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
|
package com.keuin.kbackupfabric.backup.incremental;
import com.keuin.kbackupfabric.backup.incremental.identifier.FileIdentifierProvider;
import com.keuin.kbackupfabric.backup.incremental.identifier.ObjectIdentifier;
import com.keuin.kbackupfabric.util.PrintUtil;
import java.io.File;
import java.io.IOException;
import java.nio.file.Files;
import java.nio.file.Path;
import java.util.*;
import java.util.concurrent.ConcurrentHashMap;
import java.util.concurrent.ConcurrentLinkedQueue;
import java.util.function.Consumer;
import java.util.function.Supplier;
import java.util.stream.Stream;
/**
* Incremental backup is implemented as git-like file collection.
* Files are called `objects`, the collection contains all files distinguished by their
* identifiers. Usually, identifier is the combination of hash and other short information (such as size and another hash).
* The identifier should use hashes that are strong enough, to prevent possible collisions.
*/
public class ObjectCollectionFactory<T extends ObjectIdentifier> {
private final FileIdentifierProvider<T> identifierFactory;
private final int threads;
private Exception exception = null; // fail in async
private final int minParallelProcessFileCountThreshold;
public ObjectCollectionFactory(FileIdentifierProvider<T> identifierFactory, int threads, int minParallelProcessFileCountThreshold) {
this.identifierFactory = identifierFactory;
this.threads = threads;
this.minParallelProcessFileCountThreshold = minParallelProcessFileCountThreshold;
if (threads <= 0)
throw new IllegalArgumentException("thread count must be positive.");
if (minParallelProcessFileCountThreshold < 0)
throw new IllegalArgumentException("minParallelProcessFileCountThreshold must not be negative.");
}
public ObjectCollection2 fromDirectory(File directory, Set<String> ignoredFiles) throws IOException {
final Map<String, ObjectCollection2> subCollections = new HashMap<>();
if (!Objects.requireNonNull(directory).isDirectory())
throw new IllegalArgumentException("given file is not a directory");
Set<File> files = new HashSet<>();
try (Stream<Path> walk = Files.walk(directory.toPath(), 1)) {
for (Iterator<Path> iter = walk.iterator(); iter.hasNext(); ) {
Path path = iter.next();
if (Files.isSameFile(path, directory.toPath()))
continue;
File file = path.toFile();
if (file.isDirectory()) {
subCollections.put(file.getName(), fromDirectory(file, ignoredFiles));
} else if (!ignoredFiles.contains(file.getName())) {
files.add(file); // add to the set to be processed
} else {
PrintUtil.info(String.format("Skipping file %s.", file.getName()));
}
}
}
final Set<ObjectElement> subFiles = ConcurrentHashMap.newKeySet(files.size());
// deal with all direct sub files
if (threads == 1 || files.size() < minParallelProcessFileCountThreshold) {
for (File file : files) {
subFiles.add(new ObjectElement(file.getName(), identifierFactory.fromFile(file)));
}
} else {
// use ParallelSupplier to process
ParallelSupplier<ObjectElement> parallelSupplier = new ParallelSupplier<>(subFiles::add, threads);
files.forEach(file -> parallelSupplier.addTask(() -> {
try {
return new ObjectElement(file.getName(), identifierFactory.fromFile(file));
} catch (IOException e) {
fail(e);
}
return null;
}));
parallelSupplier.process();
}
// check if any exception has been thrown in async workers.
synchronized (this) {
if (this.exception != null) {
if (exception instanceof IOException)
throw (IOException) exception;
else
throw new RuntimeException(exception);
}
}
return new ObjectCollection2(directory.getName(), subFiles, subCollections);
}
public ObjectCollection2 fromDirectory(File directory) throws IOException {
return fromDirectory(directory, Collections.emptySet());
}
private synchronized void fail(IOException e) {
this.exception = e;
}
/**
* A single-consumer, multiple-producer model.
*
* @param <Res> type of the resource to be produced and consumed.
*/
private static class ParallelSupplier<Res> {
private final Consumer<Res> consumer;
private final ConcurrentLinkedQueue<Supplier<Res>> taskList = new ConcurrentLinkedQueue<>();
private final int threads;
private final Set<ParallelWorker<Res>> workers = new HashSet<>();
public ParallelSupplier(Consumer<Res> consumer, int threads) {
this.consumer = consumer;
this.threads = threads;
}
public void addTask(Supplier<Res> valueSupplier) {
this.taskList.add(valueSupplier);
}
public void process() {
workers.clear();
for (int i = 0; i < threads; i++) {
ParallelWorker<Res> worker = new ParallelWorker<>(taskList, consumer, i);
workers.add(worker);
worker.start();
}
join(); // wait for all workers to exit before returning
}
private void join() {
while (true) {
int aliveCount = 0;
for (ParallelWorker<Res> worker : workers) {
try {
if (worker.isAlive()) {
++aliveCount;
worker.join();
}
} catch (InterruptedException ignored) {
}
}
if (aliveCount == 0)
return;
}
}
private static class ParallelWorker<V> extends Thread {
private final Queue<Supplier<V>> taskProvider;
private final Consumer<V> consumer;
public ParallelWorker(Queue<Supplier<V>> taskProvider, Consumer<V> consumer, int workerId) {
super("PutterWorker#" + workerId);
this.taskProvider = taskProvider;
this.consumer = consumer;
}
@Override
public void run() {
Supplier<V> supplier;
while ((supplier = taskProvider.poll()) != null) {
// here we do not let the consumer accept null productions
Optional.ofNullable(supplier.get()).ifPresent(consumer);
}
}
}
}
}
|
