Changelog¶

All notable changes to dash-types and dash-types-marker will be documented in this file.

The format is based on Keep a Changelog, and this crate adheres to Semantic Versioning.

Unreleased ¶

Implementation of LowerHex, UpperHex, FromStr, From<&[u8; N]> and TryFrom<&[u8]> for {derive,make}_bytes!.
ParseHexError is moved from dash-num, superseding dash_num::ParseHexError.
serialize::hex::serialize_as and serialize::hex::deserialize_as, to permit caller-defined handling of human-readable encoding while sharing common machine-readable encoding.

{derive,make}_bytes!-defined types serialize as their raw storage bytes for machine-readable formats. JSON and other human-readable formats retain hex in storage order for fwd and in reverse storage order for rev.
derive_bytes! emits From<[u8; N]> in place of impl_bytes!, and make_sbytes! emits the implementation itself. Callers of impl_bytes! without derive_bytes! must now write the conversion manually.
- As a result, both make_bytes! and make_sbytes! carry it with nocodec tag and without the codec feature and callers that wrote the conversion manually for those configurations must drop it.
The zeroize export is moved from the crate root to __deps. Callers must switch to dash_types::__deps::zeroize.
Types using derive_bytes! that implemented From<&[u8; N]>, TryFrom<&[u8]>, FromStr, LowerHex or UpperHex manually must drop those implementations, as derive_bytes! now emits them.
serialize::hex writes raw bytes for machine-readable formats instead of emitting a hex-encoded string.
serialize::hex::deserialize returns any T built from a byte slice instead of Vec<u8>. Callers that relied on Vec<u8> as the fixed return type must now mention it explicitly.
serialize::str_u64 writes a native u64 to machine-readable formats. str_u64 is a workaround for number-precision limitations in JSON and is now contained only for human-readable formats.

The codec::Checkable and codec::Hashable compatibility aliases in favour of exports from the crate root.

serialize::utf8_lossy's serializer and deserializer arms were not in sync, writing valid UTF-8 as a string for machine-readable formats but expecting byte buffers when reading, failing round trips. This has since been resolved.