be/src/exprs/aggregate/aggregate_function_histogram.h

Source
// Licensed to the Apache Software Foundation (ASF) under one
// or more contributor license agreements.  See the NOTICE file
// distributed with this work for additional information
// regarding copyright ownership.  The ASF licenses this file
// to you under the Apache License, Version 2.0 (the
// "License"); you may not use this file except in compliance
// with the License.  You may obtain a copy of the License at
//
//   http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing,
// software distributed under the License is distributed on an
// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
// KIND, either express or implied.  See the License for the
// specific language governing permissions and limitations
// under the License.

#pragma once

#include <map>
#include <memory>
#include <string>
#include <utility>
#include <vector>

#include "common/exception.h"
#include "common/status.h"
#include "core/assert_cast.h"
#include "core/column/column.h"
#include "core/column/column_decimal.h"
#include "core/column/column_string.h"
#include "core/column/column_vector.h"
#include "core/data_type/data_type_string.h"
#include "core/string_ref.h"
#include "core/types.h"
#include "exec/common/histogram_helpers.hpp"
#include "exprs/aggregate/aggregate_function.h"
#include "exprs/aggregate/aggregate_function_simple_factory.h"

namespace doris {} // namespace doris

namespace doris {

template <PrimitiveType T>
struct AggregateFunctionHistogramData {
    static constexpr auto Ptype = T;
    using ColVecType = typename PrimitiveTypeTraits<T>::ColumnType;
    const static size_t DEFAULT_BUCKET_NUM = 128;
    const static size_t BUCKET_NUM_INIT_VALUE = 0;

    void set_parameters(size_t input_max_num_buckets) { max_num_buckets = input_max_num_buckets; }

    void reset() {
        ordered_map.clear();
        max_num_buckets = BUCKET_NUM_INIT_VALUE;
    }

    void add(const StringRef& value, const UInt64& number = 1) {
        std::string data = value.to_string();
        auto it = ordered_map.find(data);
        if (it != ordered_map.end()) {
            it->second = it->second + number;
        } else {
            ordered_map.insert({data, number});
        }
    }

    void add(const typename PrimitiveTypeTraits<T>::CppType& value, const UInt64& number = 1) {
        auto it = ordered_map.find(value);
        if (it != ordered_map.end()) {
            it->second = it->second + number;
        } else {
            ordered_map.insert({value, number});
        }
    }

    void merge(const AggregateFunctionHistogramData& rhs) {
        // if rhs.max_num_buckets == 0, it means the input block for serialization is all null
        // we should discard this data, because histogram only fouce on the not-null data
        if (!rhs.max_num_buckets) {
            return;
        }

        if (!max_num_buckets) {
            max_num_buckets = rhs.max_num_buckets;
        } else if (UNLIKELY(max_num_buckets != rhs.max_num_buckets)) {
            throw Exception(ErrorCode::INVALID_ARGUMENT,
                            "histogram aggregate states have incompatible bucket counts: {} vs {}",
                            max_num_buckets, rhs.max_num_buckets);
        }

        for (auto rhs_it : rhs.ordered_map) {
            auto lhs_it = ordered_map.find(rhs_it.first);
            if (lhs_it != ordered_map.end()) {
                lhs_it->second += rhs_it.second;
            } else {
                ordered_map.insert({rhs_it.first, rhs_it.second});
            }
        }
    }

    void write(BufferWritable& buf) const {
        buf.write_binary(max_num_buckets);
        auto element_number = (size_t)ordered_map.size();
        buf.write_binary(element_number);

        auto pair_vector = map_to_vector();

        for (auto i = 0; i < element_number; i++) {
            auto element = pair_vector[i];
            buf.write_binary(element.second);
            buf.write_binary(element.first);
        }
    }

    void read(BufferReadable& buf) {
        buf.read_binary(max_num_buckets);

        size_t element_number = 0;
        buf.read_binary(element_number);

        ordered_map.clear();
        std::pair<typename PrimitiveTypeTraits<T>::CppType, size_t> element;
        for (auto i = 0; i < element_number; i++) {
            buf.read_binary(element.first);
            buf.read_binary(element.second);
            ordered_map.insert(element);
        }
    }

    void insert_result_into(IColumn& to) const {
        auto pair_vector = map_to_vector();
        for (auto i = 0; i < pair_vector.size(); i++) {
            const auto& element = pair_vector[i];
            if constexpr (is_string_type(T)) {
                assert_cast<ColumnString&, TypeCheckOnRelease::DISABLE>(to).insert_data(
                        element.second.c_str(), element.second.length());
            } else {
                assert_cast<ColVecType&, TypeCheckOnRelease::DISABLE>(to).get_data().push_back(
                        element.second);
            }
        }
    }

