This repository has been archived on 2023-10-09. You can view files and clone it, but cannot push or open issues or pull requests.
Files
blender-archive/source/blender/blenlib/BLI_vector.hh

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

999 lines
28 KiB
C++
Raw Normal View History

/* SPDX-License-Identifier: GPL-2.0-or-later */
#pragma once
2019-09-13 21:12:26 +10:00
/** \file
* \ingroup bli
*
* A `blender::Vector<T>` is a dynamically growing contiguous array for values of type T. It is
* designed to be a more convenient and efficient replacement for `std::vector`. Note that the term
* "vector" has nothing to do with a vector from computer graphics here.
*
* A vector supports efficient insertion and removal at the end (O(1) amortized). Removal in other
* places takes O(n) time, because all elements afterwards have to be moved. If the order of
* elements is not important, `remove_and_reorder` can be used instead of `remove` for better
* performance.
*
* The improved efficiency is mainly achieved by supporting small buffer optimization. As long as
* the number of elements in the vector does not become larger than InlineBufferCapacity, no memory
* allocation is done. As a consequence, iterators are invalidated when a blender::Vector is moved
* (iterators of std::vector remain valid when the vector is moved).
*
* `blender::Vector` should be your default choice for a vector data structure in Blender.
*/
#include <algorithm>
#include <cstdlib>
#include <cstring>
#include <iostream>
#include <memory>
#include "BLI_allocator.hh"
#include "BLI_index_range.hh"
#include "BLI_listbase_wrapper.hh"
#include "BLI_math_base.h"
#include "BLI_memory_utils.hh"
#include "BLI_span.hh"
#include "BLI_string.h"
#include "BLI_string_ref.hh"
#include "BLI_utildefines.h"
#include "MEM_guardedalloc.h"
namespace blender {
template<
/**
* Type of the values stored in this vector. It has to be movable.
*/
typename T,
/**
* The number of values that can be stored in this vector, without doing a heap allocation.
* Sometimes it makes sense to increase this value a lot. The memory in the inline buffer is
* not initialized when it is not needed.
*
* When T is large, the small buffer optimization is disabled by default to avoid large
2020-06-13 12:50:07 +10:00
* unexpected allocations on the stack. It can still be enabled explicitly though.
*/
int64_t InlineBufferCapacity = default_inline_buffer_capacity(sizeof(T)),
/**
* The allocator used by this vector. Should rarely be changed, except when you don't want that
* MEM_* is used internally.
*/
typename Allocator = GuardedAllocator>
class Vector {
public:
using value_type = T;
using pointer = T *;
using const_pointer = const T *;
using reference = T &;
using const_reference = const T &;
using iterator = T *;
using const_iterator = const T *;
using size_type = int64_t;
private:
/**
2020-06-13 12:50:07 +10:00
* Use pointers instead of storing the size explicitly. This reduces the number of instructions
* in `append`.
*
* The pointers might point to the memory in the inline buffer.
*/
T *begin_;
T *end_;
T *capacity_end_;
/** Used for allocations when the inline buffer is too small. */
Allocator allocator_;
/** A placeholder buffer that will remain uninitialized until it is used. */
TypedBuffer<T, InlineBufferCapacity> inline_buffer_;
/**
2020-06-13 12:50:07 +10:00
* Store the size of the vector explicitly in debug builds. Otherwise you'd always have to call
* the `size` function or do the math to compute it from the pointers manually. This is rather
* annoying. Knowing the size of a vector is often quite essential when debugging some code.
*/
#ifndef NDEBUG
int64_t debug_size_;
# define UPDATE_VECTOR_SIZE(ptr) \
(ptr)->debug_size_ = static_cast<int64_t>((ptr)->end_ - (ptr)->begin_)
#else
# define UPDATE_VECTOR_SIZE(ptr) ((void)0)
#endif
/**
2020-06-30 20:54:31 +10:00
* Be a friend with other vector instantiations. This is necessary to implement some memory
* management logic.
*/
template<typename OtherT, int64_t OtherInlineBufferCapacity, typename OtherAllocator>
friend class Vector;
public:
/**
* Create an empty vector.
