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// DEPRECATED IO INTERFACES: We have transitioned to using the Apache
19// Arrow file input and output abstract interfaces defined in
20// arrow/io/interfaces.h. These legacy interfaces are being preserved
21// through a wrapper layer for one to two releases
22
23#pragma once
24
25#include <cstdint>
26#include <memory>
27
28#include "parquet/platform.h"
29
30namespace parquet {
31
32class PARQUET_EXPORT FileInterface {
33 public:
34 virtual ~FileInterface() = default;
35
36 // Close the file
37 virtual void Close() = 0;
38
39 // Return the current position in the file relative to the start
40 virtual int64_t Tell() = 0;
41};
42
43/// It is the responsibility of implementations to mind threadsafety of shared
44/// resources
45class PARQUET_EXPORT RandomAccessSource : virtual public FileInterface {
46 public:
47 virtual ~RandomAccessSource() = default;
48
49 virtual int64_t Size() const = 0;
50
51 // Returns bytes read
52 virtual int64_t Read(int64_t nbytes, uint8_t* out) = 0;
53
54 virtual std::shared_ptr<Buffer> Read(int64_t nbytes) = 0;
55
56 virtual std::shared_ptr<Buffer> ReadAt(int64_t position, int64_t nbytes) = 0;
57
58 /// Returns bytes read
59 virtual int64_t ReadAt(int64_t position, int64_t nbytes, uint8_t* out) = 0;
60};
61
62class PARQUET_EXPORT OutputStream : virtual public FileInterface {
63 public:
64 virtual ~OutputStream() = default;
65
66 // Copy bytes into the output stream
67 virtual void Write(const uint8_t* data, int64_t length) = 0;
68};
69
70// ----------------------------------------------------------------------
71// Wrapper classes
72
73class PARQUET_EXPORT ParquetInputWrapper : public ::arrow::io::RandomAccessFile {
74 public:
75 explicit ParquetInputWrapper(std::unique_ptr<RandomAccessSource> source);
76 explicit ParquetInputWrapper(RandomAccessSource* source);
77
78 ~ParquetInputWrapper() override;
79
80 // FileInterface
81 ::arrow::Status Close() override;
82 ::arrow::Status Tell(int64_t* position) const override;
83 bool closed() const override;
84
85 // Seekable
86 ::arrow::Status Seek(int64_t position) override;
87
88 // InputStream / RandomAccessFile
89 ::arrow::Status Read(int64_t nbytes, int64_t* bytes_read, void* out) override;
90 ::arrow::Status Read(int64_t nbytes, std::shared_ptr<Buffer>* out) override;
91 ::arrow::Status ReadAt(int64_t position, int64_t nbytes,
92 std::shared_ptr<Buffer>* out) override;
93 ::arrow::Status GetSize(int64_t* size) override;
94
95 private:
96 std::unique_ptr<RandomAccessSource> owned_source_;
97 RandomAccessSource* source_;
98 bool closed_;
99};
100
101class PARQUET_EXPORT ParquetOutputWrapper : public ::arrow::io::OutputStream {
102 public:
103 explicit ParquetOutputWrapper(const std::shared_ptr<::parquet::OutputStream>& sink);
104 explicit ParquetOutputWrapper(std::unique_ptr<::parquet::OutputStream> sink);
105 explicit ParquetOutputWrapper(::parquet::OutputStream* sink);
106
107 ~ParquetOutputWrapper() override;
108
109 // FileInterface
110 ::arrow::Status Close() override;
111 ::arrow::Status Tell(int64_t* position) const override;
112 bool closed() const override;
113
114 // Writable
115 ::arrow::Status Write(const void* data, int64_t nbytes) override;
116
117 private:
118 std::unique_ptr<::parquet::OutputStream> owned_sink_;
119 std::shared_ptr<::parquet::OutputStream> shared_sink_;
120 ::parquet::OutputStream* sink_;
121 bool closed_;
122};
123
124} // namespace parquet
125