[−][src]Struct thincollections::thin_vec::ThinVec

pub struct ThinVec<T> { /* fields omitted */ }

A thin (usize) vector. Guaranteed to be a usize-sized smart pointer.

Rust's std::collections::Vec (std::Vec for short) is a triple-fat (3 x usize) pointer to the heap. When std::Vec is on the stack, it works well. However, when std::Vec is used inside other data structures, such as Vec<Vec<_>>, the triple fatness starts to become a problem. For example, when a Vec<Vec<_>> has to resize, it needs to move 3 times as much memory.

ThinVec uses a single usize value as a smart pointer, making it attractive for building larger data structures.

ThinVec is also null optimized, which makes an Option<ThinVec<_>> also usize-sized.

Capacity and reallocation

The capacity of a vector is the amount of space allocated for any future elements that will be added onto the vector. This is not to be confused with the length of a vector, which specifies the number of actual elements within the vector. If a vector's length exceeds its capacity, its capacity will automatically be increased, but its elements will have to be reallocated.

For example, a vector with capacity 10 and length 0 would be an empty vector with space for 10 more elements. Pushing 10 or fewer elements onto the vector will not change its capacity or cause reallocation to occur. However, if the vector's length is increased to 11, it will have to reallocate, which can be slow. For this reason, it is recommended to use ThinVec::with_capacity whenever possible to specify how big the vector is expected to get.

Usage Patterns

Aside from the simple single element methods, push, pop, insert and remove, ThinVec, just as std::Vec has a large number of API's that can be a bit hard to navigate, especially because it also dereferences to a slice, which has a massive api count.

Initialization Patterns

Fill a ThinVec with some clonable object:

#[macro_use] extern crate thincollections;
use thincollections::thin_vec::ThinVec;
let v: ThinVec<i32> = thinvec![7; 42];
assert_eq!(42, v.len());
assert_eq!(7, v[0]);
assert_eq!(7, v[41]);

The macro just calls ThinVec::from_elem(7, 42) which makes sure the vector is created with ThinVec::with_capacity before filling it.

Fill a ThinVec with the result of some function:

use thincollections::thin_vec::ThinVec;
let v : ThinVec<i32> = (0..42).map(|x| x+7).collect(); // x+7 is the example function here
assert_eq!(42, v.len());
assert_eq!(7, v[0]);
assert_eq!(48, v[41]);

The above patterns is also useful for types that do funny stuff when cloned. For example, when ThinVec is cloned, it only copies the values, not the capacity, so trying to create a ThinVec<ThinVec<_>> with a given capacity for the inner vectors should be done via a loop or as above, collect.

Sorting

ThinVec can be sorted because a slice can be sorted and ThinVec derefs to slice. To be sortable, the type inside the ThinVec must implement cmp::Ord. Alternatively, the sort_by method can be used to specify a custom comparator.

#[macro_use] extern crate thincollections;
use thincollections::thin_vec::ThinVec;
let mut v: ThinVec<i32> = thinvec![3, 1, 2, -1];
v.sort(); // the compiler turns this into: v.as_mut_slice().sort()
assert_eq!(-1, v[0]);
assert_eq!(3, v[3]);

Sorting floating point numbers is harder because they don't implement cmp::Ord. This is because of the weird values in floats, most importantly NaN, which has very odd semantics regarding equality (it's not equal to anything, not even itself). We can use ThinVec::transmute to get around these issues.

#[macro_use] extern crate thincollections;
extern crate ordered_float;

use thincollections::thin_vec::ThinVec;
use ordered_float::NotNan;
let mut v: ThinVec<f32> = thinvec![3.0, 1.0, 2.0, -1.0];
unsafe {
    let mut v_notnan: ThinVec<NotNan<f32>> = v.transmute();
    v_notnan.sort();
    let v: ThinVec<f32> = v_notnan.transmute();
    assert_eq!(-1.0, v[0]);
    assert_eq!(3.0, v[3]);
}

After sorting a ThinVec, you can then use binary_search to find stuff.

Iteration

ThinVec "inherits" its iteration from slice. Iterating through ThinVec is far more efficient compared to indexing operations.

