This section deals with fulltext indexing with Nuxeo 5.2 and Jackrabbit. Please read Configure Nuxeo 5.2 with Jackrabbit and PostgreSQL for generic information and PostgreSQL-specific configuration.
To allow fulltext indexing of attached documents in Nuxeo, the general information at http://wiki.apache.org/jackrabbit/IndexingConfiguration is relevant.
For Nuxeo, use the following steps:
- start Nuxeo once with the general Jackrabbit configuration described in other documents,
- stop Nuxeo,
- create the file $NUXEO/server/default/data/NXRuntime/repos/default/workspaces/default/indexing_configuration.xml as described below,
- modify the file $NUXEO/server/default/data/NXRuntime/repos/default/workspaces/default/workspace.xml as described below,
- start Nuxeo again.
The file indexing_configuration.xml should contain your chosen indexing configuration, you can take example on the following:
The first include element will find all content nodes that are children of the main document's node, this covers the structure of the file schema, where there is a child ("content") having a property holding the binary data. The second include element will cover the structure of nodes used by the files schema, where there is a child ("files") that has children (one for each attached file in the list), who themselves have a child ("file") having a property holding the binary data. The lower nodes we're interested in have type "ecmft:content".
The file workspace.xml should be modified to add the path to your indexing_configuration.xml file, usually this means having: