656 lines
24 KiB
Rust
656 lines
24 KiB
Rust
// SPDX-License-Identifier: CC0-1.0
|
|
|
|
//! Public macros.
|
|
//!
|
|
//! - [`sha256t_tag`](crate::sha256t_tag)
|
|
//! - [`hash_newtype`](crate::hash_newtype)
|
|
//! - [`impl_hex_for_newtype`](crate::impl_hex_for_newtype)
|
|
//! - [`impl_serde_for_newtype`](crate::impl_serde_for_newtype)
|
|
|
|
/// Macro used to define a tag.
|
|
///
|
|
/// Defines new struct and implements `Tag` for it.
|
|
///
|
|
/// The syntax is:
|
|
///
|
|
/// ```
|
|
/// # use bitcoin_hashes::sha256t_tag;
|
|
/// sha256t_tag! {
|
|
/// /// Optional documentation details here.
|
|
/// /// Summary is always generated.
|
|
/// pub struct FooTag = hash_str("foo");
|
|
/// }
|
|
/// ```
|
|
///
|
|
/// The `hash_str` marker says the midstate should be generated by hashing the supplied string in a
|
|
/// way described in BIP-341. Alternatively, you can supply `hash_bytes` to hash raw bytes. If you
|
|
/// have the midstate already pre-computed and prefer **compiler** performance to readability you
|
|
/// may use `raw(MIDSTATE_BYTES, HASHED_BYTES_LENGTH)` instead, note that HASHED_BYTES_LENGTH must
|
|
/// be a multiple of 64.
|
|
#[macro_export]
|
|
macro_rules! sha256t_tag {
|
|
($(#[$($tag_attr:tt)*])* $tag_vis:vis struct $tag:ident = $constructor:tt($($tag_value:tt)+);) => {
|
|
$crate::sha256t_tag_struct!($tag_vis, $tag, stringify!($hash_name), $(#[$($tag_attr)*])*);
|
|
|
|
impl $crate::sha256t::Tag for $tag {
|
|
const MIDSTATE: $crate::sha256::Midstate = $crate::sha256t_tag_constructor!($constructor, $($tag_value)+);
|
|
}
|
|
}
|
|
}
|
|
|
|
/// Creates a new newtype around a [`Hash`] type.
|
|
///
|
|
/// The syntax is similar to the usual tuple struct syntax:
|
|
///
|
|
/// ```
|
|
/// # use bitcoin_hashes::{hash_newtype, sha256};
|
|
/// hash_newtype! {
|
|
/// /// Hash of `Foo`.
|
|
/// pub struct MyNewtype(pub sha256::Hash);
|
|
/// }
|
|
/// ```
|
|
///
|
|
/// You can use any valid visibility specifier in place of `pub` or you can omit either or both, if
|
|
/// you want the type or its field to be private.
|
|
///
|
|
/// Whether the hash is reversed or not when displaying depends on the inner type. However you can
|
|
/// override it like this:
|
|
///
|
|
/// ```
|
|
/// # use bitcoin_hashes::{hash_newtype, sha256};
|
|
/// hash_newtype! {
|
|
/// #[hash_newtype(backward)]
|
|
/// struct MyNewtype(sha256::Hash);
|
|
/// }
|
|
/// ```
|
|
///
|
|
/// This will display the hash backwards regardless of what the inner type does. Use `forward`
|
|
/// instead of `backward` to force displaying forward.
|
|
///
|
|
/// You can add arbitrary doc comments or other attributes to the struct or it's field. Note that
|
|
/// the macro already derives [`Copy`], [`Clone`], [`Eq`], [`PartialEq`],
|
|
/// [`Hash`](core::hash::Hash), [`Ord`], [`PartialOrd`].
