JARQL
SPARQL construct queries on JSON files
JARQL allows to execute SPARQL Construct Queries against JSON files. The project was inspired by Tarql which provides similar functionality for CSV files. With JARQL you can easily transform JSON Documents to RDF.
Build
Instead of building yourself you can download it from the release page.
In order to create an executable .jar containing all the dependencies just run:
mvn package -Pexecutable
Usage
java -jar jarql-<version>.jar <JSON-File> <Query-File>
For example if you have a file called paperino.json with the following content:
{
"parent": [{
"name": "Paperino",
"children": ["Qui", "Quo", "Qua"],
"fiance": {"name": "Paperina"}
}, {
"name": "Topolino",
"children": ["Tip", "Tap", "Top"],
"fiance": {"name": "Minnie"}
}]
}And a file called paperino.query with the following content:
PREFIX : <http://example.com/>
PREFIX jarql: <http://jarql.com/>
CONSTRUCT {
?p :name ?n;
:child [:name ?cn].
}
WHERE {
jarql:root jarql:parent ?p.
?p jarql:name ?n.
?p jarql:children ?cn.
}Invoking jarql-1.0-pre1.jar as follows:
java -jar jarql-1.0-pre1.jar paperino.json paperino.query
will output the following RDF:
[ <http://example.com/child> [ <http://example.com/name>
"Qui"^^<http://www.w3.org/2001/XMLSchema#string> ] ;
<http://example.com/child> [ <http://example.com/name>
"Qua"^^<http://www.w3.org/2001/XMLSchema#string> ] ;
<http://example.com/child> [ <http://example.com/name>
"Quo"^^<http://www.w3.org/2001/XMLSchema#string> ] ;
<http://example.com/name> "Paperino"^^<http://www.w3.org/2001/XMLSchema#string>
] .
[ <http://example.com/child> [ <http://example.com/name>
"Top"^^<http://www.w3.org/2001/XMLSchema#string> ] ;
<http://example.com/child> [ <http://example.com/name>
"Tap"^^<http://www.w3.org/2001/XMLSchema#string> ] ;
<http://example.com/child> [ <http://example.com/name>
"Tip"^^<http://www.w3.org/2001/XMLSchema#string> ] ;
<http://example.com/name> "Topolino"^^<http://www.w3.org/2001/XMLSchema#string>
] .
Advanced usage
Add all metadata you want using BIND combined with SPARQL 1.1 functions!
Create URI...
BIND (URI(CONCAT('http://example.com/ns#', ?var)) AS ?uri)
...add language tags...
BIND (STRLANG(?string, 'en') AS ?with_language_tag)
...or datatypes!
BIND (STRDT(?num, xsd:integer) AS ?int)
Current limitations
- Nested arrays in JSON are not supported
- The order of JSon Arrays is irrelevant
Similar Projects
- JSON2RDF provides streaming conversion from JSON to RDF.