```
pub struct BinaryHeap<T, A = Global>where
A: Allocator,{ /* private fields */ }
```

## Expand description

A priority queue implemented with a binary heap.

This will be a max-heap.

It is a logic error for an item to be modified in such a way that the
item’s ordering relative to any other item, as determined by the `Ord`

trait, changes while it is in the heap. This is normally only possible
through interior mutability, global state, I/O, or unsafe code. The
behavior resulting from such a logic error is not specified, but will
be encapsulated to the `BinaryHeap`

that observed the logic error and not
result in undefined behavior. This could include panics, incorrect results,
aborts, memory leaks, and non-termination.

As long as no elements change their relative order while being in the heap
as described above, the API of `BinaryHeap`

guarantees that the heap
invariant remains intact i.e. its methods all behave as documented. For
example if a method is documented as iterating in sorted order, that’s
guaranteed to work as long as elements in the heap have not changed order,
even in the presence of closures getting unwinded out of, iterators getting
leaked, and similar foolishness.

## §Examples

```
use std::collections::BinaryHeap;
// Type inference lets us omit an explicit type signature (which
// would be `BinaryHeap<i32>` in this example).
let mut heap = BinaryHeap::new();
// We can use peek to look at the next item in the heap. In this case,
// there's no items in there yet so we get None.
assert_eq!(heap.peek(), None);
// Let's add some scores...
heap.push(1);
heap.push(5);
heap.push(2);
// Now peek shows the most important item in the heap.
assert_eq!(heap.peek(), Some(&5));
// We can check the length of a heap.
assert_eq!(heap.len(), 3);
// We can iterate over the items in the heap, although they are returned in
// a random order.
for x in &heap {
println!("{x}");
}
// If we instead pop these scores, they should come back in order.
assert_eq!(heap.pop(), Some(5));
assert_eq!(heap.pop(), Some(2));
assert_eq!(heap.pop(), Some(1));
assert_eq!(heap.pop(), None);
// We can clear the heap of any remaining items.
heap.clear();
// The heap should now be empty.
assert!(heap.is_empty())
```

A `BinaryHeap`

with a known list of items can be initialized from an array:

### §Min-heap

Either `core::cmp::Reverse`

or a custom `Ord`

implementation can be used to
make `BinaryHeap`

a min-heap. This makes `heap.pop()`

return the smallest
value instead of the greatest one.

```
use std::collections::BinaryHeap;
use std::cmp::Reverse;
let mut heap = BinaryHeap::new();
// Wrap values in `Reverse`
heap.push(Reverse(1));
heap.push(Reverse(5));
heap.push(Reverse(2));
// If we pop these scores now, they should come back in the reverse order.
assert_eq!(heap.pop(), Some(Reverse(1)));
assert_eq!(heap.pop(), Some(Reverse(2)));
assert_eq!(heap.pop(), Some(Reverse(5)));
assert_eq!(heap.pop(), None);
```

## §Time complexity

The value for `push`

is an expected cost; the method documentation gives a
more detailed analysis.

## Implementations§

Source§### impl<T> BinaryHeap<T>where
T: Ord,

### impl<T> BinaryHeap<T>where
T: Ord,

1.0.0 (const: 1.80.0) · Source#### pub const fn new() -> BinaryHeap<T>

#### pub const fn new() -> BinaryHeap<T>

1.0.0 · Source#### pub fn with_capacity(capacity: usize) -> BinaryHeap<T>

#### pub fn with_capacity(capacity: usize) -> BinaryHeap<T>

Creates an empty `BinaryHeap`

with at least the specified capacity.

The binary heap will be able to hold at least `capacity`

elements without
reallocating. This method is allowed to allocate for more elements than
`capacity`

. If `capacity`

is 0, the binary heap will not allocate.

##### §Examples

Basic usage:

Source§### impl<T, A> BinaryHeap<T, A>

### impl<T, A> BinaryHeap<T, A>

Source#### pub const fn new_in(alloc: A) -> BinaryHeap<T, A>

🔬This is a nightly-only experimental API. (`allocator_api`

#32838)

#### pub const fn new_in(alloc: A) -> BinaryHeap<T, A>

`allocator_api`

#32838)Source#### pub fn with_capacity_in(capacity: usize, alloc: A) -> BinaryHeap<T, A>

🔬This is a nightly-only experimental API. (`allocator_api`

#32838)

#### pub fn with_capacity_in(capacity: usize, alloc: A) -> BinaryHeap<T, A>

`allocator_api`

#32838)Creates an empty `BinaryHeap`

with at least the specified capacity, using `A`

as allocator.

