boost/intrusive/hashtable.hpp
///////////////////////////////////////////////////////////////////////////// // // (C) Copyright Ion Gaztanaga 2006-2015 // // Distributed under the Boost Software License, Version 1.0. // (See accompanying file LICENSE_1_0.txt or copy at // http://www.boost.org/LICENSE_1_0.txt) // // See http://www.boost.org/libs/intrusive for documentation. // ///////////////////////////////////////////////////////////////////////////// #ifndef BOOST_INTRUSIVE_HASHTABLE_HPP #define BOOST_INTRUSIVE_HASHTABLE_HPP #include <boost/intrusive/detail/config_begin.hpp> #include <boost/intrusive/intrusive_fwd.hpp> //General intrusive utilities #include <boost/intrusive/detail/hashtable_node.hpp> #include <boost/intrusive/detail/transform_iterator.hpp> #include <boost/intrusive/link_mode.hpp> #include <boost/intrusive/detail/ebo_functor_holder.hpp> #include <boost/intrusive/detail/is_stateful_value_traits.hpp> #include <boost/intrusive/detail/node_to_value.hpp> #include <boost/intrusive/detail/exception_disposer.hpp> #include <boost/intrusive/detail/node_cloner_disposer.hpp> #include <boost/intrusive/detail/simple_disposers.hpp> #include <boost/intrusive/detail/size_holder.hpp> #include <boost/intrusive/detail/iterator.hpp> //Implementation utilities #include <boost/intrusive/unordered_set_hook.hpp> #include <boost/intrusive/slist.hpp> #include <boost/intrusive/pointer_traits.hpp> #include <boost/intrusive/detail/mpl.hpp> //boost #include <boost/container_hash/hash.hpp> #include <boost/intrusive/detail/assert.hpp> #include <boost/static_assert.hpp> #include <boost/move/utility_core.hpp> #include <boost/move/adl_move_swap.hpp> //std C++ #include <boost/intrusive/detail/minimal_less_equal_header.hpp> //std::equal_to #include <boost/intrusive/detail/minimal_pair_header.hpp> //std::pair #include <algorithm> //std::lower_bound, std::upper_bound #include <cstddef> //std::size_t #if defined(BOOST_HAS_PRAGMA_ONCE) # pragma once #endif namespace boost { namespace intrusive { /// @cond template<class InputIt, class T> InputIt priv_algo_find(InputIt first, InputIt last, const T& value) { for (; first != last; ++first) { if (*first == value) { return first; } } return last; } template<class InputIt, class T> typename boost::intrusive::iterator_traits<InputIt>::difference_type priv_algo_count(InputIt first, InputIt last, const T& value) { typename boost::intrusive::iterator_traits<InputIt>::difference_type ret = 0; for (; first != last; ++first) { if (*first == value) { ret++; } } return ret; } template <class ForwardIterator1, class ForwardIterator2> bool priv_algo_is_permutation(ForwardIterator1 first1, ForwardIterator1 last1, ForwardIterator2 first2) { typedef typename boost::intrusive::iterator_traits<ForwardIterator2>::difference_type distance_type; //Efficiently compare identical prefixes: O(N) if sequences //have the same elements in the same order. for ( ; first1 != last1; ++first1, ++first2){ if (! (*first1 == *first2)) break; } if (first1 == last1){ return true; } //Establish last2 assuming equal ranges by iterating over the //rest of the list. ForwardIterator2 last2 = first2; boost::intrusive::iterator_advance(last2, boost::intrusive::iterator_distance(first1, last1)); for(ForwardIterator1 scan = first1; scan != last1; ++scan){ if (scan != (priv_algo_find)(first1, scan, *scan)){ continue; //We've seen this one before. } distance_type matches = (priv_algo_count)(first2, last2, *scan); if (0 == matches || (priv_algo_count)(scan, last1, *scan != matches)){ return false; } } return true; } template<int Dummy = 0> struct prime_list_holder { private: template <class SizeType> // sizeof(SizeType) < sizeof(std::size_t) static BOOST_INTRUSIVE_FORCEINLINE SizeType truncate_size_type(std::size_t n, detail::true_) { return n < std::size_t(SizeType(-1)) ? static_cast<SizeType>(n) : SizeType(-1); } template <class SizeType> // sizeof(SizeType) == sizeof(std::size_t) static BOOST_INTRUSIVE_FORCEINLINE SizeType truncate_size_type(std::size_t n, detail::false_) { return static_cast<SizeType>(n); } template <class SizeType> //sizeof(SizeType) > sizeof(std::size_t) static BOOST_INTRUSIVE_FORCEINLINE SizeType suggested_upper_bucket_count_dispatch(SizeType n, detail::true_) { std::size_t const c = n > std::size_t(-1) ? std::size_t(-1) : suggested_upper_bucket_count_impl(static_cast<std::size_t>(n)); return static_cast<SizeType>(c); } template <class SizeType> //sizeof(SizeType) > sizeof(std::size_t) static BOOST_INTRUSIVE_FORCEINLINE SizeType suggested_lower_bucket_count_dispatch(SizeType n, detail::true_) { std::size_t const c = n > std::size_t(-1) ? std::size_t(-1) : suggested_lower_bucket_count_impl(static_cast<std::size_t>(n)); return static_cast<SizeType>(c); } template <class SizeType> static BOOST_INTRUSIVE_FORCEINLINE SizeType suggested_upper_bucket_count_dispatch(SizeType n, detail::false_) { std::size_t const c = suggested_upper_bucket_count_impl(static_cast<std::size_t>(n)); return truncate_size_type<SizeType>(c, detail::bool_<(sizeof(SizeType) < sizeof(std::size_t))>()); } template <class SizeType> static BOOST_INTRUSIVE_FORCEINLINE SizeType suggested_lower_bucket_count_dispatch(SizeType n, detail::false_) { std::size_t const c = suggested_lower_bucket_count_impl(static_cast<std::size_t>(n)); return truncate_size_type<SizeType>(c, detail::bool_<(sizeof(SizeType) < sizeof(std::size_t))>()); } static const std::size_t prime_list[]; static const std::size_t prime_list_size; static std::size_t suggested_lower_bucket_count_impl(std::size_t n) { const std::size_t *primes = &prime_list_holder<0>::prime_list[0]; const std::size_t *primes_end = primes + prime_list_holder<0>::prime_list_size; std::size_t const* bound = std::lower_bound(primes, primes_end, n); //Tables have upper SIZE_MAX, so we must always found an entry BOOST_INTRUSIVE_INVARIANT_ASSERT(bound != primes_end); bound -= std::size_t(bound != primes); return *bound; } static std::size_t suggested_upper_bucket_count_impl(std::size_t n) { const std::size_t *primes = &prime_list_holder<0>::prime_list[0]; const std::size_t *primes_end = primes + prime_list_holder<0>::prime_list_size; std::size_t const* bound = std::upper_bound(primes, primes_end, n); bound -= std::size_t(bound == primes_end); return *bound; } public: template <class SizeType> static BOOST_INTRUSIVE_FORCEINLINE SizeType suggested_upper_bucket_count(SizeType n) { return (suggested_upper_bucket_count_dispatch)(n, detail::bool_<(sizeof(SizeType) > sizeof(std::size_t))>()); } template <class SizeType> static BOOST_INTRUSIVE_FORCEINLINE SizeType suggested_lower_bucket_count(SizeType n) { return (suggested_lower_bucket_count_dispatch)(n, detail::bool_<(sizeof(SizeType) > sizeof(std::size_t))>()); } }; #if !defined(BOOST_INTRUSIVE_DOXYGEN_INVOKED) //We only support LLP64(Win64) or LP64(most Unix) data models #ifdef _WIN64 //In 64 bit windows sizeof(size_t) == sizeof(unsigned long long) #define BOOST_INTRUSIVE_PRIME_C(NUMBER) NUMBER##ULL #define BOOST_INTRUSIVE_64_BIT_SIZE_T 1 #else //In 32 bit windows and 32/64 bit unixes sizeof(size_t) == sizeof(unsigned long) #define BOOST_INTRUSIVE_PRIME_C(NUMBER) NUMBER##UL #define BOOST_INTRUSIVE_64_BIT_SIZE_T (((((ULONG_MAX>>16)>>16)>>16)>>15) != 0) #endif template<int Dummy> const std::size_t prime_list_holder<Dummy>::prime_list[] = { BOOST_INTRUSIVE_PRIME_C(3), BOOST_INTRUSIVE_PRIME_C(7), BOOST_INTRUSIVE_PRIME_C(11), BOOST_INTRUSIVE_PRIME_C(17), BOOST_INTRUSIVE_PRIME_C(29), BOOST_INTRUSIVE_PRIME_C(53), BOOST_INTRUSIVE_PRIME_C(97), BOOST_INTRUSIVE_PRIME_C(193), BOOST_INTRUSIVE_PRIME_C(389), BOOST_INTRUSIVE_PRIME_C(769), BOOST_INTRUSIVE_PRIME_C(1543), BOOST_INTRUSIVE_PRIME_C(3079), BOOST_INTRUSIVE_PRIME_C(6151), BOOST_INTRUSIVE_PRIME_C(12289), BOOST_INTRUSIVE_PRIME_C(24593), BOOST_INTRUSIVE_PRIME_C(49157), BOOST_INTRUSIVE_PRIME_C(98317), BOOST_INTRUSIVE_PRIME_C(196613), BOOST_INTRUSIVE_PRIME_C(393241), BOOST_INTRUSIVE_PRIME_C(786433), BOOST_INTRUSIVE_PRIME_C(1572869), BOOST_INTRUSIVE_PRIME_C(3145739), BOOST_INTRUSIVE_PRIME_C(6291469), BOOST_INTRUSIVE_PRIME_C(12582917), BOOST_INTRUSIVE_PRIME_C(25165843), BOOST_INTRUSIVE_PRIME_C(50331653), BOOST_INTRUSIVE_PRIME_C(100663319), BOOST_INTRUSIVE_PRIME_C(201326611), BOOST_INTRUSIVE_PRIME_C(402653189), BOOST_INTRUSIVE_PRIME_C(805306457), BOOST_INTRUSIVE_PRIME_C(1610612741), BOOST_INTRUSIVE_PRIME_C(3221225473), #if BOOST_INTRUSIVE_64_BIT_SIZE_T //Taken from Boost.MultiIndex code, thanks to Joaquin M Lopez Munoz. BOOST_INTRUSIVE_PRIME_C(6442450939), BOOST_INTRUSIVE_PRIME_C(12884901893), BOOST_INTRUSIVE_PRIME_C(25769803751), BOOST_INTRUSIVE_PRIME_C(51539607551), BOOST_INTRUSIVE_PRIME_C(103079215111), BOOST_INTRUSIVE_PRIME_C(206158430209), BOOST_INTRUSIVE_PRIME_C(412316860441), BOOST_INTRUSIVE_PRIME_C(824633720831), BOOST_INTRUSIVE_PRIME_C(1649267441651), BOOST_INTRUSIVE_PRIME_C(3298534883309), BOOST_INTRUSIVE_PRIME_C(6597069766657), BOOST_INTRUSIVE_PRIME_C(13194139533299), BOOST_INTRUSIVE_PRIME_C(26388279066623), BOOST_INTRUSIVE_PRIME_C(52776558133303), BOOST_INTRUSIVE_PRIME_C(105553116266489), BOOST_INTRUSIVE_PRIME_C(211106232532969), BOOST_INTRUSIVE_PRIME_C(422212465066001), BOOST_INTRUSIVE_PRIME_C(844424930131963), BOOST_INTRUSIVE_PRIME_C(1688849860263953), BOOST_INTRUSIVE_PRIME_C(3377699720527861), BOOST_INTRUSIVE_PRIME_C(6755399441055731), BOOST_INTRUSIVE_PRIME_C(13510798882111483), BOOST_INTRUSIVE_PRIME_C(27021597764222939), BOOST_INTRUSIVE_PRIME_C(54043195528445957), BOOST_INTRUSIVE_PRIME_C(108086391056891903), BOOST_INTRUSIVE_PRIME_C(216172782113783843), BOOST_INTRUSIVE_PRIME_C(432345564227567621), BOOST_INTRUSIVE_PRIME_C(864691128455135207), BOOST_INTRUSIVE_PRIME_C(1729382256910270481), BOOST_INTRUSIVE_PRIME_C(3458764513820540933), BOOST_INTRUSIVE_PRIME_C(6917529027641081903), BOOST_INTRUSIVE_PRIME_C(13835058055282163729), BOOST_INTRUSIVE_PRIME_C(18446744073709551557), BOOST_INTRUSIVE_PRIME_C(18446744073709551615) //Upper limit, just in case #else BOOST_INTRUSIVE_PRIME_C(4294967291), BOOST_INTRUSIVE_PRIME_C(4294967295) //Upper limit, just in case #endif }; #undef BOOST_INTRUSIVE_PRIME_C #undef BOOST_INTRUSIVE_64_BIT_SIZE_T #endif //#if !defined(BOOST_INTRUSIVE_DOXYGEN_INVOKED) template<int Dummy> const std::size_t prime_list_holder<Dummy>::prime_list_size = sizeof(prime_list)/sizeof(std::size_t); struct hash_bool_flags { static const std::size_t unique_keys_pos = 1u; static const std::size_t constant_time_size_pos = 2u; static const std::size_t power_2_buckets_pos = 4u; static const std::size_t cache_begin_pos = 8u; static const std::size_t compare_hash_pos = 16u; static const std::size_t incremental_pos = 32u; }; namespace detail { template<class SupposedValueTraits> struct get_slist_impl_from_supposed_value_traits { typedef SupposedValueTraits value_traits; typedef typename detail::get_node_traits <value_traits>::type node_traits; typedef typename get_slist_impl <typename reduced_slist_node_traits <node_traits>::type >::type type; }; template<class SupposedValueTraits> struct unordered_bucket_impl { typedef typename get_slist_impl_from_supposed_value_traits <SupposedValueTraits>::type slist_impl; typedef bucket_impl<slist_impl> implementation_defined; typedef implementation_defined type; }; template<class SupposedValueTraits> struct unordered_bucket_ptr_impl { typedef typename detail::get_node_traits <SupposedValueTraits>::type::node_ptr node_ptr; typedef typename unordered_bucket_impl <SupposedValueTraits>::type bucket_type; typedef typename pointer_traits <node_ptr>::template rebind_pointer < bucket_type >::type implementation_defined; typedef implementation_defined type; }; template <class T> struct store_hash_is_true { template<bool Add> struct two_or_three {yes_type _[2 + Add];}; template <class U> static yes_type test(...); template <class U> static two_or_three<U::store_hash> test (int); static const bool value = sizeof(test<T>(0)) > sizeof(yes_type)*2; }; template <class T> struct optimize_multikey_is_true { template<bool Add> struct two_or_three {yes_type _[2 + Add];}; template <class U> static yes_type test(...); template <class U> static two_or_three<U::optimize_multikey> test (int); static const bool value = sizeof(test<T>(0)) > sizeof(yes_type)*2; }; struct insert_commit_data_impl { std::size_t hash; }; template<class Node, class SlistNodePtr> BOOST_INTRUSIVE_FORCEINLINE typename pointer_traits<SlistNodePtr>::template rebind_pointer<Node>::type dcast_bucket_ptr(const SlistNodePtr &p) { typedef typename pointer_traits<SlistNodePtr>::template rebind_pointer<Node>::type node_ptr; return pointer_traits<node_ptr>::pointer_to(static_cast<Node&>(*p)); } template<class NodeTraits> struct group_functions { // A group is reverse-linked // // A is "first in group" // C is "last in group" // __________________ // | _____ _____ | // | | | | | | <- Group links // ^ V ^ V ^ V // _ _ _ _ // A|_| B|_| C|_| D|_| // // ^ | ^ | ^ | ^ V <- Bucket links // _ _____| |_____| |______| |____| | // |B| | // ^________________________________| // typedef NodeTraits node_traits; typedef unordered_group_adapter<node_traits> group_traits; typedef typename node_traits::node_ptr node_ptr; typedef typename node_traits::node node; typedef typename reduced_slist_node_traits <node_traits>::type reduced_node_traits; typedef typename reduced_node_traits::node_ptr slist_node_ptr; typedef typename reduced_node_traits::node slist_node; typedef circular_slist_algorithms<group_traits> group_algorithms; typedef circular_slist_algorithms<node_traits> node_algorithms; static slist_node_ptr get_bucket_before_begin (slist_node_ptr bucket_beg, slist_node_ptr bucket_end, node_ptr p) { //First find the last node of p's group. //This requires checking the first node of the next group or //the bucket node. node_ptr prev_node = p; node_ptr nxt(node_traits::get_next(p)); while(!(bucket_beg <= nxt && nxt <= bucket_end) && (group_traits::get_next(nxt) == prev_node)){ prev_node = nxt; nxt = node_traits::get_next(nxt); } //If we've reached the bucket node just return it. if(bucket_beg <= nxt && nxt <= bucket_end){ return nxt; } //Otherwise, iterate using group links until the bucket node node_ptr first_node_of_group = nxt; node_ptr last_node_group = group_traits::get_next(first_node_of_group); slist_node_ptr possible_end = node_traits::get_next(last_node_group); while(!(bucket_beg <= possible_end && possible_end <= bucket_end)){ first_node_of_group = detail::dcast_bucket_ptr<node>(possible_end); last_node_group = group_traits::get_next(first_node_of_group); possible_end = node_traits::get_next(last_node_group); } return possible_end; } static node_ptr get_prev_to_first_in_group(slist_node_ptr bucket_node, node_ptr first_in_group) { node_ptr nb = detail::dcast_bucket_ptr<node>(bucket_node); node_ptr n; while((n = node_traits::get_next(nb)) != first_in_group){ nb = group_traits::get_next(n); //go to last in group } return nb; } static void erase_from_group(slist_node_ptr end_ptr, node_ptr to_erase_ptr, detail::true_) { node_ptr const nxt_ptr(node_traits::get_next(to_erase_ptr)); //Check if the next node is in the group (not end node) and reverse linked to //'to_erase_ptr'. Erase if that's the case. if(nxt_ptr != end_ptr && to_erase_ptr == group_traits::get_next(nxt_ptr)){ group_algorithms::unlink_after(nxt_ptr); } } BOOST_INTRUSIVE_FORCEINLINE static void erase_from_group(const slist_node_ptr&, const node_ptr&, detail::false_) {} BOOST_INTRUSIVE_FORCEINLINE static node_ptr get_last_in_group(node_ptr first_in_group, detail::true_) { return group_traits::get_next(first_in_group); } BOOST_INTRUSIVE_FORCEINLINE static node_ptr get_last_in_group(node_ptr n, detail::false_) { return n; } static node_ptr get_first_in_group(node_ptr n, detail::true_) { node_ptr ng; while(n == node_traits::get_next((ng = group_traits::get_next(n)))){ n = ng; } return n; } BOOST_INTRUSIVE_FORCEINLINE static node_ptr next_group_if_first_in_group(node_ptr ptr) { return node_traits::get_next(group_traits::get_next(ptr)); } BOOST_INTRUSIVE_FORCEINLINE static node_ptr get_first_in_group(node_ptr n, detail::false_) { return n; } BOOST_INTRUSIVE_FORCEINLINE static void insert_in_group(node_ptr first_in_group, node_ptr n, true_) { group_algorithms::link_after(first_in_group, n); } static void insert_in_group(const node_ptr&, const node_ptr&, false_) {} BOOST_INTRUSIVE_FORCEINLINE static node_ptr split_group(node_ptr const new_first_in_group) { node_ptr const first((get_first_in_group)(new_first_in_group, detail::true_())); if(first != new_first_in_group){ node_ptr const last = group_traits::get_next(first); group_traits::set_next(first, group_traits::get_next(new_first_in_group)); group_traits::set_next(new_first_in_group, last); } return first; } }; template<class BucketType, class SplitTraits> class incremental_rehash_rollback { private: typedef BucketType bucket_type; typedef SplitTraits split_traits; incremental_rehash_rollback(); incremental_rehash_rollback & operator=(const