collector.hpp

Go to the documentation of this file.

#ifndef OSMIUM_RELATIONS_COLLECTOR_HPP
#define OSMIUM_RELATIONS_COLLECTOR_HPP
 
/*
 
This file is part of Osmium (https://osmcode.org/libosmium).
 
Copyright 2013-2023 Jochen Topf <jochen@topf.org> and others (see README).
 
Boost Software License - Version 1.0 - August 17th, 2003
 
Permission is hereby granted, free of charge, to any person or organization
obtaining a copy of the software and accompanying documentation covered by
this license (the "Software") to use, reproduce, display, distribute,
execute, and transmit the Software, and to prepare derivative works of the
Software, and to permit third-parties to whom the Software is furnished to
do so, all subject to the following:
 
The copyright notices in the Software and this entire statement, including
the above license grant, this restriction and the following disclaimer,
must be included in all copies of the Software, in whole or in part, and
all derivative works of the Software, unless such copies or derivative
works are solely in the form of machine-executable object code generated by
a source language processor.
 
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT
SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE
FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE,
ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
DEALINGS IN THE SOFTWARE.
 
*/
 
#include <osmium/handler.hpp>
#include <osmium/handler/check_order.hpp>
#include <osmium/memory/buffer.hpp>
#include <osmium/osm/item_type.hpp>
#include <osmium/osm/object.hpp>
#include <osmium/osm/relation.hpp>
#include <osmium/osm/types.hpp>
#include <osmium/relations/detail/member_meta.hpp>
#include <osmium/relations/detail/relation_meta.hpp>
#include <osmium/util/iterator.hpp>
#include <osmium/visitor.hpp>
 
#include <algorithm>
#include <array>
#include <cassert>
#include <cstddef>
#include <cstdint>
#include <functional>
#include <iomanip>
#include <iostream>
#include <utility>
#include <vector>
 
namespace osmium {
 
    class Node;
    class Way;
 
    namespace relations {
 
        template <typename TCollector, bool TNodes, bool TWays, bool TRelations>
        class Collector {
 
            class HandlerPass1 : public osmium::handler::Handler {
 
                TCollector& m_collector;
 
            public:
 
                explicit HandlerPass1(TCollector& collector) noexcept :
                    m_collector(collector) {
                }
 
                void relation(const osmium::Relation& relation) {
                    if (m_collector.keep_relation(relation)) {
                        m_collector.add_relation(relation);
                    }
                }
 
            }; // class HandlerPass1
 
        public:
 
            class HandlerPass2 : public osmium::handler::Handler {
 
                osmium::handler::CheckOrder m_check_order;
                TCollector& m_collector;
 
            public:
 
                explicit HandlerPass2(TCollector& collector) noexcept :
                    m_collector(collector) {
                }
 
                void node(const osmium::Node& node) {
                    if (TNodes) {
                        m_check_order.node(node);
                        if (!m_collector.find_and_add_object(node)) {
                            m_collector.node_not_in_any_relation(node);
                        }
                    }
                }
 
                void way(const osmium::Way& way) {
                    if (TWays) {
                        m_check_order.way(way);
                        if (!m_collector.find_and_add_object(way)) {
                            m_collector.way_not_in_any_relation(way);
                        }
                    }
                }
 
                void relation(const osmium::Relation& relation) {
                    if (TRelations) {
                        m_check_order.relation(relation);
                        if (!m_collector.find_and_add_object(relation)) {
                            m_collector.relation_not_in_any_relation(relation);
                        }
                    }
                }
 
                void flush() {
                    m_collector.flush();
                }
 
            }; // class HandlerPass2
 
        private:
 
            HandlerPass2 m_handler_pass2;
 
            // All relations we are interested in will be kept in this buffer
            osmium::memory::Buffer m_relations_buffer;
 
            // All members we are interested in will be kept in this buffer
            osmium::memory::Buffer m_members_buffer;
 
            std::vector<RelationMeta> m_relations;
 
            using mm_vector_type = std::vector<MemberMeta>;
            using mm_iterator = mm_vector_type::iterator;
            std::array<mm_vector_type, 3> m_member_meta;
 
            int m_count_complete = 0;
 
            using callback_func_type = std::function<void(osmium::memory::Buffer&&)>;
            callback_func_type m_callback;
 
            enum {
                initial_buffer_size = 1024UL * 1024UL
            };
 
            iterator_range<mm_iterator> find_member_meta(osmium::item_type type, osmium::object_id_type id) {
                auto& mmv = member_meta(type);
                return make_range(std::equal_range(mmv.begin(), mmv.end(), MemberMeta(id)));
            }
 
        public:
 
