Skip to content

encoding.md

Latest commit

 

History

History
130 lines (91 loc) · 3.18 KB

encoding.md

File metadata and controls

130 lines (91 loc) · 3.18 KB
Raw
id stdlib-encoding
title Encoding Module
slug /stdlib/encoding

These functions are pure .NET runtime externs. Hosts may register PureExterns.functions without exposing IO; the CLI registers them by default. --no-default-externs removes them. JavaScript hosts must supply corresponding externs themselves.

Bytes are represented as int list; every element must be in 0..255. Functions return Some result on success and None for invalid content. Inputs are never read from files, and no system encoding or current culture is consulted.

Encoding.utf8Encode

Encoding.utf8Encode : string -> int list option

Encodes text as UTF-8 bytes without a BOM. Unpaired UTF-16 surrogates return None.

Encoding.utf8Encode "é"
// Some [195;169]

Encoding.utf8Decode

Encoding.utf8Decode : int list -> string option

Strictly decodes UTF-8. Invalid byte values or malformed UTF-8 return None. A supplied BOM is preserved as U+FEFF.

Encoding.utf8Decode [195;169]
// Some "é"

Encoding.base64Encode

Encoding.base64Encode : int list -> string option

Encodes bytes with the standard Base64 alphabet and padding, without line breaks.

Encoding.base64Encode [97;98;99]
// Some "YWJj"

Encoding.base64Decode

Encoding.base64Decode : string -> int list option

Decodes standard padded Base64; .NET Base64 whitespace is accepted. Invalid input returns None.

Encoding.base64Decode "YWJj"
// Some [97;98;99]

Encoding.hexEncode

Encoding.hexEncode : int list -> string option

Encodes bytes as lowercase hexadecimal, with two digits per byte.

Encoding.hexEncode [0;171;255]
// Some "00abff"

Encoding.hexDecode

Encoding.hexDecode : string -> int list option

Decodes even-length hexadecimal; either case is accepted. Whitespace, odd lengths, and invalid digits return None.

Encoding.hexDecode "00AbFF"
// Some [0;171;255]

Encoding.uriEncode

Encoding.uriEncode : string -> string option

Percent-encodes a URI component using UTF-8, leaving only RFC 3986 unreserved characters unchanged. Invalid UTF-16 returns None.

Encoding.uriEncode "a +/é"
// Some "a%20%2B%2F%C3%A9"

Encoding.uriDecode

Encoding.uriDecode : string -> string option

Decodes percent escapes and strictly validates UTF-8. Invalid escapes or UTF-8 return None. A plus sign remains a plus sign; this is not form decoding.

Encoding.uriDecode "a%20%2B%2F%C3%A9"
// Some "a +/é"

Register only pure content helpers

A .NET embedding host can expose all Encoding functions and the new SHA-256 functions without constructing an IO-capable registry:

open FScript.Language
open FScript.Runtime

let registration = HostInterop.ofExterns PureExterns.functions
let result =
    Parser.parseProgram "Hash.sha256 \"hello\""
    |> TypeInfer.inferProgramWithRegistration registration
    |> Eval.evalProgramWithRegistration registration

Core scalar math, String, List, Map, Option, Path, Glob, and SemVer helpers need no registration. The scalar math functions live in Int and Float.