pub trait Ord: Eq + PartialOrd {
// Required method
fn cmp(&self, other: &Self) -> Ordering;
// Provided methods
fn max(self, other: Self) -> Self
where Self: Sized { ... }
fn min(self, other: Self) -> Self
where Self: Sized { ... }
fn clamp(self, min: Self, max: Self) -> Self
where Self: Sized { ... }
}
Expand description
Trait for types that form a total order.
Implementations must be consistent with the PartialOrd
implementation, and ensure max
,
min
, and clamp
are consistent with cmp
:
partial_cmp(a, b) == Some(cmp(a, b))
.max(a, b) == max_by(a, b, cmp)
(ensured by the default implementation).min(a, b) == min_by(a, b, cmp)
(ensured by the default implementation).- For
a.clamp(min, max)
, see the method docs (ensured by the default implementation).
Violating these requirements 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.
ยงCorollaries
From the above and the requirements of PartialOrd
, it follows that for all a
, b
and c
:
- exactly one of
a < b
,a == b
ora > b
is true; and <
is transitive:a < b
andb < c
impliesa < c
. The same must hold for both==
and>
.
Mathematically speaking, the <
operator defines a strict weak order. In cases where ==
conforms to mathematical equality, it also defines a strict total order.
ยงDerivable
This trait can be used with #[derive]
.
When derive
d on structs, it will produce a
lexicographic ordering based on the
top-to-bottom declaration order of the structโs members.
When derive
d on enums, variants are ordered primarily by their discriminants. Secondarily,
they are ordered by their fields. By default, the discriminant is smallest for variants at the
top, and largest for variants at the bottom. Hereโs an example:
#[derive(PartialEq, Eq, PartialOrd, Ord)]
enum E {
Top,
Bottom,
}
assert!(E::Top < E::Bottom);
However, manually setting the discriminants can override this default behavior:
#[derive(PartialEq, Eq, PartialOrd, Ord)]
enum E {
Top = 2,
Bottom = 1,
}
assert!(E::Bottom < E::Top);
ยงLexicographical comparison
Lexicographical comparison is an operation with the following properties:
- Two sequences are compared element by element.
- The first mismatching element defines which sequence is lexicographically less or greater than the other.
- If one sequence is a prefix of another, the shorter sequence is lexicographically less than the other.
- If two sequences have equivalent elements and are of the same length, then the sequences are lexicographically equal.
- An empty sequence is lexicographically less than any non-empty sequence.
- Two empty sequences are lexicographically equal.
ยงHow can I implement Ord
?
Ord
requires that the type also be PartialOrd
, PartialEq
, and Eq
.
Because Ord
implies a stronger ordering relationship than PartialOrd
, and both Ord
and
PartialOrd
must agree, you must choose how to implement Ord
first. You can choose to
derive it, or implement it manually. If you derive it, you should derive all four traits. If you
implement it manually, you should manually implement all four traits, based on the
implementation of Ord
.
Hereโs an example where you want to define the Character
comparison by health
and
experience
only, disregarding the field mana
:
use std::cmp::Ordering;
struct Character {
health: u32,
experience: u32,
mana: f32,
}
impl Ord for Character {
fn cmp(&self, other: &Self) -> Ordering {
self.experience
.cmp(&other.experience)
.then(self.health.cmp(&other.health))
}
}
impl PartialOrd for Character {
fn partial_cmp(&self, other: &Self) -> Option<Ordering> {
Some(self.cmp(other))
}
}
impl PartialEq for Character {
fn eq(&self, other: &Self) -> bool {
self.health == other.health && self.experience == other.experience
}
}
impl Eq for Character {}
If all you need is to slice::sort
a type by a field value, it can be simpler to use
slice::sort_by_key
.
ยงExamples of incorrect Ord
implementations
use std::cmp::Ordering;
#[derive(Debug)]
struct Character {
health: f32,
}
impl Ord for Character {
fn cmp(&self, other: &Self) -> std::cmp::Ordering {
if self.health < other.health {
Ordering::Less
} else if self.health > other.health {
Ordering::Greater
} else {
Ordering::Equal
}
}
}
impl PartialOrd for Character {
fn partial_cmp(&self, other: &Self) -> Option<Ordering> {
Some(self.cmp(other))
}
}
impl PartialEq for Character {
fn eq(&self, other: &Self) -> bool {
self.health == other.health
}
}
impl Eq for Character {}
let a = Character { health: 4.5 };
let b = Character { health: f32::NAN };
// Mistake: floating-point values do not form a total order and using the built-in comparison
// operands to implement `Ord` irregardless of that reality does not change it. Use
// `f32::total_cmp` if you need a total order for floating-point values.