            Collector() :
                m_handler_pass2(*static_cast<TCollector*>(this)),
                m_relations_buffer(initial_buffer_size, osmium::memory::Buffer::auto_grow::yes),
                m_members_buffer(initial_buffer_size, osmium::memory::Buffer::auto_grow::yes) {
            }
 
        protected:
 
            std::vector<MemberMeta>& member_meta(const item_type type) {
                return m_member_meta[static_cast<uint16_t>(type) - 1];
            }
 
            callback_func_type callback() {
                return m_callback;
            }
 
            const std::vector<RelationMeta>& relations() const {
                return m_relations;
            }
 
            bool keep_relation(const osmium::Relation& /*relation*/) const {
                return true;
            }
 
            bool keep_member(const RelationMeta& /*relation_meta*/, const osmium::RelationMember& /*member*/) const {
                return true;
            }
 
            void node_not_in_any_relation(const osmium::Node& /*node*/) {
            }
 
            void way_not_in_any_relation(const osmium::Way& /*way*/) {
            }
 
            void relation_not_in_any_relation(const osmium::Relation& /*relation*/) {
            }
 
            void flush() {
            }
 
            const osmium::Relation& get_relation(size_t offset) const {
                assert(m_relations_buffer.committed() > offset);
                return m_relations_buffer.get<osmium::Relation>(offset);
            }
 
            const osmium::Relation& get_relation(const RelationMeta& relation_meta) const {
                return get_relation(relation_meta.relation_offset());
            }
 
            const osmium::Relation& get_relation(const MemberMeta& member_meta) const {
                return get_relation(m_relations[member_meta.relation_pos()]);
            }
 
            osmium::OSMObject& get_member(size_t offset) const {
                assert(m_members_buffer.committed() > offset);
                return m_members_buffer.get<osmium::OSMObject>(offset);
            }
 
        private:
 
            void add_relation(const osmium::Relation& relation) {
                const size_t offset = m_relations_buffer.committed();
                m_relations_buffer.add_item(relation);
 
                RelationMeta relation_meta{offset};
 
                int n = 0;
                for (auto& member : m_relations_buffer.get<osmium::Relation>(offset).members()) {
                    if (static_cast<TCollector*>(this)->keep_member(relation_meta, member)) {
                        member_meta(member.type()).emplace_back(member.ref(), m_relations.size(), n);
                        relation_meta.increment_need_members();
                    } else {
                        member.set_ref(0); // set member id to zero to indicate we are not interested
                    }
                    ++n;
                }
 
                assert(offset == m_relations_buffer.committed());
                if (relation_meta.has_all_members()) {
                    m_relations_buffer.rollback();
                } else {
                    m_relations_buffer.commit();
                    m_relations.push_back(relation_meta);
                }
            }
 
            void sort_member_meta() {
                std::sort(m_member_meta[0].begin(), m_member_meta[0].end());
                std::sort(m_member_meta[1].begin(), m_member_meta[1].end());
                std::sort(m_member_meta[2].begin(), m_member_meta[2].end());
            }
 
            static typename iterator_range<mm_iterator>::iterator::difference_type count_not_removed(const iterator_range<mm_iterator>& range) {
                return std::count_if(range.begin(), range.end(), [](const MemberMeta& mm) {
                    return !mm.removed();
                });
            }
 
            bool find_and_add_object(const osmium::OSMObject& object) {
                auto range = find_member_meta(object.type(), object.id());
 
                if (count_not_removed(range) == 0) {
                    // nothing found
                    return false;
                }
 
                {
                    members_buffer().add_item(object);
                    const size_t member_offset = members_buffer().commit();
 
                    for (auto& member : range) {
                        member.set_buffer_offset(member_offset);
                    }
                }
 
                for (auto& member : range) {
                    if (member.removed()) {
                        break;
                    }
                    assert(member.member_id() == object.id());
                    assert(member.relation_pos() < m_relations.size());
                    RelationMeta& relation_meta = m_relations[member.relation_pos()];
                    assert(member.member_pos() < get_relation(relation_meta).members().size());
                    relation_meta.got_one_member();
                    if (relation_meta.has_all_members()) {
                        const size_t relation_offset = member.relation_pos();
                        static_cast<TCollector*>(this)->complete_relation(relation_meta);
                        clear_member_metas(relation_meta);
                        m_relations[relation_offset] = RelationMeta{};
                        possibly_purge_removed_members();
                    }
                }
 
                return true;
            }
 
            void clear_member_metas(const RelationMeta& relation_meta) {
                const osmium::Relation& relation = get_relation(relation_meta);
                for (const auto& member : relation.members()) {
                    if (member.ref() != 0) {
                        const auto range = find_member_meta(member.type(), member.ref());
                        assert(!range.empty());
 
                        // if this is the last time this object was needed
                        // then mark it as removed
                        if (count_not_removed(range) == 1) {
                            get_member(range.begin()->buffer_offset()).set_removed(true);
                        }
 
