atomgraph/json2rdf

Streaming generic JSON to RDF converter

json2rdf's Issues

Releases not created for tagged commits

I noticed that beyond the first 1.0.1 tag, no Github releases have been created for subsequent tagged commits (1.0.2 etc) - is that intentional?

whitespace in json keys

characters.json:

{
  "characters": [
    {
      "first name": "Ash",
      "lastname":"Ketchum"}
      ]
    }

is valid json.

but:

cat characters.json | docker run --rm -i atomgraph/json2rdf 'http://tmp.com/
Exception in thread "main" org.apache.jena.riot.RiotException: <http://tmp.com/#first name> Code: 17/WHITESPACE in FRAGMENT: A single whitespace character. These match no grammar rules of URIs/IRIs. These characters are permitted in RDF URI References, XML system identifiers, and XML Schema anyURIs.
        at org.apache.jena.riot.system.IRIResolver.exceptions(IRIResolver.java:384)
        at org.apache.jena.riot.system.IRIResolver.resolve(IRIResolver.java:341)
        at org.apache.jena.riot.system.IRIResolver.resolveToString(IRIResolver.java:358)
        at com.atomgraph.etl.json.JsonStreamRDFWriter.write(JsonStreamRDFWriter.java:107)
        at com.atomgraph.etl.json.JsonStreamRDFWriter.convert(JsonStreamRDFWriter.java:66)
        at com.atomgraph.etl.json.JSON2RDF.convert(JSON2RDF.java:82)
        at com.atomgraph.etl.json.JSON2RDF.main(JSON2RDF.java:60)

should json2rdf just make the decision to delete non URI safe characters or maybe percent encode them?

Adding a Apache Spark UDF?

Hi! Would it make sense to have a small addition that makes the library usable in Apache Spark? Something along the lines of

package com.atomgraph.etl.json;
import org.apache.spark.sql.api.java.UDF1;
import org.apache.jena.rdf.model.Model;
public class Json2rdfUDF implements UDF1<String, String> {
private static final long serialVersionUID = 1L;
@Override
  public StreamRDF call(String jsonString) throws Exception {

       InputStream bis = new ByteArrayInputStream(jsonString.getBytes());
       Reader reader =  new BufferedReader(bis);

       StreamRDF rdfStream = new CollectorStreamRDF();
       new JsonStreamRDFWriter(reader, rdfStream, baseURI.toString()).convert();
       
       return rdfStream;
   }
}

# for namespace properties should be optional

From the README: "Property namespace is constructed by adding # to the base URI."

If the base URI ends in a forward slash (/), a pound/number sign (#) should not be added to the base URI.

Support for JSONL

Reading JSONL files (.ndjson) would be useful.

Incorrect version number used in Docker container

When running JSON2RDF in the Docker container is shown in the readme, I get this error message:

xxxx [~/dev/workspaces/graphstuff/JSON2RDF]> echo "{'label':'Hello'}" | head -100 | docker run -i -a stdin -a stdout -a stderr atomgraph/json2rdf urn:test:
Error: Unable to access jarfile target/json2rdf-1.0.0-SNAPSHOT-jar-with-dependencies.jar

The version number in the Dockerfile wasn't adjusted when bumping the version to 1.0.1.