A new software library is available from the MARC4J project Website (http://marc4j.tigris.org). The MARC4J Lucene API provides an easy to use and easy to configure utility for creating Lucene indexes based on MARC or MARCXML. Lucene is an open source text search engine library written in Java.
By default the library uses an index context based on the MARC to Dublin Core crosswalk, but users can create an index configuration using a simple XML format. It is also possible to store the full MARC record as binary content. A command-line utility is added to enable the creation of the Lucene index without the need to write code. The following command, for example, adds the MARC records in
input.mrc to an existing Lucene index using the given index schema:
java org.marc4j.lucene.util.MarcIndexDriver -index /home/index
-schema file:///home/schema.xml input.mrc
The library can be downloaded from the Documents and files section of the MARC4J project page at http://marc4j.tigris.org. Look for a folder called marc4j-lucene. The library is published under the LGPL license.