List of usage examples for org.apache.lucene.index DirectoryReader indexExists
public static boolean indexExists(Directory directory) throws IOException
true if an index likely exists at the specified directory. From source file:org.opensolaris.opengrok.index.IndexDatabase.java
License:Open Source License
/** * Get an indexReader for the Index database where a given file * * @param path the file to get the database for * @return The index database where the file should be located or null if it * cannot be located./*www . j a v a 2s . c o m*/ */ public static IndexReader getIndexReader(String path) { IndexReader ret = null; RuntimeEnvironment env = RuntimeEnvironment.getInstance(); File indexDir = new File(env.getDataRootFile(), INDEX_DIR); if (env.hasProjects()) { Project p = Project.getProject(path); if (p == null) { return null; } indexDir = new File(indexDir, p.getPath()); } try { FSDirectory fdir = FSDirectory.open(indexDir, NoLockFactory.getNoLockFactory()); if (indexDir.exists() && DirectoryReader.indexExists(fdir)) { ret = DirectoryReader.open(fdir); } } catch (Exception ex) { log.log(Level.SEVERE, "Failed to open index: {0}", indexDir.getAbsolutePath()); log.log(Level.FINE, "Stack Trace: ", ex); } return ret; }
From source file:org.pageseeder.flint.lucene.LuceneIndexIO.java
License:Apache License
private void open() throws IOException { // create it? boolean createIt = !DirectoryReader.indexExists(this._directory); // read only? boolean readonly = isReadOnly(this._directory); if (readonly) { this._writer = null; this._reader = new ReaderManager(this._directory); this._searcher = new SearcherManager(this._directory, FACTORY); } else {/*from w ww. j a v a 2s.c o m*/ // create writer IndexWriterConfig config = new IndexWriterConfig(this._analyzer); ConcurrentMergeScheduler merger = new ConcurrentMergeScheduler(); // merger.setMaxMergesAndThreads(maxMergeCount, maxThreadCount); config.setMergeScheduler(merger); if (createIt) config.setOpenMode(OpenMode.CREATE); this._writer = new IndexWriter(this._directory, config); if (createIt) this._writer.commit(); boolean applyAllDeletes = true; // create searcher this._searcher = new SearcherManager(this._writer, applyAllDeletes, FACTORY); // create reader this._reader = new ReaderManager(this._writer, applyAllDeletes); } // add it to list of opened indexes OpenIndexManager.add(this); // set state to clean state(State.CLEAN); }
From source file:org.pageseeder.flint.lucene.LuceneLocalIndex.java
License:Apache License
/** * Indicates whether the folder exists and is a valid Lucene index. * * <p>This method uses the Lucene {@link IndexReader} to check whether the folder corresponds to a valid Lucene Index. * * @param location the folder where the index is located. * @return <code>true</code> if the folder exists and is a Lucene index; <code>false</code> otherwise */// w w w . j a v a 2s . c o m public static boolean exists(File location) { if (!location.exists() || !location.isDirectory()) return false; boolean exists = false; // Get the last modified from the index try { Directory directory = FSDirectory.open(location.toPath()); exists = DirectoryReader.indexExists(directory); directory.close(); } catch (IOException ex) { LOGGER.error("Unable to check if local index exists", ex); } return exists; }
From source file:org.tightblog.service.LuceneIndexer.java
License:Apache License
