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
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
use std::io::{self, Read, Seek};
use std::sync::atomic::Ordering;
use std::sync::Arc;
use crate::constants::*;
use crate::core::record::Record;
use crate::error::{FeoxError, Result};
use crate::storage::format::get_format;
use super::FeoxStore;
impl FeoxStore {
/// Force flush all pending writes to disk.
///
/// In persistent mode, ensures all buffered writes are flushed to disk.
/// In memory-only mode, this is a no-op.
///
/// # Example
///
/// ```no_run
/// # use feoxdb::FeoxStore;
/// # fn main() -> feoxdb::Result<()> {
/// let store = FeoxStore::new(Some("/path/to/data.feox".to_string()))?;
/// store.insert(b"important", b"data")?;
/// store.flush_all(); // Ensure data is persisted
/// # Ok(())
/// # }
/// ```
pub fn flush_all(&self) {
if !self.memory_only {
// First flush the write buffer to ensure all data is written
if let Some(ref wb) = self.write_buffer {
let _ = wb.force_flush();
}
if let Some(ref disk_io) = self.disk_io {
// Update metadata with current stats
let mut metadata = self._metadata.write();
metadata.total_records = self.stats.record_count.load(Ordering::Relaxed) as u64;
metadata.total_size = self.stats.disk_usage.load(Ordering::Relaxed);
metadata.fragmentation = self.free_space.read().get_fragmentation();
metadata.update();
// Write metadata
let _ = disk_io.write().write_metadata(metadata.as_bytes());
let _ = disk_io.write().flush();
}
}
}
pub(super) fn load_value_from_disk(&self, record: &Record) -> Result<Vec<u8>> {
let sector = record.sector.load(Ordering::Acquire);
if self.memory_only || sector == 0 {
return Err(FeoxError::InvalidRecord);
}
// Get the appropriate format handler
let metadata_version = self._metadata.read().version;
let format = get_format(metadata_version);
// Calculate how many sectors we need to read
let total_size = format.total_size(record.key.len(), record.value_len);
let sectors_needed = total_size.div_ceil(FEOX_BLOCK_SIZE);
// Read the sectors
let disk_io = self
.disk_io
.as_ref()
.ok_or_else(|| {
FeoxError::IoError(io::Error::new(
io::ErrorKind::NotFound,
"No disk IO available",
))
})?
.read();
let data = disk_io.read_sectors_sync(sector, sectors_needed as u64)?;
// Use format to get the value offset
let offset = format.value_offset(record.key.len());
if offset + record.value_len > data.len() {
return Err(FeoxError::InvalidRecord);
}
Ok(data[offset..offset + record.value_len].to_vec())
}
pub(super) fn open_device(
&mut self,
device_path: &Option<String>,
file_size: Option<u64>,
) -> Result<()> {
if let Some(path) = device_path {
// Open the device/file
use std::fs::OpenOptions;
#[cfg(target_os = "linux")]
use std::os::unix::fs::OpenOptionsExt;
#[cfg(unix)]
let (file, use_direct_io) = if std::path::Path::new("/.dockerenv").exists() {
let file = OpenOptions::new()
.read(true)
.write(true)
.create(true)
.truncate(false)
.open(path)
.map_err(FeoxError::IoError)?;
(file, false) // Don't use O_DIRECT in Docker
} else {
// Try with O_DIRECT on Linux, fall back without it on other Unix systems
#[cfg(target_os = "linux")]
{
// Try to open with O_DIRECT first
match OpenOptions::new()
.read(true)
.write(true)
.create(true)
.truncate(false)
.custom_flags(libc::O_DIRECT)
.open(path)
{
Ok(file) => (file, true), // Successfully opened with O_DIRECT
Err(_) => {
// Fallback to regular open
let file = OpenOptions::new()
.read(true)
.write(true)
.create(true)
.truncate(false)
.open(path)
.map_err(FeoxError::IoError)?;
(file, false)
}
}
}
#[cfg(not(target_os = "linux"))]
{
let file = OpenOptions::new()
.read(true)
.write(true)
.create(true)
.truncate(false)
.open(path)
.map_err(FeoxError::IoError)?;
(file, false) // O_DIRECT not supported on this platform
}
};
#[cfg(not(unix))]
let file = OpenOptions::new()
.read(true)
.write(true)
.create(true)
.truncate(false)
.open(path)
.map_err(FeoxError::IoError)?;
// Get file size