|
|
///
|
|
/// You can also define multiple newtypes within one macro call:
|
|
///
|
|
/// ```
|
|
/// # use bitcoin_hashes::{hash_newtype, sha256, hash160};
|
|
///
|
|
/// hash_newtype! {
|
|
/// /// My custom type 1
|
|
/// pub struct Newtype1(sha256::Hash);
|
|
///
|
|
/// /// My custom type 2
|
|
/// struct Newtype2(hash160::Hash);
|
|
/// }
|
|
/// ```
|
|
///
|
|
/// Note: the macro is internally recursive. If you use too many attributes (> 256 tokens) you may
|
|
/// hit recursion limit. If you have so many attributes for a good reason, just raising the limit
|
|
/// should be OK. Note however that attribute-processing part has to use [TT muncher] which has
|
|
/// quadratic complexity, so having many attributes may blow up compile time. This should be rare.
|
|
///
|
|
/// [TT muncher]: https://danielkeep.github.io/tlborm/book/pat-incremental-tt-munchers.html
|
|
///
|
|
// Ever heard of legendary comments warning developers to not touch the code? Yep, here's another
|
|
// one. The following code is written the way it is for some specific reasons. If you think you can
|
|
// simplify it, I suggest spending your time elsewhere.
|
|
//
|
|
// If you looks at the code carefully you might ask these questions:
|
|
//
|
|
// * Why are attributes using `tt` and not `meta`?!
|
|
// * Why are the macros split like that?!
|
|
// * Why use recursion instead of `$()*`?
|
|
//
|
|
// None of these are here by accident. For some reason unknown to me, if you accept an argument to
|
|
// macro with any fragment specifier other than `tt` it will **not** match any of the rules
|
|
// requiring a specific token. Yep, I tried it, I literally got error that `hash_newtype` doesn't
|
|
// match `hash_newtype`. So all input attributes must be `tt`.
|
|
//
|
|
// Originally I wanted to define a bunch of macros that would filter-out hash_type attributes. Then
|
|
// I remembered (by seeing compiler error) that calling macros is not allowed inside attributes.
|
|
// And no, you can't bypass it by calling a helper macro and passing "output of another macro" into
|
|
// it. The whole macro gets passed, not the resulting value. So we have to generate the entire
|
|
// attributes. And you can't just place an attribute-producing macro above struct - they are
|
|
// considered separate items. This is not C.
|
|
//
|
|
// Thus struct is generated in a separate macro together with attributes. And since the macro needs
|
|
// attributes as the input and I didn't want to create confusion by using `#[]` syntax *after*
|
|
// struct, I opted to use `{}` as a separator. Yes, a separator is required because an attribute
|
|
// may be composed of multiple token trees - that's the point of "double repetition".
|
|
#[macro_export]
|
|
macro_rules! hash_newtype {
|
|
($($(#[$($type_attrs:tt)*])* $type_vis:vis struct $newtype:ident($(#[$field_attrs:tt])* $field_vis:vis $hash:path);)+) => {
|
|
$(
|
|
$($crate::hash_newtype_known_attrs!(#[ $($type_attrs)* ]);)*
|
|
|
|
$crate::hash_newtype_struct! {
|
|
$type_vis struct $newtype($(#[$field_attrs])* $field_vis $hash);
|
|
|
|
$({ $($type_attrs)* })*
|
|
}
|
|
|
|
$crate::impl_bytelike_traits!($newtype, { <$newtype as $crate::Hash>::LEN });
|
|
|
|
#[allow(unused)] // Private wrapper types may not need all functions.
|
|
impl $newtype {
|
|
/// Constructs a new hash from the underlying byte array.
|
|
pub const fn from_byte_array(bytes: <$hash as $crate::Hash>::Bytes) -> Self {
|
|
$newtype(<$hash>::from_byte_array(bytes))
|
|
}
|
|
|
|
/// Copies a byte slice into a hash object.
|
|
#[deprecated(since = "0.15.0", note = "use `from_byte_array` instead")]
|
|
#[allow(deprecated_in_future)] // Because of `FromSliceError`.
|
|
#[allow(deprecated)] // Because of `from_slice`.
|
|
pub fn from_slice(sl: &[u8]) -> $crate::_export::_core::result::Result<$newtype, $crate::FromSliceError> {
|
|
Ok($newtype(<$hash as $crate::Hash>::from_slice(sl)?))
