realm / realm-core / thomas.goyne_232

{

}

{

    size_t sz = m_children.size();

    size_t current_cond = 0;

    size_t nb_cond_to_test = sz;

    while (REALM_LIKELY(start < end)) {

        size_t m = m_children[current_cond]->find_first_local(start, end);

        if (m != start) {

            // Pointer advanced - we will have to check all other conditions

            nb_cond_to_test = sz;

            start = m;

        }

        nb_cond_to_test--;

        // Optimized for one condition where this will be true first time

        if (REALM_LIKELY(nb_cond_to_test == 0))

            return m;

        current_cond++;

        if (current_cond == sz)

            current_cond = 0;

    }

    return not_found;

}

{

    REALM_ASSERT(is_valid());

    Cluster cluster(0, get_alloc(), m_table->m_clusters);

    cluster.init(m_mem);

    cluster.set_offset(m_key.value - cluster.get_key_value(m_row_ndx));

    return func(&cluster, m_row_ndx);

}

{

    return obj.evaluate([this](const Cluster* cluster, size_t row) {

        set_cluster(cluster);

        size_t m = find_first(row, row + 1);

        return m != npos;

    });

}

{

    // aggregate called on non-integer column type. Speed of this function is not as critical as speed of the

    // integer version, because find_first_local() is relatively slower here (because it's non-integers).

    //

    // Todo: Two speedups are possible. Simple: Initially test if there are no sub criterias and run

    // find_first_local()

    // in a tight loop if so (instead of testing if there are sub criterias after each match). Harder: Specialize

    // data type array to make array call match() directly on each match, like for integers.

    m_state = st;

    m_state->set_payload_column(source_column);

    size_t local_matches = 0;

    if (m_children.size() == 1) {

        return find_all_local(start, end);

    }

    size_t r = start - 1;

    for (;;) {

        if (local_matches == local_limit) {

            m_dD = double(r - start) / (local_matches + 1.1);

            return r + 1;

        }

        // Find first match in this condition node

        auto pos = r + 1;

        r = find_first_local(pos, end);

        if (r == not_found) {

            m_dD = double(pos - start) / (local_matches + 1.1);

            return end;

        }

        local_matches++;

        // Find first match in remaining condition nodes

        size_t m = r;

        for (size_t c = 1; c < m_children.size(); c++) {

            m = m_children[c]->find_first_local(r, r + 1);

            if (m != r) {

                break;

            }

        }

        // If index of first match in this node equals index of first match in all remaining nodes, we have a final

        // match

        if (m == r) {

            bool cont = st->match(r);

            if (!cont) {

                return static_cast<size_t>(-1);

            }

        }

    }

}

{

    while (start < end) {

        start = find_first_local(start, end);

        if (start != not_found) {

            bool cont = m_state->match(start);

            if (!cont) {

                return static_cast<size_t>(-1);

            }

            start++;

        }

    }

    return end;

}

{

    MixedNodeBase::init(will_query_ranges);

    REALM_ASSERT(bool(m_index_evaluator) ==

                 (m_table.unchecked_ptr()->search_index_type(m_condition_column_key) == IndexType::General));

    if (m_index_evaluator) {

        auto index = ParentNode::m_table->get_search_index(ParentNode::m_condition_column_key);

        m_index_evaluator->init(index, m_value);

        m_dT = 0.0;

    }

    else {

        m_dT = 10.0;

    }

}

{

    REALM_ASSERT(m_table);

    if (m_index_evaluator) {

        return m_index_evaluator->do_search_index(m_cluster, start, end);

    }

    else {

        return m_leaf->find_first(m_value, start, end);

    }

{

    MixedNodeBase::init(will_query_ranges);

    StringData val_as_string;

    if (m_value.is_type(type_String)) {

        val_as_string = m_value.get<StringData>();

    }

    else if (m_value.is_type(type_Binary)) {

        BinaryData bin = m_value.get<BinaryData>();

        val_as_string = StringData(bin.data(), bin.size());

    }

    REALM_ASSERT(bool(m_index_evaluator) ==

                 (m_table.unchecked_ptr()->search_index_type(m_condition_column_key) == IndexType::General));

    if (m_index_evaluator) {

        auto index = ParentNode::m_table->get_search_index(ParentNode::m_condition_column_key);

        if (!val_as_string.is_null()) {

            m_index_matches.clear();

            constexpr bool case_insensitive = true;

            index->find_all(m_index_matches, val_as_string, case_insensitive);