* This does not do any memory allocation.
*/
Vector(Allocator allocator = {}) noexcept : allocator_(allocator)
{
begin_ = inline_buffer_;
end_ = begin_;
capacity_end_ = begin_ + InlineBufferCapacity;
UPDATE_VECTOR_SIZE(this);
}
Vector(NoExceptConstructor, Allocator allocator = {}) noexcept : Vector(allocator)
{
}
/**
* Create a vector with a specific size.
* The elements will be default constructed.
* If T is trivially constructible, the elements in the vector are not touched.
*/
explicit Vector(int64_t size, Allocator allocator = {})
: Vector(NoExceptConstructor(), allocator)
{
this->resize(size);
}
/**
* Create a vector filled with a specific value.
*/
Vector(int64_t size, const T &value, Allocator allocator = {})
: Vector(NoExceptConstructor(), allocator)
{
this->resize(size, value);
}
/**
* Create a vector from a span. The values in the vector are copy constructed.
*/
template<typename U, BLI_ENABLE_IF((std::is_convertible_v<U, T>))>
Vector(Span<U> values, Allocator allocator = {}) : Vector(NoExceptConstructor(), allocator)
{
const int64_t size = values.size();
this->reserve(size);
uninitialized_convert_n<U, T>(values.data(), size, begin_);
this->increase_size_by_unchecked(size);
}
/**
2020-06-25 23:13:02 +10:00
* Create a vector that contains copies of the values in the initialized list.
*
* This allows you to write code like:
* Vector<int> vec = {3, 4, 5};
*/
template<typename U, BLI_ENABLE_IF((std::is_convertible_v<U, T>))>
Vector(const std::initializer_list<U> &values) : Vector(Span<U>(values))
{
}
Vector(const std::initializer_list<T> &values) : Vector(Span<T>(values))
{
}
template<typename U, size_t N, BLI_ENABLE_IF((std::is_convertible_v<U, T>))>
Vector(const std::array<U, N> &values) : Vector(Span(values))
{
}
template<typename InputIt,
/* This constructor should not be called with e.g. Vector(3, 10), because that is
2020-10-14 15:24:42 +11:00
* expected to produce the vector (10, 10, 10). */
BLI_ENABLE_IF((!std::is_convertible_v<InputIt, int>))>
Vector(InputIt first, InputIt last, Allocator allocator = {})
: Vector(NoExceptConstructor(), allocator)
{
for (InputIt current = first; current != last; ++current) {
this->append(*current);
}
}
/**
* Create a vector from a ListBase. The caller has to make sure that the values in the linked
* list have the correct type.
*
* Example Usage:
* Vector<ModifierData *> modifiers(ob->modifiers);
*/
Vector(const ListBase &values, Allocator allocator = {})
: Vector(NoExceptConstructor(), allocator)
{
LISTBASE_FOREACH (T, value, &values) {
this->append(value);
}
}
/**
* Create a copy of another vector. The other vector will not be changed. If the other vector has
* less than InlineBufferCapacity elements, no allocation will be made.
*/
Vector(const Vector &other) : Vector(other.as_span(), other.allocator_)
{
}
/**
* Create a copy of a vector with a different InlineBufferCapacity. This needs to be handled
* separately, so that the other one is a valid copy constructor.
*/
template<int64_t OtherInlineBufferCapacity>
Vector(const Vector<T, OtherInlineBufferCapacity, Allocator> &other)
: Vector(other.as_span(), other.allocator_)
{
}
/**
* Steal the elements from another vector. This does not do an allocation. The other vector will
* have zero elements afterwards.
