be/src/exprs/function/complex_hash_map_dictionary.h
Line | Count | Source |
1 | | // Licensed to the Apache Software Foundation (ASF) under one |
2 | | // or more contributor license agreements. See the NOTICE file |
3 | | // distributed with this work for additional information |
4 | | // regarding copyright ownership. The ASF licenses this file |
5 | | // to you under the Apache License, Version 2.0 (the |
6 | | // "License"); you may not use this file except in compliance |
7 | | // with the License. You may obtain a copy of the License at |
8 | | // |
9 | | // http://www.apache.org/licenses/LICENSE-2.0 |
10 | | // |
11 | | // Unless required by applicable law or agreed to in writing, |
12 | | // software distributed under the License is distributed on an |
13 | | // "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY |
14 | | // KIND, either express or implied. See the License for the |
15 | | // specific language governing permissions and limitations |
16 | | // under the License. |
17 | | |
18 | | #pragma once |
19 | | |
20 | | #include <memory> |
21 | | #include <utility> |
22 | | #include <vector> |
23 | | |
24 | | #include "common/status.h" |
25 | | #include "core/block/columns_with_type_and_name.h" |
26 | | #include "core/column/column.h" |
27 | | #include "core/data_type/data_type.h" |
28 | | #include "core/data_type/data_type_nullable.h" |
29 | | #include "core/data_type/data_type_string.h" |
30 | | #include "exprs/function/complex_dict_hash_map.h" |
31 | | #include "exprs/function/dictionary.h" |
32 | | |
33 | | namespace doris { |
34 | | |
35 | | // ComplexHashMapDictionary is a dictionary used to store multiple keys and multiple values |
36 | | // Currently, it is in a relatively simple state. In the future, it will reuse the hashmap of hashjoin |
37 | | // In fact, you can think of this class as a hashjoin that will not be cleared (but the specific hashmap used is different from that used by hashjoin) |
38 | | class ComplexHashMapDictionary : public IDictionary { |
39 | | public: |
40 | | ComplexHashMapDictionary(std::string name, std::vector<DictionaryAttribute> attributes) |
41 | 688 | : IDictionary(std::move(name), std::move(attributes)) {} |
42 | | |
43 | | ~ComplexHashMapDictionary() override; |
44 | | static DictionaryPtr create_complex_hash_map_dict(const std::string& name, |
45 | | const ColumnPtrs& key_columns, |
46 | | const DataTypes& key_types, |
47 | 688 | const ColumnsWithTypeAndName& values_data) { |
48 | 688 | std::vector<DictionaryAttribute> attributes; |
49 | 688 | std::vector<ColumnPtr> values_column; |
50 | 694 | for (const auto& att : values_data) { |
51 | 694 | attributes.push_back({att.name, att.type}); |
52 | 694 | values_column.push_back(att.column); |
53 | 694 | } |
54 | 688 | auto dict = std::make_shared<ComplexHashMapDictionary>(name, attributes); |
55 | 688 | dict->load_data(key_columns, key_types, values_column); |
56 | 688 | return dict; |
57 | 688 | } |
58 | | |
59 | | ColumnPtr get_column(const std::string& attribute_name, const DataTypePtr& attribute_type, |
60 | 1.01k | const ColumnPtr& key_column, const DataTypePtr& key_type) const override { |
61 | 1.01k | return get_tuple_columns({attribute_name}, {attribute_type}, {key_column}, {key_type})[0]; |
62 | 1.01k | } |
63 | | |
64 | | ColumnPtrs get_tuple_columns(const std::vector<std::string>& attribute_names, |
65 | | const DataTypes& attribute_types, const ColumnPtrs& key_columns, |
66 | | const DataTypes& key_types) const override; |
67 | | |
68 | | size_t allocated_bytes() const override; |
69 | | |
70 | | private: |
71 | | void load_data(const ColumnPtrs& key_columns, const DataTypes& key_types, |
72 | | const std::vector<ColumnPtr>& values_column); |
73 | | |
74 | | ColumnPtr get_single_value_column(const IColumn::Selector& value_index, const NullMap& null_map, |
75 | | const std::string& attribute_name, |
76 | | const DataTypePtr& attribute_type) const; |
77 | | |
78 | | void init_find_hash_map(DictionaryHashMapMethod& find_hash_map_method, |
79 | | const DataTypes& key_types) const; |
80 | | |
81 | | DictionaryHashMapMethod _hash_map_method; |
82 | | |
83 | | // Used to save key columns, because some types of hashmaps do not hold key columns, such as MethodStringNoCache |
84 | | ColumnPtrs _key_columns; |
85 | | }; |
86 | | |
87 | | inline DictionaryPtr create_complex_hash_map_dict_from_column( |
88 | | const std::string& name, const ColumnsWithTypeAndName& key_data, |
89 | 12 | const ColumnsWithTypeAndName& values_data) { |
90 | 12 | ColumnPtrs key_columns; |
91 | 12 | DataTypes key_types; |
92 | 16 | for (const auto& key : key_data) { |
93 | 16 | if (key.column->is_nullable()) { |
94 | 0 | throw doris::Exception(ErrorCode::INTERNAL_ERROR, |
95 | 0 | "ComplexHashMapDictionary key column should not be nullable"); |
96 | 0 | } |
97 | 16 | key_columns.push_back(key.column); |
98 | 16 | key_types.push_back(key.type); |
99 | 16 | } |
100 | 12 | auto dict = ComplexHashMapDictionary::create_complex_hash_map_dict(name, key_columns, key_types, |
101 | 12 | values_data); |
102 | 12 | return dict; |
103 | 12 | } |
104 | | } // namespace doris |