Example usage for org.apache.lucene.search IndexSearcher IndexSearcher

List of usage examples for org.apache.lucene.search IndexSearcher IndexSearcher

  1. HOME
  2. Java
  3. org.apache.lucene
  4. org.apache.lucene.search.*
  5. IndexSearcher
  6. IndexSearcher

Introduction

In this page you can find the example usage for org.apache.lucene.search IndexSearcher IndexSearcher.

Prototype

public IndexSearcher(IndexReaderContext context)

Source Link

Document

Creates a searcher searching the provided top-level IndexReaderContext .

Usage

From source file:com.gitblit.LuceneExecutor.java

License:Apache License

/**
 * Searches the specified repositories for the given text or query
 * //from   w  w w  .  ja  v a  2  s .  c om
 * @param text
 *            if the text is null or empty, null is returned
 * @param page
 *            the page number to retrieve. page is 1-indexed.
 * @param pageSize
 *            the number of elements to return for this page
 * @param repositories
 *            a list of repositories to search. if no repositories are
 *            specified null is returned.
 * @return a list of SearchResults in order from highest to the lowest score
 * 
 */
public List<SearchResult> search(String text, int page, int pageSize, String... repositories) {
    if (StringUtils.isEmpty(text)) {
        return null;
    }
    if (ArrayUtils.isEmpty(repositories)) {
        return null;
    }
    Set<SearchResult> results = new LinkedHashSet<SearchResult>();
    StandardAnalyzer analyzer = new StandardAnalyzer(LUCENE_VERSION);
    try {
        // default search checks summary and content
        BooleanQuery query = new BooleanQuery();
        QueryParser qp;
        qp = new QueryParser(LUCENE_VERSION, FIELD_SUMMARY, analyzer);
        qp.setAllowLeadingWildcard(true);
        query.add(qp.parse(text), Occur.SHOULD);

        qp = new QueryParser(LUCENE_VERSION, FIELD_CONTENT, analyzer);
        qp.setAllowLeadingWildcard(true);
        query.add(qp.parse(text), Occur.SHOULD);

        IndexSearcher searcher;
        if (repositories.length == 1) {
            // single repository search
            searcher = getIndexSearcher(repositories[0]);
        } else {
            // multiple repository search
            List<IndexReader> readers = new ArrayList<IndexReader>();
            for (String repository : repositories) {
                IndexSearcher repositoryIndex = getIndexSearcher(repository);
                readers.add(repositoryIndex.getIndexReader());
            }
            IndexReader[] rdrs = readers.toArray(new IndexReader[readers.size()]);
            MultiSourceReader reader = new MultiSourceReader(rdrs);
            searcher = new IndexSearcher(reader);
        }

        Query rewrittenQuery = searcher.rewrite(query);
        logger.debug(rewrittenQuery.toString());

        TopScoreDocCollector collector = TopScoreDocCollector.create(5000, true);
        searcher.search(rewrittenQuery, collector);
        int offset = Math.max(0, (page - 1) * pageSize);
        ScoreDoc[] hits = collector.topDocs(offset, pageSize).scoreDocs;
        int totalHits = collector.getTotalHits();
        for (int i = 0; i < hits.length; i++) {
            int docId = hits[i].doc;
            Document doc = searcher.doc(docId);
            SearchResult result = createSearchResult(doc, hits[i].score, offset + i + 1, totalHits);
            if (repositories.length == 1) {
                // single repository search
                result.repository = repositories[0];
            } else {
                // multi-repository search
                MultiSourceReader reader = (MultiSourceReader) searcher.getIndexReader();
                int index = reader.getSourceIndex(docId);
                result.repository = repositories[index];
            }
            String content = doc.get(FIELD_CONTENT);
            result.fragment = getHighlightedFragment(analyzer, query, content, result);
            results.add(result);
        }
    } catch (Exception e) {
        logger.error(MessageFormat.format("Exception while searching for {0}", text), e);
    }
    return new ArrayList<SearchResult>(results);
}

