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:at.ac.univie.mminf.luceneSKOS.test.termexpansion.LabelbasedTermExpansionTest.java

License:Apache License

/**
 * This test indexes a sample metadata record (=lucene document) having a
 * "title", "description", and "subject" field.
 * <p/>/*from  www .j av  a  2  s.c o m*/
 * A search for "arms" returns that record as a result because "arms" is
 * defined as an alternative label for "weapons", the term which is
 * contained in the subject field.
 *
 * @throws IOException
 */
@Test
public void labelBasedTermExpansion() throws IOException {

    /* defining the document to be indexed */
    Document doc = new Document();
    doc.add(new Field("title", "Spearhead", TextField.TYPE_STORED));
    doc.add(new Field("description",
            "Roman iron spearhead. The spearhead was attached to one end of a wooden shaft..."
                    + "The spear was mainly a thrusting weapon, but could also be thrown. "
                    + "It was the principal weapon of the auxiliary soldier... "
                    + "(second - fourth century, Arbeia Roman Fort).",
            TextField.TYPE_NOT_STORED));
    doc.add(new Field("subject", "weapons", TextField.TYPE_NOT_STORED));

    /* setting up the SKOS analyzer */
    String skosFile = "src/test/resources/skos_samples/ukat_examples.n3";
    String indexPath = "build/";

    /* ExpansionType.URI->the field to be analyzed (expanded) contains URIs */
    Analyzer skosAnalyzer = new SKOSAnalyzer(indexPath, skosFile, ExpansionType.LABEL);

    /* Define different analyzers for different fields */
    Map<String, Analyzer> analyzerPerField = new HashMap<>();
    analyzerPerField.put("subject", skosAnalyzer);
    PerFieldAnalyzerWrapper indexAnalyzer = new PerFieldAnalyzerWrapper(new SimpleAnalyzer(), analyzerPerField);

    /* setting up a writer with a default (simple) analyzer */
    writer = new IndexWriter(new RAMDirectory(), new IndexWriterConfig(indexAnalyzer));

    /* adding the document to the index */
    writer.addDocument(doc);

    /* defining a query that searches over all fields */
    BooleanQuery.Builder builder = new BooleanQuery.Builder();
    builder.add(new TermQuery(new Term("title", "arms")), BooleanClause.Occur.SHOULD)
            .add(new TermQuery(new Term("description", "arms")), BooleanClause.Occur.SHOULD)
            .add(new TermQuery(new Term("subject", "arms")), BooleanClause.Occur.SHOULD);

    /* creating a new searcher */
    searcher = new IndexSearcher(DirectoryReader.open(writer, false));

    TopDocs results = searcher.search(builder.build(), 10);

    /* the document matches because "arms" is among the expanded terms */
    assertEquals(1, results.totalHits);

    /* defining a query that searches for a broader concept */
    Query query = new TermQuery(new Term("subject", "military equipment"));

    results = searcher.search(query, 10);

    /* ... also returns the document as result */
    assertEquals(1, results.totalHits);
}

From source file:at.ac.univie.mminf.luceneSKOS.test.termexpansion.URIbasedTermExpansionTest.java

License:Apache License

/**
 * This test indexes a sample metadata record (=lucene document) having a
 * "title", "description", and "subject" field, which is semantically
 * enriched by a URI pointing to a SKOS concept "weapons".
 * <p/>//from   w  w  w .  j av  a  2s.  c o m
 * A search for "arms" returns that record as a result because "arms" is
 * defined as an alternative label (altLabel) for the concept "weapons".
 *
 * @throws IOException
 */
@Test
public void uriBasedTermExpansion() throws IOException {

    /* defining the document to be indexed */
    Document doc = new Document();
    doc.add(new Field("title", "Spearhead", TextField.TYPE_STORED));
    doc.add(new Field("description",
            "Roman iron spearhead. The spearhead was attached to one end of a wooden shaft..."
                    + "The spear was mainly a thrusting weapon, but could also be thrown. "
                    + "It was the principal weapon of the auxiliary soldier... "
                    + "(second - fourth century, Arbeia Roman Fort).",
            TextField.TYPE_NOT_STORED));
    doc.add(new Field("subject", "http://www.ukat.org.uk/thesaurus/concept/859", TextField.TYPE_NOT_STORED));

    /* setting up the SKOS analyzer */
    String skosFile = "src/test/resources/skos_samples/ukat_examples.n3";
    String indexPath = "build/";

    /* ExpansionType.URI->the field to be analyzed (expanded) contains URIs */
    Analyzer skosAnalyzer = new SKOSAnalyzer(indexPath, skosFile, ExpansionType.URI);

