diff options
Diffstat (limited to 'core/templates')
-rw-r--r-- | core/templates/command_queue_mt.h | 10 | ||||
-rw-r--r-- | core/templates/cowdata.h | 57 | ||||
-rw-r--r-- | core/templates/hash_map.h | 779 | ||||
-rw-r--r-- | core/templates/hashfuncs.h | 118 | ||||
-rw-r--r-- | core/templates/oa_hash_map.h | 20 | ||||
-rw-r--r-- | core/templates/ordered_hash_map.h | 301 | ||||
-rw-r--r-- | core/templates/paged_allocator.h | 8 | ||||
-rw-r--r-- | core/templates/pair.h | 11 | ||||
-rw-r--r-- | core/templates/rb_map.h (renamed from core/templates/map.h) | 42 | ||||
-rw-r--r-- | core/templates/rb_set.h (renamed from core/templates/set.h) | 32 | ||||
-rw-r--r-- | core/templates/rid_owner.h | 2 | ||||
-rw-r--r-- | core/templates/safe_list.h | 38 | ||||
-rw-r--r-- | core/templates/self_list.h | 2 | ||||
-rw-r--r-- | core/templates/thread_work_pool.h | 4 | ||||
-rw-r--r-- | core/templates/vector.h | 2 |
15 files changed, 675 insertions, 751 deletions
diff --git a/core/templates/command_queue_mt.h b/core/templates/command_queue_mt.h index 1ecb81c2a2..7d3e31b5bc 100644 --- a/core/templates/command_queue_mt.h +++ b/core/templates/command_queue_mt.h @@ -215,7 +215,7 @@ T *instance; \ M method; \ SEMIC_SEP_LIST(PARAM_DECL, N); \ - virtual void call() { \ + virtual void call() override { \ (instance->*method)(COMMA_SEP_LIST(ARG, N)); \ } \ }; @@ -227,7 +227,7 @@ T *instance; \ M method; \ SEMIC_SEP_LIST(PARAM_DECL, N); \ - virtual void call() { \ + virtual void call() override { \ *ret = (instance->*method)(COMMA_SEP_LIST(ARG, N)); \ } \ }; @@ -238,7 +238,7 @@ T *instance; \ M method; \ SEMIC_SEP_LIST(PARAM_DECL, N); \ - virtual void call() { \ + virtual void call() override { \ (instance->*method)(COMMA_SEP_LIST(ARG, N)); \ } \ }; @@ -311,9 +311,9 @@ class CommandQueueMT { }; struct SyncCommand : public CommandBase { - SyncSemaphore *sync_sem; + SyncSemaphore *sync_sem = nullptr; - virtual void post() { + virtual void post() override { sync_sem->sem.post(); } }; diff --git a/core/templates/cowdata.h b/core/templates/cowdata.h index 326616b607..e760fc2176 100644 --- a/core/templates/cowdata.h +++ b/core/templates/cowdata.h @@ -86,13 +86,6 @@ private: return reinterpret_cast<uint32_t *>(_ptr) - 1; } - _FORCE_INLINE_ T *_get_data() const { - if (!_ptr) { - return nullptr; - } - return reinterpret_cast<T *>(_ptr); - } - _FORCE_INLINE_ size_t _get_alloc_size(size_t p_elements) const { return next_power_of_2(p_elements * sizeof(T)); } @@ -128,11 +121,11 @@ public: _FORCE_INLINE_ T *ptrw() { _copy_on_write(); - return (T *)_get_data(); + return _ptr; } _FORCE_INLINE_ const T *ptr() const { - return _get_data(); + return _ptr; } _FORCE_INLINE_ int size() const { @@ -150,19 +143,19 @@ public: _FORCE_INLINE_ void set(int p_index, const T &p_elem) { ERR_FAIL_INDEX(p_index, size()); _copy_on_write(); - _get_data()[p_index] = p_elem; + _ptr[p_index] = p_elem; } _FORCE_INLINE_ T &get_m(int p_index) { CRASH_BAD_INDEX(p_index, size()); _copy_on_write(); - return _get_data()[p_index]; + return _ptr[p_index]; } _FORCE_INLINE_ const T &get(int p_index) const { CRASH_BAD_INDEX(p_index, size()); - return _get_data()[p_index]; + return _ptr[p_index]; } Error resize(int p_size); @@ -190,6 +183,8 @@ public: } int find(const T &p_val, int p_from = 0) const; + int rfind(const T &p_val, int p_from = -1) const; + int count(const T &p_val) const; _FORCE_INLINE_ CowData() {} _FORCE_INLINE_ ~CowData(); @@ -249,7 +244,7 @@ uint32_t CowData<T>::_copy_on_write() { } else { for (uint32_t i = 0; i < current_size; i++) { - memnew_placement(&_data[i], T(_get_data()[i])); + memnew_placement(&_data[i], T(_ptr[i])); } } @@ -308,10 +303,8 @@ Error CowData<T>::resize(int p_size) { // construct the newly created elements if (!__has_trivial_constructor(T)) { - T *elems = _get_data(); - for (int i = *_get_size(); i < p_size; i++) { - memnew_placement(&elems[i], T); + memnew_placement(&_ptr[i], T); } } @@ -321,7 +314,7 @@ Error CowData<T>::resize(int p_size) { if (!__has_trivial_destructor(T)) { // deinitialize no longer needed elements for (uint32_t i = p_size; i < *_get_size(); i++) { - T *t = &_get_data()[i]; + T *t = &_ptr[i]; t->~T(); } } @@ -359,6 +352,36 @@ int CowData<T>::find(const T &p_val, int p_from) const { } template <class T> +int CowData<T>::rfind(const T &p_val, int p_from) const { + const int s = size(); + + if (p_from < 0) { + p_from = s + p_from; + } + if (p_from < 0 || p_from >= s) { + p_from = s - 1; + } + + for (int i = p_from; i >= 0; i--) { + if (get(i) == p_val) { + return i; + } + } + return -1; +} + +template <class T> +int CowData<T>::count(const T &p_val) const { + int amount = 0; + for (int i = 0; i < size(); i++) { + if (get(i) == p_val) { + amount++; + } + } + return amount; +} + +template <class T> void CowData<T>::_ref(const CowData *p_from) { _ref(*p_from); } diff --git a/core/templates/hash_map.h b/core/templates/hash_map.h index fa5677cc70..e5f73171a2 100644 --- a/core/templates/hash_map.h +++ b/core/templates/hash_map.h @@ -31,524 +31,557 @@ #ifndef HASH_MAP_H #define HASH_MAP_H -#include "core/error/error_macros.h" #include "core/math/math_funcs.h" #include "core/os/memory.h" -#include "core/string/ustring.h" #include "core/templates/hashfuncs.h" -#include "core/templates/list.h" +#include "core/templates/paged_allocator.h" +#include "core/templates/pair.h" /** - * @class HashMap + * A HashMap implementation that uses open addressing with Robin Hood hashing. + * Robin Hood hashing swaps out entries that have a smaller probing distance + * than the to-be-inserted entry, that evens out the average probing distance + * and enables faster lookups. Backward shift deletion is employed to further + * improve the performance and to avoid infinite loops in rare cases. * - * Implementation of a standard Hashing HashMap, for quick lookups of Data associated with a Key. - * The implementation provides hashers for the default types, if you need a special kind of hasher, provide - * your own. - * @param TKey Key, search is based on it, needs to be hasheable. It is unique in this container. - * @param TData Data, data associated with the key - * @param Hasher Hasher object, needs to provide a valid static hash function for TKey - * @param Comparator comparator object, needs to be able to safely compare two TKey values. - * It needs to ensure that x == x for any items inserted in the map. Bear in mind that nan != nan when implementing an equality check. - * @param MIN_HASH_TABLE_POWER Miminum size of the hash table, as a power of two. You rarely need to change this parameter. - * @param RELATIONSHIP Relationship at which the hash table is resized. if amount of elements is RELATIONSHIP - * times bigger than the hash table, table is resized to solve this condition. if RELATIONSHIP is zero, table is always MIN_HASH_TABLE_POWER. + * Keys and values are stored in a double linked list by insertion order. This + * has a slight performance overhead on lookup, which can be mostly compensated + * using a paged allocator if required. * + * The assignment operator copy the pairs from one map to the other. */ -template <class TKey, class TData, class Hasher = HashMapHasherDefault, class