let metadata = file.metadata().map_err(FeoxError::IoError)?;
self.device_size = metadata.len();
// Track whether this is a newly created file
let was_newly_created = self.device_size == 0;
if was_newly_created {
// New empty file - set configured size or default and initialize free space
let target_size = file_size.unwrap_or(DEFAULT_DEVICE_SIZE);
file.set_len(target_size).map_err(FeoxError::IoError)?;
self.device_size = target_size;
// Initialize free space manager with all space free
self.free_space.write().initialize(self.device_size)?;
let mut metadata = self._metadata.write();
metadata.device_size = self.device_size;
metadata.update();
} else {
// Existing file - check if it's empty
// If empty, initialize free space; otherwise it will be rebuilt during scan
let is_empty_file = {
let mut temp_file = file.try_clone().map_err(FeoxError::IoError)?;
temp_file
.metadata()
.map(|m| {
// Check if file is all zeros
if m.len() > 0 {
let mut buffer = vec![0u8; std::cmp::min(4096, m.len() as usize)];
temp_file.seek(std::io::SeekFrom::Start(0)).ok();
temp_file.read_exact(&mut buffer).ok();
buffer.iter().all(|&b| b == 0)
} else {
false
}
})
.unwrap_or(false)
};
if is_empty_file {
// Empty pre-created file - initialize free space like a new file
self.free_space.write().initialize(self.device_size)?;
} else {
// Existing file with data - free space will be rebuilt during scan
self.free_space.write().set_device_size(self.device_size);
}
}
#[cfg(unix)]
{
use std::os::unix::io::AsRawFd;
let file_arc = Arc::new(file);
let fd = file_arc.as_raw_fd();
self.device_fd = Some(fd);
// Store a clone of the file to keep it alive
self.device_file = Some(file_arc.as_ref().try_clone().map_err(FeoxError::IoError)?);
let disk_io = crate::storage::io::DiskIO::new(file_arc, use_direct_io)?;
self.disk_io = Some(Arc::new(parking_lot::RwLock::new(disk_io)));
}
#[cfg(not(unix))]
{
// Store a clone of the file to keep it alive
self.device_file = Some(file.try_clone().map_err(FeoxError::IoError)?);
let disk_io = crate::storage::io::DiskIO::new_from_file(file)?;
self.disk_io = Some(Arc::new(parking_lot::RwLock::new(disk_io)));
}
let disk_io = self.disk_io.as_ref().unwrap().read();
// Read metadata from existing files (not newly created ones)
if !was_newly_created {
if let Ok(metadata_bytes) = disk_io.read_metadata() {
if let Some(loaded_metadata) =
crate::storage::metadata::Metadata::from_bytes(&metadata_bytes)
{
// Initialize stats from metadata
self.stats
.disk_usage
.store(loaded_metadata.total_size, Ordering::Relaxed);
*self._metadata.write() = loaded_metadata;
}
}
}
}
Ok(())
}
}
impl Drop for FeoxStore {
fn drop(&mut self) {
// Stop TTL sweeper if running
if let Some(mut sweeper) = self.ttl_sweeper.write().take() {
sweeper.stop();
}
// Signal shutdown to write buffer workers
if let Some(ref wb) = self.write_buffer {
wb.initiate_shutdown();
}
// Write metadata directly without using the write buffer
if !self.memory_only {
if let Some(ref disk_io) = self.disk_io {
// Update metadata with current stats
let mut metadata = self._metadata.write();
metadata.total_records = self.stats.record_count.load(Ordering::Relaxed) as u64;
metadata.total_size = self.stats.disk_usage.load(Ordering::Relaxed);
metadata.fragmentation = self.free_space.read().get_fragmentation();
metadata.update();
// Write metadata
let _ = disk_io.write().write_metadata(metadata.as_bytes());
let _ = disk_io.write().flush();
}
}
// Take ownership of write_buffer to properly shut it down
if let Some(wb_arc) = self.write_buffer.take() {
// Try to get mutable access if we're the only owner
if let Ok(wb) = Arc::try_unwrap(wb_arc) {
// We own it exclusively, can call complete_shutdown
let mut wb_mut = wb;
wb_mut.complete_shutdown();
}
// If we can't get exclusive access, workers are already shutting down via initiate_shutdown
}
// Now it's safe to shutdown disk I/O since workers have exited
if let Some(ref disk_io) = self.disk_io {
disk_io.write().shutdown();
}
}
}