*/
template<int64_t OtherInlineBufferCapacity>
Vector(Vector<T, OtherInlineBufferCapacity, Allocator> &&other) noexcept(
std::is_nothrow_move_constructible_v<T>)
: Vector(NoExceptConstructor(), other.allocator_)
{
const int64_t size = other.size();
if (other.is_inline()) {
if (size <= InlineBufferCapacity) {
/* Copy between inline buffers. */
uninitialized_relocate_n(other.begin_, size, begin_);
end_ = begin_ + size;
}
else {
/* Copy from inline buffer to newly allocated buffer. */
const int64_t capacity = size;
begin_ = static_cast<T *>(
allocator_.allocate(sizeof(T) * static_cast<size_t>(capacity), alignof(T), AT));
capacity_end_ = begin_ + capacity;
uninitialized_relocate_n(other.begin_, size, begin_);
end_ = begin_ + size;
}
}
else {
/* Steal the pointer. */
begin_ = other.begin_;
end_ = other.end_;
capacity_end_ = other.capacity_end_;
}
other.begin_ = other.inline_buffer_;
other.end_ = other.begin_;
other.capacity_end_ = other.begin_ + OtherInlineBufferCapacity;
UPDATE_VECTOR_SIZE(this);
UPDATE_VECTOR_SIZE(&other);
}
~Vector()
{
destruct_n(begin_, this->size());
if (!this->is_inline()) {
allocator_.deallocate(begin_);
}
}
Vector &operator=(const Vector &other)
{
return copy_assign_container(*this, other);
}
Vector &operator=(Vector &&other)
{
return move_assign_container(*this, std::move(other));
}
/**
* Get the value at the given index. This invokes undefined behavior when the index is out of
* bounds.
*/
const T &operator[](int64_t index) const
{
BLI_assert(index >= 0);
BLI_assert(index < this->size());
return begin_[index];
}
T &operator[](int64_t index)
{
BLI_assert(index >= 0);
BLI_assert(index < this->size());
return begin_[index];
}
operator Span<T>() const
{
return Span<T>(begin_, this->size());
}
operator MutableSpan<T>()
{
return MutableSpan<T>(begin_, this->size());
}
template<typename U, BLI_ENABLE_IF((is_span_convertible_pointer_v<T, U>))>
operator Span<U>() const
{
return Span<U>(begin_, this->size());
}
template<typename U, BLI_ENABLE_IF((is_span_convertible_pointer_v<T, U>))>
operator MutableSpan<U>()
{
return MutableSpan<U>(begin_, this->size());
}
Span<T> as_span() const
{
return *this;
}
MutableSpan<T> as_mutable_span()
{
return *this;
}
/**
* Make sure that enough memory is allocated to hold min_capacity elements.
* This won't necessarily make an allocation when min_capacity is small.
* The actual size of the vector does not change.
*/
void reserve(const int64_t min_capacity)
{
if (min_capacity > this->capacity()) {
this->realloc_to_at_least(min_capacity);
}
}
/**
* Change the size of the vector so that it contains new_size elements.
* If new_size is smaller than the old size, the elements at the end of the vector are
* destructed. If new_size is larger than the old size, the new elements at the end are default
* constructed. If T is trivially constructible, the memory is not touched by this function.
*/
void resize(const int64_t new_size)
{
BLI_assert(new_size >= 0);
const int64_t old_size = this->size();
if (new_size > old_size) {
this->reserve(new_size);
default_construct_n(begin_ + old_size, new_size - old_size);
}
else {
destruct_n(begin_ + new_size, old_size - new_size);
}
end_ = begin_ + new_size;
UPDATE_VECTOR_SIZE(this);
}
/**
* Change the size of the vector so that it contains new_size elements.
* If new_size is smaller than the old size, the elements at the end of the vector are
* destructed. If new_size is larger than the old size, the new elements will be copy constructed
* from the given value.