Comparator = HashMapComparatorDefault<TKey>, uint8_t MIN_HASH_TABLE_POWER = 3, uint8_t RELATIONSHIP = 8> +template <class TKey, class TValue> +struct HashMapElement { + HashMapElement *next = nullptr; + HashMapElement *prev = nullptr; + KeyValue<TKey, TValue> data; + HashMapElement() {} + HashMapElement(const TKey &p_key, const TValue &p_value) : + data(p_key, p_value) {} +}; + +template <class TKey, class TValue, + class Hasher = HashMapHasherDefault, + class Comparator = HashMapComparatorDefault<TKey>, + class Allocator = DefaultTypedAllocator<HashMapElement<TKey, TValue>>> class HashMap { public: - struct Pair { - TKey key; - TData data; + const uint32_t MIN_CAPACITY_INDEX = 2; // Use a prime. + const float MAX_OCCUPANCY = 0.75; + const uint32_t EMPTY_HASH = 0; - Pair(const TKey &p_key) : - key(p_key), - data() {} - Pair(const TKey &p_key, const TData &p_data) : - key(p_key), - data(p_data) { - } - }; +private: + Allocator element_alloc; + HashMapElement<TKey, TValue> **elements = nullptr; + uint32_t *hashes = nullptr; + HashMapElement<TKey, TValue> *head_element = nullptr; + HashMapElement<TKey, TValue> *tail_element = nullptr; - struct Element { - private: - friend class HashMap; + uint32_t capacity_index = 0; + uint32_t num_elements = 0; - uint32_t hash = 0; - Element *next = nullptr; - Element() {} - Pair pair; + _FORCE_INLINE_ uint32_t _hash(const TKey &p_key) const { + uint32_t hash = Hasher::hash(p_key); - public: - const TKey &key() const { - return pair.key; + if (unlikely(hash == EMPTY_HASH)) { + hash = EMPTY_HASH + 1; } - TData &value() { - return pair.data; - } + return hash; + } - const TData &value() const { - return pair.value(); + _FORCE_INLINE_ uint32_t _get_probe_length(uint32_t p_pos, uint32_t p_hash, uint32_t p_capacity) const { + uint32_t original_pos = p_hash % p_capacity; + return (p_pos - original_pos + p_capacity) % p_capacity; + } + + bool _lookup_pos(const TKey &p_key, uint32_t &r_pos) const { + if (elements == nullptr) { + return false; // Failed lookups, no elements } - Element(const TKey &p_key) : - pair(p_key) {} - Element(const Element &p_other) : - hash(p_other.hash), - pair(p_other.pair.key, p_other.pair.data) {} - }; + uint32_t capacity = hash_table_size_primes[capacity_index]; + uint32_t hash = _hash(p_key); + uint32_t pos = hash % capacity; + uint32_t distance = 0; -private: - Element **hash_table = nullptr; - uint8_t hash_table_power = 0; - uint32_t elements = 0; + while (true) { + if (hashes[pos] == EMPTY_HASH) { + return false; + } - void make_hash_table() { - ERR_FAIL_COND(hash_table); + if (distance > _get_probe_length(pos, hashes[pos], capacity)) { + return false; + } - hash_table = memnew_arr(Element *, (1 << MIN_HASH_TABLE_POWER)); + if (hashes[pos] == hash && Comparator::compare(elements[pos]->data.key, p_key)) { + r_pos = pos; + return true; + } - hash_table_power = MIN_HASH_TABLE_POWER; - elements = 0; - for (int i = 0; i < (1 << MIN_HASH_TABLE_POWER); i++) { - hash_table[i] = nullptr; + pos = (pos + 1) % capacity; + distance++; } } - void erase_hash_table() { - ERR_FAIL_COND_MSG(elements, "Cannot erase hash table if there are still elements inside."); - - memdelete_arr(hash_table); - hash_table = nullptr; - hash_table_power = 0; - elements = 0; - } + void _insert_with_hash(uint32_t p_hash, HashMapElement<TKey, TValue> *p_value) { + uint32_t capacity = hash_table_size_primes[capacity_index]; + uint32_t hash = p_hash; + HashMapElement<TKey, TValue> *value = p_value; + uint32_t distance = 0; + uint32_t pos = hash % capacity; - void check_hash_table() { - int new_hash_table_power = -1; + while (true) { + if (hashes[pos] == EMPTY_HASH) { + elements[pos] = value; + hashes[pos] = hash; - if ((int)elements > ((1 << hash_table_power) * RELATIONSHIP)) { - /* rehash up */ - new_hash_table_power = hash_table_power + 1; + num_elements++; - while ((int)elements > ((1 << new_hash_table_power) * RELATIONSHIP)) { - new_hash_table_power++; + return; } - } else if ((hash_table_power > (int)MIN_HASH_TABLE_POWER) && ((int)elements < ((1 << (hash_table_power - 1)) * RELATIONSHIP))) { - /* rehash down */ - new_hash_table_power = hash_table_power - 1; - - while ((int)elements < ((1 << (new_hash_table_power - 1)) * RELATIONSHIP)) { - new_hash_table_power--; + // Not an empty slot, let's check the probing length of the existing one. + uint32_t existing_probe_len = _get_probe_length(pos, hashes[pos], capacity); + if (existing_probe_len < distance) { + SWAP(hash, hashes[pos]); + SWAP(value, elements[pos]); + distance = existing_probe_len; } - if (new_hash_table_power < (int)MIN_HASH_TABLE_POWER) { - new_hash_table_power = MIN_HASH_TABLE_POWER; - } + pos = (pos + 1) % capacity; + distance++; } + } - if (new_hash_table_power == -1) { - return; - } + void _resize_and_rehash(uint32_t p_new_capacity_index) { + uint32_t old_capacity = hash_table_size_primes[capacity_index]; - Element **new_hash_table = memnew_arr(Element *, ((uint64_t)1 << new_hash_table_power)); - ERR_FAIL_COND_MSG(!new_hash_table, "Out of memory."); + // Capacity can't be 0. + capacity_index = MAX((uint32_t)MIN_CAPACITY_INDEX, p_new_capacity_index); - for (int i = 0; i < (1 << new_hash_table_power); i++) { - new_hash_table[i] = nullptr; - } + uint32_t capacity = hash_table_size_primes[capacity_index]; - if (hash_table) { - for (int i = 0; i < (1 << hash_table_power); i++) { - while (hash_table[i]) { - Element *se = hash_table[i]; - hash_table[i] = se->next; - int new_pos = se->hash & ((1 << new_hash_table_power) - 1); - se->next = new_hash_table[new_pos]; - new_hash_table[new_pos] = se; - } - } + HashMapElement<TKey, TValue> **old_elements = elements; + uint32_t *old_hashes = hashes; - memdelete_arr(hash_table); - } - hash_table = new_hash_table; - hash_table_power = new_hash_table_power; - } + num_elements = 0; + hashes = reinterpret_cast<uint32_t *>(Memory::alloc_static(sizeof(uint32_t) * capacity)); + elements = reinterpret_cast<HashMapElement<TKey, TValue> **>(Memory::alloc_static(sizeof(HashMapElement<TKey, TValue> *) * capacity)); - /* I want to have only one function.. */ - _FORCE_INLINE_ const Element *get_element(const TKey &p_key) const { - uint32_t hash = Hasher::hash(p_key); - uint32_t index = hash & ((1 << hash_table_power) - 1); + for (uint32_t i = 0; i < capacity; i++) { + hashes[i] = 0; + elements[i] = nullptr; + } - Element *e = hash_table[index]; + if (old_capacity == 0) { + // Nothing to do. + return; + } - while (e) { - /* checking hash first avoids comparing key, which may take longer */ - if (e->hash == hash && Comparator::compare(e->pair.key, p_key)) { - /* the pair exists in this hashtable, so just update data */ - return e; + for (uint32_t i = 0; i < old_capacity; i++) { + if (old_hashes[i] == EMPTY_HASH) { + continue; } - e = e->next; + _insert_with_hash(old_hashes[i], old_elements[i]); } - return nullptr; + Memory::free_static(old_elements); + Memory::free_static(old_hashes); } - Element *create_element(const TKey &p_key) { - /* if element doesn't exist, create it */ - Element *e = memnew(Element(p_key)); - ERR_FAIL_COND_V_MSG(!e, nullptr, "Out of memory."); - uint32_t hash = Hasher::hash(p_key); - uint32_t index = hash & ((1 << hash_table_power) - 1); - e->next = hash_table[index]; - e->hash = hash; - - hash_table[index] = e; - elements++; + _FORCE_INLINE_ HashMapElement<TKey, TValue> *_insert(const TKey &p_key, const TValue &p_value, bool p_front_insert = false) { + uint32_t capacity = hash_table_size_primes[capacity_index]; + if (unlikely(elements == nullptr)) { + // Allocate on demand to save memory. - return e; - } + hashes = reinterpret_cast<uint32_t *>(Memory::alloc_static(sizeof(uint32_t) * capacity)); + elements = reinterpret_cast<HashMapElement<TKey, TValue> **>(Memory::alloc_static(sizeof(HashMapElement<TKey, TValue> *) * capacity)); - void copy_from(const HashMap &p_t) { - if (&p_t == this) { - return; /* much less bother with that */ - } - - clear(); - - if (!p_t.hash_table || p_t.hash_table_power == 0) { - return; /* not copying from empty table */ + for (uint32_t i = 0; i < capacity; i++) { + hashes[i] = EMPTY_HASH; + elements[i] = nullptr; + } } - hash_table = memnew_arr(Element *, (uint64_t)1 << p_t.hash_table_power); - hash_table_power = p_t.hash_table_power; - elements = p_t.elements; + uint32_t pos = 0; + bool exists = _lookup_pos(p_key, pos); - for (int i = 0; i < (1 << p_t.hash_table_power); i++) { - hash_table[i] = nullptr; - - const Element *e = p_t.hash_table[i]; - - while (e) { - Element *le = memnew(Element(*e)); /* local element */ + if (exists) { + elements[pos]->data.value = p_value; + return elements[pos]; + } else { + if (num_elements + 1 > MAX_OCCUPANCY * capacity) { + ERR_FAIL_COND_V_MSG(capacity_index + 1 == HASH_TABLE_SIZE_MAX, nullptr, "Hash table maximum capacity reached, aborting insertion."); + _resize_and_rehash(capacity_index + 1); + } - /* add to list and reassign pointers */ - le->next = hash_table[i]; - hash_table[i] = le; + HashMapElement<TKey, TValue> *elem = element_alloc.new_allocation(HashMapElement<TKey, TValue>(p_key, p_value)); - e = e->next; + if (tail_element == nullptr) { + head_element = elem; + tail_element = elem; + } else if (p_front_insert) { + head_element->prev = elem; + elem->next = head_element; + head_element = elem; + } else { + tail_element->next = elem; + elem->prev = tail_element; + tail_element = elem; } + + uint32_t hash = _hash(p_key); + _insert_with_hash(hash, elem); + return elem; } } public: - Element *set(const TKey &p_key, const TData &p_data) { - return set(Pair(p_key, p_data)); - } + _FORCE_INLINE_ uint32_t get_capacity() const { return hash_table_size_primes[capacity_index]; } + _FORCE_INLINE_ uint32_t size() const { return num_elements; } - Element *set(const Pair &p_pair) { - Element *e = nullptr; - if (!hash_table) { - make_hash_table(); // if no table, make one - } else { - e = const_cast<Element *>(get_element(p_pair.key)); - } + /* Standard Godot Container API */ - /* if we made it up to here, the pair doesn't exist, create and assign */ + bool is_empty() const { + return num_elements == 0; + } - if (!e) { - e = create_element(p_pair.key); - if (!e) { - return nullptr; - } - check_hash_table(); // perform mantenience routine + void clear() { + if (elements == nullptr) { + return; } + uint32_t capacity = hash_table_size_primes[capacity_index]; + for (uint32_t i = 0; i < capacity; i++) { + if (hashes[i] == EMPTY_HASH) { + continue; + } - e->pair.data = p_pair.data; - return e; - } + hashes[i] = EMPTY_HASH; + element_alloc.delete_allocation(elements[i]); + elements[i] = nullptr; + } - bool has(const TKey &p_key) const { - return getptr(p_key) != nullptr; + tail_element = nullptr; + head_element = nullptr; + num_elements = 0; } - /** - * Get a key from data, return a const reference. - * WARNING: this doesn't check errors, use either getptr and check nullptr, or check - * first with has(key) - */ - - const TData &get(const TKey &p_key) const { - const TData *res = getptr(p_key); - CRASH_COND_MSG(!res, "Map key not found."); - return *res; + TValue &get(const TKey &p_key) { + uint32_t pos = 0; + bool exists = _lookup_pos(p_key, pos); + CRASH_COND_MSG(!exists, "HashMap key not found."); + return elements[pos]->data.value; } - TData &get(const TKey &p_key) { - TData *res = getptr(p_key); - CRASH_COND_MSG(!res, "Map key not found."); - return *res; + const TValue &get(const TKey &p_key) const { + uint32_t pos = 0; + bool exists = _lookup_pos(p_key, pos); + CRASH_COND_MSG(!exists, "HashMap key not found."); + return elements[pos]->data.value; } - /** - * Same as get, except it can return nullptr when item was not found. - * This is mainly used for speed purposes. - */ + const TValue *getptr(const TKey &p_key) const { + uint32_t pos = 0; + bool exists = _lookup_pos(p_key, pos); - _FORCE_INLINE_ TData *getptr(const TKey &p_key) { - if (unlikely(!hash_table)) { - return nullptr; + if (exists) { + return &elements[pos]->data.value; } + return nullptr; + } - Element *e = const_cast<Element *>(get_element(p_key)); + TValue *getptr(const TKey &p_key) { + uint32_t pos = 0; + bool exists = _lookup_pos(p_key, pos); - if (e) { - return &e->pair.data; + if (exists) { + return &elements[pos]->data.value; } - return nullptr; } - _FORCE_INLINE_ const TData *getptr(const TKey &p_key) const { - if (unlikely(!hash_table)) { - return nullptr; - } + _FORCE_INLINE_ bool has(const TKey &p_key) const { + uint32_t _pos = 0; + return _lookup_pos(p_key, _pos); + } - const Element *e = const_cast<Element *>(get_element(p_key)); + bool erase(const TKey &p_key) { + uint32_t pos = 0; + bool exists = _lookup_pos(p_key, pos); - if (e) { - return &e->pair.data; + if (!exists) { + return false; } - return nullptr; - } + uint32_t capacity = hash_table_size_primes[capacity_index]; + uint32_t next_pos = (pos + 1) % capacity; + while (hashes[next_pos] != EMPTY_HASH && _get_probe_length(next_pos, hashes[next_pos], capacity) != 0) { + SWAP(hashes[next_pos], hashes[pos]); + SWAP(elements[next_pos], elements[pos]); + pos = next_pos; + next_pos = (pos + 1) % capacity; + } - /** - * Same as get, except it can return nullptr when item was not found. - * This version is custom, will take a hash and a custom key (that should support operator==() - */ + hashes[pos] = EMPTY_HASH; - template <class C> - _FORCE_INLINE_ TData *custom_getptr(C p_custom_key, uint32_t p_custom_hash) { - if (unlikely(!hash_table)) { - return nullptr; + if (head_element == elements[pos]) { + head_element = elements[pos]->next; } - uint32_t hash = p_custom_hash; - uint32_t index = hash & ((1 << hash_table_power) - 1); - - Element *e = hash_table[index]; + if (tail_element == elements[pos]) { + tail_element = elements[pos]->prev; + } - while (e) { - /* checking hash first avoids comparing key, which may take longer */ - if (e->hash == hash && Comparator::compare(e->pair.key, p_custom_key)) { - /* the pair exists in this hashtable, so just update data */ - return &e->pair.data; - } + if (elements[pos]->prev) { + elements[pos]->prev->next = elements[pos]->next; + } - e = e->next; + if (elements[pos]->next) { + elements[pos]->next->prev = elements[pos]->prev; } - return nullptr; + element_alloc.delete_allocation(elements[pos]); + elements[pos] = nullptr; + + num_elements--; + return true; } - template <class C> - _FORCE_INLINE_ const TData *custom_getptr(C