            // It is unfortunate but necessary to check the type due to Binary and String

            // having the same StringIndex hash values

            m_index_matches.erase(std::remove_if(m_index_matches.begin(), m_index_matches.end(),

                                                 [this](const ObjKey& obj_key) {

                                                     Mixed to_check =

                                                         m_table->get_object(obj_key).get_any(m_condition_column_key);

                                                     return (!Mixed::types_are_comparable(to_check, m_value));

                                                 }),

                                  m_index_matches.end());

            m_index_evaluator->init(&m_index_matches);

        }

        else {

            // search for non string can use exact match

            m_index_evaluator->init(index, m_value);

        }

        m_dT = 0.0;

    }

    else {

        m_dT = 10.0;

        if (!val_as_string.is_null()) {

            auto upper = case_map(val_as_string, true);

            auto lower = case_map(val_as_string, false);

            if (!upper || !lower) {

            else {

                m_ucase = std::move(*upper);

                m_lcase = std::move(*lower);

            }

        }

    }

}

{

    REALM_ASSERT(m_table);

    EqualIns cond;

    if (m_value.is_type(type_String, type_Binary)) {

        for (size_t i = start; i < end; i++) {

            QueryValue val(m_leaf->get(i));

            if (!Mixed::types_are_comparable(m_value, val)) {

                continue;

            }

            StringData val_as_str = val.export_to_type<StringData>();

            if (cond(m_value.export_to_type<StringData>(), m_ucase.c_str(), m_lcase.c_str(), val_as_str))

                return i;

        }

    }

    else {

        for (size_t i = start; i < end; i++) {

            QueryValue val(m_leaf->get(i));

            if (cond(val, m_value))

                return i;

        }

    }

    return not_found;

}

{

    StringNodeBase::init(will_query_ranges);

    const bool uses_index = has_search_index();

    if (m_is_string_enum) {

        m_dT = 1.0;

    }

    else if (uses_index) {

        m_dT = 0.0;

    }

    else {

        m_dT = 10.0;

    }

    if (uses_index) {

        _search_index_init();

    }

}

{

    REALM_ASSERT(m_table);

    if (m_index_evaluator) {

        return m_index_evaluator->do_search_index(m_cluster, start, end);

    }

    return _find_first_local(start, end);

}

{

    REALM_ASSERT(index);

    m_matching_keys = nullptr;

    FindRes fr;

    InternalFindResult res;

    m_last_start_key = ObjKey();

    m_results_start = 0;

    fr = index->find_all_no_copy(value, res);

    m_index_matches.reset();

    switch (fr) {

        case FindRes_single:

            m_actual_key = ObjKey(res.payload);

            m_results_end = 1;

            break;

        case FindRes_column:

            m_index_matches.reset(new IntegerColumn(index->get_alloc(), ref_type(res.payload))); // Throws

            m_results_start = res.start_ndx;

            m_results_end = res.end_ndx;

            m_actual_key = ObjKey(m_index_matches->get(m_results_start));

            break;

        case FindRes_not_found:

            m_results_end = 0;

            break;

    }

    m_results_ndx = m_results_start;

}

{

    REALM_ASSERT(storage);

    m_matching_keys = storage;

    m_actual_key = ObjKey();

    m_last_start_key = ObjKey();

    m_results_start = 0;

    m_results_end = m_matching_keys->size();

    m_results_ndx = 0;

    if (m_results_start != m_results_end) {

        m_actual_key = m_matching_keys->at(0);

    }

}

{

    if (start >= end) {

        return not_found;

    }

    ObjKey first_key = cluster->get_real_key(start);

    if (first_key < m_last_start_key) {

        // We are not advancing through the clusters. We basically don't know where we are,

        // so just start over from the beginning.

        m_results_ndx = m_results_start;

        m_actual_key = (m_results_start != m_results_end) ? get_internal(m_results_start) : ObjKey();

    }

    m_last_start_key = first_key;

    // Check if we can expect to find more keys

    if (m_results_ndx < m_results_end) {

        // Check if we should advance to next key to search for

        while (first_key > m_actual_key) {

            m_results_ndx++;

            if (m_results_ndx == m_results_end) {

                return not_found;

