be/src/format_v2/materialized_reader_util.cpp
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 | | #include "format_v2/materialized_reader_util.h" |
19 | | |
20 | | #include <utility> |
21 | | |
22 | | #include "core/block/block.h" |
23 | | #include "core/column/column_vector.h" |
24 | | #include "core/data_type/data_type_nullable.h" |
25 | | #include "core/types.h" |
26 | | #include "exprs/vexpr_context.h" |
27 | | #include "format_v2/file_reader.h" |
28 | | #include "io/io_common.h" |
29 | | |
30 | | namespace doris::format { |
31 | | namespace { |
32 | | |
33 | 4 | void update_counter(RuntimeProfile::Counter* counter, int64_t value) { |
34 | 4 | if (counter != nullptr) { |
35 | 4 | COUNTER_UPDATE(counter, value); |
36 | 4 | } |
37 | 4 | } |
38 | | |
39 | | } // namespace |
40 | | |
41 | 4 | ColumnPtr make_column_nullable_if_needed(ColumnPtr column, const DataTypePtr& target_type) { |
42 | 4 | if (target_type != nullptr && target_type->is_nullable() && column.get() != nullptr && |
43 | 4 | !column->is_nullable()) { |
44 | 4 | return make_nullable(std::move(column)); |
45 | 4 | } |
46 | 0 | return column; |
47 | 4 | } |
48 | | |
49 | | Status apply_materialized_reader_filters(const FileScanRequest* request, io::IOContext* io_ctx, |
50 | | Block* file_block, size_t* rows, |
51 | 57 | const MaterializedReaderFilterProfile* profile) { |
52 | 57 | DORIS_CHECK(file_block != nullptr); |
53 | 57 | DORIS_CHECK(rows != nullptr); |
54 | 57 | const size_t rows_before_filter = *rows; |
55 | 57 | size_t rows_after_delete_filter = rows_before_filter; |
56 | 57 | if (request != nullptr && rows_before_filter > 0 && !request->delete_conjuncts.empty()) { |
57 | 1 | { |
58 | 1 | SCOPED_TIMER(profile == nullptr ? nullptr : profile->delete_conjunct_filter_time); |
59 | | // Delete conjuncts use the opposite polarity from ordinary predicates: TRUE marks a |
60 | | // row for deletion. Multiple delete files are combined as a union, so a row remains |
61 | | // visible only when every delete conjunct returns FALSE. |
62 | 1 | IColumn::Filter keep_filter(rows_before_filter, 1); |
63 | 1 | for (const auto& delete_conjunct : request->delete_conjuncts) { |
64 | 1 | DORIS_CHECK(delete_conjunct != nullptr); |
65 | 1 | int result_column_id = -1; |
66 | 1 | RETURN_IF_ERROR(delete_conjunct->root()->execute(delete_conjunct.get(), file_block, |
67 | 1 | &result_column_id)); |
68 | 1 | DORIS_CHECK(result_column_id >= 0 && |
69 | 1 | result_column_id < static_cast<int>(file_block->columns())); |
70 | 1 | const auto& delete_filter = |
71 | 1 | assert_cast<const ColumnUInt8&>( |
72 | 1 | *file_block->get_by_position(result_column_id).column) |
73 | 1 | .get_data(); |
74 | 1 | DORIS_CHECK(delete_filter.size() == rows_before_filter); |
75 | 4 | for (size_t row = 0; row < rows_before_filter; ++row) { |
76 | 3 | keep_filter[row] &= !delete_filter[row]; |
77 | 3 | } |
78 | 1 | file_block->erase(result_column_id); |
79 | 1 | } |
80 | 1 | RETURN_IF_CATCH_EXCEPTION(Block::filter_block_internal(file_block, keep_filter)); |
81 | 1 | } |
82 | 1 | rows_after_delete_filter = |
83 | 1 | file_block->columns() == 0 ? rows_before_filter : file_block->rows(); |
84 | 1 | if (profile != nullptr) { |
85 | 0 | update_counter(profile->rows_filtered_by_delete_conjunct, |
86 | 0 | rows_before_filter - rows_after_delete_filter); |
87 | 0 | } |
88 | 1 | } |
89 | | |
90 | 57 | size_t rows_after_filter = rows_after_delete_filter; |
91 | 57 | if (request != nullptr && rows_after_delete_filter > 0 && !request->conjuncts.empty()) { |
92 | 7 | { |
93 | 7 | SCOPED_TIMER(profile == nullptr ? nullptr : profile->conjunct_filter_time); |
94 | 7 | RETURN_IF_ERROR(VExprContext::filter_block(request->conjuncts, file_block, |
95 | 7 | file_block->columns())); |
96 | 7 | } |
97 | 7 | rows_after_filter = |
98 | 7 | file_block->columns() == 0 ? rows_after_delete_filter : file_block->rows(); |
99 | 7 | const auto rows_filtered_by_conjunct = rows_after_delete_filter - rows_after_filter; |
100 | 7 | if (profile != nullptr) { |
101 | 4 | update_counter(profile->rows_filtered_by_conjunct, rows_filtered_by_conjunct); |
102 | 4 | } |
103 | 7 | if (io_ctx != nullptr) { |
104 | 5 | io_ctx->predicate_filtered_rows += rows_filtered_by_conjunct; |
105 | 5 | } |
106 | 7 | } |
107 | 57 | *rows = rows_after_filter; |
108 | 57 | return Status::OK(); |
109 | 57 | } |
110 | | |
111 | | } // namespace doris::format |