pub trait Hash {
// Required method
fn hash<H>(&self, state: &mut H)
where H: Hasher;
// Provided method
fn hash_slice<H>(data: &[Self], state: &mut H)
where H: Hasher,
Self: Sized { ... }
}
Expand description
A hashable type.
Types implementing Hash
are able to be hash
ed with an instance of
Hasher
.
§Implementing Hash
You can derive Hash
with #[derive(Hash)]
if all fields implement Hash
.
The resulting hash will be the combination of the values from calling
hash
on each field.
#[derive(Hash)]
struct Rustacean {
name: String,
country: String,
}
If you need more control over how a value is hashed, you can of course
implement the Hash
trait yourself:
use std::hash::{Hash, Hasher};
struct Person {
id: u32,
name: String,
phone: u64,
}
impl Hash for Person {
fn hash<H: Hasher>(&self, state: &mut H) {
self.id.hash(state);
self.phone.hash(state);
}
}
§Hash
and Eq
When implementing both Hash
and Eq
, it is important that the following
property holds:
k1 == k2 -> hash(k1) == hash(k2)
In other words, if two keys are equal, their hashes must also be equal.
HashMap
and HashSet
both rely on this behavior.
Thankfully, you won’t need to worry about upholding this property when
deriving both Eq
and Hash
with #[derive(PartialEq, Eq, Hash)]
.
Violating this property is a logic error. The behavior resulting from a logic error is not
specified, but users of the trait must ensure that such logic errors do not result in
undefined behavior. This means that unsafe
code must not rely on the correctness of these
methods.
§Prefix collisions
Implementations of hash
should ensure that the data they
pass to the Hasher
are prefix-free. That is,
values which are not equal should cause two different sequences of values to be written,
and neither of the two sequences should be a prefix of the other.
For example, the standard implementation of Hash
for &str
passes an extra
0xFF
byte to the Hasher
so that the values ("ab", "c")
and ("a", "bc")
hash differently.
§Portability
Due to differences in endianness and type sizes, data fed by Hash
to a Hasher
should not be considered portable across platforms. Additionally the data passed by most
standard library types should not be considered stable between compiler versions.
This means tests shouldn’t probe hard-coded hash values or data fed to a Hasher
and
instead should check consistency with Eq
.
Serialization formats intended to be portable between platforms or compiler versions should
either avoid encoding hashes or only rely on Hash
and Hasher
implementations that
provide additional guarantees.
Required Methods§
Provided Methods§
1.3.0 · Sourcefn hash_slice<H>(data: &[Self], state: &mut H)
fn hash_slice<H>(data: &[Self], state: &mut H)
Feeds a slice of this type into the given Hasher
.
This method is meant as a convenience, but its implementation is
also explicitly left unspecified. It isn’t guaranteed to be
equivalent to repeated calls of hash
and implementations of
Hash
should keep that in mind and call hash
themselves
if the slice isn’t treated as a whole unit in the PartialEq
implementation.
For example, a VecDeque
implementation might naïvely call
as_slices
and then hash_slice
on each slice, but this
is wrong since the two slices can change with a call to
make_contiguous
without affecting the PartialEq
result. Since these slices aren’t treated as singular
units, and instead part of a larger deque, this method cannot
be used.
§Examples
use std::hash::{DefaultHasher, Hash, Hasher};
let mut hasher = DefaultHasher::new();
let numbers = [6, 28, 496, 8128];
Hash::hash_slice(&numbers, &mut hasher);
println!("Hash is {:x}!", hasher.finish());
Dyn Compatibility§
This trait is not dyn compatible.
In older versions of Rust, dyn compatibility was called "object safety", so this trait is not object safe.
Implementors§
impl Hash for std_prelude::AtomicOrdering
impl Hash for std_prelude::Ordering
impl Hash for AsciiChar
impl Hash for Infallible
impl Hash for IpAddr
impl Hash for Ipv6MulticastScope
impl Hash for SocketAddr
impl Hash for ErrorKind
impl Hash for bool
impl Hash for char
impl Hash for i8
impl Hash for i16
impl Hash for i32
impl Hash for i64
impl Hash for i128
impl Hash for isize
impl Hash for !
