[hibernate-commits] Hibernate SVN: r11266 - in branches/Branch_3_2/HibernateExt/search: src/java/org/hibernate/search/backend/impl and 1 other directories.

Author: epbernard Date: 2007-03-09 12:08:56 -0500 (Fri, 09 Mar 2007) New Revision: 11266 Modified: branches/Branch_3_2/HibernateExt/search/doc/reference/en/modules/architecture.xml branches/Branch_3_2/HibernateExt/search/src/java/org/hibernate/search/backend/impl/BatchedQueueingProcessor.java branches/Branch_3_2/HibernateExt/search/src/java/org/hibernate/search/backend/impl/lucene/LuceneWorker.java Log: minor and doc Modified: branches/Branch_3_2/HibernateExt/search/doc/reference/en/modules/architecture.xml =================================================================== --- branches/Branch_3_2/HibernateExt/search/doc/reference/en/modules/architecture.xml 2007-03-09 16:28:58 UTC (rev 11265) +++ branches/Branch_3_2/HibernateExt/search/doc/reference/en/modules/architecture.xml 2007-03-09 17:08:56 UTC (rev 11266) @@ -19,7 +19,9 @@ <para><productname>Hibernate Search</productname> can also use a Lucene index to search an entity and return a (list of) managed entity saving you from the tedious Object / Lucene Document mapping and low level Lucene APIs. - The application code use the unified + The same persistence context is shared between Hibernate and Hibernate + Search ; as a matter of fact, the Seearch Session is built on top of the + Hibernate Session. The application code use the unified <classname>org.hibernate.Query</classname> API exactly the way a HQL or native query would be done.</para> Modified: branches/Branch_3_2/HibernateExt/search/src/java/org/hibernate/search/backend/impl/BatchedQueueingProcessor.java =================================================================== --- branches/Branch_3_2/HibernateExt/search/src/java/org/hibernate/search/backend/impl/BatchedQueueingProcessor.java 2007-03-09 16:28:58 UTC (rev 11265) +++ branches/Branch_3_2/HibernateExt/search/src/java/org/hibernate/search/backend/impl/BatchedQueueingProcessor.java 2007-03-09 17:08:56 UTC (rev 11266) @@ -89,7 +89,14 @@ public void add(Work work, List<Work> queue) { //TODO optimize by getting rid of dupe works if ( work instanceof UpdateWork ) { - //split in 2 to optimize the process (reader first, writer next + //split in 2 to optimize the process (reader first, writer next) + /** + * even with Lucene 2.1, use of indexWriter to update is not an option + * We can only delete by term, and the index doesn't have a term that + * uniquely identify the entry. + * But essentially the optimization we are doing is the same Lucene is doing, the only extra cost is the + * double file opening. + */ queue.add( new DeleteWork( work.getId(), work.getEntity() ) ); queue.add( new AddWork( work.getId(), work.getEntity(), work.getDocument() ) ); } Modified: branches/Branch_3_2/HibernateExt/search/src/java/org/hibernate/search/backend/impl/lucene/LuceneWorker.java =================================================================== --- branches/Branch_3_2/HibernateExt/search/src/java/org/hibernate/search/backend/impl/lucene/LuceneWorker.java 2007-03-09 16:28:58 UTC (rev 11265) +++ branches/Branch_3_2/HibernateExt/search/src/java/org/hibernate/search/backend/impl/lucene/LuceneWorker.java 2007-03-09 17:08:56 UTC (rev 11266) @@ -68,6 +68,11 @@ } public void performWork(UpdateWork work) { + /** + * even with Lucene 2.1, use of indexWriter to update is not an option + * We can only delete by term, and the index doesn't have a term that + * uniquely identify the entry. + */ Class entity = work.getEntity(); Serializable id = work.getId(); Document document = work.getDocument(); @@ -82,6 +87,11 @@ } private void remove(Class entity, Serializable id) { + /** + * even with Lucene 2.1, use of indexWriter to delte is not an option + * We can only delete by term, and the index doesn't have a termt that + * uniquely identify the entry. See logic below + */ log.trace( "remove from Lucene index: " + entity + "#" + id ); DocumentBuilder builder = workspace.getDocumentBuilder( entity ); Term term = builder.getTerm( id );