            }

            m_actual_key = get_internal(m_results_ndx);

        }

        // If actual_key is bigger than last key, it is not in this leaf

        ObjKey last_key = cluster->get_real_key(end - 1);

        if (m_actual_key > last_key)

            return not_found;

        // Now actual_key must be found in leaf keys

        return cluster->lower_bound_key(ObjKey(m_actual_key.value - cluster->get_offset()));

    }

    return not_found;

}

{

    REALM_ASSERT(bool(m_index_evaluator));

    auto index = ParentNode::m_table.unchecked_ptr()->get_search_index(ParentNode::m_condition_column_key);

    m_index_evaluator->init(index, StringNodeBase::m_string_value);

}

{

    // Don't use the search index if present since we're in a scenario where

    // it'd be slower

    m_index_evaluator.reset();

    auto& other = static_cast<StringNode<Equal>&>(node);

    REALM_ASSERT(m_condition_column_key == other.m_condition_column_key);

    REALM_ASSERT(other.m_needles.empty());

    if (m_needles.empty()) {

        m_needles.insert(m_string_value);

    }

    if (auto& str = other.m_value) {

        m_needle_storage.push_back(std::make_unique<char[]>(str->size()));

        std::copy(str->data(), str->data() + str->size(), m_needle_storage.back().get());

        m_needles.insert(StringData(m_needle_storage.back().get(), str->size()));

    }

    else {

        m_needles.emplace();

    }

    return true;

}

{

    if (m_needles.empty()) {

        return m_leaf->find_first(m_string_value, start, end);

    }

    else {

        if (end == npos)

        REALM_ASSERT_3(start, <=, end);

        return find_first_haystack<20>(*m_leaf, m_needles, start, end);

    }

}

{

    if (m_needles.empty()) {

        return StringNodeEqualBase::describe(state);

    }

    std::string list_contents;

    bool is_first = true;

    for (auto it : m_needles) {

        StringData sd(it.data(), it.size());

        list_contents += util::format("%1%2", is_first ? "" : ", ", util::serializer::print_value(sd));

        is_first = false;

    }

    std::string col_descr = state.describe_column(ParentNode::m_table, m_condition_column_key);

    std::string desc = util::format("%1 IN {%2}", col_descr, list_contents);

    return desc;

}

{

    auto index = ParentNode::m_table->get_search_index(ParentNode::m_condition_column_key);

    m_index_matches.clear();

    constexpr bool case_insensitive = true;

    index->find_all(m_index_matches, StringNodeBase::m_string_value, case_insensitive);

    m_index_evaluator->init(&m_index_matches);

}

{

    EqualIns cond;

    for (size_t s = start; s < end; ++s) {

        StringData t = get_string(s);

        if (cond(m_string_value, m_ucase.c_str(), m_lcase.c_str(), t))

            return s;

    }

    return not_found;

}

{

    if (!m_link_map)

        m_link_map = std::make_unique<LinkMap>();

}

{

    StringNodeEqualBase::table_changed();

    m_link_map->set_base_table(m_table);

}

{

    m_link_map = std::make_unique<LinkMap>(*other.m_link_map);

}

{

    StringIndex* index = m_link_map->get_target_table()->get_string_index(ParentNode::m_condition_column_key);

    REALM_ASSERT(index && index->is_fulltext_index());

    m_index_matches.clear();

    index->find_all_fulltext(m_index_matches, StringNodeBase::m_string_value);

    // If links exists, use backlinks to find the original objects

    if (m_link_map->links_exist()) {

        std::set<ObjKey> tmp;

        for (auto k : m_index_matches) {

            auto keys = m_link_map->get_origin_objkeys(k);

            tmp.insert(keys.begin(), keys.end());

        }

        m_index_matches.assign(tmp.begin(), tmp.end());

    }

    m_index_evaluator = IndexEvaluator{};

    m_index_evaluator->init(&m_index_matches);

}

{

    switch (col_key.get_type()) {

        case col_type_Int:

            if (col_key.is_nullable()) {

                return std::make_unique<ArrayIntNull>(alloc);