incremental_rehash_rollback &); incremental_rehash_rollback (const incremental_rehash_rollback &); public: incremental_rehash_rollback (bucket_type &source_bucket, bucket_type &destiny_bucket, split_traits &split_traits) : source_bucket_(source_bucket), destiny_bucket_(destiny_bucket) , split_traits_(split_traits), released_(false) {} BOOST_INTRUSIVE_FORCEINLINE void release() { released_ = true; } ~incremental_rehash_rollback() { if(!released_){ //If an exception is thrown, just put all moved nodes back in the old bucket //and move back the split mark. destiny_bucket_.splice_after(destiny_bucket_.before_begin(), source_bucket_); split_traits_.decrement(); } } private: bucket_type &source_bucket_; bucket_type &destiny_bucket_; split_traits &split_traits_; bool released_; }; template<class NodeTraits> struct node_functions { BOOST_INTRUSIVE_FORCEINLINE static void store_hash(typename NodeTraits::node_ptr p, std::size_t h, true_) { return NodeTraits::set_hash(p, h); } BOOST_INTRUSIVE_FORCEINLINE static void store_hash(typename NodeTraits::node_ptr, std::size_t, false_) {} }; BOOST_INTRUSIVE_FORCEINLINE std::size_t hash_to_bucket(std::size_t hash_value, std::size_t bucket_cnt, detail::false_) { return hash_value % bucket_cnt; } BOOST_INTRUSIVE_FORCEINLINE std::size_t hash_to_bucket(std::size_t hash_value, std::size_t bucket_cnt, detail::true_) { return hash_value & (bucket_cnt - 1); } template<bool Power2Buckets, bool Incremental> BOOST_INTRUSIVE_FORCEINLINE std::size_t hash_to_bucket_split(std::size_t hash_value, std::size_t bucket_cnt, std::size_t split) { std::size_t bucket_number = detail::hash_to_bucket(hash_value, bucket_cnt, detail::bool_<Power2Buckets>()); if(Incremental) bucket_number -= static_cast<std::size_t>(bucket_number >= split)*(bucket_cnt/2); return bucket_number; } } //namespace detail { //!This metafunction will obtain the type of a bucket //!from the value_traits or hook option to be used with //!a hash container. template<class ValueTraitsOrHookOption> struct unordered_bucket : public detail::unordered_bucket_impl <typename ValueTraitsOrHookOption:: template pack<empty>::proto_value_traits > {}; //!This metafunction will obtain the type of a bucket pointer //!from the value_traits or hook option to be used with //!a hash container. template<class ValueTraitsOrHookOption> struct unordered_bucket_ptr : public detail::unordered_bucket_ptr_impl <typename ValueTraitsOrHookOption:: template pack<empty>::proto_value_traits > {}; //!This metafunction will obtain the type of the default bucket traits //!(when the user does not specify the bucket_traits<> option) from the //!value_traits or hook option to be used with //!a hash container. template<class ValueTraitsOrHookOption> struct unordered_default_bucket_traits { typedef typename ValueTraitsOrHookOption:: template pack<empty>::proto_value_traits supposed_value_traits; typedef typename detail:: get_slist_impl_from_supposed_value_traits <supposed_value_traits>::type slist_impl; typedef bucket_traits_impl <slist_impl> implementation_defined; typedef implementation_defined type; }; struct default_bucket_traits; //hashtable default hook traits struct default_hashtable_hook_applier { template <class T> struct apply{ typedef typename T::default_hashtable_hook type; }; }; template<> struct is_default_hook_tag<default_hashtable_hook_applier> { static const bool value = true; }; struct hashtable_defaults { typedef default_hashtable_hook_applier proto_value_traits; typedef std::size_t size_type; typedef void key_of_value; typedef void equal; typedef void hash; typedef default_bucket_traits bucket_traits; static const bool constant_time_size = true; static const bool power_2_buckets = false; static const bool cache_begin = false; static const bool compare_hash = false; static const bool incremental = false; }; template<class ValueTraits, bool IsConst> struct downcast_node_to_value_t : public detail::node_to_value<ValueTraits, IsConst> { typedef detail::node_to_value<ValueTraits, IsConst> base_t; typedef typename base_t::result_type result_type; typedef ValueTraits value_traits; typedef typename get_slist_impl <typename reduced_slist_node_traits <typename value_traits::node_traits>::type >::type slist_impl; typedef typename detail::add_const_if_c <typename slist_impl::node, IsConst>::type & first_argument_type; typedef typename detail::add_const_if_c < typename ValueTraits::node_traits::node , IsConst>::type & intermediate_argument_type; typedef typename pointer_traits <typename ValueTraits::pointer>:: template rebind_pointer <const ValueTraits>::type const_value_traits_ptr; BOOST_INTRUSIVE_FORCEINLINE downcast_node_to_value_t(const const_value_traits_ptr &ptr) : base_t(ptr) {} BOOST_INTRUSIVE_FORCEINLINE result_type operator()(first_argument_type arg) const { return this->base_t::operator()(static_cast<intermediate_argument_type>(arg)); } }; template<class F, class SlistNodePtr, class NodePtr> struct node_cast_adaptor //Use public inheritance to avoid MSVC bugs with closures : public detail::ebo_functor_holder<F> { typedef detail::ebo_functor_holder<F> base_t; typedef typename pointer_traits<SlistNodePtr>::element_type slist_node; typedef typename pointer_traits<NodePtr>::element_type node; template<class ConvertibleToF, class RealValuTraits> BOOST_INTRUSIVE_FORCEINLINE node_cast_adaptor(const ConvertibleToF &c2f, const RealValuTraits *traits) : base_t(base_t(c2f, traits)) {} BOOST_INTRUSIVE_FORCEINLINE typename base_t::node_ptr operator()(const slist_node &to_clone) { return base_t::operator()(static_cast<const node &>(to_clone)); } BOOST_INTRUSIVE_FORCEINLINE void operator()(SlistNodePtr to_clone) { base_t::operator()(pointer_traits<NodePtr>::pointer_to(static_cast<node &>(*to_clone))); } }; //bucket_plus_vtraits stores ValueTraits + BucketTraits //this data is needed by iterators to obtain the //value from the iterator and detect the bucket template<class ValueTraits, class BucketTraits> struct bucket_plus_vtraits { typedef BucketTraits bucket_traits; typedef ValueTraits value_traits; static const bool safemode_or_autounlink = is_safe_autounlink<value_traits::link_mode>::value; typedef typename detail::get_slist_impl_from_supposed_value_traits <value_traits>::type slist_impl; typedef typename value_traits::node_traits node_traits; typedef unordered_group_adapter<node_traits> group_traits; typedef typename slist_impl::iterator siterator; typedef bucket_impl<slist_impl> bucket_type; typedef detail::group_functions<node_traits> group_functions_t; typedef typename slist_impl::node_algorithms node_algorithms; typedef typename slist_impl::node_ptr slist_node_ptr; typedef typename node_traits::node_ptr node_ptr; typedef typename node_traits::node node; typedef typename value_traits::value_type value_type; typedef typename value_traits::pointer pointer; typedef typename value_traits::const_pointer const_pointer; typedef typename pointer_traits<pointer>::reference reference; typedef typename pointer_traits <const_pointer>::reference const_reference; typedef circular_slist_algorithms<group_traits> group_algorithms; typedef typename pointer_traits <typename value_traits::pointer>:: template rebind_pointer <const value_traits>::type const_value_traits_ptr; typedef typename pointer_traits <typename value_traits::pointer>:: template rebind_pointer <const bucket_plus_vtraits>::type const_bucket_value_traits_ptr; typedef typename detail::unordered_bucket_ptr_impl <value_traits>::type bucket_ptr; template<class BucketTraitsType> BOOST_INTRUSIVE_FORCEINLINE bucket_plus_vtraits(const ValueTraits &val_traits, BOOST_FWD_REF(BucketTraitsType) b_traits) : data(val_traits, ::boost::forward<BucketTraitsType>(b_traits)) {} BOOST_INTRUSIVE_FORCEINLINE bucket_plus_vtraits & operator =(const bucket_plus_vtraits &x) { data.bucket_traits_ = x.data.bucket_traits_; return *this; } BOOST_INTRUSIVE_FORCEINLINE const_value_traits_ptr priv_value_traits_ptr() const { return pointer_traits<const_value_traits_ptr>::pointer_to(this->priv_value_traits()); } //bucket_value_traits // BOOST_INTRUSIVE_FORCEINLINE const bucket_plus_vtraits &get_bucket_value_traits() const { return *this; } BOOST_INTRUSIVE_FORCEINLINE bucket_plus_vtraits &get_bucket_value_traits() { return *this; } BOOST_INTRUSIVE_FORCEINLINE const_bucket_value_traits_ptr bucket_value_traits_ptr() const { return pointer_traits<const_bucket_value_traits_ptr>::pointer_to(this->get_bucket_value_traits()); } //value traits // BOOST_INTRUSIVE_FORCEINLINE const value_traits &priv_value_traits() const { return this->data; } BOOST_INTRUSIVE_FORCEINLINE value_traits &priv_value_traits() { return this->data; } //bucket_traits // BOOST_INTRUSIVE_FORCEINLINE const bucket_traits &priv_bucket_traits() const { return this->data.bucket_traits_; } BOOST_INTRUSIVE_FORCEINLINE bucket_traits &priv_bucket_traits() { return this->data.bucket_traits_; } //bucket operations BOOST_INTRUSIVE_FORCEINLINE bucket_ptr priv_bucket_pointer() const { return this->priv_bucket_traits().bucket_begin(); } std::size_t priv_bucket_count() const { return this->priv_bucket_traits().bucket_count(); } BOOST_INTRUSIVE_FORCEINLINE bucket_ptr priv_invalid_bucket() const { const bucket_traits &rbt = this->priv_bucket_traits(); return rbt.bucket_begin() + rbt.bucket_count(); } BOOST_INTRUSIVE_FORCEINLINE siterator priv_invalid_local_it() const { return this->priv_bucket_traits().bucket_begin()->before_begin(); } template<class NodeDisposer> static std::size_t priv_erase_from_single_bucket(bucket_type &b, siterator sbefore_first, siterator slast, NodeDisposer node_disposer, detail::true_) //optimize multikey { std::size_t n = 0; siterator const sfirst(++siterator(sbefore_first)); if(sfirst != slast){ node_ptr const nf = detail::dcast_bucket_ptr<node>(sfirst.pointed_node()); node_ptr const nl = detail::dcast_bucket_ptr<node>(slast.pointed_node()); node_ptr const ne = detail::dcast_bucket_ptr<node>(b.end().pointed_node()); if(group_functions_t::next_group_if_first_in_group(nf) != nf) { // The node is at the beginning of a group. if(nl != ne){ group_functions_t::split_group(nl); } } else { node_ptr const group1 = group_functions_t::split_group(nf); if(nl != ne) { node_ptr const group2 = group_functions_t::split_group(ne); if(nf == group2) { //Both first and last in the same group //so join group1 and group2 node_ptr const end1 = group_traits::get_next(group1); node_ptr const end2 = group_traits::get_next(group2); group_traits::set_next(group1, end2); group_traits::set_next(group2, end1); } } } siterator it(++siterator(sbefore_first)); while(it != slast){ node_disposer((it++).pointed_node()); ++n; } b.erase_after(sbefore_first, slast); } return n; } template<class NodeDisposer> static std::size_t priv_erase_from_single_bucket(bucket_type &b, siterator sbefore_first, siterator slast, NodeDisposer node_disposer, detail::false_) //optimize multikey { std::size_t n = 0; siterator it(++siterator(sbefore_first)); while(it != slast){ node_disposer((it++).pointed_node()); ++n; } b.erase_after(sbefore_first, slast); return n; } template<class NodeDisposer> static void priv_erase_node(bucket_type &b, siterator i, NodeDisposer node_disposer, detail::true_) //optimize multikey { node_ptr const ne(detail::dcast_bucket_ptr<node>(b.end().pointed_node())); node_ptr n(detail::dcast_bucket_ptr<node>(i.pointed_node())); node_ptr pos = node_traits::get_next(group_traits::get_next(n)); node_ptr bn; node_ptr nn(node_traits::get_next(n)); if(pos != n) { //Node is the first of the group bn = group_functions_t::get_prev_to_first_in_group(ne, n); //Unlink the rest of the group if it's not the last node of its group if(nn != ne && group_traits::get_next(nn) == n){ group_algorithms::unlink_after(nn); } } else if(nn != ne && group_traits::get_next(nn) == n){ //Node is not the end of the group bn = group_traits::get_next(n); group_algorithms::unlink_after(nn); } else{ //Node is the end of the group bn = group_traits::get_next(n); node_ptr const x(group_algorithms::get_previous_node(n)); group_algorithms::unlink_after(x); } b.erase_after_and_dispose(bucket_type::s_iterator_to(*bn), node_disposer); } template<class NodeDisposer> BOOST_INTRUSIVE_FORCEINLINE static void priv_erase_node(bucket_type &b, siterator i, NodeDisposer node_disposer, detail::false_) //optimize multikey { b.erase_after_and_dispose(b.previous(i), node_disposer); } template<class NodeDisposer, bool OptimizeMultikey> std::size_t priv_erase_node_range( siterator const &before_first_it, std::size_t const first_bucket , siterator const &last_it, std::size_t const last_bucket , NodeDisposer node_disposer, detail::bool_<OptimizeMultikey> optimize_multikey_tag) { std::size_t num_erased(0); siterator last_step_before_it; if(first_bucket != last_bucket){ bucket_type *b = (&this->priv_bucket_pointer()[0]); num_erased += this->priv_erase_from_single_bucket (b[first_bucket], before_first_it, b[first_bucket].end(), node_disposer, optimize_multikey_tag); for(std::size_t i = 0, n = (last_bucket - first_bucket - 1); i != n; ++i){ num_erased += this->priv_erase_whole_bucket(b[first_bucket+i+1], node_disposer); } last_step_before_it = b[last_bucket].before_begin(); } else{ last_step_before_it = before_first_it; } num_erased += this->priv_erase_from_single_bucket (this->priv_bucket_pointer()[last_bucket], last_step_before_it, last_it, node_disposer, optimize_multikey_tag); return num_erased; } static siterator priv_get_last(bucket_type &b, detail::true_) //optimize multikey { //First find the last node of p's group. //This requires checking the first node of the next group or //the bucket node. slist_node_ptr end_ptr(b.end().pointed_node()); node_ptr possible_end(node_traits::get_next( detail::dcast_bucket_ptr<node>(end_ptr))); node_ptr last_node_group(possible_end); while(end_ptr != possible_end){ last_node_group = group_traits::get_next(detail::dcast_bucket_ptr<node>(possible_end)); possible_end = node_traits::get_next(last_node_group); } return bucket_type::s_iterator_to(*last_node_group); } template<class NodeDisposer> std::size_t priv_erase_whole_bucket(bucket_type &b, NodeDisposer node_disposer) { std::size_t num_erased = 0; siterator b_begin(b.before_begin()); siterator nxt(b_begin); ++nxt; siterator const end_sit(b.end()); while(nxt != end_sit){ //No need to init group links as we'll delete all bucket nodes nxt = bucket_type::s_erase_after_and_dispose(b_begin, node_disposer); ++num_erased; } return num_erased; } BOOST_INTRUSIVE_FORCEINLINE static siterator priv_get_last(bucket_type &b, detail::false_) //NOT optimize multikey { return b.previous(b.end()); } static siterator priv_get_previous(bucket_type &b, siterator i, detail::true_) //optimize multikey { node_ptr const elem(detail::dcast_bucket_ptr<node>(i.pointed_node())); node_ptr const prev_in_group(group_traits::get_next(elem)); bool const first_in_group = node_traits::get_next(prev_in_group) != elem; typename bucket_type::node &n = first_in_group ? *group_functions_t::get_prev_to_first_in_group(b.end().pointed_node(), elem) : *group_traits::get_next(elem) ; return bucket_type::s_iterator_to(n); } BOOST_INTRUSIVE_FORCEINLINE static siterator priv_get_previous(bucket_type &b, siterator i, detail::false_) //NOT optimize multikey { return b.previous(i); } std::size_t priv_get_bucket_num_no_hash_store(siterator it, detail::true_) //optimize multikey { const bucket_ptr f(this->priv_bucket_pointer()), l(f + this->priv_bucket_count() - 1); slist_node_ptr bb = group_functions_t::get_bucket_before_begin ( f->end().pointed_node() , l->end().pointed_node() , detail::dcast_bucket_ptr<node>(it.pointed_node())); //Now get the bucket_impl from the iterator const bucket_type &b = static_cast<const bucket_type&> (bucket_type::slist_type::container_from_end_iterator(bucket_type::s_iterator_to(*bb))); //Now just calculate the index b has in the bucket array return static_cast<std::size_t>(&b - &*f); } std::size_t priv_get_bucket_num_no_hash_store(siterator it, detail::false_) //NO optimize multikey { bucket_ptr f(this->priv_bucket_pointer()), l(f + this->priv_bucket_count() - 1); slist_node_ptr first_ptr(f->cend().pointed_node()) , last_ptr(l->cend().pointed_node()); //The end node is embedded in the singly linked list: //iterate until we reach it. while(!