|
|
}
|
|
|
|
/// Returns the underlying byte array.
|
|
pub const fn to_byte_array(self) -> <$hash as $crate::Hash>::Bytes {
|
|
self.0.to_byte_array()
|
|
}
|
|
|
|
/// Returns a reference to the underlying byte array.
|
|
pub const fn as_byte_array(&self) -> &<$hash as $crate::Hash>::Bytes {
|
|
self.0.as_byte_array()
|
|
}
|
|
}
|
|
|
|
impl $crate::Hash for $newtype {
|
|
type Bytes = <$hash as $crate::Hash>::Bytes;
|
|
|
|
const DISPLAY_BACKWARD: bool = $crate::hash_newtype_get_direction!($hash, $(#[$($type_attrs)*])*);
|
|
|
|
fn from_byte_array(bytes: Self::Bytes) -> Self { Self::from_byte_array(bytes) }
|
|
|
|
#[inline]
|
|
#[allow(deprecated_in_future)] // Because of `FromSliceError`.
|
|
#[allow(deprecated)] // Because of `from_slice`.
|
|
fn from_slice(sl: &[u8]) -> $crate::_export::_core::result::Result<$newtype, $crate::FromSliceError> {
|
|
Self::from_slice(sl)
|
|
}
|
|
|
|
fn to_byte_array(self) -> Self::Bytes { self.to_byte_array() }
|
|
|
|
fn as_byte_array(&self) -> &Self::Bytes { self.as_byte_array() }
|
|
}
|
|
)+
|
|
};
|
|
}
|
|
|
|
/// Implements string functions using hex for a new type created with [`crate::hash_newtype`] macro.
|
|
///
|
|
/// Implements:
|
|
///
|
|
/// * `str::FromStr`
|
|
/// * `fmt::{LowerHex, UpperHex}` using `hex-conservative`
|
|
/// * `fmt::{Display, Debug}` by calling `LowerHex`
|
|
#[macro_export]
|
|
#[cfg(feature = "hex")]
|
|
macro_rules! impl_hex_for_newtype {
|
|
($($newtype:ident),*) => {
|
|
$(
|
|
$crate::impl_hex_string_traits!($newtype, { <$newtype as $crate::Hash>::LEN }, { <$newtype as $crate::Hash>::DISPLAY_BACKWARD });
|
|
)*
|
|
}
|
|
}
|
|
|
|
/// Implements `fmt::Debug` using hex for a new type created with [`crate::hash_newtype`] macro.
|
|
///
|
|
/// This is provided in case you do not want to use the `hex` feature.
|
|
#[macro_export]
|
|
macro_rules! impl_debug_only_for_newtype {
|
|
($($newtype:ident),*) => {
|
|
$(
|
|
$crate::impl_debug_only!($newtype, { <$newtype as $crate::Hash>::LEN }, { <$newtype as $crate::Hash>::DISPLAY_BACKWARD });
|
|
)*
|
|
}
|
|
}
|
|
|
|
/// Adds trait impls to a bytelike type.
|
|
///
|
|
/// Implements:
|
|
///
|
|
/// * `AsRef[u8; $len]`
|
|
/// * `AsRef[u8]`
|
|
/// * `Borrow<[u8; $len]>`
|
|
/// * `Borrow<[u8]>`
|
|
///
|
|
/// # Parameters
|
|
///
|
|
/// * `ty` - The bytelike type to implement the traits on.
|
|
/// * `$len` - The number of bytes this type has.
|
|
/// * `$gen: $gent` - generic type(s) and trait bound(s).