/** * Initialize the Lucene indexer./*from ww w . ja va2 s . co m*/ */ public void initialize() { // only initialize the index if search is enabled if (searchEnabled) { boolean indexNeedsCreating = false; try { // existing indexConsistencyMarker means this.shutdown() wasn't called at last app shutdown if (indexConsistencyMarker.exists()) { log.info("Index was not closed properly with last shutdown; will be rebuilt"); indexNeedsCreating = true; } else { // see if index directory exists, if not create File testIndexDir = new File(indexDir); if (!testIndexDir.exists()) { if (testIndexDir.mkdirs()) { indexNeedsCreating = true; log.info("Index folder path {} created", testIndexDir.getAbsolutePath()); } else { throw new IOException("Folder path " + testIndexDir.getAbsolutePath() + " could not be " + "created (file permission rights?)"); } } else { // OK, index directory exists, see if Lucene index exists within it if (!DirectoryReader.indexExists(getIndexDirectory())) { log.info("Lucene index not detected, will create"); indexNeedsCreating = true; } else { log.info("Lucene search index already available and ready for use."); } } } if (indexNeedsCreating) { log.info("Generating Lucene index in the background..."); // deletes index consistency marker if it exists createIndex(getFSDirectory(true)); // create index consistency marker for next app shutdown if (!indexConsistencyMarker.createNewFile()) { throw new IOException("Could not create index consistency marker " + indexConsistencyMarker.getAbsolutePath() + " (file permission rights?)"); } rebuildWeblogIndex(); } reader = DirectoryReader.open(getIndexDirectory()); } catch (IOException e) { log.error("Could not create index, searching will be deactivated.", e); searchEnabled = false; } } }
From source file:org.ujmp.lucene.LuceneMap.java
License:Open Source License
private synchronized IndexWriter getIndexWriter() { try {/* w ww .ja v a 2 s .co m*/ if (!readOnly && indexSearcher != null && indexSearcher.getIndexReader().getRefCount() > 0) { indexSearcher.getIndexReader().close(); indexSearcher = null; } if (indexWriter == null) { if (DirectoryReader.indexExists(getDirectory())) { if (!readOnly) { if (IndexWriter.isLocked(getDirectory())) { IndexWriter.unlock(getDirectory()); } indexWriter = new IndexWriter(getDirectory(), new IndexWriterConfig(Version.LUCENE_47, getAnalyzer())); } } else { if (!readOnly) { indexWriter = new IndexWriter(getDirectory(), new IndexWriterConfig(Version.LUCENE_47, getAnalyzer())); } } } return indexWriter; } catch (Exception e) { throw new RuntimeException("could not prepare writher", e); } }
From source file:org.ujmp.lucene.LuceneMap.java
License:Open Source License
private synchronized IndexSearcher getIndexSearcher() { try {/*from www .j a v a 2 s .c om*/ if (!DirectoryReader.indexExists(getDirectory())) { getIndexWriter(); } if (indexSearcher == null) { indexSearcher = new IndexSearcher(DirectoryReader.open(getIndexWriter(), true)); } return indexSearcher; } catch (Exception e) { throw new RuntimeException("could not prepare reader", e); } }
From source file:perf.SearchPerfTest.java
License:Apache License
private static void _main(String[] clArgs) throws Exception { // args: dirImpl indexPath numThread numIterPerThread // eg java SearchPerfTest /path/to/index 4 100 final Args args = new Args(clArgs); Directory dir0;/*from w ww . j a va2 s . com*/ final String dirPath = args.getString("-indexPath") + "/index"; final String dirImpl = args.getString("-dirImpl"); OpenDirectory od = OpenDirectory.get(dirImpl); /* } else if (dirImpl.equals("NativePosixMMapDirectory")) { dir0 = new NativePosixMMapDirectory(new File(dirPath)); ramDir = null; if (doFacets) { facetsDir = new NativePosixMMapDirectory(new File(facetsDirPath)); } } else if (dirImpl.equals("CachingDirWrapper")) { dir0 = new CachingRAMDirectory(new MMapDirectory(new File(dirPath))); ramDir = null; } else if (dirImpl.equals("RAMExceptDirectPostingsDirectory")) { // Load only non-postings files into RAMDir (assumes // Lucene40PF is the wrapped PF): Set<String> postingsExtensions = new