(first_ptr <= it.pointed_node() && it.pointed_node() <= last_ptr)){ ++it; } //Now get the bucket_impl from the iterator const bucket_type &b = static_cast<const bucket_type&> (bucket_type::container_from_end_iterator(it)); //Now just calculate the index b has in the bucket array return static_cast<std::size_t>(&b - &*f); } BOOST_INTRUSIVE_FORCEINLINE static std::size_t priv_stored_hash(slist_node_ptr n, detail::true_) //store_hash { return node_traits::get_hash(detail::dcast_bucket_ptr<node>(n)); } BOOST_INTRUSIVE_FORCEINLINE static std::size_t priv_stored_hash(slist_node_ptr, detail::false_) //NO store_hash { return std::size_t(-1); } BOOST_INTRUSIVE_FORCEINLINE node &priv_value_to_node(reference v) { return *this->priv_value_traits().to_node_ptr(v); } BOOST_INTRUSIVE_FORCEINLINE const node &priv_value_to_node(const_reference v) const { return *this->priv_value_traits().to_node_ptr(v); } BOOST_INTRUSIVE_FORCEINLINE reference priv_value_from_slist_node(slist_node_ptr n) { return *this->priv_value_traits().to_value_ptr(detail::dcast_bucket_ptr<node>(n)); } BOOST_INTRUSIVE_FORCEINLINE const_reference priv_value_from_slist_node(slist_node_ptr n) const { return *this->priv_value_traits().to_value_ptr(detail::dcast_bucket_ptr<node>(n)); } void priv_clear_buckets(const bucket_ptr buckets_ptr, const std::size_t bucket_cnt) { bucket_ptr buckets_it = buckets_ptr; for(std::size_t bucket_i = 0; bucket_i != bucket_cnt; ++buckets_it, ++bucket_i){ if(safemode_or_autounlink){ buckets_it->clear_and_dispose(detail::init_disposer<node_algorithms>()); } else{ buckets_it->clear(); } } } BOOST_INTRUSIVE_FORCEINLINE std::size_t priv_stored_or_compute_hash(const value_type &v, detail::true_) const //For store_hash == true { return node_traits::get_hash(this->priv_value_traits().to_node_ptr(v)); } typedef hashtable_iterator<bucket_plus_vtraits, false> iterator; typedef hashtable_iterator<bucket_plus_vtraits, true> const_iterator; BOOST_INTRUSIVE_FORCEINLINE iterator end() { return iterator(this->priv_invalid_local_it(), 0); } BOOST_INTRUSIVE_FORCEINLINE const_iterator end() const { return this->cend(); } BOOST_INTRUSIVE_FORCEINLINE const_iterator cend() const { return const_iterator(this->priv_invalid_local_it(), 0); } //Public functions: struct data_type : public ValueTraits { template<class BucketTraitsType> BOOST_INTRUSIVE_FORCEINLINE data_type(const ValueTraits &val_traits, BOOST_FWD_REF(BucketTraitsType) b_traits) : ValueTraits(val_traits), bucket_traits_(::boost::forward<BucketTraitsType>(b_traits)) {} bucket_traits bucket_traits_; } data; }; template<class Hash, class> struct get_hash { typedef Hash type; }; template<class T> struct get_hash<void, T> { typedef ::boost::hash<T> type; }; template<class EqualTo, class> struct get_equal_to { typedef EqualTo type; }; template<class T> struct get_equal_to<void, T> { typedef std::equal_to<T> type; }; template<class KeyOfValue, class T> struct get_hash_key_of_value { typedef KeyOfValue type; }; template<class T> struct get_hash_key_of_value<void, T> { typedef ::boost::intrusive::detail::identity<T> type; }; template<class T, class VoidOrKeyOfValue> struct hash_key_types_base { typedef typename get_hash_key_of_value < VoidOrKeyOfValue, T>::type key_of_value; typedef typename key_of_value::type key_type; }; template<class T, class VoidOrKeyOfValue, class VoidOrKeyHash> struct hash_key_hash : get_hash < VoidOrKeyHash , typename hash_key_types_base<T, VoidOrKeyOfValue>::key_type > {}; template<class T, class VoidOrKeyOfValue, class VoidOrKeyEqual> struct hash_key_equal : get_equal_to < VoidOrKeyEqual , typename hash_key_types_base<T, VoidOrKeyOfValue>::key_type > {}; //bucket_hash_t //Stores bucket_plus_vtraits plust the hash function template<class ValueTraits, class VoidOrKeyOfValue, class VoidOrKeyHash, class BucketTraits> struct bucket_hash_t //Use public inheritance to avoid MSVC bugs with closures : public detail::ebo_functor_holder <typename hash_key_hash < typename bucket_plus_vtraits<ValueTraits,BucketTraits>::value_traits::value_type , VoidOrKeyOfValue , VoidOrKeyHash >::type > , bucket_plus_vtraits<ValueTraits, BucketTraits> //4 { typedef typename bucket_plus_vtraits<ValueTraits,BucketTraits>::value_traits value_traits; typedef typename value_traits::value_type value_type; typedef typename value_traits::node_traits node_traits; typedef hash_key_hash < value_type, VoidOrKeyOfValue, VoidOrKeyHash> hash_key_hash_t; typedef typename hash_key_hash_t::type hasher; typedef typename hash_key_types_base<value_type, VoidOrKeyOfValue>::key_of_value key_of_value; typedef BucketTraits bucket_traits; typedef bucket_plus_vtraits<ValueTraits, BucketTraits> bucket_plus_vtraits_t; typedef detail::ebo_functor_holder<hasher> base_t; template<class BucketTraitsType> BOOST_INTRUSIVE_FORCEINLINE bucket_hash_t(const ValueTraits &val_traits, BOOST_FWD_REF(BucketTraitsType) b_traits, const hasher & h) : detail::ebo_functor_holder<hasher>(h), bucket_plus_vtraits_t(val_traits, ::boost::forward<BucketTraitsType>(b_traits)) {} BOOST_INTRUSIVE_FORCEINLINE const hasher &priv_hasher() const { return this->base_t::get(); } hasher &priv_hasher() { return this->base_t::get(); } using bucket_plus_vtraits_t::priv_stored_or_compute_hash; //For store_hash == true BOOST_INTRUSIVE_FORCEINLINE std::size_t priv_stored_or_compute_hash(const value_type &v, detail::false_) const //For store_hash == false { return this->priv_hasher()(key_of_value()(v)); } }; template<class ValueTraits, class BucketTraits, class VoidOrKeyOfValue, class VoidOrKeyEqual> struct hashtable_equal_holder { typedef detail::ebo_functor_holder < typename hash_key_equal < typename bucket_plus_vtraits<ValueTraits, BucketTraits>::value_traits::value_type , VoidOrKeyOfValue , VoidOrKeyEqual >::type > type; }; //bucket_hash_equal_t //Stores bucket_hash_t and the equality function when the first //non-empty bucket shall not be cached. template<class ValueTraits, class VoidOrKeyOfValue, class VoidOrKeyHash, class VoidOrKeyEqual, class BucketTraits, bool> struct bucket_hash_equal_t //Use public inheritance to avoid MSVC bugs with closures : public bucket_hash_t<ValueTraits, VoidOrKeyOfValue, VoidOrKeyHash, BucketTraits> //3 , public hashtable_equal_holder<ValueTraits, BucketTraits, VoidOrKeyOfValue, VoidOrKeyEqual>::type //equal { typedef typename hashtable_equal_holder <ValueTraits, BucketTraits, VoidOrKeyOfValue, VoidOrKeyEqual>::type equal_holder_t; typedef bucket_hash_t<ValueTraits, VoidOrKeyOfValue, VoidOrKeyHash, BucketTraits> bucket_hash_type; typedef bucket_plus_vtraits<ValueTraits,BucketTraits> bucket_plus_vtraits_t; typedef ValueTraits value_traits; typedef typename equal_holder_t::functor_type key_equal; typedef typename bucket_hash_type::hasher hasher; typedef BucketTraits bucket_traits; typedef typename bucket_plus_vtraits_t::slist_impl slist_impl; typedef typename slist_impl::iterator siterator; typedef bucket_impl<slist_impl> bucket_type; typedef typename detail::unordered_bucket_ptr_impl<value_traits>::type bucket_ptr; template<class BucketTraitsType> bucket_hash_equal_t(const ValueTraits &val_traits, BOOST_FWD_REF(BucketTraitsType) b_traits, const hasher & h, const key_equal &e) : bucket_hash_type(val_traits, ::boost::forward<BucketTraitsType>(b_traits), h) , equal_holder_t(e) {} BOOST_INTRUSIVE_FORCEINLINE bucket_ptr priv_get_cache() { return this->bucket_hash_type::priv_bucket_pointer(); } BOOST_INTRUSIVE_FORCEINLINE void priv_set_cache(const bucket_ptr &) {} BOOST_INTRUSIVE_FORCEINLINE std::size_t priv_get_cache_bucket_num() { return 0u; } BOOST_INTRUSIVE_FORCEINLINE void priv_initialize_cache() {} BOOST_INTRUSIVE_FORCEINLINE void priv_swap_cache(bucket_hash_equal_t &) {} siterator priv_begin() const { std::size_t n = 0; std::size_t bucket_cnt = this->bucket_hash_type::priv_bucket_count(); for (n = 0; n < bucket_cnt; ++n){ bucket_type &b = this->bucket_hash_type::priv_bucket_pointer()[n]; if(!b.empty()){ return b.begin(); } } return this->bucket_hash_type::priv_invalid_local_it(); } BOOST_INTRUSIVE_FORCEINLINE void priv_insertion_update_cache(std::size_t) {} BOOST_INTRUSIVE_FORCEINLINE void priv_erasure_update_cache_range(std::size_t, std::size_t) {} BOOST_INTRUSIVE_FORCEINLINE void priv_erasure_update_cache() {} BOOST_INTRUSIVE_FORCEINLINE const key_equal &priv_equal() const { return this->equal_holder_t::get(); } BOOST_INTRUSIVE_FORCEINLINE key_equal &priv_equal() { return this->equal_holder_t::get(); } }; //bucket_hash_equal_t //Stores bucket_hash_t and the equality function when the first //non-empty bucket shall be cached. template<class ValueTraits, class VoidOrKeyOfValue, class VoidOrKeyHash, class VoidOrKeyEqual, class BucketTraits> //cache_begin == true version struct bucket_hash_equal_t<ValueTraits, VoidOrKeyOfValue, VoidOrKeyHash, VoidOrKeyEqual, BucketTraits, true> //Use public inheritance to avoid MSVC bugs with closures : bucket_hash_t<ValueTraits, VoidOrKeyOfValue, VoidOrKeyHash, BucketTraits> //2 , hashtable_equal_holder<ValueTraits, BucketTraits, VoidOrKeyOfValue, VoidOrKeyEqual>::type { typedef typename hashtable_equal_holder <ValueTraits, BucketTraits, VoidOrKeyOfValue, VoidOrKeyEqual>::type equal_holder_t; typedef bucket_plus_vtraits<ValueTraits,BucketTraits> bucket_plus_vtraits_t; typedef ValueTraits value_traits; typedef typename equal_holder_t::functor_type key_equal; typedef bucket_hash_t<ValueTraits, VoidOrKeyOfValue, VoidOrKeyHash, BucketTraits> bucket_hash_type; typedef typename bucket_hash_type::hasher hasher; typedef BucketTraits bucket_traits; typedef typename bucket_plus_vtraits_t::slist_impl::iterator siterator; template<class BucketTraitsType> bucket_hash_equal_t(const ValueTraits &val_traits, BOOST_FWD_REF(BucketTraitsType) b_traits, const hasher & h, const key_equal &e) : bucket_hash_type(val_traits, ::boost::forward<BucketTraitsType>(b_traits), h) , equal_holder_t(e) {} typedef typename detail::unordered_bucket_ptr_impl <typename bucket_hash_type::value_traits>::type bucket_ptr; BOOST_INTRUSIVE_FORCEINLINE bucket_ptr &priv_get_cache() { return cached_begin_; } BOOST_INTRUSIVE_FORCEINLINE const bucket_ptr &priv_get_cache() const { return cached_begin_; } BOOST_INTRUSIVE_FORCEINLINE void priv_set_cache(const bucket_ptr &p) { cached_begin_ = p; } BOOST_INTRUSIVE_FORCEINLINE std::size_t priv_get_cache_bucket_num() { return this->cached_begin_ - this->bucket_hash_type::priv_bucket_pointer(); } BOOST_INTRUSIVE_FORCEINLINE void priv_initialize_cache() { this->cached_begin_ = this->bucket_hash_type::priv_invalid_bucket(); } BOOST_INTRUSIVE_FORCEINLINE void priv_swap_cache(bucket_hash_equal_t &other) { ::boost::adl_move_swap(this->cached_begin_, other.cached_begin_); } siterator priv_begin() const { if(this->cached_begin_ == this->bucket_hash_type::priv_invalid_bucket()){ return this->bucket_hash_type::priv_invalid_local_it(); } else{ return this->cached_begin_->begin(); } } void priv_insertion_update_cache(std::size_t insertion_bucket) { bucket_ptr p = this->bucket_hash_type::priv_bucket_pointer() + insertion_bucket; if(p < this->cached_begin_){ this->cached_begin_ = p; } } BOOST_INTRUSIVE_FORCEINLINE const key_equal &priv_equal() const { return this->equal_holder_t::get(); } BOOST_INTRUSIVE_FORCEINLINE key_equal &priv_equal() { return this->equal_holder_t::get(); } void priv_erasure_update_cache_range(std::size_t first_bucket_num, std::size_t last_bucket_num) { //If the last bucket is the end, the cache must be updated //to the last position if all if(this->priv_get_cache_bucket_num() == first_bucket_num && this->bucket_hash_type::priv_bucket_pointer()[first_bucket_num].empty() ){ this->priv_set_cache(this->bucket_hash_type::priv_bucket_pointer() + last_bucket_num); this->priv_erasure_update_cache(); } } void priv_erasure_update_cache() { if(this->cached_begin_ != this->bucket_hash_type::priv_invalid_bucket()){ std::size_t current_n = this->priv_get_cache() - this->bucket_hash_type::priv_bucket_pointer(); for( const std::size_t num_buckets = this->bucket_hash_type::priv_bucket_count() ; current_n < num_buckets ; ++current_n, ++this->priv_get_cache()){ if(!this->priv_get_cache()->empty()){ return; } } this->priv_initialize_cache(); } } bucket_ptr cached_begin_; }; //This wrapper around size_traits is used //to maintain minimal container size with compilers like MSVC //that have problems with EBO and multiple empty base classes template<class DeriveFrom, class SizeType, bool> struct hashtable_size_traits_wrapper : public DeriveFrom { template<class Base, class Arg0, class Arg1, class Arg2> hashtable_size_traits_wrapper( BOOST_FWD_REF(Base) base, BOOST_FWD_REF(Arg0) arg0 , BOOST_FWD_REF(Arg1) arg1, BOOST_FWD_REF(Arg2) arg2) : DeriveFrom(::boost::forward<Base>(base) , ::boost::forward<Arg0>(arg0) , ::boost::forward<Arg1>(arg1) , ::boost::forward<Arg2>(arg2)) {} typedef detail::size_holder < true, SizeType> size_traits;//size_traits size_traits size_traits_; typedef const size_traits & size_traits_const_t; typedef size_traits & size_traits_t; BOOST_INTRUSIVE_FORCEINLINE size_traits_const_t priv_size_traits() const { return size_traits_; } BOOST_INTRUSIVE_FORCEINLINE size_traits_t priv_size_traits() { return size_traits_; } }; template<class DeriveFrom, class SizeType> struct hashtable_size_traits_wrapper<DeriveFrom, SizeType, false> : public DeriveFrom { template<class Base, class Arg0, class Arg1, class Arg2> hashtable_size_traits_wrapper( BOOST_FWD_REF(Base) base, BOOST_FWD_REF(Arg0) arg0 , BOOST_FWD_REF(Arg1) arg1, BOOST_FWD_REF(Arg2) arg2) : DeriveFrom(::boost::forward<Base>(base) , ::boost::forward<Arg0>(arg0) , ::boost::forward<Arg1>(arg1) , ::boost::forward<Arg2>(arg2)) {} typedef detail::size_holder< false, SizeType> size_traits; typedef size_traits size_traits_const_t; typedef size_traits size_traits_t; BOOST_INTRUSIVE_FORCEINLINE size_traits priv_size_traits() const { return size_traits(); } }; //hashdata_internal //Stores bucket_hash_equal_t and split_traits template<class ValueTraits, class VoidOrKeyOfValue, class VoidOrKeyHash, class VoidOrKeyEqual, class BucketTraits, class SizeType, std::size_t BoolFlags> struct hashdata_internal : public hashtable_size_traits_wrapper < bucket_hash_equal_t < ValueTraits, VoidOrKeyOfValue, VoidOrKeyHash, VoidOrKeyEqual , BucketTraits , 0 != (BoolFlags & hash_bool_flags::cache_begin_pos) > //2 , SizeType , (BoolFlags & hash_bool_flags::incremental_pos) != 0 > { typedef hashtable_size_traits_wrapper < bucket_hash_equal_t < ValueTraits, VoidOrKeyOfValue, VoidOrKeyHash, VoidOrKeyEqual , BucketTraits , 0 != (BoolFlags & hash_bool_flags::cache_begin_pos) > //2 , SizeType , (BoolFlags & hash_bool_flags::incremental_pos) != 0 > internal_type; typedef typename internal_type::key_equal key_equal; typedef typename internal_type::hasher hasher; typedef bucket_plus_vtraits<ValueTraits,BucketTraits> bucket_plus_vtraits_t; typedef SizeType size_type; typedef typename internal_type::size_traits split_traits; typedef typename bucket_plus_vtraits_t::bucket_ptr bucket_ptr; typedef typename bucket_plus_vtraits_t::const_value_traits_ptr const_value_traits_ptr; typedef typename bucket_plus_vtraits_t::siterator siterator; typedef typename bucket_plus_vtraits_t::bucket_traits bucket_traits; typedef typename bucket_plus_vtraits_t::value_traits value_traits; typedef typename bucket_plus_vtraits_t::bucket_type bucket_type; typedef typename value_traits::value_type value_type; typedef typename value_traits::pointer pointer; typedef typename value_traits::const_pointer const_pointer; typedef typename pointer_traits<pointer>::reference reference; typedef typename pointer_traits <const_pointer>::reference const_reference; typedef typename value_traits::node_traits node_traits; typedef typename node_traits::node node; typedef typename node_traits::node_ptr node_ptr; typedef typename node_traits::const_node_ptr const_node_ptr; typedef detail::node_functions<node_traits> node_functions_t; typedef typename get_slist_impl <typename reduced_slist_node_traits <typename value_traits::node_traits>::type >::type slist_impl; typedef typename slist_impl::node_algorithms node_algorithms; typedef typename slist_impl::node_ptr slist_node_ptr; typedef hash_key_types_base < typename ValueTraits::value_type , VoidOrKeyOfValue > hash_types_base; typedef typename hash_types_base::key_of_value key_of_value; static const bool store_hash = detail::store_hash_is_true<node_traits>::value; static const bool safemode_or_autounlink = is_safe_autounlink<value_traits::link_mode>::value; static const bool stateful_value_traits = detail::is_stateful_value_traits<value_traits>::value; typedef detail::bool_<store_hash> store_hash_t; typedef detail::transform_iterator < typename slist_impl::iterator , downcast_node_to_value_t < value_traits , false> > local_iterator; typedef detail::transform_iterator < typename slist_impl::iterator , downcast_node_to_value_t < value_traits , true> > const_local_iterator; // template<class BucketTraitsType> hashdata_internal( const ValueTraits &val_traits, BOOST_FWD_REF(BucketTraitsType) b_traits , const hasher & h, const key_equal &e) : internal_type(val_traits, ::boost::forward<BucketTraitsType>(b_traits), h, e) {} BOOST_INTRUSIVE_FORCEINLINE typename internal_type::size_traits_t priv_split_traits() { return this->priv_size_traits(); } BOOST_INTRUSIVE_FORCEINLINE typename internal_type::size_traits_const_t priv_split_traits() const { return this->priv_size_traits(); } ~hashdata_internal() { this->priv_clear_buckets(); } void priv_clear_buckets() { this->internal_type::priv_clear_buckets ( this->priv_get_cache() , this->internal_type::priv_bucket_count() - (this->priv_get_cache() - this->internal_type::priv_bucket_pointer())); } void priv_clear_buckets_and_cache() { this->priv_clear_buckets(); this->priv_initialize_cache(); } void priv_initialize_buckets_and_cache() { this->internal_type::priv_clear_buckets ( this->internal_type::priv_bucket_pointer() , this->internal_type::priv_bucket_count()); this->priv_initialize_cache(); } typedef hashtable_iterator<bucket_plus_vtraits_t, false> iterator; typedef hashtable_iterator<bucket_plus_vtraits_t, true> const_iterator; static std::size_t priv_stored_hash(slist_node_ptr n, detail::true_ true_value) { return bucket_plus_vtraits<ValueTraits, BucketTraits>::priv_stored_hash(n, true_value); } static std::size_t priv_stored_hash(slist_node_ptr n, detail::false_ false_value) { return bucket_plus_vtraits<ValueTraits, BucketTraits>::priv_stored_hash(n, false_value); } //public functions BOOST_INTRUSIVE_FORCEINLINE SizeType split_count() const { return this->priv_split_traits().get_size(); } BOOST_INTRUSIVE_FORCEINLINE iterator iterator_to(reference value) { return iterator(bucket_type::s_iterator_to (this->priv_value_to_node(value)), &this->get_bucket_value_traits()); } const_iterator iterator_to(const_reference value) const { siterator const sit = bucket_type::s_iterator_to ( *pointer_traits<node_ptr>::const_cast_from (pointer_traits<const_node_ptr>::pointer_to(this->priv_value_to_node(value))) ); return const_iterator(sit, &this->get_bucket_value_traits()); } static local_iterator s_local_iterator_to(reference value) { BOOST_STATIC_ASSERT((!stateful_value_traits)); siterator sit = bucket_type::s_iterator_to(*value_traits::to_node_ptr(value)); return local_iterator(sit, const_value_traits_ptr()); } static const_local_iterator s_local_iterator_to(const_reference value) { BOOST_STATIC_ASSERT((!stateful_value_traits)); siterator const sit = bucket_type::s_iterator_to ( *pointer_traits<node_ptr>::const_cast_from (value_traits::to_node_ptr(value)) ); return const_local_iterator(sit, const_value_traits_ptr()); } local_iterator local_iterator_to(reference value) { siterator sit = bucket_type::s_iterator_to(this->priv_value_to_node(value)); return local_iterator(sit, this->priv_value_traits_ptr()); } const_local_iterator local_iterator_to(const_reference value) const { siterator sit = bucket_type::s_iterator_to ( *pointer_traits<node_ptr>::const_cast_from (pointer_traits<const_node_ptr>::pointer_to(this->priv_value_to_node(value))) ); return const_local_iterator(sit, this->priv_value_traits_ptr()); } BOOST_INTRUSIVE_FORCEINLINE size_type bucket_count() const { const std::size_t bc = this->priv_bucket_count(); BOOST_INTRUSIVE_INVARIANT_ASSERT(sizeof(size_type) >= sizeof(std::size_t) || bc <= size_type(-1)); return static_cast<size_type>(bc); } BOOST_INTRUSIVE_FORCEINLINE size_type bucket_size(size_type n) const { return this->priv_bucket_pointer()[n].size(); } BOOST_INTRUSIVE_FORCEINLINE bucket_ptr bucket_pointer() const { return this->priv_bucket_pointer(); } BOOST_INTRUSIVE_FORCEINLINE local_iterator begin(size_type n) { return local_iterator(this->priv_bucket_pointer()[n].begin(), this->priv_value_traits_ptr()); } BOOST_INTRUSIVE_FORCEINLINE const_local_iterator begin(size_type n) const { return this->cbegin(n); } static BOOST_INTRUSIVE_FORCEINLINE size_type suggested_upper_bucket_count(size_type n) { return prime_list_holder<0>::suggested_upper_bucket_count(n); } static BOOST_INTRUSIVE_FORCEINLINE size_type suggested_lower_bucket_count(size_type n) { return prime_list_holder<0>::suggested_lower_bucket_count(n); } const_local_iterator cbegin(size_type n) const { return const_local_iterator ( pointer_traits<bucket_ptr>::const_cast_from(this->priv_bucket_pointer())[n].begin() , this->priv_value_traits_ptr()); } using internal_type::end; using internal_type::cend; local_iterator end(size_type n) { return local_iterator(this->priv_bucket_pointer()[n].end(), this->priv_value_traits_ptr()); } BOOST_INTRUSIVE_FORCEINLINE const_local_iterator end(size_type n) const { return this->cend(n); } const_local_iterator cend(size_type n) const { return const_local_iterator ( pointer_traits<bucket_ptr>::const_cast_from(this->priv_bucket_pointer())[n].end() , this->priv_value_traits_ptr()); } //Public functions for hashtable_impl BOOST_INTRUSIVE_FORCEINLINE iterator begin() { return iterator(this->priv_begin(), &this->get_bucket_value_traits()); } BOOST_INTRUSIVE_FORCEINLINE const_iterator begin() const { return this->cbegin(); } BOOST_INTRUSIVE_FORCEINLINE const_iterator cbegin() const { return const_iterator(this->priv_begin(), &this->get_bucket_value_traits()); } BOOST_INTRUSIVE_FORCEINLINE hasher hash_function() const { return this->priv_hasher(); } BOOST_INTRUSIVE_FORCEINLINE key_equal key_eq() const { return this->priv_equal(); } }; /// @endcond //! The class template hashtable is an intrusive hash table container, that //! is used to construct intrusive unordered_set and unordered_multiset containers. The //! no-throw guarantee holds only, if the VoidOrKeyEqual object and Hasher don't throw. //! //! hashtable is a semi-intrusive container: each object to be stored in the //! container must contain a proper hook, but the container also needs //! additional auxiliary memory to work: hashtable needs a pointer to an array //! of type `bucket_type` to be passed in the constructor. This bucket array must //! have at least the same lifetime as the container. This makes the use of //! hashtable more complicated than purely intrusive containers. //! `bucket_type` is default-constructible, copyable and assignable //! //! The template parameter \c T is the type to be managed by the container. //! The user can specify additional options and if no options are provided //! default options are used. //! //! The container supports the following options: //! \c base_hook<>/member_hook<>/value_traits<>, //! \c constant_time_size<>, \c size_type<>, \c hash<> and \c equal<> //! \c bucket_traits<>, power_2_buckets<>, cache_begin<> and incremental<>. //! //! hashtable only provides forward iterators but it provides 4 iterator types: //! iterator and const_iterator to navigate through the whole container and //! local_iterator and const_local_iterator to navigate through the values //! stored in a single bucket. Local iterators are faster and smaller. //! //! It's not recommended to use non constant-time size hashtables because several //! key functions, like "empty()", become non-constant time functions. Non //! constant_time size hashtables are mainly provided to support auto-unlink hooks. //! //! hashtables, does not make automatic rehashings nor //! offers functions related to a load factor. Rehashing can be explicitly requested //! and the user must provide a new bucket array that will be used from that moment. //! //! Since no automatic rehashing is done, iterators are never invalidated when //! inserting or erasing elements. Iterators are only invalidated when rehashing. #if defined(BOOST_INTRUSIVE_DOXYGEN_INVOKED) template<class T, class ...Options> #else template<class ValueTraits, class VoidOrKeyOfValue, class VoidOrKeyHash, class VoidOrKeyEqual, class BucketTraits, class SizeType, std::size_t BoolFlags> #endif class hashtable_impl : private hashtable_size_traits_wrapper < hashdata_internal < ValueTraits , VoidOrKeyOfValue, VoidOrKeyHash, VoidOrKeyEqual , BucketTraits, SizeType , BoolFlags & (hash_bool_flags::incremental_pos | hash_bool_flags::cache_begin_pos) //1 > , SizeType , (BoolFlags & hash_bool_flags::constant_time_size_pos) != 0 > { typedef hashtable_size_traits_wrapper < hashdata_internal < ValueTraits , VoidOrKeyOfValue, VoidOrKeyHash, VoidOrKeyEqual , BucketTraits, SizeType , BoolFlags & (hash_bool_flags::incremental_pos | hash_bool_flags::cache_begin_pos) //1 > , SizeType , (BoolFlags & hash_bool_flags::constant_time_size_pos) != 0 > internal_type; typedef typename internal_type::size_traits size_traits; typedef hash_key_types_base < typename ValueTraits::value_type , VoidOrKeyOfValue > hash_types_base; public: typedef ValueTraits value_traits; /// @cond typedef BucketTraits bucket_traits; typedef typename internal_type::slist_impl slist_impl; typedef bucket_plus_vtraits<ValueTraits, BucketTraits> bucket_plus_vtraits_t; typedef typename bucket_plus_vtraits_t::const_value_traits_ptr const_value_traits_ptr; using internal_type::begin; using internal_type::cbegin; using internal_type::end; using internal_type::cend; using internal_type::hash_function; using internal_type::key_eq; using internal_type::bucket_size; using internal_type::bucket_count; using internal_type::local_iterator_to; using internal_type::s_local_iterator_to; using internal_type::iterator_to; using internal_type::bucket_pointer; using internal_type::suggested_upper_bucket_count; using internal_type::suggested_lower_bucket_count; using internal_type::split_count; /// @endcond typedef typename value_traits::pointer pointer; typedef typename value_traits::const_pointer const_pointer; typedef typename value_traits::value_type value_type; typedef typename hash_types_base::key_type key_type; typedef typename hash_types_base::key_of_value key_of_value; typedef typename pointer_traits<pointer>::reference reference; typedef typename pointer_traits<const_pointer>::reference const_reference; typedef typename pointer_traits<pointer>::difference_type difference_type; typedef SizeType size_type; typedef typename internal_type::key_equal key_equal; typedef typename internal_type::hasher hasher; typedef bucket_impl<slist_impl> bucket_type; typedef typename internal_type::bucket_ptr bucket_ptr; typedef typename slist_impl::iterator siterator; typedef typename slist_impl::const_iterator const_siterator; typedef typename internal_type::iterator iterator; typedef typename internal_type::const_iterator const_iterator; typedef typename internal_type::local_iterator local_iterator; typedef typename internal_type::const_local_iterator const_local_iterator; typedef typename value_traits::node_traits node_traits; typedef typename node_traits::node node; typedef typename pointer_traits <pointer>::template rebind_pointer < node >::type node_ptr; typedef typename pointer_traits <pointer>::template rebind_pointer < const node >::type const_node_ptr; typedef typename pointer_traits <node_ptr>::reference node_reference; typedef typename pointer_traits <const_node_ptr>::reference const_node_reference; typedef typename slist_impl::node_algorithms node_algorithms; static const bool stateful_value_traits = internal_type::stateful_value_traits; static const bool store_hash = internal_type::store_hash; static const bool unique_keys = 0 != (BoolFlags & hash_bool_flags::unique_keys_pos); static const bool constant_time_size = 0 != (BoolFlags & hash_bool_flags::constant_time_size_pos); static const bool cache_begin = 0 != (BoolFlags & hash_bool_flags::cache_begin_pos); static const bool compare_hash = 0 != (BoolFlags & hash_bool_flags::compare_hash_pos); static const bool incremental = 0 != (BoolFlags & hash_bool_flags::incremental_pos); static const bool power_2_buckets = incremental || (0 != (BoolFlags & hash_bool_flags::power_2_buckets_pos)); static const bool optimize_multikey = detail::optimize_multikey_is_true<node_traits>::value && !unique_keys; /// @cond static const bool is_multikey = !unique_keys; private: //Configuration error: compare_hash<> can't be specified without store_hash<> //See documentation for more explanations BOOST_STATIC_ASSERT((!compare_hash || store_hash)); typedef typename slist_impl::node_ptr slist_node_ptr; typedef typename pointer_traits <slist_node_ptr>::template rebind_pointer < void >::type void_pointer; //We'll define group traits, but these won't be instantiated if //optimize_multikey is not true typedef unordered_group_adapter<node_traits> group_traits; typedef circular_slist_algorithms<group_traits> group_algorithms; typedef typename internal_type::store_hash_t store_hash_t; typedef detail::bool_<optimize_multikey> optimize_multikey_t; typedef detail::bool_<cache_begin> cache_begin_t; typedef detail::bool_<power_2_buckets> power_2_buckets_t; typedef typename internal_type::split_traits split_traits; typedef detail::group_functions<node_traits> group_functions_t; typedef detail::node_functions<node_traits> node_functions_t; private: //noncopyable, movable BOOST_MOVABLE_BUT_NOT_COPYABLE(hashtable_impl) static const bool safemode_or_autounlink = internal_type::safemode_or_autounlink; //Constant-time size is incompatible with auto-unlink hooks! BOOST_STATIC_ASSERT(!(constant_time_size && ((int)value_traits::link_mode == (int)auto_unlink))); //Cache begin is incompatible with auto-unlink hooks! BOOST_STATIC_ASSERT(!(cache_begin && ((int)value_traits::link_mode == (int)auto_unlink))); template<class Disposer> struct typeof_node_disposer { typedef node_cast_adaptor < detail::node_disposer< Disposer, value_traits, CircularSListAlgorithms> , slist_node_ptr, node_ptr > type; }; template<class Disposer> typename typeof_node_disposer<Disposer>::type make_node_disposer(const Disposer &disposer) const { typedef typename typeof_node_disposer<Disposer>::type return_t; return return_t(disposer, &this->priv_value_traits()); } /// @endcond public: typedef detail::insert_commit_data_impl insert_commit_data; public: //! <b>Requires</b>: buckets must not be being used by any other resource. //! //! <b>Effects</b>: Constructs an empty unordered_set, storing a reference //! to the bucket array and copies of the key_hasher and equal_func functors. //! //! <b>Complexity</b>: Constant. //! //! <b>Throws</b>: If value_traits::node_traits::node //! constructor throws (this does not happen with predefined Boost.Intrusive hooks) //! or the copy constructor or invocation of hash_func or equal_func throws. //! //! <b>Notes</b>: buckets array must be disposed only after //! *this is disposed. explicit hashtable_impl ( const bucket_traits &b_traits , const hasher & hash_func = hasher() , const key_equal &equal_func = key_equal() , const value_traits &v_traits = value_traits()) : internal_type(v_traits, b_traits, hash_func, equal_func) { this->priv_initialize_buckets_and_cache(); this->priv_size_traits().set_size(size_type(0)); size_type bucket_sz = this->bucket_count(); BOOST_INTRUSIVE_INVARIANT_ASSERT(bucket_sz != 0); //Check power of two bucket array if the option is activated BOOST_INTRUSIVE_INVARIANT_ASSERT (!power_2_buckets || (0 == (bucket_sz & (bucket_sz-1)))); this->priv_split_traits().set_size(bucket_sz>>1); } //! <b>Requires</b>: buckets must not be being used by any other resource //! and dereferencing iterator must yield an lvalue of type value_type. //! //! <b>Effects</b>: Constructs an empty container and inserts elements from //! [b, e). //! //! <b>Complexity</b>: If N is distance(b, e): Average case is O(N) //! (with a good hash function and with buckets_len >= N),worst case O(N^2). //! //! <b>Throws</b>: If value_traits::node_traits::node //! constructor throws (this does not happen with predefined Boost.Intrusive hooks) //! or the copy constructor or invocation of hasher or key_equal throws. //! //! <b>Notes</b>: buckets array must be disposed only after //! *this is disposed. template<class Iterator> hashtable_impl ( bool unique, Iterator b, Iterator e , const bucket_traits &b_traits , const hasher & hash_func = hasher() , const key_equal &equal_func = key_equal() , const value_traits &v_traits = value_traits()) : internal_type(v_traits, b_traits, hash_func, equal_func) { this->priv_initialize_buckets_and_cache(); this->priv_size_traits().set_size(size_type(0)); size_type bucket_sz = this->bucket_count(); BOOST_INTRUSIVE_INVARIANT_ASSERT(bucket_sz != 0); //Check power of two bucket array if the option is activated BOOST_INTRUSIVE_INVARIANT_ASSERT (!power_2_buckets || (0 == (bucket_sz & (bucket_sz-1)))); this->priv_split_traits().set_size(bucket_sz>>1); //Now insert if(unique) this->insert_unique(b, e); else this->insert_equal(b, e); } //! <b>Effects</b>: Constructs a container moving resources from another container. //! Internal value traits, bucket traits, hasher and comparison are move constructed and //! nodes belonging to x are linked to *this. //! //! <b>Complexity</b>: Constant. //! //! <b>Throws</b>: If value_traits::node_traits::node's //! move constructor throws (this does not happen with predefined Boost.Intrusive hooks) //! or the move constructor of value traits, bucket traits, hasher or comparison throws. hashtable_impl(BOOST_RV_REF(hashtable_impl) x) : internal_type( ::boost::move(x.priv_value_traits()) , ::boost::move(x.priv_bucket_traits()) , ::boost::move(x.priv_hasher()) , ::boost::move(x.priv_equal()) ) { this->priv_swap_cache(x); x.priv_initialize_cache(); this->priv_size_traits().set_size(x.priv_size_traits().get_size()); x.priv_size_traits().set_size(size_type(0)); this->priv_split_traits().set_size(x.priv_split_traits().get_size()); x.priv_split_traits().set_size(size_type(0)); } //! <b>Effects</b>: Equivalent to swap. //! hashtable_impl& operator=(BOOST_RV_REF(hashtable_impl) x) { this->swap(x); return *this; } #if defined(BOOST_INTRUSIVE_DOXYGEN_INVOKED) //! <b>Effects</b>: Detaches all elements from this. The objects in the unordered_set //! are not deleted (i.e. no destructors are called). //! //! <b>Complexity</b>: Linear to the number of elements in the unordered_set, if //! it's a safe-mode or auto-unlink value. Otherwise constant. //! //! <b>Throws</b>: Nothing. ~hashtable_impl(); //! <b>Effects</b>: Returns an iterator pointing to the beginning of the unordered_set. //! //! <b>Complexity</b>: Amortized constant time. //! Worst case (empty unordered_set): O(this->bucket_count()) //! //! <b>Throws</b>: Nothing. iterator begin(); //! <b>Effects</b>: Returns a const_iterator pointing to the beginning //! of the unordered_set. //! //! <b>Complexity</b>: Amortized constant time. //! Worst case (empty unordered_set): O(this->bucket_count()) //! //! <b>Throws</b>: Nothing. const_iterator begin() const; //! <b>Effects</b>: Returns a const_iterator pointing to the beginning //! of the unordered_set. //! //! <b>Complexity</b>: Amortized constant time. //! Worst case (empty unordered_set): O(this->bucket_count()) //! //! <b>Throws</b>: Nothing. const_iterator cbegin() const; //! <b>Effects</b>: Returns an iterator pointing to the end of the unordered_set. //! //! <b>Complexity</b>: Constant. //! //! <b>Throws</b>: Nothing. iterator end(); //! <b>Effects</b>: Returns a const_iterator pointing to the end of the unordered_set. //! //! <b>Complexity</b>: Constant. //! //! <b>Throws</b>: Nothing. const_iterator end() const; //! <b>Effects</b>: Returns a const_iterator pointing to the end of the unordered_set. //! //! <b>Complexity</b>: Constant. //! //! <b>Throws</b>: Nothing. const_iterator cend() const; //! <b>Effects</b>: Returns the hasher object used by the unordered_set. //! //! <b>Complexity</b>: Constant. //! //! <b>Throws</b>: If hasher copy-constructor throws. hasher hash_function() const; //! <b>Effects</b>: Returns the key_equal object used by the unordered_set. //! //! <b>Complexity</b>: Constant. //! //! <b>Throws</b>: If key_equal copy-constructor throws. key_equal key_eq() const; #endif //! <b>Effects</b>: Returns true if the container is empty. //! //! <b>Complexity</b>: if constant-time size and cache_begin options are disabled, //! average constant time (worst case, with empty() == true: O(this->bucket_count()). //! Otherwise constant. //! //! <b>Throws</b>: Nothing. bool empty() const { if(constant_time_size){ return !this->size(); } else if(cache_begin){ return this->begin() == this->end(); } else{ size_type bucket_cnt = this->bucket_count(); const bucket_type *b = boost::movelib::to_raw_pointer(this->priv_bucket_pointer()); for (size_type n = 0; n < bucket_cnt; ++n, ++b){ if(!b->empty()){ return false; } } return true; } } //! <b>Effects</b>: Returns the number of elements stored in the unordered_set. //! //! <b>Complexity</b>: Linear to elements contained in *this if //! constant_time_size is false. Constant-time otherwise. //! //! <b>Throws</b>: Nothing. size_type size() const { if(constant_time_size) return this->priv_size_traits().get_size(); else{ size_type len = 0; size_type bucket_cnt = this->bucket_count(); const bucket_type *b = boost::movelib::to_raw_pointer(this->priv_bucket_pointer()); for (size_type n = 0; n < bucket_cnt; ++n, ++b){ len += b->size(); } return len; } } //! <b>Requires</b>: the hasher and the equality function unqualified swap //! call should not throw. //! //! <b>Effects</b>: Swaps the contents of two unordered_sets. //! Swaps also the contained bucket array and equality and hasher functors. //! //! <b>Complexity</b>: Constant. //! //! <b>Throws</b>: If the swap() call for the comparison or hash functors //! found using ADL throw. Basic guarantee. void swap(hashtable_impl& other) { //These can throw ::boost::adl_move_swap(this->priv_equal(), other.priv_equal()); ::boost::adl_move_swap(this->priv_hasher(), other.priv_hasher()); //These can't throw ::boost::adl_move_swap(this->priv_bucket_traits(), other.priv_bucket_traits()); ::boost::adl_move_swap(this->priv_value_traits(), other.priv_value_traits()); this->priv_swap_cache(other); this->priv_size_traits().swap(other.priv_size_traits()); this->priv_split_traits().swap(other.priv_split_traits()); } //! <b>Requires</b>: Disposer::operator()(pointer) shouldn't throw //! Cloner should yield to nodes that compare equal and produce the same //! hash than the original node. //! //! <b>Effects</b>: Erases all the elements from *this //! calling Disposer::operator()(pointer), clones all the //! elements from src calling Cloner::operator()(const_reference ) //! and inserts them on *this. The hash function and the equality //! predicate are copied from the source. //! //! If store_hash option is true, this method does not use the hash function. //! //! If any operation throws, all cloned elements are unlinked and disposed //! calling Disposer::operator()(pointer). //! //! <b>Complexity</b>: Linear to erased plus inserted elements. //! //! <b>Throws</b>: If cloner or hasher throw or hash or equality predicate copying //! throws. Basic guarantee. template <class Cloner, class Disposer> BOOST_INTRUSIVE_FORCEINLINE void clone_from(const hashtable_impl &src, Cloner cloner, Disposer disposer) { this->priv_clone_from(src, cloner, disposer); } //! <b>Requires</b>: Disposer::operator()(pointer) shouldn't throw //! Cloner should yield to nodes that compare equal and produce the same //! hash than the original node. //! //! <b>Effects</b>: Erases all the elements from *this //! calling Disposer::operator()(pointer), clones all the //! elements from src calling Cloner::operator()(reference) //! and inserts them on *this. The hash function and the equality //! predicate are copied from the source. //! //! If store_hash option is true, this method does not use the hash function. //! //! If any operation throws, all cloned elements are unlinked and disposed //! calling Disposer::operator()(pointer). //! //! <b>Complexity</b>: Linear to erased plus inserted elements. //! //! <b>Throws</b>: If cloner or hasher throw or hash or equality predicate copying //! throws. Basic guarantee. template <class Cloner, class Disposer> BOOST_INTRUSIVE_FORCEINLINE void clone_from(BOOST_RV_REF(hashtable_impl) src, Cloner cloner, Disposer disposer) { this->priv_clone_from(static_cast<hashtable_impl&>(src), cloner, disposer); } //! <b>Requires</b>: value must be an lvalue //! //! <b>Effects</b>: Inserts the value into the unordered_set. //! //! <b>Returns</b>: An iterator to the inserted value. //! //! <b>Complexity</b>: Average case O(1), worst case O(this->size()). //! //! <b>Throws</b>: If the internal hasher or the equality functor throws. Strong guarantee. //! //! <b>Note</b>: Does not affect the validity of iterators and references. //! No copy-constructors are called. iterator insert_equal(reference value) { size_type bucket_num; std::size_t hash_value; siterator prev; siterator const it = this->priv_find (key_of_value()(value), this->priv_hasher(), this->priv_equal(), bucket_num, hash_value, prev); bool const next_is_in_group = optimize_multikey && it != this->priv_invalid_local_it(); return this->priv_insert_equal_after_find(value, bucket_num, hash_value, prev, next_is_in_group); } //! <b>Requires</b>: Dereferencing iterator must yield an lvalue //! of type value_type. //! //! <b>Effects</b>: Equivalent to this->insert_equal(t) for each element in [b, e). //! //! <b>Complexity</b>: Average case O(N), where N is distance(b, e). //! Worst case O(N*this->size()). //! //! <b>Throws</b>: If the internal hasher or the equality functor throws. Basic guarantee. //! //! <b>Note</b>: Does not affect the validity of iterators and references. //! No copy-constructors are called. template<class Iterator> void insert_equal(Iterator b, Iterator e) { for (; b != e; ++b) this->insert_equal(*b); } //! <b>Requires</b>: value must be an lvalue //! //! <b>Effects</b>: Tries to inserts value into the unordered_set. //! //! <b>Returns</b>: If the value //! is not already present inserts it and returns a pair containing the //! iterator to the new value and true. If there is an equivalent value //! returns a pair containing an iterator to the already present value //! and false. //! //! <b>Complexity</b>: Average case O(1), worst case O(this->size()). //! //! <b>Throws</b>: If the internal hasher or the equality functor throws. Strong guarantee. //! //! <b>Note</b>: Does not affect the validity of iterators and references. //! No copy-constructors are called. std::pair<iterator, bool> insert_unique(reference value) { insert_commit_data commit_data; std::pair<iterator, bool> ret = this->insert_unique_check(key_of_value()(value), commit_data); if(ret.second){ ret.first = this->insert_unique_commit(value, commit_data); } return ret; } //! <b>Requires</b>: Dereferencing iterator must yield an lvalue //! of type value_type. //! //! <b>Effects</b>: Equivalent to this->insert_unique(t) for each element in [b, e). //! //! <b>Complexity</b>: Average case O(N), where N is distance(b, e). //! Worst case O(N*this->size()). //! //! <b>Throws</b>: If the internal hasher or the equality functor throws. Basic guarantee. //! //! <b>Note</b>: Does not affect the validity of iterators and references. //! No copy-constructors are called. template<class Iterator> void insert_unique(Iterator b, Iterator e) { for (; b != e; ++b) this->insert_unique(*b); } //! <b>Requires</b>: "hash_func" must be a hash function that induces //! the same hash values as the stored hasher. The difference is that //! "hash_func" hashes the given key instead of the value_type. //! //! "equal_func" must be a equality function that induces //! the same equality as key_equal. The difference is that //! "equal_func" compares an arbitrary key with the contained values. //! //! <b>Effects</b>: Checks if a value can be inserted in the unordered_set, using //! a user provided key instead of the value itself. //! //! <b>Returns</b>: If there is an equivalent value //! returns a pair containing an iterator to the already present value //! and false. If the value can be inserted returns true in the returned //! pair boolean and fills "commit_data" that is meant to be used with //! the "insert_commit" function. //! //! <b>Complexity</b>: Average case O(1), worst case O(this->size()). //! //! <b>Throws</b>: If hash_func or equal_func throw. Strong guarantee. //! //! <b>Notes</b>: This function is used to improve performance when constructing //! a value_type is expensive: if there is an equivalent value //! the constructed object must be discarded. Many times, the part of the //! node that is used to impose the hash or the equality is much cheaper to //! construct than the value_type and this function offers the possibility to //! use that the part to check if the insertion will be successful. //! //! If the check is successful, the user can construct the value_type and use //! "insert_commit" to insert the object in constant-time. //! //! "commit_data" remains valid for a subsequent "insert_commit" only if no more //! objects are inserted or erased from the unordered_set. //! //! After a successful rehashing insert_commit_data remains valid. template<class KeyType, class KeyHasher, class KeyEqual> std::pair<iterator, bool> insert_unique_check ( const KeyType &key , KeyHasher hash_func , KeyEqual equal_func , insert_commit_data &commit_data) { size_type bucket_num; siterator prev; siterator const pos = this->priv_find(key, hash_func, equal_func, bucket_num, commit_data.hash, prev); return std::pair<iterator, bool> ( iterator(pos, &this->get_bucket_value_traits()) , pos == this->priv_invalid_local_it()); } //! <b>Effects</b>: Checks if a value can be inserted in the unordered_set, using //! a user provided key instead of the value itself. //! //! <b>Returns</b>: If there is an equivalent value //! returns a pair containing an iterator to the already present value //! and false. If the value can be inserted returns true in the returned //! pair boolean and fills "commit_data" that is meant to be used with //! the "insert_commit" function. //! //! <b>Complexity</b>: Average case O(1), worst case O(this->size()). //! //! <b>Throws</b>: If hasher or key_compare throw. Strong guarantee. //! //! <b>Notes</b>: This function is used to improve performance when constructing //! a value_type is expensive: if there is an equivalent value //! the constructed object must be discarded. Many times, the part of the //! node that is used to impose the hash or the equality is much cheaper to //! construct than the value_type and this function offers the possibility to //! use that the part to check if the insertion will be successful. //! //! If the check is successful, the user can construct the value_type and use //! "insert_commit" to insert the object in constant-time. //! //! "commit_data" remains valid for a subsequent "insert_commit" only if no more //! objects are inserted or erased from the unordered_set. //! //! After a successful rehashing insert_commit_data remains valid. BOOST_INTRUSIVE_FORCEINLINE std::pair<iterator, bool> insert_unique_check ( const key_type &key, insert_commit_data &commit_data) { return this->insert_unique_check(key, this->priv_hasher(), this->priv_equal(), commit_data); } //! <b>Requires</b>: value must be an lvalue of type value_type. commit_data //! must have been obtained from a previous call to "insert_check". //! No objects should have been inserted or erased from the unordered_set between //! the "insert_check" that filled "commit_data" and the call to "insert_commit". //! //! <b>Effects</b>: Inserts the value in the unordered_set using the information obtained //! from the "commit_data" that a previous "insert_check" filled. //! //! <b>Returns</b>: An iterator to the newly inserted object. //! //! <b>Complexity</b>: Constant time. //! //! <b>Throws</b>: Nothing. //! //! <b>Notes</b>: This function has only sense if a "insert_check" has been //! previously executed to fill "commit_data". No value should be inserted or //! erased between the "insert_check" and "insert_commit" calls. //! //! After a successful rehashing insert_commit_data remains valid. iterator insert_unique_commit(reference value, const insert_commit_data &commit_data) { size_type bucket_num = this->priv_hash_to_bucket(commit_data.hash); bucket_type &b = this->priv_bucket_pointer()[bucket_num]; this->priv_size_traits().increment(); node_ptr const n = pointer_traits<node_ptr>::pointer_to(this->priv_value_to_node(value)); BOOST_INTRUSIVE_SAFE_HOOK_DEFAULT_ASSERT(!safemode_or_autounlink || node_algorithms::unique(n)); node_functions_t::store_hash(n, commit_data.hash, store_hash_t()); this->priv_insertion_update_cache(bucket_num); group_functions_t::insert_in_group(n, n, optimize_multikey_t()); return iterator(b.insert_after(b.before_begin(), *n), &this->get_bucket_value_traits()); } //! <b>Effects</b>: Erases the element pointed to by i. //! //! <b>Complexity</b>: Average case O(1), worst case O(this->size()). //! //! <b>Throws</b>: Nothing. //! //! <b>Note</b>: Invalidates the iterators (but not the references) //! to the erased element. No destructors are called. BOOST_INTRUSIVE_FORCEINLINE void erase(const_iterator i) { this->erase_and_dispose(i, detail::null_disposer()); } //! <b>Effects</b>: Erases the range pointed to by b end e. //! //! <b>Complexity</b>: Average case O(distance(b, e)), //! worst case O(this->size()). //! //! <b>Throws</b>: Nothing. //! //! <b>Note</b>: Invalidates the iterators (but not the references) //! to the erased elements. No destructors are called. BOOST_INTRUSIVE_FORCEINLINE void erase(const_iterator b, const_iterator e) { this->erase_and_dispose(b, e, detail::null_disposer()); } //! <b>Effects</b>: Erases all the elements with the given value. //! //! <b>Returns</b>: The number of erased elements. //! //! <b>Complexity</b>: Average case O(this->count(value)). //! Worst case O(this->size()). //! //! <b>Throws</b>: If the internal hasher or the equality functor throws. //! Basic guarantee. //! //! <b>Note</b>: Invalidates the iterators (but not the references) //! to the erased elements. No destructors are called. BOOST_INTRUSIVE_FORCEINLINE size_type erase(const key_type &key) { return this->erase(key, this->priv_hasher(), this->priv_equal()); } //! <b>Requires</b>: "hash_func" must be a hash function that induces //! the same hash values as the stored hasher. The difference is that //! "hash_func" hashes the given key instead of the value_type. //! //! "equal_func" must be a equality function that induces //! the same equality as key_equal. The difference is that //! "equal_func" compares