Jackrabbit dependencies

To use Jackrabbit in embedded mode you need to make sure that the JCR API and all required Jackrabbit libraries are included in your classpath. If you use Maven 2 , you can achieve this by specifying the following dependencies.

<dependency>
  <groupId>javax.jcr</groupId>
  <artifactId>jcr</artifactId>
  <version>1.0</version>
</dependency>
<dependency>
  <groupId>org.apache.jackrabbit</groupId>
  <artifactId>jackrabbit-core</artifactId>
  <version>1.5.0</version>
  <exclusions>
    <exclusion>
      <groupId>commons-logging</groupId>
      <artifactId>commons-logging</artifactId>
    </exclusion>
  </exclusions>
</dependency>
<dependency>
  <groupId>org.slf4j</groupId>
  <artifactId>jcl-over-slf4j</artifactId>
  <version>1.5.3</version>
</dependency>
<dependency>
  <groupId>org.slf4j</groupId>
  <artifactId>slf4j-log4j12</artifactId>
  <version>1.5.3</version>
</dependency>

The jcr dependency includes the JCR 1.0 API in your classpath. You need to explicitly declare this dependency as in jackrabbit-core the JCR API dependency scope is provided to work better in deployment models where the JCR API is shared between multiple applications.

The jackrabbit-core dependency pulls in the Jackrabbit content repository implementation and a set of transitive dependencies needed by Jackrabbit. See the Downloads page for the latest available version.

Jackrabbit uses the SLF4J for logging and leaves it up to the embedding application to decide which underlying logging library to use. In the example above we use the slf4j-log4j12 library which uses log4j 1.2 for handling the log messages. Note that the commons-logging dependency (which is a transitive dependency from Apache POI ) is explicitly replaced with the jcl-over-slf4j dependency that routes also all Commons Logging log messages through the selected SLF4J implementation. Jackrabbit 1.5.x uses SLF4J version 1.5.3.

The full set of compile-scope dependencies included by the above declaration is shown below. If you use a build tool like Ant where you need to explicitly include all dependencies, you can use this list to correctly configure your classpath.

+- javax.jcr:jcr:jar:1.0:compile
+- org.apache.jackrabbit:jackrabbit-core:jar:1.5.0:compile
|  +- concurrent:concurrent:jar:1.3.4:compile
|  +- commons-collections:commons-collections:jar:3.1:compile
|  +- commons-io:commons-io:jar:1.4:compile
|  +- org.apache.jackrabbit:jackrabbit-api:jar:1.5.0:compile
|  +- org.apache.jackrabbit:jackrabbit-jcr-commons:jar:1.5.0:compile
|  +- org.apache.jackrabbit:jackrabbit-spi-commons:jar:1.5.0:compile
|  +- org.apache.jackrabbit:jackrabbit-spi:jar:1.5.0:compile
|  +- org.apache.jackrabbit:jackrabbit-text-extractors:jar:1.5.0:compile
|  |  +- org.apache.poi:poi:jar:3.0.2-FINAL:compile
|  |  +- org.apache.poi:poi-scratchpad:jar:3.0.2-FINAL:compile
|  |  +- pdfbox:pdfbox:jar:0.7.3:compile
|  |  |  +- org.fontbox:fontbox:jar:0.1.0:compile
|  |  |  \- org.jempbox:jempbox:jar:0.2.0:compile
|  |  \- net.sourceforge.nekohtml:nekohtml:jar:1.9.7:compile
|  |	 \- xerces:xercesImpl:jar:2.8.1:compile
|  |	    \- xml-apis:xml-apis:jar:1.3.03:compile
|  +- org.slf4j:slf4j-api:jar:1.5.3:compile
|  +- org.apache.lucene:lucene-core:jar:2.3.2:compile
|  \- org.apache.derby:derby:jar:10.2.1.6:compile
+- org.slf4j:jcl-over-slf4j:jar:1.5.3:compile
\- org.slf4j:slf4j-log4j12:jar:1.5.3:compile
   \- log4j:log4j:jar:1.2.14:compile

Note that some of the transitive dependencies listed above may conflict with some other dependencies of our application. In such cases you may want to consider switching to a deployment model that uses separate class loaders for your application and the Jackrabbit content repository.

Starting the repository

Once you have your classpath configured you can start the repository with the following piece of code.

import javax.jcr.Repository;
import org.apache.jackrabbit.core.RepositoryImpl;
import org.apache.jackrabbit.core.config.RepositoryConfig;

String xml = "/path/to/repository/configuration.xml";
String dir = "/path/to/repository/directory";
RepositoryConfig config = RepositoryConfig.create(xml, dir);
Repository repository = RepositoryImpl.create(config);

See the [Jackrabbit Configuration] page for more information on repository configuration. See the RepositoryConfig and RepositoryImpl javadocs for more details on these classes.

