Search returning no results

sol605 · **Joined:** Tue Apr 27, 2010 3:17 pm **Posts:** 4

I'm trying to get a simple Hibernate searcher set up, and while my objects are being indexed correctly, no results are returned with any query. Here's my
code:

Code:

      EntityManager em = entityManagerFactory.createEntityManager();
      EntityTransaction tx = em.getTransaction();
      tx.begin();
      
      FullTextEntityManager fullTextEntityManager = org.hibernate.search.jpa.Search.getFullTextEntityManager(em);

      QueryParser parser = new QueryParser("questiontext", new SimpleAnalyzer());
      org.apache.lucene.search.Query query = null;
      
      try
      {
         query = parser.parse("test");
      }
      catch(ParseException p)
               {
                       System.out.println("exception parsing text...");
                }
      javax.persistence.Query persistenceQuery = fullTextEntityManager.createFullTextQuery(query, Questions.class);
      List<Questions> result = persistenceQuery.getResultList();
      tx.commit();
      em.close();
      return result;

If I run a query through Luke that says questiontext:test, I get a bunch of expected results, so I know that the index is good and there are items in there that should be returned.

Any ideas? Thanks.

amin-mc · **Joined:** Wed Oct 03, 2007 2:31 pm **Posts:** 205

Are you using the same analyzer for indexing and searching?

sol605 · **Joined:** Tue Apr 27, 2010 3:17 pm **Posts:** 4

I don't set up an analyzer for indexing, I guess I'm using whatever default analyzer is there?

Here's my indexing:

Code:

      EntityManager em = entityManagerFactory.createEntityManager();
      EntityTransaction tx = em.getTransaction();
      tx.begin();
      
      FullTextEntityManager fullTextEntityManager = Search.getFullTextEntityManager(em);
      List<Questions> questions = em.createQuery("SELECT question FROM Questions AS question").getResultList();
      for (Questions question : questions)
      {
         fullTextEntityManager.index(question);
      }

      
      tx.commit();
      em.close();

amin-mc · **Joined:** Wed Oct 03, 2007 2:31 pm **Posts:** 205

Can you try new StandardAnalyzer(), if you are using Lucene 2.9 then you'll need to pass the version into the constructor. I presume Question object is very simple with one field questiontext?

amin-mc · **Joined:** Wed Oct 03, 2007 2:31 pm **Posts:** 205

Can you try this:

Code:

@Entity
@Table(name = "QUESTION")
@Analyzer(impl = StandardAnalyzer.class)
@Indexed
public class Question {

    @Id
    @GeneratedValue(strategy = GenerationType.AUTO)
    @Column(name = "QUESTION_ID")
    @DocumentId
    private Long id;

    @Column(name = "QUESTION_TEXT")
    @Field(name = "questiontext", store = Store.YES, index = Index.TOKENIZED)
    private String questionText;

   ...getters and setters
}

for searching i used hibernate

Code:

    QueryParser parser = new QueryParser(Version.LUCENE_29, "questiontext", new StandardAnalyzer(Version.LUCENE_29));
                org.apache.lucene.search.Query query = null;
                try {
                    query = parser.parse(term);
                } catch (Exception e) {
                    throw new IllegalStateException(e);
                }

                FullTextQuery fullTextQuery = fullTextSession.createFullTextQuery(query, Question.class);
                org.hibernate.Query hibQuery = fullTextQuery;

                return hibQuery.list();

I've tested the above and get results back.

Cheers
Amin

sol605 · **Joined:** Tue Apr 27, 2010 3:17 pm **Posts:** 4

Trying to construct the QueryParser as you did throws an error for me:

"The constructer QueryParser(Version, String, Analyzer) is undefined."

If I get rid of the first argument, then I get the error

"The constructor StandardAnalyzer(Version) is undefined."

I am using Lucene-core-2.9.0, so I don't think its a versioning issue...Thoughts?

Sol

amin-mc · **Joined:** Wed Oct 03, 2007 2:31 pm **Posts:** 205

I was using 2.9.1

For 2.9.0 You need to do the following:

Code:

QueryParser parser = new QueryParser("questiontext", new StandardAnalyzer());

sol605 · **Joined:** Tue Apr 27, 2010 3:17 pm **Posts:** 4

Cool, it's now returning results as it should. I've been playing with the analyzers and this is amazingly powerful stuff!

Question: I noted that in Luke, each result of a query has a relevance "Score." Is there a way to set a cut off for the Score in my java query, so that only highly relevant results will be returned? I'm having an issue with somewhat irrelevant results being returned when I query. I've improved it with StopWordsFilter and that sort of thing, but I'm curious if I can do better...

Sol

sanne.grinovero · **Posted:** Wed May 05, 2010 5:05 pm

you can't directly set a threshold, but instead of returning the entities in a query you could use projection to return pairs (score,entity) and hide what's too low.

Generally, it's not considered a good idea: it might happen that the score is lower than usually, still the results on top should be the most relevant you can find for a given query, so best strategy is usually to let the user decide if the results are ok or not.
improving relevance by using smart analyzers like stopword filters and stemmers is your best option

noula · **Joined:** Fri May 10, 2013 5:35 am **Posts:** 2

Hello,
I have the same problem with the query result display. But, I have another error message:

Exception in thread "main" java.lang.VerifyError: class org.apache.lucene.analysis.SimpleAnalyzer overrides final method tokenStream.(Ljava/lang/String;Ljava/io/Reader;)Lorg/apache/lucene/analysis/TokenStream;
at java.lang.ClassLoader.defineClass1(Native Method)
at java.lang.ClassLoader.defineClass(Unknown Source)
at java.security.SecureClassLoader.defineClass(Unknown Source)
at java.net.URLClassLoader.defineClass(Unknown Source)
at java.net.URLClassLoader.access$100(Unknown Source)
at java.net.URLClassLoader$1.run(Unknown Source)
at java.net.URLClassLoader$1.run(Unknown Source)
at java.security.AccessController.doPrivileged(Native Method)
at java.net.URLClassLoader.findClass(Unknown Source)
at java.lang.ClassLoader.loadClass(Unknown Source)
at sun.misc.Launcher$AppClassLoader.loadClass(Unknown Source)
at java.lang.ClassLoader.loadClass(Unknown Source)
at TestTagInDocument.main(TestTagInDocument.java:54)

I don't understand this error !
Note that I use the lucene 4.2.1 version and the error is in this line:
[code] org.apache.lucene.queryParser.QueryParser q1 = new QueryParser("path",TextFileIndexer.luceneAnalyzer);

Note also that I can display the result of the indexing process without any error.
Thanks.

sanne.grinovero · **Posted:** Fri May 10, 2013 6:05 am

Hi, you should use Lucene version 3.6.2 with latest Hibernate Search: support for versions 4 isn't integrated yet.

noula · **Joined:** Fri May 10, 2013 5:35 am **Posts:** 2

sanne.grinovero wrote:

Hi, you should use Lucene version 3.6.2 with latest Hibernate Search: support for versions 4 isn't integrated yet.

Thanks.. It Works !!