HashSet<String>(); postingsExtensions.add("frq"); postingsExtensions.add("prx"); postingsExtensions.add("tip"); postingsExtensions.add("tim"); ramDir = new RAMDirectory(); Directory fsDir = new MMapDirectory(new File(dirPath)); for (String file : fsDir.listAll()) { int idx = file.indexOf('.'); if (idx != -1 && postingsExtensions.contains(file.substring(idx+1, file.length()))) { continue; } fsDir.copy(ramDir, file, file, IOContext.READ); } dir0 = new FileSwitchDirectory(postingsExtensions, fsDir, ramDir, true); if (doFacets) { facetsDir = new RAMDirectory(new SimpleFSDirectory(new File(facetsDirPath)), IOContext.READ); } */ final RAMDirectory ramDir; dir0 = od.open(Paths.get(dirPath)); if (dir0 instanceof RAMDirectory) { ramDir = (RAMDirectory) dir0; } else { ramDir = null; } // TODO: NativeUnixDir? final String analyzer = args.getString("-analyzer"); final String tasksFile = args.getString("-taskSource"); final int searchThreadCount = args.getInt("-searchThreadCount"); final String fieldName = args.getString("-field"); final boolean printHeap = args.getFlag("-printHeap"); final boolean doPKLookup = args.getFlag("-pk"); final int topN = args.getInt("-topN"); final boolean doStoredLoads = args.getFlag("-loadStoredFields"); // Used to choose which random subset of tasks we will // run, to generate the PKLookup tasks, and to generate // any random pct filters: final long staticRandomSeed = args.getLong("-staticSeed"); // Used to shuffle the random subset of tasks: final long randomSeed = args.getLong("-seed"); // TODO: this could be way better. final String similarity = args.getString("-similarity"); // now reflect final Class<? extends Similarity> simClazz = Class .forName("org.apache.lucene.search.similarities." + similarity).asSubclass(Similarity.class); final Similarity sim = simClazz.newInstance(); System.out.println("Using dir impl " + dir0.getClass().getName()); System.out.println("Analyzer " + analyzer); System.out.println("Similarity " + similarity); System.out.println("Search thread count " + searchThreadCount); System.out.println("topN " + topN); System.out.println("JVM " + (Constants.JRE_IS_64BIT ? "is" : "is not") + " 64bit"); System.out.println("Pointer is " + RamUsageEstimator.NUM_BYTES_OBJECT_REF + " bytes"); final Analyzer a; if (analyzer.equals("EnglishAnalyzer")) { a = new EnglishAnalyzer(); } else if (analyzer.equals("ClassicAnalyzer")) { a = new ClassicAnalyzer(); } else if (analyzer.equals("StandardAnalyzer")) { a = new StandardAnalyzer(); } else if (analyzer.equals("StandardAnalyzerNoStopWords")) { a = new StandardAnalyzer(CharArraySet.EMPTY_SET); } else if (analyzer.equals("ShingleStandardAnalyzer")) { a = new ShingleAnalyzerWrapper(new StandardAnalyzer(CharArraySet.EMPTY_SET), 2, 2, ShingleFilter.DEFAULT_TOKEN_SEPARATOR, true, true, ShingleFilter.DEFAULT_FILLER_TOKEN); } else { throw new RuntimeException("unknown analyzer " + analyzer); } final ReferenceManager<IndexSearcher> mgr; final IndexWriter writer; final Directory dir; final String commit = args.getString("-commit"); final String hiliteImpl = args.getString("-hiliteImpl"); final String logFile = args.getString("-log"); final long tSearcherStart = System.currentTimeMillis(); final boolean verifyCheckSum = !args.getFlag("-skipVerifyChecksum"); final boolean recacheFilterDeletes = args.getFlag("-recacheFilterDeletes"); if (recacheFilterDeletes) { throw new UnsupportedOperationException("recacheFilterDeletes was deprecated"); } if (args.getFlag("-nrt")) { // TODO: get taxoReader working here too // TODO: factor out & share this CL processing w/ Indexer final int indexThreadCount = args.getInt("-indexThreadCount"); final String lineDocsFile = args.getString("-lineDocsFile"); final float docsPerSecPerThread = args.getFloat("-docsPerSecPerThread"); final float reopenEverySec = args.getFloat("-reopenEverySec"); final