|
|
#[doc(hidden)]
|
|
#[macro_export]
|
|
macro_rules! impl_bytelike_traits {
|
|
($ty:ident, $len:expr $(, $gen:ident: $gent:ident)*) => {
|
|
impl<$($gen: $gent),*> $crate::_export::_core::convert::AsRef<[u8; { $len }]> for $ty<$($gen),*> {
|
|
#[inline]
|
|
fn as_ref(&self) -> &[u8; { $len }] { self.as_byte_array() }
|
|
}
|
|
|
|
impl<$($gen: $gent),*> $crate::_export::_core::convert::AsRef<[u8]> for $ty<$($gen),*> {
|
|
#[inline]
|
|
fn as_ref(&self) -> &[u8] { self.as_byte_array() }
|
|
}
|
|
|
|
impl<$($gen: $gent),*> $crate::_export::_core::borrow::Borrow<[u8; { $len }]> for $ty<$($gen),*> {
|
|
fn borrow(&self) -> &[u8; { $len }] { self.as_byte_array() }
|
|
}
|
|
|
|
impl<$($gen: $gent),*> $crate::_export::_core::borrow::Borrow<[u8]> for $ty<$($gen),*> {
|
|
fn borrow(&self) -> &[u8] { self.as_byte_array() }
|
|
}
|
|
}
|
|
}
|
|
|
|
/// Adds hex string trait impls to a bytelike type using hex.
|
|
///
|
|
/// Implements:
|
|
///
|
|
/// * `str::FromStr`
|
|
/// * `fmt::{LowerHex, UpperHex}` using `hex-conservative`.
|
|
/// * `fmt::{Display, Debug}` by calling `LowerHex`
|
|
///
|
|
/// Requires:
|
|
///
|
|
/// * [`hex-conservative`] to publicly available as `$crate::hex`.
|
|
/// * `$ty` must implement `IntoIterator<Item=Borrow<u8>>`.
|
|
///
|
|
/// (See also [`hex-conservative::fmt_hex_exact`].)
|
|
///
|
|
/// # Parameters
|
|
///
|
|
/// * `ty` - The bytelike type to implement the traits on.
|
|
/// * `$len` - The number of bytes this type has.
|
|
/// * `$reverse` - `true` if the type should be displayed backwards, `false` otherwise.
|
|
/// * `$gen: $gent` - generic type(s) and trait bound(s).
|
|
///
|
|
/// [`hex-conservative`]: <https://crates.io/crates/hex-conservative>
|
|
#[doc(hidden)]
|
|
#[macro_export]
|
|
#[cfg(feature = "hex")]
|
|
macro_rules! impl_hex_string_traits {
|
|
($ty:ident, $len:expr, $reverse:expr $(, $gen:ident: $gent:ident)*) => {
|
|
impl<$($gen: $gent),*> $crate::_export::_core::str::FromStr for $ty<$($gen),*> {
|
|
type Err = $crate::hex::HexToArrayError;
|
|
|
|
fn from_str(s: &str) -> $crate::_export::_core::result::Result<Self, Self::Err> {
|
|
use $crate::hex::FromHex;
|
|
|
|
let mut bytes = <[u8; { $len }]>::from_hex(s)?;
|
|
if $reverse {
|
|
bytes.reverse();
|
|
}
|
|
Ok(Self::from_byte_array(bytes))
|
|
}
|
|
}
|
|
|
|
$crate::hex::impl_fmt_traits! {
|
|
#[display_backward($reverse)]
|
|
impl<$($gen: $gent),*> fmt_traits for $ty<$($gen),*> {
|
|
const LENGTH: usize = ($len); // parens required due to rustc parser weirdness
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
/// Implements `fmt::Debug` using hex.
|
|
#[doc(hidden)]
|
|
#[macro_export]
|
|
macro_rules! impl_debug_only {
|
|
($ty:ident, $len:expr, $reverse:expr $(, $gen:ident: $gent:ident)*) => {
|
|
impl<$($gen: $gent),*> $crate::_export::_core::fmt::Debug for $ty<$($gen),*> {
|
|
#[inline]
|
|
fn fmt(&self, f: &mut $crate::_export::_core::fmt::Formatter) -> $crate::_export::_core::fmt::Result {
|
|
$crate::debug_hex(self.as_byte_array(), f)
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
// Generates the struct only (no impls)
|
|
//
|
|
// This is a separate macro to make it more readable and have a separate interface that allows for
|
|
// two groups of type attributes: processed and not-yet-processed ones (think about it like
|
|
// computation via recursion). The macro recursively matches unprocessed attributes, popping them
|
|
// one at a time and either ignoring them (`hash_newtype`) or appending them to the list of
|
|
// processed attributes to be added to the struct.