// Reflexivity requirement of `Ord` is not given.
assert!(a == a);
assert!(b != b);
// Antisymmetry requirement of `Ord` is not given. Only one of a < c and c < a is allowed to be
// true, not both or neither.
assert_eq!((a < b) as u8 + (b < a) as u8, 0);
use std::cmp::Ordering;
#[derive(Debug)]
struct Character {
health: u32,
experience: u32,
}
impl PartialOrd for Character {
fn partial_cmp(&self, other: &Self) -> Option<Ordering> {
Some(self.cmp(other))
}
}
impl Ord for Character {
fn cmp(&self, other: &Self) -> std::cmp::Ordering {
if self.health < 50 {
self.health.cmp(&other.health)
} else {
self.experience.cmp(&other.experience)
}
}
}
// For performance reasons implementing `PartialEq` this way is not the idiomatic way, but it
// ensures consistent behavior between `PartialEq`, `PartialOrd` and `Ord` in this example.
impl PartialEq for Character {
fn eq(&self, other: &Self) -> bool {
self.cmp(other) == Ordering::Equal
}
}
impl Eq for Character {}
let a = Character {
health: 3,
experience: 5,
};
let b = Character {
health: 10,
experience: 77,
};
let c = Character {
health: 143,
experience: 2,
};
// Mistake: The implementation of `Ord` compares different fields depending on the value of
// `self.health`, the resulting order is not total.
// Transitivity requirement of `Ord` is not given. If a is smaller than b and b is smaller than
// c, by transitive property a must also be smaller than c.
assert!(a < b && b < c && c < a);
// Antisymmetry requirement of `Ord` is not given. Only one of a < c and c < a is allowed to be
// true, not both or neither.
assert_eq!((a < c) as u8 + (c < a) as u8, 2);
The documentation of PartialOrd
contains further examples, for example itโs wrong for
PartialOrd
and PartialEq
to disagree.
Required Methodsยง
1.0.0 ยท Sourcefn cmp(&self, other: &Self) -> Ordering
fn cmp(&self, other: &Self) -> Ordering
This method returns an Ordering
between self
and other
.
By convention, self.cmp(&other)
returns the ordering matching the expression
self <operator> other
if true.
ยงExamples
use std::cmp::Ordering;
assert_eq!(5.cmp(&10), Ordering::Less);
assert_eq!(10.cmp(&5), Ordering::Greater);
assert_eq!(5.cmp(&5), Ordering::Equal);
Provided Methodsยง
1.21.0 ยท Sourcefn max(self, other: Self) -> Selfwhere
Self: Sized,
fn max(self, other: Self) -> Selfwhere
Self: Sized,
Compares and returns the maximum of two values.
Returns the second argument if the comparison determines them to be equal.
ยงExamples
assert_eq!(1.max(2), 2);
assert_eq!(2.max(2), 2);
use std::cmp::Ordering;
#[derive(Eq)]
struct Equal(&'static str);
impl PartialEq for Equal {
fn eq(&self, other: &Self) -> bool { true }
}
impl PartialOrd for Equal {
fn partial_cmp(&self, other: &Self) -> Option<Ordering> { Some(Ordering::Equal) }
}
impl Ord for Equal {
fn cmp(&self, other: &Self) -> Ordering { Ordering::Equal }
}
assert_eq!(Equal("self").max(Equal("other")).0, "other");
1.21.0 ยท Sourcefn min(self, other: Self) -> Selfwhere
Self: Sized,
fn min(self, other: Self) -> Selfwhere
Self: Sized,
Compares and returns the minimum of two values.
Returns the first argument if the comparison determines them to be equal.
ยงExamples
assert_eq!(1.min(2), 1);
assert_eq!(2.min(2), 2);
use std::cmp::Ordering;
#[derive(Eq)]
struct Equal(&'static str);
impl PartialEq for Equal {
fn eq(&self, other: &Self) -> bool { true }
}
impl PartialOrd for Equal {
fn partial_cmp(&self, other: &Self) -> Option<Ordering> { Some(Ordering::Equal) }
}
impl Ord for Equal {
fn cmp(&self, other: &Self) -> Ordering { Ordering::Equal }
}
assert_eq!(Equal("self").min(Equal("other")).0, "self");
Dyn Compatibilityยง
This trait is not dyn compatible.