Shutting down the repository

When your application no longer needs the content repository, you can shut it down with the following code.

((RepositoryImpl) repository).shutdown();

This will forcibly close all open sessions and make sure that all repository content is safely stored on disk.

The TransientRepository class

Jackrabbit comes with a TransientRepository class that makes it even easier to get started with a content repository. This class is especially handy for quick prototyping, but using the RepositoryImpl class as described above gives you better control over the repository lifecycle and is typically a better alternative for production code.

import javax.jcr.Repository;
import org.apache.jackrabbit.core.TransientRepository;

Repository repository = new TransientRepository();

This creates a repository instance that starts up when the first session is created and automatically shuts down when the last session is closed. By default the repository will be created in a “jackrabbit” subdirectory using a default configuration file in jackrabbit/repository.xml. See the TransientRepository javadocs for the ways to override these defaults.

Enabling remote access

Even if you mostly use the content repository in embedded mode within your application, it may occasionally be useful to be able to access the repository for example from an external administration tool while your application is still running. You can use the jackrabbit-jcr-rmi library to make this possible. To do this, you first need to add the appropriate dependency.

<dependency>
  <groupId>org.apache.jackrabbit</groupId>
  <artifactId>jackrabbit-jcr-rmi</artifactId>
  <version>1.5.0</version>
</dependency>

Make sure that you have rmiregistry running, and use the following code to export the repository. Note that you need to include the JCR API and the jackrabbit-jcr-rmi libraries in the rmiregistry classpath for the binding to work without extra RMI codebase settings.

import java.rmi.Naming;
import org.apache.jackrabbit.rmi.server.RemoteAdapterFactory;
import org.apache.jackrabbit.rmi.server.RemoteRepository;
import org.apache.jackrabbit.rmi.jackrabbit.JackrabbitRemoteAdapterFactory;

String url = "//localhost/javax/jcr/Repository"; // RMI URL of the
repository
RemoteAdapterFactory factory = new JackrabbitRemoteAdapterFactory();
RemoteRepository remote = factory.getRemoteRepository(repository);
Naming.bind(url, remote);

Use the following code to remote the repository binding from the RMI registry before you shutdown the repository.

Naming.unbind(url);

You need to keep a direct reference to the RemoteRepository instance in your code until you call Naming.unbind as otherwise it could get garbage collected before a remote client connects to it.

Note: RMI access is deprecated and will be removed in version 2.22.x (see JCR-4792 for details).

See the Repository Server page for instructions on how to access such a remote repository.

Embedded repository in a web application

If your want to embed Jackrabbit in a web application, you can use the classes in the jackrabbit-jcr-servlet library to avoid the above startup and shutdown code. To do this, you first need to include jackrabbit-jcr-servlet as a dependency.

<dependency>
  <groupId>org.apache.jackrabbit</groupId>
  <artifactId>jackrabbit-jcr-servlet</artifactId>
  <version>1.5.0</version>
</dependency>

Then you can instruct the servlet container to automatically start and stop the repository as a part of your webapp lifecycle by including the following servlet configuration in your web.xml file.

<servlet>
    <servlet-name>ContentRepository</servlet-name>
    <servlet-class>org.apache.jackrabbit.servlet.jackrabbit.JackrabbitRepositoryServlet</servlet-class>
    <load-on-startup>1</load-on-startup>
</servlet>

See the JackrabbitRepositoryServlet javadocs for the available configuration options.

You can then access the repository in your own servlet classes using the following piece of code without worrying about the repository lifecycle.

import javax.jcr.Repository;
import org.apache.jackrabbit.servlet.ServletRepository;

Repository repository = new ServletRepository(this); // “this” is the containing servlet

The benefit of this approach over directly using the RepositoryImpl or TransientRepository classes as described above is that you can later on switch to a different deployment model without any code changes simply by modifying the servlet configuration in your web.xml.

With this approach it is also easier to make your repository remotely available. Add the following configuration to your web.xml and your repository is automatically made available as a remote repository at …/rmi in the URL space of your webapp.

<servlet>
    <servlet-name>RemoteRepository</servlet-name>
    <servlet-class>org.apache.jackrabbit.servlet.remote.RemoteBindingServlet</servlet-class>
</servlet>
<servlet-mapping>
    <servlet-name>RemoteRepository</servlet-name>
    <url-pattern>/rmi</url-pattern>
</servlet-mapping>

Note that you also need the jackrabbit-jcr-rmi dependency in your application for the above configuration to work.