1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
|
/* Copyright 2015 The TensorFlow Authors. All Rights Reserved.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
==============================================================================*/
#include "tensorflow/core/lib/io/inputbuffer.h"
#include "tensorflow/core/lib/core/errors.h"
#include "tensorflow/core/platform/logging.h"
namespace tensorflow {
namespace io {
InputBuffer::InputBuffer(RandomAccessFile* file, size_t buffer_bytes)
: file_(file),
file_pos_(0),
size_(buffer_bytes),
buf_(new char[size_]),
pos_(buf_),
limit_(buf_) {}
InputBuffer::~InputBuffer() { delete[] buf_; }
Status InputBuffer::FillBuffer() {
StringPiece data;
Status s = file_->Read(file_pos_, size_, &data, buf_);
if (data.data() != buf_) {
memmove(buf_, data.data(), data.size());
}
pos_ = buf_;
limit_ = pos_ + data.size();
file_pos_ += data.size();
return s;
}
Status InputBuffer::ReadLine(string* result) {
result->clear();
Status s;
do {
size_t buf_remain = limit_ - pos_;
char* newline = static_cast<char*>(memchr(pos_, '\n', buf_remain));
if (newline != nullptr) {
size_t result_len = newline - pos_;
result->append(pos_, result_len);
pos_ = newline + 1;
if (!result->empty() && result->back() == '\r') {
result->resize(result->size() - 1);
}
return Status::OK();
}
if (buf_remain > 0) result->append(pos_, buf_remain);
// Get more data into buffer
s = FillBuffer();
DCHECK_EQ(pos_, buf_);
} while (limit_ != buf_);
if (!result->empty() && result->back() == '\r') {
result->resize(result->size() - 1);
}
if (errors::IsOutOfRange(s) && !result->empty()) {
return Status::OK();
}
return s;
}
Status InputBuffer::ReadNBytes(int64 bytes_to_read, string* result) {
result->clear();
if (bytes_to_read < 0) {
return errors::InvalidArgument("Can't read a negative number of bytes: ",
bytes_to_read);
}
result->resize(bytes_to_read);
size_t bytes_read = 0;
Status status = ReadNBytes(bytes_to_read, &(*result)[0], &bytes_read);
if (bytes_read < bytes_to_read) result->resize(bytes_read);
return status;
}
Status InputBuffer::ReadNBytes(int64 bytes_to_read, char* result,
size_t* bytes_read) {
if (bytes_to_read < 0) {
return errors::InvalidArgument("Can't read a negative number of bytes: ",
bytes_to_read);
}
Status status;
*bytes_read = 0;
while (*bytes_read < static_cast<size_t>(bytes_to_read)) {
if (pos_ == limit_) {
// Get more data into buffer.
status = FillBuffer();
if (limit_ == buf_) {
break;
}
}
// Do not go over the buffer boundary.
const int64 bytes_to_copy =
std::min<int64>(limit_ - pos_, bytes_to_read - *bytes_read);
// Copies buffered data into the destination.
memcpy(result + *bytes_read, pos_, bytes_to_copy);
pos_ += bytes_to_copy;
*bytes_read += bytes_to_copy;
}
if (errors::IsOutOfRange(status) &&
(*bytes_read == static_cast<size_t>(bytes_to_read))) {
return Status::OK();
}
return status;
}
Status InputBuffer::ReadVarint32Fallback(uint32* result) {
uint8 scratch = 0;
char* p = reinterpret_cast<char*>(&scratch);
size_t unused_bytes_read = 0;
*result = 0;
for (int shift = 0; shift <= 28; shift += 7) {
TF_RETURN_IF_ERROR(ReadNBytes(1, p, &unused_bytes_read));
*result |= (scratch & 127) << shift;
if (!(scratch & 128)) return Status::OK();
}
return errors::DataLoss("Stored data is too large to be a varint32.");
}
Status InputBuffer::SkipNBytes(int64 bytes_to_skip) {
if (bytes_to_skip < 0) {
return errors::InvalidArgument("Can only skip forward, not ",
bytes_to_skip);
}
int64 bytes_skipped = 0;
Status s;
while (bytes_skipped < bytes_to_skip) {
if (pos_ == limit_) {
// Get more data into buffer
s = FillBuffer();
if (limit_ == buf_) {
break;
}
}
const int64 bytes_to_advance =
std::min<int64>(limit_ - pos_, bytes_to_skip - bytes_skipped);
bytes_skipped += bytes_to_advance;
pos_ += bytes_to_advance;
}
if (errors::IsOutOfRange(s) && bytes_skipped == bytes_to_skip) {
return Status::OK();
}
return s;
}
Status InputBuffer::Seek(int64 position) {
if (position < 0) {
return errors::InvalidArgument("Seeking to a negative position: ",
position);
}
// Position of the buffer within file.
const int64 bufpos = file_pos_ - static_cast<int64>(limit_ - buf_);
if (position >= bufpos && position < file_pos_) {
// Seeks to somewhere inside the buffer.
pos_ = buf_ + (position - bufpos);
DCHECK(pos_ >= buf_ && pos_ < limit_);
} else {
// Seeks to somewhere outside. Discards the buffered data.
pos_ = limit_ = buf_;
file_pos_ = position;
}
return Status::OK();
}
} // namespace io
} // namespace tensorflow
|
