new Uint8Array(input.length * 4);
        const result = this.encodeInto(input, buffer);
        return Uint8ArrayPrototypeSubarray(buffer, 0, result.written);
    }
    /**
     * Encodes a string into a provided Uint8Array using UTF-8 encoding.
     *
     * @param input The string to encode.
     * @param output The Uint8Array to write the encoded bytes into.
     * @returns Object containing the number of characters read and bytes written
     */
    encodeInto(input, output) {
        let read = 0, written = 0;
        for (let i = 0; i < input.length; i++) {
            const codePoint = getCodePoint(input, i);
            if (codePoint > 0xffff)
                i++; // handle surrogate pairs
            const bytesNeeded = utf8BytesNeeded(codePoint);
            if (written + bytesNeeded > output.length)
                break;
            if (codePoint <= 0x7f) {
                output[written++] = codePoint;
            }
            else if (codePoint <= 0x7ff) {
                output[written++] = 0xc0 | (codePoint >> 6);
                output[written++] = 0x80 | (codePoint & 0x3f);
            }
            else if (codePoint <= 0xffff) {
                output[written++] = 0xe0 | (codePoint >> 12);
                output[written++] = 0x80 | ((codePoint >> 6) & 0x3f);
                output[written++] = 0x80 | (codePoint & 0x3f);
            }
            else {
                output[written++] = 0xf0 | (codePoint >> 18);
                output[written++] = 0x80 | ((codePoint >> 12) & 0x3f);
                output[written++] = 0x80 | ((codePoint >> 6) & 0x3f);
                output[written++] = 0x80 | (codePoint & 0x3f);
            }
            read++;
        }
        return { read, written };
    }
}
//# sourceMappingURL=text_encoder.js.map