p_custom_key, uint32_t p_custom_hash) const { - if (unlikely(!hash_table)) { - return nullptr; + // Reserves space for a number of elements, useful to avoid many resizes and rehashes. + // If adding a known (possibly large) number of elements at once, must be larger than old capacity. + void reserve(uint32_t p_new_capacity) { + uint32_t new_index = capacity_index; + + while (hash_table_size_primes[new_index] < p_new_capacity) { + ERR_FAIL_COND_MSG(new_index + 1 == (uint32_t)HASH_TABLE_SIZE_MAX, nullptr); + new_index++; } - uint32_t hash = p_custom_hash; - uint32_t index = hash & ((1 << hash_table_power) - 1); + if (new_index == capacity_index) { + return; + } - const Element *e = hash_table[index]; + if (elements == nullptr) { + capacity_index = new_index; + return; // Unallocated yet. + } + _resize_and_rehash(new_index); + } - while (e) { - /* checking hash first avoids comparing key, which may take longer */ - if (e->hash == hash && Comparator::compare(e->pair.key, p_custom_key)) { - /* the pair exists in this hashtable, so just update data */ - return &e->pair.data; - } + /** Iterator API **/ - e = e->next; + struct ConstIterator { + _FORCE_INLINE_ const KeyValue<TKey, TValue> &operator*() const { + return E->data; + } + _FORCE_INLINE_ const KeyValue<TKey, TValue> *operator->() const { return &E->data; } + _FORCE_INLINE_ ConstIterator &operator++() { + if (E) { + E = E->next; + } + return *this; + } + _FORCE_INLINE_ ConstIterator &operator--() { + if (E) { + E = E->prev; + } + return *this; } - return nullptr; - } + _FORCE_INLINE_ bool operator==(const ConstIterator &b) const { return E == b.E; } + _FORCE_INLINE_ bool operator!=(const ConstIterator &b) const { return E != b.E; } - /** - * Erase an item, return true if erasing was successful - */ + _FORCE_INLINE_ explicit operator bool() const { + return E != nullptr; + } - bool erase(const TKey &p_key) { - if (unlikely(!hash_table)) { - return false; + _FORCE_INLINE_ ConstIterator(const HashMapElement<TKey, TValue> *p_E) { E = p_E; } + _FORCE_INLINE_ ConstIterator() {} + _FORCE_INLINE_ ConstIterator(const ConstIterator &p_it) { E = p_it.E; } + _FORCE_INLINE_ void operator=(const ConstIterator &p_it) { + E = p_it.E; } - uint32_t hash = Hasher::hash(p_key); - uint32_t index = hash & ((1 << hash_table_power) - 1); - - Element *e = hash_table[index]; - Element *p = nullptr; - while (e) { - /* checking hash first avoids comparing key, which may take longer */ - if (e->hash == hash && Comparator::compare(e->pair.key, p_key)) { - if (p) { - p->next = e->next; - } else { - //begin of list - hash_table[index] = e->next; - } - - memdelete(e); - elements--; - - if (elements == 0) { - erase_hash_table(); - } else { - check_hash_table(); - } - return true; + private: + const HashMapElement<TKey, TValue> *E = nullptr; + }; + + struct Iterator { + _FORCE_INLINE_ KeyValue<TKey, TValue> &operator*() const { + return E->data; + } + _FORCE_INLINE_ KeyValue<TKey, TValue> *operator->() const { return &E->data; } + _FORCE_INLINE_ Iterator &operator++() { + if (E) { + E = E->next; + } + return *this; + } + _FORCE_INLINE_ Iterator &operator--() { + if (E) { + E = E->prev; } + return *this; + } - p = e; - e = e->next; + _FORCE_INLINE_ bool operator==(const Iterator &b) const { return E == b.E; } + _FORCE_INLINE_ bool operator!=(const Iterator &b) const { return E != b.E; } + + _FORCE_INLINE_ explicit operator bool() const { + return E != nullptr; } - return false; - } + _FORCE_INLINE_ Iterator(HashMapElement<TKey, TValue> *p_E) { E = p_E; } + _FORCE_INLINE_ Iterator() {} + _FORCE_INLINE_ Iterator(const Iterator &p_it) { E = p_it.E; } + _FORCE_INLINE_ void operator=(const Iterator &p_it) { + E = p_it.E; + } + + operator ConstIterator() const { + return ConstIterator(E); + } - inline const TData &operator[](const TKey &p_key) const { //constref + private: + HashMapElement<TKey, TValue> *E = nullptr; + }; - return get(p_key); + _FORCE_INLINE_ Iterator begin() { + return Iterator(head_element); + } + _FORCE_INLINE_ Iterator end() { + return Iterator(nullptr); + } + _FORCE_INLINE_ Iterator last() { + return Iterator(tail_element); } - inline TData &operator[](const TKey &p_key) { //assignment - Element *e = nullptr; - if (!hash_table) { - make_hash_table(); // if no table, make one - } else { - e = const_cast<Element *>(get_element(p_key)); + _FORCE_INLINE_ Iterator find(const TKey &p_key) { + uint32_t pos = 0; + bool exists = _lookup_pos(p_key, pos); + if (!exists) { + return end(); } + return Iterator(elements[pos]); + } - /* if we made it up to here, the pair doesn't exist, create */ - if (!e) { - e = create_element(p_key); - CRASH_COND(!e); - check_hash_table(); // perform mantenience routine + _FORCE_INLINE_ void remove(const Iterator &p_iter) { + if (p_iter) { + erase(p_iter->key); } + } - return e->pair.data; + _FORCE_INLINE_ ConstIterator begin() const { + return ConstIterator(head_element); + } + _FORCE_INLINE_ ConstIterator end() const { + return ConstIterator(nullptr); + } + _FORCE_INLINE_ ConstIterator last() const { + return ConstIterator(tail_element); } - /** - * Get the next key to p_key, and the first key if p_key is null. - * Returns a pointer to the next key if found, nullptr otherwise. - * Adding/Removing elements while iterating will, of course, have unexpected results, don't do it. - * - * Example: - * - * const TKey *k=nullptr; - * - * while( (k=table.next(k)) ) { - * - * print( *k ); - * } - * - */ - const TKey *next(const TKey *p_key) const { - if (unlikely(!hash_table)) { - return nullptr; + _FORCE_INLINE_ ConstIterator find(const TKey &p_key) const { + uint32_t pos = 0; + bool exists = _lookup_pos(p_key, pos); + if (!exists) { + return end(); } + return ConstIterator(elements[pos]); + } - if (!p_key) { /* get the first key */ - - for (int i = 0; i < (1 << hash_table_power); i++) { - if (hash_table[i]) { - return &hash_table[i]->pair.key; - } - } - - } else { /* get the next key */ + /* Indexing */ - const Element *e = get_element(*p_key); - ERR_FAIL_COND_V_MSG(!e, nullptr, "Invalid key supplied."); - if (e->next) { - /* if there is a "next" in the list, return that */ - return &e->next->pair.key; - } else { - /* go to next elements */ - uint32_t index = e->hash & ((1 << hash_table_power) - 1); - index++; - for (int i = index; i < (1 << hash_table_power); i++) { - if (hash_table[i]) { - return &hash_table[i]->pair.key; - } - } - } + const TValue &operator[](const TKey &p_key) const { + uint32_t pos = 0; + bool exists = _lookup_pos(p_key, pos); + CRASH_COND(!exists); + return elements[pos]->data.value; + } - /* nothing found, was at end */ + TValue &operator[](const TKey &p_key) { + uint32_t pos = 0; + bool exists = _lookup_pos(p_key, pos); + if (!exists) { + return _insert(p_key, TValue())->data.value; + } else { + return elements[pos]->data.value; } - - return nullptr; /* nothing found */ } - inline unsigned int size() const { - return elements; - } + /* Insert */ - inline bool is_empty() const { - return elements == 0; + Iterator insert(const TKey &p_key, const TValue &p_value, bool p_front_insert = false) { + return Iterator(_insert(p_key, p_value, p_front_insert)); } - void