#[macro_use] extern crate thincollections;
use thincollections::thin_vec::ThinVec;
let v: ThinVec<i32> = thinvec![1, 2, 3, 4, 5, 6, 7, 8, 9, 10];
let mut sum = 0;
for i in v.iter() { // i is a &i32, not i32 here.
    sum += i; // the compiler turns this into sum += *i;
}
assert_eq!(55, sum);

Because iter() produces a Rust Iterator, which has a large API, many things can be done in a functional style without the explicit loop:

#[macro_use] extern crate thincollections;
use thincollections::thin_vec::ThinVec;
let v: ThinVec<i32> = thinvec![1, 2, 3, 4, 5, 6, 7, 8, 9, 10];
assert_eq!(55, v.iter().fold(0, |sum, x| sum + x)); // same as the loop above
assert_eq!(550, v.iter().map(|x| x * 10).fold(0, |sum, x| sum + x));
assert_eq!([2, 4, 6, 8, 10], v.iter().filter(|x| *x & 1 == 0).cloned().collect::<ThinVec<i32>>().as_slice());

We can also iterate and mutate (in place):

#[macro_use] extern crate thincollections;
use thincollections::thin_vec::ThinVec;
let mut v: ThinVec<i32> = thinvec![1, 2, 3, 4, 5, 6, 7, 8, 9, 10];
for i in v.iter_mut() { // i is a &mut i32
    *i = *i + 10; // i points inside the vector!
}
assert_eq!(11, v[0]);
assert_eq!(20, v[9]);

Finally, we can move the contents of the vector out using into_iter()

#[macro_use] extern crate thincollections;
use thincollections::thin_vec::ThinVec;
let v: ThinVec<i32> = thinvec![1, 2, 3, 4, 5, 6, 7, 8, 9, 10];
let mut sum = 0;
for i in v.into_iter() { // i is i32 here (not &i32).
    sum += i;
}
// can't access v anymore here, it's been consumed.
assert_eq!(55, sum);

Concatenation

Concatenating vectors to vectors requires understanding the ownership/copy sematics of Rust. Vectors own their contents, so a simple concatenation using the ThinVec::append will move the contents (leaving one of the vectors empty):

#[macro_use] extern crate thincollections;
use thincollections::thin_vec::ThinVec;
let mut v: ThinVec<i32> = thinvec![10, 20, 30];
let mut v2: ThinVec<i32> = thinvec![40, 50, 60];
v.append(&mut v2); // we take a mutable reference, because we're about to empty out v2
assert_eq!(6, v.len());
assert_eq!(60, v[5]);
assert_eq!(0, v2.len());

Vectors can also be extended from clonable slices (and a vector derefs to a slice!). We can use this to concatenate two vectors without emptying one out:

#[macro_use] extern crate thincollections;
use thincollections::thin_vec::ThinVec;
let mut v: ThinVec<i32> = thinvec![10, 20, 30];
let v2: ThinVec<i32> = thinvec![40, 50, 60];
v.extend(&v2); // the compiler turns this into v2.as_slice();
assert_eq!(6, v.len());
assert_eq!(60, v[5]);
assert_eq!(3, v2.len());

We also use the same to concatenate a slice to a ThinVec:

#[macro_use] extern crate thincollections;
use thincollections::thin_vec::ThinVec;
let mut v: ThinVec<i32> = thinvec![10, 20, 30];
let s: [i32; 3] = [40, 50, 60]; // s is an array
v.extend(&s); // &s is a slice, not an array
assert_eq!(6, v.len());
assert_eq!(60, v[5]);
assert_eq!(3, s.len());

Splitting

You can split (and move the contents) of a vector into another vector:

let mut vec = thinvec![1,2,3];
let vec2 = vec.split_off(1);
assert_eq!(vec, [1]);
assert_eq!(vec2, [2, 3]);

You can also mutably borrow multiple parts of a vector using the split_at_mut slice method:

#[macro_use] extern crate thincollections;
use thincollections::thin_vec::ThinVec;
let mut v = thinvec![1, 0, 3, 0, 5, 6];
// scoped to restrict the lifetime of the borrows
{
    let (left, right) = v.split_at_mut(2);
    assert!(left == [1, 0]);
    assert!(right == [3, 0, 5, 6]);
    left[1] = 2;
    right[1] = 4;
}
assert!(v == [1, 2, 3, 4, 5, 6]);

Methods

`impl<T> ThinVec<T>`
[src]

`pub fn new() -> ThinVec<T>`
[src]

Constructs a new, empty ThinVec<T>.