an arbitrary key with the contained values. //! //! <b>Effects</b>: Erases all the elements that have the same hash and //! compare equal with the given key. //! //! <b>Returns</b>: The number of erased elements. //! //! <b>Complexity</b>: Average case O(this->count(value)). //! Worst case O(this->size()). //! //! <b>Throws</b>: If hash_func or equal_func throw. Basic guarantee. //! //! <b>Note</b>: Invalidates the iterators (but not the references) //! to the erased elements. No destructors are called. template<class KeyType, class KeyHasher, class KeyEqual> BOOST_INTRUSIVE_FORCEINLINE size_type erase(const KeyType& key, KeyHasher hash_func, KeyEqual equal_func) { return this->erase_and_dispose(key, hash_func, equal_func, detail::null_disposer()); } //! <b>Requires</b>: Disposer::operator()(pointer) shouldn't throw. //! //! <b>Effects</b>: Erases the element pointed to by i. //! Disposer::operator()(pointer) is called for the removed element. //! //! <b>Complexity</b>: Average case O(1), worst case O(this->size()). //! //! <b>Throws</b>: Nothing. //! //! <b>Note</b>: Invalidates the iterators //! to the erased elements. template<class Disposer> BOOST_INTRUSIVE_DOC1ST(void , typename detail::disable_if_convertible<Disposer BOOST_INTRUSIVE_I const_iterator>::type) erase_and_dispose(const_iterator i, Disposer disposer) { //Get the bucket number and local iterator for both iterators siterator const first_local_it(i.slist_it()); size_type const first_bucket_num = this->priv_get_bucket_num(first_local_it); this->priv_erase_node(this->priv_bucket_pointer()[first_bucket_num], first_local_it, make_node_disposer(disposer), optimize_multikey_t()); this->priv_size_traits().decrement(); this->priv_erasure_update_cache_range(first_bucket_num, first_bucket_num); } //! <b>Requires</b>: Disposer::operator()(pointer) shouldn't throw. //! //! <b>Effects</b>: Erases the range pointed to by b end e. //! Disposer::operator()(pointer) is called for the removed elements. //! //! <b>Complexity</b>: Average case O(distance(b, e)), //! worst case O(this->size()). //! //! <b>Throws</b>: Nothing. //! //! <b>Note</b>: Invalidates the iterators //! to the erased elements. template<class Disposer> void erase_and_dispose(const_iterator b, const_iterator e, Disposer disposer) { if(b != e){ //Get the bucket number and local iterator for both iterators siterator first_local_it(b.slist_it()); size_type first_bucket_num = this->priv_get_bucket_num(first_local_it); const bucket_ptr buck_ptr = this->priv_bucket_pointer(); siterator before_first_local_it = this->priv_get_previous(buck_ptr[first_bucket_num], first_local_it); size_type last_bucket_num; siterator last_local_it; //For the end iterator, we will assign the end iterator //of the last bucket if(e == this->end()){ last_bucket_num = this->bucket_count() - 1; last_local_it = buck_ptr[last_bucket_num].end(); } else{ last_local_it = e.slist_it(); last_bucket_num = this->priv_get_bucket_num(last_local_it); } size_type const num_erased = this->priv_erase_node_range ( before_first_local_it, first_bucket_num, last_local_it, last_bucket_num , make_node_disposer(disposer), optimize_multikey_t()); this->priv_size_traits().set_size(this->priv_size_traits().get_size()-num_erased); this->priv_erasure_update_cache_range(first_bucket_num, last_bucket_num); } } //! <b>Requires</b>: Disposer::operator()(pointer) shouldn't throw. //! //! <b>Effects</b>: Erases all the elements with the given value. //! Disposer::operator()(pointer) is called for the removed elements. //! //! <b>Returns</b>: The number of erased elements. //! //! <b>Complexity</b>: Average case O(this->count(value)). //! Worst case O(this->size()). //! //! <b>Throws</b>: If the internal hasher or the equality functor throws. //! Basic guarantee. //! //! <b>Note</b>: Invalidates the iterators (but not the references) //! to the erased elements. No destructors are called. template<class Disposer> BOOST_INTRUSIVE_FORCEINLINE size_type erase_and_dispose(const key_type &key, Disposer disposer) { return this->erase_and_dispose(key, this->priv_hasher(), this->priv_equal(), disposer); } //! <b>Requires</b>: Disposer::operator()(pointer) shouldn't throw. //! //! <b>Effects</b>: Erases all the elements with the given key. //! according to the comparison functor "equal_func". //! Disposer::operator()(pointer) is called for the removed elements. //! //! <b>Returns</b>: The number of erased elements. //! //! <b>Complexity</b>: Average case O(this->count(value)). //! Worst case O(this->size()). //! //! <b>Throws</b>: If hash_func or equal_func throw. Basic guarantee. //! //! <b>Note</b>: Invalidates the iterators //! to the erased elements. template<class KeyType, class KeyHasher, class KeyEqual, class Disposer> size_type erase_and_dispose(const KeyType& key, KeyHasher hash_func ,KeyEqual equal_func, Disposer disposer) { size_type bucket_num; std::size_t h; siterator prev; siterator it = this->priv_find(key, hash_func, equal_func, bucket_num, h, prev); bool const success = it != this->priv_invalid_local_it(); size_type cnt(0); if(success){ if(optimize_multikey){ cnt = this->priv_erase_from_single_bucket (this->priv_bucket_pointer()[bucket_num], prev, ++(priv_last_in_group)(it), make_node_disposer(disposer), optimize_multikey_t()); } else{ bucket_type &b = this->priv_bucket_pointer()[bucket_num]; siterator const end_sit = b.end(); do{ ++cnt; ++it; }while(it != end_sit && this->priv_is_value_equal_to_key (this->priv_value_from_slist_node(it.pointed_node()), h, key, equal_func)); bucket_type::s_erase_after_and_dispose(prev, it, make_node_disposer(disposer)); } this->priv_size_traits().set_size(this->priv_size_traits().get_size()-cnt); this->priv_erasure_update_cache(); } return cnt; } //! <b>Effects</b>: Erases all of the elements. //! //! <b>Complexity</b>: Linear to the number of elements on the container. //! if it's a safe-mode or auto-unlink value_type. Constant time otherwise. //! //! <b>Throws</b>: Nothing. //! //! <b>Note</b>: Invalidates the iterators (but not the references) //! to the erased elements. No destructors are called. void clear() { this->priv_clear_buckets_and_cache(); this->priv_size_traits().set_size(size_type(0)); } //! <b>Requires</b>: Disposer::operator()(pointer) shouldn't throw. //! //! <b>Effects</b>: Erases all of the elements. //! //! <b>Complexity</b>: Linear to the number of elements on the container. //! Disposer::operator()(pointer) is called for the removed elements. //! //! <b>Throws</b>: Nothing. //! //! <b>Note</b>: Invalidates the iterators (but not the references) //! to the erased elements. No destructors are called. template<class Disposer> void clear_and_dispose(Disposer disposer) { if(!constant_time_size || !this->empty()){ size_type num_buckets = this->bucket_count(); bucket_ptr b = this->priv_bucket_pointer(); typename typeof_node_disposer<Disposer>::type d(disposer, &this->priv_value_traits()); for(; num_buckets--; ++b){ b->clear_and_dispose(d); } this->priv_size_traits().set_size(size_type(0)); } this->priv_initialize_cache(); } //! <b>Effects</b>: Returns the number of contained elements with the given value //! //! <b>Complexity</b>: Average case O(1), worst case O(this->size()). //! //! <b>Throws</b>: If the internal hasher or the equality functor throws. BOOST_INTRUSIVE_FORCEINLINE size_type count(const key_type &key) const { return this->count(key, this->priv_hasher(), this->priv_equal()); } //! <b>Requires</b>: "hash_func" must be a hash function that induces //! the same hash values as the stored hasher. The difference is that //! "hash_func" hashes the given key instead of the value_type. //! //! "equal_func" must be a equality function that induces //! the same equality as key_equal. The difference is that //! "equal_func" compares an arbitrary key with the contained values. //! //! <b>Effects</b>: Returns the number of contained elements with the given key //! //! <b>Complexity</b>: Average case O(1), worst case O(this->size()). //! //! <b>Throws</b>: If hash_func or equal throw. template<class KeyType, class KeyHasher, class KeyEqual> size_type count(const KeyType &key, KeyHasher hash_func, KeyEqual equal_func) const { size_type cnt; size_type n_bucket; this->priv_local_equal_range(key, hash_func, equal_func, n_bucket, cnt); return cnt; } //! <b>Effects</b>: Finds an iterator to the first element is equal to //! "value" or end() if that element does not exist. //! //! <b>Complexity</b>: Average case O(1), worst case O(this->size()). //! //! <b>Throws</b>: If the internal hasher or the equality functor throws. BOOST_INTRUSIVE_FORCEINLINE iterator find(const key_type &key) { return this->find(key, this->priv_hasher(), this->priv_equal()); } //! <b>Requires</b>: "hash_func" must be a hash function that induces //! the same hash values as the stored hasher. The difference is that //! "hash_func" hashes the given key instead of the value_type. //! //! "equal_func" must be a equality function that induces //! the same equality as key_equal. The difference is that //! "equal_func" compares an arbitrary key with the contained values. //! //! <b>Effects</b>: Finds an iterator to the first element whose key is //! "key" according to the given hash and equality functor or end() if //! that element does not exist. //! //! <b>Complexity</b>: Average case O(1), worst case O(this->size()). //! //! <b>Throws</b>: If hash_func or equal_func throw. //! //! <b>Note</b>: This function is used when constructing a value_type //! is expensive and the value_type can be compared with a cheaper //! key type. Usually this key is part of the value_type. template<class KeyType, class KeyHasher, class KeyEqual> iterator find(const KeyType &key, KeyHasher hash_func, KeyEqual equal_func) { size_type bucket_n; std::size_t hash; siterator prev; return iterator( this->priv_find(key, hash_func, equal_func, bucket_n, hash, prev) , &this->get_bucket_value_traits()); } //! <b>Effects</b>: Finds a const_iterator to the first element whose key is //! "key" or end() if that element does not exist. //! //! <b>Complexity</b>: Average case O(1), worst case O(this->size()). //! //! <b>Throws</b>: If the internal hasher or the equality functor throws. BOOST_INTRUSIVE_FORCEINLINE const_iterator find(const key_type &key) const { return this->find(key, this->priv_hasher(), this->priv_equal()); } //! <b>Requires</b>: "hash_func" must be a hash function that induces //! the same hash values as the stored hasher. The difference is that //! "hash_func" hashes the given key instead of the value_type. //! //! "equal_func" must be a equality function that induces //! the same equality as key_equal. The difference is that //! "equal_func" compares an arbitrary key with the contained values. //! //! <b>Effects</b>: Finds an iterator to the first element whose key is //! "key" according to the given hasher and equality functor or end() if //! that element does not exist. //! //! <b>Complexity</b>: Average case O(1), worst case O(this->size()). //! //! <b>Throws</b>: If hash_func or equal_func throw. //! //! <b>Note</b>: This function is used when constructing a value_type //! is expensive and the value_type can be compared with a cheaper //! key type. Usually this key is part of the value_type. template<class KeyType, class KeyHasher, class KeyEqual> const_iterator find (const KeyType &key, KeyHasher hash_func, KeyEqual equal_func) const { size_type bucket_n; std::size_t hash_value; siterator prev; return const_iterator( this->priv_find(key, hash_func, equal_func, bucket_n, hash_value, prev) , &this->get_bucket_value_traits()); } //! <b>Effects</b>: Returns a range containing all elements with values equivalent //! to value. Returns std::make_pair(this->end(), this->end()) if no such //! elements exist. //! //! <b>Complexity</b>: Average case O(this->count(value)). Worst case O(this->size()). //! //! <b>Throws</b>: If the internal hasher or the equality functor throws. BOOST_INTRUSIVE_FORCEINLINE std::pair<iterator,iterator> equal_range(const key_type &key) { return this->equal_range(key, this->priv_hasher(), this->priv_equal()); } //! <b>Requires</b>: "hash_func" must be a hash function that induces //! the same hash values as the stored hasher. The difference is that //! "hash_func" hashes the given key instead of the value_type. //! //! "equal_func" must be a equality function that induces //! the same equality as key_equal. The difference is that //! "equal_func" compares an arbitrary key with the contained values. //! //! <b>Effects</b>: Returns a range containing all elements with equivalent //! keys. Returns std::make_pair(this->end(), this->end()) if no such //! elements exist. //! //! <b>Complexity</b>: Average case O(this->count(key, hash_func, equal_func)). //! Worst case O(this->size()). //! //! <b>Throws</b>: If hash_func or the equal_func throw. //! //! <b>Note</b>: This function is used when constructing a value_type //! is expensive and the value_type can be compared with a cheaper //! key type. Usually this key is part of the value_type. template<class KeyType, class KeyHasher, class KeyEqual> std::pair<iterator,iterator> equal_range (const KeyType &key, KeyHasher hash_func, KeyEqual equal_func) { std::pair<siterator, siterator> ret = this->priv_equal_range(key, hash_func, equal_func); return std::pair<iterator, iterator> ( iterator(ret.first, &this->get_bucket_value_traits()) , iterator(ret.second, &this->get_bucket_value_traits())); } //! <b>Effects</b>: Returns a range containing all elements with values equivalent //! to value. Returns std::make_pair(this->end(), this->end()) if no such //! elements exist. //! //! <b>Complexity</b>: Average case O(this->count(value)). Worst case O(this->size()). //! //! <b>Throws</b>: If the internal hasher or the equality functor throws. BOOST_INTRUSIVE_FORCEINLINE std::pair<const_iterator, const_iterator> equal_range(const key_type &key) const { return this->equal_range(key, this->priv_hasher(), this->priv_equal()); } //! <b>Requires</b>: "hash_func" must be a hash function that induces //! the same hash values as the stored hasher. The difference is that //! "hash_func" hashes the given key instead of the value_type. //! //! "equal_func" must be a equality function that induces //! the same equality as key_equal. The difference is that //! "equal_func" compares an arbitrary key with the contained values. //! //! <b>Effects</b>: Returns a range containing all elements with equivalent //! keys. Returns std::make_pair(this->end(), this->end()) if no such //! elements exist. //! //! <b>Complexity</b>: Average case O(this->count(key, hash_func, equal_func)). //! Worst case O(this->size()). //! //! <b>Throws</b>: If the hasher or equal_func throw. //! //! <b>Note</b>: This function is used when constructing a value_type //! is expensive and the value_type can be compared with a cheaper //! key type. Usually this key is part of the value_type. template<class KeyType, class KeyHasher, class KeyEqual> std::pair<const_iterator,const_iterator> equal_range (const KeyType &key, KeyHasher hash_func, KeyEqual equal_func) const { std::pair<siterator, siterator> ret = this->priv_equal_range(key, hash_func, equal_func); return std::pair<const_iterator, const_iterator> ( const_iterator(ret.first, &this->get_bucket_value_traits()) , const_iterator(ret.second, &this->get_bucket_value_traits())); } #if defined(BOOST_INTRUSIVE_DOXYGEN_INVOKED) //! <b>Requires</b>: value must be an lvalue and shall be in a unordered_set of //! appropriate type. Otherwise the behavior is undefined. //! //! <b>Effects</b>: Returns: a valid iterator belonging to the unordered_set //! that points to the value //! //! <b>Complexity</b>: Constant. //! //! <b>Throws</b>: If the internal hash function throws. iterator iterator_to(reference value); //! <b>Requires</b>: value must be an lvalue and shall be in a unordered_set of //! appropriate type. Otherwise the behavior is undefined. //! //! <b>Effects</b>: Returns: a valid const_iterator belonging to the //! unordered_set that points to the value //! //! <b>Complexity</b>: Constant. //! //! <b>Throws</b>: If the internal hash function throws. const_iterator iterator_to(const_reference value) const; //! <b>Requires</b>: value must be an lvalue and shall be in a unordered_set of //! appropriate type. Otherwise the behavior is undefined. //! //! <b>Effects</b>: Returns: a valid local_iterator belonging to the unordered_set //! that points to the value //! //! <b>Complexity</b>: Constant. //! //! <b>Throws</b>: Nothing. //! //! <b>Note</b>: This static function is available only if the <i>value traits</i> //! is stateless. static local_iterator s_local_iterator_to(reference value); //! <b>Requires</b>: value must be an lvalue and shall be in a unordered_set of //! appropriate type. Otherwise the behavior is undefined. //! //! <b>Effects</b>: Returns: a valid const_local_iterator