The binary heap will be able to hold at least `capacity`

elements without
reallocating. This method is allowed to allocate for more elements than
`capacity`

. If `capacity`

is 0, the binary heap will not allocate.

##### §Examples

Basic usage:

1.12.0 · Source#### pub fn peek_mut(&mut self) -> Option<PeekMut<'_, T, A>>

#### pub fn peek_mut(&mut self) -> Option<PeekMut<'_, T, A>>

Returns a mutable reference to the greatest item in the binary heap, or
`None`

if it is empty.

Note: If the `PeekMut`

value is leaked, some heap elements might get
leaked along with it, but the remaining elements will remain a valid
heap.

##### §Examples

Basic usage:

```
use std::collections::BinaryHeap;
let mut heap = BinaryHeap::new();
assert!(heap.peek_mut().is_none());
heap.push(1);
heap.push(5);
heap.push(2);
{
let mut val = heap.peek_mut().unwrap();
*val = 0;
}
assert_eq!(heap.peek(), Some(&2));
```

##### §Time complexity

If the item is modified then the worst case time complexity is *O*(log(*n*)),
otherwise it’s *O*(1).

1.0.0 · Source#### pub fn pop(&mut self) -> Option<T>

#### pub fn pop(&mut self) -> Option<T>

Removes the greatest item from the binary heap and returns it, or `None`

if it
is empty.

##### §Examples

Basic usage:

```
use std::collections::BinaryHeap;
let mut heap = BinaryHeap::from([1, 3]);
assert_eq!(heap.pop(), Some(3));
assert_eq!(heap.pop(), Some(1));
assert_eq!(heap.pop(), None);
```

##### §Time complexity

The worst case cost of `pop`

on a heap containing *n* elements is *O*(log(*n*)).

1.0.0 · Source#### pub fn push(&mut self, item: T)

#### pub fn push(&mut self, item: T)

Pushes an item onto the binary heap.

##### §Examples

Basic usage:

```
use std::collections::BinaryHeap;
let mut heap = BinaryHeap::new();
heap.push(3);
heap.push(5);
heap.push(1);
assert_eq!(heap.len(), 3);
assert_eq!(heap.peek(), Some(&5));
```

##### §Time complexity

The expected cost of `push`

, averaged over every possible ordering of
the elements being pushed, and over a sufficiently large number of
pushes, is *O*(1). This is the most meaningful cost metric when pushing
elements that are *not* already in any sorted pattern.

The time complexity degrades if elements are pushed in predominantly
ascending order. In the worst case, elements are pushed in ascending
sorted order and the amortized cost per push is *O*(log(*n*)) against a heap
containing *n* elements.

The worst case cost of a *single* call to `push`

is *O*(*n*). The worst case
occurs when capacity is exhausted and needs a resize. The resize cost
has been amortized in the previous figures.

1.5.0 · Source#### pub fn into_sorted_vec(self) -> Vec<T, A>

#### pub fn into_sorted_vec(self) -> Vec<T, A>

1.11.0 · Source#### pub fn append(&mut self, other: &mut BinaryHeap<T, A>)

#### pub fn append(&mut self, other: &mut BinaryHeap<T, A>)

Source#### pub fn drain_sorted(&mut self) -> DrainSorted<'_, T, A> ⓘ

🔬This is a nightly-only experimental API. (`binary_heap_drain_sorted`

#59278)

#### pub fn drain_sorted(&mut self) -> DrainSorted<'_, T, A> ⓘ

`binary_heap_drain_sorted`

#59278)Clears the binary heap, returning an iterator over the removed elements in heap order. If the iterator is dropped before being fully consumed, it drops the remaining elements in heap order.

The returned iterator keeps a mutable borrow on the heap to optimize its implementation.

Note:

`.drain_sorted()`

is*O*(*n** log(*n*)); much slower than`.drain()`

. You should use the latter for most cases.

##### §Examples

Basic usage:

Source§### impl<T, A> BinaryHeap<T, A>where
A: Allocator,

### impl<T, A> BinaryHeap<T, A>where
A: Allocator,

1.0.0 · Source#### pub fn iter(&self) -> Iter<'_, T> ⓘ

#### pub fn iter(&self) -> Iter<'_, T> ⓘ

Returns an iterator visiting all values in the underlying vector, in arbitrary order.

##### §Examples

Basic usage:

Source#### pub fn into_iter_sorted(self) -> IntoIterSorted<T, A> ⓘ

🔬This is a nightly-only experimental API. (`binary_heap_into_iter_sorted`

#59278)

#### pub fn into_iter_sorted(self) -> IntoIterSorted<T, A> ⓘ

`binary_heap_into_iter_sorted`

#59278)Returns an iterator which retrieves elements in heap order.