The vector will not allocate until elements are pushed onto it.

Examples

use thincollections::thin_vec::ThinVec;
let mut vec: ThinVec<i32> = ThinVec::new();
vec.push(17); // initial allocation
vec.push(42);

`pub fn with_capacity(capacity: usize) -> ThinVec<T>`
[src]

Constructs a new, empty ThinVec<T> with the specified capacity.

The vector will be able to hold exactly capacity elements without reallocating. If capacity is 0, the vector will not allocate.

It is important to note that although the returned vector has the capacity specified, the vector will have a zero length. For an explanation of the difference between length and capacity, see Capacity and reallocation.

Examples

use thincollections::thin_vec::ThinVec;
let mut vec = ThinVec::with_capacity(10);

// The vector contains no items, even though it has capacity for more
assert_eq!(vec.len(), 0);

// These are all done without reallocating...
for i in 0..10 {
    vec.push(i);
}

// ...but this may make the vector reallocate
vec.push(11);

`pub fn capacity(&self) -> usize`
[src]

Returns the number of elements the vector can hold without reallocating.

Examples

use thincollections::thin_vec::ThinVec;
let vec: ThinVec<i32> = ThinVec::with_capacity(10);
assert_eq!(vec.capacity(), 10);

`pub fn reserve(&mut self, additional: usize)`
[src]

Reserves capacity for at least additional more elements to be inserted in the given ThinVec<T>. The collection may reserve more space to avoid frequent reallocations. 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

use thincollections::thin_vec::ThinVec;
let mut vec = ThinVec::new();
vec.push(1);
vec.reserve(10);
assert!(vec.capacity() >= 11);

`pub fn reserve_exact(&mut self, additional: usize)`
[src]

Reserves the minimum capacity for exactly additional more elements to be inserted in the given ThinVec<T>. After calling reserve_exact, capacity will be greater than or equal to self.len() + additional. Does nothing if the capacity is already sufficient.

Note that when the stack storage capacity exceeds self.len() + additional, nothing is allocated on the heap and the capacity remains as the stack capacity.

Panics

Panics if the new capacity overflows usize.

Examples

use thincollections::thin_vec::ThinVec;
let mut vec = ThinVec::new();
vec.push(1);
vec.reserve_exact(10);
assert_eq!(11, vec.capacity());

`pub fn shrink_to_fit(&mut self)`
[src]

Shrinks the capacity of the vector as much as possible.

It will drop down as close as possible to the length but the allocator may still inform the vector that there is space for a few more elements.

Examples

use thincollections::thin_vec::ThinVec;
let mut vec = ThinVec::with_capacity(10);
vec.push(1);
vec.push(2);
vec.push(3);
assert_eq!(vec.capacity(), 10);
vec.shrink_to_fit();
assert!(vec.capacity() < 10);
assert!(vec.capacity() == 3);

`pub fn truncate(&mut self, len: usize)`
[src]

Shortens the vector, keeping the first len elements and dropping the rest.

If len is greater than the vector's current length, this has no effect.

The drain method can emulate truncate, but causes the excess elements to be returned instead of dropped.

Note that this method has no effect on the allocated capacity of the vector.

Examples

Truncating a five element vector to two elements:

let mut vec = thinvec![1, 2, 3, 4, 5];
vec.truncate(2);
assert_eq!(2, vec.len());
assert_eq!(1, vec[0]);
assert_eq!(2, vec[1]);

No truncation occurs when len is greater than the vector's current length:

let mut vec = thinvec![1, 2, 3];
vec.truncate(8);
assert_eq!(vec[..], [1, 2, 3]);

Truncating when len == 0 is equivalent to calling the clear method.

let mut vec = thinvec![1, 2, 3];
vec.truncate(0);
assert_eq!(vec[..], []);

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

Extracts a slice containing the entire vector.

Equivalent to &s[..].

Examples

use std::io::{self, Write};
let buffer = thinvec![1, 2, 3, 5, 8];
io::sink().write(buffer.as_slice()).unwrap();

`pub fn as_mut_slice(&mut self) -> &mut [T]`
[src]

Extracts a mutable slice of the entire vector.