    /* Define different analyzers for different fields */
    Map<String, Analyzer> analyzerPerField = new HashMap<>();
    analyzerPerField.put("subject", skosAnalyzer);
    PerFieldAnalyzerWrapper indexAnalyzer = new PerFieldAnalyzerWrapper(new SimpleAnalyzer(), analyzerPerField);

    /* setting up a writer with a default (simple) analyzer */
    writer = new IndexWriter(new RAMDirectory(), new IndexWriterConfig(indexAnalyzer));

    /* adding the document to the index */
    writer.addDocument(doc);

    /* defining a query that searches over all fields */
    BooleanQuery.Builder builder = new BooleanQuery.Builder();
    builder.add(new TermQuery(new Term("title", "arms")), BooleanClause.Occur.SHOULD)
            .add(new TermQuery(new Term("description", "arms")), BooleanClause.Occur.SHOULD)
            .add(new TermQuery(new Term("subject", "arms")), BooleanClause.Occur.SHOULD);

    /* creating a new searcher */
    searcher = new IndexSearcher(DirectoryReader.open(writer, false));

    TopDocs results = searcher.search(builder.build(), 10);

    /* the document matches because "arms" is among the expanded terms */
    assertEquals(1, results.totalHits);

    /* defining a query that searches for a broader concept */
    Query query = new TermQuery(new Term("subject", "military equipment"));

    results = searcher.search(query, 10);

    /* ... also returns the document as result */
    assertEquals(1, results.totalHits);

}

From source file:at.ac.univie.mminf.luceneSKOS.URIbasedTermExpansionTest.java

License:Apache License

/**
 * This test indexes a sample metadata record (=lucene document) having a
 * "title", "description", and "subject" field, which is semantically enriched
 * by a URI pointing to a SKOS concept "weapons".
 * /*from www.j  a va2 s  . c om*/
 * A search for "arms" returns that record as a result because "arms" is
 * defined as an alternative label (altLabel) for the concept "weapons".
 * 
 * @throws IOException
 */
@Test
public void uriBasedTermExpansion() throws IOException {

    /* defining the document to be indexed */
    Document doc = new Document();
    doc.add(new Field("title", "Spearhead", Field.Store.YES, Field.Index.ANALYZED));
    doc.add(new Field("description",
            "Roman iron spearhead. The spearhead was attached to one end of a wooden shaft..."
                    + "The spear was mainly a thrusting weapon, but could also be thrown. "
                    + "It was the principal weapon of the auxiliary soldier... "
                    + "(second - fourth century, Arbeia Roman Fort).",
            Field.Store.NO, Field.Index.ANALYZED));
    doc.add(new Field("subject", "http://www.ukat.org.uk/thesaurus/concept/859", Field.Store.NO,
            Field.Index.ANALYZED));

    /* setting up the SKOS analyzer */
    String skosFile = "src/test/resources/skos_samples/ukat_examples.n3";

    /* ExpansionType.URI->the field to be analyzed (expanded) contains URIs */
    Analyzer skosAnalyzer = new SKOSAnalyzer(matchVersion, skosFile, ExpansionType.URI);

    /* Define different analyzers for different fields */
    Map<String, Analyzer> analyzerPerField = new HashMap<String, Analyzer>();
    analyzerPerField.put("subject", skosAnalyzer);
    PerFieldAnalyzerWrapper indexAnalyzer = new PerFieldAnalyzerWrapper(new SimpleAnalyzer(matchVersion),
            analyzerPerField);

    /* setting up a writer with a default (simple) analyzer */
    writer = new IndexWriter(new RAMDirectory(), new IndexWriterConfig(matchVersion, indexAnalyzer));

    /* adding the document to the index */
    writer.addDocument(doc);

    /* defining a query that searches over all fields */
    BooleanQuery query1 = new BooleanQuery();
    query1.add(new TermQuery(new Term("title", "arms")), BooleanClause.Occur.SHOULD);
    query1.add(new TermQuery(new Term("description", "arms")), BooleanClause.Occur.SHOULD);
    query1.add(new TermQuery(new Term("subject", "arms")), BooleanClause.Occur.SHOULD);

    /* creating a new searcher */
    searcher = new IndexSearcher(IndexReader.open(writer, false));

    TopDocs results = searcher.search(query1, 10);

    /* the document matches because "arms" is among the expanded terms */
    Assert.assertEquals(1, results.totalHits);

    /* defining a query that searches for a broader concept */
    Query query2 = new TermQuery(new Term("subject", "military equipment"));

    results = searcher.search(query2, 10);

    /* ... also returns the document as result */
    Assert.assertEquals(1, results.totalHits);