From source file:com.gitblit.service.LuceneService.java

License:Apache License

/**
 * Gets an index searcher for the repository.
 *
 * @param repository/*  w  ww .java  2  s .  co m*/
 * @return
 * @throws IOException
 */
private IndexSearcher getIndexSearcher(String repository) throws IOException {
    IndexSearcher searcher = searchers.get(repository);
    if (searcher == null) {
        IndexWriter writer = getIndexWriter(repository);
        searcher = new IndexSearcher(DirectoryReader.open(writer, true));
        searchers.put(repository, searcher);
    }
    return searcher;
}

From source file:com.gitblit.service.LuceneService.java

License:Apache License

/**
 * Searches the specified repositories for the given text or query
 *
 * @param text/*from  w ww .j a  va2  s .com*/
 *            if the text is null or empty, null is returned
 * @param page
 *            the page number to retrieve. page is 1-indexed.
 * @param pageSize
 *            the number of elements to return for this page
 * @param repositories
 *            a list of repositories to search. if no repositories are
 *            specified null is returned.
 * @return a list of SearchResults in order from highest to the lowest score
 *
 */
public List<SearchResult> search(String text, int page, int pageSize, String... repositories) {
    if (StringUtils.isEmpty(text)) {
        return null;
    }
    if (ArrayUtils.isEmpty(repositories)) {
        return null;
    }
    Set<SearchResult> results = new LinkedHashSet<SearchResult>();
    StandardAnalyzer analyzer = new StandardAnalyzer();
    try {
        // default search checks summary and content
        BooleanQuery.Builder bldr = new BooleanQuery.Builder();
        QueryParser qp;
        qp = new QueryParser(FIELD_SUMMARY, analyzer);
        qp.setAllowLeadingWildcard(true);
        bldr.add(qp.parse(text), Occur.SHOULD);

        qp = new QueryParser(FIELD_CONTENT, analyzer);
        qp.setAllowLeadingWildcard(true);
        bldr.add(qp.parse(text), Occur.SHOULD);

        IndexSearcher searcher;
        if (repositories.length == 1) {
            // single repository search
            searcher = getIndexSearcher(repositories[0]);
        } else {
            // multiple repository search
            List<IndexReader> readers = new ArrayList<IndexReader>();
            for (String repository : repositories) {
                IndexSearcher repositoryIndex = getIndexSearcher(repository);
                readers.add(repositoryIndex.getIndexReader());
            }
            IndexReader[] rdrs = readers.toArray(new IndexReader[readers.size()]);
            MultiSourceReader reader = new MultiSourceReader(rdrs);
            searcher = new IndexSearcher(reader);
        }

        BooleanQuery query = bldr.build();
        Query rewrittenQuery = searcher.rewrite(query);
        logger.debug(rewrittenQuery.toString());

        TopScoreDocCollector collector = TopScoreDocCollector.create(5000);
        searcher.search(rewrittenQuery, collector);
        int offset = Math.max(0, (page - 1) * pageSize);
        ScoreDoc[] hits = collector.topDocs(offset, pageSize).scoreDocs;
        int totalHits = collector.getTotalHits();
        for (int i = 0; i < hits.length; i++) {
            int docId = hits[i].doc;
            Document doc = searcher.doc(docId);
            SearchResult result = createSearchResult(doc, hits[i].score, offset + i + 1, totalHits);
            if (repositories.length == 1) {
                // single repository search
                result.repository = repositories[0];
            } else {
                // multi-repository search
                MultiSourceReader reader = (MultiSourceReader) searcher.getIndexReader();
                int index = reader.getSourceIndex(docId);
                result.repository = repositories[index];
            }
            String content = doc.get(FIELD_CONTENT);
            result.fragment = getHighlightedFragment(analyzer, query, content, result);
            results.add(result);
        }
    } catch (Exception e) {
        logger.error(MessageFormat.format("Exception while searching for {0}", text), e);
    }
    return new ArrayList<SearchResult>(results);
}

