Using Sheet2RDF

Sheet2RDF comes both as a command line tool and as embedded tool of Vocbench (VB). If you want to use Sheet2RDF without installing Vocbench, you can then use it from command-line interface (CLI distribution).

Specific instructions for the VB tool can be found on its dedicated page while here we provide in-detail instructions for the CLI utility.

Which version to choose?

The user-interface of the VB tool allows for more interactive guidance towards the generation of the PEARL file, which will drive in turn the production of the triples from the spreadsheet content. However, there are a few scenarios where using the CLI might be preferable:

the PEARL has been already generated and finalized (possibly through the VB tool), so the CLI will just run the triples production
the PEARL generation from the spreadsheet is pretty immediate, or there is not much editing to do
the users are starting from scratch, but they are familiar with the PEARL language and they prefer to customize the PEARL rule file manually

Running Sheet2RDF from command line

To run a triplification process from command line you need to create a .properties file containing the system and process configuration. The following properties are available:

datasheetFile: path of the datasheet file to triplify;
inputPearlFile: path of the input pearl file. This is optional: if not provided, Sheet2RDF will try to automatically generate a pearl file from the available input spreadsheet
inputFile: input rdf file containing starting background knowledge which is incremented with triples generated by Sheet2RDF. This parameters is optional: if not provided, then the output file will be created and populated from scratch;
outputFile: this is the output rdf file that will contain the generated triples. This parameters is optional. If not provided, Sheet2RDF will write its output directly over the input file, in this case, inputFile must be specified, otherwise default name output.rdf will be assigned to this parameter;
outputSerializationFormat: serialization format of the output file (available values: BinaryRDF, JSON-LD, N3, N-Quads, N-Triples, RDF/JSON, RDF/XML, TriG, TriX, Turtle);
modelType: type of the model. The available choices are SKOS or OWL. If not specified, the default value is SKOS. This choice has currently no influence on the process, you can leave it unspecified;
baseUri: base URI of the model;
repositoryDir: the directory used by the triple store for local storage. This is optional, if omitted the default value is repDir, also some triple stores do not need it so in some cases it could be useless;
bundlesFolder: directory used to store CODA bundles (e.g. converters);
felixCache: cache directory of the Felix OSGi environment. Another optional parameter, if omitted felix will be set as default value.