    std::string get(const DataTypePtr& data_type) const {
        std::vector<Bucket<typename PrimitiveTypeTraits<T>::CppType>> buckets;
        rapidjson::StringBuffer buffer;
        // NOTE: We need an extral branch for to handle max_num_buckets == 0,
        // when target column is nullable, and input block is all null,
        // set_parameters will not be called because of the short-circuit in
        // AggregateFunctionNullVariadicInline, so max_num_buckets will be 0 in this situation.
        build_histogram(
                buckets, ordered_map,
                max_num_buckets == BUCKET_NUM_INIT_VALUE ? DEFAULT_BUCKET_NUM : max_num_buckets);
        histogram_to_json(buffer, buckets, data_type);
        return {buffer.GetString()};
    }

    std::vector<std::pair<size_t, typename PrimitiveTypeTraits<T>::CppType>> map_to_vector() const {
        std::vector<std::pair<size_t, typename PrimitiveTypeTraits<T>::CppType>> pair_vector;
        for (auto it : ordered_map) {
            pair_vector.emplace_back(it.second, it.first);
        }
        return pair_vector;
    }

private:
    size_t max_num_buckets = BUCKET_NUM_INIT_VALUE;
    std::map<typename PrimitiveTypeTraits<T>::CppType, size_t> ordered_map;
};

template <typename Data, bool has_input_param>
class AggregateFunctionHistogram final
        : public IAggregateFunctionDataHelper<Data,
                                              AggregateFunctionHistogram<Data, has_input_param>>,
          VarargsExpression,
          NotNullableAggregateFunction {
public:
    AggregateFunctionHistogram() = default;
    AggregateFunctionHistogram(const DataTypes& argument_types_)
            : IAggregateFunctionDataHelper<Data, AggregateFunctionHistogram<Data, has_input_param>>(
                      argument_types_),
              _argument_type(argument_types_[0]) {}

    std::string get_name() const override { return "histogram"; }

    DataTypePtr get_return_type() const override { return std::make_shared<DataTypeString>(); }

    void add(AggregateDataPtr __restrict place, const IColumn** columns, ssize_t row_num,
             Arena&) const override {
        if constexpr (has_input_param) {
            Int32 input_max_num_buckets =
                    assert_cast<const ColumnInt32*, TypeCheckOnRelease::DISABLE>(columns[1])
                            ->get_element(row_num);
            if (input_max_num_buckets <= 0 || input_max_num_buckets > 1000000) {
                throw doris::Exception(
                        ErrorCode::INVALID_ARGUMENT,
                        "Invalid max_num_buckets {}, row_num {}, should be in (0, 1000000]",
                        input_max_num_buckets, row_num);
            }
            this->data(place).set_parameters(input_max_num_buckets);
        } else {
            this->data(place).set_parameters(Data::DEFAULT_BUCKET_NUM);
        }

        if constexpr (is_string_type(Data::Ptype)) {
            this->data(place).add(
                    assert_cast<const ColumnString&, TypeCheckOnRelease::DISABLE>(*columns[0])
                            .get_data_at(row_num));
        } else {
            this->data(place).add(
                    assert_cast<const typename Data::ColVecType&, TypeCheckOnRelease::DISABLE>(
                            *columns[0])
                            .get_data()[row_num]);
        }
    }

    void check_input_columns_type(const IColumn** columns) const override {
        this->template check_argument_column_type<typename Data::ColVecType>(columns[0]);
        if constexpr (has_input_param) {
            this->template check_argument_column_type<ColumnInt32>(columns[1]);
        }
    }

    void reset(AggregateDataPtr place) const override { this->data(place).reset(); }

    void merge(AggregateDataPtr __restrict place, ConstAggregateDataPtr rhs,
               Arena&) const override {
        this->data(place).merge(this->data(rhs));
    }

    void serialize(ConstAggregateDataPtr __restrict place, BufferWritable& buf) const override {
        this->data(place).write(buf);
    }

    void deserialize(AggregateDataPtr __restrict place, BufferReadable& buf,
                     Arena&) const override {
        this->data(place).read(buf);
    }

    void insert_result_into(ConstAggregateDataPtr __restrict place, IColumn& to) const override {
        const std::string bucket_json = this->data(place).get(_argument_type);
        assert_cast<ColumnString&, TypeCheckOnRelease::DISABLE>(to).insert_data(
                bucket_json.c_str(), bucket_json.length());
    }

    void check_result_column_type(const IColumn& to) const override {
        IAggregateFunction::check_result_column_type(to);
        this->template check_result_column_type_as<ColumnString>(to);
    }

private:
    DataTypePtr _argument_type;
};

} // namespace doris

Coverage Report

Created: 2026-09-28 10:29