From source file:com.gitblit.tickets.TicketIndexer.java

License:Apache License

private IndexSearcher getSearcher() throws IOException {
    if (searcher == null) {
        searcher = new IndexSearcher(DirectoryReader.open(getWriter(), true));
    }//from   w  w w.j  a v  a 2  s . com
    return searcher;
}

From source file:com.github.alvanson.xltsearch.SearchTask.java

License:Apache License

@Override
protected List<SearchResult> call() {
    DirectoryReader ireader = null;/*w ww. j av  a 2  s .  c o  m*/
    List<SearchResult> results = null;

    updateMessage("Searching...");
    try {
        ireader = DirectoryReader.open(config.getDirectory());
        IndexSearcher isearcher = new IndexSearcher(ireader);
        isearcher.setSimilarity(config.getSimilarity());
        QueryParser parser = new QueryParser(config.getVersion(), config.contentField, config.getAnalyzer());
        Query query = parser.parse(qstr);
        logger.debug("Query: {}", query);
        ScoreDoc[] hits = isearcher.search(query, limit).scoreDocs;
        // collect results
        results = new ArrayList<>(hits.length);
        for (ScoreDoc hit : hits) {
            Document document = isearcher.doc(hit.doc);
            File file = new File(root.getPath() + File.separator + document.get(config.pathField));
            String title = document.get(config.titleField);
            if (title == null) {
                title = "";
            }
            // report metadata in `details`
            StringBuilder sb = new StringBuilder();
            for (IndexableField field : document.getFields()) {
                if (field.stringValue() != null) {
                    sb.append(field.name() + ": " + field.stringValue() + '\n');
                }
            }
            results.add(new SearchResult(file, title, hit.score, sb.toString()));
        }
        updateMessage(hits.length + " results");
    } catch (IOException ex) {
        updateMessage("I/O exception");
        logger.error("I/O exception while reading index", ex);
    } catch (ParseException ex) {
        updateMessage("Parse error");
        logger.warn("Parse exception while parsing '{}'", qstr, ex);
    }
    // close ireader
    if (ireader != null) {
        try {
            ireader.close();
        } catch (IOException ex) {
            logger.warn("I/O exception while closing index reader", ex);
        }
    }
    return results;
}

From source file:com.github.alvanson.xltsearch.SelectTask.java

License:Apache License

private Map<String, String> getHashSums() {
    Map<String, String> hashSums = new HashMap<>();
    DirectoryReader ireader = null;/*from   w w  w .j  a  va2  s  . co  m*/
    try {
        if (DirectoryReader.indexExists(config.getDirectory())) {
            // read hashsums from `directory`
            ireader = DirectoryReader.open(config.getDirectory());
            IndexSearcher isearcher = new IndexSearcher(ireader);
            Query query = new MatchAllDocsQuery();
            ScoreDoc[] hits = isearcher.search(query, ireader.numDocs() + 1).scoreDocs;
            // collect results
            for (ScoreDoc hit : hits) {
                Document document = isearcher.doc(hit.doc);
                String relPath = document.get(config.pathField);
                String hashSum = document.get(config.hashSumField);
                if (relPath != null && hashSum != null) {
                    hashSums.put(relPath, hashSum);
                }
            }
        } // else: return empty map
    } catch (IOException ex) {
        logger.error("I/O exception while reading index", ex);
    }
    if (ireader != null) {
        try {
            ireader.close();
        } catch (IOException ex) {
            logger.warn("I/O exception while closing index reader", ex);
        }
    }
    return hashSums;
}

From source file:com.github.cmisbox.persistence.Storage.java

License:Open Source License

private IndexSearcher getSearcher() throws Exception {
    this.reader = this.reader.reopen();
    return new IndexSearcher(this.reader);
}