}

From source file:at.fh_kufstein.InformationRetrievalUebung2.Main.java

public static void main(String[] args) {
    try {/*from   w w w  . ja v  a 2  s  . c o  m*/
        Directory dir = getBookIndexDirectory();
        IndexSearcher searcher = new IndexSearcher(dir);
        QueryParser parser = new QueryParser(Version.LUCENE_30, SEARCH_FIELD, new SimpleAnalyzer());
        Query query = parser.parse(getUserQuery());
        TopDocs docs = searcher.search(query, MAX_HITS);
        printResults(docs);
        searcher.close();
        dir.close();
    } catch (IOException | ParseException ex) {
        Logger.getLogger(Main.class.getName()).log(Level.SEVERE, null, ex);
    }
}

From source file:at.lux.fotoretrieval.retrievalengines.LucenePathIndexRetrievalEngine.java

License:Open Source License

public List<ResultListEntry> getImagesBySemantics(String xPath, Vector objects, String whereToSearch,
        boolean recursive, JProgressBar progress) {
    if (progress != null) {
        progress.setString("Query expansion running");
    }//from ww w  .j  a v  a 2s.com
    List<Graph> graphList = getExpandedGraphs(xPath, whereToSearch);
    LinkedList<ResultListEntry> results = new LinkedList<ResultListEntry>();
    if (progress != null) {
        progress.setMinimum(0);
        progress.setMaximum(graphList.size());
        progress.setValue(0);
        progress.setString("Querying expanded graphs");
    }
    int countGraph = 0;
    try {
        indexSearcher = new IndexSearcher(parsePathIndexDirectory(whereToSearch));
        for (Graph graph : graphList) {
            results.addAll(searchForGraph(graph, whereToSearch));
            countGraph++;
            if (progress != null)
                progress.setValue(countGraph);
        }
        indexSearcher.close();
    } catch (IOException e) {
        e.printStackTrace();
    }
    Collections.sort(results);
    LinkedList<ResultListEntry> sorted = new LinkedList<ResultListEntry>();
    HashSet<String> doublettes = new HashSet<String>();
    for (ResultListEntry entry : results) {
        String descriptionPath = entry.getDescriptionPath();
        if (!doublettes.contains(descriptionPath)) {
            doublettes.add(descriptionPath);
            sorted.add(entry);
        }
    }
    return sorted.subList(0, Math.min(sorted.size(), maxResults));
}

From source file:at.lux.fotoretrieval.retrievalengines.LucenePathIndexRetrievalEngine.java

License:Open Source License

/**
 * Searches for all available nodes with given query String
 *
 * @param queryString   query like "Mathias Lux" or some text inside a node.
 * @param whereToSearch defines the base directory for the search
 * @return a List of Matching nodes with their associated weights
 *//*from   w  w  w  .j a va 2  s. c  om*/
public List<Node> getNodes(String queryString, String whereToSearch) {
    LinkedList<Node> result = new LinkedList<Node>();
    try {
        IndexSearcher searcher = new IndexSearcher(parseSemanticIndexDirectory(whereToSearch));
        String[] fieldsToSearchInForNodes = new String[] { "label", "givenname", "organization", "familyname",
                "all" };
        MultiFieldQueryParser qParse = new MultiFieldQueryParser(fieldsToSearchInForNodes,
                new StandardAnalyzer());
        Query query = qParse.parse(queryString);
        Hits hits = searcher.search(query);
        int hitsCount = hits.length();
        if (hitsCount > maxResults)
            hitsCount = maxResults;

        for (int i = 0; i < hitsCount; i++) {
            Document d = hits.doc(i);
            StringBuilder sb = new StringBuilder(20);
            sb.append(hits.score(i));
            sb.append(": ");
            sb.append(d.get("label"));
            Node node = new Node(Integer.parseInt(d.get("id")), hits.score(i), d.get("label"));
            node.setType(d.get("type"));
            result.add(node);
        }

    } catch (IOException e) {
        e.printStackTrace();
    } catch (ParseException e) {
        System.err.println("QueryString was: " + queryString);
        e.printStackTrace();
    }
    return result;
}

From source file:at.lux.fotoretrieval.retrievalengines.LucenePathIndexRetrievalEngineTest.java

License:Open Source License

public void testSearch() {
    try {//from   w  ww . j a v a  2 s  .com
        QueryParser qParser = new QueryParser("graph", new WhitespaceAnalyzer());
        IndexSearcher search = new IndexSearcher(
                "C:\\Dokumente und Einstellungen\\Mathias\\Eigene Dateien\\JavaProjects\\Caliph\\testdata\\idx_paths");
        Hits h = search.search(qParser.parse("_*_0_1"));
        for (int i = 0; i < h.length(); i++) {
            System.out.println(h.score(i) + ": " + h.doc(i).get("graph"));
        }
    } catch (IOException e) {
        e.printStackTrace();
        fail(e.toString());
    } catch (ParseException e) {
        e.printStackTrace();
        fail(e.toString());
    }
}

From source file:at.lux.fotoretrieval.retrievalengines.LucenePathIndexRetrievalEngineTest.java