impl Hash for str
impl Hash for u8
impl Hash for u16
impl Hash for u32
impl Hash for u64
impl Hash for u128
impl Hash for ()
impl Hash for usize
impl Hash for CString
impl Hash for String
impl Hash for Layout
impl Hash for TypeId
impl Hash for CStr
impl Hash for Error
impl Hash for PhantomPinned
impl Hash for Ipv4Addr
impl Hash for Ipv6Addr
impl Hash for SocketAddrV4
impl Hash for SocketAddrV6
impl Hash for RangeFull
impl Hash for Alignment
impl Hash for OsStr
impl Hash for FileType
impl Hash for UCred
impl Hash for PrefixComponent<'_>
impl Hash for ThreadId
impl Hash for Instant
impl Hash for SystemTime
impl Hash for Duration
impl Hash for OsString
impl Hash for Path
impl Hash for PathBuf
impl<'a> Hash for Component<'a>
impl<'a> Hash for Prefix<'a>
impl<'a> Hash for Location<'a>
impl<B> Hash for Cow<'_, B>
impl<B, C> Hash for ControlFlow<B, C>
impl<Dyn> Hash for DynMetadata<Dyn>where
Dyn: ?Sized,
impl<F> Hash for Fwhere
F: FnPtr,
impl<Idx> Hash for core::ops::range::Range<Idx>where
Idx: Hash,
impl<Idx> Hash for core::ops::range::RangeFrom<Idx>where
Idx: Hash,
impl<Idx> Hash for core::ops::range::RangeInclusive<Idx>where
Idx: Hash,
impl<Idx> Hash for RangeTo<Idx>where
Idx: Hash,
impl<Idx> Hash for RangeToInclusive<Idx>where
Idx: Hash,
impl<Idx> Hash for core::range::Range<Idx>where
Idx: Hash,
impl<Idx> Hash for core::range::RangeFrom<Idx>where
Idx: Hash,
impl<Idx> Hash for core::range::RangeInclusive<Idx>where
Idx: Hash,
impl<K, V, A> Hash for BTreeMap<K, V, A>
impl<Ptr> Hash for Pin<Ptr>
impl<T> Hash for Bound<T>where
T: Hash,
impl<T> Hash for Option<T>where
T: Hash,
impl<T> Hash for Poll<T>where
T: Hash,
impl<T> Hash for *const Twhere
T: ?Sized,
impl<T> Hash for *mut Twhere
T: ?Sized,
impl<T> Hash for &T
impl<T> Hash for &mut T
impl<T> Hash for [T]where
T: Hash,
impl<T> Hash for (T₁, T₂, …, Tₙ)
This trait is implemented for tuples up to twelve items long.
impl<T> Hash for Reverse<T>where
T: Hash,
impl<T> Hash for PhantomData<T>where
T: ?Sized,
impl<T> Hash for ManuallyDrop<T>
impl<T> Hash for Discriminant<T>
impl<T> Hash for NonZero<T>where
T: ZeroablePrimitive + Hash,
impl<T> Hash for Saturating<T>where
T: Hash,
impl<T> Hash for Wrapping<T>where
T: Hash,
impl<T> Hash for NonNull<T>where
T: ?Sized,
impl<T, A> Hash for Box<T, A>
impl<T, A> Hash for BTreeSet<T, A>
impl<T, A> Hash for LinkedList<T, A>
impl<T, A> Hash for VecDeque<T, A>
impl<T, A> Hash for Vec<T, A>
The hash of a vector is the same as that of the corresponding slice,
as required by the core::borrow::Borrow
implementation.
use std::hash::BuildHasher;
let b = std::hash::RandomState::new();
let v: Vec<u8> = vec![0xa8, 0x3c, 0x09];
let s: &[u8] = &[0xa8, 0x3c, 0x09];
assert_eq!(b.hash_one(v), b.hash_one(s));
impl<T, A> Hash for Arc<T, A>
impl<T, A> Hash for Rc<T, A>
impl<T, E> Hash for Result<T, E>
impl<T, const N: usize> Hash for [T; N]where
T: Hash,
The hash of an array is the same as that of the corresponding slice,
as required by the Borrow
implementation.
use std::hash::BuildHasher;
let b = std::hash::RandomState::new();
let a: [u8; 3] = [0xa8, 0x3c, 0x09];
let s: &[u8] = &[0xa8, 0x3c, 0x09];
assert_eq!(b.hash_one(a), b.hash_one(s));