Equivalent to &mut s[..].

Examples

use std::io::{self, Read};
let mut buffer = thinvec![0; 3];
io::repeat(0b101).read_exact(buffer.as_mut_slice()).unwrap();

`pub fn swap_remove(&mut self, index: usize) -> T`
[src]

Removes an element from the vector and returns it.

The removed element is replaced by the last element of the vector.

This does not preserve ordering, but is O(1).

Panics

Panics if index is out of bounds.

Examples

let mut v = thinvec!["foo", "bar", "baz", "qux"];

assert_eq!(v.swap_remove(1), "bar");
assert_eq!(v[..], ["foo", "qux", "baz"]);

assert_eq!(v.swap_remove(0), "foo");
assert_eq!(v[..], ["baz", "qux"]);

`pub fn insert(&mut self, index: usize, val: T)`
[src]

Inserts an element at position index within the vector, shifting all elements after it to the right.

Panics

Panics if index > len.

Examples

let mut vec = thinvec![1, 2, 3];
vec.insert(1, 4);
assert_eq!(vec, [1, 4, 2, 3]);
vec.insert(4, 5);
assert_eq!(vec, [1, 4, 2, 3, 5]);

`pub fn remove(&mut self, index: usize) -> T`
[src]

Removes and returns the element at position index within the vector, shifting all elements after it to the left.

Panics

Panics if index is out of bounds.

Examples

let mut v = thinvec![1, 2, 3];
assert_eq!(v.remove(1), 2);
assert_eq!(v, [1, 3]);

`pub fn retain<F>(&mut self, f: F) where F: FnMut(&T) -> bool,`
[src]

Retains only the elements specified by the predicate.

In other words, remove all elements e such that f(&e) returns false. This method operates in place and preserves the order of the retained elements.

Examples

let mut vec = thinvec![1, 2, 3, 4];
vec.retain(|&x| x%2 == 0);
assert_eq!(vec, [2, 4]);

`pub fn dedup_by<F>(&mut self, same_bucket: F) where F: FnMut(&mut T, &mut T) -> bool,`
[src]

Removes all but the first of consecutive elements in the vector satisfying a given equality relation.

The same_bucket function is passed references to two elements from the vector, and returns true if the elements compare equal, or false if they do not. The elements are passed in opposite order from their order in the vector, so if same_bucket(a, b) returns true, a is removed.

If the vector is sorted, this removes all duplicates.

Examples

let mut vec = thinvec!["foo", "bar", "Bar", "baz", "bar"];

vec.dedup_by(|a, b| a.eq_ignore_ascii_case(b));

assert_eq!(vec, ["foo", "bar", "baz", "bar"]);

`pub fn dedup_by_key<F, K>(&mut self, key: F) where F: FnMut(&mut T) -> K, K: PartialEq,`
[src]

Removes all but the first of consecutive elements in the vector that resolve to the same key.

If the vector is sorted, this removes all duplicates.

Examples

let mut vec = thinvec![10, 20, 21, 30, 20];

vec.dedup_by_key(|i| *i / 10);

assert_eq!(vec, [10, 20, 30, 20]);

`pub fn len(&self) -> usize`
[src]

Returns the number of elements in the vector, also referred to as its 'length'.

Examples

let a = thinvec![1, 2, 3];
assert_eq!(a.len(), 3);

`pub fn is_empty(&self) -> bool`
[src]

Returns true if the vector contains no elements.

Examples

use thincollections::thin_vec::ThinVec;
let mut v = ThinVec::new();
assert!(v.is_empty());

v.push(1);
assert!(!v.is_empty());

`pub fn push(&mut self, val: T)`
[src]

Appends an element to the back of a collection.

Panics

Panics if the number of elements in the vector overflows a usize.

Examples

let mut vec = thinvec![1, 2];
vec.push(3);
assert_eq!(vec, [1, 2, 3]);

`pub fn pop(&mut self) -> Option<T>`
[src]

Removes the last element from a vector and returns it, or None if it is empty.