This method consumes the original heap.

##### §Examples

Basic usage:

1.0.0 · Source#### pub fn peek(&self) -> Option<&T>

#### pub fn peek(&self) -> Option<&T>

Returns the greatest item in the binary heap, or `None`

if it is empty.

##### §Examples

Basic usage:

```
use std::collections::BinaryHeap;
let mut heap = BinaryHeap::new();
assert_eq!(heap.peek(), None);
heap.push(1);
heap.push(5);
heap.push(2);
assert_eq!(heap.peek(), Some(&5));
```

##### §Time complexity

Cost is *O*(1) in the worst case.

1.0.0 · Source#### pub fn capacity(&self) -> usize

#### pub fn capacity(&self) -> usize

Returns the number of elements the binary heap can hold without reallocating.

##### §Examples

Basic usage:

1.0.0 · Source#### pub fn reserve_exact(&mut self, additional: usize)

#### pub fn reserve_exact(&mut self, additional: usize)

Reserves the minimum capacity for at least `additional`

elements more than
the current length. Unlike `reserve`

, this will not
deliberately over-allocate to speculatively avoid frequent allocations.
After calling `reserve_exact`

, capacity will be greater than or equal to
`self.len() + additional`

. Does nothing if the capacity is already
sufficient.

##### §Panics

Panics if the new capacity overflows `usize`

.

##### §Examples

Basic usage:

1.0.0 · Source#### pub fn reserve(&mut self, additional: usize)

#### pub fn reserve(&mut self, additional: usize)

Reserves capacity for at least `additional`

elements more than the
current length. The allocator may reserve more space to speculatively
avoid frequent allocations. After calling `reserve`

,
capacity will be greater than or equal to `self.len() + additional`

.
Does nothing if capacity is already sufficient.

##### §Panics

Panics if the new capacity overflows `usize`

.

##### §Examples

Basic usage:

1.63.0 · Source#### pub fn try_reserve_exact(
&mut self,
additional: usize,
) -> Result<(), TryReserveError>

#### pub fn try_reserve_exact( &mut self, additional: usize, ) -> Result<(), TryReserveError>

Tries to reserve the minimum capacity for at least `additional`

elements
more than the current length. Unlike `try_reserve`

, this will not
deliberately over-allocate to speculatively avoid frequent allocations.
After calling `try_reserve_exact`

, capacity will be greater than or
equal to `self.len() + additional`

if it returns `Ok(())`

.
Does nothing if the capacity is already sufficient.

Note that the allocator may give the collection more space than it
requests. Therefore, capacity can not be relied upon to be precisely
minimal. Prefer `try_reserve`

if future insertions are expected.

##### §Errors

If the capacity overflows, or the allocator reports a failure, then an error is returned.

##### §Examples

```
use std::collections::BinaryHeap;
use std::collections::TryReserveError;
fn find_max_slow(data: &[u32]) -> Result<Option<u32>, TryReserveError> {
let mut heap = BinaryHeap::new();
// Pre-reserve the memory, exiting if we can't
heap.try_reserve_exact(data.len())?;
// Now we know this can't OOM in the middle of our complex work
heap.extend(data.iter());
Ok(heap.pop())
}
```

1.63.0 · Source#### pub fn try_reserve(&mut self, additional: usize) -> Result<(), TryReserveError>

#### pub fn try_reserve(&mut self, additional: usize) -> Result<(), TryReserveError>

Tries to reserve capacity for at least `additional`

elements more than the
current length. The allocator may reserve more space to speculatively
avoid frequent allocations. After calling `try_reserve`

, capacity will be
greater than or equal to `self.len() + additional`

if it returns
`Ok(())`

. Does nothing if capacity is already sufficient. This method
preserves the contents even if an error occurs.

##### §Errors

If the capacity overflows, or the allocator reports a failure, then an error is returned.

##### §Examples

```
use std::collections::BinaryHeap;
use std::collections::TryReserveError;
fn find_max_slow(data: &[u32]) -> Result<Option<u32>, TryReserveError> {
let mut heap = BinaryHeap::new();
// Pre-reserve the memory, exiting if we can't
heap.try_reserve(data.len())?;
// Now we know this can't OOM in the middle of our complex work
heap.extend(data.iter());
Ok(heap.pop())
}
```

1.0.0 · Source#### pub fn shrink_to_fit(&mut self)

#### pub fn shrink_to_fit(&mut self)

1.56.0 · Source#### pub fn shrink_to(&mut self, min_capacity: usize)

#### pub fn shrink_to(&mut self, min_capacity: usize)

Discards capacity with a lower bound.

The capacity will remain at least as large as both the length and the supplied value.