clear() { - /* clean up */ - if (hash_table) { - for (int i = 0; i < (1 << hash_table_power); i++) { - while (hash_table[i]) { - Element *e = hash_table[i]; - hash_table[i] = e->next; - memdelete(e); - } - } + /* Constructors */ - memdelete_arr(hash_table); + HashMap(const HashMap &p_other) { + reserve(hash_table_size_primes[p_other.capacity_index]); + + if (p_other.num_elements == 0) { + return; } - hash_table = nullptr; - hash_table_power = 0; - elements = 0; + for (const KeyValue<TKey, TValue> &E : p_other) { + insert(E.key, E.value); + } } - void operator=(const HashMap &p_table) { - copy_from(p_table); - } + void operator=(const HashMap &p_other) { + if (this == &p_other) { + return; // Ignore self assignment. + } + if (num_elements != 0) { + clear(); + } - void get_key_list(List<TKey> *r_keys) const { - if (unlikely(!hash_table)) { - return; + reserve(hash_table_size_primes[p_other.capacity_index]); + + if (p_other.elements == nullptr) { + return; // Nothing to copy. } - for (int i = 0; i < (1 << hash_table_power); i++) { - Element *e = hash_table[i]; - while (e) { - r_keys->push_back(e->pair.key); - e = e->next; - } + + for (const KeyValue<TKey, TValue> &E : p_other) { + insert(E.key, E.value); } } - HashMap() {} + HashMap(uint32_t p_initial_capacity) { + // Capacity can't be 0. + capacity_index = 0; + reserve(p_initial_capacity); + } + HashMap() { + capacity_index = MIN_CAPACITY_INDEX; + } - HashMap(const HashMap &p_table) { - copy_from(p_table); + uint32_t debug_get_hash(uint32_t p_index) { + if (num_elements == 0) { + return 0; + } + ERR_FAIL_INDEX_V(p_index, get_capacity(), 0); + return hashes[p_index]; + } + Iterator debug_get_element(uint32_t p_index) { + if (num_elements == 0) { + return Iterator(); + } + ERR_FAIL_INDEX_V(p_index, get_capacity(), Iterator()); + return Iterator(elements[p_index]); } ~HashMap() { clear(); + + if (elements != nullptr) { + Memory::free_static(elements); + Memory::free_static(hashes); + } } }; diff --git a/core/templates/hashfuncs.h b/core/templates/hashfuncs.h index 2a129f97d5..1330d55270 100644 --- a/core/templates/hashfuncs.h +++ b/core/templates/hashfuncs.h @@ -31,14 +31,22 @@ #ifndef HASHFUNCS_H #define HASHFUNCS_H +#include "core/math/aabb.h" #include "core/math/math_defs.h" #include "core/math/math_funcs.h" +#include "core/math/rect2.h" +#include "core/math/rect2i.h" +#include "core/math/vector2.h" +#include "core/math/vector2i.h" +#include "core/math/vector3.h" +#include "core/math/vector3i.h" #include "core/object/object_id.h" #include "core/string/node_path.h" #include "core/string/string_name.h" #include "core/string/ustring.h" #include "core/templates/rid.h" #include "core/typedefs.h" + /** * Hashing functions */ @@ -155,6 +163,9 @@ static inline uint64_t make_uint64_t(T p_in) { return _u._u64; } +template <class T> +class Ref; + struct HashMapHasherDefault { static _FORCE_INLINE_ uint32_t hash(const String &p_string) { return p_string.hash(); } static _FORCE_INLINE_ uint32_t hash(const char *p_cstr) { return hash_djb2(p_cstr); } @@ -178,6 +189,55 @@ struct HashMapHasherDefault { static _FORCE_INLINE_ uint32_t hash(const StringName &p_string_name) { return p_string_name.hash(); } static _FORCE_INLINE_ uint32_t hash(const NodePath &p_path) { return p_path.hash(); } + template <class T> + static _FORCE_INLINE_ uint32_t hash(const T *p_pointer) { return hash_one_uint64((uint64_t)p_pointer); } + + template <class T> + static _FORCE_INLINE_ uint32_t hash(const Ref<T> &p_ref) { return hash_one_uint64((uint64_t)p_ref.operator->()); } + + static _FORCE_INLINE_ uint32_t hash(const Vector2i &p_vec) { + uint32_t h = hash_djb2_one_32(p_vec.x); + return hash_djb2_one_32(p_vec.y, h); + } + static _FORCE_INLINE_ uint32_t hash(const Vector3i &p_vec) { + uint32_t h = hash_djb2_one_32(p_vec.x); + h = hash_djb2_one_32(p_vec.y, h); + return hash_djb2_one_32(p_vec.z, h); + } + + static _FORCE_INLINE_ uint32_t hash(const Vector2 &p_vec) { + uint32_t h = hash_djb2_one_float(p_vec.x); + return hash_djb2_one_float(p_vec.y, h); + } + static _FORCE_INLINE_ uint32_t hash(const Vector3 &p_vec) { + uint32_t h = hash_djb2_one_float(p_vec.x); + h = hash_djb2_one_float(p_vec.y, h); + return hash_djb2_one_float(p_vec.z, h); + } + + static _FORCE_INLINE_ uint32_t hash(const Rect2i &p_rect) { + uint32_t h = hash_djb2_one_32(p_rect.position.x); + h = hash_djb2_one_32(p_rect.position.y, h); + h = hash_djb2_one_32(p_rect.size.x, h); + return hash_djb2_one_32(p_rect.size.y, h); + } + + static _FORCE_INLINE_ uint32_t hash(const Rect2 &p_rect) { + uint32_t h = hash_djb2_one_float(p_rect.position.x); + h = hash_djb2_one_float(p_rect.position.y, h); + h = hash_djb2_one_float(p_rect.size.x, h); + return hash_djb2_one_float(p_rect.size.y, h); + } + + static _FORCE_INLINE_ uint32_t hash(const AABB &p_aabb) { + uint32_t h = hash_djb2_one_float(p_aabb.position.x); + h = hash_djb2_one_float(p_aabb.position.y, h); + h = hash_djb2_one_float(p_aabb.position.z, h); + h = hash_djb2_one_float(p_aabb.size.x, h); + h = hash_djb2_one_float(p_aabb.size.y, h); + return hash_djb2_one_float(p_aabb.size.z, h); + } + //static _FORCE_INLINE_ uint32_t hash(const void* p_ptr) { return uint32_t(uint64_t(p_ptr))*(0x9e3779b1L); } }; @@ -186,14 +246,68 @@ struct HashMapComparatorDefault { static bool compare(const T &p_lhs, const T &p_rhs) { return p_lhs == p_rhs; } +}; - bool compare(const float &p_lhs, const float &p_rhs) { +template <> +struct HashMapComparatorDefault<float> { + static bool compare(const float &p_lhs, const float &p_rhs) { return (p_lhs == p_rhs) || (Math::is_nan(p_lhs) && Math::is_nan(p_rhs)); } +}; - bool compare(const double &p_lhs, const double &p_rhs) { +template <> +struct HashMapComparatorDefault<double> { + static bool compare(const double &p_lhs, const double &p_rhs) { return (p_lhs == p_rhs) || (Math::is_nan(p_lhs) && Math::is_nan(p_rhs)); } }; +template <> +struct HashMapComparatorDefault<Vector2> { + static bool compare(const Vector2 &p_lhs, const Vector2 &p_rhs) { + return ((p_lhs.x == p_rhs.x) || (Math::is_nan(p_lhs.x) && Math::is_nan(p_rhs.x))) && ((p_lhs.y == p_rhs.y) || (Math::is_nan(p_lhs.y) && Math::is_nan(p_rhs.y))); + } +}; + +template <> +struct HashMapComparatorDefault<Vector3> { + static bool compare(const Vector3 &p_lhs, const Vector3 &p_rhs) { + return ((p_lhs.x == p_rhs.x) || (Math::is_nan(p_lhs.x) && Math::is_nan(p_rhs.x))) && ((p_lhs.y == p_rhs.y) || (Math::is_nan(p_lhs.y) && Math::is_nan(p_rhs.y))) && ((p_lhs.z == p_rhs.z) || (Math::is_nan(p_lhs.z) && Math::is_nan(p_rhs.z))); + } +}; + +constexpr uint32_t HASH_TABLE_SIZE_MAX = 29; + +const uint32_t hash_table_size_primes[HASH_TABLE_SIZE_MAX] = { + 5, + 13, + 23, + 47, + 97, + 193, + 389, + 769, + 1543, + 3079, + 6151, + 12289, + 24593, + 49157, + 98317, + 196613, + 393241, + 786433, + 1572869, + 3145739, + 6291469, + 12582917, + 25165843, + 50331653, + 100663319, + 201326611, + 402653189, + 805306457, + 1610612741, +}; + #endif // HASHFUNCS_H diff --git a/core/templates/oa_hash_map.h b/core/templates/oa_hash_map.h index 4e712fccf2..25c21d1802 100644 --- a/core/templates/oa_hash_map.h +++ b/core/templates/oa_hash_map.h @@ -246,13 +246,17 @@ public: return