*/
void resize(const int64_t new_size, const T &value)
{
BLI_assert(new_size >= 0);
const int64_t old_size = this->size();
if (new_size > old_size) {
this->reserve(new_size);
uninitialized_fill_n(begin_ + old_size, new_size - old_size, value);
}
else {
destruct_n(begin_ + new_size, old_size - new_size);
}
end_ = begin_ + new_size;
UPDATE_VECTOR_SIZE(this);
}
Curves: Port curve to mesh node to the new data-block This commit changes the Curve to Mesh node to work with `Curves` instead of `CurveEval`. The change ends up basically completely rewriting the node, since the different attribute storage means that the decisions made previously don't make much sense anymore. The main loops are now "for each attribute: for each curve combination" rather than the other way around, with the goal of taking advantage of the locality of curve attributes. This improvement is quite noticeable with many small curves; I measured a 4-5x improvement (around 4-5s to <1s) when converting millions of curves to tens of millions of faces. I didn't obverse any change in performance compared to 3.1 with fewer curves though. The changes also solve an algorithmic flaw where any interpolated attributes would be evaluated for every curve combination instead of just once per curve. This can be a large improvement when there are many profile curves. The code relies heavily on a function `foreach_curve_combination` which calculates some basic information about each combination and calls a templated function. I made assumptions about unnecessary reads being removed by compiler optimizations. For further performance improvements in the future that might be an area to investigate. Another might be using a "for a group of curves: for each attribute: for each curve" pattern to increase the locality of memory access. Differential Revision: https://developer.blender.org/D14642
2022-04-15 10:14:54 -05:00
/**
* Reset the size of the vector so that it contains new_size elements.
* All existing elements are destructed, and not copied if the data must be reallocated.
*/
void reinitialize(const int64_t new_size)
{
this->clear();
this->resize(new_size);
}
/**
* Afterwards the vector has 0 elements, but will still have
* memory to be refilled again.
*/
void clear()
{
destruct_n(begin_, this->size());
end_ = begin_;
UPDATE_VECTOR_SIZE(this);
}
/**
* Afterwards the vector has 0 elements and any allocated memory
* will be freed.
*/
void clear_and_make_inline()
{
destruct_n(begin_, this->size());
if (!this->is_inline()) {
allocator_.deallocate(begin_);
}
begin_ = inline_buffer_;
end_ = begin_;
capacity_end_ = begin_ + InlineBufferCapacity;
UPDATE_VECTOR_SIZE(this);
}
/**
* Insert a new element at the end of the vector.
* This might cause a reallocation with the capacity is exceeded.
*
* This is similar to std::vector::push_back.
*/
void append(const T &value)
{
this->append_as(value);
}
void append(T &&value)
{
this->append_as(std::move(value));
}
2021-04-19 23:56:12 +10:00
/* This is similar to `std::vector::emplace_back`. */
template<typename... ForwardValue> void append_as(ForwardValue &&...value)
{
this->ensure_space_for_one();
this->append_unchecked_as(std::forward<ForwardValue>(value)...);
}
/**
* Append the value to the vector and return the index that can be used to access the newly
* added value.
*/
int64_t append_and_get_index(const T &value)
{
const int64_t index = this->size();
this->append(value);
return index;
}
/**
* Append the value if it is not yet in the vector. This has to do a linear search to check if
* the value is in the vector. Therefore, this should only be called when it is known that the
* vector is small.
*/
void append_non_duplicates(const T &value)
{
if (!this->contains(value)) {
this->append(value);
}
}
/**
* Append the value and assume that vector has enough memory reserved. This invokes undefined
* behavior when not enough capacity has been reserved beforehand. Only use this in performance
* critical code.
*/
void append_unchecked(const T &value)
{
this->append_unchecked_as(value);
}
void append_unchecked(T &&value)
{
this->append_unchecked_as(std::move(value));
}
template<typename... ForwardT> void append_unchecked_as(ForwardT &&...value)
{
BLI_assert(end_ < capacity_end_);
new (end_) T(std::forward<ForwardT>(value)...);
end_++;
UPDATE_VECTOR_SIZE(this);
}
/**
* Insert the same element n times at the end of the vector.
* This might result in a reallocation internally.
*/
void append_n_times(const T &value, const int64_t n)
{
BLI_assert(n >= 0);
this->reserve(this->size() + n);
uninitialized_fill_n(end_, n, value);
this->increase_size_by_unchecked(n);
}
/**
* Enlarges the size of the internal buffer that is considered to be initialized.
* This invokes undefined behavior when the new size is larger than the capacity.
* The method can be useful when you want to call constructors in the vector yourself.
* This should only be done in very rare cases and has to be justified every time.
*/
void increase_size_by_unchecked(const int64_t n) noexcept
{
BLI_assert(end_ + n <= capacity_end_);
end_ += n;
UPDATE_VECTOR_SIZE(this);
}
/**
* Copy the elements of another array to the end of this vector.
