be/src/storage/segment/variant/variant_statistics.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 <gen_cpp/segment_v2.pb.h> |
21 | | |
22 | | #include <cstdint> |
23 | | #include <map> |
24 | | #include <string> |
25 | | |
26 | | namespace doris::segment_v2 { |
27 | | |
28 | | #include "common/compile_check_begin.h" |
29 | | |
30 | | struct VariantStatistics { |
31 | | std::map<std::string, int64_t> subcolumns_non_null_size; |
32 | | std::map<std::string, uint32_t> sparse_column_non_null_size; |
33 | | std::map<std::string, uint32_t> doc_value_column_non_null_size; |
34 | | bool has_nested_group = false; |
35 | | |
36 | 635 | void to_pb(VariantStatisticsPB* stats) const { |
37 | 635 | if (stats == nullptr) { |
38 | 0 | return; |
39 | 0 | } |
40 | 635 | stats->Clear(); |
41 | 635 | auto* sparse_map = stats->mutable_sparse_column_non_null_size(); |
42 | 4.15k | for (const auto& [path, value] : sparse_column_non_null_size) { |
43 | 4.15k | (*sparse_map)[path] = value; |
44 | 4.15k | } |
45 | 635 | auto* doc_value_map = stats->mutable_doc_value_column_non_null_size(); |
46 | 635 | for (const auto& [path, value] : doc_value_column_non_null_size) { |
47 | 31 | (*doc_value_map)[path] = value; |
48 | 31 | } |
49 | 635 | stats->set_has_nested_group(has_nested_group); |
50 | 635 | } |
51 | | |
52 | | // Additive merge: only update sparse_column_non_null_size from pb |
53 | | // without touching other maps (subcolumns, doc_value, has_nested_group). |
54 | 380 | void merge_sparse_from_pb(const VariantStatisticsPB& stats) { |
55 | 2.50k | for (const auto& [path, value] : stats.sparse_column_non_null_size()) { |
56 | 2.50k | sparse_column_non_null_size[path] += value; |
57 | 2.50k | } |
58 | 380 | } |
59 | | |
60 | | // Additive merge: only update doc_value_column_non_null_size from pb |
61 | | // without touching other maps. |
62 | 10 | void merge_doc_value_from_pb(const VariantStatisticsPB& stats) { |
63 | 35 | for (const auto& [path, value] : stats.doc_value_column_non_null_size()) { |
64 | 35 | doc_value_column_non_null_size[path] += value; |
65 | 35 | } |
66 | 10 | } |
67 | | |
68 | 0 | bool has_sparse_column_non_null_size() const { return !sparse_column_non_null_size.empty(); } |
69 | 211 | bool has_doc_value_column_non_null_size() const { |
70 | 211 | return !doc_value_column_non_null_size.empty(); |
71 | 211 | } |
72 | | |
73 | 238 | bool existed_in_sparse_column(const std::string& relative_path) const { |
74 | 238 | return sparse_column_non_null_size.contains(relative_path); |
75 | 238 | } |
76 | | |
77 | 6 | bool existed_in_doc_value_column(const std::string& relative_path) const { |
78 | 6 | return doc_value_column_non_null_size.contains(relative_path); |
79 | 6 | } |
80 | | |
81 | 22 | bool has_prefix_path_in_sparse_column(const std::string& dot_prefix) const { |
82 | 22 | auto it = sparse_column_non_null_size.lower_bound(dot_prefix); |
83 | 22 | return it != sparse_column_non_null_size.end() && it->first.starts_with(dot_prefix); |
84 | 22 | } |
85 | | |
86 | 6 | bool has_prefix_path_in_doc_value_column(const std::string& dot_prefix) const { |
87 | 6 | auto it = doc_value_column_non_null_size.lower_bound(dot_prefix); |
88 | 6 | return it != doc_value_column_non_null_size.end() && it->first.starts_with(dot_prefix); |
89 | 6 | } |
90 | | |
91 | 1 | bool has_doc_column_non_null_size() const { return has_doc_value_column_non_null_size(); } |
92 | 0 | bool existed_in_doc_column(const std::string& relative_path) const { |
93 | 0 | return existed_in_doc_value_column(relative_path); |
94 | 0 | } |
95 | 0 | bool has_prefix_path_in_doc_column(const std::string& dot_prefix) const { |
96 | 0 | return has_prefix_path_in_doc_value_column(dot_prefix); |
97 | 0 | } |
98 | | }; |
99 | | #include "common/compile_check_end.h" |
100 | | |
101 | | } // namespace doris::segment_v2 |