std::base64
Base64 (RFC 4648): bytes written as text in a 64-letter alphabet, for
carrying binary data where only text fits (JSON fields, HTTP headers,
data URLs). base64.encode uses the standard alphabet (+ and /);
base64.encode_url the URL and file name safe one (- and _). Both
pad with = to a multiple of four characters; padded: False leaves it
off, as tokens such as JWTs do:
token = base64.encode_url(signature, padded: False)
Each decoder reads its alphabet with or without = padding and returns
an Err naming the first byte it cannot read. Every function here is
pure, so a typed literal’s handler may call them.
Import with import std::base64.
base64.encode
Section titled “base64.encode”fn encode(data: Bytes, padded: Bool = True) -> String
data in base64’s standard alphabet, padded with = to a multiple of
four characters, or with no padding when padded is False.
import std::base64
// RFC 4648, section 10.
assert base64.encode(String.to_bytes("")) == ""
assert base64.encode(String.to_bytes("f")) == "Zg=="
assert base64.encode(String.to_bytes("fo")) == "Zm8="
assert base64.encode(String.to_bytes("foo")) == "Zm9v"
assert base64.encode(String.to_bytes("foob")) == "Zm9vYg=="
assert base64.encode(String.to_bytes("fooba")) == "Zm9vYmE="
assert base64.encode(String.to_bytes("foobar")) == "Zm9vYmFy"
assert base64.encode(String.to_bytes("f"), padded: False) == "Zg"
assert base64.encode(String.to_bytes("foo"), padded: False) == "Zm9v"
base64.decode
Section titled “base64.decode”fn decode(text: String) -> Result<Bytes, String>
The bytes text holds in base64’s standard alphabet, with or without
= padding.
import std::base64
assert base64.decode("Zm9vYmE=") == Ok(String.to_bytes("fooba"))
assert base64.decode("Zm9vYmE") == Ok(String.to_bytes("fooba"))
assert base64.decode("Zm9v!mE=") == Err("invalid base64 at byte 4")
base64.encode_url
Section titled “base64.encode_url”fn encode_url(data: Bytes, padded: Bool = True) -> String
data in base64’s URL and file name safe alphabet, - and _ in place
of + and /, padded with =, or with no padding when padded is
False, the form JWTs and many other tokens use.
import std::base64
data = Bytes.from_list([251, 255, 191])
assert base64.encode(data) == "+/+/"
assert base64.encode_url(data) == "-_-_"
assert base64.encode_url(String.to_bytes("f")) == "Zg=="
assert base64.encode_url(Bytes.from_list([251, 255, 191, 102]), padded: False) == "-_-_Zg"
base64.decode_url
Section titled “base64.decode_url”fn decode_url(text: String) -> Result<Bytes, String>
The bytes text holds in base64’s URL-safe alphabet, with or without
= padding.
import std::base64
assert base64.decode_url("-_-_") == Ok(Bytes.from_list([251, 255, 191]))
assert base64.decode_url("Zg") == Ok(String.to_bytes("f"))
assert base64.decode_url("+/+/") == Err("invalid base64 at byte 0")