*
* This can be used to emulate parts of std::vector::insert.
*/
void extend(Span<T> array)
{
this->extend(array.data(), array.size());
}
void extend(const T *start, int64_t amount)
{
this->reserve(this->size() + amount);
this->extend_unchecked(start, amount);
}
/**
* Adds all elements from the array that are not already in the vector. This is an expensive
* operation when the vector is large, but can be very cheap when it is known that the vector is
* small.
*/
void extend_non_duplicates(Span<T> array)
{
for (const T &value : array) {
this->append_non_duplicates(value);
}
}
/**
* Extend the vector without bounds checking. It is assumed that enough memory has been reserved
* beforehand. Only use this in performance critical code.
*/
void extend_unchecked(Span<T> array)
{
this->extend_unchecked(array.data(), array.size());
}
void extend_unchecked(const T *start, int64_t amount)
{
BLI_assert(amount >= 0);
BLI_assert(begin_ + amount <= capacity_end_);
uninitialized_copy_n(start, amount, end_);
end_ += amount;
UPDATE_VECTOR_SIZE(this);
}
template<typename InputIt> void extend(InputIt first, InputIt last)
{
this->insert(this->end(), first, last);
}
/**
* Insert elements into the vector at the specified position. This has a running time of O(n)
* where n is the number of values that have to be moved. Undefined behavior is invoked when the
* insert position is out of bounds.
*/
void insert(const int64_t insert_index, const T &value)
{
this->insert(insert_index, Span<T>(&value, 1));
}
void insert(const int64_t insert_index, T &&value)
{
this->insert(
insert_index, std::make_move_iterator(&value), std::make_move_iterator(&value + 1));
}
void insert(const int64_t insert_index, Span<T> array)
{
this->insert(begin_ + insert_index, array.begin(), array.end());
}
template<typename InputIt> void insert(const T *insert_position, InputIt first, InputIt last)
{
const int64_t insert_index = insert_position - begin_;
this->insert(insert_index, first, last);
}
template<typename InputIt> void insert(const int64_t insert_index, InputIt first, InputIt last)
{
BLI_assert(insert_index >= 0);
BLI_assert(insert_index <= this->size());
const int64_t insert_amount = std::distance(first, last);
const int64_t old_size = this->size();
const int64_t new_size = old_size + insert_amount;
const int64_t move_amount = old_size - insert_index;
this->reserve(new_size);
for (int64_t i = 0; i < move_amount; i++) {
const int64_t src_index = insert_index + move_amount - i - 1;
const int64_t dst_index = new_size - i - 1;
try {
new (static_cast<void *>(begin_ + dst_index)) T(std::move(begin_[src_index]));
}
catch (...) {
/* Destruct all values that have been moved already. */
destruct_n(begin_ + dst_index + 1, i);
end_ = begin_ + src_index + 1;
UPDATE_VECTOR_SIZE(this);
throw;
}
begin_[src_index].~T();
}
try {
std::uninitialized_copy_n(first, insert_amount, begin_ + insert_index);
}
catch (...) {
/* Destruct all values that have been moved. */
destruct_n(begin_ + new_size - move_amount, move_amount);
end_ = begin_ + insert_index;
UPDATE_VECTOR_SIZE(this);
throw;
}
end_ = begin_ + new_size;
UPDATE_VECTOR_SIZE(this);
}
/**
* Insert values at the beginning of the vector. The has to move all the other elements, so it
* has a linear running time.
*/
void prepend(const T &value)
{
this->insert(0, value);
}
void prepend(T &&value)
{
this->insert(0, std::move(value));
}
void prepend(Span<T> values)
{
this->insert(0, values);
}
template<typename InputIt> void prepend(InputIt first, InputIt last)
{
this->insert(0, first, last);
}
/**
* Return a reference to the nth last element.
* This invokes undefined behavior when the vector is too short.