If the current capacity is less than the lower limit, this is a no-op.

##### §Examples

1.80.0 · Source#### pub fn as_slice(&self) -> &[T]

#### pub fn as_slice(&self) -> &[T]

1.5.0 · Source#### pub fn into_vec(self) -> Vec<T, A>

#### pub fn into_vec(self) -> Vec<T, A>

Consumes the `BinaryHeap`

and returns the underlying vector
in arbitrary order.

##### §Examples

Basic usage:

Source#### pub fn allocator(&self) -> &A

🔬This is a nightly-only experimental API. (`allocator_api`

#32838)

#### pub fn allocator(&self) -> &A

`allocator_api`

#32838)Returns a reference to the underlying allocator.

1.0.0 · Source#### pub fn len(&self) -> usize

#### pub fn len(&self) -> usize

1.0.0 · Source#### pub fn is_empty(&self) -> bool

#### pub fn is_empty(&self) -> bool

1.6.0 · Source#### pub fn drain(&mut self) -> Drain<'_, T, A> ⓘ

#### pub fn drain(&mut self) -> Drain<'_, T, A> ⓘ

Clears the binary heap, returning an iterator over the removed elements in arbitrary order. If the iterator is dropped before being fully consumed, it drops the remaining elements in arbitrary order.

The returned iterator keeps a mutable borrow on the heap to optimize its implementation.

##### §Examples

Basic usage:

## Trait Implementations§

1.0.0 · Source§### impl<T, A> Clone for BinaryHeap<T, A>

### impl<T, A> Clone for BinaryHeap<T, A>

Source§#### fn clone_from(&mut self, source: &BinaryHeap<T, A>)

#### fn clone_from(&mut self, source: &BinaryHeap<T, A>)

Overwrites the contents of `self`

with a clone of the contents of `source`

.

This method is preferred over simply assigning `source.clone()`

to `self`

,
as it avoids reallocation if possible.

See `Vec::clone_from()`

for more details.

Source§#### fn clone(&self) -> BinaryHeap<T, A>

#### fn clone(&self) -> BinaryHeap<T, A>

1.4.0 · Source§### impl<T, A> Debug for BinaryHeap<T, A>

### impl<T, A> Debug for BinaryHeap<T, A>

1.0.0 · Source§### impl<T> Default for BinaryHeap<T>where
T: Ord,

### impl<T> Default for BinaryHeap<T>where
T: Ord,

Source§#### fn default() -> BinaryHeap<T>

#### fn default() -> BinaryHeap<T>

Creates an empty `BinaryHeap<T>`

.

1.2.0 · Source§### impl<'a, T, A> Extend<&'a T> for BinaryHeap<T, A>

### impl<'a, T, A> Extend<&'a T> for BinaryHeap<T, A>

1.0.0 · Source§### impl<T, A> Extend<T> for BinaryHeap<T, A>

### impl<T, A> Extend<T> for BinaryHeap<T, A>

Source§#### fn extend<I>(&mut self, iter: I)where
I: IntoIterator<Item = T>,

#### fn extend<I>(&mut self, iter: I)where
I: IntoIterator<Item = T>,

Source§#### fn extend_one(&mut self, item: T)

#### fn extend_one(&mut self, item: T)

`extend_one`

#72631)1.56.0 · Source§### impl<T, const N: usize> From<[T; N]> for BinaryHeap<T>where
T: Ord,

### impl<T, const N: usize> From<[T; N]> for BinaryHeap<T>where
T: Ord,

Source§#### fn from(arr: [T; N]) -> BinaryHeap<T>

#### fn from(arr: [T; N]) -> BinaryHeap<T>

1.5.0 · Source§### impl<T, A> From<BinaryHeap<T, A>> for Vec<T, A>where
A: Allocator,

### impl<T, A> From<BinaryHeap<T, A>> for Vec<T, A>where
A: Allocator,

Source§#### fn from(heap: BinaryHeap<T, A>) -> Vec<T, A>

#### fn from(heap: BinaryHeap<T, A>) -> Vec<T, A>

Converts a `BinaryHeap<T>`

into a `Vec<T>`

.

This conversion requires no data movement or allocation, and has constant time complexity.

1.5.0 · Source§### impl<T, A> From<Vec<T, A>> for BinaryHeap<T, A>

### impl<T, A> From<Vec<T, A>> for BinaryHeap<T, A>

Source§#### fn from(vec: Vec<T, A>) -> BinaryHeap<T, A>

#### fn from(vec: Vec<T, A>) -> BinaryHeap<T, A>

Converts a `Vec<T>`

into a `BinaryHeap<T>`

.

This conversion happens in-place, and has *O*(*n*) time complexity.