public NewIndexer(String indexDir2) throws IOException {
// create the index
if (indexWriter2 == null) {
indexWriter2 =
new IndexWriter(
FSDirectory.open(new File("resultsList")),
new IndexWriterConfig(
Version.LUCENE_36,
new EnglishAnalyzer(Version.LUCENE_36, StandardAnalyzer.STOP_WORDS_SET)));
indexWriter2.deleteAll();
}
}
@Inject
@SuppressForbidden(reason = "System.out.*")
public NodeEnvironment(Settings settings, Environment environment) throws IOException {
super(settings);
this.addNodeId = settings.getAsBoolean(ADD_NODE_ID_TO_CUSTOM_PATH, true);
this.customPathsEnabled = settings.getAsBoolean(SETTING_CUSTOM_DATA_PATH_ENABLED, false);
if (!DiscoveryNode.nodeRequiresLocalStorage(settings)) {
nodePaths = null;
locks = null;
localNodeId = -1;
return;
}
final NodePath[] nodePaths = new NodePath[environment.dataWithClusterFiles().length];
final Lock[] locks = new Lock[nodePaths.length];
int localNodeId = -1;
IOException lastException = null;
int maxLocalStorageNodes = settings.getAsInt("node.max_local_storage_nodes", 50);
for (int possibleLockId = 0; possibleLockId < maxLocalStorageNodes; possibleLockId++) {
for (int dirIndex = 0; dirIndex < environment.dataWithClusterFiles().length; dirIndex++) {
Path dir =
environment
.dataWithClusterFiles()[dirIndex]
.resolve(NODES_FOLDER)
.resolve(Integer.toString(possibleLockId));
Files.createDirectories(dir);
try (Directory luceneDir = FSDirectory.open(dir, NativeFSLockFactory.INSTANCE)) {
logger.trace("obtaining node lock on {} ...", dir.toAbsolutePath());
try {
locks[dirIndex] = Lucene.acquireLock(luceneDir, NODE_LOCK_FILENAME, 0);
nodePaths[dirIndex] = new NodePath(dir, environment);
localNodeId = possibleLockId;
} catch (LockObtainFailedException ex) {
logger.trace("failed to obtain node lock on {}", dir.toAbsolutePath());
// release all the ones that were obtained up until now
releaseAndNullLocks(locks);
break;
}
} catch (IOException e) {
logger.trace("failed to obtain node lock on {}", e, dir.toAbsolutePath());
lastException = new IOException("failed to obtain lock on " + dir.toAbsolutePath(), e);
// release all the ones that were obtained up until now
releaseAndNullLocks(locks);
break;
}
}
if (locks[0] != null) {
// we found a lock, break
break;
}
}
if (locks[0] == null) {
throw new IllegalStateException(
"Failed to obtain node lock, is the following location writable?: "
+ Arrays.toString(environment.dataWithClusterFiles()),
lastException);
}
this.localNodeId = localNodeId;
this.locks = locks;
this.nodePaths = nodePaths;
if (logger.isDebugEnabled()) {
logger.debug("using node location [{}], local_node_id [{}]", nodePaths, localNodeId);
}
maybeLogPathDetails();
if (settings.getAsBoolean(SETTING_ENABLE_LUCENE_SEGMENT_INFOS_TRACE, false)) {
SegmentInfos.setInfoStream(System.out);
}
}
public static void main(String[] args) throws IOException {
IndexReader reader = null;
/*
* Opening the index first simplifies the processing of the
* rest of the command line arguments.
*/
for (int i = 0; i < args.length; i++) {
if (("-index".equals(args[i])) && ((i + 1) < args.length)) {
reader = DirectoryReader.open(FSDirectory.open(new File(args[i + 1])));
if (reader == null) {
System.err.println("Error: Can't open index " + args[i + 1]);
System.exit(1);
}
;
break;
}
;
}
;
if (reader == null) {
System.err.println(usage);
System.exit(1);
}
;
/*
* Process the command line arguments sequentially.
*/
for (int i = 0; i < args.length; i++) {
if ("-index".equals(args[i])) {
/*
* Handled in the previous loop, so just skip the argument.