*/
const T &last(const int64_t n = 0) const
{
BLI_assert(n >= 0);
BLI_assert(n < this->size());
return *(end_ - 1 - n);
}
T &last(const int64_t n = 0)
{
BLI_assert(n >= 0);
BLI_assert(n < this->size());
return *(end_ - 1 - n);
}
/**
* Return a reference to the first element in the vector.
* This invokes undefined behavior when the vector is empty.
*/
const T &first() const
{
BLI_assert(this->size() > 0);
return *begin_;
}
T &first()
{
BLI_assert(this->size() > 0);
return *begin_;
}
/**
* Return how many values are currently stored in the vector.
*/
int64_t size() const
{
const int64_t current_size = static_cast<int64_t>(end_ - begin_);
BLI_assert(debug_size_ == current_size);
return current_size;
}
/**
* Returns true when the vector contains no elements, otherwise false.
*
* This is the same as std::vector::empty.
*/
bool is_empty() const
{
return begin_ == end_;
}
/**
* Destructs the last element and decreases the size by one. This invokes undefined behavior when
* the vector is empty.
*/
void remove_last()
{
BLI_assert(!this->is_empty());
end_--;
end_->~T();
UPDATE_VECTOR_SIZE(this);
}
/**
* Remove the last element from the vector and return it. This invokes undefined behavior when
* the vector is empty.
*
* This is similar to std::vector::pop_back.
*/
T pop_last()
{
BLI_assert(!this->is_empty());
T value = std::move(*(end_ - 1));
end_--;
end_->~T();
UPDATE_VECTOR_SIZE(this);
return value;
}
/**
* Delete any element in the vector. The empty space will be filled by the previously last
* element. This takes O(1) time.
*/
void remove_and_reorder(const int64_t index)
{
BLI_assert(index >= 0);
BLI_assert(index < this->size());
T *element_to_remove = begin_ + index;
T *last_element = end_ - 1;
if (element_to_remove < last_element) {
*element_to_remove = std::move(*last_element);
}
end_ = last_element;
last_element->~T();
UPDATE_VECTOR_SIZE(this);
}
/**
2020-06-25 23:13:02 +10:00
* Finds the first occurrence of the value, removes it and copies the last element to the hole in
* the vector. This takes O(n) time.
*/
void remove_first_occurrence_and_reorder(const T &value)
{
const int64_t index = this->first_index_of(value);
this->remove_and_reorder(index);
}
/**
* Remove the element at the given index and move all values coming after it one towards the
* front. This takes O(n) time. If the order is not important, remove_and_reorder should be used
* instead.
*
* This is similar to std::vector::erase.
*/
void remove(const int64_t index)
{
BLI_assert(index >= 0);
BLI_assert(index < this->size());
const int64_t last_index = this->size() - 1;
for (int64_t i = index; i < last_index; i++) {
begin_[i] = std::move(begin_[i + 1]);
}
begin_[last_index].~T();
end_--;
UPDATE_VECTOR_SIZE(this);
}
/**
* Remove a contiguous chunk of elements and move all values coming after it towards the front.
* This takes O(n) time.
*
* This is similar to std::vector::erase.
*/
void remove(const int64_t start_index, const int64_t amount)
{
const int64_t old_size = this->size();
BLI_assert(start_index >= 0);
BLI_assert(amount >= 0);
BLI_assert(start_index + amount <= old_size);
const int64_t move_amount = old_size - start_index - amount;
for (int64_t i = 0; i < move_amount; i++) {
begin_[start_index + i] = std::move(begin_[start_index + amount + i]);
}
destruct_n(end_ - amount, amount);
end_ -= amount;
UPDATE_VECTOR_SIZE(this);
}
/**
* Do a linear search to find the value in the vector.
* When found, return the first index, otherwise return -1.
*/
int64_t first_index_of_try(const T &value) const
{
2020-07-03 14:52:51 +02:00
for (const T *current = begin_; current != end_; current++) {
if (*current == value) {
return static_cast<int64_t>(current - begin_);
}
}
return -1;
}
/**
* Do a linear search to find the value in the vector and return the found index. This invokes
* undefined behavior when the value is not in the vector.