License:Open Source License

public void testPrecisionAndRecall() {
    try {/*from w w w .j av  a  2  s . c  om*/
        String repository = "C:\\Java\\JavaProjects\\CaliphEmir\\testdata";
        //            String repository = "C:\\Dokumente und Einstellungen\\Mathias\\Eigene Dateien\\JavaProjects\\Caliph\\testdata";
        IndexSearcher is = new IndexSearcher(repository + "\\idx_paths");
        IndexReader ir = IndexReader.open(repository + "\\idx_paths");

        for (int i = 0; i < ir.numDocs(); i++) {
            testQuery(ir, new Graph(ir.document(i).getValues("graph")[0]), is);
        }

    } catch (IOException e) {
        e.printStackTrace();
        fail(e.toString());
    } catch (ParseException e) {
        e.printStackTrace();
    }
}

From source file:at.lux.fotoretrieval.retrievalengines.LucenePathIndexRetrievalEngineTest.java

License:Open Source License

public void testPrecisionAndRecallFullText() {
    try {// www . ja v  a 2s  . co  m
        String repository = "C:\\Java\\JavaProjects\\CaliphEmir\\testdata";
        //            String repository = "C:\\Dokumente und Einstellungen\\Mathias\\Eigene Dateien\\JavaProjects\\Caliph\\testdata";
        IndexSearcher is = new IndexSearcher(repository + "\\idx_paths");
        IndexReader ir = IndexReader.open(repository + "\\idx_paths");

        for (int i = 0; i < ir.numDocs(); i++) {
            testDirectQuery(ir, new Graph(ir.document(i).getValues("graph")[0]), is);
        }

    } catch (IOException e) {
        e.printStackTrace();
        fail(e.toString());
    } catch (ParseException e) {
        e.printStackTrace();
    }
}

From source file:at.lux.fotoretrieval.retrievalengines.LucenePathIndexRetrievalEngineTest.java

License:Open Source License

private void testDirectQuery(IndexReader ir, Graph query, IndexSearcher is) throws IOException, ParseException {
    IndexReader reader = IndexReader.open("C:\\Java\\JavaProjects\\CaliphEmir\\testdata\\idx_semantic");
    IndexSearcher searcher = new IndexSearcher("C:\\Java\\JavaProjects\\CaliphEmir\\testdata\\idx_fulltext");

    HashMap<Integer, String> node2label = new HashMap<Integer, String>();
    for (int j = 0; j < reader.numDocs(); j++) {
        String id = reader.document(j).getValues("id")[0];
        String label = reader.document(j).getValues("label")[0];
        node2label.put(Integer.parseInt(id), label);
    }//from  www  .j  ava  2s .  com
    // create results from mcs:
    LinkedList<ResultHolder> resultsMcs = new LinkedList<ResultHolder>();
    for (int j = 0; j < ir.numDocs(); j++) {
        Graph model = new Graph(ir.document(j).getValues("graph")[0]);
        float mcsSimilarity = query.getMcsSimilarity(model);
        String[] file = ir.document(j).getValues("file");
        for (int i = 0; i < file.length; i++) {
            String s = file[i];
            resultsMcs.add(new ResultHolder(mcsSimilarity, s));
        }
    }
    Collections.sort(resultsMcs);
    //            for (Iterator<ResultHolder> iterator = resultsMcs.iterator(); iterator.hasNext();) {
    //                ResultHolder r = iterator.next();
    //                System.out.println(r.getDocumentNumber() + ": " + r.getSimilarity());
    //            }

    // create results from search:
    StringBuilder qBuilder = new StringBuilder(64);
    for (Iterator<Node> iterator = query.getNodes().iterator(); iterator.hasNext();) {
        Node node = iterator.next();
        //            qBuilder.append("\"");
        qBuilder.append(node2label.get(node.getNodeID()));
        qBuilder.append(" ");
        //            qBuilder.append("\" ");
    }
    //        System.out.println(query);
    QueryParser qParse = new QueryParser("all", new WhitespaceAnalyzer());
    Query q = qParse.parse(qBuilder.toString());
    Hits hits = searcher.search(q);
    LinkedList<ResultHolder> resultsSearch = new LinkedList<ResultHolder>();
    for (int i = 0; i < hits.length(); i++) {
        String graph = hits.doc(i).getValues("file")[0];
        //            int docID = -1;
        //            for (int j = 0; j < ir.numDocs(); j++) {
        //                Graph model = new Graph(ir.document(j).getValues("graph")[0]);
        //                if (model.toString().equals(graph)) docID = j;
        //            }
        resultsSearch.add(new ResultHolder(hits.score(i), graph));
    }
    Collections.sort(resultsSearch);
    printPrecisionRecallPlotFileBased(resultsMcs, resultsSearch);
}