                        for (auto& member_meta : range) {
                            if (!member_meta.removed() && relation.id() == get_relation(member_meta).id()) {
                                member_meta.remove();
                                break;
                            }
                        }
                    }
                }
            }
 
        public:
 
            uint64_t used_memory() const {
                const uint64_t nmembers = m_member_meta[0].capacity() + m_member_meta[1].capacity() + m_member_meta[2].capacity();
                const uint64_t members = nmembers * sizeof(MemberMeta);
                const uint64_t relations_size = m_relations.capacity() * sizeof(RelationMeta);
                const uint64_t relations_buffer_capacity = m_relations_buffer.capacity();
                const uint64_t members_buffer_capacity = m_members_buffer.capacity();
 
                std::cerr << "  nR  = m_relations.capacity() ........... = " << std::setw(12) << m_relations.capacity() << "\n";
                std::cerr << "  nMN = m_member_meta[NODE].capacity() ... = " << std::setw(12) << m_member_meta[0].capacity() << "\n";
                std::cerr << "  nMW = m_member_meta[WAY].capacity() .... = " << std::setw(12) << m_member_meta[1].capacity() << "\n";
                std::cerr << "  nMR = m_member_meta[RELATION].capacity() = " << std::setw(12) << m_member_meta[2].capacity() << "\n";
                std::cerr << "  nM  = m_member_meta[*].capacity() ...... = " << std::setw(12) << nmembers << "\n";
 
                std::cerr << "  sRM = sizeof(RelationMeta) ............. = " << std::setw(12) << sizeof(RelationMeta) << "\n";
                std::cerr << "  sMM = sizeof(MemberMeta) ............... = " << std::setw(12) << sizeof(MemberMeta) << "\n\n";
 
                std::cerr << "  nR * sRM ............................... = " << std::setw(12) << relations_size << "\n";
                std::cerr << "  nM * sMM ............................... = " << std::setw(12) << members << "\n";
                std::cerr << "  relations_buffer_capacity .............. = " << std::setw(12) << relations_buffer_capacity << "\n";
                std::cerr << "  members_buffer_capacity ................ = " << std::setw(12) << members_buffer_capacity << "\n";
 
                const uint64_t total = relations_size + members + relations_buffer_capacity + members_buffer_capacity;
 
                std::cerr << "  total .................................. = " << std::setw(12) << total << "\n";
                std::cerr << "  =======================================================\n";
 
                return relations_buffer_capacity + members_buffer_capacity + relations_size + members;
            }
 
            HandlerPass2& handler(const callback_func_type& callback = nullptr) {
                m_callback = callback;
                return m_handler_pass2;
            }
 
            osmium::memory::Buffer& members_buffer() {
                return m_members_buffer;
            }
 
            bool is_available(osmium::item_type type, osmium::object_id_type id) {
                const auto range = find_member_meta(type, id);
                assert(!range.empty());
                return range.begin()->is_available();
            }
 
            size_t get_offset(osmium::item_type type, osmium::object_id_type id) {
                const auto range = find_member_meta(type, id);
                assert(!range.empty());
                assert(range.begin()->is_available());
                return range.begin()->buffer_offset();
            }
 
            std::pair<bool, size_t> get_availability_and_offset(osmium::item_type type, osmium::object_id_type id) {
                const auto range = find_member_meta(type, id);
                assert(!range.empty());
                if (range.begin()->is_available()) {
                    return std::make_pair(true, range.begin()->buffer_offset());
                }
                return std::make_pair(false, 0);
            }
 
            template <typename TIter>
            void read_relations(TIter begin, TIter end) {
                HandlerPass1 handler_pass1{*static_cast<TCollector*>(this)};
                osmium::apply(begin, end, handler_pass1);
                sort_member_meta();
            }
 
            template <typename TSource>
            void read_relations(TSource& source) {
                using std::begin;
                using std::end;
                read_relations(begin(source), end(source));
                source.close();
            }
 
            void moving_in_buffer(size_t old_offset, size_t new_offset) {
                const osmium::OSMObject& object = m_members_buffer.get<osmium::OSMObject>(old_offset);
                auto range = find_member_meta(object.type(), object.id());
                for (auto& member : range) {
                    assert(member.buffer_offset() == old_offset);
                    member.set_buffer_offset(new_offset);
                }
            }
 
            void possibly_purge_removed_members() {
                ++m_count_complete;
                if (m_count_complete > 10000) { // XXX
//                    const size_t size_before = m_members_buffer.committed();
                    m_members_buffer.purge_removed(this);
/*
                    const size_t size_after = m_members_buffer.committed();
                    double percent = static_cast<double>(size_before - size_after);
                    percent /= size_before;
                    percent *= 100;
                    std::cerr << "PURGE (size before=" << size_before << " after=" << size_after << " purged=" << (size_before - size_after) << " / " << static_cast<int>(percent) << "%)\n";
*/
                    m_count_complete = 0;
                }
            }
 
            std::vector<const osmium::Relation*> get_incomplete_relations() const {
                std::vector<const osmium::Relation*> incomplete_relations;
                for (const auto& relation_meta : m_relations) {
                    if (!relation_meta.has_all_members()) {
                        incomplete_relations.push_back(&get_relation(relation_meta));
                    }
                }
                return incomplete_relations;
            }
 
        }; // class Collector
 
    } // namespace relations
 
} // namespace osmium
 
#endif // OSMIUM_RELATIONS_COLLECTOR_HPP