In older versions of Rust, dyn compatibility was called "object safety", so this trait is not object safe.
Implementorsยง
impl Ord for ErrorKind
impl Ord for StringDelimiter
impl Ord for TokenKind
impl Ord for UnstableFeature
impl Ord for Ordering
impl Ord for AsciiChar
impl Ord for Infallible
impl Ord for IpAddr
impl Ord for SocketAddr
impl Ord for AnsiColor
impl Ord for Color
impl Ord for BigEndian
impl Ord for LittleEndian
impl Ord for Month
impl Ord for ValueSource
impl Ord for PosixFadviseAdvice
impl Ord for PrctlMCEKillPolicy
impl Ord for SigmaskHow
impl Ord for Signal
impl Ord for Utf8Sequence
impl Ord for Direction
impl Ord for Algorithm
impl Ord for ChangeTag
impl Ord for DiffTag
impl Ord for bool
impl Ord for char
impl Ord for i8
impl Ord for i16
impl Ord for i32
impl Ord for i64
impl Ord for i128
impl Ord for isize
impl Ord for !
impl Ord for str
Implements ordering of strings.
Strings are ordered lexicographically by their byte values. This orders Unicode code
points based on their positions in the code charts. This is not necessarily the same as
โalphabeticalโ order, which varies by language and locale. Sorting strings according to
culturally-accepted standards requires locale-specific data that is outside the scope of
the str
type.
impl Ord for u8
impl Ord for u16
impl Ord for u32
impl Ord for u64
impl Ord for u128
impl Ord for ()
impl Ord for usize
impl Ord for Error
impl Ord for Components<'_>
impl Ord for PrefixComponent<'_>
impl Ord for StringKind
impl Ord for OsString
impl Ord for Path
impl Ord for PathBuf
impl Ord for Utf8Path
impl Ord for ByteString
impl Ord for CString
impl Ord for String
impl Ord for TypeId
impl Ord for ByteStr
impl Ord for CpuidResult
impl Ord for CStr
impl Ord for PhantomPinned
impl Ord for Ipv4Addr
impl Ord for Ipv6Addr
impl Ord for SocketAddrV4
impl Ord for SocketAddrV6
impl Ord for Alignment
impl Ord for Duration
impl Ord for std::ffi::os_str::OsStr
impl Ord for Instant
impl Ord for SystemTime
impl Ord for aho_corasick::util::primitives::PatternID
impl Ord for aho_corasick::util::primitives::StateID
impl Ord for Ansi256Color
impl Ord for RgbColor
impl Ord for Effects
impl Ord for Reset
impl Ord for Style
impl Ord for Utf8PathBuf
impl Ord for Months
impl Ord for NaiveDate
impl Ord for NaiveDateDaysIterator
impl Ord for NaiveDateWeeksIterator
impl Ord for NaiveDateTime
impl Ord for IsoWeek
impl Ord for Days
impl Ord for NaiveTime
impl Ord for TimeDelta
impl Ord for Arg
impl Ord for clap_builder::builder::os_str::OsStr
impl Ord for Str
impl Ord for StyledStr
impl Ord for Id
impl Ord for ArgCursor
impl Ord for AtFlags
impl Ord for FallocateFlags
impl Ord for FdFlag
impl Ord for OFlag
impl Ord for RenameFlags
impl Ord for ResolveFlag
impl Ord for SealFlag
impl Ord for MemFdCreateFlag
impl Ord for SaFlags
impl Ord for SfdFlags
impl Ord for Mode
impl Ord for SFlag
impl Ord for FsFlags
impl Ord for TimeSpec
impl Ord for TimeVal
impl Ord for WaitPidFlag
impl Ord for AccessFlags
impl Ord for Pid
impl Ord for LazyStateID
impl Ord for Unit
impl Ord for NonMaxUsize
impl Ord for regex_automata::util::primitives::PatternID
impl Ord for SmallIndex
impl Ord for regex_automata::util::primitives::StateID
impl Ord for Position
impl Ord for Span
impl Ord for Literal
impl Ord for ClassBytesRange