false; } - /** - * returns true if the value was found, false otherwise. - * - * if r_data is not nullptr then the value will be written to the object - * it points to. - */ - TValue *lookup_ptr(const TKey &p_key) const { + const TValue *lookup_ptr(const TKey &p_key) const { + uint32_t pos = 0; + bool exists = _lookup_pos(p_key, pos); + + if (exists) { + return &values[pos]; + } + return nullptr; + } + + TValue *lookup_ptr(const TKey &p_key) { uint32_t pos = 0; bool exists = _lookup_pos(p_key, pos); @@ -306,7 +310,7 @@ public: bool valid; const TKey *key; - TValue *value; + TValue *value = nullptr; private: uint32_t pos; diff --git a/core/templates/ordered_hash_map.h b/core/templates/ordered_hash_map.h deleted file mode 100644 index 3d1f3a08ec..0000000000 --- a/core/templates/ordered_hash_map.h +++ /dev/null @@ -1,301 +0,0 @@ -/*************************************************************************/ -/* ordered_hash_map.h */ -/*************************************************************************/ -/* This file is part of: */ -/* GODOT ENGINE */ -/* https://godotengine.org */ -/*************************************************************************/ -/* Copyright (c) 2007-2022 Juan Linietsky, Ariel Manzur. */ -/* Copyright (c) 2014-2022 Godot Engine contributors (cf. AUTHORS.md). */ -/* */ -/* Permission is hereby granted, free of charge, to any person obtaining */ -/* a copy of this software and associated documentation files (the */ -/* "Software"), to deal in the Software without restriction, including */ -/* without limitation the rights to use, copy, modify, merge, publish, */ -/* distribute, sublicense, and/or sell copies of the Software, and to */ -/* permit persons to whom the Software is furnished to do so, subject to */ -/* the following conditions: */ -/* */ -/* The above copyright notice and this permission notice shall be */ -/* included in all copies or substantial portions of the Software. */ -/* */ -/* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, */ -/* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF */ -/* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.*/ -/* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY */ -/* CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, */ -/* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE */ -/* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ -/*************************************************************************/ - -#ifndef ORDERED_HASH_MAP_H -#define ORDERED_HASH_MAP_H - -#include "core/templates/hash_map.h" -#include "core/templates/list.h" -#include "core/templates/pair.h" - -/** - * A hash map which allows to iterate elements in insertion order. - * Insertion, lookup, deletion have O(1) complexity. - * The API aims to be consistent with Map rather than HashMap, because the - * former is more frequently used and is more coherent with the rest of the - * codebase. - * Deletion during iteration is safe and will preserve the order. - */ -template <class K, class V, class Hasher = HashMapHasherDefault, class Comparator = HashMapComparatorDefault<K>, uint8_t MIN_HASH_TABLE_POWER = 3, uint8_t RELATIONSHIP = 8> -class OrderedHashMap { - typedef List<Pair<const K *, V>> InternalList; - typedef HashMap<K, typename InternalList::Element *, Hasher, Comparator, MIN_HASH_TABLE_POWER, RELATIONSHIP> InternalMap; - - InternalList list; - InternalMap map; - -public: - class Element { - friend class OrderedHashMap<K, V, Hasher, Comparator, MIN_HASH_TABLE_POWER, RELATIONSHIP>; - - typename InternalList::Element *list_element = nullptr; - typename InternalList::Element *prev_element = nullptr; - typename InternalList::Element *next_element = nullptr; - - Element(typename InternalList::Element *p_element) { - list_element = p_element; - - if (list_element) { - next_element = list_element->next(); - prev_element = list_element->prev(); - } - } - - public: - _FORCE_INLINE_ Element() {} - - Element next() const { - return Element(next_element); - } - - Element prev() const { - return Element(prev_element); - } - - Element(const Element &other) : - list_element(other.list_element), - prev_element(other.prev_element), - next_element(other.next_element) { - } - - void operator=(const Element &other) { - list_element = other.list_element; - next_element = other.next_element; - prev_element = other.prev_element; - } - - _FORCE_INLINE_ bool operator==(const Element &p_other) const { - return this->list_element == p_other.list_element; - } - _FORCE_INLINE_ bool operator!=(const Element &p_other) const { - return this->list_element != p_other.list_element; - } - - operator bool() const { - return (list_element != nullptr); - } - - const K &key() const { - CRASH_COND(!list_element); - return *(list_element->get().first); - } - - V &value() { - CRASH_COND(!list_element); - return list_element->get().second; - } - - const V &value() const { - CRASH_COND(!list_element); - return list_element->get().second; - } - - V &get() { - CRASH_COND(!list_element); - return list_element->get().second; - } - - const V &get() const { - CRASH_COND(!list_element); - return list_element->get().second; - } - }; - - class ConstElement { - friend class OrderedHashMap<K, V, Hasher, Comparator, MIN_HASH_TABLE_POWER, RELATIONSHIP>; - - const typename InternalList::Element *list_element = nullptr; - - ConstElement(const typename InternalList::Element *p_element) : - list_element(p_element) { - } - - public: - _FORCE_INLINE_ ConstElement() {} - - ConstElement(const ConstElement &other) : - list_element(other.list_element) { - } - - void operator=(const ConstElement &other) { - list_element = other.list_element; - } - - ConstElement next() const { - return ConstElement(list_element ? list_element->next() : nullptr); - } - - ConstElement prev() const { - return ConstElement(list_element ? list_element->prev() : nullptr); - } - - _FORCE_INLINE_ bool operator==(const ConstElement &p_other) const { - return this->list_element == p_other.list_element; - } - _FORCE_INLINE_ bool operator!=(const ConstElement &p_other) const { - return this->list_element != p_other.list_element; - } - - operator bool() const { - return (list_element != nullptr); - } - - const K &key() const { - CRASH_COND(!list_element); - return *(list_element->get().first); - } - - const V &value() const { - CRASH_COND(!list_element); - return list_element->get().second; - } - - const V &get() const { - CRASH_COND(!list_element); - return list_element->get().second; - } - }; - - ConstElement find(const K &p_key) const { - typename InternalList::Element *const *list_element = map.getptr(p_key); - if (list_element) { - return ConstElement(*list_element); - } - return ConstElement(nullptr); - } - - Element find(const K &p_key) { - typename InternalList::Element **list_element = map.getptr(p_key); - if (list_element) { - return Element(*list_element); - } - return Element(nullptr); - } - - Element insert(const K &p_key, const V &p_value) { - typename InternalList::Element **list_element = map.getptr(p_key); - if (list_element) { - (*list_element)->get().second = p_value; - return Element(*list_element); - } - // Incorrectly set the first value of