|
|
//
|
|
// Once the list of not-yet-processed attributes is empty the struct is generated with processed
|
|
// attributes added.
|
|
#[doc(hidden)]
|
|
#[macro_export]
|
|
macro_rules! hash_newtype_struct {
|
|
($(#[$other_attrs:meta])* $type_vis:vis struct $newtype:ident($(#[$field_attrs:meta])* $field_vis:vis $hash:path);) => {
|
|
$(#[$other_attrs])*
|
|
#[derive(Copy, Clone, PartialEq, Eq, PartialOrd, Ord, Hash)]
|
|
$type_vis struct $newtype($(#[$field_attrs])* $field_vis $hash);
|
|
};
|
|
($(#[$other_attrs:meta])* $type_vis:vis struct $newtype:ident($(#[$field_attrs:meta])* $field_vis:vis $hash:path); { hash_newtype($($ignore:tt)*) } $($type_attrs:tt)*) => {
|
|
$crate::hash_newtype_struct! {
|
|
$(#[$other_attrs])*
|
|
$type_vis struct $newtype($(#[$field_attrs])* $field_vis $hash);
|
|
|
|
$($type_attrs)*
|
|
}
|
|
};
|
|
($(#[$other_attrs:meta])* $type_vis:vis struct $newtype:ident($(#[$field_attrs:meta])* $field_vis:vis $hash:path); { $other_attr:meta } $($type_attrs:tt)*) => {
|
|
$crate::hash_newtype_struct! {
|
|
$(#[$other_attrs])*
|
|
#[$other_attr]
|
|
$type_vis struct $newtype($(#[$field_attrs])* $field_vis $hash);
|
|
|
|
$($type_attrs)*
|
|
}
|
|
};
|
|
}
|
|
|
|
// Extracts `hash_newtype(forward)` and `hash_newtype(backward)` attributes if any and turns them
|
|
// into bool, defaulting to `DISPLAY_BACKWARD` of the wrapped type if the attribute is omitted.
|
|
//
|
|
// Once an appropriate attribute is found we pass the remaining ones into another macro to detect
|
|
// duplicates/conflicts and report an error.
|
|
//
|
|
// FYI, no, we can't use a helper macro to first filter all `hash_newtype` attributes. We would be
|
|
// attempting to match on macros instead. So we must write `hashe_newtype` in each branch.
|
|
#[doc(hidden)]
|
|
#[macro_export]
|
|
macro_rules! hash_newtype_get_direction {
|
|
($hash:ty, ) => { <$hash as $crate::Hash>::DISPLAY_BACKWARD };
|
|
($hash:ty, #[hash_newtype(forward)] $($others:tt)*) => { { $crate::hash_newtype_forbid_direction!(forward, $($others)*); false } };
|
|
($hash:ty, #[hash_newtype(backward)] $($others:tt)*) => { { $crate::hash_newtype_forbid_direction!(backward, $($others)*); true } };
|
|
($hash:ty, #[$($ignore:tt)*] $($others:tt)*) => { $crate::hash_newtype_get_direction!($hash, $($others)*) };
|
|
}
|
|
|
|
// Reports an error if any of the attributes is `hash_newtype($direction)`.
|
|
//
|
|
// This is used for detection of duplicates/conflicts, see the macro above.