boolean storeBody = args.getFlag("-store"); final boolean tvsBody = args.getFlag("-tvs"); final boolean useCFS = args.getFlag("-cfs"); final String defaultPostingsFormat = args.getString("-postingsFormat"); final String idFieldPostingsFormat = args.getString("-idFieldPostingsFormat"); final boolean verbose = args.getFlag("-verbose"); final boolean cloneDocs = args.getFlag("-cloneDocs"); final Mode mode = Mode.valueOf(args.getString("-mode", "update").toUpperCase(Locale.ROOT)); final long reopenEveryMS = (long) (1000 * reopenEverySec); if (verbose) { InfoStream.setDefault(new PrintStreamInfoStream(System.out)); } if (!dirImpl.equals("RAMDirectory") && !dirImpl.equals("RAMExceptDirectPostingsDirectory")) { System.out.println("Wrap NRTCachingDirectory"); dir0 = new NRTCachingDirectory(dir0, 20, 400.0); } dir = dir0; final IndexWriterConfig iwc = new IndexWriterConfig(a); iwc.setOpenMode(IndexWriterConfig.OpenMode.APPEND); iwc.setRAMBufferSizeMB(256.0); iwc.setIndexDeletionPolicy(NoDeletionPolicy.INSTANCE); // TODO: also RAMDirExceptDirect...? need to // ... block deletes against wrapped FSDir? if (dirImpl.equals("RAMDirectory")) { // Let IW remove files only referenced by starting commit: iwc.setIndexDeletionPolicy(new KeepNoCommitsDeletionPolicy()); } if (commit != null && commit.length() > 0) { System.out.println("Opening writer on commit=" + commit); iwc.setIndexCommit(PerfUtils.findCommitPoint(commit, dir)); } ((TieredMergePolicy) iwc.getMergePolicy()).setNoCFSRatio(useCFS ? 1.0 : 0.0); //((TieredMergePolicy) iwc.getMergePolicy()).setMaxMergedSegmentMB(1024); //((TieredMergePolicy) iwc.getMergePolicy()).setReclaimDeletesWeight(3.0); //((TieredMergePolicy) iwc.getMergePolicy()).setMaxMergeAtOnce(4); final Codec codec = new Lucene62Codec() { @Override public PostingsFormat getPostingsFormatForField(String field) { return PostingsFormat .forName(field.equals("id") ? idFieldPostingsFormat : defaultPostingsFormat); } }; iwc.setCodec(codec); final ConcurrentMergeScheduler cms = (ConcurrentMergeScheduler) iwc.getMergeScheduler(); // Only let one merge run at a time... // ... but queue up up to 4, before index thread is stalled: cms.setMaxMergesAndThreads(4, 1); iwc.setMergedSegmentWarmer(new IndexWriter.IndexReaderWarmer() { @Override public void warm(LeafReader reader) throws IOException { final long t0 = System.currentTimeMillis(); //System.out.println("DO WARM: " + reader); IndexSearcher s = new IndexSearcher(reader); s.setQueryCache(null); // don't bench the cache s.search(new TermQuery(new Term(fieldName, "united")), 10); final long t1 = System.currentTimeMillis(); System.out.println("warm segment=" + reader + " numDocs=" + reader.numDocs() + ": took " + (t1 - t0) + " msec"); } }); writer = new IndexWriter(dir, iwc); System.out.println("Initial writer.maxDoc()=" + writer.maxDoc()); // TODO: add -nrtBodyPostingsOffsets instead of // hardwired false: boolean addDVFields = mode == Mode.BDV_UPDATE || mode == Mode.NDV_UPDATE; LineFileDocs lineFileDocs = new LineFileDocs(lineDocsFile, false, storeBody, tvsBody, false, cloneDocs, null, null, null, addDVFields); IndexThreads threads = new IndexThreads(new Random(17), writer, new AtomicBoolean(false), lineFileDocs, indexThreadCount, -1, false, false, mode, docsPerSecPerThread, null, -1.0, -1); threads.start(); mgr = new SearcherManager(writer, new SearcherFactory() { @Override public IndexSearcher newSearcher(IndexReader reader, IndexReader previous) { IndexSearcher s = new IndexSearcher(reader); s.setQueryCache(null); // don't bench the cache s.setSimilarity(sim); return s; } }); System.out.println("reopen every " + reopenEverySec); Thread reopenThread = new Thread() { @Override public void run() { try { final long startMS = System.currentTimeMillis(); int reopenCount = 1; while (true) { final long sleepMS = startMS + (reopenCount * reopenEveryMS) - System.currentTimeMillis(); if (sleepMS < 0) { System.out.println("WARNING: reopen fell behind by " + Math.abs(sleepMS) + " ms"); } else { Thread.sleep(sleepMS); } Thread.sleep(sleepMS); mgr.maybeRefresh(); reopenCount++; IndexSearcher s = mgr.acquire(); try { if (ramDir != null) { System.out.println(String.format(Locale.ENGLISH, "%.1fs: index: %d bytes in RAMDir; writer.maxDoc()=%d; searcher.maxDoc()=%d; searcher.numDocs()=%d", (System.currentTimeMillis() - startMS) / 1000.0, ramDir.ramBytesUsed(), writer.maxDoc(), s.getIndexReader().maxDoc(), s.getIndexReader().numDocs())); //String[] l = ramDir.listAll(); //Arrays.sort(l); //for(String f : l) { //System.out.println(" " + f + ": " + ramDir.fileLength(f)); //} } else { System.out.println(String.format(Locale.ENGLISH, "%.1fs: done reopen; writer.maxDoc()=%d; searcher.maxDoc()=%d; searcher.numDocs()=%d", (System.currentTimeMillis() - startMS) / 1000.0, writer.maxDoc(), s.getIndexReader().maxDoc(), s.getIndexReader().numDocs())); } } finally { mgr.release(s); } } } catch (Exception e) { throw new RuntimeException(e); } } }; reopenThread.setName("ReopenThread"); reopenThread.setPriority(4 + Thread.currentThread().getPriority()); reopenThread.start(); } else { dir = dir0; writer = null; final DirectoryReader reader; if (commit != null && commit.length() > 0) { System.out.println("Opening searcher on commit=" + commit); reader = DirectoryReader.open(PerfUtils.findCommitPoint(commit, dir)); } else { // open last commit reader = DirectoryReader.open(dir); } IndexSearcher s = new IndexSearcher(reader); s.setQueryCache(null); // don't bench the cache s.setSimilarity(sim); System.out.println("maxDoc=" + reader.maxDoc() + " numDocs=" + reader.numDocs() + " %tg deletes=" + (100. * reader.maxDoc() / reader.numDocs())); mgr = new SingleIndexSearcher(s); } System.out.println((System.currentTimeMillis() - tSearcherStart) + " msec to init searcher/NRT"); { IndexSearcher s = mgr.acquire(); try { System.out.println("Searcher: numDocs=" + s.getIndexReader().numDocs() + " maxDoc=" + s.getIndexReader().maxDoc() + ": " + s); } finally { mgr.release(s); } } //System.out.println("searcher=" + searcher); FacetsConfig facetsConfig = new FacetsConfig(); facetsConfig.setHierarchical("Date", true); TaxonomyReader taxoReader; Path taxoPath = Paths.get(args.getString("-indexPath"), "facets"); Directory taxoDir = od.open(taxoPath); if (DirectoryReader.indexExists(taxoDir)) { taxoReader = new DirectoryTaxonomyReader(taxoDir); System.out.println("Taxonomy has " + taxoReader.getSize() + " ords"); } else { taxoReader = null; } final Random staticRandom = new Random(staticRandomSeed); final Random random = new Random(randomSeed); final DirectSpellChecker spellChecker = new DirectSpellChecker(); final IndexState indexState = new IndexState(mgr, taxoReader, fieldName, spellChecker, hiliteImpl, facetsConfig); final QueryParser queryParser = new QueryParser("body", a); TaskParser taskParser = new TaskParser(indexState, queryParser, fieldName, topN, staticRandom, doStoredLoads); final TaskSource tasks; if (tasksFile.startsWith("server:")) { int idx = tasksFile.indexOf(':', 8); if (idx == -1) { throw new RuntimeException( "server is missing the port; should be server:interface:port (got: " + tasksFile + ")"); } String iface = tasksFile.substring(7, idx); int port = Integer.valueOf(tasksFile.substring(1 + idx)); RemoteTaskSource remoteTasks = new RemoteTaskSource(iface, port, searchThreadCount, taskParser); // nocommit must stop thread? tasks = remoteTasks; } else { // Load the tasks from a file: final int taskRepeatCount = args.getInt("-taskRepeatCount"); final int numTaskPerCat = args.getInt("-tasksPerCat"); tasks = new LocalTaskSource(indexState, taskParser, tasksFile, staticRandom, random, numTaskPerCat, taskRepeatCount, doPKLookup); System.out.println("Task repeat count " + taskRepeatCount); System.out.println("Tasks file " + tasksFile); System.out.println("Num task per cat " + numTaskPerCat); } args.check(); // Evil respeller: //spellChecker.setMinPrefix(0); //spellChecker.setMaxInspections(1024); final TaskThreads taskThreads = new TaskThreads(tasks, indexState, searchThreadCount); Thread.sleep(10); final long startNanos = System.nanoTime(); taskThreads.start(); taskThreads.finish(); final long endNanos = System.nanoTime(); System.out.println("\n" + ((endNanos - startNanos) / 1000000.0) + " msec total"); final List<Task> allTasks = tasks.getAllTasks(); PrintStream out = new PrintStream(logFile); if (allTasks != null) { // Tasks were local: verify checksums: // indexState.setDocIDToID(); final Map<Task, Task> tasksSeen = new HashMap<Task, Task>(); out.println("\nResults for " + allTasks.size() + " tasks:"); boolean fail = false; for (final Task task : allTasks) { if (verifyCheckSum) { final Task other = tasksSeen.get(task); if (other != null) { if (task.checksum() != other.checksum()) { System.out.println("\nTASK:"); task.printResults(System.out, indexState); System.out.println("\nOTHER TASK:"); other.printResults(System.out, indexState); fail = true; //throw new RuntimeException("task " + task + " hit different checksums: " + task.checksum() + " vs " + other.checksum() + " other=" + other); } } else { tasksSeen.put(task, task); } } out.println("\nTASK: " + task); out.println(" " + (task.runTimeNanos / 1000000.0) + " msec"); out.println(" thread " + task.threadID); task.printResults(out, indexState); } if (fail) { throw new RuntimeException("some tasks got different results across different threads"); } allTasks.clear(); } mgr.close(); if (taxoReader != null) { taxoReader.close(); } if (writer != null) { // Don't actually commit any index changes: writer.rollback(); } dir.close(); if (printHeap) { // Try to get RAM usage -- some ideas poached from http://www.javaworld.com/javaworld/javatips/jw-javatip130.html final Runtime runtime = Runtime.getRuntime(); long usedMem1 = PerfUtils.usedMemory(runtime); long usedMem2 = Long.MAX_VALUE; for (int iter = 0; iter < 10; iter++) { runtime.runFinalization(); runtime.gc(); Thread.yield(); Thread.sleep(100); usedMem2 = usedMem1; usedMem1 = PerfUtils.usedMemory(runtime); } out.println("\nHEAP: " + PerfUtils.usedMemory(runtime)); } out.close(); }
From source file:pt.ua.ieeta.biomedhub.dictionaries.wrappers.CTakesLucene.java
License:Open Source License
public Multimap readBlocks(int blockSize) throws IOException { try {// w w w . j a va 2 s . c o m if (!DirectoryReader.indexExists(index)) { return null; } } catch (IOException ex) { Logger.getLogger("lucene").log(Level.SEVERE, null, ex); } String querystr = "*:*"; Query q = null; try { q = new QueryParser(Version.LUCENE_4_9, "title", analyzer).parse(querystr); } catch (ParseException ex) { Logger.getLogger(CTakesLucene.class.getName()).log(Level.SEVERE, null, ex); } int hitsPerPage = 100; IndexReader reader = DirectoryReader.open(index); IndexSearcher searcher = new IndexSearcher(reader); TopScoreDocCollector collector = TopScoreDocCollector.create(200000, true); searcher.search(q, collector); ScoreDoc[] hits = collector.topDocs().scoreDocs; System.out.println("Found " + hits.length + " hits."); for (int i = 0; i < hits.length; ++i) { int docId = hits[i].doc; Document d = searcher.doc(docId); //System.out.println((i + 1) + ". " + d.get("first_word") + "\t" + d.get("code")); System.out.print("."); synons.put(d.get("code"), d.get("first_word")); } return synons; }
From source file:start.lucene.IndexAndSearch.java
