moonbitlang / core / 6259

91%

push

github

feat(encoding): add hex package (#3625)

Adds moonbitlang/core/encoding/hex for in-memory lowercase hexadecimal
encoding and decoding, mirroring the per-codec granularity of the
existing encoding/{ascii,base64,utf8,utf16} packages.

Public API is deliberately minimal: encode(BytesView) -> String,
decode(StringView) -> Bytes raise Malformed, and the Malformed
suberror. decode accepts both letter cases; encode emits lowercase.
Go's EncodedLen/DecodedLen sizing helpers are intentionally left out:
encode and decode allocate their own results, so the helpers would
have no caller with a real job (the lengths are simply 2 * n and
n / 2, stated in the docs). They can return alongside buffer-writing
variants if those are ever added. Stream APIs are excluded because
core exposes no IO abstraction here.

Refreshed against current main (2026-08-21): negative-index string
slicing in a test replaced with an explicit start offset, and the
Malformed Debug promotion handled with the same deprecated pub extend
pattern encoding/base64 uses.

Codex CLI review (xhigh, two rounds): round 1 caught the encoder's
StringBuilder size_hint being expressed in output characters while the
hint is measured in BYTES on non-JS targets (2n reserved only n UTF-16
slots, forcing a growth-and-copy on every non-empty encode) — fixed to
4 * n with the unit documented; round 2 verified the fix and that
decode's byte-buffer hint was already unit-correct.

Signed-off-by: Codex CLI <codex@openai.com>
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>

19 of 19 new or added lines in 2 files covered. (100.0%)

16521 of 18200 relevant lines covered (90.77%)

307625.39 hits per line