be/src/format/generic_reader.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/generic_reader.h" |
19 | | |
20 | | #include <gen_cpp/PlanNodes_types.h> |
21 | | #include <gen_cpp/Types_types.h> |
22 | | |
23 | | namespace doris { |
24 | | |
25 | | Status GenericReader::init_column_descriptors( |
26 | | const TFileScanRangeParams& params, const TFileRangeDesc& range, |
27 | | const std::vector<ColumnDescriptor>& column_descs, const TupleDescriptor* tuple_descriptor, |
28 | | const RowDescriptor* row_descriptor, RuntimeState* state, bool fill_partition_from_path, |
29 | | const std::unordered_map<std::string, std::tuple<std::string, const SlotDescriptor*>>& |
30 | | partition_col_descs, |
31 | | std::unordered_map<std::string, VExprContextSPtr>* out_missing_col_descs, |
32 | | std::unordered_set<std::string>* out_missing_cols, |
33 | 24 | std::unordered_map<std::string, DataTypePtr>* out_name_to_col_type) { |
34 | | // 1. Get columns from reader (which columns exist in the file, which are missing) |
35 | 24 | RETURN_IF_ERROR(get_columns(out_name_to_col_type, out_missing_cols)); |
36 | | |
37 | | // 2. Build default value expressions for missing columns |
38 | 24 | std::unordered_map<std::string, VExprContextSPtr> col_default_value_ctx; |
39 | 141 | for (auto* slot_desc : tuple_descriptor->slots()) { |
40 | 141 | auto it = params.default_value_of_src_slot.find(slot_desc->id()); |
41 | 141 | if (it != params.default_value_of_src_slot.end()) { |
42 | 0 | VExprContextSPtr ctx; |
43 | 0 | if (!it->second.nodes.empty()) { |
44 | 0 | RETURN_IF_ERROR(VExpr::create_expr_tree(it->second, ctx)); |
45 | 0 | RETURN_IF_ERROR(ctx->prepare(state, *row_descriptor)); |
46 | 0 | RETURN_IF_ERROR(ctx->open(state)); |
47 | 0 | } |
48 | 0 | col_default_value_ctx.emplace(slot_desc->col_name(), ctx); |
49 | 0 | } |
50 | 141 | } |
51 | | |
52 | | // 3. Build missing_col_descs from missing columns |
53 | 24 | out_missing_col_descs->clear(); |
54 | 141 | for (const auto& col_desc : column_descs) { |
55 | 141 | if (!col_desc.is_file_slot) { |
56 | 0 | continue; |
57 | 0 | } |
58 | 141 | if (out_missing_cols->contains(col_desc.name) || |
59 | 141 | out_missing_cols->contains(to_lower(col_desc.name))) { |
60 | 0 | auto it = col_default_value_ctx.find(col_desc.name); |
61 | 0 | if (it != col_default_value_ctx.end()) { |
62 | 0 | out_missing_col_descs->emplace(col_desc.name, it->second); |
63 | 0 | } else { |
64 | | // No default value: fill with null |
65 | 0 | out_missing_col_descs->emplace(col_desc.name, nullptr); |
66 | 0 | } |
67 | 0 | } |
68 | 141 | } |
69 | | |
70 | | // 4. Call set_fill_columns on the reader |
71 | 24 | if (fill_partition_from_path) { |
72 | 24 | RETURN_IF_ERROR(set_fill_columns(partition_col_descs, *out_missing_col_descs)); |
73 | 24 | } else { |
74 | 0 | RETURN_IF_ERROR(set_fill_columns({}, *out_missing_col_descs)); |
75 | 0 | } |
76 | 24 | return Status::OK(); |
77 | 24 | } |
78 | | |
79 | | } // namespace doris |