|
|
#[doc(hidden)]
|
|
#[macro_export]
|
|
macro_rules! hash_newtype_forbid_direction {
|
|
($direction:ident, ) => {};
|
|
($direction:ident, #[hash_newtype(forward)] $(others:tt)*) => {
|
|
compile_error!(concat!("cannot set display direction to forward: ", stringify!($direction), " was already specified"));
|
|
};
|
|
($direction:ident, #[hash_newtype(backward)] $(others:tt)*) => {
|
|
compile_error!(concat!("cannot set display direction to backward: ", stringify!($direction), " was already specified"));
|
|
};
|
|
($direction:ident, #[$($ignore:tt)*] $(#[$others:tt])*) => {
|
|
$crate::hash_newtype_forbid_direction!($direction, $(#[$others])*)
|
|
};
|
|
}
|
|
|
|
// Checks (at compile time) that all `hash_newtype` attributes are known.
|
|
//
|
|
// An unknown attribute could be a typo that could cause problems - e.g. wrong display direction if
|
|
// it's missing. To prevent this, we call this macro above. The macro produces nothing unless an
|
|
// unknown attribute is found in which case it produces `compile_error!`.
|
|
#[doc(hidden)]
|
|
#[macro_export]
|
|
macro_rules! hash_newtype_known_attrs {
|
|
(#[hash_newtype(forward)]) => {};
|
|
(#[hash_newtype(backward)]) => {};
|
|
(#[hash_newtype($($unknown:tt)*)]) => { compile_error!(concat!("unrecognized attribute ", stringify!($($unknown)*))); };
|
|
($($ignore:tt)*) => {};
|
|
}
|
|
|
|
/// Functions used by serde impls of all hashes.
|
|
#[cfg(feature = "serde")]
|
|
pub mod serde_details {
|
|
use core::marker::PhantomData;
|
|
use core::str::FromStr;
|
|
use core::{fmt, str};
|
|
|
|
use serde::de;
|
|
|
|
/// Type used to implement serde traits for hashes as hex strings.
|
|
pub struct HexVisitor<ValueT>(PhantomData<ValueT>);
|
|
|
|
impl<ValueT> Default for HexVisitor<ValueT> {
|
|
fn default() -> Self { Self(PhantomData) }
|
|
}
|
|
|
|
impl<ValueT> de::Visitor<'_> for HexVisitor<ValueT>
|
|
where
|
|
ValueT: FromStr,
|
|
<ValueT as FromStr>::Err: fmt::Display,
|
|
{
|
|
type Value = ValueT;
|
|
|
|
fn expecting(&self, formatter: &mut fmt::Formatter) -> fmt::Result {
|
|
formatter.write_str("an ASCII hex string")
|
|
}
|
|
|
|
fn visit_bytes<E>(self, v: &[u8]) -> core::result::Result<Self::Value, E>
|
|
where
|
|
E: de::Error,
|
|
{
|
|
if let Ok(hex) = str::from_utf8(v) {
|
|
hex.parse::<Self::Value>().map_err(E::custom)
|
|
} else {
|
|
Err(E::invalid_value(de::Unexpected::Bytes(v), &self))
|
|
}
|
|
}
|
|
|
|
fn visit_str<E>(self, v: &str) -> core::result::Result<Self::Value, E>
|
|
where
|
|
E: de::Error,
|
|
{
|
|
v.parse::<Self::Value>().map_err(E::custom)
|
|
}
|
|
}
|
|
|
|
/// Type used to implement serde traits for hashes as bytes.
|
|
pub struct BytesVisitor<ValueT, const N: usize>(PhantomData<ValueT>);
|
|
|
|
impl<ValueT, const N: usize> Default for BytesVisitor<ValueT, N> {
|
|
fn default() -> Self { Self(PhantomData) }
|
|
}
|
|
|
|
impl<ValueT, const N: usize> de::Visitor<'_> for BytesVisitor<ValueT, N>
|
|
where
|
|
ValueT: crate::Hash,
|
|
ValueT: crate::Hash<Bytes = [u8; N]>,
|
|
{
|
|
type Value = ValueT;
|
|
|
|
fn expecting(&self, formatter: &mut fmt::Formatter) -> fmt::Result {
|
|
formatter.write_str("a bytestring")
|
|
}
|
|
|
|
fn visit_bytes<E>(self, v: &[u8]) -> core::result::Result<Self::Value, E>
|
|
where
|
|
E: de::Error,
|
|
{
|
|
let bytes = <[u8; N]>::try_from(v).map_err(|_| {
|
|
// from_slice only errors on incorrect length
|
|
E::invalid_length(v.len(), &stringify!(N))
|
|
})?;
|
|
|
|
Ok(<Self::Value as crate::Hash>::from_byte_array(bytes))
|
|
}
|
|
}
|
|
}
|
|
|
|
/// Implements `Serialize` and `Deserialize` for a new type created with [`crate::hash_newtype`] macro.