impl Ord for ClassUnicodeRange
impl Ord for Utf8Range
impl Ord for Opcode
impl Ord for BuildMetadata
impl Ord for Prerelease
impl Ord for Version
impl Ord for Height
impl Ord for Width
impl Ord for ATerm
impl Ord for B0
impl Ord for B1
impl Ord for Z0
impl Ord for Equal
impl Ord for Greater
impl Ord for Less
impl Ord for UTerm
impl Ord for Braced
impl Ord for Hyphenated
impl Ord for Simple
impl Ord for Urn
impl Ord for Uuid
impl<'a> Ord for Component<'a>
impl<'a> Ord for Prefix<'a>
impl<'a> Ord for Utf8Component<'a>
impl<'a> Ord for Utf8Prefix<'a>
impl<'a> Ord for PhantomContravariantLifetime<'a>
impl<'a> Ord for PhantomCovariantLifetime<'a>
impl<'a> Ord for PhantomInvariantLifetime<'a>
impl<'a> Ord for Location<'a>
impl<'a> Ord for Utf8Components<'a>
impl<'a> Ord for Utf8PrefixComponent<'a>
impl<'s> Ord for ParsedArg<'s>
impl<'src> Ord for Attribute<'src>
impl<'src> Ord for Interpreter<'src>
impl<'src> Ord for Name<'src>
impl<'src> Ord for Namepath<'src>
impl<'src> Ord for StringLiteral<'src>
impl<'src> Ord for Token<'src>
impl<A> Ord for &A
impl<A> Ord for &mut A
impl<B> Ord for Cow<'_, B>
impl<Dyn> Ord for DynMetadata<Dyn>where
Dyn: ?Sized,
impl<F> Ord for Fwhere
F: FnPtr,
impl<K, V, A> Ord for BTreeMap<K, V, A>
impl<Ptr> Ord for Pin<Ptr>
impl<Storage> Ord for __BindgenBitfieldUnit<Storage>where
Storage: Ord,
impl<T> Ord for Option<T>where
T: Ord,
impl<T> Ord for Poll<T>where
T: Ord,
impl<T> Ord for Resettable<T>where
T: Ord,
impl<T> Ord for *const Twhere
T: ?Sized,
Pointer comparison is by address, as produced by the [
<*const T>::addr](pointer::addr)
method.
impl<T> Ord for *mut Twhere
T: ?Sized,
Pointer comparison is by address, as produced by the <*mut T>::addr
method.
impl<T> Ord for [T]where
T: Ord,
Implements comparison of slices lexicographically.
impl<T> Ord for (Tโ, Tโ, โฆ, Tโ)
This trait is implemented for tuples up to twelve items long.
impl<T> Ord for ManuallyDrop<T>
impl<T> Ord for Cell<T>
impl<T> Ord for RefCell<T>
impl<T> Ord for PhantomData<T>where
T: ?Sized,
impl<T> Ord for PhantomContravariant<T>where
T: ?Sized,
impl<T> Ord for PhantomCovariant<T>where
T: ?Sized,
impl<T> Ord for PhantomInvariant<T>where
T: ?Sized,
impl<T> Ord for NonZero<T>where
T: ZeroablePrimitive + Ord,
impl<T> Ord for Saturating<T>where
T: Ord,
impl<T> Ord for Wrapping<T>where
T: Ord,
impl<T> Ord for NonNull<T>where
T: ?Sized,
impl<T> Ord for CapacityError<T>where
T: Ord,
impl<T> Ord for Change<T>where
T: Ord,
impl<T> Ord for Unalign<T>
impl<T> Ord for Reverse<T>where
T: Ord,
impl<T, A> Ord for BTreeSet<T, A>
impl<T, A> Ord for Rc<T, A>
impl<T, A> Ord for Vec<T, A>
Implements ordering of vectors, lexicographically.
impl<T, A> Ord for Box<T, A>
impl<T, A> Ord for LinkedList<T, A>
impl<T, A> Ord for VecDeque<T, A>
impl<T, A> Ord for UniqueRc<T, A>
impl<T, A> Ord for Arc<T, A>
impl<T, B> Ord for Ref<B, [T]>
impl<T, B> Ord for Ref<B, T>
impl<T, E> Ord for Result<T, E>
impl<T, N> Ord for GenericArray<T, N>where
T: Ord,
N: ArrayLength<T>,
impl<T, const CAP: usize> Ord for ArrayVec<T, CAP>where
T: Ord,
impl<T, const N: usize> Ord for [T; N]where
T: Ord,
Implements comparison of arrays lexicographically.
impl<T, const N: usize> Ord for Simd<T, N>
Lexicographic order. For the SIMD elementwise minimum and maximum, use simd_min and simd_max instead.