/** * * @param headers a list of CPeptides entries * @param ptmFactory/* w w w. j a v a 2 s . c o m*/ * @param linkerName - just name of linker to create both heavy and light * labeled versions * @param fragMode fragmentation mode * @param folder index folder * @throws IOException * @throws Exception */ public IndexAndSearch(HashSet<StringBuilder> headers, File folder, PTMFactory ptmFactory, FragmentationMode fragMode, String linkerName) throws IOException, Exception { // canCrosslinkerAttach if index files exist on given folder boolean reader = DirectoryReader.indexExists(FSDirectory.open(folder)); // if an index folder is absent, first index if (!reader) { CPeptidesIndexer indexer = new CPeptidesIndexer(headers, folder); indexer.index(); } cPeptideSearcher = new CPeptideSearcher(folder); this.ptmFactory = ptmFactory; this.fragMode = fragMode; heavyLinker = GetCrossLinker.getCrossLinker(linkerName, true); lightLinker = GetCrossLinker.getCrossLinker(linkerName, false); }
From source file:Tweet_search.Indexer.java
public Indexer() throws IOException, ClassCastException, ClassNotFoundException { GetProjetBaseDirAndSetProjectPropFile setPropFile = new GetProjetBaseDirAndSetProjectPropFile(); prop = setPropFile.prop;/*from w ww. java 2s . com*/ tweet_starts_from_date = Integer.parseInt(prop.getProperty("tweet.starts.from.date", "20")); tweet_ends_from_date = Integer.parseInt(prop.getProperty("tweet.ends.from.date", "29")); // // if(prop.getProperty("pos.tag","false").toLowerCase().contentEquals("true")) // postag=new POStag(); // if(prop.getProperty("ner.classify","false").toLowerCase().contentEquals("true")) // nertag=new NERtag(); setAnalyzer(); /* property files are loaded */ /* collection path setting */ if (prop.containsKey("collSpec")) { boolIndexFromSpec = true; } else if (prop.containsKey("collPath")) { boolIndexFromSpec = false; collPath = prop.getProperty("collPath"); collDir = new File(collPath); if (!collDir.exists() || !collDir.canRead()) { System.err.println("Collection directory '" + collDir.getAbsolutePath() + "' does not exist or is not readable"); System.exit(1); } } else { System.err.println("Neither collPath not collSpec is present"); System.exit(1); } /* collection path set */ /* index path setting */ indexFile = new File[tweet_ends_from_date - tweet_starts_from_date + 1]; Directory[] indexDir = new Directory[tweet_ends_from_date - tweet_starts_from_date + 1]; indexWriter = new IndexWriter[tweet_ends_from_date - tweet_starts_from_date + 1]; String indexBasePath = prop.getProperty("indexPath"); System.err.println(indexBasePath); if (indexBasePath.endsWith("/")) ; else indexBasePath = indexBasePath + "/"; System.err.println(indexBasePath); for (int i = tweet_starts_from_date; i <= tweet_ends_from_date; i++) { int j = i - tweet_starts_from_date; indexFile[j] = new File(indexBasePath.concat(Integer.toString(i))); indexDir[j] = FSDirectory.open(indexFile[j]); /* index path set */ if (DirectoryReader.indexExists(indexDir[j])) { System.err.println("Index exists in " + indexFile[j].getAbsolutePath()); boolIndexExists = true; } else { System.out.println("Will create the index in: " + indexFile[j].getName()); boolIndexExists = false; /* Create a new index in the directory, removing any previous indexed documents */ IndexWriterConfig[] iwcfg = new IndexWriterConfig[tweet_ends_from_date - tweet_starts_from_date + 1]; iwcfg[j] = new IndexWriterConfig(Version.LATEST, analyzer); iwcfg[j].setOpenMode(IndexWriterConfig.OpenMode.CREATE); /* */ indexWriter[j] = new IndexWriter(indexDir[j], iwcfg[j]); } } boolDumpIndex = Boolean.parseBoolean(prop.getProperty("dumpIndex", "false")); if (boolIndexExists == true && boolDumpIndex == true) { dumpPath = prop.getProperty("dumpPath"); } }