Coverage Report

Created: 2026-03-12 16:03

next uncovered line (L), next uncovered region (R), next uncovered branch (B)
be/src/format/table/hudi_reader.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
#pragma once
18
#include <memory>
19
#include <vector>
20
21
#include "format/orc/vorc_reader.h"
22
#include "format/parquet/vparquet_reader.h"
23
#include "format/table/table_format_reader.h"
24
namespace doris {
25
#include "common/compile_check_begin.h"
26
class HudiReader : public TableFormatReader, public TableSchemaChangeHelper {
27
public:
28
    HudiReader(std::unique_ptr<GenericReader> file_format_reader, RuntimeProfile* profile,
29
               RuntimeState* state, const TFileScanRangeParams& params, const TFileRangeDesc& range,
30
               io::IOContext* io_ctx, FileMetaCache* meta_cache)
31
0
            : TableFormatReader(std::move(file_format_reader), state, profile, params, range,
32
0
                                io_ctx, meta_cache) {};
33
34
0
    ~HudiReader() override = default;
35
36
    Status get_next_block_inner(Block* block, size_t* read_rows, bool* eof) final;
37
38
0
    Status init_row_filters() final { return Status::OK(); };
39
};
40
41
class HudiParquetReader final : public HudiReader {
42
public:
43
    ENABLE_FACTORY_CREATOR(HudiParquetReader);
44
    HudiParquetReader(std::unique_ptr<GenericReader> file_format_reader, RuntimeProfile* profile,
45
                      RuntimeState* state, const TFileScanRangeParams& params,
46
                      const TFileRangeDesc& range, io::IOContext* io_ctx, FileMetaCache* meta_cache)
47
0
            : HudiReader(std::move(file_format_reader), profile, state, params, range, io_ctx,
48
0
                         meta_cache) {};
49
    ~HudiParquetReader() final = default;
50
51
    Status init_reader(
52
            const std::vector<std::string>& read_table_col_names,
53
            std::unordered_map<std::string, uint32_t>* col_name_to_block_idx,
54
            const VExprContextSPtrs& conjuncts,
55
            phmap::flat_hash_map<int, std::vector<std::shared_ptr<ColumnPredicate>>>&
56
                    slot_id_to_predicates,
57
            const TupleDescriptor* tuple_descriptor, const RowDescriptor* row_descriptor,
58
            const std::unordered_map<std::string, int>* colname_to_slot_id,
59
            const VExprContextSPtrs* not_single_slot_filter_conjuncts,
60
            const std::unordered_map<int, VExprContextSPtrs>* slot_id_to_filter_conjuncts);
61
};
62
63
class HudiOrcReader final : public HudiReader {
64
public:
65
    ENABLE_FACTORY_CREATOR(HudiOrcReader);
66
    HudiOrcReader(std::unique_ptr<GenericReader> file_format_reader, RuntimeProfile* profile,
67
                  RuntimeState* state, const TFileScanRangeParams& params,
68
                  const TFileRangeDesc& range, io::IOContext* io_ctx, FileMetaCache* meta_cache)
69
0
            : HudiReader(std::move(file_format_reader), profile, state, params, range, io_ctx,
70
0
                         meta_cache) {};
71
    ~HudiOrcReader() final = default;
72
73
    Status init_reader(
74
            const std::vector<std::string>& read_table_col_names,
75
            std::unordered_map<std::string, uint32_t>* col_name_to_block_idx,
76
            const VExprContextSPtrs& conjuncts, const TupleDescriptor* tuple_descriptor,
77
            const RowDescriptor* row_descriptor,
78
            const VExprContextSPtrs* not_single_slot_filter_conjuncts,
79
0
            const std::unordered_map<int, VExprContextSPtrs>* slot_id_to_filter_conjuncts) {
80
0
        auto* orc_reader = static_cast<OrcReader*>(_file_format_reader.get());
81
0
        const orc::Type* orc_type_ptr = nullptr;
82
0
        RETURN_IF_ERROR(orc_reader->get_file_type(&orc_type_ptr));
83
0
        RETURN_IF_ERROR(gen_table_info_node_by_field_id(
84
0
                _params, _range.table_format_params.hudi_params.schema_id, tuple_descriptor,
85
0
                orc_type_ptr));
86
87
0
        return orc_reader->init_reader(&read_table_col_names, col_name_to_block_idx, conjuncts,
88
0
                                       false, tuple_descriptor, row_descriptor,
89
0
                                       not_single_slot_filter_conjuncts,
90
0
                                       slot_id_to_filter_conjuncts, table_info_node_ptr);
91
0
    }
92
};
93
94
#include "common/compile_check_end.h"
95
} // namespace doris