*/
int64_t first_index_of(const T &value) const
{
const int64_t index = this->first_index_of_try(value);
BLI_assert(index >= 0);
return index;
}
/**
* Do a linear search to see of the value is in the vector.
* Return true when it exists, otherwise false.
*/
bool contains(const T &value) const
{
return this->first_index_of_try(value) != -1;
}
2020-07-20 13:02:10 +02:00
/**
* Copies the given value to every element in the vector.
*/
void fill(const T &value) const
{
initialized_fill_n(begin_, this->size(), value);
}
/**
* Get access to the underlying array.
*/
T *data()
{
return begin_;
}
/**
* Get access to the underlying array.
*/
const T *data() const
{
return begin_;
}
T *begin()
{
return begin_;
}
T *end()
{
return end_;
}
const T *begin() const
{
return begin_;
}
const T *end() const
{
return end_;
}
std::reverse_iterator<T *> rbegin()
{
return std::reverse_iterator<T *>(this->end());
}
std::reverse_iterator<T *> rend()
{
return std::reverse_iterator<T *>(this->begin());
}
std::reverse_iterator<const T *> rbegin() const
{
return std::reverse_iterator<T *>(this->end());
}
std::reverse_iterator<const T *> rend() const
{
return std::reverse_iterator<T *>(this->begin());
}
/**
* Get the current capacity of the vector, i.e. the maximum number of elements the vector can
* hold, before it has to reallocate.
*/
int64_t capacity() const
{
return static_cast<int64_t>(capacity_end_ - begin_);
}
/**
* Get an index range that makes looping over all indices more convenient and less error prone.
* Obviously, this should only be used when you actually need the index in the loop.
*
* Example:
* for (int64_t i : myvector.index_range()) {
* do_something(i, my_vector[i]);
* }
*/
IndexRange index_range() const
{
return IndexRange(this->size());
}
friend bool operator==(const Vector &a, const Vector &b)
{
return a.as_span() == b.as_span();
}
friend bool operator!=(const Vector &a, const Vector &b)
{
return !(a == b);
}
/**
* Print some debug information about the vector.
*/
void print_stats(StringRef name = "") const
{
std::cout << "Vector Stats: " << name << "\n";
std::cout << " Address: " << this << "\n";
std::cout << " Elements: " << this->size() << "\n";
std::cout << " Capacity: " << (capacity_end_ - begin_) << "\n";
std::cout << " Inline Capacity: " << InlineBufferCapacity << "\n";
char memory_size_str[15];
BLI_str_format_byte_unit(memory_size_str, sizeof(*this), true);
std::cout << " Size on Stack: " << memory_size_str << "\n";
}
private:
bool is_inline() const
{
return begin_ == inline_buffer_;
}
void ensure_space_for_one()
{
if (UNLIKELY(end_ >= capacity_end_)) {
this->realloc_to_at_least(this->size() + 1);
}
}
BLI_NOINLINE void realloc_to_at_least(const int64_t min_capacity)
{
if (this->capacity() >= min_capacity) {
return;
}
/* At least double the size of the previous allocation. Otherwise consecutive calls to grow can
* cause a reallocation every time even though min_capacity only increments. */
const int64_t min_new_capacity = this->capacity() * 2;
const int64_t new_capacity = std::max(min_capacity, min_new_capacity);
const int64_t size = this->size();
T *new_array = static_cast<T *>(
allocator_.allocate(static_cast<size_t>(new_capacity) * sizeof(T), alignof(T), AT));
try {
uninitialized_relocate_n(begin_, size, new_array);
}
catch (...) {
allocator_.deallocate(new_array);
throw;
}
if (!this->is_inline()) {
allocator_.deallocate(begin_);
}
begin_ = new_array;
end_ = begin_ + size;
capacity_end_ = begin_ + new_capacity;
}
};
#undef UPDATE_VECTOR_SIZE
/**
* Same as a normal Vector, but does not use Blender's guarded allocator. This is useful when
* allocating memory with static storage duration.
*/
template<typename T, int64_t InlineBufferCapacity = default_inline_buffer_capacity(sizeof(T))>
using RawVector = Vector<T, InlineBufferCapacity, RawAllocator>;
} /* namespace blender */