1
|
package eu.dnetlib.data.collector.plugins.filesystem;
|
2
|
|
3
|
import java.io.IOException;
|
4
|
import java.nio.file.Files;
|
5
|
import java.nio.file.Path;
|
6
|
import java.nio.file.Paths;
|
7
|
import java.util.Iterator;
|
8
|
import java.util.Set;
|
9
|
|
10
|
import org.apache.commons.io.FilenameUtils;
|
11
|
import org.apache.commons.lang3.StringUtils;
|
12
|
import org.apache.commons.logging.Log;
|
13
|
import org.apache.commons.logging.LogFactory;
|
14
|
|
15
|
import com.google.common.collect.Iterators;
|
16
|
import com.google.common.collect.Sets;
|
17
|
|
18
|
/**
|
19
|
* Class enabling lazy & recursive iteration of a filesystem tree. The iterator iterates over file paths.
|
20
|
*
|
21
|
* @author Andrea
|
22
|
*
|
23
|
*/
|
24
|
public class FileSystemIterator implements Iterator<String> {
|
25
|
|
26
|
/** The logger */
|
27
|
private static final Log log = LogFactory.getLog(FileSystemIterator.class);
|
28
|
|
29
|
private Set<String> extensions = Sets.newHashSet();
|
30
|
private Iterator<Path> pathIterator;
|
31
|
private String current;
|
32
|
|
33
|
public FileSystemIterator(final String baseDir, final String extensions) {
|
34
|
if(StringUtils.isNotBlank(extensions)) {
|
35
|
this.extensions = Sets.newHashSet(extensions.split(","));
|
36
|
}
|
37
|
try {
|
38
|
this.pathIterator = Files.newDirectoryStream(Paths.get(baseDir)).iterator();
|
39
|
this.current = walkTillNext();
|
40
|
} catch (IOException e) {
|
41
|
log.error("Cannot initialize File System Iterator. Is this path correct? " + baseDir);
|
42
|
throw new RuntimeException("Filesystem collection error.", e);
|
43
|
}
|
44
|
}
|
45
|
|
46
|
@Override
|
47
|
public boolean hasNext() {
|
48
|
return current != null;
|
49
|
}
|
50
|
|
51
|
@Override
|
52
|
public synchronized String next() {
|
53
|
String pivot = new String(current);
|
54
|
current = walkTillNext();
|
55
|
log.debug("Returning: " + pivot);
|
56
|
return pivot;
|
57
|
}
|
58
|
|
59
|
@Override
|
60
|
public void remove() {}
|
61
|
|
62
|
/**
|
63
|
* Walk the filesystem recursively until it finds a candidate. Strategies: a) For any directory found during the walk, an iterator is
|
64
|
* built and concat to the main one; b) Any file is checked against admitted extensions
|
65
|
*
|
66
|
* @return the next element to be returned by next call of this.next()
|
67
|
*/
|
68
|
private synchronized String walkTillNext() {
|
69
|
while (pathIterator.hasNext()) {
|
70
|
Path nextFilePath = pathIterator.next();
|
71
|
if (Files.isDirectory(nextFilePath)) {
|
72
|
// concat
|
73
|
try {
|
74
|
pathIterator = Iterators.concat(pathIterator, Files.newDirectoryStream(nextFilePath).iterator());
|
75
|
log.debug("Adding folder iterator: " + nextFilePath.toString());
|
76
|
} catch (IOException e) {
|
77
|
log.error("Cannot create folder iterator! Is this path correct? " + nextFilePath.toString());
|
78
|
return null;
|
79
|
}
|
80
|
} else {
|
81
|
if (extensions.isEmpty() || extensions.contains(FilenameUtils.getExtension(nextFilePath.toString()))) {
|
82
|
log.debug("Returning: " + nextFilePath.toString());
|
83
|
return nextFilePath.toString();
|
84
|
}
|
85
|
}
|
86
|
}
|
87
|
return null;
|
88
|
}
|
89
|
}
|