Lucene users (2.3): migrate to Lucene 2.3.1

In some specific cases, Lucene 2.3.0 can corrupt your index. It happens in very specific cases:

• you use autoCommit=false on IndexWriter
  
• or multiple threads are adding documents where some have term-vector enabled fields and some don't

These particular situations do not occur in the regular use of Hibernate Search, but upgrading to Lucene 2.3.1 works fine, so go upgrade, you won't be bitten later. Native Lucene users should upgrade as well.

Here is the official Lucene team announcement

Lucene Java 2.3.1 available

This release contains fixes for serious bugs in 2.3.0 that could cause index corruptions in autoCommit=false mode or in cases where multiple threads are adding documents where some have term-vector enabled fields and some don't. The autoCommit option was added to IndexWriter with release 2.2.0. If not explicitly set to false, the IndexWriter runs in autoCommit=true mode.

See CHANGES.txt for a detailed listing of changes.

2.3.1 does not contain any new features, API or file format changes, which makes it fully compatible to 2.3.0.

We would like to encourage everyone who is currently using Lucene Java 2.3.0 to upgrade to 2.3.1 to prevent possible index corruptions!

Lucene is available at apache.org.

Hibernate Search in Action book

I am very please to announce that a book on Hibernate Search is on its way. John Griffin and I are co-authoring Hibernate Search in Action from Manning.

The goal of this book is to give a good practical understanding of Hibernate Search and guide people through the steps of adding full text search capability into their Hibernate based application. The book also covers the necessary Lucene knowledge you need to use Hibernate Search on a daily basis.

An early version of the book is already available through the MEAP program. Five chapters are already out there. If you are interested, give it a try, we welcome your feedback!