(HibernateSearch) MultiFieldQueryParser different analyzer per field

Question

Some of my indexed fields use a Greek analyzer and I want to use an English analyzer for some other fields. My problem is: When searching for results (with a MultiFieldQueryParser currently), how can I use a different analyzer per field, so that a Greek analyzer is used for Greek-indexed fields and an English analyzer is used for English-indexed fields?

Answer 1

You could build your query parser like this:

Analyzer analyzer = fullTextSession.getSearchFactory().getAnalyzer(Item.class);
QueryParser parser = new MultiFieldQueryParser(Version.LUCENE_31, fields, analyzer);

which would use the proper analyzer, as defined in the annotations of your Item class:

@Field(name = "greekTitle" analyzer = @Analyzer(impl = GreekAnalyzer.class))
public void getGreekTitle(){
 //...
}

@Field(name = "englishTitle" analyzer = @Analyzer(impl = StandardAnalyzer.class))
public void getEnglishTitle(){
 //...
}

Answer 2

Here is the solution I found. Please comment.

transaction.begin();

PerFieldAnalyzerWrapper wrapper = new PerFieldAnalyzerWrapper(new StandardAnalyzer(Version.LUCENE_30));

wrapper.addAnalyzer("greekTitle", new GreekAnalyzer(Version.LUCENE_30));
wrapper.addAnalyzer("greekDescription", new GreekAnalyzer(Version.LUCENE_30));

String[] fields = {"greekTitle", "greekDescription", "englishTitle", "englishDescription"};

QueryParser queryParser = new MultiFieldQueryParser(Version.LUCENE_30, fields, wrapper);
queryParser.setDefaultOperator(QueryParser.AND_OPERATOR);
org.apache.lucene.search.Query query = queryParser.parse(QueryParser.escape(queryString));

javax.persistence.Query persistenceQuery = 
fullTextEntityManager.createFullTextQuery(query, Item.class);

@SuppressWarnings("unchecked")
List<Item> result = persistenceQuery.getResultList();

transaction.commit();

return result;