belonging to //! the unordered_set that points to the value //! //! <b>Complexity</b>: Constant. //! //! <b>Throws</b>: Nothing. //! //! <b>Note</b>: This static function is available only if the <i>value traits</i> //! is stateless. static const_local_iterator s_local_iterator_to(const_reference value); //! <b>Requires</b>: value must be an lvalue and shall be in a unordered_set of //! appropriate type. Otherwise the behavior is undefined. //! //! <b>Effects</b>: Returns: a valid local_iterator belonging to the unordered_set //! that points to the value //! //! <b>Complexity</b>: Constant. //! //! <b>Throws</b>: Nothing. local_iterator local_iterator_to(reference value); //! <b>Requires</b>: value must be an lvalue and shall be in a unordered_set of //! appropriate type. Otherwise the behavior is undefined. //! //! <b>Effects</b>: Returns: a valid const_local_iterator belonging to //! the unordered_set that points to the value //! //! <b>Complexity</b>: Constant. //! //! <b>Throws</b>: Nothing. const_local_iterator local_iterator_to(const_reference value) const; //! <b>Effects</b>: Returns the number of buckets passed in the constructor //! or the last rehash function. //! //! <b>Complexity</b>: Constant. //! //! <b>Throws</b>: Nothing. size_type bucket_count() const; //! <b>Requires</b>: n is in the range [0, this->bucket_count()). //! //! <b>Effects</b>: Returns the number of elements in the nth bucket. //! //! <b>Complexity</b>: Constant. //! //! <b>Throws</b>: Nothing. size_type bucket_size(size_type n) const; #endif //#if defined(BOOST_INTRUSIVE_DOXYGEN_INVOKED) //! <b>Effects</b>: Returns the index of the bucket in which elements //! with keys equivalent to k would be found, if any such element existed. //! //! <b>Complexity</b>: Constant. //! //! <b>Throws</b>: If the hash functor throws. //! //! <b>Note</b>: the return value is in the range [0, this->bucket_count()). BOOST_INTRUSIVE_FORCEINLINE size_type bucket(const key_type& k) const { return this->bucket(k, this->priv_hasher()); } //! <b>Requires</b>: "hash_func" must be a hash function that induces //! the same hash values as the stored hasher. The difference is that //! "hash_func" hashes the given key instead of the value_type. //! //! <b>Effects</b>: Returns the index of the bucket in which elements //! with keys equivalent to k would be found, if any such element existed. //! //! <b>Complexity</b>: Constant. //! //! <b>Throws</b>: If hash_func throws. //! //! <b>Note</b>: the return value is in the range [0, this->bucket_count()). template<class KeyType, class KeyHasher> BOOST_INTRUSIVE_FORCEINLINE size_type bucket(const KeyType& k, KeyHasher hash_func) const { return this->priv_hash_to_bucket(hash_func(k)); } #if defined(BOOST_INTRUSIVE_DOXYGEN_INVOKED) //! <b>Effects</b>: Returns the bucket array pointer passed in the constructor //! or the last rehash function. //! //! <b>Complexity</b>: Constant. //! //! <b>Throws</b>: Nothing. bucket_ptr bucket_pointer() const; //! <b>Requires</b>: n is in the range [0, this->bucket_count()). //! //! <b>Effects</b>: Returns a local_iterator pointing to the beginning //! of the sequence stored in the bucket n. //! //! <b>Complexity</b>: Constant. //! //! <b>Throws</b>: Nothing. //! //! <b>Note</b>: [this->begin(n), this->end(n)) is a valid range //! containing all of the elements in the nth bucket. local_iterator begin(size_type n); //! <b>Requires</b>: n is in the range [0, this->bucket_count()). //! //! <b>Effects</b>: Returns a const_local_iterator pointing to the beginning //! of the sequence stored in the bucket n. //! //! <b>Complexity</b>: Constant. //! //! <b>Throws</b>: Nothing. //! //! <b>Note</b>: [this->begin(n), this->end(n)) is a valid range //! containing all of the elements in the nth bucket. const_local_iterator begin(size_type n) const; //! <b>Requires</b>: n is in the range [0, this->bucket_count()). //! //! <b>Effects</b>: Returns a const_local_iterator pointing to the beginning //! of the sequence stored in the bucket n. //! //! <b>Complexity</b>: Constant. //! //! <b>Throws</b>: Nothing. //! //! <b>Note</b>: [this->begin(n), this->end(n)) is a valid range //! containing all of the elements in the nth bucket. const_local_iterator cbegin(size_type n) const; //! <b>Requires</b>: n is in the range [0, this->bucket_count()). //! //! <b>Effects</b>: Returns a local_iterator pointing to the end //! of the sequence stored in the bucket n. //! //! <b>Complexity</b>: Constant. //! //! <b>Throws</b>: Nothing. //! //! <b>Note</b>: [this->begin(n), this->end(n)) is a valid range //! containing all of the elements in the nth bucket. local_iterator end(size_type n); //! <b>Requires</b>: n is in the range [0, this->bucket_count()). //! //! <b>Effects</b>: Returns a const_local_iterator pointing to the end //! of the sequence stored in the bucket n. //! //! <b>Complexity</b>: Constant. //! //! <b>Throws</b>: Nothing. //! //! <b>Note</b>: [this->begin(n), this->end(n)) is a valid range //! containing all of the elements in the nth bucket. const_local_iterator end(size_type n) const; //! <b>Requires</b>: n is in the range [0, this->bucket_count()). //! //! <b>Effects</b>: Returns a const_local_iterator pointing to the end //! of the sequence stored in the bucket n. //! //! <b>Complexity</b>: Constant. //! //! <b>Throws</b>: Nothing. //! //! <b>Note</b>: [this->begin(n), this->end(n)) is a valid range //! containing all of the elements in the nth bucket. const_local_iterator cend(size_type n) const; #endif //#if defined(BOOST_INTRUSIVE_DOXYGEN_INVOKED) //! <b>Requires</b>: new_bucket_traits can hold a pointer to a new bucket array //! or the same as the old bucket array with a different length. new_size is the length of the //! the array pointed by new_buckets. If new_bucket_traits.bucket_begin() == this->bucket_pointer() //! new_bucket_traits.bucket_count() can be bigger or smaller than this->bucket_count(). //! 'new_bucket_traits' copy constructor should not throw. //! //! <b>Effects</b>: //! If `new_bucket_traits.bucket_begin() == this->bucket_pointer()` is false, //! unlinks values from the old bucket and inserts then in the new one according //! to the hash value of values. //! //! If `new_bucket_traits.bucket_begin() == this->bucket_pointer()` is true, //! the implementations avoids moving values as much as possible. //! //! Bucket traits hold by *this is assigned from new_bucket_traits. //! If the container is configured as incremental<>, the split bucket is set //! to the new bucket_count(). //! //! If store_hash option is true, this method does not use the hash function. //! If false, the implementation tries to minimize calls to the hash function //! (e.g. once for equivalent values if optimize_multikey<true> is true). //! //! If rehash is successful updates the internal bucket_traits with new_bucket_traits. //! //! <b>Complexity</b>: Average case linear in this->size(), worst case quadratic. //! //! <b>Throws</b>: If the hasher functor throws. Basic guarantee. BOOST_INTRUSIVE_FORCEINLINE void rehash(const bucket_traits &new_bucket_traits) { this->rehash_impl(new_bucket_traits, false); } //! <b>Note</b>: This function is used when keys from inserted elements are changed //! (e.g. a language change when key is a string) but uniqueness and hash properties are //! preserved so a fast full rehash recovers invariants for *this without extracting and //! reinserting all elements again. //! //! <b>Requires</b>: Calls produced to the hash function should not alter the value uniqueness //! properties of already inserted elements. If hasher(key1) == hasher(key2) was true when //! elements were inserted, it shall be true during calls produced in the execution of this function. //! //! key_equal is not called inside this function so it is assumed that key_equal(value1, value2) //! should produce the same results as before for inserted elements. //! //! <b>Effects</b>: Reprocesses all values hold by *this, recalculating their hash values //! and redistributing them though the buckets. //! //! If store_hash option is true, this method uses the hash function and updates the stored hash value. //! //! <b>Complexity</b>: Average case linear in this->size(), worst case quadratic. //! //! <b>Throws</b>: If the hasher functor throws. Basic guarantee. BOOST_INTRUSIVE_FORCEINLINE void full_rehash() { this->rehash_impl(this->priv_bucket_traits(), true); } //! <b>Requires</b>: //! //! <b>Effects</b>: //! //! <b>Complexity</b>: //! //! <b>Throws</b>: //! //! <b>Note</b>: this method is only available if incremental<true> option is activated. bool incremental_rehash(bool grow = true) { //This function is only available for containers with incremental hashing BOOST_STATIC_ASSERT(( incremental && power_2_buckets )); const size_type split_idx = this->priv_split_traits().get_size(); const size_type bucket_cnt = this->bucket_count(); const bucket_ptr buck_ptr = this->priv_bucket_pointer(); bool ret = false; if(grow){ //Test if the split variable can be changed if((ret = split_idx < bucket_cnt)){ const size_type bucket_to_rehash = split_idx - bucket_cnt/2; bucket_type &old_bucket = buck_ptr[bucket_to_rehash]; this->priv_split_traits().increment(); //Anti-exception stuff: if an exception is thrown while //moving elements from old_bucket to the target bucket, all moved //elements are moved back to the original one. detail::incremental_rehash_rollback<bucket_type, split_traits> rollback ( buck_ptr[split_idx], old_bucket, this->priv_split_traits()); for( siterator before_i(old_bucket.before_begin()), i(old_bucket.begin()), end_sit(old_bucket.end()) ; i != end_sit; i = before_i, ++i){ const value_type &v = this->priv_value_from_slist_node(i.pointed_node()); const std::size_t hash_value = this->priv_stored_or_compute_hash(v, store_hash_t()); const size_type new_n = this->priv_hash_to_bucket(hash_value); siterator const last = (priv_last_in_group)(i); if(new_n == bucket_to_rehash){ before_i = last; } else{ bucket_type &new_b = buck_ptr[new_n]; new_b.splice_after(new_b.before_begin(), old_bucket, before_i, last); } } rollback.release(); this->priv_erasure_update_cache(); } } else if((ret = split_idx > bucket_cnt/2)){ //!grow const size_type target_bucket_num = split_idx - 1 - bucket_cnt/2; bucket_type &target_bucket = buck_ptr[target_bucket_num]; bucket_type &source_bucket = buck_ptr[split_idx-1]; target_bucket.splice_after(target_bucket.cbefore_begin(), source_bucket); this->priv_split_traits().decrement(); this->priv_insertion_update_cache(target_bucket_num); } return ret; } //! <b>Effects</b>: If new_bucket_traits.bucket_count() is not //! this->bucket_count()/2 or this->bucket_count()*2, or //! this->split_bucket() != new_bucket_traits.bucket_count() returns false //! and does nothing. //! //! Otherwise, copy assigns new_bucket_traits to the internal bucket_traits //! and transfers all the objects from old buckets to the new ones. //! //! <b>Complexity</b>: Linear to size(). //! //! <b>Throws</b>: Nothing //! //! <b>Note</b>: this method is only available if incremental<true> option is activated. bool incremental_rehash(const bucket_traits &new_bucket_traits) { //This function is only available for containers with incremental hashing BOOST_STATIC_ASSERT(( incremental && power_2_buckets )); size_type const new_bucket_traits_size = new_bucket_traits.bucket_count(); size_type const cur_bucket_traits = this->bucket_count(); const size_type split_idx = this->split_count(); //Test new bucket size is consistent with internal bucket size and split count if(new_bucket_traits_size/2 == cur_bucket_traits){ if(!(split_idx >= cur_bucket_traits)) return false; } else if(new_bucket_traits_size == cur_bucket_traits/2){ if(!(split_idx <= new_bucket_traits_size)) return false; } else{ return false; } const size_type ini_n = this->priv_get_cache_bucket_num(); const bucket_ptr old_buckets = this->priv_bucket_pointer(); this->priv_bucket_traits() = new_bucket_traits; if(new_bucket_traits.bucket_begin() != old_buckets){ for(size_type n = ini_n; n < split_idx; ++n){ bucket_type &new_bucket = new_bucket_traits.bucket_begin()[n]; bucket_type &old_bucket = old_buckets[n]; new_bucket.splice_after(new_bucket.cbefore_begin(), old_bucket); } //Put cache to safe position this->priv_initialize_cache(); this->priv_insertion_update_cache(ini_n); } return true; } #if defined(BOOST_INTRUSIVE_DOXYGEN_INVOKED) //! <b>Requires</b>: incremental<> option must be set //! //! <b>Effects</b>: returns the current split count //! //! <b>Complexity</b>: Constant //! //! <b>Throws</b>: Nothing size_type split_count() const; //! <b>Effects</b>: Returns the nearest new bucket count optimized for //! the container that is bigger or equal than n. This suggestion can be //! used to create bucket arrays with a size that will usually improve //! container's performance. If such value does not exist, the //! higher possible value is returned. //! //! <b>Complexity</b>: Amortized constant time. //! //! <b>Throws</b>: Nothing. static size_type suggested_upper_bucket_count(size_type n); //! <b>Effects</b>: Returns the nearest new bucket count optimized for //! the container that is smaller or equal than n. This suggestion can be //! used to create bucket arrays with a size that will usually improve //! container's performance. If such value does not exist, the //! lowest possible value is returned. //! //! <b>Complexity</b>: Amortized constant time. //! //! <b>Throws</b>: Nothing. static size_type suggested_lower_bucket_count(size_type n); #endif //#if defined(BOOST_INTRUSIVE_DOXYGEN_INVOKED) friend bool operator==(const hashtable_impl &x, const hashtable_impl &y) { //Taken from N3068 if(constant_time_size && x.size() != y.size()){ return false; } for (const_iterator ix = x.cbegin(), ex = x.cend(); ix != ex; ++ix){ std::pair<const_iterator, const_iterator> eqx(x.equal_range(key_of_value()(*ix))), eqy(y.equal_range(key_of_value()(*ix))); if (boost::intrusive::iterator_distance(eqx.first, eqx.second) != boost::intrusive::iterator_distance(eqy.first, eqy.second) || !(priv_algo_is_permutation)(eqx.first, eqx.second, eqy.first) ){ return false; } ix = eqx.second; } return true; } friend bool operator!=(const hashtable_impl &x, const hashtable_impl &y) { return !(x == y); } friend bool operator<(const hashtable_impl &x, const hashtable_impl &y) { return ::boost::intrusive::algo_lexicographical_compare(x.begin(), x.end(), y.begin(), y.end()); } friend bool operator>(const hashtable_impl &x, const hashtable_impl &y) { return y < x; } friend bool operator<=(const hashtable_impl &x, const hashtable_impl &y) { return !(y < x); } friend bool operator>=(const hashtable_impl &x, const hashtable_impl &y) { return !