From source file:com.github.jiloc.USTweetsAnalyzer.Analyzer_Index.java

public Analyzer_Index(Directory dir) throws IOException {
    ir = DirectoryReader.open(dir);//from ww  w  . ja  va2s  .c o m
    searcher = new IndexSearcher(ir);
    // universe =  new HashMap<String, HashSet<String>>();

}

From source file:com.github.le11.nls.lucene.UIMABaseAnalyzerTest.java

License:Apache License

@Test
public void baseUIMAAnalyzerIntegrationTest() {
    try {/*from  w  w w  .  j  a  v  a2 s  .c o  m*/
        // add the first doc
        Document doc = new Document();
        doc.add(new Field("title", "this is a dummy title ", Field.Store.YES, Field.Index.ANALYZED));
        doc.add(new Field("contents", "there is some content written here", Field.Store.YES,
                Field.Index.ANALYZED));
        writer.addDocument(doc, analyzer);
        writer.commit();

        // try the search over the first doc
        IndexSearcher indexSearcher = new IndexSearcher(writer.getReader());
        TopDocs result = indexSearcher.search(new MatchAllDocsQuery("contents"), 10);
        assertTrue(result.totalHits > 0);
        Document d = indexSearcher.doc(result.scoreDocs[0].doc);
        assertNotNull(d);
        assertNotNull(d.getFieldable("title"));
        assertNotNull(d.getFieldable("contents"));

        // add a second doc
        doc = new Document();
        doc.add(new Field("title", "il mio titolo", Field.Store.YES, Field.Index.ANALYZED));
        doc.add(new Field("contents", "che cosa e' scritto qui", Field.Store.YES, Field.Index.ANALYZED));
        writer.addDocument(doc, analyzer);
        writer.commit();

        // do a matchalldocs query to retrieve both docs
        indexSearcher = new IndexSearcher(writer.getReader());
        result = indexSearcher.search(new MatchAllDocsQuery("contents"), 10);
        assertTrue(result.totalHits > 0);
        for (ScoreDoc di : result.scoreDocs) {
            d = indexSearcher.doc(di.doc);
            assertNotNull(d);
            assertNotNull(d.getFieldable("title"));
            assertNotNull(d.getFieldable("contents"));
        }
    } catch (Exception e) {
        fail(e.getLocalizedMessage());
    }
}

From source file:com.github.le11.nls.lucene.UIMATypeBasedSimilarityTest.java

License:Apache License

@Before
public void setUp() throws Exception {
    dir = new RAMDirectory();
    analyzer = new UIMAPayloadsAnalyzer("/HmmTaggerAggregate.xml");
    writer = new IndexWriter(dir, new IndexWriterConfig(Version.LUCENE_33, analyzer));

    Document doc = new Document();
    doc.add(new Field("title", "this is a dummy title containing an entity for London", Field.Store.YES,
            Field.Index.ANALYZED));
    doc.add(new Field("contents", "there is some content written here about the british city", Field.Store.YES,
            Field.Index.ANALYZED));
    writer.addDocument(doc, analyzer);//w ww  .ja  v  a  2  s  . c  om
    writer.commit();

    // try the search over the first doc
    IndexSearcher indexSearcher = new IndexSearcher(writer.getReader());
    TopDocs result = indexSearcher.search(new MatchAllDocsQuery("contents"), 10);
    assertTrue(result.totalHits > 0);
    Document d = indexSearcher.doc(result.scoreDocs[0].doc);
    assertNotNull(d);
    assertNotNull(d.getFieldable("title"));
    assertNotNull(d.getFieldable("contents"));

    // add a second doc
    doc = new Document();
    doc.add(new Field("title", "some title regarding some article written in English", Field.Store.YES,
            Field.Index.ANALYZED));
    doc.add(new Field("contents", "this is the content of the article about", Field.Store.YES,
            Field.Index.ANALYZED));
    writer.addDocument(doc, analyzer);
    writer.commit();
}