Field comma separator

elendrim · **Posted:** Mon Jan 10, 2011 11:43 am

Hello,

I have made my fieldBridge to store a Set<String> :

Code:

import java.util.Set;

import org.apache.commons.lang.StringUtils;
import org.apache.lucene.document.Document;
import org.apache.lucene.document.Field;
import org.hibernate.search.bridge.FieldBridge;
import org.hibernate.search.bridge.LuceneOptions;

public class SetStringFieldBridge implements FieldBridge {
   
   public static final char SEPARATOR = ',';
   
   @Override
   public void set(String name, Object value, Document document, LuceneOptions luceneOptions) {
      
      if ( value == null ) {
         return;
      }
      
      // we expect a Set<String> here. checking for Set for simplicity
      if ( ! (value instanceof Set )) {
         throw new IllegalArgumentException("support limited to Set<String>");
      }
      
      @SuppressWarnings("unchecked")
      Set<String> set = (Set<String>)value;
      String values = StringUtils.join(set, SEPARATOR);
      
      Field field = new Field(name, values, luceneOptions.getStore(), luceneOptions.getIndex(), luceneOptions.getTermVector());
      field.setBoost(luceneOptions.getBoost());
      document.add(field);
   }
   
   

}

Code:

@Field(index=Index.UN_TOKENIZED, store=Store.YES, analyzer=@Analyzer(impl=SimpleAnalyzer.class))
@FieldBridge(impl=SetStringFieldBridge.class)
@ElementCollection
@CollectionTable(name="v_logicalitem_downloadtype", joinColumns=@JoinColumn(name="logicalitem_id", insertable=false, updatable=false))
@Column(name="downloadtype")
private Set<String> downloadtypes;

magix · **Posted:** Mon Jan 10, 2011 12:10 pm

Why don't you split up the content to more fields?
I.e. calling document.add(field) for every element in the set.

Matthias

elendrim · **Posted:** Tue Jan 11, 2011 5:08 am

Yes, you've right, I didn't know I could do that.
This is what I need, thanks !

elendrim · **Posted:** Tue Jan 11, 2011 5:23 am

Here is my new SetStringFieldBridge :

Code:

import java.util.HashSet;
import java.util.Set;

import org.apache.lucene.document.Document;
import org.apache.lucene.document.Field;
import org.hibernate.search.bridge.LuceneOptions;
import org.hibernate.search.bridge.TwoWayFieldBridge;

public class SetStringFieldBridge implements TwoWayFieldBridge {
   
   @Override
   public void set(String name, Object value, Document document, LuceneOptions luceneOptions) {
      
      if ( value == null ) {
         return;
      }
      
      // we expect a Set<String> here. checking for Set for simplicity
      if ( ! (value instanceof Set )) {
         throw new IllegalArgumentException("support limited to Set<String>");
      }
      
      @SuppressWarnings("unchecked")
      Set<String> set = (Set<String>)value;
      
      for (String string : set) {
         Field field = new Field(name, string, luceneOptions.getStore(), luceneOptions.getIndex(), luceneOptions.getTermVector());
         field.setBoost(luceneOptions.getBoost());
         document.add(field);
      }
      
   }

   @Override
   public Object get(String name, Document document) {
      Field[] fields = document.getFields(name);
      Set<String> set = new HashSet<String>();
      for (Field field : fields) {
         set.add(field.stringValue());
      }
      return set;
   }

   @Override
   public String objectToString(Object value) {
      if ( value == null ) {
         return "";
      } else if ( value instanceof String ) {
         return (String) value;
      } else {
         return String.valueOf(value);
      }
   }
   
   

}

sanne.grinovero · **Posted:** Tue Jan 11, 2011 12:29 pm

yes, Matthias is right the result is equivalent.
The only information you loose is relative positions of terms, but I guess you're not interested in that; you still have relative positions of the terms which make up each of your elements.
Relations from @IndexedEmbedded collections are encoded as your bridge is doing.