*/
i++;
} else if ("-list-edocid".equals(args[i])) {
System.out.println("-list-edocid:");
if ((i + 1) >= args.length) {
System.out.println(usage);
break;
}
;
Document d = reader.document(Integer.parseInt(args[i + 1]));
System.out.println(
"Internal docid --> External docid: " + args[i + 1] + " --> " + d.get("externalId"));
i += 1;
} else if ("-list-docids".equals(args[i])) {
System.out.println("-list-docids:");
for (int j = 0; j < reader.numDocs(); j++) {
Document d = reader.document(j);
System.out.println("Internal --> external docid: " + j + " --> " + d.get("externalId"));
}
;
} else if ("-list-fields".equals(args[i])) {
Fields fields = MultiFields.getFields(reader);
System.out.print("\nNumber of fields: ");
if (fields == null) System.out.println("0");
else {
System.out.println(fields.size());
Iterator<String> is = fields.iterator();
while (is.hasNext()) {
System.out.println("\t" + is.next());
}
;
}
;
} else if ("-list-postings".equals(args[i])) {
if ((i + 2) >= args.length) {
System.out.println(usage);
break;
}
;
listPostings(reader, args[i + 1], args[i + 2], Integer.MAX_VALUE);
i += 2;
} else if ("-list-postings-sample".equals(args[i])) {
if ((i + 2) >= args.length) {
System.out.println(usage);
break;
}
;
listPostings(reader, args[i + 1], args[i + 2], 5);
i += 2;
} else if ("-list-stats".equals(args[i])) {
System.out.println("Corpus statistics:");
System.out.println("\tnumdocs\t\t" + reader.numDocs());
System.out.println(
"\turl:\t"
+ "\tnumdocs="
+ reader.getDocCount("url")
+ "\tsumTotalTF="
+ reader.getSumTotalTermFreq("url")
+ "\tavglen="
+ reader.getSumTotalTermFreq("url") / (float) reader.getDocCount("url"));
System.out.println(
"\tkeywords:"
+ "\tnumdocs="
+ reader.getDocCount("keywords")
+ "\tsumTotalTF="
+ reader.getSumTotalTermFreq("keywords")
+ "\tavglen="
+ reader.getSumTotalTermFreq("keywords") / (float) reader.getDocCount("keywords"));
System.out.println(
"\ttitle:\t"
+ "\tnumdocs="
+ reader.getDocCount("title")
+ "\tsumTotalTF="
+ reader.getSumTotalTermFreq("title")
+ "\tavglen="
+ reader.getSumTotalTermFreq("title") / (float) reader.getDocCount("title"));
System.out.println(
"\tbody:\t"
+ "\tnumdocs="
+ reader.getDocCount("body")
+ "\tsumTotalTF="
+ reader.getSumTotalTermFreq("body")
+ "\tavglen="
+ reader.getSumTotalTermFreq("body") / (float) reader.getDocCount("body"));
System.out.println(
"\tinlink:\t"
+ "\tnumdocs="
+ reader.getDocCount("inlink")
+ "\tsumTotalTF="
+ reader.getSumTotalTermFreq("inlink")
+ "\tavglen="
+ reader.getSumTotalTermFreq("inlink") / (float) reader.getDocCount("inlink"));
} else if ("-list-terms".equals(args[i])) {
if ((i + 1) >= args.length) {
System.out.println(usage);
break;
}
;
listTermDictionary(reader, args[i + 1]);
i += 1;
} else if ("-list-termvector".equals(args[i])) {
if ((i + 1) >= args.length) {
System.out.println(usage);
break;
}
;
listTermVectors(reader, args[i + 1]);
i += 1;
} else if ("-list-termvector-field".equals(args[i])) {
if ((i + 2) >= args.length) {
System.out.println(usage);
break;
}
;
listTermVectorField(reader, args[i + 1], args[i + 2]);
i += 2;
} else System.err.println("\nWarning: Unknown argument " + args[i] + " ignored.");
}
;
/*
* Close the index and exit gracefully.
*/
reader.close();
}