            }

            return std::make_unique<ArrayInteger>(alloc);

        case col_type_Bool:

            return std::make_unique<ArrayBoolNull>(alloc);

        case col_type_String:

            return std::make_unique<ArrayString>(alloc);

        case col_type_Binary:

        case col_type_Mixed:

            return std::make_unique<ArrayMixed>(alloc);

        case col_type_Timestamp:

            return std::make_unique<ArrayTimestamp>(alloc);

        case col_type_Float:

            return std::make_unique<ArrayFloatNull>(alloc);

        case col_type_Double:

            return std::make_unique<ArrayDoubleNull>(alloc);

        case col_type_Decimal:

            return std::make_unique<ArrayDecimal128>(alloc);

        case col_type_Link:

        case col_type_ObjectId:

        case col_type_UUID:

        case col_type_TypedLink:

{

    switch (col_type) {

        case col_type_Int: {

            if (is_nullable) {

                BPlusTree<util::Optional<Int>> list(alloc);

                list.init_from_ref(ref);

                return list.size();

            }

            else {

                BPlusTree<Int> list(alloc);

                list.init_from_ref(ref);

                return list.size();

            }

        case col_type_Bool: {

            BPlusTree<Bool> list(alloc);

            list.init_from_ref(ref);

            return list.size();

        case col_type_String: {

            BPlusTree<String> list(alloc);

            list.init_from_ref(ref);

            return list.size();

        case col_type_Binary: {

            BPlusTree<Binary> list(alloc);

            list.init_from_ref(ref);

            return list.size();

        case col_type_Timestamp: {

            BPlusTree<Timestamp> list(alloc);

            list.init_from_ref(ref);

            return list.size();

        case col_type_Float: {

            BPlusTree<Float> list(alloc);

            list.init_from_ref(ref);

            return list.size();

        case col_type_Double: {

            BPlusTree<Double> list(alloc);

            list.init_from_ref(ref);

            return list.size();

        case col_type_Decimal: {

            BPlusTree<Decimal128> list(alloc);

            list.init_from_ref(ref);

            return list.size();

        case col_type_ObjectId: {

            BPlusTree<ObjectId> list(alloc);

            list.init_from_ref(ref);

            return list.size();

        case col_type_UUID: {

            BPlusTree<UUID> list(alloc);

            list.init_from_ref(ref);

            return list.size();

        case col_type_Link: {

            BPlusTree<ObjKey> list(alloc);

            list.init_from_ref(ref);

            return list.size();

{

    if (start <= m_known_range_start && end >= m_known_range_end) {

        return find_first_covers_known(start, end);

    }

    else if (start >= m_known_range_start && end <= m_known_range_end) {

        return find_first_covered_by_known(start, end);

    }

    else if (start <= m_known_range_end && end > m_known_range_end) {

        return find_first_overlap_upper(start, end);

    }

    else { // start > m_known_range_end || end < m_known_range_start

        return find_first_no_overlap(start, end);

    }

}

{

    return m_condition->find_first(rowndx, rowndx + 1) == not_found;

}

{

    m_known_range_start = start;

    m_known_range_end = end;

    m_first_in_known_range = first;

}

{

    for (size_t i = start; i < end; ++i) {

        if (evaluate_at(i)) {

            return i;

        }

    }

    return not_found;

}

{

    // CASE: start-end covers the known range

    // [    ######    ]

    REALM_ASSERT_DEBUG(start <= m_known_range_start && end >= m_known_range_end);

    size_t result = find_first_loop(start, m_known_range_start);

    if (result != not_found) {

    else {

        if (m_first_in_known_range != not_found) {

        else {

            result = find_first_loop(m_known_range_end, end);

            update_known(start, end, result);

        }

    }

    return result;

}

{

    REALM_ASSERT_DEBUG(start >= m_known_range_start && end <= m_known_range_end);

    // CASE: the known range covers start-end

    // ###[#####]###

    if (m_first_in_known_range != not_found) {

        if (m_first_in_known_range > end) {

        else if (m_first_in_known_range >= start) {

    }

    // The first known match is before start, so we can't use the results to improve

    // heuristics.

    return find_first_loop(start, end);

}

{

    REALM_ASSERT_DEBUG(start <= m_known_range_end && start >= m_known_range_start && end > m_known_range_end);

