McObject has released Perst 3.0, a major upgrade of its Android-ready, all-Java open source, object-oriented embedded database system. Developers working with the new release will benefit from new, highly efficient full-text indexing and searching within Perst databases, as well as enhanced documentation via a new 38-page Introduction and Tutorial, with ample code illustrations.
Perst’s built-in full-text search improves on the database’s earlier integration with the open source Apache Lucene information retrieval library. By making full-text search a native Perst feature, the embedded database gains greater efficiency compared to the earlier approach, and a smaller footprint (the Lucene .jar file is several hundred kilobytes in size, while including search capability in Perst adds just a few kilobytes).
Perst’s new full-text search feature uses a standard b-tree to store an inverse index of keywords, with parsing and stemming of indexed text performed either by user-defined code or by Perst’s default simple parser and stemmer. Logical operators are supported. For documents matching a query, Perst calculates the relevance of the document to the query.
Perst 3.0 marks improvements in the embedded database’s documentation. McObject has added a concise new 38-page Tutorial and Introduction that guides users through Perst’s features and design principles, using both explanation and code examples. Sections range from the basics – such as creating, opening and storing objects in a database – to advanced topics including schema evolution and database replication. Download the new Tutorial and Introduction at http://www.mcobject.com/downloads.php?step=6&id=86.
McObject offers the Android-ready Perst, as well as TestIndex, a demo application showing Perst and Android’s bundled SQLite database performing the same tasks side by side. Both are available as free downloads, with complete source code, from http://www.mcobject.com/android.
With Perst, users of Android-based phones will benefit from responsive, richly-featured embedded software made possible by a database system that delivers high performance and a small footprint, and by the efficiency that results from true Java developer-oriented features.
Perst stores data directly in Java objects. This eliminates the need for data-packing or unpacking code to map between the application's data model and the database's data model, as is required by relational and object-relational databases. The Perst API is flexible, easy-to-use and very fast compared to alternative commercial Java OODBMSs.
For more information, visit McObject's Web site at http://www.mcobject.com or contact Ted Kenney at firstname.lastname@example.org.