Examples

let mut vec = thinvec![1, 2, 3];
assert_eq!(vec.pop(), Some(3));
assert_eq!(vec, [1, 2]);

`pub fn append(&mut self, other: &mut Self)`
[src]

Moves all the elements of other into Self, leaving other empty.

Panics

Panics if the number of elements in the vector overflows a usize.

Examples

let mut vec = thinvec![1, 2, 3];
let mut vec2 = thinvec![4, 5, 6];
vec.append(&mut vec2);
assert_eq!(vec, [1, 2, 3, 4, 5, 6]);
assert_eq!(vec2, []);

`pub fn split_off(&mut self, at: usize) -> Self`
[src]

Splits the collection into two at the given index.

Returns a newly allocated Self. self contains elements [0, at), and the returned Self contains elements [at, len).

Note that the capacity of self does not change.

Panics

Panics if at > len.

Examples

let mut vec = thinvec![1,2,3];
let vec2 = vec.split_off(1);
assert_eq!(vec, [1]);
assert_eq!(vec2, [2, 3]);

`pub fn clear(&mut self)`
[src]

Clears the vector, removing all values.

Note that this method has no effect on the allocated capacity of the vector.

Examples

let mut v = thinvec![1, 2, 3];

v.clear();

assert!(v.is_empty());

`pub fn into_boxed_slice(self) -> Box<[T]>`
[src]

Converts the vector into Box<[T]>.

This function causes a bunch of copying and should generally be avoided. ThinVec is far more versatile than Box<[T]>

Note that this will drop any excess capacity.

Examples

let v = thinvec![1, 2, 3];

let slice = v.into_boxed_slice();
assert_eq!([1,2,3], slice[..]);

Any excess capacity is removed:

let mut vec = ThinVec::with_capacity(10);
vec.extend([1, 2, 3].iter().cloned());

assert_eq!(vec.capacity(), 10);
let slice: Box<[i32]> = vec.into_boxed_slice();
assert_eq!(slice.into_thinvec().capacity(), 3);

ⓘImportant traits for Drain<'a, T>
Important traits for Drain<'a, T>
`impl<'a, T> Iterator for Drain<'a, T> type Item = T;`
`pub fn drain<R>(&mut self, range: R) -> Drain<T> where R: RangeBounds<usize>,`
[src]

Creates a draining iterator that removes the specified range in the vector and yields the removed items.

Note 1: The element range is removed even if the iterator is only partially consumed or not consumed at all.

Note 2: It is unspecified how many elements are removed from the vector if the Drain value is leaked.

Panics

Panics if the starting point is greater than the end point or if the end point is greater than the length of the vector.

Examples

let mut v = thinvec![1, 2, 3];
let u: ThinVec<_> = v.drain(1..).collect();
assert_eq!(v, &[1]);
assert_eq!(u, &[2, 3]);

// A full range clears the vector
v.drain(..);
assert_eq!(v, &[]);

ⓘImportant traits for Splice<'a, I>
Important traits for Splice<'a, I>
`impl<'a, I: Iterator> Iterator for Splice<'a, I> type Item = I::Item;`
`pub fn splice<R, I>(&mut self, range: R, replace_with: I) -> Splice<I::IntoIter> where R: RangeBounds<usize>, I: IntoIterator<Item = T>,`
[src]

Creates a splicing iterator that replaces the specified range in the vector with the given replace_with iterator and yields the removed items. replace_with does not need to be the same length as range.

Note 1: The element range is removed even if the iterator is not consumed until the end.

Note 2: It is unspecified how many elements are removed from the vector, if the Splice value is leaked.

Note 3: The input iterator replace_with is only consumed when the Splice value is dropped.

Note 4: This is optimal if:

The tail (elements in the vector after range) is empty,
or replace_with yields fewer elements than range’s length
or the lower bound of its size_hint() is exact.

Otherwise, a temporary vector is allocated and the tail is moved twice.

Panics

Panics if the starting point is greater than the end point or if the end point is greater than the length of the vector.

