Decode According to the WHATWG Encoding Standard

This package provides a thin layer on top of iconv-lite which makes it expose some of the same primitives as the Encoding Standard.

1const whatwgEncoding = require("whatwg-encoding");
2
3console.assert(whatwgEncoding.labelToName("latin1") === "windows-1252");
4console.assert(whatwgEncoding.labelToName("  CYRILLic ") === "ISO-8859-5");
5
6console.assert(whatwgEncoding.isSupported("IBM866") === true);
7
8// Not supported by the Encoding Standard
9console.assert(whatwgEncoding.isSupported("UTF-32") === false);
10
11// In the Encoding Standard, but this package can't decode it
12console.assert(whatwgEncoding.isSupported("x-mac-cyrillic") === false);
13
14console.assert(whatwgEncoding.getBOMEncoding(new Uint8Array([0xFE, 0xFF])) === "UTF-16BE");
15console.assert(whatwgEncoding.getBOMEncoding(new Uint8Array([0x48, 0x69])) === null);
16
17console.assert(whatwgEncoding.decode(new Uint8Array([0x48, 0x69]), "UTF-8") === "Hi");

API

• decode(uint8Array, fallbackEncodingName): performs the decode algorithm (in which any BOM will override the passed fallback encoding), and returns the resulting string
• labelToName(label): performs the get an encoding algorithm and returns the resulting encoding's name, or null for failure
• isSupported(name): returns whether the encoding is one of the encodings of the Encoding Standard, and is an encoding that this package can decode (via iconv-lite)
• getBOMEncoding(uint8Array): sniffs the first 2–3 bytes of the supplied Uint8Array, returning one of the encoding names "UTF-8", "UTF-16LE", or "UTF-16BE" if the appropriate BOM is present, or null if no BOM is present

Unsupported encodings

Since we rely on iconv-lite, we are limited to support only the encodings that they support. Currently we are missing support for:

• ISO-2022-JP
• ISO-8859-8-I
• replacement
• x-mac-cyrillic
• x-user-defined

Passing these encoding names will return false when calling isSupported, and passing any of the possible labels for these encodings to labelToName will return null.

Credits

This package was originally based on the excellent work of @nicolashenry, in jsdom. It has since been pulled out into this separate package.

Alternatives

If you are looking for a JavaScript implementation of the Encoding Standard's TextEncoder and TextDecoder APIs, you'll want @inexorabletash's text-encoding package. Node.js also has them built-in.