encoding

Base64, hex, base32, base58, ascii85, percent-encoding and quoted-printable codecs.

Backend: native — stdlib/encoding.so, built from the Rust crate rust/encoding over base64, hex, data-encoding, bs58, ascii85, percent-encoding and quoted_printable. Import: import("encoding") → global.encoding.*

How bytes travel

The encoded form of a payload is always text and the raw form is always a bytes value: an encode takes bytes and returns a str, a decode takes a str and returns bytes. A payload that is not valid UTF-8 is therefore representable — a decode that used to yield "" now returns those bytes — so a decode failure is an empty bytes. *Valid reports whether the input is well formed in its codec.

Test a decode with bytesLength(result) is 0 rather than expecting an exception: there is no error channel across the native ABI.

Conventions

Padding is optional when decoding and always emitted when encoding, for base64 and base32 alike. base32 decoding also accepts either case.
Whitespace is never accepted inside an encoded payload.
A malformed % escape is a failure rather than a passthrough: the underlying decoder would leave %2 alone, so the module validates the escapes itself.
ascii85 output includes the Adobe <~/~> frame; decoding accepts the frame or a bare stream, ignores whitespace, and understands the z shorthand for four zero bytes.
quotedPrintable is the text encoding of RFC 2045: = is escaped, CR/LF become =0D/=0A, and long lines gain soft breaks that decoding removes.

Functions

Function	Signature	Returns
`base64Encode`	`(bytes b)`	Standard base64, padded
`base64Decode`	`(str s)`	the decoded `bytes`, or empty on failure
`base64Valid`	`(str s)`	`true` if `s` is well-formed standard base64
`base64UrlEncode`	`(bytes b)`	URL-safe base64 (`-`/`_`), padded
`base64UrlDecode`	`(str s)`	the decoded `bytes`, or empty on failure
`base64UrlValid`	`(str s)`	`true` if `s` is well-formed URL-safe base64
`hexEncode`	`(bytes b)`	Lower-case hexadecimal
`hexEncodeUpper`	`(bytes b)`	Upper-case hexadecimal
`hexDecode`	`(str s)`	the decoded `bytes`, or empty on failure
`hexValid`	`(str s)`	`true` if `s` is hexadecimal with an even length
`base32Encode`	`(bytes b)`	Upper-case RFC 4648 base32, padded
`base32Decode`	`(str s)`	the decoded `bytes`, or empty on failure
`base32Valid`	`(str s)`	`true` if `s` is well-formed base32
`base58Encode`	`(bytes b)`	Base58 (the Bitcoin alphabet)
`base58Decode`	`(str s)`	the decoded `bytes`, or empty on failure
`base58Valid`	`(str s)`	`true` if `s` is well-formed base58
`ascii85Encode`	`(bytes b)`	Adobe ascii85, framed with `<~` `~>`
`ascii85Decode`	`(str s)`	the decoded `bytes`, or empty on failure
`ascii85Valid`	`(str s)`	`true` if `s` is well-formed ascii85
`percentEncode`	`(bytes b)`	Percent-encoded URI component
`percentDecode`	`(str s)`	the decoded `bytes`, or empty on failure
`percentValid`	`(str s)`	`true` if every `%` is followed by two hex digits
`quotedPrintableEncode`	`(bytes b)`	Quoted-printable text
`quotedPrintableDecode`	`(str s)`	the decoded `bytes`, or empty on failure
`quotedPrintableValid`	`(str s)`	`true` if `s` is well-formed quoted-printable

percentEncode uses encodeURIComponent semantics: A-Z a-z 0-9 - _ . ! ~ * ' ( ) stay unescaped and everything else, including / ? = &, is escaped.

Example

global setup(){ import("encoding"); }

global main(){
    println(global.encoding.base64Encode("Lynxer"));
    println(global.encoding.base64Decode("THlueA"));   // padding is optional
    println(global.encoding.hexEncode("Lynxer"));
    println(global.encoding.percentEncode("a b/c?d=e&f"));
    println(global.encoding.percentValid("%2"));        // truncated escape
}

encoding

How bytes travel

Conventions

Functions

Example

See also