@k/encoding

Install

$ npm install @k/encoding

---

UTF-8

For simple UTF-8 encoding and decoding, there are two exported methods.

import { utf8_to_ui8, ui8_to_utf8 } from '@k/encoding';

utf8_to_ui8('Hello');  // Uint8Array<48 65 6c 6c 6f>

ui8_to_utf8([ 0x48, 0x65, 0x6c, 0x6c, 0x6f ]);  // "Hello"

Number Conversions

There are a large number of methods for converting various number types to/from byte arrays.

import { i32_to_ui8, ui8_to_i32 } from '@k/encoding';

i32_to_ui8(100000000);  // [ 5, 245, 225, 0 ]

ui8_to_i32([ 5, 245, 225, 0 ]);  // 100000000

The supported number types that can be converted are as follows:

• i8
• i16
• ui16
• i32
• ui32
• i64
• ui64
• f32
• f64

The main functions (like i32_to_ui8) force all results to use a big-endian byte order. If you need to control the byte order yourself, see below.

Avoiding Copies

To avoid needless array allocations and copies, there are also direct-write versions of all the number _to_ui8 methods that will write directly into an array you pass in, as opposed to returning a new array. These methods instead return the new write index after the value is written (ie. if you write a 4-byte type, the return result will be the index you passed in, plus 4).

import { write_i32_as_ui8 } from '@k/encoding';

let index = 0;
const yourArray = [ ];

// Write the bytes for 100000000 into `yourArray`, starting at index 0
index = write_i32_as_ui8(100000000, yourArray, index);

Byte Order

If you need to check if the system uses little or big endian byte order, there are exported constants to tell you.

import { isLittleEndian, isBigEndian } from '@k/encoding';

All of the number conversion functions also have "forward" and "backward" versions (ie. i32_to_ui8 also has i32_to_ui8_forward and i32_to_ui8_backward counterparts). To use the systems byte order, use the "forward" methods, and to use the oposite, use the "backward" methods.