Coverage Report

Created: 2026-07-12 15:11

next uncovered line (L), next uncovered region (R), next uncovered branch (B)
be/src/format/table/paimon_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
18
#pragma once
19
20
#include <gen_cpp/PlanNodes_types.h>
21
22
#include <memory>
23
#include <string>
24
#include <utility>
25
#include <vector>
26
27
#include "common/status.h"
28
#include "format/orc/vorc_reader.h"
29
#include "format/parquet/vparquet_reader.h"
30
#include "format/table/deletion_vector.h"
31
#include "format/table/table_schema_change_helper.h"
32
33
namespace doris {
34
class ShardedKVCache;
35
36
std::string build_paimon_deletion_vector_cache_key(const TPaimonDeletionFileDesc& deletion_file);
37
38
Status decode_paimon_deletion_vector_buffer(const char* buf, size_t buffer_size,
39
                                            DeletionVector* deletion_vector);
40
41
// PaimonOrcReader: directly inherits OrcReader (no composition wrapping).
42
// Schema mapping in on_before_init_reader, deletion vector reading in on_after_init_reader.
43
class PaimonOrcReader final : public OrcReader, public TableSchemaChangeHelper {
44
public:
45
    ENABLE_FACTORY_CREATOR(PaimonOrcReader);
46
    PaimonOrcReader(RuntimeProfile* profile, RuntimeState* state,
47
                    const TFileScanRangeParams& params, const TFileRangeDesc& range,
48
                    size_t batch_size, const std::string& ctz, ShardedKVCache* kv_cache,
49
                    io::IOContext* io_ctx, FileMetaCache* meta_cache = nullptr,
50
                    bool enable_lazy_mat = true)
51
1
            : OrcReader(profile, state, params, range, batch_size, ctz, io_ctx, meta_cache,
52
1
                        enable_lazy_mat),
53
1
              _kv_cache(kv_cache) {
54
1
        _init_paimon_profile();
55
1
    }
56
    PaimonOrcReader(RuntimeProfile* profile, RuntimeState* state,
57
                    const TFileScanRangeParams& params, const TFileRangeDesc& range,
58
                    size_t batch_size, const std::string& ctz, ShardedKVCache* kv_cache,
59
                    std::shared_ptr<io::IOContext> io_ctx_holder,
60
                    FileMetaCache* meta_cache = nullptr, bool enable_lazy_mat = true)
61
0
            : OrcReader(profile, state, params, range, batch_size, ctz, std::move(io_ctx_holder),
62
0
                        meta_cache, enable_lazy_mat),
63
0
              _kv_cache(kv_cache) {
64
0
        _init_paimon_profile();
65
0
    }
66
1
    ~PaimonOrcReader() final = default;
67
68
1
    Status TEST_init_deletion_vector() { return _init_deletion_vector(); }
69
70
protected:
71
    Status on_before_init_reader(ReaderInitContext* ctx) override;
72
73
    Status on_after_init_reader(ReaderInitContext* /*ctx*/) override;
74
75
private:
76
    void _init_paimon_profile();
77
    Status _init_deletion_vector();
78
79
    struct PaimonProfile {
80
        RuntimeProfile::Counter* num_delete_rows = nullptr;
81
        RuntimeProfile::Counter* delete_files_read_time = nullptr;
82
        RuntimeProfile::Counter* parse_deletion_vector_time = nullptr;
83
        RuntimeProfile::Counter* decoded_cache_hit_count = nullptr;
84
        RuntimeProfile::Counter* decoded_cache_miss_count = nullptr;
85
        RuntimeProfile::Counter* file_cache_hit_count = nullptr;
86
        RuntimeProfile::Counter* file_cache_miss_count = nullptr;
87
        RuntimeProfile::Counter* file_cache_peer_read_count = nullptr;
88
    };
89
90
    const DeletionVector* _deletion_vector = nullptr;
91
    ShardedKVCache* _kv_cache;
92
    PaimonProfile _paimon_profile;
93
};
94
95
// PaimonParquetReader: directly inherits ParquetReader (no composition wrapping).
96
class PaimonParquetReader final : public ParquetReader, public TableSchemaChangeHelper {
97
public:
98
    ENABLE_FACTORY_CREATOR(PaimonParquetReader);
99
    PaimonParquetReader(RuntimeProfile* profile, const TFileScanRangeParams& params,
100
                        const TFileRangeDesc& range, size_t batch_size, const cctz::time_zone* ctz,
101
                        ShardedKVCache* kv_cache, io::IOContext* io_ctx, RuntimeState* state,
102
                        FileMetaCache* meta_cache = nullptr, bool enable_lazy_mat = true)
103
1
            : ParquetReader(profile, params, range, batch_size, ctz, io_ctx, state, meta_cache,
104
1
                            enable_lazy_mat),
105
1
              _kv_cache(kv_cache) {
106
1
        _init_paimon_profile();
107
1
    }
108
    PaimonParquetReader(RuntimeProfile* profile, const TFileScanRangeParams& params,
109
                        const TFileRangeDesc& range, size_t batch_size, const cctz::time_zone* ctz,
110
                        ShardedKVCache* kv_cache, std::shared_ptr<io::IOContext> io_ctx_holder,
111
                        RuntimeState* state, FileMetaCache* meta_cache = nullptr,
112
                        bool enable_lazy_mat = true)
113
0
            : ParquetReader(profile, params, range, batch_size, ctz, std::move(io_ctx_holder),
114
0
                            state, meta_cache, enable_lazy_mat),
115
0
              _kv_cache(kv_cache) {
116
0
        _init_paimon_profile();
117
0
    }
118
1
    ~PaimonParquetReader() final = default;
119
120
1
    Status TEST_init_deletion_vector() { return _init_deletion_vector(); }
121
122
protected:
123
    Status on_before_init_reader(ReaderInitContext* ctx) override;
124
125
    Status on_after_init_reader(ReaderInitContext* /*ctx*/) override;
126
127
private:
128
    void _init_paimon_profile();
129
    Status _init_deletion_vector();
130
131
    struct PaimonProfile {
132
        RuntimeProfile::Counter* num_delete_rows = nullptr;
133
        RuntimeProfile::Counter* delete_files_read_time = nullptr;
134
        RuntimeProfile::Counter* parse_deletion_vector_time = nullptr;
135
        RuntimeProfile::Counter* decoded_cache_hit_count = nullptr;
136
        RuntimeProfile::Counter* decoded_cache_miss_count = nullptr;
137
        RuntimeProfile::Counter* file_cache_hit_count = nullptr;
138
        RuntimeProfile::Counter* file_cache_miss_count = nullptr;
139
        RuntimeProfile::Counter* file_cache_peer_read_count = nullptr;
140
    };
141
142
    const DeletionVector* _deletion_vector = nullptr;
143
    ShardedKVCache* _kv_cache;
144
    PaimonProfile _paimon_profile;
145
};
146
147
} // namespace doris