4.23.9 Multibase Encoding

Import :std/encoding/multibase.

4.23.9.1 Validation

(multibase-valid? str [start = 0] [end = (string-length str)]) returns #t when str[start,end) has a supported prefix and a valid canonical payload, otherwise #f. The prefix is the character at start.

Prefix	Encoding	Payload Validation
`z`	Base58 Bitcoin	`base58-valid?` with the Bitcoin alphabet
`u`	Base64 URL-safe, unpadded	URL-safe alphabet; padding forbidden
`U`	Base64 URL-safe, padded	URL-safe alphabet; required padding
`m`	Base64, unpadded	Standard alphabet; padding forbidden
`M`	Base64, padded	Standard alphabet; required padding

Base64 validation uses base64-decoded-length with the corresponding padding: and urlsafe: flags. It rejects whitespace, junk, incorrect padding, invalid lengths, and nonzero unused pad bits. Complete four-character groups need no padding in either mode. A successful length, including zero, becomes #t.

A prefix alone (z, u, U, m, or M) is valid: its payload is empty. An empty substring or unknown prefix is invalid. Validation does not allocate a substring or decoded bytes.

str must be a string; start and end must be fixnums satisfying 0 <= start <= end <= (string-length str). Invalid argument types or bounds raise ContractViolation.

There is no multibase-decoded-length API: exact Base58 decoded sizing requires numeric conversion, unlike the linear alphabet scan used here. Callers needing the bytes can decode after validation.

4.23.9.2 Encoding And Decoding

(multibase-encode encoding bytes) returns a prefixed string; encoding is one of the prefix characters above. The corresponding constants are multibase-base58-btc, multibase-base64-url, multibase-base64-url-pad, multibase-base64, and multibase-base64-pad.

(multibase-decode str) returns a u8vector. Its existing decoder semantics are unchanged; in particular, validation is stricter than the legacy Base64 decoder. Use multibase-valid? when canonical input is required. Missing or unsupported prefixes raise ContractViolation when decoding.