If your data is already in RDF, you can add your files in .ttl (Turtle) or .nt (N-Triple) format to the /rdf folder before running the Maven build and the rest will be taken care of automatically.
If you have access to a live source of data that you want to integrate, you can add the code to fetch and integrate your data to the model to the project. We love community contributions, but please be certain to conform to the code of conduct when doing so. Remember that other users may not have access to your datasources; your implementations should fail gracefully if the sources are unavailable.
If you're just looking to experiment with making your data fit, the /rdf folder also supports data dumps in CSV format. There are a few conditions you have to conform to for it to work:
- Your CSV must either be UTF-8 encoded, or only use low-ASCII characters.
- You must use commas as field delimiters, and carriage returns as record delimiters.
- You can enclose fields in quotes
"if you like. - Leading and trailing spaces will be trimmed from unquoted field values.
- You must have a header row:
- The first header is the prefix namespace for your entities. It will be prepended to the values in the first column to create the URIs of your RDF subjects.
- The second and subsequent headers are the fully-qualified names of the predicates that you'll be providing to the model.
- The second and subsequent columns are the objects you'll be adding to the
model. Parsing uses a fallback scheme:
- If the value can be parsed as a URI, it's treated as an RDF Resource URI, unless the predicate is the schema.org URL predicate, when it will be parsed as a URL.
- If the value can be parsed as a number, it will be treated as an XSD Double literal.
- The value is otherwise parsed as a String literal.
There's a working example in the test resources.