the pair with a value that will - // be invalid as soon as we leave this function... - typename InternalList::Element *new_element = list.push_back(Pair<const K *, V>(&p_key, p_value)); - // ...this is needed here in case the hashmap recursively reference itself... - typename InternalMap::Element *e = map.set(p_key, new_element); - // ...now we can set the right value ! - new_element->get().first = &e->key(); - - return Element(new_element); - } - - void erase(Element &p_element) { - map.erase(p_element.key()); - list.erase(p_element.list_element); - p_element.list_element = nullptr; - } - - bool erase(const K &p_key) { - typename InternalList::Element **list_element = map.getptr(p_key); - if (list_element) { - list.erase(*list_element); - map.erase(p_key); - return true; - } - return false; - } - - inline bool has(const K &p_key) const { - return map.has(p_key); - } - - const V &operator[](const K &p_key) const { - ConstElement e = find(p_key); - CRASH_COND(!e); - return e.value(); - } - - V &operator[](const K &p_key) { - Element e = find(p_key); - if (!e) { - // consistent with Map behaviour - e = insert(p_key, V()); - } - return e.value(); - } - - inline Element front() { - return Element(list.front()); - } - - inline Element back() { - return Element(list.back()); - } - - inline ConstElement front() const { - return ConstElement(list.front()); - } - - inline ConstElement back() const { - return ConstElement(list.back()); - } - - inline bool is_empty() const { return list.is_empty(); } - inline int size() const { return list.size(); } - - const void *id() const { - return list.id(); - } - - void clear() { - map.clear(); - list.clear(); - } - -private: - void _copy_from(const OrderedHashMap &p_map) { - for (ConstElement E = p_map.front(); E; E = E.next()) { - insert(E.key(), E.value()); - } - } - -public: - void operator=(const OrderedHashMap &p_map) { - _copy_from(p_map); - } - - OrderedHashMap(const OrderedHashMap &p_map) { - _copy_from(p_map); - } - - _FORCE_INLINE_ OrderedHashMap() {} -}; - -#endif // ORDERED_HASH_MAP_H diff --git a/core/templates/paged_allocator.h b/core/templates/paged_allocator.h index b9067e2edd..cf5911a847 100644 --- a/core/templates/paged_allocator.h +++ b/core/templates/paged_allocator.h @@ -50,6 +50,10 @@ class PagedAllocator { SpinLock spin_lock; public: + enum { + DEFAULT_PAGE_SIZE = 4096 + }; + template <class... Args> T *alloc(const Args &&...p_args) { if (thread_safe) { @@ -121,7 +125,9 @@ public: page_shift = get_shift_from_power_of_2(page_size); } - PagedAllocator(uint32_t p_page_size = 4096) { // power of 2 recommended because of alignment with OS page sizes. Even if element is bigger, its still a multiple and get rounded amount of pages + // Power of 2 recommended because of alignment with OS page sizes. + // Even if element is bigger, it's still a multiple and gets rounded to amount of pages. + PagedAllocator(uint32_t p_page_size = DEFAULT_PAGE_SIZE) { configure(p_page_size); } diff --git a/core/templates/pair.h b/core/templates/pair.h index eb86e21b03..6d33213fe3 100644 --- a/core/templates/pair.h +++ b/core/templates/pair.h @@ -31,8 +31,8 @@ #ifndef PAIR_H #define PAIR_H +#include "core/templates/hashfuncs.h" #include "core/typedefs.h" - template <class F, class S> struct Pair { F first; @@ -69,6 +69,15 @@ struct PairSort { } }; +template <class F, class S> +struct PairHash { + static uint32_t hash(const Pair<F, S> &P) { + uint64_t h1 = HashMapHasherDefault::hash(P.first); + uint64_t h2 = HashMapHasherDefault::hash(P.second); + return hash_one_uint64((h1 << 32) | h2); + } +}; + template <class K, class V> struct KeyValue { const K key; diff --git a/core/templates/map.h b/core/templates/rb_map.h index f228640a1e..c732ccd485 100644 --- a/core/templates/map.h +++ b/core/templates/rb_map.h @@ -1,5 +1,5 @@ /*************************************************************************/ -/* map.h */ +/* rb_map.h */ /*************************************************************************/ /* This file is part of: */ /* GODOT ENGINE */ @@ -28,8 +28,8 @@ /* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ /*************************************************************************/ -#ifndef MAP_H -#define MAP_H +#ifndef RB_MAP_H +#define RB_MAP_H #include "core/error/error_macros.h" #include "core/os/memory.h" @@ -39,7 +39,7 @@ // https://web.archive.org/web/20120507164830/https://web.mit.edu/~emin/www/source_code/red_black_tree/index.html template <class K, class V, class C = Comparator<K>, class A = DefaultAllocator> -class Map { +class RBMap { enum Color { RED, BLACK @@ -49,7 +49,7 @@ class Map { public: class Element { private: - friend class Map<K, V, C, A>; + friend class RBMap<K, V, C, A>; int color = RED; Element *right = nullptr; Element *left = nullptr; @@ -111,7 +111,9 @@ public: _FORCE_INLINE_ bool operator==(const Iterator &b) const { return E == b.E; } _FORCE_INLINE_ bool operator!=(const Iterator &b) const { return E != b.E; } - + explicit operator bool() const { + return E != nullptr; + } Iterator(Element *p_E) { E = p_E; } Iterator() {} Iterator(const Iterator &p_it) { E = p_it.E; } @@ -136,7 +138,9 @@ public: _FORCE_INLINE_ bool operator==(const ConstIterator &b) const { return E == b.E; } _FORCE_INLINE_ bool operator!=(const ConstIterator &b) const { return E != b.E; } - + explicit operator bool() const { + return E != nullptr; + } ConstIterator(const Element *p_E) { E = p_E; } ConstIterator() {} ConstIterator(const ConstIterator &p_it) { E = p_it.E; } @@ -178,7 +182,7 @@ public: private: struct _Data { Element *_root = nullptr; - Element *_nil; + Element *_nil = nullptr; int size_cache = 0; _FORCE_INLINE_ _Data() { @@ -344,7 +348,7 @@ private: void _insert_rb_fix(Element *p_new_node) { Element *node = p_new_node; Element *nparent = node->parent; - Element *ngrand_parent; + Element *ngrand_parent = nullptr; while (nparent->color == RED) { ngrand_parent = nparent->parent; @@ -500,7 +504,7 @@ private: Element *rp = ((p_node->left == _data._nil) || (p_node->right == _data._nil)) ? p_node : p_node->_next; Element *node = (rp->left == _data._nil) ? rp->right : rp->left; - Element *sibling; + Element *sibling = nullptr; if (rp == rp->parent->left) { rp->parent->left = node; sibling = rp->parent->right; @@ -572,7 +576,7 @@ private: memdelete_allocator<Element, A>(p_element); } - void _copy_from(const Map &p_map) { + void _copy_from(const RBMap &p_map) { clear(); // not the fastest way, but safeset to write. for (Element *I = p_map.front(); I; I = I->next()) { @@ -710,8 +714,12 @@ public: return e; } - inline bool is_empty() const { return _data.size_cache == 0; } - inline int size() const { return _data.size_cache; } + inline bool is_empty() const { + return _data.size_cache == 0; + } + inline int size() const { + return _data.size_cache; + } int calculate_depth() const { // used for debug mostly @@ -735,17 +743,17 @@ public: _data._free_root(); } - void operator=(const Map &p_map) { + void operator=(const RBMap &p_map) { _copy_from(p_map); } - Map(const Map &p_map) { + RBMap(const RBMap &p_map) { _copy_from(p_map); } - _FORCE_INLINE_ Map() {} + _FORCE_INLINE_ RBMap() {} - ~Map() { + ~RBMap() { clear(); } }; diff --git a/core/templates/set.h b/core/templates/rb_set.h index cdc6e8447d..2de816769c 100644 --- a/core/templates/set.h +++ b/core/templates/rb_set.h @@ -1,5 +1,5 @@ /*************************************************************************/ -/* set.h */ +/* rb_set.h */ /*************************************************************************/ /* This file is part of: */ /* GODOT ENGINE */ @@ -28,8 +28,8 @@ /* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ /*************************************************************************/ -#ifndef SET_H -#define SET_H +#ifndef RB_SET_H +#define RB_SET_H #include "core/os/memory.h" #include "core/typedefs.h" @@ -38,7 +38,7 @@ // https://web.archive.org/web/20120507164830/https://web.mit.edu/~emin/www/source_code/red_black_tree/index.html template <class T, class C = Comparator<T>, class A = DefaultAllocator> -class Set { +class RBSet { enum Color { RED, BLACK @@ -48,7 +48,7 @@ class Set { public: class Element { private: - friend class Set<T, C, A>; + friend class RBSet<T, C, A>; int color = RED; Element *right = nullptr; Element *left = nullptr; @@ -328,7 +328,7 @@ private: void _insert_rb_fix(Element *p_new_node) { Element *node = p_new_node; Element *nparent = node->parent; - Element *ngrand_parent; + Element *ngrand_parent = nullptr; while (nparent->color == RED) { ngrand_parent = nparent->parent; @@ -482,7 +482,7 @@ private: Element *rp = ((p_node->left == _data._nil) || (p_node->right == _data._nil)) ? p_node : p_node->_next; Element *node = (rp->left == _data._nil) ? rp->right : rp->left; - Element *sibling; + Element *sibling = nullptr; if (rp == rp->parent->left) { rp->parent->left = node; sibling = rp->parent->right; @@ -554,7 +554,7 @@ private: memdelete_allocator<Element, A>(p_element); } - void _copy_from(const Set &p_set) { + void _copy_from(const RBSet &p_set) { clear(); // not the fastest way, but safeset to write. for (Element *I = p_set.front(); I; I = I->next()) { @@ -661,8 +661,12 @@ public: return e; } - inline bool is_empty() const { return _data.size_cache == 0; } - inline int size() const { return _data.size_cache; } + inline bool is_empty() const { + return _data.size_cache == 0; + } + inline int size() const { + return _data.size_cache; + } int calculate_depth() const { // used for debug mostly @@ -686,17 +690,17 @@ public: _data._free_root(); } - void operator=(const Set &p_set) { + void operator=(const RBSet &p_set) { _copy_from(p_set); } - Set(const Set &p_set) { + RBSet(const RBSet &p_set) { _copy_from(p_set); } - _FORCE_INLINE_ Set() {} + _FORCE_INLINE_ RBSet() {} - ~Set() { + ~RBSet() { clear(); } }; diff --git a/core/templates/rid_owner.h b/core/templates/rid_owner.h index 95632cdec2..d26977380e 100644 --- a/core/templates/rid_owner.h +++ b/core/templates/rid_owner.h @@ -36,9 +36,9 @@ #include "core/string/print_string.h" #include "core/templates/list.h" #include "core/templates/oa_hash_map.h" +#include "core/templates/rb_set.h" #include "core/templates/rid.h" #include "core/templates/safe_refcount.h" -#include "core/templates/set.h" #include <stdio.h> #include <typeinfo> diff --git a/core/templates/safe_list.h b/core/templates/safe_list.h index 53fc3fe5f9..e850f3bd5e 100644 --- a/core/templates/safe_list.h +++ b/core/templates/safe_list.h @@ -75,8 +75,8 @@ public: class Iterator { friend class SafeList; - SafeListNode *cursor; - SafeList *list; + SafeListNode *cursor = nullptr; + SafeList *list = nullptr; Iterator(SafeListNode *p_cursor, SafeList *p_list) : cursor(p_cursor), list(p_list) { @@ -203,7 +203,7 @@ public: } // Calling this will cause zero to many deallocations. - void maybe_cleanup() { + bool maybe_cleanup() { SafeListNode *cursor = nullptr; SafeListNode *new_graveyard_head = nullptr; do { @@ -212,7 +212,7 @@ public: if (active_iterator_count.load() != 0) { // It's not safe to clean up with an active iterator, because that iterator // could be pointing to an element that we want to delete. - return; + return false; } // Any iterator created after this point will never point to a deleted node. // Swap it out with the current graveyard head. @@ -225,6 +225,17 @@ public: tmp->deletion_fn(tmp->val); memdelete_allocator<SafeListNode, A>(tmp); } + return true; + } + + ~SafeList() { +#ifdef DEBUG_ENABLED + if (!maybe_cleanup()) { + ERR_PRINT("There are still iterators around when destructing a SafeList. Memory will be leaked. This is a bug."); + } +#else + maybe_cleanup(); +#endif } }; @@ -253,8 +264,8 @@ public: class Iterator { friend class SafeList; - SafeListNode *cursor; - SafeList *list; + SafeListNode *cursor = nullptr; + SafeList *list = nullptr; public: Iterator(SafeListNode *p_cursor, SafeList *p_list) : @@ -353,11 +364,11 @@ public: } // Calling this will cause zero to many deallocations. - void maybe_cleanup() { + bool maybe_cleanup() { SafeListNode *cursor = graveyard_head; if (active_iterator_count != 0) { // It's not safe to clean up with an active iterator, because that iterator could be pointing to an element that we want to delete. - return; + return false; } graveyard_head = nullptr; // Our graveyard list is now unreachable by any active iterators, detached from the main graveyard head and ready for deletion. @@ -367,6 +378,17 @@ public: tmp->deletion_fn(tmp->val); memdelete_allocator<SafeListNode, A>(tmp); } + return true; + } + + ~SafeList() { +#ifdef DEBUG_ENABLED + if (!maybe_cleanup()) { + ERR_PRINT("There are still iterators around when destructing a SafeList. Memory will be leaked. This is a bug."); + } +#else + maybe_cleanup(); +#endif } }; diff --git a/core/templates/self_list.h b/core/templates/self_list.h index 7f2236fa3a..719b5f2e63 100644 --- a/core/templates/self_list.h +++ b/core/templates/self_list.h @@ -108,7 +108,7 @@ public: private: List *_root = nullptr; - T *_self; + T *_self = nullptr; SelfList<T> *_next = nullptr; SelfList<T> *_prev = nullptr; diff --git a/core/templates/thread_work_pool.h b/core/templates/thread_work_pool.h index 957af44f48..b0cebf04f1 100644 --- a/core/templates/thread_work_pool.h +++ b/core/templates/thread_work_pool.h @@ -52,7 +52,7 @@ class ThreadWorkPool { C *instance; M method; U userdata; - virtual void work() { + virtual void work() override { while (true) { uint32_t work_index = index->fetch_add(1, std::memory_order_relaxed); if (work_index >= max_elements) { @@ -68,7 +68,7 @@ class ThreadWorkPool { Semaphore start; Semaphore completed; std::atomic<bool> exit; - BaseWork *work; + BaseWork *work = nullptr; }; ThreadData *threads = nullptr; diff --git a/core/templates/vector.h b/core/templates/vector.h index d87e76139b..2ac7c7630a 100644 --- a/core/templates/vector.h +++ b/core/templates/vector.h @@ -92,6 +92,8 @@ public: _FORCE_INLINE_ const T &operator[](int p_index) const { return _cowdata.get(p_index); } Error insert(int p_pos, T p_val) { return _cowdata.insert(p_pos, p_val); } int find(const T &p_val, int p_from = 0) const { return _cowdata.find(p_val, p_from); } + int rfind(const T &p_val, int p_from = -1) const { return _cowdata.rfind(p_val, p_from); } + int count(const T &p_val) const { return _cowdata.count(p_val); } void append_array(Vector<T> p_other); |