The aperture tool is quite impressive. I am still getting my head around RDF, but it looks interesting as well.
In the meantime, can someone point me in the right direction for getting to Lucene fullText indexes created by Aperture? I seem mention of LuceneSail in the PersistentStoreCrawlingExample, but I seem to be missing the connection.
At the moment, I would like to duplicate all the content and metadata Aperture extracts in a lucene index. As I understand more about RDF, I believe I would start using that instead.
Normal RDF stores aren't very good at querying fulltext. It's the same problem you have with LIKE queries in relational databases. LuceneSail is a tool that adds efficient full-text searching capability to an RDF store. It's not a part of Aperture, more an extension of Sesame - the RDF database. If you want to know more see this blog post
If you want to use LuceneSail in your app, you'd need to provide an appropriate implementation of the RDFContainerFactory interface, and pass it to the crawlers and accessors. This implementation would have to produce containers, which encapsulate models which in turn encapsulate sesame repositories created with LuceneSail.
It's not that complicated as it sounds, it would boil down to 4 lines of code. I haven't done it myself (yet) but it shouldn't be too difficult. We have a feature request to create an example,
you may want to monitor it to see if anything has changed in this respect.
Right now the LuceneSail is looking for a new maintainer. It's officially released anywhere. You'd need to get it directly from Aduna repository, more instructions here.
Hope this helps
... and apart from the feature request there's also a wiki page
It's a thing that's been there for years, just haven't made it to the top of anyone's priority list. You could send an email to aperture-dev mailing list, most developers follow it carefully.
Log in to post a comment.