|
|
#[macro_export]
|
|
#[cfg(feature = "serde")]
|
|
macro_rules! impl_serde_for_newtype {
|
|
($($newtype:ident),*) => {
|
|
$(
|
|
$crate::serde_impl!($newtype, { <$newtype as $crate::Hash>::LEN });
|
|
)*
|
|
}
|
|
}
|
|
|
|
/// Implements `Serialize` and `Deserialize` for a type `$t` which
|
|
/// represents a newtype over a byte-slice over length `$len`.
|
|
#[doc(hidden)]
|
|
#[macro_export]
|
|
#[cfg(feature = "serde")]
|
|
macro_rules! serde_impl(
|
|
($t:ident, $len:expr $(, $gen:ident: $gent:ident)*) => (
|
|
impl<$($gen: $gent),*> $crate::serde::Serialize for $t<$($gen),*> {
|
|
fn serialize<S: $crate::serde::Serializer>(&self, s: S) -> core::result::Result<S::Ok, S::Error> {
|
|
if s.is_human_readable() {
|
|
s.collect_str(self)
|
|
} else {
|
|
s.serialize_bytes(<Self as $crate::Hash>::as_byte_array(self))
|
|
}
|
|
}
|
|
}
|
|
|
|
impl<'de $(, $gen: $gent)*> $crate::serde::Deserialize<'de> for $t<$($gen),*> {
|
|
fn deserialize<D: $crate::serde::Deserializer<'de>>(d: D) -> core::result::Result<$t<$($gen),*>, D::Error> {
|
|
use $crate::serde_macros::serde_details::{BytesVisitor, HexVisitor};
|
|
|
|
if d.is_human_readable() {
|
|
d.deserialize_str(HexVisitor::<Self>::default())
|
|
} else {
|
|
d.deserialize_bytes(BytesVisitor::<Self, $len>::default())
|
|
}
|
|
}
|
|
}
|
|
));
|
|
|
|
/// Does an "empty" serde implementation for the configuration without serde feature.
|
|
#[doc(hidden)]
|
|
#[macro_export]
|
|
#[cfg(not(feature = "serde"))]
|
|
macro_rules! serde_impl(
|
|
($t:ident, $len:expr $(, $gen:ident: $gent:ident)*) => ()
|
|
);
|
|
|
|
#[cfg(test)]
|
|
mod test {
|
|
use crate::sha256;
|
|
|
|
#[test]
|
|
fn hash_as_ref_array() {
|
|
let hash = sha256::Hash::hash(&[3, 50]);
|
|
let r = AsRef::<[u8; 32]>::as_ref(&hash);
|
|
assert_eq!(r, hash.as_byte_array());
|
|
}
|
|
|
|
#[test]
|
|
fn hash_as_ref_slice() {
|
|
let hash = sha256::Hash::hash(&[3, 50]);
|
|
let r = AsRef::<[u8]>::as_ref(&hash);
|
|
assert_eq!(r, hash.as_byte_array());
|
|
}
|
|
|
|
#[test]
|
|
fn hash_borrow() {
|
|
use core::borrow::Borrow;
|
|
|
|
let hash = sha256::Hash::hash(&[3, 50]);
|
|
let borrowed: &[u8] = hash.borrow();
|
|
assert_eq!(borrowed, hash.as_byte_array());
|
|
}
|
|
|
|
hash_newtype! {
|
|
/// Test hash.
