We have just released the 2.3.3 version of the UnboundID LDAP SDK for Java, with a number of improvements and bug fixes over the 2.3.1 release (NOTE: the only difference between the 2.3.2 and 2.3.3 releases is a fix to a javadoc formatting problem). You can get the latest release online at the UnboundID website or the SourceForge project page, and it’s also available in the Maven Central Repository.

As usual, the release notes provide a complete overview of changes made in this release, but some of the most significant updates include:

• A number of connection pooling improvements and fixes, including: It is now possible to create a connection pool even if a failure is encountered while establishing the initial set of connections, which is useful for cases in which the target directory server is unavailable when the application is starting. You can establish and/or close connections in parallel, which helps speed things up if the pool has a lot of connections or the server is remote or otherwise slow to respond. Pooled connections that cache server schema information are established more quickly. Methods have been added to help undo the effects of processing a bind on a pooled connection so that connections stay authenticated as the account used when they were initially established.
• A number of persistence framework improvements and fixes, including: The default object encoder now provides generic support for Serialized objects. A new getAll method makes it possible to retrieve all objects of a specified type below a given base DN. Fix a bug related to generating an appropriate set of modifications for an updated object, and for generating filters to define criteria to use when searching for objects. Improve the output of the generate-schema-from-source tool.
• A number of In-Memory Directory Server and LDAP listener improvements and fixes, including: It is now possible to combine multiple schema files to create the schema for the in-memory directory server. The canned response request handler now has the ability to customize the entries and references to return for a search operation. The LDAP debugger tool now has the ability to accept SSL-based connections.
• A number of LDIF-related improvements and fixes, including: Add an LDIFWriterEntryTranslator interface that can make it easier to transform entries to be written to LDIF. Fix a bug that could cause problems reading LDIF records with comments that have been wrapped across multiple lines. Add a convenience method to get the entry to be added from an LDIF add change record (whereas previously it was only possible to get the DN and attributes separately).
• A number of tool-related improvements and fixes, including: Add a new multi-server command-line tool API which makes it easier to create tools that need to interact with multiple directory servers. Make it easier to register JVM a shutdown hook to invoke code when the JVM in which the tool is running begins shutting down.
• Add the ability to specify the default SSL/TLS protocol version when creating secure connections (or using StartTLS) for cases in which the caller doesn’t explicitly specify a protocol.
• Dramatically improve the performance of the Attribute equals method for attributes that have a large number of values.