TextEncoder

TextEncoder takes a stream of code points as input and emits a stream of bytes. For a more scalable, non-native library, see StringView – a C-like representation of strings based on typed arrays.

Example

const encoder = new TextEncoder()
const view = encoder.encode('€')
console.log(view); // Uint8Array(3) [226, 130, 172]

Constructor

TextEncoder()

Returns a newly constructed TextEncoder that will generate a byte stream with utf-8 encoding.

Properties

The TextEncoder interface doesn't inherit any property.

TextEncoder.prototype.encodingRead only

Is a DOMString containing the name of the encoder, that is a string describing the method the TextEncoder will use.

Methods

The TextEncoder interface doesn't inherit any method.

TextEncoder.prototype.encode()

Takes a USVString as input, and returns a Uint8Array containing utf-8 encoded text.

TextEncoder.prototype.encodeInto()

Takes a USVString to encode and a destination Uint8Array to put resulting utf-8 encoded text into, and returns a dictionary object indicating the progress of the encoding. This is potentially more performant than the older encode() method.

Polyfill

The below polyfill will only furfill the specs demanded by the W3 (no character encodings other than UTF-8 are supported, unfortunately ☹️). It is designed to work in IE5 "out of the box". However, in IE5-IE9, it will return a regular Array instead of a TypedArray. In such circumstances as these with such memory inefficient slow browsers, this polyfill (or any polyfill for that matter) would be impractical for large strings in such old browsers. Finally, note that you should run the below code through a minifier (especially closure compiler) to turn sequences like 0x1e << 3 into 0xf0. These sequences are not already precomputed because they serve to aesthetically illustrate how the polyfill works.

if (typeof TextEncoder === "undefined") {
    TextEncoder=function TextEncoder(){};
    TextEncoder.prototype.encode = function encode(str) {
        "use strict";
        var Len = str.length, resPos = -1;
        // The Uint8Array's length must be at least 3x the length of the string because an invalid UTF-16
        //  takes up the equivelent space of 3 UTF-8 characters to encode it properly. However, Array's
        //  have an auto expanding length and 1.5x should be just the right balance for most uses.
        var resArr = typeof Uint8Array === "undefined" ? new Array(Len * 1.5) : new Uint8Array(Len * 3);
        for (var point=0, nextcode=0, i = 0; i !== Len; ) {
            point = str.charCodeAt(i), i += 1;
            if (point >= 0xD800 && point <= 0xDBFF) {
                if (i === Len) {
                    resArr[resPos += 1] = 0xef/*0b11101111*/; resArr[resPos += 1] = 0xbf/*0b10111111*/;
                    resArr[resPos += 1] = 0xbd/*0b10111101*/; break;
                }
                // https://mathiasbynens.be/notes/javascript-encoding#surrogate-formulae
                nextcode = str.charCodeAt(i);
                if (nextcode >= 0xDC00 && nextcode <= 0xDFFF) {
                    point = (point - 0xD800) * 0x400 + nextcode - 0xDC00 + 0x10000;
                    i += 1;
                    if (point > 0xffff) {
                        resArr[resPos += 1] = (0x1e/*0b11110*/<<3) | (point>>>18);
                        resArr[resPos += 1] = (0x2/*0b10*/<<6) | ((point>>>12)&0x3f/*0b00111111*/);
                        resArr[resPos += 1] = (0x2/*0b10*/<<6) | ((point>>>6)&0x3f/*0b00111111*/);
                        resArr[resPos += 1] = (0x2/*0b10*/<<6) | (point&0x3f/*0b00111111*/);
                        continue;
                    }
                } else {
                    resArr[resPos += 1] = 0xef/*0b11101111*/; resArr[resPos += 1] = 0xbf/*0b10111111*/;
                    resArr[resPos += 1] = 0xbd/*0b10111101*/; continue;
                }
            }
            if (point <= 0x007f) {
                resArr[resPos += 1] = (0x0/*0b0*/<<7) | point;
            } else if (point <= 0x07ff) {
                resArr[resPos += 1] = (0x6/*0b110*/<<5) | (point>>>6);
                resArr[resPos += 1] = (0x2/*0b10*/<<6)  | (point&0x3f/*0b00111111*/);
            } else {
                resArr[resPos += 1] = (0xe/*0b1110*/<<4) | (point>>>12);
                resArr[resPos += 1] = (0x2/*0b10*/<<6)    | ((point>>>6)&0x3f/*0b00111111*/);
                resArr[resPos += 1] = (0x2/*0b10*/<<6)    | (point&0x3f/*0b00111111*/);
            }
        }
        if (typeof Uint8Array !== "undefined") return resArr.subarray(0, resPos + 1);
        // else // IE 6-9
        resArr.length = resPos + 1; // trim off extra weight
        return resArr;
    };
    TextEncoder.prototype.toString = function(){return "[object TextEncoder]"};
    try { // Object.defineProperty only works on DOM prototypes in IE8
        Object.defineProperty(TextEncoder.prototype,"encoding",{
            get:function(){if(TextEncoder.prototype.isPrototypeOf(this)) return"utf-8";
                           else throw TypeError("Illegal invocation");}
        });
    } catch(e) { /*IE6-8 fallback*/ TextEncoder.prototype.encoding = "utf-8"; }
    if(typeof Symbol!=="undefined")TextEncoder.prototype[Symbol.toStringTag]="TextEncoder";
}

^{Source: https://github.com/anonyco/FastestSmallestTextEncoderDecoder}

Specifications

Browser compatibility

See also

• The TextDecoder interface describing the inverse operation.
• StringView – a C-like representation of strings based on typed arrays
• A shim allowing to use this interface in browsers that don't support it.
• Components.utils.importGlobalProperties
• Node.js supports global export from v11.0.0