(x < y); } /// @cond BOOST_INTRUSIVE_FORCEINLINE void check() const {} private: void rehash_impl(const bucket_traits &new_bucket_traits, bool do_full_rehash) { const bucket_ptr new_buckets = new_bucket_traits.bucket_begin(); size_type new_bucket_count = new_bucket_traits.bucket_count(); const bucket_ptr old_buckets = this->priv_bucket_pointer(); size_type old_bucket_count = this->bucket_count(); //Check power of two bucket array if the option is activated BOOST_INTRUSIVE_INVARIANT_ASSERT (!power_2_buckets || (0 == (new_bucket_count & (new_bucket_count-1u)))); size_type n = this->priv_get_cache_bucket_num(); const bool same_buffer = old_buckets == new_buckets; //If the new bucket length is a common factor //of the old one we can avoid hash calculations. const bool fast_shrink = (!do_full_rehash) && (!incremental) && (old_bucket_count >= new_bucket_count) && (power_2_buckets || (old_bucket_count % new_bucket_count) == 0); //If we are shrinking the same bucket array and it's //is a fast shrink, just rehash the last nodes size_type new_first_bucket_num = new_bucket_count; if(same_buffer && fast_shrink && (n < new_bucket_count)){ new_first_bucket_num = n; n = new_bucket_count; } //Anti-exception stuff: they destroy the elements if something goes wrong. //If the source and destination buckets are the same, the second rollback function //is harmless, because all elements have been already unlinked and destroyed typedef detail::init_disposer<node_algorithms> NodeDisposer; typedef detail::exception_array_disposer<bucket_type, NodeDisposer, size_type> ArrayDisposer; NodeDisposer node_disp; ArrayDisposer rollback1(new_buckets[0], node_disp, new_bucket_count); ArrayDisposer rollback2(old_buckets[0], node_disp, old_bucket_count); //Put size in a safe value for rollback exception size_type const size_backup = this->priv_size_traits().get_size(); this->priv_size_traits().set_size(0); //Put cache to safe position this->priv_initialize_cache(); this->priv_insertion_update_cache(size_type(0u)); //Iterate through nodes for(; n < old_bucket_count; ++n){ bucket_type &old_bucket = old_buckets[n]; if(!fast_shrink){ for( siterator before_i(old_bucket.before_begin()), i(old_bucket.begin()), end_sit(old_bucket.end()) ; i != end_sit ; i = before_i, ++i){ //First obtain hash value (and store it if do_full_rehash) std::size_t hash_value; if(do_full_rehash){ value_type &v = this->priv_value_from_slist_node(i.pointed_node()); hash_value = this->priv_hasher()(key_of_value()(v)); node_functions_t::store_hash(pointer_traits<node_ptr>::pointer_to(this->priv_value_to_node(v)), hash_value, store_hash_t()); } else{ const value_type &v = this->priv_value_from_slist_node(i.pointed_node()); hash_value = this->priv_stored_or_compute_hash(v, store_hash_t()); } //Now calculate the new bucket position const size_type new_n = detail::hash_to_bucket_split<power_2_buckets, incremental> (hash_value, new_bucket_count, new_bucket_count); //Update first used bucket cache if(cache_begin && new_n < new_first_bucket_num) new_first_bucket_num = new_n; //If the target bucket is new, transfer the whole group siterator const last = (priv_last_in_group)(i); if(same_buffer && new_n == n){ before_i = last; } else{ bucket_type &new_b = new_buckets[new_n]; new_b.splice_after(new_b.before_begin(), old_bucket, before_i, last); } } } else{ const size_type new_n = detail::hash_to_bucket_split<power_2_buckets, incremental>(n, new_bucket_count, new_bucket_count); if(cache_begin && new_n < new_first_bucket_num) new_first_bucket_num = new_n; bucket_type &new_b = new_buckets[new_n]; new_b.splice_after( new_b.before_begin() , old_bucket , old_bucket.before_begin() , bucket_plus_vtraits_t::priv_get_last(old_bucket, optimize_multikey_t())); } } this->priv_size_traits().set_size(size_backup); this->priv_split_traits().set_size(new_bucket_count); if(&new_bucket_traits != &this->priv_bucket_traits()){ this->priv_bucket_traits() = new_bucket_traits; } this->priv_initialize_cache(); this->priv_insertion_update_cache(new_first_bucket_num); rollback1.release(); rollback2.release(); } template <class MaybeConstHashtableImpl, class Cloner, class Disposer> void priv_clone_from(MaybeConstHashtableImpl &src, Cloner cloner, Disposer disposer) { this->clear_and_dispose(disposer); if(!constant_time_size || !src.empty()){ const size_type src_bucket_count = src.bucket_count(); const size_type dst_bucket_count = this->bucket_count(); //Check power of two bucket array if the option is activated BOOST_INTRUSIVE_INVARIANT_ASSERT (!power_2_buckets || (0 == (src_bucket_count & (src_bucket_count-1)))); BOOST_INTRUSIVE_INVARIANT_ASSERT (!power_2_buckets || (0 == (dst_bucket_count & (dst_bucket_count-1)))); //If src bucket count is bigger or equal, structural copy is possible const bool structural_copy = (!incremental) && (src_bucket_count >= dst_bucket_count) && (power_2_buckets || (src_bucket_count % dst_bucket_count) == 0); if(structural_copy){ this->priv_structural_clone_from(src, cloner, disposer); } else{ //Unlike previous cloning algorithm, this can throw //if cloner, hasher or comparison functor throw typedef typename detail::if_c< detail::is_const<MaybeConstHashtableImpl>::value , typename MaybeConstHashtableImpl::const_iterator , typename MaybeConstHashtableImpl::iterator >::type clone_iterator; clone_iterator b(src.begin()), e(src.end()); detail::exception_disposer<hashtable_impl, Disposer> rollback(*this, disposer); for(; b != e; ++b){ //No need to check for duplicates and insert it in the first position //as this is an unordered container. So use minimal insertion code std::size_t const hash_to_store = this->priv_stored_or_compute_hash(*b, store_hash_t());; size_type const bucket_number = this->priv_hash_to_bucket(hash_to_store); typedef typename detail::if_c <detail::is_const<MaybeConstHashtableImpl>::value, const_reference, reference>::type reference_type; reference_type r = *b; this->priv_clone_front_in_bucket<reference_type>(bucket_number, r, hash_to_store, cloner); } rollback.release(); } } } template<class ValueReference, class Cloner> void priv_clone_front_in_bucket( size_type const bucket_number , typename detail::identity<ValueReference>::type src_ref , std::size_t const hash_to_store, Cloner cloner) { //No need to check for duplicates and insert it in the first position //as this is an unordered container. So use minimal insertion code //std::size_t const hash_value = this->priv_stored_or_compute_hash(src_ref, store_hash_t());; //size_type const bucket_number = this->priv_hash_to_bucket(hash_value); bucket_type &cur_bucket = this->priv_bucket_pointer()[bucket_number]; siterator const prev(cur_bucket.before_begin()); //Just check if the cloned node is equal to the first inserted value in the new bucket //as equal src values were contiguous and they should be already inserted in the //destination bucket. bool const next_is_in_group = optimize_multikey && !cur_bucket.empty() && this->priv_equal()( key_of_value()(src_ref) , key_of_value()(this->priv_value_from_slist_node((++siterator(prev)).pointed_node()))); this->priv_insert_equal_after_find(*cloner(src_ref), bucket_number, hash_to_store, prev, next_is_in_group); } template <class MaybeConstHashtableImpl, class Cloner, class Disposer> void priv_structural_clone_from(MaybeConstHashtableImpl &src, Cloner cloner, Disposer disposer) { //First clone the first ones const size_type src_bucket_count = src.bucket_count(); const size_type dst_bucket_count = this->bucket_count(); const bucket_ptr src_buckets = src.priv_bucket_pointer(); const bucket_ptr dst_buckets = this->priv_bucket_pointer(); size_type constructed = 0; typedef node_cast_adaptor< detail::node_disposer<Disposer, value_traits, CircularSListAlgorithms> , slist_node_ptr, node_ptr > NodeDisposer; NodeDisposer node_disp(disposer, &this->priv_value_traits()); detail::exception_array_disposer<bucket_type, NodeDisposer, size_type> rollback(dst_buckets[0], node_disp, constructed); //Now insert the remaining ones using the modulo trick for( //"constructed" already initialized ; constructed < src_bucket_count ; ++constructed){ //Since incremental hashing can't be structurally copied, avoid hash_to_bucket_split const std::size_t new_n = detail::hash_to_bucket(constructed, dst_bucket_count, detail::bool_<power_2_buckets>()); bucket_type &src_b = src_buckets[constructed]; for( siterator b(src_b.begin()), e(src_b.end()); b != e; ++b){ slist_node_ptr const n(b.pointed_node()); typedef typename detail::if_c <detail::is_const<MaybeConstHashtableImpl>::value, const_reference, reference>::type reference_type; reference_type r = this->priv_value_from_slist_node(n); this->priv_clone_front_in_bucket<reference_type> (new_n, r, this->priv_stored_hash(n, store_hash_t()), cloner); } } this->priv_hasher() = src.priv_hasher(); this->priv_equal() = src.priv_equal(); rollback.release(); this->priv_size_traits().set_size(src.priv_size_traits().get_size()); this->priv_split_traits().set_size(dst_bucket_count); this->priv_insertion_update_cache(0u); this->priv_erasure_update_cache(); } std::size_t priv_hash_to_bucket(std::size_t hash_value) const { return detail::hash_to_bucket_split<power_2_buckets, incremental> (hash_value, this->priv_bucket_traits().bucket_count(), this->priv_split_traits().get_size()); } iterator priv_insert_equal_after_find(reference value, size_type bucket_num, std::size_t hash_value, siterator prev, bool const next_is_in_group) { //Now store hash if needed node_ptr n = pointer_traits<node_ptr>::pointer_to(this->priv_value_to_node(value)); node_functions_t::store_hash(n, hash_value, store_hash_t()); //Checks for some modes BOOST_INTRUSIVE_SAFE_HOOK_DEFAULT_ASSERT(!safemode_or_autounlink || node_algorithms::unique(n)); //Shortcut to optimize_multikey cases group_functions_t::insert_in_group ( next_is_in_group ? detail::dcast_bucket_ptr<node>((++siterator(prev)).pointed_node()) : n , n, optimize_multikey_t()); //Update cache and increment size if needed this->priv_insertion_update_cache(bucket_num); this->priv_size_traits().increment(); //Insert the element in the bucket after it return iterator(bucket_type::s_insert_after(prev, *n), &this->get_bucket_value_traits()); } template<class KeyType, class KeyHasher, class KeyEqual> siterator priv_find //In case it is not found previt is bucket.before_begin() ( const KeyType &key, KeyHasher hash_func , KeyEqual equal_func, size_type &bucket_number, std::size_t &h, siterator &previt) const { h = hash_func(key); return this->priv_find_with_hash(key, equal_func, bucket_number, h, previt); } template<class KeyType, class KeyEqual> bool priv_is_value_equal_to_key(const value_type &v, const std::size_t h, const KeyType &key, KeyEqual equal_func) const { (void)h; return (!compare_hash || this->priv_stored_or_compute_hash(v, store_hash_t()) == h) && equal_func(key, key_of_value()(v)); } //return previous iterator to the next equal range group in case static siterator priv_last_in_group(const siterator &it_first_in_group) { return bucket_type::s_iterator_to (*group_functions_t::get_last_in_group (detail::dcast_bucket_ptr<node>(it_first_in_group.pointed_node()), optimize_multikey_t())); } template<class KeyType, class KeyEqual> siterator priv_find_with_hash //In case it is not found previt is bucket.before_begin() ( const KeyType &key, KeyEqual equal_func, size_type &bucket_number, const std::size_t h, siterator &previt) const { bucket_number = this->priv_hash_to_bucket(h); bucket_type &b = this->priv_bucket_pointer()[bucket_number]; previt = b.before_begin(); siterator it = previt; siterator const endit = b.end(); while(++it != endit){ if(this->priv_is_value_equal_to_key(this->priv_value_from_slist_node(it.pointed_node()), h, key, equal_func)){ return it; } previt = it = (priv_last_in_group)(it); } previt = b.before_begin(); return this->priv_invalid_local_it(); } template<class KeyType, class KeyHasher, class KeyEqual> std::pair<siterator, siterator> priv_local_equal_range ( const KeyType &key , KeyHasher hash_func , KeyEqual equal_func , size_type &found_bucket , size_type &cnt) const { size_type internal_cnt = 0; //Let's see if the element is present siterator prev; size_type n_bucket; std::size_t h; std::pair<siterator, siterator> to_return ( this->priv_find(key, hash_func, equal_func, n_bucket, h, prev) , this->priv_invalid_local_it()); if(to_return.first != to_return.second){ found_bucket = n_bucket; //If it's present, find the first that it's not equal in //the same bucket bucket_type &b = this->priv_bucket_pointer()[n_bucket]; siterator it = to_return.first; ++internal_cnt; //At least one is found if(optimize_multikey){ to_return.second = ++(priv_last_in_group)(it); internal_cnt += boost::intrusive::iterator_distance(++it, to_return.second); } else{ siterator const bend = b.end(); while(++it != bend && this->priv_is_value_equal_to_key(this->priv_value_from_slist_node(it.pointed_node()), h, key, equal_func)){ ++internal_cnt; } to_return.second = it; } } cnt = internal_cnt; return to_return; } template<class KeyType, class KeyHasher, class KeyEqual> std::pair<siterator, siterator> priv_equal_range ( const KeyType &key , KeyHasher hash_func , KeyEqual equal_func) const { size_type n_bucket; size_type cnt; //Let's see if the element is present std::pair<siterator, siterator> to_return (this->priv_local_equal_range(key, hash_func, equal_func, n_bucket, cnt)); //If not, find the next element as ".second" if ".second" local iterator //is not pointing to an element. bucket_ptr const bp = this->priv_bucket_pointer(); if(to_return.first != to_return.second && to_return.second == bp[n_bucket].end()){ to_return.second = this->priv_invalid_local_it(); ++n_bucket; for( const size_type max_bucket = this->bucket_count() ; n_bucket != max_bucket ; ++n_bucket){ bucket_type &b = bp[n_bucket]; if(!b.empty()){ to_return.second = b.begin(); break; } } } return to_return; } std::size_t priv_get_bucket_num(siterator it) { return this->priv_get_bucket_num_hash_dispatch(it, store_hash_t()); } std::size_t priv_get_bucket_num_hash_dispatch(siterator it, detail::true_) //store_hash { return this->priv_hash_to_bucket (this->priv_stored_hash(it.pointed_node(), store_hash_t())); } std::size_t priv_get_bucket_num_hash_dispatch(siterator it, detail::false_) //NO store_hash { return this->priv_get_bucket_num_no_hash_store(it, optimize_multikey_t()); } static siterator priv_get_previous(bucket_type &b, siterator i) { return bucket_plus_vtraits_t::priv_get_previous(b, i, optimize_multikey_t()); } /// @endcond }; /// @cond template < class T , bool UniqueKeys , class PackedOptions > struct make_bucket_traits { //Real value traits must be calculated from options typedef typename detail::get_value_traits <T, typename PackedOptions::proto_value_traits>::type value_traits; typedef typename PackedOptions::bucket_traits specified_bucket_traits; //Real bucket traits must be calculated from options and calculated value_traits typedef typename get_slist_impl <typename reduced_slist_node_traits <typename value_traits::node_traits>::type >::type slist_impl; typedef typename detail::if_c< detail::is_same < specified_bucket_traits , default_bucket_traits >::value , bucket_traits_impl<slist_impl> , specified_bucket_traits >::type type; }; /// @endcond //! Helper metafunction to define a \c hashtable that yields to the same type when the //! same options (either explicitly or implicitly) are used. #if defined(BOOST_INTRUSIVE_DOXYGEN_INVOKED) || defined(BOOST_INTRUSIVE_VARIADIC_TEMPLATES) template<class T, class ...Options> #else template<class T, class O1 = void, class O2 = void , class O3 = void, class O4 = void , class O5 = void, class O6 = void , class O7 = void, class O8 = void , class O9 = void, class O10= void > #endif struct make_hashtable { /// @cond typedef typename pack_options < hashtable_defaults, #if !defined(BOOST_INTRUSIVE_VARIADIC_TEMPLATES) O1, O2, O3, O4, O5, O6, O7, O8, O9, O10 #else Options... #endif >::type packed_options; typedef typename detail::get_value_traits <T, typename packed_options::proto_value_traits>::type value_traits; typedef typename make_bucket_traits <T, false, packed_options>::type bucket_traits; typedef hashtable_impl < value_traits , typename packed_options::key_of_value , typename packed_options::hash , typename packed_options::equal , bucket_traits , typename packed_options::size_type , (std::size_t(false)*hash_bool_flags::unique_keys_pos) |(std::size_t(packed_options::constant_time_size)*hash_bool_flags::constant_time_size_pos) |(std::size_t(packed_options::power_2_buckets)*hash_bool_flags::power_2_buckets_pos) |(std::size_t(packed_options::cache_begin)*hash_bool_flags::cache_begin_pos) |(std::size_t(packed_options::compare_hash)*hash_bool_flags::compare_hash_pos) |(std::size_t(packed_options::incremental)*hash_bool_flags::incremental_pos) > implementation_defined; /// @endcond typedef implementation_defined type; }; #if !defined(BOOST_INTRUSIVE_DOXYGEN_INVOKED) #if defined(BOOST_INTRUSIVE_VARIADIC_TEMPLATES) template<class T, class ...Options> #else template<class T, class O1, class O2, class O3, class O4, class O5, class O6, class O7, class O8, class O9, class O10> #endif class hashtable : public make_hashtable<T, #if !defined(BOOST_INTRUSIVE_VARIADIC_TEMPLATES) O1, O2, O3, O4, O5, O6, O7, O8, O9, O10 #else Options... #endif >::type { typedef typename make_hashtable<T, #if !defined(BOOST_INTRUSIVE_VARIADIC_TEMPLATES) O1, O2, O3, O4, O5, O6, O7, O8, O9, O10 #else Options... #endif >::type Base; BOOST_MOVABLE_BUT_NOT_COPYABLE(hashtable) public: typedef typename Base::value_traits value_traits; typedef typename Base::iterator iterator; typedef typename Base::const_iterator const_iterator; typedef typename Base::bucket_ptr bucket_ptr; typedef typename Base::size_type size_type; typedef typename Base::hasher hasher; typedef typename Base::bucket_traits bucket_traits; typedef typename Base::key_equal key_equal; //Assert if passed value traits are compatible with the type BOOST_STATIC_ASSERT((detail::is_same<typename value_traits::value_type, T>::value)); BOOST_INTRUSIVE_FORCEINLINE explicit hashtable ( const bucket_traits &b_traits , const hasher & hash_func = hasher() , const key_equal &equal_func = key_equal() , const value_traits &v_traits = value_traits()) : Base(b_traits, hash_func, equal_func, v_traits) {} BOOST_INTRUSIVE_FORCEINLINE hashtable(BOOST_RV_REF(hashtable) x) : Base(BOOST_MOVE_BASE(Base, x)) {} BOOST_INTRUSIVE_FORCEINLINE hashtable& operator=(BOOST_RV_REF(hashtable) x) { return static_cast<hashtable&>(this->Base::operator=(BOOST_MOVE_BASE(Base, x))); } template <class Cloner, class Disposer> BOOST_INTRUSIVE_FORCEINLINE void clone_from(const hashtable &src, Cloner cloner, Disposer disposer) { Base::clone_from(src, cloner, disposer); } template <class Cloner, class Disposer> BOOST_INTRUSIVE_FORCEINLINE void clone_from(BOOST_RV_REF(hashtable) src, Cloner cloner, Disposer disposer) { Base::clone_from(BOOST_MOVE_BASE(Base, src), cloner, disposer); } }; #endif } //namespace intrusive } //namespace boost #include <boost/intrusive/detail/config_end.hpp> #endif //BOOST_INTRUSIVE_HASHTABLE_HPP