    // CASE: partial overlap, upper end

    // ####[###    ]

    size_t result;

    if (m_first_in_known_range != not_found) {

        if (m_first_in_known_range >= start) {

        else {

            result = find_first_loop(start, end);

            update_known(m_known_range_start, end, m_first_in_known_range);

        }

    }

    else {

        result = find_first_loop(m_known_range_end, end);

        update_known(m_known_range_start, end, result);

    }

    return result;

}

{

                       (start > m_known_range_end && end > m_known_range_end));

    // CASE: no overlap

    // ### [    ]   or    [    ] ####

    // if input is a larger range, discard and replace with results.

    size_t result = find_first_loop(start, end);

    if (end - start > m_known_range_end - m_known_range_start) {

        update_known(start, end, result);

    }

    return result;

}

{

    m_dT = 50.0;

}

{

    m_expression->set_base_table(m_table);

}

{

    m_expression->set_cluster(m_cluster);

}

{

    ParentNode::init(will_query_ranges);

    m_dT = m_expression->init();

}

{

    if (m_expression) {

        return m_expression->description(state);

    }

}

{

    m_expression->collect_dependencies(tables);

}

{

    return m_expression->find_first(start, end);

}

{

    return std::unique_ptr<ParentNode>(new ExpressionNode(*this));

}

{

}

{

    if (m_condition_column_key.is_collection()) {

        Allocator& alloc = m_table.unchecked_ptr()->get_alloc();

        if (m_condition_column_key.is_dictionary()) {

            auto target_table_key = m_table->get_opposite_table(m_condition_column_key)->get_key();

            Array top(alloc);

            for (size_t i = start; i < end; i++) {

                if (ref_type ref = get_ref(i)) {

                    top.init_from_ref(ref);

                    BPlusTree<Mixed> values(alloc);

                    values.set_parent(&top, 1);

                    values.init_from_parent();

                    for (auto& key : m_target_keys) {

                        ObjLink link(target_table_key, key);

                        if (values.find_first(link) != not_found)

                            return i;

                    }

                }

            }

        }

        else {

            // LinkLists never contain null

            if (!m_target_keys[0] && m_target_keys.size() == 1 && start != end)

                return not_found;

            BPlusTree<ObjKey> links(alloc);

            for (size_t i = start; i < end; i++) {

                if (ref_type ref = get_ref(i)) {

                    links.init_from_ref(ref);

                    for (auto& key : m_target_keys) {

                        if (key) {

                            if (links.find_first(key) != not_found)

                                return i;

                        }

                    }

                }

            }

        }

    }

    else if (m_list) {

        for (auto& key : m_target_keys) {

            auto pos = m_list->find_first(key, start, end);

            if (pos != realm::npos) {

                return pos;

            }

        }

    }

    return not_found;

}

{

    // NotEqual only makes sense for a single value

    REALM_ASSERT(m_target_keys.size() == 1);

    ObjKey key = m_target_keys[0];

    if (m_condition_column_key.is_collection()) {

        Allocator& alloc = m_table.unchecked_ptr()->get_alloc();

        if (m_condition_column_key.is_dictionary()) {

            auto target_table_key = m_table->get_opposite_table(m_condition_column_key)->get_key();

            Array top(alloc);

            for (size_t i = start; i < end; i++) {

                if (ref_type ref = get_ref(i)) {

                    top.init_from_ref(ref);

                    BPlusTree<Mixed> values(alloc);

                    values.set_parent(&top, 1);

                    values.init_from_parent();

                    ObjLink link(target_table_key, key);

                    bool found = false;

                    values.for_all([&](const Mixed& val) {

                        if (val != link) {

                            found = true;

                        }

                        return !found;

                    });

                    if (found)

                        return i;

                }

            }

        }

        else {

            BPlusTree<ObjKey> links(alloc);

            for (size_t i = start; i < end; i++) {

                if (ref_type ref = get_ref(i)) {

                    links.init_from_ref(ref);

                    auto sz = links.size();

                    for (size_t j = 0; j < sz; j++) {

                        if (links.get(j) != key) {

                            return i;

                        }

                    }

                }

            }

        }

    }

    else if (m_list) {

        for (size_t i = start; i < end; i++) {

            if (m_list->get(i) != key) {

                return i;

            }

        }

    }

    return not_found;

}