|
|
struct TestHash(crate::sha256d::Hash);
|
|
}
|
|
#[cfg(feature = "hex")]
|
|
crate::impl_hex_for_newtype!(TestHash);
|
|
#[cfg(not(feature = "hex"))]
|
|
crate::impl_debug_only_for_newtype!(TestHash);
|
|
|
|
impl TestHash {
|
|
fn all_zeros() -> Self { Self::from_byte_array([0; 32]) }
|
|
}
|
|
|
|
#[test]
|
|
fn macros_work_in_function_scope() {
|
|
use crate::sha256t;
|
|
|
|
sha256t_tag! {
|
|
#[repr(align(2))] // This tests that we can add additional attributes.
|
|
pub struct FunctionScopeTag = hash_str("It works");
|
|
}
|
|
|
|
hash_newtype! {
|
|
/// Some docs.
|
|
#[repr(align(4))] // This tests that we can add additional attributes.
|
|
pub struct FunctionScopeHash(pub(crate) sha256t::Hash<FunctionScopeTag>);
|
|
}
|
|
|
|
assert_eq!(2, core::mem::align_of::<FunctionScopeTag>());
|
|
assert_eq!(4, core::mem::align_of::<FunctionScopeHash>());
|
|
}
|
|
|
|
// NB: This runs with and without `hex` feature enabled, testing different code paths for each.
|
|
#[test]
|
|
#[cfg(feature = "alloc")]
|
|
fn debug() {
|
|
use alloc::format;
|
|
|
|
let want = "0000000000000000000000000000000000000000000000000000000000000000";
|
|
let got = format!("{:?}", TestHash::all_zeros());
|
|
assert_eq!(got, want)
|
|
}
|
|
|
|
#[test]
|
|
#[cfg(feature = "alloc")]
|
|
#[cfg(feature = "hex")]
|
|
fn display() {
|
|
use alloc::format;
|
|
|
|
let want = "0000000000000000000000000000000000000000000000000000000000000000";
|
|
let got = format!("{}", TestHash::all_zeros());
|
|
assert_eq!(got, want)
|
|
}
|
|
|
|
#[test]
|
|
#[cfg(feature = "alloc")]
|
|
#[cfg(feature = "hex")]
|
|
fn display_alternate() {
|
|
use alloc::format;
|
|
|
|
let want = "0x0000000000000000000000000000000000000000000000000000000000000000";
|
|
let got = format!("{:#}", TestHash::all_zeros());
|
|
assert_eq!(got, want)
|
|
}
|
|
|
|
#[test]
|
|
#[cfg(feature = "alloc")]
|
|
#[cfg(feature = "hex")]
|
|
fn lower_hex() {
|
|
use alloc::format;
|
|
|
|
let want = "0000000000000000000000000000000000000000000000000000000000000000";
|
|
let got = format!("{:x}", TestHash::all_zeros());
|
|
assert_eq!(got, want)
|
|
}
|
|
|
|
#[test]
|
|
#[cfg(feature = "alloc")]
|
|
#[cfg(feature = "hex")]
|
|
fn lower_hex_alternate() {
|
|
use alloc::format;
|
|
|
|
let want = "0x0000000000000000000000000000000000000000000000000000000000000000";
|
|
let got = format!("{:#x}", TestHash::all_zeros());
|
|
assert_eq!(got, want)
|
|
}
|
|
|
|
#[test]
|
|
fn inner_hash_as_ref_array() {
|
|
let hash = TestHash::all_zeros();
|
|
let r = AsRef::<[u8; 32]>::as_ref(&hash);
|
|
assert_eq!(r, hash.as_byte_array());
|
|
}
|
|
|
|
#[test]
|
|
fn inner_hash_as_ref_slice() {
|
|
let hash = TestHash::all_zeros();
|
|
let r = AsRef::<[u8]>::as_ref(&hash);
|
|
assert_eq!(r, hash.as_byte_array());
|
|
}
|
|
}
|