[ http://opensource.atlassian.com/projects/hibernate/browse/HSEARCH-444?pag... ] Emmanuel Bernard commented on HSEARCH-444: ------------------------------------------ I can get it to work (in some cases) but I need to understand the reasons why you are using an entity-name. Are you keeping a 1-1 relationship between an entity-name and a class or are you using entity names to map the same class multiple times. If the later can you explain your use case? If the former, well can you explain your reason? The current "easy" patch bind an entity-name to a class and is half done (though will require more work on the DocumentBuilder and DocumentExtractor. That will work if we keep a 1-1 relationship I wonder if we can make the second approach work (ie the same class mapped multiple times via entity-names). That's definitively a harder problem that I am not willing to fight without a justified use case :) -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://opensource.atlassian.com/projects/hibernate/secure/Administrators.... - For more information on JIRA, see: http://www.atlassian.com/software/jira

[ http://opensource.atlassian.com/projects/hibernate/browse/HSEARCH-444?pag... ] Emmanuel Bernard updated HSEARCH-444: ------------------------------------- Fix Version/s: (was: 3.2.0.Beta2) User has not provided feedback so unassigning it to a version number -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://opensource.atlassian.com/projects/hibernate/secure/Administrators.... - For more information on JIRA, see: http://www.atlassian.com/software/jira

[ http://opensource.atlassian.com/projects/hibernate/browse/HSEARCH-444?pag... ] Begot Gilles commented on HSEARCH-444: -------------------------------------- In our project we use Teneo / Hibernate 3.6.0.Final to manipulate a MySql database and we use Hibernate Search 3.3.0.Final to index data stored in database. We use entity-name in XML mappings generated by teneo and in our java code. The patch for HSEARCH-444 is not integrated in HibernateSearch 3.3.0.Final. Can you put this patch in the next version of hibernate search? Your patch allows us to initialize hibernate search. However, our application fails when we post a query with hibernate search: java.lang.NullPointerException at org.hibernate.search.engine.ObjectLoaderHelper.executeLoad(ObjectLoaderHelper.java:144) at org.hibernate.search.engine.ObjectLoaderHelper.returnAlreadyLoadedObjectsInCorrectOrder(ObjectLoaderHelper.java:124) at org.hibernate.search.engine.QueryLoader.executeLoad(QueryLoader.java:81) at org.hibernate.search.engine.AbstractLoader.load(AbstractLoader.java:70) at org.hibernate.search.query.FullTextQueryImpl.list(FullTextQueryImpl.java:317) The function getClassMetadata return null because we use entity-name in ObjectLoaderHelper.java:144. or our application, i have modified the code of executeLoad like this (for example if indexed class is com.ExampleImpl -> teneo chooses Example for entity-name): private static Object executeLoad(EntityInfo entityInfo, Session session) { Object maybeProxy; ClassMetadata metadata = session.getSessionFactory().getClassMetadata( entityInfo.clazz ); String entity_name = null; if (metadata==null) { int indexLastPoint = entityInfo.clazz.getName().lastIndexOf("."); String lastSegment; if (indexLastPoint!=-1) lastSegment = entityInfo.clazz.getName().substring(indexLastPoint+1); else lastSegment = entityInfo.clazz.getName(); int indexImpl = lastSegment.indexOf("Impl"); if (indexImpl!=-1) entity_name = lastSegment.substring(0, indexImpl); else entity_name = lastSegment; metadata = session.getSessionFactory().getClassMetadata( entity_name ); } String hibernateIdentifierProperty = metadata.getIdentifierPropertyName(); if ( entityInfo.idName.equals( hibernateIdentifierProperty ) ) { //be sure to get an initialized object but save from ONFE and ENFE if (entity_name==null) maybeProxy = session.load( entityInfo.clazz, entityInfo.id ); else maybeProxy = session.load( entity_name, entityInfo.id ); } else { Criteria criteria = session.createCriteria( entityInfo.clazz ); criteria.add( Restrictions.eq( entityInfo.idName, entityInfo.id ) ); try { maybeProxy = criteria.uniqueResult(); } catch ( HibernateException e ) { throw new SearchException( "Loading entity of type " + entityInfo.clazz.getName() + " using '" + entityInfo.idName + "' as document id and '" + entityInfo.id + "' as value was not unique" ); } } return maybeProxy; } But the best correction would be to access directly to XML mapping to get the entity-name. But we must modify API... -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://opensource.atlassian.com/projects/hibernate/secure/Administrators.... - For more information on JIRA, see: http://www.atlassian.com/software/jira