Hi, There seems to be a problem with writing Dataset into JSON-LD in a case such as the following. TRIG input
@prefix : <http://www.example.com/resource/100/v1#> . @prefix dc: <http://purl.org/dc/elements/1.1/> . :G { <http://www.example.com/archive/100/v1> dc:isVersionOf <http://www.example.com/resource/100> . } is read into Dataset and then written in JSON-LD. The resulting JSON-LD looks like this: { "@graph" : [ { "@id" : "../../archive/100/v1", "dc:isVersionOf" : { "@id" : "./" } } ], "@id" : ":G", "@context" : { "dc" : "http://purl.org/dc/elements/1.1/";, "@base" : "http://www.example.com/resource/100/v1#";, "" : "http://www.example.com/resource/100/v1#"; } } The fifth line in this JSON-LD is wrong. The problem occurs if a resource uri in a graph is the first part of the base uri (with empty string prefix).Is it a bug? Or is it not correct to have an empty string prefix? (if I change prefix ':' to e.g. 'v1:' everything works fine, but at the same time the empty prefix works fine for e.g. TRIG) My read TRIG and write JSON-LD code: InputStream is = this.getClass().getResourceAsStream(RESOURCE_FILE); Dataset dataset = DatasetFactory.createMem(); RDFDataMgr.read(dataset, is, RDFFormat.TRIG.getLang()); is.close(); sw = new StringWriter(); RDFDataMgr.write(sw, dataset, RDFFormat.JSONLD); System.out.println(sw.toString()); Kind regards, Yana