Examples

let mut v = thinvec![1, 2, 3];
let new = [7, 8];
let u: ThinVec<_> = v.splice(..2, new.iter().cloned()).collect();
assert_eq!(v, &[7, 8, 3]);
assert_eq!(u, &[1, 2]);

ⓘImportant traits for DrainFilter<'a, T, F>
Important traits for DrainFilter<'a, T, F>
`impl<'a, T, F> Iterator for DrainFilter<'a, T, F> where F: FnMut(&mut T) -> bool, type Item = T;`
`pub fn drain_filter<F>(&mut self, filter: F) -> DrainFilter<T, F> where F: FnMut(&mut T) -> bool,`
[src]

Creates an iterator which uses a closure to determine if an element should be removed.

If the closure returns true, then the element is removed and yielded. If the closure returns false, the element will remain in the vector and will not be yielded by the iterator.

Using this method is equivalent to the following code:

let mut i = 0;
while i != vec.len() {
    if some_predicate(&mut vec[i]) {
        let val = vec.remove(i);
        // your code here
    } else {
        i += 1;
    }
}

But drain_filter is easier to use. drain_filter is also more efficient, because it can backshift the elements of the array in bulk.

Note that drain_filter also lets you mutate every element in the filter closure, regardless of whether you choose to keep or remove it.

Examples

Splitting an array into evens and odds, reusing the original allocation:

let mut numbers = thinvec![1, 2, 3, 4, 5, 6, 8, 9, 11, 13, 14, 15];

let evens = numbers.drain_filter(|x| *x % 2 == 0).collect::<ThinVec<_>>();
let odds = numbers;

assert_eq!(evens, thinvec![2, 4, 6, 8, 14]);
assert_eq!(odds, thinvec![1, 3, 5, 9, 11, 13, 15]);

`impl<T: Clone> ThinVec<T>`
[src]

`pub fn from_elem(elem: T, count: usize) -> ThinVec<T>`
[src]

`pub fn extend_from_slice(&mut self, slice: &[T])`
[src]

`impl<T: Copy> ThinVec<T>`
[src]

`pub unsafe fn transmute<X: Copy>(self) -> ThinVec<X>`
[src]

Transmutes the vector ThinVec<T> into another type of vector ThinVec<X>. Consumes the original vector.

mem::size_of::<X> must equal mem::size_of::<T>

mem::align_of::<X> must equal mem::align_of::<T>

This is achieved with no copying at all, making it super fast. Both X and T must be Copy types so as to guarantee no Drop (for sanity's sake only; no copying is performed).

This is useful and safe for all primitive types that have the same width For example:

let vecu: ThinVec<u8> = thinvec![1, 2, 3, 128, 0xFF];
unsafe {
    let veci: ThinVec<i8> = vecu.transmute();
    assert_eq!(1, veci[0]);
    assert_eq!(-128i8, veci[3]);
    assert_eq!(-1i8, veci[4]);
}

Please note that this works the same way as mem::transmute. It does not do any sort of numeric conversion. It reuses the same bits for the new type. So going from f32 to i32 is not going to create sensible numbers, just the same "weird" integers that f32::to_bits produces.

It might also be useful for converting from primitives to single elements structs, including some of the special types like NonZeroU32, if you know you don't violate any of the struct's invariants (e.g. no zeros if you're converting to NonZeroU32). repr(transparent) is a good annotation to use on such types.

One thing this can help with is with a total ordered floating point wrapper struct that will then allow for sorting, max, binary search, etc. see Sorting for an example

Avoid using this for (multi-element) repr(Rust) structs. It might be somewhat useful for repr(C) structs if you know what you're doing.

Under no circumstances does it make sense to transmute to enum wrappers such as Option, as the bit pattern for these is not well specified (even if they happen to have the same size).

Panics

Panics if the the size or alignment of X and T are different.

`impl<T: PartialEq> ThinVec<T>`
[src]

`pub fn dedup(&mut self)`
[src]

Removes consecutive repeated elements in the vector.

If the vector is sorted, this removes all duplicates.

Examples

let mut vec = vec![1, 2, 2, 3, 2];

vec.dedup();

assert_eq!(vec, [1, 2, 3, 2]);

`pub fn remove_item(&mut self, item: &T) -> Option<T>`
[src]

Removes the first instance of item from the vector if the item exists.

Examples

let mut vec = thinvec![1, 2, 3, 1];

vec.remove_item(&1);

assert_eq!(vec, thinvec![2, 3, 1]);

Trait Implementations

`impl<T> Default for ThinVec<T>`
[src]